[3°]
L6é
N&7

 

 

SMA MAA MIA ACCA TTC

      
  
    

  
 
  
 

 

» ae \\ Ua. . =
oe i i Ir
) ee Seat
ety

La i

|
cl! 'TAYLOR
COLLECTION||||/

| | I
VANTEC

Coe eee ee Oe ee

LULU a

Lo
uA

     
  

  
    
   

   

    
   

H

H

H

Hi
tt
H
H
Ht
Hi
H
Tt
i
Hi
H
H
Ht
H
HI
H

i TE
i
H|

H

H
Ht

UL

ua
tJ

all

 

 
“Hiatt .

 

 
 

Qoruell Law Schou! Library
[Ve (sou ,LAwavd Williaw , (PN (9gy
MEMOIRS

OF THE

NATIONAL ACADEMY OF SCIENCES

 

Volume XVI

 

WASHINGTON
GOVERNMENT PRINTING OFFICH
1922
BSY3/G
First Memoir, PI. 1.

Memoirs National Academy of Sciences XVI.

 

 

“OOULISIP OY} Ul UMOYS PUL[s] BIIEs IVY BURG JO PUO TIoY}IOU dy} Pus II WadMZ0q PuUe yuTOd ysIY oy) PUOsEq sat] Acq
oT} 0} soueI}Ue OGL “AI}sNpUl e[[TYOIO oy} JO FYSIoy ZuLMp poyooso ‘eTusofEO JoMOTT jo Auedurog pasoyeyy oy} JO (qJor) osnoyasem pue (WYSII) sJojIeNbpeoy oy} o1e sBuTp[Ing osivl omy OL,

“HOHONY ATIVNS() SdIHS SYSHAA ‘'GNV1S] VNSIVGDVIA| ‘SA0D YVAMA-,O-NVIA] 40 SHOHS NO S3DVTTIA DNIMOHS ‘9061 NI AVG VNATVGDVIA]

 

 
 

 

NATIONAL ACADEMY OF SCIENCES.

 

Volume XVI.
FIRST MEMOIR.

 

LOWER CALIFORNIA AND ITS NATURAL
RESOURCES.

BY

EDWARD W. NELSON,

CHIEF, BUREAU OF BIOLOGICAL SURVEY, UNITED STATES DEPARTMENT OF AGRICULTURE.

 

 
CONTENTS.

‘ es Page.
General information. ..... ia Diese Reena on Sto ce Seca eR alc egeweanate Bund aes ergata tee eee caer ela 7
The Biological Survey Expedition..............---..2.-----22-+-- sad (teases Les Saat a Wate Ra gn a dee SAY bts 13
Physical characteristics...........------.-- poise aly eo arenas aie Acie cals Se aeesd se esses a (eesvecae abesrapeiane deseo 48
Notes on the recent geological history of the peninsula. ...... 2.2.22... 2.02.0. e eee cc cee e en eee ence eee 52
Mountains of the northern interior and middle gulf coast. .-.....:.-....--2--2- 02-202 e eee eee eee eee eee eee 54
Mountains of the great volcanic plateau......-.---.. 222-2 e eee en ne nen eee eee eees 59
Mountains of the cape region south of La Paz Bay....-...-..-..-.----------+--- Osher ct tclin hod tea enc becca 62
Coastal mountains of the Pacific side. .<c.ccccscse eres seenewwebes exaxcieence es exceeds lbesss ssesegeer es cence 65
Coastal and other mountains of the northeast. ..... 2... 2-22. eee ee ee eee eee bd nee e eens 68
The plains and larger valleys.........-....-.- 22-222 eee cece cece eee cece e eee ee eenees pial ates coms 70
Streams and other surface waters............--..2-- 2222-2 ee eee eee eee sctciniai inte eeeevatatele dai de Dae mceiale aie siess 78
Underground waters....-...-..--..--- Ro Siet ASS yan tlehewat Motel ceasrcts ohare ake Raia eaten teat a ett tet Cophcnen tPA eet ctee cd aches 83
Coastalislandsscsccccnnsicnaseooredhe ose x sarceeres << meeence ene sete nesamens Ss ieee phen Ve poebeeoae eee 84
OliM Stes esi wwsesvemieewscina seereteeies 6 5 cue aice hes neste dincibaelns ada aueniea eds Leiden d 2 odyeeeelede See seed 95
Plants litre cicen ctoiciistach soca san ciauinenee nme Smelt An me ya VS AS ye st oman eI Rohe etal iteldipsntd ar enerciclereay tenes aieeunuan 103
Anima) 66 es aciciane aca ccseteoe rel chorcectna mas ees eer iinacicene seamed esis ewes Sraaines Sys an ueie a ah oe ee eee 110
Subordinate faunal districts secs .02isios!s end emiieeiistee et ddldebecn ees sosoerieerey abeeeecue czadectaee as eed aee 117
Life 20n68 v ecsxisesncesisiiene daemneeepeess senses thers Shame eeebe ee au Se cqee ds ake Se omnes 6 ndecucemene ese 121
Natural resources in animal and plant life................ 2.2.22. c eee eee eee ee eee eee eee renee 132
Agriculture and stock raising...........-...-2-2- 22-0222 eee eee eee eee eee ee 2 ead tue cyav Sr orae oeeREANvoe naa 136
Scleritific-explora tions ..35 92 gece istic ted eRe ein dae kei a poe gee afew bee eia hea eres 140
Biblopraphy:is:cc-ceues ek pow aeeeseowe es aclsaesweew es essen Seaway 14 ee wer evel caeeaeeer eens Poe eedewes oases 147

INd ek vecccc vovsuacssoniaee seas eee iat tea heawee Kmaeesereeneea cd ecacebasseeeeedues cameo ee didaeee ees 178
PLATES, Page.
1; Magdalena Bay in 1906 seccessewi seas ceecicecxeae ace sv eesbtesdeasee sass eeaveses cores cbs ese sues tee Frontispiece.
2. Fig. 1.—Cocopah Indian and brush hut on the Hardy River, in the delta of the Colorado River...........- 12

Fig. 2.—San Matias Pass, altitude 3,200 feet, between the Sierra Juarez and San Pedro Martir Mountains. . 12
3. Fig. 1.—Valley of Santo Tomas, south of Ensenada..........2..0.202 2-222 e een eee cece ence e eee e eee 13
Fig. 2.—Valley of El Rosario, south of San Quintin. ......-.-.0.- 20.222 e cece eee eee eee ence eenee 13
4, View looking north across San Felipe Bay.........-.-.--2.022 22-22 ce cece eee ene eee eee eee eee eee 16
5. Fig. 1.—Stream at mouth of La Providencia Canyon, east base of San Pedro Martir Mountains............- 17
Fig. 2.—Copal tree (Elaphrium macdougalii) near San Felipe Bay................62-. cece e eee eee eee eee 17
6. Fig. 1.—North slope near Vallecitos Ranch, San Pedro Martir Mountains, about 8,000 feet altitude......... 20
Fig. 2.—View across top of San Pedro Martir Mountains from near Vallecitos............-.-...-2-.2--2--- 20
7.-Fig. 1.—Meadow at La Grulla, San Pedro Martir Mountains, altitude about 7,000 feet...............2...-- 21
Fig, 2.—Meadow at La Grulla, showing pine forest and scattered granite bowlders..............--.-.- pce 21

8. Giant cactus (Pachycereus pringlei) on the road from Santo Domingo landing to Calmalli.-.............-.225: 24

9. Fig. 1—The Llano de los Buenos Aires, a part of the great interior plateau south of the sy mine 25
Fig. 2,—San Pablo Canyon below ruins of the old mission. ........--..-0- 2 eee eee eee eee eee e eee i 25

10. Fig. 1.—Summit of lava-capped plateau near Tinaja de Santa Ana, north of San. Ignacio. sued ededenehanss 28 ,
Fig. 2.—Canyon near Tinaja de Santa Ana in lava-capped plateau. . it dacee as are 28
li. Fig. 1.—San Ignacio in 1905. Old mission church in background... i. biaibraibyastessials idinttaya laltgenataoshuleas austere BY
Fig. 2.—Water hole in Arroyo San Angel on eastern side of Vizcaino Teste, west of San Temacio.- eat aciuy 32
12. Fig. 1.—Ojo de Liebre, or Jack Rabbit Spring, near head of Scammon Lagoon, oa Vizcaino Desert....:...- 33
Fig. 2.—Sea turtle corral at Scammon Lagoon...........22 2202 ee eee eee cee cee eee eee ee ne een eeee 33

18. Fig. 1.—Western border of the Vizcaino Desert. ......-.2.20-2- 220-2 e eee ee eee eee cence et eeeeeees 38

Bip) 2.—Comondu: Valley -cs« sacimiaccuennien seaielsiwiaee sng st.04 ceccee ced ascteaceecdsaues neon seca Rts 38

14, Fig. 1.—Landing place on beach, San Jose del Cabo..........2. 2.22.20 2 ee ee cece cece eee eee ence ee eeee 40

Fig. 2.—Creeping cactus (Lemaireocereus eruca) on the Llano de Yrais, east of Magdalena Bay.............. 40

15. Fig. 1.—Servatillo cattle ranch on plain east of Magdalena Bay.’......... 2.2.20. 2 eee eee eee eens 41

Big..2:— Water front at Ue Rags oo oicscseccssresccc ccna encase sere natn nestecvera naan elsnae P54 0d aces aa Sawer ot 41

16. Fig. 1.—Jungle of Arid Tropical and Lower Sonoran vegetation on northern slopes of the Sierra de la Tagiing,: 44
. Fig. 2,—Sugar-cane fields at Todos Santos . 44
17. Fig. 1.—Coastal plain at El Pescadero, south of Todos Santos...........-..-.-- fb dete icmmiidivieniet a 45
' Fig. 2.—View westward across San Jose Valley at Santa Anita.....-..........----. AE eh ea! Ceeeegudle 4B
18) ig, 1——CapelSan: Lueaa so: exacecscsme sauce ses anwenaiisysqaamedtnacenn naga aanaaaanetimnssuecacs scotad eens 45
Fig, 2:-San Jose: deliCabo in 1906 2 oscis.cj:o:ccajarereis since ore.nreisiaisig aeons aa am aaanne Aemeaamemeniecuarecteereenny Khas 45

19, Fig. 1—Camp at La Chuparosa in live oak forest. Upper slope Sierra de la Victoria.....................- 48
Fig. 2,-Vegetation in canyon at eastern base of Sierra de la Laguna................-2..2202002 00 eeceee eee 48

20. Fig. 1.—Santa Anita ranch in valley of San Jose del Cabo............2.... 2.022. e eee e eee eee eee eee e noes 49
Fig. 2.—East slope of Cerro Mayor, the highest peak in the Cocopah Mountains.................2...0.222. 49

21. Fig. 1.—Salt bed on east side Vizcaino Desert, west of San Ignacio...........-....-....0 2000s cece eee eeee 52
Fig. 2.—Mud volcano near Volcano Lake, in delta of Colorado. ........2.2..2. 0.2... e eee e cece eee cence 52

22. Fig. 1.—Desert waterhole north of Llano de los Buenos Aires........ 2.22.2... 2c eee eee eee eee eee e eee 53
Fig. 2.—Mining onyx on the Llano de los Buenos Aires........ 2222.2 220 00ce eee ee eect ec cece eeeeeeeeees 53

23, Fig. 1.—Island overgrown with cottonwood (Populus macdougalit), delta of Coloride: 3 in process of erosion. 78
Fig. 2.—Flood plain of Hardy River at east base of Cocopah Mountains, April, 1905.....-......2.2.0.222.0.. 78

24, Fig. 1.—El Tule Arroyo, a clear stream in canyon at junction of Sierra de las Tinajas and Sierra Juarez..... 92
Fig. 2.—Tinaja Santa Clara on west side of Vizcaino Desert........-.-.2.---. 0-200 2 2-2 cence ee eee eee eee 92

25. Fig. 1.—Rattlesnake Harbor on southeastern shore of Angel de la Guarda Island...............2.......... 93
Fig. 2.—Sea Elephant Cove on the northwestern shore of Guadalupe Island.................2.2.2.0-0.2... 93

26. Fig. 1—Forest of yellow pine and sugar pine near Vallecitos, San Pedro Martir Mountains. .............. 104
Fig. 2.—Vallecitos, 8,000 feet altitude, San Pedro Martir Mountains. .............-.2.02.00.ccce cece eee 104

27. Desert vegetation at Arroyo Catavifia on granite tableland south of Llano de los Buenos Aires.............. 105
28. Fig. 1.—Jungle of cactuses, yuccas, and other thorny plants in the northern part of the Vizcaino Desert... 108
Fig. 2.—Elephant wood trees ( Veatchia discolor) near San Andres........-..... 2.0.0 2ceeceesecceeecceccece 108

ILLUSTRATIONS.

 

  

 

  

4
{Vol. XVI, No. 11 ILLUSTRATIONS. 5

Page.

29, Fig. 1.—Cirio forest (Jdria columnaris), near San Fernando..............---. 2-2-2 eee e cece e eee senses sey 109
Fig. 2.—Forest of Yucca valida near Santo Domingo Landing...............2..2--22 22 eee eee eee ee eee eee 109

30. Fig. 1.—Orchilla (Roccella) growing on fouquieria bush near San Andres.........--.-.------2eeeeeeee eres 109
Fig. 2.—Epiphytic plants growing on desert shrubs in foggy coast belt near San Andres...-.......---.----- 109

31. Map of subordinate faunal districts in Lower California..............0.2-00 20 cece eee ee eee eee cee eee eeeees 118
82. Map of life zones of Lower California. .-....-. 2.2.2.0. cece eee e cece cee eee ete tee eee e cee eee eeeeeee 120
33. Fig. 1.—Palo blanco forest (Lysiloma candida) in valley above San Ignacio, on trail to Santa Rosalia. ....-. 136
Fig. 2.—Bark of palo blanco in sacks gathered for shipment at Cape San Lucas village..................--- 136

84. Fig. 1.—Taking honey from bees’ nest in a small cave near Matomi.................0.22220220 eee eee eee 140
Fig. 2.—Cowboy near Cape San Lucas............. 0000 c eee e eee eee cece eee eee eee e eee e eect eeeeeees 140

35. General map of Lower 'California...............--...2-- eave tala Bald la ashes creieeee Glens h sas SICroN EIS Caane & be imieteialelelene 172
LOWER CALIFORNIA AND ITS NATURAL RESOURCES.
(With 35 plates.)

_ By Epwarp W. Netson,
Chief, Bureau of Biological Survey, U. S. Department of Agriculture.

GENERAL INFORMATION.’

The present account of Lower California has been prepared as one of the results of an
expedition from the Bureau of Biological Survey of the United States Department of Agricul-
ture, which traversed the entire peninsula during 1905 and 1906 under my direction, assisted
by Mr. E. A. Goldman.

The difficulty of securing definite information concerning Lower California and the erroneous
character of many published statements concerning it have led me to include, in addition
to a brief account of our journey, short chapters on various phases of its physical character,
resources, and other subjects, with a bibliography.

A variety of causes has resulted in long delaying the appearance of this report. Its publi-
cation at the present time, however, appears opportune in view of recently awakened interest
in the peninsula. Some of the changed conditions or new developments which have occurred
since the time of the expedition are noted in their proper places.!

Lower California was discovered by Cortez in 1535, soon after he completed the conquest
of the Mexican mainland. Efforts were at once made to colonize the new territory, which was
supposed to be an island, and the first colony was located where La Paz now stands.

At the time of its discovery, Lower California was inhabited by Indians in a most primi-
tive state of development. According to the early records these natives numbered about
25,000 individuals, which were grouped under three tribal designations, distributed as follows:
The Pericues were ‘said to have occupied the peninsula from Cape San Lucas to about 24° north
latitude, including Ceralvo, Espiritu Santo, and San Jose Islands. The Guaicuras were said
to have occupied the middle region from near 24° to near 26° north latitude and Cedros Island.
The Cochimies occupied the northern section from 26° to 33° of latitude, including the adjacent
islands along the western coast. From our present knowledge it is evident that this classi-
fication of the Indians was extremely imperfect, particularly as referring to those of the northern
section, in which most of the surviving native population now lives. It is known that sur-
vivors of several tribes still exist in northern Lower California, of which perhaps the Cocopahs
about the delta of the Colorado are the most notable. The Indian population of the peninsula
has decreased until only a few hundred now remain, and some of the tribes have no doubt
See forever.

Information, concerning Indian population in Lower California at various periods.

 

ae

 

 
 
 
 
 
 

 

 

. Locality. ‘Year. Number. Year. Number.
San Francisco Xavier Mission. ....-..-...- 2... 222-22 eeee cence eee ee Noassiet 1768 485 1857 56
Guadalupe Mission...........0....- 0022.0. 0 ee cee cece eee cece tececetee| 1768 520 1857 i
La Virgin de los Dolores Mission..................- pleas duatapeiaistacas aoe giaaPogectoye:y 1768 450 1857 6
San Luis Gonzaga Mission . - Soidavdes, 1768 310 1857 | 20
Santa Gertrudis Mission 1768 1,000 |. 1857 4
San Borja Mission. . 1768 1, 500 1857 3
Santa Maria Mission.: 1768 330 | 1857 0
San Fernando Mission... Fssinay ean emma --| 1710 530 1857 ~ 3
San Pedro Martir Mission.........-.-2-22.--.--- iaiuy Wiese ae OG Sealing ...-| 1796 420 1857 1

 

 

 

i: ' 1'The chapter on general information includes data received to the end of the year 1919.

i
8 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

As already stated, at the time of its discovery the Indian population of Lower California
was estimated to number about 25,000. In 1857 the number was thought to be about 2,500.
In 1919 it is reported that, there:remain,- about, 600. insthe northern, district; many of hom, no
doubt, are ‘of mixed blood. Few of these remain at the old missions, most of them being in
scattered villages. and rancherias. At Santa Catarina Mission they number 45, as against
180 reported in 1898. About 1914 the inhabitants in the Indian village La Huerta, at the base
of the Juarez Mountains, numbered 5. This 4 is the largest’ number of Indians in any village
I know in this’ region. "They: have practically disappeared from the southern district.

In connection with the Indian population of the peninsula it is interesting to note that
a considerable number of Yaquis from Sonora have for many years worked in the mines along
the gulf coast of the peninsula. About 1909 a colony of ‘these Indians, who had been working
at the placer mines at, Miramar, settled, just above the village of El Rosario, on the west, coast,
where |. they: developed. water and are cultivating a small area of land. _This colony now numbers
about 30 individuals. - eat §

Between the heatilley of ihe indians ‘and’ the severe desert conditions of the. new. land. the
early colonists suffered great dangers and privations. In the face of extreme dangers ‘and
hardships some of the missionaries, notably Padres Kino, Ugarte, and Salvatierra, animated
by a wonderful. spirit of self-sacrificing heroism, ‘gradually explored the entire peninsula and
established missions at the main watering places throughout its length, Nearly four centuries
have passed since the discovery and all but two or three of the numerous missions, reared with
such labor, are in ruins, and the sites of many of them are abandoned, while all. but a handful
of the Indian. tribes which once roamed the deserts have disappeared.

The desert character of the peninsula, combined with political handicaps, have continued
to hold back its settlement and development by Europeans. As a result, “notwithstanding
its close, proximity to California, readily available information concerning the country and its
natural resources has been scanty and frequently misleading. The main sources of detailed

‘ information concerning it in,the United States, have been the reports of various American
naturalists, including zoologists, botanists, and geologists, who from time to time have explored
parts. of the peninsula.

Two Mexican Goyernment reports, however, contain much detailed information on the
geology and resources. The first is by J oaquin | Ramos concerning an expedition to the middle
parts of the peninsula in 1884. The other is a joint report of a commission sent by the Instituto
Geologico to the northern district in 1911. The account by Venegas of the early exploration and
occupation of the peninsula has been translated into English and is full of interest.

Mines. : ine ' a eh) 8

During the last 50 years many etdapentens,- mainly Avaovioana ‘have aaa over the
country, and afew mining camps, usually of small size, have been established. Prospects
have been found on the peninsula showing that a considerable variety of minerals.exist. These
include gold, silver, copper, lead, zinc, iron, manganese, mica, salt, sulphur, Ang Repo: in
addition to turquoise, tourmaline, onyx, and marble.

The largest existing mining camp is that of Santa Rosalia, on the middle of the gulf joast
where a French company has developed El Boleo, one of the larger copper mines of the world,
which in 1905 was said to support a population or several thousand people. The largest silver
mine on the peninsula, El Triunfo, located south of La Paz and worked by an American com-
pany, once supported a mining camp of considerable size. More recently it declined, until in
1905 only a small force was employed. “

«At the time of our visit, onyx was being mined by an American company at the onyx mine
on the east side of Llano de los Buenos Aires, whence it was hauled by wagon to Santa Catarina
Landing on the west coast. and shipped to San Diego. El] Alamo, a gold and silver mine south-
east of Ensenada, once of some importance, was being worked on a small scale at the time of
our visit, but I learn that in 1912 the work ceased and now the camp is practically deserted.

Mines carrying lead, zinc,,and silver are now. being worked on a considerable scale on the
east side of the Pinta Mountains near the mouth of the Colorado River. During the war small
No. 1] LOWER. CALIFORNIA, - NATURAL RESOURCES—NELSON. 9

quantities of copper. ore. were mined and shipped from prospects near Salada, San Quintin,
‘Sausalito, San sepoainto;: Santa pases: San Julio,: eae at ‘San Jose, aout 25 miles south of
‘Ensenada: . * ty qi aubsoadd ey, My ‘ eR ;

Many schill mines a prospeots ia — corte” a time to time in all parts of the
‘peninsula, including the dry placer mines at Santa Clara, west of the Vizcaino desert; at Cal-
malli, in the middle of the peninsula, on the northern border of this same desert; and at San
Rrancisquito, onthe Gulf side, a few miles northwest of the: old mission of' Calamshue: These
placer mines' were long since abandoned. Other placer mines, notably at Miramar, near the
gulf coast, and on the west base of the San: Pedro Martir Mountains, have been working nearly
-or quite up to the present.time. ' Efforts to develop mines in Lower California are at once met
by three serious handicaps—first, lack of transportation facilities; second, the scarcity of water;
and third, the equally: important lack of fuel. As a consequence, the expense of maintaining
operations in practically all parts of the-peninsula are prohibitive except in the case of MUGS
produce mines.

The physiography: and natural resources of Lower California, andiading: game, iilaling,
florea. and other sea products, and the agricultural and natural hee products, are treated
elsewhere under special headings.

dex 4

Government.’

‘Lower California is a Federal territory officially divided for administrative purposes by the
twenty-eighth degree of north latitude into a northern and a southern district, each adminis-
tered by a governor appointed by the President of Mexico, the capital of the southern district
‘being La Paz and that of the northern district Ensenada. In addition to being the head of the
civil administration, the governor of each district is also commander of the military forces
within his jnviadiction,

The northern district has for several years been slinitaieea by Gov. Esteban Cantu. For
the first. time in the history: of the peninsula northern Lower California has an energetic and pro-
gressive governor, who is taking active steps to develop his territory. He is completing an
automobile highway from Mexicali westward across the mountains to Ensenada. He has also
built. a military road, passable for automobiles, from Mexicali southward to San Felipe Bay,
where he plans to develop a port, eventually, it is reported, to be connected with Mexicali by a

-railroad. The enormously rich agricultural lands of the delta of the Colorado lying tributary
to Mexicali are of the same quality as,.and in reality a direct extension of, the lands of the Im-
perial Valley lying immediately adjacent on the California side of the border.

‘In connection with the government of Lower California, it is of interest to know that during
two short periods the peninsula has been in the complete possession of Americans. During the
Mexican War in 1847 the forces of the United States occupied the principal points, with head-
quarters at La Paz, but relinquished them at the end of the war. | In 1853 and 1854 the penin-
sula was again occupied and a government organized there by a band of American filibusters
under Walker. This ill-advised adventure ended in disaster, and later Walker was killed on a
similar filibustering expedition in Nicaragua. :

Land Grants. he set

For many years much ‘the larger part of Lower California has been covered by land grants
of enormous size given ostensibly for colonization and developmental purposes. One company
held nearly 12,000,000 acres. Under these grants, during the last half century, attempts were
made to colonize various parts of the peninsula, mainly from the United States, all of which
have been complete failures. Twenty families of Russian immigrants colonized Guadalupe
Valley north of Ensenada in 1906, where they were reported to be doing well for several years,
growing wheat, corn, oats, and other agricultural products. In July, 1919, the colony contained
45 families, most of whom were planning to move to the United States, owing to the difficulties
of getting crops and to the uncertainties of a market. There is no question that the existence of
the land.grants, including the enormous area held by the Magdalena Bay Co., were euceodingly
detrimental to the development. of the peninsula as a whole.

 

1 Statements regarding the government of Lower California were in type before the administrative changes of the summer of 1920.
10 MEMOIRS NATIONAL ACADEMY OF SCIENCES.’ .. [Vou. XVI

From 1915 to 1917 the Mexican Government canceled ‘these grants, announcing in some
cases that owners of land who had purchased from the holders of the grants previous to this
cancellation might have their titles confirmed by proper Pepreeiiaion to Bhp BOPBLATIO de
Fomento, in the City of Mexico.

The one part of Lower California which haa become familiar to the people of the United
States in recent years is Magdalena Bay, the surroundings andthe coast of the peninsula for a
long distance to the north of which were held under grant by'the Magdalena Bay Co. (the
Chartered Co. of Lower California), an American corporation. Rumors as to the occupation
of Magdalena Bay by the Japanese and the sale of the great concession controlling the bay to
foreign interests, owing to its proximity to our west coast, have eee given rise in this
country to excited newspaper publicity and editorial comment.

Some of the companies holding grants spent enormous sums of money in ill-advised efforts
to colonize, making failures in some districts where efficient direction would certainly have ore
duced considerable results.

The delta plain of alluvial lands of Lower California bordering the lower Colorado’ River
and adjoining the southern border of the Imperial Valley of southeastern California is by far
the richest and most important agricultural section of the peninsula. During recent years
development in this area has been rapid and will continue. Long-staple cotton, melons,
alfalfa, and barley have been the main crops, and it is reported that experimental plantings of
rice and sugar cane have done well.

Population.

Owing to its desert character, Lower California has always been sparsely populated.. With
the exception of the large copper camp at Santa Rosalia, the population of the peninsula is mainly
located in the northern and southern sections. In addition to the mining camp just mentioned,
the main towns and villages are La Paz, San Jose del Cabo, Ensenada; ‘Mexicali, San Ignacio,
Mulege, Comondu, and La Purisima.. The remainder of the jabab tae are in. smaller mining
camps, at insignificant villages, and on ranches.

The World War had considerable influence on the population of this region. - This perhaps
was shown most strongly at Santa Rosalia where it is reported’ that the mining activity was
greatly increased and the population of this camp is reported to have numbered.as high as from
7,000 to 8,000: Some of the smaller mining camps were closed through the effects of the war or
previously, owing to revolutionary outbreaks. At the same time a number of the small mining
camps were stimulated to increased activity.

- According to the census of 1910 the population of the satis peninsula was about 32,600
inhabitants, of which the northern district held 9,760, and the southern district about 22,840.
The population of the principal towns at this time was as follows: —

. Northern district.—Tijuana, 273; Mexicali, 462; Ensenada, 2, 170; El Alamo, 307; San
Quintin, 141; El Rosario, 359; Calmalli, 108...

Southern district.—Santa Rosalia, 3,807; Mulege, about 7503. Sari Ignacio, 7 93; La Purisima,
609; Comondu, 1,050; Providencia, 1,653; Purgatorio, 2,057; La Soledad, 1, 551; Magdalena
Bay, 144; Todos Santos, 1,141; La Paz, 5, ‘536; Triunfo, 2 341; San Jose del Cabo, 1,397.

During the last fow years the population of the northern, district of Lower. California has
fluctuated greatly, owing to ‘rapidly changing conditions and especially to the great agricultural
development which has taken place on the delta plain of the Colorado, tributary to Mexicali.
According to the best information I can obtain from people personally familiar with the north-
ern district the population i in 1919 was approximately 16,000, distributed mainly as follows :,

Municipality of Mexicali. (September) babel dest oceneemnee Pen cewt.caseseas aan ota oe -1. 12, 000
Tijuana.......-.--- Uisipelntujanipivia sve dabiartee tad arremminchtn Mogi aa swieiR A gee Pee seasise sais * 300 .
Tecate, and adjacent ranches. gs ahs nt sapate ance shag Suna ln aihenasayensietafadsta tw vies eels gaan aah xe paises 500
Ensenada...........2.2220--06- Sie Hesrctdeaisrseacsrashara alee aid ateigloncdeaiceniee axa ves ved savesseus 2, 000
San Quintin:......... Ieee ase aly Aa On aay A ees nL I a cpornd a ean hc, icteric Gata 85
El Rosario..........- eee eee Uwe P54 ets uiuisnon RRO erne caw alewinet is <a vaheey 12h4-Re ses 300

Calmall dss. cc cacanceaaices cock sasade tea aban artd deem hcien CaeRa eet ise ebe ube teens 800
No.1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 11

‘The municipality of Mexcaliincludesall of the agricultural section of northern Lower California.
Its population is at a maximum during the cotton-picking season, in the fall, when it reaches a
total of about 12,000. Of this number about 5,000 are Chinese and 300 to 400 are Japanese.
The bulk of the remaining population is Mexican. Apparently owing to the development of
labor difficulties, Governor Cantu, on September 10, 1919, issued an official decree prohibiting
Asiatic immigration | into the northern district of the ‘peninsiile.

About 600 Indians are reported to remain in northern Lower California, living in small
scattered rancherias and isolated ranches. ae

Owing to its geographic location, Lower California i is isolated from other parts of Mexico and
has had much intercourse with California. My expedition through the peninsula was made
after I had spent a considerable number of years in Mexico proper, and it was interesting to note
certain characteristics of the people of the peninsula. We found them everywhere honest and
friendly. There was a general absence of revolvers and machetes among the people of the
peninsula, in marked contrast to the general custom of carrying one or the other of these weapons
throughout most of the other parts of Mexico. During all of this trip, owing to the obviously
peaceable and friendly character of the people, even in the remotest sections, for the first time
in Mexican territory we felt no need of weapons for self-protection.

Wagon Roads.
+» Wagon roads in Lower California are few and poorin quality. The main seat is hat extend.
ing along the Pacific coast from Tijuana southward through Ensenada and San Quintin to beyond
El Rosario. A nearly completed automobile road is being built from Mexicali westward across
the mountains to join the road from Tijuana to Ensenada. Another road has recently been
completed from Mexicali southward to San Felipe Bay. For many years a wagon road has been
used from El Gampo on the California border south along the top of the Pinal Mountains to El
Rayo, and thence to San Rafael Valley on the west slope. A wagon road also connects Ensenada
with San Felipe Bay on the gulf, passing through Trinidad Valley. San Fernando and the onyx
mines on the Llano de los Buenos Aires are connected by road with the San Carlos and Santa
Catarina Landings. A’ wagon road also connects the mining camp at Calmalli with Santo
Domimgo Landing on the Pacific coast. Another frequently used road connects La Paz with the
mining camp of El Triunfo and Todos Santos on the west coast. Numerous other short roads
have been built from time to time by mining and other companies at various points on the pen-
insula, but they are little used, many have been abandoned for many years, and none are kept
in good condition.

Automobile travel is soasible over aie of the main roads i in the northern part of the pen-
insula, but more or less frequent difficulties are encountered. A considerable part of the traffic
throughout the peninsula is still maintained by the primitive means of pack animals over trails.

Seaports.

Port facilities in Lower California are scarce and usually of a primitive character. Custom-
house officials and port facilities, such as they are, are provided at Ensenada, San Quintin,
Magdalena Bay, La Paz, and Santa Rosalia. The last-named port is the shipping point for
copper produced by the great El Boleo copper mine. There a breakwater has been built of slag
from their furnaces, within which ships may be protected from the storms which sweep the gulf.
‘A number of bays along the coasts afford a certain amount of shelter, but there is no
population back of them to encourage their use. Most landings in the peninsula are made from
open roadsteads such as exist at San Carlos, Santa Catarina Landing, Playa Maria, Santa Rosalia
Bay, Santo Domingo Landing, Cedros lsiand, San Jose del Cabo and at numerous points along
the gulf coast.

Healthfulness.

As a whole Lower California is characterized by the remarkable healthfulness of its climatic
conditions. This is ‘especially the case in the northern half of the peninsula. In the southern
half, malaria occurs in certain places and there have been outbreaks of yellow fever from time
to time, the infection being brought from Mazatlan and other points on the Mexican mainland.
12 MEMOIRS NATIONAL ACADEMY OF SCIENCES. |: | [Vou. XVI

In 1883 a' severe epidemic of yellow fever visited: the.southern end of the penis extending
as far north as Magdalena Pay: mat San cia ; coe
Map. ; : BY

The accompanying general map of Lower California (see Pl. 35) is the most, complete yet
published. The coast line and islands are accurate, being from Hydrographic Office charts based
on surveys of the United States Navy. ‘The Idgation of the mountain summits and their altitudes
throughout the peninsula are practically all ‘from the same source. The maps of Dr. Gustav
Eisen covering | the southern part of the peninsula, published by the California Academy’ of
Sciences and the American Geographical Society, have been the basis for that region. Manu-
script maps ; from David Goldbaum have supplied the main features of the northern district, sup-
plemented by others covering certain districts by John B. Burnham, Ottomar H. Van Norden,
and E. W. Funcke. To the foregoing have been added data’ obtained during the expedition by
K. A. Goldman and the author. It is fully recognized that this map contains numerous inaccu-
racies of detail, but it is hoped that it will serve a useful purpose until more careful surveys are
made. —

Through the lack of designations for numerous geographic features, have thoughti itadvisable
to propose names for some of them. The names are mainly those of some of the early Spanish
explorers and missionaries, who first made known the coast and interior of the peninsula, and
of certain American naturalists, whose scientific work has given us much of our information con-
cerning the wild life of this region. These names are as follows:

Cerro Kino (4,289 feet), in honor of Padre Kino, one of the early missionary explorers.

Cerro Salvatierra (7,062 feet), in honor of Padre Juan Maria Salvatierra, one of the early
missionary explorers. (a :
~~. Cerro Junipero Serra (3,189 fot in honor of: eee Fonipero ome one of the early
missionary explorers.

Cerro Ugarte (5,400 feet), in oe of Padre Juan Ugarto, one of the: early missionary
explorers.

Anthony Peak (4,593 feet), in honor'of the naturalist ‘Alfred W. Anthony, to whose work
through a series of years we are indebted for ure information conecening the birds and mam-
mals of the peninsula.

Bryant’Peak (4,485 feet); in honor of the iittevaligt Walter L. — whose axplorvtons
and 'publications added much to our knowledge of the bird and mammal life of the peninsula.

Goldman Peak (5,000 feet), in honor of the naturalist Maj. Haven A. Goldman, who made
the expedition with me in 1905 and 1906. =

Brewster Peak (4,485 feet), in honor of the iataralist the late William rewatar who pub-
lished a notable account of the birds of the southern district:

Llano de San Quintin, the plain bordering San Quintin Bay.

‘Vizcaino Desert, in honor of the navigator Sebastian Vizcaino.

Llano de San Bruno, the:plain on the Gulf coast where the village of Sit Bruno is located.

Magdalena Plain, the coastal plain fronting Magdalena Bay and Island.

ACKNOWLEDGMENTS.

It gives me pleasure to express my grateful appreciation of assistance received in connection
with this report. Those who contributed to the map are named above. Much of the success
of the Biological Survey expedition was due to the faithful and efficient services of Maj. E. A.
Goldman, whose photographs illustrate the report.. Mr. A. W. Anthony contributed valuable
notes made during his long experience in the peninsula. Mr. Joseph K. Hutchinson, Capt. H. W.
Cochran of the United States Army, Dr. J. W. Mixter, Mr. Dale Bumstead, Mr. Lincoln Ellsworth,
Mr. David Goldbaum, and. Mr. O. A, Pratt, of the Federal Horticultural Board, have supplied
notes and information which have been. invaluable. It isa pleasure, also, to mention the gener-
ous hospitality of Mr. W. J. Heney, resident manager of the Chartered Co. of Lower California,
which at the time of our visit. controlled the Magdalena Bay concession. We were entertained
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 2.

 

 

 

 

 

Fie. 1.—Cocopah Indian and brush hut on Hardy River in delta ef Colorado River.

 

 

 

Fia, 2.—San Matias Pass, altitude 3,200 feet, between Sierra Juarez and San Pedro Martir Mountains. The vegetation shown on
north side of pass is characterized by many viznaga or barrel cactuses ( Echinocactus) and small agayes.
Memoirs National Academy of Sciences XVI. First Memoir, PI. 3.

 

;

 

 

 

 

Fig. 1.—Valley of Santo Tomas, south of Ensenada. (Photograph from David Goldbaum.)

 

 

 

 

 

Fic. 2.—Valley of El Rosario, south of San Quintin, These are typical agricultural villages in northern Lower California. (Photograph
from David Goldbaum.)
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 13

by him during our work in the Magdalena Bay section and were given facilities which enabled
us to accomplish much that would have been impossible without his friendly assistance.

THE BIOLOGICAL SURVEY EXPEDITION.

During many years the Biological Survey had been making detailed studies of the distri-
bution of bird and mammal life and their environmental relations in the arid regions of the
southwestern United States. It finally became of considerable importance to have first-hand
information on the little-known adjacent and closely related peninsula of Lower California.
Early in 1905 it was decided that a Biological Survey reconnoissance of the peninsula should be
undertaken. I was placed in charge of the proposed expedition, with Mr. E. A. Goldman as
my assistant. The expedition began in March, 1905, and ended the 1st of February, 1906.
During this period we traversed the entire length of the peninsula and crossed it eight times from
coast to coast. With the exception of a few subsidiary boat trips in the delta of the Colorado
and to the coastal islands, the work was all done with horses and pack animals and included more
than 2,000 miles of travel in the saddle.

In order that the delta of the Colorado might be covered before the main work was begun,
Mr. Goldman joined Dr. D. T. MacDougal, botanist of the Carnegie Institution of Washington,
on a boat expedition to the delta region. ‘They left Yuma, Ariz., March 21; descended the
river and made a brief stay at Colonia Lerdo, on the Sonora side; crossed the Colorado and
camped at the mouth of the Hardy River, in Lower California. The spring of 1905 was marked
by unusually high water in the Colorado, and much of the delta country was overflowed, espe-
cially the Hardy River district. (See Pl. 23, Fig. 2.) The party worked its way up the flood
plain of the Hardy, and from April 5 to 15 camped at the east base of the Cocopah Mountains,
exploring for plant and animal life the adjoining desert plain and the Cocopahs to their sum-
mit. They then ascended the Hardy to Volcano Lake and camped on New River. Many
curious small mud cones built up by escaping gases and called mud volcanoes are located near
Volcano Lake. (See Pl. 21, Fig. 2.) On April 21 the party left the boat and proceeded by
wagon to Calexico, just within the border of California, where this section of the work ended.

The middle of May I met Mr. Goldman at San Diego, and on May 14 we proceeded by gaso-
line launch to study the bird life on the Coronados Islands, a group of rocky islets lying about 7
miles off the extreme northwestern coast of Lower California. We camped one night on the larger
island, and after visiting the other islets and securing some interesting specimens of birds and
notes on the species breeding there returned to San Diego, May 15. On the way back we were
much interested, when a few miles off the mouth of San Diego Harbor, to sce a number of
Sabine gulls scattered about feeding on the tide lines formed by the currents in the smooth sea.
This was, I bélieve, the first record of these beautiful birds at San Diego.

On the evening of May 16 we left San Diego by steamer and arrived the following morning
at Ensenada, the most northerly port and the capital of the northern district of the peninsula,
about 65 miles from San Diego. Ensenada, which contained only about 1,000 people at this time,
is situated in a small valley at the head of Todos Santos Bay and was formerly much more pros-
perous than at the time of our visit. To the northeast the town is backed by abruptly rising,
high, grass-grown hills, the steep slopes of the coast range rising farther to the east at the head
of the small valley extending from the shore of the bay a short distance inland.

At the time of our visit Col. Celso Vega was governor of the northern district. He received
me courteously and assisted in furthering the objects of our expedition. Ensenada was also
the headquarters for the Lower California Development Co., an English corporation possess-
ing an enormous land grant in the northern part of the peninsula, which they had purchased
from the former holders. Some years before our visit this company had colonized the land about
Ensenada and at San Quintin on the coast farther south for farming and fruit growing. Many.
people were attracted there from the United States and elsewhere, but the colony proved a
failure, and nearly all of its members had left, many having lost all they possessed.
14 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

We remained some time at Ensenada studying the native bird and animal life in the
country surrounding the town, and outfitted for our long overland trip down the peninsula.
From May 20 to 22 we visited the Todos Santos Islands, at the mouth of the harbor. This trip
was especially notable through the discovery that an extensive migration of Pacific loons was
taking place several miles offshore. These birds were flying northward in a broad belt passing
about a mile inside of the islands. The weather was pleasant and they were winging their way
in a continuous, procession, singly, or sometimes two or three in a group, often spaced by intervals
of several hundred yards, but during the hours in which I observed them there was no definite
break in the flow of birds. The total number which passed during the two days in which they
were observed must have been very great. The flight of these birds northward along this coast
in such numbers was especially interesting in view of the fact that theretofore they had rarely
been recorded from anywhere in the peninsula.

On account of the sparsely populated and desert character of Lower California, the poorly
marked trails, and the few and widely scattered watering places, it was important for us to secure
a good outfit for our expedition. Unfortunately, at the time of our visit Ensenada was almost
depopulated, the most energetic part of the inhabitants having gone with their teams to the
Imperial Valley, Calif., attracted by the abundance of work and high wages on irrigation enter-
prises. Thus it was impossible to obtain satisfactory. men or good horses and mules. After
much delay we succeeded in employing two camp men who claimed to know all trails throughout
the northern half of the peninsula. Four very ordinary saddle horses and four small pack
mules were secured. We were assured by several residents of Ensenada that it was useless to
try to journey down the peninsula with saddle horses, for they would soon die, since only good
mules could stand the necessary hardships of such a trip. Our experience disproved this, for
three out of the four small saddle horses made the entire trip. The poorest horse became
exhausted and was traded foy another about midway on the journey.

It is true, however, that we found conditions exceedingly hard for our stock, and it was
only by exercising the most judicious care in stopping at frequent intervals where the scanty
forage appeared most favorable, in order that the animals might rest and recuperate, that the
trip was practicable. All of our animals were forced to depend on such forage as they could
find, since the sparsely inhabited character of the country rendered it impossible to supply
them with grain or other cultivated feed of any kind, except to a small extent at long intervals.

The men we hired as guides proved less worthy than the animals, for after a few days’ travel
they were forced to admit they did not know the trails or water holes, No other-men were avail-
able, so we were compelled during a large part of the journey across hundreds of miles of desert
to depend for guidance on our own skill in following dim trails, and on the information con-
cerning the few water holes which was obtained from the occasional people encountered. Fortune
favored us, however--we found water at all of the watering places we visited throughout the
journey, and thus escaped any serious discomfort from thirst.

South of Ensenada three trails were early established by the Spanish missionaries and are
still in use. One follows near the Gulf coast and is known as El Camino del Golfo; another,
down the west coast, is FE] Camino del Pacifico; and the third, down the mountainous "backbone
of the interior, is called El Camino de en Medio. The last, although rougher than the others,
has more frequent waterholes and better grazing, and was the one mainly followed by us,
though we crossed back and forth to the others.

On May 24 and 25 Goldman made a successful side trip from Ensenada 28 miles down the
coast to Santo Tomas to obtain topotypes of a pocket gopher (Thomomys aphrastus)' de-
scribed a few years before from the locality.

On May 31 we left Ensenada with our pack train on the journey which ended about nine
months later at the southern end of the peninsula. A well-worn wagon road led us easterly
through a little valley, up a gentle slope, and thence into a canyon cutting the western face of
the low, abruptly rising range which borders the coast in this section of the.peninsula. Our

 

1 Elliott, Field Columb. Mus., Zool. Ser., III, 219, June 1903 [= Thomomys bottae nigricans Rhoads).
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 15

first camp was at about 1,800 feet altitude, near the upper end of the canyon, among picturesque
live oak trees with short massive trunks, and huge wide-spreading branches, some of which
were nearly as large as the main trunk.

June 1 we ascended the canyon about 2 miles, and, crossing a divide at about 2,800 feet
altitude, came out into San Rafael Valley. This is a pretty, elevated valley, 18 or 20 miles
long in a northwesterly and southeasterly course, and from 3 to 10 miles wide. It forms the
upper part of the drainage of the Arroyo de San Rafael, which at the lower end of the valley
passes the old Spanish settlement Real del Castillo, and on its way to the coast north of
Ensenada passes the old Guadalupe Mission. The valley forms an open plain bordered on the
west by the coast range and on the east by the higher Sierra Juarez, or Laguna Hanson Moun-
tains. Our road led a few miles up the valley, by the Ojos Negros cattle ranch, and then easterly
about 9 miles to the little Indian village of La Huerta, where we camped at an altitude of 2,800
feet, immediately below the Indian houses. Here a small stream flows through the broad
lower end of a canyon in the foothills of the Sierra Juarez. A dozen or more Indian families
occupy a few huts scattered among small fields and orchards, where they grow peaches, apri-
cots, corn, beans, and other garden vegetables.

The western slope of the coast mountains facing Ensenada has Upper Sonoran vegetation
similar to that of the same slope near San Diego, but in the southern end of San Rafael Valley
there is a mixture of mesquites, yuccas, and other Lower Sonoran species. The bottom of
the valley is covered with grass and small scattered shrubs, the latter increasing in size and
abundance in the bordering foothills on both sides. All the way from Ensenada to La Huerta
the rocks are mainly of granite, and the east side of San Rafael Valley is covered with loose

granitic gravel. We saw many signs of sheep along the way and learned that large numbers
were being grazed in the rough country on the west side of the peninsula from the northern
border south to San Quintin. Most of the shepherds employed were Basques from the south
of France.

June 3 we left the wagon road at La Huerta and followed a winding trail for about 6 miles
easterly to La Republica, an abandoned gold mine and stamp mill; then for about 12 miles
northerly to a small meadow in an open ravine near the top of the range at 4,800-feet, where we
camped. The entire west slope of the Sierra Juarez range in this section is broken only by long,
rounded, undulating ridges, rising so gently that for long distances we appeared to be traversing
a slightly broken plain, covered with low brush and scattered rocks. Our camp was in a small
group of stunted live oaks about a mile below the El Rayo ranch on top of the mountain. The
slopes all about camp were covered with granitic sand and gravel, with large rounded granite
rocks projecting here and there. Over these slopes a thin growth of small shrubs and widely
scattered scrubby pines gave the landscape a poverty-stricken appearance. The morning of
June 4 a sharp frost coated with white the vegetation in the bottom of the meadow by our camp.

June 5 we moved about 8 miles northeasterly, passing El Rayo ranch, and made camp at
an altitude of 5,400 feet on top of the mountain near Laguna Hanson. El Rayo ranch was
formerly the Hanson ranch occupied by an American of this name with his cattle. He was
murdered near here, and the place is now one of the ranches of an American company with head-
quarters at Ojos Negros, in San Rafael Valley. The Laguna Hanson still bears the name of the
original owner, and his name is often, but erroneously, applied to the mountains on which his
ranch was located. It may be of passing interest to note that the only disagreeable and
unfriendly attitude toward our work met during our entire expedition was at this American
ranch.

At El Rayo ranch the trail entered a good wagon road which follows the top of the range
from Campo, just within the California border. The summit of the Sierra Juarez is a rolling
mesa, sloping to the west with low, usually bare, granite ridges and gigantic piles of huge rounded
rocks. These rocks were often too huge and smoothly rounded to be climbed by a man. They
were evidently the results of rapid decomposition of the granite. Laguna Hanson is a
shallow pond about half a mile across, situated in a small mountain park and surrounded by
low growth of Jeffrey pine. Two smaller ponds are located near the larger one.

158961°—21——2
16 MEMOIRS NATIONAL ACADEMY OF SCIENCES. . [Vou. XVI

June 9 we left Laguna Hanson and followed the wagon road southerly about 14 miles
along the top of the mountains, repassing E] Rayo ranch and reaching an abandoned sawmill
on a small stream in a pretty little valley. Here we lost several hours trying to find the trail,
which continues southward along the mountain, but the many old log roads confused us, so we
abandoned the idea of keeping this course and turned down the west slope, following the wagon
road 8 miles, and camped at an altitude of about 4,200 feet by a little stream in the lower end
of a small canyon. This was just above the village of ome de Cristo, at the southeastern
border of the San Rafael Valley.

June 10 we traveled southeasterly about 22 miles, crossing the southern end of San Rafael
Valley and camping at 3,700 feet altitude by a small spring and marshy meadow near the base
of a hill called Cerro Colorado. This was one of the regular camping places for teamsters on the
main wagon road for freighting supplies from Ensenada to the American mining camp at Alamo,
and as a result grazing for stock was scarce in its. vicinity. ‘Cerro Colorado lies at the head of
the drainage flowing into San Rafael Valley and isin the midst of the brush-grown hill country
which limits the southern end of this valley.

June 11 we continued about 15 miles. ieee along. fhe wagon road ‘isa rolling
hill country and, crossing a divide at about 4,500 feet, descended to Alamo, on the west end of
an open plain, at an altitude of about 3,400 feet, close to the base of some low, abrupt, and
barren rocky ridges. From Alamo the plain, which has a width of from 6 to 15 miles, extends
easterly about’20 miles to the west base of the southern end of the Sierra Juarez. It is bordered
on the northern side by series of low hills and on the southern side by low rounded ridges and
mesas, which merge into the hills bordering Trinidad Valley. Most of plain and adjoining
hills are covered with a fairly abundant growth of junipers with some pinyon pines, but most
of the larger growth for a dozen miles around has been cut for the needs of the mines at Alamo
and the available supply appeared to be approaching exhaustion. —

Alamo, with its vacant houses and dilapidated appearance, was a typical = mining
camp. There were many signs of former activity here on a-considerable scale, but at this time
only a few men were working. New supplies were being sent in aud a revival of work was be:
ing announced with the usual sanguine expectations.

At Alamo we replenished our supplies and then traveled about 6 miles easterly, along one
of the several wood ‘roads which radiate from the camp, to La Posa, a muddy water hole on
the plain, where we camped. The water hole was much frequented by cattle, and the stagnant
water was very bad. The morning of June 12 we continued easterly about 10 miles, crossing
numerous low, rounded ridges of granite bowlders, to a small deserted house by a spring known
as Rancho Viejo or Los Batequis, where we camped. for a few days. This place is located at
an altitude of 3,500 feet on a small knoll among the pinyons and junipers.

Directly across the open plain from Rancho Viejo, close to the base of the Sierra Juarez,
we saw smoke rising daily from the fires of Indians living at the old Santa Catarina Mission,
but they did not come near us.. A band of wild horses was ranging about Rancho Viejo and
gave us much trouble by coming in at night and leading off our stock. A single night among
the wild horses gave our broken stock such a feral state of mind that only their hobbles pre-
vented their leaving us forever, and when finally rounded up in the morning they would gaze at
us wild eyed, and ‘snort and tremble when first touched. Once bridled and saddled ae
lack-luster eyes and plodding gait showed the force of habit over instinct.

June 16 we traveled southeast from Rancho Viejo along a good wagon road over the soil
ing and gradually rising plain covered with pinyons and junipers, for about 16 miles to the
top of the northern rim of Trinidad Valley, at an altitude of about 3,800 feet. From the rim
a dug road descended the slope abruptly to the bottom of the valley, about 1,000 feet below.
There the road forked and we turned 'to the right about 4 miles down the valley and camped
at Campbell’s cattle ranch at 2,600 feet altitude. This ranch is near the lower end of the valley
‘and was our headquarters while we were working this locality. Trinidad Valley i is a basin
about 18 miles long east and west and about 3 miles broad in the middle, narrowing gradually
toward each end: : Itis open grassland in the lower part, with bushes on the sides and much cactus,
Memoirs National Academy of Scionces XVI. First Memoir, Pl. 4.

 

 

 

 

 

 

View looking north across San Felipe Bay. Good water is found a few feet below the surface on a flat back of shore. (Photograph from
David Goldbaum.)
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 5.

 

BELA

 

 

 

Fig. 1.—Stream at mouth of La Providencia Canyon, east base of San Pedro Martir Mountains. Note huge granite bowlders
brought out of canyon by flood waters and in places appearing like a glacial moraine. (Photograph by Dale Bumstead.)

 

 

 

 

 

Fic. 2.—Copal tree (Elaphrium macdougalii) near San Felipe Bay. The northern representative of a group of Arid Tropical trees
common in western Mexico.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 17

yucca, agave, and other desert growth for miles of its upper end and in the pass. A small
stream flows down a few miles of its lower end and drains through a rocky canyon leading to
the west coast.

On June 17 we left our extra supplies at Campbell’s and started for San Felipe Bay, on
the Gulf of California. We traveled 16 miles easterly up the valley to San Matias Pass, follow-
ing a continuation of the wagon road by which we entered the valley. The pass is an open
gap about one-fourth of a mile wide with a nearly flat bottom and gradual descent each way.
(See Pl. 2, Fig. 2.) It separates the southern end of the Sierra Juarez from the northern end
of the San Pedro Martir Mountains. Just before entering the narrowest part of the pass we
turned to the right and followed a dry wash up a little side valley and then through a rocky
canyon in the west side of the northern point of the San Pedro Martir Mountains for about 3
miles in all to San Matias Spring, near the head of the canyon. The spring, the only water
near the pass, gives a small trickling stream which runs down the canyon only a short distance
before it sinks into the sand.

June 18 we crossed a low ridge of the mountain just above and to the east of the spring and
entered the head of Windy Canyon, which winds down the east slope about 8 miles to the
sloping desert plain on the east base of the San Pedro Martir Mountains. The entire bottom
of Windy Canyon, which has high rocky walls, is so filled with a wash of granitic sand and
gravel from the disintegrating sides that it presents a smooth, easily traveled, gently descending
slope. The canyon is dry and very barren throughout its course, but on the desert at its mouth
we found the basal plain covered with a vigorous growth of cactus, yucca, mesquite, ironwood,
covillea, paloverde, fouquieria, and other desert plants. The trail turns abruptly to the right
a short distance out on the plain and skirts the base of the mountain for about 12 miles to the
mouth of Esperanza Canyon, crossing slopes made up wholly of loose disintegrated granite.

The east front of the San Pedro Martir Mountains rises abruptly in a tremendous escarp-
ment of bare cliffs and rocky slopes, 4,000 to 8,000 feet above the basal plain, with several
canyons cutting huge, ragged gashes into its face. In cool weather small streams flow down
the bottom of the larger canyons, sometimes running a mile or so out on the plain before sink-
ing into the sand, but in dry hot weather commonly retreating a mile or more up the canyons.
Extending out on the plain from the mouth of Esperanza Canyon is a great alluvial fan, or delta,
built up of disintegrated granite with a ridge of bowlders along its middle. These bowlders
vary from 1 to more than 6 feet in diameter and extend a mile down the slope, showing the
enormous force of the occasional floods which pour down the canyon. At the time of our
visit a small stream of clear, pure water flowed down the wash and disappeared in the sand
just, outside the mouth of the canyon. In the morning this stream flowed several hundred
yards farther than in the evening, retreating during the day, owing to evaporation caused by
thesun. Swarms of mourning and white-winged doves and some other birds, as well as a small
band of half wild horses ranging on the adjacent desert, were watering here, but we saw no
tracks of antelope or other game animals.

June 19 we continued southerly along the sloping basal plain about 12 miles to an alti-
tude of 2,000 feet at the mouth of La Providencia Canyon, which also had a small clear stream
of pure cold water, clear as crystal, flowing to its mouth. (See Pl. 5, Fig. 1.) La Providen-
cia is another high-walled canyon scoring the east face of the range. It is located just south
of Cerro de la Encantada, the highest peak in the range, at the most outstanding point on the
desert front of the mountains. Halfway between Esperanza and La Providencia Canyons we
passed the mouth of Canyon Diablo, the wildest and most rugged canyon in this front of the
range. It heads on the west side of Cerro de la Providencia, and descends first northerly
and then easterly, thus cutting around the west and north sides of the mountain. From the
mouth of La Providencia Canyon there extends for more than a mile down the middle of its
great alluvial delta another gigantic stream of flood-piled bowlders, which vary in diameter
up to 6 or 8 feet. This great bed of bowlders renders access to the water in the mouth of
the canyon difficult except by following closely the intricate course taken by the few half wild
18 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

horses and cattle which come here to drink. This bowlder stream has an interesting resem-
blance to a low glacial moraine.

San Felipe Bay, our destination on the Gulf of California, lies about 30 miles easterly
from La Providencia Canyon. San Felipe Valley, about 10 miles across, lies along the eastern
base of the San Pedro Martir Range, bordered on its eastern side by the Sierra San
Felipe, a low barren range parallel to the gulf coast, and reaching an altitude of about
4,000 feet. Between the southern end of this range and the Sierra Santa Rosa, nearly
opposite but a little southerly from La Providencia Canyon, is Buena Vista Pass, through
which the wagon road extends to San Felipe Bay on the gulf. The weather had become
exceedingly hot, and in order to traverse the desert between La Providencia and San Felipe
Bay with as little discomfort to our stock as possible, we left camp at 3.30 a. m., June 20, and
made our way slowly through the darkness down the slope easterly to the bottom of the valley,
which has an altitude of 1,400 feet, 600 feet below the mouth of the canyon. The bottom of
the valley where we crossed is covered with a broad belt of sand dunes capped with mesquite
thickets, and with the intervening open spaces so thickly filled with burrows of large kangaroo
rats (Dipodomys deserti) that it was difficult for our animals to struggle through. As the sun
came up over the range in front we were much relieved to reach firm ground again. Near
the base of the low mountains on the east side of the valley we found traces of an old wagon
road laid out many years ago from Ensenada by way of Trinidad Valley and San Matias Pass
to San Felipe Bay, when it was planned to colonize the shores of the bay and to land there the
supplies for Ensenada and other parts of northern Lower California, which were to be transported
overland by wagon: Our course lay directly across the valley until we reached the old road
along the west base of the Sierra de San Felipe and then turned southerly among low scat-
tered foothills. After passing through these we turned again easterly at an altitude of 1,600
feet and went down the dry sandy bottom of the broad arroyo leading from the valley through
Buena Vista Pass, which forms a deep gap between the mountains and leads directly toward
the gulf.

We soon reached the east side of the mountains at a point called Buena Vista, whence
the broad expanse of the gulf could be seen. Buena Vista has an altitude of 600 feet on the
western border of the gently sloping desert plain which extends thence to the gulf, a distance of
about 15 miles. The surface is here broken by scattered, barren hills and low mountains. The
day was extremely hot and our animals were much exhausted when, late in the afternoon,
we rode through a gap among some hills and came in sight of the narrow line of green mes-
quites in the flat. behind the sand dunes on the beach at San Felipe Bay. We saw no sign
that a house had ever been built here, but a strong signal post stood on the beach just in front
of two water holes, 3 or 4 feet deep, dug in the clay flat near the mesquites. No recent trace
of human visitors was visible, but evidently numbers of coyotes and turkey buzzards visited
the water holes regularly; and the’ fluid was so foul in consequence that even our excessively
thirsty animals refused to drink it. After some hard work the holes were emptied and the
vile-smelling mud on the bottoms and sides cleaned out. The new water which quickly seeped
in was usable, although offensive in odor and taste. We made camp a short distance from the
water under a fine, wide-spreading mesquite tree. Directly offshore, in front of the bay, Consag
Rock stands out like a great sail, its high cliffs whitened by the droppings of sea fowl. San
Felipe Bay, several miles in width, is not deeply indented, and on the west and south sides is
low and bordered by a narrow belt of sand dunes. The north side is limited by a low, isolated,
rugged mountain of uniform black color, known as San Felipe Point, which projects as a bold
headland into the gulf. (See Pl. 4.) On the bay side of this mountain is a little rounded tidal
lagoon, which fills through a narrow channel at high tide. A low wall of stone had been built
nearly across the entrance, and we learned that this lagoon has served as a natural fish trap
since prehistoric times. When, at unusually high tides, the lagoon is full of water and contains
many fish the openings in the wall are closed across the entrance channel, and as the tide falls
the water runs out between the stones leaving the fish stranded. During our visit the tides
were low, although rising and falling 15 or 20 feet, and most of the time the moist bottom of
No. 11, LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 19

the lagoon swarmed with fiddler crabs (Uca crenulata). The black rocks of the headland within
a few feet of the water line were covered with black isopods (Lygida exotica), which appeared to
thrive in the full glare of the sun on the rocks when they were uncomfortably hot to touch.
The water of the gulf along shore at this time was surprisingly warm, and many small sharks
and sting-rays frequented the shallows along the beach. After 8 o’clock the heat was intense,
but about 11 o’clock a sea breeze began to blow from the gulf and the rest of the day was more
comfortable.

From San Felipe a rarely used trail leads southerly along the coast, passing several mining
prospects, to Miramar, a mining camp which was in operation i in 1905 near El Tule Arroyo, on
the gulf slope, almost east of the Onyx mine. Water is very scanty and occurs only at wide
intervals along this trail, and prospectors are reported to have died of thirst while traversing it.
Another trail leads northerly up the coast and passes through a large arroyo separating the
Sierra de San Felipe and Sierra Pinta to Poso Salado and Tres Posos, to the west side of Laguna
Maquata and on to Mexicali.

The plains from the base of the San Pedro Martir Mountains to San Felipe Bay have a
characteristic desert flora, including such species as Pachycereus pringlei, several genera and
species of other cactuses, Fouquieria splendens, Prosopis glandulosa, and Covillea glutinosa.
In addition, about 3 miles inland, near the north base of the high Sierra Santa Rosa, which rises
abruptly west of the bay, there was a narrow belt of small broad-topped trees which in foliage,
color of bark, and general superficial appearance curiously resemble an irregular growth of
short-trunked apple trees. They are Elaphrium macdougalit, one of the most northerly represent-
atives of the copals, a group of wide distribution farther south in Lower California and on the
tropical mainland of Mexico.

Day by day the heat grew more intense at San Felipe and, in order to make the return trip
to the mountains as easy as possible for our stock, we left the bay at 2 p. m., June 25, and traveled :
back along our old trail about 15 miles to Buena Vista, where we made a dry camp on the clean
white sand of the wash which we had followed on our way to the gulf. Then by moonlight, at
1a. m., June 26, we retraced our way up the wash through the mountains and across the valley
to La Providencia Canyon. During the early morning hours the air was so cool, as was often
the case on the desert in summer, that we were glad to wear our sweaters, but as we entered the
mouth of the canyon, the sun came up over the San Felipe Mountains and its blazing heat at
once foretold another excessively hot day, with the temperature far above 100° F. in the shade.
Owing to the exhausted condition of our stock from the scanty feed and the exertion of crossing
the hot desert, we remained in this canyon for a day while the animals recuperated.

June 27 we started back to Trinidad Valley, camping again at Esperanza Canyon and San
Matias Spring, and reaching Campbell’s ranch June 29. During the 10 hot days between our
first sight of Esperanza and La Providencia Canyons and our return from the gulf, the small
streams which had reached their mouths had notably decreased and now sank into the sand
much farther up the canyons. In midsummer the lower end of the water must be a mile or
more above the mouths. For some days after our return to Trinidad Valley we were occupied
in studying the birds and mammals living there. Several dens of the desert fox (Vulpes macrotis)
were found in the bottom of the valley and some specimens secured. On July 3, I left Goldman
to complete the work in the valley, and with Campbell made a rapid trip in a light wagon
over a fairly good road to Ensenada to ship our specimens and renew supplies.

We reached Ensenada July 5 and found the schooner Academy, from San Franciso, in port
on its way to the Galapagos Islands with a scientific expedition from the California Academy of
Sciences. The expedition was in command of Mr. R. H. Beck, and the party included several
enthusiastic young naturalists who made fine collections in many branches of zoology, and
in plants wherever they landed.

July 8 I was back in Trinidad Valley and found that, according to our plans, Goldman had
started with the outfit July 4 for the northwestern base of the San Pedro Martir Mountains,
where he was to remain until I overtook him.
20 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

July 9 I left Campbell’s ranch alone on horseback and followed Goldman’s trail to the south-
east across Trinidad Valley, and then through rolling and rocky hills overgrown with scattered
pinons and junipers. About 8 miles from Campbell’s I passed an Indian rancheria called Arroyo
de Leon, where asmall group of huts was located. Goldman had stopped here for his midday rest
and found a Mexican camped close by who pretended to be a trader, but was in reality robbing
the Indians. This man was reported to make frequent trips among the few Indians of this
district ostensibly to purchase beeswax for cash, or for a few trade goods he carried, but his real
purpose was to gamble with the inexperienced Indians until he had won their money and as
many horses as possible. When Goldman passed he had already won four horses. This shows
that the pitiful remnants of the tribes which once populated the peninsula are still being preyed
upon by their so-called civilized superiors. The head man among the Indians here and in
other Indian villages in Lower California is called El Capitan, a name formerly in common use
among the Mission Indians of southern California.

Continuing through the same rolling hills about 7 miles southeasterly from Arroyo de Leon
and 15 miles from Campbell’s, at an altitude of 3,600 feet, I came to a spring and two deserted
houses known as Poso Luciano, where I camped for the night. July 10 I followed Goldman’s
trailsoutheasterly across low brushy hills for about 12 miles to the mouth of a canyon in the basal
foothills of the San Pedro Martir Mountains, at an altitude of 5,300 feet. Here he was camped
by a small stream which flows through the flat at the mouth of the canyon and sinks into the
sand just below. This place is known as El Pifion, and the flat and adjacent hillsides are covered
with pinyon trees, making it a favorite resort for Indians during the harvest of pinyon nuts.
An Indian village was once located here, but water became scarce and it was abandoned
years ago.

July 11 we left El Pifion and traversed a series of low hills and ridges covered with a vigor-
ous growth of large bushes and scattered pinyon trees flanking the basal slopes of the main
range. Our course was southerly, mainly with no sign of a trail, and travel was slow on account
of the dense growth of rigid bushes, among which it was difficult to force a passage. We finally
reached the bank of a small swift stream of clear water flowing along the base of the main slope
of the mountains at an altitude of 4,800 feet, where we camped. The stream, probably the
headwaters of the Arroyo de San Rafael, was densely bordered by a vigorous growth of willows,
and was the largest we had seen on the peninsula. Here our animals fed eagerly on the juicy
grasses along the water’s edge, and we found the fresh green of the willow thicket and the sound
of water rushing over its rocky bed extremely refreshing after our weeks on the sun-scorched
deserts.

The morning of July 12 we climbed without any trail the steep open slope of the San Pedro
Martir Mountains directly above camp, and when well up near the top of the main ridge came
unexpectedly into a well-made trail along the bank of a large ditch, which wound around
the mountain side to the Socorro placer mines a few miles to the south. A trail soon left
the ditch and led up an extremely steep rocky slope to the top of the mountain. Thence
we crossed several rounded ridges covered with an open growth of yellow pines, and about 10
miles from our camp, at about 7,200 feet altitude, found two log cabins and a corral built on
the banks of a small stream in a flat near the upper end of a small canyon. Here patches of
aspens were growing among the pines on north slopes, and grass and other herbage on the open
flats. This was Vallecitos ranch, a summer cattle ranch unoccupied at the time of our wisit.
We arrived in the midst of the first storm of the summer or rainy season, a hard thundershower
of rain and hail, which continued for more than an hour.

The same afternoon I rode over the surrounding country and about a mile and a half south-
east of the ranch found the Vallecitos or ‘‘ Little Valleys,’’ the locality where we wished to locate
for our first work on the mountain. The Vallecitos are a series of beautiful flat-bottomed
grassy parks at an altitude of about 8,000 feet, with gently sloping borders overgrown by an
open forest of yellow pines (Pinus jeffreyt) wixed with a few other conifers. (See PI. 6.)

July 12 we moved camp to the lower end of the Vallecitos, where in the arroyo a little water
comes to the surface. Two brothers named Gay and a friend from Riverside, Calif., had been
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 6.

 

 

 

 

 

Fic. 1.—North Slope near Vallecitos, about 8,000 feet, San Pedro Martir Mountains. The view shows aspens (Populus iremuloides),
Jeffrey pine (Pinus jeffreyi), and lodgepole pine (Pinus contorta).

 

 

 

 

 

Fic. 2.—View across top of the San Pedro Martir Mountains from an altitude of alout 8,500 fect near Vallecitos. The forest shown
. is mainly lodgepole pine.
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 7.

 

 

 

 

Fic. 1.—Meadow at La Grulla, San Pedro Martir Mountains (about 7,000 feet). Note the thin fringe of Jeffrey pines around the
border with bare granite ridges rising beyond.

 

 

 

 

 

Fic, 2.—Meadow at La Grulla with Jefrey pines and huge granite bowlders about the border.

 
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 21

camped here for some time, hunting deer and enjoying the beautiful surroundings. They
reported deer rather common, but were shooting only what they could utilize. They hunted
specimens for us a few days and then left on their way to the border. On the easterly side of
Vallecitos a long, high, narrow ridge forms the main crest of the mountain and reaches an altitude
of about 9,000 feet. On the southwesterly slope of this ridge several little springs and stream-
lets were found in the heads of small canyons leading down toward Vallecitos, but none flowed
down the sandy wash leading through the park. This high ridge forms the westerly wall of
Canyon Diablo, and from the crest above Vallecitos we had an impressive view down the rock-
walled gorge. Canyon Diablo heads southwesterly from the Cerro de la Providencia, the culmi-
nating peak of the range, and swings around the western and northern sides of the peak to the
open desert between Esperanza and La Providencia Canyons. In front of us to the east,
directly across the upper canyon, La Providencia Peak rose to an altitude of over 10,000 feet,
forming the sharp crest of a narrow cockscomb ridge of bare granite which, cut from the main
mass on two sides by Canyon Diablo, extends northerly as a part of the eastern wall of the
range, its spur-like character apparent only from the top of the range where we stood. The
view was uninterrupted out through the mouth of the canyon to the desert valleys 8,000 feet
below and beyond over the summits of the scattered desert ranges to the lower Colorado River
and across the northern end of the Gulf of California to the high mountain ranges far in the
interior of Sonora. The view over the low desert ranges was like looking down on a gigantic
relief map. Great alluvial fans or deltas extending out from the mouths of canyons were con-
spicuous everywhere, and several fine examples of mountain ranges 2,000 to 3,500 feet high
almost completely buried in their own disintegrated material were seen.

A few stunted pinyon trees were growing in crevices in the southern faces of Canyon Diablo
walls up to 9,000 feet altitude. Many mountain quail with coveys of young of various ages
were seen above Vallecitos on the west slope of this high ridge, and mule deer were numerous
in this locality.

- On July 18 we broke camp at Vallecitos and crossing the plateau southwesterly descended
aisteep and rocky trail about 12 miles to La Grulla (The Crane) meadows. There are three of
these meadows situated on a plateau more broken than that at Vallecitos and about 1,000 feet
lower, thus having an altitude of about 7,000 feet. La Grulla is on the southwestern slope of
the mountain and, like Vallecitos, is the type locality of several species of birds and mammals.
The main meadow at La Grulla is a narrow, winding, grassy valley several miles long, amid
a beautiful.open pine forest, with two small ponds in the bottom and gigantic granite bowlders
scattered in profusion, singly or in groups, along its sides, especially near the lower end. (See
Pl. 7.) To the north and east, high ridges of granite covered with bowlders of the same for-
mation rise almost barren of vegetation above the belt of pine trees which forms a girdle about
each of the parks, thus giving the ridges the appearance of projecting above timber line.
A small stream flowing out of the lower end of the main La Grulla meadow by which we
camped is said to form the extreme head of the Rio Santo Domingo. Near the lower pond in
the meadow were two log cabins and a corral, where some Mexicans have a summer dairy camp
and make cheese. They bring their cattle up from the foothills in spring, when the grass
becomes green, and keep them until cold weather begins.

La Grulla has been visited by practically all naturalists who have been in these mountains,
and between 1888 and 1893 A. W. Anthony was here several times. He found mallards and
cinnamon teal nesting about the ponds in May, but at the time of our visit the borders of the
ponds and all the valley bottom had been so closely grazed and trampled by cattle that no
shelter was left for waterfowl, and not a duck was seen. Deer were common and remarkably
unsuspicious near our camp. The trail from Vallecitos to La Grulla was so extremely steep
and rocky in places that it was difficult to bring down the loaded pack animals, and an ascent
would have been even more troublesome. A trail leading directly down the mountain from
near our camp, at the lower end of La Grulla meadow, was reported to have some very difficult

rocky slopes.
22 MEMOIRS NATIONAL ACADEMY OF SCIENCES. | [Vou. XVI

July 24 we completed our work at La Grulla and followed an old trail about. 10 miles south-
easterly along the mountain to Santa Rosa, another beautiful mountain park at about the same
altitude (7,000 feet) as La Grulla. It is a waterless grassy valley like a shallow basin in form,
about 2 by 3 miles in extent, and is bordered by low, pine-covered hills. This park occupies
almost the entire breadth of the top of the mountain range. Santa Rosa Canyon heads on the
northeastern side of the valley and leads thence directly down between high rocky walls to the
desert at the east base of the range, draining to the gulf. An old trail dating back to the days
of the early missionaries leads down this canyon and permanent water exists near its mouth.
Mountain sheep frequent its rocky slopes and California condors are said to nest in holes in
its high walls. The old trail from Santa Rosa Canyon crosses Santa Rosa Valley to the west
side and then descends along the course of a small canyon. We left Santa Rosa by this trail,
leading: down to the west, and about 2 miles below came to a little benchlike valley in the
canyon with a willow-bordered stream flowing along its middle. This is a southern tributary
of the Rio Santo Domingo. The little valley had a house and a cattle corral at its upper end,
known as the Rancho Santo Tomas, near which we camped several days at an altitude of about
6,000 feet. The owner has a small herd of cattle which he keeps here in summer. During our
stay the carcass of a donkey in Santa Rosa Valley attracted about a dozen California condors,
and we were able to secure two specimens. Others of these fine birds were seen, and they
appeared to be rather common in this part of the mountains.

July 28 we left Rancho Santo Tomas and descended the west slope of the mountain be a well-
worn trail. A few miles below the ranch we passed the ruins of the old San Pedro Martir Mission
on a sloping bench on the mountain side, where a small stream afforded water for irrigating some
small fields. All that now remains of the mission are the bases of adobe walls where the build-
ings stood, a little graveyard near, and the low banks of old irrigating ditches. About 16 miles
below Santo Tomas we came to the bottom of the main-slope of the mountain, where the stream-
let we left above had increased to a swiftly flowing creek bordered by cottonwoods.’ From this
point the trail turned northwesterly and after crossing several foothill ridges brought us to San
Antonio ranch at an altitude of 2,000 feet on a main fork of the Santo Domingo River, where it
emerges from its canyon in the west side of the main range. The ranch consists of two small
houses occupied by a Mexican family. The stream at San Antonio at the time of our visit was
10 or 15 yards across and averaged less than afoot deep. Willows, cottonwoods, and small oaks
fringe the water courses here. A species of rainbow trout which sometimes affords good fishing
inhabits this stream from the ranch far up into the canyon. Our specimens proved this to be
an undescribed species, which Dr. B. W. Evermann has named Salmo nelsoni. This is said to be
the only stream containing trout in the peninsula, and is the only stream in Lower California
which is reputed to maintain its flow to the sea along its entire course throughout the year.
Another small stream from the mountain joins the San Antonio Creek a little below the ranch.
The people at San Antonio ranch reported that trout were much less numerous than usual in
1905, owing to tremendous floods which had swept down the canyon the year before, ' cmrying
so many bowlders that many trout were killed.

Rancho San Antonio is in a narrow valley with the main San Pedro Martir Range rising to
the east, while a high, narrow, outlying ridge attains an altitude of about 3,400 feet to the west
and southwest. This latter mountain cuts off the moist coast winds and results in giving this
valley a hot and much more arid climate than is usual west of this part of the main range.

On August 1 we left San Antonio for San Quintin on the Pacific coast. The trail crossed the
valley and led directly up the high steep ridge to the west and then southwesterly down ridges
and west-draining slopes for about 26 miles to our camp by asmall stream at the upper end
of the high-walled Canon de Nueva York, at an altitude of about 800 feet.

August 2 we continued our southwesterly course down low ridges and over sloping mesas
west-draining for about 15 miles to the head of a broad, flat-bottomed, low-walled canyon, witha
small stream at its head, down which we traveled for 5 miles towhere it ended on the bare, nearly
level coastal plain of San Quintin. Many California quail were seen near the water in the
canyon.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 23

In the lower part of the canyon we passed a number of old fenced fields and small board
houses, and at its mouth several larger and more elaborate houses, water tanks, and wire fences,
all showing evidence of having been abandoned for some years. We learned later that these
were relics of an elaborate and costly but unsuccessful effort made by the Lower California
Development Co. to establish a large agricultural colony on the coastal plain adjacent to San
Quintin Bay. Five miles of travel from the mouth of the canyon across the nearly level plain
brought us to San Quintin, at the head of the bay on the Pacific coast. A mile above the town,
at the extreme head of the bay, was a large flour mill, built to grind the wheat the colonists were
to produce, and extending from the mill a short distance northward on the plain was a scrap of
railroad that was the beginning of a line planned to extend thence northward to southern Cali-
fornia, to carry north the flour from the grain. San Quintin was the local headquarters of the
colony, which extended for 10 miles or more up and down the coast. At the time of our visit
the town contained a few wooden buildings, one of which was a combined hotel and store, the
headquarters for occasional travelers and for the American and English miners engaged in
prospecting to the southward. This place was a port of entry with a custom house and a total
population of less than 50 people.

The coastal plain of San Quintin is from 2 to 6 miles broad and extends about 20 miles north
and 15 or 18 miles south of the town, gradually narrowing near the extremes as the foothills
approach the coast. The plain is rather sandy and overgrown with a small sparse vegetation
of herbs and small shrubs. The land on a large part of the plain appears to be good, and well-
directed dry farming methods might reclaim some of it. Some of the land is subirrigated, as was
evidenced by several small growing cornfields among the deserted ranches in the canyon through
which we entered the plain and the flourishing fig orchard we saw south of the San Simon River.
No doubt there is also much underground water that might be made available by pumping. In
addition reservoirs might be built in the foothills.

We worked for some days in the vicinity of San Quintin and packed and shipped our collec-
tions by the steamer St. , Denis, which made regular round trips to this port from San Diego via
Ensenada.

Early in the morning of August 12 we left San Quintin in the gasoline launch Todos Santos
and soon after noon landed on a pretty sand beach on the northeastern shore of San Martin
Island, north of the entrance to San Quintin Bay. On our way we noted that a narrow belt
occupying the coast a short distance in both directions from the mouth of San Quintin Bay is
volcanic and is dotted with numerous small volcanic hills. San Martin Island is a part of this
belt and is wholly volcanic, the surface being made up of ragged and broken lava beds. Itisa
favorite nesting place for brown pelicans and cormorants and is one of the guano islands of this
coast A few leopard seals and sea lions frequent its shores and the island was once a breeding
resort for the sea elephant. |

... The morning of August 14 we left San Martin and ran down the coast to San Geronimo
Island, ‘where we arrived about 3 p.m. This small island is characterized by several small
rounded hills. Gentle intervening slopes contained great numbers of deserted nesting burrows
of shearwaters in which lived swarms of white-footed mice (Peromyscus). se 15 in arun
of about eight. hours we returned to San Quintin.

~ Between August 18 and 23 Goldman made a trip about 55 miles north to San Telmo to
secure topotypes of some small mammals. From San Quintin his route followed the wagon
road, which extends north paralleling the coast to Ensenada and San Diego. About 25 miles
north he reached the village of Santo Domingo, near the mouth of the river of the same name,
and a mile or so below the ruins of the old Santo Domingo Mission. Several ranches are located
here, including one owned by an American, Mr. Young, who was growing wheat, corn, peaches.
grapes, melons, and other crops.

August 24 our recent collections were shipped by steamer to San Diego, and on August 25
we left San Quintin and traveled about 8 miles in a southeasterly course over the coast plain,
crossing the broad. dry bed of the San Simon River, near where it leaves the low foothills, and

‘ @amped at. the Santa Maria ranch, a short distance beyond, and about 3 miles from the sea.
24 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

This ranch belonged to Mr. Young, the American who has a place at Santo Domingo, and,
although without surface irrigation, contains an orchard of large black fig trees several acres in
extent. At the time of our visit the heavy crop of figs was being harvested as it ripened by
hundreds of ravens which came down daily from the foothills. Fairly good crops of hay and
grain are grown here, and the fig orchard and the eucalyptus trees bordering the fields show that
there is abundant water near the surface for subirrigation. A number of deserted frame houses
in this vicinity and an iron pipe line down the canyon of the San Simon River mark the failure
of the San Quintin colonization efforts here. The Santa Maria and one or two adjoining ranches
are the only occupied parts of the considerable area once colonized.’ Near the mouth of the
San Simon River are some marshy areas, and adjacent flats overgrown with salt weeds.
August 31 we traveled south about 8 miles along the coast wagon road to Socorro, where
there is a small fresh-water pond in the bottom of a wash which empties just below into a salt
lagoon winding through a marshy flat. This part of the coast is bordered by a series of sand
dunes backed by a long salt flat. The low rolling’ foothill plateau which backs the coast plain
gradually approaches the shore from the north and reduces the plain to a narrow point, ending
a little south of Socorro. Fogs are common along this part of the coast at this season.
September 1 we continued southward, the coast plain giving way to rolling hill country
cut by many deep dry washes leading down to the sea from the interior, some of them broad
enough to carry small rivers at flood time. About 10 miles south of Socorro the hills become
so rough and broken near the coast that the road turns to the left up a broad canyon, and after
crossing gently rolling hills and mesas for about 16 miles reaches the village of Rosario. This
is a small place containing about 20 houses on the south bank of the Rosario River, some 5
miles above its mouth. (See Pl. 3, Fig. 2.) It is an unusually prosperous place, owing to its
situation in a fertile little valley with a small permanent stream for irrigating fields in the level
river bottom. Corn, wheat, barley, beans, figs, peaches, almonds, and a few other crops are
produced here. The Rosario River rises in the dry river bed a short distance above the vil-
lage and is one of the very few streams of the peninsula which are said nearly to reach the
sea throughout the year. The surrounding hills, especially inland, are so barren and clayey
that they form a kind of ‘‘badlands.” ”
September 2. The wagon road ended at Rosario and a trail led in a general easterly course
from the village up a long dry arroyo among low, barren, rolling hills, and then across hilly
plateau country, which gradually rises until, about 25 miles inland, we reached a large shelter
cave in the rocks on the side of a canyon where we camped. This is a regular camping place
known as Las Cuevas and has an altitude of about 1,500 feet. Close by, in some deep potholes
‘in the bottom of the dry rocky canyon, are large tanks or tinajas of good water. A few miles
from Rosario the barren clay hills change to a more rocky formation with more abundant
desert vegetation. About 8 miles out on the trail we saw the first giant cactuses (Pachycereus
pringlet), and 18 miles out, on the top of a mesa, at about 2,000 feet altitude, appeared the first
of the singular, polelike tree (Idria columnaris) called cirio by the Mexicans. Both of these
plants became more common as we advanced. ~ vps
September 3 the trail led easterly from Las Cuevas through an elevated rolling and rocky
plain, broken by low rounded hills and isolated mountains with occasional small valleys. All
along the trail we were amid a plentiful growth of the strange cirio, giant cactus, ocotillo (Fou-
quieria splendens), and other desert plants—a striking contrast to the scanty vegetation on the
barren hills a few miles nearer the coast. (See Pl. 29, Fig. 1.) Fifteen miles easterly from
Las Cuevas we came to San Fernando, the site of one of the’ early missions, now marked by
the ruined walls of the old buildings, some old irrigating ditches, and alittle cluster of adobe
houses sheltering two or three poverty-stricken Mexican families. San Fernando is situated
at an altitude of about 1,500 feet on the side of a narrow valley marking the course of the
San Fernando River. The bed of the stream is dry throughout most of its course, except at
flood times, but at San Fernando some large springs furnish permanent water for irrigating
small fields. The lines of the irrigating canals built by the missionariés show that in their day
several hundred acres of land must have been irrigated. The old ditches were well made,
First Memoir, Pl. 8.

Memoirs National Academy of Sciences XVI.

 

 

 

 

Giant cactus (Pachyccreus pringlei) on road from Santo Domingo Landing to Calmalli. The trunk of this

 

giant of its kind was about 3 feet in diameter and the whole tree was 60 feet high.
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 9.

 

ee

 

 

 

Fic. 1.—The Llano de los Buenos Aires, a part of the great interior plateau south of the onyx mine, Note flat-topped hills showing former
surface level, sloping gently to the Pacific coast.

 

 

 

 

 

Fia. 2,—San Pablo Canyon below ruins of old mission. This canyon is cut in the
great lava-capped plateau, the basal sandstone showing in wall.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 25

extend along both sides of the valley, and in some places were built up 8 or 10 feet with
stones. More than a century ago this was a populous and well-cultivated little valley, where
cotton and many other crops were grown, but it has now returned almost to its primitive con-
dition, only a few acres being under tillage. From the size and length of the ditches water
must formerly have been much more abundant than now. We remained here two days col-
lecting among the willows about the water and on the adjacent arid hillsides. A number of
copper mines have been prospected and abandoned here.

September 6 we continued our journey, following a wagon road easterly from San Fer-
nando up the valley, which soon broadens to an open plain called the Llano de Buenos Aires or
Llano de San Augustin. After traveling about 20 miles we reached Poso San Augustin at an
altitude of 1,800 feet, where the wagon road we followed joins the main wagon road over which
onyx is hauled from the mine to Santa Catarina Landing on the Pacific coast for shipment to
California. A well had been dug at San Augustin to a depth of about 50 feet beside a sandy dry
wash in the open plain and contained a good supply of water.

September 7 we traveled 11 miles easterly across the plain on a good wagon road to the
onyx mines which are located at 2,500 feet altitude on the edge of the low foothills skirting the
mountains which rise abruptly a little farther eastward toward the Gulf coast. The onyx
mines, then being worked by a San Diego company, are near the upper end of El Tule Arroyo,
which leads easterly through the mountains to the gulf. The location of the mines was formerly
called El Tule but has been changed to Onyx. The camp consists of a small store and a few
small houses for employees built on the bare open slope. The onyx is in horizontal beds in
limestone near the surface.

September 8 we left the wagon road at Onyx and turned southward on a trail across a low
ridge and then over a gently rolling desert plain for 8 miles to a water hole known as Agua Dulce,
situated in the head of a rocky canyon which begins abruptly in the limestone bedrock of a
little valley on the open plain. An abruptly rising, barren range of mountain walls in the eastern
side of the Llano de los Buenos Aires, extends‘ parallel to the Gulf coast far to the south-
eastward beyond Agua Dulce. (See Pl. 9, Fig. 1.)

September 9 we traveled rapidly southeasterly on an old trail leading across a gently
undulating plateau of decomposed granitic soil, over which were scattered low ridges, great
piles of gigantic bowlders, or single bowlders of pale granite. This granite is the same in ap-
pearance as that on the Sierra Juarez and San Pedro Martir Mountains. The general altitude
of the plateau was about 2,000 feet, and to the east and west lie distant ranges of mountains.
The vegetation on this granitic plateau country was a wonderful combination of desert species,
among the most conspicuous of which were Idria columnaris, Fouquieria splendens, Pachycormus
discolor, Pachycereus pringlei, and several other cactuses, yuccas, agaves, the creosote bush,
and others.

Eighteen miles from Agua Dulce we came suddenly to the Arroyo de Catavifia, a beautiful
open, shallow canyon winding through the desert. (See Pl. 27.) Most of the broad sandy wash
in the bottom was dry, but thin patches of green grass and willows and numerous tall and grace-
ful fan palms were scattered along its course. These showed the presence of water close to the
surface, and where the trail crosses the canyon a small stream rises and flows on the surface a
short distance. At this point the wonderfully picturesque combination of fan palms, cirio, giant
cactuses, agaves, and yuccas made a landscape suggesting the strange plant life of some remote
geological period and is probably unequaled elsewhere in America. From here we continued in
a southeasterly course for about 12 miles to the Arroyo de Jaraguay.

As soon as we crossed Catavifia, the formation of the country became more broken and was
made up mainly of volcanic rock, though ridges of granite continued to outcrop here and there.
The Arroyo de Jaraguay, like that of Catavifia, occupies a low, open canyon winding across the
plain, but is larger, the broad sandy wash appearing like a white river bed. Many large groups
of fan palms were growing along the Arroyo. We reached Arroyo de Jaraguay at sunset and
searched the sandy bed for water without success until after dark, when we discovered it in two
holes, 2 or 3 feet deep, dug by coyotes. The water had been visited by coyotes and buzzards
26 MEMOIRS NATIONAL ACADEMY OF SCIENCES: [Von XVI

until it was very foul, but we were forced to use it. The drainage throughout the country
traversed from Onyx to this point is into the Pacific.

September 10 we continued about 9 miles southeasterly across the same plateau country
to a curious high isolated hill of huge granite bowlders measuring 2 to 10 feet in diameter, which
stands on an oval, nearly flat-bottomed little basin a couple of miles in diameter. This hill,
aptly named the Cerro Pedregoso, is a well-known landmark on this trail. Another trail,
apparently the main traveled route, came in from the west and joined ours close to Cerro Pedre-
goso. This is the direct trail from San Fernando by way of some small mining camps and con-
tinues south down the middle of the peninsula. About 4 miles beyond Cerro Pedregoso, along
the main trail which leads through a series of basinlike flats, we stopped at midday under a group
of mesquites. At this point a dim trail branched off to the left, and we followed it for about 8
miles over gently rolling hills to the crest of the slope, which at an elevation of 2,800 feet forms
the peninsular divide at this point. From the top of the divide an abrupt descent on the east
leads to a lower plain extending to the shore of the gulf, which was in plain view from the divide.
From near Jaraguay to the crest of this ridge, owing to a change in the geological formation, the
abundant vegetation noted along yesterday’s route gave way to a scanty herbage and few shrubs.

A good trail leads from the divide down the east slope to a small, strongly mineralized stream
flowing through a series of pools in the smooth rock bottom of a broad canyon. Thence the can-
yon narrowed and, after leaving it and crossing some low hills, we reached the barren valley in
the plain along which the San Francisquito wash finds its way. We camped on the east side of the
wash, about 8 miles from the top of the divide and at an altitude of 1,400 feet. Here a well,
some old arrastras, and many other signs marked the site of the long-deserted placer mining
camp of San Francisquito. An old wagon road led away from the arrastras toward the gulf,
whence supplies must have been brought. A sloping plain, broken irregularly by low hills and
mountains, probably all below 2,000 feet altitude, extends north and south from this place and
is known as the Llano de San Francisco, or Llano de Santa Maria. Toward the gulf the plain is
broken by low ranges of hills, while on the west rise the steep escarpments of the main divide we
had crossed.

Conditions were extremely arid at San Francisquito and the vegetation was much like that
about San Felipe Bay. The San Francisquito wells are strongly mineralized and the water is
very bad, but 6 miles southerly, in the upper end of a bare granite canyon, are small pools of
fine pure water. This is the source of supply for the King Richard mine mentioned below, and
a wagon road leads from near this water easterly across the plain to the mine. Two Mexican
families employed by the mine were living in brush huts at these water holes.

Practically throughout this expedition our stock was forced to depend for food on the
plants growing on the desert, and edible species were often extremely scarce. By the time we
reached San Francisquito the animals were so worn down we were obliged to rest several days.
It was extremely hot here, and in order to save the stock as much as possible we broke camp
September 13 at 4 a. m. in brilliant moonlight and traveled 10 miles easterly, following the
wagon road from the water canyon across the plain and down a low-walled canyon leading
toward the gulf, to the King Richard mine. This is a gold mine worked by an Englishman
named Dick Daggett. To our surprise the camp was deserted, every house was locked, and
there were no signs of recent occupation. This left us in a serious position, for we had less
than a pound of flour and no beans, rice, sugar, or bacon, and were many days’ travel from
any supply point. We had seen Daggett several weeks before at San Quintin, and he had
agreed to sell us provisions, if needed, and it became necessary to find him or face a trip of
several hundred miles to the next supply point, subsisting on such small game as we might ©
shoot. Several used trails radiated in various directions from the camp, one leading down the
canyon toward the gulf showing the most recent tracks. We followed this for about 6 miles,
through a series of small canyons and over dividing ridges in a broken granitic and slate hill
country and came out into the broad, gently sloping Calamahue Valley, through which a dry
rocky wash, like a broad river bed, wound its way. The trail led us thence down the valley
about 9 miles to the mouth of the wash, where it passes through a narrow gap between two
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 27

low ridges of hills and reaches the gulf at a point on the shore known as Calamahue Landing.
This is one of several places along the gulf coast where supplies ordered from Guaymas, Sonora,
are occasionally landed for mining companies. We were much relieved to find a group of
small brush huts by a shallow well at the base of the hill on the right of the gap, where Daggett
and his miners were camped. Daggett welcomed us hospitably and said that through a mis-
understanding his supplies failed to arrive on time from Guaymas and he and his men were
forced to close the mine, come down to this point, and subsist for more than a month solely
on sea turtle and fish caught in the gulf. Fortunately for us the night before our arrival the
boat had come in and the supplies were lying on the open beach at the mouth of the wash.
Daggett told me that he and his Indian miners had been living for two weeks on fish they
managed to catch along shore and wild honey from the rocky hillsides.

Daggett kindly sold us the necessary provisions, and we turned back up the valley about
3 miles and made adry camp. The day was extremely hot, and as we traveled up the smooth
sandy flat, just after sunset, we saw two ‘‘side-winder”’ rattlesnakes lying in the warm dust
of the trail. On our approach they slid diagonally sidewise and forward about 15 feet to one
side of the trail. There they stopped and sounded their low warning rattles as each animal
of our outfit passed.

There was scarcely a shred. of feed either at Daggett’s, or where we camped, and for the
salvation of our animals it was necessary to find pasturage. The moon was shining brilliantly
after midnight and, in order to reach a place where the stock could find feed and to travel as
little as possible in the intense heat of the sun, we resumed our journey at 2 a. m., September 14.
At 10.30 a. m., after traveling 19 miles, we reached the upper end of the valley, where it closes
into the mouth of a narrow canyon. Just inside the mouth of the canyon several springs and
pools feed a small permanent stream, which sinks a short distance below. The water in the
springs is charged with varying amounts of soda, iron, and other minerals, and most of it is
undrinkable. Large gas bubbles were rising in one spring about which deposits of soda had
built up a broad rim several feet high. On the southern side of the mouth of the canyon are
the ruined walls of the old Calamahue Mission. This was one of the most desolate and trying
camps we made and we marveled at the old missionary padres who would choose such a place
to build a mission.

All along its upper half Calamahue Valley is crossed from south to north by well-marked
old trails, made by mountain sheep, leading from the projecting points of the Calamahue
Mountains to opposite points of the lower range of hills on the north. In places the sheep must
have traveled several miles across the flat valley bottom. The small scanty vegetation and the
generally barren character of the hills and plain about San Francisquito, and also in Calamahue
Valley, and the bordering mountains closely resemble conditions prevailing near San Felipe
Bay and contrast strongly with the greater abundance of vegetation in most of the country
west of the peninsular divide. A scattered growth of small trees, either pinyon pines or live
oaks, was visible on the upper 500 or more feet of the northerly slopes of the Calamahue Moun-
tains, a few miles southeasterly from the mouth of the canyon. This is the first tree growth
we had noted on the mountains south of the San Pedro Martir Range and we should have
investigated it if our saddle and pack animals had not been too exhausted to permit delay
‘ owing to the scanty feed along this part of the route.

We remained two days at Calamahue to rest the animals. It was excessively hot and each
day a strong wind blew down the canyon which seemed to come directly from a furnace. Guns
or other iron objects left for a short time in the sun became too hot to handle. The intense
heat, high winds, and bad water at this place made it such an unpleasant camp that on Septem-
ber 16 we were glad to move on. The route leads up the canyon, which has high rocky walls
along most of its course. Water comes to the surface at several points along the bottom, and
5 miles above its mouth, near the head of the canyon, we came to the last water, a spring charged
with alkali and iron, but usable. Here we stopped during the middle of the day under the shade
of some mesquite bushes. On our way up the canyon we saw mountain sheep tracks at places
in the bottom and one fine old ram came out on the sky line of a high rocky point about 500
28 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

yards away and gazed down as we filed by. I should have been glad to stop at this promising
hunting ground, but our saddle and pack animals were becoming too weak to risk the delay.

The Calamahue Mountains, through which this canyon cuts its way, are rarely visited by
anyone and sheep are probably numerous in them. Birds were scarce from the plain at San
Francisquito to Calamahue, but there were many signs of small desert mammals, such as
Dipodomys, Perognathus, and Peromyscus, and jack rabbits were numerous in the upper end
of Calamahue Valley.

From the upper spring in Calamahue Canyon the trail led to the head of the gently rising
canyon and out on the eastern border of a part of the broad interior tableland known as the
Llano de Santa Ana. The surface of the great central tableland is depressed at the head of Cala-
mahue Canyon to an altitude of about 1,200 feet, thus forming one of the lowest passes between
the east and west coasts in the entire peninsula. The plain is broken only by low scattered
hills and ridges and is bordered along its eastern side by the Calamahue Mountains, through a
low gap of which we had just come, and far to the west and southwest by several low mountain
ridges, including the Sierra Columbia, a sharp-ridged range lying nearly parallel to and not
far from the Pacific coast. From its eastern border the plain slopes gently 10 or 15 miles south-
westerly to the head of the long San Andres Valley, which lies along the east base of the Sierra
Columbia Mountains and drains southerly into the Pacific.

Along Calamahue Valley and Canyon were thinly scattered mesquites, acacias, the garam-
bullo cactus, elephant wood, Fouquieria, Cercidium, Covillea, Opuntia bigelovii, and a few other
species, but immediately after passing over the Pacific slope divide to the Llano de Santa Ana
we came into a granitic country again and found a wonderfully vigorous growth of desert vege-
tation, including the species named above, yuccas, and numerous others extending across the
plain as far as we could see, and in many places forming impenetrable thorny jungles.

Soon after we reached the Llano de Santa Ana the sun set in gorgeous colors, and a little
later the moonlight: enabled us to continue southeasterly about 23 miles from our noon camp,
through a strange forest of desert plants. About 10 o'clock the trail became too dim to follow
and we camped in an open space at 2,200 feet altitude. The horses and mules were hobbled
and turned loose to search for food, while we made a fire and after a hasty meal unrolled our
blankets and slept, as usual, under the brilliantly starlit sky.

' September 17 we made an early start from our dry camp and continued 10 miles across the
rolling plain, which reached an altitude of 2,500 feet and became more broken and rocky: before
we reached the tinaja or water hole of Yubay: This is a large pothole of rain water 15 or 18
feet across and several feet deep in the bed rock at the mouth of a small canyon which leads
through a bare granite ridge rising about 500 feet. immediately northeast of Yubay. This is
one of a scattered series of ridges which extend for miles along the easterly side of this part
of the tableland from near the head of Calamahue Canyon to a point uate. southeast
of Yubay.

San Andres Valley to the westward was full of fog on the morning of Santen 17, and at
daybreak on the 18th the fog covered the country at least to 2,500 feet altitude on the granite
ridges back of camp. At sunrise the vegetation was dripping and the entire surface of the
ground was wet from the fog, but by 8 o’clock had become completely dry. These fogs must
be of material help in carrying vegetation through long rainless ‘periods. At this time the
days were very hot throughout this section of the peninsula, but the nights toward morning so
cool that we needed heavy blankets.

From Yubay the trail continues southward along the middle of the peninsula by way of
the old mission of San Borja, and we had been informed that it was 50 miles to the nearest
water in that direction. To attempt to travel this distance without water with our worn-out
stock would be extremely hazardous, and I decided to leave the trail here and strike directly
across country to San Andres, a deserted ranch near the Pacific coast, in the lower end of the
San Andres Valley, where it was reported we could find water nearer than on the regular route.
We accordingly left Yubay early in the afternoon of September 19 and traveled until after
dark. Our course was southwesterly down the sloping and slightly rolling plain, which is cut
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 10.

 

 

 

 

 

Fic. 1.—Summit of lava-capped plateau near the Tinaja de Santa Ana, north of San Ignacio.

 

 

 

 

 

Fic. 2.—Canyon near Tinaja de Santa Ana, in lava-capped plateau. The sandstone base and lava cap are shown on the farther wall of
the canyon.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 29

into mesa sections by a series of parallel dry canyons 50 to 75 feet deep extending down the
slope. Several miles below Yubay the sloping plain becomes more broken and rocky and ends
on the border of San Andres Valley. The bottom of this valley opposite Yubay has a width
of 6 to 8 miles westerly to the base of the barren Columbia or Prieta Range, which forms one
of the series of small ranges already mentioned as extending parallel to the west coast. The
valley has a southerly course, and a broad, stony dry wash like a river bed winds down its
bottom, draining this section of the peninsula, including the Llano de Santa Ana, into Playa
Maria Bay.

All down the slope from Yubay we traversed a jungle of giant and other cactuses, yuccas,
agaves, copal, paloverde, elephant wood, the tall columnar cirio, ocotillo, and many other
desert species. In the bottom of the valley this growth became extraordinarily vigorous and
formed a veritable forest, which over considerable areas was absolutely impenetrable. Fortu-
nately these areas of dense growth were irregular in outline and varied by narrow open spaces,
through which we were able with much difficulty to make our way. We were repeatedly forced
to turn back and try a new route. Just at dark we crossed the broad dry wash in the bottom
of the valley and came into a well-marked and recently traveled trail leading down the west
side of the valley—the west coast trail. We were much pleased to escape from the untrodden
cactus forest, and turning southward continued until it became too dark to distinguish the
trail. We then made a dry camp, 20 miles from Yubay, at an altitude of about 1,000 feet, in
a small open place beside the trail, the light from a group of dry yuccas we set on fire making a
fine illumination for half an hour. -

The darkness became less intense after midnight, and in order to make the journey as
easy as possible for the stock by traveling in the cool early morning hours, we broke camp at
2 a. m..and continued down the trail. After traveling about 10 miles we reached Punta Prieta,
a deserted cattle ranch. This place had been occupied within a few weeks and we were glad
to find a well about 40 feet deep with a bucket and rope, so we were able to water our stock.
Thence we continued about 10 miles farther down the trail to San Andres Ranch, at the ex-
treme lower end of the valley. Here the hills on both sides shut in and reduce the valley to
a narrow point from which a short canyon leads through the coast hills to the shore of Playa
Maria Bay. About a mile of the lower end of the valley bad been extensively improved and the
ruins of well-made houses, machinery, good wells, irrigation ditches, and old fields showed
that a large amount of money had been spent here, but the project had evidently been a com-
plete failure and from the ruined condition of the improvements it was plain the place had
been abandoned for many years.

I have since learned that this establishment belonged to an English corporation (Henry
Louis & Co.), which was operating here in 1887 with the object of making paper from the datilillo
or yucca (Yucca valida), which abounds throughout this region.’

At the extreme lower end of the valley water lies near enough the surface to make some wet
meadows and to fill a small pond. The green flats and large mesquites gave a welcome change
from the desert country we had traversed, and a short stop was made here to rest the stock.
An ocotillo we had not seen before (Fouquieria peninsularis) grew in the valley above San
Andres and Jdria columnaris was abundant in the valley just above the ranch. Harris’s hawk,
Scott’s oriole, and the black-chinned sparrow (Spizella atrogularis), were common here. <A dozen
or more Harris’s hawks, a species which had not been seen since we left Ensenada, gathered
every day at sunset to roost in the tall cirios near the ranch.

September 24 we continued our journey along the coast trail in an east-of-south course
a short distance inland, crossing low ranges of coast hills cut at frequent intervals by deep,
rocky, dry watercourses, all nearly parallel and leading to the sea. In a small open canyon
8 miles from San Andres we passed a brackish water hole called San Dominguito. About 10
miles beyond this we came to another broad, flat-bottomed valley, which narrows abruptly
between hills at the lower end, bringing underground water to the surface in wet meadows and
aoe mene into Santa Rosalia Bay through a short canyon, thus almost duplicating the

 

1 See Lopatecki in West American Scientist, pp. 129, 130, May, 1887.
30 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI

situation at San Andres Valley. This is Rosarito Valley, and from the appearance of its lower
end it was evident that during rainy years there are large swampy flats here, but we found water
only in a single small hole. We camped on the right-hand side near the lower end of the valley,
by the ruins of several large buildings, about which parts of machinery were scattered. Here,
years ago, unsuccessful efforts were made to extract the fiber from a large species of agave, which
is common on the hills of this section. The coast hills here are covered with a scattered growth
of agaves, yuccas, ocotillo (Fouquieria peninsularis), cirio, elephant wood, creosote bush, and
several species of cactus. The growth is smaller and stunted on the hill slopes, and much more
vigorous and abundant in the bottoms of the numerous broad canyons and small valleys.
Since leaving the San Pedro Martir Mountains we had had unbroken sunshine, but to-day local
showers from large cumulus clouds were seen to the east and south.

September 25 our trail led southeasterly through the coast hills about 8 miles to San Javier
Canyon, striking it a few miles above the sea. Where we entered San Javier Wash it forms a
narrow valley, soon closing in to a canyon bordered by low sandstone walls, down which the
trail led for some distance. At first we found a series of fine water holes scattered along the
bottom, but they were soon replaced by a clear stream of excellent water, which flowed about a
mile before it was again lost in the sand. Below this the trail passed out of the canyon for about
5 miles through the coast hills and then descended near the seashore to the northern point of the
Santo Domingo Plain. From this point southward the coast hills and low mountains swing
inland, the nearly level plain extending along the coast and broadening southeasterly into the
Vizcaino Desert. Near where we entered the northern end of the plain, the shore is bordered
by a broad belt of high sand dunes lying immediately back of the beach. On the inland side
they are succeeded by a nearly level area 4 or 5 miles broad to the base of series of rugged, scat-
tered hills and ridges. Farther south, 2 to 3 miles back of the shore, a tree-yucca forest extends
over the plain for miles, sparsely intermingled with Fouquieria and other large scattered shrubs
on which a heavy growth of orchilla (Roccella) was hanging. (See Pl. 29, Fig. 2, and pl. 30,
fig. 1.) Soon after entering the plain we saw traces of a wagon road, which skirts the coast
from the landing at Santa Rosalia Bay southeasterly to Santo Domingo. We followed it for
about 9 miles’ on the flat, well back of the sand dunes, until sunset, when we made camp about
a mile from the sea, with no water and very little grazing.

September 26 the road led down the plain 8 miles from our camp, then crossed the dunes
and skirted the shore of Santo Domingo Lagoon about 5 miles to the little group of houses at
Santo Domingo Landing, where the customs inspector lives. This is the place where supplies
are landed on the open beach for the interior gold mining camps of Calmalli and Campo Aleman.
A wagon road leads from the landing a little north of east across the plain to these camps. We
followed it for 3 miles to a well at the foot of the escarpment: of a mesa fronting the coast, where
the plain rises abruptly about 50 feet. Close by, on top of the border of the mesa, was a small
house, also a corral containing a shed where we camped. The well and buildings here are for
the use of freighters hauling supplies to the mining camps. This well is the only water supply
for the customs people, as there is none near shore at the landing. When we arrived at the well
our horses and mules had been without water for 24 hours and showed such signs of exhaustion
that we stopped here a day for their benefit. The scanty forage all along our route made the
journey hard for our animals, thus rendering it necessary to watch them closely and to stop at
intervals to permit them to rest and regain their strength. Without such care we could not have
made half of the journey.

The type locality of the desert thrasher ( Toxostoma lecontet arenicola) is near where we first
heard its notes in the sand dunes when we approached them north of Santo Domingo Landing.
It was abundant between the landing and the well, and in early morning and near sunset its
beautiful song was heard on all sides.

September 28 we took a cut-off trail easterly through the yucca forest, the wagon road mak-
ing a wide detour to the south. The trail led over a sandy plain through an unbroken forest of
yuccas intermingled with Fouquieria peninsularis, which had been plentiful since we entered its
range near San Andres. The yuccas had their tops all inclined toward the southeast owing to the
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 31

prevalence of northwest trade winds. Nine miles from camp we came into the wagon road again
beside a deserted frame house and broken windmill over a dry well. This was the Rancho de
Berrendo, situated on the border of the Llano de Berrendo, an open, clayey prairie 4 or 5 miles
across with a dwarf chamiso almost the only plant growing onit. The sandy soil and the yucca
forest stopped abruptly on the border of this prairie, but began again beyond it to the east.
The plain rises very gradually inland, and about. 20 miles from the coast the soil becomes less
sandy and the yucca forest gives way to a more mixed growth of yuccas, ocotillo (Fouquieria
peninsularis), many giant cactuses, including the largest specimens of the latter we had ever
seen, and many other species. We traveled until long after dark, and at 9.30 p. m., at an alti-
tude of 850 feet, reached a solitary hill known as the Cerro Perdido, about 25 miles from our
camp at Santo Domingo. This is one of the regular camping places on this road, but there was
no water, so we were compelled to make another dry camp, which was hard for the stock.

September 29 we made an early start and a mile beyond camp came to another abrupt
escarpment about 50 feet high, capped by a plateau extending away eastward into the interior.
Before us to our left a range of hills approached our course diagonally from the northwest, while
far away to the east low mountains, some rising to an altitude of from 3,000 to 4,000 feet, stood
out boldly. About 10 miles east of this last escarpment the road led us through a range of low
detached mountains or hills about a mile wide and 5 miles long. The hills rise abruptly and
island like from the level plain, and from their relative position and formation it was evident
‘that they represented the tops of a short, partly buried mountain range. About 4 miles beyond
the road led through another group of hills, turned to the south, and at an altitude of 1,200 feet
came to the mining town of Calmalli. This is located on the foot of a group of low bare hills
and contained 25 or 30 small houses, two or three stores, and the reduction works for the mines.

Extensive gold placers were discovered at Calmalli in 1883 from which, during the next
few years, over $3,000,000 were taken. The placers were long ago worked out and at the time
of our visit the town was nearly deserted. A quartz mine nearby was being worked in a’small
way by an American, Mr. Byron Hall, who received us hospitably.

. Good trails lead from Calmalli easterly to Barril Landing, and to San Rafael, on the gulf
coast. The trail passes over a divide on the mountains at an altitude of about 2,500 feet.  Barril
Landing is the point where supplies are landed from Guaymas. They are then packed on mule:
and donkeys across the mountains to Calmalli. '

The wagon road from our camp at Cerro Perdido at first led through the same strong growth
of desert plants noted the previous day, but near Calmalli the country became more sterile and
the cirio, which we had not seen nearer the coast, appeared again. The plain extending inland
from Santo Domingo reaches its eastern limit a few miles east of Calmalli, where ranges of barren
rocky mountains rise abruptly from the general level and form a barrier. Immediately about
Calmalli, especially to the west and south, rise low round-topped barren hills or small moun-
tains.

. October 1, we left Calmalli and followed a continuation of the wagon road about 5 miles
southerly through the low mountains to the still smaller mining camp of Campo Aleman, where
another mine was being worked on a small scale. At both camps water was obtained from wells
in the bottoms of gulches. The hills from Calmalli to beyond Campo Aleman contain much
slate with some quartz and show many signs of former placer mining. Two or three miles
beyond Campo Aleman the road led down from the hills to the lower northeastern extension of
the Vizcaino Desert plain, which extends to the base of the mountains a few miles east of our
route and west to the coast about Scammon Lagoon. We made a dry camp on the plain, at
an altitude of 500 feet, about 14 miles south of Calmalli.

October 2, about 16 miles southeasterly across the plain, we came to Poso Altamirano, a
small cattle ranch at an altitude of 400 feet on the bank of an arroyo where a shallow well in
the granitic sand of the wash gave a good supply of fine water. A few miles northeasterly
across the plain from this point the light-colored granite mountains, which are practically con-
tinuous from the California border to this point, end abruptly and are replaced thence southerly
along the backbone of the peninsula to La Paz Bay by black, lava-covered mountains and

158961°—21—3
32 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

plateaus. The northern foothills of these lava mountains begin about a mile south of Poso
Altamirano. ,

October 3 we traveled about 3 miles southwesterly from Poso Altamirano and rounding a
point of the mountainous plateau éntered the mouth of San PabloCanyon. The western border
of this mountain plateau rises abruptly from the Vizcaino Desert, and the lower part of the
canyon, which has its head far back in the range, is broad and gently sloping with a dry, sandy
wash which evidently at times carries great volumes of flood waters winding down its bottom
and thence across the plain toward the sea. We ascended the canyon between increasingly
high walls of black lava, underlaid by horizontal beds of sandstone, for about 12 miles, until it
narrowed to a width of a few hundred yards, with a high bench along its southern side. Here we
camped near the lower end of a small stream of clear water just above the point where it sinks
in the sandy wash. A short distance above, close to the ruins of the San Pablo Mission, was a
small ranch occupied by a Mexican family. A few acres of land are still in cultivation below the
irrigating ditches built long ago by the missionaries. The old mission has an altitude of about
1,000 feet and is in an extremely hot location, the effect of the sun being intensified by the heat
stored in the black lava walls of the canyon, which immediately above the ranch close in and
form a deep box canyon. A vigorous growth of large cactuses and other desert vegetation
occupies the bottom of the canyon about our camping place, and a few fan palms were growing
at the ranch. The water and abundant vegetation near the old mission made an attractive
place for birds and we saw here the first representatives of such cape district species as the cape
cardinal and cape house finch. What was still more surprising was the discovery among the
cactuses at this camp of a previously unknown chipmunk (Lutamias merriami meridionalis).
We had seen no relative of this species since leaving the pine forests on the upper slopes of the
San Pedro Martir Mountains. Its presence here appears to indicate that the species formerly
ranged continuously at least thus far south but has become extinct elsewhere.

This was a promising locality to work, but cattle had eaten the feed so closely there was.
nothing for our stock and we were forced, much to our regret, to continue our journey the next
day, October 4. The trail turned southerly across the bench near the lower end of the water
and then zigzagged directly up the steep canyon side for over 1,000 feet. This brought us out
on the top of the west slope of a great lava plateau with a broad undulating surface rising toward
the east, broken here and there by scattered volcanic hills, and descending on the west to the
Vizcaino Desert. This plateau country is made up of horizontal beds of sandstone capped by
heavy beds of dark lava. The scanty soil supports only a thin growth of vegetation, as most
of the surface is covered by an almost continuous layer, and sometimes deep beds, of loose frag-
ments of lava, which give a desolate and forbidding appearance to the landscape. This exces-
sively stony plateau is cut by numerous shallow gulches and occasional deep canyons, all drain-
ing westerly into the desert. Travel through this endless bed of loose stones was very difficult
for our animals. The trail wound along the top of the mesa, crossing a number of the larger
canyons, one of which was about 2,000 feet deep.

Seven miles southerly from San Pablo on top of the undulating plateau, at an altitude of
3,400 feet, we reached a small house and corral with a large natural tank of water in the bottom
of a lava-bedded gulch near by. This place, known as Santa Ana, is occupied as a cattle ranch
part of the year but was vacant at the time of our visit. From the crest of a small ridge here
an extended view, as far as the eye could reach northwest and southeast over the top of the
plateau, showed series of volcanic ridges and peaks breaking its surface irregularly. To the west
was a superb view over the Vizcaino Desert to the Sierra Pintada and Sierra Santa Clara, the
salt beds at Ojo de Liebre, the old whaling location of Scammon Lagoon, and across the Pacific
to Cedros and other coast islands. Pt

October 5 the trail continued southeasterly over the lava plateau, crossing several canyons
and gradually swinging down toward its western border. About 10 miles from Santa Ana we
came to a large rock tank, the Tinaja de San Esteban, at an altitude of 1,200 feet in the bottom
of a small gulch only a few miles from the eastern border of the Vizcaino Desert. Near this
point we saw the southernmost representatives of the cirio (Jdria columnaris) and the first
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 11.

 

 

 

 

 

Fic. 1.—San Ignacio in 1905. The old mission church stands in the background. Many small houses shown on the flat were
destroyed by a flood in 1911.

 

 

 

 

Kia. 2.—Waterhole in Arroyo San Angel on eastern side of Vizcaino Desert, about 20 miles west of SanIgnacio. Self-planted date palms
are numerous there,
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 12.

 

 

 

 

Fic. 1.—Ojo de Liebre, or Jack Rabbit Spring, near head of Scammon Lagoon, on Vizcaino Desert. This spring is the only water within a
day’s journey in any direction and is marked by twostunted cottonwood trees which are visible only a short distance owing to surround-
ing sand dunes. °

 

 

 

 

 

Fig. 2,—Sea turtle corral at Scammon Lagoon. Sea turtles abound in Scammon Lagoon, where they arc being taken and shipped by
a weekly schooner to San Diego. About 1,000 turtles a month are shipped from this point.

(Photographs from Capt. H. W. Cochran, U.S. A.)

 
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 33

northerly representative of the white barked palo blanco, which was abundant in the bottoms
of the larger canyons. A few other plants, mainly tropical or subtropical species not noted
before, were first observed along the canyons crossed to-day. Five miles beyond San Esteban,
across an intervening lava mesa, we came to La Natividad, a large canyon in which water
usually exists. The bottoms of the empty tanks were still damp but no water could be found,
so we were forced to make another dry camp.

October 6 the trail led southeasterly about 20 miles to San Ignacio. This part of the
route crossed a lower and more undulating part of the plateau, which was even more rocky
than the sections we had passed, being characterized by numerous extended beds of large
angular fragments of lava, sometimes a mile or more broad, over which the animals had to
pick their way slowly and with great caution to avoid falling. Some of these lava beds were
made up of rough fragments 1 to 3 feet in diameter, and it would have been impossible to cross
them with animals but for the work done by the Indians under direction of the early Spanish
missionaries in clearing the trail for several miles. These made trails are parts of the old Camino
Real, or Royal Highway, or main road, which united some of the missions in early days. After
covering about 12 miles of this excessively trying road we came to better traveling, and 2 miles
beyond reached a narrow tank more than 100 yards long full of water, called the Tinaja de
San Juan, situated in the lava bottom of a small gulch. A wild fig tree grew on one side of the
tank and several coots (Fulica) were at home there despite the bare lava borders to the water.
Five miles beyond we rode out on the brow of the mesa overlooking the valley of San Ignacio
and saw the village among the date palms a mile away on the opposite side. At this point
a horse ridden by one of our camp men became too exhausted to go farther. Its saddle was
placed on one of the- pack mules and the animal left to rest for a day before it was brought
into.camp. We continued up the valley and camped by a great spring just above the town.
This is one of several springs rising in the flat bottom of the valley which supply the water
used in irrigating the palms and other cultivated crops. Below the springs are several ponds
surrounded with tules and bushes.

San Ignacio Mission was established in 1728 and has been continuously occupied from
that date. (See Pl. 11, Fig.1.) Soon after 1730 three kinds of Arabian date palms were intro-
duced and to-day it is estimated that there are from 50,000 to 60,000 trees in this valley. They
form a small forest about the town and thence down the valley for several miles and render this
one of the most beautiful oases in all the peninsula. About 50,000 pounds of dried dates are
sent to Santa Rosalia on the gulf coast each year. Ravens and Gila woodpeckers destroy the
ripening dates by day and the bassarisk, or ring-tailed cat, by night, thus doing considerable
damage. Many gray foxes and striped skunks have their dens in the rocky walls of the canyon
and come into the palm groves at night to pick up the fallen fruit.

In addition to the date palms various crops and fruits are cultivated on a small scale,
including oranges, lemons, pomegranates, olives, figs, grapes, sugar cane, corn, beans, and others.

San Ignacio is situated at an altitude of 500 feet, and at the time of our visit contained
about 800 inhabitants. It is one of the most fertile and populous oases in the peninsula. The
old mission church and the adobe houses of the principal inhabitants are on the basal slope
under the lava cliffs forming the south side of the valley, but most of the people live in thatched
huts scattered amid the date palms in the lowest part of the bottom among drainage channels
where flood waters pass. Above the town the remains of a well-made earthen dike extends
from the south side in a curve across more than half the width of the valley and then down
on the north side. This was built by the Indians under the direction of the Spanish missionaries
to turn flood waters out of their old channels and around the north side of town, thus protecting
the fields and houses located in the original drainage way. Long ago a heavy flood broke
through the dike and drowned some of the people, since which the survivors and their descend-
ants, lacking the energetic direction of the missionaries, have never had sufficient energy to
34 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

repair the break, The dike is 4 to 10 feet high with the convex curve up the valley: and with
the points where it crosses drainage channels faced up with rubble.

We were informed that this locality sometimes suffers from prolonged droughts covering
several successive years, during which the large springs continue to flow steadily but with
somewhat lessened volume. The abundant water and attendant plant life at San Ignacio
render this an exceptionally attractive place for birds and a number of cape district species
‘ werecommon. Among these were Chaemepelia:passerina pallescens, Melospiza melodia rivularis,
Richmondena cardinalis ignea, Pipilo fuscus albigula, and Geothlypis beldingi goldmani.

From October 10 to 16 I remained with one camp man at San Ignacio, while Goldman
made a side trip to the Vizcaino Desert in search of antelope. From San Ignacio he took a
westerly course about 25 miles to San Angel, a deserted ranch in the shallow Arroyo de San
Angel on the eastern side of the desert, where holes in the arroyo give an abundance of good
water. (See Pl. 11, Fig. 2.) Numerous date palms and natural forage plants were growing at
this place, which, surrounded by shifting sand devoid of vegetation, forms a typical little oasis.
Fifty miles to the westward across the desert he found the first water in that direction in the
Tinaja de Santa Clara. (See Pl. 24, Fig. 2.) This is a fine tank of good water well hidden by
the rocky walls of-a small canyon on the western side of the Santa Clara Mountains. Some
years ago, during the placer mining in the mountains to the northward, this tank was often
visited but is now rarely seen by man. Goldman found antelope extremely scarce on this
desert, where they were formerly numerous, but shot a mule deer far out on the flat open plain,

We left San Ignacio October 18 and ascended the very rocky course of the San Ignacio
Arroyo easterly about 21 miles to La Cueva, a small unoccupied cattle ranch at an altitude of
about 750 feet, backed by a huge shelter cave in the rough lava of the northern wall of the
canyon in which we camped. On the opposite side of the canyon is a fine rock tank of good
water, and water comes to the surface at a few places in the creek bed. A couple of miles above
San Ignacio the valley narrows to a low-walled canyon with an extremely rocky bed, bor-
dered on both sides by the gently sloping, rolling plateaus covered with loose stones. Large
groves of palo blanco were scattered along the bottom of the canyon and a fine growth of large
cactuses, tree yuccas, and copals covered the bordering hills. It was cloudy all day with light
showers in the afternoon and at night.

October 19 the trail continued 5 miles easterly up the sloping mesa to the west base of the
Santa Lucia Mountains, on the foot of which is a small cattle ranch known as the Rancho de
Santa Lucia. Beyond this the slope became steeper, and the mesa led rapidly into the heart
of the rugged Santa Lucia Mountains, which extend in a northerly and southerly direction.
This is a narrow, sharp-ridged range, apparently almost wholly made up of beds of black lava,
many of which rest in an almost original form. In the midst of the ragged peaks our trail
forked, and we turned northerly on the one to the left. This led us to a small ranch in the
bottom of a canyon, where we learned that we had chosen the longer trail to Santa Rosalia by
way of Cerro Verde instead of the more direct one to the right which leads by the mining camp
of Santa Agueda. From the ranch the trail led up a long, narrow, and extremely rough and
rocky canyon to a knife-blade divide at an altitude of 2,000 feet, from which we had a fine view
across the country we had traversed and ahead over lower country to the gulf. Descending
the abrupt eastern slope to the bottom of a canyon at an altitude of about 800 feet, we camped
about 20 miles from Las Cuevas under the shelter of some huge rocks just as it began to rain
heavily.

October 20 the trail led easterly across a rolling mesa formed of deep beds of coarse rubble
conglomerate traversed by deep, narrow and sharply cut canyons leading from the Santa Lucia
Mountains to the gulf. After passing through about 15 miles of this country the trail entered
a narrow canyon, along which for 3 miles were tramways and other signs of mining. At the
mouth of the canyon, on the shore of the gulf, we came to Santa Rosalia, the largest and most
prosperous town in Lower California. Santa Rosalia then had a population of about 7,000

 

1 Another flood is reported to have swept down the valley in the fall of 1911, washing away houses and drowning more than 20 people.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 35

people and is one of the larger copper mining camps of the world. It was the property of a
French company which owns 1,000,000 acres of the surrounding country. By the terms of
their concession the company exercises strict control over the town and adjacent country, the
part taken in the government by the few local Mexican officials being much subordinated to
‘the rule of the company. At the time of our visit there was regular steamer connection between
‘Santa Rosalia and Guaymas, on the opposite side of the gulf, as well as with San Francisco.
The mining company was building a breakwater of molten slag from their furnaces by means
of which a safe harbor for the ships calling here will be provided; now they lie in the open road-
stead, exposed to every storm. At this place the horse which became exhausted on the trail
near San Ignacio was traded for another one, which was able to complete the journey.

We left Santa: Rosalia October 25 and followed a trail 10 miles southeasterly across the
stony mesas and intersecting canyons bordering the shore of the gulf, the mesas having an alti-
tude of 100 feet: or more and ending precipitously at the shore. The mesa country ended
abruptly on the south and we descended to an open coast plain near the little fishing village of
San Lucas, on the shore of a salt lagoon at the head of a small bay. The first mangroves seen
‘during the journey were growing in this lagoon. They were 4 or 5 feet high and apparently the
northern representatives of the species on the east coast of the peninsula. Following the shore
‘of the bay southeasterly for about 6 miles we reached another small group of houses called
San Bruno. The plain all about this bay appears to be underlaid at a small depth with fresh
water. At San Bruno there was a little cultivated land including a small vineyard and olive
orchard planted by an English company about eight years ago. These plantings were unsuc-
cessful and were being abandoned at the time of our visit. A number of little patches of sugar
cane and other crops were irrigated by means of shallow wells, from which the water was raised
by long sweeps. The Anglo-Mexican Colonization Co. once had a colonizing concession here
which was long ago abandoned. We remained one day at San Bruno to learn what birds and
mammals occur there but found it an unproductive spot and on October 27 continued the
journey. "

From San Bruno the trail led southerly 8 miles inland across the coast plain to a water hole
at the abandoned ranch of San Marcos, a few miles from the gulf. Seven miles beyond this
place we came to a shallow well at another deserted ranch called Las Delicias, where we camped.
The coastal plain here, as well as the top of the mesa along the coast near Santa Rosalia, is
extremely arid but with a plentiful growth of desert shrubs which, with the exception of Covillea,
were mainly leafless. Near Las Delicias we passed a belt of green a few hundred yards wide,
like a broad ribbon across the brown plain, showing where a heavy shower had passed over the
country some weeks before and wet the ground sufficiently to bring up the grass and other
herbage. Several miles westerly from our route the gently sloping coast plain ends abruptly
-at the east base of the precipitous Santa Lucia Mountains.

October 28 the trail continued southerly for 10 miles across the coast plain, which was
heavily overgrown with giant cactuses, and then southeasterly 5 miles through a range of
low hills, 300 to. 500 feet high, to the town of Mulege. This place is situated between low hills
in the valley, a few hundred yards wide, of the Arroyo del Potrero, and at the head of a narrow
lagoon extending more than a mile inland from Concepcion Bay. In 1705 a mission was estab-
lished there and the town is now one of the most beautiful places in the peninsula, and at this
time had about 750 inhabitants. Some large springs in the creek bed, just above the town,
yield water for irrigation, and the place is embowered among groves of date palms, bananas,
oranges, lemons, pomegranates, and fig trees, with fields of sugar cane and other crops inter-
spersed. The old mission church stands on a hill by the town overlooking the valley, and on
the long peninsula forming the eastern side Re Concepcion Bay rises a barren rugged range of
low mountains.

‘ ' As at San Ignacio at this season the diag palms and other fruit here attract many gray
foxes (Urocyon), bassarisks (Bassariscus), striped skunks, and raccoons, which live in the shel-
ter of the rocky walls of the valley by day and search ihe orchards for fallen fruit by night.
We'eaught: three gray foxes in the small palm grove at the upper end of Mulege the night we
36 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

camped there. These were caught soon after dark and attracting our attention by husky
barks and yelps were taken from the trap by the light of a candle.

We traveled October 29 about 6 miles southwesterly, up the broad bottom of the Arroyo
del Potrero to Poso San Rafael, a small cattle ranch on the eastern side of the arroyo with a
shallow well in the dry creek bottom. The most vigorous growth of mesquites we saw in Lower
California occurs along the sides of the arroyo near this ranch, many of them being 30 or 40
feet high, thus indicating abundant permanent water at a moderate depth..

October 30 the trail led southerly up the valley bottom, which gradually narrowed to a
wide canyon with a broad dry wash, paved with water-worn bowlders, winding down its course;
the size and abundance of the bowlders gave evidence of occasional heavy floods. About
10 miles from Poso San Rafael the valley narrows for a short distance to a low box canyon,
and to the right rises a flat-topped mountain called the Mesa de Zacatecas, where numerous
mountain sheep are reported to occur. The trail forks at this point, and the branch to the
right leads about 2 miles to La Trinidad ranch near the base of the mesa. Continuing on the
main trail 5 miles farther up the rocky and narrowing canyon we came to the ruins of El Potrero,
an abandoned cattle ranch, where we camped. El Potrero is located at an altitude of 600 feet
on a rocky mesa 6 or 7 miles in diameter with mountains on three sides, especially the southern
side, which is limited by the precipitous walls of a high mountain ridge, plainly visible from
Mulege, which bars the way like a wall in that direction. The Arroyo del Potrero heads at the
base of this mountain wall, and the bottom of its narrow canyon through the mesa, imme-
diately above the old ranch, contained numerous tanks and Parniles of good water with a
trickling stream here and there.

We left El Potrero October 31 and followed the trail catitwestanly about 3 miles across the
mesa to the base of an abruptly rising mountain slope, which it ascends in a series of sharp
zigzags and along narrow ledges for about a mile to a pass on a sharp ridge at an altitude of
about 2,400 feet. On both sides of this pass narrow mountain ridges forming the peninsular
divide rise 1,200 feet or more higher and extend to the northwest and southeast. The moun-
tain is so narrow at the pass that when the summit is reached the trail descends immediately
on the other side into the narrow, cliff-bordered, and extremely rocky canyon forming the
head of the Arroyo de San Jose, in which for half a mile or more the slope is even more abrupt
than on the side we ascended. This narrow divide forms the watershed of the peninsula, the
drainage on the north being to the gulf and that on the south to the Pacific. Six miles below
the summit we passed a number of stone-walled inclosures and ruins of houses marking the
long-abandoned Guajademi ranch. Four miles below this were the ruins of El Zapote, another
abandoned ranch; 3 miles below, another one called Los Candelitos; and 3 miles still farther
down was Agua Grande, a deserted ranch, with a tank of water in the bottom of the canyon.
At all of the places named there were water holes in the rocks and traces of former occupancy,
but all had long been abandoned. A few cattle were still ranging about Agua Grande, and
the water was extremely filthy, but none of the party knew the country ahead and, owing to
the condition of our stock, we could not risk a dry camp, so we stopped there and made the
best of it. The next morning, only 2 miles below this camp, we found a large tank full of good
clear water.

A vigorous and abundant growth of giant cactuses and other desert plants occupied all the
lower slopes of the canyon, and wild fig trees were growing in several places where water was
close to the surface. A number of pyrrhuloxias and ground doves were seen near the water
holes, and at Agua Grande the call notes of the pigmy owl (Glaucidium hoskinsi) were heard
on all sides as we were breaking camp, the only time we heard this species during the journey.
This locality is worth some careful work, but feed was so scanty and our stock in such worn
condition we considered it imprudent to remain there even for a day.

About 9 miles down the canyon from Agua Grande we passed a little ranch called San Joselito,
occupied by a single family. Four miles below this we came to San Jose, a larger ranch, with
some date palms and a small field of sugarcane. Five miles below San Jose was Rancho Hondo,
which was occupied, and just below it a crossing of the canyon known as Paso Colorado, where
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 37

we camped under a large mesquite tree. Water rises in the canyon just above Rancho Hondo
and a small stream, bordered by rushes and willows flows by Paso Colorado. During the after-
noon clouds gathered, and it began to rain. From Agua Grande to Rancho Hondo the canyon
varies from one-fourth of a mile to a mile broad, but widens below the last-named point and
the walls become much lower, descending southward toward the Pacific coast.

November 2 we continued about 3: miles down the canyon to the small village of Purisima
Vieja, where the trail turns to the left and climbs the wall out of San Jose Canyon and leads
across the broad plateau thickly covered with angular fragments of lava. From the top of the
divide above Guajademi down to Purisima Vieja the trail follows the bottom of San Jose Canyon.

All the mountains in this district are of volcanic origin, and the Pacific slope forms a con-
tinuation of the lava-covered plateau we traversed near San Ignacio. This plateau is cut in
all directions by deep ragged box canyons like the upper part of the San Jose Canyon near Guaja-
demi. The rocky walls of San Jose Canyon rise almost vertically in a series of steplike mesas, the
cliffs in some places having a talus slope, but often rising as clear vertical walls. This entire
section is of black lava, overlying basal beds of sandstone and hard conglomerate, which unite
to give a gloomy and savage aspect to the rugged wildness of the scenery. This volcanic
plateau formation continues throughout the length of thé canyon, but the walls become increas-
ingly lower and the bottom broadens from above Paso Colorado to Purisima Vieja and below.
Purisima Vieja consists of a small group of adobe houses surrounded by a few date palms and
small fields, mainly planted with sugar cane and irrigated from springs, which rise in the canyon
bottom.

At the point where our trail ascended the southerly wall of the canyon the plateau has an
altitude of about 900 feet and slopes southwesterly toward the coast plain. After traveling in a
steady rain 10 miles from Purisima Vieja southerly and then southeasterly over the undulating
and broken lava beds, varying in altitude from 1,000 to 1,300 feet above the sea, we came to
the top of an escarpment nearly 1,000 feet high overlooking the picturesque valley and town
of La Purisima, the seat of one of the old missions. When we arrived at La Purisima the
storm, continuing all day, had increased to a cold, driving rain and, owing to the small size
and crowded condition of the houses, we found it difficult to secure a place for ourselves and

-outfit. Nearly every roof in town was leaking abundantly, and the room we finally secured
in which we crowded with all our saddles and the loads of four pack mules, gave but partial
shelter. The main group of houses, many of them whitewashed, was located near the old
mission at an altitude of about 300 feet in the flat-bottomed valley nearly opposite the point
where our. trail approached the edge of the mesa, but other small houses were scattered among
fields for 8 or 10 miles up and down the bottom. At the time of our visit this valley had from
400 to 500 inhabitants.

La Purisima Valley varies from one-fourth of a mile to a mile in width and is bordered by
precipitous walls, 500 to 1,000 feet high, with a lava cap on the mesas extending away on both
sides. The rocky walls are eroded into many curious forms, which add much to the picturesque
interest of the surroundings. The largest stream in central Lower California rises from springs
near. the head of. this valley and flows down its course, irrigating many small fields in which are
grown numerous date palms, orange, and other fruit trees, mission grapes, bananas, sugar cane,
and corn. One of the main industries is that of making raisins from mission grapes, which are
packed on donkeys to San Gregorio Landing, on the Pacific coast just south of the mouth of
Purisima Canyon, and shipped in small sailing vessels to the mainland of Mexico. A few miles
below town the canyon opens out on a broad coastal plain where there is said to be much fertile
land. An American company prospecting for oil some years ago is reported to have sunk a
well in the lower end of the canyon, and the natives informed me that it had been flowing good
water ever since. This indicates the possibility of getting artesian water for irrigation on
some of the coastal plains in this region.

The rainstorm continued two days and ended in a dense fog which filled the valley. The
morning of November 4 the fog cleared away and after partly drying our outfit we continued
our journey. The trail led from near the lower end of the town up about 1,000 feet to the top
38 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI

of the abrupt south wall of the canyon, and thence southeasterly across the. rolling and broken
plateau. About 6 miles from La Purisima we came to a deep box canyon, in the bottom of
which San Vicente Ranch is located, the house standing i in the midst of a few date palm, lime,
and zapote trees, near small patelies of corn and sugar cane. A small stream of water rises
in the bottom of the canyon and flows enough to irrigate this small oasis, which supports a
single family in the midst of a gloomy waste of broken lava. From San Vicente the trail led
down the canyon and then up a tributary gorge on the left to the top of the plateau, where
we entered a great series of lava beds descending from a conical crater a few hundred feet higher
than the general level. Here for miles on all sides lie beds of bare, shattered lava containing
occasional small patches of soil on which were growing various species of desert bushes with
a few giant and other cactuses. With some difficulty we found a small patch of ground partly
free from stones for our beds and camped at an altitude of about 1,300 feet near the base of the
old crater peak. During the night a solitary huge black cloud avepe over the otherwise bril-
hantly clear sky and gave us a hard shower. °°!

November 5 the trail continued about 18 miles southeasterly through the ke plateau
country among scattered volcanic crater cones which rise from about 300 to 1,000. feet above
the general level.. Each cone is the center of a series of lava beds which spread over the sur-
rounding plain. The surface of the lava beds is covered with angular masses of shattered lava
through which the animals had to pick their way with much care. Canyons of varying’ depths
gash the surface of this great voloanio plain, which gradually rises to its greatest elevation in
this district about the base of La Giganta, near Comondu. The lava beds of the surface are
underlaid by stratified beds of volcanic ashes and similar material, usually consolidated into
soft rock. The great expanse of broken lava lying between Guajademi and Comondu gives
a wild and savage aspect to the district. In places between La Purisima and Comondu the trail
was boggy from the recent rains and added to the difficulties of the road, for the general sur-
face was such a continuous bed of lava that we were confined to a narrow track winding among
the shattered rocks. About noon we crossed a broad rounded divide about 2,000 feet above
sea level and descended the farther slope to an altitude of 1,500 feet, where we came abruptly
to'the edge of the canyon wall immediately opposite the village of Comondu. This is a scat-
tered settlement of about 1,000 people located at an altitude of about 700 feet along the flat’
bottom of'a canyon varying from 200 to 800 yards in width, and with walls from 500 to 800
feet high. A stream rising in large springs near the upper end flows about 8 miles down the
canyon and irrigates a continuous succession of small fields. Date palms were scattered
irregularly along the stream in a thin line through the small vineyards and fields of’ corn, sugar
cane; and other crops, and other palms were grouped with the orange, fig; and pomegranate
trees about the houses. The view up and down the canyon from the top of either well 9 is one
of the most beautiful in the peninsula:’ (See Pl. 13, Fig. 2.) :

For several miles‘along the plateau before we came to Comondu the trail fallomed an old
wagon road over slopes on which there was a scattered growth of large agaves. Here and
there the road branched and led across the slopes to one side or the other. At Comondu we
learned that these roads were made some years ago by an American company which built
wagon roads from the landing on the coast to Comondu and over the adjacent hills, brought
machinery up from the coast, and put up buildings with the purpose of extracting the fiber
from the agaves growing there. The enterprise was foredoomed to failure and the old roads
and one or two abandoned houses in the canyon near Comondu are all that remain of it.

Bryant, Brandegee, and others who have visited Comondu speak of the kindly treatment
received from the people there, and we met with the same hospitable reception. The visit
of Bryant and Brandegee was still remembered, and the mayor of the town in 1905 was the man
who outfitted them with saddle and pack animals for their overland trip to San Quintin in
1889. Comondu has an almost purely tropical climate, although light frosts are not very
uncommon in winter.and at intervals are severe enough to kill the leaves on sugar cane. Mala-
rial sOyeh, called smorenaes is reported to occur here in BEr) and elsewhere anouve irri-
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 13.

 

 

 

 

 

Fic. 1.—Western border of Vizcaino Desert. Santa Clara Peak in background.

 

 

 

 

 

Fig. 2,—Comondu Valley. This is a broad canyon in the lava plateau, with a stream and fertile soil, making a beautiful oasis with
date palms and other tropical and subtropical fruits.
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 39

‘gated areas in Lower California as far north, at least, as Mulege, where a number of cases
occurred in 1904. These fevers are reported to be sotber: severe and often fatal to old people.

After several days of productive work at this interesting locality, we resumed our way
November 10, following the wagon road southwesterly down the canyon. A mile below town
the walls close in and the canyon for about 12 miles is narrow and very crooked, the high basal-
tio cliffs forming many ‘wild scenes. Then the canyon broadens and the walls descend gradu-
ally for 3 miles before they merge into the gently sloping coast plain, which is about 12 miles
wide at this point. The lava cap to the mesa forming the south wall of the canyon below
Comondu is largely columnar basalt, but the north wall is mainly of horizontal strata appar-
ently very different in character. The water appears to have formed the canyon by following
the line of contact between two distinct lava flows. Below the mouth of the canyon we turned
to the right from the wash for a couple of miles to a small occupied cattle ranch called El Batequi,
19 miles from Comondu, where a shallow well furnishes: the ony available water on this part
of the plain.

November 11 we crossed the broad sandy wadh where panna Arroyo spreads over the
plain, and taking a more southerly course for 10 miles over the slightly undulating surface
came to a deserted ranch called Poso Grande, where there was good water in a shallow hole
in the sandy bed of an arroyo. ‘Two families of Mexicans with nine small children were tem-
porarily camping here in small brush huts. An old cornfield close by showed that crops are
sometimes grown here by subirrigation. Poso Grande is 3 miles inland from San Jorgé, the
head of boat navigation in the long narrow lagoon Jeading north from Magdalena Bay. Dried
fruits, hides, and other products from Comondu are shipped from there by small boats to the
bay, and other articles are brought in. From our camp at Poso Grande and from far up and
down this coastal plain the high solitary volcanic peak of La Giganta, located on the volcanic
plateau between Comondu and Loreto, stands up boldly as a striking landmark.

. November 12 we remained at Poso Grande to do some work on this part of the coast plain.

Several showers had fallen during the last two nights, but inthe afternoon a sea breeze came
in and’ the sky was cleared of threatening clouds for the first time since we left Mulege about
two weeks before. It had rained at intervals throughout this time. These winter storms in
the southern half of the peninsula are called “equipatos,” and sometimes heavy rains continue
for two weeks and render the country.so boggy that travel is impossible. November 13 we
traveled 14 miles southerly parallel to the coast over the same slightly undulating sandy plain
and made a‘dry camp in a heavy forest of gigantic cactuses and other desert plants. The
gentle undulations on this’plain appear to be due to the formation of broad low swells by the
drifting sand from the coast. About 10 miles south of Poso Grande we passed ane aban-
doned ranch called Poso Medano Blanco.
November 14 we continued southerly over the same coast plain and about 8 miles — camp
passed the deserted ranch of Santo’ Domingo on the bank of a deep dry wash where some well-
made adobe houses were going toruin. Here a waterhole in the sandy wash enabled us to water
our stock. Just south of Santo Domingo, on a large flat extending along the course of the dry
wash, we passed through a fine large growth of mesquite trees, indicating permanent water at
moderate depth. There is much fine farming land here’and all along this coast plain, which needs
only water to become very: productive.

In the afternoon we crossed a broad belt.of country in which there was a good fresh growth
of grasses and other herbaceous’ vegetation, 4 to 6 inches high, most of which was ripening its
seeds. Some ranchmen living in this section we met told us that a heavy rain fell here on August
‘7? which resulted in this growth.. Just before nightfall we suddenly passed the border of this
belt of green into the brown arid desert again and in order to give our stock the advantage, of
the fine pasturage were obliged to turn back and reenter the green belt to camp.

November 15 we traveled southerly along the same coastal plain, parallel to the shore but
some miles inland, and 20 miles from camp reached Matancita, or Soledad, ranch. This is'the
headquarters" fontile ranch of the Chartered Co. of Lower California, an Soensiiccrt! ‘corporation
which holds'a concession covering a belt: about 16 miles wide all along this coast from a little
40 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI

south of Magdalena Bay north to beyond Scammon Lagoon. We were hospitably welcomed by
the local manager of the company, Mr. W. J. Heney. Formerly this ranch was headquarters
for the orchilla-gathering business spoken of below... Matancita ranch is 10 miles inland from
the shore of the coast lagoon which gives boat communication with Magdalena Bay. The ranch
has some good adobe houses and a steam pump for raising water to irrigate a small vineyard,
orchard, and garden. A shallow well in the deep channel of the large dry arroyo which passes
the ranch supplies an abundance of water for pumping and indicates the possibilities of the
numerous similar arroyos we crossed along the coast plain to the north. The discovery of oil
in Lower California, which would supply cheap fuel for pumping water, would undoubtedly
result in the reclamation of much productive land along the west coast of the peninsula.

The coast belt owned by the Chartered Co. contains practically all the land of the peninsula
on which orchilla grows and was formerly gathered. The orchilla (Roccella) is a lichen resem-
bling Spanish moss in appearance which grows abundantly on the branches of Fouquieria penin-
sularis and other shrubs and low trees along this part of the western coastal plain and was
formerly used extensively for dyeing purposes. The inventions of chemically produced dyes
destroyed this industry, in which the original company (Flores, Hale & Co.) made large profits.
The company in control at the time of our visit had cattle ranging from Matancita ranch south
to opposite Magdalena Bay, from which visiting warships in the bay and other vessels were
supplied with fresh beef.

From November 16 to 19 we remained at Matancita, working the sumoundiug plain. A
narrow belt immediately along shore has only small herbage and stunted bushes, resembling
that in the vicinity of San Quintin, but back of this the coast plain is covered with a vigorous
growth of giant and several other species of cactuses, Fouquieria peninsularis, and many other
desert shrubs and small trees, sometimes forming a dense jungle.

November 20 we left our pack outfit at the ranch and accompanied Mr. Heney in his buck-
board 10. miles westerly to Soledad Landing, on the lagoon leading to Magdalena Bay, where
we boarded the company’s sloop, Catarina. This lagoon is a riverlike channel from one-fourth
to more than 2 miles wide, which extends along shore about 60 miles north from Magdalena Bay.
It is bordered on both sides by a belt of mangroves and is separated from the sea by a low strip
of sand which forms. the northern end of Magdalena Island. The afternoon of November 22 we
arrived at the settlement of Magdalena Bay, 50 miles south of Soledad Landing. _ (See frontis-
piece.) This small village, on the eastern or inland side of the mountains, near the south end of
Magdalena Island, faces the bay 2-or-3 miles northwest from its entrance. It consists mainly
of a large, well-made two-story residence, with stores and warehouses of the company and a
number of smaller houses occupied by the employees and by Mexican customs officials. During
the continuance of the orchilla industry this place had. a much larger population. At the time
of our visit the main export was a monthly shipment. of sea turtles to San Francisco on the
Pacific Coast Steamship Co.’s boat, which called here regularly. The south end of Magdalena
Island forms a narrow rocky ridge a few miles long rising abruptly from the sea to over 1,500
feet at its highest point and presenting a bold and rocky head at the entrance of the bay.

November 27 Mr. Heney took us in his sloop across the,bay to Margarita Island. We landed
on the shore of Almejas Bay, on the east side of the island, at the site of an old canning estab-
lishment where an unsuccessful attempt was once made to can sea turtle for export. A small
frame house on the beach is all that remains to mark the spot. Immediately back of the landing
the shore is bordered with some small salt lagoons full of mangroves, in which a considerable
colony of man-o-war birds have their nests. The young of the year, though fully grown and on
the wing, were still spending much time about the nests and hundreds of young and adults
roosted there at night. . This place is on the easterly side of the low flat area which crosses the
middle of Santa Margarita, separating the mountainous northern and southern ends of the
island. The western side of this pass forms a low flat, just inside. a belt of high sand dunes
bordering the shore, where fresh water exists within a few feet of the surface. Traces of old
fields and the ruins of a ranch house were evidences that an effort had been made to cultivate
this flat.. Several small groups of date palms growing here were being buried by the advancing
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 14.

 

 

 

 

 

Fic. 1.—Open roadstead steamer-landing on beach at San Jose del Cabo, January, 1906. Crates on beach contain brown sugar in small cakes
for shipment to other points in Lower California and to Mazatlan.

 

 

 

 

 

Fie, 2.—The creeping cactus (Lemaireoccreus eruca) on the Llano de Yrais. cast of Magdalena Bay. These curious plants radiate from

acentral point, and while the parent stem dies the continual growth at the front end causes the separate stems actually to travel
away from the starting point.
Memoirs National Academy of Sciences XVI. First Memoir, PI. 15.

 

 

 

 

 

Fic. 1.—Servatjllo cattle ranch on plain east of Magdalena Bay. Beef for ships in Magdalena Bay is gathcred about this and
neighboring ranches.

 

 

 

 

 

Fig. 2,—Water front at La Paz, January, 1906. ‘Thisis the capital of the southern district of Lower California. It was the point first
occupied by the Spaniards and is the best town in the peninsula.
No.1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 41

sand dunes. We remained on Santa Margarita until December 1, when we returned to Magda-

lena Island. Our accumulated specimens were shipped on the steamer for San Francisco, and

December 5 we started with Mr. Heney on our return to Matancita ranch, but calms delayed our
sarrival until the 8th.

December 9 we resumed our journey on horseback and traveled about 16 miles southeasterly
across the slightly rolling coastal plain to a dry camp in the cactus forest. Thesky was heavily
clouded and we put up our tént between two giant cactuses, so we were sheltered from the rain
that began during the latter part of the night.

December 10 we broke camp in the rain and continued our journey in a more southerly
course. A cold southeast wind drove the rain with such force that we were quickly wet through
and thoroughly chilled, so we rejoiced to reach shelter at Servatillo ranch, 7 miles from camp.
(See Pl. 15, Fig. 1.) This is one of the Chartered Co.’s cattle ranches and consists of two small
adobe houses and a corral by a well on the plain. The altitude here, as well as all along our
course from near Poso Grande, is about 50 feet above the sea. It rained heavily all day and
most of the following night but cleared on the morning of December 11. We then traveled about
15 miles southeasterly over the plain to La Cruz, another small cattle ranch belonging to the
company and occupied by a Mexican family.

' The rain of the last 36 hours had flooded the coastal plain so that all depressions, including
the trail, were full of water, and torrential streams ran in all the ordinarily dry channels. La
Cruz ranch is located on the plain a few miles immediately inland from Magdalena Bay, on
what is known as the Llano de Yrais. In 1871 an extensive but unsuccessful effort was made
by an American company to locate an agricultural colony on this plain. The part of the coast
plain we traversed from our camp north of Servatillo ranch to La Cruz is nearly flat, with a
gentle slope toward the coast lagoon and Magdalena Bay. A belt of sand dunes along the east
shore of the bay was in plain view along this part of the route, while far to the east the broad
plain is limited by a high plateau and range of mountains paralleling the gulf coast and occu-
pying the eastern third of the peninsula. .

We remained at La Cruz December 12 and tried to secure specimens of a small fox reported
to occur in open parts of the plain. December 13 we continued in a southeasterly course over
the plain. About 3 miles after leaving La Cruz we came out of the cactus forest into an irregu-
larly shaped open prairie, about 10 miles long in an easterly and westerly direction and from
3 to 4 miles broad, with its western end near the sand dunes on the east shore of Magdalena Bay.
This is the Llano de Yrais proper, though the name has been used to cover the adjacent plain
as well. This open prairie appears to be a kind of shallow, nearly flat sink and was largely
under water at the time we crossed:it. The preceding day Goldman had set several fox traps
near burrows on the higher parts of this plain, and as we passed we took them up and secured
three foxes. These proved to represent a previously unknown form of the graceful, big-eared
desert fox. Formerly many antelope lived on these plains, but they are now extinct south of
the Vizcaino Desert. Many low places along this part of the trail were covered with broad
sheets of shallow water, in which giant cactuses and other desert plants were standing, thus
giving a grotesque appearance to the scenery. The country gradually became less muddy
and more sandy. Late in the afternoon we entered more rising and slightly rolling country,
over which small fragments of volcanic stone were scattered. About 26 miles from La Cruz
we came to the first definite dry wash or creek bed since leaving Soledad. The trail crosses
this wash at a point where a small clear stream of excellent water rises and flows a short distance
down the bottom. Here, at an altitude of about 100 feet, is located El Sauz, a small cattle
ranch, where we camped. The high bank of the wash, as well as exposed sections of some low
mesas passed on the trail, show that this part of the country is underlaid near the surface by a
thin blanket of volcanic tuff (tepetate). On the Llano de Yrais we passed beds of an extraor-
dinary creeping cactus (Lemaireocereus eruca) which we saw nowhere else. (See Pl. 14.)
December 14 we continued in a southeasterly course about 35 miles to an old cattle ranch
called Agua Colorado, on the banks of the Arroyo Colorado or Arroyo de la Junta, in which
there was a small stream. We camped at an altitude of about 75 feet in an abandoned hut
42 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI

at the foot of a bluff on the northern side of the creek. Another small house on the opposite
bank was occupied by a Mexican family and some cowboys. Our route that day led ‘alternately
across nearly level, and gently undulating, gravelly and stony mesas separated by broad washes
or little drainage valleys with abrupt banks 40 or 50 feet high. -

December 15 the trail led about 10 miles southeasterly across low, rolling hills to the little
village of San Hilario on top of the high northern bank of a small stream of the same name.
This arroyo irrigates some small vineyards, small fields of sugar cane, and other crops. A
few date palm trees and some fan palms, perhaps of two species, were also noticed. A little
wine is made here. Among the hills 3 miles beyond San Hilario we came to the junction of
our trail with the one leading from the old mission of San Luis to La Paz. We had been approach-
ing the basal slopes of the main plateau and mountains of the east side of the peninsula for
the last two days and were now in distinctly rolling country. Three miles southeasterly from
the junction of the trails was an abandoned hut on top of a bluff overlooking the deep Arroyo
Guadalupe, in which there was a small stream flowing over a rocky bed. Beyond this arroyo
the country changes and becomes much rougher and is broken into series of low, narrow hills
and ridges. Back of these toward the mountains to the east were the sharply rising fronts
of low, scattered mesas, the formation of which was similar to that of the hills and ridges we
were crossing and indicated that both were made by the erosion of a great plateau sloping
down from the high range to the eastward. The trail continued in a winding course about
12 miles from Guadalupe through this broken country to Cerro Colorado.’ This latter is a
buffy-colored conical hill standing immediately in front of and evidently weathered from the
lower border of the high interior mesa. We made a dry camp at an altitude of about 800 feet
just above the base of this hill.

The coastal plain we traversed all the way from below Comondu to the vicinity of Agua
Colorado, except the belt close along the seacoast and the restricted Llano de Yrais, was covered
with an irregular but heavy growth of desert vegetation, among which hosts of giant cactuses
and several other cactuses, large and small, Fouquieria peninsularis, and many other desert
and arid tropical species form a low scattered forest.

December 16 the trail led southeasterly from Cerro Colorado directly up a steep escarp-
ment.to.an altitude of about 1,200 feet on the top of a broad rolling plateau covered with broken
lava. From the point where we first reached the top the plateau rose gradually to an altitude
of 1,400 feet, which was the highest. point we crossed between Cerro Colorado ‘and the gulf.
Tis surface i is the roughest and most broken we had seen on any of the leva plains crossed to
this point. This is due in part to its surface being cut by deep, ragged box canyons and their
numerous small tributaries, some draining to the Pacific and others to the gulf. The plateau
rises steadily toward the gulf on the southeast and breaks off in abrupt’ escarpments, forming
cliffs fronting La Paz Bay, and from the crest slopes gradually westerly and ‘southerly to the
broad coast plain of the Pacific. For 20 miles we traveled the roughest trail we encountered
during the entire expedition, to the crest of the abrupt escarpment on the gulf side, at the head
of a large box canyon known as the Cajon de los Reyes. Here the trail led down a steep decliv-
ity, called El Salto de los Reyes, to the bottom of the canyon. Thence we followed its wind-
ing course for 10 miles, between high and picturesquely sculptured walls, to its mouth near
the shore of La Paz Bay. Throughout its course this canyon cuts through an extension of the
high plateau we traversed in the morning. The plateau, which is of sandstone overlaid by
lava beds, ends in a series of high cliffs fronting all the westerly and northerly shores of La
Paz Bay and separated from it by a sandy coast plain 2 or 3 miles broad. A vigorous
growth of giant cactuses, mesquites, and other desert species covers the sandy plain over
which we traveled about 5 miles from the mouth of the canyon to Rodriguez, a little group of
huts forming a milk ranch on the shore of La Paz Bay, where we camped.

December 17 we continued in a southeasterly and’ then northeasterly course for about 22
miles to the town of La Paz, on the eastern side of the bay. (See Pl. 15, Fig.2.) Our route
followed the shore of the large inner bay or estero, a shallow lagoon which has a narrow outlet
into the main bay close to La‘Paz. We were surprised to see so few waders and other water-
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 43

fowl along the muddy flats bordering the lagoon. La Paz is the capital of the southern district
of Lower California and next to the largest town in the peninsula, being second only to Santa
Rosalia, and contained about 5,000 people. It is a beautiful place close to the shore at the head
of the bay and embowered in coconut and date palms, orange, lemon, pomegranate, avocado,
banana, and other trees and shrubs. The country about the town is an arid desert with mes-
quites near the bay and creosote bushes and other desert species farther back. The open plain
slopes gently back from the bay and the sharp peaks of the Cacachilas, rising boldly a few miles
to the southeast, are the nearest mountains. To the southwest the low plain forms a broad
pass between the inner bay and the Pacific side of the peninsula, and thus separates the moun-
tains and high plateaus of the north from the mountain groups to the south. In front of the
town lies the broad sweep of La Paz Bay, bordered by cliffs along its northwestern shore and
with Espiritu Santo and San Jose Islands lying just outside in the gulf.

‘La Paz is the headquarters of the pearl-fishing industry of this region extending from the
Gulf of California south to the coast of Guerrero. The business is in the hands of two or three
companies holding concessions from the Federal Government. Mr. James Viosca informed
me in a letter dated May 4, 1906, that at that period these fisheries were yielding each year
about $250,000 in pearls and $120,000 in mother-of-pear! shell.

We made another shipment of specimens from La Paz and on December 23 continued our
journey toward Cape San Lucas. Our route lay nearly due south along a good wagon road over
a gently rolling and slightly rising plain. In the afternoon we camped at San Pedro, a little
cluster of ranch houses about 18 miles from La Paz at an altitude of about 700 feet. Some
fields are dryfarmed here, but as they are dependent on the irregular rainfall for a crop the
results are uncertain.

_ December 24 we traveled southerly along the wagon road for about 17 miles to Tres Pachitas
a small ranch on the sloping plain at about the same altitude as San Pedro. About 5 miles out
from San Pedro the road forked, a branch turning to the left and leading to the mining camp of
Triunfo. Our route lay over the same rolling plain we had traversed all the way from La Paz,
but we were gradually approaching the outlying foothills of the northern end of the Victoria or
Laguna Mountains. The growth of desert vegetation was very large and abundant along the
last part of this road, and a number of tropical shrubs, including an Erythrina, were seen for
the first time. The luxuriance of the vegetation near Tres Pachitas indicates a much greater
rainfall than nearer La Paz, which is evidently due to the proximity to the high northern end
of the Sierra de la Laguna, where summer rains are comparatively regular.

December 25 we journeyed about 14 miles southerly along the road from Tres Pachitas and
reached another small ranch called Valle Flojo, located about 3 miles from the coast beside a
large dry wash at the bottom of a broad canyon about 100 feet deep. At Tres Pachitas the
road approaches the north end of the Sierra de la Laguna, which forms the northern section of
the Victoria Mountains, and after leaving that place we found the country much more rolling
owing to the series of long rounded ridges which lead northerly from the base of the mountains
and drain on the southern end of the great Magdalena Plain. The road gradually swung around
the north and west base of the Sierra de la Laguna, which rises precipitously several thousand
feet on this front to Valle Flojo. Shortly before we reached this ranch we passed two low flat-
topped hills which form part of a series extending from the road northwesterly for some miles
parallel to the Pacific coast.

The abundant growth of small vegetation and the heavy foliage on shrubs and trees all
along the route from within a few miles of La Paz were evidence of abundant rains during the
past season. Near La Paz conditions are always much more arid than elsewhere along this
route and in consequence the vegetation in that vicinity is smaller and more scanty, but as the
mountains are gradually approached and a few hundred feet in altitude gained, the vegetation
becomes much larger and more abundant. Huge giant and other large cactuses were the most
common and striking elements in this junglelike forest, which varies from 15 to 40 feet high.

December 26 we crossed a continuation of the broken country for about 12 miles southerly
from Valle Flojo to the village of Todos Santos. This is a pretty place, consisting of 30 or 40
44 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

houses at an altitude of about 100 feet on a low bare ridge overlooking a narrow valley which
is the lower end of a broad canyon leading from the west base of the Sierra dela Laguna. Beé-
low the road the valley was bright green with waving fields of sugar cane, and beyond, 2 or 3
miles away, lay the blue waters of the Pacific. A number of tall fan palms about the houses
and cane fields and mango trees along the irrigating ditches added to the attractiveness of the
place. Water for irrigation comes from some large springs in the bottom of the valley. For-
merly the springs were located high up the valley above the road from La Paz, but some years
ago they suddenly became dry and then burst out lower down, where they are at present, thus
necessitating the abandonment of all fields above the road.

For about 10 miles beyond Todos Santos the road continued southerly over low rolling
ridges, which lead down from the high mountains on the east, to Pescadero, a small group of
huts a couple of miles from the sea. These huts are on two low hills overlooking some small
sugarcane fields in the flat between the houses and the sea. Immediately north of Todos Santos
the outlying foothills of the Sierra de la Laguna reach the Pacific coast and, as southward toward
Cape San Lucas where the descending foothill slopes of the Victoria Mountains reach the coast,
they usually end in cliffs along the shore.

December 27, 12 miles south of Pescadero, we came to the San Jacinto ranch, consisting
of a few huts located within half a mile of the sea beside a broad sandy wash in a baylike coastal
valley. The dry wash was bordered near the crossing by beautiful groups of tall fan palms,
their presence giving evidence of water a short distance below the surface. Our route all day
was generally parallel to the coast and within 2 or 3 miles of it, and several low ridges alternating
with small coastal valleys extending from 2 to 4 miles inland were crossed. These coastal val-
leys are mainly flat, but are sometimes broken by low hills or ridges and crossed by sandy
washes coming down from the west slope of the Victoria Mountains. In the afternoon the
trail left one of the valleys and ascended the highest and most abrupt ridge we had yet seen
along this part of the coast, bringing us to Cajoncito, a small cattle ranch located at an altitude
of about 1,000 feet on a bench just below the summit of the hills, about 21 miles southerly
from Pescadero.

All along the route from north of Todos Santos the westerly front of the Victoria Moun-
tains, 8 to 12 miles inland, parallels the coast, and rises with abrupt and inaccessible slopes of
bare rock to a ragged line of peaks from 3,000 to 6,000 feet above their basal foothills. These
mountains and the foothills and ridges leading to the coast are mainly of pale gray granite
resembling that in the high mountains of the northern half of the peninsula, but with their
most abrupt slope facing the Pacific instead of the gulf.

December 28 the trail held in a generally southerly course all day, but was extremely
crooked, swinging down to the coast and then far inland, winding among the roughest and most
broken hill country we had seen since leaving La Paz. About 12 miles by trail from Cajoncito
we turned inland from near the ocean and traveled about 8 miles southeasterly across foot-
hill ridges to Saccaton, a small cattle ranch situated at an elevation of about 1,000 feet and about
5 miles from the sea. All along this part of the route the main foothill ridges descend to the
Pacific, forming bold rocky points and bluffs with broad sandy dry washes in deep ragged can-
yons between them. The vegetation along the coast from Todos Santos to Saccaton is much
like that farther back on the road to La Paz but is smaller and less abundant.

December 29 we crossed a high narrow ridge immediately south of Saccaton, on the far
side of which the drainage flows toward the southern end of the peninsula. Descending this
south slope we followed a series of dry washes down their winding courses through the hills and
finally came out on a long, open sandy slope descending southward to the sea. This slope is
bounded on the east and west by series of granite hills, giving it the form of a valley, down
which broad sandy washes lead to the shore of Cape San Lucas Bay. The slope is covered with
a vigorous growth of large and small cactuses, mesquites, and numerous desert and arid trop-
ical plants. Early in the afternoon, about 15 miles from Saccaton, we reached the village of
Cape San Lucas on the shore of the bay, the end of our long southward journey. (See Pl. 18,
Fig. 1.) The bay is bounded on the southwest by a low granite ridge terminating in two de-
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 16.

 

    
 

Fes
SS

Sate ERE RA REE

 

 

 

Fic, 1.—Jungle of Arid Tropica] and Lower Sonoran vegetation on northern foothill slopes of the Sierra de la Laguna, between La Paz
and Todos Santos. .

 

 

 

 

 

Fic, 2.—Sugar-cane fields at Todos Santos. <A beautiful little valley near the Pacific coast, irrigated from a giant spring.
Memoirs National Academy of Sciences XVI. First Memoir, PI. 17.

 

 

 

 

Fic. 1.—A small coastal plain at E] Pescadero, south of Todos Santos, showing the trail leading south to Cape San Lucas. Victoria
Mountains in the background.

 

 

 

 

 

Fic. 2.—View westward across San Jose Valley at Santa Anitaranch. The Sierra San Lazaro in the distance.
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 18.

eT

 

 

 

 

 

Fic. 1.—Cape San Lucas. The graniteridge forming the point of the capelies on the west side ofthesmallbay. The village of
Cabo San Lucasis back of thesand dunesin the foreground.

 

 

 

 

 

Fia. 2,San Jose del Cabo in 1906. This place had about 1,500 inhabitants. It lies a little back from the shore about 23 miles

easterly from Cape San Lucas near the lower end ofafertile little valley.
No.1} LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 45

tached granite knobs standing boldly out of the sea and forming Cape or Cabo San Lucas. The
village of Cape San Lucas contains about a dozen families who occupy a small group of adobe
houses on a flat just back of the broad sandy beach and only a few hundred yards easterly from
the base of the cape. It was formerly a much more populous and prosperous place. At the
time of our visit the inhabitants were making a livelihood mainly by gathering tan bark from
small trees on the sloping plain back of the town and shipping it to San Francisco. A few
date palms among the houses, the tropical jungle behind, and the blue sea in front unite to
make this a most picturesque spot. We obtained the use of a vacant house and spent a few
days securing specimens from a place of historic interest to American ornithologists through
the work done here in 1859 by John Xantus.

January 4, 1906, we left San Lucas and followed a trail northeasterly along the coast, some-
times on the sandy beach but mainly a mile or two inland. The country was broken by numer-
ous granite ridges, 100 to 500 feet high, extending to the sea from the low southern end of the
Victoria, Mountains, a few miles inland, and separated by deep, sandy, dry washes. About 23
miles from the cape we came to San Jose del Cabo, which is a well-built town of adobe houses
and contained about 1,500 inhabitants. It is located on a low slope on the west side of the
valley and about a mile above the mouth of the Rio San Jose, which reaches the sea here. (See
Pl. 18, Fig. 2.) San Jose Valley, 1 or 2 miles wide, is very fertile, and by means of irrigation
from the river good-sized fields of sugar cane and other crops are cultivated on the flat between
the town and the coast.’ Some parts of the bottom bordering the river were overgrown with
willows and other bushes. About the town were many orange, lemon, pomegranate, banana,
and other fruit trees, and numerous tall and graceful fan palms were scattered about the fields.
The permanent supply of water in the river and the fertile valley lands make this one of the
largest and most prosperous agricultural places on the peninsula. Steamers from and to San
Francisco. and the mainland of Mexico stop here regularly twice a month, the landing place
being on the open beach near the mouth of the river. We remained at San Jose several days
to secure topotypes of birds and mammals in a locality where Xantus, Bryant, and other
naturalists have worked. The night of our arrival (Jan. 4—5) was the coldest of the season,
the temperature dropping close to the freezing point.

January 10 we traveled about 8 miles northerly up the west side of the valley to Santa Anita,
a ranch where we were hospitably. welcomed by the proprietor, F. Pazjk. (See Pl. 20, Fig. 1.)
This place has been visited by various naturalists, including Dane Coolidge, Loye Miller, and J. E.
McLellan, and is the type locality for several species of mammals. San Jose Valley is bounded
on both sides by hills and mountains and varies from a mile in width near San Jose to several
miles at Santa Anita. The west side of the valley is walled in by foothills of the high range of
the Victoria Mountains, and west of Santa Anita the peak of San Lazaro, in this range, rises
several thousand feet and forms a bold and striking landmark. To the east the valley is lim-
ited by the gentler slopes of the Sierra de la Trinidad.

Giant cactuses and other desert plants with a strong tropical mixture form a vigorous and
abundant vegetation all along the bottom of the valley, while groups of tall fan palms at Santa
Anita tower above everything else. (See Pl. 17, Fig. 2.) The top of a long mesa to the north
between Santa Anita and Miraflores is more arid than the valley bottom, and many giant
cactuses and ocotillo (Fouquieria peninsularis) give a resemblance to the vegetation of parts
of southern Arizona. The slopes of all the mountains within sight are bare and rocky.

January 19, having completed our work at Santa Anita, we continued northerly on the west
side of the valley. Our trail lay over a rolling and canyon-cut mesa from 50 to 150 feet above
the river bottom, which had a dry and rocky bed through most of its course. About 12 miles
north of Santa Anita the mesa becomes lower and the trail descends into the river bottom,
where a small stream was flowing, and scattered individuals of a small but handsome live oak
20 to 40 feet high, with small narrow leaves, and drooping branches, made an attractive addition
to the flora. ‘Thence up the valley to Miraflores, about 18 miles north of Santa Anita, these
oaks were rather common along the bottom. Miraflores is an attractive village of 30 or 40

houses at an altitude of about 400 feet on the west side of the San Jose River, just above its
46 MEMOIRS: NATIONAL ACADEMY OF SCIENCES. ° [Vou. XVI

junction with the Arroyo San Bernardo, a small stream coming down a rocky wash from a can-
yon on the east face of the Victoria Mountains, Low broad-topped oaks with.massive trunks
sometimes 2 feet.in diameter are common in the valley at Miraflores, and the scenery reminded
us of places in the live-oak belt of Californian valleys. Miraflores is the point. whence Mr.
Belding made’ his visit to the’ Laguna Motmtaines in 1883, and we secured the services of the
same guide, Francisco Amador.

January 20 we left Miraflores and tauvsed ere across the gently ri rising rocky mesa
up the course of the Arroyo San Bernardo. For 4 or 5 miles the trail led across the west side
of the valley, through an unusually tall and luxuriant growth of giant and other cactuses with
numerous species of shrubs and small trees, mainly tropical species, which form a jungle all the
way to the base of the mountains. There we entered an open canyon, the bed of the stream
being very broad and covered with masses of water-worn bowlders of all sizes, evidences of the
floods which sometimes sweep down these slopes. The trail up the canyon was extremely rough
and broken, leading repeatedly back and forth across and along the bowlder-strewn bottom or
up and down the densely brush-grown sides. A few cottonwoods of a species peculiar to these
mountains with conspicuous white bark were scattered here and there along the canyon, and in
flats along the sides were small patches of corn. About 13 miles from Miraflores we camped at an
altitude of about 1,500 feet at San Bernardo, a small milk ranch where two families were EYng:
A little land is: cultivated here and a small quantity of cheese is made.

January 21 we continued westerly some distance up the canyon and then turned to the right
in a northerly direction and crossed two high sharp ridges and numerous smaller and more
rounded ones forming the eastern slope.of this section of the main range of the Victoria Moun-
tains. From San Bernardo the mountains became much steeper androugher. The more rounded
ridges and depressions were covered with low broad-topped oaks so closely set in many places
that it was difficult to find a passage. These oak thickets were irregularly placed and inter-
mingled with open park-like grassy areas. From the tops‘of the highest ridges, from 4,500 to
5,000 feet altitude, we had fine views of both coasts of the peninsula and the intervening country!
In the afternoon, after traveling about 14 miles in a generally northern course from San Bernardo,
we came to a long abandoned milk ranch called El] Sauz. It was at an altitude of about 4,600
feet in a little flat at the bottom of a steep-sided canyon, where a small stream of clear cold water
and plenty of wood and grass completed the requirements for a beautiful camp. Scarlet-
fruited shrubs (Heteromeles) and a few tall cottonwoods grew along the stream, two species
of oaks and a madrofio formed irregular patches on the northerly slopes, and there were open
grassy areas, mainly on summits and southerly slopes, where the sotol, agaves, and yuccas
were abundant. We remained here for two days to do our first work in the Victoria Mountains,

January 24 we left El Sauz and followed the trail down the slope to another deserted milk
ranch in a pretty flat, in the bottom of a tributary oak-grown canyon. Here the trail turned,
crossed a high ridge to the north, and descended part way on the farther side to a little knoll where
a small stream issues from a side gulch. This place, 8 miles from El Sauz, has an altitude of
about 5,000 feet and is called La Chuparosa, or the Humming Bird. Many deer tracks were
seen along the route and Viosca’s pigeons were common and nesting in the live oaks, The
heart-breaking work up and down the very steep slopes was telling on our animals, so we were
compelled to camp at La Chuparosa.

January 25 we continued northerly, the trail leading along the summits of two high and
extremely narrow ridges which form the crest of this part of the range. Between these ridges
a large canyon heads close to the summit and a small stream, the Arroyo de San Antonio, flows
down the easterly slope to the gulf, passing Santiago village. The point where the trail crosses
this stream is an old camping place called San Antonio. The crest of the second ridge imme-
diately north of San Antonio was so narrow in places that there was little more than room for
the trail, the mountain dropping away precipitously several thousand feet on the west and hav-
ing a very steep slope on the east. North of this sharp ridge the mountain top broadened and
the trail led down a long north slope thickly overgrown with a forest of pinyon pines and small
oaks into the basinlike mountain valley of La Laguna, where we camped at the ruins of an old
No.4] --- LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. AT

ranch in the extreme upper end. This open, nearly flat-bottomed valley is about a mile long
and from one-fourth to one-half as wide with several small brooklets uniting near the upper end
and flowing down the middle. It is situated at an altitude of about 5,500 feet and surrounded
by steep mountain slopes covered with low oak and pinyon forest. Formerly the bottom of the
valley is said to have been occupied by a shallow lake, the lower end of which was eroded during
a season of heavy rains‘and the lake drained. This occurred many years before Belding’s visit
here in 1882, yet some of the people at La Paz and elsewhere in the cape district still believe the
lake exists. Two or three miles by winding trail and half this distance in a direct line westerly
from La Laguna lies La Aguja, a sharp needle peak of granite over 6,000 feet high standing out
boldly as the northwestern end of the Laguna Mountains and presenting a precipitous front
5,000 feet high to the lowlands between La Paz and Todos Santos.

La Laguna Valley, from which the immediately surrounding group of mountains takes the
name of Sierra de la Laguna, is a notable point from the fact that it is rich in species of plants,
birds, and mammals peculiar to the cape district and is the type locality for many of them.
Frazar, Belding, and others have made notable collections here. We remained in this camp
until the morning of January 29, when we started on our return to La Paz. We crossed
the northern rim of the valley and then for three hours descended a long, steep, and nar-
row ridge leading down the north end of the mountain range. The trail was an old one and
apparently had not been used for years. It would have been nearly or quite impossible to
ascend it, owing to washouts along some of the steepest slopes, down which our animals were
forced to slide in a half-sitting position. In many places it was obstructed by trees and bushes
which had grown in or across the trail. This abrupt and continuous descent for over 3,000
feet brought us to a deserted ranch on the bank of a small creek in the flat bottom of a deep
canyon, so densely grown up with bushes and small trees that we had trouble in forcing our
way across. The trail led directly across the canyon and 1,500 feet up the abrupt and rocky
slope on the opposite side to the crest of the ridge. The top of this ridge is very narrow and
on the farther side drops into a deep, rugged box canyon in which there is an abundant growth
of vegetation. (See Pl. 19, Fig. 2.) The trail here turned abruptly to the left and followed“
down the crest of the ridge for several miles, finally bringing us to the bottom, where the can-
yons on the two sides of the ridge, both carrying small creeks, unite. Here we found two men
cutting fan palm leaves for thatch and they both left their work and went half a mile down the
canyon to show us where the trail is forced up on the steep mountain side to the right by the
broken character of the canyon bottom. Up on the mountain side the trail soon led into a side
canyon and down it to the main canyon again, below the rough stretch we had avoided, and
brought us out at El Taraiso, a small ranch containing two or three families. We camped
here at an altitude of about 1,800 feet after traveling about 14 miles of excessively difficult and
broken trails, which were extremely exhausting to our stock.

‘We were now out of the roughest part of the mountains and on January 30 followed a good,
broad trail for 10 miles northeasterly through low hills and ridges to the small mining camp of
El Valle, where an American company was working a small stamp mill and concentrating plant.
Eight miles beyond this in the same direction, along a good wagon road, through open rolling
hill country we came to the mining town of El Triunfo at an altitude of about 1,800 feet among
rounded hills. This was formerly the largest and most thriving mining camp in the peninsula,
but at the time of our visit was working on a much smaller scale. The large reduction plant
and a huge pile of tailings at the border of the well-built little town gave evidence of its former
prosperity. Rain began soon after our arrival and continued all night. The following day,

_ January 31, the rain continued and I took the stage for La Paz, leaving Goldman to follow with
the pack outfit. The distance is about 30 miles on a good wagon road, but owing to the worn-out
condition of the animals Goldman did not arrive until the following day, February 1. This
ended our overland work. From February 1 to 5 we worked about La Paz and on February
6 left that place on a small pearl-fishing boat, with a crew of three men, to visit Espiritu Santo
Island, off the mouth of La Paz Bay. - ;
"'158961°—21—4 :
48 MEMOIRS NATIONAL ACADEMY OF SCIENCES. ‘ [Vou. XVI

February 7 we reached the island, about 20 miles from La Paz, and anchored in front of
a few houses in a small bay on the west side, where a company organized at La Paz was experi-
menting in raising pearl oysters. They had several beds planted and claimed to be getting
gratifying results. Espiritu Santo is a waterless volcanic island covered with rounded hills
and mesas where a nearly black jack rabbit has developed, evidently from the same stock as
the pale form living a few miles away on the adjacent meniead: A few mangroves occupy
a small tidal lagoon at the head of the bay.

February 8 we left Espiritu Santo and sailed centenary to the southern end of Ceralvo
Island, which is waterless and uninhabited. Owing to calms and head winds we did not reach
Ceralvo until February 11. The only good landing and comparatively level Jand on this island
is at its southern end; the rest of its shores are abrupt and rocky.

February 13 we left Ceralvo, reached La Paz on the 15th, and sailed on the steamer for
Ensenada February 19, arriving at that place on the 26th.

February 28 T proceeded to San Diego in a gasoline launch, while Goldman made the
journey overland by stage. The stage road from Ensenada leads northerly along the coast
about 7 miles to Sauzal, a ranch, facing the sea, which was at that time the home of Mrs. Robert
Louis Stevenson. From this point the road led a few miles up Sauzal Canyon, passing through
the fertile Guadalupe Valley. From this valley the road followed up the Canyon del Burro to
the top of a ridge, said to be the highest point on the road between Ensenada and San Diego,
and descended to a little valley called Vallecitos. Beyond this another low ridge was crossed
and then the Canyon de Cancio descended to a ranch of the same name about 55 miles from
Ensenada and the regular stopping place for the night. All hills along this route have rather
gentle rounded slopes; a little farther inland are ridges rising 1,500 to 2,000 feet higher than the
road which are a continuation of the coast range back of Ensenada. The hills along the road
are mainly granite and dotted with bowlders of the same material. Large oaks (Quercus agri-
folia) are numerous in the canyons, but the hill slopes facing the sea are nearly bare. In some’
of the valleys there was a thick growth of arrow weed (Pluchea sericea), but the bordering hills
above 1,000 feet were usually covered with a dense growth of chamiso ‘(Adenostoma).

March 1, Goldman left Canyon de Cancio and traveled about 40 miles, mainly along or near
the Tijuana River, to Tijuana, whence he took a train and joined me at San Diego, where the
expedition ended.

PHYSICAL CHARACTERISTICS.

- Lower California is a long, narrow peninsula, washed on the west by the Pacific and on the
east by the Gulf of California, or Sea of Cortez, as it was named by the Spaniards. The peninsula
extends in a southeasterly direction from the southern border of California, following the general
direction of the west mainland coast of Mexico, and measures about 800 miles long by from 30
to more than 100 miles in width. The peninsula as a whole is an arid desert, a large part of
which is subjected to long-continued droughts.

Lower California, while mainly a desert, exhibits a number of striking contrasts. The tops
of the higher mountains’in the northern and southern ends of the peninsula are covered with
forests. In the north the yellow pine predominates and the mountains are capped with snow
in winter, while on the eastern side the rugged and precipitous slopes rise from the excessively.
hot basal plains, supporting remarkable growths of cactuses, yuccas, and many other desert
plants. In the south-central section great lava-crowned plateaus are gashed by deep-walled
canyons, which are occasionally watered by small streams producing lovely oases where flourish
an abundance of subtropical fruits. The country is everywhere scored by rock-strewn and
sandy washes, ordinarily dry, but marking the drainage courses of occasional torrential flood
waters. The Colorado, one of the great rivers of the continent, flows into the head of. the gulf
and supplies water to irrigate the marvelously rich alluvial lands bordering its delta.

A single main chain of mountains extends from the northern border southeasterly to La
Paz Bay, a distance of nearly 700 miles, and dominates the peninsula. It is an extension of the
mountain system of southern California, and forms a continuous series of ranges varying in
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 19.

 

 

 

 

 

 

 

 

 

Fie, 2.—Vegetation in canyon at cast base of Sierra de la Laguna, showing the cape cottonwood (Populus monticola), fan palms
(Erythea brandegeci), and other species.
Memoirs National Academy of Sciences XVI, First Memoir, Pl. 20.

 

 

Fie. 1.—Santa Anita ranch in valley of San Jose del Cabo.

 

 

 

 

 

Fic. 2.—East slope of Cerro Mayor, the highest peak in the Cocopah Mountains. Thisis a typical desert range,
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 49

length and altitude but lying generally parallel to the gulf coast. This mountain chain forms
the crest of an interior plateau, the northeastern, or gulf front, of which is composed of a series
of high, precipitous escarpments, while the slope to the southwest is long and gradual. The
escarpments of the gulf side appear to be the result of a long series of fault lines, along which
the mountain slopes rise abruptly either from the low basal mesas, from the borders of narrow
coastal plains, or directly from the shore of the gulf. Where basal mesas are present they slope
gently from the foot of the escarpment to the gulf, scored by deep box-walled dry canyons and
terminating, like the mesas near Santa Rosalia, in a precipitous front on the shore. In other
places the mesas, as well as some of the coastal plains along the gulf, are broken by barren hills,
ridges, and low mountains.

The highest part of the main range begins on the northern border near the middle of the
peninsula and extends southeast about 150 miles, swinging gradually nearer the gulf coast,
which it closely approaches near the thirtieth parallel of latitude. This part of the main range
includes the Sierra Jaurez and the San Pedro Martir Mountains, which are the highest of the
peninsula, ranging from 4,000 feet near the California border to more than 10,000 feet altitude
on La Providencia Peak. From near the thirtieth degree of latitude to La Paz Bay the main
mountain ranges lie along the gulf coast wholly within the eastern third of the peninsula.

The continuity of the main chain from the border to La Paz Bay is broken by several passes
varying from 1,200 to 3,200 feet in altitude. The only great break in the mountain chain of
the entire peninsula, however, is that opposite the head of La Paz Bay, where a pass more than
15 miles broad and probably less than 200 feet high at its lowest point completely separates the
main ranges to the northwest from those to the southeast.

Southeast of La Paz Bay there is a reversal of the relations of the mountains to the coast,
and the main range there extends along the Pacific coast, to which it presents a high, precip-
itous escarpment a few miles inland with a basal slope extending to the shore with a long easterly
slope, broken by secondary mountains, between the crest of the range and the gulf. The main
mountains lying southeast of La Paz are second in height only to the ranges near the northern
border, varying from 3,500 to 8,000 feet in altitude. The intermediate mountains, between the
southern end of the San Pedro Martir Range and La Paz Bay, vary in altitude from 2,500 to
4,000 feet, with scattered peaks reaching altitudes of from 5,000 to 6,000 feet. It is of special
interest to note that the high main range southeast of La Paz is of pale gray, rapidly disintegrat-
ing, granite closely similar in appearance to that of the high main range near the northern border.

In the north, where the main chain lies near the middle of the peninsula, it is paralleled along
both coasts by much lower and more broken subordinate ranges. “Beyond 150 miles south of the
border the subordinate ranges on both coasts lose their identity, the one on the gulf shore blend-
ing with the main chain where the latter approaches the coast beyond Point San Fermin, and
that on the west swinging toward the middle of the peninsula south of the Salado River, and
thence southward lying close along the west base of the main chain, with which it merges east
of San Quintin Bay.

In places along the main chain south of the San Pedro Martir Range short offsetting spurs
and ridges occur on either side of the ranges following the same general course and usually
inclosing small valleys or plains between them and the main ranges. Throughout their extent
the ranges are scored by many canyons, especially on the gulf side, in some of which occur small
permanent streams of excellent water.

In a general way, from the northern border southeasterly for a distance of about 700 miles
to La Paz Bay, an irregular plateau of varying elevation occupies the westerly two-thirds of the
peninsula. Its highest elevations are near the top of the main range, sometimes at the very
summit, immediately beyond which descends the sharp declivity that forms the gulf slope. In
other places the upper border of the plateau rests against the westerly flank of the narrow culmi-
nating ridges of the range. The character of the plateau is well shown near the northern border
on the Sierra Juarez, where its summit is formed by the plateau which descends gradually for
50 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Von XVI

miles to the west and on the eastern front drops precipitously several thousand feet. The
prolonged western slope of the San Pedro Martir Mountains and the precipitous eastern slope
are also strong characters of this range, but its greater elevation causes a far more rapid western
slope than usual in the peninsula. From near the southern end of the San Pedro Martir Range
to near Calmalli the plateau is lower but well marked, as it is also in the great lava-capped region
from south of Calmalli to La Paz Bay.

The elevation of the plateau varies from 6,000 feet on the Sierra Juarez, and seven to eight
thousand feet on the San Pedro Martir Range, to 1,000 feet or less near the middle of the peninsula.
In addition to the differences in its elevation, the surface of the plateau is varied by small,
scattered mountain ranges, isolated mountains, hills, ridges, and mesas, as well as by many inter-
vening plains, valleys, basins, and dry watercourses. For some distance along the north coast
the plateau is cut off from the shore by narrow ranges of coast mountains; farther south here
and there it is limited by the inner borders of low, nearly level coastal plains, as by the San
Quintin plain, the Vizcaino Desert, and the Magdalena Plain. Between these coastal plains the
plateau ends along the shore in low, descending ridges, broken hills, and scattered mountain
ridges varying from 100 to 2,500 feet high. These are two unrelated to be considered as forming
parts of a definite range. From near Cape San Eugenio to near Ballenas Bay the ridges and
peaks of the Sierra Vizcaino, lying a few miles inland, form the only definite coastal range from
San Quintin to beyond Magdalena Bay. This short range is completely isolated and appears
to be unrelated to the coastal mountains farther north.

Owing to the width of the western plateau slope the usual view to the east from the Pacific
coast or the coastal hills is over a broad, rising and broken plain which culminates in a high
skyline of serrated peaks or in the long horizontal line of the swelling plateau surface which
cuts off the hills behind. From the east coast the view inland is strikingly different, the pre-
cipitous fronts of the main range rising near at hand or only a few miles inland.

West of the summit of the Sierra Juarez a series of small valleys with scattered, intervening
hills and low mountains occupy the plateau, with small coastal ranges limiting most of its seaward
border. South of this, along the west and south base of the San Pedro Martir Range, the plateau
is far more mountainous and broken, but becomes more undulating and regular as it approaches
the coastal plain of San Quintin.

From the plain of San Quintin and the southern border of the San Pedro Martir Mountains
to the northern side of the Vizcaino Desert, near Calmalli, the high interior of the plateau is
undulating and irregular, varying from 1,200 to 2,500 feet in altitude and largely occupied by a
series of small open plains, including the Llano de los Buenos Aires, the Llano de Santa Ana,
and many smaller valleys. These valleys are bordered and more or less broken with barren
ridges and scattered mountains, while the descending border of the plateau in some places
reaches the Pacific coast as rounded or flat-topped ridges separated by deep, dry washes; else-
where the plateau is limited along the coast by low, rocky ranges and scattered hills. Some
of the hills and mountains at various points in this section of the plateau have a flat, or mesa,
top. Perhaps the most striking of these is the Mesa de San Carlos. To the southeast this sec-
tion of the plateau descends and gradually merges into the low, nearly level plain of the Vizcaino
Desert. North of San Fernando rises the isolated Cerro San Juan de Dios, the highest mountain
on this part of the tableland and said to be the southernmost mountain on the peninsula on
which snow falls.

The major part of the main mountain chain and bordering tablelands from the northern
border to near Santa Gertrudis, southeast of Calmalli in about 28° N. latitude, as well as a large
part of the coastal mountains on both sides in the north, are of pale gray and easily weathered
granite with a few scattered areas of low volcanic peaks and ridges and of other geological
formations.

The main range south of La Paz Bay, as already noted, is on the Pacific side of the peninsula,
and is formed by a series of high granite ridges, culminating in sharp peaks or narrow ridges
from 3,500 to about 8,000 feet in altitude. Everywhere among the granite mountains the
disintegration is very rapid, as evidenced by the vast sloping accumulations of sand, gravel,
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES-~NELSON. 51

and bowlders, often of enormous size, on the plains along their bases and especially at the mouths
of large canyons. Some of the granite mountains of the coastal ranges east of the Sierra Juarez
and San Pedro Martir Mountairis, in one of the most arid parts of the peninsula, are almost
buried in the sloping mass of their own débris and are typical examples of ‘‘buried’”’ mountain
ranges. At the mouths of many large canyons, notably at the east base of the San Pedro Martir
Mountains, enormous alluvial fans or deltas exist, some of them hundreds of feet deep and
several miles broad. These are of still greater length where they descend to the bottom of the
adjacent valleys. They are more fully described in the accounts of the various ranges.

From a little south of the old Mission of Santa Gertrudis to La Paz Bay the main chain of
mountains and the surface of the interior plateau is almost entirely volcanic, overlying an
elevated sandstone formation. These volcanic mountains are extremely rugged, and the south-
western slope of the mountains and the plateau are scored with numerous deep and often rugged
canyons. The surface of the great lava-covered plateau is broken by numerous volcanic hills
and peaks, including many old craters, and descends to the coastal plains which border the
Pacific for hundreds of miles in this region.

The highest of all the volcanic peaks of the peninsula, El Puro Afio, has an altitude of about
6,800 feet and is located at the southern end of the Sierra Santa Lucia; next to this are the Tres
Virgines, situated a little farther north, with an altitude of more than 6,500 feet. The Tres
Virgines are the most recent of the volcanoes, one of the peaks having shown signs of activity
within a few years. In addition to these mountains forming part of the great volcanic region
of the southeast afew other smaller areas occur here and there, notably alongshore at San
Quintin Bay and on coastal islands, including San Martin and the San Benitos, the Sierra Santa
Clara on the west side of the Vizcaino Desert, and the Sierra Pinta near the mouth of the Colo-
rado River.

As already noted, the general slope of the western two-thirds of the width of the peninsula
is westerly or southwesterly directly to the Pacific coast, nearly at right angles to the main
mountain ranges. As a result there are no long drainage courses, the longer of them extending
from 25 to 50 miles. Most of the drainage channels are dry throughout the most of the year,
except for occasional floods, and sometimes during several years in succession.

‘The coast line exceeds 2,000 miles in length, more than half of which is rocky and broken,
consisting either of low bluffs formed by the precipitous borders of tablelands descending from
the interior or the eroded slopes of hills and mountains which often rise from the sea in bold
cliffs and headlands. Much of the eastern, or gulf, coast is more or less closely bordered by the
main mountain chain, giving it a generally more rugged and broken character than that of the
western coast. The southern half of the gulf coast and the many adjacent islands are notable
for their bold and often precipitous shore lines with the contours weathered and sculptured into
strange and picturesque forms, which have been likened by one author to ‘castles, towers,
peaks, and massive lines of fortifications.” The gulf shore, however, is also varied by a number
of coastal plains, mainly of limited extent, such as those at Punta Arena (south of Ceralvo
Island), Ventana Bay, La Paz, Loreto, Mulege, and San Felipe. The northern half of the gulf
coast, from a short distance south of Santa Rosalia northward nearly to San Felipe Bay, is
mainly abrupt and commonly formed either by lines of precipitous bluffs, the fronts of low table-
lands which extend back a few miles to the main mountain ranges, or by rugged lines of hills
and mountains broken at wide intervals by narrow belts of lowland. San Felipe Bay is half-
moon-shaped with lines of low sand dunes about its southern and western shore and a high
rugged hill of black rock on its northern side forming San Felipe Point. Thence to the mouth
of the Colorado the shore line is low and gently sloping. . .
The west coast of Lower California, while less rugged than that of the east, is largely rocky
and broken, being fronted by the wave-worn bases of hills and low mountains alternating with
the precipitous fronts of low table-lands, spaced here and there by broad coastal plains where the
table-lands and mountains retreat to. the interior. The principal coastal plains on the Pacific side
lie back of San Quintin Bay, near Scammon Lagoon, and opposite Magdalena Bay.

 

1 Seemann, Voyage of the Herald, Il, p. 155, 1853.
52 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

The northern part of the western shore line for more than 100 miles south of the boundary
is abrupt and rocky, being fronted mainly by rocky hills and low mountains, forming parts of a
low coast range rising 2,000 to 4,000 feet above the sea. Near San Telmo the mountains swing
back farther into the interior and the immediate coast line is lower, but beyond the San Quintin
Plain it is broken by the precipitous fronts of low mountains and mesas until the northern
border of the Vizcaino Desert is reached. From the vicinity of Cape San Eugenio, the Sierra
Vizcaino parallels the coast a short distance inland to near Ballenas Bay. Thence southward
to within about 75 miles of Cape San Lucas the mountains give way to a low coastal plain ex-
cept from south of San Ignacio Lagoon to opposite Comondu, where the plain is interrupted
by low rocky hills or the borders of low table-lands.

The shore of the southern end of the peninsula from 15 miles south of Todos Santos to San
Jose del Cabo is formed mainly by granite ridges varying from 100 to 500 feet high, which
descend from the mountains a few miles inland and form a series of narrow rocky capes. Be-
tween the capes are small half-moon bays, commonly with white sand beaches about their heads.
Cape San Lucas, the most prominent of these projecting points, constituting the southernmost
tip of the peninsula, is a ridge nearly 300 feet high forming a bold, narrow headland extending
into the sea with outstanding sugar-loaf rocks just off the point. On the shore end this ridge
descends to a beach backed by a belt of sand dunes which separate it from a continuing but
higher ridge extending inland toward the mountains. The Cape ridge forms the western side of
the shallow San Lucas Bay. A sandy plain slopes from the mountains a few miles inland to the
north down toa flat where San Lucas village is located just back of the sand dunes along the
shore of San Lucas Bay. Another jutting granite ridge forms the other side of the bay about a
mile‘east from the cape.

In addition to the considerable number of small half-moon bays around the entire coast
line a number of bays and harbors are much larger, as those at Ensenada, San Quintin,
Vizcaino (the largest of all, from which open the Black Warrior and Scammon Lagoons), Ballenas,
and Magdalena on the Pacific coast and La Paz, Concepcion, and Angeles on the gulf coast.

The gulf shore is without any long, mangrove-bordered lagoons, although about the head of
La Paz Bay and in a few other limited areas mangroves ocour as far north as the village of San
Lucas, a few miles south of Santa Rosalia. A series of long, narrow, mangrove-bordered lagoons
border the west coast from the east side of Margarita Island north to San Jorge and some small
areas of mangroves also occur near Cape San Lucas. The only grassy tidal marshes on the coast
of the peninsula appear to be about the head of San Quintin Bay and lesser ones a few miles
southward, beyond the mouth of San Simon River. :

Salt deposits exist on the coast near San Quintin Bay and especially large ones near the
heads of Scammon Lagoon, near the head of Ballenas Bay, and on Carmen Island. (See Pl.
21, Fig. 1.) Another deposit is located on the shore of the gulf between the mouth of the Col-
orado River and San Felipe Bay.

NOTES ON THE RECENT GEOLOGICAL HISTORY OF THE PENINSULA.

Both shores of Lower California bear many signs of extensive erosion in the form of wave-
worn cliffs and narrow rocky points. In addition most of the coastal islands are similarly eroded
and closely match the neighboring shores in geological structure and often in surface contours,
while the intervening waters are so shallow it is evident that. they were once joined to the shore;
Todos Santos, Cedros, and Espiritu Santo Islands are especially striking illustrations of this
former connection. Espiritu Santo, in front of La Paz Bay, forms a continuation of a point on
the mainland with which it is joined by a submarine ridge covered by only 8 fathoms of water
at the deepest point of the 4-mile channel separating them. The surface contours and structure
of the cliffs on the two sides of this channel show the closest relationship. If the peninsula for-
merly included, as appears probable, most of the bordering islands, it must have been far
broader than at present, but the general outline was not otherwise much, different. Most of the
islands on both coasts are very rocky and hilly or mountainous, but the long northern end of
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 21.

 

 

 

te

 

 

Fic, 1.—Salt bed on east side of Vizcaino Desert, west of San Ignacio. The salt is so pure that it is gathered by natives for domestic use.

 

Ges

 

 

 

 

Fie. 2.—Mud volcano near Volcano Lake, in delta of Colorado River. The mud rises and overflows the crater, as shown by signs
of last ‘‘cruption’”’ in photograph.
Memoirs National Academy of Sciences XVI. First Memoir, PI. 22.

 

 

 

 

Fig, 1.—Waterhole north of Llano de los Buenos Aires. One of numerous
tinajas which store some of the scanty rainfalland render travel possible
in these deserts. (Photograph by Dale Bumstead.)

 

 

 

 

 

Fig. 2.—Mining onyx on Llano de los Buenos Aires. Onyx from this mine, or quarry, has been shipped to
San Diego and distributed to various parts of the United Statcs for many years. (Photograph by Dale
Bumstead.)
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 53

Magdalena Island is a sandy plain, having been built up by sea sand brought in by currents
sweeping along shore.

The connection between the islands and the peninsula within recent geologic time is abun-
dantly indicated by the close similarity in fauna and flora, especially by the bird and mammal
life.

The Tres Marias Islands, lying to the southeastward, undoubtedly formed a part of the
Mexican, mainland. at no very remote period. The sea is comparatively shallow between these
islands and the mainland, being about 150 fathoms in the deepest part along a submarine ridge
extending southeasterly from the islands to the mainland near Banderas Bay. Between the
‘mainland of Mexico and the nearest point of southeastern Lower California as well as between
the peninsula and the Tres Marias Islands lies the 1,500-fathom channel leading out from the
Gulf of California. The plants, birds, and mammals of the Tres Marias are all either identical
with species living on the adjacent Mexican mainland or obviously derived from them. On the
other hand, while a strong infusion of species of plants plainly derived from the Mexican main-
and exists in Lower California, the flora as a whole is more closely related to that of the deserts
of southern California, Arizona, and northwestern Sonora, while the batrachian, reptile, bird,
and mammal faunas are, with rare exceptions, species derived from the mainland to the north.

_ From the present configuration of the land surface it is evident that an arm of the sea once
extended across the peninsula through the broad low pass from the head of La Paz Bay to the
Pacific, thus forming an island of the mountainous region lying south of La Paz.. La Paz Bay
appears to be the remaining part of this channel. When viewed from La Paz or from the west
coast the islandlike appearance of the mountain area to the south is very striking. The existence
of this area as an island could not have been greatly prolonged judging by the uniformity of its
fauna and flora with that of the peninsula farther northward. é

The present contours show that the sea formerly extended inland from the head of Vizcaino
Bay, covering all the broad Vizcaino Desert and probably across to the gulf by way of the low
plateau on which is located San Ignacio, and which extends as a nearly level pass between the
Tres Virgines volcanoes and the north end of the Sierra Santa Lucia. The submergence of the
Vizcaino Desert area must have made a long narrow island of the Sierra Santa Clara and Sierra
Pintada, which together form the Sierra Vizcaino. The trend of the Sierra Vizcaino and of the
mountains on Cedros Island indicate that at one time they may have been united in one con-
tinuous range, probably before the submergence of the Vizcaino Desert.

Many of the other islands about the shores of Lower California evidence by their position
and formation relative to the adjacent shores that they were once the points of capes on the
mainland which have been cut off from the shore and sometimes subdivided into two or more
sections or islets. The Todos Santos Islands, at the mouth of the bay at Ensenada, were evi-
dently once united and formed a continuation of Punta Banda, with which they are still con-
nected by a submarine ridge. The mountainous southern end of the Magdalena Island and the
mountain at Cape San Lazaro on the same island were once separate islands which have been
joined recently by a broad sandy plain built by ocean currents and winds to form the long, narrow
and flat sandy northern three-fourths of this island. The northern and eastern mainland shores
of Magdalena Bay are low and sandy and much of the northern end of the bay is very shallow,
being filled in with sand. This sand formation extends north from the bay along the coast
nearly 100 miles and appears to be supplied by the ocean current alongshore. Santa Margarita
Island appears to have been divided by a narrow arm of the sea into two small mountainous
Gslands. The entire Magdalena plain is so low and nearly level that it appears probable that it
also was submerged at the period when the sea occupied the pass at the southern end of this
plain, near La Paz. From such facts it appears that the peninsula was partly submerged and
then arose slowly with approximately the outline it posseses to-day but considerably greater in
extent. Since then it appears to have been steadily reduced through erosion by the sea.

The apparent exception to this is the northern two-thirds of Magdalena Island, which ap-
pears to be still growing by additional sand brought by the sea, and the gradual increase of land
‘area in the delta of the Colorado River through the enormous deposits of silt carried down by this
54 MEMOIRS NATIONAL ACADEMY OF SCIENCES.: [Vor. XVI

stream from the middle of the Rocky Mountain region. Signs of rapid erosion are everywhere
about the shores of the peninsula and of the adjacent islands.

Dr. Eisen credits the great deposits of bowlders and wash material along the eastern base
of the Sierra Victoria and near San Bartolo to glacial action. After seeing the enormous masses
of huge bowlders and other material carried down by floods and deposited at the mouths of can-
yons along the east base of the San Pedro Martir Mountains, I am inclined to credit the deposits
along the east base of the Sierra Victoria to a similar origin rather than to glacial action. Our
route, however, did not take us near the ‘‘moraines”’ seen by Dr. Eisen between Miraflores and
San Bartolo, the most notable of which at the mouth of the canyon of San Bartolo are known as
the Quebrados de San Bartolo. Nowhere during the entire expedition did I see any signs of
glacial action, and these ‘‘moraines” appear to require further examination to determine their
origin.

The partly buried mountains in the northeastern part of the peninsula, the huge bowlder
masses on the higher parts of the Sierra Juarez and San Pedro Martir Mountains are striking
evidence of the rapid erosion now in process. Other indications of this action are presented
by the flat-topped mesas existing at various points on the western slope from near Cape Colnett
south to the middle of the peninsula, including the Mesa de San Carlos, which rises high above
the surrounding levels and is several miles across. In addition numerous lower table-topped
elevations rise from the higher base level of the interior. The granite formation which occupies
most of the northern half of the peninsula as well as the granite range south of La Paz are notable
for their rapid disintegration, as shown by the vast quantities of granite, sand and gravel; which ,
cover the slopes of these mountains and the adjacent plains.

MOUNTAINS OF THE NORTHERN INTERIOR AND MIDDLE GULF COAST.

Sierra Juarez and San Pedro Martir Mountains.

The northern part of the main mountain chain of Lower California is formed by the Sierra
Juarez-San Pedro Martir Mountain groups. These mountains are a direct continuation of
the San Jacinto Range of southern California. They are composed mainly of, the Same pale
gray, rapidly disintegrating granite, and their eastern front, overlooking the desert, is a gigantic
rocky escarpment thousands of feet high, while the western slope is long and gradual. In all
of these characters these ranges bear a close resemblance to the southern end of the Sierra
Nevada in the Mount Whitney section.

The Sierra Juarez and San Pedro Martir Mountains have so much in common they may
be considered as forming a single range about 175 miles long, extending from the northern
border of Lower California southeasterly, lying a little to the east of the middle line of the
peninsula. About a hundred miles south of the border the range is divided by San Matias Pass,
a deep notch which has an altitude of about 3,200 feet. (See Pl. 2, Fig. 2.) The part, of the
range north of this pass is the Sierra Juarez (the northernmost section of which is locally often

called the Laguna Hanson Mountains or the Sierra Juarez) and that to the south the, Sierra
San Pedro Martir, or San Pedro Martir Mountains, which latter have a length of about 75 miles.
The crests of both sections are in the form of long, high ridges but little broken. by project-
ing peaks, and varying from 4,500 to about 9, 000 feet i in general altitude. ;

The east base of these mountains rises from desert valleys or plains, the surfaces of which,
broken by scattered mountains, .extend to the lower Colorado River and Gulf of California,
the plains varying from about 1,500 feet down to sea level. From these plains, often with
only a narrow belt of low foothills or a talus slope, the precipitous rocky front, of the Tange rises
boldly 3,000 to over 8,000 feet. The front of this great escarpment, is scored at ‘intervals by
narrow, rough-walled canyons which have carved a few outstanding peaks or ridges - from the
bald rock walls. La Providencia Peak, in the San Pedro Martir Mountains, is the only one of
these frontal peaks which rises prominently above the general elevation of the mountain mass
behind it.

 

‘ Proc. Calif. Acad. Sei., V, p. 754, 1895,
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 55

The western slope usually descends gradually through a broad belt of rounded foothills
to bordering elevated valleys and tablelands beyond which are coastal mountains. That part
of the summit rising above 4,500 feet in the Sierra Juarez and about 6,000 feet in the San Pedro
Martir Mountains is occupied by the only yellow pine forests in Lower California.

A wagon road follows the Sierra Juarez from Campo, just north of the border, to a few
miles south of El Rayo ranch, where it descends to the valley on the west side of the range.
Another wagon road crosses the range from Alamo Plain to Mexicali on the Colorado Desert
through a narrow pass northeast of the old mission of Santa Catarina, where a small Indian
village still exists. This is approximately the route by which Pattie and his companions reached
the west coast from the Colorado in 1827. Another branch of the wagon road from Ensenada
leads across Alamo Plain and through the range at San Matias Pass to San Felipe Bay on the
gulf. Below are more details concerning the main sections of this northern part of the main
range.

Sierra Juarez.

These mountains consist of a single main ridge, varying in height from about 4,000 feet
at the northern border to about 6,000 near Hanson Lagoon, and thence down to 4,000 or 5,000
feet near the southern end, where it finally descends rapidly to San Matias Pass. The northern
half of this range is known locally as the Sierra del Pinal or Laguna Hanson Mountains. The
highest part of the main mountain mass is near Hanson Lagoon, but the highest peak, Cerro
Colorado, in the southern part of the range, near Santa Catarina Mission, has an altitude of
about 6,677 feet.

The summit of the northern part of the range has a width of only 3 or 4 miles across the
widest part. It forms a long, narrow, gently rolling plateau, which breaks down abruptly
into bare granite cliffs on the east and descends through gently sloping foothills on the west.
Along our route up the west slope from La Huerta to El Rayo ranch the trail traversed a distance
‘of about 18 miles over such gradual slopes and rounded ridges that we seemed to be crossing
an undulating plain until near the summit, when the higher ridges became somewhat more
marked. This slope is covered with gravelly decomposed granite varied by low projecting
granite ridges and large bowlders, and is drained by shallow washes and gulches.

_ The highest part of the plateau summit, near Hanson Lagoon, is broken by many small
‘granite’ ridges, knolls, and piled masses of titanic granite bowlders, 50 to 300 feet high, with
pretty parklike forested basins and flats between.

The old Laguna Hanson ranch, or El Rayo, as it is now called, lies on the summit of the
high part of this range northeast of Ensenada. Southeasterly from this ranch the summit of
the range becomes lower and more broken, and the wagon road along the summit from the
northern border swings down the west slope, which becomes much more abrupt until through-
out the last 20 miles or more north of San Matias Pass it rises abruptly from Alamo Plain and
presents lines of cliffs and steep slopes forming the flanks of long table-topped a upon
one of which rises Cerro Colorado.

The Indian village near the site of the old mission of Santa Catarina lies at the ae base
of this steep escarpment about 20 miles northwest of San Matias Pass. This latter is a deep,
narrow, troughlike notch, where the eastern and upper end of Trinidad Yalley: cuts through
the mountains to the low desert valley on the east.

' Water is scarce all along the summit of the Sierra Juarez. The Hanson ee (or Laguna
Hanson), about half a mile'across, is the largest of several small shallow lakelets lying on the
highest part of the range northeast of Ensenada. A few very small streams rise at the heads
‘of gulches near ‘the top of the west slope and flow down to the lower border of the foothills in
the northern two-thirds of the range but are fewer in the southern third. Water is still scarcer
on'the higher parts of the east slope than on the west, but occurs lower down in most of the large
canyons near the desert. El Tule Arroyo and Arroyo Grande flow out from the east base of
Cerro Colorado and are the largest and most permanent streams on the east side of this range.
56 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI

San Pedro Martir Mountains.

The San Pedro Martir Range i is the jit init and most notable mountain group t in Lower
California. It is almost entirely of pale gray granite and extends about 75 miles southeast
from San Matias Pass, in direct continuation of the Sierra Juarez. The range rises rapidly at
first in a series of narrow, crested ridges, reaching an altitude of over 7,000 feet on Salvatierra
Peak a few miles south of San Matias Pass, and increasing in altitude and width for about 20 or 25
miles to its culmination at an altitude of over 9,000 feet on the main ridge and to approximately
10,126 feet on La Providencia Peak, the highest point on the peninsula. Beyond this the range
gradually descends and narrows until it merges and is lost in the crest of the Matomi section of
the broad interior plateau as detailed below. The broadest part of the summit of the range
measures about 6 or 8 miles across opposite La Providencia Peak and about twice as much
across the base. The peak is situated in the middle of the east front of the San Pedro Martir
Mountains and is a sharp, narrow rocky crest rising like a cock’s comb on the end of a high
granite ridge carved from the rocky front of the mountain by the tremendous gorge of Canyon
Diablo. This canyon rises on the west side of this crest and extends first northerly and then
to the east and has the form of a gigantic ragged gash in the mountain. The crest extends
northerly from its base, the peak being at the northern extremity. The east side of the crest
and peak forms part of the east wall of the mountain range, facing the desert.

La Providencia Peak is also called Santa Catalina, Calamahue, and La Encantada on various
maps. It is a bold landmark for more than 100 miles in all directions across the deserts east
of the range and becomes especially notable in winter when covered with snow. Owing | to the
high intervening ridges it is not visible from the land on the Pacific side, but it is said that its
showy summit may be seen for more than 100 miles at sea.

The eastern side of the San Pedro Martir Mountains forms the highest and boldest part of
the abrupt escarpment already described. Here the bare granite cliffs rise directly ; from the
basal valley in a series of great rock slopes for from 4,000 to over 8,000 feet. All. ‘the higher
crests and summits of the range, including La Providencia, are located close along the top or
only a short distance back of this high escarpment.

The main part of the east front of the range is scored by four great canyons, all of which
contain small permanent streams. Beginning with the northern one, these are La Esperanza,
* 10 miles south of San Matias Pass; El Diablo, 6 miles farther, south (at the base of La Provi-
dencia); La Providencia, about 6 miles beyond; and Santa Rosa, about 8 miles still farther
south. These canyons are notable for the great alluvial fans which spread over the plains from
their mouths. Southeasterly other canyons occur in the front of the high mountain plateau
which continues the end of the range as described below.

The western slope of these mountains, while much more gradual than the eastern, is high
and abrupt with sharply ridged foothills, especially in the northern half. The middle of the
range is there scored by numerous canyons which contain the headwaters of four main streams
draining to the Pacific—the San Rafael, San Telmo, Santo Domingo, and San Simon. All of
these have permanent water in parts of their courses, but the Santo Domingo is the only one
which is said to flow continuously to the sea. The higher parts of these mountains are scantily
supplied with water in the form of a few small springs and insignificant streamlets.

The summit of the San Pedro Martir Mountains, when viewed from either the east or west
side, appears almost as regular and unbroken by projecting peaks as the Sierra Juarez. La Provi-
dencia Peak is hidden from the west by the main ridge, about 9,000 feet high, which lies west

of the peak, across Canyon Diablo, and when the peak is viewed from the east this same ridge
forms a background for it, thus causing it to stand out in comparatively small relief. In travers-
ing the broad central part of the mountains we found the summit subdivided by many rocky
ridges and cross ridges with long rocky slopes, often partly inclosing beautiful grassy parklike
basins bordered by pine forests. Some of these parks are hanging valleys or plateau-like benches
scooped out of the mountain summits at various elevations, with their open sides to the west or
south and backed by granite ridges a thousand feet or more high. The intervening slopes are
commonly covered with masses of granite bowlders often of great size and here and there form-
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 57

ing high, tumbled piles of rock. They are usually so devoid of soil that only widely scattered
trees or bushes grow on them. The highest and largest of these basins is called Vallecitos and
lies at an altitude of about 8,000 feet near the westerly base of the high ridge which forms the
main crest of the mountain, west of La Providencia. To the southeast of this is the little
mountain basin of La Encantada, and below, on the westerly slope, at 7,000 feet, lies La Grulla
meadow, one of a connected group of beautiful little grassy valleys at the base of bare granite
ridges with some small ponds and a small stream along the bottom. The borders of this valley,
covered with a thin forest of yellow pines, is dotted with many huge outcropping granite knobs
and bowlders, which typify much of the scenery on the summit of these mountains. Farther
southeasterly the rolling summit of the mountains becomes lower and narrower until about 10
miles from La Grulla the oval grassy basin of Santa Rosa, about 4 miles long and 2 miles broad,
covers almost its entire width and drains through canyons leading down both slopes. A mile
southwest of Santa Rosa and nearly a thousand feet below is the benchlike Santo Tomas Valley,
and still farther down the same slope, at an altitude of about 5,000 feet, is the bench where the
ruins of the old mission of San Pedro Martir are located. About 12 miles southeast of Santa
Rosa, at an altitude of approximately 6,000 feet, lies the little valley of Santa Eulalia, marking
the southern end of the yellow pine forest.

Southeasterly from Santa Eulalia the range loses most of its mountainous character and
gradually descends to the crest of the high interior plateau, along the eastern border of
which extends a continuation of the precipitous escarpment fronting the desert and to the
west and south a long slope leading toward the Pacific. This plateau has an altitude of from
about 3,500 to over 4,000 feet along the crest and forms the southeastern extension of the
San Pedro Martir Mountains, uniting them with the higher mountains inland from Point San
Fermin, which here approach the gulf coast and continue the main range southward. Three
large canyons cut into the east front of this plateau draining toward the gulf have fresh water
along their courses before they descend into the open plain. The northernmost is Agua Cali-
ente, southeast of this Paral, and still farther is Matomi. The last appears to be the most
important. According to Heller, the Arroyo Matomi is in a wide level valley from 1,500 to
2,000 feet above sea level and walled in by cliffs 1,500 to 2,000 feet high. Here the northern
pinyon pines (Pinus parryana) appear to reach their southeastern limit in the peninsula. Matomi
Peak, with an altitude of about 4,100 feet, lies on the east front of the range near Matomi
Valley and appears to be the highest elevation on this part of the peninsular backbone.

A trail used by the early Spanish missionaries leads from the desert at the mouth of Santa
Rosa Canyon up across Santa Rosa Park and descends the west slope by Santo Tomas ranch
and the old mission of San Pedro Martir. This is the first trail leading across the mountains
south of San Matias Pass. Another trail leads up the northwest slope to Vallecitos and still
another up the west slope to La Grulla Meadows. The east slope is inaccessible north of Santa
Rosa. An irregular, and in places difficult, trail leads along the top of the range from Valle-
citos through La Grulla to Santa Rosa and thence to Rosarito ranch near the southern end of
the mountains. Owing to the high abrupt slopes on all sides of this mountain group and the
broken and rocky character of its summit, it will never be practicable at any reasonable cost
to build a wagon road through it. A considerable part of the existing trails are so steep they
are difficult to traverse with pack animals.

The summit of these mountains has the most extensive yellow-pine forest on the peninsula,
but it is made up largely of medium-sized and small trees and scattered in many separate tracts
difficult of access one from another. Two long ditches earry water from the western slope
down to the Socorro and other placer mines in the foothill belt.

Sierra de Santa Ysabel and Sierra de Santa Maria.

As already noted, the San Pedro Martir Range ends at the mese of Matomi, northwest
of Point San Fermin. The next section of the main chain forms an offset and continuation of
the main range to the north but is much nearer the coast. It is approximately 60 miles long

1 Publications Field Columbian Museum, Zool. Series, III, p. 203, 1903.
58 MEMOIRS NATIONAL ACADEMY OF SCIENCES. © [Vou. XVI

and divided near the middle by El Tule Arroyo, which heads on the border of the tableland
near the Onyx mine. The section of these mountains north of El Tule Arroyo is known as
the Sierra de Santa Ysabel and that to the south, extending. to Calamahue Arroyo; as the
Sierra de Santa Maria. This range is mainly rather low and broken, especially in its southern
part, where in places the crest of the peninsular divide is formed by the eastern border of the
tableland, which drops away precipitously 1,000 feet or more on the east front, to the low
basal plain of Santa Maria sloping to the gulf and dotted with hills and low ridges.

The Arroyo de Santa Maria, on which is located the ruins of the old mission of Santa
Maria, cuts through the southern part of this range not far from the site of the old placer min-
ing camp of San Francisquito. The ruins of this camp are on the coastal plain about 2 miles
east of the high escarpment marking the front of the main range. Between the camp and
the gulf are scattered low hills and ridges.

It is a well-formed and rather narrow range of abruptly rising, barren rocky ridges broken
by peaks from 2,500 to. over 3,000 feet high in its northern part and becoming higher near the
southern end, where, southwest of the old Santa Maria Mission, a mountain which I have
named Ugarte Peak, rises to about 4,753 feet. It is a typical barren rocky desert range with
small streams in parts of El] Tule and Santa Maria Canyons. Elsewhere water is extremely
scarce and mainly limited to small tinajas.

The west front of the range forms a high abrupt wall, with serrated skyline, limiting the
eastern border of the desert Llano de los Buenos Aires. The base of the eastern slope descends
through low tablelands and hills to the shore of the gulf, where it ends mainly in low rocky
bluffs. Ei Tule Canyon cuts entirely through the range, its upper end heading on the border
of the Llano de Buenos Aires not far from Onyx, where onyx has been mined for some years and
hauled to the Pacific coast. The eastern border of this interior plain and the low foothills
bordering the western base of the range in this section are formed of nearly horizontal beds
of limestone. :

Northwesterly from San Francisquito the range brwalia down and the eastern border of
the tableland forms the top of the main divide. Immediately south of this pass in the range
rises a small group of rounded mountains having an extreme altitude of about 3,500 feet.
Through the southeastern border of this small range Calamahue Arroyo cuts its way. The
extreme head of the arroyo extends through the range to the eastern border of that part of the
main central plateau known as the Llano de Santa Ana, thus diverting to the gulf a little of
the former drainage of the Pacific slope. The pass through the mountains at this point has
an altitude of about 1,200 feet. a
Sierra de Calamahue y San Luis.

Southeast of the extreme head of the Calamahue Arroyo rises the highest mountain in
this section of the peninsula, Dos Picachos, with an altitude of about 5,440 feet. A few scat-
tered trees were visible on the higher northern slope of this otherwise barren mountain. Dos
Picachos forms a part of a high, abruptly rising, but round-topped, range which exténds from
the upper part of the Calamahue Arroyo east to the gulf and southeasterly parallel to and'in a
broad belt along the gulf coast to a, point southeast of Yubay, where there is a break in the
range opposite Remedios Bay. This range is mainly of granite and has a few outlying hills and
ridges of the same mat terial. It rises abruptly . like a wall from the table-land along the eastern
border of the Llano de Santa Ana. The Calamahue Mission was located near the lower end
of Calamahue Canyon, close to the northwestern end of this mountain range. Throughout
most of these mountains water is extremely scarce and the slope rocky and barren, but little
vegetation being able to gain a foothold.

Sierra de Las Flores.

The Sierra de Las Flores extends from the north side of Los Angeles Bay to the south side
of San Rafael Bay. It is a broken range bordering the coast in a generally northwest to south-
east course. Throughout its length scattered peaks rise from 2,000 to nearly 4,000 feet. This
range and the southeastern part of the Sierra de Calamahue follow the gulf Goat much more
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 59

closely than the main chain farther north. Both ranges are excessively arid and barren in
character. Many miles of their west front, overlooking the eastern border of the interior
table-land, appear to be naked rock, in sharp contrast to the adjoining surface of the table-land
with its abundant desert vegetation. _

The Arroyo de Santa Ysabel has cut back through the main range and drains a small part
‘of the eastern border of the central table-land into San Rafael Bay, thus making a low pass
similar to that formed by Calamahue Arroyo.

Immediately south of San Rafael Bay the mountains are close to the coast and reach an
altitude of a little over 1,000 feet. To the southward the main range gradually trends farther
away from the coast and incieases in height to culminate in Goldman Peak, about 5,000 feet
in altitude, located close to 28° of north latitude, northeast of the old mission of Santa Ger-
trudis. Still nearer the mission lies Bryant Peak, with an altitude of nearly 4,500 feet. On
the table-land west of this part of the main range are numerous scattered knobs, hills, and
mountainous ridges, including the isolated group of small mountains in which the mining
camps of Calmalli and Campo Aleman are located, with other low hills and ridges still farther
west on this plain.

The Sierra Calmalli is a short, bare, and rocky group of desert mountains rising abruptly
from the plain a few miles north of the small mining camp of Calmalli. From the Sierra
Calmalli northwest to Yubay several abrupt, isolated mountains and small ranges are scattered
over the table-land. These mountains are mainly of the pale gray granite of which most of
the mountains are made from the northern border all the way down the main chain and on
the adjoining part of the table-land to the west, as well as to the shore of the gulf on the east.
This granite formation comes to an abrupt termination near Santa Gertrudis and is replaced
by heavy beds of sandstone overlaid by great lava beds and other volcanic material, forming
an elevated plateau extending thence southward to La Paz Bay.

MOUNTAINS OF THE GREAT VOLCANIC PLATEAU.

The great volcanic region of Lower California forms an elevated plateau, marked near its
northern point, northeast of the old mission of Santa Gertrudis, by Goldman Peak (altitude,
5,000 feet), and extends thence southeasterly to the west side of La Paz Bay. The plateau covers
most of the middle eastern parts of the peninsula within the limits named, including a distance
of about 300 miles in length by from 20 to 30 miles in width. The basal or underlying forma-
tion of the plateau is composed of sandstone, upon which are superimposed the volcanic moun-
tains, with lava beds of varying thickness covering practically all of its surface. The slope of
the plateau descends westerly and southerly from near its eastern crest down to the border of
the Vizcaino Desert and Magdalena Bay along the Pacific coast. A broad westerly extension
of the plateau separates these two great coastal plains by a series of low foothills, ridges, and
mesas. The surface of the plateau varies considerably in character and elevation from near
sea level to above. 3,500 feet altitude. Over much of its extent it is extremely rough and
broken and for hundreds of square miles its surface is buried in masses of loose, shattered frag-
ments of lava. The drainage of the plateau to the Pacific is mainly through deep, sharply cut
canyons, in which are located some of the most fertile and populous oases of the peninsula,
among which are San Ignacio, La Purisima, and Comondu, all three of which were the sites of
missions established early in the history of the peninsula.

“The mountains in this area are all volcanic in character and constitute a part of the main
shite of the peninsula. They consist of small mountain ranges, mountain groups interspersed
with ridges and isolated peaks, many of the latter being volcanic crater cones. Together these
mountains constitute a ragged crest along the eastern side of the plateau. The eastern front
of the plateau rises abruptly from near the gulf coast, ora short distance inland, to altitudes
varying from a few hundred to more than 4,000 feet. Where there is an intervening basal
plain it is more or less broken by foothills and ‘low mountains. The principal mountain groups
forming the main chain in this region, beginning at the northwest and extending southeasterly,
60 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou XVI

are the Sierra de San Francisco, Volcanes de las Tres Virgines, Sierra de Santa Lucia, Sierra
de las Palmas, Sierra de Zacatecas, and Sierra de la Giganta.

The Sierra de San Francisco.

This range of volcanic peaks and craters, forming a section of the main chain, rises from
the top of the plateau a little southeast of Santa Gertrudis, with Goldman and Bryant Peaks
as its outlying northern summits. It is formed by a series of.ridges, peaks, and craters, extend-
ing southerly to within a few miles of San Ignacio, where the range ends at the Cerro Colorado,
and the slope descends to the low table-land pass which crosses the peninsula by way of San
Ignacio and between the Tres Virgines and the northern end of the Sierra Santa Lucia. The
Sierra de San Francisco as a range is wholly separated and distinct from the other groups of
volcanic mountains known as the Volcanes de las Tres Virgines and the Sierra Santa Lucia,
the volcanic activity forming these groups of mountains apparently occurring independently
and at different periods.

The Sierra de San Francisco rises from the higher middle part of the northern section
of the plateau, which is here from 15 to 20 miles wide, with an undulating surface reaching an
altitude of at least 3,500 feet along the middle and eastern sides, above which rise the high
elevations of the mountains. The eastern slope of the plateau presents a high, bold escarpment,
the general appearance from the gulf being that of an abruptly rising range of mountains,
of which the Sierra de San Francisco and outlying peaks for a certain distance form the summit.

The general surface of this section of the plateau is entirely covered with heavy beds of
lava overlying the basal sandstone formation. Over large areas the lava is shattered into
large angular fragments which render passage almost impassable, even to a man on foot. This
section of the plateau is drained by a number of deep, rugged canyons rising high up near the
eastern border and descending westerly to the border of the Vizcaino Desert. Practically all
of the larger canyons have water somewhere along their course, but none carries a permanent
running stream to the edge of the desert. The most important of these canyons is that of
San Pablo, near the northern border of the plateau, in which, near its upper end, occurs a small
permanent stream which sinks in the sand just below the point where was located the old
mission of San Pablo. (See Pl. 9, Fig.2.) A small amount of water occurs here and there in
tinajas, in some of the other canyons, and in a few of: them small streams flow down short dis-
tances high up toward their heads.

Further details concerning the character of the section of the plateau along the western
side of the Sierra de San Francisco will be found in the narrative of our expedition.

Voleanes de Las Tres Virgines.

The Three Virgins are three high, black, volcanic crater peaks made up of black lava and
other volcanic material. They rise boldly in an islandlike group 12 or 14 miles from the Gulf
of California on the northern side of the lava plain which extends northeasterly as a low pass
from the Vizcaino Desert by way of San Ignacio across the plateau to the gulf coast. The
three peaks vary in altitude from about 4,400 to over 6,500 feet and form a strikingly con-
spicuous landmark when viewed from the gulf. These volcanoes were the most recently active
of any on the peninsula, the last eruption reported having been in 1746. They are said to still
send out vapor and gases from crevices near the summit and to contain deposits of sulphur.

Sierra de Santa Lucia and Sierra de las Palmas.

Southeasterly from the Tres Virgines and about 10 miles from the gulf shore rises abruptly
the volcanic peak of Santa Lucia, with an altitude of about 3,000 feet. This volcanic cone
forms the northern end of a range of continuous and closely set volcanic peaks extending in a
southwesterly course toward the middle of the plateau and culminating in La Bandera Peak,
with an altitude of 5,202 feet. Beyond this, near the middle of the plateau, rises a group of
high cones and peaks, known as the Sierra de las Palmas and culminating in El Puro Afio (5,828
feet), the highest volcanic peak south of the Tres Virgines. Southeasterly from this group
of mountains occurs another group of volcanic peaks known locally as the Sierra del Valle.
No.1] | LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 61

The Sierra Santa Lucia is a sharply ridged, ragged range of mountains about 20 miles
in length and of comparatively recent origin. It is formed by a series of sharp volcanic peaks
and knife-blade ridges of bare lava almost wholly devoid of vegetation except on its lower
slopes and in the bottoms of canyons. Its western slope descends sharply to the table-land
plane which extends toward San Ignacio, and the sea slope is even more abrupt. The crest
of the range where we crossed it south of Santa Lucia Peak is a thin ridge only a few yards
wide. From the eastern base of these mountains to the shore of the gulf extends a low table-land
forming an extension of the plateau through the pass which lies between the Tres Virgines and
Santa Lucia Peak. This table-land gradually slopes easterly to end on the shore in long lines
of cliffs and bluffs 75 to 200 feet high. It is drained by narrow, abruptly walled canyons.

Everywhere along our route through the Santa Lucia Mountains we passed over and
among extended beds of black lava devoid of the slightest covering and some of them with
the lava cooled in fantastic forms, with a surprisingly recent appearance. In other places the
lava beds were shattered and broken. Water and vegetation were extremely scarce throughout
these mountains, which were the most desolate and forbidding we saw during our expedition.

The great lava covered table-land on the west base of the Sierra Santa Lucia extends also
along the entire west front of the Sierra de las Palmas and the Sierra Zacatecas, becoming
gradually higher south of the Santa Lucia Range. The part of this table-land southerly from
the Sierra Zacatecas sloping westward and southward is scored by deep canyons which end
on the border of the Vizcaino Desert.

Near the mouth of Arroyo Pabellon the peninsula becomes narrow and its width is almost
entirely occupied by the mountainous plateau, the lower western border of which approaches
or quite touches the Pacific coast, thus forming the southeastern boundary of the Vizcaino
Desert and the northern border of the Magdalena Plain.

The eastern front of the Sierra de las Palmas, which is practically continuous with the east
front of the Sierra Santa Lucia, is more abrupt than the western slope and is separated from the
gulf by the low Llano de San Bruno. ‘Water as well as vegetation is extremely scarce throughout
these mountains.

Sierra de Zacatecas.

This rough and broken group of volcanic mountains forms a broad belt along the west side
of Concepcion Bay from near Mulege to the southern end of the bay and inland to include the
Cerro de Zacatecas and Cerro Guajademi. It forms the southeastern end of a long crescent of
volcanic ridges and peaks, beginning with Santa Lucia and extending in crescentic form, with
the bow to the west, to Guajademi. The Sierra de Zacatecas rises abruptly on its eastern side
and is extremely ragged and broken all along the crest, especially about the peaks of Zacatecas
and Guajademi. To the west its flank rests on the plateau and is but little less abrupt and
rugged. | . :

The Arroyo del Potrero heads on the northern slope of Guajademi Peak and drains down
through the little town of Mulege to Concepcion Bay, while the great canyon of the Arroyo de
San Jose heads on the southern slope and immediately across a narrow pass from the extreme
head of Potrero del Arroyo. The long peninsula forming the eastern side of Concepcion Bay is
covered by an abruptly rising mountain ridge about 2,500 feet high, which descends to the sea
on all sides except the south, where the range descends and ends on a low plain, giving it an
isolated islandlike appearance.

Sierra de la Giganta. .

Southeasterly from Concepcion Bay, a little north of 26° of latitude, the high eastern crest
of the plateau is marked by a few peaks along its summit which parallel the gulf coast a few
miles inland, while a series of detached ridges and peaks more closely border the gulf shore.

Twelve or fifteen miles northeast of Comondu rises the isolated Cerro de la Giganta, with
an altitude of almost 5,800 feet. This peak on the middle western slope of the central plateau
rises boldly as a huge rocky pinnacle, forming a notable landmark visible for a great distance
along the western side of the peninsula. This mountain, with Encinos Peak located 2 little
62 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI

farther to the northwest, forms the northwestern end of the Sierra de la Giganta, which extends
as a broad mountain belt thence southeasterly to the gulf. and follows the coast thence more or
less closely to La Paz Bay. ‘The range consists of a series of volcanic peaks and ridges super-
imposed on deep beds of sandstone. It is rather broader near the northern end and decreases
southward until it becomes a narrow belt along the eastern border of the plateau.

The broadest and most impressive part of the great lava-covered plateau is located between
San Ignacio and Comondu. It is an area of tremendous former volcanic activity. ‘The under-
lying sandstone is heavily capped by great beds of lava and the surface of the undulating and
broken plain is dotted with craters and over great areas is covered with beds of bare lava, some-
times in almost unbroken sheets and again in a sea of shattered fragments. Deep canyons
cut their ragged way through this dark lava plain, in the two largest and most notable of which
lie the oases of La Purisima and Comondu. (See Pl. 13, Fig. 2.) In these two canyons small
streams flow for miles down their bottoms but elsewhere water is scarce. Vegetation is scanty
on the top of the plateau but abundant in the bottoms of all the canyons.

Southeasterly from a line drawn east and west across the peninsula through the old mission
of San Francisco Xavier to San Luis the plateau retains about an equal breadth, its abrupt
and mountainous eastern border near the gulf coast, while its western border descends gradually
to the eastern edge of the Magdalena Plain. Southeast of San Luis the plateau gradually
narrows with the lessening width of the peninsula, and its surface is less and less broken with
ridges and peaks, until on the west side of La Paz Bay it becomes a gently inclined table-land
with a series of high cliffs along its eastern front bordering the bay and an undulating slope to
the west. The surface of this slope is much broken by irregular lava beds and scored by canyons.
The high cliff front of the plateau around the north and west sides of the bay is entirely of this
sandstone worn into numberless picturesque contours and scored by several deep canyons.

The highest mountain in this range south of La Giganta is El Mechudo, at the southern end
of the range, near the north end of La Paz Bay, with an altitude of about 5,000 feet. From
near this point the range decreases in altitude southward and is soon lost in the high slope of the
plateau, the entire surface of which descends gradually and merges into the pass extending
across the peninsula at the head of La Paz Bay.

MOUNTAINS OF THE CAPE REGION SOUTH OF LA PAZ BAY.

As already noted, an arm of the Magdalena Plain crosses the peninsula to the head of La
Paz Bay, thus forming a broad low pass completely separating the mountains of the northwest
from those southeast of the bay. A line drawn east of south from La Paz to Todos Santos on
the Pacific coast would mark the approximate southeastern limit of this plain. Thence to the
east and southeast the southern end of the peninsula is mainly mountainous.

Compared with the main chain to the north the mountains south of La Paz Bay have their
formation and positions relative to the coasts completely reversed. Thus the Sierra Victoria,
which here forms the southernmost extension of the main peninsular chain, and like the San
Pedro Martir Range is composed of pale gray granite, borders the Pacific coast from a few
miles inland from Todos Santos to Cape San Lucas and presents its boldest and most precipi-
tous slope to the west. The lower and more unimportant mountain groups in this section of
the peninsula fill the interior and border the gulf coast.

Sierra de la Victoria.

The Sierra de la Victoria is the continuation of the main peninsular range south of La Paz
Bay. It is high and narrow, rising boldly from a low base level, with many precipitous and
rocky slopes to numerous sharp peaks in linear series varying from 5,000 to about 8,000 feet
in altitude. Its rugged and precipitous character gives the range a notable individuality which
impresses one from every point of view. At the northern end the range begins about 12 miles
northeast of Todos Santos and extends a little east of south more than 50 miles to the extreme
southern end of the peninsula near Cape San Lucas. On the west a belt of small plains, low-
sloping mesas, and foothills from 8 to 12 miles broad separate it fromthe shore of the Pacific
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 63

south about to Mount Santa Genoveva. Thence to Cape San Lucas the western foothills
become much higher and really. form a broadening of the range and descend to form precipitous
fronts along the shore of the Pacific. -At the southern end of the peninsula the San Jose Valley
separates the southern end of the Victoria Range from the mountains bordering the gulf coast.
Farther north the Sierra Victoria is connected with the mountains of the gulf side by a lower
ridge and series of rounded hills which form a divide at the head of San Jose Valley.

Both slopes of the Victoria Range are very abrupt, but the western is much more precipitous
than the eastern side. The entire range is almost wholly granite, which on the west is almost
bare rock from base to summit. The northern end of the range rises abruptly like a gigantic
headland, overlooking the Magdalena Plain to the north and west. The extreme northwestern
point of the range is formed by La Aguja (The Needle) Peak, a sharply pointed granite pinnacle
with an altitude of 6,100 feet projecting like a buttressed spur from the main mountain mass.
This peak presents a series of pale gray cliffs and precipitous rock slopes about 4,000 feet high,
fronting the plain on the north and west, and is a conspicuous landmark from La Paz and, other
points on the plain. Similar high bare cliffs and slopes of gray granite extend around the north
end and all along the west face of the range, resembling the most precipitous slopes on the
east front of the San Pedro Martir Mountains. Mount Limantour, with an altitude of about
6,300 feet, forms the extreme northern end of the high part of the range, but foothill ridges
extend a few miles farther. Mount Troyer, with an altitude of about 5,500 feet, is the southern-
most high peak, and south of this the range becomes steadily. lower ‘through a series of hills and
small ridges to the coast near Cape San Lucas. The cape itself is a small outlying granite
ridge representing one of the terminations of the range.

Santa Genoveva, the highest peak, a bare sharp pinnacle close to the middle of the range,
has been estimated by Dr. Eisen to reach an altitude of about 8,000 feet, but has not been accu-
rately measured. The next highest peak is Mount Porfirio Diaz, a little over 7,000 feet high,
near the northern end of the range. The remainder of the high peaks vary from about 5,000
to a little over 6,000 feet.

_. Both, slopes of the range are scored by parallel series of deep ragged canyons heading
high up near the crest and forming precipitously-walled gashes down the slope. These canyons
have an easterly or westerly course, depending on the side of the range, and are separated
by. high sharp buttresslike ridges, leading from near the mountain tops down to the basal foot-
hills. On the western side the slope of the main range is more abrupt and the main canyons
shorter, but all extend down through the foothills to the sea and are separated by long, parallel
foothill ridges, which in most instances also descend to the coast. On the east side the slope,
while abrupt, is much less precipitous. From the northern half of the range deep canyons
lead easterly through the foothills and bordering mountains to the gulf. The southern half
of the range has a shorter slope on the east side than it has farther north and the main canyons
on, this side descend to the San Jose Valley. That outlying part of the southern section of the
range known as the Sierra de San Lazaro is narrow and cut into high, rocky peaks and sharp
ridges. There is no continuous crest throughout this part of the range, nor is there any extended
moderately level area.

It may be noted by a glance at the map that the main canyons on the two sides of the
entire range head almost opposite one another. In the southern half the erosion has cut through
the crest and made deep passes with altitudes varying from 3,000 to 4,000 feet at the heads of
the canyons, thus dividing the range into a series of high, sharply sloping ridges. These ridges
have nearly parallel east and west courses at right angles to the axis of the range and culminate
midway in high, bare granite peaks. In the middle of this range, owing to the precipitous
slopes of the ridges, it is practically impossible to follow near the top in traveling north or
south, but a detour must be made around the bases or middle parts of the separate ridges.

The only comparatively level elevated area in this section is north of Mount Troyer and easterly
from El Taste where Dr. Eisen found some small flats, one of which; on the banks of a small
stream, contained about 30 acres. The drainage of all the east slope of the Sierra de San Lazaro
is into San Jose Valley. oe

158961°—21——5
64 MEMOIRS NATIONAL’ ACADEMY OF SCIENCES. [Vor: XVI

The northern part of the Sierra de la Victoria from and including the Cerro San Francisquito
to Mount Limantour, and known locally as the Sierra de la Laguna, has a high and fairly con-
tinuous but extremely narrow crest. Here deep canyons head ‘opposite one another on the east
and west slopes, the same as along the range farther south, but the divides between the peaks
are not deeply cut, so that the range has a generally higher appearance and is not so definitely
separated into a series of long east and west ridges which extend across both slopes as in the
Sierra San Lazaro. The largest elevated area in the entire range is at the northern end, where
La Aguja, Mount Limantour, and Mount Porfirio Diaz form a triangular group 4 to 5 miles across,
inclosing a small basinlike valley known as La Laguna. From the valley comes the name Sierra
de la Laguna applied to this end of the range. The crest of the range south of La Laguna is
extremely narrow, in places being barely wide enough to permit a foot trail along the crest, the
slopes dropping down precipitously for several thousand feet into great canyons on the west. The
west slope is too steep for trails, but the east slope is easier, although so steep that the few trails
are difficult to ascend with saddle and pack animals. The ridges between the canyons are also
narrower and sharper on the west side. The main canyons head close up against the crest on
both slopes of this part, of the range, and on the east side many steep subordinate ridges and
spurs occur in the upper parts of the canyons, rendering travel extremely difficult. The east
slope « of the Sierra de la Laguna drains easterly through several nearly parallel’ canyons which
unite in the foothills and lead through outlying hills and mountains to the gulf. A dividing
ridge and series of low rolling hills north of Miraflores connect the east base of the Sierra de la
Laguna with the mountains near the gulf shore.

La Laguna Valley is a small, flat-bottomed basin about a mile or more long in an easterly
and westerly direction, and half a mile wide, located at an altitude of about 5,500 feet in the
middle of the triangle formed by La Aguja, Mount Limantour, and Mount Porfirio Diaz at the
north end of the Victoria Range. It is surrounded by steep slopes rising from 100 to 500 feet
above it, the highest slopes being on the south and west. Several little streams flow out of
gulches at the head of the valley and unite in a small channel down the middle of the flat. The
bottom of the valley is said to have been formerly occupied by a shallow lake, from which the
name is derived, but a flood cut down the lower end and drained it. The lake, if it ever existed,
had long been dry when Belding visited it in 1882. The bottom of the valley has a scanty growth
of herbage, but the surrounding hill slopes are overgrown with a low forest of pinyon pines and
oaks, the only mountain forest in the peninsula south of the San Pedro Martir Mountains. Itisa
beautiful place and has an attractive camping ground by the ruins of some old cabins at the upper
end where a potato ranch was established long ago, when the mines at Triunfo were in 1 active
operation.

The Sierra de la Victoria has the greatest precipitation of any section in the southern ‘end
of the peninsula and is the source of practically all of the surface and underground waters in the
region south of La Paz Bay. La Laguna has been visited by numerous naturalists and is the
type locality for a number of species of birds and mammals.

Sierra de la Trinidad.

East of the San Jose Valley the Sierra de la Trinidad rises abruptly as a low, barren moun-
tain group, the highest peak of which reaches, an altitude of a little more than 2,800 feet. The
group is made up of some flat-topped and some pointed elevations, forming a compact oval
mountain mass bordered on the south and east by the gulf, on the west by San Jose Valley, and
on the north in part by the gulf and in part by a series of low hills near Santiago, which form a
broad and comparatively low area between it and San Bartolo and other higher mountains to
the north. These mountains are said to be of limestone, formation. We did not visit them.

From San Bartolo Peak northwesterly, to the point forming the east side of La Paz. Bay
extends a short range apparently without any general name but of rugged and broken formation
so rocky and, barren that at a distance it appears to be without vegetation. Along the line of
these mountains is a series of peaks from 3,000 to over 4,600 feet high nearly paralleling the gulf
coast and beginning with San Bartolo on the southeast and ending with the Cacachilas Peaks a
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 65

few miles southeast of La Paz. This group consists of four sharp, extremely barren and rocky
peaks, varying from about 3,500 to a little over 4,200 feet in altitude. These mountains are on
the border of the Magdalena Plain surrounding La Paz and are the mountains nearest that place.
From the Cacachilas northwesterly the range descends rapidly to a low rocky and hilly ridge
forming the point of the narrow cape east of La Paz Bay. This short and broken range presents
its most abrupt and rugged front to the gulf coast, where its slopes sometimes rise directly from
the sea or are separated from the shore by a narrow belt of sandy lowland. The gulf coast from
near San Jose to the entrance of La Paz Bay is mainly bold and rocky in character. Dr. Hisen
reports that the mountains northwesterly from San Bartolo Peak are mainly of stratified red
volcanic rock.

From Santiago north to San Antonio Peak the gulf coast hills and mountains are connected
with the Sierra de la Laguna by low foothills and intervening ridges, but the two ranges form
distinct mountain groups of different geological structure. Northerly from San Antonio Peak
the westerly front of this part of the gulf coast range, of which the Cacachilas form a part, faces
the end of the broad Magdalena Plain.

COASTAL MOUNTAINS OF THE PACIFIC SIDE.

As previously stated, the only definite coastal mountain ranges on the Pacific side of the

peninsula are in detached sections separated by coastal plains of varying extent. Beginning
in the north the first of these extends from near the international border to the northern end
of the Plain of San Quintin. Next follows the section reaching from near El Rosario to Santa
Rosalia Bay, and last the Vizcaino Range southwest of the Vizcaino Desert. The hills and low
mountains between San Ignacio Lagoon and San Jorge Landing are an extension of the interior
plateau and can not be considered as belonging to distinctively coastal mountain groups. The
mountains along the Pacific coast southeast of Todos Santos to Cape San Lucas also belong to the
main central system. ‘The hills and ridges between the Vizcaino Desert and the Magdalena
Plain are the lower border of the table-land; the mountains of Magdalena and Margarita Islands
are islandlike isolated groups and the mountains along shore near the southern end of the peninsula
belong to the main range of the interior. The three distinctly coastal ranges named above
average from 1,000 to 2,500 feet in altitude and reach their greatest elevation of nearly 4,500
feet inland from Ensenada. North from San Ignacio Lagoon the coastal mountains of the
Pacific side are immediately bordering or within a few miles of the shore and occupy the lower
slopes of the great interior table-land. From near the California boundary south to Cape Col-
nett they form a rocky front, sometimes rising precipitously several hundred feet, broken here
and there by small bays and the mouths of canyons coming down from the interior; similar
conditions prevail alongshore from Rosario to Santa Rosalia Bay.
- The coastal mountains rise abruptly on the lower slopes of the central plateau and are
usually higher than most of that part of the plateau lying between them and the base of the main
chain to the east, thus inclosing interior plains and valleys. Scattered among the coastal
mountains are hills and mountains with flat-topped summits, forming mesas or table moun-
tains reaching an extreme elevation of about 2,500 feet and a length of about 15 miles and
breadth of 6 or 8 miles on the great San Carlos Mesa north of Santa Catarina Landing. Other
notable mesas are Table Mountain, not far from the northern border; the high mesa near Cape
Colnett; the broad-topped area just mentioned north of Santa Catarina Landing; and others
north of Playa Maria Bay and in the Vizcaino Range. It has been stated that these high
islandlike table mountains, with abruptly walled sides towering above and overlooking the
present plateau, are the remains of a great elevated table-land which once covered all the penin-
sula but has been eroded to the present level, except for these remaining evidences. The fact
that the many drainage ways from the present interior plateau lead down great canyons to the
sea through the higher walls of the coastal ranges indicate that these channels were formed
when the surface level of the interior was much higher than at present.

More details concerning the western coastal ranges of the peninsula are given farther on.
66 MEMOIRS NATIONAL ACADEMY OF SCIENCES. ' [Von XVI

The Northwestern Coastal Ranges.

At the northern border and for a few miles southward the fa¥ieland bordering the west
side of the main interior range descends to the Pacific coast. Beyond this to the southward,
in a belt some miles in width along the coast, rises a series of hills and low mountains, the most
notable at the northern end being Table Mountain (2,275 feet); inland from Descanso Point.
The coastal mountains gradually increase in altitude to the southward and the rocky front
alongshore becomes bolder.. Gradually on the inland side of the range, northeast of the old
mission of Guadalupe, a definite mountain crest is formed which extends for more than 75
miles in a southerly course, passing east of Todos Santos Bay and losing its identity south of the
San Telmo Arroyo. On this ridge rise the highest elevations in the northwestern coastal range,
including Three Peaks (4,432 feet), northeast of Guadalupe, and Las Animas Peak (about 4,900
feet) northeasterly from Santo Tomas. From the Salado River the main part of the coast
range, as a broken continuation of the granite ridge just described, trends gradually. more
easterly away from the coast and merges into the high, broken, outlying spur‘of the San Pedro
Martir Mountains northeast of San Quintin Bay. South of Trinidad Valley the base of the San
Pedro Martir Mountains and the coast range are connected by series of Jove foothills and ridges
until the coast range. loses its identity. as described above.

The eastern slope of the coast range is short and forms the western boundary of the interior
plateau on which is located a series of small elevated plains and valleys, notably the San Rafael
Valley, Alamo Plain, and Trinidad Valley, which lie between the coast mountains and the west
base of the Sierra Juarez. The western slope is much longer and characterized by descending
foothills and low mountains, which reach the shore and form a rocky coast line almost con-
tinuously from near the California border to near Cape Colnett.

South of Ensenada, as the main coast range trends farther inland, its western slope has a
broadening belt of foothills and ridges, including several small basal valleys, to the west of which,
parallel to the coast, is a rugged but lower range with a few peaks reaching an altitude of 2,000
feet or more. This secondary range closely borders the coast from Punta Banda on the south
side of Ensenada Bay to within 8 miles of Cape Colnett, where the mountains descend to a plateau
several hundred feet high which extends to the cape. Near here the coastal plain of Hamalu
comes in, followed to the south by the Plain of San Quintin, the base of the mountains lying
back from the coast except for a few low, narrow spurs extending to the coast. between the

lains.
The southeastern end of the coast range, northeast from San Quintin, merges into a high

and broken mountain group forming a western spur of the San Pedro Martir Mountains, in-
cluding two series of rounded ridges reaching extreme altitudes of nearly 4,000 feet. This
mountain group is 12 to 15 miles broad and on the west sends down long platéaulike slopes to
the eastern border of the San Quintin Plain. To the south slopes of these mountains merge
directly into the great central plateau which descends from the southern end of the San Pedro
Martir Mountains to the Pacific coast south of San Quintin.

Along the coast for a short distance north and south of San Quintin Bay, including San
Martin Island, is an isolated voleanic area of lava beds with low crater peaks and volcanic
hills. The Five Peaks are small, sharp crater cones which from the sea are notable landmarks
for San Quintin Bay.

The northwestern coast range has the benefit of many fogs and fairly regular winter rains,
hence water is more abundant in springs and small streams in the canyons of the west slope
than elsewhere in Lower California. In addition to their own sources of water supply, the
drainage from all the west slope of the high Sierra Juarez and San Pedro Martir Mountains is
carried through canyons traversing the coast range. These canyons form deep rocky gorges,
especially south of Ensenada, and carry more water than exists in any other area of the same
size in the peninsula.

The slopes of the mountains facing the sea from Cape Colnett northward are generally
devoid of trees and shrubs, but are green with grass and other small vegetation i in winter. On
slopes sheltered from the northwest sea winds and on sides of canyons back from the sea there
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 67

are scattered areas of live oaks and an abundant growth of shrubs. On the high seaward slope
of the mountain immediately northwest of Ensenada a small area of California juniper forest
occurs.

A wagon road leads from San Diego southward along the coast range to Ensenada and
thence down the coast by San Quintin to beyond El Rosario; another leads from Ensenada
back through a canyon in the range to San Rafael Valley, and other interior points, and across
to San Felipe Bay, on the gulf. Several little valleys located between the higher parts of the
coast range and the sea support small farming communities. Notable among these are the
valleys at Ensenada and at Santo Tomas, the latter the site of one of the old missions.

From the southern end of the plain of San Quintin south to a short distance beyond Rosario
the interior plateau descends directly to the coast through a series of gently rolling hills and
ridges which form an abrupt and rocky coast line. Beyond this to the border of the Vizcaino
Desert, beyond Santa Rosalia Bay, a broken coastal range almost everywhere shuts off the
plateau from the sea. These coastal mountains are made up of irregular and broken ranges
of hills, mountain peaks, and ridges. The most notable of these is the Mesa de San Carlos,
a broad table-topped mountain a short distance from, and parallel to, the coast between San
Carlos and Santa Catarina Landing. This great table-topped mountain rises to an altitude of
from 2,000 to 2,500 feet, the summit being about 15 miles long in a northwesterly and south-
easterly direction by 6 miles broad. Beyond this the next most important section is the Sierra
Columbia, which parallels the coast on the western and southwestern side of San Andres Valley.
This is a sharp-ridged range rising abruptly from the plateau to an altitude of about 2,500 feet.
In many places groups of low hills border the shore line, scattered higher peaks rising to varying
altitudes from 1,000 to about 2,500 feet. From the shore of Santa Rosalia Bay the coastal
mountains extend southeasterly, coming to an end on the interior plateau nearly midway
between Santo Domingo and Calmalli. The shore line from a few miles north of Santo Domingo
is occupied by a coastal plain forming the northern point of the Vizcaino Desert. The coastal
mountains between El Rosario and the end of the range on the plateau near Calmalli are
extremely rugged and barren and almost without surface water with the exception of a limited
number of tinajas. Underground water occurs at a shallow depth in a number of the larger
canyons or drainage ways.

Vizcaino Range.

_ The small mountain groups known as the Sierra Santa Clara and Sierra Pintada form a
single low range about 100 miles long, which Mr. Waldemar Lindgren has named the Vizcaino
Mountains. The range begins near Cape San Eugenio and extends southeasterly to near the
head of Ballenas Bay, throughout its course lying parallel to the coast and a few miles inland
on the southwest side of the Vizcaino Desert. The broad low desert on one side and the sea
on the other isolates these mountains and gives them, when viewed from a distance, an island-
like appearance. The relative situation, general appearance, and trend of the mountains on
Cedros Island make it evident that they were once a part of the Vizcaino Range, as has been
suggested by Mr. Lindgren.

The Vizcaino Range is mainly volcanic and is separated into two small groups by an open
pass bordered by flat-topped ridges. The Sierra Santa Clara, at the southeastern end of the
range, is a short compact group of ragged volcanic peaks 2,000 to 3,600 feet high with basal
lava beds. The Sierra Pintada is a longer and narrower group with the higher peaks near the
middle and having about the same altitudé as the Santa Clara group. The range rises abruptly
from the Vizcaino Desert on the east and has a gentler slope on the Pacific side, where it descends.
to the coast through a series of spurs and foothills. The Sierra Pintada is less rugged and
broken than the Sierra Santa Clara and much of the southeastern part of its crest and flanking
hills are table-topped. The Vizcaino Range is extremely barren, almost the only water being
in a few hidden “ tinajas”’ and still rarer small springs. Elephant wood and a few other desert
species form the principal part of the scanty vegetation.

 

“1 Proe. Calif. Acad, Sci., III, 3d sor., pp. 25-33, 1890.
68 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

Dr. Eisen states that fogs and high winds are common in these mountains. . This author
also adds (what is apparent to anyone having a general view of this district): that when the
Vizcaino Desert was submerged by the sea the Sierra Santa Clara and Sierra Pintada formed
part of a long group of islands of which Cedros, Natividad, and San Benito are the existing
representatives.’

In 1893 and later placer gold was mined by dry washing on the western slopes of the Sierra
Pintada; water was found in a number of wells near which was located the mining camp of San
Andres, a few miles from San Roque Landing. The mines were completely abandoned years
ago and this part of the peninsula has since remained without permanent occupation, but has
been visited by occasional prospectors and other wanderers. The commonly used trail to these
placer mines is from San Ignacio, by way of San Angel, and thence across the desert through the
pass between the Sierra Santa Clara and Sierra Pintada.

COASTAL AND OTHER MOUNTAINS OF THE NORTHEAST.

A low, sloping plain with scattered desert mountain ranges covers all of the northeastern
part of Lower California between the abruptly rising east base of the Sierra Juarez and Sierra
San Pedro Martir to the lower Colorado River and shore of the Gulf of California, and extending
from the northern border south to near Cape San Fermin. This plain is one of the hottest and
most arid districts in all Lower California and forms the southern extension of the Colorado
Desert. The surface of the plain is interrupted by five short, narrow ranges, four of which form
a broken coastal chain extending in an irregular, sinuous line from the California border south-
eastward approximately parallel to the course of the Colorado and of the shore of the gulf.
One of the ranges, the Sierra de la Tinaja, is a spur from the Sierra Juarez extending northeasterly
into the plain. The four ranges forming the distinctively coastal chain are the Sierra Cocopah,
Sierra Pinta, Sierra San Felipe, and Sierra Santa Rosa. All are typical barren desert mountains
rising abruptly from the surrounding plain and having very little surface water, mainly in a
few tinajas, or natural rock tanks, in the bottoms of canyons. On the east side of the Sierra
de la Tinaja are the Arroyo El Tule and Arroyo Grande, which rise on the eastern base of the
Sierra Colorado, and have surface streams of clear water for several miles down their courses,
the largest being in the Arroyo Grande.

The Cocopah and Pinta Ranges are surrounded by a low desert plain but little above sea
level. The Sierra San Felipe and Sierra Santa Rosa have their eastern bases on plains but little
above sea level, but on the western side the basal plain is in places 1,200 feet above the sea.. As
a whole, these small ranges, like the main chain, have a shorter and more abrupt eastern slope
and a more gradual western one. They do not appear to be connected with any definite moun-
tain range north of the border and toward the south are cut out by the main peninsular chain,
which approaches the coast opposite Cape San Fermin. These ranges vary from 500 feet to
nearly 4,700 feet in altitude.

Sierra Cocopah.

This range lies near the northern border of the peninsula immediately west of the lower
Colorado River and extending in a northwesterly to southeasterly course rises islandlike from
a sandy plain only a little above sea level. Owing to this low base level with its high rocky
slope the Cocopahs are an unusually striking group of mountains, the southern part of the
range rising to an altitude of more than 3,100 feet. (See Pl. 20, Fig. 2.) Signal Mountain, with
an altitude of 2,400 feet, rises as an isolated peak on the plains near the California border as an
outlying point of the range. The main group of these mountains begins in a peak about 1,000
feet high southwest of Mexicali and a high, narrow, serrated range extends thence southeasterly
about 50 miles to a point near the head of tidewater not far from the mouth of the Colorado River.
The Cocopahs are of much-eroded gray granite with the narrow backbone marked by a series
of sharp peaks. In the period when the Gulf of California extended north over the Imperial
Valley to the Salton Sea the Cocopah Range was unquestionably a great rocky island. At the

 

1 Bull. Amer. Geog. Soc., XXXII, p. 420, 1900.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 69

west base.of this range there still remains a depression | below sea level containing a saline lagoon
known as Laguna Maquata or Laguna Salada, which at times of exceptionally high water
receives drainage from the Hardy River through a low pass at the southern end of the Cocopahs.

A low volcanic peak, called Cerro Prieto, is located a short distance from the east base of
the Cocopahs near the northwestern border of Volcano Lake. The flood plain of the Colorado,
which is almost wholly overflowed during seasons of high water, extends to the eastern base along
the southern part of the Cocopahs. At present the main channel of the Colorado flows through
Volcano Lake and the Hardy River along the eastern flank of these mountains. Water occurs
in a few tanks on both slopes of the Cocopahs, and Palm Spring, the only living water in the range,
is found on the middle of the west base. The range is extremely rocky and barren of plant life.
Sierra Pinta.

The Pinta Mountains, about 20 miles in length, lie aiaitig west of the extreme head of the
Gulf of California, with almost as low a base level as that of the Cocopahs. They are separated
from the Cocopahs by a low gap more than 10 miles in width through which at times of flood
water flows into the Laguna Maquata. These mountains are mainly volcanic in character,
have a northeasterly to southwesterly direction, with their southern end separated by a low,
narrow pass from the northern end of the Sierra San Felipe. They are lowest near the northern
end and rise to an altitude of about 2,200 feet in the south. Three low passes cut the range in
the north. The range is formed mainly of a reddish volcanic rock, to which it owes its name.
The only water in the range is that in a few small rock tanks or tinajas. To the west of the
southern end lies the Arroyo Grande, in which a stream of water flows down from the main
mountains of the interior,-and the well-known water hole, Tres Pozos, which has been a favorite
camp for many hunting parties in that region, lies on the desert to the west of its northern end.

‘Along the coastal plain from the southeastern side of this range extends a-series of scattered
hills to the vicinity of San Felipe Bay. Between these hills and the coast the plain is but little
above sea level, but rises gradually toward the interior.

Sierra San Felipe.

This range begins opposite the southwestern end of the Sierra Pinta and extends about 30
miles in a southeasterly direction to the low Buena Vista pass, separating it from the northern
end of the Sierra Santa Rosa. The range forms a high, narrow ridge of granite broken by sev-
eral rounded peaks and scored by canyons along its flanks. Its northern end rises abruptly
to the highest altitude of the range, about 4,600 feet, on Cerro Borrego. The Sierra San Felipe
extends nearly parallel to and about 12 to 18 miles east of the north end of the San Pedro Martir
Mountains, from which it is separated by the San Felipe Valley. The northern end of the range
is about 18 miles inland from the gulf, but its easterly trend brings the southern end to within
about 10 miles of the shore. As already noted, the coastal plain between these mountains and
the gulf i is dotted with outlying scattered hills. Water in tinajas is extremely scarce, and the
rocky slopes are very barren of vegetation.

From the high crest of the San Pedro Martir Mountains we had a fine view over the top of
the Sierra San Felipe, and it was evident that some of the outlying spurs and ridges as well as
the northern end of the main ridge are excellent illustrations of ‘buried’? mountains. The
granite of all the ranges in this region is easily weathered and extremely friable. The resulting
fine sand and gravel has been washed down the mountain side in such enormous quantities that
gradual and continuous slopes extend up their bases. In the present range a continuous slope
extends from the bottom of San Felipe Valley, on the west side, up to the extreme head of some
of the large canyons, the slopé broadening at the canyon mouths thus appearing like huge deltas.
The base of the mountains between these deltas is much lower than at the canyons. Where
the mountain crest becomes lower, near its northerly end, some of the more scattered peaks
become more and more buried until a number of them appear as a series of small, isolated, rocky
points rising in'a linear series, like little islands on the high, smoothly rounded ridge formed of
the disintegrated sandy material into which the mountain has degenerated, and which forms an
70 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

unbroken slope from the summits of these rounded ridges to the bottoms of the ecjeren wauley
or plain.

Sierra Santa Rosa.

This is the southermost part of the irregular chain of desert mountains beginning with the
Cocopahs on the north and forming the coastal range in the northeastern part of the peninsula.
It is a narrow granite range with rounded summits beginning a few miles inland from San Felipe
Bay on the south side of Buena Vista Pass and extending southeasterly parallel to the coast for
about 45 miles to the vicinity of Point San Fermin. The highest elevation of the range is Cerro
Kino, directly west of San Felipe Bay, near the northerly end of the range, with an altitude of
4,280 feet. From this point southward the range gradually decreases in elevation. The
western side of the range faces the San Pedro Martir Mountains across the desert San Felipe
Valley. Like the Sierra San Felipe, these mountains are almost without water, even in tinajas.

THE PLAINS AND LARGER VALLEYS.

Although Lower California is mainly mountainous and hilly, numerous plains and valleys
occur within its borders, some of them covering large areas. As described in the general notes
on physiography, a long interior plateau begins at the.northern border on the west. base of the
main mountain chain and is practically continuous along the western or southwestern side of
this chain to its end at the low pass extending across the peninsula at La Paz Bay,

This plateau slopes in a westerly and southwesterly course toward the Pacific. Along its
higher parts in the middle of the peninsula, near the base of the main chain, which forms their
eastern border, extend a scattered series of small, open plains and valleys from near the California
border to the border of the Vizcaino Desert, near Calmalli. Some of the more notable of these
open areas, on the more or less hilly and broken plateau slopes, are as follows, beginning near
the California border: Las Palmas and San Rafael Valleys, Alamo Plain, Trinidad Valley, Plains
of Buenos Aires, Santa Ana, E] Principio and San Pedro, and the deserts, or plains of San Julian,
of Santa Maria, and the north end of the Vizcaino Desert. Many poallee valleys and shallow
basins occur in the intervals between those named. |

The most extensive plains of the peninsula border the Pacific coast. On the eastern or
gulf side some small coastal plains occur, only a few of which are worth special mention, except
the most important of all in Lower California, which is that at the southern end of the Colorado
Desert, in the delta region of the Colorado River.

A brief account of the physical character of the main plains and valleys which we traversed
or of which I have secured definite information i is given below. Further details concerning
these areas along our route are given in the narrative. The descriptions of the plains and. val-
leys are arranged i in their sequence, beginning at the northernmost coastal plain on the Pacific
side and following around the coast to the head of the gulf. The interior plains and valleys are
then mentioned in their sequence from north to south. All of the larger areas are included in
the following list, but there are other small plains and valleys here and there which are not
mentioned. Some of these contain little fertile oases which are occupied by small agricultural
communities, but are of slight importance. The coastal plains contain nearly all of the compara-
tively level parts of the peninsula and in them are most of the lands available in bodies of any
importance for agricultural development except that of the most limited and primitive kind.

Llano de Hamalu.

The Llano de Hamalu is a nearly level coast plain bordering” the Pacific for about 30 miles
from near San Telmo to Santo Domingo and limited abruptly, a few miles inland, by the out-
lying foothills of the interior mountains. For a few miles immediately south of Sarita Domingo
a series of low hills separates this plain from the northern end of the Plain of San Quintin.
The channel of the Rio San Telmo extends to the sea along the northern border of this plain.
It is mainly dry except during rains, but a,small permanent surface stream flows through it in
the lower foothills and along the upper border of the plain. Here a scattered group of small
ranches forms the village of San Telmo. Several small ranches are cultivated on the southern
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 71

border of the plain along the Rio Santo Domingo, below the ruins of the old mission of the
same name. Of these the ranch of an American named Young is locally noted for the variety
of fruits grown successfully, including oranges, lemons, grapes, pears, apples, peaches, figs,
and others. The wagon road from San Quintin to Ensenada crosses this plain.

Plain of San Quintin.

An extensive coast plain lies about San Quintin Bay extending from within a few miles:
of Santo Domingo, more than 20 miles northward of the port of San Quintin, to beyond Socorro
18 miles southward of the same place. The Llano de Hamalu might really be considered the
northern end of this plain, the barrier between the two being series of low hills. The San
Quintin Plain is nearly level or slightly undulating. Near the village of San Quintin it is 5 to
6 miles broad and narrows gradually to the north and south as the foothills, which limit the
plain along its eastern border, approach the coast. The plain is open and more or less sandy
in parts, especially south of the San Simon River, where a belt of sand dunes border the coast.
Several canyons containing small permanent streams open on it along the foothills and the
broad dry wash of San Simon River crosses it to the sea a few miles south of San Quintin.
No permanent fresh water exists on the surface of the plain except the pools at Socorro, near
the southern end, but considerable areas are underlaid by good water so near the surface that
fruit and other trees may be successfully grown in places, as’ shown by the fig orchard and
eucalyptus trees on the Santa Maria ranch, just south of the San Simon River and about 4 miles
back from the coast. Marshy flats occur along the coast about San Quintin Bay, at the mouth
of San Simon River, and at Socorro. The vegetation is so low and insignificant that the plain
has the appearance of an open prairie. Extensive but unsuccessful efforts were made some
years ago, by a company holding one of the large land concessions recently cancelled, to estab-
lish a large agricultural colony on this plain. A considerable area of good land exists and by
proper management much of it could no doubt be reclaimed.

The Vizcaino Desert.

A broad and largely sandy plain, which may be called the Vizcaino Desert, begins on the
Pacific coast a little south of Santa Rosalia Bay and extends southeasterly about, 150 miles
to a point on the coast southeast of Ballenas Bay. On the northern and eastern sides it is lim-
ited by the foothills and lower slopes of the great interior plateau. This plain is broadest
opposite Scammon Lagoon, where it reaches a width of about 60 miles, and narrows to a point
at both ends. It is the largest comparatively level area in the peninsula and is excessively
arid throughout its extent. The northern section, from Black Warrior Lagoon to Santa Rosalia
Bay and thence inland to beyond Calmalli, may ‘be termed the Llanos del Perdido. Near the
sea it is mainly flat and sandy, but the shore a few miles north of Santo Domingo Landing is
bordered by a belt of high sand dunes replaced inland by a low clay flat, which soon merges
into the gradually rising sandy plateau. . South of Santa Rosalia Bay the coastal plain begins
as a point, but broadens southward as the low hills along the coast gradually swing back into
the interior and break into scattered groups, leaving a broad open expanse of desert bordering
the shore as far as one can see to the southward. Inland about 3 miles across low sand dunes
to the eastward of Santo Domingo Landing the surface of the plain rises abruptly with a steep
escarpment about 50 feet high. Beyond this a sandy plain gradually rises to the eastward
about 26 miles to a second escarpment of the same height, and thence on to the distant bases
of the high mountains paralleling the gulf shore. From the first escarpment inland to Calmalli
the plain is a part of the great interior plateau and rises gradually to an altitude of about 1,200
feet at Calmalli, but its general level is from 250 to 800 feet above the sea. A considerable
number of small, islandlike barren hills and low mountainous ridges are scattered here and
there over the plain, mainly on its northern and eastern parts. Among these is the group
of low mountains in which the small mining camps of Calmalli and Campo Aleman are located.

‘The main part of the Vizcaino Desert lies south of the mesa plains of Santo Domingo. It
averages lower than these mesas, the general level being from 75 to 300 feet above the sea, and
is but, little broken by hills or other marked elevations. Its broadest part extends as a gently
72 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI

rising undulating plain from the seacoast, near Scammon Lagoon, about 60 miles inland to the
base of the main mountain range on the east side of the peninsula, not far from. the Mission of
Santa Gertrudis, where the granite formation reaches its southern termination against great
lava-capped mesas. Southward from this the desert is limited on the east by the lower border
of lava slopes descending from the interior plateau and on the west by the Sierra Pintada and
Sierra Santa Clara, which shut it off from the Pacific coast between Cape San Eugenio and
Ballenas Bay. The desert narrows southeasterly between the coast and interior plateau to the
vicinity of Ballenas Bay, whence it again becomes a coastal plain gradually narrowing to a
point southeasterly, where the foothills of the interior plateau reach the shore 40 or 50 miles
below Ballenas Bay.

The Vizcaino Desert is largely sandy and throughout its extent surface water is exceedingly
scanty and widely spaced, the Ojo de Liebre (see Pl. 12, Fig. 1) and San Angel (see Pl. 11, Fig. 2)
being about the only fresh water on it south of the well at Santo Domingo. Extensive salt
deposits exist on flats near the heads of Scammon Lagoon and Ballenas Bay.

The northern part of the desert, over nearly all of the Llano del Perdido is covered with a
remarkable growth of desert vegetation. Farther south between the coast and interior moun-
tain ranges the vegetation is much smaller and sparser. Much of the land on the Vizcaino Desert
is very fertile and with the application of water for irrigation would produce abundant crops.
Unfortunately the scarcity of rain throughout this region and consequent lack of surface water,
and the absence of underground water at a moderate depth except in certain limited areas,
appear to bar forever any considerable agricultural development. About 12 miles inland from
Santo Domingo Landing the sandy plain, overgrown with large and abundant desert vegeta-
tion, gives way to an open clayey prairie 4 or 5 miles in diameter and nearly bare of vegetation,
known locally as the Llano de Berrendo. The plain about Ojo de Liebre, at the head of Scam-
mon Lagoon, is also known as the Llano de Berrendo. Ojo de Liebre means Jack Rabbit
Spring and not Jack Rabbit’s Eye as has been stated by a recent author. Further details
concerning parts of this desert are eves in the Seep ave narrative. ,
The Magdalena Plain.

The Magdalena Plain lies ane the mainland shore of Magdalena Bay and thence up and
down the coast. Next to the Vizcaino Desert it covers the largest body of nearly level land in
the peninsula. It begins as a narrow point on the coast a little above the mouth of La Purisima
Arroyo and gradually broadens southerly and southeasterly a distance of over 200 miles along
the coast to within a few miles of the town of Todos Santos and near the base of the northerly
foothills of the Victoria Mountains. Throughout most of its length this plain is almost entirely
free from hills and rises gradually from the coast eastward a distance of 10 to 20 miles to the
western border of the foothills of the interior plateau which extends down the middle of the
peninsula. A few miles north of latitude 24° N. the main mountain chain and interior plateau
end and the Magdalena Plain spreads easterly entirely across the peninsula to the head of La
Paz Bay, thus forming a broad, low pass separating the mountains of the north from those of
the cape district south and southeast of La Paz.. Most of this plain is ‘without permanent sur-
face water except at considerable intervals, mainly in the bottoms of washes or creek beds ex-
tending out from canyons in the foothills. It is crossed by many large dry washes, and in many
places underground water exists in considerable abundance not far below the surface. Arte-
sian water was struck in one well below La Purisima some years before our journey and was
reported to be still flowing steadily. The Magdalena Plain is fertile and mostly covered with a
large and abundant growth of desert vegetation.

Immediately inland from the shore of Magdalena Bay is a shallow, nearly level sink about
20 miles long i in an easterly and westerly direction, which becomes flooded in times of heavy
rain. It is known locally as the Llano de Yrais and is mainly free from the heavy growth of
desert vegetation common to the surrounding plain, thus giving it an open, prairielike appear-
ance. This part of the Magdalena Plain was once the scene of an extensive effort by an Amer-
ican company, largely made up of men notable in American business and political life, to estab-
lish an agricultural colony. In 1870-71 500 ‘American colonists are said to have been landed
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 78

here, but the experiment was foredoomed to the prompt and complete failure which overtook it.
This plain except the southeastern end is nearly all within the lines of the grant owned by the
Chartered Co. of Lower California.’ At Soledad or Matancita, the headquarters cattle ranch of
this company, the manager had an irrigating pump installed and was successfully growing many
fruits.and vegetables, It was along the coastal border of this plain that the orchilla industry
flourished some years ago.

San Jose Valley.

The valley of San Jose has an east of south direction for about 30 miles, beginning at an
altitude of about 800 feet: near the village of Miraflores and extending to the sea at the southern
end of the peninsula, a mile below the town of San Jose del Cabo. The Rio San Jose flows down -
the bottom of the valley, which is a broad flat varying from a mile in width near San Jose to 3 or
4 miles near Santa Anita. It is bounded on the east by a range of barren round-topped foothills
and on the west by the precipitous slopes of the San Lazaro Mountains, in the midst of which
the sharp San Lazaro Peak rises as a bold landmark. Santa Anita ranch lies in a baylike western
extension of the valley about 9 miles northerly from San Jose. From Santa Anita to Mira-
flores the west side of the valley is occupied by a rolling mesa, mainly of conglomerate, which
rises abruptly in the form of a bench 50 to 100 feet above the general level of the rest of the
valley and extends westerly to the base of the bordering mountains. The valley narrows be-
tween low rolling hills at the town of San Jose and then broadens to a small coastal plain 2 or 3
miles broad and extending about a mile inland. A fresh-water lagoon is formed by the San
Jose River just inside the sandy beach at its mouth. At the time of our visit this was the most
prosperous agricultural valley in the peninsula and maintained the town of San Jose del Cabo,
containing about 1,600 inhabitants.

Plain of Ventanas Bay.

A narrow coastal belt of sandy plain, backed by abruptly rising mountains, occurs along
the shore of Ventanas Bay, opposite Ceralvo Island, where an unsuccessful attempt to establish
a colony is reported to have been made by an American company previous to 1865.

La Paz Plain.

As already mentioned, the southerly end of the Magdalena Plain extends entirely across
‘the peninsula to the shores of La Paz Bay. Good water exists close to the surface all about the
head of this bay, as evidenced by wells and by the luxuriant growth of palms and other vegeta-
tion irrigated from this source at La Paz. The plain rises gradually to the east and south
toward the bases of the Cacachilas and other mountains a few miles away and to the south and
southwest merges into the Magdalena Plain.

Plain of Loreto.

‘A fertile plain without surface water, but with an aman supply, extends 15 or 20
miles along the gulf near Loreto. Other small plains somewhat similar in character are said
to exist on the coast and inland along the trail leading to the extreme head of Concepcion Bay.
We did not. traverse these plains, so can give no details concerning them.

Llano de San Bruno.

A coastal plain overgrown with desert vegetation extends for about 25 is along the
shore of the gulf from near Mulege north to the fishing villages of San Bruno and San Lucas,
and thence inland 6 to 10 miles to the eastern base of the Santa Lucia Mountains. It has been
called the Magdalena Plain, but in order to avoid confusing it with the plain of the same name
on the west coast, the name used above is preferable. The plain rises gently inland and has
- some low barren hills scattered here and there around its borders, especially along its southern
border near Mulege. The ruins of the old mission of Guadalupe are located at the edge of the
foothills on its western side. Scattered over the plain are a number of deserted ranches where »
attempts have been made to cultivate the land. Water is plentiful a short distance below the
surface, and some was being pumped, but primitive methods of raising it for irrigation are
impracticable. As long ago as 1866, some of the land on this plain had been acquired by a San
74 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou, XVI

Francisco company for colonizing purposes, but nothing was accomplished. At the time of
our visit, in 1905, we found an olive orchard, vineyard, and a good house at San Bruno belonging
to an English company. Efforts to develop this orchard had been carried on for some years
but apparently with poor success. Surface water is very scarce in this district but a great
spring rises in the bottom of El Potrero Arroyo just above Mulege, about three miles from the
shore of the gulf, and irrigates the fertile fields about the town. A shallow well in the bottom
of the Arroyo, a few miles above Mulege, gives an abundant supply of water.

Calamahue Valley.

Calamahue Valley is a gently sloping plain about 20 miles long and one to four miles broad,
beginning at an altitude of 950 feet at the mouth of Calamahue Canyon and descending a little
east of north to the mouth of Calamahue Arroyo on the shore of San Luis Gonzales Bay on the
gulf. It is bordered on the eastern side by the high abrupt slopes of the Calamahue Mountains
and on the west by low rounded hills. The broad sandy and stony dry bed of Calamahue Arroyo
winds down the middle of the valley and gives evidence of occasional floods. The lower end
of the valley is shut off from the gulf by a single low ridge, through a narrow gap in which the
Arroyo passes to the gulf. A shallow well of good water is located on the south side of the
bottom of the Arroyo just above the gap in the hills. The only surface water in the valley is
from the group of springs located at its upper end, just within the mouth of the canyon beside
the ruins of the old Calamahue Mission. All of these springs are more or less strongly charged
with soda and other minerals, the water from some being too bad to use and in none is it good.
The desert vegetation in the bottom of the valley is small and rather sparse, being much like
that near San Felipe Bay. The surrounding mountain slopes are extremely barren. This
valley is one of the hottest places we visited during our trip. : :
The Colorado Desert. oy

Extreme nGrieeeier Lower California is occupied by the southern end of the Colorado
Desert.

This level or onl ‘sloping: area extends from the California border southward to Sah
Felipe Bay, on the Gulf of California. It is bounded on the west by the Sierra Juarez and
Sierra de San Felipe; on the east by the Colorado River and the northern end of the gulf. A
narrow belt along the gulf coast and a much ‘broader area in the delta district, especially near
the border and about the Cocopah Mountains, are nearly level and only a little above sea level.
Inland from these flats the desert plain rises gradually to elevations varying from a few hundred
to over 1,500 feet above the sea along the base of, the mountains forming its western border.
More or less disconnected and irregular desert mountains and small ranges varying from a few
hundred feet high to elevations of over 4,000 feet are scattered irregularly over this desert,
especially in its southern half, where a considerable part of its total. area is thus occupied.
These mountains are surrounded by plains or by valleys and are separated by narrow passes.

. The flood plain of the Colorado delta extends westward to the east base of the Cocopah
Mountains. On this plain, near the base of the mountain, is located a small volcanic area
centering at the Cerro Prieto, near which are located the curious mud volcanoes of this region.
They are small mud cones a few feet high which at intervals send out small gushing overflows
of warm liquid mud from craterlike openings in the tops. (See Pl. 21, Fig. 2.)

On the western side of the Cocopahs is a sink extending below sea level containing a brackish
lake known as the Laguna Maquata or Laguna Salada. At times of high water in the Colorado
River this lake is greatly increased by overflow from the river around the southern end of the
Cocopahs. This depression, like the basin of the Salton Sea, once formed part of the great de-
pression occupied by the northern extension of the Gulf of California, occupying all this part of -

the present Colorado Desert before it was filled with silt brought down by the Colorado.
Owing to their size and isolation the Cocopah Mountains form the most conspicuous moun-
tain group of this desert area. Farther south are the Sierra Pinta, the Sierra Tinaja, and the
Sierra de San Felipe; the last-named range extending in a generally’ southerly direction parallel
to the coast to a point opposite San Felipe Bay and separating the narrow southern extension
of the Colorado Desert along the gulf coast from the interior.
No.1], LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 75

_ On the California side of the border in this region part of the Colorado Desert is occupied
by the wonderfully fertile district familiarly known as the Imperial Valley. The southern part
of this marvelous agricultural district lies across the border in Lower California, where within
the last few years extensive development has taken place, with the town of Mexicali as its com-
mercial center. These delta lands lie both in this part of Lower California and the adjacent
part of California and are dependent on water obtained from the Colorado River and carried in
canals passing through Lower California before entering California. Although having one of the
hottest and most arid climates of any part of the peninsula, it is by far the most valuable agricul-
tural area in Lower California. The lower parts of the Colorado Desert are composed wholly
of silt brought down in ages past by the Colorado River. As a consequence the silt is exceed-
ingly rich, and abundant crops grow wherever water is properly applied.

Valle de San Felipe.

. Lying along the eastern base of the San Pedro Martir Mountains is a long and troughlike
valley with an altitude of about 1,200 feet in its lowest part. This valley varies from 3 or 4
miles to more than 12 miles in width and has a length of about 60 miles. The narrow bottom
in the northern part of the valley is occupied in part, especially opposite La Providencia Canyon
in the San Pedro Martir Mountains, by series of sand dunes, and the valley slopes rise steadily
to 600 or 1,000 feet on both sides to the bases of the adjacent mountains. The valley is an
exceedingly arid barren desert, limited by the closing in at each end of the bordering mountains.

The eastern border of the desert is practically waterless, but water occurs at the mouths of a
number of large canyons in the mountains along its western border. Along the basal slope of
the mountains on the western side of the valley there is an abundant desert vegetation, but else-
where the plant vegetation is small and gives evidence of the intensely arid climatic condition.

San Rafael Valley. _

San Rafael Valley lies on the interior plateau about 20 miles by road easterly from Ensenada
at an altitude of about 2,600 feet. It forms a long, slightly undulating, basinlike plain between
the Coast Range and the west base of the Sierra Juarez. It lies in a generally northwesterly by
southeasterly direction for a distance of 15 or 18 miles and has a width of from 3 to 10 miles, being
broadest in its southern half. It appears to lie wholly within a granite formation. Te the
middle of the valley, almost due east from Ensenada, are some large springs, known as Ojos
Negros, whence a series of swampy flats with some tules in them extend down the bottom
toward Real del Castillo and drain through Arroyo San Rafael or Arroyo Guadalupe to the
Pacific coast north of Ensenada. The eastern side of the valley is sandy, owing to the abundant
disintegrated granite wash brought down from the Sierra Juarez. In the bottom of the valley,
especially above Ojos Negros, mesquites, ephedras, and yuccas were common, while the adjacent
foothills were clothed with an abundant growth of Adenostoma, Eriodictyon, Arctostaphylos, Sam-
bucus, and of two oaks, Quercus wislizenii, and Quercus agrifolia. A little land is cultivated,
mainly about the border of the valley, but it is principally occupied by a large cattle ranch with
headquarters at Ojos Negros. The lower end of the valley leads to the site of. the old mission of
Guadalupe.

Alamo Plain.

The Alamo Plain is a gently rolling part of the main plateau, at an altitude of from 3,500 to
3,800 feet, along the westerly side of the southern end of the Sierra Juarez and separated on the
north from the southern end of San Rafael Valley by a range of rounded low hills. Itis about 20
miles long in an east and west direction and muchlessinbreadth. The old Santa Catarina Mission
islocated on its eastern border and the small mining town of Alamo, at the east base of the Coast.
Range, at its western end. The plain is watered with a few small springs and waterholes and a
small stream in the arroyo at Santa Catarina. The drainage is westerly through the Santa
Tomas River into the Pacific. A considerable growth of California junipers formerly covered
the plain and the bordering foothills, but the larger of these have been used as fuel for the mine
at Alamo, leaving only the smaller growth. On the north and south the plain merges into rolling
76 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

hill country. To the southeast it gradually rises to the top of the cliffs which bound the north:
ern side of Trinidad Valley.

Trinidad Valley.

The Trinidad Valley is a long oval basin with a gently sloping bottom, extending about 18
miles in an easterly and westerly, direction, and 5 or 6 miles across in the middle, decreasing in
width toward each end. It is the head of the drainage of the Rio Salado, which flows westerly
through the Coast Range into the Pacific. Its eastern end forms San Matias Pass, which, at an
altitude of about 3,200 feet, separates the south end of the Sierra Juarez from the north end of
theSan PedroMartirMountains. Thevalley thence slopes gently westward to the lowerend below
Campbell’s ranch, where, at an altitude of about 2,500 feet, it closes into a box canyon. On
the north it is bordered mainly by a high rocky escarpment, and on the south by a series of rolling
foothills, extending from the west base of the San Pedro Martir Mountains.

The valley is without surface water, except some small springs and waterholes and a small
stream at its extreme lower end. A good underground supply exists at shallow depths in the
lower half of the valley. Its western two-thirds is an open, grassy plain with some small shrubs;
the narrowing eastern third has a considerable growth of cactuses, yuccas, agaves, and other desert
plants, which come through San Matias Pass from San Felipe Valley to the east. Trinidad Val-
ley is too dry for agriculture, but with the surrounding hill slopes makes an excellent cattle range.
Llano de Les Buenos Aires, or Llano de San Augustin.

At San Fernando there is a small narrow flat-bottomed valley which extends to the east-
ward, through the open hills, and merges into a broad, slightly undulating desert plain known
as the Llano de los Buenos Aires, or Llano de San Augustin. This plain is a part of the great
central plateau lying at an altitude of from 1,800 to 2,500 feet along the middle of the penin-
sula and extends southeasterly from the Bonshern end of the San Pedro Martir Mountains. “”

The names Llano de los Buenos Aires and Llano de San Augustin have heretofore had a
local application to the more level middle parts of this plateau between San Fernando and Onyx,
but here they are used to cover all this section of the plateau. The entire Ilano.is about 28 miles
across, in an easterly direction from near San Fernando to the Onyx mines (located at the
edge of the foothills on its eastern border), and is over 40 miles long from its border north of
San Augustin Well southeasterly to Arroyo Catavina. This part of the plateau is nearly level in
places, especially between San Fernando and Onyx, but is generally more or less undulating
with small scattered hills and outcroppings of bare granite. To the north the plateau reaches
the extreme southern end of the slopes leading down from the San Pedro Martir Mountains.
On the eastern side it is walled in by the bare, abruptly rising slopes of the mountains which
parallel the gulf shore; on the western and southern sides it slopes gently southwesterly toward
the Pacific and is limited by irregular ranges and groups of low rocky hills and mountains.
Water is scarce but is found at the surface in the bottoms of several small canyons and arroyos.
The vegetation is scanty between San Fernando and Onyx, but farther to the southeastward
an extraordinarily abundant and vigorous growth of desert species prevails. This plain is
limited on the southeast by the Arroyo Cataviiia, beyond which, for 12 miles, to Arroyo Jara-
guay, the country becomes much more broken and is covered with ridges and piles of huge
granite bowlders. (See Pl. 27.) The onyx mines, which have been worked for a number of
years, are located at Onyx on the east side of the Llano de los Buenos Aires. A wagon road,
down which the onyx is hipaled fi shipment, leads thence to Santa Catarina Landing on: the
Pacific coast.
Llano de Santa Ana, Desierto de San Julian, and Llano de Santa Maria.

The Llano de Santa Ana lies southeast of the southerly end of ‘is Liano de los Buenos
Aires, from which it is separated by an area of broken and hilly country. It is one of the most
level parts of the main plateau ‘slope and lies along the middle of the peninsula from the ex-
treme head of the Arroyo Calamahue southeasterly to the vicinity of Yubay, where it is limited
by a small belt of knolls and broken rocky slopes. The upper or eastern border of this plain
has an altitude of about 2,000 feet and is walled by the bare granite slopes of the Calamahue
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. a

Mountains. The plain extends southwesterly for 12 or 15 miles down a long slope and across
the valley of the Arroyo de San Andres to the Sierra Prieta and other barren mountains parallel-
ing the Pacific coast. The plateau on the borders of the Llano de Santa Ana is mainly rolling
and marked by low scattered ridges and projecting knobs of granite. Its slopes are scored
with many low, abruptly walled gulches with sandy dry washes in their bottoms leading down
to the broad Arroyo de San Andres. The tinaja at Yubay contains the only known water on the
trail crossing the eastern border of the Llano de Santa Ana.

The surface of the llano surrounding parts of the plateau is covered with disintegrated
granite, often more or less sandy, and supports a tremendous growth of desert vegetation,
which becomes a low forest in many places, frequently forming an impenetrable thorny jungle.
Another open plain, known as the Desert of San Julian, extends from Yubay southeasterly by
the old mission of San Francisco de Borja and through low hills into still another plain called
the Llano de Santa Maria and thence descends to the northern end of the Vizcaino Desert north
of Calmalli. The areas are parts of the well-marked continuous elevated interior plateau be-
ginning at the southwestern base of the San Pedro and extending southeasterly along the west
base of the main range until it merges into the Vizcaino Desert. Surfaée water is very scarce
and mainly in small rock tanks or here and there where the underground flow in stream beds
is forced to the surface by the bedrock for short distances.

San Ignacio Valley.

San Ignacio Valley is | formed by the broadening of a low canyon which drains the west-
ern slope of the Santa Lucia Mountains to the Pacific. It is located near the middle of the
peninsula and is only about 3 or 4 miles long by a mile or so in width. The lower end narrows
into a flat-bottomed canyon 12 or 15 miles long leading down to the southern end of the Vizcaino
Desert. Near the upper end of the valley several great springs rise in the bottom and supply
abundant water for irrigating all the arable land in the valley and most of that in the canyon
below.. The valley is the site of one of the largest and most flourishing of the old missions and
at the time of our visit had a population of about 1,000, closely grouped about the large, well-
preserved mission church. (See Pl. 11, Fig. 1.) The valley has an altitude of about 800 feet
and is-located in the midst of a great lava plateau. Its drainage is southwesterly into San
Ignacio Lagoon on the coast of the Vizcaino Desert. It is avery fertile spot, notable for
the great number of date palms. The lava plain extends northeasterly from the San Ignacio
Valley past the bases of the volcanoes of the Tres Virgines to the gulf coast and forms one of
the few low passes across the peninsula.

La Purisima Valley.

La Purisima Valley is formed by the flat bottom of a high-walled canyon, varying —
about 100 to 1,000 yards wide, in the midst of the great volcanic plateau. The bottom of the
valley has an altitude of from 300 to 500 feet and is irrigated throughout its length by a stream
giving the most abundant water supply existing in any of the canyon valleys of the peninsula.
The valley is very fertile and is cultivated for a distance of about 12 or 15 miles, below which
it becomes a narrow box canyon through low foothills for some miles and then opens on the
northern end of the Magdalena Plains. The drainage is southwestward into the Estero de
San Ramon onthe Pacific coast. We were informed that a well sunk 1,200 feet in prospecting
for'oil in the lower end of La Purisima Canyon some years before struck artesian water, which
had been flowing continuously since that time. This valley has a population of about 600 and
is the site of one of the old missions.

Comondu Valley.

This is another narrow valley formed by ‘the flat bottom of a high-walled canyon in the
elevated volcanic plateau southeasterly from, and closely resembling, the valley of La Purisima.
Iti is about half a mile wide and is irrigated for 8 or 10 miles of its upper part by the water flowing
from large springs. It narrows below to a deep. box canyon which ends on the northern part of
the Magdalena Plains. It is one of the most beautiful valleys i in the peninsula and at the time of
our visit was occupied by a population of about 1,000 people. (See Pl. 13, Fig. 2.) The valley
78 MEMOIRS NATIONAL. ACADEMY OF SCIENCES. [Von XVI

has an elevation of about 700 feet and drains southwesterly across the Magdalena Plains to the
Pacific coast. Like La Purisima this valley contained one of the old missions. :

STREAMS AND OTHER SURFACE WATERS.

Owing to the scanty and irregular rainfall within its borders Lower California is poorly
supplied with surface water, the only real river touching the peninsula being the Colorado, the
delta of which lies on the extreme northeastern border. The long, narrow form of the peninsula,
with a high divide extending practically its entire length, reduces the length of all watercourses
located within its borders to the short distance from the high interior to one coast or the other.
Although the peninsula i is about 800 miles long, it is doubtful if any of its watercourses reach a
length of 90 miles. There are hundreds of drainage channels which vary from rugged rock-
walled canyons, some of which reach 2,500 feet in depth on the flanks of the mountains, to sandy
or stony washes half a mile or more broad on the open plains. Except during heavy rains most
of these channels are wholly dry, but in the upper parts of many of the larger canyons water
rises and flows permanently for distances varying from a few rods to 10 or 15 miles before sinking
again. In some canyons the water rises and sinks several times, but usually sinks finally near
the border of the plains, or in the low foothills. Usually the water thus rising forms an insignifi-
cant streamlet, but in some cases it boils up through great springs which send out a good-sized
creek, large enough to irrigate many hundreds of acres of land, as in San Ignacio, La Purisima,
Comondu, Todos Santos, and San Bartolo Canyons. Most of the cultivated oases along the
middle of the peninsula depend on the water flowing from such springs. During heavy rain-
storms, especially in winter, streams often flow down the usually dry channels to the sea, some-
times for several days or even weeks in succession, but periods of several years may pass without
a drop of water appearing in them.

The great number ‘and frequently large size of the dry canyons and other drainage channels
throughout the peninsula, and other signs of much erosion, might lead to the impression that a
greater rainfall once prevailed. The tremendous power of such floods as occur at times through
the peninsula are well shown at the mouths of the great canyons along the east side of the San
Pedro Martir Mountains. Great alluvial fans, or deltas, several hundred feet deep slope down
to the bottom of the valley from the mouths of these canyons, and the tops of the fans for half
a mile or more below the canyons are heaped with long piles. and masses of tumbled granite
bowlders varying in size; numbers of them range from 3 to over 6 feet in. diameter and
weigh several tons. In some places these rocks are in such profusion, especially below the
mouth of La Providencia Canyon on the east base of the San Pedro Martir Mountains, that they
resemble glacial moraines and render approach to the canyon mouth extremely difficult. The
friable character of much of the rock and observation of the enormous amount of material
carried down by every flood show that present conditions are sufficient to produce the erosion
so much in evidence throughout the peninsula.

As noted elsewhere, the longest slope of the peninsula is toward the west or southwest, and
the longest drainage channels descend in that direction to the Pacific. Beginning at the northern
border the larger of these western stream beds: known to have the most extended supply of per-
manent surface water in some part or parts of their courses are named in their sequence as
follows: Tijuana, Calabazas or Cancio, Guadalupe, Santa Clara, Santo Tomas, San Vicente,
Salado, San Rafael, San Telmo, Santo Domingo, San Simon, San Javier, San Ignacio, San Jose,
La Purisima, Comondu, San Venancio, Santo Domingo, Soledad, and Todos Santos. There are
numerous smaller and wes important canyons along the mountain slopes, which also have a little
permanent water in the upper parts of their courses.

' The east slope of the peninsula is narrow and has much less water than the west slope except
for the huge volume of the Colorado poured into the head of the gulf. In numerous canyons
along the east base of the Sierra Juarez and San Pedro Martir Mountains and elsewhere down
this front of the mountains small permanent ‘streams flow for short distances, almost entirely
within the canyons except during rainy periods. The most’ prominent of the water-bearing
Memoirs National Academy of Sciences XVI.

First Memoir, Pl. 23.

od
mae

5 —
eed

Ee

=

 

 

 

Fig. 1.—Island overgrown with cottonwood (Populus macdougalii), in process oferosion. Delta of Colorado, April, 1905.

 

 

Fic. 2.—Flood plain of Hardy River at east base of the Cocopah Mountains, April, 1905.
No. 1] | LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 79

drainage courses on the east side of the peninsula are as follows: Colorado, Arroyo Grande,
Canyon Diablo, La Providencia, El Tule, Matomi, Calamahue, Santa Isabel, Agueda, San Jose,
Magdalena, El Potrero, Loreto, Santiago, and San Jose del Cabo.

The extreme northern and southern parts of Lower California have higher mountain masses
and receive a greater volume of and more regular precipitation than the middle section. Con-
sequently these two areas, without considering the Colorado, have the more plentiful supply of
both surface and underground waters. The northern section includes the Sierra Juarez and Sierra
San Pedro Martir and is th larger area with the largest total of water, but the smaller southern
section, south of La Paz, ) 4s more in proportion to its size.

The Colorado is the g) 2atest of all and the only real river in Lower California. It forms a
great delta region containi ig much fertile land in the northeastern corner of the peninsula and
pours a mighty current into the head of the gulf. Every spring and early summer there is a
flood in the Colorado from melting snow in the Rocky Mountain region of the United States,
which fills the ramifying’ channels in its lower course and overflows the lowlands between the
eastern base of the Cocopah Mountains and the Colorado. (See Pl. 23.) This diverted flow
reenters the Colorado through the various old channels which drain the delta. Some of the
overflow from the Colorado passes by the southern end of the Cocopah Mountains into the bed
of the long Laguna Maquata, or Laguna Salada, as it is variously named, which lies along the
western base of these mountains. These floods build new islands and tear down old ones in the

delta each spring.

It is well known that owing to the filling up of its bed in the delta by silt the channel of the
Colorado has many times shifted to the east or west. In 1905 the middle channel was in use,
but since then the main stream has again changed and now flows through Volcano Lake and
thence by way of the Hardy Channel to the mouth.

Within a comparatively few years large irrigating ditches have been taken out of the
Colorado i in the extreme northeastern corner of Lower California and carried across the plains to
hundreds of thousands of fertile acres on both sides of the border to the westward from this point.

Excepting the Colorado, in all the coast line of the peninsula, more than 2,000 miles in
length, only five small streams flow to the vicinity of the sea throughout the year. Of these
only one, the Rio Santo Domingo, north of San Quintin, has a continuous flow throughout most
of itslength. Of the other four, the San Jose rises and sinks several times in its channel, while
the others, the arroyos of El Rosario, Todos Santos, and El Potrero at Mulege, emerge in large
springs only a few miles from the sea. None of the streams flowing permanently to the vicinity
of the coast enter the sea directly, as ordinarily the sea beach forms a sand bar across the mouth
of each, so the stream discharges into a fresh or brackish lagoon and seeps through the sand, or
sinks in its sandy bed, except El Potrero, which flows into a lagoon’ extending from Concepcion
Bay...

_ As already noted, most of tne surface water in northern Lower California is in the drainage '
system of the Sierra Juarez and San Pedro Martir Mountains. The east slope of these moun-
tains drains toward the gulf or the delta of the Colorado and is practically waterless, except for
a few small springs and small streams of clear pure water which flow down the bottoms of the
larger of the deep, short canyons which gash the precipitous front of this slope. In cool weather,
especially after rains, these streams flow out sometimes several miles on the sloping plain at the
mouths of the canyons, but during hot weather the decrease in flow and the evaporation are SO
great that they often end well up in the lower part of the canyons. Several of these springs and
little streams are located along the éast base of the Sierra Juarez, especially in Palm Spring,
Tako, Guadalupe, Palomar, A_ 18 Caliente, Tule, and Arroyo Grande Canyons (See Pl. 24,
Fig. 1). In Arroyo Grande the permanent water flows for 10 miles in the canyon bottom but
sinks well above the open dese) After torrential rains along their upper courses both of these
arroyos send their waters nort vo the basin of Laguna Maquata. ‘Throughout most of the
year along the east base of the ».n Pedro Martir Mountains and adjacent front of the table-
land to the southeast small streams reach the mouths of the larger canyons, which include
Esperanza, Diablo, La Providencia, Santa Rosa, Agua Caliente, Paral, and Matomi.

158961 °_21_6
80 MEMOIRS NATIONAL ACADEMY OF SCIENCES. © [Von! XVI

In the latter part of June, 1905, water flowed at night a little below the mouths of La
Esperanza, El Diablo, and La Providencia Canyons, but owing to the heat and consequent
evaporation by day the lower end of the surface water had retreated one-half or three-fourths
of a mile by sunset each afternoon. Later in the season the water retreats still farther and the
lower ends of the streams are said to be well up in the canyons.

On the west sides of the Sierra Juarez and Sierra San Pedro Martir nearly all the larger.
canyons on the main mountain slopes have small permanent streams flowing in at least a part of
their length and sometimes well down to the foothills to within a few miles of the sea. Some-
times the water sinks and rises more than once in its course along the rocky or sandy canyon
bottoms.

The main drainage channels leading down to the coast on the Pacific slope of these moun-
tains are the Tijuana, San Rafael, Ensenada, San Carlos, Santa Clara, Santo Tomas, San Vicente,
Salado, San Telmo, Santo Domingo, San Simon, and El Rosario. All of these drainage ¢ channels
with many of their tributaries in the foothills have permanent running water in at least part of
their courses, but with the exception of the Rio Santo Domingo and Arroyo del Rosario the water
sinks some distance back from the coast. The Rio Santo Domingo is the only stream within
Lower California which has a permanent flow from its extreme head nearly or quite to the coast.
Tits headwaters consist, mainly of several arroyos beginning high up on the western side of the
San Pedro Martir Mountains. The most important fork is known as the Arroyo San Antonio,
or Arroyo La Grulla. This drains La Grulla Park at an altitude of more than 7,000 feet and
descends by San Antonio ranch to unite with other tributaries. This is the main source of the
Rio Santo Domingo, which reaches the Pacific north of San Quintin. About 6 miles above San
Antonio ranch the arroyo has a high waterfall which, Mexican engineers have estimated would
yield 8,000 horsepower. Other tributaries are the arroyos of El Alcatraz, La Zanja, Santa
Clara, sii Valladares. One of the north forks of the Santo Domingo i is known as the Arroyo de
Valladares, on which are located the Valladares placer mines. Mr. A. W. Anthony worked these
mines at one time and informs me that the Valladares Creek can be counted upon to give about
300 miner’s inches of water for seven months in the year. A few miles south of the mines the,
Arroyo San Antonio, another large fork of the Santo Domingo, comes out of a canyon in the west
base of the mountains at the San Antonio ranch. The water in this creek at. the end of. July,
1905, was 10 or 15 yards wide. and varied from a few inches to 2 feet deep. |

The Arroyo San Antonio is distinguished as being the only stream in ‘Lower California
which contains brook trout. We found a species living here which attains a length of 12 to 15
inches and is related to the rainbow trout of California. It was abundant i in the canyon above
San Antonio ranch, which has an altitude of about. 2, 000 feet. This trout proved to be a new
species, which Dr. Evermann has named Salmo nelsoni. There is a strong flow of water in the
headwaters of the Arroyo San Telmo, which heads on the west side of the San Pedro Martir
Mountains north of the Santo Domingo. Ditches have been taken out of this stream and water
carried several miles along the foothills to the Socorro placer mines a few miles from the Valla-
dares placers. The Arroyo del Rosario, 40 miles south of San Quintin, carries a small permanent
stream which rises in the dry bed of the arroyo and flows less than 5 miles down a small canyon.
to the seashore.

Water is scarce on the high parts of the Sierra Juarez and San Pedro Martir Mountains.
A few small springs and streamlets in canyons are about all. A large pond, known as the Laguna
Hanson, is-located on top of the Sierra Juarez northeast of Ensenada. This pond is in a small
mountain park at an altitude of about 5,400 feet and surrounded by low pine forest. Itis shallow
and nearly half a mile long with an irregular outline, the arms extending between ridges of granite
bowlders. The bottom is rather hard and the cattle ranging there keep down all marginal plant
growth, so that there is no shelter for wild fowl. Two or three much smaller ponds are located
in La Grulla Meadow, a long mountain park situated at about 7,000 feet on the west side of the
San Pedro Martir Mountains. (See Pl.7.) Formerly these ponds were bordered by aquatic
vegetation and frequented in summer by breeding waterfowl, but the cattle ranging there in
1905 had grazed this growth so closely that the birds had abandoned this resort. A large
No. 1} LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 81

spring, Ojos Negros, rises in San Rafael Valley and another large spring rises near the old mis-
sion of San Fernando, but judging from evidences of former irrigation there must be much less
water at the latter point now than in the days of the early missionaries.

Here and there throughout the north as well as elsewhere in the peninsula, both in foothills
and on mountain sides, occur tinajas or natural water tanks, usually formed by large potholes
worn in the rocky bottoms of dry canyons or gulches. These natural tanks sometimes hold
water several years in succession, during which no rain falls.

The most arid part of the peninsula lies between San Fernando at about 30° and La Paz
near 24° north latitude. Within this area not a single living stream touches the Pacific coast,
except during heavy rains, although the western slopes of the mountains and bordering plains
are marked by large numbers of deep canyons and other watercourses. Most of these are dry
throughout their length, but water flows in the upper parts of some of them, and in a few, like
San Andres and Rosarito Arroyos, water rises and flows a short distance in the lower parts of
their channels within a few miles of the coast. In the interior, especially in the elevated lava-
covered mesa country from San Ignacio southeasterly, several of the large canyons have great
springs rising in their bottoms, from which enough water flows to irrigate the narrow belt of
land, amounting to several hundred acres, which extends for miles down the canyon. The can-
yon valleys at San Ignacio, La Purisima, and Comondu are the most notable of those which,
owing to the abundance of water, form beautiful oases. San Ignacio is one of the old mission
towns located in the middle of the great lava plateau. (See Pl. 11, Fig. 1.) A giant spring
boils up in the bottom of the drainage bed of San Ignacio eave just above the town and
flows for miles down the canyon to become lost in the sand before it reaches the border of the
Vizcaino Desert. A small stream comes to the surface for short distances toward the head of
the Arroyo near the Sierra de Santa Lucia. At San Ignacio and below the great date palm
groves other fruits and crops are irrigated from this stream, which, in addition, fills two con-
siderable ponds by the town.

One of the largest streams of this character in the peninsula exists in La Purisima Canyon
and irrigates the bottom sufficiently to support a community of a thousand or more people,
although the water is not well utilized. This stream flows 12 or 15 miles along the canyon and
for a considerable part of its course the bed is characterized by a series of water holes which are
always full. These water holes sometimes reach a length of several hundred feet and are re-
ported in some instances to have a depth of 100 feet or more. During dry periods a small trick-
ling stream connects these holes while the main part of the stream bed is dry. The water supply
of Comondu comes from two large springs rising close together in the upper part of the canyon.
This stream has a course several miles in length and irrigates hundreds of acres of fertile land
along the level bottom. (See Pl. 13, Fig. 2.) Smaller streams furnish water enough to irrigate
small patches of land in the interior foothills of the west slope at San Borja, Santa Gertrudis,
Purisima Vieja, San Luis Gonzaga, and a few other places.

South of the Arroyo de Comondu a number of canyons of varying size descend the western
side of the lava plateau and send drainage ways out into the Magdalena Plain, some having con-
siderable water in the upper parts of their courses which sometimes for weeks flows well out upon
the plain before sinking. These are mainly the Arroyo of San Venancio, Guadalupe, Aguajitos,
Jesus Maria, El Quepo, Soledad, San Vicente, San Gregoria, La Junta and San Hilario. The
Guadalupe and Soledad are the largest of these.

Along the entire gulf coast the only stream, except the Colorado, which permanently reaches
the shore is at Mulege, where a great spring rises in the bottom of the Arroyo del Potrero, about
3 miles above the sea, and flows enough to irrigate the fertile bottom and thus support the town.
For about 18 miles above the spring the bottom of the Arroyo del Potrero is dry and thence up
to its source among the mountains a small amount of water rises at intervals and flows short
distances along the bottom.

Numerous dry channels lead down to the coast from the bordering mountains along the
short easterly slope of the peninsula south of the San Pedro Martir Mountains. A few of these
82 MEMOIRS NATIONAL ACADEMY OF SCIENCES. | [Vou. XVI

have a little water in the upper parts of their channels. Among these may be, mentioned the
following arroyos: Matomi, Santa Maria, San Francisquito, Calamahue,, Santa Agueda, and
Loreto. The water in the Arroyo de Santa Agueda is about 10 miles inland. and through a pipe
line supplies the mining town of Santa Rosalia, which contained about 8,000 inhabitants, An-
other small stream at Loreto supplies water for irrigating a small amount of land.

The extreme southern end. of the peninsula south of La Paz is more plentifully supplied
with surface water than any other equal area in Lower California. Small streams exist in the
upper parts of all the large canyons in the main range from the Sierra de la Laguna to within a
few miles of Cape San Lucas, and i in several instances reach the basal lowlands along the east
slope. Springs are rather scarce in this region and usually occur in or near the bottoms of can-
yons and gulches. In the mountains the springs are all small, but much larger ones occur in
the lowlands. The only stream which flows permanently to the sea on the Pacific coast. of the
cape district is at Todos Santos. There, within 3 miles of the coast, a great spring gushes from
the bottom of the drainage way leading from a large canyon on the northwestern side of the
Sierra de la Laguna and irrigates hundreds of acres of sugar.cane in a pretty valley on its way
to the sea. The small town of Todos Santos, located at the upper end of. this irrigated area,
is wholly dependent upon the land irrigated by the spring. Formerly the spring flowed from, a
point a mile or more above its present location, but suddenly became dry and soon after burst
forth at the lower point, thus causing the abandonment of a broad strip of cultivated land.

South of Todos Santos several large canyons cut the precipitous west slope of the. Vic-
toria Mountains and extend through the foothills to the sea as strongly marked drainage chan-
nels. In the upper parts of the canyons small streams occur, but lower down they sink and
leave the broad sandy or rocky beds of the washes dry, except when heavy rains send down
brief floods. Throughout, the cape district, as well as elsewhere in Lower California, the deeply
cut, dry drainage channels can be readily accounted for by present climatic conditions, in view
of the friable character of the rock and the tremendous erosion caused by occasional floods.

On the east side of the cape district, nearly opposite Todos Santos, the giant spring of
San Bartolo breaks out in the dry bed of the canyon of the same name. We did not visit this
but Dr. Eisen has estimated the permanent flow to be about 5 cubic feet of clear cold water per
second. This supplies water enough to irrigate several hundred acres of cultivated land along
the steep sides of the canyon and to form a small stream about 4 miles down its bottom. There
is another smaller spring near La Palma. The San Jose River, at the southern end of the
peninsula, is the largest and most important stream in, the cape district. It is over 50 miles
long and the main branch, the Arroyo Miraflores, rises northwesterly from Miraflores in the high
Sierra San Lazaro. It rises near the summit of the Sierra and flows southeasterly for about
15 or 18 miles to the village of Miraflores and then southerly down the valley through the town of
San Jose del Cabo to the sea a ‘mile beyond: The stony bed of the river through the valley i is
over half a mile broad in places and at flood during heavy. storms for a week or more at.a time
it is sometimes covered by aroaring and impassable torrent. Ordinarily the channel is occupied
by a small creek which rises and sinks at intervals, but comes to the surface in greatest abun-
dance near the sea where there is a never-failing flow to the beach. It hasno tributaries from the
east, but a series of large drainage channels lead into it from the border of the valley on the west.
The largest of these from San Jose northward, are the Arroyos Santa Rosa, San Lazaro (passing
Santa Anita Ranch), San Miguel, San Ignacio, Caduano, and Miraflores. The latter is the largest
of all, a strong current flowing down its canyon and into the valley all the year. Most of these
streams flow permanently within their canyons, but during the dry season the smaller ones sink
in the sand on reaching the border of the valley. At flood any one of them may increase to a
roaring torrent down the steep canyon bottoms transporting great quantities of bowlders, so that
the broad stream beds, often 100 yards or more wide, are often covered with tumbled masses
of water-worn rocks that in places render passage on horseback impossible. The Arroyo
Miraflores is very rocky and shows evidences of fierce floods many feet deep, though at the time
we ascended its course in January it. was a small clear brook. Even when the courses of, the
tributary streams in the valley are dry, they always have an underground flow which feeds the
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 83

main streat. An aggregate of several thousand acres of land are irrigated along the course of
the San Jose and its tributaries, the largest single area being on the flat lying between the town
of San Jose and the mouth of the river.

The mouth of the San Jose is usually shut off from the sea by a heavy bar of sand built up
‘py the waves. The water is thus held back to form a long narrow fresh-water lagoon immediately
back of the beach. At flood time the water from the river backs up here until it spreads over
most of the adjacent cultivated fields. At such times, if the water does not break a passage
for itself through the bar at the mouth, the people dig a channel across the beach and thus drain
it into the sea.

In the southern cape district the drainage system of the Rio Santiago is next in importance
to that of the San Jose. It is separated from the valley of San Jose by a ridge lying north of
Miraflores which extends easterly from the Cerro Blanco of the Sierra San Lazaro. From the
northeasterly slope of the Cerro Blanco to and including the easterly slopes of the Sierra de
la Laguna, all the drainage leads into the Santiago, passing the village of this name and thence

into the gulf. The stream in E] Sauz Canyon is the southernmost tributary of the Santiago
from the Sierra de la Laguna; to this may be added the Arroyos de la Chuparosa, San Antonio,
and La Laguna. Asmall stream flows down the little Laguna Valley from the higher slopes of
the Sierra dela Laguna and lower down joins with a small stream from San Antonio Canyon before
reaching the Santiago. There is a permanent flow in the Santiago to a small lake about half
‘a mile long and one-fourth of a mile wide at Santiago village, but beyond this to the gulf during
the dry season the channel is usually dry. A broad dry wash leads for several miles down a
slope from the southerly end of the Sierra San Lazaro to the flat near the beach at Cape San
Lucas. Some very small streams flow a short distance in canyons at the head of this drainage
channel, but are quickly lost in the sand. The abundance of fresh water near the surface
on the flat back of the beach at San Lucas gives evidence of a considerable underflow down this
channel,

In Lower California fresh water occurs in tinajas, or natural rock tanks, much more often
than elsewhere. A tinaja is almost always a pothole or small depression worn in the rocky
bottom of a gulch or small canyon in which rain water collects, and in the larger ones it is often
held’ through several dry seasons. But for the existence of these natural tanks, which ordi-
narily vary from a few feet to 6 or 8 yards across and of any depth up to 10 or 12 feet, travel
would be practically impossible over a large part of the peninsula. Exceptionally large tinajas
occur here and there, like the Tinaja de San Juan, a few miles northerly from San Ignacio.
This is an oblong basin, more than 100 yards long and 10 to 20 yards broad in the bottom of a
shallow drainage gliannel on a flat lava-covered mesa. At Yubay the tinaja is a rounded pot-
hole in the granite bedrock at the mouth of a canyon cutting through a low granite ridge. The
Tinaja de Santa Clara is a large oblong pothole in the bed of a small canyon cut in the lava of
Santa'‘Clara Mountains. All of these tanks are readily accessible for man or beast, as are hun-
dreds of others here and there throughout the wilderness of barren mountains on the peninsula.
Others, however, are located high up in rugged and almost inaccessible canyons or in deep narrow
holes so walled about that they can be found only by accident and the water obtained only by
use of arope. The few tinajas in the Cocopah Mountains are reported to be of this character.
Tinajas are more numerous in gulches and small canyons than in large ones, because in the
latter the mass of wash material carried by floods fills up the depressions in their beds.

Ojo de Liebre, near the head of Scammon Lagoon, on the Vizcaino Desert, is permanent,
and Funcke reports that when grass is plentiful on the surrounding plain a considerable number
of half-wild horses water there This spring is marked by three cottonwood trees. (See Pl.
12, Fig. 1.)

UNDERGROUND WATERS.

Lower California is characterized by a great number of watercourses leading from the high
interior down to the sea. Many of these are broad enough to accommodate fair-sized rivers,
some even reaching more than half a mile in width, and nearly all have deep beds of sand and
84 MEMOIRS NATIONAL ACADEMY OF SCIENCES.. - (Von, XVI

gravel which at ordinary times are dry throughout the larger part of their, courses. Most of
the larger washes head well up in, the high interior and some have running water in their channels
for miles, while others have running water at one or more places where it has been forced up
by the bedrock only to sink a short distance below. An underground flow in the beds of such
streams i is thus proved, and it evidently exists in practically all of the larger channels, although’
for miles there may be no surface indication. The giant springs at San Ignacio, Mulege, Todos
Santos, and San Bartolo, all rising in dry drainage channels, are examples of the amount of
this underflow i in some cases. Nearly all of the drainage channels are dry, on the surface near
the coast, but digging at their mouths on the flats, just back of the beach, has developed good
water in many places within a few feet of the surface, as at San Felipe Bay, ( ‘Calamahue Landing,
La Paz, Cape San Lucas, and elsewhere. It is certain that water can be found i in this manner
at the mouths of many other washes. In addition some of the coastal plains are more or less
generally underlaid with water at depths varying from 10 to 60 feet, as on the San Quintin and
San Bruno Plains, about La Paz Bay, and at Cape San Lucas. Artesian water is reported to
have been struck at a depth of about 1,200 feet near the mouth of La Purisima Canyon and will
undoubtedly be found in various other localities.

The general ‘“‘lay of the land” and character of the vegetation frequently give very positive
indications of. the presence of underground water. An unusually heavy growth of mesquite is
a good indication of water. Such a fine growth of mesquites occurs a.few miles above Mulege,
in the bottom along El Potrero wash. A similar growth indicates water over a considerable.
area near Santo Domingo ranch on Magdalena Plain. The growth of fig and other planted
trees on the ranch south of the San Simon River near San Quintin also indicates. water a few
feet below the surface. Water has been found in shallow wells at a number of ranches on
Magdalena Plain and undoubtedly exists in many otber places.

The water holes known as Tres Posos between the north ends of the Sierra Pinta and. Sierra
de las Tinajas, and in the flat at San Felipe Bay are only about 3 feet deep. They contain
fairly good water and probably mark the underground flow from Arroyo Grande and from the
San Pedro Martir Mountains, respectively.

The well at Calmalli i is said to be about 50 feet deep, that at Campo Aleman 90 feet, and
that at San Augustin. on the Llano de los Buenos Aires about 50 feet, With cheap fuel or
electric power many small or medium-sized ranches might be maintained by water pumped
from these underground supplies.

COASTAL ISLANDS. tee.

A long series of islands border the coast of Lower California, beginning with the Coronados
group on the Pacific side near the northern border and extending thence around the coast, to
the head of the Gulf of California. From, their proximity to the coast and the comparative
shallowness of the intervening water, and the close similarity or identity of the geological
structure on the islands and the adjacent shore, it is apparent that most, if not all, of them
once formed part of the mainland of the peninsula. In addition, the plant and animal life in
general are so nearly identical with those of the adjacent shore of the mainland as to complete
the evidence of their origin. Some of the islands of the gulf, however, have a few notable
species not represented by close relatives on the peninsula. The most striking of these among
the plants are the viznaga (Echinocactus digueti) and among the reptiles, an iguana (Ctenosaura
hemilopha) and a chuckwalla (Sauromalus hispidus).

With few exceptions the islands on both coasts of the peninsula are. high and rocky with
precipitous shores and few landing places. Severe desert conditions usually prevail and. most
of the islands lack fresh water. The islands of the southern end of the gulf are much better
known then those farther north. For this reason few of the gulf islands are described here.
Some of the smaller islands are of special interest on acount of their breeding places for water-
fowl. Among these the island of San Ildefonso is known to be the great breeding center of
Heermann’s gulls. Although situated in the middle of the gulf, the island of San Pedro Martir
is the breeding resort of the blue-footed and Brewster’s boobies. While Guadalupe Island is
No. 1) LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 85

so far offshore that it can not be classed among ,the coastal islands, it is included here in order
to complete’ the account of the islands off the Pacific shore of the peninsula:

Coronados Islands. :

-: This group is a series of four small rocky ‘diets about:5 miles long, lying in a northwest by
southeast direction, a few miles south of the northern boundary and about 7 miles offshore on
the Pacific side of the peninsula. They are within plain view and about 15 miles southwest
from the entrance to San Diego Harbor, southern California. They all rise boldly from the
sea, their shore lines consisting mainly, of wave-washed cliffs. The southernmost and largest
of the group is about 2 miles long north and south and half a mile broad. Its shore line is
formed by high cliffs except in a small bay in the middle of its eastern shore where boats can
land. The summit of the island is covered with rounded hills rising to an altitude of 672 feet
near the southern end. The two middle islets are barren rocks rising precipitously 100 and 250
feet above thesea. The second largest, the most northern of the group, about a mile long and
one-quarter mile wide, is surrounded by cliffs except on the southern side, where a landing
can be made. This island is a narrow rocky ridge covered with sharply sloping hills rising
over 460 feet above the sea.

These four islets are evidently the remnants left from the erosion of a former single large
island. No fresh water is found on any of them. It is reported that the middle islets once
formed a favorite resort for the sea elephant. At present a few sea lions and harbor seals haul
up on the shores. The islets are treeless and have only a few small shrubs, but are covered

with an abundant growth of herbaceous vegetation. The southernmost is overgrown with
wild oats, and the northern one has much of its southern slope covered with a mat of the curious
ice plant (Mesembryanthemum erystalinum). The, only land mammal reported from here is
a white-footed mouse (Peromyscus maniculatus dubius).

The Coronados are the breeding places of petrels, auklets, and other sea fowl, and their
proximity to San Diego has caused them to be repeatedly visited by American ornithologists,
ineluding A. W. Anthony, J. Grinnell, A. B. Howell, the writer, and others. A song sparrow
(Melospiza melodia coronatorum) has been described as peculiar to this limited area. Many
land birds, especially the California Valley quail, fly back and forth over the short distance
to the mainland. Several lists of the birds of the Coronados have been published and about
50 ‘Species are known to occur there.

Todos Santos Islands.
This is a small group of two islets and outlying rocks extending in a northwesterly and
southeasterly direction immediately in front of Todos Santos Bay, off Ensenada, and only 3
miles from and in line with Punta Banda, the cape at the southern side of the bay. A long line
of detached rocks stretch from the point, of the cape directly toward the largest islet, with which
they are connected by a submarine ridge, showing that the islets were once part of a long narrow
cape, of which Punta Banda is the remaining base. The southernmost and largest of the
islands is about 1} miles long and half a mile wide. At its northern end this islet forms a low
plateau, but to the south rises into rounded hills a little more than 300 feet high. The shore
line is rocky and precipitous, rising in cliffs 20 to nearly 200 feet high, and the only landing is
on the northeastern side. The smaller islet lies northwest of the larger one, from which it is
separated by a narrow channel. It is entirely a low plateau, a continuation of that forming the
northern end of the larger islet, and has an extreme elevation of 55 feet above the sea, and its
shore line is almost continuously precipitous. There is no fresh water on either islet. Both
are covered with an abundant growth of herbaceous vegetation and in sheltered places on the
larger islet are small patches of low shrubs. Cholla cactuses (Opwntia) are extremely abundant
on the more level areas, necessitating some care in avoiding them. These islets are so near the
mainland that the conditions are nearly the same as there except that they receive a little more
moisture from the damp sea winds. The only two land mammals on the islets are a wood rat
'(Neotoma anthony) and a white-footed mouse (Peromyscus maniculatus dubius), both of which
‘are extremely numerous and are close relatives of forms living on the adjacent mainland. A
86 . MEMOIRS NATIONAL ACADEMY OF SCIENCES: [vor. XVI

few harbor seals and sea lions were seen about the shore, and sea otters were formerly numerous
in the adjacent kelp beds. There are no birds peculiar to these islets; but a few species of
waterfowl breed here and the nest of a bald eagle was found on the top of a rocky pinnacle just
off the northern end of the larger islet. Fifty. speciés of birds have been observed on the Todos
Santos Islands, most Of them of; pepomale occurrence, : cone western gulls and cormorants breed
there.

San Martin Island.

‘San Martin is a small rocky island about a mile in diameter and rounded-triangular in form,

lying a few miles offshore to the northwest of the mouth of San Quintin Bay. On the northeast
side the shore is low with a small sandy beach and a small rounded tidal lagoon where sea éle-
phants and sea lions formerly resorted in large numbers to breed. Elsewhere the shore i is rugged
and precipitous, rising from a few feet to over 150 feet above the sea, the highest cliffs facing
south. The entire surface is covered by extremely ragged and broken beds of lava, full of
“numerous pits, caves, jagged points, and sharp ridges, over which it is extremely difficult to
travel. The middle of the island is occupied by two volcanic peaks, the highest reaching an
altitude of about 470 feet, from which the. slopes extend down on all sides to the cliffs forming
the shore line. One of these peaks contains an old crater. There is no fresh water on this island.
The lava is covered with an abundant growth of lichens. The most notable flowering plants
are some cactuses and a large velvety white cotyledon with orange red flowers (Dudleya
anthonyi). .

San Martin is one of the guano islands of this coast, ,and we found it occupied as a breeding
place by colonies of cormorants and brown pelicans. Some extraordinarily large fish-hawk
nests were perched on projecting knobs of lava a few feet above the general level. On May 4,
1903, A.S. Bunnell saw gopher snakes and “alligator lizards” here. The only land ‘mammals

are a wood rat (Neotoma martinensis) and a white-footed mouse (Peromyscus maniculatus
geronimensis).

This island was once noted as a great resort for fur seals and sea elephants, both of which
are now extinct there. The exploring ship Herald was off San Martin in October, 1846, and Dr.
Seemann records that ‘‘the rocks were swarming with seals; we had never seen them in greater
abundance; and their howling, shrieking and barking rendered them most unpleasant neigh-
bors.’”! During his visit to San Martin on May 4, 1903, Bunnell found several great blue herons
nesting there, but they were absent at the time of our visit in August, 1905. . He also noted the
presence there of about 40 capreye About 30 species of birds have been recorded from this
island. soe eee
San Geronimo Island. ;

This island lies 5 miles offshore near the south side of Rosario Bay. It is a low rocky islet
about three-quarters of a mile long and less than one-third of a mile wide, surrounded by low

cliffs 10 to 50 feet high, above which the surface forms a plateau sloping up to a hill about 130
feet high near the middle and to two smaller hills near the northern side. The islet is waterless
and surrounded by kelp ‘beds and outlying rocks. Its vegetation is very scarce and stunted.
The only land mammal is the white-footed mouse (Peromyscus maniculatus geronimensis), which
is extremely abundant. Numerous harbor seals frequent the reefs on the northwestern side.
Twenty-eight species of birds have been observed here, of which auklets, gulls, shearwaters,
cormorants, and pelicans breed. The surface of most of the level parts of the island is of loose,
soft soil, and at the time of our visit in August, 1905, was honeycombed with ‘the old burrows
of shearwaters, but none of these birds were ‘there at the time. San Geronimo is another well-
known guano island of this coast.

Cedros Island.

_ Cedros Island \ was discovered by Francisco Ulloa in 1539 and named by him Isla de Cedros;
hence. the. commonly used. “Cerros”” Island i is a misnomer.” Although discovered at so early a

 

Be 1 Seeman, Voyage of the ‘Herald, Narrative, I, pp. 119-120, 1853. 2 Eisen, Bull: Amer. Geog. Soc., XX XIII, pp. 64-66; 1901.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 87

date by the Spaniards, Cedros: Island has been explored and made known almost entirely by the

-yisits of American naturalists within the last 60 years. J. A. Veatch spent some time there in
1859. Lyman Belding visited it in October, 1882, and A. W. Anthony at different times be-
tween 1887 and 1897. W. W. Brown was there during most of April, 1906, and the Townsend
expedition on the Albatross visited it in the spring of 1911. With the exception of the botanical
exploration by Dr. Veatch and Dr. J. N. Rose of the Townsend party, most of the work on the
island has been the collecting of birds and mammals.

_ Cedros Island lies about 10 miles, off Point San Eugenio on the mainland; it is roughly
wedge-shaped, having a length of 21 miles north and south and a breadth of about 3 miles near the
northern end and 9 near the southern end. The island is abruptly mountainous, Cedros Peak,
the highest point, reaching an altitude of about 3,950 feet, while other peaks give it an irregular
and broken interior. Much of the island rises above 1,500 feet and its surface is rocky and
sandy. Dr. Veatch states that the island is formed by three nearly parallel mountain ridges
extending diagonally across it from northeast to southwest, the northern ridge being of granite
and others of shale, sandstone, and other rocks.!. He adds that most of the eastern shore of
the island is precipitous, often rising abruptly several hundred feet, but its front is broken
in many places where there are good landings. Elsewhere.the coast line is abrupt and rocky.
The northern part is apparently most fertile, especially on the western slopes and crests of the
mountains exposed to the cool damp northwest winds. The crests and slopes of the mountains
on the northwestern ‘point of the island down to about 1,500 feet are covered: with a growth
of cedars (Juniperus californica) and pines (Pinus .muricata), some of the latter attaining a
height of 50 to.70 feet. . A species of dwarf oak occurs and numerous other shrubs and flowers,
including the California holly (Heteromeles arbutifolia), which grows in the pine forest. On
the more desert eastern and southern slopes of the island the elephant wood. (Pachycormus dis-
color), various cactuses, and other desert vegetation like that of the adjacent mainland prevails.
There are two main watering places on the island, the one most used being near the beach on
the eastern side, where at a point marked ‘‘watering place” on the charts fishermen and others
obtain water from a spring which trickles out of a rock a few feet above sea level and is marked
by a patch of bulrushes. Mr. Belding ; notes that there are other patches of bulrushes on the
same shore of the island, which would indicate that water might be found at these points. 2
Water is found also in the interior at a mining camp and elsewhere near the northern end.

In writing of Cedros Island Capt. Scammon tells us that seals and sea elephants once
basked on the shores of this isolated spot in vast numbers and that in years past its shores teemed
with sealers and with sea elephant and sea otter hunters. The remains of the rude stone houses
once occupied by these hardy men still exist in many convenient places. One of these houses
on the west side was of large dimensions, a rough measurement making it about 40 feet square.’

Gold and copper mining has been carried on in a small and desultory way on this island
for. years but has not. resulted in any permanent settlement.

All the known species of land mammals on Cedros are peculiar to the island, types of
which were collected’ by Mr. Anthony. They include a mule deer, a cottontail rabbit, a wood
rat, a white-footed mouse (Peromyscus), and a pocket mouse (Perognathus).. Mr. Anthony

writes me that he has seen signs on the island of a pocket gopher (Thomomys), as well as the tracks
of a fox, and has heard reports of the presence of a rock squirrel (Citellus) about the tops of the
mountains. ‘The fox tracks might have been. either those of the gray fox ( Urocyon) or the
small desert fox (Vulpes).

The exploring ship Herald stopped at Godros 3 in October, 1846, and Dr. Seemann reported
wild ‘goats. to be abundant, there.: The, deer. have been killed, to supply mining camps with
meat until the existence of. the species has been endangered. The sea otter, sea elephants, ,
and fur seals which were once so numerous about its shores are now gone. Only 40 species
of birds have been observed, on Cedros, and owing to the proximity of the island to the main-

 

1 Hesperian, III, pp. 532-533, 1860. Higgins
- -2 Proc. U.S. Nat.-Mus., V, pp. 530-531, 1883. “
‘8 Scammon, in Appendix to Browne’s “‘ Resources of the Pacific Slope,” p. 129, 1887.
88 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI

land few. peculiar forms are known. W. W-. Brown records that in April, 1906, he found birds
not.only extremely scarce on, the island but very shy. He collected. three species of snakes
and five of lizards and saw no others.! _ ;

San Benito Islands.

‘The San Benito group consists of three small barren islands evidently formed by the erosion
of a single large island, which at. remote period was undoubtedly connected with the main-
land by Cedros Island. This group lies 15 miles west of the northern point of Cedros and is
about 5 miles across from ‘east to west by nearly 2 miles from north to south. - According to
Bunnell these islands are formed of a flaky shale and the more level areas are covered with a
brown powdered earth.

West Benito, the largest island, measures stands 2 miles from east to west by 14 miles from
north to south.’ Its shores are bold and rocky and the surface consists of an elevated plateau,
with a rounded hill near its middle rising 660 feet above the sea. Middle Benito, the smallest
of the group, is a low flat island, its highest part being only 82 feet above the sea. East Benito,
the second in:size, has four prominent hills, the highest of which reaches an altitude of about
420 feet. There is no fresh water on any of these islands. -

The vegetation consists mainly of some cactuses, agaves, and small shrubs. A single
species of lizard is found here. Twenty-six species of birds are known from these islands, of
which McGregor’s ‘house finch andthe San Benito sparrow are peculiar to this limited area.
These islands are favorite nesting places for Cassin’s auklet, Xantus’s murrelet, and the black,
Socorro, and least petrels. No land mammal occurs there, but the shores formerly abounded
with sea otters, fur seals, and sea elephants, all of which are now gheent: In May, 1903, Bunnell
saw about 300 sea lions there and some harbor seals.

- The San Benitos have been: repeatedly visited by aaties ae iolonisen including
AW. Anthony, W. W. Brown, and membérs of expeditions'sent out Py the California Academy
of Sciences.

Natividad Island. ‘

_ Natividad lies about 4 miles northwestward from Point San Eugenio on the mainland and
is almost in a line between this point and Cedros Island. ‘It is 334 miles long in a northwest
to southeast direction and varies from half a mile in width at its northwestern end to about 114
miles near the southeastern end. Itis barren and hilly, rising to a height of nearly 500 feet in the
middle and.bordered mainly by steep rocky shores. On the southeastern end there is a sandy
beach about half a mile long. Large areas of the island are covered with loose sandy soil, dotted
with the burrows of the black-vented shear water and Cassin’s auklet. Colonies of cormorants
also frequent Natividad, which is one of the well-known’ guano islands of this coast. There is
no fresh water here. Vegetation is scarce, but giant cactus and a smaller cactus are found and
a few shrubs are generally distributed. The ice plant (I esembryanthemum) i is the most abun-
dant plant and carpets much of the island. One or more species of small lizards live here. The
first notice of bird life on Natividad Island is in the record of the visit made by Padre Taraval
in 1732.2. In 1897 A. W. Anthony visited the island and later published his observations on the
colony of shearwaters whose burrows are so notable a feature of this spot. An expedition from
the California Academy of Sciences visited the island in May, 1903, and in April, 1906, W. W.
Brown spent a short time there and afterwards his observations were published in the Condor
for May and June, 1907. Only 16 species of birds are known from this island, although numerous
others must occur occasionally. The principal birds are Cassin’s auklet, the black-vented shear-
water, the California brown pelican, the cormorant, and the western and Heermann’s gulls,
This island is notable as'the nesting place of thousands of black-vented shearwaters.

The only land mammal on Natividad,'a white-footed mouse (Peromyscus maniculatus
geronimensis), is extremely abundant. Sea otter, sea elephants, and Guadalupe fur seals once
frequented its shores in large numbers but are now gone. A few sea lions and harbor seals

still occur.

1 Condor, IX, pp. 77-78, May-June, 1907. 2 Venegas, Noticias de California, II, p. 436.
No.1}. | LOWER amr econ NATURAL RESOURCES—NELSON. 89

San Roque Island. _

San Roque is is a rocky: islet lying 2 miles oe nore in front of San Roque Bay. Itis approx-
imately three-quarters of a mile long by 1.mile wide and rises about 50 feet above the sea. It
has no fresh water. Like most other islands along this coast, it is a breeding place for cormo-
rants and a few sea birds.. A single land mammal, the white-footed mouse (Peromyscus mani-
culatus cineritius), is peculiar to it.

Asuncion Island. M ae

This island lies less than 6 ‘nls couthisist of San Roque and three-quarters of a mile south
from Asuncion Point on the mainland, which forms the southeastern limit of San Roque Bay.
It is only a mile long north and south aad a quarter of a mile wide. Asuncion is a barren island
of sandstone formation, the highest point rising about 165 feet above the sea at its southern end.
It.has no fresh water. It is frequented by a few species of sea birds, but has no land mammals.

Magdalena Island.

This long, narrow, irregular island, over 50 miles long and from 1 to more than 5 miles
wide, lies close inshore and nearly parallel to the coast of the peninsula. The northern three-
fifths of the island extends in a northeasterly by southwesterly direction, separated from the
mainland by a narrow tidal lagoon varying from half a mile.to 3 or more miles wide; the southern
two-fifths of the island turns sharply in a northwesterly by southeasterly direction, forming
Cape San Lazaro at the:angle and inclosing the western side of the well-known Magdalena Bay.
The narrow tidal lagoon separating Magdalena and other sandy islands to the north from the
mainland is from 1 to 5 fathoms deep. It is navigable for boats throughout its length and is
the regular route by which produce is brought down to the bay from as far north as Comondu
Valley.

With the exception of a solitary rocky hill rising to an altitude of 1,275 feet close to Cape
San Lazaro, at the western angle of the island, and a narrow rocky ridge 10 or 12 miles long and
1,250 feet high, occupying the extreme southern end, Magdalena Island is composed of low sand
dunes and sandy flats rising from tide level to an elevation of about.25 feet.. The two above-
named rocky elevations appear to have been formerly separate islands which have been united
by the low sandy plain built by the sea.

The high rocky ridge which forms the southern end of the island shuts in the west side of
Magdalena'Bay from the Pacific and ends as a bold headland on the west side of the entrance
to the bay. ' On the west side of the bay near the northern end of this ridge is the village of Mag-
dalena Bay. (See frontispiece.) A line of sand dunes form the edge of the low mainland plain
on the northern and eastern shores of the bay. The tidal lagoon separating the northern three-

fifths of the island from the mainland is extremely irregular and contains scattered islands.
At its southern end it opens out to form Magdalena Bay. It varies from 1 to 5 fathoms in
depth and at especially low tides can be forded near the middle by men and horses.

There is no good fresh water on this island, but brackish water can be obtained by digging
in the flats north of the village.

Magdalena Bay is about 15 miles across and is almost completely landlocked. For years
it has been the favorite winter practice resort for the American Navy. About the middle of
the last century it was a favorite resort for the whaling fleets operating on this coast and later
became the headquarters for the orchilla industry. An American company, known as ‘The
Chartered Co.,” of Lower California, held a concession covering a coastal strip several leagues
broad, extending north toScammon Lagoon. The only business being done at the time of our
visit was the selling of a few cattle to vessels calling at the bay and the shipment of some sea tur-
tlestoSan Francisco. Theborders of the lagoon north of the bay andislands in it arecovered with
mangroves. The remaining vegetation of Magdalena Island consists mainly of some of the
desert species which grow on the adjacent Magdalena Plain, but the giant cactus and some
other prominent species of the plain are absent.’ On sheltered slopes and benches north of the
village, near the northern end of the high rocky ridge, there is a good growth of desert vegeta-
tion, including among’ numerous others such common genera as Abutilon, Fouquieria, Jatropha,
90 ’ MEMOIRS NATIONAL ACADEMY: OF SCIENCES. :':’ [Vou XVI

Pedilanthus, Terebinthus, Wilcoxia, Opuntia, and Lamaireocereus; elsewhere, especially 01 on the
rocky slopes’ vegetation is much smaller and more sparingly distributed. °' A

The birds of'this island are the same as those on the adjacent coastal plain of the snistitadta
A few of the mammals are peculiar to the island, including a pocket mouse, a pocket gopher,
and a white-footed mouse. Jack rabbits occur here and on ‘Santa Margarita Island. Coyotes
are extremely common, and occasionally mule deer occur near the northern end. While the
climate of Magdalena Island i is somewhat modified by the cool breezes of the Pacific, yet it is
so far south that it has much less fog than the coast farther north and ‘being extremely dry arid
tropical conditions prevail.
Santa Margarita Island.

Santa Margarita Island lies southeast of the southern end of Magdalena Island and incloses
Magdalena Bay on thesouth. Its northwestern end presents a bold headland facing the southern
end of Magdalena Island across a channel a little more than 3 miles broad, forming the entrance
to the. bay. From this point Santa Margarita extends nearly 25 miles southeasterly approx-
imately parallel to the trend of the mainland coast and varies from 2 to 5 miles in width. The
island is mainly mountainous in character, the barren rocky slopes rising: to peaks along, a crest
extending lengthwise of the island and varying from 1,000 to over 1,800 feet high, but separated
into two sections as described below. The shore line is almost entirely rocky and precipitous,
except in the middle, where a low pass more than a mile broad in its narrowest. place and about
100 feet above the sea separates the island into two nearly equal.islandlike parts. On the north-
eastern side ofthe island this pass is a little more than a mile wide, but broadens to a low plain
extending more than. 5 miles along the southwestern side between the mountain ridges and the
shore. On the inner or northeastern side of the pass are some small lagoons full of mangroves
where cormorants and man-of-war birds nest. Close by are the ruins of a small experimental
factory built long ago for canning sea turtles. | This business failed after a short trial. Through
the pass on the seaward or southwestern side the low plain is bordered along:the beach by a
belt: of sand dunes. On the flats just inside the dunes, fresh water lies close to the surface.
We found there a deserted ranch with a few date palm trees pearly prinied 4 in the eens
sand dunes.

‘Santa Margarita.is generally ‘barren and desertlike in dbarenter with a scanty vegetation
derived from the mainland shore. The hills and mountain slopes are extremely barren. The
low area in the middle contains a more abundant vegetation, mainly the same species observed
on Magdalena Island. Among the most notable species seen were the, giant cactuses (Pachy-
cereus calvus and Pachycereus pectenaboriginum), the ‘‘pitahaya agria’’ and ‘‘pitahaya dulce”
(Lemaireocereus gummosus and L. thurberi), the ocotillo (Fouquieria pentneulans), Jatropha
canescens, and Pedilanthus macrocarpus. = | ii

The few land birds noted were the same as those of the mainland, such as caracaras, fish
hawks, sparrow hawks, gilded flickers, Lower California flycatchers (.Myiarchus), Xantus jays,
ravens, rock wrens, and San Lucas cactus wrens. As already noted, the. man-of-war birds and
Farallon cormorants (Phalacrocoraz dilophus albociliatus) nest on the mangroves about the
lagoon on the inner side of the island... California brown pelicans also nest in considerable
numbers farther south on the island. nee

Margarita Island, incloses, on its northeastern, side Almejas Bay, which communicates
with Magdalena Bay by a narrow channel in which lies Mangrove Island. This is a, low sandy
island overgrown with mangroves and forming a great nesting place for cormorants and some
other waterfowl. Near the head of Madgaleng: Bay is a low, flat, and bare sandy island only
alittle above tide level, which is a resort for enormous numbers of cormorants, pelicans, and gulls.
Owing to, the number of sea fowl which make this a Tesbink place and which formerly bred there,
this i is a guano island.

, Although, like Magdalena Island, so near the coast and having the same general character
of vegetation’ and climatic conditions, Santa Margarita contains several forms of small mammals
peculiar; to it... ‘The jack rabbit. (Lepus californicus magdalenge) is somewhat different from
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 91

its relative of the adjacent shore. The wood rat is similar to the one on the mainland. The
white-footed mice, pockét mice, and kangaroo rat are all forms peculiar to this island.

Ceralvo Island.
This is the southernmost of’ the islands in the Gulf of California. anal the nearest point

lies 4 miles off the eastern coast of the peninsula southeast of La Paz Bay. Ceralvo is over 15
miles long approximately northwesterly and southeasterly and 4 miles wide. The southern
end of the island is within 3? miles of Ventana Point, the island lying on the eastern side and
partly inclosing Ventana Bay. Ceralvo is occupied by a single high mountain ridge extending
nearly its entire length and reaching altitudes of 2,265 and 2,477 feet near the northern and
southern ends, respectively. The slopes of the mountains are steep, and almost the entire
shore line is high, rocky, and precipitous. At the extreme southern end the island descends
to lower slopes and ends in a limited sandy flat on the shore backed by low benches and ascending
slopes. This affords the only comfortable landing place.

The island is extremely arid and desert in character and has no fred water nor inhabitants.
Deposits of old pearl oyster shells on the sandy flat at the southern end indicate that the island
was once the resort of pearl fishers.

Ceralvo appears to have been connected with the mainland through the extension of the
southern end of the island to Ventana Point, as indicated by the shallowness of the sea between
them. There is a much greater depth between the western side of the island and the opposite
part of the mainland. That this island was once part of the mainland is indicated by the
identity of the geological | formation on their adjacent shores.

The vegetation is a. mixture of Arid Tropical and Lower Sonoran species, mostly identical
with those of the neighboring mainland. The great viznaga (Hchinocactus digueti), however,
appears to be peculiar to this and, other islands of the gulf. Other prominent plants noted
here are Olneya. tesota, Fougquieria peninsularis, Elaphrium microphyllum, Brayodendron texanum,
Ibervillea. sonorae, Jatropha spathulata, Pedilanthus a a Lemaireocereus thurberi, L.
gummosus, Pachycereus calvus, and some opuntias.

This island is noted for the occurrence of many iguanas which are not known from the
mainland of the peninsula but do occur on some of the other islands farther north in the gulf.
Fish | hawks, sparrow hawks, ravens, black-throated sparrows, San Lucas house finches, and
rock wrens were among the few land birds noted. here.

Only two land mammals are known from Ceralvo, the pocket mouse (Perognathus penicil-
latus, siccus) and the rock mouse (Peromyscus eremicus avius). Both forms are peculiar to
the island and closely related to mainland species.

Espiritu Santo Island.

Espiritu Santo Island walls in a part of the eastern sie of La Paz Bay and is about 10
miles long north and south and 4 miles wide in its broadest part. It is irregular in outline
and almost:entirely covered with rocky hills and low mountains reaching an altitude of nearly
2,000 feet above the sea. ‘The shore line is mainly rocky and precipitous in places fronted by
high cliffs. The northern quarter of the island is nearly separated from the rest by two narrow
indentations formed by the sea. On the west side, near the southern end, is a small bay with’
a small mangrove swamp’ on the flat near its head. A narrow valley extends inland from
this flat and forms part of a pass across the island, extending between low rounded hills and
mesas. ‘The southern end of the island is separated from the opposite point of the peninsula
by San Lorenzo Channel, a shallow strait about 33 miles broad and 5 fathoms deep. The geo-
logical formation of the island appears identical with that, on the nearest point of the mainland,
and the shallow water separating them indicates. the existence, at no remote period, of a former
land connection between: the two.

Espiritu Santo has no fresh water except in natural tanks after rains, but at the time of
our yisit small houses had been built on the shore of the bay mentioned above by a company
organized. at, La Paz and apparently successful experiments were being made in the artificial
propagation ¢ of pearl oysters. The valley leading back from the head of the bay and favorable
92 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

places elsewhere on lower slopes had many scattered Arid Tropical and Lower Sonoran: shrubs
and other plants, but the upper hill slopes are rocky and extremely barreti.

The vegetation of the island is practically identical with that of the Aeesies
and mostly Arid Tropical in character with a mixture of giant and other cactuses. Among the
prominent species noted were Tapirira edulis, Elaphrium microphyllum, Forchammeria watson,
Bourreria sonorae, Jatropha canescens, Pedilanthus macrocarpus, Lemaireocereus Pe:
L. thurberi, Pachycereus calvus, several opuntias, and Pereskiopsis brandegeet.

: The hands were some of ihe more common species of the adjacent part of the siete
including the fish hawk, the sparrow hawk, the San Lucas and cape woodpeckers (Dryobates
and Centurus); the raven, canyon and rock wrens, Gambel ae cape verdin, and western
mockingbird.

This island has a larger variety of mammal life than many of the other dlanida along the
coast, probably owing to its more recent. connection with the mainland. The five species and
subspecies occurring here are peculiar to this area but closely related to forms common on the
adjacent mainland.. The jack rabbit of Espiritu Santo’ is unquestionably derived from the
Lepus californicus stock of the peninsula; through isolation it has developed into an extraor-
dinary black animal, specifically distinct.. The other mammals. of the island, including repre-
sentatives of Perognathus, Ammospermophilus, Neotoma, ' Peromyscus, and Bassarisous, show no
such striking differences from their mainland relatives.

San Jose Island. : ,

San, Jose is located immediately north of La Paz’ tay and is separated from the mainland

by a channel from 3 to 5 miles broad. It is over 16 miles long northwest by southeast and
about 7 miles in greatest breadth. The island is covered with hills and low mountains reaching
an altitude of over 2,000 feet. The shore line is bold and rocky, and the mountain slopes rise
from the sea éxcept on the southern end, where the hills become low and the coast is mainly
bordered by a sandy beach. A lagoon lies back of the beach near the southwestern point and
contains a mangrove-covered island. A single small spring is known in the interior and fresh
water also exists at the head of the bay r near the lagoon at the southwestern point. Sand beaches
occur at several points along the west coast and a ranch is located on the shore near the middle
of this side of the island. The vegetation of San Jose i is practically the same as that of the
adjacent mainland but averages rather smaller. It is most plentiful i in ‘the canyons: and other
drainage ways, the rocky higher slopes being extremely barren.
“Over 40° species ‘of land and water birds are known from this ‘sland, The water birds
include all the species known from this part of the gulf. The:land'birds are the same as those
of the adjacent mainland. The mangrove island in the lagoon at the southwestern point is a
breeding place for the San Lucas blue heron and other water birds; and Cayo Island, just
offshore from the southwestern point, is a breeding place for Craveri’s murrelets. ‘The mammal
fauna of San Jose is more extensive than on any other island off the coast of Lower California
except Cedros. It includes the peninsula mule deer (Odocioleus hemionus peninsulae), the San
Jose brush rabbit (Sylvilagus mansuetus), a kangaroo rat (Dipodomys insularis), a pocket
‘mouse (Perognathus spinatus bryanti), a wood rat (Neotoma intermedia pallida), and a bassarisk
(Bassariscus astutus insulicola). The deer is supposed to be the same as that of the adjacent
mainland. The other mammals are all peculiar to the island, but are close relatives of forms
occurring on the adjacent shore.

Carmen Island.

This large island lies about 125 miles northwesterly up the coast from La Paz and about
5 miles offshore. It is long and narrow except at the northern end, where it broadens, and a
large spurlike point extends at an angle to the southeast, inclosing Salinas Bay. The island
is about 17 miles long i in'a northeasterly by southwesterly direction and over 5 miles broad at
the widest part near the north end. A range of low peaks from 500 to 1,500 feet high extends
the entire length and the shore line is mainly rocky and précipitous, but at the heads of several
small bays there are sandy beaches with the lower slojies of the hills and mountains rising behind
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 24.

 

 

 

 

 

Fic. 1.—E] Tule Arroyo, a clear stream in canyon at junction of Sierra de las Tinajas and Sierra Juarez. (Photograph by Dr. W. J. Mixter.)

 

,

 

 

 

 

Fia. 2.—Tinaja Santa Clara in the Santa Clara Mountains on the west side of the Vizcaino Desert. This is a natural tank in
: the bottom of a canyon through a lava bed.
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 25.

 

 
 
 
 

ih,

i EEE tee

 
 

 

 

 

 

 

Fig. 1.—Rattlesnake Harbor, on southeast shore of Angel de la Guarda Island. Mostof the shore line is rocky and precipitous. (Photograph
by L. C. Decius.)

 

 

 

 

 

 

Fia. 2,—Sea Elephant Cove, on northwestern shore of Guadalupe Island. Note the heads of sea elephants in the water.
(Photograph by F. E. Norton.)
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 93

them. In the angle formed by the projection on the eastern side of the island near its northern
end lies Salinas Bay, at the head of which is the noted salt lagoon, where a good quality of table
salt has been mined for many years. Here a small village is located, inhabited by people who
work the salt deposits. Fresh water is found here, and about halbyay. from Salinas Bay to
the southern end of the island on the east coast is a small stream of fresh water.

Carmen Island has a rather plentiful Arid Tropical flora consisting almost entirely of the
species common to the adjacent shore of the peninsula, including numerous large cactuses
belonging to several genera in addition to the great viznaga (Echinocactus digueti) peculiar to the
gulf islands.

The birds are the common species of land and water fowl of the gulf and its adjacent shores.

Only three mammals, all small rodents, are known from the island, all of which are forms
peculiar to it but close relatives of mainland species. These include a Perognathus, a Peromys-
cus, and a Neotoma.

Angel de la Guarda Island.

This is the largest island on the coast of Lower California, having a length of about 40
miles and an extreme width of 10 miles. It lies parallel to the coast in a northwestern by
southeastern direction, from 10 to 15 miles offshore, well up in the northern part of the gulf and
nearly opposite Tiburon Island of the Sonoran coast. The island is unusually high and moun-
tainous, with a rocky ridge extending its entire length and reaching altitudes from 3,000 to
4,000 feet. The mountain ridge is highest at the two ends and separated by a comparatively
low depression or, pass in the middle. The shores are mainly very rocky and precipitous, the
western coast being inaccessible, but there are several landing places on the eastern side and
a low area and a small bay at the northern end. The island is extremely rocky and barren,
but considerable scattered vegetation of the same species found on the adjacent mainland of
the peninsula grows there, including many cactuses. This island is reported to abound in the
large black chuckwalla (Sauromalus) and rattlesnakes. A few of the common birds of the
coast occur here, and two species of mammals have been found—a mouse (Peromyscus guardia)

and a wood rat (Neotoma insularis).

A few tinajas or natural water tanks are reported to exist on this island but no living water.

Guadalupe Island.

Guadalupe i is one of the largest and most soiehls islands lying off the coast:.of Lower Cali-
fornia. It is situated 135 miles southwesterly from Point San Antonio, south of the mouth of
Rosario River. It is roughly wedge-shaped and is nearly 22 miles long by 3 to 6 miles across,
the broadest part lying to the north. The island forms a high unbroken volcanic ridge, highest
in the northern part, where it reaches an altitude of over 4,500 feet above the sea. It is entirely
surrounded by very deep water, which, 4 miles from the northern end, reaches .a depth of 1,400
fathoms, and about the same distance from the southern extremity a depth of over 1,100
fathoms. Between the island and the mainland to the east is a great valleylike depression
of the sea bottom having a depth of over 2,000 fathoms., This shows that the summit of Gua-
dalupe is the crest of a gigantic mountain ridge rising over 16,500 feet above its base level on
the sea bottom with only a little more than one-quarter of its height appearing above the
surface of thesea. On the west the ridge of the island slopes rapidly toward the sea, ending in
the high cliffs. On this slope two great canyons lead down to the sea, separated by a barren hill
of reddish rock. On the northern part of the island the ridge is very sharp and descends pre-
cipitously on the west, but slopes less rapidly on the east. There are several small elevated
plateaus on this end of the island. The southern part is lower with more gradual descent
and ends less: abruptly. The surface of the island is covered with rough fragments of volcanic
rock from the size of pebbles to huge, ragged, outcropping bowlders.. The shore line is formed of
perpendicular cliffs, which are extremely rough and precipitous, with many volcanic caves
and holes. which were formerly the favorable. resorts of the Guadalupe fur seal.

There. are several springs on the northern plateaus but apparently no fresh water on the
southern half of the island. Wood and water may be found at a small cove near the north-

eastern end.
94 . MEMOIRS NATIONAL ACADEMY OF SCIENCES. [vor XVI

Guadalupe has the same wet and dry seasons as the California; coast. Owing to its situa-
tion in the midst of cool ocean currents and the prevalence of ‘the northwesterly winds, it, is
much cooler than any part of, the coast of Lower California at the same altitudes. The winters
are especially cool, owing to the cold sea, winds and, prevalence of fogs; the, summers are much
dryer and are comparatively very hot.. Dense fogs are more abundant in winter but prevail
also about higher parts, of the island throughout most of the year. Dr. Edward, Palmer reports
that in December, 1874, ice formed an inch thick on the middle of the island accompanied by
2 inches of snow and hail: and followed by five, days of cold rain! |,

During the winter of 1885-86 W. E. Bryant spent several months on the northern plateaus
and experienced many cold rains and some frosts.. During May and June, 1906, W. W. Brown
explored the island and noticed that nearly every time there was a gale ‘from the. northwest
it was accompanied by. heavy: fogs. He found the temperature so low that it was necessary
to keep a roaring fire in camp, especially at night. For 10 days at a time the sun was hidden
and everything was so saturated with moisture that it was difficult to preserve specimens.
During the gales, when collecting in the open was impossible, Brown found that inside the
cypress forest on the high tableland hardly a breath of air would be stirring so perfect was the
protection of the heavy growth of these trees.

' At the time of his visit Brown found that the entire plateau of the southern half of the
island was a sunburned waste with hardly a leaf of living verdure, but the higher and northern
parts were more fertile and less dependent on rainfall, owing to the constant heavy fogs envelop-
ing it. He discovered that the pines growing on the northern ridge of the island gave fine
illustrations of the ability of trees to condense moisture from the air. These trees are on the
summit of ‘a precipice nearly 3,000 feet high facing the sea’on the western side. When a
strong wind brought in a fog, the ground under the pines streamed with water, which ran down
the tree trunks, while the open ground some distance away was scarcely dampened. He
adds that long ago, when the entire northwestern part of the island was covered with dense
forest, the condensation must have been considerable, thus producing numerous springs ‘and
greatly benefiting the vegetation. Watson states that the low southern end of the island is
warm and sunny, | while the high northern end is covered by cold fog?

Vegetation of Guadalupe —The flora of Guadalupe is more closely related to that of Cali-
fornia than to that of Lower California. More than 145 species of plants are known from the
island, of which at least 29 are peculiar to it. The lower southern end of the island is much
warmer, drier, and more barren than the northern and entirely lacks trees. Watson (op. cit.)
states that vegetation in the warm eastern canyons and on the southern end is two months
ahead of the cold north and west parts. The higher northern end has scattered tracts of pines,
cypresses, oaks, and palms. »

Unfortunately, the presence on Guadalupe of thousands of goats, which, 3 increase almost
unchecked, has proved disastrous to the vegetation. Brown estimated that there were from
6,000 to 8,000 goats at the time of his visit in 1906, and'states that the few plants that have
escaped them are growing on perpendicular basaltic cliffs or are old trees with bark too hard
and woody to eat. He did not find any young trees growing in the cypress, palm, and pine
woods. The goats and house mice, which have also been introduced by man, eat the seeds as
they fall, thus preventing the trees from reproducing themselves, and if present conditions
continue the trees must become extinct. He noted that kids only a few months old were able
to eat the hard palm seeds. ‘Two species of trees are already extinct.

At the northern end of the island Brown found the following conditions: A fast-decaying
pine forest was growing along the narrow crest of the mountain at an altitude of about 3,000
feet and in some places down its almost perpendicular western slope. The ground was strewn
with dead trees, goat trails winding among them in all directions. On the highest part of
the island, north of Mount Augusta, a large cypress forest covers an area of about 3 square
miles on the western slope, ending abruptly on the east at the crest of the ridge. Hundreds of
ore trees ‘show the marks of goats’ teeth and many were Shee or dying. No fying] junipers

1 Proc. Amer. Acad. Arts and Sciences, XI, pp. “112-121, 1875 [1876]. ‘1 Tdem, pp. 105-112,

1p ge

 
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. | 95

were seen, but numerous dead stumps showed where a grove once existed. Far down on the
northwestern slope is a large grove of cabbage palms (Erythea edulis), and a small grove of the
same exists near Steamer Point on the west shore. In the large palm grove mentioned above
are some fine oak trees 30 to 65 feet high, and east of the cabins on the plateau under a cliff
several oaks were growing, but they were stunted and low branching. In May and June vege-
tation consisted generally of wild oats and foxtail grass, but with a profusion of a species of
malva about old corrals. In the main cypress forest, on the high tableland, were numerous
dead stumps of some species of shrub probably now extinct on the island.

The original vegetation of Guadalupe has held its own only along the faces of bluffs and
precipices which are inaccessible to the goats.

Birds of Guadalupe.—In 1875 the late Dr. Edward Palmer visited Guadalupe and was the
first to make known its interesting fauna and flora. From his collection the birds peculiar
to the island were first described. Several other American naturalists have since visited the
island and studied its bird life, the more notable of these being W. E. Bryant, in 1884 and
1885-86; A. W. Anthony, several times between 1892-1897; W. W. Brown, in the spring of 1906;
and several expeditions from the California Academy of Sciences at different times.

Forty-two species and subspecies of birds are known from Guadalupe, of which eight are
endemic. Since the partial destruction of vegetation by goats and the increasing great abun-
dance of domestic cats which were introduced there by goat herders, two species, Pipilo conso-
brinus and Thryomanes brevicaudus, have disappeared. The Guadalupe caracara (Polyborus
lutosus) has also become extinct. The last-named species was probably exterminated by poison
put out by goat herders owing to their depredations among the newly born kids. Mr Brown
found wings and feathers of juncos, rock wrens, and Guadalupe flickers, showing where these
birds had been killed by cats. He.adds that the mortality among the petrels and shearwaters
from this source must be great on the northern ridge where they breed, for he saw hundreds of
of their wings strewn about.

Mammals of Guadalupe.—There are no indigenous land rauehanals on Guadalupe. Domestic
goats, cats, and house mice have been introduced and have become so destructive to the plant and
bird life that the original fauna and flora appear to be doomed to almost complete destruction
within a limited period. Mr. C. H. Townsend states that many of the goats perish each year for
lack of water during the dry season, and as the vegetation decreases the available wenee: supply
will undoubtedly decrease with it.

The shores of this island were once favorite resorts for great numbers of Guadalupe fur seal
(Arctocephalus townsendi), which is now supposed to be almost if not quite extinct. A small
‘herd of sea elephants, the last representatives of the once enormous numbers of these animals,
still remains on the western side of theisland. (See Pl. 25, Fig.2). Sea otters were once abundant

‘ about these shores, but they are now gone forever. Sea lions and harbor seals still frequent its
shores in small numbers.

Life zones of the island.—We lack tne necessary detailed observations on the vertical dis-
tribution of the plant and animal life of this island to properly define its life zones. For this
reason they can be located only in the most general way. The basal slopes of Guadalupe and
practically all of the eastern and southern sides below the higher parts of the summit are Upper
Sonoran. The summit as well as most of the northern and northwestern slopes are within the
Transition zone. Oolaptes cafer rufipileus, Junco insularis, Thryomanes brevicaudus, and Sitta
canadensis appear to be Transition zone species here. The main trees characterizing this zone
on Guadalupe are Pinus radiata and Oupressus guadalupensis.

CLIMATE.

The climate of Lower California as a whole is extremely hot and dry. Much brilliantly
sunny weather with scanty and uncertain rainfall over all of the peninsula results in the preva-
lence of marked desert conditions. There are, however, considerable local climatic variations
according to the situation in relation to latitude, altitude, mountain ranges, and to the east
and west coasts. The contrast is most marked between the excessive summer heat on the low

158961°—21——7
96 _ MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

desert near the head of the Gulf of California and the cool temperatures on the highlands of the
San Pedro Martir Mountains. The greatest annual range of temperature is on the desert near
the delta of the Colorado and amounts to more than 100° F.

In winter sharp frosts occur on these deserts and. from 3 to 4 feet of snow, with severe
freezing temperatures on tops of the highest mountains. The daily extremes of temperature
on the deserts, especially on the east side, are often considerable, the days being extremely hot
and the nights comparatively cool. Immediately along the coast on both shores the daily
variations are much less than they are farther inland. Throughout the peninsula the health-
fulness of the climate is unexcelled. Sofar as health and comfort are concerned, I fully agree
with Dr. Eisen in his statement that the ‘“Cape Region of Baja California is one of the very few
places where the various conditions of temperature, moisture, and other climatic conditions
are almost perfect.”

Much the larger part of the peninsula lies within climatic conditions which characterize the
Lower Sonoran life zone, changing to, Arid Tropical conditions in the southern third, and to
Upper Sonoran. and Transition on the mountains. The climatic conditions of the northern
parts are similar to those which prevail in the immediately adjoining sections of southern
California.

Winds.

Winds have a strong influence on climatic conditions in some parts of the peninsula, espe-
cially on the western side. Northwesterly winds blow almost continuously along the entire
length of the west coast from March to October, and during this period fine weather with but
little rain prevails. Owing to the cool northerly ocean current which bathes all the west coast,
the northwest winds coming across the sea have their temperature reduced and, while everywhere
more or less moist, are here frequently laden with fog which sweeps inland and at times spreads
over the entire west slope to the sides of the mountainous backbone of the interior. From No-
vember to February the wind varies on the west coast from southeast to southwest, with occa-
sional southeast gales accompanied by heavy rains. In the cape district these heavy winter
rains, which are described below, are called ‘‘equipatos.”” In December and January heavy
northers, lasting from one to three days, sometimes occur, mainly in the cape district and on the
west. coast. ;

The winds of tee eastern or gulf shore are much more unsettled and irregular than on the
west shore. In spring and summer northwest winds are common, some southeasterly winds
occur, and light breezes and calms are prevalent. In winter, northwesterly gales occur in the
gulf. At the end of the tropical rainy season on the mainland, in autumn, fierce southerly gales
of. several days’ duration occur in the southern and middle parts of the gulf, and sometimes
increase to destructive hurricanes which sweep the gulf and the lowlands of both its coasts.
They are known as ‘‘E] Cardonazo,”’ and occur at intervals of several years, being accompanied
by thunder, lightning, and torrential rains. An early record of one of these tempests is given
by Venegas, who tells us that ‘‘in the autumn of 1717 a terrible hurricane arose and extended
all over Lower California and the gulf, accompanied by violent rains, which swept everything
before it.. The church and Padre Ugarte’ s house (near Mulege) were leveled to the ground and
he saved his life by taking refuge under a rock, where he was exposed to the severity of the
weather for 24 hours. The channel for carrying water was choked and the sluice at San Xavier
carried away. The cultivated land, both here and at Mulege, was completely ruined. and cov-
ered with stones.”! In September, 1896, one of these hurricanes, ee by floods of
rain, did great damage in the lowlands of Sinaloa.

On the night of October 4, 1817, a terrific hurricane with heavy rain Gattsed a Aeod at San
Jose del Cabo, destroying many houses and drowning 45 persons. On September 17, 1918, a
characteristic autumnal hurricane visited the northwest coast of Mexico and the southern gulf
coast of Lower California. Two United States Shipping Board vessels on their first voyage
were reported sunk off the southern end of the peninsula, and other ships were badly battered.

 

.. ! Noticia de la California, II, p. 310, 1757.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 97

The town of La Paz was partly wrecked, several people were killed, and the equipment of the
United States Naval Station at Pichilinque, in La Paz Bay, was seriously injured. The fall
and winter hurricanes of the gulf coast carry with them such floods of rain that they are difficult
to classify and some of them are described under the account of the rainfall. These fall and
winter storms render navigation of the gulf at these seasons rather hazardous, although in the
main the weather is pleasant.

Rain and Snow.

Owing to its great length and to its geographical relations to the continental mainland,
Lower California has two definite rainy seasons, both of which at times include the entire penin-
sula. The first of these is part of the tropical summer rainy season which regularly occurs on
the west coast of Mexico as far north as northern Sinaloa, the border of which extends across
the gulf to Lower California and is specially marked in the southern part of the cape district.

The second rainy season of the peninsula is a part of the winter rains of the north Pacific
coast, which extend southward from California and at times even reach the vicinity of Cape
San Lucas. As the peninsula occupies the borderland between the main areas covered by both
these rainy seasons, it is strongly affected by any departure from normal conditions in either
of them, and a scanty rainy season in either direction usually means an entire absence of rain
for the corresponding period on the peninsula.

No year passes without some rainfall in some part of Lower California and rain falls practi-
cally every year on the high mountains of the northern and southern ends. Rain is usual every
year also in some parts of the lowlands, but at times unbroken droughts continue over the low
areas of a large part of the peninsula for periods of from two to more than five consecutive
years. During such long droughts the smaller herbage disappears, the cactuses and desert
shrubs show little vegetative activity, and most of the cattle and the native mammals on the
ranges in the dry areas die. Long periods of excessive drought were recorded in the chronicles
of the early Spanish missionaries, who told of the suffering thus entailed on themselves as well
as on the Indian population. They are the main cause of the present scanty population of the
peninsula and are responsible for the many abandoned ranches and unsuccessful attempts at
colonization throughout its extent. These periods of excessive droughts are followed by years
of torrential rains which flood the country and temporarily change the dry and sandy water-
courses into booming streams.

The areas of least rainfall in the peninsula are the desert northeastern section, Vizcaino
Desert, and the lowlands surrounding La Paz Bay.! The areas of greatest precipitation are the
San Pedro Martir Mountains in the north and the Laguna Mountains in the south. Precipita-
tion is greatest along the highest mountain slopes and diminishes along the basal slopes to the
general average of the surrounding lowlands.

But few accurate records of the annual precipitation are available for any part of Lower
California. Conditions are so similar in the lowlands of the northern end of the peninsula and
adjacent parts of southern California that it is evident the precipitation must be practically the
game for a considerable distance on both sides of the border. The rainfall of San Diego averages
about 9.4 inches yearly, which is approximately that of the coast of the peninsula as far down
as Ensenada, beyond which it rapidly decreases. The rainfall at Salton, on the southern border
of the Colorado Desert, averages about 2.5 inches and probably is about the same on the lowlands
of the peninsula south to beyond San Felipe Bay.

Conditions unlike those in the north and the lack of accurate observations render estimates
of the rainfall in the middle and southern parts of the peninsula of little value, except to give a
general indication of the relative humidity. Dr. Eisen has estimated that the average annual
rainfall at La Paz i is less than 2 inches, at Miraflores about 20 inches, at San Jose del Cabo less

 

1 In the discussion, a few years ago, of the supposed change that might occur on the Colorado Desert from the humidity due to evaporation from
the newly flooded Salton Basin, those who advanced the theory of probable climatic change from this source should have considered conditions 100
miles farther south, where the humidity due to evaporation from the vastly greater area of the Gulf of California has no appreciable effect in modify-
ing the extreme desert conditions of the bordering country either in Sonora on the east or in the area between the gulf and the east base of the lofty
San Pedro Martir Mountains in Lower California.
98 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor. XVI

than 12 inches, along the coast near Cape San Lucas not over 5 inches, and-in the Sierra de la
Laguna about 30 inches.

The summer rainy season of Lower California, as already stated, is most marked in the cape

district. It begins there somewhat later than on the opposite Mexican mainland coast and ends
earlier, usually lasting from July until October or November, September being the wettest
month. Most of our information concerning these rains in this district is from Dr. Hisen.? He
found that the earliest summer rains fall on the divide at the north end of San Jose Valley, just
north of Miraflores. Rains begin at Miraflores long before they do at San Jose del Cabo, and
sometimes fall even at Santa Anita several, weeks before they reach San Jose, although the
latter is only 9 miles farther down the valley. The rainfall along the mountains bordering the
gulf shore southeast of La Paz is less than among the mountains farther inland. The rains are
more abundant and regular i in the Sierra Victoria, the mountainous area between La Paz and the
cape, than elsewhere in the southern half of the peninsula, and next to this are most copious on
the high mountains immediately bordering the gulf coast for some distance north of La Paz Bay.
_ . During the summer rains thunder may be heard and rain seen to fall daily for weeks in the
mountains of thé cape district without a drop falling on the surrounding plains. Eisen states
also that the rainfall is heaviest west of the crest of the Sierra Victoria, though the steeper slope
there causes it to run off more quickly and so its influence on the vegetation. is less evident, than
on the more gently sloping east side. He has so well stated the character and effect of thesum-
mer rains on the lowlands of southern Lower California that his account is worth quoting. He
says: ‘‘They are most capricious as regards time and locality, and while one place may be
deluged another close by may never receive a drop. This is frequently seen in passing through
the country. In some places I found flowers and grass in abundance, and animal life was
represented by thousands of snails crawling on the bushes. A few yards or a mile away every-
thing was barren, dry, and dead, no rain having fallen there for years. When the water comes
down it generally comes in torrents, washing everything before it. This torrential rain, which
may amount in a single shower to 3 or 4 inches, has thus rarely the opportunity to sink deop i in the
soil. On the contrary, it carries everything before it, and leaves its trace behind in the formation
of the precipitous walls of the canyons and in great washouts.” * In October, 1905, Goldman
found the vicinity of the Tinaja Santa Clara on the Vizcaino Desert very dry and no sign of rain
that year, but a few miles to the eastward near the Cerro de Tecolote sufficient rain had fallen
so that desert trees and bushes were in full foliage and the ground was covered with green arise
and flowering plants.

The tropical summer rains commonly extend north along the elevated interior to the siiiddls
of the peninsula and less frequently to the adjacent coasts. They are more and more irregular up
the west coast as far as'San Quintin and along the east coast to the extreme northern border
about the Cocopah Mountains. These summer rains are always more common along the moun-
tainous backbone of the peninsula than on the coasts, and are of more frequent occurrence on
the high summits of the San Pedro Martir Mountains than elsewhere in the north. Occasionally
the outskirts of the area covered by the summer tropical rains extend as far north as the southern
end of the Sierra Nevada in California, where they are manifested by sporadic thunderstorms.

In 1905 we saw the first manifestations of the approach of these rains in the north. Up to
June 11 the sky had been absolutely cloudless, but on that date cumulus clouds formed over the
summits of the San Pedro Martir Mountains and then spread northward over the Sierra Juarez
and disappeared late in the afternoon. This was repeated the following day ahd the natives
told us that these clouds indicated the approach. of the summer rains, although it was several
weeks later before rain actually fell.

The day we reached Vallecitos, on top of the San Pedro Martir Mountains, July 12, 1915,
there were several brisk thunder showers accompanied by hail. Mr. E. W. Funcke writes that
from July 2 to 6, 1906, heavy thunderstorms prevailed in these mountains and from then until
the end of August rains were plentiful throughout this section. He adds that summer rains

 

1 Calif, Acad. Sci., V, p. 742, 1895.
3 Proc. Calif. Acad. Sci., V, pp. 741-745, 1895.
3 Bull. Amer. Geog, Soc., XXXII, No. 5, pp. 405-406, 1900.
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 99

occur nearly every year on the Cocopah or other desert mountains in this section and thence
southward along the gulf side. These rains are commonly in the form of cloudbursts, usually
in July, August, and September. They may occur any year but are irregular and may not ap-
pear for several successive years. Such cloudbursts occur at times on the desert plains as well
as in the mountains between the Sierra Juarez, the San Pedro Martir Mountains, and the gulf.
Mr. Funcke reports a heavy cloudburst at Calexico, just at the northern border of the peninsula,
on August 16, 1909, when nearly 2 inches of rain fell in two hours. The summer of 1910 rains
began on the desert south of Calexico, July 13 and 14, and were followed later in the season by
heavy cloudbursts.

- On August 21, 1884, rain fell for more than 48 hours at San Ignacio, and water ran for
several days in the arroyo. On June 30, 1911, a general rain occurred in the district about
Calmalli, after which unbroken dry weather prevailed until October 4, when a heavy storm
covered the middle region of the peninsula from about Calmalli to beyond San Ignacio.
Shortly before 11 p.m. on the 3d of October a light rain fell at San Ignacio, which continued,
accompanied by a strong gale from the north during the early hours of the 4th. At the same
time an extremely heavy rain fell in the Santa Lucia and other mountains to the east and
on the table-land 20 to 30 miles to the east and north, continuing until about noon on the
4th. Most of the water drained into the San Ignacio Arroyo, down which it ran with great
force, striking San Ignacio about ll a.m. The front of this flood was reported to be from 12 to
15 feet high. Twenty-three persons were reported drowned, 39 buildings destroyed, and many
others badly injured. All the gardens and small fields in the valley were washed away, includ-
ing thousands of date palm trees. At Calmalli it only rained about an hour on the morning of
the 4th, accompanied by strong north winds. This rain extended about 35 miles east of the
last-named place and a little to the north. Considerable rain fell along the gulf coast east of
Calmalli, and a heavy gale raged at the same time over the gulf, destroying much property at
Santa Rosalia, on the west side of the gulf, and at Guaymas and Mazatlan, on the east side,
thus indicating its extended character. Neither this storm nor the one of June 30 reached as
far west as the Pacific coast. On August 25, 1912, a light shower fell at Calmalli, the first rain
since the preceding March. :

Besides the summer rains, as already mentioned, the peninsula also has a winter rainy
season of the same origin as the winter rains of California and coincident with them. These
winter rains are less regular than in California, but at times extend southward to include the
entire length of the peninsula, though more frequent and abundant in the northwestern parts,
especially in the Sierra Juarez and San Pedro Martir Mountains and thence to the Pacific coast.
On the desert east of these mountains the winter rains are reported to be very irregular and
commonly to last only a few hours except when they appear as the torrents accompanying the
widely spread storms which sometimes occur during this season. Some years no winter rains
fall. I have been.informed that during the winter of 1908-9 in the northeastern part of the
peninsula there were but four or five rainy days, while in the season of 1909-10 no rain fell. The
winter precipitation on the summits of the Sierra Juarez and San Pedro Martir Mountains is
mainly in the form of snow which some seasons, according to A. W. Anthony, falls to a depth of
6 feet or more on the higher parts and lies on the north slopes of the summits till late in May.

Snow falls every winter from the northern border south along the Sierra Juarez and San
Pedro Martir Mountains to Rosarito Ranch, Matomi, and on San Juan de Dios Mountain, but
rarely or never farther south than the latter point. On October 12, 1908, the top of the San
Pedro Martir Mountains was covered with snow and it falls on these mountains between Octo-
ber and March, the 20th of March being about the latest date. The summits of these moun-
tains are never covered with snow during the entire winter but usually only for periods of
from two to four weeks. In midwinter snow falls far below the lower border of the pine timber.
Anthony writes that it falls down to about 2,500 feet above sea level on the west base of the
mountains and it sometimes reaches as far as Valladares. November 17, 1902, Edmund Heller
was driven out of the Sierra Juarez by a snow storm. Cattlemen who occupy the range on
these mountains in summer usually drive cattle up to the higher levels early in April and down
100 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von, XVI

to the lowlands on the west side early in October. Mr. Funcke writes that he has never seen
snow on the Cocopahs nor on any of the desert mountains between thé San Pedro-Martir Range
and the gulf. Loo ; .

Although winter rains are most regular in the northern half,of Lower California, yet they
sometimes extend the entire length of the peninsula and fall in great abundance in the cape
district. They are most abundant throughout the peninsula on the western slope. In the cape
district they often accompany gales from the southeast and are called “equipatos.’”’ The
people there say they are sometimes entirely lacking for several years in succession, but often
occur in the form of prolonged storms, sometimes extending over a period of from 12 to 15, days.
At such times the entire country becomes so boggy that all overland travel is suspended. We
experienced these conditions after the rain storm of December 10 and 11, 1905, which flooded
the Llano de Yrais, opposite Magdalena Bay, where we traveled for nearly two days through
broad shallow ponds and many swiftly flowing streams in the washes. Another day of rain
would have made the country impassable. It was a curious sight to see miles of giant cactuses
and other characteristic desert plants standing in the midst of broad half-submerged expanses.
The winter of 1904-5 was one of extremely heavy rains throughout the peninsula. One storm
continued for about two weeks throughout the cape district and rendered travel temporarily
impossible. The winter of 1905-6 several rains occurred in this region in November and De-
cember and again in January and February.. Although sharp frosts are common on the Sierra
de la Laguna in winter, yet the residents of that region report that snow never falls on these
mountains.

The following fragmentary records for rainfall by inches in different localities give an indi-
cation as to the precipitation in the lowlands of certain parts of the peninsula:

Records of rainfall, by inches, in certain parts of Lower Californta.

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Year. Jan. Feb. | Mar. | Apr. | May. | June. | July. | Aug. | Sept Oct. | Nov. Dec.
Santa Rosalia:
1898 x Xx Xx x x x Xx x x x 0.53
1905... 1.44 | 1.40} 0.31 0 Xx 0 0.51 | 0.20} 0.14] X . 80
1906... -bL| T. x 0 0 0.18) .04) xX x 0 xX
1907 0 T. 0 0 0 xX |8.18] X x xX x
1908 x x. Xx x x x -08| xX x x x
1909 waa aie x x x 0 x | 126! X | 1.42 0 x x:
BOLO. cs caccwiaceaite se orgs x x 0 x x x x x x x x x
WB wo teesacegeeetye cee edee ne x x x x x 0 x x x x x x
14 vss cccccncase preter T. ae x x 0 x x x x x x x
La Paz:
1906 wis hice sewctensewaus $oeess x x x x x x 0 1.97] .24] 1.02 0 x
zs adie x 0 0 0 0 0 .87 | 4.96) X | 3.82] xX x
-15] X 0 0 |.xX x T. -99| X. x x
x |x] x 0 1o072| X {L0O1) T 0 0 2. 07
0 0 0 .| 0 0 3.06 | 2.45] xX .05 |] 0 0
x . 05 0 0 141) .54) .59|/4.00] xX x x
x x x x x Xx x x x 0 0
Indicates no record. T Indicates a trace.

Ensenada.—1888, season’s rainfall (from fall of 1887) to March 31, 12.06; 1894 to 1911,
seasonal average, 9.74; 1919, season’s rainfall to March 14, 8.10.

Ojos Negros (ranch).—1909-1911, seasonal average, 9.16.

San Quintin.—1908-1911, seasonal average, 5.24.

Fog and Dew.

Fogs in Lower California are limited mainly to the west slope within the influence of the
northwest winds. They occur at all seasons, but are more prevalent during the summer months.
As in southern California, they generally set in at night and clear late in the morning, the rest of
the day being bright and pleasant. Fogs are most common from the northern border to San Quin-
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 101

tin and are only a little less common thence south to Cedros Island. From Cedros Island to Cape
San Lucas fogs are much less frequent than to the northward and they clear away earlier in the
morning. They are rather common in fall and winter at Magdalena Bay and on the adjacent
coast. All along the west coast the fogs sweep in over the lowlands and near Ensenada up to
over 4,000 feet altitude and in the middle of the peninsula to from 2,000 to 3,000 feet. June 1,

1905, San Rafael Valley, east of Ensenada, was hidden in fog which extended up on the west
base of the Sierra Juarez. Fogs were reported to be common here at this season. At daybreak,
September 18, 1905, a heavy fog covered all the coastal plains and low mountains up to the tops
of the granite mountains back of our camp at Yubay, in the middle of the peninsula. The ist
of November, 1905, at La Purisima a heavy rain was followed by a dense fog which covered the
coast ‘plain and interior mesas up to the high interior mountains. Valleys shut of from the
coast by a low mountain, or ridge, like that at San Fernando, even though draining into the
Pacific, are usually cut off from the fogs and most of the dews so common in places open to sea
winds. The middle of October, 1905, fog drifted down from the north each night over the Viz-
caino Desert.

Fogs are much more prevalent about Cedros Island than on the mainland and still more so
at Guadalupe Island. The abundant moisture provided by the fogs has great influence in
producing the luxuriant desert vegetation on parts of the west slope. The amount of mois-
ture which may be contributed from fogs to vegetation is well Ulustrated by the observations
of W. W. Brown, the 1st of May, 1909, on Guadalupe Island.1_ Fogs were very prevalent there,
and in the morning the ground about the bases of the pines growing on the high slopes was
streaming with water which had been condensed in the tree tops and ran down the trunks to
the ground.

Copious dews are aiso common on the west slope of the peninsula and in the morning the
vegetation is often dripping with moisture and the surface of the ground dampened as from
a light shower. The blazing sun quickly dries up this surface moisture but not before the
desert plants have had opportunity to appropriate some of it. A.W. Anthony informs me that
during the years he lived in the region west and south of the San Pedro Martir Mountains heavy
dews were prevalent, especially in valleys near the coast, which made the nights chilly through-
out the year. The prevalence of coast fogs along the lowlands bordering the west shore is
shown by the abundance of orchilla and other epiphytic plants on the desert shrubs and small
trees from near Magdalena Bay northward. Between La Paz and Cape San Lucas fogs some-
times hang about valleys and mountain slopes during, or at the end of, long-continued storms,
but are rarely seen at other times.

Fogs and dews are of very rare occurrence on the east slope of the peninsula, where the
high interior mountain ranges cut off the moist northwest winds which supply them on the
west side. The water of the gulf is much warmer than that of the Pacific, and the gulf slope
of the peninsula is much hotter and drier than the west side.’ We did not see a single fog or dew
on the east slope during 1905-6.

Temperatures.

But little exact information is available concerning the temperatures in any part of Lower
California. Freezing temperatures are known to occur in winter throughout the peninsula
except in an extremely narrow belt immediately bordering the sea in the cape district and
extending northward on the west coast to San Jorge and on the east coast to beyond Santa
Rosalia. Frosts are unusual anywhere in the immediate vicinity of theshore. Within a narrow
coastal belt, rarely over 3 miles wide, from the northern border to opposite Cedros Island the tem-
peratures are much like those at San Diego, Calif., which range from 32° F. in winter to 101° F.
insummer. From Cedros Island southward to beyond Magdalena Bay the coast belt is warmer
in summer than to the north of that island, but still not so warm as the interior. This entire
coast belt owes its coolness to the northwest winds and has the most agreeable summer climate
of any part of the peninsula lowlands. Inland on the west slope the summer temperatures are

4Condor X, p. 102, 1908.
102 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

much higher than near the Pacific coast and not uncommonly go well above 100° F. At La
Paz winter temperatures are said to range from 50° to 74° and summer temperatures from 80°
to 94°.

The hottest part of the peninsula is the northeastern section, between the Lower Colorado
River and head of the gulf on the east and the Sierra del Pinal and San Pedro Martir Mountains
on the west. This area forms the southern end of the Colorado Desert and shares its extreme
summer temperatures. Observations for several years at Salton, on the Colorado Desert not
far north of the border, show extremes ranging from 18° in winter to 128° F. in summer, and
a closely similar range undoubtedly occurs throughout the lowlands of the adjacent area south
of the border.

The northern half of the east coast of the peninsula lies for months under the fierce heat
of an unclouded sun, and the surface waters of the gulf become so warm that they help main-
tain the high temperatures, thus making this throughout the summer one of the hottest and
most desert parts of the continent.

A considerable contrast exists between the temperatures both in summer and winter
at opposite points on the two coasts. The difference is most marked near the northern
border and becomes much less in the southern half of the peninsula. Southward the tempera-
tures of the east coast become less excessive in summer until at San Jose del Cabo, according
to Dr. Eisen, the maximum temperature rarely exceeds 90° F. On the west coast the summer
temperatures increase southward but are always less than at opposite points on the east shore.

Arid tropical or semitropical conditions prevail along the coast lowlands of the east side
from Cape San Lucas north to near the middle of the peninsula and on the west side north at least
to San Jorge beyond Magdalena Bay and to San Ignacio Valley in the interior. Summer tem-
peratures, as usual, decrease with increased altitude on the mountains and are cool and pleasant
on ‘the summits of the higher ranges such as the Sierra Juarez and San Pedro Martir Ranges in
the north. and the Sierra de la Laguna of the Cape District.

The winter temperatures on the Sierra Juarez and San Pedro Martir Ranges are much lower
than elsewhere in the peninsula, especially‘on the high uplands of the San Pedro Martir Moun-
tains, where a heavy cap of snow remains for some time. Occasionally throughout the summer
light frosts occur in these northern mountains. Thus on June 4, 1905, we had a light frost
at our camp near El Rayo Ranch, at 4,600 feet altitude, on the west side of the Sierra Juarez.
Frosts were common in July at the same season in the high meadows on the San Pedro Martir
Mountains. Sharp frosts occur in winter on the tops of the mountains all along the backbone
of the peninsula to the Sierra de la Laguna and other southern mountains. Temperatures
become so low at night, between December and March, on the Sierra de la Laguna south of
La Paz that ice an inch thick is sometimes formed.

In the lowlands of the northern half of the peninsula frosts occur with considerable regu-
larity and near the northeastern border and on elevated plains and valleys are often severe,
the temperature sometimes falling to 18° or less on the Colorado Desert. Mr. A. W. Anthony
informs me that killing frosts were rare at San Fernando, though the nights were usually cold
there during the winter months. In the lowlands of the southern half of the peninsula frosts
are more sporadic. At Comondu the people told us that frosts do not visit their valley every
year but sometimes are of sufficient severity to kill the leaves of growing sugar cane. Light
frosts were also reported to occur on the Llano de Yrais, a short distance inland from Magdalena
Bay. They also occurred in the bottom of the valley a few miles above San Jose del Cabo while
we were at Santa Anita, the middle of January, 1906. Dr. Eisen states that frosts occur in
January and February at Miraflores and Caduano, sometimes killing back the vegetation,
including the Palo de Arco (Tecoma stans).*

The temperatures on the coastal islands are closely like those on the adjacent mainland
coast, but islands well offshore, like Cedros and Guadalupe, average much cooler, especially in
summer. Sharp frosts occur in winter on the higher parts of Guadalupe.

 

1 Proc. Calif. Acad.Scl., V. p. 738, 1895.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 103
PLANT LIFE.

A report by E. A. Goldman on the more important plants collected during our explorations
in Lower California furnishes detailed information concerning numerous species.' In addition
notes and lists showing the zonal distribution of the species are given in the accompanying account
of the life zones of the peninsula (pp. 117-132.)

The following notes contain general information concerning the characteristic features of
the flora. ‘The photographs throughout this report convey some idea of the region but do scant
justice to the extraordinary appearance of landscapes overgrown for hundreds of miles with
strange and grotesque vegetation. In many places these bizarre combinations are so unfamiliar
in appearance that they might represent the flora of some former geological period.

The flora of Lower California appears to have been derived from three main sources, as
follows:

1. From the mountain and foothill slopes of southern California, and extending thence down
the peninsular backbone over the Juarez and San Pedro Martir Mountains and covering most
of the area between these mountains and the Pacific coast.

2. From the deserts of northwestern Sonora, southern Arizona, and southeastern California.
The plants derived from this last-named: region form the dominant and most characteristic part
of the general flora and cover approximately two-thirds of the total land area of the peninsula,
extending its entire length and in some parts its entire breadth.

3. From the arid tropical lowlands, foothills, and low mountains of Sinaloa on the mainland
coast of Mexico opposite the southern end of the peninsula. This part of the flora is mainly
limited to the southern third of Lower California and is especially prominent in the cape district
south of La Paz Bay. Many species range farther north, however, especially along the gulf
coast, one copal reaching even to the delta of the Colorado. This tropical Mexican mainland
element of the flora throughout its extent is intermingled with equally or more prominent Tep-
resentatives of the desert flora which comes in from the north. The presence in the southern
end of Lower California of so considerable a number of arid tropical genera and species of plants
common on the opposite mainland coast of Mexico appears to indicate a closer connection be-
tween the two areas than is shown by the reptile, bird, and mammal life.

Plants Derived from the Mountains and Foothills of Southern California.

The higher slopes of the Juarez and San Pedro Martir Mountains (above 4,500 feet), in
the northern part of the peninsula, are capped with the only coniferous forest in Lower California,
with the exception of an insignificant area of pinyons on the higher parts of the Victoria Moun-
tains in the cape district and small areas of coniferous trees on the northern ends of both Cedros
and Guadalupe Islands. The pine forest on the Juarez and San Pedro Martir Mountains forms a
narrow belt for about 150 miles along the top of the range and contains the only merchantable
timber in the peninsula. These pines form an open forest variable in character. It is often
much scattered and the greater part of it is composed of pinyon or other scrubby trees of little
or ho commercial value.

It begins on the top of the Sierra Juarez near the California border and extends to within a
few miles of San Matias Pass. It is absent on all the slopes of the pass below about 4,000 feet,
but begins again immediately south of the pass, near San Matias Spring, and extends continu-
ously thence south the entire length of the San Pedro Martir Mountains to their southern end
at Matomi Canyon.

Jeffrey yellow pine (Pinus ponderosa jeffreyi) is the most important of the trees and forms
the main part of the forest above about 4,800 feet on the Sierra Juarez and above 6,000 feet on
the San Pedro Martir Mountains. (See Pl. 26, Fig. 1.) In the Sierra Juarez the Coulter pine
(Pinus coulteri) is intermingled with the Jeffrey pine and a smaller number of incense cedars also
occur there. Among the Jeffrey pines on the San Pedro Martir Mountains there is a propor-
tionately small number of sugar pines (Pinus lambertiana), white firs (Abies concolor) from about
7,500 feet to the summit, and of the incense cedar (Libocedrus decurrens) from about 6,500 feet

 

| Plant records of an expedition to Lower California, Contr. U. 8. Nat. Herbarium, vol. 16, pt. 14, pp. I-VIII, 309-371, IX-XIIT, 1916.
104 MEMOIRS NATIONAL ACADEMY OF SCIENCES. ‘ [Vou. XVI

along streams up nearly to the summits of the main ridges. A few cypresses (Cupressus sp.)
occur on northerly slopes at from 8,500 to 9,000 feet.

The lodgepole pine (Pinus chntorie is the second most important, forest tree on the San
Pedro Martir Mountains. It i is freely intermingled with the Jeffrey pine from 7,500 to 8,000
feet, and on northerly slopes above the latter altitude to the summit of poe mountain it occurs
in many small almost unbroken areas.

On the lower slopes the Parry pinyon (Pinus quadrifolia) constitutes the lower border of
the forest. It occurs as a. scattered growth near San Matias Spring and along the Rosarito
divide to the tops of the cliffs at Matomi. It also forms a thin forest along the northwest base
of the San Pedro Martir Mountains, as at Pinyon, and thence out over the foothills to the Alamo
Plain and on both slopes of the Sierra Juarez below the yellow pines. The single-leaved pinyon
(Pinus monophylla) i is reported to occur below 4,500 feet on the desert eastern slope of this range.
The common pinyon (Pinus edulis) is also reported to occur in these mountains and on the
basal slopes of the San Pedro Martir range.

Extending over the foothills of the west slope of these mountains are many junipers
(Juniperus californica). This species is particularly abundant on and about Alamo Plain and
along the northwestern foothills of the San Pedro Martir Mountains. In addition to the conif-
erous trees, the quaking aspens (Populus tremuloides) and an oak (Quercus chrysolepis) occur
in the higher parts of the San Pedro Martir, and the Palmer oak (Quercus palmeri) on the Juarez
Mountains. Along the western slopes of these montains the woody vegetation is scattered and
includes four species of Ceanothus, several species of Arctostaphylos, an elder (Sambucus), the
chamiso (Adenostoma fasciculatum), the mountain mahogany (Cercocarpus), and the California
holly (Heteromeles).

The western slope is much more abundantly clothed with vegetation than the eastern
side, where the bare granite slopes facing the desert are excessively arid and have but scanty
vegetation. In sheltered places, along the south walls of cliffs on the east side at the head
of Canyon Diablo, we saw scrubby pinyon pines growing at an altitude of 9,000 feet, thus show-
ing the effect of the desert heat in carrying Upper Sonoran Zone plants far up through the
Transition Zone.

From the basal foothills on the western slope of these mountains a belt of partly brush-
grown valleys and ridges extends westward to the coastal range. There and on sheltered
slopes occur scattered thickets made up of various species of shrubs and medium-sized live
oaks, the latter mainly in canyons draining to the west. Some cottonwoods and sycamores
grow in these canyons from near Ensenada southward to opposite San Quintin. In addition
to the cottonwoods and sycamores some of the characteristic trees and shrubs occurring in the
outlying western foothills of the mountains and along the Coast Range near Ensenada are the
California live oak (Quercus agrifolia) and scrub oak (Q. dumosa), several species of Eriogonum,
Ephedra californica, a buckeye (Aesculus), an ash (Frazinus attenuaia), and two species of
black sage (Artemisia). The east slopes of the Juarez and San Pedro Martir Mountains, as
already noted, are extremely rocky, steep, and barren, and the forest commonly ends at a
higher altitude than on the west side. Fan palms are the only tree growth in the canyons
leading down through the foothills on the eastern side and even these are lacking in many
canyons.

rom the southern end of the San Pedro Martir Mountains no forests of timber exist until
the Sierra de la Laguna, at the northern end of the Sierra Victoria, in the cape district, is
reached, a distance of more than 500 miles. There the summit of the Sierra Victoria for about
15 miles along its northern end and mainly on the east slope is capped by a scrubby forest of
live oaks, picturesque madrofios, and various scattered shrubs. Over a few square miles of
the higher parts of the extreme northern end of the range, known as the Sierra de la Laguna,
an abundant growth of scrubby pinyon pines is mixed with the oaks. The forest of pinyons
and oaks on these mountains has its lower limit close to the summit on the abrupt west slope
and is above 4,000 feet on the eastern side. Bare grassy slopes are numerous in this forested
Memoirs National Academy of Sclences XVI. First Memoir, Pl. 26.

 

 

 

 

 

Fic. 1.—Forest of yellow pine (Pinus ponderosa jeffreyi)and sugar pine (Pinus lam-
bertiana) near Vallecitos, San Pedro Martir Mountains. The sugar pine in the
middle background shows the characteristic branching of this species.

 

 

 

 

 

Fig, 2,—Vallecitos, 8,000 feet altitude, San Pedro Martir Mountains. The forest is mainly lodgepole pine
with some yellow pine,
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 27.

 

 

 

 

 

Desert vegetation at Arroyo Catavifia on granite tableland south of Llano de los Buenos Aires. The picturesque plant life of
this locality gives the landscape an appearance suggesting that of some remote geological time.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 105

area, mainly on the south exposures. Below the oak forest are open hillsides with a few oaks
of another species (Quercus brandegei) which follow the canyons and range out to the bottom
of the San Jose Valley at Miraflores. In the deep foothill canyons of the Sierra Victoria are
many extremely tall, slender-trunked fan palms and a tall, handsome cottonwood of a local
species (Populus monticola), which has white bark almost like that of an aspen. (See Pl. 19,
Fig. 2.) The lower half of the mountain slopes are occupied by a scattered mixture of cactuses,
agaves, and other species belonging to the desert lowland flora, mixed with numerous Arid
Tropical species having their nearest relatives on the shores of Sinaloa across the gulf. The
pinyon pine on the Victoria Mountains has been identified as typical Pinus cembroides, the form
common on the mountains of the Mexican mainland. All three of the oaks we found on these
mountains and the adjacent lowland to the east proved to be undescribed species, apparently
related to those of the Mexican mainland and not to those of California. Growing on the
higher parts of the Sierra de la Laguna, Belding’s nolina was one of the most characteristic
species, and a willow, a hackberry, a Thalictrum, the California holly, a Garrya, and a Ribes
were other characteristic shrubby plants there. None of the forest growth on the Victoria
Mountains is of sufficient size to be of more than local use for mining timbers or some similar
purpose and is extremely inaccessible for any practical use.

The tops of the higher mountains seen by us during our journey from the southern end of
the San Pedro Martir Range to the Sierra Victoria, with a single exception, were devoid of any
sign of oak or pine trees. The one exception was on a high mountain lying southeast from the
head of Calamahue Canyon, where we saw a few scattered scrubby trees near the summit on the
north slope, which might have been either oaks or pinyon pine, but were probably the former.
They were too far away to be positively determined.

The early Spanish missionaries are reported to have cut trees for timbers used in building
the missions at Calamahue and San Borja on the tops of these high mountains near the head
of Calamahue Canyon. In the fall of 1719 Padre Ugarte found some thick-trunked trees,
which were called ‘‘Guarivos,”’ in the canyon of the Santa Lucia Range about 30 leagues north-
west from Mulege. The logs were hauled from there to the coast, where a small vessel was
built which was used to explore the coast of the gulf.1_ These trees were probably the tropical
palo blanco (Lysiloma candida) which is plentiful in some of the canyons on the west slope of
the great lava plateau.

Other trees derived from the mainland mountains of southern California grow on Cedros
and Guadalupe Islands off the west coast of the peninsula. The upper northwestern slopes and
crests of the mountains on the northwestern part of Cedros Island are exposed to the cool,
fog-laden northwest winds of the Pacific and are clothed with California junipers and scrubby
pine trees (Pinus muricata), the latter sometimes reaching a height of over 60 feet. This
forest is of no considerable extent and not of commercial importance. On Guadalupe Island
the entire northwestern summit was formerly covered with a forest of Monterey pines (Pinus
radiata), which is now reduced to a rapidly decaying strip along the northwestern crest and
down the adjacent western slope. On the highest part of the Island, north of Mount Augusta,
a large cypress forest (Cupressus guadalupensis) exists, which W. W. Brown estimated in 1906
to cover about 38 square miles. Well down the northwestern slope is a large fan-palm grove
(Erythea edulis) containing some fine oak trees (Quercus tomentella).

' Early botanical exploration (1874 and later) proved the existence on the island of 145
species of plants, including numerous shrubs, but since then goats have become so abundant
that they have greatly reduced the vegetation. Some species have already been exterminated
and others are going.

Plants Derived Mainly from the Deserts of Northwestern Sonora and the Southwestern United States.
As detailed in the preceding account of its climate, Lower California as a whole has

extremely scanty rainfall, and periods of several successive years occur during which no precipi-
tation takes place over great areas. Under the severe desert conditions thus produced the

 

1 Venegas, Noticia de la Calif., II, p. 317, 1757.
106 MEMOIRS NATIONAL ACADEMY OF SCIENCES. | [Vou. XVI

peninsula has developed the richest. and most varied desert flora known in America and prob-
ably in the world. This flora covers at least two-thirds of the total land area of the peninsula,
including most of the lowlands and nearly all of the lower mountains and mountain ranges.
The deserts extend from the northern boundary at the delta of the Colorado to the southern
extreme at Cape San Lucas and from coast to coast. They also extend from sea level up to
from 3,000 to 5,000 feet, or even higher in some places.

In areas where for various reasons the soil conditions are unfavorable, the desert is a bare
sun-baked waste or has a scanty and stunted vegetation. These extreme conditions are
especially prevalent on the slopes of the desert mountains. In other places where a deep
sandy loam affords a favorable medium for retaining water after the infrequent rains, vegeta-
tion is extraordinarily abundant and veritable forests of desert plants from 10 to 50 feet high
are found, often becoming almost impenetrable thorny jungles. (See Pl. 28, Fig. 1.) This
vigorous desert growth occurs in places along the eastern base of the San Pedro Martir Mountains,
but is especially notable along the great interior plateau extending from southeast of San
Fernando to the northern end of the Victoria Mountains south of La Paz. The plants of this
great region exhibit an amazing and fantastic variety of forms grouped in endless combinations,
thus giving the landscapes an extraordinary aspect. The cactuses are the most generally
distributed and varied family of desert plants in this region, presenting a marvelous variety,
from species creeping over the ground like huge spiny caterpillars several yards long (Lemair-
eocereus eruca), to giant arborescent species with trunks two or three feet in diameter and 50 or
60 feet high (Pachycereus pringlet). (See Pls.8 and 14.) One slender species lies prone on the
ground, another climbs and spreads like a vine among low tree tops, while still others have
single corpulent barrel-like trunks 6 or 8 feet high. All are armored with defensive spines,
some short and closely set, others like slender needles several inches long, and still others strong
and rigid with recurving hooks at the tip and so closely set as to form a powerful defensive armor
covering the entire plant.

The cactuses of the peninsula belong to 11 genera, including Bergerocactus, Echinocactus,
Echinocereus, Lemaireocereus, Lophocereus, Mamillaria, Myrtillocactus, Opuntia, Pachycereus,
Pereskiopsis, and Wilcoxia. Other striking and characteristic associated plants are numerous
species of yuccas, agaves, nolinas, ocotillos (Fouquieria), the palo verdes (Cercidiwm and Parkin-
sonia), ironwood (Olneya), creosote bush (Covillea), mesquite, acacias, and several species of fan
palms of the genera Washingtonia and Erythea. Both genera of fan palms occur from near the
northern border to the southern end of the peninsula. According to Orcutt they occur in the
basal canyons on the east slope of the Sierra Juarez and along the courses of many canyons
and gulches where water is near the surface, in the desert mountains and plateaus of the eastern.
and middle parts of the peninsula. South of Todos Santos they appear also on the Pacific
coast. In addition to the canyons along the east base of the Sierra Juarez fan palms are
abundant in Matomi Canyon, in the canyon where the ruins of the Santa Maria Mission are
located, along the arroyos of Jaraguay and Catavifia, and in numerous other places. In the
extreme south they are numerous on the coastal plain at Pescadero and on the west side of the
San Jose Valley as well as in the basal canyons of the Sierra de la Victoria.

In addition are several desert species which seem like plant monstrosities. This is pecul-
iarly the case with the elephant wood, which has a short thick trunk and contorted branches
covered with a smooth, skinlike bark which lies in thickened folds and creases, giving the tree
adropsical appearance. (See Pl. 28, Fig.2.) Contrasting with this is the cirio (dria columnaris),
which is a tapering, polelike tree growing from 20 to over 50 feet high, with short, slender, and
thorny branchlets forming a bristling armament along its trunk. This strange tree has tufts of
pale yellow flowers rising on long slender stems from the tip of the trunk, which is covered with
a smooth greenish-yellow bark. The tapering yellowish trunk and rootletlike branches give the
cirio a close resemblance to a gigantic slender parsnip bottom side up. (See Pl. 29, Fig. 1.)

The southern end of the Colorado Desert occupies northeastern Lower California, and
includes the delta of the Colorado River. Many of the islands in the delta and some of the adjoin-
ing delta plain are thickly overgrown by Macdougal’s cottonwood (Populus macdougalii). Great
No. 1} LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 107

numbers of these are washed away during spring floods when the banks where they grow are
cut away by the river, as illustrated in a photograph taken in April, 1905. (See Pl. 23, Fig. 1.)
These trees occur at Ojo de Liebre, in the Vizcaino Desert, and at occasional ranches to the
southern end of the peninsula. The delta islands and adjoining delta plain are also overgrown
with willows and the arrowwood (Pluchea sericea). Away from the river the lower and more
barren parts of the desert have a scanty growth, including the creosote bush, an ephedra (EZ.
trifurca), the screwpod and common mesquite (Prosopis pubescens and P. juliflora), a dalea
(Parosela spinosa), a copal (Elaphrium macdougalii), the ironwood (Olneya tesota), several
species of cactus, and others.

The infrequent rains of this part of the desert often come in the form of tremendous showers
which'send many temporary torrents across the sloping plains, thus making a series of washes
leading to a main drainage channel which at such times may become an impassable torrent.
At other times these drainage ways are dry washes, along which, owing to the greater amount
of moisture they receive, grow long lines of creosote bushes. mesquites, and other shrubs which

‘mark their courses.

In addition to the common cottonwood, the palo verdes (Cercidum torreyanum and Parkin-
sonia aculeata), the common mesquite (P. juliflora), the ironwood, the creosote bush, the copal
(Elaphrium microphyllum), a Simmondsia, several species of Lycium, Baccharis, and the brittle
bush (Encelia farinosa), besides others, occur in the desert throughout the length of the penin-
sula. No giant cactuses occur about the delta of the Colorado but they grow abundantly along
the east base of the San Pedro Martir Mountains, and one or the other of the three species of
giant cactuses known from the peninsula occurs continuously thence to Cape San Lucas.

Large numbers of the woody plants of the desert grow to the size of low trees'from 12
to 30 feet in height and in connection with the cactuses, yuccas, agaves, and cirios in many
places form extensive forests. These forests are largest along the middle of the elevated
interior plateau on the western slope from San Fernando to opposite Magdalena Bay, and
again about the base of the Victoria Mountains in the cape district. In the middle of the
peninsula from near San Fernando to beyond Calmalli the polelike cirio gives a remarkably
distinctive character to the forests, which consist, in addition to the species named above,
of numerous cactuses, agaves, acacias, mesquites, huisaches (Vachellia farnesiana), cassias,
euphorbias, jatrophas, fouquierias, the elephant wood, grease wood (Atriplex), and numerous
others. Over considerable areas near Santo Domingo forests occur composed almost entirely of
Yucca valida, with a mixture of only a few other prominent plants. (See Pl. 29, Fig. 2.) Other
areas in this section contain forests made up almost wholly of species of agave.

In canyons, from near the northern border on the east base of the Juarez Mountains south
to the cape district, fan palms of the genera Washingtonia and Erythea occur and add a most
picturesque element to the scenery. (See Pl. 17, Fig..2, and Pl. 27.) The large number of date
palms in the oases from the middle of the peninsula south and the coconut palms in the cape
district are beautiful introduced additions to the flora. (See Pl. 13, Fig. 2.)

The eastern and western sides of the peninsula vary in the amount of their vegetation.
The east coast is much hotter and drier than the west, which is shown strikingly throughout
the peninsula north of La Paz by the abrupt and marked change in the character of the vege-
tation from the abundant growth on the western slope to the generally sparse and often
extremely desert character of that on the east side of the main mountain ranges forming the
backbone of the peninsula. This is due mainly to the effect of the barrier formed by the
mountain ranges which extend the entire length of Lower California and thus cut off the cool,
moist, and often foggy northwestern trade winds. This aridity of the east coast is in part
due also to the proximity of the east slope of the mountains to the Gulf of California, which
is so inclosed by land, especially about its northern two-thirds, that in summer the surface
water becomes heated to a temperature far above that of the Pacific on the opposite shore
and thus helps hold the general temperature of the air at a higher point.

In the lowland areas, as well as elsewhere on the west side of the peninsula, vegetation
is decidedly benefited by the condensation upon it of moisture contained in the sea fogs which
108 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

prevail and are drifted inland at certain seasons. The observations of W. W. Brown, in early
May, 1906, on Guadalupe Island well illustrates this. The pines on Guadalupe grow along a
precipice nearly 3,000 feet high on its western side, and when the prevailing northwest wind
blows in the fog from the surrounding ocean he noticed that the ground at some distance from
the pines looked scarcely dampened, while small streams of water which had condensed in the
tree tops were running down the tree trunks and away from their bases.

In addition the cool breezes of the Pacific have a distinct effect on the vegetation of the
small coastal islands and on a narrow belt from one to several miles in breadth along the
western shore of the peninsula. This becomes gradually less toward the cape district, but is
strongly marked throughout most of the distance. The sea breezes appear to prevent the
growth of trees and shrubs, so that the hill slopes and plains close to the sea and the small islands
are clothed with grasses and other small vegetation, including a few stunted woody plants,
some of which are peculiar to these locations. The slopes of some of the islands along the
northwest coast are covered with great beds of the curious ice plant (Mesembryanthemum crys-
tallinum), and San Martin Island is characterized by the abundant beautiful velvety white
cotyledon Dudleya anthonyi.

Tn sheltered gulches and canyons of the coast hills and stameting, chin they escape the
direct effect of the sea winds, and on sheltered sides of the islands small shrubs grow abun-
dantly. The cool, moist sea winds along the northwest coast cause an abundant growth of small
lichens on the stones, and on the trunks and branches of shrubs and other objects along the
open belt immediately bordering the shore of the mainland and of the outlying islands.
Farther inland, where the heavy growth of desert shrubs and. other vegetation forms a jungle
in many places, these cool winds and fogs favor the growth of large numbers of tillandsias and
various species of epiphytic plants. Most important of these is the orchilla (Roccella), a species
of lichen which is valuable for dyeing purposes, This plant hangs from the branches of cactuses,
low trees, and shrubs, like Spanish moss, which it greatly resembles in general appearance and
color. It grows mainly along the coastal plain from Ballenas Bay to Magdalena Bay, and was
formerly the basis of a profitable industry. The accompanying photograph illustrates this
semiparasitic erowih on the desert vegetation along the seacoast near San Andres. (See Pl. 30,
Fig. 1.)

The vegetation on the islands on both coasts is made up mainly of the same species occur-
ring on the nearest adjacent part of the mainland of the peninsula with the exception of
Guadalupe. On this island the vegetation appears most closely related to that, of the. coast
of middle California, as illustrated by the appearance of a juniper, a cypress, and the Monterey
pine.

Plants Derived from the Lowlands and Mountains of Sinaloa and Southern Sonora on the Mainland of Mexico.

As shown by the lists in the account of the life zones of the peninsula, large numbers of
plants in the southern end of Lower California belong to the arid tropical genera common in
Sinaloa, on the Mexican mainland. Many of these are among the large and conspicuous species
characterizing the lowlands, hills, and mountain slopes of the cape district up to over 4,000
feet altitude. Some of the species appear to be common to both sides of the gulf, but most
of them are undoubtedly peculiar to the peninsula, Among the numerous Arid Tropical genera
of the cape district, which also characterize the flora of the mainland across the Gulf of Cali-
fornia, may be mantionad Ficus, Lysiloma, Mimosa, Cassia, Erythrina, Albizzia, Jatropha,
Heematoxyjlon, Xanthoxylon, Lantana, Elaphrium, Manihot, Linea, Bourreria, and Ciococon:

Many of the tropical species of the southern end of Lower California are included in the
account of the Arid Tropical life zone of the peninsula, and to save repetition reference is made
to that part of this work (pp. 122-123). In addition to the tropical species of the southern end
of the peninsula, the trees of the higher mountains in that region appear also to havé been
derived from the Mexican mainland rather than from the north. This is indicated by the
fact that the pinyon pine in the small forest on the Sierra de la Laguna has been identified by
Mr. Shaw as typical Pinus cembroides. The three species of Quercus appear to be related to

fe sk & & re * 9 ,
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 28.

 

 

 

 

 

Fig. 1.—Jungle of cactuses, yuccas, and other thorny plants in northern part of Vizcaino Desert, west of Calmalli. Growth of this
character in many places is impenetrable for man or horse without cutting the way.

 

 

 

 

 

Fia. 2.—Elephant wood trees ( Veatchia discolor) near San Andres. These dropsical trees are among the most extraordinary plants of
these deserts.
Memoirs National Academy of Sciences XVI. First Memoir, FI. 29.

 

 

 

 

 

Tic. 1.—Forest of cirio (Idria columnaris) near San Fernando. This remarkable tree is peculiar to Lower California and is the
prevailing species for hundreds of miles on the granite tableland.

 

 

 

 

Fig. 2.—Forest of Yucca valida near Santo Domingo Landing. This extraordinary growth extends irregularly for miles.
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 30.

 

 

 

 

 

Fig. 1.—Orchilla (Roccella) growing on fouquieria bush near San Andres. This dye plant, of the color and general appearance of
the Spanish moss growing on trees of our Gulf States, was formerly exported by shiploads to Europe from Magdalena Bay.
The production of analine dyes killed this industry.

 

 

 

 

 

Fia, 2._Epiphytic plants growing on desert shrubs in foggy coast belt near San Andres.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 109

oaks on the mainland across the gulf rather than those of the north, and a large madronyo
(Arbutus peninsularis) is not represented on the San Pedro Martir Mountains in the north
but has a close relative on the mainland mountains across the gulf. About the heads of
sheltered bays and lagoons from the boat landing at San Jorge above Magdalena Bay southward
around Cape San Lucas and up the east coast nearly to Santa Rosalia the mangrove (Rhizo-
phora mangle) is common, The most extensive growth of this well-known tropical plant occurs
in the lagoons extending northward from the head of Magdalena Bay. Elsewhere it grows
in more or less limited areas on Margarita Island, at Cape San Lucas, at the head of La Paz
Bay, and in small areas northward along the coast, as well as on several of the islands in the
southern part of the gulf.

Two species of wild cotton have been found native in the southern end of the peninsula,
one in the cape district north of La Paz (Gossypium davidsoni) and the other on Margarita Island
(G. harknessi).. The ordinary domesticated cotton (G. herbaceum) was formerly cultivated from
Cape San Lucas north to the mission of San Fernando. .

Like the deciduous trees of the low Arid Tropical forests on the west coast of Mexico many
of the low trees of the cape district, especially among the leguminous forms, have a tendency

to flattened, broadly spreading tops. In some of the valleys and low canyons of the cape
district the Washingtonia fan palm is common. The other genus of fan palm of this district
(Erythea) is represented in canyons of the Victoria Mountains by groups of excessively tall,
slender-trunked individuals growing often 100 or more feet high and only 6 to 12 inches in
diameter. (See Pl. 19, Fig. 2.) The wild fig (Ficus palmert) occurs as scattered individuals in
situations where water ag close to the surface, from near the middle of the peninsula south to the
cape district, where it is more numerous than elsewhere. It ranges up to nearly 4,000 feet in
the foothills. ©”

The Arid Tropical element probably does not include more than 50 per cent of the entire
vegetation of the lowlands of the cape district, since it is intermingled with nearly an equal
number of desert species coming down from the north. These representatives of the northern
deserts include two species of the giant cactus and many other members of the same family,
with the creosote bush, fouquierias, mesquites, acacias, palo verdes, ironwood, the brittle bush,
and numerous others. The southern end of the great Magdalena Plain, including its extension
to the head of La Paz Bay and the mountain group called the Cacachilas, are rather poorly
clothed with vegetation, the heavy growth thus being limited to the middle and northern part
of this plain.

The northern foothill slopes of the Sierra Victoria, along the wagon road from La Paz to
Todos Santos, are covered with a heavy forest growth of giant and other cactuses, mesquites,
acacias, and many species of arid tropical trees and bushes from 10 to 30 feet high, forming in
many places a dense thorny jungle. The same vigorous growth exists along the sloping sandy
valley leading down from the southern end of the Victoria Mountains to the bay at Cape San
Lucas. In the San Jose Valley, especially along the middle of its course and on the foothill
slopes along its borders between Santa Anita and Miraflores, this growth is especially abundant.
On the basal foothill slopes back of Miraflores the arid tropical element is unusually well marked
by the numberof tropical bushes and small trees.

On the part of the Magdalena Plain lying north of Matancita occur some areas of extremely
large and abundant desert growth, including vigorous mesquite trees, which indicates the
presence of water at a comparatively small. depth below the surface of the rich soil. We also
noted a heavy growth of large mesquite trees in El Potrero Valley a few miles above Mulege.

The flowering season of plants 1 in Lower California is largely governed by the rainfall, and
many of the small herbaceous species spring up quickly and flower after a heavy rainfall at any
time during the warm season. Many of the desert plants, however, have developed the power of
storing sufficient moisture and food during the occasional rains to enable them to flower and
fruit successfully, even during seasons when no rain falls. Cactusés, agaves, yuccas, nolinas,
mesquites, palo verdes, and numerous others need rain only at wide intervals to enable them to
flower and fruit-more or less regularly in spring or summer. During their period of blossoming
110 MEMOIRS NATIONAL ACADEMY OF SCIENCES. | [Vou. XVI

the mesquites and palo verdes are masses of small but attractive flowers which collect swarms
of insects. The handsome flowers of the cactus vary from white to red and yellow, and the
wonderful masses of white blossoms of the yuccas are among the beautiful sights of the desert.
The flowering season in the high mountains of the north is delayed owing to the much colder
winters, but it opens early in the Victoria Mountains of the cape district, where on the Ist of
February Belding noted a blackberry, a black currant (Ribes brandegei), the madrofio (Arbutus
peninsularis), a monkey flower (Mimulus), the painted cup (Castilleja), , a lupine (Lupinus),
and other plants in blossom. Tisen writes that the cape district is frequently ablaze with the
yellow flowers of the palo de arco (Tecoma stans) and a large morning glory (Ipomea aurea).

As noted elsewhere, a number of species of trees of economic value to the natives have been
brought to the cape district from the Mexican mainland among which the coconut and date
palm, the gourd tree (Crescentia cujete), the tamarind (Tamarindus indica), and Pithecolobium
dulce are among the most notable. The latter and the tamarind have in a number of places
escaped from cultivation and now form part of the forest growth.

ANIMAL LIFE.

Although Lower California is mainly a desert its long coast line, varied surface, and great
extent in latitude combine to give it a varied and interesting fauna. The peninsula extends
into the border of the Tropics, but its vertebrate land fauna is almost wholly derived from the
adjacent arid region to the north and contaims representatives of nearly all the genera and
species of the southwestern desert region of the United States. The considerable variety of the
vertebrate fauna of the peninsula appears from the fact that it is known to contain approxi-
mately the following: Five genera and 5 species of amphibians, 33 genera and 75 species and
subspecies of reptiles, 244 genera and 428 species and subspecies of birds, and 51 genera and
166 species and subspecies of mammals. The most notable characteristics of the fauna of
Lower California are the preponderance of desert land species of reptiles, birds; and mammals,
and the number and variety of marine mammals.

The pronghorn antelope, mule deer, mountain sheep, and mountain lion are the big game
animals. Of these the antelope, which once occupied suitable areas practically throughout the
peninsula, is gone from most of its former range and is on the way to early extinction. The
‘mountain sheep is still widely distributed on the main mountains of the eastern half of the penin-
sula, but in addition to being hunted for sport is being slaughtered without regard to age or sex
to supply mining camps and small towns with meat, and must soon follow the antelope unless
effectively protected. The deer are more generally ‘distributed and will hold their own much
longer than the antelope and sheep, but they also have become much scarcer in many places
and are practically gone in others.

The only record I have found of bears in Lower California is the reference in Pattie’s narra-
tive to the presence of grizzlies near Santa Catarina Mission in the Sierra Juarez.1. Mountain
lions, wild cats, coyotes, badgers, and gray foxes occur throughout the peninsula, while the small
desert fox ranges south to the Llano de Yrais, opposite Magdalena Bay. There is no trust-
worthy record of wolves anywhere in Lower California. The jaguar is said to have occurred
at a very early date in the delta of the Colorado. Beavers were formerly extremely
numerous in the delta, but owing to trapping are now present in much smaller numbers. The
plains and valleys in many places throughout the peninsula swarm with desert rodents, includ-
ing jack rabbits, cottontails, kangaroo rats (Dipodomys and _Perodipus), pocket mice (Perog-
nathus) of several species, various species of white-footed mice (Peromyscus), and wood rats
(Neotoma). The antelope squirrel (Ammospermophilus) is also common south of Cape San
Lucas. :

The falling fruit of the date palm groves in the middle of the peninsula, as at San Ignacio,
Mulege, and La Purisima, attract many gray foxes, bassarisks, and little striped skunks, which
make their homes in the neighboring lava cliffs.

 

1 Thwaites, Early Western Travels, XVIII, p. 221, 1905.
No. 1]. LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 111

. When Lower California was discovered its shores swarmed with whales, elephant seals, fur
seals, and sea otters, and small game abounded. Several species of whales had their breeding
grounds in sheltered bays on both coasts and these mammals were so numerous that the early
Spanish navigators named for them Ballenas Channel in the Gulf of California and Ballenas
Bay on the Pacific coast.. During the first two-thirds of the last century the abundance of whales
about these shores attracted many whaling ships, mainly under the American flag, until the
hosts of California gray and sperm whales were either killed or driven away and whaling became
unprofitable and was entirely abandoned. During the last few years whaling on the Pacific
coast has revived with the use of steam-powered ships and may have an even more disastrous
effect on the whales than the former industry.

.., Capt. Scammon states that the great abundance of whales in the bays of Lower California
became generally known to whalers during the winters of 1846 and 1848, and in the last-named
winter 52 whaling ships anchored in Magdalena Bay.? Since the whaling activity in the period
which immediately followed the whales have never recovered anything approaching their former
abundance. During the same time the Guadalupe fur seal, northern elephant seal, and sea
otter were pursued by many hunting ships and slaughtered so ruthlessly that this fur seal is now
supposed to be extinct and the elephant seal and sea otter are on the verge of extinction, if not
already gone.

Scammon Lagoon, at the fide of Vizcaino Bay, was first explored by Capt. Scammon in
1858. He found its waters alive with whales, porpoises, and fish. Turtles and seals basked on
the shores of low islands studding the lagoon, and small game was so abundant that acres of sand
bars bared by the tide were closely packed with wild geese, ducks, snipe, and other waterfowl.’
Capt. Scammon in his ‘‘Marine Mammals” has given us the only connected account of the
wonderful abundance of animal life at one time existing on the northwest coast of Lower Cali-
fornia.

-In one of the early accounts Padre Taraval is quoted concerning the abundance of sea
otters on Cedros Island at the time of his visit in 1737, and describes the ease with which they
were killed by the Indians.‘ Under Spanish rule sea otter hunting and trading in their domain
by foreign ships was prohivited, but many thousands of skins were traded for by smuggling
and hunting vessels along the Pacific coast early in the last century. One ship alone secured
1,700 sea otter skins from the natives during a single visit to Ensenada in 1809.5

The variations in the annual rainfall on the peninsula have a direct influence on its animal
life, which is strongly shown in the case of some of the mammals. During series of consecutive
heavy rains the desert vegetation flourishes exceedingly, great quantities of seeds are ripened,
tender vegetation abounds, and desert mammals, especially the small kinds, such as pocket mice
and kangaroo rats, increase enormously until favorable areas swarm with them. When the long
rainless periods follow, the herbage ceases to grow, no seeds are ripened, and there is a resulting
scarcity of food, and an enormous mortality among the small inhabitants of the desert. The
same cause also has its effect on the abundance of rabbits, deer, and antelope. Other less obvious
causes also bring about great local temporary changes in the numbers of small mammals. This
was ‘shown strikingly during the early part of our journey, when we visited localities where
Heller found certain pocket mice abundant .in the summer of 1902, and we found them
extremely rare or entirely absent during the summer of 1905, although the same or closely allied
species were plentiful in other areas of the same region.

Lower California exhibits some striking contrasts under which its animal life exists. In
the cape district'deer and many birds winter on the Sierra de la Laguna, where snow never falls,
while Anthony tells me that on the San Pedro Martir Mountains, in the north, snow at times falls
toa depth of 6 feet and the deer and quail are forced down to the basal slopes on the west side.

 

oe Several of the early visitors to the shores of the peninsula record. the abundance of small game, as in the case of the British exploring ship,
Herald, which called at San Quintin Bay, Oct. 15, 1846, and Seemann, the naturalist of the expedition, mentions the abundance of “rabbits, bares.
kam curlew, sanderling, mackerel, smelt, and cray fish. ”
‘a Overland Monthly, IV, pp. 230-238, March, 1870.
8 Scammon, Supp. App., p. 123, in Browne’s Resources af the Pacific ‘Slope, 1887.
«Venegas, Noticia de la Calif., vol. 2, p. 438, 1757.
‘Suppl. App., p. 42, in Browne’s Resources of the Pacific Slope, 1887.

158961°—-21——-8
112 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

On the islands about the coasts of the peninsula numerous species of sea birds have their
nesting colonies. On the Pacific coast islands such species as murrelets, western gulls, petrels,
shear waters, cormorants, California brown pelicans, and man-of-war birds breed in large num-
bers. On the islands in the Gulf of California are breeding colonies of murrelets, Heermann’s
gulls, elegant terns, red-billed tropic birds, Brewster’s and blue-footed boobys, cormorants,
California brown pelicans, and man-of-war birds.

The resident birds of the plains and foothills of the peninsula are mainly species or sub-
species closely related to well-known forms of the cactus deserts of the southwestern United
States and northwestern Sonora, Among the most notable of these are the elf owl (Micropallas
whitneyi), the road runner (Geococcyx), several subspecies of the ladder-backed woodpecker
(Dryobates scalaris) and of the Gila woodpecker (Centurus uropygialis), the gilded flicker (Colaptes
chrysoides), and two forms of the ash-throated flycatcher ( Myiarchus cinerascens) ; also the desert
sparrows (Amphispiza), the cardinals (Richmondena and Pyrrhulozia), the phainopepla, the mock-
ing bird, several species of desert thrashers (Toxostoma), the cactus wren (Heleodytes), the rock
wren (Salpinctes), and the verdin (Auriparus). A considerable infusion of the Californian fauna
coming in from the northwest, includes the valley quail and subspecies of the California jay which
are common south to Cape San Lucas. In the higher mountains there is a fair representation
of genera from the mountains of southern California. In summer enormous numbers of shear-
waters and petrels visit the northwest coast, and in winter vast numbers of waterfowl resort to
the delta of the Colorado, and to San Quintin and other bays on the west coast.

The general decrease of migratory waterfowl in North America is evidenced by their decrease
along this coast in winter, while some of the resident marine waterfowl breeding on the coastal
islands, as gulls, pelicans, and cormorants, have been reduced by unrestricted persecution in
the breeding season by guano hunters.

During our journey through the middle of Lower California in the autumn of 1905 the
scarcity of migrant birds was notable. At the southern end of the peninsula, the warm pleasant
winter climate and plentiful vegetation appeared well adapted to render it an attractive winter
home for species breeding in the north. We found such birds, however, in surprisingly small
numbers compared with those living at this season along the neighboring mainland coast of
Mexico. The small number of such birds as Dendroica auduboni and Vermivora celata lutescens
was difficult to understand. It was equally unexpected to find the lark bunting (Calamospiza
melanocorys) plentiful in flocks on the cactus-grown Magdalena Plain and about the base of the
Victoria Mountains. Most of the resident land birds of the peninsula occupy the same area
throughout the year, but the apparent absence of Crotophaga sulcirostris, Myiochanes richard-
sonit peninsulae, and some other summer residents of the cape district during the winter of 1906
appeared to indicate a migration of these species across the gulf to the Mexican mainland. From
the foregoing it appears that Lower California is well outside (west of) the regular bird migration
routes and has comparatively few stragglers from the great army of migrants which twice a
year sweeps over the Mexican mainland.

The most remarkable bird records ever made in Lower California are those of Lyman
Belding, who collected a specimen of Vetola lapponica baueri at La Paz, the winter of 1881-82;
one of Motacilla ocularis at La Paz, on January 9, 1882; and one of Anthus cervinus at San
Jose del Cabo, January 26, 1883—all birds of eastern Siberia and occasionally found in Alaska in
summer and in southern and eastern Asia in winter. The occurrence of these birds so far from
their natural haunts must have been due to some remarkable storm.

The waters of the gulf are warmer than those of the Pacific coast of Lower California. This
is shown by the molluscan fauna which, as stated by Dr. Cooper, is much more tropical in charac-
ter in the gulf than on the Pacific coast. In the northern part of the gulf, however,‘ the species
are largely of the temperate fauna, many of them identical with those for the same latitude on
the west coast of the peninsula. This seems to indicate that the dividing ridge [of the peninsula],
now more than 1,000 feet in altitude in the north, was crossed by one or more channels within
geologically recent times.t ’ The low passes now known to exist across the peninsula at the
head of Calamahue Canyon, and again northeast of Calmalli, probably not exceeding 1,200 or

1 Proc. Calif. Acad. Sci., 2d ser., V, p. 37, 1895.

 
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 113

1,800 feet in height, may have been the submerged routes through which northern sea shells
reached the head of the gulf. —

The southern part of the gulf coast once abounded in the pearl oyster, which supported a
profitable industry, but the beds have been so worked out that the industry has greatly declined.
The artificial propagation of the pearl oyster was begun on some of the gulf islands near La Paz,
in 1903 or 1904, and at the time of our visit in 1906 was reported to be succeeding. Some years
later these plantings are reported to have been successful and may. indicate a method of reviving
the pearl industry. Sea turtles breed in great numbers in the lagoons and bays along the shores
and some were shipped by the monthly steamers from Magdalena Bay to San Francisco in 1905.
Unrestricted exploitation of this resource will no doubt result in its destruction, as in the case of
the marine mammals. Fish of many species are very numerous on both coasts of the Peninsula.

When the first summer rains come in parts of Lower California, both animal and plant life
awakens suddenly into vigorous activity. The contrast is most impressive in the southern
third of the peninsula—the cape district—where the climate is mainly arid tropical. There
many of the birds are said to delay the beginning of their nesting period until the advent of, the
rains releases the swarms of insect life and renews vegetative activity. Dr. Eisen states that
after the first rains in the cape district ‘‘nine-tenths of all chrysalises and larvae hatch at once
and the whole country is teeming with animal life. The bushes and trees cover themselves with
leaves and flowers, giving ample food for caterpillars and insects of all kinds. As this [the rainy
season] begins in the sierra, the sierra is the proper place in which to begin explorations and
collections. A week after the first rain the fauna is at its height. Myriads of butterflies are
seen filling the air in the daytime, while during the evening hours microlepidoptera, or moths
and beetles of all kinds swarm around the lights. Countless lizards dart among the rocks on
the hillsides or in the branches of the trees. * * * I have seen so many butterflies filling
the air in one of the high valleys of the Sierra El Taste that the air seemed thick with them, and
this continued for several weeks. When at such a time a cloud over the sun caused a temporary
shadow the innumerable host of butterflies suddenly vanished, having taken refuge on the
underside of the leaves of trees, bushes and herbs.”” One day * * * ‘“‘the whole valley
between El Taste and Mt. Troyer was filled to a height of 3,000 feet, or from the bottom of the
valley to the top of the peaks, and several miles in width, with butterflies i in almost every color
of the rainbow, but principally white and yellow. * * * One of the features of the fauna
of the cape region is the immense quantity of land shells found in some places. In certain
evidently favored localities the ground is literally covered with the dead and white shells of
land mollusks. We ride along for hours through canyons where the ground is thus strewn.
Then as we turn into another canyon, we find no shells at all, not even after close search under
rocks and trunks of trees. * * * Another peculiarity as regards the land shells is that
every sierra and every canyon almost possesses peculiar forms not found anywhere else.” The
early missionaries tell of enormous flights of grasshoppers which visited the cape region, devour-
ing everything green and darkening the sky, but there is no modern record of such a visitation.

The following lists and remarks briefly ehaunonjnze the vertebrate land fauna of Lower
California and adjacent islands.

Amphibians.

_ As might be expected from the desert character of this region its amphibian fauna is
extremely limited and none of the species are of tropical origin. Van Denburgh listed only five
genera and five species;” to these must now be added two genera and three species, totaling
as follows:

Batrachoseps attenuatus. Bein couchti. Hyla regilla.
Plethodon croceater. Bufo boreas halophilus. Rana draytonit.
Aneides lugubris lugubris. Bufo punctatus.

None of these are peculiar to Lower California. The Scaphiopus ranges as far east as
Texas, while the Hyla and Batrachoseps range some distance north along the Pacific coast.
The known range of the Plethodon extends northward to the Sierra Nevadas of West Central
California.

 

1 Proc. Calif, Acad, Sel., 2d ser., V, pp. 760-762, 1895. 3 Idem, pp. 556-561.
114 MEMOIRS NATIONAL ACADEMY OF SCIENCES. : [Vou. XVI

Reptiles. Y

Like the remainder of the vertebrate land fauna, most of the reptiles evidently invaded
the peninsula from the arid land area of the adjoining part of the United States, with only a
few representatives from, the tropical mainland of Mexico. Aside from two species of sea
turtles, the tortoise shell turtle (Eretmochelys imbricata) and. the green turtle (Cheloma agassizi) ,
Van Denburgh lists 31 genera and 62 species and subspecies of reptiles for the peninsula and
adjacent islands,! to which must now be added 8 genera and 40 forms subsequently obtained
within its limits, thus making a total of 39 genera and 102 species and subspecies.

Rerrites KNown From Lower CALIFORNtA AND ADJACENT ISLANDS.

Phyllodactylus tuberculosus. Sator angustus. Lichanura trivirgata.
Phyllodactylus unctus. Sator grandzvus. Coluber aurigulus.

Coleonyx variegatus. Sceloporus graciosus vandenburgianus. | Coluber constrictor flaviventris.
Ctenosaura hemilopha. Sceloporus licki. Coluber lateralis.

Ctenosaura insulana. Sceloporus lineatulus. Salvadora hezalepis.
Dipsosaurus dorsalis dorsalis. Sceloporus occidentalis biseriatus. Phyllorhynchus decurtatus.
Dipsosaurus dorsalis lucasensis. Sceloporus orcutti. Elaphe rosaliae.

Crotaphytus collaris baileyi. | Sceloporus zosteromus. Pituophis catenifer annectens.
Crotaphytus copeit. Phrynosoma blainvillii. Pituophis catenifer deserticola.
Crotaphytus wislizenit. Phrynosoma cerroense. Pituophis vertebralis.
Sauromalus ater. Phrynosoma coronatum. Lampropeltis boylii boylii.
Sauromalus hispidus. Phrynosoma platyrhinos. Lampropeltis boylit conjuncta.
Sauromalus interbrachialis. Phrynosoma solare. Lampropeltis californiae.
Callisaurus carmenensis. Gerrhonotus multi-carinatus. Lampropeltis nitida.
Callisaurus crinitus. © Gerrhonotus scincicauda webbii. Rhinocheilus lecontet.
Callisaurus draconoides. Anmiella pulchra. Hysiglena ochrorhynchus.
Callisaurus plasticus. Xantusia gilberti. Chilomeniscus fasciatus.
Callisaurus splendidus. Verticaria cxzrulea. Chilomeniscus stramineus.
Callisaurus ventralis ventralis. Verticaria hyperythra beldingi. Natriz valida.

Uta concinna. Verticaria hyperythra hyperythra. Thamnophis ordinoides hammondii.
Uta graciosa. Verticaria sericea. Trimorphodon lyrophanes.

Uta mannophorus. | Cnemidophorus bartolomas. Tantilla planiceps.

Uta martinensis. Cnemidophorus celeripes. Crotalus atrox.

Uta mearnsi. Cnemadophorus labialis. Crotalus cerastes.

Uta microscutata. Cnemidophorus martyris. Crotalus enyo.

Uta nigricauda. Cnemidophorus maximus. Crotalus exsul.

Uta palmeri. Cnemidophorus multiscutatus. Crotalus lucasensis.

Uta parva. A Cnemidophorus rubidus. Crotalus mitchellit.

Uta repens. Cnemidophorus tessellatus stejnegert. Crotalus oreganus.

Uta squamata. Cnemidophorus vandenburght. Pseudemys ornata nebulosa.
Uta stansburiana elegans. Plestiodon skiltonianus lagunensis. Chelonia agassizii.

Uta stansburiana hesperis. Bipes biporus. Eretmochelys squamosa.

Uta stellata. Siagonodon humilis. Caretta olivacea.

Uta thalassina. Lichanura roseofusca. Dermochelys schlegelit.

 

 

All of the foregoing genera except four (Bipes, Otenosaura, Sator, and Phyllodactylus) range
down the peninsula from beyond its border in the southwestern United States, and the species
are mainly the same as those from the north or are more or less closely related to them. Bipes,
the most marked of the two genera of reptiles peculiar to the peninsula, includes a single species
known only from the cape district south of La Paz. Bipes belongs to a distinct family—
Bipedidae—and is related to the ancient family. Amphisbaenidae. Dr. Leonhard Stejneger in-
forms me that he considers its presence at the southern end of the peninsula good evidence of
the great age of that area. This is of special interest in view of the comparatively recent origin
of the remainder of the vertebrate fauna. The nearest relative geographically of Bipes is the
genus Euchirotes of the Mexican tableland. The nearest relatives of the new genus Sator are
Sceloporus and Uta. :

Ctenosaura, the well-known iguana, so abundant on the tropical west coast of Mexico,
is represented in the cape district by one insular species. The curious genus Sauromalus,
common in parts of the southwestern United States, ranges south to La Paz and is represented
in Lower California by three species. Phyllodactylus, a genus of geckos, which is abundant on
the tropical west coast of Mexico, is represented in the cape district by two species, one of

 

~ 1 Prog. Calif. Acad. Sci., 2d ser., V, pp. 77-162 with plates, 1895.
No. 1]

LOWER CALIFORNIA, NATURAL RESOURCES—NELSON.

115

which (P. tuberculosus), is common on the coast of the Mexican mainland. Owing to its habit
of commonly frequenting houses it is not improbable that this species may have been brought
to Lower California in household effects from Mazatlan or other coast ports.

The distribution and relationships of the reptiles in Lower California are still so imper-
fectly known that considerable modification of the following summary is certain to result
from future work. A limited number of species common to the southwestern United States
are not known to extend their ranges south beyond the northern half of the peninsula. Of
the 39 genera and 102 species and subspecies of reptiles known from the peninsula 32 genera
and 59 species and subspecies have been taken in the cape district. Of these 38 appear to be

peculiar to this area.

The species known in the cape district are named in the following list.
with an asterisk (*) are peculiar to it.

Those ‘ele

Those not so marked are widely ranging Lower Sonoran

species from the north, except the gecko (Phyllodactylus tuberculosus), which is common on the
opposite coast of the Mexican mainland.

Reprites KNown From THE Care District.

Phyllodactylus tuberculosus.
*Phyllodactylus unctus.
*Ctenosaura hemilopha.
*Ctenosaura insulana.
*Dipsosaura dorsalis lucasensis.

Crotaphytus copett.
*Sauromalus interbrachialis.
*Callisaurus carmenensis.
*Callisaurus crinitus.

Callisaurus draconoides.
*Callisaurus plasticus.

Callisaurus ventralis ventralis.
* Uta mannophorus.

Uta microscutata.

*Uta nigricauda.

* Uta repens.

* Uta squamate.

Uta stansburiana elegans.

* Uta thalassina.

*Sator angustus.

Birds.

 

*Sator grandzvus.
*Sceloporus licki.
*Sceloporus lineatulus.
Sceloporus zosteromus.
Phrynosoma coronatum.
*Gerrhonotus multi-carinatus,

| Anniella pulchra.

* Xantusia gilbertt.

* Verticaria caerulea. cf

* Verticaria hyperythra hyperythra.
Verticaria sericea.

*Cnemidophorus celeripes.

*Cnemidophorus maximus.

* Cnemidophorus rubidus.

*Cnemidophorus vandenburghi.

*Plestiodon skiltonianus lagunensis.

*Bipes biporus.
Siagonodon humilis.

*Tichanura trivirgata.

* Coluber aurigulus.

 

Coluber constrictor flaviventris.
Coluber lateralis.

| Salvadora hexalepis.
| Phyllorhynchus decurtatus.

*Pituophis vertebralis.

*Lampropeltts boylii conjuncta.

*Lampropeltts nitida.
Rhinocheilus lecontei
Hypsiglena ochrorhynchus.
Chilomeniscus fasciatus.

*Chilomeniscus stramineus.

* Natrix valida.
Trimorphodon lyrophanes.

* Tantilla planiceps.
Crotalus atrox.

*Crotalus enyo.

*Crotalus lucasensis.
Crotalus mitchellit.

*Pseudemys ornata nebulosa.

Following are lists of the genera of birds known from the peninsula, with notation of the

number of species and subspecies.

Of land birds there are 142 genera and 286 species and sub-

species; and of water birds there are 100 genera and 142 species and subspecies.

Genera or Lanp Brrps with Numser or SPeEcIES AND SuBsPEcIES Known From LOweER CALIFORNIA.

One sscsiceise ws sicias cicieeies ie 1
Lophortyx......0000.200eeeeceeee 2
Chioroenas.........20.0-00000000 1
Zengidurd......cccecececaccencee 1
Melo pelidscscsechosaesecensceeees 1
Chaemepelia..........0-0---2++- 1
Scardafella........2.0.0-0202000- 1
Gymnogyps.......-.------0-0200- 1
Cathartes........0000-00ceeeeeeee 1
PNG NAS S59 chcrorycterc aie 2 nueiesers 1
CU CB Soa dee jap aces haa 1
ACH NER ire wane beatae Psclarsiciets 2
Parabuteo ........2-2-000e0e0-e- 1
BU sc cmices Saiedienie’s eeiha bee 5
Archibuted......2..ceeeeee-e----- 1
AQUI Occ iien s seardteeet acess 1
Halineetus...:c...0-0-0-- eee fue, GE
Hicrofaleo.........-+.00-eeeereee 1
Rhynchodon. ....-+++-00e00++ ctecrs® “dy
Tinnunculus......ccccecereeesee- 2

Cerchneis....-... Saeecincs Secs Santee 2

2

 

Chloroceryle wiseeseawmens ss es Pass

Sphyrapicus.........---+ teeeeee

Setochalcis.........+6 wean se se

 

Phalaenoptilus...........-.-2-+-
Chordeales catssock oisis.eivb's Suid were

Archilochus... .
CON DC. win dda wasidwdase nis spsrecees
Selasphorus........022-202022000-
Basilinna.....
DYPONNUB ooo aecuce cenissescinwenaieie
‘116 MEMOIRS NATIONAL ACADEMY OF SCIENCES.

Genzra or Lanp Birps wita NuMBER or Species anD Sunspecies Known From Lower Caurornia—Continued.

Xanthocephalus..........20.000++
Agelaius ......-.0c0cceseeeceeees
Sturnella.........0.00cccceeenees

DUNCO sd seis ssw s cisrciciecagsiesatoerds
Amphispiza..........------ catelpia
Aimophila....... 2.2000. eeceeeee
Melospiza

Passerelas casa saselesiteaioxarelstacias

Ptychoramphus.....-.+--2++++++-
Brachyramphus.......---.++-++++-
Stercorarius.........02-2+202-0---
Riss sci seccwan scot pehonea nes

Hydroprogne .....+-+--0022ee eee
ThOlQSSCUS ss casnicins sees scien cs seiele
SLCPNA. «05 occa escsiewarnaieransre

Chiidonias.........+-2+22-02000-
Phoebastria......---.-+---s20+0-

 

 

Vireosylva.....---.2-0.sceccceeee
Lantovtreo.....----..secccecccees

 

 

 

Anthtsircs ssecewateseevewshiss 2
Oreoscoptes....... moans caieioig seid 1
Mimus..... 2.0022 eeee eee eeeee 1
DUN iosckesuimaencasagar’ ch
me ae 7

3

8

1

2

1

1

1

| Telmatodytes........--..2----2+-- 2
Bildiecsccseevassvecsieccceaacae 4
Bacolophus...........2-----+--- 2
Penthestes......00--0-202-000ee- 1
Psaltriparus..ccssceceeeeeeeceeee 2

AUNMPOTUS.....00ccecee ee eceeeees 2
CROMBEO corse smesenitinowseSeean “dL

Steganopus.. 1.2.22... e eee e eens 1
Recurvirostra..........2.0.00--24- 1

 

DAMO8O sie cocci wees cammrscs seiner (li
Wels cc ccascccocetesbaesweveian. OL

LON Ss ca cust wuccicewnce vidqaice 2
TING Os <oaiereneu asec cise: Bee. ok
Catoptrophorus.........---+.- seen ok
Heteroscelus..........--22222-5- 1
Acliti8 cesesseacsexies wanes ss 1
Numenius.......2cccceceeeeneees 2
Squatarola..............----2+-- 1
PRUNAUGS 05.5 ca vrecwtadecde we 1
Osryechus... 22.0.0 ceeececcceeee 1
Charadrius. ..........0.0e cee nees 2
PAQOUG s cacesincrenssesiosss sacien 1
BUp0GO occvccsecscsenaawes creas 1
ADRNGEs ooi sven tiie bias eSeaamns 1
Arenaria.. ae 2
Haematopus........02 0-00 enenee 2

[Vou, XVI
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 117

Mammals.
Following is a list of the 51 genera of mammals known from the peninsula, showing 166
species and subspecies.

GENERA OF Lanp MammMats with NuMBER or Species AND Subspecies KNown From LOWER CALIFORNIA.

Adocoileus........-.-...-0-005- 4| Thomomys.......-.....2200.00-- Bl Peat a Anta eS Tee IA Ss 1
Antilocapra............2.22-506- 1 | Dipodomys........-....2..0020- TW SON ic deo cescechaan bohesia ste acaleniaieneane 3
OB cc soseeeceeceeececasaescess 1 | Perodipuss.cccss... Sa emeaezens 2 | Nottosorer.........2..0220220205 1
SciUrU8. ssiossiecce sos snnecewes vs 1 | Perognathus.............02.0005 19 | Scapanus.............20.0-206- 1
Butamias....-.2.2.22.2-200050- 25) Le Puss cond Guswe cane sg ckignseee 6 | Balantiopteryr........-...-.22.- 1
Citellus. . seececeee--- 4] Sylvilagus.. 2.2.0.0 2 022 eee 8 | NMycteris.............0.20200 000 2
Ammospermophilus. Sale sidieradiaistele’s 8 | Pel8 es wsece susdeesies d's tanec asee 2 | Dasypterus.........22-2-0-0-020- 1
Casto. indeseteees ssi eaeneeknes Lo) Dey nats cits ceictecsendeie se gene nenbe 2 | Mormoops.............--.---+4- 1
Mis sos captceeerss aeacdesees 2) Canis... eee eee ences 5 | Otopterus.........----2-2-2005- 1
Onychomys......2-.-0-00ecee eens 2 | UPOCyOR se issn case ee estos 2 | Chaeronycteris........-..-+----+- 1
Peromyscus... 1.2222 2220e0- 0 eee ZO Vell peSe sinc walgreens vex anasiees 2 | Natal. oi ce see cc caceen ocene 1
Sigmodon..........-.2+-+0+-0-5- Vl Orb b is sie srevisieie gate hs acta as DV} MYOt8 sos sacisinad $44 Lonaaeleners 6
OnZ0MY8- 20. caess scdnsese ceeess 1 | Bassariscus...........-.-2-.--- 8) Pioneer oc Sales occn apa wocaticen 1
Reithrodontomys.......---+-+-+- 3) ProtyOts oc cianeueheassaneweees DE PUSS vias Sagi scien owen 2
Neotoma......2-0-220-2ee-e---0- 14] Taridea....... 00... e eee eee 2 Bi ptes us ie scien Me sideoeeeees 2
MPO vccccdecewctameseseaces, TL|) MEPRNS . cccacciell ss coadonseeie 2 | Antro20Us... e208 ss se soeezccws 3
DDC cep oteinlelsio ov oceania’ I | Spilogales x. oscusetexs 1 onecocs 8 | Nyctinomus....-.-..-.-..2----- 3

 

 

A glance over the foregoing list of genera reveals the greatly preponderating northern
origin of the mamalian fauna of Lower California, the small Mexican or tropical element being
represented by the rice rat (Oryzomys peninsulae) and half a dozen bats, including Balantio-
pteryx plicata, Mormoops megalophylla senicula, Chaeronycteris mexicana, Natalus mexicanus,
Dasypterus ega xanthinus, and Nyctinomus depressus. Some, at least, of these bats probably
migrate in fall to the Mexican mainland, since we saw no trace of {Meni 4 in winter but were
informed that they appear in considerable numbers in spring and summer.

SUBORDINATE FAUNAL . DISTRICTS.

As indicated on the outline map (see Pl. 31), Lower California contains five natural
faunal areas, here called ‘districts,’ which are partly independent of the general distribution
of the life zones. Some of these areas lie within a single life zone, others in two or more. These
areas are briefly characterized below and the lists of birds and mammals peculiar to each show
they are strongly marked. They are probably characterized equally well by other groups of
animal and plant life. The three northern areas are southern continuations of faunal areas
occupying adjacent parts of southern California, thus extending the range of most of the charac-
teristic species to both sides of the northern boundary. The most notable of the faunal areas
of the peninsula is the cape district, which is here much extended beyond the previously recog-
nized limits. Taking the areas in their sequence from north to south they stand as follows:

1, San Pedro Martir District. i

This district covers a narrow belt about 175 miles long extending from the northern border
a little east of south to Matomi, near the middle of the peninsula. It occupies the summits of
the Sierra Juarez and San Pedro Martir Mountains, above about 4,000 to 4,500 feet on the west
and above 4,500 to 5,500 feet on the east and includes all of the transition zone in the peninsulas,
with the upper part of the Upper Sonoran zone in this section. This area also is the only part
of Lower California on which snow falls and in which there is a yellow pine forest. The entire
area is mountainous with tremendous escarpments on its eastern border. It lies mainly between
4,000 and 9,000 feet altitude, but reaches the extreme elevation of over 10,000 feet on La Provi-
dencia Peak. The slopes above 9,000 feet are mainly rocky and barren. On the cooler upper
slopes there is a slight infusion of Boreal species, and along its lower border the Upper Sonoran
zone is strongly in evidence. The lists and zonal relationships of the species in this area are
detailed below in the account of the transition zone with which it is so nearly coincident.
118 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI

2. San Diegan District.

- The San Diegan district. océupies northwestern Lower California and like the preceding
district forms the southern extension of an area which has been characterized as a faunal sub-
division of southern California. It extends south of the California boundary from sea level
up to an altitude of about 4,000 feet on the west slope of the Sierra Juarez and to over 4,500
feet on the same slope of the San Pedro Martir Mountains, with a southern extension to the mouth
of El Rosario Arroyo on the coast, gradually narrowing beyond San Quintin to a point at its
southern end. The surface of this area is mainly an irregular plateau varied with coastal
mountains and scattered hills and ridges with small intervening valleys, and contains near its
southern end the coastal plains of Hamalu and San Quintin. It is exposed to the cool, damp
winds from the Pacific and more or less regular winter rains. This area lies almost “wholly
within the Upper Sonoran life zone, but in places back from the seacoast, as in Trinidad Valley
and the southern end of San Rafael Valley, there is an admixture of the Lower Sonoran zone,
which becomes especially marked in the southern part of the area. Further notes on this area,
with lists of some of the most characteristic plants, reptiles, birds, and mammals, are given
below in the account of the Upper Sonoran life zone.

3. Colorado Desert District. |

: The Colorado Desert district is the extension southward across the border of southeastern
California of the Colorado Desert. It covers all northeastern Lower California between sea
level and an altitude of from nearly 4,500 feet on the Sierra Juarez to 5,500 feet or over on
the bare east slope of the San Pedro Martir Mountains. To the southward of these mountains
it occupies the gulf slope from the shore to the tops of the bordering main range. Its southern
limit is about Angeles Bay. From Matomi, at the southern end of the San Pedro Martir
Mountains, and Point San Fermin southward it forms a narrowing strip east of the main moun-
tain chain which here more closely parallels the coast. It.extends westward about the southern
end of the San Pedro Martir Mountains and along the southwestern base of these mountains
where it meets the San Diegan district.

In the north it is mainly a plain containing the delta and flood plains of the Colorado
River, varied by the Cocopah, Sierra Pinta, and other desert mountains. Farther south the
plain is broken by the Sierra San Felipe, Sierra Santa Clara, and other barren hills and moun-
tains with intervening plains or valleys. Water is axteenialy scarce and occurs mainly in
natural tanks or tinajas along the base of the desert ranges, and as.small streams in the lower
part of the canyons along the east base of the main Juarez-San Pedro Martir Range.

Its plants are largely those of the Colorado Desert in California, but along the east base
of the San Pedro Martir Mountains and south from near San Felipe Bay numerous other species
come in. This desert area lies wholly within the Lower Sonoran Zone with a tinge of a sub-
tropical flora along the gulf coast. The characteristic birds and mammals of this area are
listed beyond under the heading of the Lower Sonoran zone.

$M

Cuaracreristic PLANTS or THE Cozorapo DEsERt Disreror OF Lower Caxaronnta,

Ephedra trifurca. Cercidium. torreyanum. Purlenvointe aculeata.
Acacia greggi. Encelia farinosa. Parkinsonia microphylla.
Prosopis glandulosa. — Hymenoclea monogyra. ‘| Isomeris arborea.
Covillea glutinosa. Pluchea sericea, Koeberlinia spinosa.
Fouquieria epee Chilopsis linearis, Pachycereus pringlei.

4, Vizcaino Desert District.

This district occupies the middle section of the peninsula, its northern boundary being
formed by the southern border of the San Diegan and Colorado Desert districts. It extends
thence southward to include the northern and northwestern, borders of the great lava ‘plateau
and southward along the coast for some distance on both sides. On the east of the lava plateau
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 31.

ne =
ve eae po,

 

     
   
 
  
   
  
   

 

ARIZONA

Se
NOGA! CS me on eee os ws soe

 

“Altar

 

S O N O R A

Y ch ermosillo
2 ‘

MAP

OF

SUBORDINATE Pes DISTRICTS

LOWER CALIFORNIA 2 deS.Lucas
E.W.N TEL SON

 

 

 

 

 

 

 
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 119

the extension of the Vizcaino district, is in the form of a narrow belt lying between the main
mountain chain and the coast extending to the vicinity of Santa Rosalia. On the Pacific side
the southern extension is in the form of a narrow wedge terminating near Santo Domingo
Point in about latitude 26° 20’. This district includes nearly all of the table-lands charac-
terized by granite formation from their northern border south to the northern end of the
voleanic plateau beyond Calmalli. It also includes all of the Vizcaino Desert, most of which
is a broad coastal plain with a low table-land on its eastern and northern borders, The plateau
country to the north is broken by scattered mountains, low ridges, and groups of granite moun-
tains separated by plains which in some places take the form of shallow troughlike or basin-
shaped valleys.

The Vizcaino district is practically all within the Lower Sonoran zone but has a mixture
of arid tropical plants, especially on the borders of the lava plateau and along the gulf coast.

A few of the higher mountain tops extend within Upper Sonoran climatic conditions,
although the area at these high elevations is so limited and so extremely barren that they
appear to possess little or no life to characterize the zone.

The Vizcaino district is notable for its extraordinary development of desert vegetation.
Two of the most remarkable and characteristic plants of Lower California are limited mainly
to its confines. The first, Jdria columnaris, known locally as the “cirio,’”’ forms: scattered,
polelike forests over most of the interior table-lands from near Calmalli north to beyond San
Fernando, the species ranging but little beyond these limits. (See Pl. 29, Fig. 1.) The second,
Veatchia discolor, the elephant wood, is characteristic of this region but has a somewhat wider |
range than Jdria. (See Pl. 28, Fig. 2.) In addition a large part of the interior granitic table-
lands within this district is covered with an extraordinarily large and vigorous growth of giant
and other cactuses, yuccas, agaves, palo verdes, fouquierias, and numerous other desert trees and
shrubs which over great areas form low thorny forests or jungles. The plants of this area have
been so little studied that besides Jdria and Veatchia it is difficult to list the peculiar species,
although they must be numerous.

‘As-in the case of the plants, the reptiles of this district are little known and the species
so far taken within its limits appear to range widely in adjacent regions, most of them belonging
also in the Colorado Desert district or to the cape district. The most characteristic species
of birds and mammals are listed beyond in the account of the Lower Sonoran zone.

&. Cape District.

The cape district is the southernmost of the subordinate faunal areas of Lower California.
Its strikingly localized fauna and flora have attracted the attention of naturalists sinee about
1861, when they were first made known through the collections of John Xantus. As a result
it has been visited by numerous naturalists, and has contributed more material and had more
published concerning it than any other section of the peninsula.

The mountainous southeastern end of the peninsula south of La Paz is separated from the
mountains farther north by a broad low pass extending from the head of La Paz Bay to the
Pacific. The islandlike character of this southern area and the fact that the peculiar flora and
and fauna of southern Lower California is most strongly marked there has led Bryant, Brandegee,
Brewster, and other writers to define it as the ‘‘Cape Region.” With the knowledge of the
distribution of species now available, however, it becomes necessary to extend its limits north-
ward to include all of the peninsula near Santa Rosalia on the gulf coast and thence to near
the northern boundary of the lava plateau in the middle of the peninsula and southeasterly
to the Pacific coast in about latitude 26° 20’. The cape district as thus defined is composed of
two sections: (1) A northern one, north of the pass at La Paz Bay, which comprises the Mag-
dalena Plain along the Pacific coast and the interior lava capped plateau, rising to include

 

1 This area has commonly been known as the ‘‘Cape Region,” but the use of the term ‘‘region” in a faunal sense for so limited an area appears
inappropriate, and I have substituted the word district for this and other similar faunal areas in the peninsula.
120 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

the Sierra de la Giganta along the gulf coast; (2) that part. of the peninsula south of La Paz
Bay, which is the area formerly considered. ‘ei constitute all of the ‘‘Cape Region” is mainly
occupied by granite, limestone, and volcanic mountains.. In the northern section of the cape
district taking in most of the lava-covered plateau with the main range of mountains along
the Gulf coast, including the Sierra San Francisco, Sierra Santa Lucia, and Sierra de la Giganta,
Tises in a few places to altitudes. of over 5,000 feet, but averages from 2,000 to 4,000 feet and
slopes westward across the plateau to the Magdalena Plain and the Vizcaino Desert. South
of La Paz some of the mountain peaks of the Sierra Victoria rise to elevations of from 6,000
to 8,000 feet in the form of sharp granite points, but the main range averages from 3,500 to
6,000 feet altitude. The Sierra Victoria has the most regular summer rains of any part of the
peninsula and in consequence has comparatively abundant vegetation, especially on its eastern
flanks and summit.

In climatic conditions the coastal lowlands, foothills, and lower mountains up to 4,000
fest or more of the cape district are Arid Tropical, but the plants are a mixture of Arid Tropical
and Lower Sonoran species. The reptiles, birds, and mammals are mainly Lower Sonoran with
a small tropical element, the: preponderance of Lower Sonoran species of animals in this being
due to the fact that the only:connection with the rest of the continent by land for a great. period
has been through California on the north. The foothills have a mixture of Arid Tropical and
Lower Sonoran species up to over 2,000 feet, the upper border only becoming more strongly
Lower Sonoran, which thus forms a narrow belt of this zone from 2,000 to about 4,000 feet.
From about 4,000 feet to the summit the mountains are Upper Sonoran. Heretofore no attempt
appears to. have been made to separate what has been termed the ‘‘Fauna‘of the Cape Region”’
into its zonal-elements. It should here be stated, however, that it is difficult to recognize
more than two life zones in the cape district, since, owing to its limited area and prevalence
of Lower Sonoran species, the Arid Tropical and Lower Sonoran zones are almost inextricably
interwoven. The reptilian fauna of the cape. district is much more strongly marked than that
of any other part of the peninsula, but through lack of sufficient collections and information
on distribution I have been unable to separate the species zonally, as in the case of the birds
and mammals. The reptiles characteristic of the cape district are listed and discussed under
the general treatment of the reptilian fauna of the peninsula (p. 115.).

A number of species of birds, trees, and shrubs occur in Lower California only in the moun-
tains of the cape district south of La Paz, and some of these are peculiar to that section, a fact
which emphasizes the islandlike character of this area. As stated elsewhere, there is no doubt
that the sea once separated this extremity of the peninsula - from the main body at the point
where the low pass now exists opposite La Paz Bay, and during this period of isolation .many of
the local species probably originated. Lists are subjoined of the birds and mammals of the
cape district subdivided according to their zonal distribution in this area.

CuaracrEnistic Brrps oF Tae Care District ‘or LowzR CALIFORNIA eee Accorpine To ZONAL DisTRISUTION.
ARID TROPICAL. ,

Ardea herodias sanctilucae. Biniiros cheriway auduboni. Passerina versicolor pulchra. .

Egretta thula brewstert. , 7 Crotophaga sulcirostris. Melospiza melodia rivularis.
Butorides virescens frazari. ; Pyrocephalus rubinus mexicanus, | Dendroica erithachorides castaneiceps. :
Rallus beldingi. Richmondena cardinalis ignea. , Geothlypis beldingi beldingt.
Chaemepelia ae pales: Pyrrhuloxia sinuata peninsulae.

ARID TROPICAL AND LOWER SONORAN.

Cerchneis sparveria peninsularis, Myiarchus cinerascens pertinax. Heleodytes brunneicapillus afinis.
Otus asio zantusi. = Aphelocoma californica hypoleuca. Auriparus flaviceps lamprocephalus.. .
Dryobates scalaris lueasanus. Carpodacus mexicanus ruberrimus. Polioptila plumbea plumbea.
Centurus uropygialis brewstert. Pipitlo fuscus albigula. i ;

Colaptes chrysoides chrysoides. Torostoma cinerea cinerea.
Memoirs National Academy of Sclences, XVI.

 
 

Sell
i

24?

23°

Blo2

 

1°

 

 

 

 

 

LIFE ZONES
OF
LOWER CALIFORNIA |;
BY

E. W. NELSON

ARID TROPICAL ZONE

 

LOWER SONORAN ZONE
UPPER SONORAN ZONE
TRANSITION ZONE

15° 114? 3°

3B
Cartage ° Sojued
°,

Uicaray 99: Pherig

Stony,
\PPlaceres

Teerete o

TS Tour

 Gaern

Pax
UV

12

BLS Famciscy

 
  

PAP oie
4 Espirine s}
° -} ike

First Memoir, Pl. 32.

 
       
  

 

 

|
{
|
|
|

|
|

    

courte
% ! Carrehte
aly PL lemons
ep oi

   

PPascaderes
}

% 5
7 Tadmas

PArunas
apc
by R Pectin

- Pedians
Percis i

pes eh At h

i
SinprabodeS Loess

    
 

  

|

 
No. 1)

Chloroenas fasciata vioscae.
Glaucidium gnoma hoskinsi.
Balanosphyra formicivora angustifrons.
Basilinna xantusi.

Myiochanes richardsonit peninsulae.

LOWER CALIFORNIA, NATURAL RESOURCES—NELSON.

UPPER SONORAN.

Junco bairdi.

Aimophila ruficeps sororia.

Pipilo maculatus magnirostris.
Tachycineta thalassina brachyptera.
Lanivireo solitarius lucasanus.

121

. Vireo huttoni cognatus.

Sitta carolinensis lagunae.
Baeolophus inornatus cineraceus.
Pesaltriparus minimus grindae.
Planesticus confinis.

CHARACTERISTIC MAMMALS OF THE Care District, Listep AccoRDING TO Zona DIsTRIBUTION.

ARID TROPICAL AND LOWER SONORAN.

Citellus atricapillus. Lepus californicus zanti. Natalus mexicanus.

Ammospermophilus leucurus peninsu- | Sylvilagus auduboni confinis. Myotis volans.
Sylvilagus bachmani peninsularis. Myotis peninsularis.

Peromyscus maniculatus magdalenae. Vulpes macrotis devia. Myotis micronyz.

Peromyscus eremicus eva.
Oryzomys peninsulae.

Bassariscus astutus palmarius.
Procyon psora.

Pizonyx vivesi.
Pipistrellus hesperus australis.

Neotoma intermedia pretiosa. Taxidea tarus infusca. Eptesicus fuscus peninsulae.
Neotoma intermedia arenacea. Spilogale microdon. Dasypterus ega xanthinus.
Thomomys botiae anitae. Spilogale lucasana. Antrozous pallidus minor.
Dipodomys merriami melanurus. Balantiopteryz plicata. Nyctinomus depressus.
Perognathus baileyi rudinoris. Mormoops megalophylla senicula.

Perognathus spinatus peninsulae. Chaeronycteris mexicana.

UPPER SONORAN.

Ovis canadensis. Thomomys bottae alticola.

Peromyscus truei lagunae.

Sorex lagunae.

 

 

Mammats Raneina THrRouaHout THE Forraorna ZONAL AREAS.
(Not restricted to the Cape District.)

Neotoma intermedia intermedia.
Lynx ruffus californicus.

LIFE ZONES.

Canis peninsulae.
Urocyon cinereoargenteus californicus.

Odocoileus hemionus peninsulae.
Peromyscus maniculatus coolidget.

Lower California contains four life zones and a trace of a fifth, as follows: The Arid Tropical,
Lower Sonoran, Upper Sonoran, Transition, and Canadian. Of these the Lower Sonoran is by
far the most important, covering about three-fourths of the entire land area and extending
from one coast to the other and practically the entire length of the Peninsula. It contains
the majority of the characteristic and conspicuous plants and animals. The great preponder-
ance of this zone led Dr. Merriam to make the general statement that ‘‘Lower California is a
subdivision of the Arid Lower Sonoran zone.’’!

It should be stated here that the islands along both coasts agree in zonal characteristics
with the adjacent mainland except in a few instances which are noted.?_ These life zones, taken
in their sequence, are characterized as follows: .

Arid Tropical Zone.

The southern end of Lower California lies within the Tropics and, having an extremely
dry climate, naturally belongs to the Arid Tropical zone. This zone in the peninsula, how-
ever, is not confined to that section south of the Tropic of Cancer but extends along the
gulf coast from the shore back over the foothills of the adjacent mountains, and northward
to San Lucas at the northern end of Llano de San Bruno. Thence it leaves the shore
but continues northward along the base of the mountains some distance beyond the Tres
Virgines Peaks. In addition, all of the gulf islands north to opposite San Lucas are also

 

1 Proc, Biol. Soc. Washington, VII, p. 29, April, 1892.

Dr. Paul Bartsch informs me that the littoral marine invertebrate fauna (especially the mollusks) of Lower California belong to two sub-
divisions. The first extends along the west coast from near Santa Barbara, Calif., to near Cape San Lucas. The other includes both coasts of the
Guifof California, and extends thence southward to Guatemala and is known as the “ Mazatlanic Fauna,” a subdivision of the Tropicalian.
‘122 ‘ MEMOIRS NATIONAL : ACADEMY OF SCIENCES. [Vou. XVI

within this range. From La Paz Bay to Cape San Lucas the Arid Tropical zone covers all of
the peninsula up to about 2,000 feet on the north and east slopes, and 2,500 feet on south and
west slopes of the foothills and up the canyons cutting into the mountain ranges. This leaves
only a limited Upper and Lower Sonoran area capping the northern part of the Sierra de la
Victoria. This Arid Tropical area extends through the broad low pass to the west and south of
La Paz to the Pacific coast and thence in a broad belt occupying all of the Magdalena Plain
and adjacent lower slopes of the mountains northward to the southern end of the Vizcaino
Desert about Ballenas Bay and inland along the lower slopes and deep canyons of the interior
plateau to include San Ignacio and Natividad Canyon, northwest of San Ignacio. The Arid
Tropical area gradually descends to the northward from about 2,500 feet in the southern end of
the peninsula to less than 1,000 feet near San Ignacio, The exact definition of this zone and
its relation to the Lower Sonoran zone is extremely difficult owing to the strong mixture of
Lower Sonoran species throughout the Arid Tropical area, and the extension of numerous Arid
Tropical plants well within the Lower Sonoran lines.

Above an altitude of about 1,000 feet, south of La Paz, Arid Tropical plants. become con-
spicuously less numerous, and Lewar Sonoran species increase. The few tropical birds occurring
in the peninsula all range below 2,000 feet. The Lower Sonoran species of birds and mammals
range in equal abundance from sea level up through both Arid Tropical and Lower Sonoran
zones, thus adding to the confusion of the upper limit of the Arid Tropical zone of this region
and rendering its definition possible only through the distribution of its plant life. Imme-
diately north of La Paz Bay a similar relationship exists between the zones, the Upper Sonoran
zone being represented on the upper slopes of Mechudo Peak, but, as already stated, the alti-
tude of the upper limit of the Arid Tropical gradually decreases northward. The limit of this
last-named zone extends farther north along the basal slopes of the mountains and canyons of the
great lava plateau northwest of San Ignacio than along the coast. It will be noted that the
northern extension of the Arid Tropical zone is nearly coincident with the northern limits of the
cape district, as it is here termed.

As noted elsewhere, the tropical element in the vertebrate animal life of Lower California,
practically limited to this district, is surprisingly small and is insufficient in quantity and of
too doubtful origin to prove any former land connection between the peninsula and the Mexi-
can mainland within the period in which the present species have existed.

Plants of Arid Tropical Zone.—As might be expected, the number of Arid Tropical plants
both of species and individuals decrease gradually from the extreme southern end of the pen-
insula ‘northward. By the overlapping common to adjacent zonal areas a number of Arid
Tropical species range far north of the limits of the Arid Tropical zone here given. _

Within this zone in Lower California the banana, coconut, and date palms are cultivated.
The tamarind (Tamarindus indica) and the guamuchli (Pithecolobium dulce) have also: been
introduced and have escaped from cultivation in some places south of La Paz Bay. Among
other notable Arid Tropical plants in this zone of the peninsula are Ficus palmeri, Lysiloma
microphylla. and L. candida, Mimosa zanti, several species of Jatropha, Haematozylon, boreale,
Xanthoxylon pterota, two species of Lantana, several species of Elaphrium, Manihot chlorosticta,
Ibervillea sonorae, and numerous cactuses including the extraordinary Lemaireocereus eruca, Sev=
eral opuntias, and the giant species Pachycereus pecten-aboriginum and P. calvus. The ‘fore-
going covers only some of the more important species and might be considerably extended.

It is interesting to note that a number of species of plants characteristic of the Arid Tropical
parts of southern Lower California occur also in the arid subtropical coastal belt of southern
Texas. Among these are Xanthozylon plerota, Lantana camara, Brayodendron texanum, Karwin-
skia lwmboldtiana, and Vachellia farnesiana.

The tropical element in the plant life of this zone evidently originated on the coastal plains
and foothill slopes of that part of the Mexican mainland lying directly eastward across the
Gulf of California.
No. 11 LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 123

Puants CHARACTERIZING THE ARmp Tropical ZonE IN LowER CALIFORNIA.

[Species marked with an asterisk (*) occur also on the Mexican mainland.]

Roccella sp. (?)

*Cocos nucifera (introduced).
Phoemx dactylifera (introduced).
Ficus palmeri.

Schoepfia californica.
Antignon leptopus.

Celosia floribunda.

Fresine alternifolia.
Phaulothamnus spinescens.
Steganosperma halimifolia.

* Atamisquaea emarginata.
*Forchammeria watsoni.
Acacia flexicaulis.

Albizzia occidentalis.
Lysiloma microphylla.
Lysiloma candida.

Mimosa purpurascens.
Mimosa xanti.

*Pithecolobium dulce (introduced).

Prosopis palmeri.

* Vachellia farnesiana.
Caesalpinia placida.

Caesalpinia mexicana californica.
Cassia confinis.

Cassia occidentalis.

Cassia articulata.

Cassia goldmani.

Cercidium peninsulare.
Haematozylon boreale.

* Tamarindus indica (introduced).
Erythrina purpusi.

Esenbeckia flava.

* Xanthozylon pterota.
Elaphrium rhoifolium.
Elaphrium odoratum.

 

Elaphrium cerasifolium.
Elaphrium epinnatum.
Elaphrium filicifolium.

_Elaphrium goldmani.

Malpighia diversifolia.
* Mascagnia macroptera.
Argithamnia brandeget.

. Croton magdalenae.

Euphorbia californica.
Euphorbia zanti.

* Jatropha canescens.

* Jatropha cordata.
Manihot chlorosticta,
*Sebastiana bilocularis.
Tapirira edulis.
Maytenus phyllanthoides.
* Dodonaea viscosa.
Colubrina glabra.

*Karwinskia humboldtiana.

Abutilon californicum.
Abutilon palmeri.
Gossypium davidsoni.
Gossypium harknessi.

* Melochia tomentosa.

* Turnera humifusa.
Echinocactus rectospinus.
Echinocactus falconeri.
Echinocactus digueti.
*Lemaitreocereus thurbert.
Lemaireocereus gummosus.
Lemaireocereus eruca.
Lophocereus australis.
Mammillaria roseana.
Opuntia tapona.

Opuntia comonduensis.

 

*Pachycereus pecten-aboriginum.
Pereskiopsis brandegeet.
*Wilcovia striata.
*Rhizophora mangle.
*Conocarpus erecta.
Gongylocarpus fruticulosus.
Hauya arborea.

* Brayodendron texanum.
*Plumiera mexicana.
Vallesia dichotoma.

* Ipomoea pes-caprae.
Jacquemontia abutiloides.
*Duranta plumieri.
*Lantana camara,
*Lantana involucrata.
LInppia barbata.

Lippia formosa.

*Lippia palmeri.
Mesosphaerum insulare.
Mesosphaerum laniflorum.
Mesosphaerum palmeri.
*Crescentia cujete,
Beloperone purpusi.
Beloperone californica.
Justicia insolita.

Justicia palmeri.

Ruellia sp.

* Bourreria sonorae.

| *Chiococca racemosa.

*Tbervillea sonorae.
Alwvordia fruticosa.
Bebbia atriplicifolia.
Eupatorium peninsulare.
Palafoxia arenaria.
Porophyllum confertum.

Reptiles of Arid Tropical Zone.—The only reptiles in the southern end of Lower California
which appear to have originated on the tropical lowlands of the opposite shore of Mexico area
species « of iguana (Ctenosaura hemilopha) and two species of geckos (Phyllodactylus tuberculosus

and P. unctus).

Birds of Arid T; ropical Zone.—Of the large number of species in the land-bird fauna of the
peninsula only the nine listed below probably originated across the gulf on the mainland coast

of Mexico.

Polyborus cheriway auduboni, resident throughout the year.
Chaemépelia ‘passerina pallescens, resident throughout the year.

Crotophaga sulcirostris, resident in summer only(?).

Basilinna xantusi (not tropical), resident throughout the year.
Richmondena cardinalis ignea, resident throughout the year.
. Pyrrhulozia sinuata peninsulae, resident throughout the year.

Passerina versicolor pulchra, resident in summer only(?).

Dendroica erithachorides castaneiceps, resident throughout the year.
Geothlypis beldingi beldingi, resident throughout the year. °

All of these species except Crotophaga sulcirostris and Passerina versicolor pulchra are perma-

nent residents in the cape district.

These two exceptions are well-known summer residents,

but from’ November to February, 1905-6, we saw not a single individual of either, and I am
inclined to believe that they migrate in fall to the mainland across the gulf.
124 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

Of the Mexican mainland element in the bird fauna of the cape district, Chaemepelia passe-
rina pallescens, Crotophaga sulcirostris, Passerina versicolor pulchra, and Dendroica erithachorides
castaneiceps belong to the tropical lowlands on. both sides of the gulf, the specimens from Lower
California being indistinguishable from those of the mainland. Richmondena cardinalis ignea
and Pyrrhulozia sinuata peninsulae are fairly well-marked local subspecies peculiar to the cape
district but represented on the coastal plain across the gulf by closely related forms. Geothylpis
beldingi belongs to.a group of large closely related an eas common in marshes on the
Mexican tableland.

The birds that:from their distribution near sea level or al low elevations may be said to
characterize the Arid Tropical zone of Lower California, and to distinguish it from the adjoining
and overlapping Lower Sonoran zone are:

Ardea herodias sancti-lucae. Polyborus cheriway auduboni. Passerina versicolor pulchra.

Egretta thula brewstert. | Crotophaga sulcirostris. | Dendroica erithachorides castaneiceps.
Butorides virescens frazart. Pyrocephalus rubinus mexicanus. Geothlypis beldingi beldingt.

Rallus beldingi. Richmondena cardinalis ignea. Geothlypis beldingi goldmani.
Chaemepelia passerina pallescens. Pyrrhuloxia sinuata peninsulae.

It will be noted that the foregoing list contains several species and subspecies which also
range north in the Lower Sonoran zone of Sonora and Arizona, and others which are represented
there by closely related subspecies, thus showing how slightly the really tropical element is
represented in Lower California.

Mammals of Arid Tropical Zone.—With the exception of a few bats the only represent-
ative of the tropical mammal fauna of the Mexican coast is a species of rice mouse (Oryzomys
peninsulae), the distribution of which appears to be limited to the single small marsh about
the lagoon formed by the mouth of the San Jose River back of the landing place on the shore
at San Jose del Cabo, where it may readily have been introduced in shipments of bananas or
other farm products from Mazatlan or elsewhere along the mainland coast, where a related
species is plentiful.

Of the bats known from Lower California the following seven species appear to have come
from the Mexican mainland:

Balantiopteryx plicata. Natalus mexicanus. Pizonyx vivesi.
Mormoops megalophylla senicula. Dasypterus ega xanthinus.
“Chaeronycteris mexicana. - . . Nyctinomus depressus.

Of these, Balantiopteryx plicata, Natalus mexicanus, and Dasypterus ega xanthinus msy be
classed as characteristic of the Tropical life zone. The others range at least well through the
Lower Sonoran zone.

Lower Sonoran Zone.

The Lower Sonoran life zone predominates in Lower California and covers approximately
three-fourths of the total land area of the peninsula. It includes practically all of the Colorado
and Vizcaino Desert districts and a large part of the cape district. In the interior it extends
from the northern border, near the delta of the Colorado, southeasterly nearly to Cape San Lucas.
It borders the gulf coast south to about Santa Rosalia, where it merges into the Arid Tropical
zone. Here, in the middle of the peninsula, the Lower Sonoran crosses to the west side and
borders the Pacific coast from the northern edge of the Arid Tropical zone, about latitude 26° 20’,
north to about El Rosario River, where it meets the southern border of the Upper Sonoran area
which occupies the northwestern part of the peninsula. In the northeastern and middle sections
of the peninsula the Lower Sonoran zone extends from sea level up to from 4,000 to 4,500 feet,
thus covering most of the desert mountain ranges of this region, including practically all of the
Cocopah, San Felipe, Santa Clara, Santa Lucia, Vizcaino, and other ranges. Parts of the desert
ranges and peaks rising more than 4,000 feet above sea level are commonly within the Upper
‘Sonoran climatic conditions, but these areas are so limited that they possess but few if any of
the representatives of animal and plant life characterizing that zone. On the east side of the
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 125

Sierra Juarez-San Pedro Martir Range the Lower Sonoran zone extends to about 4,000 feet and
even higher on hot southerly slopes.

The Lower Sonoran zone crosses the high Sierra Juarez-San Pedro Martir Range through
San Matias Pass and spreads over all the bottom of Trinidad Valley. A number of Lower
Sonoran plants, birds, and mammals range north from this valley across Alamo Plain to the
southern end of San Rafael Valley. There is another Lower Sonoran strip at the west base of
the southern end of the San Pedro Martir Mountains extending south from San Antonio ranch,
from which the coastal mountains immediately to the west cut off the cool damp coast winds.

From the northern limit of the Arid Tropical zone south to the end of the peninsula the Lower
Sonoran zone lies inland and its lower limit gradually rises until in the section between La Paz
and Cape San Lucas it is about 2,000 to 2,500 feet above sea level. The zone extends thence up
on the mountain slopes to an altitude of about 4,500 feet, where it is replaced by the Upper
Sonoran cap on the higher mountains. Owing to the northern’ origin of the great majority of
the reptiles, birds, and mammals of Lower California, the Lower Sonoran element in the fauna,
and to a considerable but smaller extent the same element in the flora, spreads throughout the
Arid Tropical zone of the peninsula and thoroughly dilutes it. On the other hand, many of the
Arid Tropical plants range up well into the Lower Sonoran zone, especially along the gulf coast.
The Lower Sonoran zone is strongly defined throughout its extent north of the northern limit
of the Arid Tropical zone.

It will be noticed in the subjoined lists that the Lower Sonoran zone of this region is charac-
terized by most of the same species, or close relatives, of the plants, birds, and mammals notable
in the same zone in southeastern California and southern Arizona. Numerous species and genera
of cactuses, and the yuccas, agaves, creosote bushes (Covillea), mesquites, paloverde (Parkin-
sonia), and Fouguieria are especially prominent. With these such peculiar local species as the
strange cirio (dria columnaris) and elephant wood (Pachycormus discolor) give the landscapes
of the middle region a strange individuality.

The coastal islands in the northern half of the gulf are entirely Lower Sonoran, but from
Santa Rosalia southward are largely Arid Tropical in climate, although they are mainly occupied
by Lower Sonoran species of plants, birds, and mammals. On the Pacific coast Cedros Island
is Upper Sonoran on its northern and western slopes, but appears to be Lower Sonoran on
sheltered parts of its southern end. Both Margarita and Magdalena Islands, while close to the
Arid Tropical shore belt, are intermediate between this and the Lower Sonoran zone and are
occupied by a mixture of Arid Tropical and Lower Sonoran species.

The following lists contain the most characteristic species of this zone. Owing to the divi-
sion of the peninsula into subordinate faunal areas the bird and mammal lists are subdivided to
show the distribution in different districts. The distribution of the plants is not sufficiently
well known to subdivide them in the same way, but a short list is given of species characteristic
of this zone in the cape district.

Plants of Lower Sonora Zone.—The following list contains a few of the species character-
izing the Lower Sonoran zone in Lower California.

PLaNnts CHARACTERIZING THE LoweR SonorAN Zone in LOWER CALIFORNIA, _

[Species marked with an asterisk e occur also in California, and those marked with a dagger (t) occur also on the Mexican mainland. ).

+* Ephedra trifurca. Agave goldmaniana. * Acacia greggi..
Erythea armata. Quercus brandegei.. Prosopis palmeri.
Erythea brandegeet.: ‘| Quercus idonea. t*Prosopis glandulosa.
*Nolina bigelovii. Eriogonum elongatum. *Prosopis odorata.
Yucca valida. Eriogonum orcuttianum. | Cassia purpusi.
Agave cerulata. Eriogonum trichopodum. . t*Cercidium torreyanum.:
Agave aurea. {Atriplex linearis. Hoffmanseggia macrophylla. «.
Agave consociata. Hesperonia californica. t*Parkinsonia aculeata.
Agave nelsoni. .  Quamoclidion triflorum. }*Parkinsonia microphylla.
Agave promontorii. *Isomeris arborea. t*Olneya tesota.

Agave verans. Wislizenia palmeri. {*Parosela spinosa.
126

MEMOIRS NATIONAL ACADEMY OF SCIENCES. .

(Vou. XVI

’ PLants CHARACTERIZING THE LowER Sonoran ZONE IN Lowz#R CaLirornia—Continued.

t* Covillea glutinosa.
Elaphrium macdougalit.
Acalypha californica.
Euphorbia tomentulosa.
Pedilanthus macrocarpus.

t*Stmmondsia californica. ”
Pachycormus discolor.
Viscainoa geniculata.

*+ Fouquieria splendens.

A Few or THe PLANTs CHARACTERIZING THE LowER SONORAN Z

Quercus idonea.
Quereus brandegeei.
Erythea brandegeci.
Yucea valida.

 

Agave aurea.

Fouquieria peninsularis.
Idria columnaris.
Koeberlinia spinosa. '
Echinocereus. brandegeet.

+Lophocereus schottit.
Opuntia clavellina.

'+Pachycereus pringlet.
Cordia watsoni. :

* Mesosphaerum emoryt.

Agave promontorit.
Agave vexans.
Quamoclidion triflorum.
Prosopis palmert. °
Jatropha spathulata.

Salvia californica.
*Chilopsis linearis.

{ Hymenoclea monogyra.
*Chrysoma palmeri.

Encelia farinosa. 4
*Lepidospartum squamatum.
tPeucephyllum schottii.

_ t*Pluchec sericea.
. Porophyllum gracile.

ONE IN THE Care DisTRICcT.

Pedilanthus macrocarpus.
 Fouquiera peninsularis.
Opuntia clavellina.

 

Reptiles. of Lower Sonoran Zone.—The reptile fauna of the Lower Sonoran zone is very
closely like that of the same zone in southeastern California, as shown by the imperfect list given.
below. Others undoubtedly belong in this list, but lack of definite information concerning their
distribution prevents their inclusion.

- Coleonyx variegatus.
_Dipsosaurus dorsalis dorsalis.
Crotaphytus collaris baileyi. .
Sauromalus hispidus. a

Caltisaurus splendidus.

Callisaurus ventralis ventralis.
Uta graciosa.

Uta parva.

Phrynosoma platyrhinos.
Phrynosoma solare.

Cnemidophorus bartolomas.
Cnemidophorus martyris.
Pituophis catenifer deserticola.

Birds of Lower Sonoran Zone.—Among the birds the desert thrashers (Toxostoma), cactus
wrens, verdins, white-throated towhees, Xantus jays, ash-throated flycatchers, road runners,
gilded flickers, ‘ladder-backed and Gila woodpeckers, and white-winged doves are the most com-
mon and generally distributed. In addition the ravens, Harris hawks, and Mexican falcons are

not uncommon.

Characteristic birds of the Lower Sonoran. zone listed according to the faunal districts in which they occur.

 

Northeastern section (Colorado Desert district).

Middle section (Vizcaino Desert district).

Southern section (cape district).

 

Butorides virescens anthonyi.
Ardea herodias treganzai.
Lophortyz gambelit gambelit.
Melopelia asiatica mearnsi.
Parabutéo unicinctus harrisi.

Bubo virginianus. pallescens.
Dryobates scalaris cactophtlus.
Centurus uropygialis uropygialis.
Colaptes chrysordes mearnst.
Myiarchus cinerascens cinerascens.
Otocoris alpestris leucansiptila.
Molothrus ater obscurus.

Agelaius phoeniceus sonoriensis.
Carpodacus mexicanus frontalis.
Melospiza melodia saltonis.

Dendroica aestiva sonorana.
Geothly pis trichas occidentalis.
Toxostoma bendirei. ‘
Toxostoma lecontei lecontei.
Heleodytes brunneicapillus couest.
Auriparus flaviceps ha aviceps. :
Polioptila plumbea plumbea.:

 

Lophortys californica vallicola.
Melopelia asiatica mearnsi.
Parabuteo unicinctus harrisi.

Bubo virginianus elachistus.
Dryobates scalaris eremicus.
Centurus uropygialis brewsteri.
Coan chrysoides brunnescens.
Myiarchus cinerascens pertinax.
Otocoris alpestris enertera.

Aphelocoma californica hypoleuca,
Amphispiza nevadensis cinerea.

'| Pipilo fuscus aripolius.

Toxostoma cinerea mearnst..
Toxostoma lecontet arenicola.

‘| Heleodytes a.

Auriparus flaviceps
Powapeiia plumhea. plumbea.

aplert he eure vaiticola.
ne na asiatica mearnsi. -
an uteo unicinctus harrisi. -
Cerchnets sparveria peninsularis.
Otus asio xantusi.
Bubo virginianus elachistus.
Dryobates scalaris lucasanus.
Centurus uropygialis brewster.
Colaptes chrysoides chrysoides.
Myarchus cinerascens pertinaz.
Otocoris alpestris enertera.

Aphelocoma californica hypoleuca

Carpodacus mexicanus ruberrimus.

Pipilo fuscus albigulus.

Toxostoma cinerea cinerea.

Heleodytes brunneicapitlus affinis.
fan

Auriparus flaviceps lamprocephalus.
Potwptila plumbea piumbea,

 

 
No. 1]

LOWER CALIFORNIA, NATURAL RESOURCES—NELSON.

127

Mammals of Lower Sonoran Zone.—The usual mammal genera of the southwestern deserts
are here in abundance, including Antilocapra, Ammospermophilus, Peromyscus, Neotoma, Dipo-
domys, Perognathus, Lepus, Sylvilagus, Canis, and Vulpes. On the desert ranges of the east-

ern part of the peninsula mountain sheep are numerous.

The characteristic mammals of the

Lower Sonoran zone are listed according to the faunal districts, as follows:

Characteristic mammals of the Lower Sonoran zone listed according to the faunal districts in which they occur.

 

Northeastern section (Colorado Desert district).

Middle section (Vizcaino Desert district).

Southern section (cape district).

 

Odocoileus hemionus eremicus.
Antilocapra americana _peninsularis.
Ovis canadensis cremnobates.

Citellus tereticaudus.
Ammospermophilus leucurus leucurus.
Castor canadensis frondator.
Onychomys torridus perpallidus.
Peromyscus maniculatus sonoriensis..

Peromyscus crinitus stephens.

Peromyscus eremicus eremicus.

Sigmodon hispidus eremicus.

Reithrodontomys megalotis megalotis.

Neotoma albigula venusta.
Neotoma intermedia gilva.

Neotoma ee
Fiber zibethicus pallidus.

Dipodomys deserti. ;
ipodomys merriami simiolus.

Perognathus penicillatus angustirostrts.

Perognathus fallax pallidus.
Perognathus spinatus spinatus.

Lepus californicus deserticola.

Sylvilagus arizonae arizonae.

Felis aztecus brown.
Lynz ruffus baileyt.
Canis mearnst.
Canis lestes.

158961°—21——_9

 

Odocoileus hemionus eremicus.
Antilocapra americanus peninsularis.
Ovis canadensis cremnobates.

Eutamias merriami meridionalis.
SAAN Gp EE leucurus peninsu-
ae.

Peromyscus maniculatus coolidget.

Peromyscus eremicus eva.

Thomomys bottae russeolus.

Dipodomys platycephalus.

Perodipus simulans peninsularis.
Perognathus baileyi rudinoris.
Perognathus penicillatus arenarius.

Perognathus spinatus spinatus.

Lepus californicus martirensis.
Sylvilagus auduboni confinis.
Sylvilagus bachmani exiguus.

Felis aztecus.
Lynx ruffus californicus.

Canis pentnsulae.

 

Odocoileus hemionus peninsulae.
Antilocapra americanus peninsularis.
Ovis canadensis cremnobates.

Citellus atricapillus.

Apeoipermopliis leucurus peninsu-

ae.

Ammospermophilus leucurus insularis
(Espiritu Santo Island).

Peromyscus maniculatus magdalenae
(Magdalena ae

Peromyscus maniculatus margaritae
(Margarita Island).

Peromyscus eremicus eva.

Peromyscus eremicus avius (Ceralvo
Island).

Peromyscus eremicus insulicola (Espi-
ritu Santo Island).

Peromscus eremicus polypolius (Mar-
garita Island).

Neotoma intermedia pretiosa.

Neotoma intermedia arenacea.

Neotoma intermedia vicina.

Neotoma intermedia perpallida (San
Jose Island).

Neotoma abbreviata (San Francisco
Island).

Neotoma nudicauda (Carmen Island).

Thomomys bottae anitae.

Thomomys magdalenae (Magdalena
Island). :

Dipodomys merriami melanurus.

Di oe margaritae (Margarita Is-
and).

Dipodomys insularis (San Jose Is-
land.)

Perognathus baileyi rudinoris.

Perognathus penicillatus arenarius.

Perognathus penin ian ammophilus
(Margarita Island).

Perognathus penicillatus siccus (Cer-
alvo Island).

Perognathus spinatus peninsulae.

Perognathus Tne magdalenae
(Magdalena Island).

Perognathus spinatus margaritae (Mar-
garita Island.).

Perognathus spinatus bryanti (San
Jose Island).

Lepus californicus xantt.

Lepus californicus magdalenae (Mag-

alena Island).

Le al (Espiritu Santo Is-
and).

Sylvilagus auduboni confinis.

Sylvilagus backmani peninsularis.

Sylvilagus mansuetus (San Jose Is-
land). ‘

Felis aztecus.

Lynx ruffus californicus.

Canis peninsulae.
128 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

Characteristic mammals of the Lower Sonoran zone listed according to the faunal districts in which they occur—Continued.

 

 

Northeastern section (Colorado Desert district). Middle section (Vizcaino Desert district). Southern section (cape district).
Urocyon cinereoargenteus scotti. Urocyon cinereoargenteus californicus.
Vulpes macrotis macrotis. Vulpes macrotis devia. Vulpes macrotis devia.

Bassariscus astutus palmarius.
Bassariscus astutus insulicola (San

Jose Island).
Bassariscus astutus saxicola (Espiritu

Santo Island).
Procyon psora pallidus. Procyon psora..
Taxidea taxus infusca. Taxidea taxus infusca. Taxidea taxus in, ea
Spilogale microdo
Spilogale leer.

‘ ormoops megalophylla senicula.
Chaeronycteris mexicana. Chaeronycteris mexicana.
Mephitis mesomelas estor.
Myotis yumanensis. Natalus mexicanus. ee volans.
Mrotis peninsularis.
otis micronyx.

Pipisirelius hesperus australis.
Eptesicus fuscus peninsulae.
Antrozous pallidus pallidus. Antrozous pallidus minor. Antrozous pallidus minor.
Nyctinomus depressus.

 

 

 

Upper Sonoran Zone. /

The Upper-Sonoran zone in Lower California is broken into numerous areas of varying size
distributed from the northern border south nearly to Cape San Lucas. The largest area is that
described as the San Diegan district in the account of the subordinate faunal areas (p. 118).
This is the southern extension of the coastal belt from southern California. It occupies the
northwestern part of Lower California from the Pacific coast up to an altitude of about 4,000 to
4,500 feet on the west slope of the Juarez Mountains and to 6,000 to 6,500 feet on the same
slope of the San Pedro Martir Mountains. It extends from the northern. boundary south
to near El Rosario, narrowing rapidly south of San Quintin. The Upper Sonoran zone limits
are still undetermined on the east slope of the Juarez-San Pedro Martir Range, but the lower
border must be between 4,000 and 4,500 feet. The higher mountain tops rising above an altitude
of about 4,500 feet from the southern end of the San Pedro Martir Mountains to the southern
end of the ‘peninsula are Upper Sonoran.

The Coronados, ‘Todos Santos, San Martin, San Geronimo, the San Benitos, and Natividad
Islands belong to this zone. It also covers the northern and western slopes of Cedros Island
except the upper part of the high northwest slope, which is probably Transition; the arid eastern
and southern slopes are Lower Sonoran. Guadalupe Island, still farther offshore, surrounded
by cold ocean currents, appears to be Upper Sonoran on its lower slopes, Transition in the middle
parts, and approaching Canadian on the northerly slopes of its cold foggy summit. No detailed
work has been done, however, to outline the zones on Guadalupe, so they can be stated only in a
a general way.

From the San Pedro Martir Mountaims to La Paz Bay the higher elevations occur only
as small solitary peaks, so the areas of Upper Sonoran are extremely limited, resulting merely in a
series of scattered and widely spaced barren islets of this zone with but slight traces of the flora
and fauna proper to the climatic conditions. One of the best marked of these Upper Sonoran
islands is on the summit of the high mountain south of the old mission of Calamahue, where
we saw a thin growth of small oaks or pinyon pines, but were not near enough to identify them
further. Immediately north of La Paz Bay Mechudo Peak forms one of these islets. South of
La Paz Bay the tops of the Sierra Laguna and other parts of the Sierra Victoria, above about
4,500 feet, is well-marked Upper Sonoran. This area is 30 miles or more in extent and is cov-
ered by a growth of scrubby oaks, with pinyon pines (Pinus cembroides) on the high parts of the
Sierra Laguna.

Close along the seashore in the San Diegan district the slopes exposed to the salt air are
practically without shrubby growth, the surface being overgrown with grasses and other her-
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 129

baceous plants. Farther back on the mountain slopes and borders of the interior valleys there is
commonly an abundant shrubby growth, often forming dense chaparral, especially in canyons
like that on the mountain slopes in San Diego County, Calif.

Along the slopes of the Juarez-San Pedro Martir Mountains there is a definite belt of pinyon
pines all within the Upper Sonoran life zone. Botanists have identified three species among
these pinyons, but their relative distribution is little known. The Parry pinyon (Pinus quad-
rifolia) is the most abundant and widely distributed of the species. It grows on both slopes of
the mountains and in places spreads over broad areas, especially along the northwestern slopes of
the San Pedro Martir Mountains and thence over outlying foothills; also on the southwestern
slopes of the Sierra Juarez and the outlying hills of the Alamo Plain. On the west slope of all
these mountains the pinyons extend from about 4,000 feet up into the border of the Jeffrey
yellow pines at from 4,800 to 6,000 feet. Their lower limit is rather lower on the west slopes
than on the arid eastern side of the range. On hot sheltered south slopes at 9,000 feet near the
head of Canyon Diablo, in the San Pedro Martir Mountains, we saw small stunted pinyons, but
their ordinary limits are far below this. They extend north within a few miles of the California
border and south along the Rosario divide to the tops of the mesas about Matomi Canyon, the
extreme southern limit of the pines in this section.

The birds and mammals of the San Diegan district in Lower California are the same or are
closely related to those inhabiting the coastal belt of the San Diego district north of the border.
In the Upper Sonoran part of the cape district the birds are nearly all different, mainly sub-
specifically, from those in the San Diegan part of this zone and number among them the most
striking and characteristic members of the bird fauna of what has been known as the “Cape
Region.” The mammals also differ but are much less numerous.

The subjoined lists of mammals, plants, reptiles and birds of the two main Upper Sonoran
areas of the peninsula show the characteristics of each. Additional notes on each of these Upper
Sonoran areas are given in the descriptions of the subordinate faunal areas on pages 117-121.

Puants CHARACTERIZING THE Urrer Sonoran ZoNnE IN NortTHWwESTERN Lower Catirornia (SAN DieGan
District).

[Species marked with an asterisk (*) occur also in California. ]

* Juniperus californica.
*Pinus quadrifolia.
*Pinus monophylla.

*Pinus monophylla edulis.

Ephedra californica.
* Hesperoyucca whipplei.
Nolina palmeri.
*Quercus agrifolia.
*Quercus dumosa.
*Quercus tomentella.
Quercus palmeri.

*Eriogonum fasciculatum.

Eriogonum polifolium.
*Berberis fremontii.

Grossularia quercetorum.
*Ribes indecorum.
*Platanus racemosa.

 

* Adenostoma fasciculatum.

* Adenostoma sparsifolium.

*Cercocarpus parvifolius.

* Heteromeles arbutifolia.
Rosa minutifolia.

*Emplectocladus fasciculatus.

*Prunus ilicifolia.

*Cneoridium dumosum.
Schmaltzia ribifolia.
Aesculus parryi.

*Adolphia infesta.
Ceanothus goldmani.
Ceanothus submontanus.
Ceanothus palmeri.

*Rhamnus crocea.

* Fremontodendron californicum.
Bergerocactus emoryi.

 

*Garrya veatchii palmert,
Arctostaphylos bicolor.

* Arctostaphylos glauca,
Arctostaphylos oppositifolia.

*Fraxinus attenuata.

*Eriodictyon angustifolium.

*Eriodictyon sessilifolium.

*Ramona polystachya.
Diplacus arachnoideus.
Diplacus puniceus.
Pentstemon centranthifolius.

*Pentstemon antirrhinoides.

*Lonicera interrupta.

*Sambucus glauca.

* Artemisia californica.

* Artemisia tridentata.

* Coleosanthus atractyloides.

J
PLants CHARACTERIZING THE Upper Sonoran ZONE IN SOUTHERN Lower Cauirornia (Cape District).

*Pinus cembroides.
Nolina beldingi.
Populus monticola

*Salix lasiolepis.
Quercus devia.

Thalictrum peninsulare.
Ribes brandeget.
Heteromeles arbutifolia.
Rubus sp.

Garrya salicifolia.

 

[Species marked with an asterisk (*) occur also on the Mexican mainland.]

Arbutus peninsularis.
Salvia similis.
Tagetes lacera.

Reptiles of Upper Sonoran Zone, San Diegan District.—The following reptiles are char-
acteristic of the San Diegan district of Upper and Lower California. Too little is known of
130 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

the distribution of the reptiles of this zone in the cape district to warrant trying to separate
the zonal elements there. The reptiles of the cape district as a whole are enumerated in the
general discussion of the reptiles (p. 115).

Uta stansburiana hesperis. Phrynosoma cerroense. Pituophis catenifer annectens.
Sceloporus occidentalis bi-seriatus. Cnemidophorus labialis. Thamnophis ordinoides hammondii.
Sceloporus orcuttt. Lichanura roseofusca.

Birds of Upper Sonoran Zone—The Upper Sonoranzone of the peninsula is isolated in
two main sections, the first or most northern in the extreme northwest adjoining California,
and the second confined to the mountain slopes near the southern end mainly south of La Paz.
For purposes of comparison the birds of the two areas are listed separately.

Breps CHARACTERISTIC OF THE Upper SoNoRAN ZONE.

NORTHWESTERN SECTION (SAN DIEGAN DISTRICT).

 

Rallus levipes. Passerculus rostratus halophilus.
Gymnogyps californianus. Amphispiza belli.
Balanosphyra formicivora bairdt. Aimophila ruficeps ruficeps.
Dryobates nuttallit. Melospiza melodia coopert.
Phalaenoptilus nuttallit californicus. Pipilo fuscus senicula.
Archilochus alexandri. Passerina amoena.
Tyrannus verticalis. Lantus ludovicianus gambeli.
Otocoris alpestris actia. Vireo bellit pusillus.
Aphelocoma californica obscura. Vireo vicinior.
Cyanocephalus cyanocephalus. - Toxostoma rediviva helva.
Sturnella neglecta neglecta. Toxostoma crissale.
Agelatus phoeniceus neutralis. Thryomanes bewickii charienturus.
Carpodacus mexicanus clementis. Baeolophus inornatus murinus.
Astragalinus lawrencet. Psaltriparus minimus minimus.
Astragalinus psaltria hesperophilus. Chamaea fasciata henshawt.
Passerculus beldingi. Polioptila californica.
SOUTHERN SECTION (CAPE DISTRICT SOUTH OF LA PAZ).
Chloroenas fasciata vioscae. Pipilo maculatus magnirostris.
Glaucidium gnoma hoskinsii. Tachycineta thalassina brachyptera.
Otus asio xantusi. Lanivireo solitarius lucasanus.
Balanosphyra formicivora angustifrons. Vireo huttoni cognatus.
Phalaenoptilus nuttallii nuttalli. Sitta carolinensis lagunae.
Basilinna xantusi. Baeolophus inornatus cineraceus.
Junco bairdi. Psaltriparus minimus grindae.
Aimophila ruficeps sororia. Planesticus confinis.

 

MamMas CHARACTERISTIC OF THE UPPER SONORAN ZONE.

NORTHWESTERN SECTION (SAN DIEGAN DISTRICT).

Odocotleus cerrosensis (Cedros Island).” Microtus californicus huperuthrus.
Citellus beecheyi beecheyt. Thomomys bottae nigricans.
Citellus beecheyi fishert. Dipodomys merriami parvus.
Onychomys torridus ramona. Perodipus simulans.
Peromyscus maniculatus gambeli. Perognathus pacificus.
Peromyscus maniculatus dubius (Todos Santos Island). Perognathus penicillatus helleri.
Peromyscus maniculatus geronimensis (San Geronimo | Perognathus fallax fallax.

Island). Perognathus anthonyi (Cedros Island).

Perognathus femoralis femoralis.
Perognathus femoralis mesopolius.
Lepus californicus bennetti.

Peromyscus maniculatus cineritius (San Roque Island).
Peromyscus californicus insignis.
Peromyscus eremicus fraterculus.

Peromyscus eremicus cedrosensis (Cedros Island). Sylvilagus audubont sanctidiegi.
Reithrodontomys megalotis longicaudus. Sylvilagus bachmani cinerascens.
Reithrodontomys megalotis peninsulae. Sylvilagus bachmani cerrosensis (Cedros Island),
Neotoma intermedia intermedia. Urocyon cinereoargenteus californicus,

Neotoma fuscipes macrotis. Procyon psora,

Neotoma bryanti (Cedros Island). Taxidea taxus californica,

Neotoma anthonyi (Todos Santos Island). Mephitis occidentalis holzneri.

 

Neotoma martinensis (San Martin Island). Sorex californicus juncensis.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 131

SOUTHERN SECTION (CAPE DISTRICT).

Ovis canadensis cremnobates, Thomomys bottae alticola.
Peromyscus truei lagunae. Sorex lagunae.

Transition Zone (San Pedro Martir District).

In Lower California this zone is confined to an isolated belt, nowhere exceeding 12 or 15
miles in width and perhaps 150 miles in length, along the upper slopes of the Juarez-San Pedro
Martir Mountains. It is limited on the north at a point within 15 or 20 miles of the California
boundary, where the general elevation of the Sierra Juarez is depressed and the Upper Sonoran
zone comes in. This lower belt extends to about an equal distance north of the boundary, so
the gap between the nearest definite Transition zone areas on the two sides of the border appears
to be at least 30 to 40 miles. The lower limit of the zone on the west side of the Sierra Juarez
is about 4,800 feet and a little higher on the east slope. On the west slope of the San Pedro
Martir Mountains the lower limit is from 6,000 to 6,500 feet and about 6,500 to 7,000 feet on
the east slope. It extends thence up to the top of the main ridges at over 9,000 feet. Above
this the only elevations are the bare ragged granite crests leading to the top of La Providencia
Peak, which are too limited in size and barren of vegetation to show zonal characteristics.

This Transition zone area is characterized by a scattered forest of yellow pines (Pinus pon-
derosa jeffrey) with a thin mixture of sugar pines (Pinus lambertiana), the incense cedar (Libo-
cedrus), and other species of plants also characteristic of this zone in the mountains of southern
California, including a mixture of Pinus coulteri among the Jeffrey yellow pines on the Sierra
Juarez. (See Pl. 26, Fig. 1.) The animal life here is also very similar to that of the more
northern mountains, but a few local subspecies have developed in this isolated tract, as shown
by the subjoined lists.

Sceloporus graciosus vandenburghianus is the only reptile I can locate as belonging to this
zonal area, although Gerrhonotus scincicauda webbvi is also found there and in adjacent parts
of the Upper Sonoran zone.

Intermingled with the Transition zone species on the top of the San Pedro Martir Mountains
are a few species of birds and plants belonging to the Canadian zone. Owing to the fact that
these species form only a subordinate element in the upper part of the Transition area it is
evident that they represent merely the usual overlapping along the border and no part of these
mountains can be mapped as belonging to the Canadian zone. The possible exception to this
statement is that of the bare granite crest of Santa Catalina Peak, rising over 10,000 feet above
sea level, but its area is insignificant and it appears almost wholly devoid of vegetation. The
representatives of the Canadian zone noted on these mountains are listed beyond.

PLANTS CHARACTERIZING THE TRANSITION ZONE (San Pepro Marrir District).

[Species marked with an esterisk (*) occur also in California.)

Cupressus guadalupensis (?). *Quercus chrysolepis. Arctostaphylos pringlet.
*Tibocedrus decurrens. *Philadelphus pumilus. *Ramona incana pachystachya.
*Pinus ponderosa jeffreyt. * Holodiscus dumosus. * Monardella linoides.

*Pinus coultert. *Rosa californica. * Monardella macrantha.
*Pinus lambertiana. *Ceanothus divaricatus. Pentstemon linarioides.
* Populus trichocarpa. *Rhamnus californica.

Brrps CHARACTERISTIC OF THE TRANSITION ZonE (SAN PepRO Martie Disrricr).

Oreortyx picta confinis. Pipilo maculatus megalonyz. Troglodytes aedon parkmantt.
Colaptes cafer collaris. Piranga ludoviciana. Sitta carolinensis aculeata.
Empidonaz dificilis cineritius. Progne subis hesperia. Sitta pygmaca leuconucha.
Cyanocitia stelleri frontalis. Tachycineta thalassina lepida. Penthestes gambeli batleyae.
Euphagus cyanocephalus. Lanivireo solitarius cassinit. Sialia mexicana anabelae.

 

 

Junco oreganus townsendi. Cinclus mexicanus unicolor.
132 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

Mammats CHARACTERISTIC OF THE Transition Zone (SAN PEDRO Martie District).

Odocoileus hemionus californicus. "Peromyscus boylei rowleyi. Sorex ornatus.

Sciurus douglasit mearnsi. Neotoma intermedia intermedia. Scapanus latimanus anthonyt.
Eutamias merriami obscurus. Neotoma fuscipes macrotis. Myotis orinomus.

Citellus beecheyi fisheri. Canis clepticus. Myotis evotis milleri.
Peromyscus truet martirensis. Spilogale arizonae martirensis. Eptesicus fuscus fuscus.

Peromyscus maniculatus sonoriensis.

Canadian Zone,

As already noted a small infusion of Canadian zone species of plants and animals occurs in
the higher and colder parts of the Transition zone on the San Pedro Martir Mountains.

The only mammal of Canadian zone affiliations is Sciurus douglasi mearnsi, which here, in
common with the birds named below, ranges well through the Transition zone.

The birds belonging i in this zone are:

Nucifraga eoimnbiata (rare). | Spinus pinus. | Carpodacus cassini.
The most notable plants belonging here are:

Abies concolor. Populus tremuloides. Symphoricar pos parishit.
Pinus contorta. ,

All of these are well-known species in the mountains of California.

NATURAL RESOURCES IN ANIMAL AND PLANT LIFE.

Land Game Mammals, a

- Early in its history Lower California abounded in game. Mule deer were abundant through-
out the peninsula and on some of the adjacent islands. Antelope roamed over the desert plains
from the border of California to beyond Magdalena Bay. Mountain sheep occupied the barren
desert ranges mainly in the middle and eastern parts of the peninsula from the northern border
south to the northern side of La Paz Bay.

Coyotes, mountain lions, and bobcats were numerous and are still common in places. I
have been unable, however, to find any definite record of wolves ever having occurred in the
peninsula, and conditions there do not appear to be favorable for them. The only record of
bears is that of the grizzly, mentioned as occurring in the Sierra Juarez by Pattie.

It is of interest to note that the first mountain sheep reported from America by Europeans
were those described by the Spanish missionaries in Lower California early in the occupation of
the region. These sheep, although 1 the first to become known to Europeans, were among the
latest on this continent to receive a scientific designation, having been described in 1903 from
specimens obtained ‘at Matomi as Ovis cervina cremnobates. The mountain sheep of Lower
California average smaller than those of the northern Rocky Mountain region. They wander
from sea level on the shores of the gulf to an altitude of less than 5,000 feet on the slopes of
the mountains, and occur mainly at elevations between 1,000 and 3,000 feet. They are strictly
limited to elevations below the lower limit of pine forests. Although far less numerous than
formerly and steadily diminishing in numbers, mountain sheep still occur in many of the moun-
tain ranges from near the California border south almost to La Paz Bay.

Unfortunately, the difficulties of transportation and the scarcity of live stock through
most parts of the peninsula render the securing of food supplies so difficult that great numbers
of game animals, especially mountain sheep, have been killed and the meat dried, or jerked, to
supply mining camps and other communities. This deplorable and wasteful slaughter still
continues and unless checked will ultimately result in the extermination of the sheep and all
other large game animals of that region. Mountain sheep are peculiarly endangered by this
slaughter owing to their habit-of going in bands and drinking at certain watering places, where
hunters lie in wait behind blinds built of loose stones and kill individuals of all ages and sexes,
often in a few minutes destroying an entire band. I have been informed of one party having
killed more than 100 sheep in this manner to make dried meat during a single season.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 133

On the mainland of the peninsula occur two, and perhaps three, forms of the mule deer,
none of them strikingly different from the typical animal of the Rocky Mountains, except that
in the south the horns appear to average smaller. The mule deer of Cedros Island differs strik-
ingly from the typical animal in its dwarfed antlers. Those on San Jose Island, in the gulf,
so far as known, are the same as those on the adjacent mainland of the peninsula. About 20
years ago mule deer were abundant about the delta of. the Colorado, occurring there during
the winter season in bands. At present they are almost exterminated in this area. Elsewhere
in the peninsula, although in many places reduced below their former numbers, they are still
more or less generally distributed from sea level up to 9,000 feet on the higher parts of the San
Pedro Martir Mountains, and to 6,000 feet on the Sierra Victoria in the south. They are still
numerous on the upper slopes of both these ranges and on parts of the Magdalena Plain.

Antelopes, once so numerous on the open plains of the peninsula, have become exterminated
over considerable areas. None appear to be left in the northwestern part of the peninsula
west of the mountains, and north of El Rosario. They are also gone from the great Magdalena
Plain. Small numbers still exist on the desert south of the Cocopah Mountains and back of
San Felipe Bay, as well as in various other localities, thence south to the Vizcaino Desert, where
they are still more numerous than elsewhere. It is evident, however, that these animals are

‘fated to disappear from this region in the not distant future.

In addition to the larger game of the peninsula, jack rabbits and cottontails are common
practically throughout its extent, varying in abundance according to the seasons, whether
these are favorable or unfavorable for the production of abundant food. The jack rabbits and
cottontails of the entire peninsula and the brush rabbits found in the northwestern portion are

all geographic races of the species occurring in California.

So far as known, the only game law ever issued for the protection of land game in Lower
California is one promulgated for the protection of mountain sheep, of which the translation is
given below. Unfortunately, no effective enforcement of this regulation appears to have been
undertaken.

Unrrep States or Mexico.

GOVERNMENT OF THE NORTHERN DISTRICT OF LOWER CALIFORNIA.

 

Notice.
By order of the superior authority and until futher notice the hunting or exploitation of mountain sheep (Ovis
eremnobates) in this district is strictly prohibited.
This is for the information of the authorities of the district and the public in general for its observance, with the
understanding that infractions of this order will be punished according to law.

' Mexicali, Baja California, October 10, 1917.
ios : E. Cantu,

Governor of the District.
J. BARRERA,
Secretary of the Government.
Sea Game Mammals.

From the time of its discovery until the early part of the last century the coasts of Lower
California were frequented by enormous numbers of whales, which resorted to the numerous
bays and inlets to breed. Among the more notable of these were the sperm and California
gray whales. Fur seals, elephant seals, sea otters, sea lions, and spotted, or harbor, seals also
abounded. ;

During the first half of the last century whalers, sealers, and sea otter hunters united in
such uncontrolled slaughter of these valuable animals that their numbers rapidly decreased

_to a point where their pursuit came to an end for lack of an object. The seals, sea otter, and
sea elephants have disappeared forever as an economic factor. There is no record ofa single
imdividual of fur seal or sea otter having been obtained for many years within the waters of
the peninsula, and sea elephants have disappeared except for a small number which were dis-
covered in 1911 haunting the westward shores of Guadalupe Island. The sea otter, sea ele-
134 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

phant, and fur seal ranged mainly along the northwestern coast of the peninsula from the
vicinity of Ensenada to Cedros Island. Sea lions, and to a lesser extent harbor seals, have
been killed on a smaller scale along both shores of the peninsula during the past and up to
the present time. In the Gulf of California they are hunted in small fishing boats, and on the
islands of San Pedro Martir, Angel de la Guarda, and San Lorenzo. It is recorded that in
1884 hunters from Mulege killed 287 sea lions. The hunting of these animals for their skins
and oil is unorganized and sporadic. Their numbers are too small to warrant systematic
exploitation.

Game Birds. ‘

Land game birds are poorly represented in the peninsula. No species of grouse occurs.
There are, however, Gambel’s quail in the northeastern section and California quail in the
northwest, and thence south throughout most of the peninsula wherever permanent water
occurs. The mountain quail is plentiful in the wooded and brushy parts of the higher eleva-
tions of the Juarez and San Pedro Martir Mountains (up to above 8,500 feet on the latter range).
In addition, a local form of the band-tailed pigeon is numerous in the limited wooded area
of the Victoria Mountains in the southern end of the peninsula. Mourning doves are common
everywhere in the vicinity of permanent water.

Many species of ducks and several species of geese winter in great numbers in the delta of
the Colorado, and in the San Quintin Bay and Scammon Lagoon. These two bays are the
main winter homes of the black brant. Ducks are rather common wherever sufficient perma-
nent fresh water exists in the southern part of the peninsula, but never in any very great numbers.
Fisheries.

The waters about both coasts of Lower California abound in food and game fishes to such
an extent that eventually they must become enormously valuable as a source of supply to
the adjacent regions. Dr. C. H. Townsend, who conducted fishery investigations in the Gulf
of California during the expedition of the Fish Commission steamer Albatross in 1911, states
that during his work he found numerous excellent food fishes readily obtainable with hand
lines and seines at most of the anchorages. He found that several food fishes of the Atlantic
coast occur in the Gulf of California, among which may be mentioned the Spanish mackerel,
chub, mackerel, mullet, jack, crevally, and jewfish. In addition were many native fishes of
importance. One of these, known locally as the sea bass, is the largest of the weak fishes,
reaching a weight of 175 pounds, and is a valuable food fish. It is abundant at the head
of the gulf and enters the mouth of the Colorado, where it is taken by spearing. Another
weak fish called “blue fish” in southern California is abundant. The excellent ‘‘ pez de gallo”
is abundant in the upper parts of the gulf and reaches a length of 4 feet. Among the more or
less common species are the barracuda and yellow-tailed grouper, which are of large size and
are plentiful at the mouth of the Colorado, running up the river as far as there is brackish water.
Dr. Townsend says that a few California yachtsmen have found their way into the upper parts
of the gulf and have enjoyed the best of sport with rod and reel.

A species of trout (Salmo nelsoni) related to the rainbow trout of California was dis-
covered by us in a single stream, the Arroyo San Antonio, on the west slope of the San Pedro
Martir Mountains. This stream forms the main headwater tributary on the Rio Santo Domingo.
These trout are all below the waterfall in the Arroyo San Antonio and probably occupy less
than 15 miles of its length. They do not attain a large size and are in such limited numbers
that it would not be difficult to destroy them by overfishing. Their numbers are sometimes
much reduced by floods down the canyon which fill the stream bed with moving rocks, sand,
and gravel.

Along the northwestern shore of the peninsula, especially in the vicinity of San Quintin
Bay, occur great numbers of the large crayfish, which replaces the lobster on the Pacific coast.
A cannery was once established near the mouth of San Quintin Bay to pack these crustaceans.
Although the crayfish, were found in sufficient abundance, yet certain local difficulties resulted
in the abandonment of the enterprise and the establishment was idle in 1905.. During the
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 135

season of 1918 and 1919 a cannery for packing sardines and tuna has been in operation at San
Quintin. In addition, several fishing companies of San Pedro and San Diego, Calif., have
been operating fishing boats along the northwest coasts of the peninsula, taking tuna, mackerel,
crayfish, abalone, and other sea products for the California market.

Green sea turtles are plentiful about the southern coast, especially from Cape San Eugenio
south to beyond Santa Margarita Island and also along most of the gulf shore. Many turtles
have been taken in the vicinity of Magdalena Bay and shipped to San Francisco on Pacific mail
steamers. At one time a small canning factory was established on the east shore of Santa Mar-
garita Island for the purpose of putting up turtle meat, but it is reported that through the inex-
perience of those in charge the experiment proved a failure.

Pearl Oysters.

Within a short time after the occupation of the peninsula by the Spaniards it was discovered
that pearl oysters were plentiful along the southern half of the gulf shore of the peninsula, and
from that time until the present with short intervals an annual harvest of pearls and pearl shell
has been obtained. Owing to overfishing the pearl oyster beds in many places were destroyed and
the yield throughout has decreased. Early in 1906, when we were at La Paz, we found a local
company Was experimenting in farming pearl oysters on Espiritu Island, at the mouth of La
Paz Bay. It has since been learned that this experiment has been successful, indicating that
Lower California thus presents an opportunity for an attractive new industry. The pearl
oysters are reported to do best.in from 1 to 10 fathoms of water.

In addition to pearl oysters Dr. Townsend reports that edible oysters “‘not distinguishable
from those of our Atlantic States are abundant at several points in the Gulf of California as far
north as Georges Bay.”

Pine Forests.

In all of Lower California only two limited areas contain merchantable timber. Both of
these arein the north. The first, which is about 3 miles in width by 37 miles in length, lies along
the crest of the northern part of the Sierra Juarez, in the section known locally as the Sierra
del Pinal, or Laguna Hanson Mountains; the second, about 6 miles in width by 45 miles in
length, occupies the crest of the Sierra San Pedro Martir. These areas include all of the yellow
pine, sugar. pine, lodge-pole pine, incense cedar, and fir trees of the peninsula in a total area of
less than 400 square miles.

The greater part of the pine timber in the Juarez Mountains is readily accessible by exist-
ing Wagon roads, and several small mills have already cut much of the best growth for local use.

On the San Pedro Martir Mountains the timbered area is isolated on the elevated summit
and only accessible by trail leading up over several thousand feet of abrupt and rugged mountain
slopes. The top of the mountains where it occurs is much broken and forms numerous small
pocketlike valleys and mesas separated by altitudes varying from one to two thousand feet and
by canyons and excessively rocky slopes. The pine forest grows over the ridges and in many
small isolated areas or as narrow belts about the borders of valleys and along the bases of rocky
ridges, as shown in accompanying photographs. As a result of its isolation on the top of the

‘San Pedro Martir Mountains and the scattered distribution of the timber due to the broken
and rocky character of the mountain summit, the cost of utilizing the timber on these moun-
tains would be exceedingly high and would not be practicable at present. In both of the areas
mentioned the timber in places is fairly large, but the majority of it is scrubby. The forest
is commonly mixed in character but the yellow pines and the lodge-pole pines each occur in
many places in practically unmixed growths. The sugar pines, the incense cedars, and the firs
do not occur in any considerable numbers.

Orchilla.

A lichen belonging to the genus Roccella, commonly known as the orchilla, grows in abun-
dance on the desert shrubs and other plants occurring in a narrow belt along the Pacific coast
from the shores of Magdalena Bay north to ashort distance beyond Vizcaino Desert. (See Pl. 30,
Fig. 1.) This lichen hangs in long gray strands very similar in appearance to the Spanish moss
136 MEMOIRS NATIONAL ACADEMY OF SCIENCES. | [Vor. XVI

so plentiful in the Gulf States. It was formerly the basis of a considerable industry centered
at Magdalena Bay, being gathered in great quantities, baled, and shipped to England and Ger-
many for dyeing purposes. Large warehouses were erected at Magdalena Bay and the Chartered
Co. of Lower California became a wealthy corporation in exploiting this product. With the
discovery of aniline dyes, and their cheap production, the orchilla industry was abandoned,
and commercial activity at Magdalena Bay ceased.

Tanbark.

One of the most important trees of the peninsula from an economic point a view is the
palo blanco (Lysiloma candida), a tropical species which occurs abundantly in many places i in
the cape region from Santa Ana Canyon, in the lava plateau north of San Ignacio south to Cape
San Lucas. (See PI.33.) The gathering and exportation of its bark for tanning purposes is one
of the leading industries of the district where it occurs. The main shipping point for this
product is San Jose del Cabo.

Dye Bark.

The bark of the torote (Bursera microphylla) is still gathered and shipped from the cape
district for dyeing purposes.
Kelp.

Kelp beds similar to those off the coast of southern California occur along the west coast
of Lower California from near the northern border south to the vicinity of Cedros Island. In
1918, following the development of the kelp industry on the coast of California, more than
$100,000 was invested under Gov. Cantt’s direction in the development of a plant at Sausal,
6 miles northwest of Ensenada, for treating kelp to produce potash and other products. The
plant established by Gov. Cantd, including boats, kelp-harvesting machinery, and other appa-
ratus, Was in readiness for gathering and treating between 200 and 300 tons of kelp a day when
the armistice which brought the Great War to a close ended the exigency need of potash and
caused the abandonment of this enterprise.

Wild Fruits.
The wild fruits of Lower California were of much importance to the original inhabitants
~ but have little to recommend them now. They include mainly the fruits of several species of
cactus, including the pitahaya agria and pitahaya dulce (Lamaireocereus.thurberit and L. gum-
mosus); also the ciruela (Tapirira edulis) of the south, a small yellow plum of peculiar taste
which is a relative of the pistachio nut. Pinyon nuts are still gathered as food by some of the
few surviving Indians, especially along the western base of the San Pedro Martir Mountains,
and mesquite beans are gathered by them in many localities.

AGRICULTURE AND STOCK RAISING.
Agriculture.

The soil of Lower California is fertile and yields abundant crops wherever water is available
for irrigation. In addition, the temperatures over most of the peninsula.are so mild with
severe frosts lacking that conditions favor a large variety of temperate, tropical, and sub-
tropical fruits and other crops. The prevailing desert condition, however, combined with the
broken and mountainous nature of the country, will always form effective barriers against the
agricultural development of more than a small part of the total land area. Millions of acres
are covered with an extraordinary growth of desert vegetation which, through a long process
of adaptation, is able to survive where agricultural crop growing is an impossibility.

The most important fact to be borne in mind concerning the development of agriculture
and grazing in Lower California is that, with the exception of the high mountain tops of the
southern and northern ends of the peninsula, practically all parts of that region are subjected
from time to time to dry periods of from one to more than five years in extent, during which
little or absolutely no rain falls. This necessarily limits successful agriculture to areas which
have a permanent water supply, or where such supply can be developed or stored to carry
cultivation through successive dryseasons. Another serious drawback to agricultural develop-
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 33.

 

 

 

 

Fic. 1.—Palo blanco forest (Lysiloma candida) in valley above San Ignacio, on trail to Santa Rosalia.

 

 

 

 

 

 

 

Fic. 2.—Bark of palo blanco in sacks ready for shipment on coasting vessels at Cape San Lucas village in January, 1906. This bark is used
for tanning.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 137

ment in most of the peninsula is the lack of roads and the absence of satisfactory boat service
along the coast, thus rendering the marketing of products both costly and uncertain.

Lower California is about 800 miles long and from 30 to more than 100 miles in width.
The lands available for agriculture are mainly located in the northern and southern thirds of
the peninsula. They are in isolated areas, varying from insignificant patches of 1 or 2 acres
below springs, or in the bottoms of canyons to little valleys and open plains bordering the
seacoast. The plains, as has already been described, are mainly located along the Pacific
coast. The largest single body of agricultural land, however, is that in the delta region of the
Colorado, in the extreme northeastern section of the peninsula, and contains approximately
1,500,000 acres which may be irrigated from the Colorado.

The agricultural lands of Lower California may be roughly divided into four main areas
according to the origin of the soil. The rich plains in the delta region of the Colorado have
been formed by deposits of silt brought down by the Colorado River. The agricultural lands
of the northwestern part of the peninsula, and those south of La Paz Bay have been built up
mainly from the disintegrated gray granite of which the mountains in those districts are
mainly composed. The soil of the great Magdalena Plain is the result of the erosion and
disintegration of the volcanic cap and the sandstone base forming the great lava-covered plateau
of that region.

Until recent years the agricultural development of the peninsula has been on a small scale
and limited mainly to the west side of the mountains in the northwestern part of the peninsula,
and to areas south of La Paz Bay. Following the development of the Imperial Valley in
California American capital has been successfully promoting agricultural development on a
large scale in the rich lands of the delta region of Lower California tributary to the town of
Mexicali.

Probably more than 1,000,000 acres of land are available for irrigation in this district
fully equal in quality to those of the Imperial Valley. A large part of it is owned by American
companies and about 100,000 acres are under cultivation, producing long-fiber cotton, melons,
alfalfa, barley, and lesser crops. The crops reported in this district for 1918 include 53,271
bales of cotton, more than 13,500 tons of cotton seed, 25,000 tons of alfalfa, and 15,000 bushels
of barley. Experiments in growing rice and sugar cane in this district have been successful,
indicating probable future development in these products. The main irrigating canals feeding
the Imperial Valley in California are taken out of the Colorado in Lower California territory
to supply both areas.

Tn 1918 an American company leased from the local authorities a large tract of land about
Laguna Maquata, on the west side of the Cocopah Mountains. A pumping station was estab-
lished on the Hardy River and irrigating ditches carried around to this land. Unfortunately,
the first planting experiments, which were on a large scale, proved to be on soil so saline that
no crops were produced.

A branch of the Southern Pacific Railroad extends from Mexicali, on the international
boundary, easterly through the northern part of the delta district, swinging back across the
border to join the main line of the railroad west of Yuma, Ariz. Another branch connects this
branch road with Voleano Lake. The continued growth and future prosperity of this area is
assured. -

Another agricultural district, in which the fertile areas are scattered, broken, and far less
rich in possibilities than those of the delta region, is that lying west of the Sierra Juarez and
San Pedro Martir Mountains, extending from the border of California south along the Pacific
coast to beyond San Quintin. This area has three supply points: first, in the north, at Tijuana,
on the California border; second, at Ensenada, capital of the northern district; and the third
at San Quintin. Within this area are numerous small valleys and plains, including the coastal
plains known as the Llano de Hamalu and Llano de San Quintin, which lie along the coast
north and south of San Quintin Bay. A considerable total of good lands are available in this
district, but no large tracts in any one place except on the two coastal plains mentioned above.
More rain falls here than elsewhere in the lowlands of the peninsula, and the numerous small
138 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

streams at the bases of the mountains with their underground flow might be made available
through storage and pumping to develop a region similar in character to that across the border
tributary to San Diego.

The third area is the great Magdalena Plain extending along the Pacific coast north and
south from Magdalena Bay. The Llano de Yrais in this area was once the site of a disastrous
attempt to establish an American colony and met the fate certain to overtake any ill-judged
effort to settle this region without the proper preliminary work. This plain must be developed
by means of wells and storage reservoirs to save flood waters from the mountain slopes to the
east. Its capacity for production is indicated by the list of crops grown successfully by means
of a small well and pump at Matancita (p. 140).

In addition to the foregoing large coastal areas of good land, there are a number of smaller
ones where more or less water is available, some of which are now being utilized, as at Ensenada,
Guadalupe, and Santo Domingo Valleys in the northwest and Todos Santos, San Jose Valley,
the shore of Ventana Bay, La Paz, Loreto, Mulege, and the adjacent plain of San Bruno. Aside
from these there are a number of scattered areas throughout the interior, often containing an
acre or two and rarely exceeding 1,000 acres in extent. These interior areas number among
them the most beautiful occupied oases in the peninsula, such as San Ignacio, La Purisima, and
Comondu. Many of the interior locations have been occupied but abandoned owing to drought
and the lack of properly developed water resources. The roughness of the country about many
of these locations and their isolation are difficulties to be met in their development.

With the development of mining in this region a home market for produce will be formed.
Most of the locations, however, can not be developed successfully without some capital for
preliminary expenditure.

Further details on the location and extent of the main bodies of agricultural land in the
peninsula are given in the accompanying descriptions of the plains and larger valleys. The
notes on the ‘‘Streams and other Surface Waters,’”’ on ‘(Underground Waters” and on ‘Cli-
mate” give information concerning the water resources. The surface water supply now in use
is not developed and fully utilized, while at many places underground waters exist at moderate
depth, sometimes so near the surface as to be available for subirrigation. Many of the dry
stream beds leading down from the interior have water not far underground near the coast.
The large number of ruined and abandoned missions and ranches seen along our route was a
strong warning against inconsiderate settlement of such areas. It was also evident that without
sufficient capital for proper initial expenditure in developing water failure would be certain.

The Indians who occupied Lower California at the time of the discovery practiced no form
of agriculture, but subsisted on the natural products of the soil and surrounding waters. The
Spanish missionaries very soon began to introduce fruit trees and crops from the mainland of
Mexico and the Old World. These introductions included most of the kinds now cultivated in the
peninsula, With the exception of the Upper Sonoran district in the northwestern part, practi-
cally all of the agricultural land in Lower California is within the Arid Tropical and Lower Sono-
ran life zones. As a result the climatic conditions are nearly everywhere favorable for such irri-
gated crops as oranges, lemons, figs, olives, pomegranates, dates, grapes, wheat, barley, and alfalfa.
In the southern end, where tropical conditions prevail, coconut palms, bananas, pineapples, and
other more strictly tropical species are grown on a small scale.

Citrus fruits thrive wherever they have been tried, but we were informed that the Morelos
fruit fly had become established at San Jose del Cabo and perhaps elsewhere in the south. From
the northern border south on the west coast to San Quintin these fruits should do well in many

laces.
A few bananas are grown in the south, but the arid climate prevents their being cultivated
on a commercial scale.

Cotton grows well almost everywhere in the lowlands, and in the south, where frosts are
slight, this plant becomes a perennial. A large cotton shrub 3 years old was bearing abundantly

at Matancita in 1906.
No. 1} LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 139

The date palm was introduced by the Jesuit missionaries about 1730 and except in the north-
west continues to grow and fruit abundantly wherever tried with a sufficient water supply.
Several varieties of dates are grown, all seedlings with large stones. At San Ignacio date palms
are extremely numerous and are said to number from 50,000 to 60,000 trees. This is by far the
largest number in any one valley, but they are growing generally at the small towns and at many
ranches from San Ignacio south. The abundance of date palms at San Ignacio, La Purisima,
Comondu, and elsewhere gives an Oriental cast to these beautiful oases. (See Pl. 13, Fig. 2.)
There are thousands of acres of land suitable for the cultivation of dates where water can be
readily secured, and the long, hot, and arid summer of the peninsula offers ideal conditions for
some of the most delicate varieties of this fruit. Dried dates in rawhide bales are packed from
San Ignacio to the coast at Santa Rosalia and shipped across to Guaymas and other gulf ports.

Sugar cane is the main crop grown at San Jose del Cabo and at Todos Santos, where several
hundred acres of cane fields are cultivated. Brown sugar made into small cakes, called panocha,
is exported from these places to other parts of the peninsula. This crop is grown on a smaller
scale at other localities, expecially at Comondu, La Purisima, Mulege, and San Ignacio.

At La Purisima, Comondu, and some other places in this section an inferior quality of
raisins and some wine are made from the mission grape crop, and small quantities of both prod-
ucts are shipped to other localities.

The following crops, aside from cotton, are reported to have been produced in the northern dis-
trict of Lower California during the season of 1918:

Wheatssde sacs cwaen ietelg kilos.. 1, 800, 000 | Chick peas.......2..2....0.20.00e0eeee kilos. 2, 500
Batley scecceice cane ces sce sewe dss eats do...- 150,000} Chili peppers...........-.-.2-2-.--.---- do.... 1, 500
BSaNGsecsccsecdstes s828 stents? se acedOsess 865; 000 | Olive Olle. ecco ose scac tet ota cesnissieuce liters. . 1, 100
Corn.......02---e2 eee eee eee eee eee ee +--d0.... 160, 000 | Jojoba seed........-.-2-.2.22022-22--- kilos. . _ 3800
Pinyon Nits... isc. oe ieceeke sercisers do.... 3, 000 | Beeswax (mainly wild)...............-. do.... 7, 000
Al MONS es pccisteiore ire vas esters oy wie Seis do.... 23,000 | HONEY. ccsisiseneen deka eo yscnteee eee do.... 1, 600
Walnuts......-.- (cu Sae Meese e Ss swe do...- 1, 500

The discrepancy between the amount of beeswax and honey produced is due to the fact that
most of thelatter was consumed. The beeswax is obtained both from bees kept in domestication
and from those which have escaped from capitivity and occur throughout a large part of the inte-
rior of the peninsula from the northern border south to beyond Calamahue Mission. These wild
bees are found from near sea level up to above 8,000 feet on the San Pedro Martir Mountains.
On the deserts they live in cavities among the rocks and in certain desert plants, and in the
high mountains they occupy hollow trees. In 1905 we found a number of colonies attaching
their combs to the underside of the roofs of deserted houses at the base of the foothills in the
abandoned settlement on the San Quintin Plain. (See Pl. 34, Fig. 1.)

In January, 1918, as a result of a concession to colonize and exploit the resources of Gua-
dalupe Island, granted to Gen. Manuel Perez Trevifio, it was reported that 150 soldiers and
their women were located there. It having become evident that there was no arable land,
by July, 1919, only 32 soldiers with their followers remained. The island is occupied by large
numbers of goats, which for many years have run wild upon it and are reported to number
between 30,000 and 40,000. The soldiers remaining on the island are engaged in capturing
these animals and delivering them to contractors for 60 cents each. The better grades are being
kept and some high-grade Angora bucks have been placed on the island by Gen. Trevifio for
the purpose of improving the stock. The springs have been cleaned out, huts built for accom-
modating the people, and it is planned to try to develop the goat industry. In addition to selling
goats for the market in San Diego, others are being killed for their skins, tallow, and dried
meat, which are shipped to the Mexican mainland.

The following list of localities with their agricultural products shows most of the fruit
and crops now being cultivated between the northern border and Cape San Lucas:

La Huerta (east side of San Rafael Valley).—Peaches, apricots, beans, and corn.

Ensenada.—Grapes, peaches, apricots, almonds, walnuts, olives, oranges, lemons, pears,
apples, quinces, cherries, pomegranates, figs, wheat, barley, corn, beans, chick peas, potatoes,
and most of the ordinary garden vegetables.
140 MEMOIRS NATIONAL ACADEMY OF SCIENCES, [Von XVI.

Santo Domingo (Young’s ranch).—The productiveness of this part of the peninsula is
indicated by the fruits in bearing at the ranch of an American named Young, not far from the
Santo Domingo Mission, which include small quantities of the following kinds: Oranges, lemons,.
grapes, pears, apples, quinces, cherries, apricots, peaches, eine walnuts, figs, pomegranates,
and olives.

San Borja Mission. —Oranges, pomegranates, olives, dite mission grapes, and alfalfa.

San Ignacio.—Oranges, lemons, pomegranates, olives, black figs, dates (several varieties),
mission grapes, sugar cane, wheat, melons, beans, and mescal. ‘(About 1,000 acres cultivated.)

La Purisima- and Comondu.—Oranges, lemons, limes, olives, black figs, pomegranates,
dates, bananas, mission grapes, sugar cane, cotton, wheat, peas, and beans. (About 1,000
acres cultivated at Comondu and less at La Purisima.)

Loreto.—Dates, oranges, mescal, hides, and dried beef.

San Javier.—Olives, black figs, oranges, bananas, beans, sugar cane, and aguardiente or
canerum. (About 50 acres cultivated.)

Matancita.—Black figs, olives, pomegranates, pears, apples, bananas, grapes of several
varieties, cotton, corn, peas, beans, peppers, cabbage, tomatoes, potatoes, carrots, beets, squash,
watermelons, muskmelons, and the pulque agave.

San Jose del Cabo.—Black figs, dates, coconuts, coffee, oranges, lemons, limes, pome-
granates, aguacate (alligator pear), bananas, pineapples, tobacco, cotton, sugar cane, corn,
beans, and other small crops. (About 1,500 acres cultivated.)

Mulege.—Oranges, lemons, sweet limes, dates, olives, pomegranates, figs, mission grapes,
bananas, sugar cane, wheat, corn, beans, and others. (About 800 acres cultivated.)

Mexicali (delta of the Colorado). ee barley, Egyptian corn, melons, and cotton.

Stock Raising.

_ The grazing of both sheep and cattle is fairly successful in the northwestern section of the
peninsula from the border south to near San Quintin and from the coast up to the summit of
the Sierra Juarez and Sierra San Pedro Martir. In 1918 it was estimated that this area con-"
tained from 18,000 to 20,000 cattle. A limited number of cattle also range about the delta
of the Colorado. Elsewhere in the peninsula, as might be inferred from the recurrence of series
of rainless years, stock growing is a precarious occupation. During a succession of years in
which rains occur herds increase abundantly, only to meet disastrous losses through lack of
both food and water during the succeeding dry periods, which often continue from three to
five years. The Magdalena Bay Co. and others have some thousands of cattle on the Mag-
dalena Plain and the adjacent slopes of the lava plateau, and a more limited number of cattle
range in the region south of La Paz. (See Pl. 15, Fig. 1, and Pl. 34, Fig. 2.)

The principal merchant of La Paz, a man well past middle life, informed me that he had
always lived in that region and that among his friends were several owners of large cattle ranches
whose experience has shown that cattle growing on a large scale in that region was unsuccessful
from an economic point of view. He stated that the owners of these ranches during series
of favorable years would build up large herds only to lose them during following dry periods.
This he had known to occur repeatedly.

SCIENTIFIC EXPLORATIONS.

The first detailed information concerning Lower California was made known through the
explorations of the early. Spanish missionaries, mainly the Jesuits, who began their labors
immediately after the discovery of the peninsula by Cortez in 1535. Since then the main
contributions to our knowledge of the coast of the peninsula have come from the work of the
U. S. S. Narragansett and other American naval vessels, and our knowledge of the interior
from the work of American naturalists who have visited it within the last 60 years.

The explorations by American naturalists began when John Xantus landed at Cape San
Lucas in 1859, and have been mainly devoted to the study of the plants, reptiles, birds, and
mammals, A brief but not exhaustive summary is given below covering most of these expe-
Memoirs National Academy of Sciences XVI. First Memoir, Pl. 34.

 

 

 

 

 

Fig. 1.—Taking honey from bees’ nest in small cave near Matomi. The domesticated honey bee has become wild and occurs in the San
Pedro Martir Mountains as well as on the deserts far to the south. (Photograph by Dale Bumstead.)

 

 

 

 

 

Fic. 2.—Cowboy near Cape San Lucas. The cactus and other thorny plants are so numerous that riders in this region wear leather-covered hats
and leather jackets and trousers, and have heavy leather aprons hung on the saddle to cover their legs, as shown in photograph. In addition,
a heavy leather apron is hung over the horse’s breast to protect it from spines when rushing through jungle after half-wild cattle.
Noi] © LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 141

ditions. It is believed that this information will be useful to those studying the collections
from this region and will be of historic value. The information has been gathered from a variety
of sources including published records and personal communications from some of those who
have taken part in the work here recorded. The bibliography following the list includes the
titles of most of the published results of these explorations.

1837. F. DEPPE.

In the spring of 1837, Herr F. Deppe, who previously had collected numerous types of birds and mammals
in eastern Mexico and was then engaged in business in western Mexico, landed at La Paz and traversed the
peninsula overland from that point to San Diego, Calif. Lichtenstein published a brief summary of Deppe’s
observations, among which he states that ‘‘The peninsula is poor beyond description in living forms of organic
life. Neither reptiles nor insects were seen anywhere * * * .”

Deppe noted.a species of woodpecker frequenting the cactuses and says: ‘‘In addition to this there were
only humming birds and a species of crow which indicated the presence of water in the vicinity. At the
watercourses were occasionally found deer and still less frequently hares; with these exceptions no mammals
were seen.’’!

The quotations given above exemplify the kind of erroneous statements that gained currency during the
early history of the peninsula.

1839. Richarp Brinstry Hinps.

The first botanical collection in southern Lower California was made by Surg. Richard B. Hinds, on the
voyage of H. M. 8. Sulphur 2 in 1839. About 150 species were collected, the larger part being new. They
were from San Quintin, San Bartolome Bay, Magdalena Bay, and Cape San Lucas. The new species were
described by Bentham.?

1859. Joun A. Vearcu.

Dr. Veatch sailed from San Francisco in 1859 with a prospecting party which explored Cedros Island in the
hope of finding valuable mines. During June, July, and August he collected plants on the island and others
on the adjacent mainland, many of which were new. His plants went to the herbarium of the California
Academy of Sciences, and his published observations contained much new information concerning the flora of
Cedros Island. Some of the more conspicuous species obtained by him were described by Kellogg in ‘‘The
Hesperian.’’? One of the most remarkable of these, the elephant tree (Rhus veatchiana Kellogg), of which
Dr. Veatch published a long account, had already been described by Bentham under another generic name.

1859-1861. L. J. Xantus pr Vesey.

This naturalist, commonly known as John Xantus,was employed as tidal observer for more than two years, from
April, 1859, to the middle of 1861, by the United States Coast Survey at Cape San Lucas. His letters state
that he visited San Jose del Cabo, the Sierra Laguna, Todos Santos, La Paz, and Magdalena Bay, in addition
to traveling 350 miles up the west coast. Xantus was sent to the cape by the Coast Survey in collaboration.
with the Smithsonian Institution. He made large collections in several branches of science, notably in plants,
invertebrates, reptiles, and birds. At the time of his visit the flora and fauna of this region were practically
unknown and his collections furnished numerous types and first drew attention to the peculiar and interesting
flora and fauna of the cape district.

1867. Tue J. Ross Browne Expeprrion.

On January 5, 1867, J. Ross Browne, accompanied by William M. Gabb and F. von Lohr, landed at Cape
San Lucas to explore Lower California in the interest of a colonization company. The party traveled to San
Jose del Cabo and thence northerly to Triunfo and La Paz. From the last place they crossed the peninsula to
Todos Santos and up the coast to Magdalena Bay. From this point Browne returned to La Paz and the rest
of the party under Gabb continued across to Loreto and thence to Comondu, Purisima, Mulege, San Ignacio,
Santa Gertrudis, Calmalli, San Borja, Rosarito, Calamahue, San Fernando, San Quintin, Sausal, and Tijuana.
The reports on this trip by Browne and Gabb in Browne’s ‘‘Resources of the Pacific Slope” contains much
interesting information concerning the physical features of the peninsula. During this trip Gabb collected
afew plants, several of the cactuses serving as the types of new species.

1870-1890. Epwarp PatMER.

With the exception of Mr. Brandegee, Dr. Edward Palmer made the most extensive collections of plants
ever gathered in Lower California. Ee collected mainly along the coast and on the islands. His first work
in Lower California was on Carmen Island, in the gulf, January 1-2, 1870. Twenty years later he revisited
Carmen island and remained a week (November 1-7, 1890). A report on this collection was published by
Dr. J. N. Rose.

Dr. Palmer was the first naturalist to collect on Guadalupe Island. He remained there from February to
May, 1875, and again visited the island from March 27 to April 3, 1889. During his first visit, in addition
to a series of plants he made a small collection of bird skins which furnished the types of several new species

‘and first made known the interesting bird life of this island. .
: 1 Abh. Akad. Wiss., Berlin, 1838 (1839), pp. 417-451.

2 Belcher, Voyage Round the World, 2 vols., 1843.
3 Botany of the Voyage of H. M. 8. Sulphur, 1844.

 
142 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI

1870-1890. Epwarp Patmer—Continued.

Early in June and late in December, 1887, he collected at Mulege, and from November 22 to December 20
the same year worked at Los Angeles Bay. These collections, with others from the east side of the gulf, were
reported on by Watson. He made a collection of plants at San Quintin in January and February, 1889, and
proceeded thence southward to Lagoon Head, where he spent 2 week. From there he visited Cedros and San
Benito Islands, and on March 27 reached Guadalupe Island on his second trip, as already noted. Reports
upon these collections were made by Vasey and Rose. On Christmas Day, 1889, he left San Francisco by
steamer in company with T. S. Brandegee, who landed at Magdalena Bay, and began his prolonged explora-
tions of the peninsula, while Dr. Palmer continued to Guaymas and thence returned to La Paz, where he
collected from January 20 to February 5, 1890. Leaving La Paz he sailed for Santa Rosalia, touching en route
at Raza Island, in the gulf, February 12, and at San Pedro Martir Island, Sonora, on the following day. Be-
tween February 24 and March 15 collections were made at Santa Rosalia and Santa Agueda, the latter locality
10 miles inland from the port. A list of these collections was published by Vasey and Rose.

1881-1887. Lyman BELpinc.

The first ornithological work in Lower California after that of John Kantus was done by Lyman Belding,
who made two trips to the cape district and to points and islands along the west coast between December,
1881, and the last of March, 1888. Mr. Belding collected several new species and made valuable observations
which were published by Mr. Ridgway in the Proceedings of the United States National Museum for 1882
(Vol. V) and 1883 (Vol. VI).

The itineraries for these journeys were as follows: December 15, 1881—March 21, 1882, at La Paz and a short
visit to Espiritu Santo Island; April 1-May 17, 1882, at San Jose del Cabo, whence visits were made to Cape
San Lucas and other neighboring localities; April 14-26, 1882, Cedros Island; April 28, 1882, Santa Rosalia
Bay; May 2-11, 1882, San Quintin; May 16-17, 1882, Coronados Islands; December 15, 1882—March 23, 1883,
at La Paz; and in February a visit was made to Triunfo, Miraflores, and the Victoria (=La Laguna) Mountains,
1884 to 1887, several wagon trips were made in winter and spring from Tijuana to Laguna Hanson and return;
also to Ensenada, Guadalupe, and San Rafael Valleys, and the northern base of San Pedro Martir Mountains.
The collections made in the cape district were sent to the United States National Museum.

1882. C. G. Prineus, C. C. Parry, C. R. Orcutt, and M. E. Jonzs.
The late C. G. Pringle, who for many years was well known as a collector of plants on the Mexican main-
land, visited the northwest coast region of the peninsula from Tijuana southward to Todos Santos Bay, April
5-15, 1882, accompanied by C. C. Parry, C. R. Orcutt, and M. E. Jones.

1882-1886. C. R. Orcurr.

C. R. Orcutt made several short botanical journeys into Lower California and collected the types of various
new species, including some cactuses. On some of his trips he accompanied other botanists. A trip was made
with C. G. Pringle, C. C. Parry, and M. E. Jones from Tijuana southward near the coast to Todos Santos Bay
from April 5-15, 1882; and he was with H. C. Orcutt and C. C. Parry at Todos Santos Bay in January, 1883.
In September, 1884, he collected at Santo Tomas. Later he visited the desert about Lake Maquata, between
the Sierra Juarez and Cocopah Mountains, and gave an account of this area in the West American Scientist.
On one of his longer trips he left San Diego by wagon, March 30, 1886, crossed the boundary at Tijuana, and
collected at various localities along the Pacific coast southward to near San Fernando, where, on May 3, he
was unable to proceed farther by wagon. He published several papers, including new species, and a nominal
list of plants with no definite localities.

1885. Epwarp LEz GREENE.

Dr. Greene sailed from San Diego in April, 1885, and visited the Coronados, Todos Santos, Cedros, and
Guadalupe Islands, and San Quintin on the mainland, whence he returned to San Diego about the middle
of May. The results of this botanical trip were published by him in several papers, adding 15 species to the
flora of Guadalupe Island, seven of which were described as new.

1885-1894. Waiter E. Bryant.

We are indebted to the repeated expeditions of Bryant and his subsequent publications of results for much
of our knowledge of the bird life of Lower California and the islands on its western coast.

In January, 1885, he made a brief visit to Guadalupe Island, and December 16 of the same year again landed
there, where he remained until the Ist of April, 1886. The results of this trip appeared in the best paper yet
published on the birds of Guadalupe. The work on Guadalupe was his personal venture, but subsequently
he conducted a number of expeditions to the cape district and the coastal islands in the interest of the Cali-
fornia Academy of Sciences as set forth under the account of the expeditions of that institution. Bryant
collected mainly birds, but also secured numerous reptiles and mammals and a few plants when he was not -
accompanied by a botanist. His main collections forwarded to the California Academy of Sciences, were
destroyed by fire in April, 1906.

1887. BrewstER EXPEDITION.

One of the most important ornithological explorations ever made in Lower California was that organized
by William Brewster. The field work was done by M. Abbott Frazar, who landed at La Paz January 24,
1887, to collect birds in the cape district. Heremained there most of the year, visiting a number of localities
and making a large and extraordinarily fine and complete collection of the birds of this area. These specimens

‘are in the Brewster collection and include the types of a number of new species and subspecies described by
No. 1]

LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 143

4887. Brewster Exrepition—Continued.

Brewster. They also served as the basis for the admirable volume on the “‘ Birds of the Cape Region of Lower
California,’’ by the same author. '

Mr. Frazar’s itinerary, all in 1887, was as follows: January 24-February 26 and March 16-April 7, at La
Paz; February 26-March 15, at Espiritu Santo, San Jose, Montserrat, and Carmen Islands, and a brief visit
to the mainland opposite Carmen Island and to Loreto; April 11, June 10-July 2, and December 4-11, at
Triunfo; the last of April-June 9 and November 27-December 4, in the Sierra de la Laguna; July 4-30, at
San Jose del Rancho (Pierce’s ranch), 15 miles southeast of Triunfo; last of August-November 12, at San
Jose del Cabo.

1887-1899. A. W. ANTHONY.

1888. J.

A large part of our knowledge concerning the birds and mammals of northern Lower California and the
islands off the west coast of the peninsula is the result of extended and careful field work done by A. W.
Anthony from 1887 to 1899. During this period he made numerous overland trips and visited all the islands
on the west coast, including Guadalupe. Numerous papers on the results of this work with descriptions of
new species of birds and mammals have been published by Anthony and others. The large and valuable
collection of birds gathered during these years is now in the Carnegie Museum, Pittsburgh; the mammals
are mainly in the National Museum at Washington and in the American Museum of Natural History, New York.

The data for the following brief summary of his field work have been supplied by Mr. Anthony: March,
1887, landed at Ensenada and traveled overland to San Quintin, which was his base camp for more than a
year. From this point he visited San Fernando and Valladares in the interior and San Martin, San Geronimo,
and Cedros Islands offshore, besides making several landings along the coast south of San Quintin. During
1889 he was located at Valladares, whence several short trips were made into the San Pedro Martir Mountains,
including a visit there in April and May in company with C. H. Townsend, when a number of new species
of birds and mammals were collected. In the spring of 1893 about six weeks, including the month of
May, were spent in these mountains and the results were published in Zoe. In December, 1893, he spent
a week at the copper mines near San Fernando, and in 1894 he returned there and remained from the middle
of February to the 1st of July.

Anthony’s first visit to Guadalupe Island occupied 10 days late in May, 1892; he spent a week there in
August, 1896, and another week in April, 1897.

In 1896 he chartered a schooner and, in company with Horace Gaylord, spent from the last of July to early
September along the coast of the peninsula, including the San Benitos and other coastal islands, south to
Turtle Bay. ;

In 1897 he purchased a sealing schooner and, accompanied by R. C. McGregor, H. B. Kaeding, T. 8.
Brandegee, and A. L. Stockton, explored the west coast of the peninsula and the coastal islands, also including
Guadalupe, as far as San Jose del Cabo, where Brandegee landed and the rest of the party visited Socorro,
Benedicte, and Clarion Islands southwest of Cape San Lucas. Plants were collected, mainly by Stockton,
throughout this expedition.

On February 25, 1899, Anthony sailed from San Francisco in a schooner for the west coast of Mexico and
Central America. He was accompanied by R. H. Beck, R. C. McGregor, H. B. Kaeding, Chase Littlejohn,
and H.M. Gaylord. The expedition stopped and made collections of birds on several islands along the coast
of Lower California. On March 16 the schooner was wrecked and became a total loss on the coast about
30 miles north of Magdalena Bay.

G. Lemmon.

In May, 1888, J. G. Lemmon collected plants in San Rafael Valley and at Hanson’s ranch on the upper

slope of the Sierra Juarez.

1888-1889. CHartEs F. Ponp.

Lieut. Charles F. Pond, of the U. 8. S. Ranger, collected plants at several localities during a survey of the
shores and islands of the peninsula and made several visits to the San Benito Islands between December,
1888, and February, 1889. He also visited Cedros Island in the same winter and made a small collection,
which formed the basis of a supplementary list of Cedros Island plants published by Greene. He made hie
most important collections, however, on the low plains around the southern shore of San Bartolome Bay in

March, 1889.

1888-1906. CatrrorniA ACADEMY oF ScrENCES EXPEDITIONS.

The California Academy of Sciences has devoted much attention to the scientific exploration of Lower
California, especially between the years 1888 and 1894. This work was mainly in the southern half of the
peninsula and on the outlying islands. General collections were made, including fine series of plants, shells,
insects, and other invertebrates, reptiles, birds, and mammals, but practically all these collections were
destroyed by the fire following the San Francisco earthquake in April, 1906. Numerous papers have been
published on the results of this work in the Proceedings of the California Academy of Sciences, in Zoe, and in
other scientific publications. A good review of these expeditions is published in the Proceedings of the
California Academy of Sciences for 1895. A brief summary is given below:

1888. In March, 1888, W. E. Bryant landed at Magdalena Island and traveled to San Jorge by water and thence
overland across the peninsula through Comondu and La Giganta to Loreto, on the gulf coast. The return

158961°—21——10
144 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1888-1906. Catirornia AcaDEMY oF ScreNcEs ExrepiT1ons—Continued.
was made along the same route, but included the additional localities of San Gabriel and San Juan. Birds
were the main object of this visit, but some plants and mammals were also collected.

1889. On January 11, 1889, the second expedition landed at Magdalena Bay. W. E. Bryant and an assist-
ant, Charles D. Hains, represented the academy and made general collections of invertebrates, reptiles, birds,
and mammals. T. 8. Brandegee accompanied this party and worked independently on the botany of the
region traversed. The party journeyed north to San Jorge, La Purisima, Comondu, and back té Magdalena
Bay. From there they again returned to Comondu and traveled thence northward through La Purisima,
San Ignacio, Calmalli, San Borgia, San Fernando, and El Rosario to San Quintin, where they arrived in May
and sailed thence to San Diego. This expedition was extremely fruitful in results, especially in reference
to the plant and bird life of the peninsula.

1890. In September of this year W. E. Bryant for the academy, and T. S. Brandegee on his own account,
landed at San Jose del Cabo and traversed the country through to Miraflores, the Sierra San Francisquito,
Triunfo, and La Paz, where the expedition ended in October.

1892. During March, April, and May of this year the cape district was again visited by a party from the
academy, consisting of W. E. Bryant and Gustav Eisen, accompanied by T. 8. Brandegee on his own account.
They landed at San Jose del Cabo and visited Cape San Lucas, Agua Caliente, Miraflores, Sierra San Francis-
quito, returned to San Jose, and went by steamer to La Paz and thence to California.

1893. During September and October Gustav Eisen traveled from San Jose del Cabo across the Sierra El
Taste to Pescadero and Todos Santos, then returned through Cape San Lucas to San Jose and thence to Mira-
flores and through the Sierra San Francisquito and Sierra Victoria to Todos Santos again. During part of this
expedition Dr. Eisen was accompanied by Mr. and Mrs. T. 8. Brandegee.

1894. During September, October, and November Gustav Eisen and Frank H. Vaslit visited San Jose del
Cabo and traveled thence through Santa Anita, Miraflores, La Palma, the Sierra San Lazaro, El Taste, and
back to San Jose. From this point they traveled by way of Santiago, San Bartolo, and Triunfo to La Paz.

1903. April 25 the schooner Mary Sachs (31 tons) sailed from San Francisco on a scientific expedition to the
Revillagigedo Islands, off the west coast of Mexico, in the interest of the California Academy of Sciences.
The expedition was in charge of R. H. Beck, who was specially interested in birds. The other members of
the party were A. S. Bunnell, birds; E. W. Gifford, shells; F. E. Barkelew, plants; C. H. Marks, jr., anthro-
pology. During April and May the expedition made brief stops to collect at Ensenada, on the mainland of
Lower California, and at the adjacent San Martin, San Benito, and Natividad Islands. Fine series of speci-
mens were collected during this expedition, which returned to San Francisco, August 13. These collections
were all destroyed by fire in April, 1906.

1905-6. June 25, 1905, the schooner Academy (114 tons) sailed from San Francisco on a scientific expedition
to. the Galapagos Islands, from which it returned November 29, 1906, with extraordinarily rich collections.
The expedition was in charge of R. H. Beck, who collected birds and reptiles and was accompanied by
E. W. Gifford, birds; J. 8. Hunter, birds and mammals; J. R. Slevin, reptiles; E. S. King, reptiles; F. X.
Williams, insects; Alban Stewart, plants; and W. H. Ochsner, shells and fossils. On its way south in July
the expedition made brief stops and small collections at Ensenada, Lower California, and on the adjacent
San Martin, San Geronimo, San Benito, Cedros, and Natividad Islands.

1889-1902. TowNSHEND StitH BRANDEGEE.

Mr. Brandegee has done more than any other botanist, both by field work and in published results, to make
known the flora of Lower California. Most of his field work was carried on in connection with expeditions of the
California Academy of Sciences, which he accompanied at his own expense. In addition to these expedi-
tions, detailed elsewhere, he visited the coast region adjacent to Magdalena Bay in January, 1890, working
from Matancita to Todos Santos and thence to the Sierra de la Laguna and La Paz, where the journey ended
in February.

In the early part of 1893 Mr. Brandegee turned his attention to northern Lower California. During May
he crossed the boundary to Tijuana and traveled southward along the coast past Ensenada and as far as Cape
Colnett. From this vicinity a side trip was made into the San Pedro Martir Mountains and back over the
same route to Tijuana. In September and October of the same year he returned to the cape district. From
San Jose del Cabo a trip was made with Mrs. Brandegee to El Taste, and various localities, including the high
mountains, were revisited.

In February, 1897, he cruised with A. W. Anthony in a schooner along the west coast and adjacent islands.
They stopped at Ensenada and then in turn at the islands of Todos Santos, San Martin, San Geronimo, Guada-
lupe, San Benito, Cedros, and Natividad, and at San Bartolome Bay. Continuing farther south they touched
at San Roque and Ascencion Islands and landed at San Jose del Cabo, where Brandegee left the boat. In
September, 1899, and during the same month in 1902, he made other collections at a number of localities in the
cape district. The Brandegee collection of plants is now in the herbarium of the University of California.

1892-1898. F. FRANCESCHLI.
In December, 1892, and January, 1893, F. Franceschi made a small collection of plants on Guadalupe Island.

Very few species were added to those already known, but the list published by the collector includes inter-
esting notes on the condition of the flora as affected by the goats introduced on the island many years before.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 145

1892-1894. Gustav HIsEn.

Dr. Gustav Eisen made several scientific trips to Lower California, mainly as member of expeditions sent
out by the California Academy of Sciences. These expeditions are noted elsewhere. In addition he spent
the period from June to August, 1894, in the middle parts of the peninsula, landing at Loreto and traveling
through Comondu, San Vicente, La Purisima, Guajademi, San Ignacio, and San Angel, and thence across the
Vizcaino Desert to Cerro Prieto, La Huerta, and San Andres on the coast south of San Bartolome Bay and to the
placer mines of the adjacent Sierra Pintada..

In addition to a number of scientific papers based on collections made on the peninsula, Dr. Eisen published
two important articles, referred to elsewhere, on the geography and geology of southern Lower California with
the best maps of that region which have yet appeared.

1893. StowELL aND Lunt ExrepitTion.

J. C. Stowell and 8. C. C. Lunt made a collecting trip from Ensenada to the higher parts of the San Pedro

Martir Mountains during June and July, 1893. Their collections of reptiles, birds, and mammals are in the
Museum of Stanford University.
1894. Epgar ALEXANDER Mrarns. : ;

The late Col. Edgar Alexander Mearns, U. 8. Army, while attached to the International Boundary Com-
mission as medical officer and zoologist, made extensive collections of mammals, birds, reptiles, amphibians,
fishes, plants, and other specimens of natural history in Lower California between March and July, 1894.
His investigations included an exploration of the Lower Colorado River and the lower part of the Salton River
from March 23 to early May, and trips to Nachoguero Valley from June 1 to 7, to the San Isidro Range from
June 28 to July 8, and to the Tijuana Valley, July 9 to 20, besides incidental work conducted south of the
boundary line at various other points. Some of the results of this work are published in Part I of Col. Mearns’s
report of the mammals of the Mexican boundary. The collections resulting from this work are in the United
States National Museum.

1895. Leon Diever. ;

During 1895 a few plants were collected by Leon Diguet on Catalina Island and perhaps at other localities
in the Gulf of California. The most interesting result of his work was the discovery on Catalina Island of the
giant Echinocactus bearing his name.

1895-1915. BrotoaicaL Survey ExpEpITions.

1895. March 31, 1895, J. Ellis McLellan arrived at San Jose del Cabo to collect the birds and mammals of
that region for the Biological Survey. From this point he visited Cape San Lucas and Santa Anita and pro-
ceeded overland through Miraflores, the Sierra Victoria, San Felipe, Todos Santos, and Triunfo to La Paz,
where he arrivedin June. From this base he visited the gulf islands of San Jose, Carmen, Espiritu Santo, and
Ceralvo, and closed his field work in September, 1895.

1896. In December, 1896, Frank Stephens spent about two weeks collecting birds and mammals for the
Biological Survey between the delta of the Colorado River and the Cocopah Mountains.

1905-6. Between March, 1905, and February, 1906, E. W. Nelson and E. A. Goldman explored the penin-
sula throughout its extent in the interest of the Biological Survey as set forth in this report. Collections were
made of plants, fresh-water fishes, reptiles, birds, and mammals, all of which are in the United States
National Museum. .

1915. In May and June, 1915, Luther J. Goldman visited the delta region of the Colorado River in Lower
California, where collections, mainly of breeding birds; were made at various localities from Volcano Lake
to the mouth of the river.

1896. W. W. Price ExpepIrion. :

In 1896 Mr. Price sent a party to southern Lower California to collect birds and mammals. Dane Coolidge
was in charge and was assisted by Loye Miller and J. F. Abbott. They arrived at San Jose del Cabo May 9,
and soon located at Santa Anita ranch, where they remained for six weeks. Practically all of July was spent
at La Laguna, on the Sierra dela Laguna. August was passed in the low country, mainly at San Jose del Cabo,
after which the party returned to San Francisco.

This expedition appears to have been purely commercial and the collections were sold and widely distrib-
uted, most of the mammals going to the British Museum. Among these collections were the types of several
species of mammals.

1898-1901. C. A. Purpus.

Dr. C. A. Purpus was shipwrecked at Lagoon Head in the spring of 1898. He collected at Santo Domingo,
near Lagoon Head, and, journeying to Calmalli and San Pablo, made further collections in the central part
of the peninsula. Several of his plants were described by Brandegee as new species. He collected plants
at San Jose del Cabo in 1901.

1902. Fretp Museum or Natura History Expepition.

February 28, 1902, Edmund Heller arrived at Ensenada to collect mammals in the northern part of the
peninsula for the Field Museum of Natural History. During March and April he worked south to Santo Tomas
and easterly through Salado Canyon and Trinidad Valley to San Felipe Bay on the gulf coast, returning over
the same route to Ensenada.
146

MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1902. FreLp Museum or Natura History Expepition—Continued.

1903. J.

May 7 he again left this place and journeyed through Trinidad Valley and along the west side of the San
Pedro Martir Mountains to San Antonio ranch and Rosarito. Thence he crossed these mountains to Matomi
and Rosarito Divide and down to San Quintin, whence he made a short visit to the lower San Simon River.
The last of July he returned to the San Pedro Martir Mountains and traversed their higher parts, visiting
Santa Eulalia, Santa Rosa, La Grulla, Vallecitos, and Pifion. He left the mountains from the latter point in
October and crossed Trinidad Valley to Alamo and thence passed through Ojos Negros in San Rafael Valley to
Hanson Lagoon, on top of the Sierra Juarez. From this point the top of these mountains and a part of the
neighboring eastern slope were explored, field work ending about the middle of November.

This was the first attempt to make a detailed and thorough collection of mammals in any part of Lower
California, and the report on the collection by D. G. Elliot, including quotations from Heller’s field notes,
was a valuable contribution to our knowledge of the physical conditions and mammal life in the northern
part of the peninsula.

H. Barty.

In the fall of 1903, J. H. Batty collected birds for the American Museum of Natural History in the country
lying inland from San Quintin. During October he visited San Simon and the San Pedro Martir Mountains,
and during November continued work in these mountains at Rosarito, San Jose Canyon, Fresnal Canyon, and
Santa Eulalia. :

1904-1905. Danie, TremBLey MacDovaat.

In January and February, 1904, Dr. MacDougal made a boat trip from Yuma, Ariz., to San Felipe Bay,
Lower California, where a collection was made in the interest of the New York Botanical Garden. In March
and April of the following year, in the interest of the Carnegie Institution, and accompanied by E. A. Goldman
of the Biological Survey, he again descended the Colorado River, the objective point being the Cocopah Moun-
tains, which were reached by ascending Hardy River. A few plants were obtained by MacDougal along the
route, and as many species as the time and season would permit were taken by him in the Cocopah Mountains.
This collection was also added to the herbarium of the New York Botanical Garden. Goldman, meanwhile,
devoted his attention to the fauna of the region, the collection secured being sent to the United States National
Museum.

1905. Pamapetpaia AcaApEMyY or NatTurRAL ScieNcEs EXPEDITION.

During February, 1905, Mr. S. N. Rhoads, in company with H. E. Wilder and a guide, made a boat trip to
the delta of the Colorado River in the interest of the Philadelphia Academy of Sciences. They descended to
the mouth of Hardy River, which they then ascended to the base of the Cocopah Mountains. Thence they
traveled overland to Calexico on the California border. This trip was mainly to collect specimens of the birds
and mammals, of which annotated lists were published in the Proceedings of the Academy of Natural Sciences
of Philadelphia for 1905.

1906-1910. THayzeR Musrum Exprprrions.

During the years named above John E. Thayer sent the well-known field naturalist W. W. Brown to various
parts of Lower California and adjacent islands to collect birds and a few mammals. Several papers on the
results of this work have been published by Thayer and Bangs, mainly in the Condor. The birds collected
are in the Thayer Museum and the mammals in the Museum of Comparative Zoology. Brown’s work was dis-
tributed as follows: Cedros, San Benito, and Guadalupe Islands in sequence from April to June, 1906; San
Quintin, July, 1906, and January, 1917; Rosario, October and November, 1906, and February, 1907; San An-
dres, Rosarito, Santana, and San Javier, February to April, 1907; La Paz, February to April and September,
1908; also March to June, 1910; San Jose Island, June, 1908, and February, 1909; Pichilinque and Puerto Bal-
andro, May, 1908; Sierra de la Laguna and Triunfo, August and September, 1908; El Sauz and Miraflores,
November and December, 1908; Santiago and Eureka, November, 1908, to January, 1909 ; El Callo Island,
February, 1909; San Ildefonso Island, April, 1909; near Loreto, Comondu, and La Purisima, April and May,
1909; Espiritu Santo Island, June, 1910.

1911. AMERICAN Museum or Natura. History Expepirvion.

During the spring of 1911, a party on the United States Bureau of Fisheries steamer Albatross made a bio-
logical reconnaissance of the coast and bordering islands of Lower California in the interest of the American
Museum of Natural History. The expedition was in charge of Dr. C. H. Townsend, who devoted special
attention to reptiles and fishes and who was accompanied by the following scientific corps: Dr. J. N. Rose,
botany; Dr. Paul Bartsch, invertebrates; H. E. Anthony, mammals; and P. I. Osburn, birds.

The steamer left San Diego, Calif., February 25, and proceeded to Guadalupe Island, where it remained
several days and discovered a herd of living sea elephants, a species previously reported to be extinct. Young
sea elephants were captured and taken to San Diego to be shipped to New York. March 7, the party again
left San Diego and steamed down the coast, stopping at practically all of the islands along the west coast, as
well as at some points on the mainland, including Magdalena Bay and Cape San Lucas. Thence the boat pro-
ceeded into the Gulf of California, touching at La Paz and Mulege and on most of the islands of the Gulf north
to Angel de la Guarda and Tiburon, as well as at Guaymas on the Sonora shore. The expedition was ex-
tremely productive in material, fine series of plants, invertebrates, reptiles, fishes, birds, and mammals being
collected. These results added much to our knowledge of distribution, in addition to discovering numerous
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 147

1911, American Museum or Naturat History Expzpirion—Continued.
new species of plant and animal life. The expedition ended when the Albatross arrived at San Francisco the
last of April, 1911.

The localities visited during this expedition, with inclusive dates, follow: Guadalupe Island, March 2 to 4;
East and West San Benito Islands, March 9; Cedros Islands, March 10-12; San Bartolome Bay, March 13-14;
San Roque Island, March 15; Abreojos Point, March 16* Santa Maria Bay, March 18; Santa Margarita Island,
March 19-20; Mangrove Island, March 20; Magdalena Island, March 21; Cape San Lucas, March 23-24; San
Jose del Cabo, March 25-26; Pichilinque Island, March 27; on peninsula near Pichilinque Island, March 28;
La Paz, March 29; San Jose Island, March 30-31; Cayo Island, March 31; Agua Verde, April 1-2; Carmen
Island, April 2-3; Mulege, April 4; Concepcion Bay, April 5-7; San Francisquito Bay, April 9-10; Angel de
la Guarda Island, April 10; Tiburon Island, April 11; Seal Island, April 13; San Esteban Island, April 13-14;
Catalina Island, April 16; Santa Cruz Island, April 16; Pichilinque Island, April 17; Espiritu Santo Island,
April 18-19; Ceralvo Island, April 19.

BIBLIOGRAPHY OF LOWER CALIFORNIA.

This bibliography is not complete but includes the titles of most of the general works and
scattered articles containing original matter relating to the peninsula, its history and physical
character, and the attempts at colonization within its borders. In addition, it contains most of
the publications on botany and vertebrate zoology except those on the fishes. But few papers
on invertebrate zoology have been included, this subject being outside the field of the present
work.

1686. Oronpo y ANTILLON, Isoporo. Some Discoveries Made in the Island of California in the Year 1683. <A Rela-
tion of the Invasion and Conquest of the Floridas. London, 1686; published in Paris in 1685.
Not seen. *

1705... . Letters Edifiantes et Curieuses, etc., V. Recueil, 12mo., pp. 1-287, with map.
Contains a translation of Padre Piccolo’s memoir on missionary work in Lower California to the Royal Council at Guadalajara.

1708. Prccoto, Francis Maria. An Extract of a Memoir Concerning the Discovery of a Passage by Land to California
with a Map and Description of the Country. Presented to the Royal Council of Guadalaxara in Mexico, by
Francis Maria Piccolo. Taken from the Letters of the Missionary Jesuits, printed at Paris. With map by Padre
Eusebius Francis Kino. , <Philos. Trans., XXVI, No. 318, Nov.—Dec., pp. 232-240. London.
Thisis an English translation of the original French edition printed in Paris. The original memoir is dated Guadalaxara, Mexico,
February 10, 1702. Piccolo’s account of the mountain sheep is one of the sources of the subsequent notices of this animal by Venegas,
Clavijero, and others, and is the first account of these animals from America.

1712. Rogers, Woopzs. A Cruising Voyage Round the World [from 1708-11]. Pp. 428, with appendix and maps,
8vo, London.
During this voyage the west coast of Mexico, the Tres Marias Islands, and southern Lower California were visited during 1709-10. There
are a few notes on natural history. é

1715. Bernarp JEAN Freperic. Recueil de Voiages au Nord, ete. 12mo., Vol. III, pp. 1-340.
Contains brief translations of a few of the Spanish accounts of explorations and missionary work in Lower California.

1723. Torquemapa, JuAN DE. Primera [segunda, tercera] parte de los veinte i un libros rituales i monarquia indiana,
con el origen y guerras de los Indios Occidentales, de sus poblagones descubrimiento, conquista, conuersion y
otra cosas marauillosas de la mesma tierra. 3 vols., quarto, Madrid, 1723 [original ed. Seville, 1615].
Contains a history of the Indians of Mexico and of the conquest, with an account of the missions and explorations in upper and Lower
California, especially the voyages of Vizcaino along the west coast of the peninsula (vol. 1). This work is usually cited under an abbrevi-
ated title as the ‘“‘Monarquia Indiana.”’

1726. SHetvockE, Cart. GzorcE. A Voyage Round the World by Way of the Great South Sea. 8vo., pp. 1-468,

with cuts. London.
Contains an account of a visit to the southern end of Lower California in August, 1721, and a brief account of the peninsula, including
an extremely favorable statement concerning the character of the natives encountered at Cape San Lucas.

1754, Venzcas, MievErn. El Apostol Representado en la vida del V. P. Juan Maria Salvatierra. Pp. 316, 8vo.

Mexico.
This account of the life of Salvatierra includes his work in Lower California but has little of interest.

1757, —__-——. Noticia de la California. 3 vols., pp. 240, 436, 564, with plates and maps, 8vo. Madrid.
A general account of the early settlement and physical character, people, and natural products of Lower California. The margin of

the map in the first volume contains the first published illustration ofan American mountain sheep. In appendices to vol. 3 are accounts
of voyages along the coasts of the peninsula by Vizcaino, 1602; Woodes Rogers, 1710; Lord Anson, 1740; and Padre Fernando Consag, 1746.

1759, —_——_——. A. Natural and Civil History of California. 2 vols., pp. 387, 455, 8vo. London.
A translation of the “Noticia de la California,” published in 3 vols., in Madrid, 1757, The natural history of Lower California is treated

> in vol. 1, pp. 32-52.
148 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1762. Lockman. ... Travels of the Jesuits into Various Parts of the World, etc.’ 8vo., Vol. I, pp. 1-487. London.
Contains a few letters and reports giving interesting details of the work of Padre Piccolo and other early missionaries in Lower Cali-
fornia. Mainly concerns other parts of the world.

1772. Barezrt, Jaxos. Nachrichten von der Amerikanischen Halbinsel Californien. Pp. 358, pl., and map, small
8vo. Mannheim.

Baegert was a missionary stationed at the mission éf San Luis Gonzaga and gives an account of the physical character, inhabitants,
early missions, and natural history of the peninsula, mainly the southern part, up to 1767. x;

1778. D’AutERocHE, CHaprE. A Voyage to California to Observe the Transit of Venus, etc.. 8vo., pp. 1-215. London.

A briefaccount of a voyage made in 1768-69. The party crossed Mexico from Vera Cruz to San Blas and sailed thence to San Josedel

Cabo where it landed and observed the transit of Venus as planned. The most notableincident was the virulent epidemicof yellow fever
experienced at San Jose del Cabo.

1811. Humsoxprt, ALEXANDER. Essai Politique sur le Royaume de la Nouvelle-Espagne. 2 vols., with atlas of charts
and maps, folio. Paris.

These volumes form Part IIT of the“ Voyage de Humboldt et Bonpland.”’ In Vol. I, pp. 308-327, is a brief account of Lower California,
with a referenceto mountain sheep near Comondu and to antelopenear Monterey, Calif. Anoctavo edition of this work without the charts
and maps was published the same year in Paris.

1829. Harpy, Lizsut. R. W.H. Travels in the Interior of Mexico in 1825, 1826, 1827, and 1828. Pp. 540, with plates
and map, 8vo. London.

This narrative covers the country from the City of Mexico to the mouth of the Colorado River and the shores of the Gulf of California,
The author investigated the pearl industry of the gulf in the interest of an English corporation.

1832. Morrett, Bensamin, Jr. A Narrative of Four Voyages to the South Sea, North and South Pacific Ocean, etc.
Pp. 492, 1 plate, 8vo. New York. /

Contains an account of visits to Guadalupe, Socorro, and other islands and the west coast of upper and Lower California to a point north
of San Francisco during March to June.

1834. Dunavut-Citty, AucusTe. Voyage Autour du Monde. 2 vols., pp. 409, 438, 8vo. Paris.

Narrative of the voyage of the Heros, 1826-29, including a visit to the west coast of Mexico, the cape district of the peninsula, and upper
California. The naturalist Dr. Botta accompanied this trading voyage.

1839. Forspres, ALEXANDER. California: A History of Upper and Lower California. Pp. 352, with maps and plates,
8vo. London. : ie
' Contains a briefsummary of the history of both areas; the Lower California part appears to have been taken mainly from Venegas.

1839. LicuteNstTEIn, H. Beitrag zur olisholegichet Fauna von Californiau. <Abh. Akad. Wiss. Berlin, fiir 1838,
pp. 417-451.

Contains a summarized account of an overland journey by F. Deppe in the spring of 1836 from Loreto up through the peninsula to San
Diego, California, with brief notes on the plant and animal life observed.

1840-55. Du Perit-THovars, ABEL. Voyage Autour du Monde sur la Frégate La Venus Pendant les Années, 1836-
1839. 10 vols., 8vo, with folios of plates. Paris.

The narrative is contained in the first four volumes with three folios of plates [1840-43]. The fifth volume with one folio of plates

[1855] contains papers on mammals by Geoffroy Saint Hilaire and birds by Prevost and Des Murs. The narrative describes visits to

Magdalena Bay and points in the cape district of the peninsula and to Guadalupe Island. ,

1843. Bercuer, Sir Epwarp. Narrative of a Voyage Round the World Performed in H. M. 8. Sulphur During the
Years 1836-1842. 2 vols., pp. 387, 474, with maps and plates, 8vo. London.
Includes brief notes on the west coast of Mexico and Lower California. Richard B. Hinds was surgeon and naturalist of this expedi-
tion.
1844. Benruam, Georcr. The Botany of the Voyage of H. M. 8. Sulphur, under the Command of Capt. Sir Edward
Belcher, During the Years 1836-1842. Edited and superintended by Richard Brinsley Hinds, Esq., surgeon,
R. N., attached to the expedition. The botanical descriptions by George Bentham, Esq. Pp. 195, 60 plates,

quarto. London.

The Sulphur on her her voyage around the world visited San Quintin, San Bartolome Bay, Magdalena Bay, and Cape San Lucas,
where about 150 species of plants were collected by the surgeon, Richard B. Hinds. This being the first collection from southern
Lower California, the larger part of the plants were new to science.

1844. Dz Morras, Durtor. Exploration du Territoire de l’Orégon, des Californies et de la mer Vermeille. 2 vols.,

pp. 521, 514, with plates, 8vo. Paris.
An account of a journey made during the years 1840-42, with some notes on fur-bearing and other large mammals.

1850. Burrum, Lieut. E. Gounp. Six Months in the Gold Fields in Upper and Lower California. Pp. 172, small

8vo. Philadelphia.
An account of conditions in the gold-mining districts of California, with a brief account of the occupation of La Paz by the Seventh
Regiment of New York Volunteers, which landed there July 21, 1847, and held possession of the peninsula to the close of the Mexican War.

1850. Ryan, Wu11am Repmonp. Personal Adventures in Upper and Lower California in 1848-1849. 2 vols., pp.

347, 413, with plates, small 8vo. London.
Contains notes on conditions at southern end of Lower California in 1848,

1851... . Madame la Comptesse de, Histoire Chrétienne de le Californie. 12 mo., pp. 262. Plancy.
A brief account of the natives of Lower California and the work among them of the principal Spanish missionaries.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. _ 149

1852. CLavicrro, Francisco Javier. Historia de la Antigua 6 Baja California, Mexico. Pp. 252, 3 plates, quarto.
Mexico [first published in Italian at Venice 1789].
A translation from the Spanish edition into English was made by A. G. Randall and extracts published in the appendix to Browne’s
Resources of the Pacific Slope, pp. 155-172 (1887).. This work contains some notes on the animal and plant life of Lower California.
1853. SEemanNn, BerrHoup. Narrative of the Voyage of H. M. 8. Herald During the Years 1845-51. 2 vols., pp.
22, 302, with plates and map, 8vo. London.

Contains a brief account of the coasts of Lower California and western Mexico with Seemann’s trip inland from Mazatlan to Durango
City and return.

1854. Rew, J.M. The Ensenada. <National Magazine, Vol. 4, No. 6, pp. 502-505, June.

Contains interesting notes concerning the capture of La Paz and other ports of Lower California by the filibuster Walker, in 1853,
by a member of the expedition.

1855. Leconts, Joun. Account of some Volcanic Springs in the Desert of the Colorado. <Silliman’s American
Journal of Science, 2d ser., XIX, pp. 1-6, May.

A brief popular account of a trip from the coast range across the Colorado Desert to the mud “Volcanos” near Volcano Lake. This
is the first report on these ‘‘volcanos’”’ by a scientific: observer.

1859. Barrp, SPENCER F, Notes on a Collection of Birds made by Mr. John Xantus at Cape Saint Lucas, Lower Cali-
fornia, and now in the Museum of the Smithsonian Institution. <Proc. Acad. Nat. Sci., Philadelphia,
November, pp. 299-306.

Contains original descriptions of Cardinalis igneus, Pipilo albigula, Chamaepelia passerina var. pallescens, and the name pertinar
proposed for the cape region Jfyiarchus if it proves to be distinct; 42 species given in the list. Also mentions two bats ( Macrotus cali-
Jornicus and Vespertilio pallidus) from the same locality.

1859. Lassepas, Utises Urspano. De la Colonizacion de la Baja California y Decreto de 10 de Marzo de 1857.
Primer memorial, Mexico. 8vo, pp. 250.

A brief account of the settlement of Lower California from the time of its discovery until 1857. In addition to the account of the
colonization, including the land grants to individuals covering parts of this region, the work has chapters on the physiography, resources,
and civil government of the peninsula. It is a valuable contribution to the history of Lower California.

1859. Xanrus, Jonn. Descriptions of Supposed New Species of Birds from Cape Saint Lucas, Lower California.
<Proc. Acad. Nat. Sci., Philadelphia, November, pp. 297-299. /

Contains original descriptions of Picus lucasanus, Campylorhynchus affinis, Harporhynchus cinereus, Brachyrhamphus hypoleucus.

1860. Ketuoae, A. The Tree Primrose or False Fuchsia. <The Hesperian, IV, No. 1, pp. 1-2, with plate, March.

Original description of the handsome red-flowered shrub Oenothera arborea Kellogg (Xylonagra arborea).

1860. —————. Cedar Island Juniper. <The Hesperian, IV, No. 1, pp. 2-4, March.
Original description of Juniperus cedrosiana Kellogg (=Juniperus californica). Reported as abundant in canyons at about 700 feet
elevation, on the east side of Cedros Island.
1860. —————-._Idria columnaris. <The Hesperian, IV, pp. 101-102, with plate, May.
Original description of this remarkable species, known only from Lower California.
1860. Lawrence, G.N. Descriptions of Three New Species of Humming Birds of the Genera Heliomaster, Amazilia,
and Mellisuga. <Ann. Lyc. Nat. Hist., New York, VII, p. 109, April.

Contains original description of Amazilia rantusi (=Basilinna zantusi) from Cape San Lucas.

1860. ——————. Description of a New Species of Bird of the Genus Phaeton, also a New Species of Humming Bird
of the Genus Heliopzdica. <Ann. Lyc. Nat. Hist. New York, VII, p. 145, April.

Describes H. castaneocauda (= Basilinna zantusi) from Cape San Lucas.

1860. VzatcH, Joun A. About Cerros Island. <The Hesperian, III, No. 6 pp. 529-534, with plate, February.
A brief popular account of the island and its fauna and flora based on the author’s visit there in 1859.

1860. —————-. Rhus veatchiana Kellogg: Veatch’s Sumach, or Elephant Tree. <The Hesperian, IV, No. 2, pp.
49-51, with plate, April.
This article contains the original description of this species by Kellogg with interesting notes on the tree by Veatch.
1861. Gray, Asa. Enumeration of a Collection of Dried Plants Made by L. J. Xantus, at Cape San Lucas, etc., in
Lower California, Between August, 1859, and February, 1860, and Communicated to the Smithsonian Insti-

tution. <Proc. Amer. Acad. of Arts and Sci., V, pp. 158-173, January. a
An important list of 121 species including a number described as new.

1861. Ives, Lieut. Josern C., and others. Report upon the Exploration of the Colorado River of the West, in 1857
nd 1858. In 5 parts, with maps and plates, quarto. Washington. Senate Doc.
Part I, General Report [Narrative], pp. 131, by Lieut. J. C. Ives.
Part II. Hydrographic Report, pp. 14, by Lieut. J. C. Ives.
Part III. Geological Report, pp. 154, by Dr. J. S. Newberry.
Part IV. Botany, pp. 30, by Gray, Torrey, Thurber, and Engelmann.
Part V. Zoology, pp. 6, by S. F. Baird.

This report contains interesting information concerning the Gulf of California and the Colorado River and adjacent country up to the
big bend and thence through northern Arizona to Fort Defiance.
150. MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI

1864. BarGcerT, Jacos, An Account of the Aboriginal Inhabitants of the Californian Peninsula. <Annual Report
Smithsonian Institution for 1863, pp. 352-369, 8vo. Washington.

A translation by Dr. Charles Rau from the original of those parts of Baegert’s Nachrichten relating to the natural history and native
population of the peninsula.

1864. Comprer, C. Voyage au Golfe de Californie. Nuits dela Zone Torrid, Accompagné d’une Carte de la Sonora.
pp. 544, with map, 8vo. Paris. '
Narrative of voyages in 1828-1831 to Tampico and along the west coast of Mexico to head of the Gulf of California, with an account of

the pearl fisheries in the gulf,and of a journey from Mazatlan across country to Durango and on to Vera Cruz, including a deer hunt at
Guaymas and a bear seen in the mountains of Durango. Great numbers of pelicans were noted in the gulf.

1865. Lersz, Jacop P. Historical Outline of Lower California. Pp. 46, 8vo. New York.
This pamphlet was published to exploit a colonization land grant in southern Lower California given the author by the Mexican Gov-
ernment. It contains a brief account of the southern part of the peninsula and various documents relating to the grant.
1866. Gut, TazoporE. Prodrome of a Monograph of the Pinnipedes. <(Proc. Essex Inst., V, p. 13, March.
The name and distribution of Macrorhinus angustirostris appear in the text and a footnote gives some of its characters, thus consti-
tuting the original description of the northern elephant seal.
1866. —————.. On a New Species of the Genus Macrorhinus. <(Proc. Chicago Acad. Sci., I, pp. 33-34, April.
A brief description of Macrorhinus angustirostris. Type locality, San Bartholomew Bay, Lower California. This publication is ante-
dated by that in the Essex Institute.
1868. ANonymous. Lower California: Its Geography and Characteristics, with a Sketch of the Grant and Purposes
of the Lower California Co. Pp. 1-44, with map, 8vo. New York.
Gives an account of the formation and grant of the Lower California Co., with a list of its officers. Also Taylor’s Historical Summary
of the peninsula, which again appears in the appendix to Browne’s Resources of the Pacific Slope.
1868. Browns, J. Ross. Explorations in Lower California. <Harpers New Monthly Magazine, Vol. 37, pp. 578-591
(October); pp. 740-752 (November); Vol. 38, pp. 9-23 (December).
A good popular account of the southern end of the peninsula from Cape San Lucas and La Paz to Magdalena Bay in 1866, with com-
ments on some of the former colonization schemes. Browne’s expedition was in the interest of the Lower California Co. of New York.
1868. Taytor, ALEXANDER S. Historical Summary of Lower California. <Appendix to “Lower California: Its
Geography and Characteristics.’? Pp. 24-40, 8vo. New York.
1869. Scammon, CHar.es M. [and Epwarp D. Corz]. On the Cetaceans of the Western Coast of North America.
<Proc. Acad. Nat. Sci., Philadelphia, XXI, pp. 13-63, April.

This paper isin two parts: Part I, by Cope, contains a systematic synopsis of the species. Part II, by Scammon, on habits and
distribution, includes an account of the sea elephant.

1870. Harte, Fr. Bret. The Story of an Ornithologist. <Overland Monthly, IV, pp. 168-173, February.
A good biographical notice of A. J. Grayson.

1870. Scammon, Capt. C.M. Sea Otters. <(Overland Monthly, IV, pp. 25-30, January.
‘ A popular account of sea otters and their hunting from Lower California to the Aleutian Islands, with some notes on the habits of the

Lower Californian animal.
1870. ——_-_——-. Sea-Elephant Hunting. <Overland Monthly, IV, pp. 112-117, February.
A popular account of the northern and southern sea elephants, with notes on the northern species along the coast of Lower California.
1870. —————. On the Lower California Coast. <(Overland Monthly, IV, pp. 230-238, March.

Contains interesting notes on whaling in Magdalena Bay and elsewhere on the coast of the peninsula.

1871. ANonymous. Nulidad del Contrato Leese. Folio, 23 pp. Mexico.
Nullification of the contract made in 1864 with Jacob P. Leese for the colonization of lands around Magdalena Bay in Lower Cali-
fornia. Name, age, etc.,of each colonist is given. (Not seen.)

1871. Dexay, Draxe. In Relation to Occurrences at Magdalena Bay, Lower California, Mexico. Pp. 90, 8vo. San

Francisco.
A report by the United States consul at Magdalena Bay, with accompanying documents concerning the capture of the town and mis-
treatment of American employees and colonists of the Lower California Co. in October, 1871, by insurgents from La Paz.

1872. Hernanvez, El General Jost Marta Perez. Compendio de la Geographia del Territorio de la Baja California.

12mo., pp. 1-95. Mexico.
A brief school geography of the peninsula, with an account of the local government.

1873. Evans, Tatiesan. South of the Boundary Line. <Overland Monthly, XI, pp. 157-162, August.
This is mainly an account of mining activity in the northern part of the peninsula, with notes on the country and people.

1874, Patou, Papre Fr. Francisco. Noticias dela Nueva California. 4 vols, pp. 270, 301, 815, 253, large 8vo, with
plates. San Francisco.

A compilation of Spanish missionary activities and journeys in Lower and Upper California written prior to August, 1784. The parts
referring to Lower California contain little of value concerning the country or its inhabitants. :

1874, Scammon, CHARLES Mztvitz. The Marine Mammals of the Northwestern Coast of North America. Pp. 318,
with plates and cuts, quarto. San Francisco and New York.
Contains a general account of the habits of the species, with methods of hunting them, especially along the west coast of Lower Cali-

fornia. Many of these notes were previously published in the Proceedings of the Philadelphia Academy of Natural Sciences for 1869,
pp. 13-63. These notes contain about all we know concerning the life histories of a number of species of marine mammals in this region.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 151

1876. Pater, Epwarp. (See Watson.)
Notes on plants of Guadalupe Island.

1876. Ripeway, Rogert. Ornithology of Guadalupe Island, Based on Notes and Collections Made by Dr. Edward
Palmer. <Bull. U. 8. Geol. and Geog. Surv. Terr., II, pp. 183-195.
This was the first collection of birds from the island. Nine species are listed, of which eight are described as new, and some inter-
esting notes on their habits are given.
1876. Watson, SERENO. Botanical Contributions.
1. On the Flora of Guadalupe Island, Lower California, pp. 105-112.
2. List of a Collection of Plants from Guadalupe Island made by Dr. Edward Palmer, with his Notes upon
Them, pp. 112-121.
3. Descriptions of New Species of Plants, Chiefly Californian, with Revisions of Certain Genera, pp. 121-148.
<Proc. Amer. Acad. Artsand Sci., XI, 1875, pp. 105-148.
Contains interesting general notes on the climate and physical character of Guadalupe Island and its flora based on the collections

and notes made by Dr. Palmer in 1874. The list contains 119 species of plants, including a number of new species, and notes by the
collector.

1877. Ripaway, Ropert. The Birds of Guadalupe Island, Discuesed with Reference to the Present Genesis of
Species. <Bull. Nutt. Orn. Club., II, pp. 58-66, July.
A general discussion of the origin and characters of the island forms.

1877. Strents, Taomas H. Contributions to the Natural History of the Hawaiian and Fanning Islands and Lower

California. <(Bulletin U. §. Nat. Mus., No. 7, Ornithology, pp. 9-33; Herpetology, pp. 35-42.
Contains notes on species of birds and reptiles observed on the coasts and islands of Lower California in 1874-75.

1879, FERREL, BArToLomsé. Voyage of Cabrillo in 1542. <U. 8S. Geographical Surveys west of the 100th Meridian,
Vol. VII, Archaeology, pp. 299-314. Translated from the original by Richard Stuart Evans.
A translation of a diary of Cabrillo’s voyage along the west coast of Upper and Lower California.

1882. Parry, C.C. A New North American Rose. <Bull. Torr. Bot. Club., IX, pp. 97-98, August.
Includes original description by Engelmann of this interesting species, which was discovered by Parry, Pringle, Orcutt, and Jones
at Todos Santos Bay. .
1882, Ripeway, Ropert. On Two Recent Additions to the North American Bird Fauna, by L. Belding. <Proc.
U. S. Nat. Mus., IV, pp. 414-415, April 25.
First records of the occurrence of Motacilla ocularis and Dendroica vieilloti bryanti from La Paz. (The latter afterwards described
as D. bryanti castaneiceps, Ridg.)
1882, —__———. Descriptions of Some New North American Birds. <Proc. U. 8. Nat. Mus., V, pp. 344-346,
September 5.
Contains original descriptions of Geothlypis beldingi and Rallus beldingi.

1883. Betpine, L. [and Rosrerr Ripeway]. Catalogue of a Collection of Birds Made at Various Points Along the
Western Coast of Lower California, North of Cape St. Eugenio. <Proc. U. 8. Nat. Mus., V, pp. 527-532,

March 21. [Edited by Robert Ridgway.]
A briefly annotated list with critical comments on some species by Robert Ridgway.

1883. Betpine, L. [and Ropert Ripaway]. Catalogue of a Collection of Birds Made Near the Southern Extremity
of the Peninsula of Lower California. <Proc. U. 8. Nat. Mus., V, pp. 532-549, March 21. [Edited by
Robert Ridgway.]
A briefly annotated list with critical notes on some of the species by Robert Ridgway. A footnote contains a list of 34 species
collected by Xantus in the cape district, which were not collected by Belding.
1883. Betpine, L. Second Catalogue of a Collection of Birds Made Near the Southern Extremity of Lower California.
[Edited by R. Ridgway.] <Proc. U. 8. Nat. Mus., VI, pp. 344-352, December 27.
A briefly annotated list of species observed between December, 1882, and March 23, 1883, in the cape district, with critical notes by
Robert Ridgway.
1883. Ripeway, Ropert. Descriptions of Some New Birds from Lower California, Collected by Mr. L. Belding. <Proc.

U. S. Nat. Mus., VI, pp. 154-156, October 5.
Original descriptions of Lophophanes inornatus cineraceus, Psaliriparus grindae, and Junco bairdi, all from the cape district.

1883, —_—_——-. Anthus cervinus (Pallas) in Lower California. <(Proc. U. 8. Nat. Mus., VI, pp. 156-157, October 5.
Records a specimen collected by Belding at San Jose del Cabo, January 26, 1883.
1883, ——————._ Note on Merula confinis (Baird). <(Proc. U. 8. Nat. Mus., VI, pp. 158-159, October 5.

Records the collections of the first specimen subsequent to the type, by Belding in the Laguna Mountains, February, 1883.

1884. Gray, Asa. Veatchia Nov. Gen. Anacardiacearum. <Bull. Calif. Acad. Sci., I, pp. 4-5,. February.
Description of Veatchia cedrosensis [=Pachycormus discolor].

1884, Ketzoaa, A. New Species of Cedros Island Plants. <Bull. Calif. Acad. Sci., I, pp. 6-7, February.
Descriptions of two new species collected by Dr. J. A. Veatch.
1884, Ten Katz, H. Matériaux pour servir 4 ]’Anthropologie de 1a presqu’ Isle Californienne. <Bull. de la Soc.
d’ Anthropologie, separately paged 1-23, with cuts, July, Paris.
A brief paper on the Indians once occupying southern Lower California, with cuts and notes on skulls collected by the author in 1883
on Espiritu Santo Island and elsewhere in that region.
152 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI

1885. Betpine, L. The Piricue Indians. <The West American Scientist, I, No. 4, pp. 21-22, March.

Notes on the Indians of the cape district and their method of burial. These notes refer to the same skulls described by Ten Kate
in paper mentioned above.

1885. GREENE, Epwarp Les. Studies in the Botany of California and Parts Adjacent, II.
1. Three New Genera, pp. 179-182.
2. Miscellaneous Species, mostly new, pp. 182-214.
3. Notes on Guadalupe Island, pp. 214-220.
4. A Calalogue of the Flowering Plants and Ferns of Guadalupe Island, pp. 220-228. <(Bull. Calif. Acad.
Sci., I, No. 4, pp. 179-228, August. ,

Includes notes and descriptions of new species from the coast and. islands of Lower California.

1885. GrEENE, Epwarp L. Botany of the Coronados Islands. <The West American Scientist, I, No. 10, pp. 69-71,
October.

A list of nearly 70 species, one of which ( Malacothriz insularis) is mentioned as new.

1885. Parry, C. C. New Genus of Euphorbiaceae from Lower California. <The West American Scientist, I, pp.
13-15, February.
Original description of the genus Tetracoccus.

1885. Ramos, Joaquin M. Trabajos Ejecutados por la Comision exploradora de la Baja California, el afio de 1884,
Anales del Ministerio de Fomento, 8vo, Mexico, 1885.
Mainly a report on placer and other mines but with general information about topography, towns, etc.
1885. Ripeway, Ropert. A Review of the American ‘‘Golden Warblers.”? <Proc. U.S. Nat. Mus., VI, pp. 348-350,
September 2.
Contains original description of Dendroica bryanti castaneiceps Ridgway, from La Paz, Lower California.
1886. Bryant, Water E. CerrosIsland. <Forest and Stream, XXVII, No. 4, pp. 62-64, August 19.

Contains a general account of the island with an annotated list of 27 species of birds noted there from January 7 to 11,1885. A deer
and a rabbit were also noted.

1886. Orcutt, C. R. Northern Lower California. <(West American Scientist, II, pp. 37-41, June.
A general description of the country along the route from San Diego to El Rosario.

-

 

1886. ——. New Lower Californian Cactuses. <The West American Scientist, II, No. 16, p. 46, with plate,
June.
Original descriptions of Echinocactus orcuttii and Cereus phoeniceus pacificus from northern part of the peninsula.
1886. ——————. A Botanical Trip. <(West American Scientist, II, pp. 53-58, July.

Contains interesting notes on a number of plants met with on a journey from San Diego, Calif., to near San Fernando, Lower Cali-
fornia, in spring of 1886.
1886. Ramos, Joaquin M. Informe relativo a los Trabajos Ejecutados por la Comision Exploradora de la Baja Cali-
fornia, en el Afio de 1884 Presentado a la Secretaria de Fomento, Mexico. 8vo, pp. 224, with maps and cuts.
Contains general information concerning the middle of the peninsula from Mulege to Ojo de Liebre, with notes on geology and physi-
ography; also account of the Calmalli placers, E] Boleo copper mine at Santa Rosalia, the pearl fishery, and the orchilla industry.
1886. TowNSEND, CHarLEs H. An Account of Recent Captures of the California Sea Elephant, and Statistics Relating
to the Present Abundance of the Species. <Proc. U. 8. Nat. Mus., VIII, pp. 90-93, 1885.

An account of a trip along the coast of Lower California, October to December, 1884, when specimens were secured at San Cristobal
Bay, with some historical notes.

1886. Vasey, Georce. A-New Genus of Grasses. <The West American Scientist, III, No. 20, pp. 4-6, December.
Description of Orcuttia californica—new genus and species from near San Quintin, Lower California.

1887. ANonymous. A Beautiful Peninsula: The Government and People of Lower California, etc. <Sketch of the
settlement and exploration of Lower California [in supplementary appendix to Resources of the Pacific Slope
by J. Ross Browne], pp. 174-177, New York.

An unsigned letter dated La Paz, December 22, 1867, giving information concerning local conditions, government, products, and
colonization companies.

1887. .. . Lower California, the Peninsula. 8vo folder of 24 unnumbered pages. New York City.
This is a publicity folder of the International Co. of Mexico in which the lands held by the company are described in glowing terms to

attract colonists. These efforts to colonize failed completely.
1887. Brtpine, L. Collecting in the Cape Region of Lower California. <The West American Scientist, III, No. 24,
pp. 93-97, April.
General notes on the region and a naturalist’s experiences there from 1881-83.
1887. ——————._ Reptiles of the Cape Region of Lower California. <The West American Scientist, III, pp. 97-99,
April. *
A briefly annotated list of 54 species.
1887. Brownz, J. Ross. Resources of the Pacific Slope. A statistical and descriptive summary of the mines and
minerals, climate, topography, agriculture, commerce, manufactures, and miscellaneous productions of the
States and Territories west of the Rocky Mountains with a sketch of the settlement and exploration of Lower

California. Pp. 1-678; 1-200, 8vo.. New York.
The last 200 pages of this volume contain papers on Lower California by several authors, which are cited elsewhere in this list.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 153

1887. Bryant, Waiter E. Additions to the Ornithology of Guadalupe Island. <Bull. Calif. Acad. Sci., II, pp.

269-318, January.
An annotated list 0 {85 species observed by the author during parts of 1885-86 ,with notes on the physical character and climate of the

island.
1887. —————. Description of a New Subspecies of Petrel from Guadalupe Island. <Bull. Calif. Acad. Sci., I,
p. 450, July.

Original description of Oceanodroma leucorhoa macrodactyla.

1887. Gass, Wuiu1am M. Exploration of Lower California. <A sketch of the settlement and exploration of Lower
California [in supplementary appendix to Resources of the Pacific Slope, by J. Ross Browne], pp. 82-122,
New York. :
An account of an expedition overland from Cape San Lucas to San Diego, Calif., in 1867, with notes on the country and its resources.
1887. Hawxs, J.D. Journal of the Expedition of Mr. J. D. Hawks and Party Through the Interior of the Peninsula of
Lower California, from San Domingo to San Diego. <A sketch of the settlement and exploration of Lower
California [in supplementary appendix to Resources of the Pacific Slope, by J. Ross Browne], pp. 132-142,
New York.
Thisis a briefitinerary ofa trip made between August and October, 1849, by J. D. Hawks, Daniel T. Hulett, George H. Davis, John C.
Gilsey, Henry M. Smith, and J. J. Ellis. :
1887. Lopatrcx1, Maurice. Lower California Notes. Reprint from San Diegan of May 14. <The West American
Scientist, III, pp. 129-130, May.
Notes on the flora and topography of the west coast of the peninsula south of Santa Rosalia Bay. The English company located in
San Andres Valley is mentioned here.
1887. Orcutt, C.R. A Lemonade and Sugar Tree. <The West American Scientist, III, pp. 45-47, February.
Notes on sugar producing species of Rhus in Lower California.
1887. —_———. The Oaks of Southern and Lower California. <(The West American Scientist, III, pp. 185-139,
with photograph, May.
Includes brief notes on the species of San Diego County and northern Lower California.
1887. —_-__——. Notes on Lower California Plants. <(The West American Scientist, III, pp. 189-140, May.

Contains a brief list of species with Mexican names collected in San Andres Valley and near Santa Rosalia Bay by Maurice Lopatecki
during the survey of the west coast of the peninsula by the U.S.S. Ranger.

1887. OrtEGa, José. Historia del Nayarit, Sonora, Sinaloa y Ambas Californias. Pp. I-IX, 564, 8vo, new ed.
Mexico (first ed. Barcelona, 1854).
An account of missionary work during the early Spanish occupation, including trips of Kino to northern Sonora and mouth of the
Colorado, and of Consag north from Mulege along the gulf coast.
1887. Ranpawt, A.G. Extracts from a History of Old or Lower California. <A sketch of the settlement and explora-
tion of Lower California [in supplementary appendix to Resources of the Pacific Slope by J. Ross Browne], pp.

155-172, New York.
This is a translation of parts of Clavijero’s work cited elsewhere.

1887. Ripaway, Roserr. Description of Two New Races of Pyrrhulovia sinuata Bonap. <Auk, IV, p. 347, October.
Original description of P. s. peninsulz.

1887. Scammon, Capt. C. M. Report of Capt. C. M. Scammon, of the United States Revenue Service, on the West
Coast of Lower California. <(A sketch of the settlement and exploration of Lower California [in supplementary
appendix to Resources of the Pacific Slope, by J. Ross Browne], pp. 123-131, New York.

An account of the coast and islands on the west side of the peninsula previous to 1867, with notes on the former whaling and sealing
industries.

1887. SzcRETARIO DE FomENnTo. Exposicion que hace el Secretario de Fomento sobre la Colonizacion de la Baja
California. Pp. 113, 8vo. Mexico.

An official report by the Secretary of the Interior of Mexico concerning grants for colonization purposes to Adolfo Bulle, G. Luis
Miiller, and others in Lower California and elsewhere in Mexico, in reply to criticism concerning these acts.

1887. TAYLOR, ALEXANDER 8, Historical Summary of Lower California, from Its Discoveryin 1832to 1867. <Asketch
of the settlement and exploration of Lower California [in a supplementary appendix to Resources of the Pacific
Slope, by J. Ross Browne], pp. 6-77, New York.

The history and resources of the peninsula are here compiled and to much of value considerable misinformation is added.

1887. Veatcu, Jonn A. Report of Dr. John A. Veatch on Cerros or Cedros Island. <A sketch of the settlement and

exploration of Lower California [in supplementary appendix to Resources of the Pacific Slope, by J. Ross

Browne], pp. 143-154, New York.
Contains a general description of the island with interesting notes on its natural history from observations made during a visit extend-

ing from June to August, 1859.

1888. Brewster, Win11am. Descriptions of Supposed New Birds from Lower California, Sonora and Chihuahua,

Mexico, and the Bahamas. <Auk, V, pp. 82-95, January.
Ardea virescens frazari, Haematopus frazari, Columba fasciata vioscze, and Megascops aspersus are described from southern Lower

California.
154 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1888. Brewster, Wiu1am. On Three Apparently New Subspecies of Mexican Birds. <Auk, V, p. 136, April.
Original description of Glaucidium gnoma hoskinsi from the cape district.

1888. Bryant, Waiter E. Description of a New Subspecies of Song Sparrow from Lower California, Mexico. <Proc.
Calif. Acad. Sci., 2d ser., I, pp. 197-200, September.
Describes Melospiza jaaclais rivularis from Comondu.

1888, Curran, Mary K. Botanical Notes. I. Plants from Baja California. <(Proc. Calif. Acad. Sci., 2ser., I, pp.
227-237, December 11.
A list of about 100 species collected by Walter E. Bryant in the vicinity of Magdalena Bay early in 1888, accompanied by critical
notes,

1888. GREENE, Epwarp Ler. The Botany of Cedros Island. <Pittonia, I, pp. 194-200, June.
General notes on the botany of the island made during a three days’ visit the latter part of April, 1885.

1888, ——__——_._ A List of the Known Species of Cedros Island Plants. <(Pittonia, I, pp. 200-208, June.
, An annotated list of 82 species.

1888. LinparEn, W. Notes on the Geology of Baja California, Mexico. <(Proc. Calif. Acad. Sci., 2d ser., I, pp.
173-196, with map and diagram, September 25.
Contains some interesting notes on the topography and formation of the northern end of the peninsula,

1888. NorpHorr, CHaRLEs. Peninsular California. Pp. 130, with plates and maps, 8vo. New York.
Gives a general account of Lower California, especially the northern part, mainly in the interest of a colonization company. Much
erroneous matter is given concerning resources and agricultural possibilities. One of the colonizing officials is quoted to the effect that
there are ‘hundreds of moose”’ in the mountains.

1888. Orcutt, C. R. The Gold Fields of Lower California. <The West American Scientist, V, pp. 4-5, October.
A notice of mining activity brought about by the discovery of the Santa Clara Placers.
1889. ANTHONY, ALFRED W. New Birds from Lower California, Mexico. <(Proc: Calif. Acad. Sci., 2d ser., II, pp.
73-82, October 11.
Describes new birds taken during a visit to the San Pedro Martir Mountains the spring of 1889, with notes on distribution and habits,
1889, AnrHony, A. W. [edited by Watrzr E, Bryant]. General Ornithological Notes. I. Nesting Habits of the
California Brown Pelican (Pelecanus californicus). <(Proc. Calif. Acad. Sci., 2d ser., II, pp. 83-85, October 11.
From observations made on theislands along the western coast of Lower California.
1889. —————. Occurrence of Phzxthon xthereus and Capture of Scolecophagus carolinus in Lower California. <Proc.
Calif. Acad. Sci., 2d ser., II, p. 86.

1889. Bancrorr, H. H. History of the Pacific States of North America. XI, pp.I-VI, pp. 1-814, 8vo. San

Francisco.
Chapters 29 and 30 contain a brief history of Lower California from 1800 to 1888, including Walker’s filibustering expedition and the
projects of numerous colonization companies.

1889. Betprve, L. The Deer of Southern Lower California. <The West American Scientist, VI, No. 43, pp. 26-27,
May.
Remarks on the large size of the deer of the region; estimates that one shot by him weighed 250 pounds and that they sometimes
exceed 300 poundsin weight. Calls attention to the erroneous idea of the size of the cape deer by the statement by Caton in his
Antelope and Deer of America, pp. 119 and 337 (2d edition).
1889. Branpecexr, TownsHEenpd S. A Collection of Plants from Baja California, 1889. <(Proc. Calif. Acad. Sci.,
2d ser., II, pp. 117-215, November.
Contains an annotated list of over 700 species of plants collected about Magdalena Bay and thence on an overland trip to San Quintin.
The introductory remarks contain interesting general information and descriptions of new species. An important paper on the flora of
the peninsula.
1889. Brewster, Wu1am. Descriptions of Supposed New Birds from Western North Ame-ica and Mexico. <Auk,
VI, pp. 85-98, April.
Original descriptions of Empidonaz griseus and Progne subis hesperia from Lower California.
1889. Bryant, Waiter E. Descriptions of the Nests and Eggs of Some Lower California Birds with a Description of
the Young Plumages of Geothlypis beldingt. <(Proc. Calif. Acad. Sci., 2d ser., II, pp. 20-2‘.
Notes from the southern part of the peninsula.
1889. —————._ Provisional Descriptions of Supposed New Mammals from California and Lower California. <Proc.
Calif. Acad. Sci., 2d ser., II, pp. 25-27, June 20.
Original description of Spermophilus grammurus atricapillus from Comondu.
1889. —————-. A Catalogue of the Birds of Lower California,  Merlen, <Proc. Calif. Acad. Sci., 2d series, PH,
pp. 237-320, with maps, December.
A list 0f320 species and subspecies annotated with the observations of the author and others to date, with notes on the work done
and places visited and e short bibliography.
1889. Greenz, Epwarp Lez. The Vegetation of the San Benito Islands. <Pittonia, I, pp. 261-262, March.
General remarks on the islands and their flora.

1889. —————._ A List of San Benito Plants. <(Pittonia, I, pp. 263-266, March.
Twenty-four species collected by Lieut. Charles F. Pond are enumerated, of which three are described as new.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 155

1889. Greene, Epwarp Lez. A Supplementary List of Cedros Island Plants. <(Pittonia, I, pp. 266-269, March.
Based on a small collection of plants made by Lieut. Charles F. Pond during the winter of 1888-89, _ Several species are described as
new.

1889. —————. Some Plants from the Bay of San Bartolome, Lower California. <(Pittonia, I, pp. 287-292, April.
Twenty-four species are enumerated, including a number described as new, all from a collection made by Lieut. Charles F. Pond, in
March, 1889,
1889. Harkness. H.W. Fungi Collected by T. 8. Brandegee in Lower California in 1889. <(Proc. Calif. Acad.
Sci., 2d ser., II, pp. 231-232, December 20.
A list of species collected on an overland journey from Magdalena Bay tc San Quintin, and original description of Puccinia ornata.
1889. Linparen, W. Petrographical Notes from Baja California, Mexico. <Proc. Calif. Acad. Sci., 2d ser., IT,
pp. 1-17, June.
These notes are based mainly on rocks from the northwestern part of the peninsula.
1889. Muispaues, C.F. Contributions to North American Euphorbiacee. I. Upon a Collection of Euphorbiaceous
Plants, made by Mr. T. S. Brandegee, in 1889, on the Mainland of Lower California and the Adjacent Islands
of Magdalena and Santa Margarita. <(Proc. Calif. Acad. Sci., 2d ser., II, pp. 217-230, December 17.

With notes and descriptions of new species,

1889. Orcutt, C.R. A New Species of Cereus. <The West American Scientist, VI, p. 29, May.

Description of Cereus cochel from Todos Santos Bay to San Fernando.

1889, —————. Floraofthe Alamo. <The West American Scientist, VI, pp. 132-134, October.
Botanical notes, written on a horseback trip in the vicinity of Alamo, Lower California.
1889. Parry, 0. C. A New Species of Eriogonum (EF. fastigiatum), from Lower California. <The West American
Scientist, VI, pp. 102-103, September.

1889. Vasey, Gzoras, and JoserH N. Ross. List of Plants Collected by Dr. Edward Palmer in Lower California in

1889. <Proc. U. 8. Nat. Mus., XI, pp. 527-536, September.
Annotated lists of plants collected at San Quintin and Lagoon Head, Lower California.

1889. SancnEz, FactoM. The Truth About Lower California. Pp. 94, with maps, 8vo. San Francisco.
A report with numerous documents concerning colonization in the peninsula, especially concerning difficulties of the Mexican Inter-
national Co. of Colonization and its colonies at Ensenada and San Quintin shortly before this report.

1889. Watson, SERENO. Contributions to American Botany. I. Upon a Collection of Plants made by Dr. E. Palmer,
in 1887, about Guaymas, Mexico, at Muleje and Los Angeles Bay in Lower California, and on the Island of
San Pedro Martin [Martir] in the Gulf of California. <Proc. Amer. Acad. Arts and Sci., XXIV, pp. 36-82.
An important list, including many new species from Lower California. Of the 415 species contained in the collection from both sides
of the Gulf of California, 89, or more than one-fifth, are recorded as new.
1890. AnrHony, A. W. The Nests and Eggs of Townsend’s Junco (Junco townsendi) and San Pedro Partridge (Ore-
ortyx pictus confinis). <Zoe, I, pp. 5-6, March.
The first account of the nesting of these birds in the San Pedro Martir Mountains.

1890... . Illustrated History of Southern California. Quarto, pp. 898, with portraits. Chicago.
Gives a brief historical account of Lower California, but the volume is devoted mainly to short biographies of residents of southern
California, which appears to have been the object of this publication.

1890. BranpEGEE, T. 8. An Arborescent Polygala. <Zoe, I, pp. 4-5, March.
Notes on the tree species of Lower California (Polygala apopetala).

1890. ———_—. A Cardon Forest. <Zoe, I, p. 26, March.
Contains notes on the distribution of the giant cactuses in Lower California.

1890, —————. The Effect of Rain on Fouquieria. <Zoe, I, pp. 40-41, April.
Describes the sudden appearance of leaves after the infrequent rains in Lower California.

1890, —__—_——. A New Perityle. <Zoe, I, p. 54, April.
Description of Perityle cwneata, from Sierra dela Laguna, near Todos Santos, Lower California.

1890, —_—_——. The Hedgerows of Todos Santos. <Zoe, I, pp. 179-181, August.
Contains interesting general notes on the appearance and distribution of various species of trees and shrubs in southern Lower
California.

Notes Concerning the Collection of Plants Made by Xantus at Cape St. Lucas and Vicinity.

<Zoe, I, pp. 269-272, November.
Contains notes on the cape district, its climate, and a number of little-known species of plants.

1890, —_———_-. A New Cottonwood from Baja California. <Zoe, I, pp. 274-275, November.
Original description of Populus monticola, a striking species known only from the Sierra dela Laguna.

1890, ——_——-. A New Nolina. <Zoe, I, pp. 305-306, December.
Original description of Nolina beldingi, am arborescent species peculiar to the upper slopes of the Sierra dela Laguna.

1890, ___—-. Studies in Coreopsides and Tagetinese, Especially of Lower California, with Descriptions of
Some New Species. <Zoe, I, pp. 308-314, December.

1890 Brewster, Wom. A New Subspecies of the Solitary Sandpiper. <Auk. VII, pp. 377-379, October, 1890.
Original description of Totanus solitarius cinnamomeus from San Jose del Cabo.

1890.

 
156 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1890. Bryant, Water E. Notices of Supposed New Birds. <Zoe, I, pp. 148-150, July.
Contains notes on status of Otocoris alpestris pallida and proposes Auriparus flaviceps ornatus as a new name for the verdin of Lower
California, southern California, and southwestern Arizona.

1890. —————.. El Zorillo. <Zoe, I, pp. 272-274, November.
An account of skunks (Spilogale lucasana) causing hydrophobia in the cape region.

1890. Coenraux, A. Cucurbitacearum Novum Genus Et Species. <Proc. Calif. Acad. Sci., 2d ser., III, pp. 58-60,
July 9. ;
The new genus Brandegea, and a new species Echinocystis brandegei, described from Lower California.

1890. Eaton, D.C. A New Fern from Lower California. <Zoe, I, p. 197, September.
Asplenium blepharodes described from the Sierra dela Laguna.

1890. Exzicotr, Joun M. Should We Possess Lower California? <Overland Monthly, 2d ser., XVI, pp. 473-476,
November.
Discusses adversely the value of Lower California to the United States.
1890. LinpereN, WaLpEMAR. Notes on the Geology and Petrography of Baja California, Mexico. <(Proc. Calif.

Acad. Sci., 2d ser., III, pp. 25-33, April.
Notes on the rocks collected by Bryant in the southern part of the peninsula, and comments on the geology of that area.

1890. Merriam, CO. Hart. Contribution toward a Revision of the Little Striped Skunks of the Genus Spilogale.
‘  <North American Fauna No. 4, October 8, pp. 1-16, Pl. I.
Contains original description of Spilogale lucasana from Cape St. Lucas.

1890. Orncutr, C. R. Some Notes on Echinocactus. <Garden and Forest, III, p. 238, May 14. Reprinted in West
American Scientist, VII, pp. 67-70, December.
A brief discussion of the species occurring in southern California and Lower California.

1890. Parmer, Epwarp. (See Vasey, George, and Joseph Nelson Rose, for field notes by Palmer.)

1890. Townsenp, Cuas. H. Birds from-the Coasts of Western North America and adjacent Islands, Collected in
1888-89, with Descriptions of New Species. <Proc. U. 8. Nat. Mus. XIII, pp. 136-138, September.

Contains brief annotated lists from the coast of the peninsula and adjacent islands, including Guadalupe, and the original description
of A mphispiza belli cinerea from Ballenas Bay.

1890. —___——. Reptiles from Clarion and Socorro Islands and the Gulf of California. <Proc. U. 8. Nat. Mus.,

XIII, pp. 134-144, September.
Includes a few species from the gulf coast of the peninsula and from Carmen and Angel de Ja Guardia Islands.

1890. Vasry, Grorez, and JosrrH Netson Ross. List of Plants Collected by Dr. Edward Palmer in Lower Cali-
fornia in 1889. <Contr. Nat. Herb., I, pp. 9-28, July 18.
Annotated list of collections made at Lagoon Head, Cedros Island, San Benito Island, Guadalupe Island, Lower California, and at
Colonia, Lerdo, Sonora, near the head of the Gulf of California. A number of species are described as new.

1890. Vaszy, Grorcx, and J. N. Ross. List of Plants Collected by Dr. Edward Palmer in Lower California and
western Mexico in 1890. <Contr. Nat. Herb., Vol. I, No. 3, pp. 63-92, November.

Collections were made at La Paz, San Pedro Martir and Raza Islands, Santa Rosalia, and Santa Agueda, Lower California. The list
includes notes and descriptions of a new genus and many new Species.

1891. antHony, A. W. Secondary Migration of Birds. <Zoe, I, pp. 379-381, February.

Includes notes on the migration of Harporhynchus cinereus [= Torostoma cinerea mearnsi] at San Quintin, Lower California.

1891. Branpecez, T. 8S. A New Species of Esenbeckia (with plate). <Zoe, I, p. 378, February.
Describes Esenbeckia flava from San Jose del Cabo. ar

1891__—_-—-._ The Plants Pecular to Magdalena and Santa Margarita Islands. <Zoe, II, pp. 11-12, April.
Contains notes on the physiography of the islands and the following species are given as peculiar to them; Gongylocarpus fruticulosus,
Mamillaria halei, Agave margaritee, Brickellia hastate, and Viguiera subincisa.

1891. ——————. Cactaceae of the cape region of Baja California. <(Zoe, II, pp. 18-22, April.
Includes notes on the distribution of 12 species, and original descriptions of the following: Afamiliaria roseana, Cereus striatus, and
Opuntia rotundifolia.

4891. —————-. Drymaria in Baja California. <Zoe, II, pp. 68-70, April.
Fight species are listed, including two described as new.

1891. ——————. Field Notes on the Plants of Baja California. <Zoe, II, pp. 145-152, July.
Contains interesting notes on the distribution, habits, and economic uses of a number of species in the southern part of the peninsula.

1891, ——_——. Flora of the Cape Region of Baja California. <(Proc. Calif. Acad. Sci., 2d ser., III, pp. 108-182,
July and August.

A brief description of the cape region and a list of the plants known to occur from La Paz and Todos Santos southward based upon
two collections made in that part of the peninsula in 1890, Nearly 700 species are enumerated, including some collected by Dr. Palmer
at La Paz and by Xantus at Cape San Lucas, that were not found by Brandegee. Many new species are described and others are
accompanied by field or critical notes. One of the most important papers yet published on plants of this region.

1891. BREWSTER, Wruram. Descriptions of Seven Supposed New North American Birds. <Auk, VIII, pp. 139-149,
April.
Original descriptions of Contopus richardsoni peninsulz, Pipilo maculatus magnirostris, Vireo solitarius lucasanus, and Sitta carolinen-
sis lagune, from the cape region of Lower California.
No. 1] LOWER CALIFORNIA. NATURAL RESOURCES—NELSON. 157

1891. Bryant, Watrer E. Andrew Jackson Grayson. <Zoe, II, pp. 34-68, April.
A good biographical sketch of Grayson, including his work in Mexico and notes on the life histories of some species of Mexican birds
from his manuscript.

1891. —__——_. Preliminary Description of a New Subspecies of the Genus Lepus, from Mexico. <(Proc. Calif.
Acad. Sci., 2d ser., III, p. 92, April 23.

Original description of Lepus insularis from Todos Santos Island.

1891. —_———_._ The Cape Region of Baja California. <Zoe, II, pp. 185-201, October.

A discursive account in which the cape region is defined as that part of the peninsula southward from the base of the mountains
between La Paz on the gulfshore and Todos Santos on the Pacific coast; contains author’s itinerary and interesting notes on the country
and the birds, mammals, and plants; also includes an account of the discovery of Lepus insularis, the black jack rabbit of Espiritu
Santo Island.

1891. Hempnitt, Henry. Edible Mollusks of Western North America. <(Zoe, II, pp. 134-139, July.

Includes species living along the coast of Lower California.

1891. KrELER, CHarues A. Notes on the Colors of West Coast Mammals. <Zoe, II, pp. 203-216, October.
Contains notes on the Lower California ground squirrel and the black jack rabbit (Lepus insularis). The notes regarding the resem-
blance of the latter to its surroundings are entirely erroneous.

1891. Mituspaueau, C. F. Euphorbiacez Collected by T. S. Brandegee, Principally in the Vicinity of Todos Santos,
Baja California, January and February, 1890. <(Zoe, I, pp. 346-348, January.

Sixteen species listed, two described as new.

1891. Orcutt, C. R. <A Visit to Lake Maquata. <The West American Scientist, VII, pp. 158-164, May.
Includes a few general and botanical notes on the region about this lake along the western base of the Cocopah Mountains.

1891. ——_——-._ Rosa minutifolia. <The West American Scientist, VII, pp. 181-182, May.
Notes on the discovery, distribution, and habits of this species.
1891. ——————. The Guadalupe Palm. <The West American Scientist, VII, p. 182, May.

Notes on Erythea edulis of Guadalupe Island, reprinted from Vick’s Magazine, XIV, p. 168.

1891. TowNnsEND, CHaRtEs H. Report upon the Pearl Fishery of the Gulf of California. <(Bull. U. 8. Fish Com-
mission, IX, pp. 91-94, 3 illustrations (1889).
A brief report based on the operations of the Pear] Shell Co. of Lower California along the southern shores of the peninsula and western
Mexico to the border of Guatemala.
1892. Benprire, Cuartes. Life Histories of North American Birds with Special Reference to their Breeding Habits
and Eggs. [Part1.] Smithsonian Contributions to Knowledge, XXVIII[ Special Bull. U. 8. Nat. Mus.No. 1]
pp. X, xii; 446; with 12 colored plates, quarto. Washington.

Contains the life histories of several Lower Californian birds with some original data.

1892. BRaNDEGEE,T.S. The Distribution of the Flora of the Cape Region of Baja California. <Zoe, III, pp. 223-231,

October.
An important paper including a comparative table based on 732 species showing in condensed form the distribution of the flowering
plants and ferns and their relation to the floras of neighboring regions, especially that of the Mexican mainland.

1892, —___—. A New Rumfordia from Lower California. <Zoe, III, pp. 241-242, with plate, October.
Description of Rumfordia connata. ‘
1892, -_————. Additions to the Flora of the Cape Region of Baja California. < Proc. Calif. Acad. Sci., 2d ser.,

III, pp. 218-227, November 10.
The additional species, some of them described for the first time, bring the total number listed for this region up to 739.

1892. Cooper, J. G. Catalogue of the Land and Fresh Water Mollusca of Lower California. Zoe, III, pp. 12-25,
April.
Gives notes on the distribution of 45 species.
1892-95. ——————. On Land and Fresh Water Shells of Lower California. <(Proc. Calif. Acad. Sci., 2d ser.,
III, pp. 99-103 (Apr. 23, 1892); 207-217 (Oct. 6, 1892); 338-344 (May 5, 1893); 2d ser., IV, pp. 180-148, with
plates (Apr. 28, 1894); 2d ser., V, pp. 163-165 (Jan. 8, 1895).

Contains general notes on the country with lists of species, descriptions, and annotations, including altitudes by Eisen.

1892. Merriam, C. Hart. The Geographic Distribution of Life in North America, with Special Reference to the
Mammalia. <Proc. Biol. Soc., Washington, VII, pp. 1-64, April.

A historical résumé of the life areas proposed by various authors and discussion and definition of the life zones of North America
including northern Mexico and Lower California.

1892. Rosz, J oserH NELSON. List of Plants Collected by Dr. Edward Palmer in 1890 on Carmen Island. <Contr.
Nat. Herb., I, No. 5, pp. 129-134, June 30.

An annotated list of 68 species, including several new ones, most of which are known from the adjacent peninsula.

1892. TownsEND, CHartes Hasxins. The Pearl Divers of the California Gulf. <The Californian Magazine, II,
pp. 116-125, with 7 illustrations, June.
‘A popular account of the pearl fisheries in the Gulf of California.

1893. ArtEN, J. A. The Geographical Origin and Distribution of North American Birds Considered in Relation to
Faunal Areas of North America. <Auk, X, pp. 97-150, April.
Contains some discussion of the faunal relationships of Lower California and applies the term Saint Lucas Fauna to the life of the
area south of La Paz.
158 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1893. AtnEeNn, J. A. On a Collection of Mammals from the San Pedro Martir Region of Lower California with Notes on
Other Species, Particularly of the Genus Sitomys. <(Bull. Amer. Mus. Nat. Hist., V, pp. 181-202, August 18.
An annotated list with descriptions of four new species and subspecies collected by E. C. Thurber and A. W. Anthony.
1893. AnrHony, A. W. Birds of San Pedro Martir, Lower California. <Zoe, IV, pp. 228-247, October.
Notes on the physiography of the region and an annotated list of 121 species and subspecies of birds.
1893. Branprces, T.S. Southern Extension of California Flora. <Zoe, IV, pp. 199-210, October.

General notes on the physiography of Lower California and its floral areas and relationships, with an annotated list of species collected
on.a trip from San Diego to the San Pedro Martir Mountains, in May, 1893.

1893. T.S. B.[RanpEGEn]. Perityle rotundifolia (Benth.). <Zoe, IV, p. 210, October.
A brief note on this species and the genus Amauria from San Quintin.

1893. Bryant, Watter E. A Rookery of Man-of-war Birds. <Nidiologist, I, pp. 1-3, September.
An account of the colony nesting in the mangroves on the east side of Santa Margarita Island.

1893. Eaton, D.C. A New Station for Asplenium septentrionale. <Zoe, IV, p. 185, July.
Recorded from the San Pedro Martir Mountains.
1893. Ecxretpr, J. W. List of Lichens from California and Mexico, Collected by Dr. Ecward Palmer from 1888 to
1892. <Contr. Nat. Herb., I, No. 8, pp. 291-292, October.
Includes several species from Lower California.
1893. G.E.[=Gusrav Eisen]. The Investigations of the Collections of the Expedition to Baja California. <Zoe,

III, pp. 373-375, January.
A brief notice of recent collections made in Lower California for the California Academy of Seiences and the plans for publishing on

them.
1893. Francesco, F. Erythea edulis at Home. <The Gardener’s Chronicle (London), 3d ser., XIII, pp. 507-508,
April 29.
An interesting account of this palm and its surroundings on Guadalupe Island. Observations made in January, 1893.
1893. ——————. Notes on the Flora of Guadalupe Island. <Zoe, IV, pp. 130-139, July.

Some interesting notes on the physiography of the island, with an annotated list of 60 species of plants collected by the author in De-
cember, 1892, and January, 1893.
1893. Merriam, C. Harr. Descriptionof a New Kangaroo Rat from Lower California (Dipodomys merriami melanurus
subsp. nov.) Collected by Walter E. Bryant. <Proc. Calif. Acad. Sci., 2d ser., III, pp. 345-346, January 5.
Type from San Jose del Cabo.

1893. ———_———. The Faunal Position of Lower California. Auk, X, pp. 305-307, July.
Criticism of terms used by Dr. J. A. Allen in his paper in the Auk, 1893, pp. 97-150, with reply by Dr. Allen.

1893, Merritt, Gzo. P. The Onyx Marbles: Their Origin, Composition, and Uses, Both Ancient and Modern. Ann.
Rept. U. 8. Nat. Mus., pp. 541-585, pls. 1-18.
Contains a detailed account (pp. 572-574, pls. 16-18) of the stone belonging to this class at the Tule Arroyo,some 150 miles south of San
Diego. :
1894. AntHony, A. W. Notes on the Genus Heleodytes, with a Description of a New Subspecies. <Auk, XI, pp.
210-214, July.
Contains critical remarks on Heleodytes affinis and original description of H. brunneicapillus bryanti.
1894. BranpEGEE, T. 8. Additions to the Flora of the Cape Region of Baja California, II. <Zoe, IV, pp. 398-408,
March.
Contains descriptions of new species and notes on a number of others not previously listed from this region.
1894. Bryant, Watrer E. Nesting of the Osprey. <The Nidiologist, II, pp. 17-19, with 2 cuts, October.
An account of the nesting of the osprey on Santa Margarita Island.

1894. Emmons, 8S. F., and G. P. Merritt. Geological Sketch of Lower California. <Bull. Geol. Soc. Amer., V,
pp. 489-514, pl. 19, April.
A good résumé of the knowledge available on the date of publication with original contributions by the authors from personal ob-
servations. ‘
1894. FernaLp, M. L. Notes from the Gray Herbarium. <Zoe, IV, pp. 379-380, February.
Contains original description of Habenaria lucecapensis from the Cape Mountains.
1894. Lamson-Scripner, F. Lower California Grasses: An Enumeration of the Grasses Collected by Mr. T. 8. Bran-
degee in Lower California in 1892. <Zoe, IV, pp. 385-393, February 26.
An annotated list of 69 species.
1894. Murriam, C. Hart. Description of Four New Pocket Mice from Lower California. Collected by Walter E.
Bryant. Proc. Calif. Acad. Sci., 2d ser., IV, pp. 457-462, September 25.
Al] from Comondu and southward, and including Perognathus bryanti, P. margarite, P. spinatus peninsulz, and P. arenarius.
1894. Reeve, Jamzs Knarr. The Peninsula of Lower California. <Lippincott’s Magazine, LIII, pp. 71-78,

January,.
A popular account of conditions in the peninsula, especially the northern part, with an optimistic view of the agricultural possibilities

not borne out by experience.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 159

1894. Van Densurca, Joun. Descriptions of Three New Lizards from California and Lower California, with Notes
on Phrynosoma blainvillii. <Proc. Calif. Acad. Sci., 2d ser., IV, pp. 296-301, July 12.
Original descriptions of Uta microscutata and Cnemidophorus stejnegeri from Lower California.
1894. —_———. Notes on Crotalus mitchellii and ‘‘Crotalus pyrrhus.”. <Proc. Calif. Acad. Sci., 2d ser., IV, pp.
450-455, September 25.
Proves that Crotalus pyrrhus is a synonym of C. mitchellii, with notes on Lower Californian specimens.
1894. —__—_——. Phrynosoma solaris, with a Note on Its Distribution. <Proc. Calif. Acad. Sci., 2d ser., IV, p.
456, September 25.
Its known range to Las Animas Bay, Lower California.

1895. AntHony, A.W. A New Subspecies of Harporhynchus from Lower California. <Auk, XII, pp. 52-53, January.
Original description of H. cinereus mearnsi.

1895. ——————. Description of a New Pipilo from Southern and Lower California. <Auk, XII, pp. 109-112, April.
Pipilofuscus senicula described from San Fernando, Lower California.
1895. —_—_———. Birds of San Fernando, Lower California. <Auk, XII, pp. 134-143, April.
Contains notes on the physiography and vegetation and an annotated list of birds.
1895. —————. Junco hyemalis shufeldti in Lower California. <Auk, XII, p. 183, April.
Notes on occurrence of this and other juncos in the peninsula.
1895. ——————._ The Status of Heleodytes affinis. <Auk, XII, p. 280, July.
Showing the subspecific relationship with H. brunneicapillus.
1895. ———_——. New Races of Colaptes and Passerella from the Pacific coast. <Auk, XII, pp. 347-349, October.

Original description of Colaptes chrysoides brunnescens, from Lower California.

1895. Coorrr, J. G. Catalogue of Marine Shells, Collected Chiefly on the Eastern Shore of Lower California for the
California Academy of Sciences during 1891-92. <(Proc. Calif. Acad. Sci., 2d ser., V, pp. 34-48, May 21.

A brief account of previous collections made in the gulf and a list of 191 species, with brief annotations.

1895. Ersen, Gustav. Explorations in the Cape Region of Baja California in 1894, with References to Former Ex-
peditions of the California Academy of Sciences. <Proc. Calif. Acad. Sci., 2d ser., V, pp. 733-775, with 5

maps, October 30.

A general account of the physical characteristics of the area named, with notes on the climate and natural history and a review of the
former expeditions fromthe Academy. A valuable contribution to our knowledge of the area covered, and a general map of the penin-
sula in addition to four others of the cape region.

1895. Goutp, Grorcz H. To the Gulf of Cortez. <(Hunting in Many Lands: The Book of the Boone and Crockett

’ Club, pp. 55-83. .

Contains a popular account of a mountain sheep hunt on the desert mountains east of the head of Trinidad Valley, Lower California.

1895. Mearns, Epaar A., M. D. Description of a New Heron (Ardet virescens anthony) from the Arid Region of
the Interior of North America. <Auk, XII, pp. 257-259, July.

Described from Seven Wells, northeastern Lower California.

1895. RapiKorer, L. Three New Species of Sapindaceze from Western Mexico and Lower California. <(Contr.
Nat. Herb., I, pp. 367-368, January. :

Two are from Lower California.

1895. StowrLt, Joun M. Description of a New Jack Rabbit from San Pedro Martir Mountain, Lower California.
<Proc. Calif. Acad. Sci., 2d ser., V, pp. 51-52, May 28.

Describes Lepus martirensis from La Grulla.

1895. Van Densoureu, Jonn. A Review of the Herpetology of Lower California. Part I, Reptiles. <(Proc. Calif.
Acad. Sci., 2d ser., V, pp. 77-162, with 11 plates, May 28. :

A general discussion of the relations of the reptilian fauna of the peninsula, with an annotated list of 62 species.
1895. —__—_—-. A Review of the Herpetology of Lower California. Part II, Batrachians. <Proc. Calif. Acad.
Sci., 2d ser., V, pp. 556-561, September 10.
An annotated list of six species.
1896. AntHony, A. W. Eggs of the Black, Socorro, and Least Petrels. <The Nidologist, IV, pp. 16-17, October.
An account of the nesting and the eggs of the species named, as observed on San Benito Island. (See p. 59 of same volume of Nidolo-
gist for corrected egg measurements. )

1896. BenprrE, Cuaries. Life Histories of North American Birds, from the Parrots to the Grackles, with Special
Reference to their Breeding Habits and Eggs. [Part 2]. Smithsonian Contributions to Knowledge, XXXII.
[Special Bulletin U. 8. National Museum, No. 3], pp. 518, with seven colored plates, 4°. Washington.

Contains some original information concerning several Lower Californian species. :

1896. Coutrer, Joun M. Preliminary Revision of the North American Species of Echinocactus, Cereus, and Opuntia.
<Contr. U. S. Nat. Herb., III, pp. 355-462, April 1.

Contains descriptions of the Lower California species; some described as new.

1896. Van Densurca, Joun. Additional Notes on the Herpetology of Lower California. <Proc. Calif. Acad. Sci.
2d ser., V, pp. 1004-1008, January 15.

An annotated list of species collected by Stowell and Lunt during a trip to the San Pedro Martir Mountains in June and July, 1893.

158961°—21 11

 
160 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI

1897. Antony, A.W. Land Birds at Sea. <The Osprey, I, pp. 57-58, January.

Notes on a few species seen off Lower California.

1897. —__——_. New Birds from the Islands and Peninsula of Lower California. <Auk, XIV, pp. 164-168, iy
Original descriptions of Carpodacus megregori, Thryothorus cerroensis, and Harporhynchus lecontei arenicola.
1897. Davipson, GEorGr. The Submerged Valleys of the Coast of California, U. 8. A., and of Lower aitete,
Mexico. <(Proc. Calif. Acad. Sci., 3d ser., I, Geology, pp. 73-103, with 9 plates, June 25.
The last five pages of this paper are devoted toa brief account of the submarine valleys along the northwest coast of the peninsula.
1897. Eisen, Gustav. Explorations in the Cape Region of Baja California. <Bull. Am. Geog. Soc., XXTX, No. 3,
p. 271-280, with map, September.
A brief account of the climate, physiography, geology, and a few notes on some of the useful plants of this region.

1897. GayLorp, Horace A. Noteson Cormorants. <The Nidologist, IV, p. 102, May.
Brief notes on the cormorants frequenting San Geronimo Island.

1897. —————_. The American Ospréy in Lower California. <The Ogprey, I, p. 131, June.
Brief notes on occurrence and nesting of the osprey on islands off the west coast.
1897. ———_———. Notes from Guadalupe Island. <The Nidologist, IV, pp. 41-42, January.

Brief comments on birds noted between San Geronimo and Guadalupe Islands and on the latter in July, 1896.

1897. Kazpine, H. B. In Mexican Waters. The Nidologist, IV, p. 109, May.
Brief notes on a few birds noted on Todos Santos, San Martin, and Natividad Islands.

1897. McGregor, R.C. Nest and Eggs of the San Benito Sparrow. <The Osprey, II, November.
Notes on the nesting of this sparrow and the occurrence of other species of birds on San Benito Islands.
1897. Merritt, Grorce P. Notes on the Geology and Natural History of the Peninsula of Lower California. <(Ann.

Rept. U. S. Nat. Mus. for 1895 [1897], pp. 969-994, with cuts and map.
A general account of a trip during the summer of 1892 from San Quintin to El Rosario, San Fernando, the onyx mines, and El Tule
Arroyo, near the gulf coast, and return.

1897. Miuter, Gerrit §.,Jr. Some Bats from Lower California. <American Naturalist, XXXI, 75-77, January.
Records the capture by A. W. Anthony at San Fernando, in May and June, 1894, of Cheronycteris mexicana, Antrozous pallidus [=
A.p. pacificus), Vespertilio nitidus [= Myotis californicus], and Vesperugo hesperus (= Pipistrellus hesperus}.
1897. Merriam C. Hart. A New Fur Seal or Sea Bear (Arctocephalus townsendi) from Guadalupe Island, off Lower
California. <(Proc. Biol. Soc:, Washington, XI, pp. 175-178, July 1.
Contains historical notes and scientific description of the new species, which was based on a skull. No complete specimens of this
animal are known to.exist.
1897. ———_——. Revision of the Coyotes or Prairie Wolves, with Descriptions of New Forms. <(Proc. Biol. Soc.,
Washington, vol. XI, pp. 19-33.
Contains the original description of Canis peninsulz from Santa Anita, near San Jose del Cabo.
——. Description of a New Bassariscus from Lower California, with Remarks on Bassaris raptor Baird.
<Proc. Biol. Soc., Washington, Vol. XI, pp. 185-187.

Describes Bassariscus saricola from Espiritu Santo Island.

1897. OBERHOLSTER, Harry C. Critical Notes on the Genus Auriparus. <Auk, XIV, pp. 390-394, October.
Original description of Auriparus flaviceps lamprocephalus from Lower California.
1897. Taomas, OLpFIELD. Descriptions of New Bats and Rodents from America. <Ann. & Mag. Nat. Hist., 6th ser,
XX, pp. 544-553. ;
Contains original descriptions of Dasypterus ega ranthinus and Oryzomys peninsulz from the Southern end of Lower California.
1898. AttEN, J. A. Descriptions of New Mammals from Western Mexico and Lower California. <(Bull. Am. Mus,
Nat. Hist., X, pp. 148-158, April 12.
Contains original descriptions of 13 new species and subspecies from Lower California collected by Anthony.

1898. AnTHony, A. W. Two New Birds from the Pacific Coast of America. <Auk, XV, pp. 36-38, January.

Contains original description of Oceanodroma kaedingi from Guadalupe Island.

1897.

 

 

1898. ——————.. Four Sea Birds New to the Fauna of North America. <Auk, XV, pp. 38-89, January.
The species mentioned are Diomedea immutabilis, Puffinus auricularis, Puffinus cuneatus, and Phaéthon rubricaudus.
1898, —_————. Petrels of Southern California. Auk, XV, pp. 140-144, April.

Notes on habits and distribution of petrels off the coast of Lower California.
1898, —_———-. The Pacific Kittiwake (Rissa tridactyla pollicaris) in Lower California. <Auk, XV. p. 267, July.

First record of its occurrence there.

1898. McGregor, Ricuarp C. The Guadalupe House Finch. <The Osprey, II, pp. el February-March.
Briefnotes on the nesting of this species in March, 1897.

1898. ——_——-. Phalaropes. <The Osprey, II, pp. 87-88, February-March.
Briefnotes on migration and habits, including observations off coast of Lower California.

1898, —_————. Young Plumages of Mexican Birds. <Auk, XV, pp. 264-265, July.
Descriptions of young A mmodramus sanctorum and Carpodacus megregori.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 161

1898. McGrecor, Ricnarp C. Description of a new Ammodramus from Lower California. <Auk, XV, pp. 265-267,
July.
_ Describes A. halophilus.
1898. Munnram, C. Hart. The Earliest Generic Name for the North American Deer, with Descriptions of Five New
Species and Subspecies. <(Proc. Biol. Soc. Washington, XII, April 30, pp. 99-104.
Odocoileus cerrosensis from Cerros Island described.
1898. Minter, G.S.,Jn. Description of a new Bat from Lower California. <Ann. & Mag. Nat. Hist., ser. 7, No. 8,

pp. 124-125, August.
Original description of Myotis peninsularis and mention of four specimens of Myotis californicus trom the cape region.

1898. Ripaway, Rosert. Descriptions of Supposed New Genera, Species, and Subspecies of American Birds,
I. Fringillide. <Auk XV, pp. 226-227, July.
Original description of Aimophila ruficeps sororia from the Victoria Mountains in the cape region.
1898. THomas, OLpFrzLD. On New Mammals from Western Mexico and Lower California. <Ann. & Mag. Nat.
Hist., 7th ser., I, pp. 40-46.
Five new species and subspecies are described from collections made in the cape district, in 1896, by Dane Coolidge.

1898. WrBer, A. Les Echinocactus de la Basse California. <Bull. Mus. d’Hist. Nat., pp. 99-103, Paris.
Includes description of the large insular species of the gulf, E. digueti, named for its discoverer, Leon Diguet.

1899. AnrHony, A. W. A Night at Sea. <Bull. Cooper Orn. Club, I, pp. 101-102, November-December.
Observations on birds off the coast of Lower California.

1899. Bscx, R.H. Tropical Expedition Wrecked. <Bull. Cooper Orn. Club, I, p. 55, May-June.
An account of the loss on March 16, 1899, of the schooner Stella Erland on the coast 30 miles north of Magdalena Bay, with a scientific
expedition on board consisting of A. W. Anthony in charge, H. B. Keading, Chase Littlejohn, R. H. Beck, R. C. McGregor, and H. M.
Gaylord. The party sailed from San Francisco February 25.

1899. Branpeaes, T.S. New Species of Plants from Mexico. <Erythea, VII, pp. 1-9, January.
A list of 19 new species from various localities, including a few collected by A. W. Anthony and C. A. Purpus in Lower California.

1899. Brenincer, Grorce F. The Passing of Alfonse Forrer. <(Bull. Cooper Orn. Club, I, pp. 66-67, July-August.
Brief biography and portrait. Forrer made extensive collections in Lower California, especially near La Paz.

1899. Diavet, Lron. Rapport sur une Mission Scientifique dans la Basse-Californie. <Nouvelle Archives Missions
Scient. et Litt., IX, pp. 1-53, pls. I-X.
Not seen.
1899. DuptEy, Wu11amM Russeiy. Report on the Plants of Guadalupe Island. <The Fur Seals and Fur Seal Islands
of the North Pacific Ocean. Part III, pp. 280-283. Washington.

Contains a short account of previous botanical work on the island and an annotated list of37 species of flowering plants observed during
June, 1897.

1899. McGregor, RicHarp OC. A New Station for the Mexican Crossbill. <The Osprey, III, p. 141, May.
Records the capture of one specimen on Cedros Island.

1399. —~———. The Duck Hawk in Lower California. The Oologist, XVI, pp. 181-182, November.
A briefaccount of the breeding and other habits of this species on the islands along the west coast of the peninsula.

1899. Merritt, GzorcE P. A Trip Across Lower California. <The Osprey, III, pp. 20-25, with 5 cuts, October.
A general account ofa trip from San Quintin to the onyx mines and return with notes on animal life oieerved.

1899. Price, W.W. Some Winter Birds of the Lower Colorado Valley. <Bull. Cooper Orn. Club, I, pp. 8-93,
September-October.

General notes on the country and an annotated list of species noted on both sides of the Colorado River from Yuma to its mouth.

1899. SourawortH, Joun R. El Territorio dela Baja California, Mexico, Tlustrada. Pp. 88, with numerous cuts and
map, folio. San Francisco.
This publication is an illustrated advertising volume printed in Spanish and English, in parallel columns, with numerous portraits
and views. It contains a general account of the peninsula, its towns, prominent people, business enterprises, resources, etc.
1899. THopuRN, Wi.BUR Wiison. Report of an Expedition in Search of the Fur Seal of Guadalupe Island, Lower
California, June, 1897. <Fur Seals and Fur Seal Islands of the North Pacific Ocean, Part III, pp. 275-278,

with map of Guadalupe Island, quarto. Washington.
A brief general account of Guadalupe Island with notes on the seals observed there, a list of fishes collected, and an annotated list
of 10 species of birds observed.

1899. TownseNnp, CHARLES H. Pelagic Sealing, with Notes on the Fur Seals of Guadalupe, the Galapagos, and Lobos

Islands. <The Fur Seals and Fur Seal Islands of the North Pacific Ocean, Part III, pp. 223-274, pls. XXII-

XXXV. Chartof the distribution of the northern fur seals and map of Guadalupe Island, quarto. Washington.

Contains a general account with statistics of pelagic sealing and an account of Townsend’s trip to Guadalupe Island in search of fur
seals in 1892, when he secured the type of Arctocephalus townsendi and several sea elephants, with notes on number of fur seals taken since
1874 in this region. ;

1900, AnrHony, A. W. A Nighton Land. <The Condor, II, pp. 28-29, March.
A popular account of observations on the birds of San Benito Islands in late July, with notes on breeding habits.
162 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1900. AntHony, A. W. Eggs of the Black Oyster Catcher. <The Condor, II, p. 43, March.
Short description of nest and eggs with photograph, from Ascension Island.
1900. ————_——_. Nesting Habits of the Pacific Coast Epestes of the Genus Pufinus. <Auk, XVII, pp. 247-252,
with plate, July.
bservations mainly on the islands off the west coast of Tower California and the mainland of Mexico.

1900. Barrows, Davip P. The Colorado Desert. << National Geographic Mag., XI, No. 9, pp. 337-351, wate map

and illustrations, September..
Historical popular account of the region of the Lower Colorado River.

1900. Betpive, Lyman. A Part of My Experience in Collecting. <The Condor, II, pp. 1-5, January-February,
with portrait.
An autobiographical sketch with notes on the author’s work in Lower California.

1900. Branprecrsr, Karuarine. Notes on Cactee, II. <Zoe, V, pp. 1-9, June.
Notes on various species, mainly from Lower California, and descriptions of five new species and subspecies.

1900. BranpEcesz, T.S. Voyage of the Wahlberg. <Zoe, V, pp. 19-28, July.

An account of the scientific expedition made in this schooner under A. W. Anthony sailing in March, 1897, from San Diego and visiting
the islands along the west coast of Lower California and the Revillagigedo group. Contains brief notes on the‘islands visited and lists
of the species of plants collected. :

1900. —_——_——. A New Tapirira from Baja California. <Zoe V, pp. 78-79, September-October..

The original description of the Ciruela of the cape district. ‘

1900. Corr, Epwarp Drinker. The Crocodilians, Lizards, and Snakes of North America. <Rep. U. 8. Nat. Mus.
for 1898, pp. 153-1270, 800. Washington.
In the section on geographical distribution the author names the southern end of the peninsula, the “ Lower Californian District’’ and
gives the relations of the reptilian fauna on pages 1220-1221. The species are described and distribution given.
1900. Digurt, Lzon, La Basse-Californie. <(Annales de Geographie, IX, pp. 243-250, Paris, May.
Not seen.
1900. Ersen, Gustav. Explorations in the Central Part of Baja California. <(Bull. Amer. Geog. Soc. XXXII, No. 5,
pp. 397-429, with map.
A general account of the climate and physiography with the best map ofsouthern Lower California published to that date.

1900. Stoxes, Susan G. A New Species of Chorizanthe from Lower California. <Zoe, V, p. 60, August.
“ Describes Chorizanthe inequalis from San Telmo.

1901. AnrHony, A. W. The Guadalupe Wren. <The Condor, ITI, p. 73, May.
An account of the extinction of this species on Guadalupe Island,
1901. BranpEaes,T.S. New Species of Plants, Mainly from Baja California. <Zoe, V, pp. 104-109, February—April.

Contains descriptions of 10 new species from Lower California.
1901. Ersen, Gustav. Cerros, or Cedros Island. <Bull. Amer. Geog. Soc., XX XIII, No. 1, p. 64.
‘A discussion of the proper spelling, proving that the form Cedros is correct, :
1902. Anuen,J.A. The Hair Seals (Family Phocidae) of the North Pacific Ocean and Bering Sea, <Bull. Amer. Mus.
’ ‘Nat. Hist., XVI, December 12, pp. 459-499.
Contains original description of Phoca richardii geronimensis from San Geronimo Island.

1902. Bzxpinc, Lyman. The Fall Migration of Oreortyx pictus plumiferus. <The Condor, V, p. 18, January.

Mentions the occurrence of Anthus pensilvanicus at Cape San Lucas.

1902. Brewster, Wu.14Mm. Birds of the Cape Region of Lower California, <Bull. Mus. Comp. Zool., XLI, No. 1,
pp. 1-241, with map, September. :

This excellent summary of the bird life of southern Lower California is the best and most complete account of the birds ofany part of
the peninsula. Itis based on the collections and notes of M. Abbott Fraser and contains much original matter on life histories and the
original descriptions of Megascops zantusi, Bubo virgini Jachistus, and Tachycineta thalassina brachyptera.

1902. HyproerapHic Orrice, Unrrep States Navy Department. The West Coasts of Mexico and Central America
from the United States to Panama, including the Gulfs of California and Panama. 3d edition, p. 488, 800.
Washington.

Contains a great amount of detailed information concerning the islands, coasts, and adjacent interior in addition to sailing directions.

1902. Lumuis, ©. F. Early Western History. Diary of Junipero Serra: Loreto to San Diego, March 28—-June 30, 1769.
<Out West Magazine, Vol. XVI, pp. 293-296, 399-406, 513-518, 635-642; Vol. XVII, pp. 69-76.

A translation of the original mmanaaaiphs Contains some interesting notes on the county and Indians but nothing of value on the
natural history.

1902. Muze, Gernit S.,3r. Twenty New American Bats. <Proc. Acad. Nat. Sci., Philadelphia, pp. 389-412, May 20.

Includes original rieserinitlon of Antrozous minor from Comondu and Natalus mezicanus ree Santa Anita.

1902, Stave, Cartes B. Hunting Sheep and Antelope in Lower California. <Outing, XXXIX, pp. 505-512,

February.
Narrative ofa hunting trip from San Quintin to Matomi and San Juan de Dios.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 163

1902, TreLEasE, Wiutuiam. The Yuccee. <Thirteenth Annual Report, Missouri Botanical Garden, pp. 27-133,
with 99 plates, including maps,
A monographic revision of the group with descriptions of the species, photographs, and maps showing their distribution, including
Lower California.
1903. Branpscex, T. 8. Notes and New Species of Lower California Plants. <Zoe, V, pp. 155-174, May.
Thirty-seven species and varieties are described as new from the cape district.

1908. Brewster, Witu1aM. Breeding Grounds of the Black and Socorro Petrals—a Correction. <Auk, XX, pp.
63-64, January.
Corrects erroneous statement that these birds breed on Los Coronados Islands in the Gulf of California.~ The group near San Diego
wasintended. (See Birds of the Cape Region of Lower California, p. 33.)
1903. Exziot, D. G. Descriptions of Apparently New Species and Subspecies of Mammals from California, Oregon,
the Kenai Peninsula, Alaska, and Lower California, Mexico. <(Field Columb. Mus. Pub., Zool. ser., III, pp.
SS April.

h macrotis, Per 8 hemionotis, P. gaurus, P. homochroia, P. oresterus, Microtus californicus huperuthrus, Reithrodontomys

y ¥

i. Perognathus helleri, P. baileyi rudinoris, P. femoralis mesopolius, P. knekus, Spilogale arizonae martirensis, Sorer oreinus,
and Myotis milleri described from northern Lower California.
1903. ——————.. A List of Mammals Collected by Edmund Heller in the San Pedro Martir and Hanson Laguna Moun-
. tains and the Accompanying Coast Regions of Lower California with Descriptions of Apparently New Species.
<Field Columb. Mus. Pub., 79, Zool. ser., III, No. 12, pp. 199-232, pls. XX XIII-XXXVII, with map, June.
An important paper giving the results of the work done in northern Lower California during 1902 with a description of the country,
map, and annotated list of mammals from Heller’s notes.
1903. —__———. Descriptions of Twenty-seven Apparently New Species and Subspecies of Mammals. <Field Co-
lumb. Mus., Zool. ser. III, No. 14, pp. 239-261, December.
Contains original descriptions of Ovis cervina cremnobates and Dipodomys m. arenivagus from Lower California.
1903. GRINNELL, J.,and F.S. Dagcerr. An Ornithological Visit to Los Coronados Islands, Lower California. <Auk,
XX, pp. 27-37, January.

Contains a description: of the islands and an annotated list of 22 species of birds, including description of a new subspecies of song spar-
row ( Melospiza melodia coronatorum), observed therein August, 1902.

1903. Partsu,S.B. A Sketch of the Flora of Southern California. <Bot. Gaz., XXXVI, pp. 203-222, 259-279, October.
Contains a general description of the physiography of the region and characteristics and relationships of the flora with a list of 16 Lower
California species which enter California along the coast (p. 261).

1904. CuincH, Bryan J. California and its Missions. 2-vols. 8vo. Vol. 1, pp. 228, with cuts. San Francisco.
The first volume is devoted to an account of Lower California and the work of the Spanish missionaries there.
1904-5. FrezMaN, Louis R. A New Sportsman’s Paradise. <Western Field, V, pp. 191-197, 269-271, 352-360, 1904;
445-451, 1905. Sketch maps and photographs.
A popular account of a hunting trip in the delta region of the Colorado with some notes on the game birdsand mammals
1904. MacDoveat, Dante, TremBiy. Botanical Explorations in the Southwest. <Journal New York Botanical

Garden, V, No. 53, pp. 89-98, with plate and cuts, May.
Contains a briefaccount ofa boat trip from Yuma, Ariz., to San Felipe Bay, Lower California, and return in January and February,
1904.

1905. BranpEczE, Kataerine. Noteson Cactee. <Zoe, V, pp. 189-195, August.
Contains descriptions ofseveral new species from Lower California and other parts of Mexico.

1905. Branpeces, T.S. Palms of Baja California <Zoe, V, pp. 187-189, August.
Notes on the various native species and original description of Erythea brandegeei from the cape district.

1905. Exzison, O. C. In the Sea of Pearls. <Sunset Magazine, pp. 165-175, December.
A popular account of the southern gulfcoast of Lower California and its pearl fisheries.

1905. Fisher, Watrer K. In Memoriam: Walter E. Bryant. <Condor, VII, pp. 129-131, with portrait, September.
A brief biography giving dates of Bryant’s work in Lower California.
1905. Gotpman, E. A. Two New Wood Rats of the Genus Neotoma. <Proc. Biol. Soc., Washington, XVIII,
February 2, pp. 27-34.
Includes original descriptions of Neotoma martinensis, from San Martin Island, and N. nudicauda from Carmen Island.

1905. Grinnett, J. Where Does the Large-Billed Sparrow Spend the Summer? <Auk, XXII, pp. 16-21, January.
A general discussion ofits distribution and possible breeding places.

1905. GRINNELL, JoszPH. The Ornithological Writings of Walter E. Bryant. <Condor, VII, pp. 131-132, September.
This bibliography contains numerous titles of papers on Lower California subjects.
1905. Karpinc, Henry B. Birds from the West Coast of Lower California and the Adjacent Islands. <Condor,

VII, pp. 105-111, 134-138, July and September.
A short account ofislands visited, including the Revillagigedo group and briefly annotated list of birds seen, with a short bibliography.

 

1905. MacDouagat, Danie, TremBiy. Botanical Exploration in Arizona, Sonora, California, and Baja California,
<Journ. N. Y. Bot. Garden, VI, No. 66, pp. 91-102, with cuts, June.
Contains a brief account of the expedition down the lower Colorado, and tributaries to near the base of the Cocopah Mountains, in
company with E. A. Goldman, March and April, 1905.
164 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1905. Muzx, Seta Evaznr. An Annotated List of a Collection of Reptiles from Southern California and Northern
Lower California. <Field Columbian Museum, Zool. ser., VII, No. 1, pp. 1-19, 3 plates, November.
A list of the species collected by E. Hellerin Lower California north of Matomi, southern end of the San Pedro Martir Mountains.

1905. Parriz, James O. Pattie’s Personal Narrative of a Voyage to the Pacific and in Mexico, June 20, 1824~August
30, 1830. <Early Western Travels, 1748-1846, XVIII, pp. 1-364, 800. Cleveland.

This is a reprint of the original edition of 1831. The narrative contains an interesting account of an expedition of an American trap-
per through what is now the southwestern United States, northern Lower California, and parts of the mainland of Mexico. _

1905. Sronz, Wirmer. Ona Collection of Birds and Mammals from the Colorado Delta, Lower California, with Field

Notes by Samuel N. Rhoads. <Proc. Phil. Acad. Nat. Sci., pp. 676-690, September.
A briefly annotated list of 22species ofmammal]sand 107 birds. The results ofa trip to the mouth of the Coloradoearly in 1905 by S. N.
Rhoads.

1905. TownsEnp, CHarLes Hasxrns. Notes on Certain Pinnipeds, with Data Respecting Their Present Commercial
Importance. <Ninth Ann. Rept. New York Zoological Society, pp. 105-116, with 3 plates, (1904).
Gives brief notes concerning the northern sea elephant and the Guadalupe fur seal.

1905. Van Dyxe, T.S. Sport on the Lower Colorado. <Western Field, VI, pp. 3-7, February.
A popular account of a hunting trip from Yuma, Ariz., to the head of the Gulf of California, in December, 1904, during which great
numbers of ducks, geese, and waders were noted.

1906. ANTHony, A. W. Random Notes on Pacific Coast Gulls. <Auk, XXIII, pp. 129-137, April.
A general account of the distribution and habits of gulls including those seen along the coast of Lower California.

1906. ——————. Where Does the Large-Billed Sparrow Spend the Summer? <Auk, XXIII, pp. 149-152, April.

Discusses the probable breeding ground ofthis species, especially in northwestern Lower California.

1906. Berterscu, Hans von. Ona New Form of Oceanodroma Inhabiting San Benito Island, Off the Coast of Lower
California. <Auk, XXIII, pp. 185-186, April.

Original description of Oceanodroma monorhis chapmani. é
1906. Branpecez, T.S. New Species of Mexican Plants Collected by Dr. C. A. Purpus. <Zoe, V, pp. 231-241,
September.

Includes origina! description of Cassia purpusi from Lower California.

1906. MacDoueat, Danie, TRemsBiy. The Delta of the Colorado, with Map by Godfrey Sykes. < Bull. Amer,
Geog. Soc., XX XIII, No. 1, pp. 1-16, with cuts and maps, January.
A general account ofthe delta and of the adjacent area in Lower California, with notes on the principal plants observed. This paper
was reprinted in ‘Contributions from the New York Botanic Garden,” No. 77, pp. 1-16.
1907. MacDoveat, D. T. The Desert Basins of the Colorado Delta. <Bull. Amer. Geog. Soc., XXXIX, No. 12, p.
705-729, with map and illustrations, December. 4
Popular account of the region with discussion ofearly history.

i
1907. Mearns, Epcar ALEXANDER. Mammals of the Mexican Boundary of the United States. A descriptive cata-
logue of the species of mammals occurring in that region, with a general summary of the natural history and a
list of trees. United States National Museum Bulletin 56, pp. I-XV, 1-530, April 13.
Includes narrative of explorations conducted along the boundary line between the United | States and Lower California from March to
July, 1894. Many species of mammals and plants of Lower California discussed in the annotated lists.
1907. Merriam, C. Hart. Descriptions of Ten New Kangaroo Rats. <Proc. Biol. Soc. Washington, XX, pp..75-80,
July 22.
Contains original descriptions of Dipodomys platycephalus, D. maragarite, D. insularis and Perodipus simulans peninsularis from Lower
California.
1907. Nexson, E. W. Descriptions of New North American Rabbits. <(Proc. Biol. Soc. Washington, XX, pp.
81-84, July 22.
Contains original descriptions of Lepus californicus magdalenz, Sylvilagus mansuctus, and 8. bachmani eriguus from the peninsula.

1907. Nort, ARTHUR Wa sripce. Hunting the Bighorn. <Sunset Magazine, October, pp. 523-532, with cuts.
A popular account of hunting mountain sheep in various parts of Lower California, including the killing of an extremely dark specimen
on the east base of the San Pedro Martir Mountains.

-1907. ey The Uncharted Sierra of San Pedro Martir. <Bull. Amer. Geog. Soc., XXXIX, No. 9, pp. 544-554,
September.
A general account of these mountains.
1907. ——_———. Map of San Pedro Martir Sierra, Lower California. <Bull. Amer. Geog. Soc., XX XIX, No. 12,

facing page 768, December.
This is a small outline map to accompany the author’s article on the same region in this bulletin, Vol. XX XIX, No.9, for September
1907.
1907. ——————. Resources of Lower California. <Monthly Bull. Intern. Bur. Amer. Republics, XXV, No. 6,
pp. 1374-1376, December. :
A brief genera? account of the peninsula and its main resources.
1907, ————_——. The Mother of California. 8vo, pp. 169, with map and halftone illustrations. New York and

San Francisco.
A general historical account of the peninsula with chapters on the phySical character of the country, mines, and Dresent conditions,
with a bibliography.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 165

1907. OsERHOLSER, Harry C. Description of a New Otocoris from Lower California. <(Proc. Biol. Soc. Washington,
XX, pp. 41-42, March 27.

Describes Otocoris alpestris enertera from near Magdalena Bay.

1907. Oscoop, Wirrrep H. Four New Pocket Mice. <Proc. Biol. Soc. Washington, XX, pp. 19-22, February 23.
Contains original descriptions of Perognathus penicillatus siccus, P. p. ammophilus, and P. spinatus magdalenx, from Lower California.
1907. PanisH, 8S. B. A Contribution, Toward a Knowledge of the Genus Washingtonia. <Bot. Gaz., XLIV, pp.

408-434, with illustrations, December.
A discussion of the status and origin of the types of the various forms, most of which are probably from Lower California. Retains
the generic name Washingtonia.

1907. THaveR, JoHn E., and Ourram Banes. A New Race of the Californian Thrasher from Lower California.
<Proc. New England Zool. Club, IV, pp. 17-18, April 30.
Describes Torostoma rediviva helva from Rosario.
1907. —————. Birds Collected by W. W. Brown, jr., on Gerros, San Benito, and Natividad Islands in the Spring
of 1906, with Notes on the Biota of the Islands. <(Condor, IX, pp. 77-81, May.
Briefly annotated lists with some general notes on the islands.
1907, —__———. Catalogue of Birds Collected by W. W. Brown, jr., in Middle Lower California. <Condor, IX,
pp. 135-140, September.
A briefly annotated list of species observed during the autumn of 1905 and spring of 1906.
1908. Enee.Harpt, O. F. M. (Fr. Zepuyrin). The Missions and Missionaries of California, Vol. I, Lower California.

Pp. 654, with cuts and maps.
This publication contains a good account of the work of the Spanish missionaries in Lower California from their first arrival in 1535
to 1859. Itis based on original records and published material.

1908. LypEKKER, R. Sea Elephants for the British Museum. <(London Field, III, p. 276, February 15.
A brief general account of the species based on specimens secured on Guadalupe Island, by Harris in May, 1907.

1908. Nort, ArtHuR W. Magdalena Bay. <Sunset Magazine, March, pp. 411-420, with plates and cuts.
A popular account of the history and surroundings of Magdalena Bay.

1908. Norra, ARTHUR WaLBRipcE. The Native Tribes of Lower California. <American Anthropologist, X, pp.
236-250, April-June.
A résumé of the historical accounts of the natives of this region with original notes on present conditions, and notes and cuts of petro-
glyphs in various parts of the peninsula.
1908. Rorascuitp, Hon. Watrer. Mirounga angustirostris (Gill). <(Novitates Zooligice, XV, pp. 393-394, Plates
I-VIII, November 18.
Contains a brief account of thissea elephant based on the series of seven specimens secured for Tring Museum, by Charles Harris, at
Guadalupe Island, June 2-17, 1907, accompanied by reproductions of photographs of the living animals taken by Harris.
1908. THayeR, Joun E., and Ourram Banos. The Present State of the Ornis of Guadalupe Island. <Condor,
X, pp. 101-106, May-June.
Contains a brief description of the island and an annotated list of the birds observed there, by W. W. Brown, jr., during May and
June, 1906.
1909. Gotpman, E. A. Five New Wood Rats of the Genus Neotoma from Mexico. <Proc. Biol. Soc. Washington,

XXII, pp. 139-143, June 25.
Neotoma intermedia pretiosa, from Matancita, 50 miles north of Magdalena Bay; N. i. perpallida, from San Jose Island; N. i. vicina,
from Espiritu Santo Island; and XV. abbreviate, from San Francisco Island are included in the new forms described.

1909. GrinneLt, J. The Zone-tailed Hawk in California. <Condor, XI, p. 69, March.
Records a specimen from southeast of Tijuana, Lower California.
1909. Harris, CHARLES Miutzr. A Cruise After Sea Elephants. <(Pacific Monthly Magazine, April, pp. 321-339,

with cuts.
A popular account of a trip to Guadalupe Island in May, 1907, when a number of specimens of sea elephants were collected for Tring
Museum.

1909. Lams, Cuester. Nesting of the Xantus Murrulet as Observed on Los Coronados Islands, Lower California.
<The Condor, XI, pp. 8-9, January.
Date and character of nesting places at this locality visited in June, 1908.
1909. LypexKer, R. On the Skull Characters in the Southern Sea Elephant. <Proc. Zool. Soc. London, May 25,
' pp. 600-606. :
Southern and northern sea elephants regarded as specifically distinct.
1909. N. ELSON, E. W.,.and E. A. Gotpman. Eleven New Mammals from Lower California. <(Proc. Biol. Soc. Wash-

ington, XXII, pp. 23-28, March 10.

Original descriptions of Eutamias merriami meridionalis, A mmospermophilus leucurus insularis, Thomomys magdalenz, T. botts russe-
olus, Vulpes macrotis devius, Bassariscus astutus palmarius, B. astutus insulicola, Sorez lagunz, S. californicus juncensis, Myotis capi-
taneus, M. micronyz.

1909. OspuRN, PinerrEe I. Notes on the Birds of Los Coronados Islands, Lower California. <Condor, XI, pp. 134—

138, July. 3
An annotated list of species observed in the spring of 1908
166 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1909. Oscoop, Witrrep H. Revision of the Mice of the American Genus Peromyscus, North American Fauna No.
28, pp. 1-285, pls. 1-8, April 17.
Includes original descriptions of Peromyscus maniculatus margaritz from Margarita Island, P. m. magdalenz trom Magdalena Island,
P. eremicus avius from Ceralbo Island, P. e. polypolius from Margarita Island, P. einsulicola from Espiritu Santo Island, and P. truei
lagunae from La Laguna.
1909. Tuaver, Joun E. Some Rare Birds and Sets of Eggs from the Cape Region of Lower California. <(Condor,
XI, pp. 10-11, January.
A few notes on some species taken by W. W. Brown, jr., at La Paz and in the Sierra Laguna.

1909. ————. Correspondence. <Condor, XI, pp. 142-143, July.
Several letters written from La Paz, giving details of ornithological collecting by W. W. Brown, jr., during the winter of 1909.

1909. ————-. Frazar’s Oyster Catcher’s Eggs. <Oologist, XX VI, pp. 188-189, November 15.
Note on the number of eggs laidin a set by this species.

1909. THaysr, Jonn E., and Ourram Banas. Description of a New Subspecies of the Snowy Heron. <Proc.
New England Zool. Club, IV, pp. 39-41, April 29. ;
Describes Egretia candidissima brewsteri from San Jose Island.
1909. Van Rossem, A. Some Unrecorded Species from Los Coronados Islands, Mexico. <Condor, XI, p. 208,
November.
A briefly annotated list of nine species.
1909. Wricut, Howarp W. An Ornithological Trip to Los Coronados Islands, Mexico. <(Condor, XI, pp. 96-100,
May.
An annotated list of species observed in June, 1908.

1910. Howeit, ALFRED B. Notes from Los Coronados Islands, <(Condor, XII, pp. 184-187, November.
Notes on breeding habits of birds found there the spring of 1910, especially of Xantus murrelet.

1910. Lams, Cuester. A Glimpse of Bird Life on the West Coast of Mexico. <Condor, XII, pp. 74-79, March.
A popularaccount of a trip and birds seen from San Diego Calif., toSan Blas and the Las Marietas Islets off Cape Corrientes, in the spring
of 1909, with notes on the birds.
1910. Norra, ARTHUR WaLBRIDGE. Camp and Camino in Lower California. 8vo., pp. 346, with map and halftone
plates. New York.
A popular account of journeys throughout the length of the peninsula in 1905-6 with a bibliography.
1911, AnpREws, Roy C. Description of an Apparently New Porpoise of the Genus Tursiops with Remarks upon a
Skull of Tursiops gilli Dall. <Bull. Amer. Mus. Nat. Hist., XXX, pp. 233-237, with plate, August 26.
Original description of Tursiops nuuanu and record ofits occurrence and that of T. gillii in Lower California.
1911. Dickerson], M[ary] C. Rare Elephant Seals for the Museum. <American Museum Journal, XI, No. 4, pp.
109-112, April.
A brief notice of the capture of elephant seals on Guadalupe Island by the Albatross expedition the 1st of March, 1911, with the first
photographs ever published showing the “trunk” of the old males, taken by C. H. Townsend.
1911. GrinnELL, GEorGE Brrp. The Newly Discovered Elephant Seals. <Forest and Stream, LXXVI, No. 17,
p. 652, April 29.
A briefaccount of the discovery by the Albatross expedition of elephant seals on Guadalupe Island, in March, 1911, and the capture’of
six young for the New York Aquarium, with photographs of the old males on the beach.
1911. Howet, ALFRED B. Some Birds of the San Quintin Bay Region, Baja California. <Condor, XIII, pp. 151-
158, September, 1911. :
General notes on species observed during April, May, and July, 1910 (on p. 211 of same volume the name Numenius americanus of
of this article corrected to N. hudsonicus).
1911-12. Humpurey, W. E. Shooting the Vanishing Sheep of the Desert. <Outdoor Life, XXVIII, No. 6, pp.
477-487, December, 1911; XXIX, No. 1, pp. 3-16, January, 1912; No. 2, pp. 95-105, February, 1912.
A good popular account of a hunt in the Sierra Tinaja in northeastern Lower California with notes on the country and habits of the
sheep, with illustrations from photographs. ;
1911. Morr, Russext (‘‘Douste Barre”). Boregas and Tanakas. <Forest and Stream, LXXVI, No. 12, pp.

448-451, March 25.
A good account of a hunting trip after mountain sheep to the Tinajas Mountains west of the mouth of the Colorado River, Lower

California.
1911. Netson, E.W. A Land of Drought and Desert—Lower California. Two Thousand Miles on Horseback Through
the Most Extraordinary Cacti Forests in the World. <The National Geographic Magazine, XXII, 443-474,

with 28 illustrations, May. 7
A popular account of the journey of the author and E. A. Goldman the entire length of the peninsula in 1905-6,

1911. OseRHoLseR, Harry C. A Revision of the Forms of the Ladder-backed Woodpecker (Dryobates scalaris [Wagler]).

<Proc. U. 8. Nat. Mus., XLI, pp. 139-159, June 30.
Contains original description of D. s. eremicus from San Fernando, Lower California.

1911. Ossurn, Prrcrze I. Collecting Socorro and "Black Petrels in Lower California. <Condor, XIII, pp. 31-34,

January.
Notes on breeding habits on Los Coronados Islands, Lower California, mainly in 1909 and 1910.
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 167

1911. OssurNn, Raymonp C. California Elephant Seals at the New York Aquarium. <Zoological Society Bulletin
No. 45, pp. 759-762, with photographs, May.
An account of six young elephant seals captured on Guadalupe Island, Lower California, and received at the New York Aquarium,
March 13, 1911, via San Diego.
1911. Rmeway, Rosert. Diagnoses of Some New Forms of Picide. <(Proc. Biol. Soc. Washington, XXIV, pp.
31-35, February 24.
Contains original description of Centurus wropygialis brewsteri from the cape district of the peninsula.
1911. Sranptzy, Paut C., and E. A. Gotpman. Two New Shrubs from Lower California. <Contr. U. 8. Nat. Herb.,
XIII, pt. 10, p. 375, June.
Original descriptions of Manihot chlorosticta and Mesosphaerum insulare from the southern part of the peninsula.
1911. TaayeR, Joun E. Eggs and Nest of the San Lucas Robin (Planesticus confinis). <Oologist, XXVIII, p. 77,
April, 15.
Two nests and sets of eggs described. :
1911. ——_——. A Nesting Colony of Herrmann Gulls and Brewster Boobies. <(Condor, XIII, pp. 104-106, with
photographs, May.
Observations made on San Ildefonso Island, Lower California, the last of March, 1909, by W. W. Brown, jr.

1911. —————_._ Eggs of the Elegant Tern (Sterna elegans). <Oologist, XXVIII, p. 171, November 15.
Describes a series of eggs collected on Ceralbo Island in April, 1910, by W. W. Brown, jr.

1911. TowNsEND, CHARLES H. Oceanic Work on the Albatross. <The American Museum Journal, XI, No. 5 p. 159,

May.
Quotations from letters from the Albatross expedition along the coasts of Lower California the spring of 1911, giving notes on
itinerary, etc.
1911. —————. Zoological Results of the Albatross Voyage. <(Zoological Society Bulletin No. 46, pp. 774-776,

New York, July.
A brief popular account with photographs.
1911. TRELEASE, Witt1AmM. The Desert Group Nolinee. <Proc. Amer. Phil. Soc., L, pp. 404-442, 2 maps and 17
plates, July-August.
A systematic revision of the group, including five forms occurring in Lower California, of which two are described as new.
1912. Arten, J. A. Historical and Nomenclatorial Notes on North American Sheep. <Bull. Amer. Mus. Nat. Hist.,
XXXI, pp. 1-29, March 4.

Contains notes on the first published records of mountain sheep from Lower California, which was also the first for North America.

1912. Howetr, A. B. Notes from Todos Santos Islands. <Condor, XIV, pp. 187-191, September.
An annotated list of the birds observed in April, 1910. .
1912. Nexson, E.W. A Correction of Two Recent Names forMammals. <(Proc. Biol. Soc. Washington, XXV, p. 116,
June 29. :
Perognathus penicillatus seri substituted for P. p. goldmani and Perognathus spinatus occultus for P. 8. nelsoni.
1912, —————-. A New Subspecies of Pronghorn Antelope from Lower California. <(Proc. Biol. Soc. Washington,
XXV, pp. 107-108, June 29. =
Original description of A ntilocapra americana peninsularis from 45 miles south of Calmalli.

1912. Pururs, J.C. A New Puma from Lower California. <Proc. Biol. Soc. Washington, XXV, pp. 85-86, May 4.
Original description of Felis improcera from Calmalli.
1912. TuayeEr, JoHN, and Ourram Banas. A New Race of Great Blue Heron from Espiritu Santo Island, Lower Cali-
fornia. <(Proc. New England Zool. Club, IV, February 23, pp. 83-84.
Description given of Ardea herodias sancti-lucae from Espiritu Santo Island.
1912. TownseND, CHARLES Hasxins. The Northern Elephant Beals <Zoologica, I, No. 8, pp. 159-173, with numer-
ous plates, April 15.
Contains an account of the rediscovery of:a herd of these animals on Ganibdiune Island, Lower California, March 2, 1912, with a fine
series of photographs of the living animals taken by the author.
1912, ———_—-. The Elephant Seal not Extinct. A Herd Discovered on Guadalupe Island, West of Lower ele
fornia. <The Century Magazine, LX XXIV, pp. 205-211, with photographs, June.
An account of the author’s search for these animals in previous years, the discovery of the herd of more than 100 on Guadalupe in
March, 1911, and the capture of some young, which were sent to the New York Aquarium.
1912, ——___—. Mammals Collected by the Albatross Expedition in Lower California in 1911, with Descriptions
of New Species. <Bulletin American Museum of Natural Hilstcry, XXXI, pp. 117-130, with 2 plates, June

14,
Contains a list of species collected or observed during the expedition, including the following species or subspecies supposed to be new:
Lepus alleni tiburonensis, Perognathus baileyi insularis, P. p icillatus gold |, P.8p i, Ne albigula seri, Neotoma insu-

 

laris, Peromyscus guardia, P, stephani, P, eremicus carmeni, and Canis jamesi.

1912. Tretzasz, WiuiamM. The Agaves of Lower California. <Twenty-second Annual Report of the Missouri Botan-
ical Garden, pp. 37-65, pls. 19-72, February 14, including descriptions of new species.

Contains descriptions of 25 species, of which 17 are described as new.
168 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1913. Bonttras, Y.S., and others. Memoria de la Comision del Instituto Geologico de Mexico que Exploro la Region
Norte de la Baja California. <Parergones del Instituto Geologico de Mexico, V, Nos. 2-10, pp. 89-534, laminas
I-CXII, Secretaria de Fomento, Colonizacion e Industria, Mexico, as follows:

Bonillas, Y. 8., and Urbina, F. Informe acerca de los Recursos Naturales de la porte Norte de la Baja Cali-

fornia, Especialmento del Delta del Rio Colorado, pp. 162-235, laminas I-XXIV.

Flores, Teodoro, and Gonzalez, Pedro, jr. Exploracion dela porte Central Elevada de la Porcion Norte de

la Peninsula de la Baja California, pp. 237-275, laminas KXV-LXV.

Engerrand, Jorge, and Poredes, Trinidad. Informe Relativo a la Porte Occidental de la Region Norte de

la Baja California, pp. 278-306, laminas LX VI-LXXXVII.

Bose, E., and Wittich, E. Informe Relativo a la Exploracion de la Region Norte de la Costa Occidental de

la Baja California, pp. 307-529, laminas LK XXVIII-CXII.

This report contains the results obtained during 1911 and 1912 in the northern part of Lower California by an expedition divided into
four sections sent out by the Geological Institute of the Department of Interior, Mexico, to investigate the physiography, geology, and the
water, agricultural, and other resources of this region. The reports are accompanied by maps and plates and give the best and most
detailed information extant on the northern half ofthe peninsula. .

This report was reprinted with title page bearing a different vignette and with the following cover title: Memoria de la Comisién del
Tustituto Geolégico de México que Exploré la Region Norte de la Baja California.

1913. Pumurs, J. C. The Lower California Pronghorn Antelope. <Science (N. 8.), XXXVII, No. 5, pp. 717-718,
May 9. .

A note giving skull and horn characters of Antilocapra americana peninsularis and confirming the validity of this subspecies.

1913. SmurtcH, Frep. The Land of American Cannibals. <Travel, XXI, No. 5, pp. 34-87, with cuts, September.
A brief, interesting account of Lower California, including some references to the physical character and people of Tiburon Island.
The illustrations are from photographs taken both on the peninsula and on the island.

1913. WessteR, E.Bartietr. Report on the Northern District of Lower California, Pp. i-xxiii, 1-359, 8vo, Decem-
ber. San Diego, Calif.

A general account of northern Lower California, including the physiography and natural resources.

1913. Wmiet, GEorGE. Bird Notes from the Coast of Northern Lower California. <(Condor, XV, pp. 19-24, January—
February.

An annotated list of 98 birds observed from April 4 to 21, 1912, during a coasting trip from San Diego to the south end of Cedros Island,

1913. Wricut, Howarp W. The Birds of San Martin Island, Lower California. <(Condor, XV, p. 207-210, Novem-
ber. :

Annotated list of 12 species.

1914. GRINNELL, JosepH. An Account of the Mammals and Birds of the Lower Colorado Valley with Especial Refer-
ence to the Distributional Problems Presented. Univ. Calif. Pub. Zool., vol. 12, pp. 218-294, with map and
illustrations, March 20. ~

Extended discussion of physiography and associational areas, with annotated lists of birds and mammals.
1915. Funcxs, E.W. The Sheep Hunting of Lower California. <(Outdoor Life, September, pp. 221-228, with pho-

tographs and map.
A general popular article on the habits and hunting of mountain sheep in Lower California with information concerning other game
animals and a fairly accurate map showing the distribution of large game on the entire peninsula.

1915. Hem, Dr. Arnotp. Sur la Géologie de la Parte Meridionale de la Basse Californie. <Comptes Rendu, vol.

161, p. 419, L’ Academie des Sciences, Paris, October.
Gives briefnotes on the geology of the peninsula south of La Purisima.

1916. ——_———. Reisen im Siidlichen Teil der Halbinsel Niederkalifornien. <Zeitschrift der Gesellschaft fiir Erd-
kunde zu Berlin, No. 1, 4 Tafeln.
A briefaccount of an expedition through a large part of the peninsula during the summer of 1915 by a party of Swiss geologists and
botanists to investigate the character of the country, largely in relation to its agricultural possibilities.
1916. —— Charakter-Pflanzen der Halbinsel Niederkalifornien. <(Vegetations-Bilder, 13 Reihe, Heft 3-4,
Tafel 13-24, Jena. ‘

A short introduction on the general character of the peninsula, followed by 12 handsome quarto plates illustrating some of the most
notable plants with text.
1916. GoLDMAN, Epwarp A. Plant Records of an Expedition to Lower California. <Contr. W. 8. Nat. Herb.,
vol. 16, part 14, pp. i-viii, 309-371, ix—xiii, pls. 104-133, with map, 8 vo., February 10.
An annotated list of the principal plants collected or noted by E. W. Nelson and E. A. Goldman during the Biological Survey
Expedition of 1905-6. Includes descriptions of three new species of oak—Quercus brandeget, Q. idonea, and Q. devia,
1916. Poder Ejecutivo, Secretarfa de Fomento, Colonizacién y Industria, Direccién Agraria, 18 de mayo de 1916.
Permiso concedido al Sefior Miguel L. Cornejo para la explotacién de las salinas ubicadas al Sur de la Isla de
Pichilinque, Bahia de La Paz, Baja California. <(Boletin Oficial de la Secretaria de Fomento, Colonizacién

y Industria, Mexico, I, No. 4, pp. 133-135, octubre.
Permit to exploit the salt beds south of Pichilinque Island in La Paz Bay for two years from May 18, 1916,

 
No. 1] LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 169

1916. Poder Ejecutivo, Secretaria de Fomento, Colonizacién y Industria, Direccién Agraria, 28 de octubre de 1916.
Contrato celebrado entre el C. Ing. Pastor Rouaix, Secretario de Fomento, Colonizacién e Industria, en
representacién del Executivo de la Unién, y los Sefiores Gerardo Baltan4s y Salvador Catafio, para el arrenda-
miento de los *‘Islotes Coronados,”’ sitios en el Océano Pacifico, Distrito Norte de la Baja California. <Bole-
tin Oficial de la Secretarfa de Fomento, Mexico, I, No. 3, pp. 97-98, octubre.

Leases the Coronado Islands for five years from October 28, 1916, for the establishment there of a hotel, whaling station, and stone
quarries.

1916. Poder Ejecutivo, Secretarfa de Fomento, Colonizacién y Industria, Direccién Agraria, 10 de noviembre de 1916.
Contrato celebrado entre el C. Ingeniero Pastor Rouaix, Secretario de Estado y del Despacho de Fomento,
Colonizacién e Industria, en representacién del Executivo-de la Nacién, por una parte, y el Sefior Nicolas
Fossum, como apoderado del Sefior Ragnar M. Fossum, por la otra, para establacer, en la Baja California,
una fabrica de productos obtenidos mediante el. beneficio de Ja ballena. <Boletin Oficial de la Secretaria
de Fomento, Colonizacién e Industria, Mexico, I, No. 4, pp. 137-141, noviembre.

This concession permits Ragner M. Fossum {Norwegian] to establish a whaling industry on the west coast of Lower California for 10
years from November 10, 1916. The products are to be oil, guano, and salted, smoked, or dried whale meat for food. No whaling is
permitted in the bays nor during the breeding season.

1916. Rotnanp, M. C. Problema de Ja Baja California, Canal de California. Pp. 1-12.

This pamphlet without date or imprint gives on its front cover an outline map of Lower California with a canal laid down from the
mouth of the Colorado River westward to Ensenada Bay. The author gravely advocates the building of this canal, apparently oblivious
of the great north and south mountain range about 6,000 feet high directly across its course, in order to develop the commerce of Lower
California and northern Mexico. This is a good illustration of the impractical folly shown in a large part of the writings advocating the
development of Lower California and shows complete ignorance of the real conditions there.

1916. TowNsEND, C. H. The Guadalupe Fur Seal: A Contribution to Its History. <(Zoological Society Bulletin

+ (New York), XIX, No. 2, pp. 1330-1331.
A brief record of the number of skins of this seal known to have been taken, including 8,338 taken by the whaler Port aw Prince at
San Benito Islands in 1806.

1916. TownsEND, CHARLES Haskins. Voyage of the Albatross to the Gulf of California in 1911. <Bulletin American
Museum of Natural History, XX XV, pp. 399-476, with map and plates, August 2.

A general account of the expedition with list of places visited and a brief notice of the character of many localities and the work
done there. Of special note are the location of elephant seals on Guadalupe Island and the account of the pearl oyster farm on Espiritu
Santo Island.

1917. BatarEzo, M. Colonizacién de Ja Baja California. <Boletin Oficial de la Secretarfa de Fomento, Mexico, IT,
No. 2, pp. 110-133, Mayo.
Contains a brief summary of the physical character of the peninsula, its products, and resources, with a plan for colonization under
theland system of New Zealand. Owing to obvious climatic differences between the two regions, this plan is wholly impractical.

1917. D=E 1a Veca, R. Informe sobre las Riquezas Naturales de las Islas Tres Marias. <Boletin Oficial de la Secre-
taria de Fomento, Mexico, II, No. 1, pp. 101-105, abril.

A brief report on the resources of the Tres Marias Islands off San Blas, Mexico, by the head of a commission sent by the Mexican

Government for the purpose ofstudyingthem. Contains a brief résumé of the changing ownership of the islands and their exploitation.

1917... . El Hispano Americano. Edicion bilingiie, Fiesta de Amistad, pp. 1-96, mapa, San Diego, Calif., junio.
This publication fhas the character of a periodical without serial number. The present issueis practically all devoted to shortsigned
and unsigned articles concerning the northern district of Lower California, its resources, and its progress under the present governor,
Col. Esteban Cantu. It was published to commemorate a “Friendship Fiesta’’ held during June, 1917, in San Diego, Calif., for the
purpose of bringing people on both sides of the border into closer contact.

1917. Murray, RoBert CusHMAN. The Desert Life Group, and an Account of the Museum Expedition into Lower
California. The Brooklyn [N. Y.] Museum Quarterly: V [=IV], No. 4, pp. 179-210, with pldtes and cuts.
October.

An interesting account ofa trip into the desert northeastern part of Lower California in search of pronghorn antelope and other speci-
mens for the fine group of desert life now installed in the Brooklyn Museum.

1917. —_——-. Natural History Observations from the Mexican Portion of the Colorado Desert with a Note on the
Lower Californian Pronghorn and a List of the Birds. Proc. Linnaean Soc. New York, No. 29, for the vear
ending March 13, pp. 43-101, pl. 1, December 11.

A very full physiographic and faunal description of the valley of the Lower Colorado River, with an annotated list of 134 species of
birds.

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién Agraria, No. 460, 12 de abril de 1917. ‘Acuerdo de nulidad
relativo al contrato de fecha 29 de junio de 1906, celebrado con el Lic. Emilio Velasco en representacién de
las compafiias llamadas ‘‘Compafiia Mexicana de Terrenos y Colonizacién” y ‘‘ Compaifiia del Desarrollo de la
Baja California.’”” <Diario Oficial, Mexico, VI, No. 4, pp. 507-508, 7 de mayo; also in Periodico Oficial,
Ensenada, B. C., XXX, No. 16, pp. 2-3, 10 de junio. ©

Cancellation of the grant for colonization of June 29, 1906, to Emilio Velasco, representing the Mexican Land and Colonization Co,
and the Lower California Development Co., of the lands formerly covered by an original grant to Luis Huller y Cia, dated July 21, 1884,
This grant covered most of the lands in the northwestern part of Lower California, including an area of 5,606,504 hectares.

Occupants of lands deriving titles from any of the companies named are notified that they may have their equity recognized by appli-
cation to the Secretary of the Interior, Mexico.
170 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

1917. Poder Ejecutivo, Secretarfa de Fomento, Colonizacién e Industria, Direccién Agraria, No. 455, 12 de abril de
1917. Declarando nulo el contrato de 31 de marzo de 1883, celebrado con el C. Conrado Flores para ai y para
Santiago P. Hale y Ca., y al de abril de 1895 sobre deslinde y colonizacién de tierras en la Baja California.
<Diario Oficial, Mexico, V, No. 99, pp. 483-484, 28 de abril; Boletin Oficial, de la Secretaria de Fomento,
Mexico, II, No. 1, pp. 59-63, abril; also i in Periodico Oficial, Ensenada, B. C., XXX, No. 17, pp.1 and 2,
20 de junio.

Annulment of land grant of 1883 and 1895 to C. Conrado Flores and Santiago P. Hale & Co.,in southern Lower California, on which
was based the famous Magdalena Bay Co., of Hale, Flores & Co., and its successors, including 2,158, 427 hectares of land.

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién Agraria, No. 648, 16 de abril de 1917. Acuerdo que declara
nulo el contrato de 11 de deciembre de 1886, celebrado con el C. Pablo Macedo y traspaso a ‘‘La Mexicana”
Ca, Europea para la compra-venta y colonizacién de terrenos en la Baja California. <Boletin Oficial de la
Secretaria de Fomento, Mexico, I, No. 1, pp. 72-75, abril; also in Diario Oficial Mexico, 4 de mayo, pp.
504-505.
Annulment of grant to survey and colonize vacant lands in Lower California, from latitude 28° south to Cape San Lucas, in favor
a and conveyed by him to companies entitled “La Mexicana” and “The California (Mexico) Land Company,
1917. Poder Ejecutivo, Secretaria de Fomento, No. 468, 24 de abril de 1917. Acuerdo que declara nulo el contrato
de fecha 14 de septiembre de 1888 con Ricardo Kent, a nombre de ‘‘La Mexicana” Ca. Europea de minas
y terrenos en Mexico para la colonizacién de terrenos en la Baja California. <(Boletin Oficial de la Secre-
tarfa de Fomento, Mexico, II, No. 1, pp. 69-72, abril; also in Diario Oficial, Mexico, 4 de mayo, pp. 503-504.
Annulment of grant dated September 14, 1888, authorizing colonization and sale of lands in Lower California by Richard Kent, for
the company “La Mexicana” and, through transfer from it, by “The California (Mexico) Land Company, Limited.” Occupants of
lands holding titles under the grant may have their equity recognized by proper application to the Secretaria de Fomento, Mexico.
1917. Poder Hjecutivo, Secretarfa de Fomento, Direccién Forestal y de Caza y Pesca, 30 de abril de 1917. Contrato
celebrado con Jos C. C. General Manuel Pérez Trevifio y Coronel Juan C. Zertuche, para colonizar la Isla de
Guadalupe, explotar sus productos y los de los islotes ‘‘Zapato” y ‘‘Toro,’’ ubicados en el Océano Pacifico.
<Boletin Offcial de la Secretarfa de Fomento, Mexico, II, No. 1, pp. 96-100, abril.

A contract for the colonization and exploitation of the natural products of Guadalupe Island to remain in force for 10 years from April
30, 1917.

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién de Bosques, 30 de abril de 1917. Contrato celebrado entre
el C. Ingeniero Pastor Rouaix, Secretario de Fomento, representando el Executivo de la Nacién y el C. Juan
H. Mendoza para establecer, sin derecho exclusivo, en la Baja California, dos fabricas empacadoras de productos
obtenidos mediante la explotacién de fs pesca. <(Boletin Oficial de la Secretaria de Fomento, Mexico, IT,
No. 4, pp. 359-363, julio.

Contract to establish two packing houses for annens products on the east and west coasts of Lower California, at a cost of at least
$50,000, within two years of April 30, 1917.

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién Forestal y de Caza y Pesca, 31 de mayo de 1917. Convoca-
torfa para la explotacién de la Isla de Cedros. <Boletin Oficial de la Secretaria de Fomento, Mexico, II,

No. 2, pp. 158-159, mayo.
Awanmbeneeriant calling for proposals for the colonization and exploitation of the resources of Cedros Island dated May 31, 1917.

1917. Poder Ejecutivo, Secretarfa de Fomento, Direccién de Bosques, 28 de junio de 1917. Contrato celebrado con
el C. General Fortunate Zuazua para la colonizacién y explotacién de la Isla de ‘‘Cedros” situada en el Océano
Pacifico. <Boletin Oficial de la Secretarfa de Fomento, Mexico, II, No.3, pp. 267-271, junio.

This contract authorizes the establishment: of agricultural colonies and the exploitation of the natural resources of Cedros Island.
Permission included to establish a port and installa wireless telegraph station. The concession to extend for 10 years from June 28, 1917.

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién de Bosques, 3 de julio de 1917. Contrato:celebrado entre
el C, Ingeniero Pastor Rouaix, Secretario de Fomento, en representacién del Ejecutivo de la Nacidn, y el
C. Luis M. Calderén, para establecer, sin derecho exclusivo, en la Baja California, tres fabricas empacadoras
de productos obtenidos mediante la explotacién de la pesca, <(Boletin Oficial de la Secretarfa de Fomento,
Mexico, II, No. 4, pp. 365-370, julio.
Permit to establish three packing houses for fish products in Lower California at a cost of at least $30,000. The permit covers a period
of 10 years from July 3, 1917.

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién Forestal y de Caza y Pesca, 11 de agosto de 1917. Contrato
celebrado con el C. Gastén Vives por la Compaiiia Perlffera de la Isla de San Jose, para explotar la, concha-
perla en dicha isla. <Boletin Oficial de la Secretaria de Fomento, Mexico, II, No. 5, p. 455, agosto.

An emendation of the contract granted July 5, 1893, to the Pearl Fishing Co., of San José Island, permitting the cultivation of,
and fishing for, pearls and pearl shell about San Jose Island and along the coast of the peninsula from Cape Pulmo to Cape San Lucas.
Dated August 11, 1917.
No. 1]

LOWER CALIFORNIA, NATURAL RESOURCES—NELSON. 171

1917. Poder Ejecutivo, Secretaria de Fomento, Direccién Forestal y de Caza y Pesca, 31 de agosto de 1917. Dispone

esta Secretarfa que los placeres perliferos a que se refiere el decreto de 25 de abril de 1874, asi como los que
se encuentran a lo largo de las costas del Pacifico, quedan divididos para los efectos del buceo de la concha
madre perla, las dos zonas o regiones que se explotarén alternativamente cada dos afios. <Boletin: Oficial
de la Secretaria de Fomento, Mexico, II, No. 5, pp. 460-461, agosto.

An order that the pearl-fishing grounds on the west coast of Mexico be divided into two sections which are to be fished alternately
every second year. The first zone extends from the mouth of the Colorado River to Point San Marcial on the east coast of Lower Cali-
fornia and to the mouth of the Rio Balsas on the mainland coast of Mexico. The second zone extends from Point San Marcial to Cape
San Lucas, and to the border of the United States on the west side of Lower California and from the mouth of the Rio Balsas, on the
mainland, to the border of Guatemala.

1917. Poder Ejecutivo, Secretarfa de Fomento, Direccién Forestal y de Caza y Pesca, 6 de septiembre de 1917. Se

1917.

1917.

declara insubsistente el contrato que celebré el sefior licenciado Luis M. Calderon, el 3 de julio del corriente
afio, por no haber cumplido con el articulo 31 del referido contrato, en que deberia, en el termino de trienta
dias, enterar en la Tesorerfa General de la Nacidn, $1,500, oro Nacional, para la ereccién de. tres empacadoras
y explotacién de la pesca en la Baja California. <Boletin Oficial de la Secretaria de Fomento, Mexico,

II, No. 6, p. 522, septiembre.
Cancellation on September 6, 1917, of contract dated July 3, 1917, in which Luis M. Calderon is authorized to establish three fish-packing
plants in Lower California.

Poder Ejecutivo, Secretaria de Fomento, Direccién Agraria, 10 de noviembre de 1917. Se declara insubsistente

el contrato que celebro el C. General Fortunato Zuazua, en 28 de junio del corriente afio para Ja colonizacién
y explotacién de la Isla de ‘‘Cedros’’ situada en el Océano Pacifico, y haber faltado 4 lo estipulado en el articulo

32 del mismo contrato. <(Boletin Oficial de la Secretaria de Fomento, Mexico, II, No. 8, p. 608, noviembre.
Cancellation of grant to General Zuazua for the colonization of Cedros Island, owing to failure to comply with the terms of the contract.

Stincer, Dan J. Desert Trails. <Field and Stream, September, pp. 413-416; October, pp. 503-506; No-

vember, pp. 588-590; December’, pp. 678-681.
A sportsman’s account of a hunt for mule deer, antelope, and mountain sheep in the desert plains and in the mountain ranges ofnorth-
eastern Lower California. The gray wolf is erroneously mentioned as having occurred in this region up to 1903.

1918. Secretario de Fomento, Como Estaba Dividida la Baja California. <Mercurio, I, No. 22, pp. 1-2, 16 de marzo

1919.

1919.

1919.

1919,

1919.

(Seminario), Mexico.

A statement by the Secretary of the Interior, with a map showing the location of the land grants formerly held by companies and
‘individuals covering nearly all of Lower California, but which have recently been annulled owing to their illegal character and to the
failure of grantees to comply with the terms of the grants.

Asuurst, Henry F. The Purchase of Lower California. <(Congressional Record, vol. 57, No. 29, pp. 1148-

1158, January 7.

Speech of Senator Ashurst in support of a resolution introduced by him in the United States Senate, January 2, 1919, that the President
be requested to negotiate with Mexico for the purchase of Lower California, and a tract of about 10,000 acres on the northwestern border
of Sonora, which would give access to the Gulf of California on its eastern shore. The resolution was referred to the Senate Committee
on Foreign Relations and no further action taken. The speech contains considerable information from various sources concerning the
physical character and resources of the peninsula. :

Dickerson, M.C. Diagnoses of Twenty-three New Species and a New Genus of Lizards from Lower California.

<Bulletin American Museum of Natural History, XLI, Art. X, pp. 461-477 (New York), October 2.

The new genus Sator and 23 new species are described as follows: Ctenosaura conspicuosa, C. insulana, Sauromalus interbrachialis, S,
townsendi, S. varius, Calli, US COT nsis, C. inusitatus, C. plasticus,C. splendidus, Sceloporus lineatulus, Sator angustus, S. grandz-
vus, Uta concinna, U. mannophorus, U. parva, U. squamata, Verticaria cxrulea, Cnemidophorus celeripes, C. disparilis, C. estebanensis,
C. punctilinealis, C. bartolomas, and C. vandenburghi.

The new reptiles described above were obtained by the Albatross expedition of the American Museum of Natural History and the
United States Bureau of Fisheries in 1911, and are mainly from islands in the Gulf of California.

ForsBes, ALEXANDER. California: A History of Upper and Lower California, from their First Discovery to the

Present Time, etc. Royal 8vo, pp. xii, 372, with plates. San Francisco.
A reprint of the original published in London in 1839. It contains a general account of Lower California, its Indians, resources, and
history.

Funcge, E. [W.]. Hunting Antelope for Museum Specimens. <Field and Stream, March, pp. 834-836, with

photographs.

A short, popular account of a hunt in northeastern Lower California made for the Brooklyn Museum of Arts and Sciences, with brief
description of the desert and the birds and animals found there. This trip was more adequately described by Robert Cushman Murphy.

Gorpon, P. R. Filming the Sea Elephant. <The Wide World (magazine), October, pp. 485-489, with

illustrations.

A popular account of an expedition to Guadalupe Island in July, 1913, to secure moving pictures of elephant seals. A second trip to
theisland the same season revealed the fact that sealers had raided the herd soon after the first visit by the photographers and had killed
or driven away all of the animals.

1919. TowNsEND, CHARLES Hasxins. Lower California. Access to the Gulf of California Versus the Acquisition of

the Peninsula. <New York Times (newspaper), January 26, editorial section.

The author opposes the acquisition of the peninsula as unwise but favors the acquisition of a strip in northwestern Sonora soas to give
access to the Gulf of California for commercial purposes and in order that the rich fishery resources of the gulf may become available to
the interior of the United States. Some ofthe valuable food fishes and their abundance are mentioned.
Memoirs National Academy of Sciences, XVI.

  
 
   
  

 
 
 

117

°
qe

 

 

: 8
Coronados 1s. 4

  
 

Sack
de CANGIO

  
   
   
  

 

 

           
 
  
 
  
 
 

 

 

    
     

Cl ze a ENCING He Sx
c eae ee
eNta Grulla yack.
Deine Linstead ORAS
Bto Tomas aay! ae \
aD ea ah
~
\
\
oN
ens
See
a

Consag cock \
ae \
“tay Pr San Felipe acd \

San Felipe 3 x _ \ a,

San Felipe Bay ~

\

\

Z

5 Gi Ysiderc NY \

Mw = oS ma \

 

   
 
    
   
 
 
 
 
  

   

i
\

ey SY Rosarito a
: : C8

rihed

> :

Vigo) Pt. San Fermin ‘
. Sd

SA3

   

- Se °

ge, ee krarar landing

\ be Ysan Luisi. we
Mesqitite

we
&
a Pt. Final
Na

 

 

 
  
 

ANGEL

4 }
DE LA
GUARDA |. oe

 
4 Orvetrpec Bas
1} P u

f

4

1

|
:

aN, Pr. San Felipe

Consag rock
.

 

 

 

  

First Memoir, Pl. 35

 

 

  

 

San Felipe
San felipe Bay

 

 

 

 

 

 

 

 

 

30
 

 

 

SXernando Pt.

 
   
    
       
   
  
 
  
 
  
 
 
   
   
  
 
 
    
  

AU

= PRN ae
amay ues, Wi,
5°

  

$.Carlos Pr.
Biuff Pt,

   

   

 

 

BSs x
IS { Chapala,
‘Lax’Paloriti

 
       

 

 

 

 

 

Ss =~ and, é

TIBURON].

 

  

 
  

Woe
x
pu SanFsteban |.
Rosalia Pt. %
o
sta RP”
San Benito ls.
& g Cabeza de!
Caballo Pr
i. Santo Gomingo landin
Jick
’
meee SEBASTIAN VIZCAINO _//\fibekWe
aa =~ __ RedondoPt = : i
BAY
Nativids |
¢

 

es

x

Pt.San Pablo
= B
s

ue
ai Roque” 4 ENS
=o *y
Sc Asuncion |, BY Paes 7
civ BONN at = ait
Asun x ~P-- ° 2a de StaClara
1 }— Its ee aClara

 

 

 

 

      

  
    
       

FtSarHipolito

Sten

sil Ra

“sgl ce S4
yyy Ay

 

         
 
  
    
 

  

flallenas
f Bay

  

AbrecjosPt

 
  
 

AAS
AH Ss
= Pe

Pt San Juanico

 

 
     
 
 

? Crvwt-pec Baz
1
:
F

|

Consag, rock

o
Sy Pt. San Felipe

 

Blanca Pt! \:
Blanca. B.

 
 

 

Cabeza de!
Caballo Pt!

Santo Domingo landing

ASTIAN VIZCAINO aha veer

BAY oO
S_

 

 

    
 

 

   
  
 

 

Sanfisteban!

Miia eS

TIBURON].

 

 

 

 

 

 
 

 

 

 

NATIONAL ACADEMY OF SCIENCES
Charles D.Walcott , President

Reconnoissance Map of

LOWER CALIFORNIA , MEXICO

COMPILED FROM U.S.HYDROGRAPHIC OFFICE CHARTS WITH

ADDITIONS FROM ORIGINAL MAPS PUBLISHED BY GUSTAVEISEN

DAV1D GOLDBAUM AND GODFREY SYKES, MANUSCRIPT MAPS

BY E.W.FUNCKE, AND OBSERVATIONS BY E.W.NELSON AND
E.A.GOLOMAN

UNDER DIRECTION OF
_ EW.NELSON CHIEF,BUREAU OF BIOLOGICAL SURVEY
U.S.DEPARTMENT OF AGRICULTURE
1919

Scale 1:2,000 000

° 10 20 30 40 50 Mi.
i ge ee
Legend
O Town ====Wagon Road
o Ranch -~--- Trail
@ Mission — Rail Road
pa aWaterhole

 

 

 

 

 

 

Route traversed by Biological
Survey Expedition, 1905-1906

115

RAS.
Pt San Juanico \4 :

Pozo Gras
SanJorge

 

Shue

a

%

Ve

 

C.SLazara / wy SSNS
an SQ
Hughes mY xy
elo
© Cerse (
'
|
Entrada
Red

 

114° n3°
CE LE CNL LT ES TIT BE TE SPE IP TEE IPS OT TT EE i I I TE EE IEE Ea ETI NE DS I
 

 

SanJorgé

ls a orngo”

 

oO FA Quepe wv

Ihe

at

Si

i By
Solednay at se
an of TS \
| favances URE

  

=
uo

 

 

 

 

Entrada Pt
Redondg

 
 

 

 

 

 

 

 

 

 

  
  

2 NG
+) eee

-" ~
Pe

  

  

‘} Sta Catalinal.

 

P.SanCarivs
imbwliehe
*RocaNegra

Sta Cruz i

«ls Morena

 
    
    
      
  

  
 
 
 
 
   

 

    
  
 
  
   
    
 

 

ei
ey

&

BS AES
A ROSETS
Saas war

 

 

ista Partide

  
   

Espiritu Santo |.

 

   

 
  

 

 
INDEX.

 

Page.
AbeOn Ge eicesecseiisiees cae na olee yds aauwvececace + 135
Abbott, J. F., work by.............- 2 Soveaem eres 145
Abies concolor.............02..c.0c eee eee ee eee ee 103, 182
Ab Wtilon espe siccssc ono. senda note sac souvedas nee sank 89
californicum..........2.00..0.00. 2 cece eee eee 123
AIMEE Fe ister Sereda lenre wing eases See 123
MCR Gia so Socata ia te sceashlels ne leek: 28, 106, 107, 109
MeRUCA USS: sie ve vgdweiene shi excueekeeeeiss 123
BLOOM ccm tires u/s8 vderd ascynleie dele Gama ceceieee 118, 125
Acalypha californica...........0......-220-02000- 126
ACCIPIVGT i220) oseeines 2c nnceaiedercmanoa tees eneeas 115
Acknowledgments.............2..20ceeeeeeeeeeee 12
ACUUIB 22 osha ha nde BosAd Vo. Ganiele eek es 116
Adenostomascc3jcck ess oheee ou aa etene sees 48, 75
fasciculatum :o..-cedececcewexs 3 eaeeteekeres 104, 129
sparsifolium.....-.......-------e-e eee eee eee 129
Adolphia infesta........-.......-200-2--ee eee ee eee 129
Aechmophorus...............-------2---eee cee eee 116
Aegialitis (see Charadrius, 116).
BNC) 0 41.1016 2 eae ee ee ear 115
Aescultiéscsseueteassssemene Sue sie slsere sek ese 104
PANTY s sae cseiecess sstisetsee ss eoetsiasereses 129
Agave..... . 25, 30, 106, 107, 109, 119
AUPREAd 2 cashthaese teaches scene ewiaaans 125, 126
Cerulatacveccceuy.cchanscmseeesoewass Mheseoes 125
CONSOCIR Gh oss sax cee Ske aera anaes ts 125
goldmaniana.............22-.0-0 02-02 scene neers 125
MAPCATIAG sic 2 ones vedivetie sui e sss ees eieniels es 156
NELSON...) wa sadecvesaaw esses ses edeevereeee ses 125
promontorii..........-....---------0- 22+ + 125, 126
NE < eK Gl acncd sae oasoes eens sedeeaeaicetee's 125, 126
Agelaiuscec: vaseinc'ss og ectecene ees ewenee semen ede 116
“ phoeniceus neutralis......-...-------+--+-++-4 130
AONOTICNSISi cs scwasccar os ecwieeses eee ess 126
Agriculture......-.-.------ 0-2-2 2e eee eee eee eee 136
* COloradé delta oc ccccccc cogs es nsiseeeeese estes 137
Magdalena Plain..........--------+---+-+++-+- 138
San Pedro Martir Mountains......---------- sae 187
Agua Caliente Canyon....-..-------+-+2--22-2-2-++ 57
Colorado ranch....---------------------- teeue S41
Dile@ccccaetes cece se bse se oedesse eee eee sets 25
Grande ranch....----------+----+--- 2-2 etree 36
Aguacate......------- DhGais aie Yee eSeen 2 leas 140
Aguardiente........-------++2-cee ert errr eee 140
Aimophila........2200-02¢2eec eee eee ee eee 116
™ wuficeps ruficeps..---------- tence eee ee eeeees 130
< Banlieue cs os loam deceteae _ 121, 130, 161
Ajaia.. 1.0... 20 ee cece eee etter erent 116
Alamo mining camp
© Plain. ....---- 2-22 eee ere rr terete cece
Albatross (boat)
Albizzia.......--------+- 22-222 -- seeeeee “eee tees
'' oecidentalis. ...-----2-0- esse cece eee eee e eens 128
Aleman Campo.....-----+-++++-++++++ we seeeeeeee 30

 

Page.
Ua fais cssies ceo sisacd nek cake doe sulague 137, 138, 140
Alligator lizard ...-...222.2.02020.0.0c0e cece eee eee 86
AM iyvial tans tee oo dee eeelce toys ckaeeeeeeces 78
Almeida Bay ocin accu tawiinsd duckie Bere eemanrnactane 40
Almonds 22 cece s ssc. An eeneindie ae teeanuiass 24, 139, 140
Aluco (see Tyto, 115).
Alvordia fruticosa.......2..2-00- 0020 ee ee eee ee eee 123
Amador, Francisco.......-.....-----eee-e ee eer eens 46
Ama laics cscsceeeenc cameaeademes seed se eer caciinee 149
RANGUS 6 o,.6 ouios Jeeta ca astaca fener eee 149
American Museum of Natural History, work by... 146
Ammmodramus...........--20--0e cece eee e eee eeee 116, 161
MAIOPHUUSs |osesie oo: scatateist vahvdieseiare sv ag sisleteiee aieiave 161
BANCHOTUM aos cesacecwee ed eae vee see aeccvee sce 160
Ammospermophilus.............--.--------- 92, 110, 117
leucurus insularis...........-..-.---.-------- 127, 165
leUGUIUSsnseccceseseenw os suelo salaneinaeeies 127
peninsulae............2-..-22----2 2-6. 121, 127
AMPhiblans jaiddeieseases vWetceatewkaan ese eee 110, 113
Ataphishaeni dae sic 2cccisceicwasistec e's o-cpeeleciebedele 114
Amphispiza.sss:sccnscescaske cen 45k Sewereeeees 112, 116
Dbelitieciccsice es ccepeeerdeme se es seadweeseeece 130
CINELOD secisceicitiieeeseaniensnenadacdeekauis 156
nevadensis cinerea...........--.-----0- eee 126
ANAS ssacesvseveswdisseatiees sdese ve sess coe xes seers 116
Angel de Ja Guarda Island..............---.2.22--- 93
Anglo Mexican Colonization Co.............-..--- 35
Animal ie oe ccasecssjssetehenecasts samcess ies 132
Anniella pulchra.........2..22.2..22.220222005 114, 115
ANSO? o/s sawietesectcssis cee sesedeeetesse sess seems 116
Antelope........-2:--220ee20e00: 34, 41, 110, 111, 132, 133
SQULNTEL -ajsic peadeeeitesecgkeiia eeelcsaeseeee 110
Anthony,A.W.,work by.. 12, 21, 85, 87, 88, 95, 143, 144, 146
Anthony Peak (named)............-..----------- 12
AMGRUS: Ssicasse Sei cakse gure nocoed ts vices cua heen 116
CORVINUSS .cs.2 ae ea ean eadi sas oe aaemeR nad 112
Antignon leptopus,....:........-.0---.----------- 123
ATUGIOCA Taio: wn dcecnetededcadee Uisnudecee essere 117
americana peninsularis................-. 127,167, 168

Antrostomus (see Setochalcis, 115).

Antrozous......-..-2----- 2202s e eee eee eciekiar ee 117
WMMIDOL sep sass tach seewdtacsewanwaesseeae eee 162
pallidus. Jo saosewatiean ees slciesinesaees's es 160

WINS ¢ sires ade te cunleee sv eexsebescers + 121, 128
PAClficus.: scswsesieeweese es boeedeaaatee 160
pallidiss: endear nin seceseekisdretcet 128

Aphelocoma..........-0..--00eeeeeeeeee eee eee .. (108

californica hypoleuca.................-2.--- 120, 126
OWSCUTAS 3s fo s2kieed ce att tesa cece Sess 130

AMhri gain Ra gaes eee A Accene ss ewes iauice dss 116

Apples ssaiei wince ooh Btn ase dela cae aecmnmnasies 71, 139, 140

APUICOtS cos Festina ele nrg de Gialehete 139, 140

Aquila sive sea eems v's dae eee oe) Se eeeeeee cores 115
 

 

174 INDEX.
Page. Page.
Arbutus peninsularis.............-.......-- 109, 110,129 | Auriparus........... 2.02.20. 2 cece e ee eee ee eee 112, 116
Archibuteo........-.......-2-2-2-5-- clueless 115 MaVICOpSssicisehicarceemeeceA yas sees seeds 156
Arehilochtis.coccciee nets cosscessacessostciwaks 115 MA VICE PS ise Seeee cokers beiea 6 bene 126
BORAN ATs ge. csrsintgas wee aiiatlarey eter ielskbly -- 130 ‘lamprocephalus............2..22-+-- 120, 126, 160
Arctocephalus townsendi. ...........---.--- 95, 160, 161 | Automobile roads. .....-........0--2-2-- eee eens ‘
Arctostaphylos...............2.2-22. ssineiiaedaeie WO;,LO& | Avocado sexcscscseeeeewstieanese ssc. sce peenceses 43
Dicoloresccdaxe stews os ewe ete coaceeeees & 129) || Baccharis.. ccc. ccb5 9 8oses eve ciscad caemine ve ede 107
PlAUCRs. Sesh es ee ee edd asad eaokeeneeices 42D | BRAS OR i7 sie ieissc tis vada td Caco ceesiegeices nave sk Seasons 110
oppositifolia ........... 22.2222 eee e cece eee eee 129 } Baeolophus..........-------- AG ee gamenaaemuapes 116
PHNglel sees s 25522 ccs aeneeesaae sseudicablccis 131 inornatus cineraceus...............-.-02+5 . 121,130
PAT OC Bh sake yi 2 aoe Sty esa ete ett A Ae 116 MUPINUG: cc cccccccccadsesccccccasaagews 180
herodias sancti-lucae......---.----.----- 120, 124,167 | Balanosphyra.........-...-.-2--22-+-02020-e ee eee 115
treganzai.........2-..2--- sea got eaeenmn 126 formicivora angustifrons.......-...-.....-++- 121, 130
virescens anthonyi........ ee eens 159 bairdl , js.aseoes ae hee vetes eons use 130
WAZATE si vesejdsduss un ckee cemeee Weeieniae 153 | Balantiopteryx...........-2------02eeee eee eee eee 117
ATG ena: too nico pee ee ieee aes ae siete eee H16:|' . plicataveses veces secdacnceare ss oeeye5 ese 117, 121, 124
Aronaria. = c's <4 steewier dered iadumneeaeee seme 416) Ballenas: Bayicn. ose: vcees se eeeeaeee sas Suede 72
Argithamnia brandegei.........-.+-2--22--0-+--+5 128, | Bananas. o2220-scoxsegesseeveseweceees 35, 37, 43, 138, 140
Arid Tropical Life Zone..........-..---+-----+--- 121,123 | Barkelew, F. E., work by.....-.--.-.....-------- 144
binds? se scscss yas cee ees eee eee ess , 228) | Barley. o.oc oc osc ccainsaicecesicccie 24, 137, 138, 139, 140
MaMa Bo s.2hsci5c1nsicpayacc bs Roath orerecieaeradesianienio’ 1A. | Dairaeuda:. coe sacieeds cegnadcc ceases sees ce eees 134
Plants: = 0.521. 5 acacia uaseleiietice cheeiee eae ess 122; | Baril Landing: «5.0: cccinccccccce caees adancisieieeiees 31
POPES Eg cc.2cand ccm natanoexacnnalseanmanadas 123 | Bartsch, Paul, work by............-.-.-.-+-+------ 146
SATTABUTHS 325 ku Genoa ase wae ee BAe 26 | Basilinnay 32420005 paces sesicsdceea ye wisest : 115
ATTOW WE6d sco scsde52 du onaiunsemonesc scale vege 48 KANG eastcse wees seeacceseeees 121, 123, 180, 149
APrOWWO0d oo 20% )22 sosecccnassscceersesessiseeciwns 107 | Basin, La Encantada...............------+----0-- 57
Artoyo Calamahue............-----.-----2-2-0005 74 Salita Rosa esen ses de i occteatasmmdeedann sei 57
© WALAVIMA Sco. ta sseidongestie esas ncceneianes 76 Vallecitog: s.ce0cchec geese ee gusasacttassaas 57
Colorade sui siesastenssatek ia ekeieen eh eet As | BOSQUeS 2 s'scicisic vin vocieattewioe sacs eceutienemieeeane 15
COMONA We secede eacaeeaa seine cieiacatels 39: || BassarIsCUs:). i. ecccleaas sie s cc ee s c'sre aisisiante dee 35, 92,117
BI Tule ss: aeisaesneteseestecesuds 19, 25, 55, 68 astutus insulicola......................-- 92, 128, 165
palmarius..cccssseseek ss see Sacesde 121, 128, 165
SAXI1COlA: os seener sess oo se csieeaniser eats 128
saxicola: < 22: cesassecewex eee sess saceeeeeeweee 160
Bassarisk .:.cness sosvecveaseeeagse eoeasaceave 33, 35, 92, 110
MOON 2 sk GeeeGiiee Gee be eeedeeee soe 20 | Bassaris raptor.......-.-.--- Guietaeraechatwusedecis 160
Matomi.......... Psaecaosiaiuneseseke 57 | Batty, J. H., work by................-..0-200000- 146
Pabellon..........-.-- Er nmeriebmiaseee nee 61;-|) Bay: Alme asian ec ccsse ewes sesmaceeeccs heeces 40
Potrero.......22..00200020ee 35, 36 (desc.), 61, 74 ~Ballenas.............0-.0-002cecceeeeeceeeee 71
BanvANgel soi ecsetaveganuiewwmmincacemes oe 34 | CONCEPCION 3.55 'o:sclsceeintieiare'v aa. ty Seavvocienss welsls 78
San Antonio...... Pic Siecceasecwed eee 46 TG Pag os scat ssasereacaegeeas new secewesee ses 42,73
San Bernardo.........--2-/..0se-e2ee eee 46 Magdalena, desc...........-------2-2-2-e00ee 89
San Tenaciosai0. wucwt wanda cchietee Se 34 Playa Marie j:ccccey ote s oe seine eteiseais 29
Ban Jose ouaiite ida euntatclsss 36, 61 (desc.) San Felipe.............220.22000- 17, 18, 19, 51, 69, 70
San Rafael..........----- itsjedsreaetstete wetels 15, 20, 75 San Luis Gonzales.......-.........00..-.22005 74
58 San Quintin 2. ssoe0s cs ee sradaasinnastencamens 71
104 Santa Rosaliasicos- <'x's's's's'cesinstvews oaenkeuas 29, 71
129 Todos Santos.-.:ccesc:e7s52 4444 eeesedeccewede 13
129 Ventanase-.cneees iseccs sxe oes cdouseessewcces 51
104: | “Beans yaecteeicee etc eeulsewcts <cesenis 24, 33, 139, 140
es 115 DiesqUitesa/025s2censveee ee eetiseceees eae 186
Aspen, quaking..........-..2..22.0--02 eee seen eee 104 | Bear, grizzly........-.---- 22. - eee eee eee eee es 110, 182
Asplenium blepharodes...........-.---.-------+-- 156: || Beaver scecccssisasd sche sincesibame aoe wes teens 110
septentrionale..............2.2202-22ee eee eees 158 | Bebbia atriplicifolia.................-...222.220-- 123
Astragalinus..............------- sd viel ol Setarahdieeaisteiat 116 | Beck, R. H., work by..................02202- 19, 143, 144
LO WYO COI sar. auie teiea tiie c-ss's- 2c dsia's's She g's oecccrsiate 130 ||| Be@SWAX sc csis cane sg os aa wisinsivineainntaaaniscis cheilsieale o 1389
psaltria hesperophilus....................20055 180 || Beets. sc actseas se ses scecccemiscse eter se iesd sis es 140
Asuncion Island.........--.-.----- Soames 89 | Belding, Lyman, work by..............--.. 46, 47, 87, 142
Atamisquaea emarginata.............---2.2--25005 123 | Beloperone californica....................0.- 00.00 128
‘Atriplex o2s.cesaccatataaedaus ctx ceed seeacaeunces 107 PULPUSl veces sees vs cosas oseeetwaews see. see 123
Vinéaridesecceeccdacsdeduesatdanaas same tme 125 | Berberis fremontii..................0....200-0-0- 129
Auklétsccoeksnscsckechiie Sv ktte tia aasaeneanes 85, 86 | Bergerocactus............... 2... s eee eee eee eee eee 106
aes Bi. sista edieaus elec 'e o'ga ea Sesser 88 CMON Lise bc da sateakie eae tiemedsaeidddomass 129

 
 

INDEX. 175
Page. | Cactus—Continued. Page.
Berrendo, Lan 02 oeeeacn ge ossceem Maes saeco 72 gata pitbulls: kecancvoiwuieekonweteeeweune 28
Biological Survey Expedition, Narrative pdeeeees 13-48 giant........... 24, 25, 31, 35, 36, 41, 42, 43, 89, 109, 119
Bipedidae:..........-..----.-+--22se22 eee e eee 114 | Caesalpinia mexicana californica..........------ -- 128
Bi pes sinisis ssiaia sig eclerteeenies oa damtiedwes ica cease 114 DIRGUEtaanusnsisddeses aco ateieeiicses bias 123
St ee 114,115 | Cajon de los Reyes...........--------+-+-0ee 2s 42
Bird lists.......... 90, 91, 92, 95, 120, 121, 123, 124, 126,130 | Galamahue Canyon............+.2.2-2eeeeeseeeees 74
Birds, Cape Faunal District..............-...------ 120 Tandine canes y oo iclecnessedey aioe pine sianeecee 27
Arid Tropical Zone.................--0--+-+ 123, 124 MissiOiinc crset vu cmclblcetlemn eo weet ied 2 27, 74
Canadian Zone...............-.0252eeee creer 182 MOU TALIA. cee con ciel etic te gwar eh oeetes 27, 74
Lower Sonoran Zone..............2-2--00-0- 2-8) 126 Baa lossce paces sar geese wisi hc Bese 56
Transition Life Zone............-..--.-+++++- 131 VAR carded anasanumiieioeewiat 26, 27, 74 (desc.)
Upper Sonoran Life Zone.........-.+----++++- - 130 | Calamospiza melanocorys......------+++++-+-+-+- 112
Blackberty sco c0.4gceisdiesce ese Neesesedeee: EVO "lal epi is Wace etc idte etd sensi ON ea ata eas 13
Black Warrior Lagoon............-----+++++++++-- 71 | Calidris..........-1--.e0--- Boyne a RP eS 116
Blasipus.......-.-.--- 22222220. eee eee e eee eee 116 } California Academy of Sciences, work by...--..-- 88, 143
Bluefish....... pte eee e cer eee eee eee eee eee de 184 | Callisaurus carmenensis...........-.------- 114, 115, 171
BODCAt gcse us sceaens Y yyse ewe we coment aie cole gts 132 PHT GUS iets dees ane AA ae cts 8 he tle ae 114, 115
Bombycilla...........2.2---22 22220 eee eee eee c ene 116 | draconoides.......222.-.--+2eeeeeeeeeeceees 114, 115
Booby, blue-footed...............-..2---2-+-050+- 84,112 |: inusitatus.......ccc-sccecceccecencecseecetes 171
\ Brewster’s.--..--..2--- +000 2e eee eet ee eee eee 84, 112 Plas ea dceeusichasium een iwededsnce lay ele el
Botaurus.....-..---- 20-22-2002 eee eee eee eee eee 116 splendidus.........-----++-22eeeee2e+ 114, 126, 171
Bourreria............-. eee e esate eee eee cere e eee 108 ventralis ventralis.........-.-.------+--- 114, 115, 126
<>) gSomorae.....-. cc earstbara tein 2's, elabanshayaenrouue yee sel e aae 925 123%) “Gal pia Ud csccarsmoctege heawusicissisieis eee b xa emetacateees 9,30
Brachyramphus..............--+-------+- ait egstcieias 116 nT So Oe tates, cettaatetnee nD nese Ro ed an hate 81
is bypoleucus. ...-.-.-.-------+-+- 222s eee eee 149 population =isevcescccadssessrnagbesunaviases 10
Brandegee, T.S., work by...--.---..-----+ 142, 143, 144 | Calypte.......-...----- 222 ee eee eeee eee eee eee 115
Brant, black ....::.-.---2-----+-++22+-e eee '.. 184 | Camino de en Medio.........-..------00-eeeeee= 14
Branta . . 2... 2. - 222g ee eee e ee eee ee cee eee eee es 116 AEN COM lO fockiceiniot tenia oss aaah Gores 14
Brayodendron texanum......002.--2-.-.----- 91, 122, 123 del Pacifico.........2.2-20202cec eee e cece ee eee 14
Brewster Expedition, work by......--.---------+- EEN ey Oia ceca ete ee pentane ie ok ata cea eal. fol ical 33
Brewster Peak (named). ...--- eee wera eens 12 | Campbell’s cattle ranch.....-......-.--.---- 16, 19, 20, 76
Brickellia hastata.........-------+++2+--e++-+- weg IDB | (Campos. c.csatnensnesagenece oes cdamsmeedee <k'ts 15
Brittle bush...... didinie Simm ninieid ne So Sele e seid seat ss 107,109 | Campo Aleman...........2--20.-220002b ence eee 30, 71
Browne, J. Ross, work by.-.-.------ beseiald ind: Siecese cee 141 MINE xiut-oos ataweanwen esas ensekRUoees ose ns 31
Brown, W. W., work by..--.-----------+-- 87,88, 95, 146 | Campylorhynchus affinis........-....----- Veteee. = =149
Bryant Peak (named) ., ahs ay Sia's sxlaapemenee es ane 12 | Canadian Life Zone.........------------e-- eee ee 132
Bryant, W. E., work by..---.-------- 45, 95, 142, 148, 144 | Ganerum.......------ eee eee cece eee eect cece eees 140
BibO. accesses. cc temas es se eeeeeeees 115 Sugar, see Sugar cane.
))) vVirginianus lNGHIBbWN, 5. ccigaceeaseaeeses 126 162! lh - Gana =. .sctciole aac on oa bala Relate eo aee Memes 117
be pallescens........-------+--2eeeeeeeeeeees 126 ClepHOUd sicicic ashen st adeaneeaete z. eas ctans 132
Buckeye.-....----------2222eee cece eeereee poetee 104 OMSL oS lus seed t rh naurganicoacareacaer sede 167
Buena Vista....--..-------22----+- acne eee ee eee 18, 19 VesteS vias c's’ sv neateuseoeequeesen stk agnsheains 127
¢ Bae... p.ceccaees 2 Seeeecau's c++ sp erisentcneee 18, 70 MGAaTNSl ee yo. vases eheea nie eee reas 127
Buenos Aires, DGG: Gn sesc causa uss reo nema » 25 peninsiilae. .. ss sssecccbseeeese cs exe's 121, 127, 160
Bumstead, Dale, work by..-------------+---+++-- 12 | Canyon, Agua Caliente............-----.0220220-- 57
Bunnell, A. S., work by.....----------+-2--20++ . 86,144 Arroyo Matomi.............---220222-eee02-- 57
Bunting, lark.........------------+--eeeeee eee --- re Calamahue.......-.----- 2-02 ee cece eee teen eee 74
Buried mountains ........-------+---+¢22eeceee- 69 Comond thse veeeestencaacsexonaceaieans 62
Bursera microphylla.......--------++++++-r800e 5 136 de Cancio Ranch.........--..+++--++2e+2eeee- 48
Buteoisiedectose snesasaeceses aed fnoaee saws saves 115 Diablos. g224ses ss se speek aces 17, 21 (dese.), 56
Butorides......-.-----------eeee eect teecteee 116 Esperanza........-2--2-20-2222222++-+++ 17,19, 21, 56
_ virescens anthonyi.......----------+:-+2++++> 126 La Providencia......2::..--:------- 17,18, 19, 21, 56
123 GHAZANL dieerbodseuiate se onanoaenne SoS 120,124 Ta PUPS oc ccenecagcceannnshenticececyawes 62
Butterfly . ...-.---------- 2-22 e cece reece 113 Paral. ccsocccke kirsis spinnin apenas 57
Buzzard.......------2-2- 02 e eects t eres 25 San Ja Viele sees seh ese ncened angdasseuks 2s Gsecey 30
Papbage oa nee eee eneececee rnc eecncregrrsresarctees 140 San S086. + s2serek avec ene se tssece bere thee canes 37
palms...../.-2---¢00esee eee eee ee seteeeeceees 8B San Pablo sencchninons sane danas weve eset 32
Cabo San Lucas, desc...........-- Nets eaeiste SYS iapatiss 45 San Rafael............2-0.02- 2c cece ete eee eens 56
Cacachilas Peaks .....-----------------ereccrrtee 64 DAD! SUNON essen os bag a cheeet ania eo oy eeleares 56
CaCHI.: ooo earesec eet amen eid 17, 25, 44, 87, 109 Sam Telm0... cc ceeceeccc sccvaeses vend scsueeeaes 56
py ihe ad on celle eeoeeennce arent ase 85 Santa Rosa.......2-2.202-0lenee nee eee eeeeees 22, 56
‘ HGCPING arcinecee 5 deere sda vaeeeeni ee sae eeeee 41 Santo Domingo.........-.--.-----------+---- 56

158961°—21——_12
  

 

 

 

176 INDEX.

Canyon—Continued. Page. Page.
Sauzal etek ts dee less Lae cat AG || MOPCAAUI Tos sh cess c ameseiel es oats oprtarerrniaie 28, 106
Wind Vie. daiwa ser gaccahe Teueaencsesens 17 peninsulare.......0.20.. 0. ccevedeceesecedecess 128
Writs cr aos Becca yane naan Moonee ete 92 torreyanum.......-..........-...+------ 107,118, 125

Cape birds... LAnlwhimidnacsiewsccbeldemsceseas 1204) Cetcocarpus-.ees 602 oe kisagocsce's clacesceaces.sees
Faunal Biswiees. Sj ntstcaeeste I eee pach dae as cchchia 119 ParVOlUses <- cscs swieee mond oe ob decease
MAMIMAIS...c ccc cceceucmocccckee dee ccese ced. 121 | Cereus cochal.............00200 020 e cee eee eee

Cape San Bugenio...........2...0000--ecec eee eee 52 phoeniceus pacificus
San Fermin..........2...0..0 0c cece eee cece 68 F353 1116 \- Se ne ae
Sam Tal CaS ayes. cyesiets niseeh elas aes ocaisinccarsicarne 45 | Cerorhinca..............22.02002eeeeeee :

Cara Cara yeaa tl Inte i ea eke Od 90 | Cerro Borrego............2--0.000202ee- Bee dace as
Quadalupe.............02.02 2 cece eee eee ee eee 95 Colotad 0 sijocos suarabineweesack yess sae

Cardinal se 64 ic sonar qanoaen tesa celceane aan Soars 112 de la Encantada. “

CAPO sais 5 oes Sasa yenvececeaneducecceneneses 32: de la Providencia.

Cardinalis igneus (see also Richmondena).......... 149 de Zacatecas

Carmen Island.............-0.--202cecceeceeeecaee 92 Guajademi..............0. 20.0. e eee eee eee

Oar pod §CUB <i6-2.\accis asexd Rees eek shania nines ea 116 INO stein evacuees be te Radwes sek eee ees
CASEI risers aot ct eer ec Sacha ae hee 132 PegrepOse sn <.c,anaavad poedhecieise bina eee ed cumecines
INCRE CRON. 12: 0i6 aghoreustsyz teaver eae ieksioke terete ue 160 Perdido.25 452. snaheeaaichiemancceeeernes cea
mexicanus clementis..................0--e--- 130 PHC cetcauian yeah anniauencsnameer tes oreo

ATOT GANG 5 ofa) yee oo apf hares ct op ce ae te ate 126 San Juan de Dios
ruberrimus.............0022200-202600- 120, 126 Ugarte (named)

OOrrO tae apleres oisraaGt ti eataeaanaty an eN eh tiny au hed eo 140 MeOPd6 2552522 oasis svasencictnasveteransraihixiale naw eters

Casi erod is s:sscicis ssarcacmeecaieweecseewwiticnex cents 116 | Ceryle. (See Streptoceryle, Ghlomearile) Dita 115

Cassa ie acca peu anche crsiene meee ene 2incees 107, 108 Chaemepelia, esta sta ase 22d Ae hpaeeievintlal ills si v's neers 115
OTICUALaA Adee ease eo eee etree eae ae 123 passerina pallescens................. 34, 120, 128, 124
CONANIS a siaeseceesssaceae leah tic ccec ewes 123 | Chaeronycteris...............0.0.202ccecceeeeee ees’ «17
goldmani..............----2-2220) Lees ee ee eee 123 MOXICANG-..ceeeseeeedanees 117, 121, 124, 128, 160
occidentalis. |. ..2..cecccedecieseccsndiacend  123)) Chactura.ic.cs.vccsccuswenaecersccenee see tee enn 115
PUTPUBIS j.cci domes s wis cade ce mmancan ener 125,164 | Chamaea........... 0.00.2. e eee eee eee eee eee 116

Castille jac acacccecseasen ee adesa see ncecenaaceanesedn 110 fasciata henshawi...............-.-.-..--.--- 130

Castillo, Real del..............2-20.0- By iioeitiatiaee 75 passerina var. pallescens... 149

CastOD se cincc Bicetecodsd Aswisa cms nenhowaenouesen 117 Gigi cee gees anes ae A ceasarueetiaaam Bc tes
canadensis frondator 427 | Channel, Hardy.._............2.2.22.2.

CAG Serene eRe oc eae 95 | Charadrius (see also Pluvialis) ee
ring-tailed............2.......20200000- 93. | Oharitonetta. . oo. cccccratonenvanacas ais oon oa hen
Wild. gacwowcasouee ee ekie eee cdads 110 | Chartered yen of Lower California. . .

Catarina (boat) 40 Chaulelasmus...............0.20.000$ cee cece cnee

Cate Vail s-isc cic tsoleetes ice oe et nis w cketaeeteceermvatinn 25 | Chelonia agassizii.................22.20--0.200-0--
IATTOVO:GiteracG ss de bisiemiett atstoe inna elle 16) | (Chen. :s.2:o.ssceereits ieseinaiteautc

Cathartes:gce-scntcuce wali ote ed Sane aaa beg saat Ti) Cherries: co ccacc. wu auisluasseaeaneeace tes

Catherpes iacccuvaeueitess se dexeceeeeeaeersse ns T16: | (CHICK Peaa ero. tier astigmatic

Catoptrophorus 116 | Chilomeniscus fasciatus.................... ieveiais

Cattle. csice.ceesetkees soe ses Seeman 140 stramineus............222...2222--02-2.6-- 114, 115
PANO. co-cinndeeaduiededcees deca semennsed 76 | Chilopsis linearis.....................-..2002.225 118, 126

Ceanothus......-..- 00-202 e cece eee eden eee e eee deee 104 | Chiococes. <. 2.222201. sceeeseeececceceenes ene 108
divaricatus 131 TAU ON ck ecko nnn od eee ew eeoma Kaa «um 123
goldmani... SUnecastitauminchateGadecas JIM) Chipmiunke sooo 2s cic.0.cmunnnanwascaacae casks 32
Palm 6rd sais Soc Gers aaa danararo ews 129 | Chlidonias........... 2.22... c cece eee eee eee 116
submontanus....-..2...022.0 eee eee eee eee 129 | Chloroceryle...........2... 022.2 222222 0000022 e eee 115

(Cedarcdctaiesds manccuetdedre emdead axed alent es 87 | Chloroenas...............0..0000 0.02 e eee eee 15
IN COUKG ypceetc nas Gawadiequeiueneaxcesncse-ciwdext 103, 135 fasciata vioscae..... 22.22.2222. eee eee eee 121, 130

Cedros Island....... 11, 32; 86, 87 (dese.); 141 (work at) | Cholla cactus......................2- fatctetet aed oa ae, 85

Celosia floribunda............--...222.cceeeeeeeee 123 | Chondestes...............02....222 00200022008 .... 116

CEl8O: Vidal a oisieis deal opin hee abak aor aeedancemaaes 18 | Chordeiles..... 2.0... .....0 2. cee cee eee 115

CentiTUs hoc sisescad sede cw ocadselidsancaaenean 92,115 | Chorizanthe inequalis...............2....2......-- 162
ULOPY Cissy 2c Sc Ada cle cena See 112 | Chroicocephalus..................22--..-..220220- 116

brewsteri.................. . 120,126,167 | Chrysoma palmeri....................22.22222000- 126
EOD YBa lavaiiucdbicey ener wei o 3 IDG ONUD sch ceacart ett a let a dabea dete ee aettada chsh te 134

Ceralvo Island..........--....-..2-+--- 48, 8, 91 (desc.) |.Chuckwalla....... 2.2.22... 220200200 cece eee eee 84

Cerchneisies exe seddddssseneiele vues eax tears 115) Ginette cs terecwcerieale see unten seis ene Ace mcietmets os 116
sparveria peninsularis...................//.. 120,126 mexicanus unicolor................2.-....+--- 131

 
INDEX. Ei.

- Page. Page.

QOireuBievesseciac ves poaeteewe ss - gossnceee st Passe see 115 | Consag Rock...........0.2..0020c cece eee cee eee 18
Oiriors. jee esate a3 ¥ 24, 25, 29, 30, 31, 106, 119, 125 | Contopus richardsoni peninsulae.............----- 156
Cirlelaes seins sed cess Saneeaeinies Cemmnircsems 136 | Coolidge, Dane, work by...........-.-----+-+--- 45
OMellis.2ccceccss gue sesend tone edaeen tee ee ace BALL I | [COO erscceattelans Seles eyelets a A are aceite a a aaah 33
Atricap Ula: (cs wees sane eel sega T2127 | Copallienwis 25 2 x vec ey c ad seeaiceaee: e's 19, 34, 107
beecheyi beecheyi............-.-2.-..--0----- 180") ‘Coppers. e222 ditdiauicnt ci4ea eteeaceede seis pads 8
BBR ge eee a eerie asl satel bleed 130, 132 THAT ea vaseline 1 cae oa ocheuaan cee eneeeees 87
tereticaudus...........-.... 2.0 e eee eee eee 127 | Cordia watsoni................0.2.020.0000 cece eee 126
Clangulaiccie. sss coniscetese ewes cone posuere oe 116 | Cormorant...........-..2.0..0..2200005 23, 86, 88, 89, 112
CHMAte: s/s sissies chek oecis lawbad antes eared 95 Farallotis.sccsscees execs va ssed deseo tes san 90
Cnemidophorus bartolomas......-.....-... 114,126,171 | Corn... 2.2.2.2. 22.2022 e eee eee 23, 24, 33, 37, 38, 189
Geller pés.,.g 205 sc. concent ancees oedan 114, 115, 171 Bey ptian. 2ssece 2c 3< 8 nkedekadienetaeseas eee 140

GIS PATS iic.2 on eiegs Deane ORGe ae aoe noes 171 | Coronados Islands. ............--..----.-- 13, 85 (desc.)
estebanensis............-...-2--0.0.0-2ee0 eee DAE WV COrter- Sean econ coutss cspelstndateebuhn’ ccigtse tacey 48
labialis. certains. aa hale wed serls 114130: |, Corthylloscceecuade ss fv teeieetashn eesti sy oneee 116
Martyris.. . <i ssetedas se eweeccteds basen ends LTE 126 ||. Corvild 3:3,cs-ciaiece pce ead ok eecaswene ats banca ee 115
MAKIN US 255050 ape ss scents os 0d Severe EU4. U1 || (COtMOM 5.2 ia Sach tei doar 04 ad Wediow weed aut naa aes 138, 140
multiscutatus................2--2- 20-2 ee eee 114 domesticated.............0...0.2002ee eee eeee 109
punctilinealis. ...........20..2.2220222-022-- 171 Tongs ber cts fence spacatepateians slat Act dena 137
FUDLGUS 2 osc itowink oc certian dciahts oe naeusine 114, 115 SOC0 2 vyanenel eure: bee wagseesiccee ty test acees 137
StE(NOPON 2. ja veces ooteeeat ys < sees dnere 159 Wildisies eamoedied oNeie deed ds Atekaltia es iascanks 109
tessellatus stejnegeri...........-.--..----2.--- 114 | Cottontail rabbit. ..........2..2.--.--2------- 87, 110, 133

-: vandenburghi.............--.----.---- 114, 115, 171 | Cottonwood............2....2.2..-2002-200- 104, 105, 107
Cneoridium dumosum. ................-2------2--- 129 MacDougal’s. .....2.-....-:2.--+---+--- Aneee 106
Coastal Telands: ois... ss inset nee saipinweinn es 84 | Cotyledoniisicscccccaxe os. si aeeieekedew esi Fexenes 108
COCCY GUS ais 'o' 2 casiainces Ss ve easaeeke ass euneeeeues 55 16 | Covilles)c ce .geccenecex bess dusiscaccee 17, 28, 35, 106, 125
Cochimi Indians...............2.-.---022222-000- 7 glutinosa........2...222--2-2-2--2------ 19, 118, 126
Cochran, H. W., work by..........--.------------ V2: |; COV Ot a. serie oa.g pa eaureyeeteealt es 25, 90, 110, 132
COCONUT aio5. dsb aoeee ed Seow s ae 48, 107, 138, 140 | Crab, fiddler. ..........2..0.2.22.-2-0022-02220++ 19
Cocopah Mountains. .......-.--..-.--.----+------ 13274: || Crayosheees3ocec esis 2s Jotamiuc cease ieee 184, 135
Cocos nucifera (introduced)...............2--2----- N23 :|) CreCisCUs 32\-jele cat's one's s aeieeey ss Peanipauients 116
Cotheenge aloes we ine eee ves eet Eo es ice 140 | Creek, San Antonio ....... anita oun eedasowaneees 22
Colaptess ccc se cjecccoses Poaceae cect onesies cee 115 | Creosote bush..................--.---- 25, 30, 106, 107, 109
cafer collaris.............-2.0222022-eeeeeeeee 181 | Crescentia cujete...........-.-...202--0-2-e22ee 110, 123
rufipileus. ...........2...0222-0e220 ee een, 400 | Crevallycicnsscinceats enced erect iceecanncosanns 134
chrysoides brunnescens....-..---..-----4-- 1265169) CROPS s.cesnise eee ke iastugeatccl 23, 39, 45, 71, 137, 189
Chryscldesixésa: 25s oaednsee ess enees 112, 120, 126 BOUACALG Senses ose Serene os eae 140

MCAINSL «io :sssee ds oeizakececos eke eeeeeee sss 126 alfaltaicc: cue. cbc espa ae Seales 137, 138, 140
Coleonyx variegatus. ......-.-.-------2+-2-e+0- 114, 126 almonds.......-..--.--2.-2---0--eeeeeee 24, 139, 140
Coleosanthus atractyloides. .......-.--.------+--- 129 AD PlOAs oc ceecpnewin conojantadsdasew waned 71, 189, 140
Colonia Lerdo......- 2-22-22. 2-22 eee e eee eee eee ee 13 ADHICOtA acca eGo son aneeeaor ee eee ae an 139, 140
Colony, attempted. ......-..--------22-22 eee eee 72 AVOCAKO! si eerie res cc aciidled tekios vale secede 43
. San Quintin Bay...........-.2----2----- 0086. 23 bananas. 22s sc5 2. ses eenes dagen eee 35, 37, 48, 138, 140
Colorado, Arroyo...........22-.20.eeececeeeees | Harley casa tsi wecedoteana oe 24, 137, 138, 139, 140
. Delta, Agriculture... - -. eoseCasab hse emees 137 PeaNs secs ee Saas Geiteeae ss saad 24, 33, 189, 140

ig DOSORE, eae 2 2tsiscceisinee ed eesie ete 74 D668 a2 ce Av atett at tammena antes oeareawenees 140
‘Faunal District... .........2.-222e0eeeeeeeee 118 Cab bag 68 acces sia Soceramccce rte: oc cteatacan 140
Riverie ss facdien Soe c sea cee cee Se Saree 10, 79 (desc.) COTO: dock duties Osi vitdawe cease enseeeee 140
Coluber aurigulus...........-.-----2--0e-2-00+- 114, 115 Cherries 2iewies snes aeaewnedad esd esasinnecnde 139, 140
constrictor flaviventris........-..-.-------- 114, 115 chick peas. ccc ivvesceceeka sc css ennes sees 139
lateralis... ........-------eceeeeeeeeeeeeee 114, 115 COCONULS... 2-2-2. e eee eee e eee cece ee eee 43, 138, 140
Colubrina glabra.........--.------- Bh ide ae hoe ee 123 COCO. sccreccaddt sa eeatadteses beat ccas ty 140
Columba (see also Chloroenas, 115). COM...------------ eee eee eee eee 23, 24, 37, 38, 139
fasciata Vioscae.....-..-.--2----- eee ee eee ee ee 153 GOMOD. secs sss eed oense eee eee cage viecias 138, 140
Oolumbia Range......-..-------00eeee eee eee eeee 29 long-fibers. ....vessces ewes ees seancdeenes 137
Col¥m bus: occu os deseediins whetene sean diieatoeeas 116 NOOK 2s Ge ymcie a a ceive Das ganduinceue 137
Comond We). ase 2 feidewles ve steric pes eceeeeneet 10 Gabe8 oo s saxisis nicis p diam ativiselemniate.s Retinataties 35,
Arroyo © Srey Side laa ite ates bchanstcncbayedfeVen dante anayavavenacetet 39, 62 36, 37, 38, 40, 43, 77, 138, 139 (introduction), 140

4 population upecoaee + theses ses e Bua eiseve- es 10 Egy ptian COPD 5,225 fo. shsts aie Sakdevevelcusleia ave evararmciee bie 140°
Concepcion Bay... -----------------ne eee ee ee eee 73 BEB hiss see ee a's oe -- 24, 33, 35, 38, 71, 138, 139, 140
Condor, California........--------------+-+-+-+--- 22 black esti dcue sen cece sed sass camealeane 140
Conocarpus erecta. .----------- 2-2 s eee eee eee ee 123 BTapesS..........-.- 2s ence ee 23, 33, 37, 71, 138, 139, 140

 
 

 

 

178 INDEX.

Crops—Continued. Page. Page.
[OIO DSCC nici tsees ne oe wiecieneeie elcn eit ecee 139i) DatilWO ss 32 secice oven aneeweencen secsteslts . 29
IeMODSiwaeincesseesaedans 4, 33; 95, 43, 71,188, 199; 140 || Déer 2. co ccdestwe ele tas sv a deasdenesectemes 46,111
Lim Gs ies pedo hbancisclichaess at bacon Raed 38, 140 mule...............2.2. 21, 34, 87, 90, 92, 110, 132, 133

BWEEbs cd casincce nema dinomidoansamuaaaaoemed 140 | Dendrocygna.............0..0..22 2 eee eee eee eens 116
MANGOES... --. 2. e cece ee eeeeees 44 | Dendroica..........0.00 0.00 cece cece eee cere ee ences 116
MOLONGS . sscinicloatedemivearamentecseaiisienecers 23, 137, 140 aestiva sonorana...........2...22-2 ee eee 126
muskmelons............. 2020.2 0ccee eee eeeee 140 BUM UDONE nc. ccc cence eee aes gecce seni elecen 112
OMNES xtincccawaihatwustgnsar nae oe 33, 74, 138, 139, 140 bryanti castaneiceps..............220---0-0+- ‘152
oranges..............-. 33, 35, 37, 38, 43, 71, 138, 139, 140 erithachorides castaneiceps.......-.---- 120, 123, 124
PCACHES se serch sei feeseiatartereerasdets ise 23, 24, 71, 139, 140 vieilloti bryanti................-.2-2---2-0--+-- 151
PeOars:+ 2: cease de deee aeeted s sess cence 71,139,140 | Deppe, F., work by...............0. 0200-02 e eee eee 141
DOAS ete caetees eainriens nin Suicdeleeseeen ane gene 140 | Dermochelys schlegelii.............-..2-.-----+-- 114
POPPE Wags s5e cies sn asieese ned: Seen ME anes 140 | Desert, Colorado, desc.............0.0e0e2eeee eee 74

COUN 3s 25 os Sos dasa deaseeatd Gaon aire eens 139 Sand Ulin: dese senso oewenigavaveeseceetes 77
pineapples..........-............ 2002 cence ee 138 Vizeai need (ic oss gs eae 30, 32, 34, 50, 71
pinmyon nuts............. 202.0 eee eee eee e eee TIO Wa O Wie eicyasvacssosc al wausid aes etescisie oie decor eee ee ee 100
pomegranates............-.- 33, 35, 38, 43, 188, 139,140 | Diablo Canyon................-2--+--+- 17, 21 (desc.), 56
POTATOOS. cisicn a nsccrsivintreenty eaveicaisicrnclonitelhslansye 139,140 | Dichromanassa...............2.220202 222 e eee eee 116
PUIGUS APAVE noon ccieieccececnars cee seaeen 140 | Diguet, Leon, work by...............--.--------+- 145
GUUINCEB <a dciccersiassiiannaeicte lore nctnesineleoase mr sasae 139, 140 | Diomedea (see also Phoebastria)..............-..--. 116
PRUSUNS Sas sas yceiciatelocis cindeuanneea SEER 37, 139 TMMUtADUIS. .csacenasawsecemonsecce scat dames 160
PLCC bicvoss tea ie re ececeqercesra tchataf kero eatessemureedeteeneoetae 187 | Diplacus arachnoideus..............-2.--+---+--- 129
BQ UASIOS tet gla fatageictasanteuicteiwieduidta eh slapensycyy scene 140 PUNICCUS se cosas oewdenrcoecacsiseea asa mcanuans 129
sugar-cane........ 33, 35, 36, 37, 38, 44, 45, 137, 139, 140 | Dipodomys..............222.-202 22 eee ee eee 28, 110, 117
LOMA LOCKS asensere same iies alate eon tees eters 140 EBERET soroychche catteestshuc atone nate ea edanoneaeetys 18, 127
Wats seems emisev wien a sened veeeoeeaes 139, 140 insularis....... weeeee-- 92,127, 164
watermelons ivccceevicwscsteweans ae eevee emenmes 140 MATPOTRC x. csceue ds rewenavouessceeceser ess 127, 164
Wheat. . 2. sesssnesnses Pee hea wea 24, 138, 189, 140 merriami arenivagus...........2.-.-2.-----+- 163

Crossbill, Mexican..........-2..2...2.-20--0eeeeee 161 MelaNUIS scenes dyse sce Secs eenee see 121, 127

Crotalus atroxs 022.0052. 2cacssedaraceeenenc oxy 114, 115 PANVUS ee Sess ea canes se ee enias vas 130
cerastes 114, 126 simiolus.......-.-......20002.-2-0ee ee ee ee 127
ODY Osos cgisirece tenis Bopinseataueee ae ties 114, 115 platycephalus..............--...-.2.22-.-- 127, 164
OXS Uo ccasj Sulcus ode Soe aeesels asians 114 | Dipsosaurus dorsalis dorsalis...........-....---. 114, 126
WNCRSCTBIS 25). cisco ce esmreedoces Gene atind dd 114, 115 dorsalis lucasensis...................-2-0.-- 114, 115
mitchellit. «..22,0c08c082 deen SS Doan io sisaseenoe 115, 159 | Dodonaea viscosa................-2.2-220ee eae eeee 123
oreganus......-.-----2--+---+-- eter c lon islet 114 | Dos Picachos Peak................22-222-2- 0-22 eee 58
pytrhus..........-- dprasaeabaaaan anaes ee v---- 159 | Dove, mourning................202.22-2-22-00-0-- 134

Crotaphytus collaris baileyi...................-- 114,126) | (Drought .ccG. ce are acted oe Beweeeienie ects 97
COPCU se ccwccvccesccteoreesdeciamncmese eee 114,115 | Dryobates..............220 0220-0 e eee eee eee eee 92,115
WISLIZODN -s.sniccicaw an inneaeacua de Sxcrn iene mes 114 THIttallil .-. ss. -cltusion seseaeseeence seed 130

Croton magdalenae.............-----.0---2-- eens 123 BCAAPIAS caste nol ipasancasncemuunmene/ einer 112, 166

CTOtO Phage esi deecnsoneesee ceed os stages cae 115 cactophilus...............2.2....2.2. Say 126
BUI CIOS EIS 32 soe ersietes' te secteisiesinisee 112, 120, 123, 124 CLCMICUS +2 20h ieee Sic oaoe 126, 166

CienosaUt a cox:2esocescecceseorescdere theese 114 TUcasanUBreces sent cece. eee cee ee tn 120, 126
CONSPICUO8S s<ss25 02 2es ccs maeewnsseeaeee ee eee VG: Da Cede cele esiecie tie .domestacin outed i cavcseaw 11
hemilopha.........0.522 cess cece acnee se 84,114, 115,123 | Dudleya anthonyi........-...2.......220..-2.... 86, 108
TTISUL AIG oes asd siaarceisioverate nse-epe/esieereee eee 114,115,171 | Dumetella.......... Beata tn byes sata thee Recmtatabeio nacre 116

Cupressus.............--.-2----55 Spieeccinesneeaee 104 | Duranta plumieri.............2....0.22.222.2.--.- 123
guadalupensis...........--..---..------ 95, 105,131 | Hagle, bald...........2...2-.2.222.222.. Eiht twecteas 86

Currant, black............2-.0-0----+--e--e--e--- 110 | Echinocactus..............200002.022..2005 106, 145, 161

Cyanocephalus............2-0-222222eeeeeee eee eee 116 digueti............2.22222.22...... 84,91, 98, 128, 161
cyanocephalus...........-.....4.0.eeee-cee eee 130 falconeri........- S22 DR Gk ptngahe alan wien wiped 123

OV anOClttais seaveissiies vue dees pee ieee eee cme tie 115 OP CUNT axis Susi eeee soem ety dertela a 24 decyeybe 152
stelleri frontalis.........-.--.2-2..-2--2-ee eee 131 TECtOBPINUS -< vic seseeessescadew kere eceece cme 123

CY Press. a1 -scteeteseseesimeenseeade 94, 104, 105, 108 | Echinocereus brandegeei................. Meets 126

Cypseloides (see Nephoecetes, 115). Echinocystis brandegei................-.22222222. 156

TD ft Bsa ssc arcesccp cn. fasa’s cacqeact adenine adln- casera iaddedies zea aes 116) | Weretisis cesses cndiceismaailemaninaoawencemen aeons 116

Dagoctt,, Dik sca. ineveraciesss Wooneaedeesanea 26 candidissima brewsteri....................... 166

DY Bi ed ecsesans ed ster sees bereeetctets burst omen Meera VRS 107 thula brewsteri........... hela dee kV Reeth Ar 120, 124

Dasypterieecnwy swrocns senavannedeag secnernieliewnt 117 | Eisen, Gustav, work by................. acheseay ra teeeats 145
ega xanthinus..........-.-.-----++ 117, 121, 124, 160 on glacial action...........--...2......0 0c eee _ 64

Dt@Ss eee ccbeicceaen cece 35, 36, 37, 38, 40, 43, 77, 90,107, | El Alamo.............2-2.ee0-e eee ees 8, 10 (population)

138, 139 (introduced), 140 | El Batequi Ranch...............002.- 202 eee eee eee 39
INDEX, 179

El Diablo Canyon
El Mechudo Peak

El Rosario, population
El Sauz Ranch

ee ee ee a ee ee ee ee

Ellsworth, Lincoln, work by.
difficilis cineritius

10 (population); 11, 13, 14, 18

39, 96, 100 (desc.)

eee

95, 105, 157, 158

 

. Page.
Brythiinas souicneeddak haa sau dubs: ose eaaws -.. 48, 108
PUNDUBLS 255.68 cocteeaieaeuisa skns¥e esata cue 123
Esenbeckia flava...........-..022-2---2cecececees 123, 156
Esperanza Canyon................02.2-02ee0e+ 17,19, 21
Espiritu Santo Island................... 47,48, 91 (desc.)
Estero de San Ramon...............2-.--2eeeeees 77
Eupatorium peninsulare..................2..222-- 123
Huphapus..<.6:5 oss canevedsvssoed noc. ss eevee 116
cyanocephalus.............--2--2.2.0-022-00ee 131
Euphorbia. .......... 2.02. 2.02. 0202 eee eee eee eee 107
CalifOIrmiC ayo eos ab eeec ea ceee eee cetaceans 123
tomentulosa.............2.. 002 eee ee eee eee 126
PRATT oleh Jelena does plete A ye aaa ena 123
BUDO a sevsunwcevmae ss eee eoatye doen vo oa cence 116
Wiitamilags ivi. 2 Soccccac as ¢ ance encodeses ee wehetes 117
merriami meridionalis.................-.. 32, 127, 165
ODSCUTUB aan <2. osidgceueeee es ces cea 132

Falco (see Hierofalco, Rhynchodon, Tinnunculus,

Cerchneis, 115). :

Fan, allivial ascscoes cx vss seignewaieeex ees cscicckis 78
Faunal District............. opeeeeamneiare seodas 117
Cape 9 eek tintutea ooh sn Soa aaa desea scne 119
Colorado Desert.............20.0.00202ee ee eee 118
San Diegan..............2.2.2-02020000 202 e ee 118
San Pedro Martir.............-..2.2.22-2..-. 117
Vizcaino Desert... ... 118
Helis suse ssees sotteteacewe wos ak ka ta siue aed oe 117
BZUCCUS semis tis fed oosechoed 6d deptiaceedieeaaneaees 127
aztecus browni........-.....22.2222-22eeeee-e 127
IMpPrOCer as. se ncadesses.c cise oeeciie ss esas 167
Fever, malarial..............2.2-022222002 ce eens 38
PID OP sae Sasa aetna haat tnd oA NAMEN eae er te 117
zibethicus pallidus..................2.22..22- 127
BICUS ic ccsunmueinser eelgee oe ches dabiecadeknee Ganon 108
PalMer weeecsccc cece ss koko ieneeecazeess 109, 123
Field Museum of Natural History, work by........ 145
Figs ssc2acsen sess bauaneosocenin 24, 33, 55, 71, 188, 189, 140
Dae hsieceucta sted cete hi evel gagtiestnnd eles 24, 140
Wild heehee etd coke, 33, 36, 109
Finch, house, Cape.........2.2222.02.0.0202000ee 32
MeGregors ic secant 505s s ante watsmessiieas 88
Wan LUCAs ec os.cseesakeneeeeedeece ees 91
WiBidasciienssee onebroaadeancees hahaa gion a: 103, 135
PBI 2 hoe ecto PREC RY wel trea 111
bles 5a5,.'seisjatorecn a Segoe ee seen mr ecaeeke 134
Five. Peaksewcescs: csasouceme diss cae acigciins amma 66
Flicker, gilded...............--2---22- ceeereees 90, 112
Guadalupe......-.------.---22.-.22e eee eee eee 95
HIGOG scouts aedernavaniin mashes sp dahaaiceysieet 96
Flora, derivation........... 02.20.2200. .0c2ceeeee es 103
Guadalupe......-.---- 2.22... 02. e eee cece 94
Flores, Hale & Co..-.... 2.2.0.2 022. e cece eee eee 40
WIGrid ass scteeien saws cette didi asiondenwdddsudasoawien 116
Flycatcher, ash-throated. .............2200020 20005 112
Lower California.......2.2.00000.00000..0-00. 90
HOG st deed ecwiie date Sada ateaise eos. Seenwbaaeckh 100
Forchammeria watsoni..............20..2...202.. 92, 123
Forests; pine seins. Su. ca eaniecawenipes cadena Seees 135
Fouquieria............ 17, 28, 30, 89, 106,107, 109, 119, 125.
' peninsularis........... 29, 30, 31, 40, 42, 45, 90, 91, 126

splendens.................... +--+ 19, 24, 25, 118, 126
 

180 INDEX.
Page. Page.
ONG pst tte cpt tea abate ate orate i ance 87 | Grazing, cattle.......0....0.000.20.20e eee eee eee eee 149
COBY oe cep raeihiine aralatadeie aes chee coe afers 19, 87, 110 SHEED 232 vasa csanadtindcn acorn mee eneaenee 140
PTAs es coisas scare areeeed eieleceideecs 33, 35, 87,110 | Grease wood...............0..--02 eee eee eee eeee 107
Hombatl:prasas.0c6<jocexe gbbivocae ou Csazn aneacind 95 | Greene, Edward Lee, work by..........-.-..------ 142
Franceschi, F., work by........---2.2..2..-02.20+- 144 | Grinnell, J., work by.............2..0-0220202--2005 85
Francisquito..........2220.2. 0.2220 e eee eee 9 | Grossularia quercetorum...............-.---.----- 129
Fraxinus attenuata ...............-22.220----ee- 104, 129 | Grouper, yellow-tailed..................-22------ 134
Frazar, M. Abbott, work by...............-.-205 AT NAD. | IGTUS cwsecls, give asic este ptecadda ek Seapateh et alesec nz alo 116
HYG Rt jose oie iste taclene youre MAvew os ees cei 116 | Guadalupe Arroyo............2.2-02.2-2002ee eee 42,75
Fremontodendron californicum..............-.+++ 129 NOs 252 8255 0 Geet omer mMadee tae aeee 94
WTOSt sweep seers cha had ns dewusaiesnnae be 94, 96, 102 Helen de 22-25 2.02.2 inenmacoaenemenee ek 84, 93 (desc.)
Pie sige ss eiexisqgesiaccuwnci dg aebieieiietinos 622 ba 33, 116 Lif6 ZONES jc niacin oattaeaguaanraencet ee 95
Hullmarus? 2 jj 2d tenatwiedlece Geswdminw ane 3 alse 116 MISSI OMS 252 ioaecsce stars ceataralorociediere Ace RES 7, 15, 73, 75
Fur seal (see Seal). VallOY. oisatccsnkinteniler es hottest Aue raat 9
Gabb, William M., work by..-......---.---.-006 141 | Guaicura Indians......................2---2-+-+- 7
Gallinago.........-.---...--2--22-- jedia tia madiate eieieie’es 116 | Guajademi Ranch.................-...---00-4002+ 36
Gallina soos. aca cceseacisceaaot de aeccysuutoneciers s 116 | Guano Island...............0..2.00-002 eee eee eee ee 90
Game IAW ss. co coctapipsscosieeeesesdbenmameces « ASS] Guaraiciey- cee Veo sve sen dances eeeeearee ey ys 116
mammals, land..................2-02-0---0-- 132) WGUATTY O8.c'5-¢54's SR ec ecsvesee na cee ebrae dy se 105
BOD - ici asteige tae sect pee Sue e wen lced VBS MW GQUiacay ee cis cee pc ba os 4 diese diaicvassraie ence eee SE 85 116
Garambullo cactus. ........--.2..0- 2022s eee p eee QB NS Gea Me oreo 20d dooce nasreigseng se fea le'ely a ybiaecesatereeaigyersaa dees 86, 112
Garry a cali cifolia zoas.cacueecesese candeaeeesexeckes 129 Heermann’ajnc2c.42 ys sy nas eee 84, 88, 112
veatchii palmeri..............2-2.-222--20-05- 129 Sabine:.s:2ecies2ss 2024 sve veneveeeeseeeek sess 13
GaVitiecccess 645 te vexemeneesewsee ee aceite ‘116 Wester os.ccies sexe eceese teen nee dee 86, 88, 112
Gay Brothers..............-.------.---- actin gence 20 | Gymmogypsss. ss... 2.s-e2-eeneseuverscaseccesas 115
Gaylord, Horace A., work by.............--....4- 143 californmianus......-.- 2+ +2 -.0s2.skeenc ences eee 130
J. M., work by.........22220202 202-2 eee eee eee 143 | Habenaria lucaecapensis............-.----.----+--- 158
GOCK Oia escie secnGrck 22d Sea abaiaaansaee Wisesine 116; 123! | GC DOrh yl. iosiee bs one sete wie eomnteasoig aes as 105
Gelochelidon.......... 2.2... 2...00.-00e cece eee 116} | Haemato pus: cci..c0se 20s os Secs acenddaeceacede eae 116
GeOcOCt yxy .c50 a5n2205 Gsacmcawagyonaeneaee 112, 115 WACAM 2s sss os Sh cin eee gs eee 153
GeOlORY sac vc aacwiass scare onemnemernarensatcnatucsers , 52 | Haematoxylon............-..2-2-2.-0-2-0---2-2-2- 108
Geothl ¥ Pls jasccccs dew cepeetinnynucarcncecamenmames 116 Dorel s.3.ccceci ge sie 2 Rawearnaaedeass 122, 123
DOMAIN GT Sc 5505 oro cen nase oseer eae 151, 154 | Hains, Charles D., work by............--.-.--.--.- 144
belditist.24s5ncccase eases 120, 128, 124 | Haliaeetus..............2....2222022- 22 eee eee 115
POlMMANT os eco eds mapas aeneseewene 4.) Hell BytOlsnc tesco accaeaseeeeemoaeeeen tse neg 31
trichas occidentalis...............-222-------- 126 | Halocyptetias. sc 226 2 hse pe eceaysewecask doe 116
Gerrhonotus multi-carinatus...........---..---- 114,115 | Hamalu, Llano. .........--.---2---22 2-22 e eee 70
scincicauda webbii..........-.-22-22-+2++--+ 114, 181 | Hanson Lagoon.....-...-.-.--..2------ 15, 55, 80 (desc.)
Gifford, E. W., work by.............--.---2----45 144 RANCH os dueciais eres orchataes tebe SysgeaeeRe tacts ans ince 15
Glacial action, Eisen on...........--..--.----+---- 54 | Hardy Channel................-22..222222222-20-. 79
Glavicidiimi:eeees oes weaker tacins astighine Helge wate 115 RAV OP ooo sieks os ar oeeieell Rea SAS Seles ce 2s 13, 69
gnoma hoskinsi........-.------------ 36, 121, 130, 154 | Harporhynchus cinereus...............-......... 149,156
Gbate cian Oo ose oon de ene mee eels 94, 95, 139 cinereus mearnsi..................2-2222--0.20- 159
GOlEs seoidawiwegsind sobee Asana seebebee es cemaee 8, 87 lecontei arenicola..........-...-.-..22-2.2-0--- 160
Goldbaum, David, work by......-...--.-------+--- 12 | Hauya arborea..........-.-.-2-----..2220 eee eee 123
Goldman, E. A., work by...........------- 12, 18,19, 145 | Hawk, duck.............2..-- 2.202222 e ee eee eee ee 161
Luther J., work by......-2--.202--2220eeeeeeee 145 Pidhis 24, vclneeess sai, ev aeeenadieletnnes 86, 90, 91, 92
‘Goldman Peak (named)....-....--..--+--+-+-+-+- oe 2 ATV IS’ BS 22 32225 Syetal peer ownnes eee eth tes 29
Gongylocarpus fruticulosus.........-.----------- 123, 156 SPANO Wesais oes i oceedacdimbiuanncwmmees > o 90, 91, 92
Goose, SW ne scictotre a'e ic clomesesenehe yee oad eiseaiaevele 111 | Healthfulness...............2..020.0.00 0.0.02. 2 cease ll
Gopher, pocket............--------++++ eee eee 14, 87,90 | Hedymeles.......-...-.---2-.---2.-202.20222-220- 116
SAK. ec eheae secs ee Ce eeaeeen ean eae 86 | Helodytes.....-...-..----.--- 02... 22 eee eee eee 112, 116
Gossypium davidsoni..............-2--2-222066+ 109, 123 affinis.......... 2.0... -- 2. esses eee ee eee ee 158, 159
Papel. co. sce ice cit weet saaemeeleicnde oe: 123 brunneicapillus.........................22222. 159
herbaceum........-..-------- esse eee ee eee ee 109 affinis. ..-- 2-2-2... eee eee eee eens 120, 126
Gourd tree. .... 22-22 eee eee eee eee 110 bryanti Be er ee eee ee eee 158
Pa ee else aah EARLS A ae 25 COUCSI.... 0.226. eee eee eee eee 126
Grants, land secsccre os 2 cnevonncas Alt saa tess 9 | Heliomaster....-.........-.--....0- 222 eee eee .-. 149
Grapes.........- ese eiesllicteb rotted 23, 33, 71, 138, 189, 140 | Heliopaedica castaneocauda..................... 149
WhissIOU Gee Sha cee eens nee eteoads 37 | Heller, Edmund, work by.............2.2.....2.. 145
Grass, foxtail........ 2.220... eee eee e eee eee es 95 | Heney, W. J., work by. .........--....----...00e 12, 40
GreashopPet coca. secs aneenceansdas oe een tees 113 | Herald (boat) ......................2 02 eee. 86, 87
INDEX. 18]
Herodias (see Casmerodius, 116). Page. | Island—Continued. Page.
Heron, San Lucas blue.......-.-.-.-.---0-0--2 2 ee 92 Natividad oo.2< 2 cc:ccd.0 oo ccenane Suistis sé sideeles 88
Hesperonia californica ..........-...-.-+.--0--0-+ 125 San Benito ..........20.2. 202 eee eee ee eee eee 88
Hesperoyucca whipplei............-.-----.-----5- 129 San Geronimo (desc.).....-..-.-----+-+--2+05 23, 86
Heteractitis (see Heteroscelus, 116). San Ildefonso... 2. 2.22.2 ee eee eee eee eee 84
Heteromel 68 csc sicietstisas veins dat ddee wiley ss savanienes 46, 104 Sam Jose's. ceetues cet eew oe ca oeeren es 2 rege 92
arbutifolaa 26.22 secconace eed cee veto dadee 87, 129 San Martin, desc..............2..22222-22-eee 23, 86
Heteroscelus..-.....-....-- 222-20 -ee cece ee eee ees 116 San Pedro Martir...............0.2222.20000- 84
Hierotaled,s -2osscistend cs Saeeseeaee se senses --- 15 Ban ROGUE scc0 oceeee ete cdemeerecaatens once 89
FimantO pus. ici cc canine sieviass seaae vice canescens 116 Santa Margarita ....-..-..--..--.2--...eeeee- 90
Hinds, Richard Brinsley, work by... ......------ 141 Todos Santos. .2 42-4 secs eeesssereeee 14, 85 (desc.)
Hind 0 ses3s dues cee ieee vyueeceek yess daboeendax 3 116 | Isomeris arborea.........-.-..-------2---2-2-+-- 118, 125
Hoffmanseggia microphylla...............-...-.-- 125 | Isopod, black..........-------++++-ee eee ee eee eee 19
Holly, California........-.....-..-.-22--+-- - 87,104,105 | Ixobrychus........-...-.-----+2+-202 eee eee eee 116
Holodiscus dumosus.............-----20-ee scenes VSD TROP OUB es oe yacaSa ean dttkns arson bleed cei ota 116
Hondo Ranch sivisic cpa eevionie dso saaelaeseneces 36, 37 | Jack rabbit (see Rabbit)
Honey ssisicene soe cemicae yee ssc anaeteseeae see 139 | Jacquemontia abutiloides........-...-......-..-- 123
Howell, A. B., work by... ...--.--.-------.-005- 85: | JaguaPeccc as <a ceeekad ees ey esacene hes eames 110
Siilbreels cases iste aeaadieett (oemlonsee ses 107 | Jaraguay, Arroyo.......--.-2 222222 e eee eee eee 25, 76
Hunter, J. 8., work by..-.......-. hci toad re. S Saas 144, | Jatropha.; 2-2 sncccieeeecsie nd es ceeneeeces 89, 107, 108, 122
Hutchinson, Joseph K., work by. ...-...-.--..--- 12 CANESCENS jc eccidhmie soa man asinsienea eee 90, 92, 123
Ly dranassa.o2.2stconwk ti viccempen taian eee eeciescs 116 COLMMB ices nice a Gees oldatsniireee ter weiss 123
Hydrochelidon (see Chlidonias, 116). 5 Spathulataectcei2s (cs ca gander ese penis 91, 126
Hydroprogne.«.- a2. 02s sessnceads eee ses ea eae ds 16 | Jay, Xantus’s. .2escceee cs se sseesteseee veces cena ss 90
Hylocichla. « <ésc0c0..v ss caseteue eee ses tanec 116 | Jewish ::...s00sstcceseuney es sccexttioneeans se beeser 134
Hymenoclea monogyra winate erceretanncrie cemented 118,126 | Jojoba seed. ...-....-. 2-22-2022 cece eee eee eee eee 139
Hypsiglena ochrorhynchus...............----- 114, 115 | Jones, M. E., work by..........---2----2---0e2200- 142
Thervillea sonorae........-..----2---20---0-- 91, 122;123; | PUN COL serccinsenisecewwee sts hemetieseaea uk desiree 95, 116
TGS aie oo. cicecicicc ease 2 sie steisieidisi se aeeiee eos cepsicisna gases 94 Dalle sscecccnu coe eeeocen coves 121, 130, 151
PLAN Gos icssiies eis Siatejemiee egarte cies weeidiacis neawe 85, 108 hyemalis shufeldti.........-........--.--+--- 159
Tebertassuciee dans o's gunn's dleiawiern a's ese argldanwes 116 PMNSUVATIG 42. seuieet i he's vous ae cisetince es 54d See 95
Tetenisissnssccsceess os tenses ceceee ans soewawsdwes 116 oreganus townsendi...........--.-..-------4-- 131
Idria columnaris............-. 24, 25, 29, 106, 119, 125, 126 ToWnseldleseeeus ese se seee dee ce eecaenee 155
AQUANA: 2) cicracisicalzj= o sletretsdinied Se iaeicis vets ae aradis 84591) | JUNI Pel sicci2ssinicinescins Sogo seiacinae soos 6saa 94, 104, 108
Imperial Valley... ...-.....2.202 22 ee eee eee 10, 14, 75 Califomis .cjcnecdcc nat nab eet aiieauecaasnwes 75
Incense cedar... ......---- 222-2 eee eee eee eee 103, 135 | Juniperus californica...............2.--. 87, 104, 129, 149
Indians: COCTORIAN A ss iainicinie sede Sanip rece eeibiee s oslgaaised 149°
Cocbimies: #:cajs's's's'sieintisecounine viel tdgaensiond 7 | Justicia insolita............... di ara Bee vis ws aarp 123
Guailcurasids css ove esse sedeses ys: s ees opiee 7 PAM oo s0cssasecen see setae deueedcsvanices 123
Pebicweés «.<.2..ceg). esses dense sess ssmseeeees 7 | Kaeding, H. B., work by..........-.........2..- 143
Yaquis. -..- biacysod ama cesses cece neers eS 8 | Kangaroo rat (see Rat).
Indian population...........--.-.---------5---- 7, 8,11 | Karwinskia humboldtiana...-................- 122, 123
v WEN a Ses cases eo ciiirkeee nent tae teenie nee DB: 1h ROIp secc sn crareareereseveisteie pints sictetisie Siessia Woiplacteasies 136
Apomieaaureae. vce es ss lasaseedics ees sister caee 110 | King, E.8., work by.......2...-22222.222-22000- 144
, POs Capraeiees cscs ss cumesteeeeers -seagcesdevs 123 | King Richard MING ay oe yee eee cd wicked 26
Tresine alternifolia. ........--.+-+---+2+--+-eeee ee 123 | Kino, Cerro........--.---- 0-2-2 ence eee eee nee 70
Iridoprocne .. - - - nidiicae saecineceegi te ae hepatitis 116 AONE siuis niceties ayatanbhune iets Sues 8
Tron......----- Sreseseceenaes beeende EAmanancideci 8 | Koeberlinia spinosa... ......2..2.2.2-222222..- 118, 126
Tronwood ........---------------+-- ---- 17,106, 107,109 | La Aguja Peak............222202.222.002 20 eee ee 47, 63
Isla de Cedros.....--.----------4+-- idisiele sierwigits a 86 | La Bandera Peak...........--.-.2-0-22-222022205 60
Island, Angel de la Guarda...........-.--2.--200- 93 | La Chuparosa............0.0-0-02 22222 e eee eee ee 46
ASUNCION 2... + 2266s scacteebtaweee ess vec 89) La Crug rane. ccycccsiee.cccncdineiaye aus ntevans 41
Carmen: nciraediaie ss seceue teciweee esas 5 92 | La Cueva ranch. .....-2 222.222. 34
Cedros........-0----+e05 32, 86 (desc.), 87 (work at) | La Encantada basin.............2222.. 0000.22 57
Coral Voi sencin's ceecipne soe siecees 48, 73, 91 (desc.) DORK so steicidicjaes: Se sahelats aie Saeneeee «atx eexeeeke 56
Coastal 2. <i maeticoe ys Ss eaestoeeech seed sens 84 | La Esperanza Canyon. ...... cts Ah NORA AS eS 1, 56
COROTaM OS! « . aisicieieieieie ses oes onc aitere tee 18, 85 (desc.) | La Giganta... 2.2.2.2 2 222 eee eee ee eee 39
Espiritu Santo......--...-.------++-- 47, 91 (desc.) | La Grulla meadow (desc.).............-..2..20 21, 57, 80
Guadalupe. .......--.2.e.00- 066 Sere 84, 93 (desc.) | La Huerta........... 2.20222 15
Guano. .....-.--- tad hasecterc & Syeranelneyohaeerarive w0ats ceeveyete ‘90 | La Junta, Arroyo.......... 222 22 c eee eee ee 41
Magdalena. - . ...-.. 2.2.20... e eee ee ener e dened 89 | La Laguna...................22-2--004. 46, 47, 64 (desc.)
| Margarita .....-. 2-2-0. .e cee ee eee eee eee ences ¢ 40) Lea Natividad ois. s 2 ccinc cd ederice scicimemeticcien « 33

 
 

 

 

 

 

182 INDEX.
Page. | Lepus—Continued. Page.
La Paz. crete e cece eee crete tec e eee reer reece ee 10, 51 californicus—Continued.
Bay csawexee vaxeey sods conanmmeeaee 42,43 (desc.), 73
Plaine c:,0 cee sereeseee poe emeaua rie ee 73
population. «. .unccececeuneceenae esas ee ess 10, 43
settlement of ..-....-..--2--- 2-2-2 eee eee eee 7
Tia, Boba a 2.-.-ccien ns semnaninneednicmaadlicerineeas ‘sent 16 martirensis............----- Bence tedicwieys 159)
La Providencia Canyon..............--- 17,18, 19, 21,56 | Libocedrus. . ...-ecccccccccccccccecvscvceeeecees 131
Paiva wesae's veisisiema cent eeteackesommesns 21, 54, 56 decurrens.........--02e-cceeeeeeeceeeeeeee 108, 131
La Purisima........---..--2----222-+-+ Poebetibilee 10 | Lichanura roseofusca..........-2---.222e0ee eee 114, 130
CONYON anise ds vemedessomeseeecddmanesiaiat sey 62 trivirgata ...0..00006. 2 ce. ee ebeeeteciveceee- 114,115
‘ MAS8ION sccssewenseennsnet es ensorekiel seein te 37 | Life zones:
population ess sycccesenudneseeteniowsxeeceans 10 Arid Tropieal socccs00s0<s000cecebessa ena 121, 123
Valley (dese) -s.-:22...5..2seancmeertes saceane 37,77 Canadian
Lai Republica... -:...icceisiae sects nce e nts enteeeseees 15 Lower Sonoran
La Soledad, population. ..........-4---------+-+- 10 Transition
La Virgin de los Dolores Mission......-..-..--.---- 7 Upper Sonoran
Lagoon, Black Warrior..............-------+------ HELE li STAMOS: «eaters ek hss ts ata dhatra akd ge ee ane
San TNR. 6.22. ocanccceccemnan tesa «eben 77 | Limnodromas
Santo Domingo.......-.....-20+--0-2e22e+ 30, 32,71 | Limosa (see also Vetola).....--.--- eee ee 116
PeaMMON 40.520 ececcceuiiceie sven eee we caiewere 32,71 | Lion, Mountain .........0....-.---0--2005 ae 110, 132
Laguna Hanson..........--..-.-+------ 15, 55, 80 (desc. ) " gea (see Sea lion). ah
Ma GUS tA s <inmincesioiterciete suas cess 19, 69, 74 (desc. ), 79 | Lippia 108
Mountainessocse wow esnue es sele's soc ewemecs L.. 48,55 ~ parbata 123
Ranh. ececseseeeetsenerssess ves ceuewecemos 55 formosa 123
Saladavccceeseusere ces eeas eeeeeees 69, 74 (desc.), 79 palmeri 123
Lampropeltis boylii boylii..............------+-- 114 | Littlejohn, Chase, work by 143
CONUNCHAs os ccccemwewss cs sccmeereaees 114,115 | Lizard 88
CAHIOMIAGC... 0s0oc.50'5 2 22 os cee 114 alligator 86
MIA on Gieeisiciitcee esas ee teed eecdeees 114,115 | Liano de Berrendo 31, 72
Land grants.........-..-02---222ee eee eee rete eee 9 de Buenos Aires 25, 76 (desc. )
Landing, Barril..........-...---- path ca capusereveuadanelee 31 de Hamalu. .........-.----ee eee cee ceceeeee 70
CalaMaA Us cccicoin:e ss coc decuteddaeeemees Bed 27 de San Augustin. .....--..----22-2--- 25, 76 (desc. )
Santa Catalina. ...........-..2...---------- a 76 de San Bruno (named) ...-...-.-.---- 12, 73 (dese.)
San ta: Catarina sca cicccia: 6’ onsets seiesenieulsicies 25 de San Francisco...........-.-2++-2+-- ee eee 26
Santo Domingo..........---.------+---+---+- 30, 71 de San Quintin (named) pepe cada od ees = ee 12
Pansija cts oes nae Baatelas neat nore siete leinteeuense 116 de Santa Ana.............--.--.0-- 28, 29, 76 (desc.)
ludovicianus gambeli...........---------+--+- 130 de Santa Maria......-.2...s2-.2..06-- 26, 76 (dese. )
Ean vite dinates oeivte a scomeneeacworsseneaenannae 116 GOs Vira ese atee ages epee pe a ost 41, 72
solitarius cassinii. .....-..--.:----------2---- 131 del Perdido..........--------ceeeceeeceeeeece 71
WW cas@nus. nc sinc eect eneecaseneeenes se 121,130 | Lobipes. .....---.------- 222 ee ee eee eee eee ees 116
Lantana. ... 22. cee cece ee eee ecco rece teenees 108 | Lohr, F. von, work by..--...-.-...2:22-:-c220.54 UAL
CaMAara...--+-----++--- soe -+-- 122,123 | Lonicera interrupta.............0.22.002eeeeee ws. 129
involucrata....--.-..---- + eeseeeeee eee e eee 123 | Lophocereus.........-.------22-020eeeeee eee ite. 106
Larus (see also Blasipus, Chroicocephalus) ......-..- 116 australis...........--2-20-e- eee eeeeeee Lave. ©6198
Las Animas Peak.........-.----+-+----eee seen eee 66 Me 126
DAs CUCVES: 10.2082 sos saeeeeeeccdicieeioce neces 24 116
Las Delicias Ranch. ........-----------+---- Netavete’s 35 151
Late ov. cok oe cane cnner ees oo Pee RO ew Ee aes 117 1b
Lead www emer mere eet eee rete tres eee ‘bawelee eee z 8 126
Lemaireocereus....-..-----------+--+-- +2222 eee 90, 106 126
CPU CBee o.55p 4 yee iter aiererelswnedieeaeeiaicce 41, 106, 122, 123 51
gummosus...-...-.-------------- 90, 91, 92, 123, 136 15 73
tHUPbetlmccesecstaceeoesswenaxewene 90, 91, 92, 123, 136 | Los Batequis...........-26 2222 eee eeeee eee e eee 16
Lemmon, J. G., work by. .....-------+-+++-++--+++ 143 | Los Buenos Aires, Llano..........-..-...0.2.00005 76
Wie Ons s.seciocs ckaweere neste 33, 35, 43, 71, 188, 139, 140 | Los Candelitos ranch........ Balad alata pone 36
‘Leon, Arroyo... ---.----+++++-++ susie iSiejerstaree pines 20 | Los Reyes, Cajon..........- ious 0ie ue adhiebenise ' 42
Lepidospartum squamatum....---...---. waetesee + 126 | Louis (Henry) & Co........... PSR CRS Ait ate ne ciatciaias 29
Lepus...c..- 222-2 scene eee ee eee etc ee ee ecemece cece ‘| 117 | Lower Sonoran Life Zone... .-.......-.. cece ee eee: 124
alleni tiburonensis.........--------++8-- Ieee 167 Ditds deme seem kce tsa atie ce eaenenmlaeearearet
californmicus.......------2--- eee cere eee eres 92 mammals
bennetti..........-2- sere eee reer eee eee 130 plants
deserti Colaes<sc2ece xGsnewienigee ican enn nace 127 reptiles

 
 

 

 

 

INDEX. 183
i hy Page. Page.
DOs, 3 2 tob seek ceguatin eiapetcecatd st hhaetess 116 | Mearns, Edgar Alexander, ware Dye vseese ee geek 145
TMCIANO, POZO cbittoccasaanasecuehleny<ieunlaoeeeea 20 Megascops ASPOLSUB. os JG alk sods yectaaeeen cease 153
Lunt, S..6.-O;, work byicvascces 2s aeetieseeeedes 145 RANGUGL 2.5 fs ocsyriweaccsecak eos nasedoe mene 162
Lupinus sosdese ce ests seneetdeees ss itdsetaewewee 110 | Melanerpes (see Balanosphyra, 115).
Pyelumisceivececense occ casecaeer ss i cagecuven eave SLOP | Melanittac22sccc.cacuceee oes oy eguenwaseubisess: 116
Lygida exotica..........-.. se dite Along beeeveme stale. D2) Melisti oa 5. oct tpt oetave ee eh cr sn eprtleens 8 149
Lynx......... Shins Wise Sa eee Se engnceeuteee 117 | Melochia tomentosa..............000 0.0002 e eee eens 123
ruffus baileyi........-.2-2.----2-ee eee eee eee 127 | Melonipyece's ses -csaitn cides cei os saceenghaee 23, 137, 140
Californiclsscscsssine wae sy se cere dices 12), 127 | Melopelia..... ..<. ..osecetweaws ceeseccuivnsessecux es 115
Teyelom ay 4 3jo)ssmessvssened Sass easswaneoneeceess 108 asiatica Mearmsi..............2..222 2022 eee 126
CANCIDS jijoncndoadiaeweeidere na ene 105, 122, 123, 136 | Melospiza...........2. 2.220000 eee eee eee cece 116
microphylla: 2. asic cca see ese eeminies + 122, 128 fasciata rivularis............--.2-2-0-2--2-20- 154
MacDougal, D. T., work by...............-.2.---- 18, 146 melodia cooperi........--.-----22e2ee eee ee eee 130
McGregor, R. C., work by. ..-.-....-------5-505- 143 coronatoruM........-.2---2-e2eeee ee eee 85, 163
McGregor’s house finch. ..........-..-2-24---2--- 88 TUVUIATIS. iced Sele otc eecaeeceee sees 34, 120
McLellan, J. E., work by...............-.---6- 45, 145 Balto Mie edsacnd ocd a emaeeinn oe oees 126
Mackerel sc o5scccceccscedesie oe asousxseesesects 134,139'| "Mephitis: :0c6 30. 33.:\s stejeon e's tenourenewee ese o8 4 117
SPAMIBls. soe qe sea tlyjaiedee eda ees 134 mesomelas estor.........----0-e2ee eee reer 128
Macrorhamphus (see nesiotius, 116). occidentalis holzneri................--------- 130
Macrorhinus angustirostris...............-. feyn sence VOO | Mereus! oie cc os ass baauleteancunpexisbeeceeceue esa 116
Macrotus californicus........-...-.-2-----e+-++ ++ 149 | Merula confinis..............22--eece cence eeeeeee 151
Madrond..ccsevsewgcccs cis damagsdendes 46, 104, 109, 110 | Mesa de San Carlos...........--------- 50, 54, 67 (desc.)
Magdalena Bay... -. Giascedexs 10 (population), 89 (desc.) de Zacatecas.......-..0..2202 cece sence eee ma sce 36
COMPAR Ys. c0ccncesswass neaevewsemecsacnt 9, 10,140 | Mescal.... 2.22.2 ccc e cece eee e ec cceceeeeeceeees 140
Malan dinjnesje cits ssttanceetiiesecatle seas 89 | Mesembryanthemum. ...............-.02222000-- 88
Plains wove hh oie secseesieneaies 12, 72 2 (dese) 138 crystallinum.........2..2222..0222-eeee eee 85, 108
Magnesite.......-------+-++-+e eee e cece teens 8 | Mesosphaerum emoryi..............2000eeeeeeee+s 126
Mahogany, mountain............-.-+----+-+++-+-- 104 insulare......-2---0.02e cece eee eteedeceeeeee 123, 167
Malacothrix insularis. ...........-...0.2---+2-+-- 152 Lal AOPUin ioscan tea eee ese peels oe 123
Malarial fever...........--.-222++02-reere eee ees 38 PalMerh no. cece caesenesawuee socmswersenaeces 123
Mallard.......-.--.-.+--+222 220 ceee ee ee eee eee 21 | Mesquite........ 17, 28, 36, 42, 44, 75, 106, 107, 109, 125, 136
Malpighia diversifolia...........2.--22:22+02200 IO | Manet al ive suv itu Svab nce suee daadeesdalanes 10, 19, 75
Mamillaria. Pea 106 population Sub arb tarreconecs eee Wiiae Ss wats ets eke eS 10, 11
. haleb. : 2.20.2 - ence eee cee ee eee erence ees OO Noah cise ie ot Serene or ta or atch oe Ae 8
_. FOSCANA. «20.222 ee ee eee neers p> 128, 156 | wricropallag. ...2....0ceceeeeceececeececseceseere 115
Mammal lists.. 87, 88, 90, 92, 95, 110, 121, 124, 130, 131, 182 whitneyi...........020202000- Benen ae eae 112
oe BAME..----- 2-0 see eee eee eres reer es 132, 133 Miltro tus eee ne boy 4 ii aeaic wh bele's ata verdes naee es 117
Arid Tropical BONG, itivensonce neiennnese can 124 californicus huperuthrus. ..........-.------ 130, 163
Canadian Zone............---22--00-+-eeeeees 132 Miller, Loye, work by........-- ah eR ES Son 45, 145
, Cape Faunal Pai Seetoensies spoece aeons EN MBER Le cored a aedacutdanhaeen dene heceuae: 108
Lower Sonoran Zone.......-.-.----+--+-+-- , 127, 128 PUFPUFASCENS...... 2... eee eee ee de eeeeee ee 123
Transition Zone. .....-------------- patdbbaie! Taz RE ET RR nie eee 122, 123
BOP Sit tli. 4.cemen « vapicienonenean vexed isco: 110
O ANd cannas sae alec anuien es ds deaeaaenn 116
Bs Mine, Alamo........-..-..---+--0--20-e ete eee eee 16
108 Oalmallt cic ncn Chauete Bae qecdeidenied Fe
nes pee | Taine sects oo 3
Man-of-war birds..........----22----+- wereere 40,90, 112 AE Bo leGieres estate lies tua iin, tue deatdeisines 8
Map, desc....--..-----+-0++2eeer seer retest eres ees 12 EL Trinto :.3-csdec seuseesenees ese yaaa’ 8, 43, 47
Maquata, Laguna.........--------- BL Vial icscadew-at secotacceuswneice eo cumauns AT
MAtble-segnushartaits vu: a eternrentlagicesioneret a fk
Mareca. ...-..----+2+--20- eect eeerrers erste King Richard..............-..----- Adee” 26
i "Llano de los Buenos Aires............2.. Sema 8
Bete sae Cee tence hed Adee ee
Marks, C. H., work by ; an NG aE
Mary Sachs (boat) og ad agen ggPNOND 4 Pinta Mountains.............2.....2202220055 8
a ee sR ea eR ce SS Salar ocd s torsades fenced / 9
Matancita ranch......------+--2++07+07+- rat oe Sam Avid ress. ois)os seek eo Buc B Sot ha lose ceca es 68
Matomi Peak.......----- Bette eer esses aN
OMe Gye acanasi aoncandacasaanrbes aie, BI San Fernando Lagi Sawaecnea diggs iahnaesiche &
, Re ae eae, 12S San Francisquito.........20..2-.222-200202 008 58

 

Maytenus phyllanthoides..
184 INDEX.

 

 

 

 

 

 

 

Mine—Continued. Page. Page.

San Jose.... 2.00.00 ccc ceeecce eens a eee ' g | Mother-of-pearl.........000.0 000000000202 cesses 48

| Mount Limantour one th 68

Porfirio Diaz......2....000...b bee fat 63

Santa Genoveva Be tus raecieceeet 638

FP POV OPS cesses oe oS iu as cs bsisicshiced reostevies 63

Mountain lion... 2.222.222. ee eee ecb eee 110, 182

mahogany.........- Sore Redeemer Rigtione. Od

Mountains, buried.................2....22. ecin ae 69

Calamahue........ dasa ecicicnn ee eeaehotetornerns rorisveteve 27,74

Cocopah ...........2-2..cceceeeeee 18; 68 (desc.), 74

Columbia.........2......222-222-000- 28, 29; 67 (desc.)

TGP WHE 6.5 acne etevietete x eats . 48,47; 64 (desc.)

Laguna Hanson re ER acd acerca 55

' Prieta Range.......... rae Sins 29

San Felipe..........-222:2224- 18,19; 69 (desc.), 74

San Lazaro.......... sedan AaEeised “22... 63,78

San Pedro Martir...............2.2..... coud. 17; 38

19; 20 (work at), 25, 49, 54; 56 (desc.)

Santa Clara...........0.0220..- 32, 34; 67. (desc.), 72

Santa Lucia................ 34 (desc. 4 35; 61 (desc.)

’ Santa Rosa... ...2.2.220222200.02- 18,19; 70 (dese.)

Signal......... eee t: 68

Tres Virgines ... 2... es ~ 60

WI CHOTG acrcarsiosinnewcanteaahaces 43, 44, 62 (desc.)

Vizeaino . 2.22.22. .2. ccc ence ceed cee eeeeeees 67

Sierra Columbia................... 28, 29, 67 (desc.)

Sierra de Calamahue y San Luis ......:..-... 58

San Francisco.............2..2-..02220-- 60

Santa Maria (desc.)......... ieee. BT, 5B

Santa Ysabel (desc.).......:..2.......... 57,58

Guadalupe. ..........2222....2222225+-2. 7, 15, 78, 75 Zacatecas.......--..222.-22sciseeeeeeeeee GL

iat PBL ace ese ccs cers occ ocaccccos 37 Sierra de la Giganta............ Beijeaearehe ve 61

La Virgin de los Dolores Muniemmememaaee nice A teks 7 Laguna.............2...22--..- 43, 47, 64 (desc.)

San Borja................2-2-2222-000- Lb ate 7, 28,77 _ Trinidad...-.-........2+22.2--+-+- 45, 64 (desc.)

San Fernando...:.....-..----.0-ceeeeeeeeeee 7 _ Vietoria...........-.-+:+---..- 48,44, 62 dese.)

San Francisco Xavier............. celal enenyt: 7 : Mates 58

San (lenacio....dossesenntincassteseeewsmenty bs 33 61

» San Lisi case nanevenese ess erectue dee welts 42 55

San Luis Gonzaga...-..........0.-20.202206- i 7 : 60

San Pablotcns: cn exnesseviseesehe seneReececes 82 Sierra Juarez.........- 15, 16, 17, 25, 49, 54, 55 (desc.)

San Pedro Martir.................. 7,22 (desc.), 57 PiNtho.<..B rye cacee un aeealdte 19, 69 (desc.), 74

Santa Catarina...........2.22.222-eeeee eee 16, 55, 75 Pibteda..oicsssereecesde eccens 82, 67 (desc.), 72

Santa Gertrudis............22..2.002eeeeeeeee 7,51 Santa Olara..............:...... 82,84, 67 (dese.), 72

Santa Maria.............0. 042 ce cece cece ences 7 Wind ja cwneker itis oie se accdanaeseesvaecenns 74

Santo Domingo............2022202222ee0000et 23 | Mouse, house............- aieaas aeons ABR ee nice 94, 95
Mixter, J. W., work by........---------22-2+02-05- 12
Mniotilta. . duced crawortelauaivelenn ine aeenann 116
sels areca erect beers fasaade: 112
western.......-...- Sauget Sev SwAeateaee dat 92
Molothrus....... - Hicisioarens gata caclsaraneate etnemene) IG
ater obscurus.....-.- kite Perea oor .--. 126
Monardella linoides ats ee 181
macrantha......-.------c¢geeeeeeeeeeeeeeeeee DBL
Monkey flower.............2.2-.-0eeeeeeeeeeeeeeee 110
Moraines. ...:......¢-eessiesrtewsstc seasuensous ees 54
MOrM00pS..... 060 oScnkeiseens PeESe ees ase aeesenee, LY
megalophylla senicula..........---. 117, 121, 124, 128
Morning glory....-...-.-.----.eeeee eter eee teen . ho
Morongo........-- costed page epec you scugus asatora eistar trocar dee 188
Motacilla..ccc-cnc-csecsccecenseeeeeceeeeeseseegees 116
OCUIATIS Sasierrec sents a vetvioles vem ards acevo cbL2; 161.

 
INDEX. 185
Page. Page.
Myloch anes ooo o.03c020 scene's w/a eelgigendeieis v'o o'elecipahaw WD | (Oakecoc oki ee eh Le CO Ny tah inte Cah 104, 105
richardsonii peninsule.......... 112, 121 WANT soe OA ta to ad dame el oh 87
MYVOUS so awn cee dala en one aaa eae ees 117 Live oes ee Jos Sooke acini Ee ERS 104
californicus..............0.0..- Speen: ace 160, 161 Palmer soso: cntasionaeee soa anak Ue ee 104
CAPILANE UB )26 020 znd adiacncecdieewacenceaxe? 165 SCI Dips adhe a carmen Ce.oe cng rcciends sates 104
evotis milleri-.............0..0020.00 20020000 132: | Oats, Wild cnc sarawtae Resse eee ceelne s Vaoesloned 95
MIGHONY x seedy ys hy eee 121, 128,165 | Oberholseria..............200.02.. 02222 eee eee ee 116
Da Ort spy heeetan a hte teeta eh Akan ats 163 | Oceanodroma...............2.02. cece e eee eeee 116
OVIMOMIUG Sialic Poo wees se ds y Rhamaiedes 2S 132 leucorhoa macrodactyla................2----- 153
peninsularis..................222....-- 121, 128, 161 monorhis chapmani......................---- 164
VOIANS oecisies Soe. oerehe ee aaeebetiat 121,128 | Ochsner, W. H., work by.......-----.-..--.--+--- 144
YVUTMIANENSIS swede uve se sos ge eeaee desk eS 128 | Ochthodromus (eae sien oo
Myrtillocactus............20. 000200222 e eee eee 106 | Ocotillo............... ee . 24, 29, 30, 31, 45, 90, 106
Names, new (see also Place names)...............- 11-12 | Odocoileus..-ccs0s0c0s2sessenetcauease esasmeetes 117
NGNNUB2:s.cccceassasdancaces ceemce keene Seeaore 116 COITOSEDSIS.... 2.22.02. e eee eee eee eee eee 130, 161
Narragansett (boat)..........2.0....00002-eeeeeee 140 hemionus californicus...............-.----0++ 132
Natal tigi acc fal eae why we hie ieee ences 17 OF OMILCUS: ¢ cc.a4.3pmaeed taeets himansene 127
mexicanus..................... 117, 121, 124, 128, 162 peninsulae............----------6-- 92, 121, 127
Natividad Island.................2..0.202000000 88 | Oenothera arborea..........-. 222.2202 e eee eee eens 149
Natrix valida.........---222.2.2.22020-0-2--220- 114,115 | Oidemia (see Melanitta, 116).
Needle Peak. ..........0. 00.00 c cece cece cece eecees 63)'| Ojo de Laebre: ss: os -ccieeisc Sacco dwice’s bovis 32,72
Nelson, E. W., work by.........--.----------0--- P45) Of08, NearOs os <i a's aici Paani seaia a2 5 cee aseles 15, 75
Neotoma..........--2eceeeeeeeeeeeee cece 90 Ae WIG IAG, | OU cncatepenacd neaeas et aamaaeemecsiaeaharnaneee 139
abbrevistascasccossecseedsceete'ses sees 127,165" |) OWVES sco 5sccaciccunet os eeietslst aot ee iotele's 33, 74, 138, 139, 140
albigula seri.......... 167: | Oli Ga since euuseldene os oy comeeuts Boas de eee 106
VONUStH jae ei asralia crane ios demesne een 127 TORO tise ssa ew Sess bcs kaise sees ee se ates 91, 107, 125
ANtHONVLE a5 .ceondiecsecnaneneens ee canes 85,180 | Onychomys.....-...-.--. 2222-022 eee eee eee eee 117
Dryanll se sos frag seid See earns aie 130 MORCPOUIGE jis 2s sje dee de veisiaieapiatisiene S5.5-2 coc Nien 163
fuscipes macrotis..............--...-.-----.- 180, 132 torridus perpallidus..................-2--.0--- 127
insularis...................-5 te hl estates 93, 167 TAMONG satus cnaanciecashasee vonoadeee es 130
intermedia arenacea: <2 2<<+4s5--4ssecsexcsee 121,127 | ONYX 2..cccces ee: reece sete eee eee esas ee 8, 25, 76
Palas cee) setae neces $27 | ODUM ths Joel ov atdseccceaunlcnce sommes 85, 90, 106
intermedia.................2-2.20-- 121, 130, 182 PIPEIOV sso keen eee seas Ye x Vek eeeeen 28
perpallida..........-22..22..-2222-22---+.- 92,127 Olav ellinaiscee sso datiicinecineiiesia doe Seema 126
pretiosa............- wiieiteeeees 121, 127, 165 comonduensis............------ 2.22.2 eset eee 123
VACUND pore a cictraiaksrirs feta ssid-ae yralata'g Meneteaiees 127, 165 rotundifolia............2.-..22-2-2222222-0005 156
« @ depidacccsosccsonscaw ses ectacins ss Asc cemnsscese 127 THPONS co cccce eee ttc kab beeeunendevowes 123
martinensis.............222--222200-002+ 86, 130,163 | Oranges...........-..-.-- 33, 35, 37, 38, 43, 71, 138, 139, 140
MUGICAUT G.... .0ccie ee aceecioses ss 24 sees ence 127; 168! | Ovchilla scncccccccceesexwovedsesestes iasseae 40, 108, 135
Nephoecetes............--.22-22-.2--02 22202 ee eee 115 | Orcutt, C. R., work by-...........---2--22.-0205- 142
Nettionssscc oecets 3S: agaaoesue see emigedaceace 116 | Orcuttia californica...........2..22-2-2-22- 2-20 eee 152
New Riveric6 ee y a5 oooh we tenee nc o's sale ke 19 HOTCONY E ocnc scenes somieeslelnd cele Se ce nn eeiee diene 115
NOM Nas dais oo eae ns Seas de eS ER 106, 109 Picta confinis...........2.. 20-20-22 ee ee ee ee ee 131
Delditigt. on225.5524 2 ¢ceiseceueist seeks 129, 155 pictus confinis...........-.--.---....---2.22. 155
HigelO Vi. ccescsneascdaaietenined sasenoouen ees 125 plumiferus.......--.---,--.02---0eeeeeee 162
palneris a: 2. Sgsc eso sd edeenee eae 129 | Oreoscoptes.-....--------- 2-02 eee eee eee e eee eee 116
Notiosorex: ceva: eee eed eee Ss see eeiarsee 117 | Oreospiza (see Oberholseria, 116).
Nucifraga:,....0c.c03 500+ ssaccn eens ss osesesendeese 115 | Oriole, Scott’s..........-.-.0.0-2- 0-2 e eee eee eens 29
columbiana........-..22-222022eeeee eee e ee eeee 182 | Ory zoMys.<..20020<-scssonenne scam sess posi tewes 117
NUGMOnIUE. 6 34h nei cc i.gonsctines oetaseedede cen 116 peninsulae...-....-.-.-------.-+-. 117, 121, 124, 160
americanus........-.------eec cece nce eeeeeees 166 | Osburn, P. I., work by.-.-.......222.2..2.22200028 146
Wud sonics: cccvnessw ence cass os eeieceiwe vineeies 166 | Osprey secsccasoceasodecewek ase ce ess ceteedames 86
Nut, pinyon.........2-220-2200+2eeeeee eee ence 136 | Otocoris.........---.--- +--+ 2022 eee cece eee eee eee 115
pistachio...........20.0..-222eeeeee reer cee 136 alpestris @ctia. ..-.- ncedscwase ee es aceendcasess 130
Nuttallornts: cis ccaxexsesneneseee (422 cnwsilsexne 115 eMertera...... 222.22 .e eee eee eee eee ee 126, 165
Nyctanagsa.........--2222200ee cece cece eee eee 116 leucansiptila..............22..2022202200. 126
Ri gitisriag 25 wteaeapeicns sosacndadwnd caecmpanties 117 | Otopterus yess. snaceno see. ihece se se jaa geidced ss i7
DH PORE AS wioeecancana v9 ve aomeecanceaambantasndes 116 | Otter, sea..........2. 2.0.22 86, 87, 88, 95, 111, 133
Ri pcGsnt chee oa teen eosin jE OMB a nat ie sen ss mewn nncienica ng sdbndelaien st 115
depressus.......-.-------2-- 222 eee 117, 121, 124, 128 asio Xantusi.........2...2-20200220200- 120, 126, 130

 
 

 

 

 

 

186 INDEX.
: Page. | Peaks—Continued. Page.
OVS as ed, sacesccnsaninndeceeeneaac antag es 17 ify Band Gas « saiciescseyoins yacaisin/asaigts\aioreisacie asec eaa's ne 60
CANA ONSIS: 2. jew ascends Dessimeemcwanasbnet 121 La Encantada......-.-.--.2-02celececeeeeee 56
canadensis cremnobates.....-.....-- 127, 131, 182, 163 La Providencia............22-2220-222e-- 21, 54, 56
Owl elisecc.cuhtide dete Veda cu snel ee eva tne 112 Las Animas...........----200---202-0- me. 66
PIPMY socerses cae ves veekewee see eee Ereee seed 36 Mato ictal Bee geet one ceais 57
OXY eChUs seitie sees xc eeeeeetiocccuweae eae 116 Mount Limantour..........-..--.- pias ee dads 63
Olystery edibles a aie 52.5.5 sab srcsaiasateresareseietuenenesecedeiete 135 Mount Porfirio Diaz. ...........2.-------0--- 63
pearl ie cacececccanninhooteteceddeee 48, 91, 113, 135 Mount Santa Genoveva
Pabellon, Arroyo...........2.....200202-0008 raseyeicnie 61 Mount Troyer...........-.---------:
PaChyicereus siniee st teiercacecwenecanees re sressioie 106 NGO eS css; ce aacinhovcccunale aiiterayemaesens
CAV ois ck Fa siclevs cht pagent eeterlexre 90, 91, 92, 122 San Bartolo.................--------ee eee eees
pecten-aboriginum........----+---+--+-- 90, 122, 123 SAT WAZA <:-c.csacccers eee averace cases a hee er suei’
PRIDE OL? eso eye ds eee ere weretats 19, 24, 25, 106, 118, 126 Santa Catalina. .........22...22.022--222 2 eee
Pachycormus discolor.............--- 25, 87, 125, 126, 151 Santa Lucia.............- 0-02 ce cee eee eee eee
Padre: KimOizcsecen sx cxewsstaceerexscae sees ceasenus 8 WRG cop: eesdad de faethe py toheton tanh esc di teresa nase Ste
Salvatiarra: ..224 ..ceenccnaveatereress seceees 8 Werdrtesjsa- beaten nyc tasciant ie'cis apse wachamaene
Upgarte.... 22.2222. e ee eee ee eee eee eee ee 8) Pearl Hebi npc: santa wanoc aeicly deveGariosee we Ueaisela
Papa goic.2 cievsicicis tee transits eiesmmaisereisie acy Sees aietarensiebie 116 OVALCTB sys cyan earcices Saleh a cian re sluts 48, 91, 113, 135
PAINCSA CUP a:2.5.5. ss nid aooracsscrerredas. ain sie Mopac eenteratw OF Pedra sila oe eens tren Daas tells 71, 139, 140
Palafoxia arenaria..................22222202 eee eee 123; | Pea ssean-wanseenseemeuecs Saeed RU cea na Nod ot edaialshalids 140
Palit sacesacccewewcie ede anna aa aenneiaianas 94)" Pedilanthus:o:. 0s seSudc . oeo deere een 90
CADDABC saa oc wcncnsnemascueebnnemamnisseine® 95 macrocarpus.......-------- Leese less. 90, 91, 92, 126
COCO ths tce canines sena vile 107 | Pelecanus...........0.00-002 cee cece ee eee eens 116
OU fst eared apa luttve Wtarergiotesstoheuabatoiel a2 fa eeeee 35, 36, californicus.......-.- 2-22-00. cece e eee cece e eee _ 154
37, 38, 40, 43, 77, 90, 107, 189 (introduction) | Pelican.............-.-------eeecleeeeeceeeeee “42
PAN sn a tisceieemienientcts 25, 32, 44, 45, 104, 105, 106, 109 PrOWIs . acscceassraaasnenesansaee 23, 86, 88, 90, 112
Palmer, Edward, work by.....--..-...-+.+-+- 95, 141, 142 | Pelidna........22. 2022 eee eee ee ee eee eee 116
Palo blanc sssccssnewecacevesaeesncsseses 33, 34, 105, 186 | Penthestes............0000-2ece cece eee cece ener 116
OBICO scecese ewe vac ds oeemeeeeneeeeeeeses 102, 110 gambelii baileyae..........--..- EA cecauneuseines 131
Verd Oasewamudae sed daesnreees 17, 106, 107, 109, 119, 125 | Pentstemon antirrhinoides..........-..---- Se areca 129
Pandionsaccscceesnsseewec acca seeeesecteuweeeaeess 115 centranthifolius................--...-.-.------ 129
Panochais 2... -sseeioe oc5ce ca cat eeueee eerste 139 linaroides.............---------- fb eastowinteurome's 131
PALADULO ss... cne.iecie cise ecacScuie seem Seteeeeeeceet 19 | Poppet. - sos se cccevceeeewe scarce, pishepatepinsyoncre 140
unicinctus harrisi...........-.....-.---.-00--- 126 Chili. ......-. Bod Sante spatraeas dulce 2c Cease dt ncenahs 139
Paral Camyon....-......--200-.25 2-022 eben e eee tees 57 | Perdido, Cerro............ Seis sprcrtalergehtdeitta cs 31
Par iriOD Las 2.5. o.c.c.2-s.av's Shaun eiaraiclnisvete sins s elaceeesnecscrdsers 106, 125 AGMA OB pr sesh seers ete eee ees reel Salcbe ete At 71
aculeata...-.-.--..22++eeee eee eee eee ee 107, 118, 125 | Pereskiopsis. .............-..-22-22+-- See@ieveca 106
JUIMORS La ska ct noon ced akmtaineuunimcacien 107 brandegeei. ...-.-.2 2-22-22. eee eee Bheess 92, 123
Microphy ldine. go sgesscense nese ecewnn 118,125 | Pericue Indians..............-2-----.022005- re 7
Parosela spinosa... ......----.2---20beeee eee eee 107,125 | Perityle cuneata...........2..2022222eeceeee eee 155
Parry, C.C., work by+«o-c0.s20cscnsaeeeesnaces 142 POtUNGLO Lia si excess oe oe eitvine we ieee a sine 158
Paso Colorad 0. nccccccancinnesesemesscaiecsncnnne 36, 37 | Perodipus....-.-...-2----222222eeee eee eee ees 110, 117
Pass, Buena Vista. .......--.-.---- Bassi ecaian 18,70 simulans... .-.- 2-22-22. eee eee eee e eee eee 180
"Sar Matias «<1 susceeuvaceamncienine cence 17, 18, 76 peninsularis...........22.2..22... eee 127, 164
116 | Perognathus...........2...-2.020+- 28, 87, 92, 93, 110, 117
180 AMGHOT YR ooh 5 oisis ose loraw uaa damcnadrna teers 130
130 APONBIUURS 603 edu aceon mnncnnun ace acassasad 158
116 baileyi insularis..............2.....2....0200- 167
116 rudinoris. ....-..........2.--.000- 121, 127, 163
130 DrySiltiswnas rend cannnmanmasiteecnad senses 158
. 120, 128, 124 fallaxcfall ax eric einer mda tercrewie ci geraanieeteeeiio 180
45 pallidus. ..-coucreesessesesewesdie teen 127
23, 24, 71, 189, 140 femoralis femoralis..................2..22--- 130
a: mesopolius................-.2222- 130, 163
64 Helleriesjasncisstosssecs ese wataeaas 168
56 RMOVGUSS occ cieereicnimeacsie ke ea aeeeeamanen 163
17 margaritae. ............ 2222 e eee e eee 158
58 PACH CUPS dccinctsnsmasannnsoabemcacesanimies 130
62 penicillatus ammophilus.........--......... 127, 165
61 angustirostris.................2...02....- 127
ae 66 VON ATINUG Soros choses, ea wcie creisujererenctecoie uomiays 127
Sg A ih are nein area oe ass eaekaens 47, 63 Bold mant sos 2 syed iow scwarem nner ele a de wh 167
 

INDEX. 187

Perognathus—Continued. Page. Page.
penicillatus ammophilus—Continued. PHO DASHA as ceivsie weiciseerres scaeeies sakeduwes oeex 116
helleni2e 2c sei epee aise tald Mel Semone 130 | Phoenix dactylifera (introduced).........-------- 123

BTL igs re taht a eS Oe haa ae hs 167 | Phrynosoma blainvillii..................-.----- 114, 159

BICCUS S22 eCrectshesds cule Ayelet 91, 127, 165 COMOCN SCs occa sno vec yeiasen kha, 114, 130
spinatus bryanti..................0..06000-5 92, 127 coronatum..........2....2..2.2----------- 114,115
magdalenae................2.2.2.0000-- 127, 165 platyrhinos. . 2.2.22. 2.2 200 eee eee eee 114, 126
margaritae.....2. 2.2.22 e eee eee eee eee 127 SOLA os aie seve ates wed oats bs Ldn anaunes 114, 126

Melson so sachs cece s tk scteacccouicen vlciclere 167 SOLAS x2 2's 5 «-chidain ONeames pana Mokarersee seis 159
OCCULUE re Saiie cloacae dedemeusie oe ves 167 | Phyllodactylus tuberculosus................ 114, 115, 123
peninsulae.......................- 121, 127, 158 WN CHUB scctu yaad nosey use eeeorenae 114, 115, 123
BPINAUB wise wisSeve weiss Seiecern des it ati 127 | Phyllorhynchus decurtatus.........2.....2.-.. 114, 115
Peromyscus..........---...---- 23, 28, 87, 92, 93,110,117 | Picus lucasanus..............2-.22--2022.02 22 eee 149
boylei rowleyi....................-2-. Bee 132 | Pierce’s ranch, work at..........-.2-----0200-05- 143
californicus insignis. ............2-.22-++00-+ 180 | Pigeon, band-tailed................22222--.-2... 134
crinitus stephensi.............-...2.-2--02208- 127 VAOSCH' 6 tui cniers et clon decide c's doll dsvieehinn 46
eremicus avius...........------2---2--- 91,127; 166: | Pins, 02 wwavecddey seats cuessewte vee d eens 87, 94, 105
COMMON See ceccks ea honest oats 167 Coulter. ......... Da ec pak a Benge enemies ates 103
GODTOSENBIS, 232 ccs des oeetauiekeaceeeasie 130 J OMG Y iis acecenead sees Mea ees ecs 15, 103, 104
SLEMICUB s.oseeeed se eiancroeesaa ei 127)" lod Sepolé:. niccccc.e cc cies Seeitie aes wu a’ ev cieten 104, 135

OVE SC aua ete brnee recited selene 121, 127 MOntGrOYy esc cetite: a dadedccdent senates 105, 108
fraterculus...............222002000020- ee 130 PINON Ly osadscuerss2245 ay dsene 57, 104, 108, 128, 129

INSU COlA seep screenees s axineeewee es = 127, 166 BUPAS > ceca veiweedesee so deta deeneess 103, 131, 185
polypolius...........-..---..0.-0.0+--- 127, 166 VWellOWericccssiiteadagies ee saucer 20, 48, 55, 181, 135

PAUMUB 2 oo Paine id eres peed BO aaamurciae es 2588 163; |-Pitiea pplesis..jcacina iuinae voenanwnia adeno vires 138
PURI a se actcaannwiinauiiales duoeeeaarein se 93, 167 | Pinus cembroides..................-.-..--- 108, 128, 129
NEMO OAS osc esses aweGek Ss wlasidlend woaeeaee see 163 GOTTA 3 usrcntees 345 otha e se ees 105
homochroia.............-.2. 222-2 - eee eee eee eee _ 163 CONLOMA Wc.c: wccisicisteacleives oy ss Ghar eee eee ee < 104, 132
maniculatus cineritius....-....-..-.2.--.+-- 89, 130 COUNT 3c. css ccninus ttegee save a oeeRee oe 108, 131
COOMA ge. soe eseceveiee esis ds eeweeede 121, 127 AUN 1 sais): vias Seeyotedege eels veclensengau ooo 4 104

ubaiis sores sentieteee stew cs asepeoo sii 85, 180 JelHlCydeiswscaes4ssesanesiedeei se actcawseeses 20
PamMbelt es 58 oon, saeeckenaee ssa seaeeuars 130 lambertiana............2...02..222.22220- 103, 131
geronimensis.................---+--- 86, 88, 130 monophy lai: 2: .usgnckimiceea re eeense ee 104, 129
magdalenae..........-.------+-+-+- 121, 127, 166 OC UNIS). 52s cuteratccakwen so aseaeeeees 129
MIBIGANAC Soa si isseesncekeeenne ee eon 127, 166 MUPCAG aia vote ees detexiee sau eened donc 87, 105

BON OLICNSIS sco 552 2 fcdtidadet Secs yess 127, 1382 POITYV ADD. <ciosl oe ces ceeeseawee cdi cccedeocund 57
OLEStOPUS ss es age cwnee se sy aee Snes cee ey celts 163 ponderosa jeffreyi..............2..2....... 103, 131
stephanivescecidasitecce ssi. scesaakcecte sees 167 quadrifolia.........-.... 22222222 eee 104, 129
truei lagunae..............2..2220200ee 121, 131, 166 Pada tase fod clges cs eawaveaaceen aeeancde 95, 105
Martirensis........2..2-02.seecceeweenee 182 | Pinyon nut.......... 22222222222 ee eee eee eee 136, 139

POBCAd CFOs wis cscs Sosieiorrsera sass Se sishce ae nis ate BA PIPL Os 25.5549 seersiccscaisle ang itch dee erda eh pe's Seis 116
Petrelli dacvcasta owes neetetahin es as Sane 85, 95, 112 albigila:.sovagales esse deeeeeee ges evecare 149
black. ices seeccoud « onensesecces ty haeaenexs 88, 166 consobrinus...........-0-2+0202e2e scene eee eee 95

Ne o8 bys tiscmctrte taisiats cerning Set coe 88 fuscus albigula........-..-.-2.-.2.-.--. 34, 120, 126

G08 BOCOTEO oye iejarainibwierca 8 ONE tisieeye coe ele 88, 166 BMPOLNUS yi ssoe cies ds 25 se nencinawene cocky 126
Petrocholidon:. so... eiisc ocscs senasccacwen eeaesie ee 116 BONICUIA ... rssuecaiarssddaowewes eis Bes 130, 159
Peucephyllum schottii...........22-.22-202-2026+ 126 maculatus magnirostris.........-...... 121, 130, 156
Pez de allo. cn. coccasnun ks sees cecese ees cdeceee 134 MOPAIONYRs 2 n.. casciee oes see ewadaecy sss 131
Phaethon (see also Scaeophaethon)........----.-. 116 | Pipistrellus...-...........-..2... 22222222 cece. 117
» aethereus a dsdaseibads Koel eoeuek see stadeats 154 hespetusee.: « s+). oscecbeeae ses caaineeuueates 160
ihiteamGhe. se: PA Aiea ad et urd Boas a 160 australis..........222.....0 222 e cele 121, 128
Phainopeplaccs ssc. sceow sede sos sete eten ges bees 112, 116 | Piranga........------2 2-02 eee eee eee eee eee 116
Phalacrocorax. .. .--cecececcccccececcceceeeeeees 116 ludoviciana.....-........2.... 022 e eee ee eee 131
dilophus albociliatus.........-.--+--+--+++-- 90 | Pisobia.......... 22-222 eee cece eects 116
Phalaenoptilus....-...-------------+22+ee eee eee ee 115 | Pistachio nut........2....2.0.00000 0000.02 e eee 136
nuttallii californicus.......-..----------++---- 130 | Pitahaya agria.....-..... 22.2022... 90, 136
Hitttallid. s.-cace02ls bs eswes sense ase eens 130 Gules acthee din cies saneemecuieces auenaces 90, 136
Phalaropus. ..-.------2-eeeeeeee eee e rene 116 | Pithecolobium dulce,......................... 110, 123
Phaulothamnus spinescens.......-----------++-+++ 123 | Pituophis catenifer annectens.........-........... 114, 130
Philadelphia Academy of Natural Sciences, work deserticola..............02.20.002.000005 114, 126
(DY woscccsicceleaetin va dewascieee se a -meemeeee 146 vertebralis...........02..0 0000 ccc cece eee ee 114, 115
Philadelphus pumilus...........-2--2--22-22024- LON: | PARON YR oi ccan a iaseeeun yds ecurdedaay es ¥echaean 7
Phoca richardii geronimensis abe CS. See Fase ET SS Aes 162 vivesi. BS RES SAGE RUSS TRENS: FSS S.8 PREP RIAN eS sede, bie: Scsh evel a 121, 124
 

 

 

 

188 INDEX.
Place names, new: Page. | Population—Continued. Page.
" Anthony Peak... csc ceediewic cece senses ee 11 La Soledad.......... b auiddedn es Lausanne uasste 10
Brewster Peak. .-........------- Rae enans 12 Magdalena Bay....-....-.....2------02022 80> 10
Bryant Peake. ....c sec ceeeweeeeeancdanee aire 1 MéXiCali. 2 .ccsanccesstoneuoasasccacaeeetenced 10, 11
Cerro Junipero Serra... .......22-2-2--2-0--- 12 MiUleg6 secs ainda Gus asieaauedge ce iciescens 10
CaO KANG saacnesncae cad deemciawebinineeyes 12 PHOVId GDCIS sje osecccenceceebuaseuauiesweawiens 10
Cerro Salvatierra,.........-.-..2------------ 2 PUreAtorl ivi vsieintetelateiced atcha aecidaniane easiest 10
COrTO UU BAtt6: conc dgccusan sie tors sagitiseaee | San Tonacio yw visse x2 eretewnesex lasses eemen 10
Goldiian Peale. joe sta cee.cuee seis ee eeleelewisis ae 12 San Ignacio Valley........------.-----e-eeeeee 77
Llano de San Bruno. .............------+---- 12 San Jose del Cabo...........---- ‘sanarueeeees 10
Llano de San Quintin. ........-......------- 12 San Quintin scoot oc ei eee see RoR 10
. Magdalena Plain s 12 Sarita Rosalia. )< ssc: se2ce ieaeecaecaneesecsed 10, 34
Vizcaino Desert. ...........-----2--2-5- ecu, 42 DUR Na: go. 6:2iedauinncinteccennenasenet eae ‘i... 10
Plain, Alamosecc2sccicc..cossecees eed ideewsucse 75 TOdOS BANGS... ain cn:cecrncecgiece nicieisicieis ope: wsinw seine bie 10
a PA) oo ssecsedcarseoree deeds anetemenewas 73 PYVOHGs eusaaedes ewaceetnncesemmsauseaneees 10
LOTCtO itisgeiciectmneminate Ho Sess teetee me niseeee 73 | Populus macdougalii.............-2.----------+++ 106
Magdalena BN Ro se etd 9 eeepc nae 72 TUONGICOl Ase oxcccscceccecciteasceaeeees 105, 129, 155
SebTn AEN ORT soo 2c ciscey debs Serandie aed weshclarsisaes wale 71 tremuloides.........-....--------2----02-++ 104, 132
Santo Domingo...........--.------2----200-- 30 tHICHG CALA 2.2. cccew cee gains ciecisirscieses 131
Ventana Bay.........-.-..2.-- 2.2 eee ee ee eee 73 | Porophyllum confertum..............222222-2+++++ 123
Plariestieus ..5..2.60.02c-csceagniaioenseeneiied nee ones 116 126
confinis. .......-.----.0. 2 ee eee eee eee 121, 180, 167 | Porpoise...........---000eceeeesceeceeebenteseeee 111
PUAH ES. Sococie npsncecccinnece ys eceyanierealaineietrere 103, 132 | Porzana............00.22-202 cece eee eee eens bs teeak 116
Plant H6te 05 32 ).,.2¥25:c1c1c-cscssererawvormdviatciare 118, 125, 126,129 | Poso Altamirano.............0.-.200eceeeeeeeeeeee 31, 32
Plants, derivation. .......-.-..-----+--++-+-++-+- 103 Grande, work at...........22.0.202ee esse eee 39
Arid Tropical Zone.........-..-.---20--0-0-05 122 Cian ecceancte esdes sods caenvere vee menaees 20
Canadian Zone.........- Pindeaaamemaaakee aetna 132 Medano Blanco Ranch..............-------45- 39
’ Lower Sonoran Zone.......-..---- severe sen 125, 126 Salad ovcescces seasionsieseesceh eee emer ewes 19
Transition Zone. ..........--.-----2--2- 220s 131 San Augustine cescsecscseedsetdastecewes een 25
Upper Sonoran Zone............--2-2-2+2---- 129) | ‘Potatoeds-:cossewewesiee saa svcd ee coneu deen esc 4 139
Platanus racemosa... .--.-- 2-2-0220 eee eee eee ee 129 | Potrero, Arroyo......-..-.-------0--2+++ 35, 36, 61 (desc.)
Plateaus, desc........-.---. 2-02 s cece eee eee ees 49 | Pratt, O. A., work by.......-...-2--.--22222-2--- 12
Playa Maria Bay.........--.-.-------02- 02-2 e eee 11, 29 | Precipitation..........2..2220022 2022 eee cece eee eeee 97
PROPOGIB 5.20 nici acer ennecscicsicceneaseecaaimese 116 | Price, W. W., work by.........-...-....-- cece 145
Plestiodon skiltonianus lagunensis..........-.... 114,115 | Prieta Range.............2.0.22200- eee eeeeeeeeee 29
Pluchea sericea.......-.-....-2--2------ 48, 107, 118, 126 erro sc. ocs Seasecanas dae hatialcacenwunns 69
Plumiera mexicana... ....-...-.-...--2--2--2--- 123 | Pringle, C. G., work by...........2..00.220-eeeee 142
PHVB) IS i ajesnissctscencicreieie Aas ceishciayrrs manos Dawe ots TG!)  PROCY OM ses 2ecieacindineciedernnaasodesoeeueceneeseels 117
Pocket gopher.......-..--.-------+-+-++++++ s-2-. 14, 87,90 PROTA. nec cetienchsewaneautwenctegienes 121, 128, 130
Podasocys (see Eupoda, 116). PANIGUS esiiictaivececadakinelwasentaneser 128
Poi yMbusq o:a:e <2: denie sn one oe sin die secineranesbe e LG e | SPROBNE seid Sahat secorene ne aianclelard acrgra te haherlments 116
Point, San Felipe. «...0ic-seteceeee see mecesneeuces 18 subis hesperia..............22.2-220.000208 131, 154
San Fermin: ecccs 22 s+ oc dvesssadeoceenses Bee 70 | Pronghorn antelope (see also Antelope)............. 110
Polioptilasveccssseveantxvecwsssesveerds coaseews de 116 | Prosopis glandulosa.............2...2222220. 19, 118, 125
Califormicasscccewiecccekssesetencresesaatmes 130 juliflora........... sete dacalvlnoommbers se sce aicice 107
plumbea plumbea...........--.22...---2... 120, 126 ODOPALE 5c oc cteiiacuecisie cise nemaeeeee cause 25
Polyborus.....--...---- 2-0-2022 ee eee e eee e eters 115 PAIMEN1, ose vers ex awareseseeaens sce 128, 125, 126
'' cheriway auduboni..................--- 120, 128, 124 PUbCSCONG sis cecsilsiow wenwcwusimesser sea bewecalen 107
TWtOS U6 iis 2c etndeiecaiseaiocanterietinaaias 95 | Prunus ilicifolia. .......62 20.0000 000 20.0 ele eee 129
Polygala apopetala.................-.2--2--005-- 155 | Psaltriparus............2-2 2.0.22. ee cee eee eee 116
Pomegranates......-....---.--++ 33, 35, 38, 48, 138, 139, 140 PTIN SOs ses cee Ses x.cie sae. nie ded eo make dew ane 151
Pond, Charles F., work by..........-.----.2++-+- 143 minimus grindae.................22222202.. 121, 130
Podecetesincunu ait noose eee ehlaterers raeieiatele 116 DMADUIA US seis 2595 5:0 oscv5; 5 bcarecasacaarevacesdiaidec oiorweniee 130
Population, Calmalli.................--22--2--22-- 10 | Pseudemys ornata nebulosa.................... . 114,105
Comondu. .......-. 2.22 cece eee ee eee cece eee 10 | Ptychoramphus.........-........ 000 00e cece eee ee 116
Comondu Valley Puffinus (see also Ardenna, Thyellodroma).......... 116
BU Alamo ss3csocecccccnataaen as ewnsceaneosien auricularis................ 000000. e cece eee 160
El Rosari6 vic deciee catvenoiedvencc es aeed Beweccenteee CUNCADUB. =. coc nccciienas tens aaeae sees cee ee 160
Ensenada.......2.---.---+.--: Pulque agave..........--..2- 2 eee eee eee eee ee 140
THA a cats a ene ee eel, Punta: Arena... sccinsgevecsy 22a ¢ Paibaescuni ei shana imioracens 51
Va) Pade. sc Ssccscxen eee eee eaenee ceetess ee Prieta Ranch..............-.06c-eceee ee ee eee 29
Tea? PHAR oe tcc heh bi ee tanto Purisima Vieja............. 0.2... eee eee 37
 

 

 

INDEX. 189

“ Page. Ranches—Continued. Page:
Purpus, C. A., work by............0...2220-020005 145 Poso Grande.........-2-ceeeceeeeceeeeereeees 39
Pyrocephalus: 0. -22c:ceceee ses csuseessi ees cease 15 Poso Medano Blanco.......-...-++ sadapaagiecia 39
rubinus mexicanus............2.2::-2-+---+ 120, 124 Punta Prieta...........2.22.- eee eee eee sees 29
Pyrrhuloxia...........2.-. 2.062200 aa 36, 112, 116 Rodriguez...... PN ie tena eae eater nea 42
~  ginuata peninsulae...............-- 120, 123, 124, 153 SACCAtON scud $i. haosciu ates cAdseeawead seen 44
Qilall cs Awcwscecehencsn We Seeaesen See hewaielaass 111 San Andiesic.: oo cccc cede vec adegescwwas ee 28, 29
California. . ..oj.-cis 220-5 geeteeeiedessetete es 22, 85 San Angeles. Lives weal ble eegekesihls woo 34
Gamibel Sa. snesntenes s vaxngarsdes xes.e scene 134 San Antonio, work at..............2..20.0000 22
MOUNtAIN. 2. 2. josie sea ee oe wes Sete 21, 134 San Bernardo.......-.--------+2eeeeeeeeeeeee 46
Quamoclidion triflorum............+-.-----+++- 125, 126 Sait FaCiNtOvascece scccsscducieeieaee ears cose 44
QuerCUs nn 2nccdectecee nde cece eens ack ieseees 108 Sati Tose vs.n.gsbeaenucths ut satnecinasgecuans 36
SPTUOMA citi see se cetehactse.<% 48, 75, 105, 129 San Joselito was cejienee ss cotieccecneee see seine 36
brandegei.........----..----------- 105, 125, 126, 168 San Marcos.......-..--+-++-- Soeieticaaieswdee:. OO!
chrysolepis..........--.--------0+---++-+;--' 104, 181 San Pedros2-saccerenenae sess eccsecee saacanes 43
devia. .....-.. bal ssdidaiinecesinne hae 129, 168 San Vicentessicc aesnsiees «o's aware suisieleie se aieneres 38
GUTOBA sd chicas hd apnecioed Siee So deeRaeekews 104, 129 Santa Ande. eee Soot eeenes ote 32
idonea....- Duidsoune sane ass -- 125, 126, 168 Sata ANI ta atecwcsccsay.coeccacdeaseme aes sce ess 45. 73
palteri.c oc dccccn ode pice el Binet denncen 104, 129 Santa LUG ccccccccaws cas oavosecsecineemeeciels 34
tomentellavesceveisccseceenvicorenecckurss 105, 129 Santa Maria......2......-00.-2 eee ee eee eee 23, 24,71
WISLIZeNI ioe 020 vevstes sex gadsosseteediens se enes 75 Santo Domingo.............-- Raines. alee 39
Gerustule a JgneateCasiawSeeeceG vi ere 116 Santo Tomas, work at..........---.2-2-+--0-- 22
1389, 140 Saiz al so ase see ein eee 48

111 |. Servatillo........0.220 22. 2c eee eee eee eee eee 41

at allavoni erie A DRA ae Saree mee ema aeer  oLoo Soledad wc cacccendnccacanewiiee ieee 2a sseacsess 39, 73

spied saab eeiahianriedade sas Seas 92| °° Tres Pachitas.............-02.2--ceeeeceeeeeee = 4

87, 110, 133 WalleCitog sas. <i Zseis'sie sociceew gene vee dodece sat 20

28, 90, 92, 110, 133 Valle: Ploio: 2sescbhabansaeetee ase Sanateweis 43

seieausdiawgbasesca se pee deseeeeeee 133 Vil€jOovsncccacaslesssuekestesevevcsedsueseccs 16

igstewette ras tenseeeeeesssceadeseeceedes 3D } Ranper(boat)...cc-secc aces dee ses saeeledenes 153

dougie cawesee Melee ewee es godenlate seeks 97 | Rat, kangaroo.........--.----------+-- 18, 91, 92, 110, 111

Rainbow trout..............------------ amanda Says 22 TICOs een a ceeeen aac t dhmulaee Reewe werner sitet 117

Rainfall, records...........--.---------------+--- 100 WOO): carsaiaciaw Sc Mew aeeeeet 85, 86, 87, 91, 92, 110

RAISINS ee Sosa cddancrcomints ceaiicaciasn aa qamareneneieelse 37,139 | Rattlesnake.............2.20 000.0. e eee eee ee eee ee 93

Pea U8 wages cot ara'sis gieigsisincaeiie ns oreeerseaeediedn 116 Aid O-WANd Or ss 3 4% wesewed sy 9s dae eran sve’ 27
heldingi.......-.------------------- +++ 120, 124,151 | Raven........2...2--.-222-2-2 cece e eee eee 83, 90, 91, 92

* levipes......2<+-+02.- berueeswvetewseeces: “AIO Real del Castillo Peso Ue EES Deon senaeeee 15, 75

Ramona incana pachystachys. pesedieekeemeweseedes 131 | Recurvirostra..........22020..0020ce cece eee eens 116
polystachya...-.-..---------+---------2 eee eee 129 | Red-billed tropic bird.................0..-.--000- 112

Ramos, Joaquin.....--.--------0--- eee ee eee eee ee 8 | Regulus (see Corthylio, 116).

Ranches: Reithrodontomys................--0.ceeeeeeeenees 117
Agua Colorado.......--.--+----++-+-+-+eeeeee 41 megalotis longicaudus.....---.-...--2-e2eee0-- 130
Agua Grande....-.--.--------2-02teceeeeee -- 36 megalotis.......2.....0200.02 le eeeeeceeee 127

_ Berrendo............------ eieeceselneceeteies = = SL peninsulae........ 2.22.22. eee eee eee e eens 130

| Cajoncito.........-.--------- casezesegestes Aa peninsulae...... 2.222.222 cee cence e ee eeee 163

» Campbell’s......--------------e-e- eee 16, a 20,76 | Reptile lists...................22.-- 118, 114, 115, 126, 130

‘\ Canyon de Cancio........-.--- atone eee eee eens 48 | Reptiles, Arid Tropical Zone..........-.------++- 123

‘El Batequil.......-..--- 2-20-22 eee eee eee eee ee 39 ’ Lower Sonoran Zone.....-.....--------eeeee0 126

., El Potrero......-.. sips aieain a Seer pe eS eee Stisas 36 Transition Zone. .......-.-.-.2.-0--.0eeeeeee 131
El Rayo.....-.---------2- 222222 wien ence eens 15, 55 Upper Sonoran Zone..............2.---.200 129
El Sauz.....-------2 22 eee ee eee ence eee eee eee Rhamnus californiea..............-2222.--0-0000- 131
El Taraiso CLOCCR i sts aciaabes ious 2 ech eRee els sb nay 129
El Zapote....-.------------5+- Rhinocheilus lecontei............-....2.20.2+2005 114, 115
Guajademi Rhizophora mangle................-...- es aethaees 109, 123
Hanson. ...------ 2-22 ee cee cece cee ence enone Rhoads, S. N., work by...............20.202-02000- 146
Hondo.....---- 2-2-2222 eee eee cece ee eee ee Rhus veatchiana.. ihe Santee are Sina reee tate eee 141, 149
La Crug.....- 202. - eee nee e nes Rbynchod Ofte. sasecciis pos uslnauetg eae vv se paras 115
La Cueva 3 Rabes ss ceciccies been dene sate creeds ts sees 105
Laguna Hanson. ..--.i.--:-----2---5 ec eee denne 55 Drandepel av siec ake cous eleds oie ses 110, 129
Las Delicias......-..---------eseee eee eee isles 35 indecorum............. 00020020. ee eee eee ee eens 129
Los Candelitos...........--222-2202- eee eeee DOI PACES a asia te bavataa cen te eel na oro aen aoe 137
Matancht@ ce sceescscccee peices oscnd Stes 89, 40,73 | Richmondena.................-.2-2222-e- eee 112, 116
Pierce’s, work at.........2.060--0 202 ee eee ee 142 cardinalis ignea............2..-.-.--- 34, 120, 123, 124
 

   
  

 

 

 

190 INDEX.
Page. i Page.
Riparian aonb An ee Ee eas - 116 | San Bruno...............-.. noon eee 35, 73 (desc.)
RRISSA i oc ccmacrciarsercc eames Seaigemeneaien euetAaes 116 | San Carlos....... sh Glog eciteteta nated cy tad haan nAB ate 1
tridactyla pollicaris.................-2.... vee HGON i) Mesa ttcdinespbeemtee cease ice. 50, 54, 67 ee, )
Rivers: ~ ; San Diegan Faunal District...............-------» 118
Coloradostt.siiitiavccs ahve yew eoeeaamasety 10,79 (desc.) | San Dominguito..................--- ere miami 229)
ard Yio cie.s src onstatioiee tia aie eke es oy cuales 13,69 | San Esteban........... scacbiedavt deatesadechs exile See BEERS 33
San Eugenio, Cape.........-...-...-------eedeees 52
Sant, PGW pe. .oscesagacnccechuswessansnneeea eres 19, 51
BAY cacti tae armenian eons Shae 17, 18,19, 51, 69, 70
MOUT tain 5.)-isiatasfelsdosixrox w actaclsemeneettic aver 19
f ROLD tis sechage Actors pesn ere pevctiverarnedtein aa ueiaommate siete . 18
WE Cs cra rcin iaiaestseapatcicirs ayayne esate eee 18, 69, 75. Pure )
San Hermin, Cape.............-----------0- ee eee 68
Santo Domingo...............-------- Seaatinn —2i p POM aeepeasgseceis eo lerate etere ere ap eees teehee 70
TjUaNG 0. ccs eeee ses fecdteisiais cisarscewsecean 48 | San Pemandoys sc sn saqssens 9, 24 (desc. 2 25 (work
Roadrunner............-. exseusevaeecerieanegans 112 MUSSION &.icicsinciciacaseminyscias + ciuasieemeaesseeone
RoaBiscncdeededsecccecieasev ses socskemacseeee 11 RIVEF se scee ssc Hares atin RESETS ey a
Roccella........-..-.- eiddicce nies s 30, 40, 108, 123, 135 | San Francisco de Barja Miswinw..,..sssensiczeans 77
Rodriguez ranch 3 42 § DManO sooo 5.25 ecacscce base naceacceiwess eee ee 26
Rosa californica. Xavier Mission. . idence BER ae SERIE 7
San Francisquito idining camp Sctnsceeanek aoce capes 58
WASH so :ctonigrccint lamacenematamaatecseresinisne incest 26
\ (naka eucamaaens San Geronimo Island.....-..-.--- 23 (work at), 86 (ile: )
Rosarito Valley.......... iidaaheheae genoa ensans 80: |) Sad, HilarOs.s sececccncsceacacieiiacerrarnine oie ae mck seine 42
Rose, J. N., work by...... sapeahgemsauadetncue Bip l4O: | San DERECOr ceninceseeecd dans camels “10, 33 (desc.)
Royal Highway........ Safe serena eseg otersishorarrsja piaieerses 33  TTOY O wii nicisue siesta anoles eieeaaeniers 34
RUD BS oi.pccts cictcie aceyareyoictorctersiniom steieje 129 LAGOON Faisistsietssctetatarcharacersrareignaicigis Jtpegeenaunres 77
Ruelligs sada ceumieeavna-oectes 123 MABBLON So ioco J acetess cl ererotdives ane pute anata 33
RUIN CanG seen viciecrasalecuelenteisuiantatetlad tcemaiatos 140 Valle yi(dese: yoo ce rcaticrstianeramaiuiceiceeiesieies 4 77
Rumfordia. .....).. 2200.0 c cece cece cece eee eeeeees 157 | San Ildefonso Island..........-2.-.2-422..2------ 84
connata......-.-..- Mece ccd eeeseaoskeesanGes 157 | San Jacinto Ranch............2....2-22---eee eee) | 44
Russian colonists............-.. pis ra era dea as 9 | San Javier Camyon...........22--2222e008)u. 1 S. 30
Saccaton ranch.........---.222.202--02022 eee eeeee Ad Washi usin aausa dame eeduecameneraesas. | 30)
Sage; black. 2sscesessossccccwsreeee yy see eceaurse 104.) San S080 seiscec se caewenerseeysceesdeseon ase evs oes 9
Salada.........-.- deemed daca eea beets es 9 AYTOY 0! arscevsesivemeemeesisesixcaweness 36 (desc.), 61
Lapuniti.... .-.cecciecsce ae vs - 69, 74 (dean, ), 79 Can YON ees wes seivinciaivasaubeeenicesmasesseeds ws 87
Salix lasiolopis oa. ss.ciodinccshesennccdccceneteed 129 del, CABO: sisic--inieneniniee ee wate 10, 11, 45. oa ), 73
Salmo nelsoni.....-...---------2---+--+eeeeee 22, 80, 134 Tesla secrereseie eesin ce nyois ash si evkseicssusicidia ace a latecduese 08 ofan 92
Salpinctes................-.-- Lis Usone paces 112,116 Rane vciecick vse sient edece PT ee ee 36
Salts. ccsccincaccomasasctcns iiditeh diene Dawscaumodmmance 8; 52 RIVOPY on ciicntesconcsenugne ee eee 73
Salto de los Reyes......-.-----enteseceeceeeeeeee 42 Valley: <c.i<s2te-0sccassesaasmeniad was 45, 73 (dese.)
Salvadora hexalepis..........------------+--+-+- ., 114,115 | San Joselito Ranch........22. 2.22 ee eee ee ee eee
Salvatierra, Padre.......-.....-----+-+ pe escizpie isthe 8 | San Juan de Dios, Cerro...:...............2...22.
Salvia californica.........-...-2---22-+20005 asldiovete 126 | San Julian, Desert... -. sooaw eon daekteigaeeaaes
SIMI «:.-n:c(hydegscatenevecscsanuecdwanceie dec. NADA Sai TAMG o1o8 «a os. cra cierryarneniecose on caine weinanes
Sambucus scaswcsareseensntecotescesiateenieen 75, 104 | San Lazaro Mountains............2..0....... t
SIAM CHies a xineamerrndaetindde Peigecieca cee emia 129 POA cess rcinsecrs arsrenae age Ow sie eer ene aps
San Andres, mining camp............-.-2+--+--+- 68 | San Lucas.........------. sears
Ranch nx ica telaeianee aw petals ecdenctec evaetsterne 28,29 | San Luis Gonzales Bay..-.....-......2..004 Si
Valles. cisciwiee icies kel eerepeebene stateteea ask 28 MISSION iar oss ssrsramussiia anges totam eeee ieee
San PAtiPel sjjis.dciae sceaadeneneom acetn ear Rece ee eees 72 | San Marcos Ranch.........---..-..22--.eeeee eee
Ranchi saiscviaaevervedicciae cine oslo busemieulse 34 | San Martin Island, desc..............-..... ‘aeeees 23, 86
San Antonio, Arroyo...-....-2----e2.eeee eee ee ene 34,46 | San Matias Pass...............2---2. 022002 . 17,18, 76
Oreck: gicsacveceame vas sancanige Pieiaddhe Gace 22 SPIN G sicperscceacioeteniieiatinceseekentanuna tons 17,19
Ranch, work at...........---2----2005 Peace 22 | San Pablo. ........-.-.-0 2-2. e eee eee eee eet 32
San Augustin, Llano. ...-.....-..-------- 25, 76 (desc.) CANYON oic.a ascaiersasnes ae daa aware saesaamcaleeis 32
Well sci cies acusawine wiweeiabcee itt ruieleitaeteis 76 MISSIOD oo:sre n.d 6 abe ie iaiboaneneucac onan’ 32
San Bartolo Peak.....-...--.-.2-.+.2-0+-e00 .--- 64 | San Pedro Martir Faunal District................. 117
San Benito Islands.........-.------0--eee eee econ: 88 Mislan Giag dscveocisceretechd So geire evel etcetera dee 84
San Bernardo, Arroyo....--.---2-.22+-02-eeeeeeees 46 Mission... .---.-.--+- usinaewenea te 7, 22 (desc.), 57
Ranh cindcsiussswaarade cabidiucemineae wane 46 Moutains. «2a coscsieeccaeecetaes ees a... 17,18,
San Borja Mission.........--------+++ Be dibokscee 7,28 19, 20 (work at), 25, 49, 54, 56 (desc.)
INDEX. 191
Page. Page.

San Pedro Ranch. .......-.---0--.--2-- eee eee 43 | Sauromalus.... 2.2... cece cece ee cee eceees 93, 114, 115

San Quintin. ................-2-. 5-5 9, 10, 11, 23 (desc.) LOR Se A setae tush tel pened a ache i Ae ae te han 114
BAY scwcidcnnee ssiqerieacisicisise oe 23 (old colony), 71 ISPIGUS: 2h ei so ed eemcnconsaZamiee soak 84, 114, 126
Pai sie jsseiceiveet eae tated eeereseen eens 71 interbrachialis...................---- 114,115,171

San: Raiaels <j. secceseew ees cceeesciceeies os esses 31 townsendi..... 2.22.2... ee eee eee ee ee ee 171
ATTOYO: %.ceccnexcccaeeps tsk myeecbereexsedccs 20, 75 WAT es octet ey aa kal sheds hdd Scab 171
CBTY ON: o:c cveis avcidiceie sess BER eke Sapciciek 96 | SAUGALO s o-o.sce nck ees mesawds exe aus owseemseces 9
Wall@yiscccceseswieciswinn 50 15 (desc.), 16, 75 (desc.) | Sauzal Canyon. ..........2.0. 00-0 2e cece cece cess 48

San Ramon, Estero, 77....-.---...-.-----------6- 77 TRAN Chissee Beeisjemicis sicerscice cian Oe auamnaesateite 48

San Roque Island ..........------2-- eee eee ee eee 89. | BaVOrnis ris. -e cies 2h 2 no neaehewsex eas muennes 115

San Simon Canyon. ...-.-.---.---+-+-- Bah ety bye 56 | Scaeophaethon.................20--0--202 eee eeeee 116
Rivers vee skies Saline eesidness tesecas eis 23, 24,71 | Scammon Lagoon..............-.-.----20222 eee eee 32,71

Sain’ Telok ecicew sae elas icceendess 23 (work at), 70 | Scapanus...........-2-2 2220220222 eee eee eee eee 117
Can yOt wise cs calediccasrasasswase's veined 56 latimanus anthonyi...........-.------------+- 132

San Vicente ranch..........2-.--2-0-e- eee eee eee 38: | Seardafellay bo uses ae acme eticegne Sad eedeimirtle 115

Sandstone... 2-22. .e cece eeee eee e ee geen eens 89 | Sceloporus.......... 22-0222 222.222 eee eee eee eee 114

Sangre de Cristo... .....2.-----0--- 202s eee ae 16 graciosus vandenburgianus........-..------- 114, 181

Santa Agueda mine............----2-02- eee ee eee 34 M2 S23 sweet laislare'sl a eaves heads we a ele 114, 115

Santa Ana, Llano............---------- 28, 29, 76 (desc. ) linea tlisiesccaenecc sss caacasecees sees 114, 115, 171
Rane x. occas eecacenataeeteddeccs saan 32 occidentalis bi-seriatus...............--.--- 114, 130

Santa Anita Ranch.......-...------------------ 45, 73 OFCUUtl: ieee etea.5 5 tanner eeea Ded bow wled 114, 130

Santa Catalina Peak...........-.--------------- 56 ZOBLETOMUS ssc2e cancun eae Sheivamaw enti slekise 5 114, 115

Santa Catarina... ...-..e.eec cece eee ee ee eee ene 9 | Schmaltzia ribifolia.............2..0..0200.002000- 129
Landing’ oshangsuetes soos cea seduaedeeioecs 11, 25, 76 | Schoepfia californica.....-.......2.2-220-2-22205+ 123,
Missionte..o% coszcnecngnosnndsiane sexta sss 16:55, 75.) SCwUPUs ws ceodeiaeBaekiicone sacar cinco see sess 117

Santa: Clara: oco-dc.ceecleagcess te eee eeetuss ose es 9 douglasii mearnsi................2.-2-2-200-0- 132
Mountains. .. 2.2.0... e eee eee eee ence eee 34 | Scolecophagus carolinus..................-2------- 154
MMA jaacke Sond tokesweese eeenwenuhewcce nts 34 | Sea elephant................ 23, 85, 86, 87, 88, 95, 146, 164

Santa Eulalia Valley.............2222-2-20--00- 57 lion..................------- 23, 85, 86, 88, 95, 188, 134

Santa Gertrudis Mission............2.222.-2...0+- 7, 50 GUAR sec ava dion white etesctdses 86, 87, 88, 95, 111, 133

Santa Lucia Mountains. ....... receane sgt 34 (desc.), 35 turtles: ssnscssavieen ss sa asexeed 40, 90, 113, 114, 185
Pé@ak 2 oi4 teceu vesetieitiwedies eee ceeeuode 60 | Sea of Cortez........2..22. 200. cece eee ee eee eee 48

Santa Margarita Island..........--.--------0-205- 90 | Seal, elephant..............2-.-22.0.2.22255 111, 133, 169

Santa Maria, Arroyo. .............--2------.-+--- 58 MUP. ok toks yeaa ecaesseenes 86, 87, 88, 111, 133, 134
WlanG sc see rngeare sens Gene sows 26, 76 (desc.) Guadalupe..............-.--- 88, 95, 111, 164, 169
MESSI 2 ds--/os,bheeicisio assured e's eadesds 7,23, 24, 71 harbor.........-22--222eeeeeeeeeee 85, 86, 88, 95, 133
Rane +255 saseea so ssoseseewte ss + see cee 23, 24, 71 leopard waves vac eoivecuer ee yey ss saei ene se 23

e BANTA ARORA. c:.5-siciceisis cada sarin Soir v ge emareeeeees © 22 Spotted ss cccsccssycuckecdawies 25s cade deiceused 2 133
Basil scans sc sob aatenianinsmaicanmanec ss Gunes 57 | Seaports, discussion..................-....22----- 11
GSTY ON nieiececiean vs teieecaihe sodennaeece 22,56 | Sebastiana bilocularis.....................-..-.-- 123

Sanita Rosalia 2: 1. s ciis's oes aeioswinae vciesgene sce TO | SONS so cse cece oo seee vee eecansececiece snide 116
Baysierasstcieietenness shueseeunss Baal 17529, 70 \ Selasphonises. .ios4ieeu ees ssceeeouukine sicieceaseieass 115
mining camp......-..--22-22-02-22-ee eee eee 8 | Servatillo ranch.............-2--- 2-2 -22--eeee eee 41
population... cc scincdsc seme siece.steeuemtes 10; 34 | Setochaleis 2c 2202. cec2seseseecke sess seciesie ccc « 115
Walle yoo. a60s ctewoecteue acdametteson somes 22 | Setophaga.......-2-- 2-22-22 eee eee eee cece eee eens 116

Santiago village. ..........20--2-0-eeeeeeeeeeeeee dg} CORP WATER ccscannunnaaienvesalatn arc kaslieien 23, 86, 95, 112

Santo Domingoeis3<sscxieide's <vsiaciorske ss heer 23,30 | Sheep......... Gases anmeans id stad. eects tase 140
Cahyon.. ois s deer chmeceenssere actset e's sone 56 mountaiD . .....---------e.eeeee ee eee 22, 27, 36, 110, 182
EagO0N 3.0 oss nace sega s shee ene eet 30 | Shrike, Gambel’s.............------.-.---..0-000- 92
Landing. ........0--22-0- ee eee erence rece 11, 30, 71 | Siagonodon humilis....-..--..---......222.22.. 114, 115
Miesiots. 0c 5 Harceddccke mess neciucene cov acne 23 | Sialia...2ccjesae cokeeseee ys saaieee ss acecae anaes 116
Plain he es ecigeieiieiss ston Seeee Somes 30 mexicana anabelae....-..2. 22.0.2. e eee eee 131
RANCH ooose ised os Sa RE Se ea eEERe EE ors 39 | Sierra (see Mountains).

GaitO TOMS so /a:coeoaeksee ds veeseieeeesesesees | 14) SIBMOUON:.c.s.occ sce cnet an gasi eevee doseteeedee. 117
REVO: o ciecisin'e 'v- athietbcinn es HeSae Saige vs PaRGHS 75 hispidus eremicus........-..-........2002000- 127
Valley... .---eeeceeeeecece eens ence ceees -... 657 | Signal Mountain...................200.0.2200 ce eee 68

Sardines... oc-snhecscasvecedesesaeeaadewtee ee 195.4] SUVer caus aessetic ce atsade$ Ceca ned ancien 8

QptOF cc. ose cuca se SemeSE RCS Seo REE ERM Rate: 114,171 | Simmondsia..............2 00.0000 e eee eee ence 107
ANQUSEUS . ..--- eee eee eee rece eee eeee 114, 115, 171 Caliiormics ci.iseteGiliend odamed eden? ¢eeakets 126
grandaevus. .--------0e- essere eee TIA I15; 174 SUOM YB ae ecteeias caecee ac = aca dminwiss taken gaceee 158

158961°—21——13

 
192 INDEX.
Page. Page.
Bitlet het aah deat re etter ea eae. 116 | Sugar cane.......-.-. 33, 35, 36, 37, 38, 44, 45, 137, 139, 140
CANAMONGIS.2 2 2cc52 25222 ss ce eeeee secs eeees O5C Bula. ccs esis o haenetedennenie a waetabuelel's cise 116
carolinensis aculeata...............2-2-2--+--- W310) Sulphurs: sco s2escacgece cee eee edesdecek eeee ses 8
lagiinge 5 ccacheaceiciacieuisinntas 121, 130, 156 | Sulphur (boat) soil Sravaccraas tues tak hin ae auc eae ds . 141
pygmaea leuconucha..........-...-.-------- "481 Sweet limes...........-..----2-: 20 eece eee eee e eee » 140
Skunk, spotted........-..2..2..2222--2--2--- 33, 35, 110 | Sycamore........-----+02----02ce eee eee eee ee eee 104
Slevin, J. R., work by......-.-.2----22---2---00-- 144 | Sylvilagus...........-.. 2222220222 e eee eee eee eee 117
Stak Goose ecek norco teatro at eat eeudecmnin 88 arizonae arizonae,,,.-----------2-++++- ecaees 127
POPN Ob vies cpeeeie ce sicts vg stance ane sa ssclestis 86 auduboni confinis.............--.--.e0+0+- 121, 127
SnOWee eit heel atc cea etter a Aha dd 96, 97, 99 sanctidiegi..........-...- ebieatihtn ceo 130
DOCOMO wees aekew seskees sche ee tee eeeeneeen ee 24 bachmani cerrosensis....-...-...-.------<++- 130
placer Minese. 000.2525. ce eek emcees 20 CINCMASCONS! 5 joyseseece set enedewhne exy's's 130
Soledad Ranch..................22---2-2-0- eee eee 39, 73 ORIGSUUB: op .ivicsccmnmnaadeseeeraeeaKee 127, 164
HOLE oes ie oe ns ess se caamacas bee ooneadereteeh ee 117 peninsularis............--------------+ 121, 127
californicus juncensis..........------------- 130,165 |  =mansuetus...........-.-------.--2---0 92, 127, 164
lapunde:.. 2h cassencnbnns aie neanw ass 121, 131, 165 | Symphoricarpos parishii...........----.-------+- 132
OVEINUS sc :ccsie:s 2 Aacsnamna nents atisemeuerdade de 163 | Syrnium (see Strix)........-.--..-- 2 eee eee eee ee 115
OMMAUS ss fossecne ccc secemaeete vals cnn 132 | Table Mountain................2..0-2-----0---0-+ 66
Sparrow, black-chinned. . sbivieGarccanccmmacsewat ee es 29. | Tach yCineta icc. ccc sci vacs cece tee te aed tees 116
black-throated.......-....-----2---.2-22--05-- 91 ' thalassina brachyptera.........-..----- 121, 130, 162
AOBEF i: kciote sre cy pte enenaces Seakeawawacie 112 : lepida esi cpcdcseneleehaaeaetaseaennt 131
San Benitos...secseisoe aot pak osetia hie see 88 Tagetes Ta COr ai saactaiiscukeegieewieceqenmucies ened anak 129
song. - 85: Tamarind s.i40 ate sadoseantvaeebeamien yeas Jassie 110
Spa tila. ccscvessiecicemsmectetbeetiwsines shies + vee es 116 | Tamarindus indica. ........-...22.22---.---0++ 110, 123
Speoty to 2225 scacvevesewt ie caceereeueeecss se dess D115: Vahbarkees. sald. fed ea ke hed teiade 136
Spermophilus grammurus atricapillus.......-...--- 154 | Tank (see also Spring, Tinaja, Well) .....-.. a weabelate 91, 93
Sphyrapicus. ena ccoscseeeeracecceddeseenseesess's 115 | Tantilla planiceps.............--.0.---2-0----245 114, 115
Spilogale easiest pleura uae eeiolea eee ee pee sexes 117 | Tapirira edulis. 0.06.0. s062+0+2eeseceewens 92, 123, 136
arizonae martirensis. Svduicndepeet ugar ees eee ss 192,168: |: Taxidéd...i.ccssic.e edeeewteieed esses sodueciees 117
lucasana......220 lee eee eee ee 121, 128, 156 taxus californica. .........--..---- cainkulGfarceBorshe 130
MICTOM ON a acco frecieinese setae eeu seeee ees 121, 128 TTIPUSCAR AS = Gers hie aaa essen Nere we ee cies 121, 128
BPs ao. 28h Sees ciate Secu aeieeeeer reads ees Se 116 | Teal, cinnamon. .............2--200.eeee eee e eee 21
PMU 5. waireaslanieucce: shee eee tes see ee 182 | Tecoma stans.............----0202 222 eee ee eee 102, 110
Pigeon dee eecieeeswamenneeenes sous ssce ee zioe 116 | Telmatodytes..................00000 20 2e cece eee ee 116
Spizella ss enncsseweteeteas eee en cd shea saceee 116 Temperature records........-----.-----++-+-+--++ 101
atrogularis........-...---..0-02 002s ee eee eee 29 Tepetate.. ie pewuns eel ewe' aaise ge eeeerae 41
Spring (see also Tank, Tinaja, Well)............-- 92,93 | Terebinthus,...........----------2-e-e2e eee eee 90
San Matias: vsjccnseceesdes ics eeenaencesos 17,19 | Tern, elegant Ueki terstoh vinta'e ware Setter en erp ata ak 112
Squash. o..:4eswegswocseceseGedeescaude stag tieecen 140 di imcheriie Bis eicttettchond oa siete tetera ne eee dt 152
Bquatavola cin evieisvioswcaeeeee eens enec sessed wane 116+) Phalasseus.sceecatuceis see ved secant neers 116
Squirrel, antelope. ..........----------2e0tee-00+ 110: | Thalictrum : cjeccnchecdut sss sev odseunoeneeets 105
TOG ceraersvidachiets sg: hee teat seers 87 poesia ER coh eon sede ono sterile 129
Stepanopus....ss0ccsceseeccewe ees ese gees ese oyoees 116 | Thamnophis ordinoides hammondii...........-- 114, 130
Steganosperma halimifolia..........-.---------.-- 123 | Thayer Museum..................2---.202- 2220205 146
Stelgidopteryx. .......----------2s2e eee eeee ee eee 116 | Thomomys...............----. iostubas celina 87,117
Stella Erland (boat)........-..2-2-2--2-22-2----5+ 161 APHPABLUSs ese sceeanaraanceecacss acaadeeecee 14
Stephens, Frank, work by......------------+----- 145 bottae alticola..................... eecen 121, 131
Stercorarius..........2-0--- 2-00 eee eee eee eee 116 AMLCAG ss oa53. coh aaadosatabsecamewatee. 121, 127 -
Sterna (sce also Hydroprogne, Thalasseus, Sternula). 116 MIPTICANS no. nnacianas- oes SauceweAien's 14, 130
elégans. .........---- +0222 22 eee ee eee eee ee “18? |. FUSBOOIUB 2 paseivxieaieisore tek iceisls ves 28 , 127, 165
Sterile! gentec sso ccencteewgean tates ceanne 116) |, magdalenae............ isha oats a Aone 127, 165
Stewart, Alban, work by..........------+++-+++++- 144 | Thrasher, desert..............------.22-2--22008. 30, 112
Stock raising. ..-...----- 2-2-0 eee eee eee eee eee 136, 140 | Three Peaks. .......---.--.+-2----2-- 0-22 e eee 66
Stopkton, A. L., work by......-----+-++++- eaeere TAD | PHM VOIR 56 2 saa cies ncewenuew ana chose 116
Storms,'historic. ...----.------+-22ee eee ee hestetes 96 bewickii charienturus. ....-................. 130
Stowell; J. C., work by....-.---+-++++2+00e e+e ee: 145) _ brevicaudus. .....-...---...-+2 22020. e eee 95
Streams and other surface waters. ............:... 78 Thryothorus. sewoMe ne hisltindee ReeNRERE A Ses ¥ae-<tislees 116
Streptoceryle........--------++++-eeee eee eee e eee 115 COITORNSIB...- .--- ++ e cece ee eee eee eee eee 160
Strix...:..---- ides o aians Gaipedepeaeele nwa sbecwates LG. | We pesllod roma <agisccnisaaesconnanteneoip hie se maaan’ 116
Stuinielld. ..csscsews ted ess ecscewiue peewee ecneae 116 | Tijuana... ,.,.-.20-- 22-2 eee eee eee ee eee ee enone 10, 48
neglecta neglecta.....---------+-++eee-e eee 130 RAVGES sone wisnashsonla be bultiines war ee tence 48

 
INDEX.

 

- Page.
Tillandsia. 2... ...0e-eeee eee eee eee ee ee eee 108
Tinaja (see also Spring, Tank, Well) ......-.------ 24, 93
de San Tian oqas weenie tals daneceins 33
de Santa Clara... .....2--.02e eee eee eee eee 84
Tinnunculus...........------2- 22 eee reece eee - 115
Todos Santos ........-----+-0seeee reer ee 10, 43 (desc.), 72
Bay... 22. ec cece eee cence ccc erent eceeerenece 13 |
boat......--- boise skcaeeeemuaeeseometeees 23
Islands (desc.). ..-.---------e-ee sere e teeters 14, 85 |
TOMALOCS: oocveelelccie Audiewrssaaneaeeeceseeey see tee 140°}:
Tor0tesnssacewe vies suneasc tcc cdcedcesee eee 208 136°
Totanlis....cecs ses ceewneecesesnancedee +5 tse 116
solitarius cinnamomeus......---------- beers 155
Tourmaline. ......22...2-22--2 222 ee eee eee eee eee 8
Townsend, ©. H., work by......--------+-+--+-0°+° 146
Townsend expedition, work by..-..-----------+--- 87
Toxostoma........-..---ese20e---e eee apace .-- 112,116
Vendir et cceecenicesninceevecwestiinesinnrasesces 126
cinerea Cinerea.........------ 22 eee eee ee eee 120, 126
MEANS]... 22s css oe saesneeeseersiewes 126, 156
crissale.....- 22-22-22 eee eeeereeee eee tees 130
lecontei oni OGIis.cS unica iehsuercatudesGee 30, 126
leConteliecssaceeiree eaiisisemeracamanenaeee ss 126
TOCLVAVAsuisentcoasecsoseceasstee cates eee ee eee 165
» DOW. eceeraccc seeee Rleldde latte ey ee 1380
Trails, E] Camino del Golfo. aad hace ects aeeG 14
El Camino de en Medio.........-.-----++++++- 14
El, Camino del Pacifico..........--.----+----+ 4
Transition Life Zone.......-.---. eee e eee eee ee » 131
i 181
132
181
131
43
Posos...--- Pe seteg nomi ae metorsemesae eAees 19, 69, 84
51
114, 115
116 |
16 (desc.), 18, 19, 76 (desc.)
10, 43
116
131
112
80, 134
22
135
123
PUTQUOIC. ..0..0c005 cee sa ce sleceaeedceciec tenes ds 8
TTursiops gillii................2022222ee eee eee eee es 166
TUUANU.... 2c cee eee eee ee eee eee ee ees 166
Turtle: seAicoceourswieds §digeiesassaer 40, 90, 113, 114, 135
tortoise shell............---------ecee eee e eer eee 114
Tyrannus. .......22-.002.ee cece eee cece etree 115
verticalis..........-..------eee eee e errr eeeee 130
sis asp arebus eves clic te & wnietauaries Sea tesees 3 ersierSilerereGleiaiel=t= 115
Uca crenulata.......--------eeee eee etree 19
Uparte, Padre.....--------2--2ce eee er etree 8
Peak sccccvctocishices v na asauiik 0s Raeeee sens 58
Upper Sonoran Life Zone....---------+2+++255200+ 128
birds......------ceeeeee ere r erent 130
mammals......------0--e ee cce sete eesesree 130

 

 

193
Upper Sonoran Life Zone—Continued. Page.
Plantes qsosecesd knees sss ate eeea ee aes aes 129
POPs oe esis bs cw ea agacsceueeneysaeuesios 129
UWrocyon cxcnniagiecCecods sazaebhaeldiisiad s cisiejels 35, 87, 117
cinereoargenteus californicus.........-..- 121, 128, 130
SCO til acrceg ios ae idee eeedeies onesie aes 128
Ursts soc ssisehttee vetoed e suddawe ents tenant 117
Utter vce cscee ses cegenienecey ess eee teeees 114
CONMCINNG:.2cce os anes d seco tes corns 114, 171
BTACIOHHe acs ccett-byeaiea tenia 25.0 0Geeeee 114, 126
mannophorus..........---.-----+----+- 114, 115,171
MAStiNensilss205 20's cael gasccReses Melrose coke 114
PITY ois ayes seats a ak Seed tats atch ints 114
114, 115, 159
Sedan nana ue oaete ck meatus 114, 115
Sai Nae Ue eaeeet eet Aas ss 114
114, 126, 171
Bere los LINN ae SACO S 2a slee 114, 115
SQUAMAta. 22. .0c0sesseeeeseseewss es eeve 114,115,171
stansburiana elegans... niseane skas seme eeteReS 114, 115
TheS POPS gers scleral x2 code te seid ss 114, 130
Stell atay gn. screeeecc meade sen ninseawldes ta 114
thalassiniai ca icui lee eas d swine Sgeew ecm tE’ 114, 115
Vachellia farnesiana.......-.----------------- 107, 122, 123
Vallecitos...........22-220eeeseeeeeees eraakeens 21
1 DORs os ce eaesesacee sts se Reebee eee ee este 57
RANCH. cansecswnacushiee soe os caamacaeeure segs 20
Wallleyiiccinicecitccetcnc uscd ea ngnneskedancs.ecs 48
Valle de San Felipe, desc..........-.----2+-+2-++++ 75
Valle Flojo ranch......-....----------+-- eee e eee 43
Vallesia, dichotoma.........-.------2+---222--2+5+ 123
Valleys: —
Calamahue.......-.-------+-+-+++ 26, 27, 74 (desc.)
,  COMONGU Ea: ccaeees totes oo lnc etiniesy yaesieiee 77
Smperial 03. s.geu eee sees meeesteaeiey eia's 10, 14, 75
Ta, Laguna 2 wi ceo os eee sees 46, 47, 64 (dese.)
La Purisima..-..-. 2... 2222-22 s cence eens cece 77
Matoml ssc arcsscduecesaseaebtbecee tones 57
RosaritOwsosse sviss's is eis sistea maa sue coos seman se 30
San Andres. cscce0ss ss castece cece ge roses 28
» ‘Gan Felipe..............---------+- ola 18, 69
Pan TO86: «<2 so aeascteeeeaets Seeaeaed oe 45, 73 (desc.)
San Rafael ......-..-.------- 15 (desc.), 16, 75 (desc.)
Santa Bulalia........2.---22222eeeee eee ee eee 57
Santa Rosin. ce sense scene nccecwseare cnssceweas 22
Santo Tomas..........---------se eee e eee eee 57
Trinidad......-...------- 16 (desc.), 18, 19, 76 (desc.)
Vall eCitos ja. ocnccie nse beicaw wines ass adinieieeciciere 48
Veatch, John A., work by.....---.-------------- 87, 141
Veatchia cedrosensis.....-.---..---.-+----2------- 151
GisCol Ore sn dcvesseeecersewsiee vow ee sie geiaieewiee 119
Ventana Bay.....-------------- 22-22 e eee ee ee eee 51
PIS 30.02 oc eat ee eee et ese Se ade se bewca sees 73
Werditis.vacines ssc ceed Se Gied ees aacee Reeds 112
CAPOA Aaa os esas ah naeen ede eemon emit - 92
Wermivorais ss so 'istncssotenel ee sis acicinecacuneeheele'a sere 116
celata lutescens. ......-....--...-.-22.40-56- 112
Verticaria caerulea ...-............2.-22--- 114, 115, 171
hyperythra beldingi. .........-.........+-+-- 114
YPC VENTA: oye eddewaadswescues 114, 115
BOVICCR suse 5s czsiee ebsites S's decisis se aie 114, 115
 

194 INDEX
Page. Page.
Vespertilio nitidus. ....----------+----++ +22 eee ee 160! | Whale; orays ccc cc.ecocnccccccnaandiacuesteeeseeic® 111, 133
PALIAUS s22.2..00a ecw remand mmmee ene eins 149 APOriil.. .icienwewieccasentne raceercbeacnlauaterene 111, 133
Vesperugo hesperus....-.--.----------+---------- GOK) WCAt acatccuctacenwecinuaininatsrcets 23, 24, 138, 139, 140
Vetolas.: .aasecacuas «ais eeieiesieutseggeeesgeesees L1G |. WilCOxIa... ..c.0c0 cece powcr eseeeceee eee ewns 90, 106
lapponica baueri. ...........-----22----e eee 112 SUTRA lefts forsncass yet rcharsrehetarepae eee a SEE 123
Victoria Mountains...........--.-.-- 48, 44,46 (work at) | Wildcat... .........222-0- 22202 e ee eee eee eee eee 110
Viguiera subincisa.....-.---.---. +22 eee eee renee 156 | Wilder, H. E., work by....-......------0-+-+-0++ 146
Village, Indian..............-.------------2e- 05+ 55 | Williams, F. X., work by....-..--------+----+++ 144
SSTAGO sso seneceecase ania nediebislea mediaset 46 | Willow.........-..--- M javejaicibisieinio eis URS AR CREE 105
Wises SAMOS: ast sarcicnaersiave ies kidd nn alec idceecesdinieieceie AZ: | WGIBONIA: «2-5 sccarcesceiniaenrenainmnd ane one see cee eeeees 116
WACO iceptsrncticeta ca meetas cease beadcemmei as TG) | WANs oc.c.cicieaveaieciiemoneadmroeinmanaenast tasccicetee 96
belli pusillvs...cccciciciscciseress sossseaadeaes 1380 | Windy Canyon..............------ eee ence eee eee 17
huttoni cognatus.......-.---------------+- P21 TBO: WANG 2 crcess aco ierereatreiatoanetstarsicnaysicjmnionw'S acne blelnictesate rand, 139
solitarius lucasanus.......-.----+---+---------- 156 | Wislizenia palmeri..........-.-.----------+- eee 125
WICINION: «...c.cwwarscasavecissctadededetes suey PO SWol facia. rocetetdcartecens see cceed ened sane 110, 132
Vireosylva.-..--.---- 2 cence elec eee e ence eee e eens 116 | Woodpecker, Cape.......-.-.---------e-22-2ee eee 92
Viscainoa geniculata....-.-.---------- eee eee eee 126 Gils asters tateictslstenerspacieia eres oewisieexeretirete 33, 112
Vizcaino Desert. ....---.------------ 12, 30, 32, 34, 50, 71 ladder-backed........-.--..-0----+--eeeeeeee 112
Faunal district...-...2.-.--------- pac nie biareieraiets 118 San 688i eee yeep crnse caters diccisancrecs 92
MOUNTBIIB  cicccseanincmiiieccnderaen cece tata 67 | Wren, Cactus. cccsccsunussessuessesceeckile'nee ces 90, 112
Warm ag ales oisiscenis scene snc eeeicemin minis: sieminsinenatis 84, 91, 93 CANYON seep sere samamenens wen ces ewan eines 92
Voleano Lake. .....-.-------------ee cere eee 13, 69, 79 fOCK 2s acxnsexescsserceesess 90, 91, 92, 95, 112
MU Meaissc ere neces cd pkemmmonerseeenecmsce 74 | Xanthocephalus............--.-----22-2- eee eee eee 116
Volcanoes de Las Tres Virgines, desc. ....-.------ 60 | Xanthoxylon.......):.0-.5.s.2.e0eeesssnevesewwewe 108
SG Pie ures koe ecio pao seuate eon 87, 117 APR Pir cctee tail ae reahsalva beans Sooke 122, 123
TARGTO Be aeges se reistete le oiontrcehoemnacrae anaes 19 | Xantus de Vesey, L. J. (John Xantus), work by... 45,
IL SVN iakseerhaatct ctratnc inser aclatorande nerd 121, 128 140, 141
ASTIN ow ee iors ststcies exh stel sic shane) ep otc 165 ae Pilberthe 25 weeeeeeniccwhiminsessicecine 114, -
MMACT ObIS apie wees tercesratarer taraisyzienas stocstea sree 128 OMB. .-- 20-2 eee ene eee eee eee ee ee eee reece
Aspeh eos edt bate tae teed eo 11 | Xylonagra arborea.......--...---.++-2s2-e2e eee eet 149
waar Sieh iess dn ticks vata etacegs eda 162 | Yaqui Indians...........-----+-2-+-2-2+2e ee bees 8
Wellow f6V Gl: sinie0s.cccwnccmascececcccacdceuubindee 12
Walnuts......-...-.0-2-0202ee ener eee e eee eecee 139, 140 Yrais, Li 41,72
Washingtonia.......-.2--0-0-eeeeeeees 106 10751005 168 eet nec ree neon seme nicicinie ts irsestincrintenite ’
Water, streams, and other surface 78 SMB Yisssasewantinie eeeteraann se scttamnpeNRRREN Tt 28
, ? Perey aaa anes NV UGCA) con eee eceece 17, 25, 28, 30, 31, 75, 106, 109, 119
tanks, natural........----------++-----2---55- 93 Pee otha dees jo uahiskbe ete ee 34
Weatottt 608 nee cna ores hn snninnnnna rex seen 140 | Yucca valida.......2.2.202-2-2eeeeeeee ee 29, 107, 125, 128
Weak fish. ......------ ae SG ae rs 134 | Zacatecas Mesa.........------2ceeeeeeeeeececeeeee, 36
Well (see also Tank, Spring, Tinaja). Zamelodia (see Hedymeles)...............-..--25- 116
Calmalli. ......------ +2022 cece cece reece ee eee 84)| Zenaidura:.-...o-csccaverenteuseemeenacuecawerees 115
Campo Aleman......---------e2+ee2 eee eee ene 84) ZanGr sce wrsseeerncsesecceewamaacseeaemeanencats 8
Llano de los Buenos Aires.....--.------------ 84 | Zone, See Life zones.
San Augustin. .........-.ecee cece eee e eee eee 76 | Zonotrichia.........0.eeece cece sence cece ee eeecees 116

"iO
 

 

NATIONAL ACADEMY OF SCIENCES.

 

Volume XVI.
SECOND MEMOIR.

 

STUDIES UPON THE LIFE CYCLES
OF THE BACTERIA.

Parr I.

REVIEW OF THE LITERATURE,
1838-1918.

 

BY

F. LOHNIS.

—

 

 
CONTENTS.

Page

TtPOM WHOM so. cares secenesceche cies oeviSinwedeate daa cia ancien eleisin bielsietiieine creme d SEO EE Roe ete s eee aeseeeame 5
i. Different: Gell LOrnis i: ccsaaecsadekeoranasaenen ee rome nana neneGbudedeatiaeeaname Maecenas setenieesd 7
1, General: discussion ys 1205s osicinisiale'c nteiaineeieistemiecates ge sie ias cea eae igi ene haee aman’ 7

(a) Monomorphistic and pleomorphistic theories.............-------+--+--+-+-+-e- ee eee eee ee eee 7

(b) Causes of apparent and of genuine pleomorphism.............-.------++--+--+-+e eee eee eee 17

(1) (Conitamin ation ...4..00scaiciesccanecsic cin seidiieniniantionin ne ntaedeucesadeeesdedeeeeene 20

(2) Incorrect classification of pleomorphous organisms among the bacteria. ......-.... 21

(3) Abnormal (‘‘involution”’) forms............2-2- 2-2-2222 eee eee eee eee eee 24

(4) Variatione.and mutations: «< scvess< scene sciieiey paceeta semen occ stakea eee 29

(5) Morphological changes during the life cycle............-.-..-2.2+2+2-2202ee eee ee 33

(c) The study of pleomorphistic problems. ............------.0- 220.22 - sees ee eee ee eee eee eee 37

2. Morphology of the different groups of bacteria...........-.- 2-2-2. 22-2 e eee eee eee eee eee 43

(@): COCOl. os catertconoeacatccaaudae nieces amncennaeneasesend Dp arbeBeastegiate one dass Ga aieoa dace 43

(b) Nonspore-forming rods..........-.----2-2-2222 20222 e eee eee eee eee eee Soe eames ati 49

(c) ‘Spore-forming rods... c2vcsnyssecauxices wees beeen oodp eee sewrsameeen eae sisted Raia 62

(d) Spirilla;and spirochsets.<. vecs12cseweesdeexersess ssaereeaewerceesscteecesehilseceeaaes 68

(e) Higher bacteria (trichobacteria, red Sign ggneee Renee oeesa epee eeateeee see 75

i OMG VIBEOIES a2 eo oh achig ne Lcas oan jzrnuss<veusysiaisy2yansustdvs lousqaus says Sc acatc a2 Spat Sys Susce chy zich cau Seng cemea sche eaten ee aer step 88

II. Reproductive organs. (Gonidia. Regenerative bodies. Spores. Microcysts) ..... Scere Dac t ane ty a bers 90
De, General CiSCUsSON si corsni aie eiepcarete wiateninie Sei erewwraiaenieweleiee Lov aaeReeAnacca eRe meee ea 90

(a) The discovery of the different modes of reproduction...........2..22.2--- 2020 e eee e eee eee 90

(by Differentiation between reproductive organs and other cell products and artefacts......:.. 108

(c) The different types of reproductive organs........2-2.2 02.20. e ee eee eee eee eee eee eens 119

(DY GOT aa ox fie tesmicicte tered eicle eile escyecaicenertesdecenatepeie Ses ig HES CSA Senses ie Seelam wn 123

(2) Regenerative bodies and exospores.......-.---.------ 22-2022 e eee eee eee eee ee 131

(3): HM GBP Ores oya:aiewiara.cics moraine ne dine Wee ntsndudabirm aerial w atdbnemsladn eaten alum aidoninniene 138

(4) Arthrospores and microcysts..............200 ccc eee ccc eee eee eee een ence 141

(d) Gonidia and filterable Vira (cell inclusions, Aphanozoa, etc.)—Gonidia and heterogenesia... 148

2. Reproductive organs in the different groups of bacteria..............22.2-.22-.0202 eee eee eee eee ee 151

(4): Coch evans cuales se esemtiererts costars eeeecehe acts weed. ndiswe ws clean meee ne iere selevess 151

(6) Nonspore-forming rods. .......... 2-2-2222 - 2222 eee eee eee tence eeeeees 152

(ce) Spore-forming rods 2.2 2. eciecec ce cceeadamseedee ope nao ban ealeae oud hpetbesnaulacuronew sae 155

(d) ‘Spinlla‘and spirochaete iss osccctmecesscecucecamsenese pec ceae eenie re meekeeielaelelgaee 157

(e) Higher bacteria (trichobacteria, mycobacteria).............2.02 22222022202 cece eee eee eee 159

3, Conclusions s254.i0jc ass stacidew nieces oem shouuidesoremew veecdex sea Beeeee ame suena eeeetewra tenors 162

III. Formation of the symplasm and the regeneration of cells...............2.-22- 2-20-20 ee eee eee ee eee eee eee 166
1, Greme ral GisCUsel ON s 22a) e Stesssaiain’a;o 5s! nsshalainracolsiecatsishorsinisianes dads gue ages Heed Hutamauiaaroedonnene 166

(a) The discovery of the symplastic stage..........2....22-22-2 222 eee ee eee eee eee 167

(6) Formation and appearance of the symplasm...............2 0.000 e ee eee eee ee eee cee eee 181

(c) Reproduction of regenerative bodies and of vegetative cells from the symplasm............. 185

2. The symplastic stage in the different groups of bacteria...............2--0- 0222 eee eee eee eee eee 190

3. Conclislons icc hobueac ee netes sted te ikte sca niernne aia sears Zan ene eae see aesncaiel acca owe’ 195

DY... Conf UM Ct OM sd 25:05 sa's/ssse chia sta oheetsteew,thece 4 ose nn ieaeead nem oo alddae leeching gS aaa ele eS ho aR an REneecmecodea 197
Vo Methods. sccasasnasgin scwncnnate eae ememecersemintinmnsaceeeeetyiid Gao acen aniitacnawasoyulehtckleens 205
1. ad I: Different cell forms............-...-...2.-.--- i ilusss ket Nadi A erahere mine esaherats da cap varare eevee ale een 205

2. ad Il: Reproductive organs .200022ccererseneds veesestbage+igoedddeeeseusesdeelekuwe eeenercans 207

$4 ad TIT: Symplasniy 2 ssc-cecc tenes secede sneha ee cae We eeee seco oe ee Ss bail aaeawmaimoanns 210

Aid Vee! COT J UPL CTL OU secs cdec ed sce aecircas evs epoca s Subst aead cto oye 2 SpthdadieE SSO Aan coemauteaiueraeee 212
Biblogtaph ys oniaiccmcsocesaneawcancrenmecaanecnaetGand 4 thessse we jecue bed een itawas dzone billy eons 213
SUDJeCtAN MEX: acs co prteinini seid ctat eeiarelalelsle eens ac mmanisat nanan Sannin acca Gines asdugdelua seedeneees 247
Plates:.A to:S:and. I to XAT ec bees cceeeiess a ctene eee seen mena iiee dee gniy yaue yEeeEee Roe REL aeme 6 253,
STUDIES UPON THE LIFE CYCLES OF THE BACTERIA.
INTRODUCTION.

Results obtained in comparative studies upon the morphology and the life history of some
important soil bacteria, especially of those of the Azotobacter group, made it advisable to extend
these investigations, so that representatives of other prominent groups of bacteria were also
included. Short reports of these investigations have been made in two preliminary communica-
tions (Léhnis and Smith, 1916 a and b), wherein it was pointed out that undoubtedly the complete
life history of all bacteria} is much more complicated than has been generally assumed. Our
findings were summarized in the first paper (1. c., p. 700) as follows:

All bacteria studied live alternately in an organized and in an antarptiis stage. The latter has been called the
“symplastic” stage, because at this time the living matter, previously inclosed in the separate cells, undergoes a thor-
ough mixing either by a complete disintegration of cell wall, as well as cell content, or by a “melting together” of the
contents of many cells which leave their empty cell walls behind them. In the first case a sendy ee in the latter
case an unstainable ‘‘symplasm” is produced.

According to the different formation and quality of the symplasm the development of new individual cells from
this stage follows various lines. In all cases at first ‘regenerative units” become visible. These increase in size, turning
into “regenerative bodies,”’ which later, either by. germination or by stretching, become cells of normal shape. In some
cases the regenerative bodies also return temporarily into the symplastic stage.

Besides the forniation of the symplasm, another mode of interaction between the plasmatic substances in bacteria
cells has been observed, consisting of the direct union of two or more individual cells. This “conjunction” seems to
be of no less general occurrence than the process first mentioned. Its physiological significance remains to be studied.

All bacteria multiply not only by fission but also by the formation of ‘“‘gonidia”; these usually become first regenera-
tive bodies, or occasionally exospores. Sometimes the gonidia grow directly to full-sized cells. They, too, can enter
the symplastic stage. The gonidia are either liberated by a partial or a complete dissolution of the cell wall, or they
develop while still united with their mother cell. In the latter case the cell wall either remains intact or it is pierced
by the growing gonidia, which become either buds or branches. :

Some of the gonidia are filterable. They also produce new bacteria either directly or-after having entered the symn-
plastic stage...

The life cycle of each species ‘of bacteria studied i is composed of several subcycles showing wide morphological and
physiological differences. They are connected with each other by the gpmplantio stage. Direct changes from one sub-
cycle into another occur, but they ‘are rather rare exceptions. The transformation of spore-free into spore-forming
bacteria seems to'be dependent on the conditions acting wpon the symplasm and regenerative bodies.

The fact that all the different Phases i in the life history of the bacteria, i. e., conjunction,
formation and development of regenerative bodies, symplastic stage, and the occurrence of
different subcycles with peculiar cell morphology, were to be recorded constantly and regularly
in all cases studied by us, made the conclusion inevitable that these apparently new facts must
have been observed in the course of many previous investigations, though they usually have
not attracted much attention, and hardly any of the bacteriological textbooks furnishes
information upon them. The practical aims of most of the bacteriological investigations, to-
gether with the widespread adherence to standardized methods and theories, have been dis-
tinctly unfavorable to thorough studies along such lines. Nevertheless, at least a few references
could be given in our preliminary communications, and it was to be expected that a sys-
tematic search of the literature would lead to the rediscovery of more confirmative reports, which
so far had not found any acknowledgement, or which have been rejected on account of their being
not in agreement with the prevalent theory. This expectation has been fully verified.
At the present time it is out of the question to make a complete collection of all such
details, which are widely scattered in a gigantic literature. When in 1875 Ferdinand Cohn pub-
5
6 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI.

\

lished the second part of his ‘‘Untersuchungen iiber Bacterien’’ he was of the opinion that the
literature even at that time was so voluminous that a complete survey were hardly possible even
for the specialist. In 1880 Prazmowski made a similar remark when he reported on his “‘ Unter-
suchungen tiber die Entwicklungsgeschichte und Fermentwirkung einiger Bakterien.”” Since
then, every year has added thousands of new bacteriological publications. It would be an inex-
cusable waste of time to study all of them again in the light of the newly discovered facts. ButI
earnestly hope that in the discussion of the literature, as presented on the following pages, no
important contribution has been overlooked, so that the general situation is presented with fair
accuracy, and all those authors have been mentionad who have tried to shed more light upon
the life history of the bacteria.

Complete reports upon our own experimental investigations will be given later. For making
a proper arrangement of this review of the literature, however, our two preliminary communi-
cations, published in 1916, will be used. As will be seen from the following pages, every phase
in the life history of the bagien 4 has been observed by several, sometimes even by many, inde-
pendent workers, though only in comparatively few cases the complete life cycle of a distinct
species has been discovered.

It seems best to arrange the whole matter in the following five chapters:

I. Different cell forms.
II. Reproductive organs. (Gonidia. Regenerative bodies. Spores. Mictooytts. )
III. Formation of the symplasm and regeneration of cells.
IV. Conjunction.
V. Methods.

In chapters I and II a considerable number of details concerning the different groups: of
bacteria will have to be discussed. For this purpose, and in accordance with the general °
standpoint taken at present by most bacteriologists, the following five groups of pratieme may
be accepted:

(a) Cocci.
(6) Nonspore-forming rods.

(c) Spore-forming rods.

@) Spirilla and spirochaets.

(e) Higher bacteria (trichobacteria, mycobacteria).

As will be seen later, this arrangement is not in accordance with the true chorncter of and
the natural relations existing among the bacteria. But as long as these facts and relations are
not more completely known, it will be best to adhere to that more familiar grouping. At the end
of Chapters I, II, and III some conclusions will ibe drawn from the observations reviewed
therein, which m might be helpful for further. investigations. - For the same reason the methods
have been treated.separately in Chapter V..

It is hardly necessary to point out that the < data collected from the literature are of very
different value, and we are often left in doubt as to. their correct interpretation. Much of the
material discussed on ‘the following pages is awaiting further critical experimental study.

Special attention has been given to the morphological side of the subject. Unquestionably
the physiological side is. of equal or even. greater importance. However, as the situation is at
present, with most, bacteriologists on the one ‘side adhering to the dogma of simplicity and con-
stancy of the form of the bacteria, while on the other side admitting a more or less considerable
variability in their physiological behavior, it appears to be of prominent importance that in the
first place a more correct knowledge of the morphological basis of bacteriology will be secured.
As soon as the morphology of the different phases i in the life cycles of the bacteria will have re-
ceived adequate treatment, physiological investigations will also furnish much more satisfactory
results than are obtainable now. And only after the full life history of the different types of
bacteria will have been studied it will become possible to define their natural relations and to
come ultimately toa correct systematic arrangement of the various species, genéra, and families
of the bacteria.
‘I. DIFFERENT CELL FORMS.

1. GENERAL DISCUSSION.

It would be preferable, if the different forms of the vegetative cells alone could be
treated in this chapter, leaving to Chapter II all details concerning formation, appearance,
and development of the various reproductive organs (gonidia, regenerative bodies, and spores);
but this is not possible at the present time. In many cases, of course, a distinct separation is
quite feasible; sometimes, however, we are left more or less in doubt whether the forms men-
tioned in the literature are those of vegetative or of reproductive cells. This holds true espe-
cially for many of the globular bodies. The reproductive organs, which we call gonidia and re-
generative bodies, have been frequently considered to be micrococci, a mistake which indeed is
quite excusable. Both of them are able to multiply as such, and as this fact is more easily to be
observed than their germination, usually their real character has not been discovered. On the
other hand, it does not infrequently happen that typical micrococci almost imperceptibly
change from the vegetative into the reproductive state by thickening their cell wall and assuming
the character of cysts, which fact, however, can be discovered only by a very close study.
Other cells, too, may display either vegetative or reproductive activities without undergoing
conspicuous changes in their appearance, so that their correct classification may also cause
some difficulties. But even if all reproductive cells could be as clearly distinguished and sepa-
rated from the vegetative cells, as is the case with the endospores, still a considerable number
of trustworthy observations would remain, proving beyond doubt that, if not all, undoubtedly
very many bacteria have shown themselves to be able to appear in different forms, while in
the vegetative state.

(a) MONOMORPHISTIC AND PLEOMORPHISTIC THEORIES.

The questions concerning monomorphism or pleomorphism of the bacteria have been
extensively treated by European as well as by American authors during several decades. The
new point of view, which is furnished by the fact that the bacteria, like other microérganisms,
exhibit different forms and activities in the various stages of their life cycles, makes it necessary
to review briefly the opinions promulgated in this respect, for at least some of them may eventu-
ally become of considerable value in future researches.

' ‘In the French literature pleomorphistic ideas have been always en vogue. At first, before
the theory of spontaneous generation was fought by Pasteur, the transformation of the “ globules
élémentaires,”’ found in the residues of plant and animal life, into “infusoires”’ was the main
point of discussion. The writings of Dujardin (1841), Pineau (1845), Pouchet (1863), Trécul
(1865-1867), and Béchamp (1883) are to be mentioned especially in this respect. Robin (1871,
p. 931) was inclined to consider all micrococci, bacteria, and vibriones simply as stages in the
development of his Leptothrix buccalis. Davaine (1876), however, correctly emphasized, as
had been already done by Dujardin, that undoubtedly many different species exist; bacterium
and vibrio are declared to be related to Leptothrix, but not to be identical with it. When
studying a distinct species, it is said to be of special importance to pay attention to the varia-
bility of its forms ‘qui changent suivant diverses conditions.”’ Duclaux (1883) wrote in his
“Chimie biologique” (p. 33):

En résumé, toute classification fondée sur des caractéres purement morphologiques semble impossible . .
Pour donner une idée de l’incertitude qui en resulte, je dirai que la qualification de Bacillus subtilis a été donnée
déja a une dizaine d’étres, tous identiques si on ne consulte que leur nom, tous différents si on ne consulte que les

propriétés qui leur sont attribuées dans les divers mémoires qui en parlent.
7
8 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The following statement, made by Artigalas (1885, p. 23), forecasts our present standpoint
with remarkable accuracy:

Nous verrons qu’un microbe donné peut, 4 une certaine période de son évolution, étre rapporté 4 un type dont
ilse sépare complétement dans la suite. En un mét, un microbe peut passer par des transformations telles que, comparé
a lui-méme, l’expérimentation seule en puisse démontrer Pidentité. - .

Correcting an erroneous view and tendency, which was rather common at that time and
later in the German as well as in the American literature, Guignard and Charrin wrote, reporting
on their interesting experiments “Sur les variations morphologiques des microbes”’ (1887):

Ce polymorphisme ‘ n’ébranle en rien la notion généralement admise pour l’espéce; il n’en doit pas
moins attirer de plus en plus Pattention sur l’'influence des milieux, . . . et mettre en garde contre certaines
tendances & trop multiplier les espaces en se fondant sur des données morphologiques, insuffisantes.

More evidence against. the assumed constancy ‘of form and function was furnished by
Wasserzug’s publications (1888 a@ and 6b). Metchnikoff, too, in various contributions (1889-
1894), dwelt upon the “pléomorphisme, si répandu dans le monde des bactéries.”’

In the third edition (1890) of ‘‘Les bactéries et leur réle dans 1’étiologie, l’anatomie et
Vhistologie pathologique des maladies infectieuses’”’ Cornil and Babes defined their standpoint
as follows (Vol. I, pp. 18 and 30):

Un grand nombre. de botanistes affirment que les bactéries présentent des formes diverses suivant le degré de
leur développement et le milieu nutritif ot elles sont placées. Cette polymorphie est bien démontrée pour un grand
nombre d’entre elles . . . On doit méme se demander si toutes les bactéries ne présentent pas des formes variées
et si l’on ne décrit pas aujourd’hui un méme microbe sous plusieurs noms différents—Il ne faudrait pas croire que
les formes des bactéries soient toujours constantes, immuables, de facon 4 caractériser des genres distincts. Tout au
contraire, on sait que certains micro-organismes revétent des formes diverses pendant leur. développement de telle
sorte qu’ils se présentent comme un coccus, un batonnet, un filament ou une spirale dans les états successifs de leur
accroissement.

_ Similar to Artigalas,. Billet (1890) emphasizes the necessity of investigating the complete
life-history of the bacteria, because, as he says (p. 217):

On reconnaitra peut-étre alors qu’un grand nombre des formes que l’on décrites jusqu’ici comme des espéces
absolument distinctes étrangéres |’une 4 l’autre, ne sont, en réalité, que des formes d’éléments appartenant & la méme
espéce.

When Rodet in 1894 ptblihed’ his book entilled “La variabilité dans les microbes,” Arloing,
who wrote the preface for it, made the following reference to the dogmatism of R. Koch and
his pupils: 28

Tl fut un temps ot l’on croyait qu’un microbe avait des fonctions immuables comme sa forme . . Les
microbes étaient donc specifiquement et rigoureusement determinés par la forme et les propriétés. Cela était un dogma
sur lequel semblait reposer toute la microbie. Un novateur ne pouvait le menacer sans soulever des protestations ou
faire naitre un sentiment de défiance a l’égard de ses travaux.

Further proof supporting the PeeonNp ine view was also furnished by AAoing himself
in cooperation with Chantre (1894). . “

Macé (1897) is the only French author who a sometimes been quoted as standing on the
monomorphistic side. He is, indeed, of the opinion that much of what has been found by other
French bacteriologists in their investigations upon the variability of the bacteria has been
obtained under special, distinctly abnormal conditions (p. 13). But he also writes, very
similar to Cornil and Babes, concerning the one-sided theory held at that time by most of the
German bacteriologists (p. 12):

Cette opinion a été fortement battue en bréche, lorsqu’on est arrivé & prouver que certaines espéces pouvaient,
selon les circonstances de milieu ou la phase de leur cycle évolutif, donner tantét des cellules sphériques, des Coccus,
tantét des bitonnets courts, tantéts des filaments droite, tantét des filaments spirales.

The same more or less pleomorphistic standpoint is also taken in the more recent French
publications, e. g., in the textbooks of Miquel and Cambier (1902), Burnet (1911), who even
says (p. 66): ‘Les bactéries sont variables au point de dérouter les bactériologistes;”’ Le Blaye
and Guggenheim (1914), and Kayser (1914). The latter once more clearly emphasizes the
point which in the German, as well as in the American, literature has caused much misunder-
' standing (p. 28):

Cette grande variété de formes n’est nullement en contradiction avec l’unité de l’espace.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 9

In the British bacteriological literature the pleomorphistic point of view is not quite. so
generally accepted as was the case in France. Husley (1870), who unfortunately believed
in the accuracy of the work done by Hallier and Liiders in Germany, thought that really Peni-
cillium could produce Torula and this bacteria, a hypothesis which was at once refuted by
Burdon-Sanderson (1871). Bastian, on the other hand, became a persistent advocate (1872-
1914) of the theory of heterogenesis as it was upheld by the earlier French writers (Pineau,
Pouchet, Trécul, Béchamp). Distinctly pleomorphistic ideas were also promoted by Lister
(1873) and by Lankester (1873-1876). Lister himself modified his statements later (1878) as
far as they were based on incorrect conclusions drawn from results obtained with mixed cultures.
But he was right when he said (1873) that a classification of the bacteria based upon absolute
morphological constancy ‘‘is entirely untrustworthy;’’ and his clear drawings of branching and
budding bacteria also deserve our interest to-day. Lankester, too, undoubtedly worked with
impure cultures, but it must not be overlooked that much of his work is based on continuous
microscopic observations of the changes in morphology exhibited by his Bacterium rubescens.
And at least in some cases it still can be shown, as will be done in Chapter ITI, that in fact his
observations have been much more accurate than those of his critics.

A distinctly monomorphistic standpoint was taken by E. Klein (1885) in his textbook on
“‘Microérganisms and Disease,”’ obviously under German influence. He points out that true
micrococci never elongate to form rods, and mentions as such “true micrococci’’ F. Cohn’s
“ Micrococcus’’ prodigiosus and violaceus, which he also found to be always globular and immo-
tile, contrary to now well-known facts. But his own observations on anthrax, cholera, and other
pathogenic organisms equally militated against his standpoint. And some years later, after
he had found that Bacillus anthracis sometimes assumes a morphological character which
resembles very much that of some Saccharomyces or Oidium, he even did not hesitate to say
(1894): “Probably it returns to an atavistic stage in its evolutional history.”

H. M. Ward (1887) clearly states that ‘‘Schizomycetes are pleomorphic,’’ but, as he adds,
this is “‘no reason to deny the existence of distinct species.”’ Woodhead (1891), on the other
hand, is of the opinion that the doctrine of pleomorphism “can not be accepted as in any way
proved except in the case of a few well-known nonpathogenic forms,” whereas, according to
Crookshank (1896, p. 478), this doctrine is “widely accepted’? and—

Researches by competent observers have more recently clearly demonstrated that several microérganisms in their

life cycle exhibit successfully the shapes characteristic to the orders of Cohn. . . These facts obviously shake the
very foundation of Cohn’s classification, and we are left without possessing a sound basis for classification into genera

and species. . .

Hewlett (1902) : ae that “a certain amount of pleomorphism andodbtedly occurs in
some organisms,” and Muir and Ritchie (1903) equally admit that ‘there are cases where
evidence appears to exist of the occurrence of pleomorphism.”

Harris (1893) found that one of two aerobic organisms isolated from oedema was able to
exhibit ‘‘a complete series from cocci to bacilli.”’ The other, too, was ‘‘remarkable for its
pleomorphism.” Durham (1898) studied closely the dimorphism of Micrococcus melitensis.
Adami, (1892-1899) furnished other important contributions concerning the variability of the
bacteria. Together with WU. E. Abbott and Nicholson (1899) he was able to show that especially
B. coli may grow as a minute coccus, which also differs entirely in its cultural character from the
typical form. Another report upon these investigations, to which we have to refer later, was

published by M. E. Abbott (1900).
Dobell (1911, p. 484) believes ‘‘that the greater number of bacteria are pleomorphic,” and

he continues: .
The bacteria aré in no way a group of simple organisms, but rather a group displaying a high degree of morpho-
logical differentiation coupled in many cases with a life cycle of considerable complexity.

Revis (1912 a) changed a Coli strain experimentally to such an extent ‘‘that it now was
neither physiologically, morphologically nor culturally a Colon bacillus.’’ Rowland (1912) suc-
ceeded in imposing all marks of B. pseudotuberculosis rodentium upon a typical strain of B.
pestis. And results secured with many kinds of organisms, led Young (1914) to the hypothesis
10 MEMOIRS NATIONAL ACADEMY OF. SCIENCES. [Vou. XVI.

“that all bacteria formed merely stages in the life history of a group of organisms with an
extremely complicated life cycle.”

The most important contributions to our subject, ee in modern British literature
are those of Hort (1915-1917) and his collaborators, to which we will have to refer frequently.
It was with good reason that Adami (1916), after having become acquainted with these new
observations, raised the question:

Have we all these decades been wrong in our faith, that bacteria are the simplest of all living forms of life, pos
sessed of asexual multiplication by fission only? May there not be intercalated other phases in the life history of the
schizomycetes?

The German and Austrian bacteriological literature shows an interesting change from
monomorphistic to extremely pleomorphistic views, then back to the rigorously monomor-
phistic standpoint of R. Koch and his pupils, and at last an almost general acceptance of the
moderate pleomorphism as it has always been taught by French bacteriologists.

Although there is no special discussion upon this problem, the monomorphistic point of
view is already clearly discernible in Ehrenberg’s (1838) as well as in F. Cohn’s (1853) early
publications. It is, however, by no means, that strict monomorphism which was advocated
20 years later by the same author. Bact. termo, e. g., was declared to be able to grow in globular,
curved, or rodlike form.

The extreme pleomorphism, on the other hand, promulgated especially by Hallier (1865-
1878) and by Liiders (1866), undoubtedly was not supported by sufficiently reliable observa-
tions, but it can also not be considered fair to condemn summarily all what has been written by
Hallier as being “scientific nonsense’’ (Winter, p. 33), or as merely belonging to the scientific
‘chronique scandaleuse”’ (DeBary, p. 137). As much of this work was based on long con-
tinued direct microscopic studies, some of his results, though obtained with impure cultures,
still retain their value, and will even regain new interest in connection with other data which
are to be discussed in Chapter II. The same holds true in respect to various facts observed
by Karsten (1869). This author, as well as H. Hoffmann, already in 1869 and with good reason
rejected Hallier’s erroneous assumption of a direct connection existing between fungi and
bacteria. Hoffmann’s publication is of special interest, because it contains the first drawings
of branched bacteria, which later were rejected by the extreme monomorphists.

Morphological changes of single bacteria cells have been studied directly under the microscope
by Rindfleisch (1872), as well as by Billroth (1874). The latter author has been repeatedly
called the champion of the older pleomorphism, usually on account of no other reason than
that he named all bacteria together Coccobacteria septica. But he left. it explicitly (p. 35)
for the botanists to decide, whether or not the different forms which he observed were merely
different types of growth of this one Coccobacteria septica or whether several genera and species
should be made.

It is a similar case with Ndgeli’s standpoint. Usually the following sentence (1877, p. 20)
has been quoted as convincing proof of his alleged error:

Ich habe seit zehn Jahren wohl tausende von verschiedenen Spaltpilzformen untersucht, und ich kénnte (wenn
ich Sarcina ausschliesse) nicht behaupten, dass auch nur zur Trennung in zwei specifisch verschiedene Formen Néthi-
gung vorhanden sei.

; However, the following statement in the same publication (1877, p. 22) also must be taken
under consideration:

‘Wenn ich sage, dass die uns bekannten morphologischen Eigenschaften der Spaltpilze und ihr Vermégen, ver-
schiedene Zersetzungen zu bewirken, eine generische und specifische Unterscheidung nicht rechtfertigen, und. dass
selbst die Méglichkeit vorliege, alle Formen in eine einzige Species zu vereinigen, so liegt es mir doch fern, diese
Behauptung wirklich auszusprechen. In einer Sache, in welcher die morphologische Beobachtung und der phy-
siologische Versuch den Forscher noch so sehr im Stiche lassen, ist es tberhaupt gewagt, eine bestimmte Ansicht
auszusprechen.

In his later work (1882, p. 180) he defends himself against those who assert that he be
inclined to consider all bacteria one species. But at thesame time he points out that also accord-
ing to his new experiments with pure cultures, all morphological and physiological qualities of
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 11

the bacteria may vary to a considerable extent. His standpoint is best defined by the follow-

ing quotation (p. 138):

Es wird bei den Spaltpilzformen die nimliche Erfahrung sich wiederholen, die in neurer Zeit an den iibrigen
Pilzen gemacht wurde, wo die verschiedenartigsten morphologischen und physiologischen Erscheinungen als ver-
schiedene Generationen einer und derselben Species erkannt wurden. Die Species wird nicht durch absolute Merk-
male kenntlich sein, sondern dadurch, dass sie unter bestimmten dusseren Umstinden bestimmte Modificationen des
morphologischen und physiologischen Verhaltens, unter anderen Umstinden andere Modificationen zeigt. Ein
System der Spaltpilze nach Gattungen und Arten mit den jetzigen Hilfsmitteln aufzustellen hat keinen wissenschaftli-

chen Wert.

That, in fact, bacilli may assume sometimes distinctly coccoid shape, and spirilla may
temporarily look like straight rods, Négeli was able to ascertain with pure cultures. Some
years earlier Klebs (1875 6) had made a similar statement, also based on direct microscopic
observations. :

When Ferdinand Cohn began the publication of his ‘Untersuchungen tiber Bacterien”
(1872), he also, by no means, advocated those extreme monomorphistic ideas, which later,
mostly on account of the writings of his fo!lowers, have been attributed to him. He strongly
emphasized that genera, as well as species, proposed by him were to be considered form genera
and form species, not natural ones. The following lines (1872 b, p. 130) clearly define his
original standpoint:

Wir sind gendtigt, bei den Bacterien in vielen Fallen ein Verfahren anzuwenden, das auch in der Mycologie so
lange festgehalten wird, als nicht durch vervollkommnete Culturmethoden die gesamte Entwicklungsgeschichte
der Arten festgestellt werden kann.... Es besteht darin, dass jede Form, die sich durch hervorstechende Merkmale
auszeichnet, mit einem besonderen Gattungsnamen belegt. wird; jede kleinere Abweichung wird als Species unter-
schieden. Es soll damit nicht die Méglichkeit ausgeschlossen werden, dass nicht verschiedene solcher Species aus
einer und derselben Mutterform hervorgehen, ja dass nicht selbst verschiedene Gattungen nur Entwicklungszustinde
eines und desselben Individuums sein kénnen. ,

In his later contributions (1875, p. 142; 1876, p. 274) Cohn changed his mind in so far as he
now declared his genera to be naturel, not only form genera. His own investigations, however,
did not support this statement. There was indeed no possibility at that time to study the
complete life history of these smallest organisms, and Cohn’s classification is clearly incorrect
in many cases. HE. g., he mentions in his “natural” genus Micrococcus, which is said (1872 b,
p. 151) to be made up of immotile globular or oval cells, the following species:

Micr. prodigiosus, luteus, aurantiacus, chlorinus, cyaneus, violaceus, ureae, candidus, vacetnae, diphteriticus, sep-
ticus, bombycis, fulvus, phosphoreus.

According to our present knowledge only 5 of these 14 species (I. luteus, aurantiacus,
cyaneus, wreae, and candidus) are really Micrococci. That such conspicuous characters as the
motility of B. prodigiosus or the pleomorphous nature of Myzxococcus fulvus—Jahn (1909-1911,
pp. 191 and 198) has shown that this, indeed, is what Cohn called Microc. fuluus—have not
been noticed by F. Cohn sufficiently proves our point.

The German botanists who, like Winter (1884) and Schroeter (1886), otherwise closely
adhered to Cohn, did not accept his genera as “‘natural” ones. Wwnter (p. 36) says that several
of Cohn’s genera are merely developmental stages of other genera and that many of his species
probably are nothing else than more or less constant modifications, split up from one species
under different environmental conditions. Schroeter, too, was of the opinion (p. 140) that the
form of the individual cells very often change in the course of their development. They all
must be known before the true character of a species can be defined.

As Robert Koch in the early part of his bacteriological work was greatly influenced by F.
Cohn just at the time when this author had left his earlier, more cautious standpoint, naturally
the monomorphistic view was readily adopted. In his ‘‘Untersuchungen iiber die Aetiologie
der Wundkrankheiten” (1878) the following statement is made (English translation, p. 71):

In all my experiments not only have the form and size of the bacteria been constant, but the greatest uniformity
in their action on the animal organism has been observed.

Constancy of form and action became a dogma, which was upheld for the next decades
more by R. Koch’s pupils than by himself. That in many cases constancy of action is missing
12 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Von XVI.

was soon shown by French bacteriologists under the leadership of Pasteur. And the. ques-
tion whether or not separate species should and can be based on the frequently very small
differences found has been left unanswered by R. Koch. On account of practical reasons he
emphasized the necessity of making as many divisions as were indicated by differences of form
or action. But he did not care whether these now might represent ‘‘species, or varieties, or
forms, or what else one should like to call them” (1881, p. 31). And later he was even quite
willing to admit the variability of the bacteria (1890): :

Innerhalb gewisser Grenzen kénnen Abweichungen von dem gewéhnlichen Typus der Art bei den Bakterien und
insbesondere auch bei den pathogenen Bakterien vorkommen.

Strictest constancy of form and action was first proclaimed by Gaffky (1881). To him it
was unquestionable that any change which might occur was caused by the overgrowing of one
species by another. Though Friedlander (1883) found that his ‘‘Mikrokokken der Pneumonie”
(now usually called B. pnewmoniae) can grow either in distinctly globular, oval, or in rod-like
form of different length, Fliigge (1884) did not hesitate to declare:

Niemals haben wir beobachten kénnen, dass wirkliche Cokken in Bacillen sich umwandeln und umgekebrt.

In the same publication this author makes also another characteristic statement which
later has caused much unnecessary confusion:

Wenn Thatsachen gefunden waren, aus welchen die Wandelbarkeit der Infectionserreger gefolgert werden miisste,
so wiirden wir .. . auf eine weitere experimentelle Erforschung der Infectionskrankheiten verzichten miissen.

Buchner (1885), however, correctly pointed out that the disputed constancy of form and
action has nothing whatever to do with the indisputed constancy of species. Nevertheless,
this logical mistake is even to-day not yet quite eliminated.

The ‘Bacteriologische Diagnostik,” published by J. Eisenberg in 1890, gives a very char-
acteristic picture of the German literature as it was written at that time by R. Koch's pupils.
Even such highly pleomorphous organisms like the diphtheria and tubercle bacilli were con-
sidered to be correctly described as follows:

. Diphtheria bacillus (p.237): ‘‘Teils gerade, teils leicht gebogene Stabchen von der Lange der Tuberkelbacillen,
jedoch etwa doppelt so dick. In gefarbten Préparaten durch tiberwiegende Farbstoffaufnahme an den Enden han-

telformig.”
Tubercle bacillus (p. 253): ‘‘Sehr diinne, 2-5y lange Stabchen, meist leicht gekrimmt.”

The following two statements of CO. Frinkel (1891, pp. 12 and 169) are equally interesting.
The constancy of action is not longer upheld, but Fliigge’s logical mistake is renewed:

Bisher ist eine vielférmige Bakterienart nicht zur Beobachtung gelangt, und der Satz ““Man kann unter den
Bakterien nach Wirkung und Form scharf unterschiedene Gattungen und Arten erkennen, welche nicht in einander
tibergehen” bleibt zu Recht bestehen.—Freilich ist die Grenze, die wir zwischen pathogenen und nicht pathogenen
Bakterien ziehen, keineswegs eine feste und unverriickbare. Eine ganze Anzahl von Mikroorganismen, die
gewébnlich ganz harmlosen Character zeigen, vermégen unter Umstinden auch eine pathogene Rolle zu spielen, und
ebenso kann manche pathogene Art ihre gewéhnlichen Eigenschaften ablegen und in die Reihe der unschidlichen
Bakterien iibertreten.

Gotschlich (1903-1909) and Giinther (1906) once more tried to support the “ morphologische
Grundgesetz” of the constancy of the form as proclaimed by Cohn and Koch, though a con-
siderable number of contradictory results now had to be admitted. Giinther accepts already
for the different species not a constant form, but a constant form cycle. But he still thinks
the constancy of the form is indisputable, because there is no “unlimited” morphological varia-
bility (1906, p. 22).

Among the German botanists the monomorphistic theory also gained some temporary
foothold.

When Brefeld in 1881 first entered the subject, he wrote quite correctly, similar to Nageli

(p 50):
Die Untersuchung von Spaltpizformen muss . . . entwicklungsgeschichtlich' . . . in Angriffgenommen
werden, wenn eine natiirliche Systematik . . . begriindet werden soll. Die vorliufigen Einzelheiten reichen

daftir auch nicht von ferne aus.

Later, however, his standpoint was quite different (1908, pp. 99-100):

In rein morphologischer Beziehung | bieten die Bakterien nach ihrer einfachen Formbildung und der stereotypen
Art ihrer Teilung und Vermehrung wenig Bemerkenswertes dar . . . Die Ziele der Kultur bei den Fadenpilzen,
No.2.) | LIFE CYCLES OF THE BACTERIA—LOHNIS. 13

die Hohe der vegetativen Bildung zu erreichen und dann die verschiedenen Fruchtformen auf dem Wege der Kultur
aus den vegetativen Zustanden zu gewinnen, ihre Zusammengehdrigkeit zu den einzelnen Formen der héheren Pilze
sicher festzustellen, liegt bei den Bakterien von vornherein ausserhalb der Fragestellung.

A. Fischer (1903, p. 44), as well as A. Meyer (1912, p. 61), take also their stand against

the pleomorphism, though, of course, they do not dispute the variability of the cellform. Benecke
- (1912, p. 216) says:
Jede Art hat ihren Variabilitatskreis, der grésser oder kleiner sein kann.

The very numerous papers on variation, mutation, etc., contained in the German litera-
ture of the last decade, will have to be discussed later. Taken together, they furnish convincing
proof that the monomorphistic theory of Cohn and Koch has become utterly untenable.

However, more or less opposition has always been put forward by other German and
Austrian bacteriologists, though the very important practical results attained by R. Koch and
his pupils have materially helped to keep their antagonists temporarily in the background.

Neelsen (1880) drew the following conclusion from his studies upon B. cyanogenes, which
are of special interest because this work was done in F. Cohn’s laboratory and published as the
tenth contribution to his ‘‘Untersuchungen iiber Bacterien.” It is stated therein (p. 242):

dass die von Cohn aufgestellten Gattungen insofern zu eng begrenzt sind, als ein Organismus in seinem Lebens-
cyclus verschiedene der als getrennt hingestellten Formen vereinigen kann.

‘Haberkorn (1882) emphasized correctly that ‘the hypothesis of simplicity and constancy of
form was not well founded. Unfortunately his own experiments, though very interesting, are
open to the same criticism. But he was right when he wrote:

; Richtiger, d. h. der Wissenschaft férderlicher ist, beim Forschen noch Lticken im Entwicklungsgange vorauszu-
setzen, als voréilig den genetischen Cyclus abzuschliessen.

Miller's (1882) investigations upon the bacteria of the human mouth, Kurth’s (1883)
isolation of Bacterium Zopfii, growing in globular, rod-like and spiral form, Hauser’s (1885)
discovery of his three Proteus forms, first considered to be different species, later (Hauser, 1892)
acknowledged as modifications of one species, Biedert’s (1885) and Malapert-Neufville’s (1886)
studies upon other pleomorphic bacteria from mouth and water, furnished additional proof
that pleomorphism does by no means obscure, but even enhances the well defined character of
aspecies. DeBary (1884) also declared, that doubtless different species have.to be distinguished
with the bacteria as with the higher organisms. Some of them may exhibit a relatively mono-
morphous character, while others may be distinctly pleomorphic.

The most prominent place among the German adversaries of Cohn’s and Koch’s stand-
point has been taken by Zopf (1881-1885). Itis true that many of his ideas have been gathered
from studies done with species (especially with thread-bacteria) which have not very much in
common with the pathogenic microbes Koch was interested in. Nevertheless, his opposition
against the extreme monomorphism was sufficiently supported by well ascertained facts, and
he was equally right when he emphasized, like Nageli and Buchner, that as long as the complete
life history of the bacteria remained unknown no correct systematic arrangement could be
made. '
More well-founded criticism concerning the alleged constancy of form, was furnished by
Gruber (1885, 1894), Firtsch. (1888), and Escherich (1886). The last-named author says, that
now, after Koch’s methods have made it possible to study rigorously pure cultures, more and °
more bacteriologists are inclined to accept the pleomorphistic point of view.

Rosenbach’s publications on the causative agent of the so-called erysipeloid exemplify in
a ‘very interesting manner how extended research may completely change the situation. In
1884 he described the organism as a Micrococcus; in 1887 he declared it to be a peculiar branch-
ing organism probably related to Cladothria (Streptothrix), and in 1909 he showed that it rep-
resented a variety of Bact. ees suum, of which he discovered the full very polymorphous
life cycle.

The following quotations from Hueppe! s- writings are equally interesting. In 1883 he
criticized Zopf’s standpoint in a very sharp manner, very much like Fiiigge (1884). No change
from a coccus to a bacillus or vice versa could be accepted, and the idea that the tubercle bacillus
might be related to any ‘‘thrix” was declared to be entirely absurd. Ten years later F. Fischel
14 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

(1893) secured conclusive proof for this relationship by his investigations carried out under
Hueppe’s direction. But already in 1886 Hueppe’s own standpoint had undergone consider-
able modification. The “‘dogmatic onesidedness” (p. 85), which knows nothing but strict
constancy of form or unlimited variability, was now refuted and it was said (p.61):

Wenn auch die einzelnen Formen sich bald mehr bald weniger mit den Aussenbedingungen andern kénnen . ,
so tritt doch unter denselben Verhiltnissen immer ein ziemlich scharf Renney bald enger, bald weiter For-
menkreis auf.

In 1891 (pp. 22, 27) the occurrence of all different cell forms Siti the life cycle of the
same organism was still more readily admitted by Hueppe. And in 1896 Zopf’s standpoint
has been practically reached (p. 15):

Die starren Formarten, wiesie . . . Cohn, Schroeter, und Koch angenommen hatten, haben sich nicht aufrecht
erhalten lassen. Die Anpassungs{ihigkeit der. Bakterienformen bei Wechsel der Ernéhrungsbedingungen ist .
betrichtlich grésser, als man frither mit dem Begriffe konstanter Arten fiir vereinbar gehalten hatte.

Migula was at first also very much opposed against all ‘‘pleomorphistischen Geliiste’
(1897, Vol. I, p. 18). He taught (1. c., p. 50):

Die Bakterien zeigen eine grosse Einférmigkeit. Drei Gestalten sind es, welche uns immer wieder bei ihnen
entgegentreten: die Kugel, das cylindrische und das schraubig gekritmmte Stabchen.

But in the same book (p. 236) he also wrote:

Inwieweit die einzelnen Arten verschiedene Formen zu bilden vermégen, muss der weiteren Forschung vorbe-
halten bleiben; unsere gegenwartige Kenntnis davon ist im héchsten Grade liickenhaft.

In 1904 he admitted in Lafar’s ‘‘Handbuch”’ (Vol. I, p. 42) that the monomorphistic theory
had to be abandoned. The same statement had been made by Lafar himself in 1897 (p. 36),
and since then it has been amply confirmed by the results of practically all investigations made
along these lines by German and Austrian bacteriologists. The present situation is correctly
described in the following quotation from Lehmann and Neumann (1912, pp. 146-149):

Die Cohnsche Lehre von der Konstanz der Arten . . . wird heute in immer weiterem Umfange unhaltbar.
Denn die fortgesetzte, immer tiefer gehende Forschung hat zur Evidenz erwiesen, dass fast alle Eigenschaften einer
wohlumgrenzten Art sehr schwanken. . . . Wir glauben sicher, dass es der Zukunft noch in heute kaum
geabnter Weise gelingen wird, Bakterienarten in einander tiberzufithren. . . . Die Lehre von der absoluten Unver-
anderlichkeit der Bakterien, die vor 20 Jahren noch fast als Dogma galt, wird heute kaum mehr ernsthaft vertreten.

The old extreme pleomorphism of Hallier has been also revived in the German literature
from time to time. Hallier himself reiterated his former statements in 1895 and 1896, which
he now thought to be “in complete agreement with R. Koch's discoveries.” J. Miiller published
a similar work in 1898. To him the bacteria are still the offspring of the spermata of different
fungi, though already in 1869 H. Hoffmann had proven that this is not the case. Equally
erroneous were the discoveries of Stutzer and Hartleb (1897) concerning their highly pleomorphic
‘‘Salpeterpilz’” (a complex mixture of various fungi and bacteria, as was shown in 1898 by
Gaertner, Fraenkel, and Krueger) and their ‘‘Bakterium der Maul- und Klauenseuche.” Nies-
sen’s ‘“‘Syphilomyces” (1908) belongs in the same class. Dunbar’s book ‘Zur Frage der
Stellung der Bakterien, Hefen und Schimmelpilze im System” (1907), which apparently con-
firmed many of the old statements of Trécul, Béchamp, Bastian, Hallier, and others, elicited
the following characteristic remark from Pringsheim (1910, p. 139):

Bei solchen Behauptungen fasst man sich an den Kopf, um sich zu versichern, dass er noch auf den Schultern
sitzt. Das ganze Gebiude unserer Wissenschaft droht zusammenzustiirzen.

As I will have to show in Chapter II, the knowledge of the formation and development
of the bacterial gonidia helps to make these so far unexplainable observations quite intelligible.

The standpoint taken by the bacteriologists in other European countries is mostly on the
pleomorphistic side.

In Switzerland Perty (1852) has furnished one of the earliest contributions to the knowl-
edge of the life history of the bacteria, especially concerning their upgrowth from the gonidial
stage. More recently Diiggela (1905) discussed the question of the Ferane arrangement: -of
the bacteria, closely agreeing with Lehmann and Newmann. ”

In Italy the view, once held by Robin, that the different cocci and rods merely represent
stages in the development of Leptothrix buccalis was renewed by Vincentini (1893), to whom
pneumococci and tubercle bacilli were only types of growth of a Leptothrix racemosa. In Naples

 
No.2.) LIFE CYCLES OF THE. BACTERIA—LOHNIS. 15

Schroen made several important investigations, to which we will have to refer later. This: also
will be the case with the work.done by Ferrdn in Spain.

Gedoelst (1899) in Belgium shared the standpoint taken by his French colleagues. Fokker
in Holland took his place at the side of Trécul and Béchamp. He was one of the earliest oppo-
nents (1882) of the constancy dogma of; Cohn and Koch. Recently E. de Negri in Utrecht
discovered the complete life cycle of the Corynebacteria (1916).

In Denmark the work done by Lankester in England was confirmed by. Warming, who
concluded (1876, p. 10):

- Les bactéries sont données en réalité d’une plasticité illimitée, et je crois qu’il faudra renoncer au systéme de M.
Cohn et de'quelqués autres savants, qui caracterisent les genres et les espéces d’aprés leur forme.

E. Chr. Hansen in Copenhagen discovered the pleomorphism of the acetic acid bacteria.
Schmidt and Weis (1902, p. 78) also discussed the inconstancy of the cell form. Jorgensen
(1911) wrote in his textbook (English translation, p. 78):

A single species of bacteria may occur in various forms, and this may happen either dthine the normal growth of
the vegetation, or in growths on different media.

In Norway the strictly monomorphistic side was represented by G. A. Hansen. As late
as in 1903 he still rejected as unreliable all reports on irregular, especially on branched forms
of tubercle and leprosy organisms. On the other hand, Olsen (1897) has furnished very inter-
esting data supporting the pleomorphistic point of view. In his opinion, probably all bacteria,
at least all bacilli, are able to exhibit under favorable conditions a branched, mycelial growth.

Still more important contributions have been made by Almquist and his pupils in Sweden.
He is one of the few authors, who discovered independently the complete life cycles of different
pathogenic bacteria. Much valuable material is contained in a long series of his publications
(1893-1917).

In Russia as early as 1877 Cienkowski has proved convincingly that Cohn’s form genera were
untenable. Winogradsky, on the other side, strongly defended in his early papers (1887-1889)
the strictest monomorphism against Lankester, Warming, Zopf, Metchnikoff, and others, though
he admits that further experimenting perhaps might change the situation. In fact, his own
nvestigations upon the nitrite bacteria (1892) did not support his monomorphistic standpoint.
Within the same species the cell forms varied to such an extent that, as he says—

On croirait sirement avoir affaire 4 des organismes différents. I] n’en est pourtant rien.

In America at first the French standpoint won some influence by G. M. Sternberg’s edition
of Magnin’ s textbook (1884). Later, however, following Baumgarten (1890) the bacteria were
divided | by ‘Sternberg (1896) into two groups: I. Relatively monomorphous; IT. Pleomorphous
organisms. H. L. Russell (1892) found among marine bacteria considerable morphological vari-
ation. Th.Smith (1890-1900) studied the various factors modifying the development of different
pathogenic organisms and took his stand against the reckless ‘species’ making. Hopkins
(1898) isolated apparently the same pleomorphous mouth bacterium which had been studied
in Germany by Biedert (1885). Although it clearly grows either in globular or in rod-like form
according to the environment, the author emphasizes that “it is not positively asserted that a
rod can be produced from a coccus, nor a coccus from a rod.’ H. W. Conn (1900) made some
interesting observations upon the wide physiological variability in pure cultures of micrococci.
N.G. Davis (1901) got a typical coccus from Bac. rosaceus metalloides. Ohlmacher (1902) isolated
from a colon septicaemia a colon bacillus exhibiting a “truly astonishing” pleomorphism.
The task of making a detailed description of all the various shapes assumed by this bacillus, he
declares ‘‘well-nigh impossible.”

In spite of these and other contributions of American bacteriologists, there are, however,
only comparatively few American textbooks which do not adhere to the rigid monomorphism
which had been taught in Germany.

The inconstancy of form and action has been advocated by Chester (1901), to whom (p. 51)
‘tone form appears to merge into another,” and by E. F. Smith (1905), who says (p. 177):

We know that bacteria are much more responsive to changed environment than was supposed by Koch and his
followers in the eighties, and we are prepared to believe anything respecting their origin and their polymorphism which
can gain the suffrage of the great body of critical workers who now cultivate this field.

160266°—21——2
16 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.
On the other hand, A. 0. Abbott, (1902, p..51) emphasizes: — # opt De 4

One can never produce bacilli from micrococci, nor vice versa; and any' evidence which nay” be’ pee to the
contrary, is based upon untrustworthy methods of observation. , 4 ie “TE?

Like to Cohn, Koch, and their followers, to Ball (1908), as well as to Frost ‘and MeCampbell
(1910), a change in cell morphology is synonymous to a enenest 2 Spee or eens Accord-
ingly, statements like these are made: “°° ‘ .

A bacillus does not arise from a micrococcus or ‘the typhoid fever bacillus produce the bacillus of tetanus. s(Ball,
p. 23.) i ;

It is not possible to cause any permant change in the morphology of a bacterium and thus originate a new species
of bacteria. For example, it is not possible to cause a bacillus to change into a coccus and vice. versa.—(Frost and
McCampbell, p. 25. ).,

Though not quite in such dogmatic terms, eens me form i is diss sagonnted -by Jordan
(1909, p. 54; 1916, p. 65):

Under normal and uniform conditions of life each form breeds ee the Zonas forms producing only spheres
and the rods, again, only rods.

Concerning the plague bacillus, however, it is said (1916, p. 321):

Many morphologic variations, coccus shapes'and large rods, are ‘found.

The standpoint taken by Park and Williams (1914 p. 31) is similar:
Micrococci always, under suitable conditions, produce micrococci, bacilli produce bacilli, and spirilla produce
spirilla.
Nevertheless, the variability of the cell form is more adequately discussed in this work.
Of special interest is a picture (p. 294, fig. 119) showing a variety of diphtheria bacilli which grows
exclusively in the form of a large coccus.
Hiss and Zinsser (1914, p. 19) say:

Variations from the basic forms may occur, but are not common among bacteria under normal conditions.

Kendall (1916, p. 21) repeats the statement made by Jordan and adds (p. 23):

Given constant conditions, bacteria growing in a favorable environment exhibit constancy of form and size,
although a few organisms in every culture are somewhat larger or smaller than their fellows, appearing as occasional
giants or dwarfs.

Pleomorphism is said (p. 24) to be “rarely or never met with among the pathogenic bac-
teria.” However, the diptheria bacillus is declared (p. 389) to be “highly pleiomorphic” and
the plague bacillus (p. 408) “very pleiomorphic.”

In recent American literature generally the interest has been centered more upon
physiological than morphological problems. Orla Jensen’s so-called natural system of the
bacteria, almost entirely based on mere biochemical hypothesis, has been accepted, for instance,
by Jordan (1916 p. 197) as “a promising beginning in classifying bacteria.” Accordingly,
morphological questions are not infrequently left without adequate consideration. "When
describing the characters of the different Colon varieties Rogers says, in the first of two papers
published in 1914, that in the morphology of B. colt there is “so little variation that no attempt
was made to use this character in differentiation;” whereas he states in the second: “There is
a wide variation in both the size and the form of the cell, but since these variations frequently
occur within the limits of a single culture they are without varietal significance.” The micro-
scopic tests have been made only once with the cultures when 24 hours old, and the motility
was not determined at all, as it was considered to be too unstable.

Several important contributions to our subject, however, have been made more re-
cently by American authors.. Mogucht (1910) proved for the first time that it is possible
to transform a nonspore-forming lactobaccillus into a spore-forming member of the Mesen-
tericus group, which fact, though quite probable (Léhnis, 1910, p. 200), has never before been
tested experimentally. Maher (1910-1915) furnished some interesting details concern-
ing the complete life cycle of the tubercle bacillus. Unfortunately, there are some
entirely unfounded hypotheses interspersed in his work, which obviously have caused
the valuable observations contained therein to be passed practically unnoticed.

zal
No. 2.]

LIFE CYCLES OF THE BACTERIA—LOHNIS.

17

The very important contributions, which have been made by Rosenow and his coworkers
(1912-1917), deserve special discussion on the following pages, like those of Almquist and of

Hort.

(6) CAUSES OF APPARENT AND OF GENUINE PLEOMORPHISM.

Before proceeding to review the various explanations given by different authors concerning
the much disputed pleomorphism of the bacteria, it seems advisable, to arrange for ready ref-
erence in Table I a list of papers wherein data about the pleomorphism shown by the different
groups of bacteria may be found. As the presence of branching forms has been frequently
considered to be of special importance in regard to the morphology and the systematic position
of the bacteria concerned, observations made in this direction have been mentioned in a spe-

cial column.

 

 

 

 

TaBLeE I.
Organisms, Different cell forms observed. Branching observed.

Meningococcus. Lehmann and Neumann, 1912; Hort et al, 1915-1916.............

Gonococcus. .. . ol) Herz0g 51913 .oc05. osscsstescse cass wt cide ce ccwar yerowececeasecnetey

M. melitensis. . Durham, 1898; Babes, 1903; Gimther, 1906; Saisawa, 1911; Leh-

mann and Neumann, 1912; Jordan, 1916, Evans, 1918.
M. candicans.................-- Lohnis and Smith, 1916..-......-..02 0.22.22 eee eee cece eee eee Loéhnis-Smith, 1916.
M. flavus. ..... .--| Matzuschita, 1900............. .| Matzuschita, 1900,

 
  
   

M. suboornelis.:
M. rubefaciens. .
M. cyaneus......
M. sense
Sarcinae spec..
Urosarcina. . 8
Sarcinastrum..........-.-.2--+
Pneumococcus and Streptoc.

Streptoc. lactis...............-

Poliomyelitis
Enterococcus
Leuconostoc.
Proteus...

Bact. Zo’
B, acidophi

Tackobwallt from milk, cheese,
etc,

B, acidi propionici b.....,.....

B. funduliformis group. ....... Hallé, 1898; Ghon and Sachs, 1905; Ghon, Mucha and Miter,
1906; Kisskalt, 1906.

B. erysipelatos......--.......-- Kar linski, 1889; *Rosenbach, 1909 sictctinxceeiv sauunct de denmecndesane

B. septicaemiae (Pasteurella). - Rosenfeld, 1901; Nowak. , 1908, a Sisise Gye eid wits, a e/avntareimyate-sje.ate

B. pseudotubere. rodentium. . Galli-Valerio, 1903, 1913; Zlatogoroff, 1904

Bact. pestis...............-.---

B. radiobacter and radicicola..

B. rubiacearum..............-.
B. tumefaciens...
Acetic acid bacteria.
Photobacteria. .
Bact. solare..

B, prodigiesu
B, violaceum..........-.-.-.-.
B, fluorescens and pyocyaneum|

  
 
  
 

 

Ps, cerevisiae............-.-.-- Fuhrmann, 1906
B. ferrugineum...............- Rullmann, "1897-1900... ..
Nitrosomonas............--.--- eee 1892......--
Nitrobacter.............2...--- Winogr: BOL cosckodasadiies
B. mycoides. . Olsen, 1897; Nadson and Adamovic, 2
B. anthracis E. Klein, 1883, 1885; DeBary, 1884; Braem, 188
Crookshank, 1896; ’ Matzuschita, 1 1900; Ligniéres and Dirrieu,
1902; enri, 1914.”
B, subtilis Klein 1885; Léhnis and Smith, 1916
: thrix. .| Duclaax, 1883; Winkler, 1895 2
B. cohaerens. GOtth ei]; 19OM. cs si.cc-scnsyaier sesiousenien semsnes SY ENCORE Het neeses sca

aid Matzuschita, 1900 si2:csciaeicerenejsies
a povernek, AQ «, cesiciarscients ois
Lehmann

71.] Lehmann and Neumann} 1912; Léhni:
.| Beijerinck, 1901; Miquel and Cambier, 1902 oo js cisssicice asd,e

 

Rosenow etal. ,do16-1917; Nuzum, 1916; Mathers, 1917
.| Thiercelin, 1899-1903

.| Van Terhem,
-| Hauser, 1885;

   

 

 

Hopkins, 1898................-

   

and Neumann, 1912.

     
 

sand Smith, 1916.

Lagerheim, 1900.
Pansini, 1890; Kruse and Pansini, 1892; Arloing and Chantre,
1894; Babes, 1895; Crookshank, 1896; Stolz, 1898; Ohlmacher,
190: 2; Taddei, 1909; Broadhurst, 1915, Mallory ‘and Medlar,

1916,
Billroth, 1874; Maddox, 1885; Burri, 1898; Diiggeli, 1905; Léhnis,
; Wo,

  
 

1879 c; Ludwij
jacger, 1892;
zuschita, 1902;

Kurth, 1883;

Sternberg, 1898; Moro, 1900; ‘Tissier, 1900; Cahn, 1901; Rodella,
inna Isus. Sandberg 1904; ‘Noguchi, 1910; Blihdorn, 1910; Ber-
Tan

Hemneberg, 1901; Holliger, 1902; Freudenreich and Théni, 1905;

Diigeelt ; Léhnis, 1907; Heinemann and Hefferan, 1909;
Ru insky, 1910; Chatter 1910; Koegel, 1914.
Freudenreich and Jensen, 19

 

 

vm 1896; Hlava, 1902
ohlbrugee, 1901; H6fli

 

‘| Albrecht and’ Ghon, 1900; Skschivan 1900; Cacace, 1903; Dieu-
donné, 1903; Galli-Valerio, 1903; Ziatogoroff, 1904; Rowland,
Crookshank, 1896; Grassberger, 1898; Davis, 1907................-
‘| Bordoni-Ufireduzzi, 1888; Simoni, 1900; Jehle, 1902; Lepeschkin,
1904; Léhnis, 1905; Péju and Rajat, 1906;. Toenniessen, 1913:

Ward, 1917,

Haslam, 1898; Adami et al., 1899; M. E. Abbott, 1900; Matzus-
chats OhImacher, 1902; Péjuand Rajat, 1906; ‘Wilson, 1907;
Revis, 1912; Kellerman and 8 cales, 1916; Prell, 1917,

Péju and Raj at, 1906; Hata, 1908; ‘Almquist, DONT o:. aise Sie s

Almquist, 1! 93; 1894, 1916; ‘Walker and Murray, 1904; Péju and
Rajat, 1906; Wilson, 1907.

Beijerinck, 1388, 1890; Prazmoski, 1890; Hiltner and Stérmer,
1903; Léhnis, 1908; Rossi, 1907; Harrison and Barlow, 1907:
Georgewitch, 191!

Faber, 1912; Géorgevite, DOIG os wrelecriare Pactacsvind alee arate Sadnimemcien

E. F. Smith, 1 QUT eeencca ecw wines emer sinaie siesineisssavinielec neta cram nicic

Hansen, 1879 1894; Zejdler, 1896; Henneberg, 1897, 1898; Janke, 1916.

ie erinck, 1889; Barnard, 1899; Molisch, 1903, 1904; Reinelt, 71905.

olff, IROB ise oh gid aatahenscse uae bagatecocunene sagas

Hashimoto, DBD oan wis ei oitiaiaiibie weiariawinne sitrociare rae wien

prasserzug, 1888; Slater, 1891; Davis, 1901............. 2
Zopf, 1885; ‘Harrison and Barlow, 9005 osc crengee seedousen meat aser

Guignard ‘and Charrin, 1887; Schiirmayer, 1895; Wassermann,
1903; Wilson, 1907; Léhnis and Smith, 1916.

 

 

 
 

 

 

 

i Matzuschita, 1900.

Babes, 1896; Seitz, 1896; Crookshank, 1896;
Stolz, 1898; Vincent, 1902, Kraskowska and
Nitsch, 1918.

‘ Matzuschita, 1902; Stefansky, 1902.

pete aes 1887.
Moro, 190; Tissier, 1900; Cahn 1901; Bliihdorn,

Freudenreich and Th6ni, 1905; Heinemann
and Hefferan, 1909; Rubinsky, 1910.

Hallé, 1898; Ghon and Sachs, 1905.

Rosenbach, 1

Rosenfeld, "Psa, 1902, Nowak.

Galli-Valerio, 1

Albrecht and’ Ghon, Skschivan, Cacace, Dieu-
fo tons Galli-Valerio, Kodama, 1908; Row-
an

Grassberge +1898,

Barone Fareauas, 1888; Lepeschkin, 1904;

Walker and Murray, 1904; Wilson, 1907.

Hata,1908.
Gamaleia, 1900; Casagrandi, 1901; Almquist;
‘Walker and Murray, 1904; Wilson, 1907.
Be Dawson, 1899; Hiltner and Stormer;
Léhnis; Harrison and Barlow, Conn, 1909.

Faber, Georgevitch.

E. F. Smith, 1911.

Henneberg, 1897, 1898.
Belerinok 18 , 1889; Barnard, 1899.

""| Hashimoto, 1309,
.| Slater, 1891,

Harrisonand Barlow, 1905.
Gamaleia, 1900; Wilson, 1907; Léhnis and
Smith, 1916

-| Rullmann, 1897-1900,

-| Olsen. 1897,

Lignieres and Durrieu, 1902; Henri, 1914.

Léhnis and Smith, 1916.
Gottheil, 1901; A. Meyer, 2901.
18

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

Taste I—Continued.

[Vou. XVI.

 

Organisms.

Different cell forms observed.

‘

Branching observed.

 

B. mesentericus
B. megaterium
B. tumescens
B. oxalaticus
B, malabaren:
B. azotobacter

 
 
 
 
  

Other aerobic spore-forming
bacilli.

B. alvei
B. amylobacter.....--..-----..

B. Chauvoei

B. oedematis...........-.---..
Cellulose bacilli -
B. tetani
Vibrio cholerae

V. Cardiissasecseeee
i proteus Buchner.

roteus ue Moulins
¥. peenate

   
   
  
 
 
   
   
  
  

Sp. endoparagogicum
Spir. bataviae .
Spir. pyogenes.
Sp. tyrogenum. .
Spirochaeta pallida

 

Dp.
Sp. icterohaemorrhagica.
Spiroch. suis....-..-..
Abortus bacillus.....
Necrosis bacillus . ...
Streptob.
Glanders

B. diptherias

Diphtherold bacilli

B. leprae

B. tuberculosis

Pseudo-tuberculosis bacilll....

Mycobacterium spec..'.....-.-.
Actinomyces spec

Chalothrix

Beggiatoa
Crenothrix. . .. :
Clonothrix. .....----..-----+-+

 

ag Prazmowski, 1886;
‘| Kanthack and Connell, 1897; ‘Hibler, 1908

 

‘} Finkler and Prior, 1884, 1885; Firtsch, 1888; Fuerst, 1914. .
-| Kohlbrugge, 1900......-..-.-----+- 220+ ee eee reece eee

-.| Weibel, 1888. -
2 Weibel, 1888...

| Tizzoni and

 

Léhnis and Smith, 1916
DeBary, 1884; Stoklasa, 1898; Lehmann and Neumann, 19;
Zopf, 1883; A. "Koch, 1888; Garbowski, 1907
qonnis and Pillai, 1907.
Beijerinck, 1901; H. Fischer, 1905; sand Wes
zemieniewski, 1908; Peklo, 1910; PraatnoWsi, 1912; ‘Léhnis
and Hanzawa 1914; Lohnis and Smith,

 

916.

Pansini, 1890; ‘Russell, 1892; Niessen * Te08: RE an 1901;
Miquel and Cambier, 1902; Neide, 1904; Maassen, 1904; Gar-
bowski, 1907; Proca and Danila, 1910; Viehoever, 1913;
Léhnis and Smith, 1916.

Harrison, 1900-2250 sseeeecewiciceea ares Se sweeacediocaeceeceneee ae

em, 1884; Schatten-
1802: Grassberger,

Prazmowski, 1880; Buchner, 1882; Van Tie
froh and Grassberger, 1960; Winograds
1902, et Grassberger and Schattenfroh, 19

Ehlers, 1884; Fraenkel and Pfeiffer, 1895; Pini and Galli-
Valerio, 1895; Ghon and Sachs, 1903; Grassberger, 1903, 1905;
Grassberger and Schattenfroh, 1907; Hibler, 1908.

Grassberger, 1903; Ghon and Mucha, 1905

Omelianski, 1902

 

E. Klein, 1885 a; Ferran, 1885; Ermengem, 1885; Dowdeswell,
1889, 1890; Friedrich 1392; Metchnikoff, 1894; Gruber, 1894
Fraenkel and ese er ge Cunningham 897; N
1900; Gamaleia, 1 olle and Gotsehlich, 1903; Almquist,
1904, 1907; aaron 1906; Baerthlein, 1912;, Lehmann and
Neumann, 1912; Stamm, 1914; Popoft- Tscherkasky, 1914;
Olsson, 1915.

E. Klein, 1905.

  
 
 
  
  
 
 
 
  
 
 
 
 
  
  
 
     

ishi, 1901..
Maassen, aaua ¥

Weibel, 1
Teutecier, 1895; Lehmann and Neumann, 1912. -
Meirowsky, 1914.

Doerr, 19
Meirowsky, 191
Wechselmann and Loewenthal, 19 + Sob
pealoiey and Siedlecki, 1908; Meirowsky,
Handle, 1911

  

ingeit

Th. Smith, a lis, out 1895; Marx, 1899; Galli-Valerio, 1899,
1900; Conr: , 1900.

Zarniko, 1889; "Escherich, 1894; Zupnik, 1897; Madsen, 1897;
Meyerhof, 1898; Kurth, 1898; ‘Wesbrook et al., 1900; Ohl-
macher, 1902; Beck, 1903; Spirig, 1903; Goodman, 1908; Smir-
now, 1908; Dale, 1910; Balfour, 1911d; Morse, 1912; Baerthlein,
1913; Bernhardt and’ Paneth, ‘1913; Park and Williams, 1914;

Heinemann, 1917; pone, 1918
Escherich, 1894; Seitz, 1 : E. Klein, 1900; Nakanishi, 1900;
forse, 1912; Bernhar dt na Paneth, 1913; Trautmann and
Gactneros 1913; Negri and Mieremet, 1913; Lanford, 1914;
E. Negri, 1916; Walker and Adkinson, 1917; Mellon, 1917.

Babes, 1883; Lutz, 1886, Bordoni-Uffreduzzi, 1888; Levy,
1897; Cza, lewski, 1898; Teich Aes pen 1900, 1902
Redrowski, 1901, 10; Babes, 1907; T. §. Williams, 1911; Galli-
Valerio, 1912; Réenstjorna, tog, "Wolbach and Honeij, 1914.

Klebs, 1883; Babes, 1883; Z opt, ; Malassez and Vignal, 1884;
Lutz, 1886; Metchnikoff, ; F Coral and Babes, 1890; F.
Fischel, 1893; Coppen-Jates 1893, 1895; Bruns, 1895; Semmer,
1800; Olsen, i897; Babes and Levaditi, 1897; Marpmann, 1897;
Ferran, 1397; Schumowski, 1898, 1899; Schuermayer, 1898:
Schulze, 1809; Lubarsch, 1899; Arrig 0, 1900; Courmont and
Descos, 1902; "Herzog, 1903; Wolback and tat 1903; Haw-
thorn, 1903; Schroen, 1904; Piery and Mandoul, 1904; Arloing,
1908; Betegh, 1908; Much; 1908, 1909; Maher, 1910-1915; Kry-
loff, ‘1911; alli- Valerio, 1912; Wherry, 101

| Preisz, 1894; Korn, 1899; Moéller, 1899; steel 1899; Bongert,
1901; ‘Abbott and ee, 1902; Courmont and Descos,
1902' Schabad, 1904; Sanfelice, 1905.

Sohngen, ee, Por emesie Se al., LOT nocitele wins jecectenswcccas

Bollin: 877; J. Israél, 1878; 0. isrséi, 1884; Nocard, 1888;
Mc eaten et BaahGen 1890; E pinger, 1890; Kedzior,

1896; Rullmann, 1896, 1898; ‘Lachner- andoval, 1898; Schuer-

mayer, 1900; Sil erschmidt, 1901; are 1902; Feistmantel,
1902; MacCallum, 1902; Doe ep ke, 1902; N ma, 1006; Hol-
landt, 190 06; Loele, 1908; Claypole, 18 th. Cohn, ’ 1913;
Krainsky, 19

Miller, 1883; Zop f, 1883, 1885, Dobrzyniecki, 1897; Beust, 1907.
aoe eo Zep) , 1881, 1882; Billet, 1890; Biisgen, 1894; Migula,
; s,
Zopf, 1881-1883, 1895; Engler, 1882; Billet, 1890
zor 1879; Schorler, 19
rler, 1904

 

 

  
 
 

Léhnis and Smith, 1916.
Gamaleia, 1900,

Hi Fischer, 1905; Léhnis and Westermann,
908; Léhnis and Smith, 1916.

Niessen, 1898; Maassen, 1904; Léhnis and
Smith, 1916. i
Harrison, 1900.

Ghon and Sachs, 1903.

Grassberger; Ghon and Mucha.

:| Kanthack, 1896; Tavel, 1898.

Dowdeswell, 1889; Metchnikoff, 1894; Kohl-
brugge, 1901; Shibayama, 1902, Hammerl,
1906; Stamm, 1914.

:.] Kohlbrugge, 1901.

Bonhoff, 1896.

-| Maassen, 1904.

“_| Bajardi, 1903.

Kutscher, 1895.

Te en Meirowsky, 1914.
Sorokin, 1

Faber, 19 ae

Doerr, 1905.

3 Meirowsky, 1914.

Meirowsky, 1914.
Meirowsky, 1914.

..| Meirowsky, 1914.
:.| Inada et al., 1916.
-| Preisz, 1903.

Schmorl, 1891; W. Ernst, 1902.
Tizzoni and Angelis.
Semmer, Marx, Galli-Valerio, Conradi.

E. Klein, 1890, 1894; Fraenkel, 1895; Bernheim
and Folger, 1896; "Kantha ck, 1895; Meyerhof,
1398; Kurth, 1898; Hill, 1899, 1902; Cache,
1901; Concatti, 1901; Beck, 1968; Cappellani,
1910; Park, 1914; Bergstrand, 1918.

Babes, 1895; Kanthack, 1896; Nakanishi, 1900.

Czaplewski, 1898; Kedrowski, 1901, 1910; T. 8.
illiams, 1911; Johnston, 1917.

Motchnikoif, 1 1888; E. Klein, 1890, 1894; Mafucei,
1992; F. Fischel, 1893; Coppen-Jones, 1893,
1895; Semmer, 1495; Kanthack, 1896; Olsen,

; Marpmann. 1897; Schuermayer, 1898;
Dorset, 1901; Wolbach and Ernst, 1903;
Sehroen, 1904; Miehe, 1908; Galli-Valerio,

Korn, 1899; Schabad, 1904; Sanfelice, 1905.

Séhngen, 1913; Chantemesse et al., 1917.
All authors mentioned.

Dobrzyniecki, 1897.
Cohn, 1875; Zopt, 1882; Migula, 1900

Schorler, 1904.

 
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 19

There is no doubt, and the facts to be discussed later will amply support the statement,
that a wide pleomorphism has been observed within all groups of the bacteria by numerous
authors. And if at the present time several authors still firmly defend the morphological
constancy and simplicity of the bacteria, arid do not hesitate to declare beforehand that ‘any
evidence which may be presented to the contrary is based upon untrustworthy methods of
observation,” we are reminded of many cases in the history of science where newly discovered
facts have been strongly opposed by scientific dogmatism. When, e. g., Sir Thomas Brown
first attacked the time-honored dogma of the spontaneous generation of mice and other vermin
from putrefying material Alecander Ross commented on this point: “To question this is to ques-
tion reason, sense, and experience” (Frost and Me Campbell, p. 9). Oznam wrote still in 1820

against the theory of the “contagium animatum,”’ though it was supported by men like Kircher,
Vallisneri, Réaumur, Plenciz, and Frémont: “Ich will keine Zeit verlieren, diese abgeschmackten
Hypothesen zu widerlegen”’ (Loeffler, 1887, p. 11). Pineau’s (1845 a) remark that frequently
the facts recorded “sont entachés d’un esprit de systéme qui perce dans les resultats”’ fits also
remarkably well to many a report on microscopic studies done from a monomorphistic point of
view.

But a great deal of the opposition which has been raised against the pleomorphistic theory
was much more based on misunderstanding than on adverse facts. The old mistake to consider
the acknowledgment of the occurrence of different cell forms within one species as equivalent
to an acceptance of the theory of spontaneous generation or to an inclination to cancel all
differentiation of species, still takes a prominent place in some textbooks despite its absurdity.
In many other cases a different use of the word pleomorphism has been responsible for much
useless debate. Migula (1897, p. 229; 1904,.p. 47) and Schmidt and Weis (1902, p. 86) want
to have the term restricted, as had been done by DeBary (1884, p. 136) for the fungi, to such
cases where the complete life history of a bacterium is composed of several subcycles character-
ized by different morphology. Occurrence of any of the so-called involution forms should not
be taken as indicative of a pleomorphous character, according to these authors. To Migula
(1904, p. 37), however, involution forms are all cell forms differing from the normal appearance
of the species or variety concerned, irrespective of their origin (“ohne Ricksicht auf das Zustande-
kommen dieser Abweichung’’). Such reasoning naturally would never allow the discovery of
any pleomorphism, how evident it may be. Also, in this case it is “le systéme qui perce dans
les résultats.”

Hueppe (1886, p. 38) points out that pleomorphism and variability of the cell form should
be kept entirely separate. Rodet, on the other hand, in his book entitled ‘‘La Variabilité dans
les Microbes’’ (1894, p. 38), speaks of ‘‘pléomorphisme’’ when temporary, of “ variation’? when
constant variations of the form are observed. The occurrence of involution forms is character-
ized by him (p. 39) as “pléomorphisme monstrueux ou morbide.” In Pringsheim’s book on
“Variabilitat niederer Organismen’’ (1910, p. 25) pleomorphism is taken as part of morphological
variability in about the same manner as advocated by Migula.

Considering such inconsistencies and differences of opinion, it seems best to use—at least
at the present time—the term pleomorphism in the common though somewhat loose sense,
merely indicating that various cell forms have been observed with one or the other species of
bacteria. In face of the unsatisfactory situation of the whole problem it seems useless to ascribe
to the term a sharply defined meaning, while in too many cases we have not yet secured the
correct interpretation of the facts observed.

When explaining an apparent or a real pleomorphism of a bacterium one or the other or
several of the following possibilities will have to be taken under consideration:

(1) Contamination.

(2) Incorrect classification of the pleomorphous organism among the bacteria.

(3) Presence of abnormal (“‘involution’’) forms.

(4) Occurrence of variations and of mutations.

(5) The normal development of different forms in the various stages of the complete
life cycle of a species.
20 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

(1) CONTAMINATION.

To Gaffky (1881), as to many other followers of R. Koch, Nes change of a culture in form
or virulence was an indisputable proof of ‘‘contamination.” Of course, in many cases this
suspicion was and will be fairly well founded, and the practical aims of the bulk of bacteriologi-
cal work done will often justify a speedy discarding of such cultures without further test.
However, as far as scientific investigations come into view, this widespread practice should not
be followed too readily. At least in research laboratories the benevolent institution of the
“suspended sentence”’ should be granted in some cases. Already in 1885 Gruber quite correctly
questioned the dogmatic standpoint taken by R. Koch and his pupils, asking:

Muss man wirklich sobald man eine Aenderung der geziichteten Bacterien wahrnimmt . . . die unreine
Kultur fortwerfen?

Our own studies upon the life cycles of the bacteria started originally from two apparently
contaminated cultures of Azotobacter (Léhnis and Hanzawa, 1914). And it is hoped that this
review of the literature will help to stimulate a somewhat more cautious and more scientific
treatment of this special point whenever feasible.

Hort (1916 a) says that the neglect of bacterial morphology has ‘‘led many a worker to
assume that perfectly genuine examples of pleomorphic activity are merely examples of con-
tamination,” and “a careful scrutiny of alleged contaminations is absolutely fundamental.”

When Winogradsky (1888) contested Zopf’s (1881-1885) standpoint concerning the sulphur
bacteria, he, like Zopf, drew his conclusions from mixed cultures, wherein he, like Zopf, saw
all kinds of different and intermediate forms. But he did not hesitate to declare all these
to be different species, and the monomorphistic view prevalent at that time did neither ask
for proofs, nor did it pay any adequate attention to Zopf’s answer (1895), which hardly ever
has been quoted.

When Gasperini (1889) first studied the pleomorphism of the Actinomycetes (Streptothriz
Foersteri), the preparates made up of coccoid forms, short and long rods, threads and screws,
appeared to him like “une culture de bactéries fortement impure,’”’ and the interesting Acti-
nomyces (Cladothrix) asteroides found by Eppinger (1890) in a case of pseudotuberculosis
(cladothrichica) was at first thrown away as contamination. Kitt (1897-1898) discarded as
“contamination” the Streptothrix growth which he obtained in virulent cultures of B. ery-
sipelatos suum. No word is said about an adequate test, and analogous results secured by
Rosenbach (1909) make it indeed very probable that also here the facts were vanquished by
the theory. H. U. Williams (1912) isolated from a case of tuberculous pericarditis a spore-
forming bacillus and a “‘Streptothrix,” also spore-forming. Only after years a “separation”
could be achieved, and then the bacillus frequently became so similar to the “Streptothrix”
that the author further remained in doubt whether both are really different species or only
stages of one organism. When I (1905 a) first studied the various morphological changes of
B. pneumoniae, lactis viscosum, radiobacter, and radicicola, it also took some time before I
was satisfied that all these alterations were indeed not the result of contamination. The close
parallelism of the results obtained with the various cultures helped much to solve the problem.

With the Coryne- and Mycobacteria such puzzling cases are especially frequent. To
Spirig (1903) we owe a thorough discussion of these questions as relating to the diphtheria
organism. A “contamination” with a “Streptothrix,” first (1899) considered by him to be
a type of growth of the diphtheria bacillus, led to a close study of the subject (1903, p. 448).
Duval (1914), like G. A. Hansen (1908), thinks that the lepra organism must invariably repre-
sent itself is a simple, constantly acid-fast bacillus. The now very numerous findings which
strongly militate against this assumption had therefore to be discredited. The following
quotation is very characteristic (Duval, 1914):

In the cases showing diphtheroids and streptothrical forms and acid-fast rods, the recognition of these as separate
and distinct species, or the determination of any one or all of them as contaminators was often extremely difficult.
So much so that in these cases the cultures were discarded.

Harris and Wade (1915) found diphtheroid organisms to be a rather common occurrence
not only in and on the body, but also in air, water, etc. The strains isolated by them were
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 21

(apparently) quite constant. Accordingly, Kedrowski’s very extensively. studied leprosy
cultures, as well as others of pleomorphous character, are declared to be mixtures of different
organisms. Bayon (1915, p. 29) remarks very aptly that to discard in leprosy studies all
diphtheroids because of their ubiquity—

would considerably simplify pathology. The microérganisms of syphilis, of tuberculosis, of cholera could also, on
similar ground, simply be ruled out of court as far as their etiological significance is concerned.

The most recent studies upon the variations and mutations, shown by the bacteria, have
helped to make one point clear, which is. very important in this connection. As was
probably first emphasized by Firtsch (1888), it does not infrequently happen that changes
shown by some cells of a pure culture may persist for a more or less considerable length of
time. Such cases naturally are most liable to be mistaken for contaminations. One of Bern-
hardt’s publications (1915) furnishes interesting details relating to this point.

Usually, however, proper experimenting will soon succeed in bringing about such retro-
gressive changes which will settle the question. Otherwise, parallel tests will be very helpful.
If they are incorporated in bacteriological routine work as they are used, e. g. by the chemist,
the decision concerning presence or absence of contamination becomes not too difficult, provided,
of course, that reliable methods are judiciously used.

(2) INCORRECT CLASSIFICATION OF PLEOMORPHOUS ORGANISMS AMONG THE BACTERIA.

The second possibility mentioned above, i. e. the assumption of an incorrect classification
of an organism as a bacterium, has also been used quite extensively to eliminate pleomorphic
organisms, rightly or wrongly.

F. Cohn (1872b, p. 139) stipulated that a bacterium can only multiply by fission. As
true branching necessarily implies budding, it was declared by him never to occur with bacteria,
despite the positive findings of Dujardin (1841) and H. Hoffman (1869), which were soon after
confirmed by Lister (1873). The monomorphistic theory, as it developed on the foundation
laid by G@ohn, necessarily enhanced this stipulation. Undoubtedly true branching was often
declared to be ‘“‘false’”’ branching, or if this was not feasible, the occurrence of branched forms
was frequently considered to be sufficient reason to declare an organism a fungus, though it
otherwise showed all characters of a genuine bacterium. The heavy weight of dogmatism is
clearly demonstrated, e. g. in Eppinger’s paper on his Cladothrix asteroides. As we know now,
this organism is much inclined to exhibit true branching, and this was quite correctly described
by Eppinger (1890) in the following manner, based on continuous microscopic observation:
“A short thread first bends itself, a small granule becomes visible at this place, this develops
into a little wart and further into a lateral sprout.’’ Yet, curiously enough, the author calls
this process a pseudo-ramification.

The tubercle bacilli were the first to show clearly true branching so frequently that it
could not be simply discarded as “false.” Metchnikoff (1888 a) separated them, therefore,
from the bacilli and gave them the new name Sclerothrix Kochii. F. Fischel (1893) declared
the tubercle bacilli to be the parasitic growth of a pleomorphic organism, probably belonging
into a genus of higher fungi, an assumption which had been made before by E. Klein (1890)
in regard to tubercle as well as to diphtheria bacilli. Coppen-Jones (1893-1896), who first
was of the opinion that a special fungus accompanied the tubercle bacilli, later proposed the
new genus name Tuberculomyces, and thought that also other pathogenic bacilli might be merely
a type of growth of some higher fungi. Conradi (1900) declared the glanders bacillus to be a
fungus, and wanted to have all bacteria which would show branching incorporated among
the Hyphomycetes. Levy and Klemperer (1900), Cornet and Meyer (1903), Jordan: (1909),
Carpano (1913), and others took a similar standpoint. ‘Lehmann and Neumann (1912) empha-
size the many points which suggest a true natural relationship of the diphtheria, glanders,

tubercle, and leprosy organisins, including their so-called pseudo forms, with the Actinomycetes,
‘and consider the whole group either as a connecting link between bacteria and fungi, or as some
lower Hyphomycetes. .
22 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The latter question has always been answered in a different:manner. Bollinger (1877),
Gasperini (1889), Domec (1892), Lachner-Sandoval (1898), E. Levy (1898), Levy and Klemperer
(1900), Neukirch (1902), A.’ Fischer (1908), Miehe (1908), and others consider the enon
cetes to be true fungi. Kruse (1896 d), on the other hand, says: .

Es kann kaum einem Zweifel unterworfen sein, dass die Ahnlichkeit der Streptotrichen mit den Saiean ive:
nur eine dusserliche ist, wihrend sie den Bakterien nahe verwandt sind.

Meyerhof (1898), Haas (1905), Peklo (1910), and Th. Cohn (1913) also are not in favor
of uniting Actinomycetes and fungi. Bostroem (1890) concludes from his extensive study of
Actinomyces hominis, that this organism—though in the paper always called ‘Pilz” (fungus)—
should be placed among the lower algae, next to Cohn’s Cladothrix. Data concerning the
reproductive organs, which will be discussed in Chapter II, indeed strongly indicate a natural
relationship of Actinomyces, Cladothrix, and Crenothrix. Cunihe (1906), like Fraenkel (1891),
still adheres to the older view, placing Cladothrix, Crenothrix, and Beggiatoa among the lower
algae. Zopf (1881-1882), Lachner-Sandoval (1898), A. Fischer (1903), Ellis (1912), and others,
however, prefer to accept them as higher or trichobacteria. The various facts concerning the
formation and development of the reproductive organs (to be discussed i in Chap. IT) support
this view.

When considering these problems it must be kept in mind that fie characters which
have been accepted for a long time as quite specific for the Actinomycetes and related organ-
isms—viz., budding, branching, and apical growth—are by no means restricted to this particular
group, but are quite common in all groups of the bacteria. This relationship is enhanced on
the one side by the occurrence of motile stages within the life cycles. of Actinomycetes and
trichobacteria, on the other hand by a distinct similarity of the reproductive organs, formed
by these organisms, as well as by the so-called true bacteria.

. The fact that all bacteria show budding unquestionably militates somewhat against the
time-honored conception of the “‘Schizomycetes,’”’ but it can not longer be justly disputed.
Bacteria. multiply in their: ee state not only by ‘fission but to some extent also, by
budding.

That budding and fission are bye: no means so different as is usually emphasized m bac-
tariological textbooks had been already clearly pointed out by Charles Darwin (1868), who said
(Vol. II, p. 430) that both are “essentially alike.” They. pass into each other almost imper-
ceptibly. And he quotes Husley, who wrote: “Fission is little more than a peculiar mode
of budding.”

The data collected in Table I (p. 17) showed that true branching has been actually observed
with representatives of all groups of the bacteria. The smallest number of positive findings
had to be recorded, quite naturally, with the cocci; but with these organisms distinct budding—
the beginning of- all true branching—has been found to be quite common. As early as in 1874
Billroth noticed that lactic acid streptococci exhibit occasionally ‘ ‘yeast-like, branching”’ (p. 73).
Hueppe (1891) thought that the budding forms which he found with cocci, rods, and threads,
might indicate some relationship to Torula and Saccharomyces. Kruse and Pansini (1892),
as well as Adametz (1895), brought further proof concerning the large yeast-like forms pro-
duced by streptococci. Garnier (1907) named one species explicitly B. moniliformis. The
so-called bottle bacillus of Unna, though it usually grew in the form of a coccus of 1 » diameter,
and also showed culturally all marks of a genuine bacterium, has been removed by Dold (1909-
1910) under the name of Dermophyton, Malassez to a place between Hyphomycetes and Blas-
tomycetes. Meirowsky, as well as Kraus (1913), are much inclined to see some “yeast” in
it, on account of its budding. £.de Negri (1916) developed a similar culture of “ Blastomycetes”
(up to 54X14 ») from a typical Corynebacterium. . Hort (1917 b) placed the meningococcus
among the Hemiascomycetes on account of its gemmation and multiple endosporulation.
Laurent (1891) suspected the budding and branching nodule bacteria to be some form of a
yeast or a Ustilago. The same assumption was made by Calmette (1893) in regard to a |

 

1 Metchnikoff (1888 a) wrote 30 years ago: ‘‘Aus den mitgeteilten Thatsachen kann man u. A. sehen, dass die Vermehrung durch Knospung der
Familie der Bacteriaceen keineswegs so principiell abgeht, wie es gewéhnlich angenommen wird.’
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 23

spirilliform organism found by him in cases of typhus, because it grew temporarily in a yeast-
like form. And Ferrén (1885) saw in the cholera vibrio a type of growth of his so-called
Peronospora Barcinonae.

The fact that probably all bacteria, especially all bacilli, show true branching has apparently
first been pointed out by Olsen (1897), who was also of the opinion that bacteria represent
some type of growth of Hyphomycetes. Exactly the opposite standpoint was taken by E.
Wolff as late as in 1898. Even in the case of branching tubercle, diphtheria, and glanders
bacilli he was inclined to see only a casual apposition of normal rods. Gamaleia (1900, p. 18)
stated that all bacteria are able to produce branched forms; but no details are given. That
he was right, however, is sufficiently proved by the facts collected in Table I (p. 17).

Lachner-Sandoval (1898, p. 20) thinks that it is “‘impossible” to classify an organism
among the bacteria if it is able to exhibit true branching:

Die Unterbringung eines typisch verzweigten Organismus unter die Bakterien ist eine Unméglichkeit.

If this standpoint would be generally accepted, no bacteria would be left at all, and all
discussion upon their pleomorphism would become superfluous. However, I believe that there
are important scientific, as well as practical, reasons to leave the bacteria in their present place.

Budding, branching, and apical growth alone are undoubtedly no sufficient reasons to
advocate any change in their systematic position. In regard to the last-named fact it may
be emphasized that apical growth has been always observed at least with all sporulating bacilli
during their germination. A closer study of the growing bacteria, under less artificial con-
ditions than customary, in all probability will reveal more interesting details in this respect.

It is not to be denied that these facts enhance to some extent the relations existing between
bacteria and fungi. But the relations existing between bacteria, algae, and protozoa should
not be overlooked. All bacteria are able to assume more or less the microscopical appear-
ance of an Actinomyces. It has been pointed out why this organism will better take
its place at the end of the bacteria, where their line leads over into the realm of the fungi, than
among these organisms themselves. That branching occurs more often in young and vigorous
than in old and weakened cultures has been shown by F. Fischel (1893), Lubarsch (1899),
Schulze (1899), Migula (1900), Concetti (1901), Lepeschkin (1904), Heim (1906), Cappellani
(1911), and others. The opposite statement made by Géinther (1906) can not be accepted as
correct. This lends support to a hypothesis, proposed by Claypole (1913), that the Actinomy-
cetes are “representing the ancestral-type that gave rise to both the higher fungi and true
bacteria.”’

It has been mentioned above and will be discussed more completely later that a rather
close relationship between streptococci and diphtheroids becomes more and more apparent.
While the latter in their Actinomyces-like growth approach the fungi, the former have always
been considered to be a distinct indication of the relations between bacteria and algae (Nostoca-
ceae). Azotobacter furnishes a similar case. In its typical form it resembles the Chroococca~-
ceae very much. Its spore-forming growth, however, is like that of other bacilli, easily to be
induced to assume themi croscopical appearance of an Actinomyces. Crenothrix, also, exhibits
clearly this double relationship.

Motility, as well as the formation of endospores, may be taken, as has been done by Hueppe
(1891) and others, as an indication of phylogenetic relations existing between bacilli and flagel-
lates, and it is not to be denied that much of what has to be discussed in Chapters II-IV con-
cerning other modes of bacterial reproduction is duplicated by analogous facts in the life history
of protozoa. Cytological reasons, too, have been repeatedly mentioned in favor of relations con-
necting bacteria and protozoa. H. L. Russell (1892) obtained distinctly monad-like forms from
‘typical rod forms of marine bacilli. Especially in the case of his Bac. halophilus (a small motile,
asporogenous bacterium) he studied those transformations in the hanging drop successively.
Most of the large forms were motile, and it seemed probable to him that monads occurring in
the water, but not growing on artificial substrates, were merely types of growth of this bacillus.
Fusiform bacilli and spirochaets have been claimed frequently by protozoologists, though also in
this case the bacterial nature is well discernible. ~
24 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou.-XVI.

s

On the other hand several resemblances of Schizosaccharomycetes and large bacilli have been
pointed out by Lepeschkin (1903).

Taking all these facts under consideration, it seems well justified to retain the bacteria in
their entirety in their present position, and not to advocate the removal of one species or the
other on account of its being inclined to show true branching or another kind of distinct pleo-
morphism.

(3) ABNORMAL (“INVOLUTION") FORMS.

The third possibility of making pleomorphous bacteria agree with the monomorphistic
theory is to declare all “illegitimate” forms to be involution forms, teratologic, crippled, hyper-
trophic, or otherwise abnormal growth. Only with due recognition of the more or less practical
aims of most of the work done in bacteriology it becomes somewhat conceivable how it was pos-
sible that so very many unscientific statements have been made in regard to this point.

Lafar (1887, p. 34), Gotschlich (1903, p. 43), Bergey (1907), Jordan (1909, p. 55), and others
have correctly emphasized that unusual forms should be termed involution forms only when
their degenerate condition is clearly demonstrated by their growth, propagation, and motility,
being nearly or completely at a standstill and their cell content usually showing vacuolization
granulation, and reduced stainability. .A. Fischer (1903, p. 47) states that real involution forms
are either dying or dead.

Frequently, however, the term is more or less grossly misused. I mentioned that Migula
(1904, p. 37), for instance, calls involution forms all forms which he considers to be abnormal
irrespective of their origin and of their vitality. Levy and Klemperer (1900, p. 25), Maassen
(1904), and others take a similar standpoint. Maassen enumerates among the causes leading to
“teratologic” growth: High or low temperature, liquid or dry condition of the substrates, their
reaction, the stimulative effect of large amounts of carbohydrates, and of various salts. He
points out specifically that those of the pathogenic organisms are also to be found in the host,
sometimes just “when the bacterial development is at its height.”

While H. L. Russell (1892, p. 194) correctly stated that the big globular forms which he
found in old cultures of B.-granulosus could not, despite this fact, be termed involution forms
because they still were multiplying and forming spores, other authors do not hesitate to see
‘“‘mvolution”’ forms even in quite fresh cultures, in spite of the full vitality of these cells. Ra-
venel (1896, p. 30) registered “involution forms’’ with his B. cinctus in cultures 18-20 hours old.
Nothing is said about their appearance and behavior. A. Fischer (1897, p. 94) made the absurd
demand that morphological and cytological studies of the bacteria should never be made with
cultures older than 24 hours, because these ‘‘do not feel well’’ after this time:

Man sollte tiber einen Tag alte Kulturen niemals zu Studien tiber den Bau der Bakterien verwenden, da schon
nach dieser Zeit . . . die Bakterien anfangen, sich tibel zu befinden.

M. Jones (1908) found that ‘evidence of degeneration’? was “conspicuous” with the
organism studied after 2448 hours. But the author also reports that the colonies were only
1-1.5 mm. in diameter after 48 hours, and 5 mm. after 5 days. Frost and Penne (1910,
p- 26) state when discussing the involution forms:

Too rapid reproduction may cause aberrant shapes among some species.

Plotz, Olitzky, and Baehr (1915) say that in the cultures of the anaerobic bacillus, which they
isolated from typhus exanthematicus, “degeneration and involution forms appear early, so that
after repeated transplants the organism may assume a different appearance.” McFarland (1916,
p. 411) goes even so far to assume in regard to the enlarged and clubbed forms of the diphtheria
bacillus:

These probably represent an involution form of the organism, for (!) they are present in perfectly fresh cultures
. The involution of the diphtheria bacillus seems to occur in proportion to the rapidity of its growth.

Accordingly, he finds the “involution,” i. e. degenerate forms, prevalent on good substrates,
whereas bad substrates furnished him only “short spindle and lancet shapes.”’

That, in fact, branched and clubbed forms of the tubercle and diphtheria bacilli are not
involution forms has been pointed out in 1890 by £. Klein, who apparently. was the first to
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 25

emphasize the now well-confirmed fact that they are to be found in young cultures as well as
in the body. Similar statements have been made more recently in regard to B. tuberculosis
by Wolbach and Ernst (1903), in regard to B. malle: and B. diphtheriae by Hiss and Zinsser
(1914, p. 19), though in the latter case an accompanying picture (fig. 6) still bears the legend
‘degeneration forms”’ of B. diphtheriae.

Bongert (1901), Gotschlich (1903, p. 38), Kolle and Hetsch (1911, p. 24) accept branching
and club formation as a normal stage of growth, whereas Schmidt and Weis (1902, p. 82), A
Fischer (1903, p. 49), Abbott and Gildersleeve (1903) take the opposite standpoint. The last-
named authors do this especially because such forms were not seen regularly on ‘‘standard”’
media. Migula (1900, Vol. II, pp. 495, 499) thinks that a priori all branched forms of B. tuber-
culosis and of B. diphtheriae are to be considered ‘‘involution” forms and ‘‘degenerate”’ +).
Accordingly, he (Migula, 1904, p. 38) declares the swollen and branched forms of B. radicicola
to be ‘‘undoubtedly” involution forms. To Stefan (1906) it was even quite certain that they
are pathologic. Also EZ. F. Smith (Vol. II, 1911, p. 114) believes that they are degeneration
forms, because ‘‘such forms are either dead, or, if capable of further growth, return to the
original form when placed on suitable media.”’

More than 30 years ago, however, Prazmowski (1888 6, 1890) found in the course of his
thorough studies upon this organism that the so-called ‘‘bacteroids”’ appear when the develop-
ment is at its height, and that they reproduce small rods. The latter fact had first been
directly observed, 20 years earlier, by Woronin (1866), and it was afterwards confirmed by
Frank (1890), later by Hiltner and Stérmer (1903) and others. Siichting (1904) also noticed
that ‘‘bacteroids” are abundant in the cultures when these are fully developed, whereas in
cultures older than 8-14 days numerous small motile rods are present, originating from the
branched forms. My own investigations (1905 a), as well as those of Gage (1910) and Zipfel
(1911), confirmed and extended these findings. That according to Morck (1891), Harrison
and Barlow (1907) the branched forms are prominently located in the growing, active part
of the nodules, while in the youngest as well as in the oldest nodules only cocci and rods
are present, agrees very well with the results obtained in cultures kept under suitable con-
ditions. Miehe (1913), on the other hand, because he could not see further development of
branched forms under the microscope, does not hesitate to call them ‘‘cripples” and all posi-
tive results recorded in the literature ‘‘fiction.” But the normal rods of the organism used
in these experiments (B. foliicola) also refused to grow under Miehe’s microscope, and that
his statement concerning the positive findings secured by other investigators is entirely
unwarranted, goes without saying.

When Hauser (1885) first studied his pleomorphic Proteus, large globular, pear and club
shaped forms (3-74) were described by him as ‘‘involution forms.” But he pointed out himself
that these very motile forms were already present, and sometimes even prevailed in the
colonies when only a few days old, while they were replaced in old liquefied gelatine cultures
by the small rods and cocci, characteristic of this group. By direct reinoculation the ‘‘invo-
lution”’ forms could be stabilized, so that they reproduced themselves, e. g., in 30 successive
transfers, made during seven months.

The well-known ‘‘involution’’ forms of the plague bacillus, which are still virulent, appear
in artificial culture within a few days, while again in really old cultures the small ‘‘normal”
forms alone are present, as has been shown especially by Albrecht and Ghon (1900). In accord-
ance with this fact, the same forms have been found by Sata (1900), Dieudonné (1903), and
others in the body, showing full vitality and virulence. Gall-Valerio (1903) states explicitly
that the various forms of B. pestis (globules, clubs, threads, branched cells, etc.) are not
involution but evolution forms, because they appear first and the short rods later.

The meningococcus produces its ‘‘degenerate” giant forms in large numbers in only two
days old cultures (Kutscher 1907). Micrococcus melitensis grows, according to Eyre (1907), in
bacillary form only in old cultures or when the reaction of the substrate is not right, while

 

1In his own words dL. C., Dp. 499): “‘ Dass diese Verzweigungen ebenso wie bei zablreichen anderen Bakterien nichts weiter als Involutionsfor-
men sind, braucht wohl nicht erst besonders hervorgehoben zu werden.”
26 MEMOIRS NATIONAL, ACADEMY OF SCIENCES. [Vor XVI.

Saisawa (1911) found the rods in young, the cocci in old cultures. Durham (1898) has demon-
strated that merely a change.in the temperature suffices to turn the cocci into bacilli or vice
versa; both are equally vital. Jordan (1916, p. 416), accepting the rod form as the normal
type, ‘eaills the organism Bac. melitensis. Kruse and Pansini (1892) were not inclined to think
that the rod-like cells produced by pneumo- ard streptococci are involution forms, for they
were found to be the result of an especially luxuriant development of the cultures. The large
yeast-like cells of Streptococcus lanceolatus were found by Azelrad (1903) in young colonies on
the plate.

That the large swollen forms produced by acetic-acid bacteria are not properly termed
“involution” forms had been pointed out by E. Chr. Hansen as early as in 1879, and it was
recently confirmed in the course of detailed studies by Janke (1916). Bordoni-Uffreduzzi

_ (1888 6) said the same concerning the large globular, spindle-shaped and triangular forms occur-
ring in cultures of his Proteus hominis. He sees in them an expression of luxuriant growth.
Old and weakened cultures showed only the small ‘‘normal’’ rod-like cells. With the ozaena
bacillus—another member of the B. pneumoniae group—Simoni (1900) got analogous results.
And these were further confirmed by my own experiments with various nitrogen-fixing strains
of the B. pneumoniae group (1905 a), which shed some interesting light on the relations existing
between B. radicicola, radiobacter, and pneumoniae.

B. Chauvoei is another example of conspicuous pleomorphism. Its characteristic forms
occur everywhere, in the cultures as well as in the tissue. But they, too, have been declared
by Fraenkel and Pfeiffer (1895, legend to fig. 60, Pl. XXX). to be degenerate, crippled, involu-
tion forms, merely because they are entirely unlike the typical cells which ‘‘obey the rule”
of the monomorphistic dogma.

In their interesting studies upon the morphology ‘of the tetanus bacillus Kanthack and
Connell (1897). correctly emphasize that the branching, club formation and mycelial growth
shown by this organism can not be set aside as ‘‘involution forms’’, because they appear in the
cultures when 18-24 hours old, and they disappear with increasing age.

The multitude of monstrous forms found by B. Fischer (1894, p. 36) to be characteristic
for his Halibacterium polymorphum becomes visible in cultures not older than 2-3 days,
when grown on most suitable substrates.

Baumgarten (1890) called the attention of cholera investigators to the important réle
which the so-called involution and degeneration forms of V. cholerae may play in cultures, as
well as in the body. Hammer! (1906) and Popoff-Tscherkasky (1914) have furnished convincing
proof that in this case, also, those terms have been greatly misused. All these globular, trian-
gular, spindle-shaped and ramified forms were found to be very motile, abundant, and some-
times even exclusively present in young cultures; they could be propagated as such for weeks
and months, and produced again ‘‘ typical” commas and spirilla. Of special importance, how-
ever, is the fact demonstrated by Bittrolf (1912) that great variation may be shown in this respect
by different strains, especially in the animal test.

Severin (1897, p. 513) wrote in regard to the characteristic triangular branched cells of
his Vibrio denitrificans:

Es kann kaum bezweifelt werden, dass wir es hier nicht mit Involutionsformen zu tun haben, erstens finden sie
sich nicht in alten Kulturen, sondern im Gegenteil nur in frischen, und . . . derartige Formen farben sich sehr
gut und weit gleichmissiger als die Mebrzabl der sich nicht verzweigenden Formen und besonders der Spirillen.

Reichenbach (1901) took the same standpoint concerning Spirillum rubrum, and Bajardi
(1903) obtained branching so frequently in young cultures of Vibrio lingualis Weibel. that he
thought this organism should be transferred to the Actinomycetes as “ Streptothria lingualis.”

Spirochaeta pallida is, according to Sobernhevm (1907), equally inclined to exhibit “atypical”
and “involution” forms within the tissue. And the same holds true in regard to different
species which are pathogenic to plants, as has been recorded by Macchiatti (1898) and E. F.

 

1“ Sehr auffallend ist die ausgesprochene Neigung des Rauschbrandbacillus zur Bildung von Involutionsformen. ™ma ibm auch begegnen
mag,im Gewebssaft der Tiere, in Bouillon, Gelatine oder Agar, fast stets trifft man zablreiche, wunderlidh verzerrt Misswiichse, degenerierte
Zellen an, deren verkriippelte Gestalt kaum noch eine Spur von Ahnlichkeit mit den typischen, vorschriftsmissigen Gliedern besitzt.””
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 27

Smith (1911, Vol. II, pp. 73, 256, 288). The former author says concerning his Bacillus Bac-
carinii (the causative agent of the ‘‘mal nero” of grapevine):
Die Grésse wie auch die Gestalt dieses Bacillus variieren sehr . . hier haben wir also einen schénen Fall

von Polymorphismus; . . . die verschiedenen Gestalten folgen auf einander, wenn die Mikrobien tiber giinstige
Entwicklungsverhiltnisse verfiigen.

More details to be found on the following pages will amply support the conclusion, which
can be drawn from the cases discussed, viz., wherever the so-called involution forms have been
made the object of more discriminate studies, nearly invariably it has turned out that it was a
mistake to presume their being degenerate, merely on account of their “atypical” appearance.
Actually, in many cases they have shown themselves to be constantly occurring stages in the life
history of the bacteria. The“ typical”—not the so-called involution—forms have been often found
to prevail in old cultures, whereas the ‘ ‘atypical ” forms were predominant when the development
was atits height. Wasserzwg remarked in 1888 (0) that involution forms were sometimes more
normally visible than the so-called normal forms. And still earlier Finkler and Prior (1885)
emphasized correctly that it is usually by no means easy to decide definitely whether mor-
phological changes of the bacteria—the so-called involution forms—are to be considered
pathologic or normal occurrences within their life history.‘

The distinctly unscientific trend of opinion, which has been promulgated in this respect
by some of R. Koch’s overenthusiastic pupils, is clearly substantiated by the following quota-
tion, taken from C. Fraenkel’s “Grundriss der Bakterienkunde” (1891, p. 10). The involution
forms, says he—
gehéren nicht in den Entwicklungsgang der betreffenden Bakterienart, sie sind kein notwendiges, unentbehrliches
Stiick des Weges, welches das einzelne Individuum von Anfang bis zu Ende durchmisst, und verlieren damit den

Anspruch auf Legitimitat, sind uns nichts weiter als ein Anzeichen dafiir, dass unter dem Einflusse ungiinstiger Ver-
haltnisse eine Degeneration des Bakterienprotoplasmas stattgefunden hat.

Yet this standpoint has been fully shared by authors like Macé (1897, p. 323). A. Fischer
(1903, p. 49), Maassen (1904), Giinther (1906, p. 22), and others.

The real value of Fraenkel’ s reasoning becomes quite evident when we consider the following
two points: In all cases, where purely vegetative propagation is successful, as with all lower,
as well as with many higher, plants, all reproductive organs, flowers and fruits, are entirely
‘unnecessary,’ can not be accepted as “legitimate” and must be considered as indicating
degeneration, because the vegetative growth comes to a standstill at this time. Further
Fraenkel admits that B. anthracis (ike many other bacilli) also produces coccoid forms. But

-he is absolutely certain that these do not form a part of the life cycle of the bacillus, “because”
they either do not grow at all, or if they do, they produce again the “regular, typical” rods, i. e.
they behave like spores.?) The endospores had been studied by Rk. Koch, while other reproduc-
tive organs remained unknown to him. Therefore, to Fraenkel and to other faithful adherents
the former are of greatest importance and the latter entirely ‘‘illegitimate.”

A comparison of the so-called involution forms with the endospores is of special interest.
Duclauz observed several decades ago (1883, p. 654) that his Tyrothria virgula and scaber pro-
duced characteristic swollen forms, resembling spermatozoids, but that only later, within these
swellings, the endospores appeared. Our Bac. malabarensis (Léhnis and Pillai, 1907) behaved
in exactly the same manner. Braem (1889) made some interesting experiments which clearly
show that inoculation into water is as useful to stimulate the appearance of the so-called invo-
lution forms as it is to induce spore formation. A. Meyer (1901 a) points out that also branch-
ing occurs at an earlier stage of growth than spore formation; and Heymann (1903) obtained

 

1“ Wir halten es fiir ausserordentlich schwierig, sichere Beweise dafiir liefern zu k6nnen, wo die Grenze liegt zwischen den mannigfachen Varia-
tionen der Form, welche als pathologische Zustinde der Vibrionen aufzufassen sind, einerseits, und andrerseits solchen Bildungen, welche als

physiologische Entwicklungszustiinde angesprochen werden miissen.”’
2In Frdnkel’s own words (1. c., p. 9): “Gehéren nun derartige ‘Kokken’ in den Entwicklungskreis der Milzbrandbacillen? Gans gewiss

nicht. Denn. . . dieselben sind entweder tiberhaupt nicht mehr fortpflanzungsfihig . . . oder aber sie lassen sogleich wieder die
. typische Wuchsform, das Langstébchen von Tegelmiissiger Gestalt aus sich hervorgehen. Es sind diese Gebilde eben nur der Ausdruck
tir eine stattgefundene Entartung der betreffenden Bakterien,es sind Degenerations- oder . . . Involutionsformen, Misswiichse, die fiir die

Beurteilung des normalen Wachstums gar nichtin Betracht kommen kénnen.”

\
28 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

analogous results concerning the “involution” forms of diphtheria bacilli and the (later)
appearance of the granules in them giving Neisser’s staining reaction.

Endospore formation and Neisser reaction have been always accepted by the followers of
BR. Koch as of very great systematic and diagnostic importance. Therefore, the adverse valua-
tion of the ‘‘involution” forms, as taught by them, is by no means in accordance with the facts.
It would be even more correct to call the endospores involution forms, because they are to’
be found in old cultures, when motility and vegetative propagation is at a standstill. It is
true, some of them, though not all, will produce new rods when transferred to a new substrate.
But many of the so-called involution forms act in the same manner, and just this was the reason
why several authors discarded them as being of no interest or importance.

Some authors, like Kruse (1896 a, pp. 61- 64), W. Winkler (1899), Gamaleia (1900), “Pubr-
mann (1906), Taddei (1909), Léhnis (1913, p. 24) and Ambroé (1915), have advocated a closer
study of and amore scientific attitude against the so-called involutionforms. That their. appear-
ance is dependent on external (chemical and physical), as well as on internal (biological) in-
fluences, is clearly indicated by the interesting results secured by Gamaleia (1900), Matzuschita
(1900), Rosenfeld (1901), Maassen (1904), Péju and Rajat (1906), Hata (1908), and Taddei
(1909), on substrates containing various salts, relatively large amounts of carbohydrates, or
possessing an unusual reaction. Kruse (1910, p. 8) points out that a special stimulation of
growth is frequently responsible for their appearance, whereas A. Meyer (1901 a) emphasizes
that undoubtedly “inner reasons” are the more important ones, because those forms often
develop more or less irrespective of substrate, temperature, etc. Accordingly, some species
are very much inclined to produce such special growth, while others under analogous conditions
show little or no alteration at all (Migula, 1897, Vol. I, p. 52). Fuhrmann (1908) found, in
accordance with Matzuschita (1900) that moderate salt concentration merely accelerates the
normal growth, while high concentrations, of course, act detrimentally or may eventually
cause the death of the cells.

If the vitality of so-called involution forms is not tested very carefully, it may easily happen
that incorrect results are recorded. The uniformity of the methods usually employed is liable to
cause misleading observations. The difficulties, which have often arisen. when new species
had to be isolated, should be kept in mind. That a marked ‘periodicity in the progress of cell
growth may often ‘become noticeable, when tested on the ordinary substrates, has been demon-
strated with pyogenic staphylococci and with dysentery bacilli by Reimer (1909) and with
diphtheria bacilli by Springer (1913). The initial increase only lasts a few days, then (appar-
ently) a reduction takes place, followed again by another maximum, and eventually by more of
such waves. The following examples from the last-named author’ 8 paper (in round figures
and abridged) may elucidate this point:

Germs per ce. Experiment No. 4. Experiment No. 5.
At the beginning. ..... 2.2.2.0 0.. 2000000. e eee cece 3, 800 / 4,176
SARE SY ac 59 cid on di ete pase acces ccemeyemeneGn Sa 33, 000, 000 14, 000, 000
After O:dayssccxe0 ces essen eased s sexeneaewerinmesematacey 1, 000, 000 4, 600, 000
Miter DP aaa pes (nd can Oo ataoeeaaenca tec cea tneneeate 23, 000, 000 46, 000, 000
After 42 days........ Yep ARORA TANI ARTEL tyre Season 19,000,000 366, 000, 000
After 62daya.o.: ccorcgents se eve ossiinsnsgianomonmseedcans 63, 000, 000 20, 000, 000

Tt would be well worth while to combine such experiments with a close study of the appear-
ance and disappearance of the ‘‘involution” forms. It can hardly be questioned, that such
investigations would greatly help to modify the prevalent theories concerning the meaning and
importance of the different types of bacterial growth.

Looking back upon all those conflicting and often very incongruous statements, which
have been made by the bacteriological writers during four decades, one might feel much inclined
to agree with Fokker, who said as early as 1888 (p. 125) that the term “involution form”’ is
only “a name without any significance.” It would be best to drop the term entirely. As
this, however, in all probability will not be done, it might be advocated to use the word at least
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOBNIS. 29

exclusively in its correct sense. Webster’s New International Dictionary (1910) contains the
following explanations:
Involution=retrograde development or retrograde evolution; degeneration.
Degeneration=a progressive deterioration, as a return to a simpler and less highly
organized condition in the evolution.

Accordingly, it would be quite correct to call involution forms the small ‘‘normal” forms,
which in really old cultures often replace the larger, more highly organized cell types, which have
so far been termed involution forms, evidently without sufficient reason. But such practice
would only lead to still more confusion. All that can be expected in the future seems to be a
somewhat more cautious use of this greatly overworked expression. Speciatized studies of
the subject would prove very helpful in this respect.

(4) VARIATIONS AND MUTATIONS.

The fourth explanation, which has to be taken under consideration in regard to apparent
or true pleomorphism among the bacteria, is the occurrence of temporary, as well as of more
or less permanent, changes in the type of growth, called adaptation, modification, variation and
mutation. Again much differences of opinion are to be found in the literature concerning the
correct use of these expressions. This point has been discussed in the publications of Pringsheim
(1910), ‘Nyberg (1912), Dobell (1913), Toenniessen (1913), Jollos (1914), Jordan (1915), and of
Prell (1917).

For our purposes it will suffice to use preferably the general term “variability,” as in
practically all cases observed so far, quite naturally cause as well as stability (or instability)
of the occurring changes have remained more or less in doubt. In some cases the influence of
external conditions and the instability of the change induced by them are evident, in other cases
the opposite holds true. How a change of substrate may influence the form of the bacteria
has been clearly demonstrated, e. g., by Wasserzug as early as 1888. By studies upon the
variability of the streptococci, Lemoine (1896) was led to the conclusion that not only the in-
fluence of the substrate had to be considered, but also the ‘‘ modifications subies au sein méme
de l’organisme.”’ Neuwmann’s (1897) investigations upon the variability of pigmentation of
the micrococci still more indicated a predominance of internal against external causes. Luck-
hardt (1901), Retiger and Sherrick (1911), Jordan (1915), and others have furnished more exam-
ples in this direction. That, on the other hand, external conditions occasionally may assume
very great importance has been exemplified, e. g., by Bordet and Sleeswyk (1910) concerning
the change of agglutinability in various species, by Geisse (1914) in regard to the pathogenicity
of saprophytic staphylococci, by Heinemann (1915) in regard to pathogenicity and morphology
of Streptococcus lactis, and by Bernhardt (1915) with B. typhi, dipthheriae, and others.

With some of the variants obtained in these experiments considerable stability has been
recorded. Hill (1899) isolated a strain of B. diphtheriae which showed branching very persist-
ently. Lepeschkin (1904) also succeded in splitting off an equally inclined strain of his B.
Berestnewi, making use of single-cell cultures. Barber (1907), too, obtained from single cells
strains possessing morphological peculiarities which were (apparently) constant. Whether the
constancy of a new character isin fact real or apparent must necessarily remain in doubt in many
cases. For instance, a white Micrococcus strain isolated by Neumann grew white for six years
and returned then to the original yellow type. One of Luckhardt’s white Prodigiosus cultures
became suddenly red after several months’ experimenting, only to return to white afterwards.
One of my own stock cultures, a brown variety of B. fluorescens, became entirely white after it
had been, kept for about two years in the collection, always being transferred in approximately
three months intervals to beef agar made up in the same manner. The next inoculation pro-
duced the brilliant green of a typical Fluorescens, but the next one was brown again, and so it
remained through several years. As most observations naturally must be. of much shorter
duration, the question concerning the stability of a new form will always remain more
or less in doubt. It is well known that frequent transfers to the same substrate repeated after
30 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI,

a few days tend to keep a strain constant, whereas older cultures are much more liable to
furnish, either new variants or to exhibit atavistic tendencies. First (1914), e. g., who studied
anew and in single-cell cultures the remarkable variability of the Finkler-Prior vibrio, first
investigated in mass cultures by Firtsch (1888), obtained constant variants when these were
transplanted after 5-6 days, whereas by 24-4 months’ intervals the number of atavistic changes
was increased, and 4-months-old cultures produced again a growth exhibiting the normal orig-
inal characters. Burckhardt (1914) was able to restore motility in some immotile stock cultures
of B. colt not by frequent transfers, but by starting from very old material. Immotile strains,
on the other hand, could be easily obtained by keeping them under less suitable conditions.
These facts are in good agreement with some observations mentioned above concerning the so-
called involution forms; we will have to refer to them when discussing the various changes
occurring in the different phases of the life cycles of the bacteria.

While the doctrine of constancy, as promoted by R. Koch and his pupils, was rather short-
lived as far as the function of the bacteria comes into view, it has found, on the other hand, up
to the present time considerable support in regard to the form exhibited by this special group of
organisms. As Life, however, is characterized, according to Spencer, by the ‘‘continuous adjust-
ment of internal relations to external relations,” it is, indeed, self-evident that variations must
occur under changed conditions (Spencer, 1898, p. 334). That all, morphological as well
as physiological, characters of the bacteria may vary has been explicitly stated, probably for the
first time, by Négeli (1882, p. 135). The following of his sentences (p. 138) is remarkably
accurate when we consider that it was written more than 35 years ago:

Die Species wird nicht durch absolute Merkmale kenntlich sein, sondern dadurch, dass sie unter bestimmten dus-

seren Umstinden bestimmte Modificationen des morphologischen und physiologischen Verhaltens, unter anderen
Umstinden andere Modificationen zeigt.

Ziegler (1885, p. 203) also admitted that variability occurs ‘‘within definite limits,” though
it had been more correct to say ‘‘indefinite”’ limits, as he expresses himself as follows:

As to the extent of the cycle of variations, through which any one of the known bacteria may pass, we know
indeed but little.

Among the German and Austrian bacteriologists it was especially Escherich (1894), Hueppe
(1896), Kruse (1896, 1910), and Lehmann and Neumann (1912), who have insisted upon an
adequate study of these problems and have done much in reducing the chaos in systematic bac-
teriology caused by the widespread habit to incorporate into the literature many badly studied
instable varieties as constant species. Even Migula, who at first was one of the most ardent
supporters of the latter practice, had to admit later (1904, p. 35) that several organisms, e. g., the
cholera vibrio, are characterized by a very great variability:

Among the French authors Macé (1897) takes a distinctly contradictory standpoint, similar
to his position in regard to the pleomorphic character of the bacteria. On the one side he says
(p. 323):

On peut affirmer 4 l’heure présente, avec la probabilité la plus grande, qu’il existe chez les bactéries des espéces
vraies, 4 charactéres fixes, se produisant et se perpétuant sans varier.

“But only a few pages further on we are informed (p. 329):

D’uneautre cété, il n’est guére possible d’admettre une fixité absolue des charactéres que l’on considére comme
specifiques. De nombreux exemples preuvent, au contraire, qu’a cété des charactéres physiologiques que nous savons
variables, les caractéres morphologiques eux-mémes ne nous montrent pas une fixité absolue.

Otherwise, all French bacteriologists are accepting unanimously the variability of the bac-
teria as a necessary correlative to their pleomorphism, as has already been shown by some of
the quotations given above. The following paragraph taken from Duclaua’s ‘‘Traité de Mioro-
biologie” (1898, p. 605) may serve as another characteristic example:

A ces variations de fonctions viennent se supposer des variations de formes au sujet desquelles les savants avaient
fort discuté, les uns soutenant qu’elles étaient assez constantes pour entrer dans la diagnose de l’espéces, les autres

 

”

1“ o zeigt besonders der Organismus der asiatischen Cholera eine ausserordentlich grosse Zahl von Varietiiten, dle von stark gekriimmten bis
fast geraden, von sebr kurzen bis sehr lang gestreckten Zellen fast alle Zwischenformen zeigen und sich dabeiin Kulturen auffallend konstant ge-
zeigt haben . . . Diese Varietiiten behalten auch auf verschiedenen Nihrbédenihre Formen bei,sind also keine ‘Ernahrungsmodifikationen,”
und sie wiirden, bei unserer gegenwiirtig sehr unsicheren Kenntnis von der Umgrenzung naturhistorischer Arten bei den Bakterien, sicher als
verschiedene Arten betrachtet werden, wenn sie nicht eben als Erreger derselben Krankheit gefunden wiiren.”
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 31

qu’elles étaient indéfinies. On voyait bien maintenant que cette question n’en était pas une, ou plutét qu’elle
disparaissait derniére la question plus grosse et plus profonde des variations de la fonction protoplasmique . . . Bref,
a Vidéo de spécificité, encore dominante ily a dix ans . . . s’en est substituée une autre, . . . celle dela
plasticité de la fonction protoplasmique.

More recently, LeBlaye and Guggenhewm (1914, p. 12) have strongly emphasized the very
great variability exhibited by most of the species, which are forming groups of more or less
atypical ‘‘ races,’

Among the British authors, it was especially Adami who has repeatedly (1892-1916)
called the attention of his fellow-bacteriologists to this important problem. In his second
paper on this subject (1894) he pointed out that experimentally changes can be produced
which would be considered a priori as characteristics of a totally different species, and that
on account of the wide variability to be observed under different conditions “the first step
must be in the direction of the establishment of a system of broad groups.”

In the American literature some discussions upon the variability of the form of the bac-
teria are to be found in the works published by Chester (1901), Park and - ‘Williams (1914), and
others. However, as has been illustrated by the quotations given above, the doctrine of sim-
plicity and constancy of the form has been supported by many of the American bacteriologists
until quite recently. Moreover, there is even a tendency now ‘to establish not only species,
but genera on some biological characters, though the still greater variability of these marks,
clearly indicated in the early publications of the French authors (Guignard and Charvin,
Wasserzug and others), has been so often confirmed since then that it is now well beyond ques-
tion. C.-E. Winslow and A. Rogers-,Winslow (1908), e. g., have advocated dividing the micro-
cocci into different genera mainly on the basis of their pathogenicity and pigment-production.
The systematic relationship of the streptococci, too, should be established, according to the
first-named author ( Winslow, 1912), on the outcome ‘of biometric studies of their fermentative
powers, a method which had been advocated, though only for practical purposes, by Andrewes
(1906) and other British authors. That these very variable properties can not be used
for classification has been demonstrated anew by Broadhurst (1915) in regard to the strep-
tococci, and by Burton and Rettger (1917) for the Colon-Aerogenes group. Kligler (1913), on
the other hand, fully confirms the correctness of Winslow’s standpoint and claims that there is
“a remarkable general correlation between pigment production and other properties.” How-
ever, white cocci sent to the collection used for these studies were found by Kligler to be orange.
An organism, originally exhibiting all characters of Micr. agilis, was declared by him to be
either UW. citreus or roseus, indicating beyond doubt that also in this case the generally observed
instability of the physiological properties has exerted its influence.

Numerous papers have been published during the last 10 years, mostly in Europe, wherein
the morphological as well as the physiological variability of some groups of bacteria have been
more or less thoroughly discussed, usually under the aspect of mutations. This point was first
emphasized in the publications of M. Neisser (1906), Massini (1907), Penfold (1911-1912),
and R. Miiller (1912) concerning lactose fermenting “mutants” of B. coli. The preliminary
communication made by the first-named author was immediately followed by still more inter-
esting remarks by Schottelius (1906) and Gotschlich (1906) concerning B. pestis, and by Gruber
(1906) in regard to Finkler-Prior’s vibrio. The so-called mutation of B. pestis has been further
studied by Markl (1914). Toenniessen (1913- 1914) added important data upon the variability of
B. pneumoniae. Nyberg (1912) succeeded in producing artificially with all of 140 strains of
different lophotbrichic bacteria (Fluorescens, Cyanogenes, etc. ) those different types of colonies
which have been observed i in many other Senesera on “mutations.”  Baerthlein ae

river §

considerable changes of the cell form. Kanlisch (1916-1918) and Gildemeister (1917) confirmed
and extended Baerthlein’s findings with regard to tne Coli- -Typhosus group. The great varia-
bility occurring in the diphtheria group has been elucidated to some extent by the investiga-
tions of Ohlmacher (1902); Goodman (1908), Bernhardt and Paneth (1913), and others. Stamm

160266°—21——3
32 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

(1914), Olsson (1915), and Feldmann (1917) made the cholera vibrio the object of their studies.
P. Eisenberg (1914-1918) worked with representatives of various groups of bacteria (Bact. pro-
digiosum, violaceum, fluorescens, pneumoniae, typhi, coli, dysenteriae, and Sareina tetragena).
However, it would be entirely without foundation to assert that the two foremost marks of
true mutation, viz., the sudden appearance and the constancy of the alteration observed, had
been sufficiently demonstrated i in any or all of these cases. That there may have been some
true mutation is not to be denied. But the examples given above indicate clearly, how
difficult, nay even impossible, it will be to find out in the ordinary course of examination
which only lasts some weeks or months, whether the change in fact is constant or not. And
whether the new form or function has appeared gradually or indeed suddenly is a problem
still’ more difficult to solve. Accordingly, several authors, like J. Klein (1912), Bernhardt
(1912), Berry and Banzhaf (1912), Saisawa (1913), Feldmann (1917), and Baerthlein (1918), have
emphasized that the alterations’ actually did not represent mutations, but only temporary
adaptations, fluctuating variations or atavistic reactions. As older cultures generally are more
inclined to produce aberrant strains, it has been repeatedly assumed that accumulated meta-
bolic products are the foremost cause of the effect. To some extent this is undoubtedly true.
But, as with the involution forms, it must be also kept i in mind with regard 1 to morphological
and physiological variations shown by cultures of various age, first, that only some of them
show variability and these again ‘to.a different degree, while others do not, and i in the second
place, that with i increasing age of the culture there may be a complete return to the original,
so-called typical, character.

, Bernhardt (1915), Olsson (1915), and others have quite correctly spoken of the ‘ ‘eyele of
variation” characteristic for every species or group of bacteria. This cyclic alteration of form.
and function i is, indeed, an important attribute of the normal life cycles of the bacteria. And.
this fact easily explains, why in all cases more or less variability | can be observed ‘from the
start, when a sufficiently large number of strains, belonging to one species, is isolated at the
same time. - Accumulation experiments, as recommended by Beijerinck (1898, 1901 a), are of
very great importance in this respect. Kruse (1910, Pp. 1153) | says: .

Es ist von vornherein zu erwarten, dass die ‘natitrliche Ziichtung i in ahnlicher Weise Aber erzeugen wind: wie
die kinstliche. Die Erfahrung hat das auch immer wieder bestitigt.

_ Grassberger (1905) was equally right when, he wrote:

Dartiber kann kein Zweifel bestehen, die Natur versteht das Variieren noch viel besser als wir, und auch das
Misshandeln, speciell dort, wo es sich um die Wechselwirkung zwischen verschiedenen Organismen dreht.

Dye 3 ° BES

Some authors have, therefore, strongly advocated to study freshly isolated strains, so as
to avoid those “artificial” alterations caused by long continued growth on the substrates used
for laboratory work. For instance, LeBlaye and Guggenheim (1914, p. 10) write:

Un diagnostic méthodique n ost possible que si l’on opére’ avec ‘des cultures fratchement retirées de habitat
naturel. Ainsi- se trouve eliminées les erreurs innombrables qui résulteraient des modifications que subissent les
caractéres morphologiques et biologiques des ‘espéces sous Vinfluence du séjour dans les milieux artificiels.

On the other hand, the “standardization” not only of the methods to be applied, but of
the bacteria themselves to the extent of eliminating all natural varieties, has been ‘also recom-
mended. &. Koch's pupils, the representatives’ of the extreme monomorphism, laa took
the leadership in this respect. Fraenkel (1891, p. 10) said:

Wenn man dafir sorgt, dass ein solches (i. e. apparently pleomorphic) Bakterium dauernd unter gleichmissig
ginstigen, unter den besten’ Erndhrungs- und Wachstumsbedingungen steht, so bleiben die abnormen Formen aus,
und wir bekommen einen einheitlichen, fest umschriebenen Eindruck von seinem dusseren Verhalten.

However, it is by no means certain, whether the ordinary standardized laboratory methods
will afford the most suitable conditions of growth. In fact, another statement in Fraenkel’s
book (p. 132) explicitly admits that undoubtedly many microdrganisms do not grow’ at all on
the media commonly used, and—

Die Binténigkeit des. ‘Vertahrens ist gewiss haufig die Ureche fir das Fehlechlagen der Galturvemuchs,
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 33

Kruse (1896, p. 476; 1910, p. 1125) says:

Die Milzbrandbacillen von Maus auf Maus tibertragen, das Essigbakterium immer frisch mit alkoholischer Nahrung
gespeist, der Prodigiosus von Kartoffel auf Kartoffel tibertragen, verindern sich nicht, so lange man auch das Experi-
ment fortsetzt.

But as early as 1879 E. Chr. Hansen showed that the acetic acid bacteria in alcoholic
fluids, where they grow best, are very much inclined to undergo conspicuous changes in their
morphology, while they are ‘‘typical” small rods in the weakly growing layers on artificial
substrates. The anthrax bacillus, on the other hand, will never show its very characteristic
spore formation, when it is transferred in quick succession from mouse to mouse. But if we
would follow the advice given by A. Fischer (1897) to study only 1-day-old cultures, then it
would be indeed possible to get many uniform, though, of course, very incomplete, results.

McFarland’s Textbook on Pathogenic Bacteria (1916) contains another statement of
interest in this respect (p. 214):

While the appearances of the freshly isolated organism should be carefully noted, too much stress should not be laid
upon them, and before beginning the systematic study of any new organism it should be made to grow for several suc-
cessive generations upon two or three of the most important culture media. Its saprophitic existence being thus estab-
lished, the characteristics manifested become the permanent peculiarities of the species.

Some doubt, however, will be justified whether it is, indeed, a correct scientific practice
first to change artificially a parasitic organism, which naturally may be very little inclined to
grow on our laboratory substrates, no matter how much ‘‘importance’”’ we may attribute to
them, so that it will assume a different, distinctly saprophitic character, eventually lose its
pathogenicity more or less, and to conclude then that now ‘‘the characteristic peculiarities of
the species’”’ have been demonstrated. It seems, indeed, quite probable that the ‘‘domestica-
tion” of:the bacteria will often alter their morphology, as well as their physiology, relatively
to no less, degree than it has been the case with the domesticated higher plants and animals.

. How deeply the changes caused by continuous artificial cultivation may affect the general
appearance of a strain is e. g. clearly reflected in the statement, made by Migula (1900) in the
foreword to the second volume of his ‘‘System der Bakterien,” saying that many of the cultures
tested by him differed so widely from the original description that they were hardly recognizable.
If the advice had. been followed which was given by Adami (1894), viz., that all systematic
studies should be continued not, less than a year, and a second description of a so-called new
species should be published 12 months after the first one in the same journal, the situation would
‘be much better to-day in many directions. It needs hardly to be emphasized that such work as
was done, for instance, by Bredemann (1909) with the Amylobacter group, where by persistent
“‘training”’ of a large number of varieties a general species character is clearly brought out and
made more or less permanent, will always undoubtedly have its great value. The peculiarities
of aberrant strains, however, should also be carefully studied and recorded. This holds true
especially as far as alterations of form and of function occur as the result of natural variability

(6) MORPHOLOGICAL CHANGES DURING THE LIFE CYCLE.

The fifth and last explanation, which eventually may help to solve some puzzling problem
concerning bacterial pleomorphism, is the normal development of different forms in the various
stages of the complete life cycle of a species.

So far, this point has met with but little consideration, though it will undoubtedly win in
importance, as investigations go on. Some valuable material, however, is already available.

As mentioned before (pp. 30, 32), studies upon the variability of the bacteria have shown
that, at least in some cases, a distinctly cyclic appearance and disappearance of the ‘‘variants”’
was quite evident. And as far as the so-called involution forms have been considered to be
worth studying, the same fact has been noticed, viz. the original form first produces various
“involution”? forms, which later reproduce the ‘‘typical’’ form, so that eventually a several
months’ old culture again contains only ‘‘normal”’ cells. Numerous references have been given
34 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

on pages 24-28. At the latter place the investigations of Riemer (1909) and of Springer (1913)
have been recorded, which clearly indicate that a distinct periodicity of bacterial growth is
noticeable under suitable conditions. It may be added here that the result of our own studies
upon the life cycle of the Azotobacter group (Léhnis and Smith, 1916 a and 6) are in complete
agreement with these findings. In soil as well as in nutrient solution a precise succession of
the different characteristic stages in the development could be followed not only once, but
repeatedly, without making any transfers or any other alteration than adding water aseptically,
from time to time, when this became necessary.

The rather unnatural ‘‘overstuffed” substrates which are used most frequently in the
laboratories will often interfere with such observations. Large amounts of metabolic products
may kill the bacteria at an early date, and their presence has also often led to the assumption that
they are the cause of the appearance of ‘‘variants’’ and “involution” forms. It is, of course,
readily to be admitted that they, indeed, may have exerted some influence, and will often be able to
do so.. However, as the same or similar alterations are obtainable under quite different condi-
tions, and asuccessive alternation of ‘‘degeneration” and ‘‘regeneration”’ can be easily established
on suitable substrates, the natural tendency of the living cells themselves has also to be kept in
mind. Moreover, it should not be overlooked that such changed forms, produced apparently
under the influence of the metabolic products accumulated in the old substrate, not infrequently
retain their peculiar shape very persistently through many transfers on fresh substrates. | ~

In this respect some observations made by Stamm (1914) upon the variability of Vibrio
cholerae are of great interest: 13 typical cultures were kept in water; 6 remained constant, 7
showed profound alterations, and these new forms remained again constant in more than 540
transfers made during 30 months, provided that the transfers were made frequently. Material
taken from older cultures, however, either went back to the typical form, or showed further
alteration. Bernhardt (1915) recorded analogous results with meningococci, typhoid, para-
typhoid, septicaemia, and diphtheria bacilli. Daily transfers kept the ‘‘variants” obtained
from old cultures constant for several generations. As has been demonstrated by Preisz (1904)
and others, the so-called secondary colonies, frequent especially in old growth of spore-forming
bacilli on agar, are also much inclined to produce rather constant strains of different morpholog-
ical and physiological character. That giant growth and increased resistance of the cells are
typical for this new development, is obviously not in accordance with the theory of the dominat-
ing influence of the accumulated metabolic products. The ‘‘tertiary” colonies, which some-
times develop after another interval, may be accepted as additional indication of the periodicity
of growth occurring under suitable conditions.

That the pleomorphism of the bacteria is closely connected with their life cycle was already
the opinion of some early French writers, like Artigalas (1885), Cornil and Babes (1890), and
Macé (1897), who have been quoted on page 8. Two more references may be added here.
Billet (1890, p. 208) stated that every organism studied by him exhibited ‘‘un cycle évolutif
bien defini,” and Duclaux (1898, p. 607) said:

Tl faut prendre un microbe comme un 6tre 4 générations alternantes multiples et variées, se succédant, non suivant
une formule reguliére, mais suivant les conditions de l’ensemencement. Le lien de l’espéce, c’est la loi qui préside
& chacun de ces changements, et la variété des formes et des fonctions n’est pas du tout en contradiction'avec unite de
l’espéce.

In England Lankester wrote as early as 1873:

The existence of true species of bacteria‘ must be characterized . . . by the ensemble of their morphological
and physiological properties as exhibited in their complete life histories. ;

The same standpoint has been taken later by several British authors like H. M. Ward
(1892, 1910), Crookshank (1896), Walker and Murray (1904), Dobell (1911), Young (1914);
and Hort (1915-1917). The changes observed by Walker and Murray in cultures of B. typhi,
coli and V. cholerae, were accepted by them as signs of ‘‘an unexpected complexity in the life
history of these microérganisms.”’ Dobell wrote (1911, p. 484):

It appears to me probable that . . . the majority of bacteria may possess a wide range of variation in their
outward form at different stages in their life histories.
No. 2.] ; LIFE CYCLES OF THE BACTERIA—LOHNIS. 35

Young emphasized that it is necessary ‘‘to try to obtain as many stages of each life his-
tory as possible.’ Of special importance, however, is the following statement made by Hort
(1917a), because it is based on a broad experimental basis:

Evidence of complex bacterial life cycles is constantly before us, even in ordinary standardized laboratory media,
though it may, and often does, require persistent looking for. The truth, in fact, appears to be that we have gone
astray in this matter, because we have in the past invoked too easily the theories of contamination, of involution forms,
nd of mutation, and have forgotten that the natural environment of bacteria, whether as saprophytes or as agents of
disease, isin a perpetual state of flux.

In the older German literature the life cycles of the bacteria have been discussed, of course,
more or less hypothetically, by those authors who like Ndgeli, Neelsen, Zopf, and Haberkorn, tried
to defend the polymorphistic standpoint against the strict monomorphism of Cohn and Koch.
Some quotations were given on pages 11 and 13. Négeli (1882) was undoubtedly right when he
wrote:

Auf das leichte und subjectiv willkiirliche Geschaft der beschreibenden Unterscheidung muss nun erst die
eigentliche wissenschaftlich objective Arbeit der exacten experimentellen Untersuchung folgen (p. 139). Durch

Ziichtungen, die hinreichend lange unter den verschiedensten dusseren Umstinden fortgesetzt werden, ist zu bestim-
men, welche Formen sich in einander iiberfiihren lassen und welche nicht (p. 138).

The same holds true concerning the following statement made by Zopf (1883, p. 47):

Eine Systematik im Sinne der anderen Pflanzengruppen ist fiir die Spaltpilze zur Zeit insofern nicht méglich,
als es an einer entwicklungsgeschichtlichen Durcharbeitung des Gebietes noch ginzlich fehlt.

The curious change in Brefeld’s view, which led him (1908) to the assumption that it is
a priori erroneous to undertake any investigations upon the life history of the bacteria, though
he, too, had advocated such studies at an earlier date (1881), is a good example of the blight-
ing effect exerted by the rigid monomorphistic dogmatism propagated by the orthodox fol-
lowers of R. Koch.

Even F’. Cohn has once (18726, p. 180) emphasized the necessity of thorough investigations
upon the life cycles of the bacteria (see quotation on p. 11), though he abandoned later this
standpoint.more and more. But it is still more surprising that an author like Fraenkel (1891),
who did not hesitate to declare any type of growth or any reproductive organ which was not
approved by #. Koch as an entirely superfluous, ‘‘illegitimate,” degenerate by-product (see
quotation and critique on p. 27), nevertheless admitted that no definite arrangement of the
system of the bacteria could be made, because:

Der nétigen entwicklungegeschichtlichen Durcharbeitung entbehrt das ganze Gebiet noch zu sehr, als dass ein
solches Vorgehen schon am Platze wire (p. 13).

The more correct opinion, which during the last two decades has gained some ground
among German and Austrian bacteriologists, has naturally evoked new interest in such studies.
It was especially the Austrian authors W. Winkler (1899), Mencl (1905), and Fuhrmann (1906-
1913) who recommended a renewed close study of the complete life cycles of the bacteria and
themselves contributed some important findings from this field.

Although the full discussion of the data secured by such investigations upon the life history
of certain species will better find its place on the following pages, a short compilation of the
names of the authors and of the organisms, studied by them, may be inserted here. It is of
great interest for three reasons: First, practically all these studies have been made quite inde-
pendently and the good agreement of the results obtained is, therefore, of increased importance.
In the second place, it is of great value that representatives of all groups of bacteria have been
tested and analogous results have been recorded with all of them. In the third place, it
becomes evident that the necessity of such studies has been felt quite generally; among the
authors who have done such work practically all nationalities are represented.

The careful studies of Thtercelin (1899-1903) on the various developmental stages of his
Enterococcus will prove helpful for further investigations upon the streptococci. W. Winkler
(1899) obtained very important data with B. coli, fluorescens, aquatilis, and with a yellow,
nonspore-forming rod (probably Bact. herbicola), which he rather incorrectly named B. mesen-
36 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

tericus aureus. Hauser’s early work upon his Proteus (1885) furnishes a practically complete
picture of the life history of this group of organisms. Fuhrmann (1906-1908) studied the
life cycles of his Pseudomonas cerevisiae, several Fluorescentes, and of some other rods. Alm-
quists investigations (1893-1917) were especially concentrated on the typhoid bacillus and
the cholera vibrio. The latter had also been the object of the little known work of Ferran
(1885) and of Dowdeswell (1889-1890). Finkler and Prior (1885) gave a fairly complete report
upon the life cycle of their vibrio, isolated from Cholera nostras. Albrecht and Ghon (1900),
N. K. Schultz (1901), as well as Rowland (1914), have furnished valuable contributions to the
complete knowledge of the life history of B. pestis. Rosenbach (1909) did the same concerning
the bacteria causing erysipeloid, rouget and septicaemia of mice. The life history of some
organisms closely related to Friedlander’s Pneumonie bacillus was described by Bordoni- Uffre-
duzzi (1888) and by Jehle (1902). Johan Olsen (1897) investigated the life cycle of B. mycoides
and of other spore-forming bacilli. The glanders organism has been studied along similar
lines by Carpano (1913), the leprosy organism by Lutz (1886), Barranikow (1900), and by
Kedrowski (1901), the tubercle bacillus by Schroen (1886-1904) and by Maher (1910-1913)
and the Corynebacteria by E. de Negri (1916), and by Bergstrand (1918). Krzysztalowicz and
Siedlecki (1908), as well as McDonagh.(1912-1913) made the life history of Spirochaeta pallida
the object of their investigations. In Meirowsky’s “Studien iiber die Fortpflanzung von Bak-
terien, Spirillen und Spirochaeten”’ (1914 6) for the first time representatives of very different
groups (B. tuberculosis, leprae, paratyphi, enteritidis, various spirilla, and spirochaets) have
been made the object of a thorough comparative research. Of equal or even greater impor-
tance are the investigations of Hort and his collaborators (1915-1917), who worked with
meningococci, typhoid, paratyphoid, dysentery, and colon bacilli. And in our own experi-
ments upon the life cycles of the bacteria (1916) representatives of practically all groups of
bacteria, except mycobacteria and spirochaets, have been included.

Fig. 1 on Plate A taken from our first paper (Léhnis and Smith, 1916 a) shows in a sche-
matic arrangement the rather complicated life cycle of B. azotobacter, and a glance at the two
Tables II A and II B, taken from the same paper (pp. 683 and 689) and revised with regard
to some additional findings mentioned in our second preliminary communication (1916 5),
reveals at once the important fact that essentially the same types of growth occur in the life
cycles of all different kinds of bacteria.

Taste II A.—Types of growth observed with 24 ‘cultures of Azotobacter.

[The laboratory numbers of the cultures are given at the head of the columns.]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

28g ° ® |s ae 3 |s |s jo. eB y ga
ey +. wl ~ > Pr eo » a
‘/S8e| sh feelgel S82 [Salcececs| 283 a8
asa ag ag |o 233 agles|og\og 208 ao8
of growth. oo of of of 208 o£lo8 oflos og 00%
cer SE i 223 ee jeglo” SEE ez/oelos |e 383 soe
5 nQ neis nN aAQleoiNnAla n <q
a 4 led let | let fe let le <0 ao"
1/2/3]4]5 {6 ) 7] 9 | 10) 21] 12] 13] 14] 15 | 16} 17] 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25
tlel+]theft]e fete) etl eye] elaeleleleleleieielaele
t]t}tit]+y4itl]t]t]4]4+]4)4 4 [4 ]4]4]4]4]4)4)4 14] 4
tye {t+ 4fe--+ lade] a [ese] tf pony a he yaya yas
+]t]+]e)e/e (+ } 4] 4] +t +] ee Pee e yy] 4 Sys a el elas
PAST SIS (SITS S Se SSIS Seti ei sie
cal hee ae ae ach bed ae tel abba Earatl de [die Mane cd ee +
chetere reread eelare Eerste 2 epee el altit
EVE) + FVETE VETS] ELE PETE EVE TS 4 VEPEVS] $4 LEYS
FP |e ele pe lela sie lela ele pale |e ee be ela ey +

 

 

 

 

 

 

 

 
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 37

Tasie II B.—Types of growth observed with 18 representative bacteria.

[The laboratory numbers of the cultures are given at the head of the columns.}

 

 

 

 

    
 
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

é Siele lig 3 ale le |g ;
pa i” » a A eed ;
ale] TE ae lalele EB 2lalé

lP/ele/ils Belg 12/9 /8| 8 esesled 2/4 | a

a) eee lea legal el2]2|¢ escsee */a/é

Types of growth. 2/32 q a aia ata le | # 3 203 A 8/5
BlSle/Sleleig jglglelgieg & & /elel¢

sialelaiajam |alals>l|si/€S8 sa lolala

31 | 32 | 33 | 34 }.35 | 36 | 88 | 89 | 40 | 41 | 42 | 43 | 44 | 45 1,46 | 47 | 48 | 49

A Jarge globules and ovals) Sobeaaerineiese Sedbeewetie cen HH fees efeeee] hf Hb [ene nteeeeleeeel FL El] +] | + focee}eeee| Ff...
B (thick walled forms of type 4. te a ie a ares sseteuswetades +l] +t] tpt f---}eeeelee ef BP] ee] beeen [eee epee ee] HE
CC granular decomposition of A, B, L, M).. eeeft le] te] ef fee eee) EP PL ET Pb fone e [ee fee efee ee]
D feymaplesm sinbie Gabaie waite Gace fr pel tp tpt Peet Tt] epee ep ep eee a aes
ere globules and ovals).- PEPPY RP RP EP HEP EP te PePty ate ey] ty t yeti
F (small rods and threads). . APP HRLEP RPE RPE] EP EP ep tl ep ey apt ey sis
7 iélime thresds with cocd)). salsrcisia waseisisioaste a allewate eR Rae] Be ee te bed <Babok [ee fh [ecce|esesfeese| Ee fecech
secular Sopmpes Han ob e enact spores At hbel ble Pe Pe pay ey ey 4] 4 eee) 4 feed fee deep

T gegenerative bod O8)s scensmetnvenewenecansaic see e-fHPT LEE P te P ele] ep ae Pe tp ea eat
qipeeraa! cella developing inside) PERRY She ce locncnvamapesenrad AB feces AR | A? (vece [seca lemes|eecction el a Petwelencele se t|oesclenwe]enee|soncdscee
pene) efwthiaicied sisi tlt yt tpt pep eee Pe Pe ep ap a ap yp
eee end tesa, tt Pt] a feeeef eb PE] TH ]eeee] | cB fee e eee ee] +
ie Is with pointed en s) ease] AP faces] AP fanaa] “PP SR | AR De pecselese et Eo] | Ae fecweleeeef Eft
N (starlike growth)............2-. <Griciens ole Rha mammicas Sis amsaeitiu SEs +) 4+ ankol SEP ae Sep dep ge Fate gfawend 4p ll aed de bla oe fe

 

The detailed discussion reserved for the following pages will bring additional proof
that a fairly correct insight into the little-known life history of the bacteria is obtainable
at the present time, provided that those observations are carefully considered which practi-
cally always have been laid aside because they were incompatible with the prevalent doctrine,
and in their isolation they had not sufficient weight to command attention and to instigate
renewed experimental investigations.

It is beyond doubt that such investigations will lead to:an incomparably better knowledge
of the real morphology and physiology of the bacteria. The present situation in bacteriology
is very much like that confronted by Tulasne (1851) in mycology, in regard to whose work
De Bary (1884, p. 129) said:

Ankniipfend, an wenige und immer wieder zuriickgedringte Beobachtungen unternahm er zu zeigen, dass die
Formspecies der bisherigen Mycologie in vielen Fallen nicht fiir sich allein Speciesreprasentanten sind, sondern dass
eine solche Formspecies mit anderen dem Entwicklungskreise einer wirklichen Species angehéren kann.

‘What has been written by Sachs nee p. 214) about the algae holds equally true to-day
concerning the bacteria:

The discovery of alternation of generations and polymorphism in some sections justifies the supposition that
certain forms not as yet accurately known may be merely conditions of development of unknown cycles of forms,
although hitherto considered distinct species and genera.

That a complete knowledge of bacterial ontogeny is absolutely necessary for securing a
well-founded insight, into the phylogeny of these organisms is beyond question, since Haeckel
(1876) has shown how the development of the tribe is reproduced in the life history of its mem-
bers. Hueppe (1886, p. 41, 149), Kruse (1896c, p. 492), and Fuhrmann (1906) have already
emphasized the importance of this fact in regard to bacteriology.

(c) THE STUDY OF PLEOMORPHISTIC PROBLEMS.

-It goes without saying that in every case where morphological and phyaiolopical alterations
of the bacteria are made the object of more or less thorough studies all five explanations
discussed on the foregoing pages will have to be taken under consideration. Thus
far dogmatism and standardization have acted too frequently as some kind of ‘‘theoretical
blinders” in bacteriological investigations. Moreover, the practical aims of most of the work
done, and the resulting impossibility to follow up interesting new observations in a correct
scientific manner, have also to be borne in mind as important reasons why the whole situation
is still so unsatisfactory. To discard indiscriminately all unexpected cases as ‘‘contaminations”
and ‘‘involution forms” is a practice which certainly should not be followed more frequently
than is absolutely necessary.
38 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

To decide correctly which explanation will be applicable in one or the other case is
not always easy. It: may be’ done either by continuous, direct, microscopical examina-
tion or by isolating and studying single-cell cultures, or by comparing results of suitably arranged
and sufficiently extended parallel experiments. Every one of these three ways has its advan-
tages and disadvantages; none of them is principally superior to the others.

The continuous direct microscopic examination has been repeatedly declared to be the only
manner in which convincing results can be secured. However, this categorical statement is
more the outgrowth of unscientific dogmatism than the result of experimental experience. In
some special cases direct continvous observation is, indeed, highly recommendable, or even
indispensable. Formation and germination of spores and of other reproductive organs, budding '
and branching, the occurrence and meaning of apparently sexual processes, for instance, are
some points which will be studied advantageously with the living material under the micro-
scope. In other cases other ways will be preferable for securing conclusive results. What
De Bary (1884, p. 137) said is, of course, indisputable: ;

Das erste Postulat einer morphologisch-entwicklungsgeschichtlichen Untersuchung ist der Nachweis der organi-
schen Continuitét successiver Entwicklungszustinde.

But it is very disputable whether the highly unnatural conditions aouannes in a hanging
drop or agar block preparation on the stage of a microscope are best suited to bring out the
normal developmental stages characteristic for the bacterial life under natural conditions.’

It should also not be overlooked that many of the older publications upon the pleomorphism
of the bacteria have been the result of patient continuous microscopic studies. And yet prac-
tically all of them have been discarded as not convincing, some of them certainly with good
reason. The latter holds true, e. g., for many of Hallier’s discoveries. But even the reports
of such well-trained. investigators like Rindfleisch (1872), Klebs (1875 a and 6), and Cienkowski
(1877), who unanimously stated that they directly observed the change of coccus to rod-forms
and vice versa, were later not accepted as conclusive proof against the monomorphistic dogma:
that a coccus is always.a coccus and a bacillus a bacillus, ‘‘and any evidence which may be
presented to the contrary is based upon untrustworthy methods of observation’ (A. C. Abbott,
1902, p. 51). Zopf’s fundamental researches upon the pleomorphism of the bacteria (1881) -
were all based on direct continuous observation, and that Winogradsky’s opposite findings,
(1887-1888), which by monomorphistic writers have always been quoted as final evidence
against Zopf’s work, though they were obtained with the same method and with impure cultures,
too, that they, in fact, are by no means very accurate and convincing, has been sufficiently
shown by Zopf (1895) in a special reply, which hardly ever has been quoted. That it is possible
to keep a form constant for some time under constant conditions, as Winogradsky has shown in
these experiments, is beyond question. Millions of cultural experiments made according to
standardized methods have confirmed this fact. But it is equally beyond question that such
highly artificial experiments, in which no attention is paid to the innumerable modifying
influences of time and environment, constantly present and active in nature, can at best be
accepted only as part of the truth: As early as in 1880 Neelsen emphasized quite correctly,
after having discussed the various difficulties which may reduce the value of experiments based
on direct observations, especially in the case of motile organisms:

Endlich aber wire selbst nach gliicklicher tberwindung aller dieser Schwierigkeiten noch nicht viel gewonnen;
man wiirde zwar ein Bild des Entwicklungsganges haben, wie er in einem bestimmten, sich gleich bleibenden Medium
ablauft, aber dadurch keine Kenntnisse erlangen von den Modificationen dieses Cyclus, respective den anderen
Formen des Wachstums und der Fortpflanzung, wie sie in Medien von anderer und wechselnder Zusammensetzung
vorkommen.

Gruber (1885) and others have published similar. critiques. Metchnikoff (1889 6b) wrote
against Winogradsky that the continuous, observation i in the living state can not be considered
to be a ‘‘conditio | sine qua non’’ .

Tl ne faut pas oublier qu’il existe un ae nombre de parasites dont l’évolution ne 2 peut presque pas étre étudiée
sur Je vivant et dont le developpement est néanmoins suffisement connu.
ae

 

1Jn 1874 Billroth wrote in this respect: “Man wiirde auf diesem iciinstlichen Wess nur die unvollkxommensten, diirftigsten Formen
erzeilen, welche von den ausgebildeten so verschieden sind wie Hafer im Zimmer auf feuchte Watte ausgesiet von ippigem Hafer auf gutem,
eultivierten Feldboden.”’
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 39

More recently Meirowsky (1914 b, p. 84) has pointed out that especially in the case of the
protozoa nobody has been able to follow the whole life cycle of any species by continuous direct
microscopic observation:

Auch hier hat kein Auge an einem Exemplar den Zyklus ablaufen sehen, sondern man hat auf die Ritwicktang
aus verschiedenen, auf einander folgenden Bildern geschlossen.

The use of single-cell cultures has also been strongly recommended more recently, espe-
cially for solving problems in bacterial variability. A paper by Cole and Wright (1916) upon
the application of the pure-line concept to bacteria is a characteristic example how easy it is to
put forth such purely theoretical demands without furnishing any experimental support. They
believe that ‘‘pure lines” would show great constancy, but it is not mentioned in their paper
that actually all experiments which have been made so far have shown without any exception
that single-cell cultures give exactly the same results in pleomorphism and variability as do pure
cultures obtained from plates. The investigations of Kowalenko (1910), Baerthlein (1911),
Bernhardt (1912), Nyberg (1912), Penfold (1912), First (1914), Rosenow (1914), and of Almquist
(1917) have brought out this fact sufficiently. Baerthlein, Bernhardt, Nyberg, Penfold, and
Almquist confirmed their own experiments, but it is of special interest that First was able to
confirm with his single-cell cultures of the Finkler-Prior vibrio the results obtained by Firtsch
(1888) about 25 years earlier. The same holds true concerning Rosenow’s new experiments
upon the streptococci, as a counterpart to those of Kruse and Pansini (1892), and in regard to
Kowalenko’s Coli studies, as compared with those of M. Neisser (1906) and Massini (1907).

That by no means every colony appearing on the plate represents the pure progeny of
a single cell is an indisputable fact. But it is equally certain that repeated plating, controlled
by careful microscopic studies, and eventually supplemented by passage cultures in other
substrates, will finally furnish exactly the same “pure lines” as any of the single-cell methods.
Itis difficult to understand why a method, which always has been accepted as quite reliable
as long as the results obtained with it have been in accordance with the monomorphistic
theory, suddenly should become so highly suspicious, as soon as some conflict arises between
preconceived ideas and the facts before our eyes.

It is not to be denied that the study of single-cell cultures will occasionally prove very
helpful, or even necessary. All methods thus far devised for this purpose, however, have
their conspicuous disadvantages. The frequently very unsatisfactory growth of the isoluged
cells is one of them. In Barber’s (1907) experiments, e. g., 40-70 per cent of B. coli, otherwise so
readily propagating on all substrates, failed to grow. iirst (1914) succeeded with V. proteus
in only 10 per cent of all cases. That the chances for contamination are considerably increased
should also never be forgotten.

Generally, therefore, the standard methods which otherwise have proven their reliability
during the last decades will also have to be considered in the first place for conducting experi-
ments upon the life history of the bacteria. If we only abstain from the unscientific practice
to discard, without adequate testing, every culture showing some unexpected development
as “contaminated” or as containing “merely involution forms,” and every slide as “not
sufficiently clear” which shows something more than the monomorphistically trained eye is
inclined to see, those methods will usually prove quite satisfactory, provided that a sufficient
number of parallel experiments is suitably arranged, long enough continued and several times
repeated.

This latter point has been strongly emphasized by Gruber as early as in 1885, when
he first attacked the monomorphistic dogma then proclaimed by R. Koch’s followers (with
“religious fanaticism” as Gruber says). His own experiments showed him that globular,
rod-like and spiral forms are merely types of growth, not distinctive marks of species and of
genera. Changes from one form into the other were observed with pure cultures as often
as the experiments were repeated, and it was always possible to retrace the intermediate
steps to the starting point. In his “mutation” studies Baerthlein (1911) also accepted the
uniformity of the results obtained in a great number of parallel experiments as full proof of
their validity. Nyberg (1912) equally pointed out that’ it would be entirely improbable to
40 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

assume, that accidental contaminations would cause the uniform results which he secured
with 140 strains of various lophotrichous rods, all producing the same types of pleomorphous
colonies. Bernhardt (1915, p. 219) made the following remark upon this subject:

Wer zahlreiche derartige Umwandlungsversuche an verschiedenen Bakterien Monate hindurch selbst verfolgt
und sich immer wieder von neuem davon tiberzeugt hat, dass bei jeder Bakterienart die Verinderungen doch immer nur
in ganz bestimmter Richtung erfolgen, wird kaum je Schwierigkeiten haben, die Verunreinigungen, die selbstver-
standlich auf alten Platten nicht ausbleiben, als solche zu erkennen.

That experiments of short duration are hardly of any value has been discussed before
(see the quotations from Adami on p. 33, and from Stamm on p. 34).. The very great persistency
exhibited by some strains, as observed, e. g., by Stamm, may occasionally test the patience
of the investigator very much. However, appropriate experimenting usually will eliminate
difficulties arising from this source.

It has also been mentioned before that not infrequently strict adherence to monotonous
standardized laboratory methods has hindered, and will perhaps further delay such prog-
ress as should and could have been made in bacteriology long ago. There is unquestionably
a very wide gap between the highly artificial conditions under which the pure cultures
are often compelled to live in the laboratory and the actual situation the bacteria, are
confronted with in nature. In this respect it should be borne in mind that-various strains
of the same species of bacteria when recently taken from their natural habitat not only differ
more or less from strains of the same species kept for some time in “solitary confinement” in
the laboratory, but they also show very often considerable incongruencies when compared
with each other. Sets of accumulation experiments, as recommended by Beijerinck and others,
demonstrate this fact conclusively. The interesting results obtained by H. M. Ward (1895 a)
and by Stamm (1914) in their studies upon the development of bacteria in water, also show
clearly the sometimes deeply alterating influences of a more natural surrounding. In our
first preliminary paper upon the life cycles of the bacteria (1916 a, p. 685) it was pointed
out that it was very easy to obtain full insight into the complete life cycle of B. azotobacter
by inoculating all the strains on hand (numbering. 22) into their natural substrate, viz., into
soil. The important results obtained by Almquist in regard to the life cycles of pathogenic
bacteria are also mostly due to a studious application of the conditions to which these organ-
isms have to adapt themselves when living naturally as saprophytes.

A comparison of the different influence of the natural and of the more or less artificial
conditions as given in the laboratory would be rather incomplete, however, if the effect. caused
by symbiotic interaction would not find due consideration. That physiological activities. can
be, and often are, deeply altered, when bacteria are growing in symbiosis has been repeatedly
noticed. Details are to be found in the publications of Cornil and Babes (1890, Vol. I, pp.
229-237), Th. Smith (1894), Kruse (1910, p. 168), and others. But so far comparatively few
observations have been recorded which indicate that in morphological studies equally unex-
pected results may be caused by symbiosis. Lorenz (1892) obtained a clearly actinomyces-
like growth of B. erysipelatos suum, when he inoculated this organism in a filtered culture
broth of B. swisepticus. In gelatin, as well as in the animal, the bacilli reverted promptly
to their “typical” appearance. But more recently Rosenbach (1909) has demonstrated that
the complete life cycle of the organism named, as well as of those closely related to it (ery-
sipeloid and septicemia of mice), embraces, indeed, such an actinomycotie phase. Metchni-
koff (1894) found large club and dumb-bell-shaped forms with lateral buds in a cholera culture,
which contained a white coccus. From other researches, especially from those of Dowdeswell
and of Stamm, it is known that these forms, too, belong to the normal life history of V. cholerae.
Metchnikoff points out:

La constance de cette modification indique une influence particuliére du coccus blanc sur les fonctions du
vibrion.

Kurth (1898) discovered that the diphtheria bacillus is much inclined to change its appear-
ance whenever other organisms are present. Symbiosis with Streptococcus lanceolatus caused
the development of very short forms, which were hardly discornible from the pneumococci
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 4]

themselves. Ten years later Smirnow (1908) made exactly the same observation, from which
the following conclusion was drawn:

‘The change in morphology of the Klebs-Loeffler bacillus to a coccus and the return to its bacillary form is inter-
esting, as-it indicates the possibility that bacteria may change their morphology under certain favorable or unfavorable
conditions.

Similar changes were also observed in these experiments when the two organisms were
cultivated in ‘‘double test tubes” (one within the other), so that only the metabolic products
could penetrate into the other culture and no streptococci could be mistaken for diphtheria
bacilli, and vice versa. As practically all cases of diphtheria present mixed cultures of B.
diphtheriae, streptococci, and other symbionts, the importance of these findings is evident,
especially when ‘combined with the fact that several entirely independent observations
have been recorded, where nothing but coccoid forms were found in cases of true diphtheria
(Dale, 1910; Balfour, 1911 d). Some observations of Heinemann (1917), on the other hand,
clearly indicate that also in this case the ‘‘illegitimate”’ forms are, in fact, part of the normal
life cycle of this species, and analogous findings have been made with diphtheroid organisms
by Walker and Adkinson (1917), as well as by Mellon (1917).

The olive-tubercle organism studied by Petri (1907) quickly assumed characteristic ‘‘invo-
lution” forms when grown together with Ascobacterium lutewm. The changed forms remained
also in this case capable of further propagation.

Babes (1893) secured development of the fusiform bacilli found in cases of scurvey, as well
as in the normal mouth, only in such agar where streptococci had been grown before, and here
the bacilli also produced curved threads “longer and thinner than cholera bacteria.”” Proca
(1908) found anew that B. fusiformis grows much better in symbiosis with B. subtilis or strep-
tococci than alone. The typical spirochaete-like spiral forms with pointed ends found by other
investigators in pure cultures of B. fusiformis appeared readily in broth cultures of the symbionts.

The highly interesting diphtheria-like forms (club-, dumb-bell-, spindle-shaped, etc.) first
seen by Babes (1895) in pure cultures of the streptococci were also produced much more abund-
antly by the so-called Bact. Guentheri (Streptococcus lactis) when grown in symbiosis with
B. putrificus (A. Wolff, 1908). Lorenz (1909) noticed that the very pleomorphous strepto-
coccus, isolated by him from horses afflicted with ‘“Brustseuche,” gave a distinctly Strepto-
thrix-like development when cultivated together with an Aspergillus. Rosenow (1914), too,
emphasizes the importance of symbiotic effects in his report on the transmutations within the
Streptococcus-Pneumococcus group.

_ That occasionally such experiments with mixed cultures may be of great advantage in
solving special problems in regard to the life history of one or the other organism may be seen
from the following experience. One of my two Azotobacter strains, which first had passed
over into thin spore-forming rods, reverted readily back to the characteristic large round Azo-
tobacter type when cultivated under suitable conditions (in mannite soil extract). The other,
however, remained absolutely stable despite repeated transfers. Considering the fact that
in nature Azotobacter always is to be found in symbiosis with B. radiobacter, this organism
was added with the result that in very short time the large form reappeared, though it turned
out to be very unstable after renewed isolation (Léhnis and Hanzawa, 1914, culture No. 6,
figs. 15-19). In another experiment made by Kuntze (1904) a similar case has been observed.
B. oxalaticus, whose pure cultures when kept for some time on artificial substrates quickly
assumed the “normal” Subtilis-like appearance, looked very much like Azotobacter when grow-
ing in symbiosis with the so-called B. denitrificans agilis, which is probably a denitrifying variety
of B. radiobacter. Moreover, as early as in 1884 DeBary (p. 503) reported that his B. Megaterium
grew in rows of large cocci, which also broke up into single cells, and which never produced
spores (all this like Azotobacter) when this organism was kept in mixed culture with a small
bacterium whose identity has not been established, but that they reverted to the typical spore-
forming bacilli when separated from this symbiont.

‘~ ‘The remarkable uniformity of the results obtained in these three cases may be accepted
as another indication that it seems to be recommendable for a correct study of the complete life
42 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

history of the bacteria to take under due consideration also such biological influences, per-
manently active in nature, in addition to the many physical and chemical factors which may
often cause important reactions on the side of the bacteria. These, perhaps, never would be
discovered if we continue to adhere too closely to the ordinary laboratory methods.

It is beyond question that our knowledge of the true life history of the bacteria is still very
inadequate, and that we are far from knowing all facts about them which we should know.
That the ascertainment of these data is absolutely necessary before any correct decision can
be reached concerning the true character and the systematic position of the various species of
the bacteria has been emphasized by numerous authors, like Klebs (18756), Davaine (1876,
p- 20), Gruber (1885), Hueppe (1886, p. 84; 1891, p. 27), Cornil and Babes (1890, Vol. I, p. 18),
R. Koch (1890), Beijerinck (1898), H. M. Ward (1910), and others. With-regard to Zopf’s
classification, which at least tried to make use of what was known at that time upon the life
history of the bacteria, H. M. Ward made the following remarks in his important paper ‘‘On
the Characters, or Marks, Employed for Classifying the Schizomycetes” (1892), which also at the
present time deserves our attention:

Zopf’s classification, admirable as it is in many respects, is difficult to work in practice, because it is necessary to
have all the stages of development before we can decide on the position of a species; at the same time it should be
noted that in this very respect it is far ahead of the merely tabular classifications, used for hurriedly determining the
name of 4 form, as a good flora is ahead of a mere museum catalogue of plants . . . It is, in fact, just in respect
of this particular attention to all the facts in the development of the species that Zop/s classification is scientifically
so far in advance of his predecessors . . . The matter of difficulty of application can not be urged as a reason for
desisting from obtaining and recording all that can be discovered regarding an organism . . . On the other hand,
pathologists, hygienists, chemists, etc., often do not care, what vagaries the organism exhibits, so long as they can
recognize it when they meet with it. As a'matter of experience, however, it is just these vagaries that bring about
the sources of error which beset them on all hands, and hence they are equally interested with the botanist in having
them cleared up, and explained.

It has been pointed out before, that it is a fundamental mistake to assume that the acknowl-
edgement of the pleomorphous character of the bacteria be equivalent to an attempt to oblit-
erate all bacteria species. That many of the so-called species, introduced into the literature,
often by the dozens, by quite inexperienced recruits in bacteriology, will have to disappear is
of course, beyond question. Such an elimination of entirely worthless ballast, however, has,
nothing at all to do with a cancellation of well described species. The narrow limiting of species,
as advocated by F. Cohn, R. Koch, Winogradsky,; Migula and others, should have been
abandoned long ago, and, as was shown by numerous quotations, it has indeed already met
with much well justified opposition. Jordan (1909, p. 339) aptly remarks:

The features that should characterize a bacterial race or species have not yet been established, and until some
consensus of opinion on this point has been reached, discussion of specific identity is futile. ”

The following lines by E. F. Smith (1905, p. 155) are equally worth quoting in this con-
nection:

Many names have been given unaccompanied by any proper description of the organism . . . To found, for
example, a new species of rabbit on the observation that a small jumping animal about the size and shape of a rabbit
has congregated in certain turnip fields and caused great damage, and apparently had destroyed no other plants,
would only serve to provoke a smile or to raise a doubt as to the author’s mental condition, and yet descriptions equally
worthless are not at all uncommon in systematic bacteriology.

A still earlier statement made by Hill (1902a) has also lost nothing of its validity:

It is easy to say that the whole subject needs much development before consideration of the names to be employed
is necessary. Names, indeed, can not be intelligently selected until the facts themselves are thoroughly established
and correlated.

So far, the variability of the bacteria alone has been more closely considered in connection
with these problems, and it was usually from this point of view that the establishment of well-
defined groups has been advocated by Kuropean and American authors. That truly natural
species can be founded, however, only after a much more thorough knowledge of their com-
plete life history will be available, hardly needs to be emphasized. _ The details discussed on the
following pages will help to make clear how much work remains to be done along these lines,
and what new insight into the natural relationship among the bacteria may be gained from
No. 2.1] LIFE CYCLES OF THE BACTERIA—LOHNIS. 43

such studies. Moreover, these facts in their entirety furnish a welcome support to the follow-
ing statement made in our first preliminary report on this subject (1916a, p. 677):

The development of the bacteria is characterized not by the irregular occurrence of more or less abnormal forms,
but by the regular occurrence of many different forms and stages of growth, connected with each other by constant
relations.

Unquestionably many so-called species, frequently described in the most superficial manner, will have to be can-
celled, because they merely represent fragments of the life cycles of other bacteria. Good species, on the other hand,
will not only keep their position, but they will receive a much more complete and sharper definition than they now
have.

2. MORPHOLOGY OF THE DIFFERENT GROUPS OF BACTERIA.
(2) COCCI.

That the old form-genera Micrococcus, Sarcina, and Streptococcus each comprise severa
groups of organisms which will present themselves as natural genera as soon as their full life
cycles will have been studied is already clearly indicated by several observations recorded thus
far. That, however, such highly variable marks as pathogenicity and pigment production
can not be considered to be of fundamental value in this respect has been discussed.
(Concerning pigmentation of micrococci see especially Neumann, 1897; in regard to patho-
genicity of micrococci, Geisse, 1914; of streptoccocci, Heinemann, 1915, and upon carbohydrate
fermentation of streptococci, Broadhurst, 1915, and Henrici, 1916.) It is quite true that Micro-
cocci as well as Streptococci exhibit a rather well recognizable general character. But there is
a number of cases where it is difficult or even impossible to decide clearly whether the organism
should be classed as a Micrococcus or as a Streptococcus. When I (1907) tried to arrange the
numerous so-called species of lactic acid bacteria and their relatives in some fairly well
defined groups, I had to dwell upon this particular point. Soon afterwards Babes (1908)
showed experimentally that even an otherwise typical Staphylococcus aureus may occasionally
assume the morphological appearance of a Streptococcus. To the same author we owe the
first distinct results concerning the close relationship existing between Streptococci and the
diphtheria group (Babes, 1895). The analogous relations connecting micrococci and actinomy-
cetes have been more recently emphasized by Beijerinck (19142). Comparatively the most
conspicuous heterogenicity is noticeable among the Sarcinae. Some of them are undoubtedly
merely types of growth of micrococci. (See Lehmann and Newmann, 1912, p. 198; Léhnis,
1907, p. 146.) Others, however, evidently belong in the life cycle of typical rods, especially of
spore-forming bacilli.

That most of the so-called Micrococcus species created by F. Cohn were no micrococci at
all has been mentioned on page 11. And it has also been pointed out that statements made by
Fraenkel and Pfeiffer (1895), A. C. Abbott (1902), and others, maintaining that an absolute
constancy of the cell form is to be found with the micrococci, are in open conflict with numerous
well established facts.1)

It may happen, of course, that one or the other strain will be temporarily constant,
sometimes even for a comparatively very long time. In our experiments, e. g., two strains of
Micrococcus candicans, one from Russian soil, the other from evaporated milk, exhibited very
readily and in a quite ‘uniform manner the characteristic stages of their life eycle, of which some
forms are reproduced in figs. 1-4, Pl. I, while on the other hand a Micrococcus luteus was quite
irresponsive; but a careful eéueoh: revenlad’ in this case, also, at least some inclination to form the
large club forms characteristic for this group. The use of salted substrates has helped Matzu-
schita (1900) to get the peculiar rod-like forms with Micrococcus rubefaciens and M. flavus,
as shown in figs. 5 and 6 on Pl. I”, while Maassen (1904) and Péju and Rajat (19062) did not
see any alteration of the micrococci tested on similar, substrates. The yellow Kokkobacillus

 

_dIn the axpianstion to Table I of their “Atlas,” Fraenkel sl Preifer say: Mit grosser Zahigkeit halten die Mikrokokken unter allen
Umsténden an der ihnen eigéntiimlichen Form fest, und mit Recht werden sie deshalb auch als der beste Beweis fiir das vielumstrittene Gesetz
von der Constanz der Form bei den Bakterien angesehen. Aussere Einfitisse jederArt bleiben dieser Eigenschaft gegeniiber vollkommen machtlos,
und ein echter Mikrokokkus erscheint stets nur in seiner streng kugligen Gestalt.”

2 In these two cases the rod-like forms were abundant, whereas with M. luteus, candicans, aurantiacus, coronatus, flavus desidens, aerogenes
Sarcina lutea, and agilis only occasionally stretched forms became visible among the cocci.
44 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI,

zymogenes, isolated by Biedert (1885) from saliva, grew as a coccus, rod, or thread. Some of
the straight forms contained coccoid bodies, whose further development was not observed.
The yellowish ‘‘Mikrokokkus B” of Malapert-Neufville (1886, p. 57) from water showed at
first. typical globular forms, but stretched later into distinct rods which in their turn broke
again into micrococci. The yellow Micrococcus subnormalis of Hopkins (1898) from the
human mouth grew at first on agar, gelatin, broth, and potato as a rod, later as a coccus; on
serum, however, it showed the form of a coccus from the start. Whether these three organisms
were identical or not must naturally remain an open question, though it seems quite possible.
The photographs accompanying Hopkins’ paper are very much like our pictures of M. can-
dicans. Fven branching and budding are clearly visible, although they were not mentioned
by the author. Changes from cocci to rods, diphtheroid and yeast-like forms also have been
recorded by Young (1914).

Best known, however, are the transformations exhibited by M. melitensis, which species,
on account of its morphological instability, has been transferred more recently by some authors
to the bacilli. Durham (1898), who published the first fairly correct: description of this organ-
ism, found out that both agar and broth gave exclusively micrococci when kept at 37° C, while
at 18-20° after a few weeks almost exclusively bacilli were visible, about two to four times
longer than broad, and sometimes curved. But brought back to 37° they reverted to ‘‘the
ordinary coccus form in pure culture.” Babes (1903) gave another description and ‘good
pictures of Micrococcus melitensis. Part of the forms, especially those of club-like appearance,
resemble very much certain pseudo-diphtheria bacilli. According to Babes’ opinion, relations
may exist with bacteria occurring in bronchitis and whooping cough, perhaps also with the
influenza bacillus. Gunther (1906, p. 771), as well as Lehman and Neumann (1912, p. 231),
consider the organism to be a coccus which occasionally produces rod-like, clubbed, and other
forms. . Jordan (1916, p. 416) calls it Bac. melitensis. The picture accompanying his descrip-
tion, reproduced as figure 7 on plate I, makes an interesting counterpart to the rod-like forms
of other cocci (figs. 2-6); in ‘the center of figure 7 cocci are seen to bud out of a rod-form, and
very similar things are visible in figure 8. The rods exhibit in all these cases the same irregular
appearance, as is common with the Mycobacteria (especially with those of the diphtheria
group). yre (1907) is strongly opposed to the idea of placing: M. melitensis among the bacilli.
He thinks that all bacillary forms are either stretched cocci or irregular “involution” forms.
His figure 2 shows all types pictured in our photographs of M. candicans. Exactly opposite
is Satsawa’s (1911) standpoint. This author is certain that the organism is a typical short
rod, not a coccus. In his photographs cocci budding forth from rods are again very con-
spicuous, but in the paper itself nothing is said about this interesting fact. Evans (1918) made
the interesting discovery that six strains of M. melitensis behaved morphologically and
biochemically, as well as in the agglutination test,,as did B. abortus.

The organism introduced by F. Cohn (1872) as Micrococcus phosphoreus is at present
classified, probably by all authors, as Bacterium phosphoreum. According to Molisch (1903,
1904, p. 60) typical cocci are exclusively formed on salted agar, while in salted gelatin all kinds
of irregular rods, clubs, etc., become visible. On agar and potato cocci were found in the begin-
ning, which were later supplanted by rods, clubs, and branched forms.

_ F. Cohn’s Micrococcus cyaneus has been studied anew by Beijerinck (1914 a) and renamed
Actinococcus cyaneus on account of its relationship to Actinomyces. It grows usually blue,
but red in acid-producing varieties, and white when reinoculated from old material. Accord-
ing to the Dutch author the new genus Actinococcus should embrace all micrococci which ex-
hibit clearly Actinomyces-like characters, and it should be placed in the neighborhood of Coryne-
bacterium, Mycobacterium, and Actinomyces.

In this connection, and in addition to the confirmative results mentioned above, the old
observations of Rabe (1886) concerning his Micrococcus botryogenus regain new interest. Growth
within the body looked very much like the well-known clusters of Actinomyces. (See fig. 276
on Pl. XXII.) On the artificial substrate, however, only normal micrococci grew, though in
the clusters short chains were seen, as were also obtained in cultures by Babes (1908).
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 45

Several other little known species, viz., M. coronatus Fliigge, M. coralloides Zimmermann,
M. viticulosus Katz, M. polypus Mig., M. nubilus Fontin, and M. vesicae Heim (see Lehmann
and Neumann, 1912, p. 234-235) are interesting on account of their tendency to form branches
around the edge of their colonies and along their stab cultures in gelatin. The inclination to
assume rod-like forms seems also to be not uncommon in these cases. I. viticulosus reminded
Lehmann and Neumann (1912) somewhat of their Bacterium vulgare (Hauser’s Proteus).

-Meningococci and gonococci are also by no means the simple constant cocci, which they
are often supposed to be. The meningococci are especially inclined to exhibit a great varia-
bility in their cell form, as well as in their stainability. The occurrence of very small up to
very large cells has been declared by Kutscher (1907) to be characteristic for this species.
Small cocci were seen by Lehmann and Neumann (1912, p. 223) attached like buds to the large,
deeply staining “giant” forms. That the positive or negative reaction with the Gram method
can not be used to distinguish two different species (the Diplococcus intracellularis meningitidis
Weichselbaum and the Meningococcus intracellularis Jaeger) is proven by the investigations of
Kohlisch (1915) and of Hort, Lakin and Benians (1915). According to Kéhlisch, not only the
Gram staining, but also the form of the colonies and the behavior in the agglutination test,
may vary greatly. Hort and his collaborators called the attention to the important fact that
the smallest granular forms, produced by the meningococcus, are able to pass the Berkefeld
filter. On “nasgar” plates minute colonies were grown from the filtrate,and in one well isolated
colony the following cell forms were found: (a) Gram-negative diplococei (Weichselbaum’s
cocci), (6) small Gram-negative bacilli, (c) Gram-positive and Gram-negative cocci (Jaeger’s
cocci), (d) Gram-negative rods (“biscuit bacilli’) with mostly Gram-negative, sometimes
Gram-positive granules, (e) Gram-positive bacilli of unequal size.’ ‘The apparent multiple
spore-formation in, and the budding of the “giant” cells, was accepted later by Hort (1917 6)
to be sufficient reason to remove this organism entirely from the bacteria and to place it among
the Hemiascomycetes. As was mentioned on page 22, and will be more fully discussed in Chapter
II, both processes are quite common with all bacteria.

H. Herzog (1913) has published another highly interesting paper upon similar observations
made with Neisser’s Gonococcus. Analogous changes between large and small (filterable)
forms take place, which will be more fully discussed in Chapter II, their reproductive character
being quite apparent. A reproduction of his figure 6, shown as figure 2 on Plate A, throws an
interesting light on the relations existing between giant coccus-, dumb-bell-, and micro-forms,
which Herzog also observed with meningococci.

In the genus Streptococcus irregular rod-, club-, and dumb-bell-shaped forms are also by no
means rare... Rindfleisch (1872) probably was the first to observe directly under the microscope
the breaking up of rods into globular forms and the melting together of streptococci to rods.
Billroth (1874) discovered, by direct observation, that the lactic-acid streptococci are able to
produce buds and clubbed rods. He distinguished the streptococci from the micrococci gener-
ally by their being inclosed in a cylindrical sheath, which was sometimes seen to be partially
emptied. Rosenbach (1884) pointed out that size and staining quality of pathogenic streptococci
may vary widely, even within the same chain.. Maddox (1886) found clublike forms of Strep-
tococcus lactis frequently in cultures one month old. An organism described by Cornil and Babes
(1890, Vol. I, p. 506) as causative agent of an ‘‘endocardite ulcéreuse’”’ seems to have been a
rod-forming Streptococcus:

. Batonnets de lu d’épaisseur, de 2 4 3 ou 4u de.longueur, elles se disposent parfois en chainettes; les batonnets se
changent en diplococci ou en microcoques plus ou moins arrondis . . . Parmi ces chatnettes on trouve aussi des grains
ronds colorés de la méme facon, plus gros, de ln, 5 de diamétre.

Pansini (1890) recorded with various streptococci (his Nos. 4 and 5) large oval and rod-like
“involution” forms. Two years later, however, in Kruse and Pansini’s important paper upon
the relationship existing between pneumococci and pyogenic streptococci (1892, p. 283) it is said
concerning their bacillary forms:

Involutionsformen méchten wir dieselben nicht nennen, im Gegenteil schienen sie in diesem Falle binge Excess
im Wachstum ihren Ursprung zu verdanken, denn die Culturen gediehen viel besser als anfanglich.
46 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

All intermediate stages were seen, rods showing beginning segmentation, and also large
yeastlike cells. The fact, clearly brought out for the first time in this paper, that pneumococci,
as well as other pathogenic and nonpathogenic streptococci, can not be separated as true species,
but must be looked upon as more or less stable varieties, occasionally passing over into each other,
has since then found ample confirmation by publications of Lemoine (1896), Lehmann and
Neumann (1912), Baerthlein (1912), Rosenow (1914), Heinemann (1915), and others.

Arloing and Chantre (1894) noticed that the rod forms, whose greater vitality Kruse and
Pansini had demonstrated culturally, were also more virulent in the animal test; the French
authors also observed the regressive metamorphosis from the rods to the streptoconci, and they
concluded therefore:

Il est probable que certains bacilles pyogénes ne sont que des streptocoques modifiés . . . Quand on trouvera des
bacilles associés au streptocoque pyogéne, il sera prudent de ne pas conclure nécessairement 4 une association mi-
crobienne.

j The formation of clubs, branches, rods, large and small ovals, rhomboid bodies, apical
growth, etc., with streptococci has been fully demonstrated by Babes (1895), who draws from his
observations the following conclusion (p. 418):

Ich stehe nicht an, eine nahe Verwandtschaft besonders jener kurzen Streptokokken, welche hiufig Kolben bilden,
mit den Diphtheriebacillen und Sbnlichen Bakterien zuzugeben.

His figure 2 contains so many interesting details, which undoubtedly will gain in a Tipertanod:
when studied more closely in further investigations, that a reproduction of it is given as figure
3 on Plate B.

That contrary to a widespread assumption the streptococci do not only divide transversely,
but also longitudinally within the chain, was also discussed by Babdes and indicated in his drawings.
The same fact and its connection with the formation of branches was further emphasized by
Crookshank (1896, p. 179), who also dwelled upon the variable size and form of the cells, espe-
cially of those at the end of the chain. Evidently not being acquainted with Billroth’s early
findings, he continues:

Another character which is very striking, may be seen when the individuals in a chain have become separated; an
unstained or faintly stained membrane may be found bridging across the interval.

Such sheathed streptococci are also visible in Babes’ sketches. They too furnish an inter-
esting counterpart to similar sheathed granular forms produced by the Mycobacteria.

The Streptococcus aggregatus, isolated by Seitz (1896) from the mouth, was found to possess
a wide pleomorphism of cells and of colonies. The latter were either small, looking like dewdrops
or large, whitish, similar to the colonies of B. pneumoniae, or wrinkled and folded, like those of
B. mesentericus. The size of the cells varied greatly. Regular ‘‘nests’”’ were not rare, which
contained all kinds of cocci, from the very smallest up to large ‘‘giant”’ forms, sometimes united
in tetrades. Branching also was noticed. The globular forms changed into egg, onion, spindle,
and lentil shapes, as well as into distinct short and long rod forms. All these various cells were
sometimes present within the same chain. They were only partially Gram-positive. With this
Streptococcus a diphtheroid organism was found in the mouth, whose cells and colonies remarkably
resembled those of the Streptococcus.

A Leuconostoc Lagerheimi is, according to Ludwig (1896), dimorphous like Leuconostoc (Strep-
tococcus) mesenterioides, viz., within the slime appearing as a regular Streptococcus, but when
liberated, a motile short rod. Beijerinck (1898, p. 211), however, thinks that “Leuconostoc lane
heimé” belongs actually to B. xylinum Brown.

The so-called Micrococcus Sornthalit of Adametz (1895), on the other hand, shows all marks
of a true Streptococcus. On solid subtrates after some time large yeastlike cells became visible,
which stained much better than the ‘‘normal’’ cocci. '

Stolz (1898) found the large cells of pneumococci and pyogenic streptococci most frequently
on serum agar. Rodlike forms, as well as clubs with short stems, singly or in short chains, were
also present. Small cells and clubs showed often distinctly longitudinal division which led to
branching, as was described before by Babes and by Crookshank. Vincent (1902) saw in pleuritic
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 47

exudate numerous bifurcate chains of streptococci, which continued growing in this manner
when kept in a mixture of broth and human serum. Secondary branches were also formed,
which, however, were very fragile like the others. :

The slime-producing Streptoccoccus, which Hlava met in scarlatina, and for which the name
Leuconostoc hominis was proposed by him, makes another excellent example of the relations con-
necting the streptococci and the diphtheria group. Some of Hlava’s drawings, reproduced as
figure 4 on Plate B, represent clearly all forms enumerated in the papers mentioned before. The
transverse budding and branching, as well as the diphtheroid and yeastlike forms, are of special
interest.

Thiercelin’s Enterococcus is another example of a more thoroughly studied Streptococcus
strain which exhibited all the various cell forms, characteristic of this group. In his first paper
(1899) Thiercelin enumerates: Streptococci of very variable size, ovals with or without capsules,
tetrades, staphylococci, short and long bacilli, and very large oval elements. He says:

Ce microbe est doué d’un polymorphisme des plus remarquables et dans les cultures et dans |’organisme.

In his second paper (1903) the author reports that the very different forms abundantly
present in old cultures continued to grow as such after being transferred to fresh substrates.
On agar and in serum long threads, sometimes filled with granules, were also found. Thiercelin
comes to the conclusion that the coccoid forms represent the young form of a bacillus:

“L’entérocoque est la forme jeune d’une bactérie (entérobacille ou entérobactérie) . . La forme d’involu-
tion semble étre un effort fait par le microbe vers cette forme bactérie.”’

Michaelis (1902) discovered large accumulations of clubbed forms of streptococci
in pleuritic exudate. Because he was unable to obtain any development from them on his
substrates, he feels sure that they are ‘‘degenerate” and the beginning of “bacteriolysis.”

In a diphtheria test Ohlmacher (1902) encountered a streptococcus of such peculiar
behavior that a doubtful diagnosis concerning the diphtheritic or nondiphtheritic nature of
the case resulted. In broth typical streptococci appeared. On Loeffler’s serum, however,
very polymorphous rods with clubs or swollen centers were seen, together with other forms,
which looked very much like diphtheria bacilli. The author agrees with Babes that such
growth of streptococci may cause mistakes in the diagnosis of diphtheria.

The pronounced rod form occasionally assumed by Streptococcus lactis was the cause that
this Streptococcus as Bacterium lactis acidi Leichmann, or Bacterium Guentheri Lehmann et
Neumann, has been wrongly placed in another genus, though its close relationship especially to
Pneumococcus has been pointed out already by Leichmann (1900, p. 324), and then more
fully by Kruse (1903) and by Léhnis (1907-1911). Its changes in morphology have been
also discussed by Diiggeli (1905) and by A. Wolff (1908), who found out that the presence
of B. putrificus greatly enhanced the tendency of the lactic-acid streptococci to pass over into
spindle-, dumb-bell-shaped, and otherwise diphtheroid forms.

The Streptococcus isolated by Lorenz (1909) from horses afflicted with ‘Brustseuche’
gave also small and large coccoid cells, some of them growing very similar to Staphylococcus
pyogenes aureus, fine rods resembling B. erysipelatos swum, and branched diphtheroid and
Actinomyces-like forms. The appearance of the last-named type of growth was especially
conspicuous in the presence of an Aspergillus, and it was most frequently noticed on the surface of
solid substrates, while in their depth, as well as in liquid media, the typical streptococci persisted.
The changes in morphology were accompanied by equally marked alterations in virulence.

Taddet (1909) tried Streptococcus pyogenes, choreae, and erysipelatos on different substrates
(broth diluted with water, or broth mixed with 0.25-1 per cent tartaric acid or lithium chloride)
and found that all his strains, though rather stable on ordinary broth, developed on the changed
media to rods, very small and very large ovals (up to 6-7u). These altered forms remained
constant, except when brought back into ordinary broth. The morphological changes were
accompanied by a speedy loss of virulence. That essentially the same effect was secured
by diluting the broth as by adding the substances named is of special interest.

160266°—-21—_-4
48 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

In sugar media, used by Broadhurst (1915) in experiments upon the fermentative reactions
of streptococci, also often considerable morphological changes became apparent. Rod-like,
clubbed, pear-shaped, and obtusely diamond-shaped cells, as well as longitudinal dividing,
were observed.

Mallory and Medlar (1916) say of their new B. scarlatinae that it approaches more nearly
the strepto-pneumococcus group than the diphtheria group. It varies from coccus to long
bacillary forms, grows best anaerobically, and is gram-positive. Its etiological significance
remains to be studied. .

Quite recently Kraskowska and Nitsch (1918) published some very good illustrations and
a fairly complete description of the pleomorphism of streptococci, isolated from sick and from
healthy persons. Their assumption that no such observations had been made before is, how-
ever, far from being correct.

The pleomorphous Streptococcus, which Rosenow, Towne and Wheeler (1916), Nuzum
and Herzog (1916), Rosenow and Towne (1917), as well as Mathers (1917), consider to be the
causative agent of poliomyelitis, a view which is not shared by Amoss (1917), Bull (1917), and
others, shows once more the morphological changes characteristic to the streptococcus group.
In this case the attention naturally was centered upon the production of minute, filterable
coccoid forms, which, however, are by no means absent in the cultures of other streptococci.
In this respect, too, the observations of the earlier investigators are in complete agreement
with those of the authors named. Some of their photographs are reproduced as figures 9 and
10 on Plate I. For comparison one of our photographs of Streptococcus lactis (Léhnis and
Smith, 1916 6) is added as figure 11, showing analogous differences in form and size. The
picture of Streptococcus tyrogenus, reproduced as figure 12 on Plate I, was published by Migula
(1900, Vol. II, Pl. I, fig. 2) as an illustration to his statement that the cells of this species are
2uin diameter. The photograph, however, exhibits clearly large and very small forms, budding
ovals and other details which were not mentioned by the author. The Streptococcus equi, repro-
duced as figure 13 on Plate II from another photograph of Migula (1. c., Pl. II, fig. 5) makes
a very interesting counterpart to Rosenow’s poliomyelitis picture.

It is evident that these observations will lead to discoveries of still more fundamental
importance, especially when connected with those of Herzog and Hort, mentioned above. As
the true character of these minute, filterable forms, however, seems to be more that of repro-
ductive organs than of regular vegetative cells, the full discussion of their part in the life cycles
of streptococci, micrococci, as well as of all other kinds of bacteria, will better be taken up
from a general standpoint in Chapter IT.

That the old form genus Sarcina is a rather hetereogeneous mixture of more or less similar
cell forms, belonging to the life cycles of other organisms which are only partially related to
each other, has been indicated above. Some sarcinae are obviously a special type of growth
of various micrococci, as has been emphasized by Lehmann and Neumann (1912, p. 198) and
others. The differences in the size of the cells are usually still more conspicuous than with
micro- and streptococci, so that these authors explicitly state (1. c., p. 199):

Angaben iiber Zellgrésse haben wir bei den Sarcinen nicht gemacht, weil wir hier besonders unregelmiissige
Resultate fanden. Es macht den Eindruck, als ob die Zellen oft gewaltig wiichsen und dann rasch hintereinander
in acht Teile zerfielen.

Two copies taken from our second preliminary paper (1916 b) and reproduced as figures 14
and 15 on Plate II, may illustrate these differences in the cell form of Sarcina flava, when grown
on beef agar 1 day and 3 months, respectively. Two others, figures 16 and 17 on the same
plate, show the same organism after 6 days’ growth in ammonium-citrate-glycerine solution.
The breaking up of the larger forms into packages of small cells is fairly well discernible. The
irregular outlines and the very dark staining of the large cells of fig. 16 are caused by their
being newly formed from the symplasm, which progess will be discussed i in Chapter ITT.

Other sarcinae, however, will probably turn out, when studied more thoroughly, to be a
special type of growth of various bacilli, Baermnce (1901 a, p. 43, footnote) has already
pointed out that Miquel’s Urosarcina is no sarcina at all, but a bacillus, and a glance at the
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 49

drawing of Urosarcina Hansenii Mig., reproduced as figure 6 on Plate C from Miquel and
Cambier’s Traité (1902, p. 628, fig. 172), confirms this standpoint. The Dutch author
himself isolated some strains from soil which are said to represent intermediate forms con-
necting Sarcina and Bacillus Megaterium (l. ¢., p. 53, footnote). Planosarcina ureae Beij.,
differing already from most other sarcinae by their ability to form endospores, is also distinctly
inclined to produce small as well as large rods (Léhnis and Smith, 1916 a), though Ellis (1902)
was unable to get a single rodlike cell with this or with any other sarcina.

As fig. 18 on Pl. II a photograph made by Matzuschita (1900, original fig. 6) is reproduced,
showing the growth produced on salted substrates by a so-called Bact. bruneum, i. e. a brown
variety of B. fluorescens. These cells have the typical appearance of regenerative bodies, yet
they can be easily mistaken for sarcinae. A close study of Sarcina mobilis Maurea, the only
organism within this group which is able to produce fluoresceine, might be of interest in this
connection. Lehmann and Neumann (1912, p. 206) obtained the typical growth of Sarcina tetra-
gena from an old culture of Duclaux’s Actinobacter polymorphus. Contamination, of course,
is also in this case not absolutely excluded, though not very probable. At least, there are too
many parallels in the biological characters of Sarcina tetragena and of the B. pneumoniae group,
to which Duclaux’s Actinobacter belongs, to discard this finding lightly as contamination.
Ph. Eisenberg (1914) noticed that also in mutation experiments Sarcina tetragena and Bact.
pneumoniae resembled each other closely.

Figures 19-21 on Plate II are copies of some pictures of Azotobacter vitreum, published in
1914 as figures 22-24 in a paper by Léhnis and Hanzawa. This Azotobacter has a very firmly
fixed tendency to show all morphological characters of a Sarcina. Yet it is no longer doubtful
to me that this Azotobacter, like all others, is a type of growth of a spore-forming bacillus.
But it was for the first time, after this strain had been under cultivation and observation for
fully 10 years, that at least some large rods were produced and some inclination to form endo-
spores became evident.

Billet (1890, p. 213) made the statement that all bacteria may appear as sarcinae by dividing
themselves in three directions while in their ‘‘état zoogléique.” This statement, however, is
not sufficiently supported by facts. That, indeed, large packages of globular regenerative bodies
may develop from the symplasm of every kind of bacteria, will be shown in Chapter III. But
these irregular clumps bear, as a rule, only a slight resemblance to the typical sarcina forms,
and it seems not to be a general rule that such ‘‘pseudo-sarcinae’’ are produced, as with B.
fluorescens and azotobacter.

Sarcinastrum urosporae Lagerhewm is another example of a bacillus with coccoid reproduc-
tive organs, which occur in packets like Sarcina. The rods, according to Lagerheim (1900)
divide transversely and longitudinally, thus forming small cocci, which first increase in size and
then germinate again to rods. Similar processes have been seen with other bacilli, e. g. with
Bact. Fraenkelii Hashimoto, with Matzuschita’s ‘‘proteusartigem Luftbacillus” and with Bac.
amylobacter by Bredemann, to which we will have to refer on the following pages.

(b) NONSPORE-FORMING RODS.

Among the nonspore-forming rods the different Proteus varieties have been the first good
examples of a distinct pleomorphism, morphologically as well as physiologically. It is true that
Hauser (1885) at first made some concessions to the monomorphistic doctrine by creating three
different species (Pr. vulgaris, mirabilis, and Zenkeri); few years later, however, he admitted
that the differences existing between them were only sufficient to classify them as ‘physiological
varieties” of one species. The typical forms, according to his description, are small cocci and
short rods (0.4-0.6 X0.9-1.2 »). But there were also present in young cultures large rods
and threads (1 x7.5-37.5 mw), large round cells (of 1.6 « diameter), well developed spirals
(with 2-4 turns), long twisted ‘‘spirulines,” and large pear and club shaped so-called involution
forms (3-7 u), which, however, were actively motile, multiplied as such, and were replaced in
old cultures by the ‘‘normal” small rods and cocci.
50 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Proteus fluorescens, discovered by Jaeger (1892) as a causative agent in Weil’s disease, shows
otherwise all morphological and cultural characters of this group, but produces fluoresceine.
Kohlbrugge (1901) met with an interesting symbiosis of a proteus-like rod and a vibrio in feces.
Both grew better when united, and the vibrio simulated the rod in its different forms (cocci,
short rods and waved threads) to such an extent that only single-cell cultures secured the final
decision. The Bacillus proteus denitrificans of Héflich (1902) furnished always at first long thin
threads with pointed ends (0.3-0.4 20-100 u); they divided into long rods (5-10 p) and spirals,
these into short rods and vibrios, at last coccoid and yeastlike forms appeared. All short
forms reproduced long threads after each transfer. On account of its having several polar
flagella Lehmann and Neumann (1912, p. 428) are somewhat in doubt, whether it really represents
a Proteus form. Possibly the same dimorphism in flagellation (polar and peritrichous) occurs
here, as it does with B. coli.

The “proteusartige Luftbacillus,” briefly described by Matzuschita (1902), deserves our
special attention not only because of its inclination to longitudinal fission, but still more because
the various “involution” forms, as shown in the reproduction of some of Matzuschita’s draw-
ings in figure 5 on Plate B, will prove to be of special interest, when viewed (in Chap. II) in
the light derived from other observations on hitherto very little known reproductive processes
in the life history of the bacteria.

Under the designation ‘“ Bacterium pyogenes ramosum,” Stefansky (1902) added another
so-called species to the Proteus group, which in the pus from an abscess of the skin showed
exclusively the form of a coccus, while on the plate nothing but short rods became visible,
They produced within 24 hours longer rods and threads with spindle- and club-shaped inflations,
in broth large globules also, and Y-forms. On agar containing 5 per cent NaCl the highly pleo-
morphous picture, presented already by the other forms, became still more complicated by the
presence of ring-shaped, spirilla and spirochaete-like forms, as well as of branched cells.

Bacterium Zopfii, described by Kurth (1883) two years before Hauser published his mono-
graph on Proteus, is closely related to Proteus Zenkeri, as I have shown (1905 a, p. 720) by a
comparative study of a number of varieties. Kurth was not able to obtain a multiplication of
the cocci as such. They always reproduced rods, and were therefore considered by him to be
resting forms. They originated by fragmentation of the rods. Schediler (1887) confirmed these
findings by direct continuous microscopic studies. He also saw branched spirals. .

Whether Bacterium merismopedicides of Zopf (1883) belongs to this group can not be
decided, as only insufficient data are available. Its breaking up into cocci, forming tetrades
and sarcina-like arrangements, as shown in fig. 7 on Plate C, indicate, however, a.morphological
relationship to Bact. Zopfi Kurth, as well as to Matzuschita’s protean bacillus from the air.
Zopf saw motility of the “cocci’’, whereas those of Bact. Zopfii were found to be immotile.

The Lactobacilli furnish many more details, which do not fit the monomorphistic theory.
Long slender rods are typical, but short rods and cocci are by no means rare, and branching
can be observed so readily that some investigators not only discussed the relations connecting
lactobacilli and actinomycetes, but directly transferred the lactobacilli into the last-named
group. 0. Sternberg (1898) is one of the few authors who claims to have seen motile lophotri-
chous rods of the Boas-Oppler bacillus. The same author also recorded clostridia, plectridia,
and regular endospore-formation. (See Chap. II.) The two best-known lactobacilli from the
intestinal tract, Bacillus acidophilus Moro (1900) and B. bifidus Tissier (1900), are both very
much inclined to exhibit the characteristic branching. Moro found water cultures very useful
for this purpose. So-called degeneration forms (ring, club-shaped forms, spirals and others)
were frequent on agar after a few days. Some of Tissier’s drawings, reproduced as figure 8 on
Plate C,-show many features to be found with B. radicicola and other members of the B. pneu-
moniae group. That there are connections linking the lactobacilli on the one side with these
organisms, and on the other side with the streptococci, has been discussed in my paper on the
lactic acid bacteria (1907). Tissier (1. c., p. 91) declares the “formes vésiculeuses’”’ to be signs
of degeneration. But it seems more probable that they represent a special intermediary step
in the life cycle of this organism, as well as of others, e. g. with B. pneumoniae (see Toennies-
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 51

sen’s findings below). Branched forms are, according to Tissier, frequent in feces, on good
substrates after they are somewhat exhausted or have turned acid, and often under the in-
fluence of symbiosis with other bacteria. Just as with B. radiciola, the three ends of a Y form
are not infrequently the only parts which take the stain, and which are able to give further
development.

Another strain of the Acidophilus type, isolated by Cahn (1901), was given the special name
“B. aérobicus ramificatus’”’ on account of its abundant branching. Rodella (1901) found the
pleomorphism of this group so bewildering that he was unable to decide whether the different
strains studied by him should be given species or only varietal character. When treated with
the Gram method, the rods behaved very much like diphtheria bacilli, showing gram-positive
and gram-negative parts. Continued investigations led to the conclusion (Rodella, 1908) that
practically all lactobacilli from the stomach (Boas-Oppler bacillus, B. gastrophilus, acidophilus
and bifidus) are merely representatives of one single species. Bliihdorn (1913), on the other
hand, vigorously maintained that there be no similarity whatever between the two last-named
bacilli; they are said to be entirely different, morphologically as well as physiologically. Un-
doubtedly both extreme views are incorrect. There is neither a complete identity, nor an abso-
lute nonidentity. All lactobacilli together form a natural group, and not before their full life
history is known, we will be able to say, whether it is made up of different species or only
of a chain of varieties. At present, the latter view seems to be more probable. What surprises
are still in store for more thorough investigations is indicated by the discovery made by Noguchi
(1910), that Bac. bifidus is “an anaerobic phase of life of an aerobic sporogenous organism

closely resembling Bac. mesentericus fuscus.’”’ Noguchi was able to complete experi-
mentally the transformations both ways. Sandberg (1904) who also saw motility, made some
important observations upon the correlations existing between the different variable types of
colonies and the form of rods, produced by the “‘long bacilli” from the stomach.

The lactobacilli in milk, cheese, and other fermented food, exhibit the same instability of
appearance when growing under different conditions. Besides the “normal” slender rods,
long threads with inflations, as well as chains looking very much like streptococci, are equally
frequent. Henneberg (1901), Holliger (1902), Freudenreich and Théni (1905) and many others
have furnished valuable information upon these organisms, which was collected some years ago
in my papers on lactic acid bacteria (1907, 1911). From our present standpoint the following
publications are of special interest: Diiggeli (1905) and Weigmann, Gruber and Huss (1907) on
the organisms in the Armenian sour milk (matzoon); Rubinsky (1910) upon the koumiss bac-
terium; Chatterjee (1910) on the lactobacillus (mistaken by this author for a Streptothrix) in
the fermented milk of British India, called dadhi; Heinemann and Hefferan (1909) and Koegel
(1914) on the bacillus from Bulgarian sour milk (yaourt), usually, though wrongly, termed B.
bulgaricus. Undoubtedly the polymorphism of this group surpasses by far that of Proteus,
and is perhaps even more pronounced than that of the plague bacillus and of the diphtheria
group. Starting from the short, coccoid and club-shaped cells, practically all intermediate
forms may become visible up to the very thin threads with pointed ends, sometimes only 0.2-0.3
thick, not infrequently waved and coiled, and often branched. The plates attached to the
papers of Weigmann et al. and of Rubinsky may be consulted for good illustrations.

Among the organisms ot this group Bact. acidi propionict b of Freudenreich and Jensen
(1906) is perhaps the best example thus far known, exhibiting marks of relationship between the
lactobacilli and the diphtheria group. At room temperature appearing as a small short rod
(% x1), it becomes a long rod at 37° C., forming clubs which stain like the barred type of the
diphtheria group. Granules showing the Neisser reaction may be found occasionally with all
lactobacilli. The. “K6rnchenbacillen’ ’ of some German authors can not be accepted as a
special type.

That at least some of the old, poorly described Leptothrix forms are to be classified with the
lactobacilli can hardly be doubted. In some of the early publications of W. Miller (1882-1889)
the pleomorphism of this group has been already fairly well indicated, though, of. course, no
definite information can be gained from these necessarily very incomplete investigations
52 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor. XVI.

Bac. funduliformis Hallé (1898, pp. 28-31) and some other anaerobic nonspore-forming
bacilli seem also to be naturally related to the lactobacilli, especially to those displaying a
distinctly anaerobic character. On the other hand, they are also related to anaerobic spore-
forming bacilli. The same, however, holds true for the lactobacilli, as will be more fully
discussed in Chapter II. Hallé’s vaginal bacillus grows in the pus as a slightly curved thin
immotile rod, but in the cultures a conspicuous pleomorphism was noticeable, as illustrated
in figure 9 Plate C (from original fig. 3). Large rods and threads, with inflations, branches,
buds, clubs, balls, etc., were formed; in the animal tests oedema and gangrene were caused.

An anaerobic bacterium, found by Ghon and Sachs (1905) to be connected with peritonitis,
and either identical or closely related to B. funduliformis, gave also prominently small immotile
rods, slightly longer and thicker than the influenza bacillus, accompanied by cocci, threads
and inflated forms, when taken from the exudate. In the cultures the pleomorphism became
more enhanced by the presence of thick waved threads, club- and pear-shaped cells, ovals,
yeastlike, and spindle forms, all of which were Gram-negative, and whose development was
stimulated by adding 1 to 2 per cent glucose to the substrate. Kisskalt (1906) studied a similar
organism, isolated from an abscess, and emphasized its relationship to the Streptotrices.
Three other small Gram-negative anaerobic rods, isolated from meningitis by Ghon, Mucha
and Miiller (1906) are also to be included in this group. They are either identical or closely
related with each other. Their appearance after 4 days growth at 37°C. is illustrated in
figure 10 on Plate C (reproduced from original figs. 6, 9 and 20).

Several other organisms found in cases of typhus exanthematicus, whose etiologic sig-
nificance is still more or less questionable, are also to be mentioned here. The immotile
diplobacillus of M. Rabinowitsch (1909), showing bipolar staining, and the organism described
by P. Th. Miiller (1913), are apparently either closely related or even identical. In the blood
they appear as cocci, diplococci, and short rods; but in the pericardial liquid large globules,
ovals, yeast like and big dumb-bell or club-shaped cells occur. The figures reproduced
as Nos. 22-24 on Plate II (from original figs. 2, 5, and 1, respectively) show coccoid forms
from gelatin, 2 days old, rods from glycerine agar, 2 days old, and so-called degeneration
forms from serum broth, 8 days old. The latter especially indicate relations to the mycobac-
teria. In their paper upon the diphtheria group Trautmann and Gaethgens (1913) have pub-
lished a photograph (fig. 8, p. 72) demonstrating exactly the same type of clubs, cocci, and
short rods. Gram staining was either positive or negative with Miiller’s organism. Ploiz,
Olitzky and Baehr (1915) also isolated a pleomorphous, Gram-positive, small bacillus from
typhus cases, which grew either as a coccus, or as a short straight or curved rod, and which
was very much inclined to produce, in young cultures, so-called involution and degeneration
forms. The authors use these expressions, though they found that these special forms can
be made constant by frequent transfers, which fact undoubtedly does not fit their assumed
“degenerate” character. The filterability of the typhus virus is not admitted by these authors.
However, here as in other cases the observations of Hort (1915) seem to be an important step
toward the solution of the problem. It is very probable that his small Gram-positive or
Gram-negative filterable organism is nothing else than the gonidial stage of the pleomorphous
rod found by the other authors. In Chapter IT this point will be more fully discussed.

B. erysipelatos suum, B. murisepticum, and B. erysipeloides, considered by Rosenbach to be
different, while other authors accept them as identical (see Lehmann and Neumann, 1912,
p. 430), furnish another prominent example against the monomorphistie theory. First de-
scribed by Rosenbach (1884) as micrococcus, soon after, however, declared by him (1887) to be
neither a micrococcus, nor a bacterium, nor a bacillus, but an organism of peculiar systematic
standing, perhaps related to Cladothrix, it has been more recently studied anew by the same
author (1909), whereupon it was made the type of a new genus Erysipelothrix. Four of the
photographs accompanying the last-named paper have been reproduced as figures 26-29 on
Plate III. Figure 26 (original fig. XIII, 5) represents B. erysipeloides from blood serum, 7
days old. ‘Figure 27 (original fig. XII, 5) shows the same organism from broth, only 9 hours

aiding 905 3 3 pe :
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 53

old; branching, terminal swelling, as well as budding, are clearly visible. Figure 28 (original
fig. XIII, 3) demonstrates a spiral form of B. murisepticum in gelatin, and figure 29 (original
fig. XII, 3) a thin mycelial growth of B. erysipeloides from a 6-weeks-old gelatin culture,
resembling very much a similar development of lactobacilli (shown e. g. in Rubinsky’s koumiss
paper). Another picture of B. erysipelatos swum, taken from the “Atlas” of Itzerott and
Niemann (1895, fig. 40), has been added for comparison as figure 25 on Plate III. That also
in this case very different cell forms, as well as budding, are easily discernible, demonstrates
sufficiently how difficult it is to get ‘‘legitimate’’ monomorphous material; for it may be
safely assumed that the authors have selected a field as uniform as they could find. As early
as in 1889 Karlinski introduced a ‘new pathogenic schizomyces’’ under the name Bac. muri-
septicus pleomorphus into the literature, which already clearly exhibits most of the marks
mentioned by Rosenbach 20 years later as characterizing his Erysipelothrix. Karlinski empha-
sized the protean nature of his bacillus. The short rod was considered to be typical, but
“nearly all forms became visible, from small, globular and oval cells up to slender spirals,”
and in old cultures “terminal swellings appeared, often five times as thick as the rods and
threads,” which the author felt ‘‘compelled” to classify as “involution” forms. That un-
der special conditions, as mentioned on p. 40, Lorenz (1892) obtained an actinomyces-like
growth of B. erysipelatos suum is quite in line with these other findings. And it is very prob-
able, too, that the “Streptothrix” growth which Kitt (1897) found in his “Rotlauf” cultures,
when kept in serum broth, was not a contamination, as he later (1898) assumed, but really
the analogous type of growth, as in the other cases. (See also p. 20.)

» B, septicaemiae haemorrhagicae has been represented in the “Atlas” of Itzerott and Nie-
mann (1895, fig. 36, ‘‘Hiihnercholera’’) by the photograph reproduced as fig. 30 on Plate III.
It is certainly not easy to understand how the authors actually photographed unknowingly
something quite different from the typical bipolar rods, which they undoubtedly intended to
show. Evidently they were so well trained to see only ‘‘bipolar rods’’ that all the single
globular bodies, their budding (in the center of the picture), and their Staphylococcus-like
arrangement were passed unnoticed. It is a good example of well-developed regenerative bod-
ies, photographed but not seen. A comparison with figure 18 on Plate II (regenerative bodies
of a Fluorescens variety) might be recommended. Rosenfeld (1901) noticed that the addition
of salt stimulated the formation of ring-shaped, globular, thread-like, and branched cells in the
septicaemia group asin others. G. Phisalix (1902) found the mycelium-like growth of Pasteurella
in the body, while in the cultures the cocco-bacillary form reappeared; the virulence was the
same in both cases. Bang’s abortus bacillus has been classed by Nowak (1908) as a pleomor-
phous member of the septicaemia group, showing coccoid, branched, and various other forms,
but it seems more correct to place this organism among the mycobacteria, Klepzoff (1912)
succeeded in getting acid-fast growth from septicaemia organisms in the course of experiments,
which led to the discovery of relations existing between B. tuberculosis and Pasteurella groups.

B. pseudotuberculosis rodentium has been studied by Galli-Valerio (1903) and Zlatogoroff
(1904) in regard to its pleomorphism and its relation to B. pestis. Very similar small and large
as well as branched cells were found to be frequent with both organisms, A ‘‘new species”
Bacterium pseudopestis murium described more recently by Galli-Valerio (1913) may be also
accepted as an aberrant strain of B. pseudotuberculosis rodentiwm. Its plague-like pleomor-
phism deserves full attention with respect to the plague diagnosis.

The wide variations in the microscopical appearance of B. pestis have been made the
object of numerous studies; those of EZ. Klein (1897) upon the correlations existing between
cell forms and colony types of B. pestis apparently being the first of its kind. A very detailed
report upon plague morphology has been furnished by Albrecht and Ghon (1900). The repro-
ductions of a few of their drawings, given in figure 11 on Plate D, present a good illustration
of the very wide-pleomorphism characteristic for this organism.

Coccoid forms are considered to be the initial type of growth. They pass over into the
short rods showing bipolar staining, and therefore occasionally looking like diplo- and strep-
tococci. Later the cells increase in size, some swelling up to yeast-like, pear-, club-, or spindle-
54 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vox XVI.

shaped appearance, others stretching out into threads, which sometimes show branching or
a conspicuous swelling in the center. All these forms are present in the body, as well as in
the cultures. That the large cells are no ‘‘involution” forms, is not only proved by their
vitality, but still more by the fact that in several months old cultures the coccoid forms
exclusively are to be found. Budding and branching is frequent after not more than 48 hours
on glycerin agar. Skschivan (1900) saw it in only 1-day-old growth, and he is therefore
inclined to place the plague bacillus among the actinomycetes, next to B. mallei. Dieudonné
(1903) calls again all forms surpassing the ‘‘legitimate”’ coccoid or short-rod type, ‘‘involution
or degeneration forms,’ though they are, according to his own statement, fully virulent,
present already in the bubo after a short time, and most frequently obtainable in fresh isola-
tions. from the animal. Figures 31-33 on Plate III are reproductions from the textbook of
Muir and Ritchie (1903). Figure 31 (original fig. 148) shows the so-called normal coccoid and
rod-like cells from a young agar culture, figure 32 (original fig. 150) the ‘‘involution” forms
from 4 per cent salt agar, and figure 33 (original fig. 147) the microscopical appearance of the
organisms as taken from the tissue. There is unquestionably much likeness between figures
33 and 32, but hardly any between 33 and 31.

’ The difficulties occasionally arising from the microscopical similarity of B. pestis and B.
pseudotuberculosis rodentium have been discussed by Galli-Valerio (1903) and by Zlatogoroff
(1904) in papers referred to above. The conditions favoring one or the other type of growth
have been the object of special studies by Cacace (1903). Long and sometimes branched
Streptothrix-like threads were e. g. easily obtained in broth, containing 0.01-0.05 per cent
potassium chromate, while the addition of 0.5-1.5 per cent carbolic acid to the broth stimu-
lated the growth of cocci, which equally predominated on gypsum plates saturated with broth,
Abundant and uniform branching was secured by Kodama (1908) with 10 strains of B. pestis
when these, were cultivated on coagulated. egg albumen, Gotschlich (1906) reported upon an
avirulent plague “mutation” which grew in smooth, circular colonies, but reverted in two
of three cases to the original type, after having been kept for two months in the icebox. Row-
land (1912) was able to transform experimentally the plague bacillus into a strain showing
all characters of B. pseudotuberculosis rodentium.. In another paper he states that ‘‘probably
no organism presents.so marked a pleomorphism as the bacillus of plague,’ and he demon-
strates this by a large collection of beautiful photographs, two of which are reproduced on
Plate III: Figure 34 (original fig. 12, Pl. XIX) mycelial growth in immune serum, and figure
35 (original fig. 17, Pl. XXT) yeast-like: cells from the seat of inoculation in the rat. The cor-
relations existing between cell form, virulence, and appearance of the colonies have been studied
anew by Markl (1914). Highly virulent, freshly. isolated strains gave small thin colonies,
while others, more or less reduced in virulence by prolonged cultivation on artificial substrates,
produced large succulent colonies. The single cells, however, were larger and thicker in the
virulent than in the saprophytic strains. An entirely, avirulent pseudo-plague bacillus from the
soil, which I described (1905 6) under the name Bact. agreste, exhibited a surprising similarity
to the true plague bacillus, morphologically as well as physiologically, but differed from it by
being, at least temporarily, actively motile.

The various cell forms of B. influenzae have been demonstrated by Crookshank (1896) in
the drawing reproduced as figure 12 on Plate D (from original fig. 122). The filaments are com-
posed of long and short rods, coccus forms of different size, and of irregular elements. According
to Grassberger (1898) the latter are present especially in luxuriantly growing cultures. Thick
wedge-, spindle- and club-shaped cells are produced. True branching also was observed and
photographed. On horse serum all 40 strains tested developed uniformly very long threads,
which occasionally showed one or more inflations. D. J. Davis (1907) found that also in this
case the presence of 2-3 per cent sodium chloride in the substrate stimulated the tendency to
pass over into irregular large forms, somewhat resembling the diphtheria organism. But these
alterations were not thought by him to be characteristic.

A very remarkable polymorphism has been described by Waelsch (1905) for his B. involutus,
which Lehmann and Neumann (1912, p. 274) place close to the Morax-Axenfeld conjunctivitis
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 55

bacillus, but which also shows signs of relationship to the B. pneumoniae group. Very different
forms were already present in the colonies. After a few days’ growth representatives of practically
all of Cohn’s form genera appeared in the field. Figure 13 a on Plate D is a reproduction of the
original figure 4, showing the development on glycerin agar after three days. Figure 13 6
(original fig. 6) gives the picture of a 5-days-old broth culture.

Among the species or varieties belonging to the group of Friedlénder’s B. pneumoniae the
so-called Proteus hominis capsulatus, as described by Bordoni- Uffreduzzi (1888 b) represents a
form which in regard to its pleomorphism comes very near to B. involutus Waelsch. Besides
the cocci, rods, and threads, which are characteristic for the group, and large globules acting as
reproductive organs, very different spindle-shaped, curved, and triangular forms became visi-
ble on potato, which might be easily mistaken for ‘‘degenerate”’ spirilla. Figure 14 on Plate D
(a reproduction of original fig. 3 on Pl. VII) shows them, as well as some branching. As the
author explicitlysays, they are no ‘‘involution’’ forms, because they are formed when the develop-
ment is at its height, and they are replaced by the simpler forms when the culture grows older.

The B. mucosus isolated by Simoni (1900) from ozaena is another interesting example of
the pronounced pleomorphism of the B. pneumoniae group. In the body typical cocci predomi-
nated, while in the cultures, especially on solid substrates, the bacilli came to the foreground.
At the same time, however, so many other forms were already visible in young cultures, that
only by repeated experiments full certainty about the absence of contaminations could be
secured.

The ‘“‘new species,”’ isolated by Jehle (1902) from sputa in different cases of pulmonar affec-
tions, possesses all marks of Friedlinder’s bacillus. Cocci pass over into short rods, these
stretch out into long, thick, often curved threads, which eventually exhibit spindle-shaped or
globular inflations. All these various forms are present in cultures not older than 2-3 days.
Sachs’ ““Kapselbacillus’’ (1903) displayed practically the same pleomorphism.

B. Beresinewi of Lepeschkin (1904) should probably also find its place here, according to its
origin (sputum from a pneumonia case) and its general morphological and cultural characters.
It. produced the typical round, white, soft colonies, but was avirulent. The change between
cocci, rod-like, and thread forms followed the common lines, though the size of the cells was
larger than usual, and there was a very pronounced tendency to produce branched rods and
threads. This tendency could be increased and stabilized in single-cell cultures, and it was
much more noticeable with young than with old material. My own investigations (1905 a, p.
587) revealed very clearly that all members of the B. pnewmoniae group are easily induced to
display their ability of forming branches (so-called bacteroids). Péju and Rajat (1906 ¢ and e)
found potassium iodide, as well as urea, especially useful for stimulating the characteristic changes
in appearance; 1-2 drops KI favored the development of very large round cells (of 3 to 4 times
the normal size), while 10 drops increased the formation of long threads.

Some very interesting work upon the transformation of the large encapsulated ovals, not
infrequently seen with the Friedlander bacillus, into slender small coli-like rods, has been done
by Toenniessen (1913-14), who accepted these changes as ‘‘ mutations,” though they evidently
represent different stages in the life history of this organism. Six of his photographs (Nos.
III-VIID) are reproduced as figures 37-42 on Plate IV. The irregular intermediate forms, in
which the chromatine granule only takes the stain, deserve our special attention. They, too,
indicate the relationship existing between B. pneumoniae and radicicola. Their exact place in
the life cycle of members of this group will have to be discussed in Chapter III. The thin coli-
like rods proved to be entirely avirulent, but they reverted to the original short virulent type,
when transfers were made from 6-8 weeks old cultures. The Coccobacillus foetidus ozaenae
Perez, recently studied by H. C. Ward (1917), furnishes another example of the pleomorphous
nature of this group. Besides the well-known differences in shape and size of the cells, varia-
tions in the motility were observed. Some strains remained permanently immotile, while
others gained motility, thereby indicating, also in this respect, their tendency to approach the
Coli type. That in old cultures a dark brown pigment became noticeable, is not surprising in
56 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVE

view of the fact that the same behavior is often ascertainable with typical strains of B.
pneumoniae.

Two highly pleomorphous organisms, both probably related to B: colt, have been isolated
in a case of oedema by Harris (1893). One showed a brownish, the other a pink color; the
former was pathogenic, the latter was not. Both of them produced at first cocci and diplococci,
which later developed into paired stout rods (1 X 4-6), or thin long rods and threads (0.5 X 4-12y),
respectively. Sometimes the pink organism exhibited distinctly Leptothrix-like forms.
Budding and branching are also shown in the photographs accompanying the paper, but there
is no word about them to be found in the text. In a paper entitled ‘‘The Pleomorphism of the
Common Colon Bacillus,” Haslam (1898) discussed the influence of quality and quantity of
nitrogenous food upon the length and the width of the cell form. Adami, Abbott, and Nicholson
(1899) discovered in cirrhotic, and also in normal, livers of men and animals minute spherical
or oval bodies, which could be easily mistaken for pigment granules in the liver cells, occurring
singly or in pairs, sometimes in strings of three and four, which on artificial substrates produced
small colonies quite different from those of B. coli. Yet successive transfers led back to the
typical colony and cell form of B. coli. M. E. Abbott (1900) has published some more data upon
these investigations, to which reference will be made in Chapter II. Cultivation of B. coli
on salt agar furnished Matzuschita (1900) large monstrous inflated threads. Figure 36 on Plate
III is a reproduction of his figure 14. Another Coli strain gave large globules. Both forms.
are, as will be shown in Chapter II, of the same physiological value. W. Winkler (1900)
observed granulated spindle-shaped ‘‘swarming bodies’’ of B. coli, which are probably of the.
same rank as analogous forms discovered by Matzuschita as a type of growth of his protean
bacillus from air (fig. 5 on Pl. B). Ohlmacher (1902) isolated from a colon bacillus septicaemia
a highly pleomorphous coli strain which only after several transplantations (‘‘rejuvenation’’)
in broth assumed again the ‘‘typical’”’ behavior. ‘‘All gradations from minute coccoid or diplo-
coccoid to long coarse filamentous forms were observed.” Other threads were quite thin and
shadowy. Some contained a row of deeply staining inclusions. Others in young cultures
made from heart’s blood showed lateral buds and short true branches. Some of the rods and
filaments had median swellings, others clubbed ends. According to Prell (1917) B. colt may
live in a “rhabdocytic” and in a “‘coccocytic”’ phase, and other bacteria may grow in a similar
manner.

Walker and Murray (1904) discovered that 0.2 per cent saturated alcoholic solution of
methyl violet added to agar, broth or gelatin, caused typhoid and coli bacilli to grow in long
undulated, sometimes very thick, threads, which showed true branching and segmented in old
cultures into chains of cocci. The authors expressed the belief that these alterations indicate
‘‘an unexpected complexity in the life history of these microérganisms.” Data to be given in
Chapter II will prove the correctness of their assumption. Péju and Rajat (1906 6 and e)
obtained analogous results by adding potassium iodide or urea to their substrates; and these
were confirmed by other tests with urea made by Wilson simultaneously (1907). Revis
(1912 a) succeeded, by making use of malachite green, in changing one of three Coli strains to
such an extent that it now was ‘‘neither physiologically, morphologically, nor culturally a
colon bacillus.”” It produced slime and grew in long filaments or as very short rods. The
same author also studied (1912 6) the coccoid forms of B. coli and came to the conclusion that
relations seem to exist with Gram-negative micro- and streptococci, analogous to those
connecting Gram-positive streptococci and Gram-positive rods producing lactic acid. Lehmann
and Neumann (1912, p. 256, footnote) observed occasionally that a monotrichous Coli strain
became peritrichous, and the opposite change was recorded for B. alcaligenes faecalis, which
first was described to be peritrichous, later monotrichous (Lehmann and Neumann, 1912, p.
357). When Pollak (1913) cultivated this organism on Dieudonné’s blood agar, he obtained
cholera-like colonies, and the rods appeared like vibrios. On the other hand, several vibrios
have been found to be inclined to assume coli-like growth, as will be discussed later. Special
investigations upon the natural relationship which seems to exist in this direction, might lead
to very interesting results.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 57

Some pictures, shown by Kellerman and Scales (1916) as illustrating certain phases in the
life cycle of B. coli, are reproduced as figures 43-45 on PlateIV. Figure 43 makes an interesting
counterpart to the preceding pictures of B. pneumoniae. The clusters of large globules and the
hyphae-like threads in figure 45 represent forms whose physiological significance will be discussed
later. The small unstained thread containing four chromatine granules (in the lower left part
of fig. 44), also deserves our attention. Similar pictures of B. coli published by Hort (1917 a)
will find their place better in Chapter II (Pl. XIII, fig. 180), as they are especially suited for
explanatory purposes. The same holds true in comand to other data upon the behavior of the
typhoid and dysentery organisms, contained in the same paper, and also in regard to drawings
of Meirowsky (1914 6) illustrating similar phases in the life history of B. paratyphosus B.

Matzuschita (1900) published a photograph of the typhoid bacillus grown on salt agar which
is reproduced as figure 46 on Plate IV (from original fig. 16). It should be compared with
figure 36 on Plate III, showing the development of B. coli under analogous conditions. Other
examples of true branching are to be found in a photograph of B. typhosus published by Jtzerott
and Niemann (1895) in their ‘‘Atlas’’ (original fig. 18), reproduced as figure 47 on Plate IV.
The material used in this case, was taken from an agar culture. Gamaleia (1900) studied the
peculiar effect exerted by lithium salts. Large globules and branched forms became very
frequent. That these and other alterations of the cell form are not merely passive, osmotic
reactions, as was later assumed by Loeb (1902), has been already emphasized by Gamaleia
(l.c., p. 211). Casagrandi (1901) found cultivation on gypsum plates very useful for develop-
ing the branched growth of B. typhosus. Walker and Murray (1904) attained the same result
by adding 0.2 per cent alcoholic methyl violet solution to agar, gelatin or broth, while Péju
and Rajat (1906 a) and Rajat and Péju (1906) were only able to induce the typhoid bacillus, when
treated with potassium iodide, to produce long immotile threads showing oval, pear- or spindle-
shaped inflations, but no branching. In old cultures fragmentation of the long forms into short
rods and coccoid forms was frequently observed. Urea agar furnished Wilson (1907) analogous,
filamentous, round, and swollen forms, and also true branching. As early as in 1893 Almquist had
pointed out that B. typhosus not only multiplies by fission, hut also by producing lateral round
buds which grow up to new rods. Later (1904) he added, that also very thin needle-like rods
may directly come out from the old rods. These, as well as his other findings concerning
the reproduction of the bacteria will be considered more specifically in Chapters II and ITI.
Bernhardt’s (1915) studies ‘‘upon the variability of pathogenic bacteria’’ confirm the earlier
findings of thick immotile threads being formed by B. typhosus in old cultures, which still later
break up into normal motile short rods; in a few cases, however, motility was lost permanently.

That occasionally paratyphoid strains can assume all marks of the true typhoid organism,
and vice versa, has been shown by KG@hlisch (1918) and by Baerthlein (1918). The first named
author also reported (1916) upon some interesting yellow variants of the typhoid-paratyphoid
group.

Valuable results with’ dysentery bacilli were secured by Hata (1908) when he studied the
so-called ‘‘degeneration” forms of B. pestis, typhosus, dysenteriae, col, and cholerae, produced
on agar containing various amounts of CaCl,, MgCl,, or NaCl. Dysentery organisms were the
only ones which ‘‘degenerated”’ already on 1 per cent CaCl, agar, i. e., the thin cells became
larger and longer, and with increased salt content they assumed giant spindle forms, Some
strains furnished in addition large globules. True branching was frequent on NaCl agar.
Disappearance and reappearance of motility in dysentery strains have been discussed in a
paper by Dunn (1904). The experiments of Péjw and Rajat (1906 d and e) with potassium
iodide and urea gave results analogous to those mentioned before. Inflated threads and glo-
bules were produced, which later partially broke up into normal short rods. Data concerning
the correlations existing between different cell forms and appearance of the colonies in the
typhoid-dysentery group are to be found in Baerthlein’s (1912) report on ‘‘mutation” among
bacteria.

The acetic acid bacteria are demonstrating their close relationship to the pneumonia-colon
group, as in other directions, so also by their inclination to form the same type of large globules
58 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Von. XVI.

and inflated threads, which later reproduce the small ‘‘normal” forms. But it is very charac-
teristic that in this special case the large forms, though put aside in all other cases without
any adequate test as uninteresting ‘‘involution forms,” have been generally accepted in the
textbooks as some important peculiarity of the acetic acid bacteria, because in this case a well-
known investigator, Z. Chr. Hansen (1879), before the monomorphistic doctrine took possession
of most of the bacteriological textbooks, has pointed out that they belong to the normal life
cycle of these organisms. He also observed directly (1894) their formation and their
segmentation into small rods in single-cell cultures of B. aceti and of B. Pasteurianum.
Figure 15 on Plate E is a reproduction from his first paper (now fig. 77 on p. 458 in his
‘“‘Gesammelte theoretische Abhandlungen,”’ 1911), afar, on the other hand, was at least for
some time of the opinion that the large forms should be considered to be ‘‘ pathologic deformities,”
because he noticed that increased acidity stimulated their development. However, an acid
substrate is the natural habitat of these organisms, and their development is at its height under
such conditions. Zeidler (1896) described a motile acetic acid bacterium, also representing the
characteristic threads and inflations. Henneberg (1897-1898) created numerous ‘‘new species,”
which all showed the same character. He emphasizes the diagnostic importance of their
‘“‘hypertrophism”’ and pays special attention to those lateral buds which are produced by the
acetic acid bacteria in the same manner as by the lactobacilli. They appear on beer after two
days and at temperatures as low as 26-29° C. Kruse (1910, p. 1126) admits that the luxurious
growth on natural substrates is made up by the large forms, while the poor growth on artificial
media shows only small short rods; and yet he still calls the former ‘‘quaint involution forms,”
In a special study upon this group Janke (1916) introduces the new term ‘‘aberration’’ forms,
though he accepts them at the same time as part of the ‘‘normal’’ life cycle of the acetic acid
bacteria, which, therefore, if these authors were right, would be ‘‘normally aberrant” or ‘‘nor-
mally degenerate”’ while growing luxuriantly under natural conditions.

Some of the so-called photobacteria form another interesting branch of the B. pneumoniae
group. Their characteristic pleomorphism has been well illustrated in a little known paper by
Barnard (1899), as well as in the more often quoted publications of Molisch (1903-1904).
Reinelt (1905) has added other details; cocci and rods were present in all his cultures. That
sometimes branched forms appear, like those of B. radiciola, had been already mentioned by
Beijerinck in 1889.

The pleomorphism of B. radicicola, especially its inclination to grow in branched or in
radiate forms, was also well established from the beginning by the publications of Beijerinck
(1888-1890) and of Prazmowski (1890). Unfortunately, the incorrect term ‘‘bacteroids”’ has
been widely adopted for this type of growth, though there is not the least. doubt that Brunchorst
and Frank were wrong when they assumed that only cell products looking like bacteria, and
not real bacteria, were present in the root nodules of the leguminous plants. The radiate
growth was considered by Beijerinck as indicating relations to the Actinomycetes. Peklo (1910)
has dwelled upon this point in a paper on ‘‘plant actinomycoses.”’ Like Shibata (1902), he
accepts especially the organisms causing nodules in Alnus and Myrica roots as true actinomy-
cetes. Bottomley (1912) and Spratt (1912), on the other hand, isolated B. radicicola from
Alnus, Myrica, and Elaeagnus nodules and see in it the causative agent. This point needs further
investigations, which probably will best start again from the experiments carried on by Hiltner
(1898), who obtained from Alnus nodules thin rods, which grew in an actinomyces-like arrange-
ment. That B. radicicola exhibits all marks of the B. pneumoniae-coli group can be easily ascer-
tained by comparative studies (Léhnis, 1905 a) and that all members of this group show
branching under suitable conditions, is beyond dispute.1 In addition, it will become apparent
from the further discussion of these points in the following chapters that other characteristics,
also accepted by many as necessitating a separate systematic position of the nodule organism,
can be generally found with the bacteria as soon as these are made the objects of adequate

 

1 Data concerning the branched forms of B. radicicola may be found in the publications of Morck (1891), Dawson (1899), Hiltner and Stormer
(1908), Stichting (1904), Harrison and Barlow (1907), Rossi (1907), Buchanan (1909), and of Gage (1910). That they are erroneously classed as “ine
volution’”’ forms has been discussed on p. 25.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 59

studies. The relations to the vibrios, already indicated by some peculiarities of other mem-
bers of this group, as mentioned above, are equally discernible with B. radicicola and its next
kinsman, B. radiobacter. It is well beyond doubt that the organism described by Severin (4897)
as Vibrio denitrificans is nothing else than a denitrifying strain of Radiobacter (see Léhnis,
1910, p. 486). The characteristic triangular forms found with it, like with all vibrios and
spirilla, have been well drawn by Bordoni- Uffreduzzi (1888) in his sketch of ‘‘Proteus hominis
capsulatus” (reproduced as fig. 14 on Plate D). A good survey of the different types of growth
of B. radicicola has been given by Conn (1909) in the drawing reproduced as figure 16 on Plate E
(from original fig. 26, p. 99). That the flagellation has been found to be either mono- or peritri-
chous agrees well with the analogous behavior of other members of this group, as mentioned
above.

Bact. rubiacearum, the organism causing nodules in the leaves of tropical Rubiaceae,
was found by Faber (1912) to resemble B. radicicola very closely. Again the branched forms
are prevalent in young nodules, and their ability to produce small coccoid bodies, which repro-
duce the ‘‘normal”’ rods, is the same as in the case of the legume organism. Results obtained
by Georgevitch (1916) confirm these findings. The causative agent of crown gall, Bact. tume-
faciens, gives a similar branched growth (EF. F’. Smith, 1911, Vol. II, p. 73).

Among the pigment forming rods some producing a yellow color show distinct relations
to the B. coli group. Fr. Levy (1904) isolated two Coli varieties from flour, of which the first
one (B. coli albido-liquefactens) deviated from the type by its inclination to liquefy gelatin,
while the other one (B. colt luteo-liquefaciens), in addition to this, produced a yellow pigment.
While, however, both still possessed the typical fermentative abilities of the Coli group, another
strain, isolated from the same source, resembled closely the second one, morphologically as
well as physiologically, except that the production of gas was absent. An interesting counter-
part to this organism is presented by an old stock culture of Schroeder's Bact. synzanthum,
studied by Lehmann and Neumann (1912, p. 392, footnote). Originally a motile yellow short
thin rod, producing alkali and staining milk yellow, it was found to be an immotile slime
producing organism, Gram-positive in 1899, Gram-negative in 1903, growing like B. coli on
gelatin, agar, and in milk, which was coagulated; gas was produced from glucose, and only
on potato a bright yellow pigment still appeared.

Bact. Fraenkelii, isolated by Hashimoto (1899) from milk, is another yellow rod which has
been frequently cited as a prominent example of pleomorphism. Growing on agar as small
rod with 2-3 polar flagella, it forms in broth long chains of thick immotile globules, looking
just like large streptococci. These divide transversely, as well as longitudinally, which leads
to ‘‘pseudo-ramification.”” Sarcina-like formations were also seen.

Distinctly Actinomyces-like branching seems to have been observed with yellow rods
only twice, i.e., with B. erythrogenes, according to Migula (1900, Vol. II, p. 458), and with Bact.
solare, according to a report made by E. Wolff (1898). That this occurrence is no reason (con-
trary to the opinion of the latter author) to remove this organism to the Actinomycetes has
been discussed (p. 23). A “‘Streptothrix polychromogéne,”’ described by Valée (1903), exhib-
iting the same rather unique double-pigment production as the first-named species (soluble
red, insoluble yellow), perhaps is to be united with B. erythrogenes.

Bact. prodigiosum, originally mistaken by Cohn for an immotile micrococcus, has been
frequently studied with regard to the variability of its pigmentation; its morphology, on the
other hand, has been much neglected, though as early as in 1888 Wasserzug (1888 a) has demon-
strated that it, too, can exhibit great differences on various substrates. Besides rods and threads,
typical staphylococci, as well as spiral forms, were produced. Large globules (1-2 u) and yeast-
like cells have also been seen. Slater (1891) observed with another Prodigiosus variety, which
he named B. corallinum, that the long threads present in liquid substrates, exhibited a marked
tendency to form bud-like projections, which in some instances became so prolonged as to
resemble branching. His drawings remind one at once of the branched forms of B. radicicola
and related organisms. WN. G. Davis (1901) in the course of studies upon the pigment produc-
tion of B. rosaceus metalloides Dowdeswell, got, after hundreds of replatings, typical cocci
60 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

(of 0.54 diameter) which further remained constant. Ph. Eisenberg (1914) says that the
morphological alterations exhibited by B. prodigiosum are “not conspicuous, usually incon-
sistent, and not uniform.’’ It seems, however, as if here, like in many other cases, very much
attention was spent upon the appearance and disappearance of the red pigment in the cultures
and only very little interest was left for microscopical studies.

Bact. violaceum is another instance where F’. Cohn used a rod for establishing his ‘‘genus’”’
Micrococcus (see p. 11). That temporarily globular forms occur also with this organism
was probably first mentioned by Zopf (1883, p. 68). A related, immotile, slime-producing
form, called Bact. visco-fucatum, was isolated from butter by Harrison and Barlow (1905),
who found that ‘elongated branched and cuneate forms were common in old cultures, kept
at temperatures below 25° C., and in young cultures at temperatures above 27° C., and in
certain special media. These cells were 4~-9u long and up to 0.9% wide. The cuneate torms
were 0.94 at the larger and 0.3» at the smaller end. The branched forms were frequently
complicated and sometimes tangled together, reminding one of a clumping formation. After
the branched forms had become numerous, they broke up by transverse segmentation into
short oval elements.”

The pleomorphism of the colonies of Bact. syncyaneum and of other lophotrichous organ-
isms has been fully studied by Nyberg (1912). Thorough experimenting showed that it was
possible to induce every one of 140 strains tested to grow either in smooth and round or in
wrinkled and branched colonies. Other biological, as well as morphological, differences were
correlated with this dimorphism, but they were not made the object of special investigations.

With B. pyocyaneus it has been shown by Guignard and Charrin (1887), a long time ago,
that conspicuous alterations in the morphological appearance, as well as in the physiological
behavior, can be evoked by slight changes in the substrate (viz., by adding small amounts of
antiseptics, like boric acid, phenol, creosote, and thymol). The rods passed over into threads,
these into spirilla, which later formed round resting forms, which in their turn were able to
reproduce the typical rods. The spiral forms did not produce pyocyanine, but the rods germi-
nating from their resting forms, proved to be normal also in this respect. It is, on the other
hand, of course, not very difficult to suppress or alternate the pigment formation more or less
permanently also with this organism. Charrin and Roger (1887) made such experiments with
Pyocyaneus, as well as with B. fluorescens. As did small quantities of antiseptics, exclusion
of air also caused a change from the brilliant bluish green to a light yellowish color. A con-
stantly colorless variety of B. pyocyaneus was obtained by Charrin and Phisalix (1892) by
cultivating at 42.5° C. After several transfers the alteration was stable also at lower tempera-
tures, but one passage through the animal reestablished the typical character. A special pub-
lication upon pleomorphism and pleobiosis of B. pyocyaneus has been contributed by Schiir-
mayer (1895). In broth both thick short rods and very thin long threads were observed.
Gamaleia (1900) noticed that branched forms became frequent with B. pyocyaneus, as with
other species, by adding lithium compounds to the substrates. Wassermann (1903) classes
B. pyocyaneus among the “ pleomorphous schizomycetes,”’ and Péju and Rajat (1906 f) as well
as Wilson (1907) found that potassium iodide and urea favored the appearance of the different
stages of growth also in this case.

Some data upon the different cell forms produced by B. fluorescens in the various phases
of its life cycle are to be found in W. Winkler’s (1899) and in Fuhrmann’s (1906-1908) papers.
That allfour principal cell forms, viz.,small and large cocci, as well as small and large rods may
sometimes be encountered in the same microscopical field, was shown in one of the photographs
of our second preliminary communication (Léhnis and Smith, 1916 6), reproduced as figure
48 on Plate IV. Curved forms, as well as star-like growth, are equally frequent with this
species. The denitrifying Fluorescentes, described by Christensen (1903) as ‘‘new species,”’
are evidently strains of Bact. putidum, exhibiting the same pleomorphism: small cocci and rods
(0.5-1.0 X 0.5-1.5u) and large cocci, ovals and rods (1.25 X1.5-3y, occasionally up to 4—-5y)

Pseudomonas cerevisiae, whose life cycle was partially investigated by Fuhrmann (1906),
probably also belongs in the neighborhood of B. fluorescens. The same holds true for the
No. 2. LIFE CYCLES OF THE BACTERIA—LOBNIS. 61

so-called Bact. bruneum, i. e. a brown variety of B. fluorescens, whose sarcina-like regenerative
bodies were photographed by Matzuschita (1900), and reproduced as figure 18 on Plate II.
An organism first called by Rullmann (1897) Nitrosobacterium, because of the erroneous
assumption that it might participate in the process of nitrification, and later (1898 6, 1900)
changed by him into a B. ferrugineus, is another brown variety of B. fluorescens, producing in
mineral solutions thin branched threads, often showing terminal inflations; but on agar at
37° C. big “involution” forms were seen to grow up to 1.2 X 7x.

That the genuine nitrite and nitrate-producing organisms, Nitrosomonas and Nitrobacter,
are also distinctly pleomorphous has been already pointed out by Winogradsky (1892), whose
attention, however, was especially attracted by the differences distinguishing the “zoogloea’”’
and the motile stages of Nitrosomonas. ° As the discussion of the true nature of this “zoogloea”’
will be taken up in Chapter III, the “‘monas”’ stage only will be considered here.

Winogradsky was of the opinion that the Russian Nitrosomonas (from Kasan soil) should
be accepted, on account of its smaller size, as a species different from the larger Nitrosomonas
from Ziirich, and also different from the Nitrosomonas from Java, which developed large and
small cells at the same time. A comparative study of Winogradsky’s photographs, however,
shows. clearly that the Nitrosomonas of Ziirich (original fig. 1, reproduced as fig. 49 on Pl. V)
has by no means a uniformly large size; small cells are budding forth from and are interspersed
between the larger bodies. The Nitrosomonas from Kasan (original fig. 4, reproduced as
fig. 50 on Pl. V), on the other hand, displays mostly small forms, but large cells are also
present, and their breaking up into small ones by dividing and budding is equally clearly rep-
resented. The Nitrosomonas of Java (original fig. 12, reproduced as fig. 51 on Pl. V) evi-
dently takes its place exactly in the middle between both other forms. as photographed; here
an even mixture of large and small forms is to be seen. Winogradsky himself declares the larger
cells in this case to be “small colonies of the small forms,” a statement which undoubtedly is
equally applicable to the Zirich organism. That the morphology of Nitrobacter follows
similar lines is to be seen from a comparison of figs. 52 and 53 on Plate V (original figs. 1 on Pl.
XVIII, 1891, and 16, 1892). Whether, however, the so-called Nitrobacter polytrophum actually
represents, as Beijerinck (1914 6b) asserts, a polytrophous modification of the Nitrobacter oligo-
trophum, must be left in doubt. The fact that this new “modification’’ differed rather radically,
morphologically as well as physiologically, from the true Nitrobacter, and that it could not be
reverted within 10 years into the original nitrifying form, makes it very probable that some
contamination has played its réle in this as in many other cases, among which the creation of
the “Salpeterpilz”’ by Stutzer and Harileb (1897) probably is the best and at the same time the
most unfavorably known example. These authors had later (1899 6, 1901 a) results which
agreed fairly well with those of Winogradsky; they proposed, however, a new name—“ Nitro-
microbium’’ for Nitrobacter—because they noticed the occurrence of budding (visible also
in Winogradsky’ s pictures) and thought,erroneously that bacteria never show this kind of prop-
agation. Their “Hyphomicrobium” obtained at the same time in their nitrification studies
(1899 a), which grows either as short motile red, or as thin branched thread, producing round
regenerative bodies, and also thought by them to be no bacterium, may have been identical
with Bevjerinck’s B. oligocarbophilus, which has been recently removed by its author (1914) into
a new genus Actinobacillus, which is closely related to Actinomyces. The colonies are very
much alike, especially with regard to the formation of aerial hyphae. Actinobacillus is, accord-
ing to Beijerinck (1914 6), devoid of branching, which statement, however, may be justly doubted.

The Actinobacillus found by Lignéres and Spitz (1902) to be the causative agent of the
peculiar “Actinobacillose” of South American cattle, also exhibited only coccoid and rodlike
cells in the cultures, but the clusters occurring in the pus are made up of richly branched forms,
looking like Metchnikoff’s (1888 6) Pasteuria ramosa. Th. Smith (1918), on the other hand,
obtained a pleomorphic bacillus from pneumonic lungs of calves which grew in the tissue only
as a fine rod, while in serum cultures flakes with large clubs and also coccoid forms were pro-
duced. The author rejects the term Actinobacillus, which he desires to replace by the name
Bacillus actinordes.
62 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI

(c) SPORE-FORMING RODS.

Among the early bacteriologists the opinion was rather generally adopted that the spore-
forming bacilli do not permanently appear as large rods, but are also able to assume temporarily
coccoid shapes. Perty (1852), e. g., pictured his Metallacter bacillus (on his Pl. XV, fig. 26 a)
in the manner reproduced in figure 17 on Plate E. Négeli (1877) illustrated his statement,
that F. Cohn was “fundamentally wrong” in assuming that the bacilli be always long rodlike
cells, by a similar drawing, reproduced as figure 18 on Plate E (from original fig. 2). Buchner
(1882, p. 221) gave another interesting picture, reproduced as figure 19 on Plate EH, showing
once more that coccoid, as well as rodlike, forms can both be produced by the same organism.

All recent studies have confirmed and amplified these early findings. Many details per-
taining to different aerobic species, which were made the object of comparative investigations,
may be found in publications by Gottheil (1901), Neide (1904), Gruner and Fraser (1912), Ford
(1916), and others. The large, globular, granular cell forms, described by these authors, are
quite similar to those considered to be characteristic of Azotobacter. Their appearance was
not stimulated by potassium iodide and urea—i. e., those two substances which Péju and Rajat
(1906) found to be very useful in bringing about analogous changes with several nonspore-
forming rods. Branching was observed by Gottheil (1901, p. 717), especially with B. cohaerens.
It takes place, according to A. Meyer (1901), earlier than the spore formation, or under condi-
tions which favor the formation of long threads and hinder the production of endospores.
That neither the large coccoid nor the branched forms should be classed as “involution forms,”
has been discussed (on p. 24-27). That motility and flagellation here as in other cases can not
be considered to be of fundamental importance for systematic purposes, may be seen, e. g.,
from a paper by Ellis (1906) upon ‘The Life History of Bacillus hirsutus.”” Henrici, who estab-
lished this’ “‘new species,’ found it immotile, while it showed to Ellis at first polar cilia like a
““Pseudomonas,”’ but later the peritrichous flagellation of a “regular” bacillus became visible.
An old stock culture of B. implexus, described by Zimmermann as immotile, assumed motility
and so became, according to Zverler (1899), a typical B. subtilis. B. mycoides, on the other
hand, when studied by Lehmann (1899), refused for a long time to show any motility and most
of the rods remained, despite prolonged experimenting, permanently immotile.

Very interesting results with B. mycoides have been recorded by Olsen (1897), as well as
by Nadson and Adamovié (1912). The first-named author noticed that aerobic spore-forming
bacilli are much inclined to produce yeast-like forms. When cultivated on wet sand and gravel,
B. mycoides gave a white dry growth on the surface, which in carefully made preparations
showed itself to be made up of frequently branched Actinomyces-like cells, forming short
plump ‘‘pseudoconidia,”’ which very easily broke off. In the liquid, however, only “normal”
rods and threads became visible. Figure 20 on Plate F is a reproduction of figure 14 on
Plate V of Olsen’s publication, illustrating these findings. Nadson and Adamovié (1912)
obtained a typical Actinomyces-like growth of B. mycoides by adding to broth and agar equal
amounts of an old liquefied gelatin culture of B. mycoides, which was previously sterilized by
being kept 15 minutes at 120° C. Furthér transfers remained constant; they were unable .
to produce spores and to dissolve the gelatin.

Some of the forms in Olsen’s sketch resemble very closely certain types of growth of diph-
theroid and tubercle bacilli. It is worth mentioning in this connection that. Niessen found a
Gram-positive diphtheroid ‘‘streptobacillus” in syphilitic lesions which is slowly motile, pro-
duces typical endospores, shows branching and budding, and appears in small, as well as in
large, globular and rod-like forms. The “Cladothriz stereotropa,”’ isolated by Proca and Danila
(1910) from the same source is either identical or closely related to Niessen’s “‘streptobacillus. ”
Here again the diphtheroid character is very distinct, yet there are motile rods, too, which
form endospores, and also globular immotile, budding cells. A pleomorphic bacillus, obtained
by H. U. Williams (1912) from a case of tuberculous pericarditis, is another diphtheroid, al-
though spore-forming, evidently belonging to this group. And the same may hold true with
regard to Mycobacillus synovialis, isolated by Ohantemesse, Matruchot and Grimberg (1917),
from a case of arthritis. It is a spore-forming bacillus, producing a salmon color in milk and
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 63

on potato, much inclined to produce in the cultures branched threads, and therefore accepted
by its discoverers as a connecting link between the sporulating bacilli and the actinomycetes,

That B. anthracis is not always so constant and monomorphous, as was proclaimed by

R. Koch, has been shown first by Buchner (1882 a-c), who succeeded in changing its morpholog-
ical, as well as physiological, behavior to such an extent that it became very similar to B.
subtilis, By adding relatively large amounts of sugar, both organisms could be induced to
produce distinctly isodiametric cells. HH. Klein (1883) got similar round forms of 1.3-2.6p
diameter in chain-like arrangement, or sometimes still connected with typical rods and threads,
when B. anthracis was cultivated in pork broth or pork-broth gelatin, kept at 20-25° C. They
either multiplied by dividing into two or four, or by true budding. Their virulence against
guinea pigs and rabbits was the same as that of the rods. The latter always reappeared in the
animal and could also be obtained from the round cells in artificial culture. The picture given
by E. Klein (as figs. 1-3 on Pl. XXI) and reproduced as fig. 21 on Plate F, makes it prac-
tically certain that, if not all, at least most of these round cells have been microcysts or
regenerative bodies. Another drawing, published a few years later (1885) in Klein’s textbook
on pathogenic micro-érganisms and reproduced as figure 22 on Plate F (from original fig. 77,
p. 109), also deserves our attention. Here again the character of reproductive organs is fairly
well indicated. Klein himself places these cells parallel to the “sporangia of nostoc algae.”
Others of the round cells, however, seem to be of a truly vegetative nature. In a later paper
E. Klein (1894) reports that he noticed, when studying very early phases of growth of
B. anthracis “when the colonies are only just visible as angular greyish spots,” that the
sprouting filaments at the edge were made up of spindle-shaped and oval elements with
much vacuolization. His pictures (original figs. 1 and 2 on Pl. I) are reproduced as figure 23
on Plate F. After 4 days only typical straight rods and threads were found.

De Bary (1884, p. 504) saw also anthrax bacilli change into large cocci, but he was
unable to secure satisfactory proof of further development. Analogous results were obtained
by Braem (1889) by keeping anthrax bacilli in distilled water. Good pictures of the different
appearance of this species in artificial cultures and under natural conditions were given by
Itzerott and Niemann (1895, original figs. 13-15). They are reproduced on Plate V as figure
54 (from gelatin), figure 55 (from blood, mouse) and figure 56 (from spleen, mouse). All pho-
tographs were made with the same magnification (x 1,000), and the smears were all treated
in the same manner (stained with methylene blue). Crookshank (1896, p. 195) reported that
especially from gelatin and glycerin agar “‘very striking preparations are sometimes obtained
with numerous large spherical and lemon-shaped elements.’’ The same facts were described
by Rodet (1894, p. 18 and 25), as follows:

A cété de filaments réguliers . . . d’autres présentent des renflements, des dilations et, intercalés aux articles
cylindriques, des articlescourts et plusoumoinsarrondies . . . Articles . . . véritablementsphériques . .

sont alors entremélés 4 des éléments moins modifiés et c’est toute une gamme entre les filaments les plus reconnaissa-
bles et ces formes singuliéres qui donnent l’impression de la présence accidentelle d’un microbe étranger.

The round forms which Matzuschita (1900) found on salt agar, together with normal spores,
are reproduced on Plate V as figure 57 (from original fig. 31). As far as can be judged from
sight, they are typical regenerative bodies. Some of them exhibit the budding already observed
by EZ. Klein. Ligniéres and Durriew (1902) noticed that a very virulent strain of B. anthracis
changed on agar within 4 days into long filaments producing round buds or assuming irregular
spiral shape, while on potato a yellowish growth appeared, made up of pear- and club-shaped
cells. Some other interesting proof of the pleomorphism of B. anthracis was secured by Mme.
Henri (1914) by treating a watery suspension of the bacilli with ultraviolet rays. Most of the
cells were either killed or remained normal; but some of them gave colonies of such forms as are
reproduced on Plate V in figure 58 (from original fig. 4), which are very similar to those seen
by E. Klein and Matzuschita. That they were regenerative bodies was proved in this case by
their reverting back to normal rods. Sometimes, however, Henri’s experiments led to a
stabilizing of the spherical growth. The typically coccoid cells reproduced in figures 59 and 60

160266°—21——-5
64 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

on Plate V (from original figs. 5 and 6) did not go back to the rod form. Moreover, an actino-
myces-like growth developed, made up of branched, Gram-negative, nonsporulating filaments,
shown in figures 61 and 62 on Plate VI (from original figs. 7 and 8), which resemble very much
the atypical growth of B. mycoides, obtained by Nadson and Adamovié. These filaments did
also not revert to the rod form directly, but only after having formed again the coccoid regen-
erative bodies, reproduced in figure 63 on Plate VI (from original fig. 9), which process agrees
closely with the behavior exhibited by B. azotobacter and other organisms tested in our own
experiments (1916), when passing over from one subcycle into another.

With B. subtilis similar round forms were studied by Buchner (1882) and by E. Klein
(1885, p. 110). The photographs of B. subtilis reproduced as figures 64 and 65 on Plate VI
from our preliminary papers (Léhnis and Smith, 1916 a, fig. 26; 1916 5, fig. 6) make inter-
esting counterparts to those of B. anthracis, as well as to the well-known pictures of ‘‘bacte-
roids” of B. radicicola. Distinctly filamentous growth, resembling Actinomyces, has also
been obtained.

An old stock culture of the so-called Tyrothrix tenuis, originally isolated by Duclauz, gave
us (1916 a) very similar results as did B. subtilis. W. Winkler (1895), who tested several of
Duclauz’s “Tyrothrix” cultures, speaks also of broad ‘“involution’’ forms, which probably
have been of the same nature, as far as can be judged from his paper and from his drawings.
Analogous findings were recorded by Viehoever (1913, p. 230) in regard to B. Pasteurii. The
globular forms produced by segmentation of the rods, which he calls “micro-oidia,” probably
are to be classed as regenerative bodies like those obtained by Henri with B. anthracis. That
the yeast-like ‘‘involution’’ forms, considered by Miquel and Cambier (1902, p. 633) as charac-
teristic of Urobacillus Maddoxti, are actually not of such a distinctive value as these authors
thought, needs no special discussion.

A bacillus isolated by Pansini (1890) from sputum and described as “No. 6” is another
example of pronounced pleomorphism within the Subtilis-Mesentericus group. Small cocci,
large yeast-like cells, and rods of different size have been recorded. With other bacilli from
the same source “‘coccoid degeneration forms’’ were also registered by this author.

A bacillus found by Maassen (1904) in “ Hackfleisch”’ gave branching under the influence
of lithium salts added to the substrate. Figure 66 on Plate VI is a reproduction of the original
figure 11 on Plate XIV. Practically the same development was shown by another not com-
pletely studied ‘yellow bacillus No. 41”’ from soil, used in our preliminary experiments on the
life cycles of the bacteria (1916 a), in this case, however, without being stimulated by lithium
or any other specific stimulant.

Gamateia (1911, p. 211) found lithium salts equally useful for producing branched growth
of B. Megaterium. The normal rods of this species, originally described by De Bary (1884, p.
500) as being 2.5u broad, showed only a width of 0.6-0.8u when reinvestigated by Lehmann and
Neumann (1912, p. 461). It is to be expected that here again small and large cell forms may
become visible, according to the subcycles entered by the organism in the course of its develop-
ment. Long chains of big spherical cells have been also recorded by De Bary (1884, p. 503)
for B. Megaterium, and it has been mentioned (p. 41) that the presence of a small bacterium
was found to be distinctly favorable for inducing this transformation, just as was the case
when B. radiobacter was growing together with B. oxalaticus or with a stable rod-like strain
of B. Azotobacter. Bac. danicus, found by Léhnis and Westermann (1908) to be a common
symbiont accompanying B. radicicola in the root-nodules of leguminous plants, seems to be in-
fluenced by the latter in an analogous manner. As with B. Megaterium and Azotobacter the
reduction of size in pure cultures has been also noted in the case of B. oxalaticus. Migula (1900,
Vol. II, p. 538) found that at first the rods measured 23-3 x 4-8y; later, however, not more
than 1.21.84.

B. malabarensis, isolated by Léhnis and Pillai (1907) from East Indian soil, is of great
morphological interest, not only because it is even more inclined than the species mentioned to
assume distinctly Azotobacter-like forms, but also because of its liability to produce peculiarly
looking beet-shaped, spermatozoon-like, and later long-tailed spindle-shaped cells, as are shown
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 65

in figures 67 on Plate VI and 266 on Plate XXI. These forms present a remarkable counter-
part to the appearance of Rhabdochromatium, just as the large Azotobacter-like cells resemble
those of Chromatium. Probably the natural relationship of these little known sulphur bacteria
will be better understood, after the life history of the large spore-forming bacilli has been more
completely studied. Migula’s remark (1904, p. 31) that no such cell forms are to be found
with typical bacteria can not be accepted as correct. And that the peculiar, swollen, and tailed
cells can not be discarded as “involution” forms, is proven by the fact that they, too, precede
the formation of spores. Duclawx (1883, p. 654) described this a long time ago, referring to
the analogous behavior of his Tyrothriz virgula, with the following words:

Ce renflement grossit pendant que la portion du bAtonnet qu’il n’a pas atteinte se résorbe et s’amincit de plus en
plus, de sorte que 1’étre tout entier prend des formes trés singuliéres . . . Quelques-uns d’entre eux rappellent
d’une maniére frappante des spermatozoides. C’est dans les renflements que se fait ensuite le travail de formation
des spores.

Bac. Baccarinit, isolated by Macchiatti (1898) as a causative agent of the ‘mal nero” of
the grapevine, is another example of distinct pleomorphism among the spore-forming bacilli.
From typical cocci up to long threads practically all forms were seen.

Bac. tumescens shows the same character, as was first pictured by Zopf (1883, p. 66)
in the drawing reproduced as figure 24 on Plate G (from original fig. 23), and later
studied anew by A. Koch (1888) and by Garbowski (1907). The last-named author made at
the same time a fairly complete study of Bac. luteus sporogenes Wood et Baker, which led to
many confirmative results. Good branching has been recorded in this case, as well as the for-
mation of those small coccus forms (of only 0.8—1y diameter), which simulate very much strepto-
cocci and sarcinae, and which have been described as “micro-oidia” by Bredemann (1909) for
B. Amylobacter, and by Viehoever (1913) for B. Pasteurii.

That all the different species or varieties of the Azotobacter group are probably types of
growth of spore-forming bacilli, is indicated by the results of our more recent investigations (Léhnis
and Hanzawa, 1914, L. and Smith, 1916). Several of the photographs accompanying our 1914
paper present interesting counterparts to those published by Henri at the same time for B.
anthracis. Especially the small round Azotobacter cells (regenerative bodies) reproduced as
figure 68 on Plate VI (from original fig. 28) look very much like Henri’s Anthrax cells in figures
59 and 60 on Plate V. Figure.69 on Plate VI (reproduced from Léhnis and Smith 1916 a, fig. 20)
demonstrates, in addition, that it is possible to induce a strain, which originally showed the
typical character of Azotob. chroococcum, to grow in a manner very much resembling that of
the normal B. anthracis. Fig. 70 (reproduced from Léhnis and Smith 1916 a, fig. 16), on the
other hand, represents an interesting intermediate stage in the development of small rods, in
which case B. azotobacter assumes forms very similar to B. radiobacter and other organisms
belonging to the B. pneumoniae group. A comparison of fig. 70 with figs. 38 and 39 on Plate
IV, illustrating the analogous stage in the life cycle of B. pneumoniae, will further prove this
fact. (The clusters visible in fig. 70 are the outgrowths from the symplasm, to which reference
will have to be made in Chapter III.) Figures 71 and 72, taken from a paper by Walton (1915,
original figs. 1, 9 and 10) show the small short rod-like forms from a one day old culture, and
the typical large Azotobacter cells from the same culture after 7 days growth. Most of the
strains studied by this author underwent an analogous change regularly, though in other cases
the small forms have proved themselves to be much more persistent. The pleomorphism of
B. azotobacter was already noticed to some extent by Betjerinck (1901 6). More data were fur-
nished by Diiggeli (1905 a), H. Fischer (1905), Léhnis and Westermann (1908) and by Krzemi-
eniewski (1908). Branched forms were first seen by H. Fischer and by Léhnis and Westermann;
but they were considered by the first-named author to be “involution” forms. Peklo (1910)
supposed that as B. radicicola and spore-forming bacilli, so also Azotobacter be related to the
Actinomycetes. Of special importance are Prazmowski’s (1912) extended investigations upon
morphology, cytology, and physiology of the Azotobacter group, which were further confirmed
and amplified by the studies made by Léhnis with Hanzawa (1914) and with Smith (1916).
66 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Our first preliminary report upon the life cycles of the bacteria (1916 a) contains the descrip-
tion and pictures of most of the various types of growth of B. azotobacter. A schematic sketch
of the complete life cycle of this organism was reproduced from the last-named paper as figure
1 on Plate A; and on page 36 two tables were reprinted, showing that essentially the same
complex morphology was not only observed with all Azotobacter strains tested, but that it
eould also be traced in all other groups of the bacteria.

Harrison (1900) has described variations in size and shape, as well as the appearance of
Y forms, with B. alvei. Like other species connecting aerobic and anaerobic spore-forming
bacilli, it seems to be much inclined to lose its ability to form spores, as was noticed by Lehmann
and Neumann (1912, p. 505).

Pleomorphism and pleobiosis of the more or less strictly anaerobic bacilli have been firmly
established, especially by the investigations of Grassberger and Schattenfroh (1900-1907) and
by Bredemann (1909). It remains to be seen, however, whether or not the study of the com-
plete life cycles of these organisms will support the standpoint taken by the latter author,
who is of the opinion that a considerable number of anaerobic species, mentioned in the lit-
erature, should be cancelled and united with B. amylobacter. That as with aerobic, so also with
anaerobic spore-forming bacilli the morphological changes can not be stimulated in the same
manner as with various nonspore-forming bacteria, i. e., by adding potassium iodide and urea,
was reported by Péju and Rajat (1906 e). Change in motility and cell form of B. amylobacter
has already been seen by Van ‘Tieghem (1884, p. 110). Besides rod-like forms, “cellules
sphériques et ovoides”’ and others “‘enroulées en helice’’ were mentioned, all of which were
found to be able to produce spores. Analogous results have been recorded by Prazmowsh
(1880) for his Clostridium butyricum and Polymyxa. His drawings, reproduced as fig. 25 on
Plate G (from original figs. 1, 2, 6, and 7 on Pl. II), illustrate that he saw changes between
“Clostridium”’ and ‘‘Plectridium”’ type, formation of curved, coccoid, and large inflated cells
and in the accompanying text it was especially pointed out that the last-named form became
visible when the development of luxuriantly growing cultures was at its height, so that they
should not be classed as “‘involution’’ forms. They were seen to return to the normal rod
form by segmentation. Spherical cells were observed by Pragmowski only with his Clostridium
Polymyxa, but it was, indeed, quite probable, as Zopf (1883, p. 70) presumed, that they would
also be discovered with Clostridium butyricum. As mentioned above, Van Tieghem has seen
them, and in figure 26 on Plate G drawings from Bredemann’s paper on B. amylobacter (1909)
have been reproduced, showing the forms called by him “‘micro-oidia.”” They were found
with all of his strains and under quite suitable conditions, but he was unable to revert them
into the typical rod-like anaerobic growth. On aerobic plates they gave small soft punctiform
colonies.

The results obtained by Winogradsky (1902) with his Clostridium Pasteurianum are very
similar, and Bredemann includes this ‘‘species’’” among the varieties of B. amylobacter. As
figure 73 on Plate VII one of Winogradsky’s photographs (original fig. 3) has been reproduced,
showing large and small coccoid forms, curved cells and apparently also some branching and
budding. The morphological, as well as physiological, variability was found to be so great
with this species that the different strains were hardly recognizable as being of the same kind.

The behavior of and the relationship existing between the various anaerobic butyric acid,
pathogenic, and putrefying bacilli have been fairly well elucidated by the investigations of
Schattenfroh and Grassberger (1900), of Grassberger (1902-1905), and of Grassberger and Schatten-
froh (1907). Motility and cell form, but also appearance of colonies and physiological charac-
ter, exhibit very wide variabilities. Some strains are comparatively stable, while others are
easily ‘“‘denatured,”’ which expression is used by the authors for those cultures which have
lost, temporarily or permanently, their motility and their faculty to form spores. As among
the latter sometimes the production of gas also ceases entirely, and the butyric acid is largely
replaced by lactic acid, the general character of such strains becomes very similar to that of
certain lactobacilli. As I have pointed out (1910, p. 200) this transformation deserves to be
investigated more thoroughly. Lava (1911) found that treatment with formaldehyde was very
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 67

efficient in bringing about such alterations, not only with butyric acid, but also with putre-
fying strains. Presence of sugar favors the production of swollen forms and stimulates the
fixation of nitrogen in the same manner as it does with aerobic spore-forming bacilli, while
substrates rich in albuminous substances increase the tendency to pass over into the putrefying
plectridium type (B. putrificus).

B. putrificus, B. oedematis maligni, as well as B. Chauvoei, may all be, according to Grass-
berger and Schattenfroh (1907) and Bredemann (1909), merely adaptations of B. amylobacter.
The results thus far collected make a thorough study of the complete life cycles of these organ-
isms very desirable and promising.

Coccoid, lemon-, and club-shaped forms of the organism of symptomatic anthrax were
observed in cultures by Ehlers as early as in 1884. His work was discredited by Kitt (1887),
but later confirmed by Grassberger and Schattenfroh (1907). Fraenkel and Pfeiffer (1895), as
well as Piant and Galli-Valerio (1895), dwelled upon the pleomorphism shown by this species
in the tissue, where small and large rods, dumb-bell shaped and coccoid cells werefound. As fig-
ure 27 on Plate H drawings of Ghon and Sachs (1903, original figs. 13, 14, 16-18) are reproduced
to illustrate this pot. Except the cells shown at the left side, which were taken from the
muscle of a pigeon, all forms were grown on the agar plate. The small branched cells resemble
certain lacobacilli very much. Two of Grassberger’s (1903) photographs, reproduced on Plate
VII as figures 74 and 75, may complete the picture. Figure 74 (from original fig. 54) gives
an analogous mixture of small and large rods, as was shown in figure 69 for one of our spore-
forming Azotobacter strains. And figure 75 (from original fig. 21, material taken from sugar
broth) could be easily mistaken for a photograph of a typical Azotobacter culture. This par-
allelism is enhanced by the variability in motility, the instability of spore formation, by the
shape of the colonies, and especially by the possibility, proved by Grassberger (1905), to adapt
B. chauvoei to aerobic life with loss of pathogenicity. Several objections have been raised by
Hibler (1908) against the work done and the standpoint taken by Grassberger and Schattenfroh.
His own work, however, seems to be much more open to criticism. On Plate VII, as figures
76 and 77, two photographs of symptomatic anthrax are reproduced from his book, where
they appear widely separated (on Pl. X, fig. 16, and Pl. XVI, fig. 7). Both show the charac-
teristic big round cells accompanying the rods. But only in the first case Hibler is inclined
to admit this pleomorphism, because here the material was taken from a pure culture. In
the second case, representing the analogous development in the peritoneum of a mouse, he
does not hesitate to state: ‘‘Die verschieden grossen dunklen kugligen Korper sind Granula
aus aufgelésten Mastzellen, der unterste Phagozyt schliesst auch zwei solche Korner ein,’’
though, of course, this statement remains entirely without proof, and the obvious likeness of
the two photographs (despite the different, not exactly defined magnification) creates consid-
erable doubt whether Hibdler’s standpoint was not too strongly influenced by the monomorphistic
doctrine with its well-known simple (but unfounded) ‘“ explanations.”

A good example of true branching of the oedema bacillus has been given by Grassberger
(1903) in the photograph reproduced as figure 78 on Plate VII (from original fig. 42). The
material used was taken from sugar agar. Ghon and Mucha (1905) isolated a closely related
organism from a case of peritonitis, which in the peritoneal exudate first looked very much
like B. coli. In sugar agar it proved to be highly pleomorphous. Besides the more normal
cells, club-shaped, curved vibrio-like rods became visible, ‘und in grosser Menge Gebilde in
den verschiedensten Gréssen und Formen, die vielfach kaum noch als Bakterienformen ange-
sprochen werden konnten.”’ Branching was frequent in hydrocele fluid.

With B. tetani, Kanthack (1896) and Kanthack and Connell (1897) have seen branchings,
especially in young cultures (18-24 hours old), while in old cultures these forms usually disap-
peared. Clubbed cells and mycelial growth also were recorded by the last-named authors.
Tavel (1898) photographed budding and branching of a pseudo-tetanus bacillus from the
intestine, but did not mention this fact in his paper. Hibler (1908) noticed that the character-
istic drum-stick form of B. tetant became regularly visible in cultures several days old, while
with younger material the spores were very often located in the middle of the rods, which not
68 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

infrequently looked like clostridia. Substrates rich in carbohydrates produced such abnormal
forms exclusively.. Even in old cultures no plectridia appeared under such conditions.

The anaerobic Vibrio Rugula, whose spore formation has been closely studied by Praz-
mowski (1880) was placed by Lehmann and Newmann (1912, p. 156, footnote) in the neighbor-
hood of B. oedematis maligni. In fact, it seems certain that this organism has nothing to do
with the genus Vibrio, as it is accepted in the modern literature. Dujardin (1841, p. 218)
placed it close to his Vibrio (now Bacillus) subtilis, and he pointed out already, that Vibrio
Rugula grows either curved or straight, occasionally forming “‘fils alternativement droits ou
flexueux.” Prazmowski’s drawings are reproduced as figure 28 on Plate H (from original figs.
10 and 11 on Pl.I). As this organism was able to decompose cellulose, and Omelianski’s photo-
graphs of his anaerobic cellulose bacilli show the same curved cells with terminal spores, it seems
correct to assume that this is the proper classification of “Vibrio Rugula.”

(d) SPIRILLA AND SPIROCHAETS,

The question has not been definitely settled whether the spirochaets are to be con-
sidered to be true bacteria, or whether they should better have their place between bacteria
and protozoa, or perhaps among the latter themselves. Morphological, cytological, as well as
physiological, reasons have been submitted in favor of the latter view. Prowazck (1907a)
mentions in this respect especially the ribbon-like structure, the undulating membrane, the
longitudinal division and the sexual differentiation, which, as well as the cysts, in his opinion
are never to be found with bacteria. Keysselitz (1907) adds that also the flexibility of the spiro-
chaets should be accepted as a proof of their protozoal nature, because he considers bacteria
to be always of a rigid structure. Hdélling (1911) and Klimenko (1914) point to cytological
reasons in favor of the protozoa theory. Krzysztalowicz and Siedlecki (1908), as well as
McDonagh (1912-13), furnished many details concerning sexuality and reproduction, which
would support this view. Carpano (1914) even believes the rosette-like arrangement, some-
times displayed by spirochaets, to be valid proof of their protozoal nature. Novy and Knapp
(1906), on the other hand, emphasized that on account of their flagellation, transverse division,
and other marks the spirochaets should be united with the bacteria. Zettnow (1906) took also
his stand against the assumed longitudinal division, and he thought, in addition, to have
obtained proof of peritrichous flagellation. This, however, was refuted by Keysselitz (1907),
Schellack (1907-09) and Fantham (1908) as being an error, caused by a separation of the fibrils
in the undulating membrane; but Schellack (1909), as well as Dobell (1912) and Swellengrebel
(1912), have shown that the cytology of these organisms is the same as that of bacteria. Trans-
verse division and other reasons are mentioned by the two last-named authors in favor of their
opinion that the spirochaets should be placed as a special group among the bacteria. The
studies of Gross (1912) and of Meirowsky (1914b) upon their modes of reproduction equally
support this view. In addition, Meirowsky (19142) has pointed out that the true branching
found by him with all spirochaets tested, should be accepted as strict proof of their bacterial
nature. It goes without saying, that he was mistaken in this direction. Nevertheless, it
seems that an unbiased consideration of all items concerned, including especially the repro-
ductive processes to be discussed in Chapters II-IV, will support the view that the spirochaets,
or at least a part of them, show much more the general character of the bacteria than that of
the protozoa. Some organisms now included in this group might perhaps be better separated
and brought into closer connection with the protozoa. At present, however, where their life
history is hardly much better known than that of the bacteria related with them, it seems more
useful to take advantage of all the results collected so far, and especially of those which might
be helpful in further investigations upon the life cycles of their bacterial kinsmen, the vibrios
and spirilla.

Vibrio cholerae, though at first also declared by R. Koch and his pupils to be strictly mono-
morphous and absolutely constant in its activity as in its appearance, can now be considered
to be one of the best known examples of a wide pleomorphism. It is very much inclined to
assume all the form types of the old form genera of F. Cohn. As early as in 1885 its ability: to
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 69

grow in giooular shape was discovered by E. Klein (1885a). He found that at room tempera-
ture, but not at 38° C., after several days the curved bacilli pass over into circular forms, which °
later split up lengthwise into two regular comma bacilli. Ferrén (1885) had analogous results
in diluted broth (Koch’s broth is declared by him to be too concentrated). He calls these
round forms “‘oogones,” and reports that observations in the hanging drop also revealed the
presence of smaller round bodies which, as ‘“pollinodes,” fertilized the oogones. Later the
acepore are said to have liberated granules in large numbers, which in their turn grew up to

“corps muriformes,”’ and that these havereproduced some long fine spirilla, which later subdivided
themselves to typical vibrios. Hrmengem (1885), who quoted these findings extensively, was
only able to confirm the report of the Spanish author as far as the granules and the “corps
muriformes” were concerned, and when one reads Ferrén’s original report, wherein he comes
to the conclusion that the cholera vibrio is a type of growth of a ‘‘Peronospora Barcinonae,”’
it appears, indeed, by no means surprising that this work could not gain any acknowledgment
among careful bacteriologists at that time. Yet, in the light of many other facts discovered
more ‘recently, and after separating facts from ehanries, it becomes very probable that Ferrdn
has seen nearly all of the complete life cycle of this organism. The data to be given in the next
chapters upon gonidia, gonidangia, regenerative bodies, symplasm and conjunction, will eluci-
date and confirm many of his findings. That Ferrén also saw long fusiform threads containing
granules in old cultures, and that he found it difficult to revert these straight forms to the normal
curved rods, is another point of interest. Moreover, it should be mentioned that Ferran,
according to his report, was able to find all the different forms not only in his cultures, but also
in the feces of diseased persons, as well as in the exudates from inoculated animals. Babes
made also some observations upon morphological changes and on differences in the appearance
of the colonies of the cholera organism as early as in 1885. Klebs, too, wrote in 1887 (p. 359):
“in sehr zahlreichen Fallen enthalt der Choleradarm nicht die characteristischen Komma-
formen, sondern Modificationen derselben,’’ and he mentions as such especially ovoid cocci
and large globules. .

Papers by Dowdeswell (1889-1890), as well as by Schroen (1891), have brought much
confirmation to the discoveries of H. Klein and Ferran. However, they also met the same
fate, though the observations recorded therein were based on continuous microscopic obser-
vation of the living organisms on the warm stage. In addition to the forms mentioned,
Dowdeswell described and made drawings of branched filaments and of large triangular cells,
which only recently have been rediscovered and photographed. Metchnikoff (1894) got large
dumb-bell shapes, showing ‘lateral buds and branches, very similar to those which he first
noticed with the bacilli of avian tuberculosis, when he grew the cholera vibrio in symbiosis
with a white coccus. Cells looking very much like Actinomyces, small ovals, cocci, short
straight rods, and spermatozoid-shaped bodies have also been seen by him, and the cholera
organism was therefore classed as an ‘“‘example frappant du pléomorphisme si répandu dans
le monde des bactéries.” Gruber (1894) pointed out that some strains differ widely from the
typical cholera vibrio, either by their pointed ends, or by assuming the shape of nearly straight
rods possessing 4-6 long polar flagella. Motility, appearance of the colonies, nitroso-indol
reaction, as well as the behavior in the animal test, were all found to be more or less variable.
Gruber pays also special attention to the ‘Vibrio Ivanoff”’ which at first was declared by
R. Koch to be a genuine cholera vibrio; later, however, an ‘‘entirely different species.” In
their “Atlas” Fraenkel and Pfeiffer (1895) published a good picture of various forms studied
by the authors mentioned above, which to them, however, are merely ‘‘involution”’ forms,
because they came from a broth culture three weeks old. Their photograph (original fig. 99
on Pl. XLVIII) has been reproduced as figure 79 on Plate VII; the ‘‘crippled involution
forms’’ stain well, while the ‘‘typical’’ vibrios are pale and dissolving. As figure 80 a repro-
duction from Itzerott and Niemann’s “‘Atlas’’ (1895, Pl. IX, fig. 53) has been added, showing
again what these authors term “‘involution” forms, though ihe very characteristic appearance
and the great similarity of the cells visible in both pictures are certainly not in agreement
with the assumed “degeneration.”
70 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor. XVI.

+ Ounningham (1897) emphasized that “many vibrionic organisms under certain circumstances
have the power of multiplying indefinitely i in the form of cocci, diplococci, and short straight
rods, although under other conditions they are capable of assuming typically vibrionic char-
acters.’ And he adds: ‘This is a fact which is well known to everyone who has practically
studied the subject.” Gamaleia (1900, p. 207) obtained giant spirilla and globules within 24
hours, when he grew the cholera cultures in broth containing }~1 or on agar with 1-2 per
cent lithium chloride. Nakanishi (1901) noticed that the globules of V. cholerae and Metch-
mkoviit, whose diameter was about four times as large as the width of a vibrio, reverted readily
to the typical comma form, even if they were taken from cultures 15 weeks old. Kohlbrugge
(1901a) got long branched Slaments of the cholera organism in blood serum, which he (erro-
neously) described as ‘‘Cladothrix-like.’” By repeated inoculation into animals Shibayama
(1902) even succeeded in developing strains producing constantly thin branching filaments,
which, however, were quite avirulent. Most of the important results recorded by Almquist
(1904-1917) will have to be discussed in Chapter II, as referring to reproductive organs and
processes. His cultures gave also small and large coccoid forms, straight rods which mor-
phologically resembled the typhoid organism, and much budding. One culture, mentioned in
his 1916 paper, grew during two years constantly in coccoid shape, though still exhibiting
the normal behavior in the agglutination test. Maassen’s (1904) work contains interesting
details concerning the stimulating effect of various salts in bringing about the morphological
changes in the cholera vibrio, which to the author are merely ‘‘pathologic deformities.” Péju
and Rajat (1906e) found that 5 drops of their potassium iodide solution added to the cholera
culture produced coccobacilli, while 8-9 drops gave long threads. Hammerl (1906) discov-
ered spherical bodies of 1-6» diameter and branched cells in many cases. They were present
already, and sometimes exclusively, in young cultures; they could be transferred for weeks
and months, and their motility was very pronounced. The author, therefore, correctly refutes
the “involution” theory. Horowitz (1911) not only observed the same pleomorphism as
recorded by the earlier authors, he reports also upon a very great variability of the agglutina-
bility of his strains, which experiments, however, can not be accepted as conclusive. Lehmann
and Neumann (1912, p. 513) say concerning the ‘‘atypical’’ forms of V. cholerae:

Unter besonders giinstigen Bedingungen . . . trifft mam . . . vorwiegend kurzovale, kokkenartige
Gebilde. . . . In salzarmen, besonders aber auch in sehr salzreichen Fliissigkeiten bilden manche Cholerastimme
auffallend geblihte bis kuglige Formen, die vollkommen fortpflanzungsfihig sind. . . Wir haben auch manch-
mal lange, schéne Faden gefunden, die von Fluorescens nicht zu unterscheiden waren, in den nachsten Culturen
aber wieder nur als lange diinne Stébchen auftraten. Wirklich gekriimmte Formen waren selten. Dies Spiel zeigt
sich bei einer echten Kultur aus Hamburg von 1892 stammend nun schon seit iiber 10 Jahren.

The most complete survey of the different types of growth exhibited by V. cholerae has
been given by Stamm (1914). Some of his photographs are reproduced as figures 81-83 on
Plate VII (from original figs. 2, 4, 5, and 15 on Pls. XI and XII, and figs. 1 and 2 from the
text, p. 512 and 524). He cultivated 13 strains under comparatively natural conditions, viz.,
in water; 6 of them remained constant, while 7 changed in their morphological as well as in
their physiological behavior to a considerable extent. They all lost their agglutinability, and
they did not return to the original type within 30 months, if frequently transferred. Inocu-
lations from old cultures, however, gave again curved forms. The cocci, though themselves
not agglutinable, produced a rabbit serum which agglutinated the typical V. cholerae. 37
strains, isolated by Popoff-Tscherkasky (1914) during the Balkan war, and kept for 4-9 months
without being transplanted, grew after the first reinoculation as a mixture of cocci, short rods,
and vibrios. The next transfer on agar gave after 24 hours at 37° C. very large irregular
threads, big globules, and large rods, but not ‘a single vibrio. These forms replated on agar
gave only colonies containing vibrios of somewhat larger size, which then again proved to be
stable. That the appearance and disappearance of the different forms is the result of the
various phases of the life cycle of the cholera organism has been ascertained by Olsson (1915)
under Almquist’s direction. Atypical wrinkled colonies always gave him immotile cells, but
these could be changed back into the motile form producing the typical colony.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 71

How the strictly monomorphistic view, as established by R. Koch, in regard to his “‘Komma
Bacillus,’ has changed within the last 30 years, even among his collaborators and followers,
may be seen from the following quotations: In 1892 Friedrich in volume 8 of the ‘“Arbeiten
aus dem Kaiserlichen Gesundheits-Amte” tried to deal effectively with all that had been dis-
covered by authors like Dowdeswell, Cunningham, and others. As far as this was not in accord-
ance with the “‘legitimate”’ findings of R. Koch and his pupils it was easily ‘explained”’ as
plasmolysis, vacuolisation, granular decomposition, degeneration, or simply fiction. Dowdes-
well’s ‘‘claims”’ were ‘‘not to be taken seriously.”’ Nevertheless, some alterations in the appear-
ance of the colonies and cells, as well as in the pigment production, are admitted, though on
account of their instability they were not accepted as true variation. In 1903 Kolle and Kolle
and Gotschlich admitted that the morphological marks, except the ‘‘legitimate”’ polar flagellum,
can not be considered specific. The same was declared to be true in regard to the colonies
on the gelatine plate, which were mostly found to be dimorphous. Only the agglutination test
was strictly upheld, and those cases discarded as not being cholera, where a quickly fatal disease,
showing the symptoms of cholera, was caused by vibrios which were not agglutinated by the
true cholera serum. In 1912 Baerthlein furnished another contribution in volume 40 of the
“Arbeiten aus dem Kaiserlichen Gesundheits-Amte’’ wherein now, however, the “mutation”’
of V. cholerae and of other bacteria was discussed extensively. Three types of colonies made
up by different cell types were found and also traces of a cyclic development were discovered.
Additional data upon this subject were published by Feldmann (1917) and by Baerthlein (1918).

As was mentioned on page 56, Pollak (1913) obtained colonies and cells looking like those
of V. cholerae, when cultivating B. faecalis alcaligenes on the blood agar recommended by
Dieudonné for cholera diagnosis.

The life cycle of the vibrio isolated by Finkler and Prior (1884) from cases of cholera nostras,
which was named by Buchner on account of its pleomorphism V. proteus, received a fairly
complete description by its discoverers already in 1885. Their very interesting, although fre-
quently doubted, observations upon the reproductive processes will be quoted in Chapters
II and III. The curved form was found to occur only temporarily. It was soon replaced
by big globules, thin and thick rods, club-shaped cells, and other forms very similar to those
observed much later also with V. cholerae. The ends of the curved cells were either rounded
or pointed, sometimes their appearance was different at the same cell. BR. Koch (18846) defended
the “strictly specific” nature of his “‘Komma Bacillus” strongly against Finkler and Prior,
whose findings were discredited as being the result of experiments with impure cultures. The
last-named authors, however, could point out in return that Koch’s own slides exhibited also a
much greater pleomorphism of the cholera organism than this author was willing to admit.
Ermengem (1885), too, did not hesitate to take his stand against Finkler and Prior, because
differences in the appearance of the colonies and the cell form were to him full proof of “impuri-
ties.” Firtsch (1888), on the other hand, showed soon that unquestionably pure cultures of the
Finkler-Prior vibrio are able to produce four clearly different colonies, which, if found alone,
undoubtedly would be considered to be those of four different species. Motile and immotile
comma forms, spirals, straight rods, spindle-shaped, ovoid, and monad forms were again present,
and Firtsch succeeded in many cases to change experimentally the various types of cells and
colonies into each other. That his observations and conclusions were correct has been proven
more than 25 years later by Fiirst (1914), who with single-cell cultures secured analogous results.
The different cell forms remained so constant, when frequently transferred to fresh substrates
(after 5-6 days), according to standardized laboratory rules, that they could be easily mistaken
for different species. When kept for 24 months and then reinoculated many of them reverted
to the original vibrio type. And 4-months’ old cultures gave 100 per cent normal cultures.

Bonhoff’ (1896) isolated an organism from cholera nostras which first appeared very much
like B. coli, macro- as well as microscopically. Later, however, it assumed comma, S and screw
forms, while other cells presented themselves as diplococci. The rods possessed two polar
flagella, representing an interesting connecting link between Vibrio and Pseudomonas.
72 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vo. XVI.

E. Klein (1905) met in Cardium edule a pathogenic ‘‘ Vibrio cardii,’”’ which he places between
V. cholerae and V. Finkler-Prior. On agar, gelatin, and serum it forms abundantly convex and
globular cells, which retain their motility and propagative ability, and they are declared to be
very similar to those found before by the same author in cases of true cholera.

Kohlbrugge (1900) described a “new” vibrio from water under the (preoccupied) name
V. proteus, which grew on agar as a short rod, in gelatin and broth as a vibrio, and in water as
acoccus. Tested later (Kohibrugge, 1901 a) on blood serum, it produced long branched threads
as did V. cholerae. Another pleomorphous vibrio from water was described by Jorge (1896),
again resembling B. coli in several respects and apparently still more closely related to V. denitri-
ficans Severin, whose character and systematic position was discussed on page 59. _ Jorge’s organ-
ism grew on agar mostly as a straight rod, in broth very much like B. coli, while on gelatin
distinctly curved forms up.to long spirals were predominant, some of them showing inflations
at their ends. The pathogenic fluorescent vibrios, isolated from water by Fuhrmann (1905),
may also be looked upon as indicative of relations connecting this group and that of B. (Pseudo-
monas) fluorescens.

Vibrio phosphorescens Dunbar gave, when cultivated by Maassen (1904) on agar containing
2-3 per cent lithium chloride, branched cells in very great numbers. As figure 84 on Plate VII
Maassen’s two photographs are reproduced (from original figs. XIII, 10 and XIV, 3). The
forms shown are very similar to some drawn by Dowdeswell in his investigations upon cholera,
mentioned above.

Several of the highly pleomorphous marine Hali- and Photobacteria described by B. Fischer
(1894) should very probably take their place here. This is especially true concerning his
Halibacterium polymorphum, which gave practically all kinds of cell forms, and all these in
different sizes, in cultures not older than 2-3 days, and on most suitable substrates. There were
recorded (l.c., p. 36): globules, short rods, comma and screw forms, irregular cells of large size
(sometimes larger than yeast cells), pear-, hook-, and club-shaped forms, very long spindles with
pointed ends, and irregularly curved threads. Halibacterium pellucidum and Photobacterium
tuberosum. exhibited also much of the characteristics of a vibrio. It is the same with Photobac-
terium balticum according to the description furnished by Barnard (1899).

Weibel (1887-1888) isolated several Vibriones saprophiles, another from the mucus of the
nose, and one from the tongue, all of which exhibited pleomorphous tendencies. The “Vibrio aus
Nasenschleim”’ is said to have furnished “‘eine Musterkarte abenteuerlicher Runen und Schnér-
kel.” Bajardi (1903), who studied Weibel’s V. lingualis anew, obtained among many other
forms branched threads, which stained like B. diphtheriae when treated with Neisser’s method.
The organism, therefore, i is declared to be a ‘‘Streptothrix.” The picture given by Bajardi
(as original fig. 1) isreproduced as figure 85 on Plate VIII. The growth on gelatin resembled
Anthrax, and in broth Actinomyces-like clusters were found. Large yeast-like cells also became
visible, and in older cultures the long threads were again replaced by short cells, often containing
polar granules.

Bonhof (1896) isolated from water, besides other vibrios and spirilla, a “Vibrio Rugula”
with 18-20 polar flagella, which also gave branching and on the plate Anthrax-like colonies.
It is, no doubt, quite different from the anaerobic cellulose-destroying organism, described by
Prazmowski (1880) under the same name, and mentioned on page 68.

Probably the first example of branching among the spirilla was the frequently quoted
Spirillum endoparagogicum, discovered by Sorokin (1887-1890). The formation of branches
in this case was due, according to the author, to the germination of ‘‘spores”’ inclosed i in the
mother cell, wherein they appeared in various numbers. How far this statement can be accepted
as correct will have to be discussed in Chapter II. The drawing given in the first paper (1887)
is reproduced as figure 29 on Plate H.

Spirillum Undula grows, as Kutscher (1895 6) found out, in a small and in a large type.
The latter one showed all intermediate stages leading over to straight rods, as did also other
spirilla. Sometimes Kutscher saw that one end of a cell was curved, while the other one was
entirely straightened. He also noticed with Spirillum Undula, as well as with Sp. Serpens,
what seemed to him to be ‘“‘ganz eigenartige Gebilde,”’ viz. eumed branches breaking forth from
No. 2] LIFE CYCLES OF THE BACTERIA—LOHNIS. 73

the side or from the end of a spirillum. He does not want to decide whether or not this is true
branching. Sorokin’s papers were evidently unknown to him.

Doerr had the opportunity to see the same process with Spirillum pyogenes Mezineseu,
taken from broth as well as from the pus, as is demonstrated by the photographs accompanying
his paper, but there is nothing said about it in the text. The “new” Spirillum bataviae described
by Faber (1912) gave again ‘‘hornartige Auslaufer,’”’ and again the author feels unable to reach
a decision whether these are branches or not.

Reichenbach (1901) met very clear examples of triangular cells in six days old broth cultures
of Spirillum rubrum. Some of his photographs (original figs. 6, 8, and 9) are reproduced as
figures 86 and 87 on Plate VIII. They are very similar to those published by Severin from his
V. denitrificans and by Bordoni- Uffreduzzi from his Proteus hominis capsulatus which have been
mentioned on pp. 55 and 59, respectively. Other branched cells of Spirillum rubrum, reproduced
as figure 30.0on Plate H from a drawing of Meirowsky (1914 b) are interesting counterparts to
Spirillum endoparagogicum and to B. radicicola. Spirillum tyrogenum Denecke gave Meirowsky
analogous results.

Concerning the morphology of the spirochaets only comparatively few data are available
at present, as the numerous pure cultures isolated by Noguchi (1911-1912) have not been
closely studied in this direction.

According to Sobernheim (1907, p. 536) atypical forms of Spirochaeta pallida are not infre-
quent in syphilitic tissue. He says:

Abgesehen von ungewohnlich zarten Spirochaeten, sowie von sehr kurzen Spiralen mit verhaltnismissig lockeren
‘Windungen, sieht man zuweilen Exemplare der Spirochaete pallida, die schleifenformig gewunden und verschlungen
aind. * * * Involutionsformen in gestalt von kurzen, deformierten Fragmenten, gekérnten spirochaeten-ahnlichen
Gebilden, aufgeknauelten und verklumpten Spivodhwaten sind mehrfach beschrieben worden.

Castellani (1905) found in yaws besides, or instead of the spirochaets, oval irregular bodies,
usually 5-8 », but also smaller or much larger ones. Analogous results were reported for
syphilis by Krzysztalowicz and Siedlecki (1908) and by McDonagh (1912), which will have to be
considered more fully in Chapter II. Miessen (1898), too, got Blastomyces-like growth from
his “syphilis bacillus” or ‘“‘Syphilomyces.”’ It is absolutely impossible, however, to separate
clearly the correct findings from the probably more numerous incorrect ones, contained in all
of this author’s publications. Contaminations, especially those of diphtheroid bacilli, have
obviously played a conspicuous réle in these investigations.

Branched growth of Spirochaeta pallida, as well as of Spirochaeta balanitidis, stomatitidis,
and gallinarum, has been recorded by Meirowsky (1914 a and b). Some photographs published
in his second paper (original figs. XI, 46, 55, and 56) are reproduced as figure 88 on Plate
VIII. They show an exquisitely myosial growth of Spirochaeta pallida. The occurrence of
coccoid bodies within the life cycles of these and of other spirochaets have been ascertained by
different authors; but, being evidently reproductive organs, they are to be considered in
Chapter Il,

With Spirochaeta Obermeiers granular and rod-like forms have been seen by Karwacki
(1912). Whether the pathogenic bacillus, isolated by Afanassiew (1899) from a recurrens
case, is to be accepted as a type of growth of the recurrens spirochaete, remains to be seen. It
is reported to have grown first as a fine motile rod (0.31.5 uw), which later passed over into
curved threads of 5-14 » length, sometimes with pointed ends and granular structure.

Inada, Ido, Hoki, Kaneko and Ito (1916) have published some very interesting pictures
of the ring forms, buds and branches of their Spirochaeta icterohaemorrhagica, found by them
to be the causative agent of Weil’s disease. Two of their drawings (original figs. 69 and
70 on pl. 61) are reproduced as figure 31 on Plate H.

The Vibrio (vel Spirochaeta) suis, which by Rtither (1910), as well as by King and his
coworkers (1913), has been connected with hog cholera, grows according to the first-named
author in comma or in the typical spirochaete form, which, however, occasionally divides
itself into fine rods, resembling B. influenzae or septicaemiae, or assumes a Streptococcus-like

appearance.
74 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI.

A problem which remains to be solved despite a considerable amount of work already done
in this direction, is the question, whether the fusiform bacilli, practically constantly to be
found in symbiosis with spirochaets in healthy organs, as well as in pathogenic processes, like
angina, pyorrhea, stomatitis, gingivitis, noma, are merely symbionts or parts of the life cycle
of the spirochaets.

The bacillus found by Babes (1893) to be connected with scurvey, causing gingivitis and
hemorrhages, which grew only on agar first used by streptococci, presented the typical fusi-
form appearance, but gave also curved threads ‘“‘longer and thinner than cholera bacilli.”
In a later publication (1907) of the same author analogous results are reported about angina.
Silberschmidt (1901) described a fusiform bacillus from an abscess, exhibiting all intermediate
steps leading over to thin spirilla, of which, however, no pure cultures could be made. Beitzke
(1904) gave a summary of the work done up to that time, which together with own experi-
ments led him to the conclusion, that fusiform bacilli and spirochaets belong together geneti-
cally. On account of the occasional occurrence of branched threads he is inclined to place
these organisms in the neighborhood of Bang’s necrosis bacillus and the Actinomycetes.

Wechselmann and Loewenthal (1905) noticed that in syphilitic tissue very thin rods with
pointed ends and metachromatic granules may accompany the spirochaets. From micro-
scopic studies upon angina Mackie (1905) drew the conclusion that the fusiform bacillus, the
spirillum, certain sickle-shaped and some ribbonlike bodies, found in such material, “will
all turn out to be different stages in the life cycle of one parasite.’ The most important con-
tributions thus far made in support of the view that bacilli and spirochaets belong together
are those furnished by Ruth Tunnicliff (1906-1915). Pure cultures of fusiform bacilli isolated
(1906) from material taken from the mouth proved to be ‘“‘extremely pleomorphous.” Spiro-
chaets appeared only temporary in large numbers, and being not easily stained they may
have often remained unobserved, although being actually present. During the first 24 hours
delicate pointed rods of 3-10» length, often resembling the barred type of diphtheria bacilli,
became visible. Some plumper bacilli (1.54) with deeply stained rounded ends also were
seen. The next day long filaments developed, and soon afterwards spirals appeared, some
of them also containing dark bodies. They usually showed pointed ends, were Gram-negative
like the bacilli, and sometimes clearly made up of a chain of rods. "Work done with the anaerobic
Bac. rhinitis (1913-1915) confirmed and extended the earlier findings. In another paper by
Rosenow and Tunnicliff (1912) the same dimorphism was described in regard to an anaerobic
organism isolated from a fatal case of pyemia. The spiral forms were frequent in the fluid
of condensation of serum agar. Coccoid forms, too, were present which, however, obviously
are to be classed as reproductive organs. Proca (1908) saw satisfactory development of B.
Jusiformis, when it was grown in symbiosis with B. subtilis or with streptococci. Such mixed
cultures produced in broth many spirilla with large, often unequal undulations and thin ends.
On agar and gelatin, especially at the surface, the typical straight forms prevailed, while in
the depth of the agar curved forms also appeared. The “ Cladothria stereotropa,” found by
Proca and Danila (1910) in syphilitic products, is said to have produced besides diphtheroid
forms, fusiform bacilli as well as spirilla.

Shmamine (1912) thinks that the “nadelférmigen” bacilli ocurrring in syphilitic material
are a type of growth of Spirochaeta pallida. Both organisms produce the same colonies, are
identical in staining qualities, and the transformation of the bacilli into spirochaets was observed.
These ‘‘needle-shaped” bacilli are considered to be different from B. fusiformis. Carpano
(1914) obtained analogous results in his investigations upon the “‘fuso-spirillire Symbiose”
in necrotic gangrenous affections; and pure cultures of the spindle bacilli, isolated by Klimenko
(1914) from scarlatina, gave also spirilla after a few transfers. Again both organisms stained
in the same manner, and salvarsan acted upon these bacilli as it does upon the spirochaets.
Ozaki (1915) found that a spirochaete from the human mouth jn pure culture always pro-
duced straight bacilli, which, however, were taken to be not identical with B. Susiformis Vincent.
In their report upon a fatal case starting from pyorrhea, in which the fusiform bacillus was
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 75

isolated from the blood, Larson and Barron (1913) also took the standpoint that bacilli and
spirochaets belong together.

On the other hand, the opinion has been repeatedly expressed that spirochaets and bacilli
are two different organisms, because they could be separated and kept for some time in pure
cultures without showing any change in their morphology. Hllermann (1904-1907), Vincent
(1905), Miihlens and Hartmann (1906), Veszprémi (1907), Doflein (1909), Peters.(1911), Leh-
mann and Neumann (1912, p. 578), Krumunede and Pratt (1913), Kendall (1916, p. 531), for
instance, share this view. However, taking under due consideration in the first place the
experimental difficulties, so often encountered with in growing these organisms on artificial
substrates, in the second place the abnormal conditions prevailing in artificial cultures, and
in the third place the analogous outcome of the very numerous experiments carried on with
other organisms, apparently though incorrectly proving their monomorphism and constancy,
it seems best not to discard too rashly the positive findings mentioned above, and to wait for
further thorough and unbiased research. It can, e. g., not be accepted as justified that Arnheim
(1914) rejects Tunnichff’s results in their entirety, though he himself observed the trans-
formation of spirochaets into rod-like forms (‘‘staibchenartige Gebilde’’), as well as the preva-
lence of the latter forms in some types of syphilitic eruptions. And when he lightly discards
the round bodies, which he also had the opportunity to see, as “‘Zerfallsprodukte,’”’ he merely
adds another unfounded assumption to the multitude of similar assertions concerning ‘“invo-
lution forms, degeneration,” etc. It is quite characteristic that Tunnichf in her first work
upon this subject, done together with Weaver (1905) also did not find any connection between
fusiform bacilli and spirochaets, but that her later more searching investigations led to the
opposite standpoint.

One point which needs emphasizing in this connection is the necessity to distinguish
more sharply between the different bacilli included in this group than has been often done
so far. The papers published by Lewkowicz (1906), Babes (1907), Veszprémi (1907), Costa
(1909), Shmamine (1912), Larson and Barron (1913), Ozaki (1915) and others, furnish valuable
material in this direction. It is very probable that some of these incompletely studied: so-
called fusiform bacilli have nothing whatever to do with spirochaets, while in other cases the
opposite will be true.

A peculiar organism with a creeping motion, isolated by Miehe (1913) from leaves of tropical
plants, and described by him as Bact. repens, seems to be related to those members of the Fusi-
formis group, whose character exhibits protozoon-like traits (cf. Babes, 1907). As figure 32
on Plate H two of Miehe’s drawings are reproduced (from original figs. 9 and 10 on PI. II)..
The curved, branched, and swollen, so-called involution forms duplicate very closely the similar
growth of the spirilla.

(e) HIGHER BACTERIA (TRICHOBACTERIA, MYCOBACTERIA).

As organisms are included among spirochaets and fusiform bacilli, which evidently act as
links connecting bacteria and protozoa, so among the so-called higher bacteria the relations
become more and more apparent, which lead over to the fungi on the one side, to the algae
on the other.

Many of the details collected on the preceding pages amply support the statement made
on pp. 21-23 that all bacteria studied more thoroughly exhibit traits which so far have been
generally considered to be characteristic for Actinomyces and some closely related organisms.
Special genera (Actinococcus, Actinobacter, Actinobacillus) have been proposed, and it is a
fact always to be kept in mind that especially the large endospore-forming bacilli of the
Anthrax-Subtilis-Mycoides type are easily induced to enter a development very much resembling
that ofa typical Actinomyces. Some of the pros and cons in regard to a transplantation of
Actinomyces and its relatives among the fungi have been mentioned already (on pp. 21-23),
and others will be discussed in Chapters II and III. As some of the Actinomycetes are known
to enter temporarily a motile stage, while others have thus far always shown themselves to
remain permanently immotile, one might feel induced to accept this difference as the dividing
76 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

line between those belonging to the bacteria and those belonging to the fungi. In many cases,
however, presence or absence of motility can not even be used as a means to separate bacteria
(like B. coli, dysenteriae, mycoides, subtilis, fulwwm, and others) into different species. There-
fore it would be at least premature to use this mark as a basis for separating those organisms,
at present all grouped together as Actinomycetes, into bacteria and into fungi. Thorough
investigations upon their complete life cycles are also in this case quite indispensable before
any well-founded decision can be rendered.

The same holds true concerning Beggiatoa, Cladothrix, Crenothrix and related organisms,
which repeatedly have been declared to belong to the algae, not to the bacteria. Again this
may be the case, but much of what is known at present seems to indicate their bacterial nature.
As will be shown in Chapters IT and III, it is especially their modes of multiplication and repro-
duction which connects them more closely with the bacteria than has been thought before.
In fact, as Actinomyces has been declared (by Claypole, 1913) to be an “ancestral type”
for bacteria, as well as for fungi, so Crenothrix, Cladothrix, Beggiatoa and related forms
may be looked upon as prototypes, showing clearly such reproductive processes as occur also
with all bacteria, though they nearly always have been overlooked thus far. It will be much
more useful, therefore, if also these organisms will be made the objects of further bacteriological
investigations, than to exclude them prematurely on account of insufficient systematic reasons.
The general character of an alga, of a fungus, or of a protozoon is rather different from that
of a Beggiatoa, of an Actinomyces, or of a Spirochaeta: Certainly these organisms are also
more or less different from the majority of the bacteria, but just as some of their characteristic ©
traits can be found with true bacteria, too, it seems best to retain them in their present posi-
tion as “higher bacteria.”

The organism described by Bang under the name abortus bacillus grows in the body as a
fine small rod similar to B. erysipelatos suwm. That the latter species is related to the Acti-
nomycetes, has been emphasized by Rosenbach, who established for it the genus Erysipelothrix
(see p. 52). That the abortus bacillus should take its place among the Corynebacteria, in-the
neighborhood of glanders, diptheria and diphtheroid organisms, has been pointed out by
Preisz (1903). The cells growing in the cultures are either short, coccoid, sometimes very
similar to the double-pointed forms of Streptococcus lanceolatus, or they are longer, thin,
irregularly stained like the barred type of B. diphtheriae, passing over into wedge-, spindle-,
and club-shaped forms. Y and more complex branching was found to be not infrequent.
The Gram staining is said to have always given negative results. Interesting relations connect-
ing this species with M. melitensis have been mentioned on page 44,

The necrosis bacillus of Bang is also to be placed here as being identical with the so-called
Streptothria cuniculi, described by Schmorl (1891). Cocci, motile and immotile rods, threads,
branching and the formation of drumstick-like cells have been recorded by this author. Two
of his photographs are reproduced as figures 89 and 90 on Plate VIII (from original figs. 2
and 5 on Pl. VII). W. Ernst (1902) did not find any motile rods, but he gives many details
concerning the formation of branched threads, large inflations, ball-, pear-, and bottle-shaped
cells. 4 5
The Streptobacillus described by Tizzont and Angelis (1913-1915) as causative agent of
pellagra, grows at first as motile thin rods, which break up into typical, but gram-negative,
streptococci. These pass over into gram-positive staphylococci, which eventually reproduce
rods. The latter were often found interspersed within the streptococcus chains. These,
too, show branching; and clusters (‘‘Drusen”) with granular center and clubs at the outside
also have been seen. The colonies made up by the cocci were flat, white, glistening, and liquefied
gelatin.

That the glanders bacillus may be easily induced to change its character was emphasized
by Th. Smith as early as in 1890. Acid reaction of the substrate caused a yellow-orange growth,
while alkaline reaction made it grayish. A great variety of ‘puzzling involution forms” was
noticed by this author. Semmer (1895) reported that glanders, as well as tubercle bacilli, are
much inclined to form under saprophytic conditions threads which often show inflations, clubs,
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 77

and branches. ‘The short rods are declared to be a type of growth adapted to the environmental
conditions within the body. Marz (1899) secured further information upon the pleomorphism
of this organism. Figure 33 on Plate H is a reproduction of his figures 1-4.

Galli- Valerio (1899, 1900) confirmed and extended these findings by a close study of the
development of the glanders bacillus on different substrates and in the hanging drop. Small
and large rods, clubs, dumb-bell-shaped and drumstick-like cells, as well as branching, were
observed. The drumsticks alone are declared to be “involution” forms, not the others. The
short small rods were found to appear in old cultures, after the large branched threads had
disappeared. More data and pictures have been furnished by Conradi (1900). Branches
appeared in his cultures already on the second day; basis and apex of the threads could be
easily discerned. The organism is declared to belong to the Hyphomycetes. Excellent photo-
graphs, clearly illustrating the wide pleomorphism of B. mallei, have been published by
Carpano (1913) and are reproduced as figures 91-96 on Plate VIII. As was emphasized by
this author all the various forms appeared on the same substrate (acid glycerin horse-meat
agar), which fact will be of special importance for future investigations upon the life cycle
of this organism.

Before the etiologic significance of the diphtheria bacillus was fully established, repeatedly
eoccoid forms have been made responsible for this disease. It may be readily admitted that
the presence of true micro- and streptococci might have caused some erroneous conclusions.
Nevertheless, as new and well-founded observations have shown that the diphtheria bacillus is
able to appear in a distinctly coccoid shape, those older findings deserve once more some interest.
When Qertel (1871) directed the attention to the ‘‘micrococci” as causative agents of diph-
theria, he cautiously pointed out that the “purely botanical problem concerning their true
nature and relationship’”’ should not be touched by him. He only made use of that name
in the general sense as was done by Haller, and he mentioned already that not only round
but also long forms became visible which even were seen to sprout forth from round cells.
With less precaution F. Cohn (1872 b) created ‘a distinct species Micrococcus diphtheriticus,
which later was transferred by Zopf (1885, c p. 53) into the genus Streptococcus. R. Koch (1878)
also emphasized ‘‘that in diphtheritic deposits large numbers of micrococci are present,’ and
he thought that probably the same cocci were active in pyemia as in diphtheria. Baumgarten
(1890, p. 353) says:

Es ist kein stichhaltiger Grund vorhanden, dem Streptococcus pyogenes die Anerkennung als eines Erregers der
Hals- und Rachen-Diphtherie zu verweigern.

Guenther (1906, p. 577) takes a similar standpoint.

The pleomorphism of the true diphtheria bacillus was first studied by Zarniko (1889).
He observed club-shaped cells, three- to four-fold longer than the normal rods, segmentation
of the original bacilli, and the separation of these segments, which assumed spherical and
ovoid forms. All these ‘‘atypical” cells were declared to be ‘“‘involution”’ forms, though
they were well and uniformly stained. E. Klein (1890-94) refuted this assertion. Club-
shaped, as well as budding and branching forms, which he was the first to observe, are
said not to be involution form, because they are present in young cultures and at the out-
break of the disease. The same holds true, as was stated in the second paper (1894), with
regard to the spherical forms. In his monograph upon the diphtheria bacillus Escherich
(1894) expressed the opinion that the branched cells present in the diphtheritic membranes,
as well as in other parts of the afflicted throat, should be classed as secondary infection caused
by a “‘Cladothrix,’’ and he was inclined to accept as erroneous diagnoses those findings where
cocci alone had been seen. Typical is, according to his opinion, the small wedge-shaped rod
with rounded ends (0.3-0.5 x 1.2-2u), which may pass over into elongated and clubbed
forms. Spindle-, dumb-bell-, pear-shaped, as well as curved cells, were also seen. Moreover,
from old cultures, as well as from the animal test, very long irregularly curved threads were
obtained, which exhibited various inflations, some of them assuming the shape of chains of
giant cocci. Weakly stained rods are declared to simulate sometimes streptococci, the chro-
78 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

matic granules taking the stain much better than the surrounding sheath. Concerning the
relations existing between true diphtheria and diphtheroid bacilli Escherich takes the stand-
point that both are strictly different, although he admits, of course, that the diphtheria bacillus
may lose its virulence entirely. ; .

His mistake concerning the absence of branching was soon corrected by a short paper pub-
lished by Fraenkel (1895), wherein it is pointed out that the method usually followed in making
the smear nearly always destroys the branched forms, which, however, are produced at least
by some strains quite regularly in fresh serum or egg cultures. Bernheim and Folger (1896)
soon confirmed Fraenkel’s observations, while Kanthack (1896) directed the attention of the
German bacteriologists to the fact that analogous results had been previously obtained in Eng-
land by F. Klein and by himself. Zupnik (1897) noticed that not only the cell form, but also
the form of the colonies of the diphtheria bacillus can vary to a great extent. Some of his
strains were found to be motile. In 3-4 weeks old broth cultures Madsen (1897) observed
many small and large, regular and irregular globules, which looked like a contamination by
cocci. But all intermediate stages leading over to the clubbed rods were visible, and plate
cultures showed that the strains were pure diphtheria bacilli. More details concerning the branch-
ing of the diphtheria bacilli were furnished by Kurth (1898). The club-shaped form of the
branches shown in his pictures makes an exact duplicate to the drawings of branched strep-
tococci, published by Babes (1895) and reproduced as figure 3 on on Plate B. He thinks that
such forms are restricted to the true diphtheria bacillus, and that they could be used, therefore,
in differentiating from the pseudo-diphtheria group. Evidently he was unaware that Babes
(1895) had already given the analogous pictures of diphtheroid organisms, reproduced on
Plate J as figure 34. Very interesting is the observation of Kurth that in symbiosis with Strepto-
coccus lanceolatus the diphtheria bacilli became so similar to them, that they hardly could be
distinguished. Meyerhof (1898) once more discussed the great pleomorphism of the diphtheria
bacillus, and confirmed that branching is very frequent on egg substrates. Very careful studies
upon dividing and branching of the diphtheria bacilli have been made by Hill (1899, 1902),
who for this purpose made use of his newly developed ‘‘hanging agar block” method. With
young cultures he succeeded in observing directly the ‘‘snapping off” of oval side branches and
their stretching out to normal rods.

It was by no means in accordance with the facts known at that time, and in part without
logical foundation, too, that Migula (1900, Vol. II, p. 499) wrote concerning the diphtheria
bacillus:

Dass diese Verzweigungen . . . nichts weiter als Involutionserscheinungen sind, braucht wohl nicht erst

besonders hervorgehoben zu werden . . . Uberhaupt ist der Diphtheriebacillus sehr leicht zur Degeneration in
morphologischer Hinsicht geneigt, und dies umso mehr, je besser er sich an die ktinstliche Kultur gewdhnt hat.

Weesbrook, Wilson, and McDaniel (1900) arranged the different forms shown by the diph-
theria bacillus into seven types; each of them containing cells of granular, barred, and solid
staining. Granular forms were found to predominate at the beginning of the disease. The
resemblance to Streptococcus lanceolatus is very pronounced in types D and G. As A. C. Abbott
remarked in the discussion following the reading of this paper, representatives of all seven types
probably occur in every culture. Many of them, however, are according to his opinion “simply
involution forms.” Their appearance is greatly influenced by the hydrogen ion concentration
of the substrate (Bunker, 1917). Cache (1901) reported upon another case of considerable
persistency in branching in successive subcultures, and Conceiti (1901) secured further infor-
mation upon the reciprocal relation between virulence on the one side and club formation,
branching, and luxuriance of growth on the other side. Under anaerobic conditions the
Actinomyces-like forms could be reverted to the typical rod of high virulence. Additional
data in this direction are to be found in a paper published by Ohimacher (1902), some of which
were accepted by the author as proof of the possibility of changing true diphtheria into
pseudo-diphtheria bacilli. As was mentioned on p. 47, a puzzling ““streptococcus’”’ also was
found, which in various directions came very close to the character of a member of the
diphtheria group.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 79

Beck (1903) thought that the club-shaped form should be considered to be the normal type
of growth of the diphtheria bacillus; branching again was found to be frequent in young cul-
tures. A picture of a 1-day-old growth on agar (1. c. p. 765, fig. 1) shows clearly small and large
globular and oval forms. Spirig (1903) got typically Actinomyces-like growth from old diph-
theria cultures: long thin threads dividing up into rods and cocci, forming a white powdery

- (“Ixreidig’’) layer on the surface. But at the same time he revoked an earlier statement of
his (1899) concerning the development of branched threads. They were now declared to have
been those of a contaminating “Streptothrix,’”’ just as had been assumed by Escherich (1894)
in analogous cases (see above, p. 77). Goodman (1908) discussed fully the great morphological
and physiological variability of the diphtheria, as well as of the pseudo-diphtheria, bacilli. He
comes again to the conclusion that all be ‘but variants of a single species, B. diphtheriae, which
constitutes the entire group.”” The facts secured by Smirnow (1908) concerning the growing of
B. diphtheriae in coccoid form, when living together with streptocci and other bacteria, confirm
the analogous findings of Kurth (1898), mentioned above. Cappellani (1910) encountered re-
peatedly in the beginning of the disease in fresh exudate highly virulent diphtheria bacilli showing
good branching, which was reproduced in the transfers on different substrates. He supposes
again, like H. Klein (1890), that the rods are merely adaptations to the parasitic life, while the
actinomycotic growth is the normal development under saprophytic conditions.

In the course of diphtheria tests made in Hamburg, J. Dale (1910) repeatedly met with
unusual miscroscopic pictures, as shown in figure 97 on Plate [IX (reproduced from original
fig. 6). This photograph makes an interesting counterpart to figure 91, coccoid forms of
B. matlei. The secondary colonies, which developed from such coccoid growth. were again
made up of typical diphtheria bacilli. Soon afterwards Balfour (1911 d) reported that in the
tropics clinically true diphtheria cases frequently are caused by pure cultures of curious cocci,
showing Neisser’s granules and developing characteristic diphtheria colonies on Loeffler’s serum,
which, however, are composed exclusively by Gram-negative cocci. In subcultures on agar
typical rods were found within 48 hours, while on serum the cocci remained constant. The
photographs appended to the paper are quite similar to that of J. Dale.

A study of the diphtheria group (including pseudo-diphtheria bacilli) by the biometric
method was made by Morse (1912). Six morphological types are described and correlated with
cultural and biochemical characteristics (chromogenesis, acid production, virulence, toxin pro-
duction and complement fixation). They are arranged in two main subgroups: (1) true diph-
theria bacilli—(@) virulent, (6) avirulent; (2) diphtheroids, to which reference will have to be
made below. Berry and Banzhaf (1912) were unable to confirm the results recorded by Good-
man (1908) mentioned above. They explain them by assuming that Goodman had worked
with a ‘mutating organism.’’ Their own experiments show very wide and irregular variations
in the production of acid, despite the title of the paper: ‘‘Nonvariability of Diphtheria Bacilli.”’

That, in fact, the variability of these organisms is by no means inferior to that of other
bacilli has been fully demonstrated anew by Baerthlein (1913), by Bernhardt and Paneth
(1913), and by Schmitz (1916). Baerthlein noticed many changes in 2-4 weeks old cultures,
which seemed to him to be examples of true ‘‘mutation.’’ Three types of colony and cell form
were found to be regularly present. Bernhardt and Paneth, as well as Schmitz, conclude from
their observations upon the morphological and physiological variability of the diphtheria
bacillus in cultures and in the body, that true and pseudo-diphtheria bacilli are to be accepted
as strains of one species. Secondary colonies were especially inclined to furnish aberrant
forms. That the inoculation into the animal is very efficient in bringing about such changes,
as was observed by these authors, has been confirmed by Rémer (1914). Park and Williams
(1914, p. 38) mention the fact that ‘many large, intensely staining forms with one to several
apparent branches and many metachromatic granules” were usually present in 6-12 days old
cultures, and that they were the only ones which showed active growth and division. Thesame
authors also give (1. c., p. 294, fig. 119) an interesting picture of a strain which grew exclusively
in the form of a large coccus. Heinemann (1917) found irregularities in toxine production
connected with variations in growth and in microscopical appearance. Strains of low toxicity

160266°—21 ——6
80 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

gave thick yellowish films, which were composed of diplococci, resembling strepto-and meningo-
cocci. The great importance of: this observation for the correct diagnosis of diphtheria is
emphasized by the author, who evidently was not aware of the fact that he once more confirmed
what had repeatedly been discovered by several independent workers. A very complete
description of the wide pleomorphism of the diphtheria bacillus was published by Bergstrand
(1918). The occurrence of various globular forms was also thoroughly studied in this case.

The relations existing between streptococci and pseudo-diphtheria bacilli were pointed out
by Babes in 1895. As figure 3 on Plate B his drawings of branched streptococci, showing club
formation, were reproduced. They make an interesting counterpart to his sketches of clubbed
pseudo-diphtheria bacilli, reproduced as figure 34 on Plate J (from original fig. 3). How the
bacilli may pess over into ‘streptococci,’ is also well discernible in these early drawings. LEsche-
rich (1894 p. 190-225) was the first to show that the diphtheroid organisms are not less pleo-
morphic than the true diphtheria bacilli. When Seitz (1896) isolated several streptococci from
the mouth, he also got one strain which was considered to be an avirulent diphtheroid.
Tts colonies were entirely like those of the streptococci, its cells showed globular, triangular,
rodlike, club-shaped, and many other irregular forms. The staining with Gram’s method gave
positive, as well as negative, results. Early observations upon branching of pseudo-diphtheria
bacilli have been recorded by Kanthack (1896). A ‘‘Bacterium diphtheroides,” isolated by
E. Klein (1900) from the pus of a diseased udder, grew in serum mostly in globular or ovoid
form, while only a few wedge-shaped rods indicated its true character. Nakanishi (1900 c and
d) made a diphtheroid from the skin the object of very careful studies, because for some time he
believed to have discovered the causative agent of variola. It started its growth as triangular
coccus, later stretched out into short and long rods, which became beaded and eventually
turned over into club-,dumb-bell-shaped or branched forms. These sometimes remained con-
stant in successive transfers. In the animal test bright globular or oval bodies appeared,
which in artificial substrates assumed again the bacillary form. Those shown in figure 98 on
Plate [X (reproduced from original fig. 8) invite a comparison with the coccoid growth of diph-
theria and glanders bacilli (figs. 96 and 91).

When discussing the question of virulence among pseudo-diphtheria bacilli Hamilton
(1904) admitted that usually the differentiation from the true diphtheria bacillus is not too
difficult. But he also noticed that some virulent pseudo-diphtheria bacilli may become so
similar to true diphtheria bacilli, that they can only be distinguished by animal experiments.
Clark (1910), on the other hand, is of the opinion that so-called virulent pseudo-diphtheria
bacilli are true diphtheria bacilli of atypical morphology. Bac. Hofmanni, however, is declared
to be a different species, not merely a variety of the Klebs-Loeffler bacillus, though, according
to the author, both ‘‘doubtless” came from ‘‘common ancestors.” Kolmer (1912) considers
the Hofmann bacillus to be a “‘mutation”’ of the diphtheria bacillus, connected with it by many
intermediate forms. Wesbrook’s types were studied anew, and the sugar test accepted as of
practical value, ‘‘when used in conjunction with tests for virulence.’’ Morse (1912) added to
B. Hofmanmi and zerosis as two new types of diphtheroids the pink growing ‘‘B. hoagii”’ and
the yellow ‘B. favidus.”” As mentioned above, Bernhardt and Paneth (1914) concluded from
their experiments, that true and pseudo-diphtheria bacilli are strains of one species. Likewise
Trautmann and Gaethgens (1913) report to have succeeded in changing experimentally avirulent
diphtheroids to fairly virulent typical diphtheria bacilli.

From cases of Hodgkin’s disease Lanford (1914) isolated a pink, highly pleomorphous
organism of doubtful etiological significance, which assumed various shapes and sizes, from a
cocco-bacillus, 1-2u, to a large raquet-shaped bacillus, 8-10y in length. In older cultures rods
of about the size of B. Hofmanni were predominant. Occasionally relatively enormous oval
forms became visible. The most complete work upon the morphology and the life history of
various Corynebacteria, especially of several strains causing malignous granulom (Hodgkin’s
disease), which has been done so far, is that of E. de Negri and Mieremet (1913) and E. de Negri
(1916). Single-cell cultures were used for these investigations which, as was already pointed
out in our first paper (Léhnis and Smith, 1916 a, p. 696), represent a very welcome confirmation
No.2.) | LIFE CYCLES OF THE BACTERIA—LOHNIS. ; 81

and extension of our own observations upon the life cycles of the bacteria. Some of Negri’s
photographs, published in her second paper, are reproduced as figures 99-101 on Plate IX
(from original figs. 11, 12, 30, and 73), showing granular forms from whey agar, threads from
serum, round cells and thin rods from serum, together with a yeast-like cell from granulom
(magnification in all these cases X 560). The pathogenicity of the last-named cell form is
especially noteworthy. That Negri erred in placing her organism among the Blastomycetes
has been discussed before (Léhnis and Smith 1916, a, p. 697). It is not improbable that at
least some of the so-called bottle bacilli from the skin, which are of a rather uncertain sys-
tematic standing (see p. 22), are analogous types of growth of common diphtheroids

The close relations existing between this group and the streptococci have been made the
object of more thorough studies by Walker and Adkinson (1917) and by Mellon (1917). The
organism, isolated by the first-named authors from sputum of patients with bronchial asthma,
produced different types of colonies on plain and on dextrose agar. Some of them were very
small, opalescent, looking like tiny dew drops; they contained streptococcus-like forms, Gram-
positive, as well as Gram-negative, within the same chain. Under certain conditions they
passed over into diphtheroid bacilli, while in other cases they remained fairly constant. Fre-
quently a chain was made up of Gram-positive and Gram-negative cocci, oval forms, and true
rod-shaped bacilli. The authors are convinced that ‘‘many of the streptococci found in sputum
may be really diphtheroid organisms rather than true streptococci, and this may even apply to
the hemolytic streptococci.” Another type of colony was dirty white, raised in the center,
and with undulate margin. It contained chiefly Gram-positive bacilli, but also Gram-negative
bacilli, cocci and diplococci. Still another type was large, round, flat, thin, whitish and very
tenaceous. In it usually Gram-positive cocci predominated, some of which were very large.
Mellon's strain was isolated from the lung (pulmonary fibrosis). "When grown in broth con-
taining a few drops of human or rabbit serum, the bacilli of the barred type disintegrated, and
their chromatine masses became diplococci, as shown in figure 35 on Plate J (reproduced
from original fig. 2, p. 88), which later grouped themselves either as staphylo- or as strep-
tococci. It was difficult to revert these cocci into the original barred rods, but this was done
successfully on slightly acid, as well as on comparatively strongly alkaline, substrates. In the
animal this organism also acted like a streptococcus of low virulence. Mellon mentions espe-
cially the relations connecting his strain with Thiercelin’s Enterococcus. A glance over the
foregoing pages, as well as over the references given on pages 46-48 in regard to the diphtheroid
growth of various streptococci, will suffice to demonstrate in addition, that analogous relations
between numerous members of both groups are indicated in many of the earlier publications.
In the third part of Mellon's paper the name B. enzymicus was proposed for the subgroup related
to the streptococci; it was found to be present everywhere in the body. Other diphtheroids
displayed some proteus-like characteristics; they were motile and liquefied gelatine. Eberson
(1918) believes that the diphtheroids should be separated into not less than nine groups, all of
them having the character of true species. With regard to B. enzymicus it is assumed that the
presence of coccoid forms is either due to a contamination, or they are, indeed, produced by the
diphtheroid organism; in this case, however, the author does not want to have them classed as
“true” cocci.

B. leprae was introduced into the literature by G. A. Hansen (1880-1882) and A.
Newsser (1881) as a slender motile bacillus, forming round “spores.” Babes (1883) saw fre-
quently round swellings at the end of the rods, and in the tissue between the rods small round
elements (0.5-1.54) in small groups or chains, which stained the same way as therods. Another
early but little known contribution was made by Lutz (1886), who noticed that after careful
decolorization (with alcohol +nitric acid) of a strongly stained preparation (Gram’s method was
especially useful for this purpose) all bacilli presented themselves as chains of well stained cocci
within weakly stained sheaths. Really homogeneous rods were never seen by him. Some-
times at the end of the rods, or separated from them, other larger round bodies became visible,
either very darkly stained or not at all, or reacting like metachromatic granules. That some
82 - MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

of these were reproductive cells was discussed by the author and will be considered in
Chapter II, while other still more interesting observations are to be reserved for Chapter III.
Some of Lutz’s drawings are reproduced as figure 36 on Plate J (from original figs. 7-11 on
p. 81). He proposed to place the leprosy organism in a special genus Coccothrix, which
also had to embrace the tubercle bacillus and related forms. Bordoni- Uffreduzzi (1888 a)
noticed that homogeneous or granulated rods became visible in his preparations, according to
the mode of staining employed. In addition to the round swellings at one end of the rods,
he met also dumb-bell-shaped cells.

In 1897 E. Levy described for the first time an only slightly acid-resistant Actinomyces-like
organism, related to B. tuberculosis, which was isolated from a case of leprosy. More details
upon this organism were given by Czaplewski (1898). Young cultures contained homogeneous
acid fast rods; in older material the granules only were stained within the decolorized bacilli.
In broth thick round bodies were formed in the middle or at the end of the rods; clubs and
branches became also visible, the latter especially in gelatin. The etiological significance of the
organism was left in doubt. Similar diphtheroid strains were secured by Teich (1899) in five
different cases, which grew either as thin rods, like in the tissue, or displayed the wide pleomor-
phism characteristic for this group.

Barranikow (1900-1902) was the first to state that the leprosy organism is characterized
by a ‘‘complicated life cycle.” Form, staining reaction, acid-resistance, and virulence were
found to be different in the various phases of life. Washing with water was sometimes suffi-
cient to take away all stain. Kedrowskt (1901) confirmed that leprosy like other so-called
acid-fast bacilli are by no means constantly acid-resistant, and he also confirmed the wide
pleomorphism of this species. Cocci, rods of different shape and size, clubs and sometimes
typically Actinomyces-like growth were observed. Young bacilli proved to be motile. The
author: states that like plague, glanders, diphtheria, and tubercle bacilli so also the leprosy
organism passes through a complicated life cycle outside of the body, and that the branched
forms are one of the stages of growth entered at this time. Kedrowski continued his investi-
gations for a long time, and in a second paper (1910) he was able to furnish much, additional
proof concerning the pleomorphism, but also concerning the etiological significance of* his
organism. Figure 37 on Plate J (a reproduction of original fig. 37) demonstrates clearly the
‘“‘Coccothrix-,’”’ as well as the Actinomyces-like, habit of growth.

The discoverer of the leprosy bacillus, G. A. Hansen (1903), however, wasnot inclined to accept
any of these cultures as genuine leprosy organisms. He could not turn his mind from the striot
monomorphism, in vogue at the time when he made his discovery (1879). Changes in form,
acid-resistance, and especially the occurrence of branching were to him convincing proof of
untrustworthy investigations. This anachronism was rectified a few years later by another
paper on the same subject, published in the same ‘‘Handbuch,”’ by Babes (1907 a).

The results of extensive investigations carried out in British India, published by 7.8.
Williams (1911), are in complete agreement with Kedrowski’s work. The ‘‘Streptothrix lep-
roides,” as it is called here, grows as bacillus or Streptothrix (i. e. Actinomyces) of changing
acid-resistance. The excellent pictures accompanying the paper show also coccoid forms.
In his ‘‘Etudes sur les Actinomycates”’ Galli-Valerio (1912) gave some interesting drawings of
clubbed and spindle-shaped cells of Mycobact. leprae, as taken from the tissue. Making use of
single-cell cultures, Reenstjerna (1912) proved definitely that a culture of the leprosy organism,
whose pathogenicity was established by experiments on monkeys, was able to grow temporarily
as a not acid-resistant streptococcus. Kritschewsky and Birger (1912) made some serum tests,
which indicated that Kedrowski’s culture was the true B. leprae, while another culture from
Duwal did not react with these sera. A very thorough discussion of the whole subject is con-
tained in two papers by Wolbach and Honeij (1914), who are of the opinion that in the different
parts of the world apparently different varieties are acting as causative agents: (a) the diph-
theroid, (6) the acid-fast pigmented, and (c) the acid-fast not pigmented groups. Their own
culture once more exhibited an ‘‘extreme pleomorphism.” In young cultures short coccoid,
not acid-fast forms in pairs were prevalent, but acid-fast granules and segments were also al-
No.2] ~ LIFE CYCLES OF THE BACTERIA—LOHNIS. 83

ways visible. In addition, not acid-fast slender rods with acid-fast granules (i. e. Lutz’s ‘‘Coc-
cothrix’’) and club-shaped cells, whose swollen ends were usually acid-fast, were present. Inoc-
ulation experiments and agglutination tests were considered not to be conclusive.

On the other side, the standpoint taken by G. A. Hansen was defended by Duval and his
collaborators. In 1912 Duval and Wellman described a nonchromogenic, constantly acid-fast
rod, morphologically resembling the diphtheria bacillus, which, grew only on specially prepared
substrates, and whose etiological significance seemed to be probable. The pleomorphism ob-
served by other authors was contested. In 1913 Duval and Harris pointed out anew that the
acid-fast character of their culture was ‘‘as constant as that of B. tuberculosis” (which they
apparently assume to be constant) and that no streptothrical and filamentous stages became
visible; only diphtheria-like ‘‘involution’’ forms were recorded. In 1914 Duval extended and
confirmed these statements: All his cultures grew constantly as acid-fast rods. Those, how-
ever, which did not adhere to this ‘‘legitimate” course, were discarded as contaminations,
though no proof was available to support this assumption (see the quotation given on p. 20).
In addition, in 1915 Harris and Wade did not only demonstrate that diphtheroids are to be
found everywhere on and in the body, in air, water, etc., and that those isolated by them were
again entirely constant, they also did not hesitate to take these findings as a basis for the asser-
tion that the opposite results, obtained by all other authors, were simply to be explained as
caused. by impure cultures. A fitting remark, made by Bayon (1915) in regard to such an as-
sumption, was quoted on page 21. More recently Duval’s standpoint has been made quite
untenable by the work done by Johnston (1917), who declares himself wholly in accord with
Kedrowski and Bayon. He states that the leprosy organism has two stages in its life history:
a streptothrical not acid-fast form, and a bacillary acid-fast one.

_. The pleomorphism of B. tuberculosis was emphasized as early as in 1883 in two publi-
cations by Klebs. While R. Koch (1882) strongly insisted upon the constancy and the
monomorphism of his bacillus, and the presence of these bacilli was declared by him to be the only
valid proof of true tuberculosis, Klebs pointed out, that always, even in Koch’s own cultures,
coccoid forms were regularly to be found, some of which seemed to be motile. The bacilli are,’
according to his opinion, one type of growth of the tubercle organism, and their presence is
declared to be undoubtedly of great diagnostic value. As causative agent, however, the cocci
are considered to be of the same, or even of higher, importance. Observations published by
Babes (1883) at the same time fully confirmed this view. Not infrequently no bacilli could
be found in tuberculous tissue, but ‘‘grains bien colorés, ronds ou cubiques, se rapprochant de
Papparence des sarcines”. These round bodies were, as in leprosy, 0.5-1.5u in diameter; they
were stained in the same manner as the rods. The bacilli were also seen to contain sometimes
round inflations in the middle or at the end. In the same year (1883, p. 67) Zopf emphasized
that, despite Koch’s statement, the tubercle bacillus may assume the form of a coccus. The
critique directed by Fliigge (1884) against Zopf remains an interesting document of the manner
in which the monomorphistic dogma was established and defended at that time. But at the
same time Biedert and Sigel (1884) emphasized once more that in chronic cases the regular
bacilli may be entirely replaced by granular rods and accumulations of ‘‘granules”’ not stained
by aqueous aniline dyes, but stained by a modified Gram method.

In 1883 and 1884 Malassez and Vignal also published their important papers on “tuber-
culose zoogléique, forme ou espéce de tuberculose sans bacilles.”” Like Klebs and Babes, they
often found during several transfers in the animal test only ‘masses zoogléiques de micro-
coques’’; then the bacilli appeared, and later they disappeared again. In their first paper
the authors state clearly: ‘Tl existe une parenté entre nos zoogloées et les bacilles de Koch.”
Tn the second paper, however, they leave this point open to further research. Their drawings
show, that the micrococci, which they observed, were sometimes considerably larger than the
round bodies inside of the bacilli. Several authors have later assumed that this ‘‘tuberculose
zoogléique”’ of Malassez and Vignal had been some kind of pseudo-tuberculosis, an assumption
which is evidently not in accordance with the facts reported by the French bacteriologists, and
which is likewise contradicted by the more recent observations of Spengler, Much and others
84 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

to be mentioned below, which fully confirm those older discoveries. After Lutz (1886) had
noticed that the tubercle bacilli, when treated in the same manner as he did with the leprosy
bacilli, also assumed the appearance of a “‘Coccothrix,”’ Klebs (1887) modified somewhat. his
earlier standpoint; he now thought that the coccoid forms which he had seen, should be also
declared as ‘“Coccothrix.” This explanation, however, is not in full agreement with those
observations which revealed the presence of single, comparatively large, globular, acid-fast
bodies in tuberculous tissue. Fig. 102 on Plate IX is a reproduction of a sketch published by
Cornil and Babes (1890, Vol. II, p. 402, fig. 342) of a cut through the pia mater in tuberculous
meningitis, showing such “grains ronds qui se colorent par le méme procédé que les batonnets.”

That the tubercle bacilli may pass over into branched forms was first ascertained by
Metchnikoff (1888a) for the avian type. Cultures kept at 44° C. exhibited a great pleomorph-
ism, as shown in figure 38 on Plate J (reproduced from original fig. 20 on Pl. V). That the
branching is the result of budding is clearly visible in the picture, as is also the occasional separa-
tion of oval side branches (s and n—n’), i. e., the same process which was described later by Hill
in regard to the diphtheria bacilli. It was emphasized by Metchnikoff that the branched forms
should not be classed as “involution”? forms. More data upon branching also of the human
and bovine types of the tubercle bacillus have been contributed during the next years by
E. Klein (1890-1894), Mafucci (1892), Fr. Fischel (1893), Coppen-Jones (1893-1895), Bruns
(1895), Semmer (1895), Kanthack (1896), Olsen (1897), and Marpmann (1897). Most of these
authors reached the conclusion that the bacilli are merely a type of growth of an organism
which should be classed among the Actinomycetes or even among the higher fungi. (See p. 21.)
Club-shaped cells were very frequently found by Coppen-Jones in sputis from far advanced
cases. Babes and Levaditi (1897) were the first to produce experimentally typical Actinomyces-
like clusters (‘‘Drusen”’) with clubs in the animal tissue. Their results were soon confirmed by
Schulze (1899) and by Lubarsch (1899).

Ferrén (1897a), on the other hand, reported that by successive adaptation of the tubercle
bacilli to ordinary broth (without glycerine) he had been able to reduce their acid-fastness and
‘to induce them to grow at 20° or even at 10° C. as motile coli-like rods, which, however, were
still able to produce typical tuberculosis in guinea pigs, simultaneously reverting to the original
type of growth. Temporary motility in young homogeneous cultures was also recorded by
Schumowski (1898-1899), by Courmont and Descos (1902) and by Hawthorn (1903a).

Schiirmayer (1898) encountered small and large nonacid-fast globular forms of the avian
type, which reproduced typical rods. Some pictures presenting Streptococcus-like, but still
acid-fast growth in scrophulous glands, which were published by Arrigo (1900), are reproduced
as figures 103 and 104 on Plate IX (from original figs. 1a and 26), where they make a sharp
contrast to the richly ramified thread in figure 105, reproduced from a photograph made by
Migula (1900, Vol. II, fig. 3, Pl. V), who had ‘‘no doubt” (though, of course, also no proof)
that such abundant growth is “merely degenerate.’ Dorset (1901a) found another good
example of budding and branching of the human type in a 6-weeks-old glycerine broth culture.
The bovine type, when cultivated in the frog by Herzog (1903), changed its morphology to such
an extent that, as the author says, it seemed to be very doubtful at first whether the forms
observed were really merely transformations of the organism tested. -Wolbach and Ernst
(1903), too, found the morphological variability of the tubercle bacillus under different con-
ditions and especially in rapidly growing cultures “most extensive,” leading from small cocci
up to long threads and large branched forms. .

The greatest deviation from the ‘‘legitimate” form, which had been declared by R. Koch
and his followers to be the decisive indication of true tuberculosis, was probably that described
by Schroen (1904). According to this author the tubercle bacillus passes through three different
developmental cycles whose complete description, though promised in this paper, apparently
has not been published. The causative agent of phthisis is considered by Schroen to be an
organism of peculiar standing and not to be connected with the tubercle bacillus, which is
accepted as only causing tuberculosis, not phthisis. Description and pictures of this “new
phthisiogenic microbe,” as it is called, leave, however, not much doubt that they actually refer
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 85

to a type of growth of the tubercle bacillus, analogous to that produced by B. radicicola within
the root nodules, but never on artificial substrates. As this will be discussed in Chapter ITI,
the discoveries of Schroen will also better find their place there.

In a paper entitled ‘‘ Polymorphisme du bacille de Koch dans les produits de l’expectoration
phthisiques” Piery and Mandoul (1904) established the following four morphological types:
“(1) les homogénes courts, (2) les homogénes longs, (3) les moniliformes courts, (4) les monili-
formes longs.’ Like Lutz, these authors are of the opinion that the tubercle bacilli are made
up of chains of well stainable granules, which are surrounded by sheaths of variable stain-
ability. According to Péju and Rajat (1907) successive cultivation of the tubercle organism
in broth containing up to 4 per cent potassium iodide proved to be very efficient in stimulating
lateral budding and multiple branching, as well as the production of other aberrant nonacid-
fast forms. Large inflations of 6-8 u diameter and typical mycelial growth were also observed.
Analogous results were secured by Arloing (1908) by keeping the cultures at 44.5-45.5° C,
and under increased atmospheric pressure (2.5 atmospheres). Miehe (1908) made some obser-
vations upon branching of the tubercle bacillus in the hanging drop. The globular and ovoid
bodies were again accepted by Betegh (1908) as the initial stage in the development of the tubercle
bacillus. They were seen by him to produce very thin, straight, nonacid-fast rods, which in-
crease in size, become acid-fast, and assume usually a somewhat curved shape. These rods,
first homogeneous, later become granular, and finally disintegrate again into these round bodies.
Much (1907-1909) made some new attempts to convince the German bacteriologists, that
Koch’s monomorphistic dogma was indeed not in accordance with the facts. Once more the
infectious nature of the coccoid type of growth and the occurrence of tuberculosis without the
“legitimate” tubercle bacilli were clearly demonstrated. The passing over of acid-fast into
not acid-fast forms and the opposite process were also closely studied.

Probably thus far the most complete investigation upon the life cycle of B. tuberculosis
has been made by Maher (1910-1915). The transformation of the typical rods into granular
forms, the loss of acid-resistance, the passing over into filamentous, coccoid, and yeastlike
cells, and the regeneration of not acid-fast, as well as of acid-fast, rods were carefully followed.
Of special interest, however, are the discoveries made by this author in regard to the symplastic
stage of the tubercle organism, which will be discussed in Chapter III. The coccoid, as well
as the yeastlike, cells could again be cultivated as such, in the same manner as it was done,
for instance, with the Corynebacteria by E. de Negri. Unfortunately, only rather short reports
have been published by Maher; and several hypotheses incorporated in his papers (assuming,
e.g., the tubercle bacilli to be the offspring of the acid-fast spores of B. subtilis, and meningo-
cocci, as well as pneumococci, to be the progeny of the tubercle bacilli) evidently have had the
effect that his discoveries did not elicit adequate attention.

Much’s work has been confirmed and extended by KArylow (1911), who noticed that in
vitro, as well as in vivo, very young tubercle bacilli were not acid-fast and also Gram-negative,
but could be easily stained with aqueous aniline dyes. The Gram-positive substance, i. e., the
granular mass, is the next to be developed, while the acid-fast material appears later within
the membrane. The pleomorphism again was found to be very conspicuous already in young
cultures. Analogous results were recorded by Klepzoff (1912), who points out that these changes
seem to indicate relations connecting the B. tuberculosis and the Pasteurella groups. From
the latter he was able to get acid-fast growth, too. Several interesting drawings. of branched
threads and of clubbed forms of Mycobact. tuberculosis have been published by Galli- Valerio
(1912); and more details concerning the variation in acid-resistance and cell morphology may
be found in a paper by Wherry (1913 a) upon the ‘‘Chemical Conditions Influencing Acid-Proof-
ness and Non-Acid-Proofness in a Saprophytic Culture of B. tuberculosis.” Minute, as well as
large, coccoid bodies, short and long, thick and thin, straight and curved rods, which were or
were not acid-resistant, were all seen to occur in an old culture originating from R. Koch’s
laboratory.

The pleomorphism of the pseudo-tubercle bacilli is very similar to that of the genuine
tubercle bacillus. Preisz (1894) published some interesting pictures of a strain isolated from
86 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

sheep, reproduced as figures 106 and 107 on Plate IX (from original figures III-V on Plate IX,
all x 1,000), showing their diphtheroid growth, as well as a Streptococcus-like development,
which practically duplicates that found by Arrigo with B. tuberculosis (fig. 103, Pl. TX).
The pseudo-tubercle bacillus isolated by Korn (1899), from butter, showed in the cultures, as well
as in the animal test, traits characteristic of the tubercle bacillus on the one side, of B. coli on the
other. The ‘‘Coccothrix” form dominated in old agar and serum cultures. Branched, club-
shaped and spermatozoid cells were also seen. Motility in young cultures was noticed by
Moéller (1899) with his ‘‘Grass bacillus II.” E. Klein (1899) did not see active motility, but
also his culture showed two polar flagella when stained with Ermengem’s method. Typically
actinoymyceslike clusters (‘‘Drusen’’) in the animal test were obtained by Lubarsch (1899)
with pseudo-tubercle as with genuine tubercle bacilli. The organism described by Bongert
(1901) under the name Corynethrix pseudo-tuberculosis murium exhibited an analogous behav-
ior, and the author points out that this club formation is a normal mode of growth, not a sign
of degeneration. Moéller’s grass and hay bacilli were also seen by Abbott and Gildersleeve (1902) to
grow in the tissues very much like Actinomyces and Bac. tuberculosis. All pseudo-tubercle
bacilli tested by Courmont and Descos (1902) proved to be actively motile, like B. tuberculosis,
in young liquid, homogeneous cultures. The partially acid-proof ‘‘Streptothrix,”’ isolated by:
Schabad (1904) from a case of pseudo-tuberculosis, produced neither clubs, nor clusters,
but it showed very clearly true branching. A close relationship to the true tubercle organism
was noticed by Sanfelice (1905) when he studied another ‘‘Streptothrix,”’ causing pseudo-
tuberculosis.

Mycobacterium hyalinum and related species, whose peculiar physiological activity as
benzine, petrol and paraffin destroyers has been investigated by Séhngen (1913), appeared in
many respects very much like Actinomyces. On the other hand, however, they also resembled
B. radicicola, growing in the same type of colonies and in the form of Y-shaped rods.

A fairly complete description of the morphology of Actinomyces in man was published by
J. Israél as early as in 1878 in.a paper discussing for the first time human actinomycosis. He
enumerates as elements building up the clusters found in the pus: (1) thin threads; (2) granules
of various size, some small like typical cocci, others larger, very bright, globular or oval; (8)
peculiar pear- or club-shaped bodies of very variable size and form, apparently dividing longi-
tudinally. Sometimes the granules alone were present in the tissues. All intermediate stages
connecting threads and clubs were seen. Formation of buds was noticed, too. One of
J. Israél’s drawings (orig. fig. 4 on Pl. II) has been reproduced as figure 108 on Plate IX.
The swellings visible therein are of special interest, because very similar formations have
been later seen by Metchnikoff’ (1888 b) with his Pasteuria ramosa. When O. Israél (1884)
first succeeded in cultivating Actinomyces hominis on serum, he also obtained the characteristic
club formation in vitro for the first time. The causative agent of the ‘‘farcin de boeuf’ was
named by its discoverer Nocard (1888) ‘‘bacille de farcin;” but already this first description
leaves no doubt that this organism is a true Actinomyces. A special paper upon ‘“‘The Morpho-
logy of Actinomyces,” by McFadyean (1889), adds to the cocci, rods, threads and clubs already
known, large circular bodies, which are believed to be related to the clubs. The latter were
found to react differently when treated with Gram’s method; and the cocci, too, were seen to be
able to grow differently, viz., they either multiplied as such (in very young colonies) simulating
strepto- and staphylococci, or they stretched themselves to rods and threads (in older colonies).

The very complete studies by Bostroem (1890) upon human actinomycosis brought much
confirmation especially to the first findings of J. Israél. The clubs were seen to divide longi-
tudinally as well as transversely, within the threads cocci, short and long rods were observed,
and with young rods motility was ascertained. The pathogenic organism, isolated by Eppinger
(1890) from a case of pseudo-tuberculosis and named by its discoverer Cladothriz astercides,
undoubtedly is also a true Actinomyces, though no clubs became visible. Eppinger’s erroneous
view concerning the mode of branching has been mentioned on p. 21. Coccoid as well
as rod-like cells were found to be actively motile in young broth cultures. In old cultures
the rods produced within the threads were seen to slip out, leaving the empty sheath behind.
The thermophilic ‘Cladothrix,”’ met by Kedzior (1896) in water, showed also true branching
No. 2.) LIFE CYCLES OF THE BACTERIA—LOHNIS. 87

and the characteristic powdery layer, so frequently produced by the Actinomycetes on solid
substrates. Coccoid forms and young rods again exhibited motility. The same behavior
was observed by Rullmann (1896, 1898 a) with his Cladothriz odorifera (i. e. Actinomyces
odorifer), as well as with a yellow ‘‘Streptothrix,” isolated from saliva. Another Actinomyces
obtained by Schtirmayer (1900) from tuberculous tissue, and described by him as “Oospora
(Streptothrix) proteus’? on account of its pleomorphism, grew temporarily in the form of
motile cocci or motile rods. The cocci in this case again were able to multiply as such for
some time, when kept at a low temperature. They then appeared like streptococci, but the
tendency to produce branched compounds was much more pronounced than with genuine
streptococci.

Feistmantel (1902) studied more closely the variable acid-fastness of ‘‘ Streptothrix farcinica”’
and of other Actinomycetes. His results are very similar to those quoted above for
B. tuberculosis. Full descriptions of the life history of Eppinger’s organism, now correctly
called Actinomyces asteroides, were published by MacCallum (1902) and by Nakayama (1906).
According to Doepke (1902) different species may cause actinomycosis in man, some of them
are aerob, some anaerob, clubs may be furmed or may be absent, the coccoid forms may grow
as such, or they may revert to rods and threads. Much variation in the formation of clubs
and in their staining reaction has been also recorded by Hollandt (1906) and by Loele (1908).
Gram staining and acid-fastness may even vary with the different parts of branched clubs.
Occasionally threads were found by the first-named author up to 6y broad, producing different
kinds of gonidia, which fact was accepted as indicating a closer relationship between Actino-
myces and Crenothrix. A detailed description of an Actinomyces variabilis sp. n. was given
by Th. Cohn (1913). With the pleomorphism of the cells the formation of different colonies,
was also studied in this case. Comparative tests of numerous Actinomycetes, contributing
some additional] details in regard to their pleomorphous character, have also been made by
Glaypole (1913) and by Krainsky (1914).

Among the so-called thread- or trichobacteria the old genus Leptothrix has been frequently
mentioned as a prominent example of bacterial pleomorphism. However, the early observa-
tions of Robin (1871) and his contemporaries undoubtedly can not be accepted as sufficient
proof in this respect, and several so-called Leptothrix species probably have been nothing else
than a thread-like growth of other bacilli, principally of lactobacilli and of aerobic, as well as
of anaerobic, spore-forming bacilli. Leptothrix gigantea, as described by Miller (1883), appears
very similar to Crenothrix. The very different size of the threads, the slipping out of coccoid
and of rod-like cells from the sheath, and their growing up, are the same in both cases. Zopf
(1883) added to cocci, rods and threads, spiral forms as another type of growth of Lepto-
thrix. With Leptothrix buccalis he ascertained later (1885) that the coccoid forms again may
multiply as such for some time, before returning to the thread-like growth. The Leptothrix
cultivated by Dobrzyniecki (1897) from the mouth, gave colonies similar to those of B. anthracis.
One of the photographs accompanying the paper seems also to show some branching. Vin-
centini (1893), too, studied a so-called Leptothrix racemosa, which, however, has been clearly
a complex mixture of different organisms. Beust (1907) thought that Bacillus buccalis mazi-
mus, Leptothriz maxima, Lept. innominata and Lept. racemosa should be considered to represent
merely stages in the life history of one organism.

Oladothriz dichotoma is characterized, according to F. Cohn (1875, p. 185), by its tendency
to exhibit “false’’ branching only. But repeatedly the genus name has been erroneously used
for naming species which belonged to the genus Actinomyces. Zopf (1881-1882) described
as different types of growth of Cladothrix dichotoma: Cocci, rods, spirilla and threads; but his
findings, though based on continuous miscroscopic observations, were strongly contested by
Winogradsky (1888, p. 111). The results recorded by Billet (1890), however, are again in good
agreement with those of Zopf, while Biisgen (1894) took his stand on Winogradsky’s side.
Spiral forms were not seen by him; small and large globular cells were discarded as “merely
involution forms;” rods and threads alone were accepted as normal types of growth. Migula
(1900, p. 39) published an interesting drawing (original fig. 45), which is reproduced as figure
39 on Plate K, presenting Cladothrix (a-f), as well as Actinomyces (g-7). Form and flagella-
88 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

tion of the motile cells are like those of Azotobucter agile, as seen by Beijerinck (1901 6).
But while the picture given of “‘false’’ branching (a) is evidently a duplicate of that made by
Cohn, the other figure (¢), showing branched material grown on agar, seems to indicate, as far
as can be seen from it, that also true branching may occur with this organism. The position
of many of the branches, as sketched, is. obviously not compatible with the assumption of
“false” branching. If the growth which Migula had on his agar has really been that of Clado-
thrix it would demonstrate that under such conditions the resemblance between Cladothrix
and Actinomyces (e and f) is indeed much greater than is generally believed. To Migula
not only the cocci, but also the rods in old agar cultures are ‘‘merely involution forms.” The
frequently contested observations by Zopf, including those relating to the liberation of spirals
with polar flagella from the Cladothrix threads, have been more recently confirmed by Ellis
(1912) in his ‘Investigations into the Life History of Cladothrix dichotoma (Cohn).”’ He comes
to the conclusion that Cladothrix is related on the one side to Crenothrix and Clonothrix, on
the other to Pseudomonas and Spirillum.

Cocci, rods, threads, and spiral forms were also found by Zopf (1881-1883) with Beggiatoa.
All short forms exhibited temporary motility. The coccoid cells either propagated as such,
or they reverted into rods. Engler (1882) confirmed these results, but they were again utterly
discredited by Winogradsky (1888). That, however, Winogradsky’s statements can not be
accepted as well founded, has been discussed (on p. 38). The answer given by Zopf
(1895) deserves full attention, especially because new results are recorded therein, which confirm
those obtained before. The findings of Ballet (1890) were also in complete agreement with those
of Zopf.

Crenothrix, too, can form, according to Zopf (1879), cocci, rods, and threads, and the cocci
may either multiply as such or grow up to threads. Crenothrix is said to be unable to produce
branches, Schorler (1904), however, who otherwise confirmed Zopf’s results, also encountered
a very similar, but branching, organism, which was made the type of the new genus Clono-
thrix. Only future experiments with pure cultures will make it possible to decide, whether this
separation has been correct, or whether branching is equally common within this group as itis
with all other bacteria.

3, CONCLUSIONS.

In regard to the occurrence of different cell forms within the various groups of bacteria
the following conclusions can be drawn from the facts discussed on pp. 7-88:

(1) The monomorphistic theory of F. Cohn, R. Koch, and their followers, still accepted by
many Dacteriologists as being correct, is irreconcilable with numerous facts, which prove
conclusively that all bacteria are pleomorphic. All well studied species—like Micr. candicans,
gonorrhoeae, and meningitidis; Strept. pyogenes, lanceolatus, and lactis; Bact. pestis, pneu-
moniae, radicicola, colt, acidophilum, bifidum, proteus, erysipelatos suum, pyocyaneum, and fluores-
cens; Bac. anthracis, subtilis, Azotobacter, Amylobacter, and Chauwvoei; V. cholerae, S pirochaeta
pallida; Corynebact. mallet and diphtheriae; Mycobact. leprae and tuberculosis—have shown
themselves to be able to grow in various, round, straight and curved, small and large, regular and
irregular, cell forms. Budding, branching, and apical growth are common with all bacteria.

(2) That it was possible to uphold the monomorphistic theory for a considerable length of
time, despite its erroneousness, is due to the following reasons:

(a) Differences in the cell form and in the appearance of the colonies have been usually
accepted as a priori evidence of difference in species; and, therefore, most of the cultures
exhibiting such differences, have been rejected as being contaminated.

(b) Genuine bacteria have often been laid aside as being no bacteria at all, because they
showed budding, branching, or other peculiarities, which, according to the monomorphistic
theory, should not occur with bacteria.

(c) In nearly all cases where undoubtedly pure cultures of true bacteria have exhibited
various cell forms, only one of them has been accepted by the monomorphistic theory as
‘‘typical”’ or “legitimate,” while the others were discarded, without adequate test, as being
‘‘atypical,” ‘‘degenerate,”’ or ‘‘involution forms.”
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 89

(dZ) The cultural methods, as developed by the founders of the monomorphistic doctrine,
make it often possible to keep a culture fairly constant for a considerable length of time,
and to secure a continuous development of only one type of cells. Natural variations, as
well as the different phases in the life cycles of the bacteria, are not easily noticeable under
these conditions; but wherever they have been observed, the characteristic changes in cell
morphology caused by them usually were again mistaken as growth of a ‘‘contaminating”’
organism or as some uninteresting ‘‘involution” form.

(3) In practically all cases where the so-called involution forms have been properly studied,
it has been found that it was a mistake to classify them as such. This has been fully proved
especially with Micr. gonorrhoeae and meningitidis; Strept. pyogenes and lanceolatus; Bact.
pestis, pneumoniae, radicicola, aceti, coli, typhi, proteus, and pyocyaneum; Bac. anthracis, mycoides,
Azotobacter, Chauvoei, and tetani; v. cholerae, Spirochaeta pallida; Corynebact. ‘mallei, and
diphtheriae, Mycobact. leprae, and tuberculosis. The endospores alone exhibit all marks of real
involution.,, Many of the so-called involution forms, however, are most frequent when the
bacterial development is at its height, part of them can be propagated as such, and later they
are again replaced by the smaller cells of simpler morphology. The latter forms are generally
best adapted to parasitic growth, though in the body, too, branching and otherwise ‘‘atypical’”’
development is by no means absent.

(4) How far the alterations in morphological as well as in physiological behavior are

due to variation and mutation among the bacteria can not be decided at present, because only
insufficient information, or none at all, is available concerning the regular changes ane
in the different phases in the life cycles of the bacteria.
(5) The life cycles of probably all bacteria are composed of several subcycles sound
wide morphological and. physiological differences. To secure all data pertaining to the life
history of a species is much more difficult than to describe a bacterium in the ordinary
mannér. Long-continued investigations, made under different conditions, and controlled by a
sufficient number of parallel tests, are indispensable in this case. Uniform cultural conditions
and frequent transfers, made in regular intervals, tend to keep one type of growth constant;
while successive transfers, made repeatedly during several weeks from and to various substrates,
will always reveal more or less completely the pleomorphic character of every bacteria species.
The best starting point for such experiments seems to be the symplastic stage, to be discussed
in Chapter Ii.

(6) It is a grave, though often repeated, logical mistake to believe that the acknowledg-
ment of the pleomorphism of the bacteria be equivalent to a negation of the existence and
constancy of bacterial species. Just as all studies upon the pleomorphism of the fungi, alge,
and protozoa did not minimize, but have increased the accuracy in distinguishing between the
various genera and species, so also systematic bacteriology will win considerably by every
thorough investigation upon the life cycles of the different species. It is to be expected, and
already clearly indicated by numerous well-confirmed results, that the old monomorphistic
form genera and form species of F. Cohn, R. Koch, and of their followers will have to be thor-
oughly revised; and many of the so-called species, often described in the most superficial man-
ner, will have to be canceled entirely as being merely fragments of the life cycles of other species.
Only those species can be accepted as natural ones whose life history is fairly well known, and a
correct'systematic arrangement of the bacteria is not possible except upon such a natural basis.

(7) As far as can be concluded from the data available at the present time, it is probable
that an increased knowledge of the life history of the bacteria will not lead to any change in
the general position of the bacteria within the system. Their budding, branching, and apical
growth may be considered as presenting additional evidence of their being related to the
fungi, but it would not be justified to incorporate all bacteria on account of these characters
among the fungi themselves, though this has been recommended by different authors. Other
facts, especially with regard to the various modes of reproduction, undoubtedly strengthen the
relations existing between bacteria and lower alge, as well as between bacteria and. protozoa.
Many parallelisms are noticeable in the life histories of all these organisms; all of them are
more or less pleomorphous.
I], REPRODUCTIVE ORGANS.

(GONIDIA. REGENERATIVE BODIES. SPORES. MICROCYSTS.)

1. GENERAL DISCUSSION.

In our first preliminary report (Léhnis and Smith, 1916 a, p. 700) we said concerning the
reproductive organs:

All bacteria multiply not only by fission but also by the formation of ‘(gonidia”; these usually become first regen-
erative bodies, or occasionally exospores. Sometimes the gonidia grow directly to full-sized cells.

The gonidia are either liberated by partial or complete dissolution of the cell wall, or they develop while still
united with their mother cell. In the latter case the cell wall either remains intact or it is pierced by the growing
gonidia, which become either buds or branches.

Some of the gonidia are filterable. They also produce new bacteria either directly or after having entered the
symplastic stage.

The transformation of spore-free into spore-forming bacteria seems to be dependent on the conditions acting upon
the symplasm and regenerative bodies.

The data collected on the foregoing pages will have shown that our observations concern-
ing the pleomorphism exhibited by all bacteria in their vegetative stage, though at variance
with the predominant, monomorphistic doctrine, are amply confirmed by results obtained by
many investigators, who in most cases have worked quite independently. In relation to the
different modes and organs of reproduction of the bacteria the situation is very similar. If
only the numerous findings to be quoted on the following pages would have met with adequate
consideration, and if they had been brought into proper correlation, the doctrine still adhered
to in most textbooks of bacteriology, that the endospores are the only special organs of repro-
duction of the bacteria, probably. would never have been established.

As was said on p. 7, it is not always possible to reach a definite decision as to the
vegetative or the reproductive character of certain cell forms described in the literature. There-
fore, it will be necessary to refer occasionally back to one or the other item mentioned in
Chapter I.

Before the special discussion of the character, the formation, and the aevalootiont of the
different reproductive organs is taken up, a brief historical resumé upon the discovery of the
various modes of bacterial reproduction may be given first. In the second place the differen-
tiation between reproductive organs and other cell products and artefacts must be considered.
A few remarks upon some special problems (filterable vira, cell inclusions, aphanozoa,
chlamydozoa, ultra-microorganisms, and heterogenesis), which become more easily accessible
from our newly-won standpoint, will be added at the end.

(a) THE DISCOVERY OF THE DIFFERENT MODES OF REPRODUCTION.

When E. Chr. Ehrenberg in 1838 described the ‘‘infusoria” as ‘complete organisms,’’ the
genus Monas was reported as being able to produce “‘eggs,’”’ which in the case of Monas Okenit
were seen to be liberated by the cells. With the small species of Monas, like Monas termo, as
well as with spirilla. and spirochaets, no analogous details could be made out. In the genus
Vibrio, however, such granules were observed, as was also the case with the comparatively
large Ophidomonas. In Gallionella a special organ, built up by four or more globoid bodies,
was discovered, which the author considered as some kind of ‘‘ ovarium.”

'Perty in 1852 was able to furnish much more information. Speaking upon protozoa, he
says (p. 66 and 67):

Ich glaube nach zahlreichen Untersuchungen die Uberzeugung hegen zu dirfen, dass eine gewisse Klasse von
Blaschen und Kérperchen, welche man in den Wimperinfusorien entstehen sieht, zur Fortpflanzung dienende, den
Sporen vergleichbare Keime seien, fiir welche der Name Blastien (ré 6ddoreov, Keim, Trieb) vorgeschlagen wird.

Die Fortpflanzung geschihe hiernach nicht durch gewéhnliche Substanzteilchen, wie Dujardin, noch Eier, wie
Ehrenberg annimmt, sondern durch eigentiimliche, im Innern der Tierchen entstehende, allmablich zahlreicher
werdende Kérperchen.

90
No. 2.] LIFE CYCLES OF THE BACTERIA—LOUNIS. 91

Concerning the Phytozoidia the following statement was made (p. 76):

Bei ihnen kommen auch die beiden Fortpflanzungsarten durch Teilung und durch Blastien vor; freilich ist man
manchmal zweifelhaft, ob man Blastien oder nur den durch fortgesetzte Teilung in zahlreiche Parzellen zerfallenen
Inhalt des miitterlichen Geschépfes vor sich hat.

When discussing the monads (p. 83) the blastia are identified with Ehrenberg’s ‘‘eggs’’;
the Vibrionida, too, are declared to grow up (p. 104):

. aus Anfiingen, welche verschwindend klein sind.

The great difficulties arising from the minute size of these blastia and of the newly developed
organisms are repeatedly pointed out, as for instance in the following sentence (p. 67):

Diese Kleinheit verhindert meist ihre Wahrnehmung, wenn sie zerstreut ausser dem Infusorienkérper vorkommen;

aus gleichem Grunde kann man auch die Unterschiede der Anfange vieler mikroskopischer Wesen nicht genau wahr-
nehmen.
. Perty has been credited with the discovery of tiie bacterial endospores, and his drawings of
“‘ Snoronema gracile,” reproduced as figure 40 on Plate K (from original fig. XV, 26), prove that
his observations have been very accurate in this respect. But evidently.it has heen overlooked
so far that he also gave admirable pictures of the formation and development of what we now
call gonidia. His drawings. of spirilla, reproduced as figure 42 on Plate L (from original figs.
XV, 28, 29B, and 31), deserve our full attention. The formation of the gonidia within the
organisms and the upgrowth of the normal spirilla from minute beginnings are illustrated as
clearly as in only very few pictures published in the modern literature. Concerning the large
globular bodies, visible at some of the spirilla, Perty says (p. 132):

Brachte man ein Trépfchen (of the culture) in einen Tropfen Brunnenwasser auf die Objectplatte, so bildete
alsbald ein Teil der Spirillen durch Gerinnung Kugeln an sich, mit denen sie sich anormal bewegten, manche noch
leicht, andere miihsam.

That is, a description of the so-called plasmoptysis has been given in this early work,
more accurate than much of what has been written more recently upon this subject.

About the small and large ‘‘spores,”’ i. e., micro- and macro-gonidia, of Gallionella ferru-
ginea Perty made the following remark (p. 215):

Die kleinsten Kérnchen zeigen oft Molecularbewegung; gréssere bewegen sich wie Monaden scheinbar willkiirlich.
Sollen sie zu Faden erwachsen, so bildet sich zuerst eine Hiille um sie, dann teilt sich der Inhalt und die Hiille ver-
langert sich.

Notwithstanding many inaccuracies to be found in Ch. Robin’s ‘‘Histoire des végétaux
parasites’ (1853), the following references concerning his Leptothriz buccalis seem worth being
quoted:

On peut quelquefois observer dans leur intérieur de trés petits granules ronds, placés de distance en distance

p. 347).

These round bodies were thought to slip out of the rods, to make part of the ‘‘granula-
tions moléculaires’’ floating i in the saliva, and eventually to develop to new threads (p. 351).

That Hallier (1865 a) and Liiders (1866) have dwelled upon the same subject has been also
of no avail, though Hallier was right when he reported that the ‘‘granules”’ not always repro-
duce new threads immediately, but may for some time multiply as such, or may develop into
larger round bodies, i. e. , Tegenerative bodies. His favorite hypothesis, however, misled him to
see ‘‘Penicillium spores’”’ in them, which, of course, were also present in his very impure cul-
tures, as were yeasts and all kinds of molds. To Hallier (1866-1896) all these various fungi
were the result of the upgrowth of the bacteria and their granules, first called by him ‘‘micro-
cocci’’ and later ‘‘plastids.”’

Karsten (1869) seems to have been the first author who used the term ‘‘micro-gonidia”’
for the small bodies, called blastia by Perty. Apparently he was influenced by the idea which
he shared with Hallier, that bacteria and their ‘‘seed’’ were the offspring of vegetative and
reproductive fungus cells; for the latter he used the term ‘‘Pilzgonidien.”

Burdon-Sanderson (1871) saw ‘‘spheroids’’ (probably spores) elongate to rods, but he
assumed also that bacteria might develop from ‘‘ultra-microscopic’’ bodies, not visible in
92 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor. XVI.

homogenous liquids. Oertel (1871) reports that in his diphtheria investigations he not ‘only
found cocci and rods, but also ‘‘Zellchen mit sprossendem Keimschlauch,” which, according
to his drawing, have been very similar to those observed by Rindfleisch (1872) and reproduced
as figure 41 on Plate K (from original fig. 1 on Pl. XVIII). That these cells were not germina-
ting spores:has been made clear by Rindfleisch’s statement that the ‘‘punctiform’’ bodies were
actively motile, probably by one flagellum. He observed also for the first time that micrococci
sometimes do not divide regularly, but increase in size, forming giant cocci, whose réle will
be discussed later.

When F’. Cohn (1870) gave the complete description of Crenothriz polyspora, the formation
of its motile ‘‘micro- and macro-gonidia”’ was also pictured. In figure 43 on Plate L two of his
drawings (original figs. 11 and 13) are reproduced, one showing ‘‘ein Sporangium in der
Mitte bandférmig verbreitert,’”’ the other a lateral club, filled with finely granulated protoplasm,
considered by Cohn to have a spore-like character. To the same author (Cohn, 1872 b) we also
owe the first fairly accurate investigation upon the bacterial spores, which was supplemented
(1876) by R. Koch’s study upon the spores of B. anthracis. As the bacterial gonidia are to many

modern authors ‘‘merely fat,” it is not uninteresting that Koch thought the spores were made
u ——

r . . aus einem stark lichtbrechendem Trépfchen, vielleicht einem 61, welches von einer diinnen Protoplasma-
schicht eingehiillt ist. Letztere ist die eigentliche entwicklungsfahige Zellsubstanz, wahrend ersteres.vielleicht einen
bei der Keimung zu verbrauchenden Reservestoff bildet.

Much more important, however, is the fact that the very first photographs ever made of
bacteria show in an. excellent manner gonidia and regenerative bodies, i. e., those reproductive
organs of the bacteria which, at the present time, are practically unknown to most bacteriologists.
One of the pictures made by R. Koch (1877, original fig. 4 on Pl. XV) has been reproduced
as figure 109 on Plate X. It shows, according to Koch (1. c., p. 423), ‘‘Bacillen mit mehreren
seitlichen Sporen,” and it is stated, in addition:

Die Sporen . . . sind dicker als der Bacterienkérper und treten kugelartig aus disse hervor.

The bacilli concerned are said to be related to. Bact. termo, i. e., Bact. fluorescens.| Figure 110
is a reproduction of one of our own photographs (1916 a, fig. 40) ‘showing formation of gonidia
and regenerative bodies by B. fluorescens. Koch's picture was made with 500-, ours with 1,000-
fold magnification; otherwise both are very much alike, and it is, indeed, surprising that this
old observation neither has been followed up by R. Koch or by one of his pupils, nor has met
with adequate attention elsewhere in the bacteriological literature. Another photograph
by R. Koch (1877, fig. 5 on Pl. XVI), reproduced as figure 111 on Plate X, demonstrates the
analogous stage in the development of B. anthracis (X 700). In this case, however, nothing
has been said in the text concerning these round lateral and terminal buds and ae round
bodies, clearly visible in the picture. Again a comparison with a more recent photograph may
be helpful. Figure 112 on Plate X is a reproduction from a picture made by Giinther (1906,
fig. 29, smear from the spleen of a mouse, X 1000), which presents the same type of gonidia and
regenerative body, and also an interesting side branch above the latter; but again all is passed
without any remark by the author.

Billroth (1874) was, like Perty, of the opinion that besides spores (‘‘Dauersporen’’) small
coccoid bodies are playing an important réle in the multiplication of the bacteria. He states
(on p. 16)—
dass die zuerst in einer bacterienfreien Flissigkeit sich bildenden Bakterien aus feinem und feinstem Coccos
entstehen.

Concerning the formation and development of this ‘‘Coccos,” we are told (pp. 21, 22):

Eine weit hiufigere und deutlicher zu verfolgende Umwandlung des Bakterienkérpers ist die zu kleinen runden
Kiigelchen, zu blassem Coccos.

Der auf diese Weise in den Bacterien entstehende blasse Coccos vermehrt sich zuweilen gleich nach seiner
Entstehung.

 

1 Some authors, like Hauser (1885) and Pfejfer (1888, p. 590) have been of the opinion that the old Bact. termo should be identified with B. proteus;
Beijerinck (1897, p. 40) related it to Bact. punctatum, while Macé (1897, p. 913), Z. F. Smith (1905, p. 170), and Lehmann and Neumann (1912, p. 411,
footnote) define it as Bact. fluorescens. That this view is correct is evidenced by aremark made by F. Cohn (18728, p. 197) upon the green pigment
produced by B. termo, and by a comparison of R. Koch’s and our own photographs (figs. 109 and 110 on Pl, X).'
No. 2.] LIFE CYCLES OF THE BAOTERIA—LOHNIS. 93

Two of Billroth’s drawings (original figs. 36 and 42), reproduced as figures 44 and 45
on Plate Is, present several interesting details. Formation, multiplication, liberation, as well
as upgrowth of the gonidia within the mother cell, are well discernible. The big inflations
with their granules, usually discarded by later authors as “merely involution forms,” make
interesting counterparts to Cohn’s Crenothrix “sporangium,” shown in figure 43. It is especially
noteworthy that the cells visible in figure 45, which are shedding their granules, were obtained
from whey kept at 40-45 C., i. e., under conditions leading to an accumulation of lactobacilli,
and just these organisms have furnished more recently excellent examples for this little known
mode of reproduction. Robin’s mixed species Leptothriz buccalis, mentioned above, probably
also included some lactobacilli. Bullroth’s drawings should be further compared with those
made by E. Chr. Hansen (1879) of acetic acid bacteria, which were reproduced as figure 15
on Plate E. The similarity of the reproductive organs produced by both groups of bacteria
will be discussed on another page.

Bacterium putridinis, described by Davaine (1876, p. 23) as the cause of ‘une pourriture
et ulcerations séches”’ in plants, exhibited a very similar behavior. It is said to have presented
itself in three different forms: ‘‘(1) en corpuscules amorphes, infiniment petits et innombrables,
constituant un tourbillon mouvant dont la plupart des individues se perdent aux limites de

la vision; (2) en filaments minces, courts, droits. . . atteignant au plus 0.005 mm. de
longeur, . . . (3) en filaments generalement plus long. . .” And Davaine adds:

Tl est remarquable que les corpuscules les plus petits qui échappent presque 4 notre vue aient une vitalité
supérieure & celles des bactéries bien développées. Cesontsansdoute . . . des germes.

The assumption that these germs might have been spores, on account of their ‘“vitalité
supérieure,” would collide with the other fact mentioned before, viz., that ‘un tourbillon
mouvant” was noticeable. That the resistance of the gonidia may be distinctly superior to
that of the vegetative cells will be shown later.

In 1877 some good descriptions of the formation of gonidia by Cladothrix and Crenothrix
were published by Cienkowski. He pointed out that.it is often difficult to distinguish between
gonidia and ordinary cell inclusions (“‘gew6hnliche Inhaltskérper’’), and that the reproductive
character of the former can be only ascertained by studying their further behavior and develop-
ment. Letzerich (1878) saw that in blood taken from typhoid cases, isolated cocci grew up to
large spheroids, called ‘‘Plasmakugeln,” and that later new bacteria were formed within each
of them. In 1878 Ewart recorded several equally interesting findings, secured by continuous
observations of hanging drop preparations. With Bact. termo (1878 a) he found like R. Koch
that bright, extremely small, almost spherical ‘“‘spores” ‘‘escaped’’ from the cells, but he found
in addition, that they were able to reproduce new rods. Not always, however, a direct up-
growth was observed; sometimes these “spores” first divided themselves into four ‘“sporules,”
which multiplied as such for some time, but then gave again a new growth of rods. Analogous
results were secured with Anthrax (Hwart 1878 6). Also in this case the “spores” either
reproduced the bacilli directly, or at first ‘“‘sporules’’ were formed, which grew as such for
some time, before reassuming their regular function. That these so-called spores were no
true endospores is proved, except by this behavior, by their not being heat resistant like genuine
spores, according to Ewart, and by their taking the stain very easily, as was found out by E.
Klein, who in 1883 made the same observations. The drawings made by the last-named
author, which were reproduced as figure 21 on Pl. F, leave no doubt that these “spores” have
been what we now call regenerative bodies and microcysts, which partially acted as gonidangia.
Perty’s discoveries upon the reproduction of spirilla have been confirmed by Geddes and Ewart
(1878). Again the “spores” were seen to “escape” at the end or at the side of the cells. They
were of unequal size, multiplied either as such, or grew up to new spirilla, and sometimes they
were seen to be inclosed within larger spheres. Their multiplication took place by fission as
well as by budding, inside as well as outside of the parent cells.

In the course of his investigations upon Actinomyces, J. Israél (1878) discovered that the
“bright spores,” formed in a lateral or in a terminal position by the threads, may multiply
for some time as such before reproducing new rods and threads, and that the clubs, which
94 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

developed at the end of the threads, may either break up and liberate their granular content, or
may produce “blasse Kérnchenhaufen”’ by fission or by budding.

Van Tieghem (1879 a and 6) studied carefully formation and germination of so-called spores
of Spirillum amyliferum, of a vibrio, and of a spirochaete found in oysters. Unfortunately,
their resistance against drying and heating has not been examined. In accordance with results
obtained by Heydenreich, Guttmann (1880) ascertained that also Spirochaeta Obermeieri pro-
duces “‘spores,’’ which occurred singly or in pairs, were actively motile, but could not be seen to
grow up to new spiral forms. Similar motile ‘granules’ were noticed by the last-named
author in the blood of patients afflicted with pneumonia crouposa, scarlatina, typhus abdom-
inalis, typhus exanthematicus, diphtheria, and erysipelas. Special attention was paid to their
differentiation from other blood granules. More data relating to Spirochaeta Obermeieri
were furnished by Albrecht (1881), who found nothing but granules in the blood during the
intermittent stage, while during the attack all forms, from granules up to full grown spirochaets,
became visible. He called the motile granules germs (‘‘Keime’’).

Formation, multiplication and germination of the gonidia produced by the large tricho-
-bacteria (Crenothrix, Cladothrix, Leptothrix, Beggiatoa) have become well known by the studies
made by Zopf (1879-1885). Figure 47 on Plate M is a reproduction of his drawings of Crenothrix
Kihniana (1883, original fig. 6), figure 48 one of Beggiatoa alba (1883, original fig. 27).

Engler (1882) confirmed Zopf’s findings and added a new form, Phragmidiothrix multi-
septata. The picture he made from it has been reproduced as figure 46 on Plate L (from original
fig. 24). It is of special interest because of the small branches, apparently produced by gonidia
germinating within the sheath. However, no final evidence could be secured in this respect.

Unfortunately, the results obtained with these comparatively large organisms have not
been used as a help for testing the smaller bacteria along similar lines. The trichobacteria
were either declared not to be bacteria at all, or they were removed to a special position as
‘“‘higher” bacteria (see p. 22). But all recent, as well as many of the earlier, observations
prove conclusively that formation and development of the gonidia are very much alike with
all—‘higher” and ‘‘lower”—bacteria.

Another opportunity to become acquainted with gonidia and regenerative bodies was
given to R. Koch (1881) when he made the two photographs reproduced as figures 113 and
114 on Plate X, showing anthrax and oedema bacilli in the tissue accompanied by globular
forms. However, again, as in 1877, no attention whatever was paid to these forms, which
fact is especially surprising, because at that time Koch’s standpoint concerning the etiological
significance of the anthrax bacilli was strongly contested by Fokker (1881-82), who pointed
out that in numerous cases no bacilli but only small coccoid bodies had been found, and that
the latter probably were produced by the former and reproduced them in their turn. More
evidence was furnished in this direction by Archangelski (1883), Roloff (1883), and Toussaint.
, According to a communication made by the last-named author to Magnin (1884, p. 150),
he noticed, like Ewart and E. Kleim, that the anthrax bacillus is able to produce smaller and
larger spherical or elliptical ‘‘sporangia,” especially when kept i in blood at 37-40° C. Three
to six ‘‘spores”’ were seen to escape from each ‘‘sporangium,” and a new development of rods
from these ‘‘spores” was also directly observed by Toussaint in Ranvier’s chamber.

In regard to the bacillus of symptomatic anthrax Ehlers (1884) came to similar conclusions;
he says:

Der Rauschbrandpilz bildet zwei getrennte Entwicklungscyklen, welche entweder in der Sporen- oder in der
Coccen-Gonidienbildung ihren Abschluss finden.

These ‘‘cocco-gonidia” were also found to be able to reproduce typical symptomatic anthrax
in the animal.

However, all these findings were discarded by the R. Koch school of bacteriologists. Loeffler
(1887, p. 171) declared that Fokker, Archangelski, Roloff, et al. had been ‘‘wrong without
exception,” though he does not furnish any evidence supporting his statement. The absurd
dictum made by Fraenkel in regard to the ‘‘unnecessary” and ‘‘illegitimate” nature of the
coccoid reproductive organs of B. anthracis was quoted on p. 27. LEhler’s results were dis-
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 95

credited by Kitt (1887), but later confirmed by Grassberger and Schattenfroh (1907). The inter-
esting manner in which Hibler (1908) tried to get rid of the embarrassing presence of the large
round forms (regenerative bodies or gonidangia) of B. Chawvoei has been mentioned on p. 67.
Figures 115 and 116 on Plate X are reproductions of two others of his protographs (original
figs. 17 and 18, Pl. VIII, x 2,000), which demonstrate clearly that with this species it is not so
easily possible to overlook these ‘‘illegitimate”’ formations, as R. Koch and his followers could
do with B. anthracis and the smaller bacteria. The convenient ‘‘explanation,’”’ that they are
‘“‘merely inflations” or ‘‘simply involution forms,” has, of course, proved helpful here as in
many similar cases.

On the other hand, 2. Koch (1882-1884 a) and several of his pupils did not hesitate to ascribe
‘‘spores”’ to tubercle, typhoid, and other bacilli, usually on account of no other reason than that
these bacilli sometimes show inclusions, which are not stained by aqueous dyes, and that their
resistance against drying may be sometimes rather high. Koch declared that the tubercle
bacilli produce spores ‘‘like the anthrax bacilli,” though he usually saw 4-6 unstained inclusions
in each tubercle bacillus. And he also asserted that occasionally these not stainable ‘‘spores’”’
alone were present within the tissue:

Thre Anwesenheit verrat sich nach dem Verschwinden der Bacillen nur durch die infectidsen Eigenschaften der
kisigen Substanz, in welche sie eingebettet sind,

He also noticed that in some preparations only stained granules became visible instead of
the bacilli, but there was no doubt to him that these were ‘‘granules of plasma cells.’”’ However,
Klebs (1883 a and 6), Babes (1883), Malassez and Vignal (1883-1884), Biedert and, Sigel (1884),
Artigalas (1885), E. Klein (1885), Amrusch (1886), and Lutz (1886) furnished soon a large
amount of evidence that these stainable granules may be, and often are, the true offspring of
the bacilli, and that they are able to reproduce the long forms. -

The typhoid ‘‘spores’”’ observed by Eberth (1880) and Gaffky (1884) have been refuted by
Buchner (1888) and by Pfuhl (1888). Similar claims made by Hueppe (1884) concerning B.
acidi lactict and B. cyanogenes were retracted by his pupil Epstein (1890). What A. Neisser
(1881) and G. A. Hansen (1882) declared to be lepra ‘‘spores” have been undoubtedly regenera-
tive bodies, as is to be seen especially from the description given by Neisser. The results of the
leprosy studies of Babes (1883), Lutz (1886), and Bordoni-Uffreduzzi (1888 a) fully confirm this

* conclusion.

The florid hypotheses, developed by Béchamp (1883) in his book on ‘‘Microzymas” in
relation to the bacteria and their various reproductive organs, have been distinctly unfavorable
to further studies upon these problems; and the usage established by Zopf (1883) to take the
term ‘‘cocci” as practically equivalent to ‘‘gonidia,”’ has also caused considerable misunder-
standing. Hueppe (1886, pp. 109, 125) blamed Zopf for this incorrect denomination; not every
globular or ovoid body should be called coccus, and this name should never be used for repro-
ductive organs. But Hueppe also emphasizes on the other hand (1. c., p. 133) that globular
bodies which are able to multiply as such should be called cocci, not gonidia. So again Zopf
would have been correct, because all gonidia undoubtedly can multiply as such by fission, as
well as by budding, and often do so before reproducing the larger cells.

In his ‘‘ Traité de Botanique” Van Tieghem (1884, p. 1105) introduced the name ‘‘cyst”’
as designating those resting cells, which later have been also called ‘‘arthrospores.” He defines
them as follows: :

Les kystes ne sont que des cellules ordinaires . . . qui grandissent, changent de couleur, épaissent leur
membrane et passent enfin 4 1’état de la vie latente. Ala germination, le corps protoplasmique du kyste, revétu de
la couche interne de la membrane, reprend sa couleur primitive, se cloisonne dans la méme direction que lorsqu’il
faisait partie de filament, déchive la couche externe cutinisée de la membrane et s’allonge au dehors.

ée

The exact meaning of the term ‘‘arthrospore,’’ introduced into bacteriology by DeBary
(1884), has been given neither by him nor by Hueppe (1886), who made it popular. The latter
author only points out that arthrospores are always of globular form and do not multiply as
such. That he also recommended to include the gonidia among the arthrospores (1. ¢., p. 128)
was an evident mistake.

160266°—21——7
96 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

Arthrospores were especially attributed by Hueppe (1885) to the cholera bacilli. His
drawings, reproduced as figure 49 on Plate M (from original fig. 2), leave no doubt that he saw
those reproductive organs, which we now call regenerative bodies. Germination by stretching
was repeatedly observed; in one case formation, as well as germination, of this organ was
continuously studied with the same cell. Increased resistance of these bodies, when dried,
was also noticed. Kitasato (1889) was unable to confirm Hueppe’s findings, and this negative
result has been often quoted in the textbooks (e. g., by Pfeiffer 1896, p. 536; Heim, 1906, p. 187;
Lehmann and Neumann, 1912, p. 517) as being full proof of the inaccuracy of Hueppe’s work; but
it has not been quoted, on the other hand, that there are recorded many more positive results,
all confirming and extending Hueppe’s discovery. Babes (1889) saw that terminal round bodies
are produced by the cholera vibrio, which after separation grew up to new bacilli; within
long spirilla he found several of such round bodies. Their higher resistance against drying was
carefully tested by Finkler and Prior (1885), as well as by Firtsch (1888), with V. proteus, by
Weibel (1888) with V. lingualis and saprophiles, and by Esmarch (1887) with his Spirillum
rubrum, where the granules, as their author says, “‘are arranged like the peas in an opened pod.”
Sorokin (1887-1890) called these bodies “‘spores,” though they did not stand the heating, but
he saw them germinate while still within the parent cell, which process led to those interesting
branched forms reproduced in figure 29 on Plate H, and induced him to call his organism Spirillum
endoparagogicum.

Development of new cells from the gonidia, still inclosed in the old cell, has also been
described and drawn by Kiinstler (1884) for his Bacterioidomonas sporifera. Nearly 20 years
later the same occurrence was studied by Chutton and Pérard (1913). Their drawing, repro-
duced as figure 50 on Plate M (from original fig. on p. 1233), leaves no doubt that their Metabac-
tervum polyspora is identical with Kiinsiler’s organism, whose existence was evidently unknown
to them. Like Kiéinsiler they call the 1-8 new rods which were seen to develop from ‘‘masses
chromatiques,”’ ‘‘spores,’’ obviously an erroneous interpretation.

Analogous endogenous upgrowth of the gonidia was discovered by Finkler and Prior
(1884-1885) with their vibrio. They either saw the motile “spores” coming out of the mother
cells and developing outside into new vibrios, or the ‘spores’ remained and grew inside of the
old cells, which increased in size and often assumed a club-shaped appearance. Later these
“nurses” (Ammen), as the authors call them, broke up and set the new bacilli free. Ferrdn °
(1885) described large round or mulberry-like bodies of the cholera vibrio, which grew up from
the granules contained in the thread-like cells, and which either send out thin threads becoming
new spirilla or produced new spirilla inside of their cell walls, which later dissolved. The
Spanish author indicates in addition that similar modes of reproduction probably occur with
cocci and bacilli, too, and he mentions a Dr. Colvée in Valencia who observed similar things with
the tubercle bacillus. In another paper by Kiinstler (1885) “De la position systématique des
bactériacées” the following important statements have been made concerning spore formin
bacilli and spirilla: ’

D’aprés certaines découvertes récentes il semblerait que ces formes se transforment souvent en vésicules claires, de
dimensions considérables rélativement 4 leur taille primitive. Dans ces vésicules se produisent une foule de spores,
analogues 4 des microcoques, qui sont mises en liberté par la déhiscence des parois.

Les spores des Spirilles se divisent au sein de leur enveloppe et finissent par acquérir alors la faculté dese mouvoir.
Ce n’est qu’ultérieurement que les sporules deviennent libres et germent.

In the same paper a spirochaete-like organism has been described under the name Protero-
monas Regnardi, which was placed between bacteria and flagellates, and which was found to
produce an inflation at or near the middle of the body, from which small “buds” were seen to
separate, which became motile and grew up to new individual cells of ‘‘ Proteromonas.”’

The picture reproduced as figure 122 on Plate X from Artigalas’ “Microbes pathogénes”’
(1885, original fig. 1 on Pl. 5) shows the giant cells formed by Pneumococcus, which the author
calls ‘‘spores,” producing and liberating ‘‘sporules.’’ Maddox (1885) published at the same
time a similar drawing of inflated, club-shaped cells of Bacterium lactis (i. e., Streptococcus lactis),
which let out their deeply staining granular content, and which were accepted by the author as
‘sporangia,’ not as ‘‘degenerate’’ cells.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 97

Evidently without knowledge of the work done in France, in England, and in Spain, Schroen
(1886-1891) in Italy found out that the tubercle bacillus may produce globular ‘“‘spores,”
which eventually increase in size and produce internally several ‘‘daughter spores,” which later
leave the cyst (‘‘Kapsel”) singly or in chains, forming again new granulated bacilli. With
numerous other species (B. anthracis, Megateriwm, V. cholerae, etc.) he secured analogous
results. The content of the large inflated cells, the ‘involution forms” of the German authors,
was seen to transform itself into small coccoid and rod-like bodies, which later were liberated.
In a footnote to an abstract of Schroen’s first paper, Bawmgarten in his ‘‘ Jahresbericht’’ (1886)
strongly doubts the accuracy of this work and mentions as foremost reason for this that neither
BR. Koch nor he himself had ever noticed such modes of reproduction.

Klebs (1887, p. 76) was of the opinion that in many cases the deeply staining granules
(“chromatophilen Kérner’’) visible in the cells, which he named “‘microsomes,’’ may serve as
starting points for new bacterial development. Hauser (1885) noticed that the so-called involu-
tion forms of his Proteus vulgaris, whose interesting photograph is reproduced as figure 120 on
Plate X (from original fig. 16 on Pl. X) often contained one or more small, dark granules, or, more
frequently, a globular, bright body, whose nature remained problematical to the author. All
these large, globular, pear- or dumb-bell shaped forms were actively motile, reproduced quickly
(within 12 hours) luxurious colonies, could be made constant in their characteristic appearance,
and proved to be fairly resistant against drying, i. e., they exhibited in no way any sign of “invo-
lution,” but presented themselves as an interesting connecting link between rudimentary goni-
dangia and sporogeneous cells. The lateral bud attached to the long rod at the right side of
the picture and the many small forms apparently growing up from the gonidal stage are equally
noteworthy. ;

The organism described by Bordoni- Uffreduzzi (1888 6) under the name Proteus hominis
capsulatus, which, however, undoubtedly is to be placed in the neighborhood of B. pneumoniae,
produced within the body rods and large numbers of gonidia, as may be seen from the liver
cut, reproduced as figure 123 on Plate X (from original fig. 6 on Pl. VIII); in the cultures also
large globules were found, which reproduced rods, and many other irregular forms, mentioned
already on page 55. ,

Of special interest is a report made by A. Neisser (1888) upon ‘‘spore” formation of B.
zerosis. The bodies concerned, for which the special treatment, known as Neisser’s spore-
staining method, had been developed, are those granules or disks in the so-called barred type
of the diphtheria group, which are now called gonidia, while Neisser used this term for the
whole, club-like structure. The most important findings were these:

Jedes einzelne Teilglied wichst wieder zu einem neuen Bacillus aus. Die Wachstumsrichtung dieser Teilchen

steht aber senkrecht auf derjenigen, in welcher sich der einzelne Bacillus zu der keulenférmigen . . Kette
ausbildete.

This mode of development leads to the parallel arrangement of short rods of uniform
length, so characteristic for this group. Bacilli containing 3 or 4 granules, ‘‘welche als erste
Anfiange der Bacillen ganz kurze Ansatzstiicke zeigen,’ were also present. And it was further
emphasized that these ‘‘spores” are produced in greatest numbers when the development
reaches its maximum, not when it comes to its end, as with B. anthracis.

All these observations are in good agreement with some of our recent findings concerning
the development of the gonidia of B. azotobacter and of other species. Still more complete,
however, is the coincidence of facts incorporated by Tomaschek (1888) in the drawings of his
large B. muralis, reproduced as figure 51 on Plate N (from original fig. on p. 183), and those
secured in our Azotobacter studies. In fact, every detail visible in this picture could be
accepted as representing some stage in the development of B. azotobacter. The large ovals sur-
rounded by concentric slimy layers, the formation of the small globular gonidia, their liberation,
as well as their upgrowth inside of the mother cell, are all as with Azotobacter. Hansgirg
(1889) vigorously maintained against Tomaschek that this so-called B. muralis be no bacillus
at all, but a colorless alga. That again is exactly what has been asserted by several authors in
98 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

regard to the large cells of Azotobacter, though our more recent, findings leave no doubt that
they are merely a type of grows in the life cycle of spore-forming bacilli of the Subtilis-
Mesentericus group.

Another interesting picture of the formation of gonidia within a bacillus has been pub-
lished by W. Miller (1889, p. 54), when he described what he wrongly called Jodococcus vaginatus.
A glance at figure 52 on Plate N (reproduced from original fig. 15) shows that it makes a good
counterpart to some of the older drawings made by Billroth (see fig. 44 on Pl. L) and by others
of the trichobacteria, and that Baumgartner (1909) was perfectly right when he removed this
organism as Bacterium iogenum to its proper place.

What Beijerinck (1888), Frank (1890), and Prazmowski (1890) described as liberation of
‘‘swarming bodies” (‘‘Schwirmer’’) by B. radicicola is, at least partially, to be classed among
the early observations upon bacterial gonidia. The last-named author compared them to small
bright globular spores (‘‘Sporenktigelchen”’); Atkinson (1893) called them zoospores.

Dowdeswell’s investigations (1889-1890) upon the cholera organism brought much con-
firmation to those of EH. Klein, Ferrén, and Schroen. ‘‘Sporangia” producing ‘‘sporules,”
rods, and threads splitting up into granules, and the regeneration of normal forms from these
sporules were once more studied continuously with the living material. The drawing of a
stained preparate of cholera, published by Cornil and Babes (1890, original fig. 275) and repro-
duced as figure 124 on Plate X, may be compared with Hueppe’s sketches, reproduced as figure
49 on PlateM. An analogous picture of the first-named authors, showing gonidia and regen-
erative bodies of the tubercle bacillus, was reproduced as figure 102 on Plate IX.

Mac Fadyean (1889) noticed similar large globular bodies with Actinomyces and thought
they were related to the clubs, while the smaller ‘‘cocci”’ were declared by him to be the ‘‘seed”
for new development in the animal. Bostroem (1890) called the bright, darkly staining, pear-
like granules which he found in and outside of the Actinomyces threads ‘‘spores.” Within the
clubs he also observed such granulated threads, which were seen later to disintegrate, liberating
the ‘‘spores.”’ That they may cause budding and branching was found out by Eppinger (1890),
who noticed that the granule visible in the thread at first develops a small ‘‘wart,’’ which then
grows out into a branch, just as had been seen by Sorokin with his Spirillum endoparagogicum.
The interesting forms of ‘‘Streptothriz cuniculi” photographed by Schmorl (1891) were repro-
duced as figures 89 and 90 on Plate VIII. The granules within the threads, but also their
liberation are well discernible, especially their coming out of one of the inflations visible in
figure 90; lateral buds and the beginning of branching are equally conspicuous. Mafucci (1892)
recorded similar findings with the avian type of B. tuberculosis, and Fr. Fischel (1898) reported
concerning the human type of this organism:

In manchen der kléppelartigen Verdickungen sieht man beildufig in der Mitte derselben dusserst kleine,
hellglinzende, runde oder ovale, an sehr verkleinerte Milzbrandsporen erinnernde Gebilde.

Marpmann (1893) found in sweepings from the street round spore-like germs of the tubercle
bacillus, which gave him normal cultures after having been kept for one hour in the steam.
Coppen-Jones (1895) recommended to place the ‘‘spores”’ of the tubercle organism next to the
chlamydospores of Mucor and similar fungi. Semmer (1895) met again the bright granules in
the threads, as well as in the big inflations, of tubercle and glanders bacilli, and saw them when
liberated grow up to new rods and threads. And the observations made by A. Neisser (1888)
with B. xerosis were confirmed and extended by Escherich (1894) in his work upon the diph-
theria bacillus.

A permanent loss of endospore formation was first recorded by Lehmann (1888) with
B. anthracis. The “‘microspores,” which were found in such asporogenous rods, were killed
when kept 2-3 hours at 60°C. Hous (1890) collected additional experimental evidence upon
the same subject, while Novy (1894) noticed with an asporogenous strain of B. oedematis malign
that moderate heating (1 hour at 58° C.) had no deleterious effect, thereby proving beyond
doubt that the easily stainable granules, which were present in and outside of the rods, were
more resistant than the vegetative cells and able to act as reproductive organs. The subsidiary
réle played by the ‘‘granules” in such cases was already indicated to some extent by analogous
No. 2.) LIFE CYCLES OF THE BACTERIA—LOHNIS. 99

findings with B. anthracis.and Megaterium, made by DeBary (1884). Weibel (1888), on the
other hand, found once more large globular bodies and big club-like inflations within the threads
of an asporogenous culture of B. anthracis, and L. Klein (1889) reported that he obtained, as
a ‘‘contamination” from a culture of B. Megaterium a spore-free ‘‘new” B. allantoides, which
produced large ‘‘sausage-shaped”’ accumulations of ‘‘cocci”, probably similar to those seen
by Billroth (1874) and reproduced in figure 44 on Plate N.

A photograph published in the ‘‘Atlas”’ of Fraenkel and Pfeiffer (1895) of what the authors
call ‘crippled degeneration forms” of B. Chauvoei, widely differing from the typical ‘‘vor-
schriftsmissigen’”’ cells, has been reproduced as figure 117 on Plate X (from original fig. 60 on PI.
XXX). It makes an excellent object for comparison with the analogous ‘‘involution forms”
in Hauser’s Proteus picture (fig. 120). Rudimentary spore formation is well noticeable in
both. Inaddition, the three figures, Nos. 118, 119, and 121, are equally worth studying. They
were reproduced from Maassen’s (1904) paper on the ‘‘involution forms’ of the bacteria
(original figs. 4 and 7 on Pl. X, and fig. 1 on Pl. XI, respectively), and demonstrate the develop-
ment of B. pestis, B. lactis aerogenes, and of V. cholerae on lithium chloride agar. The same
hardened granules appear in the inflated globular and club-shaped cells. A picture of B. pyo-
eyaneus, from salt agar, made by Matzuschita (1900) and reproduced as figure 125 on Plate
X (from original fig. 26), illustrates the same fact, in reference to which A. Fischer has made an
interesting hypothetical remark already in 1891, which, however, has never been tried experi-
mentally. It is very probable, indeed, that A. Fischer was right when he assumed that by an
increase in salt content of the substrate perhaps many bacteria could be induced to form spores,
which otherwise are not so inclined. According to our experiments the spores are closely re-
lated to the regenerative bodies, and the formation of these reproductive organs is greatly
favored by drying. A higher salt concentration in the agar causes the analogous osmotic effect,
and it is, therefore, easy to understand why practically all the pictures made by Matzuschita
and Maassen from their salty cultures show various phases in the development of regenerative
bodies and gonidangia. The figures 5, 6, 18, 36, 46, 57, 84, 118, 119, 121 and 125 o0n Plates I-V,
VII, and X, which were.made from their photographs, are by no means pictures of some quaint
and unimportant ‘ ‘teratologic” type of growth, as Maassen assumed; in fact, they all illustrate
interesting phases i in the formation and development of the reproductive organs, as character-
istic of the various species.

The same holds true concerning other so-called involution forms, mostly from old agar
cultures, shown in the photographic atlasses of Fraenkel and Pfeiffer (1895) and of Itzerott and
Niemann (1895), some of which have been reproduced as figures 25, 30, 47, 79, and 80 on Plates
III, IV, and VII. Others are copied as figures 126-130 on Plates x and XI (from original figs.
7, 16, 46, 60, and 104 in the atlas of the last-named authors) illustrating analogous stages in the
development of various spirilla (126), B. anthracis (127), B. capsulatus Pfeiffer (128), Vibrio
Bonhoff (129), and Spirillum rubrum (130). The characteristic budding of the globular darkly
stained bodies is clearly visible in all of them.

Other exceptionally good photographs of regenerative bodies, budding out in a lateral
a a were made in 1891 by Zettnow; one of them is reproduced as figure 131 on Plate

XI (from original fig. 14). But they, too, were discarded as being probably ‘‘merely in-
volution forms.”’ Further studies made with impure cultures of Spirillum Undula (Zettnow,
1896) furnished among others the interesting picture reproduced as figure 132 on Plate XI (from
original fig. 16). Round darkly stained and also not stained globular bodies are visible inside,
as well as outside, of the spirilla; some of them seem to sprout sideways, and a contaminating rod
exhibits also its typical lateral buds. Zettnow, however, did not pay any attention to the last-
named fact, and the globules of the spirilla are to him simply ‘‘vacuoles,” though vacuoles of a
truly astonishing behavior. Zettnow informs us that these ‘‘vacuoles”’ transformed themselves
into ‘‘globules”’ of various size, which separated themselves from the parent cell and became
free! After having liberated themselves these transformed ‘‘vacuoles’’ exhibited another equally

 

'He says literally, that ‘‘die Hohlraume sich in kleinere und gréssere Kugeln umwandeln, diese sich abtrennen
und schliesslich frei werden.”
100 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

surprising feature, viz., they were now difficult to stain, and only after having been treated with
a mordant (‘‘Loeffler’s Beize’’) they took the stain at least at the outside. Because as Zettnow
states (contrary to the facts mentioned above) nobody had ever spoken of spores in connection
with spirilla; he thinks that he is justified to class these vacuole-globules as signs of degener-
ation. In another paper (Zettnow, 1897) he returns to the subject, after having now worked
with pure cultures, which were studied living, stained with methylene blue. Spirilla contain-
ing granules were especially actively motile; nevertheless the granules are to Zettnow still the
mate of degeneration. They are declared now to have been gas bubbles, which persist and

gglomerate after the spirilla have died. They gave ‘‘fat reaction,” too, though they were
ie stained by methylene blue; and because they also appeared as ‘‘Chromatinmassen,”’ the
author is inclined to think—

dass sie bestimmt sind, die Art tiber die ungiinstige Zeit, in welcher die Vermehrung stockt, zu erhalten.

Buds, however, are still to him “krankhafte Degenerationsformen.”  Kutscher (1895 a),
who had isolated the pure cultures used in these studies, obtained analogous results. He
also saw the bright granules within the spirilla and the same characteristically branched forms,
as described and drawn by Sorokin (see pp. 72 and 96), but because the granules did not show
a high resistance against heating, and germination tests apparently were thought to be super-
fluous, they were declared to be nospores. Cunningham (1897), on the other hand, noticed once
more that cholera spirilla may be reproduced by coccoid forms. Kohlbrugge (1900-1901) re-
corded analogous results with several water vibrios, and Nakanishi (1900) with V. cholerag,
V. Finkler-Prior, and V. Metchnikovii. In the latter case the globular bodies were found to
predominate in 4-day-old cultures, and after being transferred to new substrates they ger-
minated readily to new vibrios. Migula (1897, Vol. I, p. 157), however, was of the opinion that
the round bodies, which Perty had found with the spirilla, had been—

. andere Organismen, da sie teils endsténdige, von dem Spirillum durch eine Scheidewand deutlich getrennte,
teils irgendwo dem Ricken aufsitzende, keinerlei genetischen Zusammenhang mit dem Spirillum bekundende Kugein
sind.

Undoubtedly, Migula had not read the careful description given by Perty, quoted on
page 91, otherwise he would not have made such an incorrect statement. Furthermore, itis
not without interest, that he himself (Migula, 1904) published a photograph of some spirilla,
reproduced as figure 133 on Plate XI (from original fig. 8 on Pl. II), showing exactly the same
body, or ‘‘foreign organism’’ according to his opinion, attached to the ‘‘back’’ of one of the
spirilla in his own preparate. This picture should also be compared with figure 130 (Spir.
rubrum, photographed by Jizerott and Niemann). Another photograph given by Migula (1904)
on the same plate, has been reproduced as figure 134 on Plate XI. It was destined to illustrate
the flagellation of B. proteus, but it exhibits in addition the round bodies characteristic of this
species. - How little Migula knew, indeed, of the reproductive processes common among the
baeteria, and about the literature covering this subject, is very clearly indicated by the manner,
in which he refutes in the last-named publication (1904, p. 61) some observations made by
Trambusti and Galeotti (1892) concerning the formation and liberation of gonidia, though they
were in complete accordance with the earlier findings of Perty, Billroth, Ewart, Toussaint, Kiinst-
ler, Finkler and Prior, Ferran, Artigalas, Maddoz, Schroen, T omaschel, Dowdeswell, and others
cited above. To Migula such an occurrence is simply impossible:

Ein solcher Vorgang kommt bei den Bakterien der ganzen Art und Weise ihrer vegetativen Entwicklung nach
nicht vor.

In figure 53 on Plate N another example of this “impossible” mode of reproduction is
shown, as it was drawn by Babes (1895, original fig. 13 B) when he studied Ascobacterium
luteum. Formation, multiplication, and upgrowth, as well as liberation, of gonidia and young
rods are all well visible.

More evidence concerning the occurrence of globular reproductive organs with various
nonspore-forming bacteria, especially with B. typhi, dysenteriae and V. cholerae, was furnished
in a series of important contributions by EF. Almquist (1893-1917), who especially emphasized,

‘
No. 2.) LIFE CYCLES OF THE BACTERIA—LOHNIS. 101

that if we keep the bacteria in the laboratory under conditions more similar to their nat-
ural environment, they will reveal at once much more of their life history than they are
able to exhibit under the monotonous artificial conditions under which they are usually com-
pelled to live. At first it was shown (Almquist, 1893) that B. coli, as well as B. typhi, when
grown at a low temperature (10-11° C.) in manure soil, produce by fragmentation, as well as
by budding, very small round bodies, which transferred to new substrate (broth) at room
temperature quickly reproduce new normal, often also curved, rods. Further studies
(Almquist, 1904) added to the small ones larger round bodies, which were called “‘conidia”’ by
the Swedish author, and which were also found to bud out at the side or at the end of the vege-
tative cells, often supported by a short stem. They were discovered in both cholera and
typhoid cultures, especially when 2-3 per cent NaCl had been added to the substrates. They
were easily stainable and reproduced readily on new substrates each one or two small, very
motile, young vegetative cells. With cholera the germinating rod looked at first very much
like typhoid, while in the latter case the first generation was extremely thin (‘‘nadelartig’’).
It was noticed, too, that these ‘‘conidia” are able to multiply as such by budding. All these
findings were confirmed and extended in the publications of 1906-1908; they are mostly based
on direct observations of the living material. In the 1907 paper Almquist compares his conidia
with the reproductive organs of the iron bacteria, also called by him “conidia.” Some drawings
published in 1908 are reproduced as figure 54 on Plate N (from original figs. 5,13, and 21). It
must be admitted that they are not very impressive, but everybody who has had some experi-
ence in this direction will acknowledge their accuracy. But it was not until 1916 and 1917
that good photographs were also furnished by Almquist. Five of them are reproduced as
figures 135-139 on Plate XI (from original figs. 4, 5, and 23 in the 1916 and figs. 1 and 4 in the
1917 papers), illustrating formation and development of the “‘conidia” of B. typhi (figs. 135,
136, and 138), B. dysenteriae (fig. 137), and V. cholerae (fig. 139). Besides these species,
B. diphtheriae, B. acidi propionici c, several micrococci and coli-like organisms were also
studied. A considerable number of the experiments was made with single-cell cultures; dried
agar kept at 10-14° C. proved to be especially favorable for stimulating the development of
these round reproductive forms. It is obvious that Almquist’s discoveries are in complete
agreement with many earlier, as well as with more recent, observations made along similar
lines; but probably nobody has followed so patiently and extensively the formation, multipli-
cation, and germination of what we now call gonidia and regenerative bodies, as Almquist did.

That his findings were contested by authors like Gotschlich (1909) and Kruse (1910, pp.
8 and 31) is not surprising, but does certainly not reduce their great value to any extent. On
the other hand, it is again rather interesting to compare these pictures of Almquist with some
others contained in various papers and textbooks, which reveal the same details, though as a
rule, this was not noticed or at least not mentioned, by the respective authors themselves.
Figure 140 on Plate XI is a reproduction of a’ photograph made by Friedrich (1892, original
fig. 4 on Pl. V) from a 4-day-old cholera culture. Figure 141 is the picture of a young Pyocy-
aneus culture, published by Muir and Ritchie (1903, original fig. 69); it should be compared
with figure 31 on Plate III, showing the same process, i. e., germination of gonidia and of
regenerative bodies, with the plague bacillus, photographed iy the same authors, and with
figure 136 on Plate XI (germinating gonidia and regenerative bodies of B. typhi, Almquist).
Figure 142 was made from Giinther’s textbook (1906, original fig. 54), presenting the develop-
ment of regenerative bodies of B. enteritidis. Figure 148 illustrates the same process with
B. ochraceus as photographed by Migula (1900, Vol. IT, Pl. XI, fig. 3), and as figure 144 another
picture made by the same author (Migula, 1900, Vol. II, Pl. XVII, fig. 5) has been added,
presenting a Spirillum “sporiferum.” . Of special interest, however, is figure 145, made after
a picture published by Kitt (1899, p. 393), wherein numerous regenerative bodies are visible,
budding out at the side or at the end of the rods, belonging to a Bac. phlegmasiae uberis (a
member of the Aerogenes group).

That the colonies growing on the plate sometimes contain nothing but such round regener-
ative bodies, is illustrated by two photographs made by Azelrad (1903), reproduced as figures
102 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

146 and. 147 on Plate XI. The former shows a contact preparate of Streptococcus lanceolatus;
its large budding cells are especially noteworthy. The latter was taken from a colony of B. coli.

A réle played by the gonidia in the multiplication of the bacteria, and the parallelism indi-
cated thereby between tricho- and other bacteria, are also shedding some more light upon the
physiological significance of the clubs formed by Mycobacteria and Actinomycetes, as well as
by many other bacteria. As figure 43 on Plate L an old drawing, made by F. Cohn (1870), was
reproduced, showing an interesting “club,” attached in a lateral position to a Crenothrix
thread, and considered by the author to be of a spore-like character. On Plate J as figure 34,
a drawing by Babes (1895) was given, presenting clubs produced by a diphtheroid organism,
one of them containing four small granules, while in the other one a larger spore-like body is
visible. On Plate B as figure 3 another drawing made by the same author may be found; one
of the Streptococcus chains sketched therein, has also attached to its side a club, which resem-
bles very closely the Crenothrix club of Cohn. That the gonidia, or “spores,” as A. Neisser
called them, which are contained in the club-shaped cells of the diphtheriod bacilli may according
to this author’s observations (A. Neisser, 1888) germinate therein and grow out sideways was
mentioned on page 97. In addition, on page 98, quotations from MacFadyean (1889), Bo-
stroem (1890), and Eppinger (1890) have been given, which also indicated that relations seem to
exist between the production of gonidia (‘‘spores”) within the threads of Actinomyces and
the formation of the characteristic clubs. Concerning the latter Crookshank (1896, p. 437)
made the following important statement:

These club-shaped bodies represent organs of fructification, rather than the results of degeneration or death. The
difficulty in accepting the view of their being entirely lifeless forms lies in the fact that the author has observed daughter
cells growing from the mature clubs; and, further, in the bovine fungus the author has been able to trace the stages
in the development of a single club to a completely formed rosette.

As figure 55 on Plate N a drawing, made by the same author (Crookshank, 1896, original
fig. 12 on Pl. XI), has been reproduced, showing tubercle bacilli taken from the liver of a Rhea,
with swellings at the ends of the rods, which were said to be “very suggestive of spores.”
Whether the apparent germination of some of the liberated “spores” visible in the picture has
been a true sprouting was not ascertained. Some years earlier, however, Coppen-Jones (1893)
had reported that he saw clubs of the tubercle bacilli grow out into numerous short hyphae,
and later Conradi (1900) recorded the same occurrence with B. maillei: fine short lateral threads
being produced by clubs, as well as by the older broad threads.

On the other hand, Olsen (1897) has emphasized that spore-forming bacilli, too, may assume
the appearance of an Actinomyces. His drawing of B. mycoides grown in sand was repro-
duced as figure 20 on Plate F. It should be compared with Babes’ sketch of a clubbed
diphtheroid bacillus mentioned above, as well as with the picture of B. tuberculosis made by
Metchnikoff (see fig. 38 on Pl. J) and with the description of B. diphtheriae given by Hill (quoted
on p. 78). Asin these two cases, the breaking off of clubs in the shape of oval bodies from
the end of side branches of B. mycoides is clearly visible in figure 20, but in addition the forma-
tion of endospores within the clubs is also well illustrated. According to the Norwegian
author the same phenomenon, viz., the production of one endospore within each club, was
noticed with several cocci, sarcinae, and with other bacilli. A drawing made by Dangeard
(1891) of his ‘‘ Hubacillus multisporus,” reproduced as figure 56 on Plate N (from original fig. 4 on
Pl. VIII), wins new interest in this connection. And it is worth mentioning, too, that on the
one side Olsen’s view is supported by analogous findings with spore-forming bacilli reported by
Cozzolino (1900) and by Johnson (1912), on the other side by some observations made by
Babes (1907 a) with B. leprae, and by Schiitze (1908) in regard to Actinomyces monosporus.
According to the first-named author the branched forms of B. leprae are inclined to produce
terminal clubs, staining at the outside like the normal cells, while the bright center reacts like
a spore; and Schiitze’s Actinomyces monosporus is characterized by always producing only
one oval spore at each branch, which, though still easily stainable, was found to display a
higher resistance than other Actinomyces spores. Loele (1908) at the same time secured
further evidence, that at least some of the Actinomyces clubs clearly exhibit their character as
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 103

reproductive organs, viz., those containing numerous granules. They were classed by the
author as ‘‘sporangia.”’

A very interesting drawing of what we now call regenerative bodies, made by Henneberg
(1898) of an acetic acid bacterium (Bact. orydans), is reproduced as figure 57 on Plate N (from
original figure 1, whose magnification evidently was not properly determined). It furnishes a
very clear insight into the mode in which these bodies are formed. The same reproductive
organs were seen by Henneberg (1901) also with several lactobacilli, where they are, indeed,
most frequent, butit is not easy to understand how the author, after having made those sketches
of the acetic acid bacteria, now can state without hesitation:

An den Zellen kleben hiufig mehr oder weniger grosse tropfenférmige Ausscheidungen aus den Nahrfliissigkeiten,
die den Zellen ein ganz verandertes Aussehen geben kénnen. 3

B. Lindneri is declared to have shown ‘‘fettartige Ausscheidungen aus den Nihrfliissig-
keiten an den Zellen festgeklebt.”’

Tissier (1900) obtained the same round bodies with his B. bifidus, but he was careful
enough not to discard them without trial; therefore, he was able to report:

De ces boules partent en rayonnant des corps bacillaires d’une grande finesse qui se subdivisent.

Sandberg (1904) made some good photographs of regenerative bodies attached to lacto-
bacilli taken from the stomach, though not noticing them. Weigmann, Gruber, and Huss (1907)
made pictures of the same ‘‘bulb-like inflations’’ growing at the end or at the side of lacto-
bacilli from matzoon, and Kuntze (1908) photographed those of B. bulgaricus. None of these
authors, however, saw them germinate. As figure 148 on Plate XI one of our photographs of
B. bulgaricus (Léhnis and Smith, 1916 a, fig. 29) has been reproduced, showing two regenerative
bodies in their characteristic position and also (below the upper one) two thin threads parting
at the same place from a broader rod.

Analogus things have been repeatedly photographed. Maassen (1904), for instance,
published a picture of a spore-forming bacillus isolated from chopped meat (‘‘Hackfleisch’’),
which was reproduced as figure 149 on Plate XI (from original fig. 4 on P]. XI), where it makes
an interesting counterpart to our photograph of B. bulgaricus, as well as to Olsen’s drawings of
spore-forming clubs. In the textbook of Hiss and Zinsser (1914, p. 576, fig. 125) B. subtilis is
photographed as reproduced in figure 150 on Plate XI, but in the text nothing is said con-
cerning this remarkable outgrowth.

Small coccoid reproductive organs of B. coli, like those discussed by Almguist (1893), were
closely studied by Adami, Abbott, and Nicholson (1899) and by M. E. Abbott (1900). Tosome
extent they behaved like vegetative cells and were mentioned, therefore, already on page 56
But their ability to reproduce normal rods was also well noticeable. The authors connect
them with the so-called polar granules, so common in B. coli and related species, and also with
the ‘‘beading”’ often shown by the tubercle bacilli. Furthermore, M. E. Abbott points out
that like B. coli other bacilli, too, may undergo such changes in the body, so that in fact many
of the ‘‘granules’’ occurring within the tissue may be reproductive organs of different bacilli.
Asimilar view has been expressed by Wertheim (1899) in regard to the gonococcus, as well as
to staphylo- and streptococci. The small granules produced by these organisms are considered
to be the reason why occasionally infections may be caused, though no cocci can be found.
N. K. Schuliz (1901) obtained analogous results concerning the plague bacillus. She observed
directly in the hanging drop the upgrowth of the granules into normal rods, as shown on
Plate O in figure 58, which was reproduced from her first paper (1901 a, original fig. I). Other
interesting data have been furnished by Rothert (1902), unfortunately in a not easily accessible
Russian dissertation entitled ‘‘Degeneration and Regeneration of the Bacteria.” It was
pointed out in this paper, probably for the first time, that all bacteria during their period
of ‘‘degeneration’’ produce more or less resistant granules, which are enabled to act under
suitable conditions as organs of ‘‘regeneration.”” The following species were studied, all with
positive result: Staphylococcus pyogenes aureus, Streptoc. pyogenes, B. coli, B. typhosus, B.

‘pneumoniae, B. capsulatus Pfeiffer, B. rubidis, B. arborescens, Vibrio cholerae, V. Milleri; and
104 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

it was especially criticized, with good reason, that too many bacteriologists are always ready to
discard all cultures showing ‘“‘granular decomposition” as dead, without having them first
thoroughly tested. Fedorowitch (1902), another Russian bacteriologist, added important
observations upon the germination of the ‘‘granules’’ produced by several cocci, B. pyocyaneus,
 typhosus, coli, septicaemiae murium, cholerae gallinarum, diphtheriae, and tuberculosis, which
he studied in the hanging drop and for which he proposes the name ‘‘protospores,’’ because they
are, according to his opinion, reproductive organs similar to the regular endospores, though not
reaching the full development and the high resistance of these latter forms. With his Entero-
coccus, Thiercelin (1903) found out that small granules, which the French author calls ‘‘micro-
blastes,” increase in size step by step until they reach again the normal size of the coccus. In
addition, Thiercelin and Jowhaud (1903 a) discovered that the microblasts are produced by the
liberation of the ‘‘granulations périphériques,’”’ which are growing within the cell alongside
with the ‘‘taches centrales,” which participate in the cell division. In some cases ‘‘un fin
pédicule”’ was seen connecting parent cell and microblast. Grassberger (1903) saw and dis-
cussed formation and liberation of similar granules by B. Chauvoei. He describes the process
as ‘‘Abschniirung unreifer Sporenanlagen”’ and makes in this connection the very appropriate
remark:
Gewohnlich werden bei mikrophotographischen Darstellungen sporulierender Bakterien solche Stellen vermieden,

in denen sich derartige losgetrennte Sporenanlagen vorfinden, da sie eine verdachtige, allerdings nur morphologische
Ahnlichkeit mit Kokken aufweisen.

His photograph (original fig. 67) has been reproduced as figure 151 on Plate XI, where
it furnishes 1 good counterpart to figure 152, a copy of a photograph made by Winogradsky
(1902, original fig. 1) from his Clostridium Pastorianum. The coccoid forms are easily dis-
cernible, in reference to which the author says:

An den Faden beobachtet man manchmal einen seltsamen Vorgang, nimlich die Abschniirung von kleinen, kok-
kenartigen Gliedern in missiger Menge, welche, soviel ich weiss, nicht mehr entwicklungsfihig sind.

Experiments with several nonspore-forming bacteria (Pseudomonas cerevisiae, fluorescens
and related forms) led Fuhrmann (1906-1913) to the conclusion that some of the metachro-
matic granules appearing within the cells, especially within the ciub-shaped ones, are repro-
ductive organs to be compared with the endospores produced by the spore-forming bacilli.
That he was not correct in regard to the second statement is beyond question. All bacteria,
spore-forming and nonspore-forming alike, use their gonidia as one mode of multiplication and
reproduction, while the more or less highly resistant endospores have their special task as
resting forms, in which respect they can be replaced by the gonidia only to a very limited extent.
But Fuhrmann was undoubtedly right in regard to the reproductive character of the “gran-
ules’’ studied by him, and it is only to be regretted that also his careful studies have not been
considered and tested adequately. The development of new rods from the granules in the
clubs and their liberation, as studied by him in the hanging drop, prove once more that this
process is by no means so ‘“‘impossible,”’ as it was declared to be by Migula (1904).

That coccoid reproductive organs may play an important part in the life history of the
tubercle bacillus was already indicated by those early observations of RKlebs, Babes, Malassez
and Vignal, Biedert and Sigel, and others, mentioned on page 95, by which it was also ascer-
tained that some of these granules are acid-fast like the bacilli,while others are not, but that
these can be stained by a modified Gram method, worked out by Biedert and Sigel. Schilrmayer
(1898 a), too, noticed that smaller and larger not acid-fast granules may be formed by the
tubercle bacillus, which in their turn reproduce typical rods and are, according to this author,
‘‘undoubtedly spores.’’ To globular bodies of the tubercle bacilli seen by Arrigo, which were
reproduced from his drawings as figures 103 and 104 on Plate IX, are probably also to be classed,
at least in part, as reproductive organs. The Italian author called them “germ granules”
(Keimkérnchen). Chester (1901, page 14) was of the opinion “that these granular particles”
(replacing the tubercle bacilli in old lesions) “are resting bodies of the nature of gonidia, which
are capable of reproducing the species.”” What Spengler (1905-1907) described as fragments
(“Splitter”) of the tubercle bacilli is evidently of the same class; later they were declared by
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 105

him and by Betegh (1908) to be “spores.’’ Much (1907-1909) paid special attention to the not
acid-fast but Gram-positive granules. He emphasized strongly that they never had been seen
before and that they were “entirely” different from the acid-fast globular bodies studied by
Spengler and others. That both statements are not valid can be easily seen from the earlier
publications, as well as from those contributed more recently by Fuchs-Wolfring (1908), Knoll
(1910), Fontes (1910), Kirchenstein (1912), Heinrich (1912), and others. The acid resistance
of these globular bodies naturally changes like that of the rods. Their slipping out of the
sheath and their ability to reproduce small rods, which afterwards grow up to normal size,
has been repeatedly observed by these authors. Fontes’ paper is of special interest, because
the filterability of a part of these “granules” was also studied in this case, and Heinrich
has shown that ‘Much’s granules”’ are equally present in Corynebacteria, B. erysipelatos suum
and other not acid-fast, but Gram-positive species. Rosenblat (1905), too, observed in the
hanging drop, that granules were formed within tubercle, leprosy and other bacilli, which were
liberated after 7-9 days and then grew up to new rods and threads. But later (1911) she made
the following somewhat surprising statement:

Es liegen keine Beweise dafiir vor, dass die Muchschen Granula Entwicklungsformen oder Dauerformen der Tuber-
kelbacillen sind.

Leschke (1911) tried to correct this mistake.

With the diphtheria bacillus and its relatives the situation is very similar; but the con-
tributions upon this subject are less numerous. ‘“Cocci’’ have always played an important
role in the diagnosis of diphtheria, as well as in the cultures obtained in such cases, from the
time of Oertel (1871) until recently (Walker and Adkinson, Mellon, Heinemann, 1917), as was
discussed on pages 77-81. A. Neisser’s work on the “spores” of B. xerosis was mentioned
in this chapter (p. 97). In addition, it may be quoted that Babes in a few cases found heat-
resistant spores with genuine diphtheria bacilli (Cornil and Babes, 1890, Vol. II, p. 59), and
Chester (1901, p. 14) thought it very likely “that the granular segments which they often pro-
duce are of the nature of gonidia.”” Additional interesting data are to be found in a paper
upon the “‘Klebs-Léffler bacillus,” written by Bergstrand (1918).

Coccoid reproductive bodies in leprosy have been also seen quite frequently. Besides
those early discoveries of A. Neisser, G. A. Hansen, Babes and Lutz, the more recent findings
of Betegh (1908) and Kedrowski (1910) are to be mentioned. Both authors studied especially
those only partially acid-resistant forms, which were reproduced in figures 153 on Plate XI
and 154 on Plate XII (from Kedrowskv’s original fig. 38 on Pl. VII). The latter are presenting
an interesting counterpart to the round bodies found by £. de Negri (1916) in serum cultures

‘of Corynebacteria, reproduced as figure 101 on Plate IX, and to similar round cells discovered
by Ghon, Mucha and Miiller (1906) in meningitis, here acting as reproductive organs of the
anaerobic bacilli, shown in‘figure 10 on Pl. C. Figure 155 on Plate XII is a reproduction of
the growth found in cuts from the brain (original fig. 3); figure 156 presents the typical appear-
ance of the regenerative bodies grown anaerobically on glucose agar (original fig. 18); and
figure 157 is a copy of a drawing made by Neelsen as early as in 1880 (original fig. 9 on Pl. XI) of
the regenerative bodies of B. cyanogenes. This author already found out that the round bodies,
which he compares with small yeast cells, are the result of the upgrowth of the gonidia formed
by the rods, that they are either motile or immotile, that they stain very darkly, and that they
are able to reproduce the normal rods when transferred at the right time.

The round bodies visible in figure 158 on Plate XII, photographed by Hammerl (1906,
original fig. 4), belong to V. cholerae. Bittrolf (1912) and Stamm (1914) studied the same
motile, budding globules, which were seen to germinate under suitable conditions to typical
cholera bacilli.

Most extensive studies upon these ‘‘granular” and ‘“‘globular’’ reproductive organs, which
we call the gonidia and regenerative bodies of the bacteria, have been made within the last
decade in regard to the spirochaets. It is evident that so much valuable material was secured
in this case, because the investigators, most of them coming from the protozoological side,
have not been blinded by the adverse bacteriological doctrine, and, therefore, they succeeded
106 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

despite the fact that these modes of reproduction are much more difficult to study with spiro-
chaets than with any of the common bacteria. Dutton and Todd (1905), Castellani (1905),
Wechselmann and Léwenthal (1905), Leuriaux and Geets (1906), Krienitz (1906), Breinl and
Kinghorn (1906), as well as Perrin (1906), have furnished the earliest contributions, establishing
the fact that ‘‘metachromatic”’ or ‘‘chromatine”’ granules produced within the spirochaets may
either multiply as such and replace the spirochaets temporarily in the organism, or eventually
reproduce the long forms, and that not infrequently larger ‘‘bulbs” are visible at the side or
at the end of the spiral organisms, which also seem to act as reproductive organs. That cysts
formed by coiled spirochaets may equally participate in this process was made probable by
the investigations of Breinl (1907) and of Dutton and Todd (1907). Among the numerous
further studies which tried to elucidate these difficult problems those of Leishman (1909-1918),
Balfour (1911-1913), Fantham (1911), Hindle (1911), Noguchi (1911-1912), and Henry (1913)
may be mentioned here. An illustration by Fantham (original fig. 5) has been reproduced as
figure 59 on Plage O. The formation and liberation of the ‘‘spores” as shown here, invites a
comparison with Perty’s old drawing of granulated spirilla visible in figure 42 on Plate L, and
with the liberation of the gonidia by some bacteria as it was drawn by Billroth and reproduced
in figures 44 and 45 on Plate L. Two drawings made by Bosanquet (1911 original figs, 21 and
22) of Spirochaeta anodontae may be added as figure 60 on Plate O. It is interesting to see
how very closely the results of these new studies upon the life history of the spirochaets agree
with those recorded nearly 40 years ago by Heydenreich, Guttmann and Albrecht, which were
mentioned on p. 94.

As figure 159 on Plate XII a photograph of an anaerobic bacillus, probably beloibine
in the Fusiformis group, has been reproduced, which was made by Rosenow and Tunnicliff
(1912, original fig. 3 on Pl. I). Liberated ‘‘granules”’ are well visible, as is also their escaping
and budding out from rods and threads.

That the gonidia of pathogenic bacteria may play an important réle as filterable vira and
cell inclusions has been emphasized by Herzog (1910-1913) with regard to the Gonococcus.
The observations upon the ‘‘infective granules” of the spirochaets in tick fever, and spiro-
chaetosis of fowls, recorded by Leishman, Balfour and others of the authors just mentioned, the
relation between spirochaets and filterable virus in hog cholera, as discussed by Riither (1910)
and King et al. (1913), and the recent discoveries of Rosenow et al. (1916-1917) upon the small
forms of a Streptococcus as causative agent in poliomyelitis, as well as those of Hort and his
collaborators, to be discussed presently, indicate beyond doubt that these problems will become
the more important the more thoroughly the gonidia of the bacteria will be studied.

In some experiments carried out with B. bifidus Noguchi (1910) was able to demonstrate
for the first time that it is possible to induce regular spore formation with otherwise not spore
forming bacilli, and also to cause their reversion to the original type.

The first special information upon gonidia and regenerative bodies (‘‘Regenerationsfor-
men’’), produced by Azotobacter, was furnished by Prazmowski (1912, p. 145, 167), and of still
greater value are additional investigations of the same author (Pragmowski, 1913), wherein he
‘reached the conclusion that probably all bacteria are able to liberate small granules of nuclear
material which act as reproductive organs. The first photographs of the flagellated gonidia of
Azotobacter made by D. H. Jones, have been, reproduced as figure 160 on Plate XII (from
original figs. 9 and 10 in Pl. V); although not clearly defined they illustrate the liberation of the
gonidia from the disintegrating larger cells, and with some difficulty their long cilia may be dis-
cerned. Another photograph, reproduced as figure 161 on Plate XII, was made by Marras-
sini (1918, original fig. 28 on Pl. I); it shows several gonidia of B. su BRE, a few of them
with long single cilia, though the author himself had only in mind to show ‘the capsule and
flagella around the Subtilis rods, apparently not noticing the many round bodies, despite their
predominance.

A casual observation concerning endospore formation in some or our old Azotobacter cul-
tures led to the discovery of the striking pleomorphism of this group of organisms (Léhnis and
Hanzawa, 1914). Copies of three of the photographs published in this paper were given as
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 107

figures 19-21 on Plate II; the breaking up of the larger cells into the smaller ‘‘granules,”’ first
discussed by Prazmowski, is clearly visible in the first two of them. The perfectly round forms
of medium size, which were also shown in one of the photographs of that paper, reproduced as
figure 68 on Plate VI, are to be classed as regenerative bodies. A comparison of this picture
with figure 35 on Plate III (B. pestis, Rowland, 1914), figure 59 on Plate V (B. anthracis, Henri,
1914), figure 91 on Plate VIII (B. mallei, Carpano, 1913), figures 101 and 104 on Plate IX
(Corynebacterium, E. de Negri, 1916, and B. tuberculosis, Arrigo, 1900), figures 154, 156, and
158 on Plate XII (B. leprae, Kedrowski, 1910, the anaerobic meningitis bacillus of Ghon, Mucha
and Miiller, 1906, and V. cholerae, Hammerl, 1906) will make it clear that the surprising uni-
formity of this globular type of regenerative bodies as produced by all groups of bacteria is
certainly a point of considerable importance.

In his ‘‘Studien tiber die Fortpflanzung von Bakterien, Spirillen and Spirochaeten”’ Mei-
rowsky (1914 6) has furnished very valuable information upon the production and development
of gonidia and regenerative bodies by representatives of different groups of bacteria, once more
demonstrating the general occurence of these modes of reproduction. Many details secured by
him, will have to be discussed on the following pages. As figures 162-167 on Plate XII some
of his drawings are reproduced, illustrating what he calls the formation of ‘‘buds”’ and ‘‘umbels”
(‘‘Knospen”’ and ‘“‘Dolden”) by B. tuberculosis (figs. 162-164), B. leprae (figs. 165-166), and by
Spirillum rubrum (fig. 167). But Meirowsky, too, furnishes an excellent example of what dis-
advantageous effect may result from an uncritical submission to the prevalent theory, when he
concludes from his observations, that tubercle and leprosy bacilli should be separated from the
bacteria on account of their forming ‘‘buds”’ and ‘‘umbels,” and should be placed among the
fungi, because, according to his opinion, such modes of multiplication and reproduction are
entirely absent in bacteria; and yet he himself found exactly the same facts with B. paratyphi B,
B. enteritidis, spirilla and spirochaets, and concludes, on the other hand, that the spirochaets,
therefore, ought to be classed among the bacteria. The following quotations will prove this
curious lack of logic.

Das Auftreten von Seitenknospen, Dolden and Seitenzweigen stellt die Tuberkelbacillen ausserhalb der Reihe
der Bakterien und lisst sie zu héheren Pilzformen in Beziehung treten, in deren Entwicklungskreis sie vermutlich
gehéren (p. 11).

Betrachtet man den Leprabacillus vom rein morphologischen Standpunkt, so ergiebt sich, dass er nicht zu den
Bakterien gerechnet werden kann, da bei diesen nur die Vermehrung auf dem Wege der Querteilung und Sporenbil
dung stattfindet (p. 15).

Concerning B. paratyphi B:

Auch hier zeigt sich wieder, dass ein kugliges Gebilde ausserhalb des Bacillenleibes liegen und mit ihm durch
einen feinen Stiel verbunden sein kann; (und es) sind teils gegliederte, teils ungegliederte Faden vorhanden, die an
ihren Enden eine Dolde tragen (p. 15, i. e., on the same page bearing the statement in regard to the leprosy organism),

Concerning B. enteritidis:

Neben ihm lag hiufig ein kugliges oder auch mehr lingliches Gebilde, das in zahlreichen Fallen mit ihm durch
einen Stiel verbunden war . . . Die weiteren Bilder zeigen, . . . dass diese Knospen . . . zu langen

. . Faden auswachsen kénnen. Diese Gebilde weisen haufig . . . kuglige Auftreibungen, Seitenknétchen
und vor allem Seitensprossen und Verzweigungen auf (p. 16)..

Concerning spirochaets:

Es muss . . . betont werden, dass dieser Knospenbildung eine ausserordentliche Bedeutung fiirdie Vermeh-
rung der Spirochaeten zukommt (p. 59).

That, in fact, the processes studied by Meirowsky are quite general with and typical for
all bacteria, and that, therefore, they can not be used as a reason for separating tubercle and
leprosy organisms from other bacteria, has been presented, moreover, by Meirowsky in the
same paper very clearly in a graphic summary (Pl. XVIII), which has been reproduced as
Plate XIV.

Evidently without knowledge of this German publication Hort in England published a series
of reports (Hort et al., 1915-1917), which on the one side furnish much confirmative material
with regard to Meirowsky’s and Almquist’s discoveries, and on the other side open a wide new

field for further research upon pleomorphism and reproduction of the bacteria, especially in
108 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor. XVI.

connection with the so-called filterable vira. On Plate XIII as figure 178 a collection of draw-
ings made by Meirowsky (original plate III), illustrating the various phases in the life history
of B. paratyphi B, has been placed side by side with two pictures made by Hort (1917a, origi-
nal figs. 1 and 2) of B. typhosus (fig. 179), and of B. coli (fig. 180), which deserve to be studied
closely. The remark made by Adami (1916) after having become acquainted with Hort’s
investigations, which was quoted on p. 10, will present itself as being, indeed, very appropriate,
when considering the many data discovered or rediscovered. recently with regard to bacterial
reproduction, which have always been discarded by the ‘‘simplicity’”’ doctrine, still widely
prevailing among bacteriologists and in bacteriological literature. With the meningococci
Hort (19176) once more discovered that type of multiple production of gonidia within the
enlarged parent cell, which was declared by Migula to be ‘‘impossible,” though, in fact, it has
been studied by many investigators and is so common with all groups of bacteria that Hort
undoubtedly was mistaken when he thought the Meningococcus must be placed among the
Hemiascomycetes on account of its ‘‘gemmation and multiple endosporulation.”’

A short summary of our own investigations upon the different modes of bacterial repro-
duction (Léhnis and Smith, 1916a) was quoted on p. 90, and several details will have to
be mentioned on the following pages. Some of our photographs, which were demonstrated
at the New Haven meeting of the ‘‘Society of American Bacteriologists’’ (19165), are repro-
duced as figures 168-176 on Plate XII, illustrating formation and liberation of gonidia of Mier.
candicans (fig. 168) and 1. luteus (fig. 169), development of the regenerative bodies by Streptoe.
lactis (fig. 170), B. pneumoniae (fig. 171), and B. bulgaricus (fig. 172), germination of regen-
erative bodies of Azotobacter Beijerinckvi (fig. 173), development of new small organisms from
the gonidia while still connected with the parent cell (fig. 174, B. fluorescens; fig. 175, Azotob.
vinelandii) and a star-like upgrowth of small rods from an agglomeration of unstainable gonidia
liberated from the large darkly stained cells of Azotob. chroococcum, visible in the lower part of
the picture (fig. 176). The many relations to and confirmations of earlier observations quoted
on the preceding pages are evident. And it is equally obvious that the various forms visible
in figure 177, which is a reproduction of a photograph made by Goadby (1917, original fig. 14)
of a bacillus isolated from gas gangrene, which could not be identified ‘‘owing to the curious
method of growth,” are an interesting collection of gonidia, regenerative bodies and endospores.

It goes without saying that, like the pleomorphism of the bacteria in their vegetative
stage, so also their different modes and organs of reproduction will necessitate much more
thorough investigations than those generally en vogue now as at Robert Koch’s time. It is
easy, but not scientific, to get quickly rid of all unexpected facts by declaring them to be invo-
lution forms, contaminations, artefacts, plasmolysis, plasmoptysis, etc., without making any
special test. Such special tests, however, are of course indispensable if correct results are
to be obtained.

(b) DIFFERENTIATION BETWEEN REPRODUCTIVE ORGANS AND OTHER CELL PRODUCTS AND ARTEFACTS.

Undoubtedly, the reproduction of a new bacterium is the only fact by which it is possible
to decide with absolute certainty whether or not a problematic body is a reproductive organ
of a bacterium. The observation of active motility often shown by the gonidia, and some-
times by regenerative bodies, too, furnishes also fairly conclusive proof. In many cases, how-
ever, the examination of the stained preparate must be used as basis for judgment, and here
the road is beset with numerous pitfalls. Thus far many investigations have been made upon
the bacterial cell and its content, without paying any attention whatever to the possibility,
that some of the cell products might have been reproductive organs other than endospores
in different stages of their development. It is not only probable, but even certain to some
extent, that among the cell inclusions described as nuclear material, chromatin, metachro-
matic granules, volutin, glycogen, fat and other reserve material, as well as vacuoles, growing
gonidia and regenerative bodies have been included, because their staining reactions, changing
in the course of their development, have established an apparent relationship to the one or to
the other class of cell products. Detailed studies, as have been made in this respect with
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 109

growing endospores, are nearly entirely missing at the present time with regard to the other
bacterial reproductive organs. Nevertheless, the cytological results collected so far also
contain several valuable indications, which may be helpful for further research; therefore,
they will have to be considered first.

The time is not far back when there was much discussion concerning the presence or ab-
sence of a nucleus in the bacterial cell. The results available now, though by no means final,
can be accepted, however, as sufficient evidence that arrangement or distribution of the nuclear
material changes during the life of the bacterial cell and varies to some extent with the different
species. Uniform distribution, similar to the ‘‘chromidia’’ in the protozoa, arrangement in
circles, spirals, zigzags, rods, chains, or filaments, as well as concentration into one or two dis-
tinct bodies, displaying the qualities of true nuclei, have all been observed within the same cell
and with representatives of very different groups of bacteria. Publications by Zukal (1896),
Vejdovsky (1900-1904), Mencl (1905-1911), Perrin (1906), Swellengrebel (1906-1909), Gusi-
liermond (1908), Raditka (1908-1909), Hélling (1910), Dobdell (1911), A. Meyer (1912), Praé-
mowski (1912), and Luska (1914) have accumulated much valuable material in this respect,
against the older widely divergent views held by Biitschli (1890-1896), A. Fischer (1897-1903),
and Migula (1897). It can not be further doubted that nuclear material plays the same
important réle in regard to growth, division, and multiplication of the bacterial cells as it
does in the cells of the higher organisms. That it will always participate in the formation
of reproductive organs is equally beyond doubt, despite the scantiness of experimental proof
available at present.

Some interesting observations concerning the production of zoospores have been made
by Zukal (1896) in connection with his cytological studies. He noticed that the nuclear
granules, which he calls ‘‘microsomes”’ and which he found to be frequent in bacteria, protozoa,
and Cyanophyceae, are able, especially in the latter case, to assume under certain circum-
stances the character of regular zoospores. That the chromidia of protozoa may reproduce
new cells has been recorded very often. Motility, fission, and participation of ‘chromatin
granules” in the division of the bacterial cell have been studied by Babes as early as in 1885.
Their predominant part in the formation of bacterial endospores has become known especially
by the investigations of Zetinow (1899), Schaudinn (1902-1903), Guilliermond (1908), Rasiéka
(1908-1909), A. Meyer (1912), and Swellengrebel (1913). That the spores of Actinomyces
are also of nuclear character was pointed out by Neukirch (1902), and the granules (i. e. gonidia)
of the tubercle bacilli were declared by Fontes (1910) to be made up prominently of para-
nucleo-albumin. The so-called sporoids, replacing sometimes the regular endospores in spore-
forming bacilli, are also chemically equivalent to them, as far as can be derived from micro-
chemical analysis, according to Rizitka (1908-1909) and Petschenko (1913). The formation
of the so-called spores (i. e. gonidia) of Spirillum volutans went on, as was pointed out by Amato
(1908), practically in the same manner as the formation of the endospores in B. mesentericus,
subtilis, and mycoides. That some of the nuclear material may leave thé old cells has been
noticed by Rowland (1899) and by Menel (1911), but it is to be doubted whether this is merely
an excretory process, as these authors are inclined to believe, or whether these liberated granules
have not been actually gonidia. Probably the work done by Praémowski (1912-1913) will
prove most valuable as a basis for further investigations in this direction, as here the reproduc-
tion of new cells from the nuclear granules has been studied inside as well as outside of the
parent cell. In addition some very suggestive work done with yeasts also deserves our atten-
tion. Hisenschite (1895) collected valuable data concerning the staining qualities, motility,
division, and arrangement of nuclear granules in yeast cells and about their participation in
budding; and Hest (1907) contributed a highly interesting article on ‘““pseudo-vacuoles” and
“‘pseudo-nuclei” in yeasts, showing that the frequently mistaken ‘‘granules” in this case too
acted clearly as reproductive organs, very much like the bacterial gonidia. Some of his draw-
ings and photographs could be readily accepted as illustrating analogous processes in the life
of Azotobacter.
110 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

The chromatin granules have usually attracted most of the attention of the various inves-
tigators, whose results, however, diverge widely, a fact not difficult to understand. With
regard to the higher organisms the chromatin has been declared to be ‘‘the vehicle in which
the constitutional structure, primarily of the species and secondarily of recent ancestors and
parents is represented’’ (Spencer, 1898, p. 264). With the bacteria at least in the case of endo-
spore formation the concentration of the chromatic substance has been always noticed, though
especially in ripe old spores the presence of other substances may occasionally greatly affect
the microchemical reactions, which fact, studied especially by Zettnow (1899), has misled some
authors like Ri&gitka (1913) to believe that the chromatin ‘‘vanishes”’ in old spores. Again,
with regenerative bodies and gonidia the analogous change in stainability may occur, though
it is generally less frequent than with endospores. Growing gonidia and regenerative bodies,
however, will always, like growing endospores, exhibit the chromatin reactions very clearly.
Thus far the chromatin granules present in the bacterial cell have been accepted by the various
authors as being of very different physiological value, an opinion which is undoubtedly justified
to some extent, but which at the present time is usually much more based on assumption than
on evidence. This is particularly true in regard to such cases where the ever-ready hypothesis
of ‘‘involution forms’’ has been invoked indiscriminately. Beck (1903, p. 777), for instance,
does not hesitate to declare:

So viel steht fest, dass die Differenzierung des Plasma des Bakteriums und damit auch die Bildung der chro-
matischen Substanz in demselben mit der Vermehrung und Fortpflanzung des Bacillus nichts zu tun hat.

But in the same ‘‘Handbuch,” which contains this highly surprising statement, Gotschlich
(1903, p. 83) connects the granules with an increased resistance of the cells and Wladimirof
(1903, p. 722) is certain about their reproductive character. A. Fischer (1897, p. 122; 1903,
p. 8) classes them as reserve material; to Migula (1897, vol. I, p- 89) they are the beginning
of nucleus formation. Swellengrebel (1907) saw round bodies of a ‘‘very complicated inner
structure” develop in the center or at the end of the cell, but they, too, are to him ‘‘merely
involution forms.’ Petschenko (1913) distinguishes those chromatin parts which are becoming
“‘sporoids”’ (gonidia), from others participating in the metabolism of the cell, by the fact that
the latter according to circumstances may increase and decrease in size, while the former show
only an upgrowth, a fact evidently not easily to be ascertained and probably not always valid.
Whether an observation made by Biitschli (1896, pp. 40-43) will be a helpful hint for recog-
nizing growing gonidia, remains to be seen. This author noticed that the centrally located
chromatin granules turned blue when stained with methylen blue, while those at the periphery,
though reacting with Delafield’s hematoxylin like those in the center, exhibited metachromatic
qualities, becoming red with methylene blue. It is not impossible that there may be some
relations between these findings and those recorded by Thiercelin (1903) concerning the ‘‘taches
centrales” and ‘‘granulations périphériques,’’ mentioned on p. 104, though it would be undoubt-
edly a serious mistake to assume that gonidia and regenerative bodies will always originate
close to the cell wall; there is sufficient evidence on hand that at least the growing gonidia are
often to be found in a central position and their diameter is not infrequently as large or even
a little larger than the width of the parent cell.

The polar granules have been frequently looked upon as being enabled to assume a repro-
ductive character, though the results obtained have been much at variance. So it has been
the case with B. typhi, coli, V. cholerae, and V. Finkler-Prior, as may be seen, for instance, from
the papers by Ermengem (1885), Funkler and Prior (1885), Buchner (1888), Adami, Abbott
and Nicholson (1899), and by Escherich and Pfaundler (1903). That in the textbooks the
polar granules are usually classed as ‘‘merely degenerative” or ‘‘simply artefacts” is in ac-
cordance with the generally prevailing practice; and it is, of course, indisputable that rough
handling of dried smears may cause the appearance of plasma clumps looking like polar gran-
ules, but not having their physiological value, and, therefore, furnishing no proof whatever in
regard to the meaning of those polar granules, which are present in the living cell. N. K. Schultz
(1901 a) made some interesting observations concerning the reproductive function of the polar
granules in old cells of the plague bacillus, as may be seen from her drawings reproduced as
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 111

figure 61 on Plate O (from original fig. Il), illustrating the development of the polar granules
to a uniformly distributed cell content. Baur (1905) has shown that polar granules participate
in the formation of the arthrospores produced by Myxococcus; and that they are equally con-
nected with the development of endospores has been frequently observed. Their independent
acting as reproductive organs, however, remains to be studied more thoroughly, before any
well-founded answer can be given.

The metachromatic granules, which Babes (assg) first found to be frequent with all kinds
of bacteria, especially in older cultures, have been characterized in that paper, besides by their
being stained violet by methylene blue, by their ability to reproduce new vegetative cells and
by their acting in this direction as counterparts to endospores, though not exhibiting such an
exceptional resistance against heat and drying. In a second paper (Babes, 1895) they were
declared to participate in cell division, as well as in the formation of spores, buds, and branches;
and in a third contribution (Babes, 1914) the so-called Much’s granules of the tubercle bacillus
were identified with them, and their réle as reproductive organs was once more emphasized.
To A. Fischer (1897) their metachromatic behavior was merely caused by the assumed fact
that the accumulation of the stain within was said to prevent the passage of light, so that only
the reddish reflex characterstic of methylene blue remained noticeable. Guillermond (1910), .
who next to Babes studied the subject most intimately, contests this author’s interpretation,
as well as that preferred by A. Meyer (1903), who identifies the metachromatic granules with
his so-called volutin. According to the French author, metachromatic granules are produced
not only by all bacteria, but also by all protozoa, yeasts, other fungi, as well as by higher organ-
isms; which statement leaves hardly any doubt, that, in fact, rather different objects have been
grouped together. If, therefore, other authors like Jordan (1909, p. 58), Frost and McCampbell
(1910), and Dobeil (1911) assume that the metachromatic granules are some kind of reserve
material, but do not participate in the life of the cell, though Fedorowitsch (1902) and Fuhrmann
(1906) have demonstrated that growing gonidia, regenerative bodies, as well as spores, also
exhibit, at least temporarily, metachromatic reactions, such conflicting statements are, indeed,
not so irreconcilable as they may seem to be at the first moment. The large metachromatic
globules, which Hélling saw in old spirilla cultures and which he interpreted as volutin, de-
spite their ability to divide themselves, will also have to await a more correct classification. The
same holds true concerning those metachromatic granules which were found in and outside of
‘‘Mastzellen’’ and were classed as metabolic products by Bargum (1903) as well as by Guilliermond
and Mawas (1908). That some of them may be gonidia or regenerative bodies of bacteria is
just as well possible as was the case, e. g., with the round bodies seen by Hibler in and outside
of ‘‘Mastzellen” accompanying the bacilli of symptomatic anthrax, and also in the pure cultures
of this organism, as was discussed on p. 67. Ross (1912) pointed out specifically that the
large spherical bodies studded with chromatin granules, which are produced by spirochaets
(as well as by other bacteria in their symplastic stage) can be easily mistaken for mast cells.

The granules first described by P. Ernst (1888), and usually called accordingly, have been
often identified with Babes’ metachromatic granules, for instance by Biitschli (1890), A. Fischer
(1897), Marz and Woithe (1900), Kamimura (1903),. Bonazzi (1915), and others, though their
solubility in boiling water differentiates them from Babes’ granules, whose relation to repro-
ductive organs is certain, at least to some extent, as with the granular form of tuberculosis,
mentioned above. When Coppen-Jones (1895) maintained that Ernst’s granules be growing
“spores” of the tubercle bacillus, he also evidently meant Babes’ granules, and the great dif-
ference which exists between the claims made by Marz and Woithe (1900) concerning the réle
played by the ‘‘Babes—Ernst” granules in the life of pathogenic bacteria, and the results
recorded by their opponents like Picker (1903), is due undoubtedly to the study of different things,
all included under the same conventional term. It is true that P. Ernst himself in his first
paper (1888) dealt prominently with B. zerosis and its sporeformation, as studied by A. Neisser,
and in his second paper (1889) he calls the inclusions directly ‘‘sporogenous granules,” though
evidently many of them, as described and illustrated in the paper, have nothing to do with the
formation of spores or other reproductive organs; in a more recent paper (P. Ernst, 1902), too,

160266°—21——8
112 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou, XVI.

some of the bodies show again marks of having been gonidia (budding, fission, and motility after
liberation). Nevertheless, it has become such a general practice to ascribe to “‘ Ernst’s granules”
solubility in boiling water, since Bunge (1895) and Miihlschlegel (1898) have differentiated their
‘‘sporogenous” granules from them in this way, that it will be preferable to adhere to this
practice, and to use this method for differentiating between soluble and insoluble metachro-
matic granules, and for obtaining eventually more information upon their relations to the va-
rious reproductive organs.

That the staining reaction not infrequently changes during the development of the gonidia
and regenerative bodies, though usually not to such an extent as with the endospores, has been
frequently noticed in the course of our studies. While usually well accessible to aqueous dyes
especially to fuchsin, gonidia as well as regenerative bodies sometimes become hard to stain,
® point which has been studied to some extent by Kuntze (1904) with regard to the gonidia
(‘‘Schwarmer’’) of B. oxzalaticus. In most cases, however, the comparatively very dark color
of the stained regenerative bodies is very conspicuous, as may be also seen from the photo-
graphs reproduced on several of our plates. In addition, aqueous fuchsin often causes a
peculiar tint of red with regenerative bodies, which is well discernible after sufficient training
of the eyes.

As no sharp microchemical differentiation is possible at present, and possibly never will
be, between chromatin and metachromatic granules on the one side and growing reproduc-
tive organs on the other, so also fat, glycogen, and other so-called reserve material will be used
in smaller or larger quantities by the bacterial cell for building up its reproductive organs, and,
therefore, the more or less characteristic reactions of these substances will be also noticeable
with growing gonidia, regenerative bodies, and spores. It would be, of course, a serious mis-
take to draw rashly final conclusions from such tests, and it should also be constantly borne
in mind when making microchemical experiments with bacteria, that these reactions are by
no means so prompt and reliable as one might be inclined to believe after having read, for
instance, the books of A. Meyer (1903-1912). On the other hand, it would be equally unwise
to share such an extreme standpoint as was taken by A. Fischer (1899), who felt compelled to
fight against the ‘staining mania” (‘‘Farbungstaumel’’), mostly because he got all kinds of
reactions and artefacts under conditions which were rather different from those prevailing in
bacteriological tests. The following statement made by Kruse (1910, p. 48), however, may
be accepted as fairly correct:

Mantraut . . den meisten mikrochemischen Identifizierungsverfahren doch zuviel zu.

Fat has always taken the most prominent place among the substances used for “‘explain-
ing” the presence of various granules within the bacterial cell. R. Koch (1876) was quite
positive that the Anthrax spore be made up of a bright droplet of some kind of oil, covered
by a thin protoplasmatic layer. And at present often a staining with sudan, etc., is considered
to be sufficient evidence to establish the presence of “fat droplets” inside or outside of the
cell, though it may have been gonidia containing some fat, but, of course, also all other
substances necessary to a continuation of bacterial life. Béisgen (1894) felt completely justi-
fied to reject the findings of Zopf (1881) and Billet (1890) concerning the globular gonidia
produced by Cladothriz dichotoma, because he was ablé to “solve” these “fat granules” by
treating them with hot alcohol. In fact, however, he merely extracted some fat from what
he calls a “vacuoliges Protoplasmageriist,” and the direct observations, made by the earlier
authors upon the germination of these bodies, are, of course, also not eliminated by this extrac-
tion. The manner in which Henneberg (1901) classed as ‘‘fat droplets” what has been evi-
dently the very conspicuous regenerative bodies of lactobacilli, was criticized on page 103. In
publications of P. Eisenberg (1909) and Vahle (1909) the ‘‘fat”’ hypothesis plays an equally
- prominent réle. Kruse (1910, p. 48), too, accepts the fat reactions as full proof that, e. g.,
the so-called sporogenous granules not only contain some fat, but that they are nothing else
than ‘fat droplets.” On the other hand, Altmann (1894, pp. 101-104) has already pointed
out that the cell granules which participate actively in the cell life of the higher organisms
may also assimilate considerable amounts of fat without reducing their vital activity even if
No.2.) LIFE CYCLES OF THE BACTERIA—LOHNIS. 113

they appear micorochemically like large fat drops. With Blastomycetes, Casagrandi (1897)
found out that such apparent fat granules react like nuclein and protein substances after the
fat has been extracted by alcohol. The ‘‘swarming bodies” of B. oxalaticus have been tested
by Kuntze (1904) also with the result, that they may contain, but that they are not, fat; and
the same has been said concerning similar “sporoids” by Réditka (1909), Vay (1909), and
Ambroé (1909), especially against P. Hisenberg. The protein reaction has been quite evident
as far as tests have been made in this direction. That especially the granules of the tubercle
bacilli contain albuminous together with fatty substances was ascertained by Deyke (1910)
and confirmed by Babes (1914), who compares them also in this respect with the endospores
of the bacilli.

Besides fat, according to A. Meyer and his pupils, volutin is to be credited with playing
an important réle as reserve material in the bacterial cell. The fact, however, that it is de-
clared (A. Meyer, 1912, p. 204) to be absent in spore-forming bacilli, Pseudomonas, Strepto-
coccus and Sarcina, does not support this view, and as this “‘volutin” admittedly belongs to
the protein substances, more thorough tests of the so-called volutin globules especially in
spirilla seem to be necessary; they bear a suspicious resemblance to those other globules which
were found by some authors to be without any significance, while others have seen them ger-
minate quite readily and grow up to new spirilla. Dietrich and Liebermeister (1902), as well as
Riézctka (1908), have already critized the volutin hypothesis. Prazmowski (1912, p. 157), too,
calls this substance “highly problematical;” and Minchin (1915) is convinced,that the ‘“in-
fective granules’’ (i. e. gonidia) of trypanosomes (spirochaets) have been frequently mistaken
as ‘‘ volutin.”

The “granules” and ‘‘globules,’”’ visible in and outside of the bacteria, have also been
often readily accepted as glycogen, merely on account of the brown color sometimes exhibited
by them when treated with iodine. That this reaction is by no means very stable has been
frequently noticed. The round buds (‘‘Aussprossungen”’) of B. radicicola, e. g., showed, accord-
ing to Hiltner and Stérmer (1903), often but not always glycogen reaction. The inflated forms
of B. Chauvoei were found by Hibdler (1908) to turn brown, violet, or yellow. Similar incon-
sistencies have been discussed by Prazmowski (1912, p. 158) with regard to Azotobacter. Kuntze
(1904) noticed that the swarming bodies of B. oxalaticus turned yellow at the outside, brown
in the center. To assume that bodies exhibiting the glycogen reaction be nothing else than
glycogen, would be equally erroneous as to class them as “merely fat droplets,”’ because they
behave in a similar manner.

Menel (1910) preferred to interpret the brown color, caused by iodine in buds of Azoto-
bacter, as indicating amyloid, and rejected the assumption that they are plasmatic substances.

The coccoid bodies of sulphur bacteria, whose motility and multiplication had been studied
by Zopf, were claimed by Winogradsky (1888) to be nothing else than “sulphur granules,” an
assertion which has been sufficiently answered by Zopf (1895). Concerning the blastia of Perty
(1852), Migula (1897, Vol. I, p. 8) made an analogous, equally unfounded statement.

The “‘Schwebekérper” or ‘‘Aérosomen,”’ attributed by Molisch (1907) to Rhodobacteria
and Phycochromaceae, will also have to await renewed critical study.

The astonishing properties of the ‘‘vacuoles,’”’ discovered by Zettnow (1896-1897) inside,
as well as outside, of various spirilla, have been mentioned on page 99; and the interesting inter-
pretation given by Hest (1907) in regard to the ‘‘pseudo-vacuoles” of yeast cells, was quoted
on page 109. A. Fischer (1903, p. 7) was of the opinion that deeply staining inclusions in spirilla
were to be explained as vacuoles, wherein the stain became accumulated, while other investi-
gators usually have interpreted those parts of the body to be vacuoles, which were only weakly
stained or not at all. :

Some special réle in oxydative processes going on in the cell has been attributed to granules
present within the bacterial cells by Dietrich and Liebermeister (1902), as well as by Brandt
(1913). If this hypothesis should be correct, the fact itself would not militate against an
eventual participating of such granules in the processes of multiplication and reproduction.
114 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

All cell inclusions mentioned above, part of which will undoubtedly be recognized as identical
with gonidia or with regenerative bodies, when thoroughly studied, have been classed as the re-
sult of either the normal or the abnormal metabolism of the cell. The “fatty degeneration”
especially plays a conspicuous réle with many authors, despite the rather insufficient basis on
which this hypothesis rests. There are numerous other cases, however, where it also has been
considered ‘to be quite sufficient, simply to state that the objects seen were ‘‘merely involution
forms,’’ and to abandon all further research upon them. With regard to the vegetative cells
it has been shown in the first chapter (pp. 24-29) to what extent this term has been misused very
frequently, so that it is not necessary to dwell upon this point once more. Only this may be
emphasized again, that among the reproductive organs the endospores most clearly display all
marks of true involution, but they hardly ever have been classed as such. They too show,
however, that as in many other cases, so also here, involution may be followed by evolution,
and the same holds true for bacterial reproductive organs generally. That, for instance, the
sporoid bodies in vibrios and spirilla should not be laid aside as ‘involution forms,” as was
done by Doyen (1885) and others, has been pointed out by Weibel as early as in 1888. He says
directly:

"Es bedarf fast einer gewissen Selbstiiberwindung, sie als Sporen zu verleugnen.

That the exceptional resistance, often, though not always, exhibited by endospores, is not
to be found with the other reproductive organs, has evidently also caused some misunderstand-
ing. Woodhead (1891), for instance, makes the following remark concerning the so-called
arthrospores of the cholera bacillus, notwithstanding their ability to germinate:

As any apparent spores that have been formed invariably failed to resist the action of drying, such ‘‘arthrospores”
containing bacilli must, for the present, be looked upon as involution forms.

The same standpoint was taken more recently by Hiss and Zinsser (1914, p. 16), as well as
by Kendall (1916, p. 30). Wertheim (1899), who first observed the breaking up of the Gonococ-
cus into minute coccoid bodies, was careful enough to test their further behavior and was able,
therefore, to find out that also in this case the ‘involution’ was followed by a new evolution
of typical cocci. Swellengrebel (1907), on the other hand, classed and discarded the round bodies
within spirochaets as ‘‘involution forms,’’ though he saw them growing and noticed their ‘very
complicated inner structure, ’’ both undoubtedly no marks of degeneration. With regard to the
“‘conidia’’ (i. e., gonidia and regenerative bodies) described by Almguist, Gotschlich (1909) made
the unique remark that they can not be accepted as reproductive organs, because they do not
have “einen bestimmten, streng eingehaltenen morphologischen Character’; an interesting
proof to what extremes an author may be misled by strict adherence to the monomorphistic
doctrine. That the German author, in fact, erred in making this statement, is sufficiently
proven by an inspection of Almquist’s drawings, reproduced as figure 54 on Plate N, and of his
photographs, to be found on Plate XI as figures 135-139. The form of these reproductive organs
is at least as distinct as that of the vegetative cells, and it has been emphasized already
that especially the round regenerative bodies of very different species may assume a vexatiously
uniform appearance. But even if Gotschlich’s claim would be right, his remark would still have
to be rejected, because the true character of a reproductive organ is always determined by its
physiological behavior, viz., by its ability to reproduce vegetive cells, but never by its mor-
phology. ,

The term ‘‘granular decomposition” has been often used as affording another chance to
bring bacteriological researches to an early end. The process was described by Hueppe (1886,
p. 108) as “‘retrograde metamorphosis’’; the granules were declared by him to be entirely unable
to enter on a new development, and it was pointed out by him specifically that they should not
be put parallel to gonidia. Less dogmatic, but more correct, Cornil and Babes wrote in their
textbook (1890, Vol. I, p. 33):

On peut constater, 4 un moment donné dans la chambre humide, que des bactéries de diverses espdces se trans-
forment en granulations qui resemblent 4 des microbes ronds. Mais souvent ces granulations ne sont plus vivantes
ni inoculables,
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 115

The experiments made by Pfeiffer (1894) and by Pfeiffer and Kolle (1896) upon the granular
decomposition of cholera and typhoid bacilli in immunized guinea pigs, as well as those of Hm-
merich and Saida (1900) upon the analogous behavior of anthrax bacilli, when treated with
pyocyanase, have especially created the widespread belief that all ‘“‘granular decomposition”
without exception be indicative of the death of the bacteria. However, already in 1898 Can-
tacuzéne pointed out that the granules in Pfeiffer’s experiment are, at least sometimes, living
and able to propagate; and he also discovered the interesting fact that only living, not dead,
organisms showed this kind of transformation. Bacilli killed by heating were still agglutinated
within the animal, but were otherwise unchanged after 3-4 hours. Kohlbrugge (1901 a) noticed
that hemoglobin stimulated the granular transformation of V. cholerae, and he also observed the
reproduction of regular bacilli from these granules. Gotschlich (1903, p. 58), too, admits—
dass die sonderbaren kérnigen Degenerationsproducte der Cholerabacillen beim Pfeifferschen Phanomen, sowie der
Pestbacillen in Bubonen . . . eine Zeit lang ihre Lebens- und Regenerationsfahigkeiten in Kulturen behalten.

That the same may not only happen in cultures but also in the organism, was indicated by
an observation made by Nakayama (1906) in animal tests with Actinomyces asteroides, where a
new growth of threads started from the granular detritus within the phagocyts. Much (1909)
emphasizes with regard to the dissolution of bacteria within the leucocyts:

Auflésung ist nicht gleichbedeutend mit Abtétung . . . Bringt man Pneumokokken mit einem agglutin-
ierenden Serum in Beriihrung, so quellen sie auf und verlieren volkommen ihre Farbbarkeit. Trotzdem sind sie
aber nicht abgetotet.

The breaking up of Spirochaeta gallinarum into granules, when kept in fowl’s blood, is
another instance where granular decomposition does not mean that life is at its end. Accord-
ing to investigations made by Hindle (1911) and others, these granules are undoubtedly repro-
ductive organs, though Swellengrebel (1912) tried to apply the volutin hypothesis in this case,
too.

A very thorough discussion of the whole subject has been published by Herzog (1913),
who once more confirmed that ‘‘granular decomposition’’ and ‘‘partial bacteriolysis” of gono-
cocci, plague, cholera, hog cholera, and other pathogenic bacilli within the body is by no means
always equivalent to their ultimate destruction; the granules are eventually able to reproduce
a new generation of typical bacteria. It may be added that Simonini’s (1914-15) studies upon
the influence of certain elements (lantham, cer, thorium) upon B. coli, typhi, dysenteriae, cholerae,
subtilis, anthracis, diphtheriae, meningococci, gonococci, etc., have also furnished several inter-
esting details, which may become helpful for further work upon the meaning of this ‘‘granular
decomposition.” Here the granules once more have proved themselves to be able to act as
reproductive organs, though the process of regeneration was often greatly. delayed, and not
infrequently such properties as virulence, motility, gelatine liquefaction, were absent for a long
time in the new cultures, developed from the granules.

Closely related to the facts just mentioned are other observations like those recorded
in 1881 by Fokker and in 1883 by Archangelski concerning the temporary replacement of an-
thrax bacilli by coccoid bodies within the organism. Vignal (1889) noticed that in artificial
cultures, too, and already from the second day on, some of the rods of B. mesentericus began to
contract their plasmatic content into similar granules, leaving vacuoles in the cells, which fact
was interpreted by the French author as follows:

Tl est impossible d’attribuer cet état 4 un appauvrissement du milieu nutritif ou a la formation de matiéres nuisibles,
car & cété des bacilles vacuolisés on en trouve d’autres qui se divisent énergiquement.

_ Many authors, however, were only too willing to declare a priori any formation of granules
to be degenerative, while only few were inclined to admit that degeneration might be followed
by regeneration and both processes are part of the normal life cycles of the bacteria.

__ The following remark made by Chester (1901, p. 14) with regard to the granules in Mycobac-
teria, which he already correctly classed as gonidia, holds true for many analogous cases:

Certain bacteriologists have considered many of these so-called gonidia as degeneration forms; but it is more likely

that they are distinct morphologic elements, inasmuch as degenerative elements could not be expected to produce new
vegetative cells.
116 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

On the other hand, it was a fundamental mistake when  Wladimiroff (1903 6) declared the
granules in spirochaets to be merely degenerative, ‘‘because’”’ they are not resistant against
heating. Gotschlich (1903, p. 43), Kruse (1910, p. 31), as well as Lehmann and Newmann (1912,
p- 517), also classify this granula formation as ‘‘degeneration,”’ though they admit at least the
possibility of a succeeding regeneration.

The terms ‘‘segmentation’’ and ‘‘fragmentation’’ have been used occasionally, especially
in regard to the formation of arthrospores by Actinomycetes. Kruse (1896 a, p. 55) wanted
to have both terms sharply separated:

Nicht zu verwechseln mit der Segmentierung, durch welche lebensfihige, normale Elemente geschaffen werden,
ist der unregelmiissige Zerfall von kiirzeren und lingeren Bakterienzellen in ungleiche und oft abnorm gebildete
Teilstticke, die Fragmentierung, die in alten Kulturen zu beobachten ist.

+ » But he defined himself the fragmentation in the following manner:

Es ist das offenbar ein regressiver Vorgang, der hier nur erwihnt sein mag, weil die Méglichkeit nicht ausge-
schlossen werden kann, dass unter giinstigen Umstinden aus dem Zerfall noch lebensfahige Keime hervorgehen, die
sich durch eine Art von Ver jiimgungsprozess . . . zu normalen Elementen regenerieren kénnen.

Both processes, therefore, are reproductive; and already Zopf (1883, pp. 10-12) has described
the fragmentation of bacteria as ‘‘some kind of propagation,” placing it parallel to the
production of the so-called hormogonia by fission-algae. Accordingly, both terms have been
used in the bacteriological literature, especially in papers on Actinomycetes, practically synony-
mously, as may be seen from the publications by Lachner-Sandoval (1898), Neukirch (1902),
Gilbert (1904), Lepeschkin (1904), Haas (1905), Schiitze (1908), and Lehmann and Neumann (1912,
p. 623).

Riéitka (1907) has correctly emphasized that the regeneration of normal bacteria from living
fragments has so far not met with adequate attention, despite the important réle played by the
analogous process in the life of other primitive plants and animals.

In the same manner as the terms ‘‘degeneration,” ‘‘involution,” ‘‘reserve material’”’ or
other ‘‘cell inclusions’ had to serve undoubtedly much too frequently for ‘‘explaining’’ the
observations made upon various types of reproductive organs and for uniting them with the
rigid monomorphistic doctrine, which allowed only one constant form of the vegetative cell
and only one mode of reproductive organ, the endospore, as studied by R. Koch and his pupils,
so also the genuine or merely hypothetical artefacts have been often invoked to allow a quick
discarding of unexpected facts. 7

Among the so-called artefacts those introduced by A. Fischer (1891, 1900) as being the
products of plasmolysis and plasmoptysis have been widely accepted by bacteriologists as really
important facts. Though A. Fischer’s claims were supported by only very little and obviously
unsatisfactory experimental work, and though only very few confirmative, but many contradic-
tory, results have been recorded since then, especially the plasmoptysis, characterized by A.
Meyer already in 1905 as an ‘‘outgrowth of A. Fischer’s vivid imagination,” i is still treated by
writers of bacteriological ‘fentbouks as a very important fact. There is no doubt, of course,
that bacteria like other cells may be induced by a hyperosmotic environment to contract their
cell plasma, i. e. to exhibit so-called plasmolysis; and it is equally beyond question, that a sudden
change from high to low pressure may cause an increase of-the inner tension to such an extent,
that the cell wall breaks and part of the content is forced out with more or less vigor. A Fischer
(1900) himself, however, has practically eliminated this osmotic basis of his theory by reporting
first that plasmoptysis may occur with some but not with all cells, when they are transferred
from alow to a high concentration, and, second, that plasmolysis and plasmoptysis may occur
simultaneously in the same single cell (A. Fischer, 1900, p. 10). Accordingly, Kolle and Hetsch
(1911, p. 30) readily accept both terms as correct explanations of all kinds of processes going
on in cultures, as well as in the organism, which they consider to be degenerative, and state
explicitly:

Die Ursachen der Plasmoptyse und Plasmolyse kénnen sehr verschiedenartig sein . . . Es ist daher nicht
erlaubt, aus dem Vorgang selbst Schliisse auf das auslésende Agens zu ziehen.
No. 2.] -LIFE CYCLES OF THE BACTERIA—LOHNIS. 117

If every plasma contraction should be called plasmolysis, then naturally all gonidia
formation would have to be identified with it, but as the term can be used correctly only in
those cases where, in fact, the change in osmotic pressure acts as causative agent, gonidia
formation may occasionally be mistaken for plasmolysis, especially if the last-named authors’
advice should be accepted, but more careful research will always reveal the inherent difference
between both processes. Podwyssoteky and Taranoukhine (1898) went even so far as to link
the spore formation of B. anthracis with A. Fischer’s plasmolysis theory, though this species,
like several other spore-forming bacilli, was declared by A. Fischer (1903, p. 25) to be not
subject to plasmolysis on account of its permeability. As many other species (sarcinae,
streptococci, staphylococci, etc.) were also found by the same author to be not inclined to show
any plasmolytic reaction, this process unquestionably can not be used as a correct explana-
tion for the formation of those granules by all bacteria, which, though contracted plasma,
demonstrate by their further behavior that they are actually reproductive organs.

That genuine plasmoptysis has still less to do with the production of reproductive organs
than has plasmolysis, is equally evident. A. Fischer never furnished photographic pictures
of what he saw in his experiments; the drawing reproduced as figure 62 on Plate O from
his textbook (1903, original fig. 27) is obviously highly schematized and is rather different
from the drawings given in another paper (1906), one of which has been reproduced as figure
181 on Plate XV (from original fig. 8 on Pl. III). This ejecting and dissolving of irregular
clumps of plasma is, indeed, what might be expected, and is also in aceordance with the original
description given by Fischer (1900), while all his other data concerning the formation of a
new membrane around the ejected plasma and the motility of these bodies, evoke a strong
suspicion, that he actually has had side by side in his preparates plasmoptysis, as well as gonidia
and regenerative bodies, and that indiscriminate ‘‘explaining”’ of the latter fact by the former
has caused the development of his incorrect theory. Furthermore, it has to be remembered,
that many of the globules, considered by A. Fischer to be ‘‘Plasmoptyse-Kugeln,”’ have been
merely droplets of some organic substances (‘‘dirt droplets,” as A. Meyer called them), which
are indeed quite common with coverglasses which have not been thoroughly cleaned. Leuchs
(1905) has directed the attention to this fact, which has been admitted by A. Fischer
(1906). Additional data concerning genuine and so-called plasmoptysis were furnished by
Blaw (1905), A. Meyer (1905-1906), Hammerl (1906) and Garbowski (1906-1907). The last-
named author, though working in A. Fischer’s own laboratory, was equally unable to confirm
this authors statements. A fairly pronounced plasmoptysis was recorded only by Schuster
(1910), who cultivated some bacteria in a 10 per cent ammonium sulphate solution. Under
normal conditions plasmoptysis is of no importance whatever; a study of the slow growth of
budding gonidia and regenerative bodies will, as Meirowsky (1914) has emphasized, at once
demonstrate that these organs are by no means the result of a sudden osmotic eruption, and
their further behavior confirms this fact.

The occurrence of artefacts in stained preparations is a point which needs constant atten-
tion in all investigations upon gonidia and regenerative bodies. Very deceptive artefacts
may be obtained occasionally with nutrient solutions containing albuminous or fatty sub-
stances or with certain staining methods; but it would be wrong, of course, to conclude from
such happenings that the same artefacts are also present in other cases. Important informa-
tion in this respect may be derived from A. Fischer’s book entitled ‘‘Fixierung, Farbung and Bau

_ des Protoplasmas’’ (1899), though again the conclusions drawn by the author are often out
of all proportion to their experimental basis. Weibel (1888), who was among the first
to study formation and development of bacterial regenerative bodies, has already selected
the one way which avoids the disturbing influence of artefacts in stained preparations, viz. :
the examination of the living organism in the hanging drop. That vital staining with methylen
blue may be helpful, was shown by Zettnow (1897), Nakanishi (1900a), Riditka (1903),
Meirowsky (1914), and others. Absolutely clean coverglasses are, of course, equally impor-
tant in this case, as they are in regard to ‘‘plasmoptysis.”’
118 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

A particularly disturbing effect may be caused occasionally, as was pointed out in
our second preliminary report (Léhnis and Smith, 19166), by the fact that many of the thick
walled regenerative bodies are killed, but not dissolved by thorough heating of the substrates.
Therefore, it may and will happen, especially with substrates containing pepton or other ingre-
dients which were exposed to strong bacterial infection, that dead foreign bodies are present
in perfectly prepared and sterilized media, which may enter the field, well stained and evi-
dently being genuine regenerative bodies, though actually representing an unwelcome con-
tamination. The use of various substrates, kept in thoroughly cleaned vessels, will usually
suffice to eliminate this possible cause of error, though occasionally filtering of the medium
through a reliable bacteria filter will become necessary to secure definite results.

Granules in blood and tissue may be easily mistaken for reproductive organs of bacteria
and vice versa. The granules in ‘‘Mastzellen’”’ have been discussed on p. 111. That those
in liver and other tissue may be bacterial gonidia has been mentioned repeatedly (see e.g.
Bordoni- Uffreduzzi, p. 97, and M. E. Abbott, p. 103). The production of ‘‘infective granules”
by spirochaets has directed the attention of bacteriologists to the hemoconia and other granules
in blood. As early as in 1880 M. Wolff and Guttmann have dwelled upon this subject, which
has been more recently discussed by Rabinowtisch (1909), Balfour (191le and 1912), Gleits-
mann (1913), Herzog (1918), and by King, Baeslack and Hoffmann (1913). As with blood, the
dark field method may also cause considerable trouble when used in other cases; comparative
examinations of the sterile media are necessary to avoid mistakes due to the presence of other
granules in the substrates.

Crenkowski already emphasized in 1877 that the direct observation of the development of
gonidia to bacteria is the only final proof that these small round bodies are not merely cell
inclusions, but actually reproductive organs. Zopf (1879), Toussaint (1884), and other early
authors took the same standpoint, which, of course, is beyond dispute.

But Zopf (1895) was also right when he refuted Winogradsky’s exaggerations concerning
the identity of gonidia with ‘‘sulphur granules” on the ground that he had seen them multiply
as such, though they did not revert in his experiments into the original rods and threads. At
the present time, where so little is known upon the conditions favoring the upgrowth of gonidia
and regenerative bodies to fully developed bacteria, it will have to be considered to be, indeed,
quite satisfactory, if only the formation and the multiplication of these reproductive organs
as such can be directly followed. _Adami, Abbott, and Nicholson (1899), for instance, were not
always able to get a regular upgrowth from the gonidia of B. coli; sometimes they reverted
promptly, in other cases they merely multiplied. A culture of motile ‘‘cocci’” (gonidia),
isolated by Stamm (1914) from V. cholerae, which was shown in figure 82 on Plate VII, grew
as such in 65 transfers during 7 months, then it reproduced again large curved forms. Similar
results will have to be expected frequently, but they can at least be accepted as decisive answer
in regard to the various problems concerning artefacts, reserve material, etc., though, of course,
the possibility of a contamimation by cocci has also to be kept in mind. This point, however,
can be eliminated without great difficulty. Microscopically, as well as macroscopically, the
growth of gonidia and regenerative bodies is so different from that of micrococci, that a some-
what experienced investigator will not be led astray.

The situation is much less satisfactory if the problematic bodies refuse to grow under the
conditions selected for the experiments. Such negative findings have been recorded, e. g.,
by Résitka (1907), Fontes (1910), Leishman (1910), Th. Smith (1913), Wolbach (1915), and
many others. Leishman and Smith, however, though failing in the cultural experiment, suc-_
ceeded in the animal test, and there is no doubt that as our knowledge will increase also cultural
experiments will become more and more successful. To specify the conditions beforehand
and, if they prove to be not suitable, to draw from such a negative result the final conclusion
that no growth is possible, is, of course, a very serious, though unfortunately not infrequent,
mistake. Miehe (1913), for instance, was absolutely certain that the globular forms pro-
duced by his B. repens were “merely involution forms,” because they did not germinate in
his hanging drop. But the normal rods of B. foliicola, which he also studied in the hanging
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 119

. drop, were equally not inclined to grow and to multiply under his eyes; yet there is nothing
said about their being ‘merely involution forms.’ With endospores the irregularities in ger-
mination are well known, and in this case a negative finding will never be accepted as decisive.
Exactly the same standpoint has to be taken in regard to the other reproductive organs of the
bacteria.

The motility often shown by the gonidia, and sometimes by regenerative bodies, as well,
may be accepted as a fairly conclusive proof, though the Brownian movement of small foreign
bodies, especially conspicuous in the dark field, makes ‘eternal vigilance’’ imperative. The
use of antiseptic substances was found to be helpful by Kuntze (1904). The staining of flagella
is still more conclusive, but, of course, also much more difficult. The photographs reproduced
as figures 160 and 161 on Plate XII, show clearly that satisfactory pictures of these minute
details are not easily obtainable. One long flagellum seems to be very common with the motile
gonidia of practically all groups of bacteria. It has been so recorded by Giard (1882) for Creno-
thriz, by Beijerinck (1888) for B. radicicola, by Plaut (1907) for B. fusiformis, by D. H. Jones
(1913) for Azotobacter, by Marrassini (1913) for B. subtilis, and by Stamm (1914) for V. cholerae.
As is the case with the motility of the bacilli themselves, so also the motility of these repro-
ductive organs may vary according to circumstances. According to Migula (1897, Vol. I,
pp. 202-205) the gonidia of Crenothrix are immotile, while others saw motility, and Cladothrix
gave him immotile as well as motile gonidia. Adami and his collaborators (1899) found those
of B. coli immotile, while Kellerman and Scales (1916) noticed a very conspicuous motility.
The gonidia and regenerative bodies of V. cholerae, which practically always exhibit a lively
motility, as was seen by Hammerl (1906), Kolle and Hetsch (1911), Stamm (1914), and
others, may show variation, too; Almquist (1908) found them to be motile at 20° C., immotile
eve, () THE DIFFERENT TYPES OF REPRODUCTIVE ORGANS.

Tt goes without saying that a sharp distinction between the different types of reproductive
organs is not possible in every case. They all serve the same fundamental purpose: cell repro-
duction; but in addition they may either participate in multiplication, or they may act as more
or less resistant resting forms, or they may combine these two supplementary functions.
Their formation, development, and character will differ accordingly, but some overlapping
will also be noticeable. Minute spots of nuclear material represent the beginning of all repro-
ductive organs; the addition of plasmatic substances, of reserve material, and of a more or less
resistant membrane determine their final appearance; but though this will suffice for making a
correct classification in most cases, sometimes a close study of the actual development of the
bodies concerned will be necessary to distinguish correctly, for instance, between regenerative
body, arthrospore, and microcyst, or between endo- and exospore.

The names used by us at present for the four groups of reproductive organs are:

(1) Gonidia.

(2) Regenerative bodies and exospores.
(3) Endospores.

(4) Arthrospores and microcysts.

The meaning of these terms, and why they have been selected, may be discussed first;
the literature relating to character and behavior of the different types of reproductive organs
will be reviewed afterwards.

What we call gonidia of the bacteria is probably identical, at least partially, with what
has been called eggs (“Hier’’) of the bacteria by Ehrenberg (1838); some of his observations,
however, may also relate to endospores, and the name itself would be, of course, ‘unacceptable
now. The term “blastia’’ used by Perty (1852) is much better, but its original meaning has
also been too broad. The endospores have been kept separate by the Swiss author, but besides
the bacterial gonidia, his blastia also enclose regenerative bodies of the bacteria and various
reproductive organs of protozoa. The ‘‘micro-gonidia,”’ which Karsten (1869) observed to
grow up to bacteria, have been probably also gonidia, as well as regenerative bodies. In relation
to Crenothrix F. Cohn (1870) made use of the terms micro- and macro-gonidia, though he
120 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

admitted that no sharp distinction between both can be made. According to our present «
arrangement his macro-gonidia would have to be classed among regenerative bodies or micro-
cysts. What Sachs (1875) named swarm-spores or zoospores of lower algae is equivalent to
our gonidia of the bacteria; he calls them also micro- and macro-gonidia. In Zopf’s paper
on Crenothrix (1879) the terms gonidia and spores are used synonymously. According to
Béchamp’s description (1883) some.of his ‘‘microzymas” are undoubtedly identical with the
bacterial gonidia, but it is equally obvious that many other things, even part of the bacteria
themselves, would be covered by the same name. Much of what should have been termed
gonidia, has been called “cocci” by Zopf (1883); an inaccuracy which has been justly criticized
by Hueppe (1886). Unfortunately, this author erred by asserting that gonidia be resting
forms and unable to multiply as such, the latter fact being characteristic for cocci. If this
would be true, Zopf, not Hueppe, would have been right.

The application of the term gonidia to the club-shaped cells of Actinomyces and of other
organisms, as was done by Bollinger (1877) and by A. Neisser (1888), can also not be accepted
as correct. These cells may contain gonidia, but they are not gonidia (i. e., seed) themselves.

“What Kiebs (1887) called ‘‘microsomes” of the bacteria have been evidently true gonidia,
and this term might be used had it not been applied by other writers, e. g., by Spencer (1898,
p. 253), in a much more general sense to all plasmatic granules to be found in the cells.. The:
motile gonidia of B. radicicola have been called zoospores by Atkinson (1893) and by Hartleb
(1900). Thiercelin (1903) chose the designation ‘‘microblastes”’ for the gonidia of his Entero-
coccus, The so-called spores of Spirillum endoparagogicum (Sorokin, 1887), Actinomyces bovis
(Bostroem, 1890), B. coli and typhi (Almquist, 1893), B. fusiformis (Tunnicliff, 1906), and of
Leptothrix (Chlamydothrix) ochracea (Petschenko, 1916) exhibit all marks of genuine gonidia,
Those produced by the trichobacteria have been called by Migula first (1900) conidia, later
(1904) gonidia. Hillis (1907) and Benecke (1912) also wrote upon ‘‘conidia” in trichobacteria,
while Chester (1901), Schmidt and Weis (1902), as well as Frost and McCampbell (1910), gave
preference to the term ‘‘gonidia.” The so-called conidia of Almquist (1904-1917) include
gonidia as well as regenerative bodies. Whether the ‘‘sporoids” of Rizitka (1909) have been
gonidia must be left undecided. What Jones (1913) described as ‘‘gonidia spores” of Azoto-
bacter, however, have been undoubtedly true motile gonidia. The expression ‘‘infective
granule,” used by Balfour (1911-1913) and others in relation to the gonidia of spirochaets, as
well as other protozoological terms, like ‘‘marginal points,” ‘‘anaplasma,” are evidently not
preferable to the old, fairly well established and defined term ‘‘gonidia.” The terms blastia,
microblasts, etc., I believe, do not deserve preference. As not all gonidia are motile, the use
of the name zoospores can also not be recommended. And the rather indefinite expression
‘‘spores”’ should certainly be avoided. That ‘‘gonidia” (derived from yévos = offspring) should
not be associated with or replaced by ‘‘conidia”’ (derived from xovia = dust) needs hardly to be
emphasized. Conidia are, according to Lindau (1904, p. 191):

Sporen, welche exogen, d. h. ausserhalb der Zelle durch Abschniirung entstehen.
And:
Die Konidien sind die recht eigentlichen Fortpflanzungszellen des Pilzreiches, denn sie zeigen die Anpas 1ng
der Pilze an das Landleben in der héchsten Form.

It is true that under abnormal conditions conidia may be also formed internally, as has
been described, for instance, by Klécker and Schiénning (1899) for Dematium pullulans and
other fungi, and one might say that the arthrospores produced by Actinomyces at its aerial
hyphae are genuine conidia and the gonidia formed within its cells are such abnormal endo-
genous conidia. However, it seems to be much more preferable and recommendable to avoid,
as far as bacteria are concerned, the term conidia entirely and to speak alone of gonidia. With
the trichobacteria the latter term has been applied by some authors also to full-grown cells,
when these are formed within the sheath, an extension of its original meaning, which better
will be avoided as far as possible.

According to the old definition, given by Sachs (1875, p. 211) with regard to the lower
algae, which, however, is equally applicable to the bacteria, gonidia are organs of asexual repro-
No, 2.1, LIFE CYCLES OF THE BACTERIA—LOHNIS. 121

duction, formed by the contraction of the plasmatic cell content, which leave the parent cell
either by breaking the cell wall, or which become liberated when the cell dissolves. Motility
is frequently, though not generally, noticeable with bacterial gonidia. Usually two, four, or
more gonidia are formed within the same cell. Their resistance, as a rule, is not very different
from that of the full-grown, vegetative cell, as they are serving much more for multiplication
than for preserving the life under unsuitable circumstances. They often seem to be made up
nearly exclusively of chromatin substances; occasionally, however, the presence of reserve
material (fat, starch, glycogen, etc.) may become so conspicuous that errors in the microchemical
analysis may be caused. .

As was mentioned above, the so-called infective granules, marginal points, etc., of protozoa
are very similar to bacterial gonidia. What Minchin (1915) said in regard to protozoa
holds also true in our case:

Infective granules are true endogenous chromidial buds.
And: :
Each bud, when complete, has the morphological and cytological value of a true cell, very minute in size and
reduced almost entirely to its chromatin elements.

That the bacterial gonidia sometimes may multiply as such for a long time before they
revert to normal cells is in good agreement with this view.

Sometimes the endogenous production of gonidia assumes such proportions that a con-
siderable increase in size and an alteration in the shape of the parent cell becomes visible.
Toussaint called such cells ‘‘sporangia” (according to Magnin and Sternberg, 1884, p. 150) or
‘‘pseudothéques” (according to Rodet, 1894, p. 117). Finkler and Prior called them first
““Ammen”’ (1884), later ‘‘Keulen”’ (1885). -Artigalas (1885) spoke of ‘‘spores”’ containing
“‘sporules,” and Carpano (1913) also used the term ‘“‘spores.’’ That the latter term should
not be applied in this manner is beyond question. But the names used by Finkler and Prior,
as well as by Toussaint, are equally not acceptable. Therefore, it seems best to introduce the
term “‘gonidangium,”’ already used in mycology, but to reserve it for those ‘‘giant cells’’ pro-
ducing numerous gonidia, and not to apply it to normal cells containing a few gonidia, though
this, of course, would be etymologically equally correct. As will be seen, very many of the
so-called involution forms of spherical, pear, and club shape are, in fact, gonidangia.

What we now call ‘‘regenerative bodies’’ is a collection of different reproductive organs
whose proper separation and classification must be left for the time when more data will be
available. Among the reproductive organs as described by earlier authors the following may
have been regenerative bodies: Some of the macro-gonidia of F. Cohn (1870), the ‘‘arthrospores”’
of V. cholerae as described by Hueppe (1885), the round reproductive organs of Actinomyces
growing singly on short side branches, which were called ‘‘chlamydospores” by Schiirmayer
(1900), some of the “bacteroids” of the nodule organism and of other bacteria, the ‘‘zygospores”’
of B. radicicola studied by Hartleb (1900), the ‘‘zygo- or stylo-spores”’ of B. tuberculosis described
by Droba (1901), the ‘‘Dauerkapseln” of B. tuberculosis mentioned by Schroen (1904), part
of. the ‘‘conidia”’ of Almgquist (1904-1917), probably some of the ‘‘unreife Sporenanlagen’’
frequently found by Grassberger (1903) with anaerobic bacilli, and the ‘‘Regenerationsformen”’
of Azotobacter discovered by Prazmowski (1912, p. 145).

Generally the regenerative bodies are characterized by their being easily and deeply stained
by aqueous dyes, by their different appearance as compared with the vegetative cells (globular,
oval, or irregular), by their ability to reproduce normal cells immediately or after having prop-
agated as such by fission or by budding, by their distinctly increased resistance against drying,
heat, or other detrimental influences. But, as has been said above, this grouping can be only
accepted as some temporary arrangement, which will have to be discarded as soon as the different
components of this group will be more completely known,

Some of them are evidently related to what we call terminal exospores and probably will
later have to be united with them. Others seem to be genuine zygospores, as will be dis-
cussed more fully in Ohapter IV. Many of the round forms, as well as some of those of
irregular shape, usually characterized by an especially solid appearance, are produced directly
122 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

from the symplasm, as will be demonstrated in Chapter III. Perhaps the name ‘‘sclerotia’’
might recommend itself for the latter forms. But all final action upon this point, as well as
the proper naming of the remaining ‘‘regenerative bodies,’’ must be left to future experimental
investigations.

“‘Exospores’’ are those reproductive organs which are produced either terminally or later-
ally and whose staining reaction and resistance are similar to those of the endospores. As to
their mode of being formed, it seems not improbable, that they, too, are zygospores, and perhaps
this will be their final denomination.

At present it is customary to call the ‘‘endospores” of the bacteria simply ‘‘spores’’, but
this custom will have to be abandoned, as it is beyond doubt, that other bacterial reproductive
organs are equally true “‘spores.’”’ The demand of some authors, like Gamaleia (1888), Bruns
(1895), Hawthorn (1903), Mertens (1903), Frost and McCampbell (1910), that only those repro-
ductive organs should be called spores, which are characterized by their special staining reaction
and their exceptional heat resistance, can not be accepted as being correct. Zropd means
“‘seed,’’ and therefore even the gonidia might be called spores, as has been done, indeed, by
several authors. But as their morphological, as well as their physiological, character is so
widely different from that of the typical spores, undoubtedly a special name was preferable in
that case. On the other hand, it seems not to be justified to confine this term strictly to only
one type of reproductive organs, and to exclude all others, which according to formation, develop-
ment and general character, are very similar to them, though not quite so resistant to staining
and heating. The use of the distinct term ‘‘endospore,’’ therefore, is to be recommended; the
term ‘‘cyst,”’ as used by Gamaleia (1900), on the other hand, should not find application in this
case.

The term ‘‘arthrospore,” introduced by Caspary into the mycological literature, has been
refuted by Bail as early as in 1857 on the ground that only those reproductive organs should be
called spores which are the results of some process of fructification. The opposition was still
more pronounced after De Bary (1884) had taken up this term for bacterial reproductive organs
and Hueppe (1886), as well as De Toni and Trevisan (1889), had accepted and extended its use.
Kruse (1896 a, p. 60), Migula (1897, Vol. I, pp. 37, 168; 1900, p. 8; 1904, p. 123), Schmidt and Weis
(1902, p. 58), ‘A. Fischer (1903, p. 42), Gotschlich (1903, pp. 82, 83), Muir and Ritchie (1908, p. 8),
Giinther (1906, p.21), as well as Hiss and Zinsser (1914, p.16), have all taken a more or less pro-
nounced stand against this theory. However, a careful study of all reports concerned leaves no
doubt, that the special morphology, increased resistance, and the germination of these spores
have been fully ascertained at least in some cases, and, therefore, it is not admissible to discard
them lightly as being ‘‘simply involution forms,’’ as was done repeatedly. The term has been
fairly correctly used by G. VV. Sternberg (1893, p. 17), Crookshank (1896, p. 19), Lafar (1897, p. 62),
Gamaleia (1900, p. 17), and Benecke (1912, p. 179), though all these authors paid more attention
to those cases, where the whole vegetative cell was transformed, than to the others, where by
a preliminary segmentation joints were formed which then became arthrospores. In our first
preliminary paper (Léhnis and Smith, 1916 a, p. 679, footnote) we also said:

We think it best to reserve the name ‘‘arthrospore’’ exclusively for those cases where the whole cell acquires
the character of a spore.

However, a more thorough consideration of the problem has made the conclusion inevitable,
that whole cells turnitig into resting forms should be named ‘‘microcysts,’’ and the term ‘‘arth-
rospore”’ should be used for those spores which ‘‘like a chain of beads are formed by fission”’
according to Jackson’s definition (given in his Glossary of Botanic Terms), which also agrees
best with the etymology of the term, viz: 7d &pdpov=the joint. It has been mentioned
before that in this way the term ‘‘conidia,’’ sometimes used for the aerial spores of Actinomyces,
can be avoided, which is very desirable on account of its being frequently misused and also easily
mistaken for ‘‘gonidia.’”’ If these reproductive organs of Actinomyces are accepted as type of
arthrospores, their being produced by segmentation or fragmentation, their increased resistance,
as well as their germination, can all be accepted as thoroughly studied by various investigators.
Whether some process of fructification precedes the formation of these spores, must be left in
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 123

doubt at present, in the same manner as with regard to the endospores. That other bacteria
are able to produce such arthrospores is beyond question. Those of the Proteus group are
most generally known, but others formed by saprophytic spirochaets are also quite typical, as
may be seen from the drawings made by Gross (1912, original figs.8 and 9), which were reproduced
as figure 63 on Plate O.

If the whole cell becomes more resistant by thickening its membrane, and sometimes also
by reducing the amount of water present within the cell, the resting form produced may be
termed “microcyst.’’ The analogous cells of Nostoc have been called ‘‘heterocysts” by Sachs
(1875, p. 215), those of Leuconostoc by Van Tieghem first “spores” (1879c), later ‘“‘kystes”
(1884, pp. 1105, 1108), and those of several spore-forming bacilli ‘‘chlamydospores” by A. Meyer
(19016), whose drawings, reproduced as figure 64 on Plate O (from original figs. 4 and 15 on
Pl. XX), illustrate the difference of the microcysts when compared with Gross’s picture of
the arthrospores of spirochaets. The ‘‘spores’’ of Myxococcus exhibit also more the character
of microcysts than that of arthrospores, though some similarities are noticeable in the latter
direction, too, as may be seen from the drawing made by Zukal (1897), reproduced as figure 65
on Plate O (from original fig. 3).

Kvoris means cavity, bag, and a cyst is according to Jackson ‘‘a cell of nonsexual origin
which reproduces the plant by germination after a resting period.’’ The character of a bacterial
microcyst fits this definition completely. That, however, the special name microcyst seems to be
more recommendable in this case than the more general term cyst, is due to the fact, that the
latter expression has been used in a different sense with the Myxobacteria, and that, as will have
be shown in Chapter III, similar large cysts, i. e. ‘‘macrocysts,’’ may also be formed by all
bacteria when in the symplastic stage.

By some authors typical microcysts have been named endospores, a usage which can not
be approved. Migula (1897, Vol. I, p. 169), for instance, insisted that Van Tieghem’s Leucon-
ostoc cysts had been endospores, because they had been brighter in the center and had germinated
like endospores, though Van Tieghem clearly stated that the wall of the whole cell ‘‘se cutinise
en déhors.”” Prazmowski expressed the analogous opinion in regard to resistent thickwalled
cells of Micrococcus ureae (1888a) and of Azotobacter chroococcum (1912), despite the fact that the
typical character of microcysts is especially noticeable in these cases.

The coiled and ‘‘encysted’’ spirochaets furnish another good example of the mode of forma-
tion of this type of reproductive organ, as well as of the correct use of the term.

That arthrospores and microcysts can not be easily distinguished when isolated is readily
to be admitted. But the same holds true with regard to endo- and exospores, as well as in other
cases, where the knowledge of the formation of the body in question is necessary in order to
reach a scientifically correct classification.

(1) GONIDIA.

The formation of the gonidia is caused by a contraction and fragmentation of the content
of the parent cell; though, naturally, the details of this process vary with the different groups of
organism and to some extent with the cells themselves.

With the large trichobacteria, where F. Cohn (1870) first studied the formation of the
gonidia, the cell content contracts itself into larger and smaller globules. When the cells break
up within the sheath into 16 or more of such motile globular or ovoid granules, Cohn speaks of
micro-gonidia; while he uses the term macro-gonidia, when only 2-4 are formed. But he
himself emphasizes that both types of gonidia can not be sharply separated. The drawings
made by Billroth (1874), which were reproduced as figure 45 on Plate L, demonstrate very
clearly the characteristic formation and liberation of gonidia by some long bacilli, probably
lactobacilli, as they were accumulated in whey kept at 40-45° C. The analogous description
given by Ewart (1878) for the ‘“‘sporules” of B. anthracis, as well as for the “bright, unusually
small, almost spherical spores” of B. termo (fluorescens), and by Geddes and Ewart (1878)
for the motile ‘‘spores”’ of spirilla, are also of considerable interest. A comparison of the last-
named paper with those of Perty (1852), of Fantham (1911), and of Bosanquet (1911) leaves no
124 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

doubt that essentially the same process has been studied by all these authors (see figs. 42, 59,
and 60, on Plates L and O). If the gonidia-do not slip out at one end, as was sketched by
Fantham, but are simply liberated by a complete dissolution of the cell wall, as has been
observed with spirochaets, e. g., by Bosanquet and by Riither (1910), the picture honored very
similar to what is a common occurrence with the Mycobacteria. Iniz (1886) made the
first pictures of this process, as exhibited by B. leprae, reproduced as figure, 36 on Plate J.
Those published by Kedrowski (1910), reproduced as figure 37 on Plate J and as figure
153 on Plate XI, illustrate some other equally interesting phases in the formation of
gonidia by this organism; they may be compared with those of the diphtheroid organism,
drawn by Babes (1895) and reproduced as figure 34 on Plate J. The description given by
Bostroem (1890) concerning the analogous behavior of Actinomyces also deserves our attention.
The threads were seen to contain spherical, bright, ‘“‘bead-like’’ granules, having the same
diameter as the threads and being separated from each other by regular distances; they either
slipped out of the sheath or became free by the dissolution of the cell wall. More recently
Hollandt (1906) noticed that occasionally Actinomyces is able to produce micro- and macro-
gonidia in the manner characteristic for Crenothrix. That the same fact can be sometimes
observed with smaller bacteria, too, has been demonstrated by Miller (1889), when he made
his drawing of the so-called Jodococcus vaginatus, which was reproduced as figure 52 on Plate N.
Almquist (1893) recorded analogous findings when he studied B. coli and typhi; Prazmowski
(1912) and we (Léhnis and Hanzawa, 1914; L. and Smith, 1916) saw the same facts with
Azotobacter.

The sketches made by Morck (1891) of nodule bacteria forming gonidia, which are repro-
duced as figure 66 on Plate P (from original figs. III, 4, and V, 12) show an interesting ring
formation, which may be found with other bacteria, too; P. Ernst (1902), for instance,
described this fact with B. fluorescens. That the gonidia may arrange themselves into such a
circular grouping within the parent cell, while this retains its rod-form practically unchanged,
has been recorded by Trambusti and Galeotti (1892) in the case of a bacillus from water.

Usually, however, the gonidia formed by such small rods as B. colt, fluorescens, ete., or
by the larger ones of the Subtilis-Mesentericus group, appear simply as 2, 4, or more granules,
arranged in one row. A photograph of B. coli, made by Kellerman and Scales (1916) from a
contact preparate, has been reproduced as figure 182 on Plate XV; it allows one to recognize very
clearly how abundant the gonidia may already be in young colonies, According to our own
observations their number usually reaches the maximum, when the development of a culture
is at its height. Figure 183 on Plate XV, reproduced from our first preliminary report (Léhnis
and Smith, 1916 a, original fig. 39), pictures this phase in the life of a spore-forming bacillus
(yellow bacillus No. 41) and makes a good object for comparison with the preceding photograph
of B. colt.

With micro- and streptococci the gonidia formation becomes most conspicuous, when these
very small bodies bud out of the parent cells, but remain temporarily attached to them by a
very fine, rather long stem, a peculiarity which probably has been first seen by Thiercelin and
Jouhaud (1903 a), and which is illustrated by our photographs of Micr. candicans, luteus, and
Strept. lactis, reproduced as figures 168-170 on Plate XII. But again the analogous behavior
can be traced through all groups of bacteria, as was ascertained by Meirowsky (1914 6), whose
most characteristic drawings have been reproduced on Plate XIV.

Data relating to the formation of a single gonidium are not available at present. Some
observations reported by Fuhrmann (1908), Amato (1908) and Rizicka (1908) indicate, however,
that this process is very similar to the initial steps in the formation of an endospore.

The production of gonidangia by the bacteria is quite general. Usually they are the result
of a simple swelling of the parent cell; sometimes, however, the gonidia bearing rod first forms a
circle, which later develops into a large coccoid body, as has been recorded for V. proteus by
Firtsch (1888), for B. radicicola by Morck (1891), and for B. fluorescens by P. Ernst (1902).
This process resembles very much another one noticeable with spirochaets, where the coiled
forms first produce microcysts, which then act as receptacles for the gonidia, liberated inside
No. 2] LIFE CYCLES OF THE BACTERIA—LOHNIS. 125

by the dissolution of the coiled spirochaets. Perrin (1906), Dutton and Todd (1907), and others
have studied this transformation. The gonidangia of spirochaets, described by Ross (1912)
and by Moolgavkar (1912), seem to have been, partially at least, of the same origin. They
were either “large spherical cells, studded with small chromatin granules” or more irregular,
pear-shaped bodies. In figure 67 on Plate P gonidangia of B. radicicola are reproduced from
drawings made by Morck (1891, original figs. 2 b-d on Pl. IIT), to demonstrate an occurrence
which has been also studied by Hartleb (1900), Hiléner and Stérmer (1903), Rossi (1907), Arz-
berger (1910), and others.

That numerous observations upon the formation of bacterial gonidangia were made
by some of the pioneers in bacteriology at the time, when the unscientific discarding of
all ‘involution forms’ was not yet an established rule, was mentioned in the historical résumé
given on pages 90-108, especially on page 100, together with'a very characteristic remark
made by Migula. The early drawings of Billroth (1874), reproduced as figures 44 and 45 on
Plate L, are excellent illustrations of the manner in which gonidia bearing rods may develop
to voluminous gonidangia. The description given by Toussaint of the gonidangia of B.
anthracis is equally worth quoting. In the textbook of Magnin and Sternberg (1884, p. 150)
this otherwise unpublished observation was described as follows:

In cultivating spores of the bacteria of charbon . . . Toussaint has seen the filaments take a transverse diame-
ter almost double the ordinary diameter, then the protoplasm of the filament to gather together at certain points

. . Finally the points occupied by the condensed protoplasm augment considerably in volume and form some
ovoid organs, more or less elongated, or swollen into a ball, or in the form of a gourd at one extremity. In the interior
of these sporangia from three to six spores afterward form, . . . finally by breaking up of the membraneous en.
velope the spores become free.

Still earlier findings, recorded by Leézerich (1876-1878) in his experimental investigations
upon diphtheria and typhoid fever, are not less interesting. He saw isolated cocci in the
blood grow up into large plasmatic globules, which reproduced internally new bacteria. Some
sketches, published in the first paper (1876, original figs. 1-4, on Pl. XII) are reproduced
as figure 68 on Plate P, because they will possibly regain new importance in further in-
vestigations upon the “Goccoid” form of diphtheria, mentioned on page 79, as well as in
studies upon the relations existing between Bacteriaceae and Myxobacteriaceae. At the same
time Geddes and Ewart (1878) gave the analogous description of the gonidangia formed by
spirilla, while J. Israél (1878) noticed that the club-shaped cells of Actinomyces may serve
the same purpose. The pear-shaped cells were seen to break up and to liberate their granular
content. Exactly the same fact has been recorded by E. Chr. Hansen (1879-1894) with regard
to the acetic acid bacteria, whose globular forms grew up to lly diameter. V. cholerae and
related organisms have also frequently exhibited their large globular or club-shaped gonidangia,
which have been fairly well studied by Finkler and Prior (1884-1885), Ferrdn (1885) Schroen
(1886-1890), and by Dowdeswell (1889-1890), though they remained to R. Koch and his fol-
lowers always uninteresting ‘‘involution forms.” A reference made by Kiéinstler (1885) in
regard to the gonidangia of spore-forming bacilli has been cited on p. 96. It is again in com-
plete agreement with those mentioned before. In fact, Kinstler (1885), as well as Schroen
(1886), have already reached the conclusion that the production of gonidangia is a common
occurrence within all groups of the bacteria. More data upon this subject have been furnished
by the last-named author in two other papers (Schroen, 1890: Zur Genese der Microorganismen,
and Schroen, 1904: Der neue Mikrobe der Lungenphthise). Unfortunately the author seems
to have been unable to finish a special publication upon ‘‘The Genesis of the Microbes,’ ’ which
he promised in the second paper. Concerning the tubercle bacillus he wrote in 1890 that as
with the cholera vibrio so also in this case globular forms may increase in size and become

‘mother spores,” whose content divides itself into numerous coccoid ‘‘daughter spores.” And
with B. anthracis and Megaterium he noticed that the normal cells swell up into large somewhat
irregularly curved ‘‘Bacillen-Sackchen,” containing coccoid reproductive organs, which either
are liberated as such, or transform themselves already internally into new small bacilli. The
dividing up of the content of such gonidangia into only two large curved bacilli has been also
126 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

recorded by Schroen (1890), a possibility which was mentioned by E£. Klein (1885) and by
Dowdeswell (1890) with regard to V. cholerae, and later by Hartleb (1900) in his studies upon
B. radicicola.

In his last paper Schroen (1904) comes to the conclusion that besides the tubercle bacillus
another fungus-like organism acts as causative agent of phthisis, of which he gives among
others the picture reproduced as figure 192 on Plate XV (from original fig. VIII). Apparently
genuine hyphae are visible in the picture. However, it is more probable that these thick
threads are no hyphae at all, but slime threads very similar to those produced by B. radicicola
and the Myxobacteria. In Chapter III this point will have to be considered more fully. The
round ‘‘capsules’”’ of Schroen’s organism, too, seem to be clearly bacterial gonidangia. Ac-
cording to his own description they produce 20 to 40 thin bundles of very fine threads, which,
after their liberation, appear just like tubercle bacilli. As figure 189 on Plate XV Azotobacter
gonidangia have been reproduced, which were interpreted by Beijerinck (1901 6, original fig. 4)
as involution forms; they make a highly interesting counterpart to the ‘‘capsules”’ of Schroen’s
‘‘phthisiogenic microbe,” as well as to the gonidangia of the Pneumococcus, discovered by
Artigalas (1885) and reproduced as figure 122 on Plate X. That with the streptococci again
pear-shaped gonidangia may replace these perfectly round ones, has been shown by Maddox
(1885), who observed the liberation of the deeply staining granular content of large club-shaped
cells, present within or at the end of chains of Streptoc. lactis. With regard to his Entero-
coccus, Thiercelin (1899) wrote:

On voit le coccus devenir trés volumineux et donner naissance 4 un grand nombre de petits diplocoques vivement
colorés au Gram, contenus dans une sorte de gangue amorphe, vaguement colorée par 1’ eosine.

The upgrowth of new small bacteria within the gonidangia, described by Schroen, has been
also recorded by Babes (1895), whose drawings were reproduced as figure 53 on Plate N,
as well as by Gamaleia (1900), who called these small forms ‘‘Mikromiten.” That Schmorl’s
photographs of his Streptothriz cuniculi (B. necroseos), which were reproduced as figures 89
and 90 on Plate VIII, illustrate the formation of gonidia and gonidangia very well, will become
apparent in this connection. Hibdler’s pictures of B. Chauvoei, shown on Plate X as figures
115 and 116, as well as that of B. mallet made by Carpano and reproduced as figure 96 on Plate
VIII, deserve also to be recommended for a renewed comparative study. Semmer (1895)
reported that tubercle as well as glanders bacilli produce big inflations filled with bright gran-
ules, like those to be found within the rods, which are able to reproduce rods and threads after
being liberated. Stoklasa (1898) saw B. Ellenbachensis form ‘‘Blasen”’ and ‘‘Kugeln”’ with
numerous motile granules, and Riditka (1908) obtained analogous results with B. anthracis;
here the large globules were seen to liberate their granular content, which was interspersed
with very small rods. In view of all these facts it is not easy to understand how Fuhrmann
(1906-1913) could reach the standpoint and adhere to it that the process observed by him,
viz., the formation of large pear-shaped cells or clubs containing and liberating either small
motile granules or young rods, i. e., the development of typical gonidangia, be restricted to
that group of bacteria which he happened to study (some Pseudomonas species), while it is,
in fact, quite general with all bacteria.

_ The many inflations observed by Hollandt (1906) within thick threads of Actinomyces
revealed their true character by the regular manner in which their content divided itself into 4
and 16 parts (gonidia). The resemblance to what F. Cohn called a “sporangium” of
Crenothrix, reproduced as figure 43 on Plate L, is obvious.

On the other hand, Herzog (1910-1913), Lehmann and Neumann (1912), and others have
noticed that the so-called giant forms of micrococci, especially those of gono- and meningo-
cocci, produce small granules, which the first author found able to reproduce normal cells.
The similarity to the gonidangia of the streptococci, mentioned above, is again very clear.
And if all these important facts would not have been so utterly neglected by the textbooks
Hort (1917 b) certainly would not have been inclined to remove the meningococcus to the
Hemiascomycetes, because once more he observed the upgrowth of normal cells to “giant
forms,” producing small granules, which in their turn became again normal cocci. A photo-
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 127

graph published in our first preliminary report (Léhnis and Smith, 1916 a, original fig. 5), illus-
trating the liberation of the gonidia from gonidangia of Azotobacter has been reproduced as fig-
ure 191 on Plate XV. With B. fusiformis Rosenow and Tunnicliff (1912) saw small “cocci”
escape from the swollen ends of the threads. Almquist (1916) obtained small globules, as well
as straight and curved rods, from the gonidangia of B. typhi. That the characteristic ‘“involu-
tion forms”’ of B. pestis are also gonidangia is clearly indicated by the drawings made by Albrecht
and Ghon (1900), which were reproduced as figure 11 on Plate D. The same holds true concern-
ing the inflated big threads of B: coli, photographed by Matzuschita (1900) and reproduced as
figure 36 on Plate III; and it is not to be doubted that many other so-called involution forms,
especially most of the globular, pear-shaped, clubbed or otherwise inflated cells, so common
within all groups of the bacteria, will find a similar explanation, as soon as adequate attention
will be granted to them.

The form of the gonidia is usually, but not always, exactly spherical. Sometimes they
are more or less ovoid, or even rectangular or square, as a result of the limited space wherein
they are often compelled to grow. Their staining reaction changes necessarily according to
their varying chemical composition. (See pp.112-113.) Sometimes they refuse entirely to be
stained by aqueous dyes, as was recorded, for instance, by Esmarch (1887) with Spirillum
rubrum, by Kedzior (1896) with Actinomyces thermophilus, by Kuntze (1904) with B. ozalaticus,
by Riditka (1908) with B. anthracis, by Mencl (1911) with Azotobacter, and by Balfour (1911 a)
with spirochaets. But more often they stain as well as or even better than the parent cells.
Thus were the results obtained, for instance, by Finkler and Prior (1884-1885) with V. proteus
and cholerae, by Bostroem (1890) with Actinomyces hominis, by Stoklasa (1898) with B. Ellen-
bachensis, by Thiercelin and Jouhaud (1903) with Enterococcus, by Tunnicliff (1906) with B.
fusiformis, and by D. H. Jones (1913) with Azotobacter. When treated with Gram’s
method, the gonidia are often stained, even if the cells themselves are Gram-negative. BD.
radicicola and its relatives are prominent examples in this direction (Léhnis, 1905 a). Occa-
sionally special staining methods may give good results, as were secured e. g. by A. Neisser (1888)
with B. xerosis, and by Bliesener (1901) with V. cholerae, while the gonidia of Esmarch’s Spirillum
rubrum, on the other hand, could not be stained by any of these methods. The metachromatic
reaction of the gonidia to methylen blue seems to be fairly constant, according to Babes (1895,
1907 a, 1914) and Fuhrmann (1906-1908). The Romanowsky mathod gave Balfour (1911 a)
only negative results with the “infective granules” of spirochaets. Herzog (1910), on the other
hand, noticed that the gonidia of gono- and meningococci turned red when treated with Giemsa’s
modification of the Romanowsky stain; but when growing up to the full size of the parent cell
they assumed more and more the normal deep violet color.

Active motility seems to be more frequent among the gonidia than among their parent
cells. Geddes and Ewart (1878), Kiinstler (1885), and Dowdeswell recorded this fact with gonidia
produced by spirilla. Heydenreich, Guttmann (1880), and Albrecht (1881) were the first to report.
the same occurrence with spirochaets. Beijerinck (1888), Frank (1890), Atkinson (1893),
ee (1900), and many others have studied the ‘“‘swarming bodies”’ or “zoospores”’ of

B. radicicola. Motile gonidia of Actinomycetes were observed by Eppinger (1890), Rullmann
(1896) and Kedzior (1896), of B. anthracis by Podwyssotzky and Taranoukhine (1898), of B.
ozalaticus by Runtze (1904), of B. subtilis by Marrassini (1913), and of Azotobacter by D. H.
Jones (1918). Photographs made by the two last-named authors were reproduced as figures
160 and 161 on Plate XII. It is hardly necessary to emphasize, that the minute size of the
gonidia makes it especially imperative to distinguish sharply between active motility and
Brownian movement. Comparative tests without and with antiseptics are often advisable.

The development of the gonidia may start either inside or outside of the parent cell. If
the gonidia are not liberated by the partial or complete dissoiution of the cell wall, but remain
confined within the cell, they develop into either buds or branches. The great multitude of
forms which results from such development, may be seen from the pictures of B. coli and typhi,
made by Hort (1917 a) and reproduced as figures 179 and 180 on Plate XIII. Zopf (1879) was
apparently the first to notice this formation of branches in the case of Crenothrix. Migula

160266°—21——9
128 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. KVL

(1900) has later assumed that the gonidia do not germinate inside, but outside, of the cell wall,
to which they were said to be sticking; and he illustrated his theory by the drawings repro-
duced as figure 69 on Plate P (from original fig. 43 f and g). However, observations made
by Rullmann (1907) and by Ellis (1907-1910) have confirmed Zopf’s findings, not Migula’s
assumption. Hollandt (1906) got analogous results with the large Crenothrix-like threads of
Actinomyces, which he discovered. Spirillum endoparagogicum, as described by Sorokin (1887),
is the only example of this kind of bacterial development which has been frequently quoted in
the textbooks. A glance at his drawings, reproduced as figure 29 on Plate H, will suffice to
show that, in fact, it was by no means such an exceptional occurrence as it has been usually
considered to be. Kutscher (1895) recorded similar findings with Spirillum Undula, and
Meirowsky (1914 b) has proved the general occurrence of such branching. His drawings, repro-
duced as figure 30 on Plate H and on Plate XIV, make interesting counterparts to Sorokin’s
sketches. The lateral germination of the gonidia in B. xerosis, as studied by A. Neisser (1888),
has been another early and exceptionally clear observation of the same principle. Eppinger
(1890), too, followed carefully the gradual development of the gonidia, appearing within the
threads of his Cladothria (i. e. Actinomyces) asteroides, first into buds, and later into branches.

In regard to the typhoid bacillus Almquist (1893) was the first to point out that its multiplication
is not only caused by fission, but also by producing new rods by budding and branching. His
later publications (Almquist, 1904-1918) confirmed and extended these findings, now also with
respect to V. cholerae, B. dysenteriae and paratyphosus. The sprouting of very thin, needle-like
rods, to which the Swedish author paid special attention, is also well noticeable in Hort’s draw-
ings of B. typhosus on Plate XIII (fig. 179).

The results recorded by Escherich (1894, p. 87) with B. diphtheriae are practically duplicating
those of A. Neisser with B. zerosis, and the description given by Fraenkel (1895) of the formation
of branches by the diphtheria bacillus is in complete agreement with that given by Eppinger for
Actinomyces asteroides. Hill (1902) has pointed out that not always a granule is visible at the
base of a branch in B. diphtheriae and that its presence, therefore, is not closely connected with
the formation of the branch. This point needs further study. It may be that the bacteria are
also able to produce branches in the same manner as the fungi do, but this fact, of course, would
not interfere with the special réle played by the growing gonidia in the life of the bacterial cell.

A close relation between the ‘‘metachromatic granules” and the formation of buds and
branches by various bacteria has been emphasized by. Babes (1895). Kedzior (1896), too,
noticed how bright granules contained within the threads of his thermophilic actinomyces
extruded into buds and branches. The observations made by Stolz (1897) with a diphtheroid
organism, by Craig (1898), Cornet and Meyer (1903, p. 83) and by Fontes (1910) with B. tuber-
culosis, by Conradi (1900) with the glanders bacillus, by Neukirch (1902) and by E. Levy (1902)
with Actinomycetes, by Greig-Snuth (1900), Faber (1912) and Georgevitch with B. radicicola and
Mycobact. Rubiacearum, and by Bajardi (1903) with V. lingualis, all confirm this relation.
Most of these authors report that the chromatin granule, which Georgevitch called “le centre
cinétique,” slips into the young branch, where it is to be found at its apex.

A photograph made by E. de Negri (1916, original fig. 27), showing excellent budding of
her Corynebacterium causing malignous granulom, has been reproduced as figure 184 on Plate
XV. Another one of B. coli, made by Kellerman and Scales (1916) is to be seen in figure 185
on Plate XV. The next two pictures (figs. 186 and 187) are reproductions from our first
paper (Léhnis and Smith, 1916 a, original figs. 3 and 4); both present budding Azotobacter
cells, the last one furnishing an especially clear insight into the relation existing between branches
and gonidia retained within the parent cell.

The development of the gonidia within or while still united with the parent cell may also
lead to some other modes of development. Figure 188, reproduced from our second preliminar 'Y
report (1916 8, original fig. 1) illustrates.a case where the budding gonidia grew up to full sized

cells, which now give the parent cell a queer, turtle-like appearance. More frequent, however,
although rare compared with the common budding and branching, is the development of full-
sized cells from the gonidia while still confined within the parent cell. Naturally, only large
No.2]. LIFE CYCLES OF THE BACTERIA—LOHNIS. 129

cells like Azotobacter offer sufficient space for such direct internal upgrowth. Figures 190 and 193
on Plate XV, reproduced from our first paper (1916 a, original figs. 6 and 21) may illustrate this
possibility. Prazmowski’s (1912) report on Azotobacter contains some similar drawings. The
giant Bacterioidomonas, discovered by Kiinstler (1884), which was later described anew as
Metabacterium by Chaiton and Pérard (1913), whose drawings were reproduced as figure 50 on
Plate M, furnishes an interesting counterpart. With the smaller bacteria such kind of upgrowth
is only possible, of course, after the parent cells have enlarged themselves to gonidangia, becom-
ing the ‘‘Bakteriensickchen” of Schroen (1890) or the “Ammen” of Finkler and Prior (1884).
The sketches made by Tomaschek (1888) and by Babes (1895), which were reproduced as figures
51 and 53, Plate N, should be compared in this respect. The large cells of the protean .air
bacillus studied by Matzuschita (1902) furnish another example in this direction. (See fig. 5
on Plate B.)

If the gonidia are liberated as such, their further development, quite naturally, is subject
to many more uncertainties. They may become free by a complete dissolution of the cell wall
or they may emerge at one point as a ‘‘cloud’’ or as a string of granules embedded in a small
or large quantity of other plasmatic material. What Meirowsky (1914 6) called ‘‘umbels’’
(Dolden) have been evidently in some cases such masses of liberated gonidia, while in other cases
it seems more probable that flakes of symplasm have been present, as may be seen from his
drawings reproduced on Plate XIII (fig. 178) and on Plate XIV.

A sufficient supply of nutrient material furnished by the parent cell undoubtedly assures and
facilitates the new development considerably. The starlike arrangement occasionally noticeable
with all bacteria may be caused by the outgrowth of young rods from such liberated cell content,
as was illustrated already by one of our Azotobacter photographs, reproduced as figure 176 on |
Plate XII. Hort’s pictures on Plate XIII (figs. 179 and 180) contain some similar stars, pro-
duced by B. typhosus and B. coli. They have been frequently observed with B. radicicola
and radiobacter by Beijerinck (1888), Beijerinck and Delden (1902), and myself (Léhnis, 1905 a).
Not all stars, however, are the result of this kind of development. In Chapter III a radiate
outgrowth from the symplasm will have to be mentioned, and in Chapter IV a similar grouping
of full-grown bacteria in the conjunct stage.

If the amount of nutrient material supplied by the parent cell is insufficient, or if it is quickly
dispersed by dissolution, the development of the liberated gonidia depends preeminently on the
substrate which they will find. Lagerhewm (1900) observed that the gonidia of his Sarcinastrum
Urosporae increased in size and germinated to new rods only if they had a chance to settle down
on their natural substrate, viz., a thread of Urospora. When discussing the behavior of the
filterable gonidia more examples will have to be mentioned which furnish some valuable infor-
mation concerning the important effect which the substrate exerts in such cases. In artificial
cultures often no upgrowth was noticeable on account of unsuitable environmental conditions;
not a few failures, however, have been caused, as may be safely assumed, by the widespread
inclination to consider ‘‘granular decomposition”’ and ‘‘autolysis” of the bacteria as full proof
of their death and therefore justifying the discarding of all such cultures. That this custom does
not. take into account the important réle played by the symplasm in the life history of the
bacteria will be proved by the facts to be discussed in Chapter III. At present it may suffice
to refer only to the observations made by Fuhrmann (1906) with regard to the behavior of the
metachromatic granules (i. e. gonidia), imbedded in the “‘detritus,” which results from the
disintegration of the club-shaped gonidangia of Pseudomonas cerevisiae and related forms. He
reports that at the end of the life cycle there were present only—
die stiérker lichtbrechenden, mit Methylenazur rot-violett firbbaren Kiigelchen, aus denen sich wieder bewegliche
Kurzstébchen rtickbilden, sobald der Detritus auf einen neuen Nahrboden tibertragen wird.

If the conditions are not favorable to an immediate upgrowth of normal cells, the gonidia
may multiply as such for a considerable length of time, either by fission or by budding. This
mode of development was reported in 1865 by Hallier for the so-called Leptothrix buccalis.
In 1878 Geddes and Ewart studied the same process with some water spirilla, and at the same
130 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

time J. Israél with Actinomyces. Zopf (1879-95) saw the gonidia of Crenothrix and of Beg-
giatoa behave in the same manner. Analogous results were recorded with B. anthracis by
Archangelski (1883), with B. Chawvoei by Ehlers (1884), with B. allantoides by L. Klein
(1889), with B. coli by Almquist (1893), by Adami, Abbott and Nicholson (1899) and by WU. E.
Abbott (1900), with B. typhosus, paratyphosus, dysenteriae, and pseudodysenteriae by Almquist
(1893-1908), with V. cholerae by Cunningham (1897), by Almquist (1904-8), and by Stamm
(1914), with various spirochaets by Wechselmann and Léwenthal (1905), Henry (1913) and others,
with B. tuberculosis by Fontes (1910), and by Maher (1910-1915), with Azotobacter by Prazmowski
(1912). Macroscopically this kind of development is usually very inconspicuous; on solid sub-
trates nothing more will be seen than a very thin dewy layer without any pigmentation, while in
liquids either a weak turbidity or a fine cloud of sediment becomes noticeable. Nevertheless,
the pathogenicity may be fully retained as was discovered by Archangelski in 1883.

In those cases where the environmental conditions allow an upgrowth of the liberated

gonidia, they may, as it is also very often done by the gonidia budding out from the parent cell,
first develop to regenerative bodies before reproducing normal vegetative cells. This possi-
bility will be discussed later. The direct germination of a new vegetative cell from a gonidium
has been first recorded by Perty (1852). Haller (1865) and Karsten (1869) probably observed
' the same process, which then was firmly established by F. Cohn (1870) by his investigations
upon the germination of the gonidia of Crenothrix. Rindfleisch (1872) made some interesting
drawings of germinating gonidia, which were reproduced as figure 41 on Plate K. Zopf (1879)
confirmed Cohn’s findings, but he also pointed out that by no means all gonidia can be expected
to germinate promptly when placed under the microscope. In his important investigations
upon the gonidangia formation of B. anthracis, Toussaint (1884) also studied the upgrowth of
new rods from the liberated gonidia. The same has been done with V. cholerae by Dowdeswell
(1889), Bhiesener (1901), Fedorowitsch (1902) and Almquist (1904), with V. proteus by Finkler
and Prior (1884-1885), and by Firtsch (1888), with other spirilla by Kiinstler (1885), with Acti-
nomyces hominis by Bostroem (1890), with B. coli and typhosus by Almquist (1893)and by Fedo-
rowitsch (1902), the latter in addition recorded analogous results with B. pyocyaneus, cholerae
gallinarum, septicaemiae murvum, diphtheriae, and tuberculosis, with B. diptheriaealso by Cache
(1901), with B. tuberculosis by Arloing (1908), with B. mallet by Marx (1899), with Entero-
coccus by Thiercelin and Jouhaud (1903), with B. fusiformis by Tunnicliff (1906) with Spiro-
chaeta gallinarum by Hindle (1911), and with Azotobacter by Pragmowski (1912). Practically
this germination consists merely in an expansion and stretching of the small gonidium to the
normal size of the vegetative cell, as it was pictured, for instance, by NV. K. Schultz (1901 a)
with B. pestis. (See fig. 58 on Pl. O.) Sometimes the original membrane may be left behind,
which fact naturally is more easily discovered with the large trichobacteria than with smaller
species. Drawings made by Billet (1890) of germinating Cladothria gonidia, reproduced as
figure 70 on Plate P (from original figs. 4 and 5 on Pl. IV) may serve as an illustration. Wolo-
schin (1913) once observed that under certain conditions the capsules of B. anthracis break up,
and that then from these apparently empty broken sheaths new bacilli may germinate, a fact
which induced him to formulate the somewhat vague hypothesis, that the substance of the cap-
sules itself be endowed with an “‘embryonic life’. Undoubtedly, it is much more probable
that those unstainable gonidia, especially frequent with spore-forming bacilli, after budding
out from the parent cell have remained within the equally unstainable capsule, thus apparently
becoming a part of it. ,

That the resistance of the gonidia is not much higher than that of the vegetative cell has
been pointed out before. They are organs for multiplication and reproduction and do not
possess any special protection against deleterious influences, except that given by the higher
concentration of their plasmatic body and by the resulting reduction in moisture content.
Naturally they can not withstand boiling water, but it has been frequently noticed that other
noxious influences active in nature which destroy the life of the vegetative cell can be endured
by the gonidia, which, therefore, also in this respect may occasionally play an important réle.
Finkler and Prior (1885), for instance, ascertained that the gonidia of their vibrio proved to be
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 131

only a little more resistant at 65° C., but much more so against sharp drying at room temper-
ature (above P,O,); and in old cultures they retained their reproductive power for months.
According to Esmarch (1887) dried cultures of his Spirilum rubrum, which had not yet pro-
duced their gonidia, died within 6-8 days, while the vitality under the same conditions was fully
retained for more than five weeks, when the gonidia were formed, though these also in this case
were killed in water of 52° C. within five minutes. Bliesener (1901), too, found that the gonidia
of V. cholerae were killed if exposed to 50°C. for 30 minutes. Sharp drying also proved to be
detrimental; but in water the vitality was retamed for nearly three years, and the author,
therefore, thinks that under natural conditions they may occasionally conserve the life for a con-
siderable length of time. The results obtained by N. K. Schultz (1901) with B. pestis, and by
Rothert (1902), as well as by Fedorowttsch (1902). and by Martini (1910) with several other non-
spore-forming bacilli (Staphyloc. pyogenes, Streptoc. pyogenes, B. coli, typhosus, dysenteriae
pneumoniae, pyocyaneus, septicaemiae, and others) were quite similar. Old sealed cultures of
the plague bacillus were found by Schultz to be still alive after four years, and Fedorowitsch
reported that a culture of B. pyocyaneus died within 3-8 days, if it was dried when only 5-6
hours old, but that it lived for 38 days when it was allowed to grow 24 hours, so as to form goni-
dia before being dried. With Bac. cholerae gallinarum the following data were recorded:

Age of culture before drying...............-.--.-.-+-- 4-8 hours. 24 hours. 5 days.

Limit of vitality was reached after..............-.-..- 22-25 days. 97 days. 84 days.

B. coli was completely killed, according to Schultz and Ritz (1910), if a 3-6 hours culture
was kept for 25 minutes at 53° C., while 12-24 hour cultures gave about the same number of
colonies on the plates made from the not heated and from the heated suspensions.

The very considerable resistance of diphtheria and tubercle bacilli against drying, which
Chester (1901) connects with the presence of ‘‘gonidia”’, is evidently due to their having changed
into the more resistant regenerative bodies, though a somewhat greater resistance of the genuine
gonidia in this case is also quite probable. Jessen and Rabinowttsch (1910)reported that the
small acid-fast granules of tubercle bacilli remain intact, when the bacilli themselves are solved
by caustic alkali. The so-called granular decomposition of cholera and of other organisms in the
serum is undoubtedly, partially at least, another indication of the increased resistance of the
gonidia. And the same holds true with regard to the fact that slightly detrimental influences,
which reduce the vitality of the vegetative cells, stimulate the formation of the gonidia, a fact
which has been studied by Ruata (1905) with V. cholerae, which was treated with free ammonia.
Balfour (1911 a) has emphasized the greater resistance of the gonidia of spirochaets against
salvarsan, and the resulting difficulty in curing chronic spirochaetosis. This point deserves
to be studied, mutatis mutandis, in other chronic diseases, too.

(2). REGENERATIVE BODIES AND EXOSPORES.

According to the definition given on page 121 the so-called regenerative bodies are produced
either by the gonidia or by the symplasm. It would be preferable if only the first category of
them could be considered here and the others be left to Chapter III. But the data available
at present often do not allow such a sharp distinction, and therefore the subject must be treated
in a somewhat loose manner. Details relating to the transformation of the symplasm into
regenerative bodies will be discussed in Chapter III; the appearance and behavior of the fully
developed regenerative bodies, however, will be better considered here, as it is often not only
difficult, but even impossible to decide exactly whether observations recorded in the literature
refer to regenerative bodies developed from the gonidial or from the symplastic stage.

The first type of formation of regenerative bodies, consisting in the upgrowth of gonidia
to bodies of larger size and of more solid structure, is very common among the bacteria and
has been noticed, therefore, by many investigators. Especially the development of the large
deeply staining globular regenerative bodies at the side or at the end of the normal bacteria,
has often evoked more or less interest. The photographs reproduced as figures 116, 124, 126—
129 on Plate X, nearly all on Plate XI, figures 170-172 on Plate XII, those on Plate XIII,
and the drawings made by Henneberg (1898), which were reproduced as figure 57 on Plate N,
132 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor. XVI.

will convey some idea of their general occurrence. In addition, a few drawings made by Niessen
(1898) from his so-called syphilis bacillus have been reproduced asfigure 194 on Plate XV,
because they demonstrate especially well the characteristic appearance of this type of repro-
ductive organs. The following description, given by Fuhrmann (1906) also deserves to be
quoted; after having discussed the réle played by the ‘‘clubs’’ (i. e. gonidangia) in the life
history of Pseudomonas cerevisiae and related species he continues:

Es sind noch eigentiimliche Bildungen zu nennen, die dann auftreten, wenn die verlangerten Stabchen in opti-
male Lebensbedingungen gebracht werden. Es bilden sich dann am Stabchen kleine Warzchen, die sich vergréssern
und endlich ablésen und frei in der Flissigkeit schweben. Bei manchen von ihnen kann man nun eine Aufteilung
des Inhaltes beobachten. Die Teilstiicke werden dann frei und gleichen allerkleinsten Stébchen und Piinktchen.

When the formation of the regenerative body takes place at the end of the cell, the resulting
picture resembles very much that of a plectridium, but the regular bright spore is replaced
by the deeply staining regenerative body. The drawings reproduced as figure 71 on Plate P may
illustrate this possibility. They were taken from a paper upon Spirochaeta icterohaemorrhagica,
published by Inada, Ido, Hoki, Kaneko and Ito (1916, original figs. 40-47 on Pl. 61).

With micro- and streptococci it is often difficult to discern exactly whether the body
formed should be classed as regenerative body or as microcyst. Our photographs of Microc.
candicans, reproduced as figure 168 on Plate XII, and the drawings made by Babes (1895)
of branching streptococci, reproduced as figure 3 on Plate B, may serve as examples in this
direction. On the other hand, our photograph of Streptoc lactis, reproduced as figure 170
on Plate XII, demonstrates equally well the budding out of the gonidium and its upgrowth
to a lateral regenerative body. A picture of Streptoc. pyogenes published by Hewlett (1902)
and reproduced as figure 72 on Plate P (from original fig. IIa) shows the same details, as does
also the first sketch of those made by Hlava (1902) of his Leuconostoc hominis, which have been
reproduced as figure 4 on Plate B.

Among the nonspore-forming short rods very instructive preparates are. easily. obtainable
especially with B. radicicola and with the lactobacilli (B. acidophilus, casei, etc.). Two charac-
teristic photographs of B. radicicola made by Hiliner and Stérmer (1903, original figs. II, 1
and 2) are reproduced as figures 195 and 196 on Plate XV. Bac. bulgaricus has been shown
in figures 148 on Plate XI and 172 on Plate XII; other pictures of closely related organisms
may be found in papers of Weigmann, Gruber and Huss (1907), Kuntze (1908), and White and
Avery (1909). The picture made by Tissier (1900) of B. bifidus, reproduced as figure 8 on
Plate C, is also of interest in this connection. Some of the photographs by Almquist (1916~
1917), showing regenerative bodies of B. typhosus, dysenteriae and of other nonspore-forming
bacteria, were reproduced as figures 135-138 on Plate XI.

The branched threads of B. subtilis reproduced as figure 197 on Plate XV from our second
preliminary paper (Léhnis and Smith, 1916 b, fig. 27) exhibit at the same time small, pale,
budding gonidia, well developed regenerative bodies in lateral, as well as in terminal, position,
and also an unstainable body, apparently a growing exospore. Interesting details concerning
the formation of regenerative bodies by the gonidia of Azotobacter have been published by
Prazmowski (1912, pp. 145-147), who paid special attention to their development from the
liberated gonidia. The so-called micro-oidia, described by Bredemann (1909) for B. Amylo-
bacter and by Viehoever (1913) for B. Pasteurii, are probably also to be classed as regenerative
bodies, though they have not been seen to reproduce normal rods. Similar round forms have
been observed by F. Klein (1883) with B. anthracis, but here their formation, as well as their
ability to reproduce normal rods, have been carefully studied and ascertained.

With spirilla and vibrios lateral, as well as terminal, regenerative bodies are equally
frequent. The so-called arthospores of V. cholerae found by Hueppe (1885), whose drawings
have been reproduced as figure 49 on Plate M, as well as those described by Firtsch (1888) for
V. proteus, have been either regenerative bedias orexospores. The deeply staining resistant
organs described by Weibel (1888) for his V. lingualis are also to be classed as regenerative
bodies.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 133

With spirochaets these lateral or terminal swellings are still more conspicuous, and they
have been noticed accordingly by practically all authors, who have made such studies. éinstler
(1885) probably has been the first to mention this occurrence, when he described his ‘‘ Protero-
monas Regnardi.” ‘The spirilliform organism isolated by Calmette (1893) from typhus cases,
whose ‘‘spores’’ were found either to multiply as such (by budding) or to reproduce new
‘‘spirilla,”. apparently also belongs in this group. More recent and more complete, though
often misinterpreted, evidence relating to spirochaets has been furnished by Prowazek (1906),
Mihlens and Hartmann (1906), Schaudinn (1907), Fantham (1911), Hindle (1911), Noguchi
(1911-1912), Wolbach (1914), Metrowsky (1914), Inada, Ido, Hoki, Kaneko and Ito (1916), as
well as by Leishman (1918). Noguchi (1912) has pointed out with good reason, that these
formations ‘‘should not be viewed as the result of plasmoptysis’’ because ‘‘the number of the
organisms with round bodies . . . seems to become greater where the growth of the
culture approaches its maximum”’ and the development of new spirochaets from these bodies
also has been observed. Some very interesting pictures made by Inada et al. (1916) have
been reproduced as figure 31 on Plate H; they make a good counterpart to our Subtilis photo-
graph (fig. 197 on Plate XV).

That fusiform bacilli behave in the same manner as spirochaets do, was noticed by Eller-
mann (1907), but he discarded the round bodies once more as ‘‘involution forms.”

Most frequently their presence has been recorded with the Mycobacteriaceae. The fairly
resistant, but easily stainable round reproductive organs formed usually at the ends, but also at
the sides of B. tuberculosis, leprae, mallet, diphtheriae, and pseudo-diphtheriae have attracted
considerable attention. The earliest observations in this direction have been those of A. Neisser
(1881) and G. A. Hansen (1882) upon the formation of globular ‘‘spores” of B. leprae. They
were confirmed by Babes (1883), who in addition discovered the analogous organs of B. tubercu-
losis, and more completely by Lutz (1886). Similar results were later recorded for B. leprae
by Bordoni- Uffreduzzi (1888a), Czaplewski (1898) and by Meirowsky (19146), for B. mallet by
Semmer (1895), Gallt- Valerio (1900) and by Carpano (1913), for diphtheroid organisms by
Nakanishi (1900c) and by E. de Negri (1916), for B. diphtheriae by Bergstrand (1918), for
B, tuberculosis by Sander (1893), Semmer (1895), Crookshank (1896), Droba (1901), W. Ernst
(1902), Meier (1903), Schroen (1904), Rosenblat (1905), Betegh (1908), Meirowsky (19146), and
by others.

The deeply staining ‘‘spores” seen by Schiitze (1908) to grow singly at the end of the
branches of Actinomyces monosporus evidently have also been regenerative bodies, and the
same seems to hold true with regard to the spherical spore-like body found by Metchnikoff (18886)
to be produced within the club-shaped cells of his Pasteuria ramosa.

The form of the terminal regenerative bodies is either globular, oval, or pear-shaped, while
those in lateral position are either spherical or kidney-shaped. After being liberated, they all
seem to assume a perfectly spherical form. Whether the regenerative bodies of irregular shape,
frequently to be seen in the cultures, are exclusively produced by the symplasm or by normal
vegetative cells, also, can not be decided at present. The first-named possibility, however, seems
to be more probable. -

All regenerative bodies take the stain very readily, so that under the microscope, as well as
in the photographs, these comparatively large dark bodies are very conspicuous.

Motility seems to be absent as a rule. In those cases, where motile gonidia grow up to
regenerative bodies after being liberated, these too, may retain motility for some time, though
it is, of course, under such conditions a more or less disputable point, at what moment the
“‘gonidium” actually becomes a ‘‘regenerative body.” Especially with V. cholerae repeatedly
large globular bodies have been observed, e. g. by Dowdeswell (1889), Hammer! (1906), Almquist
(1908) and Bittrolff (1912), which, indeed, may have been motile regenerative bodies. Almquist
(1907) reports, that also in the case of B. typhi he saw motile globules separate themselves from
the bacilli, which being able to germinate promptly apparently have been genuine regenerative
bodies.
134 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vau: XVI.

Further development from the regenerative bodies is naturally more easily to be ascertained
than it is usually the case with the gonidia, But here again occasionally fragmentation and
dissolution may be seen, which processes should not be taken by themselves as proof of ‘‘involu-
tion” and death. Generally the irregularly shaped regenerative bodies seem not to be able to
reproduce directly normal vegetative cells. One often sees them either breaking up into round
regenerative bodies, which germinate later, or melting together to a flake of symplasm, which

in its turn may produce either directly normal vegetative cells or another set of regenerative
bodies of regular shape. Such fragmentation and dissolution have been repeatedly reported,
for instance with so-called bacteroids of the root-nodule bacteria by Beijerinck (1888), Moeller
(1890), Hiltner and Stérmer (1903) and others. The irregular regenerative bodies of spore-
forming bacilli, as well as those of micrococci, behave in the same manner, as was discussed in
our preliminary papers (Léhnis and Smith, 1916 a and 6). That Buchanan and others have
seen the ‘‘bacteroids” of B. radicicola proliferate as such or produce small motile rods by
segmentation, should not be considered to be irreconcilable with the facts just mentioned.
‘‘Bacteroids” behaving in this manner are merely branched vegetative cells, which may be
morphologically more or less similar to those other forms, though they are different physiologically.

The globular regenerative bodies are much inclined to multiply as such, before reproducing
normal vegetative cells. Like the gonidia, they multiply by budding as well as by fission; both
processes may be often seen to proceed side by side in the same preparation. The resulting
‘‘pure cultures” of budding, relatively large, coccoid cells may look very similar, though
actually belonging to entirely different organisms. Eyre and Washburn (1897) obtained such
a luxuriantly growing, avirulent, gelatin liquefying culture of resistant micrococci from a typical
pathogenic Pneumococcus. Numerous photographs, reproduced on our plates, illustrate the
analogous behavior of various species as observed by different authors. These are:

II, 18: Bact. brunewm (fluorescens var.), Matzuschita (1900).
III, 30: Bact. cholerae gallinarum, Itzerott and Niemann (1895).
III, 35: Bact. pestis, Rowland, (1912).
V, 59: Bac. anthracis. Henri (1914).
VI, 68: Bac. Azotobacter, Léhnis and Hanzawa (1914).
VIII, 91: Bac. mallei, Carpano (1913).
IX, 97: Bac. diphtheriae, J. Dale (1910).
TX, 98: Bac. variabilis lymphae vaccinalis, Nakanishi (1900c).
IX, 101: Corynebacterium of malignous granulom, E. de Ne (1916).
IX, 104: Bac. tuberculosis, Arrigo (1900).
_ XI, 146: Streptococcus lanceolatus, Axelrad (1903).
' XI, 147: Bact. coli, Axelrad (1903).
XI, 154: Bac. leprae, Kedrowski (1910).
XII, 155 and 156: Meningitis bacillus, Ghon, Mucha and Mittler (1906).
XII, 157: Bact. cyanogenes, Neelsen (1880). g
XII, 158: Vibrio cholerae, Hammerl (1906).

That the presence of these globular regenerative bodies in practically pure growth within
the organism may sometimes cause serious diagnostic errors, has been already mentioned with
regard to diphtheria on pages 79 and 80. The general occurrence of such deceptive forms within
all groups of bacteria is certainly of considerable importance, and a careful study of their
morphological and physiological behavior is much to be recommended. In this connection an
observation made by Calmette (1893) in regard to a spirilliform organism, isolated from typhus
cases, is of interest. The round regenerative bodies grew constantly as such, multiplying by
budding, when they were kept on an acid substrate, while an alkaline reaction and substrates
rich in albumines stimulated their germination and the regeneration. of spiral forms,

It seems not unlikely. that several descriptions of so-called cocci actually refer to regenera-
tive bodies of other species. If these ‘‘cocci”’ later changed to rods, or if in old cultures of
bacilli such ‘‘cocci’”’ have appeared, the ‘‘contamination’’ theory undoubtedly has been often
invoked. erroneously. I myself have to plead guilty, that for nearly two years I have rejected
as ‘‘impure” actually pure cultures of aerobic cellulose destroying bacteria, which had been
isolated by Grant Lochhead (unpublished investigations 1912-1914), because éavatul examination
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 135

of his preparations invariably revealed the reappearance of ‘‘cocci,’”’ which, as our later studies
(Lohnis and Smith, 19165) have brought out, are constantly to be found in this as in every other
group of bacilli. What perplexing pictures are to be had especially from these aerobic cellulose
destroying rods may be seen from figure 265 on Plate XXI, which was reproduced from the last-
named paper (orig. fig. 52). It was made from a frequently tested pure culture of Bact. acidulum,
originally described by Kellerman, ‘McBeth, Scales and Smith (1913) as a slender short rod
(average dimensions 1.0 0.3y).

Even with regard to the Nitrosococcus of Winogradsky it is by no means improbable that this
is no coccus at all, but a pure culture of regenerative bodies of Nitrosomonas. The photograph
made by Winogradsky (1891), reproduced as figure 198 on Plate XV, shows besides round cells
of rather different size, i. e., a peculiarity much more frequent with regenerative bodies than with
genuine cocci, some small short rods, which according to Winogradsky represent an ‘‘impurity;”’
the photograph is said to have been made from a ‘‘vieille culture sur silice, déjaé devenue un peu
impure.” However, in the lower right corner of figure 198 we see a larger rod connected with a
round body exactly in the same manner, as it is shown by all bacilli when producing regenerative
bodies and in the upper left corner something like a torn, empty membrane is visible, whose
presence also fits much better to the theory that regenerative bodies, not genuine cocci, are
represented in this picture. Additional evidence will be furnished in Chapter III. A com-
parison of figures 198 and 195 (Nitrosococcus and regenerative bodies of B. radicicola) is also
to be recommended.

The multiplication of irregularly shaped rod-like regenerative bodies is sometimes to be
observed. with micrococci, and occasionally it becomes so firmly established, that the cultures
present a picture much more resembling an Actinomyces than a Micrococcus. Figures 2-6 and
8 on Plate I illustrate this possibility. It is readily to be admitted, of course, that our present
knowledge often leaves us in doubt, whether such forms should be classed as regenerative
bodies or as truly vegetative cells. Their staining reaction, as well as their ability to reproduce
regular micrococci, may often serve as a valuable indication, and their development from the
symplasm may supply additional evidence. Further investigations, however, are indispensable
to learn more upon the vegetative or reproductive character of such forms. .

The germination of rod-like forms from round regenerative bodies has been observed quite
frequently viz. with B. cyanogenes by Neelsen (1880), with B. typhosus by Almquist (1904), with
B. paratyphosus, dysenteriae, and pseudodysenteriae by the same author (Almquist, 1908), with
B. colt by Kellerman and Scales (1916) , with B. radicicola by Spratt (1912), with Proteus hominis
capsulatus by Bordoni- Uffreduzzi (1888b), with V. cholerae by Hueppe (1885), Babes (1889),
Nakanishi (1901) and by Almguist (1904), with various other vibrios by Firtsch (1888), Kohl-
brugge (1900) and by Nakanishi (1901), with spirochaets by Noguchi (1912 d), with diptheroid
organism by Nakanishi (1900c) and by E. de Negri (1916), with B. leprae by Ozaplewski (1898)
and by Meirowsky (19146), with B. tuberculosis by Rosenblat (1905) and by Meirowsky (19146).
The pictures made by the last-named author of B. tuberculosis, leprae, and Spirillum rubrum,
which were reproduced as figures 164, 166, and 167 on Plate XII, as well as his drawings shown
on Plate XIV in the column ‘‘Freie Knoapen and Jugendformen,’ ’ should be compared with
some photographs made by Kellerman and Scales of germinating regenerative bodies of B.
coli, reproduced as figures 199 and 200 on Plate XV. That two or three sprouts may break

‘forth simultaneously, as demonstrated in the last-named picture, seems to be no rare occurrence.
Almquist (1906) mentioned this fact in regard to B. typhi, and a photograph made by Stamm
(1914), reproduced as figure 201 on Plate XV, illustrates the same type of growth with V.
cholerae. The picture is of special interest, because it shows side by side with germinating
regenerative bodies others, which evidently are developing to gonidangia.

The higher resistance of the regenerative bodies against heating, drying, and other unfavor-
able conditions has evidently exerted its influence where it was noticed that cultures of non-
spore-forming bacteria sometimes exhibit a quite considerable resistance. Details in this direc-
tion have been repeatedly collected, e. g., by Ficker (1898), Rothert (1902), P. Eisenberg (1908),
Lehmann and Newmann (1912), and by Bernhardt (1915). Own experiments with V. cholerae
136 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

furnished the first-named author the following interesting data. Cultures of different age
died, when. kept in the humid chamber at 37° C, after 14, or after more than 50 days in ‘the
following’ succession :

, fie HR? : Days. . Days. Days. Days.
Age of the cultures...... caehee sen on eeeade amareinns Bene vonets 1 2 3 8
Duration of life.../... 5.00522... eset ss bt hrae aoe 3 14. More than 50

Microscopic studies apparently have not heen mada in this case, though older observations
by Firtsch (1888) and by Weibel (1888) upon the increased resistance of the globular: bodies
present in old cultures of other vibrios should have suggested such investigations. Firtsch
had noticed that the vegetative cells of V. proteus died when dried more than }-4 hour above
sulphuric acid, while their ‘‘granules”’ still germinated after having been dried for. 12} hours;
and Weibel had found that the ‘‘spores” of his V. saprophiles withstood the influence of moist
heat of 53° C. for more than 24 hours. Cantacuzéne (1898) reported that the round forms of
V. cholerae were killed only by temperatures higher than 80° C., and Almquist (1916) succeeded
by heating cholera cultures up to 60° C. in obtaining a strain, which for two years grew con-
stantly in the round form, but still showed regular agglutination.

Gage (1903) noticed that most cells of B. coli and typhosus were killed by temperatures of
45-55° C., while some, however, withstood heating up to 85°C. Freezing acted in an analogous
manner upon cultures of B. fyphosue, campestris, and others, when tested by EZ. F. Smith and
Swingle (1905); the authors suppose that ‘ ‘arthrospores” were present in their cultures. Many
round and irregular forms were seen by Preisz (1904) in the secondary colonies, developing i in
old cultures of numerous bacteria, which also showed a considerable increase in longevity.
Spratt (1912) reported that the round regenerative bodies of B. radicicola were even not killed
by boiling water, and Rodella (1908) made use of the higher resistance of the analogous organs
of the lactobacilli for isolating them: he inoculated material 4-5 days old into melted agar kept
at 80° C. and left it therein for 2-3 minutes before lowering the temperature.

. With diptheroid organisms similar results have been recorded by A. Neisser (1888), hone
B. werosis in 6-days-old cultures withstood 70° C. for 15 minutes, by E. Klein (1900), whose
Bacterium diptheroides died usually within one week, but lived for many weeks in serum, where
it regularly produced oval or spherical cells instead of the normal rods. The Corynebacterium
studied by F. de Negri (1916) was also killed at 60° C. within one hour, when tested in the rod
form, while the budding round cells remained alive. The round deeply staining bodies produced
by Streptobacillus pellagrae resisted 80° C., according to Tizgont and Angelis (1913). B. mallet
showed only low resistance when tested hs. Marx (1899), but Bondme, according to Lehmann
and Newmann (1912, p. 548), found that occasionally a temperature of 70° C. was endured for
6 hours and that of 90° C. for 3 minutes. Carpano (1913) considered the round deeply staining
“‘spores’’ of the glanders bacillus to be the reason of its remaining alive in the cultures for more
than a year. Schiiize (1908) noticed that the regenerative bodies (‘‘spores’’) of Actinomyces
monosporus endured the temperature of 75° C. for 40 minutes. _

Sometimes the reproductive organs growing at the side or at the end of the bacteria do not
show the characteristic appearance of the normal deeply staining regenerative bodies, but assume
more the character of the endospores, especially with regard to their resistance against the ordi-
nary staining methods. On account of their external position we call such bodies exospores.
Our photographs of B. subtilis, reproduced as figure 197 on Plate XV, and of B. Azotobacter, .
figure 69 on Plate VI, may illustrate this occurence. The simultaneous production of endo-
and exospores in the latter case is of special interest. On the other hand, the replacement of
deeply staining regenerative bodies by unstained exospores, as shown with B. subtilis, seems to
be very frequent with spirilla and vibrios. It is not improbable that the so-called arthro-
spores of V. cholerae, described by Hueppe (1885), Babes (1889), and others, have been, partially
at least, exospores. Weibel (1888) ) observed with his different vibrios side by side stained and
unstained round bodies, just as we did recently with some salt-water spirilla (Léhnis and Smith,
19164). Of his Spirobacillus Cienkowskii Metchnikoff (1889) made the drawing reproduced
as figure 73 on Plate P (from original fig. 14 on Pl. 1). Spirillum monospora, described by
Dobell (1908), seems to have exhibited a similar appearance. The spore-forming anaerobic
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 137

cellulose destroying bacteria, studied by Praémowski (1880) and by Omelianski (1902), are
another example to be mentioned here. As they are often curved, the first-named author used
the old name Vibrio Rugula for them, but probably they should be better placed in the neigh-
borhood of B. oedematis maligni. His drawings, showing the terminal spore formation, have
been reproduced as figure 28 on Plate H. Omelianski’s photographs seem to indicate that also
in his preparations some of the terminal bodies took the stain, and apparently some lateral
regenerative bodies were present, too. In addition his findings concerning the heat resist-
ance of these spores deserves our attention; they withstood 95°, but died quickly at 100° C.
It is true that the terminal spores of B. tetant and related anaerobic bacilli are characterized by
a.similarly low resistance. But it is a disputable point whether it is altogether correct to class
them as endospores, as is usually done. Jn my opinion, they too should be accepted as exo-
spores. Their position, forms, and degree of resistance links them obviously more closely to
those exospores mentioned above than to the centrally located, highly resistant endospores,
formed by the bacilli of the Subtilis-Mesentericus group.

Exospores are also produced occasionally by many other bacteria. Lutz (1886) noticed
that B. leprae sometimes furnishes such bodies. With B. tuberculosis terminal spores were
recorded by Ff’. Fischel (1893), Hawthorn (1903 6), and by Wherry (1913), with B. pseudo-diph-
theriae by Simons (1898) ; in the latter case they survived 80° C for more than 10 minutes. Prob-
ably those spores of B. tuberculosis and diphtheriae, which according to the observations made by
Marpmann (1893) and by Babes (see Cornil and Babes, 1890, Vol. II, p. 59), respectively, were
even not killed by boiling, are also to be classed here.

Furthermore, the following so-called nonspore-forming bacteria have shown themselves
to be able, under conditions not yet well known, to assume the faculty to produce fairly resist-
tant exospores (or, as they have been usually called, polar endospores). They are B. acidi
lactict and B. cyanogenes (Hueppe (1884), B. typhi (Gaffky, 1884), B. cola (Piccola, 1896), Bact.
lactis acidi ( Weigmann, 1899) and Bact. Zopfir (Swellengrebel, 1904). It has been mentioned
on page 95 that some of these observations, especially with regard to typhoid and lactic acid
bacilli, have been repeatedly refuted. Nevertheless, they, too, deserve a renewed study. That
in all these cases invariably, and in most of them quite independently, polar spores have been
found, is certainly not without importance. On the other hand, some spore-forming bacilli
have been described, showing polar spores, too, whose general character brings them very close
.to.usually nonspore-forming bacteria. These are: B..esterificans Maassen (1899), closely related
to B. coli; B. violarius acetonicus Bréaudat (1906), evidently a spore-forming strain of Bact.
violaceum; and an unnamed morphologically tetanus-like bacillus, growing yellow or red (Mor-
genthaler, 1916), just like Bact. herbicola. A closer search among the thousands of species de-
scriptions would undoubtedly furnish more such evidence. But it is of much greater interest,
that. also the so-called denatured butyric acid bacilli, which have lost their ability to produce
normal endospores, which, however, by certain methods can be induced to turn again to sporula-
tion, were found by Grassberger and Schattenfroh (1907) to form round polar spores, like B.
tetanit. On the other hand, an old weakened culture of B. anthracis has been studied by Chau-
veau and Phisaliz (1895), which had lost its faculty to produce endospores, but instead of these
formed quite constantly terminal round bodies, which, however, were killed by temperatures
above 70° C., and which could be stained by aqueous dyes, i. e., they were normal round regen-
erative bodies.

My own findings with Azotobacter and various other usually nonspore-forming bacteria
are in full agreement with all these observations. It seems as if the production of terminal
round regenerative bodies and of exospores is a necessary step which has to be passed when
the formation of endospores is to be developed. Some-genera (Spirillum, Vibrio) seem to be
unable to reach the latter stage, but in regard to the so-called nonspore-forming rods it will be
well worth trying whether or not the newly won insight will open experimental possibilities,
which so far never have been tried. A few additional remarks upon this problem will be
made on the following pages.
138 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor, XVI.

(3) ENDOSPORES.

Formation of endospores may occur within the vegetative rods, as well as within gonidangia
and regenerative bodies. Thus far only the first possibility has been studied more thoroughly.
According to Zopf (1883, p. 66) the granules first visible within the cell unite in making up
the spore, but L. Klein (1889) studied some species which did not show any granulation of the
cell plasma preceding the formation of the spore. As Hueppe (1891, p. 29) pointed out, either
the whole content of the cell or only a part of it may be used up for that purpose. Bunge (1895)
accepted only a special class of cell granules as the true elements of spore formation; others,
like those seen by P. Ernst, were declared to be in no way connected with it. That, however,
here as in other cases staining reactions are not very conclusive is clearly indicated by some
results recorded by Zettnow (1899), who saw the spores, when they were treated with Romanow-
sky’s method, turn red or blue or remain unstained, sometimes only showing a blue spot in the
center. Nakanishi (1900 a and 6) declared that the endospore be nothing else than a changed
nucleus, but he soon modified this statement, saying that the nucleus surrounds itself with
chromophilic cytoplasm and so becomes a spore. The spore formation of the large B. Biitschhit
and sporonema was accepted by Schaudinn (1902-1903) as some primitive sexual process, to
which reference will be made in Chapter IV. The granules present in the cell were seen
to increase in number and in size before the spores were formed. Riditka (1909) defined
spore formation as concentration of the chromatin substance of the cell; later, however, he was
of the opinion that the chromatin center ‘‘vanish”’ in old spores. With Bac. anthracis Péneau
(1911) observed that young cells appeared fairly homogeneous with only some small granules
present, which after about 12 hours formed one definite nucleus; this in its turn, about 8 hours
later, dissolved into a chromidial system, which ultimately, in the 3-4 days old culture, concen-
trated itself again, now making the spore. Swellengrebel (1913) saw that in his B. deliensis the
chromatin granules arranged themselves first into straight centrally located or into zigzag
filaments, which then in part and under assistance of part of the cytoplasm formed the spore,
while the ‘‘vegetative”’ plasma and chromatin, thus separated from the ‘‘generative’” plasma
and chromatin, later dissolves with the rest of the cell. The unequal and in most cases unstable
distribution of the nuclear material within the bacterial cell, which has been discussed on page
109, makes all these more or less divergent findings concerning spore formation quite
comprehensible.

Occurrence of two endospores within one cell has been recorded by A. Koch (1888) for.
B. inflatus and ventriculus, and by Vuillemin (1903) for Clostridium disporum. The spore-
forming cells in these cases have been all more or less inflated. The last-named author dwells
upon this point, reaching the conclusion that spores formed under such conditions should be dif-
ferentiated from the normal endospores as ‘‘cystospores,’’ and the sporulating inflated cell
should be called accordingly ‘‘sporocyst.’’ That this point of view was fairly correct is proved
by the fact that double and multiple spore formation may be shown by regenerative bodies, as
well as by those inflated globular, pear- or club-shaped cells, which normally act as gonidangia.
As it is fundamentally the same nuclear material, which, in the form of gonidia, regenerative
bodies, exo- or endospores, serves the one purpose of reproduction, it is by no means surprising,
that one type of reproductive organ may replace or may pass over into another, especially when
we keep in mind that all these separations and classifications are much more artificial than
natural. The photographs reproduced as figures 57 on Plate V, 65 on Plate VI and 117-121
on Plate X, showing the apparent or real beginning of spore formation within irregular cells
(so-called involution forms, i. e. regenerative bodies and gonidangia) of B. anthracis (Mat-
guschita, 1900), B. subtilis (Loéhnis and Smith 1916 a), B. Chauvoei (Fraenkel and Pfeiffer, 1895),
B. pestis and lactis aerogenes ( Maassen, 1904), Proteus vulgaris (Hauser, 1885) and of V. cholerae
( Maassen, 1904), deserve to be studied closely, as they are very suggestive for further investi-
gations along these lines.

A. Fischer (1891, 1903, p. 39), as well as Migula (1897, Vol. I, p. 166), were of the opinion
that the co-called nonspore-forming bacteria would be eventually found to be able to produce
endospores, too; to the first-named author this fact was ‘‘beyond doubt.” His suggestion that
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 139

increased salt content of the substrate will be favorable in this direction, seems to be correct, as
was discussed on page 99. And, as far as the nonspore-forming rods come into view, it is indeed .
probable that they can be induced to form regular endospores or to show themselves to be
parts of the life cycles of spore-forming bacilli. With micro- and streptococci, spirilla, spir-
ochaets, and mycobacteria, however, the situation seems to be different. In addition to what
has been said above (p. 137) concerning the beginning of spore formation by so-called nonspore-
forming rods, it may be pointed out that formation of spores has also been recorded with lacto-
bacilli by C. Sternberg (1898) and by Fokker (1901), with B. piscicidus, a member of the Proteus
group, by WN. Sieber (1895), with members of the Coli-Aerogenes group by Rogers, Clark and
Davis (1914), as well as by Burton and Rettger (1917), and with B. fluorescens by Tanner (1918).
Microscopic and cultural characters showed otherwise little or no alterations in these cases;
similarly, the small monotrichous Bac. leyuminiperdus, described by Oven (1906), displayed
much more the appearance of a common ‘‘Pseudomonas’’ than that of an endospore-forming
rod. The situation becomes more complicated when the establishment of the ability to form
endospores is accompanied by profound changes in morphology and cultural behavior of the
organism. Yet this is to be expected and perhaps the a in all those cases, where genuine
endospores are developed, not only those polar ‘‘exospores”’, which stand between regenerative
bodies and true endospores. When Noguchi (1910) transformed experimentally the anaerobic
nonspore-forming B, bifidus into an aerobic bacillus closely resembling B. mesentericus fuscus,
the ever-ready ‘‘contamination” theory was eliminated by the fact that the process could be
reverted at will, Our own observations are in complete agreement with these findings. There
are evidently clearly defined subcycles within the life cycles of the bacilli, which are not only
characterized by the absence or presence of endospore formation, but by numerous other dif-
ferences, too, which fact has thus far practically always misled to the assumption that two or
more different species had to be separated, because a close study of the link connecting the
different subcycles, viz., of the symplasm, has never been recommended before.

The observations made with cultures of otherwise endospore-forming bacilli, which have
lost this ability are of special interest in connection with the facts just mentioned. Grassberger
and Schattenfroh (1907) and many other investigators have noticed that anaerobic butyric
acid bacilli not infrequently turn over into quite different aerobic nonspore-forming rods,
which, like the asporogenous strains of the Anthrax-Subtilis-Mesentericus group, are micro-
scopically, as well as culturally, much more similar to B. coli or aerogenes or to some lacto-
bacillus than to their own sporulating parent organism. Sometimes they may also exhibit a
more or less Actinomyces-like growth, as was obtained, for instance, from B. anthracis by
Nadson and Adamovié (1912), as well as by Henri (1914). Of course, there is also the possibility
merely to suppress the spore formation temporarily without otherwise changing the character
of a strain to any considerable extent, as is often easily performed, e. g., by keeping the bacilli
in milk; but this case should not be confounded with the real transformation of a strain, which
change is always accompanied by more or less fundamental, morphological, as well as physi-
ological, alterations, such as were observed, for instance, by Winogradsky (1902), Passini (1905),
Grassberger (1905), and Grassberger and Schattenfroh (1907).

When the accidental loss of spore formation was studied for the first time by Lehmann
(1887), it was found that instead of the regular endospores only “‘micro-spores” were pro-
duced, which, however, were killed when kept for 2-3 hours at 60° C. This observation was
soon confirmed by Rows (1890) and others. More recently Rédicka (1908-1909) paid special
attention to these ‘“‘sporoids,” which he often found enclosed in large, hypertrophic cells. The
same occurrence, i. e., the replacement of endospore formation by the development of goni-
dangia, has been already reported by Weibel in 1888 in regard to asporogenous anthrax bacilli,
and by E. Klein in 1899 as a result of his studies upon B. caiaverts Sporogenes; though, again,
in the opinion of the latter author, these giant cells were “probably involution forms.” P.
Eisenberg (1908) noticed that asporogenous strains of B. anthracis, tumescens, Megaterium,
and ramosus occasionally may exhibit a comparatively high resistance; a temperature of 90° C.
was endured in his experiments for 5-15 minutes. He assumes that the capsules of B. anthracis,
140 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor XVI.

and in the other cases ‘‘Ausnahme-Zellen,’”’ are to be made responsible for this result. That
the latter have been regenerative bodies is very probable. According to Garbowski (19076)
the secondary colonies of B. luteus sporogenes Smith et Baker produced, instead of endospores,
irregular forms, which were declared to be vegetative cells, but which evidently also have
been in reality gonidangia and regenerative bodies. This conclusion is supported by the fact
that Preise (1904) found the irregular, mostly round or pear-shaped cells, present in the sec-
ondary colonies of otherwise spore-forming bacilli, to be fairly resistant; and earlier findings
by Chauveau and Phisaliz (1895), as well as Henri’s more recent observations, leave no doubt
that true regenerative bodies may be produced in large numbers by asporogenous strains of
endospore-forming bacilli. Furthermore, it has been ascertained by Olsen (1897), Cozzolino
(1900), and by Peklo (1910) that club-shaped cells may produce one endospore each, instead
of becoming normal gonidangia. All these facts together may serve as additional evidence
concerning the close relationship existing between gonidia, regenerative bodies and endospores,
and in regard to the possibility of mutual replacement in the one or in the other direction,
How the resistance gradually increases, when gonidia and regenerative bodies revert again to
endospores, has been already studied.to some extent by Phisalix (1892). Addition of. blood
to broth cultures of an asporogenous Anthrax strain was found to be very useful for promoting
this process, but just.as the bright “pseudospores”’ (gonidia) were quickly. killed at 65°C., so
the first new spores also withstood this comparatively low temperature only for a few minutes.

Later, however, the usual high resistance was regained.

That the form and also the resistance of the endospores is by no means so constant, as is
assumed by some authors needs hardly to. be emphasized. With regard to anaerobic bacilli.
these facts have been made clear by Hibler (1908), but the same holds true with the. aerobic
bacilli. An average or “typical’’ form and resistance may, of course, be calculated; these,
however, are again more or less dependent on the conditions of the experiment. That the
resistance of the round terminal spores of the Tetanus-Putrificus group is generally distinctly
lower than that of the regular oval endospores, as produced by Bac. mesentericus, subtilis, etc.,
and that, in addition, “failures” are much more frequent with the former than with the latter,
once more confirms our view that such terminal spores should be better separated from the
endospores, and find their place as exospores between regenerative bodies and endospores.

The germination of the endospores has, also been accepted as being quite constant, and has
been used accordingly by some authors, like Migula, as a means of classifying the spore-forming
bacilli. That this assumption was not correct, and Migula’s procedure therefore not well
founded, has been shown by Caspary (1902). The same holds: true, of course, with regard to
such minor details, as the presence or absence of. a special membranous capsule around the
spore, to which occurrence great importance has been attached in the case of Clostridium
Pastorianum and related forms of the Soe group. The careful studies of Bredemann
(1909) have decided this question.

That endospores sometimes may germinate while still enclosed within the parent cell, as.
was observed, e. g., by Garbowski (19076) with B. asterosporus, furnishes an interesting counter-
part to the analogous upgrowth of the gonidia retained within the parent cell or gonidangium,
as was discussed on page 128.

Not only the mode of germination of the endospores, but the fact itself is by no means so
constant as is often presumed. Many endospores not only refuse to germinate when tested,
they even dissolve entirely and pass over into the symplastic stage, just as is done by many
gonidia and regenerative bodies (see Chap. III). Whether they are also able to multiply by
budding or by fission has never been studied, and the mere assumption that such a case might
be possible is quite liable to be rejected as simply absurd by the well-established bacterio-
logical dogmatism. However, as the complete dissolution of these apparently very solid forms
is not difficult to ascertain, so it is also to be expected that positive results may be secured
occasionally in the last-named direction. I have often seen round darkly staining buds attached
to liberated endospores, looking exactly like the regenerative bodies growing at the side or at
the end of the vegetative cells. The sprouting of young exospores from Azotobacter cells
No. 2] LIFE CYCLES OF THE BACTERIA—LOHNIS. 141

containing at the same time endospores, as shown in figure 69 on Plate VI, is another fact to
be kept in mind in this connection. Maher (1915) speaks of a breaking up of Subtilis spores
into “cocci” or ‘vital granules,” and as early as in 1878 it was reported by Ewart, that he
saw what he calls Anthrax “spores” divide themselves into 2 and 4 parts, while others germi-
nated directly. It is more probable, as was said on page 93, that these spore-like bodies, which
have also been observed by E. Klein (1883) and more recently by Réadzitka (1908), have been
regenerative bodies or microcysts. Nevertheless, they also might be considered to have been
rudimentary endospores, still showing some of the features characteristic for the less resistant
reproductive organs. It is likewise to be admitted, that not the well developed ‘“‘ripe’’ endo-
spores are inclined to enter the symplastic stage, but that this is more a kind of subterfuge
for rather weak endospores, which otherwise would be complete ‘failures. ”’

(4) ARTHROSPORES AND MICROCYSTS.

The formation of arthrospores, i. e., the segmentation of the vegetative bacterial cell and
the transformation of these segments into fairly resistant reproductive organs, seems to have
been first observed by Van Tieghem (18796) with a spirochaete found in oysters. Soon after
Zopf (1881) and later W. H. Hoffmann (1912), as well as Gross (1912), reported upon similar
findings. Some characteristic drawings made by the last-named author have been reproduecd
as figure 63 on Plate O. Kurth (1883) studied the same process with B. Zopfii, where he saw
each rod produce two arthrospores. Hueppe (1885), who like DeBary (1884) believed the
formation of arthrospores to’be very widespread among the bacteria, also accepted the break-
ing up of the rods into coccoid resistant bodies as the characteristic feature of this process.
Both authors, however, have used the term in a rather loose manner; regenerative bodies, as
well as microcysts, have also been called by them arthrospores. The same holds true with
regard to Zopf (1892), who ascribed, for instance, arthrospores to his Bact. vernicosum, merely
because he observed a slightly increased resistance, which, of course, may just as well have
been due to the presence of regenerative bodies. A good example of typical arthrospore for-
mation has been contributed by Matzuschita (1902), when he described his protean bacillus
from air. His drawings, reproduced as figure 5 on Plate B, show clearly side by side forma-
tion of gonidia and their upgrowth within the gonidangia, the presence of lateral regenerative
bodies, the production of arthrospores, and the transformation of whole cells into microcysts.
Most regular and characteristic, however, is the arthrospore formation with the Actinomycetes.
Gasperini (1889), Domec (1892), Kedzior (1896), Doepke (1902), E. Levy (1902), Gilbert (1904),
Lehmann and Neumann (1912, p. 622) have furnished descriptions of this process. Most
typical is the dividing up of the short side branches of aerial hyphae into the coccoid spores,
sometimes also called conidia, in accordance with mycological nomenclature. In the long
threads the development of reproductive organs often stops with the formation of the gonidia,
which then slip out, leaving the empty sheath behind them. This is still more frequently the
case with the Mycobacteriaceae (see p. 124). Some of the papers upon tubercle bacilli, however,
also contain reports on genuine arthrospore formation. And the slight differences recorded by
the numerous authors dealing with “fragments,” ‘‘spores,”’ and “granules” of the tubercle
bacilli, are undoubtedly caused in part by the fact that, on account of incomplete observa-
tions upon the manner in which these: bodies were formed, it could not be ascertained exactly
whether the round forms seen and tested have been ‘eonidia or regenerative bodies or
arthrospores.

The round thick-walled resting cells of the genus Myxococcus have also been repeatedly
called arthrospores. But as most authors who studied the subject have reported that the
whole cell contracts itself, forming one resting body, the term microcysts seems to be preferable.
It is true that it has been observed in some cases, for instance by Kruyff (1908), that the rods
first divide themselves, and it is to be admitted, as was said on page 123, that such occurrences
may be also considered to represent arthrospore formation, but it seems best to place them
in an intermediate position, connecting arthrospores and microcysts.

The form of the arthrospores is, according to their mode of formation, not always exactly
pee or ovoid, Dat more of cubical shape, and as the thickening of the cell wall i is most
142 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI

characteristic for this process, the staining reaction does not show any, or at least not much,
alteration, compared with that of the vegetative cells.

Germination of the arthrospores of spirochaets has been studied first by Van Tieghem
in 1879, while more recently Gross (1912-1913), as well as W. H. Hoffmann (1912), were unable’
to observe this process again. Positive results have been also recorded by Kurth (1883) with
his B. Zopfit, and by Gasperini (1889), Domec (1892), Lachner-Sandoval (1898), Feistmantel
(1902), F. Levy (1902), Glbert (1904), and others in regard to Actinomycetes. These findings
invalidate a statement made by Jordan (1909), p. 407; 1916, p. 463), that the coccoid bodies
producer by Actinomycetes are not “in anywise related to the normal reproduction of the
species.” Not infrequently two to four young threads break forth simultaneously, creating a
picture very similar to that shown of B. coli in figure 200 on Plate XV. The membrane of the
germinating arthrospore seems to be used again, if not always So, at least in most cases.

Multiplication as such has been never recorded with genuine arthrospores.

The resistance of arthrospores has been always found to be much greater against drying
than against heating, although in this direction again fully developed arthrospores are clearly.
more resistant’ than vegetative cells. According to Kurth (1883) the dried rods of B. Zopfit
died after 5-7, the arthrospores after 17-26 days, but heating experiments revealed no differ-
ence. Abnormally high resistance was recorded by Kedzior (1896) for a thermophilic actino-
myces, whose spores are said to have withstood 100° C. (steam) for 34-4 hours. Neukirch
(1902) found 5 minutes at 70° C. to be the limit; Feistmantel (1902) fixed the death point for
spores of ‘Streptothrix farcinica’”’ at 60° C. for 15 minutes. MacCallum (1902) noticed that
young cultures of Actinomyces asteroides were killed at 65° C. within 5 minutes, while older
ones endured 65° for 30, 68° for 5-10 minutes. The vegetative cells of Actinomyces thermo-
philus died according to Gilbert (1904) at 65° within 5 minutes, but the spores withstood 70°
C. for more than an hour. Old dry cultures of Actinomyces violaceus, grown on straw and
practically completely transformed into spores, were found by Berestneff (1907) to be still alive
after more than 10 years. Lehmann and Neumann (1912, p. 623) state that vegetative cells
of Actinomycetes are killed at 60-65° C., whereas the arthrospores endure 75-85° C. for 5 or
3 minutes.

The formation of a microcyst, i. e., the transformation of a whole vegetative cell into one
usually relatively large resting body, is characterized in most cases by a swelling up and round-
ing of the cell, followed by a more or less considerable thickening of the cell wall. With the
curved forms, especially with spirochaets, a coiling up of the body initiates the process. Micro-
cyst formation seems to have been first seen with Streptococci. As early as in 1876 Salomonsen
reported that Streptococci, which otherwise did not produce resistant cells, furnished great
numbers of them when kept in blood at 40°C. Soon afterwards Van Tieghem (1879c) published
his first illustrated description of Leuconostoc mesenterioides, whose ‘‘spores’’ were shown to be
analogous to the heterocysts of the Nostocaceae, in which group of organisms Leuconostoc
itself was later placed by Van Tieghem (1884, p. 1108). Its “spores” were now called “‘ kystes.”’
When studying his Micrococcus ochroleucus, Prove (1887) noticed that the cells, whose normal
diameter was 0.5-0.8u, sometimes increased to 1.6-1.84 diameter and became more resistant
against heating, as well as against, staining. Some of the large dark cells visible in our photo-
graphs of Streptococcus lactis (Plate I, fig. 11) and of Sarcina flava (Plate IT, fig. 15) look very much
as if they had also been microoysts, but no special tests upon germination and increased resist-
ance have been made so far. The drawings of the acetic acid bacteria, made by H#. Chr. Hansen
(1879), show within the chains some large thick-walled cells, which Hansen considered to be
possibly “spores.” (See fig. 15 on Pl. E.) Likewise, the pictures of the “ proteusartige Luft-
bacillus,” drawn by Matzuschita (1902; see fig. 5 on Pl. B), leave hardly any doubt that micro-
cysts were also formed by this.species. An excellent illustration of the same process, as seen
with spore-forming bacilli, has been furnished by A. Meyer (1901); it was reproduced as figure
64 on Plate O. The drawing made by EL. Klein (1883) of the round forms of B. anthracis, which
has been shown as figure 21 on Plate F, furnishes an interesting counterpart. The thick-
walled large cells of Azotobacter are equally typical microcysts. A. Fischer (1905), Krzemien-
iewsk (1908), and especially Prazmowski (1912, p. 162) have studied them fairly thoroughly.
No. 2.], LIFE CYCLES OF THE BACTERIA—LOHNIS. 143

_ Vibrio proteus has been seen by Firtsch (1888) to coil up occasionally, as it is rather frequent
with spirochaets, according to the observations made by Breinl and Kinghorn (1906), Prowazek
(19060, 1907a), Breinl (1907), Schaudinn (1907), Gonder (1909), Karwacki (1911), Wolbach
and Binger (1914), Inada et al. (1916), and others. Unquestionably not all of these coiled
forms develop actually to regular microcysts. The drawings made by Prowazek (1907a,
original fig. 8b), which are reproduced as figure 74 on Plate P may illustrate this possibility,
while some other sketches, published by Inada and his collaborators (1916, original figs. 71
and 72) and reproduced as figure 75 on Plate P, probably are to be accepted as showing the
formation of genuine microcysts.: |

Germination of microcysts has been recorded with those produced by cocci and by Azo-
tobacter. Van Tieghem (1879c, 1884) saw that the ‘“‘kystes” of Leuconostoc when still
inclosed in the chain germinated laterally and the membranes of the microcysts were left behind
by the young cocci. The analogous behavior of the heterocysts of Nostoo and of other Cyano-
phyceae has been described by Brand (1901). Germination of the microcysts of Micrococcus
ochroleucus has been observed by Prove (1887). H. Fischer (1905) reported that the germi-
nating Azotobaoter left no empty membrane behind, but Krzemieniewski (1908), Prazmowski
(1912), and D. H. Jones (1913) recorded opposite results.

In thesame manner as some of the vegetative cells increase in size and become gonidangia,
so also microcysts may occasionally undergo a similar change, and they, too, may either liberate
a number of gonidia, or these may reproduce new vegetative cells while still inclosed within
the microcyst. The liberation of the gonidia seems to be the rule with the microcysts of spiro-
chaets (Breinl, 1907); direct germination apparently never has been observed in this case.
The microcysts of streptococci not infrequently break up into 3 or 4 parts, each of them acting
as a new vegetative cell. Babes (1908) probably was the first to give an account of this inter-
esting ocourrence. If such a cyst remains within the chain, it may act as a very conspicuous
starting point for branches, as was noticed by Vincent (1902). The early report of E. Klein
(1883) upon what has been at least in part microcysts of B. anthracis also refers to their divid-
ing into 2, 3, or 4 parts, each of which proved to be able either to multiply as a round budding
vegetative cell, or to grow up again to a normal rod. The drawings made by Matzuschita
(1902) from his protean bacillus from air, reproduced as figure 5 on Plate B, shows also some large
round cells, looking like microcysts, which clearly exhibit the same mode of fission. More
recently Hort (1917b) noticed that the “giant cells” of Meningococcus may either become
gonidangia or divide themselves into 3 or 4 equal parts.

Concerning the resistance of the microcysts very little is known at present. Similar to
the arthrospores, they are evidently more apt to withstand the influence of drying than that
of heating. However, in a few cases Prove (1887) was able to record a surprisingly high resist-
ance; some of the microcysts tested by him are said to have endured 100° C. for more than
one-half hour. Baumgarten (1890, p. 295) mentioned that old cultures of staphylococci were
not always killed, when kept at 99° C. for 15 minutes, and A. C. Abbott (1912) succeeded in
selecting especially vigorous strains of Staphylococcus aureus by keeping the original culture
for 5-20 minutes in fairly strong antiseptic solutions (0.1 per cent HgCl,, 0.75 per cent phenol
or saturated Na,CO,). Holman (1914) noticed that in one case streptococci were not even
killed when kept three times for 20 minutes:in the Arnold apparatus, a result which evidently
must be attributed to the presence of protecting albuminous substances in the substrate.
Special experiments upon the heat resistance of the microcysts of Azotobacter, Bac. anthracis
and of related species apparently have not been made so far. A rather considerable resistance
against drying, however, has been repeatedly observed with the first-named organism.

() GONIDIA AND FILTERABLE VIRA (CELL INCLUSIONS, APHANOZOA, ETC)—GONIDIA AND
* HETEROGENESIS.

The size of the bacterial gonidia shows wide variations, especially when they have grown
within a gonidangium. With the smaller species the diameter is usually 0.1-0.3 ». Therefore,
a.more or less considerable number of gonidia is able to pass Chamberland, Berkefeldt, and
similar bacteria filters, and when transferred to a suitable medium, such a filtrate, though freed

160266°—21——10
144 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor XVI.

from all bacteria, may still cause an effect on account of the gonidia present\ therein. This
possibility makes the filterable gonidia an important subject in connection with the study) of
filterable vira, certain cell inclusions, so-called aphanozoa, chlamydozoa, etc. Some of these
may, indeed, belong to the protozoa; but there is no reason why the same assumption should
be accepted as the only working hypothesis in all cases. In fact, several observations are
already available, which make it certain that sometimes bacterial gonidia, too, may play an
important part, and it is very probable that they will win in epee as soon as more
attention is centered upon this point.

General reviews of the whole subject have been published by Prowazek (1907), Lapel
(1909), Loeffler (1911), Doerr (1911), Wolbach (1912), Fontanel (1913), and others. According

to the first-named author the small bodies concerned are less than 0.25 » in diameter, of globular
shape, occurring singly, in pairs, in chains or small clumps, difficult to stain, usually best with
Giemsa’s method, and fairly resistant against physical and chemical effects. In addition
Fontanel has pointed out that they are probably all motile. All these characters would fit the
filterable bacterial gonidia: size, shape, motility, staining qualities, as well as increased resist-
ance, are the same. That especially glycerin is a good preservative for gonidia, as it is for
filterable vira, has been already ascertained by Neelsen in 1880, and more recently by Mathers
(1917). With regard to alcohol analogous results have been obtained by Jwanowski (1899).

The development within the cells of the host as “cell inclusions,’’ ‘‘ Kern-Kappen,” etc., has
been often assumed as indicating the protozoal nature of the intruders. But already in 1889
Mac Fadyean reported upon the interesting fact that the small ‘‘cocci” (gonidia) produced by
Actinomyces invade the animal cell, sometimes even the nucleus, and develop therein. The
gonidia of spirochaets were seen by Balfour (1911 5), Ross (1912), and others, to behave in the
same manner. And Duischenko (1914) was able to ascertain that smallest inclusions, found by
him within the red blood cells of rodents, which they destroy, and which therefore at first were
considered to be some protozoa related to Theileria parva, developed, after pene liberated, in
the blood to bacteria, similar to those of the plague.

That gonidia, and sometimes regenerative bodies, too, may occur within the mast cells,
has been mentioned on page 111. As Herzog (1913) has pointed: out, it still remains to be studied,
whether the ‘eosinophilic’ granules, which are present in blood cells, are actually cell products
or bacterial gonidia. That the latter are much inclined to grow as such, and to replace the
bacteria within the body, has been pointed out by M. E. Abbott (1900) and others. The fact,
that thus far only in a few cases development of normal bacteria cells has been obtained from
these filterable coccoid bodies, does not exclude the possibility that improved methods. will
secure positive results also in other cases. As far as filterable vira have been cultivated as such,
their behavior did not differ from that shown by cultures of filterable gonidia. 5

The organisms of vaccine and variola, classed by F. Cohn (1872 a and b) somewhat pre-
maturely as Micrococcus vaccinae, were declared by Calkins (1904) to belong to the Micro-
Sporidia; but all his findings can be just as well accepted as descriptions of bacterial gonidia, re-
generative bodies, and symplasm. Distinctly more in favor of the protozoa hypothesis are the
results obtained by Howard and Perkins (1904), while the various communications made by
Prowazek (1905, 1906 a, 1907 ¢c) upon “‘Initialkérper’’ and Chlamydozoen”’ leave. the matter
again entirely in doubt. The findings of Bonhoff (1905) upon what he called Spirochaeta vac-
cinae, seem never to have been followed up; they were very similar to results obtained in vari-
ous spirochaetoses. That the causative agent is filterable has been shown by Stegel (1905 a)
and by A. Negri (1906). The cultures made by Siegel (1911) of his Cytorrhyctes vaccinae would
suggest relations to some Micrococcus. However, no conclusive results have been secured.

The various coccoid bodies found by Nencki, Sieber and Wijnikewitsch (1898) to be connected
with ‘‘Rinderpest”’ also failed to exhibit any sign, which would be strictly against their bacterial
nature.

The small motile globules first grown by Nocard and Roux (1898) 1rom pleuropneumonia
of cattle, seem to have been the gonidia of a pleomorphous bacterium, described by Bordet
(1910), by Borrel, Dujardin-Beaumetz, Jeantet, and Jouan (1910) under the name Asterococcus
No. 2] LIFE CYCLES OF THE BACTERIA—LOHNIS. 145

mycoides, and by Martzinowski (1911) as Coccobacillus mycoides peripneumoniae. Confirmative
reports are still outstanding; and Kallert (1911) is of the opinion that colloidal substances from
the serum have been mistaken for organisms.

After Dorset, Bolton, and McBride (1905) had ascertained that hog cholera is caused by
a filterablev irus, Lourens (1907) showed that Bact. cholerae suum is able to produce filterable
“‘oranules,’’ which were found to be able to reproduce the bacilli in the animal, but not in the
cultures. They were, therefore, declared to be the primary, the bacilli the secondary cause of the
disease. Hitibner (1908) contested these observations strictly, as did Uhlenhuth, Xylander,
Htibner and Bohtz (1908-1909), who confirmed and extended the findings of the American
authors, not only with regard to hog cholera, but also concerning swine plague. The multipli-
cation of the filterable virus as such was especially studied by Pfeidler and Lentz (1913). Riither
(1910), on the other hand, is of the opinion that the granules present in the infectious filtrate are
produced by spirochaets, which were seen to be present in blood, urine, and in the organs of
diseased, but not in healthy hogs. After 10 days they also reappeared in the filtered virus, and
their reproduction from the filterable granules is reported to have been observed on solid sub-
strates, too. The granules were found in great number in lice and intestinal parasites (nem-
atodes and others), to which the author ascribes, therefore, an important réle in transmitting
the disease. Sometimes the spirochaets were seen to break up into small rods, looking like B.
septicaemiae or influenzae. Very similar results have been obtained by King in cooperation with
Baeslack and Hoffmann (1913) with a Spirochaeta suis, which again was found only in diseased
hogs and was seen to break up into filterable infective granules. Healy and Gott (1916) were
able to grow the filterable forms as such in 1 per cent glucose broth, to which ground mesenteric
glands of hogs had been added, and Proescher and Sei! (1917) described anew the small coccoid
bodies present in endothelium cells, blood, and urine.

Trachoma, as well as blennorrhoea non gonorrhoica, are caused, according to Haiber-
stédter and Prowazek (1909), by the Chlamydozoa discovered by them. As Heymann (1909)
has pointed out, however, only the ‘‘Kern-Kappen”’ are to be considered of diagnostic value,
while single and double granules were declared by him to-be present in various affections of
the conjunctiva. On the other hand, extended studies of this subject led Herzog (1910) to the
conclusion, that the so-called trachoma bodies should not be connected with protozoa, but
are formed by ‘‘micro-gonococci,’’ which are produced by ‘‘macro-gonococci.’’ Béing (1912)
was not willing to accept any of these findings as being correct, and also Noguchi and Cohen
(1913) at first were only able to isolate and cultivate anaerobically small coccoid bodies, whose
cultures showed all stages of the so-called inclusions, but did not give definite results in inocu-
lationexperiments. Herzog (1913), however, furnished further proof concerning the réle played
by the minute ‘‘involution forms” (gonidia) of the gonococci; an interesting picture, pre-
senting these smaller and larger, globular and dumb-bell-shaped bodies, has been reproduced
as figure 202 on Plate XVI (from original fig. 5 on Pl. IIT). It should be compared with figure
2 on Plate A. The matter was brought beyond doubt by the extensive investigations made
by Williams, Wilson and Gurley (1914). Trachoma inclusions are, according to these authors,
“simply intracellular nests of growing bacteria’’; besides the gonococcus a hemoglobinophilic
bacillus was found to be able to produce in cultures, as well as in the eye, all forms described
by Prowazek, especially ‘‘dense clumps of extremely minute and irregular coccoid forms.”
A second contribution by Noguchi and Cohen (1915 added some interesting data, suggesting
analogous relations existing between the Koch-Weeks bacillus and cell inclusion conjunctivitis.

With regard to lyssa Babes (1907c) has pointed out that while the so-called Negri bodies
are not always visible and probably represent merely cell products, very small granules, which
stain like bacteria, are constantly present. A. Negri (1909), however, maintained, that the
cell inclusions be parts of the life cycle of a protozoon, Neurorrhyctes hydrophobiae. The
standpoint taken by J. Koch and Rissling (1910) is more that of the first-named author, and
S. R. Klein (1911) reached the conclusion that rabies is to be considered as a very acute
streptococcal infection, and the Negri bodies as remnants of these streptococci. Staining

x
146 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

experiments misled Tanakamuru (1913) to assume, the granules present in rabietic brain to
be merely lipochrom and degeneration products of the gangliar cells. Noguchi (1913), as
well as Williams (1913), were able to cultivate the virus as such. They both think that the
bodies grown looked like protozoa, not like bacteria; but the photographs made would
equally fit the appearance of bacterial gonidia and regenerative bodies.

From scarlatina filterable, motile granules have been obtained by Siegel (1905 b), who
classes them as Cytorrhyctes scarlatinae among the protozoa, while Bernhardt (1911) has pointed
out, that the cell inclusions visible are similar to those seen in trachoma, and this holds true
also with regard to the fine filterable granules, which are often dumb-bell-shaped. That they
may be the gonidia of the streptococci, which are always present as ‘‘secondary invasion”
in scarlatina as in measles, has become very probable by Hort’s studies (1915-1917).

Poliomyelitis is another case where it seems to be very probable that the filterable small
bodies first obtained by Flexner and Noguchi (1913) are gonidia of streptococci. After they
have become adapted to artificial substrates, they behave very much like small streptococci,
according to the data furnished by Flexner, Noguchi and Amoss (1915). That, however,
the small forms are, in fact, parts of the life cycle of a polymorphous Streptococcus, has been
first proclaimed by Rovencit, Towne and Wheeler (1916), and it was soon afterwards con-
firmed by Nuzum and Herzog (1916). These, as well as the further contributions by Rosenow
and Towne (1917), Mathers: (1917), Mathers and Howell (1917), Rosenow and Wheeler (1918),
Rosenow, Towne and Hess (1918), leave hardly any doubt that it is also in this case the filter-
able gonidia, which act as the specific virus. -Amoss (1917), as well as Bull (1917), have con=
tested Rosenow’s and Nuzum’s results, but their own experiments can not be accepted as con-
vincing negative proof, especially when the peculiar staining reactions, as well as the charac-
teristic cultural behavior, of the filterable gonidia are duly considered. Rosenow and Towne
are undoubtedly right when they point out that the whole subject concerning the filterable
vira needs a comprehensive study from these new standpoints.

The aphthe virus has been classed by Siegel (1905 a) as a Cytorrhyctes aphtharum in the
neighborhood of Cytorrhyctes vaccinae, but later experiments (1910) led him to the belief that
this Cytorrhyctes is a type of growth of a micrococcus. Further investigations (Siegel, 1912)
strenghtened this view; his photographs are exactly of the same type as those made by
Rosenow, Nuzwm and by Mathers of the poliomyelitis germ. Pictures published by Betegh
(1911) of what he considers to be the causative agent of foot-and-mouth disease, are very
similar. Stegel’s as well as Betegh’s claims have been refuted by-Wehrle and Zwick (1913)
and by Kallert (1913). A reply was published by Siegel (1913). The problem needs further
unbiased study, as was emphasized by F. Winkler (1906).

As causative agent of syphilis a Cytorrhyctes luis has been introduced by Siegel (1905 ce),
which was accepted by Leuriaux and Geets (1906) as the granular form of Spirochaeta pallida.
The important réle played by the gonidia in other spirochaetoses has been repeatedly empha-
sized. Riither’s and King’s reports upon the granules of Spirochaeta suis were quoted above,
those made by Breini (1907) upon the filterable gonidia of Spirochaeta Duttoni, by Balfour
(1911 @ and 6) upon Spirochaeta granulosa, and by others have been discussed on pages 105-106.
Wolbach and Binger (1914) pointed out correctly that sometimes errors may be caused by
spirochaets passing the filter, just as is done occasionally by other bacteria; but the first-
named author’s statement (Wolbach, 1915), ‘‘that there is no evidence of epirochaeta multi-
plying by any other method than single fission,” is not in accordance with certain observa-
tions, and that in his experiments the granules did not multiply as such, does not invalidate
the fact, that under suitable conditions all bacterial gonidia act in this manner. Balfour's
discovery concerning the higher resistance of the ‘‘infective granules” against salvarsan and
the importance of this finding in regard to chronic spirochaetosis have been mentioned on

age 131.
ve The widely varying observations upon typhus exanthematicus may be accepted as another
instance where the study of the presence and action of bacterial gonidia will help solve the
problem. The diplococcus, described by Rabinowitsch (1909), as well as the pleomorpho
No. 2:] LIFE CYCLES OF THE BACTERIA—LOHNIS. 147

organism isolated by P. Th. Miller (1913), whose pictures have been reproduced as figures
22-24 on Plate II, or the equally pleomorphic anaerobic bacillus discovered by Plotz, Olitsky
and Baehr (1915), may all act either as such or by means of their gonidia. Hort’s (1915-1916 a)
experiments have already brought much light in this direction.

With meningitis the situation is quite similar. Here again, according to Hort (1916 a), the
micrococci themselves are important as a “danger signal,” that the true infective agent, their
filterable gonidia, may be present. The fundamental importance of these meningitis investiga-
tions, made by Hort and his collaborators (1915-17), has been justly emphasized by Adami
(1916), who also pointed out that besides to meningitis, typhus fever, scarlatina, and hog cholera
the same principle may apply to tuberculosis. That, indeed, filterable gonidia are produced
by B. tuberculosis, too, had been demonstrated already by Fontes (1910).

That in common colds once more filterable gonidia are of etiologic significance, is practically
beyond doubt. The anaerobic organism isolated by Tunnicliff (1913-1915) from the nose, as well
as Micr. catarrhalis, and perhaps other bacteria, too, again seem to be less important directly
than indirectly by their producing infective, partially filterable gonidia. What Kruse (1914)
named Aphanozoum coryzae, the virulent filtrate causing the cold, has been cultivated as such
by Foster, jr. (1917). Morphology and staining reactions of the smaller and larger budding
bodies, which the author compares with Fleaner’s globoid poliomyelitis germs, are exactly those
of gonidia and of regenerative bodies.

That the virulence of the gonidia may be more or less different from that shown by the vege-
tative cells, from which they originate, isnot surprising. Of equal interest, however, is the possi-
bility, that distinctly antagonistic activities may be displayed by the parent cell and its repro-
ductive organs. Almquist (1911) obtained, when studying the growth of filtered gonidia of B.
typhosus, occasionally instead of the usual, very scant development, a thick yellowish layer
on lactose agar, formed by small immotile oval forms, which were not pathogenic, but whose sera
agglutinated the typhoid bacilli. That these peculiar organisms have been no contamination
has become certain by further investigations made by Almquist (1917), which in no case gave
an upgrowth of normal typhoid bacilli from the filtrate, while in 10 per cent of the tests the same
“ Bact. antityphosum”’ was obtained, which remained stable for 7 years, and which on account of
its behavior in the agglutination and in the Pfeiffer test is believed to be a “mutation” of B.
typhosus. Herelle (1917) also reports to have isolated from feces and urine of dysentery con-
valescents an “invisible” microbe of antagonistic character, which did not grow on any sub-
strate, except in the presence of dysenteria bacilli killed and solved by heating. Similar results,
though not quite so clear, were obtained by the same author in studies upon paratyphoid.

The possibility that all bacteria, perhaps with the only exception of the large trichobacteria,
may produce filterable gonidia, makes the conclusion inevitable, that they must be present
quite generally in nature, as well as in the laboratory. In fact, the omnipresence of “ultrami-
croscopic” germs of bacteria has been already considered by Burdon-Sanderson in 1871, and more
recently by Gaidukov (1906); but this hypothesis has been refuted by Molisch (1908) and by Cano
(1909). The former author points out that neither he nor anybody else had ever obtained
cultures of “‘ultra-microérganisms,’”’ and the latter recorded nothing but negative results
when he tested the filtrates of many different substances under the microscope, in cultures, or in
the animal test. These negative findings, however, are by no means decisive. As is the case
with the filterable vira, also the other, not pathogenic filterable gonidia produce in or on the
substrates used in the laboratories—provided that they grow at all—such a very scant growth
that it can be as easily overlooked as their presence under the microscope. The extremely thin
and restricted dewy layer on solid media, or the very slight cloudiness in liquid substrates, seems
to be all what can be expected. But as at least in some cases a renewed development of larger
forms has been secured, it is to be expected that more positive results will be obtained, when the
conditions allowing such upgrowth will be better known, and especially if the important réle
played also in this case by the symplastic stage, will receive adequate consideration. That posi-
tive results have been obtained more frequently within the animal than in cultures agrees well
148 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor XVI.

with the fact that genuine plasma, not changed by heating, evidently greatly favors such
upgrowth as is clearly indicated by the observations to be presently discussed.

The knowledge of the participation of the gonidia in the life history of the bacteria throws a
very interesting light upon some facts, which have been accepted so far by several authors as strin-
gent proof of the possibility of heterogenesis. It has been mentioned already (on p. 14) that some
years ago Dunbar (1907), whose experience in doing bacteriological work can not be questioned,
published a book wherein he gave many facts apparently proving that all kinds of bacteria
may be produced by the same species of algae, and that this book was accepted by authors like
Pringsheim as an attempt to undermine the whole science of bacteriology. If the general
knowledge of the literature would be better as it usually is, it had at once become evident, that
Dunbar’s results are practically identical with those recorded by many earlier authors, who all
found, that within dying or dead cells of algae, fungi or other organisms, bacteria may develop
from smallest granules, which microscopically can not be distinguished from the cell granules
themselves.

Dujardin (1841, p. 93) mentions that it has been the opinion of Miller, Dumas, Gleichen,
and other authors, at the end of the eighteenth and at the beginning of the nineteenth century,
that the organic residues of plant and animal life dissolve themselves into “globules élémentaires,”’
which by transformation may become “‘infusoires.”” According to Fokker (1887 a, p. 3), Buffon
and Needham assumed that the living molecules after the death of plants and animals may
continue their existence as such and eventually build up new microérganisms. Perty (1852),
Pouchet (1863), and Trécul (1865-1867) stood practically on the same heterogenetic standpoint,
and the victory attained by Pasteur in regard to spontaneous generation had in fact not much
influence in this direction. That life can not begin anew was soon generally admitted, but the
theory of heterogenesis still found many defenders, and as long as the existence of the bacte-
rial gonidia is not taken into account is, indeed, invincible.

The detailed description given by Trécul (1865-1867) of the upgrowth of Amylobacter within
closed dead cells from very small granules, ‘‘les derniers molécules vivantes de protoplasma,”
fits remarkably well to the results recorded by Dunbar. The same holds true concerning the
findings made by Karsten (1869) together with Harz, who by continuous direct microscopic

~ observations ascertained that within cells of yeasts and of other fungi, but only when their vital
activity had come to a standstill, “‘micro-gonidia”’ may develop, which later either internally
or externally may grow up to “ Vibrionen’”’:

Sehr iiberzeugend sieht man diese Entwicklungserscheinungen der endogenen Zellchen zuweilen an einzelnen
Gliedzellen von Pilzmycelien, die zwischen andern, welche gesund verbleiben und sich normal entwickeln,
erkranken (p. 29).

An interesting drawing made by Karsten (1869) is reproduced as figure 76 on Plate Q (from
original fig. V). The development of small rod-like bacteria within a Rhizopus sporangium,
and the upgrowth of various forms from the minute “micro-gonidia”’ is fairly well discernible.

The otherwise rather faulty observations made by Hallier (1866-1896) and by Liiders
(1866) have been correct in so far as these authors, too, have watched directly the transformation
of smallest granules within dead cells of fungi into bacteria, but it was, indeed, an astonishing
logical mistake, when Hallier still in 1895 (p. V) tried to defend the following absurd conclusion:

Die winzigen Zelibildungen, welche Ndgeli unter dem Namen der Spaltpilze zu einer Familie zusammengestellt
hat, sind keine vollsténdigen Gebilde, sondern Erzeugnisse des Plasma verschiedener Pilzgruppen. Die Néigelische
Familie der Spaltpilze ist also aus dem System zu streichen.

The voluminous contributions made by Bastian (1872-1914) to the theory of heterogenesis
become also quite interesting, when read with some knowledge of the existence and the beha-
vior of the bacterial gonidia and, in addition, of the symplastic stage of bacterial life, as
will be discussed in Chapter III. The “newly evolved specks of living matter’’ now lose at
once their mysterious character.

The “‘microzymas’’ of Béchamp (1883) present a similar case. His drawing reproduced as
figure 77 on Plate Q (from original fig. 4 on Pl. I) show formation and liberation of gonidia by
No. 2.) LIFE CYCLES OF THE BACTERIA—LOHNIS. 149

long rods about as clearly, as it was pictured, for instance, by Billroth in 1874 (see fig. 45 on Pl. L)
or 40 years later by :Meirowsky. (See Pl. XIV). It will be justly doubted, however, that
Béchamp.was right when he stated (p. 628):
Les seuls éléments anatomiques non transitoires de l’organisme qui persistent aprés la mort et qui évoluent pour
former des bactéries, sont les microzymas.
But in view of the observations mentioned above, it is already more easily to be understood,
when he says (p. 148):

- que-les microzymas sont de ceux qui produisent aisément des bactéries.

And when he points out (p. 473) that the microzyma is not by itself a bacterium, but ‘s
related to it in the same manner as a spore to a mold, and (on p. 839) that the bacterium is able
to produce again ‘‘microzymas”’ he evidently comes very close to the truth. That bacteria
easily develop from the “microzymas” in dead plant cells, blood corpuscles or fibrine, while
other substances, for instance the white of the egg, did not give such a “transformation,” is
also in full agreement with the findings of other investigators.

Engler (1882) reported, that he not infrequently saw small bacteria develop within dying
cells of a large Beggiatoa, though he did not assert, of course, as did Wigand (1884), that they
were formed there by ‘‘anamorphosis”’ of the cell plasma. The latter author thought that he
saw all kinds of bacteria grow up from the plasmatic granules within intact plant cells. but ap-
parently no cultural experiments have been made, and it remains, therefore, doubtful, whether
these bodies have been real bacteria or only cell inclusions of a similar shape, especially those
called chondriosomes.

The numerous experiments made by Fokker (1887-1903) to support his ‘‘neue Bakterien-
lehre,’’ which was, however, in fact the old heterogenetic hypothesis, are in so far of greater
interest, as the author, late Professor of Hygiene at the University of Groningen, undoubtedly
has been well acquainted with bacteriological methods of investigation, and he was able to report
that he not only had under his microscope things looking like bacteria, but that he actually cul-
tivated bacteria under conditions which would invalidate the easy explanation of ‘‘ mere contami-
nation.’ He confirmed the earlier observations of Klebs (1873), Tiegel (1874), Zahn (1884) and of
others, that blood and parts of muscles aseptically won and kept, did not give bacterial growth,
but that this became visible within parts of the liver, spleen, and kidneys. Blood alone or
diluted with water remained equally sterile, but “‘hematocyts”’ appeared and developed to nor-
mal bacteria, when small amounts of blood were kept in nutrient solution (slightly acid broth
of ‘beef extract and lactose), either at 20°, or 37°, or at 50-52° C. The highest temperature
proved to be most suitable. The appearance of these hematocyts, as shown in figure 203 on
Plate XVI (reproduced from original fig. I, Pl. I) leaves hardly any doubt that they are
regenerative bodies growing up from the gonidia contained within the pale blood corpuscles.
That they are first Gram-negative, later gram-positive, that iodine gives them a brown stain,
that they are resistant against alkali and acetic acid, and that they apparently multiply by
budding, is also in full agreement with this assumption. Further experiments proved that
bacteria developed from these round bodies (see fig. 204 on Plate XVI, reproduced from
original fig. II on Pl. III); at room temperature a ‘‘comma bacillus,’”’ at 40° C. yellowish, gel-
atin liquefying bacteria, and at 52° C. another organism which showed budding and branching,
probably an Actinomyces, were obtained. With the ‘‘comma bacillus” reproduction by small
granules, produced inside of the parent cell, was also observed. okker’s discovery that in cases
of anthrax, especially during the first 12-17 hours, no bacilli are visible, but only granules,
which stain like the nuclei of the spleen cells, and like the anthrax rods which develop from
them, is of interest in this connection, though it does not, of course, support the Dutch
author’s assumption that it is the granules of the sick spleen cells which become anthrax
bacilli. Some others of his findings, especially with regard to ‘‘heterogenesis” in milk, will
be considered in Chapter III, as the development of regenerative bodies from the symplasm of
milk bacteria has obviously caused this faulty hypothesis.
150 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

The cell granula theory of Altmann (1894) is similar to the earlier ones in so far.as he, too,
accepts cell granules and microérganisms as equivalent (p. 141), but-his standpoint is. different
when he says (p. 146): et

Die Zellengranula lassen sich nicht ziichten, sie sterben mit der Zelle ab.

On the other hand, Hueppe (1896, p. 31) was not quite right when he tried to discard all
positive results by the ‘following statement:
Die Angaben iiber Bildung von Bakterien aus anderen Zellen und deren Bestandteilen durch Anamorphose des

Protoplasma beruhen auf Verwechslungen von Zellkérnern in Milch, Blut, Geweben und von Fibrinausscheidungen
und ktnstlich verinderten Kernbestandteilen mit Bakterien.

Miinden (1896-1907) also secured some positive results concerning the development of
bacteria from small granules present within plant or animal cells, but his fantastic “cytoblast”’
and “chtonoblast”’ hypotheses, founded upon this basis, are certainly of no value whatever.
The same holds true concerning the theories promulgated by J. H. Miiller (1898),. who once
more had to record negative results, when he tried to “develop” -acteria from living spores
of fungi, while dying or dead material reacted differently.

That it is the gonidia of the bacteria which grow up within the cell, making use of the
albuminous substances of the dead protoplasm, has been pointed out for the first time, and with
full certainty by W. Winkler (1899). Clean pieces of Hymenomycetes kept in steril wort gave
him many “bacterioblasts,” which developed new bacteria, just as had been described by
Karsten thirty years earlier. That, as P. Lrnst (1902) reported, bacteria often grow in a
vertical position and in thick bunches on and around a mycelium, and that Gaidukow (1906)
found his ultra-microérganisms frequently within closed cells of algae and fungi, is again in
complete agreement with the earlier observations.

And it is now easily understood why in strictly pure, single-cell éaliaiies, of an cis
Dunbar (1907) has seen development of different kinds of bacteria in a great number
of cases, though extensive controls (over 4,000) proved that these could not be simply explained
as contaminations by bacteria. It was again only the dead, not the living, alga cells which
gave this growth. Whether the bacterial gonidia had been introduced with the alga, or whether
they were, indeed, contaminations from the air, can not bedecided. The special tests made
with the isolated algae cells on gelatin and on agar are inconclusive, because under such con-
ditions bacterial gonidia do-not grow at all, or their growth remains so scant that only special
attention in this direction will detect it.

M. E. Abbott’s (1900) remark that the small coccoid bodies to be found in liver and other
cells may belong to B. coli and other bacilli seems to be, indeed, the correct explanation of
apparent heterogenesis found at such places; and it may also be considered to be beyond doubt
that bacterial gonidia have played their part in the development of bacilli within animal cells,
as observed by Saski (1907) and by Portier (1917).

The behavior of Anaplasma marginale, according to Theiler’s (1910) paers a protozoon
causing ‘‘gall sickness,’ seems to be, as far as can be surmised from his and Sieber’s (1910)
description, strikingly similar to that of Fokker’s “hematocyts.” The microscopic picture
of these small chromatine granules, growing at the edge of the blood corpuscles and multiplying
as such, is exactly like that furnished very frequently by bacterial gonidia. Comparatively
large round bodies “‘studded with chromatine granules” were seen by Ross (1912) to occur in
cases of spirochaetosis, and they are also by no means'rare with B. radicicola and Azotobacter.
Probably they will be found with all bacteria as soon as this subject will be studied more
thoroughly.

When the gonidia live within the dead cell, making use of its protoplasm, they naturally
become what Prowazek called Chlamydozoa, and if the alga used by Dunbar in his experiments
was acid-fast, it is no wonder that all bacteria which developed in his cultures proved to be
at first also acid-resistant.

It remains to be hoped that some staining reaction will be found later which would allow
a differentiation between bacterial gonidia and cell granules, but in view of the ever-changing
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 151

composition of both kinds of these smallest units not very great expectations seem to be war-
ranted. Therefore, the heterogenetic hypothesis can not be defeated definitely from this new
point of vantage; that it has lost, however, practically all of its probability, is hardly to be
disputed. Besides this, the results discussed above are of great practical interest, insofar as
valuable suggestions offer themselves for a successful attack of the otherwise extremely difficult
problem to secure a speedy upgrowth of normal bacterial cells from their gonidia, which on the
commonly used substrates is often out of the question.

2. REPRODUCTIVE ORGANS IN THE DIFFERENT GROUPS OF BACTERIA.

The summary of what is known at present concerning the occurrence of the different types
of reproductive organs among the various kinds of bacteria, which is to be given on the follow-
ing pages for ready reference, will be arranged, on account af the reasons mentioned on p. 6,
in the same manner as the review given in the second part of Chapter I along the old familiar
lines of grouping. It is, of course, somewhat absurd to write, e. g., upon the spore formation of
nonspore-forming bacteria, but we will have to tolerate temporarily such, it is to be hoped, not
misleading ‘absurdities,’ because a new, more correct classification of the bacteria can not
be attempted successfully, before many more results will have been gathered, than are available

at the present time.
(a) COCCI.

The formation of minute and of comparatively large forms of micrococci, probably their
gonidia and microcysts, were apparently first noticed by Liibbert (1886) in the course of his
studies upon Staphylococcus pyogenes aureus. Prove (1887) recorded with his Micrococcus
ochroleucus normal cells of 0.5-0.84, minute coccoid bodies of 0.1-0.3un, and large forms
of 1.6-1.84 diameters; the latter were called ‘“‘Dauersporen”’ on account of their con-
siderable resistance against heat (80 minutes at 100° C.), but they have been evidently
microcysts. Prazmowski (1888 a) found resistant forms regularly with Micr. ureae, which
survived when exposed for 1 minute to 90° C., but were killed at 100° C. The whole cell was
transformed, and no membrane was thrown off during germination. Nevertheless the author
insisted that these were ‘“endogenous”’ spores. Like this, so also the statement made by
DeToni and Trevisan (1889, p. 1072) can not be accepted as referring to genuine endospores,
although it reads: endosporae microsomae in coccis normalibus obvenientes. ;

Irregular rod-like regenerative bodies were observed by Matzuschita (1900) when he grew
Micr. flavus liquefaciens and M. rubefaciens on salt agar (see figs. 5 and 6 on Plate I), while other
micrococci, including VW. candicans, did notreactin thesame manner. The last-named organism,
however, gave analogous results when studied by Léhnis and Smith (1916 a and 6); besides the
formation of regenerative bodies the production of gonidia and of microcysts has been dis-
cussed in these papers. Some photographs were reproduced as figures 2-4 and 8.on Plate I
and 168-169 on Plate XII. The “protospores”’ found by Fedorowitsch (1902) with micrococci
as with other bacteria leave no doubt about their being gonidia. Such “exogenous globules”’
were also seen by Almquist (1917) with Micr. pyogenes, while his “macrococci”’ of Micr. Thulini
furnish another example ot microcyst formation.

The important réle played by the gonidia in the life cycle of the Gonococcus was discovered
by Wertheim (1899) and more thoroughly studied by Herzog (1910-1913), who also secured
analogous resu'ts with Meningococcus. The latter have been confirmed and considerably
extended by Hort and his collaborators (1915-1917), whose observations relating to the develop-
ment of gonidangia, which were mistaken for “‘asci,”’ are of special importance.

Some data concerning the formation of gonidia, regenerative bodies, and of microcysts by
Sarcina flava and Planosarcina ureae have been secured by Léhnis and Smith (1916 a and b).
The production of regular endospores by the last-named organism, first described by Betjerinck
(1901 a), together with other morphological data exhibited in the course of its life cycle, points
to certain relations with spore-forming bacilli, a fact which was already indicated in an earlier
note by Olsen (1897). :
152 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

That the “Dauersporen’”’ seen by Salomonsen (1876) in blood cultures of streptococci,
have been microcysts, is hardly to be doubted, and this is beyond question in regard to the
“spores” or “kystes” of Leuconostoc mesenterioides, according to the description given by Van
‘Tieghem (1879 ¢ and 1884, p. 1108). The same holds true concerning the “arthrosporae macro-
somae in filamentis huc illuc sparsae,’’ which were mentioned by De Toni and Trevisan (1889,
p. 1054) as characteristic of the genus Streptococcus. The drawings of streptococci made by
Babes (1895), which were reproduced as figure 3 on Plate B, those of Leuconostoc hominis of
Hlava (1902), reproduced as figure 4 on Plate B, as well as that of Streptococcus pyogenes, made
by Hewlett (1902) and reproduced as figure 72 on Plate P, are of great interest on account of
the details contained therein in regard to gonidia, gonidangia, regenerative bodies, and micro-
cyst formation. All these various types of reproductive organs were also found by Thiercelin
(1899-1903), when he studied his Enterococcus. His observations upon its gonidia (‘‘micro-
blastes’’) and gonidangia deserve our special attention, and have been quoted, therefore, more
fully on pages 104 and 126. Formation and behavior of these “‘microblasts” have been thor-
oughly investigated by Thiercelin and Jouhaud (1903 a and 6). The peculiar splitting up of
microcysts into 3 or 4 parts was first described by Babes (1908), and some data upon their
resistance against heating (live steam for more than 20 minutes) were obtained by Holman
(1914). The avirulent, resistant, Micrococcus-like subculture, grown by Eyre and Washburn
(1897) from typical pneumococci, has been evidently a pure culture of regenerative bodies,
such as have been photographed by Azelrad (1903) from a contact preparate of Streptococcus
lanceolaius, reproduced as figure 146 on Plate XI.

The important rdéle played by filterable gonidia of streptococci as filterable virus in diseases
like poliomyelitis, was discussed on the foregoing pages (pp. 145-146).

A good description of gonidangia produced by Streptococcus lactis has been furnished by.
Maddox (1885). The spore-like terminal bodies seen by Weigmann (1899) at rod-like forms of
this species, are probably to be classed as regenerative bodies. Our own experiments with
lactic acid streptococci proved once more the regular occurrence of gonidia, regenerative bodies
and microcysts. Photographs have been reproduced as figures 11 on Plate I and 170 on Plate
ate (6) NONSPORE-FORMING RODS.

The report and pictures given by Hauser (1885) in his Proteus monograph, of what he
calls the involution forms of this organism, are mostly to be interpreted as relating to the forma-
tion of gonidangia and to the occasional transformation of their content into one rudimentary
endospore. A characteristic photograph has been reproduced as figure 120 on Plate X. Bac.
piscicidus agilis was found by N. Sieber (1895) to produce genuine endospores, though it exhib-
ited in all other respects the character of the Proteus group. With B. Zopfii, whose arthro-
spore formation has been studied by Kurth (1883), analogous results concerning typical endo-
spore formation were recorded by Swellengrebel (1904).

The large globules produced by B. bifidus are to be interpreted, according to the descrip-
tion given by Tissier (1900), as either gonidangia or regenerative bodies. It was stated in
that paper (p. 90):

De ces boules partent en rayonnant des corps bacillaires d’une grande finesse qui se subdivisent.

, The fairly high resistance (80° C. for 1 hour) shown by the Boas-Oppler bacillus, when tested
by C. Sternberg (1898), which was believed by this author to be due to the presence of spores,
may just as well have been the result of the formation of regenerative bodies. The observa-
tions of Sandberg (1904) and of Rodella (1908) are in accordance with this assumption. Some
data furnished by the last-named author indicate that also the liberation of gonidia was noticed
by him in some cases. Fokker (1901), however, seems to have had, indeed, some really endo-
spore-forming lactobacilli. As far as can be seen from his report, his results agree closely
with those secured by Noguchi (1910), who succeeded for the first time to change experimentally
the anaerobic Bac. bifidus into an aerobic spore-forming bacillus, resembling B. mesentericus
fuscus.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 153

The very conspicuous regenerative bodies of lactobacilli have been mentioned by Weig-
mann, Gruber, and Huss (1907), Kuntze (1908), White and Avery (1909), Rubinsky (1910),
Lohnis and Smith (1916 a and b), and by others. Figures 148 on Plate XI and 172 on Plate XII
demonstrate their characteristic appearance; in the center of the last-named picture a germi-
nating regenerative body is visible. Part or all of what has been described as “‘granule’’ for-
mation by Lwuerssen and Kiihn (1908), White and Avery (1909) and by Koegel (1914), is undoubt-
edly to be explained as relating to the production of gonidia. As this is quite general, but at
the same time naturally varying to some extent, presence or absence of the “granules’’ can not
be made the basis for separating several types or species of lactobacilli, as was done by the
first-named authors. Kuntze’s experiments have already demonstrated the instability of this
feature. Pe

The picture of a pure culture of regenerative bodies of B. cholerae gallinarum, which was
made by /tzerott and Niemann (1895), who were obviously not aware of the special character
of these dark round cells, was reproduced as figure 30 on Plate III. The gonidia of this species,
as well as of B. septicaemiae murium, have been studied by Fedorowitsch (1902) under the name
of protospores.

Gonidia, gonidangia, regenerative bodies, and perhaps microcysts, too, may have been
seen by Albrecht and Ghon (1900), when they made the drawings of B. pestis, which have been
reproduced as figure 11 on Plate D. The important réle played by the gonidia in the repro-
duction of this species was pointed out by N. K. Schultz (1901). Analogous results with B.
pseudo-tuberculosis rodentium have been recorded by Zlatogoroff (1904). Maassen (1904) made
an interesting photograph of what he calls teratologic growth of B. pestis, reproduced as figure
118 on Plate X as a very suggestive object for comparison with Hauser’s Proteus picture (fig.
120).

The formation of the typical round lateral regenerative bodies of B. pneumoniae has been
shown in figure 171 on Plate XII, which was reproduced from our second preliminary paper
(Léhnis and Smith 1916 6). Gonidia, regenerative bodies and gonidangia of another member
of the B. pneumoniae group, the so-called Proteus hominis capsulatus, have been fairly well
studied by Bordoni-Uffreduzzi (1888 6). Two interesting drawings of his have been given as
figure 14 on Plate D and figure 123 on Plate X; the latter, made from a liver cut, is especially
instructive. The production of terminal exospores by B. acidi lactici has been ascertained by
Hueppe (1884) in his first investigations upon the microbial alterations of the milk by micro-
scopic tests as well as by boiling. Though some years later, with Hueppe’s consent, Epstein
(1900) has revoked these findings, it still remains probable that they have been correct (cf.
p. 137). The photograph of B. aerogenes made by Maassen (1904), reproduced as figure 119
on Plate X, equally indicates the possibility of inducing this like other “nonspore-forming”’
bacilli to produce exo- and endospores. Kitt’s drawing of B. phlegmasiae uberis, a variety of B.
aerogenes, is another good example of the mode of producing regenerative bodies within this
group. (See fig. 145 on Plate XI.) That an old stock culture of Actinobacter polymorphus
Duclaux presented itself to Lehmann and Neumann (1912, p. 206) as closely resembling Sarcina
tetragena, would also have to be explained as another case of a pure growth of round regenera-
tive bodies, provided that no error in making the transfers has occasionally happened (cf.
also p. 49).

The peewee of gonidia and regenerative bodies in the life cycle of B. coli has been
studied by Almquist (1893), Adami, Abbott and Nicholson (1899) M. E. Abbott (1900), Fedoro-
witsch (1902), Ohlmacher (1902), and more recently by Kellerman and Scales (1916), whose
very interesting photographs have been reproduced as figures 182, 185, 199, and 200 on Plate
XV. Thedrawing made by Hort (1917 a) of all forms which he saw in his Coli cultures, repro-
duced as figure 180 on Plate XIII, deserves also to be examined carefully. Terminal spore-
formation of the type seen by Hueppe with his B. acidi lactici, has been recorded for B. coli,
though only in one case, by Piccoli (1896). The very close similarity of Maassen’s (1899)
B. esterificans, which, however, despite its otherwise coli-like character, showed a more pro-
nounced tendency to develop such polar spores, is noteworthy in this connection. What
154 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

Rogers, Clark, and Davis (1914), as well as Burton and Rettger (1917), have classified as sporoge-
nous members of the Colon-Aerogenes group may have been analogous strains; but it must be
left in doubt whether the definition of this group almost entirely by biochemical features, as
given by these authors, has not led to an essential alteration in the meaning of the name of the
group. The picture made by Matzuschita (1900) of large “involution” forms of B. coli,
reproduced as figure 36 on Plate ITI, apparently demonstrates a type of gonidangia for-
mation. Another strain was reported to have furnished large globules. Transformation of
the inflated. parts of threads into large round bodies was actually observed by Wilson (1907).

The “small sporoid bodies” found by Eberth (1880) in B. typhi, 1 to 3 to the rod, have been
undoubtedly gonidia; while the round terminal unstainable bodies, described by Gaffky (1884)
as the spores of this species, seem to belong into the same class as the similar formations found
in a few cases with B. coli and B. acidi lactici. Unstainable spaces within the rods, which have
been considered for sometime by Klebs (1887, p. 174) and others to be spores, are to be explained
with great probability as vacuoles. The negative findings of Buchner (1888) and of Pfuhl (1888)
in regard to typhoid ‘“‘spores”’ confirm this view, but, despite a widespread opinion, they can
not be accepted as full proof that the terminal bodies seen by Gaffky were not spores.
This point remains to be settled by future investigations. Almquist’s (1893-1917) studies,
which were mostly centered on this species, have brought out many interesting details con-
cerning formation and further development of gonidia and regenerative bodies by B. typhosus.
Some of his photographs were reproduced as figures 135, 136, and 138 on Plate XI. Large
globular gonidangia, which are mentioned in one of his latest papers (1916), have been described
before by Gamaleia (1900). Direct microscopical investigations upon the reproduction of
normal rods from so-called: protospores, i. e., from gonidia, have been made by Fedorowitsch
(1902). The drawings published by Hort (1917 a) and reproduced as figure 179 on Plate XIII,
furnish a comprehensive picture of probably all forms which may occur in the course of the
development of the gonidia and regenerative bodies of the typhoid organism.:

Analogous results as with B. typhosus have been recorded by Almquist (1908-1917), as
well as by Hort (1917a), in-regard to paratyphoid and dysentery bacilli, and by Meirowsky
(1914 b) with B. paratyphosus B. and B. enteritidis. Meirowsky’s drawings of the paratyphoid
organism, reproduced as figure 178 on Plate XIII, make an excellent counterpart to Hort’s
composite pictures of B. coli and B. typhosus.

Radiate growth, shown in Hort’s pictures as result of the upgrowth from the gonidial stage,
has been found by Beijerinck and van Delden (1902) to be especially frequent with B. radiobacter,
which like B. radicicola is closely related to the Colon-Aerogenes (B. pneumoniae) group. Our
own experiments have furnished many confirmative results (Lohnis 1905 a, L. and Smith, 1916,
a and 6). However, not all radiate growth of bacteria is of the same origin. Sometimes a
similar development starts from the symplastic stage, in other cases a stellate arrangement of
adult cells is noticeable at the time of conjunction (see Chaps. III and IV). The small “swarm-
ing bodies”’ (gonidia) of B. radicicola are known as long as the species itself (Beijerinck, 1888).
Frank (1890), Prazmowski (1890), Morck (1891), and Atkinson (1893) soon confirmed and
extended the findings of the Dutch bacteriologist. Some characteristic drawings made by
Morck were reproduced as figures 66 and 67 on Plate P; they illustrate clearly the forma-
tion of the gonidia in the rods, as well as in the gonidangia. The following description
given by Greig-Smith (1900) concerning the liberation of a budding gonidium of B. radicicola,
pictures in an excellent manner how this occurrence looks under the microscope not only in
this case, but always when motile gonidia are making their way to liberty:

When the bud has separated from the parent protoplasm it pulls and tugs in its endeavor to free itself from

the capsule membrane containing the motionless mother cell, and we have an appearance exactly like that of an ant
attempting to drag along a twig which proves too heavy for its powers.

“Zoospores’’ (gonidia) growing up to new rods within a ‘‘mother spore” (gonidangium)
have been seen by Hartleb (1900). What Hiltner (1900) called spores may have been gonidia
as well as regenerative bodies, and the so-called sporangia of B. radicicola (Hiltner and Stérmer,
1903, pp. 210, 262) are to be interpreted either as gonidangia or as large irregular regenerative
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 155

bodies, produced by the symplasm, a type of the so-called bacteroids, which readily break up
into round regenerative bodies. Photographs were reproduced as figures 195 and 196 on
Plate XV, and more data relating to this type of regenerative bodies are to be found in the
papers of Rossi (1907), Peklo (1910), and Spratt (1912). An exceptionally high resistance of
such bodies against heating has been reported by the last-named author.

The big inflations of acetic acid bacteria, which were studied by EF. Chr. Hansen (1879, 1894,
see fig. 15 on Plate E), are typical gonidangia. Gonidia and microcysts have been also seen by
the Danish author, and excellent drawings of regenerative bodies, published by Henneberg (1898)
were reproduced as figure 57 on Plate N.

Some white and yellow short rods, found by A. Wolff (1908) in milk, furnish another
interesting example of a comparatively high resistance of regenerative bodies against heating.
At least part of the round forms withstood boiling for 15 minutes. Large pear- or bottle-
shaped cells, probably gonidangia, were also observed in the cultures.

That the yellow or red pigmented spore-forming bacillus isolated by Morgenthaler (1916)
is to be considered to represent probably a sporogenous variety of Bact. herbicola, and that
B. violarius acetonicus of Bréaudat (1906) appears to be the analogous counterpart to Bact.
violaceum, has been pointed out on page 137.

Gonidia, gonidangia, and regenerative bodies of B. cyanogenes have been correctly described
by Neelsen in 1880. But what he called the spores of this species have been evidently mostly,
if not all, vacuoles. Genuine exospores, however, have been found by Hueppe (1884).

The very interesting fact that gonidia and regenerative bodies of B. fluorescens have been
photographed by R. Koch as early as in 1877, though later overlooked entirely, has been
mentioned on page 92. Only one year later Ewart (1878 a) published another but more
complete description of the same occurrence. One of our own photographs (Léhnis and Smith,
1916 a, fig. 40) has been reproduced as figure 110 on Plate X as counterpart to the picture
made by R. Koch 40 years earlier, reproduced as figure 109. Another characteristic illustration
of the next step in the development of what Koch called lateral spores was shown as figure
174 on Plate XII (Léhnis and Smith, 1916 6, fig. 28). Genuine spore formation by some
Fluorescens strains has been recorded by Tanner (1918). Bact. brunewm,; of which Matzuschita
(1900) obtained a pure culture of globular regenerative bodies (see fig. 18 on Pl. II) may be
also mentioned here, because it is hardly to be doubted, that it is to be classed as one of the
brown varieties which are rather common in the Fluorescens group. Special studies upon
gonidia, gonidangia, and regenerative bodies of B. fluorescens and of another organism with
polar flagella (Pseudomomas cerevisiae) have been made by Fuhrmann (1906-1908).

That Nitrosomonas as well as Nitrobacter are endowed with the same types of reproductive
organs is fairly well indicated by the photographs made by Winogradsky (1891-1892), reproduced
as figures 49-53 on Plate V. And it is very probable, as was pointed out on page 135, that
Winogradsky’s Nitrosococcus is no separate species, but a pure culture of regenerative bodies

of Nitrosomonas.
(c) SPORE-FORMING RODS.

That B. anthracis is accompanied within the tissue by its gonidia and regenerative bodies
has been demonstrated by R. Koch (1877) in the first photograph which he made of this organism
(reproduced as fig. 111 on Pl. X); though no explanation was given by him in regard to the
round bodies visible therein. It was left, therefore, to Ewart (1878 6) to give the first detailed
description of the mode of formation and development of these reproductive organs. Although
he speaks of spores, it is quite evident from his general report, as well as from the fact that
these bodies did not withstand boiling, that the term was used in a different manner as is
customary to-day. Soon afterwards Fokker (1881) strongly emphasized the important rdle
played by these coccoid reproductive organs, especially in the beginning of the disease, which
had been overlooked by R. Koch, though they were once more photographed by him (Koch,
1881) in the picture reproduced as figure 113 on Plate X. Archangelski (1883), Roloff (1883),
and E. Riein (1883) soon furnished more confirmative material in this direction; pure cultures
of the round bodies were grown by the first-named author through three successive transfers
156 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

and then reverted promptly to normal rods. Equally early discoveries made by Toussaint
upon the formation of gonidangia by B. anthracis, and upon the liberation and further develop-
ment:of the gonidia produced therein, have been reported in the textbook of Magnin and
Sternberg (1884, p. 150). Similar findings were made by Weibel (1888) and by Schroen (1890).
The possibility of an occasional replacing of the regular endospores by gonidangia has become
very clear by these contributions. A curious, but highly characteristic statement, by which
‘raenkel (1891) tried to discard gonidia and regenerative bodies of B. anthracis as “unnecessary”
and. “illegitimate,” because they “merely” reproduce the “legitimate” rods, has been quoted
on page 27. Rodet (1894, p. 18) confirmed once more the correctness of Toussaint’s observations
and the results obtained by. Chauveau and Phisaliz (1895) with an Anthrax strain of low
virulence, in which the endospore formation was entirely and constantly replaced by the
production of round terminal regenerative bodies, represents another important contribution
to our knowledge of the physiological value of these reproductive organs. The photograph
made by Mateuschita (1900) of B. anthracis grown on salt agar, reproduced as figure 57 on
Plate V, shows also regenerative bodies mixed with endospores. And the photograph published
by Giinther (1906), which was reproduced as figure 112 on Plate X, makes a very interesting
counterpart to the two Anthrax pictures of R. Koch (figs. 111 and 113). Réééka (1908) came
again very close to the results secured by Ewart, Toussaint, and others of the earlier workers,
whose publications. evidently had remained unknown to him. WM. Miiller (1912) thinks that
asporogenous strains of B. anthracis occur much more frequently in nature than is generally
assumed; and McFarland (1916, p. 359) gives in his textbook an interesting picture of “ Anthrax
bacilli in glomeruli of kidney’’ showing many small round bodies close to and stained like the
bacilli, which is very similar to Koch’s early photographs. Henri’s important paper (1914)
upon the transformation of regular Anthrax bacilli into round (regenerative) bodies and into an
asporogenous Actinomyces-like growth has been quoted repeatedly. The most characteristic
photographs contained therein were reproduced as figures 58-63 on Plates Vand VI. _

A good picture of a lateral regenerative body of B. subtilis has been published without
comment by Hiss and Zinsser (1914); it was reproduced as figure 150 on Plate XI. Other
photographs showing gonidia and regenerative bodies as formed by this organism have been
furnished in our preliminary papers (Léhnis and Smith, 1916 a and 6); two of them were repro-
duced as figures 65 on Plate VI and 197 on Plate XV.

The big globules found by Blau (1905, p. 125) in cultures of his B. enjlinarionig which either
contained several small plasmatic globules or one endospore, evidently have been gonidangia,
while other round cells, seen by A. Meyer (1901) in cultures of B. cohaerens, Ellenbachensis
and ruminatus, have been typical microcysts. Two drawings made by the last-named author
have been reproduced as figure 64 on Plate. O.

A spore-forming aroma producing bacillus, isolated by Burri (1897) from Swiss cheese,
exhibited quite constantly with all six strains studied ‘‘granular differentiation’’ of the plasma,
coiling of the granular rods into ring forms and development of large globules of several micra
diameter, which appeared like a conglomerate of small cocci. In the textbook of Lehmann and
Neumann (1912, p 460), wherein the bacillus was named B. bernensis, no indication whatever
is given concerning this interesting | report of the Swiss author in regard to the formation of
gonidia and gonidangia by this species.

The large globular f forms of B. M. egaterium, described by DeBary (1884, p. 503), may have
been either gonidangia or microcysts, while the ‘‘Bacillen-Saéckchen,”’ found t by Schroen (1890)
with this species, as well as with B. anthracis, leave no doubt as to their having been gonidangia.
The same holds true with regard to similar cells of B. Ellenbachensis observed by Stoklasa
(1898, p. 120).

Regular endospore formation was recorded by Zopf (1883) with the rod-like, but also with
the large globular cells of B. twmescens. .,

Bac. ozalaticus was found by Kuntze (1904) to produce abundantly ‘‘swarming bodies”
(gonidia). Their,liberation was seen, but not their further development.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 157

That the large round cells of Azotobacter are a type of growth of endospore-forming bacilli
became evident from the behavior of some of our old stock cultures (Léhnis and Hanzawa,
1914). Investigations made by Ford (1916) upon numerous sporogenous bacilli have shown,
on the other hand, that many, if not all of them, are able to produce cells looking very much
like Azotobacter. What Beijerinck (1901) called involution forms of this species have been
probably gonidangia, as was pointed out on page 126 when discussing Beijerinck’s photograph
of these cells, reproduced as figure 189.0n Plate XV. The very characteristic microcysts of
B. Azotobacter have been described as ‘‘ spores” by H. Fischer (1905), Krzemieniewski (1908), and
Prazmowski (1912). The last-named author’s extended investigations upon this species have
added much valuable material to our knowledge of the production and development of the
gonidia and gonidangia of Azotobacter. Some of our own photographs of the different types
of reproductive organs, as found with this organism, have been reproduced as figures 19-21
on Plate II, 68-70 on Plate VI, 186-188, 190, 191, and 193 on Plate XV.

- Good development of lateral regenerative bodies was shown by a bacillus isolated from
‘* Hackfleisch ’’ (chopped meat), when it was grown by Maassen (1904) on lithium chloride
agar; a photograph of it was reproduced as figure 149 on Plate XI. A not yet completely
studied yellow bacillus, No. 41 of our collection, apparently related to Maassen’s organism,
exhibits very prominently the formation and development of gonidia and regenerative bodies,
as is illustrated by the photograph reproduced as figure 183 on Plate XV.

An interesting description of the upgrowth of B. Amylobacter from what seems to have
been its gonidial stage has been given more than 50 years ago by Trécul (18676). Winogradsky
(1902), on the other hand, noticed the production of small coccoid bodies by his Clostridium
Pastorianum, but he was of the opinion that they were unable to act as reproductive organs.
Asporogenous strains, developed from this species, were morphologically, as well as physiolog-
ically, so different from the original culture, that they were hardly recognizable as such.

The ‘“‘cocco-gonidia” of B. Chauvoei have been studied by Ehlers in 1884. In the animal
they grew up to typical rods. The photographs published by Grassberger (1903), Grassberger
and Schattenfroh (1907), and by Hibler (1908), some of which were reproduced as figures 74-77
on Plate VII, 115 and 116 on Plate X, and 151 on Plate XI, contain many interesting details
concerning gonidia, gonidangia, and regenerative bodies, though they have been grossly mis-
understood, especially by the last-named author, as was discussed on pages 67 and 95.

That according to Hibler’s observations old weakened cultures of B. Chauvoei are inclined
to produce polar spores, is of special interest when brought into parallel with the so-called
involution forms of this species, as photographed by Fraenkel and Pfeiffer (1895, reproduced
as fig. 117 on Plate X) and also with the fact found out by Grassberger and Schattenfroh (1907),
that temporarily asporogenous anaerobic butyric acid bacilli, when returning to sporeformation,
at first pass through a stage where they exhibit the appearance of a typical plectridium.

(d) SPIRILLA AND SPIROCHAETS.

The various reproductive organs of V. cholerae (gonidia, gonidangia, and regenerative
bodies) have been seen apparently for the first time by Ferrén (1885), but the various unfounded
hypotheses which he liberally added to his observations, have been evidently the cause that
the latter did not meet with adequate interest, though they were confirmed, at least in part,
by Ermengem (1885, pp. 334-342). The results secured by Dowdeswell (1889-1890) and by
Schroen (1890) are in complete agreement with those of Ferrén, but they, too, despite their
being all based on direct continuous microscopic examinations, have not been acknowledged
by the then dominating German bacteriologists. Even the careful and undoubtedly very
correct studies of Hueppe (1885) upon the formation and germination of the terminal round
regenerative bodies (or exospores %), which he called arthrospores, were not “‘officially’”’ accepted
Because Kitasato (1889) was not able to observe germination of the granules resulting from the
‘ pranular decomposition’ of the cholera bacilli in old cultures, Hueppe’s quite different positive
findings were declared by Woodhead (1891, p. 168) and by Pfeiffer (1896, p. 536) to be entirely
erroneous, though also Babes (1889) reported to have obtained the same results as Hueppe.
158 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

A good photograph of cholera regenerative bodies, reproduced as figure 140 on Plate XI, has
been published by Friedrich (1892); but no word of explanation was given by him in regard to
the round bodies clearly visible in the picture. Upgrowth of new vibrios from the gonidia
has been recorded by Cantacuzéne (1898), Bliesener (1901), Kohlbrugge (1901), and Fedoro-
witsch (1902), while the germinating globules seen by Nakanishi (1901) have been apparently
regenerative bodies. The large globules visible in the photograph made by Maassen (1904),
which was reproduced as figure 121 on Plate X, are gonidangia wherein spore-like bodies have
replaced the normal gonidia. Thorough studies upon formation, multiplication, and germina-
tion of the round regenerative bodies of V. cholerae have been contributed by Almquist (1904—
1917); they confirm once more the accuracy of Hueppe’s early work, and the photograph
reproduced as figure 139 on Plate XI deserves to be studied jointly with figure 49 on Plate M.
A photograph of a pure growth of the round reproductive organs of the cholera organism made
by Hammerl (1906) has been reproduced as figure 158 on Plate XII. Many data given in the
more recent publications of Bittrolf’ (1912) and of Stamm (1914) are in full agreement with
the early findings of Ferran, Dowdeswell, and Schroen.

Very similar results have been recorded by Finkler and Prior (1884-1885) with the vibrio
isolated by them from cases of cholera nostras. The formation of gonidia, gonidangia, and of
regenerative bodies has been fairly well studied. Germination of the round regenerative bodies
of this species, as wellas of V. Metchnikovii, was also observed by Nakanishi (1901). That the
spores attributed by Gamaleia (1888) to the last-named organism have been such indeed may
be justly doubted; the staining reaction, upon which this opinion was based, is certainly not so
conclusive as was generally assumed at thattime. Evidently the same holds true in regard to
the “spores” or “sporelike bodies” found by Van Treghem (1879 6), Weibel (1888), Hueppe
(1891), and others with various vibrios. They all have been mistaken gonidia, regenerative
bodies or vacuoles. With a Vibrio proteus, isolated by Kohlbrugge (1900) from water, the
reproduction of normal cells by gonidia and regenerative bodies has also been ascertained.

Germination of the so-called spores of Spirillum amyliferum has been recorded by Van
Tieghem (1879 a). How the sprouting of the gonidia may lead to a characteristic branching of
the parent cells has been shown by Sorokin (1887-1890). Somie drawings of his Sprlum
endoparagogicum were reproduced as figure 29 on Plate H. That the term “spores” was
incorrectly applied in this case by the author is proven by his statements that these “‘spores”’
had no distinct membrane and were not more resistant against heating than the vegetative
cells. The analogous round bodies found by Esmarch (1887) within Spirillum rubrum leave
also hardly any doubt that they are to be classed as gonidia. As Meirowsky (1914 b) has
demonstrated, they may grow up in exactly the same manner as those of Sorokin’s spirillum.
A sketch illustrating this fact was reproduced as figure 30 on Plate H; Plate XIV offers further
information. Interesting data upon the regenerative bodies of this species, as well as of
Spirillum tyrogenum, have been also collected by Meirowsky. Figure 167 on Plate XII and
the lower part of Plate XIV may be compared in this respect. Branched forms like those
described by Sorokin and Meirowsky have been seen by Kutscher (1895) with Spirillum Undula
and Sp. Serpens. But neither he nor Zetinow (1896) reached a correct insight into the real
character of the round bodies within the spiral cells, which are the cause of this branching.
Photographs made by Doerr (1905) of Spirillum pyogenes show some other case of branching
and budding; they, too, were published, however, without any explanatory remark in the text.
Two salt-water spirilla furnished us (Léhnis and Smith 1916 a) very good examples of the mode
of branching first seen by Sorokin. Formation and germination of the round terminal regen-
erative bodies were equally conspicuous.

The ability of certain spirochaets to produce arthrospores has been discovered by Van
Tieghem (1879 b) in regard to an oyster parasite, and by Zopf (1881) with Spirochaeta plicatilis.
Their findings have been recently confirmed by Gross (1912-1913), whose drawings were
reproduced as figure 63 on Plate O, and by W. H. Hoffmann (1912).

The first information upon the occurrence of motile gonidia with Spirochaeta Obermeieri
and their participation in the life history of this species has been furnished by Heydenreich,
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 159

Guttmann (1880), and Albrecht (1881). In regard to Spirochaeta pallida analogous results have
been secured by Wechselmann and Léwenthal (1905), Leuriaux and Geets (1906), McDonagh
(1912-1913), and Meirowsky (1914). Data concerning the round terminal and lateral regenera-
tive bodies of this species were also published by the last-named author, as well as by Miihlens
(1907), by Krzysztalowicz and Siedliecki (1908), and by Selenew (1910). Similar facts have
been recorded with various other spirochaets by Krienitz (1906), Miihlens and Hartmann (1906),
Gonder (1909), and especially by Noguchi (1911-1912) with all his pure cultures of different
spirochaets (Sp. pallida, micro- and macrodentium, mucosa, Duttoni, phagedenis and gallinarum).
Gonidia, as well as regenerative bodies, were also observed by Calmette (1893) with the spiro-
chaetelike organism, discovered by him in cases of typhus exanthematicus, by Bonhoff (1905)
with his Spirochaeta vaccinae, by Breinl (1905) and Leishman (1918) with Spirochaeta Duttoni,
by Fantham (1911) with different species, by Hindle (1911) with Spirochaeta gallinarum, by
Bosanquet (1911) with Spirochaeta anodontae, by Tunnicliff (1913) with a spirochaete associated
with infections of the accessory sinuses, and by Inada and his collaborators (1916) with Spiro-
chaeta icterohaemorrhagica. The réle played by the “infective granules,” and their ability to
reproduce the spirochaets has been further discussed by Leishman (1909-1918), Balfour
(1911-1913), Riither (1910), and by King, Baeslack and Hoffmann (1913).

The formation of microcysts by spirochaets has been studied by Breinl (1907), Dutton and
Todd (1907), Doflein (1909), Gonder (1909), Fantham (1911), and by Inada et al. (1916). As
the first-named authors found out, the gonida are liberated within the microcyst, which, there-
fore, acquires the character of a gonidangium.

The motile gonidia of fusiform bacilli have been discovered by Plaut (1907) and their
regenerative bodies by Ellermann (1907) who, however, thought they were to be classed as
involution forms. Rosenow and Tunnicliff (1912) observed the slipping out of the gonidia
from the bacilli, whose ends may become inflated, assuming the character of gonidangia.
Multiplication of the “cocci’’ as such was also seen.

(e) MYCO- AND TRICHOBACTERIA.

With B. tuberculosis gonidia, gonidangia, and regenerative bodies have been seen undoubt-
edly, whereas concerning arthrospores and microcysts no such definite statement can be made,
though it is probable that they, too, have been observed. The 2-6 unstainable “spores,”
attributed by R. Koch (1882-1884) to his bacillus, have been either vacuoles or gonidia. Motile
gonidia (“‘Schwaérmsporen”’) were discovered by Klebs (1883 a). Similar “grains bien colorés”
have been mentioned by Babes (1883), but: in addition swellings of the bacilli “suivant leur
longeur”’ were registered, which may have been growing gonidangia. The occurrence of these
reproductive organs with tubercle bacilli has been ascertained by Schroen (1886-1904). The
description given by Lutz (1886) of the coccoid bodies within the bacilli, and of the large round
deeply staining or not stainable thick-walled bodies at:the end of the rods, leaves no doubt upon
their being gonidia and regenerative bodies or exospores, respectively. Metchnikoff (1888 a)
observed that oval cells broke off from short side branches, but whether these should be classified
as arthrospores or as regenerative bodies, can not be decided at present. His drawings, repro-
duced as figure 38 on Plate J, do not permit a final conclusion. Some of the large
swellings shown look very much like gonidangia. Slipping out of the gonidia was seen by
Mafucci (1892). The occurrence of unstainable sporoid bodies within terminal swellings of the
bacilli has been recorded by F. Fischel (1893), while Sander (1893, p. 277) studied not only the
formation but also the germination of deeply staining, polar regenerative bodies. The excep-
tionally high resistance, ascribed by Marpmann (1893) to “spores” of the tubercle bacilli,
which are reported to have not died when kept for 1 hour in the steam, can not be fully
explained at present. If not the result of a faulty test, which seems most probable, it may be that
either sporelike bodies, such as those seen by Lutz and by Fischel, have reached an exceptionally
high degree of resistance, or comparatively large clumps of symplasm have been present,
which have shielded some of the germs enclosed therein. Simultaneous upgrowth of numerous
thin threads from clubs (i. e. gonidangia) of the tubercle bacilli has been noticed by Coppen-

160266°—21——11
160 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Jones (1893). Semmer (1895) reported that he saw the small globoid bodies reproduce
normal bacilli, after they had been liberated by rods and clubs.

A characteristic sketch of regenerative bodies made by Crookshank (1896) has been repro-
duced as figure 550n PlateN. Schiirmayer (1898 a) met with small motile not acid-fast granules, —
which either multiplied as such or grew again up to rods. The pictures published by Arrigo
(1900), which were reproduced as figures 103 and 104 on Plate IX, are very suggestive of arthro-
spores and microcysts; but as the mode of formation of these bodies is not definitely known,
the possibility remains that they, too, have been gonidia and regenerative bodies. Arthrospores
(‘‘conidia”) produced by ‘‘aerial hyphae” of the tubercle organism, were also recorded by
Droba (1901), besides lateral regenerative bodies, which he called ‘‘zygo- and stylospores.”’
Other instances of germinating tubercle gonidia have been mentioned by Fedorowitsch (1902)
and by Rosenblat (1905). In the latter case arthrospores, too, may have been seen; and the
so-called ‘‘Splitter” of the tubercle bacilli, as found by Spengler (1905-1907), apparently have
‘also included both types of reproductive organs. The description which Betegh (1908) fur-
nished of them confirms this view. That the ‘‘granules” of Much (1907) are to be interpreted
as gonidia is beyond doubt, and it is equally certain that much of what has been written about
this discovery by Much (1907-1909), Wirths (1908), Knoll (1910), Heinrich (1912) Kirchenstein
(1912-1913) and by others, could have well been spared if the earlier observations would have
been properly considered. An interesting, but not easily accessible, contribution relating to the
tubercle gonidia, whose partial filterability was ascertained in this case, was published by
Fontes (1910). That the acid-fast as well as the not acid-fast ‘‘cocci” produced by the tubercle
bacilli may develop as such has been further studied by Maher (1910-1913). The large acid-fast
coccoid bodies seen by Wherry (1913) have been evidently regenerative bodies. Probably the
most thorough investigation upon formation and development of gonidia and regenerative
bodies of B. tuberculosis thus far published has been made by Meirowsky (1914 6); some of his
pictures were reproduced as figures 162-164 on Plate XII and on Plate XIV.

That B. leprae is able to develop regenerative bodies has been ascertained by A. Neisser
(1881) and by G. A. Hansen (1882), both of whom called them spores. Gonidia, as well as regen-
erative bodies and exospores, of this organism have been studied by Babes (1883) and more
thoroughly by Luiz (1886). Bordoni- Uffreduzzi (1888 a) termed the polar swellings arthrospores.
Their germination to bacilli has been probably first seen by Czaplewski (1898). The occurrence
of sporelike inclusions within what has been either regenerative bodies or gonidangia of the
leprosy organism was recorded by Babes (1907). Very interesting pictures of the different
reproductive organs of B. leprae, made by Kedrowski (1910), were reproduced as figure 37 on
Plate J and as figures 153 and 154 on Plates XI and XII. Others published by Metrowsky
(1914 8) are to be found on Plate XIT (figs. 165 and 166) and on Plate XIV.

Liberation and germination of the gonidia produced within the threads and the gonidangia
of B. mallet have been probably first discussed by Semmer (1895) as an analogous occurrence to
the same type of reproduction of B. tuberculosis. Marx (1899) and Conradi (1900) saw only
lateral development of the gonidia into buds and branches, and reached the conclusion that no
spores were formed, because no resistance against heating could be ascertained. Wladimiroff
(1903), however, pointed out that the granules formed by the rods can preserve the life of the
species for a very long time, and he places them parallel to those granules (i. e. gonidia)
studied by N. K. Schultz (1901) and by Rothert (1902) with B. pestis and numerous other bacteria.
The rather high resistance which, according to Lehmann and Neumann (1912, p. 548), has been
occasionally observed with B. mallet by Bondme, who found that a temperature of 70° C. could
be endured for 6 hours, and 90-100° C. for 3 minutes, clearly indicates that also in this case
regenerative bodies, arthrospores, or microcysts may have exerted their influence. In the
photographs made by Carpano (1913) of this organism, reproduced as figures 91-96 on Plate
VIII, gonidia, gonidangia, and regenerative bodies are well discernible, while in regard to the
occurrence of arthrospores and microcysts no definite. answer can be secured, neither from the
picture nor from the paper itself.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 161

The gonidia of B. diphtheriae were considered by Klebs (1887) to be spores. According to
Cornil and Babes (1890, Vol. II, p. 59) reproductive organs of comparatively high resistance
against boiling water have been found by the latter author in 2 among 42 cases, but their exact
nature was not determined. That the gonidia are not spores was emphasized against Klebs by
Escherich (1894, p. 73), who m addition pointed out (p. 83) that the club-shaped cells should
not be classed as sign of degeneration, because they develop in abundance on serum within 24
hours. Radiate and parallel growth shown by B. diphtheriae was declared by him (1. c., p. 87)
to be caused by the germination of the ‘‘chromatine granules” to new rods. Good pictures of
typical lateral regenerative bodies of this species have been published by Bernheim and Folger
(1896) and by Spirig (1903). The last-named author’s publications upon this subject (1899,
1903) give also valuable information upon occurrence and germination of gonidia and arthro-
spores. Liberation and upgrowth of diphtheria gonidia have been further recorded by Cache
(1901) and by Fedorowitsch (1902). Chester (1901, p. 14) sees in them the cause of the resistance
of diphtheria bacilli against drying; but it is more probable that regenerative bodies and arthro-
spores also in this case are of greater importance. Pure coccoid growth of diphtheria, observed
by J. Dale (1910), Balfour (1911 d), and by Park and Williams (1914, p. 294), was evidently
the result of round regenerative bodies multiplying as such for some time. In secondary colonies
normal rods were seen to grow again by the first-named author. Interesting illustrations show-
ing the various reproductive organs of B. diptheria may be found in Bergstrand’s paper (1918).

What A. Neisser (1888) called the endospores of B. zerosis was obviously what we now call
gonidia. Especially their lateral germinating while still within the rods, together with their
moderate resistance against heating, leaves no doubt in this respect. What Neisser on the
other hand calls gonidia are those clubbed cells to which this name was given by Bollinger
(1877), when he first saw them with Actinomyces, which, however, when they are fully devel-
oped, become either gonidangia or regenerative bodies. The ‘‘sporogenous” pseudo-diphtheria
bacillus described by Simoni (1898) owed its moderate heat resistance (80° C. for 10 minutes)
probably to its ability to produce regenerative bodies or arthrospores, such as Nakanishi (1900 c)
found with his diphtheroid Bac. variabilis. A characteristic photograph made by the latter
author was reproduced as figure 98 on Plate IX. Others published by E. de Negri (1916),
showing gonidia and regenerative bodies of a Corynebacterium causing malignous granulom,
have been reproduced as figures 99 and 101 on Plate IX and as figure 184 on Plate XV. The
regenerative bodies were seen to reproduce either normal rods and threads or again budding
round forms, as all such regenerative bodies are able to do; the latter, however, were mistaken
by the author as being the growth of a Blastomyces.

The description of Streptobacillus pellagrae as given by Tizzoni and Angelis (1913-1915)
contains sufficient data from which it may be concluded that gonidia, regenerative bodies,
arthrospores, and microcysts have probably all been seen with this Actinomyces-like organism.

The so-called Streptothri cuniculi of Schmorl (1891), which is identical with Bang’s necrosis
bacillus, has furnished instructive pictures of gonidia and gonidangia, which have been repro-
duced as figures 89 and 90 on Plate VIII. Both were found by W. Ernst (1902) to be somewhat
more resistant than the vegetative cells.

Typical arthrospores, formed especially by segmentation of the aerial hyphae, have been
observed quite regularly with probably all Actinomycetes, for instance by J. Israél (1878),
Gasperini (1889), Domec (1892), Garten (1895), Kedzior (1896), Rullmann (1896), Lachner-Sand-
oval (1898), E. Levy (1902), Neukirch (1902), Doepke (1902), Feistmantel (1902), Gilbert (1904),
Schiitze (1908), Krainsky (1914), and others. Data referring to the liberation of the gonidia
formed within the threads or in the clubs, which act occasionally as gonidangia, are not quite
so frequent. They may be found in the papers of J. Israél (1878), Mac Fadyean (1889), Bos-
troem (1890), Eppinger (1890), Crookshank (1896, p. 487) and Loele (1908). Of special interest
is the discovery made by Hollandt (1906) that a typical Actinomyces occasionally may develop
micro- and macro-gonidia like a Crenothrix, and that these may also germinate within the widened
sheath, giving rise to big bundles of new thin threads. While in this case the relationship

’
162 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

to the trichobacteria is very pronounced, connections with the other bacilli, especially. with those
producing endospores, are indicated by the tendency of some Actinomycetes to develop only
one spore at the end of their side branches, as was observed, e. g., by Schiitze (1908) and H. U.
Williams (1912). Such ‘spores’? may look very much like the common lateral regenerative
bodies, but there is also the possibility that they may consolidate their content to a regular
endospore, then showing a growth as was obtained by Olsen (1897) with B. mycoides. A
characteristic sketch was reproduced as figure 20 on Plate F. What Cozzolino (1900) described
as Bac. filamentosus, found in a case of pseudo-actinomycosis, exhibits also very clearly at the
same time the morphological and physiological character of an Actinomyces and that of an
endospore-forming bacillus. The same holds true in regard to a spore-forming Gram-positive
organism, isolated by Gruner and Fraser (1912), which was said to have been ‘“‘regarded by others
as diphtheroid.” Mycobacillus synovialis; isolated by Chantemesse, Matruchot and Grimberg
(1917) from a case of arthritis, produced typical endospores which withstood 90° C.for 30
minutes.

Formation, liberation, and development of the gonidia of the trichobacteria are fundamen-
tally the same as in the case of all other bacteria, but the usually much larger size of these repro-
ductive organs made it more easy to find out that they really play an important réle in the life
cycles of these organisms.

The gonidia of Leptothrix have been studied by Robin (1853), Hallier (1865), Miller (1883),
Arustamow (1889), Dobrzyniecki (1897), and Beust (1907). But it is hardly necessary to em-
phasize that not only the material used in these experiments has been by no means pure, the
whole “genus”’ itself is obviously a mixture of large thread-like forms of very different bac-
teria.

Data relating to the gonidia of Crenothrix are to be found in the publications of Cohn (1870),
Zopf (1879), Giard (1882), and Rullman (1907). Those of Cladothrix have been described by
Zopf (1881), Billet (1890), and Ellis (1912), those of Beggiatoa by Zopf (1881-1883), Engler
(1882), and Zopf (1895), and those of Clonothrix and of other iron bacteria by Schorler (1904),
Ellis (1907-1910), and Petschenko (1913).

That the other types of bacterial reproductive organs, viz. regenerative bodies, arthro-
spores, and microcysts, as well as endospores, seem to be entirely absent in this group is prob-
ably to be explained by the strictly aquatic mode of life to which all these organisms have
adapted themselves. To some extent the macro-gonidia may replace those more resistant
bodies and help to preserve the life during dry spells. In addition, the comparatively thick
sheaths of the trichobacteria are also enabled to act, in case of need, as protective organs.

3. CONCLUSIONS.

According to the details discussed on pages 90-162 the present status of our knowledge
of the various modes and organs of bacterial reproduction can be defined as follows:

(1) Contrary to the monomorphistic theory, which only knows of one constant type of
vegetative cell for every species, and of not more than one type of reproductive organ, the en-
dospore, for merely one group of bacteria, it has been ascertained by numerous independent
investigations that in reality all bacteria are not only distinctly pleomorphous in their vege-
tative growth, but are also able to produce various organs of reproduction. These are: Gonidia,
regenerative bodies, exo- and endospores, arthrospores, and microcysts. All of them are made up
of nuclear substances, which are reinforced by smaller or larger amounts of reserve material and
protected by a more or less resistant membrane. Gonidia and regenerative bodies participate
actively in the process of multiplication, whereas the other reproductive organs are in the first
place resting forms. Gonidia, regenerative bodies, and probably microcysts, too, are produced
by all bacteria, while arthrospores, exo- and endospores are less common; though there are
indications that they will be discovered in many more cases, as soon as these problems are
more thoroughly investigated.

(2) That gonidia play an important réle in the multiplication and reproduction of. bacteria
has been discovered by Perty in 1852, and as the result of numerous later studies it has become
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 163

certain that they are common with all bacteria. Two to four or more gonidia may be produced
within every cell; their formation is most abundant when the bacterial development is at its
height. They may start growing while still within the parent cell, forming buds and branches
or directly new vegetative cells within the old membrane, or they may be liberated either by
actively piercing the cell wall or after the membrane has been partially broken or completely
dissolved. Sometimes they remain temporarily attached to the parent cell by a comparatively
long stem. Most of them, if not all, are enabled to display active motility. One long flagellum
has been found as organ of locomotion. Their resistance against heating is not much higher than
that exhibited by the vegetative cells, but they are more persistent under unfavorable environ-
mental conditions, especially when exposed to drought. Direct upgrowth of the gonidia to
new vegetative cells is not often noticeable in artificial cultures, on account of their minute size,
their motility, and lack of suitable nourishment; but it has been observed occasionally with
representatives of all groups of bacteria, e. g. with staphylo- and streptococci, Gonococcus,
Meningococcus and Enterococcus, with Bact. pneumoniae, typhi, coli, pestis, septicaemiae, pyo-
cyaneum, and fluorescens, with Bac. anthracis, Megaterium, Azotobacter, Amylobacter, Chauvoet,
and oedematis maligni, with Vibrio cholerae and proteus, with various spirilla and spirochaetes,
B. fusiformis, Corynebacterium diphtheriae and mallei, Mycobacterium leprae and tuberculosis,
Actinomycetes, and with the various trichobacteria. Small clumps of gonidia may give rise
to a distinctly star-like growth of bacteria, as it is very common with B. radiobacter and radici-
cola. Because of their variable chemical composition gonidia have been frequently mistaken
as fat, volutin, or other reserve material. They have been also erroneously discarded as sign
of plasmolysis when still enclosed in the parent cell, or as product of “granular decomposition”’
or of “partial bacteriolysis’’ after having been liberated by the dissolution of the cell wall.
Within the blood and tissue, where they may replace the vegetative bacterial cells more or less
completely, they have been confounded with hemoconia, cell granules, and other nonbacterial
corpuscles. On and in the generally used cultural substrates they usually multiply as such,
_ producing a very scant growth, which is only noticeable on very careful inspection. The higher
resistance of the gonidia against chemical substances, like salvarsan, may become of importance
in the medical treatment of chronic diseases. '

(3) Not infrequently the bacterial cells increase considerably in size and undergo certain
morphological changes before they begin to produce gonidia. This is the case when they become
gonidangia, which are able to give birth to a large number of gonidia. Theso-called giant forms
of spherical, oval, spindle-, pear-, or club-shaped appearance, which usually have been lightly
dismissed as “involution forms,’ have always proved to be gonidangia whenever they have
been carefully examined. Formation of gonidangia has been discovered, for instance, with
Gonococcus, Meningococcus, Enterococcus, Pneumococcus, Sireptoc. lactis, Bact. proteus, pestis,
coli, typhi, aceti, radicola, fluorescens and cyanogenes, Bac. anthracis, Megaterium, Azotobacter and
Chauvoei, Vibrio cholerae and proteus, Corynebacterium mallei, diphtheriae and necrophorum,
Mycobacterium tuberculosis, Actinomyces hominis and bovis, and with Crenothrix.

(4) Part of the gonidia of probably all species of bacteria, except those of the trichobacteria,
are small enough to pass bacteria filter. They have been found to play an important réle
as filterable vira, ‘‘infective granules,’ or as so-called chlamydozoa, in diseases like trachoma,
blenorrhoea non gonorrhoica, meningitis, scarlatina, lyssa, and various spirochaetoses. Multi-
plication as such seems to be the rule with the filterable gonidia, but under certain conditions,
especially in the animal, reproduction of normal vegetative cells has been observed. Theplasmain
dead or dying cells of plants and animals generally offers the best opportunity forsuch an upgrowth;
and the assumed transformation of cell granules into bacteria, which has been repeatedly pro-
claimed by adherents to the theory of heterogenesis, becomes easily explainable, as soon as the
omnipresence and the peculiar behavior of the bacterial gonidia is taken into account.

(5) Those reproductive organs of the bacteria which have been called (pro tempore) regen-
erative bodies are produced either by the vegetative cell or by the symplasm. They are either
globular, oval, pear-, or kidney-shaped, or more or less irregular, rod-shaped, and branched.
164 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

When developed by the vegetative cell, they may be found in a lateral or in a terminal position.
The round regenerative bodies are able to multiply as such by fission and by budding, while
the irregular forms either break up into round forms or reenter the symplastic stage. Pure
growth of round regenerative bodies has been observed with Streptoc. lanceolatus, Bact. pestis,
septicaemae, coli, cyanogenes and fluorescens, Bac. anthracis, Azotobacter and Amylobacter,
Corynebact. mallei and diphtheriae, Mycobact. leprae and tuberculosis. The uniform appearance
of these bodies may sometimes cause difficulties in the diagnosis, especially in the case of diphtheria.
Possibly several species of cocci will have to be reclassified as regenerative bodies of other
species. This is very probable in regard to Nitrosococcus. Upgrowth of normal cells from the
regenerative bodies is easily to be observed. It has been recorded with Bact. pestis, coli, typhi,
dysenteriae, radicicola, fluorescens, and cyanogenes, Bac. anthracis and Azotobacter, Vibrio cholerae
and other vibrios, Spirillum rubrum, various spirochaets, Corynebact. mallet and diphtheriae,
Mycobact. leprae and tuberculosis. 1 to 4 sprouts may be produced by one regenerative body.
While still connected with the parent cell the round regenerative bodies may be mistaken for
products of plasmoptysis, and after being liberated for contaminating cocci, whereas the rod-
like regenerative bodies of the micrococci may be confounded with bacilli or actinomycetes
and the irregular regenerative bodies discarded as “involution forms.’’ Instead of multiplying
as such or germinating to new vegetative cells, the regenerative bodies occasionally develop to
gonidangia or produce new vegetative cells by multiple fission. They are easily stainable,
sometimes motile, and usually fairly resistant. The irregularities often observed in heating
experiments with nonspore-forming bacteria are undoubtedly often due to the presence of re-
generative bodies. Slow drying, a comparatively high concentration in the salt content
of the substrates, and low temperatures have been repeatedly found to be favorable for devel-
oping a good growth of regenerative bodies.

(6) In some cases the round lateral or terminal regenerative bodies become unstainable and
assume the character of so-called exospores, which take an intermediate position between the
round regenerative bodies and the endospores. Their resistance is generally not higher than
that of the former and distinctly lower than that of the latter. They have been observed with
various vibrios and spirilla, as well as with the organisms of diphtheria, leprosy, and tuber-
culosis. In addition, however, they have been seen in a few cases with Bact. pneumoniae, coli,
typhi, acidi lactici, cyanogenes, Zopfii, fluorescens, etc., and positive results secured in changing
spore-free lactobacilli into spore-forming rods of the Mesentericus group and in reestablishing
spore formation in asporogenous strains of butyric acid bacilli, make it very probable that
bacteria producing such exospores represent, in fact, connecting links between nonspore-forming
and spore-forming bacilli, and that their appearance in the cultures indicates the time when it
becomes possible to induce a strain to develop its ability of endospore formation.

(7) Endospores can be produced besides by the vegetative cells, by regenerative bodies,
as well as by gonidangia. The latter may contain two or more rudimentary endospores; and
the formation of normal gonidangia is especially frequent in asporogenous strains of otherwise
spore-forming bacilli. The polar endospores of Bac. tetani, putrificus, and of related species,
are closely related to the terminal exospores. The same conditions which stimulate the devel-
opment of the regenerative bodies are also favorable to endospore formation, which, however,
usually takes place at a later time than the first-named process. If the endospore formation is
suspended, the strains concerned become either similar to the regularly nonspore-forming
bacteria, or to the Mycobacteria and Actinomycetes, which in their turn occasionally produce
rudimentary endospores in their club-shaped gonidangia.

(8) Arthrospores are the result of a segmentation of the vegetative cells and of the trans-
formation of these ‘‘joints’’ into fairly resistant bodies of globular shape. They are easily
stainable and withstand drying better than heating. They are most frequent with Myco-
bacteria and Actinomycetes, certain Spirochaets and members of the Proteus group. Every
arthrospore may produce 1 to 4 sprouts.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 165

(9) Vegetative cells of bacteria sometimes transform themselves in their entirety into
reproductive organs, called microcysts, by increasing their size, assuming a globular shape,
and thickening their membrane. After a period of rest, they either resume directly their
vegetative functions, or they germinate like spores, or they break up into 2, 3, or 4 segments
which become vegetative cells. Sometimes they also transform themselves into gonidangia.
Resistance and staining qualities are similar to those of the arthrospores, to which they are
closely related. They have been found with micro- and streptococci, Bact. proteus and acett,
Bac. anthracis, Azotobacter, and other spore-forming bacilli, Vibrio cholerae and proteus, various
Spirochaets and Mycobacteriaceae, and probably will be seen within all groups of bacteria,
as soon as this point has received more attention.
Ill. FORMATION OF THE SYMPLASM AND THE REGENERATION OF CELLS.

1. GENERAL DISCUSSION.

Our observations relating to that part of the life cycles of the bacteria which we proposed

to call the symplastic stage have been summarized in our first preliminary paper (Léhnis and
Smith, 1916 a, p. 700) as follows:

All bacteria studied live alternately in an organized and in an amorphous stage. The latter has been called the
“symplastic” stage, because at this time the living matter previously inclosed in the separate cells undergoes a thorough
mixing either by a complete disintegration of cell wall, as well as cell content, or by a ‘‘melting together” of the
content of many cells which leave their empty cell walls behind them. In the first case a readily stainable, in the
latter case an unstainable, ‘‘symplasm” is produced. ;

According to the different formation and quality of the symplasm the development of new individual cells from
this stage follows various lines. In all cases at first ‘‘regenerative units” become visible. These increase in size,
turning into ‘‘regenerative bodies” which later, either by germinating or by stretching, become cells of normal shape.
In some cases the regenerative bodies also return temporarily into the symplastic stage.

The life cycle of each species of bacteria studied is composed of several subcycles showing wide morphological
and physiological differences. They are connected with each other by the symplastic stage. . . . The trans-
formation of spore-free into spore-forming bacteria seems to be dependent on the conditions acting upon the symplasm
and regenerative bodies.

We pointed out in addition (Léhnis and Smith, 1916 6) that the facts recorded would
certainly not have appeared so new and extraordinary as they did, if it had not (as a result
of the prevailing “simplicity” theory) become a general habit to discard indiscriminately all
evidence offered by the bacteria in this direction as being ‘‘merely dirt,” or “simply slime,”
or ‘‘only detritus” resulting from bacterial “autolysis.”

Very few references could be given at that time. But, as it was the case in regard to the
pleomorphism of the bacteria and their different types of reproductive organs, search of the
literature has furnished additional proof in abundance. The discussion of these facts, as
recorded in the literature, though entirely neglected by the bacteriological textbooks to-day,
will show anew how thoroughly modern bacteriology must be revised, before it will rest again
on the solid ground of actual facts observed, instead of clinging to a narrow dogmatism, as
developed during the last decades.

What we propose to call ‘‘symplasm”’ has already received several names. Pineau (1845)
spoke upon a “substance granuleuse,” Perty (1852, p. 106) about an “in Schuppen angeordnete
Punktsubstanz,’”’ which occasionally formed globular agglomerations. FF. Cohn (1853) adopted
the term “zoogloea.” Lankester (1876) furnished an excellent contribution upon ‘‘macro-
plasts or reproductive disks.’ Haberkorn (1882) introduced the expression ‘‘Bacteriophytom.”’
Marpmann (1884) called the process ‘“Protoplasma-Bildung.” W. Winkler (1899) and Alm-
quist (1917) used the name “plasmodium,”’ the first-named author also the new term ‘“‘Bakterio-
blast.” Raditka (1908) classed the “zoogloea” as a “syncytium,” and Maher (1910-1913)
called the symplasm of the tubercle bacilli their ‘“matrix.”

We would, of course, readily give preference to one of these older terms if it would be
equally distinct or better than ours. But ‘substance granuleuse,” ‘Punktsubtanz,” ‘‘Proto-
plasma-Bildung,” “syncytium”? and ‘“‘matrix” can hardly come under consideration. The
term “zoogloea” is being used so generally for all slimy agglomerations of bacteria that it has
lost its specific meaning. The expression ‘‘plasmodium” implies motility, which is noticeable
only in some cases. The ‘“‘macroplast or reproductive disk,” as described by Lankester, on the
other hand, refers to a special discoid type of symplasm, which is a rather rare occurrence.
So only Haberkorn’s ‘“‘bacteriophytom” and W. Winkler’s “bacterioblast’’ would remain for

166
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 167

eventual adoption. But the latter term is very liable to collide with Perty’s ‘‘blastia”’ (i. e.,
gonidia), and the former expression has been never used as far as ] know except in a short
preliminary paper, and it is hardly to be disputed that the terms “‘symplasm”’ and “‘symplastic
stage”’ are more descriptive and more convenient.

The use of these new terms seems to be preferable also because of their adaptability to
analogous occurrences in the life of other organisms. Some facts will have to be recorded
on the following pages which indicate that undoubtedly with many of the lower organisms, but
probably with the higher organisms, too, the symplastic stage of life plays an important, though
hitherto little known réle. As far as microédrganisms come into view the sentence “Omnis
cellula e cellula” is undoubtedly only partially true. But also with the higher organisms it
does not seem to be so generally applicable as is often assumed.

That bacteria live alternately in various cell forms and in the amorphous symplastic stage,
and that this fact furnishes the explanation of bacterial pleomorphism, as well as the experi-
mental basis upon which the life history of the bacteria can be studied much more thoroughly
and successfully than has been possible thus far, these conclusions apparently have not been
reached before by any other author, evidently because even those who paid attention to the
symplastic stage have worked with only one or a few species and had no opportunity to com-
pare and unite their results with those secured by others, because nobody has ever tried to
collect these data. The following discussion of the discovery of the symplastic stage will
show conclusively how well our new observations agree with those secured already by earlier
authors mostly at a time when the eyes were not yet mistrained to see in the microscope only
the “normal, typical, legitimate”’ forms, and to overlook persistently all that which does not

fit the theory.
(a) THE DISCOVERY OF THE SYMPLASTIC STAGE.

In the ‘“‘Histoire naturelle des zoophytes” of Dujardin (1841, p. 20) the following passage
is of interest:

Quand . . . wun infusoire n’est plus dans des conditions favorables 4 son existence, il se décompose par difflu-
ence, c’est-4-dire que la substance glutineuse dont il est formé s’écoule en globules hors de la masse, laquelle, si les
mémes circonstances continuent 4 agir, se décompose tout entiére . . . mais si par une addition d’eau fraiche
ou d’un liquide convenable on change ces circonstances funestes, la reste de l’animalcule reprenant sa vivacité primi-
tive, recommence 4 vivre sous une forme plus ou moins modifiée.

The ‘Recherches sur le développement des animalcules infusoires et des moisissures”
by Pineau (1845 a) were more specific and more comprehensive. As the beginning of micro-
scopic life he always found on the surface of organic substances submerged in water “une
substance granuleuse, qui précéde toujours l’apparition des étres organisés des infusions, tant
animaux que végétaux.” He observed continuously the transformation of this granular
substance into first immotile, later motile, bacteria and protozoa, as well as into molds. That’
he accepted this transformation as heterogenesis was, of course, not correct, but in agreement
with the consensus of opinion at that time. His drawings, reproduced as figure 78 on Plate Q
(from original figs. 8-27 on Pl. 4 bis), are of great interest, as is the following description given
in a “Supplément” (1845 5) to these ““Recherches:”

On remarquait, en premier lieu, de petits amas de granulations dont les contours commencaient par étre diffus;
peu 4 peu ‘ces amas devenaient plus nettement circonscrits, et ils finissaient par acquérir l’aspect de véritables

- monades, d’abord immobiles, puis douées de mouvement.—Le phénoméne primitif du développement des cellules
et des infusoires consiste essentiellement en une agglomération de granules.

Though primarily relating to other microérganisms, these observations are equally correct
with regard to the regeneration of bacterial cells from the symplasm, while it must be left to
mycologists and protozoologists to decide how far these old findings will be supported by the
results of modern investigations upon the life history of these other microérganisms.

Perty (1852), to whom we owe the earliest detailed information upon the different types
of bacterial reproductive organs, has also furnished very:clear descriptions of the part played
by the symplastic stage in the life of protozoa, as well as of bacteria. His observations upon
168 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

the formation of Podophyra and Actinophrys cells were illustrated by the drawing reproduced
as figure 79 on Plate Q (from original figs. 9 and 10 on Pl. VIII) and described as follows (p. 74):

Im Winter 1848 nahm ich zum ersten Male Podophyra libera wahr, und zwar 5 gréssere Exemplare von etwa
1/50’” Durchmesser lagen an einer Masse aus feiner Punktsubstanz gebildet, in welcher bis zu 1/400/” herab kuglige
Zusammenballungen von Molekiilen lagen, von welchen die grésseren immer deutlicher die Beschaffenheit teils von
Podophyra, teils von Actinophrys annahmen ... In den kleinsten dieser Zusammenballungen waren erst nur
wenige Molekiile unregelmassig und locker vereinigt, in den grisseren zahlreichere zu regelmissiger Kugelgestalt
mit allmahlich sichtbar werdenden Strahlen; von diesen grésseren zeigten ein paar zitternde, hin- und herriickende
Bewegung. sie :

In old infusions, filled with bacteria and protozoa, Perty saw the motile bacteria agglutinate
and transform themselves into immotile, crumbly, granular clumps of various shape, which
when transferred into fresh solutions gave new bacterial growth (1. c., p. 109, 110). Most
characteristic and essentially very correct is the following passage (I. c., p. 113), although the
heterogenetic standpoint, which Perty shared with his contemporaries, is clearly noticeable:

Die organisierten Kérper scheinen sich bei der Faulnis aufzulésen und in eine unaussprechlich feine Punktsubstanz
umzuwandeln, welche nicht in dem lebenden Organismen vorgebildet, sondern schon eine neue Construction ist.
Diese Punktsubstanz (welche manchmal Neigung zeigt, sich in kuglige Agglomerate zu ballen ... ,auch bisweilen
zitternde Bewegung zeigt) ist aber nicht eine durchaus homogene, sondern innerlich schon wieder verschiedentlich
determiniert, so dass aus ihr Wesen verschiedener Art, namentlich Vibrioniden und Monadinen entstehen kénnen.

Evidently the same fact has been studied by F. Cohn (1853) when he described the
“Gallertkugeln” present in old bacteria infusions as being “‘ganz verschieden von den
oberflichlichen Bakterienhiuten.” ‘‘Zoogloea’ termo’’ was defined as ‘‘massae mucosae
globosae, uvaeformis,” and that later the floating films of bacteria usually have been called
zoogloea is obviously not in accordance with these original findings and statements.

When Robin (1853) described his Leptothrix buccalis he also mentioned that its long cells
were ‘“‘réunis généralement, par la base, & une gangue amorphe granuleuse,” which he thought
to be the detritus of food remnants in the mouth, which, however, as will be seen later, may
have been just as well flakes of symplasm not yet transformed into new threads. Later (1871)
in his “Traité du microscope”’ the French author wrote concerning the same subject (p. 927):

Les nombreux granules, trés fins, de volume uniforme, que le microscope montre dans le mucus, 4 la surface des
cellules épithéliales, linguales, intestinales, dans beaucoup de déjections intestinales, ont un aspect trés characteristi-
que; elles ont, d’un auteur 4 l’autre, regu des noms trés differents (zoogloea, couche muqueuse primordiale, pellicule
proligére).

It is, of course, not to be denied that this, as well as the earlier quotations given above,
may also relate in part or wholly to agglomerations of bacterial gonidia and perhaps of regenera-
tive bodies. Instruments and methods available at that time did not allow a clear distinction;
even to-day such a task is not always easy. But the data to be given on the following pages
will show that it is at least very probable that the symplastic stage has been already discovered
by these early authors.

In the same manner as the bacterial gonidia proved to be of great value in supporting
(apparently) the old theory of heterogenesis, as was discussed on pages 148-150, so naturally
the amorphous agglomerations of bacterial symplasm can be mistaken even more easily for
any other kind of vegetal or animal ‘“‘slime,” which by its ability to produce bacteria may be
readily accepted as decisive, though in fact deceiving, proof of bacterial heterogenesis. ‘This
is the obvious reason why this hypothesis could be adhered to until recently by various authors.
Their observations were usually fairly correct; but the absence of adequate knowledge upon
the symplastic stage of the lower organisms allowed them to indulge in lofty speculations and
made it impossible, on the other hand, to explain the facts recorded in an intelligent manner.

It was pointed out before (Léhnis and Smith 1916 a, p. 699) that especially those much
doubted and disputed findings recorded by Bastian (1872-1914) now are coming under entirely
new aspects. Many of his pictures prove beyond doubt that he, indeed, observed very care-
fully the upgrowth of cells from the,symplastic stage. Those reproduced as figures 205-208
on Plate XVI (from original figs. 11 B and ©, 12, and 13 on Plates IV and V, 1905, and fig.
33 A on Pl. XI, 1907) are especially characteristic and worthy to be compared with Perty’s
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 169

and Pineau’s drawings, as well as with other photographs shown on the following plates. That
Bastian’s experiments were faulty in many respects needs not to be pointed out, as he declared
himself (1905, p. 184):

I emphasize the fact that the researches which I am about to detail have no pretence to be conducted in ways
that are proper and usual in the great bulk of bacteriological inquiries.

Sterilization was often incomplete, and what he observed may sometimes have been dead
cells, occasionally also silica conglomerates, though undoubtedly not so generally as Payne
(1916) and Onslow (1917) are inclined to assume, as Bastian himself (1907, pp. 247, 275) was
fully aware of this possible source of error. He also reported (1872, Vol. II, p. 209) that he
saw the bacteria disappear, dissolving themselves into ‘‘a whitish refractive and homogeneous
protoplasm,” and more recently he added other interesting details concerning the formation
of the bacterial symplasm and the regeneration of cells from it, which will be considered later.
In view of all this, it is certainly not easy to understand how he could continue to assert (1903),
p- 3):

The lower forms of life, both animal and vegetal, are ever springing up anew . . . from matrices wholly
unlike themselves.

That Bastian did not become aware of the fact that he merely was dealing with the two
manifestations of bacterial life is especially surprising when one considers what a great number
of publications has been issued soon after his first book came out, all discussing the ability of
the bacteria to form these “‘matrices’”’ and to spring forth from them. Osler and Schéfer wrote
in 1873:

In vielen Krankheiten befindet sich in dem Blute eine kleinere oder gréssere Anzahl farbloser, granulierter Mas-
sen, die Grésse der weissen Blutkérperchen oft um viele Male iibersteigend, und unter starken Vergrésserungen aus
kleinen, blassen Teilchen zusammengesetzt erscheinend.

They also observed under the microscope that these substances were able to produce
threads and motile bacteria.

That the “ Microsporon septicum,”’ described by Klebs (1872), behaves in an analogous
manner was noticed by this author (1873), when he studied its growth in gelatin kept in Geiss-
ler’s chamber. Besides regular colonies made up of rods, homogeneous yellow agglomerations
Were seen, sometimes spherical, sometimes of irregular outline showing protrusions “similar to
the pseudopodia of amoebae.” These plasmatic masses were seen to produce new bacteria,
and, on the other hand, bacteria were seen to fuse together into such homogeneous plasma.
Diphtheria cultures were found (Klebs, 1875 a) to contain small rods and “dark brown masses,”
homogeneous in the center, finely granulated at the outside, which were seen to transform
themselves gradually into small bacilli, first at the edge, later in the center, too. In cases of
endocarditis the same author (Klebs, 1878) noticed that the hyalin substances present were
formed by micrococci. Stained with hematoxylin small granules became visible at the out-
side, while the center proved to be homogeneous, composed of round and of irregular agglom-
erations. Figure 210 on Plate XVII is a reproduction of Klebs’s drawing (original fig. 3 on
Pl. II). The author concludes:

Esscheint . . . nicht bezweifelt werden zu kénnen, dass jene oft erwihnte hyaline Masse zum Teil aus den
umgewandelten Schistomyceten besteht.

Furthermore, in 1883 Klebs pointed out in regard to tuberculosis, that Koch’s bacilli are
only one phase in the development of the causative agent, which presents itself in young inocu-
lations as finely granulated masses, which later give birth to the bacilli. The following state-
ment (1883 b, p. 131) is of great interest:

An isolierten Tuberkeln, welche in lufthaltigen, mikroskopischen Kammern der direkten Beobachtung zuganglich
gemacht werden, sieht man diese feinkérnigen Massen aus dem Gewebe selbst hervorwachsen. Dieser Umstand
sowie ihr Auftreten in den Kulturen beweist, dass sie einen integrierenden Bestandteil des Tuberkelorganismus dar-
stellen, und diese Auffassung wird noch dadurch bestitigt, dass wie schon von Koch zugegeben, seine Bacillen in
offenbar frisch wuchernden Tuberkeln oftmals nur in tiberaus geringer Anzahl gefunden werden. Andrerseits zeigt
ihr constantes Nebeneinandervorkommen auch in méglichst reinen Culturen (ich habe die Anwesenheit der kérnigen
Massen auch in einer von Koch selbst hergestellten Cultur nachweisen kénnen), dass beide Bestandteile gemeinsam
zu dem Wesen des Tuberkelorganismus gehéren.
170 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

_ The large hyalin globules noticeable in cases of rhinoscleroma within and without the cells
were declared to be (Klebs, 1887, p. 271) “wahrscheinlich Uberreste abgestorbener Bacillen-
haufen.”’ is

Another set of analogous observations has been furnished simultaneously by Letzerich.
In cases of diphtheria he found (1873-1876) large spherical or pear-shaped plasmatic bodies,
which apparently were formed by a melting together of the bacilli-lying side by side in the
membrane, and which produced in their turn small coccoid bodies, which later were set free
by rupturing the membraneous outside of these cyst-like bodies. Blood cultures furnished
in addition large irregular agglomerations, sometimes showing amoeboid movements, occa-
sionally uniting with each other or dividing into several parts. They also transformed them-
selves into ‘cocci,’ first at the edge, later in the center. The drawings reproduced as figure 80
on Plate R (from original figs. 5-10 on Pl. XII, 1876) look very queer and suspicious at
the first moment. But we will have to consider later so much additional evidence, proving
beyond doubt that, indeed, genuine bacteria may behave sometimes very much like Myxo-
bacteria and Myxomycetes, that these drawings may find a place here, despite the fact that the
trustworthiness of Letzerich’s findings is undoubtedly much impaired by the manner in which
he supported the most extravagant hypotheses & la Hallier. The plasma spheres of diphtheria,
for instance, are said to develop occasionally to fungus spores, and the diphtheria organism is
declared to be a “Tilletia diphtheritica.’’ Obviously under Klebs’s influence, a somewhat
more sober standpoint was later taken by Letzerich (1880), who now also studied under the
microscope in Geissler’s chamber the transformation of the spherical ‘‘wax-like’’ masses of the.
‘plasma cells’’ into normal colonies of bacteria.

Billroth (1874) found amorphous agglomerations in broth, serum, and various infusions,
which he thought to be perhaps “ plasmodia”’ and, on account of their showing occasionally a
development of coccoid bodies, to be possibly the cause of a “‘eigentiimliche Coccosentstehung
aus einem zu einer Platte verschmelzenden Plasma”’ (1. c., p. 10). Onthe other hand, he also
observed that in old cultures of bacteria and algae a dissolution of the cells took place. But
apparently he did not think of any connection existing between these two facts.

In cases of variola Weigert (1875) discovered in the blood, as well as in various organs of
the body, peculiar, granular; spherical, or cylindrical ‘‘Bakterien-Schliuche,” which were
embedded in amorphous masses produced by the surrounding cells, and which later themselves
disappeared completely, becoming entirely homogeneous, or transforming themselves into
smallest granules.

Probably the most important among the early contr.outions to the knowledge of the
symplastic stage of the bacteria has been made by Lankester (1876) in the course of his investi-
gations upon the red sulphur bacteria (Bact. rubescens). He noticed that repeatedly, always
when the food supply became low, “‘macroplasts”’ or ‘reproductive discs’? were formed, which
under high magnification (1,100) presented themselves to be either “truly homogeneous”
or having a “minutely punctuate structure,” or being regular colonies of globular cells, sur-
rounded by a membrane. Further studies showed that the minute granules “indicate so many
new units or centers of organization,” and it is stated (p. 36):

Just as we find exceptional cases in animal and vegetable cells in which a mass of protoplasm gives rise simul-
taneously to numerous nuclei, each of which becomes surrounded by a segregated mass of protoplasm and produces

a numerous cell progeny by multicentral segregation, so it appears that in the large discoid macroplasts of Bacterium

rubescens a formation of innumerable new plastids occur—not by a progressive division into two, four, eight, etc.—
but by 9 simultaneous multicentral segregation.

It is, indeed, highly surprising, how such an excellent description of this very interesting
process, which was accompanied by so characteristic pictures as those reproduced as figure 209
on Plate XVI, could later be entirely forgotten, and despite the various studies made by
Winogradsky. and by Molisch with the same or similar organisms nothing more was learned
about these macroplasts, which, it is true, have no place within the narrow limits of the rigid
dogma of bacterial monomorphism and constancy, which both authors felt obliged to defend
vigorously. And yet, in addition to all other evidence, it is only necessary to compare Lankes-
ter’s picture with the photograph of a Nitrosomonas “‘zoogloea” made by Winogradsky (1892),
No. 2.]" LIFE CYCLES OF THE BACTERIA—LOHNIS. 171

which is reproduced as figure 270 on Plate XXI, to show at once that despite their conspic-
uousness it must have been difficult to overlook these things persistently.

But it is not less surprising that Robert Koch, too, just as he first saw and photographed the
gonidia and regenerative bodies of various bacteria and later forgot all about them (cf. pp.
92 and 94), gave a faithful picture of the partially homogeneous, partially granular agglomerations
of the symplasm, found by him in spreading abscesses in rabbits, on Plate I of his ‘‘Unter-
suchungen iiber die Aetiologie der Wundkrankheiten” (1878), reproduced as figure 211 on Plate
XVII (from original fig. 8), but afterwards never returned to this subject. And not only this;
he did not even hesitate (1884 a, p. 54) to “explain” Klebs’ (1883) observations concerning the
formation of symplasm by the tubercle bacilli by simply asserting that the unstainable masses
seen by this author had been ‘‘serum.”” A glance at the quotation printed above suffices to
demonstrate the utter inadequacy of this ‘“‘explanation.” An equally superficial ‘‘explanation’”’
was furnished by Loeffler (1884, p. 431) in regard to Klebs’ and Letzerich’s discoveries. Their
homogeneous plasmatic agglomerations are simply declared to have been normal bacterial
“colonies” in the gelatin used, despite the very distinct differentiation, made by Klebs (1873)
as well as by Letzerich (1880), between colonies and plasmatic ‘‘wax-like” agglomerations.
Moreover, neither R. Koch’s nor Loeffler’s assumption gives any explanation for those homo-
geneous or granulated masses found by Klebs, Letzerich, Weigert, Osler and Schaefer, as well as
by R. Koch himself, to be present within the body.

Nevertheless, from now on the symplastic stage of the bacteria was excluded from the
“legitimate” bacteriology of R. Koch and his followers, as was the case with pleomorphism,
with the gonidia, and with the regenerative bodies of the bacteria. During the following
decades, however, many confirmative results have been gathered by independent workers, which,
taken together, leave no doubt that the omission from the bacteriological textbooks of all data
pertaining to this subject is quite unwarranted.

In 1880 and 1882 G. A. Hansen wrote upon ‘‘brown elements” occuring in leprous tissue
which are filled either with a finely granulated mass or with normal rods. His descriptions, as
well as his drawings, resemble those of Klebs and of Letzerich very closely, and they found an
early confirmation in Babes’ leprosy studies (1883), wherein it was ascertained that the tissue
often contains ‘‘ une masse homogéne et granulée” while practically no microbes are visible.

Albrecht (1881) noticed in the course of his investigations upon the development of Spiro-
chaeta Obermeiert that besides motile granules (gonidia) and spirochaets also ‘‘unférmliche
Haufen von dunklen Punkten” were present in the blood, which continually changed their
form.

At the same time Fokker (1881) noticed that Bac. anthracis produced within the body a
similar finely granular ‘‘detritus,” and further investigations (1882) led to the result that in
300 inoculation tests on mice in no instance bacilli were found, three times coccoid bodies, and
in all cases in blood and liver
eine bedeutende Menge grobkérniges Material, das sich gar nicht farben und weder durch Alkohol noch auch durch
Erhitzen am Glase ankleben liess.

Twenty years later Fokker (1902, pp. 16-26) returned to this subject, reporting that he found
the ‘‘dissociation” of the anthrax bacilli in water, serum, and in broth to be especially pro-.
nounced during the first hours after having made the transfers. Comparative plating on agar
gave, for instance, the following number of colonies:

At the beginning.........------+----.--222-22-02 eee ee erence eee 720 2, 000
iter I=? NOW Ss .23ccces cass ax eres oi eeecaaes sie Geereee 0 8
8-4 hours... ---- 22-2 --eee eee eee eee eee eee erent eetee ete a
L day... 0.22 c ccc eee eee ee eee ee eee teeters 10, 300 780
2 days... 02.2... 2.2 eee eee cee eee eee eect teeters neni ee 3, 500

Concerning the regeneration of the bacilli it is assumed

dass sich aus diesem gelésten Plasma Granula verdichten und aus diesen Granula wieder Bacillen hervorgehen, oder
aber, dass nachdem sich ein osmotisches Gleichgewicht ausgebildet hat, das Plasma wieder von den Geriisten auf

genommen wird.
172 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [ Von. XVI.

The latter possibility is considered to be more probable. As will be seen later, in fact both
modes of reconstruction may take place.

Owing to Fokker’s adherence to the theory of heterogenesis, it was unavoidable that in the
same manner, as he accepted the participation of the gonidia i in the reproduction of the bacteria
as full proof of the heterogenetic origin of these organisms (cf. p. 149), so also the symplastic
stage had to serve as an important part of his ‘‘neue Bakterienlehre.” The ‘‘milk granules”
and the flakes of ‘‘casein,” which gave him (1901) ‘‘new” bacteria, have been undoubtedly
flakes of symplasm and regenerative bodies growing out of it, as was discussed in our first pre-
liminary paper (1916 a, p. 692). The photograph of such a flake of symplasm, studded with the
round regenerative bodies of Streptoc. lactis, taken from milk, has been reproduced as figure 214
on Plate XVIII (from original fig. 33, Léhnis and Smith, 1916 a).

In his studies upon the ‘‘microzymas” Béchamp (1883) naturally also encountered the
symplasm and saw it taking part in the reconstruction of cells. A characteristic drawing of
his (origmal fig. 5, Pl. I) showing the regeneration of yeast cells from the agglomerations of
‘‘microzymas” in vinegar has been reproduced as figure 81 on Plate R. It should be compared
with the analogous pictures made by Pineau, Perty and Bastian (reproduced as figs. 78 and 79
on Plate Q, and fig. 206 on Pl. XVI). The description given by Béchamp (1. c., p. 472) is worth
quoting: ;

La mére de vinaigre est membraneuse et les microzymas y sont réunis par une matiére unissante, hyaline. A
mesure que les cellules apparaissent, les choses se passent comme si les microzymas consommaient en méme temps et
les aliments qui leur sont fournis par le bouillon sucré et la matiére hyaline qui les unit, et s’agglomérant, sécrétaient
la matiére qui forme 1’enveloppe, les parois de la cellule.

Very similar observations have been recorded by Cocardas (1884); and his drawings are
much alike to those of the earlier authors. But the abstruse ‘‘nouvelles idées” presented in
his paper make it, indeed, quite conceivable why the French ‘‘Société de Botanique” did not
accept his contribution for publication.

That the ‘‘Bacteriophytom”’ of Haberkorn (1882) is identical with what we call symplasm
has been mentioned above. It was always found in old cultures as large, compact, oval, or
globular bodies with granular surface of different refraction, producing either ‘‘Kérnerhaufen”’
(i. e. agglomerations of round regenerative bodies) which later grew up to rods, or directly
regenerating numerous young bacilli which made the ‘‘Mutterboden”’ look “wie gestrichelt.”’
Two of our photographs, reproduced as figures 256 and 257 on Plate XX (from Léhnis and
Smith, 1916 a, original fig. 15, and 1916 }, original fig. 48), may demonstrate the correctness of
these early observations.

Marpmann (1884, p. 44) was equally right when he wrote:

Unter Umstanden lést sich eine ganze Familie oder ein Enzelwesen ganz in Schleim, Protoplasma auf. :
Dieses Protoplasma kann wahrscheinlich unter giinstigen Umsténden wieder die urspriingliche Spaltpilzform hervor-
bringen. Man findet solche Schleimmassen in allen dlteren Kulturen, daher ist es wahracheinlich, dass der Lebens-
prozess der Spaltpilze mit der Protoplasma-Bildung abschliesst. . . . Ob diese Schleimbildung stattfindet, wenn
Spaltpilze in den gesunden tierischen Kérper gelangen und sich nicht normal entwickeln kénnen, ist nicht bewiesen,
dagegen sehr wahrscheinlich.

“Les amas enkystés de bacilles’’ which according to Babes (1883) occur in “les tissus
scléreux de tuberculose,’’ and which were often found not to contain any rods, but instead of
these ‘‘grains bien colorés, ronds ou cubiques, se rapprochant de |’apparence des sarcines,”’
may well have been in part or wholly agglomerations of symplasm transforming itself into
regenerative bodies. ‘‘La tuberculose zoogléique’’ of Malassez and Vignal (1883-1884) with
its characteristic “masses zoogléiques de microcoques,” which in old lesions reproduce the
typical bacilli of &. Koch, eliminates all doubt, if there still should be any, that Klebs’, not
Koch’s, findings were the more correct and more complete ones. In their first paper the French
authors state explicitly:

Il existe une parenté entre nos zooglées et les bacilles de Koch.

But in the second one they are again somewhat doubtful in regard to this point, and this
probably was the cause why later authors sometimes declared this type of tuberculosis to have
been pseudo-tuberculosis. The staining reaction of the ‘‘zoogloea’”’ was found by Malassez
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 173

and Vignal (1884) to be variable, and it is pointed out specifically that such clumps may have
often been mistaken for giant or mast-cells. Amrusch’s (1886) observations “‘ Uber eine Zoogloea-
Form der Tuberkel-Organismen”’ confirm fully the earlier discoveries. Itis true that the German
author was very anxious to point out that his findings had nothing whatever to do with those of
the Frenchmen, because ‘“‘his’”’ zoogloea took the stain (weakly), while that of Malassez and
Vignal is declared by him, quite contrary to their own statements, to have been unstain-
able. A study of his paper, however, reveals clearly, that he once more saw the melting together
of rods to slimy, granular agglomerations, from which later new bacilli grew out; in addition
to this he noticed, when studying the living material, that within these clumps continuous
movements took place and the outlines changed frequently, exactly in the same manner as
had been observed before by Stricker, whose results obtained with putrefying bacteria are
reported in this paper. .

In cases of ‘‘pseudo-tuberculosis’”’ of guinea pigs Eberth (1885) found that instead of rods
“‘micrococci’’ were present, especially in the liver, occurring in large clumps, stainable at the
edges, unstainable in the center, just like R. Koch’s septicaemia “‘micrococci.’”’ Nocard (1889)
recorded similar facts, and Zagari (1890) thought himself to be justified, because he, like Eberth,.
Pfeiffer, and others, obtained slimy agglomerations of short rods (0.4X0.8-1.0u), which he
incorrectly calls micrococci, in cases of pseudo-tuberculosis of rodents, to “‘explain”’ the “ tuber-
culose zoogléique,”’ found by Malassez and Vignal in man, as having been “merely pseudo-
tuberculosis,” though he, of course, never saw the Koch bacillus develop from the zoogloea,
as these authors had seen.

That spirilla may also enter the symplastic stage has first been recorded by Finkler and
Prior (1884-1885). Old cultures of their vibrio furnished ‘“‘kérnige Massen, die man als Detritus
bezeichnen kénnte,’’ which, however, transferred into new substrates (broth or gelatine) after
12-24 hours produced “‘ausserordentlich kleine Kommabacillen,” slowly growing up to their
normal size. Soon afterwards Weibel (1887) added to his description of a ‘‘ Vibrio aus Nasen-
schleim”’ the following footnote (p. 469):

Eine eigentiimliche Beobachtung habe ich noch nachtriglich gemacht. Praparate von einer 4 Wochen alten
Agar-Cultur zeigen nirgends mehr intacte Stabchen oder Faden, sondern nur einen kleinkérnigen Detritus, der die
Farbe schlecht annimmt—also das Bild abgestorbener und zerfallener Bakterien. Wohl characterisierte Sporen
findet man auch nicht. Und doch, wenn man aus diesser Masse in gewdhnlicher Weise auf neuen Agar iibertrigt, so
wichst in 1-2 Tagen die tippigste Cultur. Nahere Untersuchungen hieriiber werde ich nicht verséiumen.

But in his later publication he did not touch this point again. Maybe he did not like to
be attacked by R. Koch and his pupils in the manner as was done to Finkler and Prior.

Dowdeswell (1889-1890) found in his cholera cultures small masses of minute granules and
sometimes of fine short filaments. Some of these masses were seen slowly moving around in an
amoeboid or gyrating manner. The small granules contained therein reproduced vibrios and
filaments by uniting and filling the cells, and it is also reported that—
the cells of the filaments, in masses large or small, are seen to actually coalesce or be fused together, as if they were
evolved out of a mass of primitive plasma.

_ Curiously enough, also this British author sedan was unacquainted with the analogous
discoveries made by Lankester and the other early authors, who all had observed the same
occurrence and had described it with practically the same words.

P. Ernst (1888) reported that he got prompt development oz Bac. zerosis from the small
granules visible in a “krimligen, pulpésen Einbettungsmasse,” which was the only thing
present in 3-months-old cultures. Already some years earlier Hauser (1885) had seen that the
rods and filaments of his Proteus transformed themselves often quickly (within 24 hours) into
a finely granulated zoogloea, wherein only minutest globules remained visible, and that later
new rods and threads “germinated out of the clumps of zoogloea.’’

In the course of his leprosy studies Lutz (1886, p. 81) made the interesting drawings repro-
duced as figure 82 on Plate R, of what he, like later P. Ernst, called ‘‘Einbettungsmasse,” and
of which he says:

Diese Einbettungsmasse spielt bei der Formation der leprésen Neubildung eine wichtige Rolle, da sie den grésseren
Teil der Stabchencolonie bildet.
174 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [von XVI.

Sometimes these slimy agglomerations looked to him like—
aufgeléste Nebelflecke in Gestalt durchsichtiger Wolken, die aus lauter Piinktchen und Tiipfchen bestehen.

In cases of rhinoscleroma Cornil and Alvarez (1885) found not infrequently within the
giant cells, replacing the bacilli smaller or larger hyalin masses, which were stained more
easily than the encapsulated bacilli. Good pictures of all intermediate stages are given in the
paper, but the authors do not reach a definite conclusion, whether these masses were produced
by the bacteria themselves or by the cells under the influence of the bacteria. That Klebs
(1887) made analogous observations, has been mentioned before (p. 170), and in the textbook
of Cornil and Babes (1890, Vol. II, p. 326) the question is taken up once more, but again no
final answer is given, though the authors say (1. c., p. 327):

Existe-t-il une relation de cause 4 effet entre les microbes du rhinosclérome et les masses hyalines des grandes
cellules? Cela nous parait vraisemblable.

Billet (1890) made some interesting remarks upon the réle played by the “‘ état zoogléique”’
in the life history of Cladothrix dichotoma and of other bacteria, and stated (p. 215):

L’état zoogléique définitif . . . nous semble devoir constituer un caractére de premier ordre pour la différ-
enciation des diverses espéces de Bactériacées.

Experiments with bacteria from the root nodules of leguminous plants furnished Frank
(1890) sometimes peculiar colonies, formed by small motile ‘‘Schwirmer,” which later devel-
oped to ‘‘Zoogloeahaufen,” wherein the small bodies were visible ‘in kochender Wimmelbe-
wegung.” Figure 83 on Plate R illustrates these different types of growth (reproduced from
original figs. 34 a-c on Pl. VIII).

Similar colonies were seen by Prazmowski (1890) in pure cultures of B. radicicola, as well
as in the nodules themselves; they were surrounded, however, by a comparatively solid mem-
brance,; which was left behind, when the bacteria were liberated. The drawings made by
Morck (1891), reproduced as figure 84 on Plate S (from original figs. 3 and 12 e and f),showing the
clumps of plasma, which he saw to be formed by a melting together of the “bacteroids” in the
root nodules of various legumes, make an interesting counterpart to Lutz’s leprosy pictures,
reproduced as figure 82. These clumps were found to be able to reproduce small rods, and on
the other hand “granulated protoplasm” was seen to be present in the youngest part of the
nodules, here forming small coccoid bodies, which later developed to normal bacilli.

Tf all the foregoing observations had elicited the interest, which they actually deserved, the
important discoveries made by Thazter (1892-1904) upon the Myxobacteriaceae undoubtedly
also would have been valued by the bacteriologists in a more adequate manner, than it was
generally their fate. Instead of being left aside as some curiosity, Thazter’s studies fully deserve
to be kept in mind when planning thorough bacteriological investigations in the future. A
short discussion of the relations existing between bacteria, myxobacteria and myxomycetes
will be given at the end of this historical resumé.

In human, as well as in bovine, actinomyces “grains” Crookshank (1896) found the central
portion ‘composed as a rule of a structureless core,” but Lachner-Sandoval (1898), recording
the same fact, pointed out that Gram staining makes small round Gram-positive bodies visible
within the Gram-negative “ detritus,” which sometimes also may contain some pale shadows of
former threads. Another observation, which the author was ‘‘unable to explain,” because he,
too, was not acquainted with the earlier findings concerning the upgrowth of new cells from
such “granular detritus,” i. e., from the symplasm, was very correctly described by him (1. c.,
p. 41) as follows:

Bei ganz jungen Individuen sieht man vielfach neben einem Faden oderaufdem Endeeines solchen . . . eine
wolkenférmige Masse, die aus winzig kleinen, lichtbrechenden Kérnchen besteht.

It will suffice to compare this description with one of our photographs, reproduced as
figure 267 on Plate XXI (from Léhnis and Smith, 1916 a, original fig. 28) showing a young

thread of our “yellow bacillus No. 41,” developing from the symplastic stage, to understand
at once the meaning of Lachner’s “unexplainable” observation.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 175

Some drawings made by Messen (1898) of his diphtheroid “‘Syphilis-Bacillus,’’ which were.
reproduced as figure 217 on Plate XVIII (from original fig. 13), should be compared with those
of Morck and of Lutz given above. However, Niessen’s contributions, as well as those made
by Minden, are greatly impaired by the fact that the actual observations, discussed therein,
are enshrouded by many fantastic ideas and unfounded hypotheses upon ‘‘cytoblast,”’
“‘chthonoblast,” the transformation of bacteria into crystals, etc., which have necessarily
created a very suspicious attitude among the readers of these papers. But when we free the
findings recorded therein, from these accessories, it becomes obvious that Miinden has also
seen the bacterial symplasm in its amoeboid and in its encysted forms, as well as the trans-
formation of these hyalin masses into regenerative bodies of various shape, which ultimately
reproduce the normal bacteria.

Very meritorious investigations upon the symplastic stage of bacterial life have been made
by W. Winkler (1899); and it is hardly to be doubted that they would have attracted much
more attention if they had not been discussed rather summarily in one short paper, to which
merely drawings were added, which can not convey an adequate idea of what the author found
and of what is really to be seen in such experiments. Some of them are reproduced as figure
212 on Plate XVII. In liquid as well as on solid substrates, in mixed and in pure cultures, W.
Winkler always met with smaller and larger clumps of finely granulated or homogeneous plasma,
sometimes inclosed within a solid membrane, in other cases naked and showing amoeboid
movements like a plasmodium. Often these clumps assumed a ‘‘brain-like” structure, and a
small central body (‘‘Mittelpunktkérperchen”’) became visible within each lobe. One of our
photographs, reproduced as figure 231 on Plate XVIII (from Léhnis and Smith, 1916 a, original
fig. 12), may serve as an illustration. The transformation of these plasmatic masses into
bacteria was directly observed under the microscope and (1. c., p. 576) described as follows:

Die dem Rande zunichst liegenden Granula ordneten sich in Reihen; die einzelnen Kérner umgaben sich mit
Plasma, riickten auseinander, schwollen an, streckten sich und wurden nach und nach zu Bakterien.

Two of our photographs, reproduced as figures 215 and 216 on Plate XVIII (from Léhnis
and Smith, 1916 a, original fig. 32: Streptoc. lactis, and 1916 6, original fig. 56: Microc.
candicans) may be compared in this respect.

Sometimes the first cells formed showed very irregular outlines; they were triangular,
spindle-shaped, oval, cubic, or more or less irregularly curved. Already in 1880 Neelsen has
pictured such an occurrence in the drawing reproduced as figure 85 on Plate S (from original
fig. 10 on Pl. XT).

. Clumps of symplasm of different size were frequently seen by Winkler in colonies,
especially in those imbedded within the substrate, and here the symplasm was often inclined to
send out thread-like protrusions, which the author called “‘filidia.”” These filidia may produce
more or less normal bacteria, either by contraction or by segmentation, as shown in the right
part of figure 212 on Plate XVII (reproduced from W. Wankler’s original fig. 41), or by
furnishing the material for their formation within the more or less solid membrane of the filidium,
as illustrated by the two sketches in the upper center of figure 212 (reproduced from original
figs. 34 and 35). Sometimes large round bodies, called ‘“‘macrospores,’”’ were seen to grow in
the symplasm, or within broad filidia, which acted as “sporangia’”’; the drawings in the lower
left part and in the center of figure 212 (reproduced from original figs. 48 and 50) picture these
two possibilities. It is beyond doubt that these ‘‘macrospores” of W. Winkler are identical
with our large round regenerative bodies, whose frequent upgrowth from the symplasm will
have to be discussed later.

As far as I know, only one author, Riz¢ka (1903), has paid attention to and has continued
Winkler’s work. Unfortunately, he restricted his observations to the study of mixed infusions
and buried his data in a paper ‘Ueber die biologische Bedeutung der férbbaren Kérnchen des
Bakterieninhaltes.’’ He followed continuously the dissolution of the bacteria cells and their
reconstruction from the plasmatic masses, which he calls “zoogloea,” and concludes:

Es kann als gesichert gelten, dass die Zoogloea ein Syncytium darstellt, in welchem die beiden Komponenten,
die geformten Elemente und das strukturlose Plasma die Fahigkeit besitzen, in einander tiberzugehen.

160266—21——12
176 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

But there still remain some more recent observations, all made quite independently, which
also represent important contributions to our subject. s

When N. K. Schultz (1901 a) made her investigations upon the longevity of the gonidia of
B. pestis and their participation in the life cycle of this organism, she noticed that in old cul-
tures minute granules were present, imbedded in weakly staining flakes, which after bemg
transferred to fresh substrates, first produced agglomerations of globular bodies, which soon
after grew up to normal rods; but 14 days later the pale flakes with their minute granules
appeared again and replaced the vegetative cells. The figs. 218-221, reproduced on Plate
XVIII (from original figs. VII-IX and XII), illustrate these changes very clearly. However,
the Russian bacteriologist was not aware that she was facing the symplastic stage of the
plague bacillus. The flakes inclosing the granules were thought to be “le précipité floconneux
du bouillon,”’ and of course, especially in the several years old cultures, such an occurrence
had been quite probable. But the very pronounced similarity between the old and the new
flakes proves that this was not the case. The artist (Jvanoff), who made the pictures and merely
copied what he saw, without being blinded by any prejudice, showed very clearly that. the new
flake (fig. 221) is formed by the residues of the agglutinated cells, which had become unstain-
able. Moreover, it will suffice to compare these pictures with others reproduced on our plates
to feel convinced that also in this case it was the symplasm, not some dead precipitate, which
made such luxuriant upgrowth possible.

The actinomyces-like buds which Ligniéres and Spitz (1902) found growing out of the
hyalin homogeneous mass, which formed the grains in the pus of cases of actinobacillosis, are
very similar to those of Metchnikoff’s Pasteuria ramosa, as may be seen from the drawings of
the first-named authors, reproduced as figure 213 on Plate XVII. It is true that Metchmkoff
did not pay any special attention to the fact that the characteristic clubs of his Pasteuria were
produced by flakes of hyaline symplasm, but a photograph made by Rouz, which was added
to Metchnikoff’s (1888 6) drawings, and which is reproduced as figure 222 on Plate XVIII,
proves this point.

A bacillus, evidently related to B. pneumoniae, if not identical with it, was studied by
Jehle (1902) with the result that hardly any normal cells were seen in 5-8 days’ old cultures,
but numerous crumbly pale masses and flakes, which after being transferred to new substrates
promptly reproduced polymorphous cells.

With regard to leprosy Pernet (1902) reached the conclusion that the “ gloeal masses,’’ which
he, like G. A. Hansen (1880-1882), Babes (1883), and Lutz (1886), found within the tissue cells,
are not products of degeneration, but ‘‘a phase in the parasite’s life history, a resting stage,
during which it prepares for further proliferation,” and he pointed out that clinical observa-
tions are in good agreement with this view.

Fuhrmann’s (1906-1908) results with Pseudomonas cerevisiae are very similar to those re-
corded by Schultz with B. pestis. Here again the “detritus” containing minute granules,
showed itself to be able to reproduce new vegetative cells after long periods of rest. A good
photograph, published in the second paper (1908, original fig. 5 on Pl. I) and reproduced as
figure 238 on Plate XIX, exhibits the typical appearance of the symplasm with its regenerative
units, regenerative bodies and new rods.

Some very important contributions to our subject have been made by Rosenback (1909),
when he studied his polymorphous Erysipelothrix (Bac. erysipeloides). The four photographs
reproduced as figures 239-242 on Plate XIX (from original figs. XIV, 6; XII, 2; XIII, 2 and
XII, 1) are of special interest. The rods, whose normal appearance was shown in figures 26
and 27 on Plate III, agglomerate and enter into granular dissolution (fig. 239); the end prod-
ucts are visible in figure 240. Occasionally these granular clumps may reproduce fine threads,
as shown in figure 241, or even a distinctly mycelial growth may spring forth from them, like
that in Figure 29 on Plate III. Sometimes, however, the dissolution of the rods ends with the
formation of completely homogeneous masses, which may surround themselves with rather
solid membranes, thus forming large globules, as shown in 270-fold magnification in figure 242.
This encystment of the symplasm is of gréat interest when compared with those old reports,
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. LTT

made by Letzerich and other early bacteriologists upon the production of large cysts by typical
bacteria closely resembling the cysts of certain Myxobacteriaceae.

The symplastic stage of the tubercle bacillus, so well studied by Klebs (1883) and by
Malassez and Vignal (1883-84), but so flatly rejected by R. Koch (1884), was once more dis-
covered and fairly thoroughly investigated by Maher (1910-1913). He became aware that it
is especially due to the ability of this nonacid-fast ‘‘matrix”’ to reproduce many different forms
of various staining reactions why old cultures of tubercle bacilli are so much inclined to furnish
cells of very variable size, shape, and pigmentation. According to our own experience the
following statement, made in Maher’s second paper, allows a wide generalization:

From this matrix it is possible to grow coccal and bacillary forms that vary in morphology, in acid resistance, and

in chromogenic power, according to the variance in the age and vegetative energy of the tubercle bacilli that composed
the matrix, and according to the variance in the technique to which the matrix is subjected.

At least partially correct were also the findings of Herzog (1913) concerning the slimy dis-
solution of gonococci, or of whole colonies of these organisms, into a yellowish, transparent
mass, which at first still contained some minute globular bodies, later, however, became either
entirely homogeneous—in this case its was declared to be dead—or transformed itself into oval
or globular “giant forms.”

That Meirowsky (1914 6) in his investigations upon the development of the bacteria fre-
quently met with the symplastic stage of various bacilli, spirilla, and spirochaets is clearly
demonstrated by a number of the pictures accompanying his paper. For instance, the
sketches of Spirilum rubrum, reproduced as figure 86 on Plate S (from figs. 4 and 5 on Pl.
V 6), and some of the illustrations of ‘‘Doldenbildung,” shown on Plate XIV, leave no doubt
upon this fact. But Meirowsky himself thought these “umbels” were closely connected with
the “bud formation,” caused by the gonidia leaving the cells, and he stated at first (p. 9)
explicitly:

Die Dolden sitzen stets am Ende des Bazillus und stellen eine feine wolkige Masse dar, die aus zahlreichen sehr
kleinen, stark lichtbrechenden Kérperchen besteht.

Later (p. 16), however, he admitted that the ‘“umbels,’’ made up of small granules
embedded in a zoogloea-like matrix, were often found without being attached to a rod, and
that he also saw all intermediate stages between these masses and new vegetative cells. The
size of such agglomerations exceeds that of a bacillus or of a spirillum frequently so manifold,
that it would be quite inconceivable how they might develop from one single cell; the opposite
process is, no doubt, much more probable.

Some very characteristic photographs, published by Almquist (1916-1917), demonstrating
the development of new cells from what he calls bacterial “plasmodia,”’ have been repro-
duced as figures 243-247 on Plate XIX (from original figs. 12 and 22, 1916, and Nos. 5, 6, and
12, 1917). Figure 243 presents the upgrowth of B. typhi; besides fairly normal rods, large
yeast-like forms and peculiarly slimy threads are visible, “in statu nascendi.” Figure 244
shows the round regenerative bodies of V. cholerae being formed within a flake of symplasm.
This picture should be compared with figure 238 (symplasm of Pseudomonas cerevisiae Fuhr-
mann). Figure 245 was made from a single cell culture of B. diphtheriae; together with figure
246, B. acidi propionici c, it exemplifies the possibility of very irregular forms consolidating
themselves out of large and small flakes of symplasm. Figure 247 pictures the symplasm of
Almquist’s Bact. antityphosum (cf. p. 147); it may be compared with figure 216 on Plate XVIII,
presenting Micr. candicans in the same phase of its life cycle, and also with figure 223 on the
same plate, which is a reproduction of a picture made by Kellerman and Scales (1916) of the
first beginning of new growth of B. coli from the symplasm. Another photograph of the same
authors, reproduced as figure 224 on Plate XVIII, illustrates once more the concentration of
the pale granular mass first into an unsharp slimy thread, like those photographed by Almquist
of B. typhi (fig. 248), and the progressive consolidation of parts of this thread into normal,
deeply staining rods. Two other coli pictures, made by Kellerman and Scales (1916) and
reproduced as figures 44 and 45 on Plate IV as an illustration of the pleomorphism of this
178 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

species, will now also be better understood as to their true meaning. They show clumps of
the large yeast-like cells, so frequently originating in and from the symplasm, and another
example of such broad threads, to be compared with W. Winkler’s “filidia.”’

The summary of the results of our own preliminary studies upon the symplastic stage of
representatives of various groups of bacteria has been given on page 166. It may be added here
that by some special experiments (Léhnis and Smith, 1916, a, p. 679) the fact has been definitely
established that not only in old, but also in very young, cultures the formation of symplasm
may take place in accordance with the tendency of the bacteria to live alternately in an organ-
ized and in an amorphous stage. Some of our photographs, illustrating the various steps leading
from cells to symplasm and back to cells of the same or of a different type, have been reproduced
as figures 225-233 on Plate XVIII, figures 250-261 on Plate XX and figures 262-267 on Plate
XXI. As a detailed discussion of these pictures will better find its place on the following
pages, it may suffice to mention here that figures 225-231 on Plate XVIII and figures 250-258
on Plate XX were all made from Azotobacter. Figure 232 (Pl. XVIII) and figures 259-263
(Pls. XX-XXD) illustrate different phases in the life cycle of B. Auorescens. A flake of sym-
plasm of B. subtilis is shown in figure 233 (Pl. XVIII) for comparison with figure 230. The
upgrowth of Sarcina flava is demonstrated by figure 264 (Pl. XXI). Figure 266 shows pecu-
liarly pointed cells of B. malabarensis, whose large gonidia break out before melting together
to the symplasm. A thread of the “yellow bacillus No. 41” growing up from the ayiplesi is
visible in figure 267.

It needs hardly to be emphasized that these photographs furnish fairly accurate ifinstrations
to several of the early descriptions, mentioned above, or somewhat superior counterparts to
the rather primitive drawings, made by the earlier authors. It should be pointed out, however,
especially for those readers who have not yet seen such preparates under the microscope, that
even the best photographs can only convey a very incomplete impression of the actual appear-
ance of this type of growth. Large flakes or clumps of symplasm, which make the most inter-
esting object under the microscope, can not be photographed at all. And even so many parts
of the smaller agglomerations are necessarily out of focus, that the flat photographic picture
is always very inferior to the bright and plastic appearance of the symplasm under the
microscope.

But in one direction these photographs will hardly leave any doubt, viz., that a thorough
study of this particular phase of the life cycles of the bacteria will supply us with very valuable
and much needed information upon the occurrence of different cell forms, the real significance
of so-called involution forms, the appearance and disappearance of variations and mutations,
and many other data, which could be obtained with the customary methods of investigation.

Cultures showing “autolysis” or “granular decomposition” of their cells are usually discarded
as speedily as, or even more lightly than those producing “involution forms,’ much to the
disadvantage of bacteriological science. From the descriptions furnished by authors like
Ermengem (1885, p. 20), E. Klein (1885, p. 111), Gamaleia (1900 pp. 27, 154), Kruse (1910,
p- 20), Retiger, Berman and Sturges (1916), and many others, it may be safely concluded that
more thorough tests would have demonstrated, that also in these cases “degeneration”? would
have been supplanted by ‘ “regeneration,” as soon as suitable conditions for the revival of cell
life had been established. It is, of course, self evident that permanently adverse conditions
must end all life, but “autolysis” of the bacteria i is by no means equivalent to death; in fact,
it enables the living matter to readjust itself and to reenter cell life in a better adapted modi.
fication. Every experienced bacteriologist knows that careful testing will bring many cul-
tures back to good growth, which on superficial examination would be declared dead.
That “autolysis” may be followed by renewed development has been observed, e. g., very
clearly by Malfitano (1900) in his experiments upon “La bacteriolyse de la bactéridie char-
bonneuse”’ in full agreement with those made by Fokker, which were discussed on p.171. A
few hours after the bacilli had been transferred into distilled water only an “amas de débris
informes’”’ remained, but the author adds:

Le phénoméne est trés net, bien que, dans ces conditions de nouvelles bactéridies puissent se former.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 179

Kruse (1910 p. 21) also says:

Unter Umstinden tritt in denselben Mitteln (i. e. in old cultures, serum, etc.) Bakteriolyse und Wachstum neben
oder nacheinander auf, indem naémlich die durch Selbstverdauung schon zugrunde gegangenen Bakterien den iibrig
gebliebenen den Nahrboden . . . verbessern . . . oder . . . im Serum die Alexine neutralisieren.

A more complete knowledge of the facts recorded by the earlier authors would probably
have replaced the hypothesis in the last part of this statement by a more correct reference to
the bacterial symplasm.

That the bacteria are to be frequently found in their amorphous stage in vivo as in vitro
has been indicated by the quotations taken from the publications of Osler and Schéfer (1873),
Letzerich (1873-1876), Weigert (1875), Klebs (1878-1887), R. Koch (1878), G. A. Hansen
(1880-1882), Albrecht (1881), Fokker (1881-1882), Babes (1883), Malassez and Vignal (1883-
1884), Cornil and Alvarez (1885), Eberth (1885), Amrusch (1886), Zatz (1886), Nocard (1889),
Crookshank (1896), Lachner-Sandoval (1898), Ligniéres and Spitz (1902), and Pernet (1902).
More recently renewed attention was given to this point. A photograph made by £. de Negri
(1916) of the amorphous masses found in granulom tissue, caused by a Corynebacterium, has been
reproduced as figure 237 on PlateX VIII. What Mallory and Medlar (1916) described and pictured

“clumps of bacilli,” taken from the tonsils in cases of scarlet fever, show very clearly their
symplastic status, when we compare the photograph reproduced as figure 248 on Plate XIX
(from original fig. 23 on Pl. XX), for instance, with figures 216, 246, 247 and 254 on Plates
XVITI-XX. The same holds true in regard to a photograph made by Mellon (1917, original
fig 5, Pl. 2) of ‘‘a colony of diphtheroid bacilli lying in the interstitial tissue of the lung,” which
was reproduced as figure 249 on Plate XIX. In eases of poliomyelitis, Rosenow, Towne and
Wheeler (1916) found often in tonsillar abscesses ‘a peculiar gelatinous opalescent material
from which the peculiar Streptococcus was isolated in large numbers.”’ And when Wade (1918)
studied an unclassified chronic Philippine ulcer, he obtained results which are in complete
agreement with the numerous findings mentioned above, although according to the author’s
own judgment his ‘‘hypothesis departs radically from orthodox views.” He says:

‘Essentially it appears as if the differentiated demonstrable parasitic element in a mycotic
lesion may give rise to a derivative substance, morphologically not differentiated and perhaps
even quite amorphous, that, unlike the known toxins, soluble or insoluble, is by itself more or
less viable and may be capable, to a certain extent at least and under certain conditions, of
persisting as such and even of increasing or growing.”

This ‘substance or body” was named by Wade ‘‘cryptoplasm,” while for the large round
bodies, which were again seen to develop from it, the term ‘‘cryptococcus’”’ was introduced.

It is to be expected with certainty that more confirmative results will be secured as soon
as the hyalin and amyloid substances of the body will be made the object of new studies along
these lines. Descriptions published by Lubarsch (1903) leave no doubt that appearance and
staining reactions are very much alike in both cases, and though he does not refer to a partici-
pation of microorganisms in the formation of these substances, this has been done already in
1878 by Klebs and in 1885 by Ziegler, who pointed out (1885, p. 285) that epithelial cells, when
attacked by bacteria, swell up and liquefy or degenerate into homogeneous lumps or granular
detritus. To what degree the cells of the host on the one side, and those of the parasite on the
other, are to be made responsible for the outcome apparently nobody has ever tried to decide.
But that the living plasmatic substances of the higher organisms, especially those building up
the animal tissue, are equally able to enter and to pass through such a ‘‘symplastic”’ stage as the
bacteria do, has been pointed out by Robin about half a century ago, according to Béchamp
(1883, p. 543), in the following statements:

En anatomie génerale on appelle blastéme ou cytoblastéme des espéces de substances amorphes liquides ou demili-
quides, soit épanchées entre les éléments anatomiques préexistantes dans un tissu ou & sa surface, soit interposées
entre des éléments qui naissent 4 leur dépens au fur et 4 mesure de leur production au sein ou a Ja surface d’un tissu.

Dans le blast€me prennent ou peuvent prendre naissance des éléments anatomiques, normaux ou morbides (granu-
lations moléculaires, fibres, tubes, cellules, etc.).

Le blastéme est une matiére: complétement homogéne, amorphe, sans structure.
180 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

And quite recently Burrows and Neymann (1917) reported upon their ‘‘Studies on the
Metabolism of Cells in vitro’’ as follows:

Growth associated with mitotic cell division is seen only in (tissue) cells which have become passively placed at
the interspace between certain insoluble substances and the medium. These substances are liberated from the cells
when they are removed from their normal habitat to an oxygen containing plasma or salt solution. The substances are
almost transparent, their refraction is not very different from that of the original medium and they accumulate at the
surface of the medium to form a membrane. They are liberated in large amounts from a tissue fragment rich in cells.

A cell brought in contact with the surface of this transparent substance adheres to it and flattens over its surface.
Such a cell grows and divides by mitosis when oxygen is present and the waste products remain below a certain concen-
tration.

A single isolated cell when placed in a drop of fresh plasma does not grow and divide.

It is hardly to be-doubted that a close study of the symplastic stage of bacterial life will
yield information of considerable value for general biology. The formation of new cells
directly out of the amorphous plasma can perhaps in no case be better observed than with the
bacteria, although lower fungi, as well as protozoa, may also furnish excellent objects for such
investigations, as had been noticed by Pineau and Perty a long time ago, according to their
studies discussed on pages 167-168.

When referring to Thazter’s Myxobacteriaceae (on p. 174) it has been pointed out that
the more complete knowledge of the life cycles of the bacteria undoubtedly establishes much
closer connections between these two groups of organisms, as well as with the Myxomycetes.
W. Winkler (1889) concluded from his observations that the bacteria are related to Myxomy-
cetes as well as to Amoebae and that the Myxobacteria are true bacteria. In 1866, Woronin
placed the root-nodule organism (B. radicigola) in the neighborhood of Plasmodiophora, as did
later Kny (1879) and Prilliewx (1879). Schroeter (1886, p. 134) and Atkinson (1898) class it
among the Myxomycetes; Zukal (1897), Stefan (1906), Jorgensen (1911), and. Pinoy (1913)
emphasized its relationship to the Myxobaéteriaceae. It is not to be disputed that the forma-
tion of spores and cysts by the Myxobacteria, as it has been described in the publications of
Thazter (1892-1904), Zukal (1897), Baur (1905), Quehl (1906), Kruyff (1908), Vahle (1909),
Jahn (1909-1911), and A. L. Smith (1913), at the first moment seems to have very little in com-
mon with the general behavior of the bacteria. But it must never be overlooked that the
experimental conditions under which the bacteria are nearly always kept and studied also do
not allow the Myxobacteria to develop in their characteristic form, as has been shown, for
instance, by Quehl and Vahle. The small Bac. helixoides (0.6 X 2-4u,) isolated by Muto (1904)
from saliva, which grew on the plate in colonies made up of rods at the edge and oval forms in
the center and sent out slimy protrusions creeping forward like snails, undoubtedly has been a
maltreated Myxobacterium. Jahn was certainly right when he wrote:

” Die den Myxobakterien eigentiimliche Form der Coloniebildung ist auch bei den Bakterien nicht so ungewohn-
lich, und sie wiirde auch den Bakteriologen noch vertrauter sein, wenn sie mehr gewohnt waren, die Bakterien unter
ihren nattirlichen Lebensbedingungen zu beobachten.

A bacillus isolated by him showed in a nutrient solution a behavior very similar to that
of Myxobacteria: P

Verschlechtern sich die Existenzbedingungen, so vereinigen sich die Bacillen zu einem dichten Haufen und
tanzen lebhaft wie ein Miickenschwarm durcheinander. Dann kommen sie allmighlich zur Ruhe, sondern Schleim ab
und bilden eine characteristisché Colonie aus rundlichen, verkiirzten Individuen.

When studying the “zoogloea” stage of the bacteria in 1877 R. Koch also reported (p. 415)
to have seen the cells within the zoogloea agglomerate into groups of 10-12 individuals and to
enter spore formation. Zikes (1916) observed on dry gelatin a growth of bacteria, as well as of
yeasts, which was microscopically as well as macroscopically similar to that of certain myxo-
bacteria.

That the spore formation of the Myxobacteriaceae is fundamentally identical with the
production of arthrospores and microcysts of the bacteria has been discussed on page 123.
The same holds true in regard to the agglomeration of the vegetative cells and their slime
production preceding the encysted stage of the myxobacteria, and the symplastic stage of the
bacteria. As figure 87 on Plate S a drawing made by Baur (1905) of Myzxococcus ruber in
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 181

the beginning of fructification has been reproduced (from original fig. 3 on p. 113). It should
be compared with one of Rosenbach’s photographs, reproduced as figure 239 on Plate XIX,
showing B..erysipeloides before entering the symplastic stage. The more or less complete
dissolution of the vegetative, as well as of the reproductive, cells of the bacteria forming the
symplasm obviously constitutes the most pronounced difference although perhaps a similar
behavior may still be discovered among myxobacteria, as well.

The encystment of these agglomerations is also by no means absent among the bacteria,
The cyst of Polyangium, drawn by Zukal (1897, original fig. 8) and reproduced as figure 88
on Plate S, makes a counterpart to some of the old sketches of bacteria found in cases of
diphtheria by Leézerich (1876, reproduced as figs. 68 and 80 on Pls. P and R) and of a
leprosy zoogloea, published by G. A. Hansen (1880, original fig. 10). The ‘“macroplasts” of
Bact. rubescens, studied by Lankester (1876) and reproduced as figure 209 on Plate XVI, the
“zoogloea” form of Nitrosomonas, as photographed by Winogradsky (1892) and reproduced
as figure 270 on Plate XXI, the Asococcus of Billroth (1874) and F. Cohn (1875), the Clath-
rocystis of Cohn (1875), the Ascobacteria of Van Tieghem (1880), the Micrococcus botryogenus
of Rabe (1886), the Amoebobacter and Thiocystis of Winogradsky (1888, pp. 60-78), the Asco-
bacterium of Babes (1895), the Ascobacillus of Moreno (1901), as well as the “bacteriocysts”’
of Miiller-Thurgau (1908), which are to be discussed on the following pages, indicate all very
clearly that large encysted agglomerations of cells are not at all uncommon among the bac-
teria. Even the homogeneous symplasm may encapsulate itself, as was shown by Rosenbach
(1909) in the photograph reproduced as figure 242 on Plate XTX. That it will mostly depend
on the environmental conditions whether or not such macrocysts will be formed by the bac-
teria, may be safely assumed, though hardly any experimental results are available at present
in this direction. Pinoy (1913) pointed out that symbiotic (or antagonistic?) effects may act
as stimulants in this direction; and, indeed, we should never forget that our pure cultures
on standardized media will often fail to exhibit the characters shown by the mixed microflora
under natural conditions.

That the bacterial symplasm is able to reproduce either round regenerative bodies or
directly vegetative cells is another parallelism to the behavior of the myxobacteria, which
pass their resting stage either as spores or as vegetative cells. But with the bacteria both
modes of regeneration occur apparently with every species, while in the latter case a more
distinct differentiation has taken place.

The occurrence of bacterial plasmodia, showing amoeboid movements, indicates relations
existing between bacteria and Myxomycetes, as was pointed out by W. Winkler (1899).
De Bary (1884, p. 477) thought the Myxomycetes should be separated from the fungi and
should be called, together with the Acrasiae, Mycetozoa. However, as Henneberg (1904) has
demonstrated more recently, amoeboid stages may be also found with true fungi (yeasts);
on the other hand, amoeboid movements of the bacterial symplasm do not seem to be a
general character, and more thorough studies very probably will reveal many analoga to the
bacterial symplastic stage among protozoa, as well as among fungi and algae. Therefore
the relations existing between bacteria and protozoa should not be too much enhanced on
account of this one point.

(b) FORMATION AND APPEARANCE OF THE SYMPLASM.

According to the observations made by Lankester (1876), W. Winkler (1899), Rosenbach
(1909), and Léhnis and Smith (1916 a and 6) the formation of the symplasm always takes place
in two phases: First, the bacteria agglomerate to smaller or larger clumps; second, the cells
dissolve more or less completely to a crumbly or slimy mass, assuming the shape of flakes,
irregular clumps, or regular spheroid bodies, and being sometimes endowed with amoeboid
motility.

Apparently all species of the bacteria and all kinds of vegetative cells, as well as of repro-
ductive organs, are able to enter the symplastic stage. It occurs both in old and in young
cultures, dependent on internal as well as on external conditions.
182 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

A so-called pseudo-agglutination has been frequently noticed in connection with aggluti-
nation tests; it has been discussed by Savage (1901), Escherich and Pfawndler (1903, p. 341),
Hiss and Zinsser (1914, p. 231), and others. Hewlett (1902, p. 143) found that “agglutinins”
were present in old broth cultures of typhoid bacilli; filtrates obtained therefrom powerfully
agglutinated the bacilli in fresh broth cultures. Nocht and Mayer (1907, pp. 18, 33) report
that spirochaets often agglomerate when kept for a longer time at low temperature or in serum,
that the agglomerated cells occasionally fuse together in the center, and that sometimes “ desag-
glomeration” was observed. That similar clumps of spirochaets may be also found in the
tissue is indicated by a photograph made by Flexner (1907) of ‘‘colony-like” masses present
in the skin of a macerated syphilitic foetus. McFarland (1916, p. 721) says referring to this
picture: ‘The dense aggregation of organisms may indicate agglutination.” There is no doubt,
however, that the agglomeration preceding the formation of the symplasm is essentially dif-
ferent from the typical agglutination, though both processes present a very similar appear-
ance. Agglutination is, according to the careful review published by Kohler (1902), promi-
nently a physico-chemical reaction, whereas the active gathering of the cells, when entering
the symplastic stage, makes it quite evident that in the first place a biological cause is active
in this. case, though external, physical as well as chemical, factors will naturally also exert
their influences, hastening or retarding the agglomeration of the cells.

The dissolution of the united cells leaves either the empty cell walls as “shadows,” but
still clearly visible, in the field, or they, too, are disintegrated and participate in the formation
of a more or less homogeneous symplasm, which often becomes entirely amorphous, hyalin,
and highly refractive. Our photographs reproduced as figures 225-231 on Plate XVIII (rom
Lohnis and Smith, 1916 a and 5) illustrate these various modes of formation of the symplasm,
which may all occur with the same species, as shown here with B. azotobacter. That all types
of vegetative cells, as well as of reproductive organs, are able to enter the symplastic stage is
also. demonstrated by these photographs.

Like the structure, the staining reaction of the sraileae can also show wide variations.
In such cases where darkly staining membranes, especially those of regenerative bodies, partici-
pate in the process generally deeply staining flaky masses are produced, whereas the voluminous
clumps, resulting especially from dissolved endospores or gonidia or from the agglomerated
contents of large vegetative cells, are usually entirely unstainable by aqueous dyes. Frequently
of course, large masses of symplasm are made up by both deeply staining and unstainable parts.
W. Winkler (1899) reported to have directly observed the transformation of crumbly dark
flakes into voluminous ‘‘plasmodia.’”’ Thus far I have not seen such changes; but the presence
of very thick clusters of the peculiar bright slimy masses in old cultures rich in symplasm makes
it, indeed, quite probable that special structural alterations may take place, which remain to
be studied. Letzerich (1876), too, has reported that he observed directly a growth of bacterial
“plasmodia”’ as such; the inner circle inserted within the large irregular symplasm in the lower
left. corner of the drawing reproduced as figure 80 on Plate R was explained by the author as
indicating the original size from which the upgrowth started. It is impossible to decide at
present whether or not this old report is to be accepted as correct.

The variations in the staining qualities of the symplasms make it easily to be understood
why, for instance, the tuberculosis ‘‘zoogloea’”’ was found by Malassez and Vignal (1884) to be
not acid-fast and either not at all or with difficulty to be stained with aqueous dyes, while
Amrusch (1886) recorded a slight acid-resistance. Metchnikoff (1888 a) saw the slimy amor-
‘phous masses assume a yellow color, when they were treated like the tubercle bacilli; and Maher
(1910-1913) found again his tubercle ‘‘matrix’’ to be either not or slightly acid-fast. According
to Lutz (1886) the ‘‘gloeal masses” of B. leprae are stainable with common aniline dyes, and
turn bluish-red when treated with Gram’s method, while they were seen by Pernet (1902) to vary
in their staining qualities. Analogous results concerning the ‘‘plasmodia”’ of B. radicicola have
been recorded by Hiltner (1898), and by Bastian (1905, p. 186) with his mixed cultures.

Iodine stains the symplasm either yellow or brown, like the hyalin and amyloid substances
in the animal tissue, but this reaction may again vary, as has been shown by Morck (1891) with
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 183

B. radicicola. Letzerich (1874) saw sometimes, but not regularly, a blue color develop in his
diphtheria ‘‘Plasmaballen’’ when they were treated with iodine and sulphuric acid.

Amoeboid movements of bacterial symplasms have been seen by Letzerich (1876), Dowdes-
well (1889), W. Winkler (1899), and Miinden (1907). Dowdeswell’s observations, which were
made with V. cholerae, have been refuted by Friedrich (1892), who asserts that they can not be
accepted as correct. The German author admits, however (p. 104), that he also saw the bacilli
dissolve into irregular flakes of plasma (‘‘unregelmissige Plasmafetzen’’). But that he did not
see a new growth of bacilli arising from these flakes was very probably due to faulty experiment-
ing. Internal, oscillating or gyrating movements have been noticed by Perty (1852), Amrusch
(1886), and by Frank (1890). I have also seen this kind of movement in hyalin symplasm,
whereas I have had no opportunity thus far to confirm the findings concerning amoeboid
motility.

That the voluminous agglomerations of symplasm are endowed with a comparatively high
resistance against the solving action of acids, alkali and hot water is a priori very probable.
The prominent réle which they have played in Bastian’s experiments on ‘‘abiogenesis”’ and
‘“heterogenesis’’ shows that they readily withstand long continued heating. Metchnikoff
(1888 a) discovered that the symplasm of B. tuberculosis is very resistant against acid and alcohol.
W. Winkler (1899) found out that by adding a 4-10 per cent solution of caustic alkali the sym-
plastic masses present im thick bacterial growth become clearly visible, because they are less
easily dissolved than the bacteria. And when I encountered for the first time what later turned
out to be the symplasm of B. azotobacter, I also tried in vain to dissolve these doubtful clumps
‘by treating them thoroughly with acid and with alkali. That they are sometimes hard to fix on
the slide, by heating as well as by alcohol, has been mentioned above. This behavior, together
with their heat resistance, will be considered in Chapter V, on account of their liability to cause
faulty results.

That the increased resistance also protects the life within the symplasm to a considerable
degree has been demonstrated by Fuhrmann (1908), who discovered that what he called
the ‘‘detritus” of Pseudomonas cerevisiae remained alive for several months in a 10 per cent NaCl
solution, wherein a young agar culture of the bacilli died within a very short time. If the sym-
plasm encysts itself, as was observed by Rosenbach (1909), its resistance will evidently be still
more increased, and at the same time its function as resting stage becomes quite obvious.

That this melting together of numerous cells and the ensuing thorough mixing of their
ingredients are of great importance in regard to a better adaptation of the living substance to the
environmental condition is especially clearly indicated by the very pronounced tendency of the
liberated gonidia to enter first the symplastic stage, before reproducing new cells. As was dis-
cussed in the foregoing chapter, the gonidia grow up to new cells very readily while still
connected with the parent cell, or if they have otherwise access to genuine plasma, like that
in dead or dying cells of fungi and algae. But in other cases free gonidia have been rarely
found to be inclined to act promptly as normal reproductive organs, when kept under ordinary
laboratory conditions, whose rather unnatural character is beyond dispute. Dissolution of the
‘ gonidia has been mentioned in the early publications of Perty (1852) and of Billroth
(1874), and the same fact has induced many other investigators to believe that also in this
case dissolution means death, and no further investigation was thought to be necessary.
F. Cohn (1870) as well as Zopf (1879) have noticed, however, that the microgonidia produced
by Crenothrix, instead of germinating to new threads, as the macrogonidia always did, first
agglomerated to a ‘‘zoogloea,’’ which after a period of mest gave new vegetative growth. "Ewart
(1878 a) recorded analagous results with the ‘‘spores’’ (gonidia) of “‘B. termo”’ (B. fluorescens) ;
only after having formed the “‘zoogloea”’ they were able to reproduce rods. And it is not improb-
able that the reason why R. Koch (1877), after having made his interesting first photograph of
the ‘lateral spores”’ (gonidia and regenerative bodies) of B. fluorescens, which was reproduced
as figure 109 on Plate X and discussed on page 92, never returned to this subject, has again been
this mistaken dissolution of the small round bodies. Anthrax gonidia displayed the same
184 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

deceptive behavior, when first studied by Fokker (1881-1882), and also their réle, as well as that
of the gonidia of other parasitic bacteria, therefore, has often been overlooked.

Very conspicuous is the formation of symplasm when large gonidangia, containing numer-
ous gonidia, break up, forming a‘small flake or clump of hyalin substance. Dowdeswell (1889-
1890) has been probably the first who did not follow the generally observed easy rule to discard
quickly all such “dead involution forms.” He studied carefully the flocculent ‘‘débris,”
formed by the gonidangia of V. cholerae, and saw new curved bacilli develop from the minute
granules inclosed within the plasma flakes, which also sometimes exhibited amoeboid movements.
Mafucci (1892) obtained similar results with the club-shaped gonidangia of B. tuberculosis,
though. he only assumed that the deeply staining “granules” within the granular masses might
act as “spores,” without being in the possession of an accurate experimental basis, on which
to found his belief. Meyerhof (1898), who worked with diphtheria bacilli, was more radical
when he reported that the clubs within 8 days disintegrated—

in kriimelige Produkte, die als abgestorben angesehen werden mussten.

No explanation was given why the author felt compelled to draw this conclusion without
first studying the subject more closely. The latter was done by Fuhrmann (1906-1908), who
was careful enough not to discard lightly what he also considered to be merely “detritus” of
the club-shaped gonidangia of Pseudomonas cerevisiae and related organisms. Therefore, he
secured the characteristic upgrowth from it, as was shown in the photograph reproduced as
figure 238 on Plate XTX. Similar results have been recorded by Rédzitka (1908) with B.
anthracis and by Herzog (1913) with Microc. gonorrhoeae.

Especially noteworthy is the symplastic stage with the filterable gonidia. Authors who
have studied these smallest bodies more thoroughly have always encountered pale, slimy
masses of variable staining reaction, wherein the minute granules often showed a comparatively
conspicuous upgrowth. The small globular or dumb-bell-shaped cell inclusions, observed by
Bernhardt (1911) in scarlet fever, were seen to be embedded within the cells in a plasmatic
stratum of a peculiar “cloudy” structure, which took the stain sometimes weakly, in other
cases, however, quite readily. On the other hand Hoefer (1911), who studied the same subject
at the same time in the same laboratory, did not see the small bodies found by his colleague,
but only homogeneous, hyalin clumps of different outlines, which later contained numerous
smallest granules. He is inclined to believe that these formations represent another phase in
the development of the cell inclusions, and both descriptions taken together fit indeed exactly
the general character of the symplastic stage of the bacteria. Furthermore, the photograph
made by Mallory and Medlar (1916) of the agglomerations of their B. scarlatinae, taken
from the tonsils, which was reproduced as figure 248 on Plate XIX, also leaves no doubt about
its being a typical picture of a clump of symplasm, just beginning to reproduce larger cells.
The trachoma bodies were found by Addario (1912) to be equally embedded in a homogeneous
substance, and the coccoid bodies, isolated by Noguchi and Cohen (1913) from the trachoma
inclusions, again formed clumps of finely granular structure. The “clusters of very minute
organisms of quite indeterminable morphology,” discovered by Hort (1916 a) in meningitis |
filtrate, apparently have been very similar to the small clumps of symplasm, which were found
in our cultures of the filterable gonidia of B. fluorescens, and of which one is shown in the dark field
in figure 234 on Plate XVIII (reproduced from our first preliminary paper, original fig. 42).
The filterable organism isolated by Healy and Gott (1916) from the tissues of cholera hogs is
said to occur also ‘“‘in very small clumps, showing many minute bright points”; the plasmatio
mass was once more found to stain either deeply throughout or not at all. A paper by Foster
(1917) upon the etiology of common colds is accompanied by a photograph (original fig. 3,
Pl. 9) showing clearly the symplasm formed by the filterable gonidia just beginning to repro-
duce small globoid bodies. Very similar pictures from the poliomyelitis germ have been pub-
lished by Amoss (1917, pls. 44 and 46), and that excellent photograph of the symplastic stage
of the filterable Bact. antityphosum of Almquist (1917), reproduced as figure 247 on Plate XIX
and discussed on p. 177, deserves to be studied once more in this connection.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 185

(c) REPRODUCTION OF REGENERATIVE BODIES AND OF VEGETATIVE CELLS FROM THE SYMPLASM.

The highly interesting reconstruction of cells from the symplastic stage may proceed,
like the formation of the endospores within the cells, along two slightly different lines. Either
the agglomeration of the granular material and its consolidation to a new full-sized cell are
most conspicuous, or at first one center of reorganization becomes clearly visible, from which
the new cell evolves in a comparatively slow development. Undoubtedly the first mode of
action is the more striking one, and it has, therefore, prominently attracted the attention of
such early investigators as Pineau, Perty, Bastian and Béchamp, whose reports and drawings
have been quoted and discussed on the foregoing pages (pp. 167-172). But the second possi-
bility has been also admirably described as early as in 1876 by Lankester, whose very interest-
ing description and drawings have been reprinted on page 170 and on Plate XVI (fig. 209).
W. Winkler (1899), too, saw the separate centers of organization surround themselves with
plasma and in this way grow up to new cells (see the literal quotation on p. 175).

According to my own observations both types of reproduction may be found with the
same species. Stained preparates show invariably as the first step the reappearance of minute
deeply staining granules, which are especially conspicuous in the hyalin type of symplasm;
wé have named them “regenerative units’ (Léhnis and Smith, 1916 a, p. 680). When large
cells are formed directly, several of these units will gather and surround themselves with the
uniting membrane; in other cases one unit increases slowly in size and builds up the new cell.
It is quite obvious that these regenerative units are nuclear in character, and the same diversity
which is noticeable in the adult bacteria cell (see p. 109) becomes also evident in its embryonic
stage. This parallelism extends still further. In those cases where only one deeply stained
granule is visible within each new cell, this either stays there as such, or it spreads uniformly
throughout its inclosure.

Our photographs reproduced as figures 253-264 on Plates XX and XXI illustrate these
different modes of reconstruction, as far as this is possible by such pictures, which are neces-
sarily more or less inferior to the real appearance of the objects under the microscope. Appear-
ance, agglomeration, and upgrowth of the minute regenerative units are well discernible, and
a comparison of figures 256 and 262 with most of the others demonstrates clearly how the cells
sometimes instead of emerging slowly as separate units from and within the amorphous sym-
plasm are directly and sharply cut out from their matrix, which in this way divides itself, much
as certain protozoa do, into ‘‘schizonts” which only after separation assume again their typical
cell form. As figure 256 demonstrates, new sporulating rods may come into existence as such,
although at first of a queer tadpole shape, they display prominently their embryonic endo-
spores, which usually reach their full development together with the newly born cells themselves.
Another noteworthy peculiarity is the development of entirely pale bodies, which may multiply
as such for a long time, always refusing to take the aqueous stain. They are visible in figures
259, 262, and 263; in the last case the transformation of the deeply stained hairy symplasm
into these unstainable bodies is very conspicuous, while it is a little more difficult to discern
the pale ovals within the stained cloudy agglomerations of figure 262. All three photographs
were made from B. fluorescens, but other species exhibit sometimes an analogous behavior.
Occasionally a deeply staining granule is visible within an irregular pale cell, or within each
segment of a pale sheath of varying length, but, like in the case mentioned above, later dis-
perses itself, giving the whole new form the normal staining reaction and undoubtedly other-
wise taking part in the reconstruction of the cell. Figures 70 on Plate VI and 235 on Plate
XVIII present such dotted pale threads of B. azotobacter, partially dissolved in the latter case
by heating, and on page 177 we have already discussed the similar consolidation of slimy threads
formed by B. coli and typhi. The photographs made by Toenmessen (1913) of B. pneumoniae,
which were reproduced as figures 37-42 on Plate IV and which were explained by the author
as being ‘‘mutants,” make another interesting object for comparative studies. They all illus-
trate very clearly our preceding discussion of the reconstruction of normal cells from the sym-
plasm. A photograph of B.' colt published by Kellerman and Scales (1916), which was repro-
186 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

duced as figure 44 on Plate IV, furnishes another good example of new upgrowth from the
symplastic stage. And a comparison of figures 99 on Plate [IX (Corynebacterium, FE. de Negri,
1916), 240 on Plate XIX (Erysipelothrix Rosenbach, 1909), 223 on Plate XVIII (B. coli, Kel-
lerman and Scales, 1916), 247 on Plate XIX (B. antityphosum, Almquist, 1917), and 255 on
Plate XX (B. Azotobacter, Lohnis and Smith, 1916 8), will indicate how a simultaneous upgrowth
of the regenerative units to uniformly coccoid bodies may give rise to new forms, which in
no way fit the time-honored definitions of the old form species.

This holds equally true in regard to the round regenerative bodies, whose vexatiously
uniform appearance with the different species has been already discussed on page 134. If they
grow up in pure culture from the symplasm, pictures like those reproduced as figures 156 on
Plate XII; 219 on Plate XVIII, 244 on Plate XIX, 257 and 258 on Plate XX become visible
under the microscope, leaving the investigator much in doubt as to what species they may belong.

Often, but not always, the new cells mature first at the outside of the hyalin agglomera-
tions, and this may lead to a peculiar starlike arrangement, as has been photographed by
M. Jones (1905) in the picture of a motile water organism (probably related to B. radiobacter),
which was Tepradnced as figure 274 on Plate XXII. The author believes that this rosette
formation is ‘‘apparently effected by a uniform grouping of the descendants of a single cell and
is in no sense an agglutination phenomenon.” But a study of the photograph in the light of
my own observations upon the star formation of B. radiobacter make the explanation given at
first more probable to me. Beijerinck and Delden (1902) found also the stars of the last-named
species often, though not always, embedded in a thick slime. And Winogradsky (1888, p. 38)
described quite similar “‘radiate colonies” of his Thiothrix, wherein the young threads are
living together in clusters, developing from the same slimy basis. The analogous structure of
the clusters of Actinomyces has been discussed on page 174.

Cladothrix, Crenothrix, and other trichobacteria have been so often described as autering
a ‘‘zoogloes,” stage and as growing out of slimy flakes, which frequently serve as a basis for
the threads, by which they are anchored to the ground, that such pictures like figure 267 on
Plate XXTI, showing a young | thread of our spore-forming ‘‘yellow bacillus No. 41” developing
from the symplasm, must evoke considerable doubt whether the ‘‘common” bacteria may not
also in this respect come much closer to the ‘‘higher” bacteria, if they are only allowed to
exhibit their properties more fully than is usually granted to them. Our photograph shows
that at the side of the broad basis of the thread some regenerative units are still growing up
within the flake, and exactly similar round bodies have been also photographed by Molisch
(1910, original fig. 8, Pl. IT) in what he calls the ““Haftscheibe” of iron bacteria. That Meirowsky
(1914 6) obtained analogous results with other bacteria and that a description given by Lachner-
Sandoval (1898) of young actinomyces threads fits our picture completely has been mentioned .
on pages 174 and177. It may be added that the threads of Leptothriz within the human mouth
exhibit the same morphological character, according to Robin (1853), Miller (1889), and Beust
(1907).

That sometimes the outer part of the symplasm may gonaite and inclose the newly devel-
oped cells like a cyst has been mentioned on page 181 as an interesting parallelism to the behavior
of the Myxobacteriaceae. But it seems as if such ‘‘macrocysts” of the bacteria, as they may
be called to distinguish them clearly from the microcysts discussed in Chapter II, are no regular
occurrence and no species mark. W. Winkler (1899) and Miinden (1907) have noticed them
with B. coli and diphtheriae, where they have been missed by other authors, like Kellerman
and Scales (1916) and Almquist (1917). The environmental conditions undoubtedly play a
very important réle in this case, although Miller-Thurgau (1908) perhaps went a little too far
when he assumed that practically the action of tannic acid alone was to be made responsible
for the formation of the interesting ‘‘bacteriocysts, ” which he discovered in wine. The paral-
lelism to Billroth’s Ascococcus (1874) , Babes’ Ascobacterium (1895), Moreno’s Ascobacillus
(1901), Winogradsky’s Amoebobacter and Thiocystis (1888), as well as to the last-named author’s
zoogloea form of Nitrosomonas (1892), is so evident that the whole subject undoubtedly needs
a more thorough study, which should also try to clear up the early findings of Letzerich, G. A.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 187

Hansen and Babes. (See pp. 170 and 171). Peculiar pseudo-macrocysts seem to occur occa-
sionally in the body; J. Jsraél (1878) encountered round bodies in decayed teeth wherein some
‘‘Leptothrix” was seen to develop, and Arndt (1880) reported that he found saliva corpuscles
filled with spirochaets and that he also saw spirochaets coming out and entering these bodies.

That large clumps of symplasm may encapsulate themselves has been discovered by
Rosenbach (1909); his photograph was reproduced as figure 242 on Plate XIX. But another
photograph of B. typhi, published by Almquist (1917), which was reproduced as figure 243 on
the same plate, makes it very probable that much of what represents itself as large yeast-like
cells in the symplasm of many different species of bacteria is in reality such encysted lobes,
made up of a deeply staining membrane and a hyalin content. In some cases the investiga-
tors have been able to branch off really yeast-like cultures, as, for instance, E. de Negri (1916)
from her Corynebacteria, but in other cases no growth of these problematical forms could be
secured, e. g., by Maher (1913) with B. tuberculosis. And after having become acquainted
with Rosenbach’s findings, it seems to me, indeed, very probable that as in Almquist’s typhoid
cultures so also those very large dark bodies visible in our photographs of Azotobacter sym-
plasms, reproduced as figures 253 and 254 on Plate XX, are no genuine cells, but encysted
lobes of the symplasm.

These bodies then would be closely related to other still-more curious forms, which in
our preliminary communications have béen called “uregular regenerative bodies,” and of
which it was said on page 122 that probably ‘‘sclerotia’’ would be a more correct denomination
for them, but that final action in this respect can not be taken at the present. The two pho-
tographs of Bac. subtilis and fluorescens, reproduced as figures 65 on Plate VI and 261 on Plate
XX, may serve as illustrations. The extremely polymorphous “bacteroids” of B. radicicola
are another well-known example. At least part of the queer spiny bodies developing from
the symplasm of the micrococci also belong to this class; and the same holds true in regard
to those irregular bodies of B. cyanogenes, pictured by Neelsen (1880), which were reproduced
as figure 85 on Plate S, as well as to similar abnormal forms of Streptoc. lactis, found by Burri
(1898) in Swiss cheese. That these bodies are nothing else than pieces of symplasm surrounded
by a more or less solid, deeply staining membrane is quite obvious in many cases; and many
of them merely persist for some time and then dissolve again into a more or less homogeneous
mass. Others, however, are able to propagate as such, at least temporarily ; those cultures
of irregular Actinomyces-like rods characteristic for the micrococci belong in this class. The
abnormal Streptococcus cultures of Burri, mentioned above, also grew as such for some time.

That as in these so also in other cases a close study of the development taking place within
and from the symplasm is the best means to attack successfully many of the numerous prob-
lems concerning bacterial pleomorphism, variability, occurrence of so-called involution forms,
limitation of the species, and similar points, needs no prolix discussion. It is equally obvious
that chemical, physical, and biological reactions will often exert a much stronger influence
when they are acting upon the symplasm than during the time the living material is standing
against them in a rather stable organization within the cell. Undoubtedly, Fokker (1902-1903)
went here, as in his heterogenetic speculations, too far when he assumed that by mixing
various species of bacteria in their symplastic stage new species could be created at will, but
he was perfectly right in making the following important statement (1903):

Wo sich in einer indifferenten Flissigkeit eine einzige Mikrobenart vorfindet, entstehen nach der Dissociation
Mikroben der gleichen Beschaffenheit . . . Wosich aber zwei oder mehrere Arten vorfinden und die dissociierten
Plasmata sich vermischen, kénnen sich aus diesem Gemisch Mikroben bilden, die die Eigenschaften der urspriinglich
anwesenden in sich zu einer Einheit kombinieren, also von jeder der eingesdten einzelnen Arten verschieden sind.
Auch ist die Méglichkeit nicht ausgeschlossen, dass bei der Dissociation in einer Nahrlésung, welche Bestandteile
enthilt, die in dem Substrate, in welchem die Mikroben erzogen sind, fehlten, diese Bestandteile auf das dissociierte
Plasma einwirken, so dass letzteres Veranderungen er'sidet, welche sich in dec Eigenschaften der.aus diesem Plasma
regenerierten Mikroben kundgeben.

It is evident that the perplexing problems concerning the often so far-reaching morpho-
logical and physiological alterations of the bacteria caused by symbiosis, as well as by the change
of substrate and environment, are becoming much more accessible, if the symplastic stage is
188 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

chosen as the starting point in such experiments. That a mixing of the symplasms of naturally
related bacteria, like the streptococci and the diphtheroids, will lead to.the appearance of inter-
esting intermediate varieties, as mentioned on pages 40-41, is very well conceivable, just as
well as or even better than the analogous facts caused by cross-breeding among the higher
plants or animals. It is practically unavoidable that the reconstruction of new cells from
such a mixed symplasm leads to the incorporation of elements of different parentage, so that
the newly born variety (not species, as Fokker thought) will combine the characters of the
original strains,

With the substrate it is quite similar, as Fokker has pointed out already. The upgrowth
of the regenerative units within the symplasm is very similar to the development of bacterial
gonidia within the dying or dead plasma of other plants or animals, which has been discussed
on pages 148-150. That the natural varieties of the bacteria exhibit so often to such a marked
degree the specific influence of their habitat is not astonishing, if one bears in mind how the
very intimate contact between substrate and living matter in the symplastic stage must exert
its direct influence upon the composition of the new cells.

It may. be expected with certainty that renewed experiments on the effects of ‘cross-
breeding’ and adaptation of the bacteria will furnish results of great importance to bacteri-
ology, as well as to biology generally, if they will be centered upon the symplastic stage in bac-
terial life. It can even not be denied that the réle played by the regenerative units in the
reconstruction ‘of new cells from the symplasm displays many features, which remind us of
the activities ascribed to the hypothetical ‘‘gemmules”’ of Darwin (1868, p. 448), the “‘plasti-
dules’’ of Haeckel (1876), the “microplasts’”’ of Hanstein (1882, p. 218), the “‘pangens”’ of
Vries (1889), the ‘“‘germplasm”’ of Weismann (1886, 1892), the “‘bioblasts’’ of Alimann (1894,
p. 141), and the ‘‘physiological units’ of Spencer (1898, p. 226). The greatest resemblance is
noticeable in regard to Spencer’s physiological units, which were declared to ‘possess the prop-
erty of arranging themselves into the special structures of the organism to which they belong”’
and to be , vompists aggregations of smaller chemical units, compounds of highly compound
molecules.” But also Alimann’s hypotheses upon cell formation are bound to gain new interest
when linked up with our new point of view. In regard to the higher organisms Altmann said
(l. ¢., p. 8):

Die Zellen sind nicht Elementarorganismen, sondern Colonien von solchen mit eigenartigen Gesetzen der Colo-
nisation; die Zellen entstehen aber nicht durch das Zusammentreten der Kiigelchen, sondern sie sind daraus in jenen
geschichtlichen Perioden entstanden, die den mikroskopischen Elementen gerade so eigen sind, wie den groben For-
men der Lebewesen auch.

That the theory ‘‘omnis cellula e cellula’’ is not so generally applicable, as Altmann and
many other authors believed, has been pointed out on page 167. But astudy of the symplastic
stage in the lives of the lower organisms proves conclusively that Altmann’s opponents, like
Hueppe (1896, p..2), A. Fischer (1899, p. 295) and A. Fischel (1903), erred to a still greater
extent when they disputed the fundamental importance of those ‘‘granules’’ for the life of the
cell. One has only to behold their agglomeration or their growing up to new cells under
the microscope, to become quickly convinced of their being, indeed, the “elementary
organisms,” forming temporarily a comparatively stable association in the form of a coccus,
a rod, or aspirillum, but continuing their existence unimpaired after these cells have undergone
renewed dissolution. The staining reaction of these granules may change within the symplasm,
as it does within the cells, where this fact has caused so many differences of opinion among the
various investigators. When, for instance, Metzner (1903) reports that the small granules in
many animal cells take the fuchsin stain while young, but become unstainable when they grow
old, and that in the latter stage they serve the metabolism of the cell, he is touching evidently
the same fundamental fact which manifests itself in the development of the pale cells from
agglomerations of well-stained regenerative units, as mentioned above, as well as in the variable
behavior of the gonidia within the bacterial cell, as was discussed in Chapter II.

The physiological value of the symplastic stage for the bacterial life is very clearly indicated
by the two facts, that the symplasm is formed from time to time in cultures kept for long
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 189

periods, but that it is also often produced by very young cultures immediately after they have
been transplanted to a new substrate. Buchanan (1918) is of the opinion that the bacterial
development proceeds regularly in seven phases, which are defined by mathematical formulas.
However, his assumption that at the beginning always a ‘‘stationary’’ period and “‘lag’’ be
noticeable, and that the “logarithmic phase of growth,” the “phase of negative acceleration”’
and the ‘maximum stationary phase’’ be followed with mathematical certainty by the “accel-
erated death rate’’ and the “logarithmic death rate,” is not in accordance with many well-estab-
lished facts. When Chesney (1916) studied the so-called lag in the development of newly trans-
planted cultures, ‘he noticed that no lag occurred when the transfers were made at the time of
maximal growth, but that it was always exhibited by older material. However, not only old cul-
tures, also young ones may show this “‘lag.’””, When we transferred our Subtilis and Azotobacter
cultures every morning and evening during several weeks, always from and to the same sub-
strate (meat agar for Subtilis, mannite agar for Azotobacter), no lag was noticeable in the growth
of the subcultures for several days, while it was very pronounced during the next days, then
again immmediate upgrowth took place, again a period of lag followed, and so on. Micro-
scopic examinations showed clearly that either the cells propagated as such, or that they first
entered the symplastic stage, and it was especially these experiments which we accepted as
evidence of the alternation between both modes of life (Léhnis and Smith, 1916 a, p. 679).
That the same alternation takes place when cultures are kept for a longer period without being
transferred to new substrates is indicated by the results obtained by Riemer (1909) and by
Springer (1913), which have been quoted on page 28, and it is proven by our continued studies
of all our 24 Azotobacter strains, which exhibited very distinctly in successive microscopic tests
their persistent passing through the various phases of their life cycle, when kept permanently
in the same solution (without any transfer) for more than a year. That under such conditions
metabolic products and other external influences will act besides the internal causes, which were
practically solely responsible for the results mentioned above, is not to bedoubted. In nature,
however, the situation is the same. And it will be advisable not to be too quick in declar-
ing changes in form and function, often noticeable in old cultures, as being ‘“‘merely”’ the
result of ‘involution or degeneration,” caused by the ‘‘detrimental effect of the metabolic
products.” It is true that secondary and tertiary colonies, growing on old agar slants, often
abound in “abnormal’’ cells, but as these cells, according to Preisz (1904) and others, are usually
more resistant than those which grew first, and the so-called involution forms, as was fully dis-
cussed in Chapter I, are often more numerous in young than in old cultures, it seems best to
apply the newly won knowledge of the réle played by the bacterial symplasm, to these facts
and to make them the object of renewed and more thorough studies. Why Fiirst (1914) found
in the course of his experiments upon the variability of Vibrio proteus that in the 4-months-old
cultures a period of rest preceded the return of the original type of growth, is easily to be under-
stood, and it is equally obvious why apparently fully stabilized varieties, that have remained
constant through hundreds of transfers made during several years, eventually ‘may return to
the original type. A close investigation upon the reproduction of the different types of cells
from the symplasm will elucidate all such experiments to a very considerable degree.

That practically all of the symplasm is used up for the reconstruction of the new cells is
an interesting sign of Nature’s economy, which was first recorded by W. Winkler (1899), and
which became again very conspicuous in many of our experiments. (See Léhnis and Smith,
1916 a, pp. 680 and 692, and the photographs reproduced as figs. 215, 216, 219, 220, 223, 243,
256-261, and 263 on Pl. XVIII-XXI.)

In an interesting paper upon autolysis, wherein the participation of assimilation and dis-
similation in all vital processes is thoroughly discussed, Nicolle (1913, p. 446) makes the following
statement, which is of special importance for our subject:

Lorsque l’assimilation et la désassimilation s’équilibrent harmonieusement, - . . Cestla vie stationnaire ...
Quand la désassimilation l’emporte, mais que l’assimilation conserve une intensité suffisante, c’est Vatrophie. Lorsque

la désassimilation l’emporte, tandis que l’assimilation tombe au-dessous d’une certaine valeur minime, c’est l’autolyse.
Quand J’assimilation l’emporte, mais que la désassimilation conserve une intensité suffisante, c’est Vhypertrophie,
190 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

voire 1’ hyperplasie. Lorsque l’assimilation l’emporte, tandis que la désassimilation tombe au-dessous d’une certaine
valeur minime, c’est l’autocoagulation, phénoméne inverse de l’autolyse; c’est la naissancade formes résistantes,
rigides, eoaguil des, condensées . . . Chose curieuse, les deux tendances opposées se partagent quelquefois la meme
cellule (sporulation rapide, lors d’inanition brutale) donnant ainsi Vimage d’une action et d’une réaction également
excessives.

But in the same manner as the French proverb “Les extrémes se touchent’’ is applicable
to this latter case, it holds equally true in regard to many instances, where the new forms
emerging from the symplastic stage are not, as might be expected at the first moment, normal
vegetative cells, but, quite to the contrary, regenerative bodies, rods with endospores, and all
lands of “giant” forms, which are usually classed as signs of involution and degeneration, and
which only in the course of their further development gradually resume that harmonious equi-
librium between assimilation and dissimilation which, according to the French author, char-
acterizes stationary life.

2. THE SYMPLASTIC STAGE IN THE DIFFERENT GROUPS OF BACTERIA.

That the so-called Ascococcus of Billroth (1874) does not represent a special genus, but is
merely a type of growth of micrococci developing from the symplastic stage, is hardly to be
doubted, despite an opposite statement made by F. Cohn (1875, p. 151). A drawing published
by the latter author has been reproduced as figure 275 on Plate XXII (from original fig. 5 on
Pl. V), where it may be compared with another picture (fig. 276), published by Rabe (1886,
original fig. 2 on Pl. IV) of his Micrococcus botryogenus, which exhibited these actinomyces-like
agglomerations only within the tissue, while on artificial substrates the development did not
differ from that of any other Micrococcus.

A characteristic photograph of Micrococcus candicans growing up from the symplastic stage
has been reproduced as figure 216 on Plate XVIII (from our second preliminary paper, 1916 b).
All intermediate steps from the smallest regenerative units up to full grown cocci, like those
shown in figure 1 on Plate I, are clearly discernible. The peculiar clusters of irregular, spiny,
rodlike regenerative bodies, in figures 2 and 8 on Plate I, described in Chapter I as a charac-
teristic type of growth of the micrococci, are equally the recult of a complete transformation of
flakes of symplasm of Microe. candicans..

Figure 264 on Plate XXI (taken from our second paper, 1916 5, original fig. 53) demon-
strates the analogous behavior of a Sarcina. Again at the edge of a large flake of symplasm
of Sarcina flava the gradual development, of the regenerative units to large cells is to be seen.
And in figures 16 and 17 on Plate II the final transformation of the large cells into normal
sarcinae has been shown. Before having reached their definite shape, the packets are often
connected by more or less solid bridges, which are clearly visible in figure 16, and which have
been described before by Sauerbeck (1909) in a paper upon a Sarcina mucosa, of which several
packets also frequently were seen to be enclosed in a slimy ‘‘capsule” (probably of symplasm).

The ability of staphylococci and gonococci to produce in 3-4 weeks old cultures ‘‘white
pleomorphous clumps,” similar to those of the so-called syphilis bacillus, as shown in figure 217
on Plate XVIII, has been mentioned already by Niessen (1898). But it was Herzog (1913)
who first pointed out that this passing over of the gonococci into such ‘‘verquollene Schleim-
massen” is of considerable etiologic importance, and that these clumps should not be lightly
discarded as dead.

Analogous observations concerning streptococci were made by Billroth as early as in 1874,
upon which he reported (p. 12):

Das Plasma tritt in Form amorpher, kérnig-schleimiger Substanz aus den Kugelnhervor . . . Ob damit
das Pflanzchen abgestorben ist, will ich dahin gestelltsein lassen.

Why he did not succeed in getting a new development, with these organisms, while in other
cases he secured positive results with similar bacterial ‘‘plasmodia,” must be left undecided.
That it was undoubtedly the regenerative units and round regenerative bodies growing in and
from the symplasm of lactic acid streptococci in milk which misled Fokker (1901) to his
anachronistic hypothesis of the heterogenetic origin of milk bacteria from casein granules has
No. 2.1 LIFE CYCLES OF THE BACTERIA—LOHNIS. 191

been discussed on page 172 and illustrated by figure 214 on Plate XVIII. Figure 215 on the
same plate pictures, on the other hand, a flake of symplasm of the same species which, however,
is transforming itself directly into normal vegetative cells of Streptococcus lactis. That as with
other bacteria, so also with this species, the symplasm may break up occasionally into entirely
irregularly shaped regenerative bodies may be concluded from an observation made by Burri
(1898), who discovered in faulty Swiss cheese colonies of an organism, belonging into the group
of Bact. Guentheri (Streptoc. lactis), which besides a few normal cells contained numerous irregu-
larly curved and distorted forms, which, however, were able to multiply as such for some time,
just as the analogous irregular regenerative bodies of the micrococci are inclined to do.

The findings of Bernhardt (1911), Hoefer (1911), and of Mallory and Medlar (1916) concern-
ing the causative agent of scaralatina, which have been quoted on page 184, represent another
important contribution to our knowledge of the symplastic stage in the life cycle of the strep-
tococci. And it is very probable, too, that in the case of poliomyelitis analogous results will
be reached. The photograph published by Rosenow and Towne (1916), which was reproduced
as figure 9 on Plate I, looks much like a picture of symplasm producing new cells, and that
various filterable vira have been found in close connection with such slimy, granular, differently
staining masses, characteristic for this stage, was mentioned on page 185.

The small micrococci of about 0.15 diameter, embedded in thick zoogloea masses, which
R. Koch (1878) found instead of septicaemic bacilli in spreading abscesses in rabbits, and of
which he discovered that even the unstainable homogeneous part proved to be infective (causing
septicaemia), may well be accepted as the regenerative units of B. septicaemiae; Koch’s drawing,
reproduced as figure 211 on Plate XVII, indicates clearly that he had large agglomerations of
symplasm under the microscope.

In the case of B. radicicola the symplasm and the various growth developing from it have
been the cause of widely divergent conclusions reached by the different authors. As was men-
tioned on page 180, the organism has been placed among the Myxomycetes as well as among the
Myxobacteria. Frank (1879) thought at first that the infective threads and the branched and
budding forms developing from them should be accepted as mycelium and haustoria of a fungus.
Later (1887), with the help of his pupils Brunchorst (1885) and Tschirch (1887), he deprived it
entirely of its microbial nature; the ‘‘bacteroids’”’ were now declared to be cell products of the-
legume. Two years later, however, after Beijerinck (1888) had isolated B. radicicola, Frank
(1889) discovered that the infective threads be ‘‘plasmodia’’ of a peculiar ‘‘Rhizobium legu-
minosarum”’ and the ‘‘bacteroids”’ some kind of ‘‘mycoplasma.” But in 1890 the German
author also came to the conclusion that bacteria are to be made responsible for the various types
of growth within the plant. Moéller (1885) also classed the alder organism at first among the
Myxomycetes, because he saw the finely granulated plasma differentiate itself into round

“spores,” i. e., regenerative bodies, while he later (1890) assumed it to bea ‘“‘unicellular hypho-
myces” (Frankia subtilis),. producing globular, oval, or pear-shaped ‘‘sporangia”’ of 4-6u
diameter at the end of the ‘‘hyphae,” wherein ‘‘spores’’ are said to develop, which later repro-
duce a new ‘‘mycelium.” The analogous change of opinion is also to be found in Prazmowski’s
(1888-1890) important papers upon the nodule organism of the leguminous plants. At first the
“‘Kndllchenpilz”’ was described as forming large cysts, which after being placed in water liberated
their content, made up of ‘‘bacteroids”’ embedded in a slimy, plasmatic substance. But later
these cysts were accepted as being encapsulated colonies of bacteria. Laurent (1890) was in-
clined to place B. radicicola at the side of Pasteuria ramosa on account of the similarity noticeable
between the ‘‘bacteroids” of the nodule organism and the branched clubs characteristic of
Metchnikoff’s organism. Morck (1891) found out that their regular bodies occurring in the root
nodules of different legumes are either able to reproduce directly small motile bacteria or that
they break up at first into round regenerative bodies. Their resistance against acid and alkali
proved to be comparatively high.‘ The granular plasma, visible in the youngest part of the
nodules, was seen to produce either round or irregular forms or directly small rods. When study-
ing thenodules of Alnus Hiltner (1898) obtained very similar results, Irregular slimy clumps and
threads, containing rods, were seen, as well as ‘ ‘sporangia, produced by the rods or the threads,

160266°—21——18 >
192 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [ Vou. XVI.

which divided themselves into round ‘‘spores,”’ which in their turn reproduced new rods. A later
publication by Hiltner and Stérmer (1903) contains those interesting photographs reproduced as
figure 236 on Plate XVIII (from original fig. 8 on Pl. II), showing the ‘‘sporangia”’ of Alnus,
and as figures 271 and 272 on Plate XXI (from original figs. 5 and 6, Pl. II), illustrating the
analogous development of round bodies within the nodules of clover and pea, respectively.
Shibata (1902) reported that the round bodies of Alnus may occasionally enter a process of mul-
tiple segmentation, similar to that sometimes observed with microcysts of streptococci and of
other bacteria, which has been discussed on page 143. As figure 273 on Plate XXI a photograph
has been reproduced (Léhnis and Smith 1916 a, original fig. 34), showing a flake of symplasm
from a root nodule of red clover, which is producing clubbed and branched ‘‘bacteroids”’ as well
as small, round regenerative bodies.

An excellent photograph of round regenerative bodies, found in cultures of B. colt by
Kellerman and Scales (1916), has been reproduced as figure 44 on Plate IV. It should be com-
pared with Hiltner and Stérmer’s Alnus picture (fig. 236 on Pl. XVIII). That also in this
case the large bodies grew up from the symplasm is indicated by other details visible in the
photograph. The hyphae-like threads of B. coli, shown by the first-named authors in two other
pictures (figs. 45 and 224 on Pls. IV and XVIII), make interesting counterparts to the
“infecting thread” of B. radicicola. In figure 223 on Plate X:VIII the simultaneous and uniform
upgrowth of the regenerative units of this species is equally well demonstrated. That the
symplasm of B. coli may sometimes exhibit amoeboid movements has been observed by W.
Winkler (1899) and by Miinden (1899).

The picture published by Almquist (1916) of the various cell forms of B. typhi, which was
reproduced as figure 243 on Plate XIX, shows very clearly the consolidation of the plastic
substance into slimy threads and large globules. Meirowsky’s (1914 6) drawings of what he calls
“umbels” of paratyphoid and enteritis bacilli exemplify the upgrowth of new rods and threads
from the symplasm. A few reproductions are to. be found on Plate XIII (fig. 178, right lower
corner) and on Plate XIV (second column).

That in cases of rhinosclerom the formation. of symplasm by the causative agent is rather
conspicuous has been indicated by the early findings of Cornil and Alvarez (1885) and of Klebs
(1887).

A good illustration of the regeneration of new cells from the symplastic stage among the
lactobacilli is given in the photograph of B. acidi propionici c, published by Almquist (1917) and
reproduced as figure 246 on Plate XIX. The “‘vesicular forms’’ of B. bifidus, which Tissier
(1900) considered to be degenerative, are probably, like similar forms frequently to be found
with B. radicicola, coli, pneumoniae, Azotobacter, etc., another type of new cell formation from the
symplasm. Tissier’s drawing, which has been reproduced as figure 8 on Plate C (the “formes
vésiculeuses”’ are to the right), should be compared with the photographs made by Toenniessen
(1913) of B. pneumoniae, reproduced as figures 38-41 on Plate IV; the gradual development of
regular cells from these irregular intermediate forms is especially well discernible in this case.
Other data in regard to the symplastic stage of members of the B. pneumoniae group are given
in a paper by Jehle (1902).

A Bac. tubifex, described by E. Dale (1912) as causative agent of a disease of potato leaves,
was seen to produce infecting threads of the same appearance as those of the nodule bacteria,
and probably also of the same physiological quality.

The important findings of Rosenbach (1909) concerning the symplastic stage of B. erysipe-
latos suum (B. erysipeloides) have been fully discussed on page 176. Highly interesting photo-
graphs of his paper were reproduced as figures 239-242 on Plate XTX, and as figures 27-29 on
Plate III.

The mycelial upgrowth shown in the last-named picture invites comparison with a similar
photograph made by Rowland (1914) of B. pestis (see fig. 30, Pl. III), while other data relating
to the symplastic stage of this organism have been observed, but not correctly interpreted,
by NV. K. Schultz (1901, see figs. 218-221 on Plate XVIII and previous discussion on p. 176).

The agglomerations of growing regenerative units of the bacillus isolated by P. Th. Miiller
(1913) in cases of typhus exanthematicus, which have been reproduced as figure 22 on Plate II,
No. 2.] © LIFE CYCLES OF THE BACTERIA—LOHNIS. 193

deserve to be compared with figures 216, 223, 240, 247, and 255 on Plates XVIII-XX, picturing
the same phase in the life cycles of other bacteria (M. candicans, B. coli, erysipeloides, anti-
typhosum, and azotobacter, respectively). In addition, figure 24 on Plate II clearly represents
another type of upgrowth from the symplasm of Miiller’s organism; although a glance at the
torn flakes of symplasm visible in the field leaves no doubt that a more carefully made prepara-
tion would have presented much better objects for comparative study.

In Hauser’s first publication upon his Proteus (1885) the symplasm with its regenerative
units and their development to new bacilli has been fairly well described.

Analogous observations with various Pseudomonas species were made by Fuhrmann
(1906-1908). One of his photographs is to be found as figure 238 on Plate XTX. Various
phases in the development of different cell forms of B. fluorescens from the symplastic stage
have been reproduced as figures 259-263 on Plates XX-—XXI (from our second paper, 1916 6,
original figs. 8, 49-51, 63). The photograph shown as figure 265 furnishes another good illus-
tration of how different the cells of “typically” short rods may appear under such conditions;
the picture was made from an aerobic cellulose destroying organism, Bact. acidulum.

What Omelianski (1899) described as a new nitrite-forming organism should be studied
again, as its photographic picture, reproduced as figure 268 on Plate XXI, makes it prac-
tically certain that merely a clump of round regenerative bodies has been seen, which grew
up from the symplasm, such as is shown in figures 257 and 258 (Plate XX) from B. azotobacter,
or in figure 219 (Pl. XVIII) from B. pestis. The two other photographs of Nitrosomonas, made
by Winogradsky (1892) and reproduced as figures 269 and 270 on Plate XXII, are good illustra-
tions of macrocyst formation. The presence of the large round cells within the ‘“zoogloea”
(fig. 270) strengthens our belief, expressed on page 135, that the so-called Nitrosococcus of
Winogradsky (shown in fig. 198 on Pl. XV) is naeiing but a type of regenerative bodies of
Nitrosomonas.

Not many data are available at present concerning the symplastic stage of spore-forming
bacilli. Thatit plays an important réle in the life history of B. anthracis has been repeatedly
discussed by Fokker (1881-1882 and 1902). Young (1914) made a few remarks upon so-called
sporoblasts, apparently being agglomerations of regenerative bodies growing up from the
symplasm and developing to spore-forming rods. Our own preliminary studies upon the life
cycles of B. azotobacter and related sporogenous bacilli, however, leave no doubt that also
within this group of bacteria correct and complete results will not be reached until detailed
investigations have been carried out in this direction. The photographs of B. azotobacter
(figs. 70 on Pl. VI,'225-231 on Pl. XVIII, 250-258 on Pl. XX), of B. subtilis (figs. 65 on Pl. VI,
233 on Pl. XVIII), of B. malabarensis (fig. 266 on Pl. XXT) and of the ‘‘yellow bacillus No. 41”
(fig. 267 on Pl. XXT) will suffice to indicate that very many interesting problems are awaiting
adequate research.

That vibrios and spirilla also pass through the symplastic stage has first been noticed by
Finkler and Prior (1884-1885) and by Wetbel (1887). Dowdeswell (1889-1890), however, is to
be credited with having made the first more thorough investigations upon this subject. ‘‘Cocci”’
embedded in amorphous masses have been also found by Jorge (1896) in cultures of a vibrio
isolated from water, but their true character was not ascertained. Almquist (1916) got an
interesting culture of fairly resistant regenerative bodies from a cholera culture, heated up to
60 C.; in figure 244 on Plate XIX the upgrowth of these round forms from the symplasm is
very well illustrated. It is not improbable that Hueppe (1885) had similar objects under the
microscope when he made the sketch of a “zoogloea”’ containing round bodies which is visible
in the center of figure 49 on Plate M. Albrecht (1881) discovered in blood cultures of Spiro-
chaeta, Obermeiert irregular clumps of dark points, which permanently changed their form, and
he ‘also observed the formation of new spirochaets from them. The more recent findings,
recorded by Meirowsky (1914 6) with various spirochaets and by Leishman (1918) with Spiro-
chaeta, Duttoni, are in perfect agreement with those early discoveries. On Plate XIV, in the
lower part of the second column, some interesting drawings made by Meirowsky of spirochaets
194 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

growing up from the pian are to be seen. Figure 86 on Plate S should be compared
with them.

With the Mycobacteria some more data are available. The early findings of Letzerich
(1873-1876, cf. page 170) have, of course, not been obtained with pure cultures of diphtheria
bacilli; nevertheless, they invite further investigation. That such macrocysts as have been
seen by this author may be found, indeed, in pure cultures of B. diphtheriae has been made
very probable by results recorded by Miinden (1899). Formation of round regenerative bodies
from amorphous masses in old diphtheria cultures, i. e. from the symplasm, has been noticed by
Spirig (1903).and by Balfour (1911.d). A fairly good photograph of various cell forms of B. diph-
theriae, developing from the symplastic stage, has been published by Almquist (1917); itis repro-
duced as figure 245 on Plate XIX. That the amorphous agglomerations present in old cultures
of B. xerosis are equally able to produce a new growth has been ascertained by P. Ernst (1888).
On the other hand, it was evidently not correct, when Morse (1912) maintained that only one
group of pseudo-diphtheria bacilli (B. hoagit) be characterized by the fact that in old cultures
the bacilli “fuse into poorly staining masses,” whereas B. zerosis and Hofmanni are said not
to exhibit such changes. The photograph made by Mellon (1917) of diphtheroids from the
lung, reproduced as figure 249 on Plate XIX, furnishes additional evidence in this respect.
The publications of Niessen (1898, 1908) upon his so-called syphilis bacillus contain several
interesting data concerning the symplastic stage of diphtheroid organisms, of which, however,
only very cautious use should be made, on account of the uncritical manner in which they
have been gathered and reported.

That with leprosy the symplastic stage of its causative agent is of considerable frequency
and. importance may be seen from the publications of G. A. Hansen (1880-1882), Babes (1883,
1907), Lutz (1886), Czaplewski (1898), Barranikow (1900), Pernet (1902), and Meirowsky (1914 b),

Equally numerous are the observations upon finely granulated or entirely. homogeneous
“zoogloea’’ masses in tuberculosis and in cultures of B. tuberculosis.. The contributions made
in regard to this subject by Klebs (1883), Babes (1883), Malassez and Vignal (1883-1884),
Amrusch (1886), Schroen (1886-1904), Metchnikoff (1888a), Arrigo (1900), Meier (1903), Maher
(1910-1913), and Meirowsky (1914 b) have already touched many points, which should be thor-
oughly studied... Especially the work of Schroen, though rather incomplete in itself, is un-
doubtedly worthy to be.taken up anew and carried out in a more satisfactory manner; the
data collected with B. radicicola, and the peculiar behavior of this organism within the root
nodules of leguminous and of other plants, reveal many striking resemblances to what has
been: described and pictured by Schroen concerning his ‘‘new phthisiogenic microbe.” That
this is no fungus, as was assumed by its discoverer, is beyond doubt, despite the incompleteness
of the report. The thick slimy threads, accepted as mycelia, resemble very closely. those
produced by B. radicicola within the plant.

The granular mass, found by Crookshank (1896), Mertens (1903), and others in the center
of many clusters. of Actinomyces, is obviously an unchanged part of symplasm, and the de-
scription given by Lachner-Sandoval (1898, quoted on p. 174) concerning the development of
young threads of Actinomyces from flakes of symplasm, could be accepted as pertaining at the
same time to any other kind of rod-like bacterium. Meirowsky’s drawings (see Pl. XIII, fig.
178, No. 57, and Pl. XIV, second column, ‘‘Paratyphus B”’) and our photograph of the spore-
forming yellow bacillus No. Al (fig. 267 on. PL. XXI) fit Lachner’s description completely.

The development of new threads of Cladothrix from round gonidia (or regenerative bodies 2),
which are,embedded in a ‘‘zoogloea’’ (symplasm), has been illustrated by Billet (1890) in the
upper part of the sketch reproduced as figure 70 on Plate P. That the microgonidia of Crenothrix
often enter the symplastic stage before new threads are reproduced has been pointed out by F.
Cohn (1870) and by Zopf (1879). And ne wrote alnoay | in 1852 (p. 215) in regard to Galli-
onella ferruginea:

’ Nach der Zerstérung der Hiillen ballt sich der Inhalt, der mehr oder minder die'Form von Kérnchen annimmt,
die von 1/1500” bis zu unmessbarer Kleinheit vorkommen, in Haufen zusammen. Diese manchmal unregelmiissigen

Kérnchen scheinen eben sowohl zur Vermehrung zu dienen, als die regelmiissigen, innerhalb der Faden erzeugten,
kugligen oder elliptischen Sporen.
No, 2.] LIFE CYCLES OF THE BACTERIJA—LOHNIS. 195

That the so-called ‘‘Haftscheibe,” considered by Molisch (1910) to be quite characteristic
for certain species of iron bacteria, may be also correctly classified as a remnant of the symplas-
tic flake, which produced the thread, is hardly to be doubted. On the other hand, the statement
made by Winogradsky (1888, p.40) that Zopf had been wrong in attributing to Beggiatoa the
ability to grow sometimes attached to the substrate, and that such sulphur bacteria be repre-
sentatives of a special genus Thiothrix, will have to be reconsidered in the light of the new facts
mentioned above. Evidently all rod- and thread-like bacilli are able to make use of their
symplasm for attaching themselves to their substrates if the conditions under which they are
living are favorable to such an arrangement.

3, CONCLUSIONS.

From the details discussed on pages 166-195 the following conclusions can be drawn con-
cerning occurrence and formation of the symplasm and the regeneration of new cells from the
symplastic stage:

(1) All bacteria live, in vivo as well as in vitro, alternately in an organized and in an amor-
phous stage. By the partial or complete dissolution of vegetative and reproductive cells a
plasmatic mass, the symplasm, is formed, which after a period of rest and according to cir-
cumstances may transform itself into new cells of thesame or of a more or less modified char-
acter. It has been found, for instance, in young as well as in old cultures of Microc. gonorrhoeae,
candicans, and other species of Micrococcus; various streptococci; Bact. pneumoniae, rhinosclero-
matis, coli, typhi, paratyphi, enteritidis, pestis, septicaemiae, erysipelatos suum, proteus, bifidus,
and other lactobacilli; Nitrosomonas; B. radicicola and fluorescens; Bac. anthracis, subtilis,
and Azotobacter; Vibro cholerae and proteus; various spirilla and spirochaets; Corynebacterium
diptheriae and pseudo-diphtheriae; Mycobact. tuberculosis and leprae; different Actinomycetes
and trichobacteria. Within the body the presence of bacterial symplasm has been recorded
in cases of gonorrhoea, trachoma, variola. scarlatina, poliomyelitis, meningitis, endocarditis,
septicaemia, rhinoscleroma, anthrax, pseudo-tuberculosis, tuberculosis, leprosy, malignous
granulom, diptheria, actinomycosis, actinobacillosis, recurrens fever and syphilis. It partici-
pates in the formation of hyalin and amyloid substances within the body. Observations made
by Pineau in 1845 and by Perty in 1852 have already indicated that all microscopic life starts
from the symplastic stage, which, however, is not entirely absent in the cell life of the higher
organisms. The earliest complete description of the symplastic stage of bacterial life has been
published by Lankester in 1876. The so-called autolysis of bacteria is in part identical with
the formation of the symplasm. The dissolution of the cells should not be considered
synonymous with death of the living matter itself, and the regeneration of new cells from the
symplastic stage is not to be accepted as heterogenesis, although this has been done by
Bastian, Fokker and others.

(2) The formation of bacterial symplasm proceeds wave i in two phases: First, the cells
agglomerate to smaller or larger clumps; second, a,more or less complete dissolution of the cells
takes place, resulting in a crumbly or slimy mass, assuming the shape of flakes, irregular clumps
or regular spheroid bodies, and being sometimes endowed with amoeboid motility. The dissolu-
tion of the agglomerated cells leaves either the empty cell walls as shadows, but still clearly
visible, in the field, or they, too, are disintegrated, forming a more or less homogenous symplasm
of an either hairy, granular, or entirely amorphous, hyalin structure, of variable staining qualities,
and of fairly high resistance against acids and alkali. The different types of symplasm may be
found with the same species of bacteria; one type may pass into the other. Presence or absence
of amoeboid motility is no species mark; the same holds true in regard to an occasional
encystment of the symplasm. All kinds of vegetative cells as well as of reproductive organs
of the bacteria are able to enter the symplastic stage. Irregular regenerative bodies, gonidangia,
and liberated gonidia (including the filterable ones) are especially inclined to produce symplasm,
and even endospores may do the same.

‘(3) The reconstruction of new cells from the bacterial symplasm follows various lines, accord-
ing to internal as well as external conditions (quality of the symplasm, environment, and
196 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

technique). At first always regenerative units become visible, which either grow up separately
to new cells, or of which several may agglomerate and surround themselves with a uniting
membrane thus forming at once full-grown cells, Usually the reconstruction starts at the
edge of the symplastic flakes or clumps, occasionally leading to a distinctly star-like grouping
of the new cells; sometimes, however, the transformation takes place simultaneously throughout
the whole mass. The new cells formed are as a rule regenerative bodies of round or irregular
shape, but normal vegetative cells may be also produced immediately. Endospores, too, may
develop at once within the new cells. A uniform appearance is rather rare among the newly
constructed cells; usually they are more or less pleomorphic. A special type of growth sometimes
to be found in such cultures are broad, slimy threads, often of irregular outline, which by
contraction or by segmentation reproduce normal vegetative cells. Like other irregular rege-
nerative bodies they have been improperly classed as ‘‘involution forms.” Generally all of
the symplasm is transformed into new cells; occasionally, however, the outer part of it becomes
membraneous and incloses the whole colony of newly developed cells as a ‘‘macrocyst,’’ or a
small flake of symplasm remains at the basis of a newly formed thread, fixing it to its sub-
strate. All the different possibilities mentioned may be found with the same species of bacteria.
Macrocysts have been seen with micrococci (Ascococcus), B. radicicola, Nitrosomonas, Asco-
bacterium, Ascobacillus, Mycobact. leprae, and with various sulphur bacteria (Amoebobacter,
Thiocystis, Clathrocystis, and Bacterium rubescens); they indicate a closer relationship between
Bacteriaceae and Myxobacteriaceae.

(4) As the various subcycles within the life cycles of the bacteria are connected by the
symplastic stage, a thorough study of the regeneration of the new cells from the symplasm is
essential for securing a correct knowledge of the principles governing the pleomorphism and the
variability of the bacteria. Biological, physical, and chemical factors exert usually their most
pronounced influence during this phase of the life cycles of the bacteria, which in their turn
are at this time in a much better position to adapt themselves to changed environmental condi-
tions, than they are during their cellular life.
IV. CONJUNCTION.

Our studies upon the life cycles of the bacteria led us to the conclusion (Léhnis and Smith,
1916 a, p. 700) that besides the formation of the symplasm another mode of interaction between
the plasmatic substances in bacteria cells is generally to be found, consisting of the direct union
of two or more individual cells. We called this process ‘‘conjunction,” because frequently more
than two cells are united, and a connecting bridge (yoke) is sometimes, but not always, visible;
therefore, the terms ‘‘¢opulation” and ‘‘conjugation”’ are not well applicable.

Conjunct cells were constantly seen in cultures two to four days old. The conjunction
always preceded the formation of the gonidia, of lateral and terminal regenerative bodies, as
well as of exo- and endospores. The latter, too, sometimes entered the conjunct stage.

The formation of connecting bridges, first seen with large Azotobacter cells, reminded me
of a similar observation made occasionally by Férster (1892), who found that in freshly collected
living material of Chromatium Okenii, during the first 3 days, often a ‘‘ primitive copulation”
was noticeable, characterized by the formation of cylindric unstained bridges, leading from
and to the unstained inner part of 2, 3, or more neighboring cells. Sometimes a knoblike
inflation became visible in the middle of the bridge, and it was also no rare occurrence that
only a half bridge was protruding from a single cell. No conspicuous change was seen to follow
the act of ‘‘copulation,” but the cells were especially large at this time, while they later lost
in size as well as in vitality.

This, however, was the only confirmative finding which we could cite in our first prelim-
inary report (1916 a, p. 687), though we pointed out in addition that conjunction of the bac-
teria, like their budding and the formation of the symplasm, has not only been seen by many
bacteriologists, but also has been unknowingly reproduced in several illustrations in our daily
used textbooks. ‘

As figures 278-294 on Plates XXII and XXIII pictures are given, which may illus-
trate the appearance of various species in the conjunct stage. Figure 278 has been taken from
Féorster’s paper. It makes a good counterpart to our Azotobacter photographs, No. 279-283.
Figure 284 presents spores of Azotobacter in conjunction; the fine bridges are clearly notice-
able also at some isolated spores. Figure 285 shows B. fluorescens, figures 286 and 287 B.
subtilis. Figure 288 is a photograph of B. Chauvoei, made by Itzerott and Niemann (1895,
original figure 48); figure 289 was taken from a contact preparate of a colony of B. fluorescens,
made by Azelrad (1903, original figure 22); figure 290 shows Bact. esterificans Stralauense,
photographed by Maassen (1899, original figure IX, 1); figure 291 was made by Guenther (1906,
original figure 8) from a colony of large bacilli, isolated from a straw infusion; figures 292 and
293 are parts of a picture of the hay bacillus, published in the textbook of Hiss and Zinsser (1914,
original figure 125); and figure 294 is another Azotobacter photograph, made by Walton (1915,
original figure II, 8). By a careful examination of these-pictures the various modes of uniting,
which often replace the bridges between the cells, will be easily discovered. Especially a pe-
culiar beak-like development of an end of acell, stretching out to the side of another cell, is very
striking. Sharp bents of the rods are equally conspicuous, as are the T and V formations,
while little swellings or small droplets, sometimes noticeable at the touching points, will reveal
their presence upon close scrutiny.

It goes without saying that the process itself is much better to be studied with the living
material, and reproductions from stained preparates can only convey a more or less incomplete
impression. But good photographs of living cells showing conjunction are hardly obtainable,
and as with the symplastic stage, again direct microscopic studies are indispensable for reach-

ing a correct view of the subject. as
198 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

In his textbook on bacteriology A. Fischer (1903, p. 42) states very distinctly:
Geschlechtliche Fortpflanzung ist beiden Bakterien noch niemals, auch nicht andeutungsweise beobachtet worden.

That he did not know, at least, Férster’s paper, which was published in the ‘‘Centralblatt
fir Bacteriologie,” makes it very doubtful whether he was entitled to make such a strict state-
ment. Numerous other observations, to be recorded on the following pages, are equally in
open conflict with Fischer’s attitude as well as with the standpoint taken by many other writers.

It was pointed out in the foregoing chapters that R. Koch, Gaffky, Fliigge, and Fraenkel
were undoubtedly wrong, when they established their rigid monomorphistic. doctrine, that
Brefeld as well as Migula were not in accordance with well-established facts when they asserted
the. occurrence of various types of reproductive organs among the bacteria to be a priori
impossible, and that it was a serious mistake when R. Koch and Loeffler lightly discarded all
that was already known at their time upon the symplastic stage of the bacteria. But these
erroneous ‘views are still upheld in many bacteriological textbooks, despite the very numerous
facts which conclusively prove their inaccuracy. Concerning the occurrence of sexual pro-
cesses among the bacteria much less data are available, but they are numerous enough and of
sufficient weight to recommend the adoption of a more correct view than that held by A.
Fischer.

It is to be readily admitted that for the practical bacteriologist a knowledge of the
occurrence of sexual processes among the bacteria is of much less importance than a thorough
acquaintanceship with their pleomorphism, their various modes of reproduction, and with
their alternating between cell life and symplastic stage. This, however, does in nowise inter-
fere with the great scientific interest of the subject.

Before taking up the data collected from the bacteriological literature, it seems advisable
to discuss briefly a few points, which will be helpful for securing a better understanding and a
more correct valuation of what has been done thus far by bacteriologists.

It has been emphasized already by Darwin (1868, pp. 431, 432) that sexual reproduction is
not so very different from asexual reproduction, and that it is an error to assume that a bud
differs essentially from a fertilized germ, though mndoubiedlys ‘beings produced sexually are
much more liable to vary than those produced asexually.” According to Spencer (1898, p.
265) the terms asexual and sexual multiplication are equivalent to cell reproduction and cell
generation. Delage (1895, p. 117) makes the following statement:

Le conjugaison est bien véritablement l’intermédiaire entre la reproduction asexuelle par spores et la reproduc-
tion sexuelle par oeufs et spermatozoides .. . . La conjugaison va nous présenter une série de formes qui par une
extrémité confinent 4 la reproduction asexuelle, . . . et par Vautre passent a la reproduction sexuelle.

Sachs (1875, p. 203) distinguished between asexual generation, with asexual reproductive
cells, developed independently without foreign aid, and sexual generation, i. e., development.
of new cells with the aid of another cell. Hf in the latter case the uniting cells are uniform, the
process was called conjugation, if they are conspicuously different, fertilization.

Kruse (1896 }, p. 83). says:

Partielle Konjugation heisst der bei Infusorien weit verbreitete Vorgang, bei dem sich gleichartige Individuen
vortbergehend mit einem Teile ihres Kérpers vereinigen, je einen ihrer Mikronuclei mit einander austauschen und
sich wieder von einander trennen. ,Ausstossung von Richtungsspindeln, Kern-Auflésung und -Neubildung findet
dabei nicht statt.

It is obvious that this feanition fits-very closely to that type.of sonjunction which is most
frequently exhibited by the bacteria.

According to Prowazek (1907 6) sexuality is an elementary process inherent to the living.
matter and either already discovered or expected to be found with all lower plants and animals.
Doflein (1909, p.. 158) states that as far as the protozoa have been studied multiplication by
segmentation can not go on indefinitely, and that sometimes after a few, in other cases after
thousands, of purely vegetative generations sexual processes are taking place, either in the form
of copulation (i. e., complete uniting of two cells) or as conjugation (i. e., interaction between
the nuclear substances of various cells).
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHMNIS. 199

With special reference to sexual processes among protists Hartmann (1909) has pointed
out that fertilization is characterized more by the copulation of the cell nuclei than by the inter-
action of the cells themselves, and that, therefore, besides ‘“‘amphimixis” between different
cells, ‘‘automixis,’’ i. e., the copulation of various nucles within the same cell, should be taken
under consideration as an important mode of fertilization among protists. A copulation taking
place between sister cells is called “‘pedogamy,” and declared to be frequent among amoebae,
bacilli, and lower algae. Long-continued vegetative reproduction is accepted as causing a
gradual increase in differentiation among the cells and their nuclei, and therefore the fertiliza-
tion is viewed as a means of reestablishing the inner equilibrium, and thereby stimulating anew
vegetative growth and multiplication. The effect exerted by amphimixis upon the variability
of lower organisms has been discussed by Pringsheim (1910, pp. 120-126).

Of special interest are the following sentences, written by Barker (1901) in his first paper
on Zygosaccharomyces, whose mode of conjugation is duplicated in a striking manner by the
behavior of the large bacilli in their conjunct stage. The yeast cells, obtained from commer-
cial ginger, produced beaks which touched each other with their tips, causing a fusion of the cells:

The appearance of the compound cells, thus formed, was that of two ordinary cells attached to one another by
an elongated neck with complete communication of their cavities through the neck, and at this stage with protoplasm
filling up the whole of the cells and the neck.

A few hours later the formation of bright granules and of spores in both cells became
noticeable. Sometimes the tubes did not conjugate, but turned into buds or branches.
Satisfactory results concerning the behavior of the nuclei were not obtainable, but the author
is inclined to accept the process as a sexual one, not merely as a fusion of cells, and he continues:

There is one aspect of all simple sexual acts that appears worthy of further attention, and which is suggested very
definitely by the present case. It would appear that whenever nucleated spore-producing protoplasm is being stored
in neighboring cells, or whenever it has been accumulated and is parted by a septum, a strong tendency exists for
union. Numerous examples can be given, such as the fusion of sporidia of Ustilagineae or of Protomyces, the nu-
clear fusions in teleutospores, aecidiospores, etc., or those in the ascogenous cells of Erysipheae, Exoascus, Peziza,
etc., and in the basidia of Hymenomycetes, also the fusion of sister cells in some Conjugatae, e. g., Spirogyra.

Many additional data concerning the copulation of yeasts have been secured by Guillier-
mond (19106—1912), who found the process rather widespread among the spore-forming yeasts
and Schizosaccharomycetes; spores, too, sometimes showed fusion, occasionally already while
still within the ascus. But numerous cells produced spores without previous copulation. Col-
onies of Debaryomyces globosus exhibited clearly the copulation occurring between sister cells.
Often the two cells were quite different in appearance, ‘‘mais on ne saurait voir dans cette
‘inégalité l’indice d’une tendance a l’hétérogamie.” Other species, however, especially a Zy-
gosccharomyces Chevalieri, furnished good examples of true heterogamic copulation; the whole
content of a smaller cell (microgamet) passes over into a larger one (macrogamet), which trans-
forms itself into an ascus.

According to Marchand (1912) the copulation of the spores by making beaks and bridges is
fairly frequent with yeasts, though a great number of them germinate and develop independ-
ently.

That many parallels are presenting themselves between these observations and those enu-
merated above, relating to the behavior of the bacteria, is hardly to be denied. And this will
become more evident when the following findings are also taken under consideration.

As early as in 1872 Rindfleisch pointed out that he often saw two bacteria united with
each other, and that this union was not the result of fission. Im 1875 Klebs reported in his V.
contribution “Zur Kenntnis der pathogenen Schistomyceten” that motile rods often at first
have the shape of short coccoid or oval bodies, and that these, after repeated touching and
separating, at last enter a permanent conjunction, and then grow up into long rods and threads,
which later segment themselves again into round cells. He calls the process explicitly ‘‘ einen
Act der Verschmelzung oder der Copulation.”

.. Still more important is the following discovery, published by Albrecht in 1881. The cells
of Spirochaeta Obermeieri were seen, when kept in blood at 37° C., to touch each other first with
200 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

their ends and then fuse together more or less completely, so that they presented the appearance
of one organism. After a shorter or longer interval they separated again; but the author em-
phasizes that a true fusion is characteristic for this. occurrence, which should not be mistaken
for a casual apposition of two cells.

Buchner (1882) declared the occurrence of rods in V form to be characteristic of B. subtilis,
while it was pointed out by Escherich (1894) that the long forms of the diphtheria bacillus
are often to be found in V form or in rosettes, and that the ends touching each other are often
broader than the rods themselves. Kurth (1898) and others have obtained analogous results
with this species. Upon the behavior of the so-called Vibrio denitrificans, Severin (1897, p.
512) reported as follows:

In 1-2 tagigen Bouillonkulturen kann man hiufig Stébchen beobachten, die in sehr verschiedenartiger Vereinigung
miteinander verbunden sind, bald zu zwei, bald zu drei, zuweilen noch mehr, bald in form von T, bald legt sich das
eine Stabchen an ein anderes in form eines Zweiges an, oder es vereinigen sich zwei derselben mit ihren Enden.

Under the title “‘An unusual bacterialg rouping” Hefferan (1912) discussed the various
modes of rosette formation, as found by her with B. rosaceus metalloides. She is inclined to
connect it with a peculiar mode of cell division, but the experimental support furnished for
this assumption can hardly be accepted as convincing. The author herself explicitly admits
that—
there remain numerous possibilities, yet untouched by investigation, as to the actual cause of this peculiar phenomenon,
which I am inclined to believe is a problem closely related to the vital processes of the organism.

It was also noticed that not infrequently rods were swarming together to angles or to a
rosette-like position, and that they then were either staying or again swarming away; ‘‘a
peculiar clear space at the center”’ became often visible in this case.

One of our photographs of B. radiobacter (Léhnis) and Smith, 1916 }, original fig. 29) illus-
trating this mode of star formation has been reproduced as figure 277 on Plate XXIT. A
comparison with figure 274 on the same plate, showing the star formation from the symplasm,
will be helpful for accurately distinguishing between both processes.

Apposition of motile cells in.a T position to other cells has been further recorded by
Bajardi. (1902, with V. lingualis), who being evidently not acquainted with the analogous
findings published before, wrote upon this fact:

Man findet hier die eigentiimliche Erscheinung, dass haufig kiirzere Stiibchen oder Fadenstiicke sich rechtwinklig
an die langeren Faden anlegen, woraus man den Eindruck einer seitlichen Knospung gewinnen kénnte.

Cytological investigations led Rayman and Kruis (1904) to the belief that copulation
between young bacterial cells is by no means infrequent. And results obtained in ultramicro-
seopic studies induced Gaidukov (1906) to make an analogous statement.

~ Obviously without knowledge of Albrecht’s early work, the conjunction shown by spirochaets
has been repeatedly discussed during the last 10 years, and several authors have declared
themselves to be inclined to interpret this process, whose last phase was often mistaken for
longitudinal fission, as genuine copulation. See, for instance, Sobernheim (1907, p. 582),
Breinl (1907), Schellack (1907), Meirowsky (1914 6), and F. Levy (1916). Especially the last-
named author’s dark field studies, made with Spirochaeta Obermeieri, are in complete agree-
ment with those findings recorded by Albrecht 35 years before.

When studying the development of bacterial colonies Graham-Smith (1910) found that
always a characteristic ‘‘post-fission movement” was noticeable, such as had been discovered
by Hill (1902 a) with B. diphtheriae, indicating a decided tendency of the newly formed rods
to place themselves parallel each other. This was realized either by “loop forming” (seen
with B. anthracis), by ‘‘folding” (bacilli related to Anthrax, and B. pestis), by ‘‘snapping”
(B. diphtheriae) or by slipping (B. coli, pneumoniae, pyocyaneus, pseudo-tuberculosis, subtilis,
various vibrios and spirilla). The physiological meaning of this parallel arrangement has not
been considered by the author; it seems, however, as if in well-made contact preparates of
bacterial colonies always numerous cells are to be found in the characteristic conjunct stage,
being connected by bridges or by beaks, such as are visible in figures 279-283, and especially
in figures 289 and 291.
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 201

According to some observations made by Faber (1912) with B. rubiacearum pairs of rods
in parallel arrangement may turn themselves by double folding into a cross-like position.
This occurrence is similar to that described by Hefferan.

That the various modes of conjunction may be followed by the formation of lateral and
of terminal regenerative bodies has been noticed by several authors.

Some of the drawings made by Hueppe (1885) of what he called arthrospore formation of
V. cholerae (see fig. 49 on Pl. M) leave hardly any doubt that it is more correct to assume that
the long spirilla isi the characteristic V position, which later produce the round bodies at their
ends, were in conjunction, than to accept this as a prefission phase. Similarly, Babes (1889)
pointed out that the round regenerative bodies frequently seen by him with spirilla and vibrios
are either located at one end of the cell or at the point where two cells touched each other, or,
as he thinks, were they divided. The photograph of V. cholerae, made by Friedrich (1892)
and reproduced as figure 140 on Plate XI, demonstrates especially clearly the appearance of
conjunct cells and their relation to the formation of regenerative bodies.

The sexual processes ascribed by Vincentini (1893) to his so-called Leptothrix racemosa are
probably to be interpreted as an example of conjunction, resulting in the production of round
regenerative bodies. But the short abstract of the apparently rather confused paper, which
was only accessible to me, allows no definite conclusion.

The very interesting drawings of regenerative bodies produced by acetic acid bacteria in
2-days-old cultures, made by Henneberg (1898) and reproduced as figure 57 on Plate N, repre-
sent the first clear indication that the formation of this type of bacterial reproductive organs is,
indeed, very much alike to the zygospore- or to the ascus-formation of certain fungi. For
instance, a picture of Eremascus fertilis published by Guilliermond (1910b) could be readily
mistaken for a drawing of conjunct lactobacilli producing a lateral round regenerative body;
even the septa visible within the mycelium find their counterparts in Henneberg’s sketch. The
following remark made by Guilliermond fits both cases equally well:

L’anastomose s’accompagne d’une résorption de la parois séparatrice des deux articles réunis et permet la mélange
de leur contenu dans la cellule mére de l’asque.

A photograph made by Maassen (1899) of his Bact. esterificans Stralauense, reproduced as
figure 290 on Plate XXIII, exhibits, besides other interesting details, in the middle of its upper
edge an exceptionally good example of the initial step leading to the formation of a lateral regener--
ative body, which should be compared with the upper part of Henneberg’s drawing. Tissier
(1900) found the ends of his B. bifidus often inflated, especially those connecting two united
rods; and at this junction frequently a round regenerative body (“une boule”) became visible.
Analogous observations were made by Fuhrmann (1906) with 2-days-old cultures of his Pseu-
domonas cerevisiae. Rods with terminal inflations were seen to touch each: other sideways,
and a connecting bridge was found to exist between these bulbs. Fuhrmann says:

Diese ganze Erscheinung erinnert sehr lebhaft an Bilder, wie wir sie bei der Zygosporenbildung zu sehen gewohnt
sind.

Droba (1901) reported that he discovered “‘Zygo- and stylospores” with Bac. tuberculosis,
and Miihlens and Hartmann (1906) thought that the round bodies which they found with
spirochaets might be the products of some preceding act of fertilization. Some drawings made
by Prowazek (1906a), reproduced as figure 89 on Plate S (from original figs. 10 c, 11a and 6 on
Pl. II), illustrate very clearly how two separate regenerative bodies or one compound body (of
the character of a zygospore) may be formed by two conjunct cells of Spirochaeta gallinarum,
very much as it was seen and drawn by Hueppe (with V. cholerae) and by Henneberg (with Bact.
oxydans). White and Avery (1909) noticed also quite correctly that the oval or kidney-shaped
“nodules” of lactobacilli are located either at the point where two bacteria touched each other,
or that each of them is supported by a short stem, branching out of the side of a rod. It was,

however, far from being correct when they asserted: ‘
Their formation may undoubtedly be considered a true plasmolysis.

The photograph of B. subtilis, published in Hiss and Zinsser’s textbook (1914, original fig.

125), which contains those cells in the characteristic conjunct position, which were reproduced
202 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVL

in figures 292 and 293 on Plate XXIII, also presents at the same time a well-developed lateral
regenerative body, visible in figure 150 on Plate XI. A photograph of B. coli, made by Keller-
man and Scales (1916), which was shown as figure 185 on Plate XV, is of still greater interest.
With most cells the conjunction is easily noticeable, and a pair in V position in the upper part
of the field shows again the regenerative body being formed at the touching point. Almguist’s
pictures (especially figs. 135, 138, and 139 on PI. XI) offer some additional confirmative data,
relating to B. typhi and V. cholerae, as does figure 172 on Plate XII (a reproduction of one of our
photographs of B. bulgaricus).

The formation of endospores has been connected with copulation, probably for the first
time, by Hueppe (1891, p. 29), who wrote:

Die Endosporenbildung erscheint mir als wirkliche Fruktifikation, als einfachste Form einer Art geschlechtlicher
Fortpflanzung, der Copulation, zur besseren Anpassung an die Art bedrohenden Aussenbedingungen.

This view was shared by Gamaleia (1900, p. 23); and on account of cytological results
obtained with the large new species B. Bueischlit and sporonema, Schaudinn (1902) promoted
the theory that the endospores represent the result of autogamy between neighbored cells, a
theory which later has been confirmed by Rizitka (1910), but contested by Dobell (1909-1911)
and by Swellengrebel (1913). Investigations carried out by Baur (1905) upon the formation of
arthrospores (or microcysts?) by Myxococcus led to results which are in several directions
similar to those of Schaudinn, and the same holds true with regard to observations recorded
by Gross (1913) concerning the arthrospore formation of spirochaets. Park and Williams (1914,
p. 38), too, are inclined to interpret as ‘‘a primitive sexual process, a sort of dutogamy”’ the
fusion of metachromatic granules, which was seen to precede cell fission in the large, : often
branched, cells of diphtheria bacilli.

Another mode of interaction between separate bacterial cells has been described by Vuil-
lemin (1888) in his studies upon the “zoospores” of the nodule organism from the legumi-
nous plants as follows: ;

Parfois les zoospores se rapprochent et restent appliquées l’une 4 ]’autre parle bout pointu . . . Les boules
brillantes sont souvent placées sur les faces opposées. Nous ne savons trop quelle interprétation attribuer 4 cette
sorte de conjugaison. A un certain moment il y a, semble-t-il, une soudure véritable sans que le corps des conjoints
subisse des modifications visbles. =

After separation one body was seen to move rapidly away, while the other became immotile,
increased in size and turned over into what seems to have been a gonidangium or a large round
regenerative body. The French author concludes (1. c., p. 193):

On se demande s’il n’y a pas 1a une sorte de fécondation, dont le resultat est de transformer la zoospore en spore.

Like Vuillemin, but apparently without knowledge of this author’s findings, Hartleb (1900)
was inclined to believe that the copulation of the “zoospores” of B. radicicola leads to the
production of ‘‘zygospores.”

That also a distinct sexual dimorphism may be found sometimes among the bacteria has
been discussed for the first time by Haberkorn (1882), though it is not to be denied that his report
does not look very trustworthy. Small bodies were seen to fuse with larger ones, which later
developed to gonidangia (called ‘‘Samenkapseln”’ by Haberkorn). Essentially the same fact
has been recorded by Ferrdm (1885) in regard to the cholera vibrio. Small bodies, termed
“antheridia,” were seen to fertilize larger round “ oogones,”’ which afterwards became gonidangia,
The Spanish author, who was hardly acquainted with Haberkorn’s short communication in the
“Botanische Centralblatt,” points out that he found similar “ oospheres”’ also with several bacilli.
But his results differed in many respects so widely from those of R. Koch and his pupils, which
then were accepted everywhere as the only reliable ones, that they probably would have been
passed practically unnoticed, also if they had not been mixed with several quite unwarranted
hypotheses, which were bound to create considerable distrust in Ferrén’s whole work (cf. our
discussion of his publication on p. 69). Dowdeswell (1889-1890), however, has fully confirmed
Ferrdn’s discovery. Observations of the cholera organism in Ranvier’s moist chamber showed
him that, indeed, large round ‘“‘oospores”’ are fertilized by small pear-shaped “zoospores,” and
that they afterwards break up into “sporules.”’
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 203

Firtsch (1888), on the other hand, who also paid attention to the large globules produced
by Vibrio proteus, is of the opinion that they are simply the result of a fusion of the loops, fre-
quently formed by the curved cells especially during the second day; despite the complete
absence of thorough 4nvestigations in this direction, he feels justified to state:

Mit Fruktifikation haben diese Gebilde . . . natiirlich nicht das Geringste zu tun.

Sexual dimorphism among spirochaets has been repeatedly discussed by protozoologists.
For instance, Perrin (1906) interpreted the conjunction of thinner and thicker cells of Spirochaeta
Balbiami as “conjugation of male and female gametes.” Prowazek (1907 a), Krzysztalowicz and
Siedlecki (1908), Gonder (1909), and. others have shared this view, while it has been refuted by
Fantham (1908), Doflein (1909, p. 315), Dobell (1912), and others.

A process of fertilization, very much like that described by Ferrén and Dowdeswell with
V. cholerae, has been reported by McDonagh (1912) with Spirochaeta pallida. A round
“macrogamete’”’ was seen to be fertilized by a “‘microgamete,” showing the typical form of a
spirochaete, and afterwards to develop to a “‘zygote,”’ which at first divided into four “sporo-
blasts” and later into numerous ‘“‘sporozoites.”’

An unbiased consideration of all these more or less incoherent findings and opinions, col-
lected on the foregoing pages, makes it undoubtedly imperative to abandon the a priori negative
standpoint at.present en vogue in regard to sexualism among the bacteria. It is readily to be
admitted that a great deal of what has thus far been recorded upon this problem is rather unsatis-
factory. But enough data remain which clearly indicate that more thorough investigations
will yield important results.

Besides the normal mode of conjunction, which is practically constantly noticeable in all
young cultures, especially from the second to the fourth day, I myself have seen occasionally,
when examining the living material in the hanging drop, that two cells of about equal size unite
themselves permanently in such a manner as has been described by Klebs (1875 a), and I have
also occasionally met single large cells of B. azotobacter and of other species, which were attacked
again and again by small motile bodies, exactly as it has been reported by Haberkorn (1882),
Ferran (1885), Dowdeswell (1889-1890), and by McDonagh (1912). Not knowing of any of
these publications when these casual observations were made, they have not yet been followed
up by special studies, but they may be mentioned here, because they perhaps will help to direct
the attention of other investigators to this difficult subject.

More light upon this phase of the life cycles of the bacteria may be expected from careful
researches upon the formation and behavior of the bacterial gonidangia, which at the same
time will do away with much useless talk about “involution forms.’ A peculiar ring formation
has been described by Van Tieghem (1884, p. 110) with B. amylobacter, by Firtsch (1888) with
V. proteus, by Morck (1891) with B. radicicola, by Schmorl (1891) with Streptothriz cuniculi, by
Burri (1897) with B. bernensis, by Moro (1900) with B. acidophilus, by Rosenfeld (1901) with
several members of the B. septicaemiae group, by Stefansky (1902) with his B. pyogenes ramosum,
by P. Ernst (1902) with B. fluorescens, and by numerous authors with various spirochaets.
Several of these reports seem to indicate that these rings or loops are much inclined to turn into
globules, as was already observed by Perty (1852), Firisch (1888), Morck (1891), and by P.
Ernst (1902), and it is not impossible that these globules are able to act later as the ‘‘macro-
gametes” of McDonagh and his earlier colleagues before developing to gonidangia.

Whether the conjunction of the young bacterial cells has anything to do with the appear-
ance of the gonidia within these cells, remains to be seen. That, however, the round lateral
and terminal regenerative bodies are frequently true zygospores becomes increasingly probable,
though it is quite possible that, as with certain fungi, so also with the bacteria, ‘“azygospores’’
may sometimes take the place of zygospores.

The peculiar triangular forms found with spirilla and other bacteria whose center gradually
assumes globular shape while the ends are fading away (see figs. 86 and 87 on Pl. VIII) present
another problem which perhaps wa find its solution when attacked from this new point of

vantage.
204 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

How far the formation and germination of endospores is dependent on a previous act of
autogamy or heterogamy will have to be decided by further research. Our figure 69 on
Plate VI indicates fairly well that the formation of endo- as well as of exospores may be influenced
by the conjunction of the parent cells, and the conjunction between the liberated endospores
themselves, as visible in figure 284 on Plate XXII, should be accepted at present at least, as an
interesting occurrence of unknown physiological value. It should be kept in mind in this con-
nection that analogous facts are known from various fungi, and it seems to be especially suggestive
that, for instance, the sporidia of Ustilagineae have been found by De Bary (1884, p. 192) to
copulate before germinating when kept in a poor substrate (water), but that on a rich substrate
they germinate without previous copulation.

It has been pointed out in Chapter III that sometimes endospores, like the bacterial cells
themselves, may enter the symplastic stage before reproducing new cells, and it is quite probable
that the results obtained by us in this direction have been partially due to our keeping the
cultures usually in much less concentrated substrates than are generally used in bacteriological
laboratories, substrates, however, which are very favorable to a long-continued development of
the bacteria, as was exemplified by the data given on page 189. It is well known that trans-
plantation of well-grown cultures into water exerts a stimulating influence upon the forma-
tion of reproductive organs. It also strengthens the tendency of the bacteria to enter the
symplastic stage. And the same may hold true in regard to the conjunction of the bacteria,
which despite our very incomplete knowledge can be looked upon, like the formation of the
symplasm, as a means of placing the continuity of bacterial life upon a broader and safer basis
than that afforded by a single cell.

That studies on the behavior of the nuclear material within the bacterial cells during and
after conjunction are not very promising at the present time, is hardly to be denied, the diffi-
culties in this case being so much greater than with the yeasts and other lower fungi. However,
careful and patient research will undoubtedly secure more evidence that, as has been made
probable by the more or less casual observations quoted above, sexual processes also play
their réle in the lives of the bacteria, just as they do in the lives of all other, lower as well as
higher, organisms, and that, in this respect also, the bacteria are not those unnaturally rigid
things which they were declared to be by the preconceived doctrine of constancy and simplicity,
but that they again behave much more like other living beings and, therefore, deserve to be
studied like these in all the vagaries which they exhibit in the course of their life cycles.
V. METHODS.

|. ad 1; DIFFERENT CELL FORMS.

On pages 37-43 the main points of importance for studies upon pleomorphistic problems
have been discussed. In addition, the following references occasionally may prove helpful
in further investigations.

In regard to microscopic observations R. Koch (1881, p. 6) and Fraenkel (1891, p. 86-89)
have published most valuable hints for avoiding possible errors due to artefacts, contaminated
staining solutions, etc. Equally worth reading is A. Fischer’s (1899, p. 30) discussion of the
vexatious réle played sometimes by peptones, albuminous, and other substances in dried prepa-
rates. For staining, the Congo red adsorption method, as worked out by Benians, is highly
recommended by Hort (1917a). According to Knack (1915) nigrosin is superior to India
ink, and especially suitable for studying living material. Many details concerning dried smears
and hanging drop preparations are given in Hueppe’s book on bacteriological methods (1891,
pp. 48-51, 86-99). Prazmowski (1912, p. 110) found wax preferable to fat or vaseline for
sealing hanging drops, because it does not shut off the air so completely as the last-named sub-
stances do. Schouten (1905) pointed out that hanging drops of broth often become too acid
to allow further growth; placing some 2 per cent KOH below the drop is recommended as pre-
ventive measure. Th. Cohn (1913) found it useful to change the hanging drop into a thin
liquid layer by placing a small cover glass on it, an arrangement which seems, indeed, to be
very practical for certain purposes. The hanging agar block method of Hill (1902 b) gave
Baur (1905) less satisfactory results than the hanging drop; Graham-Smith (1910), who used
it extensively, had also to record frequent failures.

The continuous observation of bacterial development under the microscope is often impos-
sible on account of the disinclination of many bacterial cells to grow under such abnormal
conditions; moreover, their motility may spoil the result. On pages 38-39 it has been pointed
out why the statement, made by some authors. that only those changes should be accepted
as correctly ascertained, which have been observed directly and without interruption under the
microscope, can not be considered as appropriate. Even Winogradsky (1888) had to admit
that the movements of the bacterial cells often hindered the continuous observation, which he
declared absolutely imperative for solving pleomorphistic problems. And though DeBary
(1884, p. 137) was undoubtedly right when he said that the continuity of the different phases
in the growth of fungi (and bacteria) must be as accurately ascertained as the fact that the apple
is a developmental product of the apple tree, certainly neither he nor anybody else would insist
that only if this apple has grown from its beginning to full maturity in the laboratory under
prescribed conditions and under the never flinching eye of one observer the fact can be accepted
as properly proven.

The use of single-cell cultures is another desideratum which is greatly exaggerated by some
theorists. Repeated plating coupled with careful microscopic tests furnishes, as has been
discussed on page 39, exactly the same results as the single-cell cultures, which generally are
less satisfactory on account of lack of growth and greater chances for contamination. Occasion-
ally the fishing methods recommended by Schouten (1905-1907), Barber (1907-1911), Mieu-
wenhuis (1911), and Hecker (1916) will be useful. Usually, however, the India ink method
of Burri (1909) has much in its favor, especially in connection with the improvements worked
out by Viehoever (1913, p. 241: dilutions to be made in small test tubes) and by Troili-
Petersson (1904: droplets to be placed upon gelatin or agar on hollow slide or in Boettcher’s
chamber). The little-known drop plate method of Holten (1893) deserves also further testing.

205
206 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

After having made the dilutions in a good nutrient solution, small droplets are placed upon a
glass plate possessing a net work of glass ridges (1 mm. high); another glass plate is put on as
cover, and those drops containing only one cell can be marked and the growth studied directly
under the microscope. The early method introduced into bacteriology by Klebs (1873) to
keep the inoculated gelatin in a Geissler chamber, which allows the microscopic study of the
isolated germs, offers other obvious advantages for investigations upon the life history of the
bacteria. It has already been recommended by H. VM. Ward (1895), who also made gelatin
cultures on cover glasses.

Replacing the Petri dishes by flat square bottles tends further to reduce the possibilities
of contamination; and often repeated plating, together with a sufficient number of parallel
tests, whose great importance has been pointed out on page 39, makes the experiments as safe
as, or even safer than the various modes of single-cell cultures, but generally much more
successful.

Frequent transfers from and to the same substrate usually hinder the bacteria to display
their natural pleomorphism. Therefore, if not such incomplete results are wanted for some
special purpose, long-continued tests on a greater variety of substrates, kept at different temper-
atures and under otherwise modified conditions, are generally preferable, and they are indis-
pensable if complete and correct data upon the life history of the bacteria are to be secured.

As a rule the substrates recommended for laboratory use are too concentrated, and the
large amount of metabolic products accumulating therein often causes the premature death
of the cultures, which in less concentrated media continue to exhibit their characteristic alter-
nating mode of life at least for several months. Some recipes for such weak nutrient solutions,
suited for soil bacteria, have been given in our first preliminary paper (Léhnis and Smith, 1916 a,
pp. 686 and 689), and it was also mentioned there that sterilized soil itself proved to be a very
suitable medium, allowing B. azotobacter to display all its characteristic cell forms with the
greatest precision within a few weeks. “ Rejuvenation”’ in sterilized soil has also been found
by Bredemann (1909) to be very helpful for reestablishing the common character in many
different Amylobacter strains. Cultivation in water has helped Stamm (1914) and others to
discover important parts of the life cycles of V. cholerae and of other organisms, which are

much less conspicuous on the standardized substrates. Broth diluted with water furnished
Ferran already in 1885 all the different cell forms of the cholera bacillus and gave later Taddei
(1909) the various types of growth of Streptoc. pyogenes, choreae, and erysipelatos, Distaso
(1916) recommended diluted serum (1 part serum diluted with 1 part tap water, and digested
by trypsin or pancreas).

Increased ‘aeration, which is often desirable when working with solutions can be established
either by immersing plates or strips of gypsum partially into the liquid, as has been done with
good results, for instance, ‘by Casagrandi (1901) and by Cacace (1903), or simply by adding a
strip of filter paper before sterilizing, so that it stands out about an inch above the solution.
Increased atmospheric pressure, together with relatively high temperature, was found by
Arloing (1908) to be very effective in bringing about the development of clubbed, branched,
and of giant forms of B. tuberculosis. On the other hand, anaerobiosis allowed Concetti (1901)
to change the actinomycotic type of growth of B. diphtheriae quickly back into the typical,
highly virulent rod form, and Almquist obtained many of his interesting results with B. typhosus,
V. cholerae, and with other ‘pathogenic organisms by keeping his cultures at temperatures as
low as 10-12°C.

The influence of a wider range in acid and alkaline reaction of the substrates deserves to be
made the object of detailed studies. Already in 1888 Wasserzug noticed that B. prodigiosus
always produced rods and threads in acid, but coccoid and spiral forms in alkaline broth.
Th. Smith (1890) saw B. mallei grow with a yellow-orange color on acid, but as a gray layer on
alkaline substrates. According to Thiercelin and Jouhaud (1903 6) the Enterococcus appears as
a typical Streptococcus in slightly acid broth, but alkaline reaction stimulates the development
of tetrades. Anaerobic bacilli were more ready to produce their ‘‘giant’’ forms in Hibler’s (1908)
experiments when 1-2 per cent Na,CO, was added to agar, serum, or potato. On the other
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 207

hand, B. typhosus, coli, and related forms presented their wide pleomorphism to Hort (1917 a),
when kept in distinctly acid broth (usually +10 to+20, sometimes up to +60 phenolphthalein).
Bunker (1917) noticed that varying hydrogen-ion concentration induced the appearance of
all different types of diptheria bacteria desscribed by Wesbrook.

For special purposes, especially for securing a quicker development of the various forms,
it will sometimes become advisable to increase the amount of certain substances in the substrates,
though under such conditions special care must be taken to prevent an early death of the culture.
Albrecht and Ghon (1900), for instance, pointed out that the addition of 5 per cent glycerin or 5
per cent sugar proved to be very useful for studying the pleomorphism of B. pestis. And it was
with the same organism that Hankin and Lewmann (1897) first found out the stimulating
influence exerted by a higher concentration (24-3 per cent) of common salt. Analogous results
with this and numerous other species have been recorded by Matzuschita (1900), Skschivan (1900)
Rosenfeld (1901), Stefansky (1902), Maassen (1904), Hata (1908), and others. That most of
the cell forms seen by these authors have been wrongly classed as involution forms and that
common salt merely speeds up the normal development and induces often, on account of its
physiologically drying effect, the production of resting and reproductive organs has been dis-
cussed in Chapters I and II.

Other salts, like the lithium compounds tested by Gamaleia (1900, p. 207), the rubidium,
caesium, magnesium, strontium, baryum chlorides used by Maassen (1904), potassium iodide
and urea tested by Péju and Rajat (1906-1907), and by Wilson (1907), exert, besides their
general osmotic action, more or less specific influences. And this is still more the case with the
various antiseptics, such as 0.5 per cent potassium bichromate (Thiercelin, 1903), 0.5-1.5 per
cent carbolic acid (Cacace, 1903), 0.2 per cent saturated alcoholic solution of methyl violet
(Walker and Murray, 1904), or small amounts of CuSO, (Garbowski, 19075).

Special substrates have been used and recommended in many: cases, so for developing the
branched forms of B. radicicola by Stutzer (1900-1901), Hiltner and Stérmer (1903), Buchanan
(1909), Zipfel (1911), and many others, for B. diphtheriae by Escherich (1894, p. 81), Abbott
and Gildersleeve (1903), for B. tuberculosis by Wolbach and Ernst (1903) and by Wherry (1918).
Egg substrates proved to be of special benefit in investigations upon branching made by Fraenkel
(1895), Meyerhof (1898), Dorset (1901 6), and Kodama (1908), while serum was found to be most
useful by Kohlbrugge (1901, for water vibrios), by Vincent (1902 for branching streptococci)
and by Carpano (1913 for developing the full pleomorphism of B. mallei); continued cultivation
of B. coli in ascitic fluid enabled M. EF. Abbott (1900) to fix the coccoid growth, so that it did not
revert to the rod form. It goes without saying that the skillful use of many varied substrates,
together with such an arrangement of the experiments that the bacteria are allowed to grow
under conditions as natural as possible, will lead to the most satisfactory results.

As has been pointed out in Chapter I, the study of symbiotic effects is also of very great
importance. Instead of keeping different strains in mixed cultures, as was already done
with good result by DeBary (1884, p. 503), it may be sometimes preferable to adopt the ‘‘double
test tube’? method of Smirnow (1908, the apertures of the smaller inner tube being covered by
celloidin), or to add merely the sterilized metabolic products of other cultures, a method which
has given remarkably good effects, even when the material was taken from old gelatin cultures
of the same species (B. mycoides, according to Nadson and Adamovié, 1910).

2. ad II: REPRODUCTIVE ORGANS.

It has been discussed in Chapter II, especially on pages 108-119, that numerous difficulties
may arise in the course of investigations upon the formation and the further behavior of the
various types of reproductive organs of the bacteria, and that generally their characteristic
function, viz., the reproduction of new vegetative cells, will have to be studied directly, toeliminate
all doubtful points, which can not be definitely settled otherwise. It must be borne in mind,
however, that the conditions usually prevailing in the hanging drop or agar block preparations
are often not favorable to the development of new cells, and that, therefore, a few negative
results obtained in this manner can not be accepted as a safe basis, on which to found final

160266°—21——-14
208 MEMOIRS NATIONAL ACADEMY OF SCIENCES, [Vou XVI.

conclusions as to the reproductive or nonreproductive character of some problematical body.
It needs hardly to be emphasized that especially the study of the small gonidia, as well as
that of the more conspicuous regenerative bodies, necessitates great care in making micro-
scopic examinations. When discussing A. Fischer’s much exaggerated plasmoptysis hypothe-
sis, it has been pointed out that a very thorough cleaning of the slides and cover glasses (by
potassium bichromate+sulphuric acid) is absolutely necessary. Great care in making and
fixing smears is equally essential, as many of these frail things will not stand the ordinary rubbing
and heating; the round regenerative bodies are especially liable to be materially altered by
excessive heat. Lack of care in rinsing the stained preparate may wash away many of the
slightly fixed bodies.

More than ordinary cleanliness is also essential for cultural experiments. Sterilization of
vessels and substrates kills the regenerative bodies more easily than the endospores, but they
still may become visible in the microscopic field and cause considerable error. As figure 295
on Plate XXIII a photograph has been reproduced (Léhnis and Smith, 1916 }, original fig. 57),
showing various vegetative cells and regenerative bodies of B. Azotobacter developing from the
symplastic stage, and in figure 296 (reproduced from original fig. 58, 1.c.) the same material is
demonstrated after it had been heated for 30 minutes in the autoclav at 20 pounds pressure.
Jt is evident that substances and vessels, containing such bodies, are very liable to lead to
erroneous results. Careful comparative studies of steril controls will always be advisable, and
occasionally the treatment of the glassware with chromic acid and the filtration of the sub-
strates through Berkefeld, Chamberland or asbestos filter (Hew, 1906, p. 35) will become
necessary.

Transplanting vigorously growing cultures: of several days age into water, mineral solu-
tions, on substrates containing a relatively high amount of salt, or keeping them at com-
paratively low temperatures, are the means which have been used most successfully to stimu-
late the development of reproductive organs. Transfers made into water gave Van Treghem
(1879 6) very satisfactory results in his studies upon the spore formation of the bacteria, includ-
ing the arthrospores of spirochaets. In the same way Braem (1889) got many different types
of reproductive organs (regenerative bodies, gonidangia, etc.), which he mistook for signs of
“degeneration,’’ and Rullmann (1900) recorded analogous findings in mineral solutions as
being “Hungerformen”. Many of the “abnormal”’ cell forms appearing on salt substrates
leave no doubt about their being regenerative bodies or gonidangia, though they, too, were
usually classed as “involution forms.’’ Especially the investigations made in this direction
by Mateuschita (1900) and by Maassen (1904) have furnished numerous data, which may
be used as a basis for further, more thorough research. Almquist (1904), Fuhrmann (1906),
and Hammerl (1906) ascertained that these various forms, whose development is favored by
increased salt content of the substrate, are indeed reproductive organs. The first-named
author (Almquist, 1904-1916) has also contributed much material illustrating the stimulating
effect of low temperatures (10-15° C.) upon the production of regenerative bodies, especially
by pathogenic organisms. Results secured by Schtiirmayer (1900) and by Arloing (1908) in
regard to the formation and multiplication of gonidia, regenerative bodies, and of gonidangia
of the tuberculosis group are in good agreement with Almquist’s observations.

Change in reaction naturally will also exert its influence. Garbowski (19076), for instance,
noticed in the course of his studies upon variability and spore formation of B. luteus sporogenes
that on nitrate agar endospores were readily produced, while on sugar or glycerin-ammontar-
trate agar only darkly staining inflations became visible (either regenerative bodies or goni-
dangia), and that this difference was due to an alkaline reaction developing in the nitrate
agar and an acid one in the glycerin or sugar agar. Bredemann (1909) has collected more
evidence in regard to the favorable influence of an alkaline reaction upon spore formation.
On the other hand, it seems equally advisable to study more closely the specific effect of a more
or less acid reaction upon the production of the other types of reproductive organs.

Simonds (1915) has pointed out that besides composition and reaction of the media the
influence of symbiosis should not be overlooked. Experiments with the double test-tube
No. 2. LIFE CYCLES OF THE BACTERIA—LOHNIS. 209

method (see p. 207), or the use of substrates containing the metabolic products of the same
or of another species, seem to be very recommendable also in this case.

Increased or decreased aeration may be equally helpful. Archangelski (1883), e. g.,
found that the small round reproductive organs (gonidia and regenerative bodies) of B. anthracis
multiplied readily as such in absence of air, but reproduced promptly rods in presence of air.
Bredemann (1909), on the other hand, had a constant aerobic culture of what he calls “micro-
oidia’”’ of the anaerobic B. amylobacter.

That the gonidia show much variation in their staining reaction has been emphasized in
Chapter II. Often the common staining, especially with fuchsin, is applicable; sometimes,
however, these bodies are very little inclined to take any stain at all. LZngol’s solution has
been found helpful in such cases by Kuntze (1904) and by Réditka (1908); but many authors
give preference to one or the other kind of vital staining. Some valuable hints upon the
principles of this method may be found in papers written upon this subject by Pfeffer (1886),
Campbell (1888), A. Fischel (1903), and Ph. Eisenberg (1913, pp.487-446). Methylene blue
has been used as early as in 1885 by Babes in his studies upon the motile “granules,’’ which he
discovered in the cells of V. cholerae, and it has been equally preferred by Zetinow (1897),
Nakanishi (1900 a, 1901), P. Ernst (1902), Menel (1910), Meirowsky (1914 6), and others.
Amaio (1908) recommended “Brillantkresylblau,”’ Balfour (1912), especially for blood prepa-
rates, toluidin blue, and Prazmowski (1912 p. 148) got most satisfactory results with a mixture
of methylene green and fuchsin. Identical results were obtained by Swellengrebel (1906)
when the vital staining was replaced by the Heidenhain method, applied to the dried prepa-
rate, fixed with 35 per cent formalin. And besides this, the various modifications of the Roman-
owsky-Giemsa method have been repeatedly found to be of special value for such studies, as
may be seen from the publications of Zetinow (1899), Feinberg (1900), Giemsa (1904-1914),
Herzog (1910), and of Betegh (1911), while A. Meyer (1912, p. 61) takes the opposite standpoint.
That the cytological theories and the microchemical methods of the last-named author (1. ¢.,
pp. 209-247) have not met with general approval was mentioned in Chapter II (p. 113).

A special staining method for gonidia has been recommended by Fedorowitsch (1902).
It is a modified Gram method and deserves further testing, as frequently the gonidia have
shown positive reaction with this method, even if the parent cells themselves are Gram-nega-
tive. Occasionally one or the other of the various methods may become applicable which
have been developed for staining the ‘‘granules”’ in diphtheria, tubercle, and leprosy organisms
by Lutz (1886, p. 87), Unna (1887, pp. 12, 15), A. Netsser (1888, p. 175), Spengler (1907),
Betegh (1908-1909), Knoll (1910), and Fontes (1910).

- The dark field method, which Meirowsky (1914 6, p. 15) and others found to be very use-
ful for such investigations, is not restricted to living preparates. Dried smears also often give
very good pictures, as was pointed out recently by Leishman (1918), and as we have shown
in the photograph reproduced as figure 234 on Plate XVIII (from Léhnis and Smith, 1916 a,
original fig. 42).

Because the gonidia often, though not always, display a slightly but distinctly higher
resistance than their parent cells against various deleterious influences, it is to be expected
that more detailed investigations will also provide us with means to separate the gonidia from
the vegetative cells. Observations made by Fuhrmann (1908) and others indicate, for instance,
that a comparatively high concentration of NaCl is much less detrimental to the gonidia than
to the cells, and the same seems to hold true in regard to a weak KOH solution, according to
some findings by Frank (1890) and by Jessen and Rabinowttsch (1910).

As the ordinary laboratory substrates very seldom allow the direct upgrowth of isolated
gonidia, the positive results obtained in many experiments upon so-called heterogenesis, where,
in fact, dying or dead plasma of fungi and algae, not altered by heating, proved to be a very
suitable medium for securing new upgrowth of bacteria from their gonidia, seem to promise
equally good results for new methods to be developed along these lines. In addition to what
has been said upon this subject on pages 148-150, it may be pointed out that especially
210 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVL

the algae experiments made by Dunbar (1907) and the blood tests of Fokker (1887, p. 72)
deserve to be carefully studied anew. Autolyzed tissue (liver, kidney, brain, etc.) may be
useful, too, in some cases; Couret and Walker (1913) obtained good growth of pure cultures
of parasitic amobae on it.

When experimenting with the filterable part of the gonidia, still further difficulties are to be
overcome. Faulty bougies or too long continued filtering under too high pressure are especially
liable to cause incorrect results. Doerr (1911) and Fontanel (1913) have published valuable
papers upon the technique, thus far developed, and its proper use. But especially in regard
to securing a new development of normal bacterial cells from the filtered gonidia practically
all remains to be done. However, results obtained by Nocard and Roux (1898), Nocard and
Leclainche (1903-1905, Vol. I, p. 449), Bordet (1910), Borrel et al. (1910), Almquist (1911-1917),
Hort (1917 a), and in our own experiments (Léhnis and Smith, 1916 a, p. 695), though still very
incomplete, call for a more thorough study of these problems. That occasionally colloid
granules may become a source of error was discussed by Huntemiiller (1916) and by Hallen-
berger (1917).

Observations upon the formation and germination of the lateral and terminal round
regenerative bodies and of arthrospores generally present no more difficulties than do analogous
experiments with endospores. As soon as their increased resistance is more completely
known, their separation from the vegetative cells by moderate heating or by thorough drying
will become an easy task. Some data recorded by Berestneff (1907); Rodella (1908), and by
Almquist (1916) are already quite satisfactory. Cultivation on serum favored sometimes
distinctly the pure growth of round regenerative bodies (Z. Klein, 1900; EH. de Negri, 1916).

Because in stock cultures the endospore formation often shows a tendency to vanish
more or less completely, repeated pasteurization may be used to prevent this change, as was
recommended by Beijerinck (1901 a). To suppress the spore formation experimentally num-
erous methods are available. Often the use of small amounts of antiseptics, first tried success-
fully by Rous (1890), will give the best results; in other cases cultivation in milk (Grassberger,
1902), adding of eosin to the substrates (Noguchi, 1908), keeping the cultures at nearly maximal
temperatures (Ph. Hisenberg, 1914) or other means may be preferable.

For the regeneration of endospore formation C. Phisalix (1892) found the cultivation in
blood most successful with B. anthracis, while Grassberger (1905) and Grassberger and, Schatten-
froh (1907) saw best results with their anaerobic butyric acid bacilli when sugar was added to the
substrates. As the regeneration of a lost spore formation is fundamentally not very different
from the new development of this faculty in a strain (or so-called species) which thus far has
appeared as being entirely and constantly asporogenous, the same method which promises the
best results in the latter case, viz.'a careful study and repeated heating of the intermediate
stages between regenerative bodies or gonidangia and endospores developing from the symplasm
will also undoubtedly in many cases secure quicker and always less haphazard results than the
methods first mentioned.

3, ad Ill: SYMPLASM.

For obtaining well developed bacterial ‘‘plasmodia” W. Winkler (1899) found it most
useful to place pieces of raw meat or potato into sour beer-wort, slightly acid bouillon,
or a decoction of mushrooms. After 24-36 hours the ‘‘plasmodia”’ were seen to grow forth and
soon after they began to form bacterial cells, which later separated themselves, On page 169
it has been mentioned that Klebs (1883 6) has directly observed under the microscope how
the symplastic masses of the tubercle bacilli sprouted out from cuts of tubercular tissue kept
in a well aerated chamber. In most cases it is by no means difficult to find out by daily testing
of cultures one to several weeks old when the formation of the symplasm reaches its maximum
and the regeneration of new cells begins to take place, provided that the points mentioned on
page 206, concerning suitable cultural conditions have not been overlooked.

Transfer of small amounts of the symplasm in the common manner, or after picking out
single flakes or clumps with one or the other of the various single-celt methods (discussed on
No. 2.] LIFE CYCLES OF THE BACTERIA—LOHNIS. 211

page 205, will occasionally give more conclusive results than the continued study of the material,
kept in the old substrate.

Observations made with the unstained living material in the hanging drop have usually
not given satisfactory results in the cases where I have employed this method. The irregular
surface and density of these clumps and flakes causes nearly always such abnormalities in
refraction and reflection of the light that the new formation of bacterial cells could not be accu-
rately followed in this way. Movements within the slimy masses increase these difficulties.
On the other hand the possibility can not be denied that occasionally good results will be
secured. Rizitka (1903), for instance, seems to have been more successful with this method.

Small flakes of symplasm, especially those formed by filterable gonidia, give very good
pictures in the dark field; but, as was pointed out by Leishman (1918), the strong light
seems to check all further development. Possibly some mode of vital staining will help to over-
come the difficulties standing in the way of studying the living material.

However, the photographs reproduced_on Plates XVIII to XXI will prove sufficiently, I
believe, that stained preparations can furnish many interesting data, especially when they are
made from the same culture in frequent intervals. That the staining qualities of the
symplasms vary widely has been discussed on pages 182-183, and on page 178 it was pointed
out that and why photomicrographs of such preparations are always rather inferior to the
original. According to Weigert (1875) hematoxylin has proved to be most useful for staining
bacterial ‘‘zoogloeas”’ within the tissue, but Klebs (1878) saw it act only upon the ‘‘granules”’
(the regenerative units) within the slime. Herzog (1913) found besides methylene blue, giving
the metachromatic reaction, Giemsa solution best suited, which stains the regenerative units
red, like the ‘‘eosinophilic’”’ granules in leucocytes. It is not impossible that one or the other
of the color reactions used for amyloid and hyalin substances, described by Lubarsch (1903),
may also find application in investigations upon bacterial symplasm.

The comparatively high resistance of the symplasm against acids and alkali, which W.
Winkler (1899) has made use of for separating bacterial ‘‘plasmodia”’ from surrounding bac-
terial cells, also allows, if necessary, a thorough cleaning of the fixed smear, especially the
solving of precipitates, which otherwise might disturb the picture. Great care in rinsing,
however, is essential, because, as Fokker (1882) has already noticed, the fixation of the
symplasm by heat, as well as by alcohol, is sometimes very incomplete.

Certain substrates, especially those made up from meat, peptone, etc., may often present
vexatious flakes and clumps, which react like genuine symplasm, but which are either coagula
from the ingredients used, or, though real symplasm, do not belong to the cultures under
experiment, but were produced by bacteria which have grown on the ingredients before they
were used for making the substrates. As with the regenerative bodies to be found in
many substrates, here also careful comparative tests of the steril media are indispensable.
Figure 297 on Plate XXIII shows such a doubtful flake from sterile beef agar (Léhnis and
Smith, 1916 b, original fig. 61). It looks exactly like some real symplasm, for instance, like
that of B. fluorescens, whose transformation into pale oval cells was illustrated in figure 263
on Plate XXI. Figure 251 on Plate XX (symplasm of rods and spores of B. azotobacter)
may serve as another object for comparison. Naturally, a final decision can be reached with
such dead or ‘‘fake’’ symplasm only by carefully ascertaining whether or not it is able to
produce new cells.

It is easily understood that transfers made at the same time from one culture, con-
taining large amounts of symplasm, into various substrates which are to be kept under
different conditions, will offer excellent opportunities for securing an accurate insight into
the characteristic behavior of a species. Careful, repeated heating of the symplasm and of the
regenerative bodies developing from it, has proved to be the most promising way for inducing
experimentally sporulation in a strain, which had never formed endospores before. That a
very good starting point for experiments on the effects due to symbiosis will be furnished by
cultures containing the bacteria in their symplastic stage has been mentioned above (p. 188);
212 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

and it was pointed out, too, that long-continued observations upon the same culture (in parallel
sets) will contribute much valuable material toward the solution of problems concerning ‘the
so-called involution forms, bacterial variation, mutation, etc.

4, ad IV: CONJUNCTION.

Hardly any special data are available at the present time as to the best manner of studying
bacterial conjunction. Experiments with the living material at the right time (usually from
the second to the fourth day) are most recommendable. Because the motility of the cells is
often a rather disturbing factor, their incarceration into the meshwork of fine cuts of elder
pith, as recommended by Fuhemann (1908), may be tried. Contact preparates give often, as
was shown by several of the photographs, much better pictures than do smears, wherein the
fine connections are easily destroyed. If by such tests the right moment has been fixed, the
study of the living material should be started. Occasionally vital staining may be helpful.
BIBLIOGRAPHY.

Aszotr, A. C. 1902. The principles of bacteriology.
6th ed. xz and 641 pp. 111 ill. Philadelphia
and New York. :

1912. On induced variations in bacterial func-
tions. An experimental study. In: Journ. Med.

Res., v. 26, No. 3, pp. 513-521.

and GILDERSLEEVE, N. 1902. On the actinomy-

ces-like development of some of the acid-resistant

bacilli (streptothrices ?). In: Centralbl. f. Bact.

I. Abt. Orig. v. 31, pp. 547-550, with 1 pl.

19038. On the branching occasionally
exhibited by Bacillus diphtheriae. In: Centrabl.
f. Bact.. I. Abt. Orig. v. 35, No. 3, pp. 273-280.

Aszort, M. E.
progressive portal cirrhosis. In: Journ. Pathol. and
Bact., v. 6, No. 3, pp. 315-326.

Avametz, L. 1895. Uber Micrococcus Sornthalii. In:
Centralbl. f. Bact. II. Abt. v. 1, No. 13/14, pp.
465-473, m. 1 Taf. att

Apami, J. G. 1892. On the variability of bacteria and

the development: of races.

v. 16, No. 6, pp. 366-389.

1894. How is variability in bacteria to be re-
garded? In: Publ. Health Papers and Rep., Amer.

Publ. Health Assoc., v. 20, pp. 415-418.

1916. On the life history of the Meningococcus
and of other bacteria. In: Brit. Med. Journ., No.
2911, pp. 525-526.

——., Axssotr, M. E., and Nicwoxson, F. J. 1899.
On the diplococcoid form of the colon bacillus.
In: Journ. Exper. Med., v. 4, pp. 349-372, with
4 pls.

Appario, ©. 1912. Uber das Vorkommen der Prowazek-
schen Kérper in den Follikeln und Papillen des
Trachoms. In: Centralbl. f. Bact. I. Abt. Orig.
v. 66, No. 5/6, pp. 393-405.

Aranassiew, 8. M. 1899. Uber einen aus dem Korper
eines Recurrenskranken erhaltenen Bacillus. In:
Centralbl. f. Bact. I, Abt. v. 25, No. 12, pp.
405-415.

Aprecut, H., und Gaon, A. 1900. Uber die Beulen-
pest in. Bombay im Jahre 1897. II. Wissenschaft-
licher Teil des Berichtes. C. Bakteriologische
Untersuchungen tiber den Pestbacillus. In: Denk-
schriften d. math.-naturw. Classe d. Akad. Wien,
v. 66, pp. 581-827, m. 6 Taf.

Auprecat, R. 1881. Beitrag zur Kenntnis der Ent-
wicklung der Spirochaete Obermeieri. In:
Deutsch. Arch. f. klin. Med., v. 29, No. 1/2, pp.
77-112.

Atmaquist, E. 1893. Zur Biologie der Typhusbakterie

und der Escherichschen Bakterie. In: Zeitschr.

f. Hyg., v. 15, pp. 283-290.

1904. Neue Erscheinungsformen des Cholera-

spirills und der Typhusbakterie. In: Centralbl. f.

Bact. I. Abt. Orig. v. 37, No. 1, pp. 18-28.

 

 

 

 

 

 

 

1900. On the bacteriology of a case of |;

In: Med. Chronicle,
fa m. 9 Abb. u. XXXV Taf.

 

1906. Cultur von pathogenen Bakterien
In: Zeitschr. f. Hyg., v. 52, No.

Atmautst, E.
in Diingerstoffen.

2, pp. 179-198.

1907. Neue Tatsachen zur Biologie der Typhus-
bakterie. In: Centralbl. f. Bact. I. Abt. Orig.
v. 45, No. 6, pp. 491-495.

1908. Studien iiber das Verhalten einiger
pathogenen Mikroorganismen bei niedriger Tem-
peratur. In: Centralbl.f. Bact. I. Abt. Orig. v.
48, No. 2, pp. 175-185.

1911. Studien tiber filtrierbare Formen in
Typhuskulturen. In: Centralbl. f. Bact. I. Abt.
Orig. v. 60, No. 3/4, pp. 167-174.

1916. Wuchsformen, Fruktifikation und Varia-
tion der Typhusbakterie. In: Zeitschr. f. Hyg.,
v. 83, No. 1, pp. 1-18, m. 5 Taf.

1917. Patogena bakteriers omdaning under
saprofytisk vaixande. In: Svenska Likaresillsk.
Handl., v. 43, No. 3, pp. 543-552, m. 2 tavl.

Attmann, R, 1894. Die Elementarorganismen und ihre
Beziehungen zu den Zellen. 2. Aufl. 160 pp.,

Leipzig.

Amato, A. 1908. Uber die feine Struktur der Bakterien.
In: Centralbl. f. Bact. I. Abt. Orig. v. 48, No. 4,
pp. 385-393, m. 2 Taf.

AmpBroZ, A. 1909. Entwicklungszyklus des Bacillus
nitrin. sp., als Beitrag zur Cytologie der Bakterien.
In: Centralbl. f. Bakt. I. Abt. Orig. v.51, No. 3,
pp. 193-226, m. 2 Taf.

——. 1915. Cytologische Beitrage zur Morphologie
und Aetiologie von sogenannten Involutions- und
Degenerationsformen der Bakterien. Vortrag, ge-
halten auf dem V. Kongress béhmischer Natur-
forscher und Arzte zu Prag (czechisch). Abstr.
in: Centralbl. f. Bact. II. Abt.. v. 44, No. 5/8,
p. 173.

Amoss, H.L. 1917. The cultivation and immunological
reactions of the globoid bodies in poliomyelitis.
In: Journ. Exper. Med., v. 25, No. 4, pp. 545-555,

; with 3 pls.

AmruscH, E, 1886. Uber eine Zoogloea-Form der
Tuberkel-Organismen. In: Med. Jahrb., v. 82,
pp. 291- 298, m. 1 Taf.

AnpReweEs, F. W. 1906. The evolution of the strepto-
cocci. In: Lancet, 1906, II, No. 4343, pp. 1415-1420.

ARCHANGELSKI, A. 1883. Ein Beitrag zur Lehre vom
Milzbrandcontagium. In: Centralbl. f. d. med.
Wissensch., v. 21, No. 15, pp. 257-261.

Artoine,S. 1908. Variations morphologiques du bacille

de la tuberculose de ’homme et des mammiferes,

obtenues artificiellement: In: Compt. rend. Acad.

Paris, v. 146, No. 3, pp. 100-104.

et CHANTRE, E, 1894. Infection purulente chi-
rurgicale, variations morphologiques et pathogénes

de son agent. In: La Semaine méd., v. 14, No. 22,

p. 179.

 

 

 

 

 

 

213
 

214

Aenpt, R. 1880. Beobachtungen an Spirochaete den-
ticola, der Spirochaete des Zahnschleimes. In:
Virchow’s Arch. f. path. Anat., v. 79, No. 1, pp.
76-86.

Arnuem, G. 1914. Spirochitenuntersuchungen. In:
Zeitschr. f. Hyg., v. 76, No. 8, pp. 407-442, m.
2 Taf.

Arrico, G.p’. 1900. Uber die Gegenwart und iiber die
Phasen des Koch’ schen Bacillus in den sogenannten
skrofulésen Lymphdriisen. In: Centralbl. f. Bact.
I. Abt. v. 28, No. 16, pp. 481-485, m. 1 Taf.

ArtTiaaLas, ©. 1885. Les microbes pathogénes. 260 pp.,
av.6pl. Paris et Bordeaux.

Arvustamow, M.J. 1889. Zur Morphologie und Biologie
der Leptothrix. In: Wratsch No. 3, 4 (Russian).
Abtr. in: Centralbl. f. Bact., v. 6, No. 13, p. 349.

ARZBERGER, E. G. 1910. The fungus root tubercles of
Ceanothus americanus, Elaeagnus argentea, and
Myrica cerifera. In: Ann. Rep. Missouri Bot.
Garden, v. 21, pp. 60-102, with 9 pls.

Atxinson, G. F. 1893. Contributions to the biology of
the organism causitig leguminous tubercles. In:
Bot. Gazette, v. 18, No. 5, p. 157-166; No. 6, pp.
226-237; No. 7, pp. 257-266, with 4 pls.

Axetrap, ©. 1903. Uber Morphologie der Colonien

pathogener Bakterien. In: Zeitschr. f. Hyg., v. 44,

No. 8, pp. 477-497, m. 3 Taf.

Bases, V. 1883. Comparaison entre les bacilles de la
tuberculose et ceux de la lépre. In: Compt. rend.
Acad. Paris, v. 96, No. 18, pp. 1323-1326.

1885. Untersuchungen iiber Koch’s Komma-

bacillus. In: Virchow’s Arch. f. path. Anat., v. 99,

No. 1, pp. 148-163.

 

———. 1889. Uber isoliert firbbare Antheile von Bak-’

terien.

In: Zeitschr. f. Hyg., v. 5, No. 2, pp.
173-196. :

 

rhagien verursachenden Bacillus bei Skorbut. In:
Deutsche med. Wochenschr., v. 19, No. 43, pp.
1035-1037.

1895. Beobachtungen tiber die metachroma-
tischen Kérperchen, Sporenbildung, Verzweigung,
Kolben- und Kapselbildung pathogener Bakterien.
In: Zeitschr. f. Hyg., v. 20, No. 3, pp. 412-437, m.
2 Taf.

 

—. 1903. DasMaltafieber.- In: Kolle-Wassermann’s | .

Handbuch der pathogenen Mikroorganismen, v. 3,
pp. 438-445.
—. 1907a. Lepra. In: Kolle-Wassermann’s Hand-
buch der pathogenen Mikroorganismen. I. Erg.
Bd. pp. 155-187, m. 1 Taf.

19076. Spindelférmige Bazillen. In: Kolle-
Wassermann’s Handbuch der pathogenen Mikro-
organismen. I. Erg. Bd. pp. 271-290, m. 1 Taf.
1907c.
Kérper and ihre Beziehungen zu dem Virus der
Wutkrankheit. In: Zeitschr. f. Hyg., v. 56, No. 3,
pp. 435-452, m. 2 Taf.

1908. Sur la formation des chainettes chez le
Staphylococcus aureus. In: Compt. rend. Soc.
Biol., v. 65, pp. 265-267.

1914. Uber metachromatische. Rieechien in
den acido-resistenten Bacillen. In: Berl. Klin.
Wochenschr., v. 51, No. 11, pp. 501-503.

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

1893. Uber einen die Gingivitis und Hamor-

 

 

 

Untersuchungen iiber die Negrischen.

 

[Vou. XVI.

Bases, V., et Levaprt1, C. 1897. Sur la forme actino-
mycosique du bacille de la tuberculose. In
Arch. méd. exp. et d’anat. path?, v. 9, No. 6,
pp- 1041-1048, av. 2 pl.

Barrtatzein, K. 1911. Uber Mutationserscheinungen
bei Bakterien. In: Centralbl. f. Bact. I. Abt.
Ref., v. 50, Beiheft, pp. 128-134.

——. 1912. Uber Mutationserscheinungen bei Bak-
terien. In: Arb. K. Gesundh. Amt., v. 40, No. 4,
pp. 433-536, m. 8 Taf.

1918a. Uber die Mutation bei Bakterien und

die Technik zum Nachweis dieser Abspaltungs-

vorginge. In: Centralbl. f. Bact. I. Abt. Orig.

v. 71, No. 1, pp. 1-13.

1913b. Uber Mutation bei Diphtherie. In:
Centralbl. f. Bact. I. Abt. Ref., v. 57, Beihett,
pp. 89-92.

——. 1918. Uhber bakterielle Variabilitat, insbeson-
dere sogen. Bakterienmutationen. In: Centralbl.
Bact. I. Abt. Orig. v. 81, No. 6, pp. 369-435.

Bam, TH. 1857. Uber Hefe. In: Flora, v. 40, No. 27,
pp. 417-430; No. 28, pp. 433-444, m. 6 Taf.

Basarpi, A. 1903. Die “Streptothrix lingualis’’ (Syn. »
Vibrio, Spirosoma linguale) im Munde der Gesun-
den und: der Diphtherischen. In: Centralbl. f.

 

Bact. I. Abt. Orig. v. 35, No. 2, pp. 129-137,
m. 1 Taf. .
Batrour, A. 19lla. The infective granule in certain

protozoal infections, as illustrated by the spiro-
chaetosis of Sudanese fowls (preliminary note),
In: Brit. Med. Journ., 1911, I, No. 2622, p. 752.

19116. Spirochaetosis of Sudanese fowls. In:
Fourth Rep. Wellcome Trop. Res. Labor. Vol. A:
Medical, pp. 76-107, with 2 pls.
191lc. Fallacies and puzzles in blood examina-
tion. In: Fourth Rep. Wellcome Trop. Res.
Labor. Vol. A: Medical, pp. 109-125, with 3 pls.
. 191ld. Diphtheria in the Tropics. In: Fourth
Rep. Wellcome Trop. Res. Labor. Vol. A: Medical,
pp. 239-245, with 1 pl.

1912. The value of vital blood staining in the

study of the so-called “infective granules’ (pre-

liminary note). In: Brit. Med. Journ..1912, I, No

2668. p. 362.

1913. A contribution to the life history of
Spirochaetes. In: Centralbl. f. Bact. I. Abt.
Orig. v. 70, No. 3/4, pp. 182-185.

Batt, M. V. 1908. Essentials of bacteriology. 6th ed.
280 pp., with 135 illustr. Philadelphia and London.

Barser, M. A. 1907. On heredity in certain micro-

organisms. In: Kansas University Science Bulle-

tin, v. 4, No. 1, pp. 1-48, with 4 pls.

———. 1908. The rate of multiplication of B. coli at
different temperatures. In: Journ. Infect. Dis-

eases, v. 5, No. 4, pp. 379-400.
1911. A technic for the inoculation of bacteria
and other substances into living cells. In: Journ.
Infect. Diseases, v. 8; No. 3, pp. 348-360.
Barcum. 1903. Mastzellen. In: Encyklopaedie d.
mikrosk.Technik, herausgeg. v. P. Ehrlich u. a.
1400 pp., 184 Textabb. Berlin und Wien.

Barker, B. T. P.. 1901. A. conjugating “yeast.” In:
Philos. Transact. Roy. Soc. (London), Ser. B,
v. 194, No. 203, pp. 467-485, with 1 pl.

 

 

 
No. 2.] LIFE CYCLES OF THE

BarnarpD, J. E. 1899. Photogenic bacteria. In:
Transact. Jenner Inst. Prevent.. Medic., v. 2,
pp. 81-112, with 2 pls. |

Barranixow, J. 1900. Beitrag zur Bakteriologie der
Lepra. In: Centralbl.f. Bact. I. Abt. v. 27, No.
20/21, pp. 709-710.

——. 1902. Zur Kenntnis der saiurefesten Mikroben.
In: Centralbl.f. Bact. I. Abt. Orig. v. 31, No. 7,
pp. 282-285.,

Bastian, H.C. 1872. The beginnings of life. v. I:
xxxv and 475 pp., with 38 ill. v. II: 1x and 640
pp., with 96 ill. Append. A-E: civ pp., with 6
ill. New York.

1908. Studies in heterogenesis. 1x and 352 pp.,
with x1x pls. Append. 1-nt and Index xxxvii pp.
London and Oxford.

——. 1905. The nature and origin of living matter.
344 pp., with xxxu pls. ‘Philadelphia and London.

——. 1907. The evolution of life. xvuiand 319 pp.,
with xi pls. and 12 figs. in the text. New York.

——. 1914. The production at will of either fungus

 

germs, flagellate monads, or amoebae from the ulti- |

mate segments of small masses of zoogloea.
Nature, v. 94, No. 2356, pp. 462-468.

Baupet, E. A. R. F. 1911. Asporogene Milzbrand-
bacillen. In: Centralbl. f. Bact. I. Abt. Orig.
v. 60, No. 6, pp. 462-480.

‘Baumearten, P. 1890. Lehrbuch der pathologischen
Mykologie. 1x and 973 pp., m. 101 Abb. Braun-

_ schweig.

Bauma@artner, E. 1909. Bacterium iogenum. In:
Deutsche Monatsschr. f. Zahnheilk., v.27, pp. 764-
771.

In:

Baur, E. 1905. Myxobacterien-Studien. In: Arch. f..

Protistenk., v. 5, No. 1, pp. 92-121, m. 1 Taf.
Bayon, H. 1915. Leprosy: A perspective of the results
of experimental study of the disease. In: Annals

of Trop. Med. and Parasitol., v. 9, No. 1, pp. 1-90,

with 6 pl.

Biécuamp, A.’ 1883. Les microzymas dans leur iscsouit

avec l’hétérogénie, l’histogénie, la physiologie et la |

pathologie. xxxvut and 992 pp., av. 5 pls. Paris.
Brecx, M. 1903. Diphtherie. In: Kolle-Wassermann’s

Handb. d. pathog. Mikroorg., v. 2, pp. 754-831.
BryveErincx, M. W. 1888. Die Bacterien der Papiliona-

ceenknéllchen. In: Botan. Zeitg., v. 46, No. 46,

pp. 725-735; No. 47, pp. 741-750; No. 48, pp. 757-,
' 771; No. 49, pp. 781-790; No. 50, PP: 797-804, m.1)

Tat.

1889. Le Photobacterium luminosum. In:

 

Arch, néerland. des sci. exact. et natur., v. 23,

pp. 401-415.
——. 1890. Kiinstliche Infection von Vicia Faba mit
Bacillus radicicola. Ernihrungsbedingungen dieser

Bacterie. In: Botan. Zeitg., v. 48, No. 52, pp. 837~

843.

. 1894. Uber die Natur der Faden der Papiliona-
ceenknéllchen. In: Centralbl. f. Bact., v. 15,
No. 19/20, pp. 728-782.

——. 1897. Emulsions- und Sedimentfiguren bei
beweglichen Bakterien.

II. Abt. v. 3, No.1, pp. 1-5; No. 2/3; pp. 40-47.

1898. Uber die Arten der Essigbakterien. In:
Centralbl.f..Bact. 1. Abt. v.4, No.6, pp. 209-216.

 

 

-BACTERIA—LOHNIS.

 

 

In: Centralbl. f. Bact. |

 

215

Bewerinck, M. W. 1901 a. Anhaufungsversuche mit
Ureumbakterien. In: Centralbl. f. Bact. mu. Abt.
v. 7, No. 2, pp. 33-61, m. 1 Taf.

1901 b. Uber oligonitrophile Mikroben. In:

Centralbl. f. Bact. u. Abt. v. 7, No. 16, pp. 561-

582, m. 1 Taf.

1914 a. Uber Schréter und Cohn’s Lakmusmi-

crococcus. In: Folia microbiol., v. 2, No. 3, pp.

185-200, m. 2 Taf.

. 19146. Uber das Nitratferment und iiber physio-

logische Artbildung. In: Folia microbiol., v. 3,

No. 2, pp. 91-113, m. 1 Taf.

und Detpen, A. vAN. 1902. Uber die Assimila-
tion des freien Stickstoffs durch Bakterien. In:
Centralbl. f. Bact. m. Abt. v. 9, No. 1/2, pp. 1-43.

Berrzxe, H. 1904. Uber die fusiformen Bacillen. In:
Centralbl. f. Bact. I. Abt. Ref., v. 35, No. 1/2,
pp. 1-15.

BENECKE, W. 1912. Bau und Leben der Bakterien.
xir and 650 pp., m. 105 Abb. Leipzig und Berlin.

Brrzstnew, N. 1898. Uber Pseudoaktinomykose. In:
Zeitschr. f. Hyg., v. 29, pp. 94-116, m. 3 Taf.

. 1907. Uber die Lebensfihigkeit der Sporen von

Strablenpilzen. In: Centralbl. f. Bact. 1. Abt.
Ref., v. 40, No. 9/10, p. 298.

Brereey, D. H. 1907. Involutions- und Degenerations-
formen. In: Centralbl. f. Bact. 1. Abt. Ref., v
40, No. 11/12, pp. 360-361.

Berestranp, H. 1918. Uber den ‘‘Klebs-Léfflerschen
Bazillus.’’? In: Acta Oto-Laryngologica, v.1, No. 1,
pp. 131-186, 6 Taf.

Brrnuarpt, G. 1911. Experimentelle Untersuchungen
tiber die Scharlachitiologie. In: Deutsche Med.
Wochenschr., v. 37, No. 17, pp. 791-792; No. 23,
pp. 1062-1063.

. 1912. Beitrige zur Morphologie und Biologie der

Ruhrbakterien. In: Zeitschr. f. Hyg., v. 71, No.2,

pp. 229-248.

1915. Uber Variabilitit pathogener Bakterien.

In: Zeitschr. f. Hyg., v. 79, No. 2, pp. 179-248.

und Panera. 1913. Uber die Variabilitat des

Diphtheriebacillus. In: Centralbl.f. Bact.. I. Abt.

Ref., v. 57, Beiheft, pp. 83-89, m. 1 Taf.

Bernuem, J., und Foraer, C. 1896. Uber verzweigte
Diphtheriebacillen. In: Centralbl. f. Bact. I. Abt.
v. 20, No.1, pp. 1-3, m. 1 Taf.; No. 25, pp. 905-906.

Berry, J. L., and Banzuar, E.J. 1912. Nonvariability
of diphtheria bacilli. In: Journ. Infect. Diseases,
v. 10, No. 3, pp. 409-415.

Bertranp, D.M. 1913. Etude d’un bacille lactique de
Vappareil digestif du faisan. In: Compt. rend.
Soc. Biol., v: 74, pp. 96-97.

Brrrex, L. von. 1908. Neue differentialdiagnostische

Farbemethode fiir Tuberkel-, Perlsucht- und an-
dere siurefeste Bacillen, nebst Strukturstudien
bei verschiedenen siurefesten Bakterienarten. In:
Centralbl. f. Bact. I. Abt. Orig. v. 47, No. 5, pp.
654-664.

——. 1909. Uber eine neue Methode zur Darstellung
der Tuberkelbacillensporen. In: Centralbl. f.
Bact. I. Abt. Orig. v. 49, No. 3, pp. 461-463.

——. 1911. Beitrige zur Aetiologie der Maul- und
Klauenseuche. In: Centralbl. f. Bact. I. Abt.
Orig. v. 60, No. 1/2, pp. 86-90, m. 2 Taf.

 

 

 

 

 

 
216

Beust, To. Fear. von. 1907. Leptothrix buccalis. In:
Centralbl. f. Bact. I. Abt. Ref., v. 40, No. 4,

¢ p. 114.

Briepert, Px. 1885. Beitrag zur Frage nach der Constanz
der Spaltpilze (Kokkobazillus zymogenus und
Bacterium Termo). In: Virchow’s Arch. f. pathol.
Anat., v. 100, No. 4, pp. 439-459.

und SiezL, G. 1884. Chronische Lungenentziin-
dung, Phthise und wmiliare Tuberculose. In:
Virchow’s Arch. f. pathol. Anat., v. 98, No. 1, pp.
91-159, m. 1 Taf,

Buet, A. 1890. Contribution 4 1l’étude de la mor-
phologie et du développement des bactériacées. In:
Bull. scient. de la France et de la Belgique, v.
21, pp. 1-288, avec 9 pls. et 19 figs. dans le texte.

Burots, Ta. 1874. Untersuchungen tiber die Vegeta-
tionsformen der Coccobacteria septica. xiv and
244 pp., m. 5 Taf. u. 1 Fig.im Text. Berlin.

Brrrrotrr, R. 1912. Zur Morphologie des Choleravibrio.
In: Centralbl. f. Bact. I. Abt. Orig. v. 66, No.
7, pp. 496-500.

Buav, O. 1905. Uber die Temperaturmaxima der

 

Sporenkeimung und der Spotenbildung, ete. In: |

Centralbl. f. Bact.
97-143, m. 1 Taf.

BueseneR. 1901. Beitrag zur Lehre von der Sporen-
bildung bei Cholerabacillen. In: Zeitschr. f. Hyg.,
v. 36, pp. 71-74.

Buiiiuporn, K. 1910. Untersuchungen iiber den Bacil-
lus bifidus communis und den sogenannten Bac.
acidophilus (Streptobacillus faecalis). In: Jahrb.
f. Kinderheilkunde v. 72, p. 693-704,

Béna, W. 1912. Uher Zelleinschliisse bei Trachom und
Conjunctivitiden. In: Arb. K. Gesundh. Amt, v.
40, No. 2, pp. 235-244. ;

Bouunerr, A. 1877. Uber eine neve Pilzkrankheit
beim Rinde. In: Centralbl. f. d. medic. Wis-
sensch., v. 15, No. 27, pp. 481-485.

Bonazzi, A. 1915. Cytological: studies of Azotobacter
chroococcum. In: Journ. Agric. Res., v. 4, No. 3,
pp. 225-239, with 2 pl.

Bongert. 1901. Corynethrix pseudotuberculosis mu-
rium, ein neuer pathogener Bacillus fiir Mause. In:
Zeitechr. f. Hyg., v. 37, pp. 449-475 m. 2 Taf.

Bonnorr, H. 18962. Untersuchungen iiber Vibrionen
und §Spirillen. I. Vorl. Mitt. In: Hyg. Rund-
schau, v. 6, ‘No. 8, pp. 351-356.

II. Abt. v. 15, No. 4/6, pp.

 

Spirillen. I. In: Arch. f. Hyg., v.26, No. 2/3, pp.
162-202, m. 1 Taf.

— —. 1905. Die Spirochaeta vaccinae. In: Berl.
| Bruns, H. 1895. Ein Beitrag zur Pleomorphie der

Klin. Wochenschr., v. 42, No. 36, p. 1142-1144.
Borpet, J. 1910. La morphologie du microbe de la

péripneumonie des bovidés. In: Ann. Inst. Pas- |

teur, v. 24, No. 3, pp. 161-167.
et StezswyK. 1910. Sérodiagnostic et varia-

 

pilité des microbes suivant le milieu de culture. |

In: Ann. Inst. Pasteur, v. 24, No. 6, pp. 476-494.
Borpont-Urrrepuzzi, G. 1888a. Uber die Cultur der
Leprabacillen. In: Zeitschr. f. Hyg., v. 3, pp.
178-188, m. 5 Taf.
——. 1888b. Uber den Proteus hominis capsulatus
und iiber eine neve durch ihn erzeugte Infections-

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

1896b. Untersuchungen iiber Vibrionen und |

 

[Vou. XVI.

krankheit des Menschen. In: Zeitschr. f. Hyg.,
v. 3, pp. 333-354, m. 2 Taf.

Borret, Dusarpin-BEAUMETZ, JEANTETetJouan. 1910.
Le microbe de la péripneumonie. In: Ann. Inst.
Pasteur, v. 24, No. 3, p. 168-179.

BosAnaquet, W. C. 1911. Brief notes on the structure
and development of Spirochaeta anodontae Keys-
selitz. In: Quart. Journ. Microsc. Science, n. s.,
v. 56, No. 222, pp. 387-393, with 1 pl.

Bostroem. 1890. Untersuchungen tiber die Aktinomy-
kose des Menschen. In: Beitr. z. pathol. Anat.
u, allg. Pathol., v. 9, No. 1, pp. 1-240, m. 10 Taf.

Borromizy, W. B. 1912. The root-nodules of Myrica
Gale. In: Annals of Botany, v. 26, No. 1, pp. 111-
117, with 2 pls.

Brazm, C. 1889. Untersuchungen tiber die Degenera-
tionserscheinungen pathogener Bakterien im destil-
lierten Wasser. In: Beitr. z. pathol. Anat. u.
allg. Pathol., v. 7, No. 1, pp. 11-46.

Branpb, F. 1901. Bemerkungen itiber Grenzzellen. In:
Ber. d. Deutschen Bot. Ges., v. 19, No. 3, pp.
152-156. oe

Branot, R. 1913. Beitrag zur Kenntnis der Morphologie
oxydierender Bakterienfermente. In: Centralbl. f.
Bact. I. Abt. Orig. v. 72, No. 1/2, pp. 1-22, m.
1 Taf.

Brkavpat, L. 1906. Sur un nouveau microbe pro-
ducteur d’acétone. In: Ann. Inst. Pasteur, v. 20,
No. 10, pp. 874-879.

Brepemann, G. 1909. Bacillus amylobacter A. M. et
Bredemann in morphologischer, physiologischer und
systematischer Beziehung. In: Centrabl. f. Bact.
II. Abt. v. 23, No. 14/20, pp. 389-568, m. 2 Taf.

BreFetp, O. 1881. Bacillus subtilis. In: Bot. Unters.
tb. Schimmelpilze, Heft IV, pp. 36-54.

———. 1908. Untersuchungen aus dem Gesamtgebiete
der Mykologie, v. XIV.

Brett, A. 1907. On the morphology and life history of
Spirochaeta Duttoni. In: Ann. Trop. Medic. and
Parasitol., v. 1, No. 3, pp. 433-438, with 1 pl.

and Kineporn, A. 1906. An experimental study
of the parasite of the African tick fever (Spirochaeta
Duttoni). In: Mem. Liverpool School of Trop.
Medic., No. 21, pp. 1-52.

Broapuurst, J. 1915. Environmental studies of strep-
tococci with special reference to fermentative re-
actions. In: Journ. Infect. Diseases, v. 17, No. 2,
pp. 277-330.

Bruncnorst, J. 1885. Uber die Knéllchen an den
Leguminosenwurzeln, In: Ber. d. Deutschen Bot.
Ges., v. 3, pp. 241-257.

 

‘Tuberkelbacillen. Diss. med. Strassburg. 24 pp.,
m. 1 Taf.

Bucuanan, R. E. 1909. The bacteroids of Bacillus
radicicola. In: Centralbl. f. Bact. II. Abt., v. 23,
No. 1/5, pp. 59-91.

——. 1918. Life phases in a bacterial culture. In:
Journ. Infect. Diseases, v. 23, No. 2, pp. 109-125.

Buowner, H. 1882a. Uber die experimentelle Erzeu-
gung des Milzbrandcontagiums. I. u. II. Mitt.
In: Nageli’s Unters. iib. nied. Pilze, pp. 140-177,
186-204.
LIFE CYCLES OF THE

No. 2.]

Bucuner, H. 1882b. Beitrige zur Morphologie der Spalt-
pilze. In: Niageli’s Unters. tib. nied. Pilze, pp.
205-224.

——. 1882c. Kritisches un Experimentelles tiber die
Frage der Constanz der pathogenen Spaltpilze.
In: Nageli’s Unters. tib. nied. Pilze, pp. 231-285.

——. 1885. Beitriige zur Kenntnis des Neapler Cho-
lerabacillus und einiger demselben nahestehender
Spaltpilze. In: Arch. f. Hyg., v. 3, pp. 361-442,
m. 1 Taf.

——. 1888. Uber die vermeintlichen Sporen der
Typhusbacillen. In: Centralbl. f. Bact., v. 4, No.
12, pp. 353-358; No. 13, pp. 385-390.

Btseren, M. 1894. Culturversuche mit Cladothrix di-
chotoma. In: Ber. d. Deutschen Botan. Ges., v.
12, No. 6, pp. 147-152, m. 1 Taf.

Birscmu, O. 1890. Uber den Bau der Bakterien und

verwandter Organismen. 37 pp.,m.1Taf. Leipzig.

1896. Weitere Ausfiihrungen iiber den Bau der
Cyanophyceen und Bakterien. 87 pp., m. 5 Taf.

u. 6 Textfig. Leipzig.

Butz, C.G. 1917. The pathogenic effect of streptococci
from cases of poliomyelitis and other sources. In:
Journ. Exp. Med., v. 25, No. 4, pp. 557-580, with
7 pls.

Bunaet, R. 1895. ther Searching bei Bacterien.
In: Fortschr. d. Med., v..13, No. 20, pp. 813-826;
No. 21, pp. 853-863, m. 1 Taf.

Bouncer, J. W. M. 1917. Some effects of hydrogen ion
concentration of media on the behavior of the diph-
theria bacillus in culture.
No. 1, pp. 31-32.

Burcxuarpt, J. L. 1914. Untersuchungen tiber Bewe-
gung und Begeisselung der Bakterien, etc. I. Teil.
In: Arch. £. Hyg., v. 82, No. 5/7, pp. 235-320.

Burpon-SanpeRson. 1871. The origin and distribution
of microzymes (bacteria) in water, etc. In: Quart.
Journ. Micr. Science, n. s., v. 11, pp. 323-352.

Burnet, E. 1911. Microbes et toxines. x1 and 349 pp.
av. 71 figs. dans le texte. Paris.

Burr, R. 1897. Aromabildende Bakterien im Emmen-

taler Kise. In: Centralbl.f. Bact. II. Abt. v. 3,

No. 23/24, pp. 609-615.

1898. Uber das Vorkommen relativ grosser
Bakterienkolonien in fehlerhaftem Emmentaler
Kase. In: Centralbl. f. Bact. II. Abt. v. 4, No.

15/16, pp. 608-615.

——. 1909. Das Tuscheverfahren, etc. 42 pp., m. 3
Taf. Jena.

Burrows, M. T., and Neymann, C. A. 1917. Studies
on the metabolism of cells in vitro. I. In: Journ.

obs Exp. Med., v.25, No. 1, pp. 93-94.

Burton, L. V., andl Rerteer, L. F. 1917. Correlations
in the Colon-Aerogenes group. In: Journ. Infect.
Diseases, v. 21, No. 2, pp. 162-195.

Cacace. 1903. Sulla variazioni morfologiche del bacillo
della peste. In: Giorn. dell’Assoc. Napol., v.12,
No. 2. Abstr. in: Centralbl. f. Bact. I. Abt.
Ref., v. 34, No. 8/9, p. 242.

Cacue, A. 1901. De la culture du bacille de diphthérie
croissant en fils ramifiés. In: Centralbl. f. Bact.
I. Abt. v. 29, No. 25, pp. 975-980.

 

Abstr. of Bact. Vol. 1,

 

217

1901. Uber die nach Gram farbbaren Bacillen
des Sauglingsstuhles. In: Centralbl. f. Bact. I.
Abt. v. 30, No. 19, pp. 721-726.

Cauxins, G. N. 1904. The life history of Cytorrhyctes
variolae Guarnieri. In: Journ. Med. Res., v. 11,
No. 1, pp. 136-172, with 5 pls.

CatmetTE, E. 1893. De la présence d’un microorga-
nisme dans le sang, les crachats et les urines des
malades atteints de typhus exanthématique. In:
Annal. de microgr., v. 5, No. 2, pp. 85-104, av. 1 pl.

CampBeLL, D. H. 1888. The staining of living nuclei.
In: Unters. a. d. Bot. Inst. Tiibingen, herausgeg.
v. W. Pfeffer, v. 2, No. 3, pp. 569-581.

Cano, U. 1909. Untersuchungen tiber die Verbreitung
der ultramikroskopischen Keime in der Natur. In:
Centralbl. f, Bact. I. Abt. Orig. v. 49, No. 1,
pp. 78-79.

CantacuziNne, J. 1898. Nouvelles recherches sur le
mode de déstruction des vibrions dans l’organisme.
In: Annal. Inst. Pasteur, v. 12, No. 4, pp. 272-300.

CappeLtant,S. 1910. Sulle ramificazioni del bacillo di
Loeffler. In: Ann. d’Igiene sperim., n. s., v. 20,
No. 3, p. 301. Abstr. in: Centralbl.f. Bact. I. Abt.
Ref., v. 48, No. 25, p. 772. —

Carpano, M. 1913. Beitrag zur Kenntnis des B. mallei.
Morphologisches und Biologisches. In: Centralbl.
f. Bact. I. Abt. Orig. v. 71, No. 4, pp. 267-285,
m. 3 Taf.

——~. 1914. Die nekrotisch-gangranédsen Affectionen
in der Veterinirpathologie. Die fuso-spirillire
Symbiose. In: Centralbl. f. Bact. I. Abt. Orig.
v. 74, No. 3/4, pp. 225-241, m. 1 Taf.

CasaGranpi, 0. 1897. Uber die Morphologie der Blas-
tomyceten. In: Centralbl.f.Bact. II. Abt. v.3,
No. 23/24, pp. 634-639; No. 25/26, pp. 718-722.

——. 1901. Sulla relazioni tra bacteri proto, meta e
paratrofi, etc. In: Ann, d’Igiene sperim., n. 8., v.
11, No. 2, p. 161 con tav. Abstr. in: Baumgarten’s
Jahresber., v. 17, p. 804.

Caspari,G. 1902. Uber die Konstanz der Sporenkei-
mung bei den Bacillen und ihre Verwendung als
Merkmal zur Artunterscheidung. In: Arch. f.
Hyg., v. 42, No. 1/2, pp. 71-106, m. 1 Taf.

Casrzntani, A. 1905. Further observations on parangi

' (yaws). In: Brit. Med. Journ., 1905. 1, No. 2342,

pp. 1330-1331.

Cuantemesse, A., Matrucnor, L., et Grimpzre, A.
1917. Un microbe nouveau, Mycobacillus synovi-
alis,etc. In: Compt. Rend. Acad.,. Paris, v. 164,
No. 17, pp. 652-655.

Cuarrin et Puisatrx. 1892. Abolition persistente de la
fonction chromogéne du Bacillus pyocyaneus. In:

Comp. rend. Acad. Paris, v. 114, No. 26, pp. 1565-

1568.

et Roczr, G. H. 1887. Des modifications qu’on

peut provoquer dans les fonctions d’un microbe

chromogéne. In: Compt. rend. Soc. Biol., v. 39,

pp. 596-598.

CuatrensEE, G. C. 1910. A new lactic acid producing

Streptothrix, found in the fermented milk of India,

called the dahdi. In: Centralbl. f. Bact. I. Abt.

Orig. v. 58, No. 2, pp. 103-112.

BACTERIA—LOHNIS.

CAHN.

 
218

Cuaron, E., et Pérarp, Cx. 1913. Schizophytes du
caecum du cobaye. II. Metabacterium polyspora

n. g.,n. sp. In: Compt. rend. Soc. Biol., v. 74,

pp. 1232-1234.
Cuauveat, A., et Pursarix, C. 1893. Contribution a
_ étude de la variabilité et du transformisme en
microbiologie, 4 propos d’une nouvelle variété de
bacille charbonneuse (B. anthracis claviformis).
In: Comp. rend. Acad. Paris, v. 120, No. 15, pp.
801-807.
CuEesnEY, A.M. 1916. The latent period in the growth
, of bacteria. In: Journ. Exp. Med., v. 24, No. 4,
pp. 387-418.
Cuester, F.D. 1901. Manual of determinative bacteri-
' ology. vrand 401 pp., with 13 figs. New York.
Curistensen, H.R. 1903. Zwei neue fluoreszierende
Denitrificationsbakterien.
II. Abt. v.11, No. 6/7, pp. 190-194, m. 2 Taf.
Ciznxowsk«1, L. 1877. Zur Morphologie der Bacterien.
“In: Mém. Acad. St. Pétersbourg., 7e sér., v. 25, No.
2, 18 pp., av. 2 pls.

CuaypoLz, E. J. 1913. On the classification of the |

Streptothrices, particularly in their relation to
bacteria. In: Journ. Exp. Med., v. 17, No. 1, pp.
99-116, with 3 pls.

Cocarpas, E. 1884. Idées nouvelles sur la fermenta-
tion. II. Le Penicillium ferment dans les eaux
‘distillées pharmaceutiques. In: Journ. de micro-
graphie, v. 8, No. 7, pp. 390-397, av. 1 pl.

Coun, F. 1853. Untersuchungen iiber die Entwicke-.

lungsgeschichte der microscopischen Algen und
Pilze. In: Nova acta Acad. Caes: Leopold.-Caro-
linae Nat. cur., v. 24, I, pp. 101-256, m. 6 Taf.

———. 1870. Uber den Brunnenfaden. In: Beitr. z.
Biol. d. Pflanzen, v. 1, Heft 1, pp. 108-132, m. 1 Taf.

——. 1872a. Organismen in der Pockenlymphe. In:
Virchow’s Arch. f. pathol. Anat., v. 55, No. 1/2, pp.
229-238.

——. 1872b. Untersuchungen iiber Bacterien. In:
Beitr. z. Biol. d. Pflanzen, v.1, Heft 2, pp. 127-224,
m. 1 Tai.

1875. Untersuchungen tiber Bacterien. II. In:

Beitr. z. Biol. d. Pflanzen, v. 1, Heit 3, pp. 141-207,
m. 2 Taf.

———. 1876. Untersuchungen iitberBacterien. IV. In:
Beitr. z. Biol. d. Pflanzen, v. 2, Heft 2, pp. 249-276.

Coun, Ta. 19138. Ein Aktinomycet der minnlichen
Urogenitalorgane. In: Centralbl.f. Bact. I. Abt.
Orig. v. 70, No. 5/6, pp. 290-306, m. 6 Taf.

Cots, L. J., and Wrient, W. H. 1916. Application of
pure-line concept to bacteria. In: Journ. Infect.
Diseases, v. 19, No. 2, pp. 209-221.

Concerti, L. 1901. Wher die aktinomykotische Form
des Loefflerschen Bacillus in gewissen Zustéinden
saprophytischen Lebens. In: Arch. f. Kinderheil-
kunde, v. 31, No. 3/4, pp. 227-246.

Conn, H. W. 1900. Natural varieties of bacteria. In:
Journ. Boston Soc. Med. Sciences, v. 4, No. 6, p. 170.

——. 1909. Agricultural bacteriology. 2ded. xand
331 pp., with 64 figs. Philadelphia.

Conrapl, H. 1900. Die Hyphomycetennatur des Rotz-
bacillus. In: Zeitschr. f. Hyg., v. 33, pp. 161-176,
m. 2 Taf.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

| Cornet, G., und Meyer, A. 1903. Tuberkulose.

In: Centralbl. f. Bact. |

 

[Vor XVI,

Corren-Jonzs, A. 1893. Uber einen neuen, bei Tuber-
kulose haufigen Fadenpilz. Vorl. Mitt. In: Cen-
tralbl. f. Bact., v. 13, No. 21/22, pp. 697-701, m. 1
Taf.

———. 1895. Uher die Morphologie und systematische

Stellung des Tuberkelpilzes und tiber die Kolben-

bildung bei Aktinomykose und Tuberkulose. In:

Centralbl. f. Bact. I. Abt. v.17, No. 1, pp. 1-16;

No. 2/3, pp. 70~76, m. 1 Taf.

1896. Uber die Nomenklatur des sogenannten

“Tuberkelbacillus.”” In: Centralbl. f. Bact. 1.

Abt. v. 20, No. 10/11, pp. 393-395.

In:
Kolle-Wassermann’s Handb. d. pathogen. Mikroorg.,
v. 2, pp. 78-177.

Corn, A.-V., et AtvArEz. 1885. Mémoire pour servir
& Vhistoire du rhinosclérome. In: Arch. physiol.
norm. et pathol. 3™¢ sér., v. 6, No. 1, pp. 1140,
av. 1 pl.

, et Basses, V. 1890. Les bactéries et leur réle,
etc. 3™¢ éd. v. 1: vm et 582 pp. v.2: 608 pp.,
385 figs. dans le texte et 12 pl. Paris.

Costa, 8S. 1909. Mobilité du bacille fusiforme de Vin-
cent. In: Compt. rend. Soc. Biol., v. 67, No. 24,
pp. 866-867.

Couret, M., and Watker, J. 1913. The cultivation of
amoebae in pure culture upon autolyzed tissues.
In: Journ. Exp. Med., v. 18, No. 3, pp. 252-258.

Courmont, P., et Descos, A. 1902. Cultures liquides
homogénes et mobilité des bacilles ‘‘acido-résis-
tants”. In: Compt. rend. Soc. Biol., v. 54, pp.
1355-1357.

Cozzouino, V., 1900. Ein neues Fadenbacterium, eine
pseudoactinomycotische Erkrankung erzeugend.
Zeitschr. f. Hyg., v. 33, No. 1, pp. 36-52, m. 1 Taf.

Crata, Cu. F, 1898. The branched form of the Bacillus
tuberculosis in sputum. In: Journ. Exp. Med.,
v. 3, pp. 363-370, with 1 pl.

CrooxsHank, E. M. 1896. Textbook of bacteriology.
4th ed. xxxand 715 pp., 273 figs. in the text and
22 pls.

CunnineHam, D.D. 1897. Choleraic and other commas;
on the influence of certain conditions in determin-
ing morphological variations in vibrionic organisms.
In: Scient. Mem. Med. Officers of the Army of
India, v. 10, pp. 1-29.

CzarLewsk1, E. 1898. Uber einen aus einem Leprafalle
geziichteten alkohol- und siurefesten Bacillus aus
der Tuberkelbacillengruppe. In: Centralbl. f. Bact.
I. Abt. v. 23, No. 3/4, pp. 97-107; No. 5/6, pp.
189-194, m. 2 Taf.

Date, E. 1912. A bacterial disease of potato leaves. In:
Annals of Botany, v. 26, No. 101, pp. 1338-154, with
2 pls.

Datz, J. 1910. Uber eine ungewéhnliche Form des
Diphtheriebacillus. In: Centralbl. f. Bact. I.
Abt. Orig. v. 56, No. 5/6, pp. 401-410.

DaneearD, P.A. 1891. Contribution 41’ étude des Bac-
tériacées vertes. In: Le Botaniste, v. 2, No. 4, pp.
151-160, av. 1 pl.

Darwin, Co. 1868. The variation of animals and plants
under domestication. American ed. by A. Gray,
v. 2, vir and 568 pp. New York.

 
No. 2.] LIFE CYCLES OF THE

Davatne, C. 1864. Recherchessur les vibrioniens. In:

' Compt. rend. Acad. Paris, v. 59, No. 15, pp. 629-

633.

. 1876. Bactérie. Bactéridie. In: Dictionn. En-

cyclop. d. Sciences Méd., éd. par A. Dechambre.

I. Sér., v. VIII, pp. 18-39. Paris.

Davis, D. J. 1907. Hemophilic bacilli—their mor-
phology and relation to respiratory pigments. In:
Journ. Infect. Diseases, v. 4, No. 1, pp. 73-86.

Davis, N. G. 1901. Variations of Bacillus rosaceus
metalloides (Dowdeswell). In: Journ. Boston Soc.
Med. Sciences, v. 5, No. 7, pp. 384-385.

Dawson, M. 1899. Further observations on ‘‘Nitragin”
and on the nature and functions of the nodules of
leguminous plants. In: Proceed. Roy. Soc. (Lon-

_ don), v. 66, No. 425, pp. 63-65.

DeBary, A. 1884. Vergleichende Morphologie und
Biologie der Pilze, Mycetozoen und Bacterien.
xvi and 558 pp., m. 198 Textabb. Leipzig.

Detace, Y. 1895. La structure du protoplasma et les
théories sur ’hérédité, ete. xvz and 878 pp., av.
39 figs. dans le texte. Paris.

De-Tont, J. B., et Trevisan, C. V. 1889. Schizomyce-
taceae Naeg. In: Sylloge Fungor. dig. P. A.
Saccardo, v. VIII, pp. 923-1037. Patavii.

Deyxe, G. 1910. Zur Biochemie der Tuberkelbazillen.
In: Miinchn. med. Wochenschr., v. 57, No. 12, pp.
633-636.

Drerrich, A. und Liesermetster, G. 1902. Sauer-
stoff iibertragende Kérnchen in Milzbrandbacillen.
In: Centralbl. f. Bact. I. Abt. Orig. v. 31,
No. 12, pp. 858-865.

Distaso, A. 1916. Digested and diluted serum as a
substitute for broth for bacteriological purposes.
In: Brit. Med. Journ., 1916, II, No. 2912, pp. 555-
556.

Dosext, C. C. 1908. Notes on some parasitic protists,

In: Quart. Journ. Microsc. Science, n.8., v. 52, No.

205, pp. 121-138, with 1 pl.

1909. On the so-called ‘‘sexual” method of
spore formation in the disporic bacteria. In:
Quart. Journ. Microsc. Science, n.s., v. 53, No. 211,
pp. 579-596, with 1 pl. :

——. 1911. Contributions to the cytology of the bac-

teria. In: Quart. Journ. Microsc. Science, n. s.,

v. 56, No. 223, pp. 395-506 with 4 pls,

. 1912. Researches on the Spirochaets and related

organisms. In: Arch. f. Protistenkunde, v. 26, No.

2, pp. 117-240, with 5 pls.

1913. Some recent work on mutations in micro-
érganisms. II. Mutations in bacteria. In: Journ,
Genetics, v. 2, No. 4, pp. 325-350.

Dosrzyniecer, A. R. von. 1897. Uber Leptothrix.
In: Centralbl. f. Bact. I. Abt. v. 21, No. 6/7,
pp. 225-229.

Dorrce, K. 1902. Beitrag zur Kenntnis des Erregers
der menschlichen Aktinomykose. In: Miinchen.
Med. Wochenschr., v. 49, No. 21, pp. 873-875.

Doerr, R. 1905. Uher Spirillum pyogenes Mezinescu.

‘Tn: Centralbl. f. Bact. I. Abt. Orig. v. 38, No. 1,
pp. 15-19, m. 1 Taf.

1911. Uber filtrierbares Virus. In: Centralbl.

f. Bact. I. Abt. Ref., v. 50, Beiheft, pp. 12-23 .

 

 

 

 

 

 

219

Doruzin, F, 1909. Lehrbuch der Protozoenkunde. x
und 914.pp., m. 825 Abb. im Text. Jena.

Doup, H. 1909. The cultivation of the so-called bottle
bacillus. (Preliminary note.) In: Journ. Roy.
Inst. Public Health, v.17, No. 12, p. 748.

1910. On the so-called bottle bacillus (Dermo-
phyton Malassez). In: Parasitology, v. 3, No. 3,
pp. 279-287, with 1 pl.

Domec, TH. 1892. Contribution 4 l’étude de la mor-
phologie de Vactinomyces. In: Arch. de méd. exp.
et d’anat. pathol., v. 4, No. 1, pp. 104-113.

Dorset, M. 1901a. Note on branched forms of tubercle
bacilli found in cultures. In: Ann. Rep. Bureau
of Animal Industry (U. 8. Dept. of Agric.), v.
18, pp. 572-573.

19016. The use of eggs as a medium for the cul-

tivation of B. tuberculosis. In: Ann. Rep. Bur.

Anim. Ind. (U. 8. Dept. of Agric.), v. 18, pp. 574—

578.

. Botton, B. M., and McBrypz,C.N. 1905. The
etiology of hog cholera. Bur. Anim. Ind. (U. 8.
Dept. of Agric.), Bull. 75, 102 pp., with 6 pls.

DowpEsweELt, G. F. 1889. Sur quelques phases du
développement du microbe du choléra. In: Ann.
microgr., v. 2, No. 12, pp. 529-544, av. 1 pl.

1890. Note on the morphology of the cholera
comma bacillus. In: Lancet, v. 68, 1890, I, No.
3487, pp. 1419-1423.

Doyen, E. 1885. Recherches anatomiques et expéri-
mentales sur le choléra épidémique. In: Arch.
physiol. norm. et pathol. 3me sér., v. 6, No. 2,
pp. 179-236, av. 3 pls.

Drosa, St. 1901. Die Stellung der Tuberkuloseerreger
im System der Pilze. Vorl. Mitt. In: Bull, inter-
nat. Acad. Cracovie. Cl. sci. math. et natur., 1901,
No. 6, pp. 309-311.

Ductaux, E. 1883. Chimie biologique. In: Encyclo-
pédie chimique, publ. par M. Frémy, v. 9, viret 908
pp. Paris.

1898. Traité de microbiologie. v. 1:Microbio-
logie générale, mz et 632 pp., av. 61 figs. Paris.

Dteeru, M. 1905a. Der Speziesbegriff bei den Bak-
terien. In: Verh. Schweiz. Naturforsch. Ges.

(Luzern), v. 88, pp. 287-299, m. 5 Taf.

19050. Bakteriologische Untersuchungen tiber
das armenische Mazun. In: Centralbl. f. Bact.
II. Abt. v.15, No. 19/20, pp. 577-600.

Dusarpin, F. 1841. Histoire naturelle des zoophytes.
xm et 684 pp. Paris.

Dunzpar, W. P. 1907. Zur Frage der Stellung der Bak-
terien, Hefen und Schimmelpilze im System. vm
und 60 pp., m. 3 Fig. u. 5 Taf. Miimchen und
Berlin.

Dunn, C. H. 1904. A motile culture of Bacillus dysen-
teriae. In: Jour. Med. Res., v. 11, No. 2, pp.
451-453.

Duruam, H. E. 1898. Some observations gn the Micro-
coccus melitensis (of Bruce). In: Journ. Pathol.
and Bact., v. 5, pp. 377-388.

Durscuenko,J.8. 1914. Eigentiimliche Einlagerungen
in die Erythrocyten einer Nagetierart, etc. In:
Centralbl. f. Bact. I. Abt. Orig. v. 74, No. 3/4,
pp. 241-243.

BACTERIA—LOHNIS.

 

 

 

 

 

 
220

Dutton, J. E., and Topp,J.L. 1905. The nature of hu-
man tick fever in the eastern part of the Congo Free
State. In: Mem. Liverpool School of Med., No. 17,
pp. 1-18.

 

 

1907. A note on the morphology of Spi-
rochaeta Duttoni. In: Lancet, v. 85, 1907, m, No.
4396, pp. 1523-1525.

Douvat,CH.W. 1914. Thebehavior of the Hansen ba-
cillus in vitro. In: Amer. Journ. Trop. Diseases,
v. 2, No. 3, pp. 185-190.

and Harris, W. H. 1913. Further studies upon

the leprosy bacillus, its cultivation and differentia-

tion from other acid-fast species. In: Journ. Med.

Res., v. 28, No. 1, pp. 165-198.

and WELIMAN, C. 1912. A critical study of the

organisms cultivated from the lesions of human

leprosy, etc. In: Journ. Infect. Diseases, v. 11,

No. 1, pp. 116-139.

Esrrson, F. 1918. A bacteriologic study of the diph-
theroid organisms with special reference to Hodg-
kin’s disease. In: Journ. Infect. Diseases, v. 23,
No. 1, pp. 1-41, with 3 pls. -

Exserrn, C. J. 1880. Die Organismen in den Organen
bei Typhus abdominalis. In: Virchow’s Arch. f.
pathol. Anat., v. 81, No. 1, pp. 58-74, m. 1 Taf.

1885. Zwei Mykosen des Meerschweinchens.

I, Chronische, durch Micrococcen erzeugte LHite-
rung. In: Virchow’s Arch. f. pathol. Anat., v.100,
No. 1, pp. 15-22.

Exsuers, T. 1884. Untersuchungen tiber den Rausch-
brandpilz. Diss. phil. Rostock., 48 pp., m. 1 Taf.

ExrRenserc, Cur. G. 1838. Die Infusionstierchen als
vollkommene Organismen. xvmt und 547 pp., m.
64 Taf. Leipzig.

E1senBERG, J. 1891. Bakteriologische Diagnostik nebst
Anhang: Bakteriologische Technik. 3 Aufl. xxr
und 509 pp. Hamburg und Leipzig.

ErsenBerc, Po. 1908. Uber die Thermoresistenz der
vegetativen Formen der aéroben Sporenbildner.
In: Centralbl. f. Bact. I. Abt. Orig. v. 48, No. 2,
pp. 187-191.

1909. Weitere Untersuchungen iiber Fettein-
schliisse bei Bakterien. In: Centralbl. f. Bact.
I, Abt. Orig. v. 51, No. 2, pp. 115-121.

——. 19138. Untersuchungen iiber halbspezifische
Desinfektionsvorginge. I. Mitt. In: Centralbl. f.
Bact. I. Abt. Orig. v. 71, No. 5/7, pp. 420-503.

—— 1914a. Untersuchungen iiber die Variabilitat
der Bakterien. III. Mitt. Weitere Untersuchun-
gen tiber das Sporenbildungsvermigen bei Milz-
brandbacillen. In: Centralbl. f. Bact. I. Abt.
Orig. v. 73, No. 2, pp. 81-123.

—. 1914. Untersuchungen itiber die Variabilitat
der Bakterien. IV. Mitt. Uber den Variations-
kreis des B. prodigiosum und violaceum. V.
Mitt. Uber Mutationen in der Gruppe des Bact.
fluorescens, B. pneumoniae, bei Sarcina tetragena
und bei Bact. typhi. In: Centralbl. f. Bact. I.
Abt. Orig. v. 73, No. 7, pp. 449-488.

——. 1918. Untersuchungen tiber die Variabilitit der
Bakterien. VI. Mitt. Variabilitét in der Typhus-
Coli-Gruppe. In: Centralbl. f. Bakt. I. Abt.
Orig. v. 80, No. 7, pp. 385-413, m. 3 Taf.

 

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

[Von. XVI,

E1senscuirz, §. 1895. Uber die Granulierung der
Hefezellen. In: Centralbl. f. Bact. II. Abt.
v. 1, No. 18/19, pp. 674-680.

ELLERMANN, V. 1904. Uber die Kultur der fusiformen
Bacillen. Vorl. Mitt. In: Centralbl. f. Bact. I.
Abt. Orig. v. 37, No. 5, pp. 729-730.

——. 1907. Zur Kenntnis der Spindelbazillen. In:
Zeitschr. f. Hyg., v. 56, No. 3, pp. 453-461.
Eis, D. 1902. Untersuchungen iiber Sarcina, Strep-
tococcus und Spirillum. In: Centralbl. f. Bact.
I. Abt. Orig. v. 33, No. 1, pp. 1-17; No. 2, pp.

81-96; No. 3, pp. 161-166, m. 2 Taf.

—— 1903. On the discovery of cilia in the genus
Bacterium. In: Centralbl. f. Bact. II. Abt.
v. 11, No. 8/9, pp. 241-251.

1906. The life history of Bacillus hirsutus. In:
Annals of Botany, v. 20, No. 79, pp. 233-257, with
1pl.

——. 1907. A contribution to our knowledge of the
thread bacteria. I. In: Centralbl. f. Bact. II.
Abt. v.19, No. 16/18, pp. 502-518, with 2 pls.

1910. A contribution to our knowledge of the

thread bacteria. II. In: Centralbl. f. Bact. II.

Abt. v. 26, No. 10/12, pp. 321-329, with 1 pl.

1912. An investigation into the life history of

Cladothrix dichotoma (Cohn). In: Proceed. Roy,

Soc. (London), Ser. B, v. 85, No. 580, pp. 344-858.

with 1 pl. ,

Emumericy, R., und Sama. 1900. Uber die morpholo-
gischen Veriinderungen der Milzbrandbacillen bei
ihrer Auflésung durch Pyocyanase. In: Centralbl.
f. Bact. I. Abt. v.27, No. 22/23, pp. 776-787, m.
1 Taf.

Enauer, A. 1882. Uber die Pilzvegetation des Weissen
oder Todten Grundes in der Kieler Bucht. In:
IV. Ber. d. Commission z. wissensch. Unters. d.
Deutschen Meere. VII-XI. Jahrg. I. Abt. pp.
187-194, m. 1 Taf.

Errrvcer, H. 1890. Uber eine neue pathogene Clado-
thrix und eine durch sie hervorgerufene Pseudo-
tuberculosis (cladothrichica). In: Beitr. z. pathol.
Anat. u. allg. Pathol., v. 9, No, 2, pp. 287-328, m,
2 Taf.

Erste, St. 1900. Untersuchungen iiber Milchsaure-
gérung und ihre praktische Verwertung. In:
Arch. f. Hyg., v. 37, pp. 329-859.

ERMENGEM, E. van. 1885. Recherches sur le microbe
du choléra asiatique. XII et 374 pp., av. 13 pls.
Paris et Bruxelles.

Ernst, P. 1888. Uber den Bacillus xerosis und seine
Sporenbildung. In: Zeitschr. f. Hyg., v. 4, No.
1, pp. 25-46, m. 1 Taf.

——. 1889. Uber Kern- und Sporenbildung in Bac-
terien. In: Zeitschr. f. Hyg., v. 5, No. 3, pp-
428-483, m. 2 Taf.

——. 1902. Uber den Bau der Bakterien. In: Cen-
tralbl. f. Bact. II. Abt. v. 8, No. 1, pp. 1-7;
No. 2, pp. 34-36; No. 3, pp. 65-73; No. 4, pp.
97-107, m. 2 Taf.

Ernst, W. 1902. Uber Nekrosen und den Nekrose-
bacillus (Streptothrix necrophora). In: Monatsh.
f, prakt. Tierheilk., v. 14, No. 5, pp. 193-228, m.
2 Taf.

 

 

 
No. 2.] LIFE CYCLES OF THE

Escuericu, TH. 1886. Beitrage zur Kenntnis der Darm-
bacterien. I. MHelikobacterium (Klebs). In:
Miinchn. med. Wochenschr., v. 33, No. 1, pp. 2-7.

——. 1894. Aetiologie und Pathogenese der epide-

, mischen Diphtherie. I. Der Diphtheriebacillus.
xu und 294 pp., m. 2 Taf. Wien.

, und PraunpieR, M. 1903. Bacterium coli
commune. In: Kolle-Wassermann’s Handb. d.
pathog. Mikroorg., v. 2, pp. 334-474.

Esmarca, E. 1887. Uber die Reinkultur eines Spiril-
lum. In: Centralbl. f. Bact., v. 1, No. 8, pp.
225-230.

Evans, A. C. 1918. The relationship between Bact.
abortus and the organism causing Malta fever. In:
Abstracts of Bact., v. 2, No. 1, p. 18.

Ewart, J. C. 1878a. The life history of Bacterium
termo und Micrococcus. In: Proceed. Roy. Soc.
(London), v. 27, No. 188, pp. 474-480, with 1 pl.

——. 1878b. On the life history of Bacillus anthracis.

In: Quart. Journ. Microsc. Science, n. s., v. 18,

No. 70, pp. 161-170, with 1 pl.

J. H. 1907. Maltafieber. In:
mann’s Handb. d. pathog. Mikroorg.
pp. 601-625.
and WasHsurgn, J. W. 1897. Resistant forms of

the Pneumococcus. In: Journ. Pathol. and Bact.,
‘v. 4, pp. 394-400.

Fazer, F.C. von. 1912a. Daserbliche Zusammenleben
von Bakterien und tropischen Pflanzen. In:
Jahrb. f. Wissensch. Botanik, v. 51, No. 3, pp.
285-375, m. 3 Taf.

1912b. Spirillum Bataviae n. sp. In: Cen-
tralbl. f. Bact. II. Abt. v.36, No. 1/5, pp. 41-42.

Fantuam, H. B. 1908. Spirochaeta (Trypanosoma)
Balbianii (Certes) and Spirochaeta anodontae
(Keysselitz); their movements, structure and affini-
ties. In: Quart. Journ. Microse. Science, n. 8.,
v. 52, No. 205, pp. 1-78, with 3 pls.

1911. Some researches on the life-cycle of
Spirochaetes. In: Annals of Trop. Medic. and
Parasitol., v. 5, No. 3, pp. 479-496.
Frevorowirscu, A. 1902. Uber die Kornigkeit der

~ Bakterien. In: Centralbl.f. Bact. II. Abt. v. 8,
No. 15/16, pp. 481-495, m. 1 Taf.

Frrnpere. 1900. Uber den Bau der Bakterien. In:
Centralbl. f. Bact. I. Abt. v. 27, No. 12/13,
pp. 417-426, m. 5 Taf.

FEIstMANTEL, ©. 1902. Sdure- und Alcoholfestigkeit
der Streptothrix farcinica, etc. In: Centralbl. f.
Bact. I. Abt. Orig. v. 31, No. 10, pp. 433-445.

Fetpmann, I. 1917. Uber choleraibnliche Vibrionen
mit besonderer Beriicksichtigung ihrer Mutations-
vorginge. In: Centralbl. Bakt. I. Abt. Orig. v.
80, No. 4, pp. 129-160.

Frrrdn, J. 1885. Uber die Morphologie des Komma-
bacillus. In: Zeitschr.f. Klin. Med. v. 9, No. 4,
pp. 361-373, m. 1 Taf.

————. 1897a. Note relative aux aptitudes saprophytes
du bacille de la tuberculose, etc. In: Compt. rend.
Acad. Paris, v. 125, No. 15, pp. 515-518.

1897b. Nouvelles découvertes sur le bacille de

tuberculose.- Barcelona. Abstr. In: Centralbl. f.
” Bact. I. Abt. v. 22, No. 16/17, pp. 483.

 

Kolle-Wasser-
I, Erg. Bd.

Eyre,

 

 

 

 

 

 

221

Ficxer, M. 1898. Uber Lebensdauer und Absterben
von pathogenen Keimen. In: Zeitschr. f. Hyg.,
v. 29, No. 1, pp. 1-74.

1903. Zur Frage der Kérnchen und Kerne der
Bakterien. In: Arch. f. Hyg., v. 46, pp. 171-199.

Finger, D., und Prior, J. 1884. Uber den Bacillus
der Cholera nostras und seine Cultur. In: Deutsche
Med. Wochnschr., v. 10, No. 39, pp. 632-634.

1885. Forschungen tiber Cholerabacte-
rien. In: Erg. Hefte z. Centralbl. f. allg. Gesundh.-
Pflege, v. 1, No. 5/6, pp. 279-445, m. 7 Taf.

Firtscu, G. 1888. Untersuchungen iiber Variationser-
scheinungen bei Vibrio Proteus. In: Arch. f. Hyg.,
v. 8, pp. 369-401, m. 2 Taf.

Fiscuzr, A. 1903. Farbungen. In: Encyclop. d. mi-
krosk. Technik, herausgeg. v. P. Ehrlich et al.
1,400 pp. m. 134 Textabb. Berlin und Wien.

Fiscnzt, Fr. 1893, Untersuchungen tiber die Morpho-
logie und Biologie des Tuberkulose-Erregers.
28 pp., m. 3 Taf. Wien und Leipzig.

Fiscuzr, A. 1891. Die Plasmolyse der Bakterien. In:
Ber. tiber d. Verhandl. d. Kgl. Sachs. Ges. d.
Wissensch. zu Leipzig. Math.-phys. Cl., v. 43,
pp. 52-74, m. 1 Taf.

1897. Untersuchungen tiber den Bau der
Cyanophyceen und Bakterien. 1x und 136 pp. m.
3 Taf. Jena.

1899. Fixierung, Farbung und Bau des Pro-

toplasmas. x und 362 pp. m.1 Taf. u. 21 Abb.

im Text. Jena.

1900. Die Empfindlichkeit der Bakterienzelle

und das bakterizide Serum. In: Zeitschr. f.Hyg.,

y. 35, No. 1, pp. 1-58, m. 1 Taf.

1903. Vorlesungen tiber Bakterien. 2. Aufl.

x und 374 pp., m. 69 Abb. Jena.

1906. Uber Plasmoptyse der Bakterien. In:
Ber. d. Deutschen Bot. Ges., v. 24, No. 2, pp.
55-63, m. 1 Taf.

Fiscuer, B. 1894. Die Bakterien des Meeres. In:
Ergebn, d. Plankton-Exped. d. Humboldt-Stif-
tung, herausg. v. V. Hensen., v. 4, M. g., 82 pp.,
m.1Taf. Kiel und Leipzig.

FiscHer, H. 1905. Uber Stickstoffbakterien. In: Ver-
handl. d. Naturhist. Ver. d. preuss. Rheinlande,
etc., v. 62, pp. 135-145, m. 1 Taf.

Fiexner, 8S. 1907. Spirochaeta (Treponema) pallida
and syphilis. In: Journ. Exp. Med., v. 9, pp.
464-472, with 1 pl. °

,and Nocucmt, H. 1913. Experiments on the culti-

vation of the microérganisms causing epidemic

poliomyelitis. In: Journ. Exp. Med., v. 18, No. 4,

pp. 461-485, with 4 pls.

, and Amoss, L. 1915. Concerning sur-
vival and virulence of the microérganisms culti-
vated from poliomyelitic tissues. In: Journ. Exp.
Med., v. 21, No. 1, pp. 91-102, with 6 pls.

Friiags, C. 1884. Sind die von Zopf in seinem Hand-
buch tiber die Spaltpilze gelehrten Anschauungen
vereinbar mit den Ergebnissen der neueren For-
schungen tiber Infectionskrankheiten? In: Deut-
sche Med. Wochenschr., v. 10, No. 46, pp. 741-743.

Forster, F, 1892. Uber eine merkwiirdige Erscheinung
bei Chromatium Okenii Ehrbg. sp. In: Centralbl.
f. Bact., v. 11, No. 9/10, pp. 257-264, m. 1 Taf.

BACTERIA—LOHNIS.

 

 

 

 

 

 

 

 
222

Foxxer, A. P. 1880. Die Identitat von Bacillus sub-
tilis und Bacillus anthracis. In: Centralbl. f. d.
_ med. Wissensch., v. 18, No. 44, pp. 817-819.

1881. Milzbrand ohne Stibchen. In: Central-
bl. f. d. med. Wissensch., v. 19, No. 2, pp. 20-22.
1882. Zur Bacterienfrage. In: Virchow’s Arch.

f. pathol. Anat-, v. 88, No. 1, pp. 49-60.
———. 1887a. Untersuchungen tiber THeterogenese.

 

 

Heft I: Protoplasmawirkungen. 53 pp. Gronin- |

gen.

18876. Untersuchungen iiber Heterogenese.
Heft II: Die Haematocyten. pp. 55-87, m. 1 Taf.
Groningen.

1888. Untersuchungen tiber Heterogenese.
Heft III: Weitere Untersuchungen, pp. 89-155,
m. Taf. II-V. Groningen.

. 1901. Untersuchungen  itiber

Heft IV: Die Granula der Milch.

Taf. Groningen.

1902. Versuch einer neuen Bakterienlehre.

49 pp.,m.2 Taf. Haag.

1903. Versuch einer neuen Bakterienlehre. In:
Centralbl. f. Bact. I. Abt. Ref., v. 33, No. 1,
pp. 1-6.

Fontanet, G. 1913. Les microbes invisibles.
méd. Paris. 120 pp.

Fonres, A. 1910. Estudos sobre a tuberculoze. In:
Mem. do Instit. Oswaldo Cruz, v. 2, No. 2, pp. 186-
205, c. 2 estamp.

Forp, W. W. 1916. Studies on aerobic spore-bearing

' nonpathogenic bacteria. PartI. In: Journ. Bact.,

v. 1, No. 3, pp. 273-319, with 26 pls.

Foster, yx., G. B. 1917. The etiology of common
colds. In: Journ. Infect. Diseases, v. 21, No. 5.
pp. 451-474, with 1 pl.

FrRAENKEL, C. 1891. Grundriss der Bakterienkunde.
3. Aufl. vurund 515 pp. Berlin.

——. 1895. Eine morphologische Eigentimlichkeit

des Diphtheriebacillus. In: Hyg. Rundschau, v,

5, No. 8, pp. 349-350.

1898. Untersuchungen iiber den von Stutzer
und Hartleb beschriebenen Salpeterpilz. In:
Centralbl. f. Bact. II. Abt. v. 4, No. 1, pp.
8-13; No. 2, pp. 62-67.

und Preirrer, R. 1895. Mikrophotographischer
Atlas der Bakterienkunde. 2. Aufl. xx und 50
pp., 76 Taf. m. Text. Berlin.

Frank, B. 1879. Uber die Parasiten in den Wurzelan.
schwellungen der Papilionaceen. In: Botan-
Zeitg., v. 37, No. 24, pp. 377-387; No. 25, pp.
393-400, m. 1 Taf.

——. 1887. Sind die Wurzelanschwellungen der
Erlen und Elaeagnaceen Pilzgallen? In: Ber. d.
Deutsch. Bot. Ges., v. 5, No. 2, pp. 50-57, m.
1 Taf.

——. 1889. Uber die Pilzsymbiose der Leguminosen.

In: Ber. d. Deutsch. Bot. Ges., v. 7, No. 8, pp.

332-346.

1890. Uber die Pilzsymbiose der Leguminosen.
In: Landw. Jabrb., v. 19, pp. 523-640.
Frevpenreice, E. von, und Jensen, Orta. 1906.

Uber die im Emmentaler Kise stattfindende Pro-
pionsiuregirung. In: Centralbl.f. Bact. II. Abt.
v. 17, No. 17/18, pp. 529-546.

 

Heterogenese.
102 pp., m. 3

 

 

These

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

[Vou XVI.

FREUDENREICH, E. von, und TH6n1, J. 1905. Uber die
Wirkung verschiedener Milchsiurefermente auf die
Kasereifung. In: Centralbl. f. Bact. II. Abt. v.
14, No. 2, pp. 34-43, m. 1 Taf.

FriepiANvER, C. 1883. Die Mikrokokken der Pneu-
monie. In: Fortschr. d. Med., v. 1, No. 22, pp.
715-733.

Frreprich, P. 1892. Vergleichende Untersuchungen
tiber den Vibrio cholerae asiaticae (Kommabacillus
Koch), etc. In: Arb. a, d. K. Gesundh. Amte, v.
8, No. 1, pp. 87-134, m. 3 Taf.

Frost, W. D., and McCampsuit, E.F. 1910. Textbook
of general bacteriology. xvi and 340 pp., illustr.
New York.

Fucus-Wotrrina, §. 1908. Die Much’schen ‘Granula”
und die Carl Spengler’schen ‘‘Splitter.’’ In: Beitr.

_ 2, Klinik d. Tuberkul., v. 10, pp. 175-182.

First, Tu. 1914. Untersuchungen iiber Variationser-
scheinungen beim Vibrio Finkler-Prior. In: Arch.
f. Hyg., v. 83, No. 7/8, pp. 350-390, m. 3 Taf.

Fuurmann, F. 1906a. Zur Kenntnis der Bakterienflora
des Flaschenbieres. I. Pseudomonas cerevisiae.
In: Centralbl. f. Bact. II. Abt. v.16, pp. 309-
325, m. 1 Taf. ;

——. 19060. Entwicklungszyklen von Bakterien. In:
Verb. d. Ges. Deutsch. Naturf. u. Aerzte. 78.
Vers. (Stuttgart). II. Teil, 1. Halfte: Naturw.
Abhandl., pp. 278-279. Leipzig, 1907. ;

——. 1908. Entwicklungszyklen bei Bakterien. In:
Beihefte z. Bot. Centralbl., v. 23, Abt. I, pp..1-18,
m. 1 Taf. ; :

1913. Vorlesungen iiber technische Mykologie.
vin und 454 pp., m.140 Abb. im Text. Jena. (spec.
pp. 46-50, m. 2 Abb.).

GArtNER, A. 1898. Untersuchungen tiber den von
Stutzer und Hartleb beschriebenen Salpeterpilz.
In: Centralbl. f. Bact. II. Abt. v.4, No.1, pp.
1-7; No. 2, pp. 52-61; No. 3/4, pp. 109-119, m. 2
Taf. f

Garrxy, G. 1881. Experimentell erzeugte Septicaemie
mit Riicksicht auf progressive Virulenz und acco-
modative Ziichtung. In: Mitt. a.d. K. Gesundh.
Amte, v. 1, pp. 80-133.

———. 1884. Zur Aetiologie des Abdominaltyphus.
In: Mitt.a.d. K. Gesundh. Amte, v. 2, pp. 372-420.

Gacz, G. E. 1910. Biological and chemical studies on
nitroso-bacteria. In: Centralbl. f. Bact. II. Abt.
v. 27, No. 1/3, pp. 7-48.

Gagz, St. pe M. 1903. Uber die relative Lebensfiihig-
keit von B. coli und B. typhosus unter gewissen
Bedingungen. In: Centralbl. f. Bact. I. Abt.
Ref., v. 33, No. 9/10, pp. 279-280.

Gawuxrov, N. 1906. Uber die ultramikroskopische

' Untersuchung der Bakterien und iiber die Ultra-
mikroorganismen. In: Centralbl.{. Bact. II. Abt.
v. 16, No. 20/21, pp. 667-672, m. 9 Abb.

Gatu-Vaterio, B. 1899. Contribution 4 étude de la
morphologie du Bac. mallei. In: Centralbl. f.
Bact. I. Abt. v.26, No. 6, pp. 177-180, av. 5 figs.

——. 1900. Seconde contribution & l’étude de la
morphologie de B. mallei. In: Centralbl. f. Bact.
I. Abt. v. 28, No. 12/13, pp. 353-359, av. 29 figs,

 
No. 2.] LIFE CYCLES OF THE

Gatii-Vauerio, B. 1903. Contribution 4 J’étude des
caractéres morphologiques et des cultures de Bact.
pestis et des rapports de ce bacille avec Bact. pseudo-
tuberculosis rodentium. In: Centralbl. f. Bact.
I. Abt. Orig. v. 33, No. 5, pp. 321-330, av. 2 pls.

——. 1912. Etudes sur les actinomycdtes. In: Cen-
tralbl. f. Bact. I. Abt. Orig. v. 63, No. 7, pp.
555-564.

——. 1913. Bacterium pseudopestis murium n. sp.
In: Centralbl. f. Bact. I. Abt. Orig. v. 68, No.
2, pp. 188-194.

Gamati1a, N. 1888. Vibrio Metchnikovi (n. sp.). In:
Annal. Inst. Pasteur, v. 2, No. 9, pp. 482-488.

——-. 1900. Elemente der allgemeinen Bakteriologie.
vi und 242 pp. Berlin. .

GarBowsxzI, L. 1906. Plasmoptyse und Abrundung bei
Vibrio Proteus. In: Ber. d. Deutsch. Botan. Ges.,
v. 24, No. 9, pp. 477-483, m. 1 Taf.

1907a. Gestaltsinderung und Plasmoptyse. In:
Arch. f. Protistenk., v. 9, No. 1, pp. 53-88, m.1
Taf,

——. 19076. Uber Abschwichung und Variabilitat des
Bacillus luteus Smith et Baker und Bacillus tumes-
censZopf. In: Centralbl.f. Bact. II, Abt. v.19,
No. 21/23, pp. 641-655; No. 24/25, pp.737-749; v. 20,
No. 1/3, pp. 4-20; No. 4/5, pp. 99-118, m. 2 Taf.

Garnier, M. 1907. Sur un microbe particulier. In:
Arch, de méd. exp. et d’anat. pathol., v. 19, No. 6,
pp. 785-792.

Garten, J. 1895. Uber einen beim Menschen chroni-
sche Eiterung erzeugenden pleomorphen Mikroben.
In: Deutsche Zeitschr. f. Chirurgie, v. 41, No. 4/5,
pp. 257-285, m. 1 Taf.

GasPERINI,G. 1889. Recherches morphologiques et bio-
logiques sur un microorganisme de 1’atmosphére,
la Streptothrix Foersteri Cohn. In: Annal. mi-
crogr., v. 2, No. 11, pp. 449-474, av. 3 pls.

Geppes, P., and Ewart, J.C. 1878. On the life-history
of Spirillum. In: Proceed. Roy. Soc. (London),
v. 27, No, 188, pp. 481-485, with 2 pls.

Grepoztst, L. 1899. Traité de microbiologie appliquée 4
la médecine vétérinaire. 2me éd. xv et 535 pp.,
av. 97 figs. dans le texte. Lierre.

Geissz, A. 1914. Erziehung pathogener Eigenschaften
bei saprophytischen Staphylokokken. In: Zeit-
schr, f. Hyg., v. 77, No. 3, pp. 482-494.

GrorcevircH, P, 1910. Dela morphologie des microbes

des nodosités des légumineuses,

Soc. Biol., v. 69, pp. 276-278.

1916. De la morphologie des microbes des no-
dules des feuilles d’une Rubiacée, Pavetia coffra.

In: Compt. rend. Soc. Biol., v. 79, pp. 411-413.

 

 

Gaon, A. und Mucna, V. 1905. Beitriige zur Kenntnis |
der antaéroben Bakterien des Menschen. III. Zur |

Aetiologie der Peritonitis. II. Mitt. In: Cen-
tralbl. f. Bact. I. Abt. Orig. v. 39, No. 5, pp.
497-508; No. 6, pp. 641-648; v. 40, No.1, pp. 37-48,
m. 1 Taf.

 

und Miitzer, R. 1906. Beitrage zur
Kenntnis der anaéroben Bakterien des Menschen.
IV. Zur Aetiologie der akuten Meningitis. In:
Centralbl. f. Bact. I. Abt. Orig. v. 41, No. 1, pp.
1-14; No. 2, pp. 145-152; No. 3, pp. 305-314; No.

160266°—21——-15

In: Compt. rend. |

223

4, pp. 401-405; No. 5, pp. 504-510; No. 6, pp. 606-
610; No. 7, pp. 689-694, m. 2 Taf.

Guon, A., und Sacus, M. 1903. Beitrige zur Kenntnis
der anaéroben Bakterien des Menschen. II. Zur
Aetiologie des Gasbrandes. In: Centralbl. f. Bact.
I. Abt. Orig. v. 34, No. 4, pp. 281-311; No. 5, pp.
398-406; No. 6, pp. 481-488; No. 7, pp. 609-617, m.
3 Taf.

BACTERIA—LOHNIS.

 

 

1905. Beitrage zur Kenntnis der anaé-
roben Bakterien des Menschen. III. Zur Aetiolo-
gie der Peritonitis. I. Mitt. In: Centralbl.f. Bact.
I, Abt. Orig. v. 38, No. 1, pp. 1-10; No. 2, pp.
131-136, m. 1 Taf.

Grarp, A, 1882. Sur le Crenothrix Kihniana (Raben-
horst), cause de l’infection des eaux de Lille. In:
Compt. rend. Acad. Paris, v. 95, No. 5, pp. 247-249.

Giemsa, G. 1904. Eine Vereinfachung und Vervoll-
kommung meiner Methylenazur-Methylenblau-
Eosin-Farbemethode zur Erzielung der Romanow-
sky-Nochtschen Chromatinfaérbung. In: Centralbl.
f. Bact. I. Abt. Orig. v. 37, No. 2, pp. 808-311.

——. 1914. Zur Schnellfarbung (Romanowsky-Far-
bung) in Trockenausstrichen. In: Centralbl. f.
Bact. I. Abt. Orig. v. 73, No. 7, pp. 493-496.

Gisert. 1904. Uber Actinomyces thermophilus und
andere Aktinomyceten. In: Zeitschr. f. Hyg., v.
47, No. 3, pp. 383-406, m. 1 Taf.

GILDEMEISTER, E. 1917. Weitere Mitteilungen iiber
Variabilitats-Erschein ungen bei Bakterien, etc. In:
Centralbl. f. Bact. I. Abt. Orig. v. 79, No. 2,
pp. 49-62.

GLEITsMANN. 1913. Beitrag zur Entwicklungsgeschichte
der Spirochaeten (Borrelien). In: Centralbl. f.
Bact. I. Abt. Orig. v.68, No.1, pp.31-49, m.1 Taf.

GoapBy, K. W. 1917. Bacteriology of septic war
wounds. In: Journ. Roy. Microsc. Soc. 1917, pt. 3,
No. 238, pp. 269-278, with 4 pls.

GonpeR, R. 1909. Die Stellung der Spirochaeten unter
den Protisten. In: Centralbl. f. Bact. I. Abt.
Orig. v. 49, No. 2, pp. 190-196, m. 2 Taf.

——. 1910. Die Entwicklung von Theileria parva,
dem Erreger des Kiistenfiebers. In: Arch. f. Pro-
tistenk., v. 21, No. 2, pp. 1438-164, m. 5 Taf.

Goopman, H. M. 1908. Variability in the diphtheria-
group of bacilli. In: Journ, Infect. Diseases, v. 5,
No. 4, pp. 421-442.

Gortscuticn, E. _1903. Allgemeine Morphologie und Bio-

logie der pathogenen Mikroorganismen. In: Kolle-

Wassermann’s Handb. d. pathog. Mikroorg., v. 1,

pp. 29-222 (spec. pp. 29-131).

1906. Discussionsbemerkung zu M. Neisser’s
Bericht. In: Centralbl.f. Bact. I. Abt. Ref., v.
38, Beiheft, p. 98.

1909. Nachtrige zur allgemeinen Morphologie
und Biologie der Bakterien. In: Kplle-Wasser-
mann’s Handb. d. pathog. Mikroorg. Erg. Bd. 2,
pp. 1-61. :

Gorrnzm, O. 1901. Botanische Beschreibung einiger
Bodenbakterien. In: Centralbl. f. Bact. II. Abt.
v. 7, No. 12, pp. 430-485; No. 13, pp. 449-465; No.
14, pp. 481-497; No. 15, pp. 529-544; No. 16, pp.
582-591; No. 17/18, pp. 627-637; No. 19, pp. 680-691;

 

 

 

No. 20, pp. 717-730, m. 4 Taf.
224

Granam-Smrru, G. §. 1910. The division and post-
fission movement of bacilli grown on solid media.
In: Parasitology, v. 3, No. 1, pp. 17-58, with 5 pls.

GrassBEeRGER, R. 1898. Zur Frage der Scheinfiden-
bildung bei Influenzakulturen. In: Centralbl. f.
Bact. I. Abt. v. 23, No. 9/10, pp. 353-364,
m., 1 Taf. 5

——. 1902. Zur Morphologie des beweglichen Butter-
sdurebacillus. In: Arch. f. Hyg., v. 42, No. 8, pp.
219-250, m. 4 Taf.

1903. Morphologie des Rauschbrandbacillus und
des Odembacillus. In: Arch. f. Hyg., v. 48, No. 1,
pp. 1-76, m. 11 Taf.

——. 1905. Uber Anpassung und Vererbung bei
Bakterien. Zugleich ein Beitrag zur Aérobiose
anaérober Bakterien. I. Mitt. In: Arch. f. Hyg.,
v.53, No. 2, pp. 158-179, m. 2 Taf.

‘und ScHarrenrrou, A.
sduregirung. IV.Abhandlung. In: Arch. f. Hyg.,
v. 60, No. 1, pp. 40-78, m. 2 Taf.

Greia-SmirH, R. 1900. The nodule organism of the
Leguminosae. In: Proceed. Linn. Soc. N. 8.
Wales, v. 24, No. 4, pp. 653-678, with 2 pls.

Gross, J. 1912. Uber Systematik, Structur and Fort-
pflanzung der Spironemaceae. In: Centralbl. f.
Bact. I. Abt. Orig. v. 65, No. 1/3, pp. 83-98.

1918. Sporenbildung bei Cristispira. In: Arch.
f. Protistenk., v. 29, No. 2, pp. 279-292, m. 1 Taf.

Gruser, M. 1885. Uber die als ‘“‘Kommabacillen”
bezeichneten Vibrionen von Koch und Finkler-
Prior. Vorl. Mitt. In: Wiener med. Wochenschr.,
v. 35, No. 9, pp. 262-263; No. 10, pp. 298-302.

1894. Cholera-Studien. II. Uber die bak-

teriologische Diagnostik der Cholera und des

Cholera-Vibrio. In: Arch. f. Hyg., v. 20, pp.

123-151.

1906. Discussionsbemerkung zu M. Neisser’s
Bericht. In: Centralbl. f. Bact. I. Abt. Ref., v.

38, Beiheft, p. 98.

Gruner, O. C., and Fraser, J. R. 1912. Observations
on Bacillus mesentericus and allied organisms. In:
Journ. Infect. Diseases, v. 10, No. 2, pp. 210-225.

Ginruer, C. 1906. Einfiihrung in das Studium der
Bakteriologie. 6. Aufl. xm und 904 pp., m. 15
Taf. Leipzig.

GuienarpD, L., et CHarrin. 1887. Sur les variations
morphologiques des microbes. In: Compt. rend.
Acad. Paris, v. 105, No. 24, pp. 1192-1195.

Gurturrmonp, A. 1908. Contribution 4 l’étude cyto-
logique des bacilles endosporés. In: Arch. f.
Protistenk., v. 12, No. 1/2, pp. 9-43, av. 3 pls.

—. 1910a. Les corpuscules métachromatiques ou
grains de volutine.
No. 3, pp. 289-309.

——. 19100. Quelques remarques sur la copulation
des levires. In: Annales mycologici, v. 8, No. 3,
pp. 287-297.

——., 1911. Sur un example de copulation hétéro-
gamique observé chez une levfre. In: Compt.
rend. Soc. Biol., v. 70, pp. 442-444.

———. 1912. Nouvelles observations sur la sexualité de

- levdres. In: Arch. f. Protistenk., v. 28, No. 1,
pp. 52-77, av. 4 pls.

 

 

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES,

1907. ther Butter-

In: Arch. f. Protistenk., v. 19, |

 

(Von. XVI.

GuitueRMOND, A., et Mawas. 1908. Caractéres histo-
chimiques des granulations des Mastzellen, etc. In:
Compt. rend. Soc. Biol., v. 64, pp. 307-309.

Gurat, V. 1898. Sur la phylogénie et le pléomorphisme
des bactéries. Montevideo. (Not seen.)

Gurrmmann, P. 1880. Zur Histologie des Blutes bei
Febris recurrens. In: Virchow’s Arch. f. pathol.
Anat., v. 80, No. 1, pp. 1-9.

Haas, E. 1905. Beitrag zur Kenntnis der Aktinomy-
ceten. In: Centralbl. f. Bact. I. Abt. Orig. v.
40, No. 2, pp. 180-186.

Hazperxorn, Tu. 1882. Vorlaufige Mitteilung iiber
Genese, Morphologie und Eigenschaften sogenannter
pathogener Bacterien. In: Botan. Centralbl., v. 10,
No. 16, pp. 100-106.

Haxrcxet, E. 1876. Die Perigenesis der Plastidule oder
die Wellenzeugung der Lebensteilchen. 80 pp.,
m.1Taf. Berlin.

Harsersraiprer, L., und Prowazex, 8S. von. 1909.
Uber Chlamydozoenbefunde bei Blennorrhoea
neonatarum non gonorrhoica. In: Berl. Klin
Wochenschr., v. 46, No. 41, pp. 1839-1840.

Hatt£, J. 1898. Recherches sur la bactériologie du
canal génital de la femme. Thése méd. Paris,
108 pp., av. 1 pl.

HALLENBERGER. 1917. Beitrag zur Aetiologie der Vari-
ola. In: Centralbl. Bakt. I. Abt. Orig. v. 80,
No. 1/3, pp. 89-105, m. 2 Taf.

Hauer, E. 18650. Uber Leptothrix buccalis. In:

Botan. Zeitg., v. 23, No. 23, pp. 181-183, m. 1 Taf.

1865b. Beobachtungen iiber Leptothrix und
Hefe. In: Botan. Zeitg., v. 23, No. 38, pp. 282-284;
No. 39, pp. 289-291, m. 1 Taf.

——. 1866. Die pflanzlichen Parasiten des mensch-
lichen Kérpers. rv und116 pp.,m.4Taf. Leipzig.

. 1867. Garungserscheinungen. vi und 112 pp.,
m.1Taf. Leipzig.

——. 1868. Parasitologische Untersuchungen. vi und

80 pp., m.2 Taf. Leipzig.

1878. Die Plastiden der niederen Pflanzen, ihre,
selbstandige Entwicklung, ihr Eindringen, etc.
92 pp., m.4 Taf. Leipzig.

——. 1895. Die Pestkrankheiten (Infektionskrank-
heiten) der Kulturgewachse. xv und 144 pp., m.
7 Taf. Stuttgart.

——-. 1896. Die Hefe der Alkoholgérung, insbeson-
dere der Biergirung. rv und 61 pp., m. 2 Taf.
Weimar.

Hamitron, A. 1904. The question of virulence among
the so-called pseudo-diphtheria bacilli. In: Journ.
Infect. Diseases, v. 1, No. 4, pp. 690-713.

Hammert, H. 1906. Studien tiber die Morphologie dea
Vibrio cholerae asiaticae. In: Centralbl. f. Bact.
I. Abt. Orig. v. 41, No. 6, pp. 611-616; No. 7, pp.
695-700; No. 8, pp. 785-788; v. 42, No.1, pp. 1-4,
m. 1. Taf.

Hankin, E. H., and Leumann, B. H. F. 1897. A
method of rapidly identifying the microbe of bu-
bonic plague. In: Centralbl. f. Bact. I. Abt. v.
22, No. 16/17, pp. 438-440.

Hansen, E. Cur. 1911. Untersuchungen tiber Essig-
bakterien. I. (1879\ and II. (1894). In E. Chr.
Hansen’s Gesammelte Theoretische Abhandlungen
uber Garungsorganismen, herausg. v. A. Klécker
pp. 457-513. Jena.

 

 

 
No. 2.] LIFE CYCLES OF THE
Hansen, G. A. 1880. Baeillus leprae. In: Virchow’s
Arch. f. pathol. Anat., v. 79, No. 1, pp. 31-42, m.

1 Taf. ‘

. 1882. Studien iiber Bacillus leprae. In: Vir-
chow’s Arch. f. pathol. Anat., v. 90, No. 3, pp.
542-548. .

— —. 1903. Lepra. In: Kolle-Wassermann’s Handb.
d. pathog. Mikroorg., v. 2, pp. 178-203.

Hanserra, A. 1889. Noch einmal iiber Bacillus muralis
Tom. und iiber einige neue Formen yon Grotten-
Schizophyten. In: Botan. Centralbl., v. 37, No. 2,
pp. 33-39.

Hanstein, J. von. 1882. Beitrige zur allgemeinen Mor-
phologie und Biologie der Pflanzen. In: Botan.
Abhandl. herausg. v. J. Hanstein, v. 4, No. 3, pp.
1-244,

Harris, A.B. 1893. A new microorganism of spreading
oedema. In: Journ. Pathol. and Bact., v. 1, pp.
303-317.

Harris, W.H., and Wave, H.W. 1915. The widespread
distribution of diphtheroids and their occurrence in
various lesions of human tissues. In: Journ. Exp.
Med., v. 21, No. 5, pp. 493-508, with 1 pl.

Harrison, F.C. 1900. Thefoul brood of bees. Bacillus
alvei (Cheshire and W. Cheyne). In: Centralbl.
f. Bact. II. Abt. v. 6, No. 13, pp. 421-427; No.
14, pp. 457-469; No. 15, pp. 481-496; No. 16, pp.
513-517.

——— and Bartow, B. 1905. A new chromogenic slime
producing organism. In: Centralbl. f. Bact. II.
Abt. v. 15, No. 17/18, pp. 517-538.

1907. Thenodule organism of legumino-
s2e—its isolation, cultivation, identification and com.
mercial application. In: Centralbl.f. Bact. II. Abt.
v.19, No. 7/9, pp. 264-272; No. 13/15, pp. 426441
w. 9 pl.

Harries, R. 1900. Die Morphologie und systematische
Stellung der Knéllchenbakterien. In: Chemiker
Zeitg., v. 24, No. 82, pp. 887-888.

Hartmann, M. 1909. Autogamie bei Protisten und ihre
Bedeutung fiir das Befruchtungsproblem. In:
Arch. f. Protistenk., v. 14, No. 2, pp. 264-334.

Hasnmoto,§. 1899. Ein pleomorphes Bakterium. In:
Zeitschr. §. Hyg., v. 31, No. 1, pp. 85-88.

 

 

Hastam, H.C. | 1898. The pleomorphism of the common |

Colon bacillus. In: Journ. Pathol. and Bact., v.
5, pp. 189-194.

Hara, §. 1908. Uberdiedurch bestimmte anorganische

' Salze verursachten Degenerationsformen bestimmter

Bakterienarten. In: Centralbl. f. Bact. I. Abt.
Orig. v. 46, No. 4, pp. 289-292. :

Hauser, G. 1885. Uber Faulnisbacterien und deren
Beziehungen zur Septicaemie. 94 pp., m. 15 Taf.
Leipzig. a

——. 1892. Uber das Vorkommen von Proteus vul-
garis bei einer jauchig-phlegmonésen Eiterung. In:
Miinchn, med. Wochenschr., v. 39, No. 7, .pp.
103-105.

Hawrnorn E. 19082. Cultures homogénes du bacille
de la tuberculose en eau peptonée. In: Compt.
rend. Soc. Biol., v. 55, pp. 398-399.

——. 19036. De V’apparition de corps sphériques
ressemblant 4 des spores sur le bacille tuberculeux

 

225

In: Compt. rend. Soc.

BACTERIA—LOHNIS.

cultivé en eau peptonée.
Biol., v. 55, pp. 399-400.

Heaty, D. J., and Gorr, E.J. 1916. A filterable organ-
ism isolated from the tissues of cholera hogs. In:
Journ. Infect. Diseases, v. 18, No. 1, pp. 124-128,
with 1 pl.

Hecker, F. 1916. A new model of double pipet holder
and the technic for the isolation of living organisms.
In: Journ. Infect. Diseases, v. 19, pp. 305-314, with
6 figs.

Herreran, M. 1902.
In: Centralbl. f. Bact.
689-699.

Hzm, L. 1906. Lehrbuch der Bakteriologie. 3. Aufl.
xu und 550 pp., m. 233 Abb. im Text u. 13 Taf.
Stuttgart.

Heiwemann, P.G. 1915. The variability of two strains

of Streptococcus lacticus. In: Journ. Infect. Dis-

eases, v. 16, No. 2, pp. 221-239.

1917. The morphology of a strain of B. diph-

theriae. In: Journ. Bact. v. 2, No. 4, pp. 361-363,

Hetnrico, 1912. Vergleichende Untersuchungen iiber
die granuliren Formen der Tuberkelbazillen bei
Haustieren. In: Monatsh. f. prakt. Tierheilk. v. 23,
No. 10/11, pp. 483-513, m. 1 Taf.

HenNEBERG, W. 1897. Beitrige zur Kenntnis der
Essigbakterien. In: Centralbl. f. Bact. II. Abt.
v. 3, No. 9/10, pp. 223-228.

1898. Weitere Untersuchungen iiber Essigbak-

terien. In: Centralbl.f. Bact. II. Abt. v. 4, No.1,

pp. 14-20; No. 2, pp. 67-73; No. 3/4, pp. 138-147.

1901. Zur Kenntnis de Milchsiurebakterien der
Brennereimaische, der Milch und des Bieres. In:
Wochenschr. f. Brauerei, v. 18, No. 30. «Autoref.
in Centralbl. f. Bact. II. Abt. v. 8, No. 6, p.
184-190.

1904. Abnorme Zellformen von Brennereihefen.
In: Centralbl. f. Bact. II. Abt. v. 13, No. 5/7,
p. 150-153, m. 1 Taf. ;

Henri, Muze. Vicror. 1914. Etude de l’action méta-
biotique des rayons ultraviolets. Production de
‘formes de mutation de la bactéridie charbonneuse.
In: Compt. rend. Acad. Paris, v. 158,pp. 1032-1035,
av. 1 pl. o

Henrict, A. T. 1916. The specificity of streptococci.
In: Journ. Infect. Diseases, v. 19, No. 4, pp.
572-605, i

Henry, A. 1913. A consideration of the infective
granule in the life history of protist organisms. In:
Journ. Pathol. and Bact., v. 18, pp. 250-258.

Herewe, F, pz. 1917. Sur un microbe antagoniste des
bacilles dysentériques. In: Compt. rend. Acad.
Paris, v. 165, No. 11, pp. 373-375.

Herter, C. A. 1907. The common bacterial infections
of the digestive tract. x and 360 pp. New York.

Herzog, H. 1903. Die Abschwaichung der Saugetier-
tuberkulosebacillen im Kaltbliiterorganismus. In:

Centralbl..f. Bact. I. Abt. Orig. v. 34, No. 6, pp.

535-539; No. 7, pp. 675-685, m. 1 Taf.

1910a. Uber die Natur des Trachomerregers.
In: Deutsche med. Wochenschr., v. 36, No, 23,
p. 1076.

An uxusual bacterial grouping.
II. Abt. v. 8, No. 22, pp.

 

 

 

 

 
226

Herzoe, H. 19106. Uber die Natur und Herkunft des
Trachomerregers, etc. 52 pp., m. 2 Taf. Berlin
und Wien. Rev. by G. Meyer in Centralbl. f. Bact.

I. Abt. Ref., v. 48, No. 9, pp. 281-283.

1913. Uber die Involutionsformen des Gonococ-

 

cus Neisser, etc. In: Virchow’s Arch. f. pathol. |

Anat., v. 212, No. 2/3, pp. 248-367, m. 2 Taf..

Hest, J. J. van. 1907. Pseudovakuolen in Hefezellen |

und Ziichtung von Pseudozellkernen ausserhalb
der Hefezellen. In: Centralbl. f. Bact. II. Abt.
v. 18, No. 24/25, pp. 767-790, m.3 Taf.

Hewrett, R. T. 1902. Manual of bacteriology. 2d
ed. xi and 533 pp., with 20 pls. and 64 figs. in the
text. Philadelphia.

Heymann. 1903. Diphtheriebacillus. In: Encyclopae-
die d. mikrosk. Technik, herausg. v. P. Ehrlich et
al. 1400 pp., m. 134 Textabb. Berlin und Wien.

Hermann, B. 1909. Uber die ‘‘Trachomkérperchen.”’
In: Deutsche med. Wochenschr., v. 35, No. 39, pp.
1692-1694.

Hipster, E. von. 1908. Untersuchungen iiber die
pathogenen Anaéroben. 438pp.,m.17 Taf. Jena.

Hux, H. W. 1899. Branching forms of Bacillus diph-
theriae. In: Journ. Boston Soc. Med. Sciences, v.
3, No. 4, pp. 86-92, with 1 pl.

19022. Branching in bacteria with special

reference to B. diphtheriae, In: Journ. Med. Res.,

v.7, No. 2, pp. 115-127, with 2 pls.,

1902b. ‘‘Hanging block” preparations for the
microscopic observations of developing bacteria.
In: Journ. Med. Res., v. 7, No. 2, pp. 202-212,
with 3 figs.

Hurner, L. 1898. Uber Entstehung und physiologische
Bedeutung der Wurzelknéllchen. In: Forstl.-
naturw. Zeitschr., v. 7, No. 12, pp. 415-423.

——. 1900. Uber die Bakteroiden der Leguminosen-
knéllchen und ihre willkirliche Erzeugung ausser-
halb der Wirtspflanzen. In: Centralbl. f. Bact.
II. Abt. v. 6, No. 9, pp. 273-281.

und Srérmer, K. 1903. Neue Untersuchungen

tiber die Wurzelknéllchen der Leguminosen und
deren Erreger. In: Arb. a. d. Biol. Abt. f. Land-
u. Forstw. am K. Gesundh. Amte, v. 3, No. 3, pp.
151-307, m. 4 Taf.

Hinpiz, E. 1911. On the life cycle of Spirochaeta galli-
narum. Preliminary note. In: Parasitology, v. 4,
No. 4, pp. 463-477.

Hiss, sz., P. H., and Zmsser, H. 1914. Textbook of
‘bacteriology. 2d ed. .xv1 and 766 pp., with 153
illustr. New York and London.

Hava. 1902. Leuconostoc hominis und seine Rolle bei

' den akuten exanthematischen Krankheiten. In:
Centralbl. f. Bact. I. Abt. Orig. v. 31, No. 4, pp.
263-270. :

Horrer, P. A. 1911. Uber intrazellulére Einschluss-
kérper bei Scharlach. In: Deutsche med. Wochen-
schr., v. 37, No. 23, pp. 1063.

Horticu, C. 1902. Vergleichende Untersuchungen tiber
die Denitrifikationsbakterien des Mistes, des Strohes
und der Erde. In: Centralbl. f. Bact. II. Abt.
v. 8, No. 8, pp. 245-248; No. 9, pp. 273-278; No. 10,
pp. 305-308; No. 11, pp. 336-339; No. 12, pp.
361-367; No. 13, pp. 398-406.

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES,

 

[Vou. XVI.

Hozttme, A. 1910. Die Kernverhaltnisse von Fusi-
formis termitidis. In: Arch. f. Protistenk., v. 19,
No. 3, pp. 239-245, m. 1 Taf.

1911. Vergleichende Untersuchungen tiber Spi-
rochaeten und Spirillen. In: Arch. f. Protistenk.,
v. 23, No. 1/2, pp. 101-124, m. 4 Taf.

Horrmann, H. 1869. Uber Bacterien. In: Botan.
Zeitg., v. 27, No. 15, pp. 233-244; No. 16, pp. 249-
257; No. 17, pp. 265-272; No. 18, pp. 281-291; No.
19, pp. 305-314; No. 20, pp. 321-332, m. 1 Taf.

Horrmann, W.H. 1912. Zur Stellung der Spirochaeten
im System. In: Centralbl. f. Bact. I. Abt. Orig.
v. 66, No. 7, pp. 520-523.

Hoitanpt, R. 1906. Die Zungenaktinomykose des
Schweines, etc. Diss: vet.med. Giessen. 49 pp.

Hotucer, W. 1902. Bakteriologische Untersuchungen
tiber Mehlteiggirung. In: Centralbl. f. Bact. II.
Abt. v.9, No. 9, pp. 305-312; No. 10, pp. 361-371;
No. 11/12, pp. 395-425; No. 13, pp. 473-483; No,
14/15, pp. 521-537.

Horman, W. L. 1914. The relative longevity of differ-
ent streptococci and possible errors in the isolation
and differentiation of streptococci. In: Journ.
Infect. Diseases, v. 15, No. 2, pp. 298-308.

Horren, K. 1893. Zur Reinkultivierung auf fliissigem
Nahrboden. In: Centralbl. f. Bact., v. 18, No. 28,
pp- 752-753.

Horxiys, 8S. A. 1898. A peculiar mouth bacterium.
In: Journ. Boston Soc. of Med. Sci., v. 2, No. 10,
pp. 163-166, with 1 pl.

Horowrrz, A. 1911. Contribution 4 l'étude des pro-
priétés biologiques des vibrions cholériques. In:
Arch. d. sci. biol. St. Pétersbourg, v. 16, No. 5,
pp. 445-489. —

Hort, E. C. 1915. Typhus fever. In: Brit. Med.
Journ., 1915, I, No. 2833, pp. 673-675.

——. 1916a. Epidemic cerebro-spinal fever: The
place of the meningococci in its etiology. In:
Brit. Med. Journ., 1916, I, No. 2874, pp. 156-160.

19166. Studies in pleomorphism in typhus and

other diseases. In: Journ. Roy. Microsc. Soc.,

1916, No. 6, pp. 528-533.

. 1917a. Morphological studies in the life-history

of bacteria. In: Proceed. Roy. Soc. (London), Ser.

B., v. 89, No. 620, pp. 468-480, with 5 pls.

- 19176. The meningococcus of Weichselbaum.

‘In: Brit. Med. Journ. 1917, II, No. 2960, pp. 377-381.

and CauLremp, A.H. 1916a. Epidemic cerebro-
spinal fever: The place of the meningococcus in its
etiology. In: Brit. Med. Journ., 1916, II, No. 2911,
pp. 522-524.

— 19166. Epidemic cerebro-spinal fever:
The place of the meningococcus in its ers
In: Journ. Roy. Army Med. Corps, v. 26, No.
pp. 312-339.

, Lakin, C. E., and Brnzans, T. H. C. 1915.

Epidemic cerebro-spinal fever: The place of the

meningococcus in its etiology. In: Brit. Med.

Journ., 1915, I, No. 2830, pp. 541-543; No. 2884, Pp-
715-718.

 

 

 

 

 

 

3,

 

 

 

1916. The relation of the menin-
gococcus of Weichselbaum to the true infective
agent in epidemic cerebro-spinal fever. In: Journ.
Roy. Army Med. Corps, v. 26, No. 2, pp. 153-188.
No. 2.] LIFE CYCLES OF THE

Howarp, W. T., and Perks, R.G. 1904. A study of
the etiology of variola. In: Journ. Med. Res.,
v.12, No. 3, pp. 359-383, with 3 pls.

Htsner. 1908. Ist der Bacillus suipestifer der Erreger
der Schweinepest oder nicht? In: Centralbl. f.
Bact. I. Abt. Orig. v. 47, No. 5, pp. 586-590.

Huezpre, F. 1883. Kritik von Zopf’s ‘Morphologie der
Spaltpflanzen.” In: Fortschr. d. Med., v. 1, No.
6, pp. 204-208.

———. 1884. Untersuchungen iiber die Zersetzungen
der Milch durch Mikroorganismen. In: Mitt. a. d.
K. Gesundh. Amte, v. 2, pp. 309-371.

——. 1885. Uber die Dauerformen der sogenannten
Kommabacillen. In: Fortschr. d. Med., v. 3, No.
19, pp. 619-627.

——. 1886. Die Formen der Bacterien und ihre
Beziehungen zu den Gattungen und Arten. vu
und 152 pp., m. 24 Abb. Wiesbaden.

——. 1891. Methoden der Bakterienforschung. 5.
Aufl. vor und 495 pp., m. 2 Taf. u. 68 Abb. im
Text. Wiesbaden.

——. 1896. Naturwissenschaftliche Einfihrung in
die Bakteriologie. vu1 und 268 pp., m. 28 Abb. im
Text. Wiesbaden.

Huntemttier. 1916.- Kritische Studien iiber Morphol-

. ogie und Ztichtung von filtrierbaren Virusarten.
Centralbl. Bakt. I. Abt. Orig. Vol. 79, No. 1,
pp. 36-40.

Horyna, F., and Marex, J. 1916. Special pathology
and therapeutics of the diseases of domestic animals.
2d Amer. ed., by J. R. Mower and A. ErcHHorn,
v. I: xvr and 1213 pp., with 12 pls. and 231 ill.
v. II: xvm and 1108 pp., with 7 pls. and 207 ill.
Chicago.

Hoxizy. 1870. On the relations of Penicillium, Torula
and Bacterium. In: Quart. Journ. Microscop.
Science, n. 8., v. 10, pp. 355-362.

Inapa, R., Ivo, Y., Hox, R., Kanexo, R., and Ivo, H.
1916. The etiology, mode of infection and specific
therapy of Weil’s disease (Spirochaetosis ictero-
haemorrhagica). In: Journ. Exp. Med., v. 23,
No. 3, pp. 377-402, with 7 pls.

Isra&L, J. 1878. Neue Beobachtungen auf dem Gebiete
der Mykosen des Menschen. In: Virchow’s Arch.
f. pathol. Anat., v. 74, No. 1, pp. 15-53, m. 4 Taf.

Isratn, O. 1884. Uber die Cultivierbarkeit des Acti-
nomyces. In: Virchow’s Arch. f. pathol. Anat.,
v. 95, No. 1, pp. 140-145.

Irzerorr, G., und Niemann, Fr. 1895. Mikrophoto-
graphischer Atlas der Bakterienkunde. 115 pp.
und 21 Taf. Leipzig.

Iwanowsk1, D. 1899. Uber die Mosaikkrankheit der
Tabakpflanze. In: Ceritralbl. f. Bact. II. Abt.
v. 5, No. 8, pp. 250-254.

JazGER, H. 1892. Die Aetiologie des infectidsen fieber-
haften Icterus (Weilsche Krankheit). In: Zeitschr.
f. Hyg., v. 12, pp. 525-596, m. 7 Taf.

Jann, E. 1909/1911. Myxobacteriales. In: Kryptoga-
menflora der Provinz Brandenburg., v. 5, No. 1/2,
pp. 187-206. ;

Janke, A. 1916. Studien ter die Essigsiurebakterien-
flora von Lagerbieren des Wiener Handels. In:
Centralbl. f. Bact. II. Abt. v. 45, No. 1/5, pp.
1-50, m. 2 Taf.

 

BACTERIA—LOHNIS. 227

Junie, L. 1902. Uber eine neue Bakterienart im Spu-
tum. In: Centralbl. f. Bact. I. Abt. Orig. v. 32,
No. 3, pp. 192-200, m. 1 Taf.

Jensen, C. O. 1908. Die vom Nekrosebacillus (B.
necroseos) hervorgerufenen Krankheiten. In:
Kolle-Wassermann’s Handb. d. pathog. Mikroorg.,
v. 2, pp. 693-706.

JessrN, F., und Rasrinowirscn, L.- 1910. Zur Frage
der Léslichkeit von Tuberkelbacillen. In: Cen-
tralbl. f. Bact. I. Abt. Orig. v. 54, No. 5, pp.
454-457.

Jéreznsen, A. 1911. Microorganisms and fermenta-
tion. Transl. by S. H. Davies. 4thed. xz and
489 pp., with 101 figs. in the text. London.

Jounson, J. Cu. 1912. On well-marked aérotropic
growth of Bacillus Megaterium. In: Annals of
Botany, v. 26, No. 108, pp. 949-950.

Jounston, J. A. 1917. The varying morphology of
Bacillus leprae, etc. In: Philippine Journ. Sci.,
v. 12, Sect. B, No. 3, pp. 115-118, 1 pl.

Jomtos, V. 1914. Variabilitat und Vererbung bei Mi-
kroorganismen. In: Zeitschr. f. indukt. Abstamm,
u. Vererbungslehre, v. 12, pp. 14-35.

Jones, D. H. 1913. A morphological and cultural
study of some Azotobacter. In: Centralbl. f. Bact.
II. Abt. v. 38, No. 1/6, pp. 14-25, with 5 pls.

Jonzs, M. 1905. A peculiar microorganism showing
rosette formation. In: Centralbl. f. Bact. II.
Abt. v.14, No. 15/16, pp. 459-463, with 3 figs.

Jorpan, E. 0. 1909. Textbook of general bacteriology.
557 pp., 163 figs. Philadelphia and London.

——. 1915. Variation in bacteria. In: Proceed. Na-
tion. Acad. Sci. U. 8. A., v.1, No. 3, pp. 160-164.

———. 1916. A textbook of general bacteriology. 5th
ed. 647 pp., with 177 figs. in the text. Philadel-
phia and London.

Joraz, R. 1896. Uber einen neuen Wasservibrio. In:
Ceniralbl. f. Bact. I. Abt. v. 19, No. 8, pp.
277-281.

Katizrt, E. 1914. Untersuchungen iiber Maul-und
Klauenseuche. I. Mitt. In: Arb. a. d. K. Ge
sundh. Amie, v. 47, No. 4, pp. 591-602, m. 2 Taf.

Kammura, J. 1902. Uber ein polychromes Kérperchen
bei einer noch nicht bekannten Art von Mikroor-
ganismen. In: Mitt. d. Med. Ges. zu Tokio, v. 16,
No. 4, p. 1 (Japanese). Abstr. in: Centralbl. £.
Bact. I. Abt. Ref., v. 34, 1903, No. 2/3, p. 50.

Kanrnack, A. 1896. Uber verzweigte Diphtherie-
bacillen. In: Centralbl. f. Bact. I. Abt. v. 20,
No. 8/9, pp. 296-297.

——— and Connztt, T. W. 1897. The flagella of the
tetanus bacillus and other contributions to the mor-
phology of the tetanus bacillus. In: Journ. of
Pathol. and Bact., v. 4, pp. 452-459, with 3 pls.

Karudsxi, J. 1889. Ein neuer pathogener Spaltpilz
(B., murisepticus pleomorphus). In: Centralbl. f.
Bact., v. 5, No. 6, pp. 193-207, m. 1 Taf.

Karsten, H. 1869. Chemismus der Pflanzenzelle.
90 pp., m. 9 Abb. im Text. Wien.

Karwacki, L. 1912. Uber die Morphologie des Spiro-
chaeta Obermeieri, kultiviert im Blutegel. In:
Centralbl. f. Bact. I. Abt. Orig. v. 62, No. 3/4,
pp. 250-268, m. 1 Taf.
228

Kayser, E. 1914. Microbiologie agricole. 3™°.
vin et 572 pp., av. 95 figs. dans le texte. Paris.

Keprowszr,W.J. 1901. Uber die Cultur der Lepraerreger.
In: Zeitschr. f. Hyg., v. 37, No. 1, pp. 52-69, m.
1 Taf.

“1910. Experimentelle Untersuchungen iiber
Lepraimpfungen bei Tieren. In: Zeitschr. i. Hyg. ”
v. 66, No. 1, pp. 1-100, m. 12 Taf.

Kepzior. 1896. Uber eine therméphile Cladothrix.
In: Arch.-f. Hyg., v. 27, No. 4, pp. 328-338.
Keiierman, K. F., McBers, I. G., Scauzs, F. M., and
’ Surra, N. R. 1918. Identification and classifica-

tion of cellulose dissolving bacteria. In: Centralbl.

éd.

 

f. Bact. II. Abt. v. 39, No. 20/22, pp. 502-522, |

with 2 pls.
and Scaues, F. M. 1916. Life cycles of B. coli.

 

Preliminary paper, read at the New Haven meet-

ing of the Soc. Amer. Bacteriologists.

Kenpatt, A.J. 1916. Bacteriology, general, patholog-
ical and intestinal. x and 651 pp., with 9 pls.
and 98 ill. in the text. Philadelphia and New
York.

Keysseritz, G. 1907. Uher die undulierende Mem-
bran bei Trypanosomen und Spirochaeten. In:
Arch, f. Protistenk., v. 10, No. 1, pp. 127-138, m.
1 Taf.

Kine, W. F., Bazstack, F. W., and Horrmann, G. L.
1913. Studies on the virus of hog cholera. In:
Journ. Infect. Diseases, v. 12, No. 2, pp. 206-235.

——— and Horrmann, G. L. 1913. Spirochaeta suis,
its significance as a pathogenic organism, In:
Journ. Infect. Diseases, v. 13, No. 3, pp.463~-498,
‘with 1 pl.

Krecuenstein, A, 1912. Ein Beitrag zur Sporenfrage
und Sporenfirbung der Tuberkuloseerreger. nl:
Centralbl. f. Bact. I. Abt. Orig. v. 66, No. 1,
pp. 144-159. s

——. 1913. Uher “Splitter” im Sputum von Phthi-
sikern. In: Deutsche med. Wochenschr., v. 39.
No. 28, pp. 1871-1372.

Kissxaut. 1906. Ein Fall von Erkrankung durch B.
funduliformis. In: Ber. d. Oberhess. Ges. f.
Natur- u. Heilk. zu Giessen. N.F. Med. Abt.,
v.1, pp. 68-74. Abstr. in: Centralbl. f. Bact. I.
Abt. Ref., v. 40, No. 1/3, p. 44.

Krrasato, 8. 1889. Die Widerstandsfihigkeit der Cho.
lerabacterien gegen das Eintrocknen und gegen-
Hitze. In: Zeitschr. f. Hyg., v. 5, No. 1, pp,
184-140.

Kirt, Ta. 1887. Der Rauschbrand. In: Centralbl. f,
Bact., v. 1, No. 23, pp. 684-690; No. 24, pp. 716-723;
No. 25, pp. 741-757.

———. 1888. Der Micrococcus ascoformans und das
Mykofibrom des Pferdes. In: Centralbl. f. Bact.
v. 3, No. 6, pp. 177-182; No. 7, pp. 207-210; No. 8,
pp. 246-248.

———. 1897. Die Streptothrixform des Rotlaufbacil-
lus. In: Centralbl. f. Bact. I, Abt. v. 22, No.
24/25, pp. 726-732.

——. 1898. Berichtigung meiner Mitteilung tiber
die Streptothrixform des Rotlaufbacillus. In:
Centralbl. f. Bact. I. Abt. v. 23, No. 14, pp.
601-602.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

[Von XVI.

Kuess, E. 1872. Beitrige zur Pathologie. der Schuss-
wunden. xund137pp.,m.10Taf. Leipzig.
——. 1878. Beitrige zur Kenntnis der Micrococcen.
In: Arch. f. exper. Pathol. u. Pharmacie, v. 1, No.

1, pp. 31-64, m. 4 Taf.

——. 1875a. Beitrige zur Kenntnis der pathogenen
Schistomykosen. IV. In: Arch. f. exper. Pathol.
u. Pharmacie, v. 4, No. 1/2, pp. 107-136, m. 1 Taf.

. 18750. Beitrige zur Kenntnis der pathogenen

Schistomyceten. V. In: Arch. f. exper. Pathol.

u. Pharmacie, v. 4, No. 8, pp. 207-247, m. 2 Taf.

. 1878. Weitere Beitrage zur Entstehungsgechichte
der Endocarditis. In: Arch. f. exper. Pathol. u.
Pharmacie, v. 9, No. 1/2, pp..52-94, m.-1 Taf.

——. 1883c. Weitere Beitrige zur Geschichte der
Tuberkulose. In: Arch. f. exper. Pathol. u. Phar-
macie, v. 17, No. 1/2, pp. 1-52, m. 3 Taf.

18836. Tuberkulose. In: Real-Encyclopaedie

der gesamt. Heilkunde, herausg. v. A. Eulenburg,

v. 14, pp. 123-144. Wien und Leipzig.

1887, Allgemeine Pathologie. I. Teil. x1v
and 514 pp., m. 8 Taf. u. 66 Abb. im Text. Jena,

Kizm, E. 1883. On a morphological variety of B. an-
thracis. In: Quart. Journ. Microsc. Science N. S.,
v. 28, No. 90, pp. 260-264, with 1 pl.

1885a. The relation of bacteria to asiatic chol-
era.. In: Brit. Med. Journ., 1885, I, No. 1265, pp.

650-652.

1885. Microorganisms and disease. 2d ed.
xu and 201 pp., with 108 engravings. London.
1890. Ein weiterer Beitrag zur Aetiologie der

Diphtherie. In: Centralbl. f. Bact., v..7, No. 25,

pp. 785-794.

——. 1894. A contribution to the morphology of bac-
teria. In: Quart. Journ. Microsc. Science N. S.,
v. 36, No. 141, pp. 1-9, with 1 pl.

1897. Ein Beitrag zur Morphologie und Biologie
des Bacillus der Bubonenpest. In: Centralbl. f.
Bact. I. Abt. v.21, No. 24/25, pp. 897-901.

———. 18992. Ein Beitrag zur Bakteriologie der
Leichenverwesung. In: Centralbl.. f. Bact. I,
Abt. v. 25, No. 8/9, pp. 278-284,

1899b. Ein Beitrag zur Kenntnis der Verbrei-
tung des Bacillus pseudotuberculosis. In: Cen-
tralbl. f. Bact. I. Abt. v. 26, No. 9, pp. 260-261.

——. 1900. Uber zwei neue pyogene Mikroben: Strep-
tococcus radiatus und Bacterium diphtheroides.
In: Centralbl. f. Bact. I. Abt. v. 28, No. 14/15,
pp. 417-419.

1905. Uber einen neven tierpathogenen Vibrio-
Vibrio cardii. In: Centralbl. f. Bact. I. Abt.
Orig. v. 38, No. 2, pp. 173-174.

Kizm, J. 1912. ther die sogenannte Mutation und die
Verinderlichkeit des Garvermégens bei Bakterien,
In: Zeitachr. f. Hyg., v.73, No. 1, pp. 87-118.

Kustn, L. 1889a. Uber einen neven Typus der Spo-
renbildung bei den endosporen Bakterien. In:
Ber, d. Deutschen Bot. Ges., v. 7, Gen. Vers. Heit,
pp. (57)-(72), m. 1 Taf.

——. 1889). Botanische Bakterienstudien. 2. Zur
Entwicklungsgeschichte einer neuen ““pleomor-
phen” Spaltpilzart: Bacillus allantoides n. sp. In:
Centralbl. f. Bact., v. 6, No. 14, pp. 383-386.

 

 

 

 

 

 

 

 

 

 
No. 2.] LIFE CYCLES OF THE

Kiet, 8. R. 1911. Negri-Kérperchen, bloss Uberbleib-
sel von Streptokokken. In: Allg. Wiener med.
Zeitg., v. 56, No. 29, p. 326.

Kuzpzorr, K. 1912. Evolution des Tuberkelbacillus.
In: Centralbl. f. Bact. I. Abt. Ref., v.53, No. 11,
p. 331.

Kurerer, I. J. 1913. A systematic study of the Cocca-

ceae in the collection of the Museum of Natural His- |.

tory. In: Journ. Infect. Diseases, v. 12, No. 3, pp.
432-482.

Kumenxo, W. N. 1914. Die Bedeutung der Spindel-
bacillen in der Pathologie des Scharlachs. I. Mitt.
In: Centralbl. f. Bact. I. Abt. Orig. v. 74, No.
5/6, pp. 487-494.

Kiicker, A., und Scménnine, H. 1899. Uber Durch-
wachsungen und abnorme Konidienbildungen bei
Dematium pullulans De Bary und bei anderen Pil-
zen. In: Centralbl. f. Bact. II. Abt. v. 5, No.
14, pp. 505-507.

Kwnacx, A. V. 1915. Die Untersuchung im kiinstlichen
Dunkelfeld. In: Centralbl. f. Bact. I. Abt.
Orig. v. 76, No. 2/3, pp. 235-236.

Kno, W. 1910. Morphologisches und Biologisches
tiber mit Methylviolett-Fuchsin gefiirbtes Tuber-
kulosevirus. In: Beitr. z. Klinik d. Tuberkulose,
v.15, pp. 211-217, m. 1 Tat.

Kny, L. 18794. Uber die Wurzelanschwellungen der
Leguminosen und ihre Erzeugung durch Einfluss
von Parasiten. In: Bot. Zeitg. v. 37, No. 4, p. 58.

18796. Zu dem Aufsatze des Herrn Prof. Frank
“Uber die Parasiten in den Wurzelanschwellungen

der Papilionaceen.’”’ In: Bot. Zeitg., v. 37, No. 34,
pp. 537-541.

Kocs, A. 1888. Uber Morphologie und Entwicklungs-
geschichte einiger endosporen Bacterienaiten. In:
Bot. Zeitg., v. 46, No. 18, pp. 277-287; No. 19, pp.
293-299; No. 20, pp. 309-318; No. 21, pp. 325-332;
No, 22, pp. 341-350, m. 1 Taf.

Kocu, J., und Risstine, P. 1910. Studien zur Aetiologie
der Tollwut. In: Zeitschr. f. Hyg., v. 65, No. 1,
pp. 85-112, m. 3 Taf.

Kocu, R. 1876. Untersuchungen iiber Bacterien. V.
Die Aetiologie der Milzbrandkrankheit, etc. In:
Cohn’s Beitr. z. Biol. d. Pflanzen, v. 2, Heft 2, pp.
277-310, m. 1 Taf.

1877. Untersuchungen iiber Bacterien. VI.

Verfahren zur Untersuchung, zum Conservieren

und Photographieren der Bacterien. In: Cohn’s

Beitr. z. Biol. d. Pflanzen, v. 2, Heft 3, pp. 399-434,

m. 3 Taf.

1878, Untersuchungen itiber die Aetiologie der
Wundkrankheiten. Translat. by W. W. Cheyne.
(‘Investigations into the etiology of traumatic
infective diseases,” v. LXXXVIII. The New
Sydenham Soc.) xm and 74 pp., with 5 pls.
London, 1880.

. 1881. Zur Untersuchung von pathogenen Orga-

nismen. In: Mitt. a. d. K. Gesundh. Amte, v. 1,

pp- 1-48, m. 14 Taf.

1882. Die Aetiologie der Tuberkulose. In:

Berliner klin. Wochenschr.,v. 19, no. 15, pp. 221-230.

1884a. Die Aetiologie der Tuberkulose. In:

Mitt. a. d. K. Gesundh. Amte, v. 2, pp. 1-88, m.

10 Taf.

 

 

 

 

 

 

 

229

R. 18845. Uber die Cholerabacterien. In:
Deutsche med. Wochenschr., v. 10, no. 45, pp. 725-
728.

——. 1890. Uber bacteriologische Forschung. In:
Verhandl. d. X. Internat. Med. Congresses (Ber-
lin), v.1, pp. 35-47.

Kopama. 1908. Uber die Morphologie des Pestbacillus.

In: Mitt. d. Japan. Hyg. Ges., v. 2, No.4, Abstr.

BACTERIA—LOHNIS.

Koca,

in: Centralbl.f. Bact. I.Abt. Ref., v.42, No. 4/6,
p. 120.
Koresrt, A. 1914. Zur Yoghurtkontrolle. In: Cen-

tralbl.f. Bact. II. Abt. v.42, No. 17/18, pp. 449-
479.

Koéutzr, Fr. 1902. Das Agglutinationsphinomen. In:
Klin. Jahrb., v. 8, pp. 39-160.

K6uuiscu. 1915. Bakteriologische Befunde bei einem
Fall von Meningokokkensepsis, etc. In: Zeitschr.
f. Hyg., v- 80, No. 3, pp. 404-430, m. 1 Taf.

——. 1916. Gelbwachsende, den Bacillen der Typhus-
Paratyphus-Gruppe 4bhnliche Bakterien. In:
Centralbl. Bact. I. Abt. Orig. Vol. 78, N. 3, pp.
136-141.

1918.  Herausziichtung eines Paratyphus-B

Stammes aus einem Typhusstamm in Rinderdarm-
schleim. Centralbl. Bact. I. Abt. Orig. Vol. 81,
No. 1/2, pp. 6-9.

Kousrvees, J. H. F. 1900. Vibrionen-Studien. II.
Panmorphismus und erbliche Variationen. In:
Centralbl. f. Bact. I. Abt. v. 28, No. 24, pp. 833-
842.

. 190la. Vibrionen-Studien. III. Agglutina-

tion durch Toxine zur Bestimmung der Virulenz.

In: Centralbl. f. Bact. I. Abt. v. 30, No. 18,

pp. 689-696.

. 1901b. Symbiose zweier pleomorpher Faeces-
Bakterien. In: Virchow’s Arch. f. pathol. Anato-
mnie, v. 163, No. 3, pp. 365-379, m. 1 Taf.

Komwzumi1, M. 1912. On the nature of the ‘‘margina-
points” occurring in the blood corpuscles of cattle.
In: Centralbl. f. Bact. I. Abt. Orig. v. 65, No.
4/5, pp. 337-340, with 1 pl.

Koutz, W. 1903. Cholera asiatica. In: Kolle-Wasser-

mann’s Handb. d. pathog. Mikroorg., v. 3, pp. 1-74.
und GortscuticH, E. 1903. Untersuchungen

tiber die bakteriologische Choleradiagnostik und

Specifitét des Koch’schen Choleravibrio. In: Zeit-

schr. f. Hyg., v. 44, No. 1, pp. 1-128.

und Hetscu, H. 1911. Experimentelle Bakterio-
logie, etc. 3. Aufl, 2 Bde. xvr and 968 pp., m.
98 Taf., 180 Textabb. u. 10 Karten. Berlin und
Wien.

Kotmer, J. A. 1912. <A contribution to the bacteriology
of diphtheria. In: Journ. Infect. Diseases, v. 11,
No. 1, pp. 56-70.

Korn, O. 1899. Zur Kenntnis der siurefesten Bakte-
rien. In: Centralbl. f. Bact. I. Abt. v. 25, No.
15/16, pp. 532-541.

Kowatenxo, A. 1910. Studien iiber sogenannte Muta-
tionserscheinungen bei Bakterien, etc. In: Zeit-
schr. f. Hyg., v. 66, No. 2, pp. 277-290, m. 1 Taf.

Krainsry, A. 1914. Die Aktinomyceten und ihre
Bedeutung in der Natur. In: Centralbl. f. Bact.
II. Abt. v. 41, No. 24/25, pp. 649-688 m. 2 Taf.

 

 

 

 

 
230

Krasxowsxka, L., und Nirscu, R. 1918. Zur Mor-
phologie der Streptokokken. In:.Centralbl. Bakt.
I. Abt. Orig. v. 82, No. 3/4, pp. 264-270, 1 Taf.

Kraus, A. 1913. Uber das Wesen des sogenannten
Unna’schen Flaschenbazillus. In: Arch. f. Der-
matol. u. Syphilis, v. 116, No. 3, pp. 723-736, m.
2 Taf.

Krienirz, W. 1906. Uber morphologische Verinde- |

rungen an Spirochaeten. In: Centralbl. f. Bact.
I. Abt. Orig., v. 42, No. 1, pp. 48-47.

Krrrscuewsky, J., und Brraer, O. 1912. Uber die
Beziehung des B. leprae Hansen zu einigen bei
Lepra isolierten Mikroorganismen. In: Charkoff
Med. Journ., v. 14, No. 7, p. 101. Abstr. in: Cen-
tralbl. f. Bact. I. Abt. Ref., v. 56, 1913, No. 14,
p. 98.

Kritezr, W. 1898. Uber den Salpeterpilz von Stutzer
und Hartleb. In: Centralbl. f. ‘Bact. II. Abt.
v. 4, No. 5, pp. 184-188.

Krumwirpe, Cu., and Prarr, J. S. 1913. Fusiform
bacilli: Cultural characteristics. In: Journ. Infect.
Diseases, v. 13, No. 3, pp. 438-441.

Kruse, W. 1896a. Allgemeine Morphologie der Bak-
terien. In: Fliigge’s ‘‘Mikroorganismen.” 38. Aufl.

v. 1, pp. 44-79.

18966. Allgemeine Morphologie der Protozoen.

 

In: Fliigge’s ‘‘Mikroorganismen.”” 3, Aufl. v. 1,
pp. 79-83.

———. 1896c. Variabilitét der Mikroorganismen. In:
Fligge’s ‘‘Mikroorganismen.”” 3. Aufl. v. 1,
pp. 475-493.

——. 1896d. Systematik der Streptotricheen. In:
Fligge’s ‘‘Mikroorganismen.” 3. Aufl. v. 2,
pp. 48-66.

1903. Das Verhiltnis der Milchsiéurebakterien

zum Streptococcus lanceolatus (Pneumoniecoccus,

Enterococcus u. s. w.). In: Centralbl. f. Bact.

I. Abt. Orig. v. 34, No. 8, pp. 737-739.

1910. Allgemeine Mikrobiologie. xv and 1184
pp. Leipzig.
1914. Die Erreger von Husten and Schnupfen.

In: Miinchn. med. Wochenschr., v. 61, No. 28,
pp. 1547.

—— und Pansim1, S. 1892. Untersuchungen tiber
den Diplococcus pneumoniae und verwandte Strep-
tokokken. In: Zeitschr. f. Hyg., v.11, pp. 279-380.

Kroyrr, E. pz. 1908. Die Lebensgeschichte von Myxo-
coccus javanensissp.n. In:Centralbl.f. Bact. IT.
Abt. v.21, No. 13/14, pp. 385-386.

Krrtow, 0. 1911. Uber die Bedeutung und das Vor-
kommen der Much’schen Granula. In: Zeitschr. f.
Hyg., v. 70, No. 1, pp. 135-148.

KrzemMreniewskl, 8. 1908. Untersuchungen tiber Azoto-
bacter chroococcum Beij. In: Bull. Intern. Acad.
Cracovie. Cl. d. sci. math. et nat., 1908, No. 9,
pp. 929-1051, m.1Taf.

KrzyszTatowicz, Fr., et Srepticxr, M. 1908. Etude
expérimentale de la syphilis. Morphologie de
Spirochaeta pallida. In: Bull. Intern. Acad. Cra-
covie. Cl. d. sci. math. et nat., 1908, No. 3, pp.
173-245, av. 2 pls.

Ktwstizr, J. 1884. Bacterioidomonas sporifera. In:
Journ, de microgr., v. 8, No. 7, pp..376-380, av. 1 pl.

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

[Vou. XVI,

Kinstier, J. 1885. De la position systématique des
bactériacées. In: Journ. de microgr., v. 9, No. 6,
pp. 248-258; No. 7, pp. 295-307, av. 1 pl.

Kuntze, W. 1904. Beitrige zur Morphologie und Phy-
siologie der Bakterien. In: Centralbl.f. Bact. IT.
Abt. v.13, No. 1/3, pp. 1-12, m. 1 Taf,

——. 1908. Studien iiber fermentierte Milch. I.
Yoghurt und Mazun. In: Centralbl. f. Bact. IT.
Abt. v. 21, No. 24/25, pp. 737768, m. 3 Taf.

Kurrs,H. 1883. Bacterium Zopfii. Ein Beitrag zur
Kenntnis der Morphologie und Physiologie der
Spaltpilze.. In: Bot. Zeitg., v. 41, No. 23, pp. 369-
386; No. 24, pp. 393-405; No. 25, pp. 409-420; No.
26, pp. 425-435, m. 1 Taf.

——. 1898. Uher die Diagnose des Diphtheriebacillus
unter Beriicksichtigung abweichender Culturfor-
men desselben. In: Zeitschr. f. Hye. + V.28, pp.
409-438, m. 1 Taf.

Kurscurer, H. 1895a. Die Vibrionen- und Spirillen-
flora der Diingerjauche. In: Zeitschr. f. Hyg., v.
20, No. 1, pp. 46-59. ‘

——. 1895b. Spirillum Undula minus und Spirillum
Undula majus. In: Centralbl. f. Bact. I. Abt.
v.18, No. 20/21, pp. 614-616.

——. 1907. Epidemische Genickstarre. In: Kolle-
Wassermann’s Handb. d. pathog. Mikroorg. I.
Erg. Bd., pp. 481-521.

Laconer-Sanpovat, V. 1898. Uber Strahlenpilze.
Diss. med. Strassburg. 75 pp., m. 1 Taf.

Larar, F. 1895. Physiologische Studien iiber Essig-
garung und Schnellessigfabrikation. In: Centralbl.
f. Bact. II. Abt. v.1, No. 4/5, pp. 129-150.

1897. Technische Mykologie. v. I. xm und
362 pp. Jena. :

Laceruem, G. 1900. Mykologische Studien. III.
Beitrige zur Kenntnis der parasitischen Bakterien
und der bakteroiden Pilze. Bihang till Svenska
Vet.- Akad. Handl., v. 26, afd. III, No. 4, 21 pp.,
1 tav.

Lanrorp, J. A. 1914, Studies on diphtheroid organisms
isolated from lymphadenomatous structures of the
body. In: Amer. Journ. Tropic. Diseases, v. 2, No.
3, pp. 191-198.

Laneser, R. 1904. Untersuchungen itiber einen mit
Knétchenbildung einhergehenden Prozess in der
Leber des Kalbes und dessen Erreger. In: Zeit-
schr. f. Hyg., v. 47, No. 3, pp. 353-382, m. 1 Taf.

Lanxester, E. R. 1873. On a peach-coloured bacter-
ium—Bacterium rubescens. n. sp. In: Quart.
Journ. Microsc. Science N. 8., v. a8; pp. 408-425,
with 2 pls.

1876. Further observations on a peach-coloured
bacterium. In: Quart. Journ. Microsc. Science N.
8, v. 16, no. 1, pp. 27-40, with 1 pl.

Larson, W. P., and Barron, M. 1918. Report of a
case in which the fusiform bacillus was isolated
from theblood stream. In: Journ. Infect. Diseases,
v.18, No. 3, pp. 429-457.

Laurent, E. 1890. Sur le microbe des nodosités radi-
cales des légumineuses. In: Compt. rend. Acad.
Paris, v. 111, No. 20, pp. 754-756.

——. 1891. Recherches sur les nodosités radicales
deslégumineuses. In: Annal. Inst. Pasteur, v. 5,
No. 2, pp. 105-139, av. 2 pls.

 

 
No. 2.] LIFE CYCLES OF THE

Laxa,O, 1911. La désinfection dans la laiterie par la
voie séche. In: Rev. génér. du lait, v. 9, No. 1,
pp. 8-16.

Lz Braye, R., et Guecennzem, H. 1914. Manuel pra-
‘tique de diagnostic bactériologique. 444pp. Paris.

Lenmann, K. B. 1887. Uber die Sporenbildung bei
Milzbrand. In: Miinchn. med. Wochenschr., v. 34,
No. 26, pp. 485-488.

——. 1899a. Einige Bemerkungen zur Geisselfrage.
In: Arch. f. Hyg., v. 34, No. 8, pp. 198-199.

——. 1899b.  Notiz:iiber den Bacillus mycoides.
Arch.f. Hyg., v.35, No.1, p.10.

——und Neumann, R.O. 1912. Atlas und Grundriss
der Bakteriologie. 5. Aufl. II. Teil., xv und
777 pp. ‘Miinchen.

Leicamann, G., und BazarEwsx!, 8. von. 1900. Uber
einige in reifen Kasen gefundene Milchsiéurebak-
terien. In: Centralbl.f. Bact. II. Abt. v.6, No.
8, pp. 245-253; No. 9, pp. 281-285; No. 10, pp. 314-
331. 5

Leisoman, W. B. 1909. Preliminary notes on experi-

ments in connection with the transmission of tick

fever. In: Journ. Roy. Army Med. Corps, v. 12, No.

2, pp. 123-135, with 1 pl.

1910. The mechanism of infection in tick fever
and on the hereditary transmission of Spirochaeta
~Duttoni in the tick. In: Lancet, v. 88, 1910, I, pp.

11-14,

——. 1918. A note on the “granule clumps” found in
Ornithodorus moubata and their: relations to the
spirochaetes in African relapsing fever (tick fever).
In: Annal, Inst. Pasteur, v. 32, No. 2, pp. 49-59.

Lemoine, G.H. 1896. Variabilité dans la forme et dans
les caractéres du culture du streptocoque. In:
Arch, de méd. exp. et d’anat. pathol., v. 8, No. 2,
pp. 156-167.

Lerrescuxin, W. W. 1903. Zur Kenntnis der Erblich-
keit bei den einzelligen Organismen. In: Cen-
tralbl. f. Bact. II. Abt. v.10,No. 5, pp. 141-151,
m. 1 Taf.

In:

einzelligen Organismen. Die Verzweigung und
Mycelbildung bei einer Bakterie (Bacillus Berest-
nevin.sp.). In: Centralbl. f. Bact. II. Abt.

Vv. 12, No. 22/24, pp. 641-648; v. 13, No. 1/3, pp.
13-22, m. 20 Fig.

Lescuxe, E. 1911. Uber die granulare Form des Tuber,
kulosevirus. In: Centralbl. f. Bact. I. Abt-
Orig. v. 59, No. 4, pp. 365-371. :

Lerzericu, L. 1873. Die Entwicklung des Diphtherie-
pilzes. In: Virchow’s Arch. f. pathol. Anat. v.
58, No. 2, pp. 303-310, m. 1 Taf.

—. 1874. Die lokale und allgemeine Diphtherie.
In: Virchow’s Arch. f. pathol. Anat., v. 61, No. 4,
pp. 457-484, m. 1 Taf.

—. 1876. Experimentelle Untersuchungen. In:
Virchow’s Arch. f. pathol. Anat., v..68, No. 4, pp.
519-546.

. 1878. Experimentelle Untersuchungen  tiber

 

Typhus abdominalis. In: Arch. f. exper. Pathol.
u. Pharm., v. 9, No. 3/4, pp. 312-328.

1880. Untersuchungen iiber die morphologi-
schen Unterschiede einiger pathogenen Schistomy-

 

1904, Zur Kenninis der Erblichkeit bei den |

 

231

ceten. In: Arch. f. experim. Pathol. u. Pharm.,
v. 12, No. 5/6, pp. 351-365, m. 1 Taf.

Lzucus, G. 1905. Sind bei der bakteriziden Wirkung
des Blutserums osmotische Vorgingeim Spiel? In:
Arch. f. Hyg., v. 54, No. 4, pp. 396-412.

Leuriaux, C., et Geers, V. 1906. Culture du Tré-
ponema pallidum de Schaudinn. In: Centralbl.
f. Bact. I. Abt. Orig. v. 41, No. 6, pp. 684-688.

Levine, M., Wexoin, J. C., and Jounson, B. R. 1917.
The Voges-Proskauer and correlated reactions of
Coli-like bacteria. In: Journ. Infect. Diseases, v.
21, No. 1, pp. 39-46.

Levy, E. 1897. Ein neues, aus einem Fall von Lepra

geziichtetes Bacterium aus der Klasse der Tuber-

kelbacillen. Studien tiber diese Klasse. In: Arch.

f. Hyg., v. 30, No. 2, pp. 168-183.

1899. Uber die Actinomycesgruppe (Aktinomy-
ceten) und die ihr verwandten Bakterien. In:
Centralbl. f. Bact. I. Abt. v.26, No. 1, pp. 1-11.

1902. Die Wachstums- und. Dauerformen der
Strahlenpilze (Aktinomyceten) und ihre Bezie-
hungen zu den Bakterien. In: Centralbl. f. Bact.
I. Abt. Orig. v. 33, No. 1, pp. 18-23.
and Kurmperer, F. 1900. Elements of Clinical
Bacteriology. 2d ed. Auth. transl. by A. A.
Eschner. 441 pp., 92 figs. Philadelphia.

Levy, Fr. 1904. Hygienische Untersuchungen iiber
Mehl und Brot. In: Arch. f. Hyg., v. 49, No. 1,
pp. 62-112.

1916. UWher Copulationsvorgiinge-(?) bei Spiro-
chaeta Obermeieri. In: Arch. f. Protistenk., v. 36,
No. 3, pp. 362-363.

Lewxowicz, X. 1906. Uber die Reinkulturen des fusi-
formen Bacillus. In: Centralbl. f. Bact. I. Abt.
Orig. v. 41, No. 2, pp. 153-155, m. 1 Taf.

Lienrzres, J., ET DuRRizEv, H. 1902. Formesanormales

de la bactéridie charbonneuse isolées dans le sang

d’un taureau mort de maladie naturelle. In: Bull.

Soc. centr. méd. vét. N.S., v. 20, No. 5, pp. 102-107.

et Sprrz. 1902. Actinobacillose. In: Bull. Soc.

centr. méd. vét. N.8., v. 20, No. 18, pp. 487-535.

Linpau, G. 1904. Allgemeine Morphologie, Entwick-
lungsgeschichte, Anatomie und Systematik ‘der
Eumyceten. In: Lafar’s Handb. d. Techn. Myko-
logie, v. 1, pp. 150-221.

Lirscutitz, B. 1908. Uber mikroskopisch sichtbare, fil-

trierbare Virusarten (Uber Strongyloplasmen). In:

Centralbl, f. Bact. I. Abt. Orig. v. 48, No. 1,

pp. 77-90.

1909. Uber mikroskopisch sichtbare, filtrierbare
Virusarten. In: Centralbl.f. Bact. I. Abt. Retf.,

v. 44, Beiheft, p. 101.

Lister, J. 1873. A further contribution to the natural
history of bacteria and the germ theory of fermenta-
tive changes. In: Quart. Journ. Microsc. Science
n.8., v. 13, pp. 380-408, with 3 pls.

1878. On the nature of fermentation. In:
Quart. Journ. Microsc. Science N. 8., v. 18, No. 70,
pp. 177-194.

Lors, L.M. 1902. On branching forms of certain bac-
teria. In: Journ. Med. Res., v. 8, No. 2, pp. 415-
423,

BACTERIA—LOHNIS.

 

 

 

 

 

 
232

-Lozrrier, F, 1884. Untersuchungen iiber die Bedeu-
tung der Mikroorganismen fiir die Entstehung der
Diphtherie, etc. In: Mitt. a..d. K. Gesundh.
Amte, v. 2, pp. 421-499.

1887. Vorlesungen tiber die geschichtliche Ent-
wicklung der Lehre von den Bakterien. I. Teil.
xu und 252 pp., m. 3 Taf. u. 37 Abb. im Text.
Leipzig.

1911. Uber filtrierbares Virus. In: Centralbl.
f. Bact. I. Abt. Ref., v. 50, Beiheft, pp. 1-12.

Loéunis, F. 1905a. Beitrage zur Kenntnis der Stick-

stoffbakterien. In: Centralbl. f. Bact. II. Abt.

v.14, No. 18/20, pp. 582-604; No. 22/23, pp. 713-728.

. 19056. Bacterium agreste n.sp. In: Centralbl.

f. Bact. I. Abt. Orig. v. 40, No. 2, pp. 177-180.

1907. Versuch einer Gruppierung der Milch-

sdurebakterien. In: Centralbl. f. Bact.

v.18, No. 4/6, pp. 97-149.

1910. Handbuch der landwirtschaftlichen
Bakteriologie. xm und 907 pp. Berlin.

———. 1911. Zur Kenntnis und Benennung der in
Milch und Molkereiprodukten vorkommenden
Bakterien. In: Centralbl. f. Bact. ITI. Abt. v.
29, No. 12/14, pp. 331-340.

1913. Vorlesungen iiber landwirtschaftliche
Bakteriologie. vi und 398 pp., m. 10 Taf. u. 60
Abb. im Text. Berlin.

—— und Hanzawa,J. 1914. Die Stellung von Azoto-
bacter im System. In: Centralbl. f. Bact. IT.
Abt. v. 42, No. 1/4, pp. 1-8, m. 2 Taf.

—— und Priuar, N. K. 1907. Uber Stickstoff fi-

xierende Bakterien. II. In: Centralbl. f. Bact.

II. Abt. v.19, No. 1/3, p. 87-96, m. 1 Taf.

and Surro, N. R. 1916a. Life cycles of the
bacteria (Preliminary communication). In: Journ.

Agric. Res., v. 6, no. 18, pp. 675-702, with 7 pls.

. 19166. Life cycles of the bacteria.
Paper read at the New Haven meeting of the Soc.
Amer. Bacteriologists. Dec. 27, 1916.

——— und Westermann, T. 1908. Uber Stickstoff
fixierende Bakterien. “IV. In: Centralbl. f. Bact.
II, Abt. v. 22, No. 7/10, pp. 284-254, m. 1 Taf.

Lorre, W. 1908. Beitrag zur Morphologie der Acti-
nomycesdruse. In: Zeitschr. f. Hyg., v. 60, No.
2, p. 227-244, m. 2 Taf.

Lorenz. 1892. Beobachtungen tiber die Mikroorga-
nismen des Schweinerotlaufs und verwandter
Krankheiten. In: Arch. f. wissensch. u. prakt.
Tierheilk., v.18, no. 12, pp. 39-62.

—. 1909. Einiges tiber die Entwicklungsformen
des Erregers der Brustseuche. In: Berl. tierarztl.
Wochenschr., v. 25, no. 36, pp. 659-664.

Lovrens, L.F.D.E. 1907. Untersuchungen iiber die
Filtrierbarkeit der Schweinepestbacillen (B. sui-
pestifer). In: Centralbl. f. Bact. I. Abt. Orig.
v. 44, no. 5, pp. 420-427; no. 6, pp. 504-512; no.
7, pp. 630-648.

Lusarsco, O. 1899. Zur Kenntnis der Strahlenpilze.
In: Zeitschr. f. Hyg., v. 31, No. 1, pp. 157-220,
m. 1 Taf.

——. 1903. ‘‘Amyloidentartung” und ‘‘Hyalin.”
In: Encyklopaedie der mikroskop. Technik, her-
ausg. v. P. Ehrlich et al. 2 Bde. 1400 pp., m.
134 Textabb. Berlin und Wien.

 

 

 

 

 

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

II. Abt.

 

[Vou XVI,

Lucxuarpt, A. E. 1901. Uber Variabilitét und Bedin-
gungen der Farbstoffbildung bei Spaltpilzen.
Diss. med. Freiburgi. B. 55 pp.

Lupwic, F. 1896. Die Genossenschaft der Baumfluss-
organismen. In: Centralbl.f. Bact. II. Abt. v.
2, No. 10/11, pp. 337-351.

Litssert, A. 1886. Biologische Spaltpilzuntersuchung.
Der Staphylococcus pyogenes aureus und der
Osteomyelitiscoccus. vr und 102 pp., m. 2 Taf.
Wiirzburg.

Livers, J. 1866. Uber Abstammung und Entwicklung
des Bact. Termo Duj., Vibrio lineola Ehrb. In:
Bot. Zeitg., v. 24, No. 5, pp. 33-39; No. 6, pp.
41-46, m. 1 Taf.

Luerssen, A., und Kian, M. 1908. Yoghurt, die bul-
garische Sauermilch. In: Centralbl. f. Bact. IT.
Abt. v. 20, No. 8/9, pp. 234-248.

Lusxa, Fr. 1914. Morphologisch-biologische Untersu-
chungen iiber die firbbaren Kérnchen im Inhalte
des Micrococcus ochraceus. In: Arch. f. Protis-
tenk., v. 33, no. 3, pp. 272-312, m. 3 Taf.

Lurz, A. 1886. Zur Morphologie des Mikroorganismus
der Lepra. In: Dermatol. Studien, herausg. v.
P. G. Unna, Heft 1, pp. 77-100 (Erg. Heft I z.
Monatshefte f. prakt. Dermatologie, Jahrg. 1886).

MacCattum, W. G. 1902. On the life history of Acti-
nomycesasteroides. In:Centralbl.f. Bact. .I. Abt.
Orig. v. 31, no. 12, pp. 529-547.

McDonacu, J. E. R. 1912a. The life cycle of the
organism of the syphilis. In: Lancet, v. 183, 1912,

II, No. 4650, pp. 1011-1012.

1912b. The complete life history of the organism
of syphilis. In: Lancet, v. 188, 1912, II, No.

4659, pp. 1650-1651.

1913. Der Lebenszyklus des Mikroorganismus
der Syphilis. In: Dermatol. Wochenschr., v. 56,
No. 15, pp. 413-420, m. 5 Taf.

McFapygean, J. 1889. The morphology. of the actino-
myces. In: Brit. Med. Journ., 1889, I, No. 1485,
pp. 1339-1344.

McFaruanp, J. 1916. Textbook upon the pathogenic
bacteria and protozoa. 8th ed. 807 pp., with 323
illustr. Philadelphia and London.

Maassen, A. 1899. Fruchtather bildende Bakterien.
In: Arb. a. d. K. Gesundh. Amte, v. 15, no. 3,
pp. 500-513, m. 1 Taf.

1904. Die teratologischen Wuchsformen (In-
volutionsformen) der Bakterien, etc. In: Arb.
a. d. K. Gesundh. Amte, v. 24, no. 3, pp. 385-402,
m. 6 Taf.

Maccmiarti, L. 1898. Uber die Biologie des Bacillus
Baccarinii Macchiatti. In: Centralbl. f. Bact. II.
Abt. v. 4, no. 8, pp. 332-340.

Mact, E. 1897. Traité pratique de bactériologie. 3™°
éd. vu et 1144 pp., av. 240 figs. dans le texte.
Paris,

Macaig, F.P. 1905. Vincent's angina and the Bacillus
fusiformis. In: Lancet, v. 88, 1905, II, no. 4271,
pp. 110-111.

Mappox, R. L. 1885. On some unusual form of lactic
ferment—Bacterium lactis. In: Journ. Roy.
Microsc. Soc., Ser. IT, v. 5, pt. 2, pp. 205-206.

Mapsen, Tx. 1897. Zur Biologie des Diphtheriebacillus.
In: Zeitschr. f. Hyg., v. 26, p. 157-192, m. 1 Taf.

 

 

 
No, 2.] LIFE CYCLES OF THE

Marvcct, A. 1892. Die Hihnertuberkulose. In: Zeit-
schr. f. Hyg., v. 11, pp. 445-486, m. 1 Taf.

Maenin, A., and Sternsperc, G. M. 1884. Bacteria.
xvmi and 494 pp., with 12 pls. New York.

Maenus. 1903. Zellchemie. In: Encyklopaedie der
mikrosk. Technik, herausg. v. P. Ehrlich et al.
2 Bde. 1400 pp., m. 134 Textabb. Berlin und
Wien.

Mauer, St. J. 1910. The ee of the acid-fast
tubercle bacillus to other forms of bacterial life.
In: Med. Record, v. 78, No. 20, pp. 845-849.

——.. 1913. The progeny of the tubercle bacillus.
In: Med. Record, v. 84, No. 26, pp. 1162-1163.

——. 1915. The bacterial flora of trees and men. In:
Scientif. American, v. 80, Suppl. No. 2070, p. 158.

MaaPert-NEUFVILLE, R. von. 1886. Bakteriologische
Untersuchung der wichtigsten Quellen der stidt-
ischen Wasserleitung Wiesbadens, etc. In:
Zeitschr. f. analyt. Chemie, v. 25, p. 39-88.

Mauassez, L., et Vienat, W. 1883. Tuberculose
zoogléique. In: Archiv. physiol. norm. et pathol.
3™° sér. v. 2, No. 4, pp. 369-412, av. 2 pls.

—. 1884. Sur le microorganisme de la
tuberculose zoogléique. In: Arch. physiol. norm.
et pathol. 3™° sér. v. 4, No. 2, pp. 81-105, av.
2 pls.

Matrrrano, G. 1900. La bactériolyse de la bactéridie
charbonneuse. In: Compt. rend. Acad. Paris, v.
131, No. 4, pp. 295-298.

Matiory, F. B., and Mepiar, E.M. 1916. The etiology
of scarlet fever. In: Journ. Med. Res., v. 35, No. 2,
pp. 209-229, with 11 pls.

Marcuanp, H. 1912. Sur la conjugaison des ascospores
chez quelques levfres. In: Compt. rend. Soc.
Biol., v. 72, pp. 410-412.

Mark, J. G. 1914. Zur Frage der Mutation bei Pest-

 

bacillen. In: Centralbl. f. Bact. I. Abt. Orig.
v. 74, No. 7, pp. 529-540.
Marpmann, G. 1884. Die Spaltpilze. 193 pp., m. 25

Abb. Hallea. 8.

1893. Die Untersuchung des Strassenstaubes
auf Tuberkelbacillen. In: Centralbl. f. Bact., v.
14, No. 8, pp. 229-234.

1897. Zur Morphologie und Biologie des Tu-

berkelbacillus. In: Centralbl. f. Bact. I. Abt.
v. 22, No. 20/21, pp. 582-586, m. 1 Taf.

Marrassini, A. 1913. Uber das Vorhandensein einer
den Kérper einiger Bakterien umgebenden Hiille,
etc. In: Centralbl. f. Bact. I. Abt. ous: v. 71,
No. 2/3, pp. 113-128, m. 2 Taf.

Martint. 1910. Uber hohe Grade von Hebansdatee bei
Typhus, Paratyphus B, Aertryck-, Girtnerschen
Enteritis- und bei Ruhrbakterien des Typus
Shiga-Kruse, Flexner und Y. In: Zeitschr. f.
Hyg., v. 65, No. 1, pp. 121-126...

Martznowsx!, E. J. 1911. De l’étiologie de la péri-
pneumonie. In: Annal. Inst. Pasteur, v. 25, No.
12, pp. 914-917, av. 1 pl.

 

 

Marx, H. 1899. Zur Morphologie des Rotzbacillus.
In: Centralbl. f. Bact. I. Abt. v. 25, No. 8/9,
pp. 274-278,

und Worrne, Fr. 1900. Mykologische Unter-
suchungen zur Biologie der Bakterien. In:

 

 

233

Centralbl. f. Bact. I. Abt. v.28, No.1, pp. 1-11;
No. 2, pp. 83-39; No. 3, pp. 65-69; No. 4/5, pp.
97-111, m. 3 Taf.

Massart, J. 1902. Recherches sur les organismes
inférieurs. V. Sur le protoplasme des schizophytes.
In: Recueil de Jl'Inst. Botanique (Univ. de
Bruxelles), v. 5, pp. 251-294, av. 6 pls.

Massin1, R. 1907. Uher einen in biologischer Beziehung
interessanten Kolistamm (Bact. coli mutabile).
In: Arch. f. Hyg., v. 61, No. 3, pp. 250-292, m.
1 Taf.

Martuers, G. 1917. The etiology of acute poliomyelitis.

In: Journ. Infect. Diseases, v. 20, No. 1, pp. 113-

124, with 4 pls.

and Howett, K. 1917. Immune reactions in
rabbits injected with micrococci from acute polio-

myelitis. In: Journ. Infect. Diseases, v. 21,

No. 3, pp. 292-297.

Matzuscuita, T. 1900. Die Einwirkung des Kochsalz-
gehaltes des Nahrbodens auf die Wuchsform der
Mikroorganismen. In: Zeitschr. f. Hyg., v. 35, pp.
495-510, m. 6 Taf.

1902. Beobachtungen iiber einen merkwiirdigen
Teilungsprozess bei einem proteusartigen Luft-

bacillus. (Vorl. Mitt.) In: Centralb. f. Bact. II
Abt. v.9, No. 8, pp. 257-260, m. 1 Taf.

Meier, O. 1903. Uber das Wachstum der Tuberkel-
bazillen auf vegetabilischen Nahrbéden. Diss. .
med. Freiburgi. B. 63 pp., m. 2 Taf.

Metrowsky, E. 1914a. Untersuchungen iiber die Stel-
lung der Spirochaeten im System. In: Miinchn.
med. Wochenschr., v. 61, No. 11, pp. 592-596.

19146. Studien iiber die Fortpflanzung von
Bakterien, Spirillen und Spirochaeten. v1 und 95
pp., m.19 Taf. Berlin.

Metton, R. R. 1917. A study of the diphtheroid group
of organisms with special reference to their relation
to the streptococci. In: Journ. Bact., v. 2, No. 2,
pp. 81-107, with 2 pls.; No. 3, pp. 269-307; No. 4,
pp. 447-500.

Mencz, E. 1905. Cytologisches iiber die Bakterien der
Prager Wasserleitung. In: Centralbl. f. Bact. II.
Abt. v.15, No. 17/18, pp. 544-564, m. 4 Taf.

1910. Ober den Kern und seine Teilung bei
Sarcinen und Micr. ochraceus. In: Arch. f.
Protistenk., v. 19, No. 2, p. 127-148, m. 1 Taf.
1911. Die Kerniguivalente und Kerne bei

Azotobacter chroococcum, etc. In: Arch. f.
Protistenk, v. 22, No. 1, pp. 1-18, m. 1 Taf.

Mertens, V. E. 1903. Beitrage zur Aktinomykosefor-
schung. In: Zeitschr. f. Hyg., v. 42, No. 1, pp.
45-89, m. 1 Taf.

Mercunixorr, E. 1888. Uber die phagocytire Rolle
der Tuberkelriesenzellen. In: Virchow’s Arch. f.
pathol. Anat., v. 113, No. 1, pp. 63-94, m. 2 Taf.

1888). Pasteuriaramosa. In: Annal. Inst. Pas-
teur, v. 2, No. 4, pp. 165-170, av. 1 pl.
1889a. Contribution 4 l’étude du pléomor-
phisme des bactéries. In: Annal. Inst. Pasteur,
v. 3, No. 2, pp. 61-68, av. 1 pl.

1889 b. Note sur le pléomorphisme des bac-
téries. In: Annal. Inst. Pasteur, v. 3, No. 5, pp.
265-267.

BACTERIA—LOHNIS.

 

 

 

 

 

 

 

 
234

Mertcunixorr, E. 1894. Recherches sur le choléra et
. les vibrions. In: Annal. Inst. Pasteur, v. 8, No. 5,
pp. 257-274, av. 1 pl.; No. 8, pp. 529-589.
Metzner. 1903. Altmann’sche Granula-Methoden. In:
.,, Encyklopaedie d. mikrosk. Technik, herausg. v. P.
Ehrlich et al. 2 Bde.
Berlin und Wien.
Meyer, A. 1901c. Uber die Verzweigung der Bakterien.
In: Centralbl. f. Bact. I. Abt. v. 30, No. 2, pp.
49-60, m. 2 Taf.

——. 1901. Uber Chlamydosporen und iiber sich.

mit Jod ‘blau farbende Zellmembranen bei den

Bakterien. In: Ber. d. Deutschen Botan. Ges., v.:

19, No. 6, pp. 428-432, m. 1 Taf.

1903. Practicum der botanischen Bakterien-
kunde. vm und 157 pp., m. 1 Taf. u. 31 Abb. im
Text. Jena.

1905. Uber Kugelbildung und Plasmoptyse der
Bakterien. In: Ber. d. Deutschen Bot. Ges., v.
23, No. 8, pp. 349-357, m. 1 Taf.

1906. Uber Alfred Fischer’s Plasmoptyse der
Bakterien. In: Ber. d. Deutschen Bot. Ges., v.
24, No. 5, pp. 208-213.

1912. Die Zelle der Bakterien. v1 und 285 pp.,
m, 1 Taf. u. 34 Abb. im Text. Jena.

Meryer, K. F. 1911. Beitriige zur Genese und Bedeu-
tung der Kochschen Plasmakugeln in der Pathoge-.
nese des afrikanischen Kiistenfiebers. In: Cen-
_tralbl. f. Bact. I. Abt. Orig. v. 57, No. 5, pp.
415-432, m. 3 Taf.

Meyernorr, M. 1898. Zur Morphologie des Diphtherie-
bacillus. In: Arch. f. Hyg., v. 33, No. 1/2, pp. 1-

- 84, m. 1 Taf.

Micnazuis, L. 1902. Uber Degenerationsformen von
Pneumokokken in pleuritischen Exsudaten. In:
Berl. klin. Wochenschr., v. 39, No. 20, pp. 463-466.

Misuse, H. 1908. Beitrige zur Biologie, Morphologie
und Systematik des Tuberkelbacillus. In:
Zeitschr. f. Hyg., v. 62, No. 1, pp. 181-156.

——. 1913. Weitere Untersuchungen iiber die Bak-
teriensymbiose bei Ardisia crispa. In: Jahrb. f.
wissensch. Botanik, v. 53, No. 1, pp. 1-54, m. 2 Taf.

MieuLta, W. 1897-1900. System der Bakterien.
(1897): vir und 368 pp., m. 6 Taf. v. II (1900): x
und 1068 pp., m. 18 Taf. u. 35 Abb. im Text. Jena.

1900. Schizomycetes. In: Die natiirl. Pflan-

 

 

 

 

zenfamilien, begr. v. A. Engler u. K. Prantl. I.:

Teil, Abt. la, pp. 1-44, m. 47 Fig. Leipzig.

1904. Allgemeine Morphologie, Entwicklungs-
geschichte, Anatomie und Systematik der Schizo-
myceten. In: Lafar’s Handb. d. Techn. Mykolo-
gie, v. 1, pp. 29-149. Jena.

Muer, W. D. 1882. Der Einfluss der Microorgan-
ismen auf die Caries der menschlichen Zihne. In:
Arch. f. exper. Pathol. u. Pharmacie, v. 16, No.

3/4, pp. 291-304, m. 1 Taf.

1883. Uber einen Zahn-Spaltpilz, Leptothrix
gigantea. In: Ber.d. Deutsch. Bot. Ges., v. 1,
pp. 221-226, m. 1 Taf.

1889. Die Mikroorganismen der Mundhéhle. xx
und 305 pp., m.1Taf.u.112Abb.im Text. Leipzig.

Mincutn, E.A. 1915. Remarks on the nature and signifi-
cance of the so-called “infective granules” of pro-

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

1400 pp., m. 134 Textabb..

v. 1.

 

(Vou. XVI.

tozoa. In: Annal. Inst. Pasteur, v. 29, No. 11, pp.
537-544,

Miquet, P., et CamBirer, R. 1902. Traité de bactério-
logie pure et appliquée. xv et 1059 pp., av. 224
figs. dans le texte. Paris.

Mofiter, A. 1899. Ein neuer siure- und alkoholfester
Bacillus aus der Tuberkelbacillengruppe, etc. In:
Centralbl. f. Bact. I. Abt. v. 25, No. 11, pp. 369-
373, m. 1 Taf.

Moétxter, H. 1885. Plasmodiophora Alni.

Deutsch. Bot. Ges., v. 3, pp. 102-105.

1890. Beitrag zur Kenntnis der Frankia sub-
tilis Brunchorst. In: Ber. d. Deutsch. Bot. Ges.,
v. 8, No. 7, pp. 215-224.

Motrscu, H. 1903. Uber das Leuchten des Fleisches,
insbesondere toter Schlachttiere. In: Bot, Zeitg.,

v. 61, I. Abt., p. 1-18.

1904, Leuchtende Pflanzen. 1x und 168 pp.,

m, 2 Taf. u. 14 Textfig. Jena.

1907. Die Purpurbakterien. va und 95 pp.,
m.4 Taf. Jena.

——. 1908. Uber Ultramikroorganismen.
Zeitg., v. 66, I. Abt., pp. 131-189,

1910. Die Eisenbakterien. vm und 83 pp., m.
3 Taf. Jena.

Mooteavxar, 8. R. 1912. On certain bodies found in
syphilitic lesions, etc. In: Brit. Med. Journ., 1912,
II, No. 2711, pp. 1655-1656.

Morcx, D. 1891. Uber die Formen der Bakteroiden bei
den einzelnen Spezies der Leguminosen. Diss.
phil. Leipzig, 44 pp., m. 5 Taf.

Moreno, M. 1901. Eine neue Art von Ascobacillus,
entdeckt im Wasser des Lozayakanals bei Madrid.
In: Centralbl. f. Bact.I. Abt.v. 30,No.3, pp.111-114.

MoreentoHater, O. 1916. Ein farbstoffbildender Bacil-
lus aus Bienenlarven. In: Centralbl. f. Bact. IT.
Abt. v.46, No. 17/21, pp. 444-450, m. 1 Taf.

Moro, E. 19002. Uber die nach Gram farbbaren Bacil-
len des Sduglingsstuhles. Vorl. Mitt. In: Wien.
klin. Wochenschr., v. 18, No. 5, pp. 114-115.

== 19006. Wher den Bacillus acidophilus n. sp.
‘In: Jahrb. f. Kinderheilk., v. 52, No. 1, pp. 38-55,
m. 1 Taf.

Morsz, M.E. 1912. A study of the diphtheria group of
organisms by the biometrical method. In: Journ.
Infect. Diseases, v. 11, No. 2, pp. 253-285.

Mucn, H. 19072. Uber die granule, nach Ziehl nicht’
farbbare Form des Tuberkulosevirus. In: Beitr. z.
Klinik d. Tuberk., v. 8, pp. 85-99.

——. 19076. Uber die nicht siurefesten Formen des
Koch’schen Tuberkelbazillus. In: Beitr. z. Kli-
nik d. Tuberk., v. 8, pp. 357-365.

——. 1908a. Die nach Ziehl nicht darstellbaren For-
men des Tuberkelbacillus. In: Berlin. klin. Wo-
chenschr., v. 45, No. 14, pp. 691-694,

19086. Granula und Splitter.

Klinik d. Tuberk., v. 11, pp. 67-71.

——. 1909. Immunitit und Immunitatsreaktion. In:
Mitteil.a.d. Hamburger Staats~Krankenanstalten,
v. 9, pp. 1-66.

Mtuzens, P. 1907. Untersuchungen tiber Spirochaeta
pallida etc. In: Centralbl. f. Bact. I. Abt. Orig.

v. 43, No. 6, pp. 586-592; No. 7, pp. 674-693, m. 2
Taf.

In: Ber. d.

 

 

 

In: Bot.

 

 

In: Beitr 2,
No. 2.] LIFE CYCLES OF THE
Mustens, P., und Hartmann, M. 1906. Uber Bacillus
fusiformis und Spirochaeta dentium. In: Zeitschr.
f. Hyg., v. 55, No. 1, pp. 81-112, m. 4.Taf.
Mittuiscntzcet, A. 1898. Ein Beitrag zur Morphologie
' und Entwicklungsgeschichte d. Bakterien, etc. In:
Arb. a. d. K. Gesundh. Amte, v. 15, No. 1, pp.
131-153, m. 1 Taf.

Miizr, J. H. H. 1898. Bakterien und Eumyceten,
oder was sind und woher stammen die Spaltpilze?
v und 48 pp., m.1 Taf. Berlin.

Mtiurer, M. 1912. Uber das Vorkommen von asporo-
genem Milzbrand unter natiirlichen Verhdltnissen.
In: Centralbl. f. Bact. I. Abt. Orig., v. 66, No. 7,
pp. 501-503.

Miter, P. Ta. 1913. Bakteriologische Untersuchun-
gen beiFlecktyphus. In: Arch. f. Hyg., v. 81,
No. 6, pp. 307-332, m. 1 Taf.

Mier, R. 1912. Bakterienmutationen. In: Zeitschr.
f. indukt. Abstammungs- u. Vererbungslehre, v. 8,
pp. 305-324.

Mitier-Tuureav, H. 1908. Bakterienblasen (Bacte-
riocysten). In: Centralbl. i. Bact. II. Abt. v.20,
No. 12/14, pp. 353-400; No. 15/17, pp. 449-468,
m. 4 Taf.

Minpen, M. 1896a.
Arch. f. Anat. u. Physiol.
pp. 22-35.

——. 1896b. Zweiter Beitrag zur Granulafrage. In:
Arch. f, Anat. u. Physiol. Physiol. Abt. 1896,
pp. 269-293.

——. 1897. Dritter Beitrag zur Granulafrage.
Arch. f. Anat. u. Physiol. Physiol. Abt. 1897,
pp. 340-369, m. 2 Taf.

——. 1899. Vierter Beitrag zur Cytoblastenfrage. In:
Centralbl. f. Bact. II. Abt. v. 5, No. 11, pp.
398-408; No. 12, pp. 447-456; No. 13, pp. 490-499,
m. 3 Taf.

——. 1907. Der Chtonoblast, die lebende biologische

“~ ‘wnd morphologische Grundlage alles sogen. Beleb-
ten und Unbelebten. vr und 167 pp., m. 9 Taf.
u. 11 Abb. im Text. Leipzig.

Murr, R., and Rrronte, J. 1903. Manual of bacteriology.
Amer. ed. by N. MacLeod Harris. xx and 565 pp.,
with 170 illus. New York and London.

Muro, T. 1904. Ein eigentimlicher Bacillus (B. heli-
xoides). In: Centralbl. f. Bact. I. Abt. Orig. v.
37, No. 3, pp. 321-325, m. 1 Taf.

Napson, G. A., und Apamovid, §.M. 1910. Uber die
Beeinflussung der Entwicklung des B. mycoides,
etc. In: Bull. Jardin Imp. Bot. St. Pétersbourg., v.
10, pp. 154-165. Abstr. In: Centralbl. f. Bact.
II. Abt. v.31, 1912, No. 5/10, pp. 287-288.

Nieeu, C. von. 1877. Die niederen Pilze in ihren
Beziehungen zu den Infectionskrankheiten und
der Gesundheitspflege. xxxm und 285 pp., m. 6
Abb. im Text. Mitnchen.

——. 1882. Untersuchungen  tiber siiadtane Pilze.
285 pp. Miénchen und Leipzig.

Naxanisat, K. 19002. Vorliufige Mitteilung tiber eine
neue Farbemethode, etc. In: Miinchn. med. Wo-
chenschr., v.:47, No. 6, pp. 187-188. :

19006. Beitrage zur Kenntnis der Leukozyten

und Bakteriensporen. In: Miimchn. med. Wo-

chenschr., v. 47, No. 20, pp. 680-683.

In:
1896,

Ein Beitrag zur Granulafrage.
Physiol. Abt.

 

 

In: |

 

235

NaxanisH, K. 1900c. Bacillus variabilis lymphae vac-
cinalis, etc. In: Centralbl. f. Bact. I. Abt. v. 27,
No. 18/19, pp. 641-661, m. 2 Taf.

1900d. Nachtrag zu meiner Arbeit “Bacillus
variabilis lymphae,”’ etc. In: Centralbl. f. Bact.

I. Abt. v. 28, No. 10/11, pp. 304-305.

1901. Uber den Bau der Bakterien. In: Cen-
tralbl. f. Bact. I. Abt., v. 30, No. 3, pp. 97-110;
No. 4, pp. 145-158; No. 5, pp. 193-201; No. 6, pp.
225-232, m. 5 Taf.

Nakayama, H. 1906. Impfversuche mit Actinomyces
asteroides Eppinger an Meerschweinchen. In: Arch.
f. Hyg., v. 58, No. 3, pp. 207-312, m. 4 Taf.

Neetsen, F. 1880. Untersuchungen iiber Bacterien.
X. Studien iiber die blaue Milch. In: Cohn’s Beitr.
z. Biol. d. Pflanzen, v. 3, Heit 2, pp. 187-247, m. 1
Taf.

Neer, A. 1906. Uber Filtration des Vaccinevirus. In:
Zeitschr. f. Hyg., v. 54, No. 3, pp. 327-346.

1909. Uber die Morphologie und den Entwick-
lungszyklus des Parasiten der Tollwut. In: Zeit-
schr. f. Hyg., v. 63, No. 3, pp. 421-448, m. 3 Taf.

Neer, E. E. A. M. ve. 1916. Untersuchungen zur

Kenntnis der Corynebakterien, etc. In: Folia mi-

crobiologica, v. 4, No. 2, pp. 119-188, m. 8 Taf.

und Mizremet, C. W. G. 1913. Zur Aetiologie
des malignen Granuloms. In: Centralbl. f. Bact.
I. Abt. Orig. v. 68, No. 3/4, pp. 292-309, m. 4 Taf.

News, E. 1904. Botanische Beschreibung einiger spo-
renbildender Bakterien. In: Centralbl.f. Bact. IT.
Abt. v.12, No. 1/3, pp. 1-32; No. 6/8, pp. 161-176;
No. 11/16, pp. 337-352; No. 19/21, pp. 539-554, m.
3 Taf.

Neisser, A. 1881. Weitere Beitrige zur Aetiologie der
Lepra. In: Virchow’s Arch. f. pathol. Anat., v. 84,
No. 3, p. 514-542, m. 1 Taf.

1888. Versuche iiber Sporenbildung bei Xerose-
bacillen, Streptokokken und Choleraspirillen. In:
Zeitschr. f.. Hyg., v. 4, pp. 165-196.

Neisser, M. 1906. Ein Fall von Mutation nach de
Vries bei Bakterien. In: Centralbl. f. Bact. I.
Abt. Ref., v. 36, Beiheft, p. 98.

Nencst, M., Srzzer, N., et WiuNIKEWITcH, W. 1898.
Recherches sur la peste bovine.. In: Arch. sci. biol.
St. Pétersbourg, v. 6, pp. 374-396, av. 3 pls.

Nevgrce, H. 1902, Uber Strahlenpilze (Actinomy-
ceten). Zweite Folge. Diss. nat. phil. Strassburg.
72 pp., m. 1 Taf.

Neumann, R. 1897. Studien tiber die Variabilitat der
Farbstoffbildung bei Microc. pyogenes a aureus
(Staphylococcus pyogenes aureus) und einigen an-
dern Spaltpilzen. In: Arch. f. Hyg., v. 30, No. 1,
pp. 1-31, m. 1 Taf.

Nicotiz. 1913. L’autolyse. (Etude de biologie générale.)
In: Annal. Inst. Pasteur, v. 27, No. 2, pp. 97-117.

Niessen, M. von. 1898. Ein neuer Beitrag zur Syphi-
lisaetiologie. In: Centralbl. f. Bact. I. Abt. v.
23, No. 2, pp. 49-59; No. 3/4, pp. 108-117; No. 5/6,
pp. 194-205; No. 7, pp. 258-266, m. 2 Taf.

——. 1908. Der Syphilisbacillus. Seine Geschichte,
Literatur, Kultur und spezifische Pathogenitat fiir
Tiere und Menschen. 84 pp., m. 37 Taf. Leipzig.

Nrevwenauis,.A. W. 1911. Method to cultivate micro-
organisms from one cell. In: Proceed. Sect. Sci. K.
Akad. Amsterdam, v. 13, pp. (566)-(576), with 2 pla.

BACTERIA—LOHNIS.

 

 

 

 
236

Nocarp, E. 1888. Note sur la maladie des beeufs de la
Guadeloupe connue sous le nom de farcin. In:
Annal. Inst. Pasteur, v. 2, No. 6, pp. 293-802, av.
2 pls. -

1889. Sur la tuberculose zoogléique. In: Compt.

rend. Soc. Biol., v. 41, No. 34, pp. 608-610.

et Lectaincne, E. 1903-1905. Les maladies mi-

crobiennes des animaux. 37° éd. v. I: 1 et 668

pp. v. II: 645 pp. :

et Roux, E. 1898. Le microbe de la péripneu-
monie. In: Annal. Inst. Pasteur, v. 12, No. 4, pp.
240-262.

Nocut, B., und Meyer, M. 1907. Trypanosomen als
Krankheitserreger. In: Kolle-Wassermann’s Handb.
d. pathog. Mikroorg. I. Erg. Bd. pp. 1-74, m.3
Taf.

Noeucat, H., 1908. On the inhibitory influence of eosin
upon sporulation. In: Journ. Exp. Med., v. 10,
No. 1, pp. 30-35.

——. 1910. Pleomorphism and pleobiosis of Bac. bifi-
duscommunis. In: Journ. Exp. Med., v.12, No. 2,
pp. 182-195, with 4 pls.

1911. A method for the pure cultivation of
pathogenic Treponema pallidum (Spirochaeta pal-
lida). In: Journ. Exp. Med., v. 14, No. 2, pp. 99-
108, with 5 pls.

—. 1912a.
(Treponema microdentium and macrodentium). In:
Jour. Exp. Med., v. 15, No. 1, pp. 81-89, with 5 pls.

——. 1912b.. Treponema mucosum (new species)
grown in pure culture. In: Journ. Exp. Med., v.

16, No. 2, pp. 194-198, with 1 pl.

1912c. The pure cultivation of Spirochaeta
Duttoni, Sp. Kochi, Sp. Obermeieri and Sp. Novyi.

 

 

 

 

 

In: Journ. Exp. Med., v. 16, No. 2, pp. 199-210, |

with 2 pls.
1912d. Pure cultivation of Spirochaeta phage-
denis (new species), etc. . In: Journ. Exp. Med., v.

16, No. 3, pp. 261-268, with 2 pls.
—-. 1912e.

 

In: Journ. Exp. Med., v. 16, No. 5, pp. 620-628,

with 1 pl.

—. 1918. Contribution to the cultivation of the

parasite of rabies. In: Journ. Exp. Med., v. 18,
No. 3, pp. 314-316, with 1 pl.

 

vation of so-called trachoma bodies. In: Journ.

Exp. Med., v. 18, No. 5, pp. 584-587, with 1 pl.

1915. Bacteriological and clinical studies
of an epidemic of Koch-Weeks bacillus conjunc-
tivitis associated with cell inclusion conjuncti-
vitis. In: Jour. Exp. Med., v. 22, No. 3, pp. 304-
318, with 3 pls.

Novy, F.G. 1894. Ein neuer anaérober Bacillus des
malignen Oedems. In: Zeitschr. f. Hyg., v.17, pp.
209-233, m. 2 Taf.

and Knapp, R. E. 1906. Studies on Spirillum

Obermeieri and related organisms. In: Journ.

Infect. Diseases, v. 3, No. 3, pp. 291-293, with

7 pls.

Nowax, J. 1908. Le bacille de Bang et sa biologie. In
Annal. Inst. Pasteur, v. 22, No. 6, pp. 541-556, av.
8 pls.

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

Cultural studies on mouth spirochaetae | _

Cultivation of Spirochaeta gallinarum. |.

and Conen, M. 1913. Experiments on the culti-

[Vou. XVI,

Nuzum, J. W., and Herzoc, M. 1916. Experimental
studies in the etiology of acute epidemic poliomye-
litis. In: Journ. Amer. Med. Assoc., v. 67, No. 17,
pp. 1205-1208, with 4 figs.

Nvyzere, 0. 1912. Uber die Kolonien der lophotrichen
Stébchenbakterien. Akad. Abhandlung. Helsing-
fors. 86 pp., m. 4 Taf.

OxrrTet. 1871. _Experimentelle Untersuchungen iiber
Diphtherie. In: Deutsches Arch. f. klin. Med., v.
8, No. 3/4, pp. 242-354, m. 2 Taf.

Onmmacuer, A. P. 1902. Observations upon the mor-
phologic variations of certain pathogenic bacteria.
In: Journ. Med. Res., v. 7, No. 2, pp. 128-137,
with 2 pls.

Ourrzxy, P. K. 1917. The non-filterability of typhus
fever virus. In: Journ. Infect. Diseases, v. 20,
No. 4, pp. 349-356. .

Otsen, O. J. 1897. Zur Pleomorphismusfrage. In:
Centralbl.f. Bact. II. Abt. v. 3, No. 11/12, pp.
273-284, m. 2 Taf. :

Otsson, P. G. 1915. Zur Variation des Choleravirus.
In: Centralbl. f. Bact. I. Abt. Orig. v. 76,
No. 1, pp. 23-37.

Omztiansx1, V. 1899. Uber die Isolierung der Nitri-

fikationsmikroben aus dem Erdboden. In: Cen-

tralbl. f. Bact. II. Abt. v.5, No. 15, pp. 587-549,

m. 1 Taf.

——. 1902. Uber die Garung der Cellulose. In:
Centralbl.f. Bact. II. Abt. v. 8, No. 7, pp. 193-
201; No. 8, pp. 225-231; No. 9, pp. 257-263; No.
10, pp. 289-294; No. 11, pp. 321-326; No. 12, pp.
353-361; No. 13, pp. 385-391, m. 1 Taf.

‘Onstow, H. 1917. The nature of growths in colloidal

silica solutions. In: Nature, v. 98, No. 2469, pp.

489-490.

| Oster, W., and Scuirer, E. A. 1873. Uber einige im

Blute vorhandene Bakterien bildende Massen.

Vorl. Mitt. In: Centralbl. f£. d. med. Wissensch.,

v. 11, No. 37, pp. 577-578.

‘Oven, E.von. 1906. Eine neue Bakterienerkrankung

der Leguminosenfriichte. In: Centralbl. f. Bact.

II. Abt. v.16, No. 1/3, pp. 67-74, m. 1 Taf.

}Ozax1, Y. 1915. Zur Kenntnis der anaéroben Bakterien

: der Mundhdhle. III. Mitt. In: Centralbl.f. Bact.
I. Abt. Orig. v. 76, No. 7, pp. 469-476.

‘Pansini, S. 1890. Bakteriologische Studien tiber den
Auswurf. In: Virchow’s Arch. f. pathol. Anat.,
v. 122, No. 3, pp. 424-469, m. 2 Taf.

Par, W. H., and Wmitams, A. W. 1914. Pathogenic
microorganisms. 5th ed. vm and 684 pp., with
9 pls. and 210 figs. in the text. Philadelphia and
New York.

Passini, Fr. 1905. Studien tiber faulniserregende an-
aérobe Bakterien des normalen menschlichen
Darmes und ihre Bedeutung. In: Zeitschr. f.

| Hyg., v. 49, No. 1, pp. 185-160.

Payne, S. G. 1916. On the supposed origin of life in
solution of colloidal silica, In: Annals of Botany,
v. 30, No. 99, pp. 383-388, with 1 pl.

‘PzaRMAIN, T. H., and Moor, C. G. 1898. Applied

 

bacteriology. 2nd ed. xv and 464 pp. with 9
pls. and 38 figs. in the text. London.
No. 2.] LIFE CYCLES OF THE

Pésu, G., et Rasat, H. 1906a. Polymorphisme ex-
périmental du bacille d’Eberth. In: Compt. rend.

 

 

Soc. Biol. v. 60, pp. 336-338.

19066. Le coli-bacille dans les milieux
salins. In: Compt. rend. Soc. Biol. v. 60, pp.
628-630.

———

 

1906c. Quelques cas de polymorphisme

de bacilles par l’iodure de potassium, etc. In:

Compt. rend. Soc. Biol., v. 60, pp. 816-817.

1906d. Quelques nouveaux cas de poly-

morphisme de bacilles, etc. In: Compt. rend. Soc.

Biol., v. 60, pp. 1013-1014.

1906e. Note sur le polymorphisme des

bacilles dans l’urée. In: Compt. rend. Soc. Biol.,

v. 61, pp. 477-479.

. 1906f. Variations morphologiques et

biologiques des bactéries, etc. In: Journ. de

physiol. et de pathol. génér., v. 8, pp. 868-876.

1907a. Morphologie du bacille du la

tuberculose humaine dans les milieux salins. In:

Compt. rend. Soc. Biol., v. 63, pp. 427-429.

19076. Cytologie du bacille de la tuber-
culose humaine dans les milieux salins, In:
Compt. rend. Soc. Biol., v.63, pp. 681-682.

Pexto, J. 1910. Die pflanzlichen Aktinomykosen. In:
Centralbl. f. Bact. II. Abt. v.27, No. 17/21, pp.
451-579.

Pénav, H. 1911. Cytologie de Bacillus anthracis. In:
Compt. rend. Acad. Paris, v. 152, No. 10, pp. 617-
619.

Penrotp, W. J. 1911. Studies in bacterial variation.
In: Journ. of Hyg., v.11, No. 1, pp. 30-67, with 1 pl.

——. 1912. On the specifity of bacterial mutation.
In: Journ. of Hyg., v. 12, No. 2, pp. 195-217, with
4 pls.

aan 1902. The extra-or intra-cellular location
of Hansen’s bacillus. In: Lepra, v. 2, No. 4, pp.
203-205.

Perrin, W. S. 1906. Researches upon the life-history
of Trypanosoma balbianii (Certes). In: Arch. f.
Protistenk., v. 7, No. 1, pp. 131-156, with 2 pls.

Perry, M. 1852. Zur Kenntnis kleinster Lebensformen.
vut und 228 pp., m. 17 Taf. Bern.

Peters; W. H. 1911. “Hand infection apparently due

to Bacillus fusiformis. In: Journ. Infect. Diseases,
v. 8, No. 4, pp. 455-462.

Perri, L. 1907. Untersuchungen tiber die Identitat
des Rotzbacillus des Olbaumes. In: Centralbl. f.
Bact. II. Abt. v.19, No. 16/18, pp. 531-538.

PerscHenko, B. pe. 1918. Sur le cycle évolutif de
Chlamydothrix ochracea (Ktitz.) Mig., etc. In:
Arch. f. Protistenk., v. 28, No. 2, pp. 239-312, av.
3 pls.

Prerrer, W. 1886. Uber Aufnahme von Anilinfarben
in lebende Zellen. In: Unters. a. d. Bot. Inst.
Tiibingen, v. 2, No. 2, pp. 179-331, m. 1 Taf.

——. 1888. Uber chemotaktische Bewegungen von
Bakterien, etc. In: Unters. a.d. Bot. Inst, Tibin-
gen, v. 2, No. 3, pp. 582-661.

Premrer, R. ‘1904. Weitere Untersuchungen tiber das
Wesen der Choleraimmunitat, etc. In: Zeitschr.
f. Hyg., v. 18, No. 1, pp. 1-16.

 

 

 

 

 

 

 

237

Premrer, R. 1896. Die Spirillen. In: Fligge’ a‘ ‘Mi-
kroorganismen. ” 3, Aufl. v. 2, pp. 527-599.

und Kore, W. 1896. Uber die specifische Im-
munititsreaction der Typhusbacillen. In: Zeit-
schr. f. Hyg., v. 21, pp. 203-246.

PFEILER, W., und Lentz, W. 1913. Die Zichtung des
Virus der Schweinepest. In: Berliner tierarztl.
Wochenschr., v. 29, No. 39, pp. 689-692.

Prusy. 1888. Zur Sporenbildung der Typhusbacillen.
In: Centralbl. f. Bact., v. 4, No. 25, pp. 769-776.

Puisauix, C. 1892. Regénération expérimentale de la
propriété sporogéne chez le Bacillus anthracis,
etc. In: Compt. rend. Acad. Paris, v. 115, No. 4,
pp. 253-255.

Pursatix,G. 1902. Polymorphisme des Pasteurella. In:
Compt. rend. Soc. Biol., v. 54, pp. 645-648.
Prant, G. P., et Gatti-Vanerio, B. 1895. Sur une va-
riété du Bactérium Chauvoei. In: Annal. Inst.

Pasteur, v. 9, No. 4, pp. 258-264.

Piccour, E. 1896. Sulla sporulazione del Bacterium coli
commune. In: Centralbl.f. Bact. I. Abt. v.19,
No. 9/10, pp. 307-3138.

PreryetManpout. 1904. Polymorphisme du bacille de
Koch, ete. In: Compt. rend. Soc. Biol., v. 57, pp.
586-588.

Pivzav, J. 1845a. Recherches sur le développement
des animalcules infusoires et des moisissures. In:
Annal. sci. nat. Zoologie. 37° sér. v. 3, pp.
182-189, av. 1 pl.

1845. Supplément aux recherches sur le dével-
oppement, etc. In: Annal.sci. nat. Zoologie. 3™°
sér. v. 4, pp. 103-104.

Prnoy, E. 1913. Sur la necessité d’une association bac-
térienne pour le développement d’une Myxobac-
térie, Chon dromyces crocatus. In: Compt. rend.
Acad. Paris, v. 157, No. 1, pp. 77-78.

Puavr, H. C. 1907. Uber die Geisseln bei fusiformen
Bacillen. In: Centralbl. f. Bact. I. Abt. Orig.
v. 44, No. 4, pp. 310-316, m. 1 Taf.

Protz, H., Ourrzxy, P. K., and Banur, G. 1915. The
etiology of typhus exanthematicus. In: Journ. In-
fect. Diseases, v. 17, No. 1, pp. 1-68, with 1 pl.

PopwyssoTzky, W., et TARANOUKHINE, B. 1898. Con-
tribution 4 l'étude de la plasmolyse chez les bac-
téries. In: Annal. Inst. Pasteur, v. 12, No. 8, pp.
501-509, av. 1 pl.

PottaK, R. 1913. Uber Formenwechesel bei dem Bacil-
lus faecalis alcaligenes. In: Centralbl.f. Bact. I.
Abt. Orig. v. 68, No. 3/4, pp. 288-291.

Pororr-TscHERKASKY, D. 1914. Quelques observations
sur la morphologie et la biologie du V. cholerae
(Koch) Buchner, etc. In: Centralbl. f. Bact. I.
Abt. Orig. v. 74, No. 5/6, pp. 382-392.

Portier, P. 1917. Recherches sur les microorganismes
symbiotiques dans la série animale. In: Compt.
rend. Acad. Paris, v. 165, no. 5, pp. 196-199.

Poucuet, F. A., Jory, N., et Musser, Cu. 1863. Ex-
périences sur l’hétérogenie. In: Moniteur scien-
tifique, v. 5, pp. 760-763.

Prazmowse«I, A. 1880. Untersuchungen tiber die Ent-
wicklungsgeschichte und Fermentwirkung einiger
Bakterien-Arten. 59 pp., m. 2 Taf. Leipzig.

BACTERIA—LOHNIS.

 

 
238

Prazmowsk!, A. 1888a. Uber Sporenbildung bei Bak-
terien. In: Biol. Centralbl., v.8, No. 10, pp. 301-307.
1888). Uber die Wurzelknéllchen der Legumi-
nosen. In: Botan. Centralbl., v. 36, no. 7, pp. 215-
219; No. 8, pp. 248-255; No. 9, pp. 280-285.
——. 1890. Die Wurzelknéllchen der Erbse. In:
Landw. Versuchsstat., v. 37, pp. 161-238, m. 2 Taf.

1912. Azotobacter Studien. I. Morphologie
und Cytologie. In: Bull. intern. Acad. Cracovie.
Cl. sci. math. et nat. Sér. B, 1912, No. 3, pp. 87-
174, av. 3 pls.

1913. Die Zellkerne der Bakterien.
intern. Acad. Cracovie. Cl. sci. math. et nat.
B, 1918, No. 4, pp. 105-151, av. 1 pl.
Preisz, H. 1894. Recherches comparatives sur les

pseudotuberculoses bacillaires, etc. In: Annal.
Inst. Pasteur, v. 8, No. 4, pp. 231-255, av. 2 pls.
——. 1903. Der Bacillus des seuchenhaften Verwer-
fens. In: Centralbl.f. Bact. I. Abt. Orig. v. 33,
No. 3, pp. 190-196.
——. 1904. Studien iiber Morphologie und Biologie
des Milzbrandbacillus, etc. In: Centralbl. f. Bact.
I, Abt. Orig. v. 35, No. 3, pp. 280-293; No. 4, pp.
416-434: No. 5, pp. 537-545; No. 6, pp. 657-665,
m. 2 Taf.
Preitt, H. 1917. Die Vielgestaltigkeit des Bacterium
coli. In: Centralbl. f. Bact. I. Abt. Orig. v. 79,
No. 6, pp. 324-339,
Prituieux. 1879. Sur la nature et sur la cause de la for-
-mation des tubercules qui naissent sur les racines
deslégumineuses. In: Bull. Soc. Botan. de France,
v. 26, pp. 98-106. =
PrinesHem, H. 1910. Die Variabilitat niederer Orga-
nismen. vu und 216 pp. Berlin.
Proca, G. 1908. Sur quelques particularités du bacille
fusiforme (Vincent), etc. In: Compt. rend. Soc.‘
Biol., v. 64, pp. 771-772.
et Dantua, P. 1910a. Sur la présence dans les
produits syphilitiques d’une thrichobactérie patho-
géne (Cladothrix stereotropa n. sp.). In: Compt.
rend. Soc. Biol., v. 68, pp. 79-81.
19106. Sur le polymorphisme de la
thrichobactérie des produits syphilitiques. In:
-Compt. rend. Soc. Biol., v. 68, pp. 190-192.
ProzscHer, F., and Sem, H. A. 1917. The etiology of
hog cholera. (Preliminary report.) In: Journ.
_ Amer. Veterin. Med. Assoc., v.51, no. 1, pp. 64-69,
with 2 pls.
Prove, O. 1887. Micrococcus ochroleucus, eine neue
chromogene Spaltpilzform. In: Cohn’s Beitr. z.
Biol. d. Pflanzen, v. 4, Heft 3, pp. 409-440, m. 1 Taf.
Prowazex, 8. von. 1905. Untersuchungen iiber die
Vaccine. I. In: Arb. a. d. K. Gesundh. Amt., v.
22, No. 3, pp. 535-556, m. 1 Taf.
——. 1906a. Untersuchungen iiber den Erreger der
Vaccine. II. In: Arb. a. d. K. Gesundh. Amt,
v. 23, No. 2, pp. 525-541.
19066. Morphologische und _ entwicklungsge-
schichtliche Untersuchungen tiber Hithnerspiro-
chaeten. In: Arb. a. d. K. Gesundh. Amt, v. 23,
No. 2, pp. 554-569, m. 2 Taf.
—. 1907a. Vergleichende Spirochaeten-Untersu-
chungen. In: Arb. a. d. K. Gesundh. Amt, v. 26,
No. 1, pp. 23-31, m. 1 Taf.

 

 

In: Bull.
Sér.

 

 

 

 

 

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

[Vou. XVI.

Prowazexk,S.von. 19076. Die Sexualitét bei den Pro-
tisten. In; Arch. f. Protistenk., v.9, No.1, pp. 22-32.

 

. 1907c. Chlamydozoa. I. Zusammenfassende
Ubersicht. In: Arch. f. Protistenk., v. 10, No. 2/3,
pp. 336-358.

Quen, A. 1906. Untersuchungen iiber die Myxobac-
terien. In: Centralbl. f. Bact. II. Abt. v. 16,
No, 1/3, pp. 9-84, m. 1 Taf.

Razz, C. 1886. Uber mykotische Bindegewebswuche-

, rungen bei Pferden. In: Deutsche Zeitschr. f.
Thiermed., v. 12, No. 3, pp. 137-162, m, 1 Taf.

Raginowirscu, M. 1909. Zur Aetiologie des Fleckty-
phus. In: Arch. f. Hyg.,.v. 71, No. 4, pp. 331-379.

Raszat, H., et Ptsu,G. 1906. Variations morphologiques
du B. d’Eberth, etc. In: Compt. rend. Soc. Biol.,
v. 60, pp. 468-469.

Ravenet, M. P. 1896. Notes on the bacteriological ex-
amination of the soil of Philadelphia. In: Mem.
Nat. Acad. Science, v. 8, First Memoir, p. 1-41.

Rayman, B., et Kruis, K. 1904. Des noyaux des bac-
téries. In: Bull. intern. Acad. de Bohéme. Roc-
nik XII, Trida 2, Cirlo 31. Abstr. in: Centralbl. f.
Bact. II. Abt. v.13, p. 645.

ReenstyeRNa, J. 1912. Uber die Kultivierbarkeit des
Lepraerregers und die Ubertragung der Lepra auf
Affen. In: Deutsche med. Wochenschr., v. 38,
No. 38, pp. 1784-1785.

Retcnenpacu, H. 1901. Uber Verzweigung bei Spiril-
len. In: Centralbl. f. Bact. I. Abt. v. 29, No.
13, pp. 553-557, m. 1 Taf.

Reineit, J. 1905. Beitrag zur Kenntnis einiger Leucht-
bakterien. In: Centralbl. f. Bact. II. Abt. v.15,
No. 10/11, pp. 289-300.

Retreer, L. F., Berman, N., and Sturces, W.S. 1916.
Further studies on bacterial nutrition, ete. In:
Journ. Bact., v. 1, No. 1, pp. 15-33.

and SHerrick, J. L. 1911. Studies on bacterial
variation. In: Journ. Med. Res., v. 24, No. 2, pp.

265-284.

Revis, C. 1912a. The production of variation in the
physiological activity of B. coli, ete. In: Proceed.
Roy. Soc. (London), Ser. B, v. 85, No. 578, pp.
192-194. ‘

19126. Coccoid forms of B. coli, etc. In: Cen-
tralbl. f. Bact. II. Abt. v.33, No. 17/19, pp. 424-
428.

Riemer. 1909. Beitrag zur Kenntniss des Stoffwechsels
des Micrococcus pyogenes aureus. In: Arch. f.
Hyg., v. 71, No. 2, pp. 131-222.

RINDFLEISCH. 1872. Untersuchungen tiber niedere Or-
ganismen. In: Virchow’s Arch. f. pathol. Anat.,
att No. 1/2, pp. 108-120: No. 3, pp. 396-407, m.

af.

 

 

Rosin, Cu. 1853. Histoire naturelle des végétaux para-
sites. xvi et 702 pp.,av.un atlas del5pl. Paris.

1871. Traité du microscope. xx et 1028 pp.,
av. 3 pl. et 317 figs. dans le texte. Paris.

Ropuia, A. 1901. Uber die sogenannten siurelieben-
den Bacillen im Sduglingsstuhle. In: Centralbl.
f. Bact. I. Abt. v. 29, No. 18, pp. 717-724, m.1

Tat.

1908. Magencarcinom und Milchsdurebacillen,
etc. In: Centralbl. f. Bact. I. Abt. Orig. v. 47,
No. 4, pp. 445-466.

 

 
No. 2.1 LIFE CYCLES OF THE

Roper, A. 1894. La variabilité dans les microbes. 224
pp. Paris.

Roemer, P. H. 1914. Bakteriologische Diphtherie-
studien. In: Berl. klin. Wochenschr., v. 51, No.
11, pp. 503-506.

Rogers, L. A., Cuarx, W. M., and Davis, B. J. 1914.
The Colon-group of bacteria. In: Journ. Infect.
Diseases, v. 14, No. 3, pp. 411-475.

and Evans, A.C. 1914. The characteris-
tics of bacteria of the Colon type found in bovine
feces. In: Journ. Infect. Diseases, v. 15, No. 1,
pp. 99-123.

Rotorr, F. 1883. Uber die Milzbrand-Impfung und die
Entwicklung der Milzbrand-Bakterien. In: Arch.
{. wissensch. u. prakt. Tierheilk., v. 9, No. 6, pp.
459-470.

Rosensace, F. J. 1884. Mikro-Organismen bei den
Wundinfections-Krankheiten des Menschen. x,
und 122 pp., m. 5 Taf. Wiesbaden.

1887. Uber das Erysipeloid.. In: Arch. f. klin.

Chirurgie, v. 36, pp. 346-350.

——. 1909. Experimentelle, morphologische und
klinische Studie tiber die krankheitserregenden
Mikroorganismen des Schweinerotlaufs, etc. In:
Zeitschr. f. Hyg., v. 63, No. 2, pp. 343-371, m. 7
Taf.

RosensraT, St. 1905. Zur Kenntnis der zur Gruppe der
Tuberkelbazillen gehérenden saurefesten Mikro-
organismen. In: Flora, v. 95, No. 2, pp. 412-467.

1911. Vergleichende Untersuchungen iiber
neuere Farbungsmethoden der Tuberkelbacillen,
etc. In: Centralbl. f. Bact. I. Abt. Orig. v.58,
No. 2, pp. 173-192.

Rosenrerp, A. 1901a. Uber die Involutionsformen der
Pestbacillen und einiger pestihnlicher Bakterien,
etc. Diss. med. Kénigsberg i. Pr., 25 pp.

. 19015. Uber die Involutionsformen einiger pest-
abnlicher Bakterien, etc. In: Centralbl. f. Bact.
I. Abt. 30, No. 17, pp. 641-653.

Rosenow, E. C. 1914. Wechselseitige Mutation von
Pneumokokken und Streptokokken. In: Centralbl.
f. Bact. I. Abt. Orig. v.73, No. 4/5, pp. 284-287,
m. 2 Taf.

. 19146. Transmutations within the Streptococ-

cus-Pneumococcus group. In: Journ. Infect. Dis-

eases, V. 14, No. 1, pp. 1-32.

and Towne, E. B. 1917. Bacteriological obser-

vations in experimental poliomyelitis of monkeys.

In: Journ. Med. Res., v. 36, No. 1, pp. 175-186,

with 5 pls.

and Hess, C. L. von. 1918. The elective

localization of streptococci from epidemic polio-

myelitis. In: Journ. Infect. Diseases, v. 22, No. 4,

pp. 313-344, 5 pls. ;

and WHEELER, G. W. 1916. The etiology
of epidemic poliomyelitis. In: Journ. Amer. Med.
Assoc., v.67, No. 17, pp. 1202-1205, with 6 figs.

—— and Tunnicurrr, R. 1912. Pyemia due to an
anaérobic polymorphic bacillus, probably Bacillus
fusiformis. In: Journ. Infect. Diseases, v. 10, No.
1, pp. 1-6, with 1 pl.

and WHEELER, G. W. 1918. The etiology of epi-

demic poliomyelitis. In: Journ. Infect. Diseases,

v. 22, No. 4, pp. 281-312, 3 pls.

160266°—21——16

 

 

 

 

 

 

 

 

 

 

 

239

Ross, E. H. 1912. An intracellular parasite developing
into spirochaetes. In: Brit. Med. Journ., 1912, II,
No. 2711, pp. 1651-1654.

Rossr, G. pz. 1907. Uber die Mikroorganismen, welche
die Wurzelknéllchen der Leguminosen erzeugen.
In: Centralbl. f. Bact. II. Abt. v.18, No. 10/12,
pp. 289-314; No. 16/18, pp. 481-489, m. 2 Taf.

Rotuert, K. A. 1902. Degeneration and regeneration
of bacteria (Russian). Diss. med. St.-Péters-
bourg. 79 pp., with 1 pl.

Rovx, E. 1890. Bactéridie charbonneuse asporogéne.
In: Annal. Inst. Pasteur, v. 4, No. 1, pp. 25-34.

Row1anpd, 8S. 1899. Observations upon the structure of
bacteria. In: Transact. Jenner Inst. Prevent.
Med., v.-2, pp. 143-161, with 1 pl.

1912. The relation of pseudotubercle to plague,

etc. In: Journ. of Hyg., v. 12, Plague Suppl. IT,

pp. 350-357.

. 1914. The morphology of the plague bacillus.

In: Journ. of Hyg., v. 18, Plague Suppl. III, pp.

418-422, with 7 pls. ;

Ruata, G. 1905. Les formations des granulations dans
les cultures des vibrions. In: Annal. Inst. Pas-
teur, v. 19, No. 10, pp. 661-672.

Rusinsky, B. 1910. Studien iiber den Kumiss. In:
Centralbl. f. Bact. II. Abt. v. 28, no. 6/8, pp.
161-219, m. 1 Taf.

Riruer, R. 1910. Zur Sichtbarkeit des Schweine-
pesterregers. 63pp. Hannover. —

Rutimann, W. 1896. Weitere Mitteilungen iiber Clado-
thrix dichotoma und odorifera. In: Centralbl. f.
Bact. II. Abt. v. 2, No. 22, pp. 701-706.

1897. Uber ein Nitrosobacterium mit neuen

Wuchsformen. In: Centralbl. f. Bact. II. Abt.

v. 3, No. 9/10, pp. 228-231. =

1898a. Uber eine aus Sputum isolierte patho-

gene Streptothrix. In:Mimchn. med.Wochenschr.,

v. 45, No. 29, pp. 919-922. ;

1898b. Uber einen neuen chromogenen Bacillus

aus stidtischem Kanalwasser. In: Centralbl. f.

Bact. I. Abt. v. 24, No. 13, pp. 465-467.

1900. Uber einen neuen chromogenen Bacillus

aus stidtischem Kanalwasser. In: Centralbl. f.

Bact. II. Abt. v. 6, No. 5, pp. 129-131.

1907. Photogramme von Crenothrix polyspora
Cohn. In: Centralbl.f. Bact. II.Abt. v.20, No.

_ 4/5, pp. 97-99, m. 1 Taf.

Russeit, H. L. 1892. Untersuchungen tiber im Golf
von Neapel lebende Bakterien. In: Zeitschr. f.
Hyg., v. 11, pp. 165-206, m. 2 Taf.

Robzitxa, V. 1908. Uber die biologische Bedeutung
der farbbaren Kérnchen des Bakterieninhaltes. In:
Arch, f. Hyg., v. 46, pp. 337-389, m. 2 Taf.

—. 1908. Sporenbildung und andere biologische
Vorgiinge bei dem Bact. anthracis. In: Arch. f.
Hyg., v. 64, No. 3, pp. 219-294, m. 3 Taf.

——. 1909. Die Cytologie der sporenbildenden Bak-
terien, etc. In: Centralbl. f. Bact. II. Abt. v.
23, No. 10/13, pp. 289-300. ,

——. 1910. Uber die experimentelle Autogamie der
Bakterien. In: Arch. f. Entwickl. Mechanik d.
Organismen, v. 30, I. Teil, pp. 443-460.

BACTERIA—LOHNIS.

 

 

 

 

 

 
240

Rtzéxa, V. 1913. Eine Methode zur Darstellung der
Struktur fertiger Bakteriensporen, etc. In: Cen-
tralbl. f. Bact. II. Abt. v. 36, No. 19/25, pp. 577-
587, m. 1 Taf.

Sacus, J. 1875. Textbook of botany. Translat. by
A. W. Bennett and W. T. Dyer. xn and 858 pp.,
with 461 figs. Oxford.

Sacus, M. 1903. Ein Beitrag zur Kenntnis der Kapsel-
bacillen. In: Centralbl. f. Bact. I. Abt. Orig.
v. 33, No. 9, pp. 657-678.

Saisawa, K. 1911. Uber den Erreger und die Diag-
nose des Maltafiebers. In: Zeitschr. f. Hyg., v. 70,
No. 2, pp. 177-203, m. 1 Taf.

1913. Uber den modifizierenden Einfluss von
kohlehydrathaltigen Nahrbéden, etc. In: Zeitschr.
f. Hyg., v..74, No. 1, pp. 61-72.

Satomonsen, C. J. 1876. Zur Isolation differenter
Bacterienformen. In: Botan. Zeitg., v. 34, No. 39,
pp. 609-622.

SanpBerG, G. 1904. Ein Beitrag zur Bakteriologie der
milcheaaten Garung im Magen, etc. In: Zeitschr.
f. klin. Med., v. 51, No. 1/2, pp. 80-94, m. 1 Taf.

SanpeR. 1893. Olen ds Wachstum von Tuberkel-
bacillen auf pflanzlichen Nahrbéden. In: Arch.
f. Hyg., v. 16, pp. 238-311, m. 1 Taf.

Sanreticz, F. 1905. Streptothrix-Pseudotuberkulose.

| Tn: Centralbl. f. Bact. I. Abt. Orig. v. 38, No. 1,
pp. 30-41, m. 1 Taf. a

Sasxr, St., 1907. Uber anaérobe Mikroben in normalen
Korpergeweben. In: Bull. intern. Acad. Cracovie.
Cl. sci. math. et nat., 1907, No. 4, pp. 2552258.

Sava, Sr. 1900. thar Fatterungspest und das Verhalten
des Pestbacillus im tierischen Kérper, etc. In:
Arch. f. Hyg., v. 39, no. 1, p. 1-30.

SAUERBECK, E. 1909. Sarcina mucosa nova species.
In: Centralbl. f. Bact. I. Abt. Orig. v. 50, No.
3, pp. 289-295.

Savace, W. G. 1901.
the typhoid bacillus. °
v. 7, pp. 888-899.

Scuapap, J. A. 1904. Actinomycosis atypica pseudo-
tuberculosa. In: Zeitschr. f. Hyg., v. 47, No. 1,
pp. 41-80, m. 1 Taf.

ScHatrenrrou, A., und GRASSBERGER, R. 1900. Uber
Buttersiiuregirung. I. Abhandlung. In: Arch. f.
Hyg., v. 37, No. 1, pp. 54-104, m. 1 Taf.

Scuaupinn, Fr. 1902. Beitrige zur Kenntnis der
Bakterien und verwandter. Organismen. I.
Bacillus Biitechlii n. sp. In: Arch.f. Protistenk.,
v. 1, No. 2, pp. 306-343, m. 1 Taf.

1903. Beitrige zur Kenntnis der Bakterien,
etc. II. Bacillus sporonema n. sp. In: Arch. f.
Protistenk., v. 2, No. 3, pp. 416-444, m. 1 Taf.

——. 1907. Zur Kenntnis der poe pallida
und anderer Spirochaeten. In: Arb. a. d. K.
Gesundh. Amt., v. 26, No. 1, an 11-22, m,.2 Taf.

Scuzeptier, H. 1887. Beitrag zur Morphologie der
Bakterien. In: Virchow’s Arch. f. pathol, Anat.,
v. 108, No. 1, pp. 30-44, m. 1 Taf,

Scuzntack, C. 1907. Morphologische Beitrige zur
Kenntnis der europidischen, amerikanischen und
afrikanischen Rekurrensspirochaeten. In: Arb. a.
d. K. Gesundh. Amt, v, 27, No, 2, pp. 364-887, m.
1 Taf.

 

Pseudo-clumping in cultures of
In: Journ. Pathol. and Bact.,

 

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

[Vou. XVI.

ScHEuiack, C. 1909. Studien zur Morphologie-und Sys-
tematik der Spirochaeten aus Muscheln. In: Arb.
a. d. K. Gesundh. Amt., v. 30, No. 2, pp. 379-428,
m. 6 Taf.

Scummpr, J., und Weis, Fr. 1902. Die Bakterien.
vu and 416 pp., m. 205 Fig. im Text. Jena.
Scumrrz, K.E.F. 1916. Die Verwandlungsfihigkeit der
Bakterien. In: Centralbl. f. Bact. I. Abt. Orig.

Vol. 77, No. 5/6, pp. 369-417, 3 Taf.

Scumort, G. 1891. Uber ein neues Fadenbakterium
(Streptothrix cuniculi), In: Deutsche Zeitschr. f.
Thiermed., v. 17, No. 5/6, pp. 375-408, m. 2 Taf.

Scunzmer, A. 1902. Contributions to the biology of
Rhizobia. I. Rhizobium mutabile in artificial
culture media. In: Botan. Gazette, v. 34, No. 2,
pp. 109-113, with 1 pl. Pe

1903. Contributions to the biology of Rhizobia.
II. The motility of Rhizobium mutabile. In:

Bot. Gazette, v. 35, No. 1, pp. 56-58.

Scnorter, B. 1904. Beitrige zur Kenntnis der Eisen-
bakterien. In: Centralbl. f. Bact. II. Abt. v.
12, No. 22/24, pp. 681-695.

Scuorrenivs. 1906. Diskussionsbemerkung zu M. Neis-
ser’s Bericht. In: Centralbl. f. Bact. I. Abt.
Ref., v. 38, Beiheft, p.98.

Scnouten, 8. L. 1905. Reinkulturen aus einer unter
den Mikroskop isolierten Zelle. In: Zeitschr. f.
wissensch. Mikroskopie, v. 22, pp. 10-45. —

——. 1907. Methode zur Anfertigung der glasernen
Isoliernadeln, gehérend zu dem Isolierapparat fir
Mikroorganismen. In: Zeitschr. f. wissensch. Mi-
kroskopie, v. 24, pp. 258-268.

ScuroEzn, O. von. 1886. Uber Tuberkelbacillen und
die Tuberkelspore. In: Tagebl. d. 59. Vers.
Deutsch. Naturf. u. Arzte (Berlin), p. 144. Abstr.
in: Baumgarten’s Jahresber. d. pathog. Mikroorg.,
v. 2, p. 198.

1890. Zur Genesis der Mikroorganismen, etc.
In: Verhandl. d. X. Intern. Med. Congresses
(Berlin), v. 2, Abt. III, pp. 120-123.

1904. Der neue Milrobe der Lungenphthise,

etc. vi und 81 pp., m. 21 Mikrophot.. Minchen.

ScuroeTer, J. 1886. Schizomycetes. In: F. Cohn’s
Krytogamenflora von Schlesien, v. 3, erste Halfte
(Pilze), pp. 186-174. Breslau.

Scutrmayer, B. 1895. Beitrige zur Beurteilung der
Bedeutung und des Verhaltens des Bacillus pyo-
cyaneus. In: Zeitschr. f. Hyg., v. 20, No.-2, pp.
281-294.

18982. Uber Entwicklungscyklen und die

verwandtschaftlichen Beziehungen héherer Spalt-

pilze. In: Verhandl. d. Ges. Deutscher Naturf. u.

Arte, v. 70 (Diisseldorf), II, 2, pp. 404-406.

18988. Artenconstanz der Bakterien und
Descendenztheorie. In: Verh. d. Ges. Deutach.
Naturf. u. Arzte, v. 70 (Diisseldorf), II, 2, pp.
406-408.

——. 1900. Uber Aktinomykose des Menschen und
der Tiere. In: Centralbl. f. Bact. I. Abt. v. 27,
No. 2, pp. 49-61; No. 8, pp. 101-107, m. 2 Taf.

Scutirzz, H. 1908. Beitrige zur Kenntnis der thermo-
philen Aktinomyceten, etc. In: Arch. f. Hyg.,
v. 67, No. 1, pp. 35-56, m. 2 Taf.

 

 

 

 

 
No. 2.] LIFE CYCLES OF THE

Scuuutz, J. H., und Rrrz, H. 1910. Die Thermoresis-
tenz junger und alter Coli-Bacillen. In: Centralbl.
f. Bact., I. Abt., Orig. v. 54, No. 3, pp. 283-
288.

Scuurrz, N. K. 190la. De la vitalité du microbe de la
peste bubonique dans les cultures. In: Arch. sci.
biol. St.-Pétersbourg, v. 8, No. 4, pp. 373-389,
av. 1 pl.

19016. Uber die Lebensdauer von Bacillus
pestis, etc. In: Centralbl. f. Bact. I. Abt. v.
29, No. 5, pp. 169-174.

Scuuize, O. 1899. Untersuchungen iiber die Strahlen-
pilzformen des Tuberkuloseerregers. In: Zeitschr.
f. Hyg., v. 31, No. 1, pp. 153-186, m. 1 Taf.

Scuumowsx1, W. 1898. Uber die Beweglichkeit der

Tuberkelbacillen. In: Centralbl. f. Bact. I. Abt.

v. 23, No. 19, pp. 838-839.

1899. Studien an auf eiweissfreien Nahrbéden
geztichteter Tuberkulose. In: Arb. a. d. Labor. f.
allg. Pathol. d. Univ. Warschau, v. 1 (Russian).
Abstr. in: Centralbl. f. Bact. I. Abt. v. 26, No.
4/5, p. 155.

Scuuster, J. 1910. Uber einen Fall von Bakterien-
Plasmoptyse. In: Ber. d. Deutschen Bot. Ges., v.
28, No. 9, pp. 488-496.

Scuwatpe, E. 1900. Uber Variabilitit und Pleomor-
phismus der Bakterien. In: Mimchn. med. Wo-
‘chenschr., v. 47, No. 47, pp. 1617-1621.

Szrrz,: J. 1896. Streptococcus aggregatus. In: Cen-
tralbl. f. Bact. I. Abt. v.20, No. 24, pp. 854-862.

Setznew, O.F. 1910. Zur Morphologie der Spirochaeta
pallida, Ring-. und Sternformen derselben. In:
Centralbl. f. Bact. I. Abt. Orig. v. 54, No. 1,
pp. 7-11, m. 2 Taf.

Semmer, E. 1895. Uber die Morphologie des Tuberkel-
und Rotzbacillus und den Ursprung der patho-
genen Schizomyceten. In: Deutsche Zeitschr. f.
Thiermedicin, v. 21, No. 3/4, pp. 212-214.

Severin, S. A. 1897. Zur Frage tiber die Zersetzung
von salpetersauren Salzen, etc. In: Centralbl. f.
Bact. II. Abt. v. 3, No. 19/20, pp. 504-517; No.
21/22, pp. 554-563.

 

 

SEIPATA, K. 1902. Cytologische Studicn iiber die endo- |

' trophen Mykorrhizen. In: Jahrb. f. wissensch.
Botanik, v. 37, pp. 643-684, m. 2 Taf.

Surpayama, G. 1902. Uber die Veristelung der Chole-
‘ravibrionen und deren Bedeutung. In: Mitt. d.
Med. Ges. zu Tokio, v. 16, No. 75, p. 1 (Japanese).
Abstr. in: Centralbl. f. Bact.
No. 4/5, p. 111.

Sumammeg, T. 1912. Uber die Reinztichtung der Spi-
rochaeta pallida und der nadelférmigen Bakterien
aus syphilitischem Material, etc. In: Centralbl.
f. Bact. I. Abt. . Orig. v.65, No. 4/5, pp. 311-337,
m. 4 Taf.

Sieper, H. 1910. -Uber. Anaplasma marginale. In:
Berl. klin. Wochenschr., v. 26, No. 50, pp. 993-998.

S1zeser, N. 1895. Zur Frage nach dem Fischgifte, etc.
In: Gazetta lekarska, 1895, No. 13-17 (Russian).
Abstr. in: Centralbl. f. Bact., v. 17, p. 888.-

Srecen, J. 1905a. Untersuchungen iiber die Aetiologie
der Pocken und der Maul- und Klauenseuche. In:

I. Abt. Ref., v. 34, |

 

BACTERIA—LOHNIS. 241

Anhang z. d. Abhandl. d. K. Preuss. Akad., 1905.
Physik. Abhandl., I,' pp. 1-34, m. 2 Taf.

Stzcet, J. 1905b. Untersuchungen tiber die Aetiologie
des Scharlachs. In: Anhang z. d. Abhandl. d. K.
Preuss. Akad., 1905. Physik. Abhandl., II, pp.

1-14, m. 1 Taf.

1905c. Untersuchungen iiber die Aetiologie der
Syphilis. In: Anhangz.d. Abhandl. d. K. Preuss.
Akad., 1905. Physik. Abhandl., III, pp. 1-25, m.
2 Taf.

1910. Gelungene Reinkultur des Cytorrhyctes
luis. Vorl. Mitt. In: Centralbl. f. Bact. I. Abt.
Orig. v. 57, No. 1, pp. 68-81, m. 1 Taf.

1911. Gelungene Reinkultur des Cytorrhyctes

vaccinae. In: Centralbl. f. Bact. I. Abt. Orig. v.

59, No. 4, pp. 406-415, m. 2 Taf.

1912a, Einige erginzende Bemerkungen zum
Nachweis der Cytorrhyctescoccen bei Maul- und
Klauenseuche. In: Berlin. tierarztl. Wochenschr.,
v. 28, No. 2, pp. 27-29.

19126. Zur Aetiologie der Maul- und Klauen-

seuche. In: Berlin. tierarztl. Wochenschr., v. 28,

No. 45, pp. 821-822, m. 1 Taf.

1914. Untersuchungen iiber die Aetiologie der
Maul- und Klauenseuche, etc. In: Berlin. tier-
airztl. Wochenschr., v. 30, No. 1, pp. 1-3; No. 2,
pp. 25-27.

Smperscumipt, W. 1901c. Uber Aktinomykose.

Zeitschr. f. Hyg., v. 37, pp. 345-380, m. 2 Taf.

19016. Uber den Befund von spiessférmigen
Bacillen, ete. In: Centralbl.f. Bact. I. Abt. v.
30, No. 4, pp. 159-166.

Srmonps, J.P. 1915. The effect of symbiosis upon spore
formation by Bacillus Welchii. In: Journ. Infect.

- Diseases, v. 16, No. 1, pp. 35-37.

Smonr, A. pz. 1898. Uber einen sporogenen Pseudo-

Diphtheriebacillus. In: Centralbl. f. Bact. I. Abt.

v. 24, No. 8, pp. 294-297, m. 1 Taf.

1900. Beitrige zur Morphologie und Biologie
der Mucosusbacillen der Ozaena und iiber ihre Iden-

titat mit den Pneumobacillen. In: Centralbl. f.
Bact. I. Abt. v. 27, No. 12/13, pp. 426-436; No.
14/15, pp. 493-502, m. 2 Taf.

Smmonin1, A. 1914. Uber die Einwirkung seltener Er-

: den auf Bakterien. In: Centralbl. f. Bact. I.
Abt. Orig. v. 74, No. 3/4, pp. 3438-348, m. 1 Taf.

1915. Einwirkung der seltenen Exiien auf Bak-
terien. Zweite Mitt. In: Centralbl. f. Bact. I.
Abt. Orig. v. 75, No. 5/6, pp. 398-407.

 

 

 

 

 

 

In:

 

 

 

| Sxscurvan, T.-1900. Zur Morphologie des Pestbak-

teriums. In: Centralbl. f. Bact.
No. 10/11, pp. 289-292, m. 4 Fig.

Sxuprensxl, F.X. 1918. Sur lasexualité chez les cham-
pignons myxomycétes. In: Compt. rend. Acad.
Paris, v. 167, No. 1, pp. 31-84.

I. Abt. v. 28,

‘Starer, CH. 1891. Onared pigment forming organisms,

Bac. corallinus (?). In: Quart. Journ. Microsc.
Science, n.s., v. 32, No. 127, pp. 409-416, with 1 pl.

Surrnow, M. R. 1908. Some symbiotic relations of the
Bacillus diphtheriae. In: Journ. Med. Res., v. 18,
No.' 2, pp. 257-276,
242

Smira, A. L. 1901. Myxobacteria. In: Journ. of Bot-
any British, and Foreign, v. 39, No. 458, pp. 69-72.

Smita, E. F. 1905-14.. Bacteria in relation to plant
diseases. 3 vols. Carnegie Inst. Publ. No. 27.
Washington, D.C. V.I: x1 and 285 pp., with 146
figs. and 31 pls. (1905). V. II: vm and 368 pp.,
with 148 figs. and 20 pls. (1911). V. III: vu and
309 pp., with 138 figs. and 45 pls. (1914). .

1916. Studies on the crown ‘gall of plants.

In: Journ. Cancer Res., v. 1, No. 2, pp. 231-258,

with 24 pls.

and Swinetz, D. B. 1905. Der Einfluss des

Gefrierens auf Bakterien. In: Centralbl. f: Bact.

I. Abt. Ref., v. 37, No. 11/14, pp. 357-359.

Suir, R. W., and Baker, J. L. 1898. Bacillus luteus
sporogenes. In: Centralbl.f. Bact. II. Abt. v.4,
No. 21, pp. 788-789.

Smrru, Ta. 1890. On the influence of slight modifica-
tions of culture media, etc. In: Journ. Comp.
Medic. and Veterin. Archives, v. 11, No. 3, pp.

159-161.

1894. Modification, iemporary and permanent,
of the physiological characters of bacteria in mixed
cultures. In: Transact. Assoc. Amer. Physicians,
v. 9, pp. 85-106.

1900. Variation among pathogenic baeteria. In:
Journ. Boston Soc. of Medical Sciences, v. 4, No.
5, pp. 95-109.

——. 1913. Notes on the biology of the tubercle
bacillus. In: Journ. Med. Res., v. 28, No.1, pp.
91-110.

—, Wik. A pleomorphic bacillus fron, pneumonic
lungs of calvessimulating Actinomyces. In: Journ.
Exp. Med. v. 28, No. 3, pp. 333-344, 4 pls.

SopernHem, G. 1903. Milzbrand. In: Kolle-Wasser-
mann’s Handb. d. pathog. Mikroorg., v. 2, pp. 1-77.

-——. 1907. Spirillosen. In: Kolle-Wassermann’s
Handb. d. pathog. Mikroorg. I, Erg. Bd. pp.
522-600, m. 1 Taf.

Séunezn, N. L. 1913. Benzin, Petroleum, Paraffinél
und Paraffin als Kohlenstoff- und Energiequelle fir
Mikroben. In: Centralbl.f. Bact. II. Abt. v.37,
No. 22/25, pp. 595-609, m. 3 Taf.

Soroxm, N. 1887. Eine neue Spirillum-Art. Vorl.
Mitt. In: Centralbl. f. Bact., v. 1, No. 16, pp. 465-
466, m. 1 Abb.

———. 1890. Noch einmal tber Spirillum endopara-

 

 

 

gogicum. In: Centralbl. f. Bact., v. 7, No. 4, pp. |

123-124, m. 1 Abb.

Spencer, H. 1898. The principles. of biology. 2d ed. | ~

v. 1, xm and 706 pp. New York.

Spenater, C. 1905. Uber Splittersputa Tuberkuléser.

. In: Zeitschr. f. Hyg., v. 49, No. 3, pp. 541-544, m.
‘1 Taf.

——. 1907.
Tuberkelbacillen, etc. In: Deutsche med. Wo-
chenschr., v. 33, No. 9, pp. 337-338,

Srrric, W. 1899. Die Streptothrix-
Natur des Diphtheriebacillus.
Bact. I. Abt, v. 26, No. 18/19, pp. 540-541.

——. 1908. Studien iiber den Diphtheriebacillus.
In: Zeitschr. f£. Hyg., v. 42, No. 3, pp. 420-460, m.
3 Taf.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

Neue Farbamothodan fir Perlsucht- und

(Actinomyces-) |
In: Centralbl. f.

 

[Vor. XVI.

| Spratt, E. R. 1912. The morphology of the root-

tubercles of Alnus and ‘Elaeagnus, and the poly-
morphism. of the organism causing their formation.
IV: Annals of Botany, vol. 26, No. 101, pp. 119-128,
with 2 pls.

Sprinager. 1913. Beitrag zur Kenntnis des Stoffwechsela
. des Bacillus diphtheriae. In: Arch. f. Hyg., v.79,
No. 1, pp. 25-58.

Stamm, J. 1914. Zur Frage der Veranderlichkeit der
Choleravibrionen in Wasser. In: Zeitschr. f. Hyg.,
v. 76, No. 3, pp. 469-542, m. 2 Taf.

Sreran, J. 1906. Studien zur Frage der Leguminosen-
knéllchen. In: Centralbl. f. Bact. II. Abt. v.16,
No. 4/6, pp. 131-149, m. 2 Taf.

Sreransky, W. K. 1902. Uber ein neues, Eiterung
hervorrufendes, verzweigtes Bakterium. In: Cen-
tralbl. f. Bact. I. Abt. Orig., v. 31, a 8, pp.
86-92.

SternsBeERG, OC. 1898. Zur Biologie des Boas’schen Milch-
saurebacillus, etc. In: Wien klin. Wochenschr., v.
11, No. 31, pp. 744-747.

Srernsperc, G. M. 1893. Manual of pacteriology. xi
and 886 pp. illustr. New York. .-

1896. Textbook of bacteriology. x1 and 693
pp., with 200 figs. and 9 pls. New York.

Sroxzasa, J. 1898. Biologische Studien iiber ‘‘Alinit.’’
In: Centralbl. f. Bact. II. Abt. v. 4, No. 1, pp.
89-41; No. 2, pp. 79-86; No. 3/4, pp. 119-130; No.
7, pp. 284-289; No. 12, pp. 507-513; No. 13, pp.
535-540,

Srouz, A. 1897. Uber einen Bacillus mit Verzweigun-
gen. In: Arch. f. Hyg., v. 30, No. 2, pp. 156-167.

1898. ber besondere Wuchsformen bei

Pneumo- und Streptokokken. In: Centralbl. f.

Bact. I. Abt. v.24, No. 9, pp. 337-345.

Sturzer, A. 1900. Beitrige zur Morphologie der als
“Bacterium radicicola” beschriebenen Organis-
men. In: Mitt. d. Landw. Inst. Breslau, v. 1, No.
3, pp. 57-72, m. 1 Taf.

——. 1901c. Die’ Organismen der Nitrifikation. In:
Centralbl. f. Bact. II. Abt. v. 7, No. 5/6, pp.
168-172.

190iv. Die Bildung von Bakteroiden in
kimstlichen Nahrbéden. In: Centralbl. f. Bact.
II. Abt. v. 7, No. 25, pp. 897-912.

—— und Harriss, R., 1897a. Der Salpeterpils.,
Centralbl. f. Bact, II. Abt. v. 3, No. 1, pp.
_ No. 2/3, pp. 54-57; No. 7/8, pp. 161-177; No. 9/10,
pp. 235-245; No. 112, pp. 311-321; No. ‘13/14, pp-
pp. 351-354.

18976. Das Bacterium der Maul- und

Klauenseuche. In: _Arch. f. Hyg., v. 80, No. 4,

pp. 372-404.

—. 18992. Untersuchungen tiber die bei der

Bildung von Salpeter beobachteten Mikroorganis-

men. I. Mitt. In: Mitt. d. Landw. Inst. Breslau,

v. 1, No. 1, pp. 75-100, m. 2 Taf.

"18995. Untersuchungen liber die bei
der Bildung von Salpeter beobachteten Mikro-
organismen. II. Mitt. In: Mitt, d. Landw. Inst.
Breslau, v. 1, No. 2, pp. 197-282, m. 1 Taf.

Stcurine, H. 1903. Kritische Studien wber die Knill-
chenbakterien. In: Centralbl. f. Bact. II. Abt.
v.11, no. 12/13, pp. 377-888; No. 14/15, pp. 417-441;
No. 16/18, pp. 496-520.

 

 

 

 

 
No. 2.] LIFE CYCLES OF THE

SwELLENGREBEL, N. H. 1904. Quelques notes sur la
morphologie et la biologie du Bactérium Zopfii
(Kurth). In: Annal. Inst. Pasteur, v. 18, No. 11,
pp. 712-720.

——. 1906. Zur Kenntnis der Cytologie von Bacillus

"maximus buccalis (Miller). In: Centralbl. f. Bact.
II. Abt. v. 16, no. 20/21, pp. 617-628; no. eet,
‘pp. 673-681, m. 1 Taf.

- 1907a. Zur Kenntnis der Cytologie der Bakte-

rien. II. Bacterium binucleatum. In: Centralbl.

f. Bact, II. Abt. v. 19, No. 7/9, pp. 193-201,

m, 21 Abb.

19076. Surlacytologie comparée des spirochétes
et des spirilles. In: Annal. Inst. Pasteur, v. 21,
No. 6, pp. 448-465; No. 7, pp. 562-586, av. 2 pl.

——. 1909. Untersuchungen iiber die Zytologie eini-

’ ger Fadenbakterien. In: Arch. f. Hyg., v. 70, No. 4,
pp. 380-404, m. 2 Taf.

——. 1912. Trypanosomen, Spirochaeten und Bak-
terien. In: Centralbl. f. Bact. I. Abt. Ref., v.
51, No. 5/6, pp. 129-158.

——. 19138. Zur Kenntnis der Sporenbildung bei Bak-
terien. In: Arch. f. Protistenk., v. 31, No. 3,
pp. 277-285, m. 1 Taf.

Tappgr, D. 1909. Beitrag zum Studium der Morpho-
logie des Streptococcus. In: Centralbl. f. Bact.
I. Abt. Orig. v. 50, No. 6, pp. 561-571.

Tanakamury, J. 1913. Uber die Natur der sogenannten
Lyssakérperchen.
207 (Japanese). Abstr. in: Centralbl. f. Bact. I.
Abt. Ref., v. 60, No. 9, pp. 266-268.

Tanner, F.W. 1918. <A study of green fluorescent bac-
teria from water. In: Journ. Bact., v. 3, No. 1, pp.
63-101.

Tavet, E. 1898. Uber den Pseudotetanusbacillus des
Darmes, In: Centralbl. f. Bact. I. Abt. v. 23,

No. 18, pp. 588-541, m. 1 Taf.

Tetcu,M. 1899. Beitrage zur Kultur des Leprabacillus.
In: Centralbl. f. Bact. I. Abt. v. 25, No. 21/22,
pp. 756-761.

Tuaxrer, R. 1892. On the Myxobacteriaceae, a new
order of Schizomycetes, In: Bot. Gazette, v. 17,
No. 12, pp. 389-406, with 4 pls.

 

 

——. 1897. Further observations on the Myxobacte-

riaceae. In: Bot. Gazette, v.23, No. 6; pp. 395-411,

with 2 pls.

1904. Notes on the Myxobacteriaceae. In: Bot.
Gazette, v. 37, No. 6, pp. 405-416, with 2 pls.
Tuerter, A. 1910. Anaplasma marginale (gen. and spec.
nov.). In: Report of the Govt. Veterin. Bacteriolo-
gist for 1908/09. Pretoria, 1910, p. 7. Abstr. in:
Centralbl. f. Bact. I. Abt.

p. 10.

THIERCELIN, E. 1899. Morphologie et modes de repro-
duction de V’entérocoque. In: Compt. rend. Soc.
‘Biol., v. 51, No. 26, pp. 551-553.

1903. Formes d’involution de l’entérocoque.
Entérobactérie. In: Compt. rend. Soc. Biol. , v.55,
“No. 1, pp. 24-26.
etJouHaup,L. 1903a. Reproduction de l’entéro-
-coque: taches cenirales, granulations périphériques
“et microblastes. In: Compt. rend. Soc. Biol., v. 55,
pp. 686-688.

 

 

 

In: Saikingaku-Zasshi, 1913, No. |

 

243

THIERCELIN, E., et Jounaup, L. 19036. Discipline des
variations de formes de l’entérocoque. In: Compt.
rend. Soc. Biol., v. 55, p. 701-703.

Trzaet, E. 1874. Uber Coccobacteria septica Billroth
im gesunden Wirbelthierkérper. In: Virchow’s
Arch. f. pathol. Anat., v. 60, No. 3/4, pp. 453-470.

Tisster, H. 1900. Recherches sur la flore intestinale des
nourrissons. Thése méd. Paris. 253 pp., av. 17
figs. dans le texte.

Tizzoni, G., und ANGELIS, G. pz. 1913a. Studien iiber
die Biologie und die Morphologie des pleomorphen
Streptobacillus der Pellagra. Vorl. Bericht. In:
Centralbl. f. Bact. I. Abt. Orig. v. 69, No. 1/2,

pp. 5-8.

BACTERIA—LOHNIS.

 

19136. Uber den Entwicklungszyklus

des pleomorphen Streptobacillus der Pellagra.

Zweite vorl.. Mitt. In: Centralbl. f. Bact. I. Abt.

Orig. v. 70, No. 1/2, pp. 24-29.

1914, Hauptcharaktere des Streptoba-

cillus pellagrae, etc. In: Centralbl.f. Bact. I. Abt.

Orig. v. 74, No. 3/4, pp. 219-224.

1915. Bedeutung des Pleomorphismus
bei der Identifikation und Klassifikation des Strep-
tobacillus pellagrae (T.). In: Centralbl. f. Bact.
I, Abt. Orig. v. 76, No. 1, pp. 47-48.

TornnisssEeN, E. 1913. Uber Weeen und Ursache der
Mutation bei Bakterien. In: Centralbl.f. Bact. I.
Abt. Orig. v. 69, No. 5/6, pp. 391-412, m. 1 Taf.

1914. Uber Vererbung und Variabilitét bei

Bakterien, etc. In: Centrabl. f. Bact. I. Abt.

Orig. v.73, No. 4/5, pp. 241-277, m. 2 Taf.

TomascHEK, A. 1888. Uber Bacillus muralis, etc.
Botan. Centralbl., v. 36, No. 6, pp. 180-186.

TramBustt, A., und Gatzort1, G. 1892. Neuer Beitrag
zum Studium der inneren Struktur der Bakterien.
In: Centralbl. f. Bact., v. 11, No. 23, pp. 717-722.

TRAUTMANN und GagETHcEeNns. 1913. Einige schwe-
bende Fragen der bakteriologischen Diphtherie-
diagnose. In: Centralbl. f. Bact. I. Abt. Ref.,
v.57, Beiheft, pp, 61-78. |

Tricut, A. 1865. Production de plantules amyliféres
dans les cellules végétales pendant la putréfaction.
In: Compt. rend. Acad. Paris, v. 61, No. 11, pp.
432-436.

1867a. Réponse A trois notes de M. Nylander

concernant la nature des Amylobacter. In: Compt.

rend, Acad. Paris, v. 65, No. 13, pp. 513-521.

1867b. Examen de quelques objections qui

pourraient étre faites 8 mon travail sur l’origine des

Amylobacter. In: Compt. rend. Acad. Paris, v.

65, No. 23, pp. 927-984.

 

 

 

 

In:

 

 

| Tromi-Perersson, G. 1914. Ein-Zell-Kultur von lang-
Ref., v. 50, No. 1/2, | :

sam wachsenden Bakterienarten, etc. In: Central-
pl. f, Bact. II. Abt. v.42, No. 17/18, pp. 526-528.

Tscuircu, A. 1887. Beitrige zur Kenntnis der Wurzel-
knéllchen der Léguminosen. In: Ber. d. Deutschen
Bot. Ges., v. 5, No. 2, pp. 58-98, m. 1 Taf.

Tutasne, L.-R. 1851. Note sur appareil reproducteur
dans les Lichens et les Champignons: In: Compt.
rend. ‘Acad, Paris, v. 32, No. 12, pp. 427-430; No.
18, pp. 470-473.

Tunnicurr, R. 1906. The identity of fusiform bacilli

' and gpirilla. In: Journ. Infect. Diseases, v. 3, No.

1, pp. 148-155, with 1 pl.
244

Tunnicurr, R. 1911. Further studies on fusiform ba-
cilli and spirilla. In: Journ. Infect. Diseases, v. 8,
No. 3, pp. 316-321.

1918a. A spirochaete associated with infections

of the accessory sinuses. In: Journ. Infect. Dis-

eases, v. 13, No. 2, pp. 280-282.

19136. An anaérobic: organism associated with
acute rhinitis. In: Journ. Infect, Diseases, v. 13,
No. 2, pp. 283-288.

1915. Further observations on the bacteri-
ology of rhinitis with special reference to an an-
aérobic organism (Bacillus rhinitis). In: Journ.
Infect. Diseases, v. 16, No. 3, pp. 493-495.

UnLenuutH, XYLANDER, Hipener, und Boutz. 1908.
Untersuchungen tiber das Wesen und die Bekim-
pfung der Schweinepest. In: Arb. a. d. K. Ge-
sundh. Amt, v. 27, No. 3, pp. 425-671.

1909. Weitere Unter-
suchungen iiber das Wesen und die Bekampfung
der Schweinepest, etc. In: Arb. a.d. K. Gesundh.
Amt, v. 30, No. 2, pp. 217-329.

Unwa, P.G. 1887. Die Rosaniline und Pararosaniline.
In: Dermatolog. Studien, herausg. v. P. G. Unna,
Heft 4, pp. 1-73 (Erg. Heft I z. Monatshefte f.
pract. Dermatologie, Jahrg. 1887).

Uny,J. 1894. Uher die Schwankungen des Bact. coli
commune in morphologischer und kultureller Be-
ziehung. Diss. med. Strassburg. 47 pp.

Vane, ©. 1909. Vergleichende Untersuchungen tiber
die Myxobakteriazeen, etc.

Ii. Abt. v. 25, No. 5/9; pp. 178-260, m. 2 Taf.

Van Trncuim, Pa. 18792. Développement du Spiril-

 

 

 

 

 

 

lum amyliferum sp. nov. In: Bull. Soc. Botan. de |

France, v. 26, pp. 65-68.
1879b. Sur les spores de quelques bactéries.
In: Bull. Soc. Botan. de France, v. 26, pp. 141-142.
———. 1879c. Sur la gomme de sucrerie (Leuconostoc
mesenterioides). In: Annal. sci, nat. Botanique.
6me sér., v. 7, No. 3/4, pp. 180-208, av. 1 pl.
1880. Sur quelques bactéries aggrégées. In:
Bull. Soc. Botan. de France, v. 27, pp. 148-153.
-——, 1884. ‘Traité de Botanique. xxxm et 1656
pp., av. 803 grav. dans le texte. Paris,
Vay, F. 1909. Uber kornchenartige Bildungen i in Pest-
pakterien. In; Centralbl. f. Bact. I. Abt. Orig.
v. 52, No. 3, pp. 305-318, m. 1 Taf.
Vuspovskt, F. 1900. Bemerkung tiber den Bau und
die Entwicklung der Bakterien. In: Centralbl. f.

 

 

Bact. II. Abt. v. 6, No. 18, pp. 577-589, m. 1
Taf. ;
———. 1904. Uber den Kern der Bakterien und seine

‘Teilung. In: Centralbl. f. Bact.
No. 16/18, pp. 481-496, m. 1 Taf.

VeszPRiMt, D. 1907. Ziichtungs- und ‘Tierversuche mit
Bacillus fusiformis, etc. In: Centralbl. f. Bact.
I. Abt. Orig. v. 44, No. 4, pp. 332-339; No. 5,
pp. 408-415;,No. 6, pp. 515-523; No. 7, pp. 648-665;
y. 45, No. 1, pp. 15-33, m. 4 Taf.

Vrenorver, A. 1913, Botanische “Untersuchung harn-
stofispaltender Bakterien, etc. Centralbl. f. Bact.
II. Abt. v. 39, No. 8/14, pp.. 209-359, m. 2 Taf.

Vienat, W. 1889. Contribution 4 l'étude de bactéria-
cées, 181 pp. Paris.

II. Abt. v. 11,

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

In: Centralbl. f. Bact. }

 

{ Vou. XVI.

Vincent, H. 1902, Sur les variations morphologiques
du streptocoque et sur un streptocoque ramifié. ,
In: Arch, de méd. expér. et d’anat. pathol., v. 14,
No. 5, pp. 521-524.

1905. Sur la non-identité du bacille fusiforme
et du Spirillum sputigenum. In: Compt. rend.
Soc. Biol.,.v. 58, pp. 499-501.

Vincentini1, F. 1898. Della Leptothrix racemosa. In:
Atti R. Accad. Med.-Chirurg. Napoli, n.s., v. 46,
No. 4. Abstr. in: Baumgarten’s Jahresber. d. pa-

* thog. Mikroorg., v. 9, p. 430.

Vries, H. pz. 1889. Intracellulare Pangenesis, v1und
212 pp. Jena.

Vuniemin, P. 1888. Les tubercules radicaux des légu-
mineuses. In: Annal., sci. agron., v. 5, 1888, I,
pp. 121-212, av. 2 pls.

1903. La famille des Clostridiacées ou Bactéria-
cées cystosporées. In: Compt. rend. Acad. Paris,
v. 136, No. 25, pp. 1582-1584. .

Wave, H. W. 1918. Studies on cryptoplasmatic infec-
tion. 1. Development ofa Cryptococcusin cultures
from an unclassified chronic Philippine ulcer. In:
Philippine Journ. of Science v. 13, Sect. B, No. 4,
pp. 165-189, with 5 pls.

Wasruscu, L. 1905. Uber einen eigenartigen Mikroor-
ganismus im Praeputialsekret. In: Centralbl. f.
Bact. I. Abt. Orig. v. 38, No. 6, pp. 645-649,
m. 1 Taf.

Waker, E. W. A., and Murray, W. 1904. The effect
of certain dyes upon the cultural characters of the
Bacillus typhosus and some other microorganisms.
In: Brit. Med. Journ., 1904, II, No. 2270, pp.16-18.

Waxxer, I. C., and Apxinson, J. 1917. Studies of a
diphtheroid organism isolated from the sputum of
patients with bronchial asthma. In: Journ. Med.
Res., v. 35, No. 3, pp. 391-402.

WaxTon, 1. H. 1915. Azotobacter and nitrogen fixation
in Indian soil. In: Mem. Dept. Agric. India, Bac-
teriol. Ser., v. 1, No. 4, pp. 97-112, with 5 pls.

Warp, H.C. 1917. Coccobacillus (foetidus) ozaenae of
Perez. In: Journ. Infect. Diseases, v. 21, No. 4,
pp. 338-340, with 1 pl.

Warp, H. M.. 1887. Schizomycetes. In: Encyclopae-
dia Britannica. 9thed. v. 21, pp. 415-425.
—. 1892. On the characters, or marks, employed
; for classifying the Schizomycetes. In: Annals of

Botany, v. 6, No. 21, pp. 103-144.

1895a, The formation of bacterial colonies.

Annals of Botany, v. 9, No. 36, pp. 653-657.

1895b. A false bacterium. In: Annals of Bot-

any, v. 9, No. 36, pp. 657-658. ;

1910. Bacteriology. In: Encyclopaedia Britan-
nica. llithed. v. 3, pp. 156-171. es

Warminc, E. 1876. Observationssur quelques bactéries
qui se rencontrent sur les cétes du Danemark. In:
Vidensk. Medded. fr. d. Naturhist. Foren. i Kjg-

benhavn, v. 28, pp. 1-36.

WASSERMANN, A. 1908. Bacillus pyocyaneus. In:
Kolle-Wassermann’s Handb. d. pathog. Mikroorg.,
v. 3, pp. 471-488,

Wasszrzuc, E. 18880. Variations de forme chez les
bactéries. In: Annal. Inst. Pasteur, v. 2, No. 2,
pp. 75-83. =o

 

 

 

In:

 

 
No.2.1 LIFE CYCLES OF THE

Wasserzue, E. 18886. Variations durables de la forme
et de la fonction chez les bactéries. In: Annal.
Inst. Pasteur, v. 2, No. 3, pp. 153-157. ;

Weaver, G. H., and Tunnicurr, R.. 1905. The occur-
rence of fusiform bacilli and spirilla in connection
with morbid processes. In: Journ. Infect. Diseases,
v. 2, No. 3, pp. 446-459, with 1 pl.

Wecusetmann, W., und Léwenrnat, W.
Kenntnis der Spirochaeta pallida. In: Med. Kli-
nik, v. 1, No. 33, pp. 838-841.

Wesrte und Zwick. 1913. Verlauf und Ergebnisse der
Ubertragungsversuche, die im Kaiserl. Gesund-
heitsamte . angestellt wurden. In: Arb. a.
d. K. Gesundh. Amt, v. 45, No. 4, pp. 522-583,

Wewet,E. 1887. Untersuchungen tiber Vibrionen. In:
Centralbl. f. Bact., v. 2, No. 16, pp. 465-472, m. 1
Taf.

——. 1888. Untersuchungen tiber Vibrionen. ITI. In:
Sfatedet f. Bact., v. 4, No. 8, pp. 225-232; No.

9, pp. 257-264; No. "10, pp. 289-295, m. 10 Abb.

Weicert, CO. 1875. Uber ‘pockonslintiche Gebilde in
parenchymatésen Organen und deren Beziehungen
zu Bacteriencolonien. Habil. Schrift Breslau. 46

p.

Weiemann, H. 1899. Versuch einer Einteilung der
Milchsiurebakterien des Molkereigewerbes. In:
Centralbl. f. Bact. II. Abt. v.5, No. 24, pp. 825-
831; No. 25, pp. 859-870, m. 1 Taf.

—, Guvaze, Ta., und Huss, H. 1907. Uber armeni-
ches Manun, In: Centralbl. f. Bact. II. Abt. v.
‘19, No. 1/3, pp. 70-87, m. 3 Taf.

Weismann, A. 1886. Zur Annahme einer Continuitat
des Keimplasmas. 11 pp. Freiburg i. B. (Re-

print from: Ber. d. Naturf. Ges. Freiburg, v. 1,
No. 4.)

——. 1892. Uber die Vererbung. 2. Aufl. 49 pp.
Jena,

Wertueim, E. 1899. Der Gonococcus auf kimstlichen
Nahrbéden. In: Verhandl. d. Ges. Deutscher
Naturf. u. Arzte, v. 71, II, 2-pp. 403-406.

Wessroor, F, F., Wmson, L. B., and McDanret, O.
1900. Varieties of Bacillus diphtheriae. In:
Transact. Assoc. Amer. Physicians, v. 15, pp. 198-
223, with 6 pls.

Wuerry, W. B. 1918a. Some chemical conditions in-
fluencing acid-proofness and nonacid-proofness in a
saprophytic culture of B. tuberculosis. In: Journ.

Infect. Diseases, v. 18, No. 1, pp. 144-154.

19136. Some chemical conditions favoring the
production of “spores” in B. tuberculosis. In:
Centralbl. f. Bact. I. Abt.
pp. 115-118, with 2 pls.

‘Wut, B., and Avery, O. T. 1909. Observations on
certain lactic acid bacteria of the so-called Bulga-
ricus type. In: Centralbl. i. Bact.’ II. Abt. v.
25, No. 5/9, pp. 161-178, with 2 pls.

Wieann, A. 1884. Entstehung und Fornientwitkiang
der. Bakterien. Vorl. Mitt. 2. Aufl. 40 pp.
Marburg.

Wuuams, A.W. 1918. Cultivation of the rabies crgan-

, ism. In: Journ. Amer. Med. Assoc., v. 61, Ne: 17,
pp. 1509-1511.

Wrson, M. A., and Guriny, C. R. 1914, Aataay

of trachoma, he:

 

 

1905. Zur

Orig. v.70, No. 3/4, |

 

V. Microscopic and cultural |

BACTERIA—LOHNIS.

2 3
studies. In: Journ. Infect. Diseases, v. 14,,No.. 3,.7
pp. 312-329. BS. Ai

Wiitams, H.U. 1912. A pleomorphic bacillus growing
in association with a streptothrix. In: Journ. Med.
Res., v. 27, No. 2, pp. 157-161, with 1 pl.

Writams, T.8.B. 1911. Leprosy. Supplement to the
Indian Med. Gazette, May, 1911, 8 pp., with 4 pls.

Wison, W. J. 1907. Pleomorphism, as exhibited by
bacteria grown on media containing urea. In:
Journ. of Pathol. and Bacteriol., v. 11, No. 4, pp.

394-404, 1 pl.
WinkiER, F. 1906. Der gegenwirtige Stand der Cytor-
rhyctesfrage. In: Wien. klin. Wochenschr., v. 19,

No. 12, pp. 340-342.

Wingier, W. 1895. Zu Charakterisierung der Duclaux’
schen Tyrothrixarten, etc. In: Centralbl. Bact.
II. Abt. v. 1, No. 17, pp. 609-619; No. 18/19,
pp. 657-674, m. 2 Taf.

1899. Untersuchungen iiber das Wesen der
Bakterien und deren Einordnung im Pilzsystem.
In: Centralbl. f. Bact. II. Abt. v. 5, No. 16/17,
pp. 569-579; No. 18/19, pp. 617-630, m. 2 Taf.

Wivograpsxy, §. 1887. Uber Schwefelbacterien. In:
Bot. Zeitg., v. 45, No. 31, pp. 489-507; No. 32, pp.
513-523; No. 33, pp. 529-539; No. 34, pp. 545-559;
No. 35, pp. 569-576; No. 36, pp. 585-594; No. 37,

pp. 606-610.

1888. Beitrage zur Morphologie und Physiologie
der Bacterien. Heft I. 120 pp.,m.4 Taf. Leip-
zig.

1889. Sur le pléomorphisme des bactéries. In:

Annal. Inst. Pasteur, v. 3, No. 5, pp. 249-264.
——. 1891. Recherchessur les organismes de la nitri-

fication. 5me mém. In: Annal. Inst. Pasteur,
v. 5, No. 9, pp. 577-616, av. 1 pl.

1892. Contributions 4 la morphologie des orga-
nismes de la nitrification. In: Arch. sci. biol.
St. Pétersbourg, v. 1, pp. 87-187, av. 4 pl.

1902. Clostridium Pastorianum, seine Morpho-
logie, etc. In: Centralbl. f. Bact. II. Abt. v. 9,
No. 1/2, pp. 48-54; No. 3/4, pp. 107-112, m. 1 Taf.

1904. Die Nitrifikation. In: Lafar’s Handb. d.
Techn. Mykologie, v. 3, pp. 132-181, m. 3 Taf.
Winstow, C.-E. A. 1912. The classification of thestrep-

tococci by their action upon carbohydrates; etc. In:
Journ. Infect. Diseases, v. 10, No. 3, pp. 285-293.
—and Roczrs-Winstow, A. 1908. The systematic
relationship of the Coccaceae. With a discussion of
the principles of bacterial classification. 300 pp.,
Ipl., 3 figs., and 14 tables. New York and London.
Winer, G. 1884. Die Pilze. In: Rabenhorst’s Kryp-
togamenflora von Deutschland, etc. 2. Aufl. v.1,
vin und 924 pp. Leipzig (spec. pp. 33-67).
Wierns, M. 1908: Uber die Much’sche granulare Form
des Tuberkulogevirus. In: Minchn. med. Wo-
chenschr, v. 55, No. 32, pp. 1687-1690.
Wrapiuirorr, A. 1903a, Rotz. In: Kolle-Wassermann’s
Handb. d. pathog. Mikroorg., v. 2, pp. 707-753.
1903b. Riickfallfieber. In: Kolle-Wasser-
mann’s Handb. d. pathog. Mikoorg., v. 3, pp. 75-
101.
Worpacu, 8. B. 1912, The filterable viruses. A sum-
mary. In: Journ. Med. Res., v. 27, No. 1, pp. 1-25.

 

 

 

 

 

 

 
246

. Woxsdén, §.B. 1914. The distribution and morphology
“of Spirochaeta Duttoni, etc. In: Journ. Med. Res.,
v. 30, No. 1, pp. 37-48, with 3 pls.
1915. On the filterability and biology of Spiro-
chaetes. In: Amer. Journ. Tropic. Diseases, v. 2,
No. 8, pp. 494-505, with 2 pls.
———and Bineer, C.A.L. 1914. The cultivation of a

 

free living filterable spirochaete (Spirochaeta elusa, |

new species). Prelim. report: In: Journ. Med.

Res., v. 30, No. 1, pp. 9-22, with 3 pls.

and Ernst, H. C. 1903. Observations on the

morphology of B. tuberculosis from human and

bovine sources. In: Journ. Med. Res., v. 10, No. 3,

pp. 315-330, with 9 pls.

and Honey, J. A. 1914e. A critical review of

the bacteriology of human and rat leprosy. In:

Journ. Med. Res., v. 29, No. 3, pp. 367-423.

19146. The diphtheroid bacillus from
leprosy lesions. In: Journ. Med. Res., v.30, No. 1,
pp. 1-8, with 2 pls.

Wour, F. 1909. Uber Modifikationen und experimentell
ausgeliste Mutationen von Bacillus prodigiosus und
anderen Schizophyten. In: Zeitschr. f. indukt.
Abstamm, u. Vererbungsl., v. 2, pp. 90-132.

Worrr, A. 1908. Zur Kenntnis der Verinderungen in
der Bakterienflora der frischen Milch, etc. In:
Centrabl. f. Bact. II. Abt. v. 20, No. 18/20, pp.
545-563; No. 21/23, pp. 651-675; No. 24/25, pp. 737-
780.

Wotrr, E.
gen fiir die Systematik der Bakterien.
Wiirzburg. 19 pp.

Wourr, M. 1880. Zur Bakterienlehre bei accidentellen
Wundkrankheiten. In: Virchow’s Arch, f. pathol.
Anat., v.81, No. 2, pp. 193-277; No. 3, pp. 385-420.

Wotoscuin, A. D. 1918. Zur Morphologie und Biologie

 

 

 

1898. Uber die Bedeutung der Verzweigun-
Diss. med.

des Milzbrandbacillus, etc. In: Centralbl.f. Bact.

I. Abt.
Taf.

Woopneap, G. 8. 1891. Bacteria and their products.
xu and 459 pp., with 20 illustr. London.

Woronin. 1866. Uber die der Schwarzerle (Alnus gluti-
nosa) und der gew6hnlichen Garten-Lupine (Lu-
pinus mutabilis) auftretenden Wurzelanschwel-
lungen. In: Mém. Acad. St. Pétersbourg, 7me
sér., v. 10, No. 6, 13 pp., av. 2. pls.

Youne, M.C.W. 1914. Variability among bacteria and

; its bearing on diagnosis. In: Brit. Med. Journ.,
1914, II, No. 2808, pp. 710-711.

ZaaaRi, G. ~ 1890.
“‘zoogloeica” oder Pseudotuberculose. In: Fort-
schr. d. Med., v. 8, No. 15, pp. 569-578; No. 16,
pp. 629-637. ke

Zaun, F. W. 1884. Untersuchungen iiber das Vorkom-
men von Fiulniskeimen im Blut gesunder Tiere.

Orig. v. 72, No. 4/5, pp. 312-327, m.1

In: Virchow’s Arch. f. pathol. Anat., v. 95, No. 3,

pp. 401-407.

ZaRniKo, C. 1889. Zur Kenntnis des Diphtheriebacillus.
In: Centralbl. f. Bact., v. 6, No. 6, pp. 153-162;
No. 7, pp. 177-182; ‘No. 8/9, pp. 224-298,

Zurpier, A. 1896. Uber eine Essigsiure bildende Ter-
mobakterie. In: Centralbl. f. Bact.
No. 23/24; pp. 729-739.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

 

Uber die sogenannte Tuberculosis |

II. Abt. v. 2, |

 

[Vou XVI.

ZemueR, A. 1897. Bemerkung zu der Arbeit von Dr. W.
Henneberg, etc. In: Centralbl. f. Bact. II. Abt.
v. 8, No. 15/16, p. 398.

Zerrnow, E, 1891. Wher den Bau der Bakterien. In:
Centralbl. f. Bact., v. 10, No. 21, pp. 689-694, m.
1 Taf.

1896. Bilder von Spirillum Undula majus bei
freiwilligem Absterben. In: Centralbl. f. Bact.
I. Abt. v.19, No. 6/7, pp. 177-180, m. 1 Taf.

——. 1897. Uber den Bau der grossen Spirillen, In:
Zeitschr. f. Hyg., v. 24, pp. 72-92, m. 2 Taf.

——. 1899. Romanoswki’s Farbung bei Bakterien.
In: Zeitschr. f. Hyg., v. 30, pp. 1-18, m. 1 Taf.

——. 1906. Farbung und Teilung bei Spirochaeten.
In: Zeitschr. f. Hyg., v. 52, No. 3, pp. 485-494, m.
1 Taf.

ZIEGLER, E. 1885. The Schizomycetes or Bacteria.
Chap. XXX in “Textbook of Path. Anatomy,’
etc., transl. by D. MacAlister. 2d ed. v. 1, pp.
262-305. ‘ :

ZmrueR, Fr. 1899. Uber die Beziehungen des Bacillus
implexus Zimmermann zum B. subtilis Cohn. In:
Arch. f. Hyg., v. 34, No. 3, pp. 192-197.

Zxes, H. 1916. Uber abnorme Kolonienbildungen bei
Hefen und Bakterien. In: Centralbl. Bakt. II.
Abt. Vol. 46, No. 1/5, pp. 1-4.

ZereL,H. 1911. Beitrige zur Morphologie und Biologie
der Knéllchenbakterien. In: Centralbl. f. Bact.

ITI. Abt. v. 32, No. 3/5, pp. 97-187.

Zuatocororr, 8. J. 1904. Zur Morphologie und Bio-
logie des Mikroben der Bubonenpest, etc. In:
Centralbl. f. Bact. I. Abt. Orig. v. 37, No. 3,
pp. 345-358; No. 4, pp. 513-526; No. 5, pp. 654-666.

Zorr, W. 1879. Entwicklungsgeschichtliche Untersu-
chungen tiber Crenothrix polyspora, etc. 21 pp.;
m.3Taf. Berlin.

1881. Uber den genetischen Zusammenhang

von Spaltpilzformen. In: Monatsber. d. Akad.

Berlin, 1881, pp. 277-284, m. 1 Taf.

1882. Zur Morphologie der Spaltpflanzen. v1

und 74 pp.,m. 7 Taf. Leipzig.

1883-1885. Die Spaltpilze. 1. Aufl. (1883).

vi und 100 pp., m. 34 Text-Abb. 3. Aufl. (1885).

vi und 127 pp., m. 41 Abb. im Text. Breslau.

1892. Zur Kenntnis der Organismen des ameri-
kanischen Baumwollsaatmehles. In: Zopf’s Beitr.
z. Physiol. u. Morphol. niederer Organismen, Heft 1,
pp. 57-97, m. 1 Taf.

———. 1895. Zur Kenntnis des regressiven Entwick-
lungsganges der Beggiatoa nebst einer Kritik der
Winogradsky’schen Auffassung, etc. In: Zopf’s
Beitr. z. Physiol. u. Morphol. d. nied. Organismen;
v. 5, p. 37.

Zoxar, H. 1896. Uber den Bau der Cyanophyceen und
Bacterien, etc. In: Ber. d. Deutschen Bot. Ges.,
v. 14, No. 9, pp. 331-339.

1897. Uber die Myxobacterien. In: Ber. d.
Deutschen Bot, Ges., v. 15, No. 10, pp. 542-552,
m. 1 Taf.

Zupnrx, L. 1897. Uber Variabilitat der Diphtherie-
bacillen. In: Berliner klin. Wochenschr., v. 34,
No. 50, pp. 1085-1087.

 

 

 

 

 
SUBJECT INDEX.

{Page numbers marked by * refer to illustrated discussions. Abbreviations following species names refer to chapters
on morphology (M), reproduction (R), symplasm (8), conjunction (C).]

Page.

Abnormal cell forms....................202ee eee 24-29
abortus bacillus....... CL guadteaneepeteecseteteees 58, 76
accumulation experiments.-..........-..-.------- 40
acetic acid bactéria.......-.- M *58, R *108, 125, 155, 201
acid-fastness, variability. ............-..2---. 84-87, 104
Actinobacillus...............20202 0202 eee eee 61
Actinobacter polymorphus. ............--------- 49, 153
™ ACUNOCOCCUB 6). ov eniedsoecbies yal So maascieaens ¥ « 44
Actinomyces Monosporus.............-----22+-- 102, 133
thermophilus: ............-.......-- 142

VIGIACOUB soia)s22 tiduiee ea Sinnaes sale cen 142
Actinomycetes, classification.................. 21-23, 75
: morphology......-.-.---....-64. *86-87
reproduction........-.--- 93, 98, 102, 120,

125, 130, 141, 142, 161
symplasm...............--- 174, *176, 194

AGhNOPDTY 6 ssa tsjonlsste eceteses Sethi o's's siteiclgiemiewieis *168
adeiptatittig \Genecaees seein ceaee 29-33, 188
MOTAULON oo sioosieiuls aie icisinis + gpleing ee tubisn so sionieareiees 206, 209
POLOBOMEK. 522 anes sf awake nedodas 2 aangenniene 113
agglutination..............02.002 eee cece ee ence ee 182
alnus nodules..........-...-2002 0c e eee ee eens 191, *192
Altmann’s granules..........-..-- See dce eashesers 112, 150
American authors on bacterial morphology......... 15-17
Amoebobacter.....-......-2-2--2+-5 eden SesneGireth 181
amphimixis...........2...-2000eeceeeceeeceeeees 199
AmylGbacter suisse ieane <2 veaseeteiwll oesedaceuls *66, 157
amyloid: 25 ss.c¢scccueeiscssawtamenees stances 113, 179
anamorph Osis. 2 1s dcc dec e Suen cine: oe eGasaaeesdts 149
ONAplAsMS wo oe eross cecewses onesceed sane’ 120, 150
Antherididy se sucess vsicre bain oaisteiaseniae ele iadon eg 202
Aphanozoum coryz@.......-..2-- 22 -e ee ee eee eee 148
sph li Vitus ceca souk eae Hae 146
apical growth of bacteria.........-.---.----.--+5- 23, 46
artefacts.........-.---.0000eeeeecceeeeeeeees 116,117, 211
arthrospores.........-.------2-2--+- Gyalgisioieteiie 95, 122, *123
formation...........----------- 111, 141, 208
germination.......-..-..------------- 142
resistance.......-.2022.2- eee ee elec 114, 142
Ascobacillus........-..00c-2e scence cece eee ee ee eens 181
Ascobacteria..........00---eeee eee ee cece eee Masene 18
Ascobacterium luteum.............---..------- *100, 181
AACOCOCCUS. .....02.0---e ence cence ene e cee senee 181, *190
Asterococcus mycoides.........--------+-+--eeee- 144
Austrian authors on bacterial morphology-......-.-. 10-14
PULOBAMY ... 2-22 e wee ec c eee e ewer eee e eens 202
Gitta Pei dos cas nanccenwagameey daeeneee ss Foe? 178, 189
AULLOMIKIB® owns ve sciwia lela weisitie'e a aieraninwneiaiee ts tex 199
Azotobacter.......--- M *36, *49, *65; R *106, *108, *127,

*128, 143, 157; 8 *178, 193; C *197, 203
relation to Actinomyces.........--.-- «65

 

Page.
AZYVPOBPOLES os ona; swiosiew'e ease eeeakeseake Weds 203
Babes’-orantil es ses cece se e's-es acss yacmixieg SOL ecieee 11
bacilli, relation to actinomycetes..............- *102, 162
B.=Bacillus and Bacterium.
BS DOrtuse ss cits ection teint ites adden aes 53, 76
BvC@tls.sc2ee5 a staanccewenk ess 44 eee ee se ecede *58
B. acidi lactici..........2000ececececeeececeeeee 137, 153
B. acidi propionicib............. pSzaasuRtceds 51
B. acidi propionicic......-.......---.22-20005 *177, 192
B. acidophilus..............0..0200.ece cece eeeeee 50, 139
Be Meld alumi is te reise y os eaves Pede rack *135
B, actinoides..........0.0..0. ccc e ence cee ee cece 61
B. eerobicus ramificatus.............2.-0eeeeee eens 51
B. erogenes. (See B. lactis aerogenes.)
B allantoid 682s crest G ce hechie'e's chrasaaeselermters 99, 130
BealVeli ewes pees sewdsaeeeehde sic cee seeene se’ 66
B, Amylobactetsc_. 2... 2:.cc2- denis oe a eeaeieeeccces *66, 157
B. antityphosum............2.2.2.2022.220200e- 148, *177
B. anthracis.............2-2.-2.-- M *63, R *92, *94, *99,
125, 130, 187, 142, *155; S 171,193; C 200
B. arborescens SeisigessVenenedess ys aceabiereee des 103
By asberosporuse sss ccsnestaceas's via sects nceeicee 140
B. Azotobacter. (See Azotobacter.)
Be Baeearnily. ciccese cs ostecencaeanes geipareweewls 65
B. Berestnewi. ..-..-.-.----0-eesce cence ccccocece 55
DB. Demensis fc 2h i's s''nistoud to deniaguss esau Seas 156
By bifidusicscecece se ssincsaued ais boas gctieekaicienis M *50;
R 103, 106, 152; S 192; C 201
B. bruneum: hse ccues ss ce nen lesaee M *49, 60; R *155
DB. Dutschlits:<.2cc.¢2<biostetetesee ne aielaeee 202
B. bulgaricus.........-..----.-- M 51; R *103, *108, *153
By capswlatuss.cccc 2s o Maeeiekecieew ks + ouside este *99, 103
B. Chauvoei.......-. M *67; R *95, *99, *104, 157; C *197
B. cholerze gallinarum................. M *53; R 104, *153
B. CONACTONS os 0s cecineree ee se sateen ns M 62; R *123, 156
B. coli...-.--.------ M *156, *157; R *101, 108, *108, *124
*128, 130, 187, 139, 153; S *177, *192; C 200, *202
B. coli albido-liquefaciens...........2.--.....0.008 59
B. coli luteo-liquefaciens.............2..2..2..005- 59
BL Corallimis..2ccijce sean canenswdscca at wienreeoeens 59
B. cyanogenes...........-- M 60; R *105, 137, 155; S *175
By CYUMANCUS. foe cices Sse ee Sagaaseace sas Soacees ‘156
By-GaMCUs yas oeauae evs eeNGeee bass scsi cece 64
By, delionsis: scccsiicc Gee eeacinesedcaie c2.canunarie te 138
B, diphtherie...........2..........02. M *77-79; R 104,
105, 137, 161; S 169, *177, 194; C 200
B. diphtheroides....... M *80; R 98, 102, 137, 161; S *179
B. dysenteriz....... ated aioe es caicarperece M 57; R *101, 154
B. Ellenbachensis...................2---2-2-22008 156
B. enteritidis............-2-2..2..4..- *101, 107, 154, #192
Bi. COZY MICUS ssi oicels habe aennes ceanueeecade *81
248

 

Page.
B. erysipelatos suum.............. M *52, *53; 8 *176, 192
B. erysipeloides.........-......-.. M *52, *53; S *176, 192
B. erythrogenes.........-.2-2..020eeeeecee eee e eee 59
B. gastrophilus..........2220.....00eeeee sees cess 51
Behe xordes,. siccjcciee Ginemcentdndeaaaaeocaseaet 180
By. Werbicola. 2. o.2.é0:5.6 see.c.ciare aie desea enneeaueae 137, 155
Be hitsubiis:2 «sans dasasaonovnrs sagnannannaneans 62
By HOM GW sos cis enna ie swe nunnnriresamacasaneenaeed 80, 194
B. Hofmanni.............2.2----2-2e- eee ee SS asap cia 80
B. implexus lateral cree chcrah echaheh cb vate Sie ooe leach net cr nasioiars 62
BAS FUG sss cictsie w cictevatccenaisrescisrsrerstancisiraestahcis adatclarsionede 138
Bu influenzae. «0:0. . sce cic eos oes seni seesectgic cisterns *54
By ANVOMGUS sce cccig cs pews secede cca ceetioneceeg *55
B opens sscc ssn eis sb eles s See vee cele e's ssses *98
B. lactis aerogenes........-...--2.-------+-- *99, 139, 153
B. leguminiperdus.........-..2.2---2-02-eee eee eee 139
Ute pieiseaaee biatch M *81-83; R 95, 102, 105
*107, 133, 160; S 171, *173, 176, 194
B. malabarensis..........---------- M *64; S *178, 193
B. mallei.............. M76, *77; B98, 102, 126, 133, 160
B. Megaterium.........2-.22.--22-2005 M 64; R 125, 156
B. melitensis..............-. spin fara raraiarenereraiierRiato's *44
B. merismopedioides................-2-2--20--0-- *50
cB STUCOSUBS hse vce sore eas ciepaeiccitrorseciapeeaia iia ince wiererore 54
Be murals 20. cc csexce ese nsacesanceunaiacesanmees *97
B. murisepticus..........-....-..- M *52, *53; R 104
B. murisepticus pleomorphus. ......-....2.2....-. 53
Bi MY COGS i.5 cence tied scam taewe amen mmneniemnne *62, 102
Bi Mecrose0s suet icon sass saeeenCawav esc eeawsnSee *76
B26 1.6 wre sscisrsiomacrivre sna sande none a mncvaw acorniele ey 64
B, no. 41, yellow... M 64, R *124, 157; S *174, #178, *193
B. ochraceus........2.cceceeecceecccceeccececees *101
B. oedematis maligni.............22-.020eeeeeeee *67, *97
B. oligocarbophilus......--....---++----++e-e+0+- 61
B. On aa MCB saci isicieccicizrciate sareis aroueelsieie eels edacens’s 64, 156
Bu ORY Mang ecaey ss seeds: secesieaauscsaasagayen *103
B. paratyphosus..........22.222-0e eee eee *107, 154, *192
Bis Pasteur pas aee latratela a serene scree mca tects 64
By pestisss o.0s20c see ee scene ooeee M *53, *54; R *99, *101,
*103, *111, 153; S *176, 192; © 200
B. phlegmasice uberis...............2...22206- #101, 153
B. phosphoreum..........22-- 2.0 eee eee e eee e eee 44, 58
By PISCICldUs <0. ssa sete nrnannarenn asain eerieraniareies 139, 152
B. pneumonie.... M *55; R 103, *108, 153; § 176; C 200
B. prodigiosus......-...---+++-+--+eeeeeeeeeee eee 59
B. proteus. (See Proteus.)
B. proteus denitrificans. ...-...-.-.-----+-+-.05:. 50
B. pseudopestis murium...............-.2.....0-- 53
B. pseudotuberculosis..........-..2-2----e0000- *86, 200:
B. pseudotuberculosis rodentium............ safes 53, 153
By PuUduss.cceecs ys 2edhsseaecs026 2504 eke eee 60

 

SUBJECT INDEX.

 

Be putridintes 6.2% cuts cisiaicisieie niece's wae ets bewee 93
B. putrificus...... Gepcemeecenetedddenes demtinemees 67
B. pyocyaneus........-..------ M 60; R *99, 104; C 200
B. pyogenes ramos ...... 2... e ee eee cece ener ecece 50
B: radicicola)scxseccecenviedewsadsaksise 622 'seies M 58, *59;
R *124, *125, *132, 154; S *174, 180, *191
B. radiobacter........-.--+ M 59; R154; S 186; C *200
BistOPens: ss mescguce dene ches cehesemeisbis aeisiemiare’ *75
Be Phinitis: sjcevesGcuad ved esscacedesesa ances 74
B. rosaceus metalloides...........-2+---2-0-0+0-8 59, 200
B PUD OSCONS ss cocci ase ccaeeaseiiedene oe scicioeces *170
Be Rublacea rim sis. aaccretejasistesiva see v's ad sievsigis.eieceis 59, 201
BTU Gis sesh ee a eacetieteettdet ls eee are eae enn es 103
Be TUMing FUssencncwcnce nerntuiesancexdx sii es dace *123, 156
B. scarlatine........-----...-- M 48; R 146; S *179, 184
B. septicaemiz heemorrhagice............--.-- —*53, *191
Bespordnemias. ds ja.0cacenweeeeeataes'sees sk euean ~ 202
Beeubtiliss ccc... 6. bdsdeese seer saneceecee sic M *64;
R *103, *106, *132, *156; S *178, 193; C *197, 200

B. syncyaneum. (See B. cyanogenes.)
B. synxanthum. ......... 220-2242 eee ee eee eee ene 59
Briermogcors. sie ceccnedes 2oe seen teeee cateua *92, 93
B. termo syn. fluorescens........22.-0---seeseceee ‘92
Bvtetapleccchcaccstece decease cates camemnes focuses 57
B. tuberculosis... M *83-85; R 95, 97, *98, *102, *104, 105,
*107, 133, 137, 159, 160; S 169, 177, 194; © 201
Bitubilex. co shoccc ocak ies rotons tetera tewayang 192
B tumefaciens: 2 seceqneey ss eggs eo aciosieemesess a)
Bs HUMeCSCONS 3... ee ceuieecdece ddieee ieamgereenie *66, 156
B. typhosus...............-. M “57; R 95, *101, 103, 104,
¥*108, 133, 154; S *177, *192; C *202
B. variabilis. ......2002cceeececeeececccccceees *80, 161
By ventriculusiscccesscseceaeease ives sec cwacacews 138
B. vernicosum.............0--22.ee scenes dnie eras 141
By VIOINCCUB sc oe. c ee wiciiewi eae ee sees eee dbemimsie ee 60
B. violarius acetonicus............---..--eeee08 187, 155
B. visco-fucatum...........20-es.seeeeereeeeoeeee ~ 60
By R6TOBIS: wince visas care caaiecerere M 80; R 97,161; S 173, 194
Be LOPal. acminiveeadandeccsassacs M 50; R 187, 141, 152
bacterioblasts.........-....---2-2------.2---+-- 150, 166
DECLELLOC YRS ccc Sees ecards ee 181, 186
Bacterioidomonas sporifera..........-..... ravetsve chins *96
bacteriolysis. .. 2.222.222... cece cece cee eae e eee 115, 178
bacteriophytom.......... 2.222222. -ceeeeesnnes 166, 172
Dacteroidsi.dnckve cn sivemccwmiaiiee’sse ve 25, 58, 134, 187
Beggiatoa coc. eu esse sesaecesss M 88; R *94, 130; S 195
systematic position. ................... 76
Dioblasts-ccs sewsstedigee ress as es gees ctaciaeaceuaicio 188
Dlaistemancic vee cso voces ee ela asamnacane nee seule 179
DISS ast iccakincignncwiten as ed ees anonetewsocuses 90, 119
Blastomycetes, relation to bacteria............. ede 2B
blood granules...............sseeeeesee neers Le. . 94,118
Boas-Oppler bacillus..................20224 pisietets 50, 152
bottle bacillus of Unna................. 0220.00 eee 22
branching of bacteria. ..............2.0..22.. 21~23, 128
British authors on bacterial morphology . te satenwne 9540
budding of bacteria... 22.2222 ......... 2.00008 se. 22, 128
butyric acid bacilli....... § bo ce pannaeesmmied Se eee 66
cell formation.......... sipeeathigissciaicbatctaes oaseas 167, 185, 188
cell granules Nea Iara AA ctealosdane 148-150
COU STOW sciscvcose ietee 12s cseanmaersekeor eee ys 180
CélancluBiona sey sau’ cstv se canprdoa ice iccniec « 144
SUBJECT INDEX.

Page.
chlamydospores.............0.-eeceeeeeeceees 121, *123
chlamydozoa...........----2--0- 22sec eee ee eeee 144, 150
chondriosomes....-......2..-0-2--0e-eeee eee eeeee 149
chromatine granules...............0.2-02 202 eee 109-111
Chromatium...........2...2-2.2.220-022222e2-2- 65, *197
ChrOMIGIA ie wecsseveanes saasevewseess Seer en cuees 109
chtonoblast........-. 20.0. cence eee n eee e cece nee 150
Cladothrix......... M *87, 88; R 93, *180, 162; S 186, *194
systematic position....................- 76
asteroides. i ssccvexddecsesssecuas os cae 86
dichotoma........-. M *87, 88; R *130; 8 *194
OdOTMORM Esai. essere salen seu see veieeieee eee 87
SLELCOLTODR.« © a oi0is nee scien ese caanee eens 62,74
thermophila soc necsvessveseceveeeses can 86
classification of bacteria............-.----- 21-24, 42, 181
Clathrocystis. i vcccere soascencceenceds Saeceeaen 181
Clonothra xc 52 oct tnisonsinaacunsioayieeis'srewa set 88, 162
Clostridium butyricum.............-....20.00005 *66
CISPOFUNi. 5:2). cis eieeaeisigs ec owcee Sack 138
+ Pastorianum...... - M *66; R *104, 157
POLVINY RAieiswe cnn baicceae saaheae See *66
club-shaped cells...........22...2200-- 102, 104, 121, 125
COCCI jesse Week vets sere M *43-49; R *151-152; 8 *190
Coccobacillus foetidus ozene...-.-....--. 2.20.00 55
mycoides peripneumoniz........... 145
Coccobacteria septica................. M10; R 92; S$ 170
COCCO-BONIM A. ss cieeecze ooo abenwlaiede nied srmeees 94
CoCCOt brik ac ieciecccwcwwinee ceieisin scanty eieleldaye *82, 84
Congo-red method......... de etc. ayapalutsl aid eibiaee = creda eyes 205
GANT TID: sc covecinmawensmawdaon s wowssevenady av HIOL, 114, 100
Conjugation...........0. eee ee eee eee eee 198
conjunction..........0-0. 00 e eens cece eee e ee eee *197, 212
conjunctivitis................ JAtatsecense ites Saas 145
CODSLATICYisis's'e's is. cteieiale cuiomiene b o.crerasiaehatG > sectoral 12
(See also monomorphism. ‘
contamination................-.. 20, 39, 118, *208, 210, 211
continuous observation...............0e2e.eeeeeee 38, 39
copulation...........0.02.-eecceeeeeeceeees *197, 199, 202
Corynebacteria..... M *76-*81; R *105, *128, 161; 8 *179
CORY 28) ots caced chudieee ad @uck toe eauaneaee 148, 184
Crenothrix........ M 88; R *92-*94, *128, 130, 162; S 186
systematic position.............----...+- 76
Cristispira tapetos..........--.---.0----eeee cence *123
CRY PLOCOCCUB apis stein c co ninesie hanes oo gamene Seieenals 179
Chy PtO PlaSM oasieniec e's acids asinine so wnwisieiteeiewieeeret 179
cultural conditions...............----2-202eeeee eee 40
CYSUOSPONCS oe esisc esos s oyeveeeeccses cecneeseeceas 138
CYSUS. oo eeeeeeee cece eee 95, *123, *142, *143, 152, *181, 186
eytoblasbs: voce cam.ace sued los eeacve noes eoalased 150
cytoblastema.............0- 0020 c eee cee eee eeeeee 179
Cytorrhyctes APHtharuM so) cciojosc daiersssis cise slain e semis 146
Wiis 52 vensdos gabeeeenne se Shaeceeceses 146
scarlatine!. sso. 32 sescis soseeeeeceek 146
WACCING 616s ccinesismece ceanaesersice 144
Danish authors on bacterial morphology.......-... 15
dark-field method..........---.-----------e+0- esx 209
Dermophyton Malassez...............20--00ee eee 22
degeneration.......--.....--.2-----2-2-0-5 24-29, 34, 115
diphtheria group. ....-.-----.--.------+-+-- M *77-*81;

R 102, *125, 131, 161; § *170, *177, *179, 194
diphtheroids, systematic position............. 62, *80, #81
drop plate method..........-.......--2----205- 205, 206

 

249

Page.
drying effect on reproductive organs. ............ 99, 101
duration of experiments..............-2---02- 24, 39, 40
Dutch authors on bacterial morphology..........- 15
Enchelys ovata: severe. xcs vv senddasecedes ay oeees *167
endospore conjunction..................-- .... *197, 199
CehniNIONs 08s sas neeecccwatstetel nets ek 122
formation............ 109, 138, 139, 202, 204, 208
beginning. ............. 98, *99, 137
experimental induction...... 106,
139, 140, 208, 210
LOSS Ej tosis /sttetddandaeecsis c= 98, 139, 210
germination:.<.-sscssrs se sec eisseeee eee 140
symplastic stage............-----...-4 182, 204
Enterococcus. .......--------- M 47; R 104, 124, 126, 152
eosinophilic granules......-..-.--------2+--2eeeeee 144
Ernst's granules.........-.2-0-0-.seee cee eee eeenee 111
HErysipelothrix.. 25 cscaciacseuditac a6 pc asoaane *52, *176
Eubacillus multisporus.........-.--.-+-2-e0200-- *102
OXOSPOLES ayes cicsee ew cited ae ee heed 122, *136-137
fat, differentiation.............-....-000-- eee eeee 112
fatty degeneration..............-.--2---------0-- 114
fertUizAtiOU <n. ceee iss caisdsesadue 2.254 wenaesiece 198, 199
PIG. cose oeiniciceowen a sind nneeadiancees tceames wesc *175
-filterable gonidia. ........2.....2202000000- 143, 184, 210
fil€erable Vitae aiccicce oe eaicepiaces eis otis enisisine -. 143-148
flagellation, inconstancy......-.-..-..------+-- 56, 62, 66
fragmentation .<<.0<5sjccssnvi cress eseessyeeeoeerse 116
French authors on bacterial morphology...-....-. 7,8
fusiform bacilli.............. M 74; R *106, 127, 133, 159
relation to spirocheets......-......- 74
gametes...-.....----- oleatasl hada coker erst tact t Anais! 203
Geissler’s chamber..............-2-..2---2-2-+---- 206
POM MUST 22 cde. wcs setemcamee doce es coed adamee 188
German authors on n bacterial morphology......--.. 10-15
germination, irregularities............-..---.-...- 118
Berny PlasM 2 vaicicewggs oe ose ciaiceteaeate yee searteels 188
glanders bacillus. (See B. mallei.)
ploeal Massesicc5-crscecuee sss setae deel secs 176
Glycerin-Aethylbacterie bia Closers: ci diefa'd Slyayoareibiva Maelo *62
SIV CORED. o es sccsceauiuneses deneannaredkeneese oe 113
gonidangia, definition... ..........-------2.200-- 121
formation.......-.-.-- *124-127, 202, 203, 208
spore formation...........-..---.+.... *138
symplastic stage.........-.--.---..+-- 184
gonidia, definition...............-..2-.22-2225-- 119-121
{OTMAON.Ss5.ckeen eta cece ees eee *123-125
BELMINAON 2. e-ccsies ses eceseeeeees *128, *130
MOWMY sais s Sosa os vee deel wcis etaerara 127
multiplication......-...--...22.-2.2.--- 129, 130
POSIBTANCO. oc cica cca entice cwiete 130, 131, 209
MCOiacrasccwmemmvs \ieicuivacetan alas ses Saves 143
staining reactions. ...................- 127, 209
symplastic stage......-.2..2---2..0-220 0s 183
Genococcus.......-.-. M *45; R 103, 126,151; S 177,190
granular decomposition..............-.-. 104, 114-116, 178
prass baCUl ce sola cya chetele sac taagcuchisiss sees vanes 86
Halibacterium pellucidum...............-2--+--+- 72
polymorphum......-..----.------ 72
hanging agar block. .........-...---2--------+0+- 205
haNPING Crop sciwieestc si qels Ss os Aelitee aan essary 205
hematocystss ie steer gdey voy es eceeeewws see he *149
H@MOCOD A s63. ses see degmce ss Hes eeeeeueelees eas 118
250

Page.
hheteroCystas gin osgsisegssd ais eck chee 128, 148
heterogenesis........-----------e-eeeeceees 149, 168, 172
hog cholera. s snncwsneeaudindwetemeceireasets see 145, 184
hhyaliny jolc.si laden eteeter ee datiecceeeezeesx 179
Hyphomicrobium. ...-....-......222 2 cee eee eeeee 60
India-ink method. ........2. 222-22 eee cece eee 205
infective granules. ............-....006 106, 113, 118, 121
involution forms... ........200222 cece eee 24-29, 114, 187 |
iron bacteria... 2.2.0.2... eee ee cece ence eee e eee 186
Italian authors on bacterial morphology......-..-. 14
Jodococcus vaginatus....-.-.-.--------2-0-ee eee #98 -
Kapselbactllus:ciecsceecesisaecieee ved dsdes seed 55
Kérnchenbacillen............-...2.-2-2e-eee eens 51
Kokkobacillus zymogenes.....-.-...------------ 43, 44
lactobacilli........... M 50,51; RB *108, 152, *153; © 201
systematic position. ..........-...-..- 50, 51
lag insprowth sjas<2ccc20cs occ dee se tonece semis es 189
leprosy. (See B. lepre.)
Leptothtix... M 51,87; R 91,129, 162; § 168,186; C 201
Leuconostoc hominis.................2--+0-+-0- *47, 152
Lagerheimi........--..--.---.-2----- 46
mesenterioides.........----------- 142, 152
life cycles of bacteria. ........02...--0-2.2 ee eee 33-37
lithium salts, effect on morphology. mideelsision ss ees *99, 207
longitudinal division of bacilli. ...........--.--.. 50
spirochets........-------- 68
streptococci.......... Sees 46
Vyasa socce sss sesss suauseeduesseukedeyuuduegi sees 145
MACTOCYStS ..30 usa seeds sess eseace emesis 123, *181, 186
Macropamete.... 2c .s.ccs.ce cee seccceeeeeesees 203
macrogonidia....-.-..----------2-22 22 lene 92, 119, 120
macrogonococci..........- atucescecoseenates eset 145
macroplast..<2)iccescie es ede eed escieeces ees 166, *170
MACTOBPOLES 2:2 '« siscisie og voice esess sek eeesseswee cee *175
MArgiNal Points... -i.csic sc aeseiersicie wesieusiee coe sess 120
MastZellentscviccandansapepeainiaiesine eee ciaedeee.s 111, 118
PARUIRE cocedeceatacs nies Rice eee ee es saaee 166, 169, 177
meningitis, aneerobic bacilli...............-.-- *52, *105
filterable virus. ...-..---------------- 148
Meningococci.........-------- M 45; R 108, 126, 143, 151
Metabacterium polyspora.........---.-.--------- *96
Metallacter Bacillus............-..-2.---2---2---- *62
metachromatic granules .........-..-..---------- 111
Microb laste. 5 Aesu nso lee tists eo g sts 104, 120, 152
micrococci, general character... ....-....---..+-- 43
relation to actinomycetes............- 43
SPCClOSecaeceeeessese estes tecdeseeess 11, 48
Micrococcus. Bascesceswsineseesecscdsecescsasesess 44
botryogenus..........---. M 44; § 181, *190
candicans. ..M *43; R *108, 151; 8 *175, *190
Oatarrbalises «.. a:sicicisie:sccisiwieieicseseseisiniisicie’s 148
Coralloides.....25 226265 cceced eee cela nes 45
COTODATIB so o2.5.2. nc pcnacodnecdyersirmeroec 45
CYANCUB 2. osc2.5 ced ca cies ve qQeeacanee 45
diphtheriticus............-.....5..-- 77
PA VUS seit raster sews eee tierwe *43, 151
gonorrhoex. (See Gonococcus.)
WteUsivensceciedssceieeksasessses *43 , *108
melitensis....0s52762sca0 oe sesaescacse / #44
meningitidis. (See Meningococci.).
MU DUS eo a, '2 5's aisle darciets bees crane 45
“ ochroleucus........-..0---0-eeeeene 142, 151

 

SUBJECT INDEX.

Micrococcus phosphoreus.........-----+-+2+---005 44
Poly puss ncevceic ict eewasbstineesien 45
pyogenes......-.....-.-. M 43; R 103, 151
rubefaciens. .........----.-------+ *43,151
Sormthalitisc ccs cecune ses acs enciccsiens 46
subnormalis........------.+-----+-- 44
PRU iiss. caeaceeiegestesstase 151
UPC. oS eheewe esse se saiiderescadee so 151
VACCIIR 66 0e bes ve ceeeceweneue ce eee 144
VORI C4 alls hanctane trea cue Ceeeies 45
viticulosus.......--.-------eeeeeeees 45

mnicrocysts, definition. ....-..------.----2------- *123

formatiOnsesa: osscsspesesereetaees *142, *143
fragmentation........-..-..-.2022-000- 152
germination......-.-....----.------ 148
resistance........-...----- eeeee die. 148, 152

THICTOBAM CUE i.e esc. nscietesnveresarereiaiore viataibe-waemasaaiaeecs 203

microgonidia..........---. ee eee eee ee eee 91, 92, 119, *148

micro-gonococcl. ...- 2.22. eee eee eee eee ee ee eee *145

TMICTO-OLG 18s oso oo san dca araseravereirere siete orerarareeceves 64, *66, 132

MictOplasts. 0:5. 6. a sciacieninwsstweviecees oo issues 188

INICKOSOMNES ops ie'acinterniaraiarga cee bwiceeweedemee 97, 109, 120

MICTOSPOLOS s x a:sisieraisisiwrsinieigisisrsicincin ss seein es 98, 139, 140

Microsporon septicum. . ogttne tuaeialete se ommaasaee 169

MICTOZVMAB ss s.3cc0e- oss eeainesinkslewieleis 95, 120, *148, *178

milk pranules...:sjc:0000se0eseednscdcneseeescees *172, 190

Monas lens ss <2 seescceasasareiewteedectakekeeenen *167

monomorphism...........---- Saito caked te 9, 11, 12, 15, 16

mother spores. .....-- poecrwerist ned Saeed Res eases 154

motility of gonidia..........--..-.... dissin tenes’ 19

regenerative bodies..........-...----.- 119
BY MPlABMN ss 2 A occsicieeeceadesieins o2 oie 183

mutation of bacteria.........-..---.-------eeeeee 29-33

Mycobacillus synovialis.................-2--- ‘.. 62, 162

Mycobacteria ...... M *81-*86; R 141, 159-162; S *194

classification. ..-.......2.22------- 21, 107

Mycobacterium hyalinum~.........-..-.-.-.--..- 86

MyxobaCtera:<... cose sioddssaedsewemieseunssis3 3 *180, *181

Myxococcus.......-..-- R *123, 141; S *180, *181; C 202

Myxomycetess csescssceceinesevceessccs's se sase 180, 181

necrosis bacillus. <2 s<issedscnsexeavssessaccaeees *76

Negri bodiesscs2..sdexce ets dicewicewse ca ececdacecs 145

Neurorrhyctes hydrophobie......--..-----+----- - 145

i IN dey dieietorantdclvre diclinisan gee es en ee Shanta 205
Nitrobacte?iccssosssececewseiczes ones cccecane *61, 155
’ oligotrophum........ psvdjaoie srsratcinitaee Seeeeee 61
polytrophum..........2.2-0---...0065 61

Nitromicrobium..........2.202 200-202 e eee eee ee 61

Nitrosobacterium.........22.2.-22-- cece cece e ee 61

Nitrosococcus..........------. cece cece cece *135, *193

Nitrosomonas. ......... M *61; R155; S *170, *181, *193

nodule bacteria. (See B. radicicola.)

Norwegian authors on bacterial morphology... ..... 15
nuclei in bacterial cells...................22.008 109, 204
nutrient solutions........2.... 00.0.0. 22 ce eee e ee 206
QOPONES -srseeeaeisw send abe eamiddnennes twaleniebinn 202
OOBPHGVES sivas d daisin sini sic seesynapteeeecanmetene’s 202
Oospora proteus. ....-. 2.2.22. 2. e cece e eee eee eee 87
PADQONS 2c s:ci- 2 laisse 5d ca nrnnveee scswceune see es 188
Pasteurella. ...........---..4- ses gaat eie's et day *53
Pasteuria ramosa.............. niiewccwewsee sss 188) S176
Ped ORAM. coca ss xeleesseushaccseudss sss ede 199
SUBJECT INDEX.

Page.
Penicillium glaucum. ..........-..0202eeeeeeeees *167
periodicity of bacterial growth............ 28, 33-39, 189
Peronospora Barcinonee. ...........-2-0-2-ee eens 23, 69
photobacteria........--.----- 2-2-1 eee eee ee eee eee 58
Photobacterium balticum..........-............-- 72
tuberosum... .............0200.. 72
Phragmidiothrix multiseptata. ...........-2....-. *94
phthisiogenic microbe. .....--.---.-----+-+++- *126, 194
physiological units............-.-..0.-2.-0..020e- 188
Planosarcina urew........20....00cceeeeeeeeeeees 49, 151
Plasmakupeln 25. ocis cece cajsc eee cee n ccs cones 93
plasMAdiUGl. .......ncenescenesene 166, 170, *177, 181, 210
plasmolysis. ...-----.-0. eee eee eee e eee eee eee 116, 117
plasmoptySis........----- 0-0 ee eee eee eee ee eee 116, *117
DIMRHIEL Ss si. cemes x behaneamauen Suemaseewos tutes 91
DUBRG TINGE ion! ainda sciee so eamacstenieiswe re Use 188
pleomorphism... ..............2.2- 00 ee ee 7-43, 187, 188
pleuropneumonia.............22-2-- 2222s eee eee 144
Pneumococcus..........-.-...---- M 45, 46; R *102, 152
Podophyra. .....-------+ 02-222 eee eee eee eee *168
polar granules..........-.-------------0-- 103, 110, *111
poliomyelitis. ..........-.-.------5-- *48, 146, *179, 191
Polyangium......0...0.... 0c cece cece cece ene en eee *181
polymorphism. (See pleomorphism.) -
post-fission movement. ..........-.....--20-22--- 200
Proteromonas Regnardi......-----....-.-------- 96, 133
Proteus.....2....20-- M 49; R *97, *100, *152; S 173, 193
MUOTESCONE. cose se emceddt seed Sv eeoiees 50
hominis capsulatus.......-.- M *55; R *97, *153
proteusartiger Luftbacillus. ..................-- *50, 141
protospores. . ...----.- 2-22 eee eee eee eee 104
PYOLO208 sic cia cies ceciscc oscars do au nusiene S *167, 181; C 198
pseudo-agglutination......-...-.---.-2----- ee eee 182
pseudo-diphtheria group. .M *80, *81; R *161; S *179, *194
pseudo-macrocysts........--.-.0--- 22 -e eee eee eee 187
Pseudomonas cerevisie......-.-------------- eee M 60;
R 104, 132, 155; 8 *176, *193; C 201
pseudo-nuclei................---0--2 eee e eee neers 109
pseudo-tuberculosis...............--------- *86, 172, 173
pseudo-vacuoles.........2.2--0--202 cece eee eee 109, 113
PADIOB sco cccauee sekeaan ae naneeaeeke s <'ssanaiy 145
reaction, influence on growth...........-.-.-.-- 206, 208
regeneration...........--.--- 34, 118, 171, *175, *185, *186
regenerative bodies, definition. ............-..... 121
formation..........-...-. *129-133,
*186, 190, *201-*203, 208
germination..........-...-.-- 134
motility........-.-.--2--..--- 133
multiplication.............. 134, 135
resistance............-.-- 136, *208
staining: ..25:<ses008.ssesesss 133
regenerative units. .....-.--..---0--02e eee e eee ee *185
reproductive disks.............2.00.eeeeeeeeeeeee *170
resistance of reproductive organs. .........-.-.--- 114,
118, 130, 131, 136, 143, *208
Rhabdochromatium. ..............0.2202e2 cece ee 65
thinoscleroma.........2....0-seececeeeeee ee eecs 170, 174
Rhizobium leguminosarum............------2---- 191
Binder pest ace nce's sicjeisicinis-wicieisieis weisigesiacisiwie se seni 144
ving formation.........0.0c0eece cece cece eee e eee 124, 203
rosette formation...............22-22-22005- 68, *186, 200

Russian authors on bacterial morphology......-.-. 15

 

251

Page.

Salpeterpilas cwicwmccis vss xaewes seea ds amaiied aeons 14,61
salt substrates, effect on morphology.......... *57, *99, 207
Saprospira grandis.........-.-----.---ceeeeeeeeee #123
Sarcina flava.........-..---- M *48; R151; 8 *178, *190
MObiis se ses ses awetddee sds sa seeeekanies 49

WA CO8A ieee seo a ota crabs bisicere Sw teas weees 190
TOTAQONA. 2.sis once bwicieten dees saiieaisie ees 49, 153
sarcinz, relation to cocci and bacilli. ...........-- 43, 48
Sarcinastrum Urospore. .......-.--2----- eee eee 49, 129
ACOTIAHNG) c\.oascwackes sew eeeieeeseee 5o5 ees 146, *179, 184
PCHIZONS: co sscmnineeed tect eidewess same sewiee ce 185
Schizosaccharomyces......-...-.++2+-eeeeeeceeeee 24
Schipeirinen a, aue iors asegisceuae vases 112, 154, 174
SclérOtia ceeemce wines sos exenas bkicues seine s ce 122, *187
secondary colonies......-...-.-.02.+-sseeeecoeee 34, 189
segmentation......---.---.----- eee eeeeee detneideies 116
sexual dimorphism.......-..-....---+eeceeee 68, 202, 203
sexual reproduction...........-2..0-seeeeeceeeeee 198
single-cell cultures.....-........200seeeeeeceenee 39, 205
species, definition. ..........-.-.---se-eeee 380, 31, 33, 42
spirilla........-.-..-- R *91, 94, *99, *107, 158; S *177, 193
Spirillum amyliferum. ...........2220-s0eee eee 94, 158
DACA VIE osc SiS aicini wicia aretaiele’e wisielsteieiere aise 73
endoparagogicum.....-...-.-.26+. *72, *96, 158

PYOLENESs = = case ecsedear we wececve cscs 73, 158
rubrum.... M *73; R96, *99, *107, 158; S *177

Serpenss . 2 cccccsseecmmsseecccetemece 72, 158
SPOT CCU Mas jansacik sees tieateesieiceced *101
$yPOSONUIs.. ceca nccaune secs ae 73, *158

Undue cis soitepeseiels'es'e's care M 72; R*99, 158
Spirobacillus Cienkowski............-.---eeeseeee *136
Spirocheta anodonte: -.... 22.2.2... 2 eee eee eee eee *106
balanitidis........2..0.0-cnecnscsecee 73

ANNUM toca iescaw dures deae tie sisisicte 159

DUO. ccd cic iedsiciaiels's sievsee,ns'pseieeiecs 159, 193

gallinarum. -........... M 73; R159; C *201
icterohzmorrhagica..... M *73; R *182, *143

TNUCOB Rs acs iejcissein ds sinje'e's 2 cieiesaig eco aSisinin'e 159
Obermeieri....M 73; R 94, 158; S 171, 193;

C 199, 200

Pallids.. s.c.osececiaices ese M *73; R 159; C 203
phagedenis.........-..--2.sesceeeeees 159

PICAHIS civic ie cs aictsicis djaisislsinaiesieesicie 158

SOMA LIGIS 55a ise'wis ccisicisrenewsteiewisies. 73

BUIS: ss Asasdiskesiewiss <sigaiiowa Souicice 78, 145

WOCCIND ssi vsssvvesteses tcdeedeeores. 159
spirochets..M *73; R94, *106, 107, 125, 183, 141, *143, 158;

8 193; C 199-*201
classification.............+2.---+---2-4 68, 107
spores. (See arthrospores, cysts, endospores, goni-

dia, regenerative bodies. )
sporoblasts....-..------- 2c seen eee eee e eee e ee 198, 203
100K as 138
sporogenous granules... ........-...2--2. esse eee 112
SPOVOMGS. sins ainieieins J owrciaieveis ateeaieicie's si e'gistars 109, 110, 120, 139
Sporonema gracile. ..........2. 222s eee e cence eens *91
SPOTOZOMES sviecien sou Soseerewm serve awe eeesaweees 203
SPOT UO <i ic.sseidseiecase-< atsievas Sasa bicteisie d.0 sarees eiewes aa 93, 202
stability of aberrant strains.................-2-0-- 29, 34
staining methods. .........2.-- 2-2-2. -eeeeeeeee 205, 209
staining reactions, inconstancy.............- 112, 182, 209

staphylococci. ............----. M43; R 103,151; 8 190
252

Page.
Staphylococcus aureus.......-...22-2005 M 48; R 108, 151
star formationisc:s-vseccceseseves emcee 129, *186, *200, 201
Streptobacillus pellagree.....-.........--2-+-4+-- 76, 161
streptococci... M 43, 45, *45; R 103, 104, *142, 152; S 190,
*191
relation to B. diphtheric group . .43, *46, *47,
3B y 77-81
Streptococcus aggregatus...........-----.-2--- ee 46
causing ‘‘Brustseuche” .......-....- 47
causing poliomyelitis...-..........-. *48
Choved ie nsecccerssveaves <'ssiesece oe 6 47
OGUL: secewieeseniveesseeicvsy cess *48
verysipelatos. ........2--.20-0---06- 47
lactis..... M 45; 47, *48; R 96, *108, 126,
¥152; S #172, *175, 187, *191
lanceolatus. (See Pneumococcus.)
pyogenes... M 45, *46, 47; R 103, *132, 152
PYTOPONUB i 2 os sete ses eeeweitsincisioces *48
Streptothrix cuniculi-..........-..-.------ *76, 126, 161
Dadhissccceces ss se2 tees cae egennaees 51
farCiNICd. 22.5 ccereseees ses Posed deevite 87
leproidés...... 232s sca0ssscxnaasaeess 82
polychromogenes.........-..-.-.---- 59
PROlelist io s.c vie fs 2s Seis eget stwatennctes 97
pseudo-tuberculosis. ..-....-----.--- 86
SUV LORPORE = jo:a 5.2.5, c1ecs stascmieinras arash BER Selects 122, 201
UL SURAT SS ay c.c:-<: = 5. Guavssensesanaynsaussave ainss srswrorsie eevee. Sieteieve 206
sulphur granules...........-..-2..-0-2 2 ee eee 113, 118
SWAXIM-BPOTES «eee eee eee cee 120 |
SWING PLAGUE sacra scaisrssrecinerremencerncrcls sav s'slees dais 145
Swiss authors on bacterial morphology............. 14
BYMDIORM Be onic sisters narsianciaisisigistsisieraws oe ares wets 40, 41, 207
symplasm.........222..220-02-- 166, *175, *176~*178, 210
APPCATANCO! ccc tc dete eed 182, 210 |
fOPMSMON sieves on -dseeenecewenceiessd #182, 189 |
PLOW LD oo iopreicns bec natiaeditsscisiieblels *182 |.
GiGi ates ece kusiee aera nee. 183
occurrence in the body. BS sieielewle caer 179, 182 |.
TOSIStANCE 2s 22 cess veces cowrseeeee sacs 183, 211
syphilis. (See Spirocheta pallida.)

 

Syphilisbacillus...:....-... M 62, 73; R *132; § *155, 194 |
Syphilomyces.........-.-------- 20s eee eee eee eee 14, 73
temperature, influence on reproductive organs.... 101,
206, 208
teratologic growth...........2.0.2.2-020-eeeeee eee 24
tertiary colonies.......-...0---- 0. eee eee eee eee 189
Thiocystis. .............- Pececanieese sede beads oa 181
186, 195
seibiawicigucte hiten ent lia ate die oie ote 170
trachoma bodies. ...-........-2e--ee ete eee ees 145, 184

 

SUBJECT INDEX.

Page.

trichobacteria...... M *87, 88; R *92, #94, 123, 162; S 186
classification. .......-..----------- 22,76

tubercle bacilli, resistance. .....-..------------ 131, 159
Tuberculomyces. ..-.----------2-+-eeee ener eee 21
tuberculose zoogléique....-.-.----+-++-25ee-eee 83, 172
tuberculosis, filterable gonidia..........---------- 148
tuberculosis group. ...-------------------++ M *83-*86;
R 159, 160; 8 169, 177, 194; C 201

typhus fever. ....---------+e+ eee eee eee eee *52, 146, 147
Tytothrix. ...--..-- 22 ee eee eee eee e cence eee eeee 64, 65
ultra-microérganisms. ...-------+-+++--+-e0e -» 148, 150
urea, effect on morphology...------------+-+-++++- 56, 207
Urobacillus Maddoxii.......--.-------------0--- 64
Urosarcina Hansenii......-...---.------ mitasiiecercs #49
VACCING sos acasene do smadioneneees reese Ee 144
vacuoles in bacterial cells........-------------+ *99, 113
variability of bacteria. .....-------------- 19, 29-33, 187
VaTIOla sx = avauccwcieceeiccekiedees omens Bl eeienie 144, 170
Vibrio Bonhoff..........--.-----0-22eee cree eceee *99
CONG. ce sis vedere wipes oie cie'e Sistsrsrse s Beaten 72

cholerze. .M *69-71; R *96, *98—*101, 103, *105,125,
133, 157, *158; S 173, *177, 193; C *201, *202

 

denitrificans. .....-...-----------+--+--- 59, 200
Finkler-Prior..........-.--------0-+-----+- M71;
R 96, 100, 125, 143, 158; S 173, 193
TWanolis.s secceieieistececccdensceseumarts es 69
lingualis........-...--..---- M *72; R96; © 200
Metchnikovil...........-..-.-------- 70, 100, 158
Miller ccc wt tatrnnneeameteccescaas ee 103
NAGASE. lect oceiseemescctmwcateee cus 72, 173
phosphorescens. -...-..-------2+---eee-ee- *72
proteus. (See V. Finkler-Prior. )
proteus Kohlbrugge............--.--..264- 72, 158
Rugula (Bonhoff)........-.2------ isk ae ahh 72
Rugula (Prazmowski). ......---..---.----- *68
saprophiles...........-..--..-- M 72; R96; 8193
SUIS: < 0/54)" jae teditbeneceuwednnd eee ketae
VONGCII 205 cise inlstaccin asians Gis ceovewies wad 23s si seaanee
'Vorticella infusionum
_ yeasts, copulation...........-....-. es
CHTOMOBY. . cs anasnsereceae ke cdimiria Sag weleieces
regeneration. .......- Ricci i eCNeRs ads SONS *172
BVM Plas ec. sso eoelcceetecletewe sss oases 181
POONER fins iodo dhbloonts 166, *168, 173, #174, *181, 186
ZOOSPOLEB sc. = serene nance de vaweie ved chemoneses 120
CON] WPAtION ji attend eee eee 202
formations sj. sseeaneeedier ete sscicenicnces 109
Zygosaccharomyces.-.....-.-. 2.2. eee cece e eee ee 199
ZYPOBPOLESs sisisysrsrareaieceieieecienwiewiieeaas 121, 122, *201-203
ZY RONG sissies cer beaiacieimede tines aus sanasaae weeds 203
PLATES A TOS AND I To XXIII.

[N. B.—The page numbers in () refer to the pages where
the illustrations have been discussed. |

 

160266°—21—_17 (258)
PLATE A.

Fig. 1 (p. 36).—Life cycle of B. azotobacter. The broken straight lines divide the different types of growth indicated by
the letters A toM. The Greek letters a to d refer to subdivisions, The single and double pointed arrows indicate the
development of one form from another. The four circles confine, in every case, all those forms which represent together
a rather constant mode of life, and which have been usually considered as bases for establishing separate species,
Lohnis and Smith (1916 a, p. 678).

Fic. 2 (p. 45).—Morphological changes of Micrococcus gonorrhoeae, Herzog (1913, original fig. 6).

254
Plate A.

National Academy of Sciences.

 

 

 

'
Gi
'

4
fo

 

 

 

Fig. 1.

 

 

sot

boos

a

<

 

 

Fig. 2.

255
PLATE B.

Fig. 3 (p. 46).—Pleomorphism of streptococci. Babes (1895, original fig. 2). 800.

4 (p. 47).—Leuconostoc hominis Hiava (1902, original figs. 3-6). 1,000.

5 (p. 50).—Proteusartiger Luftbacillus.

Matzuschita (1902, original figs. II-V).

1,000.
 

 

Plate B

National Academy of Sciences.

 

 

EPeaDq, ab 4 :

ero
Beye eeee meeees

x

eo
Sa |
x “see ae \
oS Sh
eae Ye 5 et

 
 

 

 

Fig. 5.

 

 

“y iy
? . ; = 4 ,
18 OO eo m” = ’ <
ee eS Set enw ¢ ic
e a Ww
SP PC2000 : ee 3 = .
oD ae. ota
Mee, Ws tore e
eee ees ,
Ss = a {lig
Rete ean
i Ve at
orn
aa, es
ce “ a i
fo Hay é ns i u
1 ty ‘ \ NY i
De, y Ube ‘
Co EE vem PVN A
@3¢0 ', « qeowy us
a Cae fi
8 4 wn ‘wens
a6 3 omens
Seeeses
So SO

 

 

 

 

 
PLATE C.

Fic. 6 (p. 49).—Urosarcina Hansenii Miquel et Cambier (1902, original fig. 172) 1,000.
7 (p. 50).— Bacterium merismopedioides Zopf (1883, original fig. 19). 900.
8 (p. 50).—B. bifidus communis Tissier (1900, original figs. 4and 6). Branching and ‘ “formes vésiculeuses”. X1,000.
9 (p. 52).—B. funduliformis Hallé (1898, original fig. 3). 1,000.
10 (p. 52).— Anaerobic bacilli from meningitis. Ghon, Mucha and Miiller (1906, original figs. 6, 20,and 9). 1,000.

258
National Academy of Sciences.

Plate C.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sg S358
0 §
1S
8
o
i
0 3
Pee e
=
BES
ie
30
al
0

 

 

 

 

Fig. 7.

 

 

 
PLATE D.

Fic. 11 (p. 53).—B. pestis. Albrecht and Ghon (1999, original figs. IT, 2-4; III,5 and 9). 1,000.
12 (p. 54).—B. influenzae. Crookshank (1896, original fig. 122, p. 249). >.1,200.
13 (p. 55).—B. involutus Waelsch (1905, original figs. 4and 6). 1,000.
14 (p. 55).—Proteus hominis capsulatus Bordoni-U ffreduzzi (1888 b, original fig. VII, 3). 750.
Plate D.

 

 

National Academy of Sciences.
oe
®,
@
ed eZ
Lf
4
Re : Tate
2 OF seo,
a of %
a if a
e
i. eee
® @
la One é
® ° en
s. 5s )e
y % a HV
o eG. Ofo |
fe OD . @ ?
ed g 804
Bago ak xy g) >
ey wo
oa. ‘ fe" we
an ;
Egret ial

 

 

 

 

 

 

 

 

 

‘
‘5 we
ene
Tae ’
eoerne _ woot
Ce ret a Soa a
‘ < ~ oer \
‘ if nasi, .
*\ eds wo 9
a é
s = é
ae as ‘
fs ios $oerrms
eee ees ‘
Nee . --=
vn wa ee
5 Wee a
. ‘
? i a
a Ly ah es a»
Baliga 2 Fig. 14.

 

  

 

 

 

a
Fig. 13.
262

PLATE E.

Fic. 15 (p. 58).—Bacet. aceti E. Chr. Hansen (1879, original fig. 77, 1911). 1,000.
16 (p. 59).—B. radicicola. Conn (1900, original fig. 26, p. 99). 1,000.
17 (p. 62).— Metallacter Bacillus Perty (1852, original fig. XV, 26a). 1,000.
13 (p. 62).—Different cell forms of bacilli. Ndgcli (1877, original fig. 2).
“19 (p. 62).—* Glycerin-Aethylbacterie”’ (Fitz). Buchner (1882, original fig. 10). x 4,000.
National Academy of Sciences.

Plate E.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

( fie
ney,
™ ( / /
OO? fo
MOEN) BE ee bid
Cay eased 7 ‘ok yet i |
( yd ) ey \ 4
Gis): ) Nee
ut yr LA ver i J
oi :
(Gay U/A\)
BS ES) iy oy
Se i ip Nous

 

 

 

263,
PLATE F.

Fig. 20 (p. 62).—Bac. mycoides. Olsen (1897, original fig. V, 14). 900.
21 (p. 63).—B. anthracis. E. Klein (1883, original figs. XXI, 1-3) 1,000.
22 (p. 63).-—B. anthracis, F. Khin (1885, original fig. 77, p. 109). 450.
23 (p. 63).—B. anthracis. E. Klein (1894, original figs. I, 1 and 2). 1,000.

264
Plate F.

 

 

 

265

 

 

 

 

 

 

 

 

 

 

National Academy of Sciences.

 

 

rue A)
ro e soot Ree
eee *e porn ee ae \
\ .. ate, e
= \ oecs. B ‘@, °
Q. & j _teare cog. ‘% oe
SS Ky \ od Sow. 2 Vo-.e8'
ITO : ‘ i ee Ot =
ays ~. 1 ? e-° Sse Se j
& £ aR ‘ ° ie Sep Oren, iS

. . \ @ See ue ns 9.

t e PE - = Se, ati
mb | t o , SeNOMS2e, > ~ 88, . =
ANS ae / et 2 vo O80 eae Sate=é SSNS % re

¢ ’ A Fete On, DS ®
“ oo” ee. <2. is
as . NJe, a
On * NOey e
& *e “ s
af “8, e
ws "© os 40004
S
A
bo
ie .
=a
A
~a
Ss
c}
q
XT, = bp
Ae? ry . re
CE Seymore 5
ME’ teks a
a :
Gas, g ot
Saat Book

 

 

 

 

 

 

Fig. 23.
266

PLATE G.

Fie. 24 (p. 65).—B. tumescens Zopf (1883, original fig. 23, p. 66). 900.

25 (p. 66).— Clostridium butyricum (above) and Clostridium Polymyrza (below), Prazmowski (1880,
original figs. II, 1, 2, 6,7). 1,020.

26 (p. 66).—‘‘ Micro-oidia’’ of B. amylobacter, Bredemann (1909, original figs. II, 32-36). 2,450.
National Academy of Sciences.

 

 

Plate G.

 

 

 

 

 

 

 

 

 

 

\I | :
> ‘ SS o8\
\ @ (3 ‘
i S ws t NNN
] % |
; \ s
\ [| n . 1D) \
3) =
\\ OX» 2 Q
: STH.H0 © :
89° 6 Oo i eee OCS
Fig. 24. Fig. 26.
0
e101 2) (0) Ae ,
My Ni j i rik
\ bY | <— Cae)
fey iis Meahh 0) “) ;
| {} Vo } y --

 

 

Fig. 25.

267

 

 
PLATE H.

FIG. 27 (p. 67).—B. Chauvoci, Ghon and Sachs (1903, original figs. 13, 14, 16-18). 1,000.
28 (p. 68).— Vibrio Rugula Prazmowski (1880, original figs. 1,10,and 11). 1,020.
29 (p. 72).—Spirillum endoparagogicum Sorokin (1887). 1,875.
30 (p. 73).—Spirillum rubrum. Meirowsky (1914 b, original figs. Vb, 6 and 7). 2,000.
31 (p. 73).—Spirochacta icterohacmorrhagica Inada et al. (1916, original figs. 69 and 70).
32 (p. 75).— Bact. repens Miche (1913, original figs. II, 9,and 10). 1,120-1,200.
33 (p. 77).—B. mallei. Afarzr (1899, original figs. 1-4).
 

National Academy of Sciences.

Plate H.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a
¢¥ Q.
ne x 4 Vd
pare wF
On” e,. o ’
~¥, ° t
0
ey 4
p \ 4
Sey
Fig. 27.
°
°
°
‘0
°
Fig. 28. Fig. 29.
a ~~
NX Si =
\anios we We Be, ~ ee
NO A

cy

Fig. 30.

aR
AX

 

 

 

 

 

 

 

ae lee
igs er ay Se)

 

 

 

His Fig. 33.

 

 

 

Fig. 31.

160266 ° —21——_18

Fig. 32.
270

PLATE J.

Fic. 34 (p- 80).—Pscudo-diphtheria bacilli. Babes (1895, original fig. 3). ><800.
35 (p. 81).—B. enzymicus Mellon (1917, original fig. 2, p. 88).
36 (p. 82).—B. leprae. Lutz (1886, original figs. 7-11, p. 81). 1,000.
37 (p. 82).—B. leprae. Kedrowski (1910, original fig. 37). 1,000.
38 (p. 81).—B. tuberculosis. Metchnikoff (1888 a, original fig. V, 20). 1,200.
National Academy of Sciences.

 

 

 

Plate J.

 

 

 

rsdieesce 3! oes
on RAG
so@ed ge © aan ¢ fF

 

 

 

 

 

Fig. 34
of
x
=e
“Nye =, S fate
Prete < *Saass:
angteeec >” Fr %e
o¢ om Qai
Sones S
we Oe
vont,
Fig. 35.

Fig. 36.

 

 

 

 

Fig. 37.

 

 

 

 

 

 

271

 

 
PLATE K.

Fie. 39 (p. 87).—Cladothrix (a-/) and Actinomyces (g-i). Migula (1900, original fig. 45, p. 39). e and h 500; all
others <1,000.

40 (p. 91).—Sporonema gracile Perty (1852, original fig. 26 on P]. XV). 1,000.
41 (p. 92).—Germinating gonidia. Kindfleisch (original fig. XVIII, 1).

272
National Academy of Sciences. Plate K.

 

“

Moe

       

 

CD
>
Se es
Soe,
ky

 

é

   

Le Sor age

 

 

 

Fig. 39.

 

 

 

 

 

  

Fig. 41.

273

 
PLATE L.

Fic. 42 (p. 91).—Gonidia and Spirilla. Perty (1852, original figs. 28, 29B, and 31 on Pl. XV). 1,000.
43 (p. 92).—‘*Sporangium”’ and “‘spore”’ of Crenothrix. F. Cohn (1870, original figs. 11 and 13). > 500-800.
44 (p. 93).—Formation of gonidia and gonidangia. Billroth (1874, original fig. 36). 1,185.
45 (p. 93).—Formation and liberation of gonidia. Billroth (1874, original fig. 42). 1,185.
46 (p. 94).—Phragmidiothrir multiseptata. Engler (1882, original fig. 24). 400.

274
National Academy of Sciences. Plate L.

 

 

 

eorseee er12 ‘
i ae z

4 e

¢ , ‘

 

 

 

 

 

 

 

 

 

 

Fig. 43.

 

 

 

 

 

 

 

 

 
PLATE M.

Fia. 47 (p. 94).— Crenothrit Kiihniana. Zopf (1883, original fig. 6). 600.
48 (p. 94).—Beggiatoa alba. Zopf (1883, original fig. 27). 900.
49 (p. 96).— Vibrio cholerae. Hueppe (1885, original fig. 2, p. 622).
50 (p. 96).— Metabacterium polyspora Chatton et Pérard (1913, p. 1233). 2,000.

276
Plate M.

 

C6615 Qa. TAL Sous, {ee
C538)5 QSsy We 2So; Gn, Mei

 

Oye
Elijekekeh

SRA RSPAS ea

ue

              

         

 

   
 

APT

Pegs aise Eelei dt

 

    

29a ood
P20 02 2900090
fone Po 329298989905

Fig. 48.

       

 

aie.
oo — 2-550 DM MeO. I Oa
See OD LOR 2 OYOr SQ OV To”
Soe 2 —— Oo IQD OBS 2
a SO 6F o LAS, o.OLS Ss
EB BEC SE OES 2009300908 F:5.O° SF

 

 

 

 

 

 

National Academy of Sciences.

 

 

Fig. 47
0 0)
S553
€
ove
g
Fig. 49

 

 

 

 

 

 

 

 

Fig. 50.

 
Fig. 51 (p.
52 (p.
53 (p.
54 (p.
55 (p.
56 (p.
57 (p.

PLATE N.

97).—B. muralis. Tomaschek (1X8x, p. 183). 1,000.

98).—Jodococcus vaginatus. W. Miller (1889, original fig. 15, p. 54). 1,100.
100).—Ascobacterium luteum. Babes (1895, original fig. 13 B). 800.
101).—B. typhi and V. cholerae with “conidia.” Almquist (1908, original figs. 5, 13, and 21).
102).—B. tuberculosis. CrookshanJ: (1896, original fig. KI, 12). 1,200.
102).— Fubacillus multisporus Dangeard (1891, original fig. VIII, 4). 1,000.
103).— Bact. orydans. Henneberg (1898, original fig. 1). 1,000 (?).
National Academy of Sciences.

 

 

(

=)

Li FED
CD QS» _S
Cee) SS) Cees
A ED Rt
ise ex
e (Sy Sy
CEN \ eZ =\ oe)
© By
AC Be Use
co @ ~~
NS)
es
(Sor?

 

 

 

 

 

 

 

 

 

 

 

Fig. 55.

 

 

 

 

 

 

 

 

 

Plate N.

 

 

=
42
>
aN
ZA g
Po 4, of
io
LEO, Vo
y 7 f 3

 

 

 

 

 

Fig. 57.

 

 
280

Fia. 58 (p.
59 (p.
6) (p.
61 (p.
62 (p.
63 (p
64 (p.

65 (p.

PLATE O.

103).—Upgrowth of B. pestis. N. K. Schultz (1901 a, original fig. I). 1,000.

106).—“‘Spore” formation by spirochaets. Fantham (1911, original fig. 5, p. 489).
106).—Spirochaeta anodontae. Bosanquet (1911, original figs. 21 and 22). 2,000.
111).—Regeneration of B. pestis. N. K. Schultz (1901 a, original fig. IL). 1,000.

117).—“ Plasmoptysis” of V. cholerae. A. Fischer (1903, original fig. 27). 1,500.
123).—Saprospira grandis (left) and Cristispira tapctos (right). Gross (1912, original figs. 8 and 9).

123).—Microcysts of B. cohaerens (1,000) and of B. ruminatus (3,500). A. Afeyer (1901 b, origi-
nal figs. XX, 4and 15).

123).— Myxococcus macrosporus. Zukal (1897, original fig. 3). 1,400.
National Academy of Sciences.

 

 

=
eeee
S@nw

 

 

 

we

Plate O,

 

 

 

pueU ics
ae od

eo
a

 

 

Fig. 58.

Fig.

 

 

 

 

 

64,

Fig. 61.

 

 

 

 

 

 

 

 

 

@ @ €
go €
® 6 o-
Sa B
a
a
i
q
Fig. 59, @
; q
a &
Se ea eta nie ss
ty ca _
2 i ~
ce, eC C
LY, 5 mo
°; ors +
Sop aqn = Gp S
“es
BIC Fig. 63
koa.
2
wo 4a) e Qo -
ay © Orman OG 0
©0 ox, ED, ©
) SF OEE ©

 

 

 

 

- 62.

 

 

 

 

281
282

Fie. 66 (p.
67 (p.
68 (p.
69 (p.
70 (p.
71 (p.

72 (p.
73 (p.
74 (p.
75 (p.

PLATE PP. 4

124).—Gonidia formation of B. radicicola. Morck (1891, original figs. III, 4, and V, 12).
125).—Gonidangia of B. radicicola. Morck (1891, original figs. III, 2 b-d).
125).—Gonidangia from diphtheria. Letzerich (1876, original figs. XII, 1-4). 480.
128).—Crenothrix branching. Miguila (1900, original fig. 43 fandg). 1,000.
130).—Upgrowth of Cladothriz dichotoma. Billet (1890, original figs. 1V,4and 5). 1,600.

132).—S pirochaeta ictcrohaemorrhagica with regenerative bodies. Inada, Ido, Hoki, Kaneko,
and Jto (1916, original figs. 40-47 on Pl. 61).

132).—Streptococcus pyogenes. Hewlett (1902, original fig. Ila). 1,500.
136).—Spirobacillus Cienkowskii. Metchnikoff (1889, original fig. 14, P11). 2,020.
143).—Coiled spirochaets. Prowazck (1907a, original fig. I, 8h).

143).—Encysted spirochaets. Inada et al. (1916, original figs. 71 and 72).
National Academy of Sciences.

Plate P.

 

 

 

 

 

 

 

 

 

aR 0 ©
‘er we 3 &
é 0. P Oe 6 m0, 60 po
QS are & é % ia 5 o 0
~ \ —aty? t wer
O78 sa 3 , ales OAR Oo Pe QO
3 oO 7 ‘b o 300 )
Cane ft 8 eo 2 Benes ; * vy co ‘Bo
. ¥ > oy oT 0.
Plo) t Si e ~
a 8 a a R Ye Nd
ot AS o Ry L g
20) %
O_ Fig. 67
Fig. 66.
1
a
ie0 es
S ey: 6s

 

 

 

 

 

 

 

 

Fig. 68.

 

 

 

 

 

Fig. 70.

 

 

 

 

 

Fig. 72.

 

 

 

 

 

 

 

i<j t=

 

 

 

 

. Yo \ iy e,
= / ~ ft
j \ a 5 Py
Fig. 71. Fig. 73.

 

 

 

 

Fig. 74.

Fig. 75.
283
PLATE Q.

Fie. 76 (p. 148).—Bacterial development from “micro-gonidia.’’ Aarsten (1869, original fig. V).

77 (p. 148).—Rods and “ microzymas.’’ Béchamp (1883, original fig. I, 4). 650,

78 (p. 167).—Development of protozoa and fungi. Pineau (1845 a, Pl. 4 bis). 400. Figs. 8-10, Monas lens; 11-13,
Enchelys ovata; 14-20, Vorticella infusionum, 21-27, Penicillium glaucum.

79 (p. 168).— Development of protozoa (Podophyra and Actinophrys). Perty (1852, original figs. VIII, 9.and 10). 1,000,

284
Plate Q.

National Academy of Sciences.

 

 

 

 

 

 

 

 

 

[Plein the

 

 

Fig. 78,

19

 

—21

160266°

 

 

 
286

PLATE R.

Tia. 80 (p. 170).—Growth obtained from diphtheria. Letzerich (1876, original figs. XII, 5-10). 480.
81 (p. 172).—Regeneration of yeast cells in vinegar. Béchamp (1883, original fig. 1,5). 650.
82 (p. 173).—Symplasm from leprous tissue. Lufz (1886, original figs. 1-6). 1,000.

83 (p. 174).—Growth of bacteria from root nodules. Frank (1890, original figs. VIII, 34 a-c).
aand bX660, c X1,030.
National Academy of Sciences. Plate R.

 

 

 

 

 

Fig. 80.

 

 

2 DOL —- Wrvege

 

   

 

 

 

 

 

 

 

 

Fig. 83.

to
a
“1
288

Fia. 84 (p.
85 (p.
86 (p.
87 (p.
88 (p.
89 (p.

PLATE 6&.

174).—Symplasm from root nodules. Morck (1891, originals figs. 3 and 12 c and/).
175).—Irregular cells of B. cyanogenes. Neelsen (1880, original fig. XI, 10). 650.
177).—Spirillum rubrum. Meirowsky (1914 b, original figs. V b, 4-5). 2,000.

180).— Myxococcus ruber. Baur (1905, original fig. 3, p. 113).

181).—Cyst of Polyangium. Zukal (1897, original fig. 8). 50.

201).—Spirochaeta gallinarum. Prowazek (1906 b, original figs. II, 10c,;1la@andb). 1,560.
National Academy of Sciences.

Plate S.

 

 

fie sees
L { 4 Se ) i
CaaS Vas CT) Sard
Nw y bea
C x50 f } (
Gs ) lo
WA

 

 

 

 

 

 

 

Fig. 84.

 

 

 

 

 

Fig. 86.

 

 

 

 

 

Fig. 88.

 

 

    

- >)
ay NAAN

Py dW Vee
Ne Zh he
Sa hei

 

 

Fig. 87.

 

 

oS >

 

 

 

Fig. 89.
289
Fie.

Naot wnwo wv

290

PLATE I.

. (p. 43): Microc. candicans. Léhnis and Smith (1916 a, original fig. 13). 1,000.

. (pp. 43,135): Microc. candicans. Léhnis and Smith (1916 a, original fig. 37). 1,000.
. (pp. 43,135): Microc. candicans. Léhnis and Smith (1916 b, original fig. 14). 1,000.
. (pp. 48,135): Microc. candicans. Léhnis and Smith (1916 6, original fig. 15). 1,000.
. (pp. 48, 135): Microc. rubefaciens. Matzuschita (1900, original fig. 2). 1,000.

. (pp. 43, 135): Microc. flavus. Matzuschita (1900, original fig. 4). 1,000.

. (p. 44): Microc. melitensis. Jordan (1916, original fig. 99). 1,000.

. (pp. 44,190: Microc. candicans. Léhnis and Smith (1916 b, original fig. 40). 1,000.
. (pp. 48,191): Poliomyelitis. Rosenow and Towne (1917, original fig. X, 7c). 1,000.
10.
Ls
12.

(p. 48): Poliomyelitis. Mathers (1917, original figs. I, land 3). 1,200.
(pp. 48, 142): Streptoc. lactis. Léhnis and Smith (1916 b, original fig. 10). 1,000.
(p. 48): Streptoc. tyrogenus. Migula (1900, v. IJ, original fig. I, 2). 1,000.
National Academy of Sciences. Plate I.

 

  

tn.

“=
ie
s
nat
&
wa
4
& x7
= ‘
\
wt

 

 

 

 

291
PLATE II.

Fic. 13. (p. 48): Streptoc. equi. Migula (1900, v. II, original fig. IT, 5). 1,000.

14. (p. 48): Sarcina flava. Léhnis and Sm2th (1916 b, original fig. 11). 1,000.

15. (pp. 48, 142): Sare. flava. Léhnis and Smith (1916 b, original fig. 12). 1,000.

16. (pp. 48,190): Sare. flava. Léhnis and Smith (1916 b, original fig. 54). 1,000.

17. (pp. 48,190): Sare. flava. Léhnis and Smith (1916 b, original fig. 55). 1,000.

18. (pp. 49, 134): Bact. bruneum. Matzuschita (1900, original fig. 6). 1,000.

19. (p. 49): Azotob. vitreum. Léhnis and Hanzawa (1914, original fig. 24). 800.

20. (p. 49): Azotob. vitreum. Jéhnis and Hanzawa (1914, original fig. 22). 800.

21. (p. 49): Azotob. vitreum. Léhnis and Hanzawa (1914, original fig. 23). 800.

22. (pp. 52,192): Typhus exanthematicus. P. Th. Miiller (1913, original fig. 2). 1,000.

28. (p. 52): Typhus exanthematicus. P. Th. Miiller (1913, original fig. 5). 1,000.

24. (pp. 52,193): Typhus exanthematicus. P. Th. Miller (1913, original fig. 1). 1,000.
292
Plate Il.

National Academy of Sciences.

 

 

 

24

nN

a

 

 

293
Fie. 25

294

26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.

PLATE III.

(p. 53): B. erysipelatos suum. J#zerott und Niemann (1895, original fig. 40). 1,000.
(p. 52): B. erysipeloides. Rosenbach (1909, original fig. XIII, 5). 1,000

(p. 52): B. erysipeloides. Rosenbach (1909, original fig. XII, 5). 1,000.

(p. 53): B. murisepticum. Rosenbach (1909, original fig. XITI, 3). 1,000.

(pp. 53,192): B. erysipeloides. Rosenbach (1909, original fig. XIT, 3). 1,000.

(pp. 53, 134): B. cholerae gallinarum. Itzerott und Niemann (1895, original fig. 36). 1,000.
(pp. 54,101): B. pestis. Muir and Ritchie (1903, original fig. 148). 1,000.

(p. 54): B. pestis. Muir and Ritchie (1903, original fig. 150). 1,000.

(p. 54): B. pestis. Muir and Ritchie (1903, original fig. 147). 1,000.

(pp. 54,192): B. pestis. Rowland (1914, original fig. XIX, 12). 1,000.

(pp. 54,134): B. pestis. Rowland (1914, original fig. XX1I, 17). 1,000.

(pp. 56,127): B. coli. Matzuschita (1900, original fig. 14). 1,000.
National Academy of Sciences.

Plate Ill.

 

 

c
~ aba ,
- Fy he
& ih a
28
ter seer cao py 4
op Say Ns ys
‘ wha’ Se Noten eat bt
Ste ee a ON N
wee a 4 _ In if
Pine “7a™] a
RON Stayt eg s}
Ser Nema ain ca ial neg) Wie te maa
\ NS) i~ Sie
. ' ~ ,
( ee 027
le Na 08 fred
\we soy
» , a ’ are
Vy tie a re oily
/ . ee orn |
‘ MS ety Aad
of . «emai ‘
u Ney ‘ |
tte ee
“ en «
ae a \
31

i \ AS
=
a
, '
a }
ad Br
i Gee SS eae

“loap
Fs) gO %
° eg
et
Bae “aa
¢ < ort o ae :
ie. ;
Poe » San
a
Scie
i wy D
‘ + ge
ow a?
32

 

  
 
 

-
a 1%
®

   

na
s
yee
a fo
=~ =
=z st
a
= oUt g
ee
‘ae:

 

 

295
PLATE IV.

Fie. 37. (pp. 55,185): B. pneumoniae. Toenniessen (1913, original fig. III). 1,000.
38. (pp. 55,185): B. pneumoniae. Toenniessen (1913, original fig. IV). 1,000.
39. (pp. 55,185): B. pneumoniae. Toenniessen (1913, original fig. V). 1,000.
40. (pp. 55,185): B. pneumoniae. Toenniessen (1913, original fig. VI). 1,000.
41. (pp. 55,185): B. pneumoniae. Toenniessen (1913, original fig. VII). 1,000.
42. (pp. 55,185): B. pneumoniae. Toenniessen (1913, original fig. VIII). 1,000.
43. (p. 57): B. coli. Kellerman and Scales (1916, original fig. 8). 1,000.

44. (pp. 57,177): B. coli. Kellerman and Scales (1916, original fig. 11/12). 1,000.
45. (pp. 57,177): B. coli. Kellerman and Scales (1916, original fig. 9). 1,000.

46. (p.57): B. typhi. Matzuschita (1900, original fig. 16). 1,000.

47. (p. 57): B. typhi. Itzerott und Niemann (1895, original fig. 18). 1,000.

48. (p. 60): B. fluorescens. Léhnis and Smith (1916 b, original fig. 7). 1,000.

296
National Academy of Sciences. Plate IV.

 

 

39

  
    
  
 

BEA Ye 0 Meld HRS ay Oe eee? “wy eed
SNe ne Ss Re I ag

f ety D tte
a AE SIP
< Pee ‘

      
   

set dee i "3 5
ey ST glen AABN) te 2
Seite ei ee Mage jas at
0 EE GATE NS

40 41 42

=
fi, We)

4m

 

 

 

 
298

Fie. 49.
59.
Bl.
52.
53.
5A.
55.
56.
57.
58.
59.
60.

PLATE V.

(p. 61): Nitrosomonas Ziirich. Winogradsky (1892, original fig.1). 1,000:
(p. 61): Nitrosomonas Kasan. Winogradsky (1892, original fig. 4). 1,000.
(p. 61): Nitrosomonas Java. Winogradsky (1892, original fig. 12). 1,000.
(p. 61): Nitrobacter. Winogradsky (1891, original fig. XVIII, 1). 1,000.
(p. 61): Nitrobacter. Winogradsky (1892, original fig. 16). 1,000.

(p. 63): B. anthracis. Jtzerott und Niemann (1895, original fig. 13). 1,000.
(p. 63): B. anthracis. Jtzerott und Niemann (1895, original fig. 14). 1,000.
(p. 63): B. anthracis. Ttzerott und Niemann (1895, original fig. 15). 1,000.
(pp. 63,138): B. anthracis. Matzuschita (1900, original fig. 31). 1,000.

(p. 63): B. anthracis. Henri (1914, original fig. 4). 925.

(pp. 63, 134): B. anthracis. Henri (1914, original fig. 5). 925.

(p. 63): B. anthracis. Henri (1914, original fig. 6). 925
Plate V.

National Academy of Sciences.

 

30

   

‘dy “si
. ah
we! Sy OF
oot WES

Bf het! weet i

Pg er eee
ers Soe
@

  

 

 

 

60

 

 

299
Fig. 61.
62,
63.
64.
65.
66.
67.
68.
69.
70.
V1.
#2.

300

PLATE VI.

(p. 64): B. anthracis. envi (1914, original fig. 7). 925.

(p. 64): B. anthracis. /fenri (1914, original fig. 8). 925.

(p. 64): B. anthracis. /Tenri (1914, original fig. 9). 925.

(p. 64): B. subtilis. ZLéhnis and Smith (1916 b, original fig. 6). 1,000.

(pp. 64,138): B. subtilis. Léhnis and Smith (1916a, original fig. 26). 1,000.

(p. 64): Bac. aus Hackfleisch. Maassen (1904, original fig. XTV, 11). 1,000.

(p. 65): B. malabarensis. Léhnis and Pillai (1907, original fig. 2). 1,200.

(pp. 65, 134): B. Azotobacter. Léhnis and Hanzawa (1911, original fig. 28). 800.
(pp. 65, 136): B. Azotobacter. Léhnis and Smith (1916 a, original fig. 20). 1,000.
(pp. 65, 185): B. Azotobacter. Léhnis and Smith (1916 a, original fig. 16). 1,000.
(p. 65): Azotobacter. Walton (1915, original fig. I, 9). 1,000.

(p. 65): Azotobacter. Walton (1915, original fig. I, 10). 1,000.
National Academy of Sciences. Plate VI.

 

66

 

i¢ We . , = ' ie
as 6 he oe

68 69

thane

Ma lNeegeese

 

 

 

 

301
160266°—21 20

 

 
Fic. 73.
74,

7s

75.
76.
Ts
78.
79.
80.
81.
82.
83,
84.

. 66):
. 67):
. 67):
. 67):
. 67):
. 67):
. 69):
. 69):
. 70):
. 70):
. 70):
. 72):

PLATE VII.

Clostr. Pastorianum. Winogradsky (1902, original fig. 3). 1,000.

. Chauvoei. Grassberger (1903, original fig. 54). 1,000.

. Chauvoei. Grassberger (1903, original fig. 21). 1,000.

. Chauvoei, Hibler (1908, original fig. X, 16). 800(?).

. Chauvoei. Hibler (1908, original fig. XVI, 7). 1,000(?).
oedematis maligni. Grassberger (1903, original fig. 42). 1,000.
cholerae. Fraenkel und Pfeiffer (1895, original fig. 99). 1,000.

. cholerae, Jtzerott und Niemann (1895, original fig. 53). 1,000.
. cholerae. Stamm (1914, original fig. XI, 2and 4). 1,000.

. cholerae. Stamm (1914, original fig. XI, 5 and NII, 15). 1,000.
. cholerae. Stamm (1914, original figs. land 2). 1,000.

. phosphorescens. Maassen (1904, original figs. 10 and 3). 1,000.

addddabnood
Plate VII.

National Academy of Sciences.

 

 

 

ne WA SgTe

=
Kg oo 6 ON

   

 

fe ro
y t ! 7

= J. KOM,
we SIN, SG

Bos Ne
ON Re
: 4," NONE yy) &

es Bee eG
Sy* Ae. Prt:
s 5 ~ . ry .
x - ae g
ih ’ a \ :
y oft Ny
‘- 8 “
ie

 

303

Aoi \ ee)
es
C ¥
. ¢ ae
ar 7.
aS

 

 

 
Fic.

304

85

86.
87.
88.
89.
90.
“91.
92.
93.
94.
95.
96.

PLATE VIII.

. (p. 72): V. lingualis. Bajardi (1903, original fig. 1). 1,000.

(pp. 72, 203): Spirillum rubrum. Reichenbach (1901, original fig. 6). 1,000.

(pp. 73, 203): Spirillum rubrum. Reichenbach (1901, original fig. 8/9). 1,000. 5
(p. 73): Spirochaeta pallida. Meirowsky (1914 6, original fig. XT, 46, 55, and 56). >< 2,000-3,000.
(pp. 76, 98): Streptothrix cuniculi. Schmorl (1891, original fig. 2). 1,000.

(pp. 76, 98): Streptothrix cuniculi. Schmorl (1891, original fig. 5). 1,000.

(pp. 77,134): B. mallei. Carpano (1918, original fig. 1). 1,000.

(p. 77): B. mallei. Carpano (1913, original fig. 2). 1,000.

(p. 77): B. mallei. Carpano (1913, original fig. 3). 1,000.

(p. 77): B. mallei. Carpano (1913, original fig. 4). ><1,000.

(p. 77): B. mallei. Carpano (1913, original fig. 5). 1,000.

(pp. 77,126): B. mallei. Carpano (1913, original fig. 6). 1,000.
National Academy of Sciences. Plate VIII.

 

 

 

 

 

305
306

Fic.

97.

98.

99.
100.
101.
102.
103.
104.
105.
106.
107.
108.

PLATE IX.

(pp. 79,134): B. diphtheriae. J. Dale (1910, original fig. 6). 1,000.

(pp. 80, 134): B. variabilis. Nakanishi (1900 ¢, original fig. 8). 1,000.

(p. 81): Corynebacterium. £. de Negri (1916, original fig. 12). 560.

(p. 81): Corynebacterium. 2. de Negri (1916, original fig. 11). 560.

(pp. 81, 134): Corynebacterium. JZ. de Negri (1916, original fig. 30/73). 560.
(pp. 84, 98): Pia mater tubercul. (Corni/ et Babes (1890, original fig. 342). 500.
(pp. 84, 134): Tuberculosis. Arrigo (1900, original fig. 1 a).x670.

(pp. 84,134): Tuberculosis. Arrigo (1900, original fig. 2 b). 670.

(p. 84): B. tuberculosis. Migula (1900, v. II, original fig. V, 8). 1,000.

(p. 86): Pseudo-tuberculosis. Preisz (1894, original fig. III/IV). 1,000.

(p. 86): Pseudo-tuberculosis. Preisz (1894, original fig. V). 1,000.

(p. 86): Actinomyces. J. Israél (1878; original fig. II, 4). 750.
Plate IX.

 

 

 

National Academy of Sciences.

99

98

97

 

}

10:

101

 

105

104

103

 

 

 
Fie. 109.
110.
lil.
112.
113.
114.
115.
116.
117.
118.
419.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.

308

(p.
(p.
(p.
(p.
(p.
(p.

(pp.
(pp.
(pp.
(pp.
(pp.
(pp.
(pp.
(pp.

PLATE X.

92): B. termo. R. Koch (1877, original fig. XV, 4). 500.

92): B. fluorescens. Léhnis and Smith (1916 a, original fig. 40). 1,000.
92): B. anthracis. R. Koch (1877, original fig. XVI, 5). 700.

92): B. anthracis. Gtinther (1906, original fig. 29). 1,000.

94): B. anthracis. R. Koch (1881, original fig. VI, 33). 700.

94): B. oedematis maligni. &. Koch (1881, original fig. VIII, 46). 700.

95,126): B. Chauvoei. J/ibler (1908, original fig. VIII, 18). 1,900.

95,126): B. Chauvoei. Hibler (1908, original fig. VIII, 17). 2,000.

99,138): B. Chauvoei. Fraenkel und Pfeiffer (1895, original fig. 60). 1,000.

99,138): B. pestis. Maassen (1904, original fig. X, 4). 1,000.

99,138): B. lact. aérogenes. AMaassen (1904, original fig. X, 7). 1,000.

97,138): Proteus vulgaris. Hauser (1885, original fig. X, 16). 1,000.

99,138): V. cholerae. Maassen (1904, original fig. XI, 1). 1,000.

96,126): Pneumococcus. Artigalas (1885, original fig. 1, pl. 5).

. 97,153): Proteus hominis capsulatus. Bordoni-Uffreduzzi (1888 6, original fig. VIII, 6). 750.

. 98): V. cholerae. Cornil et Babes (1890, original fig. 275). 1,500.

. 99): B. pyocyaneus. Matzuschita (1900, original fig. 26). 1,000.

. 99): Spirilla. Itzerott und Niemann (1895, original fig. II, 7). 1,000.

. 99): B. anthracis. Jtzerott und Niemann (1895, original fig. III, 16). 1,000.

. 99): B. capsulatus. Jtzerott und Niemann (1895, original fig. VITI, 46). 1,000.
. 99): V. Bonhoff. Jtzerott und Niemann (1895, original fig. X, 60). 1,000.
National Academy of Sciences. Plate X.

vic {NW
WIS cN
fy? } Ls

 

     
  

Tie:

120 121

 

 

  

 

309
PLATE XI,

Fie. 130. (p. 99): Spirillum rubrum. Itzerott und Niemann (1895, original fig. XVIII, 104). 1,000.
131. (p. 99): Bacillus sp. Zettnow (1891, original fig. 14). 1,650.
132. (p. 99): Spirillum Undula. Zettnow (1896, original fig. 16). 1,125.
133. (p. 100): Spirillum sp. Migula (1904, original fig. II, 8). 1,000.
134. (p. 100): B. proteus. Migula (1904, original fig. II, 4). 1,000.
135. (pp. 101, 202): B. typhi. Almquwist (1916, original fig. 4). 1,000.
136. (p. 101): B. typhi. Almquist (1916, original fig. 5). 1,000.
137. (p. 101): B. dysenteriae. Almquist (1916, original fig. 23). 1,000.
138. (pp. 101, 202): B. typhi. Almquist (1917, original fig. 1). 1,000.
139. (pp. 101, 202): V. cholerae. Almquist (1917, original fig. 4). 1,000.
140. (pp. 101, 202): V. cholerae. Friedrich (1892, original fig. V, 4). 800.
141. (p. 101): B. pyocyaneus. Muir and Ritchie (1903, original fig. 69). 1,000.
142. (p. 101): B. enterititidis. Gunther (1906, original fig. 54). 1,000.
143. (p. 101): B. ochraceus. Migula (1900, original fig. XI, 3). 1,000.
144. (p. 101): Spirillum sporiferum. Migula (1900, original fig. XVIT, 5). 1,000.
145. (p. 101): B. phlegmasiae uberis. Kitt (1899, original fig., p. 393). 1,000.
146. (pp. 102,134): Streptoc. lanceolatus. .4zelrad (1903, original fig. 4). 1,000.
147. (pp. 102,134): B. coli. Azelrad (1903, original fig. 19). 1,000.
148. (pp. 103, 153): B. bulgaricus. Léhnis and Smith (1916 a, original fig. 29). 1,000.
149. (p. 103): Bac. aus Hackfleisch. Maassen (1904, original fig. XI, 4). 1,000.
150. (pp. 103, 202): B. subtilis. Hiss and Zinsser (1911, original fig. 125).
151. (p. 104): B. Chauvoei. Grassberger (1903, original fig. 67). 1,000.
152. (p. 104): Clostridium Pastorianum. Winogradsky (1902, original fig. 1). 1,000.
153. (p. 105): B. leprae. Kedrowskz (1910, original fig. 38 a). 1,000.

310
National Academy of Sciences. Plate XI.

| é i :
I =a
4

ut) »!

 

 

a \
ieee

y oa

’ i:

2 a lr a)
134 135

     

g °° RET Peewee
+ oO Mette

    

 

   

Yr
ff
t a ay . e

Mie — Bn senlee®
Vel |

152 153

\
i
LN

 

 

 
Fie. 154

312

PLATE XII.

. (pp. 105,134): B. leprae. Kedrowski (1910, original fig. 38 b). 1,000.

155. (pp. 105, 134): Meningitis. Ghon, Mucha und Miiller (1906, original fig. 3). 1,000.
156. (pp. 105, 134): Meningitis. Ghon, Mucha und Miiller (1906, original fig. 18). 1,000.
157. (pp. 105, 134): B. cyanogenes. Neelsen (1880, original fig. XI, 9). 650.

158. (pp. 105, 134): V. cholerae. Hammeril (1906, original fig. 4). 1,000.

159. (p. 106): B. fusiformis (?). Rosenow and Tunnicliff (1912, original fig. I, 3). 750.
160. (p. 106): B. Azotobacter. D. H. Jones (1913, original fig. V, 9/10. 1,000.

161. (p. 106): B. subtilis. Marrassini (1913, original fig. I, 28). 1,000.

162. (p. 107): B. tuberculosis. Meirowsky (1914 6, original Pl. IIa). 2,000.

163. (p. 107): B. tuberculosis. Meirowsky (1914 6, original Pl. IIa). 2,000.

164. (pp. 107,135): B. tuberculosis. Meirowsky (1914 6, original Pl. IIa). 2,000.

165. (p. 107): B. leprae. Meirowsky (1914 }, original Pl. IIb). 2,000.

166. (pp. 107,135): B. leprae. Meirowsky (1914 b, original Pl. IIb). 2,000.

167. (pp. 107,135): Spirillum rubrum. Metrowsky (1914 b, original Pl. Va). 2,000.
168. (pp. 108,151): M. candicans. Léhnis and Smith (1916 8, original fig. 31). 1,000.
169. (pp. 108, 124): M. luteus. Léhnis and Smith (1916 5b, original fig. 32). 1,000.

170. (pp. 108, 132): Streptoc. lactis. Léhnis and Smith (1916 6, original fig. 30). 1.000.
171. (pp. 108, 153): B. pneumoniae. Léhnis and Smith (1916 b, original fig. 37). 1,000.
172. (pp. 108,153): B. bulgaricus. Léhnis and Smith (1916 b, original fig. 36). 1,000.
173. (p. 108): Azotob. Beijerinckii. Léhnis and Smith (1916 b, original fig. 33). 1,000.
174. (p. 108): B. fluorescens. Léhnis and Smith (1916 b, original fig. 28). 1,000.

175. (p. 108): Az. vinelandii. Léhnis and Smith (1916 a, original fig. 24). 1,000.

176. (pp. 108, 129): Az. chroococcum. Léhnis and Smith (1916 b, original fig. 26). 1,000.
177. (p. 108): Bacillus sp. Goadby (1917, original fig. 14). 1,000.
National Academy of Sciences.

Plate XII.

 

158

162

174

 

 

159

175

 

164

 

‘60

 

 

 

169

 

v aN
mae
\'
ee
rn in
1 Ne
\ A
,
Mie

 

 

313
PLATE XIII.

Mie. 178. (pp. 108, 192): B. paratyphi B. Meirowsky (1914 b, original Pl. III). 2,000.
179. (pp. 108, 127): B. typhosus. Hort (1917 a, original fig. 1). 1,500.
180. (pp. 108,127): B. coli. Hort (1917 a, original fig. 2). x 1,500.

314

 
National Academy of Sciences. Plate XIII.

 

 

4
Ig - - f t} \'
a fio GJ / "
. 0 ¢ be Us Q 4 f 5 6 @ } 8 | g 9

Cr gin

 

 

 
PLATE XIV.

(pp. 107, 128, 158, 177): Budding and branching of bacteria. Meirowsky (1914 b, original Pl. XVITI). 2,000.
316
National Academy of Sciences.

 

 

 

 

Plate XIV.
Freie .
Seiten- und Kn Teilung der Stiten-
Se tet | Davin | omen, | du | Sen
Knospen formen eat ace
Tubercul. hum. % @
Reinkultur i : 4 x ;
Tubercul. bov. e
Reinkultur i ° 7 .. 4

 

Leprabazillen aus

 

 

 

einen elie nach x 9
Behandlung mit %
10°/o Antiformin
Paratyphus B.
Reinkultur ts aS
ok §,
Girtner’sche Abe kh e
Enteritisbazillen f- A , "
Reinkultur ‘ef A
ay aot “

 

 

 

 

 

 

 

 

 

 

 

 

Spirochaeta gallinar. © p ' 9 Maa,
( Z
s +
Mundspirochiaten ) g
(Stomatitis mercur.) e ( 3
b ae
is is OR ( Le ee
( “
Spirill. rubr, ) GP ae ue
& a, f pe
4, ate a g e Han
Spirochaeta pallida ) 4 7 9 Geel , Nel
~ Reinkultur { } e 5 3 Weare ‘
/ “He A ey
, a <
160266°—21——21 at

 
Fie. 181

318

182.
183.
184.
185.
186.
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.

IL

PLATE XV.

. (p. 117): Plasmoptysis. 1. Fischer (1906, original fig. ITI, 8). 1,500.

(p. 124): B. coli. Kellerman and Scales (1916, original fig. 24). 1,000.

(p. 124): Bac. no. 41. Léhnis and Smith (1916 a, original fig. 39). 1,000.

(p. 128): Corynebacterium. £. de Negri (1916, original fig. 27) 560.

(pp. 128, 202): B. coli. Kellerman and Scales (1916, original fig. 23). 1,000.
(p. 128): B. Azotobacter. Léhnis and Smith (1916 a, original fig. 3). 1,000.
(p. 128): B. Azotobacter. Léhnis and Smith (1916 a, original fig. 4). 1,000.
(p. 128): B. Azotobacter. Léhnis and Smith (1916 b, original fig. 1). 1,000.
(p. 126): Az. chroococcum. Beijerinck (1901 6, original fig. 4). 1,000.

(p. 129): Az. vitreum. Léhnisand Smith (1916 a, original fig. 6). 1,000.

(p. 127): B. Azotobacter. Léhnis and Smith (1916 a, original fig. 5). 1,000.
(pp. 126, 194): ‘‘Phthisiogenic microbe.’’ Schroen (1904, original fig. VIII). 500.
(p. 129): B. Azotobacter. Léhnis and Smith (1916 a, original fig. 21). 1,000.
(p. 132): “Syphilis Bacillus.’’ Miessen (1898, original fig. 13). 900.

(p. 132): B. radicicola. Hilter und Stérmer (1903, original fig. II, 1). 1,000.
(p. 132): B. radicicola. Ailter und Stérmer (1903, original fig. II, 2). 1,000.
(pp. 132, 136): B. subtilis. Léhnis and Smith (1916 }, original fig. 27). 1,000.
(p. 135): Nitrosococcus. Winogradsky (1891, original fig. XVIII, 2). 1,000.
(p. 135): B. coli. Kellerman and Scales (1916, original figs. 19 and 20). 1,000.
(p. 135): B. coli. Kellerman and Scales (1916, original figs. 21 and 22). 1,000.
(p. 185): V. cholerae. Stamm (1914, original fig. 20). 1,000.
National Academy of Sciences. Plate XV.

 

 

 

 

 

319
Fie. 202.
203.
204.
205.
206.
207.
208,
209.

320

(p.

145):
. 149):
. 149):
. 168):
. 168):
. 168):
. 168):
. 170):

PLATE XVI.

Microc. gonorrhoeae. Herzog (1913, original fig. III, 5). 1,000.

Hematocyts. Fokker (1887, original fig. I, 1). 770.

Bacilli from blood. Fokker (1888, original fig. III, 2). 770.

Zoogloea with embryonal areas. Bastian (1905, original fig. 12). 500.
Regeneration of cells. Bastian (1905, original fig. 11, B/C). 500.

Zoogloea in transformation. Bastian (1905, original fig. 13). 750.

Cell formation. Bastian (1907, original fig. 33). 500.

Macroplasts of B. rubescens. Lankester (1876, original fig. III, 8, 9, 15, 20). 1,100.
National Academy of Sciences. Plate XVI.

 

 

205 206

 

207 208

 

 

 

 

321
PLATE XVII.

Fic. 210. (p. 169): Endccarditis. Klebs (1878, original fig. II, 3).
211. (p.171): Septicaemia. R. Koch (1878, original fig. I, 8) .x700.
212. (p. 175): Plasmodia and filidia. W. Winkler (1899, original figs. 13, 15, 16, 34, 35, 40, 41, 48, 50) |
X800-1,000. .
213. (p. 176): Actinobacillosis. Ligniéres et Spitz (1902, original fig. 3 A and B).
322
Plate XVII.

National Academy of Sciences.

 

         

  

=
ee
ees

2 =.

 
 

212

 

 

 

 

323
Fie. 214.
215.
216.
217.
218.
219.
220.
221.
222:
223.
224,
225.
226.
227%
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.

324

(pp. 172,191): Str. lactis.
(pp. 175,191): Str. lactis.

PLATE XVIII.

(pp. 175, 190): M. candicans.

(pp. 175, 190): ‘‘Syphilis Bacillus.’
(p. 176): B. pestis.
(p. 176): B. pestis.
(p. 176): B. pestis.
(p. 176): B. pestis.
(p. 176): Pasteuria ramosa.
(pp.
(pp.
(pp.
(pp.
(pp.
(pp.

177, 186):
177, 192):
178, 182):
178, 182):
178, 182):
178, 182):
. 178, 182):
. 178, 182):
_ 178, 182):
. 178): B. fluorescens.
. 178): B. subtilis.
. 184): B. fluorescens.

. 185): B. Azotobacter.
. 192): Alnus nodule.

ww

BDawBnwuw

N. KE. Schultz (1901 a, original fig. VII).

N. K. Schultz (1901 a, original fig. VIII).
N. K. Schultz (1901 a, original fig. TX).

N. K. Schultz (1901 a, original fig. XTI).
Metchnikoff (1888 b, original fig. IIT).
. coli. Kellerman and Scales (1916, original fig. 29). 1,000.
. coli. Kellerman and Scales (1916, original fig. 7).
. Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
Léhnis and Smith (1916 a, original fig. 30). 1,000.

Léhnis and Smith (1916 a, original fig. 33). 1,000.
Léhnis and Smith (1916 a, original fig. 32). 1,000.
Léhnis and Smith (1916 b, original fig. 56). 1,000.
Niessen (1898, original fig. 13). 900.
1,000.

X1,000.
1,000.

X1,000.

850.

1,000.

Léhnis and Smith (1916 a, original fig. 9). 1,000.
Léhnis and Smith (1916 a, original fig. 7). 1,000.
Léhnis and Smith (1916 6, original fig. 41). 1,000.
Léhnis and Smith (1916 a, original fig. 8). 1,000.
Léhnis and Smith (1916 a, original fig. 11). 1,000.
Lohnis and Smith (1916 b, original fig. 43). 1,000.
Léhnis and Smith (1916 a, original fig. 12). 1,000.

Léhnis and Smith (1916 a, original fig. 27).
Léhnis and Smith (1916 a, original fig. 42). 1,000.
Léhnis and Smith (1916 a, original fig. 17). 1,000.
Hiltner und Stérmer (1903, original fig. II, 8). 1,000
. 179): Granulom. £. de Negri (1916, original fig. 62). 560.

1,000.
National Academy of Sciences. Plate XVIII.

 

 

 

 

 

325
Fie. 238.
239.
240,
241.

242.

243.

244,

245.
246.

247,
248.
249,

326

PLATE XIX.

(pp. 176,193): Ps. cerevisiae. Fuhrmann (1908, original. fig. I, 5). 1,800.
(pp. 176,181): B. erysipeloides. Rosenbach (1999, original fig. XIV, 6). 970.
(pp. 176, 186): B. erysipeloides. Rosenbach (1909, original fig. XII, 2). 970.
(p. 176): B. erysipeloides. Rosenbach (1909, original fig. XIII, 2). 970.

(pp. 176,181): B. erysipeloides. Rosenbach (1909, original fig. XII, 1). 270.
(pp. 177, 187): B. typhi. Almguist (1916, original fig. 12). 1,000.

(pp. 177,193): V. cholerae. Almguist (1916, original fig. 22). 1,000.

(pp. 177,194): B. diphtheriae. Almquist (1917, original fig. 5). 1,000.

(pp. 177,192): B. acidi propionici c. Almquist (1917, original fig. 6). 1,000.
(pp. 177,184): Bact. antityphosum. Almguist (1917, original fig. 12). 1,000.
(pp. 179, 184): Clumps from tonsils... Mallory and Medlar (1916, original fig. XX, 23). 1,000.
(pp. 179,194): Diphtheroids from lung. Mellon (1917, original fig. 5, pl. 2).
National Academy of Sciences. Plate XIX.

 

a ee sy

Sieg eg

 

 

 

 

327
Fic. 250.

328

251.
202.
233.
204.
255.
256.
257.
258.
259.
260.
261.

(p. 178): B. Azotobacter.
(p. 178): B. Azotobacter.
(p. 178): B. Azotobacter.
(pp. 178, 185):
(pp.
(pp.
(pp.
(pp.
(pp.
(pp.
(pp.
(pp.

B.

178,185): B

178, 185):
172, 185):
172, 185):
178, 185).
178, 185):
178, 185):
178, 185):

Bown

PLATE XX.

Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
. Azotobacter.
. fluorescens.
. fluorescens.
. fluorescens.

Lohnis and Smith (1916 b, original fig. 44).
Léhnis and Smith (1916 a, original fig. 19).
Léhnis and Smith (1916 a, original fig. 18).
Léhnis and Smith (1916 b, original fig. 47).
Léhnis and Smith (1916 a, original fig. 14).
Lohnis and Smith (1916 b, original fig. 46).
Léhnis and Smith (1916 a, original fig. 15).
Léhnis and Smith (1916 b, original fig. 48).
Loéhnis and Smith (1916 a, original fig. 13).

Léhnis and Smith (1916 6, original fig. 49).
Léhnis and Smith (1916 6, original fig. 51).
Lohnis and Smith (1916 b, original fig. 50).

1,000.
1,000.
1,000.

X<1,000.
1,000.
1,000.
1,000.
1,000.
X1,000.

1,000,

1,000.

X1,000.
Plate XX,

National Academy of Sciences.

 

 

 

ad

On

 

260

 

 

329
Fie. 262.
263.
264.
265.
266.
267.
268.
269.
270.
271.
272.
273.

330

PLATE XXTI.

(pp. 178, 185): B. fluorescens. Lédhnis and Smith (1916 6, original fig. 8). 1,000.
(pp. 178, 185): B. fluorescens. Léhnis and Smith (1916 5, original fig. 63). 1,000.
(pp. 178, 190): Sarcina flava. Léhnis and Smith (1916 b, original fig. 53). 1,000.
(pp. 135,193): Bact. acidulum. Jéhnis and Smith (1916 b, original fig. 52). 1,000.
(pp. 65,193): B. malabarensis. Léhnis and Smith (1916 b, original fig. 4). 1,000.
(pp. 174, 186): Yellow bacillus no. 41. Léhnis and Smith (1916 a, original fig. 28). 1,000.
(p. 193): Nitrosomonas. Omelianski (1899, original fig. 2). 1,000.

(p. 193): Nitrosomonas. Winogradsky (1892, original fig. 11). 1,000.

(p. 193): Nitrosomonas. Winogradsky (1892, original fig. 15). 1,000.

(p. 192): Clover nodule. Hiltner und Stérmer (1903, original fig. II, 5). 1,000.

(p. 192): Pea nodule. Hiltner und Stérmer (1903, original fig. II, 6). 1,000.

(p. 192): Clover nodule. Léhnis and Smith (1916 a, original fig. 34). 1,000.
National Academy of Sciences. Plate XXI.

 

=

 

266 267

 

268 269 utd

F
a
“ye

 

  

 

 

331
Fie. 274.

332

275.
276.
277.
278.
279.
280.
281.
282.
283.
284.
285.

(p. 186):
(p. 190):
(p. 190):
(p. 200):
(p. 197):
(p. 197):
(p. 197):
(p. 197):
(p. 197):
(p. 197):
(pp. 197,
(p. 197):

PLATE XXII.

Rosette jormation. M. Jones (1905, original fig. 1). 1,500.
Ascococcus Billrothii. F. Cohn (1875, original fig. V, 5). x65.

M. botryogenus. Rabe (1886, original fig. IV, 2). 100.

B. radiobacter. Léhnis and Smith (1916 b, original fig. 29). 1,000.
Chromatium Okenii. ocrster (1892, original fig. 26). 750.

B. Azotobacter. Léhnis and Smith (1916 a, original fig. 1). 1,000.
B. Azotobacter. Léhnis and Smith (1916 a, original fig. 2). 1,000.
B. Azotobacter. Léhnis and Smith (1916 b, original fig. 18). 1,000.
B. Azotobacter. Léhnis and Smith (1916 b, original fig. 19). 1,000.
B. Azotobacter. Léhnis and Smith (1916 b, original fig. 25). 1,000.
204): B. Azotobacter. Léhnis and Smith (1916 6, original fig 20). 1,000.
B. fluorescens. Léhnis and Smith (1916 b, original fig. 23). 1,000.
National Academy of Sciences. Plate XXII.

 

 

 

 

 

333

 

160266°—21 22
Fria. 286.
287.
288.
289.
290.
291.
292.
293.
294.
295.
296.
207,

334

PLATE XXIIT.

(p. 197): B. subtilis. Léhnis and Smith (1916 b, original fig. 22). 1,000.

(p. 197): B. subtilis. Léhnis and Smith (1916 5, original fig. 21). 1,000.

(p. 197): B. Chauvoei. Iterott und Niemann (1895, original fig. 48). 1,000.

(p. 197): B. fluorescens. Axelrad (1903, original fig. 22). 1,000.

(pp. 197,201): Bact. esterificans. Maassen (1899, original fig. XI, 1). 1,000.

(p.197): Bacilli from straw. Gnther (1906, original fig. 8). 1,000.

(pp. 197, 202): B. subtilis. Hiss and Zinsser (1914, original fig. 125).

(p. 197): B. subtilis. Hiss and Zinsser (1914, original fig. 125).

(p. 197): Azotob. chroococcum. Walton (1915, original fig. II, 8). 1,000.

(p. 208): Azotobacter cells not heated. Léhnis and Smith (1916 b, original fig. 57). 1,000.
(p. 208): Azotobacter cells heated. Léhnis and Smith (1916 b, original fig. 58). 1,000.
(p. 211): Flake from agar. Léhnis and Smith (1916 b, original fig. 61). 1,000.
National Academy of Sciences.

Plate XXIII.

 

 

 

 

He
¢

rk
o 4.”
\ ,

is

   

a
or o.
288
SEEMS A,
TMS
= SEW INSS \
See ees NAY \\Wa
ee Nil) yf
eZ AWN
(LN Hy
WES wy! NS
\ 4 Wy it aie
SEE GE
YU pay,
eee ZF
RWW ee
MS

es :

 

 

335
 

 

NATIONAL ACADEMY OF SCIENCES.

 

Volume XVI.
THIRD MEMOIR.

 

THE CONSTANTS OF NATURE.

Parr V.

A RECALCULATION OF THE ATOMIC WEIGHTS.

FOURTH EDITION, REVISED AND ENLARGED.

BY

FRANK WIGGLESWORTH CLARKE, LL. D., D. So.,

CHEMIST, U. 8. GEOLOGICAL SURVEY.

 

 
TABLE OF CONTENTS.

Page
Introductions. excswes sacuciecoes sweets sepeceewees < 5
Formule for the calculation of probable error -....-. 9
The fundamental ratios, O, H, Ag, Cl, Br, I, K, Na,
TE, ON Ss Caccadaewk sec sartodoenneeeaeGcaaadacuen 10
The oxygen-hydrogen ratio ...............-.--. 11
The nitrogen-oxygen ratio........-..--------- 31
The carbon-oxygen ratio..........-.---..------ 4l
Syntheses and density ofhy drochloric acid ....- 48
Analyses and density of nitrosyl chloride .-... 52
The ratio HCl: NH3......-.--------------- eee 53
Syntheses and density of hydrobromicacid. .-. 54
Iodine pentoxide ratios............--..--.----- 56
Analyses of sodium halides.............------- 57
Analyses of chlorates.-.............---------- 58
Analyses of bromates and iodates..........-.. 61
Thesilver-chlorine ratio..............-.---..-- 62
The silver-bromine ratio...........-...------- 71
The silver-iodine ratio..............--------- 80
Ratiosconnecting the silver halides.........-.. 84
The potassium chloride-silver ratios .........- 87
Potassium bromide and iodide ratios.........- 91
The sodium halide-silver ratios..........------ 92
The ammonium halide-silver ratios...........- 94
The silver nitrate ratios ........-..-.-.-+----- 96
Potassium and sodium nitrate ratios..........- 99
The lithium ratios.......-..----------------- 101
The silver-carbon ratios.........----.-------- 104
The sulphur ratios.............-..-2-22-0.2-5- 109
General discussion......... ded oecedaaateees 114
Rubidititisscc2s-jonaserke ecceeeeseeeseasaeaseeans 120
CRS i's Sdn ga eocse ce een oe Sees sate 123
Coppel is ccncsiseenenseoe a redumamecs as’. donee neces 125
GOl dixie Ss sae epeuavtint see teeieedeeaes <4 s eae eey 134
Cale: sasnses 32 eeesc ene eewSerlesielne ices os dee'ss 140
SHONHUM sang be cael dese nsedscp nites sr ies ot ce 145
Barut clo etfs sos og carsecie eh ost ciblaspeteap ets gies 149
Radic. acicices Gace acvas oeeeceebeceeneseeus ce 156
Vieads osccemieeeoss tetamonseetenteweneetedume sts 158
GlucintiMn +2352 sie. eee dexewits seccewnsneae hits 170
Magnesium 20s c3 eves posses ouscidteneeberdde disci 172
ZiDC oversee teweeieee ess sess 55 Meee eee sees ess 179
Cadmiumis.).532 osscsee vey eee bs chcieeeeeen eae 187
MGLCURY sacs. aaenainnieideiek Cine creas aowaeeele 200
BOrobeeeecseesescases saceceeue shoes oseemneaeerse 205
AVGMINUM osceed sabes ce tceGecioiee ss ose seseetesas 211
Gallhumesccsssxensiwesaeecseeecenie sess xd andyes 216
Unda: owtapseasisaeeieiedeee tiwsceaiiese + sie siesta 216
Thallium: 2:04.05 cs scscsoesssawsedecwcweeees secs 218

 

Page.
Silicone. screws soos no astnedeciosemes seeadgoesacnts 222
TitANM 32:05. ioecnces sete ee ewes ae ves ceeeaeeees 224
Ger MADIUM 35/3265 0s ovccacins SERS eee eeeTesse 226
AT COMMUD sig arg.sseoseaCaisoaesneasene sn.c ae nee «ee 226
IT ATA ios rclck tes ceapes ge vst ake c eisrep d speieee ieee oe Se 228
Thorium (Ionium)........-....----2---2e-eee eee 233
Phosphoriss:is9s4 0's aegaaes scabies e's sole enate 236
Vanadiumicces:ecetececc de se cxeeesseeebecee ess ee 240
APSENICS 2.25 sbedakisios Se Saeed dae ee see eee eS 243
ANIGIMODY $312.02 tc eaes aes ease meee eee 247
BiBMUtH aint gcaddcs awanweeaseweeetesteeteaasaaes 256
Colimbitin win5 5 cones ee ete ce sewn wenn amass 260
"Pan tals sea dcyara sctocs cletrape teed Saeielicioe eieiere aon 262
Chromium seo ec8 occas ecm hevalnawieleiNselsler o's's's'shetsietets 264
Molybdenum ssc c0s< eeecns Koscceece cess oes shane 272
TUNPStEN <.scisdscujeeieivs So snneeeerate ce shes see 276
ID PADME 2 so2ee scien 62242 Seneeins Coosa eee sess 285
Selenite: occ. cocccasionwnnaagecwiawceiceeet.csans 292
TOMUPUM senses caedsence ce onwescesGeeceaceessns 297
NOM G7 so Sees ek so oe es mee eee eS 312
Man Panese ic.csic55.5 xis havebiare vic's soles hecdwoiseesisale e's 316
TiOD is siiaiwes ccd pameee sins 1a eheaheeeeeescne seeds 321
Nickel and cobalt ..............-..0-se0eeeeeeeee 327
Ruthenium s:. gc.523ccecaches we sascmacess senesece ces 345
Rhodium: -2:262c¢scecseee voce vannenasiceecaesane 346
Palladium 2ccscsees icecccesse ss soeasescdssdeew ae 348
OSMIUM fected es tkeclcieictans ope senna: ab Seelxice Sivals 358
Dr iin sg sh oe eras ns recat e nde 4 Sar oe eee soe 360
Platinutisc< sceseete dessa sess cs Seeeeeriaesic ee ses 362
DCANGAUM <5. occSacietis Srswie tices Seqqseetbiniiwieeaeeie 80s 372
MAMI Ss sinccineititien nadie scat oneanseoeeedeecans 374
anthan win ss iecsestace shea cceds anccaccescewsass 378
COT os Sa SEA is Bd Soe es 384
Praseod Vy MIUMs 2.552 sieelsess oi eS caonsekewecws 390
Neod Ym sieves cesscieecseccsesrscreceaewee naeewts 394
SAMATIUM : siA5 < vis, h tee ereeg se seieic hints SS wie ietand austeee Siaualove 397
EUPOPLU Mi fss-ccyt aes sisia eco Serta 4 8d ease ieee 400
Gad olmitiec. <<<cccmsoceseeccednsehattaceeneuus 401
Terbium........-...-..- Piasemathi ates senihesaie s 403
DyaprosliM sjcscancasesvoseeekginsd sess eeeeece ae 403
Holmium: sees es & sraseires eect sS ciceeaeeins 404
Erbium, Thulium..................0..0. 02. e eens 405
Ytterbum and Lutecium (Aldebaranium and
Cassiopeium)......-----.-2.- 2c eee ee ence ee eee 407
The Helium-Argon Group................-..--.--- 408
Table of Atomic Weights.....................- 411
INDO Xs ssGed es cece Ads i ase ae reac sausaneeawee 413
A RECALCULATION OF THE ATOMIC WEIGHTS.

(Fourts Epition, REVISED AND ENLARGED.)

By Frank WiaeLeswortH CLARKE.

INTRODUCTION.

In the autumn of 1877 the writer began collecting data relative to determinations of atomic
weight, with the purpose of preparing a complete résumé of the entire subject, and of recalculat-
ing all the values. The work was fairly under way, the material was collected and partly
discussed, when I received from the Smithsonian Institution a manuscript by Prof. George F.
Becker, entitled “Atomic Weight Determinations: A Digest of the Investigations Published
since 1814.”’ This manuscript, which has since been issued * as Part IV of the “Constants of
Nature,” covered much of the ground contemplated in my own undertaking. It brought
together all the evidence, presenting it clearly and thoroughly in compact form; in short, that
portion of the task could not well be improved upon. Accordingly, I decided to limit my own
labors to a critical recalculation of the data; to combine all the figures upon a common mathe-
matical basis, and to omit everything which could as well be found in Prof. Becker’s “ Digest.”’

In due time my work was completed, and early in 1882 it was published.?, About a year
later Meyer and Seubert’s recalculation appeared, to be followed later still by the less elaborate
discussions of Sebelien and of Ostwald. All of these works differed from one another in various
essential particulars, presenting the subject from different points of view, and with different
methods of calculation. Each one, therefore, has its own special points of merit, and, in a
sense, reinforces the others. At the same time, the scientific activity which they represent
shows how widespread was the interest in the subject of atomic weights, and how fundamen-
tally important these constants undoubtedly are.

The immediate effect of all these publications was to render manifest the imperfections
of many of the data, and to point out most emphatically in what directions new work is needed
to be done. This led to an extraordinary activity in the determination of atomic weights,
and so much new material accumulated that in 1897 a second edition of the work, revised to
date, became necessary. A third edition,‘ much enlarged, appeared in 1910, and since then
a large amount of new work has been done, with great improvements in technique, especially
by. Richards and his colleagues at Harvard University, by Edgar F. Smith in Philadelphia,
and by Guye at Geneva, not to mention many other workers of high merit. The assimilation
of this new material, and its combination with the older data, is the object of the present
volume.

At the very beginning of my work, a fundamental question confronted me. Should I
treat the investigations of different individuals separately, or should I combine similar data
together in a manner irrespective of persons? For example, ought I, in estimating the atomic
weight of silver, to take Stas’s work by itself, Marignac’s work by itself, and so on, and then
average the results together; or should I rather combine all series of figures relating to the

composition of potassium chlorate into one mean value, and all the data concerning the com-
position of silver chloride into another mean, and, finally, compute from such general means
the constant sought to be established? The latter plan was finally adopted; in fact, it was

 

1 Smithsonian Miscellaneous Collections, vol. 27, serial No. 358, pp. 152.

2 Smithsonian Miscellaneous Collections, vol. 27, serial No. 441, pp. 279.

8 Smithsonian Miscellaneous Collections, vol. 38, serial No. 1075, pp. vi, 370.
4Thid., vol. 54, No. 3, pp. iv, 548.
6 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

rendered necessary by the method of least squares, which, in a special, limited form, was
chosen as the best method of dealing with the problem.

The mode of discussion and combination of results was briefly as follows. The formule
employed are given in another place. Beginning with the ratio between oxygen and hydrogen,
each series'of experiments was taken by itself, its arithmetical mean was determined, and the
probable error of that mean was computed. “Then the several means were combined according
to the appropriate formula, each one receiving a weight dependent upon its probable error.
The general mean thus established was taken as the most probable value for the ratio, and at
the same time its probable error was mathematically assigned. In the first two editions of this
work it was used to give the atomic weight of oxygen referred to hydrogen as unity. In the
third edition the oxygen standard was adopted, and the atomic weight of hydrogen is referred
to it. This is an accordance with the decisions of the International Committee on Atomic
Weights; although my personal preference, on theoretical grounds, is for the hydrogen stand-
ard. The subsequent computations, however, are rendered simpler by assuming that O=16,
and that is another reason for my change of policy.

Next in order came a number of elements which were best considered together; namely,
silver, chlorine, bromine, iodine, potassium, sodium, nitrogen, sulphur, and carbon. Their atomic
weights, with those of hydrogen and oxygen, form a fundamental group, by means of which
other atomic weights are determined. Direct comparisons with oxygen or hydrogen are
relatively few; indirect determinations with the aid of silver and the halogens are many. For
the elements in question there were data from many experimenters. All similar figures, that
is, the figures for each ratio, were first reduced to a common standard, and then the individual
means were combined into general means. Thus all the data were condensed into seventy-
one ratios, from which a number of values for each atomic weight could be computed. The
ratios represent the actual experimental work; the atomic weights are inferential. Finally,
the several values for each atomic weight are treated as if they were means of the usual type,
and combined by the method of least squares into a general mean, which is suppoesd to repre-
sent the most probable value for each constant. The fundamental values having been deter-
mined, they are next applied to the calculation of what may be called the secondary atomic
weights, and in this work the probable error of each term in each ratio is taken into account.
This will appear more clearly evident in the subsequent actual calculations.

But although the discussion of atomic weights is ostensibly mathematical, it cannot be
purely so. Chemical considerations are necessarily involved at every turn. In assigning
weights to mean values I have been, for the most part, rigidly guided by mathematical rules;
but in some cases I have been compelled to reject series of data which were mathematically
excellent but chemically worthless because of constant errors. In certain instances there
were grave doubts as to whether particular figures should be included or rejected in the calcu-
lation of means, there having been legitimate reasons for either procedure. Probably many
chemists would differ with me upon such points of judgment. In fact, it is doubtful whether
any two chemists, working independently, would handle all the data in precisely the same way,
or combine them so as to produce the same final results. Neither would any two mathema-
ticians follow identical rules or reach identical conclusions. In calculating the atomic weight
of any element those values are assigned to other elements which have been determined in
prévious chapters. Hence a variation in the order of discussion might lead to slight differences
in the final results.

_ As'a matter of course the data herein combined are of very unequal value. In many
series of experiments the weighings have been reduced to a vacuum standard; but in other
cases this correction has been neglected. In a majority of instances the errors thus introduced
are slight; nevertheless they exist, and interfere more or less with all attempts at a theoretical
consideration of the results.’

 

1 For a discussion of these vacuum corrections see Guye and Zachariades, C. R. Acad. Sci., 149, 598. Theerrorsin reductions to a vacuum are
larger than has been commonly supposed. Forareply by Richards and Baxter see Journ. Amer. Chem. Soc., 32, 507. Also Guyeand Renard,
Journ. Chim. Phys., 14, 55, and Guye, pp. 25, and 83. In thesame volume, pp. 195, 204, Guyeand Germann discuss other sources of error in atomic
weight determinations. See also Van Laar (Ibid., 17, 266, 1919) on the calculation of atomic weights.
No. 3.] ATOMIC WEIGHTS—CLARKE. 7

Necessarily, this work omits many details relative to experimental methods, and particulars
as to the arrangement of special forms of apparatus. For such détails original memoirs must
be consulted. Their inclusion here would have rendered the work unwarrantably bulky. There
is such a thing as overexhaustiveness of treatment, which is equally objectionable with under-
thoroughness.

Of course, none of theresults reached in this revision can be considered as final. Every one
of them is liable to future corrections. To my mind the real value of the work, great or little,
lies in another direction. Thedata have been brought together and reduced to common stand-
ards, and for each series of figures the probable error has been determined. Thus far, however
much my methods of combination may be criticized, I feel that my labors will have been useful.
The ground is cleared, in a measure, for future experimenters; it is possible to see more distinctly
what remains to be done; some clues are furnished as to the relative merits of different series of
results.

On the mathematical side my method of recalculation has obvious deficiencies. It is special
rather than general, and at some future time, when a sufficiently large mass of evidence has
accumulated, it must give way to a more thorough mode of treatment. For example, the
ratio Ag, : BaBr, has been used for computing the atomic weight of barium, the atomic weights
of silver and bromine being supposed to be known. But these atomic weights are subject to small
errors, and they are superimposed upon that of the ratio itself in the process of calculation.
Obviously, the ratio should contribute to our knowledge of all three of the atomic weights
involved in it, its error being distributed into three parts instead of appearing in one only. The
errors may be in part compensatory; but that is not certainly known.

Suppose now that for every element we had a goodly number of atomic weight ratios, con-
necting it with at least a dozen other elements, and all measured with reasonable accuracy.
These hundreds of ratios could then be treated as equations of observation, reduced to linear
form, and combined by the general method of least squares into normal equations. AI errors
would thus be distributed, never becoming cumulative; and the normal equations, solved once
for all, would give the atomic weights of all the elements simultaneously. The process would
be laborious, but the result should be the closest possible approach to accuracy. The data as
yet are inadequate, although some small groups of ratios may be handled in that way; but in
time the method is sure to be applied, and indeed to be the only general method applicable.
Even if every ratio was subject to some small constant error, this, balanced against the similar
errors of other ratios, would become accidental or unsystematic with reference to the entire
mass of material, and would practically vanish from the final means.

Concerning this subject of constant and accidental errors, a word may be said here. My
own method of discussion eliminates the latter, which are, in great part at least, removable by
ordinary averaging; but the constant errors, vicious and untractable, remain, at least partially.
Still, where many ratios are considered, even the systematic errors may in part compensate
each other, and do less harm than might be expected. They have, moreover, a peculiarity
which deserves some attention.

In the discussion of instrumental observations, the systematic errors are commonly con-
stant, both as to direction and as to magnitude. They are therefore independent of the acciden-
tal errors, and computation of means leaves them untouched. But in the measurement of chemi-
cal ratios the constant errors are most frequently due to an impurity in one of the materials
investigated. If different samples of a substance are studied, although all may contain the
same impurity, they are not likely to contain it in the same amount; and so the values found for
the ratio will vary. In other words, such errors may be constant in direction, but variable in
magnitude. That variation appears in the probable error computed for the series of obser-
vations, diminishes its weight when combined with other series, and so, in part, corrects itself.
It is not removed from the result, but it is self-mitigated. The constant errors familiar to the
physicist and astronomer are obviously of a different order.

That all methods of averaging are open to objections, I am, of course, perfectly aware. I
also know the doubts which attach to all questions of probable error, and to all combinations
8 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [VoL. XVI.

of data which depend upon them. I have, however, preferred to face these objections and to,
recognize these doubts rather than to adopt any arbitrary scheme which permits of a loose
selection of data. After all, the use of probable error as a means of weighting is only a means
of weighting, and perhaps more justifiable than any other method of attaining the same result.
When observations are weighted empirically—that is, by individual judgment—far greater
dangers arise. Almost unconsciously, the work of a famous man is given greater weight. than
that of some obscure chemist, although the latter may ultimately prove to be the best. But
the probable error of a series of measurements is not affected by the glamor of great names; and
the weight which it assigns to the observations is at least as good as any other. In the long run,
I believe it assigns weight more accurately, and therefore I have trusted to its indications, not
as if it were a mathematical fetish, but regarding it as a fairly safe guide, even though it is not
infallible.

One possibly weak point in the method adopted deserves to be mentioned. Its fairness
depends in part upon the fairness of the experimenter. One chemist, making a series of measure-
ments, gives all of his determinations. Another chemist selects those which are most concord-
ant, and suppresses others which seem to him less trustworthy. The latter series, therefore, is
likely to receive higher weight than belongs to it; while the former series will be underweighted.
The rejection of data, even by the man who is most familiar with them, is always a dangerous
proceeding, and one which should be discouraged.

The other and more usual method of adjusting the atomic weights, that of selecting deter-
minations in accordance with their apparent chemical merit, has recently been followed by
Brauner. In his excellent and critical discussion of the subject, now appearing in Abegg’s
Handbuch der anorganischen Chemie, he gives all the determinations for each element, and then
assigns preference to those which most appeal to his judgment. In most instances his findings
agree with mine, and therefore our conclusions reinforce each other. Sometimes we differ,
and in such cases it would seem that new determinations are desirable. When values derived
from different sources, and computed by different methods are concordant, they may be regarded
as probably well established; but even then certainty is not attained. The history of atomic
weight determinations bears abundant witness to this assertion.

For example: Until within very recent years the work of Stas, emphasized by that of
Marignac, was regarded as almost final. Now, however, some of the ratios measured by these
chemists are found to be out of harmony with the best modern investigations, and there is a tend-
ency toward rejecting the older work altogether. But the researches of Stas give a homogene-
ous and concordant group of atomic weights, which cannot be entirely thrown aside without
much more evidence against them than as yet exists. It is probable that the silver used by
Stas contained occluded oxygen, as was pointed out by Dumas;' and this would account for
some, but not all of the variations from recent revisions of the ratios. It is also probable, as
Richards has shown, that Stas underrated the solubility of silver chloride. How large these
errors may be in Stas’s work, assuming them to exist, is uncertain; and to assign zero weight to
his determinations would be too extreme a procedure. His data and Marignac’s are therefore
retained in the present recalculation, with the proper mathematical weight; and the final results
seem to be satisfactory. Indeed, the Stas values for silver, chlorine, and bromine, applied to
the determinations of other atomic weights, sometimes give more concordant results than the
modern figures. This is especially true in the cases of cesium, barium, and magnesium, although
the discrepancies are not large. The data of Richards and his collaborators for the chlorides of
these metals give a ratio between silver and chlorine in agreement with the measurements by
Stas, and lower than that which Richards and Wells have established. If, therefore, the work
of Stas is in error, the same error inheres in the atomic weights of the three metals above men-
tioned, and the latter, even if the uncertainty is small, ought to be revised. The sharp concord-
ance found when the atomic weights were computed with Stas’s figures is either illusive, or else
the modern data for silver and chlorine are wrong. The first of these alternatives is the more

 

1 Ann, Chim. Phys. (5), 14, 289. See also the Appendix to the first edition ofthis Recalculation, in which the influence of a correction for
occluded oxygen is considered.
No. 3.] ATOMIC WEIGHTS—CLARKKE. — 9

probable. In spite of the discordance now evident, the determinations for cesium, barium, and
magnesium are by far the best we have, and their uncertainties need not be regarded as serious.

In Meyer and Seubert’s recalculation, weights are assigned in quite a novel manner. In
each series of experiments the maximum and minimum results are given, but instead of the
mean there is a value deduced from the sum of the weighings—that is, each experiment is
weighted proportionally to the mass of the material handled in it. For this method I am
unable to find any complete justification. Of course, the errors due to the operations of
weighing become proportionally smaller as the quantity of material increases, but these errors,
with modern apparatus, are relatively unimportant. The real errors in atomic weight deter-
Minations are much larger than these, and due to different causes. Hence an experiment
upon ten grammes of material may be a little better than one made upon five grammes, but it
is by no means necessarily twice as good. The ordinary mean cf a series of observations, with
its measure of concordance, the probable error, is a better value than one obtained in the
manner just described. If only errors of weighing were to be considered, Meyer and Seubert’s
summation method would be valid, but in the presence of other and greater errors it seems
to have but little real pertinency to the problem at hand.

Although all reasonable efforts have been made to ensure the completeness of this work,
it is possible that some recent determinations of atomic weights have been missed. Because
of the war in Europe there has been great irregularity in the receipt of foreign journals, and
many numbers of German and Austrian publications have not yet come to hand. It is possible
that some of the missing journals may contain data which should be utilized here.

In addition to the usual periodicals, the following works have been freely used by me
in the preparation of this volume: ©

Berzeuvs, J: J. Lehrbuch der Chemie. 5 Auflage. Dritter Band. SS. 1147-1231. 1845.

Van Geuns, W. A. J. Proeve eener Geschiedenis van de Aquivalentgetallen der Scheikundige Grondstoffen
en van hare Soortelijke Gewigten in Gasvorm, voornamelijk in Betrekking tot de vier Grondstoffen de
Bewerktuigde Natuur. Amsterdam, 1853.

Mover, E. Historisch-Kritisch Overzigt van de Bepalingen der Avquivalent-Gewigten van 13 Eenvoudige
Ligchamen. Utrecht, 1853.

Muuper, L. Historisch-Kritisch Overzigt van de Bepalingen der Aquivalent-Gewigten van 24 Metalen. Utrecht

1853.
Ovpemans, A. C., Jn. Historisch-Kritisch Overzigt van de Bepaling der Mquivalent-Gewigten van Twee en
Twintig Metalen. Leiden, 1853.
Brecker, G. F. Atomic Weight Determinations: a Digest of the Investigations Published since 1814. Smith-
'  gonian Miscellaneous Collections, Vol. 27, No. 358. Washington, 1880.
Sras, J. 8. Untersuchungen iiber die Gesetze der Chemischen Proportionen tiber die Atomgewichte und ihre
gegenseitigen Verhaltnisse. Uebersetzt von Dr. L. Aronstein. Leipzig, 1867.
See also his ‘‘Oeuvres Complétes,’’ 3 vols., published at Bruxelles in 1894.1
Meyer, L., and Seusert, K. Die Atomgewichte der Elemente, aus den Originalzahlen neu berechnet. Leipzig,
1883.
SeBEuen, J. Beitrage ziir Geschichte der Atomgewichte. Braunschweig, 1884.
OstwatD, W. Lehrbuch der allgemeinen Chemie. Zweite Aufl. I Band. SS. 18-188. Leipzig, 1891.
Marienac, J.C. G. Dz. Oeuvres Complétes. 2 vols. Geneva, 1902.
Ricwarps, T. W. Experimentelle Untersuchungen ueber Atomgewichte. Hamburg and Leipzig, 1909.

Abegg’s Handbuch, containing Brauner’s recalculation, has already been mentioned.
Its value is very great. The four Dutch monographs above cited are also especially valuable.
They represent a revision of all atomic weight data down to 1853, as divided: between four
writers.
FORMULZ FOR THE CALCULATION OF PROBABLE ERROR.
The formula for the probable error of an arithmetical mean, familiar to all physicists, is
as follows:

(1.) ea 0.67454] ao

1 The citations used in the present Recalculation are all from the Oeuvres Completes.

 
10 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou: XVI.

Here n represents the number of observations or experiments in the series, and S the sum
of the squares of the variations of the individual results from the mean.

In combining several arithmetical means, representing several series, into one general
mean, each receives a weight inversely proportional to the square of its probable error. Let
A, B, C, etc., be such means, and a, 5, ¢ their probable errors respectively. Then the general
mean is determined by the formula:

4 +p+5 aah
(2.) = a7 ¢
at it e . ’
For the probable error of this general mean we have:
Poteet
3.) Vatitae

In the calculation of atomic and molecular weights the following formule are used: Taking,
as before, capital letters to represent known quantities, and small letters for their probable

errors, respectively, we have for the probable error of the sum or difference of two quantities,
A and B:

(4.) e= Var? +B?
For the product of A multiplied by B the probable error is
5.) e= (Ab) + Ba? -
For the poe of three quantities, ABC:
(6.) ~~ e= ¥(BCa)? + (ACB)? + (ABe)?
For a quotient, B, the probable error becomes
Se
(7.) _v x) +?
A
Given a proportion, A:B::C:z, the probable error of the fourth term is as follows:
es : wv PC) + Cort Bor
A

This formula is used in nearly every atomic weight calculation, and is, therefore, excep-
tionally important. Barely a more complicated case arises in a proportion of this kind: **
A:B::C+2:D+2
In this proportion the unknown quantity occurs in two terms. Its probable error is
found by this expression, and is commonly large:

(.) oem Saye SB a? + Am) + ee

 

When several independent values have been calculated for an atomic weight they are
treated like means, and combined according to formule (2) and (3). Each final result is, there-
fore, to be regarded as the general or weighted mean of all trustworthy determinations. This
method of combination is not theoretically perfect, but it seems to be the one most available

for present oses.
: per THE FUNDAMENTAL RATIOS.

In the determination of atomic weights, a small number of values are to be regarded as
fundamental. They are the standards of reference; and by comparison with them all the other
atomic weights are established. Two of these values, the atomic weights of hydrogen and
No. 3.] ATOMIC WEIGHTS—CLARKE. il

oxygen, are primary; that is, one or the other of them is the basis of the entire system; hydrogen
as unity in the older arrangements; oxygen equal to 16 in the more modern scheme. Over
the relative merits of these two ultimate standards there has been much controversy ; but with
discussions of that sort the present work has nothing to do. The oxygen standard is now
recognized by international agreement, and will therefore be accepted here.

Comparatively few of the atomic weights, however, are fixed by direct comparison with
either oxygen or hydrogen. In most cases other values intervene, and especially the atomic
weights of silver, chlorine, bromine, iodine, nitrogen, carbon, sulphur, potassium and sodium.
These constants are first to be determined, and their establishment may be compared to a
primary. triangulation, of which the hydrogen-oxygen ratio is the base line. The lithium
ratios may also be included here, for the reason that they give one of the best determinations
of the atomic weight of silver. The ratios connecting these 12 elements with one another
are to be discussed in the following pages.

THE OXYGEN-HYDROGEN RATIO.

Leaving out the account of the earliest researches, which now have only historical interest,
the first, determinations of this ratio worth considering are those by Dulong and Berzelius,’
who, like some of their successors, effected the synthesis of water over heated oxide of copper.
The essential features of the method are in all cases the same. Hydrogen is passed over the
hot oxide, and the water thus formed is collected and weighed. From this weight and the loss
of weight which the oxide undergoes, the composition of water is readily calculated. Dulong
and Berzelius made but three experiments, which gave the following percentages of oxygen
and hydrogen in water:

ls i.

88. 942 11. 058

88. 809 11.191

88. 954 11. 046
From these figures the ratio H : O becomes—
16. 124
15. 863
16. 106

Mean, 16.031, + .057.

As the weighings were not reduced to a vacuum, this correction was afterwards applied by
Clark,? who showed that these syntheses really make O=15.894; or, in Berzelian terms, if
O=100, H=12.583. The value 15.894,+.057 we may therefore take as the true result of
Dulong and Berzelius’ experiments, a figure curiously near that reached in the latest researches.

In 1842 Dumas* published his elaborate investigation upon the composition of water.
The first point was to get pure hydrogen. This gas, evolved from zinc and sulphuric acid,
might contain oxides of nitrogen, sulphur dioxide, hydrosulphuric acid, and arsenic hydride.
These impurities were removed in a series of wash bottles; the H,S by a solution of lead nitrate,
the H,As by silver sulphate, and the others by caustic potash. Finally, the gas was dried by
passing through sulphuric acid, or, in some of the experiments, over phosphorus pentoxide.
The copper oxide was thoroughly dried, and the bulb containing it was weighed. By a current
of dry hydrogen all the air was expelled from the apparatus. and then, for 10 or 12 hours,
the oxide of copper was heated to dull redness in a constant stream of the gas. The reduced:
copper was allowed to cool in an atmosphere of hydrogen. The weighings were made with the
bulbs exhausted of air.. The following table gives the results:

‘ Column A contains the symbol of the drying substance; B gives the weight’ of the bulb and
copper oxide; C, the weight of bulb and reduced copper; D, the weight of the vessel used for
collecting the water; E, the same, plus the water; F, the weight of oxygen; G, the weight of
water formed; H, the crude equivalent of H when O= 10,000; I, the equivalent of H, corrected

 

1 Thomson’s Annals of Philosophy, July, 1821, p. 50. 3 Phil. Mag. (3), 20, 341. ®C, R. Acad. Sci., 14, 537.
12 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

for the air contained in the sulphuric acid employed. This correction is not explained, and
seems to be questionable.

A, B. ¢. Dz. E. F. a. i. I.
HSOg: : i eiees 291.985 278.806 480.807 495.684 13.179 14.827 1250.5 1249.6
FSO j.c3.c6ec ee 344.548 324.186 488.227 511.132 20.362 22.905 1249.0 1248.0
H,SQ,.....202.5. 816.671 296.175 439.711 462.764 20.495 23.053 1248.1 1247.2
P,O5------------- 625.829 568.825 884.190 948.323 57.004 64.044 1250.6 1249.0
H,SQ,.....- aeet 804.546 728.182 887.331 973.291 76.364 85.960 1256.2 1254.6
H,SQ,......-.... 533.726 490.155 867.159 916.206 43.571 49.047 1256.3 1255.0
H,SQ,...... wei. 661.915 627.104 839.304 878.482 34.811 39.178 1254.6 1253.3
PLOpesee'esceces x 612.625 566.738 824.624 876.244 45.887 51.623 1250.0 1249.0
P.Og------ ++ eens 904.643 844.612 822.660 890.246 60.081 67.586 1258.3 1255.1
H,S04..-.0gc00se 642.325 590.487 741.095 799.417 51.8388 58.320 1250.4 1248.9
P,0,-.----------- 587.645 535.187 874.832 983.910 52.508 59.078 1251.2 1249.0
P,O,.------------ 673.280 613.492 931.487 998.700 59.789 67.282 1253.3 1250.8
H,SQ,.....-22.0. 660.855 598.765 682.374 752.273 62.090 69.899 1257.7 1254.8
HS0,.........2- 642.325 590.847 741.097 799.455 51.888 58.360 1258.1 1256.2
FSO cesenuciees 937.845 881.362 1064.762 1128319 56.483 63.577 1255.8 1252.2
PLO gecvsasgedleekt: 756.352 719.563 878.640 920.030 36.789 41.390 1250:6 1249.1
PO rer sctera dive 754.162 720.000 887.817 926.275 34.162 38.458 1257.3 1255.1
BO earnest \.... 759.762 727.632 888.662 924.887 32.183 36.175 1257.5 1254.7
PO; seccven tects 747.652 716.825 877.862 912.539 30.827 34.677 1248.8 1248.0
Means......... 1253.3 1251.5

In the sum total of these 19 experiments, 840.161 grammes of oxygen form 945.439 grammes
of water. This gives, in percentages, for the composition of water—oxygen, 88.864; hydrogen,
11.136. Hence the ratio H : O, calculated in mass, is 1 : 15.9608. In the following the values
are deduced from the individual data given under the headings F and G:

15. 994 15. 984
16. 014 15. 958
16. 024 15. 902
‘15. 992 15. 987
15. 916 15. 926
15. 916 15. 992
15. 943 15. 904
16. 000 15. 900
15, 892 16. 015
15. 995

Mean, 15.9607. +: .0070

In calculating the above several discrepancies were noted, probably due to misprints in
the original memoir. On comparing columns B and C with F, or D and E with G, these anom-
alies chiefly appear. They were detected and carefully considered in the course of my own
calculations; and, I believe, eliminated from the final result.

The investigation by Erdmann and Marchand‘ followed closely after that of Dumas.
The method of procedure was essentially. that of the latter chemist, differing from it only in
points of detail. The hydrogen used was prepared from zinc and sulphuric acid, and the zinc,
which contained traces of carbon, was proved to be free from arsenic and sulphur. The copper
oxide was made partly from copper turnings and partly by the ignition of the nitrate. The
results obtained are given in two series, in one of which the weighings were not actually made
in vacuo, but were, nevertheless, reduced to a vacuum. In the second series the copper oxide
and copper were weighed in vacuo. The following table contains the corrected weights of
water obtained. and of the oxygen in it, with the value found for the ratio in a third column.
The weights are given in grammes.

 

1 Journ, prakt. Chem., 26, 461. 1842.
No. 3.]_ ATOMIC WEIGHTS—CLARKE. 13

First series.

Wt. Water. Wt. 0. Ratio.
62. 980 55. 950 15.917
95. 612 84, 924 15. 891
94, 523 84. 007 15.977
35. 401 31. 461 15.970

Mean, 15. 939+.014
Second series.

Wt. Water. wt. oO. Ratio.
41. 664 37.034 15. 996
44,089 89.195 16.018
53. 232 47.321 16.011
55. 636 49. 460 16.017

 

Mean, 16. 010-+.0036

The effect of discussing these two series separately is somewhat startling. It gives to
the four experiments in Erdmann and Marchand’s second group a weight vastly greater than
their other four and Dumas’s nineteen taken together. Forso great a superiority as this there
is no adequate reason; and it is highly probable that it is due almost entirely to fortunate
coincidences rather than to greater accuracy of work. We will, therefore, treat Erdmann
and Marchand’s experiments as one series, giving all equal weight, the mean now becoming
O=15.975+.0113. If we take the sum of the eight experiments, 483.137 grammes water and
429.352 grammes oxygen, and compute from these figures, then O = 15.966.

It would be easy to point out the sources of error in the foregoing sets of determinations,
but it is hardly worth while to do so in detail. A few leading suggestions are enough for present
purposes. First, there is an insignificant error due to the occlusion of hydrogen by metallic
copper, rendering the apparent weight of the latter a trifle too high. Secondly, as shown by
Dittmar and Henderson, hydrogen dried by passage through sulphuric acid becomes per-
ceptibly contaminated with sulphur dioxide. In the third place, Morley' has found that
hydrogen prepared ‘from zinc always contains carbon compounds not removable by absorp-
tion and washing. Erdmann and Marchand themselves note that their zinc contained traces
of carbon. Finally, copper oxide, especially when prepared by the ignition of the nitrate, is
very apt to contain gaseous impurities, and particularly occluded nitrogen.? Any or all of
these sources of error may have vitiated the three investigations so far considered, but it would
be useless to speculate as to the extent of their influence. They amply account, however, for
the differences between the older and the later determinations of the constant under discussion.

Leaving now out of account all measurements of the relative densities of hydrogen and
oxygen, to be considered separately later, the next determination to be noted is that published
by J. Thomsen in 1870.3 Unfortunately this chemist has not published the details of his work,
but only the end results. Partly by the oxidation of hydrogen over heated copper oxide,
and partly by its direct union with oxygen, Thomsen finds that at the latitude of Copenhagen,
and at sea level, one liter of dry hydrogen at 0° and 760 mm. pressure will form .8041 gramme
of water.. According to Regnault, at this latitude, level, temperature, and pressure, a liter of
hydrogen weighs .08954 gramme. From these data, O=15.9605. It will be seen at once that
Thomsen’s work depends in great part upon that of Regnault and is therefore subject to the
corrections recently applied by Crafts and others to the latter. These corrections, which
will be discussed further on, reduce the value of O from 15.9605 to 15.91. In order to com-
bine this value with others it is necessary to assign it weight arbitrarily, and as Thomsen made
eight experiments, which are said to be concordant, it may be fair to rank his determination
with that of Erdmann and Marchand and to°assume for it the same probable error. The

' value 15.91 +.0113 will therefore be taken as the outcome of Thomsen’s research.

 

1 Amer. Chem. Journ., 12, 469. 1890. ' 4See Richards’s work cited in the chapter on copper. ® Ber, Deutsch. chem. Ges., 3,928, 1870,
14 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

In 1887 Cooke and Richards published. the results of their elaborate investigation! These
chemists weighed hydrogen, burned it over copper oxide, and weighed the water produced.
The copper oxide was prepared from absolutely pure electrolytic copper, and the hydrogen
was obtained from three distinct sources, as follows: First, from pure zinc and hydrochloric
acid; second, by electrolysis, in a generator containing dilute hydrochloric acid and zinc-
mercury amalgam; third, by the action of caustic potash solution upon sheet aluminum. The
gas was dried and purified by passage through a system of tubes and towers containing potash,
calcium chloride, glass beads drenched with sulphuric acid, and phosphorus pentoxide. No
impurity could be discovered in it, and even nitrogen was sought for spectroscopically without
being found.

The hydrogen was weighed in a glass globe holding nearly 5 liters and weighing 570.5
grammes, which was counterpoised by a second globe of exactly the same external volume.
Before filling, the globe was exhausted to within 1 mm. of mercury and weighed. It was then
filled with hydrogen and weighed again. The difference between the two weights gives the
weight of hydrogen taken.

In burning, the hydrogen was swept from the globe into the combustion furnace by means
of a stream of air which had previously been passed over hot reduced copper and hot cupric
oxide, then through potash bulbs, and finally through a system of driers containing successively
calcium chloride, sulphuric acid, and phosphorus pentoxide. The water formed by the com-
bustion was collected in a condensing tube connected with a U tube containing phosphorus
pentoxide. The latter was followed by a safety tube containing either calcium chloride or
phosphorus pentoxide, added to the apparatus to prevent reflex diffusion. Full details as to
the arrangement and construction of the apparatus are given. The final results appear in three
series, representing the three sources from which the hydrogen was obtained. All weights
are corrected te a vacuum.

First series. Hydrogen from zine and acid.

Wt. of H. Wt. H,0. Ratio H:0.
- 4233 3. 8048 15.977
- 4136 3. 7094 15. 937
- 4213 8. 7834 15. 960
- 4163 38.7345 15, 941
4131 38. 7085 15. 954

Mean, 15. 954+.0048
Second series.— Electrolytic hydrogen.

.4112 3. 6930 15. 962
. 4089. 3. 6709 15.955
4261 3. 8253 15. 955
. 4197 3. 7651 15. 942
4144 3.7197 15.953

Mean, 15. 9534: .0022
Third serves.— Hydrogen from caustic potash.

42205 3. 7865 15. 943
4284 3. 8436 15. 944
- 4205 3.7776 15.967
43205 1 0, 8.8748 15. 937
4153 3. 7281 15. 954

4167 3. 7435 15. 967

: Mean, 15. 952+. .0035
Mean of all as one series, 15. 953+ .0020

 

Shortly after the appearance of this paper by Cooke'and Richards Lord Rayleigh pointed
out the fact, already noted by Agamennone, that a glass globe when exhausted is sensibly con-

 

1 Proc, Amer. Acad., 23,149. Amer. Chem. Journ., 10, 81.
No. 3.] ATOMIC WEIGHTS—CLARKE. 15

densed by the pressure of the surrounding atmosphere. This fact involves a correction to the
foregoing data, due to a change in the tare of the globe used, and this correction was promptly
determined and applied by the authors. By a careful series of measurements they found that
the correction amounted to an average increase of 1.98 milligrams to the weight of hydrogen
taken in each experiment. Hence O equals not 15.953, but 15.869, the probable error remain-
ing unchanged. The final result of Cooke and Richards’ investigation therefore is

O=15. 869 + .0020
Keiser’s determinations of the ratio were published almost simultaneously with those of

Cooke and Richards. He burned hydrogen occluded by palladium, .and weighed the water
so formed. In a preliminary paper? the following results are given:

Wt. of H. Wt. of H,0. Ratio H: 0.
- 65100 5. 81777 15. 873
- 60517 5. 41540 15. 897
- 33733 3. 00655 15. 822

 

Mean, 15. 864 + .015

Not long after the publication of the foregoing data Keiser’s full paper appeared.? Palla-
dium foil, warmed to a temperature of 250°, was saturated with hydrogen prepared from dilute
sulphuric acid and zinc free from arsenic. From 100 to 140 grammes of palladium were taken,
and it was first proved that the metal did not absorb other gases, which might contaminate
the hydrogen. Before charging, the foil was heated to bright redness in vacuo. After charg-
ing, the tube containing the palladium hydride was exhausted by means of a Geissler pump
to remove any nitrogen which might have been present. In the preliminary investigation
cited above, the latter precaution was neglected, which may account for the low results.

Between the palladium tube and the combustion tube a U tube was interposed, contain-
ing phosphorus pentoxide. This was to determine the amount of moisture in the hydrogen.
The combustion. tube was filled with granular copper oxide, prepared by reducing the com-
mercial oxide in hydrogen, heating the metal so obtained to bright redness in a vacuum, and
then reoxidizing with pure oxygen.

Upon warming the palladium tube, which was first carefully weighed, hydrogen was given
off and allowed to pass into the combustion tube. When the greater part of it had been burned
the tube was cut off by means of a stopcock and allowed to cool. Meanwhile a stream of nitro-
gen was passed through the combustion tube, sweeping hydrogen before it. This was followed
by a current of oxygen, reoxidizing the reduced copper; and the copper oxide was finally cooled
in a stream of dry air. The water produced by the combustion was collected in a weighed bulb
tube, followed by a weighed U tube containing phosphorus pentoxide.

A second phosphorus pentoxide tube served to prevent the sucking back of moisture from
the external air. The loss in weight of the palladium tube, corrected by the gain in weight of
the first phosphorus pentoxide, gave the weight of hydrogen taken. The gain in weight of the
two collecting tubes gave the weight of water formed. All weights in the following table of
results are reduced to a vacuum:

Wt. of H. Wt. #0. Ratio H: 0.

0. 34145 3. 06338 15. 943
- 68394 6. 14000 15. 955

‘i . 65529 5. 88200 15. 952
. 65295 5. 86206 15. 954

. 66664 5. 98116 15. 944

. 66647 5. 98341 15. 955

. 57967 5. 20493 15. 958

- 66254 5. 94758 15. 952

. 87770 7. 86775 15. 950
77215 6. 93036 15. 951

 

Mean, 15.9514 + .0011

 

1 Proc, Amer. Acad., 23, 182. Amer. Chem. Journ., 10, 191. 3 Berichte, 20, 2323. 1887. 3 Amer. Chem. Journ., 10, 249. 1888.
161566 °—20——_2
16 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Tn sum, 6.55880 grammes of hydrogen gave 52.30383 of water, whence O= 15.9492.

In March, 1889, Lord Rayleigh published a few determinations of the ratio obtained by
still a new method. Pure hydrogen and pure oxygen were both weighed in glass globes. From
these they passed into a mixing chamber, and thence into a eudiometer, where they were gradu-
ally exploded by a series of electric sparks. After explosion, the residual gas remaining in the
eudiometer was determined and measured. The results, given without weighings or explicit
details, are as follows:

15. 93 15. 93
15. 98 15, 92
15. 98

Mean, 15.948 + .009

Correcting this result for shrinkage of the globes and consequent change of tare, it becomes
O=15.89, + .009.

In the same month that Lord Rayleigh’s paper appeared, W. A. Noyes? published his first
series of determinations. His method was to pass hydrogen into an apparatus containing hot
copper oxide, condensing the water formed in the same apparatus, and from the gain in weight
of the latter getting the weight of the hydrogen absorbed. The apparatus devised for this pur-
pose consisted essentially of a glass bulb of 30 to 50 cc. capacity, with a stopcock tube on one
side and a sealed condensing tube on the other. In weighing, it was counterpoised by another
apparatus of nearly the same volume, but somewhat less weight, in order to obviate reductions
toavacuum. After filling the bulb with commercial copper oxide (90 to 150 grammes), the ap-
paratus was heated in an air bath, exhausted by means of a Sprengel pump, cooled, and weighed.
It was next replaced in the air bath, again heated, and connected with an apparatus delivering
purified hydrogen. When a suitable amount of the latter had been admitted, the stopcock
was closed, and the heating continued long enough to convert all gaseous hydrogen within it
into water. The apparatus was then cooled and weighed, after which it was connected with a
Sprengel pump, in order to extract the small quantity of nitrogen which was always present.
The latter was pumped out into a eudiometer, where it was measured and examined. The
gain in weight of the apparatus, less the weight of this very slight impurity, gave the weight of
hydrogen oxidized.

The next step in the process consisted in heating the apparatus to expel water, and weighing
again. After this, pure oxygen was admitted and the heating was resumed, so as to oxidize
the traces of hydrogen which had been retained by the copper. Again the apparatus was
cooled and weighed, and then reheated, when the water formed was received in a bulb filled with
phosphorus pentoxide, and the gaseous contents were collected in a eudiometer. On cooling
and weighing the apparatus, the loss of weight less the weight of gases pumped out, gave the
amount of water produced by the traces of residual hydrogen under consideration. This
weight, added to the loss of weight when the original water was expelled, gives the weight of
oxygen taken away from the copper oxide. Having thus the weight of hydrogen and the weight
of oxygen, the ratio sought for follows. Six results are given, but as they are repeated, with
corrections, in Noyes’ second paper, they need not be considered now.

Noyes’ methods were almost immediately criticized by Johnson, * who suggested several
sources of error. This chemist had already shown in an earlier paper‘ that copper reduced in
hydrogen persistently retains traces of the latter, and also that when the reduction is effected
below 700°, water is retained too. The possible presence of sulphur in the copper oxide was
furthermore mentioned. Errors from these sources would tend to make the apparent atomic
weight of oxygen (referred to hydrogen as unity) too low.

In his second paper *® Noyes replies to the foregoing criticisms, and shows that they carry
no weight, at least so far as his work is concerned. He also describes a number of experiments
in which oxides other than copper oxide were tried, but without distinct success, and he gives
fuller details as to manipulations and materials. His final results are in four series, as follows:

 

1 Proc. Roy. Soc., 45, 425. 3 Chem. News, 59, 272. 6 Amer. Chem. Journ., 12, 441. 1890.
2 Amer. Chem. Journ., 11, 155. 1889. 4 Journ. Chem, Soc., May, 1879.
No. 3.] ATOMIC WEIGHTS—CLARKE. 17
First Series. —Hydrogen from Zine and Hydrochloric Acid.

Wt. of H. Wt. of O. Ratio H: 0.
. 9443 7. 5000 15, 885
. 6744 5. 3555 15. 882
. 7866 6. 2569 15. 909
. 5521 4, 3903 15. 904
4274 3. 3997 15. 909
. 8265 6. 5686 15. 895

 

Mean, 15.8973 + .0032
This series appeared in the earlier paper, but with an error which is here corrected.

Second Series. —Electrolytic Hydrogen, Dried by Phosphorus Pentowide.

Wi. of H. Wt. of O. Ratio H:0.
0. 5044 4.0095 15. 898
. 6325 5. 0385 15. 982
. 6349 5. 0517 15. 918
. 5564 4.4175 15. 879
. 7335 5. 8224 15. 876
. 6696 5. 3181 15. 885

Third Series —Electrolytic Hydrogen, Dried by Passage Through a Tube Packed with Sodium Wire.

 

Mean, 15. 8971-40. 0064

Wt. of H. Wt. of O. Ratio H: 0.
0. 9323 7.4077 15. 891
9952 7.9045, 15. 885
3268 2.5977 15. 898
7907 6. 2798 15. 884
_ 7762 6.1671 15. 891
1.1221 8. 9131 15. 887

 

Mean, 15. 8893-40. 0014

At the end of this series it was found that the hydrogen contained a trace of water,
estimated to be equivalent to an excess of 3 milligrams in the total hydrogen of the six
experiments. Correcting for this, the mean becomes O = 15.899.

Fourth Series—Electrolytic Hydrogen, Dried over Freshly Sublimed Phosphorus Pentoxde.

Wt. of H. Wt. of O. Ratio H:0.
1. 0444 8.3017 15. 898
. 7704 6. 1233 15. 896
- 8231 6. 5421 15. 896
. 8872 7. 0490 15. 890
. 9993 7. 9403 15. 892
1.1910 9. 4595 15. 885

 

Mean, 15. 8929+0. 0013

The mean of all the 24 determinations, taken as one series, with the correction to the third
series included, is O=15.8966+0.0017. In sum, there were consumed 18.5983 grammes of
hydrogen and 147.8145 of oxygen; whence 0 = 15.8955.

Dittmar and Henderson,! who effected the synthesis of water over copper oxide by what
was essentially the old method, begin their memoir with an exhaustive criticism of the work
done by Dumas and by Erdmann and Marchand. They show, as I have already mentioned,
that hydrogen dried by sulphuric acid becomes contaminated with sulphur dioxide, and also
that a gas passed over calcium chloride may still retain as much as 1 milligramme of water per
liter. Fused caustic potash they found to dry a gas quite completely.

 

Z Proc. Roy. Soc. Glasgow, 22, 33. Communicated Dec. 17, 1890.
18 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI

In their first series of syntheses, Dittmar and Henderson generated their hydrogen from
zinc and acid, sometimes hydrochloric and sometimes sulphuric, and dried it by passage first
through cotton wool, then through vitrioled pumice, then over red-hot metallic copper to
remove oxygen. In later experiments it first traversed a column of fragments of caustic soda
to remove antimony derived from the zinc. The oxide of copper used was prepared by heating
chemically pure copper clippings in a muffle, and was practically free from sulphur. In weighing
the several portions of apparatus it was tared with somewhat lighter similar pieces of as nearly
as possible the same displacement. The results of this series of experiments, which are vitiated
by the presence, unsuspected at first, of sulphur dioxide in the hydrogen, are stated in values
of H when O=16, but in the following table have been recalculated to conformity with the
earlier determinations:

Wt. of water. Wt. of O. Ratio H:0.
4. 7980 4, 26195 15, 901
7. 55025 6. 71315 16. 039
6. 2372 5. 53935 15. 875

11. 29325 10. 03585 15. 963
11. 6728 10. 3715 15. 940
11. 8433 10. 5256 15. 976
11. 7317 10. 4243 15. 947
19. 2404 17. 0926 15. 916
20. 83435 18. 5234 16. 031
17. 40235 15. 4598 15. 917
19. 2631 17. 11485 15. 934

 

' Mean, 15. 949+0. 0103

Reducing to a vacuum, this becomes 15.843, while a correction for the sulphur dioxide
estimated to be present in the hydrogen brings the value up again to 15.865. Still another
correction is suggested, namely, that as the reduced copper in the combustion tube, before
weighing, was exposed to a long-continued current of dry air, it may have taken up traces of
oxygen chemically, thereby increasing its weight. As this correction, however, is quan-
titatively uncertain, it may be neglected here, and the result of this series will be taken as
O=15.865+0.0103. Its weight, relatively to that of some other series of experiments, is
evidently small.

In their second and final series Dittmar and Henderson dried their hydrogen, after
deoxidation by red-hot copper, over caustic potash and subsequently phosphorus pentoxide.
The copper oxide and copper of the combustion tube were both weighed in vacuo. The results
were as follows, vacuum weights being given:

Wt. of water. Wt. of O. Ratio H:0.
19. 2057 17. 0530 15. 843
19. 5211 17. 3342 [15. 853]
19. 4672 17. 2882 15. 868
22. 9272 20. 3540 15. 820
23. 0080 20. 4421 [15. 934]
23. 4951 20. 8639 15. 859
23. 5612 20. 9226 [15. 859]
23. 7542 21. 0957 15. 870
23. 6568 21. 8994 15. 884
23. 6179 21. 8593 15. 848
24. 6021 21. 8499 15. 878
24. 3047 21. 5788 15, 832
23. 6172 20. 9709 15. 849

 

Mean, 15. 861-0. 0052

The authors reject the three bracketed determinations, because of irregularities in the course
of the experiments. The mean of the 10 remaining determinations is 15.855+0.0044. Both
means, however, have to be corrected for the minute trace of hydrogen occluded by the reduced
copper. This correction, experimentally measured, amounts to +0.006. Hence the mean of
No. 3.] ATOMIC WEIGHTS—CLARKE. 19

all the experiments in the series becomes 15.867 +0.0052, and of the 10 accepted experiments
15.861+40.0044. The authors themselves select out seven experiments, giving a corrected
mean of 15.866, which they regard as the best value. Taking all their evidence, their two series
combine thus:

IRifst S6r68 socci4 ote cand Saeed skh ueedeudstmngneeueses 15. 865 +0. 0103
Second SeTieS.........-2.2-222225-- blebs Soh to te cra eapiarnde ie 15.867 + .0052
General meatiicnssnicsesicwes es 22 sseersebcawswiss 15. 8667+ .0046

Leduc,’ who also effected the synthesis of water over copper oxide, following Dumas’s
method with slight modifications, gives the results of two experiments, as follows:

 

Wt. Water. Wt. O. Ratio H:0.
22.1632 19. 6844 15. 882
19. 7403 17. 5323 15. 880

Mean, 15. 881

These experiments we may arbitrarily assign equal weight with two in Dittmar and
Henderson’s later series, when the result becomes 15.881, +.0132, the value to be accepted.
Leduc states that his copper oxide, which was reduced at as low a temperature as possible,
was prepared by heating clippings of electrolytic copper in a stream of oxygen.

To E. W. Morley? we owe the first complete quantitative syntheses of water, in which
both gases were weighed separately, and afterwards in combination. The hydrogen was
weighed in palladium, as was done by Keiser, and the oxygen was weighed in compensated
globes, after the manner of Regnault. The globes were contained in an artificial “cave,”
to protect them from moisture and from changes of temperature; being so arranged that they
could be weighed by the method of reversals without opening either the “cave” or the balance
ease. For each weighing of hydrogen about 600 grammes of palladium were employed.
After weighing, the gases were burned by means of electric sparks in a suitable apparatus,
from which the unburned residue could be withdrawn for examination. Finally, the apparatus
containing the water produced was closed by fusion and also weighed. Rubber joints were
avoided in the construction of the apparatus, and the connections were continuous throughout.
The weights and derived ratios are as follows:

 

 

H taken. O taken. #0 formed. H:0. #:H,0.
3. 2645 25. 9176 29.1788 15. 878 17.877
3. 2559 25. 8531 29. 1052 15. 881 17. 878
3. 8193 30. 3210 34. 1389 15. 878 17. 873
3. 8450 30. 5294 lost 15.880 22...
3. 8382 30. 4700 34. 3151 15. 877 17. 881
3. 8523 30. 5818 34. 4327 15. 877 17. 876
3. 8298 30. 4013 34. 2284 15. 877 17. 875
3. 8286 30. 3966 34. 2261 15. 878 17.879
3. 8225 30. 3497 34.1742 15. 879 17. 881
3. 8220 30. 3479 34. 1743 15. 881 17. 883
3. 7637 29. 8865 33. 6540 15. 881 17. 883
3. 8211 30. 3429 34. 1559 15. 882 17. 878

Mean, 15. 8792, 17.8785,
+. 00032 +. 00066
Combined, these data give:
Brom ratio. HO). 2 iseces ese setisas lonely ds caiasteee 15. 8792, + . 00032
From ratio H:H,0..... 0020222: suesnsecesasecsecececnc 15. 8785, + .00066

 

General mean.........--.22.-0- cece ee eee eens 15. 8790, + . 00028

 

1 Compt. Rend., 115,41. 1892, See also the complete memoirin Ann, Chim. Phys. (7), 15, 48. 1898. Inthelatter Leduc gave a preliminary
determination which made O=15.860.

3«On the Density of Oxygen and Hydrogen, and on the Ratio of their Atomic Weights,” by Edward W. Morley. Smithsonian Contributions
to Knowledge, 29, 1895, 4to, xi + 117 pp., 40 cuts. Abstract in Am. Chem. Journ., 17, 267 (gravimetric), and Ztschr. phys. Chem., 17, 87 (gaseous
densities); also note in Am. Chem. Journ., 17, 396. Preliminary notice in Proc. Amer. Association, 1891, p. 185. See also a discussion by Morley
ofall the earlier determinations, in the Western Reserve University Bulletin, for April, 1895.
20 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

For details Morley’s original memoir must be consulted. No abstract can do full justice
to it.

Two other series of determinations, by Julius Thomsen, are radically different in method
from all the previous work. In the first seriest he determined the ratio between HCl and
NH,; and thence, using Stas’ values for Cl and N, fixed by reference to O=16, computed the
ratio H:O. This method was so indirect as to be of little importance, and gave for the atomic
weight of oxygen approximately the round number 16. I shall use the data further on for
another purpose. The paper has been sufficiently criticised by Meyer and Seubert,? who have
discussed its sources of error. The ratio has also been studies by Baume and Perrot; whose
work will be considered in another section, later.

In Thomsen’s later memoir* a method of determination is described which is, like the
preceding, quite novel, but more direct. First, aluminum, in weighed quantities, was dis-
solved in caustic potash solution. In one set of experiments the apparatus was so constructed
that the hydrogen evolved was dried and then expelled. The loss of weight of the apparatus
gave the weight of the hydrogen so liberated. In the second set of experiments the hydrogen
passed into a combustion chamber in which it was burned with oxygen, the water being retained.
The increase in weight of this apparatus gave the weight of oxygen so taken up. The two
series, reduced to the standard of a unit weight of aluminum, gave the ratio between oxygen

 

and hydrogen.
The results of the two series, reduced to a vacuum and stated as ratios, are as follows:
First. Second.
Weight of H Weight of O
Weight of Al Weight of Al

0. 11180 0. 88788
0.11175 0. 88799

0. 11194 0. 88774

0. 11205 0. 88779

0. 11189 0. 88785

0. 11200 0. 88789

0. 11194 0. 88798
0.11175 0. 88787

0. 11190 0. 88773
0.11182 0. 88798

0. 11204 0. 88785

0. 11202

0.11204 0. 88787, + 0.000018
0.11179

0. 11178

0. 11202

0. 11188

0. 11186

0. 11185

0. 11190

0. 11187

 

0.11190, + 0.000015

Dividing the mean of the second column by the mean of the first,we have for the equiv-

alent of oxygen: seis aie
0. , £0. i,
0.11190, £0. 000018 “9245, +0. 0011

Hence O=15.8690, +0.0022.
The details of the investigation are somwhat complicated and involve various corrections
which need not be considered here. The result as finally stated includes all corrections and

is evidently good.

 

1 Zeitsch. physikal Chem., 13, 398. 1894.
3 Ber. Deutsch. chem. Ges., 27, 2770. See also Acree and Brunel, Amer. Chem. Journ., 36, 117.
8 Zeitsch. anorg. Chem., 11,14. 1895.
No. 3.] ATOMIC WEIGHTS—CLARKE. 21

The syntheses of water reported by Keiser! in 1898, involved the direct oxidation of
hydrogen occluded in palladium, with subsequent weighing of the water so produced in the
vessel in which it was generated. That vessel was tubular in form, and divided into two
compartments; one containing phosphorus pentoxide, to absorb the water, the other holding
the palladium hydride. Each determination required five weighings, as follows: First, of
the vessel, containing only the drying agent, and exhausted of air. Second, the same as the
first, plus the palladium. Third, the gain in weight was measured after saturating the pallad-
ium with hydrogen. Fourth, the entire apparatus after complete oxidation of the hydrogen
to water. The gain in weight gave the oxygen absorbed. Fifth, like the fourth, but with
the palladium removed. The difference between the first and fifth weighings gave the amount
of water formed. All the operations were thus performed in a single piece of apparatus, and
troublesome corrections were avoided. The data obtained were as follows, with weights not
reduced to a vacuum:

 

 

H taken. O taken. HO formed. H:0. #:H,0.
- 27549 2.18249 2.45975 15. 8444 17. 8573
. 27936 2. 21896 2.49923 15. 8860 17. 8925
. 27091 2.15077 2.42355 15. 8781 17.8919
. 26845 2.13270 2. 40269 15. 8890 17. 9005

Mean, 15. 8744 15. 8855

Taken as one series, the two sets of values, eight determinations in all, give for the ratio
H : O the number 15.8799+.0046. This figure is slightly higher than Morley’s average, but
below bis maximum.

Late in 1907, an elaborate investigation by Noyes? was published, covering five series of
syntheses. The first series of 20 experiments, however, was found to be affected by a small
constant error, and it was therefore rejected. The other series gave the subjoined results, with
all corrections, including the reduction to a vacuum, applied.

Second series.—Electrolytic hydrogen, from sulphuric acid, was weighed in palladium,
and again in the copper oxide tube in which it was oxidized to water. The apparatus was
similar to that used in his former research, and so, too, but with differences in detail, was the
procedure.

 

 

Ff taken. O taken. #0 formed. H:0. HT: H,0.
3. 72565 29. 57891 33. 30408 15. 8785 17. 8783
3. 80318 30. 18400 33. 98748 15. 8730 17. 8732
3. 75873 29. 83358 33. 59127 15. 8743 17. 8737
2. 96328 23.5197 26. 48379 15. 8742 17. 8746
2.11395 lost 18.89214 see 17. 8734
3.53136 28. 02910 31. 56024 15. 8744 17. 8743
3. 53959 28. 09619 31. 63554 15. 8754 17. 8753
Mean, 15.8750 17. 8747

+.00052 +. 00045

Third series—Hydrogen from sulphuric acid was passed directly into the copper oxide
bulb and there converted into water.

 

 

H taken. O taken. HO formed. H:0. Hf: H,0.
2.44279 19. 39757 21. 84042 15. 8813 17. 8815
2.18739 17. 36305 19. 55117 15. 8756 17. 8763
2.75129 21. 84345 24. 59389 15. 8787 17. 8781
4.00062  —......-e 35.750838  —— ....... 17. 8726
4.04057 32. 07689 36. 11762 15. 8774 17. 8775
Mean, 15. 8782 17. 8772

+. 00091 +. 00097

 

1 Amer, Chem. Journ., 20, 783. 1898.
2 Amer. Chem. Journ., 29,1718. 1907. In Vol. 30, p. 4, 1908, Noyes discusses all determinations of the atomic weight of hydrogen, and proposes
criteria for the rejection of doubtful data.
22 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Fourth series.—Hydrogen and oxygen, both obtained by electrolysis of sulphuric acid,
were directly combined by means of palladium, somewhat as in Keiser’s determinations. The
use of copper oxide was thus avoided.

 

 

taken. O taken. #f,0 formed. H:0. #: H,0.
2. 27916 18. 08455 20. 36128 15. 8695 17. 8674
4. 12734 32. 76527 36. 89043 15. 8772 17. 8761
4.17556 33. 13449 37. 30787 15. 8707 17. 8696
4.19346 33. 27384 37. 46453 15. 8694 . 17. 8681
2. 30746 18. 30863 20. 61357 15. 8691 17. 8669
4.59692 36. 48543 41. 08162 15. 8739 17. 8735
4. 63625 36. 79354 41. 42905 15. 8721 17. 8718
4.57274 36. 28696 40. 85834 15. 8710 17. 8704
Mean, 15. 8716 17. 8705

+. 00066 +. 00074

Fifth series.—Essentially like series four, except that the two gases were prepared by
electrolysis of barium hydroxide.

 

 

EF taken. O taken. H,0 formed.” #H:0. AT: H,0.
4. 61180 36. 60909 41. 22105 15. 8763 17. 8763
4. 62358 36. 69575 41. 31647 15. 8733 17. 8721
4.59853 36. 50484 41. 10212 15. 8768 17. 8762
4. 55832 36. 17887 40. 73904 15. 8738 17. 8746
4, 20399 33. 37000 37. 57336 15. 8754 17. 8751
Mean, 15.8751 17. 8749

+. 00046 +. 00051

Since these series of measurements represent different methods, and are evidently of unequal
value, it is best to combine them mathematically, giving each-mean a weight inversely propor-
tional to the square of its probable error. Noyes computed all of the experiments on the basis
of the oxygen standard, giving for each one its expression as the atomic weight of hydrogen. J]
have chosen the present form for simplicity of calculation, and for greater ease of combination
with previous determinations.'

 

Ratio H:0. Atomic weight H.
Series 2): PO. sass nscisciee eee sa ode pease eee aticccemarns 15. 8750+. 00052 1. 00787
EGP Os Sete aicagaroecyesda spay ceayatvorsases tia vacie/areacsbore nmi atersudec accel 15. 8747 +. 00045 1.00789
Dries: dy cel $C 2 ana nye eivs tye nies cere aa laterg aah na ite tyes 15. 8782+. 00091 1. 00767
He HO sa: amas censa nan naannnbaesaacksauaremicded 15. 8772+. 00097 1. 00774
Series 4. OH Ose ccna sed conuawiig ter eestons deed Seema 15. 8716 +. 00066 1. 00809
A 3 OL aoe eedo eed os See came eceee ee ee 15. 8705+. 00074 1. 00815
Series b..H POs is wnciseenaosniseeceeeesnaacdwasieaee ese. 15. 8751+. 00046 1. 00786
HH? He Ois sas sess ocissoeg neat emeneseamea ess 15. 8749+. 00051 1. 00788
General MeCN eee caisiesiaisisigeareaeaeeesenbeeus bee eeh es = 15. 8745 +. 00021 1.00783 +. 000013

The details of Noyes’ investigation are too voluminous for repetition here. It goes almost
without saying that every precaution was taken which his own previous experience and the
experience of others could suggest, and that his materials were of the highest degree of purity.
The true value of the ratio must lie somewhere within the range of variation shown by his
individual determinations, which it may be observed, overlap those of Morley.

We have now before us, for combination, 15 sets of determinations of the hydrogen-oxygen
ratio. JI have arranged them in the order of descending magnitude, and computed their general
mean as follows:

 

1 See Noyes’ memoir for the results of combining his data in different ways. The 48 experiments, taken as one series, give H = 1.00793, + .00002.
Referring to his determinations published in 1890, Noyes points out a constant error in them, the elimination of which reduces the value of the
ratio to 15.879, in agreement with the measurements by Morley. .
No. 3.] ATOMIC WEIGHTS—CLARKE. 23

 

Ratio. Atomic weight H.
1, Erdmann and Marchand....-..-.......220--e--eeeee eens 15.975 +.0113 1.00156
D- DUMB cinacwarcasies Be sor anes we cesadce ene ee sree eens 15. 9607 +. 0070 1. 00246
8, Keiser, 1888..0..0-.c00 ss sdesxtecsecemes 25 ee ee secieeces 15. 9514+. 0011 1. 00305
4. Thomsen, 1870... -----26seieconcsesesiee tees s eeeeesess 15.91 +.0113 1. 00565
5. Noyes, 1890, uncorrected..............-----+--e-e2 cee e eee 15. 8966+. 0017 1. 00650
6. Dulong and Berzelius............------+---+-++-+2----+-- 15.894 +.0570 1. 00667
pe Ws lei eli, 48 unguarded Oa cachaeacieeSnenengenes 15.89 -£..0090 1. 00692
Bo LedUG. acco neds 2c eee aee eRe eake eye vere aserpases. 15.881 +.0132 1. 00750
GO) Keiser, 1898 -.enzadenccecewsassesees sets seed aleeateaeey 15. 8799+. 0046 1. 00756
10. Morley.....------ Shes ssdeabinw hy ees oh sates eam eS 15. 8790 +. 00028 1. 00762
V1. Noyes) 190 (ccedsossc tea cacnnan esse nt nc centee ee cs 15. 8745+. 00021 1.00783
12. Cooke and Richards... 20....02-.s0+6 200 sessiriewseaeeenes 15. 8690+. 0020 1. 00825
13. Thomsen, 1895... seeecesee soos seis os ips peeeee scenes 15. 8690 +. 0022 1. 00825
14. Dittmar and Henderson....-.....-------------+++-----+- 15. 8677 +. 0046 1. 00834
U5. Keiser: 1887 oona%cqecoens seas os ss ce eebe eeeebes ess 15.864 +.0150 1. 00857
Genera UMeancscnescatewso0650 54 Seas getaicgeddasaciet 15. 8779+. 00016 1. 00769 +. 00001

In this combination, which includes all the syntheses, good or bad, the general mean lies
between the values found by Noyes and Morley. It is, therefore, not far from the truth. If
we reject the high values, Nos. 1 to 7, the general mean becomes 15.8760+.00017, and
H=1.00781+4.00001. Values 10 and 11, combined, give 15.8761 +.00017, and H=1.00780+
.00001. That is, the Morley and Noyes determinations control all the others, and practically
eliminate them. The high and low figures tend to balance one another, and so to disappear
from the final combination.

In discussing the relative densities of oxygen and hydrogen gases we need consider only
the more modern determinations, beginning with those of Dumas and Boussingault. As the
older work has some historical value, I may in passing just cite its results. For the density of
hydrogen we have .0769, Lavoisier; .0693, Thomson; .092, Cavendish; .0732, Biot and Arago;
.0688, Dulong and Berzelius. For oxygen there are the following determinations: 1.087, Four-
croy, Vauquelin, and Séguin; 1.103, Kirwan; 1.128, Davy; 1.088, Allen and Pepys; 1.1036, Biot
and Arago; 1.1117, Thomson; 1.1056, De Saussure; 1.1026, Dulong and Berzelius; 1.106, Buff;
1.1052, Wrede.*

In 1841 Dumas and Boussingault? published their determinations of gaseous densities.
For hydrogen they obtained values ranging from .0691 to .0695; but beyond this mere state-
ment they give no details. For oxygen three determinations were made, with the following

results:
1. 1055
1.1058
1.1057

Mean, 1.10567 +-.00006

 

If we take the two extreme values given above for hydrogen, and regard them as the entire
series, they give us a mean of .0693+.00013. This mean hydrogen value, combined with the
mean for oxygen, gives for the latter, when H=1, the density ratio 15.9538 +.031.

Regnault’s researches, published four years later,3 were much more elaborately executed.
Indeed, they long stood among the classics of physical science, and it is only recently that they
have been supplanted by other measurements.

For hydrogen three determinations of density gave the following results:

: . 06923

. 06932
. 06924

 

Mean, . 069263 +.000019

 

1 For Wrede’s work, see Berzelius’ Jahresbericht for 1843. For Dulong and Berzelius, see the paper already cited. All other determinations
are taken from Gmelin’s Handbook, Cavendish edition, v. 1, p. 279.

1Compt. Rend., 12, 1005. Compare also with Dumas, Compt. Rend., 14, 537.

3 Compt. Rend., 20, 975.
24 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

For oxygen four determinations were made, but in the first one the gas was contaminated

by traces of hydrogen, and the value obtained, 1.10525, was, therefore, rejected by Regnault as
too low. The other three are as follows:

1. 10561

1. 10564

1. 10565

Mean, 1. 105633 +.000008

 

Now, combining the hydrogen and oxygen series, we have the ratio H: O: : 1: 15.9628 +.0044.
According to Le Conte,' Regnault’s reductions contain slight numerical errors, which, corrected,
give for the density of oxygen, 1.105612, and for hydrogen, .069269. Ratio, 1: 15.9611.

A much weightier correction to Regnault’s data has already been indicated in the discus-
sion of Cooke and Richards’s work. He assumed that the globes in which the gases were weighed
underwent no changes of volume, but Agamennone,? and after him, but independently,? Lord
Rayleigh showed that an exhausted vessel was perceptibly compressed by atmospheric pres-
sure. Hence its volume when empty was less than its volume when filled with gas. Crafts
having access to Regnault’s original apparatus has determined the magnitude of the correction
thus indicated. Unfortunately, the globe actually used by Regnault had been destroyed,
but another globe of the same lot was available. With this the amount of shrinkage during
exhaustion was measured, and Regnault’s densities were thereby changed to 1.10562 for
oxygen and .06949 for hydrogen. Corrected ratio, 1: 15.9105. Doubtless Dumas and Bous-
singault’s data are subject to a similar correction, and if we assume that it is proportionally
the same in amount, the ratio derived from their experiments becomes 1: 15.9015.

In the same paper, that which contained the discovery of this correction, Lord Rayleigh
gives a short series of measurements of his own. His hydrogen was prepared from zinc and
sulphuric acid, and was purified by passage over liquid potash, then through powdered mercuric
chloride, and pulverized solid potash successively. It was dried by means of phosphorus pent-
oxide. His oxygen was derived partly from potassium chlorate, and partly from the mixed
chlorates of sodium and potassium. Equal volumes of the two gases weighed as follows:

Hl. 0.
- 15811 2, 5186 +..00061 °]
- 15807
. 15798
. 15792

 

Mean, . 15802 4-.000029

Corrected for shrinkage of the exhausted globe these become—H, 0.15860; O, 2.5192. Hence
the ratio 1: 15.884 +.0048.

In 1892 Rayleigh published a much more elaborate determination of this ratio. The gases
were prepared electrolytically from caustic potash, and dried by means of solid potash and
phosphorus pentoxide. The hydrogen was previously passed over hot copper. The experi-
ments, stated like the previous series, are in five groups; two for oxygen and three for hydrogen;
but for present purposes the similar sets may be regarded as equal in weight, and so discussable
together. The weights of equal volumes are as follows:

H. 0.
. 15807 2. 5182
. 15816 2.5173
First set}. 15811 2. 5172| First set
Mean, . 15808}. 15803 2. 5193{Mean, 2.51785
. 15801 2. 5174
. 15809 2.5177)

 

1 Private communication. See also Phil. Mag. (4), 27, 29, 1864, and Smithsonian Report, 1878, p. 428.
4 Atti Rendiconti Acad. Lincei, 1885. ;

8 Proc. Roy. Soc., 43, 356. Feb., 1888.

4Compt. Rend., 106, 1662.

6 Arbitrarily assigned the probable error of a single experiment in Rayleigh’s paper of 1892.

6 Proc. Roy. Soc., 50, 448, Feb. 18, 1892.
No. 3.] ATOMIC WEIGHTS—CLARKE. 25

 

H. 0.
. 15800 2. 5183,
. 15820 2. 5168
Second set |. 15792 2.5172] Second set

Mean, .15797}. 15788 2.5181/Mean, 2.5172
. 15783 2.5156
. 15807
. 15801 Mean, 2.5176+.00019
. 15817

Third set Hier

Mean, .15804 15798
. 15802
. 15807

 

Mean, . 15804 -£.000019

These weights with various corrections relative to temperatures and pressures, and also
for the compression of the exhausted globe, ultimately become for H, .158531; and for O,
2.51777. Hence the ratio 1: 15.882+.0023. For details relative to corrections the original
memoir should be consulted.
In his paper ‘On a new method of determining gas densities,’”’ 1 Cooke gives three measure-
ments for hydrogen, referred to air as unity. They are:
- 06957

- 06951
. 06966

Mean, . 06958 + .000029

Combining this with Regnault’s density for oxygen, as corrected by Crafts, 1.10562,+
.000008, we get the ratio H: O:: 1 : 15.890 +.0067.

Leduc, working by Regnault’s method, somewhat modified, and correcting for shrinkage
of exhausted globes, gives the following densities:+

 

Ef. 0.
.06947 1.10501
-06949 1.10516
06947

 

Mean, .06948 + .00006745

The two oxygen measurements are the extremes of three, the mean being 1.10506 + .0000337.
Hence the ratio 1 : 15.905, +.0154.

In a later memoir Leduc * gives two more measurements of the density of oxygen. They
are 1.10527 and 1.10521. If we include these in series with the other values the mean becomes
1.10514 +.0000321. The use of this figure in subsequent combinations of data has an insig-
nificant effect upon the computations. It raises O from 15.905 to 15.906.

The first two hydrogen determinations were made with gas produced by the electrolysis
of caustic potash, while the third sample was derived from zinc and sulphuric acid. The
oxygen was electrolytic. Both gases were passed over red-hot platinum sponge, and dried by
phosphorus pentoxide.

Much more elaborate determinations of the two gaseous densities are those made by
Morley.* For oxygen he gives three series of data; two with oxygen from potassium chlorate,
and one with gas partly from the same source and partly electrolytic. In the first series, tem-
perature and pressure were measured with a mercurial thermometer and a mano-barometer.
In the second series they were not determined for each experiment, but were fixed by com-
parison with a standard volume of hydrogen by means of a differential manometer. In the
third series the gas was kept at the temperature of melting ice, and the mano-barometer alone

 

1 Proc. Amer. Acad., 24, 202. 1889, Also Am. Chem. Journ., 11, 509.

2 Compt. Rend., 113, 186. 1891 :

§ Ann. Chim. Phys. (7), 15,29. 1898. InC. R., 148, 42, Leduc claims that the probable error of his H is only + .00001.
4 Paper already cited, in the synthetic portion of this chapter.
26 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

was read. The results for the weight in grammes, at latitude 45°, of one liter of oxygen are
as follows:

 

 

First Series. Second Series. Third Series.
1. 42864 1, 42952 1. 42920
1. 42849 1, 42900 1. 42860
1, 42838 1, 42868 1. 42906
1. 42900 1. 42853 1, 42957
1. 42907 1. 42858 1. 42910
1. 42887 1. 42873 1. 42930
1. 42871 1. 42913 1, 42945
1. 42872 1, 42905 1. 42932
1. 42883 1. 42896 1. 42908
1. 42880 1. 42910
Mean, 1. 42875 + . 000051 1. 42874 1. 42951
Corrected, 1. 42879 + . 000051 1. 42878 1. 42933
1. 42872 1. 42905
1. 42859 1, 42914
1, 42851 1. 42849
1. 42894
Mean, 1. 42882 + . 000048 1. 42886

Corrected, 1. 42887 + . 000048

 

Mean, 1. 42912 + . 000048
. Corrected, 1.42917 + . 000048
General mean of all three series, 1.42896 + .000028.

Morley himself, for experimental reasons, prefers the last series, and gives it double weight,
getting a mean density of 1.42900. The difference between this mean and that given above
is insignificant with reference to the atomic weight problem.

In the case of hydrogen, Morley’s determinations fall into two groups, but in both the
gas was prepared by the electrolysis of pure dilute sulphuric acid, and was most elaborately
purified. In the first group there are two series of measurements. Of these, the first involved
the reading of temperature and pressure by means of a mercurial thermometer and mano-
barometer. In the second series, the gas was delivered into the weighing globes after occlusion
in palladium; it was then kept at the temperature of melting ice, and only the siphon barometer
was read. In this group the hydrogen was possibly contaminated with mercurial vapor, and
the results are discarded by Morley in his final summing up. For present purposes, however,
it is unnecessary to reject them, for they have confirmatory value, and do not appreciably
affect the final mean. The weight of one liter of hydrogen at 45° latitude, as found in these
two sets of determinations, is as follows:

 

First Series. Second Series.
0. 089904 0. 089977
- 089936 * — , 089894
. 089945 . 089987
. 089993 . 089948
. 089974 . 089951
. 089941 . 089960
. 089979 - 090018
. 089936 . 089909
. 089904 . 089953
. 089863 . 089974
. 089878 . 089922
. 089920 . 090093
. 089990 . 090007
. 089926 . 089899
. 089928 . 089974
. 089900
Mean, . 089934 + . 000007 - 089869
Corrected, . 089938 -+ . 000007 . 090144
. 089984

 

Mean, . 089967 + . 000011
Corrected, . 089970 + . 000011

 

1 Correction applied by Morley to all his series, for a slight error, so355, in the length of bis standard meter bar.
No. 3.] ATOMIC WEIGHTS—CLARKE. 27

In the second group of experiments, the hydrogen was weighed in palladium before transfer
to the calibrated globe; and in weighing, the palladium tube was tared by a similar apparatus
of nearly equal volume and weight. After transfer, which was affected without the intervention
of stopcocks, the volume and pressure of the gas were taken at the temperature of melting ice.
A preliminary set of measurements was made, followed by three regular series; of these, the
first and second were with the same apparatus, and are different only in point of time, a vacation
falling between them. The last series was with a different apparatus. The data are as follows,
with the means as usual:

 

 

 

Preliminary. Third Series. Fourth Series. Fifth Series.
. 089946 . 089874 . 089972 . 089861
- 089915 - 089891 . 089877 . 089877
. 089881 - 089886 . 089867 . 089870
- 089901 . 089866 . 089916 . 089867
. 089945 . 089911 . 089770 . 089839
. 089856 . 089846 . 089874
Mean, . 089918, . 089912 - 089864
+ . 0000271 . 089872 Mean, . 089875, . 089883 '
Corrected, . 089921 + . 0000187 . 089830
Mean, . 089883, Corrected, . 089880 - 089877
+ . 0000049 . 089851

 

Corrected, . 089886
Mean, . 089863,

+ . 0000034

Corrected, . 089866

Now, rejecting nothing, we may combine all the series into a general mean, giving the
weight of one liter of hydrogen as follows:

PUPStBCHIOS s oice <s/a's pam aunantaerecsecnhatiaie este oe 0. 089938 +0. 000007
Second SOMES iasi82 sa cconnanetaesnne seve cemes ct cae .089970+ .000011
Preliminary series, second method.......-.-.------- .089921+ . 0000271
Thitd SCT1CS 5 210 s'ee's gee sie cisicigiaees ee eeceanie se eee . 089886+ . 0000049
Fourth serieg.s.. 5. 5.23 sseesennecdeaiwiccanials sess . 089880+ . 0000187
Fifth series scccc32 ve sss s sneeeweetesceceee ee eees .089866+ . 0000034
General Mean). soos see ceca ecs eee eectewes sss . 0898974 . 0000025
Rejecting the first three..............2.2-0- 02 ee eens . 089872+ . 0000028

This least mean value for hydrogen will be used in succeeding chapters of this work for re-
ducing volumes of the gas to weights. Combining the general mean of all with the value found
for the weight of a liter of oxygen, 1.42896, + .000028, we get for the ratio H : O.

O=15.8955 + 0.0005

If we take only the second mean for H, excluding the first three series, we have—
O= 15.9001 + 0. 0005

The value is undoubtedly nearest the truth, and is preferable to all other determinations
of the density ratio. Its probable error, however, is given too low; for some of the oxygen
weighings involved reductions for temperature and pressure. These reductions involve, again,
the coefficient of expansion of the gas, and its probable error should be included. Since, how-
ever, that factor has been disregarded elsewhere, it would be an over-refinement of calculation
to include it here. Other corrections, of a mathematical character, have been recently applied
to Morley’s data by Guye and Mallet.1 They find for the normal weight of 1 liter of each gas,
O = 1.42886, and H=0.089875. The difference between these figures and those given by Morley
is so small as to be negligible.

Still more recently, by a novel method, J. Thomsen has measured the two densities in ques-
tion? In his gravimetric research, already cited, he ascertained the weights of hydrogen and of

 

1 Compt. Rend., 138, 1034. 1904, 2 Zeitsch. anorg. Chem., 12,4. 1806.
28 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

oxygen equivalent to a unit weight of aluminum. In his later paper he describes a method of
measuring the corresponding volumes of both gases during the same reactions. Then, having
already the weights of the gases, the volume-weight ratio, or density, is in each case easily
computable. From 1.0171 to 2.3932 grams of aluminum were used in each experiment. Omit-
ing details, the volume of hydrogen in liters, equivalent to 1 gramme of the metal, is as follows:

1, 24297 1. 24283
1. 24303 1. 24260
1, 24286 1, 24314
1. 24271 1. 24294

Mean, 1.24289+0. 00004.

The weight of hydrogen evolved from 1 gramme of aluminum was found in Thomsen’s
gravimetric research to be 0.11190+0.000015. Hence the weight of 1 liter at 0°, 760 mm., and
10.6 meters above sea level at Copenhagen is:

0.090032 + 0.000012
or at sea level in latitude 45°,
0.089947 + 0.000012 gram.

The data for oxygen are given in somewhat different form, namely, for the volume of 1
gramme of the gas 0°, 760, and at Copenhagen. The values are, in liters:

0. 69902
- 69923
- 69912
. 69917
- 69903
. 69900
- 69901
- 69921
. 69901
- 69922

Mean, . 69910+0. 00002
At sea level in latitude 45°, .69976+0. 00002
Hence 1 liter weighs 1.42906 + 0.00004 gramme.
Dividing this by the weight found for hydrogen, 0.089947 + 0.000012 we have for the ratio
H:0O,

 

15.8878 + 0.0022
The density ratios, H :O, now combine as follows:

Dumas and Boussingault, corrected........---- re eiatceeiaes 15. 9015 +40. 031

Regmault,, Corretéd <..osnsasckcneen winnidesacidemeced: 24 15.91054+ .0044
Sls ghi 108 iso cu ocinsse-oa nomomnmmsncainenenteen 2% 15.884 + .0048
Rayleigh, 1892)...c.ccoscosetoueues seseestascaseawades 15. 882 + .0023
C00Keuiiee. sac esses ve peiarsceaemeeesonss ee 15.890 + .0067
TSG CUIG oe eens Usd ape salrceerare de oareiennetok ata pee ehiiiate x12 15.906 + .0154
Morley, including all the data....-.....-...----2+-+-- 15. 89554 .0005
TPHOMSEN «(2 oice-0'ciessienneici oe See ee Sede sem aheey < ees she xs 15. 8878+ . 0022

General Mean! ..2¢ ss fesan sch ede senaees sess 15. 89484 .00048

[f we reject all of Morley’s data for the density of hydrogen except his third, fourth, and

fifth series, the mean becomes
15.8991 +0. 00048

In either case Morley’s data vastly outweigh all others.

If oxygen and hydrogen were perfect gases, uniting by volumes exactly in the ratio of 1 to 2,
then their relative densities would also indicate their relative molecular weights. But, in fact,
the two gases vary from Boyle’s law in opposite directions, and the true composition of water by
volume diverges from the theoretical ratio to a measurable extent. Hence, in order to deduce
the atomic weight of hydrogen from its density, or that of oxygen, if the hydrogen scale is
No. 3.] ATOMIC WEIGHTS—CLARKE. 29

preferred, a small correction must pe applied which depends upon the amount of the divergence.
Until modern times our knowledge of the volumetric composition of water rested entirely upon
the determinations made by Humboldt and Gay Lussac'! early in the last century, which gave a
ratio between H and O of alittle less than 2 : 1, but their data need no farther consideration here.

In 1887 Scott? published the first series of experiments, 21 in number, finding as the most
probable result a value for the ratio of 1.994:1. In March, 1888,3 he gave four more determina-
tions, ranging from 1.9962 to 1.998 : 1; and later in the same year‘ another four, with values
from 1.995 to 2.001. In 1893,° however, by the use of improved apparatus, he was able to show
that his previous work was vitiated by errors, and to give a series of measurements of far greater
value. Of these, 12 were especially good, being made with hydrogen from palladium hydride,
and with oxygen from silver oxide. In mean the value found is 2.00245 + 0.00007, with a range
from 2.0017 to 2.00380.

In 1891 an elaborate paper by Morley * appeared, in which 20 concordant determinations
of the volumetric ratio gave a mean value of 2.00023+0.000015. These measurements were
made in eudiometer tubes, and were afterwards practically discarded by the author. In his later
and larger paper,’ however, he redetermined the ratio from the density of the mixed electrolytic
gases, and found it to be, after applying all corrections, 2.00274. The probable error, roughly
estimated, is 0.00005. Morley also reduces Scott’s determinations, which were made at the tem-
perature of the laboratory, to 0°, when the value becomes 2.00285. Leduc’s * single determina-
tion, based upon the density of the mixed gases obtained by the electrolysis of water, gave
2.0037; but Morley shows that some corrections were neglected. This determination, therefore,
may be left out of account.

There is also a corroborative measurement by Rayleigh,’ who assigns to the ratio the value
2.0026. This agrees well with the figures given by Scott and Morley. Rayleigh also gives
measurements of gaseous densities at very low pressures, and obtains molecular ratios differing
considerably from those ordinarily found. At atmospheric pressure, for example, H=1.0075;
and at very low pressures its atomic weight becomes 1.0088.

In 1917 an elaborate redetermination of this volume ratio, by Burt and Edgar,!° appeared.
Five series of determinations were made, with varying conditions, and with all known pre-
cautions against error. Omitting details, the mean values for these series are as follows:

Series 1, twelve measurements........-...-.------- 2.00294 +0. 00002
Series 2, ten measurements......-..........------- 2.00292 +0. 000014
Series 3, ten measurements...........-...-2------- 2.00292 +0. 00001
Series 4, ten measurements... .-...--...-.-2------- 2.00301 +0. 00002
Series 5, twelve measurements..............-...--- 2.00287 +0. 00001
Getieral q6an 2 z..ci0 acieiiee's os 2 geeipedexeaworee 2.00294 +0. 0000058

This mean is subject to two small corrections, which reduce it to 2.00288, +0.0000058.
For comparison this may be combined with the previous determinations, neglecting those of
Leduc and Rayleigh, thus:

COL, COMeCted a ead ct du ce bcmscctecencleleiencs 3 eildee 2.00285 +0.00007

MONG Y seed wecea tees See ta gee ees ey oeed  od 2.00274 +0. 00005

Puriiand Hdpat. ix. ley eugene de stants 2.00288 +0. 0000058
General mean....................22.-2------- 2.00287 +0. 0000057

 

1 Journ, de Fhys., 60, 129.

2 Proc. Roy. Soc., 42, 396.

3 Nature, 37, 439.

4 British Assoc. Report, 1888, 631.

5 Proc. Roy. Soe., 58,180. In fullin Philosophical Transactions, 184, 543, 1893.
6 Amer. Journ., Sci. (3), 46, 230 and 276.

1 Already cited with reference to syntheses of water.

®Compt. Rend., 175, 311. 1892, Ina later, more complete menoir, Ann. Chim. Phys. (7), 15, 49, Ldeuc gives the figures 2.0034. He also

criticizes Morley’s deductions.

9 Proc. Rou. Soc., 73, 153. 1904.
10 Phil. Trans., series A, vol. 216, pp. 393-427.
‘

30 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

This is practically Burt and Edgar’s value alone. The others vanish. Combining their
value for the volume ratio with Morley’s values for the densities of oxygen and hydrogen,
Burt and Edgar conclude that

H=1.00772.

Now, including all the data so far considered, we have as a mean value for the density
ratios:

(A) H:0::1:15.8948 + .00048,
or, omitting Morley’s rejected series for hydrogen,
(B) H:0::1:15.8991 +.00048.

Correcting these by the volume ratio, 2.00287 +.0000057, the final result for the atomic
weight of oxygen, in terms of the hydrogen unit, and as computed from the gaseous densities

becomes—
Brom Avis aaveare nance enactaauinly sede eeeka hades O=15. 8720 +.00048

From. Bic2ssccecidceeesek vy oes vei ties edviseess O=15. 8763 +. 00048

Combining these figures with the values deduced from the syntheses of water, rejecting
nothing, we have—

By syntheses of water...-.---..---------0--+eee eee O=15. 8779 +.00016
By gaseous densities... .-..........---.-225 eee eee O=15. 8720 +. 00048

General mean..........------.---------+--- O=15. 8773 +.00015
Hence, on the oxygen scale. ......-..---------+--- H=1.00773 +.00001

If we reject the seven highest values under the first heading, and omit Morley’s defective
hydrogen series under the second, we get—

 

By syntheses of water............---.---------- O=15. 8760, +. 00017, or H=1.00781 +.00001
By gaseous densities. ...........-------------- O=15. 8763, +.00048, or H=1. 00779 +. 00003
General Meal... 2... e cence ene es aes O=15. 8761, +.00016, or H=1. 00780 +.00001

The two component values of the last mean are practically identical, and fix the atomic

weight of hydrogen as very close to
H=1.0078 +.00001.

The actual uncertainty of this value, however, is greater than the so-called ‘probable
error.” The latter, it must be borne in mind, is a mathematical expression which should not
be used in a colloquial sense. For computations of this kind the probable error is essentially
a coefficient of concordance, which merely indicates the relative value or weight assignable to
a given series of observations in comparison or combination with others.

In the foregoing discussion both chemical] and physical data have been taken into account.
In recent years, however, much has been done on the purely physical side; and the molecular
weights of gases have been deduced from physical evidence alone. In these calculations the
weight of a normal liter of oxygen is a fundamental standard of reference, and a number of
investigators have sought to determine it accurately. Jaquerod and Pintza,' for example,
give as the mean of five determinations, the value 1.4292 grams, but without the individual
figures. Jaquerod and Tourpaian? also, but without details, find it to be 1.4290 grammes.

Guye, from data given in different forms by other investigators, has computed the follow-
ing values for this constant: ‘

From data by Rayleigh. .....-.....-.---- 2-2. eee eee eee eee eee 1.42905
From data by Wiedues sccsersccecedesvee y seis 35224 Sheaseceeees 1. 42876
From data by Gray......... MeBuiebuientinuee s exe bss dae daaies 1. 42896

The arithmetical (unweighted) mean of these five values is 1.42899 grammes.

 

3 Compt. Rend., 139, 129. 2 Arch. Sci. Phys. Nat., (4) 31, 20. § Journ. Chim. Phys., 5, 203. 1907.
No. 3. ATOMIC WEIGHTS—CLARKE. 31

The most recent determinations of the constant now under consideration are those by
Scheuer! and Germann.? Scheuer made 24 determinations, which gave in mean as the crude
weight of a liter of oxygen the value 1.42954+.000016 grammes. Corrected for latitude and
elevation the weight of the normal liter becomes 1.42906 grammes. By Germann there are 15
determinations, in mean 1.42924+.000058 grammes crude, or 1.42906 normal. Continuing
these values with the preferred figure given by Morley and the determination by Thomsen,
we have—

Morley, preferred... .....2--2--2- 220-02 e ee eee eee eee 1. 42900 +. 000028
Thomsen 2.0: isiurtsiacoa Micali sss sestamacemacy See eee 1.42906 +. 00004
Schetet..:.< cs:cnintc-Suswewies cos ve madawerearedewa s sie, _ 1.42906 +. 000016
Germann.........-.----- Gig Piece sneer eende ee 24 1. 42906 +. 000058
"General MCSD so hcea meses naseanie a Seicetee dacs 1. 42905 + .000013

This value, combined with Morley’s figure for the weight of a liter of hydrogen, and Burt

and Edgar’ s volume ratio gives
H=1.00768

as compared with the 1.00780 of my general combination, which includes the purely chemical
determinations.

Several recent discussions of the physical data relative to the atomic weight of hydrogen
are worth citing here. Jaquerod and Scheuer * reduced the gaseous densities by measuring
the compressibilities of hydrogen and oxygen, and found H=1.0078. Grinnell Jones,‘ in a
very careful reduction of the density ratios by several methods found H=1.00775. Guye,s
from the established densities of the two gases, with various corrections and a discussion of
sources of error, concludes that the true value lies between 1.00767 and 1.00773; or, in mean,
that H=1.00770. Hinrichs,* in a calculation which was partly. theoretical, decided that
H=1.0081. These various values compare as follows:

Clarkete a od iol dat tate et A Satin sata a cdimauintnuasas H=I. 0078
Jaquerod and Scheuer.......... 2. sce cee cece ee eeeese Giisieee H=1.0078
JONES sos sais cree ee eattienews bs Vawsaisninetseeseuree sss sane H=1. 00775
GUY Ci. s nc ecscie Gee Seto ent cee ee acemien Nawarscieneme de acaies H=1.0077
Burt and Edgar RUMCe mace Kine a eee meee ein Pe eee ewe NROAS H=1.00772
MICH Petsmieeeetadeandacnnssetous nonneateetecced cokagen H=1. 0081

The first five of these values are fairly concordant, and justify the belief that the atomic
weight of hydrogen is fixed with an uncertainty not greater than 1 part in 10,000. My own
value, which involves both chemical and physical determinations, will be used throughout this
work. For reductions of data relative to gases other than oxygen and hydrogen, the Morley
value for hydrogen, and the normal figure for « oxygen, 1.42905 grammes at 0°, 760 mm. pressure,
latitude 45 N., and sea level will be employed. 7

THE NITROGEN-OXY GEN RATIO.

_ The direct ratio between nitrogen and oxygen has been determined by analyses of the
oxides of nitrogen and by measurements of gaseous densities. The different methods may be
considered in regular order.

The exact analysis of nitrous oxide, with reference to the atoms weight. of nitrogen, was
effected by Guye and. Bogdan. 8 The gas itself was condensed in carefully purified charcoal, and
so weighed; it was then passed slowly through a tube containing a spiral of iron wire, which

 

1 Anzeiger Wien Akad., 50, 378. .

2 Journ. Chim. Phys., 12, 66. Journ. Phys. Chem., 19, 437. Germann gives an exhaustive summary of the earlier investigations.

3 Compt. Rend., 140, 1384.

4 Journ. Amer. Chem. Soc., 32, 513.

5 Journ. Chim. Phys., 15, 208.

6 Rév. Gén. Chim., 13, 361.

1 Two additional ales for the weight ofa liter of oxygen at 0° and 760 mm., but uncorrected for latitude and elevation, are worth putting on
record here. By Guye Kovacs and Wourtzel at Geneva (Journ. Chim. Phys., 10, 332), 1.42819 grammes, the mean of 15 determinations. By Bruy
lantsand Bytebier at Louvain (Bull. Acad. Belge, 1912, 856) 1.4295 grammes, the mean of 10 determinations: |

§ Compt. Rend., 138, 1494. 1904. Journ. Chim. Phys., 3, 537. 1905.

161566°—20——3
32 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

was heated to redness by an electric current, Thé iron was oxidized, and its gain in weight
gave the amount of oxygen in the N,O,. The results obtainod were as follows:

Wt. N,O. Wi. 0. At. Wt. N.
1.1670 4242 14. 0085
_ 9498 - 3453 14. 0052
. 8652 8145 14, 0083
1, 2247 . 4455 13. 9924
1. 4202 . 5159 14. 0229

 

* Mean, 14. 0075+. 0033

For the complete gravimetric analysis of nitric oxide we have the elaborate data furnished
by R. W. Gray.t The gas was weighed, and then reduced by electric heating over finely divided
metallic nickel. The'gain in weight of the nickel represented the amount of oxygen absorbed.
In some of the experiments the liberated nitrogen was condensed, at the temperature - ‘of liquid
air, in cocoa-nut charcoal, and its weight also was determined. "Two. series of determinations
were made, on nickel from ‘different sources, but for present purposes these may be treated as one.
For three of the measurements corrections are given for the nitrogen occluded by the mixed
nickel and nickel oxide, which corrections I have applied in the following table of Gray’s results:

 

 

 

Wt. NO. Wt. O. Wt. N.

. 31384 -167290 kee

. 64304 . 34300 . 80010

. 50672 . 27025 Eset
54829 e202

. 61862 . 32981 . 28885

. 62622 . 33401 . 29234

- 62128 e8ONNT asta

. 54469 . 29029 . 25432

. 52001 . 27715 . 24270

. 62103 . 33103 . 28998

From these weights the subjoined values for N are derived.

'NO:0,. N03. NO: Nj.
14, 0164 witb Sata

_ 18.9960 13. 9988 13. 9996
SEM! atest igh
14. 0217 gegen Seen
14, 0110 14, 0129 14. 0146
13. 9997 14. 0039 14, 0094
14. 0217 Sie,  t«‘C
14. 0218 14. 0174 14, 0136
14. 0204 14. 0112 14. 0031
14. 0169 14. 0159 _ 14,0151

Mean, 14.01264.0022 =  _ —- 14, 0100-4. 0020 14, 0092-4. 0018

The general mean of the three series is N=14,0104 + .0011

The analyses of nitrogen peroxide, by Guye and Drouginine,? are similar in character to
those of nitrous oxide by Guye and Bogdan. A spiral of iron wire was electrically heated in a
bulb containing a known weight of the peroxide, and the increased weight of the spiral gave
the amount of oxygen released by the complete decomposition of the initial compound. For
the elaborate details relative to manipulation the original memoir should be consulted. The
weights, in grams, are as follows:

We. NiO, Wi. 20>. Wt. Ny. At. Wi. N.
0. 8856" 0. 6160 0. 2696 14. 005
1. 8494 1, 2860 0.5634 14, 019
2.1281 1. 4801 0. 6480 14. 010
1.7191 1. 1957 0.5234 14, 008
0. 9951 0. 6921 0. 3030 14. 010
1. 5324 1. 0658 0. 4666 14. 009
1.2425 ° 0.8642 0.3783 14, 008

Mean, 14. 0099+. 0011
1 Journ. Chem. Soc., 87, 1601. 1905. 2 Journ. Chim. Phys., 8, 473, 1910.

 
No. 3.]

ATOMIC WEIGHTS—CLARKE. © 33

An entirely different process for determining the atomic weight of nitrogen was adopted
by Wourtzel,! who oxidized nitric oxide, NO to the peroxide, N,O,. The nitric oxide was first
dissolved in liquefied NO,, forming nitrous anhydride, N,O,.

a stream of dry oxygen. All the operations were effected at very low temperatures.

weights were as follows:

The latter was then oxidized by
The

Wi. NO. We. O, absorbed. At. Wt. N.
1. 3653 0. 72805 14. 005
1. 76655 0, 9419 "14. 008
1, 5983 0. 85225 " 14, 006
1, 9562 1. 04305 14, 007
2. 1854 1. 1652 14. 008

Mean, 14. 0068+ .0004

The five series of determinations of the atomic weight of nitrogen, by Scheuer,’ appear to
have been published only in abstract, without the details needed for their critical discussion.
Other work by Scheuer, however, has been of such good character, that these determinations may
be accepted as trustworthy.

nitrogen oxides.

Three of the series were. ana the other two give analyses of
The data are as follows:

Series 1. Syntheses ‘of N,O paces NO and O,.

Wt. NO. Wt. .NO,. At. Wt. N.
1. 22627 1.88011 14. 0078
1. 23426 1. 89086 14. 0073
1, 22527 1. 87861 14. 0063
1. 22906 1, 88437 14, 0087

Mean, 14. 0075+. 0005

Series 2. Syntheses of N,O,; from NO and O,.

Wt. NO. Wt. N20; At. Wt. N.
2. 45259 3. 10643 14. 0085
2. 46280 3. 11940_ . 14. 0087
2. 45063 3. 10397 14. 0074
2. 45809 3. 11840 14. 0083

Mean, 14. 0077+. 0004

Series 3.—Syntheses of N,O, from N,O, and O,.

Wt. NO, Wi. NO, At. Wt. N.
1. 20468 1. 45825 14. 0070
1.53106 1.85331 14, 0092
1, 64258 1. 98829 14. 0083
3. 11340 3. 76873" 14. 0071
. Mean, 14.0079 +. 0005
Serves 4.—Analyses of N,O,.
Wt. NO, Wt. 0, Wt. N, At. Wt. N.
0. 92456 0. 64304 oy 0. 28152 14. 0095
. 99827 . 69434 . 30393 14. 0072
. 95147 . 66177 . 28970 14. 0085
. 89762 . 62341 . 27331 14. 0090
“4 Mean, 14. 0085-4. 0005
Series 5.—Analyses of NO.
Ws. N.O, Wt. 0, We: Nz At. Wi. N.
0. 97434 “0. 61523 0. 35911 14, 0083
. 92776 . 58582 84194 14, 0086
. 96347 . 62733 . 36614 14.0075
97145 . 61339 . 35806 14. 0097
- 94570 . 59714 . 34856 14. 0092

Mean, 14. 0086-+. 0004.

 

1 Compt. Rend., 154, 115, 1912.

2 Anzeiger Wien Akad., 49, 38, 1912.
34 MEMOIRS NATIONAL ACADEMY OF ‘SCIENCES. [Vou. XVI.

The atomic weights and probable errors are all as given by Scheuer. They have not been
recalculated here. '

We now have nine series of determinations, which combine as Follawa

Guye and Bogdan... ....-...2-22200-.0-222.022.--+-N=14. 0075+.'0033
Giay:teitcse 2 ye cicsbisn neti letclldd cunmtacmink ete N=14. 0104+. 0011
Guye and Drouginine............. Benaniideivrcicneeres N=14. 0099+. 0011
Wourtzel.....-..-...02 002 cece pile te tacmieng Saye ie taiare baie N=14. 0068 +. 0004
Scheuer, Series 1..............-..- Bocce swaths wae dee N=14. 0075+. 0005
Scheuer, Series 2..........--.. 20.0 cece eee c ee cee ee eee N=14. 0077-4. 0004
Scheuer, Series 3...... 222.22... eceeeeeceee cece eee eeeee N=14. 0079. 0005
Scheuer, Serles:4 ss-jasewss cetecessaccskbieeeatenieions N=14. 0085+. 0005
Scheuer, Series 5..-......2.... 2222-2 eee eeeeeeeeeeee ee N=14. 0086+. 0004
General mean..........-.--..-.-2--002.020---+--N=14, 0079+. 00018

The accurate calculation of molecular and atomic weights from gaseous densities is really
an affair of very recent times. The gases, as measured, show divergencies from Avogadro's
law, and the crude density ratios therefore require correction, as we have already seen in refer-
ence to the atomic weight of hydrogen. It seems best, however, to assemble the actual measure-
ments first and to apply the corrections to the entire mass of data afterwards.

For nitrogen there are abundant measurements made upon the element itself, and also
good data for nitrous oxide, nitric oxide, and ammonia,

The earlier determinations of the density of nitrogen were all made upon nitrogen derived
form the atmosphere. But the supposed nitrogen contained, as we now know, the heavier
argon, and the value obtained was therefore incorrect. It is, however, worth while to examine
the data, and to see whether a correction for argon may not be advantageously made. The
very early work of Biot and Arago, Thomson, Dulong and Berzelius, Lavoisier, and others
can be neglected, and, as in the case of oxygen, we need consider only the results obtained by
Dumas and Boussingault, Regnault, and several more recent investigators.

Taking air as unity, Dumas and Boussingault * found the density of atmospheric nitrogen

to be—
.970

972
974
a eae

: Mean, . 972+. 00078
For oxygen, as was seen in our discussion of the O:H ratio, the same investigators found a
mean of 1.10567, +. 00006. The ratio between this and the nitrogen figure is 16:14.0657, +. 0118.

By Regnault? much closer work was done. He found the density of atmospheric nitrogen
to be as follows:

. 97148 . 97155
. 97148 . 97108
. 97154 . 97108

Mean, . 97137 +. 000062

For oxygen, Regnault’s mean value is 1.105633, +.000008. Hence, ee as before,
N = 14.057, + .0009.

Both of the preceding values are affected by a correction for the difference i in volume between
the weighing globes when full and when empty. This correction, in the case of Regnault’s data,
was measured by Crafts,? who gives 1.10562 for the density of oxygen, and 0.97138 for that of
nitrogen. The changes are so small that the ratio remains practically unaltered. The correc-
tion in this particular instance is negligible.

Von Jolly,‘ working with electrolytic oxygen and with nitrogen prepared by passing air
over hot copper, compared the weights of equal volumes of the two gases, with results as follows:

 

1 Compt. Rend., 12, 1005. 1841. 8 Compt. Rend., 106, 1664.
2 Compt. Rend., 20, 975. 1845. ‘Annalen der Physik. (2), 6, 520. 1879.
No. 8.] ATOMIC WEIGHTS—CLARKE. 35

 

 

Oxygen. Nitrogen.
1. 442470 1. 269609
1. 442579 © 1, 269389
1. 442489 1. 269307
1. 442570 1. 269449
1, 442571 1. 269515 .
1. 442562 1. 269443
1, 442478 1. 269478
Mean, 1. 442545 +. 000013 Mean, 1. 269455+. 000024

The ratio, when O=16, is N=14.0802 + .0003. Corrected by Rayleigh, the ratio between
the weights becomes 14.0805.

The next determination in order of time is Leduc’s.1 He made nine measurements of the
density of atmospheric nitrogen, giving a mean of .97203, with extremes of .9719 and .9721;
but he neglected to cite the intermediate values. ‘Taking the three figures given as represen-
tative, and assuming a fair distribution of the other values between the indicated limits, the
probable eroor of the mean is not far from 0.00002. For oxygen he found 1.10514 + .000032.
The ratio between the two densities is 16 : 14.0729 +. 0005.

Lord Rayleigh,? who prepared nitrogen from the atmosphere by several methods, and
weighed it in a standard globe in direct comparison with oxygen, obtained the following weights:

 

 

Oxygen. Nitrogen... |
2. 6272 2.310385
2. 6271 2. 31026
2. 6269 2. 31024
2. 6269 2. 31012
2. 6271 _ 2.31027
Mean, 2. 62704+. 00004 2. 31025 +. 000025

In a later paper * Rayleigh gives the following additional weights for atmospheric nitrogen,
which are directly comparable with the foregoing series:

2. 31017 : 2. 31001 2. 31028
2. 30986 2. 31024 2.31163
2. 31010 2. 31010 2, 30956

Including these figures with those of the first series, the mean of all becomes 2.31023 + .00008.
Correcting these data for the compression of the empty globes, the mean weights become,
foroxygen, 2.6276+.00004, and for nitrogen, 2.31079+.00008. The ratio between them is
16 : 14.0704 + .0005.

The combination of these determinations is as follows:

Dumas and Boussingault............---.-2222eece eee eee 14. 0657-£. 0113
Resnaul th o.c2c2.8 cesnsisseeaseesaneeeeemaseresanogueskes 14. 0570-4. 0009
Voit JOY sccassvcctiinnneiecsteae node na lddueee see ee Saas 14. 0805 +. 0003
TUG socio istinade lng soda araisie sae Meee eeeres meeeas ees 14. 0729+. 0005
Rayleigh. ..........2-22202ece cece cece eee ee eee eee e ees 114, 0704+. 0005
General mean. .......-.---.2----- eee eee eee eee eee 14. 0758+. 00022

Now, to correct this mean for the argon contained in the nitrogen. Good measurements
have shown that normal air contains, by volume, 0.937 per cent of argon, and 78.122 of nitrogen.
The density of argon, referred to the oxygen standard, is 19.940. Applying these values, the
final figure for nitrogen, derived from air, becomes 14.0052 + .00022, a result which is in harmony
with others to be considered presently.

In Rayleigh’s investigation of the density: of nitrogen it was found that nitrogen from
chemical sources was lighter than that extracted from the atmosphere. This led to the dis-

 

1 Compt. Rend., 113, 186. 1891. 2.Proc. Roy. Soc., 53, 134. 1893. 8 Proc. Roy. Soc., 55, 340. 1894.
36 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

covery of argon, to which reference has already been made. In two of his memoirs Rayleigh
has given determinations of the density of ‘‘chemical nitrogen’’ obtained from nitrous oxide,
nitric oxide, ammonium nitrite, urea, and magnesium nitride, and the gas from all these sources
is precisely the same. His weights, given now as one series, and representing the same volume
as those previously cited, are as follows:

2. 30148 from nitric oxide

2. 29890 from nitric oxide

2. 29816 from nitric oxide

2. 30182 from nitric oxide

2. 29869 from nitrous oxide

2. 29940 from nitrous oxide

2. 30074 from nitrous oxide

2. 30054 from nitrous oxide

_ 2, 29849 from ammonium nitrite ©

2. 29889 from ammonium nitrite
2. 29870 from ammonium nitrite
2. 29850 from urea
2. 29918 from magnesium nitride

Mean, 2. 29949 +. 00024

 

Corrected for the compression of the empty globe, this mean becomes 2.30005. The weight of
an equal volume of oxygen wasfound tobe 2.62760 + .00004. Hence theratiois16 : 14.0055 + .0015.

Two determinations of density for ‘‘chemical nitrogen’’ are given by Leduc.? In mean,
the value found, referred to air as unity, is 0.96717 +.00008. This, combined with the figure
already cited for oxygen gives, as the value for the ratio under discussion, 16 : 14.0025+.0012.
There are also two determinations by Gray,? whose comparative weights at 0° and 760 mm.,
are as follows:

 

 

Nitrogen. Oxygen.

. 32286 . 36889

- 32275 . 36879
Mean, . 322805+. 00004 Mean, . 36884+. 00003

Hence the ratio 16 : 14.0030+.0021.

It is evident that the data given by Leduc and Gray are overvalued in comparison with
Rayleigh’s much larger series of determinations. The general mean, however, as shown in the
following combination, can not be far from the truth:

Rayleigh. 20.2.5... cil eee e eevee ee den eee 2-14, 0055 -L. 0015

WiOdUC@ sas sca eisieels ea naeagese a eeneeeeee eee eel seer eewly 14. 0025+. 0012

GUBY siete oss cs basewidesdeaddaddesemeseesstde eee aivieteineeds 14. 0030+. 0021
General mean......-..-------- 0-222 ee eee eee eee eee 14, 0036 +. 00085

Hence the normal liter of nitrogen weighs 1.25066 grams.
For the density of nitrous oxide there are several series of measurements. Leduc * gives
three figures, as follows, referred to air as unity:

1. 5304
1. 5298
1.5301

Mean, 1.5301 +. 00012

 

Combined with Leduc’s value for oxygen, this gives the density ratio
O, : N,O : : 32 : 44.3050 -+.0037.
By Rayleigh there are two series of determinations,® made at different times. In the earlier
series the gas was possibly contaminated by traces of nitrogen, in the second series the nitrous

 

1 Proc. Roy. Soc., 55, 340, 1894, and 57, 266, 1895, 4 Ann. Chim. Phys. (7), 15, 35. 1898,
2 Ann. Chim. Phys. (7), 15, 33. 1898. 5 Proc. Roy. Soc., 62, 204, 1897, and 74, 181, 1904.
8 Journ. Chem. Soc., 87, 1601. 1905.
No. 3.] ATOMIC WEIGHTS—CLARKE. 37

oxide was purified by condensation at the temperature of liquid air. The weights of nitrous
oxide filling his standard globe are subjoined.

 

 

1897 series. 1904 series.
3. 6359 3. 6368
8. 6354 3. 6360
8. 6364 3. 6362
3. 6358 3. 6363
3. 6360 3. 6367
3. 6366
Mean, 3. 6359 3. 6354
Mean, 3. 6363

These are so nearly together that I venture to treat them as one series, in mean
3.6361 +.000093. The weight of the same volume of oxygen was 2.6276+.00004. The value
of the ratio, therefore, is 32 : 44.2819 + .0037.

The measurements by Guye and Pintza ‘are stated so as to show the weight of a normal
liter of nitrous oxide. The figures are, in grammes—

1, 97762

1. 97707
1. 97760

Mean, 1. 97743+. 00015

 

The weight of a normal liter of oxygen is 1.42905+.000013. Combining this with Guye
and Pintza’s figure the ratio becomes 32 : 44.2796 + .0040.
The three independent values for the density ratio O,:N,0, combine as follows:

DedtG is cacw did edd oweganyenmielss Gebietes 44.:300022.0037
Ray leah ci Gai catanwasnannsensisansancane eens es 44, 2819 +. 0037
Guye and Pintza............2..-. 2 eee cee cence eee ee... 44, 2796+. 0040

General mean.......-..---2.---+---------22--2 ++ 2+ 44, 28894. 0023

This mean corresponds to a normal liter-weight for nitrous oxide of 1.97785 grammes.

It is convenient at this point to consider the volumetric analysis of nitrous oxide made
by Jacquerod and Bogdan.? A measured volume of the gas was decomposed by an electrically
heated spiral of iron wire, and the volume of the residual nitrogen was measured afterwards.
Then, with the known densities of the two gases, the ratio between them was easily calculable.
Reduced to uniform conditions, one liter of nitrous oxide gave the following volumes of nitrogen:

1. 00737
1, 00698

1. 00714
1. 00718

Mean, 1. 00717 ++. 000054

 

To this value, however, a correction is yet to be applied, namely, for the increase in volume
of the iron wire consequent upon oxidation. This demands a deduction of 0.00030, which
reduces the mean to 1.00687. That is, one liter of nitrous oxide, decomposed, yields 1. 00687
liters of nitrogen. Hence the following calculation:

1 liter N,O weighs......---.--- 2-224... eeee eee ce eee 1.97785 grammes.3
: 1.00687 N, weighs.........-.-.-...- 1. 25066 X1. 00687=1. 25925‘

Oxygen in ‘NO, 0.71860 8“

 

 

1 Compt. Rend., 139, 677, 1904. Corrected in C. R., 141, 51. 1905. The corrected figures are used here.

2 Journ. Chim. Phys., 3, 562. 1905.

3 Jaquerod and Bogdan assume, for the liter-weights of Nz and N,O, 1.25045 and 197772, respectively. I here use the weights previously com-
puted in this chapter. Jaquerod and Bogdan find N=14.015.
38 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI

From these data, O: N, : : 0.7186 : 1.25925 =28.0388, and N=14.0189+.0030. The
probable error is computed from the figures already given relative to the densities of the gases.

For the density of nitric oxide there are three modern investigations. First, by Gray;
second, by Guye and Davila,? and third, ,by Scheuer.2 Gray gives two series of weights, in which
nitric oxide is directly compared with an equal volume of oxygen. Two supplementary de-
terminations are cited as additions to series 2.

Oxygen. * NO 1. ' NO 2.
- 38230 . 85845 . 35851
. 38229 . 35852 - 35848
- 88227 . 35851 . 35852
. 38225 . 85849 . 35850
. 38226 . 85859 . 35848
. 38230. . 85856 - 35855 |

Mean, . 38228-+. 0000058

Mean of all, .35851 +. 0000076

From niiene wees the crude density ratio is
| O, : NO : : 32 : 30.0102. 0007.

Guye and Davila prepared their nitric oxide by three distinct methods, and ‘obtained the
following figures for the normal liter-weight.

I. II. IIL.
1, 3406 1. 3408 1. 3399
1.3402 1, 3398 1. 3403
1. 3401 1. 3400
1. 3407 1. 3408
1, 3398 1. 3402
1, 3402 1, 3402

The mean of the 14 determinations, taken as one series, is 1.3402 +.000056. With the
accepted weight of the normal liter of oxygen, 1.42905 + .000013, the density ratio becomes—

O,: NO : : 32 : 30.0104+. 0011.
Scheuer’s s figures for the weight of a liter of nitric oxide are as follows:

hy , 34047 ae 1. 34053 . 1. 34064 1. 34059
1. 34073 1. 34058 1. 34039 1. 34076
1. 34067 1. 34060 1. 34066 1. 34062
1534049 1. 34062 1. 34068 1. 34042

Mean, 1. 34059+. 000018

Corrected (Scheuer) the weight of the normal liter is 1: 34014 grammes, and—
O?: NO : : 32 : 30. 0091+. 0005.
The three series of determinations combine thus—

GIO scenes exes sos awe wiv ecmusscaienevensbeceseres NO=30. 0102+. 0007
Guye and Davila... ...-......-2--.-2- eee eee ee .--. NO=30. 0104+. 0011
SChOUGE eects oe se Soca s cee ss eearadnesaeem men NO=30. 0091+. 0005

General mean... .. 21s i siea meen ones eadiees NO=30. 0095 +. 00038

This corresponds to a normal liter-weight of 1.34016 + .000017.

The density of gaseous ammonia, according to Leduc,‘ is 0.5971, referred to air as unity.
But this figure represents only a single determination, with material of doubtful purity, and
need not be considered further. Guye and Pintza,' with carefully purified ammonia, made
five determinations of ee which gave the subjoined results for the weight of the normal
liter:

77080 -77099
-77069 .77076
.77073

Mean, .77079 + .000035

 

3 Anzeiger Wien Akad., 50,378. 1913. 5 Compt. Rend., 141, 51. 1905.
4 Ann, Chim, Phys. (7), 15, 39. 1898,

1 Journ. Chem. Soc., 87, 1601. 1905.
2 Compt. Rend., 141, 826. 1905.
No. 3.] ATOMIC WEIGHTS—CLARKE. 39

_ Perman and Davies‘! made three series of determinations, by two methods; but the first
series, with commercial ammonia, is to be rejected. The other series gave the following figures
for the weight of one liter of the gas:

 

I, II.

-7709 77094
7711 77094
7712 -77090
7713 -77088
7711 -77091
7713

— Mean, .77094

Mean, .77115

Corrected, by reduction to latitude 45° and sea level these two series taken as one, give in
value 0.770855 + .000034 grams as the weight of a normal liter of ammonia.
There is also a later series of determinations by Scheuer,? as follows:

-77098 77121
-79124 -77088
-77083 77115
-77102 -77091

Mean, 0.77108 + .000056.
Corrected (Scheuer), this gives as the weight of the normal liter,

0.77077 + .000056 grams.
The three series combine thus:

 

 

Guye and Pintza......... .77079 +.0000385
Perman and Davies........ 770855 + .000034
Scheuer. ...-.-..----.----- .77077 +.000056

General mean...... .77081 + .000022

Hence, with the accepted value for the normal liter of oxygen, 1.42905 +.00013, the crude
density ratio is—
O, NH, : 732 :17.2605 + .00043.

The law of Avogadro, that equal volumes of gases contain equal numbers of molecules, is
rigorously true only for ideally perfect gases. For gases as they actually occur it is approxi-
mately true, but with varying degrees of divergence. The approximation is close for the
so-called permanent gases, while those which are easily liquefiable conform less nearly to the
law. In order, therefore, to compute molecular weights from observed gaseous densities, it is
necessary to apply corrections to the experimental data, or else to employ methods of deter-
minations of great manipulative difficulty. By measuring densities at very low pressures,
quite close approximations to the truth may be obtained, and observations at high temperatures
are also nearly valid. For example, Rayleigh * from gaseous densities at very small pressures,
obtained the following value for nitrogen, as compared with the standard, oxygen:

N,=28.018, and N=14.009.
On the other hand, by measuring the density of nitrogen at 1067.4°, Jaquerod and Perrot *
found—
— N,=28.0155, and N=14.0077.
' These values are probably not far from the truth, and are obviously well in accord. At low
pressures and at high temperatures gases are more nearly in agreement with Avogadro’s law
than they are under ordinary conditions.

In the case of the oxygen-hydrogen ratio, the density corrections were determined by actual
measurement of the volumes in which the two gases combined, a method which is not always

 

1 Proc. Roy. Soc., 78A, 28. 1906. Perman and Davies give still other density determinations for different temperatures and pressures.
2 Anzeiger Wien ‘Akad., 49,38. 1912,

3 Proc. Roy. Soc., 73. 153. 1904.;
4 Compt. Rend., 140, 1542. 1905.
40 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

applicable, or at least not conveniently so. It is easier to compute the corrections from physical
data, and for this purpose various methods have been proposed.1

The following formule, based upon the celebrated gas equation of Van der Waals, are
according to Guye,? available for the reduction of gaseous densities to true molecular weights:

1. It,
_ (RxmT)L RL
Mea Oe M=(ya,) d—b,)

Equation I applies to the permanent gases, that is, to those which are liquefiable only
below 0°. Equation II applies to the easily liquefiable gases. R is the gas constant, and
according to Berthelot ® its value is 22.412. The constant m, of equation I, is given by Guye
the value 0.0000623.' T represents the critical temperature, on the absolute scale; L is the
weight of one liter of gas at 0°, 760 mm., sea level, and latitude 45°; and M is the molecular
weight. The symbols a and 6 are the constants of the Van der Waals equation, which vary for
different gases, and in II are brought to the standard temperature and pressure. ,

In any given case the use of these formule requires a knowledge of the constants a and 6.
These can be deduced from the compressibilities and coefficients of expansion of a gas, or from
the critical constants. The latter method is the one adopted by Guye, and with one exception
it will be followed here. Guye gives the required data in form ready for use,* and they yield
results which appear to be trustworthy. Applied to the densities given in the preceding pages
they give the following reductions:

Nitrogen, chemical.—L = 1.25066. T=127.5°. (1+a) (1—b=1.00100. Hence N= 14.0058
+.00085. From the figures given for atmospheric nitrogen, N=14.0074+.00022. The
weighted mean is N= 14.0073 +.0002.

Mitrous oxide.—Li = 1.97785. (1+a,) (1—6,) = 1.00733. Hence N,O = 44.0060 + .0023.
The crude density ratio gives 44.2889 + .0023, showing that the correction is large.

Nitric oxide.—_L = 1 34016. T=179.5°. ‘@=0.00257. 6=0.00115. Hence NO=30.0081 +
-00038.

Ammonia.—The crude density ratio gave NH, = 17.2605 + .00043. This has been reduced
by means of compressibility data. Perman and Davies, who measured the compressibility,
give the multiplying factor 0.9867, whence NH,=17.0309. Jaquerod and Scheuer,’ by a
different formula, and using only the density determinations of Guye and Davila, find NH,=
17.0148. If H=1.0078, N=14.0075, Perman and Davies’ method, or 13.9914 by Jaquerod
and Scheuer. The first value is apparently the best and will be adopted here.

There are now five independent values for N, as follows:

From analyses‘and syntheses...............2.--.--4-- N=14.0079 + .00018
hom @pndily Gl Necaasnwersncumcutmanmeaswanakeneee N=14,0073-+.00020
From density of N,O........-...2..----+-2--ee eee N=14.0030+ .00105
From density of NO............-2--+2-+2--eeeee eee . -N=14.0081 + .00038
From density of NH;...-. ieee SEea@ eee es N=14.0075 + .00043

Gootietall Tea DY. <a. .25 22s ere. S esos citisicis erate: eis Sele zis. N=14.0076+ .00012

From compressibility data Rayleigh’ found from N,, N= 14.008, and from N,O, N = 13.998.
His low pressure value, as previously cited, was N=14.009, and J aquerod and Perrot, at high
temperatures, found N=14.0077. To include these values in the general mean would change
the final result inappreciably, if at all, and they may therefore be disregarded. They have,
however, confirmatory significance.

 

1 See D. Berthelot, Journ. Physique (3), 8, 263. 1899. Leduc, Ann, Chim. Phys. (7), 15,1. 1898. . Guye and Friederich, Arch. Sci, Phys. Nat.
(4), 9, 505, and 13, 559. Buys, Journ. Chim. Phys., 8, 321, and 5, 203, also Compt, rend., 138, 1213, and 140, 241, There is a copious literature upon
this subject.

2 Journ. Chim, Phys., 3,321. 1905.

3 Zeitsch. Elektrochem., 1904, 621. In Journ, Physique (3), 8, 527, Berthelot gives values of @ and 0 for several gases,

4 When Guye gives two or more figures for (1+a) (1—b) I take the average.

6 Compt. Rend., 140, 1384. 1905. From NO Jaquerod and Scheuer found N=14.005.

Proc. Roy. Soc., 74, 446. 1904
No. 3.] : ATOMIC WEIGHTS—CLARKE. Al

Some of the determinations utilized in the foregoing combination are evidently overvalued,
especially the figure derived from atmospheric nitrogen. The “probable error” here assigned
to N, simply indicates the weight which it should receive in calculating other atomic ratios.’

As a check upon the other determinations of the atomic weight of nitrogen, Guye and
Pintza? have determined the composition of ammonia by volume. The gas was decomposed
by a spiral of platinum wire heated to redness, and from the density of the mixed gases, N,+3H,,
compared with the known densities of nitrogen and hydrogen, the required datum was
calculated. For the weight, in grammes, of a normal liter of the gaseous mixture, the follow-

ing figures were obtained:
0.38044
2 0.38055
0.38046

Mean, 0.38048 + .000024

 

Corrected for traces of SO, and SO,, this becomes 0.37989. If the weights of the normal
liters of N, and H, are 1.2507 and 0.08987, respectively, the two gases in ammonia are com-
bined in the ratio 1 : 3.00172. Applying this datum to the densities of nitrogen and hydrogen,
and assuming H=1.0078, N=14.017, with a probable error, not exactly calculable, greater
than +.0017. To combine this figure with the value already found would change the latter
inappreciably. Indeed, Guye and Pintza regard their determinations as inferior to those
made by other methods, and publish their results only as a confirmation of the low value for
N, as compared with the value 14.04 which had been in general acceptance form any years.
The aberrant value derived from the volumetric analysis of nitrous oxide can also be neglected.

THE CARBON-OXYGEN RATIO.

The ratio between carbon and oxygen, or in other words, the atomic weight of carbon, has
been directly determined by several methods. It has also been indirectly computed from
analyses of silver salts, such as the acetate; but that group of ratios will be considered under
another heading. The early attempts to estimate it from analyses of hydrocarbons, have
now only historic value, and can be omitted from the present discussion. The direct measure-
ments of the ratio represent three distinct processes: .

First, by the combustion of carbon itself.

Second, by the combustion of carbon monoxide.

Third, by determining the density of gaseous compounds of carbon.

The first of these methods was used by Dumas and Stas * in 1840, and a year later by
Erdmann and Marchand.‘ In both investigations weighed quantities of diamond, of natural
graphite, and of artificial graphite were burned in oxygen, and the amount of dioxide produced
was determined by the usual methods. The graphite employed was purified with extreme
care by treatment with strong nitric acid and by fusion with caustic alkali. I have reduced
all the published weighings to a common standard, so as to show in the third column the
amount of oxygen which combines with a unit weight (say 1 gramme) of carbon. Taking
Dumas and Stas’ results first in order, we have from natural graphite:

. ©. C0>. Ratio.
1.000 3.671 2.6710
.998 3.660 2.6673
.994 3.645 2.6670
1.216 4,461 2.6686
1.471 5.395 2.6676

 

Mean, 2.6683 +.0005

 

. 1 Fora general discussion of the atomic weight of N, see Guye’s lecture delivered before the Chemical Society of Paris, June 10, 1905. Also
Compt. Rend., 140, 1386, and 144, 1360; and Ber., 39, 1470. Two papers by Berthelot, of a controversial nature, are in Compt. Rend., 144, 76 and
260, and one by Leduc in Compt. Rend., 146, 399. . ; ‘

4 Compt. Rend., 147, 925. 1908.
*Compt. Rend., 11, 991. Ann. Chim. Phys. (3), 1, 1.
‘Journ. prakt. Chem., 23, 159.
42 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

With artificial graphite:
C. CO;. Ratio. ‘
-992 3.642 2.6714
998 3.662 2.6693
1.660 6.085 2.6657
1.465 5.365 2.6621

rae ‘ Mean, 2.6671+.0014
And with diamond:

C. CO,. Ratio.
708 2.598 2.6695
864 3.1675 2.6661
1.219 4.465 : 2.6628
1.232 4,517 2.6664
1.375 5.041 2.6662

Mean, 2.6662 + .0009
Erdmann and Marchand’s figures for natural graphite give the following results:

Cc. Cc0,. Ratio.
1.5376 5.6367 2.6659
1.6494 6.0384 2.6609
1.4505 5.31575 2.6647

In one experiment 1.8935 grammes of artificial graphite gave 6.9355 grammes CO,. Ratio
for O, 2.6628. This, combined with the foregoing series, gives a mean of 2.6636 + .0007.
With the diamond they found:

C. CO,. Ratio.
8052 2.9467 2.6596
1.0858 3.9875 2.6632
1.3557 4.9659 2.6629
1.6305 5.97945 2.6673
-7500 2.7490 2.6653

Mean, 2.6637 + .0009

In more recent years the ratio under consideration has been carefully redetermined by
Roscoe, by Friedel, and by Van der Plaats. Roscoe 1 made use of transparent Cape diamonds,
and in a sixth experiment he burned carbonado. The combustions were effected in a platinum
boat, contained in a tube of glazed Berlin porcelain; and in each case the ash was weighed and
its weight deducted from that of the diamond. The results were as follows, with the ratios
stated as in the preceding series:

C. C0... Ratio.

1, 2820 4, 7006 2. 6666

: 1, 1254 4.1245 2. 6649
1. 5287 5. 6050 2. 6665

. 7112 2. 6070 2. 6656

1.3842 , 5. 0765 2. 6675

- 4091 1, 4978 2. 6612

© Mean, 2. 6654 + .0006

Friedel’s work,? also upon Cape diamond, was in all essential particulars like Roscoe’s.
The data, after deduction of ash, were as follows:

Cc. C0,. Ratio.
. 4698 1. 7208 2. 6628
- 8616 3. 1577 2. 6649

Mean, 2. 6638 + .0004.

By Van der Plaats * we have six experiments, numbers 1 to 3 on graphite, numbers 4 and .
5 on sugar charcoal, and number 6 on charcoal made from purified filter paper. Each variety

 

1 Ann, Chim. Phys. (5), 26, 136. Zeit. anal. Chem,, 22, 306. 1883. Compt. Rend., 94, 1180. 1882. 8 Compt. Rend., 100, 62. 1885.
2 Bull, Soc, Chim., 41, 100. 1884.
No. 3.) ATOMIC WEIGHTS—CLARKE. 43

of carbon was submitted to elaborate processes of purification, and all weights were reduced
to a vacuum standard. The data, with ash deducted, are subjoined:

C. CO>. Ratio.
1. 5.1217 18. 7780 2. 6664
2. 9.0532 33, 1931 2. 6664
3. 13.0285 47. 7661 2. 6663
4. 11.7352 43, 0210 2. 6660
5. 19. 1335 70. 1336 2. 6655
6. 4.4017 16. 1352 2. 6657

; ; ; Mean, 2. 6660 + .0001.
This combines with the previous series thus:

Ratio. Atomic weight C.

Dumas and Stas, first set... -.---. 0-22 e eee eee eee eee eee 2. 6683 + .0005 11. 9926
Dumas and Stas, second set........... 22-22-00 - eee e eee eee eens 2. 6671 + .0014 11. 9981
Dumas and Stas, third set.........-2.. 2-22-22 e ee eee eee eee ee ee 2. 6662 + .0009 12. 0031
Erdmann and Marchand, first set..........-.----------0e2-- ees 2. 6636 + . 0007 12. 0138
Erdmann and Marchand, second set.....-.--.---------e+---+- 2. 6637 + . 0009 12. 0134
TROSC0G sro. seanciessle heave tanzes. gate oe Ses os Geared ME ie bide eee naiee eens 2. 6654 + . 0006 12. 0057
Priced else aducne asiialee vc sie Ser dager ecneme eee tht nate eine 2. 6638 + . 0007 12. 0129
Vani der: Plaats scseeweas s vit's cd Soseetiocn ieee meenee eee d -..-- 2.6660 + .0001 12. 0030 ‘

General Miéah. 22.0 pecseaedes does oss eetaiansews seca 2. 6660 + . 0001 12. 0030 + . 0005.

The effect of this combination is to give the work of Van der Plaats overwhelming weight,
to which it is perhaps not entitled. The other determinations practically vanish.

According to Scott! all of the foregoing determinations are subject to an important cor-
rection, namely, a reduction to weight in vacuo. This correction was applied by Van der
Plaats, at least partially, but Scott lays emphasis upon the change in volume of the potash
solution in which the carbon dioxide was absorbed and weighed. The corrections, as applied
by Scott, are given in the following table, in which the total reduced weights of carbon and
dioxide are used instead of the individual weights of the separate experiments:

Total C. Total CO,. Ratio. Atomic weight.

Dilias and Stasinccscedseccvsicineecicsscsecee 16. 1994 59. 4201 2. 66804 11. 9938
Erdmann and Marchand..........-.....----- 12. 1636 44, 58537 2. 66547 12. 0054
ROSCOC! Sons icct Sheek bee ee oi sy s8 Sasa ... 6.4428 23. 6275:° 2. 66727 11. 9973
Bried @lieins2acetenciuwenes idesas sesemensses 1. 33185 4. 8818 2. 66543 12. 0056
Van det Plaster cccccceecee seis tess comers 62. 5115 229. 1836 2. 66630 12. 0017

If to these figures we assign the relative weights given in the previous combination the
final mean will be identical with that of Van der Plaats as before, and C=12.0017, +.0005.
Scott adopted the unweighted average of the five series given above, and made C= 12.0008.

The second method for determining the atomic weight of carbon was employed by Stas?
in 1849. Carefully purified carbon monoxide was passed over a known weight of copper oxide
at a red heat, and both the residual metal and the carbon dioxide formed were weighed. The
weighings were reduced to a vacuum standard, and in each experiment a quantity of copper
oxide was taken representing from eight to twenty-four grams of oxygen. The method, as
will at once be seen, is in all essential features similar to that usually employed for determining
the composition of water. The figures in the third column, deduced from the weights given by
Stas, represent the quantity of carbon monoxide corresponding to one gramme of oxygen:.

0 CO,. Ratio.
9. 265 25. 483 1. 75046
8. 327 22. 900 1. 75010
13. 9438 38. 351 1. 75040
11. 6124 31. 935 1. 75008
18. 763 51. 6055 1. 75039
19. 581 53. 8465 1. 74994
22. 515 61. 926 1. 75043
24, 360 67. 003 1. 75053

Mean, 1. 75029 + .00005.

 

1 Journ, Chem. Soc., 71, 550. 1897. 2 Bull. Acad. Belg., 1849 (1), 31. Oeuvres Complates, 1, 287.
44 MEMOIRS NATIONAL ACADEMY OF ‘SCIENCES. [Vou. XVI.

Hence CO = 28.0046, and C = 12.0046 + .0008.

This work of Stas was also criticized by Scott,! in connection with the determinations by
the first method. The process employed is subject to several possible errors, two of them
being especially serious. First, the carbon monoxide may have contained hydrogen or hydro-
carbons. Secondly, the copper oxide, which was prepared by calcining copper nitrate, almost
certainly occluded nitrogen. The value found for C, however, is probably not very far from
the truth, and it is not unlikely that errors in opposite directions tended to compensate one
another.

For the density of carbon monoxide there are available determinations by Leduc? and
Rayleigh. Leduc used a globe which had a capacity of 2.9440 grammes of air. Filled with
CO it held the following weights, giving the accompanying densities:

Weight CO. Density.
2. 8470 . 96705
2. 8468 - 96698
2. 8469 - 96702

 

. 96702 + .000015.

This density, combined with Leduc’s determination of the density of oxygen, 1.10514+
BOGIES gives the crude ratio—
O, : CO : : 32 : 28.0007 +. 0010.
Rayleigh’s determinations may be stated in the following form: A globe which held
2.62760 + .00004 grams of oxygen, held of carbon monoxide—

2. 29843
2. 29852
2. 29854

Mean, 2. 29850 + . 000024.

 

Corrected for the compression of the globe when empty this becomes 2.29906 + 000024.
From these data the crude value for CO is 27.9989+.0012. Combining this with Leduc’s
determination the general mean becomes—

CO = 28.0000 + .00077.

Rayleigh, it must be observed, prepared his three samples of carbonic oxide by three
distinct methods, and the concordance in their weights gives strong assurance of their purity.

For the calculation of the true molecular weight of carbon monoxide from this crude density
ratio, the critical data cited by Guye‘ are available. The mean of two sets of critical constants
gives (1+a) (1—6)=1.00109, and T=132.7°. Applying these figures by the formula ‘given
under nitrogen, the molecular weight becomes .

CO=238. 0031 + .00077.

The density of carbon dioxide has been. determined by many investigators,° but the earliest
measurements have now only historical interest. In 1845 Regnault * published five determina-
tions of the density, referred to air as unity, and they were the first to be worth consideration
now. His figures are as follows:

1.52915 1.52906

a Mee

1.52900 1.52915
1.52915

Mean, 1.52910 +.000032

 

1 Loc. cit. See also Richards, Amer. Chem. Journ., 20, 701. 1898.

1Compt. Rend., 115, 1072. 1893.

3 Proc. Roy. Boe, 62, 204. 1897.

4 Journ. Chim. Phys. 7 8, 342, 1905.

6 The early determinations are well summarized in Van Geun’s monograph.
Compt. Rend., 20, 993. 1845.
No. 3.] ATOMIC WEIGHTS—CLARKE. 45

Corrected by Crafts,! for compression of the empty globe, this becomes 1.52897 + .000032.
For the density of oxygen, Regnault’s corrected value is 1.10562 + .000008.

Hence O, : CO, : : 32 : 44.2530 + .00098.

In three concordant measurements, which are not given separately, Leduc ? found for CO,
the density 1.52874. This figure, combined with his value for oxygen, already cited, gives for
CO, the density ratio 32 : 44.2667. Rayleigh * gives a single figure for the density of CO,,
namely, 1.52909. For oxygen he found 1.10535. Hence the ratio 32 : 44.2673.

The three determinations by Guye and Pintza‘ are stated in the form of normal litre-

weights, as follows:
1.97684

1.97676
1.97681
Mean, 1,97680 + .0000176

With the accepted figure for the weight of a litre of oxygen, 1.42905 +4 .000013, the ratio
‘becomes’
O, : CO, ::32 : 44.2654+.00071.

To the values deduced from Leduc’s and Rayleigh’s data we may arbitrarily assign equal
weight with the mean of Regnault’s series. The four determinations then combine thus:

Regnal tact cae cicjindideise it ela salideriacegesiues 44.2530 +.00098
eduereke oicce snes seek eee ce ds annesureeetan 44,2667 + .00098
Rayleigh........ ie it aE cal aa iedae a a 44.2673 + 00098
Guye and Pintza...:..............2-22.------- Siancicie Bee 44.2654+ .00071

General MEAN... - 2.00.0. e ee eee eee ee eee 44,2632 + .00038

Regnault’s figure is probably too low. Its omission would raise the general mean to 44.2674
but such a procedure is questionable. I prefer therefore to leave the combination unchanged,
except for the necessary reduction by means of the critical constants. For these, based on the
mean of determinations by Amagat and Keesom, Guye® deduces (1+a,) (1—6,) =1.00687.
Applying this value we have for the molecular weight under consideration,

CO, = 43.9966 + .00038.

There are a few other data relative to the oxides of carbon yet to be considered.
Rayleigh * has compared the density of carbon monoxide at atmospheric pressure with its
density at pressures between 75 and 150 millimeters of mercury. The molecular weights
deduced are, for normal pressure, CO=28.000, for low prussures, 28.006, when O=16.
Hence C= 12.006.

A comparison of the gases at high temperatures has been made by Jaquerod and Perrot.’
They measured the expansion of the two carbon oxides up to 1067.4°, applied their results to
the mean densities found by Leduc and Rayleigh, and obtained the following molecular values:

CO =28.009, and C=12.009.
CO, = 43.992, and C=11.992.

These figures are interesting for comparison with those previously discussed, but can hardly
be used in a general combination.
. Another group of data from which the carbon-oxygen ratio can be deduced is found in the
density determinations of certain organic compounds. The older measurements need not be
considered, but three recent investigations have some value.

 

1Compt. Rend., 106, 1664. § Journ. Chim. Phys., 3, 337. 1905.
4 Ann. Chim. Phys. (7), 15, 34. 1898 6 Proc. Roy. Soc., 73, 153. 1904.
® Proc. Roy. Soc., 62, 204. 1897. 7Compt. Rend., 140, 1542. 1905.

‘Compt. Rend., 141, 51. 1905.
46 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVL

First, the density of methyl oxide (CH,),0, as determiried by Baume.’ Two series are
given, with the subjoined values for the weight of a normal litre:

I. I.
2.10912 2.10925
2.10886 2.10941
2.11045 ‘2.11026
2.10920 2.10936
2.10948 - . 2,11005
2.11003 2.10977
2.10947

 

 

3: Mean, 2.10968+.00011
Mean, 2.10951+.00014

A small correction raises these means by 0.00001. Combined, the final value is 2.10961 +
.000084. With the critical data given by Baume, a,=0.03111, and 6,=0.00382. Reducing
these figures by means of the formula already cited, the molecular weight of methyl oxide
becomes 46.0306 +.0021. Hence, with H=1.0078,

C=11.9919 +.0010

a value which is almost certainly too low.
For the weight of a normal litre of methane, CH,, Baume and Perat 2 find the following
values: :

0.71690 0.71636
0.71657 0.71672
0.71633 0.71678.
0.71669 0.71725
0.71751

Mean, 0.71689 + .000098

Reducing this with the critical constants determined by Guye, CH, = 16.034 as 0022.

Hence
C= 12,0028 + .0022. ot

The density of ethane, C,H,, has also been determined by Baumé and Perrot.’ Two series
of measurements were made, giving the subjoined figures for the weight of the normal liter:

I. IT.
1.35671 1.35600
1.35679 1.35610
1.35671 1.35653
1.35652 1.35640
1.35700 1.35590
1.35640 1.35640

Mean of all as one series, 1.356455 +.000065. Reducing their data by means of the critical
constants, the authors find C,H,=30.119, and C=12.036. This value is evidently too high.

From the density of toluene, Leduc‘ has recently deduced the value 12.003. The determi-
nation, however, is not sufficiently explicit i in detail to admit of its use for present purposes.
Another value is calculable from Parson’s glucinum ratios; * namely, C = 12.007..

There is also a preliminary note, by Scott,® which gives, without details, the results of com-
bustions of naphthalene and cinnamic acid. In six analyses, 17.6175 grammes of naphthalene
gave 60.5355 of CO, Hence C=11.999. ‘In two analyses, 8.6153 grammes of cinnamic acid
gave 23.0413 of CO,. Hence C=12.0015.

 

1 Journ. Chim. Phys:, 6,46. 1908. Baume also gives data for methyl chloride, but they are not available for a good determination of mol
weight. : lecular

1Compt. Rend., 148, 39. 1909.

3 Compt. Rend. 148, 832. 1909.

4Compt. Rend., 148, 832. 1909.

6 See section on glucinum, later.

6 Proc. Chem. Soc., 25, 310.
No. 3.] ATOMIC WEIGHTS—CLARKE. 47

' Rejecting the aberrant value for C derived from the density of ethane, we have six available
values to combine, as follows:

From combustion of carbon...........02.eeeeeeeeeeeeee C=12.0017 + .0005
From combustion of CO............-.--+---e++e++--+-- +s C=12.0046 + 0008
From density of CO............------+ +02 eee serene ee eee C=12.0031+.0008
From density of CO,....-.-.-- 2+. 2-22 e eee eee eee eee ee O=11.9966 + .0004
From density of CHy....2......2-2-+2eeceeeeeeeeee+++-C=12.0028 + 0022
From density of ee =11.9919+.0010
General’ mean... 6.262.565 sceetessaeese sss .++--C=12.0000+ .00026

Later, in the discussion of all the ‘i sdamca ratios this value for carbon will be slightly
modified by values derived from data based upon the study of several silver compounds, and
also of sodium carbonate.

ADDENDA.
Since the foregoing calculations were completed, two additional memoirs on the atomic
weight of carbon have appeared. Stahrfoss* has determined the density of acetylene, ethylene,
and ethane. For ethylene, C. aH, the following values Aor the weight of the normal liter are—
1.26061
1.26178
1.26063
Mean, 1. "T.26101

Hence, according to Stahoposs, who reduced his data by the sete of critical constants,
C=12.004.
For ethane two series of determinations are given, as follows:

1. 2.
1.35632 1.35684
1.35621 1.35701
1.35668 1.35687 .-
1.35585

 

Mean, 1.35690

 

Mean, 1. 35627

From, the mean of all seven determinations, ‘reduced by means of the critical constants,
Stahrfoss finds C=12.006. He also applies corrections to the data given by Baumé and Perrot,
and from, them finds for the weight of the normal liter of ethane the value 1.35650, which is in
good agreement with his own measurements.

The determinations by Stahrfoss of the density of acetylene were not used in computing
the atomic weight of carbon, because of the tendency of the gas to polymerize and so to become
abnormally heavy. —

_ The density of ethylene. has also been determined by Batuocas," who obtained the following
values for the weight of the normal liter. —

Preliminary series. f°

 

1. 25985
1,26033
Final series.
1.26047 ant 1.26047
1.25999 a 1.26050
1.2604. 1.26052
1.26002 BP sz 1.26012
1.26022 : 1.26037
1.26052 1.26047
1.26045 ce 1.26029
1.26029 1.26046
1.26003 1.26031
1.26031
. Mean of. both series, 1.2603 +.0001.
1 Journ. Chim. Phys., 16,175. 1918. 2 Journ. Chim. Phys., 16, 322. 1918,

161566°—20-—_4
48 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

From the figures, by two. methods of reduction, Batuecas obtains the following values for
the molecular weight of ethylene:

By molecular volumes...........02.0.cceceeece cee ceeceeene ‘ot,=28 022
By critical constants..........2.00..00 02 cc cece eee ceceececees C,H, =28. 030
Meant sen ciesesoe ene eSeees Lcacuabaee ace SGcatecaeceeuees C,H,=28. 026

Hence, if H=1.0078, C=11.9974.
To combine these new values for carbon with the general mean found in the finai compu-
tations is pany necessary. They would change the ultimate value LETC

SYNTHESES AND DENSITY OF HYDROCHLORIC ACID.

The gaunidativs synthesis of hydrochloric acid, with reference to the. atomic weight of
chlorine, was first effected by Dixon and Edgar.' Chlorine, prepared by the electrolysis of
fused silver chloride, was weighed in liquid form. Hydrogen, obtained electrolytically from
barium hydrate, was occluded by palladium, and so weighed. A combustion globe was filled
with the chlorine, and the hydrogen, ignited by a spark, was burned in it... The excess of chlo-
rine was determined by absorption in potassium iodide, and subsequent titration of the liber-
ated iodine with thiosulphate solution. With corrected weights the following results were
obtained:

Weight H. Weight Cl. Ratio Cl:H.
0. 9993 35. 1666 35. 191
1.0218 . 85, 9621 35. 195

. 9960 35. 0662 35. 207
1. 0243 86.0403 35. 185
1.0060 "85.4144 ' 85. 208

- 9887 34. 8005 35. 198
1.0159 35. 7639 ' 35. 204
1.1134 39. 1736 : 35. 184
1.0132 35. 6527 35. 188

Mean, 35. 195 +.0019

The determinations by Noyes and Weber” were differently éiddunted. The hydrogen
was weighed in palladium ; but the chlorine was taken in the form of potassium chloroplatinate.
That salt was heated in a stream of hydrogen, and the loss in weight gave the weight of chlo-
rine taken. The hydrochloric acid produced was also collected a and weighed.

_ Two series of experiments were made, differing in detail. ‘In series I, the hydrogen was
prepared by electrolysis ¢ of sulphuric acid, and in series II from barium, hydroxide. The hydro-
chloric acid of series I was collected in water directly; but in series IT it was first condensed to
a solid by cooling with liquid air. The. two series, however, were concordant, and may there-
fore be treated here as one. The data obtained, reduced to a vacuum basis, were as follows:

 

 

A. Cl. HCl. Cl: ACL: H
0. 25394 8. 93293 9. 18695 35. 177 36. 178
. 28004 9. 85590 10. 13259 35. 195 36. 183
- 91821 18. 23468 18. 75359 35. 188 36. 189
. 67631 23. 79587 24, 47128 35. 185 36. 183
- 58225 20.48158  — .....--- 35.177 we eee
- 47989 16. 88423 17. 36310 35. 184 36. 181
- 64132 22. 55816 23. 20054 35. 175 36. 176
- 81608 28. 71691 29. 53167 35. 188 36. 187
. 83194 29. 28055 30. 11207 35. 195 36. 195
. 39074 13. 74926 14. 14078 35. 188 36. 188
. 75560 - 26. 58427. 27. 33926 _ 385.183 36. 182
. 77518 27, 26746 28. 04110 35. 176 36.174

Mean, 35. 1843 36. 1835

+. 0014 +. 0013

 

1 Phil. Trans., 205A, 169, 1905. ye 2 Journ. Amer. Chem, Soc., 30, 13. 1908.
No. 3.1 ATOMIC WEIGHTS—CLARKE. 49

Edgar’s! syntheses of hydrochloric acid resembled those of Dixon and Edgar, so far as
the preparation and weighing of the initial substances were concerned. The chlorine was
then burned in the hydrogen, at the end of a quartz tip, and the hydrochloric acid so produced
was condensed to solid form by means of liquid air. It was afterwards allowed to evaporate,
and passed through a quartz tube filled with mercury vapor, which removed any free chlorine.
The purified hydrochloric acid was finally condensed, either in a steel bomb or by absorption
in water, and weighed. The corrected weights and ratios are subjoined:

H. Cl. HCl. CLE. HCUH.
2. 1452 75. 5026 77. 6469 35. 196 36. 196
2. 0387 71. 7504 73: 7880 35. 194 36. 194
1.7762 62.5008 cee 35.188 eae
1. 9935 70. 1638 72. 1565 35. 196 36. 196
1. 6469 57.9671 cee eee SRSs aie
2. 1016 73.9662  — .....e ORI: uae
1.7254 60. 7162 62. 4401 "35. 190 36. 189
2. 0885 73.4991 "75. 5859 35. 192 36. 191

Mean, 35. 194 36. 193
+0008 = £0009

Upon reducing the HC1:H ratios to the Cl:H form the five sets of determinations combine

thus:
Dixon and Edgar, H:Cl........-.. etbreteceoe ie wcles eeo.srs's'sss BD. 190- =. 0019

Noyes and Weber, H:Cl..........+.---------- acresreses 35, 1843-4. 0014
Noyes and Weber, H:HCl.......-.---------- cea ... 85. 18354. 0013
Eider, HG) otsdcnedvede cts dawera wn newousdsiedie nd eee 35.194 +. 0008
Edgar, H:HOl..... 22.20.0202 20 ee eee eect ed ence ees eee 35.193 4.0009

General mean, 35. 1911+. 00049

That is, the atomic weight of chlorine, when H=1, is 35.1911. If O=16, then Cl=35.4652
+.0005. noe? “8g
For the weight of a normal liter of hydrochloric acid, Guye and Ter Gazarian* give the
subjoined figures: .
1. 6404 1. 6389
1. 6397 1. 6401
Mean, 1.6398 +'.00007

Reducing these by the method of critical constants, in which the term (1+a,) (1—},)
=1.00773, the molecular weight of HCl becomes 36.4693+.0015. Hence, if H=1.0078,
Cl=35.4615 + .0015.

The work of Gray and Burt* on the chlorine-hydrogen ratio is much more elaborate.
First, three series of determinations of the density of HCl are given, with the weight of one
liter of the gas at 0°, 760 mm., and at London, as follows:

 

 

 

I. Il. ITI.

1. 64053 , 1. 64022 1. 63950
1. 64004 1. 63999 1. 64013
. 1.64020 1. 63976 1. 68984
1. 63986 1. 64083 1. 64069
1. 64030 1. 64031
Mean, 1. 64016 1. 64021 1. 64017
+.00010 1. 64027 1. 64050
1. 64051
Mean, 1. 64023 1. 63992
+. 00008 1. 64001
Mean, 1. 64016

+. 00008

 

1 Proc. Roy. Soc., 81A, 216. 1908.

? Compt. Rend., 143, 1233. 1906. <A single determination of the density of HCl by Leduc (Ann. Chim. Phys. (7), 15,1. 1898) can hardly be
utilized here.

* Journ. Chem. Soc., 95, 1633. 1909.
50 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

The three series represent hydrochloric acid prepared by three distinct methods. Series I and
II are to be corrected by —0.00013 gramme, which represents gas absorbed by the walls of the
containing glass bulb. The corrected mean becomes 1.64011 -+.00005 grams, which, at sea level
and latitude 45° is equivalent to 1.63915 grammes, the weight of the normal liter. If we reduce
this with the critical constants used by Guye and Ter Gazarian,? it, gives HCl =36.4548, and with
H=1.0078, Cl=35.4470 + .0011, a very low value.

In order to ascertain the atomic weight of chlorine, Gray and Burt have measured the
compressibility of the HCl, and also determined its composition by volume. In the latter case
the gas was decomposed by heated aluminum, and the volume of hydrogen liberated from two
volumes of hydrochloric acid was measured. The volumes thus found were as follows:

1.00797 1.00781
1.00795 1.00779
_ 1.00790 1.00787
1.00790 1.00798

Mean, 1.00790 4.000017.

Using Morley’s value for the weight of a liter of hydrogen, 0.089872 + .0000028 gramme,
and with the atomic weight H =1.0078 + .00001, abs molecular weight of HCl is given by the

following equation: =)

1.639152
0.089872 * 1.00790 * 1.0078 = 36.4738

and Cl=35.4658+.0013. Gray and Burt, calculating with H=1.00762 (Morley’s value), find
Cl=35.459.

The data given by Gray and Burt for the compressibilities of oxygen and hydrochloric
acid are too complex to admit of detailed reproduction here. The normal liter of oxygen, weigh-
ing 1.42905 grammes, gave a, limiting density of 1.42762 grammes. That of HCl, 1.63915
grammes, gives a limiting density of 1.62698 grammes. The ratio between these limiting
densities is, the true ratio between the molecular weights according to Avogadro’s law. Hence,

1.62698 x 32 ,
amen © Weight HCl = 36.4687

Hence, assuming the probable errors. 0.00005 for HCl, and 0.000018 for O,, Cl=35.4609 +

.0014.
The three values derived from Gray and Burt’s determinations, now give the subjoined
values for the molecular weight of HCl, when H=1.0078.

By critical constants.........+...-+.+ ante eeeeeeereee HC1l=36.4548 + .0011
By volumetric analysis.....-.-..-.+.---2+-+2220e seen eee 36.4738 + .0013
By compressibility ineaeurenients, Soa radsaeewcemuezoeees 36.4687 + .0014

General Meat... sia scadanas Sussaaniea aueean HC1=36.4640+ .0007

Hence, Cl=35.4562+4.0007. Gray and Burt, from their two methods alone, find Cl
=35.460. The difference between their figure and the foregoing ePapeaayn is only one part
in 9332, which is less than the actual uncertainty.

Determinations of the density of HCl have also been made by ee By simultaneous
weighings of the gas in six globes, 28 values were obtained for the weight of the normal liter,
as follows:

1.63035 1.63983 1.63943 1.63941
1.63959 1.63932 1.63943 1.63944
1.63939 1.63887 1.63933 1.63895
1.63940 1.63977 1.63933 1.63928
1.63968 1.63951 1.63931 1.63962
1.63945 1.63892 1.63942 1.63932
1.63987 1.63938 1.63968 1.63928

Mean of all, as one series, 1.63941 + .000031.

' Gray and Burt do not make this calculation. It is useful, however, for purposes of comparison.
2 Journ. Chim, Phys., 8,289, 1910, Zeitsch phys. Chem. 58, 575.

 
No. 8.] ATOMIC WEIGHTS—CLARKE. 51

Scheuer reduces his measurements by three methods, with the following results:

By critical constants.........-2225220.e.e ee bee bee e cece cece ee HC1=36.455

By limiting densities...........-..-.--- 2-02 -- eee eee ee eee eee HCl=36.474

By molecular volumes............-.----- ei pseeuiot Gace HCl=36.473 -
Mean.. sors heedance Se edniadsantete ss snoaeete - -HCl=36.4673

Hence, if H=1.0078, Cl=35.4595 + .00065. “Sclheahe: himself rejects the value derived from the
critical constants, and concludes that Cl=35.466. The lower value as given above is more
consistent with other values for chlorine, and will be retained here.

In 1911 Burt and Gray published a new series of determinations of the density of hydro-
chloric acid, with new precautions against error. Their figures for the weight of the normal
liter at 0°, 760 mm., and at London are as follows:

1.63977 1.63892
1.63999 ~ 1.64009
1.64007 1.64016 .
1.63999 1.64037
1.64049 1.64030

Mean, 1.64011-+ .00005.

Reduced to sea level and latitude 45° this becomes 1.63915 + .00005, a value identical with
that found in, the earlier investigations. The value for chlorine derived from it, and calculated
in the same way, is also the same, namely, 35.4562+.0007. Burt and Gray, by their method
of reduction, conclude that Cl= 35.460.

Two years later Scheuer ? issued another series of determinations, of which the full details
seem not to have been published. His figures for the weight of a liter of hydrochloric acid
are as follows:

1.63980 1.63986 1.63965 1.63952
1.63949 1.63963 : _ 1.63984 1.63978
1.63967 1.63980 1.63973 1.63968
1.63970 1.63942 1.63970 1.63975

Mean, 1.63969 + .000021.

Corrected for latitude and elevation this becomes 1.63917 + .000021, the weight of the nor-
mal liter. Reduced in the same manner as were his former determinations, this gives HCl
= 36.4687, and Cl=35.4609+.00047. The difference between this value and that obtained by
Burt and Gray is due to the different methods of reduction. If reduced in the same way nearly
identical values for the atomic weight of chlorine should be obtained. The physical constants
employed by Scheuer are not the same as those used by Burt and Gray.

With the six values for the atomic weight of chlorine, as given in the preceding pages,
the determination derived by Guye and Fluss from analyses of nitrosylchloride may also be
combined. The details of this determination are given in the next section of this work. The
general combination is as follows:

By syntheses of HCl..........2.....2eeeceeeees pte C1=35,4652+ .0005
Guye and Ter Gazarian................-..2.2200-04-- Cl==35.4615+.0015
Gray and Burt.......2..2.20. 222202 cece eee eee cee eee Cl=35.4562 +. .0007
Scheuer, earlier series.........:.-..22-0-2002eeeee eee C1=35.4595 + .00065
Bort and Greyec coun vv vs vagucwancesvcecwersers2 .. C1=35,4562-+ .0007
Scheuer, later series...........-.---- 2. 2-2-0202 eeeeee C1=35.4609 + .00047
Guyerand Piss... cco ecie ss one ea semie scenes se .-« Cl=35.4680+.0010
General meal.......-.--------- eee ence eee cence C1=35.4598 +- 00025

This final mean is practically identical with the value adopted in recent years by the Inter-
national Committee on Atomic Weights, namely, Cl=35.460.

 

1Chem. News, 103, 161 and 170. 1911. 2 Anzeiger Wien Akad., 50, 378. 1913
52 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

In the foregoing discussion, and also in the preceding discussion of the data relative to
nitrogen, it is apparent that determinations of atomic weight by physical methods are capable
of a high degree of accuracy. They are not, however, superior to the purely chemical determi-
nations. In both general methods there is the fundamental difficulty of insuring the absolute
purity of the substances that are employed. In the physical methods there are also the uncer-
tainties as to the exact values of the physical constants. Critical constants, measurements of
compressibility, and measurements of volunie are all subject to experimental errors; and their
determinations by different workers do not always agree. Such determinations impose their
errors upon the values given by them for atomic or molecular weights; and the “probable
errors’”’ assigned to the latter are therefore too low. When both lines of investigation converge
to a common value for an atomic weight the results obtained are probably close to the truth,
but neither line can claim superiority over the other.

ANALYSES AND DENSITY OF NITROSYL CHLORIDE.

The analyses of nitrosyl chloride, NOCI, by Guye and Fluss,’ are of special interest, because
they give direct ratios between the three component elements. The carefully purified chloride
was first weighed, and then distilled over heated silver, which absorbed the chlorine. The
weight of the latter was given by the gain in weight of the silver. It was next passed over
heated copper, which retained oxygen, and finally over metallic calcium to fix the nitrogen.
The sum of the three components was generally a little less than that of the nitrosyl chloride,
but whether the loss represents undetermined impurity, or failure to collect all the products
of decomposition, seems to be uncertain. The weights obtained were as follows:

NOCI. Cl. 0. , N. Loss.
. 5341 . 2893 - 1805 . 1142 - 0001
- 4284 . 2319 . 1046 . 0916 . 0003
. 7995 . 4331 . 1954 . 1710 - 0000
. 5639 . 3048 . 1375 . 1204 - 0012
- 5121 . 2773 . 1251 . 1095 - 0002

From these figures, with O=16, the atomic weights of N and Cl are directly calculable, by
comparison with O=16. In a third column I also give the value of the ratio Cl: N : : 100: 2,
computed from columns 2 and 4.

 

 

N. Cl. CUN.
14. 001 35. 470 39. 475
14, 011 35. 472 39. 500
14. 002 35. 464 39. 483
14. 010 35. 468 39. 501
14. 005 35. 466 39. 488
Mean, 14.006 + .0017 35. 468 + . 0010 39. 489 + .0033

Several other ratios are calculable from the data given, and indeed, were computed by Guye
and Fluss; but they are not needed here. They involve to a greater extent the uncertainties due
to the losses from the initial substance. The direct value for chlorine has already been utilized
in the preceding section of this work; .that for nitrogen is negligible. Its influence upon the
general combination does not even affect the fourth decimal place.

The synthesis of nitrosyl chloride has been effected by Wourtzel.2_ By the direct union of
chlorine and nitric oxide he obtained the following weights; the ratio Cl : NO : : 100: z has
been added by myself. a

NO. Ratio.
2. 1670 1. 83375 84, 622
1.7751 1. 5022 84. 626
2. 5632 2. 16915 84. 627
2. 25865 1. 91135 84, 624
2. 8994 2. 45385 84. 633

 

"Mean, 84. 626 + .0013.

 

1 Journ. Chim. Phys., 6,732, 1908. 2 Journ. Chim. Phys., 11,214. 1913.
No. 3.) . i was ATOMIC WEIGHTS—CLARKE. aM 53

The density of nitrosyl chloride has also been determined by Wourtzel.' For the weight
of one liter of the gas (at 0° and 760 mm. ?) he gives the following figures, each one the mean
of three series. Three discordant values are rejected.

2. 9880 2. 9877
2. 9886 2. 9888
2. 9880 2. 9875

 

Mean, 2.9881 + .00014.

Wourtzel also gives a series of determinations at 287 mm. pressure, from which in mean
one liter of the gas weighs 2.9466 grammes. With these data he determined the weight of the
normal liter of nitrosyl chloride to be 2.919 grammes and the molecular weight as—

NOCI = 65.457 + .0031,

The probable errors are computed by myself.
We now have, apart from the direct values found for nitrogen and chlorine, three new
ratios to be used in this final discussion of these atomic weights. They are

Cl: N :: 100 : 39.489 +.0033.
Cl: NO :: 100 : 84.626+.0013.
C1+N=49. 457-4. 0031.

The last of these ratios is of small significance. The sum of the valacs already assigned to
N and Cl is .
35.4598 +- 14.0076 = 49.4674,

which differs from the third ratio by 1 part in 3410. This. divergence i is too large for the accu-
rate determination of atomic weights...

THE RATIO HCI: NH;.

Julius Thomsen,’ for the purpose of fixing indirectly the ratio H : O, has made a series of
determinations of the ratio HCl : NH;, which may properly be used toward establishing the
atomic weight of ‘nitrogen. | First, pure, dry, gaseous hydrochloric acid is passed into a weighed
absorption apparatus containing pure distilled water. After noting the increase in weight,
pure ammonia gas is passed in until a very slight excess is present, and the apparatus is weighed
again. The excess of NH;, which is always minute, is measured by titration with standard
hydrochloric acid. In weighing, the apparatus is tared by one of similar form, and containing
about the same amount of water. Three series of determinations were made, differing only in
the size of the absorption apparatus; so that for present purposes the three may be taken as
one. Thomsen considers them separately, and so gives greatest weight to the experiments
involving the largest masses of material. I give his weighings, and also, as computed by him,

 

. HCl
the ratio NH,’

HCl. NH. ; Ratio.

First SHA ertnanshniiens dendetasliguoeas Raaketes ete atl 5.1624 | 2.4120 © 2.1408
3. 9425 1. 8409 2. 1416

4, 6544 2. 1739 2.1411

3. 9840 1. 8609 2. 1409

5.8295 2. 4898 2. 1406

' “49517 °°. 1.9863 2. 1405
4, 8287 2. 2550 2, 1414

6. 4377 3. 0068 2.1411

4, 1804 1, 9528 2. 1407.

5. 0363 2. 3523 2. 1410

4.6408 2.1685 2, 1411

 

 

1 Journ. Chim. Phys., 11,29. 1913.
2 Zeitsch. physikal. Chem., 18,398. 1894. Fora criticism sas ene ro see Acree and Brunel, Amer. Chem. Journ., 36,117. See also
Meyer and Seubert, Ber. 27, 2770.
54 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

 

Aci. NE. Ratio.

Second series... --......2eeeesecccccceccccencercesrees 11. 8418 _ 5, 5302 2. 14130
14, 3018 6. 6808 2. 14073

12. 1502 5. 6759 2. 14067

11. 5443 5. 3927 2. 14073

12. 3617 5.7733 2. 14118

Third @eries 53:25 /0/.s4i doc bwdocadaaseusececsesan awe 19. 3455 9. 0360 2. 14094
19. 4578 9. 0890 2. 14081

 

Mean of all, 2. 14093+ . 000053
Reduced to a vacuum, 2. 1394

From the sums of the weights Tienieat finds the ratio to be 2.14087, or 2.13934 in vacuo.

Thomsen, it will be noticed, performed. his syntheses i in an aqueous solution. Baumé and
Perrot ‘ in a later investigation affented the direct union of the two gases. _ Omitting determina- ,
tions, which they rejected as defective, their weights are as follows: .

HCl. NG;. Ratio.
1. 50409 : 0. 70245 2, 1412
1, 28613 4 . 60046 2.1419

. 56766 . 26514 2, 1410
1, 47740 . 68995 2.1413
2. 08181 - 97219 2.1414

Mean, 2. 14136+. 000092.

With H=1.0078 and N=14.0076, Thomsen’s determinations give Cl=35.4283. Baumé and
Perrot’s figures, on the other hand, give C1=35.4617, a much better value. It seems justifiable,
therefore, to reject the work of Thomsen as of no present significance. In the final discussion
of the fundamental ratios this one will be used to determine the atomic weights of both chlorine
and nitrogen, and so distributing the errors.

SYNTHESES AND DENSITY OF HYDROBROMIC ACID.

The synthesis of hydrobromic acid has been effected by Weber? by essentially the same
method as that followed by Noyes and Weber in their work on hydrochloric acid., There are,
however, differences of detail, for which the original memoir should be consulted. The weights
and ratios obtained are as follows:

 

 

A, Br.” HBr. Hi: Br. HA: HBr,
0. 77300 61. 28837 62. 06052 » 79. 2863 80. 2853
- 86060 68. 25033 69. 11144 "79. 8055 80. 3061
. 77607 61. 54733 62. 32198 79. 3064 80. 3046
- 96927 76. 88221 77. 85135. 74, 3197 80. 3195
1, 07545 (85. 29562 - 86. 37092 _. 79.3114 80. 3114
. 99689 79. 06834 80. 06424 79. 3150 80. 3140 .
. 74966 59. 45275 60. 20500 79. 3063 80. 3097
- 98161 77. 85054 78. 83758 | 79. 3141 80, 3145
1.00131 79. 39633 80, 39658 79, 2915 80. 2914
. 81983 65. 02140 65, 83867 79. 3108 80. 3077
Mean, 79. 3067 80. 3064
+. 0022 +. 0023

With H=1, these two ratios give Br= 79,3067 and 79.3064, respectively. Combined they give,
with 0=16 and H=1.0078,
Br= 79.9260 + .0016 -
The only determinations of the density of hydrobromic acid suitable for modern atomic weight
reductions are those made recently in Guye’s laboratory by MM. Moles,* Reiman,‘ and Murray.®

 

1Compt. Rend., 155, 461. Journ. Chim. Phys., 11,53. 1913.

2 Journ. Amer. Chem. Soc., 34, 1204,

_§Journ. Chim. Phys., 14,389. 1916. A critical discussion by Guye precedes Moles’ Paper.
4Compt. Rend., 164, 44and180. 1917. Journ. Chim. Phys., 15, 293.

5 Ibid., p. 182. Tourn, Chim. Phys., 15, 334.
No. 3.] ..: > ATOMIC WEIGHTS—CLARKE. 55

Moles prepared his hydrobromic acid by four distinct processes, namely: By the hydrolysis
of PBr,, the reaction between Br and H,S, and the reactions between Br and naphthalene and
Br and paraffin. With a gas thus obtained he made 33 determinations ‘of the weight’ of one
liter at 0° and 760 mm., as follows:

3. 64497 3. 64439 3. 64411
3. 64530 - 8.64490 3. 64370
3. 64514 8. 64473 38. 64420
3. 64365 3. 64424 8.64874 ©
38. 64298 3. 64600 3. 64480
3. 64608 8. 64437 3. 64420
3. 64374 8. 64462 3. 64423
3. 64528 3. 64404 3. 64480
3. 64400 3. 64395 3. 64477 |
3. 64569 3. 64475 3. 64326
3. 64369 38. 64403 3. 64450

Mean, 3.64442 + .00013.

Moles also made 11 determinations of the weight of a liter of the gas at 506.67 mm. pressure,
and 17 at 253.33 mm. The mean weights were 3.63314 and 3.62190 grams, respectively. From
these data, by formule which he gives in detail, he finds— ~
Berets hase foc, HE ie , aot
HBr =80.9333,

where, if H=1.0078, Br=79.9255. Moles assumes that H=1.0076, which gives a slightly
higher value for bromine. ‘

Reiman prepared his hydrobromic acid by the action of syrupy phosphoric acid on
potassium bromide. He gives 12 determinations for the weight of one liter, at standard tem-
perature and pressure, as follows:

_ 8.6436 3, 6435

3.6442 3. 6433

we 3. 6442 38. 6445

8, 6461 3, 6442

_ 8.6457 3. 6439

8. 6435 3. 6440
epee Mean, 3. 644224. 0002.

ih ,
He, like ee also gives series of weights at reduced pressure. |
Murray employed hydrobromic acid produced by the hydrolysis of AIBr,, ond gives 16
determinations of the weight of one liter as follows:

3. 6497 3. 6482
3. 6415 3. 6439
3. 6402 38. 6322
3. 6418 3. 6422
3. 6467 3. 6453
3. 6432 : 3. 6463
3. 6410 ; 3. 6492
3. 6401 3. 6430

eo Mean, 3.6435 + .0005.

Combining these series, we have—
GIRS, us 5sichis5 cc taed GAN ARR a Sea COARAGNGAASe lehaeonann te 3. 64442+. 00013

RelMan.<.ctecawwy ec teescesaaks ane phrase Siiaceoii ahead : 3. 64422+. 0002
MUPPBY sac oeseeabuies oes testes seeae Ns shsiaitiete Re islets eeie seein ceeceedee xe 3. 6485 +. 0005
General mean...........---------+2-200 Dee ee sete teeter eee ee 3. 64431+. 00011
as the weight of a standard litre of hydrobromic acid. Hence, reduced | proportionally to Moles’
oe HBr = 80.9310 + .0024, :
ani nests Br = 79.9232 4.0024.

 

1 From the data relative to the density of HBr, Guye (Journ. Chim, Phys., 17, 171) finds Br=79.920. In a previous paper (ibid, p. 141) he dis-
tusses the application of Avogadros’s law to the density of HBr.
56 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

As a final combination we have, as the atomic weight of bromine,

By syntheses of HBr. .................. iddidasedgee Br=79.9260+ .0016.
By density of HBr......... 2.0.2.2... cece ccc e ee ee cee Br=79.9232 + .0024.
General mean... ......0.2.0.020eceeeceseeeeee Br=79.9251+.0013.

IODINE PENTOXIDE RATIOS.

The complete analysis of iodine pentoxide has been effected by Guichard,' who determined
thereby the direct ratio between iodine and oxygen. The pentoxide was decomposed at a high
temperature, and the iodine set free was collected and condensed in a suitable tube, in which
it was weighed. The oxygen was fixed on heated copper, and also weighed. The weights
obtained are as follows:

1,0. L 0.

5.18051 8.93910 1.24183
7.80826 5.93688 1.87138
9.33281 7.09689 2.23637
6.23324 4.73922 1.49333
7.08076 5.38441 1.69714

From these data three sets of values for the atomic weight of iodine can be computed.

 

 

I-1,0;. 0:1,0;. O:1.
126.923 126.867 126.880
126.897 126.898 126.898
126.960 126.928 126.936
126.885 126.962 126.943
126.964 126.887 126.905
Mean, 126.926-+.0108. Mean, 126.908+.0112. Mean, 126.912

Mean of all as one series, I[=126.915+ .0056.

The ratio between iodine pentoxide and silver has been measured by Baxter and Tilley.?
The oxide was prepared by the careful dehydration of iodic acid, the latter having been made
from purified iodine. After weighing, the pentoxide was dissolved in water, and the acid so
formed was reduced to hydriodic acid by means of hydrazine. By final titration of the solution
with a solution of pure silver, the ratio in question was determined. The ultimate data, with
vacuum, weights and all corrections applied, are as follows:

Series I.—Tilley.

Weight I,0;. Weight Ag. Ratio.
6. 06570 3. 92027)3
9. 48035 6. coal 64. 6234
7. 73052 4.99564 64, 6223
12. 63909 8. 16777 64. 6231
9. 49913 6. 13841 64, 6208
8. 34369 5, 39202 64, 6239
8. 83155 5. 70715 64, 6223
6. 77487 4, 37803 64. 6216

 

: Mean, 64. 6225+. 0003
Series II.—Bazter.

Weight 1,05. Weight Ag. Ratio.
12. 09036 7. 81320 64. 6234
6. 29744 4. 00957 64, 6226
10. 89880 7. 04309 64. 6226
9. 33895 6. 03505 64, 6222
10. 15370 6.56169 : 64. 6236

 

1 Compt. Rend., 159, 185, 1914. With details in Annales Chim. Phys., (9) 6, 279 and (9) 7,5. 1916 and 1917.
2 Journ. Amer. Chem. Soc., 31, 201. 1909.
4 These analyses were inadvertently mixed, and are combined in the table.
No. 3.] ATOMIC WEIGHTS—CLARKE. 57

Weight I,0;. Weight Ag. Ratio.
11. 00453 7.11141 64. 6226
7. 01649 4, 53431 64. 6236
9. 33573 6. 03304 64. 6231
8. 72163 5. 63619 64. 6231
9. 01524 5. 82591 64. 6229

 

ae Mean, 64. 6230+. 0001
Combining both series, the mean value for the ratio is

1,0,:2Ag::100 : 64.2229 + .0001
ANALYSES OF SODIUM HALIDES.

The chloride and bromide of sodium have been analyzed electrolytically by Goldbaum.1
The salts were electrolyzed in a cell having mercury as the cathode, with a weighed anode of
silver, The sodium formed an amalgam with the mercury, and the chlorine or bromine wes
firmly deposited on the silver. A trace of chlorine remained in solution, and was determined
by means of the nephelometer. All weights were reduced to a vacuum.

Sodium chloride was analyzed in two series; the first one being “preliminary.” The weights
and percentage of Cl in NaCl are as follows. The percentages were computed by myself.

NaCl. Cl. Per cent Cl.
0. 81495 0. 49482 60. 656

. 79370 - 48135 60. 646

. 88940 . 538962 60. 672

. 45642 . 27679 60. 644

- 48884 . 26618 60. 655

- 44239 - 26828 60. 643

. 64942 - 39387 60. 650

. 53706 . 82081 60. 665

 

. Mean, 60. 654+. 0025
For the final series we have—

NoCl. Cl. Per cent Cl.
1. 02234 0. 62014 60. 6587
1. 02221 . 62006 60. 6587
: 2. 48474 1. 47692 60. 6601
1, 46370 . 88789 60. 6607
- 56934 . 84534 60. 6562
1. 00793 - 61141 60. 6600
1. 06501 . 64600 60. 6567
2. 16720 1. 31460 7 60. 6589
2. 75219 1. 66939 60. 6568
. 92900 - 56349 60. 6555
1, 83527 1, 11324 60. 6581

 

Mean, 60. 65824-. 00035
Weighted mean of both series, 60. 6581+. 00035

For sodium bromide only one series of analyses was made, as follows:

Na Br. / Br. Per cent Br.
1. 05343 “0.81808 77. 6538
1. 33360 1. 03561 77. 6554
1. 95652 1.51936 77. 6583
5. 02976 3. 90586 77. 6550
2. 09332 1. 62554 77. 6538

. 6: 46697 5. 02178 77. 6528
5. 54733 4. 30768 77. 6582

7. 03901 5. 46606 77. 6538

 

Mean, 77. 6543-4. 00029

1Journ. Amer. Chem. Soc. 33, 35. 1911.
58 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

From these series of analyses the cross ratio between chlorine and bromine is calculable,
giving ;
Br: Cl::100:44. 3674 +. 0006.

If Cl=35.4598, the value previously found in this recalculation,
: Br=79. 9221.
If Br=79. 9251, the value previously found, then
Cl=35. 4607. |

This computation, based upon the new ratios gives strong evidence in favor of the accuracy
of Goldbaum’s work. ;
: ; ANALYSES OF CHLORATES.

Until recent times. the fundamental values for the atomic weights of silver, chlorine and
potassium, were best determined through analyses of chlorates. Modern, direct determinations
of the chlorine-hydrogen and nitrogen-oxygen ratios have in great measure supplanted the
chlorate work, which, however, must still be taken iato account, and may even regain some of
the lost: ground.

The first good series of analyses of potassium chlorate was made by Berzelius.' All the
earlier estimations were vitiated by the fact that when potassium chlorate is ignited under
ordinary circumstances a little solid material is mechanically carried away with the oxygen
gas. Minute portions of the substance may even be actually volatilized. These sources of
loss were avoided by Berzelius, who devised means for collecting and weighing this trace of
potassium chloride. All the successors of Berzelius in this work have benefited by his example,
although for the methods by which loss has been prevented we must refer to the original papers
of the several investigators. In short, then, Berzelius ignited potassium chlorate, and deter-
mined the percentage of chloride which remained. Four experiments gave the following

results:
60. 854 60. 850
60. 850 60. 851
Mean, 60.851, -. 0006

The next series was made by Penny,? who worked after a somewhat different method.
He treated potassium chlorate with strong hydrochloric acid in a weighed flask, evaporated to
dryness over a sand bath, and then found the weight of the chloride thus obtained. His per-
centages are as follows, in six trials: ,

60. 825 60. 820
60. 822 , 60. 823
60. 815 60. 830

Mean, 60. 8225, +. 0014

In 1842 Pelouze* made three estimations by the ignition of the chlorate, with these
results:
60. 843
60. 857 |
60. 830

 

Mean, 60, 843+. 0053

Marignac, in 1842,4 worked with several different recrystallizations of the commercial
chlorate. He ignited the salt, with the usual precautions for collecting the material carried off
mechanically, and also examined the gas which was evolved. He found that the oxygen from
50 grammes of chlorate contained chlorine enough to form 0.003 gramme of silver chloride.
Here are the percentages found by Marignac:

 

1 Poggend. Annalen, 8,1. 1826, 3 Compt. Rend., 15, 959.
2 Phil, Trans., 1839, p. 20. ¢ Ann. Chem. Pharm., 44,18. Oeuvres Complates, 1, 57.
No. 3}. ATOMIC WEIGHTS—CLARKE. 59

In chlorate once crystallized. .......------ 02-022 e eee eee rere ees 60. 845
In chlorate once crystallized. .......----------+- sreeysssvosnnse 6-835
In chlorate twice crystallized... ......+-+-+-2-+-2-e:eeereeteees 60. 833
' In chlorate twice crystallized. .......-.----++-+++-++ Fe ees pat 60. 844
In chlorate three times crystallized.........-..--+--++- ees 60. 839

_ In chlorate four times crystallized. ....-...-.------+-- popes 60. 839
3 | Mean, 60. 8392+. 0013

 

In the same paper Marignac describes a similar series of experiments made upon potassium
perchlorate, KCIO,. In three experiments it was found that the salt was not quite free from
chlorate, and in.three more it contained traces of iron. A single determination upon very pure
material gave 46.187 per cent of oxygen and 53,813 of residue. n

In, 1845 two series of experiments were published by Gerhardt.’ The first, made in the

usual way, gave these results: ee
60. 871
60. 881
60. 875

Mean, 60. 8757+. 0020

 

In the second series the oxygen was passed through a weighed tube containing moist
cotton, and another filled with pumice stone and sulphuric acid. Particles were this collected
which in the earlier series escaped. From these experiments we get—

60. 947 /

60. 947
60. 952,

Mean, 60.'9487+. 0011

 

These last results were afterwards sharply criticised by Marignac,? who seriously questioned
their value. ye .

The next series, in order of time, is due to Maumené.* This chemist supposed that particles
of chlorate, mechanically carried away, might.continue to exist as chlorate, undecomposed ; and
hence that all previous series of experiments might give too high a value.to the residual chloride.
In his determinations, therefore, the ignition tube, after expulsion of the oxygen, was uniformly
heated in all its parts. Here are his percentages of residue:

60. 788 60. 785
60. 790 60. 795
60. 793 60. 795
60. 791

Mean, 60. 791+. 0009

The question which most naturally arises in connection with these results is, whether
portions of chloride may-not have been volatilized, and so lost. :

Closely following Maumené’s paper, there is a short note by Faget,* giving certain mean
results. According to this chemist, when potassium chlorate is ignited‘slowly, we get 60.847
per cent of residue. When the ignition is rapid, we get 60.942. As no detailed experiments are
given, these figures can have no part in our discussion. _

Next in order are two series of analyses by Stas.5 In the first series are the results obtained
by igniting the chlorate. In the second series the chlorate was reduced by strong hydrochloric
acid, after the method followed by Penny: a

 

 

 

| First series. Second series.
60.8380 «4+ 60. 850
60. 8395 ‘ 60. 853
60. 8440 ' 60. 844
60. 8473 ——
60. 8450 Mean, 60. 849-4. 0017

‘Mean, 60. 84276 4-0. 0012 ;
1 Compt. Rend., 21, 1280. 3 Ann. Chim. Phys. (3), 18, 71. 1846. 5 Genvres Sompiites, 1, 395-405. |

2 Suppl. Biblio. Univ. Geneve, Vol. 1. 4 Ann. Chim. Phys. (3), 18, 80. 1846.
60 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

In these experiments very elaborate precautions were taken to avoid error and insure
accuracy. All weighings were reduced to a vacuum standard, from 70 to 142 grammes of
chlorate were used in each experiment, and the chlorine carried away with the oxygen in the
first series was absorbed by finely divided silver and estimated.

These determinations by Stas were generally regarded as conclusive until 1911, when
Staehler and Meyer’ published their analyses of potassium chlorate. Their material was
purified with scrupulous care and analyzed by a method similar to that employed by Stas in
his second series of analyses but somewhat modified. The chlorate, after heating to fusion in
a quartz glass vessel in order to secure complete dehydration, was dissolved in water and then
reduced to chloride by a stream of pure gaseous hydrochloric acid. The potassium chloride
was afterwards fused in an atmosphere of nitrogen in order to remove all traces of water and
acid. Two sets of analyses with vacuum weights are reported by Staehler and Meyer; one pre-
liminay, the other final. The data for the preliminary series, with percentages calculated by
myself, are as follows: :

KCIO03. KCl. Per cent KCl.
12. 38248 _7. 58218 60. 829
11. 28213 6. 86340 60. 834
14. 22480 ‘8. 65366 60. 835 -
11. 52268 : 7. 00963 60. 833

12. 44913 7. 57331 60. 834

vaette

; Mean, 60. 833+. 0007
For the final series the figures are—

KCl0;. KCl. Per cent KCl.

10. 26355 6. 24370 60. 8337

10. 08261 6. 18362 60. 8337

10. 03177 6. 10269 60. 8337

10. 63651 6. 47073 60. 8351-

12. 05095 7. 33096 60. 8351

soy es . Mean, 60. 8339+. 00022
The several series of analyses of potassium chlorate now combine as follows:

'  *  Berzelitis. ..... O yaat FAs Sake nnbenstanmekedmanemee rex: 60. 851 +. 0006 |
Penny...-i.....-- ice253 eeeeoeSs Canes ae@ademaa2 2s 60. 8225+. 0014
Pluses wie ice ones eos adnpacicniaaioer pegeunceiswiaie ey .. 60.843 +. 0053
Marignac....... 2.2.0.0. 0.0ececee cence ec eeecececeeeeeee 60. 8392. 0013
Gerhardt, first..........22.....--  laseucnene avo diecast abaavate seers 60. 8757+. 0020
Gerhardt, second...... ch cistetes a rane ease pngenenesmieras eehantaci aaa 60. 9487+. 0011
Maumené......-....- ie ideas mehaoesueeee eee tevenene yee 60. 791 +. 0009
BS tna Maris te nae assoc sdaselslpaaeesmsenaaeeahiocesens 60. 8428+. 0012
Stas, S6CONG: sone ve esc ds sateen seewccueceee ye yess 60. 849 +. 0017
Staebler and Meyer, first......--.----.+-++-eeeeee eee ee ee 60. 833. +, 0007
Staehler and Moyer, second... weetee eee ee eee regulelgeiastaee ses 60. 8339+. 00022
General man. oe eee tee eset es etetereeee ec eceeececees 60, 8365+. 00018

Hence if O=16, KCl= 74.5631 +.0004.

Guye and Ter Gazarian? have called attention to the fact that potassium chlorate tends
to retain some chloride as an impurity. Staehler and Meyer, however, claim to have avoided
that source of error in their analyses. The correction for such an impurity probably applies
to the earlier analyses of the chlorate.

The percentage of oxygen in sodium chlorate has been determined only by. Penny,* who
used the same method which he applied to the potassium salt. Four experiments gave the

following results: - 45, 060 45, 080

45, 075 45. 067
Mean, 45. 0705+. 0029
Hence NaCl = 54.500 a 0048.

1 Zeitsch. anorg. Chem., 71,378. 1911. For acriticism by Gave, see Journ. Chim. EYE, ., 10, 145, and ay, by Meyer and Staehler, in Zeitsch.
anorg. Chem., 77, 255.

2Compt. Rend., 143, 411.. According to Guareschi the bromate is also a constant impurity in the slate and also ia theiodate. See Chem.
Zentralblatt, 84, 978. 1913. \

4 Phil. Trans. 1839, p. 25.

 
No. 3] -. ATOMIC WEIGHTS—CLARKE. . 61

For the composition of silver chlorate there are analyses by Marignac’ and by Stas.’
Marignac’s series is as follows:

AgCl03. AgCl. Per cent, AgCl.
24. 540 18, 363 74. 920
25. 809 19. 336 74, 913
30. 306 22. 709 74. 932
28. 358 21. 247 74, 924
28. 287 : 21, 185 74. 893
57. 170 42. 840 74. 934

 

; Mean, 74. 9193-4. 0041
Corrected to a vacuum, this becomes 74.917.

The determinations by Stas are only two in number, giving the subjoined percentages of
AgCl in ABO s:
74, 919
74, 922

Mean, 74. 9205+. 0010

 

Combining this with Marignac’s figure, the general mean becomes 74.9203 + .0010. Hence

AgCl = 143.390 +.0060.
ANALYSES OF BROMATES AND IODATES.

Accurate analyses of bromates and iodates, available for atomic weight determinations,
are few in number, and from a modern point of view, not satisfactory. Potassium bromate
was analyzed by Marignac,' by careful calcination, with ‘all the precautions taken to avoid loss.
His figures are subjoined:

KBr0). KBr. — - Per cent 0.
6. 801 4, 849 28. 7016
3. 480 2. 483 28. 6494
6. 320 4, 506 28. 7025

23. 186 16. 521 28. 7458

 

Mean, 28. 6998+ .0133
Hence KBr = 119.249 + .0596.

Marignac attempted to analyze silver bromate, but found difficulties in drying the salt.
He also made some experiments upon the precipitation of silver bromate by potassium chloride,
but published no details of his determinations. He merely states that from 31.32 to 31.47
parts of KCl were needed to precipitate the silver from 100 parts of bromate.

Stas ‘ effected the analysis of silver bromate by reduction with sulphurous acid, its con-
tent in water having been previously determined. After applying all corrections the subjoined
percentages of oxygen were found from the weight of the bromate and that of the residual

sneer bromide:
Aa 20. 351
20. 347

Mean, 20. 349+. 0014

 

Hence AgBr = 187.884 + .0133.

The percentage of oxygen in potassium iodate has been datartiined by Millon.’ In three
experiments he found:
. 22. 46
22. 49
22. 47

Mean, 22. 473.005

Hence KI = 165.590 + .0384.

According to Marignac ® potassium iodate loses iodine when calcined, and is therefore
unsuited to atomic weight determinations.

eee a .
See ’ . 1 Oeuvres Complates, 1, 80. 5 Oeuvres oan: 1, 84. 5 Ann. Chim. Phys. (3), 9, 400. 1843.
2 Oeuvres Completes, 1, 685. 4 Oeuvres Completes, 1, 635. 6 Oeuvres Complates, 1, 85.
62 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [voL. XVI.

Millon also estimated the oxygen in silver iodate, getting the following percentages: _

17. 05
17. 03
17. 06

Mean, 17. 047+.005

The analysis of silver iodate has also been performed with extreme care by Stas.!. From
76 to 157 grammes were used in each experiment,. the weights being reduced to a vacuum. As
the salt could not be prepared in an absolutely anhydrous condition, the water expelled ir each
analysis was accurately estimated and the necessary corrections applied. In two. of the experi-
ments the iodate was decomposed by heat, and the oxygen given off was fixed upon a weighed
quantity of copper heated to redness. Thus the actual weights, both of the oxygen and the
residual iodide, were obtained. In a third experiment the iodate was reduced to iodide by a
solution of sulphurous acid, and the oxygen was estimated only by difference. In the three
percentages of oxygen given below, the result of this analysis comes last. The figures for
oxygen are as follows: oy et Bg
5 aie ens 16. 976° ee
16. 972
~ 16. 9761

Mean, ‘16. 9747 + .0009 -

 

This, combined with Millon’ 8 series above cited, gives us a general mean of 16. 977 1 +, 0009,
Hence Agl =234.7344.0126. 0

THE SILVER-CHLORINE RATIO. ee

For the ratio between silver and chlorine there are many series of determinations, some
direct and some indirect... As with numerous other ratios, the first work entitled to any con-
sideration was done by. Berzelius.?

He made three estimations, using each tithe 20 grammes of pure silver. This was
dissolved in nitric acid. In the first experiment: the silver chloride was precipitated and col-
lected. on a filter... In the second and third: experiments the solution was mixed with hydro-
chloric acid.in a:flask, evaporated to dryness, and the residue;then fused and weighed. without
transfer. One hundred oe of silver formed of chloride: os

' 132. 700 oe
132. 780 : 2h Ey
132. 790 : » 3 Ps

sue ae i Mean, 132. 757 4.019 ae e*

 

Turner’s work ? closely resembles that of Berzelius. Silver was dissolved in nitric acid and
precipitated as chloride. In experiments one, two and three the mixture was evaporated and
the residue fused. In experiment four the chloride was collected on a filter. A fifth, experi:
ment was made, but has been rejected as worthless.

- The results were as follows: In a third column'I put the aie of AgCl proportional t to
100 parts of Ag.

Ag. AgCl. Ratio.
28. 407 37. 737 182. 844
41. 917 55. 678 182. 829
40. 006 53. 143 132. 837

30. 922 41. 070 132,818, ef
ae oe bn BPG ee a Mean, 182. 8324,0038 soul

 

The same general method of dissolving silver ia nitric ‘acid, precipitating, ‘evaporating,
and fusing without transfer of material was also adopted by Penny.‘ His results for 100 parts
of silver are as follows, in parts of chloride:

 

1 Oeuvres Complétes, 1, 628. 2 Thomson’s Annals of Philosophy, 15, 89. 1820. 3 Phil. Trans., 1829, 291. ‘ Phil, Trans., 1839, 28.
No. 8.] ATOMIC WEIGHTS—CLARKE. 63

182. 836 132. 840
132. 840 132. 830
132. 830 132. 838
143. 840

 

Mean, 132. 8363 + .0012

In 1842 Marignac * found that 100 parts of silver formed 132.74 of chloride, but gave no
available details. Later, in another series of determinations, he was more explicit. Silver
was dissolved in nitric acid, and precipitated by hydrochloric acid. The precipitate was washed
several times with boiling water, by decantation, and the chloride was finally dried and fused
in the same flask in which it had been formed. The figures are as follows:

Ag. AgCl. Ratio.
79. 853 106. 080 132. 844
69. 905 92. 864 132. 843
64. 905 86. 210 132. 825
92. 362 122. 693 132. 839
99. 653 132. 383 132. 844

Mean, 132. 8394..0024

Corrected for weighing in air the mean becomes 132.854.

The above series all represent the synthesis of silver chloride. Maumené ? made analyses
of the compound, reducing it to metal in a current of hydrogen. His experiments make 100
parts of silver equivalent to chloride:

132. 734 132. 729
132. 754 132, 741
132, 724

Mean, 132.7364: .0077

By Dumas ‘ we have the following estimations:

Ag. AgCl. Ratio.
9. 954 13. 227 182. 882
19. 976 26. 542 132. 869

Mean, 132. 8755+. 0044

Next in order are seven determinations by Stas.* In the first, second, and third, silver
was heated in chlorine gas, and the synthesis of silver chloride thus effected directly. In the
fourth and fifth silver was dissolved in nitric acid, and the the chloride thrown down by passing
hydrochloric acid gas over the surface of the solution. The whole was then evaporated in the
same vessel, and the chloride fused, first in an atmosphere of hydrochloric acid, and then in
a stream of air. The sixth synthesis was similar to these, only the nitric solution was pre-
cipitated by hydrochloric acid in slight excess, and the chloride thrown down was washed by
repeated decantation. All the decanted liquids were afterwards evaporated to dryness, and
the trace of chloride thus recovered was estimated in addition to the main mass. The latter
was fused in an atmosphere of HCl. The seventh experiment was like the sixth, only ammo-
nium chloride was used instead of hydrochloric. acid. From 98.3 to 399.7 grammes of silver
were used in each experiment, the operations were performed chiefly in the dark, and all weigh-
ings were reduced to a vacuum. In every case the chloride obtained was beautifully white.
Treating Stas’ determinations as a single series, his figures are as follows:

 

Ag. AgCi. Ratio.
91. 462 121. 4993 132. 841
69. 86735 92. 8145 182. 848
101. 519 134, 861 132. 843
108. 549 144, 207 132. 849
399. 651 530. 920 182. 846
99. 9925 132. 8382 132, 848
98. 3140 130. 602 132. 8417
Mean, 182. 8445+. 0008
1 Ann. Chem. Pharm., 44, 21. 3 Ann. Chim. Phys. (3), 18, 49. 1846. 5 Oeuvres Complateés, 1, 333-341.
2 Oeuvres Completes, 1, 79. 4 Ann .Chem. Pharm., 113,21. 1860.

161566°—20-—_5
64 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

According to Richards and Wells,; who made two very careful series of syntheses, the
work of Stas on the silver-chlorine ratio was subject to constant errors. His silver probably
contained occluded oxygen, and perhaps alkalies also, and his glass vessels were attacked and
changed in weight by the acids used in his operations, These errors were avoided by Richards
and Wells, who precipitated and fused their silver chloride either in porcelain or quartz vessels,
generally the latter, and who employed silver of the highest possible purity. A number of
minute corrections were also applied to their determinations, but these can not be considered
in detail now. The results obtained appear in the two following tables: .

Preliminary Series.

Ag. AgCl. Ratio.
9. 06843 12. 04365 132. 861
8. 39217 11. 14985 132. 860
5. 37429 7. 14056 132. 865
8. 08222 10. 73869 132. 868
7. 08517 9. 41362 132. 864
7. 97715 10. 59837 132. 859
8. 11978 10. 78767 132. 857
8. 53452 11. 33907 182. 861
6. 73284 8. 94511 132. 858
8. 91366 11. 84240 132. 857
9. 72295 12. 91769 132. 858
8. 63961 11. 47862 132. 860

11. 18795 14. 79849 132. 865

 

: ; Mean, 132. 8610+. 00065
Final Series.

Ag. AgCl. Ratio.
7. 24427 9. 62508 182. 865
8. 30502 11. 03484 132. 870
7. 29058 9. 68676 132. 867
8. 58472 11. 40614 132. 866
8. 01318 11. 64648 132. 862
9. 77160 12. 98335 132. 868
7. 98170 © 10. 60528 132. 870

11. 49983 15. 27964 132. 868
6. 25318 8. 30834 182. 866
7. 72479 12. 26360 132. 866

Mean, 132. 8668+. 0005

Five more determinations of this ratio, by Richards and Anderegg ? remain to be noted.
The usual Harvard methods of operation were employed. The figures are given in two distinct
groups, which are consolidated here.

Ag. AgCl. Ratio.
6, 22211 8. 26738 132. 871
5. 16568 6. 86361 132. 869
5. 18476 6. 88876 182. 866
5. 92279 7. 86933 132. 864
5. 00706 6. 65256 182. 864

 

Mean, 132. 867+. 0009

The foregoing ten series of determinations are to be classed as direct; that is, they were
made for the express purpose of measuring the ratio between silver and chlorine, and were
not complicated by other considerations. Arranged in the order of ascending magnitude,
and expressed in the form Ag : Cl: : 100 : 2, these combine as follows:

 

1 Publ. Carnegie Inst., Washington, No. 28, 1905. 2Journ. Amer. Chem. Soc., 37, 14-21, 1915.
No. 3.] ATOMIC WEIGHTS—CLARKE. 65

MANIMEN 6 i.65cccedaeecewets ieee saqneemeeanes sy ese e'seee -32. 736 +.0077
Berselivies :soaqsee chee wee bons $ on ceaeeeGnceanes sees ee 32.757 +. 0190
TUIMED: a raaiars sigovisieciainsesascas nen ads antes eeenleleees gobs 32. 832 +. 0038
Pony cee: xqsicecseeeueedsicses casei seemeceessentetes 32, 836 +. 0012
Stade. 3s .cese co teseee see cigeipaneewne oeleegies esses 32. 8445 +. 0008
Map HNC anavormancieaind oeco4 cas seme omen wanna gia 32, 854 +. 0024
Richards and Wells, preliminary..........-.-.---------- 32. 861 +. 00065
Richards and Wells, final. ..........-.--------- ice 32. 8668+. 0005
Richards and foideregy Pais shngaldets ape acce sine pee eeeess 32. 867 +. 0009
DMS. oes eee lanka sss ve ee nape awe dae etle eee sii 32. 8755+. 0044
General mean............---- prec e cece eee eeeeeeeee 32. 8670+. 00038

This general mean is almost identical with Richards and Wells’ final determination.

A second group of determinations of the silver-chlorine ratio may be termed incidental.
A chloride is balanced against silver, and the silver chloride produced is also weighed, and this
procedure, intended to fix other atomic weights, also gives values for the ratio now under
consideration. The following determinations, thus obtained, are all useful. I limit myself,
however, to work done by individual authorities, and do not attempt to combine observations,
say of RCl: Ag by one chemist, and RCl:AgCl by another, into determinations of the ratio
Ag:AgCl. The details of the several investigations will be found in subsequent chapters of
this work, in relation to what I may term the several collateral elements.

The first series of this incidental kind to be now considered is due to Lenher,' and is derived
from his data on the atomic weight of selenium. Silver selenite was converted into silver
chloride, and the latter was afterwards reduced to metal by heating i in hydrogen. The vacuum
weights and the derived ratio appear in the next table.

AgCl. Ag. Ratio.
. 21897 . 16480 132. 870
. 48522 . 86534 132. 813
. 58999 44417 132. 830
. 67532 50821 182, 882
. 82232 . 61882 132. 885
1. 08350 . 81562 132. 844
1. 36288 1. 02588 132. 850
1. 67234 1. 25884 132, 848

 

Mean, 132. 853+. 0060
Similar data are furnished by Ebaugh’s? analyses of silver arsenate, which were designed
to determine the atomic weight of arsenic. The weights are all reduced to a vacuum.

AgCl. Ag. Ratio.
. 21547 - 162175 132. 863
- 44615 . 33583 132. 850
- 48820 - 367525 132. 844
. 74517 . 56099 132. 831
- 88083 . 66318 132. 819
- 94830 . 71400 132. 815
. 98014 . 73771 132. 863

Mean, 132. 841+. 0050

In their memoir upon the atomic weight of cesium, Richards and Archibald? give analyses
of cesium and potassium chloride, balancing each salt against silver and silver chloride. In
the following table the first two determinations are derived from the potassium salt, and the
others from the cesium compound. The weights refer to the vacuum standard, as do all the
others in this group of determinations.

 

1 Journ. Amer. Chem. Soc., 20, 555. 1808. 3 Doctoral thesis, U niversity of Pennsylvania, 1901. * Proc. Amer. Acad., 38, 443. 1903.
66 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

Ag. AgCl. Ratio.
3. 61747 4, 80600 1382. 855
3. 62283 4, 81325 132. 859
2. 45600 3, 26240 132. 834
2. 53351 3. 36532 132. 832
1. 45686 1. 93555 132. 858
1, 94244 2. 58003 132. 824
2. 05023 ; 2. 72382 132. 854
1. 50720 2. 00253 132. 864
1. 32251. 1. 75678 132. 837
1. 29434 1, 71972 132. 864
1. 13743 1. 51093 132. 837
1. 97590 2. 62484 132, 835
2. 00760 2. 66720 132. 855
3. 24850 4.31570 132. 852

 

Mean, 132. 847-4. 0024

Figures of the same order are given by Archibald in his research upon the atomic weight of
rubidium,’ and again in his memoir upon potassium?

Rubidium Series.

Ag. AgCl. : Ratio.
1. 78454 2. 37070 132. 842
1.84241 © 2. 44778 132. 858
2.04710 2. 71960 132. 851

. 97702 1. 29796 132. 849
1. 91316 2. 54118 182. 826
2. 58550 3. 43475 132. 847
1. 96076 2. 60452 132. 832
1. 91462 2. 54386 132. 865
1, 89346 2.51557 132. 856
2.01515 2. 67685 132. 836
1. 94594 2. 58528 182, 855
2. 07668 2. 75878 132. 846
3. 56998 4. 74233 182. 842
2. 17233 2. 88613 182. 862

 

Mean, 132. 848+. 0020
Potassium Series.

Ag. AgCl. Ratio.
8. 20598 4, 25916 132. 850
2. 88479 3. 83250 132. 852
4. 19557 5. 57396 182. 853
6. 85280 9. 10362 182. 845

 

Mean, 132. 850+. 0012

The analyses of cobalt chloride, by Baxter and Coffin,? furnish the subjoined figures:

Ag. AgCl. Ratio.
1. 82671 2, 42676 132. 846
2, 45398 3. 26095 132, 884
6. 38081 8. 47735 132, 857
4, 92244 6. 54019 132. 865
5. 78815 7, 69084 132, 872
5. 47410 7. 27284 132. 859
2.61905 8. 48012 132, 877

 

Mean, 132. 8657 +. 0034

 

1 Journ. Chem. Soc., 85, 786, 1904. ? Trans. Roy. Soc., Canada, 1904, Sec. III, p. 47. 3 Zeitsch. anorg. Chem., 51, 171. 1906.
No. 3:1 ATOMIC WEIGHTS—CLARKE. 67

From the analyses, by Baxter and Hines,’ of manganese chloride we have—

Ag. AgCl. Ratio.
7. 93740 10. 54641 132. 870
6. 05041 8. 03868 132. 862
5. 67279 7, 53731 132. 868
6. 11818 8. 12932 182. 871
5. 91637 7. 86129 132. 873
7. 67995 10. 20372 132. 862
6. 72227 8. 93140 132. 863

 

Mean, 132. 8670+. 0012

Baxter and Wilson,? analyzing lead chloride, obtained the following data:

Ag. AgCi. Ratio. .
3. 62987 4, 82273 132. 862
3. 21408 4.27016 132. 858
3. 97568 5. 28272 132. 876
2. 99456 3. 97949 132. 891
2. 40837 3.19909 132. 832
3. 33407 4, 42982 132. 865

 

‘Mean, 132. 864:. 0054

By Hoénigschmid® there are two determinations, based upon his study of radium chloride.
They are—

Ag. AgCi. Ratio.
0. 71630 0. 95166 132. 858
. 70600 . 93804 132. 867

Mean, 132. 8625+. 0030

Baxter and Moore,! in their work on the atomic weight of phosphorus, give the following
figures:

Ag. AgCl. Ratio.
10. 82441 14. 38118 132. 858
10. 36771 13. 77434 132. 858
13. 46408 17. 88933 182. 860
12, 21814 16. 23430 132. 870
12. 14555 16. 13727 132. 884
12. 08725 16. 05962 132. 874
11. 43585 15, 19245 132. 849

8. 45324 11. 23091 132. 859
13. 94377 18. 52592 132. 862

 

Mean, 132. 864 ++ 0.0080
From Scheuer’s® analyses of silver sulphate we have—

Ag. AgCl. Ratio.

8. 63592 11. 47436 132. 868
5. 99316 7. 46296 132. 868
10. 21124 13. 56757 132. 869
8. 96085 11. 90590 132. 866
9. 70232 12. 89159 132. 871

Mean, 132. 8684 + 0.0005

Richards and Lembert® in their determinations on the atomic weight of lead from uraninite

 

give the following figures:
Ag. AgCl. Ratio.
3. 00984 3. 99877 182. 857
4. 33300 5. 75707 182. 866
1 Journ. Amer. Chem, Soc., 28, 1560. 1906, ‘Journ. Amer. Chem. Soc., 34, 1644.
aJourn. Amer. Chem. Soc., 30, 187, 1908. 6 Arch. Sci. Phys. Nat. (4), 36, 318.

3 Monatsh. Chem., 33, 253. 6 Journ, Amer, Chem. Soc., 36, 1329.
68 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.
Ag. '  AgCl. Ratio.
4. 27157 5. 67573 132, 872
3. 01600 4. 00703 132. 859
3. 14788 4, 18265 132. 872
2. 18162 2. 89816 132, 844
2. 29951 3. 06475 132. 844
8.15153 4.18670 132. 846
1. 56952 2. 08663 182. 847

and give two series of figures:

Ag. AgCl. Ratio.
2, 52563 3. 35537 132. 855
3. 07636 4, 08686 132. 847
2. 94693 3. 91519 132, 856
2. 87400 3. 81830 132. 856
3. 50343 4, 65464 132. 859

Mean, 132. 855 + 0.0014
Second series.

Ag. AgCl. Ratio.
3. 76724 5. 00512 132. 857
3. 09903 4.11741 132. 853
3.82647 5, 08425 132. 870

’

First series.

 

Mean, 132. 868 + 0.0038

 

 

Mean, 132. 863 + 0.0026

of ordinary lead. The figures for silver and silver chloride are as follows:

Ag. AgCl. Ratio.
4.79374 6. 36895 132. 860
3. 68659 4. 89837 132. 870
3.75047 4, 98328 132. 871

 

H6nigschmid and Horovitz! determined the atomic weight of isotopic lead (‘‘uranblei’’)

In another paper the same chemists? give a series of determinations based upon analyses

Mean, 132.867 + 0.0030

A long series of determinations by Baxter, Whitcomb, and Stewart? is derived from their
analyses of neodymium chloride. Two aberrant determinations are rejected by the authors
and are therefore omitted here. The figures for the present ratio are as follows:

Ag. AgCl. Ratio.

5. 21445 6. 92786 132. 859
5. 78760 7. 69005 132. 871
4. 84458 6. 43738 132. 878
7. 75892 10. 30876 132. 863
6. 41045 8. 51634 132. 851
5. 36392 7, 12743 182. 877
7. 14338 9. 49134 132. 868
6. 12054 8. 13232 182. 868 :
5. 45444 7, 24714 182. 867 ©
7. 66088 10. 17891 132. 868
7. 23524 9. 61299 132. 863
9. 33359 12. 40098 132. 864
5. 81558 - 7. 72718 182. 870

Mean, 132. 867 + 0.0013

 

1 Monatsh. Chem., 36, 355,

2 Monatsh. Chem., 36, 375.

3 Journ. Amer. Chem. Soc., 38, 302.
No. 3.] ATOMIC WEIGHTS—CLARKE. 69

These incidental series of values for the ratio Ag : Cl :: 100 : 2 now combine as follows:

Bbatigh, As series. <2... 2. e.cccis sc eee secs adele as sae 82.841 +.0050
Richards and Archibald, Cs series..............--------- 32.847, +.0024
Archibald, Rb series. . sehyulouewewouseasecdeuta des 2.848; 50020
Archibald, K series aise: elena am a'e's s'stassisiglaeeaie sieia’ ein’ oie os 32.850, +. 0012
Lenher; Se series... o2.c2c.ccevetes vesaceewecesacdhescne 32. 853, +. 0060
Hénigschmid and Horovitz, Pb series 1.........-.-.-.-- 32.855, +.0014
Honigschmid, Ra series..............-0-522 eee eee ee eee 82. 8625, +. 0030
Honigschmid and Horovitz, Pb series 2....2.....-.----- 82.863, -.0026
Baxter and Wilson, Ph series...........--.-.----------+ 32.864, +. 0054
Baxter and Moore, P series..........-.----- ee eee 32.864, +.0030
Baxter and Coffin, Co series...........-..+--.---00-0---: 82. 8657, +. 0034
Baxter and Hines, Mn series.........---.------------e-- $2. 867, +. 0012
Baxter, Whitcomb and Stewart, Nd series.............-- 32. 867, +.0013
Hénigschmid and Horovitz, Pb series 3.........-.....-- 32.867, +. 0030
Richards and Lembert, Pb series................----+--- 32. 868, +. 0038
Scheuer, AgoSO, series.....-...----. 2-2 eee ee eee eee eee 82. 8684, +. 0005

General Meats 2c cece dee b25s ees ees eeer 32. 8640+. 00036

A third group of determinations is to be classed as indirect. When two ratios, RCI : Ag
and RCI : AgCl have been measured by the same investigator, but independently of each other,
the cross ratio, Ag : Cl :: 100 : z is easily calculable from them. The examples to be given
presently are almost self-explanatory; but the details of the determinations must be sought
for in the later sections of this work, on titanium, magnesium, barium, cadmium, etc.

In his determinations of the atomic weight of titanium, Thorpe gives data from which the

subjoined ratios are derived:
4Ag:TiCl,: :100:43.999 + .0032

4AgC1:TiCL,: :100:33.118 + .0019

Hence Ag : Cl: : 100 : 32.855 + .0092.
From Richards’s analyses of barium chloride we have—

2Ag : BaCl, :: 100 :96.525 + .0010
2AgCl : BaCl, : : 100 : 72.653 +.0014

Hence Ag : Cl: : 100 : 32.8575 + .0029.
Ratios computed from the analyses of magnesium chloride by Richards and Parker:

2Ag : MgCl, :: 100 : 44.138 +0003
2AgCl : MgCl, : : 100 : 33.226 + .0013

Hence Ag : Cl: : 100 : 32.842 + .0054.
Data for cadmium chloride are given by Baxter and Hines, and also, later, by Baxter,
Hines and Frevert. Their series, combined together, give—

2Ag: CdCl, :: 100 : 84.9677 + .0008
2AgCl: CdCl, : :::100 : 63.9523 + .0004

Hence Ag: Cl: : 100 : 32.861 +.0016.
For sodium chloride the analyses of Richards and Wells give the following ratios:

Ag : NaCl :: 100 : 54.1854 4.00015
AgCl : NaCl :: 100 : 40.7797 + .00028

Hence Ag : Cl: : 100 : 32.873 + .0010.
The potassium chloride ratios of Richards and Staehler are—

Ag : KCL :: 100 : 69.1073 + .00032
AgCl : KCl: : 100 : 52.0118 + .00025

Hence Ag :Cl:: 100 : 32.869 + .0009.
70 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor, XVI.

In Archibald’s work on the atomic weight of platinum the subjoined ratios appear:
Ag : Pt ::100 : 180.965 + .0034
AgCl : Pt :: 100 : 136.203 + .0031

Hence Ag : Cl: : 100 : 32.864 +.0039.

The determinations by Richards and Willard of the atomic weight of lithium gave two
pairs of values for the silver-chlorine ratio.

Preliminary series.
Ag : LiCl : : 100 : 39.3023 + .00048
AgCl : LiCl : : 100 : 29.5802 + .00033
Hence Ag : Cl: : 100 : 32.867 + .0022.

Final series.

Ag : LiCl :: 100 : 39.2992 + .00014
AgCl : LiCl : : 100 : 29.5786 + .00014

Hence Ag : Cl: : 100 : 32.8637 + .00077
From the strontium chloride ratios of Thorpe and Francis we have—

2Ag :SrCl, :: 100 : 73.490 + .0008
2AgCl : SrCl, :: 100 : 55.311 + .0009

Hence Ag : Cl: : 100 : 32.867 + .0026.
Briscoe and Little give two ratios for vanadyl chloride, as follows:

3Ag : VOCI, : : 100 : 53.554 + .0008
3AgCl : VOCI, :: 100 : 40.3085 + .0010

Hence Ag : Cl: : 100 : .0062.
Hénigschmid and Horovitz in their earlier study of isotopic lead give data yielding the
following ratios:
2Ag : PbCl, :: 100 : 128.687 + .0021
2AgCl : PbCl, : : 100 : 96.854 + .0010

Hence Ag : Cl: : 100 : 32.864 +.0026. _

The atomic weight detcrminations by Baxter and Grover for ordinary lead give these ratios:
2Ag : PbCl, :: 100 : 128.905 + .0010
2AgCl : PbCl, :: 100 : 97.021 +.0010

Hence Ag : Cl: : 100 : 32.863 + .0017.

For praseodymium Baxter and Stewart find—
3Ag : PrCl, :: 100 : 76.4090 + .0006
3AgCl : PrCl, :: 100 : 57.5100 + .0007

Hence Ag : Cl: : 100 : 32.862 4.0019.

Similar data by Baxtcr and Chapin are given for neodymium:

. 3Ag : NdCl, : : 100 : 77.4437 + .0009
3AgCl : NdCl, :: 100 : 58.2876 + .0006
Hence Ag : Cl :: 100 : 32.865 + .0020.
No. 3.] ATOMIC WEIGHTS—CLARKE.
These indirect determinations now combine as follows:
Richards and Parker, Mg ratios............---.----------- 82.842 +. 0054
Thorpe, Vitatiog: suc vesase cis ss danaaseteceessed sous 32. 855 +. 0092
Richards, Ba ratios................-.20-0-0-20- eee eee e+ 32, 8575+. 0029
Briscoe and Little, V ratios..........-..----------2eeeeee 32. 860 +. 0062
Baxter, Hines and Frevert, Cd ratios............--------- 32.861 +. 0016
Baxter and Stewart, Pr ratios..........-.--.-.---2------+-- 32. 862 +. 0019
Baxter and Grover, Pb ratios..........-------.----------- 32.863 +. 0017
Richards and Willard, 2d Li series...........--2.-.---2++-+ 32. 8637 +. 00077
Archibald, Pt ratios..........2..222.0000--222 eee cerns 32. 864 +. 0039
H6nigschmid and Horovitz, Pb ratios................---- 32. 864 +. 0026
Baxter and Chapin, Nd ratios..............-.--+---------- 32.865 +. 0020
Richards and Willard, 1st Li series.............---.------ 82. 867 +. 0022
Thorpe and Francis, Sr ratios.......-...-.--------+--+----- 32: 867 +. 0026
Richards and Staehler, K ratios.......- Ban sy eat chal 82. 869 +. 0009
Richards and Wells, Na ratios.............---.-++---ee--++ 32. 873 +. 0010
General mean ...............2 20-2 eee ee ee eee ee 32. 8660+. 00041

71

By combining the three groups of determinations, a final value for the ratio Ag:Cl::100:z

is obtained, thus—

Direct determinations. ...................000--e0 eee ee eee 32. 867+. 00038

Incidental determinations..................---+-+-.----+-- 32. 864+. 00036

Indirect determinations..........-.....-.---------++---- 32. 866+. 00041 .
General mean of 41 series....-..---.-.--------.-- 32. 8656+. 00022

THE SILVER-BROMINE RATIO. °

The measurements of the silver-bromine ratio resemble those of the ratio between silver

and chlorine, and fall into three groups. First in order are the direct determinations.

Marignac,' to effect the synthesis of silver bromide, dissolved the metal in nitric acid, pre-
cipitated the solution with potassium bromide, washed, dried, fused, and weighed the product.

The ratio Ag: Br :: 100 : 2 is given in the third dolanin:

Ag. AgBr. Ratio.
25. 000 43. 518 74, 072
20. 120 35. 020 74. 055
15. 000 26. 110 74. 066

 

2 Sayin hte, Tia _ Mean, 74 064 +.003
Corrected for weighing in air the mean becomes 74.077.

Much more elaborate determinations of this ratio are due to Stas? In one experiment a
known weight of silver was converted into nitrate, and precipitated in the same vessel by pure

hydrobromic acid. The resulting bromide was washed thoroughly, dried, and weighed.

In four

other estimations the silver was converted into sulphate. Then a known quantity of pure
bromine, as nearly as possible the exact amount necessary to precipitate the silver, was trans-
formed into hydrobromic acid. This was added to the dilute solution of the sulphate, and after
precipitation was complete, the minute trace of an excess of silver in the clear supernatant fluid

was determined. All weighings were reduced to a vacuum. The data are as follows:

Ag. AgBr. Ratio.
53.1958 92.6042 74.0830
51.3436 89.3780 74.0790
55.0615 | 95.8505 74.0795
55.8040 97.1450 74.0805
43.3620 , 84.1904 74.0830

 

Mean, 74.0810 + .0006

1 Oeuvres Complates, 1, 83. ? Oeuvres Complates, 1, 587, 603.
72 MEMOIRS NATIONAL ACADEMY OF SCIENCES. © [Vou. XVI.

In his paper on the atomic weight of cadmium, Huntington gives three syntheses and three
analyses of silver bromide. The data are as follows, with the usual ratio given in the last
column:

Ag. AgBr. Ratio.
1. 4852 2. 5855 74, 084
1, 4080 2. 4510 ‘74. 077
1, 4449 2. 5150 74, 060
AgBr. Ag. Ratio.
4. 1450 2. 3817 74, 035
1, 8172 1. 0487 74.111
4. 9601 2, 8497 74. 057

 

Mean, 74. 071+.0072

Similar synthetic data are also given by Richards, incidentally to his work on copper.? There
are two sets of three experiments each, which can here be treated as one series, thus:

Ag. AgBr. Ratio.
1.11235 1.93630 74.073
[usr 2.74335 74.044
2.16670 * 3.77170 74.076
-9664 1.68205 74.053
-9645 1.6789 74.069
9639 1.6779 74.074

 

Mean, 74.065+.0035

In their research upon the eletrochemical equivalents of copper and silver, Richards, Col-
lins, and Heimrod $ give the following syntheses of silver bromide from electrolytic silver:

Ag. AgBr. Ratio.
0.71585 1.24567 74.013
5.43807 9.46557 74,061
3.76993 6.56216 74.066
2.29649 3.99820 74.100
2.15701 3.75473 74.071
2.37893 4.14187 74.106
2.97120 5.17218 74.077

 

Mean, 74.0706 + .0078

Tt is only fair to state in this connection that the foregoing series was intended to determine
the purity of the silver, and not as an accurate measure of the ratio.

Scott,‘ in his analyses of ammonium bromide, titrated the compound with silver. He after-
wards collected and weighed the silver bromide, in order to determine the silver bromide ratio.
The subjoined weights refer to the vacuum standard:

Ag. AgBr. Ratio.
6.82315 11.87733 74.074
9.66809 16.82816 74.090*
5.41906 9.43315 74.0735
5.51258 9.59596 74.074
5.70686 9.93346 . 74.062
5.33191 9.28093 74.064
5.62572 9.79254 74.067

 

Mean, 74.072 +.0023

 

1 Proc, Amer. Acad., 17, 28. 1881. 4 Journ. Chem. Soc., 79, 147. 1901.
2 Proc. Amer. Acad., 25, 199, 210, 211. 1890. A * The starred figure is corrected for a trace of impurity.
8 Proc. Amer. Acad., 35, 139. 1899.
No. 3.] ATOMIC WEIGHTS—CLARKE. 73

In his paper on the atomic weight of iron Baxter ! gives three direct comparisons of silver
with silver bromide, with vacuum weights, as follows:

Ag. AgBr. Ratio.
4.77783 8.31754 74.086
5.87977 10.23533 74.077
4.82995 8.40809 74.082

 

Mean, 74.082+.0018

A much more thorough and conclusive set of syntheses was published by Baxter ” in 1906.
The purest silver was dissolved in nitric acid, and precipitated by ammonium bromide. The
silver bromide, before weighing, was fused in an atmosphere containing bromine vapor. With
vacuum weights, Baxter’s figures are as follows:

Ag. AgBr. : Ratio.
4.71853 8.21363 74,072
5.01725 8.73393 74.078
5.96818 10.38932 74.079
5.62992 9.80039 74.077
8.138612 14.16334 74.080
5.07238 8.82997 74.079
4.80711 8.36827 74.081
4.27279 7.43776 74.072
5.86115 10.20299 74.078
7.91425 13.77736 74.083
6.40765 11.15468 74.084
6.38180 11.10930 74.078
6.23696 10.85722 74.079
9.18778 15.99392 74.078
8.01261 13.94826 74,079

10.48638 18.25452 74.078
8.59260 14.95797 74.079
8.97307 15.62022 74.079

 

Mean, 74.0785 -+ .00047

The direct determinations of the ratio Ag: Br combine thus:'

 

Richards os csccccweisse see Seige aceesecageeticuaesoemens 74.065 +.0035
Richards, Collins, and Heimrod 74.0706 .0078
Huntington 74.071 +.0072
SCoth. -ceceeeeses ssh ecaenccnenmacceseesees ae ek enone 74.072 +.0023
Marignae: .cs.2ssccge eoede eedaty 74.077 +.0030
Baxter, 1906. ‘ 74.0785 + .00047
GS teh a cates eye ayant te fats areca ci iaucayacereyereie iste lebanese saan 74.081 +.0006
Baxter, 1904. ..-..eeeeee tere cere cere rere er ee teen ees sce prea 74,082 + .0018
General. Means .c ink ele cewdslsceen sae seas 74.0797 + .00035

Analyses of various metallic bromides have furnished many incidental determinations of
the silver-bromine ratio, like those already described for silver and chlorine. In his work on
titanium bromide, intended to determine the atomic weight of titanium, Thorpe® gives the
following equivalent weights of silver and silver bromide:

Ag. AgBr. Ratio.
3.66122 6.375391 74.133
5.55097 9.663901 74.094
8.17645 14.227716 74.008

7.83493 13.639956 74.092

 

Mean, 74.082+.0176

 

1 Proc. Amer, Acad., 39, 250. 1904. 2Journ, Amer, Chem, Soc.,:28, 1322, 1906. 8 Journ. Chem. Soc., 47, 126. 1885.
74

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

[Vou. XVI.

Thorpe and Laurie’ compared gold with silver and silver bromide, and gave equivalent
weights as follows:

Ag.
3.38451
2.60896
2.28830
2.26415
1.97147
2.01292
2.50334
2.93608

AgBr.

5.89199
4.54261
3.98288
3.94309
3.43015
3.50207
4.35736

5.11045

Ratio.
74.087
74.113
74.054
74.153
73.989
73.980
74.062
74.057

Mean, 74.062+-.0143

 

In Richards’ ? memoir upon the atomic weight of barium, the subjoined vacuum weights
of Ag and AgBr are given as equivalent to each other. Two additional determinations are

rejected by Richards as inaccurate:

Ag.
1.71323
2.13584
1.52921
2.11740
1.72276
1.34175
4.11360
2.56010
2.51415

AgBr.
2.98230
3.71809
2.66191
3.68615
2.99868
2.33530
7.16120
4.45670
4.37669

Ratio.
74.075
74.081
74.071
74.089
74.063
74,049
74.086
74.083
74.082

Mean, 74.075+.0029

From the analyses of nickel bromide, by Richards and Cushman,’ the following figures are
derived. These, and all the subsequent series, represent vacuum weights:

Ag.
3.23910
2.66636
3.33990
1.31787
1.23482
1.30629
2.21652

AgBr.
5.63892
4.64208
5.81391
2.29435
2.14963
2.27384
3.85805

Ratio.
74.089
74.098
74.074
74.088
74.085
74.069
74.059

Richards and Cushman also give one direct determination of the ratio, in which 2.10289
grammes of silver yielded 3.66066 of bromide. Ratio, 74.078. Including this in the foregoing

series, the mean becomes 74.080 + .0030.

Similar data appear in the memoirs of Richards and Baxter‘ on the atomic weight of

cobalt.

Their analyses of cobalt bromide gave the following equivalent figures:

Ag.
1.31702
2.54585
2.80449
1.81170
2.64879
2.84891
2.29593
1.89033

AgBr.
2.29296
4.43095
4.88135
3.15368
4.61046
4.95943
3.99706
3.29053

Ratio.
74.102
74.046
74.055
74.073
74.059
74.086
74.093
74.072

 

1 Journ. Chem, Soc., 61, 565. 1887.
2 Proc, Amer. Acad., 28, 1.

1893.

® Proc. Amer. Acad., 33,97. 1897.

4 Proc. Amer. Acad., 33, 115, 1897; and 34, 351, 1899,
No. 3.]

ATOMIC WEIGHTS—CLARKE.

75

There are also in these two memoirs by Richards and Baxter three direct determinations

of the ratio, as follows:

Ag.
2.18679
2.91386
2.97097

AgBr.

3.80679
5.07226
5.17170

Ratio.
74.081
74.073
74.074

Taking these with the previous eight determinations as one series, the mean value for the

ratio is 74.074 + .0033.

From the analyses of uranium bromide, by Richards and Merigold,' the following figures

are obtained:
Ag.
1.39365
82559
1.43617

AgBr.
2.42588
1.43713
2.50009

Ratio.
74.066
74,073
74.080

Mean, 74.073+.0027

 

The following figures are derived from the analyses, by Richards and Archibald,’ of

cesium bromide:
Ag.
1.77402
3.14606
3.63740

AgBr.

3.08815
5.47673
6.33213

Ratio.
74.076
74.082
74.084

Mean, 74.081+.0017

 

Archibald’s * analyses of rubidium bromide give a similar series of comparisons, as follows:

Ag.
1.74930
1.35230
1.37061
1.70300
2.50590
2.46502
2.83340

AgBr.

3.04578
2.35401
2.38589
2.96462
4.36215
4.29084
4.93210

Ratio.
74,114
74.075
74.076
74.081
74.075
74.069
74.070

Mean, 74.080+.0040

 

Baxter, Hines, and Frevert,‘ in order to determine the atomic weight of cadmium, ana-

lyzed cadmium bromide. Their silver figures are subjoined:

Ag.
9.08379
5.40724
5.35277
5.61597
4.07226
4.63072
4.68200
4.75259

AgBr.
15.81319
9.41267

9.31830

9.77649

7.08933

8.06130

8.15070

8.27360

Ratio.
74.081
74.075
74.084
74.088
74.088
74,083
74.086
74.086

Mean, 74.084 + .0010

 

 

1 Proc. Amer. Acad., 37, 383. 1902,
3 Proc. Amer. Acad., 38, 443. 1903.

8 Journ. Chem. Soc., 85, 776. 1904.
4Journ. Amer. Chem. Soc., 28,770. 1906.
76 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI.

Baxter and Hines‘ also analyzed manganese bromide, and gave the following equivalent
weights of Ag and AgBr:

Ag. AgBr. Ratio.
6. 56765 11. 43300 74. 080
4, 83238 8. 41206 74. 077
4. 90354 8. 53642 74, 087
5. 65813 9. 85008 74, 087
5. 82600 10. 14206 74, 083
3. 61478 6. 29271 74, 083
5. 18711 9. 02959. 74. 077
3. 94042 6. 85968 74. 085
4, 51250 7. 85571 74. 088
3. 61736 6. 29740 74, 088
4. 79620 8. 34915 74. 078
3. 59319 6, 25569 74. 098
5. 72641 9. 96840 74. 078

 

Mean, 74. 0844 .0011

Baxter and Thorvaldson? in their analyses of ferrous bromide derived from meteoric iron
give the following weights for silver and silver bromide:

Ag. _ AgBr. Ratio.

3. 95631 6. 88720 74, 081
4. 67177 ; 8. 13282 74. 084
4.75550 8. 27855 74, 083
6. 95854 12. 11329 74. 078

. 8, 20904 5. 58632 , 74. 081

 

Mean, 74. 081+ .0008

e

From the figures relative to lead bromide the same chemists* give the following data:

Ag.  AgBr. Ratio.
3. 85610 6. 71241 74. 073
3. 95874 6. 89124 74. 077
2. 94968 5. 13400 74. 062
4, 53498 7. 89526 74. 067
4, 12154 7. 17470 74. 078
4, 06484 7. 07672 74. 096

Mean, 74. 075+ .0032

Lead bromide was also analyzed by Baxter and Grover, whose data give the following
ratios:

Ag. AgBr. Ratio.
2. 38398 4, 15017 74, 086
2. 02288 8, 62224 74. 119
8. 04158 5. 29498 74, 086
3. 26022 _ 8, 93446 74, O74
2. 53086 4, 40616 74. 097
2. 66549 4. 64048 74, 095
8. 40044 5. 91976 74, 088
2. 86337 4. 98467 74, 084
3. 69447 6. 43124 74. 077
2. 79011 4. 85708 74, 082
4. 01148 6. 98380 74, 095°

Mean, 74. 089+ .0025

 

1 Journ. Amer. Chem. Soc., 28, 1560. 1906. 8 Journ. Amer. Chem. Soc., 37, 1020. 1915.
2 Journ. Amer. Chem. Soc., 33, 337. 1911. 4 Journ. Amer. Chem. Soc., 37, 1027. 1915.
No. 3.1 ATOMIC WEIGHTS—CLARKE. 77

Baxter, Moore, and Boylston‘ determined the atomic weight of phosphorus by analyses of
the tribromide. Their data relative to the present ratio are as follows:

Ag. . AgBr. Ratio.
7. 11504 12. 38560 74. 076
5. 63022 9. 80058 74, 071
5. 64583 9. 82830 74. 081
7. 74048 13, 47428 74, 075
5. 52140 9. 61143 74. 076
4. 58163 7. 97563 74. 079
5. 64835 9. 83261 74. 087
3. 84649 6. 69553 74. 073
6. 90379 12. 01761 74. 073
6. 59456 11. 47946 74. 075
9. 49930 16. 53569 74. 073
5. 15074 8. 96595 74. 071
5, 25447 9. 14663 74. 070
8. 45957 14. 72642 74. 080
5. 01851 8. 73586 74. 073
8. 68405 15. 11725 74, 081
9. 26406 16. 12737 74. 085

 

Mean, 74. 076+ .0008

Hénigschmid and Horovitz? determined the atomic weight of uranium by two series of
analyses of the bromide. Their work furnishes the following data:

Series 1.

Ag. AgBr. Ratio.
3. 60418 6. 27409 74. 078
38. 09028 5. 37994 74. 092
3. 55214 6. 18353 74. 079
3. 42969 5. 97038 74, 079

Mean, 74. 082+.0011
Series 2.

Ag. AgBr. Ratio.
4. 08466 7. 11092 74. 088
4. 14755 7. 22023 74. 086
4. 56818 7. 95237 74. 084
2. 42929 4, 22884 74. 077
3. 40682 5. 93036 74. 073

 

Mean, 74. 082+ .0019

The work of Hénigschmid * on thorium bromide yields the following data for the silver
bromine ratio:

Ag. AgBr. Ratio.
4. 00046 6. 96402 % 74, 081
3. 01126 5. 24210 74, 083
3. 82643 6. 66091 74, 072
38. 67975 6. 40573 74. 080
4. 01057 6. 98157 74. 079
3. 27425 5. 69982 74, 080
2. 86343 4, 98479 74, 085
2. 92948 5. 09931 74, 069
3. 49762 6. 08888 74, 086
3. 92861 6. 83873 74. 075
4, 18375 7. 28310 74. 081
3. 38105 5. 88552 74, O74

 

Mean, 74. 079-4. 0010

 

1 Proc. Amer. Acad., 47, 585. 1912.
2 Monatsch. Chem., 37, 185. 1916. ‘
3 Zeitsch. Electrochem., 22,18, 1916. From abstract by Baxter in Journ. Amer. Chem. Soc., 39, 340.
78 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The incidental determinations of the ratio Ag : Br : : 100 : x now combine as follows:

Thorpe and Laurie, Au series.............-.-.-------20+ 74, 062 +.0148
Richards and Merigold, U series.........4-..-.-.-.-..++- 74.073 +. 0027
Richards and Baxter, Coseries...........-----2-2-0e0005 74.074 +. 0033
Richards; Ba series.......-.-.....2.2-2-2-222-022-0+2+-+-- 74,075 4.0029
Baxter and Thorvaldson, Pb series.............-....----- 74.075 +. 0032
Baxter, Moore and Boylston, P series.............-...--- 74.076 +. 0008
Ho6nigschmid, Th series............2.22-2.-0-2 ee eee eee ee 74.079 +.0010
Richards and Cushman, Niseries.............22-..2-2--- 74. 080 +. 0030
Archibald, Rb series........-.-.--..222222-ee cece ee ee eee 74,080 +. 0040
Richards and Archibald, Cs series............-..- aecien 74,081 +. 0017
Baxter and Thorvaldson, Fe series.......-..--........--- 74, 081 +. 0008
Thorpe;, Ti SCries oases :ciersic aicieieSisieia wo sie sisisisnsisiorarnioraiewiereieeis 74. 082 +. 0176
Honigschniid and Horovitz, U series1.............. ald 74.082 +.0011
Honigschmid and Horovitz, U series 2.............--.---- 74, 082 +. 0019
Baxter, Hines, and Frevert, Cd series......-........--.-- 74, 084 +.0010
Baxter and Hines, Mn series.........--..----------++----- 74.084 +.0011
Baxter and Grover, Pb series...........---------+e+--0+- 74.089 +. 0025
General tiiean 5. 2252 sake escacasneaccec ees smeaeas 74. 0803 +. 00035

Several indirect determinations of the silver bromine ratio, as in the case of the chlorides,
are deducible from analyses of metallic bromides. *
In Cooke’s determinations of the atomic weight of antimony, the ratios are as follows:

3Ag : SbBr, :: 100 : 111.114+.0014
3AgBr : SbBr, :: 100 : 63.830 +.008

Hence Ag : Br : : 100 : 74.078 4.0219.

From Huntington’s analyses of cadmium bromide we have—
2Ag : CdBr, :: 100 : 126.076 + .0052
2AgBr : CdBr, :: 100 : 72.4216 +.0028

Hence Ag : Br : : 100 : 74.086 + .0098.

The work of Richards on strontium bromide gives—
2Ag : SrBr, :: 100 : 114.689 + .0012
2AgBr : SrBr, :: 100 : 65.884 + .0006

Hence Ag : Br: : 100 : 74.077 + .0024.
The ratios deduced from analyses of zinc bromide by Richards and Rogers are—
2Ag : ZnBr, :: 100 : 104.380+.0007
2Ag : Br, :: 100 : 59.962 + .0004
Hence Ag : Br : : 100 : 74.077 4.0016.
Analyses by Baxter of ferrous bromide yield the following ratios:
2Ag : FeBr, :: 100 : 99.960 + .0027
2AgBr : FeBr, :: 100 : 57.4195 + .00044
Hence Ag : Br : : 100-: 74.087 +.0049.
Richards and Mueller studied potassium bromide with the subjoined results:
Ag : KBr :: 100 : 110.319 + .0004
AgBr : KBr :: 100 : 63.3727 + .0003
Hence Ag : Br: : 100 : 74.981+.0012.
The determinations by Archibald of the atomic weight of platinum give the following ratios:
Ag : Pt :: 100 : 180.965 + .0034
AgBr : Pt :: 100 : 103.955 +.0037
Hence Ag : Br :: 100 : 74.080 +.0070.

1 For details, see later sections of this work, on Sb, Cd, Sr, Zn, Cd, Fe, etc.
No. 3.] | ATOMIC WEIGHTS—CLARKE. 79

From the work of Thorpe and Francis on strontium bromide we have—

2Ag : SrBr, :: 100 : 114.703 +.0040
2AgBr : SrBr, :: 100 : 65.892 +.0011

Hence Ag : Br : : 100 : 74.077 4.0067.

The determinations of the atomic weight of calcium by Richards and Hénigschmid yield

the following ratios:
Ag : CaBr, :: 100 : 92.6503 + .00043

2AgBr : CaBr, :: 100 : 53.2232 + .00018

Hence Ag : Br: : 100 : 74.079+.0010.
Baxter, Thorvaldson, and Cobb made two series of analyses of ferrous bromide. From
them the following ratios are derived:

Preliminary series.

2Ag : FeBr, :: 100 : 99.9573 + .00027
2AgBr : FeBr, :: 100 : 57.4221 +.00114

Hence Ag : Br : : 100 : 74.078 4.0035.
Final series.

2Ag : FeBr, :: 100 : 99.9575 + .00057
2AgBr : FeBr, :: 100 : 57.4213 + .00029

Hence Ag : Br: : 100 : 74.078 +.0013.
Hénigschmid’s data relative to radium bromide give these ratios:

2Ag : RaBr, :: 100 : 178.812 + .0009
2AgBr : RaBr, :: 100 : 102.714 4.0019

Hence Ag : Br : : 100 : 74.087 +.0035.
From the work of Hénigschmid on uranium tetrabromide we have these ratios:

Series 1.

4 Ag : Ubr, :: 100 : 129.253+.0015
4 AgBr : UBr, :: 100 : 74.248 + .00025

Hence Ag : Br :: 100 : 74.083 + .0074.
Series 2.

4 Ag : UBr, :: 100 : 129.272 + .0006
4 AgBr : UBr, :: 100 : 74.2617 +.00014

Hence Ag : Br :: 100 : 74.076 +.0028.

In their determinations of the atomic weight of carbon, Richards and Hoover exactly
neutralized sodium carbonate with aqueous hydrobromic acid. The bromine in the solution
was then determined by titration with a standard solution of silver. The amount of silver
used, and that of the silver bromide formed were both ascertained. First, however, the hydro-
bromic acid was standardized. 100 grammes of the HBr solution were equivalent to

4.77442 +.000001 grammes of silver, and
8.31125 +.000018 grammes of silver bromide.

Hence Ag : Br :: 100 : 74.080+.0004. This value might fairly have been classes among
the direct determinations.
The other ratios derived from the work of Richards and Hoover are as follows:
2 Ag : Na,CO, :: 100 : 49.1265 + .00032
2 AgBr : Na,CO, :: 100 : 28.2206 + .00032
Hence Ag : Br :: 100 : 74.081 +.0017.
161566°—20-——6
80 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The general combination of the indirect determinations is as follows:

Honigschmid, U series 2...............2.-22 2c eee eee eee 74.076 +. 0028
Richards, Sr sevies..............--2-.---2e+-+------++-- 74.077 +. 0016
Richards and Rogers, Zn series..........-----------+-+- 74.077 +. 0024
Thorpe and Francis, Sr series...........---.-.----- t.--- 74.077 +. 0067
Cooke; Sb iseRies 5 i553 506 aemns Sat ceanminincines sotdeenies 74.078 +.0219
Richards and Hoénigschmid, Ca series..........-.-...---- 74.079 +. 0010
Archibald, Pt series...-....... vais eee ara oases ..... 74,080 +.0070
Richards and Hoover, HBr series..............-------- 74.080 +. 0004
Richards and Hoover, Na COs, series...........-----+--- 74,081 +. 0017
Richards and Miiller, K series...........--...---.------ 74.081 +. 0012
“Honigschmid, U series 1........-2..2.2-20- eee eee e eee 74, 083 +. 0074
Huntington, Cd series. .........22.. 0.0.0 eee cece eee 74.086 +. 0098
Baxter; Fe 0168 20 siiace's geese e cn gig sad eostneneaees 74,087 +. 0049
Hénigschmid, Ra series... ........ 2.0.20 eee e ee eee eee 74.087 +.0035
General mean. .... PF vigaly Seitoeteah dnd te haces 74, 0802+. 00033
The final combination of the three groups of values is as follows:
Direct determinations. ........--..-------2-2eceeeeeeee 74, 0795+. 00035
Incidental determinations.........-......--.-.----+-0-- 74. 0803+. 00035
Indirect determinations. ...--. ongeves4ses see eaeees 74, 0802+. 00033
General mean of 39 series.......-....-.--------+-- 74, 0800+. 00022

Ag : Br :: 100 : 74.0800 +.00022.
THE SILVER-IODINE RATIO.

The composition of silver iodide, first thoroughly investigated by Marignac and Stas, has
recently been the subject of elaborate researches.

Marignac ' dissolved weighed quantities of silver in nitric acid, and precipitated the silver
iodide with a solution of potassium iodide. He gives the following weights, and the ratio of

AgI to 100 parts of Ag:

Ag. AgI. Ratio.
15. 000 32. 625 217. 500
14. 790 ‘ 32.170 217. 512
18. 545 40. 339 217. 520

 

Mean, 217. 511-4. 0036
Corrected for weighing in air this becomes 217.5335.
Three series of determinations are given by Stas,’ all with weights corrected to a vacuum.
In the first series of experiments Stas converted a known weight of silver into nitrate,
and then precipitated with pure hydriodic acid. The iodide thus thrown down was washed,
dried, and weighed without transfer. His figures are as follows:

Ag. Agl. Ratio.
97. 5915 212. 2905 217. 529
43.5281 94, 6984 217. 536

Mean, 217. 5325+. 0024

In the second series a complete synthesis of silver iodide from known weights of iodine
and metal was effected. The iodine was dissolved in a solution of ammonium sulphite, and thus
converted into ammonium iodide. The silver was transformed into sulphate and the two
solutions wére mixed. When the precipitate of silver iodide was completely deposited the
supernatant liquid was titrated for the trifling excess of iodine which it always contained.
As the two elements were weighed out in the ratio of 127 to 108, while the atomic weight of
iodine is probably a little under 127, this excess is easily explained. From these experiments

 

1 Oeuvres Completes, 1, 86, 3 Oeuvres Completes, 1, 548-683.
No. 3.] ATOMIC WEIGHTS—CLARKE. 81

two sets of values were deduced; one from the weights of silver and iodine actually employed,
the other from the quantity of iodide of silver collected. From the first set we have of iodine
for 100 parts of silver:

117. 5390 117. 5430
117. 5380 117. 5420
117. 5318 117. 5300

 

Mean, 117.5373-£. 0015

From the weight of silver iodide actually. collected the following figures are given for the
ratio Ag:I. The third experiment in the foregoing column has no equivalent here:

117. 529 117. 538
117, 531 117. 530
117. 539 —

 

Mean, 117.5334+. 0014

These determinations, by Marignac and Stas, are remarkably concordant, and yet, as shown
by later investigations, they are affected by constant errors. Silver iodide, precipitated from
nitrate solutions, occludes silver nitrate, a fact which must be taken into account in two of the
preceding series. The concordance between the second and third series of Stas, however,
remains unexplained, if we suppose them to be in error also. That the errors in four sets of
determinations, by two observers and four methods, should be so exactly alike in direction and
magnitude, is difficult to understand.

With Ag=107.93, the Marignac-Stas determinations of this ratio irish I=126.85. This
value was accepted for many years, until Ladenburg, in 1902, showed that it was about one-tenth
of a unit too low. Ladenburg ! depended upon the ratio Agl : AgCl to establish this conclusion,
but he also gave one measurement of the ratio now under consideration, as follows: 50.3147
grammes Ag gave 109.4608 AgI, whence the ratio Ag : AgI = 217.552; a figure higher than those
in the foregoing tables, but not strikingly so.

~ Soon after the publication of Ladenburg’s memoir, Scott ? reported two syntheses of silver
iodide, as follows, with weights corrected to a vacuum:

Ag. Agl. Ratio.
4.6240 10.0634 117.6340
6.39978 13.92913 117. 6502

 

Mean, 117.6421+ .0054

Koethner and Aeuer,? who also studied what might be called the Ladenburg ratio, suc-
ceeded in proving the occlusion of silver nitrate by silver iodide, to which allusion has already
been made. They effected two syntheses of silver iodide, however, avoiding this error, and by
two methods. First, silver iodide was precipitated from solution with pure hydriodic acid.
Secondly, silver was directly combined with iodine, by heating in a stream of iodine vapor. The
two syntheses are subjoined, with the ratio stated in the form Ag: Agl, and the weights corrected
to a vacuum:

Ag. Adl. Ratio.
34.51789 7512752 217.6347
11.37544 24,.75691 217.6480

 

Mean, 217.6413+ .0045

The very thorough and careful experiments by Baxter * fall into several series, and represent
several distinct methods of procedure. First, pure silver was converted into nitrate, and pre-
cipitated by a solution of ammonium iodide in presence of an excess of ammonia. All ‘weighings
in Baxter’s experiments were reduced to a vacuum and various minor corrections were

 

1 Ber. Deutsch. chem, Ges., 35, 2275, 1902,

2 Proc, Chem. Soc., 18, 112, 1902.

3 Liebig’s Annalen, 337, 123. 1904.
4Three memoirs. First, Proc. Amer. Acad., 40, 419. 1904. Second,ibid. 41, 78. 1905. Third, Journ. Amer. Chem. Soc., 32,1591. 1910.
82 MEMOIRS NATIONAL ACADEMY OF SCIENCES. © [Vou. XVI.

applied, concerning which the original memoirs must be consulted. Two series of determina-
tions are given, as follows: :
Preliminary series.

Ag. AglI. Ratio.
5.23123 11.38531 217.6411
3.57039 7.77033 217.6325
4.60798 10.02804 217.6233
4.52467 9.84822 217. 6561
4.66256 10.14591 217.6039

F 1 Mean, 217.6314+ .0059
inal series.'

Ag. Agl. Ratio.
4.77244 10.38698 217.645
4.82882 10.50981 217.648
4.04262 8.79755 217. 620
1.64711 3.58515 217.663
4.85804* 10.57318 217.643
4.83482 10.52241 217.638
4.97120 10.81800 217.613
3.53858 7.70136 217.640
3.89693 8.48187 217.655
5.33031 11.60111 217.644
5.08748 11.07259 217.644

Mean, 217.6412+ .0029
Mean, rejecting the seventh value, 217.6440+ .0024

Secondly, pure iodine was weighed, converted into hydriodic acid by means of sulphurous
acid, and then transformed-into ammonium iodide with pure ammonia. As nearly as possible
the exact equivalent of silver was dissolved in nitric acid and added to the iodine solution. The
trifling excess of silver or iodine was finally determined by titration. The following results
were thus obtained:

Ag. I. Ratio.
5.54444 6.52288 117.6473
6.27838 7.38647 117.6493
4.57992 5.38814 117.6470

Mean, 117.6479+ .0005

In Baxter’s second memoir the ratio just given was redetermined with variations in the
process. Iodine was converted into hydriodic acid and precipitated by a solution of silver,
taking care to avoid an excess of the latter. The final adjustment was effected by titration,
as before. In five of the experiments the silver iodide so produced from a known weight of
iodine was collected and weighed. In the following table both ratios are expressed in the form
Ag:I::100:2,; A representing the direct comparisons and B the silver iodide syntheses,

1. Ag. AgI. Ratio A. Ratio B.
3.29308 2.79897 = nena 117.6588 eee
3.70132 3.145840 eee eee LESTE chiens
3.75641 3.19258 6.94913 117.6607 117.6555
3.24954 2.76186 6.01137 117.6576 117.6589
4.12541 3.50639 7.63204 117.6543 117.6460
3.53166 3.00165 6.53351 117.6573 117.6494
2.99835 2.54842 5.54682 117.6552 117.6529
2.00015 1.6999. eee 117.6621

Mean, 117.6573-+ .0007 117.6525+ .0015

 

1 The starred figure is erroneously given in the original. The corrected figure was kindly furnished me by Professor Baxter. The seventh
experiment in the series Baxter rejects,
No. 3.1 ATOMIC WEIGHTS—CLARKE. 83

In the final series of experiments, based upon the silver used in the preceding set, the ratio
Ag : AgI was redetermined, as follows:

Ag. AgI. Ratio.
3.19249 6.94877 217.6600
2.76175 6.01110 217.6555
3.00189 6.53399 217.6628
2.54833 5.54659 217.6558

Mean, 217.6585 + .0012

In Baxter’s third memoir the ratio of iodine to silver iodide was determined by a slight
modification of his former process. The iodine was converted into hydriodic acid by means of
hydrazine hydroxide, and then, with due precautions, precipitated by a solution of silver nitrate.
The weights (in vacuo) and the usual ratio follow:

Ag. I. Ratio.

7.65468 9.00628 117.6572
11.43179 13.45067 117.6602
10.08571 11.86648 117.6561
* 7.24498 8.52461 117.6611
5.46351 6.42840 117.6606
7.05641 8.30266 117.6613
8.45904 9.95288 117.6602
5.92491 6.97131 117.6621
7.97927 9.38852 117.6614
5.58231 6.56811 117.6594
16.03902 18.87136 117.6590
15.16249 17.84091 117.6648
12.71170 14.95666 117.6606

Mean, 117.6603 + .0004

The determinations of the silver-iodine ratio by Gallo, although numerous, are not as con-
cordant as the foregoing series. Silver was deposited electrolytically, and in the same circuit
iodine was liberated from a solution of potassium iodide, and determined afterwards by thiosul-
phate titration, With vacuum weights the results obtained are as follows:

Ag. I. Ratio.
0.18054 0.21230 117.5917
.21360 -251309 117.6541
.23103 .27181 117.6513
.24005 -28213 117.5291
615454 .18167 117.5553
.2597 .30515 117.5010
.16229 .19080 117.5673
.300988 35411 117.6490
.26819 .31528 117.5584
25877 .30425 117.5755
24422 .28703 117.5298
.20838 -24516 117.6505
.25047 .29445 117.5599
.20266 .23826 117.5664
.18316 .21533 117.5639
.37278 .43809 117.5197
28221 33207 “117.6677
.2582 .30856 117.5677
.33963 .39923 117.5485
.33461 39345 117.5846
.3360 .39502 117.5655
.37025 .43526 117.5584
30824 .36233 117.5480
.36890 .42789 117.5845

Mean, 117.5770+ .0074
1 Atti Accad. Limeei (5), 15,24. 1908, Gaz. Chim. Ital., 36, 116.

 
84 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

Two more series of electrolytic measurements of the ratio between silver and iodine remain
to be mentioned. Both series were made with the iodine coulometer; but the first one, by
Washburn and Bates’ is to be regarded as preliminary. The weights in grammes and the ratio
as adopted here are as follows:

Ag. I, Ratio.
6. 15289 7. 2341 117. 572
6. 50799 7. 6497 117. 547
5.94510 6. 9884 117. 552
5. 64232 6. 6342 117. 579

 

Mean, 117. 5625+. 0065.

The later determinations, by Bates and Vinal,? are more in accord with the values found
by Baxter. The data follow with vacuum weights:

Ag. I, Ratio.
4, 10582 4, 82959 117. 628
4, 10469 4. 82862 117. 637
4. 09903 4, 82224 117. 643
4, 39711 5. 17273 117. 6407
4. 10523 4, 82851 117. 618
4, 12310 4. 84942 117. 616
4, 10475 4, 82860 117. 634
4.18424 4. 92130 117. 615
4. 10027 4, 82247 117. 613
4. 10516 4, 82844 117. 619

 

Mean, 117. 626+. 0025

Neglecting the single determination by Ladenburg, and reducing all the series to the com-
mon form of Ag:1::100:x, the various means combine thus:

Mari onaCices 1c accice as nes ves secre ee anele aio eaten 117. 5835+. 0036
Sass fins cesta uaa aac oes amacsemaaemeanincceaasane 117. 5825-4. 0024
Stas, second........0.. 0... cece cece eee e eee cece eee eees 117. 5373+. 0015
Stas, third........ Bs ante nansauns Hacer ie lat 117. 5334+. 0014
DCOttactecasdnetenaw A aualea oa abans per itaal tats cee Se Rented 117. 6421+. 0054
Koethner and Aeuer....... 2.222.002.0020. cece cence eee 117. 6413+. 0045’
Baxter, 1904, preliminary...............-.-2-22.2-0200- 117. 6314+. 0059
Baxter, 1904, Ag:Agl........00....022ecceeeeeeeeceeees 117. 6412+. 0029
Baxter, 1904, Ag:]....0.0..0.0.0..cceeeecceeeeeeeeceeee 117. 6479-4. 0005
Baxter, 1905; AGT sciesc ceases soe ro vacuuoneiceseeenks 117. 6573+. 0007
Baxter, 1905, I:AgI................ {WR iweteeoeewenaes 117. 6525+. 0015
Baxter, 1905, Ag:Ag]..............00000c0ceeceeec ee eee 117. 6585 -+. 0012
Baxter, 1910, TiAgl scsecksugsaees st cece samereeeeede ness 117. 6603 +. 0004
Gallo siesenss sew capes ness ee debs ee cues scadesemaweinss 117. 5770+. 0074
Washburn and Bates.............. (Ege carta bragas Sahat asa aa a 117. 5625+. 0065
Bates and Vinal...........--.--..--- fe varcaaseyh eestendielsia cs ent 117. 6260+. 0025
General mean....... 0... eee eee cece eee eee eee 117. 6454+. 00026

Of these individual values, that given by Baxter’s series of 1910 is probably the best, but
it is not absolute and should not be assigned exclusive weight. The atomic weight of iodine
derived from the general mean is only 0.019 lower than that computed from Baxter’s data.

RATIOS CONNECTING THE SILVER HALIDES.

The three ratios between the silver halides, AgCl:AgBr, AgCl:AgI, and AgBr:AgI, have
all been measured with a high degreo of accuracy, and by essentially the same process.

When silver bromide is heated in chlorine gas, silver chloride is formed. In 1860 Dumas °
employed this method for estimating the atomic weight of bromine. His results are as follows.
In the third column I give the weight of AgBr equivalent to 100 parts of AgCl:

 

1 Journ. Amer, Chem. Soc., 34, 1367. 1912. 2 Journ. Amer. Chem. Soc., 36, 916. 1914. ® Ann. Chem, Pharm., 113, 20,
No. 8.) ATOMIC WEIGHTS—CLARKE. 85

AgBr. Agel. Ratio.
2. 028 1. 547 131, 092
4, 237 3. 235 130. 974
5. 769 4. 403 131. 024

 

Mean, 131. 030+. 023

The two series of determinations by Baxter + are much more elaborate, and far more con-
clusive. Before being weighed, the silver bromide was fused in a current of air saturated with
bromine. The figures given below are for vacuum weights, which is true for all of Baxter’s
data as cited in this section.

 

1905 Serves.
AgBr. AgCl. Ratio.
10. 92091 8. 33538 131. 019
18. 88062 10. 59457 131. 016
8. 21484 6. 27006 131. 017
7. 87887 6. 01352 131. 020
6. 90106 5. 26735 131. 016
9. 53704 7. 27926 131. 017
Mean, 131. 0175-4. 00045
1906 Series.
AgBr. AgCl. Ratio.
8. 03979 6. 13642 131. 0176
8. 57738 6. 54677 131. 0170
13. 15698 10. 04221 131. 0168
12. 71403 9. 70413 131. 0167
13. 96784 10. 66116 131. 0162
13. 08168 9. 98469 131.0174
12. 52604 9. 56059 131. 0175
11. 11984 8. 48733 131. 0170
8. 82272 6. 73402 131. 0172
11. 93192 9. 10721 131. 0162
12. 53547 9. 56767 131. 0190
17. 15021 13. 09009 131. 0167
10. 31852 7. 87572 131. 0168
Mean, 131. 0171+. 00013
The three series combine as follows:
DUM YS os) sich aos edaisa seas CRA Ses sd Rama ORET Sake Ree 131. 030 +. 023
Baxter; 1905: seeses vasee svete os esse eeieeae'y es 323 St 131. 0175+. 00045
Baxter; 19065. :...2sccsentaseees suse reeeeeeaeey esas 131. 0171+. 00013
General mean.............2..2-- 20 eee eee eee eee 131. 0172+. 00012

Dumas’s figures might be rejected altogether without changing the final mean.

:Silver iodide, heated in chlorine, is similarly converted into chloride. This ratio has
been repeatedly investigated, first by Berzelius.? His figures are subjoined, with the ratio
AgCl:AgI :: 100: x in the last column:

Ag. AgCl. Ratio.
5. 000 3. 062 163, 292
12. 212 7. 4755 163. 360

Mean, 163. 326+. 023

There are also two early experiments by Dumas,’ as follows:

Ag. AgCi. Ratio,
3. 520 2.149 163. 793
7. 011 4, 281 ' 163. 770

Mean, 163. 782+. 008

 

1 Proc. Amer, Acad,, 41, 82. 1905. Ibid., 42,201. 1906. 3 Ann, Chem. Pharm., 113, 28. 1860.
2 Ann. Chim. Phys. (2), 40, 430. 1829,
86 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The modern work upon this ratio began with an investigation by Ladenburg! in 1902,
which showed that the previously accepted value for the atomic weight of iodine was at least
a tenth of a unit toolow. Ladenburg made two series of determinations, with vacuum weights;
one preliminary, the other conducted with greater care. His figures are as follows:

Preliminary Series.

AglI. AgCl. Ratio.
31. 2558 19. 0817 163. 800
33. 7357 20. 5930 163. 821
49, 88229 30. 4525 163. 804
47. 8830 29. 2262 163. 836
60. 1435 36. 7154 163. 810
41. 3649 25. 2448 163. 855
50. 8916 31. 0664 163. 816
41. 3233 25. 2200 163. 851
80. 8139 49. 3181 163. 863
89. 5071 54. 6367 163. 822

: £ Mean, 163. 8278+. 0048
Final Series.

AgI. AgCl. Ratio.
62. 6658 38. 2526 163. 821
63. 8402 38. 9687 163. 824
74. 7576 45. 6324 163. 826

Mean, 163. 8237+. 00103

Ladenburg was followed, in the measurement of this ratio, by Koethner and Aeuer; ? whose
observations are of the highest significance. In a number of preliminary experiments they found
that silver iodide, precipitated from solutions of silver nitrate, was liable to contain occlusions of
the latter salt; a fact which accounts for the low values for iodine found by Marignac and Stas.
Their final determinations of the chloride-iodide ratio are as follows, with vacuum corrections:

AglI. AgCi. Ratio.
24, 88066 15. 18914 163. 8056
10. 24699 6. 25564 163. 8036
12. 57020 7. 67389 163. 8048
25. 18868 15. 37678 163. 8098

9. 62006 ; 5. 87285 163. 8057
12. 26770 7. 48901 163. 8093
22. 60660 18. 80056 163. 8093
20. 98601 12. 81160 163. 8048
22. 47667 13. 72119 163. 8099

Mean, 163. 8080+. 00057

Two series of measurements of this ratio are due to Baxter,’ who verified the occlusion of
silver nitrate by silver iodide. This source of error he obviated by fusing the iodide in an
atmosphere containing iodine. In one series, the silver iodide was first converted into bromide
and afterwards into chloride; in the other series‘the conversion was direct. Baxter’s determina-
tions appear in the two following tables:

Bromide Series.

Ag. AgCl. Ratio.
18. 65457 8. 33538 163. 815
17. 35528 10. 59457 163. 813
10. 27105 6. 27006 163. 811 :
8. 62870 5, 26735 168, 815
11. 92405 7. 27926 163, 809

Mean, 163. 8126+. 00079

 

1 Ber, Deutsch. chem. Ges., 35, 2275. 1902. ;

3 Liebig’s Annalen, 337, 123 and 367. 1904. Ibid., 338,362. See also Ladenbuirg, ibid., 338, 259. A preliminary paper by Koethner and Aeuer is
in Ber., 37, 2536, 1904.

8 Proc, Ame r Acad., 40, 431, 1904. Ibid., 41,73, 1995. Journ, Amer, Chem. Soc., 26, 1593, and 27, 879,
No. 3.]

Agl.

9. 26860

6. 72061
11. 31825
10. 07029
13. 49229

ATOMIC WEIGHTS—CLARKE.

Direct Series.

AgCl.

5. 65787
4. 10259
6. 90912
6. 14754
8. 23649

Ratio.
163. 818
163. 814
163. 816
163. 810
163. 811

87

Mean, 163. 8138+. 00101

The seven series for the ratio AgCl: AgI, arranged in the order of ascending magnitude, now
‘combine thus:

Berzelins.--<..2).s<c02s0e2 ees ccccwnes izes Qoereeeeess 163, 326 +. 023
Dumas..............-. hist cialtis dah aaclatiae tate ta tetanic aos 163. 782 +. 008
Koethner and Aeuer.......-..-.-2..- 2-0-2020 -ee ee eee 163. 8070+. 00057
Baxter, bromide series................0.000-00--20005 163, 8126+. 00079
Baxter direct: cacios Gc cuiien veo os dometnccwade teases 163. 8138+. 00101
Ladenburg, final...........2.......0--------2-+--+-- 163, 82374. 00103
Ladenburg, preliminary................--+-----+------ 163. 8278+. 0048
General M6aNl sssiescacese sss asgecesces cue ¥.8 ses 163. 8118+. 00038

For the ratio AgI : AgBr: : 100 :~ there is one set of determinations by Baxter.! Silver
iodide was converted into bromide by heating in bromine vapor. The data are as follows:

Agl. AgBr. Ratio.
13. 65457 10. 92091 79. 9799
17. 35528 13. 88062 79. 9792

9. 70100 : 7. 75896 79. 9812
10. 27105 8. 21484 79. 9805
9. 85688 7. 88351 79. 9798
8. 62870 6. 90106 79. 9780
11. 92405 9. 53704 79. 9816
7. 56933 6. 05389 79. 9792

THE POTASSIUM CHLORIDE-SILVER RATIOS.

Mean, 79. 9799+. 00028

These ratios have been repeatedly measured. First, let us consider the ratio Ag : KC1::
100 :z. Marignac? dissolved pure silver in nitric acid, and determined the ratio by titration

with a solution of potassium chloride.

The data are as follows:

Ag. KCl. Ratio.

4. 7238 3. 2626 69. 067
21. 725 15. 001 69. 050
21. 759 15. 028 69. 066
21, 909 15. 131 69. 063
22. 032 15. 216 69. 063
25. 122 17. 350 69. 063

Mean, 69. 062+. 0017

Corrected for weighing in air this becomes 69.098 + .0017.

The work of Stas falls into several series, widely separated in point of time. His earlier
experiments * upon this ratio may be divided into two sets, as follows: In the first set the silver
was slightly impure, but the impurity was of known quantity, and corrections could therefore
be'applied. In the second series pure silver was employed. The potassium chloride was from
several different sources, and in every case was purified with the utmost care. From 10.8 to 32.4
grammes of silver were taken in each experiment, and the weighings were reduced to a vacuum.
The method of operation was, in brief, as follows: A definite weight of potassium chloride was
taken, and the exact quantity of silver necessary, according to Prout’s hypothesis, to balance it

 

1 Proc, Amer, Acad., 41, 73. Journ. Amer, Chem. Soc., 27, 878. 1905, 8 Oeuvres Completes, 1, 363, 364.
2 Oeuvres Completes, 1, 77.
88 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

was also weighed out. The metal, with suitable precautions, was dissolved in nitric acid, and
the solution mixed with that of the chloride. After double decomposition the trifling excess of .
silver remaining in the liquid was determined by titration with a normal solution of potassium

chloride.
First series. Second series

69. 105 69. 105
69. 104 69. 099
69. 103 69. 107
69. 104 69. 103
69. 102 69. 103
69. 105
Mean, 69. 1036+. 0003 69. 104
69. 099
69. 1034
69. 104
69. 103
69. 102
69. 104
69. 104
69. 105
69. 103
69. 101
69. 105
69. 103

Mean, 69. 1033+. 0003

 

 

In these determinations Stas did not take into account the slight solubility of precipitated
silver chloride in the menstrua employed in the experiments. Accordingly, in 1882,1 he
published a new series, in which by two methods he remeasured the ratio, guarding against the
indicated error, and finding the following values:

69. 1198 69, 121
69. 11965 69. 123

Mean, 69. 1209+. 0003
Corrected for a minute trace of silica contained in the potassium chloride, this mean
becomes
69.11903 + .0003 ?

Still later, in order to establish the absolute constancy of the ratio in question, Stas made
yet another series of determinations,’ in which he employed potassium chloride prepared from
four different sources. One lot of silver was used throughout. The values obtained were as
follows:

69. 1227 © 69, 1228 69. 1249
69. 1236 69. 1222 69. 1288
69. 1234 69. 1211 69. 1225
69. 1244 69. 1219 69. 1211
69. 1235

A series was also begun in which one sample of potassium chloride was to be balanced
against silver from various sources, but only one result is given, namely, 69.1240. This, with
the previous series, gives a mean of 69.1230 + .0002.

The difference between the highest and the lowest of Stas’ series corresponds to a difference
of 0.021 in the atomic weight of potassium. The rejection of the earlier work might be quite
justifiable, but would exert a very slight influence upon our final result.

In 1903, incidentally to their work on cesium, Richards and Archibald‘ published two
analyses of potassium chloride, in which both ratios were determined. That is, the silver

 

1 Oeuvres Complates, 762-767, 775-777, * Oeuvres Completes, 3, 516, 539,
§ Ann. Chim. Phys. (6), 7, 513. 1886. ‘4 Proc, Amer. Acad., 38, 456, 1903.
89

chloride was weighed, giving data for the second ratio, AgCl:KCl::100:z. The results, with
vacuum weights, follow:

ATOMIC WEIGHTS—CLARKE.

No. 3.)

 

 

KCl. AgCi. Ag. Ag ratio. AgCl ratio.
2. 50019 4, 80600 3. 61747 69, 114 52. 022
2. 50391 4, 81325 3. 62283 69. 115 52. 021

Mean, 69. 1145 52. 0215
+. 0003 +. 0003

In 1904 Archibald ' gave an additional series of determinations of these ratios, also with
vacuum weights, as follows:

 

 

KCl. AgCl. Ag. Ag ratio. AgCi ratio.
2, 21586 4, 25916 3. 20598 69. 116 52. 026
1. 99379 3, 83250 2. 88479 69, 114 52. 023
2. 89977 6. 57396 4, 19557 69. 115 52. 024
4. 73606 9. 10362 6. 85280 69. 111 52. 024
Mean, 69. 114 52. 024

+. 0007 +. 00055

The measurements of these ratios by Richards and Staehler ? are probably the most con-
clusive, for every care was taken to detect and avoid constant errors, such as the authors believe
were present, despite all precautions, in the work of Stas. The occlusion of silver nitrate by
silver chloride is an error of this kind. The figures for the silver ratio are as follows, with
vacuum weights:

KCl. Ag. Ratio.
3. 88074 5. 61536 69. 109
7. 44388 10. 77156 69. 107
5. 00681 7. 24614 69. 106
5. 04833 7. 30515 69. 107
8. 19225 11. 85412 69. 109
4. 99795 7. 23230 69. 106
5. 16262 7. 47042 69. 107

 

Mean, 69. 1073-4. 00032

For the silver chloride ratio Richards and Staehler give the subjoined figures:

KCl. AgCl. Ratio.
4, 36825 8. 3986 52. 012
5. 56737 10. 7038 52,013
6. 41424 12, 3323 52,012
3. 27215 6, 2913 52, 011
4, 83028 9, 2870 52, O11

 

Mean, 52. 0118+. 00025

Several earlier measurements of the silver chloride ratio remain to be mentioned. First,
Berzelius * found that 100 parts of KCl were equivalent to 192.4 of AgCl, a value of which,
corrected for weighing in air, becomes 192.32. Hence AgCl : KCI :: 100 : 51.997.

In 1842 Marignac ‘ published two determinations as follows:

KCl. _AgCl.
17, 034 32. 761
14, 427 27, 749

Ratio.
51. 995
51. 991

 

Mean, 51. 993

 

1 Trans, Roy. Soc. Canada, 1904, Section III, p. 47.

1 Publ. Carnegie Inst., Washington, No. 69, p. 7. 1907. An advance publication in Ber. Deutsch. chem. Ges., 39, 3611, contained also the

figures of some preliminary experiments, which the authors discard in their final report.
8 Poggend. Annal.,8,1. 1826,
# Oeuvres Completes, 1, 60.
‘

90 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

Four years later, Marignac ! published a second series of determinations. The new figures
Pp

 

are:
KCl. AgCl. Ratio.
15. 028 28. 910 51, 982
15, 131 29. 102 51. 993
15. 216 29, 271 51. 983
Mean, 51. 986

The mean of both series, taken as one, is 51.989; which, corrected to a vacuum, becomes
52.011-+.0018.

In three determinations Maumené ? obtained the following figures:

KCl. AgCl. Ratio.
10. 700 20. 627 51. 874
10.5195 20. 273 51. 892

8.587 16. 556 51. 868

 

Mean, 51. 878-+.0049

These figures seem to represent weights in air, but they are hardly worth correcting.
Two other analyses, with vacuum reductions, were made by Thiel * incidentally to his
research upon indium:

KCl. AgCl. Ratio,
7. 4314 14, 2903 52. 003
7.4321 14, 2939 51. 995

 

Mean, 51. 999+ .0027
Assembling the data for both ratios, we now have the following combinations:

Ratio Ag: KCl:: 100:2.

Marenae ss Se ieex ccc aseceeacssmncacsene se geseseceekc 69.098 +. 0017
Stag) firsts. 2s... wi «Ue vee ca wesraemnaseeamaneaneeaecece 69. 1036+. 0003
Dtas, BCONd accu wwce5240 6 sa5acecdeneenansseaneeeeine 69. 1033+. 0003
Stasi phird ..0occ joergens saniat asses eu ceeencen 69. 1190+. 0003
Stas, MUP hes sins rrnchacasresccas seed caeenceete 69. 1230+. 0002
Richards and Archibald. .............-.-- qeseehuntinciels 69. 1145+. 0003
Archibald... ......----2e2ececeeeee eee eudgiellg vena 69.114 +.0007
Richards and Staehler......... Pee ecieisierea's eels raat 'S 69. 1073+. 00032
General MEAN 6400s cee cece seesecacededseeses 69. 1188+. 00011
Ratio AgCl:KC1::100:x
Berzelius: cc: s,s adsoasedaseariioeearenceue etic sees 51.997 +. 0049
Marignaes: 2:cciecicis cisacisiecie dan ini tae beaanteedaeuiaces 52.011 +. 0018
Mam en 3.00. < sisis cic ccassaicscibe dita gasie suas Saisie edena aa 51.878 +.0049
Richards and Archibald. ...... keemacauenmececeauned 52. 0215+. 0003
Piola a se et oir swins aeypaauca aa Qa Cathal neemeeanes 51.999 +. 0027
Archibald . naa ciate tiseicrersraiSa Tis) see ee Sie eS SSeS Ee wiasety 52.024 +.00055
Richards and Staehler...... petedecmecaoekekee es seed 52. 0118+. 00025
General Mean cass Jans sattedicekerscaress a4 52. 0163+. 00018

In the last combination the single experiment by Berzelius is given equal weight with
Maumené’s series. Both general means differ from Richards and Staehler’s averages by less
than one part in 10,500, or 0.01 per cent.

 

1 Oeuvres Complttes, 1, 78. The figures of the first set are republished in this series, but not repeated here. Marignac treats the five experi-
ments as one series.

3 Ann, Chim. Phys. (3), 18,41. 1848.

8 Zeitsch. anorg. Chem., 40, 313. 1904,
No. 3.] ATOMIC WEIGHTS—CLARKE. 91

POTASSIUM BROMIDE AND IODIDE RATIOS.

The ratio between silver and potassium bromide was first accurately determined by Marig-
nac. I give, with his weighings, the quantity of KBr proportional to 100 parts of Ag:

Ag. KBr. Ratto.
2.131 2. 351 110. 324
2. 559 2. 823 110. 316
2. 447. 2.700 110. 339
3. 025 3. 336 110. 283
3. 946 4. 353 110. 314
11. 569 12. 763 110. 321
20. 120 22. 191 110. 293

 

Mean, 110. 313+. 005
Corrected to a vacuum this becomes 110.343, +.005.

Stas,? working in essentially the same manner as when he compared potassium chloride
and silver, and with bromide from several distinct sources, found the following values for
this ratio: =

110. 361 110. 336
110. 360 110, 344
110. 360 110. 332
110, 342 110. 343
110. 346 110. 357
110. 338 110. 334
110. 360 110. 335

 

Mean, 110. 3463+. 0020

In his paper on the atomic weight of nitrogen, Dean * gives three measurements of the
Ag : KBr ratio, but with a bromide which was supposed to be not quite pure. His results,
however, are so close to later determinations that they are worth citing:

Ag. KBr. Ratio.
8. 52439 9. 40336 110. 311
7. 83113 8. 63900 110. 316
8. 92432 ; 9. 84450. si 110. 312

 

Mean, 110, 3134. 0010

The recent measurements of this ratio by Richards and Mueller ‘ differ considerably from
the concordant results of Stas and Marignac. The modern work was probably based upon
purer materials, especially in the case of the silver employed. For details upon this side of
the discussion the original memoirs must be consulted. The figures, with vacuum weights,
given by Richards and Mueller, are as follows:

KBr. Ag. Ratio.
4. 33730 3. 93164 110. 318
4, 18763 3. 79587 110. 320
4, 15849 3. 76943 110. 321
3. 67867 38. 33450 110. 321
3. 60484 3, 26776 110. 315
4. 78120 4, 33387 110. 322
5. 67997 5. 14860 110. 321
6. 41587 . 5. 81571 110. 320
2. 88134 2. 61184 110. 318
3. 64383 2. 30309 110. 316
3. 12757 2. 83504 110. 318

 

Mean, 110. 319+. 0004

 

1 Oeuvres Complétes, 1, 82. Four preliminary analyses are discarded. 3 Journ. Chem. Soc., 77, 177. 1900.
3 Oeuvres Complétes, 1, 747. A ; 4 Publ. Carnegie Inst., Washington, No. 69, p. 27. 1907.
92 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

This combines with the former determinations thus:

Marignac........... Ue ic batura saiaigebeseceeaiodetek ed 110. 343 +. 0050
SERB a ck eee aieeeecre te stetaleiuldtoteere Sd eeaceaiahee 110. 3463+. 0020
DCBD ies sss 8.38 aarryeerde ang ee einen Eee aas <eeeleoeas 110. 313 +. 0010
Richards and Mueller. .................---0eeeeeeeeee 110. 3190+. 0004
General ican sie jcceie soca cece cesses senate 110. 3193 +. 00033

Richards and Mueller also determined the second ratio, AgBr:KBr::100:x. Their figures
are—

KBr. AgBr. Ratio.
2. 19027 3. 45617 63. 3728
4, 19705 6. 62285 63. 3723
2. 06723 3. 26206 63. 3719
2. 58494 4. 07889 63. 3736

 

Mean, 63. 3727+. 0003

When applied to the determination of the atomic weight of potassium, the Richards and
Mueller ratios yield almost absolutely identical results, which also coincide with the figures
obtained by Richards and Staehler with the chloride. This agreement is strong evidence in
favor of the new determinations.

The ratio between silver and potassium iodide seems to have been measured only by Marig-
nac,! but without remarkable accuracy. The figures are as follows:

Ag. KI. Ratio.
1.616 2.483 153.651
2.503 3.846 153.665
3.427 5.268 153.720
2.141 3.290 153.667

10.821 . 16.642 153.794

 

Mean, 153.6994+ .0178
Corrected to a vacuum. by Marignac, this becomes 153.800.

THE SODIUM HALIDE-SILVER RATIOS.

The ratio between silver and sodium chloride has been fixed by several investigators.
Pelouze? dissolved a weighed quantity of silver in nitric acid, and then titrated with
sodium chloride. Equivalent to 100 parts of silver he found of chloride:

; ' 64.158

54.125
54.139

Mean, 54.141+ .0063

 

By Dumas * we have seven experiments, with results as follows:

NeCcl. Ag. Ratio.
2.0535 3.788 54.211
2.169 4.0095 54.097
4.3554 8.0425 54.155
6.509 12.0140 54.178
6.413 11.8375 54.175
2.1746 4.012 54.202
5.118 9.434 54.187

 

Mean, 54.172+ .0096

Stas,‘ applying the method used in establishing the similar ratio for potassium chloride,
and working with salt from six different sources, found of sodium chloride equivalent to 100
parts of silver:

 

1 Oeuvres Complétes, 1, 86. 3 Ann. Chem. Pharm., 113,31. 1860.
Compt. Rend., 20, 1047. 1845. 4 Oeuvres Complates, 1, 370.
No. 3.] ATOMIC WEIGHTS—CLARKE. 93

54.2093 54.2060
54.2088 54.2076
54.2070 54.2081
54.2070 54.2083
54.2070 54.2089

Mean, 54.2078-+.0002

Asin the case of the corresponding ratio for potassium chloride these data needed to be checked by
by others which took into account the solubility of silver chloride. Such data are given in
Stas’ paper of 1882,! and four results are as follows:

54.2065 54.2091

54.20676 54.2054
jMean, 54.20694 + .00045

Corrected for a trace of silica in the sodium chloride, this mean becomes 54.2047 x .00045.

The elaborate research of Richards and Wells? upon this ratio, gave a lower value than that
found by Stas. According to Richards and Wells, the silver by Stas probably contained occluded
oxygen, and his silver chloride carried down occlusions of sodium salts. The new data, with
vacuum weights as usual, are as follows, the last two experiments forming a small supple-
mentary series:

NaCl. Ag. Ratio.
3.96051 7.80896 54.187
2.32651 4.39355 54.186
5.36802 9.90699 54.184
4.00548 7.39210 54.186
4.69304 8.66101 54.186
3.27189 6.03842 54.185
5.08685 9.38795 54.185
3.66793 9.76952 54.183
5.48890 10.12993 54.185
3.55943 6.56909 54.185
3.38684 6.25046 54.185
4.68529 8.64634 54.188

‘ ile ‘ Mean, 54.1854+ .00025
The five series of determinations combine thus:

> PelOUZE lca sinss cesses «eel die Laie ole bse eeieinspartgsesmiees es 54.141 +.0063
Dinas’ siec cateteca ee tees taceaneaeneh ares Maat eee: 54.172 +.0096
i tee earliest ie haves plea means tapouaeecken aes 54.2078 + 0002

Stas, datenaccsecsc sienocacisweciwmnics Sanne teeeeseae sss 54.2047 + .00045

Richards and Wells .......-....-.2-222eeeeeer cece ceeee 54.1854+ .00025

General: Mean sec secsiwiesiercasne cis se cgtoassaeu ones 54.1995+ .00015

In this combination the work of Pelouze and Dumas counts for almost nothing. Stas’
determinations carry high weight, and it is not easy to understand how their supposed syste-
matic errors could have been so uniform in magnitude. Such errors should vary from experi-
ment to experiment, and so tend to increase the ‘probable error”’ of the mean.

In their research upon the atomic weight of boron, Ramsay and Aston® converted borax into
sodium. chloride. In the latter the chlorine was afterwards estimated gravimetrically by weigh-
ing as silver chloride on a Gooch filter. Hence the ratio, AgCl: NaCl: : 100: 2, as follows:

NaCl. AgCl. Ratio.
3.0761 7.5259 40.874
2.7700 6.7794 40.859
2.8930 7.0804 40.859
2.7360 6.6960 40.860
1.9187 4.6931 40.863

 

Mean, 40.867 + .0033

 

1 Oeuvres Complétes, 1, 768, 778. | 2 Publ. Carnegie Inst., Washington, No. 28, pp. 52, 56. 1905. 8’ Chem. News, 66,92. 1892.
94 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

The same ratio was also measured, much more exactly, by Richards and Wells. The
occlusion of sodium salts by the silver chloride was especially considered and guarded against.
The figures obtained are as follows:

NaCl. AgCl. Ratio.
3. 27527 8. 03143 40. 781
5. 56875 13. 65609 41.779
4. 18052 10. 25176 40. 779
4, 54319 11.,14095 40, 779
1, 97447 4, 84196 40.778
3.97442 9. 74547 40. 782
6. 69495 16. 41725 40, 780
2. 88692 7. 07955 40. 778
5. 56991 13. 65833 40. 780
5. 85900 14, 36693 40. 781

 

Mean, 40. 7797 +. 00028

This mean, combined with that of Ramsay and Aston, gives a general mean of 40.7803 +
.00028, which falls within the limits of variation of Richards and Wells’ series.

For the ratio between silver and sodium bromide we have one set of measurements by
Stas.! The bromide was prepared by saturating Na,CO, with HBr. The HBr proportional to

100 parts of silver was—
95. 4420 95. 4426

95. 4383 95. 43892
Mean, 95. 4405+. 0007

 

The second bromide ratio, AgBr:NaBr, is represented by two experiments, made by Richards
and Wells? in a research upon the transition temperature of sodium bromide. With vacuum
weights the figures are—

NaBr. AgBr. Ratio.
5. 49797 10. 03253 54, 8014
3. 64559 6. 65248 54, 8005

 

Mean, 54. 8010 +. 0005
THE AMMONIUM HALIDE-SILVER RATIOS.

Ratios connecting silver with the chloride and bromide of ammonium have been repeatedly
determined by methods essentially the same as those adopted in the similar analyses of
potassium and sodium halides.

For the ammonium chloride equivalent to 100 parts of silver, Pelouze® found:

"49, 556
49. 517

Mean, 49. 5365+. 013

 

Marignac ‘ obtained the following results. The usual ratio for 100 parts of silver is given
also:

Ag. . N#H,Cl. Ratio.
8. 063 - 8, 992 49. 510
> 9, 402 4, 656 49, 521
10. 339 5. 120 49. 521
12. 497 6.191 49.540
11. 337 5.617 49. 546
11. 307 5. 595 49. 483
4, 326 2.143 49. 588

 

; Mean, 49. 523+. 0055
Corrected to a vacuum this becomes 49.556 + .0055.

 

1 Oeuvres Complétes, 1, 796. 3 Compt. Rend., 20,1047, 1845.
Proc. Amer. Acad., 41, 443 4 Oeuvres Completes, 1, 89.
No. 3.] ATOMIC WEIGHTS—CLARKE. 95

Stas ' made three series of determinations of this important ratio at different times and
under varying conditions. All of his weights, as usual, were reduced to a vacuum standard.
The third series, published in 1882, was undertaken in order to correct for the solubility of
silver chloride, which was not sufficiently guarded against in the earlier work. The values
found for the ratio Ag:NH,C1::100:z are as follows:

 

 

First series. Second series. Third series.
49. 568 49. 598 49. 599
49. 581 é 49.597 49. 600
49. 572 "49.593 49. 597
49. 577 49. 597 49. 5987
49. 595 49. 5974 49. 597
49. 588 49. 602 -

49. 591 49, 597 Mean, 49. £98+. 0005
49. 593 49. 598

49. 600 49. 592

49. 599

49. 598 Mean, 49. 597+. 0006

49. 597

49. 591

49. 592

Mean, 49. 589+. 0018

The first four determinations in the first series are rejected by Stas as unsatisfactory.
By Scott ? two determinations of this ratio are available, with vacuum weights, as follows:

Ag. N#,Cl. Ratio.
9. 64484 4. 78257 49. 587
11. 12810 5. 51744 49. 581

 

Mean, 49. 584+. 0020

Scott also made one determination of the ratio AgCl:NH,Cl. 4.7850 grammes NH,C]
balance 12.82048 of Ag. The ratio, therefore, is 100:37.3234.
The several values for the ratios Ag:NH,CI::100:2 now combine as follows:

PelOU Ze sctS.ceccece Soe ase aaaad og ldddoeeea ey 49. 5365 +. 0130
Margin. taiice Sar os se or enheetidoonns sx ees peeeonees 49.556 +. 0055
Dlas i firsth Series ssc onic aoaweee ae eed sew seta eddie 49.589 +. 0018
Stas, second series... . 2... .:5----2e eee cee ec eens ee ee cane 49.597 +. 0006
tas; thitd series 6352s vegas aeainkhss 204 iawaeeensans 49.594 +. 0005
SCOlb des deccsetnaacinalsdsece anahioneteecec cose esees 49.584 +. 0020
General mean........-..-.2.22-2. 2200 e eee ee eee 49. 5965 +. 00038

For the ratio between ammonium chloride and silver chloride there is a series of nine
determinations by Richards, Koethner and Tiede.* The values are found as follows:

NG,Cl. AgCl. Ratio.
2. 02087 5. 41469 37. 3220
2. 23894 5. 99903 37.3217
1. 55284 4. 16076 37.3211
1. 36579 3. 65959 37. 3209
1. 61939 4, 33914 37. 3205
1. 93795 5. 19219 37, 3243
2. 89057 7. 74498 37. 3219
1. 31405 38. 52082 37. 3223
1, 82091 4, 87921 37. 3198

 

Mean, 37. 3217

 

1 Oeuvres Completes, 1, 378, 478, 781.
2 Journ. Chem. Soc., 79, 147. 1901.
-* Journ. Amer. Chem. Soc., 31, 6. 1909. The actual analyses were made by Tiede.

161566°—20 7

 
96 MEMOIRS NATIONAL ACADEMY OF SCIENCES.

[Vou. XVI.

Scott’s single determination of this ratio, 37.3234, falls within the limits of variation of the
foregoing series. Including it in the computation, the ratio becomes

AgCl : NH,C1 :: 100 : 37.3218 + .0008

All weights were reduced to a vacuum basis. ;

Three additional determinations, by Tiede, were not published in the oueiaal memoir.
I am indebted to the kindness of Professor Richards for them and include them here. The-
data are as follows:

NH,Cl. ~  AgCl. Ratio.

1.10489 2. $607 37.3185
-95997 2.5723 37.3195
-75831 2.0186 37.3184

 

Mean, 37.3188 + .00025

Combined with the previous series of determinations, the general mean becomes 37.3205 +
0.0002.

For the ratio Ag : NH,Br : : 100: z there are determinations by Stas and Scott.

Stas! obtained the following values for z:

‘ 90.831 90.8317
: 90.831 90.8311

90.8297 90.8302

90.823 —_——_

Mean, 90.8297 4.0008

Scott’s ? aia rejecting three preliminary experiments in which the ammonium bromide
was distinctly acid, are as follows, with vacuum weights:

Ag. NH,Br. Ratio.
4.92273 4.46957 90.795
4.20661 4.63303 90.796
4.23664 4.66644 90.790
4.31464 4.75175 90.801
6.19233 6.82047 90.790

, 8.77664 9.66708 90.789
10.47233 11.53416 90.794

4.91997 5.41834 90.802

5.00442 5.51164 90.797

5.17914 5.70390 90.800

4.84099 5.33177 90.795

5.10677 5.62515 90.784

 

Mean, 90.7944 .0011
Combining this with the series by Stas, the general mean is 90.8175 + .00065.

THE SILVER NITRATE RATIOS.

The quantity of silver nitrate which can be formed from a known weight of metallic silver

has been determined by several investigators.
evaporated to dryness without transfer, and weighed.

gave of nitrate:

Marignac’s * results were as follows.
standard of 100 parts of silver:

157.430 157.440
157.437 157.430
157.458 157.455

 

Mean, 157.4417 + .0033

Penny ° dissolved silver in nitric acid in a flask,
One hundred parts of silver thus

In the third column they are reduced to the common

 

1 Oeuvres Complétes, 1, 801.

3 Journ. Chem. Soc., 70, 147, 1901.

8 Phil. Trans., 1839.

For a criticism by Richards, see Proc. Amer. Phil. Soc., 43, 116, 1904,

4 Oeuvres Completes, 1,88. From the sum of the weights, corrected to 4 vacuum, Marignac computes the ratio 1 : 157.455,
No. 3.) ATOMIC WEIGHTS—CLARKE. 97

Ag. Ag NO;. Ratio.
68.987 gave 108.608 157.433
57.844 91.047 157.401
66.436 104.592 157.433
70.340 110.718 157.404

200.000 314.894 157.447

Mean, 157.4236 + .0061

Corrected for weighing in air this becomes 157.449.

Stas,’ employing from 77 to 405 grammes of silver in each experiment, made two different
series of determinations at two different times. The silver was dissolved with all the usual
precautions against loss and against impurity and the resulting nitrate was weighed, first after
long drying without fusion, just below its melting point, and again fused. Between the fused
and the unfused salt there was in every case a slight difference in weight, the latter giving a
maximum and the former a minimum value.

In Stas’s first series there are eight experiments, but the seventh he himself rejects as
inexact. The values obtained for the nitrate from 100 parts of silver are given below in two
columns, representing the two conditions in which the salt was weighed. The general mean
given at the end I have deduced from the means of the two columns considered separately:

Onfused. Fused.
157.492 157.474
157.510 157.481
157.485 157.477
157.476 157.471
157.478 157.470
157.471 157.463
157.488 157.469

Mean, 157.4857 Mean, 157.472

General mean, 157.474+ .0014

In the later.series there are but two experiments, as follows:

Onfused. Fused.
157.4964 157.488
157.4940 157.480

Mean, 157.4952 Mean, 157.484

General mean, 157.486 + .0003

The reverse ratio, namely, the amount of silver obtainable from a weighed quantity of nitrate,
has been determined electrolytically by Hardin.? The data obtained, however, are reducible
to the same form as in the preceding series and all properly combinable together. Pure silver
was dissolved in pure aqueous nitric acid, and the crystalline salt thus formed was dried, fused,
and used for the determinations. The silver nitrate, mixed with an excess of pure potassium
cyanide solution, was electrolyzed in a platinum dish. The results obtained, reduced to
vacuum weights, were as follows:

AgNO;. Ag. Ratio.
‘ 0.31202 0.19812 157.490
47832 .30370 157.498
-56742 36030 157.485
.07728 36655 157.490
-69409 44075 157.479
-86367 54843 157.479
86811 -55130 157.466
-93716 -59508 157.485
1.06170 -67412 157,494
1.19849 76104 157.477

Mean, 157.484 + .0020

The most thorough and recent investigation of this ratio is that by Richards and Forbes.®
They effected the synthesis of the nitrate from, the purest silver, the nitrate having been fused

1 Oeuvres Complétes, 1, 346, 724. 1Journ. Amer. Chem. Soc., 18, 995. 1896. 3 Publ. Carnegie Inst., Washington, No, 69, p. 47. 1907.
98 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

and tested for such impurities as dissolved air, retained water and ammonia, and nitric or nitrous
acids. Only two of these were found, and in minute traces, between 0 001 and 0.002 per cent
in all. Tho final data, with vacuum weights, are as follows:

Ag. AgNO;. Ratio.
6. 14837 9. 68249 157. 481
4, 60825 7. 25706 157. 480
4. 97925 7. 84131 157. 480
9. 07101 14, 28503 157. 480
9. 13702 14. 38903 157. 481
9. 01782 14. 20123 157. 480

Mean, 157. 480+:.0001

The impurities above mentioned may lower this value to 157.478, their maximum effect.
The authors accept the intermediate figure, 157.479.
Combining the several determinations, we have—

Penny nectar cheat ean uae eked 157. 4417+. 0033
Marignae sc sdschaiddecdandaciiedeiecie vies Vsdecloaaise sas 157. 449 +.0061
Stas, earlier. ...-.......--------4-- sg oiSeesaldi dia ceaesrsyers 157. 474 +. 0014
Bias) Tater ccc cuinns iclldisadesmare goes ict eaats caaseuss 157. 486 +. 0003
Padi st cccamacacamecadecatsdesngns tone aaataWensiac 157. 484 +. 0020
Richards and Forbes...........---------++--e+eeeeeee ee 157.479 +. 0001
General mean....--....------------2+ eee eee eee 157. 479 +. 000095

For the direct ratio between silver nitrate and silver chloride there are two series of esti-
mations. A weighed quantity of nitrate is easily converted into chloride, and the weight of
the latter ascertained. In two experiments Turner ' found of chloride from 100 parts of nitrate:

84. 357
84, 389
Mean, 84 873-4. 011

Penny,’ in five determinations, found the following percentages:

84. 370 84. 367
84. 388 84. 370
84. 377 Mean, 84. 3744+. 0025

The general mean from both series is 84.3743. + 0025.

The ratio directly connecting silver nitrate with ammonium chloride has been determined
only by Stas. The usual method of working was followed, namely, nearly equivalent quanti-
ties of the two salts were weighed out, the solutions mixed, and the slight excess of one esti-
mated by titration. In four experiments 100 parts of silver nitrate were found equivalent to
chloride of ammonium, as follows:

31. 489 31. 487
31. 490 31. 486

Mean, 31.488 + .0006
The similar ratio between potassium chloride and silver nitrate has been determined by
both Marignac and Stas.
Marignac‘ gives the following weights. I add the quantity of KCl proportional to 100
parts of AgNO,:

KCI. Ag NOs3. Ratio.
1. 849 4. 218 43. 836
2. 473 5. 640 43. 848
3. 317 7, 565 43. 847
2. 926 6. 670 43. 868
6. 191 14. 110 43. 877
4. 351 9. 918 43, 870

Mean, 43. 858+. 0044
Corrected to a vacuum this becomes 43.874 +.0044.

‘Phil, Trans., 1833, 537. 2 Phil. Trans., 1839. 3 Oeuvres Complétes, 1, 382. 4 Oeuvres Complates, 1, 88,

 
No. 3.) | ATOMIC WEIGHTS—CLARKE. 99

Stas’s! results are given in three series, representing silver nitrate from three different
sources. In the third series the nitrate was weighed in vacuo, while for the other series this
correction was applied in the usual way. For the KCl equivalent to 100 parts of AgNO, Stas
found:

First series. Second series. Third series.
43. 878 43, 864 43. 884.
48, 875 43. 869 43, 878
48. 875 48. 876 43. 885
48. 874

 

 

 

Mean, 43. 8755+ .0005 Mean, 43. 8697 + .0023 Mean, 43. 8823-t .0015

Combining all four series we have—

Marignaes «ce tees ia see S dictates ase eeu eee 43. 874 +. 0044
Stas, firstseries:.:.52 s<sesccsesee es sss cx ewneeeeeces eyes 43. 8755+. 0005
Stas, second series. .....-.-..22-2 22-222 ee eee eee eee eee 43, 8697 +. 0023
Stake, Ghird sorted <5 ss4s @aninwaunndcenaa sue paconeeet 43, 8823-4. 0015
General Mean, : 0. sesia eens eee ce ese ecceuen: 48. 8759+. 00046

POTASSIUM AND SODIUM NITRATE RATIOS.

Ratios connecting the alkaline nitrates, chlorates, and chlorides have been determined by
Penny, Stas, and Hibbs.

The general method of working upon these ratios is due to Penny.? Applied to the ratio
between the chloride and nitrate of potassium, it is as follows: A weighed quantity of the
chloride is introduced into a flask which is placed upon its side and connected with a receiver.
An excess of pure nitric acid is added, and the transformation is gradually brought about by
the aid of heat. Then, upon evaporating to dryness over a sand bath, the nitrate is brought
into weighable form. The liquid in the receiver is also evaporated, and the trace of solid matter
which had been mechanically carried over is recovered and also taken into account. In another
series of experiments the nitrate was taken, and by pure hydrochloric acid converted into chlo-
ride, the process being the same. In the following columns of figures I have reduced both
series to one standard, namely, so as to express the number of parts of nitrate corresponding
to 100 of chloride:

 

 

First series. Second series.
KCl treated with HNOg. KNO, treated with HCI.
135. 639 135. 628
135. 637 135. 635
185. 640 135. 630
185. 635 185. 641
135. 630 135. 630
135. 640 135. 635
135. 630 135. 630
Mean, 135. 636+. 0011 Mean, 135. 633-4. 0011

Stas,* who converted potassium chloride into nitrate, gives the following figures:

KCl. KNO;. Ratio.
50. 7165 68. 7938 135. 643
80. 2610 108. 8665 135. 638
72. 1022 99. 8050 135. 647
50. 2175 68. 1200 135. 649
48, 9274 63. 3675 : 135. 645
69. 8836 94. 7900 135. 640
14. 2578 19. 3415 185. 655

 

Mean, 135. 6453+. 0014

 

10euvres Completes, 1, 381. 2 Phil. Trans,, 1839. 3 Oeuvres Completes, 1, 683.
100 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

These figures by Stas represent weighings in the air. Reduced to a vacuum, this mean
becomes 135.6423. :

The determinations made by Hibbs! differ’slightly in method from those of Penny and
Stas. He converted the nitrate into the chloride by heating in a stream of gaseous hydro-
ehloric acid. His results were as follows, vacuum weights being given:

Weight KNO,. Weight KCl. Ratio.
0.11090 0.08177 135.624
14871 10965 135.622
-21067 .15533 135.627
23360 17225 135.620
24284 17903 135.642

 

eee Mean, 135.627+ .0026
Now, combining, we have:

Penny) Ist serlesscsoesccs', sos s snancadedeneeweceeeewecs 135.636 +.0011
Penny; 2d SeV168s:o:0 23 cosndyeek nos xanexareeeinewesce es 135.633 +.0011
Stags visie ses sv seas e ss sasweeseeeviweeeme lcd cue eane ees 135.6423 + .0014
HibbSiccsscces vise ee eotert ficawadenkdvecqueieaoances sd 135.627 + .0026

General mean...........- hes (spay oGpeeo pennies sieeneineydiare a 6 135.636 + 0007

By the same general process Penny’ determined how much potassium nitrate could be
formed from 100 parts of chlorate. He found as follows:

82.505 82.498
82.497 82.500

Mean, 82.500+:.0012

 

For 100 parts of sodium chlorate he found of nitrate:

79.875
79.882
79.890

Mean, 79.8823 + .0029

 

For the ratio between the chloride and nitrate of sodium Penny made two sets of estima-
tions, as in the case of potassium salts. The subjoined figures give the amount of nitrate
equivalent to 100 parts of chloride:

First series. Second series.

NaCl treated with HNO. NaNO, treated with HCl.
145.415 145.419
145.408 145.391
145.420 145.412
145.424 145.415
145.410 145.412
145.418 145.412
145.420

 

 

Mean, 145.410+ .0026
Mean, 145.4164+..0015

The sodium chloride to nitrate series of Stas? is as follows:

NaCl. NaNO. Ratio.
120.0110 174.5590 145.453
32.4837 47,2550 145.468
68.1295 99.1045 145.465
47.9226 69.7075 145.459
14.5380 21.1465 145.443

 

Mean, 145.4576 + .0030

 

1 Doctoral dissertation, University of Pennsylvania, 1896. Work done under the direction of Professor Edgar F. Smith.
3 Phil. Trans., 1839.
% Oeuvres Complates, 1, 688.
No. 3.] ATOMIC WEIGHTS—CLARKE. 101

Reduced to a vacuum basis this becomes 145.4526.
Hibbs’ data, obtained by the method employed in the case of the potassium compounds,
are as follows, vacuum weights being stated:

Weight NaNO . Weight NaCl. Ratio.
0.01550 0.01066 145.403
.20976 14426 145.404
.26229 .18038 145.410
66645 45829 145.429
93718 64456 145.399

 

os Mean, 145.407 + .0026
Combining, we have as follows:

Penny, Ist series........--. 22-0... c eee eee eee eee eee eee 145.4164+ .0015
Penny, 20 SOLES... .-cacceyesecevangcscuvantagenanene 145.410 +.0026
Staeee ocnnds ai orosceceaeeseseaeaes Geet ene eu oemaes 145.4526 + .0030
Hibbsivscciie.ce skcsyseweticees os hes stegsee seeks e's ose 145.407 +.0026

General Mean wccsccsesouse si seeeieedisns eves oerdees 145.418 +.0012

One other potassium nitrate ratio has been measured by Richards and Archibald? On
heating the nitrate with silica, potassium silicate is formed, and the equivalent of N,O, is vola-
tilized. The vacuum weights are given in the following table together with the ratio N,O;:
K,O ::100: z:

KNO, taken, N,0, lost. Ratio.
1.81034 0.96692 87.227
3.14564 1.68005 87.235
2.55598 1.36512 87.235

 

Mean, 87.232-+.0017
THE LITHIUM RATIOS.

The earlier determinations of the atomic weight of lithium by Arfvedson, Stromeyer, C. G.
Gmelin and Kralovanzky were all erroneous, because of the presence of sodium compounds
in the material employed. The results of Berzelius, Hagen, and Hermann were also incor-
rect, and need no further notice here. The only investigations which we need to consider are
those of Mallet, Diehl, Troost, Stas, Dittmar, and Richards and Willard.

Mallet’s ® experiments were conducted upon lithium chloride, which had been purified as
completely as possible. In two trials the chloride was precipitated by nitrate of silver, which
was collected upon a filter and estimated in the ordinary way. The figures in the third column
represent the LiCl proportional to 100 parts of AgCl:

LiCl. AgCl.
7.1885 24.3086 29.606
8.5947 29.0621 29.574

In a third experiment the LiCl was titrated with a standard solution of silver. 3.9942 grm.
LiCl balanced 10.1702 grm. Ag, equivaleht to 13.511 grm. AgCl. Hence 100 AgCl=29.563
LiCl. Mean of all three experiments, 29.581 + .0087.

Diehl,* whose paper begins with a good résumé of all the earlier determinations, describes
experiments made with lithium carbonate. This salt, which was spectroscopically pure, was
dried at 130° before weighing. It was then placed in an apparatus from which the carbon
dioxide generated by the action of pure sulphuric acid upon it could be expelled, and the loss
of weight determined. From this loss the following percentages of CO, in Li,CO, were deter-
mae 59.422 59.440

59.404 59.401
Mean, 59.417 + .006

 

1 Dissertation, University of Pennsylvania, 1896. 8 Amer. Journ. Sci., November, 1856. Chem. Gazette, 15, 7.
§ Proc. Amer. Acad., 38, 462. 1903. «Ann. Chem. Pharm., 121, 93.
102 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI

Diehl’s investigation was quickly followed by a confirmation from Troost.'' This chemist,
in an earlier paper,? had sought to fix the atomic weight of lithium by an analysis of the sulphate,
and had found a value not far from 6.5, thus confirming the results of Berzelius and of Hagen,
who had employed the same method. But Diehl showed that the BaSO, precipitated from
Li,SO, always retained traces of Li, which were recognizable by spectral analysis, and which
accounted for the error. In the later paper Troost made use of the chloride and the carbonate
of lithium, both spectroscopically pure. The carbonate was strongly ignited with pure quartz
powder, thus losing carbon dioxide, which loss was easily estimated. The subjoined results
were obtained:

Li,CO3. C0,. Per cent.
0. 970 0.577 59. 485
1. 782 1.059 59, 427

Mean, 59. 456+. 020

The lithium chloride employed by Troost was heated in a stream of dry hydrochloric acid
gas, of which the excess, after cooling, was expelled by a current of dry air. The salt was
weighed in the same tube in which the foregoing operations had been performed, and the
chlorine was then estimated as silver chloride. The usual ratio between LiCl] and 100 parts
of AgCl is given in the third column:

LiCl. AgCl. Ratio
1. 309 4.420 29. 615
2.750 9. 300 29. 570

 

Mean, 29. 5925-4. 0145

Next in order is the work of Stas,? which was executed with his usual care. In three
titrations, in which all the weights were reduced to a vacuum, the following quantities of LiCl

balanced 100 parts of pure silver:
39. 356
39, 357
39. 361

Mean, 39. 358+. 001

In a second series of experiments, intended for determining the atomic weight of nitrogen,
LiCl was converted into LiNO,. The method was that employed for a similar purpose with
the chlorides of sodium and of potassium. One hundred parts of LiCl gave of LiNO,:

162. 588

162. 600
162. 598

Mean, 162. 5953+. 0025

 

The determinations of Dittmar’ resemble those of Diehl; but the lithium carbonate
used was dehydrated by fusion in an atmosphere of carbon dioxide. The carbonate was treated
with sulphuric acid, and the CO, was collected and weighed i in an absorption apparatus, which
was tared by a similar apparatus after the method of Regnault. The following percentages
of CO, in Li,CO, were found:

59. 601 59, 517

59. 645 59. 663

59. 529—rejected 60. 143—rejected
59. 655 59, 794

59. 683 59. 584

59. 604

Mean of all, 59. 674

Rejecting the two experiments which Dittmar regards as untrustworthy, the mean of the
remaining nine becomes 59.638+.0173. This combines with the work of Diehl and Troost
as follows:

 

1 Zeit, Anal. Chem., 1, 402. 4 Oeuvres Complites, 1, 710-716.
2 Ann, Chim. Phys., 51, 108. 4 Trans. Roy. Soc. Edinburgh, 35, II, 429. 1889.
No. 8.] ATOMIC WEIGHTS—CLARKE. 103

Diehl eens ess bee lsd 8s SAMAR Se oh erat ccton Stee Sees 59. 417+. 0060
NYOOBU lake sca ada Lee pelle at Gelade Fi pele ee vg oe yeees 59. 456+. 0200
Dittmar..................- peepee dhe tees Siete alec red ad ea cee 59. 638+. 0173

General mean............0.... 2-0. ee cence ence eee 59. 442 +. 0054

The unique merit of the determinations by Richards and Willard ' is, not only that their
work was done with scrupulous accuracy, but that their ratios give simultaneous values for
the atomic weights of lithium, silver and chlorine, which are independent of all other data.
Analyses of lithium perchlorate gave directly the molecular weight of lithium chloride, with
reference to oxygen alone, and with that their other ratios are reducible. The data for the .
perchlorate are as follows, with vacuum weights:

Preliminary series.

Liclo,. Lic. Per cent LiCl.
10. 64596 4, 24171 39. 8434
12. 77683 5. 09073 39. 8435
10. 12750 4. 03587 39. 8506
13, 04021 5. 19638 39. 8489

 

Mean, 39. 8466+. 00125
Final series.

Iiclo,. ict. Per cent LiCl.
12, 79265 5. 09744 39. 8466
10. 55416 4. 20534 39. 8453
11. 39912 4, 54205 39. 8456
11. 17008 4, 45070 39. 8448
17. 84842 7.11167 39. 8448
22. 58273 8. 99846 39. 8466

 

Mean, 39. 8456+. 00023

The two series combined give a general mean of 39.8457 -+.00023. Hence LiCl = 42.393.

The two other ratios determined by Richards and Willard are those between lithium
chloride, silver chloride and silver. With vacuum weights, their data, first for the ratio
Ag: LiCl are these:

y

Preliminary series.?

Lict. Ag. Ratio.
8. 99620 22. 89013 39. 3017
5. 25395 13. 36777 39. 3030

 

. ; Mean, 39. 3023+. 00048
Final series.

ic. Ag. : Ratio.
5. 82422 14, 82035 39. 2988
6. 28662 15. 99687 39. 2991
5. 82076 14, 81122 39. 2996
6. 70863 17. 07038 39. 2998
6. 24717 , 15. 89620 39. 2998
7. 75349 19. 72977 39. 2984
7. 99108 20. 33415 39. 2988

 

: Mean, 39. 2992+. 00014
Combining the figures for this ratio we have—

BOG: ee wiessiseisieicid Oe pa itiainas Sie as HS eau a SSeS ss 53 see 39.358 +.0011

Richards and Willard 1.............-.-.00-- eee e eee eee 39. 3023+. 00048

Richards and Willard 2.................220-eeeeeeeeeece 39. 2992+. 00014
General meatts <s)c0.5e008 3 ss sccewesewely x ox eee ee 39. 30024. 00013

 

1 Journ, Amer. Chem. Soc., 32, 4. 1910. 2The material used in the preliminary series contained a trace of sodium.
104 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.
For the silver chloride ratio Richards and Willard give the following data:

Preliminary series.

Lic. AgCl. Ratio.
4.01994 13. 59125 29. 5774
6. 32840 21. 39635 29.5770
8. 99620 30. 41341 29. 5797
4. 66824 15. 78111 29. 5812
5. 43032 18, 35734 29, 5812
5. 10725 , 17. 26504 29. 5815
5. 74000 19. 40375 29. 5819
5. 42038 18. 32417 29. 5805
5. 21573 17. 63280 29. 5797
6. 56925 22, 20617 29. 5817
4, 84268 16. 37121 29. 5805

 

Mean, 29. 5802+. 00033
Final series.

LiCl. AgCl. Ratio.
6. 28662 21, 25442 29, 5779
5, 82076 19. 67875 29. 5790
6. 70863 22, 68030 29, 5791
6. 24717 21. 12073 29, 5784
5, 50051 18. 59600 29. 5790
8. 34521 28, 21438 29. 5779
6. 65987 22. 51564 29. 5788

 

oad . . ; Mean, 29. 5786+. 00014
Combining the several series for this ratio we have—

IMGIVG EH ja vosvevvers. 3 ene oo & Sa poc@rare we sumed abe bubs astpwles eared 29.581 +. 0087
WLPOORGarscint oteda teed Aaa ocens Son Aeon seer .. 29,5925+. 0145
Richards and Willard 1.............-------------2---00-- 29. 5802 +. 00033
Richards and Willard 2...............-------------+-++- 29. 5786 +. 00014
Genéral means 22-08. isteneceanescocecatenemire 29. 5789+. 00013

The older work, with its high probable errors, vanishes.
Richards and Willard, from their three final series of determinations, deduce

Ag=107. 871
Cl = 35. 454
Li= 6,939

From the final silver and silver chloride series of Richards and Willard, the ratio
Ag : Cl :: 100 : 32.8637 is derivable. This is a little lower than the value determined by Rich-
ards and Wells directly.

In a supplementary paper Richards and Cox ' discuss the purity of lithium perchlorate, as
used by Richards and Willard, and prove its availability for atomic weight determinations.

THE SILVER CARBON RATIOS.

The determination of atomic weights by the analysis of organic silver salts has been re-
peatedly attempted. The measurements of this class may, for present purposes, be conven
iently grouped together.

In 1840 Redtenbacher and Liebig? sought to determine the atomic weight of carbon, that of
silver being assumed as known, by analyses of the acetate, tartrate, racemate, and malate of
silver. There were five determinations with each compound, the salt being ignited and the
residual silver weighed. From 1 to 9 grammes of material were used in each experiment.

 

1Journ. Amer. Chem. Soc., 36, 819. 1914. 2 Ann. Chem. Pharm., 38,118. 1841. Mem.Chem. So¢.,1,9. Phil. Mag., (3) 19, 210.
No. 3.] ATOMIC WEIGHTS—CLARKE.. 105

In the acetate the following percentages of silver were found:

64. 615 64, 614
64. 624 64. 610
64, 623

Mean, 64. 6172 +. 0018

After applying corrections for weighing in air, this mean becomes 64.6006.
In the tartrate the silver was as follows:

59. 297 59. 293
59. 299 59. 293
59. 287

Mean, 59. 2938 +. 0014
Or, reduced to a vacuum, 59. 2806.

In the racemate we have:

59. 290 59. 283
59, 292 59. 284
59. 287

Mean, 59. 2872, +. 0012
: Or, corrected, 59. 2758.
And from the malate:

61. 996 62. 059
61. 972 62.011
62. 015

Mean, 62.0106 +. 0096
Or, corrected, 62. 0016.

These results are by no means unimpeachable. They involve two possible sources of
constant error, namely, impurity of material and the volatility of the silver. These objections
have both been raised by Stas, who found that the silver tartrate, prepared as Redtenbacher
and Liebig prepared it, always carried traces of the nitrate, and that he, by the ignition of that
salt, could not get results at all agreeing with theirs. In the case of the acetate a similar im-
purity would lower the percentage of silver, and thus both sources of error would reinforce each
other and make the atomic weight of carbon apparently too high. With the three other salts
the two sources of error act in opposite directions, although the volatility of the silver is proba-
bly of much greater influence than the impurity. Even if we had no other data relating to the
atomic weight of carbon, it would be clear from these facts that the results obtained by Red-
tenbacher and Liebig must be decidedly in excess of the true figure.

Strecker,1 however, discussed the data given by Redtenbacher and Liebig by the method
of least squares, using the Berzelian scale, and assuming H=12.51. Thus treated, they gave
C=75.415, and Ag=1348.79; or, with O=16, C=12.066 and Ag=107.903. These values, of
course, would change somewhat upon adoption of the modern ratio between O and H.

Observations upon silver acetate, like those of Redtenbacher and Liebig, were also made by
Marignac.? The salt was prepared by dissolving silver carbonate in acetic acid and repeatedly
recrystallizing. Two experiments gave as follows:

Acetate. Ag. Per cent.
3. 3359 2.1561 64. 633
3. 0527 1. 9727 64. 621

 

Mean, 64. 627+. 0040
Reduced to a vacuum, this becomes 64.609.
In a second series, conducted with special precautions to avoid mechanical loss by spurting,

Marignac found:

Acetate. Ag. Per cent.
24. 717 15. 983 64. 665
21. 202 13.709 64. 661
31. 734 20. 521 64. 666

 

Mean, 64. 664+. 0010

 

1 Ann. Chem. Pharm., 59, 280. 1846. 3 Ann. Chem. Pharm., 59, 287. 1846. Oeuvres Completes, 1, 184.
106 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

Other experiments, comparable with the preceding series, have more recently been pub-
lished by Hardin, who sought to redetermine the atomic weight of silver. Silver acetate and
silver benzoate, carefully purified, were subjected to electrolysis in a platinum dish, and the
percentage of silver so determined. For the acetate, using vacuum weights, he gives the fol-
lowing data, the percentage column being added by myself:

Acetate. Ag. Per cent.
0. 32470 0. 20987 64. 635
. 40566 . 26223 64. 643
. 52736 . 34086 64. 635
. 60300 . 88976 64. 637
. 67235 . 43455 64, 631
- 72452 . 46830 64. 636
. 78232 . 50563 64. 632
. 79804 . 51590 64. 646
. 92101 . 59532 64. 638
1.02495 - 66250 64. 637

 

Mean, 64. 687, +.0011

Combining this series with those of the earlier investigators we have:

Redtenbacher and Liebig.............-.------------ee- 64. 6006+. 0018
Marignac, Ist series............-. 2-5 020s eee e ese e nese eee 64.609 +. 0040
Marignac, Id/series....-:..s..20+ 2. s22- cae snc serbunenes 64.664 +.0010
Hardliinnandataeeaiancidsed ceeds ses ieacatas teeta 64. 637 +.0011

General mean ...........0-.-- 22-2 e eee eee eee eee 64, 64344. 0007

With silver benzoate, C,H,AgO,, Hardin’s results are as follows:

Benaoate. Ag. Per cent.

0. 40858 0. 19255 47,127
. 46674 . 21999 47.133
. 48419 . 22815 47.120 3
. 62432 . 29418 47.120
. 66496 . 31340 47.131
. 75853 . 35745 47,124
. 76918 . 36247 47,124
81254 . 38286 47.119
. 95673 . 45079 47.118

1. 00840 . 47526 47.1380

 

Mean, 47. 125, +0012

A different method of dealing with organic silver salts was adopted by Maumené,? in 1846,
for the purpose of establishing by reference to carbon the atomic weight of silver. He effected
the combustion of the acetate and the oxalate of silver, and, by weighing both the residual metal
and the carbon dioxide formed, he fixed the ratio between these two substances. In the case of
the acetate his weighings show that for every gramme of metallic silver the weights of CO, were
produced which are shown in the third column:

Ag. C0).
8, 083 6. 585 0.8147
11. 215 9, 135 8186
14, 351 11. 6935 8148
9.030 7. 358 8148
20. 227 16. 475 . 8145

 

Mean, . 81448

The oxalate of silver, ignited by itself, decomposes too violently to give good results; and
for this reason it was not used by Redtenbacher and Liebig. Maumené, however, found that
when the salt was mixed with sand the combustion could be tranquilly effected. The oxalate

 

1 Journ, Amer. Chom, Soc., 18, 990. 1896. 2 Amn, Chim. Phys, (8), 18, 41. 1846,
No. 3.] ATOMIC WEIGHTS—CLARKE. 107

employed, however, with the exception of the sample represented in the last experiment of the
series, contained traces of nitrate, so that these results involve slight errors. For each gramme
of silver the appended weights of CO, were obtained:

Ag. C0,.
14. 299 5. 835 0, 4081
17. 754 7.217 - 4059
11. 550 4.703 - 4072
10. 771 4. 387 - 4073
8. 674 3. 533 - 4073
11. 4355 4. 658 . 4073

 

Mean, . 40718

Now, one of these salts being formed by a dibasic and the other by a monobasic acid, it is
well to reduce both to a common standard. Doing this, we have for the ratio between carbon
dioxide and 100 parts of silver the following combination:

From the acetate... 2.2.2.2... eee cee eee ee cee eee eee eneee 40. 724+. 0076
From theoxalates 1502 oe asipaweidictas oi cide sclndosieeielees oe. 40. 718 +. 0185
General meatls «0s ¢ sciverteetes «eeu teaeeeesces ss 40. 723 +. 0071

That is, Ag : CO, : : 100 : 40.723 4.0071.

The experiments of Dean‘ on silver cyanide, may be conveniently summarized here,
although they involve nitrogen as well as carbon. Dean’s object was to determine the atomic
weight of nitrogen, the values for silver and carbon being supposedly known. The cyanide
was dissolved in nitric acid, or, in the last experiment in sulphuric acid, and its content of silver
was determined by titration with a standard solution of potassium bromide. The silver equiv-
alent of the latter compound was previously fixed by titration against a definite solution of silver.
The weights obtained, corrected to a vacuum, are subjoined, together with a column giving the
percentages of silver:

AgCN. Ag. Per cent Ag.
6. 2671 5. 0490 80. 564
17. 60585 14, 18496 80. 570
17. 1049 13. 7801 80. 561
17. 9210 14. 43881 80. 569
12. 11215 9. 75875 80. 570
14, 6672 11. 81727 80. 569

 

Mean, 80. 567+. 0010

Still another pair of ratios, involving bromine, were measured by Scott. Tetraethylam-
monium bromide, purified with great care, was titrated with silver solutions of known strength.
The results obtained, with vacuum weights, were as follows:

(C,H;),NBr. Ag. Ratio.
5. 07039 2. 60146 194, 906
5. 26380 2. 70142 194. 853
7. 10662 3. 64683 194, 876
6. 79951 3. 48976 194, 842
2, 72225 1.39695 194. 871
6. 24530 3. 20481 194, 873
5. 74581 2. 94853 194. 870
5. 21663 2, 67699 194, 869

 

Mean, 194. 870+. 0045

A single experiment with the corresponding tetramethyl compound was also made. 8.64585
grammes of (CH,), NBr are equivalent to 6.05348 of silver. Ratio, 142.824+.0123, when the
probable error is assumed equal to that of one experiment in the ethyl series. From these
figures Scott deduces a value for the atomic wetgae of carbon much higher than that given by
the direct O: C ratio.

1 Journ. Chem. Soc., 77,117. 1900. : 2 Journ. Chem. Soc., 95, 1200. 1909,

 
108 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

In a criticism of Scott’s work, Thorpe! has pointed out the possibility of errors due to the
vacuum reductions; errors discussed long ago by Marignac, and recently, in more detail, by
Guye and Zachariades.? The substances analyzed were weighed in powder, under which con-
ditions they are liable to condense and occlude air. A probable correction, applied to Scott’s
weighings, reduced the atomic weight of carbon to 12.008, in harmony with other good determi-
nations. To this criticism Scott * published a rejoinder, seeking to show, on the basis of experi-
mental evidence, that the supposed errors do not, in fact, exist. According to Guye and Zach-
ariades, the errors noted by them in the study of 26 compounds may amount to as much as, or
even more than, 3 parts in 10,000.

The determinations of the atomic weight of carbon by Richards and Hoover‘ are quite
unlike any of the previous investigations. Their starting point was sodium carbonate, which
was first converted into sodium bromide by a standardized solution of hydrobromic acid. The
ratios between the sodium bromide, silver, and silver bromide were then determined by the
usual methods. The first table following gives the ratio 2AgBr: Na,CO,::100:x. The hydro-
bromic acid was measured by weight, 100 grammes being equivalent to 8.31135 grammes of
silver bromide. The ratio stated in the third column was computed by me from the data
given by Richards and Hoover. The weights, of course, refer to a vacuum.

Wt. Na,CO3. Wt. HBr solution. AgBr ratio.
4. 75555 202. 744 28. 2211
4. 80081 204. 673 28. 2217
4. 88936 208. 457 28. 2204
5. 63157 240. 119 28. 2184
4, 49516 191. 646 28. 2210
4, 86256 207. 307 28, 2211

 

Mean, 28. 22064: .00032

The second table gives the ratio 2 Ag:Na,CO,: : 100:x, as determined from the amount of
silver used in precipitating the sodium bromide.

Wt. Na,COz. Wt. Ag. Ratio.
4, 75555 9. 68023 49. 1264
4, 80081 9. 77222 49. 1271
4, 88936 9. 95301 49, 1244
5. 63157 11. 46316 49, 1276
4, 49516 9. 15003 49, 1273
4, 86276 9. 89811 49. 1262

 

Mean, 49. 1265+ .00032

From these two ratios the cross ratio Ag:Br: :100:74.081 directly follows. That ratio has
already been used in a previous section of this work.

Since silver tartrate and silver racemate are isomeric compounds, their figures may be
consolidated into one series. We then have the following ratios in this group, to be discussed
in connection with other ratios later:

Ag(C,H,O,Ag:: 100: 64.843 + .0007
Ag,C,H,0,: 2Ag:: 100: 59.2778 + .0009
Ag,C,H,O,: 2Ag:: 100: 62.0016 + .0096
AgC,H,O,: Ag:: 100: 47.125 + .0012
Ag: CO,:: 100: 40.723 + .0071
AgCN: Ag:: 100: 80.567 + .0010
Ag: (C,H;) ,.NBr:: 100: 194.870 + .0045
Ag: (CH,),NBr:: 100: 142.824 + .0123
2Ag: Na,CO,:: 100: 49.1265 + .00032
2AgBr: Na,CO,:: 100: 28.2206 + .00032
1 Proc. Chem. Soe., 25, 285. °
2 Compt. Rend., 149, pp. 593, 1122.

3 Proc, Chem. Soc., 25, 286. See also Richards and Baxter, Journ. Amer. Chem. Soc., 32, 507. Renard and Guye, Journ. Chom. Phys., 14, 55.
4Journ. Amer. Chem. Soc., 37,95. 1915. : :

 
No. 3.3 ATOMIC WEIGHTS—CLARKE. 109

THE SULPHUR RATIOS.

The atomic weight of sulphur has been determined by means of several ratios connecting
it with silver, chlorine, oxygen, hydrogen, nitrogen, sodium, and carbon. Other ratios have
also been measured, but they are hardly available here. The earlier results of Berzelius were
wholly inaccurate, and his later experiments upon the synthesis of lead sulphate will be used
in discussing the atomic weight of lead. Erdmann and Marchand determined the amount of
calcium sulphate which could be formed from a known weight of pure Iceland spar; and later
they made analyses of cinnabar, in order to fix the value of sulphur by reference to calcium and
to mercury. Their results will be applied in this discussion toward ascertaining the atomic
weights of the metals just named.

First in order let us take up the composition of silver sulphide, as directly determined
by Dumas, Stas and Cooke. Dumas’? experiments were made with sulphur which had been
thrice distilled and twice crystallized from carbon disulphide. A known weight of silver was
heated in a tube in the vapor of the sulphur, the excess of the latter was distilled away in a
current of carbon dioxide, and the resulting silver sulphide was weighed.

I subjoin Dumas’ weighings, and also the quantity of Ag,S proportional to 100 parts of
Ag, as deduced from them:

Weight Ag. Weight S. Ratio.
9. 9393 1.473 114, 820
9. 962 1. 4755 114, 811

30. 637 4, 546 114. 838

‘80. 936 4. 586 114. 824

30. 720 4. 554 114. 824

Mean, 114. 8234+ .0029

Dumas used from 10 to 30 grammes of silver in each experiment. Stas,? however, in
his work employed much larger quantities. Three of Stas’ determinations were made by Dumas’
method, while in the other two the sulphur was replaced by pure sulphureted hydrogen. In
all cases the-excess of sulphur was expelled by carbon dioxide, purified with scrupulous care.
Impurities in the dioxide may cause serious error. The data are as follows, with vacuum

weights: Weight Ag. Weight Ag,S. Ratio.
59, 4225 68, 24823 114, 854

104, 139 119. 6078 114, 853

191. 9094 220, 4158 114. 854

150. 000 172. 2765 114. 851

249. 076 286. 061 114. 849
: Mean, 114. 8522+.0007

The experiments made by Cooke with reference to this ratio were only incidental to
his elaborate researches upon the atomic weight of antimony. They are interesting, however,
for two reasons: They serve to illustrate the volatility of silver, and they represent, not
syntheses, but reductions of the sulphide by hydrogen. Cooke gives three series of results.
In the first the silver sulphide was long heated to full redness in a current of hydrogen. Highly
concordant and at the same time plainly erroneous figures were obtained, the error being
eventually traced to the fact that some of the reduced silver, although not heated to its melting
point, was actually volatilized and lost. The second series, from reductions at low redness,
are decidedly better. In the third series the sulphide was fully reduced below a visible red
heat. Rejecting the first series, we have from Cooke’s figures in the other two the subjoined
quantities of sulphide corresponding to 100 parts of silver:

AgS. 8. Ratio.

_ 7.5411) 0. 9773 114. 889
5. 0364 . 6524 114. 882
2.5815 3345 114. 886
2. 6130 . 3387 114. 892
2. 5724 3334 114. 891

Mean, 114. 888+ .0012
1 Am. Chem. Pharm., 113, 24. 1860, 2 Oeuvres Complétes, 1, 349. Proc. Amer. Acad., 13, 47-52. 1877.

 
110 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.
1.1357 0. 1465 114. 810
1. 2936 1670 114. 823

Now, combining all four series, we have—

Mean, 114. 8165-+.0044

WD iiMas scsi Gi na ttainglnead setts aca 114. 8234+. 0029
BEE iS A cee Conch sug cares lay a aroaatheatdad 114, 8522-4. 0007
Cooke's:2dsuiui aves ou yess baw s sass sede vs eee 114, 888 +. 0012
GookelsiS dna: scassesecceesy sess deed dade atone een 114, 8165+. 0044

ge csoees peeeu peek ea eee see ese sees 114. 8581+. 0006

The percentage of silver in silver sulphate has been determined by Struve, by Stas, and by
Scheuer. Struve! reduced the sulphate by heating in a current of hydrogen, and obtained

these results:
Ag,S0,. Ag. Per cent Ag.
5. 1860 3. 5910 69. 244
6. 0543 4.1922 69. 243
8. 6465 5. 9858 69. 228
11. 6460 8. 0608 69. 215
9. 1090 6, 3045 69. 212
9. 0669 6. 2778 69. 239

Mean, 69. 230+ .004

Stas,’ working py essentially the same method, obtained the following figures, which imply

vacuum weights:

Ag,S0,. Ag. Per cent. Ag.
72.137 49. 919 69. 200
60. 251 41, 692 69. 197
81. 023 56. 071 69. 204
33. 115 57, 523 69. 209
55. 716 38. 5595 69. 207
63. 922 44, 2355 69. 202

The third sulphur ratio to be considered is one of minor importance.

 

Mean, 69. 203+ .0012

When silver chloride

is heated in a current of sulphuretted hydrogen the sulphide is formed. This reaction was
applied by Berzelius* to determining the atomic weight of sulphur. He gives the results of
‘four experiments; but the fourth varies so widely from the others that I have rejected it. I
have reason to believe that the variation is due, not to error in experiment, but to error in
printing; nevertheless, as I am unable to discover the cause of the mistake, I must exclude the
figures from our discussion.

The three available experiments, however, give the following results. The last column
contains the ratio of silver sulphide to 100 parts of chloride:

AgCl. Ag,S. Ratio.
6. 6075 5. 715 86. 478
9. 2323 7. 98325 86. 471
10.1775 8. 80075 86. 472

 

Mean, 86. 4737 +. 0015

We have also a single determination of this value by Svanberg and Struve.t After con-
verting,the chloride into sulphide they dissolved the latter in nitric acid. A trifing residue of
chloride, which had been inclosed in sulphide, and so protected against change, was left undis-
solved. Hence a slight constant error probably affects this whole ratio. The experiment of
Svanberg and Struve gave 86.472 per cent of silver sulphide derived from 100 of chloride.

 

' Ann. Chem. Pharm., 80, 203. 1851,
2 Oeuvres Completes, 1, 410.

3 Berzelius’s Lehrbuch, 5 Aufi., 3, 1187.
4 Journ. prakt. Chem., 44, 320. 1848.
No. 3.] ATOMIC WEIGHTS—CLARKE. 111

If we assign this figure equal weight with the results of Berzelius, and combine, we get a general
mean of 86.4733. +0011.

The work done by Richards! relative to the atomic weight of sulphur is of a different
order from any of the preceding determinations. Sodium carbonate was converted into sodium
sulphate, fixing the ratio Na,CO,:Na,SO,::100:2. The data are as follows, with vacuum
weights:

Na,CO3. Na,SO,. Ratio.

1. 29930 1.74113 134. 005

3. 18620 4, 26790 133. 950
1.01750 1. 36330 133. 985
2.07680 2. 78260 183. 985

1, 22427 . 1. 63994 183. 952

1. 77953 2. 38465 134. 005

2. 04412 2. 73920 134. 004

3. 06140 4, 10220 133. 997

‘ Mean, 133. 985-4. 0055

The exact value of this ratio was subject to slight uncertainties due to a trace of silica in
the sodium carbonate, and doubt as to its complete dehydration. Accordingly a new series of
determinations has recently been made by Richards and Hoover,? in order to secure greater
precision. The new data are as follows:

Na,CO3. Na,S0,. Ratio.

5. 25191 7. 03829 134. 014
4.50977 : 6. 04382 134. 016
5. 04233 6. 75737 134, 016
3. 67340 4. 02304 134. 019
4.18724 5. 61151 134. 015
4. 55100 6. 09916 134. 018

Mean, 134. 0155 +. 00065.

The general mean of this series, combined with the earlier one, is 134.015+.00065. That
is, the old series practically vanishes.’

Another method for fixing the atomic weight of sulphur was adopted by Richards and
Jones.‘ Silver sulphate was converted into chloride by heating in a current of pure, dry hydro-
chloric acid gas. The data obtained, with vacuum weights, were as follows:

AgS0,. AgCl. Per cent AgCl.
5. 21962 4. 79859 91. 934
5. 27924 4. 85330 91. 932
5. 08853, 4. 67810 91. 934
5. 36381 4.93118 * 91. 934
5. 16313 4, 74668 91. 934
‘5. 08383 4. 67374 91. 933
5. 138372 4, 71946 91. 931
5. 16148 4. 74490 91. 929
5.19919 4.77992 91. 936
5. 374386 4, 94088 91. 934

 

Mean, 91. 933+. 0004

The work done by Burt and Usher‘ on the relative atomic weights of nitrogen and sulphur is
in its methods entirely unlike any previous investigation. Nitrogen sulphide, N,S, was decom-
posed by sublimation over red hot quartz wool in a quartz tube, and the liberated nitrogen was
collected and measured. From its volume, assuming the weight of a liter of nitrogen to be
1.2514 grammes at 0° and 760 mm. pressure, the weight of the nitrogen was computed. The

 

1 Proc. Amer. Acad., 26,268. 1891. Incidental to work on the atomic weight of copper.

® Journ. Amer. Chem. Soc., 37, 108. 1915.

8 For a discussion of Richards and Hoover’s sodium carbonate ratios, see Moles, Journ. Chim. Phys., 15, 51. 1917.

¢ Publ. Carnegie Inst. Washington, No. 69, p. 69. 1907. The authors give a thorough summary and criticism of the previous work on sulphur.
§ Proc. Roy. Soc., 85 A, 82. 1911. :

161566°—20—_8
114 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Hence, using Guye’s value for L, which is sensibly identical with that of Baume, we have—

By Guye’s critical data..... 22.2.2. cece eee cence eee cee eeee S0,=64.068
By Baume’s critical data. ...:....-..22.--5. Se icnumseayeee SO,=63.955

The difference between these figures shows the uncertainty of the method as applied to
sulphur dioxide If we accept Guye’s figures, as yielding results more nearly in harmony with
the chemical methods of determination, the general mean for sulphur dioxide gives SO,=
64.068, +.00067, and S=32.068 +.00067.

Another value for the atomic weight of sulphur is derivable from the density of hydrogen
sulphide, as determined by Baume and Perrot.' Their crude values for the weight of a liter
of the gas are as follows:

1. 53934 1. 53860
1. 54126 1. 53943
1. 53843 1. 53900
1. 53862 1. 53917
1. 53789 1. 53921
1. 53843 1. 53960
1, 53798 1. 53938
1. 538890 1. 53964
1. 53929 1, 54069

Mean, 1.53916 + .00013.

Corrected to the usual standards, the weight of the normal liter becomes 1.5392 grammes.
With the critical constants determined by Olzewski, a,=.01438, and 6,=.00240. Hence the
molecular weight of H,S is 34.0893, and S=32.0737+.0030. This, combined with the value
deduced from the density of sulphur dioxide, gives a general mean of S=32.0683 + .00064.

GENERAL DISCUSSION.

There are now before us, as developed in the preceding pages, 71 ratios, from which the
atomic weights, of 11 elements are to be computed. These elements are hydrogen, silver,
chlorine, bromine, iodine, nitrogen, carbon, sulphur, lithium, sodium, and potassium. The first 12
‘‘ratios” are really positive values, referred to O=16, which can be regarded as first approxi-
mations to the true quantities. These values are applicable to the reduction of the remaining
ratios, by which they are themselves to be adjusted in turn.

The rigorous method of dealing with such a mass of data is well understood.? The several
ratios should be transformed into linear equations, and each one weighted inversely as the square
of its “probable error.” The 71 equations should then be combined into 11 normal equations,
which, when solved, would give the 11 atomic weights now under consideration. But that
method of reduction is exceedingly laborious, and would possibly be premature. There is
great activity at present in the measurement of fundamental ratios, and for that reason the
rigorous discussion of them may well be deferred. There is, moreover, one practical disad-
vantage in it, namely, that the specific influence of each individual ratio is more or less obscured,
except to the computer himself. The extent to which a given ratio affects the final results is
not readily seen in a general combination of all the data, whereas for present purposes some such
insight is likely to be helpful in guiding future work. An approximate method of reduction is
therefore adopted here, which will give highly probable values for the several atomic weights,
even if it does not yield the ‘‘most probable values” of the method of least squares. The
uncertainties will not be large, and perhaps no larger in reality than if the strict mathematical
procedure were followed rigorously.

The 71 ratios may now be tabulated, and numbered for reference, as follows:

(1) H=1.0078+ .00001. (4) S=32.0683 + .00064.
(2) C=12,0000-+ .00026. (5) Cl=35.4598 + .00025.
(3) N=14.0076 + .00012. (6) Br=79,.9251-+ .0013.

 

1 Journ. Chim. Phys., 6, 610. 1908. Baume and Perrot reduce their data with the aid of the constant 22.410, instead of the 22.412 adopted here.
Leduc’s single determination of the density (Ann. Chim. Phys. (7), 15, 35) may be neglected. His gas was not certainly pure.
2 See Clarke, Amer, Chem. Journ. 27, 321, 1902,
No. 3.] ATOMIC WEIGHTS—CLARKE. 115

(7) [=126.915+ .0056.

(8) LiCl=42.393 + .00029.

(9) NaCl=58.500+ .0048.
(10) KC1=74.5631 + .0004.
(11) KBr=119.249+ .0596.
(12) KI=165.590+ .0384.
(18) AgCl=143.390 + .0060.
(14) AgBr=187.884+ .0133.
(15) AgIl=234.734+ .0126.
(16) 1,0, : 2Ag :: 100 : 64.6229+..0001.
(17) Ag: Cl :: 100 : 32.8656 + .00022.
(18) Ag: Br :: 100 : 74.0800+ .00022.
(19) Ag: I :: 100 : 117.6424+ .00026.
(20) AgCl : AgBr :: 100 : 131.0172+ .00012.
(21) AgCl : AgI :: 100 : 163.8118+ .00038.
(22) AgI : AgBr :: 100 : 79.9799+ .00028.
(23) Ag : LiCl :: 100 : 39.3002+ .00013.
(24) AgCl : LiCl :: 100 : 29.5789 —.00013.
(25) Ag : NaCl :: 100 : 54.1995—.00015.
(26) AgCl : NaCl :: 100 : 40.7803-+-.00028.
(27) Ag: NaBr :: 100 : 95.4405+ .0007.
(28) AgBr : NaBr :: 100 : 54.8010+ .0005.
(29) Ag : KCI :: 100 : 69.11388+.00011.
(30) AgCl : KCl :: 100 : 52.01634.00018.
(31) Ag: KBr :: 100 : 110.3193 + .00033.
(32) AgBr : KBr :: 100 : 63.3727+.0008.
(33) Ag : KI :: 100 : 153.800+-.0178.
(34) Ag : NO, :: 100 : 57.479+ .000095.
(35) AgNO, : AgCl :: 100 : 84.3743+ .0025.
(36) AgNO, : KCl :: 100 : 43.8759 + .00046.
(87) AgNO, : NH,Cl :: 100 : 31.488+ .0006.
(88) Ag : NH,Cl :: 100 : 49.5965 .00038.
(39) AgCl : NH,Cl :: 100 : 37.3205+ .0002.

 

(40) Ag : NH,Br :: 100 : 90.8175+: .00065.
(41) NH, : HCl :: 100 : 214.136+ .0092.

(42) Cl: N :: 100 : 39.489+ .0033.

(43) Cl: NO :: 100 : 84.626+.0013.

(44) N-+Cl=49.457-+.0026.

(45) Br : Cl :: 100 : 44.3674+.0006.

(46) NaCl : Cl :: 100 : 60.6581 + .00035.

(47) NaBr : Br :: 100 : 77.6543+ .00029.

(48) NaCl : NaNO, :: 100 : 145.418+.0012.
(49) NaClO, : NaNO, :: 100 : 79.8823-+.0029.
(50) KCl: KNO, :: 100 : 135.636 + .0007.
(51) KC1O, : KNO, :: 100 : 82.500+.0012.
(52) N,O, : K,O :: 100 : 87.232+.0017.

(53) LiCl : LiNO, :: 100 : 162.5953 .0025.
(54) Li,CO,; : CO, :: 100 : 59.442+ .0054.

(55) 2Ag : Na,CO, :: 100 : 49.1265+ .00032.
(56) 2AgBr : Na,CO, :: 100 : 28.2206-+.00032.
(57) Ag : CO, :: 100 : 40.7234 .0071.

(58) AgC,H,O, : Ag :: 100 : 64.6434+ .0007.
(59) Ag,C,H,O, : 2Ag :: 100 : 59.2778+.0009.
(60) Ag,C,H,O, : 2Ag :: 100 : 62.0016+ .0096.
(61) AgC,H,O, : Ag :: 100 : 47.125+.0012.
(62) AgON : Ag :: 100 : 80.567-.0010.

(63) Ag : CgH,)NBr :: 100 : 194.870+.0045.
(64) Ag: C,H,,NBr :: 100 : 142.824+ .0123.
(65) 2Ag: S :: 100 : 14.8581 .0006.

(66) 2Ag : SO, :: 100 : 29.692+.0007.

(67) Ag,SO, : 2Ag :: 100 : 69.1932+ .0002.
(68) Ag.SO, : 2AgCl :: 100 : 91.934+ .0008.
(69) 2AgCl : Ag,S :: 100 : 86.4733+.0011.
(70) Na,CO, : NasSO, :: 100 : 134.015-+ .00065.
(71) 8: N :: 100 : 43.687+ .0006.

In the discussion of these 71 ratios itis evidently desirable to utilize each one as completely as
possible. That is, each ratio should contribute to the determination of every element whose
atomic weight is involved init. Of these atomic weights that of silver is especially important,
for it appears in 43 of the ratios, and is, moreover employed in determining the atomic weights
of many other elements. In order to do this, however, it is necessary to find preliminary values
for lithium, sodium, and potassium, and these can be computed from several ratios in which
silver does not appear. For sodium, for example, we have—

From ratios—

Biatid Abb s socewsrenesaceaw ceca edehoreeckastee
Gian Bris nce ones hd deveeistrasd athancpeitte so 5
era's agence a set 22.9924+. 0017
eieeeacoubasmaaes ses 23.1736+ .0050
ot Ae aera cnsreheuste ate 23.0182+ .0014

big restas eee Na=22.9984+ .00025

3, 5, and 49.......... umes eie
2,4, and 70..............-----

General mean..........

For potassium there are the following values:

From ratios— —

B:ON0 Ose cen ceeie ees sacieees

General mean. .......-.

w-e-e----Na=23.0402+0.0048
22.9987 .0003
22.9990+ .0005

names K=30.1033 0.0009
iltevarewetoacste. SOSH 2008
tem ener aecesee 39,1240 .0085
suid sila eases 39.1105: .0010

Skate ea K=39.0927+ .00075

Values for potassium can also be derived from ratios 10 and 11, but their probable errors
are so large that they would not appreciably influence the gerieral mean.
118

mine.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

From ratios—

2; 72, 66:80 AB visia so'5 sig sig siyicilareiainig weiss sinieinie e's Br=79.9204+ .0027

 

 

20, Cl, and Ag 79.9205 + .00035
MOG AT psa ccd eae ee ep ara ta ie ae 79.9225 + .0014
78, 31, and Ag 79.9234 + .0026
45 and Cl... ....2...22200---- ie ecen scales 79.9247 + .00024
Gipsenaoedc cuss oss suee seen Labamene yennesadte 79.9251 + .0013
1, 3, 40, and Ag...... 22.0.0 o cee e cece eee eee eee eee 79.9351 + .00085
9,27, and Ag.......c.cc02 eee eee Eaaes eae Goucie 79.9626 + .00081
1, 2, 3, 64, and Ag...........0.22.... Tisha ebneete 80s 79.9771 +. .0166
a Sle aule weeeee  80.0040+.0134
WS amid) Ws eA toc tat ale Ae at fae a airs 80.1563 -+ .0596
T, 2,-8,,68, and Age oo wesc. cecesxwe bierdesawenceaws 80.1621 + .0166
General mean.........0..--0--- eee eee ee eee Br=79,9228 + .00015

[VoL. XVI,

The general mean agrees well with the best determined values for the atomic weight of bro-

The last four values, with their higher probable errors, are without significance.

Iodine appears in only eight of the ratios, from which the following values are derived:

From ratios—

TOON 2 a adie mhaeenaeinn dhe 6a SmieeGieweuurtes ks T==126.497 + .0385
135005 MOG AB sieve secu amarisseiciieia ose ob/ob Sercicwin 126.826 + .0192
AB aNd ACY. ccc Sectecacieise cee aici piaieeieeietioete a oes 126.854+ .0127
OVATE A Gis oor cticje wlaireic be ehatene Shas Sat helage eaeeateyavettarsotere 126.913 + .00034
WU dsc eimaeodpebedcicmiicirseitinsltolssalettelles's Sa eaaraee 126.915 + .0056
21 Cl Bnd Ags. ins. wacctestoecceeennbekeeeeceee 126.929+ .00067
D2, SEs ANON Gs, lores eh verse toe ate ata a cllera 126.932 + .00098
Giand Ages. coves sane te pesados efesinewewe meses 126.938 + .00036
General mean..................22202-208 ....-1=26.926+ .00023

The three low values, with their large probable errors, have only slight influence upon this
combination. At most they effect only the fourth decimal place.
Nitrogen appears in 22 of the ratios from which the following values are derived:

From ratios—

79 ONG 192 sch a0 vaccines WAM cee saarscsciele N=13.9928+ .0012
BABU. Cl xo cata eceateauieea ee ame ee kes Sale teas 13.9965 + .0027
74;:68, 80d Clics civcune nanecasakeaaknniece eave ee 13.9976 + .0013
1,37, Ol, and Ago. cvesesaseckcicsecyeneieeh aces 14.0002 + .0015
42 Gnd (Cl. cae er ganscries andere cetera obe6 14.0030 + .0012
1,439; Ol eaind AG. osock. ao eon sw seabeencaeens 14. 0037 +00032
Vc4 Vy Andy Ci esc os so aciieisnemnmeven's's x wes 4 acres 14.0070 + .00074
Dieiseccicde une ene dae Wee ch anemeeemens ses 14.0076 + .00012
8b, Chand Ap. cee sonnei uee eens soceeeeerecesieas 14.0078 + .0078
S4 and Agiaws cvs se seca sawmensiescrse ves eeeaec ess 14.0083 + .00014
AS and Clewssveas. cas anteereees ccceedweciees Tie ais 14.0088 + .00047
AGUNG 7 des hie se wo Sed ac anierow cet caise Seles oe 14.0097 + .00034
72; 48 and Clivetciccaanchs nicosed s a.daiciGiccaededes 14.0114+ .00085 .
1:38,Cl, and Age. a2 ceascusseisanne onieineaecens 14.0130+ .00043
73,01, O00. Clon scccnmeetannnenassawaiemnis Ves 14.0137 +.0018
1,40. Brand AG ciccsecce acted anmcicliiten sess oe 14.0199+ .00074
2, 62, and Ag......... pysooetaceeeemeney oes co's ae 14.0210+ .0014
L204; and Ae ccd: ev sasslsadsteciensnaiad cadens 14.0266 + .0148
73, 50, and Cl............. BAS Bees Wame enema 14.0283 + .0014
A43,36,:Cl, and Ag sco. cwcsuetereuts aneerkmacceees 14.0383 +- .0025
72, 48, and Cl. scaveesssccence ane Betas bliss eeheds 14.0484+. .0082
4,2, 63, and Agicecs dee scBiangecsdshedanvderces ' 14,1469+ .0212
General mean....-. 2.2... .2. cece eee cence N=14.0081 + .00008

In this combination the value found for the atomic weight of nitrogen lies between that

directly determined and represented by ratio 3, and that. derived from the composition of
silver nitrate as given by ratio 34.. All the other values might be rejected and the final result’
would be the same as that given in the foregoing table.
No. 3.] ATOMIC WEIGHTS—CLARKE. 119

The atomic weight of potassium is deducible from 12 ratios, as follows:

From ratios—

12 and. [ee ce Sacd hahaa eaee ewnaeeeaesesiuce K=88.6440 + .0384
50, N, and Cl....... BN Eat aa en oh Nae ee 39.0361+.0015
36, N, Cl, and Ag.......-...2.-.- Sat eedlaedite 39.0794 +.00090
Ol, BrpAM eA G ss cirecteaepaciasies ws Casacneceers.e€ 39.0896 + .00042
69 Vir and ered fat ene tea ce 39.0929 + .00026
8 A A goin aca ncecorenc mee vencseus 39.0931+.0181
29, (Cl ;-etid AG. cee ead ek occus os ge bet dehonniece 39.0995 + .00020
30; Cl, and Noah tees hemseeds i Maas su bnwions 39.0999 + .00020
l0cand: Cliece ses sccncecs bigitg s Sas avecteaded pistes es oe eg 39.1026 + .00042
52 ANC IN ooescved sc cestdeaeds deeawweee si ovavasaveve-a 39.1123 + .00092
ls N cand Clic actc ite ae scistececeacs uacasemewis 39.1246 + .0085
Dan) Brssew tos Gisiiet ow teitiss oe eadiaasah eiidele'v's 39.3262 + .0596
General mean............. nd oata eiamatnerm airs K=89.0975+ .00011

Here again, the extreme values, with their high probable errors, count for nothing. The
final value is determined by the ratios which represent analyses of potassium chloride and bro-
mide. The preliminary value for K (ratio 73) was much lower.

The atomic weight of sodium is derivable from 12 ratios, as follows:

From ratios—

BS ee eh) ts) ths ete gar Na=22.9913 +.0016
DG Cl, Wad AO. ci cunahniess saanvuudwneredeonk 22.9942 + 00042
28, Br, and Ag... cc scsceseesens seceecceeew eas Aah 22.9952 + .00096
DBs A MOG dctetaadasdtr tudor neaGea ons on 22.9976 + .00038
AT aNd Bioees cel itetceies «shee. gtdapieeashw veises 5 22.9985 + .00032
46 and Clo eacevs OPA Cea esta nes ReeRE ees XSe 22.9989 + .00025
2;.56, Br, and Agiescecsssessa scsaaeendexews ees 22.9990 + .00062
Q5 CU and AG 2c ge teen ceed aut nyneeeeelee se te 23.0099 + .00022
Oe Anand 7Obois sk anscueac eee a wee aeacnee nae 23.0182 +.0014
27, BY, BHT AP 2s iccopscevessianoancecanees cue 23.0384-+ .00078
9 AN Cl wae ceirensos cena tee Seine rg hate e's 23.0395 + .0049
ADIN: BndiCL sec caiatd. isla ee ate dana ea see's 23.1741 + .0050
General mean. ............20.2-0 cee eeeee Na=23.0029+ .00012

This value is perhaps a little too high, on account of the low probable error derived from
ratio 25, the ratio between silver and sodium chloride. The four highest values have very slight
influence upon the general mean. The last value in the series is without significance.

Lithium appears in only five of the ratios, which give the following values:

From ratios—

8 and Chine sss cawie es osc Sanceca eae oe sees Li=6.9325+ .00031
23; Cl, and Agi. oicaccicdteeeannaadsecess S ochaet 6.9366 + .00019
24. Cl and Ag. cox eccdeitey eadsnwun poe cutee 6.9380 + .00022
a, Nijand Cle s:sstscu Masco sactetdddcuicieiine siecogeaes 6.96704 .0017
ONG, OF fox cis ees es eerie ae whale ae aks eee 7.0108+.0022
Gremeral TOAD oo. oec cies ws spelen eee ob e oe Li=6.9368+.00013

For sulphur the following values are found:

From ratios—

65 and Ag...... igi alata o'elsheisidie gine lures o'e dipisjeessaueiele S=32.0578 + .0013
2,70; AG: Ne scjsreiserasie a g:s ciereregciwicla wes eiarsiainie eainitie 32.0579 + .00088
G7 ANG AB se sctsicic celine Saisie Savee sane nae asaeiee 32.0625+.0011
G6 ON AG. og scsi se cciseenance ce men neetinedass 32.0635 + .0015
Tad IN licceisiecices seas sad catellaeiecess Banquets 32.0647 + .00045
We eh Natats siajarayantve Hseeraie 6 «ot nln Gis Siavaia, Sia ce dieietaj toils 32.0683 + .00064
68, Cl, and Ag............-.-.-...-- dajata ci eioetraiereiavoisis 32.0735 +.0015
69 Cl, and Ag.........-.. svabessetastnsa a.a mses assatatatonts 32.1453 + .0033

General mean..........-..---24-2-0--2++---- 5=32.0650+ .00012
The final value is identical with that adopted by Richards and Hoover.
 

118 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.
From ratios— ;
25°79) 5G,jand Api cs ss oe ede cemauenieas eles Br=79.9204+ .0027
20, Cl, and Ag...... Sea ad teamed ene 79.9205 +- .00035
72 and 47........ hen tine At Js See iils adebee ie Le "79.9225 + .0014
932 SL and Agi. ase o oa dantaandedins Seameuad 79.9234-+- .0026
Abiomd Clic. se apy aa nena enn talainyiiazeeokoisrsieaiesse 79.9247 4+- .00024
Gress astentetnede ciatal dat hile edyares bine energie 79.9251 + .0013
J, 3, 40, and Ag.... 02.2000 2c eee e cece ce ccee eee eee 79.9351 + .00085
9,27, and Ag........ee cece eee eee edaade aa aewintes 79.9626 + .00081
1, 2, 8, 64, and Ag.....2..22...2..... Bet de scents 79.9771 + .0166
DA and AG cceticuceed ose ons Seawieaoue ee eae’ 80.0040 + .0134
TS anid selina cits ded acihsh henaey L a aineyte nec 80.1563 + .0596
19) 8 569) and Ag since este eterna ema 80.1621 +: .0166
General mean..............02.20 eee ee ee eee Br=79.9228 + .00015

The general mean agrees well with the best determined values for the atomic weight of bro-
mine. The last four values, with their higher probable errors, are without significance.
Iodine appears in only eight of the ratios, from which the following values are derived:

From ratios—

Tad Wa .5552aas soso assea nsec aeeeeeeeenueee See T=126.497 + .0385
Ty OO BN AP a csciciarcicie wiardwiclaiaie csamisaieoe ia sie o'eie a eter 126.826 +t .0192
US fADd) AG eo wae eek oe ag hes haseincenk Meeiaepiate vie oe 126.854+ .0127
DQ Famd AGS oats Sere otk ahcie' che crashersparcrherd cree teases a6 126.913 + .00034
Unedid aes clviathdaarpvalaiemnaeaele Oe weI Mewes s SSS SR base 126.915+.0056
2120) and Agi ives vee aes ma eameteunee 126.929-+ .00067
22). Br; ON AG occ inree vegrvautiakenciceweeewease wes 126,932 + .00098
16. and Ags seavs seen cress senawsccaG ade ccaecw disc 126.938 + .00036
General mean...............0 02.2.2 cece e eee T=26.926 + .00023

The three low values, with their large probable errors, have only slight influence upon this
combination. At most they effect only the fourth decimal place.

Nitrogen appears in 22 of the ratios from which the following values are derived:

From ratios—

73 BUC 02 sc c0iedanianvascaeaoememesue des ioUt eats N=13.9928 + .0012
44:anid Cliccstsanssocsneectehccinaalcaes eeces's's 13.9965+ .0027
7A, BB, ANG Cli Stecsedvieectao en awn cme chilastsss 13.9976+4 .0013
U8 Cl end Apes seceding ancnd eee eanion S 14.0002 + .0015
A? an Clk eat ett ache en ecmemeeeee seek 14.0030 + .0012
2,99, Ol. dnd Agi. cogs. sedech ee eee Sawn shew 6% 14, 0037 +00032
TAU ai Ol seat cyued teed tinea oe cae budis 14.0070+.00074
Siseccee Ganges ys xe sk se eens See ee ae 14.0076 + .00012
35, Cly and Agiemecss:scetscuacstccsevscaeedeeee 14.0078 +..0078
S4and Agvecss exe sese ns. ss4scceciucieeseedccweste 14.0083 + .00014
AS and ' Clee dosaies ven wei a paniccewaice ree 14.0088 + .00047
BONG: Wai ccc eco aelds Soh ca densa gaonesekiaeeacur 14.0097 + .00034
(2: ABS ANG Clos ec seins earn nacsin aalelaieis alee oaueieiels 14.0114+ .00085
1, 38;'C); and ARs won race heakedancamsceices eens 14.0130+ .00043
73,01; (800.- Ol. a. ssid 2.s4.45-gotidekinagaaseeab sien ees 14.0137 + .0018
1,40) Br, and Agios Ads Soosgagetenwscwesweie ds 14.0199 + .00074
2, G2 and ABe oo. ccie tesa duguesiteakawteesske 14.0210 + .0014
1, 2, 64, and Ag.....-..-..02.0--ee ee eee sasperiecaretes 14.0266+ .0148
73, 50, and Cl............. pecs Cupane a. 14,0283 4.0014
73,36; Cl, and Ag. ccccnnasajssawsievsscccacaseseees 14,0383 + .0025
72, 48, and Cl..............020.. Fthaxcoiehe dialalsianeaceeie 14.0434+4 .0032
1, 2, 63, and Ag....... act ce syecn ene Recietae cceaakorses ' 14.1469 + .0212
General mean.............00. cece cee ee eee N=14.0081+ .00008

In this combination the value found for the atomic weight of nitrogen lies between that

directly determined and represented by ratio 3, and that derived from the composition of
silver nitrate as given by ratio 34. All the other values might be rejected and the final result’
would be the same as that given in the foregoing table.
‘

i ATOMIC WEIGHTS—CLARKE.

The atomic weight of potassium is deducible from 12 ratios, as follows:

From ratios—

31, Br, and Ag
32, Br, and Ag
33, I, and Ag

General mean....................

SR NG and Mie ye eee a

K=38.6440+ .0884

39.0361+.0015
89.0794+ .00090
39.0896 + .00042
39.0929 + .00026
39.0931+4.0181
39.0995 + .00020
39.0999 + .00020
39.1026 + .00042
39.1123 + .00092
39.1246 + .0085
39.3262 + .0596

-e--ee-. K=39.0975+.00011

119

Here again, the extreme values, with their high probable errors, count for nothing. The
final value is determined by the ratios which represent analyses of potassium chloride and bro-

mide. The preliminary value for K (ratio 73) was much lower.

The atomic weight of sodium is derivable from 12 ratios, as follows:

From ratios—

48, Ny Bnd Cle ioc cee neri ks vadaaeieh el eae eits
26, Chand A Gocco daceeee hs okie ccciey cece
28; Br, aNd AG. .ccccesseeceacds senteduenennns ne
2, 55, and Ag..... iisiata le elcieye csistnenavslsle alameda

Na=22.9913 + .0016

22.9942 + .00042
22.9952 + .00096
22.9976 + .00038
22.9985 + .00032
22.9989+ .00025
22.9990 + .00062
23.0099 + .00022
23.0182 + .0014

23.0384 + .00078
23.0395 + .0049

23.1741 + .0050

Na=23.0029 + .00012

This value is perhaps a little too high, on account of the low probable error derived from
The four highest values have very slight

ratio 25, the ratio between silver and sodium chloride.
influence upon the general mean. The last value in the series is without significance.
Lithium appears in only five of the ratios, which give the following values:

From ratios—

28. Cy ANDi A Gia ealos sister svercnre late sieeipenarage onic alate teres
240Cl, and Agssuvcswns siewsans so wavecscuet < o'eieve
58, Ny atid (Ol s.2.0:[.e cused ses seamemiecess es eh

General mean...........-.---0-e eee eee eee

For sulphur the following values are found:

From ratios—

66 and Ag........--...0+- Spee ane

68, Cl, and Ag......-...........-6-- shea Ne EAS test att

69. CL andi AG. hs. entascence crculAsceihlecntgess

General mean................20-0-2----24+- S=32.0650 +: .00012

Li=6.9325+ .00031

6.9366 + .00019
6.9380+ .00022
6.9670 .0017
7.0108 + .0022

Li=6.9368 +: 00013

S=32.0578 + .0013

32.0579 + .00088
32.0625+ .0011
32.0635+.0015
32.0647 + .00045
32.0683 + .00064
32.0735 + .0015
32.1453 + .0033

The final value is identical with that adopted by Richards and Hoover.
118 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [VoL. XVI,
From ratios—
2,72, 56; anid A®.: occscuuanncwdsvaaugionsess tes Br=79.9204-+ .0027
20, Cl, and Ag...... i Seta nee cee cums aata rece ae 79.9205 + .00035
72-ONG ATG cos isdho sh von) a cawsimenenaaenie so8 79.9225 + .0014
78, 31, and Ag....:..-.. BA dleAee Ree Se tes difacnete's 79.9234 + .0026
45 and Cl....... 0... eee ee ee Ride ireee stele Sone 79.9247 + .00024
Boson tac cetera it Sn trca, Fae hs tears 79.9251 +.0013
159.40 and Agio.c.cntde so wud noise dicecetemess’ 79.9351 + .00085
9,27, and Ag......20..-000-020ee- ene ius gees 79.9626 -+ .00081
1, 2, 8, 64, and Ag............2..22200- Oo. telat bes 79.9771 + .0166
V4 Bid Ni. oo Soc ec ore awetsleaueeatioecendicacens 80.0040 .0134
79 anid “Whe tend $3 bine ncsep hi oneaoensusmenateetes 80.1563 + .0596
1, 2, 8, 63, and Ag... 2... c eect e cece cence eee 80.1621 + .0166
General mean.............-.----20--2ee eee Br=79,9228 + .00015

The general mean agrees well with the best determined values for the atomic weight of bro-

mine.

The last four values, with their higher probable errors, are without significance.

Iodine appears in only eight of the ratios, from which the following values are derived:

From ratios—

73 atid 12. wee ceceene vases isee esa esta nome I=126.497 + .0385
13300) ONG APL. ccccese uh sbinbgmetins aalnee eee Sine se 126.826 + .0192
15 And AG. tice cccicac ce pe cestsec dn dd-ajeraeeanineres 126.854+ .0127
W9-and AG os as cactessg cued cnadea bese Gaetan 126.913 + .00034
VSiy csisaisihSis ated s aks AES OK Aen eieei sane meas 126.915 + .0056
21, Cl, and Ag........... MegOR ERIS US a Lae Sa cts aera cine 126.929+ .00067
22) OBE ANd AG ce cckwannst toot cea e ou omeaemaaeianee 126.932 + .00098
16 ad AGsc ccmscavssarngartses s+ SA cede eiogsecns 126.938 + .00036
General mean................22-000-- ee eee I=26.926+ 00023

The three low values, with their large probable errors, have only slight influence upon this
At most they effect only the fourth decimal place. _

combination.
Nitrogen appears in 22 of the ratios from which the following values are derived:

From ratios—

73-900 52) caveswen bs evecais Kets aa aeineteeeeeauaweieie N=13.9928+ .0012
Ad and Olicask dacs +44 ededenara ae seaehews ceewene 13.9965+ .0027
pe SA rented We ede ac cheats Chie toa geeietraie 13.9976 4:.0013
1p BE Cl cand Weiss seentetsr eee aden sccaihaeeeee 14,0002 .0015
42. and Cle oie sexe cs fees 2 exe sce aewe wrecks 14.0030 + .0012
189, Oli wad Ag spc cds veces alas seeewerewmien 14. 0037 4.00032
Aad Meme eee cele eie shtick e 14.0070-+ .00074
poe mulieeaye reels Fake Bd eres Siac Eatetae a Hien De eae 14.0076 + .00012
80, Cl sand AG acs esss ss.6e saa Aaduuananeess eee 14.0078 + .0078
BA eal bass: esha neuseavent aa seseusauasaiens 14.0083 +.00014
AS GN Olio seis 26.2 28 soda wa sNwelekenisedee Soret 14.0088 + .00047
WON TWMisoictssoucsccact net hekccmmwewaonseses 14.0097 + .00034
72; 48, O00 Cli css sees cc as Saijanenceaeacewiwisicte 14.0114+ .00085 z
dy 28) Chenier eee eda 14.0130-+ 00043
73 bl wand Clie veces oor scie Hi komgaugaeaeweneews 14.0137 + .0018
yy, Bia, Ba Nel naceuahadanardensade Baticaid 14.0199-+.00074
ae ne ere ade hl nba e eee 14,0210+.0014
Sg Sys IG each Lacial a nett rn apiece 14.0266 .0148
73, 50, and Cl............. a a eteet aa 14,0283-4.0014
78, 36, Cl, and Ag..!:.......0.... hedcctescmrt 14.0383 + .0025
72, 48, and Cl........... Maat catchavwnasinacded 14.0434 4.0032
12,63, and Agi. cccccccdaravoassasctgasekeunien 14.1469 + .0212
General mean...... Heiter ewaecn beak N=14.0081+ .00008

In this combination the value found for the atomic weight of nitrogen lies between that
directly determined and represented by ratio 3, and that derived from the composition of
silver nitrate as given by ratio 34.. All the other values might be rejected and the final result’
would be the same as that given in the foregoing table.
No. 3.] ATOMIC WEIGHTS—CLARKE. 119

The atomic weight of potassium is deducible from 12 ratios, as follows:

From ratios—

UZ FAD Tso Boh SsMe ya ati rants 2 ts pitti Pore K=38.6440+ .0384
50, N, and Cl....... MS lafclacsvaig care Gite e's wea Sielnereow artis 39.0361+.0015
S6,0N5 Cl ivand Ag. lou aeicaceecie vid vx waspioeecien 39.0794 +- .00090
$1, .BY, OND AG cc mewn dose eee gcd os 8 39.0896 + .00042
52; -BYy ANd AP. signee daceineaae eek Fokaieneeeedene 39.0929 + .00026
83, I, and Ag.............. ana dupeta rina agetcieaeras aye ietete 39.0931 4.0181
Pec id Ne noes eee el atta es 39.0995 -+.00020
BOC sjeinicl Np cic utnceerdauleace'e a Aa de ocicbeanunta wk 39.0999 + .00020
Wand Clovs ciegseedewausss tess sda hemmbeicin Sea ao 39.1026 + .00042
52 and N........ Ranges Saad a ESE 39.1123 + .00092
Ol; Ni and Clo ceo23 2 shawniretecs dos onoosacenand 39.1246+ .0085
Dan Bro 2scesiscttndcsuau ds oasswscacameane ss 39.3262 + .0596

General mean..............--0..-000.2-++- K=39.0975+.00011

Here again, the extreme values, with their high probable errors, count for nothing. The
final value is determined by the ratios which represent analyses of potassium chloride and bro-
mide. The preliminary value for K (ratio 73) was much lower.

The atomic weight of sodium is derivable from 12 ratios, as follows:

From ratios—

AS) N; anid Ole: sntica: eves waeeeeenes coarse Na=22.9913-+ .0016
26. Cl and Aes chil ee epee Lo 22.9942+- 00042
28, Br, and Ag............ccee ccc ee eee ee eens ae 22.9952+ .00096
9. 56, and Ag cos eecenc guecueweensscce 2A 22.9976 -+ .00038
47 and Br.............-- 2h pavoaadatianetes sae 22.9985 -+ .00032
46 and Cl.............-.- ngs aainscaseoeee 22.9989 + .00025
2, 56, Br, and Ag............ 2. cece cece eee e eee 22.9990 -+ .00062
25, Cl, and Ag..............- abate cet i buah lent 23.0099+ .00022
Di A cami] Osea Jescmieate orale aera cane 23.0182+ .0014
D7 Brand Neo. ssicie sie sxe ercsree dice dies ob 3 ES8 23.0384+ .00078
Orarid Ol sie. se avacarssssipiels 2 ebeara del aaacaban bee oo 23.0395 + .0049
AD, N cand: Clo. ccxcucen-gecnamemacechaeusentieon 23.1741 -+ .0050
General mean...........-2..0..0ee eee Na=23.0029-+ .00012

This value is perhaps a little too high, on account of the low probable error derived from
ratio 25, the ratio between silver and sodium chloride. The four highest values have very slight
influence upon the general mean. The last value in the series is without significance.

Lithium appears in only five of the ratios, which give the following values:

From ratios—

Sand Ole sei Aad dee die iil eee toe Li=6.9325+.00031
25... Clyiand Ag. cossorves vs caneweewewi so taainagiens 6.9366 + .00019
24, Cl, and Ag.............00- we atest toda 6.9380 + .00022
BS: Nand Cl xen eve we cst ase Lose ds oem 6.9670-+.0017
BANC sO4 ss vekomece aaa Gasecmielnen.e aka ouceewas Stas 7.0108 + .0022
General mean.......---------.---.-----0ee Li=6.9368+ .00013

For sulphur the following values are found:

From ratios—

Go rid Nes dat amen en yam =82.0578-+.0013
DAT adil Va sctinclonaal nda aeons aancbhaciaeeaesanee 32.0579 .00088
Gi Ae Ae epee me aciecems teas ea 32,0625 +.0011
GE A ie dete eddadduhiotcasobdadceeeien ce 32.0635+.0015
PIAL IN oricgaghetwiesoolagennnectne es Scuitoncte 32.0647 + .00045
hie sO de aatnehe ds aici aatn cannon nee real eaRtantd 32.0683 .00064
Gs Ol, Gnd dagen sitar canaean dain 32,0735 +.0015
BOL, AN Benet s-nenceanacdeeoannnantnaee 32.1453 + .0033

General mean......--..2.20- 2-2 eee eee eee S=32.0650+ 00012
The final value is identical with that adopted by Richards and Hoover.
120 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The atomic weight of carbon is given’by 13 ratios, but with widely varying results. The
values are as follows:

From ratios—

Daan di Gt: elas on GtaateciPaweeees se ses eeeeeeneewe C=11.7829+ .0071
57 and Ag............. Se oe at Lt eae elves Moca rma he 11.9320+ .0070
55, Na, and Ag..... Lucicca aie see heen can 11.9895 + .00074
1258 aNd: AG... saticananie eeacneg as Ae eee 11.9905 + .0014
56, Na, Br, and Ag............2....002 2 eee eee ee 11.9922+ .0012
Qe oc cccecc nce cee ee ec ccc cence ecececeseeceeseeeeeee  12,0000-+.00026
Ly Gland Ag sc tcwceandniee ako eeu eae 2 ee 12.0009 + .00058
1,64, N, Broand Ag: oi. .tweeein sys esses 12.0046 + .0027
70;:S,and Naiecssassaveeecseasankeescesk eee xe 12.0053 + .0006
62. Nyand' Api. os. siace wails Aussce ere seed he see 12.0128 + .0014
1, 68, N, Br, and Ag...............0000.20eee eee 12.0181 -+ .0006
Ly 60nd Apo: neaseie genet eki ese ees oad oecene 12.0269 + .0098
1,09, ON AGS 22s cvaan shite enaindiaine eee ee ee ns 12.0474+ .0012
General mean...........------ Tectia Scena as C=12.0025+ .00019

That this mean is higher than the atomic weight given in ratio 2, does not prove it to be
in error. That derived from ratio 70, as determined by Richards and Hoover, is still higher
and entitled to much credit. The general mean of all determinations agrees well with the
results obtained by modern physical methods, and may, therefore, stand, until it is superseded
by something of less uncertainty.

As for hydrogen, new values for its atomic weight can be deduced from 11 of the funda-
mental ratios. The computation has been roughly made, and found to be without signifi-
cance. The combined values, so obtained, are of such small weight in comparison with ratio
(1) that they only modify it in the sixth decimal place, a change which is not worth considering.

To sum up: The subjoined values, referred to O=16 as the standard, have been computed
from all the ratios, old and new, good, bad, and indifferent:

H= 1.0078-£.00001 Br= 79.9228-+.00015
C=12.0025 + .00019 I=126. 926-+.00023
N=14.0081+ .00008 Li= 6,9368+.00013
S=32.0650-+.00012 7 Na= 23.0029+.00012
C1=35.4605-+.00011 K= 39.0975+.00011

Ag=107.880+,.00016

That these values are final, is not to be supposed. That they are, in the strict mathe-
matical sense, the most probable values deducible from the experimental data, is also question-
able. But that they are highly probable values, in harmony with the best modern evidence,
can safely be asserted. The inferior determinations, low in weight, have practically vanished,
one might almost say self-rejected, but not thrown out arbitrarily. The good measurements
overwhelm the doubtful ones, whose influence upon the final computations is almost negligible.
The 11 values as given above, will be used in calculating the atomic weights of all the other

elements.
RUBIDIUM.

The atomic weight of rubidium has been determined by analyses of the chloride and bromide.

Bunsen,’ employing ordinary gravimetric methods, estimated the ratio between AgCl and
RbCl. His rubidium chloride was purified by fractional crystallization of the chloroplatinate.
He obtained the following results, to which, in a third column, I add the ratio between RbCl
and 100 parts of AgCl:

RbCl. AgCl. Ratio.
1. 000 1.1873 84.225
1. 000 1.1873 84.225 |
1. 000 1.1850 84.388
1. 000 1.1880 84,175

Mean, 84.253+.031
Hence Rb= 85.309.

 

1Zeit. Anal. Chem., 1,136. Poggend. Annal., 113, 339, 1861,
No, 3.) ATOMIC WEIGHTS—CLARKE. 121

The work of Piccard * was similar to that of Bunsen. In weighing, the crucible containing
the silver chloride was balanced by a precisely similar crucible, in order to avoid the correction
for displacement of air. The filter was burned separately from the AgCl, as usual; but the
small amount of material adhering to the ash was reckoned as metallic silver. The rubidium
chloride was purified by Bunsen’s method. The results, expressed according to the foregoing
standard, are as follows:

RbCt1. AgCl. Ag. Ratio.
1.1587 1.372 +.0019 84.300
1.4055 1.6632 +.0030 84.303
1.001 1.1850 +.0024 84.245
1.5141 1.7934 +.0018 84.313

 

Mean, 84.290-+.0105
Hence Rb = 85.361.

Godeffroy,? starting with material containing both rubidium and cesium, separated the
two metals by fractional crystallization of. their alums, and obtained salts of each spectro-
scopically pure. The nitric acid employed was tested for chlorine and found to be free from
that impurity, and the weights used were especially verified. In two of his analyses of RbCl
the AgCl was handled by the ordinary process of filtration. In the other two it was washed
by decantation, dried and weighed in a glass dish. The usual ratio is appended in the third
column:

RbC1. AgCl. Ratio.
1.4055 1.6665 84.338
1.8096 2.1461 84.320
2.2473 2.665 84.326
2.273 2.6946 84.354

 

Mean, 84.3345-+.0051
Hence Rb = 85.425.
Heycock * worked by two methods, but unfortunately his results are given only in abstract,
without details. First, silver solution was added in slight deficiency to a solution of rubidium
chloride, and the excess of the latter was measured by titration. The mean of seven experi-

ments gave—

Hence Rb = 85.285.

Two similar experiments with the bromide gave—
Ag : RbBr :: 107.93 : 165.437
Ag : RbBr :: 107.93 : 165.342

Ag : RbCl :: 107.93 : 120.801

Mean, 165.3895 +.0320

Hence Rb = 85.390.

The determinations by Archibald‘ were made with scrupulously purified materials, and
with all of the precautions observed in the best modern investigations. The chloride and
bromide were precipitated with known weights of silver, and the silver halide produced was
also weighed. ‘Two ratios were thus measured for each salt, and checked by the cross ratios
between silver and chlorine or bromine, respectively. The weights, corrected to a vacuum, are
given below, and also the four principal ratios:

 

Weight RbC1. Weight AgCl. Weight Ag. AgCl ratio. Ag ratio.
1.99966 2.37070 1.78454 84.349 112.054
2.06480 2.44778 1.84241 84.354 112.070
2.29368 2. 71960 2.04710 84,339 112.046
1.09495 1.29796 -97702 84.360 112.070
2.14381 2.54118 1.91316 84.364 112.056
2.89700 3.43475 2.58550 84.344 112.047
2.19692 2.60452 1.96076 84.350 112.044
2.14543 2.54386 1.91462 84.338 112.055

1 Journ, prakt. Chem., 86, 454. 1862. Zeit. Anal. Chem., 1, 518. 3 British Association Report, 1882, 449,

z Ann Chem. Pharm., 181, 185. 1876. 4 Journ. Chem. Soc., 85, 776. 1904.
 

 

122 MEMOIRS NATIONAL ACADEMY OF SCIENCES. rVou. XVI.
Weight RbC1. Weight AgCl. Weight Ag. AgCl ratio. Ag ratio.
2.12164 2.51557 1.89346 84.341 112,052
2.25777 2.67685 2.01515 84.344 112.040
2.18057 2.58528 1.94594 84.346 112.057
2.32699 2.75878 2.07668 84.348 112.053
4.00035 2.74233 3.56998 84.354 112.055
2.43440 2.88613 2.17233 84.348 112.064
Mean, 84.3485, 112.0545,
+.0014 +.0016
From the AgCl ratio, Rb =85.445.
From the Ag ratio, Rb = 85.424.
And Ag : Cl ::100 : 32.847.
The values for the AgCl ratio combine as follows:
A BUNBCR Sc 30s eee can gem oe Se eae 84.253 +.031
PPC CHT 5 B26 cotacsere sd carat neal eon Roe ch ieRt oe aetige 84.290 +.0105
Godefroy soso ect fo cialatatsteieyertelecpcterateaaptis entice We ates ee 84.3345 + .0051
JATChi bald ates cvoe desea mbeiat ne onus se dele eae 84.3485+4+ .0013
General mean.........-.2----+0+220eee ee eee sees eee 84.3433 + .0013

Heycock’s single value for the Ag ratio, reduced to the usual standard, becomes Ag :
RbCl :: 100 : 111.926. It is not worth while to combine it with Archibald’s values, for its
influence would be quite negligible. In the AgCl ratio the older determinations count for
something, but the general mean falls within the range of variation of Archibald’s series.

The bromide analyses by Archibald are as follows:

 

 

Weight RbBr. Weight AgBr. Weight Ag. AgBr ratio. Ag ratio.
2.68170 3.04578 1.74930 88.047 153.301
2.07280 2.35401 1.35230 88.054 153.280
2.10086 2.38589 1.37061 88.053 153.278
2.61044 2.96462 1.70300 88.053 153.285
3.84082 4.36215 2.50590 88.049 153.272
3.77852 4.29084 2.46502 88.061 153.287
4.34299 4.93210 2.83340 88.56 153.278

Mean, 88.0533 153.283,
+.0012 +.0024

From the Ag ratio, Rb =85.439.
From, the AgBr ratio, Rb =85.444.
And Ag : Br :: 100 : 74.080.
Heycock’s mean for the Ag ratio, reduced, becomes Ag : RbBr :: 100 : 153.238 +0300.
Its probable error is so high that combination with Archibald’s data would be useless.
There are now four ratios from which to compute the atomic weight of rubidium:
(1) Ag: RbCl :: 100 : 112.0545 +.0016
(2) AgCl : RbCl : 7: 100 : 84.8433 + .0013
(3) Ag : RbBr :: 100 : 153.283 + .0024
(4) AgBr :: 100 : 88.0533 + .0012
Reducing these ratios with Ag=107.880+.00016, Cl=35.4605 +.00011,
79.9228 + .00015, we have—

From ratio—
Licsiss Satine Reet eaesd oe Ana Ane Semee Rb=85.424+ .0017

c

and Br=

2 vammacnees busses eeRRARaNsaER naan wales 85.438 + .0019
DiaicininsSicnecatesa's isin vin de tigishclalhetotersewia saimingeek’s eae selec 85.439-+ .0026
Bee csiee ce does ss Saratels ecards Sele ey wok ey ete 85.444 + .0023

General Meéatecesccdnsevess sx cescescsicieexes ess Rb=85.434+- .0010
No. 3.] ATOMIC WEIGHTS—CLARKE. 123
CESIUM.

The atomic weight of cesium, like that of rubidium, has been computed from analyses of
the chloride and bromide, and also from experiments upon the nitrate. The earliest determi-
nation, by Bunsen,’ was incorrect, because of impurity in the material studied. The first
trustworthy determinations were published by Johnson and Allen? in 1863. Their material
was extracted from the lepidolite of Hebron, Maine, and the cesium was separated from the
rubidium as bitartrate. From the pure cesium bitartrate cesium chloride was prepared, and
in this the chlorine was estimated as silver chloride by the usual gravimetric method. Reduc-
ing their results to the convenient standard adopted in preceding chapters, we have, in a third
column, the quantities of CsCl equivalent to 100 parts of Ag(l:

CsCl. AgCl. Ratto.
1.8371 1.5634 117.507
2.1295 1.8111 117.580
2.7018 2.2992 117.511
1.56165 1.3302 117.399

 

Mean, 117.499+.025

Hence Cs = 132.963.

Shortly after the results of Johnson and Allen appeared a new series of estimations was
published by Bunsen.? His cesium chloride was purified by repeated crystallizations of the
chloroplatinate, and the ordinary gravimetric process was employed. The following results
represent, respectively, material thrice, four times and five times purified:

CsCl. AgCl. Ratio.

1.3835 1.1781 117.435
1.3682 1.1644 117.503
1.2478 1.0623 117.462

 

Mean, 117.467+.013
Hence Cs = 132.917.
Godeffroy’s work * was, in its details of manipulation, sufficiently described under rubid-
ium. In -three of the experiments upon cesium the silver chloride was washed by decanta-
tion, and in one it was collected upon a filter. The results follow.

CsCl. AgCl. Ratio.

1.5825 1.351 117,135
1.3487 1.1501 117.265
1.1880 1.0141 117.148
1,2309 1.051 117.107

 

Mean, 117.164+.023

Hence Cs = 132.483.

The foregoing investigations may now be regarded as merely preliminary, in comparison
with the more elaborate determinations made by Richards and Archibald.’ Their material
was purified by fractional crystallization as cesium dichloriodide, from which the chloride,
bromide, and nitrate were afterwards prepared. The chloride and bromide were freed from
possible traces of moisture by fusion in an atmosphere of nitrogen, and analyzed’ by the usual
method. That is, they were precipitated by known weights of silver dissolved as nitrate, and
the silver chloride or bromide produced was also weighed. All the weights given are reduced
to a vacuum standard. The results obtained with cesium chloride are given in the next table:

 

Weight CsCl. Weight AgCl. Weight Ag. AgCl ratio. Ag ratio.
3.83054 3.26240 2.45600 117.415 155.967
3.95120 3.36532 2.53351 117.409 155.958
2.27237 '1,93555 1.45686 117.402 155.977
3.02935 2.58003 1.94244 117.415 155.956
3.19774 2.72382 2.05023 117.399 155.970
1 Zeitsch. Anal. Chem., 1, 137. 8 Poggend. Annalen, 119, 1. 1863. 5 Proc, Amer, Acad., 38, 443. 1903.

2 Amer, Journ. Sci. (2), 35, 94. 4 Ann, Chem. Pharm., 181, 185. 1876.
124 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

 

 

Weight. CsCl. Weight AgCl. Weight Ag. AgCl ratio. Ag ratio.
2.35068 2.00253 1.50270 117.386 155.963
2.06245 1.75678 1.32251 117.399 , 155,950
2.56372 2.183858 = ...... 117.409 sis nee ee
2.01881 1.71972 1.29434 117.392 155.972
1.77391 1.51093 1.13743 117.405 155.958
3.08160 2.62484 1.97590 117.401 155.959
3.13117 2.66720 2.00760 117.395 155.966
5.06656 4.31570 3.24850 117.398 155.966

Mean, 117.402 155.9635

\+.0016 +.0016
From Ag ratio, Cs = 132.793.
From Ag(Cl ratio, Cs = 132.820.
And Ag:Cl::100:32.846.
There is also a single analysis of cesium chloride made by Stewart + for Richards, whose
figures are as follows:

CsCl. Ag. AgCi. Ag ratio. AgCl ratio.
0.72285 0.46336 0.61583 156.002 117.378

To each of these ratios the probable error of one determination by Richards and Archibald,
+.0069, may be assigned. Combining the silver ratio with the value found by Richards and
Archibald, the weighted mean is Ag:CsCl::100:155.966 +.0016. The silver chloride ratio com-
bines with the others as follows:

Johnson and Allen. ........-....---.- ee eee eee eee eee eee 117.499+ .025
BUDE ec :cccae a atandeeriscieaedied es 258 Soe eee eae o ka eases 117.467 + .013
GodeHToy. i. ssceis. tac bak ta setere ps5 Pas S584 vaeGReae 117.164 + .023
Richards and Archibald................------------------ 117.402 + .0016
Stewart. 4.cccces stemcndeatecete dees seuss s Rabtenbeades 117.378
General mean... ...-.. 222-222 eee eee eee eee eee eee 117.403 + .0016

The older determinations practically vanish, leaving the Richards and Archibald mean

almost unchanged.
The figures for the bromide series are as follows

 

 

Weight CsBr. Weight AgBr. Weight Ag. AgBr ratio. Ag ratio.
3.49820 3.08815 1.77402 113.278 197.192
6.20409 5.47673 3.14606 113.281 197.202
2.17300 6.33213 2.63740 113.279 197.201

Mean, 113.279 197.198
+.0007 +.0045

From Ag ratio, Cs = 132.814.

From AgBr ratio, Cs = 132.818.

And Ag: Br::100:74.078.

When cesium nitrate is fused with finely powdered silica, cesium silicate is formed, and
the elements of the nitric radicle are expelled. With weighed quantities of the nitrate, the
loss of weight is equivalent to N,O,. The ratio N,O,:Cs,0::100:z is thus easily determined. In
four experiments Richards and Archibald obtained the following results:

Weight CsNO3. Weight N,05. Ratio.
3.76112 1.04273 260.699
3.33334 92416 260.689
4.81867 1.33590 260.706
5.04807 1.39960 260.679

 

Mean, 260.693 + .0089
Hence Cs = 132.795.

1 Journ, Amer. Chem. Soc., 37, 289. 1915.
No. 3.] ATOMIC WEIGHTS—CLARKE. 125

The five ratios for the atomic weight of cesium are now as follows:

(1) Ag: CsCl :: 100 : 155.96 +.0016

(2) AgCl : CsCl :: 100 : 117.403 +.0016
(3) Ag: CsBr :: 100 : 197.198 4.0045
(4) AgBr : CsBr :: 100 : 113.279+.0007
(5) NO; : Cs,O :: 100 : 260.693 + .0039

Reducing these ratios with—

Ag= 107.880-+.00016 . Br=79.9228 + .00015
Cl= 35.4605+.00011 N=14.0081+ .00008
we have—
From ratio—
Oe Ser acitedizenses x ote SiereaeiSisba nano neo ae a ee eaaweaeaS Cs==132.795+ .0021 ;
le saMeeiiataes a on aera gash e ec eeear came areoceecans 132.796 + .0017
BD eds aecu epee sesasi ee esac wx reve neaecenicaetene tei dec ns oe te ara Sse 132.814 + 0049
A Adee daceyelostetnene acta aca hone sly enn Qabeaih be 132.819 + .0013
Dah esa tarce teil 2s 4 wd Seyret Relat adic Sabet ae 132.826 + .0023
General mean.............2.... 2202 eeeeeee Cs==132.811+ .00084
COPPER.

The atomic weight of copper has been chiefly determined by means of the oxide, the sul-
phate and the bromide, and by direct comparison of the metal with silver.

In dealing with the first-named compound nearly all experimenters have agreed in reducing
it with a current of hydrogen, and weighing the metal thus set free.

‘The earliest experiments of any value were those of Berzelius,! whose results were as fol-
lows:

Cud. 0. Per cent Cuin CuO.
7.68075 1.55 79.820
9.6115 1.939 79.826

 

Mean, 79.8234 .002
Hence Cu = 63.298.

Erdmann and Marchand,’ who come next in chronological order, corrected their results
for weighing in air. Their weighings, thus corrected, give us the subjoined percentages of
metal in CuO:

CuO Cu. Per cent.
63.8962 51.0391 79.878
65.1590 52.0363 79.860
60.2878 48.1540 79.874

46.2700 36.9449 79.846
Mean, 79.8645 + .0038.

 

Hence Cu = 63.462.

Still later we find a few analyses by Millon and Commaille.* These chemists not only
reduced the oxide by hydrogen, but they also weighed, in addition to the metallic copper, the
water formed in the experiments. In three determinations the results were as follows:

Cud. Cu. #0. Per cent.
6.7145 5.3565 1.5325 79.775
3.3945 2.7085 -7680 79.791
2.7880 2.2240 cries 79.770

 

Mean, 79.7787 + .0043
Hence Cu = 63.125.

For the third of these analyses the water estimation was not made, but for the other two
it yielded results which, in sum, would make the atomic weight of copper 63.165. This figure
has so high a probable error that we need not consider it further.

 

1Poggend. Annual., 8,177. 1826. 2 Journ. prakt. Chem., 31, 380. 1844. 3 Fresenius’ Zeitschrift, 2, 475. 1863.
126 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The results obtained by Dumas! are wholly unavailable. Indeed, he did not even publish
them in detail. He merely says that he reduced copper oxide, and also effected the synthesis
of the subsulphide, but without getting figures which were wholly concordant. He puts Cu=
63.5.

In 1873 Hampe? published his careful determinations, which were for many years almost
unqualifiedly accepted. First, he attempted to estimate the atomic weight of copper by the
quantity of silver which the pure metal could precipitate from its solutions. This attempt
failed to give satisfactory results, and he fell back upon the old method of reducing the oxide.
From 10 to 20 grammes of material were taken in each experiment, and the weights were ©
reduced to a vacuum.

Cuod. Cu. Per cent.
20.3260 16.2279 79.838
20.68851 16.51669 79.835
10.10793 8.06926 79.831

 

Mean, 79.8347 + .0013
Hence Cu = 63.344. :

Hampe also determined the quantity of copper in the anhydrous sulphate, CuSO,. From
40 to 45 grammes of the salt were taken at a time, the metal was thrown down by electrolysis,
an dthe weights were all corrected. I subjoin the results:

CuSO,. Cu. Per cent.
40.40300 16.04958 39.724
44.64280 17.73466 39.726

 

Mean, 39.725+ .0007

The last series of data gives Cu=63.314, and is interesting for comparison with results ob-
tained by Richards later.

In all of the foregoing experiments with copper oxide, that compound was obtained by
ignition of the basic nitrate. But, as was shown in the chapter upon oxygen, copper oxide so
prepared always carries occluded gases, which are not wholly expelled by heat. This point
was thoroughly worked up by Richards * in his fourth memoir upon the atomic weight of copper,
and it vitiates all the determinations previously made by this method.

By a series of experiments with copper oxide ignited at varying temperatures, and with
different degrees of heat during the process of reduction, Richards obtained values for Cu
ranging from 63.20 to 63.62. In two cases selected from this series he measured the amount
of gaseous impurity, and corrected the results previously obtained. The results were as follows,
with a vacuum reduction :

CuO Cu. Per cent.
1.06253 -84831 79.802
1.91656 1.5298 79.820

 

Mean, 79.811+.0061

Correcting for the occluded gases in the oxide, the sum of the two experiments gives 79.901
per cent of copper, whence Cu=63.605. The other indirect results, similarly corrected, gave
79.900 per cent Cu in CuO, or Cu=63.603. If we assign all five experiments equal weight, and
judge their value by the two detailed above, the mean percentage becomes 79.900 + .0038.

The experiments on copper oxide, by Murmann,,* are of very doubtful utility. Copper was
oxidized by heating in oxygen, and the oxide was also reduced in hydrogen, giving values for
Cu varying from 63.513 to 64.397. The five experiments, with all corrections, including re-
duction to a vacuum, and rejecting the excessively high figure given above, may be stated in the
following form :

 

1 Ann. Chim. Phys. (3), 55, 129. 1859. 8 Proc. Amor. Acad., 26, 276. 1891.
2 Fresenius’ Zeitschrift, 13, 352. ‘ 4 Monatsh. Chemie., 27, 351. 1906.
No. 31 ite ATOMIC WEIGHTS—CLARKE. . 127

Cu. cud. ., Percent Cu.
1.13625 1.41856 80.099
2.64333 3.30923 79.878 |
1.07874 1.35045 79.880
5.12489 _ 6.41350 79.908

8.33515 4.17315 79.919
on . =: =. | Mean, 79.987 0278

fl

"Hlanoa ‘ie 63.749. Murmann. himself seledted, wahiee from his series ee —
63.512 and 63.560... 0. ic Ht

The determinations by. Oechsner de Goninck snl Ducelliez ! are of very slight value.
Copper was oxidized by' nitric acid, but without consideration of gaseous inclusions in the
oxide'so. a and with no reduction of the weights to a vacuum. The data are as follows;

Che Eee “ass CuO, Per cent Cu.
0.2200 0.2754 79.884
-2860 / -3580 79.888
3740 4682 79.881
2421 _ 8030 79.901
.2664 3335 79.880

"Mean, 79.887 4:.0026
Hence Cu = 63.551.
The several series for the percentage of copper in copper oxide now combine as follows,
although for practical purposes the combination has small significance.

 

 

 

Berzelius. & sec cccescthc sacdsncecanechorseeneesaecegserie £9820: t.0020
Erdmann and Marchand.......0.20.-00-2000 sedi cariecesses ---- 79.8645 + .0038
Millon and Commaille......\ OSB tee ae ere 79.7787 +:.0043 | oe
Hampe. sie. is..ccceee eee teele Guba) cee det tcesecaiee 79.8347 + .0013 ts
RichazdSsesinis<s ta zcenecaaewsx bas dysaeraystitesssense ss 79,900 + .0038
Murmann.. Jerds eee een ener heaitsies erkestess 79.937 +.0278
Oadhsnerand Diioeliteny. 2.08.0 a waad areas ans 79.887 +.0026
gat General mean..........- ridicha.n Shifar iataye Ske awe eat aeytisiets 79.842 +.0009

This result is not far from that of Hampe, whose work receives excessive weight, as does
also that of Berzelius. The oxide of copper is erdenaly of doubtful value in the measurement
of this atomic weight.

The composition of copper suits has esti studied, not sil by Hampe, but also by
Baubigny:? and by Richards. 7 Baubigny merely ignited the anhydrous' salt, weighing both it
and the residual oxide, as follows:

et CuSO," “2 Eud. Per cent:

7 (402 2.0085, _, 49.813
2 596 * 1.293 49. 807

SS | Nee ee te? ce

bore 2, ~~ Mean, 49.8104..002... 0 > 8, >
Hence, Cu = 63.459. AP es a ‘ ata ee
The same ratio, in reverse—that is, the,synthesis of the sulphate from the oxide—was
investigated. by Richards, who shows that the results obtained are vitiated by. the same errors
which affect the copper oxide experiments previously cited. The weights given are reduced

toa vacuum. The percentage of oxide in the sulphate is stated in the third eclam of AgHnes:

 

 

Cu0. CuSO,. Percent. 80 ts
1.0084 aa 2, 0235, _. 49. 835
2. 7292 5.4770 49. 830
wae 1.0144 , 2, 0350 49. 848
ee - i Mean, 49. 838 +. 0036
Hence, Cu = 63.548.
eRe, Ger Chia, 16 Pella SAREE BANS PTO: TERE: ga hen OC Ree a el IP

‘le1566°—20—9 - - S ysis

ro

; so? : it He eg ae ates inh
ae 78 oo EE RN ere ty - eae ane PURE
128 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The two series combine thus:

Baubigny: scecescocteetee ecw atiaiues Sessa emeacmense 49, B+. 0020
Richards. sce. swsccccnccudtesiecacass dudeanseaceceseoea oven 49. 838+. 0036
General mean. ............ Tease eeapageeacmeaeas ‘49, 816-4. 0017

Here, plainly, the rigorous discussion gigas ‘Baabigny s work weight in excess of its merits.

In the memoir by Richards, now under consideration, his fourth upon copper, the greater
part of his attention is devoted to the'sulphate, Hampe being followed closely in order to ascer-
tain what sources of error affected the work of the latter. Crystallized sulphate, CuSO,.5H,O,
was purified with every precaution and made the basis of operations. Three series of experi-
ments were ¢arried out, the water being determined by loss of weight upon heating, and the
copper being’ estimated electrolytically. In the first series the following data were found,
the weights being reduced to a vacuum, as in all of Richards’ determinations:

_ CuSO,. 5 ag. CuSO, at 250°. Cu.
tesa aetaccee es eR cereethat naar e a Booe BgRIBD!° | tua, 0. 7337
Divacsisdicuiciscuueee veers vesvsessecaseaneeemesen “2enkDe: = = Neiteers - 6911
D sinseiaisiowrels atorarelewwrels eo arsid bbe. sib g sterdestateesiatsinngase .- 3.4639 2. 2184 - 8817

Water at 250°. Cu. in Cryst. Salt. Cu in CuSO,.

 

desis eceeaweweas tres ess livee sone  eeaece 25. 462 savers

Qeswsee Ede hegce Gees Webvesteceetvodliieceeae seeees ; 25.452 j= — ......

Ssausseussacs sie erate Gu catsbd iiss sansa sdaesoanws 35. 958 25. 454 39. 745
Mean, 25. 456

In the second series of analyses, which are stated with much detail, several refinements
were introduced in order to estimate also the sulphuric acid. These will be considered later.
The results given below are numbered consecutively with the former series:

 

 

 

CuSO,.5 ag. CuSO, at 260°. CuSO, at 860°. Cu.
4s. soto gdeneeeaecees 3. 06006 1. 9597 1. 95637 0. 77886
Di decececeean tS usttoy d's BL SISO 1. 8048 cteamates .71740
Giageckgaeeneg energies Te BOO _ 4.8064 4, 79826 1. 90973
Hence percentages as follows:
Water, 260°. Water, 360°. Be in Cryst. Salt. Cu in CuSO,, 260°. Ditto, 360°.

So cssecasins 35. 959 36. 068 Ma 25. 452 39. 744 39. 811
Geis cevseas 35.964 = = = ....2.: 25. 454 39.750 sa eeee
Coasscesce 35. 957 36. 065 25. 446 39.733 39. 799

Mean, 35. 960 36. 067 25. 450 39. 742 39. 805

Hampe worked with a sulphate dried at 250°, but these data show that a little water is
retained at that temperature, and consequently that his results must have been too low. The
third of Richards’ series resembles the second, but extra precautions were taken to avoid
conceivable errors.

CuSO,5ag. °' CuSO, at 260°. CuSO, at 870°. Cu.
Tercgsaeueicshamsencee> Qeeka0r ior ene pe eae 0. 73380
Scasinannie innaavananaiey 18, 82019 BAe eseteld . 92344
Disnagacternermrnexums .-- 6, 81352 | RRR 3.71680 1. 47926
And the ueedios are:

Water at 260°. At 870°. Cu in Cryst. Salt. Cu in CuSO.
Dicacecucamnorcan qe cee ee 25452 2 nsinsae
Ructavartenienieneeene UNTO — atecis 25. 446 39. 740(260°)
D pscciensunigoaaveRie. aScden 36. 067 25. 445 39. 799(370°)

25. 448

In this series the determinations of sulphuric acid gave essentially the same results for all
three samples of sulphate, although one was not dehydrated, and the others were heated to
No. 3.] : ATOMIC WEIGHTS—CLARKE. 129

60° and 370°, respectively. Hence the loss of weight in dehydration at either temperature
represents water only, and does not involve partial decomposition of the sulphate. Between
360° and 400° copper sulphate is at essentially constant weight, but further experiments indi-
cated that even at 400° it retained traces of water, and possibly as much as .042 per cent. The
— trace is not expelled until the salt itself begins to decompose.

' Richards also effected two syntheses of the sulphate directly from the metal by dissolving
‘ing latter in nitric acid, then evaporating to dryness with sulphuric acid, and Aesking to constant
_— at 400°..

_ Cu. CuSQ,. . Per cent Cu.
0. 67720. 1. 7021 : 39. 786
1 00613 2. 5292. , 39. 781

If we include these percentages in a series. with the Ante from. analyses 4,6, and 9, which
gave percentages of 39.811,.39.799 and. 39. 799, i dei of copper in sulphate dried at 360°

and upwards, the mean becomes aa
Cuso, :Cu : : 100 : 39. 795 + .0036
Hence, Cu = 63.498.

Since even this-result is presumably too low, the other figure from sulphate dried at 250°
must be rejected. Since Hampe’s work on the sulphate is affected by the same sources of
error, and apparently to a still greater extent, it need not ‘be considered farther. As for
Richards’ nine determinations of Cu in CuSO,.5H,O, we may take them as one series giving a
mean percentage of 25.4514.0011, and Cu=63.55. This salt seems to retain occluded water,
for the percentage of copper in it leads to a value for the atomic wee which i is inconsistent
with the best evidence, as will be'seen later.

In the second and third series of Richards’ experiments upon copper sulphate, the’ sulphuric
acid was estimated by a method which gave valuable results. After the copper had been
electrolytically precipitated, the acid which was set free was nearly neutralized by a weighed
amount of pure sodium carbonate, and the slight excess remaining was determined by titration.
Thus the weight of sodium carbonate equivalent to the copper was ascertained. The resulting
solution of sodium sulphate was then evaporated to dryness, and a new ratio, connecting that
salt, with copper, was also determined. The cross ratio Na,CO, : Na,SO, has already been
utilized in a previous chapter. The results, i ignoring the weights of hydrated copper sulphate,
are as follows, with the experiments numbered as before:

 

 

Cu. Na,CO5.° - Na,S0,.
dee See beueeneny asec ea een eee . T7886 1. 2993 1.7411
B ccsalt ieccntdansnecatemaseasione sie seees 1. 90978 3. 1862 4, 2679
ceca Raden eee AES, J yotecauntand nace exeeneee . 73380 1, 22427 1. 63994
ae se saadaae hihi elakmasandn aimee . 92344 1.54075 sane
AG aisaihs iempestaanpessle phi Anaahat AA Cs cca 1.47926 ann tues 3. 30658
Hence, ee ae Oe
Cu : Na,CO;z:: 100: x. Cu: Na,SO,::100:%.. |.
166. 824 223.549 3°,
166. 840 223, 482
166. 840 . 223. 538
166. 849 223. 529
Mean, 166. 838+. 0035 Mean, 223. 525+. 0098
Hence Cu = 63.540. Hence Cu = 63.559.

In their work upon the atomic weight of sulphur, which is cited in a former section of
this recalculation, Richards and Hoover have shown that the sodium carbonate of the foregoing
series of determinations must have retained traces of water. This calls for a correction of
+0. 022 in the atomic weight of copper as derived from the sodium carbonate ratio; and raises
it from 63.540 to 63.562. In the summing up at the end of this section this correction will be

applied.
180 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [vou. XVI.

. In ene more experiment the sulphuric acid was weighed as barium sulphate, the latte
being corrected for occluded ‘salts.: 3.1902 grm. CuSO,.5H,O gave 2.9761 BaSO,; hence
CuSO,.5H,O: BaSO, :: 100: 93.289. The sulphate contained 25.448 per cent of Cu; hence
BaSO,: Cu :: 93.289: 25.448, and Cu=63.676. As this ratio rests upon a single determination
needs no further consideration:here. © ©. 1:6 Ses:

-», In his third paper upon the atomic’ weight of copper,’ Richards studied the dibromide,
CuBr,. In preparing this.salt he used hydrobromic acid made from pure materials, and further
purified by 10 distillations. This was saturated with copper oxide prepared from pure elec-
trolytic copper, and the solution obtained was ‘proved to be free from basic salts. As the
crystallized compound was not easily obtained in a satisfactory condition, weighed quantities
of the solution were taken for analysis, in which, after expulsion of bromine by nitric and sul-
phuric acids, the copper was determined ‘by electrolysis. In other portions of solution the
bromine was precipitated by silver nitrate, and weighed as silver bromide. The first prelimi-
nary series of experiments gave the subjoined results, with vacuum weights as usual:

In 2% grammes of solution.

Cu. in ! AgBr.
0. 4164 Wg ' + 39, 4599
4164 «2, 4605
4164, ; 2.4605
4165. _ 2.4599

, Hence 2AgBr: Cu::100: 16.927 +.0013. . §
The second, also preliminary series, was made with more dilute solutions, and came out
as follows: ; IeFa ol esse Media x

In 25 grammes of solution.
Cu...» AgBr.

fv es 7
0.2619 1. 5478

. 26185 : a 1.5477
1.5479

 Hence'2AgBr: Cu‘:'100: 16.9194+.0012.
In the third series, two distinct lots of otystallized bromide were dissolved, and the solu-
tions examined in the same way: ee oor

Cu. AgBr. Ratio.
0, 2500 1.4771 16. 925

- 5473 3. 2348 16. 919

Mean, 16. 922+. 0020

In the final set of analyses, the materials used were purified even more scrupulously than
before, and the process was distinctly modified, as regards the determination of the bromine.
The solution of the bromide was added to a solution of pure silver in nitric acid, not quite
sufficient for complete precipitation. The slight excess of bromine was then determined by
titration with a solution containing 1 gramme of silver to the liter. Thus silver proportional
to the copper in the bromide was determined, and the silver bromide was weighed in a Gooch
crucible as before. The results are subjoined:

y In 50 grammes of solution.
Cu. ° bape oF Ag. AgBr.

0. 54755 1. 8586 3. 2350
; . 54750 1. 8579 3. 2340
fe x . 1,8583 3. 2348

‘Hence Cu: 2Ag:: 100: 339.392 + .0108; and 2AgBr: Cu:: 100: 16.927 +.0012.
_ The latter ratio, combined with the results of the three preceding series, gives a general
mean of - x

Hence Cu= 63.567.

2AgBr: Cu: : 100: 16.924 + .0007

1 Proc, Amer. Acad., 25,195, 1880.
No. 3.) ATOMIC WEIGHTS—CLARKE. 131

In his two earlier papers ' Richards determined the copper-silver ratio directly—that is,
without the weighing of any compound of either metal. By placing pure copper in an ice-cold
solution of silver nitrate, metallic silver is thrown.down,. and the weights of the two metals
are in equivalent proportions. In the first paper the following results are given. | The third
column gives the value of « in the ratio Cu: 2Ag: : 100:2. ,

Cu taken. Ag found. Ratio..
0. 53875 1. 8292 339, 527
- 56190 1. 9076 339, 491
1. 00220 3. 4016 339, 414
1. 30135 4.4173 339. 440
- 99870 3. 39035 339.477
1. 02050 3. 4646 339. 500

Mean, 339. 475+. 0114

In the second paper Richards states that the silver of the fifth experiment, which had
been dried at 150°, as were also the others, still retained water, to the extent of four-tenths
milligramme in 2 grammes. If we assume this correction to be fairly uniform, as the con-
cordance of the series indicates, and apply it throughout, the mean value for the ratio then
becomes 339.408+.0114. This procedure, however, leaves the ratio in some uncertainty, and
accordingly some new determinations were made, in which the silver, collected in a Gooch
crucible, was heated to incipient redness before final weighing. Copper from two distinct
sources was taken, and three experiments were carried out upon one sample to two with
the other. Treating both sets as one series, the results were as follows:

Cu taken. Ag found. Ratio. a
0. 75760 ’ 2613 | 339. 40

. 95040 3. 2256 339. 39

. 75993 2, 5794 339. 42
1. 02060 3. 4640 339. 42

. 90460 3.0701 339. 39

Mean, 339. 404+. 0046

a value practically identical with the corrected mean of the previous determinations, and with
that found in the later experiments upon copper bromide. Hence Cu =63.570.

In various electrical investigations the same ratio, the electrochemical equivalent of copper,
has been repeatedly measured, and the later results of Lord Rayleigh. and, Mrs, Sidgewick,?
Gray,? Shaw,‘ and Vanni > may properly be included in this discussion. As the data are some-
what differently stated, I have reduced them all to: the common standard. adopted. above.
Gray gives two sets of measurements, one made with large and the outer with small metallic

 

 

 

 

 

plates: poe
Rayleigh and 8. Gray 1. ‘Gray 2. Shaw. Vanni.
340. 483 341. 297 340..252 339. 68 340. 483
340. 832 341. 413 339. 674 340. 05 340. 600
340. 367 340. 815 340. 020 339. 84 340. 367
_— _ 840, 252 339.905 339. 71 340, 252"
340. 561, 339. 905 339.674 ~~ 840. 04 " 340. 600
+. 0935 341. 064 839. 328 » * 339.94 340. 136
340. 832 . 340, 136 340. 35
341. 297 340. 136 339. 82 340. 406
341. 064 340,.136 340. 09 +. 0520
341. 413 340..020 339. 84
340, 020 339. 90
340. 935 340. 136 339. 98
+. 1072 340. 14
339. 953 840. 56 ie ‘
. &. 0521. , 339.82 : 3
2 ¥ 839.983 gi v
+. 0411
1 Proc, Amer. Acad., 22, 346, and 23, 177. 1886 and 1887, ‘ British Assoc, Report, 1886, Abstract in Phil. Mag. (5), 23, 138,
2 Phil. Trans., 175, 458. 6 Ann, der Phys. (Wiedemann’s) (2), 44, 214.

® Phil, Mag, (5), 22, 389.
132 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The lack of sharp concordance in these data and the consequently high probable errors
seem to indicate a distinct superiority of the purely chemical method of determination over that
adopted by the physicist. This.supposition is strengthened by the electrochemical experiments
of Richards, Collins, and Heimrod, who precipitated copper and silver simultaneously in the
same current of electricity. Their first series, with vacuum weights, is as follows:

Weight Cu. Weight Ag. Ratio.
0. 44478 1.51064 339. 64:
.81645 1. 07473 339. 63
. 24968 . 84792 339. 60
1, 02186 8. 47056 339. 63
. 66166 2, 24538 339. 36
. 68027 2. 14050 339. 63
. 45919 1. 55905 339. 72
. 39177 1. 33071 _ 339. 67
1.111030: 3. 76990 339. 54
. 67564 2. 29655 339, 91
. 48232 1. 63768 339. 54
. 83092 _ 2, 82203 339. 63
. 68491 2.15735 339.79
. 70102 2. 37868 339. 32
. 84469 2. 86608 839. 55
5 . 87462 2.97114 339. 71
: . 69405 2. 35683 $39. 58

Hence Cu = 63.531.

In a second series. of experiments the copper was deposited from solutions saturated with

cuprous sulphate:

Weight Cu. Weight Ag. Ratio.
0. 71847 2. 43935 339. 52
. 71861 2. 43940 339. 46
- 72019 2. 44603 339. 64
- 97193 3. 30100 339. 63
. 50916 1. 72859 339. 50
. 76188 2. 58664 339. 51° -

 

Mean, 339. 615+. 0230

 

eae Mean, 339. 5434. 0200
Hence Cu= 63.544. — . i
In the foregoing series the temperature of the solution was 0°. Two experiments at higher
temperatures, 56°-61°, gave lower values for the ratio, and consequently a higher atomic weight

for copper: ae 3
Weight Cu. Weight Ag. Ratio.

0. 97295 3. 30100 339. 28
. 76214 2. 58664 339. 39

 

Mean, 339. 335+. 0370
Hence Cu =63.583. .

More recent is an electrochemical series of determinations by Gallo,? of slight importance.
The figures, with vacuum weights, are— j

Weight Cu. Weight Ag. Ratio.
0.21805 0. 73937 339. 083
27153 - 92062 339. 049
19001 64571 ' 339, 829
39585 1, 34578 339. 972

 

. Mean, 339. 483-4. 164
Hence Cu = 63.555.

The electrolytic comparison of copper and silver by simultaneous precipitation in vol-
tameter cells by Shrimpton’ is of exceptional value because of the care taken to determine and

 

1 Proc, Amer. Acad., 35, 123. 1899. 3Atti, Acad. Lincel (5), 14, 23. 1905. ® Proc. Physical Soc., 26, 292. 1914.
No, 8) ATOMIC WEIGHTS—CLARKE. | 133

apply all recognized corrections. The corrected weights are as follows, with the corresponding
ratio: se ;

Weight Cu. Weight Ag. Ratio.
0. 65307 2. 21666 339, 422
1, 27349 4. 32294 339. 456

. 64410 2. 18635. 339. 443
. 68262 2. 14751 339, 463
- 95574 3. 24431 ; $39. 455
. 67704 , 2. 29784 339. 397
1, 04851 3. 54245 339. 474
1. 15428 3.91820 339. 450
1. 42326 4, 83168 339. 480

. 71831 2. 43829 339. 448

_- Mean, 339. 449+. 0052
Hence Cu = 63.562.
Still another electrolytic comparison of the same two metals by Pecheux! gave Cu=63.43.
As no details as to weights or the number of experiments are given, this determination need not
receive any further attention.
The general combination of all data relative to the copper-silver ratio is as follows:

Richards, first series, corrected.................-0002--05- 339, 408+. 0114
Richards, second series..........-...0..0.0ee00e- ee eerie 839, 404+. 0046
Richards, CuBr, series...............202- 202 ee eee eee eee 339. 3924. 0108
Richards, Collins, and Heimrod, first............. eae ue as 339. 615+. 0230
Richards, Collins, and Heimrod, second.............-..- 339. 543+. 0200
Richards, Collins, and Heimrod, third. ........-.-..-.-- 339. 335+. 0370
Rayleigh and Sidgewick............-......-- setloaile's 5% 340. 561+. 0935
Gray, large plates. <.sscsces20s sseevecwarsecavess ieeasmnkes 340. 935+. 1072
Gray; small plated. secs cs.csvedeendeeed sciecidiecd sedan eee 339. 953+. 0521
BAW: se. gcays 0.5 aisicisienduainsarnee cra Avan cisiovaes eat Sek sie stcapamlainenichagins 339. 983+. 0411
Wait occa) co cri pu etiicucuen moves eta endweaetese cone ees 340. 406 +. 0520
Gallo coc kiut Sosrecceuh hig sean eevea ten S 339. 483+. 1640
BShrimiptens )s 22 nsceiex casero’ cuisamrenerd a ise daneeecemes: 339. 449+. 0052

General mean............2-.eeeen eee bia gieraieiaieronrs 839. 481+. 0037

We now have the following ratios from which to compute the atomic weight of copper. »

(1) Percentage of Cu in CuO, 79.842 + .0009

(2) Percentage of Cu in Cuso,, 39.795 + .0036

(3) Percentage of Cu in CuSO, 5H,O, 25.451 4.0011
(4) Percentage of CuO in CuSO,, 49.816 + .0017

(5) Cu : Na,CO, :: 100 : 166.838 + .0035

(6) Cu: ee 2: 100 : 223.525 +.0098

(7) 2AgBr : Cu :: 100 : 16.924+.0007

(8) Cu: QAg :: 100 : 339.431 + 008%.

To reduce these ratios we have—

Ag=107.880+.00016 ' C=12. 0025+. 00019
Cl=35. 4605+. 00011 S=32. 0650+. 00012
Br=79. 9228+. 00015 Na=23. 0029+. 00012
H=1 0078+.00001
Hence
From ratio—

Li gcieoan eae gare damien oe vac oMinee eee pee Cu=63.3730-+ .0025
We os ecalacaltare Ssitsidietawcle: wenn en Pewee eee stench ite 63.4779 + .0038
Besiescs iamedeuman ce cdiaee ui ns fice eaae ox tann en alauals -63.4982 + 0069
Bi sasaiidnmticns SaGaae aiiceine ap eotupees cane ss a eanpesus 63.5525 + .0029
Dict ieena unless wand spews ¢ ee mmneaiandss easements Dats 0028
6, corrected os ons sesevsevecess+-2csebesiesusonse + ---63,5631 + 0013
Bb. soachasadecatiniapes Weneangamacueanieede es pitsienn 63.5652: .00059
Tusiis vv siecieive asiwros voseupeucemeet<e ee feeeseesss. -63,5675-+ .0026

General mean......... ee benae eee cease +s CU==63,5550 4.00050

 

1 Compt. Rend., 154, 1919. 1912.
134 MEMOIRS NATIONAL ACADEMY :-OF SCIENCES. [Vou XVI,

-This-value is probably a little too low. If ratio. (1) is rejected, the general mean of the
remaining seven values becomes Cu = 63.5611, which is perhaps nearer the truth.

GOLD.

Among the early estimates of the atomic: weight of gold the only ones worthy of consider-
ation are those of Berzelius and Levol. a

The earliest method adopted by Berzelius! was that of precipitating a solution of gold
chloride by means of a weighed quantity of metallic mercury. The weight of gold thus thrown
down gave the ratio between the atomic weights of the two metals. Inthe single experiment.
which Berzelius publishes, 142.9 parts of Hg’ precipitates 93.55 or Au. Hence if Hg=200,
Au= 196.397.

In a later investigation? Berzelius aged to the analysis of potassium aurihcloride,
KCLAuCl,. Weighed quantities of this salt were ignited in hydrogen; the resulting gold and
Pete chloride were separated by x means of ‘water, and both were collected and estimated.

e loss of weight upon ignition ‘was, of course, chlorine. As the salt could not, be perfectly
dried without loss of chlorine, the atomic weight, under investigation must ‘be determined by
the ratio between the KCI and the Au. If we reduce to a common standard, and compare with
100 parts of KCl, the equivalent amounts of gold will be those which I give in the last of the
subjoined columns:

KAuCl,. wo KCl. or uk Au. Raito.
4.1445 -8185 : ‘2.159 263.775
2.2495 : -44425 Se D172 263.815
5.1300 1.01375 2.67225 263.600
3.4130 .674 1.77725 263.687
4.19975 .8295 2.188 , 263.773

 

Mean, 263.730-+.026
Hence Au = 196.63.

Still a third series of experiments by Berzelius * may be included here. In order to estab-
lish the atomic weight of phosphorus he employed that substance to ‘precipitate gold from a
solution of gold chloride in excess. . Between. the weight of phosphorus taken and the weight
of gold ‘obtained it was easy to fix a ratio. Since the atomic weight of phosphorus has been
better established by other methods,, we may, ‘properly reverse this ratio and apply it to our
discussion of gold. One hundred parts of P precipitate the quantities of Au given in the third
column:

P, Ot Ala to
0,829 8.714 : 1051.15

754 7.930 1051.73

oar aes Mean, 1051. 444, 196
Hence, if P=31, Au= 195.568. Ese

Levol’s ‘ estimation of the atomic weight ander scieiaeiation can hardly hava much value.
A weighed quantity of gold was converted in a flask into AuCl,, This was reduced by. a stream
of sulphur dioxide, and the resulting sulphuric acid was determined as BaSO,. One gramme
of gold gave 1.782 grm. BaSO,. Hence Au=196.49.

All these values may be neglected as worthless except that ddurived Festi Berzelius’ KAuCl,
series.

In 1886 Kriiss* published the first of the recent determinations of “ihe atomic weight
under consideration, several distinct methods being recorded. First, in a solution of pure
auric chloride the gold was precipitated by means of aqueous sulphurous’ acid. In the filtrate
from the gold the chlorine was thrown down as silver chloride, and thus the ratio Au : 3AgCl

 

1Poggend. Annalen,8, 177. Si ea sath hp SHR.
'. §Lehrbuch, 5 Aufl., 3, 1212.
. 3 Lehrbuch, 5 Aufl, 3, 1188.
(, \; Ann. Chim. Phys. (3), 30,355. 1850.
's Untersuchungen tiber das Atomgewicht des Goldes. ‘Miincheni, 1848. 112 pp., 8vo.

ME hel oy aba!
No. 8.] ATOMIC WEIGHTS—CLARKE. 135

was measured. I subjoin Kriss’ weights, together with a third column giving the gold equiv-
_alent to 100 parts of silver chloride:

Au, ‘AgCl. Ratio.
7.72076 16.84737 45.828
5.68290 1240425 45.814
3.24773 7.08667 45,828
4.49167 9.80475 45.811
3.47949 7.59300 45,825
3.26836 7.13132 45.832
5.16181 1.26524 45.821
4.86044 10.60431 45.834

Mean, 45.824 .0020
Hence Au = 197.05. *

The remainder of Kriss’ determinations were made with potassium auribromide, KAuBr,,
and with this salt several ratios were measured. The salt was prepared from pure materials,
repeatedly recrystallized under precautions to exclude access of atmospheric dust, and dried
over phosphorus pentoxide. First, its percentage of gold was determined, sometimes by
reduction with sulphurous acid, sometimes by heating in a stream of hydrogen. For this ratio
the weights and percentages are as follows, the experiments being numbered for further refer-
ence, and the reducing agent being indicated:

KAuBr,. Au. Per cent.
I, BOs cecuonae pholeyeceeachinicnadiatels 10.64821 3.77753 35.476
@., BO pnssncces comp hne iaiemmbeeaivaamaceaGes 4.71974 1.67330 35.453
BN dat ea ceactaktrine ahaha ar ieee 7.05762 2.50122 35.440
BE catiibolinspneiailed date neennaee hah sade se 4.49558 1.59434 35.465
BO aces gksensoncetes ee ee 8.72302 3.09448 35.475
By BO ponasadecexenies wacasbdevaillbalcleie pabtelaiaheiar 7.66932 2.71860 35.448
7. Bly... cennsncdens xpiemasbonnnees secant 7.15498 2.53695 35.457
Be ie ct acca hee patil aueeneoaneanansehe 12.26334 4.34997 35.471
9. Hid wah a Sinianod a scaaynce acs ee web dees 7.10342 2.51919 35.465.

; Me , 35.461. A61-+ .0028
Hence Au= 197.14. ois

‘In five of the foregoing experiments the reductions were effected with sulphurous acid;
and in these, after filtering off the gold, the bromine was thrown down and weighed as silver
bromide. This, in comparison with the gold, gives the ratio Au : 4AgBr : : 100 : z:

Au. 4AQgBr. Ratio.
Doe side eeaeead oie Abin eee cymes 3.77753 14.39542 381.080
Ds scp thd cioaiih a Saas it tetas Samenwaiioaranes 1.67330 6.37952 381.254
BF apes sere Riayat eed Schein wha Seen eher cio d Gic teqsialarscoratetciats 3.09448 11.78993 380.999
Gua siecezasceketiccesecienmetakenweniann masses 2.71860 10.35902 381.042
asia exlrsainseseytdnsva)atsis. eisjaseereiores erase’ wi2ia sims MSI OTE SS 2.53695 9.66117 380.731

Mean, 381. 021-4. 057
Hence Au:: AgBr : : 100 : 95.255+.0142, and Au=197.16.

In the remaining experiments, Nos. 3, 4, 8, and 9, the KAuBr, was reduced in a stream of
hydrogen, the loss of weight, 3 Br, being noted: In the residue the gold was determined, as
noted above, and the KBr was also collected and weighed. The weights were as follows: ;

Au. Loss, $ Br. KBr.
cons deta a acsearle hao ew atte eo 2.50122 3.04422 1.51090
Mis oe, saslela pied. ciclavancratercte Sats Avestan SOG Se 1.59434 1.93937" ° .96243
Simcudes Yavsomenaeeew es eememuse dames earcelestans 4.34997 5.29316 2.62700
eae ectalzcacd:xie vecaeraccroterep oie wie SS RRO Res ek ee eee 2.51919 3.06534 1.52153
From these data we obtain two more ratios, viz, Au : Br, :: 100 : 2, pad Au : KBr : : 100:2,
thus: ; SBr. Au:KBr.
Tle Oe esa Tinea 121.710 60.405
ude Sapa narteaeeeccta ss cae 121.641 60.365
AT oath in sissy ahd RN eae 121.683 60.391
ie iiceasacaaean as I shu lpeama nee cise imnions! A PIIERO 60.398

 

Mean, 121.678+.0100 Mean, 60.390-£.0059

Hence Au= 197.05, and 197.08. _
136 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

_ From all the ratios, taken together, Kriiss deduces a final value of Au=197.13, if. O=16.
It is obviously possible to derive still other ratios from the results given, but to do so would: be
to depart unnecessarily frum the author’s methods as stated by himself.

Thorpe and Laurie,t whose work appeared shortly after that of Kriiss, also made use of the
salt KAuBr,, but, on account of difficulty in drying it without change, they did not weigh it
directly. After proving the constancy in it of the ratio Au : KBr, even after repeated crystal-
lizations, they adopted the following method: The unweighed salt was heated, with gradual
increase of temperature, up to about 160°, for several hours, and afterwards more strongly over
a small Bunsen flame. This was done in a porcelain crucible, tared by another in weighing,
which latter was treated in precisely the same way. The residue, KBr+Au, was weighed, the
KBr dissolved ‘out, and the gold then weighed separately. The weight of KBr was taken by
difference. The ratio Au: KBr::100.:«¢appearsinathirdcolumn:.. , .

Au, KBr. ‘Ratio. |

6. 19001 , 3.73440 60. 329
4. 76957 2. 87715 60. 323
4. 14050 2. 49822 60. 336
3, 60344 2.17440 60. 342
3, 67963 2. 21978 60. 326
4, 57757 2.76195 60. 337
5. 36659 8, 23821 60. 326
5.16406 3. 11533 60. 327

Mean, 60. 331+. 0016.
Hence Au = 197.28.

This mean combines with Kriiss’s thus:
VUES 350 ee ee Sade les eS se eyeee sees sesso ce ae osenees 60. 390+. 0059
Thorpe and Laurie...........--.- 60. 331+. 0016
General mean : 60. 338+. 0015

The potassium bromide of the previous experiments was next titrated with a solution of
pure silver by Stas’s method, the operation being performed in red light.. Thus we get the
following data for the ratio Ag : Au : : 100 : 2, using the weights of gold already obtained:

 

Ag. Au ‘ Ratio. : :
8, 38451 6. 19001 . 182, 893
2. 60896 4. 76957 "182, 818
2, 28830 4, 18266 182. 786
2. 26415 4: 14050 182. 868
1.97147» 8. 60344 182. 775
2. 01292 3. 67963 182. 801
2. 50334 4; 57757 182. 863
2.93608 - 5. 36659 182. 780

. 2, 82401 5. 16406 182. 865

Mean, 182.827 --. 0101
Hence Au= 197.24. |

Finally, in eight of these experiments, the silver bromide formed during titration was col-
lected and weighed, giving values for the ratio Au : AgBr : ::100 : z, as follows:

Au, AgBr. Ratio.
6. 19001 5. 89199 95. 186
4. 76957 F 4, 54261 95, 242
4. 18266 3. 98288 95. 224
4. 14050 3. 94309 95, 232
3. 60344 3. 48015 95. 191
3. 67963 8. 50207 95.175.
4. 57757 4. 35736 95. 189
5.36659 5. 11045 95. 227

Mean, 95. 208 +. 0061
Kriiss found, 95. 255 +. 0142

4 a8 General mean, 95. 222+. 0056
_ From Thorpe and Laurie’s mean, Au= 197.25.

Beh

1 Journ. Chem. Soc., 51, 565. 1887.
No.3), ATOMIC WEIGHTS—CLARKKE. 137

From the second and third of the ratios measured by Thorpe and Laurie an independent
value for the ratio Ag : Br may be computed. It becomes 100 : 74.072, which agrees fairly
with the direct determinations made by other chemists. Similarly, the ratios Ag : KBr and
AgBr : KBr may be calculated, giving additional checks upon the accuracy of the manipulation,
though not upon the purity of the original material studied.

Thorpe and Laurie suggest objections to the work done by Kriiss, on the ground that the
salt KAuBr, can not be completely dried without loss of bromine. This suggestion led to a
controversy between them and Kriss, which in effect was briefly as follows:

First, Kriss ! urges that the potassium auribromide ordinarily contains traces of free gold,
not belonging to the salt, produced by the reducing action of dust particles taken up from the
air. He applies a correction for this supposed free gold to the determinations made by Thorpe
and Laurie, and thus brings their results into harmony with his own. To this argument Thorpe
and Laurie ? reply, somewhat in detail, stating that the error indicated was guarded against by
them, and that they had dissolved quantities of from 8 to 19 grammes of the auribromide with-
out a trace of free gold becoming visible. A final note in defense of his work was published by
Kriss a. little later.?

In 1889 an elaborate set of determinations of this constant was published by Mallet,* whose
experiments are classified into seven distinct series. First, a neutral solution of auric chloride
was prepared, which was weighed off in two approximately equal portions. In one of these the
gold was precipitated by pure sulphurous acid, collected, washed, dried, ignited in a Sprengel

vacuum, and weighed. To the second portion a solution containing a known weight of pure
silver was added. After filtering, with all due precautions, the silver remaining in the filtrate
was determined by titration with a weighed solution of pure hydrobromic acid. We have thus
a weight of gold, and the weight of silver needed to precipitate the three atoms of chlorine com-
bined with it; in other words, the ratio 3Ag :Au::100:2. All weights in this and the sub-
sequent series are reduced to a vacuum, and all weighings were made against corresponding
tares:

Au, Ag. Ratio.
7. 6075 * 12. 4875 60. 921
8. 4212 13. 8280 60. 900
6. 9407 11. 3973 60. 898
3. 3682 5. 5286 60. 923
2, 8244 4. 6371 60. 909

 

Mean, 60. 910-+. 0034
Hence Ag : Au : : 100 : 182.730+.0102, and Au=197.13.

The second series of determinations was essentially like the first, except that auric bromide
was taken instead of the chloride. The ratio measured, 3Ag : Au, is precisely the same as before
Results as follows:

Au. Ag. Ratio.

8. 2345 18. 5149 60. 929
7. 6901 12. 6251 60. 911
10. 5233 17. 2666 60. 945
2. 7498 4.5141 60. 916
3. 5620 5. 8471 60. 919
3. 9081 . 4129 60. 941

 

Mean, 60. 927+. 0038
Hence Ag : Au: : 100 : 182. 781+. 0114, and Au=197.18.

In the third series of experiments the salt KAuBr, was taken, purified by five cGonyetalli-
zations. The solution of this was weighed out into nearly equal parts, the gold being measured
as in the two preceding series in one portion, and the bromine thrown down by a standard silver
solution as before. This gives the ratio 4Ag : Au : : 100 : 2.

1 Ber. Deutsch. chem. Gessell., 20, 2365. 1887. i 3 Berichte, 21, 126, 1888.
3 Berichte, 20, 3036, and Journ. Chem, Soc., 51, 866. 1887. 4 Philosophical Transactions, 180, 395. 1889.

 
138 MEMOIRS NATIONAL ACADEMY .OF SCIENCES. [Vou. XVI.

Au. Ag. .. Ratio.
5. 7048 12, 4851, 45, 693
7.9612 17. 4193 45, 693
2, 4455 6, 3518 45. 699
4, 1632 9, 1153 45. 673°

 

Mean, 45, 689+. 0040

Hence Ag : Au : : 100 : 182.756 +.0160, and Au= 197.16.

The fifth series of determinations, which for present purposes naturally precedes the fourth,
was electrolytic in character, gold and sliver being simultaneously precipitated by the same
current. The gold was in solution as potassium.aurocyanide, and the silver in the form. of
potassium silver cyanide. The equivalent weights of the two metals, thrown down in thesame
time, were as follows, giving directly the ratio Ag : Au: : 100 : a:

Au, ’ Ag. Ratio.
5. 2721. 2. 8849 182. 748°
6. 3088 3. 4487 182. 933
4, 2770 2. 3393 182. 832.
3. 5123 1. 9223 182.713
3. 6804 2. 0132 182. 814 ae

Mean, 182. 808-4. 0256
Hence Au= 197.22.

This mean may be combined with the preceding means, and also with the determination
of the same ratio by Thorpe and Laurie, thus:

Thorpe and Laurie. seewadshundneidemeeed es oeleieia isialemieler 182, 897 is 0101
Mallet, chloride series...........0.22000.ceeedeeececeeeees 182. 730+. 0102
Mallet, bromide series............-..0eeeceeeeeeet Eacpaaioes 182. 7814. 0114
Mallet, KAuBr, series.............2000ceeeeeeeeceeceeees 182. 756+. 0160
Mallet, electrolytic...........0-..ec2-ceeeeee dectwamske , 182. 808 +. 0256

General Mean wscesccesesessssseseweverseassaene ws 182. 778 x. 0055

In Mallet’s fourth series a radically new method was employed. Trimethyl-ammonium
aurichloride, N(CH,),HAuCl,, was decomposed by heat, and the residual gold was determined.
In order to avoid loss by spattering, the salt was heated in a crucible under a layer of fine siliceous
sand of known weight. Several crops of crystals of the salt were studied, as a check against
impurities, but all gave concordant values.

Salt. Residual Au. . ‘ Per cent Au. .
14, 9072 7.8754, 49, 475

15. 5263 7.6831 49, 484

10. 4523 5.1712 © "49, 474°

6. 5912 3. 2603 49, 464
(5.5744 2.'7579 49. 474

 

Mean, 49. a4. 0021
Hence Au= 197.73.

In his sixth and seventh series Mallet sought to establish, by direct measurement, the ratio
between hydrogen and gold. In their experimental details his methods are somewhat elaborate,
and only the processes, in the most general way, can be indicated here. First, gold was pre-
cipitated electrolytically from a solution of potassium aurocyanide, and its weight was com-
pared with that of the amount of hydrogen simultaneously liberated in a voltameter by the
same current in thesame time. The hydrogen was measured, and its weight was then computed
from its density. The volumes are given, of course, at 0° and 760 mm.

We. Au. Vol. H, cc. We. H.
4.0472 228, 64 0. 0205483
4. 0226 227. 08 . 0204046

4, 0955 231. 55 . 0208103
No. 3.] - ATOMIC WEIGHTS—CLARKE. 139

. These data, with the weight of one litre of hydrogen taken as 0.89872 gramme, give the
subjoined values in the ratio H : Au: :1 : 2.

196. 960
197.151
196. 805

Mean, 196. 972+. 0675

 

In the last series of experiments a known quantity of metallic zinc was dissolved in dilute
sulphuric acid, and the amount of hydrogen evolved was measured. Then a solution of pure
auric chloride or bromide was treated with a definite weight of the same zinc, and the quantity.
of gold thrown. down was determined. The zinc itself was purified by fractional distillation in
a Sprengel vacuum. From these data the ratio 3H : Au was computed by direct comparison
of the ee of gold : and that af the liberated hydrogen. The results were as follows:

Wi. Au. Vol. H, cc. Wt. A.

10. 3512 ; 1756. 10 . 157824
8. 2525 . 1400.38 - 125857
8. 1004 1374. 87 . 123565
3. 2913 558. 64 - 050206
3. 4835 590. 93 - 053109
‘3. 6421 618.11 . 055551

Hence for the ratio 3H: Au: 1:2 we have:

65. 587 65. 556
65. 571 65. 593
65. 557 65. 563

' Mean, 65. 571+. 00436

And H : Au ::1 196.713.0131. This, combined with the value found in the preceding
series, gives @ general mean of 196.722 +.0129.
The ratios available for gold are now as follows:
(1) KCl : Au :: 100 : 263.730+0. 26
(2) 3AgCl : Aus: 100 : 458.2440. 020
(3) KAuBr, : Au :: 100 : 35.4614 0. 028
(4) Au: AgBr :: 100 : 95.222 + 0. 056
(5) Au : 8Br :: 100 : 121.678+0. 100.
(6) Au: KBr ::.100 : 60.338+0. 015
(7) Ag : Au :: 100: acta 55
(8) NO;HyAuCl, : Au :: 100 :49.474+40. 021
f, (Be Anwds es 722 £0. 129
- The antecedent atomic weights are—

Ag=107. 880 +. 00016 N=14. 0081+. 00008
Cl = 36. 4605+. 00011 K=89. 0795+. 00011
Br= 79. 9228+. 00015 C =12. 0025+. 00019

H=1. 0078+. 00001

Hence
; From ratio— io ly at

Veosceiardneccodin e's sees eee 2 s'es ctse Reise’ Au=196. 632+. 0194

Ss 197. 052+. 0115

197. 053+. 0086

197. 137+. 0178

197. 182+. 0059

197. 226+. 0128

197. 253+. 0049

197. 7314. 0117

198. 256+. 0130

General mean.......-..2.-2.-eeeeeeeeeeeces -Au=197. 209+. 0029

 
140 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

‘Although this final value is not altogether: cae ‘it seems eens that uke. atomic
waits of gold is not far from Au= 197.2. en

The ninth or last value in the foregoing series teprbeenis Mallets radi between, gold and
hydrogen, and is peculiarly instructive. In Mallet’s paper the several ratios determined were
discussed upon the basis of O= 15.96, referred to hydrogen as unity. This, on the oxygen scale,
is equivalent to H=1.0025. On that basis the determination in question agreed well with the
others; but with H= 1.0078, the present value, it is enormously raised. The former agreement
between the several series of gold values was therefore only apparent, and shows that concord-
ance among determinations may be only coincidence, and no real proof of accuracy. It is prob-
able, furthermore, that direct comparisons of metals with hydrogen cannot give good measure-
ments of atomic weights, for several reasons. First, it is not possible to be certain that every
trace of hydrogen has been collected and measured, and any loss tends to raise the apparent
atomic weight of the metal studied; secondly, the weight of the hydrogen is computed from its
volume, and a slight change in the factors used in reduction of the observations may make a
considerable difference in the final result. These uncertainties exist in all determinations of
atomic weights hitherto made by the hydrogen method.

CALCIUM.

Much of the older work on the atomic weight of calcium, including the earliest determina-
tions by Berzelius, may be disregarded as having no present value. Baup’s’ analyses of organic
salts of calcium are interesting, but carry no weight now. ‘They led to the value Ca=39.98.
As for Salvétat’s ? determination, that was merely given as a statement of results, without such
details as would make his work available for discussion.

An important factor in measuring the atomic weight of calcium is the composition of
calcium carbonate, as determined by several investigators. This will be considered first, and
the determinations based upon other calcium compounds follow later.

In 1842 Dumas * made three ignitions of Iceland spar, and determined the percentages of
carbon dioxide driven off and of lime remaining. The impurities of the material were also
determined, the correction for them applied, and the weighings reduced to a vacuum. His
figures are as follows:

CaCO. Cad. Per cent.
49. 916 28. 016 56. 12
50. 497 28. 305 56. 04
64. 508 36. 167 56. 06

Mean, 56. 073+.016
Hence, Ca=40.169.

About this same time Erdmann and Marchand‘ began their researches upon the same
subject. Two ignitions of spar, containing 0.04 per cent of impurity, gave, respectively, 56.09
and 56.18 per cent of residue; but these results are not exact enough for us to consider further.
Four other results obtained with artificial calcium carbonate are more noteworthy. The care
bonate was precipitated from a solution of pure calcium chloride by ammonium carbonate,
was washed thoroughly with hot water, and dried at a temperature of 180°. With this prepa-
ration the following percentages of lime were obtained:

56. 03 56. 00
55. 98 55. 99

Mean, 56. 00-£.007

Hence, Ca=40.003.

It was subsequently shown by Berzelius that calcium carbonate prepared by this method
retains traces of water even at 200°, and that minute quantities of chloride are also held by it.
These sources of error are, however in opposite directions, since one would tend to diminish
and the other to increase the weight of residue.

 

1 Bull. Universelle des Sciences de Genéve, 39, 347. 1842. §Compt. Rend., 14, 547. 1842.
*Compt. Rend.,17, 318. 1843. ‘Journ. prakt. Chem., 26, 472. 1842.
No. 8.1: . ATOMIC WEIGHTS—CLARKE. 141

In the same paper there are also two direct estimations of carbonic acid in pure Iceland
spar, which correspond to the following percentages of lime:

56. 00
56. 02

Mean, 56. 01+ .007

In a still later paper‘ the same investigators give another series of results based upon the
ignition of Iceland spar. The impurities were carefully estimated, and the percentages of lime
are suitably corrected:

CaCO;. CaO. ‘Per cent.
4, 2134 2. 3594 55. 997
15. 1385 8. 4810 56, 022
» 28. 5503 13. 1958 56. 031
23. 6390 18. 2456 56. 032
42. 0295 23. 5533 56. 044

49. 7007 a 27, 8536 _ 56. 042

Mean, 56. 028 +.0047

Hence, Ca = 40.067.

Six years later Erdmann and ‘Marchand? published one more result, upon the ignition of
calcium carbonate. They found that the compound began giving off carbon dioxide below
the temperature at which their.previous samples had been dried, or about 200°, and that, on
the other hand,. traces of the dioxide were retained by the lime after ignition. These two
errors do not compensate each other, since both tend to raise the percentage of lime. In the
one experiment now under consideration these errors were accurately estimated, and the needful
corrections were applied to the final result. The percentage of residual lime.in this case was
55.998. This agrees tolerably well with the figures found in the direct estimation of carbonic
acid, and, if combined with those two, gives a mean for all three of 56.006 + .0043.

‘Hence, Ca=40.017..

‘Herzfeld,’ in his deteritinataons of atomic weight, made use of artificial calcium carbonate.
The lime was prepared by ignition of the oxalate, and then converted into bicarbonate by
treatment with solid carbon dioxide and water under pressure. By heating in a silver dish
the bicarbonate was converted into the normal salt, which was weighed, and then ignited at a
temperature of 1,300°-1,400°.. The data are as follows:

CaCO; bs Cad. . Per cent.
3. 9772 2. 2268 55. 989
2. 3614 1. 3218 55. 975
3. 2966 1. 8456 55. 985

 

Mean, 55. 983+ .0028

Hence, Ca = 39.965.

Hinrichsen,‘ in his two separate communications, gives analyses of spar from two distinct
localities, namely, Iceland and the Crimea. In each case very small quantities of impurity
were present, which were carefully determined and corrected for. The spar, previously freed
from all traces of moisture, was ignited in an electric furnace, at a temperature between 1,200°
and 1,400°. The results obtained, with all corrections applied and vacuum weights, are sub-
joined:

ee First series.

Ca CO; CaO. Per cent.
80.72157 17. 22354 _ 56. 0633
32..77791 |. 18. 375587 56. 0617
34, 45625 19. 31698 56. 0623

 

33. 36885 18. 70723 56. 0620

1 Journ..prakt. Chem., 31,269. 1844. -

Journ. prakt.Chem., 50,237. 1850.

' Zeitschr. Ver. Riibenzucker Industrie, 47,497. 1897. Partly reproduced in Ber., 34,559. 1904.
4 Zeitsohr. physikal. Chem., 39, 311, 1901; and 40,747, 1902.
142 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

Second series.

CaCO. CaO. Per cent.
31. 20762 17. 49526 59. 0608
22, 00588 12, 33642 56. 0602

 

Mean of both series as one, 56. 0617 + .0003
Hence, Ca=40.144.

Combining all these determinations, we have for the percentage of CaO from Cale

DUMAS J45 fect oc doa das aid Bonn dace nee cea ck ceeeececeetes 56.073 +. 016
Erdmann and Marchand, 1............-...-.----------ee: 56.000 +. 007
Erdmann and Marchand, 2.............-.-----2-2e2ee0-++ 56.028 +. 0047
Erdmann and Marchand, 3.............-.--...-22--0e22- 56.006 +. 0043
HGPziel@ 2 je.eeu tee ieite aie 2.55 vee o au cioequaiactlanersemeeeedte 55.983 +. 0028
HMPiChSeN ood oes av sis side's ie Ae daeeuee cas eRe OER 56. 0617 +. 0003
General ean sonciis 22s Pe enna cieevaineee skevienen 56. 0603+. 0003

The effect of this combination is practiéally to discard all of the determinations except
that of Hinrichsen. Herzfeld’s figures are certainly too low, and probably because of unde-
termined impurity in his artificial carbonate. The extreme difficulty of preparing absolutely
pure compounds of calcium is well known.!

In the earliest of the three papers by Erdmann and Marchand there is also given a series
of determinations of the ratio between calcium carbonate and sulphate. Pure Iceland spar
was carefully converted into calcium sulphate, and the gain in weight noted. One hundred parts

of spar gave of sulphate:
136. 07 136. 02

: 136. 06 . 136. 06
s "Mean, 136. 0525+. 0071
Hence Ca = 40.025.

In 1843 the atomic weight of calcium was redetermined by Berzelius,? who investigated
the ratio between lime and calcium sulphate. The calcium was first precipitated from a pure
solution of nitrate by means of ammonium carbonate, and the thoroughly washed precipitate
was dried and strongly ignited in order to obtain lime wholly free from extraneous matter.
This lime was then, with suitable precautions, treated with sulphuric acid, and the resulting
sulphate was weighed. Correction was applied for the trace of solid impurity contained in the
acid, but not for the weighing in air. The figures in the last column represent the percentage
of weight gained by the lime upon conversion into sulphate:

Ca. S03. Per cent gained.
1, 80425 2. 56735 142, 295
2. 50400 3. 57050 142, 592
' 3. 90000 5. 55140 142, 343
3. 04250 4) 32650 142. 202
3. 45900 4. 93140 142. 567

 

Mean, 142. 3998+. 0518
Hence Ca= 40.225.

The atomic weight of calcium has been several times computed from analyses of the chlo-
ride. The earliest determination by Berzelius * was based upon this compound, and Marignac ‘
also used it in some provisional experiments, to which, however, he assigns little importance.
They gave values for Ca far in excess of the truth. Dumas ‘® also published a series of deter-
minations of more than questionable value. Supposedly pure calcium chloride was first ignited
in a stream of dry hydrochloric acid, and the solution of this salt was afterwards titrated with
a silver solution in the usual way. The CaCl, proportional to 100 parts of Ag is given in a
third column:

 

1 See Stas, Oeuvres Complates, 3, 337. 4 Oeuvres Complétes, 1, 90.
2 Journ. prakt. Chem., 31, 263. Ann. Chem. Pharm, 46, 241. 6 Ann. Chim. Phys. (3), 65,129. 1859. Ann. Chem. Pharm., 113, 34.
3 Poggend. Annalen, 8, 189.
No. 3] ATOMIC WEIGHTS=CLARKE; |) 50) (0114 143
CaCl. 5 Ag. . Ratio. ,
2,738 | , 5. 309, 51.573
2.436 | “4,731 | 51. 490
1. 859 3.617 ° 51. 396
2.771 | ' 6, 3885! 51.424 |
2. 240 4, 3585 51. 394
cay 9% soa Y Mean, 51, 455 4554+. 0230

Hence Ca=40. 099. Let ip

Better results were obtained by Binkavdist “Bieien chloride;: purified by repeated crys-
tallizations, and fused in a stream of nitrogen, was precipitated by a solution of silver, and the
silver chloride so uaa was weighed. The data, with vacuum weights, are subjoined:

"CaCl; AgCl Ratio,
1. 56454 4, 0409 38. 7177
3. 57630 9. 2361' | I-38: 7209
» 3.59281 : : 9.2788 (88. 7206

5. 00880 12. 9364 38.7187 =
ae 23. 2506 38. 7197

Mean, 38. 7195+. 0004

HREt -

Hence Ca= 40.080.

The ratio between calcium chloride and. silver, previously determined by Dumas, has also
been measured by Richards and Honigschmid, 2-using the a i Harvard methods.

Their weights, reduced to a vacuum, and'the usual ratio follow:

ier Ag. Ps: CaCl, Ratio.
8. 94908 4, 60350 51. 4410 0
_ 9.37780 4 , 4.82401 , 51. 4407
9. 36688 4. 81846. 51. 4415 |
10. 29911 5. 29799 51. 4412
10. 50832 +! ' 4, 40550 51. 4402 “
10. 19715 5. 24539 51. 4398 -
10. 38328 t-5. 34110 wot o> € BL 4394

Hence Ca= 40.067.

wo
It is not necessary to combine Dumas’ determination with flit, for its high ‘probable error
Its influence in combination with the new determinations affects only the fifth

rejects it.
decimal place.

In another memoir Richards and Honigschmid * fopart determinations ee on analyses
Two series are given, for two ratios, with vacuum weights, as follows:

of calcium bromide.

Aya}, er

First, for the ratio 2Ag: CaBr, ::100:x.

Ag. hs
4.54252
4.95025 —
5. 77301
7. 81126

6s 504779 ve

* 8. 00455: fi:
+ AEA OF

Hence, Ca= 401056. 4

Purp
Tih My

CaBr,.

4. 20860
yo 4. 58644
5. 34866
7. 23724
4. 67673
_7. 41636

Second, for the ratio ‘2AgBr,: CaBr, ::100:z.

‘s » AgBni ;
19. 13778 |.
. 14. 88810
12. 73961 «
_ 1213702 ,
? 11. 18684
9. 69513

, aed i

Hence, Ca= 40.064.

CaBr,.

10. 18591

7. 92400

6. 78048

6. 45970.

5. 95390!
5.15998 |

ater

. ~ Magn, 51. 4405+-0002, ;

tr

Aon eT Be

.,, Ratio.

. 92. 6490
92. 6507
92. 6491

92. 6514 1

92. 6491
92. 6518

Mean, 92. 6502+. 00043

a

Ratio.
53. 2241
58. 2237
' 53. 2236
53. 2231
53. 2225
58. 2225

., Mean, 58. 2232 2232+. 00018 -

\

 

1 Journ. Amer. Chem. Soc., 24, 374. 1902.
161566°—20——_10

2 Journ. Amer. Chem. Soc., 33, 128. 1911.

3 Journ. Amer. Chem. Soc., 32, 1577. 1910.
144 MEMOIRS NATIONAL ACADEMY OF SCIENCES. Vou, XVI.

These bromide ratios also give the cross ratio Ag: Br::100: 74.079.
The determinations of the atomic weight of calcium by Oechsner de Coninck were published
in two short notes, without adequate details. In his first note’ he gives analyses of calcium
formate in which, by simple ignition, the proportion of calcium oxide was determined, The
data are as follows:

CaC,H,0, CaO. Per cent CaO.
0. 7123 0. 3074 43. 156
» 2374 we - 1021 43. 008
- 891 - 3826 42,941
2. 8493 1, 2254 43. 007

 

: Mean, 43. 028-4. 0307
Hence, Ca=39.904.

There is also a single analysis of calcium carbonate, of no value.
The second note? by Oechsner de Coninck deals with the conversion of calcium carbonate
into sulphate. His weights and the corresponding ratio are as follows:

- CaCO,. CaS0,. Ratio.
0. 3300 0. 4486 135, 939
. 2640 . 3590 135. 985

. 3630 . 4935 135. 950

. 2541 . 3454 135. 931

. 1524 . 2072 185. 958

 

Mean, 135. 953+. 0063
Hence, Ca = 40.303.

For the same ratio Erdmann and Marchand found the value 136.0525+.0071. The two
series combined give a general mean of 135.999 + .0043.
There are now eight independent ratios from which to compute the atomic weight of
calcium, as follows:
(1) CaCO, : CaO :: 100 : 56.0603 + .0003.
(2) CaO : SO, :: 100 : 142.3998 4.0518.
(3) CaCO, : CaSO, :: 100 : 135.999 + .0043.
(4) CaC,H,O, : CaO :: 100 : 43.028 +.0307.
(5) 2Ag : CaCl, :: 100 : 51.4405 + .0002.
(6) 2AgCl : CaCl, :: 100 : 38.7195-+.0004.
(7) 2Ag : CaBr, :: 100 : 92.6502 + 00043.
(8) 2AgBr : CaBr, :: 100 : 53.2232 + .00018.
To reduce these ratios we have—

Ag=107. 880+. 00016 Br=79. 9228-b. 00015
Cl=35. 4605+. 00011 S=32, 0650+. 00012
C=12, 0025-4. 00019

Hence,
From ratio—

BANG SS LE Eee Agar tenant ae peer Ca=39. 9040+. 0500

OF ad hada seat that na aA A ri oa Mt 40. 0564-4. 0010

Bia tee LG Ang ty Gi ee da ait ded 40. 0639+. 00073

Bi cetsaccosseic: ogee peeks ay gi etciees Muara 40. 0670-4. 00042

GC iGwcawsaads lates iat ssdnwiadenasmmouaedetos 40208044:.'0018

ESL cacenenseaedeneastatetasas aang cn santane 40. 1404+. 00053

Boa siiaiess Sa Sadat axumtuey peanonwesensnaws 40. 1722+. 0122

Beene haa! “dehy eaieige wsiastoe apt gra tonal pies 40. 22544. 0200
General mean........... plates cease terse Sere Ca=40. 0902+. 00028

Ratios 2, 3, and 4 are evidently worthless. Their rejection, however, changes the general
mean by only 0.0001. In short, they reject themselves.

 

1 Compt. Rend., 153, 1479. 1911. 2 Bull. Acad. Belg., 1913, 222,
No. 3.] ’ ATOMIC WEIGHTS—CLARKE. 145

STRONTIUM.

The ratios which fix the atomic weight of strontium resemble in general terms those
relating to barium, only they are fewer in number and represent a smaller amount of work.
The early experiments of Stromeyer,' who measured the volume of CO, evolved from a known
weight of strontium’ carbonate, are hardly available for the present discassion. So, also, we
may exclude the determination by Salvétat,? who neglected to publish sufficient details.

Taking the ratio between strontium chloride and silver first in order, we have series of
figures by Pelouze, Dumas, Marignac, Richards, and Thorpe and Francis. Pelouze * employed
the volumetric method to be. described under barium, and in two experiments obtained the
subjoined results. In another column I append the ratio between SrCl, and 100 parts of silver:

SrCly. Ag. Ratio.
1. 480 2,014 73, 486
2.210 _ 2,008 73. 471

Mean, 73. 4781+. 0050
Hence, Sr = 87.615.

Dumas,‘ by the same general method, made sets of experiments with three samples of
chloride which had previously been fused in a current of dry hydrochloric acid. His results,
expressed in the usual way, are as follows:

Series A.
SrCh. Ag. Ratio.
3. 137 4. 280 73, 2944
jo A 982 2. 705 73. 2717
3.041 4.142 73. 4186
3,099 4, 219 78. 4534
Mean, 73. 3595
Series B.
' Srl. eR Age ‘Ratio.
3.356 / 4,574 . 78,3713.
6, 3645, 8.667 -, 73, 4327
7.181. 9.718 >, 73. 4246
Mean, 73. 4095
Series C.
SrCil,.. Ag. Ratio.
7.213. 9.811 . 78. 5195
2. 206 8. 006 73. 3866
4. 268 5. 816 73. 5529.
4.018 5.477 73. 3613
Mean, 73. 4551

Mean of all as one series, 73. 4079+. 0170
Hence, Sr = 87.464.

The foregoing determinations were followed by the work of Richards,’ who fused his
strontium chloride in a stream of gaseous hydrochloric acid and nitrogen, and adopted all of
the precautions relative to the solubility of silver chloride which modern experience has shown
to be necessary. The results, with vacuum weights, follow:

SrCl,. . Ag. Ratio.

4, 2616 5. 7864 73. 476

_» 24019 3. 2688 RE 73. 480
8.5184 2 4.7886 7B. 475.
3. 0264 : 4, 1189 : 73. 73.476, ..

Mean, 3. 78. 4767. 0008
“Hence, § sr= = 87.613.

1 Schweigg. Journ., 19, 228. 1816.
> §Compt. Rend., 17,318. 1843..
Compt. Rend., 20,1047. 1845.
¢Ann.Chim. Phys. (3), 55,29. 1859. Ann, Chem. Pharm., 113, 34,
5 Proc. Amer, Acad., 40,603. 1905. Threeof the determinations were made by H. G. Parker.

 
146 MEMOIRS: NATIONAL ACADEMY OF’ SCIENCES. [Vou, XVI.

Thorpe and Francis‘ determined two of the chloride ratios, as Ape
pet inet The ratio, 2Ag : jis

' (8rCly. 1 Ags Ratio.
iL 64769 . . 2, 24208 73, 486
1.66352, 5, 2, 26356 73, 491.
1.53462 2.08817 73. 491
“1, 64619 2. 24011 73. 487
1.76006 2. 39486 73. 498
1, 56224; . 1,2, 12572, 73. 492

 

rn Mean, 73. 490+. 0008
sce ily gra87.641: ie lae “Ee b 4 SUM et
Second. The ratio oAeCl : SrC,:

SrCl,. AgCl. - Ratio.

1. 64764 2. 97899 55. 309
1. 66357 3. 00762 55. 312
1, 53467 2. 77446 55. 314
1. 64624 2. 97653 55. 307
1.76010 - ' 2, 18202 55. 314

 

Mean, 55. 311-:. 0009
Hence, Sr= 87.645.
The two chloride ratios combined give ‘Ag: Cl :: 100 : 32.867 + .0026.
The four values for the silver ratio now coibine as follows: ©

Beihai 5B csacussc macs ao aiereieegmiree adeloesitailaae 73. 4781 +. 0050
DUR, isc tecdace cmc hetet pus askieaReebenaekaaen 78, 4079+. 0170
apie ad ela sauce OES camieabearanaenas 73. 4767 +. 0008
Tipped Francis......+++00000005+ scewieeeteescacie 73. 4900-4. 0008

General TMGAT aasseesese ar cece eee 73. 4831+. 00056

The foregoing figures.apply to ate strontium chloride, .'The ratio between silver
and the crystallized salt,:SrCL.6H,O, was determined in two series of experiments by Marignac.?
Five grammes of the salt were used in each estimation, and, in the seeond series, the water was
also determined. The quantities, of the chloride corresponding to 100 parts of silver are given
in the last column:

 

 

Series A.
SrCh.6H,0. Ag. Ratio.
5.000 | 4, 0515 123, 411
5. 000 4, 0495 123, 472
5. 000 4.0505 123, 442
; ‘ Mean, 123. 442
Series B.
_ SrCl,.6H,0 | . «Ag. Ratio.
O00 4, 0490 123. 487
5000 4.0500 123.457
. 5.000 J. 4.0490 128. 487
. Mean, 123. 477

#4 Mean of all as one series, 123. 460+. 0082
Hence Sr = 87.362.

From series B, by’ deducting Marignac’s water determinations, 40.563 per cent in mean,
the ratio between silver ‘and the anhydrous chloride can be determined. The value found is
2Ag : SrCl, :: 100 : 73.3907 4.0065. This is so much lower than the measurements oe
cited eats it needs no further consideration.

 

1 Proc. Roy. Soc, 83 A., 277. 1910.
a Arch, Sci. Phys. Nat. 1,220, 1858. Journ. prakt Chem., 74, 216. Oeuvres Compldtes, 1, 668,
No; 8.) 9 1 (' ATOMIC) WHIGHTS-CLARKE. | 147

Marignac also determined the ratio between strontium chloride and strontium sulphate.
By direct conversion, 100 parts of the former salt gave the quantities of sulphate shown in
the third column of the next table:

SrCh,. SrSO,. °° Ratio." <
5. 942 6.887 © 115. 932
5. 941 6. 8855 © 115,949
5. 942 6. 884 115. 927

 

Teh, Gee WT Phd Mean, 115. 986. 004
_ Hence Sr =86.860. ¥
“0. 'The same ratio and also! the corresponding ratio. between prenidie anid sulphate —
been determined by Thorpe and Francis. anes figures are as follows:
First. Ratio SrBr, : SrSO,:

SrBr.. Sr8O,. Ratio.
7. 14570 5. 30466 74, 236.
7.64281 5; 67326 74, 230
9. 86072 7. 32047 74, 239

Mean, 74. 235+. 0018
Hence Sr= 87.700.
Second. Ratio SrCl, : SrSO,: _ sav

 

SrCl,.' Sr80,. Ratio.
7. 30246 8. 46071 116. 861
8. 71628 10. 09868 115. 861
8. 46493 9. 80743 115. 859
8.79502. 10. 18957 115. 855
Bub ie wrod Mean, 115. 859. 0010

Hence Sr = 87.629. —
3 "This value for the’chloride ratio, combined with that of Marignee, gives @ general mean
of 115.871 + .00097. wea Ub atoogs tives

Richards,' in his study of strontium bromide, followed ratty much dhe lines. laid dewa
in his work on barium. The properties of the bromide itself were carefully investigated; and
its purity established beyond reasonable doubt, and then the two usual ratios were determined.
First, the ratio 2Ag : SrBr, :: 100 : z, by titration with standard solutions of silver. For this
ratio there are three series of measurements, by varied processes, concerning which full details
are given. The data obtained, with weights reduced to a vacuum, are as follows:

First series.

We, Ag. Wt. SrBr,. Ratio.

 

1. 30755 1. 49962 114, 689 ©

2. 10351 2. 41225 ji4. 677

2. 23357 2. 56153 "114, 683 *!

5. 3684. 6. La 114.683
oe Mean, 114. 683

Second series. Sai

Wt. Ag. Wt. SrBr,. Ratio.

1. 30762 1, 49962 114, 683

2. 10322 2, 41225 114. 693

4. 57502 5, 24727 114. 694,

5, 3680 6.15663 _ 114, 691"

JD Au

_. Mean, 114. 690

 

1 Proc. Amer. Acad., 1894, 369,
148 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Von. XVI.

! Third series. - fag
WtAg. Wet. SrBr "Ration
2, 5434 2, 9172 "414. 697
3, 3957 3, 8946 114. 692
3. 9607 4, 5426 114, 692
4.5750 5.2473 114, 695

 

Mean, 114. 694
Mean of all as one series, 114. Gh: 0012
Hence Sr = 87.607.

For .the ratio, measured gravimetsionlly, 2AgBr.: ee 5 100 3.2, two series of detonate
nations are given:

 

“Were series.

Wt. AgBr. Wt. SrBro. Ratio.
2.4415. 1. 6086 65. 886,
2, 8561 1.8817 65, 884.
6. 9337: 4, 5681. 65. 883,

2 Mean, 65, 884
Second series.

Wt. AgBr. Wt. SrBro. ‘Ratio.
2. 27625 1, 49962 65. 881
8. 66140 2, 41225 65. 883
8. 88776 2. 56153 65. 887
9. 34497 6. 15663 65. 882

Mean, 65. 883

Mean of all as one series, 65. see 0006
Hence Sr= 87. 618.

From the two. bromide ratios the silver bromine ratio.can.be daiiidtad with the icles
result: Ag : Br :: 100 : 74.077.

The two ‘bromide ratios have been determined by, Thorpe and Francis, as follows:

First.. The ratio 2Ag.: SrBr, : iy

1

SrBr,. } Ag. Ratio. °
1.77884 ' 1.55073 114.710
1.86109 1.62216 ; 114,729
» 1.85254 1.61511 ' ii. 701
1.73801 1.51534 * "114.694
1.85787 1.61994 114.688

1.70563 1.48707 114.697
Mean, 114.703 4.0040

 

Hence, Sr=87. 637. iy
Second. The ratio 2AgBr : SrBr, :

'SrBr,. AgBr. Ratio. -
1.86112 2.82438 65.895
1.85261 - 2.81155 65.893
1.73807 2.63762 65.895
1.85798 2.81999 65.886:
1.70571 2.58866 65.892 :

Mean, 65.892 .0011
Hence, Sr=87.649._ oa
The two bromide rations combined give Ag : : Br : : 100 : 74.077 +.0067.

The bromide series of Richards and Thorpe and Francis may now be combined, as follows:
First: . , sat
Ratio 2.Ag to SrBr,.

RICH Ars 6:40 <isjeieiu 525 spsisiciemcayamrecicaalcrgiaec a se oa asi 114.689 + .0012
Thorpe and Francis. .......-.--2.--2--e- cece eee eee eee 114.703 + .0040

General mean.......... 2... cence ccc een ee eee ceeeee 114.6904 .00116
No, 3.] ATOMIC WEIGHTS—CLARKE. 149

Second:
Ratio 2AgBr to SrBr,.
Richards .xacgesy va arsenite 06 seek aie os sk ey aleeens 65.884 + .0006
Thorpe and Francis. ............-.... EaaGione 7 voapaenmees 65.892 + .0011
General mean...........00. cece clan ee cee teeeweee es 65.8857 + .00053

There are now seven ratios from which to compute the atomic weight of strontium.
(1) 2Ag : Srl, :: 100 : 73.4831 + .00056.
(2) 2AgC] : SrCl, :: 100 : 55.311 4.0009.
(3) 2Ag : SrCl,6H,O :: 100 : 123.460 +.0082.
(4) 2Ag : SrBr, :: 100 : 114.690 +.00116.
(5) 2AgBr : SrBr, :: 100 : 65.8857 + .00053.
(6) SrCl, : SrSO, :: 100 : 115.871 +.00097.
(7) SrBr, : SrSO, :: 100 : 74.235 + .0018.
To reduce these ratios, we have—

Ag=107.880 +.00016 Br=79.9228 + .00015
Cl= 35.4605+ .00011 S=32.0650 + .00012
H=1.0078 + .00001
Hence,
From ratio—
OD rapcietars '5.a|seygafe ceased? a's we were euek ves Vaauieeae Sr=87,363—.0177
Gecien oss + ceietekeb cea s hantieredad Gunite s Seeevewes 87.506 —.0097
BeBe lis epee heats ove so MEA Eee Sh ea RSS 87.610 — .0025
SD ciaiasd Sra evayavavivata wee waded 5059 6: SrSveraceisud nlessiiare vo Senciarepeusoews 87,625—.0020
Dard Udita ahieee sua Sasa eae ey eee te 87.636 —.0013
Bess ee aati coeames oeusce eae 87,645 —.0027
Bee ee ae eee 87,700—.0170
General mean...........00.0.0200eceeeeeeeee Sr=87.6314— .00093

BARIUM.

For the atomic weight of barium we have a series of nine ratios, established by the labors of
Berzelius, Turner, Pelouze, Struve, Marignac, Dumas, Richards, Thorpe, Gray and Ramsay, and
Oechsner de Coninck. Andrews ' and Salvétat? in their papers upon this subject gave no details
nor weighings, and therefore their work may properly be disregarded. First in order we may
consider the ratio between silver and barium chloride as determined by Pelouze, Marignac,
Dumas, and Richards.

Pelouze* in 1845 made the three subjoined estimations of this ratio, using his well-known
volumetric method. A quantity of pure silver was dissolved in nitric acid, and the amount
of barium chloride needed to precipitate it was carefully ascertained. In the last column I
give the quantity of barium chloride proportional to 100 parts of silver:

BaCl,. Ag. Ratio.
3.860 4.002 96.452
5.790 6.003 96.452
2.895 3.001 96.468

 

Mean, 96.4573 + .0036
Hence, Ba = 137.195.

Essentially the same method was adopted by Marignac ‘ in 1848. His experiments were
made upon four samples of barium chloride, as follows: A, commercial barium chloride, purified
by crystallization from water. B, the same salt, calcined, redissolved in water, the solution
saturated with carbonic acid, filtered and allowed to crystallize. C, the preceding salt, washed
with alcohol and again recrystallized. D, the same, again washed with alcohol. For 100
parts of silver the following quantities of chloride were required, as given in the third column.

 

1 Chemical Gazette, October, 1852. ; *Compt. Rend., 20, 1047. ‘Journ. prakt. ‘Chem., 35, 73
2Compt. Rend., 17, 318. 4 Arch, Sci. Phys. Nat., 8, pat Oeuvres Completes, 4, ‘219.
150 MEMOIRS: NATIONAL ACADEMY OF SCIENCES. [Vor/ XVI.

Ag. BaCi,. Ratio. stan wt ®%
3.4445 3.3190' 96.356}
A43.7480 3.6110 96.345|Mean, 96.354
(6.3446: 6.1140 96.362
- mf4-3660 4.1780 96.356)

Bi, 8390 4.6625 96. 352; Moan, BESSt
(6.9200 6.6680: 96.358 aes
5.6230 5.4185. 96. 363} Mean, 3 86.860
5.8435 5.6300 96.346) .

8.5750 8.2650 96.384| sroan’ 96.367
4.8295 4 6470 96.361
6.8460 6.5980 i : 96.377

pees ile 96.360: ae 0024 :

ie to . ey

Hence, Ba= 136.985. 7.

Dumas? employed barium chloride prepared, from. pure barium ate and took the extra
precaution of fusing the salt at a red heat in a current of dry hydrochloric acid gas.. Three
series of experiments upon three samples of chloride gave the following results:

 

Ag. be es BaCi,. \ , Ratio.
1.8260 1.7585 °° 96.303
3.9980 3.8420 96.339 ‘
Ayo 2405 2.1585 96.340(Mean, 96.333
14.1680 4,0162 - 96.358}:
1.7270 1.6625 96.265),
12.5946 2.4987 96.304] °
‘13.5790 3.4468 96.306] °
"Bd 4.2395 4.0822 96.290 Mean, 96.290
4.3683: 4.2062 96.289
4.6290: 4.4564 96.271} *
9,0310 - 8.6975 . . 96.307
2.3835 2.2957 96.316
4.2930 * ABT2 « 96.371
i 4.4300 4.2662 96.303 \Mean, 96. 338
4.6470 4.4764 ee 96.329) . eee

{5.8520 a SRE oy i? ag 18
|, , Mean, 96,316+.0055,.
| Hence, Ba =136. 890. :
The work done by Richards? was “of a much more ‘elaborate kind, for it involved some
collateral invéstigations as to the effect of heat upon barium chloride, etc. Every precaution
was taken to secure the Spectr oscopic purity of the material, which was prepared from several
sources, and ‘similar care was taken with regard to ‘the silver. " For details upon these points
the original paper must be consulted. As for the titrations three methods were adopted, a nd
a special study was made with reference to the accurate determination of the end’ point, in
which particular the investigations of Pelouze, Marignac, ‘and Dumas were at fault.’ In the
first series of determinations silver was added in excess, and the latter was measured with a
standard solution of hydrochloric acid. The end point was ascertained by titrating backward
and forward with silver solution and acid and was taken as the mean ‘between the two apparent
end points thus observed. The results of this series, with weights reduced to a vacuum, were
as follows:

Ag. _BaCi,. Ratio."'
6.1872” 6. 9717 d 96. 517
5. 6580: 1.5, 4597 96. 495.
3. 5988 @ 4728 96.499
9% 4010 , as 2 0726 if ge 96, 507 me

+7199 76960 - 96,541

a Mean, 96. 96. 512. a

Snags pease 3 as

In-the second series of experiments a small excess of silver was aided as before, and the
precipitate of silver chloridé"was removed by ‘filtration. The filtrate and wash ‘waters were
concentrated to small bulk, whereupon a trace of silver chloride was obtained and taken into

 

1 Ann, Chem. Pharm., 113, 22, 1860. Ann. Chim. Phys. (3), 55, 129. 2 Proc. Amer. Acad., 29, 55. 1893.
No 8.) | ' ATOMIG/WEIGHTSACLARKE,© ‘1 ! 151

account. The excess of silver remainitig’was then thrown down as silver bromide, and from
the weight of the latter the silver was calculated and subtracted from the original: amount: «

Ag. BaCl,. Ratio, °°“?
6. 59993 6. 36974 96. 512
5. 55229 5. 86010 96. 539
4. 06380 3.92244 if 96. 522

Mean, 96, 524+:0054 ©:

The third series involved mixing solutions of barium chloride and silver in as nearly as
possible equivalent amounts, and then ‘determining the actual quantities of silver and chlorine
left unprecipitated. The filtrate and wash waters were divided into two portions, one-half
being evaporated with hydrobromic acid and the other with silver nitrate. The small amounts
of silyer bromide and chloride thus obtained were determined by, reduction and the use of
Volhard’s method: . HOB

Ag. Ball, vale ‘
4.4355, | 4, 2815 1) 96, 528 -
2.7440 , 2.6488 96.581
6.1865 5. 9712 96. 520

3. 4028 the 3. 2841 96.526

: “Mean, 96. 526 526+. 0016
iy: V7
Two final. ernest were annals out by Stas’s method, somewhat as in the first series,

with variations and greater refinement in the observation of the end point. The results were
as follows:

Ag. BaCl,. Ratio. ..
voy, 7342 6.50022 96, 525
10. 6023 10. 23865 | 96.523 523

Sexy i Mean, 96. 16. 524-. 0007 ,

A careful study of Richards’ 3 paper will show that, although the last two experiments are
probably the nest, they are not entitled to such preponderance of weight as the ‘‘probable
error” here computed would give them. If all of the determinations are assigned equal weight
and treated as one series, the mean becomes 96.520+..0025, but this figure is not satisfactory.
The four series are unequal i in merit, and that fact may be fairly recognized by combining the
first and second series into one, and the third and fourth series smut a On this basis the
combination of. all the data assumes the following form: wsitsitl tie i) onie

Gelade Peat aaa canes 96. 457 es
Marignacs s:cmes2 ov 2 qanuiuscenm aunts: «ery caccaeeeedes 96.360 +.0024°:'.
(DUNS. s.. couse ss sas eee recatiocenbe en so saneenenwaen 96.316.4+.0055 . .
Richards, Series 1, 2.....------------++--eeeeeee ener e+ 96. 51654.0040
Richards, ‘Series 3, Boi oe aiainla cigs eageinib. 08 deaeineasoraieci stein 96. 5255+. 0010

; Neel in geebe | adieiaticadcadoniean ', 96.4947+.00086 ©

Richards’s determinations alone give Ba= 137.341. aie i

The ratio between silver and crystallized barium chloride has been fixed by Marignac.1 The
usual method was employed, and two series of experiments were made, in the second of which
the water of crystallization was also determined... Five grammes of chloride were taken in each
determination, to which the:subjoined weignis a. silver comenyonde: ‘The ratio to 100 parts of

silver is given in the second “column:

ae

‘Weight Ag. Ratio.
4; 4205 113. 109
aft 4195 113. 135
4.4210 113. 097
| (4, 4195 113.135,
se 4200 — 113. 122 ©
4, 4215 113. 060

Mean, 113. 110-&. 0070
Hence, Ba = 137.094.

1 Arch, Soi. Phys. Nat., 1,209. 1858. Journ. prakt.Chem., 74,212. Oeuvres Complates, 1, 559.
152 MEMOIRS NATIONAL ACADEMY, OF . SCIENCES. (Von. XV,

_., The direct ratio between the chlorides of silver and barium has been:measured by Berzelius,
Turner, Richards, and Thorpe. Berzelius! found of. barium chloride proportional to 100
parts of silver chloride— _ ‘

72, 432
72, 422
Mean, 72. 427

tes

 

Hence, Ba = 136.713.
Turner ? made five experiments, giving the following values for the ratio: =

72,754, / 192, 664
72, 406 | "72. 653
72, 622 eee

Mean, 72. 680+. 0154
Hence, Ba=137.438.

Of these Turner regards the fourth and fifth as the best; but for present purposes it is ‘not.
desirable to so discriminate.

Richards’s determinations ° fall into three series, and all are characterized by their taking
into account chloride of silver recovered from the wash waters. In the first series the barium
chloride was ignited at low redness in air or nitrogen; in the second series it was fused in a stream
of pure hydrochloric acid; and in the third series it was not ignited at all. In the last series it
was weighed in the crystallized state and the amount of anhydrous chloride was computed from
th data so obtained. The data, corrected to vacuum’ standards, ‘are as follows:

AgCl. BaCl,. Ratio.
8. 7673 6. 3697 72. 653
5. 1979 3. 7765 72. 654
A44, 9342 3. 5846 72.648 Mean, 72. 649
2. 0765 1. 5085 72. 646 i
4, 4271 3. 2163 72.650 }
2. 09750 1, 52384 72, 650
of 37610— 5. 36010 72. 669 ple 72, 6563
5. 89906 8. 92244 72. 650
8. 2189 5. 97123 72. 6524
oft 5199 3. 28410 7 72, esa7 Mem, 72. 6555

 

* “Mean, 72. 658 4-. 0014
Hence, Ba = 137.361. '

If we combine this with Richards’s silver series, which, in mean; may be written
2Ag : BaCl, : : 100 : 96.525, the cross ratio between silver and chlorine becomes
Ag : Cl: : 100 : 32.858.

Thorpe’s ‘ measurements of this ratio are not important, for they were merely intended as
a check upon the method he used in determining the atomic weight of radium, which involved
the manipulation of very small quantities of material. His data are given tiere for the sake of
completeness:

AgCl. BaCh. oo Ratio.
0.1578 0.1147 72.687
2368 1721 72.877
0788 0571 72.462
0861 0626 72.706
0937 0681 72.679
1914 1895 72.884
1083 0788 72.761

Mean, 72.694 +.0320
Hence, Ba = 137.479.

1 Poggend. Annalen, 8, 177. *Proc. Amer. Acad, 29, 55, 1893.
2 Phil. Trans., 1829, 291. «Proc. Roy. Soc., 80A, 298. 1908.

 
No. &] ATOMIC WEIGHTS—CLARKE. 153.

Assigning to Berzelius’ work equal weight with that of Turner, the four series representing
the ratio 2AgCl: BaCl, combine as follows:

Berzelius:.. cnc eweinccenermioredens so aueier scans eee 72.427 + 0154
SPUIMNGP 250s ios onscmanseeinnusaersas aba itetacsicnadeeeuides aed 72.680 + .0154
RICHSRCS soo rcs o aces isin serecaeesast yoeeetecuiunss 72.653 + .0014
Thorpe. is ssccansavsenwawsvss seedeteeese eases comewiencce 72.694+ .0320

Mean, 72.650+ .0014

The ratio between barium nitrate and barium sulphate has been determined only by Turner.’
According to his experiments 100 parts of sulphate correspond to the following quantities of

nitrate:
112.060
111.990
112.035

Mean, 112.028-+.014

 

Hence, Ba= 136.317.

For the similar ratio between barium chloride and barium sulphate, there are determinations
by Turner, Berzelius, Struve, Marignac, and Richards.

Turner? found that 100 parts of chloride ignited with sulphuric acid gave 112.19 parts of
sulphate. By the common method of precipitation and filtration a lower figure was obtained,
because of the slight solubility of the sulphate. This observation bears directly upon ne
other atomic weight determinations.

Berzelius,? treating barium chloride with sulphuric acid, obtained the Eanes results in
BaSO, for 100 parts of BaCl,:

 

112,17
112.18
Mean, 112.175
Hence, Ba=135.60. _
Struve,‘ in two experiments, found—
112.0912
112.0964

 

Mean, 112.0938
Hence, Ba = 136.99.
Marignac’s ° three results are as follows:

BaCl,. BaSO,. Ratio.
8.520 9.543 112.007
8.519 9.544 112.032
8.520 9.542 111.995

 

Mean, 112.011+.0071

Hence, Ba = 138.42.

Richards, in his work on this ratio, regards the results as of slight value, because of the oc-
clusion of the chloride by thesulphate. This source of error he was never able to avoid entirely.
Another error in the opposite direction is found in the retention of sulphuric acid by the pre-
cipitated sulphate. Eight experiments were made in two series, one set by adding sulphuric
acid to a strong solution of barium chloride in a platinum crucible, the other by precipitation
in the usual way. Richards gives in his published paper only the end results and the'mean of
his determinations; the details cited below I owe to his personal oe The weights are

reduced to a vaccum.

1 Phil. Trans., 1833, 538.

2 Phil. Trans., 1829, 291.

* Poggend. Annalen, 8, 177.

¢ Ann. Chem. Pharm., 80, 204. 1851.

6 Arch. Sci. Phys. Nat.,1, 209. 1858. Journ. prakt. Chem., 74, 212. Oeuvres Completes, 1, 559.
154 MEMOIRS NATIONAL ‘ACADEMY OF iSCIENCES. {Vou, XVI,

 

BaCh,, eRe 4 a : BaS0,.. . : ,, Ratio, a
1.78934 2.0056 jas 112, 086. ,
2.07670 2.3274 .* "712.079
1.58311 1.7741 112.064
Firs
t 3.27563: 3.6712 112.076.
“18.02489 3.3903 112.080 =. -
“(3.87092 - - 4.3385 112.080 ;° :
(3.02489 3.9726 112.076
pecoa fs. 87091 3.4880 112.085
. ° i RTs a at toe Nfean 112.077 +. 0017 he
Fleien, ie 137. 283. pitts epi ie BO eM iy, EOM. eee BPE , we ep ed Muplge sv

This mean is subject to a small correction, due to loss of chlorine on drying the chloride, which:
reduces it to 112.073. Omitting Turner’s single determination as unimportant, and assigning
to the work of Berzelius and of Struve equal weight with that of Marignac, the measurements
of this ratio combine thus:

Berzelitissscavueuiaanaecunenervessnes sax Bie aged 112.175 4.0071.
Struve........ 112.094 +-.0071
Marignac..... Bris. O11 .0071
theloartis. assis “112. 073+ .0017

  

Gomera mestcssccncchcccswacte-csonec, comme M2075. 0016

‘- In an earlier paper than the one previously dited, Hichardty? atndted, with prea care: the ratios
paaeree barium bromide with silver and silver bromide. ‘The barium bromide was prepared,
by several distinct processes, its behavior upon dehydration and even upon fusion was.studied,
and its specific gravity was determined. The ratio with silver was measured by titration, a
solution of hydrobromic acid being used for titrating back. The data are subjoined, with the
BaBr, equivalent to 100 parts of silver stated:

BaBr,. ‘Ag. 2 ee Ratio.

2.28760 1.66074 187.746

3.47120 2.52019 : / 187.736

2.19940 1.59687 | 137.732

2.35971 1.71323 187.735

2.94207 2.18584. 137.748
1.61191 1.17020 137.747,.

2.10633 1.52921) _ 187.740
2.19682 2.11740) °° p37 755

2.37290 1.72276 187.738

1.84822 1.84175" 137.747

5.66647 | 4.11360 187.750

3.52670 2.56010 187.756

_ 4.31690 3.13430 187.731

; 3.36635 2.44385 137.748
4 7 3.46347, 2.51415 _ 137. 759

eid e 4 aay PRE a Ce ey ee ee a ES

' | Mean, 137. TAB ob. 0015.

S. ¢
SE a Mts

Sead Ba= 137. 352.

. The silver bromide i in most of these. determinations, and in some others, “was collected and
weighed i in a, Gooch, crucible | with all necessary precautions. Vacuum réductions apply to, both
ratios.. I give in a third; column. the BaBr, equivalent to, 100 parts | of AgBr:

ey, A BaBr,.. . =e AgBr.. : a ‘Ratio, '
2. 28760 2. 89026 79, 149
3. 47120 4, 38635 79.136 °°
3. 81086 4, 81688 79. 138
2. 35971 2. 98230 - 79, 124.
2. 94207 8. 71809 79.129
2, 10633, - 2, 66191 79.128
2. 91682 3. 68615 79. 129

 

1 Proc. Amer. Acad., 28,1. 1893.
No. 8:] ' ATOMIC WEIGHTS—CLARKE. - 155

BaBr,. AgBr. Ratio,
2.37290 2. 99868 79. 131
1. 84822 2. 33530 79. 143
1, 90460 2. 40733 79. 116
5, 66647 7. 16120 .. 79. 127
38. 52670 4. 45670 79. 133
2. 87743 3. 63644 79. 127

3. 46347 4, 37669 79. 135

 

Mean, 79. 132+ .0015

Hence, Ba = 137.378. From the two bromide ratios combined, Ag: Br : : 100 : 74.070.
The last ratio was also determined by Thorpe,’ incidentally to his work on the atomic
weight of radium:

BoBr,. AgBr. Ratio.
0. 0899 0, 1136 79. 137
. 0960 . 1214 79. 077
1110 . 1403 79. 116
_. 0910 . 1149 79. 199
. 0808 sa 1021 79. 138

 

Mean, 79. 133 0134

Henes, Ba= 137, 328, in meugiciiatean of Richards’ series. On combination with ‘ie iafies
no noteworthy change is produced.

The determinations of the atomic weight of barium by Gray and Ramsay? were merely
intended as a check upon the accuracy of their work on radium. Barium bromide was con-
verted into ‘chloride, which was afterwards reconverted into bromide. The microbalance was
used, and the following weights, in ‘tmilligrammes, were obtained:

3.50557 BaBr, = 2.45668 BaCl, =3. 50883 BaBr,.

Hence, there are two values for the ratio, BaBr;: BaCl, : 100 :: 2, as follows:

70. 079
et 70.014

Mean, 70. 046+. 0220

Ne ire. : veg gunheents |

These fcieaguan: are of slight importance, and so ) too are those of Oechsner de Coninck:*
Barium carbonate was dissolved in nitric acid, and the liberated carbon dioxide was collected
in caustic potash and weighed. The weights obtained, and the percentage of CO, follow:

BaClo0;. C0... Per cent CO.
1. 6100 0. 3587 22, 280
1.7700 . 3447 22, 299
1, 4480 — . 3229 22. 300
1. 1590 . 2583 22, 286

os 2275 62738 22. 305

1 iy py
Mean, 22, 294. 0032

Hence, Ba=137.371: i! °°? _
From the following nine ratios the atomic weight of barium can now ke abigpaballe

(1) 2Ag : BaCl, :: 100 : 96.4947+. 00086.
(2) 2AgCl : Bal, : :: 100 : 72.650 +.0014.
(3) 2Ag : BaCl,. 2H,O :: 100 : 113.110+.0079.
(4) BaSO, :BaN0, :: 100 : 112.028 +.014.
- -(5)- BaCl,-: BaSO, :: 100 : 112.075 +.0016.
(6) 2Ag : BaBr, :: 100 : 137. 745+. 0015.
(7) 2AgBr : BaBr, :: 100 : 79.132 4.0015.
(8) BaBr, : BaCl, :: 100 : 70.046 +.0220.
(9) BaCO, : CO, :: 100 : 22.294+.0032.

1 Proc. Roy. Soc. 80A, 298. 1908. 2 Proc. Roy. Soc., 86A, 289, 1912. § Rev. Gén. Chim., 16, 245. 1913.

 

 
156 MEMOIRS. NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

To reduce these ratios we have:

‘Ag=107. 880 +. 00016 N=14. 0081+. 00008

Cl= 35. 4605+. 00011 S=35. 0650+. 00012

Br= 79. 9228+. 00015 C=12. 0025+. 00019

H=1, 0078+. 00001
Hence,
From ratio—

Ass sccisqatlavseted stuleie See Sues Geek Sadtcaeeeease Ba=136. 317+. 2705
Bis si at Calais ON au SA RNA Ae tll oie! 187, 023-4. 2180
Sees etree ent ieemaaaaiaeys citer et tticiaii eh tale 137. 094+. 0171
iss sanaues Suimedese enacts reps Seas rata dete Saga etal 187. 276+. 0019
6 cane habeisuacde acacia Bee weemeaneveseus ee 137. 308+. 0314
Bescueensys SNA et ocr el Ue al cate ca aed ne 187. 352-4. 0033
Da ih oes ails Ala a 8s Sea elena ene eectaiae 187. 353+. 0041
Quiet pone dooce esac elaesckeeveewsersercsseweert 187. 3714. 0224
T auniguisen ces te cease eanersdaaeeaeseeie 137. 378+. 0056
General mean................----eeeeeeeeee Ba=187, 308+. 0015

This mean is probably too low, for the value from ratio 1 is affected by the doubtful deter-
minations of several early investigators. That ratio, however, has the highest weight in the
eorabinaiion: Rejecting the first four values, the last five give a general mean of

Ba=137. 358+. 0023

which will be adopted in subsequent computations.
A few experiments are on record with reference to determining the atomic weight of barium
from the percentage of water in the hydrated chloride. This method has been carefully i inves-
tigated by Guye and Tsakalotos,’ who conclude that the chloride in question is not suited to
the purpose. Their data give Ba=139.5 approximately; while similar data by Marignac give
136.5. The subject needs no further consideration here.

" RADIUM.

The early, preliminary attempts to determine the atomic weight of radium may be ignored,
for they were made with confessedly impure material. In 1902 Madame Curie? published the
first, determinations of any value, basing them upon the following analyses of radium chloride.
The ratio 2AgCl : RaCl, is given in the third column :

a We

RaCl,. AgCl. Ratio.
0. 09192 0. 08890 103. 397

.. 02936 - 08627 103. 582

- 08839 - 08589 102. 911

 

Mean, 103. 297+.1349
Hence, Ra = 225.21.

In the foregoing determinations the radium chloride still contained appreciable amounts
of barium chloride. In a later series of determinations Madame Curie * used purer material,
and in much larger quantities. The results obtained were as follows:

Racl,. AgCl.a Ratio.
- 4052 . 89054 103. 768
. 4020 . 38784 103, 651
. 39335 . 87944 103. 666

 

Mean, 103, 695+. 0236
Hence, Ra = 226.35.

« Corrected for the weight of filter ash.

1 Journ. Chim. Phys., 7, 215. 1900. Marignac’s figures are discussed in the second edition of this work. See also Baker and Adlam, Journ.
Chem. Soc., 99, 502,

3 Ann. Chim. Phys. (7), 30, 140. 1903. Preliminary datain Compt. Rend., 135, 80. 1902,

3 Compt. Rend., 145, 422. 1907.

 
‘No. 8.) / ‘ATOMIC WEIGHTS—CLARKE. : . 157

Still more recent are the determinations by Thorpe,’ on small quantities of material :

‘RaCl,. AgCl. Ratio.
0, 0627 0. 0604, 103. 808
. 0639 . 0618 103. 398
. 0784 . 0753 104. 117

Mean, 103. 774 +:. 1399

Hence, Ra= 226. 57.

Thorpe regards his figures, however, as merely corroborative of Madame Curie’s. A com-
bination of the two series gives 103.700 for the general mean, and Ra= 226.37.

In his determinations of the atomic weight of radium, Hénigschmid had at his command
the resources of the Radium Institute at Vienna, and was able to use much larger quantities of
material than his predecessors. In his first memoir ? he gives analyses of radium chloride with
all modern precautions. His weights, reduced to a vacuum, and the. corresponding ratios

follow :

ReCl,. AgCi. oe, Ag. Ag ratio. ' AgCt ratio.
1, 06167 PO "U02584- etc 0m ces 103. 543
1. 03300 -99751 a see wwiadeibiore:s 103. 558
1. 01687 . 98205 ate feds 108, 547

. 99905 - 96473 . 72626 137. 561 103. 557

. 98555 - 95166 . 71630 137. 589 103. 561

- 97142 . 93804 . 70600 137. 595 103. 559

- 60385 SOSSIG) adeatnic: § saan iets 103. 548

. 57579 . 55606 abate wjausiataveed 103. 548

Mean, 187. 582+.0022 103. 553-4. 0016
From the Ag ratio, Ra=225.926. From the Ag(Cl ratio, Ra=225.946.
It is not necessary to combine the earlier determinations with these, for their. high probable

errors discard them.
In a second memoir by HG6nigschmid * he reports analyses of radium bromide. First, for

the ratio 2AgBr : RaBr, :: 100 : z.

RaBry. AgBr. Ratio.

1.17016 1. 13940 102. 699
- 99658 . 97022 102. 717
- 90111 . 87733 102. 710
. 75504 - 73502 102. 724
. 72731 - 70810 102. 715
- 60329 - 58734 102. 715

1. 08458 , 1. 05588 102. 777

ee Mean, 102. 7144. 0019.
Hence, Ra= 225.953.
Second, the ratio 2Ag : RaBr, :: 100 : x.

i aioe | RaBry . Ag. Ratio.
"0, 68496 - 0, 38306 178, 813

- + 60829 " § "°, 33789 178. 811
Mean, 178. 812-4. 0009
Hance, tae 205, 959. lg Ph
-.» The determinations by Gray and Hannay are quite unlike hose already coiisiderdd.
They converted radium bromide into chloride, and so sought to establish the ratio RaBr, :
RaCl, :: 100: «. The quantities of material used were very small, and weighed on the micro-
balance. The weights, in milligrammes and reduced to a vacuum, and the ratio follow:

 

* RaBr,. RaCcl,. Ratio.
2. 91458 2. 24144 76. 904

2. 91415 2. 24113 76. 905

1; 50117 1, 15474 76. 923

. 36910 . 28400 76, 944

2. 85262 2. 19586 76. 977

2. 76084 2. 12492 ‘76. 966

1 Proc. Roy. Soc., 80, A, 208. 1908. * Sitzungsb. Wien. Akad., 121, Abth IIa. Nov. 1912.

2 Monatsh. Chem., 33, 253. 1911. 4 Proc. Roy. Soc., 86 A, 270. 1912.
158 MEMOIRS: NATIONAL ACADEMY ‘O¥ SCIENCES. (Vou. XVI.

5 ie so RaBr*, ee r RaCi,.. RUber SK, + Refio,,. oy
2. 61099 2.00988 76. 978
2. 43281 1: 87275 76. 979
2. 36062 1: 81711 76.976

Mean, 76 945 +. 0072
Hence, Ra = 225. 860. a as
In his secorid memoir, which was later than that of Gray and Ramsay, Hénigschmid gives
_ two determinations of the same rae: 8 as follows: These weights are of course in grains, .

sere |

rin RGB ag ge 4 9 RaCh.. ; i ‘Ratio. ,
0.1288 0. 09915 é 76. 950
he ae f 2 ° TAT2 i 7 76. 952
ae - 3 ao Mean, 76. 951

Hence, Ra =225. 962,55 ae
ol TFwe assign this pair of values‘equal weight with the mean of Gray.and Hainaay? 8 bine)
tions, the general mean becomes 76.948 +. 0050. role
There are now five ratios from which to compute the atomic weight of radium.

(1) 2Ag: RaCl:: 100: 137.582 + .0022.
(2) 2AgC1: RaCl,:: 100: 103.553 + .0016.
(3) ,2Ag: RaBr,:: 100: 178. 812+ .0009.
(4) 2AgBr: RaBr,:: 100: 102.714 +.0019.
(5) RaBr,: RaCl,:: 100: 76.948 + .0050.

To reduce these we have—
; Ag=107. 880 +. 00016

Cl= 35,4605+.00011 ; _
{ leds { che g¢ Sv Be ho ees afte hha Hits _;BI= 79. 9228. 00015 ; nah Oe. ae Bac
Hence, oe
From ratio— ae
SE “BT Gach. bbe ees dpe Sa ote :, Ra=225. 911+. 0348...
61/}225, 926. 0048. |
225. 946+. 0042
225: 953-4. 0072
2259594. 0022
Ra=225. 9524. 0017

 

LEAD.

For the atomic weight of lead we have to consider experimients made upon the oxide,
chloride, bromide, nitrate, sulphate, and the metal itself. The researches of Berzelius upon
the carbonate and various organic salts need not now be be considered,-nor is .it worth while
to take into account any work of his done before the year 1818, The results obtained by
Débereiner: and by Longchamp? are also without present value.

For the exact composition of lead oxide we have to depend upon the researches of Berzelius.
His experiments were made at different times through quite a number. of years; but were finally
summed up in-the last edition of his famous “Lehrbuch.’”? In general terms his method of
experiment was very simple. Perfectly pure lead oxide was heated ina current of: hydrogen,
‘and the reduced: metal’ weighed. From his weighings I havey satoulited the  caieie. of oe
thus found and given them i in a ‘third column: a ee oe er

— 3 Earlier results, - 7
PbO. Pb. Per cent.
“8. 045 7, 4675 92, 8217
14, 183 18, 165 92, 8224
10. 8645 10. 084 92. 8160
13. 1465 112, 2045 92. 8346
21, 9425 20, 3695 92. 8313
11. 159 10, 859 92. 8309

 

1 Schweig. Journ., 17, 241,- 1816, 2 Ann. Chim, Phys., 34, 105. : 1827.) ® Bd, 3,5.:1218,

i a geite
No. 3.] ATOMIC WEIGHTS—CLARKE. 159

Latest.
6. 6155 6. 141 92. 8275
14. 487 13, 448 92. 8280
14. 626 13. 5775 92. 8313

 

Mean, 92. 8271+. 0013
Hence, Pb = 207.062.

For the synthesis of lead sulphate we have data by Berzelius, Turner, and Stas. Berzelius,!
whose experiments were intended rather to fix the atomic weight of sulphur, dissolved in each
estimation 10 grammes of pure lead in nitric acid, then treated the resulting nitrate with sul-
phuric acid, brought the sulphate thus formed to dryness, and weighed. One hundred parts

of metal yield of PbSO,:
146. 380 146. 440
146. 400 146. 458
Mean, 146. 419+. 012
Hence, Pb = 206.95.

Turner,” in three similar experiments, found as follows:

146. 430
146. 398
146. 375

Mean, 146. 401+. 011

 

Hence, Pb = 207.03.

In these results of Turner’s, absolute weights are implied.

The results of Stas’ syntheses,’ effected after the same general method, but with variations
in details, are as follows. Corrections for weighing in air were applied:

Weight Pb. Weight PbSO,. Ratio.
141. 9925 207. 9388 146. 443
148. 016 217. 6141 146. 427
100. 000 146. 419 146. 419
200. 000 292. 864 146. 432
250. 000 366. 0525 146, 421
250. 000 366. 0575 146. 423

 

Mean, 146. 4275+. 0024
Hence, Pb = 206.91.

Combining, we get the subjoined result:

BerZGlIUB s jitrcssi ctr ranceiervnulern e544 nem eeeeeeeseaenes 146. 419 +.012

TUM el a sects vsinie = aes s Be Liv a heel das eget naiaaxiadmve asics 146. 401 +. 011

BABS e ioreanc ast anes sere ectaueveecachi reece re x Alpe eta tea cette 146. 4275+. 0024
General mean.....-.....--...- 2202-2 e eee ee eee eee 146. 4262+. 0023

Turner, in the same paper, also gives a series of syntheses of lead sulphate, in which he
starts from the oxide instead of from the metal. One hundred parts PbO, upon conversion
into PbSO,, gained weight as follows:

35. 84 35. 79
35. 71 35. 78
35. 84 35. 92
35. 75

Mean, 35.804+.018
Hence, Pb = 207.623.

These figures are not wholly reliable. Numbers 1, 2, and 3 represent lead oxide contami-
nated with traces of nitrate. The oxide of 4, 5, and 6 contained traces of minium. Number
7 was free from these sources of error, and, therefore, deserves more consideration. The series
as a whole undoubtedly gives too low a figure, and this error would tend to slightly raise the
atomic weight of lead.

 

4 Lehrbuch, 5th ed., 3, 1187 2 Phil. Trans., 1833, 527-538. 3 Ocuvres Completes, 1, 390.
161566°—20——11
160 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI.

Still a third series by Turner establishes the ratio between the nitrate and the sulphate,
a known weight of the former being in each experiment converted into the latter. One hundred

parts of sulphate represent of nitrate:
109, 312

109. 310
109. 300

Mean, 109. 307+ .002

Hence, Pb=204.75. This value is so low that it need not be considered fitter:
In all these experiments by Turner the necessary corrections were made for weighing in

 

" In 1846 Marignac! published two sets of determinations of only moderate value. First,
chlorine was conducted over weighed lead, and the amount of chlorine so formed was deter-
mined. The lead chloride was fused before weighing. The ratio to 100 Pb is ee in the last
column:

Pb. PbCI,. Ratio.
20. 506 27. 517 134. 190
16. 281 21. 858 134. 225
25. 454 34 149 134. 159

 

Mean, 134. 191+.013
Hence, Pb = 207.426.

Secondly, lead chloride was precipitated bysilver nitrate and the ratio between PbCl, and
2AgCl determined. The third column gives the PbCl, equivalent to 100 parts of a

PbCl,. AgCt. Ratio.
12, 534 12, 911 97. 080
14. 052 14, 506 96. 870
25. 533 26. 399 96. 720

Mean, 96. 890+ .0704
Hence, Pb =206.84.

For the ratio between lead chloride and silver we have first a series of results by Marignac
and one experiment by Dumas. There are also later determinations to be considered in
chronological order. Some old data by Turner and by Berzelius are of no present value.

Marignac,? applying the method used in his researches upon barium and strontium, and
working with lead chloride which had been dried at 200°, obtained. these results. The third
column gives the ratio between PbCl, and 100 parts of Ag:

POCl. Ag. Ratio.
4, 9975 3, 8810 128. 768
4, 9980 3, 8835 128. 698
5. 0000 3. 8835 128, 750
5. 0000 3. 8860 128. 667

 

Mean, 128..721+.016
Hence, Pb = 206.81.

Dumas,’ in his investigations, found that lead chloride retains traces of water even at
250°, and is sometimes also contaminated with oxychloride. In one estimation 8.700 grammes
PbCl, saturated 6.750 of Ag. The chloride contained .009 of impurity; hence, correcting
Ag : PbCl, :: 100 : 128.750. If we assign this figure equal weight with those of Marignac, we
get as the mean of all 128.7266+.013. The sources of error indicated by Dumas, if they are
really involved in this mean, would tend slightly to raise the atomic weight of lead:

The synthesis of lead nitrate, as carried out by Stas,‘ gives better results. Two series
of experiments were made, with from 103 to 250 grammes of lead in each determination. The
metal was dissolved in nitric acid, the solution evaporated to dryness with extreme care, and
the nitrate weighed. All weighings were reduced to the vacuum standard. In series A the
lead nitrate was dried in an air current at a temperature of about 155°. In series B the drying

 

1 Ann. Chem. Pharm., 59, 289, and 200. 1846. Oeuvres Complates, 1, 186. § Ann. Chem. Pharm., 113,35. 1860.
2 Journ. prakt. Chem., 74, 218, 1858. Oeuvres Complates, 1, 574. 4 Oeuvres Completes, 1, 386,
No. 3.] ATOMIC WEIGHTS—CLARKE. 161

was effected in vacuo. The data are as follows, together with the ratio of nitrate to 100 parts
of lead:

 

Series A.

Weight Pb. Weight PbN,O,. Ratio.
103. 000 164. 773 159. 973
140. 6887 225. 0674 159. 975
110. 2672 176. 408 159, 982
141. 9927 227. 1527 159. 975
148. 616 237. 702 159. 968
128. 348 198. 924 159. 973

Mean, 159, 9743+ .0012
Series B.

Weight Pb. Weight PbN,O.. Ratio.
100. 000 159, 970 159. 970
200. 000 319. 928 159. 964
250. 000 399. 8975 159. 959
250. 000 399. 914 159. 965

 

Mean, 159, 9645+. 0015
: Mean from both series, 159. 9704+. 0010

Hence, Pb = 206.796.

There is still another set of experiments upon lead nitrate, originally intended to fix the
atomic weight of nitrogen, which may properly be included here. It was carried out by Ander-
son! in Svanberg’s laboratory, and has also appeared under Svanberg’s name. Lead nitrate
was carefully ignited, and the residual oxide weighed, with the following results:

PON,0¢. PbO. Per cent.
5. 19485 3. 5017 67. 4071
9. 7244 6. 5546 67. 4037
. 9. 2181 6. 2134 67. 4044
9. 6530 6. 5057 67. 3957

Mean, 67. 4027+. 0016

Hence, Pb = 207.35.

The direct ratio beteeen lead and silver has been roughly measured by the electrochemical
experiments of Betts and Kern.? Lead silicofluoride was dissolved in hydrofluosilicic acid,
and from the solution the lead was thrown down electrolytically, silver being simultaneously
precipitated by the same current. Two series of experiments gave the following data. The
ratio 2Ag : Pb :: 100 : z is stated in the third column:

First Series.

Weight Ag. Weight Pb. Ratio.
5. 8958 5.6221 95. 476
5. 8958 5. 6396 95. 654
5. 7863 5. 5246 95. 477
5. 7863 5. 5450 95. 830
7, 8408 7. 5108 95. 791
7. 8408 7.5168 95. 868
7. 6253 7.3191 95. 984
7. 6253 7.3221 96. 025
6. 2287 5. 9600 95. 676
6. 2287 5. 9605 95. 694

16. 6804 15. 9996 95. 919
16. 6804 16. 0014 95. 923
6. 8652 6. 5815 95. 868
6. 8652 6. 5812 95. 863
9. 3258 8. 9390 95. 858
9, 3253 8. 9419 95. 889
6. 8566 6. 5695 95. 813
6. 8754 6. 5877 95. 816

Mean, 95. 801+. 0243

 

1 Ann, Chim. Phys. (3), 9, 254. 1843. 2 Trans. Amer. Electrochem. Soc., 6, 67.
162 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI.

Second Series.

Weight Ag. Weight Pb. Ratio.
9. 0470 8. 6678 95. 809
9. 0470 8. 6663 95. 792

18. 4113 12. 8607 95. 895
18. 4113 12, 8558 95. 858
7. 2780 6. 9716 95. 790
7. 2780 6. 9755 95. 844
7. 2738 6. 9605 95. 693
7. 2738 6. 9698 95. 821
6. 5278 6. 2550 95. 821
6. 4864 6. 2168 95. 844

Mean, 95. 817+. 0109
General mean of both series, 95. 8141-. 0097

Hence, Pb = 206.73.

Baxter and Wilson? determined the atomic weight of lead by analyses of lead chloride,
which had been previously fused in an atmosphere of gaseous hydrochloric acid. The ratio to
Ag and to AgCl were both determined, and the figures obtained, with vacuum weights, are as
follows:

PbCI,. Ag. AgCl. Ag ratio. AgCt ratio.
4. 67691 3. 62987 4. 82273 128. 845 96.976
3. 67705 2. 85375 ahs aise 128.850 | — ......

4. 14110 3. 21408 4. 27016 128. 842 96. 978
4. 56988 3.54672 ss. se 128.848 = ......

5. 12287 3. 97568 5. 28272 128. 855 96. 974
3. 85844 2. 99456 3. 97949 128. 848 96. 958
4. 67244 3.62628  — .....-- 128.849  —......

3. 10317 2. 40837 3. 19909 128. 849 97. 002
4. 29613 3. 33407 4, 42982 128. 856 96. 982

 

Mean, 128.849+.0010 96.978+. 0039

From Ag ratio, Pb = 207.088.

From Ag(Cl ratio, Pb = 207.096.

And Ag : Cl :: 100 : 32.864.

The work done upon the atomic weight oflead by Meaglia ? is confessedly an approximation,
and nothing more. Metallic lead was used to precipitate silver from a nitrate solution, and in
that way the ratio Ag : Pb was determined. Two series of observations were made, with
different preparations of lead. Calculated with Ag = 107.93, the following values for Pb were
obtained:

1. 2.
206. 872 206. 866
206. 907 206. 897
206..903 206. 927
206. 909 206. 933
206. 930 206. 935
206. 903
206. 929

Treating both series as one, and reducing the figures to the form of ratio adopted in this

work, the mean becomes
2Ag : Pb :: 100 : 95.853+4. 0020.

Combined with the series by Betts and Kern, 95.814+4.0097, the general mean becomes
95.850 + .0019.

The same electrolytic ratio has also been measured by Pecheux,? who found Pb = 206.86.
As no details are given this determination can not be utilized here.

Analyses of lead bromide by Baxter and Thorvaldson,* lead to a slightly higher value for
the atomic weight of lead than Baxter and Wilson obtained from the chloride. Their corrected
weights and the usual ratio follow:

 

1 Proc. Amer. Acad., 43,365. 1907. Journ. Amer. Chem. Soc., 30, 187. § Compt. Rend., 154, 1419. 1912.
1 Thesis, University of Grenoble, 1907. 4 Journ. Amer. Chem. Soc., 37, 1020. 1915.
No. 8. ATOMIC WEIGHTS—CLARKE. 163

PbBr. Ag. AgBr. Ag ratio. AgBr ratio.
6. 55858 3. 85610 6. 71241 170. 083 97. 708
4, 83285 2.84126  —— ......- 170.095 — ......
5. 67758 3.33787 nee 170.096  — ......
6. 73361 3. 95874 6. 89124 170. 095 97.713
5. 93130 3.48660 — ....... VIO cates
5. 01729 2. 94968 5. 13400 170. 096 97. 727
5. 89902 3.46730 sts. eee 170.1838 see
7. 71526 4. 53498 7, 89526 170. 128 97.720
7, 40244 4.35157 nee 170.130 © ......
7. 01144 4, 12154 7.17470 170. 117 97. 725
6. 91460 4. 06484 7. 07672 170. 108 97. 709
8. 12623 4.77679 eee (0:

 

 

; Mean, 170.109+.0030 97.'717+. 0022
From the Ag ratio=207.181.

From the AgBr ratio Pb = 207.186.

And Ag : Br :: 100 : 74.075.

On account of the remarkable discovery, by several investigators, that the atomic weight
of lead from radioactive minerals differs from that of ordinary or normal lead, it became
desirable to determine whether the latter was constant, at least within the limits of experimental
uncertainty. For this purpose Baxter and Grover’ made elaborate series of analyses of lead
bromide and chloride from very different sources. Lead was obtained from galena, cerussite,
vanadinite, and wulfenite, and the minerals came from widely separated localities; namely,
from Idaho, Arizona, Washington, Missouri, Germany, and Australia. Four different mineral
species and seven localities furnished the material for the determinations. Commercial lead
nitrate was also included in the investigation. First in order are the analyses of lead bromide,
with corrected weights as usual.

PoBr,. Ag. AgBr. Ag ratio. AgBr ratio.
5.27845 Sl0277. «= asivaueets VW7O;124  etwrerctes
2.65094 1.558220 sis sees 170.126  —......
4.08410 2.40104 —— ....-. 170.097 = 2...
4.97468 2.924738 = ...... 170.090  — ......
4.05541 2.38398 4.15017 170.111 97.717
3.44139 2.02288 3.52224 170.123 97.705
5.17387 3.04158 5.29498 170.105 97.713
3.84497 2.26022 3.93446 170.115 97.726
4.30513 2.53086 4.40616 170.105 97.707
4.53445 2.66549 4.64048 170.117 97.715
5.78437 3.40044 5.91976 170.107 97.713
4.87079 2.86337 4.98467 170.107 97.715
6.28446 3.69447 6.43124 170.104 97.718
4,74639 2.79011 4.85708 170.115 97.721
6.82424 4.01148 6.98390 170.118 97.715
6.53689 3.84274 = = ...... 170.110 ee
4,10098 2.41088 4.19791 170.103 97.691*
2.64256 1.55852 is ee 170.10l oe
6.30683 3.70718 = ...... 170.125 = ......

 

 

. Mean, 170.111+.0015. 97.715+.0013.
From the Ag ratio, Pb = 207.186.

From the AgBr ratio, Pb = 207.177.
And Ag : Br :: 100 : 74.089.
Combining the series of Baxter and Thorvaldson with that of Baxter and Grover the ratios

become—
2Ag : PbBr, :: 100 : 170.1106+.0013
2AgBr : PbBr, :: 100: 97.7154+.0011

 

1 Journ. Amer. Chem. Soc., 37, 1027. 1915. * Rejected by the authors. Not included in the mean.
164 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Baxter and Grover also measured both chloride ratios, but separately. - First, the ratio of
lead chloride to silver chloride.

PbCi,. AgCl. Ratio.
6.96370 7.17754 97.021
6.89046 7.10231 97.017
7.33227 7.65732 97.022
6.51699 6.71690 97.024

Mean, 97.021+.0010
Hence, Pb = 207.221.

For the ratio between lead chloride and silver, Baxter and Grover give the following data:

PbCh. Ag. Ratio.

5.63827 4.37200 128.904
5.58730 4.33427 128.910
6.86319 5.32402 128.910
4.70770 3.65223 128.899
4.20222 3.25968 128.915
7.04688 5.46691 128.901
5.88935 4.56868 128.907
4.90083 3.80171 128.911
5.79300 4.49404 128.904
5.43965 4.21992 128.908
5.74504 4.45674 128.907
6.57216 5.09849 128.904
5.66330 4.39340 128.905
5.73434 4.44857 128,903
4.17445 3.23862 128.896
6.25884 4.85584 128.893
5.25882 4.07933 128.914

 

Mean, 128.905+.0010
Hence, Pb = 207.205.

The values given here for the atomic weight of lead differ slightly from those found by
Baxter and Grover because different values are assigned to chlorine and bromine.

Analyses of lead chloride have also been made by Hénigschmid and Horovitz,1 whose
data are as follows. The weights are corrected to a vacuum.

PbCI,. Ag. AgCl. Ag ratio. AgCl ratio.
6.17861 4.79374 6.36895 128.889 97.011
4.75170 3.68659 4.89837 128.891 97.006
4.83377 3.75047 4.98528 128.890 97.004

 

 

. Mean, 128.890+.0004. 97.007-+.0014.
From the Ag ratio, Pb = 207.172.

From the AgC] ratio, Pb = 207.180.

And Ag : Cl :: 100 : 32.867.

Four more datiertitnatvons of the ratio of flier to lead chloride remain to be noticed; two
by Richards and Wadsworth,’ and two by Richards and Hall.* The values found are so ‘close
together that they may be taken here as one series. The figures are as follows:

PCh. Ag. Ratio.
3.72918 2.89325 128.892
5.35111 4.15151 128. ‘390, snd
7.13723 5.53723 128.895
7.60063 5.89675 ‘oene andlH.

Mean, 128.894+.0005.
Hence, Pb = 207.181.

The various determinations of the two lead chloride ratios can now be combined as follows:

 

1 Monatsh. Chem., 36, 375. 1915. . 2 Journ, Amer. Chem. Soc., 38, 2613. 1916, 3 Thid., 39, 587. 1917.
No. 3.]. ATOMIC WEIGHTS—CLARKE.
First.
2Ag: PbCl,:: 100: x.
Marignac with Dumas..............2202.0--- cece eee eee ee
Baxter and Wilson................2.0 2000 cece cee eee ee eee
Baxter and Grover...........202 200 e cece eee cece eee cena
Honigschmid and Horovitz ........:.......0.2-. 2020202 eens

Richards, Wadsworth, and Hall
General MGAat o. 332 sonceewendsdiceduiie eles wen nests

Second.

2AgCl : PbCl, : : 100 : x.

Baxter and Wilson...............0.0002 eee e cece eee eens
Baxter and Grover......-...--cccececececccecececececeees
Ho6nigschmid and Horovitz ..................-02-20-0200-

General mean......... ae etitaa eden euaed

128.727 + .0130
128.849 + .0010
128.905 + .0010
128.890 + .0004
128.894+ .0005

- 128.860 + .00029

96.890 +.0704
96.978 +.0039

- 97.021 +.0010

97.007 +.0014
97.0147+ .00081

165

Finally the ratio between lead nitrate and lead oxide, previously measured by Anderson,
The nitrate was calcined to
oxide, and both compounds were weighed. The data are as follows:

has been redetermined by Oechsner de Coninck and Gérard.1

PON,{0..
0. 9054
. 7243
. 8149
. 6338

Poo.
0. 6099
- 4880
. 5490
- 4270

From the new determination Pb = 207.01.
The following ratios are now available from which to compute the atomic weight of lead:
(1) PbO : Pb :: 100 : 92.8271+.0013.
(2) PbN,O, : PbO :: 100 : 67.3922 +.0013.
(3) Pb : PbSO, :: 100 : 146.4262 +.0023.
(4) PbO : PbSO,:: 100 : 135.804 +.0180.
(5) Pb : PbN,O,:: 100 : 159.9704 + .0010.
(6) Pb : PbCl,:: 100 : 184.191 +.0130.
(7) 2Ag : PbCl,:: 100 : 128.860 +.00029.
(8) 2AgCl : PbCl,:: 100 : 97.0147 +.00081.
(9) 2Ag : PbBr,:: 100 : 170.1106 + .0013.
(10) 2AgBr : PbBr,:: 100 : 97.7154+.0011.
(11) 2Ag : Pb :: 100 : 95.850+.0019.

‘Computing with—

Ag=107. 880 +. 00016
Cl = 35. 4605+. 00011

we have—
From ratio—

Per cent PbO.
67. 362
67.375
67. 370
67. 371

Mean, 67. 3695+. 0023
Anderson found, 67, 4027+. 0016
General mean, 67, 3922+. 0013

Br=79. 9228+. 00015
S =82, 0650+. 00012

N=14, 00814. 00008

Pb=206. 796+. 0035

206. 806+. 0042
206. 920+. 0102
207. 062+. 0376
207. 107 +. 00069
207. 179+. 0042
207. 185+. 0028
207. 202+. 0024
207. 243+. 0099
207. 426+. 0787
207. 623-4. 1124

Pb=207. 102+. 00062

 

1 Compt. Rend., 163, 514. 1916.
166 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

This final result is probably, but not certainly, too low. The best modern determinations
indicate that the atomic weight of lead is not far from 207.2. The lowering of the value, as
given here, is partly due to the work of Stas on lead nitrate and lead sulphate and partly to that

of Baxter and Wilson on lead chloride.
ADDENDUM.

Two recent determinations of the atomic weight of normal lead, by Davis, were made in
connection with a study of isotopic lead. Analyses of lead chloride gave the following figures:

 

PbCly. AgCl. Ratio.
0. 94411 0. 97264 97. 067
1, 24607 1, 28454 . 97.005
Mean, 97. 036 1

Hence, Pb = 207.263.
These determinations are too discordant to be of much significance.

ISOTOPIC LEAD.

One of the most startling discoveries of modern chemistry has been the fact that the ele-
ments of highest atomic weight, namely, uranium and thorium, are unstable, and undergo a
slow decomposition into simpler forms. Helium and radium are among the products of this
disintegration, the process apparently ending with the production of lead. This lead differs
appreciably in atomic weight from ordinary or normal lead, the atomic weight of the latter
being nearly and perhaps absolutely constant. This constancy is clearly shown by the recent
researches of Baxter and Grover, Hénigschmid and Horovitz, Richards and Lembert, Richards
and Wadsworth, and Richards and Hall. The earlier discordances are due, partly to differ-
ences of method and technique, and partly to experimental errors.

Isotopic or radio-lead on the other hand, the lead derived from radioactive minerals, may
be either higher or lower in atomic weight than normal lead, accordingly as it originates from
thorium or uranium, and this fact has been determined by several competent investigators.
Let us consider first the lead from uranium minerals.

In 1914 Maurice Curie ? published the end results, without analytical details, of a series of
atomic weight determinations of lead from different sources. The data are as follows:

Origin: At.wt.Pb.
Pite bblend O2.35 css senate pcibeleiaisieg. vis os Saisie e cea e eis'e'e ve scidcins 206. 64
ORIOLE! sis cu ptosis cramer tee laa ele siccaidjarcialeesitiearesergarsine eis a a0's & 206. 36
Vttrotantalite cies csicscnwweiveGisaniae codeine hse eae nee ese chee 206. 54
MOnaZite: sosescnetscemsssccsnewcbeasen sees cee sa oes Peta vege ee 207. 08
Ordinary lead. :.ccccesec cacescceveeeesaseaveess Syste ke seals 207. 01

These figures speak for themselves. The three leads from radioactive minerals have much
lower atomic weights than the others. ;

At about the same time Richards and Lembert ? sibnaied much more elaborate determina-
tions than those of Curie. The material studied was derived from a variety of sources, and the
lead, after careful purification, was converted into chloride, which was analyzed. The original
minerals, which are indicated in the table of analyses by capital letters, were as follows:

D, P. Colorado carnotite.
H, M. Ceylonese thorianite.
I, K. Bohemian pitchblende.
G. English pitchblende.
R. North Carolina uraninite.
First come the determinations of the ratio PbCl, : Ag, with vacuum weights, as follows:

Origin. PbCl. Ag. Ratio. At. wt, Pd,
p< 3. 87082 3. 00984 128. 606 206. 56
D 6, 57331 4, 33300 128, 625 206, 60

 

1 Journ. Phys. Chem., 22, 631. 1918. 2 Compt. Rend., 158, 1676. 1914, 3 Journ. Amer, Chem. Soc., 36, 1320, 1914,
No, 3.]

Origin.

POAC RAH

ATOMIC WEIGHTS—CLARKE.

PbCh.
5. 49412
3. 88228
4, 05550
4. 05168
2. 80814
8. 95052
2. 95726
4, 05702
2. 01795

Ag.
4, 27157
3. 01600
3. 15061
3. 14788
2, 18162
3. 07209
2, 29951
3. 15153
1. 56952

Ratio.
128. 621
128. 723
128. 722
128. 712
128. 718
128. 594
128. 604
128. 732
128, 563

For the ratio PbCl, : AgCl the following data are given:

Origin.

AORRMAVOS

PCI.

3. 87082
5, 57331
5, 49412
3, 88228
4. 05168
2. 80814
2, 95726
4, 05702
2. 01795

AgCl.

_ 8, 99879
5. 75707
5. 67573
4. 00703
4. 18265
2, 89816
3. 05475
4. 18670
2. 08663

Ratio.
96. 799
96. 808
96. 800
96. 886
96. 870
96. 894
96. 809
96. 903
96. 767

167

Al. wt. Pb.
206. 59
206. 81
206. 81
206. 79
206. 81
206. 54
206. 56
206. 84
206. 47

At. wt. Pb,
206. 59
206. 61
206. 58
206. 84
206. 78
206. 85
206. 61
206. 88
206. 32

Two years later another series of determinations, by Richards and Wadsworth appeared.'
The sources of the lead may be indicated by number, thus: 1. Australian carnotite. 2. Colo-
rado carnotite. 3. Bréggerite, Norway. 4. Cleveite, Norway.

Origin.

RROD DE HEH

PCh.
4, 64010
5, 35517
6. 15608
4, 14770
5, 31585
4 65899
4, 29104
3, 92736
4, 45270

Ag.
3. 61118
4. 16711
4. 79072
3. 22748
, 4. 12670
3. 61707
3. 34187
3. 05913
3. 46818

Ratio.
128. 493
128. 512
128. 500
128. 512
128. 816
128. 806
128. 402
128. 382
128. 387

At. wt. Pb.
206. 318
206. 359
206. 334
206. 339
207. 015
206. 994
206. 122
206. 079
206. 090

The determinations by Richards and Hall? were all made with lead from Australian
carnotite. Their figures follow:

POC.

3. 70016
4, 79633
4. 64577
4, 64736
4, 87838
4. 68396

Ag.
2. 87854
3. 73154
3. 61412
3. 61568
3. 79579
3. 64425

Ratio.
128. 543
128. 535
128. 545
128, 534
128. 541
128. 530

At. wt. Pb,
206. 426
206. 409
206. 431
206. 406
206. 422
206. 399

The atomic weights were calculated with Ag = 107.88, and Cl=35.459. With Cl=35.4605 the
atomic weight of lead would be 0.003 lower.

In 1914 several determinations of the atomic weight of lead from pitchblende were pub-
lished by Hénigschmid,? whose data are as follows:

PbCl.

1. 97691
3. 04040
3. 05981
3, 30285
3. 33164
3, 22450

AgCl.
2. 04118
3. 13910
3. 15939
3. 41015
3. 44013
3. 32923

First series.

Ratio,
96. 851
96. 856
96. 848
96. 854
96. 857
96. 857

At. wt. Pb.
206. 733
206. 746
206. 724
206. 740
206. 719
206. 749

 

1 Journ. Amer. Chem. Soc., 38, 2618. 1916,

2 Journ. Amer. Chem. Soc., 39, 531. 1917.

8Zeitsch. Elektrochem., 20, 457. 1914.
168

PbCl.
3. 33164
3, 22459
3. 79447

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

Second series.

Ag. Ratio. At. wt. Pb.
2. 58888 128. 690 206. 748
2. 50586 128. 682 206. 730
2. 71545 128. 688 206. 744

von. XVI.

In the following year this investigation was continued by Hénigschmid and Horovitz 4
with much greater elaboration. Lead was obtained from three different sources, namely,
from the purest Joachimsthal pitchblende, from a crystallized uranium ore from Morogoro,

German East Africa, and from Norwegian bréggerite.

From this lead the chloride was pre-

pared, and analyzed by the methods followed by Baxter and Wilson in their study of ordinary

lead. The results obtained with vacuum weights from several series are as follows:

First.—Lead from. pitchblende.

PbCl,. AgCl. Ratio. At. wt. Pb.
2. 91224 3. 01042 96. 739 206. 410
2. 47886 2. 56256 36. 734 206. 396
2. 76797 2. 86126 96. 740 206. 413
Second.—Ore from Morogoro, Ag series.
PbCl,. Ag. Ratio. At. wt. Pb.
38. 24177 2. 52563 128. 355 206. 024
3. 94867 3. 07636 128. 355 206. 025
3. 78263 2. 94693 128. 358 206. 033
3. 68899 2. 87400 128. 357 206. 030
4. 49731 3. 50343 128. 369 206. 054
Third.—From Morogoro, AgCl series.
PoCl,. AgCl. Ratio. At. wt. Pb.
38. 24177 3. 35537 “96, 614 206. 054
3. 94867 4. 08686 96. 619 206. 066
3. 78263 3. 91519 96. 614 206. 053
3. 68899 3. 81830 96. 613 206. 052
4, 49731 4, 65464 96. 620 206. 070
Fourth.—From bréggerite, Ag series.
PbCl,. Ag. Ratio. At.wt.Pb.
4. 83605 3. 76724 128. 371 206. 060
3. 97826 3. 09903 128. 371 206. 060
4, 91204 3. 82647 128. 370 206. 057
Fifth—F rom bréggerite, AgCl series.
PbCI,. AgCi. Ratio. At.wt.Pb.
4. 50069 4. 65794 96. 624 206. 082
4. 49363 4. 65060 96. 625 206. 084
4. 83605 5. 00512 96. 622 206. 076
3. 97826 4.11741 96. 620 206. 072
4, 91204 5. 08425 96. 613 206. 050

In this series the first two analyses were only preliminary. The determinations of the

atomic weight of normal lead by Hénigschmid and Horovitz have already been cited.

Two determinations of the atomic weight of isotopic lead from samarskite, by Davis,? are

as follows:
PbCl.

1. 06795
. 73514

AgCl, Ratio. At.wt.Pb.
1. 10199 96. 911 206. 905
. 76011 96. 715 206. 342

 

1 Monatsch. Chem., 36, 355.

1915. 2 Journ. Phys. Chem., 22, 631. 1918.
No, 3.] ATOMIC WEIGHTS—CLARKE. 169

Although these determinations do not agree, they confirm the lower atomic weight of
isotopic lead as derived from another source.

Since the foregoing pages were put in type a short paper upon lead by Richards and
Sameshima?’ has also appeared. The usual ratio of 2 Ag:PbCl, was determined for ordinary
lead and also for lead from a radioactive Japanese mineral. The figures are as follows:

Ordinary lead.
Ag. PbCl,. Ratio. At. wt. Pb.
3. 13929 2. 43553 128. 895 207. 185
3. 09476 2. 40100 128. 894 207. 183

 

Mean, 207. 184
Japanese lead.

Ag. PbCl,. Ratio. At. wt. Pb.
2. 16756 1. 68212 128. 859 207.11
1. 14536 . 88881 128. 864 207. 12
1. 34496 1. 04538 128. 879 207. 15

Mean, 270.13

The Japanese material evidently contains very little isotopic lead.

The data relative to thorium lead are as yet not quite complete. Soddy and Hyman ?
examined lead derived from Ceylonese thorite, and found for its atomic weight the values
208.3, 208.4, and 208.5. This value is almost certainly excessive; for Hénigschmid? by
analyses of lead chloride derived from Soddy’s original material found Pb=207.78. His data
are as follows, with vacuum weights:

 

PbCl,. AgCl. At.wt.(Th)Pb.

2. 71471 2. 79264 207. 76

3. 90048 4. 01204 207. 77

4. 05682 4. 17300 207. 78

8. 80951 3. 91848 207. 78
Mean, 207. 78

In another communication Soddy‘ from the specific gravity of thorium lead, assuming
that normal and isotopic lead have identical molecular volumes, finds for the atomic weight of
thorium lead the value 207.64. Still later Hénigschmid * found for the atomic weight of lead
obtained from Norwegian thorite a much higher figure, namely, 207.90. Hénigschmid’s original
paper I have not seen. That thorium lead has a higher atomic weight than normal lead is,
however, thoroughly established.

The time has not yet come for the final interpretation of these remarkable results, and yet
they reveal some relations which are worth noting here. In the work of Richards and Lembert
relatively high values were found for the atomic weight of lead from thorianite, a mineral which
contains both uranium and thorium. This species therefore seems to contain both isotopes of
lead, and possibly normal lead also. The lowest atomic weights are shown by the crystalline
varieties of uraninite, namely, bréggerite, cleveite and the Morogoro mineral. Here, doubtless,
we have the closest approach to the ideal uranium lead, which is supposed on theoretical grounds
to have the atomic weight 206. In Hillebrand’s ° splendid series of analyses of uraninite and its
varieties the highest proportion of helium was found in those which were most brilliantly
crystalline. The massive varieties contained little or no helium, and the atomic weight of lead
derived from them is now shown to be relatively high. There is, therefore, a clear relation
indicated between the proportions of helium and isotopic lead in uraninite, which cannot be
without significance. Moreover, as shown by Hillebrand’s analyses, nearly all uraninites con-
tain more or less thorium, which suggests that both isotopes are present in them, but with

 

1 Journ. Amer. Chem. Soc., 42,929. May, 1920. 4 Nature, 94, 615, 1915.
3 Journ. Chem. Soc., 105, 1402. 1914. 6 Cited by Fajans, Zeitschr. Elektrochem., 26, 163. 1918,

8 Zeitschr. Elektrochem., 23, 161. 1917. 6 Amer. Journ. Sci., 40, 384. 1890.
170 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

uranium lead predominating. A comparison of the atomic weights given in the preceding pages
with the analyses of the minerals from which they were derived would surely be interesting.
It would also be worth while to determine the atomic weight of lead obtained from minerals
associated with uraninite, but not contained in it. It seems probable that all uraninites and
thorites contain more or less normal lead which was inclosed or dissolved in them at the time
of their formation; the two isotopes having been developed later.

It has been suggested by some writers that normal lead is a mixture in equal or nearly
equal proportions of the two isotopes; but on that basis it is difficult to account for the
constancy of atomic weight of the ordinary metal. One or the other of the isotopes would be likely
to predominate. I now venture to offer a different interpretation of the evidence, which may
seem to be extravagant, but which is at least worth considering. Normal lead may be regarded
as having been generated in the orderly evolution of the chemical elements, an evolution which
is indicated by the progressive complexity of the spectra of the heavenly bodies, as I pointed
out many years ago.’ Isotopic lead, on the other hand, is a product of degradation, and might be
looked upon as in a sense defective lead—lead which has lost or gained something from its normal
constitution. The atoms of the elements are now known to be complex, with nuclei surrounded
by electrons. If we suppose the three forms of lead to have identical nuclei, to which their
most characteristic properties are due, we can imagine that slight changes in mass and density
are ascribable to corresponding changes in their envelopes: That this is pure speculation must
be admitted, but it may help to suggest future lines of. research.

GLUCINUM.

Our knowledge of the atomic weight of glucinum is derived from experiments made upon
the sulphate and two salts. Leaving out of account the deternimations by Berzelius,? Awdejew,?
Weeren,‘ Klatzo,5 and Debray,® we have to consider only three modern investigations, for the
older work is no longer of real value. It is cited in detail in the former editions of this recal-
culation.

In 1880 the careful determinations of Nilson and Pettersson appeared.’ These chemists
first attempted to work with the sublimed chloride of glucinum, but abandoned the method
upon finding the compound to be contaminated with traces of lime derived from a glass tube.
They finally resorted to the crystallized sulphate as the most available salt for their purposes.
This compound, upon strong ignition, yields pure glucina. The data are as follows:

GIS0,.4H,0. Glo. Per cent GIO.
3. 8014 0. 5387 14.171
2. 6092 . 3697 14. 169
4. 3072 . 6099 14. 160
3.0091 » 4266 14. 176

 

Mean, 14. 169+. 0023
Hence, Gl= 9.113.

Kriiss and Moraht ° in their work followed the general method adopted by Nilson and Pet-
tersson, but with various added precautions and greater elaboration of detail. Their glucina
was derived from three sources, namely, leucophane, beryl, and gadolinite, and the sulphate
was repeatedly recrystallized. The results are as follows:

 

GISO.4H,0. Gio. Per cent Gi0,
21. 1928 38. 0008 14. 160
16. 2038 2. 29455 14, 161
15. 49345 2. 1902 14, 186
20. 1036 2. 8433 14, 143
22. 0465 3. 1167 14. 137
4.9619 - 7019 14. 146
18. 3249 2. 5921 14, 145
i Popular Science Monthly, January, 1873. 4 Poggend. Annal., 92,124. 1854, 7 Ber. Deutsch. chem, Ges., 13, 1451. 1880,
1 Poggend. Annal., 8, 1. 5 Zeitsch. anal. Chem., 8, 533, 1869. 8 Liebig's Annalen, 262, 38. 1891.

* Poggend. Annal., 56, 106. 1842, ‘Ann, Chim. Phys., (3), 44,37. 1855.
No. 3.) ATOMIC WEIGHTS—CLARKE. 171

G1S0.4H,0. Gio. Per cent GIO.
24, 3907 3. 0253 14, 143
20. 18045 2, 85255 14. 135
20. 0253 2. 8328 14. 146
18. 9840 2. 6832 14. 134
17. 0072 2. 4073 14, 155
22. 5044 3. 1805 14. 133
20. 88675 2. 95645 14, 154
19. 0591 2. 69305 14. 130
17. 8227 2. 5226 14. 153

 

Mean, 14. 144. 0017
Hence, Gl= 9.065.

The first two determinations, which give the highest percentage, were made upon sulphate
thrice crystallized. The others were made upon a salt four times crystallized, except in one
instance, when there were five crystallizations. To the data derived from the four times crystal-
lized compound Kriiss and Moraht give preferance, and so find a slightly lower value for the
atomic weight of glucinum. Combining, we have for the mean percentage:

By Nilson and Pettersson............-+-..0+222+-222 25222. 14. 169+. 0023
By Kriiss.and Morahtsojs.00 :<s:niseiessissicisissic uw eetiinev es vais 14. 1444. 0017
General mean ..........-020e eee eeeeeeee SSsaces, 14. 158+. 0014

The determinations, by Parsons, of this atomic weight, were based upon analyses of two
organic salts, namely, the acetylacetonate, G1(C;H,O,),, and the basic acetate, G1,O(C,H,O,),.
These compounds are volatile at moderately high temperatures, and can therefore be purified
by sublimation; an advantage which the sulphate does not possess. Parsons attempted to make
determinations with sulphate, also but obtained unsatisfactory results.

Weighed quantities of the two organic compounds were first decomposed, in platinum
crucibles, with nitric acid. The nitrate solutions so formed were then evaporated to dryness,
and the residual salt was converted into oxide by prolonged ignition. The oxide was
examined for occluded gases, and its weight was given the necessary correction for them.
The data obtained, with vacuum weights, were as follows:

Acetylacetonate. Oxide. Per cent GIO.
2. 62245 0. 31798 12. 125
3. 28037 . 39757 12. 119
2. 08993 . 25286 12. 099
2. 41401 . 29233 12. 109
1. 61353 . 19554 12. 118
1. 39714 - 16905 12. 100
1. 85023 . 22419 12. 117

 

Mean, 12. 1124+. 0025
Hence, Gl=9.101.

Acetate. Oxide. Per cent G10.
2. 61484 0. 64630 24. 716
2. 67721 . 66109 24. 693
3. 11534 . 76930 24. 693
1. 89291 . 46788 24, 717
1. 47931 . 86534 24. 703
1. 09012 . 26911 24, 686
1. 35642 . 33493 24, 692
1. 56787 . 88715 24, 693
1, 34465 . 88204 24. 698

 

Mean, 24. 698+. 0025
Hence, G1=9.1046.

In a later note,? Parsons combined the two series of determinations, and deduced simultaneous
values for Gland C. IfO=16, and H=1.008, then Gl=9.112, and C=12.007. These figures
furnish a good check upon the general accuracy of the manipulations.

 

1 Journ. Amer. Chem. Soc., 26, 721. 1904. 2 Journ. Amer. Chem, Soc., 27, 1204. 1905.
172 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XV1L

The atomic weight of glucinum is now fixed by the following ratios:

(1) GISO,.4H,O : GIO :: 100 : 14.153 + .0014
(2) G1(C,H,O,), : GIO :: 100 : 12.1124 + .0025
(3) G1,O(C,H,O,), : 4GIO :: 100 : 24.698 + .0025

Reducing these ratios with
C=12.0025 +.00019

S=32.0650 +.00012
H= 1.0078 +.00001

we have—
Prot TatiO Voiced ied cides. eciecsdGunaseoorerecises G1=9. 0802 +. 0026
ID cans cers vibsheee lores Ded vancogia aca ay aca ooo ane 9. 1012+. 0052
EB crerc fase Ns carts oul ciavou ncaa evan weven a ctasar ey aberare weet meaty 9. 1046+. 0026
General mean.......-----------+--- Gl=9. 0934+. 0017

For all practical purposes the atomic weight of glucinum may be taken as 9.1, which must

be very near the true value.
MAGNESIUM.

There is perhaps no common metal of which the atomic weight has been subjected to closer
scrutiny than that of magnesium. The value is low, and its determination should, therefore,
be relatively free from many of the ordinary sources of error; it is extensively applied in chemical
analysis, and ought consequently to be accurately ascertained.

The early determinations made by Berzelius, Longchamp and Gay-Lussac need not be
considered here, as they have only antiquarian value. The investigations which demand atten-
tion are those of Scheerer, Svanberg and Nordenfeldt, Jacquelain, Macdonnell, Bahr, Marchand
and Scheerer, Dumas, Marignac, Burton and Vorce, and Richards and Parker.

Scheerer’s method of investigation was exceedingly simple. He merely estimated the
sulphuric acid in anhydrous magnesium sulphate, employing the usual process of precipitation
as barium sulphate. He gives no weighings, but reports the percentages of SO, thus found.
In his calculations, O=100, SO,=500.75, and BaO=955.29. It is easy, therefore, to recal-
’ culate the figures which he gives, so as to establish what his method really represents, namely,
the ratio between the sulphates of barium and magnesium.

Thus revised, his four analyses show that 100 parts of MgSO, yield the following quantities
of BaSO,:

BaSO,. Per cent SO3.
198.575 66.573
193.677 66.608
193.767 66.639
193.631 66.592

 

Mean, 193.6625+: .0274

Hence, Mg = 24.466.

In a later note? Scheerer shows that the barium sulphate of these experiments carries
down with it magnesium salts in such quantity as to make the atomic weight of magnesium
0.039 too low.

The work of Bahr, Jacquelain, Macdonnell, and Marignac, and in part that of Svanberg
and Nordenfeldt, also relates to the composition of magnesium sulphate.

Jacquelain’s experiments were as follows: * Dry magnesium sulphate was prepared by mix-
ing the ordinary hydrous salt to a paste with sulphuric acid, and calcining the mass in a platinum
crucible over a spirit lamp to constant weight and complete neutrality of reaction. This dry
sulphate was weighed and intensely ignited three successive times. The weight of the residual
MgO having been determined, it was moistened with sulphuric acid and recalcined over a
spirit lamp, thus reproducing the original weight of MgSQ,. Jacquelain’s weighings for these

 

1Poggend, Annal., 69, 535. 1846. a Poggend. Annal., 70, 407. 8 Ann, Chim. Phys. (3), 32, 202,
No. 3.] ATOMIC WEIGHTS—CLARKE. 173

two experiments show that 100 parts of MgO correspond to the quantities of MgSO, given in

the last column:
1.466 grm. MgSO, gave .492grm.MgO. 297.968
492 grm. MgO gave 1.466 grm. MgSO,. 297.968
Hence, Mg = 24.443.

Jacquelain also made one estimation of sulphuric acid in the foregoing sulphate as BaSQ,,.
His result (1.464 grm. MgSO, = 2.838 grm. BaSO,), reduced to the standard adopted in dealing
with Scheerer’s experiments, gives for 100 parts of MgSO,, 193.852 BaSO,. If this figure be
given equal weight with a single experiment in Scheerer’s series, and combined with the latter,
the mean will be 193.700+.0331. This again is subject to the correction pointed out by Scheerer
for magnesium salts retained by the barium sulphate, but such a correction determined by
Scheerer for a single experiment is only a rough approximation, and hardly worth applying.

The determinations published by Macdonnell' are of slight importance, and all depend
upon magnesium sulphate. First, the crystallized salt, MgSO,.7H,O, was dried in vacuo over
sulphuric acid and then dehydrated at a low red heat. The following percentages of water
were found:

61.17 51.26
51.13 51.28
51.14 51.29

Mean, 51.21+ .020

Secondly, anhydrous magnesium sulphate was precipitated with barium chloride. From
the weight of the barium sulphate, with SO,=80 and Ba=137, Macdonnell computes the per-
centages of SO, given below. I calculate them back to the observed ratio in uniformity with
Scheerer’s work:

Per cent. SO3. Ratio, MgSO, : BaSO,.
66.67 194,177
66.73 194.351
66.64 194.089
66.65 194.118
66.69 194.239

In another experiment 60.05 grains MgSO, gave 116.65 grains BaSO,, a ratio of 100 : 194.254.
Including this with the preceding figures, they give a mean of 194.205+.027. This, combined -
with the work of Scheerer and Jacquelain, 193.700 + .033, gives a general mean of—

MgSO, : BaSO, :: 100 : 194.003 +.021

In one final experiment Macdonnell found that 41.44 grains of pure magnesia gave 124.40
grains of MgSO,, or 300.193 per cent.

From Macdonnell’s data the atomic weight of magnesium ranges between 24.00 and 24.43.

Bahr’s? work resembles'in part that of Jacquelain. This chemist converted pure mag-
nesium oxide into sulphate, and from the increase in weight determined the composition of
the latter salt. From his weighings 100 parts of MgO equal the amounts of MgSO, given in
the third column:

MgO. MgS0,. Ratio.
1.6938 5.0157 296.122
2.0459 6.0648 / 296.437
1.0784 ~ 3.1925 296.040

 

Mean, 296.200 .0815
Hence, Mg = 24.807.

About four years previous to the investigations of Bahr the paper of Svanberg and Norden-
feldt * appeared. These chemists started with the oxalate of magnesium, which was dried at
a temperature of from 100° to 105° until it no longer lost weight. The salt then contained two
molecules of water, and upon strong ignition it left a residue of MgO. The percentage of MgO
in the oxalate was as follows.

 

1 Proc. Royal Irish Acad., 5, 303. British Assoc. Report, 1852, part 2, p. 36. 3 Journ. prakt. Chem., 45, 473. 1848,
3 Journ. prakt. Chem., 56, 310. 1852.
174 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Oxalate. MgO. Per cent.
7.2634 1.9872 27.359
6.3795 1.7464 27.375
6.3653 1.7418 27.364
6.2216 1.7027 27.368

Mean, 27.3665 -+ .0023
Hence, Mg = 24.704.

In three of these experiments the MgO was treated with H,SO, and converted, as by
Jacquelain and by Bahr in their later researches, into MgSO,. One hundred parts of MgO
gave of MgSO, as follows:

MgO. MgSO,. Ratto.
1.9872 5.8995 296.875
1.7464 5.1783 296.513
1,7418 5.1666 296.624

Mean, 296.671-+0.072
Hence, Mg = 24.710.

In 1850 the elaborate investigations by Marchand and Scheerer'! appeared. These chem-
ists undertook to determine the composition of some natural magnesites, and, by applymg
corrections for impurities, to deduce from their results the sought-for atomic weight. The
magnesite chosen for the investigation was, first, a yellow, transparent variety from Snarum;
second, a white opaque mineral from the same locality; and, third, a very pure quality from
Frankenstein. In each case the impurities were carefully determined, but only a part of the
details need be cited here. Silica was, of course, easily corrected for by simple subtraction
from the sum of all of the constituents; but iron and calcium, when found, having been present
in the mineral as carbonates, required the assignment to them of a portion of the carbonic acid.
In the atomic weight determinations the mineral was first dried at 300°. The loss in weight
upon ignition was then carbon dioxide. It was found, however, that even here a correction
was necessary. Magnesite, upon drying at 300°, loses a trace of CO, and still retains a little
water; on the other hand, a minute quantity of CO, remains even after ignition. The CO,
expelled at 300° amounted in one experiment to 0.054 per cent; that retained after calcination
to 0.055 per cent. Both errors tend in the same direction and increase the apparent percentage
of MgO in the magnesite. On the yellow mineral from Snarum the crude results are as follows
giving percentages of MgO, FeO, and CO, after eliminating silica:

co, MgO. Fed.
51.8958 47.3278 0.7764
51.8798 47.3393 -7809
51.8734 47,3154 8112
51.8875 47,3372 -7753

Mean, 47.3299-+ 0.0037.

After applying corrections for loss and retention of CO,, as previously indicated, the mean
results of the foregoing series become—

co, MgO. FeO.
51. 9931 47,2743 . 7860

The ratio between the MgO and the CO,, after correcting for the iron, will be considered
“further on. ?
Of the white magnesite from Snarum but a single analysis was made, which for present
purposes may be ignored. As for the Frankenstein mineral three series of analyses were exe-
cuted. In the first series the following results were obtained:

co, MgO. Per cent.
8. 996 8.2245 47.760
7.960 7.2775 47.761
9.3265 8.529 47.767
7.553 6. 9095 47.775

Mean, 47. 766+ .0022

 

‘Journ, prakt, Chem. , 50, 385.
No. 3.] ‘ATOMIC WEIGHTS—CLARKE. 175

‘This mean, corrected for loss of CO, in drying, becomes 47.681. I give series second with
corrections applied:

MgCO, MgO. Per cent.
6.8195 3.2500 47. 658
11.3061 5. 3849 47.628
9. 7375 4.635 47.599
12.3887 5. 9033 47.650
32. 4148 15. 453 47.674
38. 8912 18. 5366 47.663
26. 5223 12. 6445 47.675

 

Mean, 47. 650-4 . 0069

The third series was made upon very pure material, so that the corrections, although
applied, were less influential. The results were as follows:

MgCo; MgO. Per cent.
4.2918 2.0436 47.622
i 27.8286 13.2539 47.627
14. 6192 6. 9692 47.672
18.3085 8.7237 47.648

 

Mean, 47. 642+. 0077

In a supplementary paper' by Scheerer, it was shown that an important correction to
the foregoing data had been overlooked. Scheerer, reexamining the magnesites in question,
discovered in them traces of lime, which had escaped notice in the original anlayses. With
this correction the two magnesites in question exhibit the following mean composition:

Snarum. Frankenstein.

 

BN cao ast eteneate aca seen mmeredaaeeiieagaee 52.131 52. 338
MPO nics fais cacietehiowineodcauheee Ga tiamrnceniascianee Oe 46. 663 47. 437
CAO iets aed Shas eteyatoral a Natereyseatee eh angio tater sania . 4380 . 225
WoO cccscevieseseeweveadansesee acee Bo ry ares hte TI6 ed eae

100. 000 100. 000

Correcting for lime and iron, by assigning each its share of CO,, the Snarum magnesite
gives as the true percentage of magnesia in pure magnesium carbonate, the figure 47.624. To
this, without serious mistake, we may assign the weight indicated by the probable error, + .0037,
the quantity previously deduced from the percentages of MgO given in the uncorrected analyses.

From the Frankenstein mineral, similarly corrected, the final mean percentage of MgO in
MgCO, becomes 47.628. This, however, represents three series of analyses, whose combined
probable errors may be properly assigned to it. The combination is as follows:

+. 0022
+.0069
+. 0077

 

Result, +. 0020, probable error of the general mean.
We may now combine the results obtained from both magnesites:

Snarum mineral................-........... Per cent MgO, 47. 624+ .0037
Frankenstein mineral.................----- s 47.628+.0020
General mean............-.--2-2-65- Per cent MgO, 47.627+.0018

Hence Mg = 24.015.

The next investigation upon the atomic weight of magnesium which we have to consider
is that of Dumas.? Pure magnesium chloride was placed in a boat of platinum, and ignited
in a stream of dry hydrochloric acid gas. The excess of the latter having been expelled by a
current of dry carbon dioxide, the platinum boat, still warm, was placed in a closed vessel
and weighed therein. After weighing, the chloride was dissolved and titrated in the usual
manner with a solution containing a known quantity of pure silver. The weighings which

 

1 Ann. Chem, Pharm., 110, 240, 2 Ann, Chem. Pharm., 113,33. 1860,
161566°—20——12
176 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

Dumas reports give, as proportional to 100 parts of silver, the quantities of MgCl, stated in
the third column:

MgCl, Ag. Ratio.
2. 203 4. 964 44. 380
2. 5215 5. 678 44. 408
2. 363 5. 325 44. 376
3. 994 9. 012 44, 319
2. 578 5. 834 44. 189
2. 872 6. 502 44.171
2. 080 4.710 44. 161
2. 214 5. 002 44. 262
2. 086 4, 722 44.176
1. 688 3. 823 44, 154
1, 342 3. 031 44, 276

Hence, Mg = 24.77. Mean, 44. 261+. 020

This determination gives a very high value to the atomic weight of magnesium, which is
unquestionably wrong. The error, probably, is due to the presence of oxychloride in the ma-
nesium chloride taken, an impurity tending to raise the apparent atomic weight of the metal.
Richards’ and Parker’s revision of this ratio is more satisfactory.

Marignac,' in 1883, resorted to the old method of determination, depending upon the direct
ratio between MgO and SO,. This ratio he measured both synthetically and analytically.
Fist, magnesia from various sources was converted into sulphate. The MgSO, from 100 parts
of MgO i is given in the third column:

MgO. MgSO. Ratio.

Weise eisja'sis nce saearesesearsterslataia stevens 1. 5635 4. 6620 298. 17
Dsccmvacus se emcievazsncudvecutec- 14087 4, 2025 298. 32
Bensuseueceamiaas amamedeacasmasls 1 5917 4. 7480 298. 30
Aoscecsoscmaseeu cease seeeee ets 1. 4705 4, 3855 298. 23
Dasseesesemennncmmannetiveces sal 1. 4778 4. 4060 298. 15
Gscecectevesmdsmeeescerareess 1, 6267 4. 8530 298. 33
Vee ceomenmtens zcrtyersree See oie 1. 3657 4. 0740 298. 37
Baseeeceecie uisspitacare sees 1 9575 5. 8390 298. 29
D scciciotrcewicee isn eupetiosecusmenete = 1. 6965 5. 0600 298. 26
BLO): sors isc ats 2s yahapsceie oa ts cenewais saeqsasseserese 1. 8680 5. 5715 298. 26

Mean, 298. 27+. 0149
‘ Hence, Mg = 24.382.

The magnesia for experiments 1 to 5 was prepared by calcination of the nitrate, that of 6
to 8 from the sulphate, and the remaining two from the carbonate. But Richards and Rogers?
have shown that magnesia derived from the nitrate always contains occluded gaseous impurity,
so that the experiments depending upon its use are somewhat questionable. The results tend
to give an atomic weight for magnesium which is possibly too high. Whether the other samples
of magnesia are subject to similar objections I can not say.

Marignac’s second series was obtained by the calcination of the sulphate, with results as

follows: MgSO. MgO. Ratio.
3.7705 1. 2642 298. 25
4. 7396 1, 5884 298. 39
3. 3830 1. 1845 298. 19
4, 7154 1, 5806 298, 33
4, 5662 1. 5802 298. 43
4.5640 1. 5300 298. 30
3. 2733 1, 0979 298, 14
4, 8856 1, 6378 298. 30
5. 0092 1, 6792 298. 31
5. 3396 1. 7898 298, 38
5, 1775 1. 7352 298. 38
5. 0126 1. 6807 | 298, 24
5. 0398 1, 6894 298, 32

H ie oLGr Mean, 298. 30-+. 0150
ence, Mg = 24.376.

‘1 Arch. Sci. Phys. Nat. (3), 10, 206. Oeuvres Complates, 2, 742. 3 Amer. Chem. Journ.. 15, 567. 1893,

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 177

These data may now be combined with the work of previous investigators, giving Mac-
donnell’s one result and Jacquelain’s two, each equal weight with a single experiment in Bahr’s

Beniee? Dacia oe tena teens Lic II ono 300. 193-4. 1413
DMO USAT oy asess steed sacs atercres anzsncaie occ cueecimreianarava alataldle annvebavetmatcile 297. 968+. 0999

I ch genet aaatnamae onaceikismeaendasea can tent lads 296. 200+. 0815

Svanberg and Nordenfeldt................22.22--2-02--- 296. 671+. 0720

Marignac, synthetic. ...-.-..... 2.2.0.2... 20 cece eee eee 298. 27 +. 0149

Marignac, calcination...............-++++sseeeeeeeeeee ees 298. 30 +. 0150

General mean........ ad dteraveyeipreelorassra Siete eeeereie 298. 210+. 0103

Burton and Vorce,! who published their work on magnesium in 1890, began with the metal
itself, which had been purified by distillation in a Sprengel vacuum. This metal was dissolved
in pure nitric acid, and the resulting nitrate was converted into oxide by calcination at a white
heat. The oxide was carefully tested for oxides of nitrogen, which were proved to be absent,
but occluded gases, the impurity pointed out by Richards and Rogers, were not suspected.
This impurity must have been present, and it would tend to lower the apparent atomic weight
of magnesium as calculated from the data obtained. The results were as follows, together with
the percentage of Mg in MgO:

Mg taken. MgO formed. Per ‘cent Mg.

0.33009 0. 54766 60. 273
. 34512 . 57252 60. 281
. 26058 . 43221 60. 290
. 28600 . 47432 60. 297
. 30917 ; . 51273 60. 299
. 27636 . 45853 60. 271
. 36457 . 60475 60. 284
. 32411 , . 53746 60. 304
. 32108 . 53263 60. 282
. 28323 . 46988 60. 262

Mean, 60. 2845-4. 0027
Hence, Mg = 24.287.

The best determinations of all are those of Richards and Parker,’ who studied magnesium
chloride with all the precautions suggested by the most recent reaches. The salt itself was not
only free from oxychloride, but also spectroscopically pure as regards alkaline contaminations,
and all weighings were reduced to a vacuum. Tha first series of experiments gives the ratio
between silver chloride and magnesium chloride, and I have reduced the data to the form
2AgCl: MgCl,::100:«. The weighings and values for x are subjoined:

MgCi,. AgCl. Ratio.
1. 33550 4. 01952 33. 225
1. 51601 4, 56369 33. 219
1. 32413 3. 98528 33. 226
1. 40664 4. 23297 33. 231
1, 25487 3. 77670 33: 227

itean, 33. 226. 0013
Hence, Mg = 24.332.

The remaining series of experiments, three in number, relate to the ratio 2Ag : MgCl,, which
was earlier investigated by Dumas. For the elaborate details of manipulation the original
memoir must be consulted. I can give little more than the weights found, and their reduction

to the usual form of ratio, 2Ag : MgCl, ::100:a:

Second Series.
MgCi,. Ag. Ratio.
2. 78284 6. 30284 44,152 .
2. 29360 5. 19560 44, 145
2. 36579 5. 35989 44, 130

Mean, 44, 142-4. 0043
Hence, Mg = 24.320.

1 Amer. Chem. Journ., 12,219. 1890. 2 Zeitsch. anorg. Chem., 13, 81. 1896.

 
178 MEMOIRS NATIONAL ACADEMY OF SCIENCES. IVou. XVI.

This series gives slightly higher results than the others, and the authors, for reasons which
they assign, discard it:

Third Series.
MgCl. Ag. Ratio.
1.99276 4.51554 44.181
1.78870 | 4.05256 44.138
2.12832 4.82174 44.140
2.51483, 5.69714 44141
2.40672 : 5.45294 44,136
1.95005 4.41747 44.144

Mean, 44.138-+.0013
Hence, Mg=24.311.
The fourth series, because of the experience gained in the conduct of the preceding deter-
moinations, is best of all, and the authors adopt its results in preference to the others:

Fourth Series. _
MgCly. Ag. , . Ratio.

2.03402 4.60855 44.136
1.91048 4.32841 44,138
2.09932 4.75635 44.137
1.82041 4.12447 44,137
1.92065 4.35151 44.138
1.11172 2.51876 44,138

Mean, 44.137 + .0003
Hence, Mg = 24.311. i

These series combine with that of Dumas as follows:

DOMAS iio 2s se sates sieiccae ae eacinesiod Racine ae ee Sao ROoe 44, 261-4. 0200
Richards and Parker, second series..................22..-- 44. 142+. 0043
Richards and Parker, third series......... SG sea acerca 44. 138+. 0018
Richards and Parker, fourth series..................0.00004 44.1374. 0003

General MeaNweccn ves seissssccsssswsaeeckeseet< 44. 138+. 0003

Here the first two values practically vanish, and the third and fourth series of Richards
and Parker appear alone. Combining this figure with their value for the AgCl ratio, the sub-
joined cross ratio appears: Ag:Cl: :100: 32.842.

To sum up, we now have the following Tatton bearing upon the atomic weight of magne-

ne (1) MgSO,:BaSO,: :100: 194.003 4.021
(2) MgO: MgSO,: :100: 298.210 +.0103
(3) Per cent of water in MgSO,, 7H,O, 51.21 +.020
(4) Per cent of MgO in oxalate, 27.3665 + .0023
(5) Per cent of MgO in carbonate, 47.627 + .0018
(6) Per cent of Mg in MgO, 60.2845 + .0027
(7) 2Ag: MgCl,: :100: 44.138 +.0003
(8) 2AgCl: MgCl: :100: 33.226 +.0013

The antecedent vales for reducing these ratios are—

Ag=107. 880 +. 00016 ‘ S=382. 065 --. 00012

Cl=35. 4605+..00011 , C=12, 0025+. 00019

Ba=137. 358 +. 0023 H=1. 0078+. 00001

Hence, From ratio—

Bee 5 Pada as © pilen Soar Mg=24. 0154. 0021
SraiGude bet osc ks en eseeReaee Rhein se eauest 24, 085+. 0680
A wasuarm wee aneerso ee eewek hhmew sees see ses 365 5 Bers, 24, 2544-.01381
Greseedvara se ecidnaieeetene dd Be SR e¥e saiwiaucSestevece oe © 24, 287+. 0020
Use SSK Bioih 2 src Gace eats Sia Sst Ham Bo ae eOSR 24. 311-4: 0007
Bowaciccew tees Da. Sea. Never av yaceuentiiciita dem nee ee: 24, 332-. 0037
Dowie Bd Gade es Kvn Any ee neeade SilewaubenGeeseese 24, 394+. 0021
Bs od adeasie mone w Sitratannsiavandkenccstemee eens eae Rees 24. 7034. 0053

General mean..............02.0c eee eee eee Mg=24. 296+. 0006
No. 3.] ATOMIC WEIGHTS—CLARKE. 179

This final value is. possibly a little too low, as compared with the individual values which
are presumably the best. The figures are, however, peculiarly instructive. Ratios 2, 7, and
8, representing essentially the work of Marignac and Richards and Parker, were originally
reduced with the Stas values for sulphur, silver and chlorine. These values are Ag=107.93
Cl=35.457 and S=32.074. ‘With these figures, and using only Marignac’s data for ratio 2,
the following values for magnesium are obtained:

4 ms From MgSO, : MgO, Mg=24. 383
“4 From 2Ag: MgCl,, Mg=24. 382
From 2AgCl : MgCl, Mg=24. 371

From the general mean represented by ratio 2, Mg = 24.398, a slightly higher value.

The concordance here is much greater than in the reduction with modern values, and may
be interpreted in either of two ways. Hither the Stas values are more exact than the new values
for Ag, Cl and S, or the earlier concordance is deceptive. In short, an agreement between
determinations of atomic weight made by diverse methods, is dependent in great part upon
the antecedent values used in the computations. Concordance and discordance may be equally
deceptive.’ Illustrations of this statement are not uncommon,

ZINC.

The several determinations of the atomic weight of zinc are by no means closely concord-
ant. The results obtained by Gay-Lussac‘ and Berzelius? were undoubtedly too low, and may
be disregarded here. We need consider only the work done by later investigators.

In 1842 Jacquelain published the results of his investigations upon this important constant.®
In two experiments a weighed quantity of zinc was converted into nitrate, and that by ignition
in a platinum crucible was reduced to oxide. In two other experiments sulphuric acid took
the place of nitric. As the zinc contained small quantities of lead and iron, these were estimated,
and the necessary corrections applied. From the weights of metal and oxide given by Jacque-
lain the percentages have been calculated:

Nitric Series.

“in Zn0. Per cent Zn.
9. 917 a 12. 3138 80. 536
9. 809 12. 1800 80. 534

Sulphuric Series.

Zn, ~ End. Per cent Zn.
2. 398 2.978 80. 524
3. 197 3. 968 80. 570

 

: Mean of all four, 80. 541+. 007
Hence, Zn = 66.224.

The method adopted by Axel Erdmann‘ is essentially the same as that of Jacquelain,
but varies from the latter in certain important details. First, pure zinc oxide was prepared,
ignited in a covered crucible with sugar, and then, to complete the reduction, ignited in a por-
celain tube in a current of hydrogen. The pure zinc thus obtained was converted into oxide
by means of treatment with nitric acid and subsequent ignition in a porcelain crucible. Erd-
mann’s figures give us the following percentages of metal in the oxide:

80. 247 es 263

80. 257 . 0. 274
Mean, 80. 260+. uae

‘Hence, Zn = 65.054.

Upon comparing Erdmann’s results with those of Jacquelain two points are worth noticing:
First, Erdmann worked with purer material than Jacquelain, although the latter applied
corrections for the impurities which he knew were present; secondly, Erdmann calcined his

 

1 Mémoire d’Arceuil, 2,174. 3Compt. Rend., 14, 636.
2Gilb. Annal., 37, 460. 4Poggend. Annal.,62,611. Berz. Lehrb.,3, 1219.
180 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

zinc nitrate in a porcelain crucible, while Jacquelain uséd platinum. . In the latter case it has
been shown that portions of zinc may become reduced and alloy themselves with the platinum
of the crucible; hence a lower weight of oxide from a given quantity of zinc, a higher percentage
of metal, and an increased atomic weight. This source of constant error has undoubtedly
affected Jacquelain’s experiments, and. vitiated his resulis. In Erdmann’s nore no such errors
seem to be present. igs, gues

Favre’ employed two methods of veieatinly First, zinc was aisias in i gulphaais
acid, the hydrogen evolved was burned, and the weight of water thus formed was determined.
To his weighings I append the ratio between metallic zinc and 100 parts of water:

Zn. #0. Ratio.
25. 389 ‘ 6. 928 366. 469
30. 369 8. 297 _ 366. 024
31. 776, 8. 671 366, 463

 

_ Mean, 366. 819-4. 088
Hence Zn = 65. 995.

‘The second method adopted | by, Favre was to burn pure z zine oxalate, and to weigh the oxide
and carbonic acid thus produced. From the ratio between these two sets of weights the atomic
weight of zinc is easily deducible. From Favre’s weighings, if CO,=100, ZnO will be as given
in the third column below:

Zn0. C0, " : Ratio.

7. 796 8. 365 93. 198
7. 342 7. 883 . 98.137
5. 2065 5. 588 98,178

 

Mean, 93. 169-:..012
Hence Zn = 65.993. rey
Both of these determinations are open to objections. In the water seriés it was essential

that the hydrogen should first be throughly dried before combustion, and then that every

trace of water formed should be collected. A trivial loss of hydrogen or of water would tend
to increase the apparent atomic weight of zinc.

In the combustion of the zinc oxalate equally great difficulties are encountered. Here a
variety of errors are possible, such as are due, for example, to impurity of material, to imperfect
drying of the carbon dioxide, and to incomplete collection of the latter. Indeed, a fourth
combustion is omitted from the series as given, having been rejected by Favre himself. In
this case the oxide formed was contaminated by traces of sulphide.

Baubigny,? in 1883, resorted to the well-known sulphate method. Zinc sulphate, elabo-
rately purified, was dried at 440° to constant weight, and then calcined at a temperature equal
to the fusing point of gold. These data were obtained:

Zr80, | ZnO. Per cent ZnO.
6. 699 3. 377 50. 410
8.776 4, 4245 50. 416

 

ne 60. 413+. 0020
Hence Zn = -65. 399.

Tn: Marignac’s determinations of the atomic weight of zinc, puniahea also in 1883,$ there is a
peculiar. complication. After testing and criticizing some other methods, he finally decided
to study the double salt K,ZnCl,, which, however, is difficult to obtain in absolutely definite
condition. Although the compound was purified by repeated crystallizations, it was found
to deliquesce readily, and thereby. to undergo partial dissociation, losing chloride of zinc, and
leaving the porous layer on the crystalline surfaces richer in potassium. . In order to evade
this difficulty, Marignac placed a large quantity of the salt in a funnel, and collected, the liquid
product of deliquescence as it ran down. In this product he determined chlorine by volumetric
titration with a standard, solution of silver, and also estimated zinc by precipitation, with sodium

 

1 Ann, Chim. Phys. (3), 10, 163. 1844. § Arch. Sei, Phys. Nat. (3), 10, 194. Oeuvres Complates, 2, 731.
2Compt. Rend., 97, 906. 1883,
No. 3.] ATOMIC WEIGHTS—CLARKE. 181

carbonate, and weighing as oxide. From the data thus obtained equations were formed,
giving for each analysis an atomic weight of zinc which is independent of the proportion between
ZnCl, and KCl in the substance analyzed. The data unfortunately are too bulky for repro-
duction here and the calculations are complex; but the results found for zinc, when Ag = 107.93,
Cl=35.457, and K = 39.137, are as follows:

Le One tliration dies Ssesestescices iuccisesences st sdaeueecdme ss sa Zn=65. 22
Ds, Two titrations sso ceindaa sencansuckiss oat cwnaecoeiaadoasens 65. 37
Bo PWOTUTAtONG soc. c awa se Seaendeeu ee cine sews sees 65. 31
Ai TWO CAtPATODA = 25 ctor sfe lots w'aratarsidnweiibiejs oa waren Sakishin a tada 65. 28
Dy. ONS tiation) ssieew's seeesssk woaeSews cb ckaceemesseee s shee 65. 26

Each of these values represents a distinct sample of the deliquesced material, and the
number of chlorine determinations is indicated.

A second set of determinations was made by the same analytical method directly upon
the recrystallized and carefully dried K,ZnCl,. The values for Zn are as follows:

6. WOME ATIODS hurled Sen hocee on aiGulaien ew isnolebunee cea ones Zn=65. 28
7. “TWO UMPAtONS sos tsi sade com en eueledeasiceeetla atic 65. 39
8. One titra thon ae ooo aisaecs's Srcpeyerhs careless Sto taveesehweesere warersbats 65. 32

In order to adapt these data to the uniform scheme of calculation employed in this work,
taking into account their probable error and the probable errors of the antecedent values for
K, Cl and Ag, it seems to be best to calculate them back with the atomic weights used by
Merignac into the form of the ratio 4Ag : K,ZnCl,::100:2. Doing this, and taking each
value as many times as there are titrations represented in it—that is, giving the results of a
double determination twice the weight of a single one—we have the following series of data
for the ratio in question:

PRD pete tsi ahaha Goi Badu Lesaac ara ae aecee hee voeaese 66. 090
66. 124
HIRO Dos wens heciasnaenn as asneienitn Vrs canieldrnite ny gaint 124
66. 110
ESL sna reenter oesetie Ween eon eeeen te 110
66. 104
BRDU UR sr ssi bisa sn eho ecig ioe te ect ae ntis setbee tee 104
BY O15 ocenstjnceecicncseies’s s HeSek eked so SHEN Se em aes o aeeiats Bs 66. 099
66. 104
EERO senesasionpsarae Eo siaceneni adele s saaieiraetei = ARERR SRE tee 104
66. 129
INTOIO 7 42s NARS SS Rape eae ee ees sane Geeeses | 66. 129
Hpi Soa papi anus God ease ina esa ea eterna’ 66.113

 

Mean, 66. 111+. 0023

Hence, from Marignac’s work, 4Ag : K,ZnCl, : : 100 : 66.1114.0023, a ratio which can
be discussed along with others at the close of this chapter. It corresponds to Zn =65.245.

During the years between 1883 and 1889, a number of determinations were made of the
direct ratio between zinc and hydrogen—that is, weighed quantities of zinc were dissolved in
acid, the hydrogen evolved was measured, and from its volume, with Regnault’s data, the
weight of H was computed. First in order are Van der Plaats’ determinations,’ whose results,
as given by himself, are subjoined. The weights are reduced to a vacuum. Sulphuric acid
was the solvent:

Zn, grm. H, litres. Zn=
6. 6725 1, 1424 65. 21
9.1271 1, 5643 65. 14

13. 8758 2. 3767 65. 18

Mean, 65. 177+. 0137

With the new value for the weight of hydrogen, 0.89872 gramme per liter, this becomes Zn
64.980 + .0137, when H=1.

 

1 Compt. Rend., 100, 52. 1885.
182 MEMOIRS NATIONAL ACADEMY OF SCIENCES. LVou. XVI.

Reynolds and Ramsay made 29 determinations of this ratio,’ rejecting, however, all but 5.
The weighings were reduced to vacuum, and in each experiment the volume of hydrogen was fixed
by the mean of seven or eight readings, The values for Zn are as follows:

65. 5060 65. 5522
65. 4766 65.4141 ..
65. 4450

Mean, 65. 4787+. 0161

These values were computed with Regnault’s data for the weight of H. Corrected by the
new value the mean becomes Zn = 65.280 +.0161.

A few determinations by Mallet were made incidentally to his svotke on the atomic weight
of gold, and appear in the same paper.? According to these experiments one gramme of zinc

gives— 341. 85 cc. H., and Zn=65.'158
34.91 “4 “65.146

341.93“ “65, 143

342.04“ «65, 122

Mean, 65. 142+. 0039

In this case the Crafts-Regnault weight of H was taken, 1 liter=.08979 gramme. Corrected,
the mean gives Zn = 65.082 + .0039.

Two other series of determinations of questionable value remain to be noticed before
leaving the consideration of the direct H : Zn ratio. They represent really the practice work
of students, and are interesting as an illustration of the closeness with which such work can be
done. The first series was made in the laboratory of the Johns Hopkins University, under the
direction of Morse and Keiser,’ and contains 51 determinations, as follows:

Zn= :

64. 68 65. 74 65. 40
65. 26 64. 72 64. 80
65. 32 65. 26 65. 20
65. 20 64. 74 64. 40
65. 60 64. 72 65. 00
64. 60 65. 10 64. 40
65. 00 64. 76 65. 24
65. 68 64, 90 64. 60
65. 38 64, 92 64. 80
65. 06 64. 64 65. 14
64, 84 65. 24 64. 84
64. 88 64, 72 64. 82
65. 00 65. 20 64. 80
65.08 65. 12 64. 40
65. 06 66. 40 64. 60
64. 74 64. 60 64. 80
65. 12 65. 60. 64. 74

Mean of all, Zn=64.997+. 0328

‘Corrected for the difference between Regnault’s value for H and the new value, this becomes
Zn = 64.800 + .0328.

The second student series “was published by Torrey,‘ who gives 15 determinations, as
follows:

Zn= : iA
65. 36 64. 96
65. 30 64. 70
64. 92 65. 00
64. 72 64. 78
65. 04 64. 44
64. 80 65. 24
65. 20 64. 92

64. 90
Mean, 64. 952+. 0436

Corrected as in the other series, this gives Zn = 64.755 + .0436.

 

1 Journ. Chem. Soc., 51, 854. 1887. 3 Amer. Chem. Journ., 6,347. 1884,
2 Amer. Chem. Journ., 12,205. 1890, + ., , # Amer, Chem. Journ., 10,74. 1888,
No. 3.]

ATOMIC WEIGHTS—CLARKE.

183

‘The five corrected means for the ratio H : ZN may now be combined, thus:

H=1.
Van der Plaais...... Lieaslcitets ie sicher eeu <3 64. 980-4. 0137
Reynolds and Ramsay...............2.0- - 65, 280+. 0161
Mallet sisi tice gett rat c's tanduadaieadaucuns 65. 081+. 0039
Morse and Keiser................-. eee ee ‘64. 800+. 0328
DORCY oii o5.canpactie Vie daddseemedes vt id 64. 755+. 0036
General mean............0...0000- 65. 079+. 0036

O=16.
65. 487 ©
65. 789
65. 590
65. 305
65. 260

65. 587 +. 0036

 

Morse and Burton,’ in their determinations of the atomic weight of zinc, returned essentially
to the old method adopted by Erdmann and by Jacquelain. Their zinc was obtained spectro-
scopically pure by distillation in a vacuum, and was oxidized by nitric acid which left absolutely
no residue upon evaporation. The conversion to oxide was effected in a porcelain crucible,
which was inclosed in a larger one, and the ignition of the nitrate was carried out in a mufile.
In weighing the crucible was tared by one of nearly equal weight. Results as follows:

Wt. Zn. Wt. ZnO. Per cent Zn in ZnO.
1. 11616 : 1. 38972 80. 320
1. 03423 1, 28782 80. 308
1.11628 1. 38987 80. 315
1. 05760 1. 31681 80. 316
1. 04801 1. 30492 80. 313 .
1. 02957 1. 28193 80. 318
1. 09181 1. 35944 80. 315
1.16413 1. 44955 80. 305
1. 07814 1, 34248 80. 305
1. 12754 1.40400 80. 306

- 91112 1. 13446 80. 310
1.10011 1. 36981 80. 311
1. 17038 1. 45726 80. 313
1. 03148 1, 28436 80. 310
1. 05505 1. 31365 80. 308

 

Mean, 80, 3115+. 00084

Hence, Zn = 65.266.
Morse and Burton verified by experiment the stability of oxide of zinc at the temperatures

of ignition, and found that it did not dissociate.

They also proved the absence of oxides of nitro-

gen from the zinc oxide. The investigations of Richards and Rogers,” however, have shown that
zinc oxide prepared by ignition of the nitrate always carries gaseous occlusions, so that the atomic
weight of zinc computed from the data of Morse and Burton is certainly too low. This con-
sideration led Morse and Arbuckle ° to reinvestigate zinc oxide, with the purpose of avoiding the
indicated error. The zinc used was a portion of the sample employed by Morse and Burton,
and the process was essentially the same, except that the oxide, after weighing, was dissolved
in sulphuric acid, and the gases which were evolved were collected, measured, and analyzed.
All weights were corrected for displacement of air. The crude data are as follows:

ot
i

Wt. Zn. * Wt. ZnO. Gases, cc.
1. 19573 1, 48860 - 468
1, 03381 1. 28707 . 402
1. 06519 1. 32599 - 342
1. 05802 1.31711 - 312
1. 26618 : 1. 57619 521
1. 03783 1. 29198 . 408
1. 08655 1. 35276 412

1.11364 ' 1.38647 - 456

’

Per cent Zn, uncorrected,
80. 326
80. 323
80. 332
80. 329
80. 332
80. 329
80. 321
80. 322

Mean, 80. 327+. 0011

 

 

2 Amer. Chem. Journ., 10,311. 1888.
2 Proc. Amer. Acad., 1893, 200. ‘
3 Amer. Chem. Journ., 20, 1898. Also published as a doctoral dissertation, Johns Hopkins University, by Arbuckle.
184 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vor XVI.

The collected gases contained only nitrogen and oxygen, in varying proportions, which
were determined in each case. Uncorrected, Zn=65.328; corrected, the value ranged between
65.437 and 65.489, in mean, 65.456. The last figure corresponds to 80.358 per cent of zinc in
the oxide, an increase of 0.031. If we assume that the same proportional error existed in all
the other experiments upon zinc oxide, the several series may be corrected and combined as
follows:

Jacquelainiie cess sdscehadasedy eae donde a tiesd weeks: 80. 572-4. 0070
Brdmann tos .e0ssadasquasteawesiesasseee seseeceseeend ',. 80. 291+. 0037
Morsé;and! Burton. osrscvrouyspovecessaseeeceeeeinveeucewnsys 80. 343+. 00084
Morse and Arbuckle: ee eee 80. 358+. 0011

 

General mean 80. 349+. 00065

Here the two earlier series practically disappear, end the modern determinations alone are
retained.

The determinations made by Gledaious and Hibbard ! represent still another process for
measuring the atomic weight of zinc. Zinc was dissolved in a voltameter, and the same current
was used to precipitate metallic silver or copper in equivalent amount. The weight of zinc
dissolved, compared with the weight of the other metal throw down, gives the atomic weight
sought for. Two voltameters were used in the experiments, giving duplicate estimates for
zinc with reference to each weighing of silver or copper. The silver series is as follows, with
the ratio 2Ag : Zn : : 100 : a in the third column:

Zn. Ag. Ratio.

. 7767 2. 5589 30. 353
. 7758 2. 5589 30. 318
- 5927 1. 9551 30.316
5924 1.9551 30. 300
_ 2277 . 7517 80. 291
. 2281 7517 80. 345
. 7452 2. 4588 30. 307
7475 2. 4588 30. 401
. 8770 2. 9000 30. 241
. 8784 2. 9000 30. 290
. 9341 3. 0809 30. 319
. 9347 3. 0809 80. 339

 

Mean, 30. 318. 0077
Hence, Zn= 65.414.

To the copper series I add the ratio Cu : Zn : : 100 : 2:

Zn. | Cu. _ Ratio.
. 1767 . 7526 " 103. 13
. 7758 . 7526 108. 08
. 5927 . 5737 108. 32
. 5924 . 5737 108. 26
2277 . 2209 103. 08
. 2281 . 2209 103. 26
. 8770 . 8510 : 103. 05
. 8784 . 8510 103. 22
. 9341 . 9038 108. 36
. 9347 . 9038 108. 42

 

Mean, 103. 22+. 0261
Hence, Zn = 65.601.

Richards and Rogers,’ in their investigation of the. atomio weight of zinc, studied the anhy-
drous bromide. This was prepared by solution of zinc oxide in hydrobromic acid, evaporation
to dryness, and subsequent distillation in an atmosphere of carbon dioxide. In some experi-
ments, however, the bromide was heated in an atmosphere of nitrogen, mingled with gaseous
hydrobromic acid. All water can thus be removed without formation of oxybromides.

 

1 Journ. Chem. Soc., 65, 443, 1889. , 4 Zeitsch. anorg.Chem., 10,1. 1895.
No. 3] ATOMIC WEIGHTS—CLARKE. 185

The zinc bromide so obtained was dissolved in water and precipitated with a solution con-
taining a known amount of silver in the form of nitrate. The silver bromide was weighed on
a Gooch crucible, and the ratio 2AgBr : ZnBr, thus found. An excess of silver was always
used, and in one series of experiments it was estimated by precipitation with hydrobromic acid.
Deducting the excess thus found from the original quantity of silver, the amount of the latter
proportional to the zinc bromide was found; hence the ratio 2Ag: ZnBr,. The results with
vacuum weights are as follows:

 

 

 

Series A.
ZnBry. AgBr. Ratio.
1. 69616 2. 82805 59. 976
1. 98198 3. 30450 59. 978
1. 70920 1 2. 84949 59. 984
2. 35079 3. 91941 59. 978
2. 66078 4, 43751 59. 961
: Mean, 59. 975+. 0034
. Series B.
ZnBry. Ag. AgBr. Ag Ratio. Ag Br Ratio.
2. 33882 2. 24063 3. 90067 104, 382 59. 959
1.97142 1. 88837 3, 28742 104. 398 59. 969
2. 14985 2.05971 3, 58539 104. 376 59. 961
2. 00966 1, 92476 3. 35074 104, 411 59. 977.
Mean, 104.392 Mean, 59. 967
+. 0054 +. 0027

At the end of the same paper Richards alone gives two more series of determinations made
upon zinc bromide prepared by the action of pure bromine upon pure electrolytic zinc. The
bromide so obtained was further refined by sublimation or distillation and dried by heating in
a stream, of carbon dioxide and gaseous hydrobromic acid. Thus was insured the absence of
basic salts and water. The weights and results found in the two series were as follows:

 

Series C.

ZnBry. Ag. Ratio.
6. 23833 6. 09766 104. 379
5. 26449 5. 0436 104. 380
9. 36283 8. 9702 104, 377

; Mean, 104. 379+. 0007
Series D.

ZnBro. AgBr. Ratio.
2. 65847 4, 43358 59. 962
2. 30939 3. 85149 59. 961
5. 26449 8. 77992 59. 961

 

Mean, 59. 961+. 0004

In some details of manipulation these series differ from those given by Richards and
Rogers jointly, but their minutize are not essential to the present discussion.
Combining these several series, we have—

For 2Ag: ZnBr, £2100: 2.

Series Bic ccvnces leans cissweyceeaee eas exceed eats» «ae 104. 392+. 0054
Series! Cues Syeasieinds cn coatea ads cs oe echaaeeeeteas< 104. 379+. 0007
General mean.........----2--- eee e eee e eee .ee++-- 104. 380+. 0007

For 2AgBr : ZnBr,:: 100 : 2.
Serie6 7A. oc.ce sos cca ce clecdine niomine de seaiesbanwewes tae 59. 975+. 0034
S6N6S Bie. coeur casa ctie nannies veces ehadusicaneiss Pa agatben 59. 967 +. 0027
Series Disha, ic saignesow esis sptued scene dea Peialats geist 59. 961+. 0004

General Mean... oc.ee ss ee wetesesss sacs kere es eees 59. 962+. 0004
186 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

From the Ag ratio, Zn = 65.364.

From the AgBr:ratio, Zn = 65.375.

And Ag : Br : : 100 : 74:077..

In order to determine the atomic weight of zinc, 2, Meaglia? measured the direct ratios between
that metal and silver or gold. The silver was precipitated, from a sulphate solution by zinc,
and the gold from a solution of sodium chloraurate. From the weights obtained the following
values for zinc were computed, when Ag=107.93 and Au=197.2.

 

 

Silver Series. Gold Series.
v2 65,58 65. 509
65. 45 65, 424
65. 50 65. 440
65. 41 65. 470
_ Mean, 65, 485+. 0247 Mean, 65.436--—. 0087

From the silver ratio, with Ag = 107.88, Zn = 65.455.

From the gold ratio, with Au=197.209, Zn = 65.439.

For the ratio 2Ag : Zn, Gladstone and Hibbert’s data give the value 30.3184 .0077. Meag-
lia’s figures, reduced to the same basis, give 30.337+.0115. The two series combined give

2Ag :Zn :: 100 : 30. 3244 -0064

There is also’ a eingls measurement of the silver-zinc ratio by Pecheux,? who finds Zn = 65.44
when Ag=107.88. As no details are given this value can not be utilized here.

A different.electrolytic determination of the atomic weight of zinc has been made by Baxter
and Grose,’ who subjected zine bromide to electrolysis, and collected the zinc in a weighed
mercury cathode. Their data, with, vacuum weights, are as follows:

ZnBr,. | Zn. , Per cent Zn.
6. 88781 1, 99917 29. 025
_ 6, 79973 1.97324 29.019
7. 67096 "9, 92712 ‘29. 030
8. 02839 2, 33122 29. 037
7.18458 2.08517 29, 023
7. 55005 2.19181 29. 080
"7, 71832 "2, 23937 29. 032
7. 11551 ‘2: 06599 29, 035
7, 44099 2. 16047 29. 035

7. 18828 2. 08727 29. 037
Mean, 29. 030+. 0013

Hence, Zn = 65.384.
Baxter and Grose: ceienl the first two determinations in this series, which raises the per-

centage of zinc to 29.032.’ This is equivalent to an atomic weight of Zn=65.392, an appreciable
difference from the value given by the entire series.
For computing the atomic weight of zinc we now have the subjoined ratios:

(1)' ZnO : Zn :::100 : 80.349+.00065. iis bp oa |

(2) ZnSO, : ZnO :: 100 : 50.413 + .0020 ae oe,

(3) H,O : Zn :: 100 : 366.319 + .088 ip ag?

(4) 2CO, : ZnO :: 100 : 93.169 + .012

(5) H.: Zn ::1 : 65.079 + .0036

(6) 4Ag : K,ZnCl, :: 100 : 66.111 + .0023

(7) 2Ag :Zm :: 100 : 30.824+ .0060

(8) Cu : Zn :: 100 : 103.22 + .0261

(9) 2Ag : ZnBr, :: 100 : 104.38 + .0007
(10) 2AgBr : ZnBr, :: 100 : 59.962 + .0004
(11) Au : Zn :: 197.2 : 65.436 + .0087
(12) ZnBr, : Zn :: 100 : 29.030 + .0034.

1 Thesis, University of Grenoble, 1907. 2Compt. Rend., 154, 1419. 1912. 8 Journ. Amer. Chem. Soc., 38, 868. 1916.

ihe
No. 3.] ATOMIC WEIGHTS—CLARKE. 187

The values used in reducing these ratios are:

Ag=107. 880-4. 00016 C =12. 0025+. 00019
Cl=35. 4605+. 00011 K=89. 0975+. 00011
Br=79. 92284. 00015 Cu=63, 555 +. 00050
S=32. 0650+. 00012 Au=197. 209 +. 0029
H=1. 0078+. 00001
Hence,
From ratio: ,
Gee eal tines leona rc ae enh Zn=65. 245+. 0100
Defi etie, stadia eine aes ean iabnalne gia bal vas 65. 364+. 0016
TQ meee caaacerearueione ener eisan Coc mneln tases 65. 375+. 0022
TD rehearse aeons , 65, 884-4. 0084
DAs SaaS LAL Te tae bie era ia Senedd 65. 399+. 0046
iL este anda pou saaaueuke vitke Lepnimaanay se uae edie 65. 421+. 00044
MT od etitcper ag te hee Rate eA sh OSA er se 65. 427-4. 0129
LD ruansacac nuabuinence bawaata eeegecrahaitite Rs 65. 439+. 0083
Beh io hah leas isd dealt tee hua 65. 587 +. 0036
Brace te ta At oadil etre Mester rea east 65. 601+. 0166
Beds ne an hese atta ar tal aac aa te iis idee tanec . 65..993+. 0106
s Powe ne aoe ctuns aidanguy Auakain reas Kamae aaiince - 65..995-£. 0159
General mean..............0 22 eee eee eee Zn=65. 397 +. 0004

CADMIUM.

The earliest determination of the atomic weight of this metal was by Stromeyer, who found
that 100 parts of cadmium united with 14.352 of oxygen.! Hence Cd=111.483. This value
has now only a historical interest.

The more modern estimates of the atomic weight of cadmium begin with the work of v.
Hauer? He heated pure anhydrous cadmium sulphate in a stream of dry hydrogen sulphide,
and weighed the cadmium sulphide thus obtained. His results were as follows, with the per-
centage of CdS in CdSO, therefrom deduced:

CdSO, CdS. Per cent.
7. 7650 5. 3741 69. 209
6. 6086 . 4.5746 69. 222
7. 3821 . 6.1117 69. 245
6. 8377 4. 7336 69. 228
8.1956 5. 6736 69. 227
‘7. 6039 5. 2634 69. 220
7.1415 4, 9431 69. 217
5. 8245 4. 0335 69. 251
6. 8462 4.7415 69. 257

 

Mean, 69, 231+. 0042
Hence Cd = 111.937.

Lenssen? worked upon pure cadmium oxalate, handling, however, only small quantities of
material. This salt, upon ignition, leaves the following percentages of oxide:

Oxalate. cdo. Per cent.
0. 5128 0. 3281 63. 982
. 6552 . 4193 63. 996

4017 . 2573 64. 053
| Mean, 64. 010-4. 014

 

Hence Cd = 112.064.

Dumas‘ dissolved pure cadmium in hydrochloric acid, evaporated the solution to dryness,
and fused the residue in hydrochloric acid gas. The cadmium chloride thus obtained was dis-
solved in water and titrated with a solution of silver after the usual manner.

 

1See Berz. Lehrbuch, 5th Aufl., 3, 1219. 3 Journ. prakt. Chem., 79, 281. 1860.
2 Journ. prakt. Chem., 72, 350. 1857. 4Ann. Chem. Pharm., 113, 27. 1860.
188 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

From Dumas’ weighings I calculate the ratio between CdCl, and 100 parts of silver:

CdCi,. Ag. Ratio.
2. 369 2. 791 ‘84. 880
4. 540 5. 348 84, 892
6.177 7. 260 85. 083
2. 404 2. 841 84. 618
3. 53825 4, 166 84. 794

‘

4. 042 4. 767 84. 791
Mean, 84. 843 0260

 

Hence, Cd =112.14.

Next in order comes Huntington’s' work, carried out in the laboratory of J. P. Cooke.
Bromide of cadmium was prepared by dissolving the carbonate in hydrobromic acid, and the
product, dried at 200°, was purified by sublimation in a porcelain tube. Upon the compound
thus obtained two series of experiments were made.

In one series the bromide was dissolved in water, and a quantity of silver not quite sufficient
for complete precipitation of the bromine was then added in nitric acid solution. After the
precipitate had settled, the supernatant liquid was titrated with a standard solution of silver
containing one gramme to the litre. The precipitate was washed by decantation, collected by
reverse filtration and weighed. To the ss I append the ratio betweeen CdBr, and 100
parts of silver bromide:

CdBr,. _ AgBr. Ratto.

1. 5592 2. 1529 72, 423
*3. 7456 », 6.1724 72. 415
2. 4267 3. 3511 72, 415
*3. 6645 5. 0590 72. 435
*3, 7679 5. 2016 72. 437
2. 7938 3. 8583 72. 410
*1, 9225 2. 6552 72. 405
3. 4773 4. 7593 72. 433

 

Mean, 72, 4216 + .0028
Hence, Cd =112.17.

The second series was like the first, except that the weight of silver needed to effect precipi-
tation was noted, instead of the weight of silver bromide formed. In the experiments marked
with an asterisk, both the amount of silver required and the amount of silver bromide thrown
down were determined in one set of weighings. The third column gives the CdBr, proportional
to 100 parts of silver: /

CaBr,. Ag. Ratio.
*3, 7456 2. 9715 126. 051
5. 0270 3. 9874 126. 072
*3. 6645 2. 9073 126. 045
*3. 7679 / 2. 9888 126. 067
*1, 9225 i 1. 5248 126. 082
2. 9101 2. 3079 126. 093
3. 6510 2. 8951 126. 110
3. 9782 ‘ 3. 1551 126. 088

 

Mean, 126. 076 +.0052.
Hence, Cd =112.18.

According to Huntington’s own calculations, these experiments fix the ratio between silver,
bromine, and cadmium as Ag : Br: Cd :: 108:80:112.31.

‘In 1890, Partridge? published determinations of the atomic weight of cadmium, made by
three methods, the weighings being reduced to a vacuum standard throughout. First, Lenssen’s
method was followed, viz, the ignition of the oxalate, with the subjoined results:

CdC 04. cdo. Per cent CdO.
1, 09898 . 70299 63. 966
1, 21548 . 77746 63. 962

 

1 Proc. Amer. Acad.,17,28. 1881. 2 Amer. Journ. Sci. (3), 40, 377. 1890.
No, 3.] ATOMIC WEIGHTIS—CLARKE. 189

CcdC,0,. cdo. Per cent Cd0.
1. 10711 0. 70807 63. 957

1. 17948 . 75440 63. 959

1. 16066 - 74327 63. 959

1. 17995 . 75471 63. 964

1. 34227 - 85864 63. 968

1. 43154 . 91573 63. 970

1. 53510 . 98197 63. 968

1, 41311 . 90397 63. 971 |

 

Mean, 63. 964+.0010.
Hence, Cd = 111.809.

Secondly, v. Hauer’s experiments were repeated, cadmium sulphate being reduced to
sulphide by heating in a stream of H,S. The following data were obtained:

CdS0,. Cds. Per cent CdS.
1. 60514 1. 11076 69. 204
1. 55831 1. 07834 69. 197
1. 67190 1. 15669 69. 185
1, 66976 1. 15554 69. 200
1. 40821 . 97450 69. 202
1. 56290 1. 08156 69. 205
1, 63278 1, 12985 69. 194
1. 58270 1, 09524 69. 198
1. 53873 1. 06481 69. 201
1. 70462 1. 17962 69. 201

Mean, 69. 199 +.0012
v. Hauer found, 69. 231+.0042

General mean, 69. 202 +.0012
The Partridge series alone gives Cd = 111.721.
In the third set of determinations cadmium oxalate was transformed to sulphide by heating
in HLS, giving the ratio Cd,CO,: CdS: : 100: 2:

Cd,CO,. cds. Per cent CdS.
1. 57092 1. 13065 71. 972
1. 73654 1. 24979 71. 973
2. 19276 1. 57825 71. 974
1. 24337 . 89492 71.974
1. 18743 . 85463 71. 975
1. 54038 1. 10858 71. 968
1. 38905 . 99974 71. 976
2. 03562 1. 46517 71. 979
2.03781 1. 46658 71. 970
1. 91840 1. 38075 71.971

 

Mean, 71. 973-4.0007
Hence, Cd = 111.59.

This work of Partridge was presently discussed by Clarke,’ with reference to the concordance
of the data, and it was shown that the three ratios determined could be discussed algebraically,
giving values for the atomic weights of Cd, S and C when O=16. These values are

Cd = 111.7850
C = 11.9958
‘S$ = 32.0002

and are independent of all antecedent values except that assumed for the standard, oxygen.
Morse and Jones,” starting with cadmium purified by fractional distillation in vacuo
adopted two methods for their determinations. First, they effected the synthesis of the oxide
from known weights of metal by dissolving the latter in nitric acid, evaporating to dryness,
and subsequent ignition of the product. The oxide thus obtained was thought to be completely

 

1 Amer. Chem, Journ., 13,34. 1891. 2 Amer. Chem. Journ., 14, 261. 1892.
190 MEMOIRS NATIONAL: ACADEMY OF SCIENCES. [Vou. XVI.

free from oxides of nitrogen. The weighings, which are given below; were made in tared cruci-
bles. The third column gives the percentage of Cd in CdO:

Cd taken. cdo found. Per cent Ca.
1. 77891 2. 03288 87. 507
1, 82492 2. 08544 87. 508
1. 74688 1. 99626 87. 507
1. 57000 1, 79418 87. 505
1. 98481 2. 26820 87. 506
2, 27297 2. 59751 87. 504
1. 75695 2. 00775 87. 508
1. 70028 1. 94305 87. 505
1. 92237 2. 19679 87. 508
1, 92081 2. 19502 - 87. 508

 

rans 87, 5066 +.00032
Hence, Cd = 112.068.

The second method employed by Morse and Jones was that of Lenssen with cadmium
oxalate. This salt they found to be somewhat. hygroscopic, a property against which the oper-
ator must be on his guard. The data found are as follows:

cdC,0,. cao. Per cent CdO.
1. 53937 . 98526 64. 004
1. 77483 1. 13582 63. 996
1. 70211 1. 08949 64. 008
1. 70238 1. 08967 64. 004
1. 74447 1.11651 64. 003

2G Bes Mean, 64. 003+. 0042
Hence, Cd =112.03

Lorimer and Smith, like Morse and Jones, determined the atomic weight of cadmium by
means of the oxide, but by analysis instead of synthesis. ‘Weighed quantities of oxide were
dissolved in potassium cyanide solution, from which metallic cadmium was thrown down elec-
trolytically. The weights are reduced to a vacuum standard:

CdO taken. Cd found. Per cent Cd.
. 34767 . 30418 87.491
. 41538 . 36852 87.515
1. 04698 . 91618 87. 507
1. 04066 . 91500 87. 493
1. 26447 1. 10649 87. 506
. 78493 . 68675 87. 492
. 86707 . 75884 87.518
. 67175 . 58785 87.510
1. 44362 1, 26329 87. 508

Mean, 87. 5044+. 0023
Hence, Cd = 112.042.

Mr. Bucher’s dissertation ? upon the atomic weight of cadmium, does not claim to give any
final measurements, but rather to discuss the various methods by which that constant has been
determined. Nevertheless, it gives many data which seem to have positive value, and which
are certainly fit for discussion along with those which have preceded this paragraph. Bucher
began with cadmium purified by distillation nine times in vacuo, and from, this his various com-
pounds were prepared. His first series of determinations was made by reducing cadmium
oxalate to oxide, the oxalate having been dried fifty hours at 150°. The reduction was effected
by heating in jacketed porcelain crucibles, with various precautions, and the results obtained,
reduced to a vacuum, are as follows:

Oxalate. Oxide. Per cent oxide.
1. 97674 1. 26414 63. 951

ms 1, 94912 1. 24682 63. 968
1. 96786 1. 25886 68. 971

 

1 Zeitsch. anorg. Chem., 1, 364. 1892,
1“ An examination ofsome methods employed in determining the atomic weight ofcadmium,’’? Johns Hopkins University doctoral disserta-
tion. By John E, Bucher. Baltimore, 1895.
No. 3] -ivce ., ATOMIC WEIGHTS—CLARKE., » 191

' Oxalate. . Oxide. Per cent oxide.
1. 87099 1. 19675 63. 958
1. 37550 . 87994 63. 972
1. 33313 . 85308 63. 991
1. 94450 1, 24452 . 64..002
2. 01846 1. 29210 64. 014

Mean, 63. 978+. 0052
Hence, Cd = 111.89.

Combing this with the means found by previous experimenters, we have for the percentage
of oxide in oxalate—

Lenssen.,...2..........20222005 se jasd ch acbtatanain clauee oyacoyeevevececanetars 64. 010+. 0140
Partridge......22....cs2eeeeeeeeeeeeeees ice eaueraaenct 63. 964-4. 0010
Morse and Jones... 2... 2.2 ecu. seen eee eeeee cuwien « emeeseane . 64. 003+. 0042
Bucher scngeng 2 sss saueaeere nya aS mestens se ececeseececeee 63. 9784. 0052

Guest aie ee Bd Aes eee 63. 966+. 0010

Bucher’s next series of determinations was by Partridge’s method—the conversion of cad-
mium oxilate into cadmium sulphide by heating in a stream of sulphuretted hydrogen. The
sulphide was finally cooled in a current of dry muPeen The vacuum weights and ratios are
subjoined:

Oxalate. |. _ Balphide,., Percentage.
2.56319 1.84716 72.065
2.18364 57341 | 72.055
2.11643 1.52462 72,037
38. 13105 (1 «2, 25582 72. 047

Mean, 72. 051+. 0127
Partridge found, 71. 973+. 0007
General mean, 71. 974+. 0007

Here Bucher’s mean practically vanishes. Taken alone, it gives Cd=112.14.

The third method employed by Bucher was that of weighing cadmium, chloride, dissolving
in water, precipitating with silver nitrate, and weighing the silver chloride found. The cad-
mium chloride was prepared, partly by solution of cadmium in hydrochloric acid, evaporation
to dryness, and sublimation in vacuo; and partly by the direct union of the metal with chlorine.
The silver chloride was weighed in a Gooch crucible, with platinum sponge in place of the asbes-
tos. To the vacuum weights I append the ratio 2AgCl : CdCl, : : 100 : z.

CdCl. 5 AgCl. Ratio.
3. 09183 4, 83856 63. 900
2. 26100 3. 53854 63. 896
1. 35729 2. 12431 63. 893

_ 2. 05582 3, 21727 63. 899
1. 89774 _ . 2.97041 63. 886

_ 8. 50367 , 5.48473 63. 880
2. 70292 "4, 23087 63. 886
4, 24276 6. 63598 63. 936
8. 40200 5. 32314 63. 910
4. 60659 7. 20386 63..946

., 2.40832 3. 76715 63. 930
2.19144 ‘ 4) 8.42724 63. 942
2, 84628 4, 45477 ———, 63, 893
" 2.56748 4.01651 ° ” 63. 923
2. 31008 - 3.61370 63. 924
1. 25008 1. 95652 63. 893
1. 96015 3. 06541 63. 944
2, 29787 3. 59391 63. 938
1. 94227 3, 03811 63. 915
1. 10976 1.73547 . 63. 946
1. 63080 2. 55016 63. 949

‘ Mean, 63. 916+. 0032
Hence, Cd =112.314.
161566°—20——138
192 MEMOIRS. NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Bucher gives a rather full discussion of the presumable errors in this method, which, however,
he regards as somewhat compensatory. The series is followed by a similar one with cadmium
bromide, the latter having, been sublimed in vacuo. Results as follows:

‘CdBr,. AgBr. Ratio.
4. 39941 6. 07204 72. 454,
3. 18030 4. 38831 72. 472
3. 60336 4, 97150 72. 480
4. 04240 5. 58062 72. 453
3. 60505 4. 97519 72. 461

 

Ms ou Mean, 72. 464+. 0035
Hence, Cd = 112.33. »

In order to fix a minimum value for the atomic weight of cadmium, Bucher effected the
synthesis of the sulphate from the metal. 1.15781 grammes of cadmium gave 2.14776 of
sulphate.

Hence, Cd = 112.35. '

' The sulphate produced was defadl at 400°, and aad examined for free sulphuric
acid, giving a correction which was applied to the weighings. The corrected weight is given
above. Any impurity in the sulphate would tend to lower the apparent atomic weight of
cadmium, and therefore the result is believed by the author to be a minimum.

Finally, Bucher examined the oxide method followed by Morse and Jones. The syntheses
of oxide were effected.in double crucibles, first with both crucibles porcelain, and afterwards
with the small inner crucible of platinum. Two: experiments were made by the first method,
three by the last. Weehe eae percentages (Cd in CdO) as follows:

 

Cd. : cdo. Percentage.
1.26142 1. 44144 (87. 511

{ . 99785 1.14085 - 87.504 —

Mean, 87. 5075+. 0024

“Cd. cdo. ~ Percentage.
1. 11321' 1. 27247 87. 484
1. 02412 1. 17054 87. 491
2. 80966 3. 21152 . 87. 487

 

’ Mean 87. 4873+. 0016

The two means given above, representing work done with porcelain and with platinum
crucibles, correspond to a difference of about 0.2 in the atomic weight.of cadmium. Experiments
were made with pure oxide of cadmium by converting it into nitrate and then back to oxide,
exactly as in the foregoing syntheses. In each case the oxide obtained at the end of the opera-
tion represented an increase in weight, but the increase was greater in platinum than in por-
celain. Hence the weighings of cadmium oxide in the foregoing determinations are subject
to constant errors, and cannot be trusted to fix the atomic weight of cadmium.

A different class of determinations relative to the atomic weight of cadmium are those of
Hardin,! who effected the electrolysis of the chloride and bromide, and also made a direct
comparison between cadmium and silver. |The aqueous solutions of the salts, mixed with
potassium cyanide, were electrolyzed in platinum dishes. The cadmium which served as the
starting point for the investigation was purified by distillation in hydrogen. All weights are
reduced to a vacuum. ‘The data for the chloride series are as follows, with a column added for
the percentage of Cd in CdCl,:

Weight CdCl,. Weight Cd. Per cent Cd.
0. 43140 0. 26422 61. 247
. 49165 ', 80112 61, 247
. T1752 . 43943 61. 241
- 72188 . 44208 61. 241
. 77264 . 47319 61. 245

 

1 Journ. Amor. Chem. Soc., 18, 1016. 1896.
No. 3.] ATOMIC WEIGHTS—CLARKE. 193

Weight CdCi,. Weight Cd. Per cent Cd.
. 81224 - 49742 61. 240
- 90022 +. 55135 61. 246
1. 02072 - 62505 61. 236
1. 26322 . 77365 61. 244
1. 52344 . 93314 61. 252

 

‘ Mean, 61. 244+. 0010
Hence, Cd = 112.07.

The results for the bromide, similarly stated, are these:

Weight CdBr, Weight Cd. Per cent Cd.

0. 57745 0. 23790 41. 198
. 76412 . 31484 41. 203

. 91835 . 37842 41. 207
1. 01460 . 41808 41. 206
“1.15074 47414 41. 203
1. 24751 51392 41. 196
1. 25951 . 51905 41. 210
1.51805 . 62556 41. 208
1. 63543 . 67378 41. 199
2. 15342 - 88722 41. 200

 

z Mean, 41. 2034-. 0010
Hence, Cd= 112.01.

The direct comparison of cadmium and silver was effected by the simultaneous electrolysis,
in the same current, of double cyanide solutions. Silver was thrown down in one platinum
dish and cadmium in another. The process was not altogether satisfactory, and gave divergent
results, those which are cited below having been selected by Hardin from the mass of data
obtained. I have added in a third column the cadmium proportional to 100 parts of silver:

Weight Cd. Weight Ag. Ratio.
0. 12624 0. 24335 51. 876

. 11032 . 21262 51. 886

. 12720 . 24515 51. 887

. 12616 . 24331 51. 852

. 22058 . 42520 51. 877

5 Mean, 51. 8764. 0041
Hence, Cd =111.93.

The work of Morse and Arbuckle! upon the atomic weight of cadmium was similar in charac-
ter and purpose to their work upon zinc. The presence of occluded gases in the oxide was
recognized, and in the new determinations they were extracted, measured, and analyzed.
Cadmium was converted into oxide, and corrections for the gaseous impurities were applied.
The vacuum weights of metal and oxide are given below, together with the volume of extracted
gas, and the crude, uncorrected percentage of Cd in CdO:

Weight Cd. Weight Cdo. Gas cc. Per cent.
1, 931882 2. 207639 0. 574 87. 509
1. 679348 1. 919096 . 480 87. 507
1, 484296 1. 696195 . 441 87. 507
1. 364861 1. 559717 - 402 87. 507
1. 502948 1.717441 . 419 87.511
1. 438035 ; 1. 648297 - 431 87. 509
1, 440410 1. 646037 - 406 87. 508
1. 459384 1. 667714 . 421 87. 508
1. 403791 1. 604196 . 390 87. 507

Mean, 87. 508+. 0003

This gives Cd=112.082. Corrected for occluded gases, Cd = 112.377 in mean, ranging from
112.359 to 112.395. The correction adds 0.029 to the percentage of metal; and if we assume the
same correction to the older determinations of this ratio, the several series combine as follows.

 

1 Amer. Chem. Journ,, 20, 536, 1898, See also Arbuckle, Thesis, Johns Hopkins University, 1898,
194 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Morse and Jones.....-..-.....2.202ee eee ee eee eee eeeeeee 87. 5356 +. 0003
Lorimer and Smith........00 0.0.2.2 ieee eee eee eee eee 87. 5334+. 0023
Bicher, (soc ewcumanseaecniass 2. aoa adecee seca ee kovasts 87. 5865+. 0024
IBUCh EHS? 4 Eocene tact tached Chae oes 87. 51634. 0016
Morse and Arbuckle..................2222 02 ee eee eee e ees 87. 53704. 0003

General Means s cscacicsciseekes steed deeseaneeesses 87. 5360+. 0002

This combination is equivalent to a rejection of all the data except those of Morse and his
colleagues. te

Baxter and Hines,’ in order to detsruiee the atomic weight of cadmium, reanrtad to the
analysis‘ of the chloride, with all the precautions characteristic of the Harvard laboratory.
First, the gravimetric ratio 2AgCl: CdCl, was determined, with the subj joined results, Vacuum
weights are given throughout:

Weight CdCl, Weight AgCl. Ratio.
5. 53421 8. 65356 63. 953
7. 77758 12. 16166 63. 952.
8. 87917 13. 88344 63. 955

Secondly, the ratio 2Ag : CdCl, was measured by adding to the solution of the cadmium
salt as nearly as possible its exact equivalent of a standard silver solution, and then determining
the slight excess of silver or chlorine by titration. The results are as follows:

Weight CdCl). Weight Ag. _ Ratio.
4. 92861 7 5. 80063 84. 967
3.86487" 4, 54891 - 34. 963
5, 08551 5. 98569 84. 961

_ 5, 84335. 6. 87704 84. 969 4
5. 99952 7. 06084 84. 969 d
3. 73092 4, 39095 84. 968

A year later, the same ratios were remeasured by Baxter, Hines, and Frevert.?_ I subjoin
their data:

Weight CdCl. Weight Ag. Weight AgCl. Ag ratio. . AgCl ratio.
5. 62500 6.61998 = .......-. 84.972  ..eeeee
6. 81031 8. 01496 10. 64918 84. 970 63. 9515
5. 50089 6. 47393 8. 60174 84. 970 63. 9509
6.11750 2... 9.56590 — ...... 63. 9511

These series are so nearly identical and so short that it seems well to treat both investiga-
tions as one. On this basis, 2Ag CdCl, :: 100 : 84.9677,+. 0008, and 2AgCl : CdCl, :: 100 :
63.9523,+. 0004. ,

Hence, from the Ag ratio, Cd =112.41.

From the Ag(Cl ratio, Cd =112.42.

And Ag : Cl :: 100 : 32.861.

Baxter, Hines and Frevert also made analyses of cadmium bromide by the usual Harvard
methods. Their data follow:

 

Weight CdBr.. Weight Ag. Weight AgBr. Ag ratio. AgBr ratio.

11. 46216 9. 08379 15. 81319 126, 182 72. 485
6. 82282 5. 40724 9. 41267 126. 182 72. 486

6. 75420 5. 35277 9. 31830 126. 181 72. 483

7. 08588 5. 61597 9. 77649 126. 174 72. 479

5. 13859 4. 07226 7. 08933 126. 183 72. 483

5. 84324 4, 63072 8. 06130 126. 183 72. 485

; 5. 99704 4, 75259 8. 27360 126. 183 72, 484
5. 90796 4. 68200 8. 15070 126. 183 72. 484
Mean, 126, 181 72, 4836
+. 0009 +. 0005

From the Ag ratio, Cd = 112.40.

 

1 Journ, Amer. Chem. Soc., 27, 222, 1905. 2 Journ. Amer. Chem. Soc., 28, 770. 1906.
No. 3. ATOMIC WEIGHTS—CLARKE. 195

From the AgBr ratio, Cd = 112.41.
And Ag : Br : : 100 : 74.082.
The four ratios determined by Baxter and his colleagues may now be combined with the

earlier determinations, as follows:
Ratio 2Ag : CdCl, : : 100; x.

Dumas........ Soe Renn as Roatan eek eae 84, 843 +, 0260
Baxter; €tC).a8 se Haeaieta cee san cheese dhe aga ms aa aineia tees 84. 9677-4. 0008
General mean..............------22-- eee eee eee ee 84. 9677+. 0008

Ratio 2AgCl: CdCl, : : 100 - x
Buchetscoesesei's seeensetencieansds Raed doth eae eters 63.916 +. 0032
Baxter, etc. .......-.....--.-56- ctapnteiiatn acti id a aiemtlats  Iate 63. 9523+. 0004
Goneralemean wireless coc eenkccniioatesempesecanees 63. 9518+. 0004

Ratio 2Ag : CdBr, : : 100: x. :

Huntington....... Faget a6 a PregtsacStS rated g, Ste aca paveecseisyasr oes s 126. 076+. 0052
Baxter, 6t@ is cccaccasnaceaenacsuntaswaces sabpesuieeee oat 126, 181-4. 0009
General mean..........-. 2-2. 2-222 ee eeee reece eee 126. 178+. 0009

Ratio 2 AgBr : CdBr, : : :100: x.
Huntiugtons ceccubanienvvsssidawcsal Oise adcamataoamecsd 72, 4216+. 0028
Buch OP ssi verncies sae dance derdit@ennes ss oseuiemeasasan’ 72. 464 +. 0035
Baxter, ¢te- 0% secs se saasmineawees os ives seed ease eee oe 72, 4836+. 0005
General mean...........-0--22-02e eee eee eee eee ee 72. 4813+. 0005

The earlier determinations of these four ratios practically vanish.

The determinations of the atomic weight of cadmium by Meaglia! were based upon the
quantitative precipitation by that metal of silver from a sulphate solution, and gold from a
solution of sodium chloraurate. With Ag=107.93 and Au=197.2 the following values for
cadmium were obtained:

Silver series. Gold series.

 

112. 37 112, 41

112. 56 121. 45

112, 45 112. 65

112, 38 112. 47

112, 48

Mean, 112. 44+. 0295 112. 40
112, 42

112. 41

Mean, 112. 461+. 0196

From the silver series, with Ag=107.88, Cd=112.39.

From the gold series, with Au=197.209, Cd=112.46.

The determinations made by Blum? depended upon ‘the conversion of CdO into CdS
by heating in a stream of hydrogen sulphide. His figures, with vacuum weights, are given
below, together with the ratio CdO : CdS : : 100: a: :

cdo. Cds. Ratio.
1. 80552 ° 2. 03108 112, 493

. 66349 . 74617 112. 461
1. 82460 2. 05256 112. 494
1. 88424 2. 11974 112. 498
3. 59206 4. 04081 112, 493
4. 38093 4, 92695 112. 464

 

at anioind Mean, 112. 484+. 0046
Hence Cd =112.69. This ratio is not of much value.

In 1911 Perdue and Hulett * proposed a new method for the electrolytic determination
of atomic weights, and in a second paper‘ applied it to the analysis of cadmium sulphate,

 

1 Thesis, University of Grenoble, 1907. § Journ. Phys. Chem., 15, 146. 1911.
2 Thesis, University of Pennsylvania, 1908. 4 Tbid., p. 155.
196 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

8CdSO,.8H,0. The method, briefly, consisted in the electrolysis of a. salt containing a metal
soluble in mercury, and in collecting the metal on a weighed mercury cathode. With cadmium
sulphate the electrolysis was preceded by a careful determination of the water in the hydrated
salt. With vacuum weights the figures for the water series are as follows:.

$CdS80,. 8H,0.

6, 32863
6. 72493
6. 87537
5, 65027
6, 81125
7.34977
7. 74837
7, 8843
6. 6100

Hence, Cd = 112.35.

#0.
1. 1856
1, 25986
1, 2886
1, 05822
1, 27557
1, 37703
1. 3572
1. 47713
1. 2480

The two electrolytic series, for the hydrated

3Cd80,.8H,0.
7.90902
9.07468
7.32787
6.48847
5.11684
8.02954
5.08743

Hence, Cd = 112.31.

CdSO,.
5.14303
5.46507
5.58667
5.53568
6.29717
6.40718
5.37196

Hence, Cd = 112.256.

Cd,

3.46335 .

3.97434
3.20936
2.84186
2.24157
3.51755
2.22827

Cd.
2.77196
2.94566
3.01076
2.98276
3.39295
3.45255
2.89457

Per cent H,0.
18.734
18, 734
18, 742
18. 729
18. 727
18, 736
18, 727
18, 734
18. 730:

Mean, 18, 733+:. 0011

 

and anhydrous sulphate are as follows:

Per cent Cd,
43.790
43.796 ©
43.796
43.799
43,808
43.807
43.799

Mean, 43.7994. 0024

 

Per cent Cd.
53.897
53.898
53.891
53.883
53.880
53.887
53.883

Mean, 53.888-4+ 0018.

The foregoing determinations were promptly criticized by Richards,’ who suggested that
the cadmium sulphate might have retained traces of water, perhaps in ‘‘solid solution,’’ which
would lead to a low value for the atomic weight of cadmium.

A year later Laird and Hulett ? applied the new method :to the direct comparison of cad-
mium and silver. The cadmium was deposited in duplicate cells, the silver is only one. Their

data follow; with the ratio 2Ag :: 100: x.

Ca,
2.2721
rent
2.3549
2.3534
2.6637
2. 6656
1.9987
7 asst
2.2979
2.2972
2. 4265
sont

Ag.
4, 3638

4.5225
5.1186
3.8312
4.4148

4.6605

Ratio.
52.067
ie oe
52.071
ace
52.040
19.077
52.039
52.044
52.050
ei bed
52.065
ante

 

1Journ. Amer. Chem. Soc., 33, 888,

2 Trans. Amer. Electrochem. Soc., 22, 385. 1912.
No. 3.] my

Ca.
2.1496
2. rt
2. 6629
Se

Hence, Cd = 112.31.

Ag.

‘4.1297

5.1256

ATOMIC WEIGHTS—CLARKE. 197

Ratio.
52, 052
co 064
52.064
i 043

Mean, 52.055+. 0007.

 

Combining this mean with previous determinations we have—

Hardin.......- fain bad oe aia G Separsieaedag ae ta wGamoneuwaee bee 51.876+.0041
MeAIB. cinsbnmmenecnacs mens Sdeuiititaveaseeeeneees 52. 090-4. 0136

. Laird and Hulett.................- pincers eta eoauae 52.055 +. 0007
General mean..........2. 0202 geeeeeeeeeee sree eee 52. 050-4. 00069

In the following year the investigation was continued by Quinn and Hulett, who made

electrolytic analyses of cadmium chloride and bromide.

follows:
CdCl ..
'4.10119

4.47151 9
5. 66452
5. 99291.
6.52910

4, 87415

4, 88423

Hence, Cd = 112.32.
The bromidé data are—

CdBrp.
4, 95538
4.15801
5, 31272
4, 55544,
4.31525
3. 69646
4.47748
8. 36276

Hence, Cd = 112.26.

Cd.
2.51427
2.74084
3.47180
3.67211
4.00219
2.98785
2.99384

‘Cd,

2.04468
1.71573
2.19183
1.87940
1.78013
2.34983
1.84754
2.21228

For the chins the data are as

Pert. Ca,

61.306

61.295 ;
61.290" ge ent Sith
61,274 Rejected.

61. 298

61.300

61.296

 

Mean, 61.297+.0010

oa

Per cent Cd.
41.261
41.263
41.256
41.256
41.252
41.251
41.263
41.253.

Mean, 41.257+.0010

 

The arithmetic mean of the six values thus found for cadmium is 112.30, or about one-
‘tenth of a unit lower than the value determined by Baxter and his associates. Because of
the doubt thus thrown upon ‘the Harvard determinations Baxter and Hartmann * made a
series of analyses of cadmium chloride by the method of Hulett and Quinn, ‘put with various
precautions against error. For the elaborate details of these analyses the original paper should

be consulted. The eae wage: vacuum weights, are as follows: =

CaCl,.
6.08570
4, 20489
_ 5.36208
7.50512
6.71591
_ 6.91556
6. 49323
3.14416
7.58705

Cd,
3.73181
2.57863
3.28817
4.60221
4.11839
3.62753
3.36805
1.92791
4.65173

ot

Preliminary series.

Per cent Cd.
61.321
61.295
61.323.
61.321
61,323
61.323
61.313,
61.317
61. 312

 

Journ. Phys. Chem,, 17,779. 1913.

2 Journ. Amer. Chem. Soc., 37, 113. 1915.
198 MEMOIRS NATIONAL ACADEMY ‘OF ‘SCIENCES. [Vou. XVI,

CaCl. Cd. Per: cent Cd.
2.26738 1.39006 61.307
5. 93501 3.63962 61.325
6.04122 3.70490 61.327
4.07400 2.49821 61.321

 

ce Mean, 61.3184.0016
Hence, Cd = 112.42. mete ia

ty

Final series. .

CdCl,. Cd, Per cent Cd.

" 9, 00004 5.51879 61.320"
6.56891 4.02808 * \-61,320 :
7.12956 4.37174 , 61.819
8.57291 "5.25679 61.318

_ 7.76294 a, A6OIL ot “161.820

 

Phy Mean, 61.3194 4.00027

Hence, Cd = 112.43. -

In continuation of this research, Baxter, Grose, and Hartmann * made new 7 electrolyte
analyses of cadmium chloride, and also a series of similar analyses of the bromide. For the
very short chloride series the figures are:

CdCl,. ' Cd. Per cent Cd.
6. 49746 3. 98430 ‘61. 321
6. 62682 4. 06348 61. 319
7. 35812 4, 51165 ° 61.317

 

: Mean, 61.319, 0011
Hence, Cd = 112.43. ate
For the bromide analyses the data are as follows:

CaBro. Cd, i Per cent Cd.
16. 55613 6. 83575 41, 287
6. 83764 2. 82342 41, 284
7. 16223 2. 95696 ' 41, 286
8.71886 “2.77414 41, 289
10. 31855 4, 26112 41, 296
9. 40020 8. 88142 41,291
8. 21123 3.39038. 41, 290
10. 51504 4, 34154 41, 289
10, 22411 | 4, 22157 41, 290
10. 99613 4.54013 41. 288
_ 1105166 ee 1, 41,290
19. 67134 SS 'g.19244 "41, 291°

 

un ay, fs Mean, 41, 289-4. 0008

Hence, Cd = 112.41.
The several series of analyses of cadmium halides may now be combined as follows:

Ratio CdBry: Cd:::100,: m2) :+ a ts Reso Ne aps
Hardin ese ote cow escewe ew eee eS Seas serra aaeidee aie 41. 203+. 0010

 

Quinn and Hulett............... 20.2. ene eee exonerate 41, 257+. 0010
Baxter, Grose, and Hartmann...................0...020 ee 41, 289+. 0008
ey ; ——
General mean... 0.2... cece een eee eee eee eee eee 41; 2554. 00054
Ratio CaCl, : ‘Cd: 100 3x.
Hardin. /....000eeee cee e eee ene eee eee eee ence ee ee 61.244 4.0010
Quinn and Hulett.............052. 00 cece eee eee eee eee 61.297 +. 0010
Baxter and Hartmann, preliminary...................--- 61.318 +. 0016
Baxter and Hartmann, final....................+--2+-- 61. 3194+. 00027
Baxter, Grose, and Hartmann......- oSeaachea accra ae se 61.319 +.0011
General mean. 22 ciseeoudidsie saseeauaesecnecen s 61. 31374. 00024

 

Journ, Amer. Chem. Soc., 38, 857. 1916. ,
No. 3.] ATOMIC WEIGHTS—CLARKE. 199

The determinations of the atomic weight of cadmium by Oechsner de Coninck and Gérard’
are of slight importance, but they should not be ignored. Cadmium carbonate was reduced to
metal by heating in a stream of hydrogen, with the following results:

‘CdCO, Cd. Per cent Cd.
0. 5500 0. 3585 65. 182

. 6050 . 3943 65. 174

. 6600 - 4302 65. 182

- 6171 - 4023 65, 192

. 5445 . 3849 65. 179

Mean, 65. 182+. 0025
Hence, Cd = 112.33.

For computing the atomic weight of cadmium the following ratios are now available:

Bucher’s single experiment upon the synthesis of the sulphate, although important and
interesting, cannot carry weight enough to warrant its consideration in connection with the
other ratios, and is therefore not included.

(1) CdO : Cd :: 100°: 87.536 + .0002.

(2) CdCO, : Cd :: 100 : 65.182 + .0025.

(3) CdC,O, : CdO :: 100 : 63.966 4.0010.

(4) CdC,O, : CdS :: 100 : 71.974 +.0007.

(5) CdO : CdS :: 100 : 112.484 + .0046.

(6) CdSO, : CdS :: 100 : 69.202 + .0012.

(7) CdSO, : Cd :: 100 : 53.888 +.0018.

(8) 83CdSO,.8H,O : 3Cd°:: 100 : 43.799 +.0024.
(9) 3CdSO,.8H,O : 8H,O :: 100 : 18.733 4.0011.
(10) 2Ag : CdCl, :: 100 : 84.9676 +.0008.
(11) 2AgCl : CdCl, :: 100 : 63.9518 +.0004.
(12) 2Ag : CdBr, :: 100 : 126.178 +.0009.
(13) 2AgBr : CdBr, :: 100 : 72.4813 + .0005.
(14) CdCl, : Cd :: 100 : 61.3137 + .00024.

(15) CdBr, : Cd :: 100 : 41.2555 + .00054.

(16) 2Ag : Cd :: 100 : 52.050 + .00069.
(17) Au : Cd :: 197.2 : 112.461+4.0196.

To reduce these ratios we have

Ag=107. 880+. 00016 S =32. 0650+. 00012
Cl =35, 4605+. 00011 C =12. 0025+. 00019

Br=79. 9228+. 00015 Au=197. 209+. 0029
H=1. 0078+. 00001

Hence,
From ratio—

Ae cep cits eae ayieeeeeasesiss saeco cet d Cd=111, 596+. 0051
Orasetetesedenats'ns septic pee aeeje ss ¢ pornetee eee 111. 7414. 0061

Sus wardg adenomas cacteeroacdamee cen see eee 111. 820+. 0041
Txieibacdsislcbisiels <in'geletslatstavseiciclettals's Nin cnarclahaaheleis 112, 255+. 0059

BL sates Ager Gece tale d creetasrveste erect e S Poa gaan ed hte 112. 258+. 0018

TMG ie cerseccaceicczetasate < oibionde oh ieee se at owas Recher 112. 3034. 0015

B ccccinse uncer ashe aces a 6-4 weasern eS Satna so alewarsemeneien 112. 306+. 0078

De Sas Sava steiiasanshe ona Oa VA ORR CR EET Hateee eames 112. 329+. 0091

eevee he aneia aaah seaias oa. dca ea nieieinualpasubel Rattus cial IAS ora 112. 348+. 0126

TL citceanu wise s one sade ee wena sees eonaaeeeeace 112. 3704. 0018

EB Sz) sias ate. ayatacatelasetsvo'a eis jaidysceiaricuece ese Sse! a piaubreignsuataleieieree 112. 397+. 0019

DA creo erspepeSceesre Uae Smee Mine taceraye aes Stepan erties woe 112. 402+. 0009
Toon cen cats adine sewecasee buwskivede es spivesGewiewes< 112. 402+. 0020
DO poses cane seeswesosvan ve cseseeuaseaxsaneweceee dee 112. 405+. 0018

LD casivircccceene ss os eedesisiowpeex sss ¥eaeneeeceeess 112. 407 +. 0012

TT anaes Moora dale y ss netadee pert ctst Ca Maeea tsa 112. 466+. 0197
Busienat atumcantage yar Huninad pobndvieter fice Nhe dee a 112. 688+. 0150
General MEAN is 25. occa wee oe Hesedecs 112. 360+. 0004

 

1 Compt. Rend., 161, 676. 1915.
200 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Rejection of the first three values raises the general mean to 112.376... The last two values,
with their high probable errors, count for practically nothing.

It will be seen, from, a comparison of the individual ratios, that the abet recent Pens
tions of the atomic weight of cadmium fall into two groups. Those of Hulett and his colleagues
are concordant, and give Cd=112.3 very nearly. Those of Baxter and his associates, on the
other hand, are even more closely concordant, and give Cd=112.4. This disagreement has
led to some controversy. The criticism by Richards of the work of Perdue and Hulett has
already been noticed, and the electrolytic determination by Laird, Quinn, and Hulett have also
been discussed by ] Baxter and Hartmann. To these criticisms Hulett and Quinn ' have replied,
showing that certain supposed errors in their work were avoided, and that the criticisms were,
in their opinion, unsound. At present the questions at issue between the two groups of chemists
seem, to be not wholly settled, and for that reason I prefer to adopt, at. least provisionally, the

mean value
 Cd= 112.3764. 0005;

rejecting the very.low determinations which are surely wrong. This value, moreover, relates
and approximates to the excellent determinations of Morse and Arbuckle, which should not

be wholly disregarded.
MERCURY.

In dealing with the atomic weight of mercury we may reject the early determinations
by Sefstrém ? and a large part of the work done by Turner. The latter chemist, in addition
to the data which will be cited below, gives figures to represent the percentage composition of
both the chlorides of mercury; but these results are neither trustworthy nor in proper shape
to be used.

First in order we may consider the percentage dcmaposttion of mercuric oxide, as established
by Turner and by Erdmann and Marchand. In both investigations the oxide was decomposed
by heat, and the mercury was accurately weighed. Gold leat served to collect the last traces of
mercurial vapor.

Turner gives four determinations. Two represent setts aed by the ignition of the nitrate,
and two are from commercial oxide. In the first two the oxide still contained traces of nitrate,
but hardly in weighable proportions. A comparison of the figures from this source with the
others is sufficiently conclusive on this point. The third column represents the percentage of
mercury in HgO:

144. 805 grains Hg=11. 54 grains O. 92. 619 per cent.
125.980 “ 10.08 « 92.592.
173.561. «“* 43,82 “ 92.625“

114. 294 9.101 ‘ : 92.620 “

 

Mean, 92. 614-4. 0050
Hence, Hg = 200.626.

In the experiments of Erdmann and Marchand ‘ every precaution was taken to ensure
accuracy. Their weighings, reduced to a vacuum, give the subjoined percentages:

HO. . Ag. Per cent.
82. 0079 75. 9347 92. 594
51. 0320 47. 2538 92. 597
84. 4996 78. 2501 92. 604
44, 6283 41, 3285 92. 606

118. 4066 109. 6408 92. 597

: Mean, 92. 5996+. 0015
Hence, Hg = 200.205.

Other determinations of the same ratio, being different in character, will be considered
later.

With a view to establishing the atomic weight of sulphur, Erdmann and Marchand also
made a series of analyses of mercuric sulphide. These data are now best available for discus-

 

1 Journ. Amer. Chem. Soc., 37, 1997. 1915. : “a Phil, Trans., 1833, 531-535.
3 Sefstrém. Berz. Lehrb., 5th ed. , 8, 1215, ‘Work donein 1812. : 4 Journ, prakt. Chem., 31,395, 1844,
No. 3.] ATOMIC WEIGHTS—CLARKE. 201

sion under.mercury. The sulphide was mlxed with pure copper and ignited, mercury distilling
over and copper sulphide remaining behind. Gold leaf was used to retain traces of mercurial
vapor, and the weighings were reduced to a vacuum:

gS. Hg. Per cent Hg.
34. 3568 29. 6207 86, 215
24, 8278 21. 40295 86. 206
37. 2177 32, 08416 86, 207
80. 7641 69. 6372 86. 223

 

Mean, 86. 2127+. 0027
Hence, Hg = 200.504.

For the percentage of mercury in mercuric chloride we have data by Turner, Millon, Svan-
berg, Hardin and Easley. Turner,’ in addition to some precipitations of mercuric chloride by
silver nitrate, gives two experiments in which the compound was decomposed by pure stannous
chloride, and the mercury thus set free was collected and weighed. The results were as follows:

44, 782 grains Hg=15. 90 grains Cl. 73. 798 per cent.
73. 09 se 25. 97 73. 784 sf

Mean, 73. 791+. 005

 

Hence, Hg=199.677.

Millon ? purified mercuric chloride by solution in ether and sublimation, and then subjected
it to distillation with lime. The mercury was collected as in Erdmann and Marchand’s experi-
ments. Percentages of metal as follows:

73. 87 73. 83

73. 81 73. 87
Mean, 73. 845+. 010

Hence, Hg = 200.236.
Svanberg,? following the general method of Erdmann and Marchand, made three distilla-
tions of mercuric chloride with lime, with the following results:

Hg Ci, Hg. Per cent.
12. 048 8. 889 73. 780
12. 529 9, 2456 73, 794
12. 6491 9. 3363 73, 810

 

Mean, 73. 795+ .006
Hence, Hg= 199.718.

More modern determinations of the atomic weight of mercury are due to Hardin,‘ whose
methods were entirely electrolytic. First, pure mercuric oxide was dissolved in dilute, aqueous
potassium cyanide, and electrolyzed in a platinum dish. Six determinations are published,
out of a large number, but without reduction of the weights to a vacuum. The data, with a
percentage column added, are as follows:

Ain, Weight Hg0. at Weight Hg. Per cent Hg. «|

a8

0. 26223 0.°24281 92, 594
. 23830 . 22065 92, 593
. 23200 . 21482 92, 595
. 14148 . 13100 92. 593
. 29799 . 27592 92, 594
. 19631 . 18177 92. 593

 

Mean, 92. 594+ .0003
Hence, Hg = 200.041.
Various sources of error were detected in these experiments, and the series is therefore
rejected by Hardin.
Hardin also studied mercuric chloride, bromide and cyanide, and the direct ratio between
mercury and silver, with reduction of weights to a vacuum. Electrolysis was conducted in a
platinum dish, as usual. With the chloride and bromide, the solutions ‘were mixed with dilute

 

1 Phil. Trans., 1833, 531-535. 3 Journ. Prkt. Chem. 45,472. 1848.
2Ann. Chim. Phys. (3), 18, 345. 1846. ‘Journ, Amer, Chem. Soc.,18, 1003, 1896.
202 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

potassium cyanide. The data for the chloride are as follows, the percentage column being

added by myself. } : ms
Weight HgCid,. Weight Hg. Per cent Hg.

0. 45932 0. 33912 73. 831
. A735 . 40415 73. 888

. 56002 . 41348 73. 833

. 63586 . 46941 73, 823

. 64365 . 47521 73. 831

. 73281 . 54101 73. 827

. 86467 . 63840 78, 832

1. 06776 . 78825 73. 823
1. 07945 4 79685 73, 820
1.51402 1. 11780 73. 830

Mean, 73. 829+.0012 ©
Hence, Hg= 200.070.

For the bromide Hardin’s data are—

Weight HgBr,. Weight Hg. Per cent Hg.
0. 70002 0. 38892 55, 558
- 56430 - 31350 55. 555
- 97142 - 31750 55. 563
» 77285 - 42932 55. 550
- 80930 - 44955 55, 548
. 85342 . 47416 55. 560
1, 11076 - 61708 55. 555
1, 17270 - 65145 55. 551
1. 26186 - 70107 55. 559
1. 40142 . 77870 55. 565

Mean, 55. 556+ .0012
Hence, Hg= 199.811. .

And for the cyanide— oe,
Weight HgC,N,. Weight Hg. Per cent Hg.

0. 55776 0, 44252 79. 337
. 63290 . 50215 73. 341
. 70652 . 56053 79. 337
. 80241 . 63663 79. 340
. 65706 . 52180 79. 338
. 81678 . 64805 79. 342

1. 07628 . 85392 79, 40

1. 22615 | «97282 79. 339

1. 66225 -1,31880 79,338

2. 11170 “1 67541 79. 339

Mean, 79. 339-+.0004
Hence, Hg= 199.763.

In the last series cited no potassium cyanide was used, but the solution of mercuric cyanide,
with the addition of one drop of sulphuric acid, was electrolyzed directly.

The direct ratio between silver and mercury was determined by throwing down the two
metals, simultaneously, in the same electric current. Both metals were taken in double
cyanide solution. With Hardin’s equivalent weights I give a third column, showing the quan-
tity of mercury corresponding to 100 parts of.silver. Many experiments were rejected, and only
the following seven are published by the author:

Weight Hg. Weight Ag. Ratio.

0. 06126 ' 0. 06610 92. 678

- 06190 . 06680 92. 665
07814 . 08432 92, 671
. 10361 . 11181 ; 92. 666

. 15201 . 16402 92. 678

. 26806 . 28940 92. 626

« 82808 + 89388 92. 639

: = Mean, 92. 660+ .0051
Hence, Hg = 199.923.
No. 3.] ATOMIC WEIGHTS--CLARKE. 2038

The determinations by Easley are quite unlike those made by his predecessors. First,
mercuric chloride in solution was reduced to metal by means of hydrogen dioxide, and was
precipitated partly as a globule and partly in finely divided form. The globule was washed
with water and acetone and weighed. The finely divided mercury was again dissolved, and
with a little mercury remaining in solution, was deposited electrolytically upon a gold cathode.
Its weight was then added to that of the globule. The following results, with vacuum weights,
were obtained:

HgCld,. Hg. Per cent Hg.
23. 43239 17. 30826 73. 865
12. 59751 9. 30608 73. 873
10. 94042 8. 08154 73. 869
11. 73734 8. 67044 73. 871

5 Mean, 73. 8695+ .0012
Hence, Hg = 200.490.
In the filtrate from the mercury the chlorine was precipitated as silver chloride and so
weighed. The results were as follows, with vacuum weights:

HgCl. a AgOl Ratio.
10. 50276 1.08744 - - 94. 7257
9. 03634 9.54027... 94. 7179

| 23. 43239 24, 73606, 94, 7297

10. 94042 11. 55158 94. 7093

11. 11409 11. 73470 94. 7113

16. 63910 17. 56808 94, 7121

Mean, 94. 7177 +.0023
Hence, Hg = 200.617.
In a second memoir Easley? gives two series of electrolytic analyses of mercuric chloride,
with elaborate precautions to insure accuracy, including, of course, the reduction of his
weights to a vacuum. The data are as follows:

Prelaminary series.

AgCl,. Hg. Per cent Hg.
10. 05743 7. 43123 73. 888

8. 41289 6. 21687 73. 897
10. 99056 8. 11897 73. 873
10. 28282 7. 59681 73. 879
19. 57120 14. 46032 73. 886

Mean, 73. 885-4. 0027
Hence, Hg=200. 651.
Final series.

HgCi,. Hg. Per cent Hg.

8. 14695 6. 01909” ' 73, 882
11. 03181 8. 15592 738: 884
13. 48192 9. 96129 73. 886
11. 08026 8. 18610 73. 880
11. 31231 8. 35819 73. 886
21, 44026 15. 84060 73. 882

Mean, 73. 883+. 0008
Hence, Hg = 200. 630.

Combining these series with the earlier determinations of the ratio we have:

PUR yc sesh Bess Sudo eeaees see fins dewnonnotineniestahac 73. 791° +. 0050
MillOnis2.2.2.<cuuedeccPoasasewsmnt oie s saaahstar slain yeep Se esate 73: 845 +.0100
SVan berg! cesctccer vais seriou ens ogajeweei dak sos Pees 73.795 +. 0060
Hardin. cossacainews ss st eeeeeeececs 45 vasieteddeker since 73. 839 +. 0012
Easley; first. ..c22-2222.222seedce00 02 9 Dialga aetiesS o'e.Blde 73. 8695+. 0012
Easley, second... ...-.-. 2-02. .e eee e cence eee eee eee ee 73. 885 +. 0027
Easley, third «2. ccac0c.sc0 snoesscnuns eciiceoneseai esses 73. 883 +. 0008

General mean......-.-.-.--22-22.------- wuitaees os 73.870 -+. 0007

 

1 Journ, Amer. Chem, Soc., 31, 1207. 1909. 2 Journ. Amer. Chem. Soc., 32, 1117. 1910.
204 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

In a still later investigation Easley and Brann! made analyses of mercuric bromide.
Their data, with vacuum weights, give the ratio 2AgBr : HgBr, :: 100 : z, as follows:

HoBr,. ee AgBr. ¢ Ratio.

8. 93958 ‘9. 31480 95. 972
14. 36691 14. 96940 . 95.975
10. 13638 10. 56230 95. 972

9. 94452 10. 36218 95. 969
12. 60142 13, 13051 95. 971
12. 16157 12. 67122 95. 978
11. 19762 11. 66809 95, 968
16. 16607 16. 84235 95. 985
11. 28487 ‘11. 75970 95. 962
17. 25074 17. 97597 95. 966
14, 20924 14. 80685 95. 964

Mean, 95. 971+. 0013

ea Hg = 200. 371.

A new method was adopted — Taylor sau Hulett 2 for debasaitnanie the composition of
mercuric oxide. The powdered oxide was heated in a sealed tube with finely divided iron,
and the mercury so liberated formed a perfectly clear globule which was easily weighed. Their
data, with vacuum weights, and the usual percentage follows:

Hg0. Hg. Per cent Hg.
2. 98464 2. 76390 92. 604
3. 61841 3. 35066 92. 601
3. 82184 3. 53923 92. 606
3. 40356 3. 15185 92. 605
3. 80726 3. 52567 92. 606
3. 56858 : 4. 23065 92. 603
5. 42945 5. 02819 92. 609
2. 38313 2. 20681 92. 602
3. 76766 3. 48933 92. 612

 

Mean, 92. 6053+. 0008
Hence, Hg =200.371.

Combining this series with its predecessors we have:

TUM OP sic eiseelucieiarns cen tsudusiumeeas Aekebes eae kaa 92.614 +. 0050
Erdmann and Marchand. .........-...----0..---0-0---- 92. 5996 +. 0015
Hardin 2s anadeccdiauyaeucsaneekmenectin tyveee ease ss 92.594 +. 0003
Taylor. and Hulettecwcociseeccsseessetevessusasseecoees 92. 6053 +. 0008
General mean............. 200. cece eee ee ee eee eee 92. 5952 +. 00028

The determination of the atomic weight of mercury by Baker and Watson * appeared in
1913. They effected the synthesis of mercuric bromide by the direct union of mercury and
bromine. With vacuum weights and the usual percentage column the data are as follows:

Hg. Br. Per cent Hg in HgBry.
4. 88190 3. 89084 50. 649
6. 74796 5. 37850 55. 648
4. 79430 3. 82210 55. 642
4. 52480 3. 60568 55. 652
4. 78892 3. 81642 55. 651
5. 52880 4. 40526 55. 655
4. 40148 3. 50738 55. 652
4. 52338 3. 60456 55. 652
5. 44704 4. 33980 55. 657

 

Mean, 55. 651-4. 0010

1 Journ. Amer. Chem. Soc., 34, 187, 1912, 2 Journ, Phys, Chem., 17, 755. 1913, 3 Journ. Chem. Soc., 105, 2530. 1913.

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 205

For the same ratio Hardin’s analyses gave 55.556 +.0012 per cent. The two series com-
bined give 55.612+.0008. From Baker and Watson’s series alone, Hg= 200.581.
The following ratios are now available for computing the atomic weight of mercury:

(1) HgO : Hg :: 100 : 92.5952 4.00028.
(2) HgS : Hg :: 100 : 86.2127 4.0027.

(3) HgCl, : Hg :: 100 : 73.870 +.0007.

(4) HgBr, : Hg :: 100 : 55.612 +.0008.
(5) HgC,N, : Hg :: 100 : 79.339 4.0004.
(6) 2Ag : Hg :: 100 : 92.660 +.0051.

(7) 2AgCl : HgCl, :: 100 : 94.7177 +.0023.
(8) 2AgBr : HgBr, :: 100 : 95.971 + .0013.

The antecedent atomic weights are—

Ag=107.880+ .00016 8=32.065 +.00012
C1=35.4605 + .00011 N=14.0081+ .00008
Br=79.9228 + .00015 C=12.0025+ .00019
Hence, 5 ‘
From ratio—

Din wa aa secon ceciatec aes eee eet ee ee! Hg=199.763 +.0048

Baines enacts Sinenisieie 2 -¢.da8re See Ren Dak cain nena 199.923+ .0110

De ES sd Soe ee een te eaten eee 200.076 + .0076

sea ie cicada ovata arena esac a tree ce tape atela rah etape 200.264 + .0046

OD sia niet ceslcishne g dseleie/ ev ssp paeMMaaay oe Se Sree 200.495 + .0057

Bivile sos HasildaseGovavs He yeeeherseew se eees decades 200.504 + .0399

Uses sais 3a see eee Raa be Sais Leweisee oe B84 oak 200.617 + .0067

Saeseseecs sedan atiehss bese amaecuelaye sees as 200.627 + .0049

General mean...........2-.-0-0-- 222 eee eee Hg=200.277 +.0022

This combination is by no means satisfactory. Mathematically, Hardin’s determinations
outweigh all the others, a superiority to which they are not entitled.. If we reject the two
lowest values in the combination, the others give a general value of Hg= 200.432; a value which
seems to be much nearer the truth. If we reject all determinations of the atomic weight of mer-
cury made before 1900, and use only those by Easley, Easley and Brann, Taylor and Hulett,
and Baker and Watson, we may be on safer ground. These determinations are as follows:

Easley, analyses of HgCl,.........-.-.--2.----2--+2--- Hg=200. 589+. 0053
Easley, ratio 2AgCl : HgCl,........2-2-2.-2.2.222- 2 ee 200.617 +.0067
Easley and Brann, ratio 2AgBr : HgBry..........------ 200. 627+. 0049
Taylor and Huleit, analyses of HgO..........-.-.-.--. 200.371 4.0217
Baker and Watson, syntheses of HgBr,...-.......-.... 200.581 = .0058

General meats iii. eiwlesticsisg's aes cwevametss Hg=200.600+ .0028

This value seems to be the most probable one for the atomic weight of mercury, and it is
exactly that which appears in the latest reports of the International Committee on Atomic
Weights. The value thus found, however, cannot be regarded as final. The analyses of mer-
curic oxide by Taylor and Hulett give a much lower value, and their tnode of operation appears
to be sound. The discrepancy, which is like that noticed under cadmium, is unexplained.

BORON.

In the first edition of this book the data relative to boron were few and unimportant. ‘There
was a little work on record by Berzelius and by Laurent, and this was eked out by a discussion
of Deville’s analyses of boron chloride and bromide. As the latter were not intended for atomic
weight determinations they will be omitted from the present recalculation, witich includes a
number of later researches.
206 MEMOIRS NATIONAL ACADEMY OF SCIENCES.

(Vou. XVI,

Berzelius * based his determination upon three concordant estimations of the. percentage
of water in borax. Laurent ? made use of two similar estimations, and all five may be properly
put in one series, thus:

47.10
«10 pra
47.10

47.15

47.20
Mean, 47.13+.013

\Laurent

Hence, B= 11.023.

In 1869 Dobrovolsky * published a dissertation, in Russian, on the atomic weight of boron.
The original I have not seen, and I am therefore compelled to use the data as cited by Brauner.‘
According to Dobrovolsky, borax is completely dehydrated by ignition when small quantities
of it are taken. With large quantities, some water is retained. Two series of experiments are

given to illustrate this assertion:
First series.

Borax. Water. Per cent water.

0.138 0.0651 47.174
-283 .1838 47.279
.312 1472 47.179

/ Mean, 47.211 + .023
‘ Hence, B= 10.859. :
Second series.

Borax. Water. Per cent water.
2.701 1.268 46.946
1.793 843 47.016
3.004 1.402 46.671

Mean, 46.878+-.072
Hence, B= 11.529.

In 1892 the posthumous notes of Hoskyns-Abrahall were edited and published by Ewan
and Hartog.® This chemist especially studied the ratio between boron bromide and silver,
and also redetermined the percentage of water in erystallized borax. The latter work, which
was purely preliminary, although carried out with great care, gave the following results, reduced

to a vacuum: '
Na,B,0,.10H,0. Na,B,O,.

Per cent H,0.
7.00667 3.69587 47.2069
12.95936 6.82560 47.3308
4.65812 2.45248 47.3504
4.47208 3.93956 47.2763
4.94504 2.60759 47.2686

Hence, B = 10.703.

Mean, 47.2866-+.0171

Two sets of determinations were made with the bromide, which was prepared from boron

and bromine directly, freed from excess of the latter by standing over mercury, and finally col-
lected, after distillation, in small, weighed, glass bulbs.
after all the usual precautions.
portional to 100 parts of silver stated as the ratio:

It was titrated with a solution of silver
The first series of experiments was as follows, with BBr, pro-

BBr,. Ag. Ratio.
1.31203 1.69406 77.449
4.39944 5.67829 77.478
5.04022 6.50820 77.444
6.51597 8.38919 77.433
7.75343 10.01235 77.439

Mean, 77.4494: .0053

 

1 Poggend. Annal.,8,1. 1826.

2 Journ. prakt. Chem., 47, 415. 1849,

3 Doctoral Dissertation, Kiev, 1869.

4In Abegg’s Handbuch der anorganischen Chemie, Bd.3, Abth.1, p. 6.

5 Journ. Chem. Soc., 61, 650. 1892.
No. 3.] ATOMIC WEIGHTS—CLARKE. 207

This series of data is regarded by the editors as preliminary, and not entitled to much con-
sideration. The second series, which follows, was the final one; both represent vacuum standards:

BBr;. Ag. Ratio.
4.467835 5.771268 77.415
8.423151 10.880648 77,414
1.655111 2.137593 77.429
8.032352 10.374201 77.426
4.092743 5.285949 77.427
2.389993 3.086842 77.425
7.721944 9.974054 77.420

Mean, 77.422+ .0018
First series, 77.449 + .0053

Hence, B=10.800. General mean, 77.425+.0017

Ramsay and Aston,’ in their paper upon the atomic weight of boron, suggest that Abra-
hall’s bromide may have contained hydrobromic acid, which would fully account for the low
result obtained. They themselves adopt two distinct methods, the first one being the time-
honored determination of water in crystallized borax. The latter was prepared from pure
boric acid and pure sodium hydroxide. Results as follows, reduced to a vacuum :

Na,B,0,.10H,0. . Na,B,0,. Per cent. H,0.
10, 3581602 5. 4784357 47.1099
5, 3440080 2. 8246677 47.1433
4. 9962580 2. 6378934 47, 2026
5. 7000256 3. 0101127 47.1912
5. 3132725 2. 8065646 47.1882
4. 9971924 2. 6392016 47.1865
5.2366921 2.7674672 47.1524

Hence, B=10.946. Mean, 47.1677+. 0086

The second method adopted by Ramsay and Aston was to distill anhydrous borax with
hydrochloric acid and methyl alcohol, both scrupulously pure, thereby converting it into sodium
chloride. The operation was conducted in a glass flask, and in the first series of determinations
ordinary soft glass was used. This, however, was somewhat attacked, so that the sodium
chloride contained silica; hence oxygen in the material of the flask had been replaced by chlorine,
thereby increasing its weight and lowering the apparent atomic weight of boron. It a second
series flasks of hard combustion tubing were taken, and the error, though not absolutely avoided,
was reduced to a very small amount. Both series are subjoined, together with the percentage
of chloride formed; but the weights, given by the authors to seven decimal places, are only
quoted to the nearest tenth milligramme. They are reduced to a vacuum:

First Series.

Na,B,0;. ' NaC. Per cent NaCl.
' 4.7684 2.7598 57.877
5.2740 3.0578 57.978
3.2344 . 1.8727 57.899
4, 0862 _ 2.3713 58. 032
3.4970 2.0266 57. 953

. Mean, 57.948+.0187
Second Series.

Na,B,0;. NaCl. Per cent NaCl.
5.3118 © 3.0761 57.911
4.7806 2.7700 57.943
4, 9907 2.8930 57. 968
4.7231 2.7360 57. 928
3.3138 1, 9187 ' 57.900

Mean, 57.930+.0081
First series, 57. 948+. 0187

Hence, B= 10.955. General mean of both, 57. 933+.0074

 

1 Journ. Chem. Soc., 63, 211. 1893.
161566°—20——14
208 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

As a check upon the last series of results, the sodium chloride was dissolved in water, and
precipitated with silver nitrate... The silver chloride was collected and weighed in a Gooch cru-
cible, and its weight gives a new ratio with anhydrous borax. The cross ratio between the
two chlorides, silver and sodium, has already been used in the discussion upon sodium. The
new ratio I give in terms of Na,B,O, equivalent to 100 parts of AgCl.

Na,B,0,. Ag Cl. Ratio.
5.3118 7.5259 70. 580
4.7806 6.7794 70.517
4. 9907 7.0801 70. 489
4.7231 6.6960 70. 536
3.3138 4.6931 70.610

 

Mean, 70. 546+.0146
Hence, B= 11.059.

Rimbach * based his determination of the atomic weight of boron upon the fact that boric
acid is neutral to methyl orange, and that therefore it is possible to titrate a solution of borax
directly with hydrochloric acid. His borax was prepared from carefully purified boric acid
and sodium carbonate, and his hydrochloric acid was standardized by a series of precipitations
and weighings as silver chloride. It contained 1.84983 per cent of actual HCl. The borax,
dissolved in water, was titrated by means of a weight-burette. I give the weights found in the
first and second columns of the following table, and in the third column, calculated by my-

self, the HCl proportional to 100 parts of crystallized borax. Rimbach himself computes the
percentage of Na,O-and thence the atomic weight of boron, but the ratio Na,B,O,.10H,0 :
2HCl is the ratio actually determined.

Na,B,0,.10H,0. HCI Solution. Ratio.

10. 00214 103.1951 19. 0853
15. 32772 158. 1503 19. 0864
15. 08870 155.7271 19. 0917.
10. 12980 104. 5448 19. 0922

5. 25732 54, 2571 19. 0908
15. 04824 155. 2307 19. 0883
15. 04761 155. 2959 19. 0908
10. 43409 107. 6602 19. 0868

5.04713 52. 0897 19. 0915

 

Mean, 19.0893 +. 0006
Hence, B=11.006.

Obviously, this error should be increased by the probable errors involved in standardizing
the acid, but they are too small to be worth considering. |

The work of Armitage on the atomic weight of boron was published ie in abstract.’
The data, however, were fortunately given to Brauner,’ who stated them in partially available
form. First, six determinations were made of the proportion of water in borax giving in mean,
47.1475 per cent with a minimum of 47.1224 and a maximum of 47.1637. Hence, B=10.986.
If these figures alone are considered, the probable error of the mean is +.0091. Secondly,
anhydrous borax was titrated with standard sulphuric acid, with the subjoined results:

Na,B,0;. SO,. Ratio.
1. 94033 - 924615 209. 053
1. 56303 - 748413 210. 251

 

Mean, 210. 052 +. 133

Hence, B=10.933. The determination is evidently of small significance.

 

1 Ber. Deutsch. chem. Ges., 26, 164, 1893.
2 Proc, Chem. Soc., 14, 22. 1898 The commnication was followed by several adversecritisms. Seealso Leonard, Chem. New, 77, 104.
8 Op. cit.
No. 3.) ATOMIC WEIGHTS—CLARKE.
Combining the data relative to the percentage of water in borax, we have—

Berzelius with Laurent. .............2-202e-e---eeeeeeee 47.130 +.0130
Dobrovols keys. Weta fecct's's "stay Watiltalere lalate aeraedacteink le siecle Seteesicvs 47.211 +.0230
Dobrovolaky, Qi. 2.5 ss ss0sseewes sos cexeeeedeses ees duuse 46.878 +.0720
Hoskyne-A brahall esis cesses cosa eeeecieee oeeecelneeee 47. 2866+.0171
Ramsay and Aston.....-........---- Nees Hee ean 47.1677 +. 0086
ATMINAG Oi). sites cree creas tae enls sii aageeage hea dees 47.14754+.0091

General Tia secs eae tees ont niseinieidn eves Doemnine 47.1654+. 0052

209

This mean is very close to that of Ramsay and Aston. Dobrovolsky’s figures count for

practically nothing.

Gautier’s determinations' were based upon analyses of four boron compounds. First,
boron sulphide was decomposed by caustic soda; the solution was then oxidized with bromine
water, and the sulphur was precipitated and weighed as barium sulphate. I give the ratio
3BaSO, : BS, : : 100 : x in the third column below. The weights are all reduced to a vacuum.

B,S,. BaSO,. Ratio.
0. 2754 1. 6312 16. 883
. 3380 2. 0004 16. 897
. 3088 1. 8300 16, 874
. 2637 1.5614 16. 888

Mean, 16. 8855-4. 0033
Hence, B= 11.024.

Secondly, boron carbide was heated in chlorine to expel the boron as BCl,.

The residual

carbon was then burned in oxygen, and the dioxide so produced was weighed. I subjoin the

weights, and also the ratio CO, : B,C :: 100: 2:

B,C. co,. Ratio.
0. 2686 0. 1515 177. 293
. 3268 . 1844 177. 224

: * Mean, 177. 258+. 024
Hence, B= 10.997.

Third, boron tribromide was decomposed by water, and its bromine content was then
determined as silver bromide. The following data relate to two samples of the boron compound,

with five analyses of the first lot and four of the second:

BBrs. AgBr. Ratio.
3. 1130 6. 994 44.510
3. 3334 7. 490 44, 505
3. 7456 8. 414 44, 516
3. 2780 7. 364 44.514
4. 2074 9. 452 44. 513
3. 3956 7. 628 44. 515
4. 0295 9. 052 44, 514
3. 7886 8. 512 44. 509

3.1711 7.124 44, 513
: Mean, 44. 512+. 0009
Hence, B=11.016.

Finally, the analysis of boron chloride was effected in the same way with the following

results:
BCh. AgCl. Ratio.
2. 6412 9. 682 27. 279
2. 7920 10. 234 s 27. 282
2. 4634 9. 026 27. 292
2. 4489 12. 640 27, 285.
2. 2015 8. 070 27. 280
2. 6957 9. 878 27. 289

Mean, 27. 2845+. 0014

 

Hence, B = 10.947.

1 Ann. Chim. Phys, (7), 18, 352. 1899.
210 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

The work of Gautier on boron sulphide and carbide may safely be disregarded in the final
discussion. The compounds were unsuited to accurate atomic weight determinations, and the
carbide was confessedly impure. The sulphate series of Armitage is also of no present value.

The determinations of the atomic weight of boron by Smith and Van Haagen’ are entirely
unlike those of their predecessors. Anhydrous borax, Na,B,O,, was converted into the sulphate,
nitrate, carbonate, chloride, and fluoride, and from the ratios so determined the required atomic
weight was computed. A fundamental problem was to prepare an absolutely anhydrous tetra-
borate, and that was a matter of great difficulty. Failure to accomplish this result vitiated
many of the previous determinations. _ .

The conversion of the anhydrous borax into the other sodium salts involved several dis-
tinct methods. The sulphate and nitrate were formed by direct treatment of the borax with
sulphuric or nitric acid, and the boric acid that was liberated was then expelled as methyl borate
by distillation with methyl alcohol, a well-known process. This process was followed in all the
determinations. In order to convert the borax into sodium carbonate it was dissolved in a little
water and then evaporated to dryness in a current of carbon dioxide. The residue wag then
repeatedly evaporated with methyl alcohol in a steady stream of carbon dioxide. The entire
cycle of operations was repeated twice, and the sodium carbonate, entirely free from boric acid,
was finally fused in an atmosphere of carbon dioxide.

The transformation of the sodium tetraborate into chloride or fluoride was effected by a
preliminary conversion into formate by means of formic acid, and a subsequent treatment
with hydrochloric or hydrofluoric acid. With one exception the chloride or fluoride thus
obtained was, after weighing, treated with sulphuric acid and converted into sulphate, giving
additional, although somewhat indirect, values for the borate-sulphate ratio. For the elabo-
rate details of all these operations and the great care taken to eliminate sources of error, the
original memoir should. be consulted. The data obtained, rearranged in order to fit the system
adopted in this work, are as follows. The weights were all reduced to a vacuum.

Ratio Na,SO,: Na,B,O, :: 100 : x.

Na.S0O,. Na,B,0,. Ratio.
0. 63313 0. 89853 141. 919
- 49113 . 69695 141. 907
1. 12315 1. 59374 141. 899
1. 12889 1. 60201 141. 910
1. 86597 2. 64768 141. 893

 

Mean, 141. 906+. 0030

Hence, B= 10.904.
Ratio 2NaF : Na,B,O, :: 100 : x.

NaF. Na,B,0,. Ratio.
0. 29042 0. 69695 239. 988

. 83008 1. 99197 239. 988

. 66757 1. 60201 239. 976
1. 10329 2. 64768 239, 980

 

Mean, 239. 983-4. 0020
Hence, B=10.903, when F=19.0045; a value derived from the ratio between NaF and
Na,SO,, as deduced from Smith and Van Haagen’s ratios.
Ratio Na,CO, : Na,B,O, :: 100 : «.
Na,CO3. Na,B,0;. Ratio.
1. 03946 1. 97702 190. 197
Hence, B= 10.905.
Ratio 2NaNO, : Na,B,O, :: 100 : x.

NaN0,. Na,B,0,. Ratio.
1. 57250 1. 86458 118. 574

 

1 Publ. No. 267, Carnegie Inst. Washington. 1918.
No. 3.] ATOMIC WE1GHTS—CLARKE. 211

Hence, B = 10.899.
Ratio 2NaCl: Na,B,O,: : 100 : 2.

NaCl. Na,B,0;. Ratio.
0. 52112 0. 89853 172, 423

‘Hence, B =10.901.

For this ratio inverted, that is for the percentage of NaCl yielded by Na,B,O,, Ramsay
and Aston found the value 57.933 + .0074, While Smith and Van Haagen’s value becomes 57.997.
Assigning this equal weight with the other the general mean of the two is 57.965+.0052. The
Ramsay and Aston value might be rejected entirely, for their boron was probably not abso-
lutely anhydrous; but as this study is in part one of the compensation of errors it may be
allowed to stand provisionally. .

The ratios from which to compute the atomic weight of boron are now as follows:

(1) Na,B,O,.10H,O : 10H,O :: 100 : 47.1654 +.0052.
(2) Na,B,O,.10H,O : 2HCI :: 100 : 19.0893 —.0006.
(3) Na,B,O, : 2NaCl :: 100 : 57.965 + .0052.

(4) 2AgCl : Na,B,O, :: 100 : 70.546 + .0146.

(5) 2NaF : Na,B,O, :: 100 : 239.983 + .0020.

(6) 2NaNO, : Na,B,O, :: 100 : 118.574 + .0070.

(7) Na,CO, : Na,B,O, :: 100 : 190.197 +.0070.

(8) Na,SO, : Na,B,O, :: 100 : 141.906 + .0030.

(9) 3AgCl : BCI, :: 100 : 27.2845 + .0014.
(10) 3Ag : BBr, :: 100 : 77.4254 .0017.
(11) 3AgBr : BBr, :: 100 : 44.512 + .0009.

The probable errors assigned to ratios 6 and 7 are in each case that of a single determina-
tion in Smith and Van Haagen’s sulphate series.
The values used in reducing these ratios are:

Ag=107. 880 +..00016............--2----- Na=23. 0029+. 00012
Cl= 35. 4605+. 00011.................-.- N=14. 0081+. 00008
Br= 79, 9228+. 00015.........-.....--.-- C=12. 0025+. 00019
F= 19. 0045+. 0012.............2.....-. H= 1. 0078+. 00001
Hence,
: From ratio—
VQ srerscceraere ssc crepeis inmates a dees gerne eke ss sis sare B=10. 800+. 0056
Gikc s Seiepsceet srcra Semenersneisis odes a mieeanamicisioueescisiaeene 10. 899+. 0037
hig Sei e ok oce enee es fuS eh dacbed Seek a th Seabees cclariis Bock meee ets 10. 903+. 0015
Shiskduweset cacdannasaenee 2c e cea nee eee eee 10. 904+. 0011
Waite te wee Wee ame ES a Ee 10. 905+. 0018
Se sand Eat ie atala che aude nlslnerrinn a Aleaienite eae 10. 929-4. 0050
Oh. saan ee is Cee aaeieu eed yeas eeabe tse se at ieee 10. 947+. 0061
Wise hisch Cisne Aetere hs ceepeea se anon ed bicepep ernie Gear owes S 10. 951+. 0074
DS oe aieia eS Baa er aucig eines He oe Bae eee DAES 11. 006+. 0031
ATs sescdaennd dss Wr dnt dteceed Baa neebablathe Clas eens 11. 016+. 0051
A HE ede shstta tata a cal an Seah one ha ead aS RE ES loro tReet wre ape 11. 059+. 0105
General mean..............2 22-2 e ee eee eee B=10. 912+. 0007

In this combination the determinations by Smith and Van Haagen outweigh all the others.
At the end of their memoir they give a careful discussion of all the other data, and point out
probable sources of error in the methods previously employed. Their final value, B=10.900,
appears to be very near the truth, and it may reasonably be adopted for general use.

ALUMINUM.

The atomic weight of aluminum has been determined by Berzelius, Mather, Tissier, Dumas,
Isnard, Terreil, Mallet, Baubigny, Thomsen, and Kohn-Abrest. The early calculations of
Davy and of Thomson we may properly disregard.
212 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

Berzelius’ + determination rests upon a single experiment. He ignited 10 grammes of dry
aluminum sulphate, Al,(SO,),, and obtained 2.9934 grammes of ALO, as residue.

Hence Al = 27.30.

In 1835 Mather ? published a single chalets of aluminum chloride, from which he sought
to fix the atomic weight of the metal. 0.646 grams of AICI, gave him 2.056 of AgCl and 0.2975
of Al,O,. These figures give worthless values for Al, and are included here only for the sake
of completeness. From the ratio between AgCl and AICl,, Al=28.731.

Tissier’s > determination, also resting on a single experiment, appeared in 1858. Metallic
aluminum, containing 0.135 per cent of sodium, was dissolved in hydrochloric acid. The solu-
tion was evaporated with nitric acid to expel all chlorine, and the residue was strongly ignited
until only alumina remained. 1.935 grammes of Al gave 3.645 grammes of ALO,. If we cor-
rect for the trace of sodium in the aluminum, we have Al=27.185.

Essentially the same method of determination was adopted by Isnard,* who, although
not next in chronological order, may fittingly be mentioned here. He found that 9 grammes
of aluminum gave 17 grammes of Al,O,. Hence Al=27.

In 1858 Dumas,’ in his celebrated revision of the atomic weights, made seven experiments
with aluminum chloride. The material was prepared in quantity, sublimed over iron filings,
and finally resublimed from metallic aluminum. Each sample used was collected in a small
glass tube, after sublimation from aluminum in a stream of dry hydrogen, and hermetically
inclosed. Having been weighed in the tube, it was dissolved in water, and the quantity of
silver necessary for precipitating the chlorine was determined. Reducing to a common stand-
ard, his weighings give the quantities of AICI, stated in the third column, as proportional to
100 parts of silver:

AICh,. Ag. Ratio.
1. 8786 4,543 41, 352
3. 021 7. 292 41, 459—Bad
2, 399 5. 802 41, 348
1. 922 4, 6525 41, 311
1, 697 4.1015 41, 375
4. 3165 10. 448 41.314
6. 728 16. 265 41, 365

In the second experiment the AICI, contained traces of iron. Rejecting this experiment,
the remaining six give a mean of 41.344+.007. These data give a value for Al approximating
to 27.4, and were for many years regarded as satisfactory. It now seems probable that the
ghloride contained traces of an oxy-compound, which would tend to raise the atomic weight.

In 1879 Terreil ° published a new determination of the atomic weight under consideration,
based upon a direct comparison of the metal with hydrogen. Metallic aluminum, contained
in a tube of hard glass, was heated strongly in a current of dry hydrochloric acid. Hydrogen
was set free, and was collected over a strong solution of caustic potash. 0.410 gramme of alumi-
num thus was found equivalent to 508.2 cubic centimeters, or 0.045671 gramme of hydrogen.
Hence, Al = 27.142.

About a year after Terreil’s determination appeared, the lower value for aluminum was
thoroughly confirmed by J. W. Mallet.” After giving a full résumé of the work done by others,
exclusive of Isnard, the author describes ss own experiments, which may be summarized as
follows:

Four methods of determination were anual: each one simple and direct, and at the
same time independent of the others. First, pure ammonia alum was calcined, and the residue
of aluminum oxide was estimated. Second, aluminum bromide was titrated ‘with a standard
solution of silver. Third, metallic aluminum was attacked by caustic soda, and the hydrogen
evolved was measured. Fourth, hydrogen was set free by aluminum, and weighed as water.
Every weight was carefully verified, the verification being based upon the direct comparison,
by J. E. Hilgard, of a kilogramme weight with the standard kilogramme at Washington. The

 

1 Poggend. Annal., 8, 177. 5 Ann. Chim. Phys. (3), 55,151. Ann. Chem, Pharm., 113, 26.
2 Amer. Journ. Sci., 27, 241. 6 Bull. Soc. Chim., 31, 153.
3 Compt. Rend., 46, 1105. 7 Phil. Trans., 1880, p. 1003,

4Compt. Rend., 66, 508. 1868.
No. 8.] ATOMIC WEIGHTS—CLARKE. 213

specific gravity of each piece was determined, and also of all materials and vessels used in the
weighings. During each weighing both barometer and thermometer were observed, so that
every result represents a real weight in vacuo.

The ammonium alum used in the first series of experiments was specially prepared, and was
absolutely free from ascertainable impurities. The salt was found, however, to lose traces of
water at ordinary temperatures—a circumstance which tended toward a slight elevation of
the apparent atomic weight of aluminum as calculated from the weighings. Two sets of experi-
ments were made with the alum; one upon a sample air-dried for two hours at 21°-25°, the
other upon material dried for twenty-four hours at 19°-26°. These sets, marked A and B,
respectively, differ slightly, B being the less trustworthy of the two, judged from a chetnical
standpoint. Mathematically, it is the better of the two. Calcination was effected with a great
variety of precautions, concerning which the original memoir must be consulted. To Mallet’s
weighings I append the percentages of Al,O, deduced from them:

 

Series A.
Alum. Al,03. Per cent.
8.2144 9258 11.270
14.0378 1.5825 11.273
5.6201 -6337 11,275
11.2227 1.2657 11.278
10.8435 1.2216 11.266
Mean, 11.2724+.0014
Series B.
Alum. Al,03. Per cent.
12.1023 1.3660 11,287
10.4544 1.1796 11.283
6.7962 -7670 11.286
8.5601 9654 11,278
4.8992 -5528 11.283

 

Mean, 11.2834-4.0011

Combined, these series give a general mean of 11.2793-+.0008. Hence, Al= 27.153.

The aluminum bromide used in the second series of experiments was prepared by the direct
action of bromine upon the metal. The product was repeatedly distilled, the earlier portions
of each distillate being rejected, until a constant boiling point of 263.3° at 747 mm. pressure
was noted.. The last distillation was effected in an atmosphere of pure nitrogen, in order to
avoid the possible formation of, oxide or oxybromide of aluminum; and the distillate was
collected in three portions, which proved to be sensibly identical: The individual samples of
bromide were collected in thin glass tubes, which were hermetically sealed after nearly filling.
For the titration pure silver was prepared, and after fusion upon charcoal it was heated in a
Sprengel vacuum in order to eliminate occluded gases. This silver was dissolved in specially
purifled nitric acid, the latter but very slightly in excess. The aluminum bromide, weighed in
the sealed tube, was dissolved in water, precautions being taken to avoid any loss by splashing
or fuming which might result from the violence of the action. To the solution thus obtained
the silver solution was added, the silver being something less than a decigramme in deficiency.
The remaining amount of silver needed to complete the precipitation of the bromine was added
from : a ‘burette, i in the form of a standard solution containing one milligramme of metal to each
cubic centimetre. “The final results were as follows, the figures in the third column representing
the quantities of bromide proportional to 100 parts of silver. Series A is from the first portion
of the last distillate of AIBr,; series B from the second portion, and series C from the third
portion:

i Mi ‘Series A.
AIBr,. Ag. Ratio.
6.0024 aT 2793. 82.458
8.6492 10.4897 82.454

3.1808 3.8573 82.462
214 MEMOIRS NATIONAL ACADEMY. OF SCIENCES. [Vou XVI,

Series. B. 4
ALBr;. Ag: Ratio.
6.9617 8.4429 82.456
11.2041 13.5897 82.445.
8.7621 4.5624 - 82,459
5.2842 , 6.4085 82.456
9.7338 11.8047 82.457
Series C.
_ ALBr,. Ag. _ » Ratio.
- 9.8515 11.8494 82.447
4.4496 : 5.3877 82.458

5.2750 6.3975 82.454
Mean, 82.456. 155 4 001
Hence, Al= 27.089.

The experiments to determine the annount of hydrogen acalved by the gelicn of caustic
soda upon metallic aluminum were conducted with pure metal, specially prepared, and with
caustic soda made from sodium. The soda solution was so strong as to scarcely lose a percept-
ible amount of water by the passage through it of a dry gas at ordinary temperature. As the
details of the experiments are somewhat complex, the original memoir must.be consulted for
them. The following results were obtained, the weight of the hydrogen being calculated from
the volume, reckoned at .089872 gramme per liter:

Wt. Al. Vol. H. Wt. A. At. Wt. (H=1).

0.3697 458.8 0.041234 26.898
3769 467.9 042051 26.889
3620 449.1 040362 26.907
7579 941.5 084614 26.872
7314 907.9 081595 26.891
7541 936.4 084156 26.882

; Meat, 26.890 -+ .0034
Hence, Al= 27.099, when O= 16.

The closing series of experiments was made with larger quantities of aluminum than were
used in the foregoing set. The hydrogen, evolved by the action of the caustic alkali, was dried by
passing it through two drying tubes containing pumice stone and sulphuric acid, and two others
containing. asbestos and phosphorus pentoxide. Thence it passed through a combustion tube
containing, copper oxide heated to redness. A stream of dry nitrogen was employed to sweep
the last traces of hydrogen into the combustion tube, and dry air was afterwards passed through
the entire apparatus to reoxidize the surface of reduced copper, and to prevent the retention
of occluded hydrogen. The water formed by the oxidation of. the hydrogen was collected in
three drying tubes. The results obtained were as follows. The third column gives the amount
of water formed from 10 grammes of aluminum:

Al: #0. Ratio.
2.1704 2.1661 9.9802
' 2.9355 2.9292 . 9.9785
"5.2682 5.2562 9.9867

Mean, 9,9818:+.0017 ;

Heneo, J Al=27.073. i

From the last two series of experiments an independent value for the atomic weight of
oxygen may be calculated. They give O=15.895, when H=1. The closeness of this figure to
some of the best determinations affords a good jucioation of the accuracy of Mallet’s work.

In connection with Mallet’s work it is worth noting that Torrey’ published a series of
measurements of the H : Al ratio, representing determinations made under his direction by
elementary students. These measurements are thirteen in number, and calculated with Reg-
nault’s old value for the weight of hydrogen, range from 26.661 to 27.360, or in mean, 27.049
+.323. Corrected by the latest value for the weight of H, this mean becomes 26.967, when H=1.
This figure, of course, has only confirmatory sienibeanee:

 

1 Amer. Chem. Journ., 10, 74. 1888.
No. 3.] ATOMIC WEIGHTS—CLARKE. 215

By Baubigny ‘ we have only two determinations, based upon the calcination of anhydrous
aluminum sulphate, Al,(SO,);.

Sulphate. Al,O;. Per cent.
8. 6745 1.0965 29. 841
2.539 . 7572 29. 823

Mean, 29. 832+. 0061

Hence, Al= 27.060.

Thomsen’s? value for the atomic weight of aluminum was derived from his earlier work
on the hydrogen-oxygen ratio. In that investigation one part of aluminum was found equiv-
alent to 0.11190+.000015 of hydrogen, and 0.88787+.000018 of oxygen. The aluminum,
however, was impure, and the first step in the new research was to determine its impurities.
These were, in 1 gramme of metal, 0.00819 gramme of silicon and 0.00322 of iron. Correcting for
these, and also for the change of volume in the soda solution following the solution of the metal,
the equivalent values become 0.99897 grm. Al, 0.11195 grm. H, and 0.88824 grm. O. From
the oxygen ratio Al=26.992 + .0011. From the hydrogen, the ratio H: Alt: 1 : 26.765. +.0036
is derived. For the same ratio Mallet found 26. Bop .0034. ‘The two series, combined, give
a general mean of 26.860 + .0025.

The determinations by Kohn-Abrest * are of very slender value. Impure aluminum was
dissolved in hydrochloric acid, the hydrogen evolved was burned over hot copper oxide, and
the water formed was weighed. The weights of metal taken and the percentages of meter
produced are given below:

Weight Al. Per cent H,0.
0. 7909 98. 08
= 7428 - 98.20
-5477 97.86
. 5132 98.10
. 6571 98. 44
. 4993 98. 03
. 5384 , 97.98

Mean, 98. 104.0473

Corrected for the known impurities of the aluminum, this mean becomes 99.151. Hence,
Al= 27.255.

Mallet’s value for this ratio, reduced to the same standard, is 99.818+.0170. Combining,
the general mean is 99.742,+.0160.

Kohn-Abrest also, made two determinations of atomic weight by converting metallic
aluminum into oxide, as follows:

Al. Ato” Per cent.
0.3429 0.6444 53.212

"', 4168 . 7850 58. 095
e Mean, 53. 153-+. 0387

Heiwe: Al=27. 230. This can be combined ith Thomsen’ s figure for the Al: 0. ratio, ‘but
its probable error is so high that it exerts no appreciable influence.

It is clear that the single determinations of Berzelius, Mather, Tissier, Isnard, and Terreil
may now be safely left out of account, for the reason that none of them could affect appreci-
ably the final value for Al. The ratios to consider are as follows:

"9 ) BAg : AIC, :: 100 : 41.344+. 0070.
(2) Percentage Al,O, in ammonium alum, 11.2793 + .0008.
(3) 3Ag: AIBr, :: 100 : 82.455 + .0010.
(4) H: Al:: 1: 26.860 +.0025. °
(5) 2A1:3H,O :: 100 : 99.742 +.0160.
(6) Al,(SO,), : Al,O, :: 100 : 29.832 + .0061.
(7) O: Al:: 16 : 26.9924 .0011.

1Compt. Rend., 97, 1369. 1883.
2 Zeitsch. anorg. Chem., 15, 447, 1897. See also ante, p. 25.
§ Bull. Soc. Chim. (3), 33,121. 1905. Preliminary in Compt. Rend., 139, 669.
216 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI,

The antecedent atomic weights are. Soci
Ag=107. 880+ .00016 S=82. 0650+. 00012 °

 

Cl=35. 4605 +. 00011 N=14. 0081+. 00008
Br=79. 9228+.00015 H=1.0078 +.00001
Hence, 5
From ratio—
Ui srersinvarz wapansts ete etatercvat At Weil acti a ah aver aps carta etads eh Se ama S Al=26. 992+. 0011.
ad rane eis haa ial aceite anata 27.0604. 0055
AC RSS ee tes egdexa euesmebiowoctes .  2e0T0, 0028
Scot eteteee ee . UO Secs etesesed ccc hetistes -27.089+.0014
Ossdeeiwideeawece eee has Sudha, ra eseresersyereateaScererarecaver 27. 093+. 0087.
ae cake en eae steal erences Bee ae 27, 153 +. 0087
Levscisgues o Salearcenressenaieiewsaiela gio oes 2.2 cracieare mie earns 27. 424s. 0227
General' mean...........0.0.02 see cece ce eeee Al=27. 039-4. 0008

The last value, from ratio 1, is worthless, but is of no influence in the general combina-
tion. No one of the other values i is entitled to exclusive confidence. The atomic weight of

aluminum needs reinvestigation.
GALLIUM. we

Gallium has been obtained i in sich small quantities that its atomic weight has riot as yet
been finally determined. The following data were fixed by the discoverer, _Lecoq de Bois-
baudran:!

3.1044 grammes gallium ammonium alum, upon ignition, left .5885 gramme Ga,0,.

Hence, Ga=70.12.

0.4481 gramme gallium, converted into nitrate and ignited, gave .6024 gramme Ga,O,

Hence, Ga= 69.70. .

These values, assigned equal weight, give in mean Ga=69.91, with an uncertainty of
perhaps half a unit.

Recently some preliminary determinations have been made by Richards, Craig, and
Sameshima,? who analyzed gallium chloride by the usual Harvard eon: "The some
given are as follows: te (Eee Reed

GaCl,. NgCl. ip
0. 43947 cy 1.07087, 3 0. 80587
_ Hence there are two values for the atomic weight of gallium, 70.09 and 70.11.. The mean,
70.10, may be adopted provisionally until the more complete lives ltOn 3 is published.

INDIUM.

Reich and Richter, the discoverers of indium, were also the first to determine its atomic
weight.2 They dissolved weighed quantities of the metal in nitric acid, precipitated the solu-
tion with ammonia, ignited the precipipite, and. pocotuained its weight: Two experiments
were made, as follows: el’ ge Dg

In, - 4 ¢ vy eile a ibs,
0.5135 : 0.6243
699 8515

Henoa, i in mean, In=110.61) a its known now: 6 be too low. sf
An unweighed quantity of fresh, moist indium sulphide was also dissolved i in nitric etal

ielding, on precipitation, ;
- pe ee 0. 2108 gramme In,0, and 0. 542 gramme BaSO,.

Hence, with BaSO, = 233.423, In=111.99; also too low.

Soon after the publication of Reich and Richter’s paper. the subject was ‘taken up by
Winkler. He dissolved indium in nitric acid, evaporated to. dryness, ignited the residue, and
weighed the oxide thus obtained.

 

1 Journ. Chem. Soc., 1878, p. 646. 3Journ, prakt. Chem., 92, 484,
2 Proc. Nat. Acad. Sci., 4,387. 1918, 4 Journ. prakt. Cham., 94, 8.

aes
No. 3.] ATOMIC WEIGHTS—CLARKE. 217

L

In. In.O.
0. 5574 0. 6817
- 6661 . 8144

. 5011 . 6126

Hence, in mean, In= 107.76; a result even lower than the values already cited.

In a later paper by Winkler! better results were obtained. ‘Two methods were employed.
First, metallic indium was placed in a solution ‘of pure, neutral, sodio-auric chloride, and the
amount of gold precipitated was weighed. I give the weighings and, in a third column, the
amount of indium proportional to 100 parts of gold:

 

In. Au. Ratio.

0. 4471 0. 8205 if? 57. 782
. 8445, 1, 4596 57. 858
Mean, 57. 820

Hence, if Au= 197.209, In= 114.08.

Winkler also repeated his earlier process, converting indium into oxide by solution in nitric
acid and ignition of the residue. An additional experiment, the third as given below, was made
after the method of Reich and Richter. The third column gives the percentage of In in In,O,:

In. In.03. Per cent.
1.124 1, 3616 82. 550
1.015 1, 2291 82. 581
. 6376: : . 7725 82. 537

These figures were confirmed by a single experiment of Bunsen’s,’ published simultaneously
with the specific heat determinations which showed that the oxide of indium was In,O,, and
not InO, as has been previously supposed:

1.0592 grm. In gave 1.2825 grm. In,O,. 82.589 per cent.

For convenience we may add this figure in with Winkler’s series, which gives a mean per-
centage of In in In,O, of 82.564. Hence, In= 113.646.

Recent investigations have shown that all of the foregoing determinations are untrustworthy,
and that they give values for the atomic weight of indium which are too low. Thiel * carefully
investigated the properties of indium oxide, and found it to be quite unsuited to atomic weight
determinations. Calcined at low temperatures it tends to retain gaseous occlusions; at high
temperatures it is distinctly volatile. Syntheses of the indium halides also gave unsatisfactory
results. Thiel finally made analyses of indium trichloride and tribromide, purified by sublima-
tion, and obtained the following ratios with the corresponding silver salts. First, the ratio
3AgCl : InCl, : 100: x, with weights corrected to a vacuum: a

i Weight InCi,. Weight AgCl. Ratio.
- 5. 0194 9.7526 | 51. 467

4, 7049 ‘9, 1401 51. 475

5. 7067 11, 0862 51. 476

5. 4075 10. 5055 51. 473

 

, ig Mean, 51. 473+. 0015
Hence, In= 114.96.

In the bromide series the weights were not reduced to a vacuum.

Weight InBr,. Weight AgBr. Ratio.
8.9040 14, 1531 62. 912
8, 2140 13. 0512 62. 937
9. 4016 14, 9422 62. 920

Mean, 62. 923-L. 0070
Hence, In= 114.74.
Thiel regards the chloride series as the better of the two, and attaches little importance to

the bromide determinations.

 

1 Journ. prakt. Chem, 102, 282. # Poggend. Annal.,141, 28. # Zeitsch. anorg. Chem, 40, 280. 1904. Preliminary in Vol. 39, 119, and Ber., 37, 175.
218 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

o

Mathers,' like Thiel, derived the atomic of indium from analyses of the two trihalides. His
‘weights and ratios are as follows: :

Weight InCl,... Weight AgCl. Ratio.
2. 1156 4, 11421 51. 422
4, 95920 _, 9.64176 51.435
1, 98175 8. 85125 51.457
5, 54540 10. 77904 . 51, 446
1. 46361 2. 84557 51. 435

4, 08602 7. 94054 51. 458 |
: Mean, 51. 442+. 0038
Hence, In= 114.83.
Combined with Thiel’s series, the general mean for the ratio is 51.469 + .0014.

Weight InBr,. Weight AgBr. Ratio.
2. 73494 4, 34550 62. 937
7. 69880 12, 23341 62, 933
6. 27450 9. 96917 62. 939
5, 36642 8, 52741 62. 931
5. 16112 8, 20128 - 62, 931
4, 98336 7. 92009 62. 921

Mean, 62. 932+. 0011
Hence, In=114.83.
On combination of this series with Thiel’s the general mean becomes 62.931 +.0011.
Neglecting the older work there are now two ratios from which to deduce the atomic weight
of indium: |
(1) 3AgCl : InCl, :: 100 : 51.469+ 0014
(2) 3AgBr : TnBr, :. :: 100 : 62.932 + .0011

Computing with Ag=107.880+ 00016; Cl= 35.4605 + .00011; and Br=79.9228 + .00015
we have—

From ratio De RS OE eat ade er Nia are In=114. 796+. 0062
HYOM PALO Docs -cieencarinowecds hc cnaticcweacedease oe 114, 947+: 0061
General mean........-- Song ue uaa tena renal In=114. 871+. 0044

This mean is, of course, not conclusive. Other indium ratios need to be determined before
the. atomic weight can be more than approximately known. The true value probably lies
between 114.8 and 115.0. For the present the mean value 114.8 may be accepted.

THALLIUM.

“The atomic weight of this interesting metal has been fixed by the researchs of Lamy,
Werther, Hebberling, Crookes, and Lepierre.

Lamy and Hebberling investigated the chloride and sulphate; Werther studied the iodide;
Crookes’ experiments involved the synthesis of the nitrate. Lepierre’s work is still more recent
and is based upon several compounds.

Lamy ” gives the results of one analysis of thallium sulphate and three of thallium chloride.
3.423 grms. of TI,SO, gave 1.578 grm. BaSO,; whence 100 parts of the latter are equivalent
to 216.920 of the former.

Hence, Tl = 205.14.

In the thallium chloride the chlorine was estimated as silver chloride. The following
results were obtained. . In the third column I give the amount of TICI proportional to 100
parts of AgCl:

TICl. AgGl. Ratio.

3.912 2.346 166.752
3.000 1.8015 166.528
3.912 2.335 167.466

Mean, 166.915 + .1905
Hence, Tl = 203.80.

1 Journ. Amer. Chem. Soc., 29, 485. 1907. 2 Zelt. Anal, Chem., 2, 211. 1863.

 
No.'3.] ATOMIC WEIGHTS—CLARKE. — 219

Hebberling’s * work resembles that of Lamy: Reducing his weighings to the standards
adopted above, we have from his sulphate series, as equivalent to 100 parts of BaSO,, the
amounts of T1,SO, given in the third column: .

T1,80,. BaSO,. Ratio.
1.4195 ‘ 0.6534 217.248
1.1924 5507 216.524

8560 3957 216.325

 

Mean, 216.699
Hence, Tl = 204.88.

Including Lamy’s single result as of equal weight, we get a mean of 216.754 -L.1387.
From the chloride series we have these results, with the ratio stated as usual:

TICl. AgCl. oe Ratio.
. 0.2984 0.1791 166.611
0452 3278 166.321

 

Mean, 166.465+ .097
Hence, Tl = 203.16.

Lamy’s mean was 166.915+.1905. Both means combined. give a general mean of
166.555 + .0865. 3

Werther’s? determinations of iodine in thallium iodide were made by two methods. In
the first series TII was decomposed by zinc and potassium hydroxide, and in the filtrate the
iodine was estimated as AgI. One hundred parts of AgI correspond to the amounts of TII
given in the last column:

TU. AgI. Ratio.
0.720 0.51 141.176
2.072 1.472, 140.761
.960 .679 141.384
385 .273 141.026
1.068 .759 140.711 ;

 

Mean, 141.0124 .085

In the second series the thallium iodide was decomposed by ammonia in presence of silver
nitrate, and the resulting AgI was weighed. Expressed according to the foregoing standard,
the results are as follows:

TU. Agl. Ratio.

1.375 0.978 140.593

1.540 1.095 140.639
_ 1.380 981, 140.673

 

Mean, 140.635+.016

General mean of both series, 140.648 -+.016. Hence Tl= 203.32.

In 1873 Crookes,’ the discoverer of thallium, published his final determination of its atomic
weight. His method was to effect the synthesis of thallium nitrate from weighed quantities
of absolutely pure thallium. No precaution necessary to insure purity of materials was neg-
lected; the balances were constructed especially for the research; the weights were accurately
tested and all their errors ascertained; weighings were made partly in air and partly in vacuo,
but all were reduced to absolute standards; and unusually large quantities of thallium were
employed in each experiment. In short, no effort was spared to attain as nearly as possible
absolute precision of results. The details of the investigation are too voluminous, however,
to be cited here; the reader who wishes to become familiar with them must consult the original
memoir.

1 Ann. Chem. Pharm., 134, 11. 1805. 2Journ. prakt. Chem., 92, 198. 1864. 4 Phil. Trans., 1878, p. 277.
220 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI,

The result of 10 experiments by Prof. Crookes may be stated as follows. In a final column
I give the quantity of nitrate producible from 100 parts, of thallium. The weights given are
in grains:

Thallium. TINO;+Glass. Glass Vessel. Ratio.

497.972995 1121.851852 472.557319 130.3875
293.193507 1111.387014 729.082713 130.3930
288.562777 971.214142 594,949719 130.3926
324.963740 1142.569408 718.849078 130.3900
183.790232 1005.779897 766.133831 130.3912
190.842532 997.334615 748.491271 130.3920
195.544324 1022.176679 767.203451 130.3915
201.816345 1013.480135 750.332401 130.3897
295.683523 1153.947672 768.403621 130.3908
299.203036 1159.870052 769.734201 130.3917

 

Mean, 130.3910+ .00034
Hence, Tl = 204.034.

Lepierre’s' determinations were published in 1893, and represented several distinct
methods. First, thallous sulphate was subjected to electrolysis in presence of an excess of
ammonium oxalate, the reduced metal being dried and weighed in an atmosphere of hydrogen.
The corrected weights, etc., are as follows:

TI1,S0,. Tl. Per cent.
1.8935 1.5327 80.945
2.7243 2.2055 80.957
2.8112 2.2759 80.958

 

Mean, 80.953+4.0030
Hence, Tl = 204.150.

Secondly, weighed quantities of crystallized thallic oxide were converted into thallous
sulphate by means of sulphurous acid, and the solution was then subjected to electrolysis, as
in the preceding series:

T1,03. Tl. Per cent.
3.2216 2.8829 89.487
2.5417 2.2742 89.475

 

Mean, 89. ahs 0040
Hence, T= 204.158.

In the third set of experiments a definite amount of thallous sulphate or nitrate was fused
in a polished silver crucible with 10 times its weight of absolutely pure caustic potash. Thallic
oxide was thus formed, which, with various precautions, was washed with water and alcohol,
and finally weighed in the original crucible. One experiment with the nitrate gave—

2.7591 grm. TINO, yields 2.3649 TI,0,. 85.713 per cent.

Hence, T] = 204.025.
_ Two experiments were made with the sulphate, as follows:

T1,S0, T103. Per cent.
3. 1012 2. 8056 90. 468
2. 3478 2. 1239 90. 463

Mean, 90. 465 -+. 0020
Hence, Tl= 204.012.

. Finally, crystallized thallic oxide was reduced by heat in a stream of hydrogen, and the
water so formed was collected and weighed.

T1,05. HO. | Per cent.
2. 7873 0. 3301 11. 843
3, 9871 .4716 11. 828
4, 0213 4761 11. 839

Mean, 11. 837+. 0029
Hence, Tl = 204.300.

 

1 Bull. Soc, Chim. (3), 9, 166.
ATOMIC WEIGHTS—CLARKE. 221

No. 3.)

In a supplementary note’ Lepierre states that his weights were all reduced to a vacuum
standard.

Some work by Wells and Penfield,” incidentally involving a determination of atomic weight,
but primarily intended for another purpose, may also be taken into account. Their question
was as to the constancy of thallium itself. The nitrate was repeatedly crystallized, and the last
crystallization, with the mother liquor representing the opposite end of the series, were both
converted into chloride. In the latter the chlorine was estimated as silver chloride, which was
weighed on a Gooch filter, with the results given below, which are sensibly identical. The TICl
equivalent to 100 parts of AgCl is stated in the last column.

é TICI. AgCl. Ratio.
Grate G tele 3, 9146 2, 3393 167. 341
Mother liquor...........2.26-2+20++ 3, 3415 1. 9968 167. 343

 

Mean, 167. 342
Hence, T= 204.41.

The general mean of Lamy’ s and Hebberling’s determinations of this ratio gave 166.555 2
.0865. If we arbitrarily assign Wells and Penfield’s mean equal weight with that, we get a
new genera] mean of 166.948+.0610.

The ratios to be considered are now as follows:

(1) BaSO, : TISO, :: 100 : 216.754 4.1387.
(2) AgCl : TIC] :: 100 : 166.948 +.0610.
(8) Agi : TH: : 100 : 140.648 + .016.

(4) Tl : TINO, :: 100 : 130.391 4.00034.
(5) TISO, : 2T1 :: 100 : 80.953 +.0030.

(6) TI,0, : 2T1 :: 100 : 89.481 + .0040.

(7) oTINO, : TIO, :: 100 : 85.718.

(8) T1SO, : T1,0, :: 100:: 90.465 + .0020.
(9) T1,0,': 3H,O :: 100 : 11.837 4.0029

The antecedent atomic weights are as follows:

Ag=107. 880 +. 00016
Cl = 35. 4605+. 00011
T +126. 926 +. 00023
Ba=137. 358 +. 0023
Ratio 7 rests upon a single experiment, and the atomic weight derived from it must there-
fore be arbitrarily weighted. To do this its probable error is assumed to be the same as that
given by ratio 8. Taking so much for granted, the nine values for thallium are

N=14. 0081+. 00008
S =32.065 +. 00012
H= 1. 0078+. 00001

From ratio—

De bististela’s shgerresacisiite Se sinceinaeeeade «saree cia T1=203. 298+. 0376
Qutb. wanldivn Semele 4 vince seed saa ee ee ee canoes 203. 844+. 0875
Bios iecneiaeiess a Vedlaneeieetdar yes s's eeretes 204. 012 +. 0696
Wsgas geerigatandes 230s See ceciees Sz eels aa 204. 025 +. 0695
A adie Baa cris ee tuic laevis eis Re ES Ae eee REE SLT 204. 034 +. 0035
Dude ats cie a citaiteg ais catia ad th oeurgt sc 2.t.clsiaetiaeeisie 204. 146 +. 0330
Gass ssicuinemcs iia eee tessedadeatets cccoemesed 204, 158+. 0780
Qusa vs guensetarils o's Sateen ge emeues ods. cere 204. 300 +. 0593
Lo 2isisds cease eG eee bat eat nebeceeetee ys bavesis 204. 945 +. 1619

A glance at the ‘‘probable errors’’ in this series of values will show that Crookes’ ratio,

No. 4, carries overwhelming weight.
for it could not vary much from that.

It is therefore unnecessary to compute the general mean,

The value T1= 204.03 is to be accepted as the best.

 

Heahs

1 Bull. Soc. Chim. (3), 11, 423. 1894.

2 Amer. Journ. Sci. (3), 47,466. 1894.
222 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

SILICON.

Although Berzelius* attempted to ascertain the atomic weight. of silicon, first by converting
pure Si into SiO,, and later from the analysis of BaSiF',, his results were not’ satisfactory. We
need consider only the later determinations. |

Pelouze,? experimenting upon silicon tetrachloride, employed his usual method of titration
with a solution containing a known weight of silver. One hundred parts | of Ag gave. the follow-

ing equivalences of SiCl, :
39. 4325.
39. 4570

Mean, 39. set 0083

 

Hence, Si= 28.370. ‘

Essentially the same method was adopted by Dumas.? Pure SiCl, was weighed in a sealed
glass bulb, then decomposed by water, and titrated. The results for 100 Ag are given in the
third column:

SiCl,. Ag: Ratio.
2, 899 7. 3558 39, 411
1, 242 3. 154 39, 379
3, 221 8, 1875 39. 340

 

Mean, 39. 377+. 014
Hence, Si= 28.077.

Dumas’ and Pelouze’s series combine as follows:

Pelouze........--cececeeececececeeeee abhube Se C.CoOG 39. 4447-4. 0083
Dumas ca coe Saree acer . 39. 37764. 014

39, 4265+. 0071

 

General mean

Schiel,* also studying the chloride of silicon, decomposed it by ammonia. After warming
and long standing it was filtered, and in the filtrate the chlorine was estimated as AgCl. One
hundred parts of AgCl correspond to the quantities of SiCl, given in the last column:

Sich,. AgCi. Ratio.
0. 6738 2. 277 29, 592
1. 3092 4, 418 29. 633

 

: Mean, 29. 6125+. 0138
Hence, Si= 27.945.

Thorpe and Young,® working with silicon bromide, obtained better results. The bromide
was perfectly clear and colorless, and boiled constantly at 153°. It was weighed, decomposed
with water and evaporated to dryness, the crucible containing it being finally ignited. The
crucible was tared by one precisely similar, in which an equal volume of water was also evapor-

ated. Results as follows, with vacuum weights:

SiBr,. S10. Per cent.
9. 63007 1. 67070 17. 349
12, 36099 2, 14318 17. 338
12, 98336 2, 25244 17. 349
9. 02269 1. 56542 17. 350
15, 38426 2. 66518 17, 324
9. 74550 1. 69020 17, 343
6. 19159 1. 07536 17. 368
9. 51204 1. 65065 17. 358
10, 69317 J. 85555 17. 358

Hence, Si=28:381.

Mean, 17. 347+. 0027

The determinations by Becker and Meyer * resemble the foregoing series, except that silicon

tetrachloride was used instead of the bromide.

The carefully purified substance was decomposed

 

1 Lehrbuch, 5 Aufl., 3, 1200.
3 Compt. Rend., 20, 1047. 1845.

3 Ann. Chem. Pharm., 113,31. 1860.
4 Ann. Chem. Pharm.., 120, 94.

5 Journ. Chem. Soc., 51, 576. 1887.
6 Zeitsch, anorg. Chem., 43, 251. 1905.
No. 3.] ATOMIC WEIGHTS—CLARKE 223

by water, the solution was evaporated to dryness, and the silica produced was weighed. In a
second communication Meyer discusses the possible retention of chlorine by the silica, and ,
shows that that error was avoided. The data obtained by Becker and Meyer follow, with
vacuum weights, and a percentage column computed by myself:

SiCl,. S103. Per cent.
4. 16733 1. 47597 35. 417
4, 69585 1. 66304 35. 415
4, 91918 1. 74204 30. 413
5. 37434 1. 90349 35. 418
5. 93985 2. 10364 35. 416

_ 6. 73605 2. 38570 35. 417
7. 16361 2. 53606 35. 402
7. 82779 2, 77242 35. 418

 

E Mean, 35. 4145+. 0017
Hence, Si= 28.231.

There are also two series of the determinations of the density of silicon tetrafluoride, SiF,.
Jaquerod and Tourpaian ? made their determinations at low pressures, and from the data thus
obtained computed the weight of the normal liter of the gas, namely, 4.6929 grammes. From
this, by two methods, they derive the molecular weight as follows:

By corresponding densities wield Se/eee Meas eke s ESERIES RS Sif, =104. 68
By critical constants............022..00eeeceeeeeeee mpaneetisk, 104. 64

Mean, SiF,=104. 66+. 0133

 

The determinations by Germann and Booth* are much more elaborate than those of
Jaquerod and Tourpaian, and were made with purer material and subjected to very thorough
corrections. Four series of determinations were made on gas of increasing purity, but only
the last two were used in the final discussion of the data. In these the following figures are
given for the weight in grammes of a “‘gross”’ liter of silicon tetrafluoride at approximately 760
mm pressure:

4. 68167 4, 68343
4. 68197 4. 68690
4. 68292 4, 68082

 

Mean, 4. 68295-4. 00059

From this, reduced to standard f orm, the weight of the normal liter becomes 4.68397 grammes.
Germann and Booth do not go so far as to compute from this figure the molecular weight of the
fluoride, but by comparison with the deductions of Jacquerod and Tourpaian it should be approx-
imately :

SiF, = 104.47 +.0132

From this, the atomic weight of silicon becomes

Si= 28.306 + .0142

This value may require adjustment later. Jacquerod and Tourpaian’s figure for the mo-
lecular weight gives Si= 28.496, which is certainly too high and need not be considered further.
The ratios relative to the atomic weight of silicon are now—

(1) 4Ag : SiCL, :: 100 : 39.4265 + .0071.
(2) 4AgCl : SiCl, :: 100 : 29.6125 + .0138.
(3) SiCl, : SiO, :: 100 ; 35.4145 + .0017.
(4) SiBr, : SiO, :: 100 : 17.347 + .0027.
(5) Mol. weight SiF,, 104.47 +.0132.

 

1 Zeitsch. anorg. Chem., 46,45. 1905. In Vol. 43, p. 242, Meyer discusses the problem of the calculation of atomic weights.
2 Journ. Chim. Phys., 11, pp. 28, 269. 1913.
3 Journ. Phys. Chem., 21,81. 1917.

161566°—20—_15
224 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

To reduce these ratios we have

Ag=107. 880+. 00016 Br=79. 9228+. 00015
Cl=35. 4605+. 00011 F=19, 011+. 0013
From ratio— az

Deca aaSe cele sane Sead Saba OS EES EDS SRSA EERE SS Si=27. 945+. 0792

B adedeinacntiarn waaeanasaels aie tas octets we remeeeas 28. 231+. 0045

Dae cao culesoantins caweeedd Lcks fase wanaebouneas 28, 293+. 0307

Beas cated baa cua cece kas oe eng oIehS 28. 306 +. 0142

BOSS SORT ee RAL TEE SE Cr Oe Oto eas 28. 426+. 0115
General mean.............22-2----220eeeeee ae 28. 255+. 0025

The rounded-off mean, Si= 28.25, is probably as near the truth as any of the individual
values. New determinations of the atomic weight of silicon are evidently desirable.

TITANIUM.

Our knowledge of the atomic weight of titanium is based almost if not quite entirely upon
the determinations made by Thorpe. The earlier work of Berzelius,’ Rose,? Mosander,? Pierre,‘
and Demoly * has now only historical value and for the present purposes may be neglected here.
It is sufficiently summarized in the former editions of this work.

In 1883 Thorpe published a series of experiments upon titanium tetrachloride, deter-
mining three distinct ratios and getting fairly concordant results. The first ratio, which was
essentially like Pierre’s, by decomposition with water and titration with silver, was in detail as
follows:

TiCl,. Ag. TiCl, : 100Ag.
2. 43275 5. 52797 44. 008
5. 42332 12. 32260 44. 015
3. 59601 8. 17461 44, 000
3. 31222 7. 62721 44. 003
4. 20093 9. 54679 44. 004
5. 68888 12. 92686 44. 008
5. 65346 12. 85490 43. 979
4,.08247 9. 28305 43. 978

; Mean, 43. 999+. 0032
Hence, Ti= 48.022. .

The second ratio, which involved the weights of TiCl, taken in the last five determinations
of the preceding series, included the weighing of the silver chloride formed. The TiCl, pro-
portional to 100 parts of AgCl is given in a third column:

TiCl,. AgCl. Ratio.
3. 31222 10. 00235 33. 114
4. 20093 12. 68762 33. 111
5. 68888 17. 17842 33. 117
5. 65346 17. 06703 33. 125
4, 08247 12, 32442 33. 125

; Mean, 33. 118+. 0019
Hence, Ti= 48.133.

In the third series the chloride was decomposed by water, and after evaporation to dryness
the resulting TiO, was strongly ignited:

TiCl,. TiO>. Per cent TiO.
6. 23398 2. 62825 42, 160
8. 96938 3. 78335 42. 181
10. 19853 4, 30128 42, 176
6. 56894 2.77011 42. 170
8. 99981 3. 79575 42. 176

8. 32885 3. 51158 42. 162

. Mean, 42. 171+. 0022
Hence, Ti= 48.101.

1 Gilbert’s Annalen, 1823, pp. 67 and 129. 4 Ann. Chim. Phys. (3), 20, 257. 1847.
2 Poggend. Annalen, 15, 145. 5 Ann. Chem. Pharm., 72, 214. 1849.
3 Berz. Jahresbericht, 10, 108. 1831. 6 Ber. Deutsch. chem. Gesell., 16, 3014. 1883.

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 225

Some two years later Thorpe published his work more in detail,! and added a set of deter-
minations, like those made upon the chloride, in which titanium tetrabromide was studied.
Three ratios were measured, as was the case with the chloride. In the first, the bromide was
decomposed by water and titrated with a silver solution. :

TiBr,. Ag. TiBr, : 100Ag.
2. 854735 3. 34927 85, 235
3. 120848 3. 66122 85. 241
4. 731118 5. 55097 85. 230
6. 969075 8. 17645 85. 234
6. 678099 7. 83493 80. 234

 

; Mean, 85. 235+. 0027
Hence, Ti= 48.115.
In the four last experiments of the preceding series, the silver bromide formed was weighed.
The third column gives the TiBr, proportional to 100 parts of AgBr.

TiBr,. AgBr. Ratio.
3. 120848 6. 375391 48. 951
4. 731118 9. 663901 48. 957
6. 969075 14, 227716 48. 982
6. 678099 13. 639956 48. 959

 

: Mean, 48. 962+. 0049
Hence, Ti= 48.117.
For the third ratio the bromide was decomposed by water; and after evaporation with
ammonia the residual titanic oxide was ignited and weighed:

TiBr,. TiO,. Per cent TiO,.
6. 969730 1, 518722 21. 790
8. 836783 1. 923609 21. 768
9. 096309 1. 979513 21. 762

 

Mean, 21. 773+. 0062
Hence, Ti= 48.073.
The following ratios are now available for computing the atomic weight of titanium:

(1) 4Ag : TiCl, :: 100 : 43.999 + .0032.
(2) 4AgCl : TiCl, :: 100 : 33.118+.0019.
(8) TiCl, : TiO, :: 100 : 42.171 + .0022.
(4) 4Ag : TiBr, :: 100 : 85.235 +.0027.
(5) 4AgBr, : TiBr, :: 100 : 48.962 + .0049.
(6) TiBr, : TiO, :: 100 : 21.773 + .0062.

Reducing these ratios with Ag = 107.880 + .00016; Cl=35.46054 .00011; and Br=79.9228
+.00015 we have—

From ratio—

Disieie cvieiccna amin tie seaeseeeeasunntaontesetee Ti=48. 022+. 0138
Gece din sie teraeeerisalenn ga siersaieevistais ohne Saiseattigene’e 48. 073+. 0292
Qecee aes oiegeeseed ve da vance eetek oaeeeEeee 48. 101+. 0072
4 ice ie Sak eRe acess ts SEER Sets Sees 48. 115+. 0117
1D wranctsdvassiee, ace. aenGrbremsgansicial ere eceusioveySuGhesshdesveia: Soota euecaraval gins 48. 117+. 0369
Diskivesasetarai na spsyarivolaai cee a Svante ne ene Ca ores eyes sides 48. 133+. 0110

General Mean s «acini: eviag sisisisiersie sso’ 3's se cvainie Ti=48. 099+. 0049

This value may be rounded off to Ti=48.1.

1 Journ. Chem. Soc., 47, 108 and 129. 1885.

 
226 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou,-XVI.

GERMANIUM.

The data relative to the atomic weight of germanium are imperfect, and due entirely to
the discoverer of the element, Winkler.’ The pure tetrachloride was decomposed by sodium
carbonate, mixed with a nowi excess of standard silver solution, and then titrated back with
ammonium sulphocyanate. ‘ The data given are as follows:

 

GeCl,. cl found. Per cent Cl.
0. 1067 0. 076112 66. 177
. 1258 . 083212 66.146
. 2223 . 147136 66. 188
. 2904 . 192190 66. 182
Mean, 66.173.

Hence, Ge= 72.504.
aoe ZIRCONIUM. ;

The atomic weight of zirconium has been determined by Berzelius, Hermann, Marignac,
Weibull, Bailey, Venable, and Venable and Bell. The determinations by Berzelius? and
Hermann’ may be neglected for present purposes.

Marignac’s determinations ¢ were based upon analyses of the double fluoride of zirconium
and potassium. His weights are as follows:

1. 000 grm. gave.0. 431 grm. Zr0, and 0.613 grm. K. 50,

2.000 “ 864 = 1, 232 es
664 “ 982 ‘i . 399 a
5.000“ 2,169 i 3.078 ts

These figures give us three ratios. A, the ZrO, from 100 parts of salt; B, the K,SO, from
100 parts of salt; and C, the ZrO, proportional to 100 parts of K,SO,:

 

 

 

A, B. a
43. 100 “61. 300 70: 310
43. 200 61. 600 70. 130
43.119 61. 000 70. 677
43. 380 61. 560 70. 468
Mean, 43. 200+. 043 Mean, 61. 365. 094 Mean, 70. 3964. 079
Hence Zr=90. 03 91.53 90. 67

Weibull * ignited zirconium sulphate, and also made a similar set of experiments with the
selenate, obtaining results as follows:

Sulphate. Zr(SO,)2-

Sulphate. ' Zr0,. Per cent.
1. 5499 0. 6684 43. 126
1. 5445 . 6665 _ 43.153
2. 1683 . 9360 43. 168
1. 0840 . 4670 43. 081

. 7913 . 3422 43, 321
. 6251 . 2695 43. 113
4704 . 2027 43. 091

 

Mean, 43. 150+. 0207
Hence, Zr=89.54.
Selenate. Zr(SeO,)>.

Selenate. Zr0,. Per cent.
1. 0212 0. 2323 32. 540
. 8418 . 2744 ; 32. 597

. 6035 . 1964 32. 544

. 8793 . 2870 32. 640

. 8089 . 1003 32. 470

 

Mean, 32. 558+. 0192
Hence, Zr= 90.79.

1 Journ. prakt. Chem. (2), 34,177. 1886. 4Ann. Chim. Phys. (3), 60, 270. 1860.
2 Poggend. Annal., 4,126. 1825. 6 Lund. Arsskrift, Vol. 18, 1881-’82.
8 Journ. prakt. Chem., 31,77. Berz. Jahresb., 25, 147.

 
No. 8.) moe ATOMIC WEIGHTS—CLARKE. 227

Bailey! also ignited the sulphate, after careful investigation of his material and of the
conditions needful to insure success. He found that the salt was perfectly stable at 400°,
while every trace of free sulphuric acid was expelled at 350°. The chief difficulty in the process
arises from the fact that the zirconia produced by the ignition is very light, and easily carried
off mechanically, so that the percentage found is likely to be too low. This difficulty was
avoided by the use of a double crucible, the outer one retaining particles of zirconia which
otherwise might be lost. The results, corrected for buoyancy of the air, are as follows:

Sulphate. Zr0,. ; Per cent.
2. 02357 0. 87785 43. 381
2. 6185 1, 1354 43. 360
2. 27709 . 98713 43. 350
2. 21645 - 96152 43. 385
1. 75358 - 76107 43. 402
1. 64065 .7120° 48. 397
2. 33255 1.01143 43. 361
1. 81105 . 78485 43. 337
Hence, Zr = 90.65. Mean, 43. 372+. 0056
This, combined with previous determinations, gives—
Weibilll ss sae sesiteny twee sepexike htaccess s sacesese esos wis 43. 150+. 0207
Bailey. ¢sscexesaes ed Psd aeaca diate aeies dies sUswiee ain eh deren eS 43. 372-4. 0056
© General mean......... 2.0 cceeeeeeee eee ceceeeeceees 43, 358-£. 0054

Venable? determined the atomic weight of zirconium by analysis of the oxychloride,
ZrOCl,.3H,0. This compound was purified by crystallization from hot hydrochloric acid and
dried in a stream of hydrochloric acid gas. Itwas then dissolved in water, and after evaporat-
ing the solution to dryness in a platinum crucible the residue was converted into zirconia by
prolonged ignition. The data are subjoined:

ve Zr0Cl,.8H,0. Weight-Zr0,. Per cent Zr0,.
5. 25762 2.78450 52. 961
3. 53994 1. 87550 52. 981
8. 25036 1. 72435 53. 051
1. 52245 . 80708 53. 012
2. 98802 1. 58274 * 52.969
2.11371 1.11920 52. 949
2. 38139 1. 26161 52. 978
1. 90285 1. 00958 53, 055
2. 61847 1. 38658 52. 954
1. 07347 . 56840 52. 951
Hence, Zr = 90. 819. Mean, 52. 986-+. 0085

The determinations by Venable and Bell ’ were based upon. analyses, by the standard
method, of zirconium tetrachloride. Their weights, reduced to a vacuum, and the ratio
4Ag : ZrCl, :: 100 : 2 are as follows:

ZrCl, Ag: Ratio.

5. 55372 10. 26178 54. 121

3.78445 6. 99198 54. 125

2. 97546 5. 49645 54, 134

7. 22480 - 18. 34222 54, 150

6. 35532 11. 73094. 54.175

6. 78989 7 12, 54259 BA 185

5. 65352 10. 44718 ‘B4, 115

2. 76975 5. 11980 54. 099

4. 30337 7. 95180 54. 106

4, 31934, 7. 98115 54. 119

2.42721 4: 48516 54. 116

4. 80724 8. 87607 54. 160

2. 35598 4, 35052 54, 154
Hence, Zr=91.746. | Beat, ba 113,008

 

1 Proc. Roy. Soc., 46, 74. Chem. News, 60, 32. 2 Journ. Amer. Chem. Soc., 20,119. 1898. 3 Journ. Amer. Chem. Soc., 39,1598. 1917.
228 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

For computing the atomic weight of zirconium we now have the tonne ratios:

(1) Percentage ZrO, in Zr(SO),, 43.358 + .0054.

(2) Percentage ZrO, in Zr(SeO,),, 32.558 + .0192.

(3) Percentage 7x0, from K,ZrF,, 43.200 + .043.

(4) Percentage K,SO, from K,ZrF,, 61.365 +.094.
(5) Percentage ZrO, in ZrOCI,.3H,O, 52.986 + .0085.
(6) 4Ag : ZrCl, :: 100 : 54.1315 +.0043.

(7) K,SO, : ZrO, :: 100 : 70.396 4.0079.

The antecedent values for reduction of these ratios are—

Ag=107. 880+. 00016 K=89. 0975+. 00011
Cl=85. 4605+. 00011 F=19. 011+. 0013
S=32. 0650+. 00012 Se=79. 176+. 0029
H=1. 0078+. 00001
Hence,
From ratio—
Siccs os 5aa.be See eaeese a edeeweaeNe sen dahens Zr=90. 206+. 2390
Is. g.cnscres 22 ab geonaauadeed aoemewes emma eae 90. 575+. 0193
Paces asasitey oe deedeltonaore sc oeeee lean Miata ene: 90. 672+. 0138
2 siesta cies AAO Aa eer laiiaratsione eae 90. 790+. 0804
Doct else ve wunc ga semenauuiad thse pares z 90. 819+. 0302
Boas von cscs does eueaadiesnaseRuneosiansueunmie : 91. 606+. 4350
Gi seers ule eis vice ec aes ean eee ee CCCs 91. 746+. 0116
General mean.............--- suvbekeaeeres Zr=91. 1745+. 0080

This mean is unsatisfactory because the values are too divergent. That from ratio 6 ap-
pears to be much the best, but it is not absolutely final. More work on zirconium is needed.

TIN.

The atomic weight of tin has been determined by means of the oxide, the chloride, the bro-
mide, the sulphide and the stannichlorides of potassium and ammonium.

The composition of stannic oxide has been fixed in two ways: by synthesis from the metal
and by reduction in hydrogen. For the first method we may consider the work of Berzelius,
Mulder and Vlaanderen, Dumas, Van der Plaats and Bongartz and Classen.

Berzelius! oxidized 100 parts of tin by nitric acid, and found that 127.2 parts of SnO, were
formed. Hence, Sn=117.65.

The work done by Mulder and Vlaanderen? was done in connection with a long investi-
gation into the composition of Banca tin, which was found to be almost absolutely pure. For
the atomic weight determinations, however, really pure tin was taken prepared from pure tin
oxide. This metal was oxidized by nitric acid, with the following results. One hundred parts

of tin gave of SnO,:
127. 56—Mulder
127. 56—Vlaanderen
127. 43—Vlaanderen

. Mean, 127. 517+. 029

 

Hence, Sn= 116.3.

Dumas? oxidized pure tin by nitric acid in a flask of glass. The resulting SnO, was strongly
ignited, first in the flask and afterwards in platinum. His weighings, reduced to the foregoing
standard, give for dioxide from 100 parts of tin the amounts stated in the third column:

Sn. Sn0,. Ratio.
12. 443 15. 820 127.14
1. 976 20. 301 t 127. 07

 

Mean, 127. 105+. 024
Hence, Sn = 118.06.

1 Poggend. Annal., 8, 177. aJourn. prakt. Chem., 49, 35. 1849. 8 Ann, Chem, Pharm., 113, 26,

 
No. 8.] ' ATOMIC WEIGHTS—CLARKKE. 229

In an investigation later than that previously cited, Vlaanderen! found that when tin was
oxidized in glass or porcelain vessels, and the resulting oxide ignited in them, traces of nitric
acid were retained. When, on the other hand, the oxide was strongly heated in platinum, the
latter was perceptibly attacked, so much so as to render the results uncertain. He therefore,
in order to fix the atomic weight of tin, reduced the oxide by heating it in a porcelain boat in a
stream of hydrogen. Two experiments gave Sn=118.08, and Sn=118.24. These become, if

reduced to the above common standard,
127. 100

127. 064

Mean, 127. 082+. 012
Hence, Sn=118.16. =

Van der Plaats? prepared pure stannic oxide trom Banca tin, and upon the material obtained
made two series ot experiments; one by reduction and one by oxidation. The results, with
vacuum weights, are as follows, the ratio between Sn and SnO, appearing in the third column:

Oxidation series.

Sn. Sn0,. Ratio.
9. 6756 12. 2967 127.091

12. 7356 16. 1885 127.114

23. 4211 29. 7667 127, 093

Reduction series.

Sn0,. Sn. Ratio.
5. 5015 4, 3280 427, 114
4. 9760 8. 9145 127.117
3. 8225 38. 0078 127. 086
2. 9935 2. 3553 127. 096

Mean of both series as one, 127. 102+. 0033
Hence, Sn = 118.07.

The reductions were effected in a porcelain crucible.

Bongartz and Classen® purified tin by electrclysis, and oxidized the electrolytic metal by
means of nitric acid. The oxide found was dried over a water-bath, then heated over a weak
flame, and finally ignited for several hours in a gas-muffle. Some reduction experiments gave
values which were too low. The oxidation series was as follows, with the usual ratio stated in

a third column: Be Sn0,. Ratio.
2. 5673 3. 2570 126. 865
3. 8414 4, 8729 126, 852
7. 3321 9. 2994 126. 831
5. 4367 6. 8962 126. 845
7. 3321 9. 2994 126. 831
9. 8306 12. 4785 126. 935

11. 2424 14. 2665 126, 896
5.5719 7. 0685 126. 860
9. 8252 12. 4713 126. 932
4, 3959 5. 5795 126. 925
6. 3400 8. 0440 126. 877

Mean, 126. 877-4. 0080
Hence, Sn = 119.06.

We now have six series of experiments giving the amount of SnO, formed from 100 parts
of tin. To Berzelius’ single determination may be assigned the weight of one experiment in
Mulder and Vlaanderen’s series :

Berzelive. vc ciccicjee s ccicanee s dewiees bs s axeeeeewewee ses ece 127. 200-4. 041
Mulder and Vlaanderen.......-.-.....------------ soanast 127. 517+. 029
UMS vise idvineie sock So aso ae Syd. dan cuted Deiat Boh e 127. 105+. 024
Vinanderen oieieic ss os axnaeeccigis does eects aeewceews wees sins 127. 082-4. 012
Vander Plaats..«.cccvex scueiseeseex se ovoddaeietens o's'2es 127, 102+. 0033
Bongartz and Classen.......---------- 20. 2se cece eee eee 126. 877+. 0080
General mean .......)..¢2ce08 v= ees eeeeaadewese se seus 127. 076+. 0026

 

1 Jahresbericht, 1858, 183. 2Compt. Rend., 100, 52. 1885. 8 Berichte Deutsch. chem. Gesell., 21, 2900. 1888.
230 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

Dumas, in the paper previously quoted, also gives the results of some experiments with
stannic chloride, SnCl. . This was titrated with a solution containing a known weight of silver.
From the weighings given, 100 parts of silver correspond to the quantities of SnCl, named i in
the third column :

SnCl,. ‘Ag. Ratio.
1. 839 3. 054 60. 216
2. 665 4, 427 60. 199

Mean, 60. 207+. 006
Hence, Sn = 117.96.

Tin tetrabromide and the stannichlorides of potassium and ammonium were all studied
by Bongartz and Classen; who, in each compound, carefully purified, determined the tin elec-
trolytically, The data given are as follows, the percentage columns being added.by myself:

Tin Tetrabromide.
SnBr, taken. Sn found. Per cent Sn.

8. 5781 ~ 2.3270 ° 27.127

9, 5850 2. 6000 27.126

9. 9889 2.7115 27.145
10. 4914 2. 8445 27.113
16. 8620 4. 5735 27.123
16. 6752 4. 5236 27.119
11. 1086 3. 0125 27.116
10. 6356 2. 8840 27.113
11. 0871 3. 0060 27.123
19. 5167 5. 2935 27. 128

Mean, 27. 123+. 0020
Hence, Sn = 118.98.
Potassium Stannichloride.

K,SnCl,. Sn found. Per cent Sn.
2. 5718 0.7472 29. 054
2. 2464 . 6524 29. 042
9. 3353 2.7100 29. 030
12. 1525 3. 5285 29. 035
12. 4223 3. 6070 29. 036
15. 0870 4, 3812 29. 040
10. 4465 3. 0330 29. 034
18. 9377 5. 5029 29. 058
18. 4743 5. 3680 29. 029
17. 6432 5, 1244 29. 045

Mean, 29. 040+. 0021
Hence, Sn = 119.07. Pe t
Ammonium. Stannichloride.

Am,SnCl,. Sn found. Per cent Sn.
1. 6448 0. 5328 32. 393
1. 8984 6141 32. 347
2.0445 . 6620 32. 381
2. 0654 . 6690 32. 391
2. 0058 . 6496 32. 386
2. 4389 ~ 7895 32. 371
4, 0970 1. 3254 32. 351
8, 4202 1.1078 32. 390
8. 6588 1. 1836 32. 349
1. 5784 . 5108 32. 362
7. 3248 2. 3710 32. 370

13. 1460 4, 2528 $2. 351
11. 9483 3. 8650 32. 348
18. 4747 5. 9788 32. 362.
18. 6635 6. 0415 32. 371
17. 8894 5. 7923 32. 378

Mean, 32. 369+. 0088
Hence, Sn= 119.1.
No. 2.1 ATOMIC WEIGHTS—CLARKE. 231

One other method of determination for the atomic weight of tin was employed by Bongartz
and Classen. Electroytic tin was converted into sulphide, and the sulphur so taken up was
oxidized by means of hydrogen peroxide, by Classen’s method, and weighed as barium sul-
phate. The results, as given by the authors, are subjoined :

Sn taken. Per cent of S gained.

2. 6285 53. 91

- 7495 53. 87
1. 4785 53. 94
2. 5690 53. 94
2.1765 53. 85
1.3245 53. 88

- 9897 53. 83
2. 7160 53. 86

Mean, 53. 885+-. 0098

This percentage of sulphur, however, was computed from weighings of barium sulphate.
What values were assigned to the atomic weights of barium and sulphur is not stated, but as
Meyer and Seubert’s figures are used for other elements throughout this paper, we may assume
that they apply here also. Putting O=15.96, S=31.98, and Ba=136.86, the 53.885 per cent
of sulphur becomes 392.056 + .0713 of BaSO,, the compound actually weighed. This a us

the ratio—
Sn : 2BaSO, :: 100 : 392.056+.0713

as the real result of the experiments, from which, with the later values for Ba, S and O, the
atomic weight of tin may be calculated.

A single determination of the atomic weight of tin, made by Schmidt,’ ought not to be
overlooked, although it was only incidental to his research upon tin sulphide. In one experi-
ment, 0.5243 grm. Sn gave 0.6659 SnO,. Hence, Sn=118.49. This lies about midway between
the two sets of values already computed.

The determinations of the atomic weight of tin by Briscoe? were made by, decomposing
stannic chloride by water, and titrating the solution with a standard solution of silver. Oxalic
acid, or Rochelle salt was used to prevent the precipitation of stannic oxide. His weights,
reduced to a vacuum, and the ratio 4Ag : SnCl, :: 100 : « follow :

SnCl,. Ag. Ratio.
9. 02435 14. 9475 60.3736
9.39855 15. 5666 60. 3764
10. 32760 17. 1058 60. 3748
9.17005 15. 1880 60. 3769
: 6.68995 11. 0808 60. 3742
9.33130 15. 4570 60. 3694
' 7.77450 12. 8764 60. 3779
9. 22730 15. 2840 60. 3723
7, 55785 12. 5183 60. 3744
9.11535 15. 0983 60. 3732
9. 22875 15. 2866 60.3715
9.21890 15. 2696 60. 3742
11. 29055 18.7017 60.3715
9.89635 16.3912 60. 3760
9. 43385 15. 6250 60.3767

 

Mean, 60. 3742-4. 0004
Hence, Sn= 118.691.
. For the same ratio Dumas’: mes give the value 60.207 +.0060. Combined with Bris-
coe’ s value, the general mean is 60.3735 + .00039.

Her, Deutsch, Chem. Ges., 27, 2743, 1894, 3 Journ. Chem. Soc., 107, 63. 1915,

 
232 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI,

Baxter and Starkweather‘ also analyzed stannic chloride, but by electrolysis; collecting
the tin on a weighed mercury cathode. With vacuum weights their data are as follows:

SnCl,. Sn. Per cent Sn.
11. 64269 5. 30498 45, 565 >
18.5943 6.1935 45. 560
10. 0897 4.5971 45, 561
11. 4319 5. 2080 45. 557
12. 2869 5. 5983 45. 563
12. 20889 5. 56286 45. 564
10. 7469 4.8965 45, 562
11. 54233 5. 25846 45. 558
15. 65437 7.18198 45.559
16. 23310 7.39664 45. 565
17.29151 7.87850 45.563
15. 04889 6. 85695 45. 565
18. 36074 8.36507 45. 560
21. 58929 9. 83676 45. 563
17. 22210 7.84718 45. 565
15. 70516 7.15589 45. 564

 

Mean, 45. 562+ .00045
Hence, Sn= 118.715.
For tin there are now the following available ratios:

(1) Sn : SnO, :: 100 : 127.076 +.0026.

(2) 4Ag : SnCl, :: 100 : 60.3735 + .00039.
(3) SnCl, : Sn :: 100 : 45.562 + .00045.

(4) SnBr, : Sn :: 100 : 27.123 +.0020.

(5) K,SnCl, : Sn :: 100 : 29.040+.0021.

(6) (NH,),SnCl, : Sn :: 100 : 32.369 +.0088.
(7) Sn : 2BaSO, :: 100 : 392.056+.0713.

To reduce these we have—

Ag=107. 880+.00016 N= 14.0081 +. 00008
Cl= 35. 4605+. 00011 S= 32.0650+.00012 ,
Br= 79. 9228+.000]5 Ba=137. 358 +. 0023
K= 39.0975+. 00011 H= 1.0078+.00001
Hence,
From ratio—

D2 here eats atinad arn aneasaan nna nie ae ee nee sino Sn=118. 186+.0113
Dh ct nis meee ne oda Mem Auer sate anatanesauieace 118. 681+. 0019
Qoustetaadh seaieoen tun owareeeesees cease gamete: 118. 715+. 0016
As a brace a cue eine Rissasaiars ig -a-a dete EEE wee eee eR Mets 118. 981+. 0094
Digcos aeS I eeA MaeMaddaa een eate 119. 074+. 0093
Tiwi sdeanculs ee dd aca eee eE noe pe cand emai wen 119. 076+. 0876
Blisdelajioectesdigd se gc+'<2.2:3 aaetasraraigpalvise stab avone: eeouete att 119. 099+. 0359
Goneral mean, «sss 27625 <ecceeerseie seeks 118. 708+ .0012

This general mean is almost identical with the values found in the two most recent investi-
gations. Their low probable errors give them controlling weight in the combination. For a
careful discussion of sources of error in the earlier determinations, the paper by Briscoe should

be consulted.
ADDENDA.

Since the preceding pages were put in type, two new papers on the atomic weight of tin
have appeared.’ First, Brauner and Krepelka, by a series of. electrolyses of ammonium stan-
nichloride, obtained unsatisfactory values. By analyses of stannic bromide, however, they
secured better results. The data, with vacuum weights, are as follows:

 

1 Proc. Nat. Acad., 2,718. 1916. In fullin Journ, Amer. Chem, Soc., May, 1920.
2 Jour. Amer. Chem. Soc., 42, pp. 917-928. 1920.
No. 3.] ATOMIC WEIGHTS—CLARKE. . 233

SnBr,. Ag. At.Wt. Sn.
2. 11624 2. 08430 118.55 Rejected.
1.11964 1. 10206 118. 73
1, 97428 1. 94359 118. 67
2, 35469 2. 31788 118. 71

Mean of last 3, 118. 70

This series of determination is prelimimary. In the second paper, by Krepelka alone, the
following figures were obtained:

SnBr,. Ag. At. Wt. Sn.
. 5.11788 5. 03796 118. 702
2. 46875 2. 43035 118. 674
. 99510 . 97961 118. 679
1. 69834 1. 67172 118. 727
38. 54265 3. 48737 118. 697
3. 18280 3. 76199 118. 717

Mean, 118. 699

The values found for tin in these two series confirm the determinations made by Briscoe and by

Baxter and Starkweather.
THORIUM.

The earlier determinations of the atomic weight of thorium by Berzelius, Chydenius,
Delafontaine, and Hermann, have now only historical interest, and need no detailed considera-
tion here.

The modern determinations begin with those of Cleve,? who analyzed the sulphate and
oxalate of thorium. The anhydrous sulphate, calcined, gave the subjoined percentages of
thoria:

62. 442 62. 470
62.477 62. 357
62. 430 62. 366

Mean, 62.423+.014
Hence, Th= 234.01.

The oxalate was subjected to a combustion analysis, whereby both thoria and carbonic
acid could be estimated. From the direct percentages of these constituents no accurate value
can be deduced, there having undoubtedly been moisture in the material studied. From the
ratio between CO, and ThO,, however, good results are attainable. This ratio I put in a fourth
column, making the thoria proportional to 100 parts of carbon dioxide:

Oxalate. ThO,. CO,. Ratio.

1. 7135 1. 0189 0. 6736 151, 262
1. 3800 . 8210 . 5433 151. 114
1.1850 . . 7030 . 4650 151. 183
1. 0755 . 6398 . 4240 150. 896

Mean, 151. 114+. 053
Hence, Th = 234.02.

In 1882 Nilson’s determinations appeared.? This chemist studied both the anhydrous
sulphate and the salt with nine molecules of water, using the usual calcination method, but
guarding especially against the hygroscopic character of the dry Th(SO,), and the calcined
ThO,. The hydrated sulphate gave results as follows:

Th(SO,)>.9H,0. TRO». Per cent ThO,.
2. 0549 0. 9267 45. 097
2. 1323 . 9615 45. 092
3. 0017 1. 3532 45. 081
2.7187 1, 2235 45. 086
2. 6280 11849 45. 088
1. 9479 . 8785 45. 099

Mean, 45. 091+. 0019
Hence, Th= 232.65.

1 These determinations are fully discussed in the former editions of this work, q. v. 3 Ber. Deutsch. chem. Gesell., 15, 2519, 1882.
3K. Svensk. Vet. Akad. Handl., Bd. 2, No, 6, 1874.

 
234 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The anhydrous sulphate gave data as follows:

Th(SO,)o. ThO.. Per cent ThO,.
1. 4467 0. 9013 62. 300
1. 6970 © 1. 0572 62, 298
2. 0896 1. 3017 62. 294
1.5710 . 9787 62, 298

“Tantes: Th— 239, 58.

 

Mean, 62. 297+. 0009

The last four determinations appear again in a paper published fives years later by Kriiss
and Nilson, who, however, give four more made upon material obtained from a different source.

The new dats are subjoined:

Th(SO,4)o- ThO,. Per cent ThO,.
1. 1630 0. 7245 62. 296
- 8607 . 5362 62. 298
1. 5417 - 9605 62. 301
1, 5217 . 9479 62. 292

 

Mean, 62. 297+. 0013
Hence, Th = 232.58.

Urbain,’ who purified his material by crystallizing thorium acetylacetonate from solution
in chloroform, gives the following analyses of the anhydrous sulphate, effected by calcination:

Th(SO,4)5. TRO». Per cent ThO,.
1. 0925 0. 6815 62. 374
. 5926 . 3699 62. 420
1. 0230 . 6384 62. 405

 

Mean, 62. 400+. 0096
Hence, Th= 233.75.

Meyer and Gumperz,’ in order to determine whether thorium is complex or not, prepared
the octohydrated sulphate from material of diverse origin, and analyzed it by dehydration and
calcination. Their data, which I give as one series, represent, first, six experiments upon prep-
arations obtained by fractional precipitation as chromate; and, secondly, six analyses of the
sulphate prepared from three samples of thorium chloride. I give here only the weights of the
anhydrous sulphate and the oxide, for the reason that the hydration of the compound was too
irregular to yield good values for the atomic weight of thorium:

Th(SO,)>- TRO». . Per cent ThO>.
0.9301 0.5793 62, 284
. 9927 . 6184 62, 295
1. 0344 . 6442 62. 278
. 9349 _ 5821 62. 263
. 6680 . 4160 62. 276
. 4296 . 2676 62, 291
. 9199 . 5730 62, 289
. 7647 . 4764 62, 299
1, 0650 . 6638 62. 300
. 1758 . 4884 62. 310
. 8824 . 5496 62. 285
. B45 . 3454 62. 290

Hence, Th = 232.48.

Mean, 62. 288+. 0024

In an early paper Brauner * and later Baskerville § claimed to have fractionated thoria into

different component earths, which varied widely in atomic weight and physical properties.
These claims, however, were not fully substantiated. Recently, Brauner,® in an exhaustive
critical discussion of the atomic weight of thorium, has published the details of his work, and
offered an explanation of the discordances. Omitting variant series of determinations, and

 

4 Proc, Chem. Soc., 17, 67.
5 Journ. Amer. Chem. Boc., 23, 761, and 26, 022,
6 Abegg’s Handbuch, 3, (2), 805,

1 Ber. Deutsch, chem, Gesell,, 20, 1665, 1887.
2 Ann. Chim. Phys. (7), 19, 223. 1900.
83 Ber. Deutsch. Chem. Ges., 38, 817. 1905.
No. 3.] ATOMIC WEIGHTS-—CLARKE.

235

also his analyses of thorium oxalate, Brauner gives two sets of analyses of the sulphate which

he regards as satisfactory. The data are as follows:

Series 1.
Th(SO,)o. TRO. Per cent ThO>.
1. 7165 1, 0699 62. 330
- 94702 . 59075 62. 380
Mean, 62. 355+. 0170
Hence, Th = 233.12.
Series 2.
Th(SO,)>. ThO>. Per cent ThO>.
1.93405 1. 2044 62. 282
1, 14252 . 71152 62. 276
1, 89330 1. 17925 62. 285
1, 96495 1, 22395 62. 289

Hence, Th = 232.43.
These figures combine with those of previous investigations thus:

Mean, 62. 283+. 0027

CIOV6 neeinctee ie hgnad be oa eae a hesedeads soe eccaneeels 62.4234 .0140
INDSOT 24 cious hercs Pecaan Deeuastekkee sl. pel seaused Wao: 62.297 + .0009
Kriss and Nilson. .... 22.2.0. 0 2-2 e eee cece ee eee cece ee 62.2974 .0013
UP bali. scchioeseni add aes 5 we ealsae Sees Saeed cc 62.400+ .0096
Meyer and Gumperz................- poco hael Rist Rit sale, 62.288-+ .0024
Pan Ord cheeses ait 020 aed NS le alan rial mae 62.3554 .0170
Brauner 2) Ale csis.e cisco 5 qovnb ste eee teas Suosaroseseace 62.283+ .0027

General méanen.-2 cee css iieeeeuiexice sdccaceeweweds 62.299 + .0007

A different procedure for determining the atomic weight of thorium was adopted by Hénig-
schmid,! who made analyses of the bromide by the standard method. His figures, with vaccum

weights, and the usual ratios are as follows

ThBr,. Ag. AgBr. Ag ratio. AgBr ratio.
5.11577 4.00046 6.96402 127.879 73.460
3.85077 3.01126 5.24210 127.879 73.459
4.9327 3.82643 6.66091 127.881 73.462
4.70546 3.67975 6.40573 127.874 73.457
5.12854 4.01057 6.98157 127.876 73.458
4.18710 3.27425 5.69982 127.880 73.460
3.66181 2.86343 4.48479 127.882 73.460

' 3.74590 2.92948 5.09931 127.869 73.459
4.47249 3.49762 6.08888 127.872 73.452
5.02409 3.92861 6.83873 127.885 73.465
5.34994 4.18375 7.28310 127.874 73.457
4.32353 3.38105 5.88552 127.875 73.460

 

Mean, 127.877 + .0009
Hence, from Ag ratio, Th = 232.124.

From AgBr ratio, Th=232.141.

The ratios from which to compute the atomic weight of thorium are now as follows:

(1) 4CO,:ThO,::100:151.114 + .053.

(2) Th(SO,),.9H,O: ThO,:: 100: 45.091 + .0019.
(3) Th(SO,),: ThO,:: 100: 62.299 + .0007.

(4) 4Ag: ThBr,:: 100: 127.877 + .0009.

(5) 4AgBr: ThBr,:: 100: 73.459 + .0006.

To reduce these ratios we have—

Ag=107.880+ .00016 S=32.0650+.00012
Br= 79.9228 + .00015 C=12.0025+.00019
H=1.0078+.00001

73.459 4.0006

 

1 Zeitsch. Elektrochem., 22, 18, 1916. From abstract by Baxter in Journ. Amer. Chem. Soc., 39, 340.
236 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.
Hence,
From ratio— Heo
Ah tas sym Sis iaigein gfe gy cie'y' ps ese gra aiaisievoretero aes acid od wine Th=282.124+ .0040
Be woes eae eA Eh: eat toad oree eet 232.141.0056
Oi ,ake are eObaistada tread suet SioSk Gererealtarantons eI 232.607 + .0058
Dina Ce ceiiveeeuees Mie lads da mbemawen eae ees 232.647 + .0145
Vea uneweebene ss 2elct seve dees saieeeeeeeeeeeee 234.021 + .0933

General mean... ..........-.--+-------- Th=232,.192+ .0028

This general mean is slightly higher than the value derived from Hénigschmid’s analyses
alone, which, on chemical grounds, appear to be the best. It would hardly be wise, however,
in the light of present knowledge, to reject the values found by analyses of thorium sulphate.
They, therefore, receive proper recognition in the final combination. The difference between
the general mean and the value given by Hénigschmid’s work alone, is only about 1 part in 4,000.

The foregoing discussion relates only to what may be called normal thorium, which appears
to be a perfectly definite element. The thorium contained in uranium ores, however, contains
an admixture of its radioactive isotope, ionium, an element which is chemically undistinguishable
from thorium, except that it has a lower atomic weight. With thorium from uranium ores
Hénigschmid has made several atomic weight determinations and obtained the following results:

 

Th(Io)Br,. Ag. AgBr. Ag ratio. AgBr ratio.
3.21504 2.51702 4.38166 127.731 73.373
2.73569 2.14178 3.72848 127.729 73.372
4.27574 3.34762 5.82951 127.724 73.371
Mean, 127.728 73.372

From the Ag ratio Th(Io) = 231.481.

From the AgBr ratio Th(Io) =231.488.

The atomic weight of ionium free from thorium would doubtless be smaller still, perhaps
as low as 231.2, or even a little lower. On this question further evidence is needed.

PHOSPHORUS.

The material from which to calculate the atomic weight of phosphorus is by no means
abundant. SBerzelius, in his Lehrbuch,’ adduces only his own experiments upon the precpi-
tation of gold by phosphorus, and ignores all the earlier work relating to the composition of the
phosphates. These experiments have been considered with reference to gold.

Pelouze,’ in a single titration of phosphorus trichloride with a standard solution of silver,
obtained a wholly erroneous result; and Jacquelain,’ in his similar experiments, did even worse.
Schrétter’s criticism upon Jacquelain sufficiently disposes of the latter.*

Only the determinations made by Schrétter, Dumas, Van der Plaats, Ter Gazarian and
Baxter and his colleagues remain to be considered.

Schrétter® burned pure amorphous phosphorus i in dry oxygen, and weighed the pentoxide
thus formed. One gramme of P yielded P.O; in the following proportions:

2.28909 2.28788
2.28783 2.28848
2.29300 2.28856
2.28831 2.28959
2.29040 2.28872

Mean, 2.289186 + .00033
Hence, P=31.027.

Dumas°® prepared pure phosphorus trichloride by the action of dry chlorine upon red
phosphorus. The portion used in his experiments boiled between 76° and 78°. This was titrated
with a standard solution of silver in the usual manner. Dumas published weights, from which
I calculate the figures given in the third column, representing the quantity of trichloride pro-
portional to 100 parts of silver:

 

16th ed., 1188.
2Compt. Rend., 33, 698.
3 Compt. Rend., 20, 1047.

4Journ. prakt. Chem., 57, 315.
6 Journ. prakt. Chem., 53, 485. 1851.
6 Ann. Chem. Pharm., 113, 29. 1860.
No. 3.J ATOMIC WEIGHTS—CLARKE. 237

PCl;. Ag. Ratio.

1. 787 4, 208 42. 4667
1. 466 3. 454 42. 4435
2. 056 4. 844 42, 4443
2. 925 6. 890 42. 4528
3. 220 7. 582 42. 4690

Hence, P=31.021. Mean, 42. 4553 +.0036

By Van der Plaats‘ three methods of determination were adopted, and all weights were
reduced toavacuum. First, silver was precipitated from a solution of the sulphate by means of
phosphorus. The latter had been twice distilled in a current of nitrogen. The silver, before
weighing, was heated to redness. The phosphorus equivalent to 100 parts of silver is given in
the third column: :

P Ag. Ratio.
0. 9096 15. 8865 : 5. 7256
- 5832 10. 1622 5. 7389

Mean, 5. 7322-+.0045
Hence, P=30.920. nae =

The second method consisted in the analysis of silver phosphate; but the proceess is not
given. Van der Plaats states that it is difficult to be sure of the purity of this salt.

Ag3;P 0,4. Ag. Per cent.
6. 6300 5. 1250 77. 300
_ 12.7170 9. 8335 77. 326

Hence, Pp ae 30.970. Mean, Ts 3138+ -0088

In the third set of determinations, yellow phosphorus was oxidized by oxygen at reduced
pressure, and the resulting P,O, was weighed:

P. P,0;. Ratio.
10. 8230 24. 7925 2. 29072
7. 7624 17. 7915 2. 29201

Hence, P=30.975.

As these figures fall within the range of Schrétter’s, they may be averaged in with his series,
the entire set of 12 determinations giving a mean of 2.28955 +.00032.

Ter Gazarian? determined the density of gaseous phosphine, from which its molecular
weight is deducible. For the weight of the normal liter, in grammes, he found—

1. 52955 1. 52944
1. 52907 1, 52907
1. 52933 1. 52933

Mean, 1. 52930+.000054

From these figures, reduced by means of the critical constants, Ter Gazarian finds
PH, =33.931. The probable error is 0.0012. Hence, P=30.908.

Baxter and Jones * based their determinations of the atomic weight of phosphorus upon
analyses of silver phosphate. This salt was dissolved, and the silver precipitated and weighed
as bromide, and in one experiment as chloride.

The weights, in vacuo, and the ratios are as follows:

Ag;P0,. AgBr. Ratio.

6. 20166 8. 34490 134, 558
6. 35722 8 55419 134. 559
5. 80244 7. 80819 134. 567
5. 05845 6. 80685 134, 564
7, 15386 9. 62694 134, 570
7, 20085 9. 68947 134. 560
6. 20182 8. 34522 134. 561
5. 20683 7. 00605 134. 555

Mean, 134. 562+.0012
Hence, P=31.058.
1 Compt. Rend., 100, 52. 1885. _— -
2 Journ. Chim. Phys., 7,837. 1909. Seealso notein Journ. Chim. Phys,, 9, 109; in which Ter Gazarian replies to criticisms.
8 Journ. Amer, Chem. Soc., 32, 298. 1910.
238 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Ag,P0,. AgCl. Ratio.
3. 34498 3. 43544 102. 704 +.0034

Hence, P =31.060.
The probable error assigned to the last ratio is that of one experiment in the bromide series.
Two years later Baxter, Moore, and Boylston! published three series of analyses of phos-
phorus tribromide, with all the usual precautions to insure accuracy. Their figures, with
vacuum weights, are as follows:

 

 

 

 

Series 1.
' PBry. Ag. AgBr. Agratio. . +‘ AgBr ratio.
8. 95293 7. 11504 : 12. 38560 83. 6668 48. 0633
4. 71056 5. 63022 9. 80058 83. 6657 48. 0641
4. 72373 5. 64583 9. 82830 83. 6676 48. 0624
6. 47622 7. 74048 13. 47428 83. 6669 48. 0636
4. 61956 5. 52140 9. 61143 83. 6665 48. 0632
7. 62060 9.10849 see 83.6648 = ......-
3. 83321 4, 58163 7. 97563 83. 6648 48. 0615
4.72578 5. 64835 9. 83261 83, 6666. 48. 0623
Mean, 83. 6662-+.00024 48, 0629--.00022
Serves 2.

PBr;. Ag. AgBr. Ag ratio. AgBr ratio.
3. 21808 3. 84649 6. 69553 83. 6628 48. 0631
5. 77604 6. 90379 12. 01761 83. 6648 48. 0631
5. 51730 6. 59456 11. 47946 . 83, 6644 48. 0624
7. 15048 8.546538 = ........ 83.6658. = ......-
7. 94753 9. 49930 16. 53569 83. 6644 48. 0629
4. 30924 5. 15074 8. 96595 83. 6625 48. 0623

Mean, 83. 6640+.00031 48. 0628+ .00010
Series 3.

PBr;. Ag. AgBr. Ag ratio. AgBr ratio.
4. 39626 5. 25447 . 9. 14663 83. 6671 i - 48. 0643
7. 07758 8. 45957 14. 72642 83. 6636 48. 0604
4, 19854 5.01851 8. 73586 83. 6611 48. 0610
7. 26540 "8. 68405 15. 11725 83. 6637 48. 0603
7. 75072 9. 26406 16. 12737 83. 6644 48. 0594

 

 

a5 y Mean, 83. 6640+.00065 48. 0611 + .00057
Combining these series we have—

3Ag:PBr, ::100:83 :6653 + .00019
Hence, P=31.006.

Hence, P=31.021.

The preceding investigation was shortly followed by one by Baxter and Moore,? who made
similar analyses of phosphorus trichloride.

Their data are as follows:

3AgBr:PBr, ::100:48.0627 + .00009

Series 1.

PCl, Ag. Ratio.
3. 60897 8. 50148 42, 4511
3. 49255 8. 22734 42. 4505
3. 80549 8. 96427 42. 4517
4. 64502 10, 94212 42. 4508
3. 99011 9. 39943 42. 4505
38. 38036 7. 96301 42. 4508

 

Mean, 42. 4509 + .00012
Hence, P=31.004.

1 Proc. Amer. Acad., 47, 585. 1912, 2 Journ. Amer. Chem. Soc., 34, 1644,

 
No. 8.]

PCL
4,59507
4.40117
5.71561
5.18668
5.15583
5.13108
4.85457
3.58844
5.91915

ATOMIC WEIGHTS—CLARKE.

Series 2 and 8.

Ag. AgCl.
10.82441 14.38118
10.36771 13.77494
13.46408 17.88933
12.21814 16.23480
12.24555 16.13727
12.08725 16.05962
11.43585 15.19245

8.45324 11.23091
13.94377 18.52592

From the Ag ratio, P =31.006.
From the AgCl ratio, P=31.005.
The two silver ratios combine with Dumas’ determinations as follows:

Ag ratio.

42.4510
42.4508
42.4508
42.4506
42.4504
42.4504
42.4505
42.4505
42.4502

 

239

AgCl ratio.
31.9520 —
31.9506
31.9506
31.9489
31.9498
31.9502
31.9538
31.9515
31.9506

 

Mean, 424506+ .00006 31.9509+ .00032

Dumas__.....-.------------ 42.4553 + ,0036

Baxter and Moore, 1__....... 42.4509 +.00012

Baxter and Moore, 1____._-.-- 42.4506 +.00006
General mean._____.-_ 42.4515 +.000054

The following ratios are now available for computing the atomic weight of phosphorus:
(1) 2P : P,O, :: 1 : 2.28955 +.00032.

(2) 3Ag : PCI, :: 100 : 42.4515 + .000054.
(3) 3AgCl : PCI, :: 100 : 31.9509 + .00032.
(4) 3Ag : PBr, :: 100 : 83.6653 +.00019.

(5) 83AgBr : PBr, :: 100 : 48.0627 +.00009.
(6) Ag,PO, : 3Ag :: 100 : 77.313 +.0088.

(7) Ag,PO, : 3AgCl :: 100 : 102.704 +.0034.
(8) Ag,PO, : 3AyBr :: 100 : 134.562 +.0012.

(9) 5Ag : P :: 100 : 5.7322 +.0045.
(10) PH, =33.931 £.0012.

To reduce these

Hence,

we have—

Ag=107.880- .00016
C1=35.4605 + .00011

From ratio—

General mean........-.---.--.-------

Br=79.9228 + .00015
H = 1.0078+.00001

P=

ee re

30.908 + .0012
30.920+ .0243
30.970 + .0384
31.005+ .0014
31.006 + .00086
31.009 + .00043
31.018 + .0077
31.021+ .00075
31.058 + .0039
31.060 + .0140

31.004+ .00032

If the lowest of all these values, which has, apparently, excessive weight be rejected, the
general mean of the remaining nine values becomes P =31.011 4.00033.
161566°—20-——16
240 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.
VANADIUM.

Roscoe’s determination of the atomic weight of vanadium was the first to have any scien-
tific value. The results obtained by Berzelius 1 and by Czudnowicz ? were unquestionably too
high, the error being probably due to the presence of phosphoric acid in the vanadic acid em-
ployed. This particular impurity, as Roscoe has shown, prevents the complete reduction of
V,0, to V,O, by means of hydrogen. All vanadium ores contain small quantities of phosphorus,
which can only be detected with ammonium molybdate—a reaction unknown in Berzelius’
time. Furthermore, the complete purification of vanadic acid from all traces of phosphoric
acid is a matter of great difficulty, and probably never was accomplished until Roscoe under-
took his researches.

In his determination of the atomic weight, Roscoe * studied two compounds of vanadium,
namely, the pentoxide, V,O;, and the oxychloride, VOCI,. The pentoxide, absolutely pure,
was reduced to V,O, by heating in hydrogen, with the following results:

7.7397 erm. V,0; gave 6.3827 grm. V,0;. 17.533 per cent loss.
6.5819 5.4296 9 © 17.507 “
5.1998“ 4.9819 “ 17.489 “
5.0450 4164 17.51.

5.4296 grm. V202» reoxidized, gave 6.5814 grm. V,0;. 17.501 per cent difference.
Mean, 17. 509+. 005

 

Hence, V =51.381.

Upon the oxychloride, VOCI,, two series of experiments were made—one volumetric, the
other gravimetric. In the volumetric series the compound was titrated with solutions con-
taining known weights of silver, which had been purified according to the methods recommended
by Stas. Roscoe publishes his weighings, and gives percentages deduced from them; his
figures, reduced to a common standard, make the quantities of VOCI, given in the third
column proportional to 100 parts of silver. He was assisted by two analysts:

Analyst A.
VOC Ag. Ratio.
2.4322 4.5525 ; 53.425
4.6840 8.7505 53.528
4.2188 7.8807 53.533
3.9490 7.3799 53.510
9243 1.7267 53.530
1.4330 2.6769 53.532
Analyst B.
voc, Ag. Ratio.
2.8530 5.2853 53.980
2.1252 3.9535 53.755
1.4248 2.6642 53.479

 

. Mean, 53.586+-.039
Hence, V =51.045.
The gravimetric series, of course, fixes the ratio between VOCI, and AgCl. If we put the
latter at 100 parts, the proportion of VOCI, is as given in the third column:

Analyst A.

VOCl, AgCl. Ratio.
1.8521 4.5932 40.323

7013 © 1.7303 40.531

« .7486 1.8467 40.537
1.4408 3.5719 40.337
9453, 2.3399 40.399

1.6183 4.0282 40.174

 

1 Poggend. Annal., 22,14. 1831. 2 Poggend. Annal., 120,17. 1863, * Journ. Chem. Soc., 6, pp. 330 and 344. 1868,
No. 3.] |" ATOMIC WEIGHTS—CLARKE. 241

Analyst B.
VOCI, AgCl. Ratio.
2.1936 5.4039 40.391
2.5054 6.2118 40.333

 

Mean, 40.378 + .028

Hence, V =51.253.

The determinations of the atomic weight of vanadium by Prandtl and Bleyer ' appeared
in two papers. Following Roscoe they analyzed vanadium oxychloride, and also studied the
reduction of the pentoxide by nydiogen In their first paper the following figures, with vacuum
weights are seeps

 

Series 1.
wn VOC. | AgCl. Ratio.
5. 47218 13. 54724 40. 393
5. 85234 14. 50771 40. 346
3. 23175 8. 00636 40. 365
5. 24732 18. 01359 40. 322
3. 56589 8. 83375 40. 367
Mean, 40. 359+. 0080
Hence, V =51.171.:

Series 2. ;
VOCI,. AgCl. Ratio.
4.91432 12. 18494 40. 331
3. 64470 9. 04685 40. 286
4, 96088 12. 30438 40. 318
6. 46766 16. 04232 40. 315
4, 33158 10. 74624 40. 308
4. 05060 10. 04498 40. 325

 

Mean, 40. 314+. 0043
Hence, V =50.978.
Four additional determinations given in Prandtl and Bleyer’s second paper I cite here as

series 3.

Series 8.
VOCh. | AgCl. ° Ratio.
5. 53988 18. 73492 40.:334
7. T7585 19. 26836 40. 356
8. 41904 20. 86554 40. 343
10.66137 26. 42699 40. 343

Mean, 40. 345+. 0037
Hence, V=51.111.
As in Roscoe’s determinations vanadium pentoxide was reduced to trioxide by heating in
a current co hydrogen. The following figures are given:

oe: 'V05 / |" V,05 Per cent lost.
9, 11431 7. 51639 - 17.532
9. 85727 8. 13127 17. 510
8. 70923 7. 18456 . 17. 506

12. 26426 10, 11721 17.507,

 

. Mean, 17. 514+. 0041

' Hence, V=51:357.
Combining this series with that of Roscoe, the general mean becomes 17.5117 +.0034.

An entirely different method of determining the atomic weight of vanadium was adopted

by McAdam.2 Sodium metavanadate was heated in a stream of gaseous hydrochloric acid,

 

1 Zeitsch. anorg. Chem., 65, 152. 1909. 67, 257. 1910. See also the academic dissertation by Bleyer, Miinchen, 1910,
3 Journ, Amer. Chem, Soc., 32, 1603. 1910.
242 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou XVI.

the vanadium was thus volatilized as chloride, and sodium chloric remained behind. The
following data, with vacuum weights, were thus obtained:

NaVO;. NaCl. Per cent NaCl.
4. 8564 2.3277 47. 981
5. 6404 2. 7033 47.927
4, 4263 2. 1220 47. 941
5. 7805 2.7710 47. 937
9. 4902 4, 5478 : 47, 921

Mean, 47. 981+. 0032

Hence, V=50.971. :

The determinations by Briscoe and Little! were based on analyses of the agyanlatide: with
all the precautions to ensure accuracy which have been recognized in recent times. Two sets
of determinations are given, with vacuum weights, as follows: Two analyses, rejected by the
authors, are omitted here.

Series 1.
VOCI. Ag. Ratio,
3. 98162 7. 43451 53. 556
5. 68804 10. 52736 53. 556
3. 88228 7. 24884 53. 557
3. 28485 6. 18377 53. 554
3. 45920 : 6. 45900 53. 556
3. 17088 5. 92089 53. 554
3. 63990 6. 79670 538. 554

Mean, 53. 555+. 00036
Hence, V = 50.944.

. Series 2.
VOCI. AgCl. Ratio.
4, 42279 10. 97187 40. 310
2. 62557 6. 51459 40. 307

Mean, 40. 3085+. 0010
Hence, V=50.941.

These VOCI, ratios can now be combined with the previous determinations thus:
Ratio 3Ag : VOCI;:: 100: x.

PUOSC00 Fa ais eons dashes oniecie ceiisaunee sen aeae os cicak eateaslaaiee 53. 586+. 039
Briscoe and Little........... 2.222220... ee cence eee eee eee 53. 555+. 00036
General mean..........22.0.. 2022 eee eee eee ee eee 53. 555+. 00035
Ratio $AgCl: VOCI, : : 100: x.
Resend exunsngrnesa sds acauidvennie RpapuPnnesasexetdaeuet 40.378 +. 028
Prandtl and Bleyer 1.....-..----22222--cceeeeeeeeeeeeees 40.359 + .0080
Prandtl and Bleyer 2...........02 2.22. seeee eee eee cence eee 40. 314 +. 0043
Prandtl and Bleyer 3............20-.0---1 cece eee eee eee es 40.345 +. 0037
Briscoe and Little. ...-...-s---acecrescrnetingespercees -- 40.3085+. 0010
General MEBs essseragerepazecereweneeenpeaennp es 40. 3119+. 00094

There are now four ratios relative to the atomic weight of vanadium. They are—
(1) V,O; : 20 :: 100 : 17.5117 =.0084.
(2) 3Ag: ‘VOC, : : 100 : 53.555 + .00035.
(3) 3AgCl : VOCI, : : 100 : 40.3119 + .00094.
(4) NaVO, : NaCl :: 100 : 47.931 +.0032.

Reducing these with Ag=107.880+.00016; Cl=35.4605+.00011; and Na=23.0029+
. 00012, we have—

sigh a ina aden ei ide ahi a V=50. 944+. 00043

Basco cucscn’ cide he tug ewes a eee eerie 50. 969. 0018
Boe wdrn Hoda wrlew eee eee eee mmm ete ee nn anaes 50. 971+. 0030
Diese Sea oe Su Ewe Sraceravalde€ So kee eae aeeeiene Se SSR RSS 51. 367+. 0299

General mean.....-....----.. eee eee eee ee 50. 947+. 0004

 

1 Journ. Chem. Soc., 105, 1310, 1914.
No. 3.] ATOMIC WEIGHTS—CLARKE. 243

The determinations by Briscoe and Little dominate the entire combination. The value
derived from vanadium oxide practically vanishes.

ARSENIC.

For the determination of the atomic weight of arsenic five compounds have been studied—
the chloride, the trioxide, and three arsenates. The bromide may also be considered, since it was
analyzed by Wallace in order to establish the atomic weight of bromine. His series, in the
light of more recent knowledge, may properly be inverted, and applied to the determination of
arsenic.

In 1826 Berzelius' heated arsenic trioxide with sulphur in such a way that only SO, could
escape. 2.203 grammes of As.O,, thus treated, gave a loss of 1.069 of SO,. Hence, As=75.02.

In 1845 Pelouze? applied his method of titration with known quantities of pure silver to
the analysis of trichloride of arsenic, AsCl,. Using the old Berzelian atomic weights, and put-
ting Ag =1349.01 and Cl=443.2, he found in three experiments for As the values 937.9, 937.1,
and 937.4. Hence, 100 parts of silver balance the following quantities of AsCl,:

56. 029

56. 009
56. 016

Mean, 56.018 +-. 004

 

Hence, As=74.91.
Later, the same method was employed by Dumas,’ whose weighings, reduced to the fore-
going standard, give the following results:

AsCl;. : Ag. Ratio.
4, 298 7. 673 56. 015
5. 535 9. 880 56. 022
7. 660 18. 686 55. 970
4. 680 8. 358 55. 993

 

Mean, 56.000+. 008

Hence, As=74.86. .

The two series of Pelouze and Dumas, combined, give a general mean of 56.014 +.0035,
as the amount of AsCl, equivalent to 100 parts of silver. Hence, As=74.91, a value closely
agreeing wito that deduced from the single experiment of Berzelius.

The same process of titration with silver was applied by Wallace ‘ to the analysis of arsenic
tribromide, AsBr,. This compound was repeatedly distilled to insure purity, and was well
crystallized. His weighings show that the quantities of bromide given in the third column are
proportional to 100 parts of silver:

AsBr;,. Ag. Ratio.
8. 3246 8. 58 97. 023
4, 4368 4, 573 97. 022
5. 098 5. 257 96. 970

 

Mean, 97.005+. 012

Hence, As=74.18. Why this value should be so much lower than that from the chloride
is unexplained.

The volumetric work done by Kessler,’ for the purpose ot establishing the atomic weights of
chromium and of arsenic, is described in the chromium chapter. In that investigation the
amount of potassium dichromate required to oxidize 100 parts of As,O, to As,O, was determined
and compared with the quantity of potassium chlorate necessary to produce the same effect.
From the molecular weight of KCIO,, that of K,Cr,O, was then calculable.

 

1 Poggend. Annalen, 8, 1. 4 Phil. Mag. (4), 18, 270.
2 Compt. Rend., 20, 1047. 6 Poggend. Annal., 95, 204. 1855. Also 113, 134, 1861.

8 Ann, Chim. Phys. (3), 55,174. 1859.
244 MEMOIRS NATIONAL ACADEMY OF ‘SCIENCES. [Von. XVI.

'' From the same figures, the molecular weights of KCI1O, and of K,Cr,O, being: both known,
that of As,O, may be easily determined. The quantities of the other compounds proportional to
100 parts of As,O, are as follows:

E070, : KCI0,.

a eg. 98. 95 os 41. 156
; : 98.94 | 41.116 -
99.17 a 41, 200.
98. 98 41, 255
99. 08 41.201 ~
99.15 41, 086
— 41. 199
Mean, 99. 045+, 028 41, 224
; 41.161
41, 193
41.149
41. 126

 

Mean, 41. 172+. 009
Another series with the dichromate gave the following figures:
‘99. 08
99. 06
99. 10

98. 97
98. 97

Mean, 99.036+. 019
Previous series, 99. 045+. 028

General mean, 99. 039+. 016

Other defective series are given to illustrate the partial oxidation of the As,O, by the action
of the air. From Kessler’s data we get two values for the atomic weight of As, thus:
‘From KiCl Oj SeTles ou daesecceoseeeteiesciecankaks As=75.224
From K,Cr,O, series........-.-..-+-- eee eciedatate}ate _- -As=75.029
’ The determinations made by Hibbs are weseid upon an altogether different process from
any of the preceding measurements. Sodium pyroarsenate was heated in gaseous hydrochloric
acid, yielding sodium chloride. The latter was perfectly white, completely soluble ‘in water,
unfused, and absolutely free from arsenic. The vacuum weights are subjoined, with a column
giving the percentage of chloride obtained from the pyroarsenate.

Na,As,0;. NaCl. "Percentage.
0..02177 0. 01439 66. 100
. 04713 . 03115 66. 094
. 05795 . 03830 66. 091
. 40801 . 26981 66, 128
. 50466 . 33345 66. 092
. 77538 . 51249 66. 095
. 82897 . . 54791 66. 095
1, 19124 . 78731 66. 092
1. 67545 1. 10732 66. 091
3. 22637 2. 13267 “66. 101

 

Mean, 66.098-4. 0080
Hence, As=74, 893. 2

1 Doctoral thesis, University of Pennsylvania, 1896. Work done under'the direction of Professor E, F. Smith. In the fifth experiment a
weight of NaClis printed 0.33045, This is evident)va misprint, which I have corrected by comparison with the other data. The rejection of thts
experiment would not affect the final result appreciably.
No. 8.] ' ATOMIC WEIGHTS—CLARKE. 245

The determinations by Ebaugh+ are analogous to those of Hibbs. First, silver arsenate
was converted into silver chloride ‘by heating in gaseous hydrochloric acid, and the chloride
was afterwards reduced to metal in a stream of hydrogen. The data obtained are as follows:

Ag,As0Q,. AgCl, Ag. Per cent AgCl. Percent. Ag.
0. 23182 0. 21547 0. 162175 92. 947 69. 957
. 47996 - 44615 . 33583 92. 956 69. 970
- 52521 . 48820 . 367525 92. 953 69. 977
- 80173 . 74517 . 56099 92. 945 69. 972
. 94782. . °8083 . 66318 92. 932 69. 969
1. 02047 . 94830 . 71400 92. 928 69. 968
1. 03558 -96258 see O290,- = ae ead

_ 1, 05462 - 98014 - 73771 92. 938 69. 950
Mean, 92.9444.0025 ‘69. 966+.0024

 

 

From Ag series, As = 74.928.
From Ag(Cl series, As =75.027.
A similar series of experiments with lead arsenate gave the subjoined figures:

‘Pb,(As04)>. PbCl. Per cent PbCl.g
0. 38152 0. 35381 92. 737

. 436197 . 40449 92. 731

. 57218 . 53065 92. 742

. 60085 . 55717 92. 730

. 74128 . 68736 92. 732

77107" . 71494 92. 721

. 88282 . 81858 92. 723

. 97779 . 90674 92, 734

 

Mean, 92. 731+. 0019
Hence As= 75.077.

Lead arsenate was also transformed into lead bromide, by heating in a stream of hydro-

bromic acid:
Pb,(As0,).. PobBr,. Per cent PbBr,.

0. 59704 0. 73092 122. 424
. 61712 - 75567 122, 451
. 65799 . 80569 122. 447

 

Mean, 122. 441+. 0076
Hence As = 74.877.

All of Ebaugh’s weights are reduced to a vacuum.

Silver arsenate was also chosen by Baxter and Coffin? for their determinations of the
atomic weight of arsenic. In some experiments Ebaugh’s method of heating in gaseous hydro-
chloric acid was adopted; in others the arsenate was dissolved in nitric acid, and the silver
then precipitated as chloride or bromide. Corrections were applied, not only for weighing in
air, but also for traces of moisture in the initial substance. Different samples of the arsenate
were prepared, which gave slightly varying results for the atomic weight, and the determina-
tions, for that reason, fall into two groups. In series 1 and 2, which may be treated as one
here, the first five determinations were made by Ebaugh’s method, and the last two by solaion
and Brecipitaeioat Ths figures thus obtained are as follows:

“Ag,As0,. as AgCl. Ratio,
3.17276 ftind tg 2.94922 92. 9544
2. 65042 2. 46367 92. 9539
3. 51128 8, 26396 92. 9564
5. 83614 - “5, 42503 92. 9558
5. 72252 5.31947 ° 92. 9568
4, 59149 4, 26796 92. 9537
3. 38270 3, 14436 92. 9542

 

Mean, 92. 9550+. 00036
Hence As = 74.973.

1 Doctoral thesis, University of Pennsylvania, 1901. 2 Journ, Amer. Chem. Soc., 31, 297. 1909.

 
246 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Series 4 and 5, with silver arsenate of different origin from that previously used, gave the
subjoined figures. Only the last experiment was conducted by the precipitation method:

Ag,As0,. AgCl. Ratio.
4, 67268 4, 34389 92. 9636
7. 71882 7.17597 92. 9672
5. 28049 4. 90908 92. 9664
4. 25346 8. 95424 92. 9652
3. 47340 3. 22893 92. 9616
5. 17269 4. 80879 92. 9650
4. 10766 3. 81858 92. 9624
5. 47133 5. 08643 92. 9646

 

Mean, 92. 9681+. 00044
Hence As= 74.907.

These series combine with Ebaugh’s thus:

Ebaugh winqesioiateiajayaiegele Sibis S'S ere Biate leg © sale ais wt orate Aree emery 92.944 +. 0025

Baxter and:Coffin, 1oscic.25 65 s''3045 4 oog a nredansaatueaier 92. 9550+. 00036

Baxter and Cofin, 2s<c0scaeseeyesexessessé s pees eeenes 92. 9681+. 00044
General: méan sss is sas seo eeeeresieweakeeies 92. 9614+. 00028

Baxter and Coffin also determined the ratio between silver arsenate and silver bromide
by the solution and precipitation method. Here again two series of analyses are given, num-
bered 3 and 6, representing different preparations of the arsenate. The two series are as follows:

 

Series 8,
Ag,As0,. AgBr. Ratio.
8. 75751 10. 66553 121. 787
6. 76988 8. 24545 121. 796
5. 19424 6. 32590 121. 787
5. 33914 6. 50258 121. 791
8, 24054 10. 03552 121. 782
7. 57962 9, 23147 121. 793
6. 05230 : 7. 37106 121. 789
Mean, 121. 789+. 0016
Series 6.
Ag,As0,. AgBr. Ratio.
4, 96261 6. 04440 121. 7988
5. 81743 6. 47658 121. 7991
4, 46882 5. 44300 121. 7995
4. 16702 5. 07539 121, 7990

 

Mean, 121. 7991+. 00015

In the last mean the probable error is so low as to give it inordinate weight, especially as
Baxter and Coffin suspect the presence of basic impurities in the arsenate. It is better, there-
fore, to treat both series as one, giving in mean Ag,AsO,: 3AgBr : : 100 : 121.793 +.0012.
Hence As= 74.955.

There are now the following ratios from which to compute the atomic weight of arsenic.
The single determination by Berzelius has been arbitrarily assigned equal weight with that of
Wallace’s series: :

(1) 2As,0, : 350, :: 100 : 48.525+.012.

(2) 3Ag : AsCl, :: 100 : 56.014+ .0035.

(8) 3Ag : AsBr, :: 100 : 97.005+.012.

(4) 8As,0, : 2K,Cr,O, :: 100 : 99.039 + .016.
(5) 3As,0, : 2KCIO, :: 100 : 41.172+.009.
(6) Na,As,O, : 4NaCl :: 100 : 66.098 + .0030.
No. 8.] ATOMIC WEIGHTS—CLARKE. 247

(7) Ag,AsO, : 3Ag :: 100 : 69.966 +.0024.

(8) Ag,AsO, : 3AgCl :: 100 : 92.9614 + .00028.
(9) Ag,AsO, : 3AgBr :: 100 : 121.793+.0012.
(10) Pb,As,O, : 3PbCl, :: 100 : 92.7314 .0019.
(11) Pb,As,O, : 3PbBr, :: 100 : 122.441+.0076.

To reduce these ratios we have—

Ag=107. 880+. 00016 Na=23. 0029+. 00012
Cl=35. 4605+. 00011 K=89. 0975+. 00011
Br=79. 9228+. 00015 Cr=52. 019+. 0013

S=32. 0650+. 00012 Pb=207. 102+. 00062
Hence,
From ratio—
Died caw esas deen beet ov iso deeraterase Selutere eae wavlaln As=74. 179+. 0389
Deeatel. y bes ines deems clas Fae AO eERe See Se cA 74, 877+. 0234

Ge sagen oer Re ope pee Apu arena Sd ating epee Naam tes ra ade 74, 893+. 0080

che Gy At Leask cE NN A clea SAU Sa yr 74. 901+. 0113

Tesi ce fata aos dasa atm oases oN eswos sar anaT ets wa a eas WAS aa caran da OS 74, 928+. 0160

Brose schalasei tan tare MaNay Maes sha fetet tres hd Seen eee eet te ae 74. 941+. 0015

Oecd us Sse elot tine ee Dad ae ee oe extant 74, 955+. 0026

Li seewicten ns dix See Someihss is 5 eee se eReRs 75. 019+. 0077

os fags Shan tne ei cee Sainseyead eaters Be a Stirs cage eevansaseenet svete ese 75. 029+. 0157

VO ec caszisicade Sak aan sta ov ateneedge. 5 dvo-o Sabai eee Se Ae 75. 077+. 0062

Deedes ares SiN aD is eect che Sok ee aya Soke a orere 75. 224+. 0217

General mean...............22.200-0200000e- As=74. 956+. 0012

This final mean is practically identical with the value derived from the analyses of silver

arsenate by Baxter and Coffin.
ANTIMONY.

After some earlier, unsatisfactory determinations, Berzelius,! in 1826, published his final
estimation of the atomic weight of antimony. He oxidized the metal by means of nitric
acid, and found that 100 parts of antimony gave 124.8 of Sb,O,. Hence Sb=129.03.. The value
129 remained in general acceptance until 1855, when Kessler,’ by special volumetric methods,
showed that it was certainly much too high. Kessler’s results will be considered more fully
further along, in connection with a later paper; for present purposes a brief statement of his earlier
conclusions will suffice. Antimony and various compounds of antimony were oxidized partly by
potassium dichromate and partly by potassium chlorate, and from the amounts of oxidizing
agent required the atomic weight in question was deduced:

’ By oxidation of Sb,0, from 100 parts of Sb..............2...-- Sb=123. 84
By oxidation of Sb with K,Cr,0,.. seeSiukiee seaeeeccea © 2123: 61,
By oxidation of Sb with KC10,--K,Cr,0;.. diewesehie ane aes 1238.92)
By oxidation of Sb,0, with KC10,+K,Cr,0, bh Geers SFG ha Re ** =123. 80
By oxidation of Sb,S, with Re Gr Os So) es area: “ =123. 58
By oxidation of tartar emetic..........2...0 02 cece eee cece ec eee ** =119. 80

The figures given are those calculated by Kessler himself. A recalculation with our newer
atomic weights for O, K, Cl, Cr, S and C would yield slightly different values. It will be seen
that five of the estimates agree closely, while one diverges widely from the others. It will be
shown hereafter that the concordant values are all vitiated by constant errors, and that the ex-
ceptional figure is also worthless.

Shortly after the appearance of Kessler’s first paper, Schneider ® published some results
obtained by the reduction of antimony sulphide in hydrogen. The material chosen was a very
pure stibnite from Arnsberg, of which the gangue was only quartz. This was corrected for, and
corrections were also applied for traces of undecomposed sulphide carried off mechanically by
the gas stream, and for traces of sulphur retained by the reduced antimony. The latter sulphur

 

1 Poggend. Annalen, 8, 1. 2Poggend. Annalen, 95, 215. 8 Poggend. Annalen, 98, 293. 1856. Preliminary note in Bd. 97.
248 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

was estimated as barium sulphate. From 3.2 to 10.6 grammes of material were taken in each
experiment. The final percentages of S in Sb,S, were as follows:

28. 559 28. 532
28. 557 28. 485
28. 501 28. 492
28. 554 28, 481

Mean, 28. 520+. 008
Hence, Sb = 120.55.

Immediately after the appearance of Schneider’s memoir, Rose! published the result of a
single analysis of antimony trichloride, previously made under his supervision by Weber. This
analysis, if Cl=35.5, makes Sb = 120.7, a value of no great weight, but in a measure confirmatory
of that obtained by Schneider.

The next research upon the atomic weight of antimony was that of Dexter,? published in
1857. This chemist, having tried to determine the amount of gold precipitable by a knowu
weight of antimony, and having obtained discordant results, finally resorted to the original
method of Berzelius. Antimony, purified with extreme care, was oxidized by nitric acid, and
the gain in weight was determined. From 1.5 to 3.3 grammes of metal were used in each experi-
ment. The reduction of the weights to a vacuum standard was neglected as being superfluous.
From the data obtained, we get the following percentages of Sb in Sb,O, :

79. 268 79. 260
79. 272 79. 286
79. 255 79, 274
79. 266 79. 232
79. 253 79, 395
79. 271 79. 379
79. 264

Mean, 79. 283-4. 009
Hence, Sb = 122.46.

The determinations of Dumas were published in 1859. This chemist sought to fix the ratio
between silver and antimonious chloride, and obtained results for the atomic weight of antimony
quite near to those of Dexter. The SbCl, was prepared by the action of dry chlorine upon pure
antimony; it is was distilled several times over antimony powder, and it seemed to be perfectly
pure. Known weights of this preparation were added to solutions of tartaric acid in water, and
the silver chloride was precipitated without previous removal of the antimony. Here, as Cooke
has since shown, is a possible source of error, for under such circumstances the crystalline argento-
antimonious tartrate may also be thrown down and contaminate the chloride of silver. But be
that as it may, Dumas’s weighings, reduced to a common standard, give as proportional to 100
parts of silver, the: pariaties of SbCl, which are stated in the third of the following columns:

SCly. Ag. Ratio.

1. 876 2. 660 70. 526
4. 336 6. 148 70. 527
5. 065 7.175 70. 592
3. 475 4.930 70. 487
3. 767 5. 350 70.411
5.910 8. 393 70. 416
4, 828 6. 836 70. 626

; Mean, 70. 512+. 021

Hence, Sb = 121.82.

In 1861 Kessler’s second paper‘ relative to the atomic weight of antimony appeared.
Kessler’s methods were somewhat complicated, and for full details the original memoirs must
be consulted. A standard solution of potassium dichromate was prepared, containing 6.1466
grammes to the liter. With this, solutions containing known quantities of antimony or of anti-
mony compounds were titrated, the end reaction being adjusted with a standard solution of fer-
rous chloride. In some cases the titration was preceded by the addition of a definite weight of

 

1 Poggend. Annalen, 98, 455. 1856. 2 Ann. Chim. Phys. (3), 55, 175.
2 Poggend. Annalen, 100, 363. 1857. 4Poggend. Annalen, 113,145. 1861.
No.3.) : : ATOMIC WEIGHTS—CLARKE. 249

potassium chlorate, insufficient for complete oxidation; the dichromate then served to finish the
reaction. The object in view was to determine the amount of oxidizing agent, and therefore
of oxygen, necessary for the conversion of known quantities of antimonious into antimonic
compounds.

In the later paper Kessler refers to his earlier work, and shows that the values then found
for antimony were all too high, except in the case of the series made with tartar emetic. That
series he merely states, and subsequently ignores, evidently believing it to be unworthy of further
consideration. “ For the remaining series he points out the sources of error. These need not be
rediscussed here, as the discussion would have no value for present purposes; suffice it to say that
in the series representing the oxidation of Sb,O, with the dichromate and chlorate, the material
used was found to be impure. Upon estimating the impurity and correcting for it, the earlier
value of Sb = 123.80 becomes Sb = 122.36, according to Kessler’s calculations.

In the paper now under consideration four series of results are given. The first represents
experiments made upon a pure antimony trioxide which had been sublimed, and which consisted
of shining colorless needles. This was dissolved, together with some potassium chlorate, in
hydrochloric acid, and titrated with. dichromate solution. Six experiments were made, but
Kessler Teeels the first and second as untrustworthy. The data for the others are as follows:

S§b,03. KCIO0;. K,Cr,0, sol. in ce.
1, 7888 0. 4527 19..2
1. 6523 . 4506 3.9
3. 2998 . 8806 16.5
1. 3438 . 3492 10.2

From these figures Kessler deduces Sb = 122.16.

These data, reduced to a common standard, give the following quantities of oxygen needed
to oxidize 100 parts of Sb,O, to Sb,O;. Eack cubic centimetre of the K,Cr, O, solution corres-
ponds to one milligramme of O :

10. 985 10. 951
10. 939 10. 936
Mean, 10. 953+. 0075

_ Hence, Sb = 122.08.

‘In the second series of experiments pure antimony was dissolved in hydrochloric acid
with the aid of an unweighed quantity of potassium chlorate. The solution, containing both
antimonious and antimonic compounds, was then reduced entirely to the antimonious condition
by means of stannous chloride. The excess of the latter was corrected with a strong hydro-
chloric acid solution of mercuric chloride, then, after diluting and filtering, a weighed quantity
of potassium chlorate was added, and the titration with dichromate was performed as usual.
Calculated as above, the percentages of oxygen given in the last column correspond to 100
parts of antimony : Oe
Sb. KCi0;. . KA Cr,0; sol. ec. Per cent O.

1. 636, 0. 5000 18.3 13. 088
3. 0825 .» 9500 30. 2 13. 050
4, 5652 1. 4106 45.5 13. 098

: Mean, 13. 079+. 0096
Hence, Sb = 122.33.

The third and fourth series of experiments were made with pure antimony trichloride,
SbCl, prepared by the action of mercuric chloride upon metallic antimony. This preparation,
in the third: series, was dissolved in hydrochloric acid, and titrated. In one experiment solid
K,Cr,0, in weighed amount was added before titration; in the other two estimations KCIO,
was taken as usual. The third column gives the percentages of oxygen corresponding to 100

parts of SbCl,: a. =

ae . Per cent O.
1. 8576 grm. SbCl, needed . 5967 grm. K,Cr,O, and 33. 4 cc. sol. 7. 0338
1. 9118 i .3019 ‘* KClO; “© 162 “ 7. 0321
4. 1235 ey .6801 ‘* KClO, “ 23.2 = * 7. 0222

Mean, 7. 0294+. 0024
Hence, Sb = 121.24.
250 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The fourth set of experiments was gravimetric. The solution of SbCl,, mixed with tartaric
acid, was first precipitated by hydrogen sulphide, in order to remove the antimony. The
excess.of H,S was corrected by copper sulphate, and then the chlorine was estimated as silver
chloride in the ordinary manner. 100 varts of AgCl correspond to the amounts of SbCl, given
in the third column:

SbCl. AgCl. Ratio.
1. 8662 3, 483 53. 580
1. 6882 3.141 53. 588
2. 7437 5. 1115 53. 677
2. 6798 5. 0025 53. 569
5. 047 9. 411 53, 629
3. 8975 7. 2585 53. 696

Mean, 53. 623+. 015

The volumetric series with SbCl, gave Kessler values for Sb ranging from 121.16 to 121.47.
The gravimetric series, on the other hand, yielded results from Sb=124.12 to 124.67. This
discrepancy Kessler rightly attributes to the presence of oxygen in the chloride; and, ingen-
iously correcting for this error, he deduces from both sets combined the value of Sb= 122.37.

The several mean results for antimony agree so fairly with each other, and with the values
obtained by Dexter and Dumas, that we can not wonder that Kessler folt satisfied of their
general correctness, and of the inaccuracy of the figures published by Schneider. Still, the old
series of data obtained by the titration of tartar emetic with dichromate contained no evident
errors, and was not accounted for. This series, if we reduce all of Kessler’s figures to a single
common standard, gives a ratio between K,Cr,O, and C,H,KSbO, .4H,O. 100 parts of the
former will oxidize of the latter : .

336. 64 337, 93
338. 01 338. 59
336. 83 335. 79

Mean, 337. 30+. 29
From this Sb = 118.67.

The newer atomic weights found in other chapters of this work will be applied to the dis-
cussion of all these series further along. It may, however, be properly noted at this point that
the probable errors assigned to the percentages of oxygen in three of Kessler’s series are too low.
These percentages are calculated from the quantities of KCIO, involved in the several reactions,
and their probable errors should be increased with reference to the probable error of the molec-
ular weight of that salt. The necessary calculations would be more laborious than the
importance of the figures would warrant, and accordingly, in computing the final general mean
for antimony, Kessler’s figures will receive somewhat higher weight than they are legitimately
entitled to.

Naturally, the concordant results of Dexter, Kessler and Dumas led to the general accep-
tance of the value of 122 for antimony as against the lower figure, 120, of Schneider. Still, in
1871, Unger ? published the results of a single analysis of Schlippe’s salt, Na,SbS,.9H,O. This
analysis gave Sb=119.76, if S=32 and Na=23, but no great weight could be attached to the
determination. It served, nevertheless, to show that the controversy over the atomic weight
of antimony was not finally settled.

More than ten years after. the appearance of Kessler’s second paper the subject of the
atomic weight of antimony was again taken up, this time by Professor Cooke. His results
appeared in the autumn of 1877* and were conclusive in favor of the lower value, approxi-
mately 120. For full details the original memoir must be consulted; only a few of the leading
‘points can be cited here.

Schneider analyzed a sulphide of antimony which was already formed. Cooke, reversing
the method, effected the synthesis of this compound. Known weights of pure antimony were
dissolved in hydrochloric acid containing a little nitric acid. In this solution weighed balls
of antimony were boiled until the liquid became colorless; subsequently the weight of metal

 

1 Poggend. Annalen, 95, 217. 9 Archiv der Pharmacie, 197, 194. Quoted by Cooke. § Proc. Amer, Acad., 5, 13.
No. 3.} ATOMIC WEIGHTS—CLARKE. . 251

lost by the balls was ascertained. To the solution, which now contained only antimonious
compounds, tartaric acid was added, and then, with a supersaturated aqueous sulphhydric
acid, antimony trisulphide was precipitated. The precipitate was collected by an ingenious
process of reverse filtration, converted into the black modification by drying at 210°, and
weighed. After weighing, the Sb,S, was dissolved in hydrochloric acid, leaving a carbonaceous
residue unacted upon. This was carefully estimated and corrected for. About two grammes
of antimony were taken in each experiment and 13 syntheses were performed. In two of these,
however, the antimony trisulphide was weighed only in the red modification, and the results
were uncorrected by conversion into the black variety and estimation of the carbonaceous.
residue. In fact, every such conversion and correction was preceded by a weighing of the red
modification of the Sb,S,. The mean result of these weighings, if S=32, gave Sb=119.994.
The mean result of the corrected syntheses gave Sb=120.295. In these eleven experiments
the following percentages of S in Sb,S, were established:

28.57 28.50
28.60 ; 28.49
28.57 28.58
28.43 28.50
28.42 28.51
28.53

Mean, 28.5182 + .0120
Hence, Sb = 120.58.

These results, confirmatory of the work of Schneider, were presented to the American
Academy in 1876. Still, before publication, Cooke thought it best to repeat the work of
Dumas, in order to detect the cause of the old discrepancy between the values Sb=120 and
Sb=122. Accordingly, various samples of antimony trichloride were taken, and purified by
repeated distillations. The final distillate was further subjected to several recrystallizations
from the fused state; or, in one case, from a saturated solution in bisulphide of carbon. The
portions analyzed were dissolved in concentrated aqueous tartaric acid, and precipitated by
silver nitrate, many precautions being observed. The silver chloride was collected by reverse
filtration, and dried at temperatures from 110° to 120°. In one experiment the antimony was
first removed by HS: Seventeen experiments were made as follows. If we reduce to a com-
mon standard, Cooke’s analyses give, as proportional to-100 parts of AgCl, the quantities of
SbCl, stated in the third column:

SbCl. AgCl. Ratio.
1.5974 3.0124 53.028
1.2533 2.3620 53.061.

.8876 1.6754 52.978

8336 1.5674 53.184

5326 1.0021 53.148

7270 1.3691 53.101
1.2679 2.3883 53.088
1.9422 3.6646 52.999
1.7702 3.3384 53.025
2.5030 4.7184 53.048
2.1450 4.0410 53.081
1.7697 3.3281 53.175
2.3435 4.4157 53.072
1.3636 2.5813 53.020
1.8638 3.5146 53.030
2.0300 3.8282 53.028
2.4450 4.6086 53.053

Mean, 53.066 + .0096
Hence, Sb = 121.81.
This mean may be combined with that of Kessler’s series, as follows:

Kessler ons oyaa dead ea cena neaaenaoed Lee iectnwaned 53.623 +.015
WOOK 02 ene Soap dadseaunare i eauie sescanetatcohaierscesey . 53.066 + .0096

General Meats cc: sss coseeceeeee secs cecassieeeone s 53.2311 + .008
252 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

The results thus obtained with SbC13, confirmed Dumas’ determination of the atomic weight
of antimony as remarkably as the syntheses of Sb,S, had sustained the work of Schneider.
Evidently, in one or the other series a constant error must be hidden, and much time was spent
by Cooke in searching for it. It was eventually found that the chloride of antimony invari-
‘ably contained traces of oxychloride, an impurity which tended to increase the apparent atomic
weight of the metal under consideration. It was also found, in the course of the investigation,
that hydrochloric acid solutions of antimonious compounds oxidize in the air during boiling
as rapidly as ferrous compounds, a fact which explains the high values for antimony found
by Kessler.1
In order to render ‘“‘assurance doubly sure,” Cooke also undertook the analysis of the
bromide and the iodide of antimony. The bromide, SbBr,, was prepared by adding the finely
powdered metal to a solution of bromine in carbon disulphide. It was purified by repeated
distillation over pulverized antimony, and by several recrystallizations from bisulphide of car-
bon. The bromine determinations resemble those of chlorine. Reduced to a common standard,
the 15 analyses give the subjoined quantities of SbBr, proportional to 100 parts of silver bromide:

SOBr5. AgBr. Ratio.

* 1.8621 2.9216 63.736
-9856 1.5422 63.909
1.8650 2.9268 63.721
1.5330 2.4030 63.795
1.3689 2.1445 63.833.
1.2124 1.8991 63.841
9417 1.4749 63.848
2.5404 3.9755 63.901
1,5269 2.3905 63.874 '
1.8604 2.9180 63.756
_ 1.7298 2.7083 63.870
3.2838 5.1398 63.890
2.3589 3.6959 63.825
1.3323 2.0863 63.859
2.6974 4,2285 63.791

Mean, 63.830+- .008
Hence, Sb = 119.86.

The iodide of antimony was prepared like the bromide, and analyzed in the: same way.
At first discordant results were obtained, due to the presence of oxyiodide in the iodide studied.
The impurity, however, was removed by subliming the iodide in an atmosphere of dry carbon
dioxide. With this purer material seven estimations of iodine were made. Reduced to a
uniform standard, Cooke’s weighings give the following quantities of SbI, proportional to 100
parts of silver iodide:

Sb. Agl. Ratio.
1.1877 1.6727 71.005

.4610 .6497 70.956
3.2527 4.5716 71.150
1.8068 2.5389 71.165
1.5970 2.2456 71.117
2.3201 3.2645 71.071

.3496 4927 70.956

 

Mean, 71.060+ .023
Hence, Sb= 119.78.

Although Cooke’s work was practically conclusive, as between the rival values for anti-
mony, his results were severely criticized by Kessler,? who evidently had read Cooke’s paper
in a very careless way. On the other hand, Schneider published in Poggendorff’s Annalen
a friendly review of the new determinations, ‘which so well vindicated his own accuracy. In

 

1In Amer. Journ. Sei. (3), 21, 220, Cooke pointed out the'errors due to the solubility of silver chloride, and gave two series of analyses of SbCls
to illustrate their magnitude.
‘Berichte Deutsch. chem. Gesell. ,12, 1044. 1879.
No. 3.] ATOMIC WEIGHTS—CLARKE. 253

reply:to Kessler, Cooke undertook still another series of experiments with antimony bromide,’
and obtained absolute confirmation of his previous results. To a solution of antimony bromide
was added a solution containing a known weight of silver not quite sufficient to precipitate
all the bromine. The excess of the latter was estimated by titration with a normal silver
solution. Five analyses gave values for antimony ranging from, 119.98 to 120.02, when Ag=
108 and Br=80. Reduced to a common standard, the weights obtained gave the amounts of
SbBr, stated in the third column as proportional to 100 parts of silver:

SOBr;. Ag. Ratio.
2.5032 2.2528 111.115
2.0567 1.8509 111.119
2.6512 2.3860 111.115
3.3053 2.9749 111.106
2.7495 2.4745 111.113

 

Mean, 111.114+.0014
Hence, Sb=119.84.

Schneider,? also, in order to more fully answer Kessler’s objections, repeated his work
upon the Arnsberg stibnite. This he reduced in hydrogen as before, correcting scrupulously
for impurities. The following percentages of sulphur were found:

28.546

28.534
28.542

Mean, 28.541+ .0024

 

Hence, Sb = 120.42.
These figures confirm his old results, and may be fairly combined with them and with the
percentages found by Cooke, as follows:

Schneider, early series... ..-...--.-----.-22- eee eee eee ee 28.520 -+.008

Schneider, late series... .. 2.2.12... 2222s eee eee eee eee 28.541 +.0024

COOK Cis cisiee, osseks aioe se iaideiseeiessk Giese Deb eee lees uae 28.5182 + .0120
Gerieral mean. oo wasse ces ciated emeemnueucinels 28.5385 + .0023

In 1881 Pfeifer * determined electrolytically the direct ratios between silver and antimony,
and copper and antimony. With copper the following data were obtained:

Sd. Cu. 8Cu:28b::100:%.
1.412 1.1008 128.270
1.902 1.4832 128.236
3.367 2.6249 128.272

 

Mean, 128.259+.0077
Hence, Sb = 122.27.

With silver he found—

Sb. Ag. 8Ag:S8b::100:%
5.925 15.774 37.562
6.429 17.109 37.577

10.116 26.972 37.506
4.865 13.014 37.383
4.390 11.697 37.531
9.587 25.611 37.433
4.525 12.097 37.406

 

Mean, 37.485+.0198
Hence, Sb = 121.32.

The latter ratio was determined by Popper,‘ several years afterwards. The two metals
were precipitated simultaneously by the same current; and in some experiments two portions

 

1 Amer. Journ. Sci., May, 1880. Berichte, 13, 951. 3 Ann. Chem. Pharm., 209, 161.
2 Journ. prakt. Chem. (2), 22, 131. 4 Ann. Chem., 233, 153.

3
254 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

of antimony were thrown down against one of silver. These are indicated in the subjoined table
‘by suitable bracketing, and the ratio is given in the third column:

Sb. Ag. Ratio.
1.4856 ] 37.463
i areor 0505 { 37.292
2.0120 37.503
2.0074 | SedGse | 37.417
3.8882 37.480
3.8903 | ata { 37.500
4.1893 37.455
4.1885 } ae 37.447
4.2710 37.507
4.2759 | Ti eaes { 37.545
5.6860 37.460
5.6901 } 16.1788 | 37.487
4.4117 11.8014 37.383
4.9999 13.3965 37.322
5.2409 14.0679 37.250

 

Mean, 37.434+.0149
Pfeifer found, 37.485+.0198

General mean, 37.452+.0119

Popper’s figures give in mean Sb=121.15.

The recent investigations by Cohen and Strengers! seem to prove that these electrolytic
determinations are worthless. They effected the simultaneous precipitation of silver and
antimony, using solutions of SbCl,, and found that the apparent atomic weight of antimony
increased with the concentration of the solutions. They give the results of 24 determinations,
with full details, but only the end results need be cited here:

With 2.3 grm. SbCl, in a cc. of solution, So=120.84 to 120.87
” 83.3 Sb=121.81 to 121.92

These values are calculated with old values for Cl and Ag, but they show the failure of the
process to yield trustworthy figures. In any final discussion of the atomic weight of antimony,
therefore, the work of Pfeifer and Popper must be disregarded.

The work done by Bongartz? in 1883 was quite different from any of the determinations
which had preceded it. Carefully purified antimony was weighed as such, and then dissolved in
a concentrated solution of potassium sulphide. From this, after strong dilution, antimony
trisulphide was thrown down by means of dilute sulphuric acid. After thorough washing,
this sulphide was oxidized by hydrogen peroxide, by Classen’s method, and the sulphur in it
was weighed as barium sulphate. The ratio measured, therefore, was 2Sb : 3BaSO,, and the
data were as follows. The BaSO, equivalent to 100 parts of Sb is the ratio stated:

Sb taken. BaSO, found. Ratio.
1.4921 4.3325 290.362
.6132 1.7807 290.394
.5388 1.5655 290.553
1.2118 3.5205 290.518
.9570 2.7800 290.491
.6487 1.8855 290.349
-7280 2.1100 289.835
.9535 2.7655 290.036
1.0275 2.9800 290.024
.9635 2.7980 290.399
9255 2.6865 290.275
-7635 2.2175 290.438

 

Mean, 290.306 + .0436
Hence, Sb = 120.61.

1 Proc. Amsterdam Acad., Section of Sciences, 5 (2), 543. 1903. See also Cohen, Collins and Strengers, Zeitsch. phys. Chem., 50, 291.
2 Ber Doutsch. chem. Ges., 16, 1942. 1883.

 
No. 8.] 0 oa 1) MATOMIC: WEIGHTS—CLARKE. © ° 255

Still another method of determination was adopted by Friend and Smith.’ . Potassium
tartrylantimonite, KSbC,H,O,, was heated in a stream of dry, gaseous hydrochloric acid, and
so converted ‘into potassium chloride. The results obtained, with vacuum weights, are sub-
joined:

K8bC,H,0, KCl. Per cent KCl.
1.19481 0. 27539 23. 049
1. 57004 . 36186 23. 048
2.00912 . 46307 23. 048
2. 04253 .47073 23, 046
2. 16646 . 49935 23. 049
2. 25558 . 51982 23. 046
2. 61255 . 60215 23. 048
. 95272 . 68064 23. 051

: Mean, 23. 048+. 0006
“anon, Sb= 120.351.

We have now before us the following ratios, good and had, from which to calculate the
atomic weight of antimony. . The single. analyses by Weber and Unger, being unimportant,
are not included: :

(1) Percentage of S in Sb. ‘Se 28. 5385 0023.
(2) Percentage of Sb in Sb,0,, 79.283 +. 009.
(3) O needed to oxidize 100 parts SbCl,, 7.0294 + .0024.
(4) O needed to oxidize 100 parts Sb,O,, 10.953 + .0075.
(5) O needed to oxidize 100 parts Sb, 13.079 + .0096.
(6) K,Cr,0, : tartar emetic :: 100 : 337.30+.29..
(7) 3Ag : SbCl, :: 100 : 70.5124 .021.

- (8) 8AgCl : SbCl, :: 100 : 53.2311 + .008.
(9) 3Ag : SbBr, :: 100 : 111. 114+. 0014.

(10) 3AgBr : SbBr, :; 100 : 63.830 + .008.

(11) 3AgI : SbI, :: 100 : 71.060+ .023.

(12) 3Cu : 2Sb-:: 100 : 128,259 + .0077.

(13) 3Ag Sb 2: 100 : 37.452 + .0119.

(14) 2Sb : 3BaSO, : 100 : 290.306 + 0436.

(15) KSbC,H,0, “KCL : :: 100 : 23.048 + .0006.

To reduce these ratios we have—

Ag=107. 880 +. 00016 C = 12,0025+. 00019
Cl= 35. 4605+. 00011 K = 39. 0975+. 00011
Br= 79. 9228-£. 00015 Ba=137. 358 +. 0023
T=126. 926 +. 00023 Cr= 52. 01934. 0013

S= 32. 0650+. 00012 a Cu
H=1. 0078+. 00001

63. 555 +. 0005

Hence, ee
‘om ratio—
62 SepbeemtniehleS ti ieeiseattnsts tat ener ata Sb=118. 6714. 2844
“119. 7814. 1621
‘ae 119. 841-4. 0047

- 119. 856+. 0451
120. 351+: 0085
120. 437 +. 0105

. 120. 609+. 0182

— 121, 210-4. 0386
121. 244+.0777
121. 824+. 0680
122. 078+. 0100
122. 272+. 0074
122. 333+. 0898
122. 462+. 0550

: 122. 524+. 0345
General mean........-.-----.---2--- seen Sb=120. 687+. 0031

1Journ. Amer. Chem. Soc., 23, 502. 1901.

 

 

161566°—29——17
256 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

This mean has obviously very little significance except in so far as it shows the relatively
low weight attaching to the higher values. The latter, say all over 121, are almost certainly
in error, and ought to be rejected. Taking only the seven lowest values, they give a general
mean of Sb=120.062+.0037. Even this figure, however, is not quite satisfactory, for the
values derived from ratios 1 and 15, which seem to be good, are not adequately accounted for.
It is highly desirable that more work should be done upon the atomic weight of antimony by
modern methods, and for the purpose, in part at least, of explaining some of the evident dis-
crepancies which appear in the foregoing table.

BISMUTH.

Early in the last century an attempt was made by Lagerhjelm ‘ to determine the atomic
weight of bismuth, but the values he obtained were from four to seven units too high. They
therefore need no farther consideration here.

In 1851 the subject was taken up by Schneider,? who studied the oxidation of the metal
with nitric acid. The work was executed with a variety of experimental refinements by means
of which every error due to possible loss of material was carefully avoided. For full details
the original paper must be consulted; there is only room in these pages for the actual results,
as follows. The figures represent the percentages of Bi in Bi,O,:

89. 652 89. 656
89. 682 89. 666
' 89.644 89. 655
89, 634 89. 653

5 : Mean, 89. 6552+. 0034
Hence, Bi= 208.05.

Next in order are the determinations made by Dumas.*? Bismuth trichloride was pre-
pared by the action of dry chlorine upon bismuth, and repeatedly rectified by distillation over
bismuth powder. The product was weighed in a closed tube, dissolved in water, and pre-
cipitated with sodium carbonate. In the filtrate, after strongly acidulating with nitric acid,
the chlorine was precipitated by a known amount of silver. The figures in the third column
show the quantities of BiCl, proportional to 100 parts of silver:

BiCl,. Ag. Ratio.
3. 506 3. 545 98. 900
1.149 1. 168 98, 373
1. 5965 1. 629 98, 005
2.1767 2, 295 97. 829
3, 081 3.144 97. 996
2, 4158 2. 470 97. 806
1. 7107 1. 752 97. 643
3. 528 3. 6055 97. 712
5, 241 5. 361 97. 762

Mean, 98. 003+. 090
Hence, with Ag=108 and Cl=35.5, Bi= 211.03.

The first three of the foregoing experiments were made with slightly ‘tecotored material.
The remaining six percentages give a mean of 97.791, whence, on the same basis as before,
Bi=110.79. Evidently these results are now of no real wale, for it is probable that the chloride
of bismuth, like the corresponding antimony compound, contained traces of oxychloride. This
assumption fully accounts for the discordance between Dumas’ determination and the deter-
minations of Schneider and still more recent investigators.

In 1883 Marignac‘ took up the subject, attacking the problem by two methods. His
point of departure was commercial subnitrate of bismuth, which was purified by re-solution
and reprecipitation, and from which he prepared the oxide. First, bismuth trioxide was reduced

 

1 Annals of Philosophy, 4, 358, 1814. 2 Ann. Chim. Phys, (3), 55,176. 1859.
? Poggend, Annalen, 82, 303, 1851, 4 Arch. Sci. Phys, Nat. (3), 10,10. Oeuvres Complétes, 2, 71”,
No. 3.1 ATOMIC WEIGHTS—CLARKE. 257

by heating in hydrogen, beginning with a moderate temperature and closing the operation at
redness. The results were as follows, with the percentage of Bi in Bi,O, added:

Bi,0;. 0. : Per cent.
2. 6460 0. 2730 89. 683
6. 7057 . 6910 89. 696
3. 6649 - 3782 89. 681
5. 8024 - 5981 89. 692
5. 1205 . 5295 89. 658
5. 5640 . 5742 89. 680

Mean, 89. 682+. 0036

Hence, Bi= 208.60.

Marignac’s second method of determination was by conversion of the oxide into the sul-
phate. The oxide was dissolved in nitric acid, and then sulphuric acid was added in slight
excess from a graduated tube. The mass was evaporated to dryness with great care, and
finally heated over a direct flame until fumes of SO, no longer appeared. The third column
gives the sulphate formed from 100 parts of oxide:

Bi,0; ; BiS0,)3. Ratio.
2. 6503 4, 0218 161. 749
2. 8025 _. 4, 2535 161. 775
2. 710 4,112 151. 734
2. 813 4, 267 , 151. 688
2. 8750 4. 3625 161. 739
2. 7942 4, 2383 151. 682

 

; Mean, 151. 728+. 0099
Hence, Bi= 208.17.

This result needs to be studied in the light of Bailey’s observation, that bismuth sulphate
has a very narrow range of stability. It loses the last traces of free sulphuric acid at 405°, and
begins to decompose at 418°, so that the foregoing ratio is evidently uncertain. The concor-
dance of the data, however, is favorable to it.

Two analyses of bismuth sulphate, rather vaguely stated, are given by Bailey. The
weights found, and the ratio derived from them are as follows:

Bix(SO,)s- ‘Bi,0;. Ratio.
2. 2155 1. 4615 151. 591
1. 5635 1. 0267 152. 284

 

Mean, 151. 937+. 231

Hence, Bi= 207.24. Combined with Marignac’s series, the general mean becomes 151.729
+.0099. Bailey’s figures practically disappear.

The next determination of this atomic weight was by Léwe,? who oxidized the metal with
nitric acid, and reduced the nitrate to oxide by ignition. Special care was taken to prepare
bismuth free from arsenic, and the oxide was fused before weighing. In the paper just quoted
Bailey calls attention to the volatility of bismuth oxide, which doubtless accounts for the low
results found in this investigation. The data are as follows:

Bi taken. Bi,0, found. Per cent Bi.
11. 309 12. 616 89. 640
12. 2776 13. 694 89. 656

 

Mean, 89. 648+. 0040
Hence, Bi= 207.84.
In Classen’s * work upon the atomic weight of bismuth, the metal itself was first carefully
investigated. Commercial samples, even those which purported to be pure, were found to be
contaminated with lead and other impurities, and these were not entirely removable by many

1 Journ. Chem. Soc., 51, 676. 1887. Bailey deduces from his analyses Bi=208.33 and 208.43. There may be some error in his printed figures,
for his deductions do not agree with the data as given.

2 Zeit. anal. Chem., 22, 498.

8 Ber. Deutsch. chem. Ges., 23, 938. 1890.
258 MEMOIRS. NATIONAL ACADEMY OF SCIENCES. [Vor. XVI

successive precipitations as subnitrate. Finally, pure bismuth was obtained by an electro-
lytic process, and this was converted into oxide by means of nitric acid and subsequent ignition
to incipient fusion. Results as follows, with the percentage of Bi in Bi,O, added:

Bi taken, Bi,0, found. Per cent Bi.
25. 0667 27. 9442 89. 703
21. 0691 23, 4875 89. 7035
27. 2596 30. 3922 89. 693
36. 5195 40. 7131 89. 700
27.9214 31. 1295 89. 6944
$2. 1188 35. 8103 89. 692
30. 1000 33. 5587 89. 694
26. 4825 59.5257 89. 693
19. 8008 22, 0758 89. 695

9 i ‘Mean, 89. 696+. 0009

Hence, Bi= 208.92, or, reduced to a vacuum, 208.90. mes

Classen’s paper was followed by a long controversy between Schneider and Classen,’ in
which the former upheld the essential accuracy of the work done by Marignac and himself.
Schneider had started out with commercial bismuth, and Classen found that the commercial
bismuth which he met with was impure. Schneider, by various analyses, showed that other
samples of bismuth were so nearly pure that the common modes of purification were adequate;
but Classen replied that the original sample used by Schneider in his atomic weight investiga-
tion had not been reexamined. Acéordingly, Schneider published a new series of determina-
a made by the old method, but with metal which had been scrupulously purified. Results
as follows:

Bi. Bi,O;. Per cent Bi.
5. 0092 5, 5868 89. 661:
3. 6770 4.1016 89. 648
7, 2493 8. 0854 89. 659
9. 2479 10. 3142 89. 662
6. 0945 6.7979 89. 653

12, 1588 13. 5610 89. 660

Mean, 89. 657+. 0015

Hence, with O=16, Bi=208.05, a confirmation of the earlier determinations.

According to Adie* the differences between the low and high values for bismuth are due
to the presence of silicon in the metal. A preliminary determination of the atomic weight,
made with pure bismuth, gave Bi=208.8, approximately. Adie’s explanation of the discrep-
ancies remains to be substantiated by others.

- Birckenbach,! working under the direction of Gutbier, effected the synthesis of bismuth
oxide, and also studied its reduction. Bismuth from three distinct sources was employed in
the investigation. First, the metal was converted into nitrate, and then calcined to oxide,
which latter was proved to be free from’ occluded gases. The data obtained were as follows:

Preliminary Series. :
Bi. , Bi,0;. Per cent Bi.

10. 2899 11. 4782 89. 647
8. 1023 9. 0372 89. 655

Mean, 89. 651+. 0027
Hence, BI = 207.905.

Final Series.
Bi. ——-Bt,O,. : _ Per cent Bi,
9. 63289 10. 74328 ' 89. 664
10. 41101 Tl. 61288 . 89. 651
"10. 97914 os "12, 24528 89. 661

 

1 Journ. prakt. Chem., 42, 563; 43, 133; 44, 23, and 411.”

3 Journ. prakt. Chem., 50, 461. 1804.

3 Proc. Cambridge Phil. Soc., 12, 240. 1903.

4Inaug. Diss., Erlangen, 1905. The oxidation series also appears under the authorship of Gutbier and Birckenbach, in Journ. prakt. Chem.

(2), 47, 457. 1908,
No. 8.) ATOMIC WEIGHTS—CLARKE
Bi. Bi,0,. | Per cent Bi.
10. 11990 11, 28800 89. 653
18. 96770 21, 15541 89. 659
11. 99601 18. 38001 89. 654
27, 23022 30. 37392 89. 651
24, 98170 27. 86431 89. 655
10. 11284 11. 27998 89. 653
28. 35991 31. 63053 89. 660

Hence, Bi= 208.02.

The reduction of bismuth oxide to bismuth gave Birckenbach the following results:

Bi0,.
2, 43105
2, 9547
1. 65199
2, 4108

Hence, Bi= 208.11.

Bi,O,.
1. 45827
2. 12432
3. 0021
2.1012
3. 0182
1. 9091 -

Hence, Bi= 208.153.

Preliminary Series.
Bi,
2.17994
2. 6488

1, 4810
2. 1609

Final Series.

Bi.
1. 30751
1, 90461
2. 6918
1. 8840
2. 70620
1.71171

Mean, 89. 656+. 0010

Per cent Bi.
89. 671
89. 647
89. 671
89. 653
Mean, 89. 660+. 0041

Per cent Bi,
89. 662
89. 657
89. 664
89. 663
~ 89. 663
89. 661
Mean, 89. 662+. 0007

The data for the percentage of Bi in Bi,O, now combine as follows:

Schneider, 185] ascescasccessteae connaetassGnered + eoaae 89. 655+. 0034
Marignacy 5..0.ijgisig nesters sactutiee scene ceaeeuaa eae gins 89. 682+. 0036
LOW sisal weeds c 3 telacetnajac ative Mecreicttgaes toenane See ee 89. 648+. 0040
Classen sae2ede 2 soastcex cede e's o's medals ticee e's 89. 696+. 0009
Schneider, 1894................-------- estinereeuwave ses 89. 657+. 0015
Birckenbach, preliminary oxidations................-.--- 89. 651+. 0027
Birckenbach, final oxidations............--.------------ 89. 656+. 0010
Birckenbach, preliminary reductions.........-....------ 89. 660+. 0041

Birckenbach, final reductions

General mean

sca te 89. 662+. 0007
Sees 89. 6683 +. 00044

259

If we omit the high value found by Classen, the general mean becomes 89.6594 + .00052.

Mehler,! also under Gutbier’s direction, studied the composition of bismuth tribromide,
which was prepared by direct union of the metal with bromine, and afterwards sublimed. The
bromine was precipitated with silver solution, and the silver bromide was weighed. The weights
are given as reduced to a vacuum. © In the third column I give the ratio 3AgBr : BiBr, : : 100 : 2:

BiBr,.
3. 77071
4, 37676
3. 64088
4, 57894.
4, 53204
2. 85054
4, 58310
6. 47910

Hence, Bi= 207.96.

AgBr.

4, 74323.

6. 50932

4.58160 ©
5. 76183
5. 70410
8. 58682
5. 76618
8. 15465

Ratio. -
79. 497
79. 443
79. 467
79. 470
79, 452.
79.473
_ 79, 482
79, 458
Mean, 79. 467+. 0042

1 Inaug. Diss., Erlangen, 1905. Sitzungsb. phys. med. Soz. Erlangen, 37, 343.
260 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Another research, carried out under Gutbier by Janssen,! involved the synthesis of bismuth
sulphate. Bismuth was first dissolved in nitric’ acid, and then, with sulphuric acid, converted
into sulphate. The latter compound was freed from moisture and excess of acid by heating to
380°, at which temperature its weight was constant. The results obtained were as follows:

Bi. Bi(80,)3. Per cent Bi.
2. 4045 4. 0706 ' 89.070
2. 41900 4. 09445 59. 081
2. 20280 3. 72745 59. 096
2. 57206 4, 35444 59. 066
5. 79241 3. 79987 ’ 59. 106 oy
3. 65233 6. 18143 59. 086

Mean, 59. 084-4. 0042
Hence, Bi= 208.081.
The recent determinations of the atomic weight of bismuth by Oechsner de Coninck and
Gérard,? are based upon the reduction of the trichloride to metal by heating in a stream of
hydrogen. Their data are as follows:

Bicl;. Bi, Per cent Bi.
0. 6100 -0. 4038 66. 197

- 5490 . 3634 66. 193

. 4880 . 3230 66. 188

. 4930 . 3263 66. 187

Mean, 66..191+. 0017
Hence, Bi= 208.273.
Rejecting the work of Dumas, the following ratios are now available for discussion:

(1) Bi,O, : 2Bi :: 100 : 89.6683 +.00044.

(2) Bi,(SO,), : 2Bi :: 100 : 59,084 +.0042.
(3) Bi,O, : Bi,(SO,), :: 100 : 151.729 +.0099.
(4) BiCl, : Bi :: 100 : 66.191 +.0017.

(5) 3AgBr : BiBr, :: 100 : 79.467 + .0042.

To reduce these ratios we have—

Ag=107. 880+. 00016 Br=79, 9228+. 00015
Cl =35. 4605 +. 00011 S=32. 065 +. 00012
Hence,
From ratio—

Di ssinseutaues seis se geumma sethantes eeeeneweaeeescd s+ Bi=207. 956+. 0238
2 sei isa smi s 1 kaaomnenirneacedamnaaeaee + assess 208, 0814. 0260
Soveveseederes as ve sesddaanmee Settee mene ¢ senses. 208, 1174. 0444
As, Sownesuais be Sansa GSadiemheeiuncuekGeiesbiecsees ‘22. 208. 273+. 0118
Digs crs ee ts cea ete Serchlit teh nedechacte tad ALE . 208. 295 +. 0095
General Meah.sid2cccateeece sees eos sees Bi=206. 240+. 0065.

This value is probably too high, mainly because of Classen’s determinations. Rejecting

them the general mean becomes
Bi= 208.063 + .0071

siliiah value is to be accepted. It is also sustained: by. Brauner’s* statement that Kuzma, by
syntheses of bismuth sulphate from bismuth oxide, has obtained the value Bi=208.0+.1. The
details of Kuzma’s work are yet to be published.

. COLUMBIUM.’*

The atomic weight of this metal has been determined by several investigators. Rose §
analyzed a compound which he supposed to be chloride, but which, according to Rammelsberg,’
must have been nearly pure oxychloride. If it was chloride, ‘tien the widely varying results
give approximately Ch=122; if it was oxychloride, the value becomes nearly 94. If it was
chloride, it was doubtless contaminated with tantalum compounds.

 

1 Inaug. Diss., Erlangen, 1906. 4 This name has 40 years’ priority over “niobium,” and therefore deserves preference.
2 Compt. Rend., 162, 252. 1916. 6 Poggend, Annal., 104, 439. 1858.
8 In Abege’s “ Handbuch,”’ 3 (3), 634. 6 Poggend. Annal., 136, 353. 1869.
No. 3.) ATOMIC WEIGHTS—CLARKE. 261

Hermann’s ' results seem to have no present’ value, and Bloomstrand’s? are very discordant.
He analyzed columbium pentachloride and sodium columbate, but his compounds could not have
been pure. They evidently contained some ee for the atomic ngere deduced from his
analyses are several units too high.

Marignac*® made about 20 analyses of iockaaaiuia intaeehanniuth CbOF,.2KF.H,0.
One hundred parts of this salt give the following percentages:

ChjOpiiasiessds cectincedcc ap saree Extremes 44. 15 to 44. 60 Mean, 44. 36
FSO joes. sede owdagaed ete oe a 57. 60 to 58, 05. '
HO, tee rabeseremcesevcudapee « 5.75 to. 5.98.

HR san eapececawewve icovstoadeee 9 “sy “SO.02 32,20"

From the mean percentage of Cb,0,, Cb = 93.418.

From the mean between the extremes given for K,SO,, Cb = 93.95.

The recent determinations by Balke and Smith* are much more satisfactory than those
already cited. Their material was certainly purer, and the results obtained were highly con-
cordant. Columbium pentachloride was decomposed by water, with the aid of a little nitric
acid, and the oxide so produced was finally ignited and weighed. Their data, with vacuum
weights, are as follows: ,

CbCI,. Cb,0;. Pr. cent Cb,05.
9. 56379 4, 71539 49, 305
5, 42742 2. 65730 49, 292
5, 15992 2, 54364 49, 296
9. 64854 4. 75641 49, 297
7, 24572 3, 57222 49, 301
8. 00559 3. 94746 49, 309
9. 60763 4, 73852 49, 324
9. 19732 4, 53638 49, 323
4. 27456 2. 10734 49, 300

 

Mean, 49. 305. 0026
Hence, with Cl=35.4605 + .00011, Cb = 93.538 + .0139.

The determinations of the atomic weight of columbium by Smith and Van Haagen * were
made by heating sodium columbate in the vapor of sulphur monochloride. The columbium
was volatilized as chloride or oxychloride, and sodium chloride remained behind. The data,
with vacuum weights, are as follows: .

NaCbo;. .. NaCl. Pr. cent NaCt.
0. 36419 0. 12975 ‘ 35. 627
. 69113 24617 . 35, 618
1. 04904 . 37368. . 85, 621
1. 64337 . 58516 _ 35. 607
1. 33367 . 47487 35. 606
- 88051 . 31366 35. 623

1. 29947 46289 35, 622
, Mean, 35. 618-4. 0021

 

Hence, with Na =23.0029 + 00012, Cb = 93.137 + .0097.
Combining this value with that derived from the chloride, the general mean for columbium

becomes
Cb = 93.277 + .0078

This final mean is not altogether satisfactory. As Smith and Van Haagen, point out,
columbium pentachloride is likely to contain some oxychloride, which tends to increase the

 

1 Journ. prakt. Chem., 68, 73. 1856.

2 Acta Univ. Lund, 1864.

3 Arch. Sci. Phys. Nat. (2), 23. 1865. Oeuvres Complates, 2, 259.
«Journ. Amer. Chem. Soc., 30, 1644. 1908.

§ Journ. Amer. Chem. Soc., 37, 1783. 1915.
262 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

atomic weight derived from it.. On'‘the other hand it is not absolutely certain that the sodium
columbate did not contain traces of water, not entirely. removable by ordinary drying. That
error, if it existed, would lead to an atomic weight slightly too low.: In the final combination
of the two values these possible errors should tend to compensation. More evidence is needed
before the atomic weight of columbium can be asoutataly: fixed.

TANTALUM.

The determinations of the atomic weight of tantalum by Berzelius,1 Rose,? Hermann,’ and
Blomstrand ¢ may be fairly left out of account as valueless. These chemists could not have
worked with pure preparations, and their results are sufficiently summed up in Becker’s “‘ Digest, ui
and also in the previous editions of this work. - GIN 14

In 1866 Marignac’s determinations appeared. 5 He made four analyses of. a. ure potaseium
fluotantalate, and four more.experiments upon the ammonium salt. The potassium compound,
K,TaF,, was treated with sulphuric acid, and the mixture was then evaporated to dryness.
The potassium sulphate was next dissolved out by water, while the residue was ignited and
weighed as Ta,O0;. One hundred parts of the salt gave the following quantities of anes and
K,SO, :

 

Ta05. K,80,.
56. 50 44. 37
56. 75 44. 35
56. 55 44, 22
56. 56 ‘ 44. 24
Mean, 56. 59:4. 037 a Mean, 44. 295+. 026
From these figures, 100 parts of K,SO, eGrenddad to the subjoined quantities of Ta,O,:
127. 338 ; 128. 178
127. 960 127. 848

Mean, 127. 8314. 120
The ammonium salt ‘(NH,),TaF,, ignited with sulphuric acid,. gave these percentages of
Ta,O;. . The figures: are corrected for a trace of K,SO, which was always pee

63..08 ~ : 63, 27

63.24 2. ° 63. 42
Mean 63. 25+. i

Hence we have four values for Ta:

From potassium salt, per cent Ta,O,......-.--.-..-------2-0-- Ta=183. 55
From potassium salt, per cent K,SO,.-......--.-..----------5 ‘ =182. 93
From potassium salt, K,SO, : Ba, Og csr cictinntidoe x's wie eee S25 182. 76
From ammonium salt, per cent T'a,O0,.......-.--------e-0----- {f = 182. 66

Avy tite cccpesc see lawn ieeebuce Ph dws Reals Hate emer ’ Ta=182. 975

The determinations ‘by Hinrichsen and Sahlbom ° were much simpler. Metallic tantalum
was converted into pentoxide by heating in oxygen, and the composition of the oxide was SO
ascertained. The weights and percentages of tantalum are as follows:

Pe

Ta. Ta,05. Pr. cent Ta. :
0. 37200 _, 0.45437 81. 872 7
. 41278 ” , 50364 81. 959
. 33558 . 40975 81. 899
. 85883 «43807... 81.912,
. 47554 . 58087 81.868 ..

, Mean, 81. 902+: 0111
Hence, Ta= 181.019 + .0857..

1 Poggend. Annalen, 4,14. 1825. 4 Acta Univ. Lund, 1864. ot We
2 Poggend. Annalen, 99, 80. 1856. § Arch. Sci .Phys. Nat. (2), 26, 89. 1866. Gatien Completes, 2, 314.
3 Journ. prakt. Chem., 70,193. 1857. 6 Bev. Deutsch. Chem. Ges., 39, 2600, 1906.

 
No. 8.] ‘ss ATOMIC WEIGHTS—CLARKE. 263

This mean value has very little significance, because the individual determinations give
values for the atomic weight of tantalum ranging from 180.60 to 181.72; a difference of 1.12
units.

The determinations by Van Haagen ' are confessedly soptoxirante; because the material
studied may not have been absolutely pure: Tantalum pentabromide was hydrolyzed to oxide,
giving the ratio TaCl, : Ta,O, :: 100: 2. The bromide, however, may have contained traces
either of an oxybromide or, as Van Haagen supposes, of an addition product of TaBr, and
HBr. Two series of determinations are given, but without correction to a vacuum. ‘The data
are as follows; both series taken as one.

TaBr,, Ta,0;. Per cent Ta,0,.
1. 11049 0, 42528 38, 297

. 68125 . 24187 38. 316
1, 64424 : . 62951 38. 286
1.545381 . 59209 38, 315

. 25094 . 09616 38. 320

2. 13436 . 81737 38. 296

1. 08818 . 41678 38. 301

2. 21255 . 84779 38. 317-

 

: Mean, 38. 306+. 0044
Hence Ta = 183.284 +. 0871.

Two years later, also in the laboratory of the University of Pennsylvania, Chapin and
Smith ? made a series of analyses of tantalum pentabromide, by: the method of hydrolysis.
Their data, with vacuum weights, are as follows

oy Sg TaBr,. Ta,0;. Per cent Ta,0,..
3 0. 86837 0. 33117 38. 137
1. 50903 . 57570 38. 150
1. 56554 . 59718 38, 145
1. 23239 . 47080 38. 162
1. 31815 . 50295 38, 156
1. 31702 . 50244 38. 150
1. 20090 . 45830 38, 163
‘1. 04050 . 39688 38. 143

 

Mean, 38. 151+. 0022
Hence Ta = 181. 8204 0414.
Similar hydrolyses of tantalum pentachloride were effected by Balke.* Hig data, with

vacuum weights, are as follows:

TaCl,. , Ta,05, Per cent ens: es ak

12. 99680 8. 02326 61. 733 & 44 ta
9. 24957 5. 71104 61. 744
10. 17456 6. 28133 61. 736
17. 99542 11. 11014 “61. 739
11. 70588 7. 22693 , - 61739
6, 24767 8, 85658 61. 728
. 7.26375 4, 48398 61. 731
15. 88270 9. 80465 61. 732

 

Mean, a. 735:t. 0012
Hence Ta'= 181.524 + .0225.

Sears and Balke ‘ first studied the ici of tantalum pentachloride, and pte two
important sources of error in the process. Some nitric acid was retained by the tantalic oxide,
and some of. the. latter was lost during ignition. They therefore abandoned that method as
unsuited to the accurate determination of the atomic weight of tantalum, and resorted to a
study, which they term preliminary, of the ratio between the chloride and silver. Three sets
of eee are given, with vacuum weights, as follows: A Boe atk se, fas

 

7 1 Thesis, ‘Uhiversity of Pentisylvania, 1909. en * Journ. Amer. Chem, Soc., 32, 1127. 1910.
a Journ. Amer. Chem. Soc., 33, 1497, 1911. / 4 Journ. Amer. Chem. Soc., 37, 833. 1915.
264 MEMOIRS ‘NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

 

ae Series. 1 'e .
TaCl,. | ~ AE Ag. ‘Ratio.
1L 66795 17. 51344 66. 622
7. 61398 coe. +11, 42585 66. 638
os 095 yy 8, 15647. 66.518
8. 09039 : 12. 16628 66. 498
tes yon 66. 569+. 0252
Hence Ta=181.771. es dae
Serves 2. a. ana
TaCl;. Ag. Ratio.
3. 53605 5. 31834 66. 488
2. 99083 4, 49851 66. 485
6. 21118 9. 34345 : 66. 476
‘ Mean; 66. 483+. 0031
Hence Ta = 181.307. res
Series 8.
TaCi,. Ag. Ratio.
4, 08892 6. 07417 66. 493
3. 32164 5. 00133 66. 417
2. 68304 4. 04002 “66. 412
3. 01453 4, 53953 66. 406
3. 95846 . ©. 95931 66. 425

Mean, 66. 4314-. 0107 ww

Hence Ta=181.027. The three series combined give a general Mean value for the ratio
of 5Ag : TaCl, : : 100 : 66.479 +.0029.

Hence—
Ta = 181.290 +.0157,

In this combination series 2 is evidently much overvalued. Series 3 is underrated because
the first determination in it varies much from the others. These considerations, however, carry
little weight, for. the method of determination, as Sears ' has shown, is subject to serious errors.
Indeed, according to Sears, tantalum pentachloride is not a epuewonthy compound to use in
atomic weight determinations.
~:1v In the present state of our knowledge it is not: necessary to each the various ‘datarmine:
tions of the atomic weight of tantalum into a general mean. The following table,, however,
may be useful as a comparison of the results so far obtained. In it the determinations by
Marignac are arbitrarily vassigned equal eee wath those by Van Haagen.

 

ioriach.<cecv eae aoe eRe Ta=182. ‘975+. 0871
Hinrichsen and Sahlbom 181. 019+. 0857
Van Haagen...........-.....- “| “ 183. 284-4. 0871
Chapin and Smith............. fae Spa Sideedimiapecheniseemtoeeia 181. 8224. 0414
Bale pss sccctoie.ttetooeictote «ays eaten aneeeroaras Se AIG 181. 524+. 0225
Sears and Balke: cjlin cso oecreecue raisiialeipiavearere ilsewiees 181. 290+. 0157

From these figures no final conclusion can be drawn. So far as present evidence goes, it
seems probable that the atomic weight of tantalum is somewhere near 181, with an Unrortaynhy

of at least a a unit.
ce CHROMIUM.

+ fp Cistenitig tie atomic weight of chromium there has been much discussion, and many
experimenters have sought to establish the true value, The earliest work upon it having any
importance was that of Berzelius,? in 1818 and 1826, which led to results much in excess of the
correct figure. His method consisted in precipitating a known weight of lead nitrate with an
alkaline chromate and weighing the lead chromate thus produced. The error in his determina-

 

1 Journ. Amer. Chem. Soc., 39, 1852. 1917, 2 Schweigg. Journ., 22, 53, and Poggend. Annal., 8, 22.
No. 8.] - ATOMIC WEIGHTS—CLARKE. ©. 265

tion arose from the fact that lead chromate, except when thrown down from very dilute solu-
tions, carries with it minute quantities of alkaline salts, and so has its apparent weight notably
increased. When dilute solutions are used, a trace of the precipitate remains dissolved, ent
the weight obtained is too low. In neither case is the method trustworthy.
«Jn 1844 Berzelius’ results were first seriously called in question. The figure for secant
deduced from his experiments was somewhat over 56; but Peligott now showed, by ‘his analyses
of chromous acetate and of the chlorides of chromfam, that’ the true number was near 52.5.
Unfortunately, Peligot’s work, although:good, was published with insufficient details to be useful
here. For chromous acetate he gives the percentages of carbon and hydrogen, but not the
actual weights of.salt, carbon dioxide, and water from which they were calculated. ‘His-figures
vary considerably, moreover—enough to show that their mean would carry but laetle, nusightt
when combined with the more explicit data furnished by other chemists. tof
Jacquelain’s? work we may omit entirely. He gives an atomic weight for Siecsontiih which
is notoriously too low (50.1), and prints none of the numerical details upon which his result
rests. The determinations by Moberg,? Lefort,* and Wildenstein,® which give values for chro-
mium in excess of Cr=53, may also be rejected, and given no further consideration. The
first researches to command our attention are those by Berlin,* His starting point was normal
silver chromate; , but in one experiment the. dichromate Ag,Cr,0, was used. These salts, which
are easily obtained. i in.a.pure condition, were reduced i in a large flask by: means of hydrochloric
acid and. alcohol. . The chloride of silver thus formed was washed by decantation, dried, fused,
and weighed without transfer. The united washings were supersaturated with ammonia, evapo-
rated to dryness, and the residue treated with hot water. The resulting chromic oxide was
then collected upon a filter, dried, ignited, and weighed. The results were as follows:

4. 6680 grm. Ag,CrO, gave 4.027 grm. AgCl and 1. 0754 grm. Cr,Ox,.

3. 4568 i 2. 983 a .7980
2. 5060 a 2. 1605 is sr
2.1530 > « 1. 8555 ie 494500
4.3335 grm. Ag,Cr0, gave 2. 8692 “ 1.5300“

From these weighings three values are calculable for the atomic weight of chromium. The
three ratios upon which these values depend we will consider separately, taking first that between
the chromic oxide and the original silver salt. In the four analyses of the normal chromate

the percentages of Cr,05 deducible from Berlin’s weighings are as follows:

1

23,037
_ 23.027

Vig : ie ae! _ Mean, 23.0144. 011 °

Hence, r= 52. 46.

23. 025
22, 968

And from the single experiment with. ‘Ag,Crs 0, the percentage of 010, was 35. 306. Hence,

Cr=52.34,00 00 rt
pr the-ratio Eeeeen Ag,CrO, and AgCl, putting the latter at 100, we have for the. former:
Be Pens Babs ty 115.917 115. 992 i ae
ee $ 115. 883 ae . 116. 033 eutal
Mean, | 115. 956+. 023 - hin Bis ;

Hence, Cr= 52. 66.

In the single experiment with dichromate 100 AgCl is formed from 151, 035 Ag,Cr, O;.

Hence, Cr= 52.61.

Finally, for the ‘ratio between Ag(Cl and Cr,0,, the five experiments of Berlin give, for 100

parts | of the loner the following quantities of the latter:

26. 705 26. 650 rn
= 26. 685 26. 662 :
eo ee 26. 707

rit Mean, 26. 682-4. 0076

Hee Cr= 52, 50.

OE a : al

 

+ .1 Compt. Rend.,.19, 609, and 734; 20, 1187; 21, 74.

2 Compt. Rend., 24, 679. 1847.
8 Journ. prakt. Chem., 43, 114.

. 4Journ. prakt. Chem., 50;.261. 1850.

6 Journ. prakt. Chem., 59, 27.
6 Journ. prakt. Chem., 37, 509, and 38, 149. 1846.
266 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

Next in order are two papers by Kessler, who employed a peculiar volumetric. method

_ entirely his own. In brief, he compared the oxidizing power of potassium dichromate with
‘that of the chlorate, and from his observations deduced the ratio between the molecular weight
of the two salts.

In his earlier. paper? hs mode of procedure was about as fellows: The two salia, weighed
out'in quantities having approximate chemical equivalency, were placed in two small flasks,
and to each was added 100 ce. of.a ferrous chloride solution and 30 cc. hydrochloric acid. The
ferrous chloride was added in trifling excess, and, when action ceased, the amount unoxidized
was determined by titration with a standard solution of dichromate. As in each case the
quantity of ferrous chloride was the same, it became easy to deduce from the data thus obtained
the ratio in question. I have reduced all of his somewhat complicated figures to a simple com-
mon standard, and give below the amount of chromate equivalent to 100 parts of claret:

120. 118 120. 096
120. 371 120. 241
120. 188 120. 181

Mean, 120. 191+. 028
Hence, Cr=52.21.

In his later paper? Kessler substituted arsenic trioxide for the iron solution. In one series
of experiments the quantity of dichromate needed to oxidize 100 parts of the arsenic trioxide
was determined, and in another the latter substance was similarly compared with the chlorate.
The subjoined columns give the quantity of each salt proportional to 100 parts of As,0,:

K,Cr,0,. KCIO0;.
98. 95 41. 156
98. 94 41. 116
99. 17 41. 200
98. 98 4]. 255
99. 08 41. 201
99. 15 41. 086

Mean, 99. 045+. 028 7 a
41.161
41. 193
41.149

41. 126
Mean, 41. 172. 009

 

Hence, Cr= 52.32.
Reducing the later series to the standard of the earlier, the two combine as follows:
(1) 2KCIO, : K,Cr,O, :: 100 : 120.191 +.028
(2) 2KCI0, : K,Cr; 0,: 100 : 120.282 + .043-
General mean, 120.216 + .0235
‘Siewert’s determinations, which do not seem to have attracted general attention, were
published in 1861.2 He, reviewing Berlin’s work, found that upon reducing silver chromate
with hydrochloric acid and alcohol, the chromic chloride solution always retained traces of silver
chloride dissolved in it. These could be precipitated by dilution with water; but, in Berlin’s
process, they naturally came down with the chromium hydroxide, making the weight of the
latter too high; hence too large a value for the atomic weight of chromium. In order to find
& more correct value Siewert resorted to the analysis of sublimed, violet, chromic chloride. This
salt he fused with sodium carbonate and a little nitre, treated the fused mass with water, and
precipitated from the resulting solution the chlorine by silver nitrate in presence of nitric ‘acid.
The weight of the silver chloride thus obtained, estimated after the usual manner, gave means
for calculating the atomic weight of chromium. His figures, reduced to a common standard,
give, as proportional to 100 parts of chloride of silver, the quantities of chromic enlence stated
in the third of the subjoined columns:

 

1 Poggend. Annalen, 95, 208. 1855. ‘ 2 Poggend. Annalen, 113, 137. 1861. § Zeit. gesammt. Wissenschaften, 17, 530.
No. 3] ATOMIC WEIGHTS—CLARKE 267

CrCl. AgCl. Ratio.
0. 2367 0. 6396 37. 007
. 2946 . 7994 36. 853
. 2593 . 7039 36. 838
- 4935 1. 3395 36. 842
. 5850 1. 5884 36. 830
. 6511 1. 76681 36. 852
. 5503 1, 49391 36. 836

 

Mean, 36. 865+. 0158

The first of these figures varies so widely from the others that we are justified in rejecting
it, in which case the mean becomes 36.842+.0031. Hence, Cr=52.048.

‘Siewert also made two analyses of silver dichromate by the following process. The salt,
dried at 120°, was dissolved in nitric acid. The silver was then thrown down by hydrochloric
acid, and, in the filtrate, chromium hydroxide was precipitated by ammonia. Reduced to a
uniform standard, we find from his results, corresponding to 100 parts of AgCl, Ag,Cr, 0, as in
the last column:

 

Ag,Cr.07. AgCl. Cr,03. Ratio.
0. 7866 0. 52202 0. 2764 150. 684
1. 089 . 72249 . 3840 150. 729

Mean, 150. 706

Hence, Cr=52.14.

Berlin’s single determination of this ratio gave 151.035. Taking all three values together
as one series, they give a mean of 150.816 +.074.

Siewert’s percentages of Cr,O, obtained from Ag,Cr,O, are as follows, calculated from the

above weighings:
35. 139

35. 262
Mean, 35. 2005-4. 0415

 

Hence, Cr=51.983.
- Combining, as before, with Berlin’s single result, giving the latter equal weight with one of
these, we have a general mean of 35.236 +.0335.
For the ratio between silver chloride and chromic oxide, Siewert’s two analyses of the
dichromate give as follows. For 100 parts of AgCl we have of Cr,O,:

52. 948
53. 150

Mean, 53. 049+. 068

Hence, Cr = 52.041.

This figure, reduced to the standard of Berlin’s work on the monochromate, becomes
26.525+.034. Berlin’s mean was 26.682+.0076. The two means, combined, give a general
mean of 26.676 +.074.

By Baubigny! we have only three experiments upon the calcination of anhydrous chromic
sulphate, as follows:

Cr,(SO,)3- Cr,03. Per cent.
1. 989 0. 7715 38. 788
3. 958 1. 535 38. 782
2. 6052 1. 0115 38. 826

 

; Mean, 38. 799+. 0092
Hence, Cr = 52.14.

In Rawson’s work? ammonium dichromate was the substance studied. Weighed quan-
tities of this salt were dissolved in water, and then reduced by hydrochloric acid and alcohol.
After evaporation to dryness the mass was treated with water and ammonia, reevaporated,

 

1 Compt. Rend., 98, 146. 2 Journ. Chem. Soc., 55, 213.
268 MEMOIRS NATIONAL. ACADEMY ' OF SCIENCES. [Vou XVI.

dried five hours at 140°, and finally ignited in a muffle. The residual chromic oxide was bright
green, and was tested to verify its purity. The corrected weights are as ‘follows:

Ars0r,07. Cr03. Per cent Cr203.
1. 01275 0. 61184 60. 365
1. 08181 . 65266 60. 330
1, 29430 . 78090. 60. 334
1,,13966 . 68799 60. 368
. 98778 - 59595 60. 332
1. 14319 . 68987 60. 346

 

: aa 60. 346 +. 0046
Habe Cr= 52.15.
_-{ Stilllater we come to-the elaborate Aotaeninabons of the ipa weight of chromium ee
by. Meineke,! who studied the chromate and ammonio-chromate of silver, and also the. dichro-
mates of potassium and ammonium. For the latter salt he measured the same ratio that Raw-
son determined, but by a different method. He precipitated its solution with mercurous nitrate,

and ignited the precipitate, with the subjoined results. Vacuum weights are given:

Am,Cr.0,. Cr.03. ‘tats Per cent Cr,0,.
2. 0416 : 1. 2316 60. 325
2.1618 1. 3040 60. 320
2. 0823 1, 2562 60. 328
2.1913 1. 3221* 60. 335

2. 0970 1. 2656 60. 353: >

‘Mean, 60. 332+. 0037
Rawson found, 60.346 +. 0046

General mean, 60. 337-E. 0029

 

1

ral

From Meineke’s figures Cr=52.11.

The chromate of silver, Ag,CrO,, and the ammonio-chromate, Ag,CrO,. 4NH,, both h prepared
with all necessary precautions to insure purity, were first treated essentially as in Berlin’s experi-
ments, except that the traces of silver chloride held in solution by the chromic chloride: were
thrown,out by. sulphuretted hydrogen, estimated, and their amount added to the main portion.
Thus the chief error in Berlin’s work was avoided. I subjoin. the data obtained, with vacuum
standards, as usual. All of Meineke’s results are so corrected:

Ag,CrO,. AgCl. Cr.O5-
2. 7826 2. 4047 0. 6384
3. 2627 2. 8199 . 7480
3. 6362 3. 1416 . 8338
4, 6781 4. 0414 1. 0726
3. 2325 2. 7930 ~741E > 8
: 3. 9137 3. 3805 . . 8976 |
Hence we have the following ratios, as in the case of Berlin’s data:
Per cent Cr,05. 100 AgCl:Ag,CrO,. 100 AgCl:Cr,03.
22.943 © 115.715 26. 548 ©
22. 926 115. 703 26. 526
22. 931 115. 744 26. 602
22. 928 115. 754 26. 601
22. 924 115. 736 26. 531
22. 935 115. 773 26. 552

 

Mean, 22. 931+. 0019 Mean, 115. 737+. 0072 Mean, 26. 560+. 0093
Berlin, 23. 014+. 0110 :

General méan, 22. 934, 0018
From. Meineke’ s figures Cr=52. 10, 52.04 and 52.14,

1 Ann. Chem. 3 261, 339. 1891. * Caloulated back from Meineke’s value for Cr to replace an evident misprint in the original.

'
No. 8.] .. ATOMIC WEIGHTS—CLARKE. 269

‘With the ammonio-chromate Meineke found as follows: |:

 

 

Ag,Cr0,.4NHy. “ AgCl. Cr203.
4.1518 2, 9724 0.7904
18 Po, so. 42601 3. 0592 - 8125
5. 9348 4. 2654 1.1317
And the ratios become— a
Per cent Cr,03. 100 AgCl:Salt. 100 AgCl:Cr,05.
19. 037 139. 679 26. 591
19. 072 139. 255 26. 559
19. 059 139. 138 26. 532
Mean, 19. 059+. 0074 Mean, 139. 357+. 1109 Mean, 26. 561+. 0115
Cr=52, 27° Cr=651. 61 Cr=52. 144

The first of these three analyses is rejected by Meineke as suspicious, but for the present
I shall allow it toremain. The data in the third column may now be combined with the corre-
sponding figures from the normal chromate, as found by Meineke and his predecessors:

Berlin jo5 Gs thas ae Scenic eek at caw A See eis 26. 682+. 0076
Siewert, from Ag,©r,0,.........-2-.2---2 22 eee eee eee eee 26. 525+. 0340
Meineke, from Ag,CrO,......-......2-.0 eee eee eee eee eens 26. 560+. 0093 pane
Meineke, from Ag,CrO,.4NH,.......---------- cece eee cece 26. 561+, 0115
Genoral Mean. ....0.:.- cee ceec ence cee e esc eeeeteess 26. 620-4. 0052

: 4 AgCl : Cr,O, : : 100 : 26. 620+. 0052

Obviously, this mean is vitiated by the known error in Berlin’s work, the ultimate effect of
which is serious.

In all four of the salts studied by Meineke he determined volumetrically the oxygen in ex-
cess of the normal oxides by measuring the amount of iodine liberated in acid solutions. With
the silver salts the process was essentially as follows: A weighed quantity of the chromate was
dissolved in weak ammonia, and the solution was precipitated with potassium iodide. After
the silver iodide had been filtered off, five or six grams of potassium iodide were added to the
filtrate; which was then acidulated with phosphoric acid and a little sulphuric. _ The liberated
iodine, was then titrated with sodium thiosulphate solution, which had been standardized by
means of pure iodine, prepared by Stas’ method.. From the iodine thus measured the excessive
oxygen was computed, and from that. datum the atomic weight of chromium was. found. For
present purposes, however, the data may be used more directly, as giving the ratios 31: Ag,CrO,
and. 31: Ag,CrO,.4NH;. Thus treated,..the weights are as follows, reduced to a vacuum.
Reckoning the salt as 100, the third column gives the percentage of iodine liberated:

Ag,Cr0,. I Set Free. Percentage.
0. 43838 0. 50251 -114. 628

- 90258 1, 03432 114. 595

. 89858 1. 02980 114, 603

. 89868 1. 03072 114. 693

Hence, Cr= 52.42.

The next series, obviously, gives the ratio 3I : Ag,CrO,.4NH,:

 

Mean, 114. 630+. 015

- Ag,Cr0,.4NH3. I Set Free. Percentages,'. i
0. 54356 0. 51784 95. 267
- . 64856 - 52046 "94. 877
- 54926 . 62322 95. 258
- 54906 - 52376 95. 392
. 04466 . 61910 95. 307
« 54536 . 51891 95. 150

Hence, Cr= 52.097.

 

Mean, 95. 208+. 0497

1 These figures are not wholly in accord with the percentages of oxygen computed by Meineke. I suspect that there is a misprint among his data

as published, probably in the second experiment, but I cannot trace it with certainty.
270 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVL

In dealing with the two dichromates Meineke used the acid potassium iodate in place of
potassium iodide, the chromate and the iodate reacting in the molecular ratio of 2:1. The thio-
sulphate was standardized ‘by means of the acid iodate, so that we have direct ratios between
the latter and the two chromates. The data are as follows, with the amount of iodate propor-
tional to one hundred parts of the dichromate in the third column:

K,Cr,0;. KHI,O,. Percentage.
0. 25090 0. 16609 = 66.198
. 25095 . 16613. "66. 200
. 25078 . 16601 66, 197
. 24979 . 16541 66, 220
. 24987 . 16540 66. 192
. 24966 . 16543 66. 262 -
. 25015 . 16559 66. 196
. 25012 16559 66. 204
. 24977 . 16546 66, 245
; 25034 . 16572 “> 66,198
. 25025 . 16567 66. 202
. 25015 . 16568 66, 234

 

Mean, 66. 212+. 0044
Hence Cr=52.14.

Am,Cr,0;. KHI,O,. Percentage.
“1 0, 21457 0. 16584 & 77.290
. 21465 . 16588 77.279
. 21464 . 16584 77. 264
. 21416 - 16543 77. 246
. 21447 - 16564 77. 232
. 21427 . 16559 77. 281
. 22196 .17152 77. 272
. 22194 . 17151 77, 278
. 22180 . 17139 77. 272

 

Mean, 7. 268+. 0041

Hence, Cr= 52.13.

' Baxter, Mueller, and Hines‘determined the atomic weight of Gonnin by the aniilyidd of
silver chromate. The weighed salt was dissolved in nitric acid and reduced either by sul
phurous acid or hydrazine sulphate. The silver was then precipitated as chloride or bromide,
by weak hydrochloric or hydrobromic acid, and the halide compound was weighed. All modern
precautions were taken in this work, such as determining traces of moisture in the chromate, and
also the traces of silver chloride or bromide oa in solution.- The! chloride series, wits
vacuum weights, was as follows:

Ag,CrO,. AgCl. Ratio.
10. 30985 8. 90908 115. 723
8. 26920 7, 14492 115. 735

5. 56679 5. 67444 115. 726
Mean, 115. 728+. 0024

 

Hence, Cr= 52.010.
This combines with previous series as follows:

Deri wai jaaaaGesioigennnenee sb neuacsaneug sieved Me ABH Oed

Meinok@ icc o2icccwcitiseniek sh lcsee'atis ve dsicteinenbes eons 115. 737+. 0072
Baxteretidliencsstenesaeuidewecues ¢c's se eenswasese sens 115. 728+. 0024
—
General Mean wsccecewse secs Sis scdecek ene ceaee eee 115. 731+. 0023
For the bromide series, all corrections applied, the data are these:

Ag,CrO,. AgBr. Ratio.

2. 63788 2. 98621 88. 336

2. 82753 3. 20084 88. 337

 

1 Journ. Amer. Chem. Boc., 31, 529. 1909.
No. 3.] -

ATOMIC WEIGHTS—CLARKE..

Ag,CrC,. AgBr. Ratto.
2. 33454 2. 64268 88. 340
1. 77910 2. 01402 88. 336
2. 33198 2. 63994 88. 335
3. 10402. 3. 51390 88. 336
2. 92751 3. 31427 88. 330
4, 21999 4. 77762 88. 328
5. 24815: 5. 94104 88. 337
6, 24014 7. 06484 88. 327

7. 92313 8. 96982 88. 331
Mean, 88. 334+. 0009

 

271

Hence, Cr=52.028. Similar determinations of the atomic weight of chromium were made

by Baxter and Jesse,’ who analyzed silver dichromate.
of Ag,Cr,O, gave 4.16076 of AgCl. Ratio, 150.611.

In a single experiment 6.26657 grammes
Giving this the weight of a single deter-

mination in the following bromide series + .0024, it combines with the earlier measurements by

Cr = 52.003.

Berlin and Siewert to a general mean of 150.612+.0024. The older work vanishes.
The dichromate-bromide series of determinations gave the subjoined data:
Ag,Cr,0,. AgBr. Ratio.
5. 71554 4. 97149 114. 966
4. 87301 4, 23888 114, 960
7. 45476 6. 48425 114. 967
4. 75269 4, 13420 114. 960
8. 15615 7. 09495 114. 957
6. 15412 5. 35309 114, 964
6. 83662 ‘ 5. 94768 ‘114, 963
5. 39883 4, 69631 114. 959

 

Mean, 114. 962-4. 0008

Hence, Cr = 52.022.

The following ratios are now available for computing the atomic weight of chromium:

(1) Percentage Cr,0, from Ag,CrO,, 22. 934+. 0018.

(2) Percentage Cr,0, from Ag,Cr,0,, 35. 236 + . 0335.
(3) 2AgCl : Ag,CrO, :: 100 : 115. 731-4. 0023.

(4) 2AgCl : Ag,Cr,O, : : 100 : 150. 612 +. 0024.

(5) 4AgCl : Cr,O, : 100 : 26. 620+. 0052.

(6) Percentage Cr,O, in Cr,(SO,);, 38. 799 +. 0092.

(7) 3AgCl : CrCl, :: 100 : 36. 842-4. 0031.

(8) 2KCIO, : K,Cr,0, :: 100 ; 120. 216+. 0235.

(9) Percentage Cr,0, in Ag,CrO,.4NH,, 19. 059+. 0074.
(10) 2AgCl : Ag,CrO,.4NH, :: 100 : 139. 357+. 1109.
(11) Percentage Cr,05 in Am,Cr,0,, 60. 337 +. 0029.
(12) Ag,CrO, : 8I :: 100 : 114. 630+. 015.

(18) Ag,CrO, ANH, : 81 :: 100 : 95. 208+. 0497.

(14) 2K,Cr,0, : KHI,O, :: 100 : 66. 212-4. 0044.

(15) 2Am,Cr,0, : KHI,O, :: 100 : 77. 268+. 0041.

(16) 2AgBr : Ag,CrO, :: 100 : 88. 334+. 0009.
(17) 2AgBr : Ag,Cr,O, :: 100 : 114. 962+. 0008.

To reduce these we have the following atomic, weight:

Ag= 107. 880+. 00016 $=32. 0650 +. 00012
Cl= 35. 4605+. 00011 N=14. 0081+. 00008
Br= 79. 9228+. 00015 K=39. 0975+. 00011

I= 126. 926+. 00023 E H= 1. 0078+. 00001

Hence

 

1 Journ. Amer. Chem. Soc., 31, 641. 1909,

161566°—20——_18
272 MEMOIRS NATIONAL. ACADEMY OF SCIENCES. [Vou, XVI.

Hence,
From ratio—

Dh oh osc cheesey snes 4 wetanmnecinbelemenwaae Cr=51, 624+. 3180

eS i rawlcee he ees he ese lbc te So ke 2 oad api are rtp Nt aa et a 52. 003 +. 0035

Seweeat peeeeeeelewe cvs es veesamtinwneeiensoawenawes 52. 014+. 0067
Wee ewe spediee Se ane eemeeeians basaistanizicidcannas 52. 022+. 0015

A Gus ciedais ciacuideaaaeaeanae i yosen Anda aa ans 52. 028+. 0033
Ledanae state aie ceameaeenas ioe decaaeenens 52.048 +. 0134

Weak case eStats Aste vies Geeta omepeasacesanwen’ 52. 097+. 2088

De roams elie sa ood oe ade cane eatin: 52. 104+. 0824
Darah ecctces snp eect cere ete e ence reese rn eceeneeseees 52. 111+. 0063

SD TS pat Aes enteccs eo hes ete ela aa AL ee 52. 111+. 0066

¢ LObtossecacene jae diee sve sas wesc deenapeds seeeeeeee , 52,1304. 0067

g Oligo ecaiaarets Jppeeeeeeeceneeeeeceenes edinzasaoticieueicyatses 52. 137+. 0167.

by Nias deks siabae ces Mate Les Satna realenian aaengine 52. 140+. 0098
Be esha icianae cs S2625.5.05cnanemnanmaneseuaagacs 52. 287 +. 0288

Qi visite Save e ahaa cen wa oanue See SSateedewtece 52. 265+. 0304
Dictcuaeey Spcehen oO Pa OEE Reioiaeaeeecoaceeteeene 52. 315+. 0149

Ds icieiienteentene chs ae sina vials tivicte woes SaaS BNeees wtenele 52. 420+. 0440
General: M Ca Bsn aie viescis's a's sicicisinistelsiasaae neieiecte Cr=52. 036+. 0012

In this combination the work of Baxter and his colleagues carries overwhelming weight,
and yet the work of Siewert, Baubigny, Rawson, and, in part, Meineke’s is not entirely ignored.
The high values, with their large probable errors, practically vanish from the general mean.

MOLYBDENUM.

If we leave out of account the inaccurate determination made by Berzelius,' we shall find
that the data for the atomic weight of molybdenum lead to two independent estimates of its
value—one near 92, the other near 96. The earlier results found by Berlin and by Svanberg
and Struve lead to the lower number; the more recent investigations, together with considerations
based upon the periodic classification, point conclusively to the higher.

The earliest investigation which we need especially to consider is that of Svanberg and
Struve.2. These chemists tried a variety of different methods, but finally based their conclu-
sions upon the two following: First, molybdenum trioxide was fused with potassium carbonate,
and the carbon dioxide which was expelled was estimated; secondly, molybdenum disulphide
was converted into the trioxide by roasting, and the ratio between the weights of the two
substances was determined.

By the first method it was found that 100 parts of MoO, will expel the following quantities

of CO,:
31. 4954
31. 3749

31. 4705
Mean, 31. 4469+. 0248

 

The carbon dioxide was determined simply from the loss of weight when the weighed quan-
tities of trioxide and carbonate were fused together. It is plain that if, under these circum-
stances, a little of the trioxide should be volatilized, the total loss of weight would be slightly
increased. A constant error of this kind would tend to make the atomic weight of molyb-
denum too low.

By the second method, the conversion by roasting of MoS, into MoO,, Svanberg and Struve
obtained these results. Two samples of artificial cisulpiide were ae A and B, and yielded
for each hundred parts the following of trioxide:

. 89. 7919
os . 89. ale

 

1 Poggend, Annalen, 8,1. 1826. 2 Journ. prakt. Chem., 44, 301. 1848,
No. 3.] ATOMIC WEIGHTS—CLARKE. | 273

89. 6436
s.r]
89. 7660 \B
89. 7640 |
89, 8635

Mean, 89. 7523+. 0176

 

Three other experiments in series B gave. divergent results, and, although published, are
rejected by the authors themselves. Hence, it is not necessary to cite them in this discussion.
We again encounter in these figures the same source of constant error which apparently vitiates —
the preceding series, namely, the possible volatilization of the trioxide. Here, also, such an
error would tend to reduce the atomic weight of molybdenum.

From the CO, series........-. 2... cee ee eee eeee eee roistsreigcaiessidacje Mo=$91. 93
From the MoS, series. ...... Pala date can cregainman sect weaahnee Mo=$93. 30

~ Berlin,’ a little later than Svanberg bail Struve, determined the atomic weight of molyb-
denum. by igniting a molybdate of ammonium end weighing the residual MoO,. Here, again,
a loss of the latter by volatilization may (and probably does) lead to too low aresult. The salt
used was (NH,) ae 3H; O, and in it these percentages of MoO, were found:

81. 589 81. 558

81. 612 ae 81. 555
Mean, 81. 581+. 0095

Hence, Mo=92.16.

Until 1859 the value 92 was generally accepted on the basis of the foregoing researches,
but in this year Dumas? published some figures tending to sustain a higher number. He
prepared molybdenum trioxide by roasting the disulphide, and then reduced it to metal by
ignition in hydrogen. At the beginning the hydrogen was allowed to act at a comparatively
low temperature, in order to avoid volatilization of trioxide; but at the end of the operation
the heat was raised sufficiently to insure a complete reduction. From the weights I calculate
the perenar of metal in MoQ,: Mh ee

Mod; | - ss Mo. Per cent.
0. 448 ~ 0. 299 66.741
. 484 . 323 66. 736
» 4840 ; . 322 66. 529
498 + 332 66. 667
559 .373 66.726
. 388 . 258 66. 495

 

Mean, 66. 649-4. 030_
Hence, Mo=95.924.
In 1868 the same method was employed by Debray.? His trioxide was purified by subli-

mation in a platinum tube. His data are as follows:

Mo0;: Mo. Per cent.
5. 514 3.667 66. 503
7.910 5. 265. 66. 561
9. 031 6.015 66. 604

Mean, 66. 556+. 020

Hence, Mo=95. 524,

For the same ratio we have also a single experiment by Rammelsberg,! who, closely fol-
lowing Dumas’ method, found in molybdenum trioxide 66.708 per cent of metal. As this
figure falls within the limits of Dumas’ series, we may assign it equal weight with one aie
ment in the latter.

Debray also made two experiments upon the precipitation of molybdenum ficiila in
‘ammoniacal solution by nitrate of silver: In his results; as published, there is curious eat

 

1 Journ. prakt. Chem., 49, 444. 1850. 3 Compt. Rend., 66, 734.
2 Ann, Chem. Pharm., 105, 84; and 113, 23. ‘ Berlin Monatsbericht, 1877, p. 574.
274 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

ancy, which, I have no doubt, is due to a typographical error. These results I am therefore
compelled to leave out of consideration. They could not, however, exert a very profound
influence upon the final discussion.

In 1873, Lothar Meyer ' discussed the analyses made by Liechti and Kemp ? of four chlo-
rides of molybdenum, and in the first edition of this work the same data were considered in
detail. The analyses, however, were not intended as determinations of atomic weight, and
since good determinations have been more recently published, the work on the chlorides will
be omitted from further consideration. It is enough to state here that they gave values for
Mo ranging near 96, both above and below that number, with an extreme range of over eight-
tenths of a unit.

In 1893 the determinations by Smith and Maas appeared,’ representing an entirely new
method. Sodium molybdate, purified by many recrystallizations and afterwards dehydrated,
was heated in a current of pure, dry, gaseous hydrochloric acid. The compound Mo,0.2HCl1
was thus distilled off, and the sodium molybdate was quantitatively transformed into sodium
chloride. The latter salt was afterwards carefully examined, and proved to be free from molyb-
denum. The data, with all weights reduced to a vacuum standard, are subjoined: —

Na, Mo0,. NaCl. Per cent NaCl.
1. 14726 0. 65087 56. 733
. 89920 . 51023 56. 743
. 70534 - 40020 56. 739
. 70793 . 40182 56.760
1. 26347 . 71695 56.745
1. 15217 . 65367 56.734
. 90199 . 51188 56.750
. 81692 . 46358 56.747
. 65098 . 36942 56.748
. 80563 . - 45717 56.747

Mean, 56.745+ .0017
Hence, Mo = 96.040.

In 1895, Seubert and Pollard * determined the atomic weight of molybdenum by two
methods. First, the carefully purified trioxide, in weighed amounts, was dissolved in an excess
of a standard solution of caustic soda. This solution was standardized by means of hydrochloric
acid, which in turn had been standardized gravimetrically as silver chloride. Hence, indi-
rectly, the ratio 2AgCl : MoO, was measured. Sulphuric acid and lime water were also used
in the titrations, so that the entire process was rather complicated. Ignoring the intermediate
data, the end results, in weights of MoO, and AgCl, were as follows. The third column gives the
MoO, proportional to 100 parts of AgCl:

Mo0,. AgCl. Ratio.
3.6002 7.1709 50.206
3.5925 7.1569 50.196
; 3.7311 7.4304 50.214
3.8668 7.7011 50.211
3.9361 7.8407 50.201
3.8986 7.7649 50.208
3.9630 7.8941 50.202
3.9554 7.8806 50.192
3.9147 7.7999 50.189
3.8543 7.6767 50.208
3.9367 71,8487 50.190

Mean, 50.2024 .0018
Hence, Mo= 95.92.

The second method adopted by Seubert and Pollard was the old one of reducing the
trioxide to metal by heating in a current of hydrogen. The weights and percentages of metal
are subjoined:

 

1 Ann. Chem. Pharm., 169, 365. 1873. 8 Journ, Amer. Chem. Soc., 15, 397. 1893.
2 Ann, Chem. Pharm., 169, 344. 6 Zeitsch, anorg. Chem., 8, 434. 1895,
No. 3.] ATOMIC WEIGHTS—CLARKE. 275

Mo0;. Mo. Per cent.
1.8033 1.2021 66.661
1.9345 1.1564 66.670
3.9413 2.6275 66.666
1.5241 1.0160 66.662
4.0533 2.7027 66.679

Mean, 66.668-+ .0022
Hence, Mo = 96.006.

Vandenberghe ' prepared molybdenum dibromide, which was next reduced to metal by
heating in hydrogen. The metal was then oxidized to trioxide by means of nitric acid. The
data are as follows:

Mo. MoO. Per cent Mo.
0.7143 1.0711 66.689
3453 5177 66.699
-9693 1.4533 66.696
5089 -7631 66.689
1.7212 2.5820 66.689

Mean, 66.692+ .0015
Corrected to a vacuum this becomes 66.687 +.0015.

Hence, Mo = 96.088.
The determinations of the atomic weight of molybdenum by Miller? are based upon the
oxidation of the metal to MoO,. The data, with vacuum weights, are as follows:

Mo. Mo0;. Per cent Mo.
0.52591 0.78879 66.673
56327 -84487 66.669
1.12757 1.69117 66.674
53014 .79517 66.670
1.10754 1.66130 66.667
1.62166 2.43181 66.686
1.45530 2.18259 66.678
94968 1.42428 66.678
.65659 -9870 66.679

Mean, 66.675+ .0013
Hence, Mo = 96.036.

This mean may be combined with former determinations thus:

ADUTIAS Soo isidichaisle sieisiniele Seseraraed areg renee Svlsicine CeCe Gree’ o’s'sleteheigats 66.649 + .0300
DODray seise cacwrsnaustaachsecwlele's sec aviaia teeing s xis auadoee 66.556 + .0200
Rammelsbergsneevsis ves teacse ass csneeestsckive eee ess sowse ses 66.708 + .0680
Seubert and Pollard.............-.... Rice SG A Seles cies arsenal 66.668 + .0022
Vandenberghe. ......< 5 c0cece ioe: sss caaces cannes te £5 Gianna 66.687 + .0015
MUM GP oc nse cctecnuniieies sisiewenese i ate Sakata cartels tay tease A 66.675 -+ .0013

General Mean sisscescsawease cece nace veewmesed se hoes 66.679 + .0009

Neglecting all determinations made before 1859, there are now three ratios from which
to compute the atomic weight of molybdenum, as follows:

(1) MoO, : Mo :: 100 : 66.679 + .0009
(2) 2AgCl : MoO, :: 100 : 50.202 +.0018
(3) Na,MoO, : 2NaCl :: 100 : 56.745 +.0017

Reducing these with Ag=107.880+.00016; Cl=35.4605+4.00011; and Na=23.0029 + .00012
we have—

From ratio—2.......0.-00 eee ence ee nec eeeeeee iivivw-s ss daisereiaiees Mo=95.920+ .0052
Basie Sc soceyn ava bectacd es wa Myacm ra ni bve tebe saje& bus ake bracapas we eidieaee 96.040 + .0062

Peet Lie laa enero ee rcears die ayer a raises eS eperoers ete 96.053+ .0029

General mean....:.........20.002000s002----Mo=96,028- .0024

In this combination the actual uncertainty is greater than the decimals. For practical
purposes the round number 96 can be used.

 

1 Acad. Roy. Belge. Mém. Couronnés, T. 56. 3 Journ, Amer. Chem, Soc. 37, 2046. 1915.
276 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou.- XVI.
= TUNGSTEN.

The atomic weight of tungsten has been. determined from analysts of the trioxide, the
hexchloride, and the tungstates of iron, silver, sodium, and barium. ©

The composition of the trioxide has been the’ subject of many ‘investigations. Malaguti*
reduced this substance to the blue oxide, and from the difference between the weights of the
two compounds obtained a result now known to be considerably too high,,, In general, however,
the. method of investigation has been to reduce WO, to W in a stream of hydrogen at a white
heat, and afterwards to reoxidize the metal, thus getting from one sample of material two
results for the percentage of tungsten. This mehr is probably accurate, provided that the
trioxide used be pure.

The first experiments which we need eouhided are, as usual, those of Berzelius.2 899 parts
WO, gave, on reduction, 716 of metal. 676 of metal, repxidized, gave 846 WO,. Hence these

percentages of W in WO,:
. 79. 644 by reduction

79. 905 by oxidation
Mean, 79. 7745+. 0880

 

Hence, W = 189.324.

These figures are far ‘too high, the error being: one due to the presence of allcelina
impurity in the trioxide employed.

Next in order of time comes the work of Schneider 3 hie: with. dhupnecseaie caretrinees,
took every precaution to get’pure material. His percentages of tungsten are as follows:

Reduction series.

79.336
79. 254
79. 312
79. 326
79. 350

Mean, 79. 3156

 

Oxidation Series.

79. 329.
79. 324
79. 328

Mean, 79. 327
Mean of all, 79. 320+.0068

Hence, W = 184.108.
5,8 Closely agreeing with naeee figures are those of Marchand, . published i in the ages year:
Rethiciion Series.

79, 307),
79.302.

Mean, 79. 3045:

Oxidation Series.

79. 321
“79; 352 ‘

: Mean, 79. 3365
Mean of all, 79. 3205+ .0073

 

 

Hence, W= 184,114.

1 Journ, prakt, Chem., 8, 179. 1836. 3 Journ. prakt. Chem,, 60,152. 1850.
2 Poggend. Annalen, 8, 1. 1826, 4 Ann, Chem, Pharm., 1, 261. 1851.

sid
No. 8.) ATOMIC WEIGHTS—CLARKKE. | - 277

The figures obtained by v. Borch’ agree approximately with the foregoing series. They

are as follows:
Reduction Series.

79. 310
79. 212
79. 289
79, 313
79, 225
79. 290
79. 302

en

Mean, 79. 277
Oxidation Series.
79. 359
79. 339

Mean, 79. 349
Mean of all, 79. 293+ .0108

 

Hence, W = 183.806.

Dumas? gives only a reduction series, based upon trioxide obtained by the ignition of a
pure ammonium tungstate. The reduction was effected in a porcelain boat, platinum being
objectionable on account of the tendency of tungsten to alloy with it. Dumas publishes only
weighings, from which I have calculated the percentages:

Wo; W. ‘Per cent,
2. 784 2, 208 “79.310
2. 994 2. 373 79. 259
4. 600 3. 649 79. 326

. 985 - 781 79, 289

.917 727 79. 280

-917 . 728 79. 389
1,717 1, 362 79. 324
2. 988 2. 370 79. 317

 

Mean, 79. 312+.009
Hence, W = 184.019.
The data furnished by Bernoulli ® differ widely from those just given. This chemist un-
doubtedly worked with impure material, the trioxide having a greenish tinge. Hence the
results are too high. Here are the percentages of W:

Reduction Series.

79. 556 ,
79, 526
79, 553
79. 558
79, 549
78, 736

Mean, 79. 413

Oxidation Series.

79, 558
79. 656
» 79, 555
“79. 554

Mean, 79. 581
Mean of all, 79. 480+ .056

 

Hence, W= 185.918.

1 Journ. prakt. Chem., 54, 254. 1851. 2 Ann, Chem. Pharm., 113, 23, 1860, | 3 Poggend, Annalen, 111,578. 1860.

 
278 MEMOIRS NATIONAL ACADEMY OF ‘SCIENCES [Von. XVI,

'Two.reduction experiments by Persoz? give the following results: |

Wo; W. Per cent.
1. 7999 1. 4274 79. 304
2. 249 1. 784 79, 324

 

Mean, 79. 3142-.007
Hence, W = 184.041.

Next in order is the work done by Hicsoot: 2 This chemist used a porcelain boat and tube,
and made six weighings, after successive reductions and oxidations, with the same sample of
7.884 grams of trioxide. These weighings give me the following five percentages which, for the
sake of uniformity with foregoing series, I have classified under the usual, separate headings:

Reduction Series

79. 196
79, 285
79. 308

Mean, 79. 263
Oxidation Series.

» 79, 230

' 79, 299

_ Mean, 79.2645
Mean of all, 79. 264+. 0146

 

 

Hence, W = 183.482.

In Waddell’s experiments * especial precautions were taken to procure tungstic oxide
free from silica and molybdenum. This cate, elaborately purified, was reduced in hydrogen,
with the following results:

Ws. W. Per cent.
1 4006 1. 1115 79. 359
. 9900 . 7855 79. 343
1 1479 . 9110 79. 362
. 9894 . 7847 79. 311
4, 5639 3. 6201 79. 320

Mean, 79, 339. 0069
Hence, W = 184.332.

The investigation by Pennington and Smith ‘started from the supposition that the tungsten
compounds studied by their predecessors had not been completely freed from molybdenum.
Accordingly, tungstic oxide, carefully freed from all other impurities, was heated in a stream
of gaseous hydrochloric acid, so as to volatilize all molybdenum as the compound MoO,.2HCI.
The residual WO,, was then reduced in pure hydrogen, and the tungsten so obtained was
oxidized in porcelain crucibles. Care was taken to exclude reducing gases, and the trioxide was
finally cooled in vacuum, desiccators over sulphuric acid. The oxidation data are as follows,
with the usual percentage column added. The weights are reduced to a vacuum:

Tungsten. Oxygen ee Percentage.
0. 862871 0, 223952 79, 394

. 650700 . 168900 79, 392

. 597654 - 155143 79. 390

. 666820 . 173103 79, 391

. 428228 . 111168 79. 390

. 671920 . 174406 79. 392

. 590220 . 153193 79. 394

. 568654 . 147588 79. 394
1. 080973 .. 280600 79. 392

; Mean, 79. 392-+. 0004
Hence, W = 184.92. ES

1 Zeit, anal. Chem., 3, 260. 1864. 3 Amer. Chem. Journ., 8, 280. 1886.
2 Ann, Chem. Pharm., 162, 368. 1872. 4 Read before the Amer. Philos. Soc., Nov. 2, 1894.

 
No, 8.] ATOMIC WEIGHTS—CLARKE. 279

The very high value for tungsten found by Pennington and Smith, nearly a unit higher
than that which was commonly accepted, seems to have at once attracted the attention of
Schneider,' who criticized the paper somewhat fully, and gave some new determinations of his
own. The tungsten trioxide employed in this new investigation was heated in gaseous hydro-
chloric acid, and the absence. of molybdenum was proved. The data obtained, both by reduc-
tion and by oxidation, are as follows:

Reduction Series.

Wo;. W. Per cent.
2, 0738 1. 6450 79, 323
4. 0853 3. 2400 79. 309
6. 1547 4. 8811 79. 307

Oxidation Series.

W. WO;. Per cent.
1. 5253 1. 9232 79. 311
3. 1938 4. 0273 79, 304
4. 7468 5. 9848 79. 314

 

Mean of all, 79. 311+. 0018

Hence, W=184.007.

In order to account for the difference between this result and that of Pennington and Smith,
an impurity in the molybdenum trioxide amounting to about 1 per cent would be necessary.
Schneider suggests that the quantities of material used by Pennington and Smith were too small,
and that there may have been mechanical loss of small particles during the long heatings. Such
osses would tend to raise the atomic weight computed from the experiments. On the other
hand, the losses could hardly have been uniform in extent, and the extremely low probable
error of Pennington and Smith’s series renders Schneider’s supposition improbable. The error,
if error exists, must be accounted for otherwise.

Soon after Schneider’s paper appeared, another set of determinations by Shinn 2 was
published from Smith’s laboratory. Attempts to verify the results obtained by Smith and Desi
having proved abortive, and other experiments having failed, Shinn resorted to the oxidation
method and gives the subjoined data. The percentage column is added by myself:

W. Wo. Per cent.

0, 22297 0. 28090 79. 377
. 17200 . 21664 79. 394
. 10989 . 13844 79. 377
. 10005 . 12598 79. 417

 

Mean, 79. 391-4. 0066

Hence, W = 184.908.

This figure is very close to that found in Pennington and Smith’s series.

The great discordance between the determinations so far cited, led Hardin * to a very
careful investigation of tungsten trioxide. The substance was prepared from various sources,
and manipulated by various methods; and although concordant results were sometimes obtained
in succession, the discordance between different series of experiments was very great. Hardin
therefore concluded that a discussion of his figures, with reference to the atomic weight of tungsten,
would be useless. Nevertheless, partly for the sake of completeness, and partly because this
calculation is in great measure a study of the compensation of errors, I prefer to cite Hardin’s _
determinations, in order that they may be compared with others. For this purpose I give his
64 determinations as one series. The letters o and r indicate oxidation and reduction
experiments, respectively. ‘

1 Journ. prakt. Chem. (2), 53, 288. 1896.
2 Thesis, University of Pennsylvania, 1896. ‘The atomic mass of tungsten.”
8 Journ. Amer. Chem. Soc., 19, 657. 1897.
280 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

The atomic weights found were as follows:

r 184. 05 r 184. 01 o 184. 86 o 184. 20
r 184. 04 r 184, 66 0 184.27. r 183. 58
r 183. 98 r 183, 99 0 184. 07 r 183. 51
r 184. 33 r 183. 93 | r 183. 83 5 183.83
r 183. 94 r 183. 91 r 183, 80 o 184. 05
r 183. 91 o 184. 53 r 183. 67 o 184. 22
r 183. 66 o 184.01 r 183. 56 o 184. 06
o 184, 94 o 184. 65 r 183. 72 r 184. 03
018486 r 183. 55 r 183. 71 r 183. 81
o 185. 00 r 184, 34 ; r 183. 80 o 183. 85
o 184. 91 r 184, 21 r 183. 87 o 184, 14
o 184.75 r 183. 95 o 183. 83 r 183. 89
o 184.15 o 183. 70 o 183. 75 r 183. 63
r 184. 88 o 184. 30 ' 0 184.13 o 184. 17
r 184. 85 o 183. 99 o 183. 90 o 184. 08
tT 184, 94 o 184, 07 o 183. 82 r 134. 13

The mean of all is W=184.105, +.0337... This gives a percentage of W in WO, of 79.320,
+.0185. The discordances were shown by Hardin to be due partly to impurities in his material,
such as nitrogen retained by trioxide prepared from ammonium tungstate, and partly to vola-
tility of the oxide at high temperatures. In a later memoir' he discusses these errors at some
length, and gives a few other determinations which are even more discordant, and therefore
not worth citation now. ee

‘Taylor’s thesis,? representing work done in Smith’s laboratory,'is essentially a study of
errors. He found that constant weight could not be secured during reduction experiments
with the trioxide, and he also found, like Hardin, that the oxidations generally gave the higher
values for the atomic weight of tungsten. Furthermore, he ascertained that tungstic oxide
derived from colloidal ammonium tungstate gave different values dependent upon whether the
latter compound was dialyzed or undialyzed. Oxide from the dialyzed salt gave the highest
atomic weights. Some of the discrepancies were ultimately traced to the presence in the
material studied of a complex salt containing manganese and iron, and the inflrence of these
impurities was studied. Iron, and also molybdenum, tend to lower the apparent atomic weight
of tungsten; manganese, and in much greater measure, raises it. The errors are in opposite
directions, but do not absolutely compensate one another.

One new method for measuring the atomic weight of tungsten was tested by Taylor, but
the results were not satisfactory. Sodium carbonate was heated in a glass bulb with tungsten
trioxide and water, the latter was distilled off after effervescence had ceased, and the residue
was then heated to 300° ina vacuum. The weights of carbonate and oxide being known, the
loss in weight represented carbon dioxide. The ratio between WO, and CO, was thus deter-
mined. I cite the weights, and also the values for the ratio WO, : CO, : : 100: 2:

Weight WO,.. Weight CO.. ve Ratio.
2. 0802 0. 3952 -* 18. 998
2. 1937 . 4173 ‘19. 023
4, 0818 7762 ; 19. 016

3, 3629 . 6394 19.018
7 Mean, 19.0125 + .0034

 

.. Hence, W=183.44; a determination which Taylor regards as worthless, while anon
that the method is one of some promise.

Several of the investigations so far described were carried out under the direction of, or in
cooperation with Prof. Edgar F. Smith. The experience obtained in their conduct gave a
sound basis for further researches, which were undertaken by Smith and Exner.’ These authors
discuss at length the sources of error in former determinations of the atomic weight of tungsten,

 

1 Journ. Amer. Chem. Soc., 21, 1017. 1899,
3 Thesis, University of Pennsylvania, 1901. “Atomic weight of tungsten.”
8 Proc. Amer. Phil. Soc., 43, 123. 1904.
No. 8.1 ' ATOMIC WEIGHTS—CLARKE. . 281

and:point out the difficulty of preparing pure material, a difficulty which was at last overcome.
From a pure ammonium tungstate they prepared pure tungsten, the pure trioxide, and pure
tungsten hexchloride, free from oxychloride, and with these substances their atomic weight
determinations were made. At this point only their syntheses of the trioxide will be considered,
their other series being discussed later. Their figures, with vacuum weights, and the usual
percentage column are given below:

Weight W. Weight WOs3. Per cent.

2. 24552 2. 83144 79, 306
1. 78151 2. 24619 79. 313
L. 63590 2.06270 79. 309
1. 38534 1. 74665 79, 314
1. 29903 1. 638774 79, 318
2. 01302 2, 53781 79, 321
2. 18607 2. 75632 79,311
2. 36755 2. 98478 79. 323
1, 94958, 2. 45781 79. 322
4, 43502 5. 59141 , 79. 318
2. 37603 2. 99548 79. 321
"2. 58780 3. 26260 79, 314
. 2. 58503 3. 25886 79. 322
2. 38298 3. 00441 79. 316
2. 05578 2. 59169 79. 322
3. 60828 4, 54915 79. 318
6. 22621 7. 84949 79. 320
5. 28444. 6. 66239 79. 317
38. 99095 5. 03138 79. 321
7. 30166 9. 20647 79. 309
3. 44143 4. 33870 79. 319
2. 67709 8.37541 79, 312

4, 96735 6. 26229 79. 322
Mean. 79. 3169 + .0007

 

Hence, w- 184.075, +.0064.

There are still other analyses by Riche,’ which are of no value for they give to tungsten. an
atomic weight of about ten units too low. We may, therefore, neglect his series ‘and go on to
corabme the others:

 

v1. Berzelius..........---------0 22 cece cree eee reeee -.- 79.7745 + .0880
“2. Schneider, 1850 ele 79.320 + .0068
9. Marchand<.)../.sceds ea. ded ase bok Eb aiees sed 2... 79.3205 + .0073
A, Borehivis dns s'vsieacice cab swee ee debesocatewcensetes 79.293 + .0108
5. DUM asicten sss damaeaisehelees ees scan eeneceeae ees 79.312 + .0090
6: Bernoulli. «0225 ceceeswdems tas Se eeaiaigeeedeaiess sees 79.480 + .0560
7. POreod.. 2004s scasaeeeeeeeee esses aires ere es eS 79.314 + .0070
8. RosC00.. iss sieccaincn sews esd es eeceseeechess 2 ales 79,264 + .0146
O. Waddell ccccn.05 o2e.ed:sdeat ets eas eeeecete eee seats 79.339 + .0069
10. Pennington and Smith: ....../.-.-------+-+--- meses 79.392 + .0004
11. Schneider, 1896..........------- pha teil ies 2 oa Sree 79.311 + .0018
12: Shinn a2 oi vecoseateukiss che scicicessorcedssse5ns 79.391 + .0066
18. Hardin sisjics cesses edeeds is vosminceeeeees ee sas 79.320 + .0185
14, Smith and Exner...........-.-2-2-- 22-2 eee eee eee 79.3169 + .0007
General mean........--- 2-20 - eee eee eee eee ee eee 79. 3706 +.00034

In this combination only two values carry much weight; the tenth and the fourteenth.
The series by Pennington and Smith is evidently much overvalued, and exerts’an undue influ-
ence upon the general mean. In reality the series by Smith and Exner is by far the most trust-
worthy of all, and the figures given by Schneider, Marchand, Dumas, and Persoz are in harmony
with it. The other series are more doubtful. The weighted mean of twelve series, omitting
Nos. 10 and 14, is 79.3160; a value almost identical with that of Smith and Exner. The latter,
therefore, is abundantly confirmed.

 

1 Ann, Chim. Phys., (3), 50, 11. 1857.
282 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

In 1861 Scheibler + deduced the atomic weight of tungsten from analyses of barium meta-
tungstate, BaO.4WO,.9H,O. In four experiments he estimated the barium as sulphate, getting
closely concordant results, which were, however, very far toolow. These, therefore, are rejected.
But from the percentage of water in the salt a hetiex result was attained. The percentages of

water are as follows:
13, 053

13. 054
13. 045
13. 010
13. 022

Mean, 13. 0368+ .0060
Hence, W = 184.06.

The work of Zettnow,? published in 1867, was more complicated than any of the foregoing
researches. He prepared the tungstates of silver and of iron, and from their composition
determined the atomic weight of tungsten.

In the case of the iron salt the method of working was this: The pure, artificial FeWO, was
fused with sodium carbonate, the resulting sodium tungstate was extracted by water, and the
thoroughly washed residual ferric oxide was dissolved in hydrochloric acid. This solution was
then reduced by zinc, and titrated for iron with potassium permanganate. Corrections were
applied for the drop in excess of permanganate needed to produce distinct reddening, and for
the iron contained in the zinc. 11.956 grammes of the latter metal contained iron correspond-
ing to 0.6 cc. of the standard solution. The permanganate was standardized by comparison
with pure ammonium-ferrous sulphate, Am,Fe(SO,),.6H,O, so that, in point of fact, Zettnow
establishes directly only the ratio between that salt and the ferrous tungstate. From Zettnow’s
four experiments in standardizing I find that 1 cc., of his solution corresponds to 0.0365457
gramme of the double sulphate, with a probable error of + .0000012.

Three sets of titrations were made. In the first a quantity of ferrous tungstate was treated
according to the process given above; the iron solution was diluted to 500 cc., and four titra-
tions made upon 100 cc. ata time. The second set was like the first, except that three titrations
were made with 100 cc. each, and a fourth upon 150 cc. In the third set the iron solution was
diluted to 300 cc., and only two titrations upon 100 cc. each were made. In sets one and two
thirty grammes of zinc were used for the reduction of each, while in number three but 20
grammes were taken. Zettnow’s figures, as given by him, are quite complicated; therefore I
have reduced them to a common standard. After applying all corrections the following quan-
tities of tungstate, in grammes, correspond to 1 cc. of permanganate solution:

0. 028301
. 028291
. 028311

. 028301
. 028367

‘ cad set

First set

 

. 028367
. 028367

.028438)_.
; osagey Third set

Mean, . 0283549+. 0000115

 

Hence, W = 184.40.

_ With the silver tungstate, Ag,WO,, Zettnow employed two methods. In two experiments
the substance was decomposed by nitric acid, and the silver thus taken into solution was titrated
with standard sodium chloride. In three others the tungstate was treated directly with common
salt, and the residual silver chloride collected and weighed. Here again, on account of some
complexity in Zettnow’s figures, I am compelled to reduce his data to a common standard. To
100 parts of AgCl the following quantities of Ag,WO, correspond:

 

1 Journ. prakt. Chem., 88, 324,
3 Poggend. Annalen, 130, 30.
No. 3.) ATOMIC WEIGHTS—CLARKE. 283

By First Method.

161. 665
161. 603
Mean, 161. 634+. 021

By Second Method.

161. 687
161. 651
161. 613
Mean, 161. 650+. 014
General mean from both series, 161. 645+. 012 ;

Hence, W = 183.64.

For tungsten hexchloride we have first, two analyses by Roscoe, published in the same
paper with the results upon the trioxide. In one experiment the chlorine was determined as
AgCl; in the other the chloride was reduced by hydrogen, and the residual tungsten estimated.
By bringing both results into one form of expression we have for the percentage of chlorine in

al
vse 59, 610
53. 632
Mean, 53. 621+. 0074
Hence, W = 184.03.

The investigation of tungsten hexchloride by Smith and Exner was much more elaborate.
They prepared the substance from scrupulously pure materials, and further purified it by
repeated sublimations. They decomposed the chloride by means of water, and weighed the
residual tungsten trioxide. Their figures, with vacuum weights, are as follows, with a per-
centage column added by myself:

Weight WCl,. Weight WO3. Per cent WO3.
3. 18167 1, 86085 58. 487
2. 66612 1. 55903 58. 476
3. 52632 2. 06244 58. 487
1. 52117 . 88972 58. 489
1, 22299 . 71523 58. 482
2. 28445 1. 33603 58. 484
3. 25404 1. 90337 58. 493
3. 37078 1. 97133 58. 483
7. 76488 4, 54082 58. 479
2. 08764. 1, 22114 58. 494
2. 80141 1. 63859 58. 492
3. 24328 1. 89681 58. 484
4, 97475 2. 91262 58. 489
3. 04036 1. 77838 58. 492
4. 31046 2. 52133 58. 493
3. 21201 1. 29381 58. 490
2. 70368 1. 58135 58. 489
3. 60658 2. 10934 58. 486
2. 63037 1. 53835 ; 58. 484
3. 41668 1. 99808 58. 480
3. 49940 2. 04675 58. 489
3. 86668 2. 26145 58. 486
3. 40202 1. 98970 58. 486
3. 20661 1. 87533 58. 483
3. 26386 1. 90909 58. 492
6. 73833 3. 94031 58. 476
7, 37889 4, 31643 58. 497

Mean, 58. 4868-+. 0007
Hence, W = 184.14.
1 The actual figures are as follows:

19.5700 grm. WCls gave 42.4127 grm. AgCl.
10.4326 grm, W Cl. gave 4.8374 grm. tungsten.
284 MEMOIRS NATIONAL ACADEMY OF. SCIENCES. [Vou. XVL

The syntheses of WO, by Smith and Exner give W=184.007+.0079. If we substitute
that value in the hexchloride ratio we can compute an independent figure for the atomic weight
of chlorine, namely, Cl=35.454+.0028. This figure is fairly good, and furnishes additional
corroboration of Smith and Exner’s determinations.

Smith and Exner also made a series of determinations of Taylor’s ratio between WO, and
CO,. Their figures, with the ratio added, are as follows:

Weight WO3. Weight CO,. Ratio.
2, 45645 0. 46775 19, 041
2. 72292 . 51785 18. 974
3. 32953 . 63288 19. 008
3. 97620 . 15473 18. 981
3. 44944 . 65489 18. 985
3. 41273 . 64796 18. 986
6. 10309 1. 16087 19.021.
6. 39735 1. 21644 19.015 ‘°°
2. 17450 . 41332 19.008 ‘
1. 57903 . 29966 18.977

Mean, 18. 9996+. 0046 ia

Hence, W= 183. 60. .

Combined with Taylor’s mean, 19.0125+.0034, the general mean is 19.0073 +.0027. This
ratio, however, is affected by constant errors, as Smith and Exner have shown. There is not
only a possibility of action of the sodium carbonate upon the glass bulb, but also a loss due to
slight decomposition of the carbonate itself at the temperature employed i in the experiments.
Smith and Exner therefore discard the method as too inaccurate.

The work done by Smith and Desi! probably ought to be considered in connection with
that of Pennington and Smith on the trioxide. Smith and Desi started with tungsten trioxide,
freed from molybdenum by means of gaseous hydrochloric acid. This material was reduced in
astream of carefully purified hydrogen, and the water formed was Collected in a calcium chloride
tube and weighed. To the results found I add the percentage of water obtained from 100 parts
of WO,. Vacuum weights are given:

Wo. #,0. Per cent HO.
0. 983024 0. 22834 23. 228
. 998424 . 23189 23. 226
1. 008074 . 23409 23. 221
. 911974 . 21184 23. 229
. 997974 . 23179 23, 226
1. 007024 »,23389 23. 226

Mean, 23.226 .0008

Hence, W=184.70. This method is also criticized by Smith and Exner, and rejected.

Still another method for determining the atomic weight of tungsten was tested by Thomas,
also in Smith’s laboratory. Sodium tungstate, Na,WO,.2H,O, was dehydrated between 180°
and 200°, and the percentage of water so determined. In this series of experiments the tung-
state contained traces of carbonate and silicate. With purer material other determinations were
made between 268° and 295°, and these were divided by the author into two series. I give
below the percentage of water computed from Thomas’s weights:

Preliminary. First series. Second series.
10. 920 10. 895 10. 918
10. 919 10. 886 10. 971
10, 941 10. 900 10, 800
> 10. 931 10. 894 10, 926
10. 937 10. 861 10. 860
10. 929 10. 891 —
10. 926 10. 878 Mean, 10.895 + .0200
10. 945 —
10. 924 Mean, 10.886 + .0034
10. 935

Mean, +10.931.0020

1 Read before Amer. Phil. Soc., Nov. 2, 1804.
a Journ. Amer. Chem. Soc, 21, 373, . 1899.’ Thomas cites some work on tungsten trioxide, but his figures appear in Hardin’s series,

 
No. 8.] ATOMIC WEIGHTS—CLARKE. 285

The general mean of the three series is 10.919+.0017. Hence W=183.95, but with a
very wide range of values in the individual experiments, This method, again, is rejected by the
author himself as unsuited to exact atomic weight determinations.

The ratios, good and bad, rejecting nothing, from which to calculate the atomic waighit af
tungsten are now as follows:

(1) WO, : W :: 100: 79.3706 +.00034
(2) BaW,O,,. oH, O: 9H,O :: 100 : 13.0368 + .0060
(8) WO, :3H,0 ::'100 : 23.226 + .0008
(4) Am,Fe(SO,),. 6,0 : FeWO, :: 0.0365457 +. 0000012 : 0.0283549 + .0000115.
(5) 2AgCl : Ag, WO, :: 100 : 161.645 + .012
(6) WCI, : 6Cl :: 100 : 53.621 +.0074
(7) WCl, : WO, :: 100 : 58.4868 + .0007
(8) WO, : CO, :: 100 : 19.0073 + .0027
(9) Na,WO,.2H,O°: 20,0 : 21,0 :: 100 : 10.919+.0017

The values to use in reducing these ratios are:

Ag=107.880-b. 00016 C = 12. 0025+. 00019

Cl =35. 4605-4. 00011, . Na=23. 0029+. 00012

S = 32. 0650+. 00012 Ba=137, 358+ . 0023

N =14, 0081+. .00008 Fe=55. 880 + .0012

H=1. 0078+. 00001
Hence,
From ratio—

Bootes ites Ree iit tierra tl aks iiss W=183. 503-4. 0329
Dyadic ok oe ered ne conias me eat Thee ida 183. 645+. 0345
Dear Aakers vay Bethan one Rh oo cc 183. 949+. 0514
Bicone cial ane eae RE oe AAMT S22 ac 184, 027+. 0388
2 sd oars al 2O ti ane Samet 184, 060+. 1172
Onna Sade oe emretn to Packehies oust cree 184, 136+. 0067
A sera seater hee ed Oe cel UN Ls aha atta? 184, 401+. 1236
Bag iN nset edt Atanas Alea acetals her NE Rl aS 184, 678+. 0031
Beedle he Rls Se neta hu ead te 184. 700-£. 0080
General mean..........-.-.2...22.2---2--- _ W=184 582+. 0026

This combination is evidently of very little significance. It includes data which are con-
fessedly defective, and which do not tend to compensation of errors. The abnormally high
value derived from ratio 1, which dominates the combination, is due to the excessive weight
given to the determinations by Pennington and Smith, which Smith himself has discarded. If,
in place of ratio 1 we take the determinations of Smith and Exner alone, namely, WO,: W::100:
79,3169 + .0007, we have the more trustworthy value, w= 184.075+.0064. This, combined
with the value from ratio 7, also due to Smith and Exner, gives a general mean of W =184.101 yt
.0049. This seems to be the most probable value now available, and it is checked by the fact,
already pointed out, that the two ratios of Smith and Exner, combined, give a good value for

the atomic weight of chlorine. .
URANIUM.

The earlier attempts to determine the atomic weight of uranium were all vitiated by the
erroneous supposition that uranous oxide was really the metal. The supposition, of course,
does not affect the weighings and analytical data which were obtained, although these, from
their discordance with each other and with later and better results, have now only a historical
value.

For present purposes the determinations made by Berzelius,* by Arfvedson,? and by
Marchand * may be left quite out of account. Berzelius employed various methods, while the
others relied upon estimating the percentage of oxygen lost upon the reduction of U,O, to
UO,. Rammelsberg’s ‘ results also, although very suggestive, need no full discussion... He

 

1 Schweigg. Journ., 22,336. 1818. Poggend. Annalen, 1, 359. 1825.

2 Poggend, Annalen, 1, 245. Berz. Jahr., 3,120, 1822.

3 Journ. prakt. Chem., 23,497. 1841.

4 Poggend. Annalen, 55, 318, 1842; 56, 125, 1842; 59, 9, 1843; 66, 91, 1845. Journ. prakt. Chem., 29, 324.
286 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou, XVI.

analyzed the green chloride, UCl,; effected the synthesis of uranyl sulphate from uranous
oxide; determined the amount of residue left upon the ignition of the sodio and bario-uranic
acetates; estimated the quantity of magnesium uranate formed from a known weight of UO,,
and attempted also to fix the ratio between the green and the black oxides. His figures vary
so widely that they could count for little in the establishing of any general mean; and, more-
over, they lead to estimates of the atomic weight which are mostly below the true value. For
instance, 12 lots of U,O, from several different sources were reduced to UO, by heating in
hydrogen. The percentages of loss varied from 3.83 to 4.67, the mean being 4.121. These
figures give values for the atomic weight of uranium ranging from 185.74 to 235.84, or, in mean,
216.17. Such discordance is due partly to impurity in some of the material studied, and illus-
trates the difficulties inherent in the problem to be solved. Some of the uranoso-uranic oxide
was prepared by calcining the oxalate and retained an admixture of carbon. Many such:
points were worked up by Rammelsberg with much care, so that his papers should be scrupu-
lously studied by any chemist: who contemplates a redetermination of the atomic weight of
uranium.

In 1841 and 1842 Peligot published some data’ showing that the atomic weight of
uranium must be somewhere near 240. A few years later the same chemist published fuller
data concerning the constant in question, but in the time intervening between his earlier and
his final researches other determinations were made by Ebelmen and by Wertheim. These
investigations we may properly discuss in chronological order. For present purposes the early
work of Peligot may be dismissed as only preliminary in character. It showed that what
had been previously regarded as metallic uranium was in reality an oxide, but gave figures
for the atomic weight of the metal which were merely approximations.

Ebelmen’s ? determinations of the atomic weight of uranium were based upon analyses of
uranic oxalate. This salt was dried at 100°, and then, in weighed amount, ignited in hydrogen.
The residual uranous oxide was weighed, and in some cases converted into U,O, by heating
in oxygen. The following weights are reduced to a vacuum:

10.1644 grm. oxalate gave 7.2939 grm. UO,

12.9985 9.3312 Gain on axidation, -3685.
11.8007 a 8.4690 a .3275.
9.9923 sa 7.1731 . ee -2812.
11.0887 a 7.9610 e ss 3105,
10.0830 7.2389 “
6.7940 ne 4.8766 ee
16.0594 “ 11.5290 e st 4531,

Reducing these figures to percentages, we may present the results in two columns. Col-
umn A gives the percentages of UO, in the oxalate, while B represents the amount of U,0,
formed from 100 parts of UO,:

 

A, B.
71.9240 ee
71. 787 103. 949
71. 767 103. 867
71. 621 103. 920
71.794 103. $00
71. 793 acacniabatele
T1.778 ee
71. 790 103. 930
Mean, 71. 782+. 019 Mean, 103. 913-b. 009
Hence U = 237.70. Hence U = 240.25

Wertheim’s * experiments were even simpler in character than those of Ebelmen. Sodio-
uranic acetate, carefully dried at 200°, was ignited, leaving the following percentages of sodium

uranate:

 

1 Compt. Rend., 12, 735. 1841. Ann, Chim. Phys. (3), 55, 1842. 8 Journ. prakt. Chem., 29, 209. 1843,
3 Journ. prakt, Chem., 27, 385. 1842.
No. 3.] ATOMIC WEIGHTS—CLARKE. 287

67. 51508
67. 54558
67. 50927

Mean, 67. 52331+. 0076

 

Hence U = 239.28.

The final results of Peligot’s! investigations appeared in 1846. Both the oxalate and
the acetate of uranium were studied and subjected to combustion analysis. The oxalate was
scrupulously purified by repeated crystallizations, and thirteen analyses, representing different
fractions, were made. Seven of these gave imperfect results, due to incomplete purification
of the material; six only, from the later crystallizations, need to be considered. In these the
uranium was weighed as U,O,, and the carbon as CO,. From the ratio between the CO, and
U;0, the atomic weight of uranium may be calculated without involving any error due to
traces of moisture possibly present in the oxalate. I subjoin Peligot’s weighings, and give, in
the third column, the U,O, proportional to 100 parts of CO,:

CO,. U;303. Ratio.
1. 456 4, 649 319. 299
1. 369 4, 412 322. 279
2, 209 7. 084 320. 688
1.019 3. 279 321. 786
1. 069 3. 447 . 322. 461
1. 052 3. 389 322, 143

 

Mean, 321, 443-4. 388
Hence U = 240.22.

From the acetate, UO,(C,H,O,),.2H,O, the following percentages of U,O, were obtained:

Acetate. U;035- Per cent.
5. 061 3. 354 66. 2715
4. 601 3.057 66. 4421
1. 869 1. 238 66. 2386
3. 817 2. 541 66. 5706

10. 182 6.757 66. 3622
4, 393 2. 920 66. 4694
2. 868 1. 897 66. 1437

 

Mean, 66. 3569+. 038

Hence, U = 239.73.

The acetate also yielded the subjoined percentages of carbon and of water. Assuming
that the figures for carbon were calculated from known weights of dioxide, with C=12 and
O=16, I have added a third column, in which the carbon percentages are converted into pér-
centages of CO,:

 

 

H,0. C. C0;.

21. 60 11.27 41. 323

21.16 11.30 | 41. 433

21. 10 11.30 41: 433

21. 20 11.10 40. 700
Mean, 21. 265+. 187 Mean, 11. 24 Mean, 41. 222+. 092

From these data we get the following values for the molecular weight of uranyl acetate:

From percentage of UgOg.....-.----2---- eee ee eee eee ee e+ 425, 8274 £1678

From percentage of CO,....-....-.22-02- 2c e cece eee cence 426. 993+ . 9530

From percentage of H,O.......... sseeeweae dé aweeume seis 2 <u 423. 60343. 7250
General mean..........--. We dandiade pelea eames eee 425. 861+ - 1651

Hence, U =239.77.
In the posthumous paper of Zimmermann, edited by Kriss and Alibegoff,? the atomic weight
of uranium is determined by two methods. First, UO,, prepared by several methods, is con-

 

1 Compt. Rend., 22, 487. 1846. 3 Ann, Chem., 232,299. 1886.
161566°—20——_19
288 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI,

verted into U,O, by heating in oxygen. To begin with, U,O, was prepared, and reduced to UO,
by ignition in hydrogen. When the reduction takes place at moderate temperatures, the UO,
is somewhat pyrophoric, but if the operation is performed over the blast lamp this difficulty is
avoided. After weighing the UO,, the oxidation is effected, and the gain in weight is observed.
The preliminary U,O, was derived from the following sources: A, from uranium tetroxide; B,
from the oxalate; C, from uranyl nitrate; D, by precipitation with mercuric oxide. The full
data, lettered as indicated above, are subjoined:

U0,. U;03. Per cent Uranate.
8. 9363 9. 2872 3. 927
y 7. 9659 8. 2789 3. 929
12. 4385 12. 9270 3. 927
12. 8855 13. 3913 3. 925
| 5. 7089 5. 9331 3. 927
9. 6270 10. 0051 3. 928
13. 1855 13. 7036 3. 929
9. 9973 10. 3901 3. 929
D. (te. 8996 16. 5242 3. 928
7. 4326 7. T2A5 3. 927

Mean, 3. 9276+. 0003
Ebelmen found, 3. 913+. 009

General mean, 3. 9276+. 0003

In short, Ebelmen’s mean vanishes when combined with Zimmermann’s. From Zimmer-
mann’s mean U = 239.58.

Zimmermann’s second method was essentially that of Wertheim, namely, the ignition of
the double acetate UO, (C,H,O,),.NaC,H,O,, the residue being sodium uranate, Na,U,O,.

Double Acetate. Uranate. Per cent Uranate.
4, 272984 2. 886696 67. 557
5. 272094 3: 560770 67. 540
2. 912283 1. 976428 67. 556
3. 181571 2. 149309 67. 555

 

Mean, 67. 552+. 0027
Werthelm found, 67. 523+. 0076

General mean, 67. 549-+. 0025
From Zimmermann’s figures U = 239.69.

_ An entirely different method for determining the atomic weight of uranium was adopted
by Aloy.! Pure uranyl nitrate was ignited in a suitable apparatus, and the nitrogen evolved was
collected and measured. The residual green oxide of uranium was reduced to uranous oxide,
which was weighed. From this weight and that of the nitrogen, as computed from its volume
the atomic weight of the metal was calculated. Unfortunately, Aloy gives only the volumes of
gas and the corresponding atomic weight, but not the weight of the oxide. His data, therefore,
as published, are radically defective. Assuming N=14.04, Aloy gives the following values for
uranium:

239. 3 239. 4
239. 4 239. 5
239. 6 239. 4
239. 3 239. 4

. Mean, 239, 412+. 0235
If N=14.0081, this reduces to U = 238.87.

The important memoir by Richards and Merigold ” begins with a careful criticism of former
determinations. In regard to Aloy’s work, they show that the residual oxide probably con-
tained some unexpelled nitrogen, and they also point out the difficulty of exactly measuring
small volumes of gas. Their own work was based upon careful analyses of uranous bromide

 

1 Ann, Chim. Phys. (7), 24, 418. Preliminary paper in Compt. Rend., 182,551. 1901. 2 Proc. Amer. Acad., 87, 365, 1902.
No. 3.] ATOMIC WEIGHTS—CLARKE, 289

by the best established methods, and their results, with vacuum weights, areas follows. First,
analyses to determine the ratio 4AgBr: UBr,:

Preliminary Series.

Weight UBr,. Weight AgBr. Ratio.
2. 2058 2. 9699 74, 272
1, 4418 1, 9401 74, 316
1, 4050 1. 8910 74, 299
1, 1749 1. 5818 74. 276
Mean, 74. 291+. 0070
Second Series.
Weight UBr,. Weight AgBr. Ratto.
1, 7999 2. 4226 74, 296
1, 0662 1, 4352 74, 290
1. 8551 2. 4967 74. 302

Mean, 74. 296+. 0029

General mean of both series, 74.295+.0027. Hence, U=238.431.
Second, measurements of the ratio 4Ag : UBr,:

Weight UBr,. Weight Ag. Ratio.
1. 7999 1, 3918 129, 322
1. 0662 - 8245 129. 315
1. 8551 1, 4342 129. 347

Mean, 129. 328+. 0066
Hence, U = 239.395.

The two ratios, combined, give the cross rat.o Ag : Br :: 100 : 74.074.
Oechsner de Coninck,' in order to establish the molecular weight of uranyl oxide, reduced
UO,Br, by heating. His results were as follows:

U0,Br,. U0... Per cent U0..
0. 737 _ 0. 466 63, 229
- 900 - 566 62, 889
. 720 « 452 62, 778
. 818 519 63, 447
1, 080 - 681 63. 056

Mean, 63. 080-L. 0805
Hence, U= 241.1.

In a second brief paper? he gives three reductions of the chloride, UO,Cl,, by heating in
hydrogen:

U0,Ch. U0. Per cent U0,
0. 523 0, 414 79. 159
. 5763 . 456 79. 098
1, 048 - 830 79, 198

Mean, 79. 152+. 0197
Hence, U = 237.24.
In 1911 and 1912 Oechsner de Coninck * published four more notes on the atomic weight
of uranium, which can be summarized very briefly. First, there is a new series of reductions

of UOCI, to UO,.

U0,Cl,. U0,. Per cent U0.
0. 512 0. 4053 79. 160
1. 022 . 808 79. 044
. 358 - 283 79. 050
. 092 . 0727 79. 022
277 . 2192 79. 134

Mean, 79. 082+. 0184
Hence, U= 236.10.

1 Bull. Acad. Belg., Classe des Sciences, 1907, 1041.

2 Tbid., 1908, 163. In another paper, Compt. Rend., 147, 1477, Oechsner de Coninck describes attempts to determine the atomic weight of chlo-
rine by reductions of UO.Ch. His results have no value and need not be considered here.

4Compt. Rend., 152, 711, 1179; 153, 63; 155, 1511.

 
290 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

Second. Uranic monohydrate was similarly reduced to UO,.

U0,.H,0. U0,. Per cent UO.
0. 5326 0. 4735 88, 903
. 5865 . 5215 88, 917
. 4263 . 3788 88. 858
. 4613 . 410 88. 879
. 4168 .370 88. 878

Mean, 88. 887+. 0070
Hence, U=240.07.
Third. Reduction of uranic dihydrate to UO,,

00;.2H,0. 00... Per cent UO,.
0. 6555 0. 55 83. 905
- 5483 - 46 83. 896
« 3658 - 3066 83. 816
« 4528 - 380 83. 922
- 5550 - 465 : 83. 784

Mean, 83. 865+. 0183
Hence, U= 238.45.
Fourth. Calcination of uranium oxalate to UO,.

U0,C20,. U0,. Per cent U0y.

0. 3300 0. 2490 ' 75, 455
. 2639 . 1991 75. 445

. 2903 . 2190 75. 439

. 2900 . 2188 75. 448

. 1453 . 1096 75. 430

. 3299 . 2489 75. 447

. 3961 . 2989 75. 461

Mean, 75. 446+. 0026

Hence, oe 238.41. ty

These determinations by Oechsner de Coninck are evidently of no real value. They are
given without adequate details as to purity of materials or methods of manipulation, the weights
are carried in some cases only to the nearest milligramme, and with no attempt at vacuum
corrections. The figures are given here as part of the published record, but only the last one
will be utilized in the final reductions.

The determinations by Lebeau! are based upon the careful heating of uranyl nitrate,
UO,(NO,),.2H,O, and its reduction to UO,. His meets follow, with a percentage column

added.

Nitrate. U0,. Per cent U0,.
0. 9657 0. 6067 62. 825 :
1. 6038 1..0077 62. 832
1, 8226 1. 1451 62. 828
2. 1934 1. 3783 62. 839
2. 7473 1. 7262 62. 833
2, 9402 2. 4755 62. 827

Mean, 62. 831+. 0014

Hence U = 238.55.

The most elaborate determinations of the atomic weight of uranium were made by Hénig-
schmid; first by himself alone, and later in association with Miss Horovitz. In both cases
uranium, bromide was analyzed by the standard method, with all the precautions to insure
accuracy that have been developed in recent times. In his first memoir ? two series of analyses
are given, as follows:

Series 1.

UBr,. Ag. AgBr. Ag ratio, AgBr ratio.
3.4006 — weaeeee 4.579830 = .....-. 74, 248
3.91401  — ....... 5.27180 a an ee ee 74. 246
4. 52766 3. 50286 6. 09741 129. 255 74, 249
5. 13497 3. 97256 6. 91566 129, 261 74, 251

 

1 Compt. Rend., 165, 163. 1912. 2 Zeitschr. Elektrochem., 20, 452. 1914.
No. 8.] ATOMIC WEIGHTS—CLARKE.
OBr,. Ag. AgBr. Ag ratio.
S788 avs see 5.02586 eee eee
4 68875 | 3. 62849 6. 31652 129, 249
3. 28480 2. 54131 4, 42400 129. 256
3. 91067 3. 02567 5. 26715 129. 249
4, 16254 8. 22038 etbe gs 129. 246

Mean, 129. 253+. 0015

From the Ag ratio, U=238.07.
From the AgBr ratio, U= 238.08.

Serves 2. ,
UBr,. Ag. AgBr.
2.89549 ws se ee 3. 89900
2.82823 | —— ......- 3. 80830
3.95050 — ....... 5. 31979
8.57277) cee ewe 4, 81110
3. 91948 3. 03190 . 5. 27803
3. 77074 2. 91689 5. 07761
4, 31113 3. 33505 5. 80562
8.333250 — aanwade 4, 48823
5. 14622 3. 98089 6. 93002
5. 05193 38. 90789 6. 80279
5. 33502 4, 12698 7. 18387
4. 84910 8. 75106 6. 52962
5. 24735 4, 05913 7. 06581
4. 56095 3, 52825 6. 14204

Ag ratio.

129, 275
129. 268
129, 268
129. 273
129. 275
129. 272
129. 273
129, 273
129. 269

AgBr ratio.
74, 247
74, 246
74, 249
74, 246

74, 248+. 00025

AgBr ratio.
74, 262
74. 265
74. 260
74, 259
74. 260
74, 262
74. 258
74. 266
74. 260
74. 263
74. 264
74, 263
74, 264
74, 268

Mean, 129. 2724..0006. 74. 2617+ .00014.

From the Ag ratio, U = 238.15.
From the AgBr ratio, U = 238.18.

291

In their joint memoir Hénigschmid and Horovitz ' give two more series of analyses of the
bromide. In the first series, before analysis, the bromide was fused in an atmosphere of bro-

mine vapor; in the second series it was fused in nitrogen.

Mean, 129. 245+ .0013

Series 1.
UBr,. Ag. AgBr.
4. 65837 3. 60418 6. 27409
3. 99404 3. 09028 5. 37994
4. 59078 3. 55214 6. 18353
4. 43275 3. 42969 5. 97038
From the Ag ratio, U = 238.04.
From the AgBr ratio, U=238.03.
Serves 2.
UBr,. Ag. AgBr.
5. 28043 4. 08466 7. 11092
5. 36124 4. 14755 7. 22023
5. 90561 4, 56818 7. 95227
38. 14043 2. 42929 4, 22884
4.40397 © 3. 40682 5. 93036

Mean, 129. 270+ .0020

From the Ag ratio, U=238.14.
From the AgBr ratio, U= 238.16.

Ag ratio.
129. 249
129. 245
129. 240
129. 246

Ag ratio.
129. 275
129. 263
129.277
129. 273
129. 263

AgBr ratio.
74, 247
74, 239
74, 242
74, 240
74, 242+ 0012.

AgBr ratio.
7A, 258
74, 253
74, 262
74. 262
74, 258
74, 259 +0012

These series combine with those of Richards and Merigold as follows:

Honigschmid 2......----------------- +2222 eee
Honigschmid and Horovitz 1.........----.-------
Honigschmid and Horovitz 2..........-----------

General mean.......--.-------+--++-e+e eee

129.328 + .0066
129.253 + .0015
129.272 + .0006
129.245 -+ .0013
129.270 + .0020
129.267 + .00049

 

1 Monats, Chem., 37,185. 1916.
292 MEMOIRS NATIONAL ACADEMY OF SCIENCES [Von, XVI,

Ratio 4AgBr : UBr,: : 100: x.

Richards and Merigold...........-.22.-22.0-0ceeeeeeeeeeees 74.295 + .0027
Honigschmid 1..........0...0 00 ccc cece cece cece eeeeeceeeeee 74.248 + .00025
Honigschinid 25.5. 2scd sere vakiesy ve eau avieeacneeneeeteieds 74.2617 + .00014
Hénigschmid and Horovitz 1............. 0.0.02 ceceeeeeeceee 74.242 + .0012
Honigschmid and Horovitz 2.......0....0020eceeceseeccecees 74.259 + .0012
General Means dees esl aiv's's aisttagueshhendeed 74,2582 + .00012

The following ratios relative to the atomic weight of uranium are now available for dis-
cussion:
(1) UO,.C,0, : UO, :: 100 : 75.446 4.0026.
(2) UO,.C,0,.H,O : UO, :: 100 : 71.782 +.019.
(3) 6CO, : U,O, :: 100 : 321.443 +.338.
(4) Molecular weight of uranyl acetate, 425.861 +.1650.
(5) UO,.Na(C,H,0,), : Na,U,O, :: 100 : 67.549 + .0025.
(6) UO,(NO,),.2H,O : UO, :: 100 : 62.831 £.0014.
(7) N.: U :: 14.04 : 239.412 + .0235.
(8) 4Ag : UBr, :: 100 : 129.267 +.00049.
(9) 4AgBr : UBr, :: 100 : 74.2582 +.00012.
(10) 8U0, : U,O, :: 100 : 103.9276 +.0003.

To reduce these ratios we have—

Ag=107.880+ .00016 N=14,0081 + .00008
Br=79.9228 -+ .00015 Na=23.0029+ .00012
C=12.0025-+ .00019 H= 1.00784 .00001
Hence,
From ratio—
Doucnadeda cbt cacpadsteahtecessacedueninennseciuds US2B7. 6994 TOBE
8 ccddswa cae sanahansnoonoseeheWaese Gi sam minrmas 238.132 + .0028
Dusstcacssal yeanresossect een cases tte tet reer 238.146-+ .0018
We en sat Gita GLa nacre aan anc ivee Soneaakeh tents 238.409-+ .0223
Gnas sod ea ate a agea ease e ins 238.546 + .0117
Teese ts Sea Sesl tsa Uo nanead asamp aces sales 238.868 + .0285 :
LO acai criciivg x eip eee ccna dela mnaien aidbiaieacosilalate 239.582-+ .0207
Do iicueneee ai asa eenavan dks ceeeeeeceeeeseee 289.654.0279
AEG asf AU ee Sheena Spee afaik atl 239.7734 .1651
Be ace ltt cia ie a cae nad ee th are Uae 240.218 + 2975
General mean..........--222222eeeeeeeeeeee 238.1554 .0011

This final mean is practically identical with the value derived from the work of Hénig-
schmid and his colleague on uranium bromide. All the other values vanish.

SELENIUM.

The atomic weight of this element was first determined by Berzelius, who, saturating
100 parts of selenium with chlorine, found that 179 of chloride were produced. Henge, Se=
79.24. Further on these figures will be combined with similar results by Dumas.

We may omit, as unimportant for present purposes, the analyses of alkaline selenates
made by Mitscherlich and Nitzsch,? and pass on to the experiments published by Sacc ° in 1847.
This chemist resorted to a variety of methods, some of which gave good results, while others
were unsatisfactory. First, he sought to establish the exact composition of SeO,, both by
synthesis and by analysis. The former plan, according to which he oxidized pure selenium
by nitric acid, gave poor results; better figures were obtained upon reducing SeO, with ammo-
nium bisulphite and hydrochloric acid, and determining the percentage of selenium set free.

Se0, Se. ‘Per cent.
0. 6800 0. 4828 71, 000
3. 5227 2. 5047 71. 102
4, 4870 3. 1930 71. 161

Mean, 71. 088. 032
Hence, Se = 78.68. a

1 Poggend. Annalen, 8,1. 1826. 2 Poggend. Annalen, 9, 623. 1827, 8 Ann, Chim. Phys, (3), 21, 119.

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 293

In a similar manner Sacc also reduced barium selenite, and weighed the resulting mixture
of barium sulphate and free selenium. This process gave discordant results, and a better
method was found in calcining BaSeO, with sulphuric acid, and estimating the resulting quan-
tity of BaSO,. In the third column I give the amounts of BaSO, equivalent to 100 of BaSeO,.

BaSe0;. BaSO,. Ratio.
0. 5573 0. 4929 88. 444
. 9942 . 8797 88. 383
. 2351 . 2080 88. 473,
- 9747 . 8621 88. 448

 

Mean, 88. 437+. 013
Hence, Se= 78.59.

Still other experiments were made with the selenites of silver and lead; but the figures
were subject to such errors that they need no further discussion here.

A few years after Sacc’s work was published, Erdmann and Marchand made with their
usual care a series of experiments upon the atomic weight under consideration.1 They analyzed
pure mercuric selenide, which had been repeatedly sublimed and was well crystallized. Their
method of manipulation has already been described in the chapter upon mercury. These

percentages of Hg in HgSe were found: one
71. 726
71. 731
71, 741

Mean, 71. 7327+. 003

 

Hence, Se = 79.05.

The next determinations were made by Dumas,’ who returned to the original method of
Berzelius.. Pure selenium was converted by dry chlorine into SeCl,, and from the gain in weight
the ratio between Se and Cl was easily deducible. I include Berzelius’ single determination,
which I have already cited, and give in a third column the quantity of chlorine absorbed by
100 parts of selenium:

Se. Cl. Ratio.
1. 709 3. 049 178. 409
1. 810 3. 219 177. 845
1. 679 3. 003 178. 856
1. 498 2. 688 179. 439
1. 944 3. 468 178. 395
1. 887 3. 382 179. 226
1. 935 3. 452 178. 398

179. 000—Berzelius
Mean, 178. 696-4. 125

 

Dumas’ figures alone give Se = 79.38.

The question may here be properly asked, Whether it would be possible thus to form SeCl,,
and be certain of its absolute purity? A trace of oxychloride, if simultaneously formed, would
increase the apparent atomic weight of selenium. In point of fact, this method gives a higher
value for Se than any of the other processes which have been adopted, and that value has the
largest probable error of any one in the entire series. A glance at the table which summarizes
the discussion at the end of this chapter will render this point sufficiently clear.

Still later, Ekman and Pettersson * investigated several methods for the determination
of this atomic weight, and finally decided upon the two following:

First, pure silver selenite, Ag,SeO,, was ignited, leaving behind metallic silver, which,
however, sometimes retained minute traces of selenium. The data obtained were as follows:

Ag,Se03. Ag. Per cent Ag.
5. 2102 3. 2787 62. 929
5. 9721 3. 7597 62. 954

 

1 Journ, prakt, Chem., 55, 202, 1852.
z Ann, Chem. Pharm., 113, 32. 1860.
8 Ber. Deutsch. chem, Gesell., 9, 1210. 1876. Published in detail by the society at Uprala.
294 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Ag Se03.

7. 2741
7. 5390
6. 9250
7. 3455
6. 9878

Hence, Se= 78.95.
Secondly, a warm ‘aqueous solution of selenious oxide was mixed with HCl and reduced by
a current of SO,. The reduced Se was collected upon a glass filter, dried, and weighed.

‘Se0.,

11. 1760
11, 2453
24. 4729
20. 8444
31. 6913

ee Se = 79.076.
Lenher,! in order to determine the atomic wedi of selenium, studied two of its compounds.
First, silver selenite was heated in a stream of gaseous hydrochloric acid, and so transfomed
In a second series of experiments the silver chloride was afterwards
reduced to metal by heating in hydrogen. Two ratios were thus determined. For conveni-

into silver chloride.

ence I now treat the two series as one.

Ag,
4, 5803
4, 7450
4, 3612
4, 6260
4, 3992

Se.
7. 9573
8. 0053
17. 4232
14, 8383
22. 5600

Per cent Ag.
62.967
62. 939
62. 978
62. 978
62. 955

Mean, 62. 957+. 0048

Per cent Se.
71. 199
71. 185
71. 193
71. 187
71,191

Mean, 71. 191+. 0016

Lenher’s data, with vacuum weights, and with the

corresponding percentages added by myself, are as follows:
Ag,Se0;.

0. 98992
1, 59912
2. 70573
. 26204
- 58078
. 70614
. 80811
. 98396
1, 29685
1. 63103
2, 00162

AgCi.
0. 82715
1. 33600
2. 26087

. 21897

. 48522

. 58999

. 67532

. 82232
1. 08350
1, 36288
1. 67234

From Ag ratio, Se = 79.288.

From AgCl ratio, Se= 79.332.

Secondly, ammonium, bromoselenate was studied. From this salt the selenium was pre-
cipitated by hydroxylamine hydrochloride, and then collected and weighed i in a Gooch cru-
cible. The vacuum weights and percentages of selenium follow:

Am SeBr,.

1. 00059
1, 50153
2. 00059
2. 00126
3. 00125
4, 00216
5. 00218
5.03001

Hence, Se= 79.25.

Ag.

0. 16480
. 36534
. 44417
- 50821
. 61882
. 81562

1. 02588

1. 25884

Se.
0. 13324
. 20022
. 26649
. 26657
. 39958
. 53346
. 66656
. 66998

Per cent Ag. ‘Per cent AgCl.

BoEeees 83. 557
= saitiessiniy 83. 560
GEL 83, 559
62. 891 83, 564.
62. 906 83, 546
62. 901 83. 551
62. 889 83. 568
62. 891 83. 572
62.892 83. 549
62. 898 83. 559
62. 891 83. 549

 

Mean, 62. 895+.0014 83, 558+. 0017

Per cent Se.
13. 3161
18. 3344
13. 3209
13. 3201
13. 3188
13. 3293
18. 3254
18. 3196

Mean, 13. 3224-4. 0017

 

1 Journ, Amer. Chem. Soc., 20, 355.

1898, Thesis, University of Pennsylvania.
No. 3.)

ATOMIC WEIGHTS—CLARKE.

295

Steiner’s ‘ determinations, two in number, were made incidentally to his work on tellu-
rium. Phenyl selenide was burned in a combustion tube, and the carbon dioxide so produced
was weighed. ‘To his figures I add the ratio (C,H,),Se : 12CO, ::100 : 2:

Selenide.
0. 2812
. 5371

CO).

0. 6375
1. 2158

Ratio.
226. 707
226. 365

 

Mean, 226. 536+. 0485

Hence, Se=78.91. This determination is of trifling significance.
Julius Meyer ’ analyzed silver selenite electrolytically. The silver was precipitated from
a solution of the salt in potassium cyanide. With vacuum weights the data are as follows:

AgSe03.

0. 5152

- 5237
1. 6964
1. 8793
2. 1460

Ag.
0. 3241

- 3295
1. 0672
1. 1826
1. 3503

Per cent Ag.

62. 907
62, 915
62. 910
62. 928
62, 922

 

Mean, 62. 9164-4. 0082

From the solutions 0.0002 gramme of silver was recovered, to be added to the sum of the
silver weights given above. This raises the percentage to 62.9193. Hence, Se=79.155.
Combining Meyer’s series with its predecessors we have—

Ekman and Pettersson..............-e-eceeceeeecceeeees 62.957 +.0048
Neenher cake Ok en tal th hel ai cht cole: 62. 895 +. 0014
MCYOP ion... caveidclesd 2 ode comaisiass siine debe eases eee ces 62. 9193+. 0082

General Meats: oscosisstwande ds Seomesedeadaccceds 62. 9003+. 0013

In order to determine the atomic weight of selenium Kuzma and Kfehlik ? analyzed the
dioxide SeO,. This was first sublimed in a current of pure air, then dissolved in hydrochloric
acid, and reduced to selenium by saturating the solution with sulphur dioxide and heating in
a sealed bomb. The weights given are as follows, with a percentage column added.

Se0,. Se. Per cent Se.
0. 44245 0. 31523 71. 246
- 61918 - 44122 71. 259
1. 39106 . 99109 71. 247
. 66740 «47544 71. 238
- 65154 - 46414 71. 237
. 96042 . 68417 71. 237
1. 21088 . 86256 71, 234
. 75468 . 53760 71. 236
. 38577 . 27486 71. 250
1. 51040 1. 07594 71. 235

 

Mean, 71, 2424. 0022
Corrected to a vacuum, 71. 230
Hence, Se = 79.26. ee
The determinations by Jannek and. Meyer * were made by oxidizing selenium by means
of nitric acid. Elaborate precautions were taken to insure purity of materials and also of
the dioxide formed. Their data with vacuum weights, are as follows:

Se. Se0.. Per cent Se.
2. 32307 3. 26249 71. 205
2. 00162 2. 81083 71, 211
2.14178 3. 00789 71. 205
3, 29634 4, 62932 71. 208
2. 10447 2. 95522 71. 212

 

1 Ber. Deutsch. chem. Ges., 34, 570. 1901.
2 Ber. Deutsch. chem. Ges., 35, 1591. 1902.

8 Trans. Bohemian Acad.,19 (2), No. 3. Work done under Brauner’s direction.
4 Zeitsch. anorg. Chem., 83,51. 1913,
“998 ‘ZI6T ‘ed[og “peoy “IInd 1

10000°*8200°L =H 80000" F 1800 FI=N
8200° F009'002=3H, G1000° F 8226 6L=10
$200' F see" ze1=8@ TT000° F 09" s8=10

61000°F 9z00'21=0 91000' 083° L0I=3V

—OABY OM SOI}BI OSdY} CONPel OT,

"9800° F 96T'TS ‘OS*FT JO WYSIoM IeTNOE}OPY (6)
"9840" F 9E9°9ZZ : OOT =: “OOST : S"HD (8)
"L100° F PZZE'ET ? OOT :: OG: “Igeg*urY (2)
"L100" F 8E9°8 : OOT :: 1O8VZ : “OPS8V (9)
"100° F £006°Z9 : OOT :: Vz: "OEY ()
"GZL F969'SL1 : OOT :: TOF: 9S (FH)
"200° F LZEL'TL ? OOL 2: BEL: OSS (€)
“810° F Le7'88 : OOT :: "OSta = "OPS" (2)
‘0100° F Sh0Z'TL : OOT =: 8S * “OPS (T)
SUINIU[OS JO JYSIOM OTUA0}S OT} INdUIOD 07 YOIA WOI] SOl}BI SUIU MOT o1v O1OT],
“JYSIOM OTM0Y8 BY} 07 Pousisss oq
osye ABUL WOM ‘9g00'F SI IqSIOM Ie[NOe[OUL 94} Jo JOLIe efqeqoid eyT, “OSI'6L=9G soueTM

‘96118 =9S"H

SBM PUNO} ONIBA VY, ‘SetjIsuep SuryruMT] Jo poyjour Aq svB oq} Jo YYSteM IB[NdeT;oOuU 94} poynduroo
wey} pues ‘epruefes uesorpAy jo AqiIqtssorduroo ey} poutmieyop soiqeyAg pus syue[Anig

*L¥0000° + 6P6GP' T ‘SeTIe8 OU ge |B Jo UBIO,

996CP ET W6seT ee

O862F “TL 9268h ‘T Z86ZP ‘TL 6962 “T

ZhOSP 'T SE6ZF 'T LLGZP ‘T Tr62P T
+ ‘$ ‘% ¥

{SMOT[OJ SB OIB SIYSIOM OU],
‘SUOT}IPUOD oUTeS OY} JepuNn UeSAXO jo Jez] B JO YYZIOM OY} POUTULIOJOpP Os[e sIOYINV OU],

*60000° F 6P1L9'§ ‘SeTIes Ou0 ge [][e Jo UBe,

OFZL9 'E QPL tt

9129" Zig s OIL9 ‘8 96029 °S

90TL9 “8 S9TLO '€ ZEIL9 ‘s THIL9

S¥CL9 ‘E FLOLO 8 TOTL9 ‘8 82149 &

OZ1z9 ' S1ZL9 S TOTLO ‘ 80TL9 ‘
as ‘ 8 T

' IBMOT[OF SB O18
S}YSIOM OY], “UNNIVA B 0} Poonpol o0M SIYSIOM OY} PUG PosN o10M Seqo[s Ino ‘ornssoid “wHUI
092 PUB ,0 78 ‘UTeANO'T 48 SBF OY} JO JOY] ONO Jo FYUSIOM ON} Jog ‘eptuetes ueSorpAY yo AyIsuep
oY} WOLZ WINTUS[IS JO YYSIOM OTUIOZY OY} JOF onjwA Mey} eALIop , doIqeydg pus syuspAnig

OL00 “FEPOS TL "Orns esr s esses weeur [eeuer)

S100 FOG a ne JoAo{ pus Youuve
G00 "F OB TL OTe STA Tt rap se AMOR pus vuzny
9100 °F T6L ‘TZ “TT SEE ere woss10}}0g Pus USULAY
D0 re Bye ee OCR NY eye eneec terete aveg

:SMOTOF SB OUTGUIOD Mou “Qag UT eg Jo eSBzUOOIEd OY} I0F BBP OU,

‘SPT'6L =9S ‘oouazyT
8100 “F80Z ‘TL ‘WeayT

 

L0Z ‘TL 10669 'F LSPLZ'S
G0 ‘TL E8P1 FP 90286 Z
902 "TL G&6IP ‘8 GLESF Z
01g ‘TL 02908 ‘8 STST6
602 ‘TL GE9LL'S. 0F928 F
"ag 7ua0 Jag “09g ‘ag

‘IAX “10A] ‘SHONGIOS FO ANACGVOV 'TVNOILVN SYLONAN 966
“pL “d “6181 “900 “d0g *mIeYD “WITOL 5 “Spl ‘Go “DENY “WeLAL “ASSUNZ4TS x
“eS8T “eat ‘eg ‘(e) ‘shy “MPO ‘UY e “gest “Gee ‘gz ‘MopeuMY “‘puesdod 1
‘99°6ZI 0} F9'9ZI =O], WOT, SuIBUBI sens oavs sz[Nsed OU], “ool
ULIOF 0} plow o1j1U Aq pozIprxo svA UINIINTIe} sjuouIedxo Jo selIos JSG OY} UY “[Tele}BUr Jo
Aytamd omnsur 0} pure ‘syyStem pus souvpeq Surys0q Jo AvM oY} UI SUOTINVOOId PIStI sNOLTeA TIL
‘poXojdue o10m Jonesy] UOA puB sNTeziog Jo spoyjoul oy, ; SAA 4q dn voeys} sem UINLINTIO4
JO JASIOM o1UI078 oY} Jo yoolqns oy} ‘AIO0T} SJog[epUsP 0} COUEJOJeL JOOITP YIM “6L8T UI
‘6Z1 =9J, punoy ‘srefnornsed ou Apeynjosqe seats oy YOryA Joy poujou w Aq g‘svuNg
"79 LOI = OL, ‘oouoyy
S60" F 80h POT ‘Tee .

 

GSS FOT T06 “69 000 ‘T
929 “POT : 89T0 ‘OL 169 °€
648 ‘POT STI6 69 ¥E6
18S “P9T 8h78 “69 899 ‘9
09% “F9T 946 ‘69 000 ‘2
‘owns ug by “Lg qua lag gar oN

:UUINIOO pity} ey} ut ynd ere
OA Igsy jo soytuenb on} ‘47es oy Jo syred QOT 407 ‘sosdpeue stone] UOA TOIT yo3 om ‘SUIyR]
—noywooyy« 08 pus 1'gOT Ajoatoedser Butog 1g pu SY JO pournsse sonyea oY} ‘opruoIg JeATIS
SB PozVUII}so SBM oUTUIOIG OYy punoduoo sIq} UL “IGysIgeL 38s e[qnop poziy[eysAso [joss
ey} jo sisA[Teue oT} 0} poyLOse OM ,“IonV_T UOA Aq epBUl O1OM SUOTYVUIMLIEJOp JXeU TJ,
"PE'SZI =O ‘eousyy
900°F 240 ‘08 ‘Weey
FE0 08
960 ‘08
L90 ‘08
OPIXOIP of} Ul [e}OUL Jo sesezusoIEd SULMOT[OF OY} O7B[NOTBO UBS 9M GOUT ‘SEs'zOs pues
982108 ‘FLZ°TO8=e], PAIS syuoumtIedxe oyeaedes 9014} OUT, “I1ZT'ZOS=9Z, ‘OOL=O FI ‘orga
WoIy ‘s}[Nsel Teuy sty ATUO poysyqnd oF “[TeyJouL OY} OZ ON[BA B pooNpop FYSIOM UL oseoJOUL
OY} Wolf pus ‘ePIXOIpP OF} 0} UMLIN[[e} peziprxo oy pros oLy1u jo suvou Ag =, ‘sntpeziog Aq epeu
e10M Uolysonb UI yURysUOD OY} Jo SUOTBUTUMIEXep ‘ATUO 94} SIBeA AUBUT JOT pus “4SI OTT,
*POLOAODSIP Oq 0} SUTBUMAL 4[NeF 4B
SI JUeTHLIedxe 10 A100q} JeyeG AA ‘“sesoqjodAy poydesoe A[QUeLIN oY} JepUN e[qISSIUIps ST UBT
JOYS JOQUINU UBOTT B 0} PRO[ SUOT}VUIULIOJOp jseq OY} Jo T[B Inq ‘GZT =A], JO oNTBA B I SUDISSE
poppucyy ‘Ayeoresoeyy, “QZ UY} SsoT PUG OZ UBU) Jo}VeIS YYUSTIOM OTUIOYe Ue SULARY ‘UIPOT
pues AuowUe UseeM40q ST] pfnoys wNLNITIe} ‘ueysAs 48T]} 0} Sutpioooy “wra4sAs orpored oy} 04
SUOTYBEI SI JO JUNOOG UO TANLINI[O} JO WYSIOM OTUM0}B OT} 07 SoYOI}e 4Sot0}UT Ie~NOTIe

 

“WOTHN TIAL

‘uasoyo ATIBryIqie oq FYStUA Yor onpea Aus se AYIIOMISNIY SB oq 0} SUIS YI GOUOPLAG OIOUL
jO }[MBjep UL ‘epruejes UssoIpAY jo AzIsUep OY} WO poaLop OUO oY} IVEU ST ON[BA SIG,

000 GOT GL=HOg: oe ete se se Sait eis esi uvoul [e1euex)

GSCOSPOLE GL = ESI STSS ET POP eee ee SU es nese ¥
LOO CECGE Sv OSS SRNR Ae ree acme ae ene pees 9
SLOD GSC CL oS SSeS ORR as SS See ae a aE Ae g
COLO OES Ce ee ORR RIE REESE L
QEOD AOSEIGL TNS CRS ig sie vey re neds PSRROR AR rigin sini 6
OSOOFSTEGh Rr ee ee ER De UE aS ease? I
CTO Ue kf 07 la A rc Sc g
TOSO* SE ISOBL = ES RS CES SSE SRE ER BOF EI OE Ss eee 8
SBE GEG BLOG, Tos NT ee ORE EE REST Ras si isicins Mt Ses j

—orjel Wor

‘eouexL

166 ‘HAUVIO—-SLHDIAM OINOLV U's ‘oN
298 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI,

Te. TeO;. Per cent Te.
2. 21615 2.77612 79. 828
1. 45313 1. 81542 80. 044
2. 67093 8. 33838 80. 007
4. 77828 5. 95748 80. 207
_2. 65029 3. 31331 79. 989

 

Mean, 80. 015+. 041
Hence, Te= 128.12.
In the second series tellurium was oxidized by aqua regia to TeO,, with results varying
from Te=128.10 to 128.32:

Te. Te0,. Per cent Te.

2. 85011 3. 56158 80. 024
3. 09673 3. 86897 80. 040
_ 5. 09365 6. 36612 80. 012
3. 26604 4, 08064 80. 037

 

Mean, 80. 028+. 004
Hence, Te = 128.22.

By von Hauer’s process, the analyses of TeBr,.2KBr, Wills’ figures give results ranging
from Te = 126.36 to 127.90. Reduced to a common standard, 100 parts of the salt yield the
quantities of AgBr given in the third column:

K, TeBry. _, AgBr. Ratio,
1. 70673 2. 80499 164. 349
1, 75225 2. 88072 164. 398
2. 06938 3. 40739 164. 657
3. 29794 5. 438228 164. 717
2. 46545 ~ 4, 05742 164, 571

 

Mean, 164, 588+. 048
Hence, Te= 127.10.

Combined with von Hauer’s mean, 164.408+.045, this gives a 5 ponand mean of 164. 468
+.0324,

The next determinations in order of time were those of Brauner.’ This chemist tried
various unsuccessful methods for determining the atomic weight of tellurium, among them
being the synthetic preparation of silver, copper, and gold tellurides, and the basic sulphate,
Te,SO,. None of these methods gave sufficiently concordant results, and they were therefore
abandoned. The oxidation of tellurium to dioxide by means of nitric acid was also unsatis-
factory, but a series of oxidations with aqua regia gave data as follows. The third column
contains the percentage. of tellurium in the dioxide:

Te. TeO,. Per cent Te.
. 2. 3092 2.9001 79, 625
2, 8158 3. 5332 79. 681
4.0176 5. 0347 79. 798
3. 1613 3. 9685 79. 660

 

. 8399 1.0526 79. 793
Mean, 79. 711+. 0239

Hence, Te= 125.72.

In a single analysis of the dioxide, by reduction with SO,, 2.5489 grammes TeO, gave
2.0374 of metal. If we give this experiment the weight of one observation in the synthetic
‘series, the Bega es tellurium’ found by it becomes— :
> 79. 932 +. 0534

, Steaee, Te= 127. 46, .

Brauner’s best results were obtained from analyses of tellurium: tetrabromide, prepared
from pure tellurium and pure bromine, and afterwards sublimed in a vacuum. This compound
was titrated with standard solutions of silver, and three series of experiments, made with

 

1 Journ, Chem. Soc., 55, 382. 1889.
No. 3.] ATOMIC WEIGHTS—CLARKE. 299

samples of bromide of different origin, gave results as follows. The TeBr, equivalent to 100.
parts ‘of silver appears in the third column:

First series.

TeBr,. 4Ag. Ratio.
2. 14365 2. 06844 103. 636
1. 76744 1. 70531 103. 643
1. 47655 1. 42477 103. 634
1, 23354 1.19019 103. 642

Second series.

TeBry. 4Ag. , Ratio.
3.07912 2. 97064 103. 651
5. 47446 5. 28157 103. 652
3. 30927 3. 19313 103. 637
7. 26981 7.01414 103. 645
3. 52077 3. 39667 103. 654

Third series.

TeBr,. 4Ag. Ratio.
2. 35650 2. 27363 103. 645
1. 51931 1. 46564 i 103. 662
1. 43985 1. 38942 103. 630

 

Mean of all as one series, 103. 644+. 0018

Hence, Te = 127.56. A reduction of the weighings to a vacuum raises this by 0.07 to 127.63.

Still another series of analyses, made with fractionated material, gave values for tellurium
running up to as high as 137. These experiments led Brauner to believe that he had found in

_ tellurium a higher homologue of that element, a view which he has since abandoned.! Brauner
also made a series of analyses of tellurium dibromide, but the results were unsatisfactory.

In the series of determinations by Gooch and Howland ? an alkaline solution of tellurium
dioxide was oxidized by means of standard solutions of potassium permanganate. This was
added in excess, the excess being measured, after acidification with sulphuric acid, by back
titration with oxalic acid and permanganate. Two series are given, varying in detail, but for
present purposes they | may be treated as one. The ratio TeO, : O :: 100 : 2 is given in the
third column:

TeO, taken. O required. Ratio.
0. 1200 0. 01202 10. 017
. 0783 . 00785 10. 026

. 0931 . 00940 10. 097

- 1100 . 01119 10. 149

. 0904 . 00909 10. 055

- 1065 - 01078 10. 122
~-;0910 - 00915 10. 055
. 0910 . 00910 10. 000
0911 . 00924 10.143

. 0913 . 00915 10. 022

- 0912 . 00915 10. 033

. 0914 - 00923 10. 098

 

Mean, 10. 068+. 0100
Hence, Te = 126.92.

In Staudenmaier’s * determinations of the atomic weight of tellurium, crystallized telluric
acid, H,TeQ,, was the starting point. By careful heating in a glass bulb this compound can be

 

1 Journ. Chem. Soc., 67, 549. 1895.

2 Amer. Journ. Sci., 58, 375. 1894. Some misprints in the original publication have been kindly corrected by Professor Gooch; hence the differ-
ences between these data and the figures formerly given.

8 Zeitsch. anorg. Chem., 10, 189. 1895.
300 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

reduced to TeO,, and by heating in hydrogen, to metal. In the latter case finely divided silver
was added to prevent volatization of tellurium. The telluric acid was fractionally crystallized,
but the different fractions gave fairly constant results. I therefore group Staudenmaier’s
data so as to bring them into series more suitable for the present discussion:

Forst.—H,T 0, to Tc0,.

H,Te0,. Loss in weight. Per cent TeO,,.
1, 7218 0. 5260 69. 451

2. 8402 . 8676 69. 453

4. 0998 1. 2528 69. 442

3. 0916 . 9450 69. 433

1, 1138 . 3405 69, 429

4, 9843 1. 5236 69. 432

4, 6716 1. 4278 69. 437

Mean, 69. 440+. 0024
Hence, Te= 127.16.
Second.—H,TeO, to Te.

H,Te0,. Loss in weight. Per cent Te0,.
1, 2299 0. 5471 55. 517
1. 0175 4526 55. 518
2. 5946 1, 1549 55. 488

Mean, 55. 508+. 0068
Hence, Te= 127.31.
Staudenmaier also gives four reductions of TeO, to Te in presence of finely-divided silver.

The data are as follows:

Te0,. Loss in weight. Per cent Te.
0.9171 0. 1839 79. 948
1. 9721 . 8951 79. 966
2, 4115 . 4835 79. 950
1.0172 . 2041 79. 935

Mean, 79. 950+. 0043

Hence, Te = 127.60.

Chikashige ' resorted to Brauner’s method, giving the ratio between silver and TeBr,. In
all essential particulars the work resembles that of Brauner, except that the tellurium, instead
of being extracted from metallic tellurides, was derived from Japanese native sulphur in which
it exists as an impurity. This difference of origin in the material studied gives the chief inter-
est to the investigation. The data are as follows:

TeBr,. Ag. Ratio.

4, 1812 4, 0348 103. 628
4, 3059 4, 1547 103. 639
4, 5929 4, 4319 103. 633

_ Mean, 108. 633-4. 0023
Brauner found, 103. 644+. 0018
General mean, 103. 640+. 0014
From Chikashige’s mean, Te= 127.52.
Metzner? determined the atomic weight of tellurium by two methods, using tellurium
which had been prepared from the hydride. First, tellurium was treated with sulphuric acid
and converted into the basic sulphate, Te,SO,. I give his weights and also the percentage of

_ tellurium in the compound:

Te. Sulphate. Per cent Te.
790. 2 1245. 0 63. 982
414.3 647.5 63, 985

1098. 3 1717.0 63. 966

Mean, 63. 978+. 0040
Hence, Te= 127.94.

 

1 Journ. Chem, Soc., 69, 881. 1896.
2 Compt. Rend., 126,1716. 1898. Metzner fails to state what his weights mean. Are they milligrammes?
No. 3.) ATOMIC WEIGHTS—CLARKE. 301

Secondly, Metzner prepared tellurium dioxide by hydrolysis of the tetrachloride and
reduced it to tellurium by heating in a current of carbon monoxide. The reduction was effected
in presence of silver, in order to avoid volatilization. His data follow, with the usual percentage
of Te in TeO, stated in the third column:

Te0,. Loss. Per cent Te.

743. 2 148. 8 79. 978
1106. 7 221.3 80. 004

988. 5 197.0 80. 073
1312. 5 262. 5 79. 962

Mean, 80. 004+. 0165
Hence, Te = 128.032.

The determinations by Heberlein' represent three distinct methods, starting with crys-
tallized telluric acid, H,TeO,. First, the acid was treated with hydrochloric acid, by which
chlorine was liberated. The latter was distilled off and collected in a solution of potassium
iodide. Iodine was set free and determined by titration with a tenth normal thiosulphate
solution. If W=the weight of telluric acid, and n the number of cubic centimeters of the
thiosulphate solution, the atomic weight of tellurium is given by the subjoined formula:

20000 W_wo,

 

Te=

The first term on the right of the equation obviously represents the molecular weight of
H,TeO,. The figures are as follows:?

Weight H,Te0,. Vol. thiosulphate, ce. Mol. W. H,Te0,.
0. 22911 20. 00 229. 100
. 5736 50. 02 229, 348
. 4038 35, 21 229, 367
. 4393 38. 30 229, 400
. 32331 28, 22 229, 135

Mean, 229. 270+. 0425
Hence, Te = 127.223.
Secondly, Heberlein employed Staudenmaier’s method of reducing H,TeO, to TeO, by
careful heating in a glass bulb:

Weight acid. Loss. Per cent TeO,.
1. 35236 0. 41431 69. 364
1. 76859 - 54122 69. 398

Mean, 69. 381 -b. 0115
Hence, Te = 126.72.
Finally, tellurium dioxide was reduced to tellurium by heating in a current of hydrogen
in presence of silver. Heberlein’s two experiments are as follows:

Te0;. Loss. Per cent Te.
1. 35908 0. 27353 79. 874
1. 94038 . 39050 79. 875

Mean, 79. 8745-4. 0034

Hence, Te=127.002. Heberlein’s determinations assign low values to the atomic weight
of tellurium.

Steiner’s determination’ of the atomic weight was made by combustion of pheny] telluride
(C,H,),Te. I give his weights and also the ratio C,,H,,Te : 12CO, :: 100: 2:

Weight telluride. Weight CO,. Ratio.
0. 2925 0. 5512 188. 440
. 2559 . 4811 188. 003
. 23065 . 4341 188. 207
. 2140 . 4081 188, 364
. 2578 . 4849 188. 092

Mean, 188, 221-4. 0549

 

1 Inaug. Diss., Basel. Printed at Strassburg, 1898.

2 The formula given by Heberlein probably involves the old atomic weights of chlorine and iodine. With modern atomic weights the value
for Te would be raised. The data as printed are, however, incomplete. A correction would be uncertain, and the probable error of the determi-
nations is so high that the change could exert no appreciable effect upon the final combination of values.

® Ber. Deutsch. chem. Ges., 34,570. 1901.
302 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Hence, Te= 126.43, a very low value. This determination only claims a rough approxi-
mation to the truth, and is not entitled to much consideration.

Pellini,! in his determinations of this atomic weight, returned to the time-honored methods
with the dioxide. First, carefully purified telurium was oxidized by nitric acid. Secondly,
tellurum dioxide was reduced by hydrogen in presence of metallicsilver. The data are as follows:

Oxidation Series.

Weight Te. Weight Te0,. Per cent Te.
1, 0679 1. 3353 79. 968
1, 5469 1, 9354 79. 926
2. 2386 2. 7980 80. 007
2. 4522 3. 0665 79. 967
2. 0977 2. 6239 79. 945
2. 0442 2.5575 79. 929
2. 0434 2, 5556 . 79. 957

Mean, 79. 957+. 0071

Hence, Te=127. 657. oe
Reduction Series.

Weight Te0,. Loss, Per cent Te.
1, 4680 0. 2944 79. 945
1, 9968 . 3993 80. 000
1, 9575 . 3932 79. 913 '

Mean, 79. 953+.0171
Hence, Te=127. 625.
Koethner,? after some preliminary, but inconclusive experiments with tellurium sulphate and

telluric acid, finally resorted to analyses of the basic nitrate, Te,HNO, This compound was
reduced by careful heating to TeO,. In series IT, as given below, the tellurium was purified
by distillation in a vacuum; in series I that precaution was not taken. Weights not reduced
to a vacuum:

Series I.
Nitrate. TeO.. -° Per cent Te0,.
2. 9373 2. 4522 83. 485
2. 7982 2. 3361 83. 486
2. 8554 2. 3840 83. 491
Mean, 83. 487+. 0014
Hence, Te= 127.30
Series IT.
Nitrate. Te0,. Per cent TeO,.
5. 30270 4. 42824 83. 510
6. 00600 5. 01543 83. 507
5. 58039 4, 65990 83. 505
28. 66904 23. 94259 83, 513
3. 83859 3. 20560 83. 510
5, 85449 4, 88930 : 83, 514
25. 65029 21. 412146 83. 513

mies * Mean, 83, 5103+. 0009

Hence, Te = 127.57.

In 1902 Scott? published a preliminary note on the atomic weight of tellurium. Analyses
of trimethyl! tellurium iodide and bromide were made, the ratio with silver iodide being deter-
mined in the first case and the titration ratio with silver in the second. I subjoin Scott’s figures,
with the ratio to 100AgI and 100Ag, respectively. Vacuum weights are given:

(CH;,), Tel. AglI. Ratio.
1. 7461 1, 3688 127. 564
6. 6425 5, 20575 127. 570
8, 0628 6, 3181 127. 614

Mean, 127. 583+. 0105

1 Ber. Deutsch. chem. Ges., 34, 3807. 1901. Gazz. Chim. Ital., 32,131. 1902. In Gazz. Chim. Ical. » 83.(2),35, Pellini discusses the possibility of
an admixturein ordinary tellurium, ofan element having a higher atomic weight. ©

2 Liebig’s Annalen, 319, 1. 1901Thesis, Halle, 1901. For a critical discussion of the subject, seo Koethner, Zeitsch. anorg. Chem., 34, 402, 1903,
See also Seubert, Zeitsch. anorg. Chem., 33, 247; and 35, 206.

8 Proc. Chem. Soc. 18,112. 1902.
No. 3.]

- Hence, Te = 127.57.

ATOMIC WEIGHTS—CLARKE. 303

(CH;,)3TeBr. Ag. Ratio.
2, 4294 1, 0373 284, 204
6. 8424 2, 9201 234, 321

Hence, Te= 127. 72.

 

Mean, 234, 263-b. 0391

Gutbier’s ' determinations began with telluric acid, H,TeO,. First the acid was dehydrated
by heating in a stream of dry air, dnd the water was colleched:3 in a calcium chloride tube and

weighed:
HyTeO,. H,0. Per cent H,0.
0. 4937 0. 1162 23, 587
- 9910 . 23835 23. 652

Hence, Te= 127.45.

 

Mean, 23, 550-£0083

Secondly, telluric acid was reduced to tellurium by precipitation with hydrazin hydrate.

Gutbier’ s data are as follows:

Per cent Te.

Hy TeO.. Te.

0. 9380 . 0.5204 55. 480
. 4963 . 2754 55. 491

1. 0485 . 5829 55. 594
. 8865 . 4915 55. 443
. 4339 . 2411 55. 566
. 3492 . 1937 55. 470

Hence, Te= 127. 31.
a general mean of 55. 5079 + .0067.

 

Mean, 55. 507+. 0165

This percentage, combined with that found by Staudenmaier, gives

Finally, tellurium dioxide was reduced to tellurium by the same process:

Te0,. Te. Per cent Te.
0. 1662 0. 13287 79. 946
. 3136 . 2507 79. 942
. 2799 . 2238 79. 957

 

Mean, 79. 948+. 0031

Hence, Te=127.585. All of Gutbier’s weights were reduced to a vacuum.
In a later memoir Gutbier ? gives two more series of reductions of tellurium dioxide. In
series I the oxide was. reduced by hydrogen, and in series II by hydrazin. Vacuum weights

are given:

Series I.
TeO.. Te. Per cent Te.
2. 99688 2. 39585 79. 944
1, 30740 1, 04527 79. 950
2. 04325 1. 63380 79. 955
2. 61725 2. 09249 79. 949
3. 61725 2. 89222 79. 956
| Mean, 79. 951:t. 0015
Hence, Te = 127.609.
Series IT.
Te0,. Te, Per cent Te.
1, 90601 ‘1. 52390 79. 952
1. 03532 . 82784 79. 959
2. 2200 1. 77480 79. 945

_Hence, Te= 127.617.

1 Liebig’s Annalen, 320, 52. 1902.

161566°—20——_20

 

Mean, 79. 952-4. 0081

2 Liebig’s Annalen, 342, 266. 1905.
304 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou. XVI.

In a more extended memoir, which includes the results of the last-mentioned investigation,
Gutbier! gives a series of analyses of basic tellurium nitrate, like those of Koethner. His figures
are as follows, not reduced to a vacuum standard:

Nitrate. Te0,. Per cent TeO,.
4. 70704 3. 92380 83. 360
6. 23210 5. 20285 83. 484
5. 65043 4, 71132 83. 379
2. 86977 2. 39211 83. 355
4. 43213 3. 69833 83. 443
9. 25691 7. 73205 83. 505
7. 09070 5. 91930 83. 481
. 12, 2400 10, 2216 83. 508

Mean, 83. 439+. 0156
Hence, Te= 126.75.

Gallo’s? investigation was an attempt to determine the electrochemical equivalent of tel-
lurium, in terms of silver. Silver and tellurium were thrown down by the same current, but in
different receptacles, and so were directly compared. In the third column I give the ratio 4Ag :
Te::100:2:

Weight Ag. Weight Te. Ratio.
0. 74117 0. 218412 29. 469
1. 03801 . 304514 29, 336
. 91704 . 27256 29. 722
1. 041101 . 307117 29, 499
1. 09064 . 321952 29. 519
1, 16302 . 34582 29, 666
. 968903 . 28646 29, 565
_ 1.518712 . 44767 29, 477
‘" 906561 . 26836 29, 534
. 995511 . 29586 29. 720
. 86596 . 25656 29. 627
1. 11282 . 328318 29, 503

Mean, 29. 553+. 0221

Hence, Te=127.53.. All of Gallo’s weights are on a vacuum basis. Gallo also made a
series of electrolytic analyses of tellurium dioxide, as follows. The precipitation was effected
in a hydrofluoric acid solution:

Weight Te0,. Weight Te. Per cent Te.
0. 4624 0. 3694 79. 888
. 7429 . 5938 79. 930
. 7995 . 6390 79. 925
- 9610 . 7664 79. 750
1. 0043 . 8025 79. 906
1. 9891 1. 5890 79. 885

Mean, 79. 881+. 0220
Hence, Te= 127.053.

Two determinations by Lenher,’ although not next in chronological order, may be con-
veniently inserted here. One was by reduction of TeO,, the other by oxidation of Te:

Te0,. Te. Per cent Te.
0. 85635 0. 6845 79. 932
. 2119 . 1694 79. 943

Mean, 79. 938-4. 0037
Hence, Te= 127.50.

Norris,‘ in the course of an investigation which proved the unity of tellurium as an element,
made a series of atomic weight determinations by Koethner’s method. The element itself was
purified by various modes of fractionation, and different fractions were found to be identical.
The basic nitrate was reduced by heating to TeO,, which was fused before weighing, a pre-

 

1 Sitzungsb. phys. med. Soz. Erlangen, 37,270. 1906. Gutbier regards these determinations as unsatisfactory.

2 Atti Acad. Lincei (5), 14,23. 1905. Also Gazz. Chim. Ital., 35,245. See also Pellini, Gazz. Chim. Ital., 34, 132, on the electrolytic determina-
tion of tellurium.

3 Journ. Amer. Chem. Soc., 30, 741. 1908.

4 Journ. Amer. Chem. Soc., 28, 1675. 1906
- No. 3.) ATOMIC WEIGHTS—CLARKE. 305

caution which eliminated the possibility of contamination by inclosed gases. The uncorrected
data are as follows:

Nitrate. Te0). Per cent TeO,.
2. 28215 1. 90578 83. 508
2. 35429 1. 96615 83, 513
1. 86853 1. 56042 83. 512
1. 77348 1. 48110 83, 514
2. 31048 1. 92938 83. 506
2. 14267 1. 78936 83. 511
2. 35523 1. 96676 83. 506
2. 18860 1. 82780 83. 515
3. 29158 2. 74881 83. 510
3. 27516 1. 89993 83. 508
2. 53164 2.11410 83. 507
2. 01327 1. 68121 83. 506

Mean, 83. 5097+. 0006

A vacuum correction to the weights reduces this mean by 0.0074 to 83.5023. Hence,
Te=127.47.

The investigation by Baker and Bennett! was also intended to determine the definiteness
of tellurium as an element. Different preparations from different sources were studied by
several methods, and all gave sensibly the same atomic weight. The results obtained by two
methods are given in detail, with vacuum weights throughout. First, tellurium dioxide was
heated with sulphur in tubes of glass, the two ends of the tube being packed with silver leaf
to avoid loss of tellurium. Sulphur dioxide was expelled, and from its amount, as measured
by the loss in weight of the apparatus, the atomic weight of tellurium was computed. The
determinations by this method may be arranged in three principal series, representing the
differences in the source of the initial substance, as follows: 1. Fractional crystallization of
telluric acid from barium tellurate. 2. Fractional crystallization of telluric acid produced by
oxidation of the element. 3. Tellurium dioxide prepared from tellurium hydride.

Series I.
Weight Te0,. Loss, SO2, Per cent 80,.
1, 51509 0. 60838 40. 155
1. 09875 . 44074 40. 113
1. 02150 . 40993 40. 130
. 90835 . 36472 40. 152
1. 00702 - 40451 40. 169
1, 01515 . 40733 40. 125
Series II.
Weight TeO,. Loss, SO. Per cent SO,.
1. 56837 0. 62938 40. 130
1. 07852 . 43257 40. 108
1. 72627 . 69246 40. 142
2. 09253 . 83927 40. 108
. 83335 . 33465 40. 157
1, 15372 . 46284 40.117
. 68618 . 67661 40. 127
. 90835 - 36472 40. 152
Series ITI.
Weight Te0,. Loss, SO. Per cent SO.
1, 02217 0. 41050 40. 160
. 80697 . 32392 40. 140
1. 32003 . 52992 40. 145
1, 05207 . 42221 40. 131
1. 37043 . 54969 40, 111
. 95944 . 88011 40. 139

Mean of all as one series, 40. 136-4. 0028

1 Journ. Chem. Soc., 91, 1849. 1907.

 
306 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Hence, Te= 127.62. Several other determinations concordant with these are cited, but
without the detailed weighings.

Baker and Bennett also determined the atomic weight of tellurium by synthesis of the
tetrabromide. Tellurium and bromine were directly combined in an atmosphere of nitrogen,
and the excess of bromine was expelled by a current of nitrogen at a temperature of 50°. The
determinations fall into several series, representing different samples of material, but they are
given here as one series:

Weight Te. Weight TeBr,. Per cent Te,
0, 61273 2. 14933 28. 508
. 56866 1. 99354 28. 525
- 59884 2. 09951 28. 523
- 57894 2. 03040 28. 514
. 54743 1.91899 28. 527
. 33859 1. 18732 28. 517
. 56866 1. 99354 28. 526
- 47643 1. 67025 28. 525
- 56622 1. 98597 28. 511
. 44271 1. 55205 28. 524
. 41671 1. 46177 28. 508
- 50611 1. 77489 28. 515
. 37382 1.31081 28. 519
- 31895 1.11868 28. 512
- 48931 1. 71554 28. 522
. 47156 1. 65404 28. 510
. 40748 1. 42867 28. 523
. 62013 2. 17449 28. 518
. 37382 1.31081 28. 519
- 50822 1. 78207 28. 518
. 12928 . 45354 28. 505
. 42926 1. 50540 28. 515
- 80348 2. 81715 28. 511
- 95309 3. 34193 28. 512

 

Mean, 28. 517+. 0009
Hence, Te = 127.54.

Baker and Bennett furthermore report a few analyses of tellurium tetrachloride which
gave values for Te ranging from 127.58 to 127.64, but without weights or details. These
determinations, therefore, are unavailable for discussion here.

In the determinations by Marckwald' telluric acid was the starting point. This com-
pound was reduced to TeO, by heating, as in several former investigations, with the following
results:

Weight H,Te0.. TeOy. Per cent TeO..
8. 6277 5. 9884 69. 409
12. 2680 8.5135 69. 396
13. 0051 9. 0244 69. 390
8. 6415 5. 9947 69. 371
8. 4588 5. 8696 69. 390
8. 0113 5. 5599 69. 401

 

Mean, 69. 393 +. 0035

Hence, Te=126.81, a figure which falls below the atomic weight of iodine. The error
which suggests itself is the possible retention of water or mother liquor by the telluric acid;
but Marckwald obtained an acid of constant weight after prolonged drying over phosphorus
pentoxide. Still, water may have been retained as an inclosure within the particles of acid,
so enveloped as to be prevented from escaping.

Lenher’s investigations ? had special reference to the homogeneity of tellurium. The
tellurium was obtained from three distinct sources; first, from the telluride ores of Colorado;

 

1 Ber. Deutsch. chem. Ges., 40, 4730. 1907. Fora criticism of Marckwald by Baker, see Chem. News, 97, 209.
3 Journ. Amer. Chem. Soc., 31, 20. 1909. See also Lenher’s figures for TeQ:, previously cited.
No. 3.] ATOMIC WEIGHTS—CLARKE. 307

second, from the residues of an electrolytic copper refinery; and third, from Bohemian material.
From these the double bromide K,TeBr, was prepared, and this, by heating first in chlorine
and afterwards in gaseous hydrochloric acid, was converted into potassium chloride. That
is, the ratio K,TeBr, : 2KCl was measured, all weights being reduced to a vacuum. In the
following table I have treated the three series as one, for the results obtained are sensibly
uniform:

K,TeBrg. KCl. Per cent KCl.
2. 33360 0. 50779 21. 7599
1. 27372 . 27716 21. 7599
141. 47573 . 82111 21. 7594
1. 65715 . 36059 21. 7596
1. 54006 . 33513 21. 7608
1. 82810 . 39778 21. 7592
9] 1. 87342 . 40765 21. 7595
1. 48045 . 32214 21. 7596
2. 24775 . 48911 21. 7600
2. 37899 . 51767 21. 7601
1. 79926 . 39146 21. 7562
. 94102 . 20476 21. 7594
341. 55357 . 33806 21. 7602
1. 95038 42440 21. 7599
1. 73248 . 37698 21. 7596
1. 81923 . 39586 21. 7598

 

1 Mean, 21. 7596+. 00017
Hence, Te = 127.56.

It has already been stated that several of the more important investigations relative to
the atomic weight of tellurium, have had for their purpose the establishment of its homogeneity.
Up to this point all the evidence has gone to show that it is not a mixture of two elements.
Tellurium from widely different sources, as in Lenher’s work, gives one and the same value for
its atomic weight. Fractionations by different methods have also given constant results,
and it seemed as if the question had been definitely settled. Browning and Flint,! however,
have secured evidence upon the other side, which deserves some attention. When tellurium
tetrachloride is mixed with water and hydrolyzed, a large part of it is precipitated as tellurium
dioxide. A part, however, remains in solution, from which it can be thrown down by ammonia
and a slight excess of acetic acid. Carefully purified tellurium was treated by the process thus
briefly suggested, and converted, with all due precautions, into the basic nitrate. The portion
precipitated by hydrolysis gave, on analysis of the nitrate, a mean value of Te=126.53. From
the portion ‘afterwards thrown down the value 128.97 was obtained. Other determinations,
by other methods, gave similar results. The alpha, or first precipitate, gave mean values, in
two additional series, of 126.64 and 126.31. The beta portion, that not precipitated during
hydrolysis of the chloride, gave Te=128.77 and 128.81.

In 1910 a continuation of this research was published by Flint.?, Ten fractionations of his
tellurium were made, and the tenth one gave an atomic weight of 124.32, an astonishingly
low figure. His first (or ‘‘alpha’”’) portion, however, yielded figures which are more nearly
normal. They are—

Nitrate. TeO>. Per cent Teo.
4, 26930 3. 56471 83. 496
3. 66863, 3. 06339 83. 502
3. 17350 2. 64983 83. 499
3. 43498 2. 86827 83. 502
5. 47269 4. 56963 83. 499

 

Mean, 83. 500-+. 0008
Hence, Te = 127.45.

1 Amer. Journ. Sci., (4) 28, 347. 1909. 2 Amer. Journ. Sci., (4) 30, 209. 1910.

 
308 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

One more series of analyses of the nitrate prepared from ordinary tellurium are given by
Flint in a later paper.'. The figures are as follows:

Nitrate. TO» Per cent TeO,.
3. 2059 2. 6766 83. 490
5. 1043 4, 2620 83. 498
4. 6762 3. 9046 83. 499
6. 3670 5. 3160 83. 493

 

Mean, 83. 495-4. 0013
Hence, Te= 127.39.

The work done by Marckwald and Foizik? is very obscurely described. By a volumetric
method TeO, was oxidized by potassium permanganate, but the figures given by the authors
do not show directly how much oxygen was taken up. Neither do they explain how their
values for the atomic weight of tellurium were calculated. Eighteen determinations are given,
in mean Te=127.610+.0592; a good value, but with so large a probable error that it would
count for little or nothing in the final combination with other more concordant figures.

In the same memoir Foizik gives three experiments on the dehydration of telluric acid.
The figures are as follows ae

H,Te0,. TeO,. Per ceni TeO>.
9. 4733 6. 5762 69. 418
9. 3118 6. 4656 69. 434
8. 9910 6. 2415 69. 420

 

Mean, 69. 424+. 0047
Hence, Te = 127.04.

This combines with previous determinations of this ratio as follows:

Staudenmaier: <2. seissssveses aeseeeerexaemee ee 69. 440+. 0024
Heberlein.............. ee vatetcamieweeaee 69. 381+. 0115
Marckwald.........0.-- 00 00e cece cee eee e eee eee 69. 393+. 0035
Hogar sc see hake hh ates Siac ah EN 69. 424+. 0047

General mean............2-.seeeeee reese 69, 424+. 0018

This mean is not of much significance, for Baker and Adlam? have shown that telluric
acid always contains an excess of water, probably enclosed within the crystals. Consequeatly
the analyses give tellurium an atomic weight much too low.

‘Harcourt and Baker‘ hydrolyzed tellurium tetrachloride as was done by Browning and
Flint. Four fractions were then treated by the method of Baker and Bennett; that is, by con-
version of the element into the tetrabromide. The results obtained, for the four distinct
fractions were as follows

Te. Br. Per cent Tein TeBr,.
0. 87822 2. 201038 28. 521
. 59706 1. 49640 28. 520
. 69189 1. 73442 28. 516
. 62732 1, 57254 28, 516
. 58307 1. 46162 28, 516

; Mean, 28. 518+. 0007
Hence, Te = 127.54.
No evidence was obtained to show a separation of tellurium into fractions of different
atomic weight.
In 1912 Pellini * published a considerable number of determinations of the atomic weight
of tellurium, in order to test the claims of Browning and Flint relative to the complexity of
tellurium. The tetrachloride was fractionally hydrolyzed, and the fractions were examined

 

1 Journ. Amer. Chem. So¢., 34, 1325. 1912. See also, p. 1669, a critical paper by Morgan, who failed to produce fractional differences in tellu-
rium like those of Browning and Flint.

2 Ber. Deutsch. Chem. Ges., 43,1710. 1910.

3 Journ. Chem. Soc., 99, 511. 1911.

4 Journ. Chem. Soc., 99, 1311. 1911.

6 Atti Accad. Lincei, (5), 21, 218. 1912.
No. 3.] ATOMIC WEIGHTS—CLARKE. 309

by two methods; first, by conversion of tellurium into the dioxide, and secondly, by the analysis
of the nitrate. By the first method, omitting one synthesis which is evidently misprinted, and
giving all the determinations as one series, Pellini obtained the following results The percent-
ages are calculated by myself, in order to make the data conform with other series.

Te. TeO,. Per cent Te.
1.1167 1. 3965 79. 964
1. 6507 2. 0655 79. 918
1. 0377 1. 2975 79. 977

. 8755 1, 09435 80. 002

- 5285 . 6592 80. 173

- 4366 . 5444 80. 198
1. 4939 1. 8680 79. 973
1. 6037 2. 0062 79. 937
1, 5843 1. 9804 79. 904
.1. 3866 1. 7351 79. 915

 

Mean, 79. 996-4. 0224
Hence, Te = 127.738.

By the nitrate method the following results were obtained:

Nitrate. TeO,. Per cent TeO,.
3. 0255 2. 5268 83. 517
3. 3243, 2. 7758 83. 500
8. 5758 2. 9861 83. 509
1, 3524 1, 1291 83. 491
1, 0218 . 8534 83. 519

 

. Mean, 83. 507+. 0035
Hence, Te = 127.55.

Continuing the investigation, Pellini fractionated his tellurium electrolytically, and then
made two sets of determinations as in the first part of his work. The data are as follows:

Te, Te0,. Per cent Te.
1, 5529 1, 9420 79. 964
- 9966 1. 2462 79. 971
. 8529 1, 0666 79. 965

 

Mean, 79. 967-b. 0014.
Hence, Te= 127.74.

Nitrate. TeO,. Per cent TeO,.
1. 4145 1.1814 83, 521
3. 9674 3. 3128 83. 501
2, 4144 2. 0164 83. 516

 

Main, 83. 513+.0023
Hence, Te = 127.60.

Pellini’s determinations give values ranging from Te=127.32 to 128.01, but with no such
large variations as were found by Browning and Flint.
The two nitrate series combine with previous determinations as follows:

Koethner, 1.......--...--.------ Migdawielss sa atacek ees 83. 487 +.0014
Koethner, 2..........-----.-.--- bate taba Asya aesre boom iaeueesiae ese 83.510 +. 0009
Gutbier:ccccks Sos kt ee avieiee ep ceeesacda sua sereunat 83.439 +.0156
NORTISia2<eetcaeneeies ue dame acchi ees eciaeee aac 83.502 +. 0006
Flint, Vessco ss 50 gjaiialeiSeetle a 45% eaesenauiaawis sy ssutanas 83.500 +. 0008
Flint, 2.......  aletepatureisisinizicrs ais eae Teas eto A aa hel Sat 83.495 +.0013
Pellini, 1...... Pateiisiae Suse iialeis wes a RG cata erannanarcas ements 83. 507° +. 0035
Pellini, 23.5 seco esa sete eee ese cmreueea dane seme aMenwele 83.513 +. 0023
General Mean. 5 5 siscesescee so ies ssn cece eeee sr eese 83. 5025+. 00038

Dudley and Bowers‘ attempted to determine the atomic weight of tellurium by analyses
of the basic nitrate, but obtained values which were so widely and irregularly variant that they

 

1 Journ. Amer. Chem. Soc., 35, 875. 1913.
310 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Von. XVI.

abandoned the method as untrustworthy. The salt seems to be liable to contain impurities,
such as water and tellurium dioxide. The authors finally resorted to the synthesis of the tetra-
bromide, by the method of paler and Bennett, and give the following series of determinations:

Te. TeBr,. Per cent Te.
0. 300558 1. 054251 28. 509
. 199807 - 700947 28. 505
. 22032 . 773048 28. 500
. 158161 . 554717 28. 512
436907 - 1. 532360 28. 512
. 29811 1. 045485 28. 514

 

Mean, 28. 509+. 0010
Hence Te = 127.486.

Combining this series with the previous determinations we have—.,

Baker and Bennetty: iscscc2cwenarnceancnaeae ey eevee ves ae 28. 517+. 0009
Harcourt and Baker........-...-.-0.000-0eceseeecceccceees 28. 518+. 0007
Dudley and Bowers..........--..-0. 2 eee cece ec eee eee eeee 28. 509 +. 0010

Mean, 28. 516+. 0005

Dennis and Anderson! Stouts to divide detiitehvie into substances of different atomic
weight, by means of the fractional distillation of hydrogen telluride, but without success. Tel-
lurium obtained from the several fractions, when converted into dioxide, gave discordant values
for the atomic weight, but the discordance was attributable to defects in the method. They
were not large enough to warrant the assumption that the sought for separation had occurred.
Later in their investigation they made 32 determinations of the atomic weight, by the volu-
metric oxidation of tellurium dioxide by potassium permanganate, and obtained values from
17 fractions, ranging between 127.36 and 127.76, The mean of all is Te=127.49+.0108.
Unfortunately, Dennis and Anderson do not give the actual weights, either of the tellurium
dioxide or of the permanganate used to oxidize it. Their results, therefore, can not be includéd
in the systematic discussion of the more definite ratios.

The determinations of the atomic weight of tellurium by Stachler and Tesch,? were made
by the time-honored method of oxidizing the element to TeO,. Their very concordant data,
with vacuum weights, are as follows:

Te. Te0,. Per cent Te.

4, 46603 5. 58683 79. 9385
6. 19861 7. 75415 79. 9393
6. 61465 8. 27460 79. 9392
5. 76199 7, 20817 79. 9368
6. 03396 7. 54818 79, 9393
6. 91214 8. 64677 79. 9389
6. 66175 8. 333511 79. 9393
5. 87323 7. 34707 79. 9398
6. 18154 7. 73286 79. 9386
6. 04458 7.56145 79. 9394
6. 90661 ‘8, 68959 79, 9414
. 4.91463 6. 14810 79. 9374
- 5.41885 ° 6. 77876 79. 9887

 

Mean, 79. 9390+. 00027
Hence, Te= 127.513.

The percentage of tellurium in the dioxide is now fixed by 18 series of determinations
which, ranged in the order of ascending magnitude may be combined, as usual, into a genera
mean:

Brauner, oxidation...............2- 0. eee eee ee eee eee ees 79.711 +. 0239
Hoberletttsca.02 ses tu scamadecscegsbndvoausd Sooucceiees 79. 8745-4. 0084
Gallo........... eau itech eaay A Aad tie ee) 79.881 +. 0220
Brauner, reduction................cceeceee eee e eee eee es 79. 932 +, 0534.
Whénherisesve vest emewceesicewe ee de 2's pe4nemaa ted de serseiaks 79.938 +. 0037

 

1Journ. Amer. Chem. Soc., 36, 882. 1914. 2 Zeitsch. anorg. allgem. Chem., 98,1. 1916.
No. 3.] -

ATOMIC WEIGHTS—CLARKE.

311

Staehler and Tesch...............2....- iaeniseeides 79.939 +.00027
Gutbiers T9028 eae. icl os wecis,d dice eciie coo eainediedenieeks Soe 79.948 +.0031
Staudenmaier............... Ea Aaa dd inthe eat S 79.950 +. 0043
Gutbier, 1905, 1.........0.00.eeeeeceeeeee aga dnkeitiaes 79.951 -.0015
Gutbier, 1905, 2.............- false wa o's ahapacaine eae os te 79.952 +.0031
Pellini, reduction.................00ececceeeeeeeeeceeee 79.953 +. 0171
Pellini, oxidation................-.- © a teanas isa eeesed .. 79.957 +.0071
Pellind 1912) Lis oscwmanecsemeneqewecses s ioeraimeia aaron as 79.967 +.0014
Pellini, 1912, 2...............2- DACs SD) ALLIS oe 79.996 -.0224
Metzner: ssccuae ee is 2 yaaeeeeees = ho x Seeeen ese alee a 80.004 +. 0165
SWS Ne deteec a Gtr acre erat a ys ata ok ete 80.015 +. 0410
AVE rice pt ore en aca de ead 80.028 +..0040
Berzelitiscs : cstecwoscscvessaeewauccas-os Synecmeaaeee wets .. 80.042 +. 0050
, General mean..........20..0. 2c eee eee ee ee ee ee ee 79,941 +. 00025

In this combination the determinations by Staehler and Tesch outweigh all the others.
The ratios for tellurium are now as follows:

t

(1) TeO, : Te :: 100 : 79.941 + .00025.

(2) TeO, : O :: 100 : 10.068+.0100. -

(3) H,TeO, : 3H,O :: 100 : 23.550 + .0083.
(4) H,TeO, : TeO, :: 100 : 69.434 +.0018.

(5) H,TeO, : Te :: 100 : 55.5079 + .0067.

(6) Molecular weight H,TeO,, 229.270 + .0425—(Heberlein).
(1) 4Ag : Te :: 100 : 29.553 4.0021.

(8) 4Ag :'TeBr, :: 100 : 103.640+.0014.

(9) K,TeBr, : 6AgBr :: 100 : 164.468 + .0324.
(10) TeBr, : Te :: 100 : 28.517 +.0005.
(11) TeO, : SO, :: 100 : 40.136 + .0028.
(12) Te,SO, : Te :: 100 : 63.978 + .0040.
(13) Te,HNO, : TeO, :: 100 : 83.5025 + 00038.
(14) ‘C,,H,.Te : 12CO, :: 100 : 188.221 +.0549.
(15) AgI : C,H,Tel :: 100 : 127.583 +.0105.
(16) Ag : C,H,TeBr :: 100 : 234.263 + .0391.
(17) K,TeBr, : 2KCl :: 100 : 21.7596 +.00017.

To reduce these ratios we have—

Hence,

 

Ag=107. 880 -£. 00016 N=14. 0081-4. 00008

Cl = 35. 4605-4. 00011 C=12. 0025+. 00019

Br= 79. 9228-+. 00015 S=32, 0650-4. 00012

I =126.926 +. 00023 K=89. 0975+. 00011

. H=1. 0078. 00001

From ratio—

AION Ae leicht AA ade ene ceca ahh Boer tee Te=126, 429-4. 0820
Dio cedars poem eetateeen ee ee rner 126. 919+. 1579
Girne tt atean aaa nans catered Pes deere: 127. 044+. 0102
GaP R a ase Rana eaters Atta 127, 223-4. 0425
25 au da cche Mis asctin oes Guiana Saad s ones 127. 313-4. 0247
Laren Ee pein daa itedeboone Ao eaten 127. 396+. 1350
Bet cease eae pian aa edeeeakie ona 127. 451-4. 0583
Oe ascceanmnn AT ane tae Ce temebaeciean). cI Steels
On Ee EOE 8 UNE Bee ake themes seas 127. 527+. 0091
Tie puege viens Gag roccddedacaasanneee saweesveeee | 127. 529+. 0016
0k atreariee back Ge San eu Soemseanmene 127. 580-b. 0024
BB ae ee hence akdal Aa eae wees 127. 546-4. 0062
WF alcaithesnon tex ee ea oe Senate eee 127. 556-4. 0064
VB a cicgcrais dotisern tn gunna ga sat eeeeekiees oe uaa 127. 569+. 0247
ET Licata ese Naps coe ann eine aeass mata ne ate 127. 620-4. 0089
NG sae eee Rola As VO OOS ese Rare oad 127. 7224. 0422
DY ws ceshanenenpuav boyawamusina xi emedaceewiyi es eorsue 127. 936+. 0143
312 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [VoL. XVI.

The atomic weight of tellurium is evidently very close to
Te=127.5,
which rounded-off value can be used for all practical purposes. To combine the determinations

by Marckwald and Foizik, Te=127.610+.0592, and Dennis and Anderson, Te= 127.49 +.0103,
with the others, would produce no appreciable change in the final result.

ADDENDUM.

Since this work was completed some determinations of the atomic weight of tellurium
have been published by Bruylants and Michielsen,* who made some rather complicated analyses
of the hydride TeH,.

Their data may be abbreviated as follows: —

t

H, Te. Te. #0.

1, 95159 1. 91885 0. 27096
. 79827 . 78501 . 11078
- 49887 - 0972 - 06945

As corrected by the authors these figures give the following values for the ratio between
tellurium and oxygen.

Te. 0. At. ct Te.”
1. 92094 "0, 24062 127.73
. 78578 . 09838 127.79 °°
. 49163 . 06167 127. 4—Rejected.

This investigation is to be regarded as preliminary. “Compare the work of Dennis and

Anderson, already cited.
FLUORINE.

The atomic weight of fluorine has been determined by several distinct methods; most com-
monly by the conversion of fluorides into sulphates. There are, however, some exceptions,
which will be considered in due time. ..

Excluding the early results of Davy,? we have to consider, first, the experiments of Berzelius,
Louyet, Dumas, De Luca, and Moissan with reference to the fluorides of calcium, sodium, potas-
sium, barium, and lead.

The ratio between calcium fluoride and sulphate has been aaa by the five inves-
tigators above named, and by one general process. The fluoride is treated with strong sulphuric
acid, the resulting sulphate i is ignited, and the product weighed. In order to insure complete
transformation special precautions are necessary, such, for instance, as repeated treatment with
sulphuric acid, and so on. For details like these the original papers must be consulted.

The first experiments in chronological order are those of Berzelius,*? who operated upon an
artificial calcium fluoride. He found, in three experiments, for one part of fluoride the following
of sulphate: 1. 749

= 1. 750
1.751

Hence, F=18.86. Mean, 1. 750-+. 0004

Louyet’s researches * were much more elaborate than the foregoing. He began with a
remarkably concordant series of results upon fluorspar, in which one gram of the fluoride
yielded from 1.734 to 1.737 of sulphate. At first he regarded these as accurate, but he soon
found that particles of spar had been coated with sulphate, and had therefore escaped action.
In the following series this source of error was guarded against.

Starting with fluorspar, Louyet found of sulphate as follows:

1, 742 1.744
1. 744 1. 7485
1.745 1. 7435

Hence, F = 18.99. Mean, 1. 7487+. 0003

1 Bull. Acad. Belge., 1919, Nos. 2-8, p. 119. 8 Poggend. Annalen, 8,1. 1826.
3 Phil. Trany., 64. 1814. : 4‘Ann, Chim. Phys. (3), 25, 300. 1849.

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 313

A second series, upon artificial fluoride, gave:

1. 748
1.741
1.741

Mean, 1. 7417+. 0004
Hence, F = 19.05.
Dumas! published but one result for calcium fluoride. 0.495 grm. gave 0.864 grm. sul-
phate, the ratio being 1:1.7455. Hence, F= 18.96.
De Luca? worked with a very pure fluor spar, and published the following results. The
ratio between CaSO, and one gramme of CaF, is given in the third column:

CaF, CaSO, Ratio
0. 9305 1. 630 1. 7518
- 836 1, 459 1. 7452
- 502 . 8755 1. 7440
- 6945 1. 7428

Hence, F= 18.95.

If we include Danes s single result with these, we get a mean of 1. 7459 + .0011.

Moissan‘ unfortunately gives no details nor weighings, but merely states that four experi-
ments with calcium fluoride gave values for F ranging from 19.02 to 19.08. ToS he assigned
the value 32.074, and probably Ca was taken as=40. With these data his extreme values as
given may be calculated back into uniformity with the ratio as stated above, becoming—

1, 7444
1. 7410

Mean, 1. 7427 ri

 

Hence, F=19.019. ;
If we assign this equal weight with Berzelius’ series, the data for this ratio combine thus:

DOPZOHUS)js.2ssiictsls! ris aets cletde bes Sale dias y siverneroee ei caren 1. 7500 +. 0004

_ Louyet, first series............. {Sed aWaS oss saieeeeketenees 1. 7437 +. 0003

Louyet, second series. ...---....0--- 2222s sence eee eeee ee 1. 7417+. 0004

De Luca with Dumas............ cei saierae mae aaelede 1, 7459+. 0011

Mo iS88N ...o5s.2ingaserieees aca eemnd be B asdscue. 6:4. 4:Giaters.avereloid waa relaae 1. 7427+. 0004
General mean...........000ceeeeeeeeeeeeeeeeeeeee 1. 7444+. 00018

For the ratio between the two sodium salts we have experiments by Dumas, Louyet and
Moissan. eae to pee one gramme of NaF gives of Na,SO,— Dg
: 1. 686 © + vd
1. 683 :
1, 685

Mean, 1. 6847-4. 0006

 

Hence, F= 19. 16. :
The weighings published by Dumas are as follows:

NaF. Na,S0,. Ratio.
0.777 . 1. 312 1, 689

1. 737 2. 930 1. 687
: Mean, 1. 688+. 0007

 

Hence, F= 19.03.
Moissan says only that five experiments with sodium fluoride gave F=19.04 to 19.08.
This was calculated with Na=23.05 and S=32.074. Hence, reckoning backward, ‘the two

values give for the standard ratio—.
iS ee 1. 6889

1, 6873
Mean, 1. 6881

 

Hence, F= 19.03.

1 Ann, Chem. Pharm., 113, 28. 1860. Compt. Rend., 51, 299. 1860. *Compt. Rend., 111, 570. 1890.

 
314 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI,

Dumas also gives experiments upon potassium fluoride. The quantity of sulphate formed
from one gramme of fluoride is given in the last column:

KF. K,80,. Ratio.
1. 483 2, 225 1. 5002
1. 309 1. 961 1, 4981

 

Mean, 1. 4991 +. 0007
Hence, F = 19.02,
The ratio between barium fluoride and barium sulphate was measured by Louyet and
Moissan. According to Louyet, one gramme of BaF, gives of BaSO,—

1. 332
1. 331
1, 330

Mean, 1. 331+. 0004
Hence, F= 19.01.
Moissan, in five experiments, found F=19.05 to 19.09. Assuming that he put Ba= 137,
and S=32.074 as before, these two extremes become—

1. 3311
1, 3305

Mean, 1. 3308
Hence, F= 19.02.

Giving this equal weight with Louyet’s mean, we get the subjoined combination:

Lovyets.. 2. .occsssetues ys ose ceteeaaiene, waeiweniies soa eee ce 1,331 +. 0004
MOIS88D ): .ec:ncee ccs abt ses ssa NGS ee SEN Rete eseewee Ke se 1, 3308 +. 0004
General Mean... s). J...:ncasctososceeseiedcessecaeceess 1. 3309+. 00028

The experiments with lead fluoride are due to Louyet, and a new method of treatment
was adopted. The salt was fused, powdered, dissolved in nitric acid, and precipitated by dilute
sulphuric acid. The evaporation ‘of the fluid and the ignition of the sulphate was then effected
without transfer. Five grams of fluoride were taken in each operation, yielding of sulphate:

6.179

6.178
6.178

Mean, 6. 1783 +. 0002
Hence, F= 19.12.

In Christensen’s determinations’ we find a method adopted which is radically unlike any-
thing i in the work of his predecessors. He started out with the salt (NH,),MnF,. When
this is added to a mixture, in solution, of potassium iodide and hydrochloric acid, iodine 3 is set
free, and may be titrated with sodium thiosulphate. One molecule of the salt (as written
above) liberates one atom of iodine. In four experiments Christensen obtained the following
data:

Am,MnF, I. Per cent.
3. 1199 2. 12748 68. 191
3. 9190 2. 67020 : 68. 185
3. 5005 2. 38429 68. 113
1, 2727 . 86779 68. 185

Mean, 68. 156+. 0128
Hence, F = 19.040.

Still another method for decent the atomic weight of fluorine was adopted by Julius
Meyer.? Carefully purified calcium oxide was weighed, slaked with water, and then converted
into chloride by means of hydrochloric acid. The chloride solution was then repeatedly evap-
orated with pure hydrofluoric acid. The calcium fluoride so produced was finally ignited to

 

1 Journ. prakt. Chem. (2), 35, 541. Christensen assigns to the salt double the formula here given. 2Zeitch. anorg. Chem., 36, 313. 1903.
No. 3.] ATOMIC WEIGHTS—CLARKE. 315

constant weight. On a vacuum basis his weights were as follows. The third column gives the
ratio CaO : CaF, :: 100 : 2:

Weight CaO. Weight CaF,. Ratio.
6. 1883 8. 6215 139, 320
4, 2736 5. 9548 139, 339
6. 2931 8. 7658 139. 292
5. 7767 8. 0485 139. 327
4. 9836 6, 9426 139, 309

 

Mean, 139. 317+. 0054

Hence, F= 19.026.

In a preliminary investigation McAdam and Smith‘ attempted to determine the atomic
weight of fluorine by the conversion of sodium fluoride into chloride. The conversion was
effected by heating the fluoride in a current of dry, gaseous hydrochloric acid. The determina-
tions gave the following results:

NaF. NaCl. Ratio.
4, 3274 6. 02075 139, 132
3. 39398 4, 72260 189. 146

 

Mean, 139. 139+. 0047

Hence, F = 19.012.

In the section of this recalculation upon silicon the determinations by Germann and Booth
of the density of silicon tetrafluoride are discussed.? From their figures the molecular weight
of the gas becomes Sif,=104.47+.0132. Hence, if Si=28.255+.0025, F=19.054+.0034.
This value. is of slight importance.

_ In their work on the atomic weight of boron, which has already been fully discussed,*
Smith and Van Haagen give data from which they compute the atomic weight of fluorine. These
data furnish determinations of two ratios, as follows:

Ratio Na,B,0O,: 2NaF :: 100: x.

NayB,0;. NoF. Ratio.
0. 69695 0. 29042 41, 670
1. 99197 . 83003 41, 669
1. 60201 . 66757 41. 671
2, 64768 1. 10329 41, 670

 

Mean, 41. 670-4. 0003
Ratio 2NaF : Na,SO, ;: 100 : z.

NaF. Na,S80,. Ratio.
0. 28042 0. 49113 ‘ 1, 69110

. 66757 1, 12889 1. 69101
1. 103829 1, 86597 1. 69132

 

Mean, 1. 6914-4. 00006

Hence, F = 19.002. | 7
This ratio combines with previous determinations of it as follows:

Louyet. .-.---- 2-2-2222 ee eee ee eee eee ences jore-e--- 1. 6847 +. 0006

Moissan. ...--.-.-.---+- 4 jeuesuapueeesvenayesmesie ..-- 1.6881 +. 0007

Smith and Van Haagen.......-.-.-..---- Nsiecaperarahsibeaia Sates 1, 69114+. 00006
General mean......-...----- 6 ayaa 1, 69112+. 000059

In this combination Moissan’s figure is assigned equal weight with that of Dumas.
The ratios from which to compute the atomic weight of fluorine are now—

(1) CaO : CaF, :: 100 : 139.317 + .0054.
(2) CaF, : CaSO, :: 1.0 : 1.74444.00018.
(3) 2NaF : Na,SO, :: 1.0 : 1.69112 + .000059.

1Journ, Amer. Chem. Soc., 34, 592. 1912. 2 See ante, p. —. 8 See ante, pp. —.
316 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou. XVI.

(4) 2KF : K,SO, :: 1.0-: 1.4991 +.0007.

(5) BaF, : BaSO, :: 1.0 : 1.8309 + .00028.
(6) PbF, : PbSO, :: 5.0 : 6.1783 +.0002.
(7) Am,MnF, : I :: 100 : 68.156 +.0128.

(8) NaF : NaCl :: 100 : 139.139 +.0047.

(9) Na,B,O, : 2NaF :: 100 : 41.670+. 0003.
(10) Mol. weight of SiF,, 104.47 +.0132.

To reduce these ratios we have—
Ca= 40. 0902+. 00023 S =382. 0650+. 00012.

Ba=137. 358 +. 0023 ' IT =126, 926+. 00023

Pb=207. 102 +. 00062
Mn= 54,949 +. 00045
Na = 23. 0029-4. 00012
K = 39, 0975+. 00011

Cl=35, 4605+. 00011
N=14, 0081+. 00008
B=10. 912 +. 0007
Si=28. 255 +. 0025

H=1. 0078+. 00001

Hence,

From ratio—

Sd Widow kg Mee oe os Osc EI aE F=18, 931+. 0040
19, 002+. 0015
19, 012-4. 0014
19. 012+. 0007
19, 015L. 0185
19. 024+. 0271
19. 026+. 0038
19. 040+. 0070
19. 054 +. 0034
19, 123-4. 0070

General mean.............-0-.-eeeectee eens F=19. 011+. 00055

 

In this combination the determinations by Smith and Van Haagen outweigh all the others.
Their determinations alone, taken in connection with their work on boron, make F=19.005.
For practical purposes the rounded-off value 19 falls within the limits of uncertainty.

ADDENDUM.

The determination of the atomic weight of fluorine by Moles and Batuecas! appeared
too late for the systematic discussion of their data. Methyl fluoride, CH,F, was taken as a
favorable substance to investigate. Twenty-three determinations were made of the weight of a
normal liter of the gas, giving a corrected average of 1.54542 grammes. From this, the
molecular weight 34.024 is deduced, and F =18. 996, or rounded off, 19. 0.

MANGANESE.

The earliest experiments of Berzelius’ and of Arfvedson? gave values for Mn ranging
between 56 and 57, and therefore need no further consideration here. The first determinations
to be noticed are those of Turner ‘ and a later measurement by Berzelius,> who both determined
gravimetrically the ratio between the chlorides of manganese and silver. The manganese
chloride was fused in a current of dry hydrochloric acid and afterwards precipitated with a
silver solution. I give the MnCl, equivalent to 100 parts of AgCl in the third column:

4, 20775 grm. MnCl,= 9.575 grm. AgCl. 43, 945

3. 063 ne = 9.96912 ‘ 43. 950
12.47 grains MnCl,=28. 42 grains AgC]. 43, 878—Turner.

Mean, 43. 924+0.015

\Berzelius.

Hence, Mn=55.07, Berzelius; or 54.87, Turner.

3 Berz. Jahresbericht, 9, 136. 1829.
4 Trans. Roy. Soc. Edinb., 11, 143. 1831.

 

1Journ. Chim. Phys., 17, 538. 1919.

5 Lehrbuch, 5 Aufi., 3, 1224.
2 Poggend. Annalen, 8, 185. 1826. ; :
No. 8] ATOMIC WEIGHTS—CLARKE. 317

Many years later Dumas’ also made the chloride of manganese the starting point ‘of some
atomic weight determinations. The salt was fused in a current of hydrochloric acid and after-
wards titrated with a standard solution of silver in the usual way. One hundred parts of Ag
are equivalent to the quantities of MnCl, given in the third column:

MnCl, Ag. Ratio.

3. 3672 5.774 58. 317
3, 0872 5. 293 58. 326
2. 9671 5. 0875 58. 321
1, 1244 1. 928 58, 320
1. 3134 2.251 _ 58,321

 

Mean, 58. 321d. 001
Hones, Mn = 54.912.

An entirely different method of investigation was followed by von Hauer,? who, as in the
case of cadmium, ignited the sulphate in a stream of sulphuretted. hydrogen and determined
the quantity of sulphide thus formed. I subjoin his weighings and also the percentage of MnS
formed from MnSO, as calculated from them:

MnS0,. Mn. Per cent.
4, 0626 2. 3425 57. 660
4. 9367 2, 8442 57. 613
5. 2372 3. 0192 57. 649
7. 0047 4. 0347 57. 600
4. 9175 2, 8297 57. 543
4. 8546 2. 7955 57. 585
4. 9978 2. 8799 57. 625
4, 6737 2. 6934 57. 629
4. 7240 2.7197 57. 572

 

Mean, 57. 608+. 008
Hence, Mn= 54.907.

This method of von Hauer, which seemed to give good results with cadmium, is, according
to Schneider,’ inapplicable to manganese, for the reason that the sulphide of the latter metal
is liable to be contaminated with traces of oxysulphide. Such an impurity would make the
atomic weight too high. The results of two different processes, one carried out by himself and
the other in his laboratory by Rawack, are given by Schneider in this paper.

Rawack reduced manganoso-manganic oxide to manganous oxide by ignition in a stream
of hydrogen and weighed the water thus formed. From his weighings I get the values in the
third column, which represent the Mn,O, equivalent to one gramme of water:

MnO, #0. Ratio.
4. 149 0. 330 12. 5727
"4, 649 . 370 12. 5643
6. 8865 . 5485 12. 5552
7. 356 . 5855 12. 5636
8, 9445 . 7135 12. 5361
11. 584 . 9225 12. 5572

 

‘feat, 12. 5582+ .0034
Hence, Mn= 54.08.

' Here the most obvious source of error lies in the possible ost of water. Such a loss, how-
ever, would increase the apparent atomic weight of manganese, but we see that valle found
is much lower than that obtained either by Dumas or von Hauer.

Schneider himself effected the combustion of manganous oxalate with aut of copper.
The salt was not absolutely dry, so that it was necessary to collect both water and carbon
dioxide. Then, upon deducting the weight of water from that of the original material, the
weight of anhydrous oxalate was easily ascertained. Subtracting from this the CO, we get

 

1 Ann. Chem. Pharm., 113, 25. 1860. 4 Journ. prakt. Chem., 72, 360. 1857. & Poggend. Annalen, 107, 605.
318 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

the weight of Mn. If we put CO,=100, the ee of manganese men: to it will be |
found in the last column:

* Oxalate. — #0. CO, Ratio.
1. 5075 0. 306 0. 7445 61.3835
2. 253 . 4555 1. 1135 61. 4291
3. 1935 . 652 1. 5745 61. 4163
5. 073 1. 028 2. 507 61. 3482

 

Mean, 61. 3943-4. 0122
Hence, Mn= 54.03.

Up to this point the data give two distinct values for Mn—one near 54, the other approxi-
mately 55—and with no sure guide to preference between them. The higher value, however,
has been confirmed by later testimony.

In 1883 Dewar and Scott! published the results of their work upon silver permanganate.
This salt is easily obtained pure by recrystallization and has the decided advantage of not being
hygroscopic. Two sets of experiments were made. First, the silver permanganate was heated
to redness in a glass bulb, first in air, then in hydrogen. Before weighing the latter gas was
replaced by nitrogen. The data are as follows:

Ag Mn0,. Ag+ Mn0. Per cent Ag+ MnO.
5. 8696 4. 63212 78.917
5. 4988 4, 33591 78, 852
7. 6735 6, 05395 78, 894
13, 10147 10. 31815 78. 756
9. 91065 78, 782
Oe o 91435 78. 811

 

Mean, 78. 885+. 0174
Hence, Kn=55.009.

The duplication of the last weighing is not explained.

In the second series the permanganate was dissolved in dilute nitric acid, reduced by sul-
phur dioxide, potassium nitrite, or sodium formate, and titrated with potassium bromide.
The ie ENO equivalent to 100 KBr appears in the third column:

AgMn0,. KBr. Ratio.
6. 5289 3, 42385 190. 686
7. 53878 3, 9553 190. 575
6. 1008 8. 20166 190. 559
5. 74647 3. 00677 191. 117
6. 16593 3. 23602 190. 540
5. 11329 2. 6828 190. 596
5. 07438 2. 66204 190. 624

18, 4484 7. 05603 190. 604

12, 5799 6. 60065 190. 588

12, 27025 6. 43808 190. 584 .

 

Mean, 190. 647-4. 0361

Vacuum weights are given throughout. To the first series of experiments the authors
attach little importance, and numbers 1 and 4 of the second series they also regard as question-
able. These experiments involved the use of sulphur dioxide as the reducing agent, and were
attended by the formation of an insoluble. residue, apparently: of a sulphide. Excluding them,
the remaining eight experiments of the second series give in mean—

KBr : AgMnO, :: 100 : 190.584+.0062, and Mn = 54.95

which will be used for the present calculation. Dewar and Scott also made determinations with
manganese chloride and bromide. With the first salt they found Mn=54.91, and with the
second, Mn=54.97; but they give no details.

1 Proc. Roy. Soc., 35, 44. 1883.

 
No, 3.] ATOMIC WEIGHTS—CLARKE. 319

-Marignac’s work upon the atomic weight of manganese also appeared in 1883.1 He pre-
pared the oxide, MnO, by ignition of the oxalate and subsequent reduction of the resulting
Mn,0O, in hydrogen. The oxide, with various precautions, was then converted into sulphate.
The percentage of MnO in MnSO, is appended:

MnO. MnSO,,. Per cent.
2. 6587 5. 6530 47. 032
2. 5185 5. 3600 46. 987
2. 5992 5. 5295 47, 006
2. 8883 6. 1450 47. 002

 

Mean, 47. 007+. 0025
Hence, Mn = 55.023.

J. M. Weeren, in 1890,? published determinations made by two methods, the one Marignac’s,
the other von Hauer’s. From manganese sulphate he threw down the hydrated peroxide
electrically, and the latter compound was then reduced in hydrogen which had been proved
to be free from oxygen. The resulting monoxide was cooled in a stream of purified nitrogen.
After the oxide had been treated with sulphuric acid, converted into sulphate, and weighed,
a few drops of sulphuric acid and a little sulphurous acid were added to it, after which it was
reheated and weighed again. This process was repeated until four successive weighings abso-
lutely agreed. The results of this set of experiments were as follows, reduced to a vacuum:

MnO. Un8S0,. Per cent.
15. 2349 32. 4142 47. 005
13. 9686 29. 7186 47. 004
13. 7471 29. 2493 - 47. 000
15. 5222 33. 0246 47. 001
14, 9824 31. 8755 47. 002
14, 6784 31. 2304 47. 000

Mean, 47. 002+. 0006
Hence, Mn=55.007.

Marignac’s mean, combined with this, hardly affects either the percentage itself or its
probable error. Fortunately, both Marignac and Weeren are completely in agreement as to
the ratio, and either set of measurements would be valid without the other. In order, therefore,
to give Marignac’s work some proper recognition, we can assume a general mean of 47.004+
.0006, without danger of serious error.

The manganese sulphate produced in the foregoing series of experiments was used, with
many precautions, for the next series carried out by von Hauer’s method. It was transferred
to a porcelain boat, dried at 260° to avoid errors due to retention of water taken up in the process
of transfer, and then heated to constant weight in a stream of hydrogen sulphide. Before
weighing, the sulphide was heated to redness in hydrogen and cooled in the same gas. The
results, with vacuum weights, were as follows:

MnS0,. MnS. Per cent,
16. 0029 9. 2228 57. 632
16. 3191 9. 4048 57. 631
15. 9307 9. 1817 57. 634
15. 8441 9, 1315 57. 634
16. 2783 9. 3819 57. 635
17. 0874 9. 8477 57. 633

 

Mean, 57. 633-4. 0004
von Hauer found, 57. 608+. 0080
Hence, the general mean is identical with Weeren’s to the third decimal place, which is
unaffected by combination with von Hauer’s data.
From Weeren’s figures alone Mn = 54.994.

1 Arch, Sci. Phys. Nat. (3), 10,21. 1883. 3 Atom-Gewichtsbestimmung des Mangans. Inaugural Dissertation, Halle, 1890.
161566°—20——21
320 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

The determinations by Baxter and Hines ' were based upon analyses of manganese bromide

and chloride, both fused in order to eliminate moisture.
employed, giving two ratios for each salt.

follows, first with the bromide:

MnBr,.
5. 58416
5. 63432
6. 53738
4. 81005
4, 88097
5. 63219
6. 52626
5. 79924
3. 59809
5. 16334
3. 92226
4, 49158
3. 60071
4, 77392
3. 57660
5. 69972

From the Ag ratio, Mn=54.921.
From the AgBr ratio, Mn = 54.922.

AgBr.
9. 76561
9. 85345
11. 43300
8. 41206
8. 53642
9. 85008
11, 41293
10. 14206
6. 29271
9. 02959
6. 85968
7. 85571
6. 29740
8. 34915
6. 25569
9. 96840

And Ag : Br :: 100 : 74.083.
Secondly, for the chloride series:

MnCl.
4. 62970
3. 52899
3. 30881
3. 56843
3. 45083
4. 47948
3. 92089

AgCl.
10. 54641
8. 03868
7. 63731
8. 12932
7. 86129
10. 20372
8. 93140

From the Ag ratio, Mn=54.925.

From the Ag(Cl ratio, Mn = 54.926:

And Ag : Cl :: 100 : 32.869.
The chloride ratios, as determined by different chemists, combine thus:

6. 56765
4. 83238
4, 90354
5. 65813

Ag.
7. 93740
6. 05041
5. 67279
6. 11818
5. 91637
7. 67995
6. 72227

 

Mean, 99. 539+. 0005

Ag ratio.
58. 328
58. 326
58. 328
58. 325
58. 327
58, 327
58. 327

Mean, 58. 327-b. 0003

The usual Harvard methods were
With vacuum weights the data obtained were as

AgBr ratio.
57. 181
57. 181
57. 180
57. 180
57. 178
57. 179
57. 183
57. 180
57.179
57. 182
57.178
57.176
57.178
57.179
57.174
57. 178

57. 179+. 0004

 

AgCl ratio.
43. 898
43. 900
43. 899
43. 896
43. 896
43. 900
43. 900

43. 898 +. 0005

 

Ag ratio
DUMB 3 eserecaieie ene Aecalee ss Seca aie eee eee 58. 321 +. 0010
Baxter and Hines... ..c..:.00 525.2 50008 mooaonganescewactais 58. 327 +. 0003
General mean.....-. 222... 0. ee eee eee eee ence eee 58. 3265+. 0003

Ag Cl ratio.
Berzelius with Turner. ............-..-----200- isialecere an 43,924 +. 0150
Baxter and Hines.....-....--. Bette eee eee e ees 43, 898 +. 0005
General means. 36.0 625 5 6 4c tapiectesinwes ce eceaieceaee 43. 898 +. 0005

In this instance the early work does not even effect the fourth decimal place.
We have now to consider the following ratios for manganese:

(1) 2Ag : MnCl, :: 100 : 58. 3265 +.0003.

(2) 2AgCl : MnCl, :: 100 : 43. 898 + .0005.

(3) 2Ag : MnBr, :: 100 : 99.539 + .0005.

(4) 2AgBr : MnBr, :: 100 : 57.179 4.0004.

 

1 Journ. Amer. Chem, Soc,, 28, 1560, 1906,
No. 3.] ATOMIC WEIGHTS—CLARKE. 321

(5) HO : Mn,O, :: 100 : 1255.82 +.3400.

(6) 2CO, : Mn :: 100 : 61.3943 4.0122.

(7) AgMnO, : Ag+MnO :: 100 : 78.835 +.0174.
(8) KBr : AgMnO, :: 100 : 190. 584 +.0062.

(9) MnSO, : MnO :: 100 : 47.004+.0006.

(10) MnSO, : MnS :: 100 : 57.633 + .0004.

The antecedent atomic weights are—

AgCIBr Ag=107. 880 +. 00016...................2.- S=32. 0650+. 00012

Cl= 35, 4605+. 00011...........2.0..00202. K=89. 0975+. 00011

Br= 79, 9228+. 00015. ...........2...22220. C=12. 0025+. 00019

H=1. 0078+. 00001
Hence,
From ratio—

Qiecg head cre hhas A ae tee en cama al aye a eh tert i Ie ON oh ee ee eee Mn=54. 030+. 0108
Dis aides desist aes eeepe ays Sibiu F araceasralattaleeielaiame hide @ AMO LDS esl eeees Soaeensea combab ae 54. 081+. 0610
OwissieweGa else Amis eie cay < tas Gana he aammeOuesosaeeess aunts canoes mackbietac's 54, 921+. 0012
IA fee esa ed ston ice te nat ek Si Unc ecpah hye dtc Vay aa uals NS DR A de eae dle tea 4a ae 54, 922+. 0016
Poche caernan hee is Soo eet ee aie eal asm ed aeebkimee sed Psyop te elie Sore niconlcnea Rater ces ie 54. 924 +, 00077
Dis buh jafars Sia a)e cisksizigcigai se ial as! nite Deauctalsmrenieby aaa se Seer ee eee e eile Steals eh eS ee 54, 926+. 0014
Berenice wis talars alse arsechelentebe uae atfo ta ace twa paam omiet sans wigs are ere gia ratsya ns rei ee EE SB Se si ORR ORO 54. 954+. 0074
NO kc xe ie eat Ss Sasi ge isuegsse tes Geass eu ones Reva on TA ea iad aanaaueeeias 54, 996+. 0010
Uist soe ae Or oh Seats e OSA Cd a a OR Os R eee dere neha ene wataeacitececeaeet 55. 009 +. 1522
Da gpal c Apacer esdiaye 2.5 nem iereian iciiia wa ach Deal eines wie Seie sats ors sae Cae cian wa els sabes 55. 0134. 0012
General Mean ssoscetieese fade ei eo duane er aknen so cs ouaneeseabenee ee 54. 949 4+-. 00045

In this combination the best work is evidently that of Baxter and Hines, as shown by
the concordant values derived from ratios 1 to 4. But Weeren’s work also appears to be
excellent, and ought not to be ignored. The general mean takes all the trustworthy determi-
nations into account, and seems to be preferable to any selection among them.

IRON.

The atomic weight of iron has been mainly determined from the composition of ferric
oxide, ferrous bromide, and the two chlorides. 2

Most of the earlier data relative to the percentage of metal and oxygen in ferric oxide
we may reject at once, as set aside by later investigations. Among this no longer valuable
material there is a series of experiments by Berzelius, another by Débereiner, and a third by
Capitaine. The first work deserving of present consideration is that of Wackenroder,? who
reduced the oxide in hydrogen at a moderate red heat. The following percentages of iron
were thus found:

69. 62 69. 98
69. 954 69. 99
69. 98

70, 04
Mean, 69, 927+. 0905

If we reject the first of these figures the mean becomes 69.988 + .0099, which is more trust-
worthy. Hence, Fe=55.97.

In 1844 Berzelius* published two determinations of the ratio in question. He oxidized
iron by means of nitric acid, and weighed the oxide thus formed. He thus found that when
O=100 Fe=350.27 and 350.369.

- Hence the following percentages of Fe in Fe,0,:
70, 018
70. 022

Mean, 70. 020+. 0013

 

 

1 For details concerning these earlier reseraches, see Oudeman’s monograph, pp. 140, 141.
4 Arch. Pharm., 35, 279, and 36, 22, 1843.
3 Berz. Jahresb., 25, 43. Ann. Chem. Pharm., 30, 432.
322 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

Hence, Fe=56.05. The ‘‘probable error” assigned to this pair of measurements greatly
overvalues them. It is better, therefore, to give the mean equal weight with Wackenroder’s,
making it 70.020 +.0099. |

About the same time Svanberg and Norlin ' published two elaborate series of experiments;
one relating to the synthesis of ferric oxide, the other to its reduction. In the first set pure
piano-forte wire was oxidized by nitric acid, and the amount of oxide thus formed was deter-
mined. The results were as follows:

Fe. Fe,03. Per cent Fe.
1. 5257 2. 1803 69. 977
2, 4051 3. 4390 69. 936
2, 3212 3. 3194 69. 928
2. 32175 3. 3183 69. 968
2, 2772 3. 2550 69. 960
2. 4782 3. 5418 69. 970
2. 3582 3. 3720 69, 935

 

Mean, 69. 9534+ .0050
Hence, Fe=55.875.

In the second series ferric oxide was reduced by ignition in a current of hydrogen, yielding
the subjoined percentages of metal:

Fe0g. Fe. Per ct.
2. 98353 , 2. 08915 70. 025
2. 41515 1. 6910 70. 015
2. 99175 2. 09455 70. 014
3. 5783 2. 505925 70. 030
4, 1922 2. 9375 70. 072
3. 1015 2. 17275 70. 056
2. 6886 1. 88305 70. 036

 

Mean, 70.0354-4+.0055
Hence, Fe = 56.093.

It is evident that one or both of these series must be vitiated by constant errors, and that
these probably arise from impurities in the materials employed. Impurities in the wire taken
for the oxidation series could hardly have been altogether avoided.

In 1844 there was also published an important paper by Erdmann and Marchand.? These
chemists prepared ferric oxide by the ignition of pure ferrous oxalate, and submitted it to re-
duction in a stream of hydrogen. Two sets of results were obtained with two different samples
of ferrous oxalate, prepared by two different methods. For present purposes, however, it is
not necessary to discuss these sets separately. The percentages of iron in Fe,O, are as follows:

70. 013
69. 962
69. 979A

70. 030
69, 977

70, 044
70. 015>}B
70. 055

Mean, 70. 0094+ .0080

 

Hence, Fe = 56.025.

Tn 1850 Maumené’s * determinations appeared. He dissolved pure iron wire in aqua regia,
precipitated with ammonia, filtered off the precipitate, washed thoroughly, ignited and weighed
after the usual methods of quantitative analysis. The percentages of Fe in Fe,O, are given in
the third column:

Fe. Fe,03. Per cent.
1, 482 2.117 70. 005
1. 452 2. 074 70. 010
1. 3585 1,941 69. 990

 

1 Berz. Jahresb., 25, 42. 2 Journ. prakt. Chem., 33, 1. § Compt. Rend., Oct. 17, 1850.
No. 3.] ATOMIC WEIGHTS—CLARKE. 323

Fe. Fe,03. Per cent.
1, 420 2. 0285 70. 002
1, 492 2. 1815 69. 998
1, 554 2. 220 70. 000

Mean, 70, 0008 +.0019
Hence, Fe = 56.003.

The two determinations by Rivot* are quite unimportant. This chemist reduced ferric
oxide in hydrogen, and obtained the subjoined percentages of iron:

69. 31
69. 35,

Mean, 69. 33 4.0130
Hence, Fe = 54.25.

Richards and Baxter ? also reduced ferric oxide by hydrogen. Iron was purified electro-
lytically and then converted into oxide by two processes. First, by solution, precipitation
as hydroxide, and ignition of the latter compound. With the oxide thus prepared, the two
subjoined reductions were made:

Fe,03. Fe. Per cent Fe.
3.17485 2. 22096 69. 954
3. 61235 2. 52750 69. 968

Mean, 69.961 + .0047
Hence, Fe= 55.900.
Secondly, iron was converted into nitrate, and that into oxide by calcination. The oxide
was free from occluded gases. The data, with vacuum weights for both series, are as follows:

Fe,03. Fe. Per cent Fe.
° 3.97557 2. 78115 69. 956
4. 89655 38. 42558 69. 959
4, 35955 3. 04990 69. 959
7, 14115 4. 99533 69. 951
6. 42021 4, 49130 69. 956

Mean, 69.9562+ .0010
Hence Fe= 55.883.
Still more recently, Baxter and Hoover * have published two more series of reductions.
of ferric oxide, one of material prepared from ordinary iron, the other of material from meteoric
iron. The two samples of iron were thus proved to be identical in character. The data, with

vacuum weights, are as follows: : ;
1. Ordinary iron.

Fe,03. Fe. Per cent Fe.
4, 86111 3. 39995 69. 9418
7. 59712 5. 31364 69. 9428
6. 50945 4, 55298 : 69, 9442
8, 00040 5. 59576 69. 9435
7. 27260 5, 08661 69, 9421
7. 69441 5. 38155 69. 9410
7. 33754 5. 13213 69. 9435

Mean, 69. 9427 + .0003

Hence, Fe = 55.847. ss
2. Meteoric iron.

FeO. Fe. Per cent Fe.
4, 46431 . 3. 12240 69. 9414
4. 85179 3. 39351. - 69. 9435
5. 21397 3. 64674 69. 9417
5. 99087 4. 19030 69. 9448
6. 70197 4. 68752 69. 9425

Mean, 69. 9428-+ .0004
Hence, Fe = 55.848.
1 Ann. Chem. Pharm., 78, 214. 1851.

3 Proc. Amer. Acad., 35, 253. Zeitsch. anorg. Chem., 23, 245. 1900.
3 Journ. Amer. Chem. Soc., 34, 1657. 1912.
324 MEMOIRS NATIONAL ACADEMY OF. SCIENCES. [Vou, XVI.

The eleven series of figures for this ratio combine thus:

Wackenroder..........-.----..-- sige weaean aa aaaNOnA 69.988 -+.0099
Berzelius. .....2.22222222.-e-2e0 ee! pious ca enwincase 70.020 + .0099
Erdmann and Marchand............ Ewe CeLeeeacTeeeee 70. 0094 + .0080
Svanberg and Norlin, oxidation........-.....2.....0.6-- 69. 9534 + .0050
Svanberg and Norlin, reduction. .........-.....-..-.--- 70. 0354 + .0055
Maumené...........-..-- 22202 eee eee ee genes seee bes owes 70. 0008 + .0019
RIVObs. 623 deans as niandiw ndieeeeveeaewnee cbt geseeaxe 69.33 +.0013
Richards and Baxter, 1......--...2..0.002ceeceeeeeeeeee 69.961 +.0047
Richards and Baxter, 2............00.0.c2eece cece cece ee 69. 9562 + .0010
Baxter and Hoover, 1... ........2202202-ceecee eee eeees 69. 9427 + .0003
Baxter and Hoover, 2........ PRtdinnheieceadesenes tees 69. 9428 + .0004
General mean........-...2. 2.02222 ee ence eee eee 69. 9451 + .00023

Dumas” results, obtained from the chlorides of i iron, are of so little weight that they qnight
safely be omitted from our present discussion. For the sake of completeness, however, they
must be included.

Pure ferrous chloride, ignited in a stream of hydrochloric acid gas, was dissolved ‘i in water
and titrated with a silver solution i in the usual way. One hundred parts of silver are equivalent
to the amounts of FeCl, given in the third column:

FeCl,. Ag. Ratio.
3. 677 6. 238 58. 945
3. 924 6. 675. 58. 787

Mean, 58. 866+ .053

Ferric chloride, titrated in the same way, gave these results: .
FeCls. Ag. Ratio.
1.179 2. 3475 50. 224.
1. 242 2.471 50. 263

 

>i Mean, 50. 2435-4 .0132
These give us two additional values for Fe, as follows:

Fret FeCl jocc0scsasseeoceee. Donker: ee Fe=56. 089
From FeOlg. ... 2.00022 -.2eeee ee eee cence cere cee eee ceee eee Fe=56. 226

A series of determinations of the equivalent of iron, made by students by measuring the
hydrogen evolved when the metal is dissolved in an acid, was published by Torrey in 1888.?
The data have, of course, slight value, but may be considered as being in some measure confirm-
atory. They are as follows:

56. 40 56. 06
55. 60 56. 22
s, 55. 38 55. 80
"85. 56 55. 78
55. 48 55. 60
55. 50 55. 70
55. 86

55. 94
Mean, 55. 777 + .0532

These values undoubtedly depend on Regnault’s value for the weight of hydrogen. Cor-
recting by the later value, as found in the chapter of this work relating to the density ratio
H :O, the mean becomes Fe=55.608 + .0532. With O=16, Fe=56.042. The probable error
in the weight of the hydrogen is ignored as having no practical significance.

A few determinations of the atomic weight of iron by Winkler * need to be mentioned, not
as directly significant, but as relating to the validity of a method which he applied to nickel and
cobalt.- Iron, not absolutely pure, was dissolved in a solution of iodine and potassium iodide.
The quantity of iodine was. known, and after the reaction ended the amount unconsumed was
measured by titration with thiosulphate solution. A ratio between iodine and iron was thus

 

Ann, Chem, Pharm., 113, 26. 1860, 2 Amer, Chem. Journ,, 10, 74. # Zeitschr. anorg. Chem., 8, 291, 1895.
No. 3.] ATOMIC WEIGHTS—CLARKE. 325

determined, which can be expressed as I, : Fe :: 100: 2. Two series are given, one with iron
cleaned by scrubbing, the other with iron which had been heated in hydrogen. The weights of
iron given below are corrected for known impurities:

 

Series I.
Fe. I. Ratio.
0. 5726 2. 585609 22. 146

. 5778 2. 608375 22.152 .-
. 5721 2. 582935 22. 149
Mean, 22. 149+ .0012
Hence, Fe = 56.226.

Series II.
Fe. I. Ratio.
0. 8252 3. 727316 22.139
- 8430 3. 809144 22.1381
. 8349 3. 771613 22. 137

 

Mean, 22. 186+ .0017

Hence, Fe=56.193. The weighted mean of both series is 22.145+.0010, which gives
Fe=56.216. This value is high, and so are the values found for cobalt and nickel by the same
method. The process is probably affected by serious constant errors, and the results obtained
by it are not good. For comparative purposes, however, the iodine ratio is included in the final
tabulation of ratios:

The best determinations of the atomic weight of iron appear to be those by Baxter and
his colleagues, and are based upon analyses of ferrous bromide. Of these the earlier are by
Baxter alone, who made two sets of analyses, fixing the two usual ratios, by the established
Harvard methods. His figures, with all corrections and vacuum weights, follow:

FeBry. AgBr. Ag. Ag ratio. AgBr ratio,
3. 55929 GT9B7S asec et 57. 420
3. 07448 5.354500 eee ne eee 57. 419
2. 96102 5. 15696 2. 96234 99. 956 57. 418
4. 00791 6. 97983 4. 00937 99. 964 57. 421

 

 

i Mean, 99. 960+. 0027 57. 4195+. 00044
From the Ag ratio, Fe= 55.828.
From the AgBr ratio, Fe= 55.825.
And Ag: Br:: 100: 74.087.

The analyses of ferrous bromide by Baxter, Thorvaldson and Cobb? fall into two series,
one preliminary, the other final. The data, with vacuum weights, are as follows:

Preliminary serves.

FeBr, Ag. AgBr. : Ag ratio AgBr ratio.
3. 45339 3. 45481 6. 01358 99. 9589 57. 4265
3. 04933 3. 05055 . 6. 31029 99. 9600 57, 4228
2. 9007 2.9019 a aweee 99.9586 — .......
3. 0873 3.0885 = _~—_.....-. 99.961.  — .......
3. 50278 3. 50426 6. 10033 99. 9577 57. 4195
4. 05239 4. 05404 7. 05752 99. 9593 57. 4195
4. 08516 4.08683  ——....-.- 99.9598 LLL...

 

 

' Mean, 99. 9593+. 00027 57. 4221-4. 00114
From the Ag ratio Fe = 55.826.
From the AgBr ratio Fe= 55.835.
And Ag: Br::100: 74.077.

1 Proc. Amer. Acad.,39, 245. 1903. 2 Journ. Amer, Chem. Soc,,33 319. 1911.

 
326

From the Ag ratio Fe = 55.824.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.
Final series.

FeBr,. ° Ag. _AgBr, Ag ratio. AgBr ratio.
5. 03555 5. 03774 8. 76950 99. 9565 57. 4212
6. 06309 6. 06557 10. 55889 98. 9592 57. 4216
5. 59258 5. 59482 9. 73974 99, 9599 57. 4202
5. 89767 5. 90014 10. 27057 99, 9581 ' 57. 4230
4, 48546 ~ 4.48742 Leese 99.9568 — .......

5. 41562 5. 41799 9, 43171 99. 9563 57. 4193 ~
6. 50002 6. 50277 11. 31958 99. 9577 57. 4227
3. 56564 3. 56719 6. 20987 99. 9565 57. 4189
5, 82434 5, 32642 9, 27237 99. 9609 57. 4216
6. 38845 5. 39094 11, 12536 99. 9610 57. 4224
6. 37952 6. 38213 10. 10971 99. 9591 57, 4229
8.51818 — ....... 14,.83468 ——....... 57. 4207

Mean, 99. 9583-4. 00057 57. 4213 +. 00029

From the AgBr ratio Fe = 55.832.
And Ag: Br:: 100: 74.079.

There is also another series of analyses by Baxter and Thorvaldson' of ferrous bromide
prepared from meteoric iron. Their data are as follows:

From the Ag ratio Fe = 55.819.

ed Y

FeBr,. Ag. _ AgBr. Ag ratio. . AgBr ratio.
3.95460 3. 95631 6. 88720 99.9568 —S-57. 4196
4.66954 4.67177 8. 13282 99. 9523 57. 4160
4.75335 4, 75550 8. 27855 99. 9548 57. 4177
6. 95582 6. 95854 12, 11329 99. 9609 57. 4230
3. 20762 3. 20904 5. 58632 99, 9557 57. 4192

 

Mean, 99. 9561+.00093 57. 4191+. 00078

From the AgBr ratio Fe= 55.823.
And Ag: Br::100: 74.081.
These series relative to ferrous bromide now combine thus—

eacigewei tect ss aseunasisme eesti’ se eeeeea teens 99. 960 +. 0027
Baxter, Thorvaldson, and Cobb, 1............
Baxter, Thorvaldson, and Cobb, 2.........-..-
Baxter and Thorvaldson............. Meroiats este menses 99. 9561+. 00078
ds sak becensiomeNsewoueeeesseenees 99. 9585 +. 00024

Ag ratio. AgBr ratio.

57. 4195+. 00044
57. 42214. 00114
57. 4213+. 00029

57. 4191+. 00078
57. 4203+. 00023

we--ee-- 99, 9593+. 00027
--.... 99. 9583+. 00051

The ratios relative to the atomic weight of iron are now as follows:

(Vou, XVI.

(1) Fe,0, : 2Fe : 100 : 69.9451 +.00023.
(2) 2Ag : FeBr, : 100 : 99.9585 + .00024.
(3) 2AgBr : FeBr, : 100 : 57.4203 + .00023.
(4) 2Ag : FeCl, : 100: 58.866 + .0530.

(5) 3Ag : FeCl, : 100 : 50.2435 + .0132.

(6) I, : Fe : 100 : 22.145+.0010.

(7) H: Fe: 1 : 55.608 +.0532.

The antecedent values are—

Ag=107. 880 +. 00016

Cl= 35. 4605+. 00011

H=1. 00078+. 00001
Hence, From ratio—

Br= 79, 9228+. 00015
I=126. 926 +. 00023

i aoc etnee ieee saeen cue ee euaieeeiee Fe=55, 8244. 00068
Cd edd reticle aoa ay Mantel Scalar kee 5B, 8284. 00095
Tsetse cil aek ed eth cate tac a eimai ae 55. 854+. 00093
Bo osha da nase eceon hues oe ete ee 56. 042+. 0532
Dae ieee ane, Seceaeeee eee tem aces 56. 089+. 1144
Bras cacmlny ns ana aanaeaeaciackecebastaene’ 56. 216+. 0026
Rati ore entre een ENO 56. 226-4. 0428

Fe=565. 846+. 00047

 

1 Journ, Amer. Chem. Soc., 33, 337. 1911.
No. 3.] ATOMIC WEIGHTS—CLARKE. 327

Rejection of the four highest values lowers the atomic weight of iron to 55.833. The
true value probably lies between this and the general mean as found.

NICKEL AND COBALT.

On account of the close similarity of these metals to each other, their atomic weights, so

nearly identical, have received of late years much attention.
__., The, first determinations, and the only ones up to 1852, were made by Rothhoff,’. each
with but a single experiment. For nickel 188 parts of the monoxide were dissolved in hydro-
chloric acid; the solution was evaporated to dryness, the residue was dissolved in water, and
precipitated by silver nitrate; 718.2 parts of silver chloride were thus formed; whence, Ni=
59.05. The same. process was applied also to cobalt, 269.2 parts of the oxide being found equiva-
lent to 1,029.9 of AgCl; hence, Co=58.93. These values are so nearly equal that their differ-
ences were naturally ascribed to experimental errors. They are, however, entitled to no special
weight at present, since it can not be certain from any evidence recorded that the oxide of either
metal was absolutely free from traces of the other.

In 1852 Erdmann and Marchand? published some figures, but without details, concerning
the atomic weight of nickel. They reduced the oxide by heating in a current of hydrogen, and
obtained values ranging from 58.2 to 58.6, when O= 16.

In 1856, incidentally to other work, Deville* found that 100 parts of metallic nickel yielded
262 of ‘sulphates whence Ni= 59.26.

To none of the foregoing estimations can any importance now be attached. The modern
discussion of the atomic weights under consideration began with the researches of Schneider‘ in
1857. This chemist examined the oxalates of both metals, determining carbon by the combus-
tion of the salts with copper oxide in a stream of dry air. The carbon dioxide thus formed was
collected as usual in a potash bulb, which, in weighing, was counterpoised by a similar bulb,
so as to eliminate errors due to the hygroscopic character of the glass. The metal in each oxalate
was estimated, first by ignition in a stream of dry air, followed by intense heating in hydrogen.
Metallic nickel or cobalt was left behind in good condition for weighing. Four analyses of each
oxalate were made, with the results given below. The nickel salt contained three molecules
of water and the cobalt salt two molecules.

. MC,0,.8H,0. C02. Per cent.
1. 1945 0. 528 44, 208
2. 5555 1. 12625 44. 072
3. 199 1, 408 44. 014
5. 020 2. 214 44, 104

 

Mean, 44. 098+. 027

The following percentages of nickel were found in this salt:

29. 107 29. 066
29. 082 29. 082
Mean, 29. 084, +006
CoC,0,.2H,0. co,. Per cent.

1. 63.55 0. 781 47. 753
1. 107 - 5295 47, 832
2. 309. 1.101 47. 683
3. 007 1, 485 47.722

 

Mean, 47, 7475+. 0218
The following were the percentages found for cobalt:

32. 552 32. 528

32. 619 32. 523
Mean, 32. 555+. 0149

 

\ Cited by Berzelius. Poggend. ‘Annalen, 8 194. 1826. 2 Ann, Chim. Phys. (3), 46, 182. 1856.
2 Journ. prakt. Chem., 55, 202. 1852. 4 Poggend. Annalen, 101, 387. 1857.
328 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou, XVI.

In a later paper' Schneider also gives someresu Its obtained with a nickel oxalate contain-
ing but two molecules of water. This gave him 47.605 per cent of CO,, and the following per-

centages of nickel:
31. 4115
31. 4088

Mean, 31. 4076+. 0026

 

The conclusion at which Schneider arrived was that the atomic weights of cobalt and nickel
are not identical, being about 60 and 58, respectively. .

These analyses of the oxalates are now of little value for determining the atomic weight
of two metals. Used directly, they give aberrant results; but by computing from the cross
ratio between the nickel or cobalt and the carbon dioxide, more nearly concordant values are
obtainable. With nickel the ratio 2CO, to Ni gives for the trihydrate Ni=58.042, and for the
dihydrate Ni=58.062. These figures are now known to be much too low. From cobalt oxa-
late, similarly computed, Co= 60.004, a value which is certainly too high.

The next chemist to take up the discussion of these atomic weights was Marignac, i in 18582
He worked with the chlorides and sulphates of nickel and cobalt, using various methods, but
publishing few details, as he did not consider the determinations final. The sulphates, taken
as anhydrous, were calcined to oxides. From the ration NiSO,: NiO, he found Ni=58.4 to
59.0, and from five measurements of the ratio CoSO,: Co, Co= 58.64 to 58.76. If oxygen is
taken as 16, these give for the percentages of oxide in the sulphate:

 

 

CoO in CoSO,. M0 in NiSO,.
48, 267 48, 187
48. 307 48, 387
; Mean, 48. 287+. 0135 Mean, 48. 287+. 0675 ;
Hence, Co=58.76. ; Hence, Ni=58.76.

. The chlorides were dried at 100°, but found to retain water; and in most cases were then
either fused in a stream of chlorine or of dry, gaseous hydrochloric acid, or else calcined gently
with ammonium chloride. The determinations were then made by titration with a standard
solution of silver in nitric acid. Five experiments with anhydrous CoCl, gave Co=58.72 to
58.84. Three more with CoCl, dried at 100° gave Co=58.84 to 59.02. Three with anhydrous
NiCl, gave Ni=58.80 to 59.00. If the calculations were made with Ag=108 and Cl=35.5,
then these data give as proportional to 100 parts of silver:

 

Nicl,. CoCl,.
60. 093 60. 056
60. 185 60, 111
3 60, 111
Mean, 60. 139+. 0310 60. 194
Mean, 60. 118+. 0192
Hence, Ni= 58.83. Hence, Co = 58.79.

In one more experiment NiCl, was precipitated with a known quantity of silver. The
filtrate was calcined, yielding NiO; hence the ratio 2Ag : NiO, giving Ni=59.29. This experi-
ment needs no farther attention.

In short, according to Marignac, and contrary to Schneider’s views, the two atomic weights
are approximately the same. Marignac criticises Schneider’s earlier paper, holding that the -
nickel oxalate may have contained some free oxalic acid, and that the cobalt salt was possibly
contaminated with carbonate or with basic compounds. In his. later papers Schneider rejects
these suggestions as unfounded, and in turn criticises Marignac. The purity of anhydrous
NiSO, is not easy to guarantee, and, according to Schneider, the anhydrous chlorides of cobalt
and nickel are liable to be contaminated with oxides. This is the case even when the chlorides are
heated in chlorine, unless the gas is carefully freed from all traces of air and moisture.

 

1 Poggend. Annalen, 107, 616. 2 Arch. Sci. Phys. Nat., nouv. série, 1372. 1858. Cfuvres Complates, 1, 575,
No. 3.) ° ATOMIC WEIGHTS—CLARKE. 3829

Dumas’ * determinations of the two atomic weights were made with the chlorides of nickel
and cobalt. The pure metals were dissolved in aqua regia, the solutions were repeatedly
evaporated to dryness, and the residual chlorides were ignited in dry hydrochloric acid gas.
The last two estimations in the nickel series were made upon NiCl, formed by heating the spongy
metal in pure chlorine. In the third column I give the NiCl, or CoCl, equivalent to 100 parts of
silver:

NiCl,. Ag. . Ratio.
0. 9123 1.515 60. 218
2. 295. 3. 8115 60. 212
3. 290 5. 464 60. 212
1. 830 3, 041 60.179
3. 001 4, 987 60.176

 

: Mean, 60. 1992+. 0062
Hence, Ni= 58.96.

CoCl,. Ag. Ratio.
2. 352 3. 903 60. 254
4,210 6. 990 60. 229
3. 592 ~ . , 5,960 60. 268
2, 492 4, 1405 60. 186
4, 2295 7. 0255 60. 202

 

Mean, 60. 2278+. 011
‘Hence, Co= 58.79.
These values for Co and Ni differ by less than a tenth of a unit; here, as elsewhere, the
figure for Ni being a trifle the lower.
Combining these data for nickel with Marignac’s series, we have—

1

2Ag: NiCl,:: 100: x

MBPIGI AG oo) ciarsauiviclai che’ has store oe aselsiajaialatelarsiat Awinrseibletstelewscetaerareiais 60. 139+. 0310
AUIS sjarese Nem actacratew era arenrs Pagaeteceecee cow avaaawee 60. 199+. 0062
General Mean sess sdscwiwssccsaeesedoaswer saewesees 60. 194-4. 0061

The cobalt figures will be combined with others later.

In 1863 ? the idea that nickel and cobalt have equal atomic weights was strengthened by
the researches of Russell. He found that the black oxide of cobalt, by intense heating in an
atmosphere of carbon dioxide, became converted into a brown monoxide of constant composition.
The ordinary oxide of nickel, on the other hand, was shown to be convertible into a definite
monoxide by simple heating over the blast lamp. The pure oxides of the two metals, thus
obtained, were reduced by ignition in hydrogen, and their exact composition thus ascertained.
Several samples of each oxide were taken, yielding the following data. The separate samples

are indicated by lettering:

Nickel.
Mo. Ni. Per cent Ni.
2. 0820 1, 6364 78. 597
ap 0956 1, 6468 78. 584
2. 0148 1, 5838: 78. 608
2. 2069 1. 7342 78. 581
BY 2. 2843 1, 7952 78. 589
2.1329 1, 6761 78. 583
2. 2783 1.7911 78. 616
C42, 1434 1, 6845 78. 590
2, 4215 1, 9030 78. 588
2. 1859 1.7179 78. 590
2. 0088 1. 5788 78. 594
2. 0839 1. 6379 78. 597
2. 6560 2. 0873 78. 588

Mean, 78. 593-4. 0018

1 Ann Chem. Pharm., 113,25, 1860. 3 Journ. Chem. Soc. (2), 1,51. 1863.
330 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von XVI.

Cobalt. .
Co. Per cent Co.
2 veil 1, 6670 78. 591
2, 0241 1. 5907 78. 588
As 2. 1296 1, 6673 78. 550
1. 9947 1, 5678 78. 598
3. 0628 2. 4078 78. 614
2, 1167 1, 6638 78. 603
a 7717 1, 3924 78, 591
1.7852 1, 4030 78. 591
of 6878 1. 3264 78, 588
2, 2076 1, 7350 78, 592
Df 2. 6851 2.1104 78. 597
ia 1461 1. 6868 78. 598
% 4038 2. 6752 78. 595
2. 2778 1.7901 ' 78.589
2.1837 1. 7163 78. 596

Mean, 78. 592+. 0023

These percentages are practically identical, and lead to essentially the same mean value for
each atomic weight, namely, Ni=58. 742

Co = 58. 738

Ina later paper Russell’ confirmed the foregoing results by a different process. He dissolved
metallic nickel and cobalt in hydrochloric acid and measured the hydrogen evolved. Thus the
ratio between the metal and his ultimate standard was fixed without the intervention of any
other element. About two-tenths of a gramme of metal, or less, was taken in each experiment.
The data obtained were as follows; the last column giving the weight of hydrogen, computed
from its volume, yielded by 100 parts of cobalt or nickel:

_ Nickel.
We. Ni. Vol. Hin ce. Ratio.
. 0906 153. 62 8, 420
.1017 172. 32 8, 418
. 1990 337. 06 3. 416
AS, 0997 168. 93 8, 417
. 1891 319. 86 3, 412
. 1859 314. 75 3. 415
. 1838 $11. 25 3, 416
. 1892 318, 75 8. 398
of 1806 305. 28 . . 8,409
. 2026 333. 81 3. 404
C .1933 325. 98 8, 401
. 1890 319. 77 3, 412
>| 1942 328. 15 3. 408
.1781 301. 09 3,410
Mean, 3. 411-4. 001
Cobalt.
Wt. Co. Vol. H in ee. Ratio.
1958 321. 36 8. 895
‘Ad: 1905 312. 95 38, 398
. 1946 319. 63 8. 897
. 2002 328. 96 8. 398
. 1996 328, 43 . 8. 408
B}. 2000 329. 55 8. 401
1721 290. 17 8. 401
cf. 1877 308. 97 8. 404
. 1985 318. 60 8. 405
D. { . 1909 314. 78 8. 410
. 1834 305. 40 8, 407

Mean, 3. 4017+. 0009

 

1 Journ. Chem. Soc. (2), 7, 204. 1869.
No. 3.] ATOMIC WEIGHTS—CLARKE. | 331

The weight of the hydrogen in these determinations was doubtless computed from Regnault’s
figures for the density of that gas. Correcting by the new value for the weight of a liter of hydro-
gen, 0.089872 gramme, the ratios become:

PovinieKel jo2ceciins \saaeacaitice iaeijdsa'elee Seumce atistelceenes 3. 42114. 0010
HOM CODAM is sisi wietcicisiiace Sia wip wiv iw'n sisinicle ou see's siomarniae gmaiaiaiels 38. 4112+. 0009

Hence, Ni=58. 92 and Co=59. 09.

Some time after the publication of Russell’s first paper, but before the appearance of his
second, some other investigations were made known. Of these the first was by Sommaruga,!
whose results, obtained by novel methods, closely confirmed those of Schneider and antagonized
those of Dumas, Marignac, and Russell. The atomic weight of nickel Sommaruga deduced from
analyses of the nickel potassium sulphate, K,Ni(SO,),.6H,O, which, dried at 100°, has a perfectly
definite composition. In this salt the sulphuric acid was determined in the usual way as barium
sulphate, a process to which there are obvious objections. In the third’column are given the
quantities of the nickel salt proportional to 100 parts of BaSO,:

Sulphate BaS0,. Ratio.
0. 9798 1, 0462 93. 653
1, 0537 1.1251 93. 654
1. 0802 1.1535 93. 645
1.1865 1, 2669 93. 654
3. 2100 3. 4277 93. 649
3. 2124 3. 4303 93. 648
Hence, Ni=58. 79. eee eee

For cobalt Sommaruga used the purpureocobalt chloride of Gibbs and Genth. This salt,
‘dried at 110°, is anhydrous and stable. Heated hotter, CoCl, remains. The latter, ignited in
-hydrogen, yields metallic cobalt. In every experiment the preliminary heating must be carried

on cautiously until ammoniacal fumes no longer appear:

Chloride Co. Per cent.
0. 6656 0. 1588 23. 858
1, 0918 - 2600 23. 814
- 9058 . 2160 23, 846
1. 5895 - 3785 23. 813
2. 9167 - 6957 23. 847
1. 8390 4378 23. 806
2. 5010 - 5968 23. 808
Hence, Co=59. 91. Mean, 23. 827+. 006

Further along this series will be combined with a similar one by Lee. It may here be said
that Sommaruga’s paper was quickly followed by a critical essay from Schneider,’ indorsing
the former’s work and objecting to the results of Russell.

In 1867 still another new process for the estimation of these atomic weights was put for-
ward by Winkler,? who determined the amount of gold which pure metallic nickel and cobalt
could precipitate from a neutral solution of sodio-auric chloride.

In order to obtain pure cobalt Winkler prepared purpureocobalt chloride, which, having
been four or five times recrystallized, was ignited in hydrogen. His nickel was repeatedly purified
by precipitation with sodium hypochlorite. From material thus obtained pure nickel chloride
was prepared, which, after sublimation in dry chlorine, was also reduced by hydrogen. One
hundred parts of gold are precipitated by the quantities of nickel and cobalt given in the third
columns, respectively. In the cobalt series I include one experiment by Weselsky, which was
published by him in a paper presently to be cited:

Mi. Au. Ratio.
0. 4360 0. 9648 45.191
. 4367 . 9666 45.179
- 5189 1. 1457 45, 291
. 6002 1, 3286 45. 175

Mean, 45. 209+. 019
1 Sitzungsb. Wien. Akad., 54, 2 Abth., 50, 1866. 2 Poggend, Annalen, 130, 310. 43 Zeit. anal. Chem., 6,18. 1867.

 
332 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

Hence, Ni=59. 44.

Co. Au. : Ratio.

. 5890 1. 3045 45.161

. 3147 . 6981 45. 080

. 5829 1, 2913 45.141

S111 1, 1312 45, 182

. 821 1, 2848 45. 307 °

B59 1.241 45, 044—Weselsky

Mean, 45. 1514025 sets
Hence, Co=59. 36. . :
Weselsky’ s paper,! already quoted, relates only to cobalt. He ignited the cobalticyanides

of ammonium and of phenylammonium in hydrogen, and from the determinations of cobalt

thus made deduced its atomic weight. His results are as follows:
0. 7575 grm. (NH,).CosCyia gave 0. 166 grm. Co. 21. 914 per cent.
- 5143 .113 f 21.972 “
Mean, 21. 9434. 029
Hence, Co= 59.09

(Cy H,N).Co,Cy2 Co. Per cent.
* 0. 8529 0. 1010 11, 842
- 6112 . 0723 11. 829
. 7140 - 0850 11. 905
- 9420 . 1120 11. 890

Mean, 11. 8665-L. 0124
Hence, Co=59. 02.

Next in order is the work done by Lee ? in the laboratory of Wolcott Gibbs. Like Weselsky,
Lee ignited certain cobalticyanides and also nickelocyanides in hydrogen and determined the
residual metal. The double cyanides chosen were those of strychnia and brucia, salts of very high
molecular weight, in which the percentages of metal are relatively low. A series of experiments
with purpureocobalt chloride was also carried out. In order to avoid admixture of carbon in
the metallic residues, the salts were first ignited in air, and then in oxygen. Reduction by hydro-
gen followed. The salts were in each case covered by a porous septum of earthenware, through
which the hydrogen diffused, and which served to prevent the mechanical carrying away of
solid particles; furthermore, heat was applied from above. Thé results attained appeared to be
satisfactory, and assign to nickel and cobalt atomic weights varying from each other by about a
unit; Ni being nearly 58, and Co about 59, when O=16. The cobalt results agree remarkably
well with those of Weselsky. The following are the data obtained:

Brucia nickelocyanide, NigOy,2(CgH_N,0,).H,.10H,0.

Sait. Ni. Per cent Ni.
0. 3966 0. 0227 5. 724

- 5638 - 0823 5. 729

- 4000 . 0230 5. 750

- 3131 . 01795 5.733

- 4412 . 0252 5. 712

. 4346 . 0249 5. 729

; Mean, 5. 7295+. 0034
Hence, Ni= 58.022.

Strychnia nickelocyanide, Ni,Cy,,(C,,H,,N,O,),H,.8H,0.

Salt. Ni. Per cent Ni.
0. 5858 0. 03854 6. 607

. 5489 . 0363 6. 613

. 3551 . 0234 6..589

- 4495 . 0297 6. 607

. 2030 . 0166 6. 561

. 1956 . 0129 6. 595

: Mean, 6. 595-b. 005
Hence, Ni=58, 079.

1 Ber. Deutsch. chem. Gesell., 2, 692. 1868. 2 Amer. Journ. Sol. (3), 2, 44. 1871.

 
No. 3.]

ATOMIC WEIGHTS—CLARKE.

Brucia cobalticyanide, Co,Cy,(C,3H,N,O,),H,. 20H,0.

Salt.
0. 4097
- 3951
. 5456
- 4402
. 4644
. 4027

Hence, Co=59.20.

Co.
0. 0154
. 0147
. 0204
. 0165
. 0174
. 0151

Per cent Co.
3. 759
8. 720
3. 739
3. 748
3. 747
3. 749
Mean, 3. 7437+. 0036

Strychnia cobalticyanide, Co,Cy,,(C,,H,N,O,).H,.8H,0.

Salt.
0. 4255
. 4025
. 3733
- 4535
. 2753
. 1429

Hence, Co =59.10.

Co.
0. 0195
. 0185
. 0170
. 0207
. 0126
. 0065

Per cent Co.
4. 583
4. 596
4, 554
4, 564
4.577
4. 549
Mean, 4, 5705+. 005

Purpureocobalt chloride, Co,(NH,) Cl,.

Salt.

0. 9472
- 8903
- 6084
. 6561
. 6988
. 7010

Hence, Co =59.099.

Co.

0. 2233
. 2100
. 1485
. 1547

. 1647
. 1653

Per cent Co.
23. 575
23. 587
23. 586
23. 579
23. 569
23. 581
Mean, 23. 5795-£. 0019

The last series may be combined with Sommaruga’s, thus:

SOMMALUPA saisscensseeecet sss sce sceeceeeceeee ys ys eens 23. 827 +. 006
TiG6 5. s0ecceiee siecisece Woqsted s dsbh os & canepparstanSucyducyerae a eOe wncpereee 23. 5795+. 0019

eI are pee a AIO 23. 6045+. 0018
Baubigny’s ' determinations of the atomic weight of nickel are limited to two experiments

upon the calcination of nickel sulphate, and his data are as follows:

NiSO,.
6. 2605
4, 4935

Hence, Ni=58.741.

Ni.
3. 0225
2. 1695

Per cent.

48. 279

48. 281
Mean, 48. 280

Zimmerman’s work, published after his death by Kriiss and Alibegoff,? was based, like
Russell’s, upon the reduction of cobalt and nickel oxides in hydrogen. The materials used
were purified with great care, and the results were as follows:

Nickel.

Ni. Ni. Per cent Ni.
6. 0041 4.7179 78. 578
6. 4562 5. 0734 78. 582
8. 5960 6. 7552 78. 585
4. 7206 3. 7096 78. 583
8. 2120 6. 4536 78. 587
9. 1349 7. 1787 78. 585

10. 0156 7. 8702 78. 579
4. 6482 3. 6526 78. 580
8. 9315 7. 0184 78. 580

10. 7144 8. 4196 78. 582
3. 0036 2. 3602 78. 579

Hence, Ni=58.704.

Mean, 78. 582+. 0006

 

1 Compt. Rend., 97, 951. 1883.

2 Ann. Chem., 232, 324. 1886.
334 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Von, XVI.

Cobalt.

CoO. Co. Per cent Co.
6. 3947 5. 0284 78. 634
6. 6763 5. 2501 78.638
5. 6668 4, 4560 78. 633
2. 9977 2. 3537 78. 637
8. 7446 ' 6. 8763 78. 635
3. 2625 2. 5655 78. 636
6. 3948 5. 0282 78. 630
8. 2156 6. 4606 78. 638
9. 4842 7. 4580 78. 636
9. 9998 7. 8630 78. 632

Mean, 78. 635+. 0002

 

Hence, Co = 58.889.

Shortly after the discovery of nickel carbonyl, NiC,0,, Mond, Langer, and Quincke * made
use of it with reference to the atomic weight of nickel. The latter was purified by distillation
as nickel carbonyl, then converted into oxide, and that was reduced by hydrogen in the usual
way.

NiO. Ni. Per cent Ni.
0. 2414 0. 1896 78. 542

. 3186 - 2503 78. 562

- 3391 . 2663 78. 531

 

: Mean, 78. 545+. 0061
Hence, Ni=58.575.

Schutzenberger’s experiments,? published in 1892, were also few in number. First, nickel
sulphate, dehydrated at 440°, was calcined to oxide.

NiSO,. Nio. Per cent.
3. 505 1. 690 48,217 |
2. 6008 1. 2561 48, 297

 

: Mean, 48. 257+. 027
Hence, Ni=58.671.

Secondly, nickel oxide was reduced in hydrogen, as follows:

 

Ni. Ni. Per cent.
1. 6865 1, 3245 78. 5385
1, 2527 - 9838 78. 5383

Mean, 78. 534

Hence, Ni= 58.536.

In one experiment with cobalt oxide 3.491 grm. gave 2.757 Co., or 78.975 per cent. Hence,
Co=60.1. In view of the many determinations of this ratio by other observers, this single
estimation may be neglected. The experiments on nickel sulphate, however, should be com-
bined with those of Marignac and Baubigny, giving the latter equal weight with Schutzen-
berger’s, thus: :

Mariguae..:s'snicracaassaeesiedeedses sasestes shastne sess 48. 287+. 0675

Baubigny..........--2--20eeee cence cece reece eee e eee 48. 280+. 027

Schutzenberger. . ........- 22. s cece cece eee ence eee enes 48. 257+. 027
General mean...... 2... sce e nese eee e eee teen eee ee 48, 269+. 018

From this point on the determination of these atomic weights was temporarily compli-
cated by the questions raised by Kriiss as to the truly elementary character of nickel and
cobalt. If that which has been called nickel really contains an admixture of some other
hitherto unknown element, then all the determinations made so far are worthless, and the
investigations now to be considered bear directly upon that question. First in order comes
Remmler’s research upon cobalt.’ This chemist, asking whether cobalt is homogeneous,

1 Journ. Chem. Soe., 57, 753, 1890.

2Compt. Rend., 114, 1149. 1892.
8 Zeit. anorg. Chem., 2, 221. Also more fully in an Inaugural Dissertation, Erlangen, 1891

 
No. 3.} _ ATOMIC WEIGHTS—CLARKE.  .. © 339

prepared cobaltic hydroxide in large quantity and made a series of successive ammoniacal
extracts from it, 25 in all. Each extract represented a fraction, from which, by a long series
of operations, cobalt monoxide was prepared, and the latter was reduced in hydrogen after
the manner of Russell. The actual determinations began with the second fraction. and the
data are subjoined, the number of the fraction being given with each experiment:

CoO. Co. Per cent Co.
Beisiaicine tacts Shh aun G ube edict eae aes 0. 09938 0. 07837 78. 859
Dawhesidouacsecdce eel ee eaduatear . 15021 . 11814 78. 650
apenas psionic wisn Adare . 22062 . 17360 78. 687
scents version vey sanee ecat . 39011 . 30681 78. 647
Guscswscmesechy oi seuceacdnek se . 28820 . 22661 78. 629
igh aoe apagelonsee e232 Sega TL AH . 34304 . 26968 78. 615
Baaideraiwereas ite agama eas . 43703 . 34321 78. 532
Dos Oca a etn Ae . 91477 . 71864 78. 560
MO ieee toe erctrrcietas cua aaeteent? . 63256 . 49661 .78. 508
SDD Sides ek SRM wine ai each asl eaten tor . 32728 . 25701 ‘78. 529
VD secu eduaara ise di oedees sec eeeeeiere 2. . 38042 . 29899 78. 595
TB ede cane Rote Nae See a . 16580 . 18027 78. 571
he hse eh ah i yds Dh ec yan tiga 1. 01607 . 79873 78. 610
LD eae al ean cate eis laqaaanieraale 1. 31635 1. 03545 78. 661
Gatos dicthhhidess'odeesee ees . 91945 . $2315 78. 650
Waswercktewswds v ooie's aeeeleeiaies . 53100 - 41773 78. 668
WS escecendceeees tik eecensadsaweee se qO2oSL., . 64728 78. 572
AD sense cetcstans oak Set epee . 81139 - 63754 78. 574
20 Sciminsauewensw ae a.acrardate acta aes . 76698 . 60292 78. 610
DN oo Riese Bole Heaaseieaces Ay 13693 . 89412 78. 643
OD Sa Aieatenisvainsl) bes ete oxtn ld ss 2. 00259 1, 57495 - 78.646
DB ed cca ekuadetve doe Oot hea einen p nica 1. 04629 . 82185 78. 549
DA sisiirai ee Sige ses oie Las eae Soe - 48954 . 38466 78. 576
DD wis sced Sedsaibie te Gua dieses Bre 6g Sacaracerevac alate - 69152 - 54326 78. 560

 

. Mean, 78. 613+. 0099
Hence, Co=58.812.
Considered with reference to the purpose of the investigation, this mean and its probable

error have no real significance. But it is very close to the mean of other experimenters, and a
study of the variations represented by the several fractions seems to indicate fortuity rather
than system. Remmler regards his results as indicating lack of homogeneity in his material;
but it seems more probable that such differences as exist are due to experimental errors and to
impurities acquired in the long process of purification to which each fraction was submitted
rather than to any uncertainty regarding the nature of cobalt itself.

From the same point of view—that is, with reference to the supposed heterogeneity of
nickel—Kriiss and Schmidt‘ carried out a series of fractionations of the metal by distillation
in a stream of carbon monoxide. Nickel oxide, free from obnoxious impurities, was first
reduced to metal by heating in hydrogen, after which the current of carbon monoxide was
allowed to flow. The latter, carrying its small charge of nickel tetracarbonyl was then passed
through a Winkler’s absorption apparatus containing pure aqua regia, from which, by evapora-
tion, nickel chloride was obtained, and from that, by reduction in hydrogen, the nickel. Ten
such fractions were successively prepared and studied; first, by preparation of NiO and its
reduction in hydrogen; and, secondly, in some cases, by the reoxidation of the reduced metal,
so as to give a synthetic value for the ratio Ni : O. The data obtained are as follows, the
successive fractions being numbered: _

Reduction of NiO.
NiO. Ni. Per cent Ni.
1 i 3722 0. 2926 78. 614
» 7471 - 5870 78. 571
2. { . 7659 - 60085 78, 450

. 7606 . 5961 “78. 372

 

1 Zeit. anorg. Chem., 2, 235. 1892.
161566°—20-——-22
336 MEMOIRS NATIONAL ACADEMY OF SCIENCES. LVou. XVI.

Nio. Ni. Per cent Nt.
1.0175 . 7984 78, 467
of 2631 . 99065 78. 430
1. 2582 9868 78, 429
5193 4076 78. 490
. 9200 7215 78, 494
. 4052 3179 78, 455
. 6518 -bLI 78, 414
. 5623 4399 78, 232
. BB5G 4350 78. 294
( .9831 7724 78. 568
i . 9765 7646 78, 300
. 9639 -TBBT 78. 400
_ 5756 4538 78, 839
56765 4451 78, 411
8) 5663 4438 78. 368
5449 4272 78. 400
3174 2491 78, 481
. 3148 2467 78, 367
10 .4976 3904 78, 457
4961 3891 78. 432

 

2 Mean, 78, 444+. 0166
Hence, Ni=58.225.

Oxidation of Ni.
Ni Nio. Per cent Ni.

1 . 5870 7471 78. 571
6011 . 7659 78. 372
5961 7606 78, 359
7988 1.0175 78. 506
| 9913 1.2631 78. 482
. 9868 1. 2582 78. 429

. 4093 5193 78. 818

. 7216 . 9200 78. 435
3194 . 4052 78. 825
511 6518 78. 414

‘ f 4415 . 5623 78. 517
4350 5556 78. 294
.TTB2 9831 78. 853
| 7667 9765 78. 515
7558 . 9639 78. 411
BBB 5756 79. 135
4456 . 56765 78. 499

8). 44415 . 5663 78. 430
4423 5642 78. 394

. 2508 3174 79.015
2467 3148 78. 367
10f 3918 .4976 78. 738
3891 . 4961 78. 432

 

Mean, 78. 557+. 0319

Hence, Ni=58.616.

To these data of Kriiss and Schmidt the remarks already made concerning Remmler’s
work seem also to apply. The variations appear to be fortuitous, and not systematic, although
the authors seem to think that they indicate a compositeness in that substance which has been
hitherto regarded as elementary nickel. In view of all the evidence, however, I prefer to
regard their varying estimations as affected by accidental errors, and to treat their means like
others.

In 1889 Winkler‘ published a short paper concerning the gold method for determining the
atomic weights in question, but gave in it no actual measurements. In 1893? he returned to the

 

1 Ber. Deutsch. chem. Gesell., 22, 891. 1889. 2 Zeit. anorg. Chem., 4,10. 1893.
No. 3.] _ ATOMIC WEIGHTS—CLARKE. © 7 337

problem with a new line of attack, and at the same time he took occasion to criticize Kriiss and
Schmidt somewhat severely. He utterly rejects the notion that either nickel or cobalt contain
any hitherto unknown element, and ascribes the peculiar results obtained by Kriiss and Schmidt
to impurities derived from the glass apparatus used in their experiments. For his own part
he now works with pure nickel and cobalt precipitated electrolytically upon platinum, and avoids
the use of glass or porcelain vessels so far as possible. With material thus obtained he operates
by two distinct but closely related methods, both starting with the metal, nickel or cobalt,
converting it next into neutral chloride, and then measuring the chloride gravimetrically in one
process, volumetrically in the other.

After precipitation in a platinum dish the nickel or cobalt is washed with water, rinsed
with alcohol and ether, and then weighed. It is next dissolved in pure hydrochloric acid,
properly diluted, and by evaporation to dryness and long heating to 150° converted into anhy-
drous chloride. The nickel chloride thus obtained dissolves perfectly in water, but the cobalt
salt always gave a slight residue in which the metal was electrolytically determined and allowed
for. In the redissolved chloride, by precipitation with silver nitrate, silver chloride is obtained,
giving a direct ratio between that compound and the nickel or cobalt originally taken. The
gravimetric data are as follows, with the metal equivalent to 100 pants of silver chloride given
in a final column: .

 

Nickel.
Ni. AgCl. Ratio.
0. 3011 1. 4621 20. 594
. 2242 1. 0082 20. 605
. 5166 2. 5108 20. 570
. 4879 2. 3679 20. 605
. 8827 1. 8577 20. 601
. 3603 1. 7517 20. 568

Mean, 20. 590+. 0049
Hence, Ni= 59.03.

Cobalt.
Co. AgCl. Ratio.
0. 3458 1, 6596 20. 836
. 3776 1. 8105 20. 856
. 4493 2. 1521 20. 877°
. 4488 2.1520 20. 855
', 2856 1. 3683 20. 873
. 2648 1. 2768 20. 886

 

Mean, 20. 864+. ae

Hence, Co=59.81.

In the volumetric determinations the neutral chloride, prepared as before, was decomposed
by means of a slight excess of potassium carbonate, and in the potassium chloride solution,
after removal of the nickel or cobalt, the chlorine was measured by titration by. Volhard’s
method with a standard solution of silver. The amount of silver thus used was comparable
with the metal taken.

Nickel.

Ni. Ag. Ratio.
0. 1812 0. 6621260 27. 366
. L662 . 6079206 27. 339

. 2129 . 7775252 27. 382
. 2232 - 8162108 . 27,346
- 5082 1, 8556645 27. 386
. 1453 . 5315040 27. 338

 

Mean, 27. 359+. 0059
Hence, Ni=59.03.
338 MEMOIRS NATIONAL ACADEMY OF SCIENCES. LVon. XVI.

Cobalt.

Co. Ag. Ratio

0. 177804 0. 6418284 27. 702
. 263538 . 9514642 27. 699

. 245124 . 8855780 27, 679

. 190476 . 6866321 27.741

. 266706 . 9629146 27, 696

. 263538 . 9503558 27.781

Mean, 27. 708 +-. 0064

In view of the possibility that the cobalt chloride of the foregoing experiments might
contain traces of basic salt Winkler, in a supplementary investigation, checked them by
another process. To the electrolytic cobalt, in a platinum dish, he added a quantity of neutral
silver sulphate and then water. The cobalt gradually went into solution, and metallic silver
was precipitated. The weights were as follows:

Co. Ag.
0. 2549 0. 9187
- 4069 1, 4691

On examination of the silver it was found that traces of cobalt were retained—less than
0.5 mg. in the first determination and less than 0.2 mg. in the second. Taking these amounts
as corrections the two experiments give for the ratio 2Ag : Co :: 100 : the subjoined values:

27. 706
27. 687

These figures confirm those previously found, and as they fall within the limits of the
preceding series, they may fairly be included in it, when all eight values give a mean
of 27.705+.0050. Hence, Co=59.78.

Still another method, radically different from all of the foregoing processes, was adopted
by Winkler in 1894.2, The metals were thrown down electrolytically upon platinum, and so
weighed. Then they were treated with a known excess of a decinormal solution of iodine in
potassium iodide, which redissolved them as iodides. The excess of free iodine was then deter-
mined by titration with sodium thiosulphate, and in that way the direct ratio between metal
and haloid was ascertained. The results were as follows, with the metal proportional to 100
parts of iodine given in the third column:

Cobalt.

Wet. Co. We. I. Ratio.

. 4999 2. 128837 28. 482

. 5084 2. 166750 23. 463

First series. ...(. 5290 2. 254835 23. 466
. 6822 2. 908399 23. 456

- 6715 2. 861617 23. 466

. 5185 2. 209694 23. 465

Second series. { 5267 2. 246037 23. 450
. 5319 2. 268736 23. 445

Mean, 23, 462-4. 0027
Hence, Co = 59.57.
Nickel.

Wt. Ni. Wt. I. Ratio.

. 5144 2. 217494 23, 251

- 4983 2. 148502 28. 246

First series. ...4. 5265 2. 268742 23, 260
. 6889 2. 970709 23. 243

. 6876 2. 965918 23. 237

. 6120 2. 205627 23. 267

Second series. .{. 5200 2. 240107 23, 267
. 5246 2. 259925 23, 267

; Mean, 23. 255-4. 0091
Hence, Ni = 59.03.

1 Zeit. anorg. Chem., 4, 462. 1893. §Zeitsch. anorg. Chem., 8,1. 1894,

 
No. 8] ATOMIC WEIGHTS—CLARKE. 339

In these experiments, as well as in some previous series, a possible source of error is to
be considered in the occlusion of hydrogen by the metals. Accordingly, in a supplementary
paper, Winkler + gave the results of some check experiments made with iron, which, however,
was not absolutely pure. His conclusion is that the error, if existent, must be very small.

In 1895 Hempel and Thiele’s work on cobalt appeared.? First, cobalt oxide, prepared
from carefully purified materials, was reduced in hydrogen. The weights of metal and oxygen
are subjoined, with the percentage of cobalt in the oxide deduced from them:

Co. 0. Percentage.
0. 90068 0. 24429 78. 664
. 79159 . 21445 78. 686
1, 31558 . 35716 78. 648
Hence, Co=58.998. Mean, 78. 666+. 0074

In their next series of experiments, excluding a rejected series, Hempel and Thiele weighed
cobalt, converted it into anhydrous chloride, and noted the gain in weight. In four of the
experiments the chloride was afterwards dissolved, precipitated with silver nitrate, and then the
silver chloride was weighed. The data are as follows:

_ Co. Cl taken up. AgCi.
0. 7010 0.8453 La...
- 3138 Pa A cr
. 2949 . 3562 1.4340
- 4691 . 5657 2. 2812
. 5818 . 7026 2. 8303
. 5763 694% gw
- 5096 . 6142 2. 4813
From these weights we get two ratios, thus:
Cl, : Co: 100 : x. 2AgCl:Co::100:2.
82. 929 20. 565
82. 731 20. 564
82. 791 20. 556
82. 924 20. 538
= pe Mean, 20. 556+. 0043
82. 970
Mean, 82. 873+. 0241
Hence, Co=58.77. Hence, Co=58.93.

‘The second of these ratios was also measured by Winkler, and the two series combine as
follows:

Wanklefs.cic:cic esieeee dss sae aoaeieawslcinicnd ies cacReteys 20. 864+. 0050
Hempel and Thiele. ............2 22. eee eee eee ee eee eee 20. 556+. 0043
General mean .........25 22% seccessessescaess e's sieves 20. 687+. 0033

Hempel and Thiele apply to it a correction for silver chloride retained in solution, but
its amount is small and not altogether certain. For present purposes the correction may be
neglected.

The atomic weight of nickel was determined by Richards and Cushman * from analyses
of nickel bromide. This salt, as first prepared, contained traces of oxide, which are to be
deducted from the halide compound. In a preliminary series of experiments the following
figures were obtained, representing vacuum weights:

NiBr,. AgBr. Insoluble, mg. Ratio.
2. 26113 3. 88769 3. 22 58. 161
2. 80668 4, 82431 7.08 58. 178
1, 41317 2. 42880 3. 05 58. 184
1, 71759 2. 95307 . 88 58. 163
2. 48565 4. 27357 5. 24 58. 163
4. 32997 7. 44280 15. 83 58.177
2. 18072 8.74856 wt ew 58.175

Mean, 58. 172+. 0023

1 Zeitsch. anorg. Chem., 8, 291. 1895, See preceding section of this work, on iron, for the detailed determinations.
2 Zeitsch. anorg. Chem., 11, 73.
8 Proc. Amer. Acad., 33,97. 1897.

 
340 MEMOIRS. NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

-In the second set of analyses, both ratios were determined, namely, with. silver and with
silver bromide, by the standard methods. The data follow:

oo) NiBr,. AgBr. Ag. : "Ag ratio. * AgBr ratio.’
3. 28039 "5. 63892 3. 28910 101. 275 | 58.174 |
2. 70044 5 4. 64208 2. 66636 101.278 - 58. 173
3.38230 5.81391 3.33990 101.270 «58.176.
1.33459 ° 2, 29435 | ' ly 31787, : 101. 268 "58.169
1.25054 2.14963 | 1, 23482 101.273 © 58. 175
1, 32278 2. 27384 1, 30629 101, 262 58. 174
1,24452 2. 85805 2. 21652 101. 263 58.177

Mean, 101. 270+. 0015 58. 174-4. 0007
This value for the AgBr ratio, combined with the preliminary series, ge! a poneral mean
of 55.1738 + .0007. :
From the Ag ratio, Ni=58.654. ve ae: eee
From the AgBr ratio, Ni=58.658.
And Ag : Br :: 100 : 74.082. po ey
In a second memoir,! Richards and Cushman describe a series of determinations based

upon the reduction of nickel bromide by heating i in hydrogen. The correcred data appear
in the next table:

¢
oe

MBr,. Ni: Per cent Ni.
2. 83325 0. 76081 26: 853
3. 21625 . 86358 26. 851
2.31241 . 62094. 26. 853
2. 87953 . 77380 26. 855
2. 29650 . 61679 26. 858
. 2. 98893 . 80272 26. 856
5. 51291 1. 48056 26. 856
2. 24969 . 60415 26. 855
Hence, Ni=58.687. Mean, 26. 855-4. 0005

In this series a correction was applied for traces of sodium bromide contained in the nickel
salt. A similar correction, applied to the former series of determinations, would raise the
atomic weight of nickel by 0.015.

The three memoirs upon cobalt, by Richards and Baxter? -contain data relative to the
bromide, the chloride and the oxide. It is hardly necessary to state that all of the materials
employed’in the investigation were scrupulously purified, and that all weights were reduced to
a vacuum. First, as in the case of nickel, the two silver ratios ‘to the bromide were deter-
mined. A preliminary set of analyses gave results as follows:

‘ . CoBry. AgBr. _- Ratio.
2, 25295 8. 86818 . 58, 243
"2. 88763 _4, 95732 58. 250
2 88806 / “8, 24056 58. 263

; Mean, 58. 252-4. 0040
The second and third series of analyses gave both ratios, and may be tabulated together:

Ge CoBr,. ° ' AgBr. Ag. os Ag ratio. AgBr ratio.
mss 1.33564 °° 2. 29296 1. 31702 "JO. 414 58. 250 °°
2. 58129 4. 43095 2. 54585 101. 392 58. 256
2. 84382 4. 88135 _ 2, 80449 . 101. 402 88,259
1. 83722 3, 15368 1. 81170 101. 409 58, 256
2. 68584 4.61046 ' 2, 64879 “101. 399 58, 255
3. 18990 5.47607 see "ia 58, 252
2. 88914 4. 95943 2. 84891 101. 412 58. 255
2. 32840 3. 99706 2. 29593 101. 414 58, 258
1. 91703 3. 29053 1, 89033 101. 413 58. 259
Mean, 101. 407 58. 255
+.0018 +. 0007

 

1 Proc. Amer. Acad., 34,327. 1899. This memoir contains a very full criticism of all the earlier work on nickel.

2 Proc. Amer. Acad., 33,115. 1897. Ibid. inise es 1899, Ibid. ue 61. 1899, Hore oe Cushman, and Baxter, see Winkler
Zeitsch. anorg. Chem. 417; 236. . 1898. a cai
No. 3.) ATOMIC WEIGHTS—CLARKE. 341

The two series for the AgBr ratio, combined, give a weighted mean of 58.2549 + .0007.

From the Ag ratio, Co=58.950.

From the AgBr ratio, Co = 58,959.

And Ag: Br:: 100: 74.075.

In their second memoir Richards and Baxter describe the reduction of cobalt bromide by
heating in hydrogen. Three series of experiments were made, and in two of them a correction
was necessary for small quantities of sodium bromide contained in the cobalt salt. In the
following tables the corrected weights of cobalt bromide and cobalt are given

 

 

Series I.

CoBr,. Co. Per cent Co.
5. 59023 L 50680 26. 954
4, 61518 1. 24381 26. 950
3. 74498 1. 00920 26. 948
3. 00135 . 80899 26. 954

: Mean, 26. 951+.0010
Series II,

CoBr,. Co. Per cent Co.
5. 32194 1, 438428 26. 950
7. 50786 2. 02321 26. 948
2. 32630 . 62677 26. 943
7. 44694 2. 00736 26. 956

. Mean, 26. 949 +.0021
Series IIT.

CoBr,. Co. Per cent Co.
5. 10891 1.37721 26. 957
6. 41339 1. 72850 26, 951
6. 59805 1. 77876 26. 959
3. 02854 . 81606 26. 953

 

Mean, 26. 953+.0021

The general mean of the three series is 26.9508 + .0008. Hence Co=58.973.

The third memoir of Richards and Baxter gives analyses of cobalt chloride and oxide.
First, the chloride was reduced to metal by heating in hydrogen. Hempel and Thiele worked
in the opposite direction, heating cobalt in chlorine and thereby effecting the synthesis of the
compound. For uniformity of statement I give Richards and Baxter's series in the same form,

as the ratio Cl,: Co: :100: 2:

CoCl,. Co. Ratio.
4.16483 1. 89243 83. 279
2. 30512 1, 04723 83. 253

Mean, 83. 266+ .0087
Hence, Co = 59.053.
Hempel and Thiele’s figures give for this ratio the figure 82.873 4.0241. The general mean
of both series is 83.220 + .0082. -
Five reductions of cobalt oxide in hydrogen are given, three in one series and two separate
experiments with varied methods’ of manipulation. The results obtained are regarded by
Richards and Baxter as unimportant, and they point out the difficulties of the process. Their

data, arranged as one series, follow

CoO. Co. Per cent Co.
7. 04053 5. 53779 78. 656
6. 69104 5. 26312 78. 659
7. 83211 6. 15963 78. 646
7. 74240 6. 09219 78. 686
10. 58678 8. 32611 78. 646

Hence, Co = 58.973.

Mean, 78. 659+ .0051
342 MEMOIRS NATIONAL ACADEMY OF SCIENCES.

This mean combines with former means as follows

Hempel and Thiele. ....:......£.--23-
; Richards and Baxter. satan. eee

General mean

aibd capaci iaaseorencey TeOER ae

  

iS easels cose aes 78.6324 +

Here Zimmermann’s determinations practically appear alone.
The analyses of cobalt chloride by Baxter and Coffin’ were made by the usual methods,

as refined at Harvard University, and give the two silver ratios.

weights, are as follows:

CoCi,.: -
1. 09959
1. 47733
3. 84133
8. 64342
2. 96315
3. 48418
3. 29523
1. 57655

From the Ag ratio, Co=
From the AgC] ratio, Co= 58.966.

Ag.
1. 82671
2. 45398
6. 38081
6. 05232
4, 92244
5. 78815
5. 47410
2. 61905

58.961.

And Ag: Cl: : 100 : 32.866.
For the silver ratio, 2Ag: CoCl,

AgCl.

2. 42676
3. 26095
8. 47735
6. 54019
7. 69084
7, 27284
3. 48012

Ag ratio.
60. 195
60. 201
60. 201
60. 199
60. 197
60. 195
60. 197
60. 195

Mean, 60. 1975
+.0006

 

| AgCl ratio.

.0023
-0002
0099
0074

.0051

-0002

45. 311

~ 45, 304

45. 313

 

* 45. 3070

+.0001

The three series combine as follows: ; '
Marignac. niece t ene nec e cence ence eee cence cence eens 60. 118 & -0192
DUT, ape hdc so sia dievereva\dcajais ae eR dae Ru mueRE Ree EER SeS 60. 228 ++ .0110
Bexter ard Coffinvas ..ccnsassaesesesdenesedereeteeses 60.1975 + .0006

60.1975 + .0006

IVob. XVI.

The data, with vacuum

: 100 : 2, there are two earlier. sets of determinations,

o

In this sombinatite ‘the older series vanish. Their influence is apparent only i in the fifth

decimal place.

The determinations of zie atomic weight of nickel by Oechsner de Couinek and Gérard ?
although: they give a fairly good value for this constant. Nickel
oxalate, NiC,O,.2H,O, was reduced by heating in hydrogen, with the following results:

Per cent Ni.

are of sight importance, a

Oxalate.

0. 4300
. 4730
. 3440
. 3869
.5161

Hence, Ni= 58.596.

t

M.
0. 1380
. 1517
- 1104
- 1241
. 1656

Mean, 32. 084 + .0030

32. 093
32. 072
32. 093
32. 075

32. 087

Two.sets of determinations of the atomic woight of nickel were made by Baxter and Parsons?
in order to, compare the ordinary metal with nickel obtained from meteoric iron. Nickel oxide
was reduced by heating in hydrogen, with the following results.

vacuum. se

1 Journ. Amer. Chem, Soc., 28, 1580. 1906. Zeitsch. anorg. Chem., 61, 171.

2 Compt. Rend., 158, 1345. 1914,

The weights are reduced to a

3 Ann, Report, Carnegie Inst. Washington, 1915, 351.
No. 3.]

ATOMIC WEIGHTS—CLARKE.

1. Terrestrial nickel.

343

M0. Mi. Percent Ni.
6. 41281 5. 03957 78. 5860
6. 10952 4. 80102 78. 5826
6. 91023 5. 43006 78. 5800

4. 60077 3. 61528 78. 5799
5. 57122 4. 37792 78. 5810
5, 59217. 4, 39445 78. 5822

 

Mean, 78. 5819+. 0007

Hence, Ni= 58.568.

2. Meteorite nickel.

M0... M. Percent Ni.
~ 6. 65624 5. 23021 78. 5760
4, 45609 3. 50196 78. 5882
4, 64795 3. 65212 78. 5748

 

Mean, 78. 5797 +. 0009

: _ Hence, Ni=58.561.
These percentages combine with the earlier determinations as follows, Schutzenberger’s
measurements of the ratio being assigned equal weight with those of Mond, Langer, and Quincke:

Mussel ssc jody wies telat wade emetic pavsiniign seed sigenicis's 78.593 +.
Zimmermanniccs ca ss cseinescanvecens 44 uawsiereaeseaeci'ss 78.582 +.
Mond, Langer, and Quincke............... ieee eee eee 78. 545. +.
Schutzenberger......----.-.---2+-+-+- Presicesipewelciesecisiae ‘78. 534 +.
Kriiss and Schmidt, reduction SOMOS s aissiers oie. cia/siersuisiciex sic 78.444 +.
Kriiss and Schmidt, oxidation series............-2-+-++-+ 78,557 -b.
Baxter and Parsons, Liaise Saaale Saatiow Sal Sgiisl Sawesa tases 78. 5819+.
Baxter and Parsons, Do irctadrcion Uaracocionea ss suuaeeesei wee 78, 5797+.

General MCAD ichjsivcivienienw soy se siiereisloe ceieries eis 78.576 +.

0018
0006
6061
0061
0166
0319
0007
0009

00046

After rejecting the analyses of the oxalate ic Schneider, we have the following ratios for
nickel to consider:

(1) Per cent of Ni in NiC,O,.2H,O, 32.084 + .0030.
(2) Per cent of Ni in brucia nickelocyanide, 5.7295 + .0034.
(3) Per cent of Ni in strychnia nickelocyanide, 6.595 + 005.
(4) Per cent of NiO in NiSO,, 48. 269 + .018.
(5) Per cent of Ni in NiO, 78.576 + .00046.
(6) 2Ag : NiCl, :: 100 : 60.194 + .0061.
(7) 2AgCl : Ni :: 100 : 20.590 + .0049.
(8) 2Ag : NiBr, :: 100 : 101.270+.0015. |
(9) 2AgBr : NiBr, :: 100 : 58.1738 + .0007.
(10) NiBr, : Ni :: 100 : 26.855 + .0005.
(11) 2Ag : Ni :: 100 : 27.359 + .0059.
(12) 2Au : 3Ni :: 100 : 45.209 + .019.
(13) 2BaSO, : K,Ni(SO,),.6H,O :: 100 : 93.6505 + .001.
(14) Ni: H, :: 100 : 3.4211+.001.
(15) I, : Ni :: 100 : 23.255 + .0091.

The values used in reducing these ratios are—

- Ag==107. 880 +. 00016 C= 12. 0025+. 00019

Cl= 35. 4605+. 00011 N= 14, 0081+. 00008

» Br== 79. 9228+. 00015 K= 39. 0975+. 00011
I=126. 926 +. 00023 Ba=137. 358 +. 0023
S= 32. 0650+. 00012 Au=197, 209 +. 0029

H= 1.0078+. 00001
344 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Von XVI.

 

Hence,

From ratio—
Divs boron atas oOo Maas aeaea CSLe On ore Ni=58. 022+. 0345
= 58. 079-+.'0441
58. 596+. 0061
58. 654+. 0011
oS 58. 658+. 0027
Sieur pears ae os Saris 2 euiSiaecmseoes Den sdweaceenmen 58. 683 +. 0013
NOs diecma nas Pacveormsmsmacle ais aiidsdaduesaes 58. 687+. 0011
Asana curast Masaneennaeeesiaestaacecades 58. 707 +. 0380
US secoteaces css ahoesansasneuesaouseasuauateese 58. 788+. 0068
1 atereaaai ten saeea Sop eaae aaa ate 58. 917-4. 0172
Gs hie ee a cise aa nese ena AN 58. 984+. 0131
Dirge Saws tag ys tlk 65's <'eie seiarsrenememersladwmesie's Shai 59. 027+. 0141
Di aencke aaa ss sas sete Seasons Laie eeeiine cause 59. 030+. 0113
AB iiehcheie wkd obese 3 eee € SBS ecreee secede cece 59. 033+. 0231
Vo eweiceapetuiewe ys eeees see dea 2d Rebates ieee ee ees 59. 437+. 0250
General mean..........0-.00.2eeeeeeee eee Ni=58. 676+. 00064

This mean lies within the limits of variation of Richards and Cushman’s determinations:
and may be regarded as satisfactory. Their work and Ziromermann’ 8 dominates the entire
combination.

For cobalt we have eighteen ratios, as follows:

(1) Per cent. of Co in CoO, 78.6324 + .0002 |
(2) Per cent. of Co in purpureocobalt chloride, 23.6045 + .0018
(3) Per cent. of Co in phenylammoniun cobalticyanide, 11.8665 + .0124
(4) Per cent. of Co in ammonium cobalticyanide, 21.943 + .029
(5) Per cent. ‘of Co in brucia cobalticyanide, 3.7437 + .0036
(6) Per cent. of Co in strychnia cobalticyanide, 4.5705 + .005
(7) Per cent. of CoO in CoSO,, 48.287+.0135 —
-.. (8) 2Ag : CoCl, :: 100.: 60.1975 + .0006

(9) 2AgCl : CoCl, :: 100 : 45.307 +.0011

(10) 2Ag : Co :: 100 : 27.705.+.0050 .

(11) 2AgCl : Co :: 100 : 20. 687 + .0033 _

(12) Cl, : Co :: 100 : 83. 220+. 0082; fa

(13) OAg : CoBr, :: 100 : 101.407 + 0018

(14) 2AgBr : CoBr, :: 100 : 58. 2549 + .0007

(15) CoBr, : Co :: 100 : 26.9508 + .0008

(16) 2Au : 3Co :: 100 : 45.151 + .025

(17) Co : H, :: 100 : 3.4110 + .0009

(18) I, : Co :: 100 : 23.462 + .0027

Hence, for the atomic weight of cobalt,

From. ratio—
TRAN tll 1112 Goo to een anna at tant aca Co=58. 761+. 0286
lL cicotedtenucia castes tai Uaccaeacdabcadisewean, . 58, 880+. 0006
1B vette iaas ass Dac terael cre ien ciel enema 58. 950+. 0039
4 oh on eitctaAoeenne us aac Mensa aes Pelee ik eg Sons 58. 959+. 0027
BccAo setteece cache ayaa abel ea aas hes 58. 961+. 0013
Dato ehcccceen eine e Coes sameadcene 58. 966-4. 0032
LBs cisks oisiaissctsioials u'o's wlee ves cwide samara techies 58. 973+. 0019
Bee aseaial ce atin ta det etd ace tiedd Abt a 59. 017+. 0630
DD eae tei eye a hs tarye ta tt he tstehe me tactats Boe 59. 020-4. 0058
Anata h ces AD etre Unita rs US ORANGE adele a dies 59. 086+. 0810
VPA R ete ctcce ete dy tee ae ae eds 59. 091+. 0156
Ged GAS en as alice are ana tg aoa 59. 0954-. 0647
Draenei ond aay teat aterocaeaen ede OF uu oaeanacnue 59. 1814. 0047
5 ccctauals hamesea da Soa Seared eaeaieueed AaROe 59. 198-4. 0570

Lbs sp gccemnawisnn's o¥ee alc wlersmiamaaisuwasbichelele cele's.os 59. 306+. 0095
No. 8.] bee ATOMIC WEIGHTS—CLARKE: 845

From ratio— : oat
AUG seni cca tc cat abe id ed Sas access ated Bas Co=59. 361+. 0330

LB siassiaisicycye. 2 S.cle lye bAravahitaie oa atsreetarenes tele ete 0 Seated saves 59. 569-4. 0069
RO eects es OO eh toes ids ors pee cle cray cect dS erate cos 59. 776+. 0108
General mean............0.ceceeceeeeeecees Co=58. 913+. 0005

It is evident that in this combination, ratio 1, representing principally the work of Zim-
mermann, receives excessive weight. For that reason, and also on chemical grounds, the final
mean is probably too low. If, however, we arbitrarily assign to ratio 1 the “probable error”
and weight of the next best ratio, No.8, the general mean then becomes

Co = 58.956 + .0007

This is probably not far from the truth; but the change thus effected serves to illustrate the fact
that the rigorous mathematical combination i is not always conclusive. Although the mathe-
matical method is most useful, it cannot do away with the exercise of judgment as based upon
other knowledge than that shown in the mere figures.

That the atomic weight of cobalt is higher than that of nickel clearly appears from the
evidence. Nevertheless, attempts have been made to prove the opposite. For example,
Parker and Sexton? assert that in fifteen electrolytic comparisons of silver and cobalt, they
have obtained a mean value of Co= 57.7, which is lower than the atomic weight of nickel. Barkla
and Sadler,” in studying the permeability of metals to the secondary Réntgen rays, have found
that property to be a periodic function of the atomic weights. By interpolation in the periodic
curve so obtained they find values for Ni ranging between 61.2 and 61.6, whereas the currently
accepted atomic weight appears to be anomalous, at least as regards the physical property now
under consideration. These conclusions, however, cannot weigh very heavily as against the
clear chemical evidence. As for Parker and Sexton’s work, the authors give no details which
would furnish an adequate basis for discussion.

RUTHENIUM.

The atomic weight’ of this metal has been determined by Claus, by Joly, and by Vogt.
Although Claus* employed several methods, we need only consider his analyses of potassium
rutheniochloride, K,RuCl,. Thesalt was dried by heating to 200° in chlorine gas, but even then
retained a trace of water. The percentage results of the analyses are as follows:

Ru. uo . i KCl. : scl.
28. 96 40. 80 30, 24
28. 48 41. 39 - 30. 22
28. 91° 41. 08 30. 04 04
Mean, 28. 78. oe 409 30. 17

Reckoning directly from the percentages, we get the following discordant values for Ru:

From percentage OhnetAl cudau.ccenSiiGoncnetoawnat teins . Ru=103. 21
From percentage of 2KCl.........-..-. 0. cece eee ee eee cence “* ==104, 44
From percentage of 3C].........-.0.-0-eee eee e eee e eee e eee eee “* 35101, 48

These results are obviously of little importance, especially since two of them are not in
accord with the position of ruthenium in the periodic system. The work of Joly is more satis-
factory.* Several compounds of ruthenium were analyzed by reduction in a stream of hydro-
gen with the following results:

First, reduction of RuO,:

RuO,. Ru Per cent Ru.
2. 1387 1. 6267 76. 060
2. 5846 1. 9658 76. 058
2, 3682 1. 8016 76. 075
2. 8849 2. 1939 76. 046

Mean, 76. 060+. 0040
Hence, Ru= 101.67.

1 Nature, 76, 316. 1907. 2 Phil. Mag. (6), 14, 408. 1907. 3 Journ. prakt. Chem., 34, 435. 1845. 4Compt. Rend., 108, 946.

 
346 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. rE
Second, reduction of the salt RuCl,.NO.H,O :

Per cent Ru.
39. 78
39. 66

Mean, 39. 72+. 0405
Third, reduction of RuCl,.NO.2NH,C1 :

Per cent Ru.
29, 44
29. 47

Mean, 29. 455+. 0101

The analyses of ruthenium dioxide by Vogt ' were by reduction of the compound in hydro-
gen.: The data obtained with vacuum weights are as follows:

RuO,. Ru. Per cent Ru.
_ 1.11748 0. 84998 76. 062

- 39454 . 80009 76. 060

. 67499 . 51815 76. 023

. 74681 . 56800 76. 056

. 57902 . 44043 76. 064

. 40827 . 31049 76, 050

Mean, 76. 0525+. 0042
Hence, Ru= 101.62.

Vogt rejects the third of these analyses, which varies widely from the others, and finds
Ru= 101.66 in close agreement with Joly’s determination. The two series, Vogt’s and Joly’s
combined give a mean percentage of 76.0565 + .0029.

There are now three ratios available from which to compute the atomic weight of ruthe-
nium. Reducing them with Cl=35.4605 + .00011, N = 14.0081, +.00008, and H= 1.0078 +.00001,
we have—

From RuCl;.NO.2NHC1..........0. 2c sees eee cece ee Ru=101, 625+. 0378
From BuO :\a:5:s scieis siciaiaw eialewiarn pieinisiaipinia see 'e.8 8s ose sete 101. 648+. 0129
From RuCl,.NO.H,O......22...202.eeeee eee eee eects 101. 742+. 1242

General mean, Ru=101. 649+. 0121

This value appears to be a fair approximation to the truth, but more determinations by

other methods are desirable.
RHODIUM.

Berzelius? determined the atomic weight of this rhodium by the analysis of sodium and potas-
sium rhodiochlorides, Na;RhC], and K,RhCl,. The latter salt was dried by heating in chlorine.
The compounds were analyzed by reduction in hydrogen after the usual manner. Reduced
to percentages, the analyses are as follows:

In Na,RhCI,.

Rh. 8NaCi. scl.
26. 959 45. 853 27. 189
27. 229 45, 301 27. 470
eee aca 27. 616
Mean, 27. 094 Mean, 45. 577 Mean, 27, 425
In E,RhCl,,
Rh. 2KCl. scl.
28. 989 41, 450 29. 561

 

1 Situngsb. phys. med. Soz. Erlangen, 43, 268. 1912, 2 Poggend. Annalen, 13, 435. 1828,
No. 8.] ATOMIC WEIGHTS—CLARKE. 347

From analyses of the sodium salt we get the following values for Rh:

From per cent of metal. . 2.2.2.2... ccc cence ee ence neces Rh=104, 72
From per cent of NaCl... 2.0.0... 0 cece eee cece cece ce eeeees “ =103. 05
From per cent of 3C1. . ... 2... eee eee eee eee eee e eee eeee _* 106, 13
From ratio between 3Cl and Rh................0.000eeeeeeeee * =105. 10
From ratio between 3NaCl and Rh.................2..2-2-0 ‘* =104, 27

These are discordant figures; but the last one fits in fairly well with the values calculated
from the potassium compound, which are as follows:

From per cent of metal.......... 22.2. cece cece cece cece eens Rh=104. 30
Brom per cent of KCl. cicce sc s<s sanieedcasec es st < oexeieedce “ =104. 30
From per cent of Cl..... 2.2.2.2 eee cence eee eee eee eeee “ =104. 37
Brom Rh SC) ratio. cdo. ecccncae tac culcesccsanendaes see ‘* 104. 32
From Rh : 2KCl ratio. . ....-.....-..00-- eee e eee cece eee wee { =104, 31

Meatieis sis see se etd fosiiakaicieeeiee's'e Pa staisaiteeidembnisicleisietis Rh=104. 37

The determinations by Jérgensen 1 seem to have been preliminary, but are good so far as
they go. Rhodium pentamine chloride, Rh (NH,),Cl,, was ignited in hydrogen, and the residual
metal was cooled in an atmosphere of carbon dioxide. The data are as follows:

Chloride. Rhodium. Per cent Rh.
3. 5180 1. 2310 34. 991
2. 1507 . 7517 34, 951
- 9091 . 3182 35, 002
1, 9889 - 6960 34. 994

Mean, 34. 984+. 0076

Hence, Rh = 103.06.

In a single analysis of the corresponding bromide, 1.2736 grammes gave 0.3065 of rhodium,
or 24.065 per cent. Hence, Rh=102.97. In another experiment 1.2675 grammes of bromide
gave 1.6683 of AgBr. Hence, Rh=103.12.

Seubert and Kobbe? determined the atomic weight in the same way, that is, by igniting
rhodium pentamine chloride in hydrogen and weighing the residual metal. Their results are
given below:

Chloride. Rh. Per cent Rh.
1, 8585 0. 6496 34, 953
1, 5560 . 5435 . 84.929
1. 5202 . 5310 34. 930
2.0111 . 7031 34, 961
1. 8674 . 6528 34. 958
2. 4347 . 8513 34. 965
2. 3849 . 8338 34. 962
2. 5393 . 8881 34. 974
1, 4080 . 4920 34, 943
1. 4654 . 5123 34, 960

Mean, 34. 954-4. 0032
Hence, Rh = 102.92.

In the sixth experiment the ammonium chloride formed was collected in a bulb tube and
estimated by weighing as silver chloride; 3.5531 grammes of AgCl were obtained. Hence,

Rh = 103.12.
The same process was followed by Hittlinger,? who obtained almost exactly the same

result. His figures are as follows:

Chloride. Rh. ‘Per cent Rh.
1. 60574 0. 56124 34, 951
1. 67310 . 58492 34, 960
1, 30182 . 45507 34. 956

Mean, 34. 956-£. 0020
Hence, Rh = 102.94.

‘Journ. prakt. Chem. (2), 27, 486. 1883. 2 Liebig’s Annalen, 260, 218. 1890. 3 Sitzungsb. phys. med. Soz. Erlangen, 39, 1. 1907,

 
348 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von, XVI.

Another series, somewhat later; by H. Dittmar, gave the subjoined figures with vacuum
weights: :

Chloride. Rh. ' Per cent Rh.
2. 01526 0. 70465 34. 967
1. 83589 . 64173 34, 954
1. 57210 . 64934 34, 943
2. 17528 . 76046 34, 959
2. 03911 71271 34. 952.
2. 20000 . 76890 84.950
1. 02840 . 35941 34. 948

Mean, 34. 953+. 0020
Hence, Rh = 102.94.

The four series of analyses of the chloride combine as follows:

VOrRONSON ec es ee sooo os Mwesnindeamauiaenenmeasaneetaes 34, 984-4. 0076
Seubert and Kobbe.........-.-0 2-22 eee cece cece eee eee ee ete 84, 9544. 0032
HGttlinger... 02.6. eeeceeceoensicceccseeeenes ptdciciemestis 34. 956+. 0020
Difttiar: sess sac <sdue eee Sabhneaaensasceses keges pores. 34, 953+. 0020

General mean...........--------00e- ie nesta eee 34. Ubb 0013

The work of Hiittlinger and Dittmar was done in the laboratory at Erlangen, under the
direction of Gutbier. So, too, was that of Renz,? who made similar. analyses of rhodium penta-
mine bromide, Rh (NH,),Br,. His data, with vacuum weights, are as follows:

Bromide. Rh. Per cent Rh.
0. 87624 0. 21057 24, 031
1, 56500 . 37638 24. 049
2. 04033 . 49069 24. 049
2. 00120 . 48135 24. 053
1. 89278 . 45525 24. 051
2.30210 © . 55416 24, O71
1. 02065 . 24555 24 058
1. 31485 . 31622 24. 049
1, 86060 . 44766 |, 24.059
1. 51040 . 36339 24. 059

Mean, 24. 0534. 0022
Hence, Rh = 102.907.

Ignoring the early work of Berzelius, and the single analysis by Jérgensen of rhodium
pentamine bromide, we have two ratios from which to compute the atomic weight of rhodium:

(1) Rh(NH,),Cl,: Rh : : 100 : 34.955 4.0013
(2) Rh(NH,);Br,: Rh : : 100 : 24.053 + .0022

To reduce these we have—

Cl=35. 4605+. 00011 N=14. 0081+. 00008
Br=79. 9228+. 00015 H= 1.0078+. 00001
Hence,
From ratio—

Daca ch ueewiae peeeal ea he ee eaweanay «+. Bh=102. 907 +. 0099

Vy acaigss Sees Mash deena case's s ese ss cenan --- —-- 102. 9825-4. 0044
General mean, Rh=102. 928 +. 0040

PALLADIUM.

The first work upon the atomic weight of palladium seems to have been done by Berzelius.
In an early paper * he states that 100 parts of the metal united with 28.15 of sulphur. Hence
Pd=113.91, a result which is clearly of no present value.

In a later paper 4 Berzelius published two analyses of potassium palladiochloride, K,PdCl,.
The salt was decomposed by ignition in hydrogen, as was the case with the double chlorides of

 

1 Sitzungsb. phys. med. Soz. Erlangen, 40, 184. 1909. @ Poggend, Annalen, 8,177. 1826.
Inaug. Diss., Erlangen, 1909. 4 Poggend. Annalen, 13, 454. 1828.
No. 3] . ATOMIC WEIGHTS—CLARKE. 349

potassium with platinum, osmium, and iridium. Reducing his results to percentages, we get
the following composition for the substance in question:

 

 

Pd. 2KCl. Ci;.

32. 726 46. 044 21. 229

32. 655 45. 741 21. 604

Mean, 32. 690 Mean, 45. 892 Mean, 21. 416

From these percentages, calculating directly, very discordant results are obtained:

From percentage of metal............2--.222e eee eee eee ee ee . Pd=106. 86
From percentage of KCl... ........-.. 2... cece eee eee ce eee ee Pd=104. 90
From percentage of Cl, (loss)..........-2------ 20-2 eee eee eee Pd=111. 11

Obviously the only way to get satisfactory figures is to calculate from the ratio between
the Pd and 2KCl, eliminating thus the influence of water in the salt. The two experiments
give, as proportional to 100 parts of KCl, the following of Pd:

71. 075
71. 391

Mean, 71. 233. 1066 é
Hence, Pd = 106.22. :
In 1847 Quintus Icilius ! published a determination, which need be. given only for the sake

of completeness. He ignited potassium palladiochloride in hydrogen, and found the following

amounts of residue. His weights are here recalculated into percentages:

64. 708

64. 965
64. 781

Mean, 64. 818

From this mean, Pd=112.05. This result has no present value.

In 1899 Keiser’s first determinations of this constant appeared.? Finding the potassium
paliadiochloride to contain ‘‘water of decrepitation,” he abandoned its use and resorted to
palladiammonium chloride, Pd(NH,Cl),, as the most available compound for his purpose.
This salt, heated in hydrogen, yields spongy palladium, which was allowed to cool in a current
of dry air, in order to avoid gaseous occlusions. The salt itself was dried, previous to analysis,
first over sulphuric acid, and then in an air bath at a temperature from 120° to 130°. Two
series of experiments were made, the second series starting out from palladium produced by
the first series. The data are as follows:

 

First Series.

Pd(NH,Cl)>. Pa. Per cent Pd.
0. 83260 0. 41965 50. 402
1. 72635 . 86992 50. 391
1. 40280 . 70670 50. 378
1. 57940 . 79562 50. 375
1. 89895 . 95650 50. 370
1, 48065 . 74570 50. 363
1. 56015 - 78585 50. 370
1. 82658 . 92003 50. 369
2. 40125 1. 20970 50. 378
1, 10400 . 55629 50. 389

. 93310 . 47010 50. 380

 

Mean, 50. 379+. 0008
Reduced to a vacuum this becomes 50.360.

Hence, Pd = 106.51.
“« 1Die Atomgewichte vom Pd, K, Cl, Ag, C, und H, nach der Methode der kleinsten | Quadrate berechnet.”” Inaug. Diss. Gottingen, 1847.

Contains no other original analyses.
2 Amer. Chem. Journ., 11, 398. 1889.

 
350 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVL

Second Series.

Pd(NH;,C1).. pg "Per cent Pd.
2. 61841 1. 31900 50. 374
2. 23420 1.12561 50. 881
1.73553 . 87445 50. 385
1. 69160 . 85210 50. 372
1. 72403 - 86825 50. 362
I, 12222 56535. 50. 378
1. 17457 . 59200 50. 401
2. 42760 1, 22280 50: 371

Mean, 50. 378: 378-4. 0028

Reduced to vacuum, 50. 359
Hence, Pd = 106.50.

The reductions to a vacuum are neglected by Keiser himself, but are here added in order to
secure uniformity with later results by the same author.

Bailey and Lamb‘ made experiments upon several compounds of palladium, but finally
settled upon palladiammonium chloride, like Keiser. Two preliminary experiments, however,
with potassium palladiochloride are given, in which the salt was reduced in hydrogen, and both
Pd and KCl were weighed. The data are as follows, with the ratio (calculated as with Ber-
zelius’s experiments) given in a third column: Ve ta 8

* PRR oon _ Pa. a nae:

“49767 7 /_ 1.05627 a “70.528 5 oe
90484 .63738 a 70.441

Mean, "70.485 +.0290
Hence, Pd=105.11.

The palladiammonium chloride was studied by two methods. First, weighed quantities
of the salt were reduced in hydrogen, the ammonium chloride so formed was collected in an
absorption apparatus, and then precipitated with silver nitrate. The weights found were as
follows, with the Pd(NH,Cl,) proportional to 100 parts of silver chloride given in the third
column:

| Pd(NHCD;. AgCl.’ Ratio.
1.24276 1.682249 73.879
1.08722 1.468448 74.040
1.47666 2.000164 73.828
1.34887 1.837957 , 73.390
"1.74569 . 2.362320 73,898

Méan, 73.807 +.0742°

Hence, Pd=106.61. Bailey and Lamb regard this as too high, and suspect loss of NH,Cl
during the operation.

The second series of data resemble Keiser’s. The salt was reduced in hydrogen, and the
spongy palladium was weighed in a Sprengel vacuum. The data are as follows:

Pd(NH,Cl)». Pa. Per cent Pd.
ie ‘0.947995 50.143
Alt. 874175 940271 50.170
1.307076 654687 50.088
pl 1.340045 633207 50.238
1.905536 955950 50.167
1.685582 846472 50.218
1.691028 849120 50.213
oo} 2.112530 1.059690 50.162
2.110653 1.057910 50.122
1.969100 .988155 50.184

Mean, 50.171+.0099

Hence, Pd=105.71. Bailey and Lamb’s weighings are all reduced to a vacuum.
Keller and Smith,? reviewing Keiser’s work, find that palladiammonium chloride, prepared
as Keiser prepared it, may retain traces of foreign metals, and especially of copper. Accordingly,

 

1 Journ. Chem. Soc., 61, 745. 1892. 2 Amer. Chem. Journ., 14, 423. 1892.
No. 3.) ATOMIC WEIGHTS—CLARKE. 351

they prepared a quantity of the salt, after a thorough and elaborate process of purification,
dried it with extreme care, and then determined the palladium by: electrolysis i in silver-coated
platinum dishes. The precipitated palladium was dried under varying conditions, concerning
which the original memoir must be consulted, and was proved to be free from occluded hydrogen.
By this method two sets of experiments were made to determine the atomic weight of palla-
dium; but for present purposes the two may fairly be treated as one. The data obtained are
as follows, but the weights do not appear to have been reduced to a vacuum:

Pd( NH,C1),. Pa. Per cent Pd.
1.29960 0.65630 50.504
aftosen 153253 50.510
‘(1.92945 .97455 50.509
1.94722 98343 50.504
1.08649 .54870 50.502
1.28423 .64858 50.503
By 1.68275 .85010 50.519
1.69113 “185431 - 50.517.
1.80805: 391310 oe 50.502.

ie op Herc a Mean, 50.508 4.0014

Hence, Pd =107.14, a result notably higher than Keiser’s.

Keller and Smith account for the difference between their determinations and Keiser’s
partly by the assumption that the materials used by the latter were not pure, and partly by
considerations based on the process. In order to clarify the latter part of the question they
made three sets of experiments by Keiser’s method, slightly varying the conditions. First,
the chloride was not pulverized before ignition, and slight decrepitation took place, while dark
stains of palladium appeared in the reduction tube, indicating loss by volatilization. Secondly
the chloride was prepared from crude palladium exactly as described by Keiser, but was pulver-
ized before reduction. No decrepitation ensued, but traces of palladium were volatilized.. The
third seriés, also on finely pulverized material, was like the second; but the palladiammonium
chloride was purified by Keller and Smith’s process. The three wemies ‘here treated as one, are
28 follows:

Pd(NH,C1), Pa. Per cent Pa,
62955 0.31743 50.422
ease 77210 380420 50.397
“<1” g3059 A918 50.350
99055 49895 50.371
1.02175 51468 50.372
1.10325 55590 50.388
; .66690 .33590 50.367
Second eeries..) g¢g4q 43733 50.360
1.41430 71255 50.382
; ~ (1.15234. .58050 50.376
~ 96229 48502 50.403
ae .97804 49294 50.401
Third series...) ‘94053 ATBIT - 0 §0,414
86090 43405 50.430

 

"Mean, 50.88 50. 388+. os
Hence, Pd =106.63. ss .
The three series seem to be: fark ‘in: apnaaane ‘babwsen “ihieradelees and with. Kaci

work, but diverge seriously from the.electrolytic data. 9.1. »

. Keller and Smith also attempted to determine the atomic weight . of palladiain by featug

the palladiammonium chloride in sulphuretted hydrogen, an so pie it into. ithe |

Pd8.: Thee: data were obtained: Es i:
Pd NH,CD>. _ Pads. “> Pep cent PdS.

0.71699 , 0.47066: ee. 85,644
1.31688 86445 65.659

Mean, 65.65 GDL. £0061
« 161566°—20-——23. , he NE a a
352 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

Hence, Pd=107.15. This result, however, is affected by the work of Petrenko-Kritschenko,!
who has shown the existence of the sulphide PdS to be uncertain.

Joly and Leidié,? in their determinations of this atomic weight, returned to the potesiiom
palladiochloride, K,PdCl,. In their first series of experiments the salt was dried in vacuo at
ordinary temperatures. It was then electrolyzed in a solution acidulated with. hydrochloric
acid, both the deposited palladium and the potassium chloride being weighed. The palladium
was dried, ignited in a stream of hydrogen and cooled in an atmosphere of carbon dioxide.; The
results were as follows, with the column added giving the Pd equivalent to 100 parts of KCl:.

K,PdCl,. Pa. 2KCl. Ratio.
1, 0255 0. 3919 0. 5520 70. 996
1, 2178 . 3937 . BBL 70. 924

1, 2518 . 4048 . 5687 71. 016
Mean, 70. 979-4. 0188

 

Hence Pd = 105.84.

This series was rejected by the authors, because the salt was found to contain water—in
one oase 0.23 per cent. This error, however, should not invalidate the Pd : KCl ratio. In a
second series the palladiochloride was dried in vacuo at 100°, giving the following data:

K,PdCl,. Pd. 2KCl. | Ratio.
1. 3635 0. 4422 0. 6186 71. 484
3. 0628 . 9944 1. 3929 71. 391
1. 4845 . 4816 . 6782 71. 011
1. 7995 . 5838 . 8206 ' “71.143

 

Mean, 71. 257-b. 0736

These experiments seem to be less concordant than the preceding set. It must be noted,
however, that the authors reject the KCl determinations and compute directly from the ratio
between the salt and the metal. But the ratio here chosen agrees best with the determinations
made by other observers, giving for this series the mean value. Pd= 106.26, and:i . moreover,
uniform with the data given ‘by Berzelius and by Bailey and Lamb.

Joly and Leidié also give two experiments made by reducing the K,PdCl, in uydesgen;
with the subjoined results:

K,PdCh,, Pd. 2KCl. Ratio.
2, 4481 0. 7949 1. 1168 71.177
1. 8250 . 5930 . 8360 70. 933

 

Mean, 71. 055-4. 0823
Hence Pd= 105.96.
Combining these data with previous series, we have—

Berzelius........-.....-- og Reavis ieee eevee 71. 233.4. 1066
Bailey and, Lamb. 22:0. <ccjsiecieceeennbst sce cnscgwicewiar ses 70. 485+. 0290 ‘
Joly and Leidié, first.............. eu thy ietialsiels poseneeneees 70. 979+. 0188
Joly and Leidié, second.............-.--....-222- 2022 ee ee 71. 257+. 0736
Joly and Leidié, third..............2.....----.022-.-2-22+- 71 0554. 0823
Geriéral Meany oo osccscseseiekceascascieds eeee sees 70. 865-4. 0150

In view of the discordance among the determinations hitherto cited and because of the
criticisms made by Keller and Smith, Keiser, jointly with Miss Mary B. Breed,’ repeated his
former work, with some variations, and new precautions to insure accuracy. His general
method was the same as before, namely, the reduction of palladiammonium chloride by a
stream of hydrogen. First; palladium was purified by distillation as PdCl, at low red heat-in
a current of chlorine. From this chloride the palladiammonium salt was then prepared. Upon
heating the compound gently in a stream of hydrogen, decomposition ensued absolutely without
decrepitation or loss of palladium by volatilization. Neither source of error existed. The
results obtained were these:

 

1Zeit. anorg. Chem., 4, 251, 1893. Compt. Rend., 116, 147. 1893. ®Amer, Chem. Journ., 16,20, 1804.
No. 3.] ied ATOMIC WEIGHTS—CLARKE. 353

Pd(NH,C1)3. Pa. Per cent Pd.
1, 60842 0. 80997 50. 358
2. 08295 1. 04920 50. 371
2. 02440 1. 01975 50. 373
2. 54810 1. 28360 50. 375
1. 75505 . 88410 50. 375

; Mean, 50. 370+. 0023
' Reduced to a vacuum, 50. 351
Hence Pd = 106.46.

In a second series of experiments, palladium was purified as in the earlier investigation,
but with special care to eliminate rhodium, iron, copper, gold, mercury, etc. The palladiam-
monium salt prepared from this material gave as follows:

Pd(NH,Cl)o. Pd. . Per cent Pd.
1. 50275 ; 0, 75685 50. 364
1. 23672 - 62286 50. 365
1. 34470 . 67739 50. 375
1. 49059 . 75095 50. 379

Mean, 50. 371+. 0026
Reduced to a vacuum, 50. 352

Hence Pd = 106.47.

Here, again, no loss from decrepitation or volatilization occurred, although evidence of
such loss was carefully sought for.

“Hardin, in 1899, made three series of determinations of the atomic weight of palladium,
by reduction of three palladium : salts i in hydrogen. His results, with weights reduced to a
vacuum, are as follows:

First: Reduction of diphenyl-pallad-diammmonium chloride,

Pd(C,H,.NH,Cl),:
Salt. Pd. Per cent Pd.

0. 98480 0. 28953 29. 400
1. 10000 . 32310 29. 376
1, 02820 . 30210 29. 381
1. 19230 . 35040 29. 389
1. 40550 - 41300 29. 385
1. 26000 . 37040 29. 397
1, 25510 . 66310 29, 404

Mean, 29. 390-4. 0029
Hence Pd = 107.00.

Second: Reduction of diphenyl-pallad-diammonium bromide,

Pd(C,H,;.NH,Br),:
Salt. Pd. Per cent Pd.
0. 88567 0. 20917 23. 617
1, 31280 . 31000 23. 614 ©
1. 50465 . 35540 23. 620
2. 01635 . 47635 23. 624
2. 92300 £69080 23. 633

Mean, 23. 622+. 0023
Hence Pd=107.01. a
Third: Reduction of palladium ammonium bromide, (NH,),PdBr,:

Salt. . Pd. Per cent Pa.
0. 77886 0. 18006 23. 118
1. 53109 . 35881 ~23, 108
2. 75168 . 63614 23. 118
1. 88136 . 48478 23. 110

Mean, 23. 1135+. 0018
Hence Pd = 106.95.

 

1 Journ, Amer. Chem. Soc., 21, 947.
354 MEMOIRS NATIONAL ACADEMY. OF SCIENCES. (Vou. XVI.

These determinations are notably higher than those made by other methods. After
reduction, the palladium was heated to redness for two hours in a stream: of dry air, to remove
possible carbon. It was then heated again in hydrogen, and finally cooled in a current of air.
Hydrogen could hardly have been occluded in the final product.

Amberg,’ whose determinations appeared in ‘1905, resorted to palladiammonium chloride
as his initial substance. ‘Three series of analyses were made, with scrupulously purified material,
and all weights were reduced to a vacuum. First, the salt was reduced electrolytically. The
precipitated palladium was weighed, and in the rinsings from it the chlorine was determined
as silver chloride. The:data are subjoined, with one Fatio 2AgCl : Pd(NH. i), : ::100: ain

the fifth column: . - cage) *
Salt. Pa. AgCl. ' Pa cent Pd, - AgCl1 ratio. Vb,

1. 06045 0.53609 ss eee 50.553 =. ee
1, 00028 . 50528 1. 35867 50. 493 73. 622
1. 66386 . 84085 “2, 25437 50.541 73. 807
. 83195 . 42092 1. 12282 50, 594 74, 095
1, 91591 . 96886 2. 59799 50. 569 "73.746
e Mean, 50.550 Mean, 73. 818

+.0110 +. 0677

From the percentage of metal, Pd=107.32.

From the AgCl ratio, Pd= 106.64.

Amberg’ s secondseries of analyses resembles the first, except that the palladium was Brecipi-
tated by hydrazin sulphate. The Percentage of metal i is s given by Amberg, but not the weights
actually obtained: oe, aos

Salt. ‘AgCt. ame rs AgCI ratio.

1. 32493 1.78656 50. 12 74.171 |
1. 02642 “1.39947 50. 29 73. 712
1, 30335 1. 76875 50. 36 73. 518
1.59709 2.16644 = 9 .eeeeee 73. 721
1. 88622 2. 55028 50. 49 73. 961
2. 59665 3. 51783 50. 68 73. 812
- -' Mean, 50. 388 Mean, 73. 814

? +. 2064 +. 0614

From percentage of metal, Pd = 106.63.
From AgCl ratio, Pd = 106.63.
The silver chloride ratio combines with previous determinations as follows:

 

Bailey and! Lambe ies scccvessexieeeeesiee sees svencuiesews eye « 73. 807 +. 0742

Amberg, first series.........--.2--2---eeeeceeeeeeeeeeee ees 73. 818+. 0677

Amberg, second series. . fe Spe kiteehieka can . 73. 8144+.0614 ...
General mean....... seruperpretepprysgrertetessece 73. 813+. 0388

PY Bh is
In his third series of analyses Amberg uel only the palladium, which was precipi-
tated electrolytically from a ‘sulphuric acid solution of the palladiammonium chloride, with a
rapidly rotating anode. . This series is excellent, but the preceding series of palladium determi-
nations are negligible: oe
ai

3 Pd. Per ea Pd.

0. 62446 0. 31470 - 60. 396

. 88878 . 42280 50. 407
1. 50282 £75725 50. 389
1.06704... 58763, «80.385
1. 98342 . 99971 50. 403
1. 53093 . 77153 50. 896
1. 18905 . 59971 50. 398

| 62635 . 31572 50. 406
1. 76110° . 88739 50. 388
3. 79639 1. 91298 50. 389
3.97553 2. 00333 50. 392
4. 62100 2. 32834 50. 386

Mean, 50. 395+. 0015
Hence, Pd = 106.66.

 

1 Liebig’s Annalen, 341, 255,

'

‘
No. 3.]

ATOMIC WEIGHTS—CLARKE. .

359

The atomic weight determinations by Krell, Woernle, and Haas were all made in the labora-
Krell! reduced palladiammonium (palladosamine)
chloride in hydrogen, and afterwards heated the. reduced metal in a stream of carbon dioxide.
His figures, with vacuum weights, are subjoined:

tory of Professor Gutbier at Erlangen.

Salt.
1. 83034
1. 73474
1, 92532
2. 63544
3, 23840

Woernle? also made analyses of palladiammonium chloride.
were effected in hydrogen, the other determinations were electrolytic.
weights, are as follows: —

Salt.
2. 94682
1. 83140
1. 02683
1, 22435
1, 46735
. 59796
© 2. 64584

ve

~ Hence, Pd- 106.66.

Haas * analyzed palladiammonium bromide, Pd (NH;Br),, by reduction i in hydrogen.
reduced metal was subsequently heated in carbon dioxide.
are as follows:

Bromide.
2. 06470
1. 73455
2.64773
1. 29106
2, 26758
1. 90770
1.77729

Hence, Pd = 106.69.

_ Pad.
0. 92197
. 87433
. 96524
1. 32868
1.68175

ied

| Pd.

1.48493
. 92296
- 51479
. 61708
. 73044
. 30139

1. 33329

Pd.
0.73274
| 61563
. 93978
. 45821
. 80490
. 67704
. 63082

Per cent Pd.
50. 372
50. 401
50. 396
50.416
_ 50.387
Mean, 50. 3945+. 0050.

 

Per cent Pd.
50. 391
50. 396
50. 397
50. 401
50. 393
50. 403
50. 392

 

_ Mean, 50. 396+. 0012

(ye tp ae

Hence, Pd =106.65. The first, aiaeant ie fists in the series, is rejected by Krell.

The first two reductions
His figures, with vacuum

The

His data, with vacuum weights,

Per cent Pd.
35. 488
35. 492
35. 493
35. 491
35. 495
35. 489
85. 493

Mean, 35. 492+. 0006

 

Kemmerer‘ analyzed two palladiammonium compounds, the chloride and the cyanide,

data are subjoined:

Chloride.

. 89187
77981

Aé . 66980
1. 08373

. 96048

¥

95615 |

. 94087
By 90106
cae 1.16994

Hence, Pd = 106.39.

cia

1 Inaugural Dissertation, hea 1906.

of nitrogen. Vacuum weights are given throughout.

oR

0. 44885
. 39218

. 83711

. 54641
- 48338

, 48129
«47356

. 45353

. 58908

2 Sitzungsb. phys. med. Soz., Erlangen, 38, 278. 1907.

Pd.

both by reduction in hydrogen, with subsequent cooling of the reduced metal in an atmosphere

With the chloride, two sets of determi-
nations were made, on two ae eee but both series are here treated as one. The

8 Inang, Dissertation, Erlangen, 1908.
4 Thesis, University of Pennsylvania, 1908.

seboyit t,

‘Per cent Pa.
~ 50. 327
50. 324.
50. 330
50. 327
50. 327
50. 336

, 50. 332
50. 333
50. 351

~ Mean, 50. 332-4. 0018
356 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou, XVI.
With the cyanide, Pd (NH,CN),, Kemmerer obtained the following results:

Cyanide. Pd. Per cent Pd.
0: 85860 0. 47463 ; 55. 280
1. 19378 - 66002 ' 55, 288
1. 41818 . 78408 55. 288
“1.05254 . 58206 55, 301
1, 39510 . 77153 55. 303
1. 66196 . 91881 55. 285

Mean, 55. 291-4. 0025
Hence, Pd = 106.46.

Like Haas, Gebhardt ' also made analyses of palladiammonium bromide, and by the same
method. His figures, with vacuum weights, are as follows:

Bromide. Pde Per cent Pd.
3.32462 1.17984 35. 488
2. 68383 95245 oy. , 85.488
1, 40117 . 49731 35. 492
2. 61673 . 92877 35. 494
2. 64229 . 93787 35. 495
2. 54424 . 90293 35. 489
2. 40456 . 71148 35. 491

Mean, 35.491 + .0007
Hence, Pd = 106.68.

The work of Haas and Gebhardt was done under the direction of Gutbier, who has com-
bined their material ‘in a memoir bearing their names in joint authorship with his.? In this
memoir 10 additional analyses of the bromide are given, but 6 of them are rejected by Gutbier
as unsatisfactory. I prefer, however, to include them in this discussion, but with low weight.
The 10 determinations I have divided into two sets, one containing the 4 preferred analyses,
the other the 6 questionable ones. Their unequal value appears in the probable elrors: —

Bromide, Pd. Per cent Pd,
3.09278 1.09783 35. 496
1.98039 «70288 35. 491
1. 50032 - 53253 35. 494
2. 84500 1.00992 35. 498

ers Mean, 35. 495-+.0011
Hence, Pd= 106.70.

Bromide. Pd. Per cent Pd.
0. 54402 0. 19286 35. 450

- 80237 . 28468 35. 479

- 91601 . 32509 35. 489

. 42942 . 15288 35. 461

', 76884 . 27271 '" 35.470

. 62795 “22270 35. 464

; ’ Mean, 35. 469+. 0038
Hence, Pd = 106.58.

The four series of analyses of the bromide now combine thus:

Haig 532 Sse eisai ainnwnco asin aaah sean eee 35, 4924. 0006
Gebhardt...................2005 S eidia cha i AD casyaud oad Mate ae 3 5. 4914.0007
Gutbier, Haas, and Gebbardt..............0.022000seeeeeee 35, 495+. 0011
Gutbier, Haas, and Gebhardt .............-..---2--2-.05- 35. 469 +. 0038
General mean... ... 2.2.20. cence cece ee eee eee cence 35. 491+. 00042

The influence of the determinations rejected by Gutbier is insignificant.

One more series of analyses of palladiammonium chloride is due to Shinn.’ The salt was
dissolved in water, and from the solution the metal was precipitated by ammonium formate.
The data, with vacuum weights, are as follows:

 

1 Sitz. phys.-med. Soz., Erlangen, 40,65, 1909,
- § Gutbler, Haas, and Gebhardt, Journ, prakt. Chem. (2), %, 457. 1909.
"Journ, Amer, Chem. Soc., 34, 1448, 1912.
No, 3.]

Per cent Pd.
50. 400
50. 392
50. 410
50. 428
50. 401
50. 386
50. 429
50. 385

ATOMIC WEIGHTS—CLARKE.

Chloride. Pd.
1.03021 0. 51923
1. 45506 . 73323
1. 10384 . 55645
. 88141 44448
1.53091 . 77160
1.98887 1.00212
1. 45102 . 73174
1.57909 . 79563
1.61705 . 81583

Hence, Pd= 106.704.

50. 452 .
Mean, 50, 409+. 0040

357

The 14 series of determinations « of the percentage of palladium i in palladiammonium chlo-

ride now combine as follows:

Keiser, Grot seried...ccee-<-<« eng an eeemirasi eisieadc tes sh 50.360 +.0008
Keiser, second series...... gi duis Nec ae oie icrs aOeisiamtaenas ---. 50.359 +.0028
Bailey and Lamb...... ae ete Comet cas mr Rtntesa ieee 50.171 +.0099
Keller and Smith, electrolytic.....................-....--- 50.508 4.0014
Keller and Smith, hydrogen series............-.--.-.---++-- 50.388 +.0043
Keiser and Breed, first series.............--.0-0--2-e-eee 50.351 +.0023
Keiser and Breed, second series..............-------+-+e+-+- 50.352 +.0026
Amberg, first series..........-. Sees Pee teegeeeouwes sted 50.550 +.0110
Amberg, second series........-......- aaeaaainsaaien me neee 50.388 +.2064
Amberg, third series. ......-.2....00-. 202020 e eee eee cece 50.395 +.0015
Krellicccaccotccdraicewnteceadsadcdbacnasessceccedesesad 50. 3945+ . 0050
Wert @sccc nica cits Sages Meena sai scemche ty 50.396 +.0012
Memmererssisielant sv e'sti ataants 2 Seisloteh co's ok ete eS 50.332 +. 0025
Shins. cavsersesaseevscseerteee i scerne sedis aeueres Goees 50.409 +.0040
Gonderal ‘m6an ss aisjcieis'e e ciniciore sSacbivs sabes dslacmeies 50. 3885+. 0005

Nine ratios are now available from which to compute the atomic weight of palladium, as

follows:

(1) 2KCI : Pd :: 100 : 70.865 + .0150
(2) 2AgCl : Pd (NH,Cl), :: 100: 73.813 + .0388
_ (8) Pd (NH,Cl), : Pd :: 100 : 50.3885 + .0005
(4) Pd (NH,Br), : Pd :: 100 : 35.491 +.00042
(5) Pd (NH,CN), : Pd :: 100 : 55.291 4+ .0025
(6) Pd (C,H,.NH,Cl), : Pd :: 100 : 29.390 + .0029
(7) Pd (C,H, : NH,Br), : Pd :: 100: 23.622 + .0023
(8) (NH,),PdBr, : Pd :: 100 : 23.1135 +.0018
(9) Pd (NH,Cl, : PdS :: 100 : 65.651 +.0051

- To reduce these ratios we have—

Ag=107. 880+. 00016 C=12. 0025+. 00019

Cl =35. 4605+. 00011 N=14,0081 +. 00008

Br =79. 9228+4..00015 —S=32. 0650-4. 00012

K =39.0975+.00011 H= 1.0078+.00001

Hence, eg
From ratio—

ED sales ahGecrcie ea een peat eee Cane ial te Pd=105. 671 +.0224
A vcpeune ss o> sas Guabt aeckdas est sapseeee este sees ~ 106. 465-+4.0077
9 tn Tageaantinedsceae pe eaacaenteenede se eteeenee 106. 6244.1112
PD opuisdece aun an i ulaad ne nhaanetisgusans essays 106. 628+. 0015
Ain ssgeaeg cag duauntoese seem ate yaaa RE nite 106. 6834. 0015
By ietjc ce atu MA bes NNEnGIs: ne ooe eects 106. 9514. 0087
Cea OS Se loes oaasrernepnah aed ths evaaecnk Hedi le 107.0034. 0115
Fc caguaasanacea Dyegenene shea veuseemesches aa sys 107.010.0109
Goes vice duaien sates eetne tataeemenata cane 107. 149+ .0490

‘Pd=106. 659+. 0010

The international value, Pd= 106.7, is probably not far from the truth, but may be a trifle

too high. Ratio 4i is in good agreement with it.
358 MEMOIRS NATIONAL ACADEMY OF ‘SCIENCES. [Vou, XVI.

OSMIUM.

The atomic weight:of this metal has been determined by Berzelius, by Fremy, by Seubert,
and by Seybold. He”

Berzelius * analyzed potassium osmichloride, igniting it in hydrogen like the corresponding
platinum salt. 1.3165 grammes lost 0. 3805 of chlorine, and the residue consisted of 0.401
gramme potassium chloride, with 0.535 gramme of osmium. eee only from the ratio
between the Os and the KCl, the data give Os= 198.94.

Fremy’s determination? is based upon'the composition of osmium tetroxide. No details
as to weighings or methods are given; barely the final result is stated, pbamely, Os = 199,65.

When the periodic law came into general acceptance, it became clearly ‘evident that both
of the foregoing values for osmium must be several units too high. A redetermination was
therefore undertaken by Seubert,? who adopted methods based upon that of Berzelius. First,
ammonium osmichloride was reduced by heating in a stream of ' ‘hydrogen. The residual
osmium was weighed, and the ammonium chloride and hydrochloric acid given off were col-
lected in a suitable apparatus, so that the total chlorine could be estimated as silver chloride.
The weights were as follows:

Am,08Ci,. Os. 6AgCl.

T. 8408 0. 7996 3. 5897
2. 0764 . 9029 4. 0460
2. 1501 . 9344 4. 1950
2. 1345 - 9275 4. 1614

Hence we have for the percentage of osmium and for the osmichloride proportional to
100 parts of AgCl—

Per cent Os. AgCl ; Salt.

-. 48, 446 51. 266
43. 484 _ 51.320

"43, 458 / 51. 254
43, 453 _ 51, 293

., Mean, 51. 283+. 0099

In a later paper ‘ two more reductions are given, in which only osmium was estimated:

Salt. 0s. “Per cent Os.
2. 6687 ‘11597 48, 456
2. 6937 1.1706 id 43, 457

These determinations, included with the previous four as ‘one series, give @ mean per-
centage of Os in Am,OsCl, of 43.459 + 0036.

Secondly, potassium osmichloride was treated in the same way, but the residue weighed
consisted of Os+2KCl. From this the potassium chloride was dissolved out, recovered by
evaporating the solution, and weighed separately. The volatile portion, 4HCl, was also deter-
mined by precipitation as ‘silver chloride. In Seubert’s-first paper these data are given:

E,0sCly.,. aOR 2KCl. —. $AgCl.
2.5148 hae 0. 7796 2. 9837 s
2. 1138 0. 8405 . 6547 .2, 5076
Hence, with salt proportional to 100 parts of Ag(l in the last column, we have—
Per cent Os. Per cent KCI. AgCl : Salt.
viyuad 31. 000 84. 091
39. 762 30.973 84. 102

Mean, 84. 097+. 0030

In his second paper Seubert gives fuller data relative to the potassium osmichloride, but
treats it somewhat differently. The salt was reduced by a stream of hydrogen as before, but

 

+ Poggend. Annalen, 13, 530, 1828. .,  # Ber, Deutsch. chem. Geg., 21. 1839, 1888.
2 Compt. Rend., 19, 468. Journ. prakt. Chem. al, 410. 1844, 4 Liebig’s Annalen, 261, 258.
No. 3.]

ATOMIC WEIGHTS—CLARKE.

359

after that the boat containing the Os+2KCl was transferred to a platinum tube, in which,
by prolonged heating in the gas, the potassium chloride was completely volatilized. The

determinations of 4Cl as 4AgCl were omitted. _ Two series of data are given, as follows:
K,0sCl,.

1, 1863
- 9279
1. 0946
1. 6055
- 4495
- 8646
- 7024
1, 2742
1. 0466

K,0sCl,.

2. 2032
2. 0394

2. 7596 .

2. 4934
2. 8606
2. 8668
1. 2227

The single percentage of osmium given in Seubert’s first paper is negligible.

. Os. : Per cent Os. .
0. 4691 39. 543
. 3667 39, 519
. 4330 39. 558
. 6351 39. 558
1778 39. 555
8417 39. 521
. 2781 38. 593
5041 39. 562
4141

.Mean, 39. 553-4. 0052

2KCl. Per cent KCI.
0. 6820 30. 955
- 6312 30. 950
. 8544 30. 961
. 7710 30. 922
- 8843 30. 913
. 5768 30. 898
. 3778 30. 899
Mean, 30. 931
- . {31. 000
Earlier sot {ey 973

Mean of all nine determinations, 30. 941+. 0079

39. 566

 

 

 

The analyses of ammonium osmichloride by Seybold' are of slight importance because

they are not concordant. ‘The data are as follows. They fall into two groups:

Group 1.

Am,08Cl,. Os. Per cent Os.
0. 92831 0. 40110 43. 209
1. 37982 . 59629 43, 215

. 86661 . 37448 43. 212
1.17170 . 50642 43. 221
1, 31312 . 56730 43. 203

C4 43, 212

Group 2.
4, 94681 2, 14915 43, 447
3. 78304 1. 64378 43, 452
1. 44792 . 62958 43. 412
43. 437

 

 

_ Seybold regards the high values given in the second group as due to the probable absorption
of oxygen by the osmium. It is equally probable that the osmichloride of the first group
contained water, in other words, it had not been thoroughly dried.

It would be justifiable to reject Seybold’s work arbitrarily, but is perhaps better to give
it some weight, and to regard the two groups as one, with a mean percentage of 43.296 + .0257.

This value, combined with that of Seubert, 43.459 + .0036, gives a general mean of

Am,OsCl, : Os :: 100 : 33.452 +.0035.

Seybold’s deteminations practically reject themselves. -

1 Inaug. Dissertation, Erlangen, 1912.
360 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVI.

The ratios for osmium are now as follows:

(1) Per cent Os in Am,OsCl,, 43.452 +.0035.
(2) 6AgC] : Am,OsCl, =: 100 : 51.283 +.0099.
(3) 4AgCl : K,OsCl, ::]100 : 84.097 + .0030.
(4) Per cent Os in K,OsCl,, 39.553'+.0052.
(5) Per cent KCl in K,OsCl,, 30.941 +.0079.

To reduce these values we have—

Ag=107. 880 +. 00016 N=14. 0081 +. 00008
Cl=35. 4605+. 00011 K=39, 0975+. 00011
H=1. 0078+. 00001
Hence,
From ratio—

Asse pccaise Be cehtidel lisa dieters toss rte tees Os=190. 386+. 0299
heh eravtsa cwcsansrss RSA ahacic ah einai oh ete ath cre poet eae 190. 979+. 0933
Diseryaniiian deepal ees sie dees e Ce dwaw Beas eaeiven 191. 122+. 0173
Deas las as ee ayy Soa a eee te eee are alge iate een a 191, 212+. 0194
DSneek aie selec kd SUMED SEE eR eS 88 See eeReT eke 192. 214+. 0852
: General mean............... 0-20 ee eee e eee cane Os=191. 073+. 0117

This value is by no means satisfactory. New determinations of the atomic weight of

osmium, are needed.
IRIDIUM.

The only early determination of the atomic weight of iridium was made by Berzelius,!
who analyzed potassium iridichloride by the same method which he employed with the plati-
num and the osmium salts. The result found from a single analysis was not far from Ir= 196.7.
This is now known to be too high: I have not, therefore, thought it worth while to recalculate
Berzelius’ figures, but Eve his eatiniation as it is stated in Roscoe and. Schorlemmer’s “Treatise
on Chemistry.’

In 1878 the matter was taken up by Seubert,? who had at his disposal 150 grammes of pure
iridium. From this he prepared the iridichlorides of ammonium and potassium (NH,),IrCl,
and K,IrCl,, which salts were made the basis of his determinations. The potassium salt was
dried by gentle heating in a stream of dry chlorine.

Upon ignition of the ammonium salt in hydrogen, metallic iridium was left behind in
white coherent lamine. The results obtained were as follows:

Am,IrCl,. Ir, Per cent Ir.
1. 3164 0. 5755 43. 725
1. 7122 . 7490 43. 745
1, 2657 . 5536 43. 739
1. 3676 . 5980 43. 726
2. 6496 1. 1586 43. 739
2. 8576 1, 2489 43. 705
2. 9088 1, 2724 43. 742

mm Mean, 43. 732. 0035
Hence, Ir= 193.402.
The potassium salt was also analyzed by decomposition in hydrogen with special precau-
tions. In the residue the iridium and the potassium chloride were separated after the usual
method, and both were estimated. Eight analyses gave the following weights:

K,IrCl, 4Cl, loss. Fr. KCl.
1. 6316 0. 4779 0. 6507 0. 5030
2, 2544 ~ *, 6600 . 8993 . 6953

2.1290 . 6238 . 8488 . 6560
1. 8632 . 5457 . 7430 . 5745
2. 6898 . 7878 1. 0726 . 8291
2. 3719 . 6952 - 9459 . 7308
2. 6092 . 7641 1. 0406 . 8040
2. 5249 . 7395 1. 0070 7775

 

1 Poggend. Annalen, 13, 435. 1828. 2 Ber. Deutsch. chem. Gesell., 11, 1767. 1878.
No. 3.] ~ /ATOMIC WEIGHTS—CLARKE. 361

Hence, we have the following percentages, reckoned on the original salt:

 

Ir. eKCl. 4Cl.
89. 881 380. 829 29. 290
39. 890 30. 842 29, 277
39. 868 30. 813 29. 300
39. 876 30. 835 29. 289
39. 877 30, 825 29, 287
39. 879 30. 811 29. 310
39. 882 30. 814 29. 285
39. 883 30. 792 29. 288
Mean, 39. 880-L. 0015 Mean, 30. 820+. 0037 “Mean, 29. 291+. 0024
Ir=193. 004 192. 872 193. 298

Joly 1 studied derivatives of iridium trichloride. The salts were dried at 120°, and reduced
in hydrogen. With IrCl,.3KCI.3H,O he found as follows:

Salt. Tr. KCl.
1. 5950 0. 5881 0. 6803
1, 6386 . 6037 - 7000
2. 6276 . 9689 1.1231

These data, if the weight of the salt itself is considered, give discordant results, but the
ratio Ir : 3KCl : : 100 : « is satisfactory. The values of x are as follows:

115. 677
115, 952
115, 915

Mean, 115. 848+. 0583

 

Hence, Ir= 193.277.
The ammonium salt, 1rCl,.3NH,Cl, gave the subjoined data:

Wt. of Salt. Wt. of Ir. Per cent Ir.
1. 5772 0. 6627 42.017
1, 6056 - 6742 41, 990

 

Mean, 42. 003+. 0094

Hence Ir = 193.282.

Analyses of ammonium iridichloride have also been made by Hoyermann * and Holzmann *
in order to determine the atomic weight of iridium. In both investigations the salt was reduced
to metal by heating in hydrogen, and the weights were referred to a vacuum. The data by
Hoyermann are as follows:

Am, IrCl,. Tr. Per cent Ir.
1. 72348 0. 75205 43. 635
‘1. 77984 . 77654 43. 629
. 1.78837 . 78011 43. 621
1.51161 . 50249 43. 634
1. 73794 . 75838 43. 636

ee Mean, 43. 631. 0019
Hence, Ir= 192.610.
Holzmann’s figures are—

Any, lrCl,. Jr. Per cent Ir.
1, 24179 0. 54483 43. 874
1. 09292 - 47809 48. 744
1. 15060 . 50299 43. 715
- 80243 . 35126 43. 747
- 7802) . 34119 43, 731

Mean, 43. 734. 0053
Hence, Ir= 193.418. .

1 Compt. Rend., 110, 1131. 1890. 2 Sitaungsb, phys. med. Soz. Erlangen, 42, 275. 1911. §Tbid., 44, 84. 1013,

 
362 MEMOIRS NATIONAL ACADEMY OF ‘SCIENCES. [Vou. XVI.

t

The three series of percentages for the ammonium salt combine as follows:

DOUD SLE S.kcceeesece esses vow seas ‘ehalepsipargents cere ances 43. 732+. 0035
Hoyermanns.s aspen ce ve ve ie deus ehereiaiseataaececeekinds 48. 631+. 0019
Holemantizseecceersskeeecehc ssi nss s eiv gucenaeedeesee dee 43. 734+. 0053

General mean.........eeeeeeees a avaten ld exert oe 43, 688+. 0016

The following tation are now available from which to compute the atomic weight of iridium:

(1) Per cent Ir in Am, IrCl,, 43.688 + .0035.
(2) Per cent Ir in K,IrCl,, 39.880 + .0015.
(3) Per cent KCl in K,IrCl,, 30.820 + .0037.
(4) Per cent 4Cl in K,IrCl,, 29-291 +.0024.
(5) Per cent Ir in Am,IrCl,, 42.003 + .0094.
(6) Ir: 3KCl :: 100 : 115. 848 +. -0583.

To reduce these ratios we have—

Cl==35. 4605+. 00011 N=14. 0081+. 00008
' K=39. 0975+. 00011 H= 1.0078+. 00001
Hence, :
From ratio— ee :

2b Boca ad Rete ‘oan e cece ee cece ecceeedecdecestccs. Tr=192. 8724. 0440
Deitel cache navs cinta ea mle gatas (On aueckine he Bars 193. 004+. 0087
NS acuta aa iule de leSia tel, Wale Dela ola uses aN he cca ala haan 193. 057 +. 0090
Gicdbrnsuacanioaseneas saan che Guucmeeeeeeaten se 193. 075+. 0970
Qisieiacsibcesicarsiare viscaracie- since wale die leig inte Mibigintwietale eee 193. 282+. 0534
CA ss Sa fok ac talanarah cackatcnshsieesonn aunt ate SRR nciee ers eaters 193. 293+. 0304
General mean. ee eeids . Ir=193. 037+. 0060

In a preliminary note Archibald! states that from ee of the salt K,IrCl, he has obtained
the value Ir=192.90. His complete investigation has not yet been published.

PLATINUM.

The earliest work upon the atomic weight of this metal was done by Berzelius,? who reduced
platinous chloride and found it to contain 73.3 per cent of platinum. Hence Pt=194.69. In
a later investigation,’ he studied potassium chloroplatinate, K,PtCl,. 6.981 parts of this
salt, ignited in hydrogen, lost 2.024 of chlorine. The residue consisted of 2.822 platinum and
2.135 potassium chloride. From these data we may calculate the atomic weight of platinum
in four ways: ;

1. From loss of Cl upon ignition......................2..22.-. Pt=198. 27
2. From weight of Pt in residue. ...........--..0..00.0000 eit  ==197. 42
3. From weight of KCl in residue. :2:........20.2.-2.22.22..05 5 196. 62
4. From ratio between KC] and Pt..............22..-.---.---- “* =197.10

The last of these values is undoubtedly the best, for it is not affected by errors due to the
possible presence of moisture in the salt analyzed.

The work done by Andrews‘ is even less satisfactory than the fascia: partly for the reason
that its full details seem never to have been published. Andrews dried ‘Potassium chloroplati-
nate at 105°, and then decomposed it by means of zinc and water. The excess of zinc having
been dissolved by treatment with acetic and nitric acids, the platinum was collected 1 upon a

filter and weighed, while the chlorine in the filtrate was estimated by Pelouze’s method. Three
determinations gave as follows for the atomic weight of platinum:
197, 86

197. 68
198. 12

so Mean, 197. 887
Unfortunately, hagews anes not state how his calculations were made.

 

 

1 Chent. News, 100, 150. ‘ 1909. Paperread before the British Association for the Advancementof Science. 8 Poggend. Annalen, 18, 468, 1828,
2 Poggend. Annalen 8,177. 1826. *British Assoc. Report, 1852.
No. 3.) a ATOMIC WEIGHTS—CLARKE. 363

In 1881 Seubert! published his determinations, basing them upon very pure :chloroplati-
nates of potassium and ammonium. The ammonium salt, (NH,),PtCl,, was analyzed by. heat-
ing in a stream of hydrogen, expelling that gas by a current of carbon dioxide, and weighing the
residual metal. In three experiments the hydrochloric acid formed during such a reduction
was collected in an absorption apparatus, and estimated by precipitation as silver chloride.
Three series of experiments are given, representing three distinct preparations, as follows:

 

Series I.

Am,PtCl,. Pt. Per cent Pt.
2. 1266 0. 9348 48. 957
1. 7880 . 7858 43, 948
1, 8057 . 7938 43. 960
2. 6876 1.1811 : 43. 946
4, 7674 2. 0959 43. 963
2, 0825 . 8935 43. 961

Mean, 43. 9564. 002
Series II.

AgPtCls Pt. ‘Per cent Pt.
3.0460 1, 3363 43.871
2. 6584 1. 1663 ' 43.876
2, 3334 1. 0238 43, 872
1. 9081 . 8351 43. 881
3. 1476 1, 3810 43. 875
2. 7054 1. 1871 43. 889

Mean, 43. 876+. 001

Another portion of. this preparation, reerystallized from water, of 1.4358 grm. 3 gave, 0.6311
of platinum, or 43.955 per cent. ”

Series III.

Am,PtCl,. = Pt. Per cent Pt,

"2.5274 1.1118 43, 990
3. 2758 1. 4409 , 43. 986
1, 9279 . 8483 44, 001
2.0182 . 8884 44, 020
1. 8873 . 8303 43. 994
2.2270 . 9798 43. 996
2. 4852 1. 0936 44, 004
2.5362 1. 1166 “44, 026
3. 0822 1. 3561 43. 998°

 

Mean, 44. 001+. 003

Tf these series are treated as independent and combined, giving each a weight as indicated
by its probable error, and regarding the single experiment with preparation IT as equal to one
in the first series, we get a mean percentage of 43:907+.0009. On the'other hand, if we regard
the twenty-two experiments as all of equal weight in one series, the mean percentage of platinum
becomes 43.953 +.0078.. Hence Pt=195.15. Upon comparing the work with that done later
by Halberstadt and by Archibald, the latter mean seems the fairer one to adopt.

For the chlorine estimations in the ammonium salt, Seubert gives the subjoined data. I
add in the last column the weight of salt proportional, to 100 parts of silver chloride:

Am,PtCl,. Pt. 6AgCl. Ratio.
2.7054 ° 1. 1871 5. 2226 > §1. 802
2. 2748 - 9958 th 8 4. 3758 51. 986

 

3. 0822 1, 3561 : 5. 9496 51. 805

aul) ; Mean, 51. 864+. 041
Hence Pt =197.20. 8 Ie a
1 Ber. ‘Deutsch. chem. Gesell., 14,-865.
364 MEMOIRS NATIONAL ACADEMY OF SCIENCES. (Vou, XVI.

The potassium salt, K,PtCl,, was also analyzed by ignition in hydrogen, treatment with
water, and weighing both the platinum and the’ potassium chloride. The weights given are as

follows:

K,PtCl,.

5. 0283
7. 0922
3, 5475
3, 2296
3. 5834
4, 4232
4, 0993
4, 4139

Pt.
2.0173
2. 8454
1. 4217
1.2941
1. 4872
1.7746
1.6444
1.7713

2KCl.
1, 5440
2.1793
1. 0890

. 9904
1. 1001
1.3547
1. 2589
1. 3516

Hence we have these percentages, reckoned on the original salt:

Mean, 40. 107+. 005

Hence Pt =194.84.

Pt.
40. 119
40. 120
40. 076
40. 070
40. 107
40. 120
40. 114
40. 130

 

KC.
30. 706
30. 728
30. 698
30. 666
30. 700
30. 627
30. 710
30. 621

 

Mean, 30. 682+-. 009

Hence Pt==195.05.

As with the ammonium salt, three experiments were made upon the potassium compound
to determine the amount of chlorine (four atoms in this case) lost upon ignition in hydrogen.
In the fourth column I add the amount of Ble corresponding to 100 parts of AgCl:

K,PtCl,.
6.7771
3.5834.
4.4139

Hence Pt = 195.31.

Pt. ' 4AgCl.
2.7158 7. 9725
1. 4372 4. 2270
1. 7718 5. 2144

Ratio,
85. 006
‘84. 774
' 84. 648

Mean, 84. 809-4. 071

Halberstadt,' like Seubert, studied the chloroplatinates of potassium and ammonium,
and also the corresponding double bromides and platinic bromide as well. ‘The metal was esti-
mated partly by reduction in hydrogen, as usual, and partly by eee Platinic bromide

gave the following results:

PtBr,.
0. 6396
1. 7596

. 9178
1. 1594
1. 9608
2. 0865
4. 0796
6. 8673

ve
1. 2588
'~ 1, 4937

Hence Pt = 194.67 :

I. By reduction in H.

Pt.
0. 2422
. 6659
. 8476
. 4388
. 7420
. 7898
1. 5422
2. 5985

I, By electrolysis.

Pt.
0. 4763
- 5649

Per cent Pt.
37. 867
37, 844
37. 873

37. 847
87, 842
' 87. 853
37. 852
37, 839

Per cent Pt.

37.837 _.
37. 819

Mean of ail ten determinations, 87. 847-4. 0033

 

1 Ber Deutsch, chem, Gesell., 17, 2962. 1884.
365

The ammonium platinbromide, (NH,),PtBr,, was prepared in two ways, and five distinct
lots were studied.. With this salt, as well as with those which follow, the data are given in
distinct series, with from one to several experiments in each group, but for present purposes it
seems best to consolidate the materialand so put itin more manageable form. ‘The percentages
of platinum and weights found are as follows:

I. By reduction in H.

No. 3.3 ATOMIC WEIGHTS—CLARKE.

Per cent Pt.

 

Am,PtBr,. Pt.
0. 6272 0. 1719 27. 408
1, 0438 - 2865 27. 447
1.1724 . 3215 27. 422
1. 4862 . 4076 27, 426
1.0811 - 2966 27. 435
1, 3383 - 3672 27. 437
1. 0096 - 2769 27. 426
1. 1935 . 3269 27. 390
1. 3182 . 3611 27. 393
2. 2476 . 6159 27. 402
1, 3358 . 3668 27. 451
1. 7859 . 4899 27. 431
4.1641 1.1427 27, 441
1. 1835 . 3250 27. 460
2. 4003 - 6591 27. 459
2.5293 - 6940 27. 438
1. 7147 - 4705 27. 439
2. 3014 . 6316 27, 444
3. 0052 - 8245, 27. 435
4, 8592 1. 3329 27. 430
1. 5337 - 4210 27. 449
f 0373 . 5594 27. 457
2. 0939 - 9751 27. 465
IT, By electrolysis.
Aim,PtBre. Pt. Per cent Pt.

1. 5586 0. 4272 27. 409
f 6052 - 4397 27. 392
; 3. 1229 . 8569 27, 439
1. 1612 . 3180 27. 386
2. 5817 - 7081 27. 427
A 0231 . 2809 27. 456
1. 6744 . 4691 27, 418
1. 6744 4591 27, 418
1. 6052 . 4397 27. 392

Hence Pt= 194. 88.

 

Mean of all thirty-two determinations, 27. 429+. 0027

With potassium platinbromide Halberstadt found as follows:

Pt.

I. By reduction in H.

K,PtBrg. 2KBr. Percent Pt. Per cent KBr.
2. 5549 0. 6630 0. 8071 25, 940 31. 590
2. 6323 . 6831 . 8318 25. 947 31. 599
2. 9815 . 7598 -, 9259 25. 910 31. 584
8. 4463 . 8939 1. 0895 25. 938 31. 613
4. 0081 1. 0404 1. 2653 25. 957 31. 568
3. 9554 1. 0266 ‘L 2495 25. 954 31. 589
2. 0794 . 5388 . 6558 25.911 31. 538

{2 . 5635 . 6849 25. 926 ‘31. 511
2. 3099 ', 5986 "7297 25. 914 81. 590
1. 4085 . 3645 4446 25. 880 31. 565

2 6166 6772 . 8279 25. 881 31. 640
2. 6729 . 6923 . 8469 25. 900 31. 684
366

K,PtBr,.

; i. 2110 *
(3. 1642
{ 9080
1. 6754
ir 3148
1. 5543

Pt.
0. 5726
- 8188
. 4947
- 4341
. 3403
- 4025

Mean of eighteen determinatione, 25. 915+. 0040

Hence Pt =195.09 and 195.78.
For ammonium platinchloride Halberstadt gives the following data:

Am2PtCl,.

1. 0604
1. 3846
1. 5065
2, 3266
1. 3808
1. 7396
2. 7420
[2
5. 4644
3, 4859

Am,PtCl..

1. 1069
1.5101
5845
f 6035
1.9271
{t 1046
1. 4179

f 9474

Hence Pt = 195.07.

II. By electrolysis.

‘@KBr.
0. 6997
. 9983
. 6025
. 6286
. 4160
. 4911

Pt.
0. 4662
. 6087
. 6617
1. 0227
. 6059
7638
1. 2068
1. 4019
2, 4035
1, 5321

IT. By electrolysis.

Pt.

0. 4161
. 4865
. 6634
. 2347
. 7044
. 8459
. 4858
. 6233

. Per cent Pt.

25. 898
25. 877

2 25.927 6

25. 915
25. 882
25. 895

I. By reduction in H,

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

Per cent KBr.
31. 647
31. 550
31.577
31. 550
31. 640
31. 596

81. 591+. 0068

Per cent Pt.

43. 964
43. 962
43, 923
43. 956
43. 880
43. 906
44,011
43. 971
43. 984
43. 951

Per cent Pt.

43. 920
43. 951
43. 930
43, 910
43, 928
43. 894
43.979
43. 959.

Mean of eighteen determinations, 43, 943+. 0054

For potassium platinchloride Halberstadt’s data are—

K,PtCls.
1. 6407
1. 9352
1. 5793
1. 6446
1, 0225

le, 4046
5. 83844
7. 1732

E,PtCl,.

ft 2354
2, B754
1. 0933
1. 3560
17345
2, 0054
2.0666

Pt.

~ 0. 6574

7757
- 6334
- 6595

4102.

. 9641

*2. 3412

2 8776

Pt.
0. 4958
L 0318

«4387

. 6956
. 8038
. 8291

2KCl.
0. 5029
. 6921
. 4836
- 5049
- 8133
. 7388
1. 7905
2. 1998

I, By dectrolysie

‘2KCl.

0. 3792
- 7898

. 8855
. 4167
- 5298
- 6147
- 6356

I. By reduction in H.
Per cent Pt.

~* 40. 069
40. 084
40. 106
40. 101
40. 117
40. 094
40. 127
40. 116

Per cent Pt.
40. 092
40, 063
40,126
40, 103

40.104
40. 081
40. 117

‘

Per cent KCI.
30. 651
30. 600
30. 621
30. 700
80. 640
30. 724
30. 688
80. 666

Per cent KCl.
30. 695
30. 667
30. 668
30. 730
(30, 545
.80. 652
80. 755
No, 3.] ATOMIC WEIGHTS—CLARKE. | i 367

| KPtCl,. Pt. : KCI. Per cent Pt... Per cent KCl.
1. 2759 . 5118 . 3808 40.112 © 30. 629
1. 9376 : . 7763 . 5927 .. 40.065 - 30. 589
12, 3972 1 - 9608 _ . 7355 40. 080 30. 681..
2. 7249 — 10929 8. . 8364 40. 108 30. 691

seen of nineteen determinations, 40. 098. 098 +. age! 30. 663+. 0080

ne ‘- " Seubert found, ‘80. 682+. 0090

‘General mean, ~ 30. 6714. 0060
Hence Pt=194.77 and 195.35 from Halberstadt’s data alone. “ un ee
The work of Dittmar and M’Arthur ' on the atomic weight of platinum is difficult t to is
and essentially unsatisfactory. They investigated potassium /platinchloride and came to the
conclusion that it contains traces of hydroxyl replacing chlorine and also hydrogen replacing
potassium. It is also liable, they think, to carry small quantities of potassium chloride. In
their determinations, which involve corrections indicated by the foregoing considerations,
they are not sufficiently explicit, and give none of their actual weighings. They attempt,
however, to fix the ratio 2KCl : Pt, and after a number of discordant, generally high results,
they give the following data for the atomic weight of platinum based upon the assumption that

2KCl= 149.182:
195. 54 195. 60

195. 48 ; 195. 37
_ Mean, 195. 50+. 0330

This ratio can ike be computed from Seubert’s and Halberstadt’s analyses, and also the
ratio 2KBr : Pt. It has not seemed HeCeeary to do so, in view of the overwhelming weight
of Archibald’s more recent work. ;

Dittmar and M’Arthur also discuss Sauherita dotarniinations; seeking to show that the
latter also, properly treated, lead to a value nearer to 195.5 than to 195. Seubert at once
replied to them,’ pointing out that the concordance between his determinations by very different
methods (a consordance verified by Halberstadt’s investigation) precluded the existence of
errors due to impurities such as Dittmar and M’Arthur assumed.

The recent determinations by Archibald® of the atomic weight of platinum were based.
upon analyses of the platinchlorides and platinbromides of potassium and ammonium. In
these analyses every precaution was taken which modern experience had shown to be necessary.
The possible presence of moisture in the several salts was carefully considered, and the potassium
compounds in particular were dried at 380° to 400°. For the elaborate details of manipulation
the original memoir must be consulted.

First, as to the analyses of potassium platinchloride. The salt, after thorough drying and
weighing, was reduced by heating in a stream of pure hydrogen. The hydrochloric acid so
formed was absorbed in water, and afterwards converted into silver chloride and weighed.
Known quantities of silver were used in this operation, so that two distinct ratios were deter-
mined. From the residual mixture of potassium chloride and platinum the chloride was
washed out, and its chlorine content was estimated as in the previous determinations. The
metallic platinum, converted into sponge by again heating in hydrogen, was also weighed.
Vacuum weights are given in all of Archibald’s determinations. The weights are as follows:

 

E,PtCl.' Pt. - 4AgCl. 2AgCl. 4Ag. 2Ag.
1. 43605: "- 0. 57667 1. 69324 0. 84690 1, 27475 0. 63722
1.69914 °° - 68226 2. 00402 1, 00172 1. 50834 . » . 75401
2.11830" / ~~ 585062 2. 49836 1, 24894 1. 88046 . 93993
2. as vo 1. 00287 2. mente 1, 47249 2. 21626 1. 10841
ahi eed SBO0IZ sewed 126271, ne eee 95080
ot 2. 20619 - 88588 2. 60135 : AL 30106 - 1. 95842 - 97909
“: 70600 Ue . 68486 2. 01201 Seas 1. 00580 ose eee Dae
1.74397 0° . 70018 "2.05691 ° °° - 1502820 1.54816 - . 77402 ° :
2. 06137 . 82789 2. 43096 1. 21526 1. 82982 2 i 91481
2. 34095 . 93991 2. 76105- 1, 38034 2. 07759 a 1 03868
1 Trans. Roy. Soc. Edinburgh, 33, 561. 1887 3 Proc. Roy. Soc. Edinburgh, 29,721. 1909.

3 Ber. Deutsch. chem. Gesell., 21, 2179. 1888.
161566°—20-—24 gik®
368

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

E,PtCl,. Pt. 4$AgCl.
1. 54787 0. 62150 1, 82560
1. 95944 . 78694 2. 31070
2, 28366 . 91697 2. 69304
2, 27441 . 91320 2, 68244

2AgCl.
0. 91266
1. 15522
1. 34636
1, 34093

4Ag. ©

1.37391
1, 73902
2. 02640
2. 01870

From these weights Archibald computes nine ratios as follows.
recalculated the figures into the percentage form used for previous investigations.
ratios are as Archibald gives them; but with the probable errors computed by myself:

 

 

 

 

 

 

2Ag.
0, 68702

. 86967
1. 01338
1, 00924

{Vou. XVI.

In the first ratio I have

 

 

Per cent Pt. 4AgCl : Pt. 2AgCl: Pt. 4AgCl: K,PtCh. 2AgCl.K,PtCl.

40. 157 34, 057 68. 092 84. 811 169. 57

40, 153, 34, 045 68. 109 84. 787 169. 62

40. 156 34, 047 68. 107 84. 788 169. 61

40. 158 34. 058 68. 107 84. 810 169. 60

Scammer tan eee 68. 117 ee 169. 61

40. 154 34, 055 68. 089 84, 810 169. 57

40, 144 34. 039 68. 091 84. 791 169. 62

40. 149 34. 040 68. 098 84. 786 169. 61

40. 162 34. 056 68. 125 84. 797 169. 62

40. 151 34. 042 68. 093 84. 785 169. 59

40. 152 34, 044 68. 098 84. 787 169. 60

40. 161 34. 056 68. 120 84. 799 169. 62

40. 154 34. 050 68. 107 84. 799 169. 62

40. 151 34, 045 68. 102 84. 789 169. 62
~ Mean, 40. 154 34. 049 68. 104 84. 795 169. 606
+. 0009 +. 0013 +. 0020 +. 0019 +. 0034

Pt=195, 22 195. 22 195. 27 195. 23 195. 20
4Ag ; Pt. 2Ag : Pt. 4Ag : K,PtCh,. 2Ag : K,PtCl,.
45, 238 90. 498 112. 65 225. 36
‘45, 233 90. 484 112. 65 225. 35
45, 235 90. 498 112. 65 225. 37
45, 251 90. 478 112. 68 225. 31
hie Shier 90. 510 sad awd 225. 36
45. 234 90. 480 112. 65 225. 33
45. 227 90. 460 112. 65 225. 31
45. 244 90. 499 112. 65 225. 33
45. 240 90. 488 112. 68 225. 37
45. 236 90. 463 112. 66 225. 30
45, 252 90. 487 112. 68 225. 31
45, 251 90. 486 112. 70 225. 35
45. 237 90. 484 112. 67 225. 36
Mean, 45. 240 _ 90. 486 112. 66 225. 34
+. 0016 st. 0026 +. 0030 +. 0047

=195. 22 195. 23 195. 21 195, 24

The other

For the first of these ratios, the percentage of Pt in K,PtCl,, there are ao determina-
tions. The three series combine thus:

Seubert s isis itesedie sie vessccete oS ied oa ediciekedweeuleniwe 40. 107

Hal berstad tics soscis cc ciaviessecrssacndeumeeereeiee ses 40. 098

Archibald...........-.-.---- piruiaecekadiche Evueseen age 40. 154
General mean.... 22.2.2... cece cece cence ce ceees 40. 1484

+ .0050
+ .0031
+ .0009

+ .00085

Similarly, the ratio 4AgCl : K,PtCl,, as determined by Seubert, may be combined with
Archibald’s series. Better still the two series may be reduced to uniform type with Archibald’s
ratio for 2AgCl, and given in the form Ag(l : K,PtCl, :

bine as follows:

:100:2. The three series then com-

Mati iert; TARO 16s 5cuiantacatndsatedeGesdavacoesasuaks 339, 236 + .2840

Archibald, 4AgCl.........0-.0c00ceceeeeececeeeeeesees 339. 180 + .0076

Archibald, 2AgCl....... seen Gu Nip onsldeoconanetes 339. 212 + .0068
General mean. ssieis)s:cjsis:e:2 sions iow sciesinie'e'e'e e's slsieieig's 339, 204 + .0051
Nos 81 «. >, ATOMIC WEIGHTS—CLARKE. 369

Archibald’s two series of measurements ab the ratios between silver and the platinchloride
can also be reduced to the form a K,PtCl, : : 100 : z, and combined:

4Ag series... ...2.2 00. ewaws fewen awe etierarnasinaende Reel! 450. 64 + .0120
DAG SETI CS oes seen pitta Slates icc fal des bo5 AWS isiat Eheiais OE sc 450. 66 + .0094
« General meat! cscccceve nsec Besos eeeeceavsenss 450. 654 + .0074

Archibald’s data for ammonium platinchloride are rather simpler than with the potassium
salt, since the total chlorine was determined at once, instead of in two portions. His weights
are subjoined:

Am,PtClg. Pt. 6AgCl.., ., 6Ag.
1. 75088 0..76976 _» 8.39181, 2. 56181
1. 36500 -59997 ° 2. 64317 1, 99014
1. 15060 - 50585 2. 22810 1. 67695
1. 27475 . 56049 2. 46936 1. 85794
2. 54096 1. 11688 4. 92047 3. 70426
The derived ratios are as follows: a: be
Per cent Pt. 6AgCI:Pt. - FAgCl: :Am,PtCl, .6Ag:Pt. 6Ag:Am,PtCl,.
43. 964 22.695 51. 621 _ 80. 165 68. 613
43. 954 22. 699 51. 643 30. 147 68. 588
43, 964 22. 703 51. 640 30. 165 68. 613
43. 969 22. 698 51. 623 "* 80. 167 68. 611
48, 955 22. 699 51. 641 30. 152 68. 597 ©
Mean, 43. 961 22. 699 : BL 684 , 80.159 159 68. 604
+ .0061 + .0031 + .0032 & .0027 "+ 0084
Pt=195. 21 195. 22 195. 23 195. 21 195. 23,

One of the ratios can be combined with Seubert’s determinations, as follows:

Ratio 6AgCt: Am,PtCl,: : > 100: &.

 

Seubert..... 2.0.2.2... icv sierdwtineutiboiantadaticeten. DLIB6L<e. 0410
PETA nics uated epeae opammpadiemdge ta gpinns gil Wad 0082
General mean........-.2220-0eeeeeee ee eee -++--+ 51,636 +. 0032
For ammonium platinbromide Archibald gives these data:
Am,PtBre. Pt. 6AgBr. 6Ag.'
1.83860 = * 0. 50497 2. 91430 1. 67448
2. 31057 . 63437 3. 66269 2. 10879
2. 33965 . 64272 3. 70900 2. 13049

Hence the following ratios:
Per cent Pt. 6AgBr : Pt. 6AgBr: Am,PtBr. 6Ag: Pt. 6Ag : Am_PtBre.

27.4650 17. 327 63. 089 30. 157 109. 801
27. 455 17.320 63. 084 30, 154 109. 829
27. 471 17.329 63. 080 30. 168 109. 827
Mean, 27.464 — 7 “17.325 63. 084 30. 160 109. 816

+..0032 "2 +#.0018 +. 0018 +. 0030 +. 0055

Pt=195.23 | 195.22 195. 22 195, 22 195, 22
The percentage of platinum in Am,PtBr, combines with Halberstadt’s figures thus:

Eis i otal ase shicehtcarngon seca tameniocenasaed wire Leetet 27.4294. 0027
Archibald... . crs geeciatd cil clavar avidiescioa ier sierra nec wecerecese 27,4644. 0032
General mean.....-..-..-.----+ 22+ eee e eee eee eee 27, 443+. 0021

The analyses of potassium platinbromide were like those of the chloride, the bromine baing
estimated in two pene 2Br and 4Br. The weights are these: 6 td

K,PtBre 9) Pt. 4AgBr. 2AgBr, . $Ag, 2Ag.

2. 19076 <0. 56779 2. 18543 _ 1.09273 | 125544. . 62770
2. 42094 . 62766 2, 41510 1.20758 1, 38761 . 69378
1. 78705 - 46344 1, 78284 .89156 1. 02416 - 51214
1. 81840 47156 | 1. 81430 - 90703 1, 04228 - 52105 .
2. 47056 -y -+,64063 2. 46507 1. 23246 1. 41572 . 70800

2.19017, . 56787 , © 2. 18525 1. 09260 1. 25530 . 62756
370 MEMOIRS NATIONAL ACADEMY OF SCIENCES. |Vou. XVI.

From these weights nine ratios are deducible, as in the case of the platinchloride, as follows:
Percent Pt. 4AgBr: Pt.  24gBr: Pt. 4AgBr: K,PtBrg. 2AgBr : K,PtBry.

 

 

 

 

 

 

 

 

25. 918 25. 981 51. 961 100. 244 200. 485
25. 926 25. 989 51. 976 100. 242 200. 479
25. 933 - 25, 995 61. 981 100. 236 200.441
25. 933 25. 991 51. 990 100. 226 200.478
25. 9381 25. 988 51. 980 100. 223 200. 458
25, 928 25. 990 51.'974 100. 225 200. 455
Mean, 25, 928 25. 989 51. 977 100, 233 200. 466
+. 0015 +. 0013 +. 0027 +b. 0025 +. 0048
Pt=195. 23 195. 23 195. 23 195, 23 195. 23
4Ag: Pt. Ag ; Pt. 4Ag : K,PtBrg. 2Ag : K,PtBrg.
45. 226 90. 456 174. 50 349. 01
45. 233 90. 470 174. 47 348. 95
45. 251 90. 491 174, 49 348. 94
45. 243 90. 502 174. 46 348. 99
45. 251 90. 485 174, 51 348. 95
45, 238 90. 489 174.47 349. 00
Mean, 45. 240 90. 482 174, 48 348. 97
+. 0028 tk. 0045 +. 0051 x. 0085
Pt=195, 22 195. 23 195. 20 195. 22

The percentage of platinum in the platinbromide combmes with Halberstadt’s figures as
follows: ;

Halberstadt....-.. pea ests ee h ye cease _ 25. 915+. 0040
Archibald.......---.0s00++ pre Loe ligt) e 25. 928-. 0015
General mean........... eho atapmini pias oacerniainiiiotntacalassiels 25. 927+. 0014

Several other ratios, given in divers forms by Archibald, are also capable of consolidation.
The ratio between silver bromide and potassium platinbromide, reduced to uniform type, that is,
to AgBr : K,PtBr, : : 100 : z, becomes—

2AgBr series........... Reena hese cecceeeseddececeseses 400, 932-4. 0096
AASBY gOries Acvieg ess 25 seiwinaspeaaacta seek cone lees 400. 932-4. 0100
. General mean... - i... 0... e eee eee eee eee isieinisiookre 400. 932+. 0069
For the ratio Ag : K,PtBr, : : 100 : we have—
DAG SONOS. oes saedda dee conan noas tae sate eeokenseeezs 697. 940-. 0170
4Ag series..-....--. seGSwe ase sods anarde tanya Menbmneamek 697. 920+. 0204
General mean. os ¢s ssa ssaveceperee ster ecascwene res 697. 9864. 0181
For the ratio AgBr : Pt : : 100 : s—
2AgBr series.....-.... puaduandasnes) Pi aceu teas tpe het! 103. 954+. 0054
4AgBr series.....-......02 cece cee cece ences BL Sia ayee ee 108. 956+. 0052
Genoral moan, - -... 46 snctisnisecewecteetiesesnce see 103. 955+. 0037
For the ratio AgCl : Pt : : 100 : 2—
2AgCl series with K,PtCl,....-.-.---20eeeeeeeceeeeeeeees 136, 208+. 0040
4ApCl series with Ky PtCle......--...sscecaeeeereeeeeeces 136. 196+. 0052
GAgCl series with Am,PtCly. ........- 20-2 2eceeeeecee eens 136. 194-4. 0186
General mean..........+2+------+- seseeeeeeeeeres 186.2084. 0081
For the ratio Ag : Pt : : 100 : 2—
2Ag series with KzPtClg......5ee.eeeeeeeeeee BAIS Sas ete 180. 9724. 0052
4Ag series with K,PtCl,....-2.+.-----2e-eeee eee e eee 180. 960-+. 0064
6Ag series with Am,PtCly.........20.-20ee eee cece eee eeee 180. 954-4. 0162
6Ag series with Am,PtBry..........--22---.ee eee serene 180. 960+. 0180
2Ag series with K,PtBrg. ....---- 2222-22 eeeeeeeeeeee eee 180. 964+. 0090
4Ag series with K,PtByr....... sya eae aaGeanle shinee. aie'esy 180. 960--. 0112

General mean..........-.---e cece eee e eee sjelassie esses 180, 965+. 0034
No, 8] ATOMIC WEIGHTS—CLARKE. 871

From the last two ratios the cross ratio Ag : Cl : : 100 : 32.864+.0039 is deducible, which
agrees Closely with the measurements by Richards and Wells. From the corresponding ratios
Ag : Pt and AgBr : Pt, we have the ratio Ag : Br : : 100 : 74.080+.0070. These agreements
with the best determinations of the silver-halogen ratios are good evidence in favor of Archibald’s
work.

One more series of analyses of ammonium platinchloride, by Schultz,’ remains to be men-
tioned. The data, with vacuum weights, are as follows:

Am,PtCl, Pt. Per cent Pt.
3. 53199 : 1. 54910 43. 859
2. 31493 1, 01537 43, 862
3. 84762 1, 68867 43. 888
1. 52569 . 66953 43, 883
2. 20900 . 96880 43. 857

Mean, 43.870 +. 0044:
Hence, Pt = 194.49.

This percentage combines with previous determinations thus:

 

Beubarinj casey dentonscenammieneond ie i aeenneeee 43.953 +.0078
Halberstadt..............- Pacman a seis eeceeceseerees 43,943 4.0054
Archibald soca csisers sas rrsindete SkGuies Lev asctsinges eas ..-+ 43,961 +.0061
Schultz: ssjeccssiwaies sess etacaades icine Ss iivsas --- 48.870 +.0044

General mean......2;.cseeeeeesceeeeesengeersngenes, 48, 918.0028

Rejecting the work of Berzelius and Andrews, the following ratios are now available from
which to compute the atomic weight of platinum:

(1) Am,PtCl, : Pt :: 100 : 43.9185 + .0028.
(2) 6 Ag : Am,PtCl, :: 100 : 68.604 + .0034.
(3) 6AgCl : Am,PtCl, :: 100 : 51.636 + .0032.
(4) K,PtCl, : Pt :: 100 : 40.14844.00085.
(5) Ag : K,PtCl, :: 100 : 450.654 +.0074.
(6) AgCl : K,PtCl, :: 100 : 339.204+ .0051.
(7) Am,PtBr, : Pt :: 100 : 27.443 4.0021.
(8) 6Ag : Am,PtBr :: 100 : 109.816 + .0055.
(9) 6AgBr : Am,PtBr, :: 100 : 63.084 4.0018.
(10) K,PtBr, : Pt :: 100 : 25.927 4.0014.
(11) Ag : K,PtBr, :: 100 : 697.936 +.0131.
(12) AgBr : K,PtBr, :: 100 ; 400.932 +.0069.
(13) PtBr, : Pt :: 100 : 37.8474.0033..
(14) K,PtCl, : 2KCl :: 100 : 30.671 +.0060.
(15) K,PtBr, : 2KBr :: 100 : 31.591 + .0068.
(16) 2KCl : Pt :: 149.182 : 195.50 + .0330.
(17) Ag : Pt :: 100 : 180.965 + .0034.
(18) AgCl ; Pt :: 100 : 136.203 + .0031.
(19) AgBr : Pt :: 100 : 103.955 + .0037.

The antecedent atomic weights are—

Ag==107. 880 +.00016 K=39, 0975+. 00011
Cl = 35.4605+.00011 N=14. 0081 4.00008
Br= 79, 9228+.00015 H= 1.0078+.00001
Hence,
From ratio— :
DB aed iece aa il Ws alrnmen cease i Natt aesl aA enema Pt=194. 670+.0199
Vo ccasabatasntens caatseorcnins jeaes ateeeceeres 194.878 4.0153
7 232d aiad haa besmmoaeen soe meer Giaominndes 195.022.0160
Aandi veda dawie sam aeebaele datedacmeanaeebietudate ts 195. 179-+. 0050

 

1Inaug. Dissertation, Erlangen, 1912.
372 MEMOIRS NATIONAL ACADEMY OF SCIENCES.

From ratio— 6° OOD: PO QA dtut aeore ef: ~-od deus
5 cee aasertere oc eoerinaceicin eon: eos iricenr pent Dl ESTES ate
GW acter ccctenniniot Wis bicse aeiion seh sade 195. 2014.0145,
Fh sncemnaeacgioneminen nin =i alenimnnappniinh at ainda a BC: 0081

2 . LE E595, 210-4. 0221

195. 217+.0112
1 195. 218-4 0203 °
195, 2220710.
195. 229+. 0037
195. 230 +. 0130
195. 230-4..0070
195. 2344. 0045
195. 2424.0276
195. 259-4. 0074
195, 415+. 0330
195.777 4. 1222

General mean.....,....--.-----2--eeeee eee Pt=195. 213-4. 0020

fy: bl ad: NORGE ii * = 8 Beton

SCANDIUM. hiadiuer,

 

 

+

{Vou. XVI.

x

‘Cleve,! who was the first to make serious determinations of the atomic weight of this element,
obtained the following data: 1.451 grm. of sulphate, ignited, gave 0.5293 grm. of Sc,0,. 0.4479
grm. of Sc,O,, converted into sulphate, yielded 1.2255 grm. of the latter, which, upon ignition,

gave 0. 4479 grm. of Se,0,. * Hence, for the percentage of Sc,O, in Sc,(SO,), we have:

 

 

_ 36.478 4
86.556 6%:
veg, 36-556
Mean, 36. ve 0175.
Hence, Sc= 45.12. a
Later results are those of Nilson,? stp maiverbad srandina oxide into the sulphate. I
give in a third column the percentage of oxide i in sulphate: ); pr) a “y
Sc,03. Seq(SO,)o- : ’ Per cent.
0.3379 x: 0. 9348 ». «f 36.166 .
3015 | 8330, (86. 194
- 2998 8257 * 36.187 °
. 8192 . 8823" - 36.178
we HQ. de Ve Pe Hg
ee , Mean, 36. 181, 004 np

Hence, Sc = 44.09. (51S, cf G6

In 1910 Meyer and Winter * published a paper on the atomic weight ‘of scandium. By
the sulphate method they obtained values for Sc ranging from 44.87 to 45.37, a variation of
half a unit. From analyses of the oxalate, Winter found Sc= 44.86 ‘to 45.33. One analysis
of the acetylacetonate gave Sc=45.07. These values are all high, and it is probable that the
material analysed contained traces of thorium. In a later research by Meyer and Golden-
berg,‘ in which thorium was removed, four analyses of the sulphate: gave the dolloseiing results:

8 Pee

 

 

Se(SO,)3 80.0; g 00 Per cent Se,0;"
1. 5924 0.5766 ‘ - 36. 210

, 2.4207 . 8765 jigiod 36.209 “Sp
2. 6394 . 9565 _ 86.239
1.5207 - . 5495, O82. 6 135

a aw it + GOOp ¢
AR Ae a go wwe Mean, 36. 198+-. 0150

Hence, Sc = 44.14. atte
1 Compt, Rend, 89,41. fo, I et

2 Compt. Rend., 91, 118.

3 Zeitsch. anorg. Chem., 67, 398. For tlie détatled antl yses see the Tinney dissertation Wate Berlin, 1911.

«Chem. Apes 106, a 1912, nie ese eh
done Ma ced

Be err ene
No. 81 ATOMIC WEIGHTS—CLARKE. 373

There are also two analyses of scandium sulphate by Lukens,’ as follows:

Sc,(SO,)s. Sc,05.
0. 30636 0. 11134
1. 14140 . 41553

Hence, Sc = 44.66.

Per cent Sc,03.
36. 343
36. 406

Mean, 36. 374+. 0210

Disregarding the work of Meyer and Winter, which is supplanted by that of Meyer and
Goldenberg, the four remaining series of analyses combine thus:

CIS Y Careers xericteyrtcen seer oe 4 siaidinie. ae Sin isiassvesis eadavereenn nails bievm 36. 530+. 0175
INDISOD cc cat nayantees casks penne oe eee hiickte leis ocoeeaeewen 36. 181+. 0040
Meyer and Goldenberg................2.0200eceeeeeeeeeees 36. 198+. 0150
STUDI A aay oregon scat epee cree ohare ada ee betel 36. 374+. 0210

General meanivsencwsrctcredseneeoceuweesie ence ee eaes 36. 288+. 0020

Hence, if S=32.065 +.00012, Sc =44.403+.0044. This result is by no means satisfactory.

ADDENDA.

Since the manuscript of this volume was sent to the printer, important determinations of

the atomic weight of scandium have been received. Héni
of scandium bromide, has obtained the following results.
are treated here as one.

ScBrs. Ag.

2.79839 38. 17972
3. 32001 3. 77235
3. 34311 3. 79829
3. 22367 3. 66250
3. 11816 3. 54255
2. 33284 2. 63915
3. 13467 2. 42519
1. 96690 2. 23460
3. 13887 3. 55645
2. 92675 3. 32529
3. 14889 3. 57784
2. 79650 3.17719
2. 80219 3. 18371
2. 62351 2. 98093
2. 59879 2. 95270
2. 60299 2. 95768
2. 37863 2. 70257
1. 41354 1. 60630

gschmid,? from very exact analyses
Two series, with vacuum weights,

Aé.wt.Se.
45. 079
45. 084
45. 108
45. 114
45. 120
45. 103
45. 122
45.120
45. 091
45. 103
45. 090
45.113
45. 108
45. 087
45. 100
45. 080
45.099
45. 054

Mean, 45. 099

 

The material from which Hénigschmid prepared his bromide had been purified by methods
described by Meyer and Schweig,? who also determimed the atomic weight of scandium by the
usual sulphate method. The value found ranged from 45.03 to 45.37; in mean, Sc =45.23.
They show, however, that the sulphate method is untrustworthy, and accent Hénioschmid’s

determination, rounded off to 45.1. This value should repl
the earliest of all, that by Cleve.

1 Journ. Amer. Chem. Soc., 35, 1470. 1913. 2 Zeitschr. Electrochem., 25, 93. 1919.

ace the older one with

4 Zeitschr. anorg. allg. Chemie, 108, 303. 1919.
874 MEMOIRS NATIONAL ACADEMY OF SCIENCES. {Vou XVI.
YTTRIUM.

Nearly all the regular determinations of the atomic weight of yttrium depend upon analyses
or syntheses of the sulphate. A series of analyses of the oxalate, however, by Berlin,’ is some-
times cited, and the data are as follows. In three experiments upon the salt Yt,(C, Ode 3H,O
the subjoined percentages of oxide were found: 2
Lite ig ead Rs : ee : 45. 70

: : 45. 65
45. 72

Mean, 45. 69+. 0141
Hence Yt=89.60.

The early work of Berzelius ? may beignored. The determinetions by Popp,’ Delafontaine,
and Bahr and Bunsen ° are also of no present value. They are sufficiently summarized in the
former editions of this work. , ee es

The first determinations of the atomic weight of yttrium which have any real significance
are those by Cleve and Hoeglund,® who adopted the method of converting the oxide into the
anhydrous Seen Their data are as follows: |

| F405. a ” ¥t(80,)s. ~~ Per cent Yt,05.
1. 4060 "9.8995 48, 608
1.0930 2.9515 48, B45
1. 4540 2. 9895 48. 637
1. 3285 2.7320 48, 627
2. 3500. 4, 8330 . 48, 624

2. 5780 5. 3055 -» 48, 691

Mean, 48. 605+. 0096
Hence Yt = 89.58.

In a later paper Cleve? gives syntheses, of yttrium sulphate made with yttria which was
carefully freed from terbia, . The weights and percentages are as follows:

Yt05. Yt,(SO,)3. Per cent Yt,03.
. 0. 8786 ° 1. 8113 48. 507
. 8363 1, 7234 48. 526
. 8906 1. 8364 48, 497
. 7102 1. 4645 48, 494
. 7372 1. 5194 48. 519
. 9724 2. 0047 48. 506
. 9308 1. 9197 47. 487
. 8841 1. 7204 ‘48, 483
1, 0224 © 2: 1073 ' 48, 517
. 9384 1, 9341 48, 519
9744 2. 0093 48, 494

1. 5314 3. 1586 48, 484
. Mean, 48, 503+. 0029

 

Hence Yt =89.12.

The yttria studied by Jones * had been purified by Rowland’s method—that is, by precipita-
tion with potassium ferrocyanide—and certainly contained less than one-half of one per cent of
other rare earths as ‘possible impurities. Two series of determinations were made—one by
ignition of the sulphate, the other by its aes The results were as follows, with the usual
percentage column, added:

1 Forhandlingar ved de Skandinaviske Naturforskeres, 8, 452. 1880.

2 Lehrbuch, 5 Aufl. 2, 1225,

8 Ann. Chem, Pharm., 131,179. 1864.

4 Ann. Chem. Pharm., 134, 108. 1865. Second paper in Arch. Sci. Phys. Nat. (2), 25,119. 1866.
6 Ann. Chem. Pharm., 137, 21. 1868.

6K. Svenska Vet. Akad. Handlingar, Bd.1, No. 8. 1873.

7K. Svenska Vet. Akad. Handlingar, No. 9, 1882. See also Bull. Soc. Chim., 39,120. 1883

8 Amer. Chem. Journ., 17,154. 1895.

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 875

First series. ‘Syntheses.

¥t,05. Yt,(S0,)s. Per cent Yt,03.

0. 2415 0. 4984 48, 455
4112 . 8485 48. 462
. 2238 .4617 48, 473
. 3334 . 6879 48. 466
. 3408 . 7033 48. 457
. 3418 . 7049 48, 489
. 2810 . 5798 48, 465
. 3781 . 7803 48, 456
. 4379 . 9032 48, 483
. 4798 . 9901 48, 460

 

Mean, 48. 467+. 0025

Second series. Analyses.

 

Yt(SO,)3. Yt,03. Per cent Yt,03.
0.5906 0.2862 48.459
.4918 .2383 48.455
.5579 .2705 48.485
.6430 .3117 48.478
.6953 .3369 48.454 a
1.4192 .6880 48.478
.8307 .4027 48.477
-7980 .3869 48.484
.8538 .4139 48.477 ° .
1.1890 - .5763 * 48.469
Mean, 48.472+ .0024
Prom Gyn thes6s si ci2...'su sceeki sles sid deme eercaetene ses se eaeeees Yt=88.95
From analyses, Leesesterds cheek werew ss oe eee sis etstels Bassas cise Sierra stele Yt=88.97

These data of Jones were briefly criticized by Delafontaine,’ who regards. a lower value as
more probable. In a brief rejoinder? Jones defended his own work; but neither the attack nor
the reply needs further consideration here. They are referred to merely as part of the record.

By Muthmann and Béhm there is a single determination. 2 46505 grammes Yt,(SO,),
gave 1.19511 Yt,0,. Per cent Yt,O,, 48.482, and Yt=89.02.

Ina prelimninney note, G. and E. Urbain! state that Yt=88.6, but they give no details.
Three determinations by Bodman § are as follows:

¥t,03. Yt.(SO4)3. Per cent Yt,03.

0.4381 0.8928 49.070
5929 1.2093 49.028
4062 8286 49.022

 

Mean, 49.040+ .0102
Hence, Yt = 91.57.
The little known analyses of yttrium sulphate by Postius® are as follows:

¥t,(S0,)s- Yt,03. Per cent Yt,03.

0.32243 0.15588 48.345
.20460 .09897 48.370
.68477 -33120 48.367
.24013 .11606 48.332
-61195 -24792 48.315
56165 .27161 48,354
.53310 -25770 48.340

Hence, Yt = 88.41.

Mean, 48. 347+..0160.

 

1 Chem. News, 71, 243.
Chem. News, 71, 305.

4 Ber. Deutsch. chem. Ges., 33,42. 1900.
¢Compt. Rend., 132, 136. 1901.

6 Bihang Svensk, Vet, Akad. Handl.,26(2),No.3. 1901.
® Dissertation, Techn. Hochschule, Miinchen, 1902.
876 MEMOIRS NATIONAL ACADEMY OF: SCIENCES. (Von. XVI.

Meyer and Wuorinen' also studied yttrium sulphate.

‘Five syntheses from the oxide

were first made, which gave low values for the atomic weight of yttrium, due to acidity of the
sulphate. Two more syntheses followed, in which the excess of acid was determined, with

these results:

Yt,03.

0.5544
.9008

Hence, Yt=88. 68.

Yt,(80,)s.

1.1452
1.8595

Per cent Yt,03.

48.411
48,402

 

Mean; 48.406 +.0034.

The work of Meyer and Weinhaber? is essentially a continuation of the foregoing research.
Two series of determinations were made, one synthetic and one analytic, with due precautions

against acidity of the yttrium sulphate.

Yt,03.

0.4954
-4972
.4980

Hence, Yt =88.75..

Yt,(S0,)s.

1.0248
8585
1.4060
8944
8975
5686

Hence, Yt=88.73.

The data are as follows:
Syntheses.
— Fia(SO,)s

1.0231
1.0266
1.0286

Analyses.

¥1,03.
0.4962
4157
6808
4330
ATA3
2753

Per cent Yt,03.
‘48.421

48,432
48.415

 

Mean, 48.423-+ .0034

Per cent Yt,0s.

48.419
48.422

"48,421

48,412
48.423
48.417.

 

Mean, 48.419-+.0011

., Syntheses of yttrium sulphate have also been effected by Egan and Balke.? Their deter-
qth which they regard as not altogether satisfactory, are as follows:

1.00552
.87756
.20640
19823
.26409
"36243
134745
.28690
.22910
27464
.18760

Hence, Yt=89.69.

Y2,(S0,)s.
2.06854
1.79829

42557
40993
54237
74383
71295
.58787
.47098
-56434
-38580

Per cent ae

48.612
48.912
48.500
48.357
48.357
48.725
48.734
48.803
48.643
48.666
48.626

 

Mean, 48.630+ .0334

There are also two determinations by Brill,‘ made with the microbalance. The percentages

of Yt,O, in the sulphate are:

Hence, Yt =89.70.

48.647
48,617

 

Mean, 48.632+ .0100.

For the percentage of yttria in the sulphate we now have the following data, to be com-
bined in the usual way. The one determination by Muthmann and Béhm is arbitrarily given

 

equal weight with the figure assigned to Brill:

1 Zeitsch. anorg. Chem., 80, 26, 1913.
2 Ber. Deutsch. chem. Ges., 46, 2672, 1913,

" "Journ, Amor. Chem, Soe., 35, 365. 1913.”
4 Zeitsch. anorg. Chem., 47, 464. 1905.
No. 8} - ATOMIC WEIGHTS—CLARKE.
Cleve and Hoeglund............------ pe G eas 48. 605+. 0096
love, laters. sesxcsvsenevneee gases a s sonbanies cobae eesee ¢ 48. 503+. 0029
Jones, syntheses.............--.-- hod ced oS 2 leat 48. 467-4. 0025
Jones, analyses..............2.-0-020eeeeeeee bbeeeeeeeee de 48, 472+. 0024
Muthmann and Béhm...........-...-2: 20 peeee cece e cence 48, 4824. 0100
BOdM ans. sata ern ceeten <b ve damlarmaselebtlocae ames 49. 0404. 0102
Brill geosvasns Bisa eee ditscamueto sven doheatdad 48, 632+. 0100
ees Diece teens Son een , 48. 347+. 0160
Meyer and Wuorinen......... wee e ence cece eeeeeees eee 48, 406+. 0034

Meyer and Weinhaber, syntheses... ae 48. 4234-. 0034
Meyer and Weinhaber, analyses. .. - 48. 419+. 0011
Egan and Balke................. Rowen gee ceek te .--- 48,.630-b. 0334

senate MECAR sce dss s dahssknweer seas nnanteseawees 48, 444+. 0008
Hence, Yt= 88. 848-4. 0016.

 

377

An entirely different method for determining the atomic weight of yttrium was adopted
by Egan and Balke! The oxide was converted into chloride by solution in aqueous: hydro-
chloric acid, and the chloride was afterward dehydrated by careful heating in a stream of dry,

gaseous hydrochloric acid. The weights given and the ratios 2YtCl, : Yt,O, :

follows:
Yt,03. YiCl. Ratio.
0. 72180 1. 24092 58. 167
. 80392 ‘1 38437 58. 072°
. 70050 1. 20610 58. 080
. 73030 1, 25755 58. 073

 

Mean, 58. 098-4. 0155
‘Hence, Yt=90.22.

: 100 : % are as

A second series of determinations of this ratios is due to Hopkins and Balke? Theirfigures

are—
Y7,0,. YCi,. Ratio.
0. 56492 0. 97718 57. 811
- 51059 . 88348 57. 813
. 64232 1, 11092 57. 819
. 82157 1, 42087 57. 820
. 68246 1.17975 57. 848
. 54164 . 93688 57. 812

 

Mean, 57. Sa 0042
Hence, Yt= 88. 93.

_The two series chaiiea! give a general mean of 57.839 + 0040, from which a =89. 016 a:

O80. st
The ieieagnee aA by Feit and Przibylla, by their volumetric method, are as fol-
lows:
Yt,0;. 0. Atomic weight.
0. 3677 0. 07781 89. 415
. 4928 . 10438 89. 309.
- 3660 . 07749 89. 356
. 3660 .O7751 89. 328
. 3704 . 07840 89. 387
. 8635 . 07701 89. 284

Mean, 89. 89. 346 +. 0135

ote

for the atomic weight of yttrium:

From the sulphate.........-..---200 ee ee eee ere rece eee Yt=88, 848+. 0016

From the chloride. ..... Faianee Sbsigwed « Sis sratetglet eae o 89. 016-4. 0180
From the oxide.......---+--+-+-++-+2+0+ meeeedeanssesas ., 89.3464.0135 _
General mean.....-.- ee eee Yt=88. 856+. 0016

The sulphate ratios alone dominate this combination.

Computing with S= 32.065 + .00012, and Cl=35. 4605 + 00011, we have these three values

 

1 Loc. eit. a1 Journ. Amer. Chem. Soc., 38, 2332. 1916. 3 Zeltschr. anorg. Chem., 50, 262, 1906. For the process, see under lanthanum.
378 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [V*_, XVI.

ADDENDUM.

The determinations of the atomic weight of yttrium by Kremers and Hopxms' appeared
after the manuscript of this book had passed out of the author’s hands. They determined the
ratio between yttrium chloride and silver by the standard method. Theit figures, with vacuum
weights, are as follows: ioe

'YtCl,. Ag. At.ut. Yt.
3. 31143 5. 47636 89. 32
2.31979 3. 83587 89. 34
2. 26815 3.75045 © 89/35
“2.29376 3. 79302 | 89. 33
2. 200731 3, 31977 89,31
1.97610 - 3. 26827 89.30.
2.17949 3. 60389 89. 34
— a Oy
Mean, 89. 33 :

ite

ie the method shaioyen by Kisii ‘dnd Hopkins is more trustworthy ai the older
ones, their ene for yy Strain’ should be provisionally accepted.

LANTHANUM.

The early attempts to determine the atomic weight of lanthanum by Mosander and by
Choubine were only rough approximations to the true value, and haye no modern significance.
The determinations by Rammelsberg,? Holzmann,? Czudnowicz,! Hermann,’ Zschiesche,® and
Erk,’ and also the earliest measurements by Marignac," may also be disregarded as of no present
value. They are sufficiently discussed in the previous editions of this work. A later research
by Marignac,® however, is entitled to consideration here. His starting point was the sulphate
of lanthanum, purified by repeated crystallizations. In two determinations the-salt was cal-
cined, and the residual oxide weighed; in two others the lanthanum was precipitated as oxalate,
and converted into oxide by ignition. The data follow:

La,(S0,)3. La,0>. Per cent La,03.

2. 0988 1, 2082 57. 566
2, 3504 1.3532 57, 573
2. 8113 1.6165 57.,500
3. 3385 1.9215 57. 556
Mean, 57. ei 0112
Hence, La= 138.81. 4

From Cleve we have two separate investigations relative to the atomic weight of ‘innthanam,
In his first series’ strongly calcined La,O, spectroscopically pure, was dissolved in nitric. > acid
and then, by evaporation with sulphuric acid, converted into sulphate: ©. = ..,

La,03. Sulphate. Per cent.
1. 9215 3. 3365 57, 590
2. 0570 3. 5705 57. 611°
1. 6980 2. 9445 57. 667
2. 0840 3. 6170 57. 617
1, 9565 8.3960 57. 612

 

Mean, 57. 619-b. 0085
Hence, La = 139.28.
In his second paper," published nine years later, Cleve gives results similarly obtained,
but with lanthanum oxide much more completely freed from other earths. ' The data are as
follows, lettered to correspond to different fractions of the material studied:

 

1 Journ. Amer, Chem. Soc., 41, 718. 1919. 7 Jenaisches Zeitschrift, 6, 396. 1871.

* Poggend. Annalen, 55, 65. 1942. 6 Arch. Sci, Phys. Nat., (1), 11, 29. 1849. Oouvres Complétes, 1, 230.
8 Journ. prakt. Chem. 75, 321. 1858. 9 Ibid., (4), 30, 68, 1873. 0. C. 2, 566. :

(Thid., 80, 33. 1860. wk, Svensk, Vet. Akad, Handlingar, Bd. 2, No, 7. 1874.

6 Thid., 82, 396. 1861. 1K. Svensk. Vet. Akad. Handlingar, No. 2, 1883.
6 Tbid., 104, 174. oi ayers He Sone eal Seal te. eb, cis
No. '3.]

: La,0;.
B 0. 8390
1, 1861

. 8993

: . 8685
. 8515

. 6486

p{ . 7329
E 1. 2477

1 1621

: rh 5749
1, 3367
Ut, 4455

Hence, La = 138.35.

1. 4600
2.0643
1. 645
1. 5108
1. 4817
1. 1282
1. 2746
2. 1708
2.0217
2. 7407
2, 3248
2, 5146

‘ATOMIC WEIGHTS—CLARKE.

Sulphate.

Per cent.
57. 466
57. 458
57. 482
57. 486
57. 468
57. 490
57. 500
57. 490

67. 481
57. 463
57. 497
57. 484

Mean, 57. 480+. 0040

379

Brauner, in 1882, published two sets of determinations, both based upon the conversion

of pure La,O, into La,(SO,),.

In his first paper, Brauner‘ gives only two syntheses, as follows:

La;0,.-
1. 75933
. 92417.

Hence, La = 138.87.

In Brauner’s second ? paper six determinations are given, one being affect
which is corrected by a citation in Abegg’s Handbuch: ?
: . La,(80,)s.

La,0;.

0. 7850
2. 3500
2. 1052
1. 0010
1, 3807
1. 5275'

Hence, La = 138.30.

Brauner’s weighings are all reduced to a vacuum.

La,(SO,)s.

3. 05707
1. 60589

1. 3658
4.0917
3. 6633
1.7411
2. 4021
2. 6588

Per cent.
57. 566
57. 549

Mean, 57. 5575+. 0057

 

‘Per cent.
57. 476
57. 483
57. 467
57. 525
57. 479
57. 451

Mean, 57. 472+. 0086

 

ed by a misprint,

Both Bauer and Bettendorff made their determinations of the atomic weight of lanthanum

hy the same general method. Bauer’s data‘ are as follows:

La,03.
0. 6431
. 7825
1. 0112
. 7325

Hence, La = 138.76.
Bettendorff found *—
La,0;.
0. 9146
- 9395
- 9133
‘1 0651

Hence, La = 138.65.

1 Monats. Chem., 3, 1.
4 Monats. Chem., 3, 486.

8 Band 3, Abth. 1, p. 240, Brauner’s discussion of ths atomic weight.

Sulphate.
11171
1, 3613:
1. 7571
1. 2725

Sulphate.
1. 5900
1. 6332
1. 5877
1. 8515

Per cent.
57. 569
57, 482
57. 549
57. 564

Mean, 57. 541+. 0136

 

Per cent.
57, 522
57. 525
57. 523
57. 526

Mean, 57. 524+. 0006

4 Inaugural Dissertation, Freiburg, 1884,
5 Liebig’s Annalen, 256, 168, 1890.
380 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von. XVI.

The few determinations by, Wolcott Gibbs! were made by thé'oxalate method, which is
described in the chapter on cerium. His purpose, however, was rather to test the method
than to definitely fix an atomic weight. The data given are as follows, with the ratio 30,0,:
La,O, added:

La,03. "C403. Ratio..

45. 61 30.15 151. 827

46. 64 80. 07 151. 729
a 80. 08 151. 679
Mean, 45. 625 30. 11 151. 528

 

a : Mean, 151. 566+. 0607

Hence, La = 139.70. ‘

Gibbs cites three determinations, by the same method, made by Shapleigh, who found
La=139.75, 139.72 and 139.67. The weighings, however, are not given, and the data are
unavailable for present purposes. “

- In ‘1901 Bodman ? published’ three determinations, based on ae of lanthanum
elite: The figures are:

La,0;. La,(SO,)s- “ Per cent La,0;.
0. 4038 0. 7013 57. 579 ©

- 4408 - 7660 57. 546

. 4467 . 7758 57. 579

 

Mean, 57. 568+. 0070

Hence, La = 138.94.

In 1902 Jones * published his elaborate series of determinations, based upon scrupulously
purified materials. He effected twelve syntheses of lanthanum sulphate from the oxide, and
examined his product carefully for acid sulphate, whose presence would tend to lower the
apparent atomic weight of the metal. This source of error Jones regards as excluded from
his determinations. His results are as follows:

La,03. La,(SO,)3. Per cent La.03.
1. 0122 1, 7592 57. 588
1. 1268 1. 9581 57. 546
. 94585 1. 6437 57. 543
1. 0675 1. 8553 57. 538
. 9030 1. 5692 57. 545:
1. 1273 1. 9589 .. BT. 548
«9407 1. 6347 57. 546
1, 0455 1 8168 57. 546
1.1271 : 1. 9586 57. 546
1. 3074 2. 2720 57: 544
1, 3389 2. 3267 57. 545
1, 2012 2; 0874 BT. 545

——

Mean, 57. 544+. 0006

Hence, La = 138.78.

The atomic weight determinations by Brauner and Pavlitek‘ included a study of both
the sulphate and the oxalate methods. In lanthanum sulphate the acid salt was always found
to be present, and its amount was determined by titration, with sodium hydroxide, using
ethyl-orange as an indicator. The excess of acid thus measured tends to lower the apparent
atomic weight of lanthanum, and Brauner argues very forcibly that all previous determinations
of atomic weights among the rare earths are vitiated by this error. The authors give three
series of syntheses of the sulphate, in which corrections for the acid salt are applied. First,
there is a preliminary series, with weights in air. The data, with the acid correction, are as
dcllows, representing eight different. fractions of the oxide: :

1 Proc, Amer. Acad., 28, 250. 1893.

2 Bibang, Svensk. Vet. Akad. Handl., 26 (2), No.3. 1901.

2 Amer. Chem. Journ. , 28, 28. 1902.

4 Journ. Chem. Soc., 81, 1243, 1902. Preliminary notice in Proc. Chem. Soc., 17, 63. 1901.

 
No. 3.]

Ea,03.
0. 93205
. 8416

. 85993

. 7847

. 80645
1. 0760
11, 0683
. 8721

.. 9755.

. 9188

. 9507

. 9677
“oS "5 8570

Ganesan to, a@ vacuum this mean becomes 57.564+. 0047. Hence, La=138.91..

ATOMIC WEIGHTS—CLARKE. tt

& La,(S04)3. Per cent La.,03.
1. 6198 57. 541
1. 46234 57. 552
1. 49440 57. 543
1. 3635 57. 550
1.40145 57. 544
1. 86913 57. 567

11, 51479 57. 572
1. 51449 57. 584
1, 69381 57. 592
1, 5955 57. 587
1, 65040 57. 604
1. 68062 57. 580
1. 48736 57.619

 

Mean, 57. 572. 0047

‘

381

Its sig-

nificance, however, is diminished by the fact that the fractions:show a progressive change in
composition, which appears in the percentage column. The later fractions are higher in La,O,

than the earlier ones.

The next set of syntheses, representing a different series of lanthanum preparations, gave
the subjoined results, corrected for acid sulphate, but with weights in air:

La,03.
0. 8262
. 95652
. 45780
1, 34754
1, 17280

La,(SO.4)3. Per cent La,03.
1. 4353 57. 563
1. 66133 57. 576
. 79574 57. 538
2. 34074 57. 569
2. 03804 57. 545

Mean, 57. 558+. 0048

Corrected for weighing i in air this becomes 57.550. La=138.82.
This series, like the preceding one, is given by Brauner and Pavlitek as preliminary to more

exact work, which seems to have been continued by Brauner alone.

His final series of figures,

determined with extreme care, and with all corrections applied, including the reduction to a

vacuum, follows:
La,03.
_ 1.06562
~ 1.00694
1. 12553
1, 70276
1, 02460
1, 28650
1. 06488

Hence, La= 139. 05. Ja

La,(S0,4)3. Per cent La,Ox.
1. 85054 57. 5843
1. 74856 57. 5868
1. 95457 57. 5845
2. 95707 57. 5827
1, 77943 57, 5802
2, 23419 57. 5824
.. 1.84910 57. 5891

 

Mean, 57. 5843+. 00075

The oxalate series of “determinations by Brauner and Pavlitek is less satisfactory - than this
sulphate series, although it leads to sensibly the same value for the atomic weight of lanthanum.
Rejecting one experiment in their series, which is thrown out as abnormal by the authors, their
percentages are as given in the next table, together with the usual ratios:

Per cent C03.

/ 81.041

29, 673

Per cent La,03. Ratio. ,

46. 884 151. 039

146. 876 151. 013

44, 722 150. 716

wees y {445711 150. 679
eg “444, 694 150. 622
144, 719 150. 706

Ad. 746 150, 797

 

! These weights are erroneous, either through a ao ae in the original or because of copying. The percentage given is that calculated by

Brauner and Paviiéek.
382 MEMOIRS. NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

Per cent C,03. Per cent La,03. Ratio.
1 FR 44,879 150. 828
ane 44, 845 150. 714
31. 920 48. 197 150. 993
29. 689 44, 751 150. 733
46. 719 151. 277
883
anes ‘at 678 151. 145
48, 129 151. 530
31. 762 48, 118 151. 495

48, 133 161. 543
Mean, 150, 989-4. 0543

 

Hence, La= 139.08.

The Gibbs value for the same ratio is 151.566-+. 0607. The general mean of both series
is 151.246 + .0400.

In @ criticism of Jones’ determinations by the sulphate method Brauner ' reiterates his
statements concerning the acid salt, and also suggests other sources of error, such as contamina-
tion of the lanthanum preparations by cerium, and losses by spattering.. To this criticism
Jones ? promptly replied, giving a new series of determinations as follows:

La,03. La,(S0,)3. Per cent La,03.
-- ¥, 2161 2. 1132 57. 548
1, 6311 2. 8342 57. 551
1. 7804 3. 0938 57, 547
1. 4168 2. 4619 57. 549
1, 9702 3. 4235 57. 549

Mean, 57. 549+. 0010
Hence, La=138.81.

The material was spectroscopically pure, and the sulphate was neutral and soluble. The
operations were performed in porcelain crucibles, and the oxide was perfectly white. Brauner
used Paibly di sae! and Jones found that lanthanum oxide, heated in platinum, Pecans

tas, - Lay(SO,)s- Per cent La,03.
al 1. 2820 2, 2264 57. 582
ee “atak 1. 3885. 2. 4110 57. 590

Mean, 57. 586+. 0027

Hence, La=139.06, a value in accord with Brauner’s. According to Jones the discolora-
tion and variation in atomic weight suggest the presence of some other oxide than the normal
compound in Brauner’s preparations. The controversy, however, remains unsettled, and addi-
tional investigations are needed to determine the truth.

The two determinations by Brill * are of slight value, and hardly worth considering. Small
quantities of lanthanum sulphate were calcined to oxide, and the weighings were made with the
Nernst microbalance, in order to test its applicability to work of this kind. The percentages
of oxide in sulphate are given below, more for the sake. of completeness than for any real sig-

nificance i in them:
57. 664

57. 726

Mean, 57. 695+. 0207

Hence, La =139.79, a very high value.

In 1906 Feit and Pryzibylla‘ determined the atomic weights of several rare earth metals
by a special volumetric process, which, however, seems to be approximate rather than exact.
In each case the weighed oxide was dissolved in an excess of half-normal sulphuric acid, the
excess being afterwards measured by titration with tenth-normal sodium hydroxide solution.

 

1 Zeitsch. anorg. Chem., 33, 317. 1903. —

4 Zeitsch. anorg. Chem., 36,92, 1903. Chem. News, 88, 13.

§ Zeitsch. anorg. Chem., 47, 464. 1905.

4 Zeitsch. anorg. Chem., 50, 248. 1906. Sce also an earlier paper in Vol. 43, p. 213. 1905.
No. 3.) | “'." ATOMIC WEIGHTS—CLARKE. ...° 383

From the data thus obtained, by a process which is not clearly explained, the authors compute
the proportion of oxygen in the oxides, and thence deduce the atomic weights of the several
measurements. It would have been better to have given the H,SO, equivalent to the oxide,
and then to have made a more direct calculation. However, I cite fhe determinations for what
they may be worth, their value being essentially corroborative. For lanthanum the authors
give the following determinations:

La,0». 0. ‘Atomic weight.

t

0.5125 0. 07544 139, 05
. 5256 07731 a 139. 11
. 4835 . 07116 139. 08
5285 . 07706 139, 04
. 4815 . 07088 139. 03
5156 - 07585 139. 15
. 5348 . 07867 139.15

 

Mean, 139. 09+. 0430

For the percentage of oxide i in oe we nage estimates as follows:

" Marignac......,. nage Sesas caer whuewasth<yataesenes :. 57.549 +. 0112
Cleve, earlier series..............--- hve eten'e iianiree eae 57.619 +. 0085
Cleve, later seriés....... pois aedldnieMesends a eee Hache iote Riakees 57.480 +. 0040

' Brauner, 1882, first series.........:....02.20ee0-e00- 22620 57. BSTHL. 0057
Brauner, 1882, second series.......-.--.-- oisinveutes « w+-++-:57,472 +. 0086
BAUER: co sisian:saeiaivie’y oes eisai Wcities ere sow st oasis Disckete/ts 57.541 +. 0136
Bettendorff. ............ 2022 c cece eceee cece ee eeeeceeces 57.524 +. 0006
Bodman: s.40csthew toi saveeastedecesiesleenesdsestodss: BT; 568° +0070
TONES: 1902 sae celeste di cureladiace ies eka ceed saree ces 57.544 +. 0006
Brauner and Pavlitek, first............-..2202---eeeeeeee 57.564 +. 0047
Brauner and Pavliéek, second.....-.....-.-------2-+--- 57.550 +. 0048
Brauner and Pavli¢ek, third eeanoeecee nema ee esamorene 57. 5843+. 00075
Jones, 1903, porcelain series.’............-. Peden aaTeoe ees 57.549 +..0010
Jones, 1903, platinum series.....-..........+-----------, 57.586 +. 0027
Brill ..3.:.s ssaeseuseces vss 2 vase eeeeee¥e ens oss ee wemsee 57.695 +. 0207

, Ge@neralimean aesse ass scedecweeeunees sSesecnane te 57. 5465 +. 00034

This mean agrees very closely with the figures given by Jones.

There are now three ratios from which to deduce the atomic weight of lanthanum:
(1) La,(SO,), : La,O, :: 100 : 57.5465 + .00034. :
(2) 3 ©,0, : : La,O; : 100 : 151.246 + 0400.
(3) O : La ::16 : 139.09+.0480.5

Reducing these with S=32.065 4.00012, and C=12.0025 +.00019, we have—

From ratio—
Lee oat ed once La=138, 800+. 0015
Qeccemnasivns sopaystiion corns wsaiaaensois eeeee 139, 0904. 0430
Boweeeeeesegeereneeegeeneeeees cisco neal aca 189, 357+. 0803

Tt is deine unnecessary to. combine these values into a general mean, for only one of
them, that from ratio 1, carries any appreciable weight. The other values could not modify
it to: any noteworthy extent. The value La=138.8 is essentially that found by Jones, whose

work is entitled to high credit. Brauner, however, by two distinct methods, found La=139,
with much to be said in favor of his determinations. The question as to. the true atomic weight
of lanthanum is therefore not closed; and it. should be taken up anew by means ot other
methods than those heretofore employed. ay og
161566°—20-—25

'
384 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

CERIUM.

Although cerium, was discovered almost at the beginning of the nineteenth century, its
atomic. weight could not be properly determined until after the discovery of lanthanum and
“ didymium” by Mosander. The early attempts at the determination, however, may now be
disregarded, as having only historical interest. These attempts, by Beringer,? Hermann,? Marig-
nac,® Kjerulf,* Bunsen and Jegel, ’ and Rammelsberg,® are all sufficiently summarized in the
former editions of this work.

In all of these early determinations the ceric oxide obtained was somewhat colored, the
tint ranging from one shade to another of light brown, according to the amount of didymium
present. Still, at the best, a color remained, which was supposed to be characteristic of the
oxide itself. In 1868, however, some experiments of Dr. C. Wolf’ were posthumously made
public, which went to show that pure ceroso-ceric oxide is white, and that all samples previ-
ously studied were contaminated with some other earth, not necessarily didymium, but possibly
a new substance, the removal of which tended to lower the apparent atomic weight of cerium
very perceptibly.

Cerium sulphate was recrystallized at least 10 times. Even after 20 recrystallizations it
still showed spectroscopic traces of didymium. The water contained in each sample of the
salt was cautiously estimated, and the cerium was thrown down. by boiling concentrated solu-
tions of oxalic acid. The resulting oxalate was ignited with great care. I deduce from the
weighings the percentage of CeO, given by the anhydrous sulphate:

Sulphate. Water. CeO). Per cent Ce0,.
1, 4542 0. 19419 0. 76305 60. 559
1, 4104 . 1898 . 7377 60, 437
1. 35027 . 1820 . 70665 60. 487
Mean, 60, 494

After the foregoing experiments the sulphate was further purified by solution in nitric acid
and pouring into a large quantity of boiling water. The precipitate was converted into sulphate
and analyzed as before:

Sulphate. Water. Ce0,. Per cent CeO,
1. 4327 0. 2733 0. 69925 60. 311
1. 5056 . 2775 . 7405 60. 296
1, 44045 . 2710 . 7052 60. 300
* Mean; 60. 302

From another purification the following weights were obtained:
1. 4684 grm. 0. 1880 grm. 0. 7717 grm. 60. 270 per cent.

A last purification gave a still lower percentage:
1. 3756 grm. 0. 1832 grm. 0. 7186 grm. 60. 265 per cent.

The last oxide was perfectly white, and was spectroscopically free from didymium. In
each case the CeO, was titrated iodometrically for its excess of oxygen. It will be noticed
that in the successive series of determinations the percentage of CeO, steadily and strikingly
diminishes to an extent for which no ordinary impurity of didymium can account. The death
of Dr. Wolf interrupted the investigation, the results of which were edited and published by
Professor F. A. Genth.

In the light of more recent evidence, little weight can be given to these observations. All
the experiments, taken equally, give a mean percentage of CeO, from Ce, (SO,), of 60.366 +. 0308.
This mean has obviously little or no real significance. It gives Ce= 138.73.

 

1 Ann. Chem. Pharm.,, 42, 134, 1842, 5 Ibid., 105, 45. 1858.
3 Journ. prakt. Chem., 20,185. 1843. 6 Poggend, Annal en, 55, 65; 108, 44.
' Arch. Sci. Phys. Nat. (1), 8, 273. 1848. Ouevres Complates, 1, 215. 7 Amer. Journ. Sei. (2), 46, 53. 1868,

«Ann, Chem, Pharm., 87, 12,
No. 3.] ATOMIC WEIGHTS—CLARKE. 385

The determination of Wolf attracted little attention, except from Wing,' who partially
verified certain aspects of them. This chemist, incidentally to other researches, purified some
cerium sulphate after the method of Wolf, and made two similar analyses of it, as follows:

 

“Sulphate. , Water. ned CeO... ‘Per cent. Ce0,. '
1. 2885 0. 1707 0. 6732 60. 225
1. 4090 . 1857 . 7372 60. 263

Mean, 60. 244

Hence, Ce = 137.87.

The ceric oxide in this case was perfectly white. The cerium oxalate which yielded it was
precipitated boiling by a boiling concentrated solution of oxalic acid. The precipitate stood
twenty-four hours before filtering.

In 1875 Buehrig’s? paper upon the atomic weight of cerium was issued. He first studied
the sulphate, which, after eight crystallizations, still retained traces of free sulphuric acid. He
found, furthermore, that the salt obstinately retained traces of water, which could not be wholly
expelled by heat. without partial decomposition of the material. These sources of error proba-
bly affect all the previously cited series of determinations, although, in the case of Wolf’s work,
it is doubtful whether they could have influenced the atomic weight, of cerium by more than
one or two-tenths of a unit. Buehrig also found, as Marignac had earlier shown, that upon
precipitation of cerium ‘sulphate with barium chloride the barium sulphate invariably carried
down traces of cerium. Furthermore, the ceric oxide from the filtrate always contained barium.
For these reasons the sulphate was abandoned, and the atomic weight determinations of Buehrig
were made with air-dried oxalate,. This salt was placed in a series of platinum boats in a.com-
bustion tube behind copper oxide. It was then burned in a stream of pure, dry oxygen, and the
carbonic acid and water were collected after the usual method. Ten determinations were
made; in all of them the above-named products were estimated, and in five analyses the resulting
ceric oxide was also weighed. By deducting the water found from the weight of the air-dried
oxalate, the weight of the anhydrous oxalate is obtained, and the percentages of its constituents
are easily determined. In weighing, the articles weighed were avers counterpoised with similar
materials. ‘The following weights were found: °

 

 

Oxalate. |, Water. C03. Ce0,.
9. 8541 2. 1897 3.69420 hee ee
9. 5368 2, 1269 3. 5752 Mae
9. 2956 2. 0735 8.484500 eee

10. 0495 2, 2364 3.7704 sae

10. 8249 2.4145 |. 40586 9 ce...
9. 3679 2. 0907 f 3. 5118 4, 6150
9.7646 2. 1769 . 3. 6616 4.8133
9. 9026 2. 2073 . 3, 7139 4, 8824
9.9376 _ 2. 2170 3. 7251 ‘4.8971
9. 5324 2. 1267 3, 5735 4.6974

These figures give us the following percentages for CO, and CeO, in the anhydrous oxalate:
C0. Ce0,. :
48.2560 nn eee
48.2490 nee eee
48.248 0 na nee
48.257 0 nee ee
48087 nei
48. 258 _ 63, 417
48, 257 63. 436
48, 262 63. 446
48. 249 63. 429.
48, 253 63. 430
Mean, 48. 2546+. 001 Mean, 63. 4316-4. 0082
Hence, Ce= 141.55. Boss _.. Hence Ce= 141.47.

 

1 ae Journ. rr (2), 49, 358. 1870. 2 Journ. prakt. Chem., 120, 222. 1875,
386 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

. Robinson’s work, published in 1884, was based upon pure cerium chloride, prepared by
heating dry cerium oxalate in a stream of dry, gaseous hydrochloric acid. This compound
was titrated with standard solutions of pure silver, prepared according to Stas, and these were
weighed, not measured. In the third column I give the ratio between CeCl, and 100 parts of
silver:

CeCl,. Ag. Ratio.
5. 5361 7. 26630 76. 189
6. 0791 7. 98077 76. 172
6. 4761 8. 50626 76. 133
6. 98825 9, 18029 76, 122
6. 6873 , 8. 78015 76. 164
7.0077 9. 20156 76. 158
6. 9600 9. 13930 76. 150

 

Mean, 76. 155+. 0065

Reduced to a vacuum this becomes 76.167. Hence Ce=140.13.

_ Ina later paper,? Robinson discusses the color of ceric oxide, and criticizes the work of Wolf.
He shows that the pure oxide is not white, and makes it appear probable that Wolf’s materials
were contaminated with compounds of lanthanum. He also urges that Wolf’s cerium sulphate
could not have been absolutely definite, because of defects in the method by which it was
dehydrated.

Brauner,? in 1885, investigated cerium sulphate with extreme care, and appears to have
obtained material free from all other earths and absolutely homogeneous. The anhydrous
salt was calcined with all necessary precautions, and the data obtained, reduced to a vacuum
were as follows:

Ce,(SO,)3. Ce0,. Per cent. Ce0,.
2. 16769 “1, 31296 60. 5693
2. 43030 1. 47250 60. 5707
2. 07820 1. 25860 60. 5620
2. 21206 1, 33989 60. 5721
1. 28448 77845 60. 6043
1. 95540 1. 18436 60. 5687
2. 46486 1. 49290 60. 5673
2. 04181 1. 23733 60. 5997
2.17714 1. 31878 60. 5739
2. 09138 1, 26654 60. 5605
2. 21401 1. 34139 60. 5863
2. 44947 1. 48367 60. 5711

_ 2.22977 1, 35078 60. 5771
2. 73662 1. 65699 60. 5486
2. 62614 1. 59050 60. 5642
1. 67544 1. 01470 60. 5632
1. 57655 . 95540 60. 6007
2. 72882, 1. 65256 60. 5600
2. 10455 1. 27476 60. 5716
2. 10735 1. 27698 60. 5965
2. 43557 1. 47517 60. 5692
3. 01369 1. 82524 60. 5649
4, 97694 8. 01372 60. 5537

 

Mean, 60. 5729+. 0021
Hence, Ce= 140.22.

In 1895 several papers upon the cerite earths were published by Schutzenberger.* In the
first of these a single determination of atomic weight is given. Pure CeQ,, of a yellowish-
white color, was converted into sulphate, which was dried in a current of dry air at 440°. This
salt, dissolved in water, was poured into a hot solution of caustic soda, made from sodium, and,
after filtration and washing, the filtrate, acidulated with hydrochloric acid, was precipitated

 

1 Chemical News, 50, 251. 1884. Proc. Roy. Soc., 37, 150. 3 Sitzungsb. Wien. Akad., Bd. 92. July, 1885,
2 Chemical News, 54, 229. 1886. 4Compt. Rend., 120, pp. 663, 962, and 1148, 1805.
No. 3.1 ATOMIC WEIGHTS—CLARKE. 387

with barium chloride. The trace of sulphuric acid retained by the cerium hydroxide was
recovered by re-solution and a second precipitation and added to the main amount. One
hundred parts of Ce,(SO,), gave 123.30 of BaSO,. Hence, Ce=139.87..

Schutzenberger, criticizing Brauner’s work, claims that the latter was affected by a loss of
oxygen during the calcination of the cerium dioxide.

In his second and third papers Schutzenberger describes the results obtained upon the
fractional crystallization of cerium sulphate. Preparations were thus made yielding oxides
of various colors—canary-yellow, rose, yellowish-rose, reddish and brownish-red. These oxides,
by synthesis of sulphates, the barium-sulphate method, etc., gave varying values for the
atomic weight of cerium, ranging from 137.7 to 143.3. Schutzenberger therefore infers that
cerium oxide from cerite contains small quantities of another earth of lower molecular weight;
but the results as given are not conclusive. The third paper is essentially a continuation of
the second, with reference to the didymiums.!

Schutzenberger’s papers were promptly followed by one from Brauner,? who claimed
priority in the matter of fractionation, and gave some new data, the latter tending to show that
cerium oxide is a mixture of atleast twoearths, One of these, of a dark salmon color, he ascribed
to a new element, ‘‘meta-cerium.”” The other he called cerium, and gave for it a preliminary
atomic weight determination, The pure oxalate, by Gibbs’ method, gave 46.934 per cent
of CeO,, and, on titration with potassium permanganate, 29.503 and 29.506 per cent of C,O,.
Hence, Ce=139.81. In mean, this ratio may be written—

3C,0, : 2CeO, :: 100 : 159.074

which will be combined with other corresponding expressions later.

Wyrouboff and Verneuil * determined the atomic weight of cerium by analyses of the
sulphate Ce,(SO,);.8H,O. The salt was prepared from three different sources, two samples from
monazite and one from cerite. It was dehydrated at 250°, and reduced to CeO, at 1,500°.
The latter was perfectly white. The weights were as follows:

Hydrous sulphate. Anhydrous sulphate. Ce0,.
1. 2385 0. 9875 0. 5977

1. 2730 1. 0148 - 6138

1. 2030 - 9590 . 5794

1, 5420 1, 2295 . 7430

. «9642 . 7685 - 4642
1, 3260 1. 0571 . 6389

1, 1429 ~ . 9112 . 5512

. 9072 - 7232 . 4372

1, 2114 . 9658 . 5840

1, 2411 3 - 9894 - 5984

Hence, the following percentages: .
#0. CeO, in hydrate. 00,5 in OHf00

 

 

20. 267 48, 259 60. 526
20. 282 48. 216 a 60. 484
20. 282 48. 162 60. 417.
20. 265 48. 184 60. 431
20. 296 48. 143 60. 403
- 20. 279 48. 182 60. 438
20, 278 48. 280 60. 557
20. 282 . 48. 192 60. 453
20, 274 48. 208 60, 468
20. 280 48, 215 60. 481
Mean, 20,278 + .0019 48.204 + .0095 60. 466 + .0103
Hence, Ce=139.22 139.39 189. 45,

 

1 Similar results were also obtained by Boudouard (Compt. Rend., 1 125, 772. 1897), The correctness of these conclusions is questionable.

2 Chem, News, 71, 283.
3 Bull. Soc. — (B),-17, 679. ison: Also Ann. Chim, i @), 9, 349, Moissan, Compt. Rend., 124, 123 also describes white ceric oxide,
388 MEMOIRS NATIONAL ACADEMY: OF SCIENCES. [Vou.. XVI.

: Kélle * studied anhydrous cerium ape which, he necueed oe calcination to oe . His
figures are as follows:

Ce(SO4)s. - ; w? “C005. Bais ats Per cent tat

1. 84760 1. 11648 Poh 60.429
1. 16074 -70078 , SS, , 80, 881
1, 53599 . 92722 -60, 366
"97196 | . 58661 60.353
e “4, 40874 84760 5 60,884
1. 75492 1, 05956 ‘60.377
1.53784 _ . 92853 60. 379,
1, 64233 , 99150 60.372

Mean, 60. 374.0067 _

«Hence, Ce = 138.79, an siigaatially? low ‘ail improbable value. i ON

The very careful investigation by Brauner and Baték * involved the study of two cerium
salts, the sulphate and the oxalate. The sulphate was dehydrated at 440°, and then calcined to
oxide, The figures given in the next table Tepresent a number of eciifieneltt fale of me salt,
aut I have here combined the data into one series:

Ce(SO,4)s. ‘ ~ CeO). Per cent. CeOa:
- 1.5074 0. 9130 60.568 9)
1:.7979 “1. 08945 60. 596.
1, 5937 9665 ; 60. 645
2, 6240 1. 5895 60. 575
1. 2161 . 7370. . 60. 604
1. 5074 . 9180 , 60. 568,
1.2192 7386 60. 581.

\

oe "Mean, 60.5914.0070 —
Reduced’ to a vacuum this becomes 60. 584.
~ Hence, Ce= 140.30.
Combining this series with others, we have—

60.366 + .0308
60. 244 + . 0308
as 60. 5729+ . 0021

Wyrouboff and Verneuil.......... at itisleiisiey dedeactows 60. 466 + . 0103
Kélle.. a ie AL iavaseeidishiadiaccenad Op ae Ne?
Besunerand Bates. se teeeeeteeteeseceeeeeeeees-- 60.584 + .0070
Ganeril teat go cscs acsoce: nti Ra ches Ete eg 60. 5528+ . 0019

 

Wing’s mean is here arbitrarily given equal weight with that of Wolf, but both series
practically vanish.

Cerium oxalate contains water, in proportions which are not absolutely constant; at least
not constant enough for good atomic weight determinations.’ In Buehrig’s analyses the water
was estimated, but itis doubtful whether the estimations can be made with adequate sharpness.
Cerium oxalate, therefore,:is best handled by the method of Stolbaiand Gibbs; which consists
in determining the amount of ceric oxide left after calcination; and in another portion of the
same sample, estimating the radicle C,O, by titration with potassium, permanganate. From
the ratio 3C,0, : 2CeO, the atomic weight of cerium can be calculated.

This method was followed by Brauner, in. a single determination which has already been
cited. It was also adopted by Brauner and Baték, who give five sets:of determinations, with
vacuum weights, as follows. I cite now only the percentages of CeO, and-C, O,, as computed
from the weighings, together with the required ratio:

, Sample I.
Per cent CeO,. Per cent C,05. Ratio.
46. 949 ‘ 29.423 | 150. 503

46. 939 29, 391 159, 685

 

! Beitriige zur Kenntniss des Cers. Inaugural Dissertation, Ziirich, 1898. 2 Zeitsch. anorg. Chem., 34, 103. 1903.
No. 3.] ATOMIC WEIGHTS—CLARKE. 389

 

Per cent CeO,. Per cent C03. Ratio.
46. 913 29, 393 159. 664
46. 920 29. 442 159. 398
29, 422 159. 507 -
Mean, 46. 930 29. 397 159, 642
29. 406 159. 593
29. 459 159. 306
29, 414 159, 540
29. 459 7 150. 306
29. 459 7 159. 306
29. 436 pee “159. 431°

The ratio here is computed from the individual figures for C,0, and the mean for CeO,.

Sample Il.
Per cent CeO». Percent C0;  §  ————~Ratio,
47.197 29. 601 | 159. 353
47, 089 29.564 |: (+ 159, 652.
47, 225 29.559 . +. 159.579
Mean, 47.170 . i 3
Samples III, IV, V.
Per cent CeO... Per cent C,03. Ratio.
47,161) ~"
II . . 51 :
I nee iy 29. 512 | 159. 718
46, 926
me oe oS 29, 391 159. 654
Vv 47.039 29. 531 159; 287

The mean of the 18 values for the ratio is 159.501 +.0285. Hence Ce=140.27.
In a later memoir, Brauner’ gives additional analyses of cerium oxalate and sulphate.
The oxalate figures are as follows:

Per cent CeO,. Per cent C,0,. Ratio.
47. 070 29. 548 159. 304
47. 067 29. 544 159. 325
47.077 29. 478 159. 682
47.070 29. 486 159. 638
aus Mean, 159. 487-4. 0677
Mean, 47. 071 :

Hence, Ce= 140.26.
If we give to Brauner’s earliest, single determination, the weight of one experiment-in the
Brauner and Baték series, the values for this ratio combine thus:

Brauner, early........-... Filed nee AO octeatebciees 159. 074+. 0990
Brauner and Baték........--.- igiglslele v b's gieweceetegwiacigte's sie 159. 501+. 0285
Brauner, latest  deomseaea wien S's es eee 159. 487+. 0677

General mean...........--.-- westeepetesetteetee 159. 471 4. 0254

 

th the memoir last cited Brauner also gives a series of determinations based on the cal-
cination of the octohydrated cerium sulphate. In the subjoined table I include two separate
determinations given near the beginning ¢ of the paper. All weights were reduced to a vacuum

standard: Sulphate. »  €eOz. Per cent C20.
, 1. 98989 0. 96175 48, 332
1. 99154 . 96251 48. 330
2. 33919 _ 1.13027 48. 319.
1. 95882 nt "94679 48. 335
‘1. 20961 - 58453 ' 48, 324
1. 54162 . 74048 48, 329
1.67748. . 81074: ca 48.331
_ 2, 02736 . 97985 48,331

Mean, 48. 48.329 +.0011

 

1 Zeitsch, anorg. Chem.,, 34, 207, 1903,
390 MEMOIRS ‘NATIONAL ACADEMY OF ‘SCIENCES. [Vou, XVI.

Brauner rejects the third determination, a procedure which changes the mean to 48.330 +
0009. Wemay adopt the latter and combine it with another series, thus:

Wyrouboff and Verneuil................0..2002ceeeeeeeees 48, 204+. 0095
Braun Gl fe weseew ones sexes sere bewinhaweees ansisaesseesce 48: 330+. 0009
7 iit ag ae eae
General mean............- aed 48, 329 ++. 0009

= Og

From Brauner’s series Ce = 140.26.

Brauner discusses at some length the color of ceric oxide, and describes it as nearly white,
but with a faint tinge of brownish-yellow. Deb

_ The Tatios, good and bad, for cerium now are—

(1) Percentage CeO, from Ce, (SO,)s, 60.5528 ¥ 0019.
(2) Percentage CeO, in Ce,(SO,),8H,O, 48.329 + .0009
(3) Percentage H,O in Ce, (SO,),.8H,O, 20.278 + .0019
(4) Percentage CeO, from Ce, (C,0,)5. 63.4316 +.0032
(5) Percentage CO, from Ce, (C,0,),,48.2546 + .001
(6) 3Ag : CeCl, :: 100 76.167 +.0065.

(7) 3C,0, : 2CeO, : : 100 : 159.471 2 0254

Lt eh ALY %
The antecedent atomic weights Oe aie
} yy mE ad
7 Ag=l07. 880 4.00016 8=32.065 4.00012) __
“ Ol= 35.4605+.0001)  <!" ° C=12, 0025+.00019 «Ul
— H=1. 0078+. 0001 apie 2.
Hence, He se 7 lean
From ratio— Pes die oa 2
errata pee waneeneee ee cee Ce=139, 212+. 0276
P Dpeseeseeaxs cess aetrperiewee seneores (abe ears - 140. 078+. 0088

peers ween enaengacgereeserseregeeseeceeepeeseeng, 140, 1274. 0210
ena dpybepunnne ne a2 : sieseeereeeeceeeeeseeesesere | 140, 240-4, 0275

ssraieThyassieseserouatescuciesaic aay hea yeesasd EaGisusbiaihe Srayereajatereseiaadsa 141. 473-4. 0166
Gaia y ete 2 uate ar Seavey lel tah Loic Looe ec rele 141. 559+. 0041

Ce= 140, 841+. 0027

 

This mean appears to be, on chemical grounds, too high; because of the evident over-
weighting of ratio 5. If we reject the aberrant values, given by ratios 3, ss and 5, the general
mean of the remaining values is

Ce=140. 214+. 0038.

OCs This represents mainly shy work of Brauner and Robinson. '

godin Of, 4 7 hw q

Soe Se, PRASEODYMIUM.

In 1885 Auer ¥ von Welsbach ' succeeded in proving that the old waldymia! ’ was a mixture
of two earths, one yielding green and the other rose-colored salts. ‘To the corresponding metals,
praseodymium and neodymium, he assigned the atomic weights Pr=143.6 and Nd=140.8,
respectively, values which were curiously reversed, either in printing or by the error of a copyist.
The true values are now known'to, be nearly Pr=141 and Nd=144, in round numbers. For
“didymium,” many discordant ‘atomic weight determinations had been made," which now
have only historical interest, and need therefore no consideration now. They are thoroughly
summed up in the first edition of this work, which was published about three years before
Welsbach’s brilliant discovery. $5

In 1898 Brauner? published a preliminary notice upon praseodymium. Thirteen deter-
minations of the atomic-weight, by both the sulphate and the oxalate methods, gave values
from 140.84 to 141.19, in mean 140.95, but the details of the work.were not given. These
early data, therefore, are not now available for discussion. The first fully described series of
determinations was made by Jones,’ who published his results a little later than Brauner.

 

Amer. Chem. Jourii,, 20,345. 1898.

"1 Monatsh. Chem., 6, 477, 1885. 2 Proc. Chem. Sac., 14,70, 1898. ,
No. 8.] »« » ATOMIC WEIGHTS—CLARKE. ‘391

Jones effected the synthesis of praseodymium sulphate from the oxide, the latter having
been first reduced from Pr,O, to Pr,O, by heating in hydrogen. The material, after purification,
still contained minute traces of lanthanum and neodymium, but these were too small to seriously
affect the atomic weight determination, The weights and percentages appear in the following
table: na : a ries ae
Pr,05. Pr,(SO,)3. Percent PrQs.

0. 5250 0. 9085 57. 789
. 6436 1, 1135 ia 57. 800
. 7967 i 1. 3788 57. 782
. 7522 1. 3018 57. 782 ©
. 7788 1, 3473 57. 805
. 6458 1.1172 57. 805

. . 6972 1. 2062 57. 801
. 7204. 1, 2464 57. 798
8665 1. 4990 57. 805
. 6717 1, 1624 . 67. 796
. 7439 1. 2873 . 57.788
. 6487 1. 1224 57. 796

 

Mean, 57. 796+. 0016
Hence, Pr = 140.46.

C von Schéele,! also in 1898, gave three series of determinations. First, by the synthesis
of praseodymium sulphate, which, however, contained a little lanthanum sulphate, as follows:

Pry. Pr,(SO,)s- ‘Percent PrOx.
1, 6738 2, 8926 BT. 865
1, 4327 2. 4788 57. 798
1. 1105 1, 9221 57.775
1, 0072 1. 7481 57. 782

 

Mean, 57. 805+. 0139

Hence, Pr = 140.53.

The second series by von Schéele is somewhat obscure, being a combination of ‘the oxalate
and sulphate methods, A part of the oxalate was converted into sulphate, and in another
part the C,0, radicle was determined by titration with permanganate solution. ‘The proportion
of oxide in sulphate was also found. I give below the several percentages, and also the ratio
3C,0, : Pr,(SO,)s: ' e 7 a

 

 

Sulphate from oxalate. Per cent C,03. Pr,0, in sulphate. Ratio.
81, 682 31. 07 57, 73
81. 638 : 31. 06 57.71
31,11 -
Mean, 81. 665 —
Mean, 31. 08 262. 515
77. 828 29. 60 57.77 262. 845
29. 58
29, 64
Mean, 29. 61 . ‘87, 737 262. 680
+. 0117 +. 1113
Pr=140. 06 189, 62

These two series have small’ claims to consideration, and may be regarded as preliminary.
The third series, by the sulphate method, is far better:

Pr,O3. Pr,(SO,)s. Per cent Pr,O3.
0. 6872 1, 1890 57. 796°

. 7834 1, 8550 ‘7. 815

. 6510 1, 1260 57.815 |

. 7640 1, 3216 57. 809

. 5183 . 8967 57, 801

a
dace ee

Fat con “Mean, 57. 807+.0026-
Hence, Pr=140.54 ere e

py ~. “t Zeitsch, anorg. Chem., 17, 310, 1898.
392 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

In 1901 Brauner + gave a preliminary notice of an investigation upon the atomic weight
of praseodymium, but without details.; He has since publaed his data in Abegg’s Hetdbod
der anorganischen Chemie,’ as: follows:

First, the ootohydrated sulphate was ss dehidinutee ab 500°, ven ceiaiell to ond aid the
latter finally analyzed iodometrically to determine the true proportions of Pr,O,. Percentage A

in the next table is that of Pr,O, in the hydrous ‘sulphate, and B refers ‘to the anhydrous salt:

 

 

Pr(SO,4)3. 8H,0. Pr (SOg)4. Pr.03. Percent A.. — Percent B.
1, 29269 1. 038242 0. 59747 46,219 57. 871
1, 27990 1. 02198 . 59137 46.204 . 57. 868
Mean, 46. 211 57. 8695
+. 0050
Pr= 141.09

Secondly, four samples of praseodymium oxalate were analyzed. by the method already
described under cerium and Jecrelanura. I give the ratio computed from, the percentages,

in the form 8C,0, : Pr,O, : :100:2:
Sample. Per cent Pr,O3. Per eit C03. Ratio. “Weight.
ene 45, 183 29. 581 i =
# { 29, oi Mee ias oe a
: (45.142 | a9/499p
145, 098 29: 598] -- Sk vee man let
B 445, 095 29.532) 152.671 4
‘145, 102 29, 511)
45. 063 denen
45. 032 29, 503
o is 128 i ist Toney 2
45. 136 29. a!
_,D {* 20, BAD 152. 677 2

Each value for the ratio is here weighted according to the. number of the CO; determina-

tions.
"Hence, Pr= 140.95.

‘Finally, Brauner affected the synthesis of the anhydrous sulphate.

The means of all, thus weighted, is 152.720 +. 0148. ada 3

Praseodymium c oxalate

was calcined, and the composition of the oxide produced was ascertained by iodometric titration.
It was then converted into sulphate, with the correction, described under lanthanum, for
excess of sulphuric acid. The corrected data are these:

Pr,03. Pr,(SO,)3- Per cent Pr,O3.
0. 73359 1, 26782 57. 863
. 64871 1.12059 57.890
* , 74103 1.28051 57.870
, 72894 1. 25972 57. 865
. 36559 . 63350 57. 867
. 82769 1. 43024 57.871

In his first series Brauner found, for the same percentage, 57.871 and 57.868.

Including

these, the eight determinations, taken as one series, give a mean of 57.871 4.0021.

Hence, Pr= 140.97.

vay

Combining this means with other means for the same ratio we have—

VONCB veces eeweees veces ee sclizewned es iataeeameencee Sets sehr (O12 796 0016
y HEROS es pn deotaneuaees meee eeekekek te setess sae 2. 57.805 +.0139

ee ne eee Bikes Baetecae ses saved tay cule
aie asin easement tenes 57.807 -+.0026

POLE re sirens iciieaumneenezaes anaes 57.871 +. 0021

sayayulat Gna iutaiaseatel dente cate graitya Onin eran 57. 3194+. 0010

Three other determinations, by Welsbach® were published in 1903. He used the sulphate
method and found Pr=140. 64, 140.50, and 140.56 when O=16. Unfortunately, he gave no

 

1 Proc. Chem, Soce., 17, 65.

2 Bd,3, Abth: 1), pp. 263-265. ,

see

. §Sitzungsh, Wien Akad., 112, 1037. 1903.
No. 8. _ ATOMIC WEIGHTS—CLARKE. =; 393

weighings, nor did he state what value he used for the atomic weight'of sulphur. His figures,
therefore, are unavailable for discussion now.

Three determinations by Feit and Przibylla,! who used the peculiar volumetric method
already described under lanthanum, are as follows:

P7203. 0. Atomic weight.
0. 54010 0.07879 140. 518

. 53420 07789 140. 601

. 50054 07302 140. 516

 

Mean, 140. 5454. 0189

The most extensive and thorough determinations of the atomic weight of praseodymium
are by Baxter and Stewart,? who analyzed the chloride by the standard methods, Two series
of figures are given, representing the two usual ratios, with the following results. All weights
are reduced to a vacuum:

 

Series 1. _

PrCly. Ag. Ratio.
7. 66554 10.03129 76. 4163
8.78959 11. 50311 76.4106
5. 01155 6. 55898 76.4075
6.04235 7.90820 : 76. 4061
6.14745 8.04563 76. 4073
6. 32550 8.27866 76. 4054
5. 12982 6.71359 76. 4095

: 4.59463 6.01331 76. 4077
- 4.77556 6.24994 76. 4097

5.96661 7, 80908 76. 4060

6. 87536 8.99824 76.4079
| | 5.73602 7.50707 76. 4082
. 4, 64585 6.08029 76. 4084

6. 85492 8.97148 76.4079
8.01311 10. 49250 76. 4080
6.17045 8.07425 76.4214
6.90040 9.03064 76.4110
7.08498 9.27274 76. 4065
5.81310 7.60811 76. 4066
6. 20845 8.12541 76. 4079
7.07590 9. 26045 76. 4099
5.77646 7.56000 76. 4082
5. 72002 7.48596 76.4100

Mean, 76. 4090+. 0006
Hence, Pr= 140.909. .
Series 2.

PrCly. AgCl. Ratio.
4.12848 7.17915 57. 5065
6.01605 12. 02533 57. 5128
7, 66554 18. 32981 57. 5089
5.01155 8.71379 57.5128
6.04235 10. 50744 57. 5054
6.14745 10. 69050 57. 5039
6. 32550 10. 99855 57. 5121
4. 69463 7.98940 _ 87.5090
5.96661 10, 37554 57. 5065
6. 87536 11. 95508 _ 57.5100
4, 64585 8.07758 57. 5154
6. 85492 11. 91774 ° 57, 5186
8.01711 18. 93953 57.5135
6.17045 10. 72833 57. 5155
6. 90040 11. 99946 57. 5059
7.08498 12. 32019 57. 5071:

 

1 Zeitsobr. anorg. Chem., 50, 258. 1906, Journ. Amer. Chem. Soc., 37, 516, 1915,
394 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou XVE,

Preis. > Ag Cl. ‘Ratio. +
6. 20845 10.79552 si, oo. ,,B7,'5095
7.07590 —, 1230401 4 67,5089
5. 77646 “" 10.04403 ‘57.5113

5.72002 9.94651 7, 5078!

Mean, 57: 5100+. 0007
Hence, Pr= 140.924. ea ail q

The ratios for praseodymium now are—.__ 1 att

(1) Pr,(S0,).-8H,O : Pr,O, :: 100 : 46,211 + .0050.
(2) Pr,(SO,), : Pr,O, :100 : 57.8194 + .0010.

(3) 3C,0, : Pr,(SO,), :: 100 : 262.680 4.1113.

(4) 30,0, : Pr,O, :: 100 : 152.720 +.0148.

(5) O: Pr :: 16 : 140.545 +.0189.

(6) 3Ag : PrCl, :: 100 : 76.4090 + .0006.

(7) 3AgCl : PrCl, 7: 100 : 57.5100 +.0007.

To reduce these ratios we have—

' Ag=107.880 -+.00016 S=32.065 +.00012
°Cl= 35.4605+.00011 C=12.0025-+ .00019
H=1.0078 4.0001 oe
Hence,
From ratio—

Bit wthcaseudetainets cesaubenmuer cesses see Seake Pr=139, 616-£. 1200
ee inc ad 140. 545+. 0189
140: 6244. 0048

140, 909+. 0022
140,924 +; 0032
140.:949.+. 0160
141. 088-4. 0235

General mean.........--.----2 eee eee eee Pr=140, 8 8764. 0017

This means is a little lower than the values found by Baxter and: Stewart, and the latest
determinations by Brauner. The work of Jones, however, should not ‘be entirely rejected.
The rounded off value 140.9 is probably near the truth.

NEODYMIUM.

Our knowledge of the atomic weight of: neodymium i is almost entirely based upon studies
of the sulphate and the chloride. Welsbach’s first determination was-cited under praseodym-
ium, and needs no further consideration. So also Brauner’s’ first preliminary figure Nd=
143.63, given without analytical details, may be dismissed here. The first important series of
determinations is that by Jones,? published i in 1898. The ees’ of me sulphate was effected
in the usual way, with the following results:

 

_Nd0; Na,(SO,), Per cent Na,
0. 8910 1. 5296 58. 251
7880 1. 3530 58, 241
- 9084 1, 5509 58.250
. 7668 1.3166 58.241
. 8908 1, 5296 58, 237
. 8848 1.5194 58, 234
. 8681 1. 4908, 58. 250
, 8216 1.4108 58. 257
. 8531 1, 4646 58. 248
. 8711 1. 4957 5S. 240
. 8932 1, 5832 58. 257
. 8893 1. 5268 58. 246

i Mean, 58. 246+. 0015
Hence, Nd = 143.53. meow.

1 Proe, Chem. Soe., 14, 72." 1898. 2 Amer, Chem. Journ., 20, 345. 1808. See additional note in Zeltsch. anorg. Chém., 19, 339. 1899.

 
Noo3) -. °! (ATOMIC. WEIGHTS—CLARKE. 395

One determination, by Boudouard,! was made by calcination of the sulphate. 2.758
grammes of Nd,(SO,), gave 1.605 of Nd,O,, or 58.194 per cent. Hence, Nd=143.18.
Brill,? in 1905, made two analyses of neodymium sulphate, with the aid of the microbalance.

His percentages of Nd,O, are—
58, 000
58. 180
Mean, 58. 090+. 0600
Hence, Nd = 142.45.

In Abegg’s Handbuch, Brauner® gives the details of a synthesis of neodymium sulphate,
with corrections for excess of acid. 0.93788 gramme Nd,0, gave 1.60873 N d,(SO,)s. Per cent
Nd,0,, 58.299, whence Nd = 143.90.

Holmberg,! who employed the usual synthetic method, found no serious difficulty in
obtaining a neutral sulphate. © In his series of determinations, therefore, a correction for excess
of sulphuric acid was not needed. His six syntheses are as follows:

Nd,0, Nd,(SO,)3 Per cent Na,0,
0. 9692 1. 6618 58. 322
. 6584 1. 1287 58. 333
1. 0292 1. 7648 58. 335
1, 0118 1, 7346 58. 330
vc, «8818 - 9462 58. 317
53845 . 9164 58. 326

at Mean, 58. 327+. 0019
Hence, Nd = 144.09.

In combining these various determinations of the oxide-sulphate ratio, the single experi-
ments by Boudouard and Brauner are each given the probable error of one experiment in
Jones’s series:

 

JONES ea asic sieicisiss kawiseseineececeeenatowamimwcides Suesnoee ved s DSi 246 0015
Boudouard wbash Peatatee 58.194 +. 0067
Brritliosc atendepeteciche sc cevetcs chet heii ties ae cesses 58,090 4.0600
Brauner............-. Sb datyalhat crite eye nls atu a emilee ... 58,299 +. 0067
'- Holmberg in - 58.327 +.0019
General mean..........2..02.000-ceeeeeeeee Bed fs 58. 2831+. 0011

In this combination only the series by Jones and Holmberg are important. The other
figures count for little or nothing. Welsbach’s* determinations, also by the sulphate method,
cannot be safely utilized, for lack of details. He found Nd=144.55, 144.52, and 144.57; in
mean, 144.547+.0103. ..

Feit and Przibylla® give the following cote a neodymium, obtained by their volumetric
method:

Nad,0, 0. Atomic weight
0. 5380 0. 07661 144, 542

5388: '. 07675 144, 485

. 5358 . 07632 144, 491

. 5265 . 07497 144. 547

Mean, 144. 516+. 0111

The determinations of the atomic weight of neodymium by Baxter and Chapin’ are based
upon analyses cf the chloride by the usual standard methods. Their data, with vacuum weights,
are as follows:

NdCl,. Ag. +a AgCl, Ag ratio. AgCl ratio.
8.16218 ——......- 5.42546 —— ....... 58. 2841
2.98305 heen 5.08226 —.....-. 58. 2850

 

1 Comp. Rend., 126, 900. 1898.

9 Zeitsch, anorg, Chem., 47, 464. 1908.

3 Abege’s Handbuch, Bd. 3, Abth. 1, p. 276. Preliminary note in Proc. Chem. Soc., 17, 66. 1901.
«Zeitseh. anorg. Chem., 43, 124, 1907.

5 Sitzungsb. Wien Akad., 112, 1037. 1903.

6 Zeitsch. anorg. Chem., 50, 259. 1906. For the process, see under lanthanum.

7 Proc, Amer, Acad., 46, 215. 1910.
396

NaCl. Ag. AgCi. Ag ratio.

© 2.99149 Le. 5.13195 cae ae
262468 ae 4.50888 ee
2.88489 eee ' 4.090604 cae
2.55827 oc... 4.88591 = ......-
8.59114 ee 6.16095 = .....e-
4, 27402 5. 18555 7. 38208 77, 4482
4.69459 ae 8.05444 = ....ee.
2.79574. 3.61030 4, 79630 77.4379
2. 59810 Secs ‘4, 45741 © (Rats
3. 42064 4, 41695 5. 86872 77.4435
4.11855 5. 31786 7.06518 77.4475
8.95958 5. 11289 - 6.79845 77.4431
2.71884 4.66895 = ana “
2. 77391 3.58196 Saami 77. 4379
4, 64435 5.996999 eee "77, 4447

Mean, 77. 4437+. 0009

From the Ag ratio Nd=144. 258.
From the AgCl ratio Nd = 144.268.

MEMOIRS NATIONAL ACADEMY OF SCIENCES.

AgCi ratio.
58, 2915
58. 2825
58. 2889
58. 2894
58. 2887
58. 2925
58, 2857
58, 2895
58. 2872
58. 2860
58, 2936

58. 2876+. 0006

[Von XVI.

Similar determinations are due to Baxter, Whitcomb, and Stewart, whose data are as

follows:

NdCi,. Ag. AgCl, Ag ratio.
3. 34038 © ABIOOF se aipias 77, 4514
4, 03869 5.21445 6. 92786 77.4519
4, 48251 5. 78760 7.69005 77.4503
3. 75204 4, 84458 6. 43738 77, 4482
6. 00909 7. 75892 10. 30876 77.4475
4, 96466 6. 41045 8.51634 77, 4464
4, 15431 5. 36392 7.12743 77, 4492
5.53210 7.14338 9.49134 77, 4437
4. 73995 6. 12054 8. 13232 77. 4433
4, 22430 5. 45444 7, 94714 77,4470
5.93271 7. 66088 10. 17891 77. 4416
5. 60324 7. 23524 9.61299 77.4439
7.22817 9. 33359 12. 40098 77.4426

.. 5, 16230 6. 66603 8. 86147 77.4419
4, 50348 5. 81558 7. 72718 ¥77, 4382
6. 42333 8. 29438 11. 02284 77.4419
7.59834 9.80822 wi... ¥77, 4385

Mean, 77.4461+.0006 58. 2881+. 0008

The starred determinations are rejected by the authors.
From the Ag ratio, Nd= 144.266.
From the AgCl ratio, Nd = 144.270.

Ag Ct ratio.
58. 2963
‘58, 2897
58. 2852
58. 2911
58. 2957
58. 2862
58. 2858
58. 2853
58. 2892

., 58,2844

58. 2883
58. 2871
"58, 2556
58. 2810
#58. 2729

Combining both sets of analyses, namely, of Baxter and Chapin and Baxter, Whitcomb

and Stewart we have—

3Ag: NdCl,:: 100: 77.4449 + .0005
3AgCl: NaCl, :: 100: 58.2878 + .0005

» There are now four ratios for neodymium. Reducing them with

Ag= 107.880 +.00016
Cl = 35.4605 +.00011
S = 32.0650+.00012

we have—
From the sulphate..............02.000000cecee cence Nd=148. 790+. 0054
From the silver ratio. .........0..0.20-eeceeeeeeceene 144. 2624. 0017
From the silver chloride ratio...........00...eeeeeeee 144, 269 +.. 0022
From the oxide......... Ps lo aie Mecleeaucadie ees 144, 5144. 0111
AS eiaeiil Hienii 4p 2s ky cacao Guanes er Nd=144, 2424. 0013

This final value is essentially that found by Baxter and his colleagues.

 

‘Journ. Amer. Chem. Soc., 38, 302, 1916.
No. 3.7 ' ATOMIC WEIGHTS—CLARKE. Mi 397
SAMARIUM.

According to Marignac,! the atomic weight of samarium is 149.4. He gives, however, one
analysis of the octohydrated sulphate, as follows: 1.8515 grammes gave 0.365 of water, and on
calcination lost 0.607 SO,. -Hence, the percentage of Sa,O, in the hydrous salt is 47.502, and in
the anhydrous sulphate 59.166. From these data Sa=149.87 and 150.01. The first regular
series of atomic-weight determinations was by Cleve,? who effected the synthesis of the sulphate
from the oxide. Data as follows:

» Ba03. 8a,(30j)3. ‘Per cent Sa,0,.
1.6735 7 _ 2,8278 _ 59. 180
1.9706 Oe sa 8 8. 3301 59.175
‘1.1122 "a 'gzgi7 9. 201
1.0634 1. 7966 + 59.190

. 8547. 1. 4440 59. 190

. 7447. 1. 2583 59. 183
Mean, 59. 1865-4. 0025

 

Hence, Sa= 150.16.
Another set of determinations by Bettendorff,® after the same general method, gave as
follows:

8,05. Sa,(S0,)3. Per cent Sa,03.
1. 0467 1. 7675 59. 219

1. 0555 : 1. 7818 59. 238
1.0195 1.7210 59. 225 ©

 

Mean, 59. 227+. 0038

Hence, Sa= 150. 45.

In asingle analysis of the hydrous sulphate, Brauner‘ obtained the {lewis figures: 1. 36567
grammes Sa,(SO,),.8H,O gave 1.09770 Sa,(SO,), and 0.65046 Sa,0,. . Per cent Sa,O, in erate
47.629; in anhydrous salt, 59.257. Hence, Sa= 150.76 and 150.67.

Kappel, 5 cited by Muthmann and Weiss, from 4, 12673. grammes Sa, (SO,), obtained 2, 45028
Sa,0,, or 59.376 per cent. Hence, Sa=151. 53, =:

Brill’s® analyses of samarium sulphate, with the aid of the. microbalance, need not be
considered, for his. two’ experiments, as recorded, are widely discordant. The most. thorough
investigation by the sulphate method is that by Urbain and Lacombe,’ whose material was
scrupulously 1 freed from. other rare earths, an impurity to be discussed more fullv a little later.
The samarium preparations of Urbain and Lacombe were derived from different, sources, gado-
linite, monazite sand, etc., and the octohydrated sulphate was analyzed by dehydration and
calcination in the ia way. In the next table I give their weights, and also three per-
centage columns, as follows: A, percentage of Sa,, (SO,)s, in Sa,(SO,)5. 8H,0. B, Sa,O, in the
hydrous sulphate. C, Sa,0, in the anhydrous sulphate.

vn 8H,0. Sa,(S0,)s. Sa,03. A. B. Cc.

1.0499 0. 8435 0.4996 80.341 47.585 59. 229
"4.9898 = (1.0362 .6137 80. 338 47.580 59. 225
1. 3650 1.0969 6497 80. 359 47.597 59. 230
1. 7992 1.4453 . 8557 80. 330 47.560 59. 206
1. 8636 41.4977 . 8873 80. 366 47.605 59. 244
. 8407 ‘ "6749 4001 80.277 47.591 59. 283
2.5107 2.0172 1.1948 "80.344 47.588 59. 228
3.1171 2.5045 1.4840 ~ 80.347 47.608 59. 253
2.9425 | 2. 3635. 1. 4004 80. 323 47.592 59. 251

 

1 Arch, Sci. Phys., Nat. (3), 3, 435. 1880. Oeuvres Complates, 2, 709.

Journ. Chem. Soc., 43, 302. 1883.

3 Amer. Chém. ‘Pharm., 263, 164. 1801 '

4 Abege’s Handbuch, Bd. 8, Abth. 1, p. 284. Preliminary notice in Journ. Chem. Soc., 43, 287. 1883.
6 Liebig’s Annalen, 331,16. 1904. -

6 Zeitsch. anorg. Chem., 47, 464. 1905.

7 Compt. Rend., 138, 1166. 1904.
398 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

 

 

Sa(SO,)38H,O.  Sa(SO,)3. Sa,03. . A B. C.
3.2200 2.5872 1, 5324 80. 348 47.590 59. 230
2. 8382 2, 2804 . 1.3508 80. 347: 47,594 59. 237
Mean, 80.338 47.590 59. 238

+.0048 4-.0026 +.0040
., From A, Sa= 150. 35.

From. B, Sa= 150. 49,
From C, Sa = 150.53.

Two of the ratios given by Urbain and Lacombe’s experiments may now be combined
with former series of determinations. First, for the percentage of Sa,O, in the anhydrous
sulphate, giving the single determinations of Marignac, Brauner, and Kappel each the weight of
one experiment in Urbain and Lacombe’s series:

Marignac.......---00-eeeeecece cece cece reece eteeeeeeees 59.166 +.0199
CHENG soo weuee aed See SAS Ree Gaus aeons eee een ale 59. 1865+ .0025
Bettendorf. ccnccrsscs7esnagatceeenant won cmeine dios 59.227 +.0038
Brauner idacncwase seed kawie o4c bees ones heen ee eeiies 59.257 +.0199
Képpel.........0-2 022 e cece cece renee eed ee ee een cess 59,376 4.0199
Urbain and Lacombe..........--.------0----ee ee eee eee 59.288 -&.0040

General Mean} .ccga5 i iieenennacclaceugemi 59. 2074+.0018

Secondly, for the percentage of Sa,O, in the hydrous sulphate, giving, arbitrarily, the single
determinations of Marignac and Brauner the same weight as in the anhydrous series:

Marignae's ios s2 o0's « iecowiswasisincsinc earn deesemewsieeiviels see ee = 47. 502 +.0199
Brauner: veexies o 3k asarvo cudedaw eed ns ties euicisleeiolomgrye sa o's 47.629+.0199
Urbain and Lacombe-...-.. Spenaerevioste swap sinacueaaaaese 47.590 +. 0026

General mean...../...-.......--5 Pai Ateemainets eben 47.589 +.0025

The value of these combinations is perhaps questionable. The earlier work on the atomic
weight of samariuin is affected by the discovery of europium, which was made by Demarcay.'
According | to this chemist, the original samaria contained admixtures of europia, which tended
to raise its apparent molecular weight. For samarium itself, by the sulphate method, he found
Sa = 147.2 to 148, and for europium, Eu=151. The material studied by Urbain and Lacombe,
however, w was free from europium, and still gave a higher percentage of oxide in sulphate than
the substances examined by the earlier investigators. Their material, therefore, was either free
from europium, or ‘else contained compensating impurities. At all events, the general means
are close to Urbain and Lacombe’s figures, and may be allowed to stand unchanged.

Still another method for measuring the atomic weight of samarium has been proposed by
Matignon,? who found that the normal sulphate, heated to between 500° and 1000°, yielded a
stable basic salt, Sa,SO,. In one determination, 0.7325 gramme Sa,(SO,), gave 0.5335 of
Sa,SO,. Hence Sa= 150.67.

The determinations by Feit and Przibylla,? with their special volumetric method, are as
follows:

80,05. 0. Atomic weight.
0. 5576 0. 07668 150. 522
. 5576 .07670 150. 477
. 5583 .07684 150. 378
. 5633 07747 150. 514

 

Mean, 150. 473+ .0221

More recent determinations of the atomic weight of samarium are by Stewart and
James,‘ who analyzed the chionide: Two series of determinations, with vacuum weights, are
given, as follows:

 

1Compt. Rend., 122, 728. 1900. Ibid., 130, 1185, 1469. Ibid., 132, 1484. 8 Zeitsch. anorg. Chem., 50, 259. 1906.
2Compt. Rend., 141, 1230. ‘Journ. Amer. Chem. Soc., 39, 2605. 1917.
No; 3.] - ‘ “ATOMIC WEIGHTS—CLARKE. 399

 

Series 1.

_ SaCly. Ag. Ratio.
4.05609 5. 10860 79. 3973
3.00065 3, 78044 79. 3730
5. 28891 6. 66424 79. 3625
‘B. 75782 ., 4, 73748 79. 3211
4, 76695 6.00873 79.3337
2.97764 3. 75234 79. 3542
3. 82407 4.81761 79.3769
4.47775 5. 64365 79. 3414
8.33781 4. 20712 79. 3374
; Mean, 79. 3553+ . 0055

Hence, Sa = 150.444.
Series 2.

SaCl, Ag. Ratio.

4. 65174 5. 86128 79. 3639
5. 138023 6.46577 79.3445
4, 52826 5. 70727 79. 3420
6. 65425 8. 38628 79. 3469
6.02908 7.59669 79. 3628
5.09858 6. 42520 . 79. 3528
4. 68220 5.90111 79.3444

4, 32763 4.5551 79.3258
an" % Mean, 79. 3479-4. 0029

 

no. Hence, Sa = 150.422.
The general mean of the two series combined i is 79.3497 +.0026.
In all there are six ratios relative to the atomic weight of samarium:
(1) Saq(S0,)3.8H,0 : Sa,(SO,)s : : 100 : 80.338-+.0048
(2) Sa,(SO,)3-8H,O : Sa,0, : : 100 : 47.589 4.0025
(3) Sa,(SO,)3 : Sa,O, : : 100 : 59.2074+ .0018
(4) Sa,(SO,), : SagSOg : : 100 : 72.883
(5) O: Sa: 216 : 150.4734 :0221
(6) 3Ag : SaCl, : : 100 : 79.3497+.0026
To reduce these we have— .

Ag=107. 880 +. 00016 S=32. 0650+. 00012
Cl= 35.4605 +. 00011 H= 1. 0078+. 00001

Hence, giving to ratio 4 the arbitrary weight represented by +.0075:

From ratio—, ae oe Po

Dae cciUaaosteeasde Binary de ana laaatas Sa=150. 312+. 0095
2k ee seee Rice aetna asthe pase eeiccee eee sane 150, 346+. 0720
Bier eerie Ate cis we See eae delves San ayia 150. 426-L. 0085
HF crab ceratios pais isis winlemne sc sie ee ne Bieta leteeareente nates 150. 473+. 0221
Deca dca pecs mallee ed dieses nna snaidaecedele sen 160. 481+. 0124
Marius Necaacrcceociouns Seleig aasie ea wake Coasts 150. 666+. 0750

General mean............0220ececeeeeeeees Sa=150. 403+. 0054

This mean may be a little too low, but'it ‘is near that which is derived from the work of
Stewart and J ames. ne
ny ADDENDUM.
Since the manuscript of this volume was sent to the printer Owens, Balke, and Kremers!
have published important work on the atomic weight of samarium. By analyses of the anhy-
drous chloride they obtained the following results:

1 Journ. Amer. Chem. Soc., 42, 515. 1920,
161566°—-20-—-26
400 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.

SaCl;. Ag. At.wt.Sa.
6. 47502 8. 15810 150. 49
4. 03877 5. 08083 150.90 Rejected.
' 4.25295 5, 35928 150. 45
4, 34537 'B. 47663 150. 41
« 3. 84046 4, 84002 ‘150. 42
38. 32317 4, 18820 150. 42
2. 96369 3.73504 150. 42
8. 34818 4, 21936 150. 44
3. 27999 4, 13383 150. 44
38. 00394 5. 78769 ’ 150. 40
2. 36481 2: 97970 150, 47
. 3. 68616. 4. 64577 150. 41
3.17435 4, 00020 150. 44
2. 81639 3. 54932. 150. 43
3. 18555 8. 95155 150. 43
2.91019 3. 66744 150. 44
4, 15775 -5. 23973 150. 43
3. 32505 4.18972 150. 47
2. 06632 62950 150. 43
Mean, 150. 43

All weights reduced to a vacuum. Determination No. 2 rejected because of impurity in the
SaCl,. The determinations by Stewart and James are completely confirmed.

EUROPIUM.

Demargay,' the ee of europium, found for its atomic weight the approximate num-
ber 151. The first detailed determinations, however, were those of Urbain and Lacombe,?
who analyzed the octohydrated sulphate. I give their weights, and three percentage columns,
as follows: A, Eu,(SO,), in Eu,(SO,),.8H,O. B, Eu,O, in Eu,(SO,),.8H,O. C, Eu,O,; in
Eu,(SO,),. ; fe

go 2

Euy(SO,);.8H,0. Evz(S0,)s. Eu,0;. A. B. C.
1.7787 1. 4303 0. 8500 80. 413 47. 788 59. 428
2, 4785 1. 9935 1. 1848 80. 432 47, 803 59, 433
2.4177 1. 9449 1. 1554 80. 444 47.789 59. 407
2, 4831 1, 9968 1. 1870 80. 416 47, 803 59. 445,
2. 2988 1, 8488 1. 0990 80. 425 47. 807 59. 444
Mean, 80.426. 47. 798 59.431

+. 0038 +. 0027 +. 0047

'-» Hence, Eu=151. 99 151. 97 151.98

Jantsch * determined the atomic weight of europium by calcination of the hydrous sulphate:

Suiphate. Eu,03. Per cent Eu,03.
1. 3501 0. 6455 47.811
1. 5054 . 7197 47. 808
1. 5213 . 7274 47. 814
1. 2881 . 6159 47. 815

Mean, 47.812 +. 0010
Hence, Eu = 152.05.
_, ; Urbain and Lacombe, for. the same ratio, found 47.798+.0027. The two series combined
give 47.810 + .00094.
Feit and Przibylla‘ have also applied their volumetric method to the determination of
this atomic weight, with the following results: 2
. B03. . x, O08 Atomic weight.

0. 3961 0. 05385 152. 535
- 4096 . 05566 ‘152. 615
- 4115 *, 05594 152. 546

Mean, 152. 565+. 0170

 

1 Compt. Rend., 132, 1484. 1900. 8’ Compt. Rend., 146, 473. 1908.
1Compt. Rend., 138, 627. 1904. « Zeitsch. anorg. Chem., 50, 260. 1906.
No, 3.) » ' ' ATOMIC WEIGHTS—CLARKE.. 401

-The four ratios for Eu now are— .
(1) Eu,(SO,),.8H,O : Eu,(SO,); :: 100 : 80.426 + .0038
(2) Eu,(SO,),:8H;0 : Eu,0, :: 100 : 47.810 + .00094
(3) Eu,(SO,), : Eu, : : 100 : 59.431 + .0047
(4) 0: Eu ::,16 : 152. 565-+.0170

Reducing these ratios with $=82. 0650 + .00012 and H = 1.0078 +.00001 we have:

-From ratio—

BT hehe a cate gOS LU CL cn ehh) 2 Eu=151. 935-. 0247
TL poeaseetl anus dees adeh ccd (ae wae cseaaee 151. 997+. 0542
es gy onan Sta oat tears "152, 033-4. 0048
Renee ceed bveduinnns «oo ouimeae Badiaha cn eyced ee 152. 565-4. 0170:

General mean..........--0-2--cece-c2-2-. Bu=162, 066-4. 0044

In round numbers the atomic weight of europium is 152.
— GADOLINIUM.

Marignac,! the discoverer of gadolinium, assigned to its oxide the ‘“‘equivalent” 120.5,
whence, Gd=156.75.. Boisbaudran? found. Gd=155.33, 156.06, 155.76, and 156.12, with
preference for the last:figure. Cleve, quoted: by Boisbaudran, found Gd= 154.15, 155.28, 155.1,
and 154.77. For these determinations there are no details, and all, probably, are referred to
SO, = 80.

The first chemist to publish his determinations with individual data was Bettendorff.*
He effected the synthesis of the sulphate from the oxide, and his weights were as follows. The
percentage of Gd, Os in Gd 2(SO,)s 1 is given in the third column:

Gd,0>. 'Gd,(SO,)s:, Per cent Gd,05.
1. 0682 1.7779 60.082
1. 0580 1.7611 : 60.076 >

 

1. 0796 1. 7969 60. 081

Mean, 60. 080+. 0013
Hence, Gd= 156.75.
Benedicks’‘ series of determinations were also by the synthetic process, as follows:

Gd,03. Gd,(S0,)3. Per cent Gd,03.
0. 4808 0.7171 60. 075
. 9675 Pe Oa - 9451: 60. 047
. 8726 . 9534 60. 059
6785 1, 1301 - 60. 039
. 7399 : 1. 2329 60. 013

a sets . 2, 2063 60. 069

Mean, 60. Obl 0020
Hence, Gd= 156. 52. 4
The two determinations by Marc ® are unimportant, but cannot be ererleckod: The

data are— ates oy sete ne

Gd,03. _ Gd, (S0,)s. . Per c cent it GOs.
0. 2201 eS” 9.3666 5 = "60: 014”

. 2444 4070. 960. 049 -

—_—_—

‘Mean, 60. 0322. ‘0120
Hence,.Gd= 156.389... 0 en ha >
Brauner,’ in a amgis experiment, found 0.88884 Gd,0, =] A8957 Ga, (50,),=1. 83903
Gd,(SO,),.8H,O. . Per cent. Gd,O, in hydrous sulphate, 48.332; in “anhydrous salt, 59.951.

Hance Gd = 155.753 and 155.78.

1 Oeuvres Gonapibias, 2,704. 4 Zeitsch. anorg. Chem., 22, 393. 1899.
2 Compt. Rend., 111, 409. 1890. 5 Zeitsch. anorg. Chem.,.38,121. 1904.
_ 8 Ann. Chem. Pharm., 270, 376. 1892. 6 Abegg’s Handbuch, Bd. 3, Abth. 1, p. 304,
402 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou, XVI.

The material studied by Brauner was received from Cleve, and was not perfectly pure.
The atomic weight found is too,low. In combining the figures for the percentage of Gd,O, in
Gd,(SO,),, Brauner’s determination may be given. oe weight with that of Marc. We fave
then— fe ‘

Bettendorff............-.- Adeseadenctdal eas as eheaeiet 60, 080+. 0013
Benedicks..............00.ceeeeeeeee eee Sistas qaniens 60. 050+: 0020
Marc.......... Pada mph cyt ahilsw ope Sleeps Sia dragons eciesecnrs 60. 032+. 0120
BIRQUNE! s36c.525. 29s jucmatramoweaewaweuuecaeeccea eet cones 59. 951+. 0120

General mean... .... 2... eee ne cece ares eee cnecrecnces 60. 070+. 0011

The purest gadolinium preparations were probably those studied by Urbain,' who calcined
the octohydrated: sulphate to oxide. Two series of determinations are given, representing
different groups of fractions obtained in the purification of his material. The data are as follows:

 

Series I.

Gd,(SO,)5-8H,0. Gd,0,. Per cent Gd,O».
1. 9256 0. 9350 48. 557

_ 49749 . 9589 48, 555
1. 9975 . ++ 9698 48. 551
2. 1083 ~ 1.0281 48. 528
1, 8993 . 9214 48. 514
2. 2065 : 1. 0707 48. 525
1. 9535 . 9479 48. 524

. 2, 2008 1. 0685 48, 551
2. 2482 1. 0914 48. 546

“ "2.1982 1. 0646 48. 541
Mean, 48. 539-+.0033
Series HT.

Gd,S0,)3.8H,0. Gd,0,. Per cent Gd,03.
2. 0551 0. 9974 48. 534
2.1555 1. 0469 48. 570
2. 2277 1. 0867 48. 512
2. 2559 1. 0946 48. 529
2. 2523 1. 0939 48. 569

 

Mean, 48. 543+. 0077

The weighted mean of both series is nearly 48.540+.0030. Hence, Gd=157.24. Brauner’s
single determination may be neglected.

There are also two determinations made by Feit and Przibylla,? with their volumetric
method:

Gd,0;. 0. Atomic weight.
0. 8852 0. 05097 157. 877

. 3956 . 05234 157. 398
Mean, 157. 388 4-. 0067

 

The three ratios for gadolinium are:
(1) Gd,(SO,),.8H,O : Gd,O, :: 100 : 48.540+.0030
(2) Gd,(SO,), :-Gd,O, :: 100 : 60.070+.0011
(3) O :.Gd :: 16 : 157.388 + .0067

Reducing these ratios with S=32.065 + 00012 and H= 1.0078 +.00001 we have:

| From ratio—

DY acu coesdsuadnddesbeasia ee teddssoiesacrdiaay Gd 158, 6784: 0069
ft ele ee ies tach ae ak oa le ok Nan ee casted 157, 256-4..0154
Ductal nse: 8 etranetaintne SOO eee PII Iau d 157. 388+. 0067

General mean. .............0...000ee eee Gd=157. 008+. 0044

1 Compt. Rend., 140, 583. 1905. 4 Zeitsch. anorg. Chem., 50, 260. 10906.
No. 8.] ATOMIC WEIGHTS—CLARKE. 403

TERBIUM.

The older determinations of atomic weight made upon terbium preparations of doubtful
character may well be ignored. Boisbaudran' has published two estimates of this constant.
First, for two preparations, one with a lighter and one with a darker earth, he gives Tb = 161.4
and 163.1. In his second paper he makes Tb=159.01 to 159.95; probably with SO,=80.
According to Feit? Tb=158.6. Emma Potratz,’ by various methods, found Tb=154, approx-
imately. For all of these determinations the essential details are lacking.

The series of determinations by Urbain‘ is more satisfactory. The octohydrated sulphate
was converted into the anhydrous salt by careful heating, with the following results:

 

Tb,(SO,)3.8H,0. Tb(SO;) 5. Per cent H,0.
2. 0407 1, 6489 19. 199
1, 9626 1. 5859 19. 194
2, 2580 1, 8245 19. 198
2. 2385 1: 8087 "19, 201
2.0037 - 1.6190 19. 200
Mean, 19. 198+. 0008
Hence, Th = 159.205 + .0130. = po
DYSPROSIUM. ou ig oath

The atomic weight of dysprosium was first determined by Urbain and Demenitroux.' They
reduced the octohydrated sulphate to oxide, by calcination, with the following results, taking
all their data as one series:

Dy,(SO,)3.8H,D. Dy05. Per cent Dy,O,.
1. 6966 0. 8359 49.269
2. 0926 1. 0301 49. 226
1. 8415 . 9069 49, 248
1.5519 . 7649 49, 288
2. 4955 1. 2296 49. 273
1, 8130 . 8927 49. 238
1. 8817 9271 49. 269

.1,1164 .5500 49. 266
1. 7308 . 8528 49, 272
2. 6038 1. 2820 49. 236
1, 6942 . 8346 49. 262
2.1776 1, 0726 ., 49. 256

 

; Mean, 49. 259+. 0086
Hence, Dy = 162.547 +.0190. . oe ae
Engle and Balke* determined the atomic weight of dysprosium by the conversion of the

oxide into the chloride. Their method has already been described under yttrium. The data,

with vacuum weights, are as follows: a. oo ght

hf age

Dy,03. _ DyCl,. * Ratio.
0. 50996 ‘0.73300 143. 7367
1, 05741 1.51998 .4 143, 7361
. 65617 . 94352 143. 7920
1, 22603 1. 76297 "143. 7950
1, 96935 2, 83135 143. 7707

 

Mean, 143. 7661-. 0061
Hence, Dy = 164.23. a

1 Compt. Rend., 102, 396, and 111, 474. 1886-1890.

2 Zeitsch. anorg. Chem., 43, 280. 1905. oo oe

Chem. News, 92, 3. 1905. : re ee ‘

«Compt. Rend., 142, 957. Preliminary papers by Urbain are in C. R. 141, 521; ibid., 142, 727; Bull. Sac. Chim. (3), 33, 403. See also Journ,
Chim. Phys., 4,321.  ' '

6 Compt. Rend., 143, 598. 1906,

¢Journ. Amer. Chem. Soc., 39, 67. 1917.

4
404: MEMOIRS NATIONAL ACADEMY. OF SCIENCES. [Vou XVI,

The same method was also followed by Kremers, Hopkins, and Balke,’ whose data, with
vacuum weights, are as follows:

Dy-Os- Dy Ch, Ratio.

0, 72974. 1.04979, -, 148, 859
1, 36017 _ 1. 95604 "143; 808 |

. 69716 1. 00255 148, 804
1, 02493 ° 1. 47475 148, 887

. 87688 °: 1. 26253 143, 979

. 69318 . 99677 _ 148,796 ..
1.55248 2, 23374 ; 148, 882

. 83958 1. 20763 © ; 143, 837

1. besst 2, 25129 143, 869
Mean, 143, 858. 0128

 

Hence Dy = 163. 83...

Combining this series with that of Engle and Balke, the general. mean is 143.782 + .0056
The discordance between. the two series of determinations indicates that the method is not
satisfactory. 53

Kremers, Hippie and ingle also determined the ratio. between silver and dysprosium

chloride, with the following results: 9) ).-":

Dy Ch. , Ag. . .., Ratio,

1. 04079 ~ <1, 26301 ” 120, 309
1. 95604 '- "9, 853880 120. 384
1. 47475 1. 77504 120; 362
1, 26253, 1, 51991 120. 386

. 99677, 1, 20083 120, 421
2, 23374. 2. 68806 120. 338
1, 20763: 1, 45325 120. 339

2, 25129. 2.70992 120, 371
Mean, 120, 358+. 0084

 

Hence Dy = 162.517.

There are now three ratios from which to compute the atomic weight of dysprosium.
Reducing them with Ag=107.880+.00016, . Cl= 35.4605+.00011, S=32.0650+ .00012, and
H=1.0078+.00001, we have—

From the silver ratio. .........-.--. nearest. Dy=162. 517+. 0236
From the sulphate. ....-...-.-----2. 222s ee eee eee 162,547 +. 0190
ARFOIE the Ox106 5 20's oo 2g cipisionsitia qearpiararergratarateretecoriveigee 164, 162+. 0236

General mean aah ctesseasereeseteseessee-+Dy==162, 987+: 0125

"The paid ratio, however, i is ee ertcomin The other two ratios, which are in fair agree-
ment, show that the atomic weight of dysprosium is somewhere near 162. .5, which, value may

be for the present adopted.
HOLMIUM.

For the atomic weight of holmium there is a series of syntheses of the sulphate from the
oxide by Holmberg.’ The data are as follows:

' Ho,03. Hox(80,)y- Per cent-Ho,0x.
0. 3467 0. 5687 60. 964
.3400 . 5579 60. 943
. 3960 . 6496 60. 961
. 7631 1, 2524 60. 930
. 6877 1, 1286 60. 934

‘ ae . 8822 60. 961

nee "Mean, 60. 949-4. 0041
Hence Ho = 163.443 + .0254. a

1 Journ. Amer. Chem. 8o0., 40, 698, 1918, 2 Zeltsch. anorg. Chem., 71, 226. 1911.

 

 
No. 3.] ATOMIC WEIGHTS—CLARKE. 405
ERBIUM, THULIUM.

Since the earth which was formerly regarded as the oxide of erbium is now known to be a
mixture of two or three different oxides the older determinations of its molecular weight have
little more than historical interest. Nevertheless the work done by several investigators may
properly be cited, if only for the sake of completeness.

First, Delafontaine’s! early investigations may be considered. Assulphate, regarded as
erbium sulphate, gave the following data, An oxalate was thrown down from it, which, upon
ignition, gave oxide. The percentages in the fourth column refer to the anhydrous sulphate.
Tn the last analysis water was not estimated, and I assume for its water the mean percentage
of the four preceding experiments:

Sulphate. Er03. #0. Per cent Er,0y.
0. 827 0. 353 0.177 54, 308
1. 0485 - 4475 . 226 54, 407
. 803 . 3415 -171 54, 035
1. 232 . 523 - 264 54, 028
1, 1505 - 495 stats 54, 760

Hence Er= 118.75. Mean, 54. 308+. 0915

Bahr and Bunsen? give a series of results, representing successive purifications of the earth
which was studied. The final result, obtained by the conversion of oxide into sulphate, was as
follows:

, -7880 grm. oxide gave 1.2765 grm. sulphate. 61.653 per cent oxide.

Hence Er=169.09. |

Hoeglund,? following the method of Bahr and Bunsen, gives these figures:

Er,03. Er(S0,)3- Per cent Er03.
1. 8760 3. 0360 61. 792
1, 7990 2. 9100 61, 821
2. 8410 4. 5935 61. 848
1, 2850 2.0775 61. 853
1. 1300 1, 827 61. 850
. 8475 1, 370 61. 861

Hence, Er=170.60. Mean, 61. 8375+. 0063
7

. According to Thalén,* apectroanonic avideies shows that the “erbia” studied by Hoeglund.
was largely ytterbia. ,
Humpidge and Burney ° give data as follows:
1. 9596 grm. Er,(SO,)s gave : 2147 grm. Er,0,. 61, 987 per cent.
1. 9011 1.1781“ 61.965
, Mean, 61. 1 976. 0074 |
Hence, Er=171.75. ;
The foregoing data were. all. published before the composite nature of the supposed erbia
was fully recognized. It will be seen, however, that three sets of results were fairly comparable,
while Delafontaine evidently studied an earth widely different from that investigated by the
others. Since the discovery of ytterbium, some light has been thrown on the matter. The
old erbia is a mixture of several earths, to one of which, a rose-colored body, the name erbia
is now restricted. For the atomic weight of, the supposedly true erbium Cleve ° gives three
determinations, based on syntheses of the sulphate after abe usual method., His weights were
as follows, with the percentage ratio added: .

Er,03- Er (S0,)s. Per cent Bry,
1, 0692 1, 7436 61, 321
J, 2153 1, 9820 61. 317

. 7850 1, 2808 61, 290

Hence, Er=166.30. Mean, 61. B09. 0068

1 Ann. Chem. Pharm., 134,108. 1865. “ ' ' 4 Wiedemann’s Beiblatter, 5,122. 1881.
2 Ann. Chem. Pharm., 137, 21. 1866. 6 Journ, Chem. Soc., 35, 116. 1879.
8K. Svensk. Vet. Akad. Handlingar, Bd. 1, No. 6. 6K. Svensk. Vet. Akad. Handlingar, No.7. 1880.

 
406. MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von, XVI.

The discussion over the complexity of erbia, however, did not stop with the work of
Cleve. Kriiss,' assisted by K. Hofmann, made a long series of fractionations of erbium ma-
terial, and gave crude atomic weight determinations of them, which varied’ widely. | ‘The
figures need not be reproduced here. L. Hermann? also has sttidied the subject, and states
that the old erbia is separable into two earths, one giving red and the other yellow salts.

‘More recent determinations of the atomic weight of erbium are as follows. First, two
unimportant analyses made by Brill * with the aid of the microbalance:

 Br(SO,)a. Er.Qy. _ Per cent Er,0,.
92. 35 56, 5B * 61, 234
36,75 22, 68 61. 496

Hence, Er= 166.75. Mean, 61. 365+. 0873 ge eGR Sr I

Under the name “neo-erbium” Hofmann and Burger * describe carefully purified material,
which gave a sharp and distinct spectrum. Four syntheses of the sulphate gave the subjoined

figures:

Er,0, Er,(S0,)s. Per cent Er,0,.

0. 9048 1, 4724 61. 451

-, 4666 - 7594 61, 443
1.4181 2. 3077 61. 451.

1.0789 ~ 1, 7563 es , 61, 436

“ Hence, Er = 167.40. Mean, 61.445+.0031 ....

Later determinations by Hofmann * involved both syntheses and analyses. The oxide
was converted into sulphate, which was afterwards calcined to oxide again. The weights
are as follows, giving six values for the Pereehanse. of oxide i in the sulphate. :

‘Er,0; Er,(SO,)s. BO, Per cent Er,Oy.
0.3117 0.5070 0.3117 61. 481
. 4486 . 7296 . 44863 -. 61.481
. 3718 . 6048 . 3718 61. 486
61. 490
61.475
' 61.475

Hence, Er (neo-erbium) = 167.69. Be See

Nearly all of the foregoing determinations can be neglected, as being of no present value.
Three of them, however, may be combined as follows, giving the percentage of erbia in tHe
sulphate:

Cl6V Ou seesndecawesee aces eees secon gs do siastydansees G1, 800: 0068
Hofmann and Burger... ,- -------+---+++++-++++- sceceses 61 4454. 0081
Hofmann. ......... ispeGies aeeasee bs erussvradaubecapacacas Ol; 481+. 0010

General mean........2.-2-+2++: rly Sete oa a SSO 61. 475+. 0009

Hence, Er= 167.641 +.0057. More data on ‘‘neo-erbium” are nedeed.
The atomic weight of thulium has not yet been carefully determined. Cleve * assigned
to'it the atomic weight 170.7, but without details as to weigiinee According .to Urbain,’

the atomic weight is below 168.5.
ADDENDUM.

Since these computations were completed, additional work on erbium has been publiched
by Wichers, Hopkins and Balke.® Their determinations were made by the oxide-chloride
method, and gave values for Er ranging from 178.00 to 168.84. These values are too divergent
to warrant further consideration. The method itself is defective, and the samples of erbium
oxide were probably not free from other rare cere:

1 Zeitsch. anorg. Chem., 3, 353. 1893.

2 Dissertation, ‘Technische Hochschule, Munchen, 1906.

3 Zeitsch. anorg. Chem., 47, 464.' 1905.

4 Ber. Deutsch. Chem. Ges., 41, 308. 1908, ,

5 Ber. Deutsch. Chem.Ges., 43, 2635. 1910.

€ Compt. Rend., 91, 329. 1880,

7Compt. Rend., 145, 759. 1907. According to Auer von Welsbach (Anzeiger Wien. Akad., 45, 529, thulium is really complex. The atomic
weight assigned to it has little significance.

® Journ. Amer. Chem.8oc., 40, 1615. 1918.

 
No. 3.) ATOMIC WEIGHTS—CLARKE. 407

YTTERBIUM AND LUTECIUM.

Although ytterbium was long supposed to be a definite element, is has recently been shown
to be complex, and its oxide is a mixture of at least two distinct earths. Nevertheless, the data
relative to the atomic weight of the old ytterbium are worth assembling, if only for historical
reference,

The first good series of determinations was by Nilson,1 who effected the synthesis of the
sulphate from the oxide in the usual manner., His figures are as follows:

 YO,03. Yb(S0,)3- Per cent Y6,03.
1, 0063 1, 6186 62. 171
1. 0139 1. 6314 62. 149
- 8509 1. 3690 62. 155
. 7371 1.1861: 62. 145
1. 0005 1. 6099 62. 147
. 8090 1. 3022 62. 126
1. 0059 1. 6189 62. 134

 

Mean, 62. 147+. 0036
Hence, Yb=173.18. a
Astrid Cleve,? by the same method, obtained the subjoined results:

¥b,05. Yb,(SO,\. Per cent Yb.Oy.
0.7791 1, 2535 62, 154
.5190 . 8353 62, 133
. 4905 7894 62. 136

 

¥ Mean, 62. 141+. 0044
Hence, Yb=173.13.

Brill, who used the microbalance, gives the following figures: anes
Per cent Yb,05.

Yb(SO)» Yb,05.
106. 00 65. 90 62. 170
92, 35 57.30 62. 047

 

Mean, 62. 108+. 0407
Hence, Yb = 172.85. a
Brauner, from 1.67279 grammes Yb,O,, obtained 2.69209 of Yb,(SO,),. Percentage
Yb,O,, 62.137, and Yb=173.10. There is also a preliminary note by G. and E. Urbain,® who
found Yb=172.6, but who give no details of the determination.
Feit and Przibylla,® by their volumetric method, obtained the following figures:

Yd,0,- 0. Atomic weight Yb.
0. 6424 0. 07808 173. 459

. 6408 - 07783 173. 600

- 6403 - 07779 173. 547

. 6466 . 07858 173. 485

ee

I
Mean, 173. 523+.0214 _

The complexity of the old ytterbium was proved almost simultaneously by Auer von
Welsbach and Urbain.’ Urbain, by a long series of fractionations, obtained from it two end
products, one, neo-ytterbium, with an atomic weight near 170; the other, dutecitum, approxi-
mately 174. Welsbach,* whose work was published a little later, proposed for his earths the
names aldebaranium and -cassiopeiwm, and gave more explicit figures as to their atomic weights.
His data are as follows:. °° : 4

AdjO. Ad{SO,)» Per cent Ad,Oy.

0. 4181 0. 6730 62. 125

5984. . 9634 62.113
+6178 - .9989 62. 109

Mean, 62.116+. 0011

5Compt. Rend., 132, 136.

_ 6 Zeitsch. anorg. Chem., 60, 261. 1906.
7™Compt. Rend., 145, 759. 1907. 146, 406. 1908,
8 Monats. Chem., 29, 192, 1908.

«Compt. Rend., 91, 56. 1880. Ber. Deutsch. chem. Ges., 13, 1430.
4 Zeitsch. anorg. Chem., 32, 129, 1902,

8 Zeitsch. anorg. Chem., 47, 464. 1905.

4 Abege’s Handbuch, 3 (1), 335.
408 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Vou. XVI.
Hence, Ad = 172.92.

a Cp,03. Cp(S0,)5- Per cent Cp,0,.
; 0. 3716 0. 5967 62. 276
. 3086 - 4956 62. 268
. 4026 . 6465 62. 274

 

Mean, 62. 273+. 0017
Hence, Cp = 174.24.

In a later paper,! Welsbach ‘gives additional determinations of the two atomic weights.
In this case, however, he analyzed the octohydrated sulphates, with the following results:

 

Cassiopeium.
Cp,(S0,)3.8H,0. Cp.03. Per cent Cp,03.
2, 0428 1, 0394 50. 881
2. 0197 1. 0274 50. 869
2. 4044 1. 2232 50. 873
Mean, 50. 874+. 0024
Hence, Cp = 174.99. i
Aldebaranium.
Ad;(SO,)3.8H,0. Ad,O;. Per cent Ad,O,.
2. 0530 1. 0392 50. 619
2. 0891 1. 0575 50. 620
2.1155 1. 0710 50. 631

Mean, 50. 623+. 0027
Hence, Ad =173.009.

A preliminary notice on the atomic weight of “neo-ytterbium,”’ by Blumenfeld and Urbain,
gave merely the final result of the determination, but without details. These, however, have
been sent me by Professor Urbain in a personal communication. They are as follows:

Yb,(SO,),.8H,0. Yb,0;. . Per cent ¥b,03.
0. 40073 0. 20311 50. 685
. 46594 23621 50. 695
. 95456 . 48388 50. 691
. 59774 30292 . - 50. 678
. 42429 _ . 21502 50. 678
1. 07123 . 54302 50. 691
. 48222. . 24446 50. 695
. 59187 . 30006 50. 697
. 65113 . 33008 50. 693
. 89289 . 45262 50. 692:
«47366 . 24010 50. 690
2. 04620 1. 03710 50. 684
1. 96940 . 99840 50. 696

Mean, 50. 690-+. 0012
Hence, Yb = 173.56.

None of. these determinations can be regarded as final. Nor is it necessary to enter:here
into the controversy between Welsbach and Urbain relative to priority. In the matter of
nomenclature alone, the priority of Urbain is clear.*

His name lutecium is therefore accepted, with Lu=175. For the other component of the
mixed earths the original name ytterbium would seem to be preferable to neoytterbium, and
Yb=173. These round numbers are perhaps as probable as any others which might be chosen.

THE HELIUM-ARGON GROUP.

The five inert gases, helium, neon, argon, krypton, and xenon, are apparently incapable
of forming compounds. Their atomic weights, therefore, can only be inferred from their
densities, for which the following data are available.

 

‘1 Monats, Chem:, 34, 1724. 1913.
2 Compt. Rend., 159, 323. 1914.
8 For a reclamation of priority, see Welsbach, Monatsh. Chem., 38, 695.
No. 3.] ATOMIC WEIGHIS—CLARKE. 409

For helium the earliest determinations by Ramsay‘ are too high. He obtained values
ranging from 3.89 to 4.84, and later figures above 2.13, when the density of oxygen as the stand-
ard is put at 16. Langlet,? a little later, assigned to helium the density 2.00 Ramsay and
hairs * after the discovery of neoh, krypton, and xenon, with purer helium, found its density
to be 1.98. i

More recent determinations of the density of helium by Watson,‘ Heuse,’ and Taylor *®
give the weight of the normal. liter of helium as follows:

S Two determinations by Watson give the weight as 0.17830 and 0.17814 gramme, whence
e=3.994,

Heuse gives seven determinations in mean 0.1785 gramme. This, reduced with the corre-
sponding weight of oxygen, 1.4290 grammes, and from the measured compressibility of helium,
gives He=4.002.

Taylor, using a microbalance, and by direct comparison with oxygen and hydrogen, found
He=4.0008+.0005. Twelve determinations were made and reduced by means of the com-
pressibility ofthelium. The atomic weight of helium then may be taken as He=4, without
serious danger of error. The older determinations give slightly lower values.

- The density of argon has been carefully determined first by Ramsay ’ and by Rayleigh. :

Compared with an equal volume of oxygen weighing 2.62760 grammes, Rayleigh found for
argon the following weights:

3.2710 3.2750

3.2617 3. 2748
3.2727 3. 2741
8.2652

Rayleigh accepts the last three determinations, which give 3.27463 inmean. With O= 16,
the density of argon becomes 19.9399 + .0012.
Ramsay’s figures, also referred to O = 16, are as follows:

19. 904 19. 959
19. 823 19. 969
19. 816 19. 932

Rej ecting the sedond and third of these determinations the four remaining values give in
mean a. density of 19.941 + .0099. Correcting the figures for argon by the method of limiting
densities, D. Berthelot ° assigns to argon the density 19.941.

Fischer and Hachnel % found for the density of argon as referred to O=16 the following

values:

19.93 19. 95
19. 94 19. 94
19. 95 19. 98
19. 97 a
‘ Mean, 19.951

A year later Fischer and Froboese™ purified argon by many fractionations and obtained
19.95 as their final value. Hence, A=39.90. There is also a memoir by Schultze, who deter-
mined the density of argon and from it found A=39.945. Still more recently Leduc,* from
measurements of the density and compressibility, finds A=39.91. The average of all the

values for argon is near A=39.90.

1 Proce. Roy. Soc., 58, 81. 1895. Journ. Chem. Soc., 57, 684. 1895.
4Zeitsch. anorg. Chem., 10, 289. 1895.

3 Proc. Roy. Soc., 67,329. 1900. Phil. Trans., 197; A 47. 1900,

4 Journ, Chem. Soc., 97, 810. 1910.

6 Verhandl. Deutsch. physikal. Ges., 15,518. 1913. Heuse also cites determinations by Olzewski, Schierloh, and Kamerliiigh-Onnes,
6 Phys, Review, 10, 617. 1917.

1 Phil. Trans., 186, 238, 1895.

8 Proc. Roy. Soc., 59, 201. 1896.

9 Compt. Rend., 126, 1501.
10 Ber. Deutsch, Chem. Ges., 43, 1435. 1910,
11 Thid., 44,92. 1911,
13 Annalen d. Physik. (4) 48, 269. 1915.
13 Compt. Rend., 167, 70. 1918.
410 MEMOIRS NATIONAL ACADEMY OF SCIENCES. [Von.. XVI.

The densities of three other inert aon were first determined by Ramsay and Travers. 1
Referred to O= 16, the figures become

Ne people eae eocess naiiawglaaase neee 9.99 9:94

KEE cece RRR Ce Saoenmeite 40. 88 40. 78:
ROSSA SS hctgsciiad bate Adder nA AL Reape 64. 00 63. 64

“They also assign to argon the densities of 19.93 and 19.96. :
More recent determinations of the density of neon are by Watson? and by Leduc For
the weight of the normal liter, in grammes, Watson obtained the following values:

0. 9002 0. 8997
. 9004 ‘ , 8997
. 9006 .9001
. 9005 9001
. 9002 - 9002

. 8998

The corrected mean is 0.9002. Hence, by the method of limiting densities, ‘Watson found
Ne= 20.200.

According to Leduc the density of neon, referred to air as unt) is 0.695. Hence Ne=20.15,
reduced by means of the’ compressibility’ of the gas.

Ladenburg and Kriigel,* in their determinations of the Asides of krypton, obtained erro-
neous values, namely, 29.335 and 29.405. Ramsay,® in still later experiments, found for Kr
the densities 40.81, 40.82, and 40.73, in agreement with the figures given by Ramsay and Travers.

More conclusive data for krypton and xenon are given by Moore,* who worked with residues
separated by fractionation from 120 tons of liquid air. For krypton the densities found were
41.504 and 41.509, or 41.506'in mean. For xenon Moore found the densities 65.380 and 65.328,
in mean, 65.354. Moore’s figures have since been reduced by Watson’ by the method of
limiting densities, The values thus obtained for the two atomic weights are— :

Kr= 82.92
Xe= 130. 22

For the gaseous emanation of radium, niton, Ramsay and Gray,* who weighed very minute
quantities on a microbalance, found Nt=220. In a later paper Gray and Ramsay® report
values ranging from 218 to 227. They regarded 222,4 as the most probable value. Debierne”
by a different method found Nt=220. Theoretically, the true atomic weight. of niton should
be that of radium, 226, minus 4, the weight of one atom of helium—that is, Nt=222.

For ordinary uses the following values can be used for this group of elements:

 

He= 4.0 Kr= 82.9

Ne=20. 2 Xe=130. 2

A =39.9-- Nt=222. 0
2 Proc. Roy. Soc., 67,329, Phil. Trans., 197; r 47. 1900. ; og 6 Journ. Chem. Soe., 93, 2181. 1908.
3 Journ. ‘Chem. 'Soc., 97, 810. 1900. ‘ 7 Journ. Chem. Soc., 97, 833. 1910.
Compt. Rend., 158, 864. 1914. ‘ * Compt. Rend., 151, 126. 1910.

‘Chem News, 81, 205, From Sitzungsb. Berlin Akad., 1900, 212, * Proc. Roy. Soo., 84 A, 536, 1911. ‘
* Proc. Roy. Soc., 71, 421, 1903. oe 10 Gant: Rend., 150, 1740. 1916,

fa
No. 3.]

ATOMIC WEIGHTS—CLARKE.
TABLE OF ATOMIC WEIGHTS.

411

In the following table the results of the foregoing calculations are brought together, each
atomic weight being rounded off to five significant figures, or sometimes fewer. For conven-
ience the values are given in two columns, referring to the two ultimate standards, O=16 and
H=1. Many chemists prefer the latter, and their wishes are, in a work like this, entitled to

respectful consideration:

O=16.
Aluminum............... 27.039
Antimony. ..........-.... 120. 06
Argon........ 39.9
AYSCNIC. c 2c ay seer cas 74, 956
Barium............ 187.36
Bismuth..........-.-..... 208.06
Boron essesscioeetese aes 10. 912
Bromine................-5 79.923
Cadmium................. 112. 38
Caesium................-- 132.81
Calcium. ..............-- 40.090
Carbon.....-..........22- 12.008
CerluM 1.0.00. 02 2. eceae se 140.21
Chlorine..........-...... 35. 460
Chromium............-... 52.036
Cobalt. ccseceucees secs’ 58. 956
Columbium.............. 93.277
Copper.......-...-----2-- 63. 561
Dysprosium.............. 162.5
Erbium.........--...---2 167.6
Europium............... 152. 07
Fluorine........--....-... 19.011
Gadolinium...........-.-. 157.01
Gallium .o. 000.000.0205 00% 70.10
Germanium............... 72.50
Glucinum............-... 9.093
Gold 22 wnscicicnecceccwaes 197.21
FeLi ii 5 ois seieinis e ossinigiaicie 4.0
Holmium...........-..-. 163.4
Hydrogen......-...--.-+-- 1.0078
Tndiumisseccccecend ic ese 114. 87
TOdiNG osisin:s 6 dee wine's eee 126.93
Tridium..........-.-.-.-. 193.04
TrO...ese-ccesececeesee-- 55.833
Krypton..........-.-.--++ 82.9
Lanthanum.............. 138.80
Lead 08s os eoasendeeean 207.10
Lithium... -.22.-2.0e2seses 6. 987
Lutecium............-... 175.0
Magnesium. .......-..--. 24. 296
Manganese............--+- 54. 949

H=1,.
26. 830
119.13
39.59
74. 376
136. 29
206. 45
10.827
79.304
111.51
131.78
39. 700
11.910
139.12
35. 186
51. 633
58. 520
92. 555
63. 069
161.24
166.3
150. 89
18. 864
155. 80
69. 54
71.94
9.023
195. 68
3.97
162.13
1.0000
113. 98
125. 95
191. 55
56. 401
82.3
187.73
205. 50
6. 883
173. 65
24. 108
54. 524
199.05

O=16.
Molybdenum... ......... 96.03
Neodymium. ............ 144. 24
NG@OMscissceecses caxeeeseed: 202,
Nickel.......--.-.-..---- 58. 676
INItOD 2c cies esctaras ics 222.0
Nitrogen............--.--- 14. 008
Osmium 26 «6. ss secceesss 191.07
OXYPelscecec sees ss dace 16.000
Palladium.........--.--.- 106. 66
Phosphorus...........-.-- 31.011
Platinum......---......- 195. 21
Potassium........-------. 39.097
Praseodymium...........- 140. 88
RA@AUIM 6 oie ss 5'e divest deca 225. 95
Rhodium. .......-....... 102. 93
Rubidium..........-...-. 85. 434
Ruthenium. ............. 101. 65
Samarium......-......... 150. 40
Scandium................. 45.1
Selenium........-....... 79.168
Silicon sae eei ose eas 28. 255
SilWerts siosdis os o's sasacieces 107.88
Sodium. cccceccccewesescas 23.003
Strontium..........-...-. 87.631
SUlPHUY) cose sscsses stoic, 32. 065
Tantalum.......-...-..... 181.3
Tellurium................ 127. 58
Terbium...........-...... 159. 20
Thallium.............-... 204. 03
Thorium............ 232.19
Thulium..............-.. 168.5
"DAD aerenstyseieciet's esa Sees 118.71
Titanium. ............... 48.099
Tungsten.....-.........- 184.10
Uranium............2...4 238.15
Vanadium............-... 50. 947
MONO acecccagecuet eee 130.2
Ytterbium................ 173.0
Mtr 2s s00.50 5 5 os casas 89.33
ZING. sciscewsacacax sess sas 65. 397
Zirconium............-... 91.17

H=1,
95. 29
143. 12
20.05
58.222
220.3
13.900
189, 59
15. 876
105. 83
30.771
193.70
38. 795
139.79
224.20
102.13
84.773
100. 86
149. 29
44.7
78. 555
28. 036
107.05
22. 825
86. 953
31.817
179.9
126. 54
157.97
202. 45
230.39
167.2
117.79
47.727
182. 67
236. 31
50. 553
129.2
171. 66
88. 66
64. 891
90. 464
INDEX TO AUTHORITIES.

A. Page.
Acree and Brunel. ........---.-.-..-02-eeeee eens 20, 53
AGC. 25 6 - ocnnodcunieingen ts Ghacmncieuiadnes £5 sce 258
WAQAMONNONE .....- 2-2 eens ee devine eens ecieeesaas 14, 24
‘Allen and Pepys. .....-. ee Settee eo te Pit ree 23
ATO ae elbacaa ri eee ahs lai 288
Amagat oe cee sect oes ae sca Ramaneecauenne et sage 45
WATAD ETE. Gast a honec wes eea ce namasinea Commer aoe 354, 357
ATA CMON sce scanacis ea Fapdesessaseeenee ye ei wot _ 161
ANAPOWS cctv doce bee ves oedema skeen sete eel te 362
Archibald voc scecsss specs ewsasced Geee dads sss ose 66, 69,
71, 75, 78, 89, 90, 121, 122, 362, 367, 368, 369, 370, 371
APIVOUSON. oicicice's le sissies sed eases sce 101, 285, 316
Rin sie ican pc aan snauacee oad veda 208, 209
Auer von Welsbach.........--.---- 390, 394, 406, 407, 408
AWC] OW:: 25 cicieie's o etsitie ceajoaewiein see es eeaiedasieiaie 170
B.
Babies srenidcieiard ss sarees swaciereccies vs <o es ccarmees 173,177
Bahr and ‘Bunesn suieidevotveiesbnaiecudiees os uiiaiats 874, 405
SINE adaware cs soee ee geue ea cece: 227, 257
Bailey and Lamb... ............-..--- 350, 352, 354, 357
BAKER Sis cicicisin wis cision tinea been cede wa ormeinciedinws .. 806
Baker and Adlam..........-..-------e-eeeeeeeee 156
Baker and Bennett...........-.-.-0-2-0-20-0-05 305, 310
Baker and Watson........-.-----2-2+e-eseeeeeee 204, 205
Pa heed ieee ede ad diaeree  ee 263, 264
Balke and Smith. auisiarerte siajieverseie ses ietors sie alnisisaest roisiaee 261
Barkla and Sadler............2..2--20-s see ee eeeee 345
Baskeryill i .c:iocessiesesescciwtee es osc daaialsatigr . 234
Bates and Vinal: ..........02.22000cceeeeeccceees / 84
Batulecassescasiooesonssacaniewses ss sameeeas secs 47, 316
Baubigny......-.....-.-- 127, 128, 180, 216, 267, 333, 334
Bauer. . -..------- 2-22 eee e eee e eee cece ee eens 379, 383
Baume oios sainccad oegasese sccecsee stcsitiec eens 46, 113
Baume and Perrot...........-----------26+ 20, 46, 54, 114
Baup....--ececer cece c cece cece cence ee teesesecece 140
Baxter. ......---- 73,78, 79, 81, 84, 85, 86, 87, 195, 325, 326
Baxter and Chapin. .......----.--0---------- 70, 71, 395
Baxter and Coffin..............--.--- 66, 69, 245, 246, 342
Baxter and Grose.........--.---.-+- Mi cinsaisaciees 186
Baxter, Grose, and Hartmann-...........-----.- 197, 198
Baxter and Grover.........---- 70, 71, 75, 78, 163, 164, 165
Baxter and Hartmann.............-.0-0---e20-5 198, 200
Baxter and Hines........-..------ 67; 69, 75, 78, 194; 320
Baxter, Hines, and Frevert........-..- 69, 71, 75, 78, 194
Baxter and Hoover.........---------+-- aoaoses 323, 324
Baxter and Jesse........------ 22-2 e cece cece ee ene 271
Baxter and Jones............------- etree see eae aeey 237
Baxter and Moore..........------+--+-++- 67, 69, 238, 239
Baxter, Moore and Boylston.....-..-.-.---- 2. 75, 78, 238
Baxter, Mueller and Hines.......--.--+---++--++- 270
. Baxter and Parsons......--..----.---20----eeeee 342, 343
Baxter and Starkweather...........-----------00- 232
Baxter and Stewart...........-.----.ee-eeeee 70, 71,393

 

Page.
Baxter and Thorvaldson..........---- 75, 78, 162, 163, 326
Baxter, Thorvaldson and Cobb. ........--.-- 79, 325, 326
Baxter and Tilley. ..........-.-.----------------- 56
Baxter, Whitcomb and Stewart............... 68, 69, 395
Baxter and Wilson.........-..-.+-..2.2-5 67, 69, 162, 165
Beckers Gis Pisceacesies siccaane geebsce eee saacad ce 5,9
Becker and Meyer.........--.....-.-.--- ahaa erates 222
BONOdICKG iss cesses ic base gccdcawmiieks se eee dete 401, 402
BOMNBeP: 268.5 viyiass oanadeadas oelaal = Peawaries 384
Betlinssejoscdcca ss $8 Seases 265, 267, 269, 270, 272, 278, 374
Bernoullisnicncsiciesesccevnuces cacdwss soe nanan’ » 277, 281
Berthelot, D........-2......----0-202--0-- 40, 41, 113, 409
Berges: cs 342 52cyanazeees <cacaudsteeleadue as 9, 58,
60, 62, 65, 85, 87, 89, 101, 109, 110, 125, 127, 134,
140, 142, 152, 153, 158, 159, 170, 172, 205, 206, 209,
212, 215, 222, 226, 228, 229, 233, 236, 240, 243, 247,
262, 264, 272, 276, 281, 285, 292, 297, 311, 312, 313,
316, 320, 321, 324, 346, 348, 352, 358, 360, 362, 374
Bettendorf. eiaiSiatarmare iedeeeweeils 379, 383, 397, 398, 401, 402
Betis and Kern................2-...cceceeeeee 161, 162
Biot and Arago..........02..02.ceeeee cece eee eens 23
Bitckénbach scs2.sasaeeescewaeceswels Sconce .. 258, 259
Bleyetish oe scs sc eiieereeenckcedec oc ocnvonsewe -- 241
Blomstrand <3 2222.2 sncsscce setewesiee’s oseteieerene 262
BUM... 5. so catarensiare sinew seinen cig ies we dmeuseecece 195
Blumenfeld and Urbain..............2...2.002222. 408
Bodman........-...--..-.--.----+..-...- 375, 377, 380, 383
Boishaudran....s..s000sesesececsercceceswnas 216, 401, 403
Bongartz...-....---------0 +202 eee ce eee eee eee ee 254
Bongartz and Classen...........-.22-.....-. 229, 230, 231
BOrehis. case oi oances stants Seedaeis vs eclentdee cease 277, 281
Bowtdotlatd os. nocicscccaceccwuiies ose s dees nae 387, 395
Brauner........... 8,
208, 234, 235, "298, 300, 810, 379, 382, 383, "886, 387,
388, 389, 390, 392, 394, 395, 397, 398, 401, 407
Brauner and Baték...........2.22.02.... po isso siete 388, 389
Brauner and Krepelka............22.2-.2.2.2000- 232
Brauner and Pavliéek..............2-..0..22.0005 380, 383
Brill.........---.------ 376, 377, 382, 383, 395, 397, 406, 407
BYIsC00 2 2ewiemaseeee quaeese tees s tae taeaevecauewean ~ 231
Briscoe and Little..........2.0...05e: 00020022
Browning and Flint................0.0.000.ececeee
Bruylants and Bytebier......: :
Bruylants and Michielsen
Bucher: sccsessciveceenccccs
Buehrig ec cedie on ceeithtos one scvtaneae lau ses
Biff 52 cisicsinqnasamaesntcces aot ladaaeasemmdba ce
BUNSEN. oi comcwvene service casas
Bunsen and Jegel.......... 22.2.0. 02 cece eee w eee
Burt and Edgar............ 2.2.02... c cece cece eens
Burt and Gray...... 22.2.2... ce eee cece eens
Burt and Usher............2....02.ceeceeece eens
Burton and Vorce.........0.0..0.ceceeceeeeeeeene

 
414 INDEX TO AUTHORITIES.

A Cc. Page.
Capitaine jasc. sig ersene cro Aainipcsto neice winyelslsiscisuersigjerncie 321
Cavendish. .sccsecdsiecdauuenev eesivaivesletsistteacte 23
Chapin and Smith................2.-.------2+-- 263, 264
Chikashige...............-.. oe Sea oae ees Decerhiier 300
CHOUDING si waiceinde howe sikaaciine secs IE Sse oe 378
CHrBteRSED ai, ss326 s scomanennscnsrraecew assientmnee 314
CHY GENIUS oc cscseoasacaeacens se cus cccmandcenere: 233
GMAT ja pccrsceia stew isdiinoed teakadanaeunnmannnaes 12
DINE anduadessenaathoatonn cxpcinliomapnenne BL, 188
(Clas8e....2.0.ceceececc eee e desc eeeecececeeeeees 257, 259
Claus See ee nny se beecece 345 |.
Cleve) Als eeviesigaain desscs aieereiaeierstoeis ajatstaletamratrgatace st A0T
Cleve; Pa T wcsssxssitevaccatessceetten eax - 233, 235,
372, 373, 374, 377, 378, 383, 397, ‘398, “401, 405, 406
Cleve and Hoeglund...........2-2-2-e0eseeeeees 374, 377
Cohen and Strengers.............------. foreicteac Sais . 254
Cooke..........--..---++--+ 25, 28, 78, 80, 109, 110, 250, 251
‘Cooke and Richards.......--..-----+-++-s+++++ 14, 23, 24
ASEPLaiiay cosh Sena ahaeinealaey patentee 18, 84, 45
“CROOK Gai ca gs acucsslrawiemicstcacdenssascidasurebyeilbr 219
Curie, Madame........2..2-202- 2222 eee eeeeeeeeeees 156 |.
Curie, Maurice..,.....--.. sragnacaanee eee
Czudnowicz....... ieciebigSuignia tae eteeeees 240, 378
D.
DAV iB. deesewe see tisacccnnscectindsceecsteocsiagas 166, 168
Davy vocewenwceucioreccokose scence oomncmees 23, 211, 312
DOAN... 2 cceweeratatecaes eeeiess aeennceess 91, 92, 107
Debierne.........j.c0c0cse-cee-scaxsessaeceenccate, 470
Debray........00.0.02 cece eee eee cece eeees 170, 278, 275
Delafontaine...........---.------------ 233, 374, 375, 405
DO LACH. os visiais ais sciniendaas piierese siminiseynescieseteisiest aia 313
DEOMALGBY: 5.o:aisicis wie aide nin wenn dca eiersieieiciaiermiecee jv.. 398, 400
DOMOLY cies ainci ree cnn ween Saamaranandcinacane wes 224
Dennis and Anderson...........-0edeeeeeee teense 810
De Saussure. .....--.--0 2-2 ee eee eee ec eee eee cence 23
Wie illdces dante nnd iiaatouestytied saree aeeeeccee 205, 327
Dewar and Scott......-.....20- 2-22 cece cece cere eee 318
Dexter, .csccecseeeseenen get ieiusiandsseesessesse 248
Diehl 2.20 2cccadesee sees seeeamasinwes Geese 101, 103
Dittmar; H...... 0-20. +s ease2cecaneonseseeseex 348
Dittmar? Wisocc0csdscsesen eae eee 101, 103
Dittmar and Henderson.........-.-.------++-- 18, 17, 23
Dittmar and M’Arthur......-..-.-----------+:--- 367
Dixon and Edgar......-.-----+ +002 +eeeeee seen eee » 48 |,
Dobereiner..........-0 20222 e eee ee eee reece eee 158, 321
-Dobrovolsky.......-...22-ee2 eee eee ee eee ee ae ++ 206, 209
Dudley and Bowers........-..-------+---2+---+ 309, 310
Dulong and eee srciicndelovaislcisidaniacnesieinlncisica 11, 23
DUNG oasis acing LM iis icwisiciarererniotioninys +. 11,17, 23,.68,

65, 84, 85, 87, 92, 93, 109, 110, 126, 140, 142, 145,

146, 150, 151, 160, 165, 175, 178, 187, 212, 222, 224,
228, 229, 236, 239, 243, 248, 256, 273, 275, 277, 281,
293, 297, 312, 313, 314, 315, 317, 320, 324, 329, 342
Dumas and Boussingault...........-.-- 28, 24, 28, 34, 35
Dumas and oe vd ns peesaneinenter ewe Neeswes ese's 41,43
Hasley....-.--.2--0-2ee eee e eee ee eee eect eee e + 208, 205
‘Easley and Brann................... wittteteaede 204, 205
BDAUgD so. .0:<disie csiwicinisiaitierantiars semiwiniew ae asin 65, 69, jane 246

Pbelmene 2esecibaiwina codicacadeneaceetueaeucaant 286

 

 

 
 

Page.
Hganiand Balke iss. ooicainisiaicsciesaicisctaieig aieie si sieteierere 376, 377
Ekman and Pettersson.......--....--.-..+-- 293, 295, 296
Engle and, Balkes.ienc ssc sccmsnnveseseseane sees 403
Erdmann, Av csccenusssscuussmenseecesewseeseee 179, 184
Erdmann and Marchand............- 12, 13, 17, 23, 41, 43,
109, 125, 127, 140, 141, 142, 200, 204, 293, 322, 324, 327
TEEN ss oyn's ais cieiciosaiciaiarctane areca onaisiynare sikelele 378
Ewan and Hartog...........------22002eeeeseeees 206
‘ F. :
Waget-.sccaesencstscessweaeseeaneingessawaxas BO
Pajans. o< eisevercesxenseeeeseseqainess 169
Havidevescsu2e edetecasees sare seca deena 180
Pele eerste si cicissaeh wigan cesicceirnendcras Sains (erermrceie 403
Feit and Przibylla........ 377, 382,393, 398, 400, 402, 407
Fischer and Froboese.............--.02+0-e--eeeeee 409
Fischer and Hihnel....... Sebi ee Shek Geet emerge _, 409
PDE o'es'seiisieweeeicis cess gerccmccbiomegie eet 307, 308, 309
FOU «3 ccnseeeastewuicas aphidissaleseie’e eicnravan arate staicle 308, 309
POUrCKOY ac aa:c.sei nye dares enaneeae sees pehtittrtatciees 23
Premy: ossssasisssmense sss elaswseswasex ss uecncios 358
Priedéls:ps5qasssssasiscewene eee eonenseen 5 Sateen 42,43
Friend and Smith. ...........---..2-0- 22sec ee eee 255
G.
Gallo wecescsondainaccascinetecsicis 83, 84, 132, 183, 304, 310
GAUIEH 5. cc neccmicmwes Siem sabiipilcitcncteneeimece 209
Gaby LUBBAC: oe ene eeee ee scien ed cniem einen sins 172
Gebhardt. cc.onacecinwawaiisiwia emalsmnccmuitie ice ieteieig 356
Gents cai jatisiviccnasccisisd aces cteem siren Sess 384
Gerhardt. oicsecccnsedin secs Seeger Serdelase 59, 60
Germann io. seste suanieese oro goueenonoce esis 31

 

Guthier.............-. 258, 259, 260, 308, 304, 309, 311, 355

Guthier and Birckenbach................-.---2.-- 258
Gutbier, Haas and Gebhardt........... eines ciscis 356
Guye..-.........--- 5, 30, 31, 40, 44, 45, 54, 55, 60, 113, 116
Guye and Bogdan. ............- ee eee ee eee eens 31, 34
Guye and Davila................--.. Rescue i... 38,40
Guye and Drouginine.............. se Soh inane $2, 84
Guye and Fluss............../ Bees ia cee aanseeny . 51,52
Guye and Friderich. ............. ot i ares, ~ 60
Guye and Germann... .............. eipenae neni 6
Guye, Kovacs, and Wourtzel.......... istesey 28
Guye and Mallet.................. 022.222 eee eee Ea
Guye and -Pintza..............2.2-2-0---

Guye and Renard.....:........ Sout a niamsniorcemrenie

Guye and Ter Gazarian

Guyeand Tsakalotos............01'

Guye and Zachariades

 
INDEX TO AUTHORITIES.

H. Page.
Ha asi vies spelen dcoinsed Sioa duidontdess ajaed . 355, 356
HapeMs eres thend ose dade ca Bede Jeccageeese LOL
Halberttadt) ccc caso essa cececdoag se 364, 368, 370, 371
AMG oaicioria ds Gsece sic Seeder sen samme 126, 127.
Harcourt and Baker..................00-00 0000s 308, 310
Hardin eciivascesteau tec ncddedseoseseases 97, 98, 106,
192, 193, 197, 198, 201, 202, 203, 204, 279, 281, 352
SUG! dacadpiesicacedenetemaeanks ike 187, 189, 297, 317
Hebberling..-.:.......-.- eager esis salamat 218
Heber lets iata's steiner esas piacere <Eeed 301, 308, 310
Hempel and Thiele............2..2...02222005 .. 339, 342
Hermann. ............ 101, 226, 233, 261, 262, 378, 384, 406
Herzfeld sic ciic5. ce mineticenoawsengeet teu emreisae 141, 142
666 osc cai as cane ee oe eadiawanes eee sundae 409
PeWCoe olka dceeay anes 121, 122
hice ovat a acea ail vctan dasgaiee i 99, 100, 101, 244
Hinrichs: s2escsc83 208 $02 cceiyodase ts ote tae 31
Pini ehednt je he cance esate eamnasu ase acces 141, 142
Hinrichsen and Sahlbom................----+-- 262, 264
Ose ln G sssenondssGeawete to teawoawca cs seu odees 405
Hignigechund Aandi Stare ielata'a'si vel ctarald Acerawle clatein vie'a wed 67, 69,
75, 78, 79, 80, 157, 167, 169, 235, 298, 373
Honigschmid and Horovita..%.scccseseeswees sees 68, 69,
70, 71, 75, 78, 164, 165, 168, 291, 292
Hofmann 225 ah gone Hass ceie es 5 qamieeereae eresiaieies 3 cs 406
Hofmann and Burger. .......2.-2..-2022es scene 406
Holinberg eo. s.i2igegasan tides sameneet eaten 395, 404
Hole mann p's casiciouin's ane sionisisasiarecnsicuine'y « 361, 362, 378
Hopkins and Balke........- ania as Wists stent as 377
Hoskyns-Abrahall. ........-..-..2--.0e-s ee eeeee 206, 209
Hoyermann ........... Ridaseaeeuees eee ose 361, 362
Hulett and Quinn............--..-022-eeeeeceee 197, 200
Humboldt and Gay-Lussac.............. aka Gecieais 29
Humpidge and Burney.......-..----2--eeeeceeeee 405
Huntington.........-.+-2;+++-+++++ 72,78, 78, 80, 188, 195
Hiittlinger ss: sc25.3sseccseneencs ssacedeeeee ses 347, 348
I.
TSNATE soccer ed woeyeiernciesieis 6 3 see E CRS ESS Es 212
J. ,
Jacquelain............--20+2eee2 e+ 172, 177, 179, 184, 265
Jannek and Meyer......-..-------+-e+eeeceeeee 295, 296
Vanseetl........0- +s ceceacereawisws sea ee seatee ces es 260
Jantech. ..--.-.-- 2-02 e cence ee eee ee ahegeaeSinkie'ss 400
Jaquerod and Bogdan......-------+--+++---eeeeee 387
Jaquerod and Perrot.....------:---+--++- eee 39, 40, 45
Jaquerod and Pintza.....-..---- Feerees Seater Seisienesie 113
Jaquerod and Scheuer.....-.---------+seeeeeeeeee 31, 40
Jaquerod and Tourpaian ...-..---------+-- Hhegteieig 30, 223
Jorgensen... 222-222. 2 2 eee reece cnet eee nee 347, 348
Johnson.......-.---0ccec eee e reese eee ene eeeeeees
Johnson and Allen.......----------e-e eee eres 123, 124
Jolly... eee cee cee eee eee een erence eres
Joly.....20-. 222 e cence ee ree nee cee e een n een anaes
: Joly and Leidié bosae anes sah iid medi aie/ohausaraiis eacis
Jones, G..-2.... 2c eee eee e cece ence ec een essences
Jones, H. C.....-- 374, 377, 380, 382, 383, 390, 392, 394, 395
K.
Kappel.....----------+-2 seer ee cr ctr escrscccses 397, 398 i
Keesom....-------- 0-2 c ccc c eee cece ccc cccecsenes 45
Keiser... .. 02-22-0222 cece eee eeceeeee 15, 21, 23, 343, 357

161566°—20-——27

 

415

Page.
Keiser and Breed.............--2.eeececececees 352, 357
Keller and Smith.............. 2.2.2 0sceeeeeceee 350, 357
Kemmerer is ccccucvceiae vais sce Btls insite 355, 357
Resslet s,s. sveucyoeaciens seis 2 243, 247, 248, 251, 252, 266
Kiwanis 2.22 -savsage nosy es oo a des bee's seinsiaisiine’s 23
Repent esse cance cee Set en atau aneeds 384
KatZ0 bis 2 basnscuscnasdes Se aoeaeseuwieges sesaere ‘170
RO esessSastraenctace te Scr naeeenWwecwcaaess 388
Roethnet os 2 ieee eciaieute es sseaidesiciesins sie'sis's 302, 309
Koethner and Aeuer..............-2.2----2-0-- 81, 86, 87
Kohn-A bresti.s ssscseececccs see cscees aces sos esis 215
Kralovanakyoscasc scccce cosine tcnce int teinee cde 101
Krell so ccsniccctint Hanae oucade so magnoeeee aac s 355, 357
Kremers and Hopkins. ...........-------0--+-00- 378
Kremers, Hopkins and Engle......-........-..-- 404
Kropelkaycccsec se ddonasenstennee cee nscanceecs -+ 233
Kriss s.3 os see se sesencereaseesd se 134, 136, 1B, 334, 406
Kriiss and Alibegoff.............-----+---- 2.2 287, 333
Kriies and Moraht...............22202.22eeeeeee 170, 171
Krtiss and Nilson.........-.-.--2.---- oapareeand , 254, 235
Kriss and Schmidt.............0.-0...5.0--04- 335, 343
Kuda. ciskecute nev sx ocelneutbiebacar deta daepeeen 260
Kuzma and Kfehlik..............20.0eeseeeeee 295, 296
: L.
Ladenburg iss s.sioiec! cadences eseeesaneeis 81, 86, 87
Ladenburg and Kriigel.............2.....202-e0s0e8 410
aperhjelini ooo a ddosncs tensa eee aceite . 256.
Laird-and, Hulett: ocjccciaiselewels ss ovtsemeciewes 196, 197 -
LAM Yeni cicsnece ssi vewssiesdcese sus es aaeeeemenee 218
Tangletss ccsesvies st soos citusadcaereiece ss sasiccaens _ 409
MUTED goes leis ieicias cities use caaiew aiieie oes Srvars 205, 206
LAVOIBOP so sired boc acgesnes seeRaese rca cnaaeee 23
Debéailiocsscresneeecced occu aaneeieneches naveads 290
Le COnté yo stecgetacendeeodavsestaueewaceeats dis 224
V6CUG: cciovwomsieswaasawenmciwrescucwess oe 19, 23, 25, 28,
29, 35, 36, 37, 38, 40, 41, 44, 45, 46, 49, 112, 113, 409, 410
LOG esis tieinscemas Seactcesteedeseiceucmees «-+-- 332,333
Lelortes cco dee cede tee cease sce oe cieaeinn ae oy 265
LOnHOP. o2.5 iccicisseirceindaces 65, 69, 294, 295, 304, 306, 310
TO0R90 352.5: ocscoe decdunsscemouetentnecaseaces 187,191
Deonards dees dicen po ccitieatecendsacaanen canes | 208
LOPIOITG Joes sais solare vio sareisinineaeasieweioee easels 220
LevVol escape soca catetets vs Saraaasd ieee 6 le 134
Liechti and Kemp. .............-..- stead bs semser 274
LOW6 a. cxisieseicvecs corsa ails Pu vecsle eis dis Steinidince'e: hs 257, 259
Longchamp...........2..-22ese cere cece cece eee 158, 172
Lorimer and Smith.................. said diaetee da - 190, 194
DOUY Sts o5)cecisis mwicecreiniew'e'c eweiieiel nec 312, 313, 314, 315
TUR ONG secs siatcinicinre o's wae sensicmesescanaien'’s ese geinerns 373

 

Maki Dstaaraiaye erst eisialweteveleive sare, 9, 0:j9.0yscesoceseud- ere \eldieieierave, be See 401, 402
Marchand............ 0.02.0 cce cece cece eee 276, 281, 285
Marchand and Scheerer..........2..0.......2-0-- 174
Marek wal 2. oi. sc sieeve cei carniseaenene asides cs 306, 308
Marckwald and Foizik............ 2... ee cee weee 308
 
 

 

 
 
  
 
 
 
 

Page.
Marignac........-....4. ... 8, 9,58, 59, 60, 61, 68, 65;:73,.73, °| ;
2g. 80, 84, 87, 89, 90, 91, 92, 94, 95, 96,'98,.99,.105,
106, 142, 146, 147, 149, 151, 153, 160,165,175,
“ . , 177,179,180, 226, 256, 259, 261, 262, 264,319, | «
ae . 828, 329, 334, 342, 378, 383, 384, 397, 398, 401;
Mahe? sc aciew Se cinss ened ei sasp on cuaweeen eeeees 212, 215:
Mat OS i isso cicis.csnscpalevesmisverdeeraicisteueisinleisiecs's wists veces BUT:
Matignon... -- 398:
Maumiené.........-..-2--...-- 59, “60, 65, 90, "106, “322; 324"
MGaghab. c.c.02sieeeevscinecieieasaus « «162,186 ,195):197'
Meh lets 2s fected cee wate reue tothe irae sian sans 2505
Meineke sjsaececa'sc os geanaeadeenes dassosses 268, 269; 270:;] ;
Mendobéete sa: ccidvisuaaies aeecewetaiceawuneneles 197
Métener 25s s.scs2s0nesxeeseewwes ia widen eae sis » 300, 311.7] °
Meyer, Jie seaess ies staecs ey Pasi cowemdis 295;:314 | :
Meyer, L is bled naa SSSR RSRN eae deme daeeemend Ge TAe
Meyer and Goldenberg 372, 373
Meyer and GumperZ. . . .....2...0 2.00 e la eweene eee 7234; 235
Meyeriand Schweig.................0-2 2000000 we 873°
Meyer and Seubert.....-.....2.0..-.2-. eee 5, 9,20, 53
Meyer‘and Weinhaher.................22.--0004! a
Meyer and Winter Bi
Meyer and Wuorinen..............-
Man ess tere )ate se cieiete tes vein 's'e'e'n x ste sisteraceraslesete
Millon and Commaille......
Mitsclietlich and Nitzseh..

   
 

313, 314) 315, 387 |
54; 55,111, 316

Ss paryetebesek 334, 343

oF e SECC Wa remeciisa Seeee ewe --- 410
a sides ou aeaneme eae 308
Morley..:c. 2.2.02. 2-0. ee ee eee eee 18, 19, 23, 25, 28; 29,31 |
Morse and -Arbuckle. .........-.--..--+0+- 183, 193; 194
Morse and -Burton..........---------0- ee eee cece o 183
Morseand-Jones.......- - 189, 191; 194

Morse'and Keiser... - ... . 182,183

eee en ew eee

 

 

  

275 |
9. q
228, 229 |
» 126)
Mee Esa alcie at Ota ie tae a naeedai ee 4555:
Muthmann and Béhm. =
Muthmann and Weiss......---....----045
hay iss adeteemacnuennce
Nilson and oe sdavachuane iniaraleleie Cities Mimics "170: 171 '
Noyes and Weber............ Spree ensaewcsoucesits 48
oO.
Gaines de Coninck..............-- + 144,155, 289, 290
Oeehsner de Coninck and Ducelliez........... 127.
Oechsner de Coninck and Gérard...... 165, 199, 260, ne
Olasiepldscsian vous padade saenatneneggpeseneesentn
Ostwald,
Oudemans. 1... ..2... 20+ e sees cece sete re eeeee
Copan; Balke, and Kremers......-.--.-5+++++
Pes P.
Parker oka RE REE Oa Ok Sees eee OS teigewpgea se LAG
Parker and Sexton...........--.s008 diecswsinin wispeiviels 345

 

/ ee eis paca ata ta

 

INDEX TO AUTHORITIES.

Page.
PATRONS S .cecigie sou oeiensariare bee kilccgiwens eeee 46,171)
PAM G6 oo ayinceieey igs aceite deans 188, 189, 19%
(WPecheuxs:s owcwe'sis cis odo usancaswsaaeetw ani 138, 162, 186
PHligot. -sewcecwie seveied oad eesnseeseeeees 265, 286, 287
Pellini. . oicexseses sees se ceaeeeess ~ 302, 308, 309, 311
fF elom Ea hs Ade 'any ome Riles meme Pom a meateice terse --. 58,
4 ” 60,92, 93, 94, 95,145, 146, 149; ‘151; 222, 236, 243
: IPetinington and Smith....... eo Nahioueiesaaat 278, 281, 285
(PORN Y-« a sicsnecinineaintoinweea 58, 60, 62, 65, 96, 98, 99; 100, 101
‘Perdue and Hulett..........-...00-eeceeceeen ene 195
‘Perman‘and Davvion paiatle Soaieioeste Haran sesemaeas Jee. 39,40
SPOQRBOZ 60's wine Saletan es wae ovis vied tettaine tacaeoans 278, 281
‘Pettenko- Eaiechonki.. oe SURRY ESE SERENE REESE 352
"PH6Hers cv eccccccaas seuss seuss ces eosnoewaxse ees 253
PiCeard be ncsswencuaenie sore Seeds Seeeemeesiins 121, 122
(PAOLO Si. soe eisicn dicing oases te SSS tiacraeieeGiiees , 224
appease sseancccataadacasa ss aesseeeeecacaraeneet 374
PROPID RT so ince Wibie iaesthe acest cai Rid § ciae cere e atoee, a.cecverants «. 253
(POSERUBE Ss. Sices usetioones aaa ara BBR cnace tae 375, 377
PONE 2 Shree setae sa ahe abaya dtigeleeieions 403
Prandtl and. Bleyer...........-.---2-2.220-++ we. 241, 242
‘Gotan and Hulett.......... olgiee Uni slevamaiesiaailns 197, 198
Quintus Icilius.............. Este watalaohd Aes cane nee lacey 349
Den tas R: |
Rammélsberg..............-.-- 260, 273, 275, 285,'378, 384
Ramisay-........ Siaaddeean es bees'acd anc nase 409,410
Ranisayand Aston. ....-..--.-----+---- 98, 207, 209,211
| Ramsay and’Gray........... poison eae adeee minnie 410
| Ranisay and Travers.....5...-..6.--5.2.022006- 409, 410°
Rawack:.....2...-22202 60008 iicet estas Reece 317
Rawson Shah P ata uwbs ee wteatdecaen sc ES aubigat gavevanere 267
Ruyleigh: -...2..---- 2022-2002 e eee eee erence eee 14;
ae 16,28, 24, 28, 29, 35,36, 37, 39; 40, 44, 45, 409°
Rayleigh and Sidgewick. ........2----.----. eee 131°
| Redtenbacher and Liebig..... igeisisis saree ez 104, 106
13, 23, 24, 28, 34, 35, 44
bs chante ie ala isha chelates asceeraveh vet trac 216
cceueteemeemeeeees ceased x eevee ed evenmeees 54, 55
Lah sooth GaiaweceeEneeee ve se we aca ores 334; 342
p2geueee dete ssee ese eeeuseneeceee 6
Sain Sctbid ahvichie“elaesld OS's orb'e sestsiAofenararsiarese hraderche side 348
éjnclie and Ramsay. .......---..-0.2-5-eeeee 182, 183
5, 6, 9,44; 71, 72, 73,74; 78; 80,

111, 126, 127, 128, 129,130,181, 133, 148; 145,
146, 147, 148, 149, 150, 151, 152, 158, 154, 196

| aha andhedereiee. ccicemncgees eobabumibaiieste . 64; 65

Richards and. Archibald. ........ 20s cease eee ee eee 65,
69, 75, 78, 88, 90, 101, 123, 124

Richards anid “Baxter sictacvedeedeudevaweasteeieesie 6,

74, 75,78, 323, 324, 340, 841, 342:

Richards, Collins, and- Heimrod.......... 72, 78, 132, 188:
Richards-and-Cox........ 0.0.00 cece eee e eee e ee eee: 104
Richards, Craig, and. Sameshima,............ oy... 169; 216
Richards-and-Cushman....... be wioeaea aaiviey 74, 78, 339,340."
Richards and Forbes.......0.. 00.02.0002. 2222 te 97; 98.

| Richards and Hall......................--- 164, 165, 167
|| Richards and Hénigschmid .................. 79, 80, 143.
Richards and Hoover.......-.......- . 79,80, 108, li
Richards and Jones.........-.0.-..000+s2eeeeeee 111, 12.
Richards, Koethner, and Tiede................ wee, 95
INDEX TO AUTHORITIES.

 

AAT?

Page. Page.
Richaras and Lambert......-...... oats oaisine 67, 69,166] Stas. .a:... 8, .9, 48, 44, 59, 61, 62, 63, 65,71, 73, 80, 84; 87,
Richards and Merigold.:.........2.... 75,78, 288, 291, 292° 90, 91, 92, 93, 95, 96, 97, 98, 99,100,
Richards and Mueller.......... res G a aeealaiats 78, 80, 91; 92° 101, 102, 103, 109, 110, 112, 159,160.
Richards and Parker-.................. 69, 71, 177, 178,179 | Staudenmaier.............2....2..00ee2e eee 299, 308, 311:
Richards and Rogers......-.-...--..------ 78, 80, 184,185 | Steiner. ........ 2.0... ..2 0.0 eee eee e eee rece 295, 301
Richards'and Sameshima............-..--.----- oe .d69 ) Stewarty-eccelauccsied a cuelses oe Sees ee » 124°
Richards and Staehler..........-..-----.-- 69, 71, 89, 90 | Stewart and James..............2.2--0-0-000ee eee 398
Richards and Wadeworth.............-. ne. 164,165, 167/| Strecker. ......2. 200002 c cece cee cece eee newness 105.
Richards and Wells..........-....-- 8, 64, 65, 69, 71, 93, 94 | Stromeyer.....-.......... 20.22.2002. eee eee 101, 145, 187
Richards and Willard....................- 70,71, 108, 104 | Struve............0220.22-0220 cee e eee eee eee 110, 112,
RACH Ge2 os et hee owe Pe otek ee asoee 281) Sve bere ntcrensceeseenieeses se os tscn onan 161, 202, 203
Rimbach 4. «2 f0sbest fe oss ese deeeeeecave ccc aaces 208 | Svanberg.and Nordenfeldt................-..-4- 173,177
Rivot.....---. elects hai d ca cndas ec aurntes auctnn racial 823,324 | Svanberg and. Norlin,....-...2-....2.....--.+.- 322, 324
Rebins0is osssndsons ana wasieced ess oysueheue 886 | Svanberg and Struve....-.....--.----------+--- 110, 272
ROGCO6. eviccwn ine eedanicmns 42, 43, 240, 241, 242, 278, 283 |."
ROO ssc tae ucey aan Adana aie heart 248, 260, 262 T.
Rothholts..2ts ie cane queaceeae ss xe kedeeeeke B27 || Wavlotecedscn ses eue se send Ssoekeececeee ss < 280, 409
Russell cecviseawsstsaescoeceeswexees sacs 829, 330, 342, 343 | Taylor and Hulett.................--------+-+-- 204, 205
s Ter Gazarnial. <2 0:2cccn ccc ob2058 Beatles haweieests ae 237
DaCOwisues oe nsaar ded Aganaewiedeudecehad ase 292, 296 ao - . et nea een an once aac a ie
Salvetats..hc6s5sicacassdeaceascaasesesescace 145, 149 rea gett tte Nt aan 90, 217
Scheels .cccecuves. sidaesdecdaw bes vealagedcecat OOS | ineing i Lica Le nc oe cel oted cos 2
; OM AS: s:crsic aeecisiins aac sy anaBaree ies bose eeeS 284
RCHOCTOR nites xcs cseanestebiien +acemaemnbesiaht MEE SIS | era seit cc sessmaiinsneies +++ 13, 20, 23, 27, 28, 31, 53, 215
Dehetblatny> +1 sceensstacness svesatape senivexs> eee 1) Mihov enuatecranies co een a5 uo bee aae 23, 211
Sree tases mete BY, 38; 84,88,/30, 80,81, 67, 60,112, Thorpe........ 69, 71, 73, 78, 108, 152, 153, 155, 157, 224, 225
Schmidt Pt te pont Vi batten ea 231 Thorpe and Francis............. 70, 71, 79, 80, 146, 148, 149
bap aes h 1 ey ius in, pe avg Pr ola gy Thorpe and Laurie.................22. 74, 78, 186, 137, 138
Schneider. .......---------+2--+e0eeeeeer ser ere es 247, Thorpe and Young... 222
252, 253, 256, 259, 276,279, 281, 317, 827,928, 930,331 | mag, oA
SCDTGttO Bunce stiect ident diaiiadenwcneiescensansts. 236 ia MO IA tie ge Ace 912 215
Schulz sscccssciescd cadence esas enacts Sila pita tt eee Li ley ito oe are ena :
DONWCY sovsneotecege ness <a xxairswaatscess 182, 183, 214, 324
Seb wltge se vised sade sd ochiee elses o'a'a.c ctetsele eeswales 3 409
Schutzanberger......--<.0+-+-++--e0eeceeee 334, 343, 386 cs ‘aaaties pea ees tO E08 OE
Scott. c-.-<s- 9, 43, 44, 46, 72, 73, 81, 84, 95, 96, 107, 108, 302 | — M"----- +--+ pObe tad den As, AP eNe AOb 20818
Sears ose cvc.dusehedete ees eh dss yeedsamearecaces 264 Uv.
Sears and Balke.............---2-----2-------- 263, 264
SObelION secre teas cwsaeeaneGe eos ea areas caiaees 5,9 ae Ge a MRA SAAS cc rena csr 200
BeTSUY ETc... . sare ocomsnnone sucha weemauumusucncee 200 Urbain, Guwrsteae skies Seiseasedns 234, 235, 402, 407, 408
Bae rae aera ashe aie cate re nee
Seubert..............- 302, 358, 360, 362, 363, 368, 369,371 | ae Boor een ree ra eae
Seubert and Kobbe ae cu ttieal a ANG at wn eionind delta 347, 348 Urbain and Lacombe FT ee grag ETT ae atte ae eae 397, 398, 400
Seubert and Pollard. ..............-.------4--- 274, 275 Vv.
Seybold........--2----2+++2eee eect eee eect eee ees 358, 359 | Vandenberghe............22..200.cceeeceeeeeee 275
Shapleigh......-..----+--+-+++2+++++eeeeeeeeeees 380 | van der Plaats.................. 42, 43, 181, 183, 229, 237
SHAW vices sawe we deo oeiescenee seen. ao eemeaaee 131, 133 Vea GeTUns cose acs 5 seerecsceces oes eve ea eaiavereee oe 9
Shinn.....-....-..-------------+---++-- 279, 281, 356, 357 | Van Haagen.......... 22.2. c cece cece cceeceec eee 263, 264
Shrimpton: seosase2 sseneeecudecie sees eeveciess 132,183 | Van Laar......-- 222 cece cece cece ccc ccccceccesee. 6
Slowert....2dcee. se scaetieave ee esse eee 266, 267, 269 Wath Tilers eee noite ate ad oa ee 181, 133
Smith, E. F........-....----------+---+---+-- 5, 244, 280 | Vauquelin........ 20200020 ce cece eeeece cee ceeeeee 23
Smith and Desi...........----..------- eee eee eee ee 284 WiONADIS cis sic.5ce Ge eete oc transiichrsraacmede oc beaded 227
Smith and Exner......-..--.------ 280, 281, 283, 284, 285 | Venable and Bell.........-.2----cceccceecccee eee. 227
Smith and Maas. .......-.--------+--+--+e- ee eee 274 | Viaanderen..........00 220020 cceccecececcccccccee 229
Smith and Van Haagen................- 210, 211, 261, 315 Vogt Eber otetacicte maactettGrnet att Matt hee Reged Se 5 345, 346
Sodd yes. ca: scacsack uss tv cctdeeeee eds cecieenes 169
Sodas snl BY BiaN ciccss as vabeeanzees oe <esarakie 169 Ww.
Sommaruga. is focaceuec ss onnesieeacscete cues 331, 333 | Wackenroder............0000 0. cece cece ccc cece 321, 324
Staehler and Meyer ......-..-----+------22eee eee 60): | Waddell. ssceccseues xciey i weueeewareves onan 278, 281
Staehler and Tesch...............-------------- SEO; SLL. | Wallace sci: stecucdquie esx Kees sie dad cicde 2accgeacuen 243
Stahir foes cic 5c).c eee eces sex edeeaniokess oe ean Saas 47 | Washburn and Bates...........0000 002. c eee ee eee 84

 
Page.
Watson s2)sice edo sai adicicaisinens sie denis a situa nisreis 409, 410
Weber ie os oes 5.c24 sanasiuaumiek s eawaleanaes s 54, 248
Weerens ds oceiatcutigunetwuneccccseserendeseates 170
Weeren, JM cena. ctie cae ee teceeeaeaa oes ss 319
Weibull: 14 ocveeuesedadsdawiul vec acetaniler ys 226, 227
Wells and Penfield................. 2c ceeceeeeere 221
Wolsbachcssesscnsasinconssust eeeeareesees 392, 407, 408
NV EHTBRie easeraece nace eeee AN wohl ee 218, 219
"Werte iin as jcisisceistciniscc.a:iois iste wicieia aieimsiaaie.d eee eae 286
Woselalty sciccisciacacisectsatintcememacndiecesunae 332
Wichers, Hopkins, and Balke................---.- 406
Wildenstein........-.....-... A Seiatate seis Sie etelrctatg 265
Wills uctsviceasaseaeeeneennbins EeEER eC oniene 297, 311

 

INDEX TO AUTHORITIES.

Page.
WAN Bes. soec sedate ciislnbnie vtec cine ooh biceisia 385, 388
Winkler.............--- 216, 324, 331, 336, 337, 338, 339, 340
With .cscsaneneccecowente sie anarees son keen 872
Werle aici ic tciseis'e s Veceialy viscieiiargnreseciaiieinieie's 355, 357
Wolfvscuncaniedoatees oo ecules ce sitosenaacaumee 384, 388
Wourtzeliesscosuetiewccesgeneer sss enaboes 88, 84, 52, 53
Wredessauawarncksic coceeiiine co's <ssisseieaee een 23
Wyrouboff and Verneuil............ 335.2mee 387, 388, 390

Z.

LiOthBOW, w wiscasn iasasaiare cin Wolnie a sinisreidin.wve 6 6.0, acoiajeleraleieeioin 282
Zimmermann...........-.-----20+6+ 287, 288, 333, 342, 343
ZBCHICECHE -.6):415'sicseie's/sin'e sive cia Sie Siac aicias'e 60 reverses 378

O
HC 137 L6 N27

 

Author Vol.

Nelson, Edward Wm

 

 

Title Copy

Memoirs of the National Academy

 

—of Sqtenees

 

Date Borrower’s Name

 

 

 
 

 

 

 
 

 

3
ceil

wy

Le

pid