Skatteberg


GEOGRAPHICAL REVIEW
PUBLISHED BY

THE AMERICAN GEOGRAPHICAL SOCIETY
OF NEW YORK

GEOGR
GRAPHICAL
AMERICAN
SOCIETY
BIQUE
1854
MAY
OF NEW
1918
PUBLICATIONS OF THE AMERICAN GEOGRAPHICAL SOCIETY
Books
Memorial Volume of the Transcontinental Excursion of 1912 of the American
Geographical Society of New York. Edited by W. L. G. JOERG. Twenty-four
papers, mostly on the geography of the United States, in English, French, German,
and Italian, with an introduction and a history of the excursion. With numerous
illustrations, including nearly forty portraits of European geographers. xi and
407 pp. $5.00 net.
The Andes of Southern Peru: Geographical Reconnaissance along the Seventy-
Third Meridian. By ISAIAH BOWMAN. With detailed topographic maps in colors,
many black-and-white drawings, and about one hundred illustrations from photo-
graphs. xi and 336 pp. $3.00 net.
The Frontiers of Language and Nationality in Europe. By LEON DOMINIAN. With
8 maps in color and 12 in black-and-white, and numerous photographs. xviii and
375 pp. $3.00 net.
Journal and Bulletin of the American Geographical Society, Vols. 1-47, 1859-1915.
Known as Journal from 1859 to 1886, and as Bulletin from 1887 to 1915 (designated
Journal, however, on title page of bound volumes from 1887 to 1900). The volumes
for the following years are available at $1.00 each: 1871, 1872, 1874, 1876, 1878-
1905 inclusive, 1908, 1915. A set of the Journal and Bulletin (complete except for
five volumes that lack individual numbers) may be obtained for $150.00.
Members' Book, Including the Names of the
Fellows, and Honorary and Corresponding
the Charter of Incorporation. June, 1917.
Officers and Councilors, Gold Medalists,
Members of the Society, together with
Pamphlet of 63 pp.
Maps in Color (Price, 25 cents each)
Physiographic Map of the Carpathians, by EMMANUEL DE MARTONNE. Scale,
1:2,500,000 (40 miles to 1 inch). 1917.
Vegetation Areas of the United States, by FORREST SHREVE. 1:9,600,000 (150 miles
to 1 inch). 1917.
Map of the Frontier Region of Northern Mexico. 1:4,300,000 (68 miles to 1 inch).
1917. Shows relief, temperature, rainfall, drainage, roads, and towns.
The Parliamentary Constituencies of the British Isles and the Results of the
General Elections since 1885, by EDWARD KREHBIEL. 1:2,000,000 (31 miles to 1
inch). 1916.
Map of the Arctic Regions. Prepared in co-operation with the American Museum of
Natural History. 1:6,300,000 (100 miles to 1 inch). 1912. Especially rich in
names.
Map of the Catskill Mountains. 1:190,080 (3 miles to 1 inch). 1907. Generalized
from the topographic sheets of the U. S. Geological Survey.
Map of Eastern Labrador, by MRS. LEONIDAS HUBBARD, JR. 1:1,584,000 (25 miles to
1 inch). 1906. Embodies original survey of the George and Nascaupee Rivers
across eastern Labrador from Ungava Bay to the Atlantic.
Mercator's Map of the World, 1538. Facsimile in black-and-white, reproduced from
the original in the Society's collection (only two original copies of this map are
known). 1885.
Res
L.L.
Regent L. L. Hubbard
что
11-6-1924
THE GEOGRAPHICAL REVIEW
CONTENTS FOR MAY, 1918
Jovan Cvijić. (1 insert
1 insert map in color)
The Geographical Distribution of the Balkan Peoples. By
(I
345
The Islands of Juan Fernandez.
By Carl Skottsberg.
(1 map, 19 photos)
362
The Growth of American Cities. By Lawrence V. Roth.
(3 diagrs.).
384
Some Influences of the Sea upon the Industries of New Eng-
land. By Malcolm Keir.
399
The "Old-Fashioned" Winter of 1917-18. By Charles F.
Brooks
405
Geographical Record
415
American Geographical Society
Presentation of the Cullum Geo-
graphical Medal to Professor
Frederick H. Newell: Meetings
of April .
North America
The Biogeographical Contrast be-
tween the Coastal Plain and
the Piedmont .
Spring Cold Spells in New
England.
The Forests of the Isthmus
of Panama
Europe
·
415
•
416
·
417
•
417
Franco-Italian and Franco-Swiss
Ports in France
The Internal Migrations Under-
lying the Present Distribution
of the Serbo-Croats
Geographical Publications
•
418
Africa
Commandant Tilho's Explora-
tions in the Tibesti Region,
1912-17.
•
419
Rubber Cultivation in East Africa 420
The Meteorological Station on
the Island of Mauritius
420
Educational Geography
Economic Geography in the High
School
•
421
Geographical News
Geographical News.
•
418
Personal
421
422
423
The Society is not responsible for the opinions or the statements of writers in the Review
Published monthly by the American Geographical Society
Broadway at 156th Street, New York, N. Y.
Price, fifty cents a number
Five dollars a year
Entered as second-class matter, January 22, 1916, at the Post Office at New York, N. Y.,
under the act of August 24, 1912
Press of Douglas Taylor & Co., New York
OUR CONTRIBUTORS
Professor Cvijić is the foremost geographer of Serbia. He is professor of geography at the
University of Belgrade. Since the invasion of his native land he has been residing in
Paris, where he is giving courses at the Sorbonne. Professor Cvijić's numerous re-
searches deal with both the physical and the human geography of the Balkan Peninsula.
Among the latter may be mentioned a ten-volume series entitled "The Population of
the Serbian Lands" (Acad. of Sciences, Belgrade, 1902-14), published in collaboration
with his students, which stresses the importance of the internal migrations of the South
Slavs as the key to an understanding of their present distribution and condition (see
p. 418 of this number). Professor Cvijić's name is pronounced Tzviyity—to use the
system of transliteration applied to the map accompanying his article-in which the ty
represents the sound of t in a careful pronunciation of “creature".
Dr. Skottsberg is a Swedish plant geographer of note. He is assistant professor of botany
at the University of Upsala. He has specialized mainly in the phytogeography of
sub-antarctic regions, notably the Straits of Magellan and southernmost Patagonia, for
which his connection with the Swedish Antarctic Expedition of 1901-03 and the
Swedish Magellanic Expedition of 1907-09 furnished the material. Dr. Skottsberg
is the author of numerous scientific papers and a general account of the 1907-09
expedition entitled "The Wilds of Patagonia" (London, 1911).
Dr. Keir is assistant professor of industry at the Wharton School of Finance and Commerce
of the University of Pennsylvania. Dr. Keir is the author of "Fisheries: An Example
of the Attitude Toward Resources" (Bull. Amer. Geogr. Soc., Vol. 44, 1912);
'Modern Korea” (ibid., Vol. 46, 1914); “The Causes for the Growth of Philadelphia
as an Industrial Center" (Bull. Geogr. Soc. of Philadelphia, Vol. 13, 1915); “Some
Responses to Environment in Massachusetts” (ibid., Vol. 15, 1917).
*
Mr. Roth is an assistant in history at Harvard University, where he has been specializing
in the geographical basis of that subject. He has also been giving courses at the State
Normal School of Salem, Mass.
Dr. Brooks is instructor in geography at Yale University and the author of several papers
on meteorology and geography, including "The Snowfall of the United States
(Quart. Journ. Royal Meteorol. Soc., London, 1913); "The Snowfall of the Eastern
United States" (Monthly Weather Rev., Vol. 45, 1915); “New England Snowfall ”
(Geogr. Rev., Vol. 3, 1917); “Island Nantucket” (ibid., Vol. 4, 1917).
ARTICLES ON THE GEOGRAPHY OF THE WAR
published in the Geographical Review
The Great Russian Retreat. By D. W. JOHNSON. (25
pp.; with 6 maps and 6 photographs.)
The Balkan Campaign. By D. W. JOHNSON.
with 1 map and 4 photographs.)
•
February, 1916.
(21 pp.;
The Conquest of Rumania. By D. W. JOHNSON. (19 pp.;
with 1 map and 3 photographs.)
July, 1916.
June, 1917.
In these three articles Professor Johnson discusses the influence of geo-
graphical conditions on the military campaigns in the areas affected. The
second deals with the campaign against Serbia at the end of 1915. Each
article gives an outline of the physical features of the area discussed and
is illustrated by maps showing the positions of the battle lines at various
critical junctures.
The Military Campaigns Against Germany's African
Colonies. By G. M. WRIgley. (22 pp.; with 5 maps.)
The more pronounced control of geographic conditions in the African as
compared with the European campaigns, because of the larger areas affected
and the greater dependency here of man on nature, is clearly brought out
in this article. The campaign in each colony is illustrated by a map indi-
cating the troop movements in relation to the topography. The article is
accompanied by a full bibliography.
The Rôle of Political Boundaries. By D. W. JOHNSON.
(5 pp.)
September, 1917.
A geographical discussion of the boundary problem, as set forth in sev-
eral recent articles.
Europe at Turkey's Door. By LEON DOMINIAN.
with 2 maps.)
(9 pp.;
•
April, 1916.
.
Discusses the claims of the European powers to Turkish territory, based
on political and economic grounds. Illustrated by a separate map showing
the spheres of influence and territorial claims in Asiatic Turkey of Russia,
England, France, Greece, and Italy.
Albania and the Albanians. By H. CHARLES WOODS. (17
pp.; with 1 map and 8 photographs.)
April, 1918.
A discussion of the political geography of this newest Balkan state by
a first-hand observer of the country and its people.
Russia's War-Time Outlets to the Sea. (5 pp.; with 1
map.)
•
February, 1916.
Deals with the closing of the Baltic and Black Sea outlets to Russia by
the war and her circumvention of this obstacle, especially by the construc-
tion of a railroad to Kola, an ice-free port on the Arctic Ocean. With map.
The Economic Resources of the Russian Empire. By E.
K. REYNOLDS. (17 pp.; with 1 map and 10
graphs.)
photo-
•
April, 1916.
Accompanied by a map showing the distribution of the main economic
resources of both European and Asiatic Russia.
Flanders.
By RAOUL BLANCHARD. (17 pp.; with 2 maps
and 9 photographs.)
December, 1917.
A summary of the geography, physical and human, of one of the most
important battle areas of the present war.
(Continued on next page)
WAR ARTICLES IN THE REVIEW-(Continued)
The People of Hungary: Their Work on the Land. By
B. C. WALLIS. (17 pp.; with 8 maps.) .
December, 1917.
A discussion of the human geography of Hungary, with special atten-
tion to the problem of nationalities. Two of the accompanying maps, rep-
resenting by a new method the population density and distribution of
nationalities, are especially valuable.
The Carpathians: Physiographic Features Controlling
Human Geography. By EMMANUEL DE MARTONNE.
(21 pp.; with 2 text-maps, 1 insert map in color, 3
diagrams, and 10 photographs.)
June, 1917.
A general account, illustrated by a colored physiographic map, of the
interrelations of the physical features and human life, many of which have
a bearing on the military operations of the present war in this area.
Poland: The Land and the State. By EUGENIUSZ Romer.
(20 pp.; with 12 maps.) .
•
July, 1917.
A geographical interpretation of Poland's history and development and
an exposition of her essential geographical unity. The article is accom-
panied by ten black-and-white maps illustrating the racial and territorial
development of Poland from 800 to 1795.
The Racial History of the British People. By H. J.
FLEURE. (16 pp.; with 2 maps.) .
March, 1918.
A geographical exposition of the early racial movements which led to the
settlement of Britain and their reflection in the composition of the present-
day population.
Trade Movements and the War. (2 pp.)
Our Trade in the Great War. By MARK JEFFERSON. (7
pp.; with 3 diagrams.)
•
March, 1916.
June, 1917.
The first article points out various fundamental effects of the war on
world trade. The second deals with the trade of the United States before
our entry into the war. By grouping the statistics concerning the chief
countries with which we have trade dealings into four groups, viz. the
Allies, the Central Powers, the Near Neutrals, and the Far Neutrals (with
respect to Germany), the various factors affecting the war trade are more
clearly discerned.
Aëronautical Charts. By OMAR B. WHITAKER.
with 2 maps.)
(5 pp.;
Aëronautical Maps and Aërial Transportation. By HENRY
WOODHOUSE. (22 pp.; with 4 maps, 3 drawings, and 5
photographs.)
•
July, 1917.
November, 1917.
European War Maps. By W. L. G. JOERG. (4 pp.) . July, 1917.
The first two articles discuss a phase of mapping of high importance in
the development of the aeroplane. They are illustrated by reproductions
of aëronautical maps used in this country and in Europe. The second
article also deals with the commercial possibilities of aeronautics in the
immediate future. The third article gives a selected list of maps, detailed
topographic sheets as well as general maps, of use in following the military
operations in Europe.
VOL. V
THE
GEOGRAPHICAL REVIEW
MAY, 1918
No. 5
THE GEOGRAPHICAL DISTRIBUTION OF THE BALKAN
PEOPLES
By JOVAN CVIJIĆ
Professor of Geography, University of Belgrade
[With separate map, Pl. III, facing p. 360.]
The various Balkan peoples are practically limited to distinct natural
regions. The Aegean region is inhabited mainly by Greeks. The inhabit-
ants of the Pindus region and its littoral and of the Epiro-Albanian coast
are in general of Albanian and Aruman (Vlakh) origin. The population
of the continental core and of the Adriatic coast is Yugo- (South) Slav. It
is divided into two groups, the eastern Yugo-Slavs, or Bulgarians, who
occupy the plateau of the lower Danube, the larger part of the Maritza
basin, and the valleys of the Rhodope Mountains, and the western Yugo-
Slavs, or the Serbo-Croats and the Slovenes, who inhabit the central and
western regions of the peninsula. The Yugo-Slavs outside of the peninsula
occupy a part of the Pannonian basin and some valleys of the eastern Alps.
Finally, some of the arid districts, Pontic and Aegean, are inhabited by
Turks.
The peoples of the Aegean and Pindus regions, the Greeks, the Albanians,
and the Arumani, are the most ancient peoples. Their ancestors settled in
the peninsula before the historic era. They may, therefore, in this sense
be considered as the autochthonous inhabitants. The other peoples who
entered later did not establish themselves in the peninsula until the Middle
Ages, the Yugo-Slavs at the beginning, the Turks at the end of this period.
THE PEOPLES OF THE AEGEAN AND PINDUS REGIONS
The Greeks
The Greeks are confined to the Pindus and Aegean regions, with their
Eurasian characteristics, and are closely related, ethnographically and
historically, to the populations of the Aegean parts of Asia Minor. They
Copyright, 1918, by the American Geographical Society of New York
346
THE GEOGRAPHICAL REVIEW
inhabited these regions in the remotest historical period and have not
changed their abode to this day, although their ethnic composition has been
profoundly modified. In spite of the vicissitudes of time and the change
in their ethnic composition they have remained a seafaring people, of the
Mediterranean type of life.
THEIR COMPLEXITY
The Greeks are possibly more complex in origin than any other Balkan
people. The direct descendants of the ancient Hellenes have been preserved
only on some of the Aegean islands and, to a less extent, on the mainland
coast. Numerous "Romans" of Slav and Asiatic origin filtered into the
Greek population of Thrace and Macedonia in the Byzantine period, and
this ethnic process continued during the Turkish period. With the Greeks
of Greece merged a large Slav population, which, at the beginning of the
Middle Ages, penetrated into Epirus, Thessaly, Hellas,¹ and the Pelopon-
nesus and the last remnants of which were not Grecized until after the
fifteenth century. The Greeks also absorbed important groups of the
ancient Thraco-Illyrian population in Epirus, Macedonia, and Thrace.
Thessaly from the twelfth to the fifteenth century was called Great Wal-
lachia and was inhabited mainly by Arumani. They have been described
by the Jewish traveler, Benjamin of Tudela (1173). During the same
period Aetolia was called Little Wallachia. The majority of these Aru-
mani were Grecized. The Grecization continues to our day in Thessaly
and in southern Macedonia. But beginning with the fourteenth century
the most important process of assimilation was the Grecization of the
Albanians in Greece-in the Peloponnesus, in Hellas, in Epirus, and in
southern Albania. Particularly the "metanastasic' current of the Tosks
discharged upon Greece. Even today, according to Philippson, there are
in Greece, within the boundaries existing prior to 1912, 250,000 Albanians
who are only superficially Grecized.
PREDOMINANCE OF HELLENIC CULTURE
In this ethnic conglomerate, surcharged with foreign elements, the Greek
language and, apparently, the principal traits of ancient Greek mentality
have nevertheless predominated. There is reason to believe that this re-
sult was only rarely achieved through the numerical superiority of the
Greeks. The influence of geographical environment and particularly of
1 In the restricted sense of Middle Greece, i. e. the elongated area lying between two depressions, the
northern occupied by the Gulf of Arta, the Spercheios River, and the Euboean Sound, the southern by the
Gulfs of Patras, Corinth, and Aegina.-EDIT. NOTE.
2 The author suggests this term (derived from the Greek µɛraváσ7ɑoç, meaning "change of habitat")
to designate the migrations of Balkan peoples caused by the Turkish invasion at the end of the fourteenth
century and continued during the four centuries of Turkish rule, as distinguished from the migrations
brought about by the great Slav invasion of the sixth century and the movements which resulted from
the rise of the Balkan states during the period of Slav dominion. See the abstract of his paper on this
topic under Geographical Record" in this number.-EDIT. NOTE.
DISTRIBUTION OF BALKAN PEOPLES
347
the manner of life have contributed much to its development. All the
immigrant peoples, the Slavs, Albanians, Arumani, came into the Aegean
region, with its Greco-Mediterranean type of life, from a different geo-
graphical environment. They imitated Greek ways and customs and
adapted themselves to the Greek population, although it was not more
numerous than the newcomers.
NORTHERN RACIAL LIMIT
The northern boundary of the Greek people was approximately defined
by the first explorers of the first decades of the nineteenth century, espe-
cially the section between Saloniki and Constantinople. Along this line
as well as the less known section between Saloniki and Kastoria, the Greeks
come into contact with the Slavs. This is a frontier of contrasting lan-
guages, physical habitus, mental characteristics, and manner of life. For
centuries this has been the most stable racial boundary of the peninsula.
Nevertheless it did not remain unchanged, particularly in the basin of
the Maritza River and on the coast of the Black Sea. Before the founda-
tion of the Bulgarian state in 1878, the Greeks inhabited in compact masses.
the valley of the Maritza as far as Adrianople as well as secluded areas
north of the Rhodope Mountains, in which region they have maintained
themselves to the present only at Stanimaka. Greeks also predominated
on the Pontic littoral south of Baltchik as far as the Bosporus. During
the last thirty or forty years the Greeks have emigrated or been expelled
in large numbers from Bulgaria, and only insignificant groups are now
left. Even on the Thraco-Macedonian coast the Greek population is not
compact. It is interrupted by large groups of Turks, by enclaves of
Bulgarians, and even by two enclaves of Serbs, those of Bayramity and
Karadja Alil, of whom the majority migrated at the beginning of 1913
to the environs of Skoplye.
Between Saloniki and the Adriatic Sea the northern limit of the Greeks
was always less stable. In this direction Hellenism has spread to a con-
siderable extent by assimilating the Tosks and numerous Arumani. Fur-
thermore the influence of the Greek-Orthodox church and of the Greek
language and civilization makes itself felt even outside of the Grecized
area, among the Tosks of southern Albania.
BOUNDARY BETWEEN GREEKS AND MACEDONIAN SLAVS
The ethnic boundary between the Greeks and the Slavs in southern
Macedonia may be indicated more precisely. If one proceeds from Epirus
and southern Albania along the Yanina-Lyaskovik-Kortcha road to the
Prespa basin, no Slavs are met with as far as Prevtis Pass and Ivan Moun-
tain, except for three villages in the Kortcha basin the inhabitants of which
speak an archaic Slav dialect. This pass and mountain constitute a sharply
defined boundary between the more or less Grecized Tosks and the Slavs.
348
THE GEOGRAPHICAL REVIEW
Farther east, in going from the basin of the Bistritza by way of Kastoria
into Macedonia the Slav population lies farther south. Kastoria has the
aspect of a Greek city, with a Turkish minority and the Slavs semi-
Grecized; but in the environs of Kastoria the Slav population predominates ;
an important section of it in the vicinity of Nesram is Moslemized. Side by
side with the Slav villages are to be found Greek, Aruman, Turkish, and
Albanian villages. Immediately southwest of Lake Ostrovo, in the district
of Sari Göl, the boundary separates the Greeks from the Turks, who in-
habit the environs of Kailar. It coincides practically with the longitudinal
axis of Lake Ostrovo, although even to the north of the lake groups of
Turks exist. There are besides in the northern part of the Sari Göl district
some Slav villages, mainly the large villages of Patelik, Vrbeni (Ekshisu),
and Sorovitchevo, which are connected with the compact Slav mass which
commences north of the pass across the Malka Nidje in the basin of
Bitolye (Monastir).
The Turks
There is a considerable difference between the distribution of the Turks
before the beginning of the Ottoman decline at the end of the seventeenth
century and their present distribution in the peninsula.
During the first period from the fourteenth to the seventeenth century,
the Turks constituted the majority of the population in nearly all the
cities of the peninsula lying along the main longitudinal highways. Out-
side of these highways they formed an important part of the population
of every city except in Dalmatia. There was also a numerous Turkish
population beyond the peninsula, in the cities of Hungary. Furthermore,
in addition to the rural Turkish population in the southeastern part of the
peninsula, the Turks began to found villages in other regions. Large num-
bers of the Turkish nomads, the Yürüks, natives of Asia Minor, established
themselves with their cattle in the central and eastern parts of the Balkan
Range, in the Rhodope Mountains, the Pirin Range, and in the mountains.
of Macedonia. The topographical names of these mountainous regions are
Turkish, and this is the only trace left by the Yürüks, who, together with
the Turkish population of the northern cities, have disappeared.
TURKISH ENCLAVES
The large bodies of rural Turkish population were always limited to the
Thraco-Macedonian region with its Eurasian characteristics, and to the
Pontic regions. Both these regions have a dry climate resembling that
of the homeland of the Turks in Asia Minor. Even these large islands
of Turkish population continually grew smaller, especially after the middle.
of the nineteenth century, in spite of the wealth of the muhadjiri. Consti-
tuting alone the natio militans, the Turks became reduced in the constant
DISTRIBUTION OF BALKAN PEOPLES
319
wars, and, furthermore, the armies and the cities alike were often deci-
mated by contagious diseases.
These large islands of Turkish population are as follows: the group of
eastern Bulgaria; the Thracian group, extending from Constantinople to
Lake Takhino, near Seres; the Vardar group, extending in an almost un-
interrupted strip from the Gulf of Orfano along the left bank of the Var-
dar as far as Skoplye; and the enclave of Kailar, where the Turks inhabit.
two small cities, Kailar and Djuma, and about a hundred small villages.
Such of these villages as are situated on the margin of the fertile basin of
Budjak are distinguished by a high degree of prosperity, a condition rare
among the Turkish population of the peninsula.
In eastern Macedonia, between Seres and Drama, there is found a popu-
lation of Greek-Orthodox Turks (4,500 inhabitants), followers of the Greek
patriarchate. These Turks have Greek schools and often speak both Turk-
ish and Greek. Their center is the hamlet of Zilyahovo (2,800 inhabitants).
In addition to these main islands and the other small ones which are
indicated on the accompanying map, the Turks still constitute an important
part of the population in all the cities lying outside of the limits of the
independent states of the peninsula as constituted prior to 1912.
The Albanians
In contrast with the Greeks, the Albanians, although occupying the
Epiro-Albanian coast since remotest antiquity, have not profited from their
contact with the sea. They are neither sailors nor fishermen. They have
practically no maritime intercourse along their own coast, nor with the
countries across the sea. Although the belt of Mediterranean climate and
vegetation along their coast is rather wide, the Albanians have little inclina-
tion for the Mediterranean manner of life and occupation. They have with-
drawn from the coast, which is often marshy, to the hills near by and the
mountains and valleys of the interior. They devote themselves principally
to cattle-breeding, less to agriculture. They are almost strangers to their
own coast. They inhabit mainly the mountainous ramparts of the Pindus
and the southern slopes of the Prokletiye (the North Albanian Alps).
Those who live to the north of the Shkumbi River, the Ghegs, are, by their
occupations and mode of life, related to the populations of the continental
core.
Even the Tosks, south of the Shkumbi, although influenced by the
Arumani and superficially affected by Greek civilization, are more closely
related to the continental populations of the peninsula than to those of the
Aegean regions. The majority of the Albanians adopted Islam after the
Turkish rule began, and these Albanians have been supporters of the Turk-
ish régime and the worst oppressors of the other races. The Greek-Orthodox
Albanians in the south and the Catholic Albanians in the north, particularly
the latter, constitute only an insignificant minority.
350
THE GEOGRAPHICAL REVIEW
THEIR FOREBEARS, THE ILLYRIANS
The Albanians, as is well known, are the ancient Illyrians, Latinized to
some extent during Roman rule. In addition they contain a certain Slav
element, introduced especially during the Middle Ages.
Before the Yugo-Slav invasion the Illyrians occupied the western part
of the peninsula from the middle Danube to Epirus and the central regions.
A powerful Illyrian tribe, the Liburni, inhabited the Adriatic coast and
islands; another, the Dalmatians, occupied the region between Spalato and
Zhupanyatz in Bosnia. Around Lake Scutari were the well-known Labeati;
their neighbors to the south called themselves the Prussites and were en-
gaged in mining. In the neighborhood of Kroya, in the center of present
Albania, lived the Albani. In the center of the peninsula, in the vicinity
of Nish, in the Kossovo and Metohiya districts, and about the upper Vardar
were to be found the Dardani; farther to the south the Peoni. The coun-
try of the Dessaretes lay in the region of Lake Okhrida. The Epirotes and
the Macedonians were also Illyrians. Little is known about the Illyrian
tribes. They were practically independent and were divided into numer-
ous clans; the Dalmatians, for instance, were divided into 342 clans. Re-
doubtable warriors, they offered desperate resistance to the Roman occupa-
tion of their country, as witness the conflicts between the Romans and
Queen Teuta of Scutari and also the insurrection of the Illyrians of Dalma-
tia and Pannonia, who raised against the Romans an army of 200,000 foot
soldiers and 8,000 horsemen.
SLAVICIZATION OF THE ALBANIANS
:
All the other Illyrian tribes, except those of the Albania of today, were
Slavicized during the Middle Ages. It appears that a part of the Illyrian
population withdrew from the central and northern districts of the penin-
sula to Albania. But even in this region and in Epirus at the end of the
tenth and the beginning of the eleventh century the valleys and plains were
inhabited by the Slavs, while the Albanians and the Arumani occupied only
the mountains. At the time of the reign of the House of Anjou (1250-1350)
there was still a Slav population in the coastal plains and about the Drin
River. This population, according to the studies of K. Jireček, was consid-
erably reinforced by immigrants from Serbia during the Serbian rule, par-
ticularly in the fourteenth century. In the heart of Albania today one may
still meet the last survivors of this semi-Albanized Serbian population.
According to their traditions the tribes of northern Albania are of mixed
origin, Albano-Serb. They consider themselves related to the Serbian
tribes of Montenegro. The national hero of the Albanians, Skanderbeg,
was, as is well known, of Serbo-Albanian origin. Just as the Yugo-Slavs
assimilated numerous Illyrians, from the Danube to southern Macedonia,
they were themselves Albanized in Albania, particularly during the Turkish
period.
DISTRIBUTION OF BALKAN PEOPLES
351
THE EASTERN BOUNDARY OF THE ALBANIANS
As a consequence of this process of assimilation the eastern boundary
of the Albanians, or the ethnographic boundary between the Albanians and
Serbs, has undergone important displacements, which it is necessary to trace.
4
There existed, even during the Turkish period, west of Lake Okhrida a
scattered Slav population extending as far as the vicinity of Elbasan. Of
this type were the present Albanized Slavs, who live in the Mokra district,
the headwater valley of the Shkumbi. The Opara region on the left bank
of the Devol River was inhabited by Slavs in the sixteenth century—“il
paese d'Opari, ch'è habita de Schiavoni," wrote Musachi, an Albanian, in
1510. There are Slav families who migrated from the Opara region into.
the Monastir basin. In the Kozyatchia region along the upper Pchinya
northeast of Skoplye are to be found Serbian families originating from the
vicinity of Elbasan. Lejean has called attention³ to small enclaves of Ser-
bian population on the Albanian coast near the mouth of the Viosa River,
on the Arzen River, and to the north of Durazzo. To the west of the Opara
region and of Elbasan lies the Musakiya region, around the mouth of the
Semeni-to the south of this river, Lesser Musakiya, to the north Greater
Musakiya, inhabited by the Tosks, known as Bapidi and Farsherioti, an
Aruman population whose occupation is agriculture and cattle breeding.
The geographical names of the Musakiya region, particularly of Lesser
Musakiya, are Slav. Numerous Slav herdsmen from Dibra were in the habit
of wintering in the Musakiya region only three decades ago. It is from
them or from the ancient Slav population of this region that this nomencla-
ture is derived.
REGIONS OF SLAV PREDOMINANCE
While the Slav population west of Lake Okhrida has disappeared, either
by assimilation or emigration, the Slavs have maintained themselves north
of the lake on the left bank of the Black Drin River. In this area, the
Drinkol district, there are large villages of Serbs (Labunishte, Vehtchane,
Borovatz, Yablanitza), or of Moslemized Serbs (Trebishte). Albanians and
Slavs dovetail, especially north of the gorge of the Black Drin, i. e. north
of Dabovlyane. The largest island of Serbian population is that of the
Golobrdo district (between the headwaters of the Shkumbi and the Matya),
a population of archaic character and partially Moslemized. To the south-
west and north of the Golobrdo district, in the Tchermenika and Bultchizi
districts and the valley of the Matya, one meets with numerous Albanians
of Serbian origin who have conserved Serbian and Christian customs. Sev-
eral Serbian families on the other (right) bank of the Drin, in the Mala
3 For locational details not entered on the accompanying map, especially district names, consult the
sheets of the map of the Balkans, 1:250,000, published by the Geographical Section of the General Staff, War
Office, London.-EDIT. NOTE.
4 Charles Hopf: Chroniques gréco-romaines, Berlin, 1873, p. 280.
5 G. Lejean: Ethnographie de la Turquie d'Europe, Ergänzungsheft No. 4 zu Petermanns Mitt., with
ethnographic map, 1:2,500,000, Gotha, 1861.
352
THE GEOGRAPHICAL REVIEW
Reka district (basin of the Radika River), originate from these three re-
gions of Albania. They maintain relations with their Moslemized and Al-
banized parent stocks, visit one another during festivals, and practice the
vendetta among themselves.
The Golobrdo region, west of, and the monastery of St. Yovan Bigorski,
east of the Drin, in the valley of the Radika, have served as a shelter against
Albanian penetration. To the south and north Albanians have advanced
farther into Macedonia and Old Serbia (see the map).
DIBRA AND ITS NEIGHBORHOOD
The city of Dibra is inhabited by numerous beys of Turkish and Albanian
origin, the most powerful in Macedonia. Its life is very much "à la turca."
Every Slav or Aruman petchalbar was formerly obliged to render tribute.
to the beys in the shape of furniture, harness, jewelry, etc. Severest perse-
cution was practiced against the Christians of the city and of the whole
region. There are Turks even in the villages, particularly in the zhupa of
Dibra. Through these Turks part of the tribe of the Miyatzi were Moslem-
ized, but in contrast with all the Moslemized inhabitants of these regions
they have not become Albanized. Like the Moslemized inhabitants of
Bosnia and the former sandjak of Novibazar, they speak Serbian. To the
north of their district there is to be met with, in the gorge of the Radika,
a curious population, Christian and Serbian, which is ignorant of Serbian
and speaks only Albanian and is furthermore filled with hate against the
Moslemized Albanians. The geographical names of this district are often
Serbian, and the Albanian spoken in it contains a number of Serbian words.
This population seems to represent a Serb-Aruman amalgam, having con-
served its religion but forgotten its language as a consequence of the prox-
imity of the Albanians and of some villages of Albanian immigrants.
THE LYUMA REGION
Along the eastern side of the valley of the Black Drin and of the White
Drin from Pishkopeya (44° 43′ N.) by way of Bitzan and Lyum-Kula as
far as the barrier Zhur to the west of Prizren extends the Lyuma region.
Its geographical nomenclature is generally Serbian. In the seventeenth
century it was still occupied by a Serbian population, into which penetrated
the Albanians of Albania, so that it became Albanized. In the zhupa of the
Shar Planina and in the Opolye, immediately south of Prizren, Albani-
zation took place only in the eighteenth and nineteenth centuries. The
most recent Albanization has taken place in the zhupa of Gora, whose
inhabitants still speak Serbian. There are families some members of which
are Greek-Orthodox and Serbian and others Moslem and Albanian.
THE NORTHERN BOUNDARY OF THE ALBANIANS
The part of the peninsula which has undergone the greatest ethnic
changes is the belt lying between Scutari and the Southern Morava at
DISTRIBUTION OF BALKAN PEOPLES
353
Leskovatz. The region between Scutari and Prizren (Pilot and Reka) was
occupied in the Middle Ages by a mixed population, Serbian and Albanian.
Serbs predominated in the basins and valleys, Albanians in the mountains.
Scutari and the plains surrounding the Boyana and the Drin, the Zadrin
and the Zaboyana districts, were occupied by Serbs; likewise the Metohiya
and Kossovo districts. After the Turkish invasion the Albanians came
down from the mountains, a part of the Serbians emigrated, and the rest
were Moslemized and Albanized. The Serbian emigrants called themselves
Skadrani. They are to be found in the region of the Drina in Serbia, in
the Sredska zhupa of Prizren, in the cities of the Kossovo and Metohiya
districts and of the valley of the Southern Morava.
FARTHER MOST ALBANIAN ENCLAVES
After the great Serbian migrations from the Metohiya, Kossovo, and
the Southern Morava districts an immense expansion of the Albanians
took place. The farthermost islands of Albanian population, some 300
kilometers from their place of origin, were those to the west of Leskovatz,
in the Masuritza region, to the north of Vranye, and in the Peshter region
south of Syenitza. They have been indicated on all the older ethnographic
maps. They existed until 1878. At this time, Serbia having occupied the
region of the Southern Morava, these Albanians emigrated, mainly to the
valley of the Lab (north of Prishtina), the Kossovo region, and the
Metohiya region, reinforcing the Albanian population already there. The
Albanian islands in the vicinity of Leskovatz and of Vranye exist no longer,
while the Albanian island of the Peshter district has survived the occupa-
tion of 1912. About the year 1700 the Klimentis established themselves in
the vicinity of Lake Scutari and exchanged Catholicism for Islam. They
speak Albanian and Serbian.
Regions occupied by a mixed Serbo-Albanian population are the Meto-
hiya and Kossovo.
The Arumani (or Vlakhs)
The Arumani are a disappearing people. Today there are not more than
154 Aruman communities, with 150.000 to 160,000 scattered individuals,
chiefly in the southern part of the rampart of the Pindus and in the central
and southern parts of the Balkan Peninsula. They may almost be termed
an uprooted people. They occupy no large area of their own. Even where
small groups of them occur, they do not cultivate the soil-they are not
agriculturists—and they are only slightly connected with the geographical
environment in which they live. They are above all the wandering shep-
6 On this relatively little-known people consult also A. J. B. Wace and M. S. Thompson: The Nomads
of the Balkans: An Account of the Life and Customs Among the Vlachs of Northern Pindus, New York,
1913 (reviewed in Geogr. Rev., Vol. 1, 1916, p. 319). The spelling used in this article is that given by these
authors as the native form (p. 2, and vocabulary, p. 307). Professor Cvijić, in his manuscript and in the
paper cited in footnote 8, uses the form Aromuni.-EDIT. Note.
354
THE GEOGRAPHICAL REVIEW
herds of the peninsula, veritable nomads who migrate from the highest
Balkan mountains to the coast and back again. Those who settle down in
villages do not stay there long. Their villages are not like ordinary villages.
Around them there are no cultivated fields, not even vegetable gardens, nor
cattle. They only serve as shelter for the women and children, to provide
for whose living the men scatter throughout the cities of the peninsula.
THEIR HISTORY
The Arumani are the last descendants of the semi-Latinized Balkan
populations. They are undoubtedly more Latinized than the Albanians;
and, furthermore, they seem to differ from the Albanians particularly
in being, together with the Greeks of Thrace and Macedonia, the most
Byzantinized group of the peninsula.
During the Middle Ages the Arumani were still very numerous in the
mountains and the secluded regions of the peninsula. They constituted, as
has been indicated, the majority of the population in Thessaly and in
Aetolia, or Great and Little Wallachia (Vlakhia). White Wallachia was
a well-known designation in the Balkans and north of the Balkan Range;
it was the Vlakhs of this region who, with the Bulgarians, founded the
Vlakho-Bulgarian kingdom under the dynasty of Asen in the twelfth cen-
tury. During the same period the name Black Wallachia was current in
what is now Moldavia. In the eleventh century the Arumani inhabited a
part of the Dobrudja and the vicinity of Anchialos on the Gulf of Burgas.
They were known in Thrace in the thirteenth century. In Serbian docu-
ments of the Middle Ages the presence of Vlasi (Vlakhs, Arumani) in the
mountainous regions to the southwest of the Shumadiya as well as in the
vicinity of Prizren is often referred to. However, it is not always possible
to decide whether the name Vlasi was used to designate the real Arumani
or, because of their pastoral occupation, the Serbian shepherds.
The Arumani have become Slavicized and Grecized and have completely
disappeared from the villages or entire districts where they were still to
be found two centuries or even one century ago. Several explorers have
stated that at the end of the eighteenth century the number of Arumani in
the central and Pindus regions of the peninsula amounted to between
400,000 and 500,000. It is a curious fact that they became Moslemized only
very rarely. For instance, the Arumani of Meglen in eastern Macedonia
first amalgamated with the immigrant Petchenegs and did not become
Moslemized for 200 years or more.
THEIR PRESENT CONDITION
Of the continually diminishing 150,000 to 160,000 Arumani who have
survived to this day a certain proportion are wandering shepherds, devot-
ing themselves to the raising of sheep and living during the summer in
temporary dwellings, called kalivas. They are to be found scattered
DISTRIBUTION OF BALKAN PEOPLES
355
throughout the mountains of the peninsula from the Stara Planina in
Serbia and Bulgaria to the mountains of the Peloponnesus. There they
spend the summer without having to prepare fodder for the winter, inas-
much as they come down from the mountains with their herds to pass the
severe season of the year on the coast of the Aegean or-a few-on the
Adriatic. For various reasons the Aruman shepherds are abandoning
cattle breeding and are increasing the numbers of the semi-sedentary type
of their fellows.
These Arumani have lived for centuries in large compact settlements,
comparable to cities, which are always located in sheltered positions and
at elevations of at least 800 to 1,000 meters. It is especially these Arumani
who have no connection with the ground on which are built their houses.
As a regular thing they live away from their homes, extra-territorially, so
to speak, in the cities or villages of the surrounding region, where they go
to earn their living and make money. That is the petchalba of the Arumani.
Only rarely do they go beyond the limits of the Balkan Peninsula.
In all the cities of the central and eastern parts of the peninsula there
is to be found a third kind of Arumani. These have permanently settled
in the cities, particularly along the longitudinal routes, where they practice
various trades for the time being, often as keeper of a café or in some
commercial occupation.
THE PEOPLES OF THE CONTINENTAL CORE
The Yugo-Slavs
The Yugo-Slavs (i. e. Southern Slavs) occupy the continental core in its
entirety. While no people of the Aegean and Pindus regions are agri-
culturists, the Yugo-Slavs of the continental core are true agriculturists,
a people habituated to the plow. They are often also horticulturists. How-
ever, there are mountainous regions where cattle raising predominates over
agriculture. The Adriatic group has adapted itself completely to Medi-
terranean and maritime occupations. The Yugo-Slavs are a peasant people.
They devote themselves principally to agriculture and live almost exclu-
sively in villages. Their country contrasts with the Aegean regions, where
the cities are more numerous, where the villages often have the aspect of
cities, and where the differences between the urban and rural population
are much less than in the continental core.
DIFFERENTIATIONS
The Yugo-Slavs occupy not only the largest area of the peninsula, but,
with their co-nationals, they are also very numerous in Austria-Hungary.
They number about 16,000,000, but do not form a compact population. A
differentiation took place between the eastern Yugo-Slavs, or Bulgarians,
who today number some 4,000,000 to 5,000,000, and the western Yugo-Slavs,
356
THE GEOGRAPHICAL REVIEW
who number more than 11,000,000, namely 7,000,000 Serbians and 4,000,000
Croats and Slovenes. The first differentiations that took place were due to
the topography, which is considerably different in the cast (the plateau of
the lower Danube and the large and uniform basin of the Maritza) from
what it is in the central and western districts, particularly in the Dinaric
and Karst regions. To these should be added the differences in racial com-
position. In the east it was mainly the Thracians who were assimilated
by the Slav and Bulgarian tribes. In the west it was the Illyrian tribes and
possibly remnants of the Celts who merged with the Slavs. A very impor-
tant event was the invasion of the Bulgarians and the union of these Ugro-
Finnic elements with the Slavs who inhabited the plateau of the lower
Danube. This Slav-Bulgar amalgam assimilated large numbers of Cumans
and Petchenegs.
The differentiation of the Yugo-Slavs was accentuated by the formation.
of the Serbian, Croatian, and Bulgarian states in the Middle Ages. The
differences among these Yugo-Slavs began to disappear during the long
Turkish domination. In addition during this period two developments.
took place which were of great racial import.
EVOLUTION OF SERBIAN NATIONAL CONSCIOUSNESS
After the battle of Kossovo and after the successive annihilation of the
Serbian states a strong Serbian national consciousness developed, which,
mainly by reason of the migrations, spread even beyond the areas included
in the medieval Serbian states. Apparently this national consciousness
and definite national aims developed at the end of the fourteenth and dur-
ing the fifteenth century in the region extending from the Kossovo district
to the Narenta on the north and from the Adriatic Sea to the eastern bor-
ders of the Shumadiya. From there it spread to other regions inhabited
by the Serbs. As if by a systematic education all the Serbs became imbued
with an elevated national sentiment and with the feeling of an obligation
to fight for liberty, justice, and the restoration of their independent state.
Through this community of sentiments and ideas the Serbs became closely
knit together, passionately devoted to their national ideal.
THE SERBIAN CHURCH
The Greek-Orthodox faith and Serbian nationality, the merging of which
began at the time of St. Sava at the end of the twelfth century, fused com-
pletely at the end of the fifteenth century and from then on became insep-
arable. The Greek-Orthodox faith almost lost its dogmatic and ecclesiastical
character, and, assuming almost a racial character, it became the Serbian
church par excellence. The Church was almost as integral a part of the
national character as the folklore. The Greek-Orthodox faith found a pow-
erful organ in the Serbian patriarchate of Pety, which was reinstated in
1557 by the vizier Sokolovity, a Moslemized Serb from Bosnia. The popes
DISTRIBUTION OF BALKAN PEOPLES
357
and priests identified themselves completely with the people. These were
not only their flock; they were their people, body and soul. The clergy wore
only the national costume, except during religious services. They consid-
ered themselves simply a part of the nation, fulfilling its religious functions.
The bishops and their emissaries visited frequently all the regions belonging
to the patriarchate of Pety, from the Save and the Danube as far as Skoplye
and Küstendil, and contributed to strengthening the national hopes. Around
the churches numerous printing establishmets grew up from the sixteenth
century on. The patriarchate restored numerous churches and ruined mon-
asteries. Popular meetings, the sabors, which were held about these
churches, became institutions where the national ideas and the program for
the future were developed still further. The liberation of the nation became
the supreme object of the "Serbian Church." The national motto was ex-
pressed in a few simple and clear words—"An unceasing fight for the sacred
cross and for national liberty."
The Church having become the organ of the thought of the nation, no
great migration took place without the popes. In emigrating with the peo-
ple they took with them across the mountains and beyond the Save and
Danube the ecclesiastical books, the sacred objects of the Church, and the
relics of the saints, not out of bigotry, but because they considered these
objects as national relics.
As a consequence of these migrations the Serbs have transferred and
transmitted by contact this elevated national consciousness and numerous
intense characteristics to the Yugo-Slav population living outside of the
peninsula, who, incidentally, were more advanced in material civilization.
But inasmuch as the "metanastasic” currents were directed toward the
north this new evolution of the national consciousness, which was accom-
panied by an evolution of the language, has hardly affected the southern
parts of the peninsula.
EVOLUTION OF THE BULGARIANS
An entirely opposite process took place among the Bulgarians, in the
regions of the plateau of the lower Danube and the Maritza basin. During
the Turkish rule the Bulgarians bowed their heads, submitted completely,
and became the most abject raya of the peninsula. To say nothing of their
mental traits, this complete submission was facilitated by the fact that their
country was nearer to Constantinople and that its flat and unforested char-
acter favored the development of an economic régime of fiefs, the tchifliks
of the Ottoman beys. This was the part of the peninsula where the régime
of the tchifliks was most severely applied. All Bulgarians became the raya,
cultivators of the soil for the beys. The name "Bulgarian" lost its national
significance and was used only in the sense of raya, laborer, peasant. It is
said that even as late as the middle of the nineteenth century Bulgarian
city dwellers and the more educated Bulgarians were ashamed of this name,
358
THE GEOGRAPHICAL REVIEW
and that it is only since the creation of the exarchate and the liberation of
Bulgaria that the name Bulgarian has begun to assume a national meaning.
SIGNIFICANCE OF THE NAME
During the Turkish period the name Bulgarian in the sense of raya and
as connoting servility began to spread even beyond Bulgarian territory,
among the whole population of the kmets, the raya cultivating the estates
of the beys. The region in which this strict economic régime was carried out
extended as far as Nish. The name Bulgarian was applied by the Turkish
officials and by the Greeks to this whole population without distinction of
nationality, to the population of the regions forming the center of the Serb-
ian state of the Nemanityi dynasty, which was subjected to the régime of
the tchifliks and which was the most oppressed. In this sense the name of
bugar (Bulgar) was used in the Vardar region and began to spread to the
Kossovo and Metohiya districts. A Russian traveler of the seventeenth
century even applied the name Bulgarian to the kmets of the neighborhood
of Sarayevo in Bosnia. In the extreme western part of the peninsula, in Dal-
matia and Croatia, as well as in the part of Serbia along the Morava River,
the term Bulgarian had no other meaning than that of a rustic and common
fellow, and the people of these regions call the popular songs bulgarstitze
(Bulgarian songs).7
The name Bulgarian, in the sense of an inferior and boorish social class,
thus spread during the course of the Turkish period beyond the areas
of Bulgarian nationality. Under the hard economic régime and Turkish
oppression the population of these regions almost completely forgot their
historic traditions. Their national consciousness did not develop as in the
Dinaric regions.
SERBO-BULGARIAN TRANSITION
There are especially two regions lacking in national consciousness,
although they have preserved some traces of historical Serbian traditions:
Macedonia proper, and the region of the Shopi, a people who constitute
from the ethnographic viewpoint the transition between Serbians and
Bulgarians. The Shopi of a part of western Bulgaria, particularly those
of the regions of Trn, Bryeznik, and Kula, as well as the population of
central Macedonia, are doubtless more closely related to the Serbians than
to the Bulgarians. The writer does not wish to make this general discussion
polemical and therefore refers the reader who is desirous of greater detail
to his earlier publications.
CONFUSION OF THE ISSUE
It is because of the extension of the name Bulgarian in the sense outlined
that some earlier travelers and explorers, unfamiliar with the Yugo-Slav
7 For more details see J. Cvijić: Questions balkaniques, Attinger, Paris and Neuchâtel, 1916.
8 More especially: Die ethnographische Abgrenzung der Völker auf der Balkanhalbinsel, with map.
1:1,500,000, Petermanns Mitteilungen, Vol. 59, 1913, Part 1, pp. 113-118; 185-189; 244-246.
<

Varazhdin
2
Klagenfurt
zelovetz
The American Geographical Society of New York
12
46
46
Piave R
A
Villach (Belyako
Lamento
Taglias
Udineo
(Sot
6Gorizia
(Goritza)
THIA
e
Marburg
(Maribor)
STYRIA
STRI A
AR
Lyublyana
NO
Laibach)
OL
VENICE
[VENEZIA]
TRIESTE
(TRST)
Fiume
1STRIA Adiyeka)
Scilli
Tzely)
Gottschee
Karlovatz
(Karlstadt)
A
-LO
45
44
43
I
42
41
40
40
KRK
VEGIJA
Pola
TZRES
(Puly)
(CHERSO)
D
R
PAG
(PAGO)
B
L. Balatons
19
S
B
H
Baya
Petch
Mohatch
BARANYA
20
21
22
23
24
25
Bekesh Tchaba
26
Kolozhyar
27
28
The Geographical. Review, Vol. V, No. 5, 1918, Pl. III
29
(Khuzh)
Szabadka o
(Subotica)
N
Szeged
Ma
Arad
Torda
Liver
Marosh Vasharhely
Bacau
30
31
ODESSA O
M
o Zombor
BATCH KA
Senta
B
T
Veliki Kikinda
Temeshvar
T
R
T
R
A
NY S
72
V A
Schässburg
(Shegeshvar)
N
Tergu Ocna
MO
Alba Iulia
(Karls burg)
lugosh
Sibiu o
(Hermannstadt)
Palanka
Novi Sad
Neusatz)
4
Tam
BELGRADE
BEOGRAD
Vershetz
(Vrshatz)
Bela Tzrkva o
www
Sh
Valyovo
Uzhitze
Western
a
Mora
Mor
Karanshebesh
Krashovao
L
R
A N
NS
(Biserica Alba
Targu Zhiu
Pozharevatz
R
Turnu Severin
U
W
L
RS
Fogarash
A IN
ANIA
VAN
Campulung
Kronstadt
(Brashov)
Siretu Sere
Berlad
VIA
Prutu pruth
RUSSA
Akkerman
46
BESSARA BIA
Bolgrall
Fokshani o
Kiliya
Reni
Galatz
P
Ismail
Braila
Tultehea
Buzeu
Sulina
45
Ploeshti
Piteshti
M
Slatina
L
Craiova
22720
Ardjesku
N
H
BUCHAREST
OBUKURESHT)
I A
B
Tchernavoda
S
Constantza
44
Kraguyevatz
Vidin
Kula
Silistra
Tutrakan
R
Djurdjiu Ruse
Rushtchuk)
a
Soult
K
L
A
Essegg
(Osyek)
Pozhego
Banyaluka
Tuzla
B
Yaytze
S
Travnik
N
Bihaty
• Gospity
MTS
Zadar
Zara) D
(80)
A
Klyutcho
DINARIO
Shibenik
Sebenico)
L
Vrbas
Osiny
Sply let
Spalato)
I
A
T
C
T
A
L
39-
38
d.
37
36-
B
Zhupanya tz)
Sarayevo
ALPÍS
VIS
(LISSA
BRATCH
(BRAZZA)
KHVAR (LISSINA)
KORTCHULA
(CURZOLA)
Y
S
Mostard
HERZEGOVINA
MLYET
(MELEDA)
R
Syenitza
6 Novibazar
Laft
Mitrovitza
B
Metkovity
Pety o
MONTENEGR
Nikshitch
Trebinys
Dubrovnik
(Ragusa)
olzeligye
Kotor
(Cattaro)
(Cettigne
Scatar
Susinye
NORTH ALBANIAN
Shkoder
Soutari)
Metohiya
Dyakova
Drin R.
White
Drin R
E
A
ETHNOGRAPHIC MAP
OF THE
BALKAN PENINSULA
BY
JOVAN CVIJIĆ
Scale 1: 3 000 000
Yugo (i.e. Southern) Slavs
14
25
25
50
775
100miles
Slovenes
Catholic
Serbo-
Croats
Greek-Orthodox
Moslemized
Albanized
Macedonian Greek-Orthodox
Slavs
Moslemized
Serbo-Bulgarian transition area
Greeks
Greek-Orthodox
Moslemized
Rumanians and Arumani (Vlakhs)
Moslemized Arumani
Turks and Moslem
Tatars
Greek Orthodox
Bulgarians
Greek-Orthodox
Moslemized (Pomaks)
Albanians
Czechs
Slovaks
Russians
Italians
Magyars
Germans
Black
Bar
(Antivari
Ulciny
(Dulcigno)
Boyana.
R
Lyesh
(Alessio
Dures
(Durazzo)
Matva
Kroya
Arzo
OTirana
R.
R
The names are rendered as far as possible in the language of the predominant
race of a given region. All names (except Italian and German) have been transliterated,
whether they belong to languages using Latin characters (e.g. Magyar, Rumanian, Croatian) or not.
The transliteration is according to sound, except for modern Greek, where it is according to letter
The system followed is that of the Royal Geographical Society ("Hints to Travellers, 9th. edit., Lon-
don, 1906, Vol.2), except for these changes and additions: tch for ch; dj forj; tz for ts; ö like
German, ö or French eu, ü like German ü or French u.
Political boundaries as in July, 1914.
15
16
Copyright, 1918, by the American Geographical Society of New York
17
OSSOVO
Prizren
Nish
Mara
Piroto
Prishtina
Vranyeo
SHAR PL
Teltovo
Dibra
Drin
Leskovatz
B
Svishtehov
THE
Dobritch
Baltchik
B
Plyeven
(Plevna)
STARA
Berkovitza
Vratza
ker
yantra
Trnoval
く
​K
Trn
Bryezniko
A
N
SOFIYA
Shumen
(Shumla
Varna
Kamishi
R
N
G
E
A
R
Sliven
o
JB U
LGORA G
SRYEDNA
A
R
I
Anchialos
Burgas
O Stara Zagora
M
Plovdiv (Philippopolis)
i t
Skoplye ch
plyepch
(Uskub)
Küstendil
Poretch
Köprülü
(Veles)
α
Moriho
Shkumbi Elbasan
Okhrida
Okhrida
Monastir
oBitolye
Prespa
Muspakiya
De
R.
Sement
R
Prevtis Pass
Berat
Kortchao
(Koritza)
Kastoriao
Vlore (Valona)
I
O
Viosa
Lyaskovik o
M
Argyrakastron
1
Santi Quaranta
Ioannina
Yanina
E
P
N
CORFU
I
LEUKAS
KEPHALLENIA
N
Kailaro
C
Verria
α
Stanimaka
R
H
о
D
O
E
PIRIN
R GE
Struma
Strumitza
Doirano
Bistritza
E
A
oDrama
I
++
+++
O Seres
L.Takhino
SALONIKI
Gulf of Orfano
ISOLUN, THESSALONIKE
Chalkidike
Gulf
of Saloniki
Trikkala
Larissa
THESS ALY
Arta
R
opotamos
Ho
AETO
Bolos
Volo) NORTHERN
E
U
Lamia
Mesotongion
Patras
P
ELQ
ZAKYNTHOS
Pyrgos
Gulf of Corinth
Corinth
(Korinthos)
Argos
O Tripolis Nauplion
S
E
A
NE SU
NN
S
Kalamáta.
Sparta
Kavala
Chalkis
B
R.
Arda
H
M
S
OGümürdjina
Dedeagatch
Enos
THASOS
SAMOTHRACE
IMBROS
LEMNOS
-
E
SPORADES
ATHENS
OATHENAI
KYTHERA
E
E
G
4
R
So Adrianople
(Edirne)
T
O/Kirk Kilisse
A
Midia
Gallipoli
Dardanelles
LESBOS
D
CHIOS
N
May
S
එළ
E
श्र
E
о
A
L
B
43
42
41
R
Rodosto
SEA OF MARMARA
8
Bosporus
CONSTANTINOPLE
0000
[STAMBUL
SAMOS
路
​S
P
A
S
0
00
44
R
R
amy
40
39
K
E
Y
A
E
པསུ
}
RHODOS
38
37
36
18
19
20
21
22
23
24
25.
26
27
28
Drawn by W BRIESEMEISTER
29
L-L.POATES ENGR'G CO., N.Y
DISTRIBUTION OF BALKAN PEOPLES
359
languages and the special conditions and real sentiments of the population,
have made the mistake of considering the name Bulgarian as a national
name. The ethnographic maps of this earlier time, based on this incorrect
observation, have done much to confuse the main Balkan question, namely
that between Serbians and Bulgarians.
The Northern Contact Zone of the Yugo-Slavs
It is easier to determine the limits between the Yugo-Slavs and their
neighboring peoples on the north, those where they come into contact with
the Italians, Germans, Magyars, and Rumanians. These peoples are so
different from the Yugo-Slavs that there should have been no ambiguity
in determining the boundary between them, if a confusion of the issue
through accessory questions had been avoided. Furthermore, Hungarian
statistics are often inaccurate.
WITH THE ITALIANS
Numerous Italian writers have attempted during the present war to
show that the natural frontiers of Italy on the east are not the well-known
frontiers of the Apennine Peninsula. These natural frontiers, so say these
writers, include part of the Balkan Peninsula as far as the divide between
the Adriatic drainage and that of the Save, so that a large part of the
Dinaric region as far as the highest range would, from the geographical
and natural point of view, belong to Italy. These are also, so they say,
the strategical frontiers recognized as necessary for the security of Italy.
In order to confirm these claims by ethnographical facts the Italian writers
add that the number of Italians several decades ago was greater on the
eastern coast of the Adriatic than it is now and that they have been
Slavicized because Austria has favored the Yugo-Slav propaganda.
It is hardly necessary to repeat here the results of the fundamental and
well-known geographical and geological investigations concerning the
region, that is to say that the Karst and the Dinaric system are integral
parts of the Balkan Peninsula and that the natural boundaries of a
peninsula like that of Italy cannot be sought for on the eastern side of the
Adriatic. It is very clear even to an untrained observer that this sea
marks the natural boundary between the Balkan and Italian peninsulas.
The strategic security of Italy, if it really is menaced from the Yugo-
Slav side, constitutes a political and military problem which can be
solved on this basis but which has nothing to do with scientific investiga-
tions. The only fair argument to a certain extent is that the number of
inhabitants classed as Italians and belonging to the Italian portion of
Dalmatia, in the Quarnero and Istria, was larger some decades ago than
at present. These people, however, were not Italians, but Italianized
Slavs. After the Serbo-Croat awakening they returned to their original
260
THE GEOGRAPHICAL REVIEW
nationality. Except in western Istria and in Trieste Italians did not exist
as a nationality, even in former times, anywhere in the Balkan Peninsula.
Italians have established themselves there only as individuals, except at
Zara and at Fiume, where they form an important minority.
WITH THE GERMANS: THE SLOVENES
To the northwest, toward the Germans, the Slovenes constitute the racial
frontier and barrier. This frontier shifted considerably to the south during
the centuries, and numbers of Slovenes were Germanized, especially as a
consequence of the aid which during the preceding centuries the Catholic
Church rendered to Germanism. The Slovenes were entirely Germanized
in the southeastern part of Upper Austria, in the regions of Salzburg and
Lungau, in the eastern part of the Pustertal, in the vicinity of Lienz in
the Tyrol, and in the larger part of Styria and Carinthia, where today they
only occupy the southern districts. Germanization had even begun in the
cities of Carniola, but it was stopped, from the middle of the nineteenth
century on, by the Slovenian awakening, and the same process went on in
Dalmatia. Numerous Slovenes, already half-Germanized, returned to their
original nationality. From that time on the boundary between the Slovenes
and the Germans in Styria and Carinthia has remained practically un-
changed.
WITH THE MAGYARS
The most considerable displacement toward the south was that under-
gone during the last three centuries by the northern edge of the Serbo-
Croats in southern Hungary. Numerous large Serbian islands on the right
bank of the Marosh as well as those of Komarom, Djur, St. Andrew, Buda-
pest, and others were Magyarized. Magyarization took place in the Baranya
and in the "Croatian archipelago."
Croatian archipelago." Uplifted to the height of a sacred
state policy, Magyarization made progress in the Banat and the Batchka,
particularly in the latter region. The Magyar propaganda fell with par-
ticular vehemence on the Catholic Serbs, the Bunyevtzi (Bunyevatzes), who
constitute the majority of the population in Szabadka and Baya, and on
the Serbs of Krashova in the eastern Banat. The total result of Magyari-
zation appears to be superficial. As soon as these districts become inde-
pendent the Serbo-Croats will return to their nationality, as was the case
in Dalmatia and the cities of Carniola. It is not possible, without making
studies on the spot and without using the statistics of the Serbian patriarch-
ate of Karlovtzi, to determine exactly, from Hungarian statistics, the sec-
tions where the Serbians constitute the majority or the most important
part of the population, which is often very mixed. Hungarian statistics
suppress tens of thousands of Serbians; they also reduce the number of
other non-Magyar nationalities.
DISTRIBUTION OF BALKAN PEOPLES
361
WITH THE RUMANIANS
As a result of the natural expansion of the Rumanians the Serbians have
lost ground in the eastern Banat, especially, it appears, during the last
decades of the nineteenth century. First the Serbian islands of the valley
of the upper Temesh disappeared, particularly those of Lugosh and Karan-
shebesh, still well known at the beginning of the nineteenth century. Then
during the last centuries, the Rumanians descended into the plain of the
Banat and filtered in among the Serbians. The Serbian island of Krashova
has maintained itself against the Rumanians thanks to its Catholic faith.
The racial limit between the region of Serbian predominance and that of
Rumanian predominance passes about to the east of Vershets and of Bela
Tzrkva (Weisskirchen).
THE ISLANDS OF JUAN FERNANDEZ
By CARL SKOTTSBERG
University of Upsala, Sweden
A fine morning in the month of August, 1908, found my companion and
myself climbing the steep and narrow ridges of Mas-afuera, the more remote.
of the Juan Fernandez Islands off the Chilean coast. We were eager to
reach the high crest of that little island, for we knew that no scientist had
ever set his foot on that particular spot. We were in great haste, for our
time was short, as our vessel, a large Chilean government transport, must
get away from the wretched anchorage as soon as possible. It was the
accomplishment of an old ambition: for many years I had longed to see the
famous island of Robinson Crusoe, Juan Fernandez, and the other members.
of its group. And during my expedition to Patagonia and Tierra del
Fuego, in 1907-09, I had quite unexpectedly obtained an opportunity to
visit the islands. Unfortunately only twelve days were at my disposal, and
the season was winter.
1
DISCOVERY OF MAGELLANIC FLORA
2
I was more or less familiar with the island flora, and the only investiga-
tion previously made in Mas-afuera did not hold out much hope of any
great discoveries in my line. "Grass and thick mats and beds of ferns,
right to the top"-thus ran the description of Dr. Federico Johow, the
eminent Chilean botanist. But the tracks on his sketch map clearly showed
that he never reached the high part of the island. Above the forest we
trod the grass of extensive meadows, but as we continued upwards the
scenery changed. Outcrops of hard rock formed picturesque little chains;
the ground was strewn with boulders; the peaty soil and the stunted vege-
tation resembled subantarctic heaths (Fig. 18). A small stream cut through
the fern beds. I thrust aside the coarse ferns to quench my thirst. Then I
suddenly stopped. For over the dewy moss trailed the tiny subantarctic
bramble (Rubus geoides), and close by crawled the runners of the common
subantarctic club moss (Lycopodium magellanicum). And there were the
delicate rosettes of one of the dwarf mountain asters of the south (Lagen-
ophora) and others which certainly had no right to exist in the latitude of
Valparaiso (33°). I felt I had a reason to congratulate myself, for thus
was made the discovery of the Magellanic flora in the Juan Fernandez
Islands.
My companion and I made a hasty survey of the neighborhood. There
was no time for serious work, the day was drawing to its close, and all
1 The name Juan Fernandez is applied both to the chief island, Mas-a-tierra, and to the group as a
whole.
2 Federico Johow: Estudios sobre la Flora de las Islas de Juan Fernandez, Santiago, 1896.
362
THE JUAN FERNANDEZ ISLANDS
363
material gathered was further restricted by the seasonal limitations: up
here the winter rest is marked. But when I bade farewell to the summits

THE JUAN FERNANDEZ ISLANDS
79°00'
Pra Tunquillar
plai Morro del Viudo
pta de los Negros.
Morro Juanango >
pra Lemos,
pta del Padre
33°40′
-Bª del Padre
Puen
pta Isla o
Be Carvajal
Cta Blanca
pta O'Higgins
Morro del Spartan
LSANTA CLARA
79°00'
MAS-AFUERA
Based on data submitted in 1895 to
the Chilean Hydrographic Office
by Dr. F. Johow:
Ba Tolten
Buque varado
Loberia nueva
Loberia ventana
33°45'
Loberia vieja ó de tierra
mona
Plano
de la
80°45'
BaVillagra
Morro Viñillo
Ba Chupones
Quebrada de Sanchez
Quebrada del sandalito.
Quebrada del sandalo
Quebrada del pasto
La Vaqueria
78*50*
Portezuelo
Pta Salinas
P.Ingles
pra S? Carlos
BS.Juan Bautista
(Cumberland Bayj
El Yunque
Pra Chamelo
.Pra Bacalao
Pia Pescadores
P.Frances
Cabo
Guasaballena
El Verdugo
MAS-A-TIERRA
(JUAN FERNÁNDEZ ISLAND)
Based on a survey in 1895 by the
Chilean Hydrographic Office.
Scale 1: 243 000
3
The geographic co-ordinates are based on the following approximate positions:
(1) for Masatierra: ▲ 33° 37′36″S. and 78°49′45″W.
(2) for Mas-afuera: ▲ 33° 46′ S. and 80°46′ W.
310
301
!
|Mas-afuera
20°
30
Mas-a-tierra
79
Quebrada del óvalo
Quebrada del mono
Quebrada de las casas
10
Scale 1:4 600 000
10
0
10
20 30
40
50 miles
RELATION OF ISLANDS TO EACH OTHER
PER U
BOLIV
60
A
50
40
"Quebrada del blindado
Quebrada de las vacas
Juan Fernández is
Valparaiso
ང
¡URU:
GUAY)
Buenosay
ostrocente
Playa ancha
Rio de Janeiro.
3-1
Quebrada del varadero
Vicente Porraso
Rodado del sándalo
Molino Casanova
Tierras blancas
A-R
Scale 1: 235 000
3miles
Scale 1:111 000 000
500
1000 miles
RELATION OF ISLANDS TO SOUTH AMERICA
THE GEOGR. REVIEW, May 1918
80°45'
!
FIG. 1-Maps of the Juan Fernandez Islands: Mas-a-tierra, 1:243,000; Mas-afuera, 1:235,000. The
authorities are cited in each case. Heights in meters.
of Mas-afuera I had already made up my mind to come back to camp at
the little stream and force the island to yield its hidden treasures. My
3
3 Carl Skottsberg: The Wilds of Patagonia, London, 1911 (see Chapter 9). On the Magellanic flora see
the author's article "Distribution of Vegetation in the Colder South Hemisphere," I'mer, Vol. 25, 1905,
pp. 402-427, and the reports of the Swedish Antarctic Expedition, 1901-03, especially his: Pflanzenphysiog-
nomie des Feuerlandes, Wiss. Ergeb, der Schwedischen Südpolar-Expedition, 1901-1903 (Stockholm, various
dates), Vol. 4, Part 9; Das Pflanzenleben der Falklandinseln, ibid., Vol. 4, Part 10; The Vegetation in
South Georgia, ibid., Vol. 4, Part 12; Die Vegetationsverhältnisse des Graham Landes, ibid., Vol. 4, Part 13.

FIG. 2-The northern shore of the western part of Mas-a-tierra. Punta del Padre on the right. (Figs. 2-20 from photographs by the author.)
364

FIG. 3-The Anson valley, sloping toward Cumberland Bay, Mas-a-tierra. In the right background the massif of El Yunque. The slopes in the foreground
are barren; those in the middle distance are covered with forest patches.
365
366
THE GEOGRAPHICAL REVIEW
botanical results proved to have a remarkable interest. The large per-
centage of endemic and often highly peculiar forms characterizing the
islands raises questions of a general nature. Problems of the physical his-
tory of the Pacific Ocean, its islands, and the South American mainland are
involved, as well as that of the great plant and animal migrations that have
taken place in this part of the world. The data accumulated in the short
time available in 1908¹ needed amplification.
SWEDISH PACIFIC EXPEDITION
The new enterprise, known as the Swedish Pacific Expedition of 1916-17,
was supported by the Royal Geographical Society of Stockholm and by a
number of government institutions. Its aim was announced as "an attempt.
to carry out as complete a biological investigation of the outlying Chilean
islands as possible, especially of Juan Fernandez." We left Sweden on
October 4, 1916. Mrs. Inga Skottsberg was attached as botanical assistant,
and I was further accompanied by Mr. K. Bäckström, who made zoölogical
collections.
The Juan Fernandez group consists of two larger islands, Mas-a-tierra,
360 miles west by south of Valparaiso, and Mas-afuera, 90 miles farther
west. Mas-a-tierra is about 12 miles long by 323 wide, Mas-afuera 6 by 32.
At the southwest end of the former and separated from it by a shallow
channel is the small and desolate islet of Santa Clara (see map, Fig. 1).
For many years after their discovery the islands were only occupied
temporarily. In the mid-eighteenth century, to prevent their use as a
rendezvous for pirates or as a provisioning station for other more formidable
enemies, the Chilean governor created the first permanent settlement in
Mas-a-tierra on the bay of San Juan Bautista, better known as Cumberland
Bay (Fig. 8). The fortunes of the settlement fluctuated. On several occa-
sions the island served as a penal station. Agricultural enterprises had little.
success. Today it enjoys a certain prosperity as the center of a flourishing
fishing industry. The waters are rich in excellent fish, many species being
peculiar to the place; yet this abundant variety is of minor importance. The
colony, including some 300 souls, may be said to subsist on one single prod-
uct, a large crustacean (Palinurus frontalis), known only on these islands
and on the sterile islands of San Felix and San Ambrosio 500 miles farther
north, where no settlement is practicable. This lobster is highly appreciated
in Chile. Two canning factories have been established here, and three
schooners carry the canned product as well as great quantities of live
lobsters to the mainland. Nearly all the fishermen have their gardens,
where fruit and vegetables are grown, and small herds of cattle graze in
the well-watered valleys.
The expedition arrived in Mas-a-tierra on December 1, 1916, and
4 The scientific results were published in the K. Svenska Vetenskapsakad. Handl., 1910-1916; Part III (1914)
deals with Juan Fernandez, Parts IV (1913) and V (1916) with the Andean and Magellanic floras.
THE JUAN FERNANDEZ ISLANDS
367
remained until April 30, 1917. A laboratory was established in the colony,
where we maintained our headquarters. February and the greater part of
March were dedicated to the survey of Mas-afuera. Mr. Bäckström
returned to Mas-a-tierra in July and spent a month there studying the
winter fauna. All research work had to be done on foot. Three narrow
and miserable cattle trails lead from Cumberland Bay to other parts, and
it is possible to follow them on horseback, at least in the drier season; but
some passages make the rider hold his breath and wish himself safe on
the other side. The higher parts of the island can be reached only by
following the narrow ridges that descend to the sea. Along these ridges,
crowned by an outcrop of basaltic rock, are the old goat paths, many still
visible in spite of the scarcity of these animals in later years. Goats appar-
ently do not hesitate to use the most dangerous passages, but the traveler
will sometimes hesitate to follow their trail. This explains why our survey
yielded good results. The younger fishermen are daring climbers and stop
at nothing in their pursuit of the goats or the chonta palm, which is much
sought after for its wood. Occasionally they contributed to our knowledge
of natural history, and one of them, Pedro Gutierrez, who was a great help
to us from the first, later became a member of the expedition to Mas-afuera.
THE ISLAND OF MAS-A-TIERRA
All the islands are built of Tertiary volcanic rocks, nearly always beau-
tifully layered. It is probable that they are early Tertiary: there is no
trace of craters or of any recent volcanic action. The topography of Mas-a-
tierra is very broken (Fig. 3). The eastern part is much the higher. It
reaches its greatest elevation along the south side, culminating in the famous
and truly inaccessible massif El Yunque (The Anvil), which falls abruptly
down into the sea 3,100 feet. The upper part is of a hard, dark gray rock
forming all the wild crests and peaks. The island has been greatly cut down
by erosion, which has left a curved backbone and a long line of ribs separat-
ing deep valleys. Underlying the harder rocks are innumerable layers of
tufa, red, brown, or yellow, that make a striking contrast to the vivid green
of the forests. The rainfall is concentrated on the higher eastern and cen-
tral portions of the island, where the air currents, rising suddenly from the
south shore, condense their moisture. From the high backbone forest-clad
valleys descend to the north coast and terminate in open bays. No level
ground of any extent is seen; a tract like that shown in Figure 8 is a rare
exception. Between the bays the shore is very steep, and the cliffs reach
the height of a thousand feet in many places (Fig. 2).
The southwestern portion of the island, of which Santa Clara is a con-
tinuation, has a rather different appearance. The upper layers are upraised
and form a high ridge along the northern margin, where the shores are
precipitous. The valleys run in a southerly direction and do not reach the
water; their streams are dry for several months. Accordingly, arborescent


FIG. 4-The chonta palm (Juania australis).
FIG. 5-A waterfall on Mas-a-tierra hidden by Cladium, a twig rush, and
Gunnera, whose leaves attain man-size.

368
THE JUAN FERNANDEZ ISLANDS
369
vegetation is entirely wanting, and for miles the ground is covered with
grasses and herbs. Only the climate of the central part (Cumberland Bay)
is well known. Here observations have been made many years. It shows
a combination of comparatively high temperature and considerable rainfall
not found on the coast of Chile. The temperature is extremely agreeable;
the summer is never very hot; and the winter is always mild, frost being
unknown. Gales are frequent, and local winds rush down from the preci-
pices and sweep through the valleys with tremendous force. A thick wet
fog hangs round the summits even in the drier season, and the character of
the vegetation further indicates that the climate of the higher portions is
colder and wetter. The southwestern peninsula with Santa Clara has a
different climate from the rest. It is and has always been destitute of trees.
A FOREST TRIP
In order to make the reader acquainted with the peculiar nature of Mas-
a-tierra I shall now offer to be his guide on a trip through its forests and
along its picturesque cliffs; but special note must be made of its different
types of vegetation, for it is plant life more than anything else that puts its
stamp on the scenery. Let us remember Darwin's saying: "A traveler
should be a botanist."
In the valleys near the sea work is easy. The soil lies bare, the forest
has been destroyed, erosion is very active. There one of the introduced
Chilean weeds has won a considerable victory, thanks to its most efficient
means of dispersal; it is a burnet, Acaena argentea. Cattle and sheep become
covered with its fruits and carry them everywhere. From the lower hilly
tracts we soon pass into low, pitch-dark thickets, formed by another
stranger, the Chilean Aristotelia maqui, introduced some sixty years ago.
It is able to germinate and grow in the deepest shadow, it sprouts from
the base of its stem with amazing speed, and it has entered and conquered
all the lower parts of the valleys. The berries are the staple food of the
thrush, which sows it in thousands every year. To the native flora it has
become a serious menace. Its vanguard has already reached the highest
ridges. If nothing is done to check the formidable invader, much of the
precious native flora is doomed.
At a height of 600 or 700 feet the forest still retains much of its original
character, and many places remote from the colony are quite primeval. It
is an evergreen rain-forest, much resembling, in general appearance, that of
southern Chile. Several species are identical with those of Chile, others
belong to Chilean genera, represented by endemic species. The commonest
♪ A summary of meteorological records of Chile is given and analyzed by R. C. Mossman: The Climate
of Chile, Journ, Scottish Meteorol. Soc., Vol. 15, 1911, pp. 313-346. For the first half-decade of this century the
principal observations at Juan Fernandez show:---mean monthly temperature, February, 66.9° F.; August,
54° F.; absolute minimum, 38.8° F.; prevalent winds (blowing on 70 per cent of the observations), south-
east to southwest; mean annual rainfall, 40.5 inches, of which 80 per cent falls from April to September.
To find such a degree of rainfall on the Chilean coast one must pass to a considerably higher latitude.
370
THE GEOGRAPHICAL REVIEW
is a myrtle tree, Myrccugenia fernandeziana, with a smooth light-gray bark
and a beautiful lustrous dark foliage, in early summer covered with innum-
erable white, fragrant flowers. The soil is strewn with fallen trunks,
covered by a rich carpet of mosses. Ferns are plentiful; there are large and
handsome forms, as well as minute, of the elegant Hymenophylla. Curi-
ously enough, phanerogamic climbers are wanting; they are replaced by
ferns with winding rhizomes reaching high up on the trunks of the trees
and forming a most attractive and striking picture.
NOTEWORTHY SPECIES
One of the most famous trees is a palm, Juania australis, a monotypical
endemic genus, called chonta by the inhabitants (Fig. 4). It is now rare
in the accessible valleys, for it has been sought and cut with great per-
sistence on account of its handsome wood. True, cutting it is strictly pro-
hibited, but the law is not rigidly enforced: not a vessel leaves the island
without carrying away some of this wood. The chonta has its nearest
relatives in tropical America but has no relation with the Chilean palm
growing just opposite on the mainland. There are several other trees and
shrubs altogether strange to the flora of Chile, or even to that of South
America.
The forest flora of Juan Fernandez has suffered at least one great loss
since man first settled there. The reader who knows the early history of
the Spanish possessions will perhaps remember that the islands once were
famous for their precious, perfumed sandalwood. So complete was the
destruction of this remarkable tree that the first complete botanical descrip-
tion was written merely from twigs of the last specimen-and that as late
as 1896! For almost a hundred years it had been sought in vain. I shall
never forget my journey to pay homage to this tree in 1908. It was de-
stroyed some years ago, and thus Santalum fernandezianum disappeared
from the world. The genus is unknown in America, it is West Pacific and
Indian, save for this single species blotted out by thoughtless man.
THE HIGHER VEGETATION
But let us hasten to the higher slopes and the mountain forest. We have
entered the fog belt, where there is more moisture and less heat." Even the
ordinary traveler would note the change in vegetation. Many new species.
appear. Ferns become plentiful, and some are very striking. One is a
proud tree-fern, Dicksonia Berteroana (Fig. 6), whose nearest relative is a
species in Fiji. Another is the Thyrsopteris elegans, with leaves as long as
12 or 15 feet, having doubtful relations back in pre-Tertiary time but none
in the present! Fine epiphytic ferns and large festoons of hanging moss
44
6 In order to gain some knowledge of the weather here we set up some instruments in the Portezuelo
pass (Selkirk's Lookout," where the memorial tablet was erected in 1868) and made observations every
two days for some little time. But we had to climb 1,850 feet to read the instruments.

FIG. 6.

FIG. 7.
FIG. 6-Virgin forest on Mas-a-tierra, with tree-ferns (Dicksonia Berteroana).
FIG. 7-Robinsonia thurifera, one of the rarest plants of Mas-a-tierra.

371
372
THE GEOGRAPHICAL REVIEW
cover the trees. This is the region where one of the rarest and most peculiar
dwellers of Mas-a-tierra is found: Lactoris fernandeziana, forming the mono-
typical family Lactoridaceae. It stands somewhere near the magnolias
and offers the only instance known of a natural order confined to an oceanic
island.
At the bottom of every quebrada (Fig. 5) a small waterfall is generally
encountered, where a stately twig rush (Cladium scirpoideum) hangs down
over the dripping wet rocks. Round the brook an endemic pangue
(Gunnera peltata) unfolds its gigantic leaves, borne on a stem many feet
high. The special tree of the valley bottoms is a stout Bochmeria, of the
nettle family, with distant tropical relations.
An excursion through the forest is easily described, but not so easily
made. The sides of the valleys are very steep, especially when approaching
the ridge. The soil is soaked with moisture and often gives way under
foot. Only with the aid of the trunks and roots can ascent of these ridges
be made, save in the few places where a path has been cleared, winding in
sharp turns along a slope often as steep as 30° or 40°. Many places are
impenetrable thickets of ferns, brushwood, and bamboo, so that it costs
some labor to gain the stony ridge.
Figure 9 will give a good idea of one of the ridges. The limit of the
forest is clearly visible. This is not a climatic limit. The soil is very thin
or entirely wanting; vegetation is confined to crevices and narrow ledges.
and thus is naturally broken up. Moreover the ridge is dry in summer
and very much exposed to the gales. Dense thickets of xerophytic shrubs,
rigid ferns, and patches of hard grass are often encountered.
MINIATURE TREES
Along the upper limit of the closed forest, where light is strong, we meet
with the host of plants which have helped to make these islands famous
among botanists. They are peculiar miniature trees, belonging to endemic
genera or at least species of various families, Compositae being most com-
mon. They have a simple or dichotomously branched stem, and the leaves
are crowded at the end of the branches. Here are large woody Eryngia, a
plantain tree, the rare Selkirkia (forget-me-not family), and several species
of Dendroseris, Robinsonia, and other Compositae (Figure 7 is a good ex-
ample). They constitute a perfect living museum of ancient types, either
without relations or connected only with distant Pacific islands, thousands.
of miles away.
The view from the higher parts--and east of El Yunque a summit of
2,500 feet is reached without much difficulty-is truly magnificent. All
around run the rocky ridges, sometimes as sharp as a knife edge and as
steep as a wall, sometimes barren, sometimes heavily clad with ferns and
mountain pangues; high, dominating peaks rise with a defiant air over
richly carpeted slopes, brilliant with many flowers. Everywhere valleys


FIG. 8-Cumberland Bay, Mas-a-tierra.
FIG. 8.

FIG. 9.
FIG. 9-A typical rocky ridge on Mas-a-tierra. The limit of forest is not due to altitude but to
absence of soil.
373

FIG. 10.
FIG. 11.
FIG. 10-Steppe vegetation in Mas-a-tierra. Small patches of Gunnera.
FIG. 11-Dendroseris macrophylla on Santa Clara. A miniature tree representing an ancient type.
The leaves are glaucous, the flower-heads bright orange.
374
THE JUAN FERNANDEZ ISLANDS
375
seem to descend, their bottoms hidden under luxuriant growth. Chontas
raise their shapely heads above the green roof, waving in a fresh breeze
that now and then uncovers a cluster of the scarlet berries. Down below
follow the barren slopes, bright with the colors of the earth and suddenly
falling away in precipitous cliffs, washed by the eternal surf. The
wind brings the sound of the roaring breakers, but here as faint as the
humming of a swarm of bees. With a little geographical training the
physiographical history of the island is easily read. Not so plain is its
story for the biologist. He faces a number of difficult problems, and like
difficulties confront the zoologist. The fauna, at first, does not seem to be
very rich. There are no indigenous mammals, and, save for sea birds,
winged life is poor in species. The most remarkable bird is a humming-.
bird (Eustephanus fernandensis), famous among ornithologists on account
of the sexual difference. However, of lower animals a great many are found.
such as land molluscs, insects, and spiders, and many of them are endemic.
The barren western parts of Mas-a-tierra are of less interest. Weeds,
from Chile or Europe, are very abundant, quite obscuring the original vege-
tation, of which but small areas like that in Figure 10 remain. On Santa
Clara fresh water is altogether wanting. In crevices of the high coast cliffs.
some good specimens of Dendroseris, of two species, do fairly well. A rock
close to Santa Clara, separated from it only at high tide, offers a most strik-
ing picture with the numerous Dendroseris trees, D. macrophylla prevail-
ing (Fig. 11).
THE ISLAND OF MAS-AFUERA
The island of Mas-afuera, mentioned in the opening paragraph, is dis-
tinctly inaccessible. In 1909 the Chilean Government established a penal
settlement here, and more or less ruinous buildings may be seen in many
places. Expensive landing arrangements were made, but since the place was
abandoned, in 1913, they have become completely destroyed. There is no
harbor and no safe anchorage anywhere. The surf often makes landing
impossible; it is always more or less precarious, so that the assistance of
skilled hands is necessary. We had chartered a schooner to take us there,
but our first attempt was fruitless and we lost a week. The second, however,
was successful, and we landed our impedimenta without damage. The
settlement was situated at the mouth of the Quebrada de las Casas. We
made ourselves as comfortable as possible in the best house of the now
uninhabited island.
DIFFICULTIES OF ACCESS
Mas-afuera has a rather striking appearance. It is a solid rectangular
block of sheeted lava, on all sides terminating in perpendicular cliffs. For
the most part a narrow strip of boulders along the beach permits a passage
round and makes it possible to enter all the valleys which reach the sea. The
west side is much higher than the east, and all along that side there is a


FIG. 12-The upland surface of Mas-a-tierra, its eastern border dissected into numerous rectilinear canyons (see map. Fig. 1).
376

FIG. 13-Forest patch in a Mas-afueran valley. The forest on this island is confined to the middle levels of the valleys (1,000-2,200 feet), the limits, in contrast
with Mas-a-tierra, being climatic. Above is a meadow and alpine zone; below a luxuriant fern vegetation flourishes.
377
378
THE GEOGRAPHICAL REVIEW
sheer drop of four to five thousand feet. From the longitudinal ridge a
great number of canyons radiate to the northeast and southeast (Fig. 12).
Their heads are usually occupied by high waterfalls, often nearly dry in
summer (Fig. 14). Some of these canyons, in the relative dimensions of
depth and width of bottom, hardly have their equals anywhere. A good
example is shown in Figure 17. There is practically no margin between the
stream and the perpendicular wall. Small waterfalls and heaps of huge
boulders must be passed in order to enter; it may even be impossible to
enter at all. Fresh pieces of rock, fallen down from the height of some
two or three thousand feet, show that the place is not absolutely safe. The
two central valleys, which drain the highest and rainiest part, have cut
back into the island so far and so deep as nearly to divide it in two. These
two valleys, Casas and Vacas, are separated by a high and narrow ridge,
so steep that its sides cannot be climbed. But near the sea a trail, cut by
the convicts, leads up on the ridge, and from there a goats' track leads all
round the end of the Casas valley to the northern half of the highland,
where two summits reach 4,700 feet. This promenade is one to be recom-
mended. It is a little tiresome and, of course, should not be tried by
anybody likely to get dizzy, for in several places there is just sufficient
space for your feet, if they be not too large, and on both sides you look
down into a canyon with a winding silver streak glittering below you-
there is a chance for a clean fall of two or three thousand feet on either
side. The southern portion of the upland cannot be reached in the same
convenient manner, for it is impossible to climb along the head wall of
the Vacas canyon, even for the goats I believe. Just beyond is the
highest summit, most inappropriately called Los Inocentes. It can only
be reached from two places near the sea; the easiest is the south wall of the
Vacas canyon (Fig. 15). The ascent is not without danger, for the slope
is covered with very slippery grass and loose stones. Shoes should not be
used, and the nearly lost faculty of grasping with the toes should be exer-
cised if possible. Our first attempt, made from the southeast corner of the
island, a very dangerous place, was frustrated after we had got over the
worst part, for we met a mile-wide and impenetrable forest of fern trees.
and came into a thick fog. Later we started from a point just south of the
Vacas valley and had complete success. We found the top to be 5,300 feet
high, a considerable figure in comparison with the small size of the island.
ITS VEGETATION
We have no exact knowledge of the climate. It is not greatly different
from that in Mas-a-tierra, for many vegetable products are the same. But,
of course, there is nothing in Mas-a-tierra to match the high upland of
Mas-afuera. Residents in the penal colony have reported that it not seldom
appears white with frost in winter. The coast cliffs and lower slopes are
dry and warm and have a steppe vegetation.

FIG. 14.

FIG. 15.
FIG. 14-Head wall of a deep canyon on the eastern side of Mas-afuera.
FIG. 15-The entrance of the Quebrada de las Vacas, Mas-afuera. Boulder beach in the foreground.
The ascent is made up the gully on the left and then along the south (left) wall of the canyon.
379


FIG. 16-The steep western side of Mas-afuera, falling away 4,000 or
5,000 feet to the sea.
FIG. 17-Typical canyon in Mas-afuera. Some of these canyons hardly
have their equals in steepness anywhere.

380
THE JUAN FERNANDEZ ISLANDS
381
The forest is confined to the middle part of the valley systems, forming
an interrupted belt between 1,000 and 2,200 feet above sea level. Figure
13 gives a good idea of this forest. It is largely composed of an endemic.
myrtle tree (Myrceugenia Schultzei). Beautiful groves of Dicksonia fill the
damp corners. Along the upper parts of the forest and ranging up into the
next region are a number of miniature trees of the Robinsonia type, some
only found on this island and now made known for the first time. The
forest limit here is a climatic one, probably a matter of summer tempera-
ture. Above this line we enter the foggy region, covered by extensive
meadows and vast beds of ferns. To this region belongs the marvelous pure
Dicksonia forest around Los Inocentes. The meadows must have been quite
different before the arrival of man, for they are invariably formed in the
main of two European plants.
ITS WILD SCENERY
Above 3,500 feet the scenery grows more and more alpine in character.
Mas-afuera has a really alpine region, notable for its Magellanic flora. It
is not a question of the miraculous arrival of some stray species from the
far south it is a whole flora, forming natural associations, with mosses,
lichens, ferns, and flowering plants. The nature of the locality is shown in
Figure 19.
:
There is a first-rate surprise in store for the traveler when he reaches
the upland and strolls across toward its western border. There he suddenly
stops and he is wise in doing so, for at his feet opens an abyss (Fig. 16).
Four or five thousand feet below him the breakers come rolling in; cliffs are
cut and torn to the most incredible shapes; there are perfect combs, over-
hanging shelves, oblique needles, ready to fall. It is wild scenery indeed
and may fairly be compared to the hell of Dante as pictured by Doré.
Let us hasten down into one of the canyons, where there is rest and peace.
It is nature's own greenhouse. What a variety of beautiful ferns! What
life on every ledge, in every corner where soil is to be found. Along the
stream are imposing groups of a fine Gunnera, discovered by me in 1908,
and endemic in this island. Its leaves measure as much as 10 feet across
(Fig. 20). The stalks are edible and refreshing, with a slightly acid taste.
FLORAL ELEMENTS NEEDING EXPLANATION
Such are the Juan Fernandez Islands. As soon as our collections have
been studied full account of our results will be given. I shall try to explain
the origins of this quaint life. There is one decidedly Chilean element,
related to the flora of Valdivia, but the occurrence of Chilean types is
not surprising. Then there is the Magellanic, or subantarctic, element, not
represented in Mas-a-tierra. It may have immigrated at a time when it
had a wider range on the continent, as was the case during the Glacial
Epoch. There is, further, a small tropical element without relations to the

FIG. 18.
FIG. 19.
FIG. 18-The upland on Mas-afuera where the first Magellanic plants were discovered in 1908.
FIG. 19-Alpine scenery on the upland of Mas-afuera.
382
THE JUAN FERNANDEZ ISLANDS
383
present flora of Chile. But this country once had a tropical flora. There
was no high cordillera separating it from the rest of the continent at that
time. Unfortunately, our knowledge of this flora is very small.
Finally, there is the very special element, the strange-looking endemic
plants which either have no relations at all or give doubtful hints towards

FIG. 20-Gunnera Mas-atuerae in the Quebrada de la Casas, Mas-afuera. The leaves sometimes measure
as much as 10 feet across. See also Fig. 5.
distant parts of the immense Pacific Ocean. Did they come across the vast
waters from many lands and become transformed into new species and
genera, while their ancestors disappeared altogether? The very slow rate
of evolution negatives this theory. Possibly they, together with others in
other islands, form the scattered remains of an ancient Pacific flora, of
which the bulk has perished. But where did they come from? Was there
a continent, a group of larger islands? Was Juan Fernandez once a part
of something, of some mysterious land? I shall only point to one fact.
Our islands are situated on a deep submarine ridge, which can be followed
north to San Felix and San Ambrosio. These latter islands have a very
different climate, but they have, nevertheless, some ancient endemic plants.
Unfortunately, it was not possible for us to visit this place, as no transpor-
tation facilities could be obtained from the Chilean government.


THE GROWTH OF AMERICAN CITIES
By LAWRENCE V. ROTH
This study places the larger American cities in four geographical
groups: those belonging to the Atlantic Coast, the Mississippi Valley, the
Great Lakes region, and the Pacific Coast. They may also be grouped in
chronological periods as hereafter outlined. With the French school of
geographers we may distinguish between site and situation as factors
influential in locating and developing urban centers.¹ The site of a city
is determined by local conditions, such as the nature of the ground, water
supply, and accessibility to the building materials and food needed by a
growing community. These factors Professor Blanchard calls éléments de
fixation. The present treatment deals rather with the general situation
of the city in relation to routes of communication and the natural resources
of its hinterland.
Periods of Growth
The growth of cities in the United States has passed through four
periods, each of which in its turn was the response to the commercial and
industrial development of a new geographical region. The first steps in
city development appear in early colonial days with the increasing use of
natural resources and the exchange of goods with the mother country, her
colonies, and various nations of Europe. The second period, whose activity
centered in the lower half of the Mississippi Valley, began about 1810
and reached its culmination about 1850. Flatboats and, later, steam craft
carried the products of the soil on the western waters. The consequent
expansion of trade and the movement of the population across the Appa-
lachian barrier gave swift growth to the cities of the region. Within this
period some Atlantic Coast ports grew more rapidly than the cities of the
Mississippi Valley. Urban centers of the Great Lakes region developed
rapidly about 1850 but reached their most significant growth after 1880.
The expansion of the railroad system westward along the lake plains and
the navigation of the lake waters were important factors in the growth of
these cities. These means stimulated the movement of people into the lake
plains province. The fourth and latest period of large city making conse-
quent on the development of a new region belongs to the Pacific Coast
and becomes significant after 1900. The constant westward migration of
population promoted by the building of the transcontinental railways, the
marketing of the native resources, trade with Alaska and the Orient, and
1 This view has been elaborated by Professor Raoul Blanchard of the University of Grenoble, Visiting
Professor in Harvard University, 1917.
384
GROWTH OF AMERICAN CITIES
385
favorable climate are the factors which have contributed to the concen-
trations of population on the Pacific Coast.
GRAPHICAL COMPARISON
The first twenty-one cities in population rank are chosen to show this
urban movement. Figure 1 does not show actual population" but changes
in rank regardless

of the number of
inhabitants. The
exhibit is by de-
cades from 1790 to
1910 with the in-
terval 1910 to 1917
added. Figure 2
shows the
the actual
population growth
of the largest of
these cities for the
same period. On
Figure 3 are com-
posite lines repre-
1790 1800 1810 1820 1830 1840 1850 1850 1870 1880 1890 1900 190 1947
PHILADELPHIA
▲ I NEW YORK
NEW YORK
BOSTON
CHARLESTON A
BALTIMORE
SALEM
POPULATION RANK
1790-1917
City Groups
A-Atlantic Coast
-Mississippi Valley
•■Great Lakēs
-X-Pacific Coast
2 CHICAGO
3 PHILADELPHIA

4 ST LOUIS
5 DETROIT
6 BOSTON
7 CLEVELAND
8 LOS ANGELES
9 BALTIMORE
10 PITTSBURGH
XII SAN FRANCISCO
112 BUFFALO
13 CINCINNATI
114 MILWAUKEE
senting the four
largest cities of
each group.
SECOND PERIOD
THE GEOGR REVIEW MAY 1918
PITTSBURGH
CINCINNATI
LOUISVILLE
BUFFALO
ST LOUIS
SALEM
CHICAGO
SAN FRANCISCO-
DETROIT
MILWAUKEE
CHARLESTON I
MINNEAPOLIS
LOUISVILLE ~
LOS ANGELES
SEATTLE —
15 NEWARK
16 NEW ORLEANS
17 MINNEAPOLIS
18 WASHINGTON
*19 SEATTLE
20 INDIANAPOLIS
21 KANSAS CITY
All the diagrams
show the cities of
the Atlantic Coast
in 1790 in lead-
ing positions. The
lines for the river
cities appear about
1810, and Figure 1
shows these centers of the Mississippi Valley rising sharply from 1810 to
1850. It also reveals a corresponding fall in rank of two Atlantic Coast
cities, Charleston and Salem. After 1850 all the river cities except St.
Louis are descending in rank, while the Great Lakes cities rise into prom-
inence. Two cities of the Atlantic Coast, Baltimore and Boston, yield their
rank to a river city, St. Louis, and a lake city, Chicago. On Figure 2,
showing actual population growth, New York and Philadelphia in 1850
FIG. 1-Diagram showing rise and fall in municipal population rank of
the twenty-one largest American cities according to region from 1790 to 1917.
2 The graphs are based on the United States Census reports from 1790 to 1910 and on various municipal
publications. In every case they represent municipal and not metropolitan statistics.
386
THE GEOGRAPHICAL REVIEW
2 000.000
19
18
NEW YORK
CHICAGO
PHILADELPHIA
are much in the lead. Baltimore and Boston still rank third and fourth.
Then come four river cities, New Orleans, Cincinnati, St. Louis, and Pitts-
burgh. The lake cities show a
slight ascent after 1850, while San
Francisco and Los Angeles appear
on the Pacific Coast. Figure 3
shows the Atlantic Coast group far
in the ascendant and rising sharply
after that date. The slight rise of
the river line, which in 1850 is sec-
ond, is entirely due to the rapid
growth of St. Louis. After this 500.000
date the lake line makes an ascent,
and the line for the Pacific Coast 4
has its beginning.

THIRD PERIOD
The decades from 1880 to 1900
are the most important for the
growth of the lake cities. Figure
1 shows every lake line rising dur-
17
16
I
13
12
1000 000
THE GEOGR REVIEW MAY 1916
900 000
8
7
6
500 000
4
3
1
GROWTH
POPULATION
1790-1917
City Groups
Atlantic Coast
Mississippi Valley
·Great Lakes
·Pacific Coast
ST. LOUIS
DETROIT
BOSTON
CLEVELAND
LOS ANGELES.
BALTIMORE
PITTSBURGH
İSAN FRANCISCO
BUFFALO
CINCINNATI
NEW ORLEANS
2
NEW YORK
PHILADELPHIA
100 000
NEW ORLEA
1790
1800
1810
1820
1830
1840
1850
FIG. 2-Diagram showing actual population growth of the largest American cities according to
region from 1790 to 1917.
ing these twenty years, while all the river lines save that of St. Louis are
falling. On the Atlantic Coast Philadelphia yields to Chicago, and Boston
1860
CINCINNATI
PITTSBUROK
BOSTON
ST/LOUIS
LOUISVIL
BALTIMORE
CHICAGO
SAN FRANCISCO
www.
BUFFALO
CLEVELAND
DETROIT
SEATTLE
LOUISVILLE
PORTLAND
LOS ANGELES
PORTLAND
SEATTLE
1870
1880
1890
1900
1910
1917
GROWTH OF AMERICAN CITIES
387
loses its fifth rank to St. Louis. The most significant movement indicated on
Figure 2 is the growth of the cities of the Great Lakes. Apart from the
more rapid development of St. Louis the river cities show a moderate
growth from 1860 to the present time. Figure 3 shows the cities of the
Atlantic Coast still in the lead in 1880, but the river cities give way to
those of the lakes after that date.

FOURTH PERIOD
3
In the period from 1900 to 1917, as Figure 1
shows, the first four cities, New York, Chicago, Phil-
adelphia, and St. Louis, maintain their respective.
rank. The most significant change, however, is the
growth of Los Angeles and Seattle on the Pacific
Coast. San Francisco holds eleventh rank from
1910 to 1917. As a center of population, however,
the San Francisco district ranks seventh while Los
Angeles ranks fourteenth. This movement has
caused a corresponding fall in rank of certain lake
cities, especially Buffalo and Pittsburgh. Another
remarkable change of rank shown on Figure 1 is
the rise of Detroit to fifth place, above Boston and
and Cleveland. Boston, however, by reason of the
density of population
in its surrounding
area is
is the fourth
metropolitan center."
Figure 2 pictures the
same urban
ment. The growth of
New York, if shown
to 1917, would re-
quire a diagram on a
different scale. The
sharp ascent of its
line after 1890 is
largely due to the
absorption of Brook-
move-
THE GEOGR HEVIEW] NY 19:8
1790
POPULATION
CITY
COMPOSITE
GROWTH
OF
GROUPS
CURVES
ATLANTIC
COAST
MISSISSIPPI
GREAT
1.
VALLEY
PACIFIC
1800 1810 1820 1830 1840 1850
1860
1870
1880
1890
COAST
7 200000
68
164
6 000 000
156
152
48
44
4 000 000
36
32
28
24
2 000 000
16
12
8
400 000
1900 1910 1917
FIG. 3-Composite curves showing the municipal population growth
of the four largest American cities in each of the four regions from 1790
to 1917 (Atlantic Coast: New York, Philadelphia, Boston, and Baltimore;
Mississippi Valley: St. Louis, Cincinnati, New Orleans, and Louisville;
Great Lakes: Chicago, Detroit, Cleveland, and Buffalo; Pacific Coast: Los
Angeles, San Francisco, Seattle, and Portland).
lyn and other municipalities in 1898. A most significant change is in the
growth of the Pacific Coast cities after the year 1900.
3 The author is aware that metropolitan statistics offer a fairer basis for comparison than those of
municipalities. The object of this paper, however, is to treat urban centers regionally rather than indi-
vidually. The metropolitan figures would change the rank list as shown on Figure 1, but they would not
materially affect the general movement of the lines.
4 The latter counted as a lake city after 1860. See below.
Mark Jefferson: The Anthropography of Some Great Cities. Bull. Amer. Geogr. Soc., Vol. 41, 1909,
pp. 537-566; reference on p. 555.
388
THE GEOGRAPHICAL REVIEW
Cities and Physiographic Provinces.
6
There is further geographic basis for the classification of these cities in
Atlantic Coast, river, lake, and Pacific Coast groups. The Atlantic cities
are on tidal waters within the Coastal Plain Province, Boston being the
only complete exception. The largest city of the group, New York, lies in
part on the narrow northern extension of this province. The size of the
cities becomes smaller as we go southward upon the broader part of the
coastal plain. It suffices to name Philadelphia, Baltimore, and Washington.
These great cities stand on the inner edge of the coastal plain near the
"fall line," where there is a break in transportation. Newark also belongs
in this group.
The river cities have the further advantage of belonging to the Missis-
sippi Valley plain. Three of these river cities, St. Louis, Louisville, and
Cincinnati, lie on the southern border of the Central Lowland Province.
New Orleans stands far south on the Mississippi Alluvial Plain of the
larger Coastal Plain Province. Pittsburgh, whose early development was
due to its position at the confluence of two rivers, belongs to the Allegheny
Plateau section of the Appalachian Plateau Province. Indianapolis and
Kansas City (Mo.) have been included in this group, both belonging to
the Central Lowland Province. It is apparent that the river group of cities
has not as much geographic unity as those of the Atlantic Coast.
The lake group of cities, including Buffalo, Cleveland, Detroit, Chicago,
and Milwaukee, falls within the Eastern Lake section of the Central Low-
land Province. There is a high degree of geographical unity here. The
author has classified Minneapolis as a lake city because it belongs to the
Western Lake section and its growth has been to a considerable degree
due to its relation to Great Lakes navigation. Pittsburgh has also been
classified as a lake city since 1860 because its remarkable growth after this
date has been due to the shipment of iron ore on the Great Lakes.
The Pacific Coast cities, San Francisco, Los Angeles, Portland, and
Seattle, belong to the Pacific Border Province. Seattle and Portland lie
within the Puget Trough section; San Francisco is in the California
Trough, while Los Angeles belongs to the Los Angeles Ranges section. All
these cities are on navigable waters. Los Angeles, though it is not strictly
speaking on the coast, may be classified as a coastal city.
Atlantic Coast Cities
The early centers of population in the colonial period were chiefly com-
mercial ports. Commerce with Europe, especially with the mother country
and her other colonies, made possible such northern ports as Boston, Phila-
delphia, New York, Newport, and Salem. Along this part of the Atlantic
For this classification the author has used Fenneman's division of the United States into physio-
graphic units; see N. M. Fenneman: Physiographic Divisions of the United States, Annals Assoc. Amer.
Geogrs., Vol. 6, pp. 19-98, Pl. I.
GROWTH OF AMERICAN CITIES
389
coastal lowlands good harbors abound. From these harbors went such
colonial products as fur, fish, wheat, flour, corn, wood, and rum. Rum
manufactured in the northern colonies was carried to the coast of Africa
and bartered for slaves, who were taken to the West Indies to be exchanged
for sugar, molasses, and money. Additional imports through these gate-
ways were cotton and woolen goods and other manufactured articles from
Great Britain. In the southern colonies, where the coastal plain was broad-
est, there were few harbors and fewer ports. Here flourished the planta-
tion aristocracy as contrasted with the commercial aristocracy of the North.
The wide plains were favorable to the plantation system. Tobacco, rice,
indigo, and a little cotton were their exports, while woolen and cotton goods,
household articles, and slaves were the imports. The chief ports of this
area were Charleston in South Carolina and Richmond in Virginia.
PHILADELPHIA
In 1790 Philadelphia was first in population rank. It owed this suprem-
acy to its position on the Delaware estuary, in the midst of the best
farming lands surrounding any northern city. It was the chief commercial
port of the United States as well as the national capital until the year 1800.7
NEW YORK
New York in 1790 was second in population. The situation of New
York was not as good as that of Philadelphia because, in the absence of
steam navigation and railroads, it had a more restricted hinterland. In
1760 New York was the seventh colony in population, even North Carolina
having more people. Manhattan Island was not then considered the most
favorable site for a large city; however, as a commercial port. New York
was soon to vindicate itself. With its excellent harbor facilities and its
easy access to Europe, it soon took first rank in commerce, especially in
imports.
The Erie Canal, completed in 1825, was the response to the need of the
new settlers of central and western New York who sought a market for
their farm products. While it increased the movement of population west-
ward and gave large impetus to New York it by no means made that city
the first commercial center of the United States. New York took first rank
both in number of people and in the value of its commerce before the Erie
Canal was completed. Its superb harbor and an early appreciation of its
situation in relation to the western country made New York the chief
Atlantic port. The Hudson-Mohawk highway was known as the easiest
route through the Appalachian barrier some time before the Erie Canal
was completed.
7 The seat of the National Government was moved from New York to Philadelphia in 1790.
8 "A Century of Population Growth" (U.S. Census Bureau, Washington, D. C., 1909) gives Philadelphia
as the first city until 1810. See p. 11.
390
THE GEOGRAPHICAL REVIEW
There were other factors promoting the development of New York up
to 1850. The convention of 1822 gave New York state a more democratic
government. Suffrage was given to all tax payers, and the city on Manhat-
tan Island became liberal toward merchants. The potato famine in Ireland
and economic and political unrest in Germany drove thousands of people
to the New World, many of whom settled in New York. Five sixths of those
who took up residence in New York City from 1815 to 1865 were Irish or
German." Personalities have often counted for much in the growth of
American cities. An illustration is found in Cornelius Vanderbilt, who
came to New York in 1829. He commercialized Fulton's invention of the
steamboat. Between 1829 and 1848 he owned fifty steamboats which sailed
to all parts of the world.
BOSTON AND SALEM
Boston and Salem suffered relative decline after 1790. Salem actually
declined as a commercial port while Boston lost its rank to Baltimore.
Professor Davis points out that Boston outstripped Salem in great part
because it was nearer the interior of Massachusetts, the space being inade-
quate for the development of both.10 The Napoleonic wars injured New
England shipping. Jefferson's system of embargoes and later the War of
1812 dealt hard blows to her commerce. An expression of this commercial
depression is seen in the IIartford Convention of 1814, where New Eng-
land protested against the carrying on of the war with England.11 When
the Napoleonic wars were ended and shipping was made safe, England
actively competed with New England ship owners. Many shippers in this
period found New York more liberal and emigrated to that port. Manu-
facturing now took the place of commerce as the dominant occupation of
New England, her waterfalls taking precedence of her harbors. This
transformation was effected in the period from 1800 to 1850. Many people
from the interior of Massachusetts emigrated to the better farmlands of
the Genesee country of New York. Boston could not rival New York in
this age of internal canal traffic. The Berkshires impeded communication
with the western country. Even the railways did not entirely remove the
barrier, and Massachusetts declined at the expense of the southwest.12 The
rise of the manufacturing industry and Irish immigration saved Boston as
a large center of population.
BALTIMORE
Baltimore, the fifth city in 1790, soon rose to third rank, and became
an effective rival of Philadelphia. The situation of Baltimore near the
9 Theodore Roosevelt: New York (Historic Towns, edit. by Freeman and Hunt), New York, 1891, p. 184.
10 W. M. Davis: The United States of America, in "The International Geography," edit. by H. R.
Mill, London, 1899, p. 722.
11 The report of the convention did not advocate secession, however, as it has often been said. See
S. E. Morison: Life and Letters of Harrison Gray Otis, 2 vols., Boston, 1913; ref. in Vol. 2, pp. 158-159.
12 Henry Adams: A History of the United States from 1801 to 1817, 9 vols., New York, 1889-90; reference
in Vol. 9, p. 103.
GROWTH OF AMERICAN CITIES
391
13
head of Chesapeake Bay, with the Susquehanna Valley as a contributing
region, was most favorable. The farmers of the Great Appalachian Valley
in Pennsylvania were, in 1825, indifferent to the proposals of internal com-
munication between Philadelphia and the West. They had easy trans-
port for their products down the Susquehanna Valley to Baltimore. Its
position was also favored by the overland trade from the West. The
Cumberland Highway led to the Ohio country by way of Wheeling but did
not touch Pittsburgh.
WASHINGTON
It is significant that there is no large commercial city south of Baltimore,
in the broader sections of the coastal plain. Washington's growth as a
city has been due largely to the location of the seat of the National Govern-
ment within its borders. As a commercial and industrial city it has never
reached the hopes of its founders. "Washington," says Professor Chan-
ning, "was undoubtedly influenced in choosing the site for Washington
City by the expectation that the new city just below the falls of the Potomac
would in no long time rival Philadelphia, New York, and Baltimore as a
center of industry and would become the continental mart for ocean-borne
commerce-would be, in short, the metropolis of the Western World.''14
CHARLESTON
Charleston, fourth in population in 1790, soon began to lose rank. At
first during the period of economic readjustment after the Napoleonic wars,
South Carolina was in harmony with the policy of protection of infant
industries and a system of internal improvements. When, however, other
parts of the country grew at her expense, Calhoun changed his liberal
attitude toward protection and internal improvements and declared in the
document known as the "South Carolina Exposition" in 1828 that such
measures were unconstitutional. South Carolina's clash was with New Eng-
land, the chief supporter of protection. But was not Charleston's rival
New Orleans rather than Boston? In the decade 1810 to 1820 New Orleans
moved ahead of Charleston as a center of population and continued to out-
strip her in growth until 1840.
River Cities
NEW ORLEANS
The rise of the new West was the most significant fact in the years imme-
diately following the War of 1812.15 The population was occupying the
southwest beyond the Appalachian Mountains. This region triumphed in
the election of Andrew Jackson in 1828. The cotton industry expanded
13 F. J. Turner: Rise of the New West, 1819-1829, New York, 1906, p. 38.
14 Edward Channing: History of the United States, 8 vols., New York, 1905-17; reference in Vol. 4, p. 105.
15 F. J. Turner, op. cit., p. 67.
392
THE GEOGRAPHICAL REVIEW
from the Atlantic coastal plain and Piedmont region into the Gulf plains.
It was this phenomenal extension of cotton culture which determined the
history of the Lower South.16 Soon after 1826 the new Southwest took
precedence over the old South Atlantic states in the production of cotton,
and New Orleans exported more cotton than Charleston.
It is noteworthy that in the decade 1812 to 1821, before the effective use
of steamboats on the western rivers, seven states came into the Union. Six
of these are in the West and Southwest. There was much traffic on the
central western waters before the steamboat came. Flatboats carried
cotton, cattle, corn, wheat, flour, and tobacco to New Orleans. The real
effect of the steamboat, made commercially useful by Robert Fulton in
1807, was to push the area of production farther up the Mississippi Valley.
This efficiency was in large part due to the new possibility of upstream
navigation. This invention, like the Erie Canal, gave strong impetus to
a movement of population already begun.
New Orleans had a most favorable situation for this regional develop-
ment. It naturally became the exporting center for the products of west-
ern farmers. A break in transportation, always a strong factor in the
development of cities, made necessary a port near the mouth of the Missis-
sippi River. New Orleans never became a large importing city, but in
exports it ranked first several times between 1830 and 1840.
CINCINNATI AND LOUISVILLE
Situated in southwestern Ohio, gathering the agricultural products of
the rich Miami valley and surrounding region, Cincinnati, called the Queen
City, became next to New Orleans the chief city of the West in 1850. It
was the center of river shipbuilding, a depot for corn, wheat, and cattle,
and a distributing point for corn, woolen goods, and other manufactured
articles. Meat packing and the manufacture of flour were its chief indus-
tries. As the center of the tobacco trade of Kentucky and as a point of
transshipment about the falls of the Ohio, Louisville was in later years sure
to become an important center of population.
ST. LOUIS
Still more fortunately located, St. Louis was to be the chief city of this
part of the United States and the second city of the great interior plains.
Its early rise was related to the exploration and development of the Far
West. It became the chief point of departure for explorers and emigrants
who followed the western trails. With the rise of river transportation, it
was the gathering place for the surrounding farm lands. In the decade
1840 to 1850 the growth of St. Louis was marked by a large influx of
16 In the then existing ignorance of bacterial life this wide extension of cotton over a geographically
suitable region fastened negro slavery on the South. If it had been possible to eliminate the malarial
mosquito and the hookworm before Eli Whitney's cotton gin was invented, white and not black labor
might have been used in the South (Channing, op. cit., Vol. 4, p. 436).
GROWTH OF AMERICAN CITIES
393
Germans. From 1850 to 1860, when all the other river cities were yielding
their rank to the centers of the Great Lakes region, St. Louis was able to
hold its position and thereafter began to grow more rapidly. The strongest
contributing factors of later years have been the excellent corn and wheat
fields as well as the cattle lands about it. It thus developed the manufac-
ture of flour and beer and the meat-packing industry. By virtue of its
position it became also one of the terminals in the westward expansion
of the railroad. Later, with the rise of manufacturing industries, with
abundant raw materials, lead, zinc, glass sands, and iron ore, together with
coal from the rich fields of Illinois, the growth of St. Louis was assured.
PITTSBURGH
There was another population center which up to 1860 should be classed
as a river city, viz. Pittsburgh, whose favorable situation for western
internal commerce during the steamboat period made her a rival of
Cincinnati. Manufactured goods came overland from the eastern ports
and were sent by boat from Pittsburgh down the Ohio River. While its
lot was with the river cities, it rose with them and after 1850 lost rank
with them. Pittsburgh was early called a smoky city because of its coal
and iron industries, but these were not important until the growth of the
Great Lakes region became significant. Then, with the valuable stores of
fuel in western Pennsylvania, her position in relation to the iron ore fields
of the northern Great Lakes region and the transportation route of the
lakes, Pittsburgh was sure to be the site of a large industrial city.
It was not a strange prediction in 1841, in an age of river transportation
and internal commerce, that the largest city of the country would by the
end of the century be somewhere in the southern part of the interior plains
region on a river, either at Cincinnati, Louisville, St. Louis, or Alton.17
Later, however, in 1843, from the census of 1840 and the fact that the
movement of population was toward the lakes rather than the rivers, it
was predicted that the chief city at the end of the century would be in
the northern half of the central valley somewhere on the Great Lakes.18
Maumee, now Toledo, by reason of its good harbor, its central position on
the lakes, and its connection with the Miami Canal joining Cincinnati and
Maumee, it was said, would be this great city. Why did this prophecy
fail? Chiefly because it did not forecast the importance of railway trans-
portation.
Lake Cities
An important relationship exists between the physical character of the
lake plains and the development of centers of population. This section,
with its soils rejuvenated through ice invasion, with easy means of trans-
17 Charles Cist: Cincinnati in 1841, Cincinnati, 1841; reference in Appendix, p. 283.
18 IIunts Merchants' Magazine, Vol. 9, 1843, pp. 42-46.
394
THE GEOGRAPHICAL REVIEW
portation by water and by land, and with its mineral wealth, especially
copper, coal, and salt, and the iron ore of the Superior Upland Province,
is in natural resources the richest area of the United States. The expansion
of the railroad along the lake plains and later across the Great Plains to
the Far West, gave a market for these stores of wealth. Within this great
agricultural region between the years 1850 and 1900 certain large lake
ports developed. They gathered the products of the soil such as corn,
wheat, oats, barley, and rye. These were manufactured into flour and beer
and later into cereal foods. The cattle lands, together with the salt de-
posits of Michigan and Illinois, made possible the meat-packing industry.
Lumber and building materials were near at hand to facilitate city build-
ing. Iron ore and coal were abundant, to make possible the growth of the
great iron industries of the lake cities. In these decades thousands of
immigrants came from all parts of Europe to help build these cities.
The most significant growth of the lake cities came after 1880. Although
its first construction followed 1850, the Sault Ste. Marie Canal became com-
mercially effective after it was enlarged in 1881.19 The iron ore of the
Lake Superior fields was released in larger quantities; there was a marked
increase after 1880 in the per capita production of pig iron and coal. The
Bessemer steel process gave rise to the great steel industry. In these
decades the Great Northern and Northern Pacific Railways were built,
opening up the wheat fields of the Red River Valley and the Dakotas and
extending the cattle lands over the Great Plains. The development of these
natural resources increased rapidly after 1880.
CHICAGO
Chicago by virtue of its central position in the heart of this agricultural
and industrial region has become the second city of the United States. It
is another illustration of the principle that city growth is favored by a
position far inland on a body of water. Situated where land transportation
and water carriage meet, Chicago has an all-water route from the Great
Lakes to the St. Lawrence, with easy access to the Lower Mississippi settle-
ments.20
Laid out in 1830 it has become in 88 years the fourth city in the
world. Its situation in the prairies of the glacial belt, with coal, iron,
copper, lumber, and building materials at hand, and its position in relation
to routes of transportation explain this remarkable growth.
CLEVELAND
The growth of Cleveland, the second metropolitan center on the Great
Lakes, has been favored by its situation. It was but natural that Cleveland,
with its harbor and its location at the mouth of a river, should become the
19 J. A. Farrlie: The Economic Effects of Ship Canals, Annals Amer. Acad. of Polit. and Soc. Sci., Vol. 11,
1898, pp. 53-78; reference on p. 67.
20 E. A. Riley: The Development of Chicago, Chicago, 1911, p. 33.
GROWTH OF AMERICAN CITIES
395
northern terminus of the Ohio Canal connecting river and lake waters.
The westward migration of New England and New York farmers along
the lake plains, later stimulated by the railroads, aided the development.
of Cleveland. Its commercial and industrial growth have been due to the
necessary change between land and water transportation. As it lies on the
direct route of iron-ore shipments from the Lake Superior region to the
coal fields of Pennsylvania, it is not strange that over half the value of
its manufactured products should be in its steel industries.
DETROIT
The swift growth of Detroit in recent times is chiefly due to the manu-
facture of automobiles and shipbuilding. Detroit was early important in
the manufacture of carriages, owing to the supplies of hard wood near at
hand. When the horseless carriages came into existence, Detroit began its
automobile industry. While the city is favorably situated on the Detroit
River, where the traffic of the Upper Lakes passes to the Lower Lakes, it
is impossible to explain its growth chiefly on the basis of its geographical
position.
BUFFALO
Buffalo owes its early growth to its advantageous position at the edge
of the lake plains where Lake Erie discharges into the Niagara River. The
Erie Canal connecting the Hudson River with Lake Erie and later the
railroad leading from New York to Buffalo greatly favored the growth of
a city at this point. Owing to Niagara Falls, Buffalo is situated at a point
where traffic by the larger lake steamers ceases and a transfer is necessary
from lake to canal boats or from lake steamers to the railroad. Flour
shipments to the eastern ports have made Buffalo a center for grain
elevators. The iron and steel industries and lumber traffic and more
recently the use of electric power generated at Niagara Falls have favored
the growth of Buffalo.
MILWAUKEE
Milwaukee has a favorable position for commerce, but it is too near
Chicago ever to become one of the largest cities of the United States. Its
hinterland is more limited, and it is not on the great east-and-west line of
transportation. It owes its growth to the grain trade and to its manu-
facturing industries.
MINNEAPOLIS AND ST. PAUL
The metropolitan district of Minneapolis and St. Paul has a favorable
situation in relation to the wheat fields, is at the head of navigation on
the Mississippi River, and has available water power in the Falls of St.
Anthony. These conditions have been influential in developing the leading
flour mill center of the United States. This district shows how the manu-
396
THE GEOGRAPHICAL REVIEW
facturing industries move westward, following agricultural industry. The
westward migration of meat packing from Cincinnati to St. Louis and
Chicago and more recently to Omaha, Kansas City, and Lincoln, following
the progressive utilization of the cattle lands for farming, also illustrates
this principle.
Pacific Coast Cities
Population centers on the Pacific Coast had their beginning after 1850,
showed a marked increase after 1880, and experienced their most significant
growth after 1900. The transcontinental railways played much the same
part in the development of the Far West as did the steamboat in the
Mississippi Valley region. It is significant that a second decade in which
as many as seven states came into the Union was that between the years
1889 and 1898, when the Dakotas, Washington, Montana, Idaho, Wyoming,
and Utah became states. The railroad, following and augmenting the popu-
lation movement, made possible the marketing of the products of the soil
and the minerals of this region. The chief factors resulting in large popu-
lation centers on the Pacific Coast were the development and marketing of
products such as fruits, grains, fish, lumber, and mineral substances. Trade
with the Orient and with Alaska has also contributed to this movement of
population.
SAN FRANCISCO
The chief city of the Pacific Coast is San Francisco, remarkably situated
for commercial purposes on a peninsula at the Golden Gate, the entrance
to San Francisco Bay. The wharves are sheltered on the bay side of the
peninsula. The city proper, with its surrounding cities, especially Oakland,
gives San Francisco seventh rank, next to St. Louis, in 1910, as a metro-
politan center. The early growth of San Francisco was due to the discovery
of gold in California, to its environing fields, and its trade with the eastern
United States by way of Cape Horn and with the Orient. The Asiatic
trade was an apparent motive for the annexation of California.21
virtue of its position on San Francisco Bay, with room for commercial and
city expansion, of its nearness to the natural products of the California
Valley, such as fruits, grain, and oil, and of its being already the financial,
commercial, and industrial center of the Pacific Coast, San Francisco, it
is reasonable to think, will continue to be the metropolis of the Pacific
States.
LOS ANGELES
By
Los Angeles, though larger than San Francisco as a strictly municipal
center of population, is by no means as yet a commercial or industrial rival.
It labors under the disadvantage of an artificial harbor, built at enormous
21 R. G. Cleland: Asiatic Trade and the American Occupation of the Pacific Coast, Ann. Rept. Amer.
Hist. Assoc. for 1914, Vol. 1, pp. 283-289; reference on p. 287.
GROWTH OF AMERICAN CITIES
397
expense, and its hinterland is much more restricted. Its climate and the
neighboring fruit regions and oil fields together with its railroad connec-
tions explain the wonderful growth of this city of Southern California.
It would be contrary to the earlier tendencies of population movement
in the United States for the largest metropolitan center of the Pacific Coast
to grow up in Southern California. In the steady westward movement of
population centers the northern component has always finally prevailed over
the southern. First the Southern colonies were more thickly populated than
those of the North. Then after 1820 the Northern States became more
populous. The trans-Appalachian movement of population was first toward
the southwest after 1810, and not until after 1850 did the old Northwest
(Great Lakes region) become more thickly settled. Again, with the further
westward movement of people, first the Southwest and then the far North-
west received the greater number of inhabitants.
SEATTLE
There is a rival to the California region on the Pacific Coast. It is
found in the Puget Sound trough, with its contributing regions of the
Willamette and Puget Sound valleys. Seattle is an important competitor
of San Francisco in its foreign commerce, especially in the Oriental trade,
since it is nearer to the Orient. It has the further advantage of the Alaskan
trade, but this is offset by San Francisco's long-established relations with
the eastern United States and with South America. Furthermore Portland
is certain to share with Seattle the development of the immediate hinter-
land. Portland controls the Columbia River fisheries and the products of
the Willamette valley. Seattle's large growth has been due to its ocean
commerce, its railroad connections, and the development of its back country
with its resources of lumber, coal, grains, and fruits. Also it is well situatel
on Puget Sound for greater commercial expansion, and it will necessarily
benefit by the future development of Canada.
SUMMARY
Our cities have grown in a vast country suddenly thrown into the noon-
day of civilization, a country organized only a little more than one hundred
years ago with a free constitution and an invitation to the world to come in.
The continent lies open on its Atlantic side, and the tides of population
reached the Mississippi River, then passed the continental divide, and
moved on to the Pacific Coast. Ample space, fertile soil, and a varied
climate offered room and facile adjustment to all. The steamship followed
the canoe and the arkboat, and the packhorse and prairie schooner gave
way to the iron track and the locomotive. Plentiful crops laid the founda-
tions for population centers, manufactures relieved the New World from
undue dependence on the Old, and modern means of transportation made
these developments possible and enduring.
398
THE GEOGRAPHICAL REVIEW
Notwithstanding the powerful movement toward the West, the East has
not failed to hold its supremacy in numbers. In 1790 all the large cities.
of the United States were on the Atlantic border. In 1840, counting the
ten leading centers, eight were on the Atlantic fringe of the country and
two were on the rivers of the interior. In 1900 a review of the first ten
cities shows four on the eastern coast, two on the rivers, three on the Great
Lakes, and one on the Pacific Coast. Again, in 1917, the Atlantic Coast has
four centers, the rivers one, the Great Lakes four, and the Pacific Coast
The growth of the eastern cities, however, is seen at its true value
in the count, not of cities, but of souls. The total population of the four
Atlantic cities is somewhat more than 8,000,000. The six centers lying
westward show about 5,300,000. If metropolitan districts, notably those of
New York and Boston, be added, the disparity is nearly as two to one and
goes far to explain the dilatory movement of the center of population,
which, entering Indiana soon after 1880, has not yet passed its western
border.
one.
SOME INFLUENCES OF THE SEA UPON THE INDUSTRIES
OF NEW ENGLAND
By MALCOLM KEIR
University of Pennsylvania.
Although all the thirteen colonies had access to the sea, in none of them
was the influence of tidewater so potent as in the New England settlements.
Since the Middle and Southern colonies were better endowed with agricul-
tural wealth than New England, tillage of the soil was their foremost
activity. New England alone, poor in farm produce, turned from the
niggardly soil to the generous ocean, where fishing and commerce became
the sources of her greatness and gave rise to almost all of her other activities.
SHIPBUILDING AND SUBSIDIARY INDUSTRIES
Within a year of the founding of Boston the first ship, the Blessing of
the Bay, was upon the stocks, and from that time until the Civil War ship-
building was one of the most familiar occupations in New England. The
white pine that covered the region was especially adapted to sea service
because it was buoyant, staunch, and cheap. English oak was heavier, and
Baltic fir was more expensive; there was no wood that could compete with
New England pine. In 1791 the cheapest construction in France cost $55
to $60 a ton, whereas in New England ships could be built for $33 to $35
a ton. With the sea at their doors and with fishermen or merchants calling
constantly for more craft it is no wonder that shipyards flourished. In
1760 American yards were turning out yearly from three to four hundred
commercial boats, and a host of less important ones such as dories, sharpies,
and small sail boats. Before the Revolution, out of a total population of
approximately 3,000,000 there were no less than 4,000 men engaged in
shipbuilding.
The sea not only developed New England commerce and promoted
shipbuilding, but in addition it stimulated several subsidiary industries.
Among these was the manufacture of rope and cordage. The first rope-
walk was set up at Boston in 1641, but soon thereafter every port of any
consequence had its own ropemakers. A modern survival of these colonial
rope factories may be seen in the great plant at Plymouth, Mass.; this
mill within sight of the famous "Landing Rock" is now the greatest works
of its kind in the country. Another industry corollary to commerce was
food packing. Flour mills, biscuit factories, and salt meat packers were
established at the great ports such as Boston, Salem, Providence, New
Bedford, New London, and New Haven. The first milk cracker made in
America was a product of the Boss Biscuit Company, located almost upon
399
400
THE GEOGRAPHICAL REVIEW
the docks at New London, Conn. The present-day great pork-packing
plants at Somerville, Mass., are descendants of the little establishments.
that once fitted out deep-sea sailing ships. Along the waterfront of every
port, too, there were numerous tailor shops to make the clothing needed
by sailors. The ready-made clothing factories of Boston can trace their
ancestry to these primitive businesses.
IRON MANUFACTURES
The sea, then, was responsible for New England's prominence in ship-
building and the allied supply industries, such as ropemaking, food pack-
ing, and tailoring. It was the greatest influence as well in promoting the
growth of New England iron manufacture because it constituted in its
commerce the largest consuming market. We are not accustomed to asso-
ciate iron manufacture with New England; nevertheless Massachusetts
was the leader in iron production from 1650 to 1750. In all the other
colonies iron manufacture was a strictly local business using local raw
materials and supplying a small local market; but on account of her
maritime market New England's iron business was of a more commanding
character. The biggest item in iron making was the shaping of nails from
bar iron to aid in shipbuilding and the construction of dwellings in the
thriving ports. Inasmuch as nails were handmade and the demand was
large, it was obligatory for nearly every household to devote its spare time
of evenings to hammering out these essential articles. Indeed nails often
replaced money in bartering goods at rural stores. It is significant that
in 1790 in Massachusetts, where the demand was greatest, the first nail-
slitting machine was invented. A survival of the former art is found
today at Taunton, Mass., a town now without iron resources, yet the
American leader in the manufacture of tacks.
Besides nails there was a large market for iron in Massachusetts for
anchors, bells, cannon, and cannon balls-all vital parts of a ship's equip-
ment. The boat's cargo was usually confined in casks or barrels and these
were held together by iron hoops; indeed in 1795 one-half the iron output
was devoted to the manufacture of hoops. Iron kettles for refining the
sugar or molasses brought in by ships, and iron pans for providing the salt
that preserved the food for outgoing boats, both created a demand for
furnaces and forges. So, all told, the sea provided great inducements for
New England iron manufacturers.
WOOD MANUFACTURES
Ocean commerce also gave rise to the cooperage industry. The fish,
rum, flour, and other provisions that were carried out of the colonies were
packed in casks or barrels; the imports, sugar and molasses, arrived in the
same kind of containers, and even staves and barrels themselves became
articles of commerce. Cooperage consequently was a promising shore in-
dustry created by the deep-sea traffic. New England's resources in wood
THE SEA AND NEW ENGLAND INDUSTRIES
401
gave her such an advantage in barrel-making over European rivals that it
was claimed that cheap barrels were as good as an extra 15 per cent profit
to American merchants.
In response to the demands for wood along the shore the lumber industry
of New England was promoted. At first lumbering was entirely incidental;
that is, it was a necessary preliminary to clearing a space for farming. But
after shipbuilding commenced, lumbering, although it was largely con-
fined to the winter season, became a regular part of a farmer's activities.
The lumber was not only employed for ships and used in constructing the
towns that commerce favored, but also was a not unimportant article of
commerce, especially with the timber-denuded, plantation-covered West
Indies. Where wood as such was not shipped it often passed into com-
merce in the form of potash, for which there was a constant demand.
The burning of wood to potash or pearlash was especially convenient for
farmers in new clearings, for the timber that was otherwise a valueless
nuisance could be converted without much labor into a revenue producer.
DISTILLING AND THE REFINING OF RAW MATERIALS
Thus the influence of the sea was felt from the iron furnaces and ship-
yards along the water's edge to the remote inland clearings. Its potency
is equally apparent in the industries that sprang into existence upon the
ocean's border in order to refine the raw materials brought into New Eng-
land by returning cargo carriers. Two commodities, sugar and molasses,
were the chief return freight from the West Indies. In response to
abundant raw materials and a market greater than the production, sugar
refining and the distillation of rum from molasses flourished all along the
New England shore. Rum at that time was an especially valuable product,
in demand by farmers and sailors alike. It was as customary an article
of a ship's stores as the sailor's biscuit or salt pork; in outfitting a vessel
for a voyage it was usual to allow as much ($92 to $275 a trip) for rum
as for bread. Inasmuch as the more southern colonies, notably Pennsyl-
vania, turned surplus grain crops into beer and had no raw materials for
rum, the distilled spirit was an important article of domestic as well as
foreign trade. Massachusetts Bay and Narragansett Bay were famous for
their distilleries.
Less repugnant to modern ideals was another distilling industry created
by sea trade; namely, the manufacture of fish oil, whose principal source
of material was the whale fisheries. Since New Bedford, New London, and
Stonington were the chief home ports for the whaling fleets these towns
became the leading whale-oil centers. The fishing and commercial ships.
required large amounts of salt for packing the catch or preserving pro-
visions. This salt was refined from sea water, and the process was a well-
known occupation from Long Island to Marblehead. Another item often
listed in the cargo of home-coming boats was the cacao bean, the raw
402
THE GEOGRAPHICAL REVIEW
material for the refining of chocolate and its by-product cocoa. The first
chocolate factory was erected near Boston, and it is interesting to note that
a chocolate plant still operates upon the same site.
Among other imports, New England received considerable gold and
silver in payment for fish, rum, and lumber. The precious metal carried
into Providence, R. I., gave birth to the jewelry industry in that place;
and later, when an ingenious Yankee there made use of the art of plating,
the foundation was laid for the fame of Providence in the jewelry in-
dustry. Today the nation's greatest center for the manufacture of cheap
jewelry is the town of Attleboro, Mass., just beyond the boundaries of
Providence, and the latter city is still the principal selling agent for plated
metal ware.
Rubber manufacture is likewise largely confined to New England be-
cause her ships brought crude rubber to her shores along with other
curious products from out-of-the-way places. Boston, Providence, and
New Haven each received goodly shipments of raw rubber. Very little
use was made of it, however, until Goodyear and Hayward discovered that
an admixture with sulphur prevented rubber from softening under heat
and becoming sticky. Sulphur also gave rubber new properties and greatly
extended its application to industry. The new plants for the manufacture
of overshoes, fountain-pen barrels, hose, gloves, webbing, and dozens of
other rubber products rose near Boston, Providence, and New Haven and
may be most often seen in those districts even at the present day.
COTTON MANUFACTURE
These industries-rubber, jewelry, chocolate, and refining-however,
have never been New England's greatest manufacturing assets. Since
1800 first place must be given to textiles and shoe manufacture. Here, too,
the sea with its traffic has had an important bearing upon New England's
success in these branches of manufacturing.
As everyone knows, the inventions that placed textile making upon a
machine factory basis were made in England and not in the United States.
Although we had made crude attempts to spin yarn by machines at Beverly
and Bridgewater, Mass., and at Providence, R. I., and some attempts also
at Philadelphia and at Stateburg, S. C., none were successful until the
arrival of Samuel Slater, an English immigrant, especially trained in the
art of machine spinning. Slater first went to Philadelphia, but, receiving
no encouragement there, he tried employment in New York. In that city
a captain of a Providence sailing packet told him that Brown and Almy,
merchants of Providence, were experimenting with cotton manufacture
and wanted a man who understood machine spinning. From this chance
Slater went to Providence and under Brown and Almy set up in 1790 the
first successful cotton-spinning factory in America. Within twelve years
men trained in this mill established twenty-nine other cotton factories in
THE SEA AND NEW ENGLAND INDUSTRIES
403
nearby Connecticut and Massachusetts, thus making Providence the chief
seat of cotton manufacture in America. The raw materials for these early
mills were ocean-borne. The cotton did not come from our own Southern
colonies, for England bought our entire crop; but our early New England
mills were exclusively sustained by the commerce carriers plying between
New England ports and Dutch Guiana and Santo Domingo.
Beginning in 1814, when power weaving was first applied in a complete
factory at Waltham, Mass., the cotton industry passed into a second phase.
in which it retreated inland from the port towns to water-power sites. The
principal influence of the sea in this era was the supplying of capital.
Merchants engaged in commerce were the only persons in America who
had free capital to experiment in the new manufacturing ventures.
most famous cotton mills, such as those of Lowell and Lawrence in Massa-
chusetts, and Biddeford, Lewiston, and Saco in Maine, were all established
by merchants whose fortunes had been derived from overseas trade.
The
About 1840 cotton manufacture entered a third phase, wherein the sea
again became more dominant. At this time steam-driven mills began to
appear, first at Newburyport, Mass., and then at Salem. Coastal mills
using steam engines had an advantage over inland factories because coal
could be brought to their doors cheaply by water. As steam engines were.
improved the steam-driven mills steadily encroached upon the business of
the inland water-power factories, and after the Civil War the coastal mills.
wrested the crown of supremacy from the Merrimac Valley. Lowell, long
champion, had to bow before Fall River, the new leader. New Bedford,
neighbor to Fall River, has now outstripped her in quantity production,
just as she has long claimed the primacy in quality. Cheap transportation
of fuel has been of great advantage to these tidewater mills. They also
benefit by cheap carriage of raw cotton; for, although the material now
seldom arrives by water, potential competition of the sea keeps down rail-
road rates. Furthermore, the maritime climate benefits the coastal mills.
Since the rubbing that cotton gets in manufacturing creates frictional
electricity, the yarn twists and snarls in dry air, but a moist atmosphere
helps to disseminate the current; hence a high humidity is favorable to
cotton manufacture. The coastal mills gain this humidity at all times by
their proximity to the sea, and the frequency of fogs adds to the benefits of
the location. Humidity is even more necessary to American mills than
English because our machinery is belt-driven, whereas the English is geared.
As everyone who has been in an American factory knows, moving belts
create a large amount of electricity.
Our cotton industry, then, was sent to New England by a sea captain,
its mills were first supplied by New England's regular commercial carriers.
its later mills were equipped by commercial capital, its modern plants are
advantageously situated in respect to fuel and raw materials, and, finally,
the coastal factories benefit by a maritime climate.
404
THE GEOGRAPHICAL REVIEW
SHOE MANUFACTURE
Not only the cotton industry, but the shoe industry as well, has been
able to attain first place in New England because the sea was close at hand.
Although shoemaking was a local industry in all the American colonies, it
attained distinction in Massachusetts. Everywhere else the village cobblers
were hampered by a lack of raw material and confined to a local market
by the difficulties of transportation. Only at Lynn, Mass., were these
obstacles removed. Lynn lay between Salem and Boston, two of the prin-
cipal colonial ports. The boats that carried fish and other products away
from these harbors brought back cargoes of hides collected from scores of
places touched on the voyages. Some vessels made special trips to the
west coast of South America and California with the sole purpose of
gathering hides for the New England tanneries. Lynn as a result never
suffered from a scarcity of raw materials. Likewise the coastal trading boats
putting out from New England carried shoes to the other Atlantic colonies
and states. Shoes made from California hides often took a second voyage
around South America to be sold to the Mexicans of Santa Barbara, Cali-
fornia. Lynn took precedence of all other Massachusetts coastal towns in
shoemaking because by accident the subdivision of labor in shoe manufac-
ture was first applied there. This device enabled most of the sewing upon.
shoes to be passed out to women in their homes. The women most eager
to secure extra income were those in sailors' families, because the bread-
winners were often away for months or years upon extended voyages.
Lynn owed its shoe industry to the ocean transport of raw materials
or completed shoes, and to the labor force built indirectly by overseas
commerce.
Thus from New England's carliest industries, fishing, commerce, and
shipbuilding, down to the present leadership in cotton and shoe manu-
facture, the sea has been a potent influence. Several of the minor manufac-
tures, such as rubber products, jewelry, or refining, originated in New
England because the occan enabled the raw materials to be collected at
a favorable situation.
THE "OLD-FASHIONED" WINTER OF 1917-18'
By CHARLES F. BROOKS
Yale University
In the United States east of the Rockies the past winter was remark-
ably cold. Some of the details can be explained, even though the reason
for the occurrence can only be surmised. The early appearance of freezing
temperatures in the Mississippi Valley heralded a type of weather which,
with cold waves of increasing severity, overspread the United States east
of the Rockies and then held the country in its icy grip for two months.
Finally, a sudden warm spell saved from catastrophe the almost despairing
cities of the East. In the autumn, corn was caught unripe; winter wheat
did not make proper growth; cotton picking was interrupted early; truck
crops and fruit were damaged. In the winter, the already over-burdened
railroads were blocked by snow, or their locomotives were barely able to
keep up steam in the below-zero gales; while rivers and harbors were locked
with ice of unprecedented thickness. Even Nantucket was tied to the main-
land by a 15-mile ice bridge. Many industries, without their normal income
of supplies and outgo of finished products, were forced into partial shut-
downs or complete suspensions of work. People suffered from the severe.
cold, the lack of fuel threatening serious consequences with each renewed.
snowstorm or cold wave. Moreover, other parts of the world were also
experiencing record-breaking droughts, floods, storms, cold, or heat. What
was the cause? Variations in solar energy received by the earth must have
been basically responsible. The effects of a more energetic atmospheric
circulation are manifested differently in different parts of the earth, espe-
cially since snow-covered lands and unusual ocean temperatures are fre-
quently involved in the final results. While the present degree of solar
activity lasts, further extremes of weather are not unlikely. But these
cannot yet be forecast.
A survey of the winter just passed, however, does bring out some imme-
diate connections of the weather with local water temperatures and snow
cover, as well as the general tendency to extremes in terrestrial weather.
Before taking up the phenomena of the past winter, let us take a general
1 General references:
U. S. Weather Bureau Publications:
Daily weather maps of the United States, published at Washington.
Snow and Ice Bulletin, published weekly, December to March.
National Weather and Crop Bulletin, published weekly, April to September, and monthly, October
to March.
Climatological Data for the United States by Sections, monthly.
Monthly Weather Review.
Reports of the Associated Press, and special reports in the New York Evening Post.
Symons's Meteorol. Mag., London, containing a summary of British weather each month.
405
406
THE GEOGRAPHICAL REVIEW
view of the following points: (1) how the 11-year sunspot cycle is con-
nected with the earth's weather; (2) how changes in the intensity and
position of the "grand centers of action" of the atmosphere produce great
regional abnormalities in temperature and rainfall; and (3) how local
temperatures where water or snow prevails affect the weather.
THE 11-YEAR SUNSPOT CYCLE
2
The 11-year sunspot cycle is a cycle in solar activity, during which the
sun radiates the most heat when spots are most numerous. For some reasons,
however, the earth in general is colder than when the sun radiates less. At
any rate, as we should expect, the atmospheric circulation is most energetic.
when the sun is sending the most energy. Changes in the strength of the
atmospheric circulation necessarily go hand in hand with changes in the
steepness of the barometric gradients. In other words, the great high-
pressure areas are higher than usual and the low-pressure areas lower than
usual in times like the present, when the sun is hotter than usual.³
CHANGES IN THE "GRAND CENTERS OF ACTION" OF THE ATMOSPHERE
Now, let us consider the second point: How do changes in the intensity
and position of the "grand centers of action" of the atmosphere produce
great regional abnormalities in temperature and rainfall? In consequence
of the general circulation of the earth's atmosphere there tend to be belts of
high pressure at the horse latitudes-about 30° N. and S.-and troughs
of low pressure in subpolar latitudes. The contrasts between the tempera-
tures of the continents and oceans, however, produce the well-known ten-
dency to continental high pressures in winter and low pressures in summer,
while there is the reverse tendency over the oceans. In the northern-
hemisphere winter, therefore, we have such subpermanent centers as
the Iceland and Aleutian "lows" and the North American, Eurasian,
Hawaiian, and Azores "highs." These highs control the directions and
strengths of the general winds, and in this way they control the weather
characters of the different regions. With normal positions and strengths
there occurs the normal weather in the different regions, anomalous though
this weather may be when compared with the means for the latitude. What
temperature departures will be produced by an intensification of these cen-
ters of action? The places already too warm for the latitude will become
warmer and those too cold will become colder: for instance, western North
America should become warmer and eastern North America colder than
usual. Rainfall differences will be intensified in much the same way:
interior regions usually dry may become very dry with the higher pressure,
while peripheral regions usually wet may become extremely wet with the
stronger cyclonic activity. Increased circulation had general results of
2 C. G. Abbot: The Sun and the Weather, Scientific Monthly, Vol. 5, 1917, pp. 400-410.
3 H. Arctowski: Variations in the Distribution of Atmospheric Pressure in North America, Bull. Amer.
Geogr. Soc., Vol. 42, 1910, pp. 270-282.
THE WINTER OF 1917-18
407
this sort in North America during the past winter. But with shifts in
the positions of the centers of action, wind directions are changed, and
with winds from unusual directions the character of the weather will tend
to depart from the normal, in general corresponding to the difference be-
tween the usual weather brought by the new wind and that by the normal
wind. An apparent westward shift of the Eurasian high this winter gave
to Europe a continental climate in many respects like that of the interior
of North America.
CONTROL OF THE WEATHER BY SURFACE TEMPERATURES
4
At all times, the diverse temperatures of water, land, and snow sur-
faces control weather details, which, when long continued, become large
features. These surfaces affect not only the temperatures and moist-
ure content of the winds, but also control the paths and strengths of
cyclones and anticyclones. Water surfaces, whenever warm relative to
the surrounding temperatures, always become centers of cyclonic activity
and, therefore, are stormy. Snow surfaces and water surfaces, when rela-
tively cool become, on the other hand, areas of quiet coldness. Applying
such tendencies to the past winter in the United States the weather facts
fall naturally into place. On account of the high specific heat and mobility
of water, air temperatures in autumn over water tend to remain higher
than those over land. When the land is cold, as it was this winter, the
contrast may become so strong that unusual storminess results on the
border between the cold land and the relatively warm water. The potential
energy of the water vapor helps to increase the storm energy over what
the mere temperature contrast might produce. The limited maximum
temperature of a snow surface, its reflecting qualities, and its very low
conductivity favor the excessive cooling of the air. The snow-cover, when
once established, will tend to maintain itself in a number of ways. (1)
Under these conditions melting and evaporation are slow processes. (2)
The prevailing coldness of the air favors the occurrence of snow rather than
rain with passing storms. (3) The cold air is dense, and, therefore, favors
high pressure, which in turn will aid in the formation of storms on the
edge of the snow cover, where a temperature contrast is likely also.
Here is our background for the extraordinary weather of the past
winter: (1) a hot sun, therefore a general tendency to strong continental
winter anticyclones and ocean cyclones, with corresponding storminess and
coolness; (2) for North America a strong winter anticyclone favoring cold-
ness in the east and warmth in the west; (3) where there are strong tem-
perature contrasts, particularly great storminess, but where the contrasts.
are small, relatively quiet weather.
4 See, for instance, H. J. Cox: Influence of the Great Lakes upon Movement of High and Low Pressure
Areas, Proc. 2nd Pan-American Sei. Congr., Washington, Dec. 27, 1915-Jan. 8, 1916, Section II, Vol. 2, pp. 432-
459, Washington, D. C., 1917.
408
THE GEOGRAPHICAL REVIEW
AN UNUSUAL AUTUMN IN THE UNITED STATES
The advance guard of our cold winter appeared on August 8 when a
strong winter type of anticyclone, or "high," entered the United States
from the Canadian Northwest. Freezing temperatures occurred on the
morning of August 9 over a considerable area in the northern Great Plains.
Similar dry and clear and, therefore, cold high-pressure areas entered on
August 27, September 8, and September 22. The first produced the first
killing frosts in most of North Dakota, northern Minnesota, and Wisconsin;
the second cold period, September 9 to 14, brought killing frosts east of
the Mississippi and north of the 40th parallel. A procession of three more
"highs" from the Canadian Northwest entering the United States on
October 4, 7, and 11 so reinforced one another that freezing temperatures
were flung southward into Texas, southern Mississippi, and eastern North
Carolina-far beyond the regions where killing frosts had previously been
reported so early in the autumn. The sequel was a snowstorm in the south-
ern part of the Great Lakes region. A similar group entering the country
October 18, 20, and 22 pushed the frost line to the Gulf coast and into
Florida. There was another Great Lakes snowstorm, with light snowfall
south into half of Tennessee. The climax of the wintry autumn came
from October 28 to 31, when a tremendous cold wave with unheard-of
October temperatures swept out of the northern Rockies and across the
plains south to Mexico and east to the Gulf and Middle Atlantic coasts.
The temperature at Denver fell to -2° F. and that at Pueblo, Colorado, to
-4° F. Again, a Great Lakes snowstorm occurred. The continual cloudi-
ness and unusual storminess, particularly during the last ten days of Octo-
ber in the northeastern quarter of the country, were reflections perhaps of
the temperature contrast between the warm lakes and the prevailing cold
weather about them. From the Missouri and Mississippi Rivers eastward
this October was the coldest on record in most states and the wettest in
many. The snowfall was extraordinary for October. Many stations in
West Virginia had 15 inches, and central Wisconsin received 171½ inches
during the month. More than two feet of snow fell at the higher stations
in the Black Hills.
EXTREMES OF WEATHER IN OTHER REGIONS
Unusual weather was not confined to the eastern United States nor to
this continent. Over a large part of the United States west of the Rockies
October was rainless or almost so and considerably above normal in tem-
perature. Many places had the highest October temperatures on record,
e. g. San Francisco, 96° F., Needles, Cal., 112° F. A flood in China in late
September and early October is said to have been without parallel in
destructiveness of life and property. A typhoon, particularly destructive
in Tokyo, reported October 1, seems to have formed at the same time as a
similar storm which visited the southeastern United States two days earlier.
THE WINTER OF 1917-18
409
In Natal, on October 27-28, four months of abnormal rainfall culminated
in a storm yielding more than eight inches in 24 hours and producing
disastrous floods. The weather of the British Isles was not unlike that of
the northeastern United States: "low mean temperature, and a general
excess of rainfall with frequent and unseasonable snowstorms in northern
districts. ''5
NOVEMBER WEATHER
The first twenty days of November in the northeastern quarter of the
United States stood in marked contrast to the closing decade of October.
November was generally quiet and dry throughout the country. Over the
northern Great Plains, the warmest November on record was sandwiched
in between the coldest known October weather and an extraordinarily cold.
December. In much of this region November was actually warmer than
October, locally in North Dakota the mean for November being more than
4° F. higher than that for October. Whereas at Sheridan, Wyo., a tem-
perature of -12° F. had been recorded in October, the lowest in November
was +18°. Farmers were able to resume plowing and, near the Rocky
Mountain front, to continue with few interruptions till Christmas. The
weather of the British Isles also was mild, the temperature being only very
slightly lower than that of October. The south was extremely dry, but the
west of Ireland and northwestern Scotland were exceedingly wet. The
eastern third of the United States remained below normal in temperature
during November, as for many previous months. In the last of November
killing frosts occurred generally on the Gulf coast and into central Florida,
and in the eastern Great Lakes region a heavy snowstorm was followed by
record low temperatures for November, reaching -16° F. at Moira, N. Y.
DECEMBER WEATHER
December continued the weather type of October. A trial cold wave
entered the United States from the Canadian Northwest on December 3,
and on December 7 a more substantial one entered the northern Great
Plains just as a strong cyclone was beginning to operate over the southern
Great Plains. Considerable snow fell over the Plains south to Oklahoma
and central Arkansas. Then, as the storm advanced eastward and gathered
energy from the moisture of the Atlantic and the Great Lakes, the snow-
fall became heavier over the Ohio Valley, where on a strip less than 50
miles wide one to three feet of snow fell; and in the lee of Lake Ontario
parts of western New York were buried under two feet of snow. This
snow cover with later additions helped to maintain the below-zero weather
in the Middle West until December 17. The lowest temperatures in this
cold spell, except west of Lake Michigan, occurred in the immediate
5 Symons's Meteorol. Mag., Vol. 52, 1917, p. 119.
6 See the maps of daily snowfall in the cyclone of February 20-23, 1912, which took about the same
route, Monthly Weather Rev., Vol. 42, 1914, pp. 325-330.
410
THE GEOGRAPHICAL REVIEW
vicinity of the Ohio River on December 9, 10, and 11, before the deep
snow had increased in conductivity by becoming packed. The extremes
were -25° F. in western Tennessee, -20° in western Kentucky, -30° in
southern Indiana, and -31° in southern Ohio. In western Tennessee the
four stations with minima of -20° or lower were in a narrow belt imme-
diately southeast of the heaviest snow belt but still within the region with
six inches or more of snow on the ground. Cairo, near the windward edge
of the snowy belt, had a minimum of -7°; while Nashville, with but three
inches of snow and lying a short distance southeast of the cold strip,
reported zero as the lowest.
NEW LOW RECORDS OF TEMPERATURE
Owing to a thaw in the Middle West from December 18 to 24, as some
of the northern cyclones were able to cross to the Great Lakes and east-
ward, the snow line temporarily receded to the Great Lakes region and the
Appalachians, only to return as on December 27 the greatest anticyclone
of the winter began slowly to settle over the country. In the Yukon
probably a new minimum temperature for the North American continent
(-86° F.) was reported at the mouth of the Pelly River. In the Appa-
lachian and Atlantic coast region, where the snow still remained on the
ground, this cold wave established new low December records for most
states, and at many stations (New York City and Boston among them) the
weather was the coldest ever recorded since 1870 at least. Except for Ken-
tucky and Maryland the following minima established new state records for
December, during the past twenty-five to thirty years at least: West Vir-
ginia, 37°; Maryland and Tennessee, -32°; New Jersey, -29°; Virginia,
-27°; North Carolina and Kentucky, -21°; South Carolina, -6; Michigan,
-38°; South Dakota, -45°; Iowa, -40°; and in New England a range from
-45° in Maine to -16° in Rhode Island.
The month as a whole was either the coldest or the warmest December
on record in about two-thirds of the United States. The New Jersey sec-
tion summary considers that it will "serve to refresh the memory of the
'oldest inhabitant' regarding an 'old-fashioned winter.'"' In Ohio, a 64-
year record fails to show a colder December. The "ice conditions on the
Ohio River at Cairo exceeded in severity anything in the memory of the
oldest inhabitant" (Illinois section summary). In marked contrast, all
of the United States west of a line fifty to two hundred miles cast of the
Rocky Mountain front was warm. In southern Idaho the mean for the
month was locally 12° above normal: it was so warm that winter grains,
alfalfa, clover, and grasses made considerable growth. And yet only three
hundred to four hundred miles to the northeast Havre, Montana, was 13°
below normal. The immediate cause of this sharp contrast was the prev-
alence of southwest winds across the mountains and of northeast winds
on the Plains.. For most states December was a dry month, for some the
THE WINTER OF 1917-18
411
driest December on record. From Texas to Nevada and California practi-
cally no rain or snow fell. In Texas, it was the driest of sixteen consecutive
dry months. On the other hand, western Washington, western Oregon,
northern Idaho, and northwestern Montana were extremely wet. One sta-
tion in Washington, Cedar Lake, reported 46.80 inches of rain during the
month.
JANUARY WEATHER
Like December, January was cold and snowy but dry in the central and
eastern United States. The presence of the extensive and deep snow cover
seems to have been primarily responsible for this cold weather, in spite of
subnormal pressures throughout the eastern half of the country. In fact,
there seems to be little occasion to consider the weather of this month as
being under any control beyond the immediate effects of the land, snow,
and water temperatures and moisture supply. As in the region east of the
Mississippi in December, the greatest departures of January temperature
from the normal occurred in those areas with the greatest depth of snow
on the ground, if the upper Great Lakes region and the northern Appa-
lachians, which are normally snow-covered, are left out of consideration.
As expected, a series of intense cyclones which followed in a general way
the southern edge of the snow-covered area produced snowstorm after snow-
storm for some distance southwest and south of the Great Lakes and on
the Middle Atlantic Coast. Many cyclones entered the warm, moist Great
Lakes region and produced almost daily snows on the lake shores. In the
Middle West the blizzard of January 12 will long be memorable as "that
cold Saturday" for its southwest gale with driving snow at -20° F. On
the main line of the Illinois Central Railroad in central Illinois, it was
reported that there was no traffic to or from Chicago for forty-eight hours.
In the south, snow fell into the Gulf of Mexico (the snow at New Orleans
being the first in thirteen years), and in Texas south almost to Brownsville.
The very intense cyclone responsible crossed the central United States along
the southern edge of the snow cover and then moved northward to the
favorable temperature and moisture environment of the Great Lakes. The
cold air was supplied by an intense anticyclone in Alberta on January 10.
The heaviest snowfall during the month occurred near the southern end.
of Lake Michigan, notably at Milwaukee and Chicago, where the falls of
52.6 and 42.5 inches respectively exceeded the record for any month and
the normal amount for a whole winter. There was no rain at Chicago
during the month. This heavy snow on the southwest border of Lake
Michigan was brought principally by cyclones passing on the south. It
was snowier here than elsewhere south of the Great Lakes because the
northeast snow-bearing winds had over Lake Michigan the greatest unob-
structed sweep over a water surface. Topographic influences and more
intense cyclonic activity produced another belt of extraordinary snowfall,
412
THE GEOGRAPHICAL REVIEW
locally four feet, on the Ozark Plateau and over the Ohio River. The
Middle Atlantic coast continued snowy because of the numerous cyclones
passing over or up the coast, between the Gulf Stream and the cold, snow-
covered land. With so much snow on the ground it is not surprising that
in the Ohio Valley and eastward the temperature rarely rose above freezing
and occasionally fell below zero. Extremely cold weather prevailed in New
England through the month. The average temperature for the section,
13.2°, was the lowest for any month since 1888 at least. "In a record
extending back to 1836, at Amherst, Mass., the average temperature of the
present month was lower than that of any month heretofore. The greater
portion of the precipitation was snow, which, in the interior of the north-
ern portion of the section, accumulated to between three and five feet in
depth."" Much snow, cold, relative dryness: these are interrelated ele-
ments. The Ohio section report states: "It is safe to say it (January)
satisfied to the last degree the longings of the most fastidious for an 'old-
fashioned winter.'
FEBRUARY COLD WAVE
A final severe cold wave which occurred early in February in the north-
eastern United States was the coldest of the winter at many places. Even
at Provincetown and Nantucket the temperatures fell to -6° F. The ice
in Long Island Sound was said to have been the heaviest in more than fifty
years. It was reported that for the first time known the main channel
used by Boston to New York ships between Nantucket and Cape Cod was
closed by a solid bridge of ice. Some tugs were icebound for twenty days.
THE WINTER IN EUROPE AND ELSEWHERE
European weather was not unlike that experienced in the eastern United
States. The British Isles in December had low mean temperatures and
deficient rainfall. London had the coldest December since 1890. The
winter weather continued with severity till the middle of January, after
which the weather was remarkably mild. The cyclone which brought the
first general snowstorm in the United States, December 7-10, seems to be
the same which with a northwest gale, snow, and extreme cold unroofed
houses in Queenstown on December 15 and then tied up the traffic in and
about Paris on December 16. On January 2, a Paris report mentioned
another heavy fall in central and eastern France which snowbound the
region, St. Etienne having three feet. The temperature at Lyons was at
zero (F.)-an almost unheard-of cold for that region. Paris was experi-
encing less snowy and somewhat milder weather. A January 10 report
from Stockholm told of unprecedented temperatures in northern Sweden, a
minimum of 70.6° F. occurring at Asele. Early in February Madrid
reported one of the most severe winters in years, following an autumn with
7 New England section of the "Climatological Data," January, 1918.
THE WINTER OF 1917-18
413
""
no rain. "The snowfall in Asturias has blocked up the Puerto Pajarez,
and the trains remain for days in the snow-drifts, unable to proceed.
In Madrid the government departments had no coal, Seville had no gas,
Valencia no wood. From the southern hemisphere came the reports that
late in January and again early in March tropical cyclones produced great
damage in Queensland. In the middle of February, Natal and Johannes-
burg were again flooded by rains, which at some places amounted to nine
inches or more in twenty-four hours.
With similar cold weather in North America and in Europe, it seems
reasonable to point to a world-wide connection-one possibly controlled by
unusual conditions on the sun. Proof must await the reduction of the
solar radiation values for this period. The weather proved characteristic
of a winter period under the influence of an active sun.
SUDDEN WARMTH IN THE EAST
In the eastern United States, the weather moderated immediately after
the early February cold wave. The temperature at Cleveland, for instance,
rose from -12° to +40° F. in twenty-four hours. From February 6 to
14, the temperature in the middle Mississippi Valley and the middle of
the northern Plains averaged 15° to 20° above normal. The snow cover
disappeared with almost unbelievable rapidity north nearly to the Great
Lakes and along the Atlantic coast to New England. Although there were
other cold waves, this general moderation of the weather marked the end
of the extraordinarily cold and snowy winter which had dominated the
central and eastern United States since early December. If it had not
been for the snow cover, it seems as if the change would have occurred ten
days earlier, when the first "lows" began to cross the northern United
States first to the Great Lakes and then to the Atlantic. Earlier in the
month three such storms had gone to pieces over the northern Plains and
re-formed in the warmer South.
SUDDEN CHANGE ON THE PACIFIC COAST
There was a change in the weather type on the southern Pacific coast
at the same time with that in the eastern United States. During the fall
and winter, Utah, Arizona, Nevada, California, and some of the surround-
ing regions were experiencing an almost unbroken drought and unusually
warm weather. A severe shortage of water for irrigation was threatened,
for there was scarcely any snow on the mountains. For instance, the snow
on the ground at Summit, Cal., at the end of January was but two inches
as compared with the usual six feet at this time of year. But at the same
time that the heat was turned on with such unexpected suddenness in the
East, the water was turned on in the Southwest. Central California's first
heavy rainstorm of the winter came from February 4 to 6, and heavy rains
commenced in southern California on February 17. Until these heavy
414
THE GEOGRAPHICAL REVIEW
rains occurred, San Francisco had had less than 20 per cent of the normal
winter rainfall and Los Angeles less than 10 per cent. All talk of a water
shortage was silenced by a heavy fall of snow on the Sierras, which at
Summit amounted to about ten feet for the month. Aside from any direct
solar effects the relations of water to air temperatures here may have been
sufficient to change the weather type. For a few months the winds on the
Washington and Oregon coast had been rather strong from the southwest,
especially in December. If these winds extended farther south offshore it
seems reasonable that the warm water driven north should after a time
have favored cyclonic action, especially since the lands had cooled to
temperatures probably below those of the water.
Another aspect of water temperatures is manifest in the spring follow-
ing such a severe winter. For example, in the northeastern United States
the heavy snow on the ground in the northern highlands and the coldness
of the Great Lakes and of the coastal waters north of Hatteras continued
to favor dry weather even into April. Similar conditions have been
observed in the Baltic region after a cold winter. Does this explain the
dry spring weather of the western battle front this year?
SUMMARY
8
Such were the characteristics of our remarkable winter. The causal
sequence of events seems to have been as follows: In the first cycle: great
continental anticyclone-clear, dry air-strong radiation-cold-snow with
the next passing cyclone. The second cycle is more complicated and longer:
another anticyclone-intense cooling of the snow surface and overlying
air-extreme cold-dense lower air-maintenance or increase of high pres-
sure-marked temperature contrast between snow cover and the bare
land or open water-strong cyclonic action with more snow near the border
of the snow cover. And then another intense anticyclone begins a third
cycle like the second. Finally, long-continued southerly winds eliminate
most of the snow cover, and then the warmth of the early spring sun shining
through the dry air prevents the recurrence of another winter cycle. The
anticyclones thus initiate a long train of weather reactions. Was each of
these anticyclones produced by a sharp increase in the solar energy received
by the earth? A periodic recurrence noticeable in the highest pressures
observed in the region embraced by the United States weather map is sug-
gestive of such a connection. The past "old-fashioned" winter, then, may
have been caused by a sun hotter than usual which brought on self-per-
petuating cold waves and droughts.
9
8 H. H. Hildebrandsson: Quelques recherches sur les centres d'action de l'atmosphère, Kungl. Svenska
Vetenskapsakad. Handl., Vol. 51, 1914, No. 8.
9 H. H. Clayton: Effect of Short-Period Variation of Solar Radiation on the Earth's Atmosphere,
Smithsonian Misc. Colls., Vol. 68, No. 3, May, 1917, p. 18 (reviewed in Geogr. Rev., Vol. 5, 1918, p. 244).
GEOGRAPHICAL RECORD
AMERICAN GEOGRAPHICAL SOCIETY
Presentation of the Cullum Geographical Medal to Professor Frederick H.
Newell; Meetings of April. At a monthly meeting of the Society on April 23, at the
Engineering Societies' Building, 29 West Thirty-ninth Street, Professor Frederick H.
Newell of the University of Illinois, the former Chief Engineer, later Director, and now
Consulting Engineer of the U. S. Reclamation Service, was presented with the Cullum
Geographical Medal.
President Greenough presided. After he had submitted the names, approved by the
Council, of 151 candidates for Fellowship, all of whom were confirmed as Fellows of the
Society, the award took place. In presenting the medal President Greenough spoke as
follows:
"The practice of introducing water to fertilize the land is as old as history, but its
adaptation to existing conditions illustrates the complexities of modern life. In this
country it has involved not merely the use of the ordinary flow of our streams but the
impounding of their flood waters in vast reservoirs and the diversion at times of their
entire currents, together with complicated collateral issues. It is clear that operations
of such magnitude and character could only be conducted by Government agency, and
accordingly, in 1902, Congress created the Reclamation Service, working in conjunction
with the Hydrographic Branch of the U. S. Geological Survey to carry out the national
policy for irrigation of the public lands.
"Our guest of the evening was made Chief Engineer and later Director of the
Reclamation Service, and chiefly to his labors during the past fifteen years is due the
successful prosecution of a vast system of public works unsurpassed in their design and
influence.
"In President Roosevelt's annual message in 1901 he asserts that 'the forest and
water problems are perhaps the most vital internal questions of the United States.' And
his Secretary of the Interior sets forth the importance of providing through wise
administration for the creation of homes for millions of people upon the arid but fertile
public lands.' These quotations outlined the responsibilities with which the Reclamation
Service was intrusted and which it has continuously exercised. Their execution com-
prised not only the building of great engineering structures but also the determination
of even more difficult questions of soil, location, and settlement, together with con-
siderations of an economic, legal, and social nature, often most intricate and delicate.
Suffice it to say that all difficulties have been triumphantly surmounted, and the Nation
is justly proud of the achievements of a Service which has transformed a wilderness
into a garden, with provision for its distribution in suitable holdings amongst permanent
settlers. With every one of its accomplishments the name of Frederick Haynes Newell
is linked as the principal agent, and it is our privilege tonight to pay tribute to his
career of devoted and successful labor in behalf of his country.
"And now, Sir, as President of the American Geographical Society, I ask your accept-
ance of the Cullum Geographical Medal. The inscription on it attempts to epitomize your
work and its beneficent results. It reads:
FREDERICK HAYNES NEWELL
ORGANIZER AND DIRECTOR OF THE
UNITED STATES RECLAMATION SERVICE
1907-1914
HE CARRIED WATER FROM A MOUNTAIN WILDERNESS
TO TURN THE WASTE PLACES OF THE DESERT
INTO HOMES FOR FREEMEN
"We trust that this memorial will not be unwelcome, representing, as it does, the
appreciation, by our thirty-eight hundred Fellows, alike of your conspicuous creations
and of their humanitarian accompaniment.'
In accepting the medal Professor Newell said:
"The receipt of this medal gives keen pleasure not merely as a recognition of indi-
vidual effort, but far more than this as a visible evidence of public appreciation of the
notable efficiency of a group of men and women-public servants-united in putting into
material form an ideal to which all have been devoted. This ideal, founded on geo-
415
416
THE GEOGRAPHICAL REVIEW
graphical studies, has been that of creating homes for landless citizens, not by depriving
others of their portion, but by utilizing the waste places and the otherwise lost or even
destructive forces of nature.
"This ideal, furthermore, has been embodied in an organization-the Reclamation
Service-founded in the determination to demonstrate that every one of its employees,
whether engineer or clerk, could and would consecrate his life to the highest good of the
public. As a result there have been few organizations comparable with this service in
mutual respect and unity of action combined with sound business methods. This has
been displayed in investing a hundred million dollars of the national wealth in such
way that the annual returns measured in dollars are nearly 100 per cent, but where the
less tangible but more important profits in better citizenship are immeasurably greater.
Bound together by strong ties of true missionary spirit, this Service has persisted in its
loyalty to the high ideals and to the men who have aided in making these real.
"On this occasion especial mention should be made of the lifelong devotion to this
cause of the late Senator Francis G. Newlands, whose name is inseparably linked with
the original law, and of the untiring activity of George H. Maxwell, who sacrificed the
best part of his life to make plain to the people the objects and benefits of the Newlands
bill: nor could anything have been accomplished had not President Roosevelt thrown his
strength into the passage of the measure and toward the initiation of the works which
were thus made possible and in whose successful planning and execution scores of devoted
public servants, each notable in his or her devotion, have taken part.""
After the presentation, Professor Newell addressed the Society on "Home Making
in the Arid West." In the course of his illustrated lecture Professor Newell pointed out
that over a hundred million dollars has been invested by the Government under the terms
of the Newlands bill; already food products to that value are coming back annually, and
opportunities made for thousands of homes. The important point is that experience has
been had and a demonstration made as to the practicability of operations on a still larger
scale to provide more homes and more farms for our population and for the other
thousands of men and women who must look for new occupations. Moreover, as shown
by this experience, it is possible to do this without additional taxation and without
additional public debt, by wisely using the security afforded by the developing land
values-the reclaimed lands and their products gradually paying the cost.
The future prosperity and comfort of great numbers of home-makers and an adequate
supply of the necessities of life to countless citizens will follow the systematic con-
tinuance of the work thus begun of bringing together the landless man and ❝manless''
land and making the union fruitful by the control of the otherwise destructive floods,
storing and distributing the waters to the drought-stricken soil, or relieving the fertile
swamp areas of the harmful excess.
At an inter-monthly meeting two weeks before, on April 9, at which President
Greenough presided, Dr. Toyokichi Iyenaga addressed the Society on "Japan of To-
day.",
Dr. Iyenaga's late connection with the Department of Foreign Affairs of Japan
and his present position as managing director of the East and West News Bureau fit
him peculiarly as an interpreter of his country, and the lecture gave evidence of this
qualification in a high degree.
NORTH AMERICA
The Biogeographical Contrast between the Coastal Plain and the Piedmont.
The "fall line' has often been discussed in its economic aspect. Contrasted human
environment on plateau and coastal plain has also been pointed out.
A recent paper by
Professor Spencer Trotter (The Life Features of the Coastal Plain and the Piedmont,
Trans. Wagner Free Inst. of Sci., Vol. 8, 1917, pp. 33-44, Philadelphia), shows how this
line has served as a frontier between the widely differing types of plant and animal life
that characterize the Piedmont border and the low coastal lands. The author has sought
to discover the underlying cause of such marked diversity in regions that lie side by side.
The district about Philadelphia is selected as typical of this biological borderland. At
this point, as in many places along the line that divides plateau from plain, the two types
of life are in juxtaposition, yet each is well restricted to its respective zone. On the
uplands just north and west of the city are found a fauna and a flora belonging, accord-
ing to Merriam's classification, to the Transition type that separates Boreal and Austral
life in North America, while on the lowlands immediately southeast the species of both
plants and animals are distinctly Austral in character though far north of where such
orders generally occur.
As regards the characteristic animal life of each zone it seems evident that it has
become established there simply because of having found its accustomed vegetational
GEOGRAPHICAL RECORD
417
habitat. The point of greatest interest in the author's paper is his explanation of the
sharply contrasted flora on opposite sides of the city. This is due not so much to dif-
ference in altitude or to the present difference in climate as to past geologic conditions.
The controlling influences in the location of the so-called interior hardwood forest that
characterizes the Piedmont border were the secular changes of climate that took place
over the continent during and after the Glacial Epoch. Alternating periods of cold and
warmth account for the present distribution of our northern and transitional forests, the
latter being a remnant of the warm-temperate flora which in Miocene time enveloped
even the circumpolar area, but which, in periods of lower temperature, found conditions
in the north unsuited to its survival except on the borderland of the Piedmont. The prin-
cipal factor in the development of a truly Austral type of flora along the coast so far
north as Philadelphia was the gradual accretion of the coastal plain by the uplifting of
the formerly submerged Atlantic seaboard. Upon these recently emerged lands the
ground was occupied by plants from the south that found congenial soil in the sands
and clays of the new-made territory. Thus the existing flora is due to causes reaching
far back into geologic history. Such a perspective is essential in an understanding of
the relation of the earth to the organic forms of today; it is necessary not only to con-
sider the influences now operating, but also to recognize the very ancient impress of
factors, which, originating in the distant past, have gradually produced our seemingly
stable species and varieties of life.
Spring Cold Spells in New England. In Continental Europe a cold or frost spell
believed to occur about May 10 is known as the "Ice Saints." A study of the long-
period temperature records kept at New Bedford, Mass., between 1813 and 1905 by
S. Rodman, Jr., and Thomas R. Rodman was made by the late Waldo E. Forbes, and pub-
lished since the death of the author. The object of this investigation was to discover
evidence for or against the occurrence of such a cold spell in New England (Annals
Astron. Observ. of Harvard College, Vol. 83, Part I, 1917). The result of Mr. Forbes's
analysis of the records may be stated briefly. The theory of the "Ice Saints' at present
has little significance. Cold weather as well as hot weather may be expected on May 10,
and hot weather as well as cold on May 7 or May 13. "It is nevertheless possible,
the
author concludes, "that when the pulsations of the weather are better understood, May
10 may prove to be a sort of node and may serve as a point of departure for the study
of weather waves.
R. DEC. WARD
The Forests of the Isthmus of Panama. Since 1910 Dr. H. Pittier has been
engaged in studying the flora of Panama for the general biological survey organized by
the Smithsonian Institution. In the January, 1918, number of the Journal of Forestry
(pp. 76-84) he briefly describes some of the results of his work in a paper entitled "Our
Present Knowledge of the Forest Formations of the Isthmus of Panama."
The con-
tinental divide separates Panama into two distinct vegetational regions, the rain-forest
area of the Atlantic slopes, the savana and savana-forest of the Pacific side. The dis-
tinction follows the distribution of the rainfall, which in turn is dependent on the
prevailing winds. The dominant northeast trades bring abundant rainfall, well dis-
tributed throughout the year, to the Atlantic slopes: the Pacific, on the contrary, is
dependent for rain on the south wind, blowing irregularly from May to November, and
thus suffers a long dry season. Panama City has an annual rainfall of 70 inches dis-
tributed through 180 days; Colon, with a low average for the Atlantic on account of its
location in front of the great gap through the divide, has 129 inches per year in 246 days.
The evergreen rain-forest of the Atlantic slopes is characterized by its exceedingly
varied composition. Up to the present 900 species of trees, not including shrubs and
woody vines, have been listed. Save in the vicinity of roads and settlements the forest
is entirely of primary origin. Its most complete development is attained on the alluvial
flats of Darien. Such woodland as exists on the Pacific slopes west of the Panama Canal
is best described as savana-forest, and it is more properly studied with the grass forma-
tions. East of the canal, that is in Darien, where rainfall becomes successively more
abundant and better distributed, the forest changes to the type described by Schimper as
monsoon forest. Its most distinctive tree is the cuipo, which here reaches its most
northerly limit. It is unusually abundant and gregarious on the low hills of Darien in
the Canal Zone. Eastward and southward the forest passes into the heavy rain-forest of
the intensely humid Pacific coast of Colombia.
Little work as yet has been done on the forests of high altitudes. In the western
part of the Republic of Panama where high altitudes are attained the evergreen oak
becomes dominant above 8,000 feet, from which elevation it forms extensive groves up
to the limit of the timber line.
The commercial prospects of the timber lands of Panama have been shortly described
in an article by Forbes Lindsay (Bull. Pan American Union, Vol. 36, 1913, pp. 499-510).
418
THE GEOGRAPHICAL REVIEW
EUROPE
Franco-Italian and Franco-Swiss Ports in France. Though Italy is developing
the hydro-electric resources of her Alpine terrains, her industries are still largely
dependent on imported coal. The pre-war import of coal amounted to some 10,000,000
tons, of which about 90 per cent came from the United Kingdom. To maintain the sup-
ply by the ordinary long sea route has become increasingly difficult, and relief has been
sought in the establishment of a new route via the Gironde ports, whereby the sea journey
has been reduced by more than three-quarters, and overland. A good deal of the diverted
traffic is discharged at Bordeaux, but, to facilitate handling, a new modern port has been
created out of the old and decaying harbor of Blaye on the right shore of the estuary.
All transport equipment has been supplied by Italy, the dockers are Italians or prisoners
of war, the French language has been replaced by Italian: to all intents Blaye has become
an Italian town (Auguste Pawlowski: Blaye, Port Franco-Italien, La Nature, Jan.
12, 1918).
French hospitality has been extended to the neutral as well as to the ally. Her prin-
cipal outlets, Hamburg and Antwerp, lost, Switzerland has turned for help to France.
The port of Cette, 325 miles by rail from Geneva, has been put at her disposal. Unlike
the state of affairs at Blaye imports in transit at Cette embrace almost the entire category
of merchandise, except coal, which Switzerland gets from Germany. Food-stuffs naturally
take first place, with wheat the chief single item (idem: Cette, Port Franco-Suisse,
La Nature, Feb. 9, 1918; see also Suppl. to Commerce Repts. No. 5a, March 12, 1918).
The Internal Migrations Underlying the Present Distribution of the Serbo-
Croats. Is race the great nation-making factor? Does ethnography supply the solution
of the Balkan problem? This factor is at least one of profound importance and pro-
found difficulty. The complications of Balkan ethnography are apparent enough on the
ordinary ethnographic map, but they become far more so when examined in such detail
as is shown by Professor Jovan Cvijić in a recent paper on the racial migrations within
the Balkan Peninsula (Les mouvements métanastasiques dans la péninsule des Balkans,
with map showing the migration currents, 1:3,000,000, Le Monde Slave, July 1, 1917,
Paris).
The Balkans have known no rest. Ignoring the racial diversity produced by movements
of earlier epochs we still find great modifications arising from the migrations that have
taken place continuously in the peninsula from the Turkish invasion of the fourteenth
century to the present day. On these movements historic documents shed insufficient
light. They recognize such great mass movements as that which towards the end of the
seventeenth century carried 37,000 families from Kossovo beyond the Danube, but the
smaller migrations involving perhaps only a few families are naturally unrecorded,
though their succession has been continuous. But they live in tradition, and they may
be read in customs, dialects, folk-names and even in written form in carefully pre-
served family trees."
During the last twenty years Professor Cvijić and his collaborators have carried on a
systematic anthropogeographic-ethnographic inquiry throughout the western and central
parts of the peninsula, with the result that today the origin of the greater number of
families in these regions is known and charted (the chart remains in a town of invaded
Serbia).
The migrations whose courses have been thus determined influenced a great area, but
most actively affected Serbia. They may be grouped into four main currents. The most
powerful, denoted as Dinaric, carried peoples of the Syenitza region (between the Lim
and Ibar Rivers) and of Bosnia and Herzegovina north, northwest, and west. One of
the main branches led to a new peopling of the fertile area, the Shumadiya, between the
Drina and Morava. This forest-clad country, isolated from Constantinople, always
enjoyed comparative freedom and was the first portion of the peninsula to be freed
from the Turkish yoke. The same migration continued across the swampy Morava valley
and established itself in the mountainous districts of eastern Serbia. The direction,
transverse rather than longitudinal along the great highways, is characteristic. Another
branch crossed the Save and dispersed in Croatia-Slavonia. In its western and northern
spread it even attained the confines of Laibach (Carniola) and Vienna. Today descen-
dants of its members are said to form half the population of Croatia-Slavonia. A third,
less important branch, following the transverse cols of the Dinaric Alps, reached the
Adriatic seaboard, where their settlement was strongly encouraged by Venice, seeking to
strengthen the Christian bulwarks against Islam.
Farther east originated the Kossovo movement, so designated from the principal area
where it arose. Penetrating to the Morava valley again by transverse routes, the stream
worked its way north, occupying successive portions of the valley. From western Mace-
donia and the Vardar a third great current also moved north, mingling with the Kossovo
GEOGRAPHICAL RECORD
419
current in the Morava valley. The fourth current, consisting of this combined flow,
reinforced by the earlier inhabitants, proceeded across the Save and the Danube to
settle in eastern Slavonia and the Banat. The total displacement effected is very great.
In Serbia the old population forms only an insignificant part of the population of today.
But at the same time overlapping and a certain degree of mingling must everywhere
have contributed to ethnic confusion.
AFRICA
Commandant Tilho's Explorations in the Tibesti Region, 1912-17. The greatest
blank on the current map of Africa is comprised by the Libyan Desert and its south-
western border. According to Banse (Der gegenwärtige Stand der Erforschung der
Libyschen Wüste und Tibestis, with map in 1:7,500,000, Petermanns Mitt., Vol. 60, I,
1914) the extent of unknown desert, where never a European has laid foot, occupies an
area not far short of 500,000 square miles. Besides the desert is the more elevated
region extending northwest from Darfur and including Tibesti and the Borku plain to
the west. For knowledge of Tibesti our sole authority up to today has been Nachtigal,
who also visited Borku. During the first decade of this century French penetration into
the territory east of Lake Chad was followed by military excursions into Borku and
Ennedi (see F. R. Cana: Problems in Exploration: Africa, with map in 1:20,000,000,
Geogr. Journ., Vol. 38, 1911). Since the effective occupation of Wadai and Kanem that
penetration has been continued, and now, in the report of Commandant Jean Tilho, we
are presented with the geographical first fruits of its later development (Exploration du
Commandant Tilho en Afrique centrale: Borkou, Ennedi, Tibesti, Dar-Four, 1912-1917,
La Géographie, Vol. 31, 1916-17, No. 6-8, pp. 407-417).
In 1912 Tilho was appointed to Kanem, the country north and east of Lake Chad,
with which lake his name is associated for his geographical labors during the mission of
1906-09 (see Geogr. Rev., Vol. 1, 1916, p. 237). A vigorous campaign against the
Senussi enabled him at the beginning of 1914 to start on the work of organizing and
pacifying the territory described as Borku-Ennedi. Ennedi is the broken mountainous
area 10,000 to 13,500 square miles in extent uniting Tibesti with Darfur. The greatest
elevations do not exceed 5,000 feet. It is the home of small groups of restless, unsub-
missive tribes. Borku is a confused mass of sandstone rocks more or less buried in
sand. About half its population of 6,000 to 8,000 is entirely sedentary and lives on
dates, a little millet and corn, and the milk of the small flocks of goats. Erdi near the
frontier is a part of the Libyan Desert: it has no oases and date gardens but is the
natural refuge of irreducible tribes-nomads of astonishing endurance and audacity.
Not only do they prey on adjacent caravan routes, but they cross 400 miles of desert to
raid the camels of Dongola and Kordofan.
While making his rounds of these regions Tilho was able to settle one of the last
great hydrographic mysteries of Africa-the supposed relation between Lake Chad and
the Nile. His earlier investigations of Lake Chad and the Bahr-el-Ghazal channel which
connects with it (not to be confused with the Bahr-el-Ghazal of the Nile system) had
favored the hypothesis of a connection between the lake and the great river (Documents
Scientifiques de la Mission Tilho, Vol. 2, pp. 601-606, Paris, 1911). Now, however, this
supposition has been disproved. Chad is a closed basin, and the present lake with the
lakes of Wanyanga to the northeast are fragments of a former Central African system of
lake and marsh that must at one time have covered an area as large as France.
Later reconnaissance in Tibesti yielded a rich harvest of geographical results. To the
complete reversal of our previous ideas, instead of a simple mountain ridge running
northwest-southeast Tibesti is resolved into four or perhaps even five chains radiating
from south to west-northwest, north, and northeast-a sort of fan pivoted on the massif
of Kussi. The volcano of Emi-Kussi presents several features of interest. The volcano
attains an altitude of 11,100 feet. Its crater is of extraordinary dimensions, the axes
being 5 and 7.5 miles respectively and the depth 130 feet. Within are two ancient
vents; the orifice of the larger measures 1.5 to 1.8 miles. Covering over 120 acres of
the crater floor is a deposit of natron, apparently of considerable depth in the center.
This deposit with the salt of the Wanyanga lakes and indications of copper in the moun-
tain zone suggest that the desert is not devoid of economic possibilities.
On the economic future of the territory lately under his control Commandant Tilho
has an emphatic word to say; a more intensive railroad policy in collaboration with
England must be prosecuted in the Sudanese zone. The first line to be built should be
a great Franco-English road from Dakar, the port at Cape Verde, to Port Sudan on the
Red Sea; today only the terminal portions of such a line exist. The line should follow
the axis of the zone free from the tsetse fly, where cattle raising is practicable and to
which products of the desert can be brought by camel, where the cultivation of millet,
420
THE GEOGRAPHICAL REVIEW
cotton, and groundnuts flourishes and where there is in general a considerable native
population. A railway would further permit relief to regions periodically famine-stricken.
It is variously estimated that one-half to three-quarters of the inhabitants of northern
Wadai died in the famine of 1914.
Among the scientific achievements of Commandant Tilho's five years' labors first
place must be given to the cartographic results. His surveys embrace a band of territory
previously unknown extending over 1,100 miles from northwest to southeast from the
center of Tibesti to the neighborhood of El Fasher in Darfur. A large piece of the last
great blank on the map of Africa is thus filled in. And this mapping, with its establish-
ment of the vital points in the desert-the wells and pasturages-will be of supreme value
both for military and civil ends.
A sketch map in 1:12,500,000 showing Commandant Tilho's journeys accompanies
the account of his work published in the April, 1918, number of the Geographical Journal.
Rubber Cultivation in East Africa. Between the two great southern continents
interrelations have been and are comparatively feeble. The most obvious connection
between them is the debt of South American agriculture and more especially that of
Brazil to African labor. But on the other side something must be placed to the credit
of Brazil. Many of the cultivated plants that play so important a part in African
economy that their foreign origin is forgotten came from the west. Manioc and maize
now feed a great part of Africa: tobacco, the pineapple, the tomato, and the sweet
potato are also widespread. These and others were brought from Brazil by the
Portuguese in the sixteenth century. But with a list of the early Portuguese introduc-
tions the tale is not told. Recently introduced Brazilian products are today figuring
among the economic resources of Central Africa. Ceará rubber is a striking example.
This species of rubber-producing plant (Manihot Glaziovii, of the same genus as the
manioc, Manihot utilissima) does not belong to the rain-forest, as does the better-
known Hevea brasiliensis, but flourishes in such a semi-arid, bush country as the north-
eastern corner of Brazil. Much of East Africa is occupied by country of a similar type
in which the Ceará rubber is readily acclimated. All along the railroad from Dar-es-
Salaam to Ujiji plantations of this tree are said to form a very noticeable feature of
the landscape, while in the earlier developed eastern section, from the coast to Morogoro,
the plantations are almost continuous (Board of Trade Journ., Dec., 1917, p. 572).
The Ceará species is short-lived and furnishes a rubber inferior to that of the Hevea
(Pará rubber), but compensation lies in its hardihood and quick growth. It promises
to be an ideal cultivation for lands formerly regarded as waste.
The Meteorological Station on the Island of Mauritius. The meteorological
station on the island of Mauritius, in the Mascarene group east of Madagascar, is
making some highly valuable contributions to our knowledge of meteorological conditions
south of the equator. The Royal Alfred Observatory (20° 6' S. and 57° 33′ E.) main-
tains a central station with many sub-stations. The island, 36 miles long by 23 wide,
is divided into 13 districts coincident with the 13 principal river basins. Each district
contains several observation places. The interior of the island is a rough mass of
basaltic rock reaching heights of 3,000 feet. The relief creates local diversities of
climate within the general oceanic habitus.
A secondary station which reports to the main station by cable is maintained 375
miles eastward on the island of Rodriguez. Vessels passing on their way to Indian ports
also report their observations, thus making possible the correlation of data from a wide
area.
The monthly bulletin of the observatory includes among other things a report on
"the state of the sea at stations round the coast, to estimate the direction of the sea
wave in connection with the study of cyclonic storms and extra-tropical gales. These
waves are conspicuous features during the passage of a storm. At times they are very
destructive, running far inland over the flat valley floors. They would be more destruc-
tive were it not for the coral reefs that surround the shore. The location of the storm
center is determined by noting the direction and length of these waves.
Mauritius, like the West Indies and the Philippines, lies in the path of tropical
cyclones, near the point where the disturbances turn southward away from the equator.
Along the east coast of Africa these storms generally occur in the months of January,
February, and March. The inhabitants of Mauritius have learned to expect them in
February, near the close of what they call winter, though it is so called simply because
it is the stormy season, for it is the warmest part of the year.
The character and movement of cyclonic disturbances are recorded in each monthly
bulletin of the observatory. The tracks of some storms are mapped. The tropical
cyclones often pass north of Mauritius and double back around Madagascar, going off
toward the southeast. In December, 1916, two storms followed this course, sweeping
GEOGRAPHICAL RECORD
421
through the Mozambique Channel. One noted in January, 1917, crossed the northern
end of Madagascar on the 26th, turned south through the channel on the 27th, and passed
over southern Madagascar on the 28th.
Strange to say, extra-tropical disturbances are felt in spite of nearness to the
equator. In November, 1916, a series of shallow highs and lows was observed, with an
interval of about five days between successive maxima.
Incomplete records have been kept at Mauritius since 1875, while rather detailed
descriptions of the hurricanes are contained in books of travel and in histories of the
island. The "History of Mauritius" by Charles Grant (London, 1801) contains many
such accounts, some of which date back to 1740.
Since January 1, 1916, the metric system has been used in all records made by this
observatory. (For data on the station see the monthly Results of Magnetical, Meteoro-
logical, and Seismological Observations and the Annual Reports of the Director of the
Royal Alfred Observatory.)
EDUCATIONAL GEOGRAPHY
as
Economic Geography in the High School. To the discussion of geography's
place in present-day education Prof. J. Paul Goode contributes an article on economic
geography in the February 23 issue of School and Society (A Course in Economic Geog-
raphy, pp. 216-222). After briefly tracing the influence of such men as Guyot, Chisholm,
and Cyrus C. Adams in the development of the science in this country the author pre-
sents what he considers to be the essential courses in high school geography. He advo-
cates three units: "Elements of Geography," as a ground work for all later study;
"Economic Geography," for the senior high school; and "Commercial Countries,''
a regional treatment with which to complete the work. The first unit the author would
require of every student. Of largest present interest Professor Goode finds the course
in Economic Geography. In this study he would consider the "Commodities of Com-
merce' and the "Geography of Trade,
Geography of Trade,'' attempting to show the geographic influences
operating behind the scenery of international relations. Atlases, maps, pictures, lantern
slides, and moving-picture films should be used as aids, while current periodicals, the
U. S. Commerce Reports, special studies made by consuls and trade experts, together
with the daily papers may be drawn upon to supplement the literature of formal text-
books for student as well as teacher.
In this scheme economic geography is emphasized, as being pedagogically the most
telling phase of geography. A knowledge of human activities as influenced by physical
environment is imparted in the Elements of Geography course. As to regional geog-
raphy, what better way of becoming acquainted with the countries and people of the
world, than considering them in connection with the things they contribute to the
commerce of mankind? If care is exercised to see that geographic principles are stressed
this should result in no lowering of standards.
To fully carry out this program would require at least three half-year courses. This
is more than is accorded to geography In most high schools, but not more than the subject
merits. The vastly wider field, politically and commercially, into which the United States
is entering makes this an opportune time to secure for geography the recognition which
it should have in American education. Giving the subject an economic trend should
greatly strengthen its appeal to a public accustomed to think largely in terms of com-
mercial expansion.
GEOGRAPHICAL NEWS
GRANTS have recently been made by the American Association for the Advancement
of Science to Professor A. E. Douglass, of the University of Arizona at Tucson, to
investigate the length of record of tree growth of the sequoias from about 2,200 to
3,000 years ago; to Professor Herman L. Fairchild, of the University of Rochester, for
the continuation and completion of his studies on the post-glacial continental uplift in
New England and the Maritime Provinces of Canada; and to Prof. S. D. Townley, of
Leland Stanford Junior University, for the investigation of earthquake phenomena.
THE Museum Association and the Free Public Library of Newark, N. J., are holding
an exhibition illustrating the resources and the social conditions of Colombia. The
exhibition comprises representative samples of material now exported or imported from
Colombia; notes on business methods suiting the conditions of South American trade;
maps, political, commercial, geographical, and agricultural; pictures of buildings, city
streets, factories, railway stations, docks, school houses, plantations, etc.; books and
the newspapers and periodicals of the country; and samples of its products of all kinds,
422
THE GEOGRAPHICAL REVIEW
including examples of the arts of the common people, home weaving, basket making,
pottery, metal work, wood carving, etc., both ancient and modern. The exhibition will
continue to October 31, except for a probable interruption in July and August.
PERSONAL
DR. L. A. BAUER of the Carnegie Institution on May 11 addressed the Philosophical
Society of Washington on Corresponding Changes in the Earth's Magnetic State and
in Solar Activity, 1888-1916.”
PROFESSOR W. M. DAVIS read a paper on "Recent Coast Survey Charts of the
Philippines and Their Bearing on the Coral Reef Problem" at the April meeting of
the American Academy of Arts and Sciences in Boston and at the annual meeting of
the National Academy of Science in Washington.
DR. H. J. FLEURE, who has filled the professorship of zoölogy at the University Col-
lege of Wales at Aberystwyth and has also acted as lecturer in geography, will henceforth
occupy the chair of geography and anthropology recently created at the institution
through an endowment of £10,500 by an anonymous donor. Professor Fleure contributed
an article on "The Racial History of the British People' to the March Review.
MR. PHILIP A. MEANS, whose note in the December, 1917, Review on the Guarani
invasions of the Inca Empire will be recalled, is the author of a "History of the Spanish
Conquest of Yucatan and of the Itzas'' recently published by the Peabody Museum of
American Archaeology and Ethnology of Harvard University.
PROFESSOR MARSHALL H. SAVILLE of the Museum of the American Indian, Heye
Foundation, of New York, is the author of "A Letter of Pedro de Alvarado Relating to
his Expedition to Ecuador" in 1534, presented in facsimile and accompanied by a
translation, notes, and bibliography, which has recently been published by the Museum.
GEOGRAPHICAL PUBLICATIONS
(Reviews and Titles of Books, Papers, and Maps)
For key to classification see "Explanatory Note" in Vol. II, pp. 77-81
NORTH AMERICA
GENERAL
DALY, R. A. Geology of the North American Cordillera at the forty-ninth parallel.
Part I: xxvii and 546 pp.; maps, diagrs., ills.; Part II: xxvii and pp. 547-857;
diagrs., ills., index; Part III: maps and ills. in pocket. Introduction by R. W.
Brock.
Geol. Survey of Canada Memoir No. 38. Ottawa, 1912.
This memoir is a record of geologic work subsidiary to that of the International
Boundary Survey. The field work occupied six seasons. A strip varying in width
from 4 to 12 miles was mapped on a scale of 1 inch to the mile. The topography of
most of this area is shown by contour lines at intervals of 100 feet. West of longitude
116° 30′ the strip mapped lies wholly in Canada. East of that it laps over from 2 to 4
miles into the United States. An east-west structural section accompanies each map
throughout the entire distance of nearly 400 miles. The report is primarily geological,
but the following chapters are of geographic interest: Chapter 3, Nomenclature of
Ranges, pp. 17-46; Chapter 21, Glaciation, pp. 577-598; Chapter 22, Physiographic
Notes, pp. 599-642.
"
The nomenclature of these ranges has been in a state of confusion. Even the terms
"Rocky Mountains" and "Selkirks" are given very diverse applications by different
writers. Daly's attempt to reduce the naming to a consistent system [first published in
the Geogr. Journ., Vol. 27, 1906] is therefore highly welcome and it is to be expected
that it will exert a great influence on future writings. It appears that on the whole a
serious attempt has been made to abide by prevalent usage in so far as any usage can
be said to be prevalent, provided only that the limits of the several ranges shall be
determined by some good reason. It is aptly pointed out that the mountains along the
International Boundary do not constitute distinct ranges" in any such sense as do
the mountains of Colorado and Wyoming. Almost the whole breadth of the Cordilleran
belt in our northern states is filled with mountains, not very unlike in appearance
throughout most of the distance. Under such circumstances Daly would make primary
divisions by selecting as boundaries the largest and most continuous valleys. Of such
valleys, four are deemed to be of the first order, two of them being styled trenches. A
trench, in this sense, is defined (p. 26) as "a long narrow intermontane depression
occupied by two or more streams (whether expanded into lakes or not) alternately
draining the depression in opposite directions." The four first-rank valleys or trenches
are here named in order, with the longitude at which each crosses the 49th parallel and
the river which occupies it at that place: Rocky Mountain Trench (115°, Kootenay
River flowing south); Purcell Trench (116° 30', Kootenay River flowing north); Selkirk
Valley (118°, Columbia River); Okanagan Valley (119° 30', Okanagan River). The
entire Cordilleran belt in latitude 49° east of Puget Sound is thus divided into five
mountain systems, the Rocky Mountain, Purcell, Selkirk, Columbia, and Cascade systems.
The "Rocky Mountains'' are thus reduced to a belt about 50 miles wide lying east of
Flathead Lake and the Kootenay River. Many will, no doubt, take exception to Daly's
use of this term. In judging of any proposed nomenclature the chances of securing
general adoption should always be taken into account. In this case the chances that any
man or organized body can restore the name Rocky Mountains to its former narrow
application are very small indeed. His emphasis on the term Cordillera for all the
highlands between the Pacific Ocean and the Great Plains is well placed. A more gen-
eral use of this term would save the words "Rocky Mountains" from some of the worst
misapplications. The insistence on the Spanish pronunciation seems to us less wise.
Webster's. International Dictionary sensibly adopts the English. Insistence on a foreign
pronunciation can only retard the general adoption of this much-needed term.
To the physiographer the most interesting part of this work is that which assails the
position which peneplanes have held in the interpretation of mountain topography. A
subequality of altitude among crests within a single view prevails throughout the entire
423
424
THE GEOGRAPHICAL REVIEW
area, with the single exception of the Front Range in Glacier National Park. As
approximate base leveling is affirmed by Willis for the first range west of this and since
the Cascade Mountains have been pronounced an upraised and dissected peneplane by
Willis, Smith, Russell, and others, it may safely be asserted that this would be the
crthodox interpretation of the entire Cordillera in this latitude (perfection of pene-
planing is of course not assumed). Daly combats this hypothesis and assumes on the
contrary that high mountains tend to assume equal heights by the more rapid degrada-
tion of the higher crests. Thus accordance of heights is made incident to mountain
degradation in its first cycle, and the two-cycle history is discarded. He finds reason
to think that there has been a rather general uplift of several thousand feet in late Ter-
tiary time, but this is treated as incidental and bearing no necessary relation to the
accordance of crest heights.
The main objections to the two-cycle theory are (1) the absence of peneplane rem-
nants, thus resting the whole argument for two cycles on accordance of crests; (2) the
lack of stream adjustment; (3) the shortness of the time allowed for the making and
destruction of the last peneplane when compared with the time which peneplanes else-
where have endured. Under (1) the only high level surfaces deserving consideration
are those of the Okanagan (Cascade) Range, and these Daly suspects to be determined
by batholiths recently unroofed by erosion, but capable of other explanations also.
Under (2) he finds the character of the drainage essentially the same from the Great
Plains to Puget Sound, i. e. mainly consequent, with little or no subsequent drainage
or adjustment to strong and weak rocks, even where contrasts of hardness are marked,
unless some of the streams following fault lines may in part have selected these lines
because of their weakness. He cites the well-known principle that adjusted drainage is
one of the marks of second-cycle erosion. Under (3) he points to the fact that the
Cascade peneplane is assumed to have been developed from a mountainous country since
the beginning of the Miocene and by some writers since its close, then uplifted in the
late Pliocene or possibly even early Pleistocene, since which time it has been (in the
mountains) wholly destroyed. This rate of making and destroying peneplanes he
thinks wholly out of proportion to that which is shown in the east by the well-preserved
Cretaceous and Tertiary peneplanes, even allowing for the fact that alpine conditions
must be assumed in the Cascades.
The constructive side of Daly's work appears in his provision for crests of uniform
height through other means. He considers (1) isostasy, both during uplift and after its
cessation; (2) the assumed regular forms of upper surfaces of batholiths and meta-
morphosed mountain cores; (3) cirque erosion; (4) influence of tree line, which is often
not far from the cirque level (the tendency of mountain crests to be leveled to tree line
and the cirque level was strongly stated by George Dawson in 1896); and (5) the sub-
equal spacing of streams during maturity (Richter, Tarr, and others). Many processes
are contrasted in their effectiveness above and below the tree line. He finds that all the
mountains on the boundary show accordant summits at elevations very close to their
"effective tree line' and concludes that many so-called peneplane remnants may owe
their existence to the influence of the tree line rather than to that of base level.
2
Daly's general conclusions concerning accordant alpine crests and their explanation
were published in the Journal of Geology in 1905. The present work adds discussions
of concrete cases. Although these hypotheses have been before the geologic world for
some years, no considerable tendency is observed among American physiographers to
follow them. On the other hand, it is but fair to say that no other country is so wholly
"addicted to the use' of peneplanes as is the United States. Apparently also the lack
of stream adjustment in these western mountains deserves more consideration than it
has received. So also does the great discrepancy between the assumed durations of
erosion cycles in the Cordilleran and Appalachian belts. The tendency of crests to halt
in their degradation at tree line and the cirque level is undisputed, but as yet no example
of even crests due to this process is acknowledged by all physiographers. On the other
hand there are unchallenged peneplanes even at mountain tops, as, for example, in the
Front Range of Colorado. Indeed, the Rocky Mountain peneplane in Colorado assumes
a role of some importance in considering the mountains farther north, for between South
Park and southern Canada accordant crests are found so abundantly and at such short
intervals that it would be difficult to say just where to begin to challenge the peneplane
theory.
Daly has well remarked the need of quantitative work to determine the adequacy of
forces which are well known to work in the direction of accordant crests. In the mean-
time, as between the two processes at issue, what can analytical, deductive study (such
as Davis has so forcibly urged and so ably used) do to determine more exactly the
sequence of forms involved in each method, with a view to determining criteria for dis-
criminating between their results?
N. M. FENNEMAN
GEOGRAPHICAL PUBLICATIONS
425
CANADA
Alberta, Saskatchewan, Manitoba
DENIS, L. G., AND J. B. CHALLIES. Water-powers of Manitoba, Saskatchewan,
and Alberta. ix and 334 pp.; maps, diagrs., ills., bibliogr., index. Commission of Con-
servation, Toronto, 1916.
DOWLING, D. B. The southern plains of Alberta. 200 pp.; maps, diagr., ills.,
index. Geol. Survey of Canada Memoir 93: Geol. Ser. No. 78. Ottawa, 1917.
HINES, J. The red Indians of the plains: Thirty years' missionary experience
in the Saskatchewan. viii and 322 pp.; maps, ills. Society for Promoting Christian
Knowledge, London, 1915. 6s. 9 x 6.
Mawson, T. H., AND SONS. Calgary: A preliminary scheme for controlling the
economic growth of the city. xiii and 88 pp.; maps, diagrs., ills. Published under
the auspices of the City Planning Commission of Calgary. Thomas Mawson & Sons,
London, Lancaster, Vancouver, and New York, [1914]. $2.00. 12½ x 10.
MITCHELL, A. E. The mountain maze of the upper Athabaska. Map, ills.
Travel, Vol. 29, 1917, No. 4, pp. 30-33 and 45-46.
OLIVER, EDWARD. Ninth Annual Report of the Secretary of Statistics for sixteen
months ended April 30, 1916. 58 pp.; map, index. Dept. of Agric., Province of
Saskatchewan, Regina, 1916. [For reference to section on wheat crop see note in July,
1917, Review, Vol. 4, p. 58.]
ROSE, BRUCE. Wood Mountain-Willowbunch coal area, Saskatchewan. ii and
103 pp.; maps, ills., index. Geol. Survey of Canada Memoir 89: Geol. Ser. No. 75.
Ottawa, 1916. [Abstracted in the March, 1917, Review, Vol. 3, pp. 241-242.]
TYRRELL, J. B. Notes on the geology of Nelson and Hayes Rivers. Maps, ills.
Trans. Royal Soc. of Canada, Ser. 3, Vol. 10, 1916, pp. 1-27.
Alberta. Studies in duty of water and climatic conditions. 1 in. to 42 mi. (1:2,-
600,000) approx. Irrigation Office, Dept. of Interior, Ottawa, 1913.
[Four maps:] (1) Alberta, Northern, (2) Southern Alberta, (3) Manitoba, (4) Sas-
katchewan [showing] Dominion electoral divisions. 1:792,000. Dept. of the Interior,
Ottawa, 1914 and 1915.
[Three maps:] (1) Alberta, Northern, (2) Southern Alberta, (3) Manitoba, showing
disposition of lands. 15th edit., corrected to January 1, 1917. 1:792,000. In 2
sheets. Dept. of the Interior, Ottawa, 1917.
•
[Three maps:] (1) Alberta, Southern, (2) Manitoba, (3) Saskatchewan, Cereal
maps of, showing acreage under crop in each township in wheat, oats, barley, and flax
during 1915. 1:792,000, or 1 in. to 12 mi. 4th edit. Railway Lands Branch, Dept.
of the Interior, Ottawa, 1916.
Manitoba, Saskatchewan, and Alberta, showing the land registration and judicial
districts. Preliminary edit. 1:2,217,600. Compiled from information supplied by Pro-
vincial Governments. Railway Lands Branch, Dept. of the Interior, Ottawa, 1917.
UNITED STATES
General
SCHANZ, MORITZ. Baumwoll-Anbau, -Handel und -Industrie in den Vereinigten
Staaten von Nordamerika. Bibliogr. Beihefte zum Tropenpflanzer, Vol. 15,
1915, No. 6, pp. 513-645.
The author is a member of the German colonial committee for the investigation of
cotton, and his observations are the result of first-hand observations in the cotton states.
The monograph is admirable in the presentation and arrangement of essential facts;
there are few publications in English that give in brief compass so complete a view of the
cotton industry in all its phases. Since most of the facts are readily accessible and more
or less familiar to American readers, one's interest centers mainly in (1) what the author
seems to regard as exceptional or distinctive and (2) his conclusions as to future condi-
tions.
Our somewhat familiar demonstration agent (Wanderlehrer) of the U. S. Depart-
ment of Agriculture is rightly accorded a high place in rural progress, although he
is not unknown in Germany. On the other hand, two rural conditions are rightly criti-
cized. The farm credit system is rightly described as almost unorganized, with a con-
sequent high interest rate, in contrast, of course, to the efficient credit systems in many
428
THE GEOGRAPHICAL REVIEW
Farrand, BeatRIX. The National Park on Mount Desert Island. Ills. Scrib-
ner's Mag., Vol. 61, 1917, No. 4, pp. 484-494. [A popular description of the Sieur de
Monts "Park" (though it should be said that it is officially named the Sieur de Monts
National Monument"), accompanied by good photographs. See note on this subject
in the Review, Vol. 2, 1916, p. 375.]
FORD, W. C., edit. Commerce of Rhode Island, 1726-1800. Vol. II: 1775-1800.
xi and 501 pp.; ill., index. (Massachusetts Historical Soc. Collections, 7th Ser., Vol.
10.) Boston, 1915. $3.00. 9½ x 6½.
GAGNON, ALPHONSE. Le Vinland. Bull. Soc. de Géogr. de Québec, Vol. 11, 1917, No.
2, pp. 114-119. [The conclusions of the author, identifying Vinland with the shores of
Massachusetts or Rhode Island, are in close agreement with those reached by W. H.
Babcock and stated by him at the last congress of Americanists held in Washington,
December 1916. Cf. also his "
Cf. also his "Markland, Otherwise Newfoundland" in the October,
1917, Review.]
Sieur de Monts National Monument; The White Mountain National
Forest. 30 and 34 pp.; ills. Sicur de Monts Publs. 17.
Sieur de Monts Publs. 17. Wild Gardens of Acadia
[Bar Harbor, Me.], and U. S. Dept. of the Interior, Washington, D. C., 1917.
SOUTH AMERICA
PARAGUAY, Uruguay, ArgeNTINA, CHILE
BELTRÁN, J. G. Geografía de la Argentina, física, política y económica. 332 pp.;
maps, diagrs., ills. Cabaut y Cía. and A. García Santos, Buenos Aires, 1917. 7½ x 5.
This is a textbook intended for the use of secondary schools, normal schools, and
advanced students. As such it is meager and reveals the backward state of education
in Latin America. The typography, text-maps, and illustrations are poor. Yet the work
marks a distinct advance in geographical method for those countries. Instead of being
limited to the bare enumeration of rivers, towns, ports, mountains, lakes, cities, etc., it
contains a discussion of such mattters as immigration, colonization, commercial and
industrial development, transportation, and city growth. An occasional attempt is made
to show the relation of man to his environment. Graphical representations are employed
to illustrate density of population, amount of foreign trade, meteorological conditions,
distribution of vegetation, and composition of the soil in different parts of the republic.
The author's treatment of boundaries is almost entirely historical rather than geographi-
cal. The statement (p. 148, note) that atmospheric pressure increases with the amount
of water vapor present is, of course, incorrect, as is also his assertion that pressure over
the ocean is higher in summer than in winter. The reverse is usually true, the gradient,
however, being the decisive factor in the formation of winds.
In spite of these defects the book is of special value to the foreigner as it contains
a wealth of up-to-date data regarding Argentina.
ARAÚJO, ORESTES. Historia de los Charrúas y demás tribus indígenas del Uru-
guay. Part I: Etnología salvaje. 142 pp. J. M. Serrano, Montevideo, 1911. 7 x 5.
[At the time of the Spanish discovery the Charruas resided on the northern shores of the
Rio de la Plata.]
Argentine Republic: General descriptive data. 31 pp.; map, ills. Pan
American Union, Washington, D. C., 1917.
Chubut-one of Argentina's territories. Ills. Bull. Pan American Union,
Vol. 44, 1917, No. 5, pp. 607-617. [Spanish version in Bol. Unión Panamericana, Vol.
45, 1917, No. 1, pp. 1-10.]
GROEBER, PABLO. Informe sobre las causas que han producido las crecientes del
Rio Colorado [Territorios del Neuquén y La Pampa] en 1914. 29 pp.; maps,
diagrs., ills. Bol. Direcc. Gen. de Minas, Geol., e Hidrol. No. 11, Ser. B (Geol.). Minist.
de Agric., Buenos Aires, 1916.
HUDSON, W. H. The Purple Land: Being the narrative of one Richard Lamb's
adventures in the Banda Orientál, in South America, as told by himself. With
an introduction by Theodore Roosevelt. x and 355 pp. E. P. Dutton & Co., New York,
1916. $1.50. 72 x 5. [First published in 1885 as "The Purple Land That England
Lost."
A story of life in Uruguay.]
KEIDEL, J. La geología de las sierras de la provincia de Buenos Aires y sus
relaciones con las montañas de Sud Africa y los Andes. 78 pp.; maps, diagrs.,
ills., bibliogr. Anal. del Minist. de Agric. Sección Geol., Mineral., y Minería, Vol. 11,
GEOGRAPHICAL PUBLICATIONS
429
1916, No. 3. Direcc. Gen. de Minas, Geol., e Hidrol., Buenos Aires. [The sierras of
Buenos Aires Province are composed of Paleozoic rocks resting discordantly upon ancient
crystallines. Their internal structure is attributed to movements taking place in late
Paleozoic time. Such movements were separated by a long interval from the regional
movements (Tertiary) to which the sierras owe their present elevation. Stratigraphi-
cally and tectonically they have close affinity with the Pre-Cordillera of San Juan and
Mendoza, on the one hand, and with the Cape plateau of South Africa on the other.
Their relation to the Andes may be compared with that obtaining between the ancient
blocks of Europe and the folded Alps. The report is well illustrated with maps and
diagrams and provided with a bibliography.]
MARTINOLI, GAETANO. Cattle breeding in the Argentine Republic at the present
day. Monthly Bull. of Agric. Intelligence and Plant Diseases, Vol. 8, 1917, No. 8, pp.
1073-1084. Internatl. Inst. of Agric., Rome.
MORANDI, LUIS. Frecuencia y modalidad de las calmas horarias en Montevideo,
período 1906-1914. 22 pp.; diagr. Inst. Nacl. de Agronomía, Montevideo, 1915. [A
study of air currents with special reference to their constancy as a source of power.
Uruguay, being largely a land of plains and gentle slopes, is poorly supplied with water-
power, but, the author thinks, should be able to make an extensive use of windmills.
Land and sea breezes are felt far inland, and there are few hours of absolute calm,
especially at an altitude of 25 meters above the ground. The data collected may be of
value also in aviation.]
MORANDI, LUIS. Resumen y discusión de las observaciones meteorológicas del
Observatorio Central, período 1901-1915. 19 pp.; ills. Inst. Nacl. Físico-Climatológico,
Montevideo, 1917. [Abstracted in the December, 1917, Review, Vol. 4, p. 488.]
Ross, GORDON. Argentina and Uruguay. xv and 308 pp.; map, diagrs., ills., index.
The Macmillan Co., New York, 1916. 9 x 6. $3.50.
North Shore of the Plata River: Sauce Point to Martin Garcia Island. From
British surveys in 1844 and 1856, with additions of Argentine surveys to 1911.
[1:84,000.] Inset: Sauce Harbor, 1:16,000. U. S. Hydrogr. Office Chart No. 2703.
Washington, D. C., 1917.
Uruguay River: Juncal I. to Concepcion del Uruguay. From Argentine surveys
between 1901 and 1910. [1:100,000.] Inset: Gualeguaychu River. U. S. Hydrogr.
Office Chart No. 2705. Washington, D. C., 1917.
EUROPE
BALKAN STATES, INCLUDING RUMANIA
Preface by Paul
7½ x 4½.
LEFEUVRE-MÉAULLE, H. La Grèce économique et financière.
Deschanel. x and 258 pp. Félix Alean, Paris, 1916. 3 fr. 50.
Dealing principally with the natural resources of Greece, this book explains the
country's economic position in the Near East. Its author has peculiar competence to
write on these subjects in view of a long term of residence in the Levant as French
commercial agent. This enables him to sketch in effective touches the possibilities of
developing Greece's agriculture and industry. An excellent chapter on the territorial
additions following the Balkan War shows how important was Greece 's gain. To Saloniki
and its value as a center of trade much space is devoted. The book will appeal to those
who are interested in the new trends of Greek economic life after the war.
ALMAGIÀ, ROBERTO. Note di cartografia albanese. Maps. Riv. Geogr. Italiana,
Vol. 23, 1916, No. 6-7, pp. 269-277.
DAINELLI, GIOTTO. Quanti sieno gli Italiani in Dalmazia? Riv. Geogr. Italiana,
Vol. 24, 1917, No. 3-4, pp. 132-147.
DEDIJER, JEVTO. La transhumance dans les pays dinariques. Ann. de Géogr.,
No. 137, Vol. 25, 1916, pp. 347-365.
DEDIJER, JEVTO. Traces glaciaires en Albanie et en Nouvelle Serbie. Map,
diagrs. La Géographie, Vol. 31, 1916-17, No. 5, pp. 325-336. Paris.
MANETTI, CARLO. Appunti sulla pastorizia albanese. Ills. L'Agricoltura Coloni-
ale, Vol. 9, 1915, No. 4-5, pp. 212-216. Florence.
NEWBIGIN, MARION. Balkan outlets in the present and in the future. Geogr.
Teacher, No. 46, Vol. 8, 1916, Part 6, pp. 333-340.
430
THE GEOGRAPHICAL REVIEW
NICULESCU, C. Contributions à la géologie de l'Epire (environs de Janina).
Diagrs. Bull. Section Sci. de l'Acad. Roumaine, Vol. 3, 1914-15, No. 1, pp. 23-34.
Bukharest.
PEDRAZZI, ORAZIO. La Dalmazia e gli slavi del sud. Boll. Reale Soc. Geogr.
Italiana, Vol. 6, 1917, No. 4-5, pp. 270-282.~ Rome.
PERNICE, ANGELO. Origine ed evoluzione storica delle nazioni balcaniche. xii
and 628 pp.; maps, bibliogr. Ulrico Hocpli, Milan, 1915. L.6.50. 8x 5. [A guide
to the history of the Balkan Peninsula, with a description of the geography of the area
as a factor in its history.]
RICCHIERI, GIUSEPPE. Il fato geografico nella storia della penisola balcanica.
Maps. Boll. Reale Soc. Geogr. Italiana, Vol. 6, 1917, No. 6, pp. 401-435. Rome.
STEBBING, E. P. The Serbian front in Macedonia. Map, ills. Scottish Geogr.
Mag., Vol. 33, 1917, No. 4, pp. 145-157.
STEFANO, SANTUCCI. Un viaggio nell'Albania settentrionale. Ills. Boll. Reale
Soc. Geogr. Italiana, Vol. 5, 1916, No. 8, pp. 646-672; No. 9, pp. 749-777; No. 10, pp.
805-823. Rome.
WOODS, H. C. Rumania and the Danube. Map. Asiatic Rev., No. 28, Vol. 10,
1916, pp. 373-381.
Balcani, 1916, L'intervento Europeo nei. 1:1,500,000. Inset: I Rumeni della
Transilvania e Ungheria. Istituto Italiano d'Arti Grafiche, Bergamo, 1916. [The
boundaries of the Balkan states are those of the treaties of London and Bukharest, 1913.]
Balkan States, Map of the, showing communications to illustrate the paper by
H. C. Woods. 1:1,750,000, or 1 in. to 27.62 mi. Accompanies "Communications in the
Balkans' by H. C. Woods, Geogr. Journ., Vol. 47, 1916, No. 4, pp. 265-293.
Balkan area (Philips' large-scale strategical war map of Europe). 1:1,140,000,
or 1 in. to 18 miles. George Philip & Son, Ltd., London, [1917].
ASIA
MANCHURIA, KOREA, JAPAN
MORSE, E. S. Japan day by day-1877, 1878-79, 1882-83. Vol. 1: xv and 441 pp.;
diagrs., ills. Vol. 2: 453 pp.; diagrs., ills., index. Houghton Mifflin Co., Boston
and New York, 1917. $8.00. 9 x 7.
As many-minded as Ulysses, humorous as Aristophanes, learned as Darwin, and
persuasive as Benjamin Franklin-such is the venerable professor who in his eightieth
year delights an expectant world by publishing some notes made forty years ago in the
Japan of semi-modern, semi-medieval aspect. These volumes are the work of a scien-
tifically trained scholar who united the learning of an Agassiz or a Schliemann with the
humorous charm of a Gilbert or a Mark Twain. We see him at one moment penetrating
the mystery of prehistoric man in Dai Nippon, the next convulsing with laughter the
gravest assembly in a country of cultivated gravity.
We frankly suspend criticism in treating of these precious volumes. We fall in love
with the irrepressible boyishness of this learned graybeard. We love his book as we love
Dr. Johnson-because of his faults no less than his virtues. Morse champions the
worn-out Chinese and Turkish views on inoculation for smallpox along with a superstitious
reverence for the Darwinian descent from monkeys. But what of that? Dr. Johnson
was even worse; and we read and re-read him with delight. It is this quality of frank-
ness, of enthusiasm, of explosive indignation, of profound erudition coupled with in-
finitely human sympathy and understanding that makes these volumes worthy of notice.
From a geographical point of view the work illustrates abundantly the close relation
between the habits of a people and soil and climate that influence them. Many customs
which Professor Morse was at first inclined to ridicule proved later to be the result of
forces which he had not appreciated at the first glance. Indeed, rather than applaud
Japan for her immense capacity for imitating "western" culture we are of opinion that
time and reflection will cause her soon to discard many so-called reforms which, while
suited to our conditions, are costly, cumbersome, and unsuited to a land whose physical
characteristics have no parallel in any other part of the world.
POULTNEY BIGELOW
GEOGRAPHICAL PUBLICATIONS
431
GJELLERUP, P. Dagligt liv i Tokio i vor tid. Geografisk Tidskrift, Vol. 23, 1916,
No. 6, pp. 237-243. [Daily life in Tokyo in our time.]
HANABUSA, NAOSABURO. État de la population de l'Empire du Japon au 31
Décembre 1913. [ix] and 335 and xxxviiiˇpp.
[ix] and 335 and xxxviii pp. Bur. de la Statistique Générale, Cabinet
Impérial, Tokyo, 1916.
Japon, l'Empire du, Exposé général des résultats du recensement de la
population de, à la fin de 1913. v and 179 pp. Bur. de la Statistique Générale, Cabi-
net Impérial, Tokyo, 1916. [In Japanese.]
MALAY ARCHIPELAGO, INCLUDING THE I'HILIPPINES
ABENDANON, E. C. Bijdragen tot de historische cartographie van den Ned. Ind.
Archipel. Tijdschr. Kon. Nederl. Aardrijk. Genoot., Vol. 34, 1917, No. 6, pp. 862-877.
CRAIG, AUSTIN, and CONRADO BENITEZ. Philippine progress prior to 1898. 136
pp.; maps. (A Source Book of Philippine History, Vol. 1.) Philippine Education Co.,
Inc., Manila, 1916. 9 x 62. [Part I discusses agriculture, industry, and commerce at
the time of the discovery; commerce during the restrictive period of Spanish control and
during the nineteenth century.]
MILLER, H. H. Commercial geography: The materials of commerce for the
Philippines. 121 pp.; maps, diagrs., ills., index. Bureau of Education, Manila, 1915.
RINNE, F. Reisebilder aus Java und Celebes. Maps, ills. Mitt. der Gesell. für
Erdkunde zu Leipzig für das Jahr 1913, pp. 99-118.
RUTTEN, L. Reisherinneringen uit Zuid-Oost Boelongan (Oost-Borneo). Map,
ills. Tijdschr. Kon. Nederl. Aardrijk. Genoot., Vol. 33, 1916, No. 2, pp. 236-253.
SHELFORD, R. W. C. A naturalist in Borneo. Edited with a biographical intro-
duction by E. B. Poulton. xxviii and 331 pp.; ills., index. E. P. Dutton & Co., New
York, 1917. $5.00. 9 x 6. [Chiefly biological, but containing a chapter on the natives
of Borneo and some discussion of the phytogeography and zoögeography of the island,
by the late curator of the Sarawak Museum.]
SHERFESEE, W. F. Annual report of the Director of Forestry of the Philippine
Islands for the fiscal year ended December 31, 1915. 91 pp.; maps. Dept. of the
Interior, Manila, 1916. [Contains abstract of a forest reconnaissance of the Agusan
Valley, northern Mindanao (pp. 52-60), accompanied by a map, 1:350,000. The original
detailed report will be published in Spanish as a special circular.]
SWAAB, J. L. M. De onderafdeeling Redjang der residentie Benkoelen. Map.
Tijdschr. Kon. Nederl. Aardrijk. Genoot., Vol. 33, 1916, No. 1, pp. 57-69. [Sumatra.]
WALLIS, B. C. The rainfall of Java. Maps, diagrs., bibliogr. Scottish Geogr. Mag.,
Vol. 33, 1917, No. 3, pp. 108-119.
WESTER, P. J.
The food plants of the Philippines. Ills. Philippine Agric. Rev.,
Vol. 9, 1916, No. 3, pp. 150-256. Manila. [Abstracted in the April, 1917, Review, Vol.
3, p. 326.]
Luzon, Southwestern, and Mindoro.
vey Chart No. 4714. Washington, D. C., 1916.
1:400,000. U. S. Coast and Geodetic Sur-
Philippine Is. Cotabato to Sarangani Bay, South Coast of Mindanao. Surveys
to 1916. Mindanao River region from surveys by U. S. Engineers in 1907. 1:200,000.
U. S. Coast and Geodetic Survey Chart No. 4607. Washington, D. C., 1917.
Philippine Islands, New library atlas map of. 1 in. to 36 mi. (1:2,280,960). Insets:
Manila and Vicinity [1:380,160]; Manila [1:26,400]; Batan and Babuyan Is. [1:3,168,-
000]; Concise Map of the Philippine Islands and the China Sea [1:15,200,000]. Rand
McNally & Co., Chicago, 1913.
Philippine Islands: Palawan to Culion Island, including Linapacan Strait. Sur-
veys to 1916. 1:100,000. U. S. Coast and Geodetic Survey Chart No. 4315.
ington, D. C., 1917.
Wash-
Philippine Islands, Southeastern Part.. Surveys to 1916. [1:800,000.], U. S.
Coast and Geodetic Survey Chart No. 4708. Washington, D. C., 1917.
Puerto Princesa, East Coast of Palawan, Philippine Islands.
Coast and Geodetic Survey Chart No. 4343. Washington, D. C., 1917.
1:40,000. U. S.
Western Masbate, Philippine Islands. [1:100,000.] U. S. Coast and Geodetic
Survey Chart No. 4412. Washington, D. C., 1917.
4.32
THE GEOGRAPHICAL REVIEW
PHYSICAL GEOGRAPHY
METEOROLOGY AND CLIMATOLOGY
PETTERSSON, HANS. Meteorological aspects of oceanography. Map, diagr. Monthly
Weather Rev., Vol. 44, 1916, No. 6, pp. 338-341.
The climatic effects of the ocean are widely recognized. Yet in weather forecasting
but little practical application has been made of observations of the temperatures and
movements of ocean water. The thermal functions of the ocean are the accumulation
and redistribution of solar heat more evenly in time and space. Dr. Pettersson likens
the ocean to a savings bank for solar energy, "receiving deposits in seasons of excessive
insolation and paying them back in seasons of want." This bank has a very extensive
foreign exchange. . . . the annual 'dividends' paid by this bank to its 'share-holders, ' i. e.
the continents round its borders, are emitted on a variable scale.'
'Besides trying to predict the extremely variable state of the fickle atmosphere one
should give more attention to the conservative element in meteorology, viz. the surface
sheet of the ocean, where changes may be observed months before their effect on our
weather becomes manifest." For example, "a sensible departure from the average value
of the vast amount of stored heat carried through this [Florida] strait . . . . might have
profound effects on the weather of the following months both on the European and the
North American continents" (cf. Geogr. Rev., Vol. 2, 1916, pp. 249-255). Similar
observations off the coasts of Formosa would no doubt be of corresponding value for
predicting the weather of Japan.
Atmospheric circulation is particularly intensified in regions where a cold and warm
current meet, as off Newfoundland, or where a warm current runs along the mountainous
ridges at the edge of a winter continent, as near the northwest Scandinavian coast.
Changes in air circulation intensity probably follow closely on the changes of water
temperature. Furthermore, observations of water temperatures may be of aid in the
daily forecasting because of the effects on storm tracks. The effects of changes in ocean
temperatures on rainfall are not so definite as are those on temperature and winds, but
it seems evident that a reduction of evaporation from the North Atlantic by an abnormal
lowering of its temperature over large areas might presage a dry season.
CHARLES F. BROOKS
ARCTOWSKI, IIENRYK. The pleionian cycle of climatic fluctuations. Proc. 2nd
Pan American Sci. Congress, Dec. 25, 1915, to Jan. 8, 1916, Vol. 2, Section 2: Astronomy,
Meteorology, and Seismology, pp. 172-179 (discussion, pp. 177-179). Washington, D. C.,
1917.
Barbato, GERMÁN, JR., AND PEDRO ESQUERRE. Iniciación al estudio de la relación
heliometeorológica. Proc. 2nd Pan American Sci. Congress, Dec. 27, 1915, to Jan. 8,
1916, Vol. 2, Section 2: Astronomy, Meteorology, and Seismology, pp. 561-570. Wash-
ington, D. C., 1917.
GALÁN, ANTONIO. Fluctuaciones climatológicas en los tiempos históricos. Proc.
2nd Pan American Sci. Congress, Dec. 27, 1915, to Jan. 8, 1916, Vol. 2, Section 2:
Astronomy, Meteorology, and Seismology, pp. 475-481. Washington, D. C., 1917.
HUNTINGTON, ELLSWORTH. Solar activity, cyclonic storms, and climatic changes.
Maps, diagrs., bibliogr. Proc. 2nd Pan American Sci. Congress, Dec. 27, 1915, to Jan.
8, 1916, Vol. 2, Section 2: Astronomy, Meteorology, and Seismology, pp. 411-431 (dis-
cussion, p. 431). Washington, D. C., 1917.
KIMBALL, H. II. Measurements of solar and sky radiation. Diagrs., ill., bibliogr.
Proc. 2nd Pan American Sci. Congress, Dec. 27, 1915, to Jan. 8, 1916, Vol. 2, Section 2:
Astronomy, Meteorology, and Seismology, pp. 549-561 (discussion, p. 561). Washington,
D. C., 1917.
SHAW, NAPIER. The meteorology of the globe in 1911. Diagrs. Quart. Journ.
Royal Meteorol. Soc., No. 179, 1916, Vol. 42, pp. 137-152. [Noted in the Review, Vol. 1,
1916, p. 464.]
SMITH, J. W. Predicting minimum temperatures. Monthly Weather Rev., Vol.
45, 1917, No. 8, pp. 402-407.
TAYLOR, G. I. The formation of fog and mist. Maps, diagrs., ills. Quart. Journ.
Royal Meteorol. Soc., No. 183, Vol. 43, 1917, pp. 241-268.
TOLLEY, II. R. Frequency curves of climatic phenomena. Diagrs. Monthly
Weather Rev., Vol. 44, 1916, No. 11, pp. 634-642. [Abstracted in the September, 1917,
Review, Vol. 4, p. 222.]
OFFICERS AND COUNCILORS FOR 1918
JAMES B. FORD
Term expires 1919
President
JOHN GREENOUGH
Vice-Presidents
PAUL TUCKERMAN
Term expires 1920
Foreign Corresponding Secretary
PROFESSOR WILLIAM LIBBEY
Term expires 1921
Domestic Corresponding Secretary
ARCHIBALD D. RUSSELL
Term expires 1920
Recording Secretary
HAMILTON FISH KEAN
Term expires 1919
Treasurer
HENRY PARISH
Term expires 1919
Director and Editor
ISAIAH BOWMAN, Ph.D.
ANTON A. RAVEN
Term expires 1921
Councilors
REAR-ADM. FRENCH E. CHADWICK
ARCHER M. HUNTINGTON
LEVI HOLBROOK
CHARLES A. PEABODY
Terms expire 1919
MADISON GRANT
GRENVILLE KANE
•
ALLISON V. ARMOUR
PHILIP W. HENRY
ALEXANDER HAMILTON RICE, M.D.
Terms expire 1920
BANYER CLARKSON
EDWIN SWIFT BALCH
W. REDMOND CROSS
WALTER B. JAMES, M.D.
H. STUART HOTCHKISS
Terms expire 1921
Associate Editor
W. L. G. JOERG
Assistant Editors
G. M. WRIGLEY, Ph.D.
GEORGE M. MCBRIDE
Contributing Editors
ALBERT PERRY BRIGHAM, A.M., Colgate University
WILLIAM M. DAVIS, Hon. D.Sc., Ph.D., Harvard University
WILLIAM CHURCHILL, F.R.A.I., The Carnegie Institution of Washington
EDWARD L. STEVENSON, Ph.D., Hispanic Society of America
ROBERT DeC. WARD, A.M., Harvard University
RICHARD E. DODGE, A.M., Teachers College, Columbia University
ELLSWORTH HUNTINGTON, Ph.D., Milton, Mass.
MARK JEFFERSON, A.M., State Normal College, Ypsilanti, Mich.
DOUGLAS WILSON ·JOHNSON, Ph.D., Columbia University
OBJECTS OF THE SOCIETY
The objects of the American Geographical Society are to collect
and disseminate geographical information by discussion, lectures, and
publications; to establish in the chief city of the United States a place
where may be obtained accurate information on every part of the globe;
and to encourage such exploring expeditions as seem likely to result in
valuable discoveries in geography and the related sciences.
The American Geographical Society is the oldest geographical
society in the United States. When it was founded, in 1852, there were
but twelve similar societies in the world. Now it exchanges publications
with more than four hundred scientific associations. The Society issues a
monthly magazine of unusual interest called The Geographical Review.
It has also a large and growing library-one of the most important geo-
graphical libraries of the world; thousands of maps and charts; and a
remarkable collection of atlases of the sixteenth, seventeenth, and eighteenth
centuries.
Travelers, men of science, and others properly accredited are welcome`
at the rooms of the Society and may freely use the book and map collections.
Two gold medals have been founded by the Society, the Cullum
Geographical Medal and the Charles P. Daly Medal, which are
awarded from time to time to explorers, writers, and men of science who
have contributed to the advance of geographical knowledge.
In addition it awards the David Livingstone Centenary Medal,
founded by the Hispanic Society of America.
The qualifications for Fellowship are an interest in exploration and
travel, in the spread of geographical knowledge, and in the advancement of
science.
A Fellow is entitled to the use of the Library, Reading and Map
Rooms; to admission to all lectures and exhibitions; and to the Society's
current publications, which include, besides the magazine, occasional books
and maps.
The annual dues are ten dollars.
FORM OF BEQUEST
I do hereby give and bequeath to the American Geographical Society of
New York...