mm 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 4 
 
 / 
 
 o 
 
 ^ -y^ 
 
 
 / 
 
 «/. 
 
 %'' 
 
 A ^" 
 
 1.0 
 
 I.I 
 
 1.25 
 
 ■■ m |||||22 
 
 , ." 112 
 
 u 
 
 1.6 
 
 Photographic 
 
 Sciences 
 Corporation 
 
 ^^ 
 
 #' 
 <;^ 
 
 ^i: 
 
 \ 
 
 -^ 
 
 ^% 
 
 \ 
 
 \ 
 
 23 WEST N\r STREET 
 
 WEBSTER. Nt 14580 
 
 (716) 8/i-4503 
 
 ^ 
 
 *» 
 *> 
 
 *. 
 
 a^ 
 
 <^\<^ 
 
 ^ 
 
 
 6^ 
 
M 
 
 "> 
 
 ^ . 
 
 
 & 
 
 fe 
 
 
 
 CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHM/ICMH 
 Collection de 
 microfiches. 
 
 Canadian Institute for Historical Microreproductions / Institut canadien de microreproductions historiques 
 
Technical and Bibliographic Notes/Notes techniques et bibliographiques 
 
 The Institute has attempted to obtain the best 
 original copy available for filming Features of this 
 copy which may be bibliographically unique, 
 which may alter any of the images in the 
 reproduction, or which may significantly change 
 the usual method of filming, are checked below. 
 
 L'Institut a microfilm^ le meilleur exemplaire 
 qu'il lui a ete possible de se procurer Les details 
 de cet exemplaire qui sont peut etre uniques du 
 point de vue bibliographique qui peuvent modifier 
 une image reproduite, ou qui peuvent exiger une 
 modification dans la m6thode normale de filmage 
 sont indiques ci dessous 
 
 K 
 
 7f Coloured covers/ 
 
 Couverture de couieur 
 
 U 
 
 n 
 
 Covers damaged/ 
 Couverture endommagee 
 
 Covers restored and/or laminated/ 
 Couverture restaur6e et/ou pellicul6e 
 
 
 Coloured pages/ 
 Pages de coulour 
 
 Pages damaged/ 
 Pages endo.nmagees 
 
 G Pages restored and/or laminated/ 
 Pages restaurees et/ou pelliculees 
 
 n 
 
 Cover title missing/ 
 
 Le titre de couverture manque 
 
 Coloured maps/ 
 
 Cartes geographiques en couieur 
 
 Coloured ink lie other than blue or black)/ 
 Encre de couieur (i.e. autre que bleue ou noire) 
 
 I I Pages discoloured, stained or foxed/ 
 i ! Pages decolorees, tachetees ou piquees 
 
 □ Pages detached/ 
 Pages detachees 
 
 I 'i Showthrough/ 
 I I Transparence 
 
 D 
 
 Coloured plates and/or illustrations/ 
 Planches et/ou illustrations en couieur 
 
 D 
 
 Quality of print varies/ 
 Qualite in^gale de I'impression 
 
 
 
 Bouni with other material/ 
 Relie avec d'autres documents 
 
 D 
 
 includes supplementary material/ 
 Comprend du materiel supplementaire 
 
 n 
 
 D 
 
 Tight binding may cause shadows or distortion 
 along interior margin/ 
 
 Lareliure serree peut causer de I'ombre ou de la 
 distortion le long de la marge int6rieure 
 
 Blank leaves added during restoration may 
 appear within the text. Whenever possible, these 
 have been omitted from filming/ 
 II se peut que certaines pages blanches ajout6es 
 lors dune restauration apparaissetit dans le texte, 
 mais, lorsque cela etait possible, ces pages n'ont 
 pas ete filmdes. 
 
 D 
 D 
 
 Only edition available/ 
 Seule edition disponible 
 
 Pages wholly or partially obscured by errata 
 slips, tissues, etc.. have been refilmed to 
 ensure the best possible image/ 
 Les pages totalement ou partiellement 
 obscurcies par un feuillet d'errata. une pelure. 
 etc . ont ete filmees ^ nouveau de facon a 
 obtenir la meilleure image possible 
 
 D 
 
 Additional comments:/ 
 Commentaires supplementaires 
 
 This item is filmed at the reduction ratio checked below/ 
 
 Ce document est filme au taux de reduction indique ci-dessous. 
 
 10X 14X 18X 22X 
 
 26X 
 
 30X 
 
 V 
 
 12X 
 
 16X 
 
 20X 
 
 / 
 
 24X 
 
 28X 
 
 32 X 
 
The copy filmed hero has been reproduced thanks 
 to the generosity of; 
 
 Libriiry of Cotu|r»!SS 
 Photoduplication Service 
 
 The images appearing here are the best quality 
 possible considering the condition and legibility 
 of the original copy and in keeping with the 
 filming contract specifications. 
 
 LoKL'tnoiairo film«^ fuf roproduit grSce A la 
 g6ri«^rosit«'( de. 
 
 Library of CoiKjrnss 
 PhotudupliCiitioti Service 
 
 Les images suivantes ont 6t6 reproduites avec le 
 plus grand soin, compte teriu de l.i condition et 
 de la nettHt^ de I'exemplaire film6, et en 
 conformity avec les conditions du contrat de 
 filmage 
 
 Original copies in printed paper covers a'e filmed 
 beginning with the front cover and ending on 
 the last page with a printed or illustrated impres 
 sion, or the back cover when appropriate All 
 other original copies are filmed beginning on the 
 first page with a printed or illustrated impres 
 sion, and ending on the last page with a printed 
 or illustrated impression. 
 
 The last recorded frame on each microfiche 
 shall contain the symbol — •• (meaning CON 
 TINUED"), or the symbol V (meaning ' END ), 
 whichever applies. 
 
 Les exemplaires originaux dont la couverture en 
 papier est imprim^e sont film6s en commencant 
 par le premier plat et en terminant soit par la 
 derni^re page qui comporte une empreinte 
 d'impression ou d'illustration, soit par le second 
 plat, selon le cas Tous les autres exemplaire.s 
 originaux sont filrn6s en commencant par la 
 premiere page qui comporta une empreinte 
 d'impression ou d'illustration et en terminant par 
 la dernidre page qui comporte une telle 
 empreinte. 
 
 Un des symboles suivants apparaitra sur la 
 dernidre image de chaque microfiche, selon le 
 cas: le symbole — ■*- signifie 'A SUIVRE , le 
 symbole V signifie FIN " 
 
 Maps, plates, charts, etc., may be filmed at 
 different reduction ratios Those too large to be 
 entirely included in one exposure are filmed 
 beginning in the upper left hand corner, left to 
 right and top to bottom, as many frames as 
 required. The following diagrams illustrate the 
 method; 
 
 Les cartes, planches, tableaux, etc , peuvent etre 
 film^s i des taux de reduction diff6rents. 
 lorsque le document est trop grand pour etre 
 reproduit en un seul cliche, il est film6 ^ partir 
 de Tangle sup6rieur gauche, de gauche ci droite, 
 et de haut en bas, en prenant le nombre 
 d'images necessaire. Les diagrammes suivants 
 illustrent la m^thode. 
 
 1 
 
 2 
 
 3 
 
 32X 
 
 i 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
11 
 
..( 
 
QKll 
 
 McGILL UNIVERSITY 
 
 PAPERS FROM THE DEPARTMENT 
 or 
 
 Botany. 
 
 
 iDon 
 
 N 
 
 mnn < 
 
 \'-'i 
 
 l.'^'nr.ff.- 
 
 No. 5. — Myelopteris Topekensis, n.sp. 
 
 >- > BY 
 D.Pr^'PENHALLOVV. 
 
 With Plates II. and III. 
 
 [Reprinted from the Botanical Gazette, January, 1897, pp. 15-31] 
 
 Montreal, 1897. 
 
MVKI.OITKKIS TOI'KKKNSIS, N. SI'. 
 
 A M'W CAKIIOSII'ICKOUS I'r.ANT. 
 
 I>. I'. I'K N II A I.I.OW, 
 
 (wrni I'l.ATKS II AM) III) 
 
 I)uKiN<i the past sixty years a luinibcr of plants, variously 
 described uiuler the names of Mediillosa' ( 1.S32) . I'aliiiacitcs' 
 (1845), Myeloxyloii^ ( iS4(j) , Sten/.tliaM 1S64 ),aiul Myelopteris' 
 (1874), have been obtained from the Carboniferous of France, 
 Germany and Great Britain, but, so far as I am .uvare, 110 
 representative of this group has been obtaineil heretofore from 
 any locality in America. 
 
 Recently Professor C. S. I'rosser has sent to me three small 
 specimens of flattened stems from the upper Carboniferous of 
 Topeka, Kansas. These fragments are about 6"" long and lie 
 in a matrix of calcite.^ One specimen represents the full 
 width of the original structure and is 33""" broad. A second 
 has the edges broken off, but a natural extension of the curva- 
 tures of the sides shows the probable breailth to have been about 
 6'^'". Both of these specimens have been compressed into a 
 flattened mass having a lenticular transverse section with a 
 maximum thickness of 5""" and 8'""' respectively. A third 
 specimen, flattened to an irregularly lenticular mass, represents 
 thin layers of plant residue adherent to the sides of the matrix, 
 and obviously but a small part of the original structure. The 
 dimensions of breadth here given represent very nearly the 
 
 'Cotto : Die Deiulmlitlienin liivit'luiiig auf ilircii inneren Uaii. Dresden, 1832. 
 
 'Corda : Heitr. zur Flora der \'or\velt. 1 845. 
 
 'Hrongniart :'rab. des >,'en. de vdg. foss. Diet. Univ. d' Hist. nal. 
 
 * Goeppert : Die foss. Fil. der perm. Form. 
 
 5 Renault : Ktude du gen. de^.Myelopteris. Acad, dc Paris aa : — 1875. | no. 10. | 
 
 * 1 am much indel)ted to Dr. H. J. Harrington for detenninations of tlie mineral 
 constituents of these fossils. 
 
 1897] IS 
 
 
 t 
 
 I 
 
 :CTg ?i S -S3 %i't'-^<'''^»HMtWJiswtfiW«i^,.u ' HW! ' ,' ' 
 
If 
 
 Hot.lMi.U u.i/l III. 
 
 I I^MAKV 
 
 <n.im«tfr tif Itu- Hiriii tiirr iir in original fnrni. Thr ^n MHr.il . ..|..i 
 !•> iImI <>| Iikiwii I u.iI. Tlir siirl.M r sliows ik ( iisiuii.il iir< ,i«i u| 
 lliiii ( n,il\' ni;i»l«r iniK li hrnkiii u|» iiiti» Miiall angular Iragiiitiils, 
 Itiit It IS iliiillv « li.iiailfri/»'il by » HnnH-whal fiiuly sliialti! 
 a|i|ifaraiUf iliu' Ik ri'imtval hI iIk ...iliral lavi-r, with n»iis(t|ii( nt 
 ixpnsiiif III iln- iiiKlfilyiiij^ >liafi'l-i u| M'lfrt'iichvriia. 
 
 I'lu' tLiiisvi-rsi- scttinii III thi' iiiiir«' |»rrfcrllv |irfs(rv»il 
 H|H'cmKii sliii\v> an iiiilir /niic i ,5""" tliii k. uliii li is (iiminiiniis 
 (111 ,ill sidfs. Ci'nlial tn this ami thus Imiiiiii^' tin- axis of tin- 
 original siriutiiii'. is .1 ili^tiii« tly ilaiktr ami smmwli.it im,ir 
 l>c>ii»iis mass, r«»iilaiiiiii^. luii- aiiii tlinc. small irr(rj,Milarlv 
 roiimlfd massis nl |iyrilf. l)'|i<iii Mil)Mi|iiirit mi(ros«ii|ii( al 
 c'xamiiiatidii, llicsi.' zonal a|i|i('aiaii<t> wnr luiiml tn lir din' tu 
 wi'll lirfiiH'tl (lirffri'iicrs ol striu tiiri-. 
 
 I 111' mil riisn)|iiial lUtails pifsctit iii.iiu li .ituics of iiiti rest 
 ami, allliiuij;li the m-iifral i-ffi-cts of die av ami ( oinprissioii liavc 
 iiiiii to complctfly (U-slioy tlu'^tiirral nlalioiis of |iiiits, ami in 
 many I'ascs, also, to (U-stroy stnutiiral ditails, tlusr last imvc hccn 
 preserved, in some instaiui's, in a n-markahly ptrfii t manm-r. 
 
 Tin; ii;NrKAi axis. 1 In- intiif ci-iitral portion of the stiiii 
 presents a romplete absence of strmtiiial ditail. 'llie whole 
 central area is occupied i)y a mass of dark colored male-ial so 
 disposeil as to indicate its prol)al)le derivation from thin walled 
 tissue, but much altered by tiecay and tlii' sul)se(|uent effects 
 of extreme compression. Here and there, dark colored masses 
 appear, possibly the residue of the mucilajfe ori^Miiallv present. 
 Thiouirhout this rej,non lari^^e rounded o|)eninjfs a|)pear. and 
 while some of these undoubtedly represent the displacement of 
 pyrite, many, and probably all, represent the former locations of 
 vascular bundles, in the dark color and structural character of 
 this area, we find amjile reason for its evident separation from 
 the cortical /.one, as ascertained upon microsco])ical examination. 
 Outwardly, this area is limited by a somewhat well defined but 
 narrow and irregular darker line, which is obviously composed 
 of much compressed thin walled cells, but which, nevertheless, 
 seems to sug)j[est a somewhat definite boundarv line between a 
 
(lAM'MiV 
 
 • »»'»7 I 
 
 .Ul/./iU-// fi/.\ /(>/•/ A/. X,\/S 
 
 n 
 
 tii'ral tutor 
 il an, IS of 
 Irii^'itit'iils, 
 ly sliialiil 
 
 I MIlM l|lll lit 
 
 |iri'sfrv<'i| 
 
 ( lilltilllMMIS 
 
 ixin (»f the 
 Mliat iimrr 
 irrt;>j[iilarly 
 "roscopical 
 
 he due t(i 
 
 of iiiti rest 
 Nsion have 
 rts, ami in 
 liavi' Iji'i'ii 
 ii;iiiiii-r. 
 
 I till' Slllll 
 
 riif wliolf 
 iiatr "ial so 
 hill walli'il 
 •nt rlfccts 
 t(I masses 
 y idcsciit. 
 pear, ami 
 iccmeiU of 
 icitioiis of 
 laracttr of 
 ition from 
 ;uiiinatioti. 
 
 ll'fllK'tl l)llt 
 
 composed 
 I'crthclfss, 
 between a 
 
 central tiicilnlla in wliiih thin vvalli-il (iitiilatiKtilal tis<tiir |>rt'> 
 <lomiii,it('s, ,tnil .1 soini'what ri^icl, or at least firiiur, out) r /om-. 
 
 I 111. M»KirA, No proper tortical struiiiir«' is rrprrsnitt d 
 II! ihiHf spt'titmiis I he outer liiuitH of the ttiatioitn ar«; fhTiiK il 
 \>\ more or Ichh liroki ii iIovvii slramlH of HcU'rciuliyina lelK, with 
 surroiiiiilin)^' parcm liynia tissue, making; it i tear that a certain 
 ,iiiioiiiit of struiture has tieeii ri iiio^ed ; ami thisairords witti 
 wImI has .ilreadv lieiii noted in spii itiieiis of Myelopteris, that 
 "tin- tissue layers outside the sclcreni hyma strands are very 
 rait l\ prcservftl."' In tins ease the thin surface layers of coal 
 .ilitMfiv ilest rilu'il are in .dl proliahilits to lie re^'anled as 
 rt presenliiiL,' theitirtiial struiture, wlinh must li.ive lit in < hit fly 
 or vvholK parent In iii.'itous in eharat-ter, ami of small radi.d 
 vtilmne. 
 
 I'm si 1. I o|< I |( \| I \\ I K. The outer, lontllMloUs /one. i.n""" 
 
 thick as alreailv iltsi riheil, li.is iis maertist uph ilitlen nti,ition 
 from the meilull.i explaim «l l»y the Iar|,(e ;ini<iUiil <if lihioiiH 
 elements whitli it < ont,iins nuiuLf to tlu- pre-i n« t of these 
 t lenients, ami tht pet uliar way in uhiih thi:y are tlistrihiitetl, 
 thev have ser\ed not onlv to prolei:t one .iiiotlu r, Imt they have 
 ,ilso ser\eil to prevent the tl'fti t ol i ttmprission trom f,illiiij,' 
 with full force upon the interveiiinu fuinlanientai strut lure whit h 
 in conse(|ue!ice, h,is often retained its strut tur;il ft-aiures in an 
 exceptitin.illv perfect m.inner ( /ii^'.s. / .iml -»i. 
 
 1'ak1'".N( iiVM A. — The i^rouml tissue, tor the irreattr ; .irt, is 
 much .ilteretl hv tiecav ami ctunpression, so that all structural 
 features, especi.itlv in the central area, have been pretty com- 
 pletely elimin.itetl. ( )ccasioiially. however, when prtitected hy 
 earlier inl"iltralit)n and petrif.ictioii, tir by the resistiiiLT char.icter of 
 the acct.!n|)anyin,i( ;^tramls of hard sclerenchym.i, this part of the 
 structure has been preserved in a very be.iutiful manner iji\^s. 
 I and .?). I''roin these areas it is possible to tlelermiiit; the fact 
 that this tissue consists of very vari.ible. but chiefly liir-^e ami 
 thin walled elements of such a character as to reminti me very 
 forciblv of the fundamental structure in many t)f the lar^^cr ferns. 
 
 'Si)lm>l.:iul)acli: r(i>>. H'.t. lOJ. 
 
 
 i 
 
li 
 
 HOiMXn.il I, </A/7A 
 
 hi«M 411* 
 
 l'li> iiiU'ri'i'lliil.ii sjMCCH onliiLirity mu*I \«illi in Hiich tiNNtit* tiru 
 |ircNi'iil, Intt llit-u' It ID) t-viilciiti' III lilt- I'siHti-nt'f of l.u iiii.i- 
 St i i him iuma. It l);iH Ih-cii hIiomii .ilitMilv that tln' siilf 
 curtu.tl /t)iu' \% !.«;""" tlurk. Willuu Ihi-^ region tlu-n- arc 
 iniiiii-roiiHuval or liiii^t'iiti.illy I'loii^atctl iHinilk'M i>f NtU'rciii'liyiiia, 
 wliii'l) U»\\\ lon^: ttriiiitlH travi*r>tiii^; tlic stem luii^itiiilinallv lor 
 ^'n-at ilKtaiui's <.//c>' /. •/. inul ./). I'Iuhi' straiuK, wlmli \ii\i' 
 till* |ii-( (iliarly slriati-d a|i|ii-arai)('f to tlu' Mirlaci- ol tin- >t|ic( iiiuii 
 wlti-rt'vrr cx|to>*rtl, .ir»' alwav* si'|»ar.itccl Iroiii oiU' aiiulhi i hv 
 Hcvcral lai^f ,iiitl tiiiii \vallc«l |»iirciitliyma «i'lls {fi}^. i) uliirli 
 iiri' Hccn to In- vory |tfrlicctly pnv^crvwil in ctrtain ari-as. ||ju 
 Mi'lcrfniliviiiatoiis clciiu'iits arc always vfr\ tliick wallril in those 
 strands uhiih lii' mxl the mrtix ( //^,^ >' ). t'»it luruini' nimh 
 tliiiiiit'i w.illi'd toward tlu- (inter ol llu-siini wlure thes ollcn 
 appear to III- in a forniative eoiidition. The strands an* Hcparatcd 
 radially by rather wide areas ol (iindaniental tissue (//</'. j), Itul 
 in conseipienee of the ^^i-neral and ^reat alteration in relative 
 positions el'fi-eti d liv lompressiun, it is inipossililc to ditertnine 
 their ori^Miial distrilmtioii. The radial dislnlitition ol these 
 strands thron^h a rather wide /one woidd seem to indii .ite that 
 they ntay ha\e luen developed in more or less well deliiietl con- 
 eentric layers, a relation whieh is certainly implied liy their 
 distriliiilitin within eertain areas (//(,^.'). HeNdnd a limit of 
 
 1.3 Iroiii the siirlace the development ol the strands a|ipears 
 
 to lie wholly arrested. 
 
 \'\^i riAU inMH IS. riu- vaseular bundles are not Ireipieiitly 
 represented, siuie in uiust cases they have been removed by 
 decay, or other causes, and their former positions are then 
 markeil by the presence of rather broail, irre^nilarly rounded 
 openinj^s of variable dimensions, which appear throuj^hout the 
 transverse section ( y/X""-*. /. -'. -md ,,'). and particularly internal to 
 the sclerenchyma /one. Occasionally the bundles are preserved 
 in a very perfect manner, and exhibit all their essential structural 
 features with tfreat clearness {y/^,^ /). The outermost uf the 
 two bundles seen in //{,'". /, when much enlaru;eil (//i,'-. y). is found 
 to consist of several broad scalariform ducts enclosed 011 two 
 
I MM tut 
 
 llHHtif tire 
 
 ,1111.1' 
 
 till' Hllll- 
 
 tlUTf AtK.' 
 
 renchyina, 
 
 iiially lor 
 liH li ^ivc 
 >»|H'i iiiifii 
 Itillirr l)V 
 /I uliitli 
 
 CJIH, I III 
 
 li ill tlioNC 
 illir lllticll 
 lu'\ ullcn 
 Ht'paiati'il 
 
 (C- ■'I . 'it'l 
 II relative 
 (K-tiTinitu* 
 
 m| tlu'Hi' 
 liratc that 
 fllU'il ruii- 
 
 l>y I heir 
 I limit III 
 Is appears 
 
 ■re(|iicntly 
 uoveil l)y 
 are then 
 V riiundcd 
 ,^hniit the 
 nternal to 
 |)reservtHl 
 stnictiiral 
 St ul the 
 ), is I'oiiiul 
 
 li 1)11 two 
 
 • ■•»7 1 
 
 .U r A A r»/7/ A7.S / 1 »/'A A A \.S/M 
 
 »'» 
 
 Hlili'H l>y tttws )ii lliick walleii I'lhruiiH elritHMIts, The phlucrii, 
 rtither Hiiiall tii vuliinie, in hen.* iiituh hrnki ii ilowii. hut it it 
 •titiiiiteii riiiiiiilly outwanl, while in the other hiittille { Jijf. th 
 where it in rather mure perleetly prcHervcil, it •« situated radially 
 iiuv.iri!. The prntuxvli Ml in luri' sicn as n iiTnu\> of siiialU r 
 elt iiienis iiiiuli altered l»y n»iiipre""<i«in ( /f),'. /). or i»i otlnr 
 iiisiiiiices mure perfectly preservi il (///;. /), sometimes un the 
 uuter fuce uf the veNsels, {irid sumetiiiicH un the inner face, hut 
 Hlways hetween them and the phloem. While the hundles vary 
 con><ideral)lv in si/e, tin v all < unlnrm to the en" iteral type and 
 it IS ol interest to note ili.it in .ill their structur.il features, they 
 H^:ree very cli»scly with the iKindlcii of n Hpeclen of Myclox\loi) 
 tltserilied l»y Solnis |,.uil).ich," and xhtt hy Seward.^ 
 
 l*'rom the present material I have heeii wholly un.ihie to 
 oliiiiin ^atisf.utorv det.iiN of th«' structure (»l the bundle m 
 loii^'itiuliii.il section, he;, Olid the latt that the vessels are dis 
 tiiu tlv scalarifornt, and in this respect they conform to the type 
 j^eiierally ohservi d in (mis. 
 
 The peculiar situation td these bundles is not altoj^ether easy 
 to .lecount for. Tlu \ icrtainlv appe.ir to lie between, ;ind are 
 therefore iiiitifrliil Willi, the strands of si K reiichynia, from which 
 eircumstiinee I was at first led to siippos*- them to he collateral, 
 as in the case of I'hoeiii\ .iiid titlier p.iliiis, but .t very careful 
 examination fails to disclose any satisfactory eviilence of such 
 relationship, while in some cases at least the vascul.ir bundle is 
 separ.ited from the nearest sclerenchyma stranil by a broad zciie 
 ni fundamental tissue. Indeed, the evidence, so far as obtain- 
 able from the present material, seems icj indicite that these 
 bundles and the sclerenchyma are altot,'ether independent of or\c 
 another; but in the present unsatisfactory comlition of the 
 niateri.^I now available, :io fin.il conclusion can be drawn. Kroiii 
 the eviilence at hand, however, it would seem that the vascular 
 bundles have their extreme outwanl distribution in the central 
 portion of the .sclerenchyma zone. I-'mm this position they 
 
 •^■o^!^. It'll. ii>i. //f. / / /i, 
 
 •.\nn. lint. 7: />.'. / .111.1 //, /ii't. /. V. /y. 
 
/.'(»// \/l.l/ ( :.!//■ Ill: 
 
 ( I A M \ K V 
 
 ilUllMSi- ill IIIIHlhlT illUMld tin- 1 I'lltlT .111(1 1)C( lilllr llll>->t iiuiiu'r- 
 
 oiis uilliin tlu' ci'iitiii' rii^JDii. 
 
 Si cKi.TiiKv oK.iAN.-.. A ii()t,il)U- fi-at'irr ^li llu" prcsfiil fossil 
 is tlir orciirronci- of mmuTous largi. iiiiicila^f passa^fs. As a 
 nik' tln'SL- stniiluifs ait' imuli altiTiM hv ili-iay ami c-nmpiis- 
 sioii, hilt ill two instances tlu'V wire foiiiul in a vi^y piTfi'Ct 
 state ol preservation ( //;,'-. J\. So far as it is at pii'scii possible to 
 deteriniiK', these organs occur tlirounliuut the siih-cortical region 
 where thev are in nion- or less intimate association with the 
 sclerencliyni.\ stran('s. i-Jsewhere it is not possible 1o determine 
 the (listrihiition satisfactorily, l)nt, fiom our kii iwledi^c ol their 
 occurrence in recent |<laiils, it is a fair inference tliat they must 
 also be distributeil throui'li Die entire l)ody of the fniidaiiuntal 
 structure. 
 
 Measurements of such of these passages as were in a 
 sufficient state of prcservati.m for such a pu'pose showed them 
 to ha\e the lollowinn' dimensions: i.;;xi00jw,; JC);xi3;/x; 
 2 1 1; ■ 143^. From these results it is possible to deduce an 
 averaj^e dimension of 127 I(j2/x. I-'-om this ayain it appears 
 that these passa<.jes mavbe described as of elliptical form, in which 
 the minor and major axes have a ratio ol i : 1.5. The very 
 great size of these structures, unusual except in a few irroui)s of 
 plants, seems to su'^^'nest a comparison with both Lyciclaceie and 
 Marattiace;e. In structure they are siniijle. LoiiL,n'tudinally 
 thev form liMig tubular passaf^es which traverse the stem for 
 great distances. l:i transverse section they consist of large ellip- 
 tical openings bounded by a very regular w.iU composed of par- 
 en:hyma cells often iliffering but little from those of the sur- 
 rounding tissue. Thev are mc>re commonly somewhat elongated 
 tangential Iv to the central canal, and by analogy with similar 
 structures in recent plar.ts we may infer that they contained 
 active proto])lasm. Thev thus forn: the secretory cells, or an 
 epithelium which is not .-.pecially differentiated ^fig. _^). A 
 comparison of the two canals {fig. 2) will serve to show, how- 
 ever, that the secretory cells often show little or ikj dexiation 
 from llie general character of tlie fundamental structure. 
 
( I \ M \ H V 
 st IIUIIUT- 
 
 HMit fossil 
 
 t"-. As ;i 
 
 cumprcs- 
 
 \- [HTli'Ct 
 
 iussil)|i' to 
 cal rr'^ioii 
 I witli tlu' 
 (Iftfiiniru' 
 (• ol tlicir 
 tlicv must 
 idaiiu'iilal 
 
 icrc in a 
 wed tlitin 
 D5X i3;/x; 
 IcdiKc an 
 it a|)|)(.-ais 
 1, in whicli 
 
 The \rrv 
 groups ol 
 l.iCLM' and 
 ;itudinallv 
 
 stem for 
 art^c el lip • 
 ed of par- 
 )f the sur- 
 eloiigatcd 
 itli similar 
 contain. -d 
 A\s, or an 
 
 :^- /)■ A 
 
 how, how- 
 dexiation 
 structure. 
 
 iK>7| 
 
 MVi.i.ori hRis hiI'I-.k i:\sis 
 
 31 
 
 Another important feat'ire of tliese laiials is t«. Ix- found in the 
 fact that tliev are always devoid of contents. llii-. ap|)ears to 
 justily tlie view tliat wliatever they may liave contained origi- 
 nallv was of a soiul)ie nature and thus |)assed out ol the ixuly 
 of tin- plant during the process of petrilactioii. 
 
 In all tlieir principal structural aspects these canals hear a 
 stronj,' reseml)lance to those of Ani^iof^fcris ivcctii (they are of 
 the same tyi)e), and it mav al^o l)e |)ointed out that they are 
 similar to tliose fourd in Riuliiophris / 1 '////</w,sw/// vvhi( h .Seward 
 has recently separated honi M velowlon," as also to tiiose of 
 Myeloxylon itself." 
 
 'rhroutfhoul the transverse section of the Topeka specinuii 
 there are numerous resinous or coaly masses of very variable 
 si/.e, l)Ut cvidentlv originalls contained in special channels or 
 cells, which have heconu' much disorifani/ed, aiul the details ol 
 which cannot now he made out. In huinituilinal section these 
 masses are (;f indefinite lenjfth. but ratiier freiiuently septate. 
 From these features it is |)ossil>le to refer them to the residue ol 
 resin masses which the plant oriifinally contained, and they are, 
 therefr)re, directlv c(jmparaljle with the similar resin bodiis 
 fouiul in recent plants, particularly those of Ani^iopkns tVCiUi. 
 
 It is tiui^- fairlv certain tiiat in the Topeka plant there were 
 orii(inally at least two. and possibly niore, kinds of secretory 
 origans, tiie one holding mucilage, the other resinous matter, 
 and in these resjiects our plant is once more comparable with 
 certain recent forms. 
 
 The general view of tiie internal structure thus obtained 
 permits us to gain some conception of the real composition of 
 this stem or stipe, from which we may infer that a restoration 
 would show a cortical layer of parenchyma at least several layers 
 of cells thick, containing numerous mucilage passages. P'ollow- 
 ing this is a sub-cortical zone distinguished by the presence of 
 nuiiicrous rather widclv separatetl strands of sclcrenchyma, the 
 elements of which are very thick walled toward the cortex, but 
 
 ■'.\nn. Hot. 8: />/. .V///./(V.f. 'i' // t.'"- 
 " .\nn. Uui. 7; //. I, figs. 1,1). 
 
 -r*jSJ^^^^|J(^||flffff^^BnW^?J^''"' 
 
HO/AXJCl/. UA/./.J //: 
 
 I.IANIARY 
 
 Cdiuimially tliiiitu"' walled toward the mciliilla. Tlicsc strands 
 iisuall)' liavf an accompunvin^ iiuicilajfc passaj^c on the oiiter 
 lace, inul are inconstant {':) process of formation tow.ird tlie 
 center. Within this zone, vascular bundles, distinf^uislied by 
 tiieir broad scalariform vessels, appear, anil increase in number 
 toward the central re^non. The central axis consists of a rather 
 broad tract of parenchyma tissue, throuj^h which tlie vascular 
 bu.ndles are distributed in larj, numbers. 
 
 l*rom this point of view, and with ilue allowance for the 
 effects of compression, it is possible to trace a strikinj^ similarity 
 in several respects to a species of Myeloxvlon describeil by 
 Solms-Laubach," and i lore |)articularly in certain respects to 
 specimens of Myelopte is described bv Williamson.'^ The evi- 
 dence is both clear and direct that this plant must be considered 
 as belontrinn' to that peculiar trroup for which the name 
 Myelo|)teris, pro])osed by Renault, has been most jrcnerally 
 emploved. 
 
 In 1S32, Cotta ilescribed certain fossils from the Carbonifer- 
 ous of Kuro|)e under the name of Medullosa, which has more 
 recently become mer^red in that of Myelopteris. Williamson, 
 however, informs us that Cotta's figures of M. clegans are wholly 
 misleading, the structure being represented in a much exagger- 
 ated form, while his two .>^pecies, J/, stiilata and .]/. porosa, 
 remain too obscure to be depended upon without further evi- 
 dence than has come down to us.'^ The genus Medullosa, 
 nevertheless, constitutes the basis of that group of i)lants 
 which, ])assing under several names, ha;; finallv come to be known 
 under that of Mvelopteris. 
 
 In 1845, Corda assigned to his genus Falmacites two plants 
 from the Coal Measures of Bohemia, under the names of P. 
 ntrboiiigciiiis ami P. Icptoxylon.^^ An examination of Corda's 
 figures shows that there is no very great resemblance, although 
 
 "Fos.s. Hdt. l6i, fiji. 14A. 
 
 ■'Fossi, plants of tlic Coal Measures. I'hil. Trans. 166: fin!. /,.?, 4. 
 '■•Koss. plants of the Coal Measures. I'hil. Trans. 166'. 
 'SFIora tier Vcjrweit 4c, 41. //. /9, .?(). 1845, 
 
I.IAMARY 
 
 c Strands 
 the outer 
 •ward the 
 lislicd bv 
 II iuinil)cr 
 if a rather 
 
 ■ vascular 
 
 L' for the 
 simihirity 
 :ribecl l)y 
 rspects to 
 The evi- 
 onsitlcred 
 he name 
 j^cne rally 
 
 irbonifer- 
 lias more 
 illiamson, 
 re wholly 
 exagger- 
 '/. poros(7, 
 rther cvi- 
 IcduUosa, 
 )f plants 
 be known 
 
 \\o ])lants 
 lies of P. 
 
 ■ Corda's 
 although 
 
 lSi>7 
 
 M) / lor/ / h/S /<)/7A/:XS/S 
 
 M 
 
 there is a suggestion of similarity to our plant in the general 
 eiiaracter of the fundamental tissue, and tiie presence of numer- 
 ous mucilage passages. These latter, however, are small and 
 a|)|iarentlv altogether separated from the vascular bundle.s. 
 
 Subsecpient observers have not been unmimlful of certain 
 structural aspects in these plants, which ha\e seemed to suggest 
 their possii)le relationship to the palms, and more particularly to 
 that tvpe of structure represented in the genus IJrac;ena, but much 
 doubt has always been entertaineil as to the possibility of mono- 
 cotyledons occurring so far back as the Carboniferous. These 
 doubts w-ere first ijromiuently expressed by lirongniart as the 
 result of comparing with the plants figuied by Cotta and Corda, 
 new material obtained from Autun, France."' He says "il y ait 
 des differences fort essentielles et (|ue rendent tres difficile 
 d'etablir des rapports entre ces fossiles et les vegetaux vivants." 
 He therefore preferred to regard Cotta's Midullo.ui i/t\!^'-a//s as 
 the representative of a new genus, for which he proposed the 
 name Mveloxylon. which thus seemed to indicate the leading 
 structural features indicated by the former name, the signifi- 
 cance of which was thereby jjcrpetuated. 
 
 Fifteen years later, Goeppert, in reviewing Cotta's species, 
 regartled Midiil/osa cligtiiis as possessing characters which were 
 variously represented in the gymnosperms, in i)alms, and in the 
 ferns. As a generalized type, he applied to it the name of 
 .Stenzelia. 
 
 In 1873, Williamson first drew attention to the belief that 
 the relations of these fossils had not been correctly interjireted, 
 and expressed the view that tiiey were really ferns allied to the 
 Marattiace;e.'' 
 
 In 1874. Renault reviewed the fossils obtained from the 
 Carboniferous beds at Autun, as a result of which he supports 
 the conclusions reached by Williamson, and while he regards the 
 name proposed by Corda to be wholly untenable, and those of 
 Cotta and Goeppert to be insufificiently indicative, he views that 
 
 "'Tab. des gen. ile Veg. l-'oss. Oo. 1849. (Quoted from Williamxm.) 
 "Hril. A.-is. .\<X\. Sci. 1873. 
 
 I 
 
34 
 
 /U)/.l\/l'.t/. (,.l/./:/r/: 
 
 I ia.\i;akv 
 
 of Hroii^Mjiart \vit!i lavor, l»iil ictfards a tlinVrcnt form as more- 
 expressive of tlie rclati()nslii|) wliicli lie (litirniliud.'** He 
 tluit lore says: " I'oiir coiiserver le iioiii, |ireiiiier eii tiale, doiiiie 
 |iar M. Hron^niiart a ces [jortioiis de plaiUes, et eii iiieine temps 
 |toiir rappeler leiir nature, je les desi^iierai sous !e iiom tie 
 M yeloptcris." 
 
 riie yet more recent studies of tliese plants h\ Williamson 
 led him to admit the force of the ari^uments employed Uy 
 Renault and the appropriateness of his nanie.'^ Referenct- to 
 Willianisou's figures discloses several points of resemblance 
 between his specimens and my own. This is to be noted (irsl 
 in a i,n-eat similarity with res|)ect to the Lfeneral distribution of 
 tissues, particuhiily as exhibited in his Jii^s. j and ./, as likewise 
 in the very general removal of the vascular inmdles. The vas- 
 cular bundle i,nven by him ( Williamson, y/^'-. 7*) is closely similar 
 to that derived from the Topeka s|)ecinien [ Jii^. 7), but differs 
 materially from his other representation (Williamson, y/i,''- 7) 
 taki'U from the uj)|)er end of a rachis, which is closelv simil.ir to 
 bundles observed by me in Dioon cdulc, whereby it offers some 
 basis of C(jmparison with the Cvcadacea;. 
 
 In longitudinal section the resemblance is rather close, but 
 in this aspect the To|)eka specimen offers little evidence of a 
 satisfactory nature beyond the general relations of jiarts, anil 
 the structural markings of the vessels winch are seen to be 
 scalariform, as in the ferns. 
 
 Finally, the relation of the mucilage passages to the vas- 
 cular bundles (Williamson, y?^. 14) and of the very large, elli])- 
 tical mucilage passages to the sclerenchvma strands (William- 
 son, Aa''- A))> ^^ '^'^" i''*-' ^■'-"'■y thin walled elements of the funda- 
 mental structure, all |)resent features almost identical with those 
 observed in the Topeka specimens {Jigs. /, ^, .?). 
 
 Williamson's specimens appear to differ from mv own chiefly 
 with respect to the particular distribution of the sclerenchvma 
 strands in the cortical region, a difference which, however, is 
 
 ■' Reclierclies .sur les veKctaiix silicitics d'.Viitun. ( l-'rom Willjumxtn.) 
 " Fos.sil plants of the Coal .Measures. I'liil. Trans. i66'. 
 
I i\\i;akv 
 
 II as more 
 ncl/" ifc 
 late, (Idiiiu'" 
 einc temps 
 L' iKim <lc 
 
 A'illiamsnn 
 ployed by 
 ■tereiice to 
 ;scnil)Iaiice 
 noted first 
 iil)iiti(iii of 
 as likewise 
 The vas- 
 ^ely similar 
 but dil'lers 
 
 t^'i.A'-- 7) 
 • similar to 
 
 Ifers some 
 
 ■ close, but 
 
 deuce of a 
 
 parts, and 
 
 ieeti to be 
 
 o the vas- 
 arife, ellip- 
 (Williani- 
 the funda- 
 with those 
 
 )vvn chiefly 
 ereiichvma 
 lowever, is 
 
 n.) 
 
 IM.,7 
 
 .M\ I i.oi'i r.ias i{<i'i Kissis 
 
 35 
 
 more speeilk: than y;eneric, l)iit my material has been so altered 
 bv compression tliat I slimdd liesitate t<t place much reliance 
 upon tliese aspects oi >triuture, preferring- ratiur to establish 
 the aflinity i)y nicans of the more perfectly prestrved slriutmal 
 elements. 
 
 The distribution of the vascular l)imdles in concentric zones, 
 as described by Williamson, may also be a feature of the Topeka 
 specimen, but for reasons alreaily stated this cannot lie asserted 
 with anv def^ree of confidence. 
 
 More recently Solms-I.aubach* has reviewed the entire 
 relations of this tfrouj) of i)lants, and while lie rejects Renault's 
 name because he rei^ards the evidence as not altogether satis- 
 factory, he prefers to retain HronLjni.irt's name (jf Myeloxylon 
 ••rather than .Sten/elia, because it is better known." He \(\\c^ 
 two il^nires, one of a i,a-neral transverse section, the other of a 
 separate vascular bundle, and it is r)f considerable interest to 
 note that this latter is almost the e.\act counterpart of a vascular 
 bundle obtained from the Topeka fossil (y/!,^ /i. His ^a-neral 
 vii-w of tlie structure is not so satisfactory, but it nevertheless 
 exhibits a close similarity to my own material in all its principal 
 features. 
 
 Solnis-Laul)ach dissents from the conclusions of both Renault 
 and Williamson, holdini( that there are stront^ reasons, on ana- 
 tomical trrounds. for considerin^r the alliance to be with the 
 CN-cadace;e, and cites Mcdnllosa Lciulun-ti as pnjbably affording 
 important evidence in support of this view. 
 
 The most recent contribution to our knowledL^e of these 
 |)lants is that offered i)y Mr. A. C. Seward, who has not only 
 reviewed tlie material originally described by Williamson, but 
 has made a detaileil study of specimens contained in the Hinncy 
 collection of the Woodwardian Museum. Cambrid;^^'. as well as 
 of new naterial derived from the Millstone grit of Lancaster." 
 The o.'r, ^.')ses show that his material is generically the same as 
 tiiat represented by the Topeka specimens as described. In a 
 
 ^'Koss. lii)t. 161. I^'OI. 
 =" Ann. liot. 7:1. 
 
 It 
 
a6 
 
 MO TAX/C.t I. GA/.E TTE 
 
 [jANi-A«V 
 
 scCDiid |);i|)cr, tlic same authority tuakcs a stiulv of certain spec- 
 imens containetl in the Williamson collection and originally 
 incliulcd by Williamson in the ^rcniis Myelopteris, hut which he 
 fiiuls to be in reality ([uite distinct. lie therefore He|)arateH 
 them under the name of Riuhhptcris WiUinnisoni." This species 
 is quite tlistinct from our Topeka specimen with res|)ect to the 
 character of '.he vascular bundles, which are concentric, and thus 
 show a tlistinct approach to the type represented iit Aui:;iof>tins 
 evccta. On the other hand, the mucilage passages, which are 
 also of the type found in Angiopteris, are essentially the same 
 as those of the Topeka specimen, differing only in distri- 
 bution.'3 
 
 l'>om the review thus presented, it is tpiitc clear that our 
 specimen must be regarded as a s|)ecies of myelopteris, accord- 
 ing to the name adopted by Renault anil Williamson, and 
 retained by me as expressing its probable relations, but that it 
 differs s|)ecifically from any of the specimens heretofore 
 described. It may be concluded further that the present mate- 
 rial represents the stipe of a frond, rather than the stem 
 proper. 
 
 Heretofore the representatives of this genus have been 
 derived wholly from the Carboniferous of Kuropc. The material 
 now at hand from the Upper Carboniferous of Kansas thus 
 afforils important evidence as to the wider geographical range 
 of these plants, while the well ])reserved condition of portions 
 t)f its structure permits a further discussion of its possible affin- 
 ities. I have, therefore, carefully passed in review such species 
 of living plants as are available in the botanic (iardens of 
 McGill University, as affording a possible solution of this ques- 
 tion. In jjrosecuting these studie;N I have had in view the sug- 
 gestions of earlier investigators, as well as those which natu- 
 rally arose in my own mind upon making a preliminary 
 examination of these fossils. I have, therefore, carefully exam- 
 ined Cordylinc tcrminalis. Phoenix dactylifcra , Kcntin Fosteruvia, 
 
 =-■ Anil. Hot. 8 : 207. 
 -■'.■\nn. Bot. 8://. /.^A^ *. 
 
f lAMAKV 
 
 •rtain spoc- 
 
 orijfinally 
 
 t which he 
 
 ; SL-paratcs 
 
 his spt'cit's 
 
 )ect to the 
 
 c.and thus 
 
 /hii^io/ifin's 
 
 wliiclj are 
 
 the same 
 
 in tlistri- 
 
 r that our 
 is, accord- 
 uson, and 
 )iit tliat it 
 heretofore 
 scnt niate- 
 tlie stem 
 
 have been 
 ic material 
 msas thus 
 ical ranjj^e 
 if portions 
 sible aflfin- 
 ch species 
 ardens of 
 this ques- 
 w the sug- 
 hich natu- 
 reliminary 
 Lilly exani- 
 Fosteriano, 
 
 I8.J7I 
 
 MM. I A tpTKKts ror/:Ki:\s/s 
 
 >1 
 
 LiitiNiiii borbonha, Cyciis irroliitii, IHoon cdulc, /.lonio intn^ri/oliti, 
 Cibotiiiin tr^iih, and .■hij;^io/'/iii.\ ivcitii. 
 
 A close comparison of the Uracicna type, as represented by 
 Cordvline, shows that any suj,fgeslion of resemblance which 
 might at first appear, has no real basis in structural characters, 
 while in manv essential respects there is a \ery wide difference. 
 Noteworthy points of reseniblance being absent, it is wholly 
 uiujecessarv to enter into a more iletailed consideration of the 
 structural aspects of this type. \'ery nearly the same observa- 
 tions arc applicable to the palms. In this grou|> of |)lants, how- 
 ever, there is a somewhat closer point of contact to be found in 
 the mucilage passages. Here these structures appear as tubular 
 channels of great length, and in this respect, as well as in their 
 distribution and great number, there is a general resemblance to 
 the Topeka fossil. Their detailed structure is, on the other hand, 
 (|uite different, anil it jxiints to a want of aflinity which is sup- 
 ported bv the structure and distribution of the vascular bundles, 
 as also the character of the funilamental structure, and no very 
 searching comparison is required to establish the fact that the 
 affinities of our fossil must be sought elsewhere. 
 
 Hv several authorities the Cycadace;e ha\e been suggested 
 as affording a satisfactorv basis of comparison, a view which, in 
 more recent times, appears to have been particularly urged by 
 Solms-Laubach,^* although he elsewhere agrees with other 
 observers that certain exceptions which have been taken to the 
 cycadaceous character of the Medullos;e are well fouiuled." 
 
 Mr. Seward, vet more recentlv. has given expression to the 
 same view, basing his opinion upon a very critical examination 
 of a large amount of material.-''^ While admitting the many 
 points of resemblance to ferns, he holds that in the position of 
 the protoxvlem and in the structure of the fnucilage passages, 
 as also in the distribution of the sub-cortical sclerenchyma, 
 there arc strong reasons for considering the affinity to be with 
 
 ^*l-'<)ss. Hot. Ibl. 
 'ilMii. 100. 
 "'^Aiin. Hot. 7: iS. 
 
! 
 
 «l 
 
 *. 
 
 fi 
 
 in 
 
 I 
 
 i 
 
 .1 
 
 I! 
 
 .'S 
 
 /mr.iMc.if. i;.\/.i- 1 II: 
 
 (lAMAHV 
 
 tin- cyi.uls ratlur than with thi- firns. Without liopiii^- to 
 settle this i|iicsti<tn at ihf prcsiiit tiim-, it may l)r prufitalilr to 
 coiisiilcr .some of the arguments ailvancctl by Mr. Seward in 
 the light of tvideiice obtained froii) an examination of material 
 ilerivcd from existing species, as also from the Topeka specimen 
 itself. 
 
 Vasili.ak iiiM)i.i;s. — An examination of both tyeadaceoiis 
 jilants and ////i,w/'/<m cvccto afforiis but littU- evidence contrary 
 I') the view urgeil bv Mr. .Seward. The evidence obtained shows, 
 as he contemls, that the position of the |)rotoxylem in these 
 plants is certainlv an argimieiit in favor of the cycadaceous 
 character of .Mveloxylon. On the other hand, the collateral 
 character of the vascular bundles in the latter cannot be taken 
 as final evidence of afViiiity either with the ferns or with the 
 cvcads, as Mr. Seward himself points out. Althougli the longi- 
 tudinal sections of the Topeka specimens havi' given far from 
 satisfactorv results, the evidence to be derived from them indi- 
 cates a much closer resemblance to Angiopteris than to any t)f 
 the cvcads 1 li.ive been able to study. 
 
 .Si;( KKiuKV ()i<(;.\Ns. — lu the Cycadacea', ;is represented by 
 Cyciis nvoliitii, /.oniui iiili'i^rifolid, and Dioon ciiiile, the secretory 
 organs ap|)ear to be all of j>ne kind as represented by mucilage 
 canals. These structures are distributicl throughout the funda- 
 mental tissue and are represented by broad canals v/hich are 
 chiefly limited by tangentially elongatetl parenchyma cells. 
 These latter, therefore, differ somewhat conspicuously from the 
 cells of the surrounding tissue, as already shown by Mr. .Sew- 
 ard.'' .So far as apj)ears from the species above irulicated, how- 
 e^'cr, these canals are alway.s lined with a layer of very thin- 
 walled epithelium cells, which become ru|)tured with age and, 
 shrinking back upon the main wall of the canal, give it a thick- 
 ened and very ragged ajjpearance. 
 
 In Angiopteris cvecta there are three distinct kinds of secre- 
 tory organs: {a) tannin sacs, \,b) resin canals, and U) mucilage 
 canals. 
 
 =?Ann. l{<it. 8: 214. 
 
 1 i 
 
I I A M \ K V 
 
 liopiim to 
 (ifitalili' III 
 Seward in 
 il iD.'itfrinl 
 
 I sprciilun 
 
 ■ciulfiiruiis 
 ;c contrary 
 ncd allows, 
 
 II in these 
 ,'ca(laccoiis 
 
 coilatLial 
 )t be taken 
 ir \vitl\ the 
 
 tliu longi- 
 •II far from 
 tliein inili- 
 II to any of 
 
 cscnttil l)v 
 ; secrctorv 
 y nuicilaife 
 the funda- 
 
 which are 
 yma cells. 
 V from tlic 
 y IMr. Sew- 
 atcd, how- 
 very thiii- 
 1 age and, 
 
 it a thick- 
 
 s of sccrc- 
 ) mucilage 
 
 |H.)7| 
 
 .1/ ) A l.Ont-./i/S Ton. A7. .\.s7.s 
 
 29 
 
 Toniiin stus. Ill U-iiisverM- section the tamiin sarn .ire ultrii 
 harely ilistiiigiii^hahle frtnii the resin canals, In reason of their 
 structural similarity. They occur al>un<lantly in the cortex and 
 ^l^roll^dlout the fundamental tissue, and especially in ilose |»r«).\- 
 imity to or within the limits of the vascular hundles. To me 
 these appear to he the structures rcfcrieil to by Mr. Seward in 
 his description of Kmhiopkns W'illiiiinsoni, when he says, "there 
 are smaller canals in the peripheral part of the phloem of each 
 bundle."'** In longitudinal section tliesc sacs are seen to be of 
 about the same diamet<'r as in the transverse section. e.Ncept in 
 the cortex, where thev assume the loMii of cylindrical cells about 
 three or four limes longer than broad. The contents are much 
 lighter colored than those of tlu- resin canals, and often present 
 a well defined granular appearance. They readily yield the 
 characteristic reactions for tannin, 
 
 Rfsin <v///^//.v. — Throughout the sub-corticd zone, scattered 
 among the sclereiichyma cells ami also central t«) each of the 
 isolated strands, are rather broad canals of indefinite length. 
 Throughout the fundamental tissue, particularly in the neighbor- 
 hood of the vascular bundles, there are also numerous c ;iii.ils 
 which differ but slightly in their structural aspects from the 
 surrounding cells. In all cases, however, they are at once 
 recognizable by the r.ither ilark reil resinous mass which each 
 contains. In longitudinal section the canals ;ire of indefinite 
 length. The contents are often septate. These structures 
 appear tt) mc to be comparable with the black, resinwus m.isses 
 of variable size to be met with not only in the Topeka specimen, 
 but in most of the Kuro|)ean specimens of Myeloxylon. 
 
 Miiciltigv Kimtls. — Throughout the ground tissue of Angio- 
 pteris there may be seen broad openings bounded by more or 
 less taiigeiitially elongated cells. These are the canals from 
 which issue the very large volume of mucilage freely liberated 
 when the stipe is sectioned. These canals are always limited by 
 cells which differ but little from those of the surrounding tissue, 
 except that they are more or less elongated taiigentially. Here 
 
 •-•'Ann. Hot. 8: 214. 
 
 fl 
 
i 
 
 ! 
 
 I 
 
 fli 
 
 r I 
 
 I* 
 
 M 
 
 m>t'i\n u I, i/h lit 
 
 \\SHV\*H 
 
 llun* in no «<|»»cl;illv iliff«niiliattil t|iithiliiim, ;int! in tlii«« rtN|ntt 
 w.' unit will) a Icitiiii' wliitli sirv«s tn sIi.m|iIv M|iai.ili- tlus«' 
 .stiiiitmi's Iroin Uiuhi- hI tin- l.yiatlai «',r. < >ii llir^f uniiimh I 
 mIiouIiI ti'il m» hositalioii iii ^l^•t•i(lill^; as tn wlnllitr a ^;iviii plant 
 wi'rc cvi"iulai"iM)iiH m filimiil iti its alVinitits. l-'inm llii', |miiiiI ul 
 vifW, tlu-n, It \\i>ultl sitin that tlir lt>|i«'ka s|iii iimii in inun- 
 lu-arlv alli«'<l to Iitiih. and tin* same wniild Imlil (riu- ul Myrlu.s- 
 vliin. il \vr an- tn haso an i>|iinii>n ii|Min tlir i'.\«illrnl li,niin-> u\ 
 Mr. Si'waitl. 
 
 Sin-i iiKin \i SI iiiKi'MiiVMA, — Tlu' ilistrilKilioii <>t tin- sclri- 
 i'lu-hyma lan liaidly l>o taki-ii as an arj,Minu-nt uiu- way m tin- 
 ..till r. siiui- lii luilli liTiis and cyfails tln-ir is siuli widf variation. 
 I .should cons'clcr this a sprcilii latluT than a i-fniiu' i liar ictrr. 
 In all of tiu' njycloxylons so lar studiid. tlu' m lirfiu hyiii.i is 
 dislributi'd in st-paratc strands. In the i)iads studinl l>\ inc 
 this tissue forms a continuous i)and in all lasi-s wht-rr slrnn^ly 
 di'\clo|n"d. In .\n>jio|itiTis it forms a rontiniions /oiu' of ron- 
 sidcrahli.' lliickiu'ss, with si'|>aralc stramls lyin^^ aloii^f tin.- iiiiur 
 
 face. 
 
 A rcsunu' i>f the results above- dctailo<i shows that in tin- 
 To|>i.'l<a s|ii'cinicn tluMV arc characters which directly connect it 
 with Kitiliioptcrii Willi, nnsoni, and aUo with other l-airopean spe- 
 cies of Mvelo.\ylon, and tlu- ividence would seem to indiiate 
 that few of these can be separateil ifeiierically. yUlmittinu the 
 force of sonii- of the objections raised by Mr. Sewaid respecting' 
 the tilicoid character of Myeloxylon, there are, nevertheless, 
 Strom,' arguments in favor of this \iew. which seem to me to jire- 
 poiulerate and thus to justify the retention of the name Myelop- 
 teris as a name expressive of this possible relationship; while 
 the fact that these plants do not conform closely It) any modern 
 tvpe would seem to raise a ipiestion as to the possible correctness 
 of the view oriifinally expressed by (ioeppert that these plants 
 in realitv represent a ^Generalized type occupyini;- a position 
 between the cyads and the ferns. 
 
 llolASK Al. I.AIIORAIOKV, 
 
 Ml Gil I. 1'MVl.R-iirV, M'lMKKAI.. 
 
1 1 >sr \H\ 
 
 {ivcii |il.iiit 
 MH |)niiit iif 
 11 i> mure 
 il Myil.ix- 
 
 I rinuifjt of 
 
 I till- «.i;lfr- 
 vv.iy or \]u' 
 I- variation. 
 rhaiatt«T. 
 luliyiii.i is 
 liid l)\ MM- 
 rr sliniij^^ly 
 
 ilJf ol 11)11- 
 
 f tin.' iiiiur 
 
 that in the 
 • coiiru'ct it 
 ■o|)t'an s|)c- 
 to indiratc 
 [iiittin,^: the 
 
 respecting 
 jvcrthelfss, 
 
 nio to |)rc- 
 iie Mvelo|)- 
 sliip ; while 
 my inotleni 
 correctness 
 hese plants 
 
 a |>usition 
 
 •>"•■; I 
 
 l/l / /('/■// A/s .'0/7 A / \Sl% 
 I Ml \S \ I InS Ml ri.A II n Wh III 
 
 .M 
 
 cf yn It 
 
 (•'(»», ». I i,iii>^»l'>t "U Hull 'Imwilik.' Uli •»• lilt 111 In 111,1 li.iiiil', itit liii.il.i 
 iiiriitiii liootif, iiimI Imm ttiM hIiii liiiitilliti. / l^i 
 
 III, • I i,iii«vrt-«i' •»<•' Mull -iMm III.' Ill' «i iiit II' liv iii.i -li.iiiil-, Willi l»»" 
 l;(ii.'f mm il4iK<' |'iii«MK»» • (X 
 
 I I \ I I Ml 
 
 III,, ( ri,iii»vt|i(r Ml linn "I .1 HI IciriM li\ MM -ll.ili'l 'tiumh" <li liiiN <if 
 tliih til II' III II tiiil' iliiu'i' iia-**:!!.'!' I'll il-" iiilli I liii I'. ■ t-'i 
 
 III,. 1 I t.lll'^M'tHC vcill.Ml III .1 V.f.ll|l,(» llllll'llc 'll'lWItlL' 'UirtiN •>! 
 
 >i I III iiiri'. ' iHcj. 
 
r" 
 
 k 
 
 
 t» 
 
 i 
 
mnix/t'ff umrtA, < w// 
 
 Htth M 
 
 MVI.I.OITKklS Tol'KKKNSlS IViili;illu\v. 
 
J fWW 
 
 >"* • :¥'^."^*'.-*^^_"-^'";'*VV"' i™*!*lU'5'l^£('r'*' 
 
/!(>/ l.\/l .1/ (..I// III-, Will. 
 
 I I.I 1 1 III. 
 
 MVKLOi'lKKlS I'ori'.KKNSIS I'enliallow. 
 
Ji 
 
 fel 
 
 I 
 
 I 
 
 I 
 
.) 
 
 s I 
 
 t* • 
 
 !V 
 
 i 
 
 I5 
 
 ^IMlMi^-IF 
 
 mk