pot, ae Hoga ane Me ahi ee a ‘ . ° t i) . ‘news tonnes ’ —— erp atrh be, Beye bey tah Nets" has hated Marbeigtna tas at ‘dean discs « i 4 be Md Lote Lote Heth heel ” S - 4a ‘ , a sonst: mhiaho-roqeie ses ry : boat qos oa ° si aia NGA heirala he. nn Seadely F . Naya ra ash dena tiga ci 1 ~~ othe lta e Ble ett cept tee New Ae b ate DG pir eh in ott y ‘ nd gil toa ie Netted raed At ; sta bbs iadatia sHecistte bet sgsqunen | ies ASCs taint ace Pint wight, saatona bageasd-as Wisi au gh A tory tos Dh bee pee pt PEAT foaled au ledingr ase sande sngte ebadashiaet teal pry uit “ ries Harte 92 bee: } qagrendndaud Levels toahbet At Re std meine whe “f ert ro / haeses wore] tpt i anantedle iat sina: sia an 33 sinatie ne : , : Gite Ind daaeond : 99> ow) oto tata Vitae 49a 4 ee Ne edode ett temeatibins seta 4 4 iar taietebie cis osted J eTray diaharty Me tg8 l srqate) aev eladen jd bie Math bs tight tie, aohcadkolet ade pagicas tideae tere) ta ’ pa serie ° s i LL Meteetibene Dako ty Ser "gotten ty Hay tetcineite ty} aye > - parte 1 ed At ol LA A Gell e® td ee toe rg haattir Corer in ag AtiegaGit ake | y ve at yen ter Pda altel Te tedltne wettab etry ap gga’ teh meta tT Ot hehe raed ti rb ee ‘ Serr merit yg " sonaiceas , 4 tine? ' ‘ H ive 5 ee inde, pen ry emp ~ set tae! Pre raeord) Stoigdd § i ben a abs Ieis' Khatghelt» ‘ 7 on sa eay te Re helt on woah ehid D? pas artery 14h 99 1 A } Peyneneh eh " WM Ag ler rattan OIL : aa. anaddsdah saan’ “ Nerdatell toe ered Bede hyn etattate 149 i Be pe Ved hg et hast hay bel, be sas phode athe y salle . . hediotlpthe sete sits Ait Hbegeh ‘Hianiabaab Prarie VA ry h pay tal abanatialt mo peer ne Mise’ nequoahes sheer ENO TEL f Beat sats NolAwioes! i : Et Aehedy Piek ' ib de bherd rrr te Bel ele Fy a) hd ue ee Sedat fea eataa henge etatepanastt a es Seb | ‘ sda tehne 108 SA Me hott shaielt ate dicerea | ir eek gs ol bys, 9 a pepeyy Lee d wt ecg Osi thet 4a sighs if eats iy massa die sadn te sh Heat Wiiaelta reais i brn Daas A Pate eden ee oll want ional} mibieloine Vek G: (ie Sle dsuahs eRalteneanada Phy : pra ie ata en: ’ ios ?, Othe - dient Staats ripe | sii acest ae ’ te sigh! = lett tet el anny . eyirere vishs aoa veh . ‘ feng ine mii ghsoelsh si aH fem Le tobi detent! stan fh te DTD IE Letts Peleg it at at bp ren MALY Foe bs vghatide tis: he" biti sildigaintsts , thie HH oraan 4 Ora ante pes : Viadet atti: site iate ae aude suekenge . 4he 14% 4h narabed oe Salts m ¢ ve mfaerrtt ety ” “a abet aida My iaaghai ayn tie ene Poss )9 4 nee A tails nar spheres be ie tel easy ath id 4 igh gir » re aan m hgh ae pigtiaety ciebeta ee caaeg page Tiehaisegeee dee iiuecitte aay Mates ay gota’, se Made hitgeter) TN adadtal hate eran PO are ’ Appetite ged a" me fds dali eg dy Tete le Vebelle boete are ‘a PAS CO pw VAM ott 10g 8 alta ts Prd n adm eden r Ho eryAlaeaD® Aare he +t aged “a op nad ally de si ihalret rita agree acne dame eat saat shanties ait Matiz ow Ne ' Ht Be pete nay 9A iery i a negaschigh y ks tee. ia) ani Sena ant vat ae ube ieee : ea chin Parag eng at au — ‘ i ont it te 5 Pept 445 ace rytah ys a) poured aia Cry vn 920 yu neeertenays ; et eine aoe 42 heats tak atest any 4 aia rn ot sil Bt 408 Leite eis Sritataatyete eriaar helt Lottee Ss tnd gl ta tilong ls tants nets ida (A tid ih oy the td nig Se Gait tal tee 4 ny va en ¢ el crite f ndt 4 “ Soret ea fhe a fal hia Mrgietat - ve iy ahsta4 dasete ue see +, init ‘hg tare i : i tity innate ste is Shey Naa 9 ahr at) “3 it atetiye ls ‘43 adeatineat® ae ond tadn fe aa Ag nt her Vetent "Ae, yiiegs Ht siBeantriete Ta heliashs iqiepasity ad V ‘ ‘ “toe itysteyt ee ' i ‘ ‘ bid ‘ ’ , . yy 4 ee An , , te i ‘ ‘ ‘ ' ‘ i Ite ge 4 bed : ‘ equ d aif; “ ; ae : 4 ria deg atl Byte cist 7? ’ ety coletast wie! uaeed NORTH Y ‘ ‘ 14 ‘ ' ¢ os , ‘ ‘ aoe mY ae ' ‘ ‘ 4 ' , { ‘ rt ’ ed hp y ‘ ‘ erreT Lat iy tne ' vate "y 4 aia farts iat 4H nae rt eee vs i re : wo ‘ su bites ie ta tee ot ‘'@ i aa Viv on Ve aid 2 8G Hy Ri aan calaaer “ 4 ’ sateen yy re gihew rs ed a “pein Tied iy ie comeye eel eve ae hed vad poeaiiet ' ne eta ae ‘ w “ hetieed 6h Ge Me RUNG G NOR tabewy . . iri ‘ ‘ ' The! ao sdeaseee 4 . ¥ (ous i 4 1 rt vag Ye eben ode ere Atte h seat iy ’ ‘ te eo PECL SEU EL) ™ ‘ (one wen y ‘ ‘ “ isd aye a Hitates ita he i) ae reas ag rf . ‘ Porrae peeree cee Ut sd cgeege OTOL eek FB wehadey Su deee ne st Arh ne D 4 youre ‘ 7] ' ‘ wae ee Ree Hag lo ite hee atid een a i bre) aide tine *y ’ i ' 7 tiv, a A itreeetioe 4m arab aiebslatyy eet OP UT Ere tH vy UE Aeenes w ae <0 Halt ‘ ¥ ww nb PWh Als pou OF oe ne 19 ph gu Pho: U4 Wo OF * ee us ; " * say . ' “ a) ‘ ay pits Oey ere n ” ; ‘ , +2 ALAS ee 24 yt ‘ i i Bae iy we ‘ ies sy eag ( VOB Os coe, Gres phote Wi ayiue sereey eid ve . i vqeye ie deem tue SUVs F whee bite ered a em aes iam erg oe ame ’ Veena ty Vou pleat Bk bo a R ' “ ' Wiaswen. alae GLA ee 4 ae ue mite yea siurenyayave :T0 nes. “ut We 8G un ' a’ Wiaterig ed ot aoe oenay ’ wy on wa roe yd ee ree “ . Peep eur eny eo ‘9 Uy wd tan, pHemMa ys Lo ai at ' , " a pe ees “ brit ep . 2 OE rrUr ir tt LSbin Be ‘ pon “at ’ ‘ Wi pr orweyatyaree po bs bauer ae +S vies hd vey ~ : red ieventdiubae sia gate onan nae ‘ Ba Vereen nat a i, a Deane be ¥ ‘7 ve ‘ i. He eo oy weptyana oe PPR y Hp TE» 4A ba chs hots nt en me otha tet an Se pene oh bd bh vee aN ; ’ 1 oy i at meee 8 tore ay moo “3 a ' aH i teria ame ; ; : Gi vrrehed dnqrarag v4 ea ited Wiser gta ge ues Vea tot reer tects phate ed ab aaetet “we te " vee leg! powedaat ay Puspe yates SS suena yew eal qvut stay Pnvts 4 ry ae aes ‘ ‘9 ‘ Ee ORE Poe bey M ep 3 oer 4 os ve ” Lateates het aera) 4 Mia eis way meta wi ’ iol wapeyet: . ee 4 water ‘ay aH ‘ 4 cw Hy Ser neat nw rn i saw we ny 4 4 ped Stet * « “ ' wt er baanbiney anheut See eta beta ooty 4 ew ee me a) Lehr eel ° say t ' wwe ‘ ieay ony Prrereube te si Verusnady a4 news went ite . ‘ Po ones Ay esetins : J ” Ne i$ SUR ais year ‘ 4 Fee ott iad Wout ety wae crer ites tier ears ta ah wig Yow ea ins ad Lh sd age i ‘ . ' ‘ wt bad) ret thom ty , Sa on ene , " ae a Pe Ce ’ > pedal eras ryt yeaph pn " 19 pied 4 " a we V4 t ' ’ ees Be ee +440 . f * wakaind en gsi tf oes Perea yewye wr'neqs 3 : ereer anol reiee noe ; i aeett si santa a ieceataleas Pei h bata 4st See tate ‘ ’ ' {woe wee ethepe eh > sei Jovtue ef ww y Atwen “ ad get pew Y ie4g : +) ney ' . ve Vs ' yous ue7h0¢ tae “8 ; . Nor ove teas wags TMi Lee ’ . we) ‘ J . oy ete aaa ‘ :. 1 7 OF batten! me 4 wid avtutaxaeue rage rw . ' io ‘i. wot u qv sqiaoasie ones bc uns pias i al tat 14 enw oo ee a) ov mass tie (a rscyleupoaeatvetsey ‘ ‘ H ‘ ' ’ “ seat y anys Iegemeyvad nace “ : 1 ete vote y Hoe ae doyel = Aig) Ayal nie pat * bf , hare ee Seen ne Smee Orr ic or Oc a Meter Dee ee “ws oh yond wie he a oa bahia) fabs oo ‘ “aon Wed ph are ede dan Ceteutnter Hd Eedeeelt * me pi I ered ou to ome oo Hoge reeded nen Hohe obdedadedeqed ya ainigeg! tewayey: Hawes erates iar seateat Hn 00" ew oh 6 iy ' \ tory we mere fe Oe ates StU MREELM bat TE MeC ron Sitec Lar dei ceeneds , . “nt pele ot 9 een a ¥ > yt etambanfee! Ap Ce eh ee be oc eee oy My rt Vibe neg att yd yt! ” , we ‘ ‘ ’ ta © vi? Ge l” sre se Oovaniy very ae ee bel shonel weeoyeonne 1 wow a+ ‘ ( “ on reer pe [Soe ieqadau th iued é . ‘ J 7 oF) . ‘ BULLETIN } i f, ry 1 \ i (ae fe \\ 4 \ \ 4 i i \t ; OF THE a | ; aM ’ ' MEOLOGICAL SOCIETY OF He NN IE ale 88 VOLS ¢ JOSEPH STANLEY-BROWN, Editor c% iA] Yy LIBRARY ROCHESTER PUBLISHED BY THE SOCIETY 1896 COUNCIL FOR 1896 JOSEPH LE Contr, President C.. H. HircrcocK, Vice- Presidents Kpwarp Orton, H. L. Farrcuiup, Secretary I. C. Wurre, Treasurer Class of 1898 B. K. Emerson, J. M. SArrorp, Class of 1897 | | | R. W.. Exzs, > Members-at-large C. R. Van Hiss, | Class of 1896 | F. D. ADAms, ) I. C. Russet, PRINTERS Jupp & Derwerter, Wasuinaton, D. C. ENGRAVERS THe Maurice Joyce Enaravine Co., WaAsHineton, D. C. CONTENTS Page Proceedings of the Seventh Summer Meeting, held at Springfield, Massachu- setts, August 27 and 28, 1895; Herman LeRoy Farrcni.p, Secretary......... 1 Session of Tuesday, August 27........... Bae hae Pinte crater ae Eo gai fe 1, GEC ORO aMCMONTS Reeve in a Set aerdie es laels hiv cerei sal slo vie eleiac'v sds ed ves 1 Champlain glacial epoch [abstract] ; by C. H. Hirconcock......... 2 Geology of old Hampshire county, in Massachusetts [abstract] ; by Lee mL VERS OING! a arenes c/s ed Seed ee veces oi Lon ies ds Mik aeSag ardlieg ses Ly 5 Session of Wednesday, August 28........... WS sitet ars 6 cree Sroate eta tcess 7 Bearing of physiography on uniformitarianism [abstract]; by W. lg DANA on Stee 68 Bic Baers Maree RT A 8. 2 ae ee 8 Marthas Vineyard Cretaceous plants [abstract]; by ArtHur Ho t- TLIC sete aes Sak ote ahi eee Ret Seca Aen er 12 Titaniferous iron ores of the Adirondacks [abstract]; by J. F. Kemp 15 Register Ofer spmmomeld: MeCtME 2) feu. a. os ielsbl 2 keg ed se aeneee sone: 16 Drumlins and marginal moraines of ice-sheets; by WARREN UPHAM.......... 17 Glacial deposits of southwestern Alberta in the vicinity of the Rocky moun- tains; by G. M. Dawson, with the collaboration of R. G. McConnunu...... 31 Geographical evolution of Gane lye) BAW co SPEIN@HR epic cin cutaia Ses 2c) Bote ere 67 Syenite-gneiss (leopard rock) from the apatite region of Ottawa county, Can- ada.e sr (Ob TS iRACGTOTRIDY os ies, elas ate neciegen Menai ea grt: | alta lean eee 95 Studies of Melonites multiporus; by R. T. Jackson and T. A. Jaaaar, Jr....... 135 Sauoecton laleechinordesn ; soy It. TL. JACKSON. 2). 02 aie we lc boas aloes nee 171 Decomposition of rocks in Brazil; by J. C. BRANNER.. 2.2. 60.60 cnc c een en nes 250 Relations of geologic science to education; Annual address by the President, il SS, SUREACLIETRIS B/ SERS ANG else SIRS eter neem RON Ess EL Sey reich co old Preglacial and postglacial valleys of the Cuyanees and Rocky rivers; by War- FEB TONPIBDA RI OSS Ss Ne Mane En REA fo 5.4), ea ee eR aa O27 Disintegration and Heconesicion: of diabase at Medford, Massachusetts ; by G. IO. ICR BRI ibie 35d Gea aialig ORS eect i cos Oe 5 ts cls Ske ce ee em ean 349 Geographic relations of the granites and porphyries in the eastern part of the OTT, 2 1057 Qe T Bee SI TDS SEIS A Bai AS ic 2 1 oh re en 363 Plains of marine and subaerial denudation; by W. M. Davis................ BY Ms Cuspate forelands; by F. P. Gutiiver....... Breil Verb, Man ae Sear 399 tcirinGencceenakes: iby Tl Wt HAWRORIDI 565. es ea bo ase ois wag ees 423 Proceedings of the Eighth Annual Meeting, held at Philadelphia, December 26, PME S89 > kLERMAN UBIO MATROMILD, S€Crelary... 2.202.205 60 eee ee 453 Sescomeou, Dinursdayy. December One 22 oho gC clo suce oie Macleod Sele ones 453 Report of the Council.......... Pera. der rciat, steal aie Wie eet ah cen eh yd SEO REUE LC ASAE] COA SS Gan Cee tee aS = eo eee alee Ree ne aes Leet) 454 SU ECACUIBE TIS PEC WOM am re rie clea one Cigil cars,» stake @acminte © 4 elon ciel aiehsvoueus 456 HG CbON SCRE DORUM lores ice cose ck cia scat ie lots eaten ema, tone 458 IPHCC HOG Ol Oldie Cher eps nc fk eco al v aycod: wie sthaienerereting'n etie™s wusi'eteqs © earenetapeiere 460 FASC EROMO Tem M EMG VEE eas y.nc58 Cid us of eats cietarted Coad ata Mis tes! eal Rickey Mich ataneneks 460 Memoir of James D. Dana [with bibliography]; by JosrepnH Lr CONE es hese, Salta Pei danine Sih ee a Beat ate ete Cea ee MoS ae Een ea 461 (iii) lv BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA. Page Memoir of Henry B. Nason; by T. C. CHAMBERLIN.... .......... 479 Memoir of Albert EK. Foote [with bibliography]; by Grorce F. TRUING Hee ole nian a nO ets Mee ioe Al ye 481 Memoir of Antonio del Castillo [with bibliography]; by Ezrquren ORDONEZ: «pi. 2455'S Sh Sede Siete he ec ead ne a Seed aie 486 Illustrations of the dynamic Tierartion pies of anorthosites and re- lated rocks in the Adirondacks [abstract]; by J. F. Kemp.... ... 488 The share of volcanic dust and pumice in marine deposits [abstract] ; by N..: 8, Gmiaiir), Juris <.cc ape ae eee mech at, Step natings Rigye tara 490 A needed term in petrography [abstract]; by L. V. Prrsson....... 492 Session of Friday; December 27 s022 ee ees cee ee ia eek oe ae 493 Sixth annual report of committee on photographs....... -.. . 494 Examples of stream-robbing in the CatskiJl mountains abstract: by N.. H: Darrow.) te pera eee ok 2 se ean! oe = apes ee 505 Notes on glaciers [abstract]: (oy. BoE Rem: J... 02 cee eae ee 508 Session of Saturday, December 28e 28. cii06 Vo. etal ss keels tek ee eee 509 Paleozoic terranes in the Connecticut valley [abstract]; by C. H. FLITOHCOGE:' } 05.222. shy Seok oa ee kes hee aioe te ie ea eee 510 Notes on relations of lower members of the Coastal Plain series in South Caroling’; ‘ny NaS Daas oil hee eee ee eee 512 Some stages of Appalachian erosion ; by ArrHurR KrirH........... 519 The Cerrillos coal field of New Raion [abstract]; by J. J. Srrv ENSON. 525 Register of the Philadelphia ‘meeting, .< - 6on)fe ssa Syke be ye eee 528 Officers and Fellows of the Geological Society of America........... ~» 529 Accessions to the library, from January, 1895, to March, 1896 ......... 539 Index. to volume 7.2.0). 2e500) Wawel ee att ee ee Ce 549 ILLUSTRATIONS PLATES ’ 1—Dawson: Boulder-clays and Saskatchewan gravels (2 figures)....... 49 2—JACKSON: Paleozoic echini (7 figures). : jock... 020 ose Oe ee 247 3 i. 4 CUO Gomes )e. 2 fen be 06 ng cr 248 { . * “8 AD Eipaares) cies bined xo a 249 5 7 “* . @& figures). Ms ..5':.+5.e aie ee 250 6 “s “a “* - (CLI figares)s 5°... 0a ede s eae oe 7 te a °°. BAUR ae aoe pees intl ebiepe sete 252 8 a % “(6 Benves) * ot,). tee eaeaie ae ie eee 253 9 ea Ki Sf AT UDO te Set ante chee ie a 254 10—Branner: Entrance to the bay of Rio de Janeiro.................. 269 11 os Exfoliating gneiss at Copocabana beach, Rio de Janeiro. . 272 12 cs The Gayha, 20. Sense ee ass ae cai bakes eee 273 13 a Boulders of decomposition on Paqueta island, bay of Rio de: JANEIRO (3 ta:00' dahon ond eo es oe a ES 277 14 ie Boulders of decomposition near Rio de Janeiro......... 278 15—Upnam: The Cuyahoga delta and glacial lake beaches in Cleveland, ODIO Sa ivicdset wed eee ow ee homens Se % Aes bree eee ee 327 16—Merriti: The great dike in Medford, Moamaatidesits yo Se oe 349 ILLUSTRATIONS. Vv Page Pilate 17/—Kuyrs: Hastern part of the Ozark uplift............ 000. Fee 363 ‘¢ 18—Guutiver: Sand point, Prudence island, Rhode Island............. 399 ‘¢ 19—Faircuinp: Hydrography of the Genesee valley................... 423 Pe 20 a Outlet channels of glacial Genesee lakes (2 figures)... .. 439 ors | 9 Water-plane and contorted clays of glacial Genesee - NANOS Ap GUERIN 8 atts ees aes) cela esata wines Ase e a, Swed 442 oe COnnEs POrtralt oh James Ws Da maine... occicin site ecole sAiere see Sess 461 «¢ 23—Darron: Portion of the Kaaters kill atlas sheet, U. 8. Geological PUEEN EN ecient a DU ee a dee ag ae ea 505 see ernrrens~'Rire, T Clnessee DASE... fcc occas loo oe Sa be cee vues eeneu: 519 FIGURES Dawson: Figure 1—Southwestern part of the district of Alberta................. 33 So —oechion on the. south: fork*of Oldman river....--.......$50.. 43 ‘¢ 3—Diagrammatic section from Lethbridge to base of Rocky mountains, along the valley of Oldman river............. 45 ‘* 4—Section in bank of Bow river near Calgary.................. 54 ““ 5—Section across Bow valley at Morley........... .........4.. 57 SPENCER: elope mic Mic On OU a tiers. o's Shy ead es Rigs wea gow eee MOAR 69 ee eo DASCIE Vel Valle yin. iis 2 Fein sca Gye alsa aes teoteceaeeg « Deserta e 69 ‘¢ 3—A section along railway on the east side of Havana bay...... 74 pet —-Sechlonvalone YY UMUFH-EaNyON (css 5. .e8 sels wns ge es ee eb oes 76 ‘¢ 5—The coastal chain of Tertiary mountains dislocated by faults. 80 6—View of terrace (Matanzas limestones) at entrance to Xagua (OE NPA ENS SenOe c cctelee NE Nate vane MIE cnet ea orn orm mL I toh 9 83 pee —ban de Matanzas from: the south. 5) actly 644 gece wh 87 eo ran- de, Matanzas from). the: northiv kt Saved Ss ooh e went ee 87 eo = baycand Valley. At SANtIAL Oy: cmaboss abet setae onions 90 Sm NCEA OUT Am DELIV Coe ray dynos) eco Siete) Sha GTR e's sich arid sre scheint se ehegkediets on ‘¢ 11—Cross-section of outlet of Xagua bay.................. 2... 91 pete Na, Ol ALEMULTT Wall Gy... .za ete eet ne rete plaka hea ode SS 92 ‘« 138—Longitudinal section of ridge closing Yumuri valley......... 92 GORDON: Figure 1—The apatite region of Ottawa county, Canada.... .......... 97 “« 2—Dike of coarse syenite-gneiss cutting quartzite and pyroxenite. 100 ‘« 3—Intrusion of syenite in pyroxenite and quartzite............ 101 ‘ ‘‘ 4—Dike of coarse syenite-gneiss cutting pyroxenite and inclosing POLMOns Ol bhelaLbetearme actos asa gol cleat anderen a so 102 ‘¢ 5—Section transverse to granular bands in the coarse syenite- CANETISS So. 5 155, o MN Net oe OR ERIS Dk ane ye are Ne CRE RR, inane eo 108 “« 6—Section showing granular band intersecting the rock......... 109 ‘« 7—Section transverse to the pyroxene band in the ellipsoidal SiOIUIUE OMOINS aera aun weniger atie a tiaUrylelonle Sa Hk aja Ss % pene lta ‘¢ — 8—Idiomorphic hornblende arate | in the streaked syenite-gneiss. 118 ‘‘ 9—Pocket in the pyroxenite filled with apatite and calcite...... 127 JACKSON and JAGGAR: Figure 1—Ideal arrangement of interambulacral plates in Melonites multi- FDO Siecle Sess i ear A een tones REM “yn ee UN NS ee eRe ge aa 164 v1 BULLETIN OF THE GKOLOGICAL SOCIETY OF AMERICA. Page JACKSON : Figure 1—Relations of the ambulacral plates........ AS ee aaa Pe aap Le ipe— Lepidocentrats antullert, SCM: ils acti des stein ex Jahon oe 223 ‘* 38—Young Goniocidaris caftaliculata, A. Ag., after Lovén......... 254 ‘* 4— Adult Bothriocidaris pahleni, Schmidt, after Schmidt......... 234 ‘< 5—Side view of young Goniocidaris, the same as figure 3....... ; Joe BRANNER: Figure 1—Diagram illustrating the exfoliation of cones.....,.........- 270 ne gs a a2LO WL ALS SUGARY 1, 2's «apelin breisl alee ata eee yas ata Rea pe 3,8! ‘¢ 3—The Corcovado from Botafogo, Rio de Janeiro.............. . 248 toe 4=—Dedo de Deus... . i552. eee mere. cre iaa teen rb LN, Lek 274 ‘* 5—Gneiss peaks with vertical ravines, Nova Friburgo, Brazil ... 275 ce 6—Diagram illustrating the exfoliating effect of changing tem- PPPAUUTE «:1°..5 i teioe & ote eeerene S aehme atret os eee oi sha ee UPHAM: Figure 1—Section on the lake shore across the preglacial valley of Rocky WEVOM soo: 5) «nie ore Saheb phage eter asicle ol YS wok nish Hen Ot eh Pues oe ree 32 SMU OL OrEVIOUS ODSELVALIONS. .. 2.45 c cee ec eee set dene a de eee eels 30 peewions in the valleys of Belly and Oldman rivers..............5..00-000-s08 39 Pam aroretne Porcupine hills... 55565. sec eee ea eee ede mens 44 mamancd valley west of the Porcupine bills..............2.-0. cceceee cece 48 MME CMT CT VING: WCTMUL Ys. ata sineo-cc « Mleid ox s/h onlele ¢ celsia ors bie sets cle wales Were eis du 49 ee mmr O Oe airy. OF bai sk Nien e Gialse ee alata oo 2ies aia! ial aime oreual sus sb 'p etal dials 50 Sections in Bow River valley..... oe Ae ci ARR Se irae eee Barren 51 “PLDI? SUG CHISSGUUISISTIONIV Rae? eau Reese oe ey na 58 INTRODUCTION. The western plains and the Rocky Mountain region of Canada un- doubtedly constitute one of the most important fields of investigation in’ connection with the glacial period in North America. The area there characterized by glacial deposits is an enormous one, but the facts de- rived from it have so far been accorded comparatively little weight in the construction of hypotheses for the continent. Of these hypotheses those in best standing have grown up chiefly during the detailed study of the southern portion of the glaciated region of the east. Distance, and a general unfamiliarity with the somewhat complex physical features of this western region, have undoubtedly prevented a ready appreciation of its phenomena, but these also must in the end be fully reckoned with before satisfactory conclusions of a general kind can be definitely reached. In former papers* the writer has endeavored to combine the observa- tions made by himself and others in the Cordillera and adjacent parts of the Great plains in a common scheme, although one admittedly of a char- *Am. Geologist, Sept., 1890, p. 153. On the Physiographical Geology of the Rocky Mountain region in Canada. Trans. Roy. Soc. Can., vol. viii, sec. 4, 1890, p. 4. V—Butt. Geon. Soc. Am., Vou. 7, 1895. (31) 32 G.M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. acter entirely tentative. In the following notes his purpose is merely to amplify previous observations on a particularly interesting part of this western region by the addition of new facts, given, as far as possible, apart from any theoretical considerations whatever. In the concluding pages, however, an attempt is made to indicate the more obvious deduc- tions which appear to flow directly from the examination of the particular district in question. In a report by the writer on the southern portion of the district of Alberta,* the principal facts then ascertained of the “superficial geology ” are given, but the work upon which that report was based was directed chiefly to the “solid geology ” of the country, and details respecting the superficial geology were as far as possible eliminated in the interests of brevity. Since the publication of that report great advances have been made in our knowledge of the glacial phenomena of the northern part of the continent, some of which seemed to render the region particularly referred to in this paper one of especial importance as the meeting place of the deposits (whether immediately or proximately derived) of the Cordilleran and Laurentide ice-sheets. Thus it became desirable that an attempt should be made to further investigate this region and to test the previous observations and conclusions. With this object in view, a couple of weeks in the early part of the summer of 1894 were devoted chiefly to a critical examination of the superficial deposits of that part of southwestern Alberta adjacent to the eastern slopes of the Rocky mountains. The writer was accompanied by Mr R. G. McConnell, who had previously acted as his assistant in the same field, and, while he as- sumes the responsibility for the statements made in the sequel, those observations made by Mr McConnell will be given under his own name and in his own words. He would further take this opportunity of ac- knowledging the value of Mr McConnell’s codperation, and of stating that in regard to the observations of fact, at least, there is complete unanimity between himself and that gentleman. PHysIcAL FEATURES OF THE REGION. The region treated of may be described as extending from the inter- national boundary northward to Bow river, or in latitude from 49° to 51° 20’. The eastern edge of the Rocky mountains proper (Laramide range) is defined by the line separating the Paleozoic rocks from those of the Cretaceous and Laramie, and, although this line is not a perfectly definite one, it corresponds closely with the orographic features, and the eastern front of the mountains is often particularly abrupt and striking. The want of definiteness referred to arises from the fact that embayments * Report on the Geology of the Bow and Belly Rivers region. (seol. Survey of Canada, 1882-’84. PHYSICAL FEATURES OF THE REGION. ao and infolds of Cretaceous rocks occur in this part of the mountains, while at least one isolated area of Paleozoic rocks is found to the east of the main margin of the range. Both the mountains and the adjacent foot- hills have been subjected to similar parallel folding and disturbance at the same post-Cretaceous orogenic period.* ROCKY SPRING: PLATEAU: Scale 10 $ Oo 10 20 30 40 50 60 710 80 90 1 FIGURE 1.—Southwestern Part of the District of Alberta. The foothill belt varies in width from 10 or 12 miles in its southern part to about 20 miles at the north, in the vicinity of Bow river. Funda- mentally, the foothills represent a bordering zone of folded and con- torted Cretaceous rocks, reduced by denudation to series of more or less nearly parallel ridges and valleys. The rivers and larger streams from the mountains generally cut across nearly at right angles in wide and relatively low transverse valleys, while the higher ridges and hills occa- sionally surpass 5,000 feet in elevation. * For some notes on this and on the Pliocene history of the region, see Am. Jour. Sci., June, 1895, p. 463. 34 G.M.DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. ’ On the east the boundary of the foothills proper coincides with that of the flexed strata, and is nearly always quite definite, the corrugations ceasing abruptly and being succeeded by a wide, low syncline, which is continuous between the latitudes above referred to, and is occupied by the remnants of a long elevated plateau—that of the Porcupine hills. This plateau is throughout composed chiefly of sandstones of Upper Laramie age, but the Porcupine hills proper extend only from Oldman river northward to Highwood river, a length of about 60 miles, with an average width of some 20 miles. Further north they are represented by a series of detached, lower plateau areas, which continue to border the foothills on the east, while to the south of the Oldman the same syncline is also occupied by plateaus, but still less prominent and lower. Of the Porcupine hills proper, the highest part extends northward from the Old- man for about 40 miles, and here a few points reach 5,300 to 5,400 or even 5,500 feet, while considerable areas of ridges and broken plateau exceed 4,500 feet. From the southern end of this high region, overlooking Oldman valley, the view is open to the base of the Rocky mountains, no comparable elevations of any extent existing in this part of the foothills. In the are from west to southwest the mountains are distinct from 20 to 25 miles, but from the last bearing, around to south, the line of the mountains recedes rapidly, being more than 40 miles distant where it crosses the forty-ninth parallel. From south to southeast the lower continuing pla- teaus already mentioned are overlooked, but from southeast around to north the outlook is across the sea-like expanse of the Great plains, of which the rare, low, plateau-like elevations are scarcely distinguishable. A profile drawn across any part of the country above described would show on the west the rugged front of the mountains (7,000 feet or more), next the much lower but irregular foothills, then a well marked depres- sion separating these from the Porcupine hills, then the plateau of the Porcupine hills, and lastly the long eastward or northeastward slope of the Great plains; but a profile traced along the valley of any one of the larger streams, and thus following the actual drainage level of the coun- try, would show a nearly uniform descent from the base of the moun- tains, only slightly increased in slope while crossing the foothill belt. These streams leave the mountains at an average elevation of about 4,550 feet. Along the eastern edge of the Porcupine syncline the plains have a nearly uniform height of about 3,300 feet, with which the general level of the rivers may be considered as practically coincident, although these often occupy postglacial valleys of from 100 to 200 feet in depth below the adjacent plain; thence to the northeastward the surface of the plain (with its rivers) gradually descends some 1,000 feet in a distance of about 120 miles. PHYSICAL FEATURES OF THE REGION. 30 The two most notable breaks in the continuity of the foothill belt and the Porcupine Hills plateau are those of the Bow valley and the valley occupied by the Oldman and its tributaries. The latter especially, which is not merely a wide river valley, but occurs in conjunction with the breaking off to the south of the highlands of the Porcupine hills, is an important and wide opening in the approaches to the mountains, and may be regarded as an irregular southwestern embayment of the plains, in which Laurentian erratics had already been found at an elevation of 5,280 feet above sealevel and upon the very margin of the mountains themselves. It was therefore chiefly in this region and in that of the Bow valley, taken in conjunction with the elevated tracts in their vicin- ity, that further information respecting the conditions of glaciation and the character of the western edge of the Laurentian drift seemed likely to be obtained. The southern high portion of the Porcupine hills in par- ticular, it appeared, might be of peculiar importance in relation to such questions, for here it was probable that either moraines or terraces might characterize the farthest and highest limits of the drift of eastern origin. SUMMARY OF PREVIOUS OBSERVATIONS. Before stating the results of the late investigation it will, however, be useful to give, in the form of a summary, the facts connected with the superficial deposits previously recorded in the report of 1882-'84. In the region of the Great plains of southern Alberta, to the east of the Porcupine hills and their representatives, an approximate estimate of the drift deposits as a whole makes these to average about 100 feet in thick- ness. In a few places on the line of section afforded by the Belly river all the recognized members of these deposits are together present, but in others only two or three of them are seen at a single locality. A com- plete section shows in descending order the following succession: 1. Stratified sands, gravels or silts. . Upper boulder-clay. . Stratified interglacial deposits, sometimes including lignite. . Lower boulder-clay. . Quartzite shingle, sometimes with stratified sands and silts. OU HR OO bo The absolute and relative thickness of each of these deposits varies much, and along Bow river, somewhat farther to the north, the inter- glacial beds were not noted, and no line of separation as between an upper and lower boulder-clay was in consequence determined.* The under- *This may, however, in part result from the fact that the importance of such a separation was not recognized at the time these sections were examined, but it is certain that there is here no such striking plane of division as on Belly river. Still further north, on Rosebud creek, Mr J. B. Tyrrell again found two boulder-clays separated by a thin layer of lignite. Geol. Survey of Canada’ vol. ii, new series, p. 143 E. 36 G. M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. ’ lying “quartzite shingle,” subsequently named by Mr McConnell the “Saskatchewan gravels,” * was, however, seen in a number of places along the Bow, the evidence here, as elsewhere, being such as to show that this deposit, although widespread, is generally characteristic of the relatively lower tracts of the plains. It is thus not often possible to determine, where boulder-clay is met with in isolated exposures, whether the lower or upper boulder-clay is represented, but it is probable that the upper or newest boulder-clay is that generally seen in all the more superficial excavations. ‘“*Overlying the boulder-clay are widespread stratified deposits, the distribution of which assists materially in giving uniformity to the tracts of level plain. It is, indeed, quite exceptional to find the surface soil consisting of boulder-clay disinte- grated in place, and this occurs only on the slopes of plateaus, or in hollows formed by denudation. That the beds overlying the boulder-clay have not been merely formed by its rearrangement in water without the addition of new material, is in- dicated by the fact that in many places erratics much larger than those character- izing the boulder-clay of the locality are found strewn over the surface of the country.f The beds observed in river sections and elsewhere to overlie the boulder- clay are generally gravels or sands below and sandy or clayey loams above. The latter form the subsoil over most of the region, and are generally rather pale brownish- or yellowish-gray in color.” Further study has served to verify and in some directions to amplify the statements summarized in the foregoing paragraphs. On the subject of terraces and water-leveled tracts it is said in the Same report: ‘‘ Terraces are prominent features in some parts of the river valleys in this dis- trict, but are generally clearly due to the action of the river itself at a former period. The extensive tracts of almost perfectly level prairie which occur, afford evidence of water action of some duration and may be regarded as wide terraces.” The conditions of the drift deposits in the region of the Porcupine Hills were not fully examined at this time and it is merely stated in the report that— ‘The eastern face of the Porcupine hills appears from a distance to be very dis- tinctly terraced, but this aspect was found to be due to the outcrop of the nearly horizontal sandstone beds.” Further and more extended investigation in 1894 shows that while the existence of these sandstone outcrops has contributed to the form as- sumed by the Porcupine hills, true water-formed terraces also exist and are actually found to extend to very great elevations, as more fully noticed in the sequel. Respecting the general aspect of the drift deposits in the foothill re- 7 Or “South Saskatchewan gravels.” Ann. Rep. Geol. Survey of Canada, vol. i, new series, p. 70 C. {Compare McConnell. Op. cit., p. 74 C. —— RESULTS OF PREVIOUS INVESTIGATIONS. Si gion between the Porcupine hills and the base of the mountains, little change can be made in the following statement given in the report of 1882-784 : ‘Terraces in the entrance to the South Kootanie pass, at a height of 4,400 feet, have already been described in my Boundary Commission Report (1875). In the valleys of Mill and Pincher creeks, and those of the forks of the Oldman, east of the actual base of the mountains, wide terraces and terrace-flats are found, stretch- ing out from the ridges of the foothills, and running up the valleys of the various streams. Actual gravelly beaches occasionally mark the junction of the terraces with the bounding slopes, and they have no connection with the present streams, which cut through them. The level varies in different localities, but the highest observed as well characterized attains an elevation of about 4,200 feet. Inthe Bow valley near Morley, and thence to the foot of the mountains, similar terraces are found, which are quite independent of the modern river; and in the wide valley of the Kananaskis pass a series of terraces was seen from a distance which must rise to an elevation of at least 4,500 feet.” It is important to note that in all this region there can be no doubt as to the origin of the crystalline erratics attributed to the Laurentian pla- teau of the east. Neither the Cretaceous nor Laramie rocks of the plains nor the Paleozoic strata of the mountains yield any such material, while the eastern derivation of the granitic and gneissic drift is further evidenced by its connected spread across the plains to the region of its supply. Thus the western limit of such characteristic erratics clearly indicates the extent of the drift from the Laurentian plateau. In regard to this western limit, it then was observed that it practically reaches the base of the Rocky motintains near the forty-ninth parallel, where Laurentian boulders were found at a height of 4,200 feet. Some 30 miles to the northwest and within a few miles of the mountains similar erratics were found at the mill on Mill creek (3,800 feet), and one was seen near Gar- nett’s ranch (4,200 feet). It was added: ““T did not, however, observe any Laurentian drift on the North fork of the Old- man, and it is probable that it is absent or nearly so in the district sheltered by the higher parts of the Porcupine hills. On the Bow river no Laurentian or Huronian erratics were seen west of Calgary, and even after their first appearance they were very scarce for some distance” (to the eastward). The elevation of the Bow at Calgary is 3,393.6 feet,* and in comparing this with that of the more southern localities the conclusion was drawn that ‘‘the western limit of the Lau- rentian drift cannot conform strictly to any contour line of the present surface of the country.” The later investigations tend somewhat to modify the above state- ments in showing that Laurentian drift does occur in a scanty and * This and some other elevations given here are derived from the results of the irrigation sur- vey or from railway surveys. Most of the heights are less precise, depending on barometric ob- servations reduced by comparison with Calgary. All may, however, be accepted within maximum limits of error (+) of 20 feet, and are sufficiently exact for all purposes of the present paper. 38 G.M.DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. sporadic manner behind the Porcupine hills, and also by the discovery of such erratics on hills of some height above the Bow river at Calgary, although that place still remains the western limit in so far as the valley of the Bow is concerned. The elevations just mentioned were not, however, the highest at which Laurentian erratics were found previous to the publication of the report of 1882-’84, for— ‘In 1883 several indubitable Laurentian boulders, representing three varieties of granitic and gneissic rocks, were found about 20 miles north of the forty-ninth parallel, at an elevation of 5,280 feet.’’ These boulders occur stranded upon a morainic ridge, due to local glaciers of the adjacent mountains. On a plateau to the south of the Porcupine hills Laurentian stones were found, though not abundantly, at a height of 4,390 feet, while similar erratics were observed to be scat- tered over the high country near Milk river at a distance of from 30 to 40 miles from the mountains and at an elevation of 4,200 feet. The obser- vations since made in the Porcupine hills enable considerable additions to be made to our previous knowledge of the maximum height attained by such eastern drift near the Rocky mountains. Digressing for a moment to places farther from the eastern base of the mountains, it will be useful to remember that on West butte of the Sweet Grass hills, 90 miles east of the mountains,* Laurentian fragments were found to a height of 4,660 feet, while according to Mr McConnell the drift of this origin finds its limiting height on the Cypress hills 200 miles from the mountains, at 4,400 feet.— Both the places last mentioned are not far from the forty-ninth parallel; but much farther to the north, in the Hand hills (latitude 51° 25’, longitude 112° 20’), Mr J. B. Tyrrell has found a similar upper limit for Laurentian boulders at 3,400 feet.t These ob- servations are cited here for purposes of comparison. In the report of 1882-84 it was stated that a similar limit occurred on the Rocky Spring ridge of northern Montana, 10 miles south of the bound- ary lineand 66 miles from the mountains, at 4,100 feet. The plateau only slightly exceeds this height, and, while convinced of the accuracy of the observation at the time, its wide discrepancy from other results may per- haps be regarded as leaving it open to suspicion. I have not had an op- portunity since of verifying it. Before dealing with the facts ascertained in 1894, it should be noted that Mr McConnell had in 1890 carefully examined the sections of the glacial deposits along Bow river between the mountains and Gleichen (about 80 * In this and other cases, unless otherwise noted, distances from the mountains are measured at right angles from the nearest part of the base of the range. + Op. cit.,. p. 75 C. t Annual Report, Geol. Survey of Canada, vol. ii (n. s.), p. 145 E. GRAVELS AND BOULDER-CLAYS. 39 miles eastward), and there found reason to believe that the Saskatchewan gravels of the plains represent and gradually pass into a “ western boulder- clay ’ inapproaching the mountains. This observation has remained un- published, but now appears to be-well established, and it follows from it, taken in connection with the facts already summarized, that there are no less than three distinct boulder-clays in the region here treated of, the old- est or “ western ” boulder-clay being followed in time by that previously named the “lower” boulder-clay, which is in turn distinctly separated from the “‘ upper” boulder-clay over a considerable part of the district, at least, by interglacial deposits. The western boulder-clay, asits name im- plies, contains no Laurentian or Huronian material, while such material, as well as limestone derived from the Winnipeg basin, is present in both the others. This general statement will serve as a clue to many of the observations subsequently detailed. In further presenting the results of recent observations attention will first be given to the sections found on the Belly and Oldman rivers, to the surface of the plains in their vicinity, and to the wide low area which is occupied by the tributaries of the Oldman in the neighborhood of the mountains. SECTIONS IN THE VALLEYS OF OLDMAN AND BELLY RIVERS. Although in the report of 1882-’84 the occurrence of two boulder-clays with an interglacial deposit was noted at Coal Banks (now Lethbridge) and a photograph was reproduced showing these deposits there running for miles along the bluffs of the river valley, no detailed section was re- corded for this place. A careful examination was made of this section in 1894, at a place about four miles north of Lethbridge, with the follow- ing result: The valley of the river at this place is cut down about 300 feet into the prairie. From 50 to 80 feet above the water level is occu- pied by dark shales of the Pierre formation of the Cretaceous, resting upon which, along a perfectly even line, are the Saskatchewan gravels or “quartzite drift,” with a thickness from 10 to 15 feet. The upper part of the shales, to a depth of two feet, is weathered and brownish in color. The gravels are coarse, often containing stones up to six inches in diam- eter, all well smoothed and water-worn, but often not perfectly rounded. They are generally arranged in a rather tumultuous manner; that is, not in regular layers graded according to size, and with the pebbles some- times standing on end. The interspaces are filled with a coarse gray sand, and a similar material forms occasional discontinuous layers a few inches or feet in thickness on the upper surface of the gravels. The stones are chiefly characteristic Rocky Mountain quartzites, but a con- siderable number of pebbles of limestone from the same source are in- eluded, as well as a few examples of the peculiar Rocky Mountain VI—Butt. Grou, Soc. Am., Vou. 7, 1895, 40 G.M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. ‘“‘oreenstone,” about which some remarks are made subsequently. One or two pebbles of peculiar crystalline rocks, not Laurentian, and prob- ably from intrusive masses in the mountains, were also found. The Saskatchewan gravels are sharply cut off above by a dark boulder- clay, the color of which is evidently due to the incorporation of a consid- erable proportion of the material of the Pierre shales and in which rather numerous crumbs of the coal of the vicinity are contained. The included stones are varied in origin, embracing quartzites from the mountains, Laurentian gneisses and some limestone of mountain origin, all often distinctly striated and glaciated. ‘The thickness of this boulder-clay is about 50 feet. Next in ascending order is a thickness, from 25 to 30 feet, of pale colored silty beds. often very finely stratified and in certain layers assum- ing a “leathery ” character and showing layers of almost paper-like fine- ness. Crumbs of coal are present, but no lignite or peaty layer is here seen. This well bedded intercalation preserves its place and character for miles along the valley and is continuous with that previously de- scribed lower down the river.* Overlying the last is the “ upper” boulder-clay, yellowish gray in color, and this, so far as can be ascertained, extends nearly or quite to the top of the bank or the general level of the adjacent prairie. Stones and boulders are not notably abundant in it at this place, but those which secur came both from the mountains on the west and the Laurentian plateau on the east. Summarizing this section and placing it in relation to others described in the report of 1882-’84, we obtain the following representation of the drift deposits of this part of the plains, the section on the right being that farthest from the base of the mountains: Sections on Belly River. Near Lethbridge. Driftwood bend. Wolf island. Distance from mountains, 60 || Distance from mountains, 75 || Distance from mountains, 85 miles. miles. miles. Height of base of section, 2,655 || Height of base of section (ap- || Height of base of section (ap- feet. proximate), 2,360 feet. proximate), 2,270 feet. Feet. Feet. Feet. i ps = : : ‘ : : : : : ; : : : : : : : : . . : 4 : P : S 8 . : : : ‘ _ = S =) Upper boulder-elay (@hOUt). TAO Fl cek ic ecescepeewnssonaainanseasdenteetnonth (Sands, ironstone, carbon- Interglacial deposits........... 30 ACEOUS 1AYVCVS)..0.secccencvsenss 33 || (Sandy clay, with lignite)... 8 Lower boulder-clay............ BU ll weseassoa dst Gcumnwewsud@erencrneenvesaasmete SU) i)|| Seawersse sdeesapsceneoecnstescedeeacees eee 15 Saskatchewan gravels........ 15 (Below river level.) (Gravels, sands and clay)... 49 Pierre (Cretaceous) shales.. 65 (Cretaceous rocks)....... ..... 10 “300 “158 “173 * Report of Progress, Geol. Survey of Canada, 1882-’84, p. 144 C. SURFACE MATERIALS. Al Before continuing the notes made in the deeper river sections to the westward of Lethbridge, a few words may be devoted to the character of the general surface of the plain corresponding to the sections above cited. This is well shown in numerous fresh cuttings along the line of railway between Dunmore (near Medicine Hat) and Lethbridge, a distance from east to west of 100 miles. Whether in the rolling prairie toward the east or the nearly level prairie to the west, the surface is almost uniformly composed of gray or brownish gray silty or loamy material, of which the depth may be stated to vary from two to five feet, although cer- tainly greater in some places. On the crests of knolls and ridges and in some of the valleys which have evidently: been cut out by postglacial flows of water, this deposit has been removed, leaving a grayish boulder- clay, which sometimes contains large stones at the surface. The stones are generally Laurentian, but are seldom abundant. It might be sup- posed that the prolonged action of rains or that even of the winds would in time produce a surface deposit of this kind, but much of the plain is so entirely flat that such explanations appear improbable. Neither are the projecting ridges notably bouldery, as should be the case if much denudation of their finer material had occurred, and the circumstances favor a belief that the silty deposits have been laid down in a body of rather shallow water, coextensive with the plain itself, in which some slight rearrangement of the exposed parts of the boulder-clay has taken place. There is some appearance of rolled gravelly deposits about the slopes of the ridges, but the cuttings are insufficient to show these fully. Following the axis of the main depression already alluded to, no ex- posures have been found further to the westward in which the lower and an upper boulder-clay are clearly distinguished, and as the sections are not continuous, it becomes impossible to decide in each case which is represented. In an exposure nearly opposite Rye Grass flat, 12 miles west of Lethbridge (52 miles from the base of the mountains), locally upturned Laramie beds are overlain by 10 feet of stratified sand and silt, followed by 20 feet of boulder-clay, which again is followed by 12 feet of rolled gravels, apparently replaced in a short distance horizontally by stratified sands. The whole section is capped by some feet of the loamy superficial silts above described. The boulder-clay seen in this section includes a number of discontinuous layers of sand and gravel. Another section of considerable length two miles and a half below Macleod (45 miles from the base of the mountains, elevation 3,024 feet) was carefully examined by Mr McConnell, and is described by him as follows: “The boulder-clay is here 45 feet in thickness from the river level and is overlain by 20 feet of sands and silts which contain layers of finely 42 G. M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. foliated leathery clays. The lower part of the boulder-clay is darker in color than the upper, but there is no division into upper and lower members, as dark and light layers alternate and change in color when followed along the bank. Stones both of western and eastern origin occur throughout, the former preponderating toward the bottom and the latter toward the top. The mass of the boulder-clay is in some places hard and clayey, in others soft and sandy, that of the last mentioned character passing occasionally into layers of sand and gravel.” The stratified sands, silts and leathery clays or shales of the above section much resemble the interglacial beds of Lethbridge, but, as already stated, there is here no means of certainly identifying the boulder-clay- Farther up along Oldman river, at the mouth of Beaver creek (28 miles from the mountains, elevation about 3,260 feet), a bank examined by Mr McConnell shows, above the river level, ‘* 50 feet of compact boulder- clay overlain by 6 feet of stratified silts and sands. ‘There is here a marked diminution in the proportion of eastern drift as compared with the last section, a rough estimate making it about two per cent of the whole.” In the same vicinity, on Oleson creek, about 400 feet above the river and to the north of it, a moderately indurated pale drab silty or sandy boulder-clay was found holding comparatively few stones, but some of them distinctly glaciated. Still further to the westward, at the confluence of the North and Middle forks of the Oldman (about 15 miles from the line of the base of the mountains, elevation approximately 3,650 feet), a good section was found, which may be set out as follows in descending order: Feet 1. Well rolled and rounded gravels, with some stones as much as 8 or 10 inches in diameter, apparently all of Rocky mountain origin..................- 10 2. Good typical boulder-clay, moderately indurated; matrix brownish yellow and earthy, containing glaciated stones and boulders of moderate size, mostly subangular, but some well rounded, derived from the mountains or from the Cretaceous rocks of the foothills, but chiefly quartzites ; some limestone and a few examples of greenstone. Two or three small pieces of Laurentian rocks were found which probably came from this boulder- CLG gsi otc owen hia 2, 9 bain o> prolate sw eae een ee ek 20 3. Stratified, earthy, brownish yellow sands, containing a few glaciated stones. 10 4. Obscurely stratified gravels, containing some stones 10 inches through, all well rounded and like beach or river shingle. Traces of glaciation were suspected on a few of these, but were in no case observed to be abso- lutely decisive. The line between this and the overlying deposit is quite regular and definite. Although there is an appearance of blending in a thickness of a few inches, there is no sign of any intervening condition Ol PPOLRNIGES o.oo) icine As bw pens mie oe ots etic eae ee ae ae 10 5. Laramie sandstones and ghales to river level... 2-0... 2 os ene ean 40 GEOLOGIC SECTIONS AND THEIR COMPOSITION. 43 Numbers 8 and 4 of this section are believed to represent the Sas- katchewan gravels, while number 2 may be either the lower or upper boulder-clay of the plains. Less than a mile to the northward the boul- der-clay was observed to rest directly upon the Laramie rocks, numbers 3 and 4 having run out. Number 4 has in some places a clayey matrix, thus beginning to assume the character of the “ western” boulder-clay. About two miles further north, along the North fork and well behind the southern part of the Porcupine hills (elevation about 3,900 feet), another section was examined, of which, however, the total thickness remained indetermined because of slides in the bank. This again shows boulder- clay of a somewhat earthy and soft character, but containing many stones, derived from the mountains or adjacent foothills. The limestone peb- bles are often distinctly but very lightly striated, and have apparently been well rounded by ordinary water action before the striation had been added. Two small crumbs of Laurentian material were discovered by search on the face of this exposure, but the decrease in importance of such material in the boulder-clay to the westward and where sheltered by the high ridges of the Porcupines is very apparent. The comparatively soft and earthy character of the boulder-clay seen behind the Porcupine hills was generally observable. Reverting to the main line of approach which we have been following toward the mountains, an exposure on the South fork of the Oldman, examined in 1883, may next be alluded to. This is distant from the moun- tains about 12 miles, with an approximate ele- vation of 38,700 feet. It again shows a boulder- clay, similar to the last, overlying a few feet of gravel derived from the mountains. Both de- Ficurr2—Section on the South posits occupy a hollow, possibly that of an old Fo ee ae river valley, as shown in the diagram annexed. ,4. “"™"™° ee aL In 1881 another section was noted on Mill 2=Saskatchewan gravels. creek, still nearer to the mountains (six miles ae ey Seas: distant, elevation 3,817 feet), which showed £=surface gravel. boulder-clay of the usual character underlain “=S% _ by a very hard boulder-clay or till of different Cae eres ea aspect, below which was a few feet in thickness of fine, compacted gravels. Some Laurentian stones were found on the surface in this vicinity above the level of thesection, but none were seen init. A similar instance of bouldery clay overlying thin layers of gravel was discovered in the same year high up on Pincher creek, in this neigh- borhood, within a couple of miles of the actual base of the mountains. The two last mentioned localities are within the limit of the country bats aS 44. G. M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. characterized by moraines, evidently due to local glaciers from the Rocky mountains, and the indurated boulder-clay of the Mill Creek section is believed, like the moraines, to be a deposit of these glaciers. The lower gravels in this case and in that of Pincher creek are obviously due to pre- glacial streams flowing from the mountains, and, although the name Sas- katchewan gravels may be applied to them, they here evidently antedate the eastern gravelly representative of the Rocky mountains or earliest boulder-clay. Further to the east, where this boulder-clay gradually passes into such gravels, there is no means of distinguishing between wholly preglacial beds and those which may have been formed during the main period of the Rocky Mountain glaciers. Many exposures of the Saskatchewan gravels may include both, and this without necessitating the supposition of any great chronologic break. SOUTHERN PART OF THE PORCUPINE HILLS. Having thus followed the main southern line of approach at low levels to the mountains, attention may next be given to the southern end of the Porcupine hills, which overlooks this avenue on the north side, at a dis- tance from about 15 to 30 miles from the base of the mountains. Oleson and Beaver creeks flow southward from this end of the hills, and it was chiefly in the vicinity of these streams that the observations noted were made. In traveling westward from Macleod (situated on the plains at an ele- vation of 3,070 feet) to Oleson creek by the regular trail north of Oldman river, a distance of 14 miles, a gradual ascent is made which becomes greater as the flanks of the hills are reached. ‘The following terrace- levels were noted on this route : North of Macleod an extensive gravel plain forming the angle between Oldman and Willow rivers is reached. This rises gradually from 3,130 feet in a distance of a couple of miles to 3.220 feet. Itssurface is not abso- lutely flat, but is diversified by low swells or ridges, which generally trend north and south. This plain is bounded to the west by a distinct rise leading to another similar plain or wide terrace, also gravelly, of which the eastern part is at a height of 3,275 feet, and which continues to slope up gradually to the west- ward. The gravels of this plain and the last are composed chiefly, but not entirely, of well rolled Rocky Mountain quartzites. At 3,286 feet on this second plain is found running northward a line of remarkable large boulders,* composed of quartzite or conglomerate. These are identical * These remarkable boulders are in size and composition unlike any observed in the boulder- clays. They have undoubtedly been water-borne and may probably have been derived from some particular region of the Laurentian plateau which became tributary at a later stage of the Glacial period. SECTION ALONG VALLEY OF OLDMAN RIVER. ‘payetessexe A[TWIOJIUN jou pue AjyeaiS Aida St oTVoS TBOTJSA oY, “SIUT]T UsyoIq UE UMOYS o1e MWOTO9S IY} JO Y}IOU Jy} O} ST[IYJOOF pue STII autdnoiod oy], “neozeid ueryuoime’y oy} wor [ettoyem YLIp ‘sjop {SUIeJUNOU sy} WOT [VIIoJeU YLIp DJESIPUT SOSSOI) “}SAM 9} 0} JSOY}IEF Suipusyxe yey} st sureyd ay} jo Avjo-1aptnoq saddn ay} yey} stsoyjyodAy oy} WO peseq ST WIeIseIp SIU, ‘s]]IY aurdno10g a4} UO YLp pue Ssde119} = 7 ‘Avpo-laptnoq ,,jeddn,, =@T ‘sjisodap [elpr[s19yUI =D “AeO-1appnoq ,,JOMOT,, = F “‘S[QAVIS UBMIYOJLASES O}UL pIVA}SvI OY} OF Sutssed Avjo-Jopnoq UslosoM : (a8e}s Ueyeqiy) = V ‘ani uvupio fo hay]vA 244 SUo]D ‘suzvzUnopy AyI0y JO ASV 07 A5p11QYIIT motf U01ZIAS IIDMUMDASDI—E AANOIA aua < DA © 80 Kon Oex® = x ee eeX eo 8 ge ; 2 tk Xo Se oo eee y 208 Yokoo hee eee ‘sae: soo, note" aug S ie Sd oe puny saddy 2° % Pe cree Na SIH 3001 FIGISFHLIT----° COFTIVW ~~~ 46 G.M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. in character with those noted in the report of 1882-’84 as occurring near the lower part of Waterton river at a height of between 3,200 and 3,300 feet.* and it may be added here that boulders of the same kind were found by Mr McConnell on the northern part of the Porcupine hills at a height of 3,950 feet, and on the Nose hills near Calgary at 3,940 feet. At 3,516 feet is a boulder-strewn terrace with some pretty large boul- ders, both of Rocky mountain and Laurentian origin; at 3,387 feet, another terrace similarly characterized ; at 3,532 feet, a terrace with rolled eravel on the surface and an abundance of eastern drift, and again at 3,648 feet occurs still another well marked and wide terrace with similar mixed drift. From this a descent was made to our camp on Oleson creek (8,600 feet) and from this place, in the course of a rather long excursion in the hills to the northward, the following terrace-levels at greater altitudes were ob- served. ‘These are briefly enumerated below, but it must be understood that many more such levels might have been noted had further time been given to the investigation. Possibly, at a distance of some miles, a quite different series of water-levels would have been recognized, for it appears probable that almost every stage in a gradual descent of the water-line may be found to be marked in some part of the Porcupine hills. 3,853 feet, a terrace-like flat with rolled quartzite and Laurentian gravel. 3,877 feet, an evident terrace with similar gravels, including some Rocky mountain limestone. 3,898 feet, a faintly impressed terrace with similar gravels. 4,182 feet, approximately, a terrace with similar gravels. 4,281 feet, a terrace with similar gravels. 4,349 feet, a terrace with similar gravels, many large well rounded stones, and a considerable proportion of limestone referable to the Winnipeg basin. 4,505 feet, a flat-topped hill, the highest in this vicinity, and evidently marking a terrace-level, covered with similar well rolled gravels, including Rocky Mountain quartzites and limestone, as well as Laurentian gneisses and Winnipeg limestones. It is thus evident that from the level of Macleod to the highest point above noted there is an uninterrupted series of terraces, covered with well rounded pebbles of mixed eastern and western origin. The erratics of eastern origin are not less abundant at higher than at lower levels, and while some of the Rocky Mountain stones are of considerable size, the eneissic Laurentian boulders are, on the whole, larger at high levels, being often as much as three feet in diameter, while some large pieces of Win- nipeg limestone were also seen at the highest levels. No glaciated stones were observed on these higher terraces, nor any signs of glaciation on the * Op. cit. 14 pp. 8, C, 149 C. SOUTHERN PART OF THE PORCUPINES. A7 sandstone outcrops where these occur, but the rock in place is rather too soft to preserve such traces well had they existed upon it. The peculiar ereenstone of the Rocky mountains before referred to is not infrequent at all levels, and as this particular rock occurs in place in the mountains (as an interbedded layer) scarcely as far north as latitude 49° 30’, it must have traveled in a northeastward direction in order to reach this part of the Porcupine hills. The matrix of the gravels, wherever seen, is a whitish silty or sandy material, perhaps in part composed of disintegrated sand- stones of local origin, but including grains of similar composition to the pebbles themselves. The flat outlines of the hills in all this southeastern part of the Porcu- pines appears to be in the main plainly due to water-levelling, although assisted by the practically horizontal attitude of the sandstone beds. From the highest point here reached the terracing of the hills may be finely seen for many miles to the northward, but still higher and _ partly wooded ridges to the westward showed toward their summits an alto- gether different and rougher character, although fundamentally composed of the same Laramie rocks. The highest terrace seen on the hills, near the headwaters of Beaver creek, was very well marked, and was estimated by eye from a distance to reach about 4,900 feet above sealevel. In continuing the inquiry it became evidently necessary to examine the higher ridges above alluded to, and this was accomplished from the upper valley of Beaver creek, whence an ascent was made to the highest point in that vicinity, locally known as Five-mile butte. In this region the'total amount of foreign drift is less considerable and distinct terraces are seldom observable, facts doubtless due to the shelter afforded by adja- cent highlands on all sides, but particularly to that of the wide belt of hills and ridges to the eastward. Our camp on Beaver creek was at an elevation of 4,222 feet, and in ascending from it to Five-mile butte, on the east side of the valley, the following notes were made: 4,950 feet, a few well rolled pieces of Laurentian, Winnipeg limestone and Rocky Mountain quartzites. 5,070 feet, a few small Laurentian pebbles. 5,144 feet, Laurentian boulders 2 feet 6 inches through, Rocky Mountain limestone, quartzite drift and probably a little Winnipeg limestone. 5,250 feet, a projecting point on the high ridge showing abundance of well rounded Laurentian and quartzite drift. At 5,300 feet the ridge becomes flat-topped and probably marks a terrace-level. It is strewn with numerous well rolled pebbles of eastern and western origin, including Laurentian, Winnipeg limestone, and Rocky Mountain limestone and quartzite. Some of the Laurentian boulders are 2 feet in diameter. Above this level nothing but debris of local sandstones was found, the highest point of Five-mile butte being reached at 5,365 feet. VII—Butt. Grot. Soc. Am., Vou. 7, 1895. 48 G.M.DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. It will be noted that the Laurentian drift is in this neighborhood markedly more abundant at the higher levels, the upper limit of the traveled material standing above all the hills and ridges to the eastward. A distinct terrace was observed on the opposite (west) side of Beaver Creek valley at an estimated height of about 5,130 feet. This may pos- sibly correspond with that previously noted as seen from the hills above Oleson creek, but is not the same. The levels in both cases are neces- sarily somewhat uncertain. In crossing the last ridge of the Porcupines on the west, between Beaver creek and the North fork of Oldman river, a height of 4,986 feet was reached, and here a few pebbles of Rocky mountain origin were found, although on projecting points 200 feet higher no traveled drift was ob- served. This evidence is, however, of a purely negative character. On the west slope, in descending toward the North fork, Laurentian drift was first recognized at 4,710 feet and continued sparingly down to about 4,060 feet. None was seen near the river itself (3,960 feet). PLAIN AND VALLEY WEST OF THE PORCUPINE HILLS. Between the Porcupines and the foothills proper a plain some miles in width here runs north and south. This to the eye appears almost perfectly level. It is continued southward beyond the Middle and South forks of the Oldman with increasing width and probably with a some- what decreasing elevation. The lowest part of this plain actually trav- ersed on our route is near the confluence of the North and Middle forks, with an elevation of 3,750 feet. In about three miles farther north it rises gradually to 4,140 feet, the surface being generally gravelly (num- ber 1 of section on page 42). This gravel plain resembles in character that occurring near Macleod at an elevation lower by about 1,000 feet, but no eastern drift was found among the pebbles, which appear to have been entirely brought down by rivers flowing from the mountains. In following the plain northward it becomes narrowed, but again widens about the bend of the North fork, where its average elevation is about 4,200 feet. From this vicinity (near the Upper Walrond ranch) the wide valley of North fork runs northwestward to the base of the mountains. It is floored by a regular terrace, apparently in continuation of the plain last referred to, which attaches to the bases of the neighboring hills some miles to the west at an elevation of about 4,400 feet. From the Upper Walrond ranch a continuous valley, bounding the Porcupine hills on the west, runs northward to Highwood river, a distance of 48 miles. A very few small Laurentian boulders were seen near the ranch, and one was observed about a mile and a half to the north at a BULL. GEOL. SOC. AM. VOE7, 1895, PEa de. FIGURE 1I.—BLUFF ON HIGHWOOD RIVER. Showing two boulder-clays, the higher passing above into stratified silts. ‘The base of the boulder-clay here rests directly on Laramie rocks. FIGURE 2.—PART OF SECTION ON BoW RIVER NEAR CALGARY. Showing clayey Saskatchewan gravels overlain by boulder-clay. BOULDER-CLAYS AND SASKATCHEWAN GRAVELS. PLAIN AND VALLEY WEST OF THE PORCUPINE HILLS. 49 height of 4,400 feet; but no more eastern drift of any kind was found along the valley for 30 miles northward. If not entirely absent, it must here be extremely scarce. At the distance just noted, near the chain of small lakes between the North branch of Willow creek and the South branch of the Highwood, where the wide gap of the Highwood valley begins to lay the country traversed more open to the eastward, a single Laurentian boulder was again seen. This was opposite the third or northernmost lake, at an elevation of 4,406 feet. In this vicinity a well marked terrace was also found at 4,270 feet, with several others faintly impressed on the hillsides up to 4,500 feet, but no higher. The upward limit of terracing and of thick drift deposits ap- pears here to be well defined. Large fragments of Rocky Mountain lime- stone are found here and there throughout this part of the foothills gen- erally stranded on prominent ridges of sandstone. At the head of the South branch of the Highwood, brownish earthy boulder-clay, with stones wholly derived from the mountains, was seen in the bank of a stream apparently resting directly on bed-rock. The surface of this boulder-clay forms a wide terrace-level in which the stream valley is cut out, with an elevation of 4,240 feet, rising to about 4,290 feet where it meets the siopes of the hills. In following the South branch northward to a point six miles from its confluence with the main High- wood, at a height of 3,960 feet, boulder-clay like the last was again seen, but here holding a few very small Laurentian fragments. HicHwoop RIVER AND VICINITY. To the eastward of the South branch Mr McConnell made a long detour among the northern ridges and plateaus of the Porcupine hills, the high- est of which are there about 4,740 feet. Upon these he found abundance of Rocky Mountain limestone and quartzite, but no eastern drift above 4,150 feet and very little drift of this origin anywhere. In the bank of the main Highwood, four miles above the mouth of the South fork (13 miles from the base of the mountains, elevation about 3,700 feet), Mr McConnell examined a section showing 35 feet of boulder- ' clay overlain by a considerable thickness of silts, and these in turn capped by river gravels. The boulder-clay is dark brownish below and hght yel- lowish above, with stones seldom exceeding six inches in diameter, which, so far as observed, are wholly of western origin. From the mouth of the South branch the Highwood was followed down to the crossing of the railway, and midway between these points some fine sections were found (see plate 1). The height of the river is here 50 G.M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. about 3,500 feet and the distance from the mountains 19 or 20 miles. In descending order, the bluffs here show— Feet iL. Well:stratified and: éurrent-bedded silte.2 3)... oc oe) koe ote eee 5, 2. "Pale yellowish pray boulder-clay, + phonies ets on | sian bre ce hic ae ee 15 5. Dark-eray ‘boulder-clay sn). <5. ae ees ae orien «Deh ees 20 qd. Laramie sandstones and shales *.\/.. os es eek. Coe wie e's ob bee 15 55 Both parts of the boulder-clay hold many and some large stones, often well glaciated and apparently all of western origin. The line between the two layers of boulder-clay is horizontal and quite distinct. Many of the larger stones occur about this level, and one of them was seen to lie about half in the lower and halfin the upper division. It is not certainly known that the division between two classes of boulder-clay found in this and the preceding section corresponds with the horizon of the inter- glacial deposits previously described, but it is believed that numbers 1 and 2 correspond with numbers 2 and 3 of the Calgary section (see page 53). A very few Laurentian fragments were seen in traveling from this place eastward to the town of High River, at the railway crossing (3,371 feet). They appeared to be more abundant to the east. HicHwoop RIvER To CALGARY. From the town of High River the regular road was followed north- ward to Calgary, 33 miles, crossing Sheep, Pine and Fish creeks and rising over eastward projections of the lower plateau, which here repre- sents the Porcupine hills. The highest point reached between High- wood river and Sheep creek is about 3,623 feet. Here less than one- hundredth of the drift stones are Laurentian, the rest being from the mountains. At 3,495 feet, on the northern descent toward Sheep creek, perhaps one-fiftieth of the stones are Laurentian, but at a corresponding elevation on the southern slope toward the Highwood such stones are exceedingly scarce. At the crossing of Sheep creek (about 3,400 feet) a partially stratified stony deposit, resembling boulder-clay but showing no striated stones, contains a considerable proportion of Laurentian frag- ments. Between Sheep and Pine creeks, beginning to the south at about 3,600 feet, rising to 3,790 feet and falling again toward Pine creek to 5,500 feet, is a lumpy, undulating country, comprising some hollows and swampy depressions without outlet, and repeating somewhat the characters of the Missouri Coteau on a much reduced scale. The surface is pretty thickly covered with soil, which is seen in places to be underlain by de- HIGHWOOD RIVER TO CALGARY. 51 posits of rolled gravel, but no sections of any depth occur. The extent of this country where crossed is about six miles. It is the only tract met with in this entire region which in any degree simulates the characters usually assumed as morainic. Nearly all the stones are from the west, but a very few Laurentian boulders are seen. At the north end of the railway bridge over Fish creek a cutting has been made in pale grayish yellow boulder-clay, in which most of the stones are well rounded (though some pieces of Rocky Mountain lime- stone are striated) and all are of western origin. Laurentian boulders are here, however, not uncommon on the surtace at elevations of 3,400 to 3,500 feet. The higher parts of a wide plain, through the center of which the Bow valley is trenched, in the vicinity of Pine creek, have a level of about 3,000 feet. Between Fish creek and Calgary, at heights of 3,400 to 3,500 feet, Lau- rentian boulders are found in increasing numbers. Some of them are several feet in diameter, and they are scattered over the surface appar- ently in association with deposits overlying the boulder-clay. Sections IN Bow RIvER VALLEY. At Calgary we reach Bow river, which has in the introductory pages of this paper been described as the second great avenue of approach to the mountains at low levels and the northernmost in the region here con- sidered. Following the plan already adopted in the case of the Belly and Oldman rivers, some notice will now be given of observations made along the Bow from east to west, or in order, ascending the stream toward the mountains. These observations are chiefly those of Mr McConnell, who in 1890 descended the river ina boat from Morley to the Blackfoot crossing with the special purpose of investigating the superficial deposits, and supplemented this by a critical examination of these deposits at Medicine Hat. Medicine Hat is situated at a distance of about 155 miles from the nearest part of the mountains and about 270 miles from the mountains by a line measured along the general course of the Bow and South Saskatchewan rivers. Mr McConnell writes: “The glacial deposits at Medicine Hat consist of light colored com- pact boulder-clays of the ordinary type, but showing in places faint lines of stratification, overlain by stratified sands and underlain by beds of quartzite pebbles, occasionally cemented into a conglomerate and some- times associated with sands and silts. ‘The line between the material derived from the east and that coming from the west is here drawn at the base of the boulder-clay ; above that 52 G.M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. horizon eastern gneissic and limestone boulders and pebbles, the latter often striated, are common, but no rocks of undoubted western origin were observed. The beds of well rounded quartzite pebbles below the boulder-clay, on the other hand, are derived, so far as known, entirely from the west, although they may here in part represent redistributed Miocene conglomerates like those of the Cypress hills, which were brought down from the mountains in Miocene times. “Twenty miles above the Blackfoot crossing or 175 miles above Medi- cine Hat, where the next section was examined, the conditions have entirely changed. At this particular place the underlying gravels are absent and the boulder-clay holds both eastern and western drift inti- mately commingled throughout, pebbles of unmistakable Laurentian eneisses and well characterized Rocky Mountain limestones often lying side by side inthe same hand specimen. The relative proportions of the two drifts at this point, 100 miles east of the mountains, measuring along the valley of the Bow, are nearly equal. In descending the river western drift of a recognizable character gives out in the boulder-clay before Medi- cine Hat is reached,and in ascending it the eastern drift gradually dimin- ishes in relative quantity and disappears altogether above Calgary, 40 miles east of the mountains, or about 50 miles if the Bow valley be fol- lowed. “Twenty-five miles above the Blackfoot crossing a boulder-clay section 110 feet in thickness is exposed. The boulder-clay is here separated into an upper and lower division by a band of stratified sands, the lower boulder-clay being darker colored than the upper one and differing from it also in containing a larger proportion of western drift. The junction between the two boulder-clays is not straight, but follows an irregular wavy line. ‘“At Pine canyon, eight miles above the last section, the Laramie sand- stones are overlain by the Saskatchewan gravels 10 feet thick, above which is a peculiar morainic-looking deposit 40 feet thick, consisting of: angular blocks of Laramie sandstone of local origin, gneisses and lime- stones from the east and limestones and quartzites from the west, ail mixed confusedly together in a matrix of coarse sand and clay. “ Four miles above the last exposure the boulder-clay, here 50 feet thick, rests directly on the older rocks. The ratio of eastern to western drift in this exposure was estimated at about 1 or 2. A notable feature of the section is the presence in it of a gneissic boulder of eastern origin measuring fully three feet in diameter. The ordinary size of the eastern pebbles in the boulder-clay along this portion of the river seldom exceeds three inches in diameter. “Two miles above the mouth of Highwood river the Saskatchewan —— a SECTIONS IN BOW RIVER VALLEY. 53 gravels appear again. They consist mostly of rounded quartzite pebbles and boulders, ranging in size from one to twelve inches in diameter, and have a thickness of eight feet. The pebbles increase in size toward the base of the formation. The boulder-clay above the gravel holds occa- sional gneissic pebbles, but they are small and scarce. ‘Two miles above the last exposure the pebble bed passes into dark clays filled with stones of western origin only. Above this is 170 feet of boulder-clay, alternating in places with sandy layers. A mile below the mouth of Fish creek the gravels reappear, but are replaced a mile above Fish creek by stratified sands. Two miles farther on the sands pass into gravels again, and these continue to underlie the boulder-clay as far as Calgary. West of Highwood river the western gravels underlying the boulder-clay consist of limestone and quartzite, the proportion and size of the former increasing as the mountains are approached, but east of Highwood river they are composed almost entirely of quartzite. The eradual diminution in size of the limestone pebbles and their final dis- appearance to the east, while the quartzite constituents still continue, is no doubt due to their inferior hardness and consequent inability to stand the wear attendant on a lengthy journey under the conditions in which it was accomplished. | “The Saskatchewan gravels and associated sands and clays in the neighborhood of Fish creek are everywhere overlain by boulder-clays holding scratched limestone and quartzite pebbles and boulders from the west, and at rare intervals small gneissic pebbles from the east.” In my own descent of Bow river, in 1881, attention was chiefly devoted to the underlying rocks, but to the above description by Mr McConnell it may be added that the existence of the Saskatchewan gravels, though obscured by slides, was suspected at some places below the Blackfoot crossing.* Above the crossing these gravels appear sometimes at the water-level and sometimes at heights from 20 to 60 feet above it, but it is probable that if carefully looked for they might be recognized at short intervals all the way down to Medicine Hat. At Calgary, on the north side of Bow river about a mile below the bridge, a very instructive and clear section occurs. This had been exam- ined by Mr McConnell in 1890, and was in 1894 carefully reéxamined by that gentleman and myself. It shows in descending order : ¢ Feet meberceular deposits of gravel and silty soil... 2... .....00.050.0cbee eset ee ceed 5 2. Stratified silts, with some lenticular layers of boulder-clay ; striated stones Pe Stiet le OOM ErS MMe WOtMermn Feat eee uci alelsietereisls ceisiale cnet! toa bd alee beak 35 * Report of Progress, Geol. Survey of Canada, 1882-84, pp. 141 C, 142 C. } Elevation of base of section, 3,390 feet. 54 G.M. DAWSON—GLACIAL DEPOSITS OF SOUTHWESTERN ALBERTA. Feet 3. Boulder-clay, with some stratified silty layers and pebbles arranged in lines OL StrabiMCa LOM: 0/2 tn v1 be: «a eles SAN i te eats eh Sake Se oe 20 AECRTA OLS t's ode El sieusne's 55a Sie a a eeaiehc etme aie CR eT a Se ance ata ee ee 15 5. Laramie sandstones and shales, nearly horizontal ...........0.0.02 sees ees 25 100 The following details, written down at the time, further explain what is seen in this interesting section. The order fol- £ lowed is that of deposition, beginning with the base of the section: The surface of the Laramie rocks where composed of fairly hard sandstones is Dp smooth and waterworn without any glacial strie. aay : Resting directly upon this are rather incoherent gravels with a considerable admixture of clayey or silty matter. All the stones are derived from the mountains, and most of them are quartzites (some 18 inches through), but Rocky Mountain limestone is also abundant. Nearly all are well rolled and rounded, but careful search shows traces of striation on some of the limestone pebbles. These appear to have been produced upon the already rounded stones and to have been largely obliterated afterwards by further wear. There is A little or no trace of stratification in the gravels, which resemble more the deposit found in the bars or bed of some river than anything else. PROD iy Re Slag Ae The gravels are cut off above sharply on a nearly of Bow River near Calgary, level plane, above which is a hard yellowish gray eer w i Ras oe nile, il te A = Laramie rocks. boulder-clay, often standing vertical in the face B = Saskatchewan gravels. and breaking out in prismatic fragments. This con- C = boulder-clay with silty 5 ‘ slew iis nc bo 5 nib a's oi vie wi veibin ee obese ceeaias 105 SUNUEUREG! SVC Ee NGS Oe ee re ey 106 “OUTST DG SSCS BCT eee rr a 107 eeemeoE omietits c-Si tte hy Cars) | eM Ge des esas hs sess celesatseee neces 107 Piipsoidal syenite-eneiss or leopard rock. .:...........00-.2000000eceees 114 LTE) S By ESE REESE ee Cec rr ee 117 Piemagemand relaiions Of tie TOCKS. . 0... jcqahisnscsceeccceescescscecessswes 119 (ULSD DUDE 2 ae aed Ror cy rrr 120 Paton ieelipsoidal SITUCHINE: ¢. Unban Beas ees le cek cece e beans ee enescaae’s 123 (PST ENE] INO ASSES ue a i IBA a Se gen a 123 Eeean Tea te MONON REE Er OUTS 51 ).Grih vai aha teil ha Louie uals iaydc ewe ode wo ed dee w cle’ 124 Winterentiaiion of the Coarse SyEMILEL b. .. 6) c cc ke eee ee ee eee oan 124 Veierennainonor toe ellipsoidal rock... 2... i. ose se ee ee ee ees 125 * Condensation of thesis presented for degree of Doctor of Philosophy, University of Chicago, June 13, 1895. The author is indebted to Dr A. R. C. Selwyn and his associates on the Canadian Geological Survey and to Mr A. P. Twidale, superintendent of the High Rock phosphate mine, for numerous courtesies and favors. The analyses were made by Professor William Hoskins, of the Chicago Academy of Sciences. Especial acknowledgments are made to Professor J. P. Iddings for his keen, critical analysis and painstaking assistance. Without his aid the paper would lack much of any value it may possess. XIV—Buut. Geon. Soc. Am., Vou. 7, 1895. (95) 96 Cc. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. , Page SECOndary OKI). 6.0.25 vie ee eee ee eee co ale cle rey ee 129 Similar structures in other rocks. saeco eee con ors 6 koe eee one eee 183 Conclusion. © s< f.5's So Sid ees 4 hihi be ee ere 130 INTRODUCTION. GENERAL DESCRIPTION OF THE REGION. The apatite region of Ottawa county comprises the area included be- tween the lower portions of the Du Lievre and Gatineau rivers. The chief mining districts occur in Portland and Templeton townships, but deposits of greater or less extent are found over nearly the whole of the area. The region lies upon the southern flank of the Laurentian axis, and is characterized in large part by asomewhat rugged topography. For some distance north of the Ottawa river, the surface is comparatively level, but this feature gradually gives place to hills which rise to a height of from 500 to 700 feet above the level of the adjacent rivers. The hills are covered with a meager soil, and the forest growth, originally limited, has been largely swept away by fires. The region is drained chiefly by the two rivers mentioned above, which flow southward into the Ottawa river. These streams are of considerable size, have swift currents, and rapids frequently occur. Waterfalls also constitute a picturesque feature of these streams. High falls, on the Du Lievre, has a descent of about 100 feet. The country between the streams is dotted with numerous lakes, which drain through small streams with tortuous courses into the Du Lievre or the Gatineau, or southward into the Ottawa. These lakes are extremely irregular in shape, with sharply sinuous shorelines, and often contain small islands. GEOLOGY OF THE REGION. Character and classification of the rocks ——In order to arrive at a proper understanding of the nature and occurrence of the ellipsoidal syenite- eneiss or leopard rock, it is necessary to introduce a brief description of the geology of the region. The facts upon which this description is based have been obtained chiefly from the reports of the Canadian Geological] and Natural History Survey.* * We are indebted chiefly to the reports of Vennor and Harrington, and to Professor F. D. Adams’ account of the typical Laurentian area in the Journal of Gevlogy, vol. i, No. 4, 1893, pp. 325-340, The map here given was taken from that accompanying H.G. Vennor’s report in the Annual Re- port for 1878.. ROCKS OF THE APATITE REGION. 97 In its geological structure the region consists of alternating bands of gneiss, crystalline limestones and pyroxenic rocks, in which are inter- stratified a number of zones of quartzites, rust-colored rock or fahlbands, and several horizons of magnetic iron ore. These rocks all belong to the Grenville series, the uppermost of the two main divisions into which the ‘Laurentian system is now quite generally divided. tent LENSE LESS ine A i wt el nye uth a = tt flan Yj We ( Wi atta vw aa ROUTAN tates PRS ne be hy Hy wy ag Mes eat WAT \' Kath Dal 3 Whe ay Uo AS ty Ik AA Lie I nt: ; Gin i wink ago. pa ron L sbestos. tiny donut ! 1 Wie Zo’ LEGEND. Le ESS2] Potsaam -Tre nt, ESS Limestone tt ZAG@neiss VY. \' E==eress W. SSNiqneiss I. el pr gated Un SG ES : Race (i Gueiss 5 Rpetite Region OF 1 OTTAWA COUNTY. CANADA. NY Toy FIGURE 1.—T7he Apatite Region of Ottawa County, Canada. The Grenville series includes rocks of very different petrographical development and of great variability in mineralogic composition. ‘In it are found all the mineral deposits of economic value—apatite, iron ore, asbestos, etcetera—which occur in the Laurentian.” * Apatite is found quite generally throughout the series, but the princi- pal workable deposits occur in a belt (see IV, figure 1) of rust-colored * Journal of Geology, vol. i, No. 4, p. 827. 98 ©. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. gneisses and pyroxenic and feldspar rocks above the Gatineau limestone band (IIT). As shown on the map, in the region between the Gatineau and Du Lievre rivers the rocks are arranged in the form of a great synclinal, with Big lake marking approximately the location of the axis. On passing eastward from the west line of Ottawa county, therefore, we pass over the different belts in ascending order until we reach the east line of Den- ham township. Lithological characters.—The gneisses interstratified with the limestones vary much in character, but the predominating variety consists of a more or less reddish orthoclase and grayish white quartz with little or no mica and sometimes with garnets. It is usually coarse or granitoid in struc- ture and the bedding often obscure, though in places it contains numer- ous beds or layers of quartzite from half an inch to a foot in thickness, which render the strike of the rock plainly visible. In some cases the mica is abundant and the gneiss then assumes a marked foliated char- acter. The micaceous gneisses are sometimes garnetiferous and occa- sionally exhibit the texture of the so-called augen-gneiss. Quartzites of considerable thickness occur now and then. They are often white and glassy and in places contain a little orthoclase. These strata are frequently traversed by dolerite dikes, some of which are of considerable thickness. The pyroxenic rocks associated with the apatite, and by Hunt * called pyroxenites, vary considerably in their characters. Sometimes they con- sist almost exclusively of pyroxene, though commonly quartz and ortho- clase are present. Mica and apatite are of frequent occurrence, and occasionally minute garnets may be seen. HIGH ROCK DISTRICT. General description—High Rock mine, the locality at which the ma- terial for this study was obtained, is located on a series of connected hills situated on the right bank of the Du Lievre river, about 21 miles above Buckingham. The openings cover in all about 600 acres on the tops of the hills which extend to a height of 700 feet above the level of the river. They are reached from the river by a tramway two miles long following the natural slope of the hill. The series of hills trend in a direc- tion south 30° east (magnetic). The openings are all in one wide belt of pyroxenic rock having a strike in the same direction. The main opening is number 11, the entrance to which is on the west side of the hill about 180 feet below the summit. The vein in which this pit is opened has been worked at several places along the side of the hill. Other veins * Geology of Canada, 1866, p. 185. Chemical and Geological Essays, p. 208. GEOLOGICAL STRUCTURE OF HIGH ROCK DISTRICT. 99 parallel with this occur along the top of the ridge and have been worked at various points. In all some 35 or 40 openings have been made on this property. Geological structure.—The hills are made up of quartzite, pyroxenite, and gneiss in belts whose direction corresponds with that of the ridge. The apatite occurs in veins or pockets in the pyroxenite. The quartzite occurs in beds standing in a nearly vertical position and with strike parallel to the general direction of the ridge. The pyroxenite is not distinctly banded, though occasionally parallel lines, which have sometimes been taken to represent lines of stratifica- tion, can be traced through it. Stratified and massive gneisses are reported * as often seen bordering the hills in the apatite region, but they were not made the subject of study at High Rock. The rocks all dip at high angles (nearly vertical), and are cut in vari- ous directions by small dikes. At several places on the hill bosses of feldspar rock appear protruding through the quartzites and pyroxenites, and expanding at.the surface. An instance of this is seen near the summit in front of the office. The feldspar is coarsely crystallized, of a lilac color, and is associated with a considerable amount of augite. In most of the openings, the apatite is associated with a reticulated feldspar rock, consisting of lumps of coarse feldspar (and a variable amount of quartz), separated from each other by thin anastomosing layers of green augite and a small amount of fine grained feldspar. This rock, which it isthe chief purpose of this paper to discuss, is closely associated with the coarse feldspathic rock above mentioned, and evidently belongs to the same rock body. In places it has a distinctly striped gneissoid appearance. In general it is composed chiefly of an alkali feldspar and augite, with a comparatively small amount of quartz, thus presenting the mineral composition of an augite-syenite. As the rock in all its phases shows more or less evidence of dynamic action, it is properly to be re- garded as a gneiss,and may therefore be designated as an augite-syenite- oneiss. SUBDIVISIONS OF THE SYENITE-GNEISS. The term syenite-gneiss is here used to include a peculiar assemblage of rocks occurring at High Rock and neighboring mines, among which is the so-called “ leopard rock ” or “concretionary veinstone”’ of the Cana dian geologists. The rock presents three distinct phases, which for con- * Penrose: Bulletin 46, U. S. Geol. Survey, p. 26. 100 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. venience may be treated separately, though they belong to the same rock mass and grade into each other almost imperceptibly. They are— Coarse grained syenite-gneiss. Ellipsoidal syenite-gneiss (leopard rock). Streaked syenite-gneiss. GEOLOGICAL OCCURRENCE OF THE Rock. The rock here described occurs in the form of dikes, sometimes cutting across the strike of the inclosing rocks and sometimes intercalated in them. In one form or another it is found at nearly all the apatite open- ings examined. At the top of the hills at High Rock, about 20 rods southeast of the office, the exposed surface consists of a belt of pyroxenite inclosed in EE] B yroxe nite Syenuite FIGURE 2 —Dzke of coarse Syentte-gneiss cutting Quartzite and Pyroxenite. A mass of the quartzite is inclosed in the syenite. The pyroxenite appears as a dike intercalated in the quartzite. quartzite, with a strike of south 30° east (magnetic). At one point there is a slight linear depression covered with soil transverse to the strike of the rock, which evidently represents a fault. On the south side of this depression the beds of quartzite and intercalated pyroxenite have been shifted about two feet to the east. On the north side both pyroxenite and quartzite are cut by a dike of coarse grained syenite in a direction south 20° east (magnetic). After passing the supposed fault-line the dike GEOLOGICAL OCCURRENCE. 101 bears easterly, parallel with the bedding of the quartzite. These rela- tions are shown in the accompanying sketch. The dike is about a foot wide, and consists chiefly of coarsely crystal- lized dark grayish or purplish feldspar (microcline), with grains and agoregates of augite and occasional patches of quartz. The grain is quite uniform, except for a thickness of about one centimeter next the walls, which is of a finer grain (number 1382). At one point a fragment of the quartzite is inclosed in the syenite. At an apatite opening southeast of Pa PAH, 2 oe Vy im] C__]npat ite. EES coarse Syenite EFS Lecpa rd Rock Ws Ct Quartzite. =AK Pyroxenite. FIGURE 3.—/ntrusion of Syentte tn Pyroxentte and Quartzite. The locality is at an apatite pit on the hill, about 30 rods southeast of the office. The front face indicates the relations as shown in the wall of the pit, while the top represents the surface ex- posure. this, in the line of strike, the syenite is seen cutting the pyroxenite, as shown in relief in figure 5. The pyroxenite has been broken up, and the different parts intricately involved in the intrusive mass. At this exposure the different phases shown by the syenitic rock are seen grad- ing into each other. In some parts the intruding rock is of a coarse grained character, while in other places it shows the ellipsoidal structure, grading finally into the striped gneissoid rock. At one point in the wall of the pit a large mass of the pyroxenite (A, figure 3) is in contact on one 102 oc. H. GORDON SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. side with the coarse grained variety (C), while on the other (B) there is an excellent development of the ellipsoidal structure, with the longer axes of the ellipsoids arranged approximately parallel to the adjacent boundaries of the pyroxenite. The latter at the point marked A shows an interlamination with apatite such as has been frequently noted in the apatite regions.* The surface exposure shows one branch (D-F) of the syenite intrusion narrowed and curving. At D the ellipsoids have become so flattened and attenuated that in transverse section they appear as parallel bands of reddish white feldspar alternating with thin stripes of augite. At however, the ellipsoidal structure again appears with sharp little trenches surrounding the feldspathic ellipsoids, due to the weathering of the inter- stitial augite layers (number 139). At EH a branch is given off, which mn ; iti! ' Pyroyenite. MY s yen wWviSyenite. FIGURE 4.—Dzvike of coarse Syentte-gnetss cutting Pyroxenite and inclosing Portions (A) of the latter. consists of the coarse grained syenite rock, with no indication of ellip- soidal or banded structure. Other portions of the syenite rock make their appearance at the surface here and there as lenses (G@) in the pyroxenite. On the brow of the hill above pit number 11 the pyroxenite is cut by a dike of syenite, a foot or more wide, extending from north 10° west to south 10° east. The dike dips into the hill at a slight angle with the vertical, and at one point gives off a branch (C), which soon wedges out in the pyrox- enite. Fragments of the pyroxenite are inclosed in the syenite, as shown * W. Boyd Dawkins: Proc. Manchester Geol. Soc., 1884. Nore.—The rocks referred to here are represented in the collection by the following numbers A = numbers 109, 110, pyroxenite. D= number 139, ellipsoidal gneiss, weathered. B= number 138, ellipsoidal syenite-gneiss. 1 i 140, streaked syenite-gneiss, c= * 133, coarse syenite-gneiss, GEOLOGICAL OCCURRENCE. 103 at A. The intruded rock (number 164) is medium to coarse grained, but shows a finer texture in the narrow passage between the inclosed masses of pyroxenite. At number 11 opening the ellipsoidal or leopard rock occurs in abun- dance, and constitutes a large part of the wall in places. At one point in the face of the wall a crystal of apatite, six inches or more in diameter, was seen inclosed in the leopard rock, with the ellipsoidal masses dis- posed ina concentric manner about it. Hxamples showing the same re- lations between more or less crushed apatite and the leopard rock were frequent in the refuse of the dumps, but the presence of crystallographic form is exceptional. At various places on the hill dome-like masses of the coarse grained syenite rock appear as local enlargements of the dike. One of these near the office shows fragments of pyroxenite scattered through the intrusive mass. These observations, which may be duplicated many times in the vicinity, are sufficient to demonstrate— 1. That the coarse grained syenitic rock, the leopard rock and the streaked gneiss belong to the same rock-body ; and, _ 2. That this body represents an intrusion of syenite later than that of the pyroxenite. An interesting feature of these syenitic rocks is their remarkably fresh condition. This appearance, which is prominently characteristic of the rock in the hand specimen, is also shown in the thin section, where very little evidence of decomposition is to be observed. Epidote and chlorite, generally common as decomposition products, are rare in these rocks. MREGASCOPICAL CHARACTERS. COARSE SYENITE-GNEISS. The first of the syenite-gneisses consists of a very coarse grained mix- ture of microcline and monoclinic pyroxene chiefly, with a variable amount of quartz. The rock is divided into irregular angular blocks, the largest being one or two inches across, separated by thin anastomosing sheets of granular feldspar, augite and quartz. These interstitial areas are sometimes thick enough to be readily traceable in the hand specimen. Many, however, are scarcely or not at all recognizable megascopically, but are brought out with distinctness under the microscope. The microcline is of a dark gray, often purplish color, crystallized in large individuals, frequently from one to two inches in diameter. The larger cleavage faces show a slight undulatory surface, with variable re- flection and bright pearly sheen. XV—Butt. Grou. Soc. Am., Vou. 7, 1895. 104 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. An analysis of the microline by William Hoskins, of Chicago, gave the following results : eae avai 2 Gil RSE LF Guar se 64.54 AOS S So Pia he ie rane eee ecm acne ae 20.55 eR ORoe 5 cians vie oe ee re Rae ah aware 45 CaO: aciais js St eR eae ee Rade Mie 99 MPO ie dhe hs Brrr teeaes, eee mene, Eka es trace DGG ats dhe 2 os eas nite eas ee 1.62 1 6 Fa, ee eA hes reser NE) 10.78 Baer. hss sce ee wk ye agate Ve 32 ‘Loss by-ionitioni. tos sate One eee ee 49 99.74 The pyroxene is of a dark green color, and occurs both in well formed prisms and as irregular aggregates of considerable size. The prisms are long and slender, and occur chiefly inclosed in the microcline, though they are associated sometimes with the quartz in the granular areas. In the latter case, however, they are usually shorter and sometimes show evidence of breaking. ‘The prisms are elongated in the direction of the vertical axis and have a very nearly equal development of the faces wo Pw (100), « P~% (010), P (110). The pyroxene is more abundant in places, forming aggregates, inclosing a varying amount of feldspar, pyrites,and titanite. In these areas the pyroxene is coarsely crystallized and of a lighter green than the idiomorphie individuals. An analysis of the crystals inclosed in the microcline by Mr Hoskins gave the following results : erg tL |. Ot open ae ea 49.79 * Al,O, Les. Bw 2 SS) 8 ee Oe, ee) ee ee ee) ee a ee er ee 2 93 POO ere es Oe oh eae ee 18.95 Ca ete ea ia SS eee 21.76 MaQ:cukiidate ose 5.60 Washi S )diap san’ eet abate 2d en ee G1 KAO sista m5 Seip: sin RN oe pa ee 06 100.00 The analysis shows a high percentage of iron, while alumina and mag- nesium are correspondingly low. Taken in connection with its physical characters, the analysis seems to indicate an augite closely allied to diallage. Quartz occurs chiefly in nests and strings, associated with the inter- stitial granular areas. ‘Titanite is present as grains disseminated through the rock, but is more abundant along the granular areas. A fine grained *'The per cent of silica was rendered doubtfully low by an accident, and is put in by difference. MEGASCOPICAL CHARACTERS. 105 form of this variety is represented in the collection by two specimens (numbers 128, 129), which are quite uniformly granular and without a distinct egneissic structure. The chief constituents, feldspar and augite, are more or less uniformly distributed. ELLIPSOIDAL SYENITE-GNEISS. In the second phase the rock consists of irregularly ellipsoidal or ovoid masses of feldspar, with more or less quartz, separated by narrow anas- tomosing partitions of green interstitial material, in which there is some- times observed a slight schistosity parallel to the surface of the masses they inclose. This is sometimes apparent also in the tendency of the rock to cleave at the contact between the ovoid lumps and the interstitial material. The ellipsoidal masses are usually more or less elongated and arranged uniformly, with the longer axes lying in the same direction. They are of all sizes up to two or three inches in diameter in cross-section, and several inches long. The grain of the feldspathic lumps is coarse in the more spheroidal forms, but becomes finer as the masses become more and more flattened and elongated. They are composed chiefly of feldspar, with a varying amount of quartz and disseminated grains of titanite, augite and apatite. Larger grains of feldspar often appear in the more flattened forms, inclosed in the finer grained mixture. The interstitial material is of finer texture than these ellipsoidal masses, and seems to be composed chiefly of pyroxene, along with finely granular feldspar and some quartz. The interstitial material sometimes has the appearance of two layers of pyroxene separated by a thin light colored lamina feld- spathic in character. In many cases these are seen to represent the wedging out of the ellipsoidal feldspathic masses. Often the thin seam of feldspathic material between two adjoining ellipsoids is found to be directly connected with the small masses of feldspar occupying the angu- lar space between three or more ellipsoids. Sometimes the lumps are flattened to thin lenticular or disk-like forms, which may be bent or _ folded so as to partially enwrap or inclose adjoining lumps. As these flattened lumps become thin they wrap about the larger ellipsoids, so that when the rock breaks across them the constituents appear to have a con- centric arrangement. In some cases, however, the concentric arrangement could not be traced directly into connection with the flattening of the feldspathic masses. The interstitial bands are composed chiefly of dark green pyroxene, in grains often elongated parallel to the vertical axis, | along with more or less finely granular feldspar. The pyroxene grains often lie with their longer axes transverse to the pyroxene band, with their ends projecting into the feldspathic areas, thus presenting somewhat 106 Cc. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. the appearance of a radiate structure. Usually a thin layer of the feld- spar adjoining the pyroxene bands is finer grained than the main part of the lump, but in some cases one or more large grains of microcline will appear to lie directly against the pyroxene band. On close inspection, however, they will be found to be separated by a very thin layer of fine erained feldspar. A mass of the rock will sometimes show a thin seam of the light colored constituent cutting across both pyroxene layers and feldspathic ellipsoids. In may be traced across the latter by its lighter color and finer grain. In specimen 150 a seam of this character cuts entirely through the block, and apparently represents a fracture which has been recemented by subsequent crystallization. The interstitial filling weathers more readily than the ovoid masses, leaving sharp little trenches surrounding these on exposed surfaces. Beginning with a somewhat flattened ellipsoidal form, the feldspathic masses become more and more flattened in one direction and extended in a direction normal to it. With this flattening of the ellipsoids, the pyroxene shells become thinner and arrange themselves more and more in parallel bands until the ellipsoidal structure is more or less com- pletely lost, and in its place there occurs a striped gneissoid rock consti- tuting the third phase. On cross-fracture the pyroxene layers of the latter are seen to coalesce, clearly indicating its relationship to the ellip- soidal rock. Where the rock incloses large crystals or masses of apatite the ellipsoidal lumps arrange themselves more or less concentrically about the inclosure, a feature which is characteristic also of the bands of pyroxene and feldspar in the succeeding phase. ‘This is well shown in specimens 137, 146, 148, 154 and 158. STREAKED SYENITE-GNEISS. As the ovoid masses become more flattened and disk-like the augite layers arrange themselves in parallel bands, alternating with the thicker feldspathic layers. Moreover, there is a marked diminution in the size of the grains, while quartz becomes relatively more abundant. Here and there areas appear which are coarser grained, and in these the pyroxene is less abundant, and is disseminated in grains and small masses instead of being aggregated into layers. There is also observed occasionally in the fine grained gneissoid rock large grains of feldspar, which sometimes inclose grains of pyroxene. The masses of apatite inclosed in the rock vary in size from a few inches to a foot or more across, and around them the bands of pyroxene and feldspar curve concentrically. The apatite is crushed in part or wholly to a granular, saccharoidal condition. Num- ber 145 shows a portion of such an apatite mass in connection with the adjoining rock. ‘The latter is rather fine grained and is striped with MEGASCOPICAL CHARACTERS. 107 pyroxene and feldspar for a space of 10 centimeters from the apatite, beyond which it is coarser grained and without the streaks of augite. In this part the pyroxene is much diminished in amount and occurs only in small, scattering aggregates and grains. Immediately next to the apatite is a layer of feldspar, from + to 1 centimeter thick, and much coarser in grain than the succeeding bands. Moreover, the apatite mass is inter- sected by several thin veins of feldspar and quartz, continuous with that of the surrounding layer. These seams vary in width from 2 millimeters to 5 millimeters, and vary proportionally in the size of grain. At one point a grain of feldspar occupies the full width of the seam. The apatite is granulated on the outer surface of the mass, while the remain- ing portion is irregularly fissured and broken, though not completely crushed. In number 155 the layers are sharply plicated, and the augite and feld- spathic constituents intermingle to a greater extent than in those with straight layers, thus partially obscuring the banding. MicroscopicaAL DESCRIPTION. COARSE SYENITE-GNEISS. Specimen number 127 (129) consists of coarsely crystallized microcline and a monoclinic pyroxene, with quartz, titanite, apatite and pyrite as accessory constituents. Occasionally a small prism of tourmaline is ob- served. The microcline is of a dark gray color and occurs in large grains inclos- ing numerous well formed prisms of augite, usually long and slender. The rock is intersected in various directions by granular bands which separate the mass into angular lumps of various sizes. These intersti- tial bands are usually thin, but become somewhat thicker in places, and consist chiefly of finely granular feldspar, quartz and augite. The quartz appears usually in connection with these granular bands, sometimes in ageregates of considerable size. These interstitial seams, therefore, con- stitute a rather obscure network of fine grained feldspar, pyroxene and quartz, inclosing various masses of coarse microcline and augite. The augite in the quartz areas occurs both as grains and prisms. The latter, however, are usually shorter than those inclosed in the microcline, and present less well defined crystallographic outlines. Sometimes two short prisms of the same size may be seen lying end to end, but not quite in the same straight line. The idiomorphic individuals are elongated in the direction of the vertical axis, and show a very nearly equal develop- ment of the faces co P (100), « P™~% (010), P (110). Section 129, cut transverse to the granular band seen in the hand speci- 108 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. men, shows this area to be made up of small angular and rounded grains of microcline, between which in places there is a small amount of feld- spar similar to that forming the microperthetic intergrowth observed in the large microcline grains, and which is probably albite. Quartz appears in irregular elongated grains here and there along the granular areas. It has crystallized sharply against the adjoining constituents, fitting closely into their sinuous outlines. Some grains are long and irregular in shape, while others which appear similar are resolved in polarized light into two or more grains having slightly different orientations. Rows of fluid FIGURE 5.—Section Transverse to granular Bands in the coarse Syenite-gnetss. Showing augite crystal broken in the portion projecting into the granular band. Au = augite ; M = microcline; P= plagioclase; Q0 = quartz; “ = feldspar. inclusions frequently extend uninterruptedly from one quartz grain to another, while in a few instances they were observed in direct continuity with similar lines of inclusions in the adjoining microcline plates. The large microcline and augite grains, which constitute the larger part of the section, are traversed in places by a series of irregular cracks, which lie parallel with the lines of fluid inclusions in the quartz. The augite sometimes shows the effects of mechanical movements in the fracturing and breaking of a grain lying in contact with the granular zone. The grain shows fracturing throughout, but the breaking appears MICROSCOPICAL DESCRIPTION. 109 greatest along the side adjacent to the granular area. As shown in figure 5, one fragment has been separated slightly from the main portion and small grains of augite lie distributed along the crack. Some of the small grains of augite are very slightly pleochroic, though the main part of the crystal is not. In specimen 125 (127) the chief constituents are also coarsely crystal- lized. The microcline has a purplish gray color, and incloses large prisms of augite similar to that previously described, except as to the size of the individuals. The pyroxene in the interstitial areas, however, is without idiomorphic outlines, and occurs mostly in large, irregular grains and ageregzates. These are of a lighter green than the idiomorphic indi- viduals in the microcline, and contain a larger number of inclusions of feldspar, titanite and pyrite. This becomes apparent in attempting to select the augite for chemical analysis. The anastomosing gran- ular bands separate the rock into angular lumps of irregular sizes. These bands are not always plain- ly apparent in the hand specimen, but are revealed with distinctness by the microscope. In thin section (number 127) the coarsely crystallized feldspar shows a beautiful development of the crosshatching of microcline. The augite and large microcline . ; 2 FIGURE 6.— Section showing granular Band tinter- grains meet each: other with toler- RS Pe Tee. ably sharp boundaries, though a 4—apatite; Q— quartz; P= plagioclase ; = small amount of granular micro- ™icrocline. cline and augite is sometimes observable along the line of contact. There are also narrow bands of granular microcline extending outward from the angle of the augite grain along the spaces between the large grains of microcline. In these zones of granular microcline there occur also small grains of plagioclase. Between the grains of microcline and plagioclase there appears in places, as a cement, a very small amount of feldspar in which fine striations may sometimes be detected. The microcline is sometimes clouded and contains occasional small plates of biotite, rutile needles, and apatite. The augite occurs in irregular grains, usually elongated parallel to the vertical axis. It is of a pale green color and nonpleochroic. The pris- matic cleavage is well developed. Cleavage parallel to the orthopinacoid 110 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. and clinopinacoid is common, though usually less distinct than the pris- matic. Grains in which the cleavage parallel to « P & (100) is well developed, giving a diallage-like appearance, are frequent. This is often made more apparent by the presence of thin twinned lamelle parallel to the same plane. Inclusions of feldspar, sometimes showing the cross- hatching of microcline, occurin the augite. Grains of pyrite occur along the cleavage, as also small, irregular flakes of hornblende. The horn- blende is green and pleochroic in green and greenish yellow colors. | The alteration of the augite to compact hornblende was clearly demon- strated by the presence of the latter along the fracture cracks of the former. At one point in the crack there is a small grain of compact green horn- blende partially inclosing a grain of iron ore. The hornblende shows well marked parallel cleavage and distinct pleochroism. In addition to this, the irregular fracture lines of the augite are bordered on either side by a thin greenish band, differing from the augite in its double refrac- tion and showing distinct pleochroism. These bands are in direct con- tinuity with hornblende in the larger space and must be regarded, there- fore, as the same. The zone of hornblendic substance is often observed along the fracture lines in the augite and grading into the latter. They are so narrow, however, that, though showing a slight degree of pleoch- roism, their identity is clearly established only when found in direct con- nection with larger recognizable masses. The apatite occurs both in the form of microscopic inclusions in the microline and as rather large rounded grains, both in the microline and in the interstitial areas. ‘These often inclose small grains and prisms with pyramidal terminations, having a lower index of refraction, which are probably quartz. In one of these inclusions the extinction was found to be parallel to the longer axis. In thin sections of specimen number 126 (128), which resembles the preceding, the granular zones are narrow and consist chiefly of microcline with a very small amount of plagioclase. The large grains of microcline show the characteristic microperthitic intergrowth with albite (?). The microcline holds as inclusions numerous small biotite plates, apa- tite, rutile needles, and an abundance of fluid inclusions. ‘The biotite plates are often distributed along the cleavage in parallel lines or bands. In other cases they appear in considerable numbers in prisms and hex- agonal sections in intersecting parallel lines corresponding to the cleavage. Dust-like decomposition products appear quite abundant in some areas, and especially along fracture lines. An aggregate of augite grains, with irregular, but rounded, outline, oc- cupies the space between two large microcline individuals. The augite MICROSCOPICAL DESCRIPTION. 111 and the coarse feldspar meet each other for the most part in well defined boundaries. In some places, however, a small amount of finely granular microcline occupies the space between them. Rounded grains of calcite, pyrite and microcline areinclosed intheaugite. The fracture lines affect- ing the microcline sometimes extend uninterruptedly through an adjoin- ing augite grain. | Specimen 128 (130) has a more even and finer grained texture than the preceding, and likewise a more uniform distribution of the constituents. In thin section the microcline shows a less clearly defined grating structure than in the foregoing, and occurs in irregularly angular and rounded grains from two to five millimeters in diameter, usually sepa- rated from each other by a network of finely granular microcline, inclos- ing here and there larger grains of augite and a small amount of quartz, titanite and calcite, but sometimes meeting each other along a common boundary. ‘The proportion of augite is much less than in most of these rocks. The interstitial granular areas consist of very small grains of microcline and augite, with a small amount of clear, fresh-looking plagio- clase feldspar between them in places. The augite does not appear to have suffered much disarrangement of parts, though coarse fractures are common. Asmallamount of pyroxene occurs in small grains distributed in the granular microcline bands. In one instance a grain lying at the intersection of three granular bands shows a line of fracturing extending diagonally across the parallel cleavage, while narrow zones of hornblende with faint pleochroism border a set of fractures developed transverse to the cleavage. ‘There is a shght difference in extinction and discordance in the direction of the cleavage in the two parts into which the diagonal crack divides the grain. Along this crack there appear small grains of apparently fresh feldspar and one of titanite, together with several small augite grains. A diallage-like appearance due to polysynthetic twinning parallel to the orthopinacoid appears in some grains. No differences were noted between the large idiomorphic augite grains and those appear- ingin the granularareas. They show little evidence of alteration. More or less quartz is present in the granular areas usually carrying rutile and fluid inclusions and in some cases small rounded granules of augite. Calcite, apatite, titanite and iron pyrites appear in small amounts. The apatite carries small inclusions of quartz. Specimen 129 (131) resembles the last, except that the microcline has a lighter color, giving the rock a fresher appearance. In general, the texture is medium to coarse, with much larger grains of microcline scat- tered through the mass. The augite and titanite appear quite uniformly distributed, but on close inspection the former may be observed slightly ageregated in anastomosing lines, giving an obscure net-lke aspect. On XVI—Buu.. Grou. Soc. Am., Vou. 7, 1895. 112 cC. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. one side the specimen shows a flat surface, along which shearing has taken place. In thin section in some areas a small amount of granular microcline appears in the spaces between the larger grains. Plagioclase feldspar occurs as small, irregular, unstriped grains in the interspaces and in small amount as a cement. The microcline in general shows greater alteration than in the preceding specimen. Numerous particles of epi- dote and occasional scales of biotite are scattered through it. Hpidote also occurs in connection with the augite. The augite agrees with that of the preceding specimen. In one case a grain having a pronounced twin lamination parallel to the vertical axis has been much fractured and partially squeezed in two at the middle. The two parts on either side of the major fracture are slightly changed in orientation, while small grains of plagioclase, feldspar and green horn- blende occupy the crack. A large grain of apatite inclosed in the augite adjoins the crack. The augite is cut nearly parallel to the clinopinacoid, 0 P&® (010), and shows an extinction angle of 45° for the thicker lamellee and 33° for the fine. It is evident, therefore, that the lamellar structure is not due to a fine interlamination of an orthorhombic pyroxene (hyper- sthene), as might be supposed, but represents a multiple twinning, prob- ably representing gliding planes, due to pressure. The part of the grain showing greatest fracturing also shows decompo- sition products in greater abundance. ‘These consist of epidote, calcite, and hornblende; the latterin small pleochroic grains scattered along the fracture lines. In one case, showing only parallel cleavage, the extinction measured upon the cleavage lines was 15 degrees. In some cases an aggregate of small grains of augite have a considerable amount of horn- blende in the form of small flakes and grains associated with the augite. Brown biotite appears in small amount, apparently as a decomposi- tion product. A mass of rock four feet long, observed on the dump of opening num- ber 11, showed the gradation from the first into the second or ellipsoidal variety.* Specimen 1386 (139, 140) was taken about 15 inches from the ellipsoidal end of the block. It consists of coarsely crystallized areas of feldspar and augite, with intervening areas in which the constituents form a moderately fine grained mixture. The augite occurs in large and small grains, the latter in the granular areas. The feldspar (micro- cline) has a gray color and shows a sharp but variable reflection from the cleavage faces. Twinned individuals having the greatest extent par- allel to the twinning plane are common. The augite and titanite both * This block is representec in the collection by numbers 134, 135, 186 and 137, taken in consecu- tive order, the last number being the leopard rock. MICROSCOPICAL DESCRIPTION. ples: appear in larger grains in the more coarsely crystallized portions. In the finer grained areas the constituents are quite uniformly distributed. In thin section the microcline shows a pronounced development. of microperthitic intergrowth with a more strongly doubly refracting, fresh- looking feldspar (albite ?). In some sections the albite bands show a very fine transverse striation, upon which the extinction is about five degrees. Sometimes small grains of quartz appear inclosed in the albite bands. The inclusions appearing in the microcline occur also in the albite bands and in the quartz and correspond to those already described. In some cases plates of biotite extend across the boundary between the microcline and albite and are partially inclosed in each. The space between the microcline grains is occupied by finely granu- lar microcline and a striated feldspar, with here and there larger grains of green augite. The striated feldspar is fresher in appearance than the microcline. The laminations are very fine, and according to a number of measurements extinguish at from 4° to 5° on either side of the twin- ning plane where this bisects the angle of extinction of the two lamelle, thus corresponding to albite or oligoclase. In addition to biotite. plates, these plagioclase feldspars often contain large numbers of rounded or nodular inclusions of quartz.** Specimens 134 (187) and 135 (188), taken from the same block as the above, are much finer grained and contain a considerable amount of quartz, showing with the feldspar a somewhat obscurely banded arrange- ment. Augite is present in comparatively small amount. The feldspar is pinkish, except in the coarser patches, where it is the usual gray color. Grains of microcline up to the size of peas, sometimes inclosing augite, occur scattered through the finer grained areas. Under the microscope these specimens are seen to differ from the preceding in the greater ex- tent of the interstitial granular areas, the lessened amount of augite, and ereater abundance of quartz. In addition to the microperthitic intergrowth with albite, the larger microcline grains show an abundance of the nodular quartz inclusions, with here and there similar forms made up of an aggregate of albite grains. These nodular quartz inclusions are of considerable size, varying from .035 of a millimeter to .425 of a millimeter in diameter, the larger ones sometimes reaching a length of .95 of a millimeter. The largest in- dividuals are often separated into a number of parts, each with slightly different extinction. It is noticeable that in many of them the more prominent fractures extend nearly in the same direction. In the granular ¥ * These correspond to the “ quartz de corrosion” of the French authors, a term quite inappro- priate, as they in no sense represent corrosive action. In correspondence with their peculiarities of shape, the term ‘‘ nodular quartz’ is here adopted for them. 114 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. zones there is an abundance of fresh, finely and sharply striated feldspar usually carrying nodular quartz inclusions. The quartz occurs distributed through the fine mosaic, often in lines and stringers of elongated grains, in which, in addition to fluid inclusions, the inclusions found in the microcline are abundant. Hpidote appears in particles distributed through the microcline, while the titanite often occurs in association with an opaque iron ore, probably ilmenite. ELLIPSOIDAL SYENITE-GNEISS OR LEOPARD ROCK. The collection embraces specimens showing a gradation from forms in which the constituents are rather coarsely crystallized and the feldspathic cores approximately ellipsoidal through others in which they become more and more elongated, flattened and distorted, accompanied by a de- crease in the size of the grains, to those finally in which the constituents are fine grained and arranged in parallel bands. The specimens were taken from the dumps, which offered excellent opportunities for securing an abundance of fresh material. Specimen 137 (141, 142, 143), which was taken from the block furnish- ing specimens 154, 135, 156, represents the ellipsoidal rock in contact with a mass of apatite, the ellipsoids sometimes thrusting themselves in be- tween adjoining masses of apatite which may have belonged originally to the same deposit. The interior of the feldspathic lumps is sometimes coarse grained and identical with portions of number 156. They also show a small amount of augite intergrown with the feldspar in these coarse grained parts, while the finer grained, outer peripheral portions show little, if any, augite. Quartz is present in considerable amount in lines parallel with the longer axis of the lump. In thinsection cut transverse to the pyroxene bands they are seen to consist of a fine grained mosaic of feldspar, quartz and augite, while the ellipsoids are composed chiefly of microcline in much larger grains. In some cases, however, the ellipsoid is more or less fine grained throughout and has lines of quartz extending through it parallel with the longeraxisofthelump. The augitic bands correspond in struc- ture with the granular belts observed in the coarse syenite. The constit- uents all show a pronounced tendency toward a laminated arrangement. They consist of microcline and plagioclase in about equal proportions, mostly in small equidimensional grains. Quartz is quite abundant, mostly in elongated grains and aggregates arranged in lines parallel with the lines of augite. They contain as inclusions small plates of biotite and occasionally small rounded grains of augite, unstriped feldspar, and small rhombic sections of titanite. lol MICROSCOPICAL DESCRIPTION. 115 The plagioclase is generally fresh, with sharply defined striations, but in some cases is without stripes and more or less cloudy. It contains an abundance of nodular quartz inclusions. The constituents of the granular areas show a tendency toward the micropoikiltic structure. The augite occurs in grains and aggregates distributed in a belt along the middle of the granular band. The grains are irregular in outline and often elongated parallel to the vertical axis. The grains usually lie with their long axis approximately parallel with the direction of the band, but in many cases they lie transverse to it. The augite grain is often pleo- chroic, but without distinct evidences of alteration. It does not differ essentially from that of the coarse grained rock. Pinacoidal and prismatic cleavages are usually well developed in the larger grains. Apatite occurs sometimes in s large elongated grains lying par- allel with the general lamination. They carry numerous quartz in- clusions, sometimes in_ short prisms terminated at each end by the pyramid. A small amount of calcite is present. In specimen number 142 (147, 148, 149) the feldspathic lumps are mostly small, usually much elongated, and very irregular in shape. The pyroxene is abun- dant in the interstitial areas, but is not regularly distributed. In “‘°’** Uh OB rg hs esd at Pe places it occurs in lumps of con- Showing prisms of augite lying in the granular siderable size, in which there ap- bands and transverse to it. pear grains of quartz and feldspar and a considerable amount of titan- ite. The flattened feldspathic lumps arrange themselves somewhat concentrically about these aggregates. The flattened feldspathic masses are relatively small near the central augite aggregate, but increase rapidly in size outward. Often the augite lump represents simply a local thickening of the interstitial band. The augite is of a dark green color and occurs in comparatively large prisms, those of the segregated masses occasionally exhibiting crystallographic outlines. In the inter- stitial bands the prisms sometimes arrange themselves transversely and project more or less into the feldspathic masses on either side. The feldspathic lumps are made up of a groundmass of fine grained feldspar and quartz inclosing large, often twinned, tabular grains of microcline. 116 ¢c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. The size of the large microcline grains is,in general, proportional to that of the feldspathiclump. The titanite occurs chiefly in association with the augite. It appears also in the feldspathic areas, but less frequently. In thin section the microcline grains of the feldspathic lumps are found to measure from 13 millimeters to 2 millimeters in diameter, while the granular matrix in which they lie is composed of grains of microcline and plagioclase varying in size from .3 of a millimeter?to .6 of a millimeter. The small grains of microcline in the granular areas often appear cloudy at the center, while the outer portion is clear and fresh. The same appearance is also observable in some of the plagioclase grains. In these and in many other sections the granular feldspars are often seen to meet each other in straight, sharply defined boundaries, suggesting an approach to crystallographic outlines. This tendency is especially pronounced in grains in which the outer zone is fresher than the inte- rior. The larger microcline grains are filled with particles of epidote, plates of biotite, fluid inclusions, and occasionally microscopic grains of tourmaline. They are generally much fractured and in some cases show cracks filled with calcite. Quartz, calcite and plagioclase feldspar occur in the augite, often, but not always, in connection with fractures. Small aggregates of hornblende needles, accompanied by’an opaque iron ore, appear occasionally as alteration products of the augite. The latter often shows distinet pleochroism : a=greenish yellow. b=ereen. ¢=green or slightly yellowish green. Quartz is not plentiful. In number 143 (150, 151, 152, 153) the ellipsoidal masses are larger, but are more or less flattened and contain a considerable amount of quartz. Under the microscope the interstitial areas show a pronounced deyvel- opment of the granular or “ mortel structure,” in which in some places there appears a preponderance of plagioclase beset with an abundance of nodular quartz, while in others granular microcline predominates. The constituents have the relations characteristic of the micropoikilitie structure. The larger grains of microcline sometimes show the effects of dynamic agencies in the bending of the lamelle and undulatory ex- tinction. The augite occurs in irregular grains and aggregates distrib- uted along the middle of the granular band, as usual. Quartz is quite abundant in angular, often elongated grains. They sometimes inclose small, rounded grains of pyroxene, which in one case appeared to stream out from a large augite grain adjoining. The apatite occurs, as usual, in rounded and elongated grains, some- times inclosing rounded grains of augite. MICROSCOPICAL DESCRIPTION. Vy In specimen 149 (161, 162) the feldspathic ellipsoids are drawn out into irregular, flattened, disk-like forms. The rock shows a tendency to cleave along the face of the augite bands when the breakage is parallel to the longer axes of the disks. Under the microscope both feldspar and augite are much finer grained than in the preceding. The latter especially appears in smaller and more rounded grains, which he distributed in a linear direction in the granular feldspathic matrix. The diallage-like structure, due to the presence of twinned lamellee parallel to the orthopinacoid, appears some- times in sections cut transverse to the vertical axis. Indications of oro- graphic pressure appear in one instance in the breaking of an augite grain, calcite being deposited in the crack, and, further, in the appear- ance of fractures extending across adjoining grains of pyroxene. ‘The granular feldspar grains often meet each other in sharp, straight lines, In one case two grains of microcline, one of which is bordered by a small amount of fresh plagioclase feldspar, are separated by a narrow band of the latter, which in polarized light is seen to be twinned, the part on either side of the twinning plane being very nearly, but not quite, in optical orientation with its adjacent microcline grain. In specimen 138 (144), taken from the dike intersecting the pyroxenite and quartzite, shown on page 101, the feldspathic lumps and interstitial bands are both of a uniformly fine grained granular texture. In thin section the feldspathic lens-shaped masses are shown to-consist of rather fine but relatively uniform grains of microcline, with a mod- erate amount of quartz and the usual portion of titanite. Occasionally the microchne carries nodular quartz inclusions, but they are few. In the interstitial zones the constituents are somewhat finer, and the grain lie with their longer axes extended in the same direction. The granular feldspar has the characteristic structure of microcline, while plagioclase is almost altogether wanting. However, these g¢ aetna es carry a considerable amount of nodular quartz. The augite appears in small grains disseminated in the granular feld- spar matrix. This specimen differs from the preceding in the presence of aconsiderable amount of hornblende associated with the augite; some- times connected with it zonally. Some of the hornblende presents idio- morphic outlines with well developed prism faces meeting at angles of 56° or 124.° STREAKED SYENITE-GNEISS. Specimen 140, associated with the last in the exposure described on page 101, shows a nearly complete flattening of the ellipsoids, giving a _ well developed gneissoid banding. Its identity with the leopard rock, however, is plainly evident from the anastomosing of the augite streaks 118 ©. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. visible on a transverse fracture face. Specimen 156 (186), taken from the dump, represents the same structure in a fresher rock. The constituents appear in a remarkably fresh condition. The feldspathic areas have been completely flattened, so that on a fracture face parallel with the plane of the greatest and least axes of the flattened ellipsoids the con- stituents are seen in parallel bands. When the rock breaks transverse to this and the augite bands, the latter are seen to coalesce. ‘The mass of the rock is fine grained, with large crystals of microcline scattered here and there along the feldspathic bands. Under the microscope the rock is seen to consist chiefly of finely granu- lar feldspar, with larger grains of feldspar and augite scattered through the groundmass. The granular groundmass consists of microcline and unstriped feldspar, the latter occurring both as a cement and in small grains carrying numer- ous nodular quartz inclusions. A small amount of striped feldspar is also present. The granular microcline gener- ally appears fresher than that in large grains, and in some cases ap- pears more cloudy at the center than in the peripheral portions. The augite grains vary greatly FIGURE 8.—Jdiomorphic Hornblende Crystal in the 1 Size, but in general are inter- streaked Syenite-gnetss, mediate in size between the micro- Ap — apatite; du = augite; H—= hornblende; cline of the groundmass and the Achat OR tric oe large porphyritic grains. They lie scattered along planes in alternation with the bands of feldspar. Hornblende is present in considerable amount, associated with the augite. It does not occur outside of the augite bands. Its relations to the augite, however, are not such as to clearly prove its derivation from that mineral. It sometimes appears in small flakes along the cleavage lines of the augite and is frequently in zonal relation with the latter. It shows the relations with the other constituents characteristic of the micropoikilitic structure, and frequently appears in crystals with characteristic crystallographic outlines. In figure 8 it is evident that the hornblende is the result of a separate crystallization, as shown by its idiomorphic form and relation to the adjacent minerals. ‘The manner in which it incloses the apatite shows that it has crystallized subsequently to the apatite. Moreover, its per- MICROSCOPICAL DESCRIPTION. 119 fection of outline indicates that it has not passed through the varied ex- periences to which the original constituents of the rock may have been subjected. It does not have the crystallographic relations with the augite characteristic of paramorphic development. ‘Titanite also appears here, often in small rhombic crystals. Specimen 155 (166) shows a pronounced plication of the bands. Along with this the rock shows more or less schistosity in certain directions. The direction of this cleavage, however, has no apparent relation to the direction of the gneissoid banding. ‘The augite streaks are less sharply defined than in the specimen last described (156). Under the microscope there appears a uniformly very thin, fine grained eroundmass of feldspar and quartz, in which large grains of microcline and numerous small grains of augite and hornblende arranged in bands occasionally occur. It is somewhat more granular than the preceding, but in other respects does not show any marked differences. CHARACTER AND RELATIONS OF THE Rocks. The relations of these rocks as observed in the field are thus seen to be borne out by their petrographical character. Beginning with the coarsely crystallized rock, in which the gneissic structure is but imper- fectly developed, there is a gradation into finer grained and more gneissoid forms, until we have in the last stage a well developed gneiss. In its typical form it consists of a mixture of microcline and augite as essential constituents, while quartz, titanite and pyrite play an accessory role. A peculiar feature of this rock is the large size of the microcline and augite grains and the segregation of both in lumps and patches. The plagioclase feldspar appears in proportion to the extent of the granular areas and constitutes an essential feature of these areas, though usually playing a less important part than the microcline grains. It is usually much fresher than the microcline, but in some cases the opposite is true. The augite, which occurs in large crystals and grains or segregated masses in the coarse grained rock, also appears in smaller grains as the rock becomes more granular and gneissoid, distributed along the middle of the gneissoid bands. The constituents of these bands are strictly allotriomorphic in their relations and often show a distinct micropoikilitic structure. These in- dications of recrystallization become more pronounced in the more gneissic forms. The indications of secondary enlargement of the micro- cline, and sometimes of plagioclase grains, are also more apparent in these latter specimens. In the coarse grained rock plagioclase feldspar is generally present in small amount as a cement between the grains of microcline. This cement-like arrangement is also apparent to a small XVIL—Butt. Grou. Soc. Am., Vou. 7, 1895. 120 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. extent in the more gneissoid variety, but is overshadowed by the general appearance of recrystallization. The idiomorphic character of the augite inclosed in the microcline is wholly absent in all the pyroxene grains appearing in the granular groundmass. These grains often appear pris- matic, but never have regular. outlines. They frequently lie transverse to the augite band, giving the radiate arrangement already noted. The evidences of dynamic movements appear in the coarse rock in the fracturing and breaking of the constituents and the presence of cracks extending uninterruptedly across adjoining grains. The last feature was less apparent in the more gneissic rock. The grains of augite usually show an abundance of coarse fractures and occasionally a broken grain with parts slightly disarranged. The appearance of breaking occurs in the augite in the granular bands only, and if these bands represent lines of breakage, as is believed, the fracturing of the pyroxene may be correlated with the agencies which initiated the development of the gneissic struc- ture. No distinct evidence of the derivation of the granular pyroxene from the coarse grains and crystals could be made out. Hornblende does not appear to have been present in the original in- trusion. It sometimes appears in a very small amount in connection with the augite, sometimes as an alteration of the latter. In these cases it is destitute of crystallographic form. In the gneissic rock, however, the hornblende becomes quite abundant and often presents well marked idiomorphic outlines. The quartz in the coarse granular rock, as in the case of the feldspar and augite, appears to be distributed in patches. It is apparently more abundant in the more gneissic rock, where the grains assume elongated forms and are arranged more and more in lines. The laminated gneissic arrangement of the constituents is a marked characteristic of the granular bands. In those of the coarse grained rock it is not marked. It is indicated, however, by the tendency of the quartz grains to become elongated parallel with the band. In the leopard rock there is a distinct lamination of the constituents in the interstitial zone. In cases where the feldspathic lumps are more or less granular through- out, the constituents tend to assume a laminated arrangement, but this is not common. NOMENCLATURE. The characteristics of the streaked gneiss here described correspond to what, according to Zirkel,* may be regarded as a pyroxene-gneiss (augite- gneiss). A somewhat similar rock, but apparently containing less augite, has been described by Lacroix as granulitic microcline-gneiss.t * Lehrbuch der Petrographie, band iii, p. 219. + Bulletin de la Société Frangaise de Mineralogie, April, 1889. NOMENCLATURE. 121 The term pyroxene-gneiss has been applied by the last named author to rocks, of much more basic composition than those under considera- tion, corresponding to the pyroxenites of the apatite region. It requires but a brief survey of petrological literature to become aware of a great difference in usage as to the term gneiss. By some authors the mineralogic composition of these rocks is made the basis of defini- tion, and, being regarded as having the greatest analogy with the gran- ites, they are defined as characterized by the presence of feldspar and quartz as essential constituents.* By many, however, the term is employed in a structural sense to denote the coarser schists, which so often present granitoid characters—a more comprehersive and preferable usage, since the gneissoid or foliated struc- ture may characterize rocks of very diverse composition. The difficulty attending the application of a mineralogical definition is acknowledged by Zirkel when he attempts to draw the line between certain hornblende- eneisses and amphibolite. The mineralogical definition precludes the use of the term syenite- eneiss. This name, however, has been used for a quartz-bearing horn- blende-gneiss and is given as a synonym for this rock by Geikie.t Naturally enough Zirkel does not recognize such a division and sets the term aside as misleading.§ On the whole, it seems to the writer that the broader use of the term gneiss in the structural sense is to be preferred. This usage prevails quite generally among the English and French pe- trographers. The principles applicable to the classification of the gneisses may be summarized as follows: 1. Their mineralogic composition. 2. Identity in composition and texture with the igneous or sedimentary rocks. Origin unknown. 3. Identity in origin, composition and texture with igneous or sedimentary rocks. Under this we have to consider (a) those rocks in which the gneissoid structure is due to dynamic agencies, and (6) those in which it is the result of conditions at- tending their original solidification. * Zirkel: “ Zum Wesen des eigentlichen Gneisses gehért das jedesmalige Dasein von Kalifeld- spath, Quarz und von einem triklinen Kalknatronfeldspath oder Natron-feldspath. . . . Wesent- lichen Bestandtheile der Gneisses im Allgemeinen bilden aber noch ausserdem Magnesiaglimmer, Kaliglimmer und Hornblende, welche indessen nicht in sammtlichen Gneissen vorkommen, sondern einzeln oder zu zweien auf gewisse Abtheilungen derseben beschrankf sind.” (Lehr- buch der Petrographie, band iii, p. 185, 1894.) + Es ist schwer die Grenze gegen die letzeren (feldspath-quarz haltigen Amphiboliten) zu ziehen, aber nicht wohlgethan, Gestein mit sehr vorwaltender Hornblende zu den Gneissen zu rechnen. Namentlich ist es auch nicht zu billigen gar quarzfreie feldspathaltige Amphibolite diesen Gneissen zuzuzihlen. Ferner bedingt der Name Gneiss immerhin ein gewisses planes Parallel- gefige. (Lehrbuch der Petrographie, band iii, p. 215, 1894.) { Text-book of Geology, 3d edition, p, 186. 2? Lehrbuch der Petrographie, band iii, p. 215, 1894. 122 Cc. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. All of these principles are recognized by petrographers, though usage varies greatly as to their application in the naming of the rocks. Thus, according to some, the specific designation is based upon the prevailing mineral characteristics, while others prefix the names of other rocks with which the gneiss agrees in its textural or mineralogical characters. The latter is the prevailing usage among the Canadian geologists, who apply the term quartzite-gneiss, diorite-gneiss, etcetera, as specific desig- nations to granitoid rocks possessing a parallel structure. Some geol- ogists* restrict the terms syenite-gneiss, diorite-gneiss, gabbro-gneiss, etcetera, to rocks in which there appears a gneissoid structure due to differential movements acting upon the igneous mass in the later stages of its original consolidation. _ The desirability of greater uniformity in the method of naming the gneisses is apparent. The following slight modification of methods in quite general use is suggested as a step in this direction: 1. That the term gneiss be used in its broader structural sense for all rocks show- ing a laminated or banded structure and in which the gneissoid structure is not known to be due to differential movements of the igneous mass before its final con- solidation. For the latter astructural qualifying term may be used, as has already been done by Geikie and others, as gneissoid or banded gabbros, etcetera. 2. That where the origin of the rock (whether igneous or sedimentary) is known; the class designations be made to correspond with the character of the original rock. Thus a gneiss known to have consolidated originally as a diorite may be termed diorite-gneiss. 3. In those cases where the character of the original rock is unknown, compris- ing probably the larger part of the group, the extent of knowledge with reference to this point may be indicated by the ending ‘‘ic.’’ Thus a rock with gneissic structure and corresponding in its mineralogic composition with the diorites, but whose geological relations are unknown, may be called a dioritic-gneiss. The following table illustrates these principles as applied to a few of the more important types of rocks: f Gneiss. Analogous massive type. Of igneous origin. Origin unknown. Granite : f Granite-gneiss : , Granitic-gneiss : : . Biotite-granite. Biotite-granite-gneiss. Biotite-granitic-gneiss, q Hornblende-granite. Hornblende-granite-gneiss. Hornblende-granitic-gneiss. Syenite: Syenite-gneiss: ; Syenitic-gneiss: a> ‘ Hornblende-syenite. Hornblende-syenite-gneiss. Hornblende-syenitic-gneiss. Mica-syenite. Mica-syenite-gneiss. Mica-syenitic-gneiss. Pyroxene-syenite. Pyroxene-syenite-gneiss. Augite-syenitic-gneiss. Diorite: Diorite-gneiss : Dioritic-gneiss : ; Mica-diorite. Mica-diorite-gneiss. Mica-dioritic-gneiss. Gabbro. Gabbro-gneiss. Gabbroic-gneiss, or gabbric-gneiss. Pyroxenite. Pyroxenite-gneiss. ; Pyroxenitic-gneiss. * J. G. Goodchild: The Geol. Mag., new ser, Dec. lV, vol. [, number 1, p. 27 (Jan, 1894). +As types of the sedimentary formations, there may be quartzite-gneisses and quartzitic- gneisses. { In the absence of a better name, the term “ granite” is here used in its restricted petrographic sense. NOMENCLATURE. 123 In accordance with this view, the classification of the rock under con- sideration is obvious. Itis a gneissoid pyroxene microcline rock, which in some places is almost or wholly free from quartz and corresponds to a pyroxene-syenite, while in others not far distant the increase in the amount of quartz would ally it to the pyroxene-granites. In view of the generally sparing amount of quartz present in the coarse grained forms, they are here referred to generally as pyroxene-syenite-gneiss or simply syenite-eneiss, though it is not to be overlooked that these grade into more quartzose forms, which may be more fittingly regarded as pyroxene- eranite-eneisses. W.G. Ferrier* describes a gneiss from the Chateau Richer district apparently identical with the “streaked gneisses” described above, to which he applies the name “ pyroxene-granite-gneiss.” Lawson + describes a rock from the Rainy Lake region, which he terms hornblende-syenite-gneiss and which he states is not separable geolog- ically from others of the same region which he calls hornblende-granite- gneiss. From his description of these rocks it is evident that, except in the ellipsoidal structure and the character of the prevailing ferromagne- sian constituent, these rocks show many points of resemblance to the syenite-eneisses of the Du Lievre region. If the view held by Lawson as to the origin of the Rainy Lake rocks is sustained, however, they should be called gneissoid syenites. ORIGIN OF THE ELLIPSOIDAL STRUCTURE. GENERAL HYPOTHESES. Recognizing the intrusive character of the rock, in seeking an explana- tion of the peculiar ellipsoidal structure two hypotheses suggest them- selves : A. That it is primary and represents differentiation of the magma; or, B. That it is secondary and due to dynamic movements subsequent to solidification. ) In applying any theory to the phenomena in question we have to con- sider its competency to explain— 1. The interstitial arrangement of the augite about the ellipsoidal masses ; 2. The gradation between this and the segregated lumps and crystals in the coarsely crystallized rock on the one hand, and the parallel arrangement of the streaked gneiss on the other ; 3. The presence of the gneissic microstructure and the progressively greater granulation of the constituents as we proceed from the least to the most distinctly eneissoid forms ; * Notes on the microscopical character of some rocks from the counties of Quebee and Mont- moreney, Canada. Number I4a, p. 9. 7A. C. Lawson: Annual Rept. Geol. Survey of Canada, 1887-’88, vol. iii, part 1, p. 120 F. 124 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. 4. The transverse arrangement of the augite and manner of crystallization of all the constituents of the interstitial granular bands ; 5. The concentric arrangement of the lumps and bands about inclosed masses of apatite or pyroxenite ; 6. The occurrence of the apatite usually in a more or less crushed condition, but exceptionally in large crystals imbedded in the leopard rock; and 7. The abrupt transition between the different phases in their field_relations. PRIMARY ORIGIN. Differentiation af the coarse syenite—If, as seems probable, the coarse feldspathic rock represents very nearly the character of the rock at the time of consolidation from the original magma, it is evident that in places differentiation had proceeded to a considerable extent at the time of crystallization. The crystallization of the pyroxene appears to have been often well advanced before that of the feldspar began. A consid- erable proportion of the pyroxene, however, occurs in large grains and aggregates in allotriomorphic development with the feldspar. In seek- ing an explanation of this tendency of the pyroxene to become segre- gated, two views suggest themselves : a. That it represents a primary differentiation of the molten magma; or, b. That it is due to the fusion and recrystallization of included frag- ments of the pyroxenite. In regard to the first view (a), if this tendency on the part of the con- stituents toward segregration represents a primary differentiation of the magma, it would seem to accord better with the more generally accepted theory that differentiation has taken place when the magma was quite fluid than with that which supposes it to take place by crystallization, mechan- ical accumulation and reliquefication. According to Professor Iddings,* a study of the chemical character of rocks shows that the differentiation of molten magmas is not according to stoichrometric proportions, and is therefore not a mineralogic differentiation. As to the method by which concentration has taken place, two views have been expressed: (1) that it is due to molecular diffusion, according to Soret’s principle, advocated by Vogt,7 Brogger ¢ and others; and (2) that it is the result of liquation as advocated by Durocher § and Bickstrom. || _ If, according to Soret’s principle, the differentiation of the magma be rezarded as due to differences of temperature, then, in order to explain the lumpy segregation of the coarse grained rock, it may be necessary *J. P. Iddings: The Origin of Igneous Rocks. Bull. Phil. Soc. Wash., vol. xii, p. 152. +J. H. L. Vogt: Geol. Foren, i, Stockholm Férhand., vol. 14, May, 1891, p. 476. Reviewed by J. J. H. Teall in the Geol. Mag., Feb., 1892. Zeitschr. f. prakt. Geol., 1893, p. 272. ~W.C. Brégger: Zeitschr f. Kryst. n. Min., Leipzig, vol. 16, 1890. 2J. Durocher: Ann. des Mines, Paris, vol. Ll, 1857, pp. 217-259. | H. Backstrom: Jour. Geol., vol. 1, 1893, p. 773. HYPOTHESIS OF PRIMARY ORIGIN. 125 to adopt Brégger’s suggestion that a partial crystallization of the feld- spar set in at the same time that the segregation of the basic ingredients was taking place. Backstrom considers* that the formation of basic in- clusions by diffusion is improbable, since there can be no difference in temperature between these and the surrounding magma. The second view (0), that the basic segregations may be due to the re- crystallization of fused portions of the pyroxenite, is not improbable, though evidence of this is not at hand. The occurrence of inclosures of the pyroxenite in the syenite-gneiss is frequently observed. These, how- ever, usually retain more or less angular contours and do not appear to have suffered very much from fusion. Professor Lawson has described inclosures of hornblende schist in the Laurentian gneiss from the Rainy Lake region, which occur in sharply angular, subangular or somewhat rounded blocks, or as more or less attenuated bands drawn out parallel to the foliation of the gneiss and confused with it. He finds evidence of total or partial fusion and recrystallization to such an extent often as to admit of the deformation of the fragments and their being drawn out into lenses. He thinks that where the magma had higher temperatures the inclosures were entirely absorbed, leaving no trace of their existence except a more basic local facies of the gneiss. Differentiation of the ellipsoidal rock—On the assumption of a primary origin it may be held that the distribution of the pyroxene in the ellip- soidal rock represents— a. A molecular differentiation of the original magma, or b. That it is due to the movement of a partially differentiated magma. In considering the first view (a) two hypotheses present themselves : In the first hypothesis (1), which attempts to explain by Soret’s prin- ciple the peculiar distribution of the basic minerals in the ellipsoidal rock, there arises at the outset great difficulty in conceiving conditions that would give a difference of temperature between the feldspathic lumps and the interstitial pyroxene bands. Moreover, the presence of the gneissic structure would seem to render this view improbable. In the second hypothesis (IT), based on the liquation theory, it may be assumed that there has been a separation of the magma into layers of different composition, and that owing to some disturbance they were broken up and subsequently crystallized in the forms observed. This, however, is considered improbable from the relations of the coarse grained and ellipsoidal varieties and the absence in the former of any indications of differentiation in layers such as the theory might lead one to expect, Moreover, the presence of the gneissic structure seems to preclude any * Jour. Geol., vol. 1, p. 777. 7A. C. Lawson: Annual Report Geol. Survey of Canada, 1887-88, new series, vol. iii, part 1, p.130 F. 126 c. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. view which does not take into consideration the movement of the magma during or following its differentiation. According to the second view (0), the gneissic structure is regarded as the result of the conditions attending the intrusion and consolidation of the igneous magma. Professor Lawson considers * that he has abundant evidence to show that the granite and syenite-gneisses of the Laurentian were plutonic rocks which crystallized slowly from a thickly viscid, tough, hydrothermal magma. Up to the time of its final solidification this magma is supposed to have been subjected to differential pressures, which produced a flow in the mass to which the foliation of the gneiss is ascribed. Geikie and Teall have described + a gabbro in which there appears a banded structure almost identical with that of the Lewisian gneiss. This is regarded as inexplicable, either on the hypothesis of differentiation in situ or on that of successive intrusions, but is thought to have been produced by the deformation of a heterogeneous magma during intrusion. In applying this view of the development of the gneissic structure to the ellipsoidal rock, two hypotheses are suggested based on the assump- tions that may be made as to the character of the differentiation. If in the first of these, which should be designated hypothesis III, we assume, as in the second hypothesis, a differentiation of the magma by liquation into lumps of feldspathic material separated by thin layers of basic material, then with a gradual movement of the magma it is evi- dent that the forms would be drawn out more and more until the con- stituents became arranged in parallel bands. It is obvious that on the final solidification of the rock, a structure would result comparable in many respects with that of the leopard rock. As against this explana- tion, however, we have the following considerations: 1. The apparently cataclastic character of the gneissic structure and crystalliza- tion appearing in places. In the coarse grained rock granular bands occur as thin seams, like recemented cracks, separating the rock into coarse grained patches. These seams are sometimes too thin to be readily detected on the surface of the rock, but under the microscope show as narrow, sharply defined bands of granular microcline, fresh plagioclase feldspar and pyroxene inclosing the coarsely crystal- lized areas of microcline. These granular bands become more pronounced and regular in the ellipsoidal gneiss, where they constitute a well marked band between the feldspathic lumps, with larger grains of pyroxene distributed in a line along the middle of the granular groundmass. In the streaked gneiss the rock becomes granular throughout, though even here quite large grains of microcline often appear in the feldspathic bands. 2. The character of the coarse grained rock and the absence of any indication in it of the kind of differentiation assumed. If a regular lumpy aggregation may * A.C. Lawson: Ann. Rep’t Geol. Survey of Canada, 1887-’88, new series, vol. iii, part 1, p. 139 F. +Sir A. Geikie and J. J. H. Teall: Banded Structure of some Tertiary Gabbros in the Isle of Skye. Quart. Jour. Geol. Soc., Nov., 1894, p. 645. OBJECTIONS TO LIQUATION HYPOTHESIS. 127 occur as a result of primary differentiation, then it would frequently characterize intrusive masses, either with or without the presence of a gneissic structure. The absence of observations in support of this is proof presumptive that it does not exist. | 3. The relation of the ellipsoids and pyroxene bands to the inclosed apatite crystals and masses. Their concentric arrangement about the apatite indicates that the formation of the apatite antedated the development of the gneissic structure. The difficulties in the way of considering that the formation of the apatite deposits took place in the original magma are: a. Their large size. The deposits vary from a few inches in cross-section to several feet, and often extend many feet in horizontal and vertical directions. The crystals are frequently several inches in diameter. The larger crystals usually occur in pockets in the pyroxenite, associated with pink calcite, mica, etcetera. A crys- tal obtained at the Emerald mine, on the Du Lievre, and exhibited at the London Exposition in 1886, measured 623 inches in circumference and weighed 550 pounds.* Crystals one to two inches in diameter frequently occur in the granular deposits. In pit number 11 at High Rock, a crystal five or six inches in diameter Se Se SSSTSS , : A SO ES SS LS SUNS was observed in an inaccessible part Ps = SK SS SS ° ° e ~ x Se of the wall imbedded in the ellip- ks Se soidal rock. SIVRESA S SS es ees : SSNS Sa b. Their inclusions. Doctor Hunt SSeS S SN “~~ ™S noted rounded crystals of quartz and SS SS S . . ° ° REN. =~ carbonate of lime as inclusions in the FSO RSS SS SSS SSRN S apatite.— Similar observations were ROSSSsSSaysce made by Emmons,t while Harrington f p RSSSSLIIA AS \ states? that the apatite crystals fre- [***ssS““ss“SS~ NCR , : RASA quently inclose calcite, pyroxene, | L_] Apatite. NSSSNR Sas phlogopite, zircon, sphene, fluorspar | BSc. iite SIS S St and pyrite. While it cannot be posi- SERS tively asserted that these inclusions | Pyroxenite. ' = characterize the crystals imbedded in % the ellipsoidal rock, the correspond- FIGURE 9.—Pocket in the Pyroxenite filled with Apa- ence between these crystals and the tite and Calcite. (After Harrington.) other deposits as ue their occurrence A large crystal of apatite projects from the side favors the supposition that they had of the cavity. a similar origin. c. The rounded outlines of the apatite crystals. This feature of the apatite crys- tals of the Laurentian veinstones was regarded by Emmons || as due to partial fusion, while Hunt { considered them to be the result of the solvent action of the heated watery solutions from which they were supposed to have been deposited. The appearance of rounded angles on crystals found in cavities in the pyroxenite seems to favor the latter view. * Descriptive Catalogue of the Economic Minerals of Canada, London, 1886. + T. J. Hunt: Geology of Canada, 1866, p. 203. { E. Emmons: Geology of the First District of New York, p. 57. 2B. J. Harrington: Geol. Survey of Canada, 1877-’78, report G, p. 15. || E. Emmons: Geology of the First District of New York, pp. 57, 58. q T. J. Hunt: Chemical and Geological Essays (1891), p. 213. XVIII—Butu. Grou. Soc. Am., Vou. 7, 1895. 128 co. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. d. The crushed condition of the apatite. The smaller deposits are generally more or less reduced to a granular condition. Often the peripheral portion is granular, while the interior consists of coarse grains and fragments of crystals, with fine granular apatite filling the interstices. Hypothesis IV assumes that the heterogeneous character of the magma was due to the fusion of inclosed fragments of the pyroxenite and, owing to the differential movement of the mass, the absorbed material became distributed in the magma and crystallized in the form observed. The presence of inclosures of pyroxenite in the syenite-gneiss has been referred to. In specimen 140 a somewhat rounded, subangular mass of pyroxenite appears in contact with the ellipsoidal gneiss. In some places the line of contact is well defined, but in others it is not. Under the microscope, however, the boundary is quite sharp and marked on the pyroxenite side by an abundance of scapolite (wernerite), while no trace of this mineral appears in the ellipsoidal areas. However, the pyroxene bands of the latter, which are cut off by the line of contact, are much more prominent adjoining the pyroxenite, gradually diminishing in thick- ness away from the boundary. Another instance is seen in specimen 153, not elsewhere described. In this there appears an irregular fragment of pyroxenite inclosed in the ellipsoidal rock. The structure of the latter is somewhat obscure. The pyroxene appears more abundant in the vi- cinity of the inclosure. Indications of orographic agencies subsequent to the development of the ellipsoidal structure appear in the schistosity of the rock and the presence of narrow bands of feldspathic material and rather large plates of biotite, which extend continuously across both the ellipsoidal rock and the pyroxenite. According to this hypothesis the apatite may be considered to have been derived from the pyroxenite and become more or less crushed and drawn out before the rock became wholly solidified. This hypothesis appears to be sustained by a number of considerations. ‘The greater abundance of pyroxene in the vicinity of tie inclosed masses of pyroxenite seems to indicate a partial absorption of the latter. The occurrence of interstitial pyroxene in greater abun- dance in certain areas in the ellipsoidal gneiss may indicate the position of smaller inclosures which have been entirely absorbed. Moreover, the character and condition of the apatite inclosures seem to favor this view. On the other hand, however, difficulties are encountered when we attempt to explain by this process— The shell-like distribution of the absorbed pyroxenic material prior to solidification, and The apparent cataclastic character of the gneissic structure. That a more or less completely banded structure may be produced in HYPOTHESIS OF SECONDARY ORIGIN. 129 this way, as claimed by Lawson,* is probably true, but whether the basic material may be so distributed as to crystallize out in the form shown in the ellipsoidal gneiss is doubtful. While the argument drawn from the character of the gneissic struc- ture is not beyond question, it is strongly presumptive against the primary origin of the ellipsoidal structure. SECONDARY ORIGIN. In attempting to explain the ellipsoidal structure by appealing to dynamic processes (hypothesis V), it is necessary to consider (1) the character of the original rock and (2) the agencies and their mode of operation. The character of the original rock, as inferred from the coarse grained variety, was evidently peculiar. Its heterogeneous character is indicated by the tendency of the feldspar, pyroxene and quartz to appear in large grains, lumps and segregations. In some cases, however, the rock has an even, granular structure and consists of feldspar and pyroxene quite uniformly distributed. In places the rock is mostly coarse feldspar, with scattering grains and small aggregates of pyroxene. In other places the pyroxene occurs in much greater amount, both as crystals of various sizes and aggregates of grains, apparently constituting from 15 to 20 per cent of the rock. In general the irregular pyroxene aggregates vary from the size of millet seeds up to marbles, rarely larger. They usually in- close more or less feldspar in small grains; also pyrite and titanite. In areas showing a large amount of pyroxene the latter sometimes appears in idiomorphie grains, of large size, inclosed inthe microcline. They also appear at times in the quartzareas. In thespecimens showing a uniform distribution of feldspar and pyroxene, idiomorphic forms of the latter were not seen. While the arrangement and mode of crystallization of the constituents seem to indicate partial differentiation of the original magma, it is not improbable that some of the pyroxene was derived from the inclosures of the pyroxenites, as already suggested. The agencies which may be appealed to as effecting the changes ob- served are those usually classed under the head of dynamic meta- morphism. These may be conceived to have effected the crushing of the coarse grained syenite, accompanied by solution, recrystallization and rearrangement of the constituents under the influence of water, and probably also of heat. The variation in the extent of the structural alteration may be due to the different attitudes of the inclosing rocks. It will scarcely be denied that certain portions of the intrusive mass. * A.C. Lawson: Ann. Rep’t Geol. Survey of Canada, 1887-88, new series, vol. iii, part 1, p. 134 F. 130 co. 8. GORDON SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. would be called upon to sustain a much greater pressure than others, while some parts might escape almost wholly. With the breaking of the rock under the influence of strain there would be more or less pul- verization of the minerals along the sides of the crack. This would favor the chemical action of percolating waters. With the recrystallization of the constituents taken into solution the crack would become healed. This may be conceived as taking place (1) by partial solution of the powder, with a secondary enlargement of the remaining portions; or (2) by complete solution and recrystallization of the pulverized material. In either case the filling would be fine grained, since even if wholly re- crystallized the process would probably go on synchronously with the movement of the rock. Evidence that this process has taken place at a later date is sometimes seen in the presence of healed cracks cutting across the banding. With increasing pressure the rock may be reduced to a coarsely frag- mental condition, and if the process were stopped at this stage the result would probably be a mass of irregular fragments, cemented together by fine grained interstitial material. The pyroxene within reach of the percolating waters would be dissolved to a greater or less extent, and on subsequent recrystallization would appear in grains marking the spaces occupied by the solutions. The relative solubility of pyroxene and feld- spar under the conditions here postulated is unknown. ‘The relative effects of weathering, however, are well shown by the manner in which the pyroxene decays on exposed surfaces of the ellipsoidal rock, leaving sharp trenches surrounding the feldspathic portions. Since basic min- erals melt at somewhat lower temperature than the acidic, it may be supposed that the temperature of the rock became sufficiently great to melt the pyroxene, but not the feldspar.* While the convoluted forms assumed by the ellipsoids are suggestive of plasticity, it is scarcely prob- able that the heat has ever reached the point indicated. This is inferred (1) from the occurrence occasiona’ly of crystals of augite, which are ap- parently original, inclosed in grains of microcline in the ellipsoidal areas; and (2) from the consideration that if the heat is due to the shearing movement of the rock, as generally conceived, it must be generated slowly, and hence would probably be dissipated nearly, if not quite, as rapidly as it is produced. As the ellipsoids became more and more flattened the interstitial pyroxene bands would assume a parallel arrangement, as in the case of other similar gneissoid structures. In support of the hypothesis of dynamic origin we note: * J. G. Goodchild: Geol. Mag., new series, Dec. IV, vol. 1, no. 1, Jan., 1894, p. 23. + F. Zirkel: Lehrbuch der Petrographie, band iii, p. 205. EVIDENCE SUSTAINING DYNAMIC HYPOTHESIS. 131 1. The gneissic character of the microstructure as traced from the coarse grained through the ellipsoidal to the streaked varieties. The augen-structure represented in specimen number 160 is clearly typical of the results usually credited to dynamic movements. Inasmuch as this appears to be merely a special development of a portion of the rock having little or no augite, it would not seem unwarranted to infer that the agencies operating here affected other portions of the rock as well, and hence that the development of the gneissic structure was, in both cases refer- able to the same agencies. 2. The evidence of recrystallization of the granular areas. This appears not only in the irregular interlocking outline of adjacent grains, but the frequent pres- ence of fresh plagioclase feldspar as a cement for the other constituents and the development of hornblende in idiomorphic forms in the more gneissoid rock. 3. The indication of dynamic action as shown (a) by the fracturing of the augite and microcline grains, and (0) the indication of strain appearing in the bending of the microcline lamelle, accompanied by undulatory extinction and the develop- ment of polysynthetic twinning of the pyroxene. These phenomena are chiefly confined to the coarse grained rock, and even here are not pronounced. It may be that some of them are due to movements subsequent to the development of the eneissic structure. The appearance of strain in the microcline, however, in certain cases (page 116) seems rather to be connected with the production of the gneissic structure. This is true also in certain cases in the breaking of the pyroxene grains, but in general neither pyroxene nor microcline furnish any evidence as to what proportion, if any, of the granular materials represent fragments of the orig- inal minerals. 4. The character and relations of the inclosed apatite. The significance of these deposits lies in their generally crushed condition. The crushed appearance of the apatite is more pronounced, so far as the limited data at hand shows, in the de- posits occurring in the streaked gneiss than in those in the coarser rock. In places where the gneissic structure of the rock is imperfectly developed, the crushing of the apatite likewise appears incomplete. The mass is made up of rather coarse fragments, with granular apatite filling the interstices. In some cases (page 107) the mass is intersected by thin seams of feldspathic material extending in from the walls and continuous with the surrounding layer. Often the deposit is reduced almost wholly to a granular condition. These granular deposits, called sugar apa- tite, are sometimes of considerable extent, as noted by Professors Penrose * and Harrington. t Professor Penrose says: ‘‘The granular variety known as sugar apatite is of a white or pale green color and looks like coarse sand, more or less coherent. * * * It is one of the purest forms of apatite mined. It is uncertain what could have caused the apatite to assume this granular condition.”’ Professor Harrington states that *‘ though at some localities the apatite occurs chiefly in crystals, at others it is wholly or almost altogether massive, varying from compact or crypto-crystalline to coarse granular. Frequently also it exhibits a dis- tinct lamellar texture. A friable saccharoidal variety is very common and often termed ‘sugar phosphate.’ When white it is sometimes difficult to distinguish by the eye from some forms of quartz sandstone. * * * Crystals are sometimes imbedded in this granular apatite, and frequently also rounded masses of apatite *R. A. F. Penrose: Bull. 46, U.S. Geol. Surv., p. 38. 7 B. J. Harrington: Geol. Survey of Canada, p. 14 G. 132 Cc. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. up to many inches in diameter. Similar masses of pyroxene as well as crystals are also sometimes imbedded in the apatite.’’ It is also worthy of note that the form of the deposits in both gneiss and pyrox- enite is strongly suggestive of crushing and squeezing. This was remarked by Harrington,* who says: ‘‘The apatite masses look as if they had been driven or squeezed into the curious forms which they now present during the folding or crumpling of the enclosing rock.”’ The explanation of the large apatite crystal (pages 103, 127) in the leopard rock is not without difficulty. One or two instances only were observed and we have no information that they ever occur in the streaked gneiss. From the manner in which the ellipsoids and bands of the gneiss arrange themselves concentrically about the apatite, it is clear that the crystallization of the apatite took place prior to the development of the eneissic structure. Four views may be suggested to account for the apatite crystal in its present position : 1. That it crystallized out of the original magma before the solidification of the latter. From the relations of both the crystallized and granular deposits of apatite occurring in the syenite-gneiss, it is evident that an explanation that will account for the presence of one must apply also to the other. We have considered already the objections to the view of the formation of these deposits in the original magma (page 127). These are the large size of the crystals, their inclusions and rounded outlines, and the extent and more or less crushed condition of the deposits. 2. That it is due to segregation and crystallization after the solidification of the rock. The belief is expressed by Harrington in the report above cited that in many cases there has been a segregation of apatite and other minerals which ac- company it from the surrounding rock into irregular or lenticular masses without any true cavity or crevice having ever existed. The growth of crystals by replace- ment in situ has been noted by various observers. ft Indications of such development appear in the small hornblende erystals occur- ring in the streaked gneiss, as described in another part of this paper (page 118). The development of large crystals by this process, however, has not been demon- strated and is considered improbable. 3. The third view is that they were deposited in a cavity in the syenite prior to the development of the gneissic structure. The objection to this hypothesis lies in the difficulty of accounting for the obliteration of the cavity without crushing the apa- tite. It may be supposed that the cavity was large in proportion to the size of the crystal and did not become closed at once but gradually, and that by the time the walls had closed in, the surrounding rock had become sufficiently plastic to adjust itself about the crystal without breaking it. The objections to the view that the rock had reached such a condition of plasticity have been considered (page 130). Moreover, the former presence of a cavity should be indicated by an irregularity in the arrangement of the ellipsoids about the apatite, which does not appear to be the case, though the observations upon this point were not conclusive. * B. J. Harrington: Geol. Survey of Canada, Report G. p. 7. + C. R. Van Hise: Am. Jour. Sci., 3d ser., vol. 33, 1887, p. 385. CONCLUSION. 133 4. The fourth view, and probably the correct one, is that the apatite was formed in cavities in the pyroxenite, and that it became inclosed in the syenite during the intrusion of the latter. The objection to this view, as in the preceding, lies in the difficulty of accounting for the preservation of its crystallographic form. However, as the ellipsoidal rock evidently represents an early stage in the development of the gneissic structure, it is probable that the differential movement at this point was not so great but that the crystal was able to withstand the strain. SIMILAR STRUCTURES IN OTHER ROCKS. Analogous structures in mica-gneisses are well known. A variety called stengeliger-eneiss (wood-gneiss), according to Zirkel,* is characterized in some cases by bands of mica winding about the stalk-shaped or wreath- shaped feldspar-quartz masses, so that these are inclosed on all sides by the mica layers. On cross-fracture, therefore, these will show discoidal, ellipsoidal, elongated, roundish, trapezoidal figures formed by the mica bands. Up to this point the description apples very well to the Ottawa occurrences, except in the character of the mineral forming the bands. Beyond this, however, the analogy fails, as further alteration produces an asbestos-like structure which has no counterpart in the rocks under consideration. Rothpletz has noted f a structure analogous to that here described in the greenstone-schists (actinolite-schists) of Hainchen, and Williams ¢ from northern Michigan. Both of these observers ascribe it to breccia- tion in situ, while the former explains the rounded character of the frag- ments and the production of much of the interstitial material by the rubbing together under the action of much orographic pressure of a mass already finely subdivided by cracks. Lawson has described similar occurrences in the Lake of the Woods region. § CONCLUSION. Reviewing now the different hypotheses in the light of all the evidence available, it is apparent that no one of them seems to offer a full and adequate explanation. The ellipsoidal and the gneissic structure in these rocks are clearly closely related in origin, and any conclusion affect- ing the one has a direct bearing upon the other. Our knowledge of the processes by which the ancient gneisses were formed is extremely limited. That they may be formed by dynamic processes has long been recog- nized, and itis now well established that a laminated structure com- parable to that of the gneisses may be produced in deep-seated igneous * Lehrbuch der Petrographie, band iii, p. 203. 7 Zeitschrift der deutsch geology Gesell., vol. 31, pl. ix, x, 1879, pp. 374-397. ft Bull. 62, U.S. Geol. Survey, 1890, pp. 166-177. ¢.Geol. and Nat. Hist. Survey of Canada, Ann. Rep., 1885, Rep. CC, p. 51. 134 Cc. H. GORDON—SYENITE-GNEISS (LEOPARD ROCK) FROM CANADA. rocks as the result of the conditions attending their intrusion. In conse- quence of the constitutional changes which appear to take place in rocks when subjected to great orographic pressure, it becomes in many cases extremely difficult, if not impossible, to distinguish between the two kinds of gneisses. Moreover, our knowledge of the cause of differentia- tion of igneous magmas is as yet little more than a speculation, but that the nature of this original differentiation conditions in part the character of the structure resulting from subsequent processes is obvious. While in many respects incomplete, involving, as it does, much that is as yet little understood in the metamorphism of rocks, on the whole the evidence seems to favor the last hypothesis (V), namely, that of dynamic metamorphism. Briefly summarized, this hypothesis supposes— 1. That the structure characterizing the leopard rock is due to orographic agen- cies and represents an intermediate stage in the development of a streaked augite- syenite-gneiss out of an augite-syenite which was distinguished by a coarsely crystallized structure and by a somewhat irregular aggregation of pyroxene. The character of the original magma may have been modified somewhat by the absorp- tion of included fragments of pyroxenite. 2. That the distribution of the pyroxene has been efteciads presumably by the solution of portions of the original constituents and their recrystallization along lines marking the location of the cracks. 3. That with continued pressure these lumps have been more and more drawn out, the process being accompanied by recrystallization until the rock assumes the streaked gneissoid form. While in general the evidence of crushing is rendered more or less doubtful by the extent of recrystallization, in one case (number 159) it is undoubted. This represents a partly developed gneiss in which, however, the ellipsoidal structure is but imperfectly presented. Though differing considerably from the rest of the rock in general appearance, there is little doubt but that it belongs to the same mass. The ground- mass is much finer than in any of the others and shows much less the effects of recrystallization. Large grains of microcline are partly crushed and drawn out into lens-like forms comparable to the augen-structure. A significant feature is the arrangement of the augite in irregular bands about the feldspar and quartz. In some of the augite grains the appear- ance of crushing and dragging is pronounced, while the general appear- ance of the rock, both in hand specimens and under the microscope, admits of no other conclusion than that its present structure is due to the effects of orographic pressure. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 7, PP. 135-170 JANUARY 10, 1896 STUDIES OF MELONITES MULTIPORUS* BY ROBERT TRACY JACKSON AND THOMAS AUGUSTUS JAGGAR, JR (Read before the Society August 27, 1895) CONTENTS. Page Introductory........ Ee a NY ese Mes Whe i enero te tieta ora atay 0S Geld 135 HDS SGIYOTOIO OLE SPOROAVST crete ce pric Men oe nc Pan ram ge are eee 136 Arrangement and development of the ambulacral and interambulacral plates. 138 isimucture and development of the ambulacrum.............-... ..-seeceees , 140 Structure and development of the interambulacrum.................020ee eee 142 Wawanons im the interambulacrum 5.2.2... 5. see eee eee cee eee eet oe ew eect 151 Man omuntionrOl Plabesae tev. vecgs os sere cine foe tke wee alee bade sales cieinee 154 Genital and ocular plates and discussion of orientation...................64, 155 Tabulations of plate arrangement in the interambulacral and ambulacral areas. 156 Remarks on the tables of plate arrangement.....................6. BS eA 157 pcos ave aAMEAMIOCUNCMIG: cies sce so silo ne eee ees ab gles oe gee ete Ree ste 165 INTRODUCTORY. The remarks in this prefatory statement will serve as an introduction for the present paper on Melonites multiporus under joint authorship, and also for the succeeding paper, “‘ Studies of Paleechinoidea,” by myself. Deep obligations are due to Mr Alexander Agassiz for the opportunity of studying the rich collections of Paleozoic echinoderms in the collections of the Museum of Comparative Zodlogy at Cambridge. Besides an ex- tensive series of Melonites multiporus, this museum possesses a number of generic and specific types and many rare species, as described in the following pages. 3 For the opportunity of using material, obligations are also due ‘to . Professor R. P. Whitfield, of the American Museum of Natural History in New York; Professor C. E. Beecher, of Yale University Museum ; Mr C. Schuchert, of the United States National Museum; Professor A. Hyatt, of the Boston Society of Natural History ; Professor William B. *Plates and descriptions of the same, also list of references quoted, will be found at the end of the succeeding paper: Studies of Paleeechinoidea. References are indicated by a figure in paren- thesis after names of authors cited. XIX—Butt. Geon. Soc. Am., Vou. 7, 1895. (135) 1386 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. Clark, of Johns Hopkins University ; Mr C. W. Johnson, of the Wagner Free Institute, and Professor W. B. Scott, of Princeton.* Similar privi- leges were extended to Mr Jaggar by Dr E. Newton, of the Museum of Practical Geology, in Jermyn street, London. To my friend, Professor Beecher, I am indebted for critical suggestions. Professor Whitfield and Mr Johnson very kindly had drawings made for me. Sincere thanks are due to Mr J. H. Emerton, of Boston, for the extreme pains he took in making the drawings for the plates. Each individual plate in a specimen was carefully scrutinized to ascertain its relative size and angles, all peculiar plates, the position of columns, etcetera, were measured, and all were drawn with great accuracy. No portions of specimens as figured are restorations except in the cases where specially stated and indicated by dotted lines or shading in the figures. Mr Jaggar, while working with me as a student in the Geological De- partment of Harvard University, made the very important discovery of the regular arrangement and progressive introduction of plates in the inter- ambulacral area of Melonites multiporus. His observations were based on the specimen illustrated by plate 3, figure 12, and plate 4, figure 18, which was at that time in the Student Paleontological Collection. Mr Jaggar compared this arrangement with that of Oligoporus dane, as shown in Meek and Worthen’s figure, introduced here as figure 34 of plate 6. He also studied the spines of Melonites multiporus. Mr Jaggar had not the time nor the opportunity to carry the results of his observations further. In London, at the Geological Museum in Jermyn street, he made draw- ings of Palxechinus, which are reproduced in plate 7, figures 38 and 39. I have carried on the studies of the interambulacrum of Melonites multi- porus, extending it by observations on other specimens. I have also added the studies of the ambulacral area in this species. In the suc- ceeding paper, “Studies of Paleeechinoidea,” the researches begun in Melonites multiporus are extended to other species and genera. Rospert Tracy JACKSON. DESCRIPTION OF SPINES. The spines of Melonites multiporus,t Norwood and Owen, have not been previously described, though their general character has been sug- gested by analogy to related species and genera and by the minute sur- * Many of the specimens described came from Ward’s Natural Science Establishment at Roches- ter, New York. +Asa matter of information worthy of record, it is stated where the types of Paleozoic echi- noids described in this and the sueceeding paper are deposited. The specimens of Melonites multiporus figured in the Illinois Geological Survey, vol. ii, pages 227 and 228, are in the A. H. Worthen collection, which is now in the University of [llinois, at Urbana, Illinos,as I am in- formed by Mr William F. E. Gurley, state geologist of Illinois. These specimens and many types are included in a published list of the Worthen collection. DESCRIPTION OF SPINES. 137 face tubercles which thickly cover the plates of the test in occasional well preserved specimens. The spine tubercles of this species were figured by Roemer (36), and are present in great numbers on the specimen from which plate 4, figure 12, was drawn, and by careful examination these tubercles can be seen in portions of the photographic figure (plate 5, figure 18). They served as bases of articulation for hundreds of small spines which covered thickly both the ambulacral and interambulacral areas. The specimen showing spines from which plate 2, figure 1, was drawn is a single individual of Melonites multiporus from the Saint Louis group, Subcarboniferous of Saint Louis, Missouri. It ig in the collections of — the Museum of Comparative Zodlogy (catalogue number 2988). The specimen is of the usual type from this locality ; close inspection, how- ever, reveals scattered over the test, but more especially in the protected furrows and depressions, innumerable minute spines clinging to the plates of both the ambulacral and interambulacral areas. They are most abundantly and best preserved in the longitudinal furrows formed by the depressed, irregular, lateral plates of each ambulacrum. A cluster of these spines is shown in plate 2, figure 1, magnified 6 + diameters. The well preserved spines are cylindrical, somewhat swollen at the base, and gradually taper to the distal point. Occasional very well preserved spines, however, show surface ornamentation, which consists of succes- sive faint swellings, interrupted by constrictions, that give to the spine a beaded appearance, as shown in the figure. This surface ornamenta- tion was first observed by Mr Westergren when drawing the accompany- ing figure. The dimensions of the spines are, average length 3 milli- meters, maximum thickness at the base 0.4 of a millimeter, and from this tapering to a somewhat blunt end at the distal terminus. The spines are frequently broken or ‘reduced in size by erosion; but otherwise all have approximately the same length on both ambulacral and interambulacral areas. On the interambulacra they are so much more exposed that by far the greater part of the spines are short and stubby from erosion. Only occasionally a long spine equalling the length of those in the ambulacra is found ; but there are enough of these to fully warrant the statement that the spines of both areas were of the same length. None of the spines of the interambulacra were seen show- ing the bulgings described, but this is attributed solely to erosion, for such ornate spines were only found in well protected areas. Besides the specimen figured, another specimen of Melonites multiporus in the Museum of Comparative Zodlogy (catalogue number 2996) shows spines. The features exhibited do not differ from those just described. Spines have also been observed on two other specimens. Hambach (18) has published a photographic figure of Melonites crassus, 188 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. Hambach, which shows numerous spines.’ In the description it is stated that the ambulacral areas are covered with little spines about one-eighth of an inch in length, while the spines of the interambulacra are only one- half as long. This peculiarity Professor Hambach refers to as one of the points of specific distinction from Melonites multiporus, which indicates that he was familiar with the spines of that species. This distinction of the two species is noteworthy, inasmuch as M’Coy’s (29) distinction of his family of the Paleechinoidea is based largely on the uniformity of the spines as distinguished from the two series present in the Archeocida- ride. As stated above, only a very few spines of the interambulacra of Melonites multiporus were found of the same length as those in the am- bulacra. Itis possible, therefore, that in Professor Hambach’s specimen of Melonites crassus all the spines of the interambulacra were so eroded as to give the impression that they were really shorter than those of the ambulacra. This could easily happen, and it is the only species of the Melonitide in which such a difference is noted. Besides the foregoing, the spines of American Melonitidee have been described by Messrs Miller and Gurley (84) in Melonites indianensis, Miller and Gurley, and Oligoporus bellulus, Miller and Gurley. In the succeeding paper the spinés of Oligoporus dane are described (plate 6, figure 32). ARRANGEMENT AND DEVELOPMENT OF THE AMBULACRAL AND INTERAM- BULACRAL PLATES. The next subject to which attention is called is the regular arrange- ment and method of introduction of the plates of the ambulacral and interambulacral areas in the Melonitide. The arrangement of the inter- ambulacra has been partially figured in Oligoporus danx, M. and W., and appears fragmentarily in a number of published figures of Melonites, Pa- lxechinus, and some other genera of Palzeechinoidea, but its significance and true relations seem to have been quite overlooked. Roemer (36) supposed that the intercalated columns terminated in much the same way at eitherend. Hesays: “ Die Vermehrung der Reihen von den Polen gegen die Mitte der Schale hin geschieht durch allmah- liches Einsetzen neuer Reihen zwischen die vorhandenen.” Lovén (25) says of the Perischoechinoida that the adambulacral plates alone extend from the peristome to the dorsal pole, which observation is entirely supported by the results of our studies. Miller and Gurley (84) consider briefly the method of growth of Oligo- porus blairi, Miller and Gurley. They express tentatively the belief that the number of columns of interambulacral plates does not increase with ARRANGEMENT AND DEVELOPMENT OF PLATES. 139 growth, but their view as expressed is quite the opposite of our conclu- sions. This view was based on 38 specimens studied, all of which they state have 6 columns of plates. The answer would be that while the num- ber of columns apparently does increase with age, the full number may be attained quite early in the life of the individual.* In such a case a con- siderable increase in size may take place without the addition of any new columns. These are the only statements seen in regard to the ar- rangement of interambulacral plates in Paleechinoids, except the general remark frequently met with, that the columns (ranges) of pases fail to reach the poles. Before describing the plate arrangement in detail. it may be well to state the case in brief. Echinoderms grow by the addition of new plates to the corona around the abactinal area and by the increase in size of previously formed plates. As the new plates of each ambulacrum or in- terambulacrum are formed they are inserted between previously formed plates of the area and the genital or ocular plates (see plate 8, figure: 13) ; therefore those plates on the lower part of the test were the earliest formed, and passing dorsally we get progressively the later and later added plates, built as the individual grew aborally. In Melonites multiporus, at the oral termination of the corona and in immediate contact with the peristome, each ambulacrum has a row f+ of four plates and each interambulacrum a row of two plates { (plate 2, fig- ures 2 and 3). Passing from this part progressively upward, or dorsally, we find an increase in the number of columns in both the ambulacra and interambulacra. In the ambulacra the two outer plates at the base, a, 6, with additions dorsally, form the two lateral columns of the area, and the two median plates at the base, a’ b’, with additions dorsally, form the two median columns of the area (plate 2, figure 4, and plate 4, figure 18). In the interambulacra new plates are added to form each column, and also new columns are added with great regularity (plate 2, figure 2; plate 3, figure 12; plate 4, figure 18). New columns are added rapidly at first, and they attain their greatest number at a point of some- what variable distance above the ambitus (plate 4, figure 18). Near the anal area the number of columns of interambulacra decreases as well as at the oral area, but for a different reason, as discussed. In the figures the several columns are numbered progressively, correspond- ing to the number of columns attained by the individual, and therefore * See figure of Palwechinus gigas, plate 7, figure 38. + The term row in this paper is limited in use to series of plates lying in one horizontal plane, in contradistinction to the term column, which is applied to series of plates lying in a vertical plane. t The structure and development of the ambulacrum and the position and development of the first three interambulacral columns, numbers 1, 2, 3, plate 2, figures 2, 3, etc., as described in this paper, were worked out by R. T. Jackson. 140 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. will be readily followed. It is to be observed that successive columns, passing dorsally, are typically alternated to the right and left of one another (plate 3, figure 12; plate 4, figure 18). This rule has some ex- ceptions, principally in even-numbered columns, but occasionally in odd ones. It will be seen that the plates of the adambulacral columns are pentagonal, while all the other plates of the interambulacral areas are hexagonal, with the exception of the ventral terminal plates of columns which are pentagonal* and adjacent plates which are in most cases heptagonal; also near the anal area the plates, instead of being hexag- onal, are more or less rhombic in form (plate 3, figure 13). The arrangement as sketched above has been traced in more than 100 specimens of Melonites multiporus, and in all the same method of intro- duction and growth is maintained, with only such slight variations as are discussed. A similar arrangement has also been traced to a greater or less extent in Rhoechinus, Palecchinus, Oligoporus, Lepidechinus, Lepi- docentrus, Lepidocidaris and Archxocidaris, and in a published figure of Lepidesthes, as described in the succeeding paper. It may be stated that no conflicting evidence has been found in any Paleozoic or recent type. STRUCTURE AND DEVELOPMENT OF THE AMBULACRUM. Turning to the detailed description of plates, the oral area will be con- sidered first. Specimens showing perfectly the oral termination of the corona seem to be rare, for that area has never been adequately de- scribed, and most specimens which are at first sight complete are want- ing in one or more rows of plates. A specimen showing this area satis- factorily is a Melonites multiporus, Norwood and Owen, from the Saint Louis group, Subcarboniferous of Saint Louis, Missouri.t The specimen is in the collections of the Museum of Comparative Zodlogy (catalogue number 3000), and is represented in plate 2 by figure 2. The ambulacra ventrally in each area consist of a row of four plates which lead to per- pendicular columns by the addition of plates dorsally. The plates at the ventral termination as well as the later added ones have two pores. In each area, as also observed on many other specimens, the two outer initial plates, a 6b (plate 2, figure 4; plate 4, figure 18), become the bases of the two lateral columns of the ambulacrum, and the two middle initial plates, a’ b’, become the bases of the two columns of large median ambu- lacral plates. At the ventral area all the ambulacral plates are of about the same size and shape, whereas further up the plates of the two central columns, a’ b’, are much larger than the plates of the lateral columns * Excepting the initial plate of column 3. See piate 2, figures 2 and 3. yAll the specimens of Melonites multiporus described in this paper are frogn the same formation and locality; therefore no further details on this point will be given. STRUCTURE AND DEVELOPMENT OF THE AMBULACRUM. 141 of the area; the central columns of plates also are quite regular in size and form, whereas the laterals are highly irregular, as shown in plate 2, figure 4; plate 4, figure 18. The curvature of the surface of the ambu- lacra, which gives the melon-like ribs, is not existent at the ventral area, This and the other characteristic generic features of this area were not developed in that portion of the test which lies near the peristome. Besides the four columns, a, a’, b’, b, which are found already at the peristomal border, new columns of ambulacral plates are added dorsally as the animal grew. The new columns are added_in the two lateral depressed furrows of each ambulacrum, and in each case between the lateral and median columns of the side to which it belongs, as seen in plate 2, figure 4, and plate 4, figure 18. In our specimen, as shown in the figure (plate 2, figure 4), the first new columns added, ¢ and d, origi- nate at about the same horizon in each half ambulacrum. The fourth columns of each area as added, e and f, again originate at the same point in the two sides of theambulacrum. The number of columns in the am- bulacra increases rapidly, but the plates of the ambulacral area are so irregular and usually so imperfectly defined that they are difficult to make out, and only a few specimens have been seen in which the devel- opment of this area could be satisfactorily traced. Besides figure 4, the development of the ambulacrum and the introduction of its newly added columns are shown well in areas B, D, and F of figure 2, plate 2. A section of the ambulacrum of Melonites multiporus (plate 2, figure 5) shows the relative size, thickness and position of the several columns of ambulacral plates. The specimen is quite normal in form, not being at all distorted. The adambulacral plates J J of adjacent interambulacral areas project under the lateral plates a b of the ambulacrum, not over them, as they do in Lepidocentrus. The median plates a’ b’ of the area are very thick, as well as large in other proportions. The pores in the ambulacral plates on the surface or distal side le in that portion of the plate which is nearest the interambulacrum (figures 4 and 5), but in tray- ersing the thickness of the plate they extend inward or toward the center of the area, as shown in figure 5. In this section plate XY was cut in such a plane that the pores did not show. Some of the pores in the section are seen passing quite across the plate. When their whole extent is not visible, their probable position is indicated in the figure by dotted lines. The position which new columns take as introduced in the ambulacra in relation to the initial columns a a’ and 0’ b is important, especially in connection with the condition seen in Oligoporusas discussed in the next paper under the consideration of that genus. The four initial columns of Melonites can properly be homologized with the four columns seen in adult Oligoporus (plate 6, figure 30), and this is important, because the 142 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. difference of the number of columns is the basis of separation of the genera. As Melonites ventrally has ambulacra like adult Oligoporus, and as adult Melonites has much more complex ambulacra (plate 5, figure 20), it may be considered that Oligoporus is the more primitive genus and Melonites the more specialized, being further advanced in the special line of variation of the family. Melonites may be considered as highly accelerated, for at an early period in development where Oligoporus (plate 6, figure 25), has but two plates in the ambulacrum, Melonites has four. The four initial columns of Melonites, a a’ and b’ b, are together equal to the two columns of plates of less specialized genera, as the primitive Silu- rian type Bothriocidaris, and also Archexocidaris, Cidaris, etcetera. This relation is discussed in the succeeding paper under Oligoporus (‘““Arrange- ment and Development of Plates in Oligoporus coreyi”), and is expressed in a diagram, figure 1, inserted in the text at that place. The late honored Professor Sven Lovén (27) has shown in modern echi- noids that a considerable number of the ventral plates of the ambula- crum are resorbed during the progressive enlargement of the actinostome. More or less of this resorption has also probably taken place in the growth of Melonites. These missing plates could only be obtained in very young specimens. If present, they might show that the ambu- lacrum in its inception had only two plates in a row, instead of four. This would be in accordance with the earliest stages of Oligoporus, of modern echinoids, and also with the condition of the adult in the ancient primitive type Bothriocidaris and most members of the Paleeechinoidea. STRUCTURE AND DEVELOPMENT OF THE INTERAMBULACRUM. The interambulacra of the adult, when perfect, consist of two plates at the ventral termination, as shown in three areas, A, Cand J, of plate 2, figure 2. These plates, numbers 1 and 2, are pentagonal in form, having a straight line on the ventral aspect. It is to be noted that the angles of these plates do not correspond to the angles of terminal pentagons in later added columns, as seen in this and other figures. These two first ventral plates give rise, by the progressive addition of plates dorsally, to the two outer columns of pentagonal adambulacral plates. In two areas, and G of our specimen (plate 2, figure 2), there are three plates at the ter- mination of the interambulacra. It will be observed that these plates have angular faces ventrally, corresponding to the angles of the ventral border of the second row in an area terminating ventrally in two plates (see also plate 2, figure 3). It is evident, therefore, that the lower row of two plates is absent from mutilation in these cases, which at first sight apparently terminated in three plates. This is an important point, for STRUCTURE OF THE INTERAMBULACRUM. 148 quite frequently specimens of Melonites multiporus are seen, showing three plates ventrally. Such areas are to be considered incomplete ventrally, and when well preserved show at the oral termination angles for the articulation of the lowermost missing row. Another evidence of the limits of the area ventrally is a line, M, of thin, dark colored, calcareous tissue, which runs around the outer border of the peristome, coinciding with the ventral edges of the plates. This apparently is the ventral border of the perignathic girdle, or the auricles and other perpendicular calcareous supports on the edge of the peristome for the attachment of buccal muscles. ‘This line of tissue stands out clearly in the open space if the lowest row in any area is wanting, as between the areas C to H, in- clusive (plate 2, figure 2). This perignathic girdle has only been observed in thisspecimen. Besides the specimens figured, the ventral termination of the interambulacra in two plates has been seen in a number of other specimens of Melonites multiporus, amounting to 21 areasinall. Inspeci- men number 3003, in the collections of the Museum of Comparative Zoology, there are two plates at the ventral termination in all five inter- ambulacral areas, as shown in plate 2, figure 3. These plates are suc- ceeded by three plates in the second row. The same specimen shows well the ventral termination of the ambulacral areas as seen in the figure. Two specimens from the Boston Society of Natural History (catalogue numbers 229A and 229B), also specimens in the EK. M. Museum at Prince- ton (catalogue numbers 1464 and 1466), show two plates ventrally in several areas. This fact of two ventral plates is important because of the similar number of columns of plates in the Kuechinoidea. It is also important as compared with the number of plates seen ventrally in other Paleozoic echinoids (Oligoporus, Lepidechinus, Archxocidaris). Its special bearing is the fact that it shows the limits of the encroachment of the peristome on the corona of this genus, which may be taken as the type of the family.* The extent of encroachment varies in different types from not at all, Bothriocidaris, Lepidechinus, to an extensive encroachment, Archeoci- daris, Cidaris (see figures 42, 43, 44, 48 and 55 in the plates). The only published observations seen on the extreme ventral area of Melontesis Meek and Worthen’s (80) figure of the ventral aspect of Melo- nites multiporus, showing the jaws in place. In the text they do not de- scribe or discuss the plates, but in the figure the five ambulacral areas show four plates at their ventraltermination. One of the interambulacra has two plates ventrally, three have three plates, and one has five. This discrepancy with our observations is unquestionably due to imperfections of the specimen figured, not to variations in this important feature. * See classification at end of succeeding paper: Studies of Paleechinoidea. XX—Butt. Grou. Soc. Am., Vou. 7, 1895. 144 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. Before describing further the plates of Melonites it is desirable to con- sider whether the lower row of two plates, numbers 1 and 2, plate 2, fig- ures 2 and 8, really represent the first formed or initial interambulacral plates. Professor Sven Lovén (27) has shown that at the ventral border of the interambulacra in the young of Goniocidaris* and Strongylocentrotus there is but a single plate which in the next row is succeeded by two plates (plate 3, figure 8). At the same period of growth the ambulacra have two columns of plates as in later stages. Of this, Professor Lovén says, “This is the normal constitution of the peristome in the whole class.” He shows that during later growth with the enlargement of the actino- stome that area commonly encroaches upon the corona, and the initial plate is gradually resorbed (plate 3, figure 9), and as the encroachment of the actinostome continues, some succeeding binary plates are also re- sorbed together with the corresponding portions of the ambulacra. In the Clypeastroids and Spatangoids the initial single plate is typically re- tained throughout life, as shown in numerous figures published by Lovén (25), A. Agassiz (3), William B. Clark (6), and Duncan and Sladen (10). It is obvious that the enlargement of the actinostome may take place by resorption of the plates of the base of the corona or by the progressive growth of the same, both tending in the same direction, the enlargement of the ventral circumference. Tiarechinus from the Trias is a genus which retains the initial single interambulacral plate, and so does Lepidechinus (plate 7, figure 42) and Pholidocidaris (plate 9, figure 54). From the above we gather the conclusion that resorption of ventral plates may or may not take place, and that the interambulacrum prob- ably always starts with a single plate. If Lovén is correct in supposing that the interambulacrum always terminates with a single plate, as every evidence goes to prove, then Melonites when young must have had a single plate at the ventral border. In a specimen of Melonites multiporus in Yale University Museum (diamond number 157, specimen C) we find an important feature bear- ing on the above consideration. In this specimen (plate 3, figure 10) f the two ventral plates have each an angle toward the median line, and these, together with the straight edge of the bottom, enclose a triangular space which doubtless contained the single initial plate, as in a similar stage of Strongylocentrotus plate 1’ (plate 3, figure 9). This specimen of Melonites does not actually show the initial single plate, and obviously * A reproduction of Lovén’s figure of this genus is given as an insert, figure 3, in the chapter entitled “ Conclusions” of the succeeding paper: Studies of Paleeechinoidea. + In the specimen, plates 2 and 3 (see figure) have slipped down a little from their original posi- tion, but they are restored to their places in the figure as they could be, all the angles being pre- served, by simply moving them upward. STRUCTURE OF THE INTERAMBULACRUM. 145 in its peculiar position it would very easily drop out after the death of the individual or in the processes of fossilization.* To our mind the aneles for its reception are almost as strong evidence as the plate itself; so it may be confidently stated that the interambulacrum of Melonites originates with a single plate, as shown by Lovén in Goniocidaris and Strongylocentrotus, and in the succeeding paper in Lepidechinus and Pholidocidaris. The specimen represented by figure 10, plate 3, evidently indicates a condition in which some resorption of the ventral border of the corona has taken place, and would correspond closely to the same condition in Strongylocentrotus (plate 3, figure 9). Reconstructing the ventral plate of Melonites to the condition it probably had before resorption cut into its ventral border, we have the condition shown in plate 3, figure 11, which is an adaptation from the ventral area of interambulacrum A in plate 38, figure 2. In this reconstruction of Melonites we have a ventral plate 1’ comparable to that seen in Strongylocentrotus before ventral resorption has commenced (plate 3, figure 8); also as seen in Lepidechinus (plate 7, figure 42) and Pholidocidaris (plate 9, figure 54). The single plate found at the ventral border of the interambulacrum in echinoids, as shown by Lovén in the young of modern forms and here in Paleozoic forms, points directly to the conclusion that this stage in erowth represents an ancestral form having a single column of interam- bulacral plates. The only such form known is the Lower Silurian genus Bothriocidaris (see figure 4 of the succeeding paper, in the chapter “ Conclu- sions” ), which from this fact assumes the greatest importance. In Melo- nites we show that newly added columns normally alternate to left and right as introduced, even-numbered columns typically appearing on the right of odd ones (plate 2, figure 2; plate 3, figures 12 and 14, and plate 4, figures 18 and 19). This being the case, it seems fair to argue that the initial plate 1’ of Melonites (plate 3, figure 11) was the basal member of column number 1, the left adambulacral, column 2, being introduced to the right of it. This assumption would carry with it the conclusion that the left adambulacral column of Melonztes and allies is probably the genetic equivalent of the single column of Bothriocidaris (figure 4 of suc- ceeding paper). Returning to the consideration of the interambulacrum of Melonites multiporus, in the next row above the ventral plates 1 and 2 (plate 2, figure 2) there are three plates. The median plate, number 3, is hexagonal in form, and it is the first formed plate of a new column; it makes, with the addition of new plates dorsally, the first column of median hexag- * Similar angles in the two ventral plates are shown in Oligoporus (plate 6, figure 26), and the initial plate is shown in Lepidechinus (plate 7, figure 42). 146 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. onal plates, as seen by following the dotted lines in the figure. Initial plate 3 is hexagonal in form; but, comparing it with the initial plates of later added columns (plate 2, figure 2), it will be seen to be similar in form ventrally and on the two sides. It is potentially pentagonal, if we may be allowed the expression, but is hexagonal because its dorsal border is truncated by the impinging ventral border of the terminal pentagon, number 4 of column 4. The terminal pentagon always induces by im- pact an additional side on one of its ventrally bordering plates, as seen in later added columns, and in this case it thus makes a hexagon out of a plate normally pentagonal. Plate 3 is typically hexagonal, and ex- ceptions to this form are very rare. Column 3 at its point of origin is obviously central in position, having a column of lateral adambulacral plates on either side. In the next row dorsally there are four plates, pentagon number 4 being the initial plate of column 4. The ventral as- pect of this pentagon impinges on plate 3, causing it to take on a hexag- onal form as just discussed. In four areas of the figure, A, C, EK and G, column 4 at its point of origin has two columns on the left of it and one on the right ; in a fifth area, I, the reverse obtains, there being one column on the left and two on the right. This shows at the start a variation which is not infrequent in even-numbered columns. Another case of the same kind is shown in the introduction of column 6 in area G of the same figure. Here there are two columns on the left and three on the right, whereas in the two other areas, E and I, showing the introduction of the sixth column, there are three columns on the left and two on the right. While even-numbered columns in most cases originate to the right of the center, having one more column on the left than on the right, they some- times originate to the left of the center, having one more column on the right than on the left. Other examples of this variation are seen in plate 3, figure 10, and in the tables on pages 165 to 170, inclusive. Later stages, with new columns added after the fourth, might be traced in plate 2, figure 2, but they can best be followed by comparison with the description of stages represented by figure 12 of plate 3. It will be seen in plate 2, figures 2 and 3, that columns 3 and 4 are introduced very early in passing from the ventral area dorsally. This rapid introduction is also shown wellin Melonites giganteus, sp. nov. {plate 5, figure 21). It is the common condition, but exceptions may be found, as in plate 3, figure 12, where column 4 originates much later than column 3. The rate of introduction of new columns, as shown in plate 2, figure 2, may be accepted as the average rate in Melonites multiporus, to which we have seen but two exceptions, namely, those shown in plate 3, figures 12 and 16. The rapid introduction of the new columns 3 and 4, in plate 2, figure 2, shows an accelerated development which is still more accentuated in STRUCTURE OF THE INTERAMBULACRUM. 147 Melonites giganteus (plate 5, figure 21), where the fifth column originates _in the second row above the fourth. On the other hand, the relatively later period of introduction of column 4 in the two examples of Melonites multiporus seen (plate 3, figures 12 and 16), shows a tendency toward a slower rate of development. In Tiarechinus we have a case of even greater accelerated development, for in it, according to figures by Professor Lovén (26), there are three plates in the second row succeeding the initial single plate of the first row. One of the clearest examples of the method of plate arrangement seen is a specimen of Melonites multiporus in the collections of the Museum of Comparative Zoélogy (catalogue number 2990).* Plate 3, figure 12, rep- resents the interambulacral area of this specimen, which is also shown in the photographic reproduction (plate 4, figure 18), and the nature of the plate arrangement may be seen by following up the series of columns of plates which are accentuated by dotted lines. While not complete dor- sally or ventrally, this specimen is very clear and is also very typical, showing little departure from the ideal type of plate arrangement.t Starting at the ventral end of the specimen (plate 3, figure 12, and plate 4, figure 18) there are three columns of plates, composed of two adambulacral columns of pentagonal plates, numbers 1 and 2, and a median column of hexagons, number 3. This middle column, number 3, would drop out near the oral termination of the area if it were complete, as in plate 2, figure 2. Passing dorsally,.a new fourth column of plates is added. This column is introduced by the terminal pentagonal plate, number 4. The introduction of this plate has disturbed the mechanical form of adjoining plates, so that they are somewhat distorted, as shown in the figure, and a hexagonal plate exists at A in place of one of the lateral pentagons of column 1. This introduction of an additional side to plate A, making a hexagon out of a lateral plate which is normally pentagonal, seems to be an individual peculiarity of this specimen, for in the case of other interambulacral areas in which the introduction of the fourth column has been observed, the initial pentagonal plate abutted against the initial plate of column 3, inducing an additional side on the dorsal border of that plate, as in plate 2, figure 2. In one other case ob- served (plate 3, figure 16), the initial plate of the fourth column did not * This is the specimen on which Mr Jaggar based his observations of the development 4nd arrangement of the interambulacrum. 7 Its principal departure is in the slow rate of addition of columns 4 and 5 (compared with plate 2, figure 2). Initial pentagon 4 is introduced considerably later than initial pentagon 3, instead of in the next row, as in the figure cited. In area C of the same specimen, however, as shown in plate 4, figure 18, the fourth column begins in pentagon 4 at a very much earlier stage of growth than in area A. This different rate of development in two adjacent areas is uncommon and has never been seen in any other specimen in the early added columns. 148 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. impinge upon initial plate 3 because of the intercalation of a peculiar plate in column 38. Column 4 of plate 3, figure 12, and plate 4, figure 18, is introduced as near the center as is mechanically possible with an even number of col- umns, but it falls to the right of the actual center, having two columns on the left and one on the right. Melonites, therefore, on the zone of pentagon 4 has four columns of plates, and a similar remark may be made of the progressively added columns where introduced. Continuing dorsally, especial attention is called to the intercalation of new columns, normally in regular alternation, 5 on the left of 4, 6 on the right of 5, 7 on the left of 6, etcetera. Continuing dorsally from pentagon 4, we find the fifth column introduced with a terminal pentagon, number 5. This column at its point of origin is in the middle, having two columns on either side of it. This feature is characteristic of odd-numbered columns ; they have an equal number of columns on either side. Occasional excep- tions, however, exist, as noted on pages 165-170. Adjoining pentagon 5, one of the plates, H, of column 4 has a seventh side added as a mechani- cal compensation for the form of the pentagonal plate 5. This relation of a terminal pentagon and an adjacent heptagonal plate, which is usually a member of the immediately preceding column, is characteristic of the form assumed when new columns are introduced above the fourth in the whole family of the Melonitide and in other Paleozoic echinoderms as well. Passing dorsally again, we find a sixth column introduced by the pentagon number 6, with an adjacent heptagon, H, on the left, which is a member of column 5. Column 6 at its point of origin has three col- umns on the left and two on the right and is therefore right-handed. Continuing still further dorsally, we find column 7, which is introduced by the terminal pentagon number 7, with an adjacent heptagon, H, on its right. The heptagon isa member of column6. Column 7 at its point of origin has three columns on either side. It is seen in the above that the odd-numbered columns 3, 5, and 7 are, when they originate, in the center, with an equal number of columns on either side; also it is seen that the even-numbered columns 4 and 6 are as near the center as they can be, but fall to the right of the center, with one more column on the left than on the right at their point of origin.* It is further to be noted that in the ideal case heptagonal plates fall al- ternately on opposite sides of the terminal pentagons in successively added columns, namely, on the right in odd columns and on the left in even columns, and are thus, when in their correct position, members of * In the reconstructions of the initial interambulacral plate (feares 19 and 11, plate 3) column 1 would also originate in a central position and column 2 would apparently originate with one col- umn on the left in accordance with the above law of the method of introduction of new columns. STRUCTURE OF THE INTERAMBULACRUM. 149 the preceding column to that one to which the terminal pentagon belongs (see plate 8, figure 12; plate 4, figures 18 and 19; plate 5, figure 20). This position of heptagonal plates, however, is a character which is sub- ject to somewhat frequent exceptions, as shown in the tables on pages 165 to 170. The above may be accepted as the rule in the method of introduction of new columns in Melonites, Oligoporus, and other Paleozoic echini. As new columns were added as the animal progressively grew dorsally,* and as new columns were introduced by a pentagonal plate, an apex of which pointed ventrally or toward the oral area (plate 3, figure 12), there is in this feature an important aid in diagnosing the relative position of the axes in even fragmentary specimens. It has been ascertained that this conclusion is correct by finding the termination of columns as stated in 14 specimens in which the axes were positively known from the presence of genital and ocular plates (plate 3, figure 13). This feature is one that has been overlooked by previous writers, for in almost all pub- lished figures of Melonitide, and other Paleozoic echinoids as well, in which the termination of columns is shown, the apex of the terminal pentagon points toward the upper pole of the figure, which should be the anal pole. This demonstrates that such figures were not correctly oriented, the oral and anal ends being transposed. This remark applies to Meek and Worthen’s (80, 31) figures of Rhoechinus gracilis, Oligoporus dane and Lepidocidaris squamosus, Hambach’s (18) figure of Melonites crassus, and M’Coy’s (28) figure of Palxechinus sphxricus, as well as some others. In passing toward the anal area where the younger plates are situated, we find that there is a gradual passage from the hexagonal form char- acteristic of the older plates of median columns. This change of form is shown well in a choice specimen of Melonites multiporus which Professor Wm. B.: Clark, of Johns Hopkins University, kindly loaned for study (plate 3, figure 13). It is also shown in plate 5, figure 20. In the pro- gressively younger plates, the form is gradually modified by the progres- sively shorter lengths of the upper and lower sides of the hexagon until the plates have a more or less rhombic form. At this part of the test it would be more convenient to study the plates in a descending order, for the growth is properly from the early rhombic form to the mature hex- agonal form. In passing from the rhombic form to the hexagonal it is occasionally seen that while the upper end of the rhomb retains its character the lower portion is truncated by a horizontal edge, thus form- ing a pentagonal plate with its apex directed dorsally, as in plate P, of figure 13, plate 38. There is, however, no need of confusing this type * See further discussion under ‘‘ Conclusions ”’ in the succeeding paper. 150 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. of plate with terminal pentagons of newly introduced columns. Close to the genital plates the interambulacral plates are quite irregular in form. A comparable irregularity of form in newly introduced plates may be seen in any modern echinoid, as Strongylocentrotus. Concurrently with the change in form, passing dorsally, there is a progressive reduc- tion near the abactinal area of the number of columns (plate 3, figure 13). This reduction, it seems, is to be ascribed to the progressively narrowing area which the upper part of the interambulacral area presents. To make a homely simile, it may be compared to a flock of sheep coming through a narrow pass. The small number in the pass does not mean that the flock is lessening, but that no more can get through at once, We take it, therefore, that the decreasing number of plates at the dorsal end of the interambulacrum does not mean that the number of columns is dying out, but that as a mechanical matter no more plates can get through this narrow area at once. An evidence of this crowding out of columns is shown well in plate 5, figure 20. Here the separation or stringing out of successive plates of the same column is most striking. The perpendicular dotted lines will serve to indicate which plates belong to a given column, although separated. The fact that the full number of columns may yet be traced in dissociated plates quite near the dorsal terminus is brought out by the oblique line X-Y in plate 3, figure 13, which bisects the separated plates of the 8 perpendicular columns ex- istent in this interambulacrum, and this fact is clearly brought out by the lines X, Y, Zin the diagram on page 164. This change to the rhombic form near the dorsal area is also shown by’plate 5, figures 20, 21, plate 6, figure 31, and plate 7, figure 36. Proof that the crowding out of columns is due to the narrowness of the area can only be obtained by getting young material, which several efforts failed to procure. Still, it is possible that in quite young stages, where the number of columns is smaller, all might continue to the dorsal ter- mination without interruption, because the area would be relatively less crowded with fewer columns. The crowding out is distinctly not a dying out of columns, for when columns die out or cease to be continued to the dorsal area it is commonly the middle or last added column which drops out first in the cases observed, as shown in the dropping out of column 11 in the dorsal area of Melonites giganteus (plate 5, figure 21). The dying out of columns is a change from the progressive addition or maintenance of columns, and, in so far as it goes, may be regarded as a retrogressive feature. This is not a common feature, as observations show. In Oligo- porus missouriensis (plate 9, figure 50) a sixth column of plates exists for a brief period, and in Lepidesthes wortheni (plate 9, figure 53) four columns of interambulacral plates are found ventrally, but one soon drops out. VARIATIONS IN THE INTERAMBULACRUM. 151 These are the only cases observed of a reduction in the number of columns passing dorsally in any Paleozoic echini.* VARIATIONS IN THE INTERAMBULACRUM. Having described an ideal normal type of plate arrangement of the interambulacrum, it is desired to call attention to one of the most irregu- lar specimens of Melonites multtporus seen. The specimen is in the col- lections of the Wagner Free Institute of Philadelphia, accession number 0226. Itisina large slab with several other individuals. The specimen (plate 3, figure 14) has eight columns in the interambulacrum. It thus givesan added column of plates as compared with that shown in plate 3, figure12. Part of the plates of the area are hidden in the matrix, but those visible show the essential features fora comparison. Atthe ventral end, as far as seen, there are four columns of plates (plate 3, figure 14), but a fifth column is soon introduced by the pentagon number 5. This column at its point of origin has two columns on either side. The sixth column is introduced by pentagon 6. Ithasa heptagon on the left, which is a member of column 5, and has three columns on the left and two on the right at its point of origin, as in plate 3, figure 12. The seventh column is introduced by pentagon 7, having a heptagon on the left, which is a mem- ber of column 5 instead of column 6, as in plate 3, figure 12. Column 7 has the usual number of columns on either side. Column 8 is peculiar in that it originates with a hexagonal plate 8 instead of with a pentagon, as in the cases previously mentioned and as seen in column 8 of the cases tabulated on pages 165 to 170. This column at its point of origin has four columns on the left and three on the right, the theoretically cor- rect position for an even-numbered column (compare with plate 5, figure 20, and plate 3, figure 12). Hexagon 8 has an octagonal plate, O, on the left, which is a member of column 7. The octagonal form of this plate is caused by the reéntrant angle made by the hexagonal form of plate 8. It is seen in the diagram (plate 3, figure 15) that a change to the pen- tagonal form of plate 8, together with a slight shifting of associated plates, would make both these plates of the usual form. ‘The specimen is of in- terest as showing the development of the eighth column in its correct position by our law of alternation; also as showing that the first formed plate of a newly introduced column is not always a pentagonal plate ; * In the E. M. Museum at Princeton, New Jersey, there are three specimens (catalogue numbers 1464, 1466, 14662) which show the jaws in place. These consist of 10 dental pyramids lying opposite the interambulacral areas. When complete they meet orally in acute terminal angles. In two specimens (1464 and 1466) part of the ambulacral and a few interambulacral plates of the peristome are preserved. These plates are more or less irregular, and the ambulacral plates have two pores. The specimens were seen too late to include them in this paper further than by this note. XXI—Butt. Grou. Soc. Am., Vou. 7, 1895, 152 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. this and a few other cases which are figured * or described, being how- ever, the only exceptions seen to the rule of terminal pentagonal plates (above number 3, plate 2, figure 2) seen in either Melonites or Oligoporus, which is somewhat remarkable, as in these two genera terminal plates are figured or tabulated in this paper in more than 260 columns, all but six of which were pentagonal (or hexagonal, when initial plates of column 3), and these exceptions were in all cases due to peculiar local conditions, as described. | Another feature shown by this specimen is the fact that heptagonal plates, which le next to terminal pentagons, are not necessarily members of the immediately preceding column. Other cases of this variation from the typical position of heptagonal plates, which are not infrequent, are seen in the tables on pages 165-170. From lateral thrusts and pressure a new column is introduced the initial plate normally takes on a when pentagonal form and an adjacent plate takes on a heptagonal form, this being a mechanical compensation for the missing side of the pentagon in an area where plates are either hexagonal or the equivalent. A peculiar case of irregular arrangement of plates is that shown in the ventral area of Melonites multiporus, plate 3, figure 16.7 The ventral plates 1 and 2 show on their lower border a reéntrant angle for the missing initial interambulacral plate (compare with plate 3, figure 10), as dis- cussed above; the third column originates in a pentagonal plate, num- ber 3, this being the only case observed in which this plate was not hex- agonal] (compare plate 2, figure 2); the second plate, H, in the third column is heptagonal and very irregular in form; the initial plate of the fourth column, number 4, originates in the fourth instead of the third row, its characteristic position, and is heptagonal, its ventral termination making areéntrant anglein plate H and being truncated dorsally by pentagon 5; the other plates of the area are of the usual form. Only one other case, plate 3, figure 12, has been seen in which plate 4 did not originate in im- mediate contact with initial plate 3. It has been stated that the normal form of median plates of the interambulacrum is hexagonal, and it is worth noting that this case makes only a modification of the rule. The sum total of plates 3, H, 4, and 5 gives 24 sides, which divided by four gives an average of six sides to each plate. In aspecimen inthe Museum of Comparative Zoélogy (catalogue num- ber 3021) two peculiar variations from the normal exist. In this speci- men (plate 2, figure 7) the initial plate 6 of the sixth column is tetrago- nal, a rare variation; also, the heptagonal plate H, associated with the terminal pentagon 8 of column 8, occurs not adjacent to the pentagon * Plate 4, figures 6 and 7; plate 3, figure 16; plate 5, figure 21. 7 The specimen is in Yale University Museum, diamond number 157, specimen Bb, VARIATIONS IN THE INTERAMBULACRUM. 153 but in the fifth instead of the seventh column, as usual. This heptagon, it is to be noted, is in the correct horizontal row, although in an unusual vertical column. Two or three similar cases of unusual position of hep- tagonal plates have been observed in Melonites multiporus and Melonites giganteus, Jackson (see area Jin tabulation of the latter species). In the E. M. Museum at Princeton, New Jersey, there is a specimen of Melonites multtporus (catalogue number 1462) which has nine columns of plates. The initial plate of column 5 is tetragonal with two adjacent heptagonal plates, as in initial plate 6, of plate 2, figure 7. Another peculiarity is the fact that column 7 at its point of origin has four columns on the left and two on the right instead. of being median in position. An adjacent interambulacrum of the specimen is normal, not showing the mentioned local peculiarities. A peculiar terminal plate of column 4 is shown in plate 3, figure 10. In this specimen the initial plate of its column is tetragonal in form, and this is almost the only case of the kind seen, except that shown in plate 2, figure 7, and plate 2, figure6. This plate impinges on the dorsal border of initial plate 3 in the usual manner (compare with plate 2, figure 2). It is shown that newly introduced plates near the dorsal area are rhombic in form, and it would seem that from the quick introduction of the second plate 4’ in this column the initial tetragonal plate 4 had been checked in its development so as never to assume the typical form. While in the body of the interambulacrum only terminal ventral plates of newly added columns are pentagonal, a few exceptions have been ob- served in which other plates have a pentagonal form. Such a case is seen in plate 3, figure 17, where, besides the initial pentagon of column 8 there are two accessory pentagons developed, one, P, on the left of pen- tagon 8 and one, P’, on the dorsal border of the same. We see nothing _ to account for this peculiarity, which is exceedingly rare, except chance variation. In this case it would seem that either pentagon 8 or P might be selected as the initial pentagon of column 8, but the left-hand penta- gon is evidently a member of column 7; also selecting the right-hand pentagon for the terminal brings column 8 in its correct theoretical posi- tion, with four columns on the left and three on the right. Melonites multiporus is described as :having seven or eight columns of plates in the interambulacrum. This is the rule; but in the large series of specimens examined many exceptions have been found in which nine columns existed in the interambulacrum. Thirteen of such areas are tabulated in the tables on pages 165 to 170. One of the best examples seen is that shown in plate 5, figure 20. In the right-hand interambulacrum of this specimen, columns 5, 6, 7 and 8 originate in pentagonal plates, as numbered, with heptagonal plates on their left 154 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. or right, in exact accordance with the law as worked out in this paper.* The ninth column originates in pentagon 9, with a heptagonal plate on its right ventral side, and with four columns of plates on either side, thus taking its correct theoretical position. The same statement holds good for the introduction of the eighth and ninth columns in the left adjoining interambulacrum of the same specimen, as shown in the figure. This de- parture from the rule of having only seven or eight columns is a variation which might be expected in a type of echinoderms which has already such a multiplication of parts. It is a variation in the direct line of pro- gressive ascent in the group to which it belongs, eleven columns being characteristic of Melonites giganteus (plate 5, figure 21), the next higher species of the genus. The introduction of the ninth column is shown in plate 2, figure 6, a specimen which is also included in the table on page 169. In this speci- men (number 3017) the initial plate of the ninth column is rhombic. It has only built a few plates, having originated shortly before the addi- tion of new plates ceased in the life of the individual. A heptagonal plate lies on its right lower side, as usual. While in some cases nine columns of interambulacral plates develop in several interambulacral areas of a specimen, it is not infrequent that only one or two areas acquire nine columns, while other areas of the same specimen have no more than eight. This fact is brought out in the tabu- lations of areas which are considered in succeeding pages. IMBRICATION OF PLATES. Having described the typical arrangement of ambulacral and inter- ambulacral plates in Melonites multiporus, it is desired to leave temporarily the consideration of interambulacral areas and take up the matter of imbrication and the genital and ocular plates as described in this and the succeeding sections. Inmbrication of plates is quite a common feature in Paleozoic echinoids, being most strongly marked in Lepidocentrus, Lepidechinus and allies. In the Melonitide 7 the imbrication is very slight. The ambulacral plates have practically perpendicular sides, both laterally and dorso-ventrally (plate 2, figure 5); what inclination occurs may be accounted for by the mechanical necessity for inclined edges in thick plates which form an arcuate test. In the interambulacrum the adambulacral plates imbricate under the adjacent ambulacrals (plate 2, figure 5) and not over them, as in Lepido- centrus and Lepidechinus. The median columns of plates of the inter- * Excepting that the heptagon next pentagon 5 should be on the right. + Excluding the genus Lepidesthes and allies, which are placed in a family by themselves, the Lepidesthide, Jackson. (See systematic table in the succeeding paper under ‘‘ Conclusions.’’) IMBRICATION: GENITAL AND OCULAR PLATES. 155 ambulacrum can not be said to show imbrication. They are very thick in the species Melonites multiporus (plate 2, figure 5), and thicker still in Melonites giganteus, Jackson, as shown in the broken portion of area I (plate 4, figure 19). The several sides of the plates diverge upward and outward as a mechanical necessity to fill space in a curved test. In species where the test is thin, as in Oligoporus coreyi, M. and W., the wedge-like form of the plates is less apparent than in species where the plates are thicker as described. The details of imbrication or want of it as described in Melonites multi- porus have been also observed in Oligoporus dane and Rhoechinus elegans. Therefore this type of mutual plate contact may be considered as char- acteristic of the famlly. GENITAL AND OCULAR PLATES AND DIscuUSSION OF ORIENTATION. In the Johns Hopkins specimen of Melonites multiporus (plate 3, figure 13) the genital plates are seen in all five areas. In these plates there are three, or four pores in each plate. The same condition occurs in a speci- men in the Museum of Comparative Zodlogy (catalogue number 3077), and in a specimen in the Boston Society of Natural History (catalogue num- ber 11569), in both of which five genital plates are preserved intact. In no case have five pores or less than three pores been observed in a genital plate, though altogether quite a large number of plates have been exam- ined. The same condition of pores has been observed in Oligoporus missouriensis, Jackson, as described in the succeeding paper. Messrs Meek and Worthen (30) have figured the genital and ocular plates of Melonites multiporus, and in their figure each of the genital plates has four or five pores. We would not question the accuracy of their figure, for all the work of these authors on Paleozoic echini is of the most painstaking and accurate kind; but it is felt that this large number of pores is not to be considered the normal or average condition, being rather an unusual increase. Mr C. R. Keyes (23) has recently published a figure of Melonites multiporus, in which the genital plates have three pores in two cases and four in two others. The fifth plate has a single large pore, but has a number of minute ones, and is described as a madre- poric plate. We have not seen any genital plate with less than three pores and have never seen evidence of a madreporic character in any of the plates. Some one of the plates, from analogy, one would suppose must be madreporic in nature, and it is important that Keyes has shown it may be found, although specimens showing it must be very rare. It seems possible that there is in his figure some error about the single large pore; but if it is correct, then a reduction as well as an increase from the normal number must be allowed for. 156 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. The ocular plates of Melonites multiporus have never been figured as perforated by pores, although Meek and Worthen (380) suggested that they may perhaps sometimes show pores. We have had the opportunity to examine a good many ocular plates in Melonites multiporus, and in no case were they perforated. The same observation is true of Oligoporus missouriensis, Jackson.* Itseems best to give up finally the idea of pores in the oculars of adult Melonites, unless some are actually observed. Pores in ocular plates are so constant a feature in echinoids (two are figured in Palxechinus by Bailey (5)) that Melonites and Oligoporus must be considered as exceptional in this feature, and quite likely they would be found in the young, if such material could be obtained. In the figures (2, 18, 20) and tables (pages 165-170) of Melonites multi- porus the several areas are oriented, the interambulacra and ambulacra being designated by letters from A to J, inclusive. As madreporic plates have not been seen, any interambulacral area is selected as a starting point and succeeding areas are counted from left to right, revolving like the hands of a watch, when looking down on the echinus from the dorsal side. The orientation is necessarily reversed when the echinus is viewed from the ventral side (plate 2, figure 2). It would be interesting if means of differentiating areas could be established; but at present it seems im- possible, for the areas, as far as known, present no radial specializations. If a definite method of orientation could be established from genital plates, it is feared that this would not aid one in orienting areas in speci- mens which were imperfect dorsally. Perhaps the orientation adopted in these papers is as convenient as any that could be practically applied. TABULATIONS OF PLATE ARRANGEMENT IN THE INTERAMBULACRAL AND AMBULACRAL AREAS. In order to give a description in brief form of a greater number of in- terambulacral and ambulacral areas in Melonites multiporus than could be figured, the accompanying tables are introduced. Besides giving other additional cases, these tables represent the arrangement in cases where it would be impossible to figure them adequately, as in areas on the oppo- site sides of the same specimen. In the tables the characters to be compared are arranged in perpen- dicular lines. The interambulacral areas, indicated by letters A, C, EB, G and J, are all considered as successive areas, as seen looking down on the test from the dorsal side and revolving from left to right like the hands of awatch. This is the same notation that has been pursued in the plates, as plate 4, figures 18 and 19, plate 5, figure 20. The first column * See succeeding paper, plate 9, figure 52. TABULATIONS OF PLATE ARRANGEMENT. EOL in a table gives the total number of columns of plates in the area. Un- less the area is nearly or quite complete dorsally this number is omitted, for if the area is incomplete any number stated might be too low. In the second column the number of the column is given; in the third the form of the plate in which the column terminates ventrally, indicated by Pif pentagonal and by X if hexagonal, or if otherwise by reference to foot-notes. In the fourth column is given the number of the horizontal row counting from the oral end in which the given column of plates originates. There are two plates only at the oral end of an adult, and these are called row number 1, and succeeding rows are counted from that point upward (compare with plate 2, figure 2). Obviously in most cases the oral termination is incomplete. In such cases the position of the first introduced column is assumed to be in the average position ; but the affixed number is then indicated by an asterisk, excepting when only one row is wanting and the interambulacrum begins with the initial plate of column 8. The advantage of this is to give the relative position above the first observed column of its later added fellows. The applica- tion of this notation is seen in the table of specimen number 3010. The fifth and sixth columns indicate the number of columns on the left and right of a column at its ventral termination, or when first introduced. The seventh column gives the position of heptagonal plates whether on the left or right of terminal pentagons. Any special exceptions are indi- cated in foot-notes. For the sake of an easy comparison of the several columns and to show individual variation, the characters which conform to the ideal law of growth are printed in Roman characters ; those which are variations from the ideal or normal are printed in italics, so as to be readily picked out. It may be stated that the specimens were not selected otherwise than as being sufficiently perfect to show satisfactorily the fea- tures of platearrangement. More tabulations might have been included, but they were omitted for the sake of brevity. It will therefore, we think, be conceded that the variations from the normal ideal type are relatively few. A comparison of the arrangement indicated in these tables should be made with the similar table of Melonites giganteus in the succeeding paper and with the figures of the several genera and species figured in the plates; also with the ideal reconstruction (see figure 1, page 164). REMARKS ON THE TABLES OF PLATE ARRANGEMENT. The first case (page 165) is specimen 3016 * in the Museum of Compar- ative Zoology. This specimen is remarkable in that it has the 10 ambu- lacral and interambulacral areas preserved almost in their original com- * This specimen was purchased of Ward, of Rochester, for the Student Paleontological Collection, but since its value was ascertained, it has been transferred to the Museum. 158 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. pleteness. Itis noteworthy that at first sight it is a specimen of indifferent quality, being considerably worn and compressed. By critical study all the areas were made out, as indicated in the table. The first row of inter- ambulacral plates is absent, but the second exists in each area. The ambulacra have 4 columns of plates at the ventral termination in all 5. areas, the characteristic condition at this portion of the test. Three inter- ambulacra, L, Gand H, have 9 columns of plates, but the other two areas, Aand E,haveonly 8. They might, however, have shown a ninth column added if preserved to the dorsal area where they are wanting. The initial plates of column three in the 5 areas are hexagonalas usual. The other 25 terminal plates of columns in the specimen have the regulation pen- tagonal form. Columns 3 and 4 originate in the second and third row respectively in all areas. Initial plate 4 truncates the dorsal border of initial plate 3 as usual. Column 5 originates in the fifth or sixth row, column 6in the ninth or tenth row, column 7 in the twelfth or thirteenth row, and column 8 in the sixteenth or seventeenth row. The ninth column originates in the one area where counted in the twenty-third row. The ninth column in area J originates one column too far to the right, but the other 16 odd-numbered columns originate in a median position, with an equal number of columns on either side. Of the 15 even-num- bered columns, 11 originate with one more column on the left than on the right; the other two originate with one more column on the right than on the left. Of the 18 heptagonal plates adjoining terminal pen- tagons, 13 occupy the correct theoretical position. Thus this specimen is very nearly normal as well as very perfect, showing little departure from the ideal type of plate arrangement, excepting in the unusual position of the ninth column in area J. The next specimen tabulated (page 166) is also important as showing the arrangement in 5 interambulacra for comparison. This specimen, number 3010, is in the Museum of Comparative Zodlogy. Each ambu- lacral area has four columns of plates ventrally. All the areas are want- ing in the dorsal portion, as that is firmly bedded in the rock. One interambulacral area, J, shows 8 columns of plates; the other 4 areas show only 7. It is possible, even probable, that an eighth would have been added to some of them if the dorsal portion were visible. Initial plates of column 3 are shown in areas (, FH and /. All are hexagonal, having their dorsal borders truncated by initial plate 4 of column 4, All the other 20 columns tabulated originate in pentagonal plates. The same numbered column originates in nearly or quite the same row in each of the 5 areas. A close comparison of the row in which columns originate is requested in this table and the preceding one. This will give one a good idea of the definite uniformity of growth maintained throughout REMARKS ON TABLES OF PLATE ARRANGEMENT. 159 the corona in two very perfectly preserved and quite average specimens in all details. The odd-numbered columns are all median at their point of origin. Six even-numbered columns are out of place, originating one row too far to the left ; also 6 heptagonal plates occupy the incorrect position. Otherwise the specimen conforms entirely to the ideal method of arrangement as here formulated. The specimen (number 2991) next considered (page 167) is in the Museum of Comparative Zoology. It has 4 columns of ambulacral plates ventrally in the 4 areas shown; also there are 3 interambulacral areas which are preserved from the second row of plates nearly to the dorsal termination of the areas. The initial plates of 14 columns shown are pentagonal. The initial plate of column 3 in area C is hexagonal, but in areas A and E the ventral border of the initial plate 3 is not shown ; therefore, though doubtless hexagonal, the form is not given. The several columns in each area originate in about the usual row as compared with each other and with other specimens. The 9 odd-numbered columns are all median in position at the point of origin. Four even-numbered columns are out of place, being one column too far to the left; also 5 heptagons are in the incorrect position, by rule; otherwise the plate ar- rangement is entirely normal throughout. Specimen number 3021 (page 167) in the Museum of Comparative Zoology has 9 columns in both areasshown. Area Ais complete dorsally ; one column, the eighth, is one column too far to the left at its point of origin, but otherwise this area conforms to the ideal in all details of its plate arrangement. Area C, however, is one of the most irregular areas met with in any specimen. It is figured in plate 2, figure 7. The ninth column of this area has 5 columns on the left and 3 on the right at its point of origin. This is a very unusual variation, for odd columns are almost universally median in position. A few similar cases will be noticed in other tables. The sixth column in the same area, C, termi- nates ventrally in a tetragonal plate, which is a very rare variation ; only four other similar cases have been met with in any Paleechinoid. One heptagon, that next pentagon 9,1is out of place. The heptagon which should come next to pentagon 8 lies in the correct horizontal row, but exists as a member of column 5 instead of column 7. This curious va- riation of position of the heptagon has been observed in only one other case in Melonites multiporus. It has been observed, however, in Melonites giganteus (see the table of that species, area J, column 11). Interambu- lacrum A, it is interesting to note, is almost entirely normal throughout, only one even-numbered column being out of place. The columns of plates are introduced at the same row in each area. Comparing the two areas of this specimen, it is seen that the irregularity of one area is not XXII—Butt. Gro. Soc. Am., Vou. 7, 1895, 160 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. repeated in the other area, and this remark applies to studies of varia- tions in the interambulacrum generally. In the development of the several areas of any given specimen of Melonites we find that the several columns are added at the same or nearly the same horizontal plane (as indicated by the numbers in the fourth column of the tables), but irregu- larities, such as seen in the position of heptagons or point of origin of columns, are not repeated in succeeding areas unless by chance, for on the principle of chances similar variations in two or more areas would occasionally occur.* The specimen of Melonites multiporus number 3023 (page 168) in area A has the same irregular position of the ninth column as specimen 3021, having 5 columns on the left and 3 on the right at its origin; also column 8 and two heptagons are out of place, otherwise the specimen is normal, conforming to the ideal arrangement throughout both areas. The next tabulation is of a specimen of Melonites multiporus given to Yale University Museum (page 168) as a slight return for many favors received. In the two interambulacral areas shown there are 9 columns of plates. All initial plates of columns are pentagons. The arrangement of details is entirely normal except that two heptagons, next pentagons 5 and 7, are on the wrong side, and one of these heptagons is too far to the left,as noted. -Twoambulacra show 4 columns of plates at the ventral termination. A specimen, number 3007 (page 168), in the Museum of Comparative Zodlogy shows two ambulacral and two interambulacral areas. Interambulacrum C has two even-numbered columns, one row too far to the left and one heptagon out of place ; otherwise the arrange- ment of this specimen is entirely normal in its details. The next specimen tabulated is owned by Mr T. A. Jaggar (page 169). Interambulacrum A has all the even-numbered columns left-handed or with one more column on the right than on the left at the point of origin; also and perhaps as a correlated feature all the heptagons adjacent to pentagons are on the wrong side in the several cases. Such uniform variation from the normal has not been seen in any other specimen of an interambulacral area. The next interambulacrum, C, of the same specimen is perfectly normal throughout, except that two heptagons occupy an incorrect position. Ambulacrum B has four columns of plates at the base, as usual. Specimen number 3017 (page 169) has 9 columns of plates in areas A and J and 8 columns in the other 3 areas. The arrangement is normal as far as visible, except that one heptagon is out of place and column 9 in * For this matter of variation it is desirable to study the figures and tables of Melonites with the idea of radial variation especially in mind. The figures of Palwechinus (plate 7, figure 36) and Ar- cheocidaris (plate 8, figures 43-45) may also profitably be considered under this study of radial variation, REMARKS ON TABLES OF PLATE ARRANGEMENT. 161 area J terminates in a ventral plate, which is tetragonal or rhombic in form. This plate, which is shown in plate 2, figure 6, is quite near the dorsal termination of the area. Specimen number 3019 (page 169) shows only one area, which has 9 columns of plates. The ninth originates quite close to the dorsal termi- nation of the area, as in the above case, and its terminal plate is rhombic inform. The arrangement of the areas is'‘entirely normal, excepting the tetragonal plate and one heptagon, which is out of place. Specimen number 3020 (page 169) is normal throughout, except that one column, the eighth, originated one column too far to the left; also one heptagonal plate is out of place. Specimen number 3022 is entirely ideal in its arrangement through- out, unusually normal in fact, for some variation is commonly met with. Specimen number 2992 (page 170) is peculiar in having one odd-num- bered column, the seventh in area C, one column too far to the right; also two even columns in area A are too far to the left, and three heptagons are out of place. One area, C, has but 7 columns of plates, whereas the other has 8, both areas being complete dorsally. Specimen 2999 (page 170) has 9 columns of plates and is normal throughout in its plate arrangement. Specimen 3004 (page 170) has two columns and one heptagon out of place. The irregular position of column 7 is noteworthy, because it is unusual for odd columns to be otherwise than median in position. Specimen number 3006 (page 170) is entirely normal throughout. Having stated the law of growth in the interambulacrum of Melonites multiporus, and having figured and tabulated the arrangement in many specimens, it is desirable to foot up the results of the observations and see how closely actual observations come to the assumed law of intro- duction and arrangement of plates. In this summary are included all the figures given of Melonites multtporus in the accompanying plates; also the tabulations of the same species.. In a few cases the same area is in- cluded in a figure and also in a table, when of course but one count is made. For most of the details considered we have also included the results obtained from the tabulation of Melonites giganteus, Jackson, as described in the succeeding paper. From the context it will be made plain when both species or only one species are included. ~ In considering this summary constant comparisons with the figures of Melonites should be made, especially the ideal diagram on page 164. Comparison is also requested with the plate arrangement illustrated in the succeeding paper in the genera Oligoporus, Paleechinus, Archxocidaris, Lepidocidaris and Lepidechinus, where the same method of growth pre- 162 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. vails in its principles, but varies somewhat from generic differences of structure. Summing up, 1 we find that in Melonites multiporus 35 interambulacral areas are figured or tabulated which are nearly or quite entire dorsally. Of these 2 have 7 columns of plates, 18 have 8 columns, and 15 have 9 columns. Therefore it may be stated that 7 columns is an unusually small number of columns for this species, it usually possessing either 8 or 9, with about even chances for either number, but slightly preponderat- ing in favor of 8. Nine columns of plates have never been described as existent in this species, which is somewhat strange, as this number is evidently so frequent. The introduction of new columns (succeeding pilav numbers 1 and 2) is shown in -248 cases in Melonites multiporus and giganteus. Of these the initial plates of the columns are pentagonal in 220 instances, in every case of which an apex of the pentagon is directed ventrally and a side of the pentagon is directed dorsally. Twenty-two of the terminal plates are initial hexagons of column number 8, which is the normal form in this instance. Only in 6 cases out of the whole number 248 are the initial plates of columns other than the regulation form; that is, the form of the initial plates of columns is correct by rule in 97+ per cent of the cases. In the matter of the position of introduction of columns of Melonites multiporus as expressed by the number of rows from the base in which the column originated, there is for the most part great uniformity. Col- umns 1 and 2 always are represented by the two plates found at the ventral termination,* as observed in 21 cases (page 143).f The third column originated in the second row, as shown in 26 cases. The fourth column originated in the third row, as observed in 27 cases; in one speci- men (plate 3, figure 16) this column originated in the fourth row and in one (plate 3, figure 12) it originated considerably later; but these are the only exceptions that have been seen. The fifth column originates in the fifth row in 8 cases, in the sixth row in 17 cases, in the seventh row in 8 cases, and in the eighth row in 7 cases; therefore the fifth column in the great majority of cases originates in the sixthrow. The sixth column in 10 cases originates in the ninth row, in 18 cases in the tenth row, and in #hat is, the ventral termination as normally existing in an adult. In this summing up of areas the initial first formed plate 1/ (plate 3, figure 10) is not considered. It is a feature properly be- longing to an earlier period of growth and is resorbed by the encroachment of the peristome. We are now considering the adult, and therefore consider the ventral border as it exists in that period of growth. +Excepting this statement in regard to the first two columns, all the observations cited are those figured in plates or shown in detail in the tables. Of course many other observations were made, but we limit the remarks to the published observations. REMARKS ON TABLES OF PLATE ARRANGEMENT. 163 7 cases in the eleventh row; the sixth column, therefore, in the majority of cases originates in the tenth row. The seventh column in 4 cases origi- - nates in the twelfth row, in 15 cases in the thirteenth row, in 11 cases in the fourteenth row, in 3 cases in the fifteenth row, and in one case in the seventeenth row; the seventh column, therefore, originates in the great majority of cases in either the thirteenth or fourteenth row, but occa- sionally lower or higher. The eighth column originates in the sixteenth row in 5 cases, in the seventeenth row in 7 cases, in the eighteenth row in 5 cases, in the nineteenth row in 5 cases, in the twentieth row in 3 cases, and in the twenty-second row in one case; there is considerable variation, therefore, in the period of introduction of the eighth column without special preponderance in favor of any one row; the sixteenth to the nineteenth row may be taken as the average swing in rate of appear- ance of this column. The ninth column has a similar wide swing in its period of introduction; it originates in the twentieth row in 3 cases, in the twenty-first row in three cases, in the twenty-second row in 2 cases; in the twenty-third row in 3 cases, and in the twenty-fifth row in one case; this ninth and last column added, therefore, makes its appearance in a majority of cases in the twentieth to the twenty-third row. The interesting result of this tabulation of periods of appearance of columns is to show with what regularity the several columns were intro- duced in the development of the animal, all appearing within the close limits of a perfectly definite period in growth. As might be expected, the columns added first, from 1 to 4, inclusive, are the most constant in their rate of appearance, later added columns being more variable, yet a pretty close adherence to the abstract law of specific periodicity is main- tained in all the columns asadded. Comparing this rate of introduction of columns with that of Melonites giganteus, we find most suggestive and interesting conclusions, as discussed under the consideration of that spe- cies in the succeeding paper. Turning to the results obtained in other columns of the tables, we would again include Melonites multiporus and Melonites giganteus in one considera- tion of the details. Out of a total of 188 odd-numbered columns intro- duced, all but 6, or 95-++ per cent, originate in the middle, with an equal number of columns on either side. In these 6 exceptions the column had one more column on the left than on the right, or was right-handed. In the cases of 1138 even-numbered columns 82 originated with one more column on the left than on the right, and 31 originated with one more column on the right than on the left; that is, 72+ per cent were in the theoretically correct position, or left-handed. In no case of column ar- rangement in either odd- or even-numbered columns was the newly intro- 164 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. FIGURE 1.—/deal Arrangement of tnterambulacral Plates in Melonites multiporus, duced column more than one column out of place to the left or right. The same remark applies to all the studies of Paleechinoids as presented in the succeeding paper, excepting the case of Melonites septenarius (plate 9, figure 49), as there described. The heptagonal plate adjoining the terminal pentagon should by rule lie to the right of the pentagon terminating odd-numbered columns, and to the left of the pentagon terminating even-numbered columns. The position of the heptagonal plate is shown in 171 cases, and of those it occupies the correct position in 122 in- stances and the incorrect position in 49 in- stances. In other words, it occupies the correct position in 71 per cent of the cases. From this close adherence to the theoretical method of growth we are justified in giving the annexed diagram as representing the correct method of plate arrangement in Melonites and the average number and relations of plates ex- istent in Melonites multiporus in the adult con- dition. In the figure all the plates are put in according to rule as deduced from the compila- tion of numerous critical observations. The sup- posed initial plate 1’is included in the figure so as to indicate all the plates it had at any period of growth, but it should be distinctly under- stood that this plate is typically resorbed in the adult condition. Nine columns of plates might have been introduced in this figure, but it is intended to represent the ideal, and 8 columns are somewhat more frequent in perfect speci- mens than 9. The lines X, Y, Z are drawn so as to bisect 8 columns of plates. They therefore indicate by their progressively narrowing angle, the stringing-out arrangement of plates in the dorsal area as described on page 150. (Com- pare with the line X-Y in plate 3, figure 13,) PLATE ARRANGEMENT OF SPECIMEN 3016. 165 TABLES OF PLATE ARRANGEMENT. a slg [@ | 4 Rede ae tS & rs ood t& ~ Be BP) Be |e. |e Ov Q > ae erect eet ; : ri ao om ao 4 } Melonites multiporus. a < 5 pe ee @ | 2. Bp eee Penieen i Museum of Comparative Zodlogy | 45 4 alae mae A ~ catalogue number 3016, Q I fe le g g Se ese baa tee pen We ® H a Sve ao Wit ees Interambulacrum A*.............csceeeee 8 8 1p 16 4 3 Left. 7 Pei 3 3 | Left. 6 IE 9 2 3 Indistinet. 5 14 5 2 2 Indistinct. 4 Je? 3 2 1 Truncates initial plate 3. 3 x 2 1 TEST hee eR Ason SADA ELEIDEEL OO? EOL IEP OCLC First row of plates wanting. FATVOMIACTUNO'D :..00.css000 sesccvessoecceess Four columns of plates at base. Interambulacrum C*.............00:0000 8 8 Ie 17 4 3 | Left. 7 We 13 3 3 Right. 6 12 9 3 2 | Left. 5 HY 5 2 2 Right. 4 12 3 2 1 Truneates initial plate 3. 3 x 2 1 WON epetesstnasessssasapevscexsipeesvabatess First row of plates wanting. MOV ACTIVA DD) vccvscesssecseosees ovsoesces Four columns of plates at base. Interambulacrum E*.................,... 9 9 t Pianesses 4 4 Right Sie il ennaul lapses. lees seencs|lisasesarse basesevescesnssae yas secede ia Nan ener deren 7 Ie 13 3 3 Right 6 124 10 3 2 Left. 5 P 6 2 2 | Right 4 1 3 2 1 Truneates initial plate 3. 3 x 2 i Pan ltecssavedataers szerraste te deeiers tae teens First row of plates wanting. POO MMACTUM BF ois sscesssseore coscdeoepors Four columns of plates at base. Interambulacrum G f..............000000 9 9 P 23 | a 4 | Left. 8 Ie 16 | 4 3 ight ff ice Wh |'asevesan lla Ause'ess||ca alan teeu| Satay etait renendeas eveavaassaeeer esters 6 1B i) 3 2 Left 5 P 6 2 2 | Deft. 4 1 3 2 1 Truncates initial plate 3. 3 x 2 1 Sel iesoaveshsescteasveasniere esi sitiaccnsee First row of plates wanting. PAMTUUNACE MIM EL ....s.s00espeoveoeciescesoces Four columns of plates at base, Interambulacrum I%...........cceceeeeee 9 9 t sseoseees 5 3 | Left. Sry. lee eltaneilanegeaten tenenceaa| Menderes lsteceess aes. ies saps sdte svanne Four columns of plates at base. * Interambulacrum wanting at the dorsal portion of the area, + Interambulacrum complete at the dorsal portion of the area. t Wanting at this portion of the area. 2 Plate on the left of pentagon 7 is hexagonal; on the right the plates are broken away so that angles cannot be determined. 166 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. Melonites multiporus. Museum of Comparative Zodlogy eatalogue number 3010. Interambulacrum A *..........:.0. sees. Ambul seri Biixccasscsnsesease set ameter Interambulserwm © *®....5... sseccsececos Ara bulaerinnl Dy. ccaecesccstaceckacbesveusesn Interambulacrum E *.,......sccccccccors A MmBULRer OM O . cccctets catessudacksanvnce Interambulacrum G * ...........eeeeeeees AY DE SeNU A sel s.s. Soc ctassvncoscen aJassasans Interambulacrum I *.........cc0.sssceess Arn UA OYUN rece senecuoares spe paeceaseee Oe Or =1 0 a — n : . H 32/3 ° oa) & _— 52) 8 =~» ro) S = & ae ° A o sesasenes 7 ie: ie _— Seip se by =} mH a S| Sai We rey = 4> A le 1% C0) = =e 2 S a0 = mn = c= 0 Ee | 2s) os 3 Ae neo ae |e — S Be ye x) > ° a) ie) S) P 14 3 P 10 2 So i's 2 Indistinct at lower part. (en eeeneee 7 6 5 4 3 Four columns of plates at base. Columns on right of, at origin Heptagon on left or right of terminal pentagon. 3 Left. 3 | Right 2 | Right s 13 3 3 Right fs 10 3 2 Left. P 7 2 2 Right Er 3 1 2 x 2 1 1 First row of plates hidden. Pp | 13 3 r 10 2 rE 7 2 4 3 2 x 2 1 First row of plates wanting. Four columns of plates at base. me bo ww ‘Truneates initial plate 3. Left. Right. Right. Truneates initial plate 3. Four columns of plates at base. | 5 4 if 14 3 1d 10 3 P 7 2 P 3? 2, lt oo) First two rows of plates obscure. 8 sg 20 3 P 15 3 4 lL 2 P 8 2 Pp 3 1 x 2 1 First row of plates obscure. _ Four columns of plates at base. Four columns of plates at base. Right. Left. Right. Truneates initial plate 3. ? Left. Repke. Right. Truneates initial plate 3. * Interambulacrum invisible in the dorsal portion of the area, being more or less deeply buried in the rock. ; ) | a PLATE ARRANGEMENT OF SPECIMENS 2991 AND 3021. 167 Melonites multiporus. Museum of Comparative Zoélogy catalogue number 2991. Seeeecccesereecoeses PAVANIOUN CTU: Exics sactcctecscosccesesceuscsecs Interambulacrum O%........cccccccsseede AN TEU O16U NCTA 00 led er ree Melonites multiporus. Museum of Comparative Zoélogy catalogue number 3021. interambulaeruml A‘ fz .....cecsse.cs cose) AMMO ULACTUND Bis. .ccscssescesedsconessocees imteranbulacrum Obs... .c.ceccc.csceces (Figured on plate 2, figure 7 ) Four columns of plates at base. S , » 2 Ss = E e a nN — PA iy od S e o ° ° gE a Wipes alas re ee Sra er Ree Ss Panes bien oo] 2 2 Se WN Sel ser Ow S > og O A= oy 1 5 3 o Det Srret Mace ° A oe) oe A A a lo] py no nO o~ F ra] S S| =| 2 en = Ss) | Ss S E SD S s = °° o | h °° °° A ES. WS Mirek aS dies) Four columns of plates at base. 8 8 12 19 4 3 a 12 15 3 3 6 12 11 3 2 5 12 7 2 2 4 12 3 2 1 3 t 2 1 1 Heptagon on left or right of. terminal pentagon. Left. Right. Left. Right. Truneates initial plate 3. Owe eee cee ees esses cesses seesss eee seesee First row of plates and ventral border of second row wanting. 8 8 P 18 4 Uf RB 13 3 6 LP 9 2 9) IP 6 2 4 P 3 1 3 x 2 1 First row of plates wanting. hr a DO Cg CO CO Right. Left. Right. Right. Truneates initial plate 3. Pee eee ee CHS eeEE eee EOEeEeeseres ES EEEEE® Four columns of plates at base. ee 7 Pi | 43 3 6 P | 10 2 5 P 6 2 4 P 3 1 3 t 2 1 Four columns of plates at base. mo bo Co Right. Right. Left. Truneates initial plate 3. see eee Seow ees eroseeessereresesr ess eseeses First row of plates hidden or wanting. CO Or =~1 00 © laclarhachaclachas] RPrpwwt cto, 2 | Two plates in first row. Four columns of plates at base. ell ell SO SOS ald 9 9 12 21 5 8 P Uy 4 7 P 13 3 6 2 9 3 5 Ie 6? 2 NWmwwss First three rows of plates wanting. Right. (?) Not visible. Right. Left. Right. Truneates initial plate 3. Beene Cowen tee ee OOOO wees sarees SO eee sees Left. Right. Heptagon on left and right. Right. * Interambulacrum wanting at the dorsal portion of the area. + Interambulacrum complete at the dorsal portion of the area. { Ventral border of plate 3 invisible. 2This plate is tetragonal in form (see plate 2, figure 7), and to compensate for the loss of two sides there are two heptagonal plates, one on either side of tetragon 6. || No heptagonal plate next pentagon 8, but a heptagonal plate exists in the row below pentagon § and in column 5. XXIJI—Butt Grou. Soc. Am., (See plate 2, figure 7.) Von. 7, 1895, 168 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. S| = es = S n = = pen = S| = I ° ° S ras B=) > ~~ = 5 z . o | & 2 si 2. 3 > | ao] _£ | £= | qeptagon on left or right Melonites multiporus. = a g a |e | on | se ‘of fan ponteustne fo) c- 5 ~ o~ = 5 Museum of Comparative Zodlogy | 5s = | = = ° a : catalogue number 3023. 2 a | ra & g & E = ee: a ee: = = = @ | P |e ©) 3 A Sy | al co) .S) SS) | "Fin th eee ; Interambulacrum Al* i... c.sc.scccsien: 9 ee Le 21 5 3 Right SNe 17 + 3 Right if P 13 3 3 Right 6 ae 7 1.8 2 | Lett. 5 Fr 6 2 2 Right 4 P 3 2 As PGES ay enceeeansactansdvenden es oceans First two rows of plates wanting. ATUMMIACTUM Bh. osnesdiontexxpsecewscapetnan Four columns of plates at base. Taterambulacrum © Fociisczccdaccckeccas Leceaccace 8 P 1 eg 4 | Right 7 re 13 3 + Right. 6 is Ww 3 2 Left. 5 P 7 2 2 Right 4 | 3 2 i I leeusadlsxewauetapeunery aa so sdubavGasstondne First two rows of plates wanting. Melonites multiporus. | Yale University Museum diamond number 2290. Interambulacrum A *...............c0000 9 9 P | 20 4 4 Right. 8 | 16 4 3 Left. 7 Pi). 3B deg So 1 Det fo) ee Oly ae caueaaven| acaceenenl eeeees eel seunearne Broken; indistinct. 5 P 8 2 2 Right. 4 Pp 3 2 1 eRe ee eet eee SaeERe? seeesesoere eeocse . 3 x 2 5 Bef Tams | Truneates initial plate 3. First row of plates wanting. AA DINRE YO Boss cexaancuneaes winevatdocs Four columns of plates at base. Interambulacrum C......... .cccecss-+ 9 9 P 22 4 4 Right. 8 P 18 4 3 Left. 7 Us 14 3 3 Right. 6 is 11 3 2 Left. 5 P 82] 2 2 | Left, First five rows of plates wanting. Ambalacrawmi D)., sniutascsceniears ere Four columns of plates at base. Melonites multiporus. Museum of Comparative Zoélogy eatalogue number 3007. Interambulsergnn \A Fisensssscavencctesse}innsucss 7 ie 14 3 3 Right. 6 Fs 11 3 2 Left. 5 le 8 2 2 Right. 4 Ie 3 2 1 Truneates initial, plate 3. 3 x 2 1 Lie || davon scacncernansaracideoseelees eee P First row of plates wanting. Ambulacrum B........ sepushicaaconscatesess Four columns of plates at base. Interambulacrum C F........0.ccc0sseees 8 8 Ez 19 3 4 Left. 7 rE 14 3 3 Right. 6 eae ale 2 3 | Right. 5 P 8 2 2 | Right. 4 Pp 3 2 1 Truneates initial plate 3. First two rows of plates wanting. Aum Dilan, We eietsatoranceasss xe Four columns of plates at base. *Interambulacrum complete at the dorsal portion of the area. + Interambulacrum wanting at the dorsal portion of the area. { But in second rather than adjacent column, as in heptagon next pentagon 8 of plate 2, figure 7. e PLATE ARRANGEMENT OF FOUR SPECIMENS. 169 o : S| aad ~ e Sy eo teow lias n = Be ws Ge =) CS <=: ° ° =| 3 z ; =_\|og : : Sc = > ao = | == | Heptagon on left or right Y o 4 = . . Melonites multiporus. 5 : zi g ne oe at of terminal pentagon. Specimen in possession of Mr T.| #5 | Si mle = 2 pe A. Jaggar, Jr. 2 & | g S S| 5 5 oh 5 5 = S @ | f) ‘ce Zz S) =| | © ‘S) S) Interambulacrum A*......... Re eee 8 8 ie 19 g 4 | Right. ‘i P| 14 3 3 | Left. 6 P 10 2 8 Right. 5 P 6 2 Diy \laept. 4 P 3 1 2 Truneates initial plate 3. 3 x 2 i We Necooncdoecnce dosnod hoononbnodocteneeceaoo5 First row of plates wanting. ASTM UT ACHUNI IB). cccisscsdsecescccescsecetess Four columns of plates at base. Interambulacrum C¥%.................. aS 8 12) 19 4 3 Left. 7 P 14 3 Bien dene. 6 Pe leo 3 2 | Right 5 P Ge? iene 2 | Right First three rows of plates wanting. Melonites multiporus. Museum of Comparative Zodlogy catalogue number 3017. Interambulacrum A f.........-sesese- 9 9 12) || 2} 4 4 | Right. 8 1e 17 3 L Right. 7 1p 1B? 8 3 Right. Wanting below this level. Interambulacrum C f...........0......-. 8 | 8 1p i 3 h Left. 7 Ie 1 2) 8 3 Right Wanting below this level. Interambulacrum EH f.. .... ........-... 8 foe” lleeesvae:l atau sacs [acre Goweue | Saaawcner'[adecravestetalusvesenooten chassaues cece ae ate Imberanibwlacrum G iz.--.0.c.-cescecssens 8 Aft len eoccece4s] Wodeecc eal lnoucceces| bescnocca MeeRenasssrocicocs eeu tas meres eaceeo ashen Interambulacrum I f.................000. 9 t 72 obeece a lbonecce al lesetactec Heptagon on right and left. Melonites multiporus. Museum of Comparative Zoélogy catalogue number 3019. Miner aan la CMUIN yicccesseccecsec ieee ssc. 9 9 I 25 4 4 Right. 8 P 16 4 3. | Right. U P 12 3 3 Right. 6 P 9 3 2 Left. 5 P Gel 2 2 Right. First four rows of plates wanting. Melonites multiporus. Museum of Comparative Zoélogy catalogue number 3020. MANGE TAMU ACTUMA sac scsceceeesdec cesses: 9 9 P 23 4 4 Left. 8 2 19 8 h Left. 7 Ry 15 3 3 Right. 6 12) 10 3 2 Left. 5 iz 6?] 2 2 Right. First four rows of plates wanting. *Interambulacrum wanting at the dorsal portion of the area. + Interambulacrum complete at the dorsal portion of the area. t Plates preserved at the dorsal portion of the area only. 2 This terminal plate is rhombic in form. (See plate 2, figure 6.) || This column originates close to the dorsal termination of the area. bic in outline like other plates in its vicinity. Its terminal plate is rhom- 170 JACKSON AND JAGGAR—STUDIES OF MELONITES MULTIPORUS. | : =e es 3 ea peed Pa ee S < & r) ° fai) 2 tee oN ea ron) 2 Le ST ai CSRs) : : F o 8 > ao ‘= | == | Heptagon on left or right ETEETO NEES Tats: 5 ° : 3 @ 3 og SE of terminal pentagon. Museum of Comparative Zodlogy | 35 o ae a = catalogue number 3022. 2 = oS a | iS & = z ay = = 3 > ot = = BS We Wie, Meee pie Interambulacrum A *.........0000 eeeee 8 8 1e 18 4 3 Left. 7 P 12 3 3 Right. | 6 P 9 || 8 2 | Left. fae P 62! 2 2 | Right. | First four rows of plates wanting. Interambulacrum C *.............06 somes ans 8 Pe 18 4 3 Left. 7 P 12?| 3 3 Right. Wanting below this level. Melonites multiporus. Museum of Comparative Zoélogy catalogue number 2992. Interambulacrum A Fu..e.eeeeeeeeeeees 8 8 P 20 3 4 | Right. 7 iz 14 3 3 Right. 6 E lu?| 2 3 Right. First seven rows of plates indistinct or wanting. Interambulacrum C f.............00s0000e af 7 i 14 4 2 Left. 6 1 10 3 2 Left. 5 e 6?| 2 2 Right. First three rows of plates wanting. Melonites multiporus. : | | Museum of Comparative Zoélogy catalogue number 2999. Interam bwlacru ni Fin vccvecvexsscsasyesene i) 9 1z 22 4 4 Right. 8 P 17 4 3 Left. 7 P 13 3 3 Right. 6 P 10?| 3 2 Left. First six rows of plates wanting. Melonites multiporus. Museum of Comparative Zoédlogy catalogue number 3u04. TnterambulaGr wii jeeccnscersssasweegsanche 8 8 P 22 38 h Left. 7 1s 14 2 4 Right. 6 P 10 3 2 Right. 5 P GH! 22 2 Not visible. First four rows of plates wanting. Melonites multiporus. | Museum of Comparative Zodlogy catalogue number 3u06. InterambalacruMy Fos-...s.ss>versnn-s-5- 8 | 8 P 18 4 3 | Left. ae bias al apes a ot 3 | Right. [ep Pte) Bos) teats, | BR ie 3 ae 2 | Right. sy ee SO ne gale Steel | First two rows of plates wanting. * Interambulacrum wanting at the dorsal portion of the area. 7 Interambulacrum complete in the dorsal portion of the area, BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 7, PP. 171-254, PLS. 2-9 JANUARY 31, 1896 STUDIES OF PALASECHINOIDEA BY ROBERT TRACY JACKSON (Read before the Society, August 27, 1895) CONTENTS ‘ Page amu mm MN E CTO TUT ees. Se. pists 4) a. Ss mist Site eisieietta Gia edie se lsitee wee ahi o Dab sds iz VTE CMY CLENO TB el Ry dob taal wi os coe senldsih ANE Aa lbh 308 boc teri 8 op eae a a 172 Mesemovion ot Melonites guganieus, Spo MOV.) 00... 225.0266 fe ee eee ee 172 Description of Melonites septenarius, sp. nov. (R. P. Whitfield) ............ 182 Consideration of Oligoporus and comparison of the same with Melonites....... 184 Pesempiion of Oligoporus Missouriensis, SP. NOV... 0-4... je- cece sec e ree t see 184 Mee eM OOO ONORUSIGOTCUD. a «oc sere Kpouniete sisters <6 = 0 S5a edie 's owe See 186 Arrangement and development of plates in leaner US COTLUL, = huss eotpcian ¢ 188 Description of plate arrangement in Oligoporus dan#.............. 002000 193 Gunevdersiion of Rhoechinus and Palwechinus......0.200.06 0 ccc cee ene eee 200 Mime TOMES AMET ShUCIES les a «sas che std iiaeis He ei Susca se Sew Medea staid wsicts 200 Structure and plate arrangement of Rhoechinus gracilis..............-545. 201 PRS TOMeLOCCHVTLUS OUTIIMOLOMCNSIS © a ocveoe se te wiv d tien diel cesisie ee da diene ee 203 Observations on Palzechinus gigas and Rhoechinus elegans ........+.00.04. 204 Sracuiesnernac mepidesthidee, fam. NOV... 2s nec we ews cewek ee age cece eee wee 206 EelLmone mips andr CMATACTCTISLICS. cc. . 2. ee nee ee ce cette re one on ..: 206 Description of Lepidesthes wortheni, sp. NOV......... Ere ARG Em Fae 207 Mere Tomotmenspecies!OL Lepidesihesy 5. oc. sic oe bein woes wpe ein See wel wes 209 Description of Pholidocidaris meeki, sp. NOV.........0....0 cee eee ceeeseees 210 Ree Meme MMT CMECOCIMARIGES. 4.52. sho aysispee Beis tps Sie aces Sete a argc pis ee ls wales s 213 Structure and plate arrangement of Archxocidaris worthent and other species. 213 PIMNCINEC On LCN AOCldarts SGUAMOSUS a.[.. oie. oiele'se tse dle tess oie eee wi 220 Meats ROMEO CTLOCUM TIS. it R08. seis ie we eoavs © ale CREAT eaters a Slade MVE WES Dale bJos'ord 4 aia nay 222 eM Mmmm MTC IOC DIG OCOM LETC sx. de zldn ua, Gisne Oshawa ea est wece maida s Sa leeiin See 8 as 222 rom mC MAT ACUETICUICG p< ..: cals Cnc nats aoe ee um eee ais © cae EI gles 0% Soa 2 222 Siruenure and plate arrangement of Lepidocenirus. 6.0.52. eee ee ete ee et 222 MSS POME CHISCROOONUMSL IA sos elk eats amtie key Aogyeahele eee ek eo ats Dele ek dele wle's 225 Structure and development of Lepidechinus................04- Oe ela at tite 220 SC ISIOTIS SS GE eke er enn eae ay” eee a BEA AMR A nh ra 229 semen onectias, ads Mer DEATIIG acts Aa ke < aie Sie Ge «: Suk eve w ss Sard wider e kes 229 A proposed new classification of Paleozoic Echini, with a sogisineuie table Gimunensevictoa WOrOUpomaiac {ta dere Menon cs cee Sadun doc pe e's 238 east SO lect MC Gl SSIMCALIOM sch no ere aces ee eres Griese ee Sate wee siea 238 Subclass Paleeechinoidea—Order I, Boi aiecideavatas i Ao aa Ve 238 SimwelaccsmoMechmmGi ded teec ek dete ee Accs ebene Gh cha el ets sles Sa cals 239 Subclass Paleeechinoidea—Order II, Perischoechinoida.............. 239 XXIV—Bott. Geon. Soc. Am, Vou. 7, 1895. (171) i7Z R. T. JACKSON—STUDIES OF PALAECHINOIDEA. Page IWelonitidsen 2262.0 ne. ee aa eee pea Eke cee 239 Bepidesthides oh): Bas fun Saat ee ene gana RE oa es 241 Arghzocidanide. oss. eer een eee loa aehtoee Rue ee 241 Lepidocentride..... srw) Walls nla Seis Colape in a) pet perce) ole Sega eet a eee eee 241 Order III, Cystocidaroida.......... a ERE AA ER crs, 5 242 Order IV. Plesiocidaroidace Aixvt he aa! Bass Sea OeK web ae oe ene 243 List of publications quoted and referred to in the text.....................-. 244 ixplanation of plates..¢ 25 22h e. sac eee Mele os oa ne eee 246 STUDIES OF THE MELONITIDA. INTRODUCTION. In the following studies of the several families of the Paleeechinoidea it is the intention to take up for consideration the several genera and species in their natural systematic order of sequence as expressed in the proposed new classification (see table facing page 242). This intention is carried out, as far as present knowledge admits, in families succeeding the Meloni- tide. In the present family of the Melonitidee, however, the rule is de- parted from, because the genera and species of which fullest knowledge is attained are the more specialized, and the more primitive genera are least known as far as availabie material goes. In this family, therefore, the genera and species are taken up in the order of convenience. for handling the several cases rather than in the proper systematic order. Thenatural systematic sequence of the genera would be, Rhoechinus, Palxechinus, Oli- goporusand Melonites,as shown in the table facing page 242, For sources of material, obligations due, and similar statements, the reader is referred to the introductory, page 135, of the preceding paper: “Studies of Melonites multiporus.” References to this earlier paper are frequently made by page number, without further quotation, in the present paper. DESCRIPTION OF MELONITES GIGANTEUS, SP. NOV. Plate 4, figure 19; plate 5, figures 21-24. ‘Table, page 180. This new species. Melonites giganteus, is described from a single indi- vidual, which is a truly superb, finely preserved specimen, from the Lower Subcarboniferous of Bowling Green, Kentucky. It is in the col- lections of the Museum of Comparative ZoGlogy, catalogue number 2989. The specimen is entirely silicified. The melon-like form is preserved with hardly any distortion on the oral side (plate 4, figure 19), and on the other side, though somewhat crushed, shows much structural detail. The oral and aboral terminations of the ambulacra and interambulacra are almost perfect in some areas, but orally they are wanting in the first row of plates (plate 4, figure 19); corona very large, exceeding by far any pre- viously described species of the Paleeechinoidea ; height at the point of ereatest dimension, 11.5 centimeters ; height from oral to anal area through DESCRIPTION OF MELONITES GIGANTEUS. 173 the exact center, 10 centimeters ; greatest horizontal diameter through the middle of corona, 15.5 centimeters. These measurements are somewhat affected by a slight dorso-ventral compression, which while reducing the height exaggerates the horizontal diameter. The interambulacra at the widest part measure 4.1 centimeters ; at the narrowest part, at the oral area, where three plates only exist, 8 mill- meters. Following the curve of the area in its center, the interambulacra measure 18.7 centimeters in length. At the oral end, as far as preserved, there are three plates in the first row of four of the interambulacra. Passing progressively from this point dorsally, new columns are intro- duced until we find the greatest number attained, 11, at a point about two-thirds of the distance to the apical pole (plate 5, figure 21). The am- bulacra at the ambitus are narrower than the interambulacra, measuring 3.5 centimeters at the widest part, at which area there are 12 columns of quite irregular plates. At the ventral termination the ambulacra are 1.4 centimeters in width, thus surpassing at this point the width of the in- terambulacra. Ventrally there are but four columns of ambulacral plates (plate 4, figure 19; plate 5, figure 22), asin Melonites multiporus. The plates of the test are very thick (plate 4, figure 19), those at the median zone measuring 8.5 millimetersin thickness. The sides of the plates are slightly inclined to allow for mutual contact in a curved test, but are as nearly perpendicular as the case admits. The plates of theadambulacral columns are extended under the adjacent ambulacral plates, as in Melo- nites multiporus (plate 2, figure 5); otherwise the plates of Meionites giganteus show no tendency to imbrication, and with their thickness emphasize a very considerable rigidity of the test. The interambulacral areas are very much elevated (plate 4, figure 19) and present a comparatively sharp angle where the sides dip down to meet the ambulacra. The ambulacra are sharply elevated in the median portion, depressed on the lateral borders. The elevations of these areas extend beyond those of the interambulacra in a peripheral line and give the Echinus a melon-lke form in a very accentuated degree. The two adambulacral columns of interambulacral plates are pentag- onal, as in other species of the genus, and are crenulated on their outer borders (plate 5, figures 21 and 22) by impact with adjacent ambulacral plates, 8 of which commonly abut against each interambulacral plate. The plates of the median interambulacral columns are hexagonal, except- ing the terminal plates of columns as added, which are pentagonal, and adjacent theptagonal plates ; also excepting the newly added dorsal rhom- bic plates and such others as are described in the detailed consideration of this area. : ) In the ambulacral areas the plates are, for the most part, ambiguous, 174 R. T. JACKSON—STUDIES OF PALHECHINOIDEA. as is the common condition in specimens of Melonites. From the small size and irregular outline of the plates, their contours are largely de- stroyed by the process of silicification. Starting with 4 plates at the ventral end (plate 5, figure 22), new plates and new columns are pro- gressively added, as in Melonites multiporus* (plate 2, figure 4), until in the middle of the corona 12 columns were observed. At this region no area was sufficiently clear to be figured, so that in representing the am- bulacrum (plate 5, figure 24) I was obliged to select an area considerably below the median zone which had not acquired the full complement of 6 columns characteristic of each half-ambulacrum in later stages. The portion selected for the detailed figure of the ambulacrum is indicated in plate 4, figure 19, by the letter XY in area B. At the area marked by Y, just above this point, by close inspection the outline of 6 plates may be made out in this half-ambulacrum. It is possible that a specimen more perfect in ambulacral detail might show still more columns added in a higher zone. The plates of the two median columns of the ambulacra a’ 0’ are very large (plate 4, figure 19, and plate 5, figure 24), and the pores in these, as in other plates of the area, exist in the part of the plate lying nearest the interambulacra, as in Melonites multiporus. All the plates observed possess two pores. ‘The plates of the ambulacra and interambulacra are thickly studded with tubercles that formed the base of attachment of spines (plate 5 figure 23; also seen in plate 4, figure 19). There are about 25 such tubercles on the larger interambulacral plates, but no spines are pre- served. There are obscure traces of genital and ocular plates, but no details of form or structure were made out. The interambulacrum of Melonites giganteus is interesting as a study on account of the great number of plates and columns existent and fora comparison of the arrangement of the same with other Paleechinoids. The most perfect area in the specimen, which by the adopted notation is designated as A, is illustrated by Mr Emerton’s very skillful drawing in plate 5, figures 21 and 22. It is also well shown up to and including the introduction of the ninth column in plate 4, figure 19. The arrangement of this same area is also represented, together with the arrangement of the other 4 interambulacral areas, as far as they can be ascertained, in the table (page 180). This table is a graphic proof of the perfection of the specimen, as in it we are able to tabulate the essential details of no less than 42 columns of plates throughout their entire length. In many por- tions, where the surface was eroded, the form of the plates could yet be ascertained by the middle, or lower proximal portion of the plates, which *On account of imperfections from silicification the transition is not shown so clearly as in the figure cited, but enough is visible to prove the similar method of introduction. DESCRIPTION OF MELONITES GIGANTEUS. 175 were still existent. In the area figured all the plates were preserved entire, excepting where vacancies are shown. At the ventral termination of the area, plate 5, figure 21, there is‘a row of 3 plates. This area is also shown as area A in plate 4, figure 19. The median plate 3 is hexagonal and is the initial plate of column 3, the first formed column of median hexagons. Column 3 obviously originates with one adambulacral column on either side, the universal position of this column. Studying this first row, it is seen that on the ventral border the median plate presents an angle; the two lateral plates a comparatively straight edge (plate 5, figure 22). This form of outline corresponds with that seen in Melonites multiporus at the ventral border of area EH (plate 2, figure 2). It consequently corresponds with the ventral termination of the second row of that species, as shown in area A (plate 2, figure 2) ; also the similar area of Oligoporus corey (plate 6, figure 25). From this evidence it is unquestionable that the first row of plates, in the specimen of Melonites giganteus, is absent from separation of the same after the death of the individual. The same feature is seen at the base of 4 areas, A, C, G and J, as shown in plate 4, figure 19; but area His less perfect ven- trally (see table, page 180). This first row of interambulacral plates could not be absent from re- sorption during life by enlargement of the peristome in Melonites giganteus, as discussed in Archxocidaris, on page 215, because in that case the ventral border would probably present a straight line, as in Melonites multiporus (plate 2, figure 3). The two lateral plates, numbers J and 2, at the ven- tral border are therefore the second plates in their columns and form the base of the two columns of adambulacral plates. In the reconstruction of the base (plate 5, figure 22) the ventral plates of the two lateral columns are shown at 1 and 2, as indicated by dotted lines, succeeded by plate 3 and its adjacent plates, which are the first plates shown in the specimen (plate 5, figure 21). The angulated ventral face of this first row would in itself indicate the absence of at least one row, but taken in conjunction with the studies of the same areas in Melonites multiporus and Oligoporus coreyi the evidence is incontrovertible. The fourth column is introduced in the next row above the third by the terminal pentagon 4 (plate 5, figure 21). At its origin it has one column on the left and two on the right, being thus one column too far to the left by our law of alternation. In areas C, E, G and I (plate 4, figure 19), however, it occupies its normal position, with two columns on the left and one on the right... A similar variation is shown in Melonites multiporus (area I, plate 2, figure 2). In 4 of the 5 areas in which this’ plate is shown in Melonites giganteus, it occupies the normal position, as shown in the table, page 180. It is to be noted that this column 4 is the 176 R. T. JACKSON—STUDIES OF PALHZECHINOIDEA. only one in plate 5, figure 21, which does not accord with the law of alter- nation in the introduction of successively added columns. Plate 4 trun- cates the dorsal border of initial plate 3 of column 3, inducing a hexagonal form in this plate, as shown also in Melonites multiporus (plate 2, figures 2 and 3) and Oligoporus (plate 6, figure 25). Thesame condition of affairs existed in 3 other areas (table, page 180). Column 5 is introduced in the second row after 4 by the terminal pen- tagon 5. At its point of origin this column has two columns on either side, the normal position. It has a heptagonal plate on its left ventral border. Melonites giganteus is later discussed as an extreme member of the genus on account of its great number of columns of plates, and in this early introduction of the fifth column there is an interesting bit of correlative evidence. Extreme types are quite commonly highly accelerated in their development, early acquiring features usually appearing at a later stage in less specialized, more primitive allies. Cases of such acceleration are seen in Nautilus, Baculites, Spondylus and Discinisca.* Here in Melonites giganteus we have the fifth column originating earlier than the same column in the more primitive f types, Welonites multiporus (plate 2, figure 2) and Oligoporus coreyi (plate 6, figure 25). The sixth column is introduced by the terminal pentagon 6 (plate 5, figure 21). At its point of origin it has 8 columns on the left and 2 on the right, its usual position. A heptagonal plate, H, lies on the left ventral border of pentagon 6. In two other interambulacral areas, # and G, column 6 originates as in this area A; but in two areas, Cand J, the column originates one column farther to the left with a heptagon on the right (plate 4, figure 19, and table, page 180). The seventh column begins with pentagon number 7, having a heptagonal plate on its right ventral border. This column at its point of origin has 3 columns on either side. Comparing this with the same plates in the other areas we find (plate 4, figure 19, and table, page 180) the same arrangement in areas Gand J. In Cthe arrangement is the same except that the heptagon is on the left side of the pentagon. In area E' the column originates one column too far to the right, so that there are 4 columns on its left and two on its right, the heptagon lying on the left of the pentagon. This irregular position of odd-numbered columns is of very rare occurrence, having been observed in very few specimens in my researches on this family,{ namely; in Melonites multiporus (numbers 3023, 8021, 3016, 8004 and 2992 in the tables, on pages 165-170) and a specimen of the same species observed at Princeton (see page 153). It is somewhat remarkable in these 7 cases *As shown by Hyatt, Brown, Jackson and Beecher. + More primitive according to the views expressed in this paper (see page 199). t An irregular position of column 5 is figured in Lepidesthes coreyi by Meek and Worthen. (See remarks on that species, page 209.) i i DESCRIPTION OF MELONITES GIGANTEUS. iT that the irregularity should have occurred in the seventh or ninth column. A still further irregularity of the seventh column is described under Melo- nites septenarius (plate 9, figure 49). The eighth column originates in pentagon 8, with a heptagon on its left and with 4 columns on the left and 3 on the right, thus being in its correct theoretical position. Comparing with plate 4, figure 19, and the table on page 180, we find that areas F and G are as described, except that has the heptagon on the right instead of on the left. Areas Cand J have column 8 one place too far to the left, having 5 columns on the left and 4 on the right, and the heptagon of C is on the right. The ninth column begins in pentagon 9 (plate 5, figure 21) with 4 columns on either side, asin Melonites multiporus (plate 5, figure 20) and Oligoporus (plate 6, figure 04). The plates below pentagon 9 in Melonites giganteus present a pecu- lar arrangement unlike anything seen in any other specimen of the Palee- chini. Bordering on the pentagon ventrally are two heptagons, H H’; below A there is a third heptagon, H”, and these 3 heptagons with the hexagon A,enclose a rhombic formed plate. A similar condition of affairs exists in the same relation to pentagon 9, in areas C, Hand J (plate 4, figure 19, and table, page 180). It is elsewhere stated that the plates of the median columns are all hexagonal or its equivalent as a dynamic consequence of the conditions of lateral pressure. This is an excellent proof of the principle, for the extra side of one heptagon, H, compensates for the loss of one side in pentagon 9; the two other heptagons, H’ H”, by their two added sides compensate for the absence of two sides in the enclosed four-sided plate. The enclosed rhombic plate, it is seen, termi- nates ventrally in an angle; the only plates normally doing this are terminal pentagons.* This plate, therefore, is considered as really the first formed plate of column 9 which has become separated from its next dorsal successor, which is pentagonal plate 9. Another case seen of a plate being separated from its successor dorsally is that shown in the ninth column of Oligoporus dane (plate 6, figure 31), where the terminal pentagon has become separated from the next plate of its column. The only other cases seen of plates of a column being separated dorso-ventrally (except near the dorsal pole, where separation normally occurs, plate 3, figure 13) are in the sixth column of Oligoporus missouriensis (plate 9, figure 50) and the seventh column in area I of Rhoechinus gracilis (plate 7, figure 36). Apparently this separation of the rhombic plate of Melonites giganteus (plate 5, figure 21) isa case of slightly arrested development in the ninth column—that is, after the first plate was formed, the rhombic one, no *Excepting, of course, rhombic plates near the dorsal area, where the form is otherwise ac- counted for. {78 R. T. JACKSON—STUDIES OF PALZECHINOIDEA. new plate was added in the next formed row, while plates were added in the several other adjacent columns; when the second row was formed above the rhombic plate a new plate was again added in column 9, but being separated from its ventral predecessor it formed a pentagon. This view appears correct from the fact, strongly brought out in these studies, that accessory or adventitious plates are unknown in the Melonitide, all the plates in every specimen seen being distinctly associated with others in definite columns. One area, G, does not show the peculiarity of the rhombic first formed plate as discussed and which is existent in 4 areas, A, C, Hand I (table, page 180). Up to the ninth column the arrangement of plates in Melonites giganteus can be compared with the similar arrangement in Oligoporus dane (plate 6, figures 31 and 34); also frequently with specimens of Melonites multi- porus (plate 5, figure 20). Above this point, however, it exceeds the number of columns of interambulacral plates of any other species of the Melonitide. The tenth column originates in pentagon number 10, with a heptagon on its right (plate 5, figure 21). Column 10 at its point of origin has 5 columns on the left and 4 on the right. Comparing this with the table (page 180), it is seen that the tenth column is similar in posi- tion in areas £, G and J; the position of the heptagon, however, varies, being on the left in area 2. The tenth column is broken away in area C. The eleventh column originates in pentagon 11, with a heptagon, H, on the right. This column at its point of origin has 5 columns on either side. Itis, therefore, as well as column 10, in its correct theoretical posi- tion, as deduced from the law of growth of interambulacral areas. Turn- ing to the table on page 180, it is seen that the eleventh column is want- ing in one area, C. In 3 areas A, EF and J, it occupies the same position and terminates as in plate 5, figure 21. In area G it originates one col- umn too far to the right. The position of the heptagonal plate in all these areas is on the right,* its usual position in odd-numbered columns. In the sixth plate, in column 11 (plate 5, figure 21), a peculiar pentagonal plate, VN, occurs. This plate is notewurthy on account of its pentagonal form, for it is not a terminal plate of a column. Progressing dorsally, we find in an area equal to about one-sixth the whole length of the interambulacrum that there is a gradual passage from hexagons to plates of a more or less rhombic form, as in Melonites multi- porus (page 149). These young plates have a progressively shorter line on the upper and lower sides and are relatively longer dorso-ventrally than the older hexagonal plates. This change may be seen in plate 5, figure 21. There is not so much drawing out and separation of the plates as in the same area of Melonites multiporus (plate 3, figure 13, and plate 5, * Excepting area G, where, owing to imperfections, its position was not ascertained. ie ae 2 DESCRIPTION OF MELONITES GIGANTEUS. Lugs, figure 20). This may be a specific or simply individual difference, or even a difference due to the age of the individual. There is a very dis- tinct dropping out of the middle and last added column, number 11. This column is reduced to a tiny rhombic plate at its upper limit, and is not represented at all in the last rows built by the echinus as far dorsally as they can be traced. On the other hand, the last (eighth) column in Melonites multiporus extends directly to the genital plate (plate 3, figure 13). Itis an axiom of old age characters that the last features acquired are soonest lost, and it seems that this specimen of Melonites giganteus had entered on its decline, which is shown by the dropping out of a column that at a little earlier period in growth must have been continu- ous to the genital area. During later growth, had such taken place, this last column apparently would not have been existent, so that the indi- vidual in its decline has virtually resumed the condition of building only 10 columns of plates, a feature which was initiated at a relatively early age, aS shown by pentagon 10, plate 5, figure 21 (see page 150).. There is a distinct inequilaterality of the interambulacrum near the dorsal area, as shown in plate 5, figure 21. Columns 9 and 10 are quite alike on the two sides of the center, but columns 7 and 8 are, on the contrary, very unlike, as are also columns 5 and 6. In each of the 5 interambulacral areas of Melonites giganteus the several columns, as demonstrated in the fourth column of the table (page 180), made their appearance at exactly or nearly the same horizon. The great- est exactitude in the period of introduction is maintained in the earliest added columns, especially numbers 3, 4, 5, 6, and 7, while in the columns appearing last there is more variation, as might be reasonably expected. Stating the case briefly, in all areas the third column originates in the second row and the fourth column in the third row, asin Melonites multi- porus. The fifth column originates in the fifth or sixth row; the sixth column in the eighth row; the seventh column in the tenth or eleventh row; the eighth column originates in the twelfth to fourteenth row; the ninth column originates in the sixteenth or seventeenth row; the tenth column originates in the twentieth to twenty-third row, and the eleventh column originates in the twenty-fifth to the twenty-ninth row. Compar- ing this result with that shown in Melonites multiporus, where it has been ascertained in a large number of specimens (page 162), we find that after the fourth column each of the later added columns originated much earlier in Melontes giganteus than in Melonites multiporus. This species, therefore, is not only furthest advanced in the special line of variation of the genus, but it is also highly accelerated in its development, very early passing through those stages seen in later growth in less specialized mem- bers of the genus. 180 R. T. JACKSON—STUDIES OF PALEZECHINOIDEA. Table of Plate Arrangement of Melonites giganteus, sp. nov. ‘ z 2 eae eet oe eis B > 4 = ta a iS) re is > ~~ = Sa|/ oOo a | 9° ee ieebacsneesen 2 fl > lad = | == | Heptagon on left or right evoniles gigan eus, sp. nov. s a 5 4 2S S- Bn x terminal pentagon. ef] 2] 8 | 2° | at} as ro) Ss = eS a r=} = & 3 a & & aaa Re i ag = Ja eh ve Zsa} ech Oe eee ey Latest 1 Interambulacrum A f......cscccessssess 11 11 1 28 5 5 | Right. (This area is figured on plate 5, 10 P 21 5 4 Right. figure 21.) of le 17 + 4 Right. 8 Pp, 14 f 3 Left. 7 P | lo 3 3 | Right. 6 2 8 3 2 Left. 5 P 5 2 2 Left. 4 P 3 1 2 | Truncates initial plate 3. 3 >.< 2 1 LY J Sevscnuapevesssavesssen dieogangat ove cecepae First row of plates wanting. AraGulser vin, Bi 2, ¢..sascesede sanbaenannst Four columns of plates at base. Interambulacrum C¥.......scc000essese|eseessess 9t |} P | 16 4 4 | Right. 8 P 13 3 4 | Right. 7 Pol tid, 2 3 | Left. 6 1e 8 2 8 | Right. 5 Ee 6 2 2 Lett. 4 ie 3 2 1 Trunceates initial plate 3. 3 >.< 2 1 EM Dilexte days asap buvaupane dost suppenncnieneenae First row of plates wanting. AM DUlaeriMy B).cscetesscs caseastaeperacer | Four columns of plates at base. Interambulacrum EF .........-..c0000 1L 11 Le 29 5 5 Right. 10 P 23 5 a (?) Broken away. of P 17 1 4 (2?) Broken away. 8 P| 12 4 3 | Right. a Ye 10 4 2 Left. 6 P 8 3 2 Left. 5 le: 5 ie 2 Right 4 r 3? 2 A) teiviaeies uporgundisaius tary Asnssweppeseradore | First two rows of plates wanting. Am DU1a Cristy: cccsecssezsscaesesenonss | Four columns of plates at base. Interambulacrum G F......ccccseeececees 11 HH P 27 6 4 | (?) Broken. 1,20 1% 21 6 4 (2) Broken. rg P 16 4 4 Right. 8 ie 14 4 3 Left. 7 is 11 3 3 Right. 6 P 8 3 2 Left. 5 Y 6 2 2 | Right. , 4 es 3 2 1 Truncates initial plate 3. 3 xX 2 1 Dh i Seeace vols cnstraceae an asuenqepneneeeen First row of plates wanting. * Interambulacrum incomplete at dorsal portion of area, + Interambulacrum complete at dorsal portion of area. t See foot-note, page 181. P = Pentagon. X= Hexagon. DESCRIPTION OF MELONITES GIGANTEUS. 1Sl Table of Plate Arrangement of Melonites giganteus, sp. nov.—Continue@, g b » » 3 veel Pou ae Ss = =) 2) iS) S| ro BI = ~~ = S43 ey S| ° SB 9 S 2 2 hee Wee Be Heptagon on left or right . . S) aS f : Melonites giganteus, sp. nov. % % S| 2 ae a Sa Gi ceatinlpencacon! 43 a pe 2 ne ne? 1,2 12/2 |2 |: 2 gest Tee S Ss E 50 =} 5 = & = | = = Zeta ey wert reine S| ——— Avpalouilaerwin Ei ss...0.c..sessccesserooecanes Four columns of plates at base. Interambulacrum I F..............0cce00 11 11 P 25 5 5 Right.* . E 10 12 20 5 y Left. 5 OF P 17 4 4 Right. 8 12 14 3 h Left. 1 1e 10 3 3 Right. 6 P 8 2 3 Right. 5) 12) 5 2 2 Indistinct. 74 P 3 2 1 Truneates initial plate 3. 3 x 2 1 ics, jleecs csv eesvecersecercessas'obersasesteessn First row of plates wanting. AUN DUIACTUM Ji... ce..ccssupesesessecnesesees Four columns of plates at base. In the interambulacrum A (plate 5, figure 21), in column 1, from the ventral to the dorsal end, as far as preserved, 29 plates may be counted ; in column 2, there are 34 plates. If column 1 were as perfect dorsally as column 2 there would without doubt be 34 plates also. Column 3 has oo plates, column 4 has 33 plates, column 5 has 33 plates, column 6 has 00 plates, column 7 has 28 plates, column 8 has 25 plates, column 9 has 23 plates, column 10 has 19 plates, and column 11 has 12 plates. Add- ing these, there are seen to be 801 plates existent in the interambulacrum (plate 5, figure 21), or, adding the 9 plates wanting in the first column, o10 plates in all. Supposing the other 4 areas to have approximately the same number, as is the fact, then the interambulacral plates of this specimen are no less than 1,550 in number, to which may be added the two ventralmost plates wanting in each area and a few dorsal plates not preserved clearly enough to be made out. It is believed that the columns of ambulacral and interambulacral plates as introduced represent stages in growth, as discussed in the sec- tion ‘General Results and their Bearing.” Assuming this to be correct, then Melonites giganteus when young had 4 columns of ambulacral plates, * But not in next column. Compare with heptagon H, associated with pentagon 8, of plate 2, figure 7. + Interambulacral complete at dorsal portion of area. {Terminates with accessory heptagons and tetragonal plate, as shown at this area in plate 5, figure 21, and plate 4, figure 19. There is no similar peculiarity about pentagon 9 in area G. 182 R. T. JACKSON—STUDIES OF PALAECHINOIDEA. as shown by the ventral area of the same. In this feature it, as well as Melonites multiporus, is like the adult of Oligoporus ; later these columns in- crease to 12. Thisisa higher number than is attained by any other species of the genus except Melonites etheridgii, W. Keeping* (see systematic table facing page 242). In the adult, Melonites giganteus has more col- umns of interambulacral plates than Melonites multiporus or any species of the genus. It may, then, in this salient feature be considered the ex- treme species of the genus, being furthest removed from the ancestral stock, which must have had relatively few columns, as evidenced by the stages in growth through which it has passed. To put it in other words, when very young it had at most 3 columns, then 4, like Melonites dispar, (Fischer), and next 5 columns, which is characteristic of adult Melonites crassus, Hambach ; when older still it had 7, then 8, like adult Melonites multiporus ; later 9 columns like extreme cases of Melonites multiporus ; finally it goes ahead of anything found in other species, and has 10 and 11 columns.t} In the accompanying table are shown the relations and form of the plates in the several interambulacral areas of our specimen as far as they could be ascertained. In studying this table comparison is requested with the figures of this species, with the tables of Melonites multiporus (pages 165-170; see also page 161), and with the figures of other species and genera of Paleozoic echinoids illustrated in the accompanying plates ; also it should be considered in connection with the systematic table of classification of Paleozoic Echini, facing page 242. DESCRIPTION OF MELONITES SEPTENARIUS, SP. NOV. (R. P. WHITFIELD). Plate 9, figure 49. In the American Museum of Natural History in New York, there is a specimen of Melonites from the Warsaw group, Subcarboniferous, of Buz- zard Roost, Franklin county, Alabama, lower limestone. No specimens of Melonites have been previously recorded from the Subcarboniferous of the south Atlantic states. This species differs from any previously described, and for it Professor R. P. Whitfield suggests the name Melo- nites septenarius, the name indicating the number of columns of plates in the interambulacral area. The specimen (plate 9. figure 49) is a silicified cast from the interior, but in parts shows the original thickness of the *'The type of this species is in the Museum of Practical Geology, Jermyn street, London. + In the collections of the Wagner Free Institute at Philadelphia there is a specimen, catalogue number 5255, which is ascribed to the species Melonites giganteus. It is from the Sub-carbonifer- ous of Tennessee. The specimen, which is fragmentary, corresponds in details of size and pro- portions with the type. In the ambulacrum there are 12 columns of plates, but in the interam- bulacrnm there are but 9. Ten rows of plates are added, after the introduction of the ninth column, without the introductioi of a tenth column. This fact is a striking difference from the type, as described above. DESCRIPTION OF MELONITES SEPTENARIUS. 183 plates. It is a small species, being one of the smallest known. It is not perfect ventrally ; but the height of the test as far as shown is 4 centi- meters. - The ambulacral plates are not well preserved, but can be made out in places. In the ambulacrum on the right of the figure there are 4 col- umns of plates at the ambitus ina half area. This shows that the species is characterized by 8 columns of ambulacral plates—an unusual number in the genus. Two pores exist in each plate. The width of the ambu- lacrum at the ambitus is 1.8 centimeters, narrowing toward the dorsal area. The interambulacrum ventrally has, as far down as preserved, 6 col- ums of plates. A seventh column is introduced by the pentagonal plate 7, with a heptagonal plate, H, on its left. This seventh column at its point of origin is remarkable in that there are 5 columns on the left of it and only 1 on the right. Odd-numbered columns commonly originate in a median position, with an equal number of columns on either side. Rarely exceptions are found, as shown in the tables of Melonites multiporus (pages 165-170), in which, in several cases, an odd-numbered column origi- nates to the right of the center, with one more column on the left than on the right; but no case has been observed in any Paleechinoid other than the present one in which any greater degree of irregularity existed. This interambulacrum is also peculiar in that a plate, P, normally hex- agonal, is pentagonal in outline, and an adjacent plate, A, is heptagonal. No other case of irregularity quite like this has been seen in any of the Melonitide. ‘Toward the dorsal termination of the area the interambu- lacral plates make an approach to the rhombic form seen in other species (plate 3, figure 13; plate 5, figure 21). The width of the interambulacrum at the ambitus is 2.3 centimeters. The interambulacrum is quite elevated in sectional outline, but presents a continuous curve rather than almost an angle on its lateral borders, as in Melonites giganteus (plate 4, figure 19). The nearest ally of Melonites septenarius is Melonites indianensis, Miller and Gurley * (84). It differs from that species in the proportionately much narrower ambulacra, in having 7 instead of 6 columns of interam- bulacral plates, and in the gently curving, rather than strongly melon- like form of the corona. The number of columns of ambulacral plates is an important feature in classification, and the question comes up whether species having eight columns, as Melonites indianensis and septenarius, should be separated ge- nerically. We think not, because in the genus Melonites the number of columns of ambulacral plates is quite a variable feature, as shown in the *The type of this species is in the private collection of Mr Wm. F. E. Gurley, of Springfield, Illinois. 184 R. T. JACKSON—STUDIES OF PALZECHINOIDEA. several species (table facing page 242). One species, M. dispar, has but 6 columns; another, J. giganteus, has 12 columns, and one, M. ethe- ridgit, has 12 or 14 columns of plates; also, while Melonites multiporus has 10 columns of ambulacral plates as the feature of the species, it may be that some specimens may only show 8 columns throughout the area. In this species, in one specimen figured (plate 5, figure 20) only 8 columns of plates exist at the ambitus (as shown in the lower portion of the am- bulacrum figured), whereas a little further dorsally Gin the upper por- tion of the ambulacrum figured) 10 columns may be counted. These two last added columns in this specimen are added later than usual, and it is conceivable that in some cases they might not be added at all. CONSIDERATION OF OLIGOPORUS AND COMPARISONS OF THE SAME WITH MELONITES. DESCRIPTION OF OLIGOPORUS MISSOURIENSIS, SP. NOV. Plate 9, figures 50-52. A fine new species of Oligoporus has recently come to hand from the Subcarboniferous of Webb City, Missouri. The exact horizon from which this specimen came is somewhat uncertain, but Dr C. R. Keyes, chief of the Missouri Geological Survey, kindly informs me that it is in all probability from the Augusta limestone. To this species I would give the name Oligoporus missouriensis, which is appropriate in recognition of the fact that this state has yielded such rich material for the elucidation of the complex structure of Paleozoic Echini. The specimen is a silici- fied cast free from matrix and is very slightly compressed. It is in the collection of the Museum of Comparative Zoélogy, catalogue number 3078. The specimen measures 9 centimeters in height through the dorso- ventral pole ; it is somewhat compressed, so that the dorso-ventral meas- urement is slightly exaggerated ; greatest width through the ambitus in the plane of compression, 9.6 centimeters ; width at right angles to plane of compression, 7 centimeters. The outline of the test presents an even, almost continuous, curved outline, the ambulacral and interambulacral areas presenting very little elevation beyond the outline of the whole. The ambulacra consist of 4 columns of low plates in each area; width of ambulacra at ambitus, from 1.4 to 1.6 centimeters, narrowing at the dorsal area. Accessory plates, as seen in Oligoporus danzx (plate 6, figure ~ 30), not present. Two pores are in each plate. The spinose projections | of the specimen which represent the ambulacral pores lie near the middle of each half ambulacral area (plate 9, figures 50 and 51). This is attributed to the fact that the specimen is an internal cast, and that the pores pass toward the center of the half-areas in traversing the thick- DESCRIPTION OF OLIGOPORUS MISSOURIENSIS. 185 ness of the plates, although on the outer or distal side of the plates they probably existed in that portion of the plate which was nearest the inter- ambulacra, asin Oligoporus danx (plate 6, figure 30). A similar condition of pores in the center of each half-ambulacrum is shown in the view from the interior, or proximal side, of Oligoporus coreyi (plate 6, figure 25). In Melonites multiporus the ambulacral pores in traversing the thickness of the plates pass toward the center of the ambulacral area instead of each - half-area, as shown in plate 2, figure 5 (see page 141). In Oligoporus missouriensis about 5 ambulacral plates are apposed to each interambulacral plate. The ambulacral plates are more regular in outline than in any other species of the genus seen. One peculiarity not seen in any other Paleozoic echinoid, is the fact that ambulacral plates which lie opposite the horizontal sutures between interambulacral plates are spread out in a fan-like fashion on the outer border, as shown in the figure. Interambulacral areas measure about 4.1 centimeters in width at the ambitus, narrowing toward the poles. Adambulacral plates are rounded on the ambulacro-interambulacral suture. There are 6 columns of inter- ambulacral plates at the ambitus, and no more are added in the dorsal portion. The sixth column in area A is introduced by the pentagonal plate 6, which is discontinuous from the next plate, 6’, of its series, as in the ninth column of Melonites giganteus (plate 5, figure 21). Around this plate, plate 6, there is an unusual arrangement of plates. A pentagonal plate, P, lies on its left border, and an octagonal plate, O, by its two added sides, compensates for the loss of two sides in plates 6 and 6’.. The sixth column starts a second time in pentagonal plate 6’ and adds a second plate on the dorsal border of it. Then the sixth column dies out and is seen no more in this area (see page 150). To take up the space where this column has dropped out there is an enlargement and irregular arrange- ment of plates in the fifth and fourth columns. The plates P’ P” P’”, in eolumn 5, which should be hexagonal, take on a pentagonal form, and one plate, H, of column 5 is heptagonal and extended to the right so as to cover the dorsal border of the last formed plate of column 6. Besides these, to compensate for loss of sides in the pentagons, there is a hep- tagonal plate, H’, in column 4 and another heptagonal plate, H”, in column 8. This is one of the most unusual irregularities seen in any Paleozoic Echinoid, but it is all in accordance with the laws of growth when the mechanical conditions are ascertained. Two other interambu- lacral areas, H' and G, not shown in the figure, have a similar dying out of column 6. No other case has been seen in any type of a column originat- ing comparatively early in the hfe of the individual and then dying out after building a few plates, except as shown in Lepidesthes wortheni (plate 9, 186 R. T. JACKSON—STUDIES OF PALEZECHINOIDEA. figure 55). One of the interambulacral areas, C, has only 5 columns as the greatest number, and this abortive introduction of a sixth in three areas shows that 6 columns have not become a fixed specific character. Another specimen has 5 columns in all 5 areas. One interambulacrum, I, is too imperfect for details to be ascertained. Four columns of plates are made out at the ventral border of the interambulacra as far as they ean be traced, but the specimen is imperfect at the lower portion of the corona. New columns of plates are introduced by pentagonal plates with adjacent heptagonal plates, as described in the several species of Melonites and in Oligoporus coreyi and O. danx. Dorsally the newly added plates are more or less rhombic, as in Melonites and Oligoporus danx (plate 6, figure 34). Surface ornamentation of plates is like plate 6, figure 35. Part of the genital and ocular plates are preserved, and this is impor- tant for comparison with the only other species of the genus Oligoporus nobilis, Meek and Worthen, in which they have been observed. In Oligo- porus nobilis, according to Meek and Worthen (381), 3 genital plates showed 5 pores, while 2 showed 4; the ocular plates are stated as imperforate. In Oligoporus missouriensis (plate 9, figure 52) the form of the genitals and oculars is the same as in Melonites (plate 3, figure 13). Two genitals have 4 pores and one 3, instead of 5 and 4, as in Oligoporus nobilis. The two oculars preserved are imperforate, as in that species and Melonites, and reach to the periproct. Oligoporus missouriensis differs from Oligoporus coreyt (plate 6, figure 28), which has 6 columns of interambulacral plates, by its more massive pro- portions; also by the smaller relative size of ambulacral plates, more abutting against an interambulacral plate than in Oligoporus coreyi. It differs from Oligoporus blairi, Miller and Gurley (34), in the same features ; also in being nearly circular instead of melon-like in form, as in that species. It differs from all species of Oligoporus or Melonites seen or de- scribed, in the nearly circular form and in the fact that the interambu- lacral plates at the junction with the ambulacra present a gently curving outline much as in Archxocidaris (plate 8, figure 48), instead of an in-, dented sutural line, as in Oligoporus dane (plate 6, figures 80 and 31); also in the peculiar fan-like form of certain ambulacral plates, as de- scribed. DESCRIPTION OF OLIGOPORUS COREYI. Oligoporus coreyi, Meek and Worthen (32), has never been figured, but was described by its authors from a single specimen from the Keokuk group of Crawfordsville, Indiana. A specimen in the collections of the Museum of Comparative Zoology (catalogue number 3008), after care- ful study, is referred to this species. This specimen (plate 6, figures 25, DESCRIPTION OF OLIGOPORUS COREYI. 187 28 and 29) agrees with the description of Oligoporus coreyi in the follow- ing characters: Body small. globose, moderately thick plates ; interam- bulacra twice as broad as ambulacra ; ambulacra composed of 4 distinct columns of plates; interambulacra in the middle composed of 6 columns of plates; pores of the ambulacral plates situated near the outer edge. The pores are near the center of each half-area in plate 6, figure 25, but that is because the specimen is viewed from the inner or proximal side of the test and the pores pass toward the center of the half-area in trav- ersing the thickness of the plate (see Oligoporus, plate 9, figure 50; see page 184). The height of Meek and Worthen’s type was 1.65 inches ; the specimen here described, as far as preserved (plate 6, figure 28), measures 1.87 inches in height. If perfect dorsally, it would probably add at least half an inch to this measurement. If the type was a reasonably entire specimen, which is not stated, our specimen would probably be a little longer in the dorso-ventral axis. The type is stated as being about two inches in breadth and the interambulacra are stated as twice the width of the ambulacra. The authors do not say what the width of the ambulacra and interambulacra is, but we can estimate these areas ap- proximately from their measurement of the breadth. If the breadth. which is two inches, represents fairly the diameter, as may be assumed, then the circumference may be attained by multiplying the diameter, 2 inches, by 3.14, which gives 6.28 inches, or 15.8 centimeters, for the cir- cumference. As the ambulacra are half the width of the interambulacra, therefore the width of the ambulacra must be one-fifteenth of the circum- ference, or 1.05 centimeters; the width of the interambulacra would be twice that amount, or 2.1 centimeters. Comparing the specimen with this ideal measurement, I find that the ambulacra measure at the widest part 1 centimeter and the interambulacra 1.9 centimeters. These meas- urements, taken with the number of columns of plates, the inferred size of the plates, etcetera, render it entirely probable that the specimen is Oligoporus coreyr. Inthe type the surface is described as unknown, but in this specimen the plates of the interambulacrum are thickly covered with small bosses for the attachment of spines (plate 6, figure 29). No spines are preserved. The specimen of Oligoporus coreyr here described is labeled as from Indiana, but the label does not give a detailed locality or geological hori- zon. The specimen is composed of thoroughly crystallized calcic car- bonate, which is stained reddish brown with oxide of iron. It differs lithologically from Crawfordsville material and cannot, therefore, be ascribed to that locality. The type was from the Keokuk group, and hay- ing no evidence to the contrary this specimen is provisionally ascribed to the same horizon. XXV—Butt. Georn. Soc. Am., Von. 7, 1895. 188 R. T. JACKSON—STUDIES OF PALHECHINOIDEA. The species Oligoporus corey differs from Oligoporus dane as described in the original publication, which is here confirmed, in being much smaller and apparently more depressed in form; in having more deeply furrowed ambulacral areas,* and also only 6 columns of interambulacral plates. From Oligoporus nobilis it differs by its smaller size and more deeply sulcate areas, which are proportionately wider. Oligoporus coreyt differs from Oligoporus mutatus, Keyes (23), and O. missouriensis, Jackson, in having relatively much smaller plates and in being much smaller as a whole; also in being a fine delicate species rather than robust in its proportions; it also has much less of a melon-like rotundity in its areas than O. mutatus, as well as one more column of interambulacral plates than that species. It is shown in echinoids that, as the individual grows, new columns of interambulacral plates are progressively added until the number normal to the species is attained. It may, therefore, be properly questioned whether this specimen is an adult, and not the young of another species as Oligoporus danz, which in later growth, had the animal lived, would have added more columns of plates, and the plates themselves have in- creased in size. In many cases this might bea difficult question to decide, and it is possible that species of Paleeechinoids have been based on im- mature specimens; but from the very principles of growth involved we find the answer in this case. The first 6 columns of plates are here, as in all other cases studied, introduced comparatively early in the life of the individual, in this case the sixth originating in what is probably the ninth row from the oral end. 2,503.0 .00365 | .01141 .01506 Site te Denes gens aan eee 3576.7 (1 0058) | “ aiiesh 02151 re PE rORS, 63's ee ate al 1,494.1 | O08 |” .60G8Y .00899 Alfio. Parmahy bay, . om cd ses - 96556 | .00141 | .00440 | .00580 DPELRe 4 alee ee 1,560.8 | .00227 | OO7IL | s0069 .00896 Pernam pues... os aa eee, ae SOL pe .00433 .01354 .01787 ALLE: PPR eats me ae Whe 1,050.5 £00158 .00478 .00631 Dglonia Isabel... Ji sceae canes 1,037.0 00151 | .00472 .00623 Sento das Lares...) ae. 2,179.5 .00318 | .00993 .O131LL LEAL a eg a AE st ata 2,394.8 00349 | .01091 .01440 Determinations have been made at several places in the world of the amount of ammonia brought to the earth in rain water. This ammonia must also be counted among the indirect agencies of rock decomposition, though it is perhaps no more important in Brazil than elsewhere, except. in so far as the amount is greater, as shown by Muntz and Marcano, and as the warm climate may increase its activity. Warington has shown that all nitrogenous substances which yield ammonia are nitrifiable.* Ammonia in contact with organic matter is soon converted into nitric and nitrous acids. In water analyses, for instance, it is understood that waters which give ammonia when fresh, yield only nitric and nitrous * Jour. Chem. Soc., London, vol. xlv, p. 653. INFLUENCE. OF RAINFALL. 309 acids after standing for some time. Nitrification takes place principally if not exclusively in the surface soil.* In order to estimate the nitric acid in the ammonia falling in rain in Brazil we are compelled to get our ideas of the amount in the rain-water from the determinations of Muntz and Marcano at Caracas, Venezuela. The total ammonia falling in the rain at that place on being oxidized to nitric acid amounts to 6.975 milligrams of nitric acid to the liter of rain- water.f Thus the table on page 308 gives the total depth of nitric acid fall- ing in the rains in Brazil, both in the form of ammonia and as nitric acid . RAINFALL. It is evident that the rainfall must be an important factor in all this rock decomposition ; both the amount and the time distribution are im- portant elements. At one of the stations from which we have a record— that of the Alto da Serra do Cubatao, where the Santos a Jundiabhy rail- way crosses the Serra between Santos and Sao Paulo—an average for 15 years gives a rainfall of 3,576.7 millimeters (140.81 inches, or more than 11 feet),and from this extreme it declines to about 50inches. Moreover, the -rainfall is very unevenly distributed throughout the year, most of the precipitation occurring in three or four months. This same precipita- tion, large as it is, if more evenly distributed throughout the year would do only a fraction of the eroding that it does when thus poured in torrents upon the earth. The year is roughly divided by the people into the two seasons which are known as the sunny weather (tempo de sol) and rainy weather (tempo de chuva). 'Thetempo de sol is the time of cool weather in that country— usually the months of May, June, July, August and September—the sol referring not to the heat, but to the continuity of the sunshine. The rainy season is the hot part of the year, and this is a point to be borne in mind, for the rains alternating with hot sunshine, the waters fall upon hot rocks or soils and their chemical activity is greatly increased by this increase of temperature. The concentration of the rainfall in a few months of the year is a con- stant feature of the Brazilian climate, although it often varies consider- ably from one year to another—that is to say, November may be a very rainy month one year and comparatively little rain may fall the follow- * Warington: Jour. Chem. Soc., vol. li, p. 118. - +Comptes Rendus, 1891, exiii, p. 780. t Lake Bonneville. G, K. Gilbert. Monograph I, U.S. Geol. Survey, pp. 41, 42. This principle is employed in the process of gold washing known as “ booming.” 310 J. Cc. BRANNER—DECOMPOSITION OF ROCKS IN BRAZIL. ing November, and this may be true not only of one but of several of the months. | Traveling in the interior of the country at certain seasons of the year, one is impressed by the dry beds of what are, at other times, large rivers. This is especially striking in the northeastern part of Brazil, through the interior of Bahia, Pernambuco, Parahyba, Rio Grande do Norte, Ceara, and Piauhy. Streams that are at one season large enough to float an ocean steamer and from a hundred to two or three hundred miles in length, often, toward the end of the dry season, are reduced to a series of pools or to a line of hot white sand. ‘This fact is so well known that it is scarcely necessary to do more than mention it here, and the statement is not intended to apply to years of drouth, but to the average condition of the region in question. In Bahia, in the month of May, Spix and Martius found the Rio do Peixe and the Rio Itapicurd, which is over 200 miles in length, only a string of pools. * Barao Homem de Mello, who was better acquainted with the physical geography of Brazil than any one else, says, in speaking of the plains of Ceara : f ‘‘PDuring this period (the dry season) the beds of streams, here improperly called rivers, dry up entirely. In this province, however, these are nothing more than channels or courses of torrential waters during the rainy season. Thus I crossed today the perfectly dry beds of the Bahti and Guayuba between Acarape and Paca- tiba. Along the large streams, such as the Jaguaribe, which is more than 600 kilometers long, there are barely a few pools here and there, the water ceasing alto- gether to flow.’’ Thisis borne out by Gardner, who says that at Icé the Jaguaribe, ‘“‘which during the rains is of considerable size, becomes quite dry’’ in the dry season. f In the Rio Grande do Norte Koster found the Ceara Merim in Novem- ber a dry bed its entire length above tidewater. The Rio das Paranhas, the largest stream in that state andin Parahyba do Norte (250 miles long) he found dry at Act, near its mouth, on December 1, but on recrossing it two weeks later he found it overflowing its banks and ‘‘two to three hundred yards in breadth.” § Even under the equator the rainfall is thus unequally distributed. There are dry treeless plains at many places along the Amazon. Hartt, * Reise in Brasilien, ii, p. 724. + Excursoes pelo Ceara. F.I. M. Homem de Mello. Revist. do Inst. Hist. do Brazil, 1872, xxxv, pt. 2, p. 85. t{ Edin. New Phil. Jour., April, 1841, p. 76. 2 Travels in Brazil. Henry Koster. 2d ed., London, 1817, vol. i, pp. 113, 147, 217. DISTRIBUTION OF RAINFALL IN BRAZIL. oll speaking of the dryness of the climate of the Amazon valley, says that the forests of the Monte Alegre, Eréré district, and of Santarem bespeak “during the dry season a very dry climate and a fault of moisture.” * My own observations show that the plains north and northeast of Macapa are so dry and parched during a large part of the year that trees erow only along the streams or in other favored places. The rains, even during the rainy season, have a decided torrential char- acter. Pompeo de Souza mentions a rain at Ceara in 1855 when a rain gage, 200 millimeters deep, overflowed. f Using Dr Draenert’s table of mean rainfalls, compiled from all the data obtainable at the time of its publication, in 1886, we find that the pre- cipitation was distributed as shown in the following table: Distribution of Rainfall by Semesters. t . : Six rain Sine dir No. of Stati Rainfall (in aie ee ation. millime- . aria : cao years ters) (in millime- | (in millime- : ters). ters). 29 EMORG EIS ATICILO, «5 s.2h)< 6 nisi se sus > oe 974.6 632.9 341.7 15 SS ITM ONS eee UH Ons ned a eisious cays» =, ate 2,503.0 1,708.9 794.1 15 Mito dasserra,.o. Pauls 0. a6. % 3,076.7 2,289.5 126722 + SAO) TERNDUIO) RAM iia ere eae eae er ene 1,494.1 1,152.6 341.5 1 EAU Ole INA! ch ons 6 yaer eye @ nage 29.6 965.6 748.0 217.6 BREW eraba oes kok ee a eae eee e. 1,560.8 1,292.5 268.3 25 Morro ’Velho. Minas... ....... -6. 1,637.0 1,457.0 180.0 2 Gono SOco,MiMag tis bi. scents ae 2,939.3 POEL 429.2 il Mipen toni WEMMAS: oe aie eee aywte aye we § 1,303.5 WE AZORS IS2E7 ee ouela, Minase eos. oe. see ee 1,453.1 Lae! PAO Lag 4 Mansos, AMazONAS? ss... .5.2<+-: 2,340.4 1,675.9 664.5 4 JEG Sse cae Oe ae ce ae ee 1,788.7 1,426.3 362.4 28 CIBEIIS begat’ thas Aaa ee a a og 1,491.5 1,347.4 144.1 8 FEZ CHEN ATION UT Oh i-p 4 Alay anceps in oy eeetie 4 90s 4 DOK le 2,453.0 518.7 vi Victoria, Espirito Santo.......... 1,050.5 779.4 251.1 Ge pe olonia Isabel. o.0..-02 8 cee ce ee 103750 839.8 NO EZ 5 S. Bento das Lages, Bahia ....... 2,179.5 1,562.4 617.1 5) IEG GE ae Ue UR ea Se 2,394.8 1,795.2 599.6 By taking shorter periods—say three months -— this contrast comes out still more strongly. * Bul. Buffalo Soc. Nat. Hist., 1874, p. 227. + Ensaio Estatistico da Provincia do Ceara. Thomaz Pompeo de Souza Brazil. i, 1863, p. 116. t The totals are taken from Dr Draenert’s Vertheilung der Regenmengen in Brasilien. Meteoro- logische Zeitschrift, Sept., 1886. They cover a longer and fuller series of observations than that given by Professor Loomis. Amer. Jour. Sci., vol. xxv, 1883, p. 3. 2 Revista do Observatorio do Rio de Janeiro, vol. vi, 1891, p. 169, XXXIV—But. Gron, Soc, Am., Vou. 7, 1895, ae J. C. BRANNER—DECOMPOSITION OF ROCKS IN BRAZIL. Maximum and Mininum Rainfall by Trimesters. Maximum rain- | Minimum Si . fall for three | fall for three Stations. months Gn} months (n millimeters). millimeters). RAD - We gl AMET O 2b cle oc. «aie weve, 0 ote ee meets a eee 349.3 135.9 Santos aehed Gas eed a Cac e sted eae le en an Re 1,006.2 360.7 Alto. da Senpas 25 (..-.20 iss ae oe eee 1,508.6 587.5 SAD PAO ee hice cscs shes 6 cye.k oe Rese eee Gree 810.7 111.9 AtoVParnalhya “5% sce. ccs Ses ok sear ee meter 563.2 0.0 Ulberalia si5.22% & Seek. daa ESS Oe ee 840.9 68.0 INGOT TOMRV CMO. io eice ciate! s/s wtvarccalers Se Re ee ee 923.0 39.0 GOMSMOMNO OCOD. 2 i 2 elvis a Soka eee eee Pee ets 1,715.5 109.4 GANA hates di li oe 5c shold caveat Sele ote eR 732.9 0.0 OP at AS cee ok falc cySiG eh Bae RTS eS 943.9 53.8 Mais :A Mmazonas. :. <).\. we's/s.c suche ena pee ee 955.4 176.8 aa Pet) oon celia Shawl 6 miei aera ar he eee ets 871.6 128.7 AEA das sc%) ba), ‘ats saint stv alge baie teh a ace ea eee 941.8 41.4 MORN AMDILEO fn. wes os deco kies. eae eae ee 1,682.5 106.7 I PCORTa Ge to. und OR eat se ue a tleaen Roe mation ee 469.2 61.6 Colonie (Isa hel ck che aha. was cele eee se 492.4 64.6 Dente das VaPed tet: site parasite a hiked en 981.5 235.4 ES, iste tiv Boe wicis Move ain eee pea eee 1,156.3 256.7 The little rain that falls during the dry months is not enough to fill the streams, but it all or nearly all soaks into the dry ground at a time when itis highly effective as a chemical agent. The effect of long dry periods upon the soil should not be overlooked. In many places, espe- cially in the clayey lands and in the soils derived from the calcareous rocks of the Cretaceous belt, great cracks open the soil toa depth of from 5 to 10 feet, according to the length of the dry season. ‘These crevices admit atmospheric air and gases readily to a considerable depth, organic matter is constantly falling into them, and when rains come the surface waters penetrate at once to their bottoms and fill the whole upper soil. It is worthy of note that so far as our defective statistics go they show that the rainfall is largest along the east coast of Brazil, where the south- east air currents from the Atlantic first strike the continent. This region also includes the principal gneiss and granite area of Brazil and the re- gion of decomposed schists and shales of the Minas water-shed. Rate OF DECOMPOSITION. No data are at hand for an exact determination of the rate of rock de- composition in Brazil. Some of the oldest gneiss buildings do not exhibit any marked evidences of decay, while others are clearly soft- * Observations for one year only. RATE OF DECOMPOSITION. ole ened by the decay of the feldspar. It is doubtful, however, whether these cases of the decay of building stones can be taken in evidence. The walls and buildings first constructed of gneiss in Brazil were in all probability made of rock from the surface or from near the surface, and possibly already more or less affected. The most marked evidences of decay are along the joints between sep- arate blocks in some of the old buildings. Here there is often a round- ing off of the corners and a crumbling of the mortar. Doorposts and pillars of stone are sometimes affected at the bottom and not at the top, due, no doubt, to the greater amount of organic acids and moisture reaching these lower points. Such cases, however, are not to be taken into consideration in the discussion of rocks under natural conditions. It is certainly true that the agents of rock decay are much more active in Brazil than in cooler climates. Caldcleugh expressed the opinion years ago that in Europe “the agents of destruction are feeble compared with those of a tropical country.”* It is also to be noted that waters falling upon and flowing over these bare rocks are, to begin with, unsaturated and therefore have greater dissolving power than spring or stream water. RESUME. 1. Decomposition is widespread and deep; depths of 100 feet are com- mon; some of more than 300 feet are known. 2. Land-slides caused by deep decay are abundant. 3. Decomposition is not universal, and its absence is especially to be noted in the Cretaceous and Tertiary areas. 4, Talus slopes are rare. 5. Mountains and peaks of gneiss and granite exfoliate like enormous boulders of decomposition, producing a characteristic topography which often resembles glaciated surfaces and roches mountonées. 6. The fragments of nearly all the massive homogeneous rocks tend to exfoliate. 7. Openly exposed blocks of massive crystalline rocks sometimes weather in trenches or in fluted boulders. 8. Changes of temperature cause the openly exposed rocks to crack and to exfoliate. But little decomposition is caused by the direct action of changes of temperature in Brazil, but they open crevices in the rocks which admit moisture and acids—the principal agencies of rock decay. 9. The daily range of temperature sometimes amounts to more than 100 degrees Fahrenheit. * On the geology of Rio de Janeiro. Alexander Caldcleugh. Trans. Geol. Soc., 2d ser., vol. ii, 1829, p. 69. S14 J. C. BRANNER—DECOMPOSITION OF ROCKS IN BRAZIL. 10. An important factor in the decomposition of the rocks is the fact that the rainy season is the hot season, and the waters falling upon hot rocks have their temperatures raised to about 140 degrees. . 11. The dark color of the rocks increases their absorbing and radiating power and consequently the range of temperatures to which they are subject. 12. The unequal expansion and contraction of minerals with changing temperatures hasten the disintegration of rock surfaces. 15. The mechanical effects of changes of temperature, however, are surface phenomena. 14. The coarse textured rocks seem to be more susceptible to decom- posing agencies than the more compact ones. 15. Insects living in the ground, especially the ants and termites, con- tribute large quantities of organic acids to rock decomposition. 16. Plant life is especially rank, and both growth and decay are more rapid than they are in temperate regions. Plants are the chief source of the acids which attack the rocks of Brazil. 17. Carbonic acid is also brought to the earth in large quantities by the rains. 18. Nitric acid is produced and brought down by the rains in much larger quantities than in temperate regions. 19. The rainfall of Brazil is very large, ranging from 974.6 millimeters at Rio de Janeiro to 3,576 millimeters on the Serra do Mar, in the state of Sao Paulo. 20. The concentration of the rainfall renders it more effective both chemically and mechanically. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 7, PP. 315-326 FEBRUARY 25, 1896 RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION ANNUAL ADDRESS BY THE PRESIDENT, N. S. SHALER (Read before the Society December 27, 1895) CONTENTS Page ISU IROCUNGHIOT ae een eee ee AT ESENS ARPES cag BA Ay ls ORR OR RON in CECE 315 Permuonsnip ot teaching and research defined... .. 2.2... 6.22.0. eee etre ee. 316 Interdependence between research and instruction in geology................ BI? Value of geological education and methods of transmission..............---. 318 Coanmrewensive character Of @eolo@y::. ss. oc ss ecu acess eee ewe de dee ree 318 AO MSO feb MEPS CLOIICE 5) 515 0) va lsat aja ins wea Flals ais, Ya tiale hore wie hse die oe 320 Pon O OMB TRUCE Ors e. 4 Palas yore Gis c-e\svacdicst o'a.4 ee gues od tiops alten oak ameter eee 320 PE etMOr seattle Ge AC MMO eta what). o.c Jota sfc at sis olute's oo 5) els oc wae pe 321 Difficulties encountered in field teaching.................... iohatevenatreneal ae 322 Undesirability of teaching geology to immature students................. 324 repent work and its influence and requirements. ..-..... 22.22.6060. .5-21 eee O24 DEL CAUTISTIOID ooh 5:6 ANNE) duets a5 nok ce RIE MAPS A ge Ro oe Pe 325 INTRODUCTION The custom has been established which requires the retiring President of this Society, as other societies which have for their purpose the ad- vancement of science, to set forth his views concerning matters related to the interests which the association seeks to promote. This custom evidently rests on the reasonable presumption that the officer during his term of service has been led by his duties to consider how the cause which he represents may be promoted, how its store of truth may be enlarged, and in what manner it may best be made to serve the interests of mankind. This task may be essayed either by a survey of the work which has recently been accomplished in the science, with appropriate comment on the trends and results of the endeavors, or the essayist may restrict his undertaking to some one portion of the field with which he is conversant in the hope that he may be able to present the fruits of his own labors in a way which 1s likely to be profitable to others. For various reasons I have chosen the latter of these alternatives and have taken for my subject the relations of geological science to education, XXXV—Butt. Geon. Soc. Am., Vou. 7, 1895. (315) O16 N.S.SHALER—RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION. Under this title I shall not only include those questions which pertain to pedagogy, but certain larger aspects of the matter which relate to the needs of society, both from the moral and the economic point of view. RELATIONSHIP OF TEACHING AND RESEARCH DEFINED. I have been in good part led to take up this subject for the reasons that the title itself is a protest against the modern notion that the work of re- search should be separated from that of teaching; that natural inquiry should be released from the ancient and profitable connection with edu- cation which in my opinion has advanced and ennobled. both these branches of learning. Those who seek to have inquiry endowed are led to the endeavor by a true sense of the importance of the tasks with which the path-seekers in the fields of nature have to deal. They are, moreover, ’ guided to their object by the motive which leads to the division of labor in all work which men do, whether in economics or in pure learning. Undoubtedly a certain kind of success would attend the complete sepa- ration of the students of phenomena from those whose business it is to impart knowledge; but there are gains which, though immediate, are not desirable, for the reason that they entail in the long run serious losses. It may well be apprehended that the definite separation of the inquirers in any science from those who are to teach the learning would result on the one hand in isolation of the men of the laboratory from the life of their time and on the other to a degradation of the instruction to a level where it would become mere formal tutoring, destitute of the penetrating spirit which gives value to scientific thought. It seems to me that the explorer, if he be animated by the true spirit of his class, finds himself seeking for undiscovered realms, not for personal gains, nor, indeed, merely to add to the store of things known, but always with reference to the enlargement of mankind. His motive is in the highest sense that of the teacher; he limits his opportunities of personal culture if he denies himself the chance of communicating his gains to the youth of his time. It may be held that the investigator has his means of teaching through the press and the learned societies, but I need not tell my brethren of the craft that the opportunities of sympathetic con- tact with his fellow-men which are thus to be had are very limited; that they are quite insufficient to satisfy the natural desire of an ardent stu- dent of nature for relations with the life about him. ‘The only way in which a really wholesome situation can be found for the naturalist in any of the realms of nature is to link his work with the tasks of education. Viewed from the point of view of the student of science, who ‘has to catch the spirit of inquiry from the word of the master if he is to win it at all, we see that the teaching function of the inquirer is of the utmost RELATIONSHIP OF GEOLOGY TO TEACHING DEFINED. Olt importance to his science. We all recognize and deplore the evils which arise from the fact that young people have to be introduced to most branches of learning by teachers who have little chance to gain or to preserve the spirit of inquiry. We can at most hope that the scientific motive may come to these instructors through a study of the psychology which properly underlies their work. It is unreasonable to suppose that they will be able to bring to their work the stimulating influence of those who are a part of the learning they convey. Therefore if men are to be bred in the ways of the naturalist, the task must be done by investigators. It goes, or should go, without saying that while these men may give and receive profit from their positions as teachers, they should not be called on to do the share of this work which is often inflicted on them, as it is on the teaching body of our schools in general. A condition of this combination of inquiry and instruction is that the two should be associ- ated so as to give the men of science leisure for their studies as well as an opportunity to influence youths by their teachings. INTERDEPENDENCE BETWEEN RESEARCH AND INSTRUCTION IN GEOLOGY. There are good reasons why the connection between research and _ in- struction should be preserved in geology, even if it be abandoned in the case of the other sciences. In those other branches of natural learning the subject-matter can be brought into the laboratory, or at least, as in the case of astronomy, be in some measure made immediately visible to the student, but in geology only a very small part of the fact can be demonstrated by laboratory means. Even where the teacher finds him- self in a field which is rich in illustrations, he is sure to lack examples of the greater part of the important facts which he has to bring to the understanding of his pupils. Under these conditions good teaching de- pends upon the development of the inquiring spirit without the stimulus of a satisfactory direct contact with phenomena. This task cannot be accomplished by any routine methods or by instructors who are not true men of science. It can only be done by those who have the spirit of the investigator in them, who know the range of fact in the intimate and personal way which will enable them to arouse the constructive im- aginations of the youth to the task of picturing the unseen—a task which is at the foundation of the best culture which science has +o elve. A capital instance of what can be done by a teacher who igs also an inquirer is afforded by the work of Louis Agassiz in extending the in- terest in glacial geology in this country. His lectures on the subject were so vivid, they so effectively presented the physiggnomy of the Swiss glaciers, that they quickened the imaginations of the dullest persons. They aroused an interest in the matter which was so intense and on the 318 N.S.SHALER—RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION. whole so well informed that the study of glacial geology in the larger sense of the term developed more rapidly and on better lines in this country, where existing ice fields are lacking, than in European lands, where examples abound. In such work we see the part of the master in instruction. Asa contrast I may be allowed to relate a story which gives us a notion of what science teaching is likely to become when it is left to the people of routine. The professor of mineralogy in Harvard University one day observed two young women examining his mineral cabinet, one of whom was evidently searching for some particular species. Offering his help, he found that the object of her quest was feldspar. When shown the min- eral she seemed very much interested in the specimens, expressing her- self as gratified at having the chance to see and touch them. The pro- fessor asked her why she so desired to see the particular mineral. The answer was that for some years she had been obliged to teach in a neigh- boring high school, among other things, mineralogy and geology, and that the word feldspar occurred so often in the text-book that her curiosity had become aroused as to its appearance. It will, of course, be possible to give the routine teachers some practi- cal knowledge of feldspar and of the other matters of fact with which they have to deal in their text-book work, but the motive, or the lack of it, which is indicated by the incident will always have to be reckoned on as inseparable from the mill-work of ordinary schools. So far as geology is concerned, the instruction of this text-book kind which may be essayed in the secondary schools is quite in vain; its only effect is to make the youths on whom it is inflicted quite unapproachable by the teacher who may afterwards undertake to introduce them to geology. All of us who have taught in colleges know the youth who has had somebody’s “ six weeks of geology ” rubbed in by a drudge who, if required to do so, would in a like way have applied Sanscrit. We know that the youth who has been so misused is in most cases, provided he is not blessed with a good capacity for escaping the influences of education, utterly unfit for our uses. The most economical thing to do, in the large sense of the word, is to give him the advice which the elder Agassiz was wont to give to those of his students who proved impregnable to his methods of instruction: ‘Sir, you better go into business.” VALUE OF GEOLOGICAL EDUCATION AND METHODS OF TRANSMISSION. COMPREHENSIVE CHARACTER OF GEOLOGY. Assuming, as we needs must, that as geologists it is our duty not only to extend the learning of the science, but also to take charge of its dif- fusion among the people, let us consider in general the value of good COMPREHENSIVE CHARACTER OF GEOLOGY. 319 which we have to deliver and the manner in which the transmission may best be effected. So far, doubtless for the reason that geologists are un- commonly busy people, there has been little note taken of the importance of the store of the science to society or the way in which the knowledge Should be handed down. We have been content to harvest and have hardly considered the work of cultivation; therefore the assessment which I am about to give will doubtless need much revision. In the first place, we should note well the fact that geology differs from all other divisions of natural learning in that it is not limited to a particular group of facts or modes of energy, but is in a way concerned with nearly all the work which is done in and on this sphere. We should, perhaps, except human affairs ; but if he is so minded the geolo- gist may make good his claim to a large share in interpreting that group of phenomena also. In fact, the earth lore is not a discrete science at all, but is that way of looking at the operations of energy in the physical, chemical and organic series which introduces the elements of space and time into the considerations and which furthermore endeavors to trace the combination of the various trends of action in the stages of develop- ment of the earth. It is in these peculiarities of geology that we find the basis of its value in education and in the general culture of society, which it is the aim of education to create. It should be in its province, as it is clearly in its power, to give to mankind perspectives which will serve vastly to enlarge the evident field of human action. All observant teachers know that no true success in education is possi- ble until we contrive an awakening of the youth from the sleepy accept- ance of the world about him. To rid the student of this benumbing relic of the bone-cave, the spirit of the commonplace, there is no treat- ment so effective as that which it is in the power of the master in geology to give. The story of the ages clearly told, with a constant reference of the bearing of the matter on the appearance and the fate of man, will quicken any mind that is at all fitted to profit by the higher education. Although geology can hardly be said as yet to have made any such gen- eral impression on laymen as is justified by the body of truth which it has to deliver, the close observer may notice certain important changes in the state of the public mind which seem clearly to have been due to the teachings of the science. While many things go into the making of the world’s judgments, there can be no question that the plain truths concerning the antiquity of the earth and the series of events which have led to the coming of mankind have in this generation been most effective in overturning sectarian bigotry and in other ways enlarging the spirit of all educated people. It is evident that the main contribution which geology has to make to 320 N.S.SHALER—RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION. those conceptions which may enter into the spirit of our society relates to the position of man; the abstract learning, that which is in and for itself, is for those who have the professional interest. These public values of the science are of two diverse kinds—on the one hand those which pertain to intellectual enlargement; on the other, to economic development. Therefore in considering our duty by the educational side of our work we should see what the contributions can be to these two modes of endeavor and how they should be presented. First, I shall consider the limitations of that work which may be regarded as distinctly pedagogic. DIVISIONS OF THE SCIENCE. It seems to me necessary distinctly to separate the body of the instruc- tion which is to be given in geology into two parts—that which is appro- priate to the general public and that which, though “caviare to the general,” fits the appetite of the professional-minded. We are indebted to the philosophical pedagogue Herbert for a statement of the self-evident proposition that interest in a matter must exist before information con- cerning it can be profitably communicated; therefore in our teaching we must take no end of care to provide this foundation for the attention. This care is particularly necessary in the matters of geology, for, as be- fore remarked, the facts cannot often be exhibited in the experimental way as in the laboratories of chemistry and physics, where the touch of hand or the sight of controlled actions establishes a personal relation with the problems. The teacher of our science has to avail himself of certain antecedent motives which he can presume to exist in any normal youth which may provide the required foundation of interest. What I have to say on this point is the result of nearly a third of a century of experience in teaching geology, and is based on work which has been done with more than 4,000 students. The basis for the induction is sufficiently great to make the conclusions of value. These are in brief as follows: That instruction in geology, which is meant for those who have not acquired the professional motive, must find its basis of interest on either of two foundations—on the element of sympathy with all which relates to the fate of man which is native in all of us, or on the love of the open fields which every youth who is not utterly supercivilized has as a birthright. Each of those interests is in a way primal, both may be separately reckoned on as strong in nearly all youths who are fitted for the higher education. CLASS-ROOM INSTRUCTION. To make use of the motives which may interest the beginner in geol- ogy my experience has shown that the first thing to do is to give by means CLASS-ROOM INSTRUCTION. lt of familiar lectures a general acquaintance with those series of actions which show the long continuous operations of energy in the orderly march of events, taking pains at each convenient opportunity—there are many such—to note how these processes have served to bring about the conditions on which the development of peoples or of states depends. Thus, in treating of volcanoes, the very humanized story of Vesuvius or of AXtna, especially the dramatic episode of the death of Pliny the Elder, is worth much to the teachers for the reason that it serves to bring a sense of human affairs into a subject which for lack of illustration is apt to remain remote and therefore uninteresting. The fact that the story of these volcanoes, especially that of Vesuvius, is inwoven with that of men forms a bond between the mind of the novice and an order of nature which would otherwise be utterly unrelated to him. Again, in treating of seashore phenomena, the history of harbors and their relation to the development of states, affords a basis on which to rest the account of coastline work. Yet again, in the matters connected with the forma- tion of mineral deposits, which from the nature of the subject are apt to be somewhat elusive, it is easy to fix the attention by reference to the relation of those stores to the needs of man. So, indeed, in all parts of this preliminary work of awakening and developing interest in his sub- ject the teacher of geology, if he is to be successful, must go about his task on the supposition that he has to extend existing interests to his field. When men have for some hundred generations appreciated the earth as we would have them do it, the process of selection or the inher- itance of acquired characteristics may give a birthright interest in the large problems of geology; but while here and there a youth may be found with a Hugh Miller’s taste for the science, the teacher who reckons on having his class thus inspired will fail to achieve success. METHODS OF FIELD TEACHING. As soon as the teacher through his work in the lecture-room has suc- ceeded in extending the natural inborn interests of his pupils to the problems of geology, instruction in the field should begin. In this part of the work there is need of a great change in the methods and aims of the teaching. While in the lecture-room the conditions require the didactic method and exclude that of investigation, the reverse is the case in the field. When I first essayed peripatetic teaching I made the grave mistake in endeavoring to lecture with the phenomenon asa text. In time I found that the fatigue and other disturbing conditions of the open made students unable to profit by any such didactic method, and that all such direct instruction should be done while they were in the more re- ceptive conditions of the house. The true use of the field is to awaken 322 N.S. SHALER—RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION. in the pupils the habit of seeking for themselves. The teacher may trust in this task to the existence of an observant motive in men which is at its best when they are in the open air. All of us, however dull we may be in the housed state, have when afield a discerning humor which prompts us to learn the reasons for the unexplained occurrences of nature. This precious relic of the savage life, of the original motive of curiosity, which has been the source of man’s advance on the most of his intellectual up- goings, is in average youths strong; it requires the deadening effects of a long and misspent life to eradicate it in any normal human being. It is to this element of curiosity, informed by the preliminary instruction of the lecture-room, that the teacher of field geology should mainly trust for his success. In practice it will be found impossible completely to exclude didactic teaching in the field—such arbitrary divisions of methods are generally impracticable—but when in face of an exhibition of any geological phe- nomena, with the briefest possible preliminary, designed to fix the atten- tion of the class upon the facts, the teacher should at once become a mere questioner, a goad to arouse the men to a like interrogation of the things they see. Itis important that the first problems of interpretation which are essayed should be of the simplest order, for immediately suc- cessful work in the unaccustomed harness is much to be desired. Thus the determination of strikes and dips, the identification of visible faults, and aboye all, the careful recording of such facts, should come first and the work be carried to distinct success before any effort is made to use the results in the larger interpretations as to the attitudes of strata. In my experience it is most desirable in the early part of the field train- ing to give all that can be obtained in the way of work which relates to causes of action, and thus, for the reason that men, however great their training may otherwise be, are unlikely to conceive the earth about them as a realm of continuous processes, their geology is thus not brought down to the present period. The beds and banks of the streams, the retreating escarpments, the shores of lakes and of the ocean—above all the, when rightly discerned, majestic phenomena of the soil—all may serve to impress the pupil with the activity of the earth, and thus clear his mind of the natural but blinding conception that after its creation time the sphere entered on an enduring rest. DIFFICULTIES ENCOUNTERED IN FIELD TEACHING. In my experience the difficulties which have to be met in field teach- ing, apart from the hard labor involved in the simultaneous exercise of mind and body, consists in the struggle which the instructor has to make with the incapacities which arise from the supercivilization of his pupils. DIFFICULTIES IN FIELD TEACHING. > ode These hindrances are protean in form, but they are most commonly to be found in an inability to think in three dimensions any better than we can in four, and an incapacity to continue any work when alone. As to the first of these defects there seems to be no resource except to revive the natural dimensional sense which primitive peoples have. If the student has had sound training in solid geometry he may the more quickly recover the capacity to form the special conceptions which are required of the geologist; but the natural solid is quite another thing from the ideal, and while the theoretical view of them is the same the practical experience is very different. Some youths never learn to deal with the earth problems from the solid point of view. They are there- fore cut off from the better uses of the field; yet even with this signal disadvantage they may do good work in certain parts of the science. One of the most distinguished of our American geologists, now dead, was, perhaps on account of the fact that he saw from but one eye, quite with- out the sense of the relations of the solid; yet, while in the field-work his success as measured by his talent was limited, his contributions in other departments were great and of enduring value. Nevertheless, though the people who abide in two dimensional spaces may possess abilities of a high order, they should be kept out of the science which more than any other calls for the ability to frame three dimensional conceptions. An inability to work alone in the field is a rather common and in my experience an incurable defect in certain students who would otherwise be fitted for geology. Those who are thus aftlicted appear to lose their motive of inquiry when they are parted from their fellow-men. Their malady is to be regarded as one of the many defects of body and mind which are due to over-housing—to that absolute separation from the peace of the wilderness which characterizes our city life. As soon as possible the field student should be brought to the point where he is required to make his own maps, at first as sketches, and then in the more formal way by pacing, with some methodical control, such as by a simple triangulation. One piece of such map-work where the delineation of the surface in general ground plan and contour, as well as the geological coloring, is from his own labor will often be sufficient to affirm the working power of the man. In the ideal of the system such instruction should come to every student who undertakes the study of geology, but in practice it will probably be gained by very few. In the department of Harvard University which is devoted to the science about _ 800 men each year enter on the elementary work. Of these not more than the eighth part continues the study to the point where they may begin to do work which may be regarded as independent ; yet fewer essay the training which looks forward to a professional career. As this de- XX XVI—Butt. Grou. Soc. Am., Vou. 7, 1895. 324 N.S.SHALRR—RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION. partment has been long established and is favorably conditioned to give instruction, the lack of a large attendance under a system of free election by students may be taken as an indication that while the elementary didactic presentation of the science attracts the greater number of the youths of our colleges, the higher branches are less attractive than the other similarly difficult work of the indoor learning. The conclusion is that geology in the larger sense of the term is, at least in the present con- dition of culture, an interest for a few chosen spirits who are so fortunate as to be born with a share of the world sense, or at least with an aptitude for studies which demands a measure of the primitive man which is not to be found in the most of our supercivilized folk. UNDESIRABILITY OF TEACHING GEOLOGY TO IMMATURE STUDENTS. In the demand which is now made for.a beginning of all our sciences in the secondary schools it is proposed to include geology in the list and to set boys and girls of from fourteen to seventeen years of age at work upon the elementary work of the learning. For my own part, while it seems to me that some general notions concerning the history of the earth may very well be given to children, and this as information, it is futile to essay any study in this science which is intended to make avail of its larger educative influences with immature youths. The educative value of geology depends upon an ability to deal with the large concep- tions of space, time and the series of developments of energy which can only be compassed by mature minds. Immature youths, even if they intend to win the utmost profit from geology, would be better occupied in studying the elementary tangible facts of those sciences such as chem- istry, physics or biology, sciences which in their synthesis constitute geology, rather than in a vain endeavor to deal in an immediate way with a learning which in a good measure to be profitable has to be approached with a well developed mind. The very fact that any considerable geo- logical problem is likely to involve in its discussion some knowledge of physics, chemistry, zoology and botany is sufficient reason for postponing the study until the pupil is nearly adult. Expert WoRrRK AND ITs INFLUENCE AND REQUIREMENTS. Besides the relations to society which may be established by his posi- tion as a teacher, the geologist is from the character of his studies much called on for another kind of help, that which pertains to the de- velopment of earth resources or to the litigation which concerns earth values, In this field the relations are more critical and more perplexing EXPERT WORK. B25 than in that of instruction. The results of blundering are more appar- ent and their immediate effect on the reputation of the science more unhappy. That this branch of learning has managed to retain a fair place in the esteem of the public in face of the criminal blunders which its prophets have made is indeed remarkable. It shows how much our people are disposed to pardon where they believe that men mean well, however ill may they do. There is, however, a lesson from this unhappy experience which we should all read and inwardly digest. This is in effect that what is called expert work demands other qualities of mind and another training than those which go to make a successful investi- gator or teacher. We, as well as the general public, need to recognize that fact, that there is as much reason to suppose that a noted teacher of political economy should prove successful in determining the merits of a proposed business project as that his colleague in geology should be fit to advise in regard to a mining venture. The teacher may be an ex- pert in the economics of the profession, but the proof of the fact is not to be found in his scientific work or in his success as an instructor. If he has not had the other training, it may be safely assumed that he will be totally unfitted to wrestle with the tricky fellows who try in amaz- ingly varied ways to deceive him, or even with the tendencies of his own mind, which naturally lead him to see riches where others fancy they discern them. In the interests of our science it is most desirable that all expert work should pass into the hands of a body of men who should bring to their task so much of geology as is needed for the particular inquiry, com- monly not very much, and who can join with it the more important practical acquaintance with the miner’s art and the conditions of trade which relate thereto. In certain cases the men of theory may well serve these experts; all their-inquiries are likely to be of service in the deter- minations, but on them should not be the responsibility for the business side of the problems. There is little the geologist does in the way of research which may not have some practical application to the affairs of men, but he should not mistake this possibility of usefulness as an indi- cation that it is for him to give his inquiries an economic turn. CONCLUSION. We thus see that geological science, like the most of the other branches of natural learning, has two distinct points of contact with society—that of instruction and that of economic affairs. In each of these fields of usefulness its services to man have been great and are to be far greater in the time to come. As for instruction, the task is to give to men an 326° N.S.SHALER—RELATIONS OF GEOLOGIC SCIENCE TO EDUCATION. adequate perspective for their lives. It is to ennoble our existence by showing how it rests upon the order of the ages. In the economic field it is to show the resources which these ages have accumulated in the earth for the service of the enlarged man, who is to attain his possibili- ties by a full understanding of his place in nature. To do the fit work we need to combine the functions of explorers and guides zealous to open the way to the unknown, and those of teachers who take care that the youth of our time are led into the land which we know to have so much promise for man. 7 ‘ a HOALBSOY garaoy 81H sromresey qunouriegc | ALISYBAINJ) 8 Se 3nu3sayj NUaysaM oy: J vas . . ?. : (ae *e, *) € ae hollows and relative eminences’’ (page 564). In discussing breaches in the escarpments and hill ranges of the Wealden district, the same author says: ‘The only explanation of these facts is . . . marine erosion first produced a surface of planation across the whole district while it was being slowly elevated, so that this original surface sloped gently from a central line toward the north and south. The primary streams naturally followed these slopes, . . . forming the transverse valleys’’ (page 581). Richthofen is the leading advocate of marine erosion among conti- nental geologists. He treated the origin of plains of denudation, inde- * Physical Geology, 1882, p. 577. + Second edition, 1887. { Handbook of Physical Geology, 1892, p. 620. 384 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. pendently of Ramsay’s writings, in his great work on China, attention being led to the problem by the occurrence of unconformable marine strata lying on smooth foundations, as observed in his eastern travels. He concludes that the “ oldland” platform cannot have been produced by atmospheric wasting or by running water; these agencies produce valleys separated by ridges. Truly the valleys multiply and widen and the ridges weaken, but reduction to a lowland can be reached only locally and in small dimensions. Moreover, change in the altitude of land works against complete denudation ; yet, although such a result is unattain- able by subaerial agencies, it may be accomplished by the waves of the sea beating onthe coast. Three cases are considered: astill-stand of the land for an indefinite period, a slow elevation and a slow depression. The still-standing land would be cut inward to a limited distance, after which the waves would be exhausted on the platform of their own cary- ing. During elevation slight effect could result, for the work would always be beginning anew. Slow depression alone can produce regional abrasion, for then the power of the waves is maintained by the continued sinking of the bottom, while detritus accumulates on it. In contrast to structural plateaus (Schichtungsplateaus), plateaus of denudation have no relation to the structures across which they are cut or to the valleys which are sunk beneath their level after general elevation. As examples, the Ardennes and the uplands of the middle Rhine are first mentioned, these being explained as producible only by sea waves; never by flowing water or other subaerial agents. Another example given is the western slope of the Sierra Nevada of California, now uplifted and dissected.* The substance of the above is repeated in Richthofen’s ‘‘ Fiihrer fur Forschungsreisende,”’ | emphasis being given to the association of plains of denudation with unconformably overlying sediments, to which the English school directs insufficient attention. Subaerial agents are de- scribed as excavating valleys in uplifted plains of denudation, but not in producing the plains (pages 171-173, 670, 671). The prevalence of superposed streams in certain dissected uplands of abrasion is noted (pages 671, 672), but no contrast drawn between these examples and others in which the streams are systematically adjusted to the structures. Cornet and Briart have made special study of the greatly deformed Paleozoic rocks of Belgium, which they believe once rose in lofty moun- tains. Although they regard subaerial agencies competent to produce the “complete ablation” of a land surface, they conclude that it was probably the waves of an encroaching sea that contributed largely to destroy what remained of their ancient mountains in Cretaceous time. * China, 1882, vol. ii, chap. xiv, sec. 3. + Berlin, 1886, pp. 353-361. tLe relief du sol en Belgique. Ann. Soc. Geol., Belg., iv, 1877, pp. 72-113. VIEWS OF AMERICAN WRITERS. 885 Philippson follows Richthofen in treating plains of denudation— ‘obrasionsflachen ”—as the result of wave action.* Tar AMERICAN SCHOOL. Few American writers accept the belief of the English school. The first clear recognition of the importance of subaerial baseleveling should, I believe, be credited to our geologists in the western surveys.t Powell’s “ Exploration of the Colorado river” (1875) brought the American view of the capabilities of subaerial erosion more prominently forward, yet the text does not furnish brief explicit statement directly to the effect that lowlands of denudation may be produced by subaerial agencies. Extracts would lose their flavor apart from their context, but in figuring a section of the wall in the Grand canyon the beveled sur- face of the tilted older strata on which the horizontal Carboniferous strata he is drawn smooth and even. The overlying beds “are records of the invasion of the sea; the line of separation the record of a long time when the region was dry land” (page 212). Here the implication is that the sea gained entrance by depression of the baseleveledland. The overlying strata are regarded as the ruins of some unrepresented land, not of the locally buried land. The explanation is precisely opposite to that given to similar structures by Richthofen. In Powell’s ‘ Geology of the Uinta mountains ” (1876) there is a similar absence of explicit account of baseleveled plains, apparently because it was not necessary to expand truisms so simple; but the chapter on degradation very clearly implies the capacity of subaerial forces to wear down mountains, however high ; indeed, its burden is to show that the destruction of a lofty range is so much accelerated by steep declivity that its life cannot be much longer than that of a low range. Mountains are “ephemeral topographic forms ;”’ all existing mountains are geologically recent (page 197). All this without once calling on the aid of sea waves. Dutton’s monograph on the “ Tertiary History of the Grand Canyon district”? (1882) is most characteristically American in treatment as in theme. Referring to the great unconformity near the base of the canyon walls in the Kaibab and Sheavwits plateaus, he says, on page 207, that— “The horizontal Carboniferous beds appear to have been laid down upon the surface of a country which had been enormously eroded and afterward submerged.” * Studien iber Wasserscheiden, 1886, p. 100. +Marvine briefly presented the essence of the idea in 1873, but he made mention of marine action in a late stage of the process, somewhat after the fashion of the English school. Deserib- ing the east slope of the Rocky mountain front range, he wrote: “ The ancient erosion gradually wore down the mass of Archean rocks to the surface of the sea, . . . the mass was finally leveled off irrespective of structure or relative hardnesses of its beds, by the encroaching ocean, which worked over its ruins and laid them down upon the smoothed surface in the form of the Triassic and other beds” (Hayden’s Survey, Rept. for 1873, p. 144). XLV—Boun. Gron. Soc. Am., Von. 7, 1895. 386 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. The erosion followed uplift, the deposition followed submergence when the erosion was essentially completed. Along the surface of contact there are— ‘*A few bosses of Silurian strata rising higher than the hard quartzitic sandstone which forms the base of the Carboniferous. These are Paleozoic hills, which were buried by the growing mass of sediment. But they are of insignificant mass, rarely exceeding two or three hundred feet in height, and do not appear to have ruffled the parallelism of the sandstones and limestones of the massive Red Wall group above them ” (page 209). On another page (181) Dutton says: ‘“The meaning of this great unconformity obviously is that after a vast body of early Paleozoic strata had been laid down they were distorted by differential ver- tical movements, were flexed and faulted,and were elevated above the sea. They were then enormously eroded. . . Still later the region was again submerged.” Over the rugged country thus ravaged, the later strata, perhaps 15,000 feet thick, were laid down. Many other examples of the American view may be given. Most of them, as in the cases already cited, take no account of the possibility that the evenly abraded surface of the older terrane might be essentially the product of wave work, but tacitly assume that it resulted from subaerial erosion, followed by depression, with more or less tilting, so that the sub- merged area comes to be sheeted over with waste derived from some non- submerged area. Irving concludes that in Wisconsin— ‘An amount of material vast beyond computation was removed from this ancient land before the encroachment upon it of the sea within which the [Potsdam] sand- stone was deposited.” * The buried oldland is referred to as a “sub-Potsdam land surface.” + Van Hise, writing of the great unconformities below and above the Penokee series of Wisconsin and upper Michigan, implhes great subaerial erosion, by which an uplifted region was reduced to a peneplain ; depres- sion, submergence and deposition of material eroded elsewhere then fol- lowed. The essentials of the explanation are that the Penokee series rests upon an ancient land surface, more or less modified by wave action at the time of submergence, but worn down from its constructional form almost entirely by subaerial agents. Walcott, recognizing wave work at the margin of an encroaching sea as contributing to the formation of basal conglomerates, nevertheless ex- plains the great pre-Cambrian land area of our country as “ approaching *Seventh Ann. Rep. U.S. Geol. Survey, 1888, p. 402. + Ibid., p. 409. { U.S. Geol. Survey, monograph xix, 1892, pp. 454-466. VIEWS OF AMERICAN WRITERS. 387 the baselevel of erosion over large portions of its surface.”* Moreover, it was a result of continental depression and not of erosive encroachment of the waves that the upper Cambrian sea gained its extension over the great interior of the continent (page 565). The relation of subacrial and marine agencies are here, as in so many instances, just reversed from their proportionate activities in Richthofen’s scheme. McGee was the first to present a clear statement of the vast subaerial denudation of our Atlantic slope in Mesozoic time: | ““ Before the initiation of Potomac deposition, but subsequent to the accumula- tion of the Triassic and Rhaetic deposits and to the displacement and diking by which they are affected, there was an eon of degradation during which a grand mountain system was obliterated and its base reduced to a plain which, as its topog- raphy tells us, was slightly inclined seaward and little elevated above tide. y There followed a slight elevation of the land, when the rivers attacked their beds and excavated valleys as deep as those today intersecting the Piedmont plain. Then came the movement by which the deposition of the Potomac forma- tion was initiated; the deeply ravined baselevel plain was at the same time sub- merged and tilted oceanward.” Tf It appears from the foregoing examples that, in denuded plains over which unconformable sediments have been deposited, some late and small share in the work of denudation may be allowed to the shore waves as they advance over an already prepared peneplain when depression occurs ; but it is otherwise with those uplifted and dissected plains of denudation upon which there is no reason to think that unconformable sediments: have ever been deposited. The plateau in which the Grand canyon of the Colorado is cut is an extraordinary example of this kind. It is, moreover, notable from consisting of nearly horizontal strata, where acute observation has been needed to detect evidence of the long cycle of ero- sion passed through before the region was uplifted to its present altitude. The great plateau is beveled obliquely across the Carboniferous and Permian strata, so that the undulating surface of the upland in its medial part presents Permian beds on the hills and Carboniferous beds in the hollows; but to the south, where the strata gently rise, the whole surface is Carboniferous; to the north, where the strata sink, the surface is en- tirely Permian. ‘We may suppose that this entire region, at the epoch at which the great denuda- - tion of the Mesozoic system approached completion, occupied a level not much above the sea. Under such circumstances it would have been at what Powell terms baselevel of erosion. The rivers and tributaries would no longer corrade their channels. The inequalities which are due to land sculpture and the general process of erosion would then no longer increase, and the total energy of erosion would be * Twelfth Ann. Rep. U.S. Geol. Survey, 1591, p. 562. + Three formations of the middle Atlantic slope. Am. Jour. Sci., vol. xxxv, 1888, p. 142. 388 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. occupied in reducing such inequalities as had been previously generated. During periods of upheaval, and for a considerable time thereafter, the streams are cutting down their channels, and weathering widens them into broad valleys with ridges between. The diversification so produced reaches a maximum when the streams have nearly reached their baselevels; but when the streams can no longer corrade, » and if the uplifting ceases, these diversifications are reduced and finally obliterated. Such, I conceive, was the case here. . . . The entire region was planed down to a comparatively smooth surface.’’ * Willis first called attention to the occurrence of an uplifted and dis- sected peneplain of subaerial denudation in the mountains of North Carolina,t and Hayes and Campbell have since then shown the’ great extent and area of this ancient land surface. { Willis and Hayes have lately described the northern and southern Appalachians, § giving much attention to the essential extinction of the mountains, except in the Caro- lina highlands, in late Cretaceous time. The first author writes of the lowland thus produced: “The land was flat, featureless and very slightly elevated above the sea” (page 189). The second author writes: “The whole region was reduced to a nearly featureless plain, relieved only by a few groups of monadnocks where the highest mountains now stand” (page 330). Emerson writes of the Berkshire hills in western Massachusetts : “ Erosion planed away the mountains to the general level, which can still be seen in the average level of the plateau, pitching slightly east. * * * When this peneplain was formed it was doubtless horizontal and near the sealevel, and was what is called a baselevel.”’ || Salisbury says that the even crest-lines of the New Jersey highlands tell of ‘mountainous elevations reduced to a peneplain near the level of the sea.” 4] Not only the tilted rocks of the Allezhenies and of the older Appa- lachian belt, but the horizontal strata of the Allegheny plateau are regarded as having been baseleveled, or almost so, before their present uplift and dissection was gained. See, for example, the account of the Cumberland plateau in Tennessee by Hayes.** Griswold has recognized a greatly dissected peneplain in the even crested ridges of the Arkansas novaculites, and has associated the warp- ing of the great peneplain of which his special district was a part with the origin of the lower course of the Mississippi in late Mesozoic time.ft *Grand Canyon District. U.S. Geol. Survey Monogr., IJ, 1882, p. 119. + Round about Asheville, Nat. Geog. Magazine, vol. i, 1889, p. 297. t Geomorphology of the southern Appalachians, ibid., vol. vi, 1894, p. 69. 2 Nat. Geog. Monographs, vol. i, 1895, nos. 6 and 10, } Hawley sheet, Geol. Atlas U.S., 1894. 4 Geol. Survey New Jersey, 1894 (1895), p. 8. ** Sewanee sheet, Geol. Atlas U.S., 1895. ++ Geol. Surv. Arkansas, 1890, vol. iii, p. 222; Proc. Bost. Soc, Nat, Hist., vol. xxvi, 1895, p. 478. VIEWS OF AMERICAN WRITERS. 389 Keyes * and Hershey ¢ have recently described the upland of the Ozark plateau in Missouri as an uplifted and dissected peneplain. The region has an essentially horizontal structure, like the Allegheny plateau, with which it isin many ways homologous. ‘The latter author tells of residual hills or monadnocks which still surmount the upland plain, and of faint inequalities of form that seem to mark “ the hydrographic basins of the streams which flowed on the Cretaceous lowland plain;” but as a whole the region was ‘“‘a low, marshy plain of very slight relief, probably nearly at sealevel.” ; Darton describes the Piedmont area of Virginia as— *“An undulating plateau carved in greater part in crystalline rocks . . traversed by rivers which flow in gorges. . . It is now very clearly recognized that the Piedmont plateau is a peneplain of Tertiary age. . . There is a system of very low, flat divides coincident with those of the present drainage system.’’ t Keith also describes the formerly even surface of the Piedmont belt in which the valleys of today are incised, as a Tertiary baselevel of subae- rial origin. § The bevelled western slope of the Sierra Nevada, regarded as an up- turned plain of marine abrasion by Richthofen, is ascribed by Gilbert, || Leconte,4] Lindgren, Diller}}+ and others to subaérial denudation , but Lindgren makes it clear that when the region stood lower it was not worn smooth enough to be called a peneplain; “the declivities and irregularities of the old surface are too consid erable for that.” Diller describes a peneplain formed on the upturned Cretaceous rocks of northern California and now dissected by various streams: ““The production of such a broad, uniform plain by the erosion of rocks varying greatly in hardness could only be accomplished on a very gentle slope near the level of the controlling water body, and we may therefore properly consider this plain a baselevel of erosion.’’ tt Lawson presents an instructive account of an uplifted and dissected peneplain beveled across upturned strata in northern California. Water- worn gravels occur on the ridges of the dissected upland. They “ can only be interpreted as remnants of the stream gravels of the ancient peneplain.” $§ , * Geol. Surv. Missouri, vol. viii, 1894, pp. 330, 352. + American Geologist, vol. xvi, 1895, p. 338. t Chicago Jour. Geology, 1894, vol. ii, pp. 568-570. 2 Fourteenth Ann. Rept. U.S. Geol. Survey, 1894, p. 369. || Science, vol. i, 1883, p. 195. q Bull. Geol. Soe. Am., vol. 2, 1891, p. 327. ** Tbid., vol. 4, 1893, p. 298. ++ Chieago Jour. Geol., vol. ii, 1894, p. 34. tt Fourteenth Ann. Rept. U.S. Geol. Survey, 1894, p. 405. 22 University of California; Bull. Dept. Geol., vol. i, 1894, p. 244. 390 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. G. M. Dawson describes an ancient peneplain, now an elevated and dissected plateau, in the Rocky Mountain rezion of Canada: ‘*Climbing to the level of this old plateau, or to that of some slightly more ele- vated point about the fiftieth or fifty-first parallel of latitude, the deep valleys of modern rivers with other low tracts are lost sight of, and the eye appears to range across an unbroken or but slightly diversified plain, which, on a clear day, may be observed to be bounded to the northeast, southwest and south by mountain ranges with rugged forms, and above which in a few places isolated higher points rise, either as outstanding monuments of the denudation by which the plateau was pro- duced, or as accumulations due to volcanic action of the Miocene or middle Tertiary period.’’ * After explicitly considering the alternatives of marine and subaerial erosion, the author decides against the former, because the plateau dis- trict is not accessible to the sea, and because there are no marine strata thereabouts referable to the period when the peneplain was formed. The river system of the region— ‘‘aided by other subaerial agencies, cut down almost its entire drainage basin till this became a nearly uniform plain, with some slight slope in the main direc- tion of the river’s flow, but of which the lowest part approximately coincided with the sea-level of the time. . . . After reaching this baselevel of erosion the rivers would, of course, be unable to do more than serve as channels for the con- veyance of material brought into them from the surrounding country, which, wherever it stood above the general level, was still subject to waste. The val- leys became wide and shallow, and the surface as a whole assumed permanent characters.’’ T My ownstudies lead me to believe that subaerial denudation has reduced Various mountainous or plateau-like uplifts to lowland peneplains. A considerable number of extracts might be presented from the works of foreign writers to show that the idea of marine denudation is on the whole less favorably received by continental than by English geologists ; but the features of land form and the processes of land sculpture have not been studied in Europe with the attention that has been given to stratigraphic succession or to the problems of paleontology and petrog- raphy. Regions that are known to be uplands of denudation are often described with abundant detail as to their structure, but with the scantiest reference to the conditions of their topographic development. * . . . The Rocky Mountain regionin Canada . . . , Trans. Roy. Soc. Can., viii, 1890, p. 11, + Loe. cit., p. 13. t The following articles may be referred to: Relation of the coal of Montana to the older rocks (Tenth Census U. S, vol. xv, 1886, p. 710); Topographic development . . . of the Connecticut valley (Am. Jour. Sci., vol. xx xvii, 1889, p. 480) ; Geographic devetopment of northern New Jersey (with J. W. Wood, Proc. Boston Soe. Nat. Hist., vol. xxiv, 1889, p. 373); Rivers of northern New Jer- sey (Nat. Geog. Mag., vol. ii, 189), p. 6); Topographic forms of the Atlantic slope (Bull. Geol. Soc. Am., vol. 2, 1891, p. 557); Physical geography of southern New England (Nat. Geog. Monogr., vol, i, 1895, p. 276); Development of certain English rivers (Loudon Geog. Jour., vol. v, 185, p. 140). VIEWS OF AMERICAN WRITERS. 391 A characteristic example of this.manner of treatment is found in the valuable works by Lepsius on the mountains of the upper and middle Rhine,* in which the Schiefergebirge and other ancient mountains are fully treated as to structure, although little is said of their form and still less of the origin of their form. The following citations are from works in which land form and sculpture are more fully considered. The increasing importance attributed by Sir A. Geikie to subaerial agencies in his later writings has already been noted. Professor James Geikie goes further in this direction and says: ‘* Valleys continue to be deepened and widened, while the intervening mountains, eaten into by the rivers and their countless feeders and shattered and pulverized by springs and frosts, are gradually narrowed, interrupted and reduced until eventually what was formerly a great mountain chain becomes converted into a low-lying undulating plain.’’ fT Gosselet, in his comprehensive monograph on the Ardennes, says that the tilted, folded and faulted strata of their uplands haye been, as it were, planed down by the combined action of atmospheric disintegration and pluvial wearing. Both the Jurassic and Cretaceous formations are de- scribed as lying on oldland soils, where they overlap the Paleozoic strata. The elaborate treatise on “ Les formes du terrain” (1888), by de la Noé and de Margerie, clearly maintains that pluvial denudation may not only produce valleys, but it may wear down the divides between the val- leys (page 106). The escarpments or cross-valleys of the Weald in south- ern England may be explained without calling on marine erosion, as most of the English geologists have done (pages 135, 136). Plateaus of abra- sion, without a cover of unconformable strata, may be “ simply the result of prolonged subaerial erosion.” If unconformably covered, it still re- mains to be seen how far the abraded surface is— ‘““The modification by wave action of a hardly different surface, produced by the prolonged work of streams which had long before attained faintly graded slopes, and which had by the aid of atmospheric agents almost completely destroyed pre- existing inequalities of form ’’ (page 188). Penck concludes that the final aim of subaerial denuding agents is to reduce a land almost completely to a plain,§ but his account of the Schiefergebirge of the middle Rhine does not explicitly state whether the “abrasionsplateau ” of their uplands is of marine or subaerial origin. || * Die Oberrheinische Tiefebene und ihre Randgebirge, Forschungen zur deut. Landeskunde, i, 1885, 35-91; Geologievon Deutschland, 1887. +Mountains, their origin, growth and decay: Scot. Geog. Mag., vol. ii, 1886, p. 160. { L’Ardenne. Mém. Carte géol., France, 1888, pp. 802, 808, 837. @ Das Endziel der Erosion und Denudation, Verh. viii deut. Geographentag, 1889, pp. 91-100. || Landerkunde des Erdtheils Europa, i, 1887, p. 316. 392 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. In his compendious volumes on the ‘‘ Morphologie der Erdoberfliiche ”’ (1894), he considers plains of marine and of subaerial denudation, both as to process of origin and as to derivative forms, after elevation and dis- section, but criteria for their discrimination are not discussed.* De Lapparent, president of the French Geographical Society, has advocated subaerial erosion as the means of denuding the Ardennes and the Central plateau of France, f and later says: “ La notion des pénéplaines est extremement féconde, et ce n’est pas un de ses moindres mérites d’avoir porté le coup de grace 4 la théorie des plaines de dénuda- tion marine, si fort en honneur de l’autre coté du détroit.” t CoMPARISON OF THE TWO SCHOOLS. It is noteworthy that, with few exceptions, the more recent writers here quoted do not discuss both processes by which smoothly abraded plains, whether buried or bare, may be produced, but directly announce their conclusion as to the origin—by marine or by subaerial agencies—of the surface under consideration. This, of course, implhes that they regard the question as settled, just as for some time back it has been the habit of geologists on finding marine shells in stratified rocks to conelude, with- out reviving the discussions of earlier centuries, that the strata are of marine origin, aud that their present position indicates a change in the relative attitude of the land and sea. But in this latter example all geologists are today agreed, while in the problem of the origin of plains of denudation each writer follows only the conclusion of his own school, not the conviction of the world. Jt is chiefly to arouse attention to this aspect of the problem that the present review is undertaken. It is further noteworthy that, with few exceptions, the authors who discuss the matter at all do not attempt to discriminate between the two possible classes of denuded surfaces by searching for features peculiar to one or the other, but content themselves with @ priori argument as to the possibility of producing plains by marine or subaerial agencies. There is, however, a certain difference of attitude in the two schools regarding the doctrine of the other. The English school hardly considers at all the ability of subaerial agencies to produce smooth plains of denu- dation; their discussion of the question turned really on the possible origin of valleys by subaerial agencies. ‘The American school does not, as far as I have read, deny the ability of marine agencies, but attributes ereater ability, especially far in continental interiors, to subaerial agencies ; their discussion of the question postulates the subaerial origin of ordinary * Vol. ii, pp. 145, 181, 489. + L’age des formes topographiques, Rev. des quest. scient., Oct., 1894. {La géomorphogénie, ibid., April, 1895. COMPARISON OF ENGLISH AND AMERICAN SCHOOLS. 393 valleys as a matter already proved, and goes on from this to the possible ultimate result of the valley-making processes. Again, the English school denies, tacitly or directly, the probability or even the possibility of a period of still-stand long enough for essentially complete subaerial denu- dation close to sealevel, but assumes the possibility of a period of still- stand or of slight depression continuous and long enough to allow the sea waves to plane off the sinking lands. The American school tacitly questions the occurrence of great erosive transgressions of the sea during either a still-stand or a slow depression of the land, but admits the possi- bility of essentially complete subaerial denudation to an average sealevel, above and below which the land long hovers in many minor oscillations before a new attitude is assumed by great depression, elevation or de- formation. It should be borne in mind that the depressed and buried or the uplifted and dissected plains of denudation whose origin is in question are in no cases geometrical planes; they nearly always possess perceptible inequalities, amounting frequently to two or three hundred feet; but these measures are smallcompared to the inferred constructional relief of earlier date, or compared to the deep valleys often eroded be- neath the plain if it has been uplifted. By whatever process the so-called “ plain of denudation” was produced, an explanation that will account for a peneplain of moderate or slight relief is all that is necessary. Abso- lute planation is so rare as hardly to need consideration here. In no respect is the contrast between the two schools more strikingly shown than in the beliefs concerning the cover of unconformable strata that lie or are supposed to have lain upon an oldland. The continental members of the English school generally regard these strata as an essen- tial result of the process of marine denudation during slow depression ; if such strata are absent from a dissected plateau, their absence is ex- plained by denudation after uplift. The American school does not give the cover of unconformable strata an essential place in the problem; if present, it is generally ascribed to deposition following the submergence of a region already for the most part baseleveled by subaerial agencies. REVIEW OF THE A PRIORI ARGUMENT. It may be noted that the value of marine agencies gained a high repu- tation for effective work before subaerial agencies were recognized as significantly affecting the form of the land, and that from that time to the present the importance of the latter agencies has been steadily in- creasing in the minds of geologists. The manifest work of waves on a bold coast was perceived ata time when the production of valleys by rain and rivers was scouted. Today it is not so much that the absolute streneth of marine erosion is given a smaller value than heretofore, but XLVI—Butt Grow. Soc. Am., Vou. 7, 1895. 394 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. that the relative importance of subaerial erosion is rated much higher than at the beginning of the century. While the sea works energetically along a line, subaerial forces work gently over a broad surface. Chiefly for this reason Geikie concludes that “ before the sea, advancing at a rate of ten feet a century, could pare off more than a mere marginal strip of land between 70 and 80 miles in breadth, the whole land might be washed into the ocean by atmospheric denudation.” * A slight movement of elevation usually sets the sea back to begin its work anew on the seaward side of its previous shoreline, but such an elevation only accelerates the work of subaerial denudation all over the elevated region. The waves on the seashore shift their line of attack with every slight vertical movement of the coastal region ; but the sub- aerial forces over large continenta] areas gain no notice of slight move- ments until a considerable time after they have been accomplished, and hence they perform their task only with reference to the average atti- tude ofthe land. Observers near a shoreline naturally have their attention directed to the unsteadiness of the land, as indicated by marks of many recent changes of land level; hence they are perhaps indisposed to admit that any land has ever stood still—or oscillated slightly above and below an average attitude—long cnough to be nearly or quite baseleveled by subaerial agencies. They prefer to think that the sea is, in spite of its many stops and starts, the great leveler of the lands. Some have intimated that the insular position of English observers has led them to exaggerate the relative power of the sea. Thus W. T. Blanford, after much experience in India and elsewhere, as well as at home in England, writes: “It is not surprising that the power of rain and rivers should be recognized with difficulty in regions where their effects are comparatively so dwarfed as in the British isles, while the power of marine denudation is at its maximum from the enormous coastline exposed and the small amount of detritus furnished for its protection by rivers of small length and in which floods are of exceptional occur- rence.” fT But even this weil practised observer contended only for the subaerial origin of valleys, not of plains also. On the other hand, those whose studies have been directed chiefly to large interior areas seldom have occasion to observe the action of energetic shore waves, and hence are apt to attribute relatively little importance to their work. The small share of attention recently given by Powell to shore waves and coastal forms in a general discussion of physiographic processes and features is perhaps thus explained. { The citation from Dawson, given above, is * Text-book, 1885, p. 432. + Geol. and Zo6l. Abyssinia, 1870, p. 158, note. jt Nat. Geog. Monographs, vol. i, 1895, nos. 1 and 2. THE A PRIORI AND A POSTERIORI ARGUMENTS. 395 an especially good illustration of the manner in which large continental surroundings may affect the opinions of an observer who, from certain associations, might be expected to follow the insular school. Although mature deliberation and good judgment may lead through a priori argument to a safe conclusion in many problems, the method is of difficult application here on account of the great number of variable factors whose appropriate values can be hardly determined. It is prob- ably by reason of assigning different values to variable factors that the opposite conclusions summarized above have been reached. STATEMENT OF THE A POSTERIORI ARGUMENT. In attempting to decide by arguing from effect to cause whether evenly denuded regions have been worn down by subaerial or marine agencies, let us try to stand on a provisional Atlantis, hoping that it may give steady support long enough for us to gain an unprejudiced view of the opinions that are so generally accepted on the lands to the east and west. From this neutral ground let us attempt to deduce from the essential conditions of each explanation of the problem as many as possible of its essential consequences, and then confront these consequences with the facts. The measure of accordance between consequences of theory and facts of observation will then serve as a measure of the verity of the theory from which the consequences are derived. No final decision can be reached in many cases; for, however clearly the consequences may be deduced, the facts with which they should be compared are often beyond the reach of observation. In such cases it is advisable to announce in- decision as clearly as decision is announced in the others. As far as I have been able to carry the analysis of the problems, it is more difficult to find positive criteria characteristic of plains of marine denudation than of plains of subaerial denudation; hence I will take up the latter class first. It should be remembered, however, that in each class of plains both classes of agencies may have some share, one pre- ponderating over the other. CONSEQUENCES OF SUBAERIAL DENUDATION. Imagine a region of deformed harder and softer strata raised to a con- siderable elevation. Then let the land stand essentially still, or oscillate slightly above and below a mean position. ‘The rivers deepen their val- leys, the valleys widen by the wasting of their slopes, and the hills are slowly consumed. During this long process a most patient and thorough examination of the structure is made by the destructive forces, * and whatever is the drainage arrangement when the rivers begin to cut their *See Bearing of physiography on uniformitarianism. Bull. Geol. Soc. Am., vol. 7, 1895, pp. 8-11, 396 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. valleys a significant rearrangement of many drainage lines will result from the processes of spontaneous adjustment of streams to structures. This involves the adjustment of many subsequent streams to the weaker structures and the shifting of many divides to the stronger structures. Adjustment begins in the early stages of dissection, advances greatly in the mature stages, and continues very slowly toward old age, while the relief is fading away. Indeed, when the region is well worn down some of the adjustments of maturity may be lost in the wanderings of decrepitude, but this will seldom cause significant loss of adjustment except in the larger rivers. Now, if a region thus baseleveled or nearly baseleveled is raised by broad and even elevation into a new cycle of geographical life, the rivers will carry the adjustments acquired in the first cycle over to the second cycle. Still further adjustment may then be accomplished. The master streams will increase their drainage area in such a way that the minor streams will seldom head behind a hard stratum. In a word, the drainage will become more and more longi- tudinal and fewer and fewer small streams will persist in transverse courses. All this is so systematic that I believe it safe to assert that the advanced adjustments of a second cycle may in many cases be distin- guished from the partial adjustments of a first cycle. It should be noted further that in the early stages of the second cycle the residual reliefs of the first will still be preserved on the uplands. and that they will be sys- tematically related to the streams by which the dissection of the upland is In progress, as noted in the examples described by Darton and Hershey. It is manifestly impossible to apply what may be called the river test to plains of denudation upon which a cover of unconformable sediments is spread; but, before assuming that such buried plains are of marine origin, their uppermost portion next beneath the cover should be exam- ined to see if it presents indications of secular decay before burial; and, if so, a subaerial origin for the plain may be argued. Certain aspects of this division of the subject have been discussd by Pumpelly.* Another matter of importance is the character of the undermost layers of the cover. If these are fresh-water beds a subaerial origin for the plain on which they rest may be inferred. The ‘Potomac formation offers an example of this kind. + CONSEQUENCES OF MARINE DENUDATION. Now suppose that a region of disordered structure is partly worn down by rain and rivers and is smoothly planed across by the sea during a time of still-stand or of gradual depression. The land waste gained in the * Bull. Geol. Soe. Am., vol. 2, 1891, p. #11. + McGee: Am. Jour. Sci., vol. xxxv, 1888, p. 187; Fontaine: Monogr. xv, U.S. Geol. Survey, 1889, p. 61. ‘ CONSEQUENCES OF MARINE DENUDATION. Bulk later attack will be spread off-shore on the platform abraded in the earlier attack. The basal strata of the unconformable cover thus formed must indicate their marine origin and must be appropriately related in com- position and texture to their sources of supply. ‘The drainage systems of the land will be essentially extinguished by the encroaching sea. When the region rises, with the cover of new sediments lying evenly on its smoothed back, a new system of original consequent streams will take their way across it. If the elevation be sufficient, the streams will in- cise their valleys through the cover of new sediments and in time find themselves superposed on the ‘“oldland” beneath. As time passes, more and more of the cover will be stripped off; at last it may disap- pear far and wide, although the stripped surface of the oldland may still retain a generally even sky-line as a memorial of its once even denuda- tion. Now, in this case, the rivers by which the dissected plateau is drained will have at most only a very slight adjustment to its structure. Their courses will have been inherited from the slope of the lost cover ; they will at first run at random across hard and soft structures; a little later some adjus tment to the discovered structures will be made, but as long as the even sky-line of the upland is recognizable, only the incom- plete adjustments appropriate to the adolescent stage of denudation can be gained. EXAMPLES OF DISSECTED UPLANDS WITH ADJUSTED DRAINAGE. This essay has already reached so much more than its expected length that it will not be possible to give extended space to the consideration of specific examples. This is, however, no great disadvantage, inasmuch as the number of examples in which the problem has been considered in relation to drainage arrangement and other discriminating features is very small. The various articles already referred to concerning the geo- graphical development of the Appalachian region treat this aspect of the subject with some care; to these may be added my paper on “ Certain English rivers,” in which it seems to me that there is shown some ground for the consideration of the.alternative to the usual English view. Of the Ardennes it may be briefly said that systematic longitudinal and transverse streams are well developed in certain areas, and in those parts, at least, there does not appear direct evidence of marine transgression. Sheets 48 and 54 of the Belgian topographical map (scale, 1 : 40,000) ex- hibit these features very clearly. On the other hand, the branches of the Rhine and the Moselle in the Schiefergebirge suggest superposition from a lost cover, as mapped on the sheets of the Karte des Deutschen Reichs (scale, 1 : 100,000). It is manifest that many plains of denudation, now uplifted and more 3998 DAVIS—PLAINS OF MARINE AND SUBAERIAL DENUDATION. or less dissected, may be found in which no simple test based on the presence of superposed streams will serve to settle the question of marine origin. Indeed, it appears to mea difficult matter to adduce any ex- amples of extensive plains of denudation whose origin is demonstrably marine and to whose planation subaerial agencies have not contributed the greater work. A region may be almost reduced to baselevel by sub- aerial denudation when the transgressing sea completes the work, extin- euishing the adjusted valleys and introducing superposed streams in the next cycle of denudation. A region well baseleveled under the air may by quick depression suffer rapid ingression of the sea, whose shore waves will during depression nowhere reside long enough to perform a signifi- cant amount of abrasion. When the region is thus submerged and stands again relatively quiet, the waste from a non-submerged area, gained both by marine and subaerial denudation, may be spread over the denuded and depressed plain, and when afterwards elevated with an unconform- able cover that will induce superposed drainage, all trace of former adjustments will be lost; yet here the planation was not marine. A district of superposed drainage in central New Jersey, where the Amboy clays once spread over the red shales and sandstones of the Trias, may probably be taken asan example of this kind. Superposed rivers cannot, therefore, always be taken to prove that the uplands which they dissect are uplifted plains whose denudation was chiefly performed by the sea. Regions of essentially horizontal structure normally have wandering streams; no systematic arrangement of drainage is here to be expected. Discrimination in such regions has seldom been attempted between ex- amples of one cycle of subaerial denudation, now adolescent or mature, and examples of two cycles, the first having reached old age and the second now being in its adolescence or maturity. The sky-line would — be smooth and even in examples of either class: in the first, because its original constructional form was a plain; in the second, because it was planed down essentially smooth at the close of the cycle preceding the current cycle. It is, however, sometimes possible in regions of horizontal structure to recognize the records of old age reached in a former cycle by a slight discordance between the general upland surface and the attitude of the strata; or by the association of the region with an adjacent region of tilted structure where indications of an earlier cycle of subaerial denu- dation are manifest, both these tests being applicable in the Allegheny plateau ; or by the arrangement of the faint residual relief of the uplands, where not trenched by young or adolescent streams, this test having been applied in the Piedmont district of Virginia, in the Ozark plateau of Missouri, and in the Great plains of eastern Montana. Further study of many other examples is desirable. GNV1S! 30OHY ‘GNVISI 3ONS0NYd ‘iNIOd GNVS Jas Sl Id ‘G68! ‘Z “OA ‘WV ‘OOS ‘1039 ‘11Ng BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 7, PP. 399-422, PL. 18 MARCH 28, 1896 CUSPATE FORELANDS BY F. P. GULLIVER (Read before the Society December 27, 1895) CONTENTS Page LULSTECG. TI EIRIG Ds «6/4 PALA 7 SUE RO SIO ee EE MRE a ING PDT a oP 400 “NAITIAUTTS oo 0 aye. Ar Riri etc ee agate Om AL Ca ee 401 Bre eA RAPE M TR DRE IAP ih ye) 0.12 haath ake u ot's) bball abel g'ate leyiler vie b-ghetey ® were Am ala ha clone 401 rem CMM PHN ALOE Ge) seeker atts aN han A aah acy. AS Bis ele! dala ah oho wid WS He ble Ae ciacae altel Selb dlets 402 PVR OR OA Piney PSN OF Ss ch Se figiele isl, ai chac totor'es gray wichwaninne & Av Dyerau Dlecro ithe ayn's aetebe lar’ 402 Sette Gat PIP EON se Sie sa aera eS e cca! chins act’ yn! a)/aie, ep 5) nid a! ooe/ ai mletauey wee ee Oem sem eels 402 LEVEE ESIEL, SII CIERIA & rie Sc Aa ve I va tere sea 402 Mamma MON shOr CUCM i. 5. oes vied h se yee Ga eee we ca see eelade snc 403 Monee MUCMGEeSONOMCORY A 2 sks. eet eek dese lle bs Bolev eet ane tees 404 MEAT 6 Shh tio een g dial afer e lw cwae VIO ei Ed bs HIS AON Rie hee ane es 406 fee remnmemine OATOMWMAS Ns 15) tana ' ol wise bone e/t gb Sates eedts adi Mlges ola Side 407 rere nn TN GIN © TOMI fe Pea o exci teyeteyen viel seeak's wove » we dhalal en sie com tees 7 ee 408 emesmmaenections tm the Carolinas... 2.200.000. seednty igen deccdeede sts 408 Seta OTE The OO OKOUUUN oe ei one siepe: 2 9% yal > wicuiew anelis ee eie/ bul s) weave ars, slemengrs 409 Srlgra MMT e NG INTCE CAWOS case 8s Lc. oes aise #0) ee er o bine Gro noe who Seite etree ia vole 409 PETE e OMMEGOMN LOG WHINY PACT 66.6! 2 eis.0 eure jaf sie rnn eyo s Sesie Blaceeidlietace a Grae ears aac 409 Val GP ZSUMTI ON SS ARN eR ano RR eS a . 410 SOMOLEM, CIDOB! 9 © 6/6 GAL See Age ee eo eet em ee sapewn' ALL Maman rMGkdeSerioGlOM, sa wi). . 4. nl Seen PUB ah lathes walwery ode alt 411 Tepe ee sta tO et Fe rath a Ae 411 Beg NONSENSE Ale a ledeirk wads) 1 on cae plein syavonsuien eae a ayer whateva Bee een oat 413 SIDES? SUEBYOSr gS SO aS aN cen ee Sr eS PO et cesT 413 Lagoon-marsh stage.......... Sash Dt HORA Rom efrSt Se MunTe ELEY? ean iat S oi Mehay acest 414 ATTN WMC MMR NRA 2 ofa) ate aie JhokSie eo bitin eRveh oneeaaadena cm eee uel ic Sack esa 'o ets als 415 Drie ama chem@leeo GOV ner, x2 Are tists nsttsw-cishe aalalave Mie ehQ NSE. ate lelew’ aleve Wee ce cs 416 Meomymcomeromved With tacts 215.2 biloiee nee sale slewier dees ae sie eecle belts oo 417 Lote TE, CICS gc di SE i OO RGD Car Sa eR a 417 PNAC LOM VES UAE TINCT, sino 2

ZESI 4 2 dIHSONZYNIE se) VLA LNOW139 PEP ) Pf) = A oa, Sdijj\ud = oe Ny on ‘ 3) ‘S ‘ ? CN 5 fay a \ hw a é Las ‘ Q i EF» P % VIITRONV 4 a H fo) * SS a soit QNOW1V/S L be ele) ) ‘ ATTIASTISNYOH & : E Svitaqo tases Bike : s a) oS) gez!* 661) : viav oe) LuodHuy © nv We MEISE CSE ipyee OS Yap aaa Se auosnisny: A 4 G02 ° elites 79 MOLHONOH f SNUNgo@ A H s® 22, o ff a0e = S Pp _ 26 - of vSfti RO IXOW TTI ry ? VOVHISYNV YS a h~ 5 1 8 SNIVMSO f W? 249 | H ee LOEI ( ee Saenn Pjo5 ‘ H f @OADIS Ip Hy 1; —~ 4h 3 a ‘ C\ 0881 f Ra \ sNOLiv (, oo? cence \ 2@ $ Pos r C) » SINMH _f veil 7 / anniasnva ae CSE ee : > ©) ere 7s, LN oreo cle) oy 1 £ WY ; § ‘S02, H 7 D & ON S\ounasivia} ge” > x vl WYHONID WS ENESEE VALLEY. insverse to the water-parting. ine. ‘Ce BULL. GEOL. SOC. AM. VOL. 7, 1895, PL. 19 Hi é lA we ncottnene ma! ENS YLVAM SCALE oF MES EW 1 \ } gscortsauns oanveonT Ho ge epee ES CONESUS L RUSHFORD HYDROGRAPHY OF THE GENESEE VALLEY. Wat ng shown by heavy broken line. Gl outlets indicated by bars transverse to the water-parting. Figures indicate altitude above mean tide. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 7, PP. 423-452, PLS. 19-21 APRIL 14, 1896 GLACIAL GENESEE LAKES BY H. L. FAIRCHILD (Read before the Society August 27, 1895) CONTENTS Page LULD'GC UVCLILOIIS & oy ceBIS 2 Be Oe eae er aa ean RAL ae 424 Meee omaGenesce Valley! ices eee ow ee ee ee 425 PimaUuOctpMycamd: GopoeraAphy.!.i.5 028s oes oe Be von peace gees ee 425 LOTS) CIBTOO RINSE 8 UR Cis Oe Rei an apenas ea ae er ean Aa AEN Fl ee ae 426 LES DENCH SIONS Cre SES Es) eS ea Se 427 Postglacial channels of the Genesee........ Bs Taate Aisa Cetin caked A PD 427 Peostelaeial channels: of tributary streams..........00..02.. 0000000. ae. 428 iseMICMRC AMICI SKOlNMmMe GEESCCs.).c.0.) ve eek ee etl dese. eee fae e as 429 Sequence of events in the geological history of Genesee valley........ ss Sema) ihaprodmetony Stabementy 00... desc. o ee ee lee ee LOR Shy Pes inde iis 8 2,8 429 irae preelacrlosubaerial Crosion: 2 oy .0. ce eee ee ee we eee ees 429 Ppisedesorlaikes! by ice-ad Vance... 66 ee. ee dal Set eee eee ena eee 430 MSOC eKOIMe AClanOMs At. yas iayh on sta se oe ee aL ae Sy ae aaa eae 48 orcaederotslakesviony ICC-remre at. .\.). 65s. ween le ieee een ee ea de sae ae edel 430 MOM OnuMNOraTMal MAKES. (os) isUels. Lie eis s eed e aed eee Ob eb pa see les 430 raver postelacial: subaerial erosion.....2. 20.055 .0. 00 cee eee neve eee 430 Peer Otte rola cla lla wes so... sien) 2 dase et ies Wa bagays @babe eon cma 431 OEM OL Oi Wiabel laMCGs —.\.' She's cs hinds iia asa ois 2 oth Slt cieraded Seana sty ble Woke 431 Mier tier Memo USP meet vary ce al, ba acasnisie Xx byoudies Goa. wate ancealelecson Muures uaa ey eo 432 Bionic meres tenvemMeOM ba cls <2. a cys erbatenieed.ckestin hos ais eisteeie | Sod ata ee ae saccn © ous « 432 Wecemipnon of the glacial lakes........3....... oe PU te oele aes EAT RCN Rs S ene 433 hiiemouace: wlbrce primary KiukeS. 20. o ihe dames dees Pee se se ade domes 433 (LTBI) sisted 2 Sealed 0 ARR Ms Re Me Sore BO oT) A) A A 433 Lah S INTIS OUT Scar oss ie Wee MCR RS GAMERS “E aT RAR Nac ec TA HAC 417 Sn ee 433 Sccondustee arenmnsylvania lakevsi,..cooerr eile. eS 434 (CUIUL EST IES, ON oa ta 8 ad ee gee LOR Pe ee eS ict op ie 434 RSP SCOT ho) che. onsy cya ees x a peeeasei yee ORI de REIL Peace at cao! SURO Ge 434 Peete aN Velie he Males 8 sl 5. ahha Wye Aleit ots cee ater e colons cs Wed Slabea ay 34 OMT Gs sb SSE ea re Bc TAP St Pas ec ag a 434 WVEIPCENCIS Ros chloe: oA eee P MERI AREER yh iG a fiche a ulte' poe eles 435 Houmiarstce: beltast=Hillmore lakes Mowe eco e se ties Cee Poe sa use tele 43 OTE LS Garnet cys Sea ye ale oc cp inks cat Pee Ne ken cn NEN aoe wl af Daud 436 NVPAULGIE LOW CIS aie Sitters ox < alo) Aa emis Sete Ntts athe Waletai a -eleia wane 4 hed 437 L—Butt. Grot, Soc. Am., Vou. 7, 1895. (423) 424 H. L. FAIRCHILD—GLAUIAL GENESEE LAKES. Page Fitth stage: Portageville-Numda lake! sve ee eter ee een, ae ee ee 438 Outleta coe costs es BAS SS ee ee en eee 438 General deseription:.<\.f.< 542 See see eee eee ee ee 438 Primary OWwuetig iso oo. siais dna ae ee eee Nace ea nee ee 439 UWiltimate outlets... o's. «0:5 as0e he ethabere, sparen See eer aay etree ce en ae 439 Water levels aici. cau 5 t's, bi oieeee theory ten ete arene hee ates er hee 439 Sixth stage: Darisville lake. i -o5. Aah Boe penis cre bo on Oe cp re 440 OUClete sale eed Sik vcs SS eee Ce bese nen ge Mop Spates © .. 440 Woater-levels s,s occa: sc egs set wis) o ten wl nelelate sus ae eee S Fleet oe ae eee 441 Seventh stage: Warren tributary lakes: 2.2 ps9 45--e oe te hee ee eee 442 Oatletig oc cis sak sein oe See aye ela ae eae tere pe 442 Water-levels............. sls nh SN BS NEE Ne ate eee hee ae 443 Highth stave: Warren waters. ..: oi. 23h. 2 ones saa eee ee ee ee 443 Ninth stage: Iroquois waters. 2.240: iw epee same bays cine eee oe 445 Tenth stage: lake Ontario (non-glawial): >. 7. soeeeec +s oan ee 446 SUIDIMLATY «sb iss a sine een m Mew a cle lereee mein ole emia eit eine ee ee 446 Contemporary local glacial lakes... ... 02... «- «ches soe ee ,. 446 In Genesee hydropraphic area... 2s. sacs ewinne ake inl oe 446 Conditions affecting formation of local lakes.......,.. .< i¢ina0 =e 446 Knight Greek lake.’ sia. in sped oaenie sets eel ee ie 447 Friendship (Van Campens Creek) lake...........0.eeceeeceersseeet 447 Black Creek lake..i:.. 0... sss staseute seeming hid eee ae ed eee 448 Rushford Lake sev iv spo 050d sv re dew de empl ers eis Ride Se Gee 448 WW re W Deb is jx onic s- anip wom my red gga ops wees Sete a a a 448 Dansville lake « 0s. 2» as's'ss 54 ose ot kines onl s arene ree ae ee 448 Seothabirre: Dales: x i.9 cress ute wv awn e's ta Phar ee Ree 448 Springwater lake. . acca. ssa. becom devine v temo ab ale ee 449 Tributary to Genesee HIER.) 05 2.55. nice wie ss sya a opie ple an ee 449 Subsequent morainal lakes . . 2 62002 ci eas shu pe odes wale nly ea 449 Tn Gemesee Piver. oa. .dsis aap 'sctaw Copp» pun 6 & oiecaiy bine Oe ee 449 In tributary streams.; i05. 50.0 <0). ss. Vol.4..) Vol-5. ~ Vol.'6: Total. From Fellows............s066 Soacdde $2 45 $1 55 $0 35 $0 70 $2 75 $2 25 $10 05 From the public..............s.0 0 40 2 05 IPOD uanbedecko. eboccdaaseuan ecedaecates 3 95 Total for 1895.............. 2 85 3 60 1 85 0 70 2 75 2 25 14 00 By last report (1894).............. 18 05 13 00 4 00 4 85 MOD Digescam access 41 05 MNotal to date-:-2.......-.-.. $20 90 $16 60 $5 85 $5 55 $3 96 $2 25 $55 05 CTEM ClpetOltel leeenemomcuerecctetence toa ences casseaiccuvecois cet sciieensugecsudseeseeduccds eusae oeeeaneueben beceeeune $2,248 65 EVE COUVC CMO LEO LUMMCH pelea CL WELING Crancsevstidceossnerocenciansiseceatecstccsenctncceceaera sosecocerees 35 00 PRO Vale EC CM USHLORG AL Cleersenstccrcsweacl! saa ctosliecisss oso e sw case atlsasosioce Su alia Spaceeaedoecase steubecaasosane $2,283 65 BA ommineharsed. amc we MACOS Cte ses..cc.ccc2.. cues deevessnncue «ocesacdeaseucossuedaharcanacksouses 131 90 BRO cae Ulle Cinies AVE SHUORMAUCInccer scenes sccesiiscerter es scececrecesuscestcaehuonte a ennn steccemeeconeetees $2,415 55 Exchanges.—The list of institutions to which the Bulletin is donated now numbers 85, of whom 10 receive brochures and 75 the completed volume. Since the printing of this list, in connection with the list of the library at the end of volume 6, the following changes have been made: Library of University of Toronto and Library of McGill University have become subscribers; Engineering and Mining Journal removed from the list; Museo de la Plata, Geographical Society of Finland, Geological Survey of Sweden, Cincinnati Society of Natural History, National Geo- graphic Society, and American Geographical Society added to the list. 456 PROCEEDINGS OF PHILADELPHIA MEETING. Library.—The books, pamphlets and maps belonging to the Society are deposited with the Case Library, Cleveland, Ohio, under a contract which was outlined in the Secretary’s report of last year. ‘The lst of material deposited up to January, 1895, is printed in the Bulletin, volume 6, pp. 001-516. The list of material deposited during the present year should be printed in volume 7. The library is available for the use of the Fellows under the following rules: 1. Fellows are permitted to draw out material in reasonable quantity for a period not exceeding two months. 2. The transportation charges both ways and other expenses are to be paid by the Fellow so borrowing. d. The Fellow is held responsible only for such loss or damage as may occur through his fault, as, for example, by insufficient wrapping or mis- directions. EXPENDITURE OF SECRETARY’S OFFICE FOR THE SOCIETY’S FISCAL YEAR, NOVEMBER 30, 5 1894, TO NOVEMBER 30, 1895 Account of Administration Postage wicca ae yh ao axipe ole» tet es ak ee ere oe $36 70 WX PVesSAGe..< 5.. a: dad oe shad, Soest en eee Re ee ee 4 77 Stationery-aned TeCOrdg. . ../¥ip. cas = tranche a bet oie ae ort hg cane ees 4 95 Printing, including stationery. tcc. =p seine aaa ie eRe ee 105 61 Mieetini oy oc e005 aa eee, cre de oe nk eee a La eee 17 50 Library. ..sss'. the southward. These head in a plateau of Eocene overlain by the Lafayette formation, and they cat more or less deeply into clays and sands of the Potomac formation before they reach the low, swampy flats adjoining the river. At Fort Motte there isa high bluff !ying a short way back from the river, in which the Potomac beds are seen overlain by the Eocene marls and dark clays containing abundant fossils. In my trip down the Peedee river from Cheraw I traveled in a small skiff, and although [ made a careful search for outcrops I found very few. It was not long after a freshet and the water was still quite high, so that possibly a greater num- ber of exposures would have been seen at low water. Low banks of Columbia formation and wide areas of swamps were the only features that I observed except- ing a short searp at Gardners bluff, 10 miles below Cheraw, and Hunts bluff, 20 miles farther down. In Gardners bluff the following section was noted : 1: Orariee aril doit said Loam. ).-d/ee aoa ee ee oh ees yi a Raps eS 5 feet 2. Orange sand with pebbles, cross-bedding |. «03 4-b. se. View oe eee Gis 3. Coarse gray sand with butt streakaync.. aie, 2a ey ane be eee one Sade 4. Coarse gray sand, somewhat cross-bedded; lines of clay and quartz pebbles... \ 2S 264 nokia GS Sark Oe ae Bin eee ches ne ier eA iin ge 16% 5. Fine gray sandy clay, massive; contains indistinct lenses of purer clay.. 8 “ Numbers 2 and 3 are sharply separated throughout by a slightly waving line ; numbers 4and 5 are separated by a moderately sharp break, which is quite irregu- lar at several points. Beds 1 and 2 are probably Lafayette in age and the under- lying deposits are undoubtedly Potomac formation. At Hunts bluff there is a 20- foot exposure for about 200 feet along the northeast bank of the river. At its base it exhibits gray, cross-bedded sands, with a few scattered quartz pebbles; next there is an irregular bed of pebbly sands and then stratified gray to brown sands, merging upward into brown-buff loams at the surface. The basal beds appear to be Potomac, but the evidence is not conclusive. A short distance eastward in the higher lands about Bennettsville there are observed the marls of marine Cretaceous age, Which overlies these Potomac beds. The marls come to the river bank at Mars bluff, below the railroad bridge, 10 miles east of Florence, in a fine series of exposures which have been briefly described by Tuomey. The Potomac beds there are sands and clays of the usual character, and they pass beneath the river a short distance below. The marine Cretaceous marl and marlstone contain abundant distinctive fossils, and they are in turn overlain by fossiliferous Eocene marlstone for some distance. : Several deep wells in South Carolina have penetrated the Potomac beds and their records throw some additional light on the relations. The well in the village of Aiken was bored thfough the formation at a point near the western edge of the Eocene, and consequently it exhibits the full thickness of the Potomac. The fol- lowing record is given : 0-45 feet red clay. Lafayette. 7} 45-100 ‘* sand. ; | 100-130 ‘ ‘‘chalk.” Mixture of fine white sand and kaolin. | Potomac. 130-465 ‘* sand and soft sandstone; some clay. 465-741 ‘‘ granite. J A well at Florence, having a depth of 1,335 feet, passed through Miocene, marine Cretaceous and Potomac formations, and at 608 feet entered typical red sandstone ‘ ; RELATIONS OF LOWER MEMBERS OF COASTAL PLAIN SERIES. 517 of the Newark formation. The Potomac beds begin at a depth of about 110 feet, as nearly as I can recognize them, and consequently have a thickness of about 500 feet. They consist of gray sands in greater parts, presenting considerable variety in coarseness. Considerable lignite was reported, but no clay appeared in the few samples of borings which were saved. For the well at Marion I was unable to obtain many definite data, but learned that the lower members of the Coastal Plain series were alternating beds of sand and tough clay, and that their floor of crystal- line rock was found at a depth of 700 feet. At Darlington the Potomac sands were found underlying the Cretaceous marls, but the depth of the contact was not ascer- tained. At Orangeburg the Potomac sands were entered for some distance in a well which has a depth of 1,160 feet. i was only able to obtain meager informa- tion for this well, and vould not determine the nature or age of the beds which lie next above the Potomac sands, but I should expect them to include a considerable portion of the marine Cretaceous marls which extend from 430 to over 1,800 feet in the Charleston well. I might here add that I am strongly inclined to believe that at least a portion of the lower beds in the Charleston wells may represent an offshore phase of the Potomac formation. However, as higher Cretaceous mol- luscan remains were reported from a depth of 1,955 feet in the first well, the water-bearing sands and sandstones from 1,960 to 1,980 feet may be the top of the Potomac formation. Although I observed plant remains in the Potomac beds at many points * which would no doubt settle any question as to the age of the beds, I have depended en- tirely on the structural relations and physical characteristics for my correlation of the formation.. There could not be any doubt as to the continuity of the great series of sands, clays and sandstones which underlie both the Eocene buhrstone and the Cretaceous marls and lie on the surface of the crystalline rocks. As this series lies below quite old marine Cretaceous and above the Newark formation, the most obvious correlation would be with the Potomac formation, which occupies this position for hundreds of miles along the Atlantic slope, and moreover their physical characteristics fully bear out the correlation. Of course in speaking of the Potomac formation I refer to that formation as a whole, comprising the Tuscaloosa beds, which I believe are eventually to be separated as an independent formation. Marine OCreraceous ForMAtTION I have nothing of general interest to add to the statements of Tuomey regarding the Cretaceous marls and clays, for I did not extend my observations very far into their area. The formation appears to thin out before reaching the Wateree river, and on Black river it appears to be buried under the Tertiary, excepting possibly for a short distance near Kingstree, where it is indicated on the geologic map issued in 1883. The enormous expansion of the formation in the well at Charleston is a rather surprising feature, but, as above suggested, it is possible or even probable that the lower beds in this well are offshore deposits of Potomac age. EocENE ForRMATIONS The lowest Eocene beds westward are the buhrstone and some argillaceous marls which underlie the buhrstone at certain localities. To the eastward there are sev - *J have been informed by Professor Lester F. Ward that he has discovered plant remains of Potomac age in the extension of these beds in North Carolina oh the Cape Fear river and also at various points in eastern Alabama. LXII—Butt. Grou. Soc. Am., Vor, 7, 1895, 518 PROCEEDINGS OF PHILADELPHIA MEETING. eral hundred feet of overlying marls, and the buhrstone appears to lose its char- acteristics in the extreme eastern part of the state. In the northern portion of the state the Eocene formations thin rapidly as the marine Cretaceous beds rise to the surface, and they are finally represented by thin outliers, often lying quite widely scattered over the irregular surface of the Cretaceous marls. The western edge of the buhrstone passes from Aiken to within 10 miles of Columbia, and thence to the eastward to below the confluence of the Congaree and Wateree rivers. In the wells at Charleston the Eocene members have a thickness of about 370 feet, and are sup- posed to lie at about 60 feet below the surface. They there consist of marls of vari- ous kinds, which are mainly argillaceous above and more calcareous below. The buhrstone is a very hard silicious rock, often 15 to 20 feet thick, and usually filled with shells. The overlying marls and marlstones are known as the Santee beds, which consist mainly of light colored marls, with some beds of marlstone of con- siderable extent, and the Ashley and Cooper marls, which are of darker color. MiIo0cENE FORMATIONS The Miocene deposits consist of sands and marls, which occur in scattered areas mainly in the northern and eastern counties. Lately Dr Dall has found evidence that the phosphate deposits are also of this age. The thickness of the sands and marls is usually not over 30 feet, and they le on an irregular surface of the Eocene or marine Cretaceous formations. LAFAYETTE FORMATION * This is a superficial mantle of orange sands and loams which covers the higher plateau regions. The elevation of this plateau is about 650 feet along the western border of the coastal plain. There it has a thickness of from 30 to 80 feet, and its more loamy portions give rise to the greater part of the ‘“‘ Red Hills.” Its eastern extension has not been traced, but it is thought to be the same as some of the younger Pliocene marls. . CoLuMBIA FORMATION This is a thin capping, mainly of sands and loams, which covers the lower lands and appears to extend as high as 400 feet or more in the higher region, giving rise to some portions of the ‘‘Sand hills.” RESUME OF GENERAL STRATIGRAPHIC RELATIONS IN THE ATLANTIC COASTAL PLAIN FROM NEW JERSEY TO SOUTH CAROLINA BY N. H. DARTON Remarks were made by D. W. Langdon, Jr. An abstract of the latter paper is published in the American Geologist, volume xvii, page 108. * Described at several localities by McGee. Twelfth Annual Report of the Director of the U.S, Geol. Survey, 1892, pp. 347-521. rit i t NISV@ SASSANN31 SHL = +o “Id ‘G68l ‘Z “IOAN ‘Av ‘OOS 10359 ‘11N¢g SOME STAGES OF APPALACHIAN EROSION. 519 The following paper was then read : SOME STAGES OF APPALACHIAN EROSION BY ARTHUR KEITH Contents Page MEMO CUMCLONY os ionscse: sncistassfenesscdsiocacaboseccevosvctesesecs ASS nGOROSHObOOsC0GR OCHA AaBeueb bo nCoOEEObOuOOE elibonsascensnewarastamtecets 519 Coneraliteatures of the southern A ppalachiaMs..cr.ccccsacs-eccecccccercccersa=+seccscccsicavss cocaciseeeecescuecswcieceses 519 UD AMINA SHO ASHMS ewan tsicseveerssjoe senor eeccaccerstcciescinassestsedceest cece acimorcuudanuc seas autumersaciecdsaievewenswaschecceas'essses 519 MUN AC Cm OMNIS ssreoccasievessoceaenscoecualoueconeeuecaeetasulona cous tebenctlson Tocaem ee cuaeiren cuGcaawuicen sdwdclddivey suc sscaepeeasedecaice 520 VE Mela OTUSUOMMLE NO IWCs ceo ae wich e sc ceesacSuese cou niewecccawa ca esest cceine tease tareerdoswncesteseussccedcucted osceesbescisaevedventrcin 521 Heatunes of the NennesSee (Pasi... ...c..cccececessewececaccuccs onenascoacess Ecol dnauce teen aes aden scecsaaahieteecesctscsabeieetes 522 AZEING MLA SN OUMPS accseesasweseacwesossccessenssetesclacescatecisccccoee Ree eae nee meaatsh cb eeiarescausace tancionelidenceacsncensaesaacas 522 ON CHSC CLVO Mead eee e ete ceccecuoee sc salvo ececseeat as eSeGn ee enecedtanncceacuce ovuveanuctestueseeesa owes csudear tebe Peat na esc 523 Mal RAHI UME MICAS SSOION cdeoacnsceed sadcoc boseasabeeLeod bacses 500000700 SAOLIIODOF OHO SoBSS HALE ao JUdOcbode OR PUD BONE CBB SHOaEG 523 UMMM MIR Vere amas entNle cus trae ccsccuacassiccscuaisucs tt veces eolnaiiessagnoeeesesarsoonente Heuoneeceae Oscna sbueaudeDaculs Mateaseuciosnaccsmcecenee 524 INTRODUCTORY In the southern Appalachians the phenomena of subaerial erosion are shown under every phase except those of arid and glacial conditions and in nearly every stage of development from alpine forms to complete reduction. The work of degradation, which is controlled by the characteristic features of Appalachian structure and stratigraphy, itself emphasizes these features most strongly. Various publications have been made of facts and theories connected with the erosion and uplift of the Appalachians. Willis* has described a characteristic Appalachian baselevel plain; Davist has published theories and descriptions of peneplains in New Jersey and Pennsylvania; Hayes and Campbell t have published a descrip- tion, with a theory of the deformation of two peneplains shown at intervals over © the southern Appalachians; the present author 2 has described the nature and de- formation of five Appalachian peneplains in northern Virginia and Maryland, and various other publications have touched upon minor features of erosion. The sub- ject is still far from fully grasped, however, and even the broad processes of the production of the present surface are a subject for discussion. The purposes of this paper are to classify the peneplains of the southern Appalachians according to the ideas expressed previously by the author, and to oppose the extreme application of the theory of deformation as advanced by Messrs Hayesand Campbell. For these objects a systematic presentation of details need not be given. GENERAL FEATURES OF THE SOUTHERN APPALACHIANS DRAINAGE BASINS Four great groups of streams drain the southern Appalachians and are carrying on the work of erosion. First of these are the tributaries of the Tennessee river * National Geographic Magazine, vol. i, no. 4, pp. 291-300. Ibid., vol. i, 1889, pp. 183-253. + Proce. Bost. Soe. Nat. Hist., vol. xxiv, 1889, pp. 365-423. National Geographic Magazine, vol. ii, 1890, pp. 81-110. Bull. Geol..Soe. Am., vol. 2, 1890, pp. 545-581. t National Geographic Magazine, vol. vi, 1894, pp. 63-126. 2 Fourteenth Annual Report of the Director of the U.S. Geol. Survey, 1892-93, pp. 366-394, 520 PROCEEDINGS OF PHILADELPHIA MEETING. draining into the Ohio river from northern Alabama, Tennessee, southwest Vir- ginia and western North Carolina; these streams head upon the Blue Ridge, flow northwest through the Unaka mountains, southwest along the Great valley of Ten- nessee and Virginia and northwest through the Cumberland plateau. The second group form the drainage of the Ohio river in southwest Virginia, West Virginia and Kentucky; these flow northwest in converging lines through New river and various arms of the Ohio. Third are the streams of middle and northern Virginia, Mary- land and Pennsylvania, such as the James, Potomac and Susquehanna. The fourth, or Atlantic group, comprises streams rising east of the mass of the Appa- lachians and flowing directly into the Atlantic and the gulf of Mexico through eastern Virginia, North Carolina, South Carolina, Georgia and Alabama. Their waters run southeast into the Atlantic and south into the gulf of Mexico. SURFACE FORMS The typical surfaces produced by the cutting of these streams are well known through literature and need only a brief mention here. Inthe Great valley a series of long, straight valleys alternate with straight, narrow ridges. As compared with their length, the valleys are narrow,even when most conspicuous in size. In the Unaka mountains and Cumberland plateau the divides are irregular, the valleys show little systematic arrangement beyond a normal convergence into the great rivers, and the basins are broad in comparison with their width. Divides of the Atlantic drainage have small relief except near the stream heads, and the streams drain comparatively narrow, parallel basins. Two general types of divides exist—those whose summits rise to nearly equal heights and those which show great diversity. The latter prevails in the Unaka mountains and along the borders of the chief river systems. The characteristic, even crests prevail in the vicinity of the larger streams in all regions and are most pronounced inthe Atlantic drainage, the Great valley and the Cumberland plateau. Thus upon a broad view the Appalachians are most uniformly reduced near the larger streams and are most irregular near the major divides. Such a result is normal in ordinary erosion and would be expected in this case. As deduced in theory and as exhibited in the Appalachians in hundreds of cases, erosion produces a regular sequence of forms. Beginning on an unreduced sur- face, a stream cuts a narrow trench steadily downward until it reaches its baselevel of erosion. Then the sides of the canyon are attacked, the downward cutting ex- tending meanwhile up the larger and smaller tributaries until an approximate base- level grade is established, increasing in slope as the streams diminish in power. Continued wear opens out the canyons into valleys and peneplains, which in time occupy the entire area adjacent to the larger streams and extend up the tributaries. Toward the heads of the streams these peneplains contract into valleys with floors rounding upward at their borders, and these in turn give place to series of terraces and bottoms. With the division and weakening of the streams, debris becomes coarser near its source, the little falls over individual pebbles accumulate into steeper grades, bottoms are replaced by planation slopes, and these by ravines and gullies. Above all project the unreduced masses or residuals forming the main divides. The details and successive steps of this process can be seen to perfection in the streams flowing into the Atlantic, and are there rendered especially clear because the streams flow over rocks of quite uniform powers of resistance. There the uniformity and omnipresence of the concave curve establish it beyond a doubt SOME STAGES OF APPALACHIAN EROSION. | 521 as the law, and the variation of the altitudes, hand in hand with the cutting power of the streams, defines the whole series as the result of subaerial erosion. In the Cumberland plateau the appearance of peneplains is often simulated by the out- crop of flat beds of hard rock which may lie at various altitudes and represent no period of reduction. In the Great valley also the peneplains are overshadowed and masked by the great differences of the rocks in point of resistance to wear. When once the criteria are established, however, it needs no extended search to discover successive forms of degradation as distinct as in the Atlantic streams and grouped in the same manner. Difficulties in the paths of the streams are more localized, however, by the differences in the rocks, and need consideration in coordinating the forms. VARIATIONS OF LEVEL During any extended period of degradation minor difficulties of erosion would be reduced, but the major barriers, such as are produced by unusual groups of rocks and affect entire river basins, can retard reduction so seriously as to produce considerable discordance of elevation in different basins. An excellent instance of this is furnished by the Pigeon and French Broad rivers in North Carolina. These streams flow in concentric curves, and the larger or French Broad meets the least obstruction ; 1t has accordingly reduced its peneplains and valleys which lie above the barrier 400 feet lower than the corresponding ones of the Pigeon. Re- sults in the same direction would ensue from warping of the surface, which would give added slope and power to one stream and decrease the grade of an opposite flowing stream. ‘This would retard final reduction merely by the added amount of material to be removed, but would leave no permanent differences of altitude in similar forms. An unfailing factor in producing differences of altitude at like periods of reduction is the distance from the sea. A peneplain produced by a stream 500 miles from its mouth will be higher than one produced only 100 miles from the sea, whether by the same or a different stream and whether in the same or a different region. A certain amount of fall, however slight, is necessary to make the rivers flow and will be expressed in the accumulated altitudes. On this account the Yadkin river, flowing direct to the Atlantic, has formed its peneplain at 2,300 to 2,500 feet close up to the main residual of the Blue Ridge, while the Nolichucky, 30 miles to the west, taking a longer course through the Tennessee, has cut its peneplain along its main valley only down to 2,600, or 200 feet higher. Most potent of all factors of variation is the amount and nature of debris fur- nished to the streams. Insoluble rocks clog the streams with debris and large re- siduals furnish great quantities of waste, so that, as residuals in general are composed of the more obdurate strata, the two unite to raise the grades and perpetuate divides. Soluble rocks yield debris which moves with a minimum grade, so that they seldom remain as residuals. Accordingly grades over soluble rocks are low entirely to the divides, which are thus at the mercy of local variations in the rocks and in the stream powers, and which are correspondingly unstable. . It is to be expected, therefore, that widely dissimilar basins will have peneplains formed at the same time but at somewhat different altitudes. Such expectation is amply borne out by the facts of the field, and is in fact exceeded. The least in- spection of peneplains shows differences of altitude amounting to 3,000 feet. Two explanations can be made of such great differences, either that one or two pene- plains have been warped out of their original plane or that many peneplains are 22 PROCEEDINGS OF PHILADELPHIA MEETING. Or represented which were produced at different periods and successively elevated with little warping. FEATURES OF THE TENNESSEE BASIN PENEPLAIN GROUPS Inspection of the Tennessee basin in the Great valley reveals well developed peneplains at four altitudes. The uppermost series appears in peneplains and plateaus near the heads of the main streams from 3,300 to 3,700 feet high, falling slightly away from the divides. As the streams increase in size these plains are more and more dissected by sharp gorges of later origin, until along the northwest front of the Unakas only the most insoluble rocks approximate the original height at 3,000 to 3,200 feet. In the valley of Tennessee some ridges of hard rock remain at this height, but are considerably dissected. On Cumberland plateau northwest of Knoxville a large area of knobs and level ridges on one of the main divides remains at 3,100 to 3,200 feet. As the draining streams grow larger this ancient plain is dissected and supplanted by another and lower plain. This second great group of forms is found at altitudes of 2,000 to 2,600 feet. It begins as a series of peneplains along the upper Holston at 2,300 to 2,500 feet. Behind the barrier of the Unakas the tributaries of the Tennessee, the Nolichucky, French Broad and Pigeon rivers, have cut out broad peneplains from 2,300 to 2,700 feet. These are very well preserved, and in every case they show a slight but steady rise upstream, whatever the direction of flow. As before stated, the Pigeon pene- plain is uniformly higher than the adjacent French Broad plain. Passing down the Great valley the second peneplain is much dissected, and appears only in sand- stone ridges in the valley or in peneplain remnants along the foothills of the Unaka mountains at elevations ranging from 2,100 to 2,400 feet. The Clinch and Bays mountains and the Cumberland front, especially the former, are fine examples of baseleveled ridges. Inthe lower Tennessee valley all traces of this peneplain have been removed. The third group of surface forms attains prominence in the Tennessee valley after the confluence of Watauga river and the north and south forks of Holston river at altitudes of 1,600 to 1,800 feet. Above the junction it is only manifest in broad flood-plains, bottoms and similar features fingering between the remnants of the peneplain last described. For 50 or 60 miles southwest down the valley this altitude of 1,600 to 1,800 feet is prominent in broad peneplains. These become more and more dissected, until only seattered ridges attain that height, and the country stands at 1,000 to 1,100 feet. Along either side of the Great valley many remnants of this peneplain appear; on the Hiawassee drainage through the Unakas it is finely developed at 1,700 to 1,800 feet, and on the opposite side of the valley Waldens ridge and Cumberland plateau exhibit broad areas at 1,500 to 1,700 feet. The last series appearing in the Tennessee valley becomes predominant after the union of Nolichucky, French Broad and Holston rivers above Knoxville. Broad bottoms and gravel-covered terraces and valleys mark the emergence of the rivers at 1,000 feet, their courses between that altitude and 1,600 to 1,800 feet being largely confined to narrow valleys and gorges. Below these points broad valleys appear, widen out into peneplains and soon occupy the entire valley at 1,000 to 1,100 feet, extending southwestward at that height for many scores of miles. In the course of still more recent erosion the streams have carved narrow canyons, which slowly SOME STAGES OF APPALACHIAN EROSION. O2e open out downstream and are bordered by terraces and bottoms from 600 to 700 feet above sea. Thus in the Tennessee valley are seen four distinct groups of peneplains and associated features, marking four periods of stable land and long degradation. The greatest of these is the first, because it extended to the headwaters of the main rivers, and only the most obdurate and remote masses escaped reduction. Each successive period was less important than the preceding as measured by the results accomplished. The forms of any minor period would have been obliterated, how- ever, by a greater subsequent one, so that the record can only be expected to pre- serve those which were in descending order of magnitude. At the present day the most conspicuous are the 1,600 to 1,800 and the 1,000 to 1, 100-foot peneplains, which occupy much of the Great valley, and, swinging around the south end of the Unaka mountains and the Blue ridge, pass northeast along the heads of the Atlantic basins. CLINCH SECTION The relations of Clinch mountain, the typical baseleveled ridge of Tennessee, furnish an epitome of the whole basin. Rising abruptly from the 1,000-foot pene- plain and flanked on both sides by ridges of the next peneplain at 1,600 to 1,700 feet, its summits stand at 2,100 to 2,200 feet; a few points rise to 2,500 feet and a few wind gaps are cut down to 2,000 feet. This average height of 2,200 feet is maintained for 100 miles northeastward to Moccasin gap, near the state boundary in Virginia, the flanking ridges continuing at uniform heights. Northeast of that gap the mountain rises within three miles to 3,200 feet, and its summits continue at that height for 30 miles to Little Moccasin gap. From this point northeastward the mountain is very irregular in height and loses its identity in a great mass, which is for the most part over 4,000 feet above the sea. In this group of ridges the 1,000-foot peneplain is perfectly obvious ; the same characteristics that are con- ceded to Clinch mountain at 2,200 feet are precisely repeated in the portion stand- ing at 3,200 feet and in the flanking ridges at 1,600 feet. Therefore the same rea- soning that identifies a baseleveled ridge at 2,200 feet must recognize the abruptness of the jump from level to level and must identify three baselevel periods instead of one. The linear profiles of the ridges are shown on the accompanying map. UNAKA-BLUE RIDGE SECTION A profile with similar features but less compact in form is taken along the head- waters of the Tennessee branches between the Blue Ridge and the Unaka moun- tains. It starts with a series of plateaus in Virginia and North Carolina, near the state boundary, at heights of 3,100 to 3,200 feet. These are considerably inter- rupted by the residuals between the Ohio streams and the Tennessee basin, but can readily be traced over into the Tennessee basin at heights of 3,300 to 3,500 feet. Into this surface are sunk the fingers of the lower system at 2.600 to 2,700 feet. Onthis part of the section the peneplains are much interrupted by residuals, and on the divides between the Watauga-Nolichucky and the Nolichucky-French Broad basins the upper peneplain again. appears. The second plain is well shown in the French Broad and Pigeon basins at 2,200 and 2,600 feet, and again on the Tuckaseegee at 2,300 and the Little Tennessee at 2,200 to 2,100 feet. The Nantahalah has barely produced a plain at the upper level of 3,500 to 3,600 feet, while into the edges of this the two Hiawassee peneplains are cut at 2,100 feet and at 1,800 to 1,900 feet, with bottoms and terraces sloping up the streams. These plains with small residuals continue in the Nottely and Toccoa basins, and the lower is carried 54 PROCEEDINGS OF PHILADELPHIA MEETING. on through the Etowah basin, descending to 1,700 and 1,600 feet. In the latter basin the 1,600-foot peneplain is shortly cut out by the 1,100 to 1,200-foot plain, which gradually descends to 1,100 and 1,000 feet with the fall of the river, con- tinuing at that height for nearly 100 miles. In this section the abundance of re- siduals has necessarily increased the grades of the streams and produced the sem- blance of warped surfaces, but the abruptness of the breaks from plain to plain and the direction of the slope away from the divides in most cases remove the possi- bility of warping and make it necessary to distinguish the plains as separate. Northwest of Knoxville, Tennessee, the four series show the well developed steps within a radius of eight miles, and in innumerable instances groups of three appear in close connection. In each group of forms the sequence from dissected peneplain through the peneplain into the broad valley, bottom and canyon is normal and complete, and the beginnings of each lower plain cut deep into the heart of the plain next above it. In its broad reaches each plain is remarkably constant in level, but in its narrow portions, where grades are raised by debris from neighbor- ing residuals, the slope upstream is invariably found, regardless of the direction of flow. Areas occur in which peneplains are indubitably warped, but they are readily recognized on the gronnd and are distinctly the exception. In short, ero- sion has produced in this basin at least four peneplains, each approximately level and each swinging around the neads of the lower plains in successive steps. SUMMARY Study of other river basins reveals similar series in every case, and without citing the countless details thus far known it is sufficient to state in brief that Appalachian degradation was marked by at least seven periods of approximate reduction. Each of these produced a vast series of peneplains which appear in various forms at the present day ; the oldest lie along the main divides and the youngest along the mar- gin of the sea. In some cases these plains have been warped from their original level, but far the greater portions of them retain nearly their original attitudes. It follows, therefore, that the disturbances which caused the revival of erosion were characterized by broad, uniform uplifts with local zones of warping quite subordi- nate in area. This is typified by the peneplain in Maryland and eastern Virginia, which ranges for 100 miles east and west and many times that distance to the south- west at an altitude of 500 to 550 feet, but which along the Staunton river in southern Virginia slopes up northwest 400 feet in 30 miles and‘then remains level for the next 30 miles. This view departs considerably from the theories of other authors, who have de- fined the peneplains as dome-shaped or warped surfaces, making the warping the predominant feature ofthe uplift. Further differences of view exist in the distinction of many peneplains instead of a few. It is agreed that the higher peneplains are associated with the main divides; that fact, however, is as well explained a priori by the usual sequence of erosion as by the theory of warping. The peneplains must have been deformed, otherwise the land never could have risen; but, on the other hand, the peneplains must rise with the streams and in regions of massive residuals with a considerable slope; therefore the differences of level must be studied in each group and may be referred to one or both causes. The normal peneplain sur- face is slightly sloping, and for each slope to which a different origin is ascribed proof must be furnished to account for the abnormal condition. The present con- fusion has been caused, in part at least, by correlating as parts of the same pene- SOME STAGES OF APPALACHIAN EROSION. 925 plain features which in the field can be traced past one another as parts of different peneplains. In other cases a peneplain actually slopes, but its slope is with the fall of the stream and in a direction contrary to that demanded by the extreme theory of warping. In short, the slopes of the peneplains are so slight in the great majority of cases as to distinguish warping as the exceptional form of uplift. Al- though the process was a simple one, the succession of uplifts was long and quite complex. A full understanding of the different stages demands the expenditure of much time and connected field work, and will be made the subject of future publications. The following paper was read by title: THE CERRILLOS COAL FIELD OF NEW MEXICO BY JOHN J. STEVENSON [ Abstract ] During August, 1895, the writer had an opportunity to revisit the Placer coal field of New Mexico, now known as the Cerrillos coal field. It is about 25 miles south from Santa Fé and directly beyond the Galisteo river. The field is small, appar- ently a detached portion of the Laramie area extending far southward within the Rio Grande region. The district of especial interest is that lying south from Cerrillos and Waldo, stations on the Santa Fé railroad. It is less than two miles wide. and reaches south- ward to little more than five miles from the Galisteo, but it contains evidently all of the workable coal beds, and exhibits the transition from bituminous to anthra- cite in a very satisfactory manner. The mines are all on Coal canyon, which ex- tends from the Placer or Ortiz mountains at the south to Waldo at the north, some- what more than six miles. The Ortiz mountains are largely trachytic; from them there extend northward two plates, each about 200 feet thick, which pass between Laramie strata and follow very closely the dip of the stratified beds. The upper plate covers the area east from Coal canyon, and is now the surface rock, the overlying beds having been removed. It extends northward to somewhat less than two miles south of Waldo, terminating opposite the lower end of the village of Madrid, where are the offices of the Cerrillos Coal Company. The lower plate, about 400 feet below the upper, does not come to the surface on Coal canyon, but it was reached in a boringon the mesa immediately west and comes out in an arroyo within a few rods west from the boring. Several dikes extend upward from this plate, one evidently very large being shown west from Coal canyon, which mus: have been connected with the upper plate, as it rises very high above the mesa; a second is seen in Coal canyon, not more than 10 or 12 feet wide; if, does not reach the upper plate ; a third, very narrow, found in the same canyon at a mile and a half above Madrid, passes dis- tinctly into the upper plate. Professor Kemp examined the specimens from several exposures and recognizes the close resemblance in composition throughout. The only stratified rocks within the district examined belong to the Laramie, and the exposed section is somewhat more than 1,000 feet thick. The rocks resemble those of the same age in the Trinidad coal field, but shale is present in greater pro- portion. Limestone is wholly absent, apparently, and the sandstones are unusually non-fossiliferous. The coal beds are numerous, but most of them are very thin and several are not persistent in all of the sections. LXIII—Butt, Grou. Soc. Am., Vou. 7, 1895, 526 PROCEEDINGS OF PHILADELPHIA MEETING. The only coal beds of interest here are those within the interval between the trachyte plates. They are White-ash coal bed: .. :... sto ee: 2 feet 6 inches to 7 feet. Tritenyal eters oo, Ne a ee i) rhe Cokine coal bed....>. «2. santa 1 foot to 2 feet 6 inches. Tbe oe ics Soe ee 80 feet. Cook-White coal bed. :....:...... BR Interval, about... ../. 2c hee ent LEO "* Waldo-coal ped.c2 .. ice anemones anne The White-ash bed is not more than 15 feet below the upper plate, and the Waldo bed as found in the bore-hole about 10 feet above the lower plate of trachyte. The White-ash bed has been mined at many pits along Coal canyon for a distance of nearly three miles, beginning at about a mile and a half from Waldo. It is the important bed of the region and the only one now mined. It was examined in four pits, two of which are now in operation. At the old Boyle mine, about a mile and a half above Madrid, the coal is a hard dry anthracite, varying much in char- acter. It is slipped and jointed throughout. Some portions closely resemble the graphitoid anthracite of Rhode Island. The Lucas mine at Madrid was idle when visited, but work had been stopped for only ashort time. The southerly levels of this mine yield an anthracite of ex- cellent quality, equal in appearance and composition to the average anthracite of Pennsylvania, but the northerly levels show arapid change. Jointing becomes an- noying at a little distance from the slope and the coal is wasted in the breaker. Within 350 feet evidences of great pressure and disturbance accumulate and the coal soon is laminated, like that from some Vespertine mines of southwest Virginia, with the polished surfaces, often curved, frequently not more than one-fourth of an inch apart. This, however, is still anthracite, and work was stopped in these northerly levels only because of great waste in breaking. The Cunningham mine, at the lower end of Madrid, entered a tender coal at the crop; the slope was pushed 1,100 feet, but no anthracite was found. The coal burns with flame. The White-ash mine, about half a mile north from the Lucas, is the important pit. At one time trains might be seen coming from its slope made up of cars car- rying, some of them, anthracite, others the tender, semi-bituminous, and others still the rich bituminous coal which has given this mine its reputation. The bitu- minous coal, containing 39 per cent of volatile combustible, is obtained from the northerly levels, but the southerly levels yield for the most part what is called tender coal. The latter is dull, very tender and much of it has an almost cone-in- cone structure. It is reached in the southerly levels at varying distances from the slope. The passage from bituminous into anthracite through this tender coal is shown in the sixth level, southerly, where the tender coal was reached at 125 feet from the slope and the anthracite at 450 feet. The passage is gradual. The an- thracite makes its appearance at the bottom and thickens gradually, crushed coal being replaced by laminated and that by the harder almost homogeneous coal, the change being completed within 50 feet. The Coking bed was worked some years ago at about two miles above Madrid, where its coal was coked in ricks. The Cook-White coal is no longer mined, but it has been opened at many places along Coal canyon, and the changes in character of the coal are clearly shown, CERRILLOS COAL FIELD OF NEW MEXICO. BK Above Madrid fragments on the old dumps show that the coal is anthracite. A pit at the lower end of Madrid, almost midway between the Cunningham and White-ash mines, shows a tender coal which resembles that from Pocahontas, in Virginia. Analysis shows that it contains about 30 per cent of volatile, which is about what should be expected if its changes are similar to those of the White-ash. The Waldo bed is not reached in the upper part of Coal canyon, but it has been mined extensively further down. The only interest it has here is its existence in the bore-hole west from Coal canyon, where it is not more than 10 feet above the lower plate of trachyte, and shows no evidence of any metamorphism whatever. Long ago Newberry and afterward Stevenson regarded the coal as metamorphosed hy heat from a great dike of eruptive rock following the northerly side of the Placer (now Ortiz) mountains. This, which then was but a suggestion, is sufficiently clear as an explanation now. As the center of eruption was in the Ortiz moun- tains, the metamorphism should be most notable near those mountains. That is distinctly the condition, for at the most southerly point showing the White-ash bed well the anthracite is very hard, but the change is less and less toward the north until normal coal is reached in the White-ash mine below Madrid. The gradation is equally clear in the Cook-White bed, but the small bed between the main seams appears to contradict the hypothesis, as it is decidedly bituminous at half a mile above the pit where the White-ash bed vields the hardest anthracite observed. This condition is easily explained by the fact that the small bed is broken by clay seams several feet wide, which sometimes cut out all of the coal ; these seams would prevent the passage of heat from one portion to another. The conditions at several localities show that mere proximity to the mass of eruptive rock was insufficient to produce change. The lower plate of trachyte is but 10 feet below the Waldo coal bed in the bore-hole west from Coal canyon, but, though 200 feet thick, it had no appreciable effect upon the coal. The interval between the White-ash bed and the upper plate of trachyte shows insignificant variations along Coal canyon, and it must be approximately the same in the newer parts of the White-ash mine, yet in the Lucas mineand at all localities south from it the coal is anthracite, whereas at all points north from it to the border of the eruptive rock one finds only transition coal. It seems clear that direct contact is necessary to produce change. Professor J. F. Kemp describes the eruptive rock as a trachyte closely allied to andesite. Its outflow then was early, possibly at the time of the Laramide eleva- tion, when great outpourings of andesite occurred in Colorado, Utah, Wyoming and Montana. The coal was completely formed prior to this elevation, prior to any disturbance, there being not only no evidence of pulpiness, but every evidence that the coal was thoroughly hard. It was crushed into minute fragments, slickensided like the Utica shales of Franklin county, Pennsylvania, or laminated and rolled into leaves like the Vespertine coals of southwestern Virginia. The process of con- version was complete before disturbance, not merely in the lowest beds, but also in the White-ash bed at nearly 900 feet above the bottom of the Laramie. The scientific program was declared finished. Vice-President Charles H. Hitchcock offered the following resolution, which was unanimously adopted : ‘““Resolved, That the sincere thanks of the Geological Society of America are hereby tendered to the officers of the University of Pennsylvania for their kindness 528 PROCEEDINGS OF PHILADELPHIA MERTING. and courtesy, particularly in granting the use of rooms in the Department of Arts building and for the midday luncheon; also to the ‘ Local Committee’ for their efforts to make the meeting a success.”’ With a few appropriate remarks, the President declared the meeting adjourned. REGISTER OF THE PHILADELPHIA MEETING, 1895 The following Fellows were in attendance at the meeting: FLORENCE Bascom. Rospert BEL. A. S. BickMORE. H. D. CAMPBELL. M. R. CAMPBELL. W. B. CLARK. EK. D. Cops. WHITMAN Cross. N. H. Darton. W. M. Davis. J) Oo. ne E. V. D’INVILLIERS. E. T. DUMBLE. B. K. EMERSON. S. F. Emmons. H. L. FAarrcuixp. PERSIFOR FRAZER. G. K. GILBERT. Hi; Pe Gunurvar: C. W. HAYEs. ANGELO HEILPRIN. C. H. Hircucock. JED. HorcHkKiIss. EK. O. Hovey. L. 1 -Hueearp, J. P. IppInes. R, T. JACKSON: ARTHUR KEITH. J. F. Kemp. Hy taba H. B. KUMMEL. Total attendance, 61. C. R. Keyes: A. C. LANE. D. W. Lanepon, JR. FRANK LEVERETT. By, Meee, EpwArp ORTON, L. V. Presson. H. F. Rem. W. N. Rice. Hernricu Ries. R. D. SAisBury. CHARLES SCHUCHERT. W. B. Scort. N. S. SHALER. C. H. Smyru, Jr. J. STANLEY-BROWN. T. W. STANTON. J. J. STEVENSON. H Oe Sas MN 8 R. S. Tarr. C. R. Van Hise. M. E. Wapsworta. W. H. WEED. I. C. Warre. H. S. WILLiAMs. BaItLey WILLIs. J. E. Wo.irr. G. F. WricHt. Fellows-elect Re fB Seay Tor: J. B. WoopworruH. OFFICERS AND FELLOWS OF THE GEOLOGICAL SOCIETY OF AMERICA OFFICERS FOR 1896 President JosEPH Lr Contr, Berkeley, Cal. Vice-Presidents C. H. Hircucock, Hanover, N. H. EpwArpD Orton, Columbus, Ohio. Secretary H. L. Fatrcuiip, Rochester, N. Y. Treasurer I. C. WuitE, Morgantown, W. Va. Editor J. STANLEY-Brown, Washington, D. C. Councillors (Term expires 1896) F. D. ApAms, Montreal, Canada. I. C. Russerz, Ann Arbor, Mich, (Term expires 1897) R. W. Ets, Ottawa, Canada. C. R. VAN Hise, Madison, Wis. (Term expires 1898) B. K. Emerson, Amherst, Mass. ' J. M. Sarrorp, Nashville, Tenn. (529) 530 PROCEEDINGS OF PHILADELPHIA MEETING. FELLOWS, APRIL, 1896 * Indicates Original Fellow (see article III of Constitution) . Frank Dawson Apams, Ph. D., Montreal, Canada; Professor of Geology in McGill University. December, 1889. TruMAN H. Aupricn, M. E., Birmingham, Ala. May, 1889. Henry M. Amt, A. M., Geological Survey Office, Ottawa, Canada; Assistant Paleon- tologist on Geological and Natural History Survey of Canada. December, 1889. Harry Foster Barn, M.S., Des Moines, Iowa; Assistant Geologist, lowa Geological Survey. December, 1895. S. Prentiss Batpwin, Cleveland, Ohio. August, 1895. ALFRED E. Bartow, M. A., Geological Survey Office, Ottawa, Canada; Geologist on Canadian Geological Survey. August, 1892. GeEoRGE H. Barron, B. 8., Boston, Mass.; Instructor in Geology in Massachusetts Institute of Technology. August, 1890. FLORENCE bascom, A. M., B.S., Ph. D., Bryn Mawr, Penn.; Instructor in Geology, Petrography and Mineralogy in Bryn Mawr College. August, 1894. Wituiam S. Baytey, Ph. D., Waterville, Maine; Professor of Geology in Colby University. December, 1888. * GrorGE F. Becker, Ph. D., Washington, D. C.; U. 8. Geological Survey. Crarues E. Bercuer, Ph. D., Yale University, New Haven, Conn. May, 1889. Ropert Bett, C. E., M. D., LL. D., Ottawa, Canada; Assistant Director of the Geological and Natural History Survey of Canada. May, 1889. AvBert S. Brckmore, Ph. D., American Museum of Natural History, 77th St. and Eighth Ave., N. Y. city; Curator of Anthropology in the American Museum of Natural History. December, 1889. Wituram P. BiaKke, Tucson, Ariz.; Professor of Geology, Metallurgy and Mining in University of Arizona. August, 1891. * Jonn C. Branner, Ph. D., Stanford University, Cal.; Professor of Geology in Leland Stanford Jr. University. Abert Perry Bricuam, A. B., A. M., Hamilton, N. Y.; Professor of Geology and Natural History, Colgate University. December, 1893. * GARLAND C. BroapHEAD, Columbia, Mo.; Professor of Geology in the University of Missouri. Henry P. H. Brumecy, Ottawa, Canada; Manager of N. A. Graphite and Mining Company. August, 1892. * SamMuEL Cavin, Iowa City, lowa; Professor of Geology and Zoology in the State University of Iowa, ‘State Geologist. Henry Donatp CAMPBELL, Ph. D., Lexington, Va.; Professor of Geology and Biology in Washington and Lee University. May, 1889. Marius R. CamMpsBE.., U. 8. Geological Survey, Washington, D.C. August, 1892. FRANKLIN R. Carpenter, Ph. D., Deadwood, South Dakota; Superintendent Dead- wood and Delaware Smelting Company. May, 1889. LIST OF FELLOWS. 5381 Rosert Cuatmers, Geological Survey Office, Ottawa, Canada; Geologist on Geo- logical and Natural History Survey of Canada. May, 1889. *T. CO. Cuamperiin, LL. D., Chicago, Ill.; Head Professor of Geology, University of Chicago. CLARENCE Raymonp CraGcHorn, B.S8., M. E., Vintondale, Pa. August, 1891. * WittiAM B. Ciark, Ph. D., Baltimore, Md.; Professor of Geology in Johns Hopkins University. * Epwarpb W. Cuaypotg, D.Sc., Akron, O.; Professar of Natural Science in Buchtel College. Juxius M. Ciements, B. A., Ph. D., Madison, Wis. ; Assistant Professor of Geology in University of Wisconsin. December, 1894. CouLier Coss, A. B., A. M., Chapel Hill, N. C.; Professor of Geology in University of North Carolina. December, 1894. * THropore B. Comstock, Tucson, Ariz.; President of the University of Arizona. * Epwarp D. Cops, Ph. D., 2102 Pine St., Philadelphia, Pa.; Professor of Geology in the University of Pennsylvania. *Francis W. Cracin, B. S., Colorado Springs, Col.; Professor of Geology and Natural History in Colorado College. * ALBERT R. CrRaNDALL, A. M., Milton, Wis. * WituiAM O. Crossy, B. S., Boston Society of Natural History, Boston, Mass. ; Assistant Professor of Mineralogy and Lithology in Massachusetts Institute of Technology. Wurman Cross, Ph. D., U. S. Geological Survey, Washington, D. C. May, 1889. Garry E. Cuniver, A. M., 1104 Wisconsin St., Stevens Point, Wis. December, 1891. * Henry P. Cusnina, M. S§., Cleveland, Ohio; Associate Professor of Geology, Adelbert College. ) T. Netson Date, Williamstown, Mass. ; Geologist, U. S. Geological Survey, In- structor in Geology, Williams College. December, 1890. * Netson H. Darron, United States Geological Survey, Washington, D. C. * WitiiAM M. Davis, Cambridge, Mass. ; Professor of Physical Geography in Har- vard University. ¢ GrEorGE M. Dawson, D. Sc., A. R.S. M., Geological Survey Office, Ottawa, Canada; Director of Geological and Natural History Survey of Canada. May, 1889. Sir J. Witttam Dawson, LL. D., Montreal, Canada. May, 1889. Davin T. Day, A. B., Ph. D., U.S. Geological Survey, Washington, D.C. August, 1891. . OrvitLeE A. Dersy, M.S., Sao Paulo, Brazil; Director of the Geographical and Geological Survey of the Province of Sao Paulo, Brazil. December, 1890. *JosepH S. Dinier, B. S., United States Geological Survey, Washington, D. C. Epwarp V. p’Ixvinuers, E. M., 711 Walnut St., Philadelphia, Pa. December, 1888. *Kpwin T. Dumsie, Austin, Texas, State Geologist. CLARENCE E. Dutron, Major, U. S. A., Ordnance Department, San Antonio, Texas. August, 1891. *Wintram B. Dwicur, M. A., Ph. B., Poughkeepsie, N. Y.; Professor of Natural History in Vassar College. Cuartes Rk. Easrman, A. M., Ph. D., Cambridge, Mass. ; Assistant in Paleontology in Harvard University. December, 1895. 532 PROCEEDINGS OF PHILADELPHIA MEETING. *GrorGe H. Evpripasr, A. B., United States Geological Survey, Washington, D. C. Rozert W. Exts, LL. D., Geological Survey Office, Ottawa, Canada; Geologist on Geological and Natural History Survey of Canada. December, 1888. * BengaMiIn K. Emerson, Ph. D., Amherst, Mass. ; Professor in Amherst College. *SamuEL F. Emmons, A. M., E. M., U.S. Geological Survey, Washington, D. C. JOHN Everman, F. Z. S., Oakhurst, Easton, Pa. August, 1891. Haroutp W. Farrpanks, B. S., Berkeley, Cal.; Geologist State Mining Bureau. August, 1892. 2 * Herman L. Farrcuip, B. 8., Rochester, N. Y.; Professor of Geology and Natural History in University of Rochester. J. C. Faves, Danville, Kentucky; Professor in Centre College. December, 1888. Eucenrt Rupotper Faripaurtr, C. E., Geological Survey Office, Ottawa, Canada; Geologist on Geological and Natural History Survey of Canada. August, 1891. P. J. Farnswortu, M. D., Clinton, Iowa; Professor in the State University of Iowa. May, 1889. Outver C. Farrtneron, Ph. D., Chicago, Ill.; In charge of Department of Geology, Field Columbian Museum. December, 1895. Sanprorp Fiemine, LL. D., Ottawa, Canada; Civil Engineer. August, 1893. WitiramM M. Fonrarne, A. M., University of Virginia, Va.; Professor of Natural History and Geology in University of Virginia. December, 1888. * Persiror Frazer, D. Sc., 1042 Drexel Building, Philadelphia, Pa.; Professor of Chemistry in Franklin Institute. * Homer T. Fuuier, Ph. D., Springfield, Mo.; President of Drury College. Henry Gannett, S. B., A. Met. B., U. 8. Geological Survey, Washington, D. C. December, 1891. * Grove K. Gitpert, A. M., United States Geological Survey, Washington, D. C. ApAM Capen Git, A. B., Ph. D., Ithaca, N. Y.; Assistant Professor of Mineralogy and Petrography in Cornell University. December, 1888. N. J. Giroux, C. E., Geological Survey Office, Ottawa, Canada; Geologist on Geo- logical and Natural History Survey of Canada. May, 1889. Cuarves H. Gorpon, M.8., Beloit, Wis. August, 1893. Untysses SHERMAN Gran’, Ph. D., Minneapolis, Minn.; Assistant on Geological Survey of Minnesota. December, 1890. WILLIAM StuKELEY Gres_ery, Erie, Pa.; Mining Engineer. December, 1893. GEORGE P. Grims_ey, M. A., Ph. D., Topeka, Kan.; Professor of Geology in Wash- burn College. August, 1895. Leon S. Griswotp, A. B., 238 Boston St., Dorchester, Mass. August, 1892. Freperick P. Guiiiver, A. M., Cambridge, Mass. August, 1895. *WitiiaAmM F. E. Gurury, Springfield, Ill.; State Geologist. ARNOLD Hagur, Ph. B., U. 8. Geological Survey, Washington, D.C. May, 1889. *Curistropuer W. Haut, A. M., 803 University Ave., Minneapolis, Minn.; Pro- fessor of Geology and Mineralogy in University of Minnesota. * James Haut, LL. D., State Hall, Albany, N. Y.; State Geologist and Director of the State Museum. Henry G. Hanks, 1124 Greenwich St., San Francisco, Cal.; lately State Mineralo- gist. December, 1888. Joun B. Hastrnas, M. E., Boisé City, Idaho. May, 1889. * Jonn B. Hartcuer, Ph. B., Princeton, N. J.; Assistant in Geology, College of New Jersey. August, 1895. * ErAsMus Haworrn, Ph. D., Lawrence, Kan. LIST OF FELLOWS. 533 C. WiLLARD Hayes, Ph. D., U. S. Geological Survey, Washington, D.C. May, 1889. *ANGELO HeEtLprin, Academy of Natural Sciences, Philadelphia, Pa.; Professor of Paleontology in the Academy of Natural Sciences. * KuGENE W. Hitearp, Ph. D., LL. D., Berkeley, Cal.; Professor of Agriculture in University of California. Frank A. Hitut, Roanoke, Va. May, 1889. * Ropert T. Hit, B. S., U. 8. Geological Survey, Washington, D. C. RicHarpD C. Hius, Mining Engineer, Denver, Colo. August, 1894. * CHARLES H. Hircucocr, Ph. D., Hanover, N. H.; Professor of Geology in Dart- mouth College. WittiaAmM Herspert Hopspss, B. Sc., Ph. D., Madison, Wis.; Assistant Professor of Mineralogy in the University of Wisconsin. August, 1891. * Levi Horsroor, A. M., P. O. Box 536, New York city. ArrHur Houricx, Ph. B., Columbia College, New York ; Instructor in Paleontology. August, 1893. * JoserpH A. Hoimes, Chapel Hill, North Carolina; State Geologist and Professor of Geology in University of North Carolina. Mary E. Hormrs, Ph. D., 201 S. First St., Rockford, Illinois. May, 1889. THomas ©. Hopkins, A. M., State College, Center county, Penn. December, 1894. * JEDEDIAH Horcuxiss, 346 E. Beverly St., Staunton, Virginia. * EpMUND Otis Hovey, Ph. D., American Museum of Natural History, New York city, Assistant Curator of Geology. * Horace C. Hovey, D. D., Newburyport, Mass. * Epwin E. Howe, A. M., 612 17th St. N. W., Washington, D. C. Lucius L. Hupparp, A. B., LL. B., Ph. D., Houghton, Mich.; State Geologist of Michigan. December, 1894. *ArtpHeus Hyatt, B.8., Bost. Soc. of Nat. Hist., Boston, Mass.; Curator of Boston Society of Natural History. JosepH P. Ippines, Ph. B., Professor of Petrographic Geology, University of Chi- cago, Chicago, Ill. May, 1889. Exrric D. Inca, Geological Survey Office, Ottawa, Canada; in charge of Mineral Statistics and Mines. August, 1894. A. WreNDELL Jackson, Ph. B., 407 St. Nicholas Ave., New York city. December, 1888. Rosert T. Jackson, 8. B., 8. D., 33 Gloucester St., Boston, Mass.; Instructor in Paleontology in Harvard University. August, 1894. Tuomas M. Jackson, C. E., 8. D., Clarksburg, W. Va. May, 1889. * JosepuH F. JAmMes, M. 8., Department of Agriculture, Washington, D. C. * Wrinttarp D. Jonnson, United States Geological Survey, Washington, D. C. Avexis A. Jutien, Ph. D., Columbia College, New York city; Instructor in Co- lumbia College. May, 1889. Artruur Kertu, A. M., U. 8. Geological Survey, Washington, D.C. May, 1889. * James F. Kemp, A. B., E. M., Columbia College, New York city; Professor of Geology. Cuartes Roriin Keyes, A. M., Ph. D., Jefferson City, Missouri; State Geologist. August, 1890. Frank H. Knowuton, M. S., Washington, D. C.; Assistant Paleontologist U. S. Geological Survey. May, 1889. LXIV—Buttu. Grou, Soc. AM., Vou. 7, 1895, 534 PROCEEDINGS OF PHILADELPHIA MEETING. Henry B. Ktimmer, A. M., Ph. D., Trenton, New Jersey; Assistant on the State Geological Survey of New Jersey. December, 1895. * GeorGE F. Kunz, care of Tiffany & Co., 15 Union Square, New York. Ravpu D. Lacog, Pittston, Pa. December, 1889. GerorGE Epa@ar Lapp, A. B., A. M., 81 Oxford St., Cambridge, Mass: August, 1891. J.C. K. Lartamnoe, M. A., D. D., Quebec, Canada; Professor of Mineralogy and Geology in University eet Quem August, 1890. Lawrence M. Lampe, Ottawa, Ganaaee Artist and Assistant in Palohaialeee Geo- logical Survey of Canada. August, 1890. Atrrep C. Lang, Ph. D., Houghton, Mich.; Assistant on Geological Survey of Michigan. December, 1889. DanieL W. Lanapon, Jr., A. B., 6 Wall St., New York city; Geologist of Chesa- peake and Ohio Railroad Company. December, 1889. AnprEwW C. Lawson, Ph. D., Berkeley, Cal.; Assistant Professor of Geology in the University of California. May, 1889. * JosepH Lr Conte, M. D., LL. D., Berkeley, Cal.; Professor of Geology in the University of California. * J, Perer Lestry, LL. D., 1008 Clinton St., Philadelphia, Pa.; State Geologist. FrANK LEVERETY, B. S., Tee Iowa.; eee U.S. Geyienes Survey. Au- gust, 1890. WALDEMAR LinpGREN, U. 8. Geological Survey, Washington, D. C. August, 1890. Roperr H. LouGHrinGe, Ph. D., Berkeley, Cal.: Assistant Professor of Agricultural Chemistry in University of California. May, 1889. Apert P. Low, B. S., Geological Survey Office, Ottawa, Canada; Geologist on Canadian Geslapthal Survey. August, 1892. Tuomas H. Macsripe, Iowa City, lowa; Professor of Botany in the State University of Iowa. May, 1889. Henry McCatury, A. M., C. E.. University, Tuscaloosa county, Ala.; Assistant on Geological Survey of Alabama. May, 1889. Ricaarp G. McConneti, A. B., Geological Survey Office, Ottawa, Canada; Geolo- gist on Geological and Natural History Survey of Canada. May, 1889. James Rreman MAcFArRLANE, A. B., Pittsburg, Pa. August, 1891. *W J McGer, Washington, D. C.; Bureau of North American Ethnology. Wiustam McInngs, A. B., Geological Survey Office, Ottawa, Canada; Geologist, Geological and Natural History Survey of Canada. May, 1889. Perer McKe var, Fort William, Ontario, Canada. August, 1890. Oxtver Marcy, LL. D., Evanston, Cook Co., Ill.; Professor of Natural History in Northwestern University. May, 1889. Oruniet C. Marsu, Ph. D., LL. D., New Haven, Conn.; Professor of Paleontology in Yale University. May, 1889. Vernon F. Marsters, A. B., Bloomington, Ind.; Associate Professor of Geology in Indiana State University. August, 1892. Epwarp B. Matuews, Ph. D., Baltimore, Md.; Instructor in Petrography in Johns Hopkins University. August, 1895. ~P. H. Ment, M. E., Ph. D., Auburn, Ala.; Professor of Geology and Natural History in the State Polytechnic Institute. December, 1888. * Joun C. Merriam, Ph. D., Berkeley, Cal.; Instructor in Paleontology in University of California. August, 1895. LIST OF FELLOWS. 535 * Freperick J. H. Merrinu, Ph. D., State Museum, Albany, N. Y.; Assistant State Geologist and Assistant Director of State Museum. Grorce P. Merritt, M. S., U. S. National Museum, Washington, D. C.; Curator of Department of Lithology and Physical Geology. December, 1888. James KE. Mirus, B. S., Quincy, Plumas Co., Cal., December, 1888. THomas F. Moszs, M. D., Urbana, Ohio. May, 1889. *Frank L. Nason, A. B., 5 Union St., New Brunswick, N. J.; Assistant on Geo- logical Survey of New Jersey. *Prrer Nerr, A. M., 361 Russell Ave., Cleveland, Ohio; Librarian, Western Re- serve Historical Society. FREDERICK H. Nrewe t, B. S., U.S. Geological Survey, Washington, D.C. May, 1889. WiiAm H. Niuzs, Ph. B., M. A., Cambridge, Mass. August, 1891. Wituiam H. Norron, M. A., Mt. Vernon, Iowa; Professor of Geology in Cornell College. December, 1895. . Caries J. Norwoop, Frankfort, Ky.; State Mine Inspector of Kentucky. August, 1894. * EDWARD Orton, Ph. D., LL. D., Columbus, Ohio; State Geologist and Professor of Geology in the State University. * Amos O. OsBporn, Waterville, Oneida Co., N. Y. CHARLES Patacur, B. S., Mineralogical Laboratory, Harvard University, Cam- bridge, Mass. August, 1894. * Horace B. Parron, Ph. D., Golden, Col.; Professor of Geology and Mineralogy in Colorado School of Mines. RicHarp A. F. Penross, Jr., Ph. D., 1331 Spruce St., Philadelphia, Pa. May, 1889. JosEPH H. Perry, 176 Highland St., Worcester, Mass. December, 1888. *WintiaAM H. Perrer, A. M., Ann Arbor, Mich.; Professor of Mineralogy, Eco- nomical Geology and Mining Engineering in Michigan University. Louis V. Prrssox, Ph. B., New Haven, Conn.; Assistant Professor of Inorganic Geology, Sheffield Scientific School. August, 1894. * FrRankiin Piatt, 1617 Chestnut St., Philadelphia, Pa. * Juttus Pontman, M. D., University of Buffalo, Buffalo, N. Y. Wii B. Porrer, A. M., E. M., St. Louis, Mo.; Professor of Mining and Metal- lurgy in Washington University. August, 1890. *Joun W. Powstt, Bureau of Ethnology, Washington, D. C. * CHARLES S. Prosser, M. S., Schenectady, N. Y.; Professor of Geology in Union College. * RAPHAEL Pumpetty, U. 8. Geological Survey, Newport, R. I. Freperick L. Ransome, B. §., Berkeley, Cal. August, 1895. Harry Frevprne Rei, Ph. D., Johns Hopkins University, Baltimore, Md. De- cember, 1892. Wittram Norra Rice, A. M., Ph. D., LL. D., Middletown, Conn.; Professor of Geology in Wesleyan University. August, 1890. Heryricu Ries, Ph. B., Fellow in Mineralogy, Columbia College, New York city. December, 1893. ; Craries W. Rorrs, M. 8., Urbana, Champaign Co., Ill.; Professor of Geology in University of Illinois. May, 1889. 536 PROCEEDINGS OF PHILADELPHIA MEETING. *TsragEL ©. Russevu, M. §., Ann Arbor, Mich.; Professor of Geology in University of Michigan. * James M. Sarrorp, M. D., LL. D., Nashville, Tenn.; State Geologist; Professor in Vanderbilt University. Orestes H. St. Joun, Topeka, Kan. May, 1889. * Routiin D. Santspury, A. M., Chicago, Ill.; Professor of General and Geographic Geology in University of Chicago. Freperick W. Sarpeson, University of Minnesota, Minneapolis, Minn. Decem- ber, 1892. * CHARLES SCHAEFFER, M. D., 1309 Arch St., Philadelphia, Pa. . CuHarueEs ScaucHert, Washington, D. C.; Assistant Curator in Paleontology, U.S. National Museum. August, 1895. WiiuraAM B. Scott, M. A., Ph. D.,56 Bayard Ave., Princeton, N. J.; Blair Professor of Geology in College of New Jersey. August, 1892. Henry M. Seery, M. D., Middlebury, Vt.; Professor of Geology in Middlebury College. May, 1889. Aurrep R. C. Setwyn, C. M. G., LL. D., Ottawa, Canada. December, 1889. * NaTHANIEL S. SHaver, LL. D., Cambridge, Mass.; Professor of Geology in Har- vard University. Wii H. SHerzer, M.S., Ypsilanti, Mich.; Professor in State Normal School. De- cember, 1890. _* Freperick W. Srwonps, Ph. D., Austin, Texas; Professor of Geology in Univer- sity of Texas. * EuGene A. Smuiru, Ph. D., University, Tuscaloosa Co., Ala.; State Geologist and Professor of Chemistry and Geology in University of Alabama. JaMeES Perrin Smiru, M. S., Ph. D., Palo Alto, California; Assistant Professor of Paleontology, Leland Stanford Jr. University. December, 1893. * Joun C. Smock, Ph. D., Trenton, N..J.; State Geologist. Cuarues H. Smyrn, Jr., Ph. D., Clinton, N. Y.; Professor of Geology in Hamilton College. August, 1892. . Henry L. Smyru, A. B., Cambridge, Mass.; Instructor in Mining Geology in Harvard University. August, 1894. * J. W. Spencer, A. M., Ph. D., 1820 Corcoran St., Washington, D. C. JostAH KE. Spurr, A. B., A. M., Gloucester, Mass. December, 1894. JosEPH STANLEY-Brown, 1318 Massachusetts Ave., Washington, D.C. August, 1892. Timotay WruttaAm Sranron, B.S., U. S. Geological Survey, Washington, D. C.; Assistant Paleontologist U. S. Geological Survey. August, 1891. * Joun J. Srevenson, Ph. D., LL. D., University of the City of New York; Pro- fessor of Geology in the University of the City of New York. JoseeH A. Tarr, B.S., Washington, D.C.; Assistant Geologist U. 8. Geological Survey. August, 1895. Ravpg §. Tarr, Cornell University, Ithaca, N. Y.; Assistant Professor of Geology. August, 1890. Frank B. Taytor, Fort Wayne, Ind. December, 1895. * Asa Scorr Tirrany, 901 West Fifth St., Davenport, Iowa. * James E. Topp, A. M., Vermillion, S#Dak.; Professor of Geology and Mineralogy in University of South Dakota. * Henry W. Turner, B. S., U. S. Geological Survey, Washington, D. C. LIST OF FELLOWS. yon JosEPpH B. Tyrresi, M. A., B. Sc., Geological Survey Office, Ottawa, Canada; Geologist on the Canadian Geological Survey. May, 1889. * Epwarp O. Uric, A. M., Newport, Ky.; Paleontologist of the Geological Survey of Minnesota. *WarreN Upnam, A. M., Librarian Minnesota Historical Society, St. Paul, Minn. *CHARLES R. Van Hisz, M. S., Madison, Wis.; Professor of Mineralogy and Petrography in Wisconsin University ; Geologist U. S. Geological Survey. * AnrHony W. Voeprs, Alcatraz Island, San Francisco, Cal.; Captain Fifth Artil- liginyy, Wash ainahig *MarsHmMan E. Wapswortu, Ph. D., Houghton, Mich.; State Geologist ; Director of Michigan Mining School. * CHARLES D. Watcory, U.S. National Museum, Washington, D. C.; Director U.S. Geological Survey. Watrer H. Weep, M. E., U.S. Geological Survey, Washington, D.C. May, 1889. Lewis G. Wesrcarr, 1303 Chicago Ave., Evanston, Ill. August, 1894. Tomas C. Weston, Ottawa, Canada. August, 1893. Davip Wuirs, U.S. National Museum, Washington, D. C.; Assistant Paleontolo- gist, U. S. Geological Survey, Washington, D. C. May, 1889. *TsrarL C. Wuirsz, Ph. D., Morgantown, W. Va. * CHARLES A. Wuits, M. D., U. S. National Museum, Washington, D. C.; Paleon- tologist U. 8. Geological Survey. JOSEPH FREDERICK WHITEAVES, Ottawa, Canada; Paleontologist and Assistant Di- rector Geological Survey of Canada. December, 1892. * Ropert P. Warrrietp, Ph. D., American Museum of Natural History, 77th St. and Eighth Ave., New York city ; Curator of Geology and Paleontology. * Kpwarp H. Wiriams, Jr., A..C., E. M., 117 Church St., Bethlehem, Pa.; Pro- fessor of Mining Engineering and Geology in Lehigh University. * Henry S. WitiraMs, Ph. D., New Haven, Conn.; Professor of Geology and Paleon- tology in Yale University. Battey Wiis, U. 8. Geological Survey, Washington, D.C. December, 1889. * HoRACE VAUGHN WINCHELL, 1306 S. E. 7th St., Minneapolis, Minn.; Assistant on Geological Survey of Minnesota. *Nrwton H. Wincneiit, A. M., Minneapolis, Minn.; State Geologist; Professor in University of Minnesota. *ArtHur Winstow, B. §., Roe Building, 5th and Pine streets, St. Louis, Mo. Joun EK. Worrr, Ph. D., Harvard University, Cambridge, Mass.; Assistant Pro- fessor in Petrography, Harvard University. December, 1889. Raosert Stimpson Woopwarp, C. E., Columbia College, New York city; Professor of Mechanics in Columbia College. May, 1889. Jay B. WoopwortH, B. S., Cambridge, Mass.; Instructor in Harvard University. December, 1895. Ausert A. Wricut, A. B., Ph. B., Oberlin, Ohio; Professor of Geology in Oberlin College. August, 1893. *G. FrepERIcK WricHt, D. D., Oberlin, Ohio; Professor in Oberlin Theological Seminary. Lorenzo G. Yatrs, M. D., Los Angeles, Cal. December, 1889. Wim §, Yuatss, A. B., A. M., Atlanta, Ga.; State Geologist of Georgia. August, 1894. 538 PROCEEDINGS OF PHILADELPHIA MEETING. FELLOWS DECEASED * Indicates Original Fellow (see article III of Constitution) * CHARLES A. ASHBURNER, M. S., C. E. Died December 24, 1889. Amos Bowman. Died June 18, 1894. * J. H. Cuapin, Ph. D. Died March 14, 1892. GerorcE H. Cook, Ph. D., LL. D. Died September 22, 1889. ANTONIO DEL CastinLo. Died October 28, 1895. * JamEs D. Dana, LL. D. Died April 14, 1895. *ArBerr E. Foote. Died October 10, 1895. * Ropert Hay. Died December 14, 1895. Davip Honeyman, D. C. L. Died October 17, 1889. Tuomas Srerry Hunt, D. Sc., LL. D. Died February, 1892. * Henry B. Nason, M. D., Ph. D., LL. D., Troy, N. Y. Died January 17, 1895. * Joun S. Newsperry, M. D., LL. D. Died December 7, 1892. * RicHarD Owen, LL. D. Died March 24, 1890. CHarLes WacumurH. Died February 7, 1896. * Grorce H. Wiiitams, Ph. D. Died July 12, 1894. * J. Francis Witurams, Ph. D. Died November 9, 1891. * ALEXANDER WINCHELL, LL. D. Died February 19, 1891. Summary Onipitial Kellows .. ..20% beens cence nea s vane ebay 87 Mlected: PEUGwWs. vo. 0h) ees athe heehee ae eee 141 NESTS TP: ¢.01~ anon ots tae eR Sp LE On 228 Doceased Fellows. 25.0. feces th wee ade ehGas os eal. ACCESSIONS TO LIBRARY FROM JANUARY, 1895, TO MARCH, 1896 BY H. L. FAIRCHILD, Secretary and Acting Librarian Contents’ Page (A) From societies and institutions receiving the Bulletin as donation (* Exchanges”’)........... 539 (@)). ATC VRT OE CLARA SB ERENCE SHORES BOLE RPS ae ene ea eta ne Seb ry ie TR An car | Mamet ie aC OU eT) feng oe 539 MOS MDESTUDICE) DC reek Aathea ct ats du ac dehs'- Sos vs of Aura deay Gas hcceh acca eres Lomas aoe aN Buea obs stra setae wanes ed voubinne duces Bumeewaeste 541 MENMMENGH leccans oo caa tai Nisin desed ome chances Seteutree closes aba eeea sea aac SAS aS Smear Roe tamed oc ok a So Sein veShiuuatas Tiros etal Sas cece meet 544 GL) AMUISITEY ERSTE aay are erecta Wire ns Wo Sie Gar mene gH aS OR 9. 1 0 Tied dak ne ae EO On 544 ReipEnomi state SHEcueel (Corba lOresc3s¢ ccagssedo opp ooocooeooeAsene 460 —; Geology of old Hampshire county, in INES SACHS CEES LAVA ict econ cho casecstenc soem tees 5 —, Reference to discussion of paper by 7, 11, 507, 509, 512 EMERTON, J. H., Acknowledgments to....... 136 —, Reference to drawings by ....... 174; 216, 246 Emmons, E., cited on inclusions in apatite 127 — — — rounded apatite crystals ....... Hhnatee sea 127 EMMONS, S. F., continued on Mount Ranier Forest Reserve commiittee........... ....s008 2 ENGLAND, Figures of fossils from the Car- [SORT RRESRONDIS Oyosocc qsoceudonecoscopadsdocosOocnce 252, 253 —, Reference to drumlins im................200-0000 27 == =} SLACIALION IY. 620 jinoessoncccesceecuh toneectes 28 —, Shore forms on coast Of..............0sceseeeeeses 415 EOCENE formations of the Coastal ies ae 517 — fossils from Cuba, Reference to............... 8 = history Of Cuba... ceticd decatesseschesdetessecees 75-81 ESCHWEGE, W. I. VON, cited on Brazilian DOMMES teas aticcwteos sees oseeceeceesetae neem 277, 278 ext OlatlOm Ol nOCkKSiy -e-seeeeen eee eee 292 ETHERIDGE, H. G., cited on denudation...., 383 ETHERIDGE, R., cited on relations of Olizo- WPOTGUS sadctiyae ss deo valdsaaneiesesstanaste sects casceee cs 189, He EUROPE, Reference to ice-sheets of.............. EXAMPLES of stream-robbing in the Cats- kill mountains ; N. H. Darton........ eh 505 FAILYER, G. H., cited on nitric acid in rain.. 307 FAIRCHILD, H, sla elected Secretary........... 460 ; Glacial Genesee lakes! ve wie on 423 _, ’ makes TAS] OXOP ALE US) SISO RSA BENG ondacacoocsouaine 456 —; Proceedings of the Eighth Annual Meeting, held at Phitadelphia, Decem- JOYE AS; 27) BONGL As}, WUBIN), cco seecaaceotodd Gadosouddco 453 - Proceedings of the Seventh Summer ” Meeting held at Springfield, Massachu- setts, August 27 and 28, 1895...........0...0000 I —, Reference to ‘“‘ Glacial lakes of western INS Mots 2) MD jcntachiescsven sectereccescenaances sstoce 424 —— — ‘The kame-moraine at Rochester ”’ by SEO Sac OORDDACOLEDEOOUICCROUCODOSUOTaErEic odode Dnesenar Raare 445 Title of a DRA ecient eae ee 4, 510 FARRINGTON, O .C., Announcement of elec- TLO TO Lae catenisciessveshe ones ce ate te cunpledeniin es 454 =P EOE CHIOMIO NM iiicietescoscdcodsedee se clewess lees eeu Os T —, Reference to memoir of Dana by........... 473 FERGUSON, EK. L., List of photographs pre- ~ SGM BSCR yas tn nt atebawarteooceun uaclees ses ukeoeanumn at 495 IENDICILONAS), JS) (CI BUONO) BennassecopdabaeeooeaboEcHoeRsoee> I FERRAND, M. P., cited on rock decay.... 262, 263 FERRIER, W. G., cited on gneiss..............05. 123 FINLAND, Morainic drift hills in................. 28 FISCHER, FERDINAND, cited on carbonic acid in the air.. ; O5 FLORIDA, Shore forms on coast of... . 406, ‘407, 408 FONSECA, J. S. DA, cited on ant nests.......... 299 ae roel decay OUND ieoee tit dablaiasteenis Bouananean ene 264 —, quoted on Brazilian temperatures.......... 286 FONTENELLE, J. F., cited on Brazilian to- pography. Nea Up eANceeee ck eM Liisa Ste Ala BRA 277 Foote, A. E., Announcement of death of.... 454 502 Page Foore, A. E., Bibliography Of........c.s...sssss0s 485 —. IWLeaT ac OF Sake Ree a Re ea on 481 FORBES, J. D., cited on temperatures......... 287 FORCE, C. G., ‘Altitudes determined by.. 339, 340 FOREL, A. cited, on cause of rock decay..... 295 FORELANDS, CUTS Paley eaecscrsanesahecne Schenken eee 399 FOssEY, MATYEN DE, Reference to............- 486 Fossiv fauna of Labrador, Xen Oliconssece bee —— Plant Prom Cu Dahenscesecicce-stesacesrastves ie-senndndpnenocem. ceeaiere 416 GILBERT, G. K., cited on changes of level of lake Iroquois See One day ska coger eae Vee Reg « 446 at LS INO TY oo espun nro nanph es eann ae peaeneeaaeaa 359 — — — erosion. gesgarecersaghes G00 —-—— vegetations effect on rock Ascagt seen 302 — — — Pleistocene glacial lakes.. PS ners le) — — — rock decay ........ Sivotencne der seb csesentnneeeens 265 — —— Sheridan beach .:. 2.5: scccesccesdcncoctowans 342 — — — shore forms in lake Lahontan......... 410 === WAIVE ACHON TOOK GIGCENG sGaaonocech os Paced coaeceeneen beaeeo 261 HUSSAK, EUGEN, cited on tock decay... 260, 263 HUTTON, W., cited on guiding pape in geological STG TSS hoe sce sspcoed sae ac oee oe —, Reference to views of, on geology. scenes sooo Aho Seas te VUELOS OL. .ctsc-.cccestccsece-voeeehnaetweceoe ‘9, II HYAtTT, A., Acknowledgments to................ 135 — cited on accelerated development i in Pa- NSS Chiit OUMSA ces sotvsccccesseeassnsssecuseanehelees 176 —, Reference to discussion by....... Bape eewiaie sckise | IcE age, Correlation of the stages of the, in- UI CAESC BBM ue Nr smactesaiciiseuch ah ciceocesae Seewes os! ——, Reference to subdivisions GME HE u3 cose 65 ICE- SHEETS, Drumlins and marginal mo- raines of... weedeat 7 IDDINGS, J. P. , Acknowledgments TOR atat 95 —, cited on differentiation of magmas........ 124 —, Reference to discussion of paper by....... 507 ILL INOIS, Figures of Subcarboniferous fos- sils from. A tanon cation decskescesetatececueseseeereeas 251, 254 —, Mapping of morainic material in........... 24 ILLUSTRATIONS of the dynamic metamor- phism of anorthosites and related rocks in the Adirondacks; J. F. Kemp........... 488 INDIANA, Figure of Subcarboniferous fossil from... Saab easter aso occctnicinc cues tienes trond apescliosesueeeens 251 —, Mapping of morainic material in......--.-. 24 Iowa, WO MRUUTTAN TING O foe sess eases eoteeterccsdecotes sceete 21 _—, Figures of Subcarboniferous fossils FLOM veseees BADEN sh tciac cece eae ana mocseeeeines eS e252 IOWAN stage correlated with Polandian...... 3 — —, Reference to.. Reel seeks 2 IRELAND, Figures of. Carboniferous’ fossils si COMA eos hi Siete ae SN LOSES sguatativoucaanevecacws 252 —, Reference to glaciation i Aft oseese reese icevetesss 28 ——— drumlins i LMM eteetececoececcss SA es ceenad See 27 IRVING, R. D., cited on denudation drcasbeeee +. 300 ITALY, Shore currents ON COASUIOl ese 420, 421 Page Jackson, R. T., and T. A. JAGGAR, JR.; Studies of Melonites MUULPOVUS, .s000+.e0re- 135 —, cited on accelerated development in Paleeechinoidea. Meda aaea Saweaalstaguete nome ce oerees 176 — Sy ee Busses term ‘ “phylembryo”... e285 WOtidies ons Lalcechinoldedesescrssseseesasse-e- 171 —, =? Title of paper by .. 7 JAEKEL, OTTO, cited on. ‘pores ‘in Bothrio- (BOOP AUS soces paqcean90 20603000090: bo RgaDOoRDOdT Oda. 212, 234 JaGGaR, T. A., JR., and R. T. JACKSON ; Studies of Melonites MULLLPOYUS.....-4 02000 135 —, compares plate arrangement of Melo- nites multiporus with that of Oligoporus OND TRE see COSC OPEC EDOCEEEPELE DEEL EEE ee 197 —, Reference to specimen owned by............ 160 —, Studies of Palg@echinus gigas DY..........0++ 207 == ISIE EVO LED Apel (DWansse-d Gecttesssseecescocm-losas ce 7 JAYNE, HORACE, Announcement of recep- tion to Society 18S /eetoabacnascooc00a7, caeaataozaachod 493 Jouns HopxkKINS UNIVERSITY collection, JEMAUEKE ONE SHO oNSsOl Tho soc psc saccocsarcanaonce 248 Jounson, C. W., Acknowledgments to........ 136 JOHNSON, Miss I. L., Acknowledgment LCOS ened deride Sane Seen OCOR OAC IRE LCACEScareaabc oct caaces 207 JOHNSON, Wry Cruel Cral, Che obtay bipae 6 sonearosneeoose 20 JOHNSON, S. Wit cited on influence of hu- mus acid on rock GIS CANY Breese deeeessitesesncecess 302 ———— vegetation on rock decay......... 301 EEN CE Sache cece oe 306 ' JUKES, J. B., cited on denudation............... 379 JUKES- ‘BROWN, A. J., cited on denudation.. 383 —, cited on radiolarian Carths......cccscscesee 81 JULIEN, A. A., cited on influence of humus acid on rock GECAY....cerseccneeeee seveee senses ere 302 FHS EKONOL AGA) SHAN acoecsnooaconoeneodosune 306 ee im TO CK CC CA a2- cence ccooeconestoaiseccre 287, 288, 292 JUTLAND, Shore forms on coast Of............... 405 KALKOWSKY, ERNST, cited on sillimanite.. 284 KANSAN STAGE correlated with Saxonian GDOCL, sesaccacsotcooscoososnesenqosose easooshosone coddHaseN 3 Se Ey IRE SIIVOS 1H), o -Scoanacneccus odo psacodBbaObo Beeede 23 KEEPING, W., cited on Perischodomus bise- VULUUS rareenecciti ck sive Saesteaigetee, sect soeobereleatees ensece 226 — founded genus Rhoechinus... scgsscsen200 KEITH, ARTHUR, cited on denudation......... 389 — , Reference to discussion of paper bye Loh as 512 ; Some stages of Appalachian erosion...... 519 KELLNER, O., cited on nitric acid in rain... 307 KELLER- LEUZINGER, Reference to “A ma- zon and Madeira Rivers”’ by........ ....... 278 KELVIN, ious cited on conductivity of rocks.. Sresneesssh LOO KEMP, J. iF, “cited on ‘Cerrillos Coal... 525 —_——-—— trachyte K cboteessiuueite tess ssuasacectivcsteessecosece 527 ; Illustrations of the dynamic metamor- hism of anorthosites and related rocks Am telne PAI OT al CiShiiasesseccunec-seoseesceveresns: 488 — offers resolution of thanks... .........,....sse00 ) 16) —, Reading of paper DY..........---eeseesesree sree 494 —, Reference to discussion of paper by.. ..... 507 ante Study of Brazilian gneiss by............ 283, 284 —,; Titaniferous iron ores of the Adiron- MOSLEY SS SL SAD Ly hao PO, gat ati gt er ER I KENTUCKY, Figures of Subcarboniferous FOSSINS MMO MIN cea ee Gene Panes bececsese teres sees 249, 252 KERR, W. C., cited on agencies affecting rock decomposition meadneueCotiscecunsevle senoeeeeens 359 KEYES, C. R., cited on Arch@ocidarts......... 214 wes EH erica tons aela sone dats sc tesdina eseess 389 — — — genital plates of Melonztes multi po- Gale Sapte sacvac ee eicac Ne ucles tide desist ae ee denlascenes esha 155 — — — horizon of ‘Oligoporus MISSOUVLENSLS 184 — = — FLYBOECHINUS..0 vere ssnneneereenneronenoevenseess 207 ; Geographic relations of the granites and porphyries in the eastern. part of LINE (OZB0 RE So onescane Ben EH DECOR SEO oCOECU IOC OOUCIOS 363 TUTE Ost FORT OETE hi econ ccosedesnccosodaeeosdoe Feces ASS KINAHAN, G. H., cited on drumlins...... ..... 27 KING, H.,.cited on geology of Missouri........ 369 504 Page KING, CLARENCE, cited on glaciation in Massachusetts.............- Ssgaenteutrss cain eucdaweanas P35 | IGT RETAIN Hips iINe he nGMCe LOysuerr. secret waned: 336 KNOWLTON, F. H., Identification of fossil PVAME, Diy.5< caewecanape cmmdocceteencccm Jbiancedeana eters 73 KOENIG, G. A., cited on diamond-carbon in meteorites. ec caladeicn es acjanaaahad acs deve saawerwonwawaes 485 — — — some recently discovered minerals. 484 KOSTER, HENRY, cited on Brazilian boul- GES ee pasta Sac cciseananvanaascestancos cele oinnes apa 279 Sa Ss TAUPO ce. cas wt vara iatave caus svaswsedoeassce 310 KUMMEL, H. B., election of... See AGT KUNZ, G. F.; - Memoir of Albert E. Foote.. . 481 LABRADOR fossils... 3 LACROIX, A., ated on “microcline- eneiss.. 120 LAERNE, C. V. VAN D., cited on ANESsjicncense .. 298 LAFAYETTE nee correlated with Seamiati PITOCCHE cise eco ance tae caavscenocsnure «Janta 2 =~ Of SOUL: CALOlIMAr..coccenssusdavdesccce eeuecanns 518 LANE, A. C., Reference to discussion of paper by... pele cuanto eneawe - 490, 507 —, Title of paper by. epee contiscun seh cannon suareenna cea 507 LANGDON, D.: W., JR., Reference to discus- sion of paper by... 512, 518 LAPPARENT; A. DE, cited on ‘denudation... - 392 LARAMIE rocks of Alberta, Reference to.. 32, 34 LAWES, SIR J. B., cited on composition of SAU Oleg) foe Gl ee ee ere oe ye ae mee ra NDS 307 Lawson, A. C., cited on banded structure... 129 — — — CENUAAtION......cccessensnesvesecesern ins sernce 389 —— — gneiss.. re oe es — — — hornblende- achist- wake a. eee 125 ——\-——" Lawrentiam TOCKS: 4 cewsders-sascbwulekyn aces 126 J /— FOCK SEMMICEULE:, . seasrconsdeaneneoss oobwips foxes 133 LAURENTIAN system, Reference to.. owe (QT Le CONTE, JOSEPH, cited on denudation... .- 389 a= === WAVE ACTION. ssw ccebseteeswevtersceacsiarnces 402 _- elected PHESIDENE. fassct in ceabew pcaeer nti bauatecee de 460 ; Memoir of James Dwight Dana............. 461 LEGARRA, SALTERAIN Y, cited on geology of Cuba... 68, 71, 74, 78, 80 LEOPARD rock ‘from. Ottawa. ‘county, Can- BGA: sakes ovssoaystunksBeerue vussanegerneten dees en eoehe aude 95 LEPpsIvus, G. R., cited on denudation... 391 LESLEY, J. P., Resolution of greetings 'to.. 09 LEVERETT, FR ANK, cited on Crittenden DE ACM Eta ccceun tt ispebewicen cine dem eaeyoubend ponue semana ce 344 — — — drumlins.. pesweer | 20 == ——-—— AStory Of glacial lakes in Ohio....... 344 aS | ELD SIC) DEAC His scan awese pee vayaten Maaseneeers 341 — — — moraines and raised beaches of lake DEV TV GD searedeee dhe caters eacannanpisVebeeomecdas oyeie bn 443, 444 —-—— Pleistocene glacial lakes.. - 340 — — —retreatal moraines in Ohio.. "330. 337, 345 = —\— Sheridan DEACHI...b...-cscerenunsot cveseeeesrs 342 Mapping of morainic material in the WIESE YE eee isles anceabrSvinutenpoubde pL mesennia eee 24 = SRererenee to discussion of paper by. sees 509 DIRIES Of PADELS DY vn asc. an ccemsnshnenstenncaeeiay LEW Is, H. C., cited on glaciation in Great Britain and Ireland... ccssscccccssesee covees 28 — — — glaciation in Pennsylvania.............. 27 —, Reference to work of, asa glacialist...... 471 Lew y, L., cited on amount of carbonic acid Pere or le WP BER Tin I 03 Liars, E., cited on landslides in Brazil.. 267 — — ~ rock oer: hee eee ees 261, 4 LINCOLN, D. F., cited on drumlins............. LINDGREN, W., cited on denudation........... 389 LITHOLOGY of Missouri granites and por- PHY TIES: esiaecc hho eee eeneaic es toes 366, 367 LIVERMORE, S. T., cited on former mates ah Sandy point........ » 422 LOGAN, SIR W. E., cited on ‘Erie clay. Jatin «9/330 LONG ISLAND, Cretaceous strata Of... sees 12 —, Shore forms on coast of... eee. "405, 406 Loomis, ELIAs, cited on Brazilian rainfall. 311 Lovin, ‘SVEN, cited on ambulacral plates of echimaids:: .cekeat see Agios BA Seaen® nee eOMeEe BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA. Page Lovén, SVEN, cited on Arachnoides pla- CENLES Hise sG sansvaxs senda oesbreseae outer assekes 230, 231 — — — Clypeastroids and Spatangoidg........ — ——ccorona of Echinoidea ........ 229, 231, EOC YCLICR « wscn cub tte bute stavvarsotawae ae eee eae — — — Goniocidaris and Str ong viocentr o- tus.. bs --- 144, 145, 192 — — — modern echinoids... a> pap sicmeemeyeemeLne — — — Perischoechinoida.......sc0-cssscsseceeeeee 138 = LIAKECK IN US sv coondiananveuss caucsvy cvexunmlAi areas Luc, J. A: Dr, cited on torelandSas-ssesaes +» 400 LUND, AM,, cited on antsic.c.cuscnsuewdoe eee 298 LYELL, SIR CHARLES, cited on denudation.. 379 —-—— wave action pis 0B sha mio'tuNn sa Sun do slap sev aviansies 402 —, Reference to views of, on geology......... 463 — — — WTItiNgS Of... ceceee sees eye ohiwianigele sre OL AEE MCCONNELL, R. G., and G. M. Dawson; Glacial deposits of southwestern AI- berta in the vicinity of the Rocky moun- TAINS «. ssssiosensinsl iia: eacnadelvads cece g ie eR eee 3I —, invents term ‘‘ Saskatchewan neni Me we =) Title of paper Dyn. -.3- is scatnnarces cs says temeeaen 12 McCook, H. C., cited on ant burrows.......... 297 M’Coy, F., cited on Arch @octdarts....cs.0..00+0 213 — — — Paleechinoidea ..........0.sececcees ; meee — — — PAIGCECRANUS )vcvcniessessevasysaceues nencs. s TAN UbAsNtn Rb seniiee 3 SCANDINAVIA, Reference to drumlins in..... 27 SCANIAN Pliocene correlated with Lafayette 2 ScHMIDT?, F., cited on Bothrtocidarts.....cc000 234 — — — pOres 1n BOlhrOcidarts wccccsseereceoees 212 SCHUCHERT, C., Acknowledgments to... 135, 229 —, Announcement of election Of... 2, 454 SCHULT ZE, I,., cited on Lepidocentrus miil- LOWE, vcscns cus soveks inisons¥qasdeoy dagioa sides ine pena tae ee ES 224 SCOTLAND, Reference to drumlins in............ 27 Scott, W. B., Acknowledgments to............. 136 _ , Announcement Ofecture Dyt..<..panacees 493 SCOULER, JouN, cited on influence of vege- EATIOUIONTOCK CECH Y.-.isc0nccenensascenpoveeenneee 302 Scropg, G. P., cited on exfoliated rocks...... 291 SEELEY, R., cited on denudation... . 383 SELWYN, A. R. C., Acknowledgments to..... 95 SHALER, N. S., cited on action of shore cur- LEDUES hicvaaivdecereuuaebesaakeoaato oes saddvcnssyoeouhiass 408 — — — Boston drumlins. .........cceeeeeeseeeeeeeeeees 20 — — = 'COAate CHSDS. ie sseveso ves cnsumess aces siemens aon AO cw (eee bt AT UT ITIS 457 seve sccwdvssunussankescenieueenatees 27 — — — exfoliation Of rOCKS............ccceeseeceesees 291 — — —. headwaters of Genesee river.. » A383 — — — sand movement on Atlantic coast... 404 — opens Philadelphia meeting ... . ...........+ 453 — — Seventh Summer Meeting........ ......... I — placed on Mount Rainier Forest Reserve Conimittee ss: Gees, Viecss. ae ruees eer sreeciees 2 —, Reference to discussion of paper by.. 4, 7, I1, 14, 15, 493, 505-509 ; Relations of geologic science to educa- ’ tion.. RAISE OERUOTIES aso oronperonosanbacreuabeckeace 444 —; Drumlins and marginal moraines of ICE [SMSSES site slink nad eden eee ee I — ; Preglacial and postglacial valleys of the Cuyahoga and Rocky rivers.. Be —, Reference to correlation of the Lafayette with the Saskatchewan by.............., sce = OLS Of (asa SlaCialisty.c. ewasceeeer sews 471 = atlerok pape4»n, Dy iy weit. cee aceeceteaveus 12, 510 URSEL, CHARLES a , cited on rock decay..... 261 VAN HIsgE, C. R., cited on denudation... ..... 386 Sa MOSEL EH RANON IDES Seco naccen eeeccnpenH sae 375 — — — secondary origin of crystals laine 132 — — — titaniferous ores. 15 —, Reference to discussion of paper “Dy... “7 504, 507, 512 Title of Paper LD iene ee cee rt rah at ae aa II, 507 VAN Nuys, T. C., cited on carbonic acid in JELENA sc a eee ait Oe cen A eee TER, 304 VARIGNY, H. DE, cited on evaporation from WSS most esetobo, aso! 2 sedan ecOupeEEd Saban L eee eee RBAEESG 361 VENNOR, H. Ge “cited on geology of Canada. 96 VER MONT argillites be anassease rea lceine nesta aye SII, 512 Sy DETOn Mat TO mMOted sratiyn- te nseeesecs — suggests name Melonttes seplenarius........ 182 WHITNEY, MILTON, Analysis of rock by 351-353 — cited on residual clays of Wisconsin... 359 WHITTYLESEY, CHARLES, cited on Cuy ahoga AKAN AMES [WAS s wccccsuencienensndctay svectieun tices: 330 — — — Pleistocene glacial MQICES Ts casaxetesesecees 340 WIGGINS, JOHN, cited on forelands.............. 400 WILKES, CHARLES, quoted on Brazilian LER PEHALITES Hi. manu pecedereustetenvesbcaeendes ua 286 WILKINS expedition, Reference to ............ 336 WILLARD, J. T., cited on nitric acid in rain 307 WILLIAMS, G. Hi; cited on rock structure... 133 Eee oo: SAMA Batts RDO Sesh 1a ssa tas ay reeks . 284 WILLIAMS, H. E., cited on ant nests............ 299 WILLIAMS, H. S., euearas of memoir of Dana by ey Se onbern ead —, Reference to memoir ‘on Dana by.. Lc meawnchy 467 -- — — discussion of paper by...... ..... 4, 504, 512 WILLIAMSON, E., cited on Brazilian boul- BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA. Page WILLIs, BAILEY, cited on Appalachian ero- sion.. Soe en aesenee set Saubiviei tesa dy cee wens SLO — Se dentidation cnt eases eee 388 — continued on Mount Rainier Forest Re- serve Committee .. made 2 -~-, Reference to discussion of - paper ‘by. 504, 507 WINCHELL, N. H., cited on Belmore ridge... 341 — — — drumlins in the northwest............. Oe — --— Pleistocene glacial lakes... .,4\.......... 340 WINOGRADSKY, S., cited on acters le + 303 WINSLOW, ARTHUR, cited on cr line TOCKS OF MiSSOUTi........ec00cceccecneneh gfe ooeese 369 WISCONSIN correlated with the Me len- PUES 2, * . ’ 2 ’ ‘ a 3 : i i ’ ua ‘ ; e ' ' 5 » heehee if - tee J * mn wn” i : ‘ LD Sie es : r So * f ‘ < - A . a? ab * f é "A hi : af Ab ha’ Als ” he fete ont ek Bie Tete hee Bete tS te deh ete het) 4 mead Shee Hed Ab ete we) emey WPT fer Oe) Me bepctietypebon Devel DW hepepeh eevebet every Oe ee ee ee es Co hte eh ee A ee a nt Sete 7 Pot 4 OU ee ee ee ee wetel vies et wey Bypveyegey i : eee os Ce ee fe a ee er ny Nhe helt Di pe mete, ow Pbdibide dade sar i) ew igh olir Dole <=: LE EES a Ore HP ene idtuctytprebntatisetsestneitetet asians echt | og al : b eet My, ah ‘iaaihadibeek pas UN Ad SMITHSONIAN INSTITUTION LIB ts tae di et mie ¥), Sb Peet ‘ i Y hese Matt Hea T ty yeep: } uw . z ‘ Rininceien wheher, ta . ty HDD. ailvendatte pel puny: : ue, » no aos ah be be Webi ehehemen wih hey sy ee Uh > « « “e whet 4, " Ae: We eusbaiecey buehehb tabuis sbenen atnrccoren ee Howe uinie i saa? 3 90 Nip emery ont by i wh : 88 01309 1764 eeeve te hehe Meh petit oT ' He Soebend * . fvenens ‘ ithe indbleate be we 4 } ‘ Hea Mahl ord Pete he N ek FOR em, ¢ J . " pobetrmte Met moat op etuod 4 ye oe a ps Cap Mahe ee eM Me Geel, be iT we o, rt * ur : we pe , cece ee iid cdi ad_ audience coal eT ar > wee pret att Cs Satdbatar yp iecbat he ? q We POF gh pe teeter gure 4 ry to 0 fhe ten” Cr ae Vay ' eh ED es: oe why 4 satiety Meh ele tw pO pele ge beh ete loeb ro - 1 . Lae yeette dass phe , p ' ee Te ee ere a ’ , aan) ' aobailh we it festhek neta eR teehiel cmepepeteedew seep raed fea: 4 : bentpon ab be We pete ; ' cs FE oie Ae Meme t DE hem phe, bel te Me pare) Walesa LIAl LIL SU ee Tr Heb pedte det Fi rhe hehenen eaterh cared rer arate CA at if paye Pend eed) bee eA AE betas we bw wp 6 iL) ehh! hehe oe 8 Ferre W heb pone } ome wh ibeeme thes SOORDY DEM ALReHebeligvn) Rr bor lebet betel ob) Avan! pabbytbnst hotel Orr ehatel: ‘ SMe hehe ew HEN rE Betete hee ear a PPTRARe Re oe hebew i rane eed eh Ree bee wen Nott Meher er wee Porte! OM Memehet tn jmp Water) " 0 Rermmnged oe le feb etree rhe PME we hath he: re ' ’ » + i 4 b etet te! + (Ong: Mamet eb (ognbont os 8 haw ak At onihes arsaeenes. Ohtani beh |, & Bgdyl «t-edeteinn + okp! - Rr pop bed phat de 4 1 | Swt epempetemtel wih eabeqae pig ale tpeteek ow « - hobel b! burt nebehbObtaiet ia, hd eteth: h . " bite vm ete! Moe Qhatieh.\ pe tebute deeibeatathillionteet | AYP tysegt sd po fod PADALMLL ULL Le Teri 4 y Hh Oe mele foe ated pe ete pe peeked U ° ' rt HY 47 O44 nrwe wt el Co hathobe tert peng Dated titer pe’ Mp! hebm taker wh a os . : Pri eter te wires my ee ae errr ae be tow 4 . wha . "09 BS OU a Ba btin Se x wh w ‘ Ve ae ee de habe ne hetetptnt pa aleab athe ened Rote PEheee & ae Le ee ran . Ld ee oe , ererest Stet ge _ ’ ve ee al a de ee are ‘ : 0 bts > Dette. eae babepouss ue i) 4“ ‘ ~ . f . . mi ¢ : . . he +7 Ns AES hha vi) by + Ley : f | lO ay 8 mere. eat hy wn p+ eet | 4 : ok: ome . es i yecth dd bas rarer * z , ’ * phones tafiers ‘ ahah tote oud Tigtoel P verbipt tos bend . . ’ , parses abe ~- aed fob shen Sh hn-t FF os, wihitat ’ reer PEPHDE PO ome he HO VOLE ee oleh Re ph tet bnet er erie tf, er tare ~ Pepager PERE HeT bom sremeeye he Pht tote ama bee wim, ey» SOP Leh be phat: beh Aes herent, ~ ne edo * MOMeO eRe OLOdEh etre h Medien, oF peeeebecatie hen 5 + ois eter er Me oP Rabitiglehne: Sn obs ome rb pbalhe ot » PiU ebeRlrpebet om Memet 2 genie) ot te, ae sh jo a brett ote ee heb “+ * Get out tie biter « be be of . oo) Martie du Ten Tae area a « ' ’ i flee on obytnamethgeesa se erry as Pah eh Cede neh hen ehetne lobe eres he . © Reng ot pee ‘ ty Vel gt) ree ot ear as ol phot t tot he Opel ipod “ a ‘ ack shea) ‘ , Habe ie ph As aral elds aAete ar $449 4 OP) Rp tetetier se ; = C eased Y one" 14 oa vs whl ng Caan Poe . * . * OMe te hee: . : PheDe t besheseurepeen. fi kak aeteb hee bchpatemee hen. tpt . = fam.t habe ainteel atadad cle / fel bev are hed Tet oh in BOD OOMeds Be nbie Pte bess 9e ‘ r ti : ; . ° ‘ . Me ’ ° ‘ a ‘ His Carla cat ta th fate : site ee ts ne teng? se e mH Bt oI sae Deingtene . H {be ster opp as wats gig el: nadat ot In. " a, pre Kenel yh pen " tha baad ; hvertee’ r r . : 7 ! ab ne HME! AST EDR tObiD ME hehehe hebetel woe 6. L | Uh , raf , : ween . Shee! gt A oa Ba eab eof sabe Hetiks Bane f i ‘ : paw Bobi heh neh ohne eh : rae Pgh aH59 thet whois rm! fe bet Me ial = por es mens ny ' - rpm ars pte SOPeE Leh Bebeh en Ly DeRet oh Aah of ol i ‘ fT opt pt Del Mel pieden ¢ OP Ab" Ot BD OE Hf ¥ ef bone ete etal ahet ety pew Mey hey 1 hheg . vey © Be: fenet hak oe who ite tend wg ot ‘ ase eos 40-48 Abd ok ee os NM OBs | oh ol wed we utp tthe beh MR Eb ore jhke we. . . - eae PINDER PE he hal oe 4 ether neal ity wali te ot el ol > , | Ass belek Al eh ea a ee fot whe * ‘ . ‘ae i» & ‘ Y * ot he he cae bee fet ’ ‘ i Ae 2. Ce oi ohne . . ‘ Pt te Mo bente ey Reha tonhid PE DAP UAGeA DD Meds” Teli ge ee 4 oe a) And ‘ ; . } ape Add al Ur aT area e ee | * » Natal be to8, * . ' ‘ ey a a Coe t) ’ to “ tp ome of . bp . b ‘ ' ee ae) i . oh beve Orb we te a ‘ ~ © Wome ' . * + © tee 4 ' ’ f TOL thet Bh peArt ey mepeteleds by tere Gepspm ‘ . a OPPS LOL + Ped RRR Tbe eh PALO peber ee eh ° pa? fork. o-* ‘ PeBetel Be) Ppet oh: oser ptt Rene eRe fete we te! te ohh de = » ° ‘ ‘ Ei Peet rte h! bee rep ith hr PET PPD ROS el OA me at om ge bay a: Vere © > pe . ’ grr hibep! eth! eheraee bie A tad bab aod a ah ae OHA rhabel Orb b phe ael otal nh betel age es ee eee ' " r re rhelere'«lelebe . LO hped ae! ml oh et tet abe ee ae ~~ we , . + Ap taheme bi be bpm behitetied pi itehet phd tug ep te od ther haw , ont [Oe t? prree Sr aera Senn a! 4 phepetyenad - ‘ . ‘ yen A 4 se trvarepts . . a . bokge ho . ’ ' * i of “ > -~ dome & w + re i} bebe ~» e ’ . Cad | tr 4 * => . Mab > » . . & . Deb ob ‘ Pletel mel Se i] b- . “. > | a seeds roe ae abel obdel Manes aelankot ant a ee ee . » _> * mo . o Be rept ot ihe Depart ehert § Od) ete! bf pe | ot . * . a - at nf ' « HES Pee eh PP Per shat at hab py wl bp! whebiemeny pid tbe we + tet ee ' te pe o- . 744 \ 6 0al otalt en, tt 0) PO ele e «le ‘ ae Sa ee r oe - +» = . - op? be DPE De aeOe pe! Leb pete ohne et * +! rey rir ye § trby ' - a t. cs . PEST B ve FLeeh ee pel abt bptn Hf ot etal ple 5 he spetit pe, Orbit, phe -, “ fortebep 4 oe . - ee bee « oie ‘ Pa olla tee et we hptete " al Soe wb how <2 Pot one ‘ ‘ Por elei toe eph) PO b Oop pbel Pe leh et || pepe ob & prher ' rf - a tht tee Hpeel. \ » to prey 7 o ‘ . ty of phew ” o- bo ° e ' ’ i ooh . . - . at wet ° is n e ! i“ - ee eens Pt Pp be pee oom ' » ee ’ - MEE ab) eh O4, be; , { n mo - . » ‘ “PAP fal obey te bt ane ema t « { . ca le op , . Pte porte, Sele tel tye fbb as hed - - $ se rster sett ° anes et Cae _~ M Te pF . ele + bee Me ‘ * 4 1405 At "j ph tsbelveype J a ‘ - 4 e ry up post Shei re ah 4. Mtraedsg : bp, > ns ' » ia ‘ " ele a5) er ' hotmepe ; - - ‘ . , bi 44 "y iat Led pt j 4 be!+ a» + e0bel — » eet ORatew | me! ob BE tei tay ial i ee P'¢ ool edeb et ber ot ohete.p whe oe & ‘a Be ee fh 4 ' " 1 oh ; oe Ne baci . bon . Py n rhe elees loo Te Selbemen . as wi eeiies o) grit ‘th, et PE LD Pret er 008, Lele Side PePel oy Moby ie hd Laat od 8 fi ; . i - . . oy obete 1 f Na siletens atid a ‘ f e2 boas theres : ; : : ~ — : ’ : ye fart dite MS ae , . " bbetan aie Seber i-oe i bk itp ; wigs r ft} 4 ibe. ety ‘ ‘ ny F . . eer c : el! ~ . " ithe rete repeat j - f ' ' oh ° F - ‘ o tee ts) Ds de ee HS ey vhoi bw berh ‘ ‘ wep toe we . . ¥ aagholeeienn 1 ’ e ‘ rane | fy : : vig eX we ae Parr i ae of eer Pa cee tet eee) feel 4 ” ree Prhsee fer ye | ad erie >» 7 e be A . ~ - . Wine ATAeNMu gal, ett oT? ahah! ae hae ctr toe rind ibys Wye a> Fler eter pe tele ehepeke re opr ous . ‘ ’ ddr, bh d bb Obl eel ae O44. PEEL bey bpeyy - * o~en n PAP PEE eh ome de) | tbe. - t-* ~ » : - ehaskd tele ee eee j “ ; “ . ‘ a . » ‘ epee r ‘ i ’ ‘ , . b- - . oud hop > > ere ohh pe » ° 4 at Pye he tr rp Prine - . ohhe a) DP apa) ow . ' ri wri ye . me 6 ag io OER E Ut PDE i onein ror breerere. ©: > ~« oe Toul td de Kad cs tae Ol aad tee es Oram OPa pet Re ye ~ ow Py Pate tty ber ut aL orc euee, cuetee ad i a Eee I ea ed - --* probe ” BA ageRD eye) PN 0! Hh etLOLL Het ony teeke abe ‘ . ‘ . ‘ on aa aan ore? 0 bet + . » Fy ‘ - . rs a ed al ed hd Bet Be oa le cts ee Lt ae | eT ee Ee ot ” a ae au, ary CR fee ponene Rear mary eel ye Bete , en ee . SY sect dd LALA bad cl dt ee ee es rae ome & ee . rer ~ wee Hat bedbArved ib oa veh Haber ss imecnocrl supeprodpheetirnricpsses tron reer ahi ney oon mt a r uhm Fy de by: OF Ret Peep are atprnedevel bebeeeh, bat “ o> ere 7 be rt bhoe Sabra al aatie hdd Mie cat veel uty eye re ae < -eeee Powe * — ' wry . ; t + ey tL Perit Ges beliees oteter dec epeln — - Ai me. “pes ’ pala A ielols eae ott Bete te -“; (me bus @ . ’ ’ "e+ @ o - ’ tore oo-e ’ ' es or ‘ Weer dertes be! che ef yey ores Oke ey > cnet i s*vew* ’ or eal ’ , ; n a gy, th 44 hed dd el end ee de ek ce Be en ee ee Se ee . 2 eS . . - re yey RUM Ad Ard) cd RE LCL RT Le od ai cl ded td LALIT al ee ee (ree © Gea . ya * - ‘ - fOr y ete y ebeie) Fh rhehebemeh eqenel egrets rpbebiees Fic eben rete vrerer, + of j OR ppt. . ewe aie Vere i yh ER ORTH: pate ry eytees re FR e® Ghee wedeeel Pew ptee terre ® hoor ens n * ' ‘ 0 te bled hohe Bid 00 bd oo he a ad oP le ee oa r) Pr ° ree - ad i ham ns to eee, Revrey cree phverg-s\ sue ee ers ’ ~~» j a » : “rey ‘ ‘ aM SOF vtee Ariens Che) ob rer yti Ph OTL & ow ' ~ . vipybubelise im edd bh ddl Moedek ake 2d le te Lea eee ne he OT TY eS eben tee 4 ‘ * ab js on ’ Pa ® the 1thth 5 v N46 PL eh Py eEh LoL Er) b4 4 Hed wel Poet gee Crees em ae ' Pe+> * bey Habs tod thd dh meat et Ata dhe bled kas a ee Cea dd Ge tobe Ooi Oe oe el ' 7 art. on ’ ri th th chet hdd Sd Pena rAR ARTE data ter eeeT ae hes ane ee oan £0 6 eh ry Sa et yt . ‘ . o the hoe a ee oe ee Prats ied ae ’ Pig .e or ’ ine rides Cree. ee rte ee ee ete ° ‘ + - . . ’ it AAAs he Bed Be ea ie ee “ ‘ tee il oad Ree . . aM abaels rhene Wb oR eet e, eae, Pane ye ope i . . . * ’ * UUM OE EE HE hene: dU rU EN ye RRR Peder A rede be ive ' wre or ales ° UU hae Bs ee Oe pe ae ery Mri et eyet . -h - oe ’ ~ AdLihth dt aLhLe dd od be th Se CEP Veber hem vee mped Oty cee ot er « - ‘ e SUF PEON Pi tire bury ThE ten yes b is oe OO ee rr od b . e . ‘ = ° : : . : Sid das od td ee a es sa aluk Hud hls Sadak ed el er ee ee CPR} opared ouebotut oe ered) - 4 2! eee a Al ie La a eae Be ees te ee « hs thal ad GEL ee ee ae ~~ » ° * —_ . ay rare, - Eee eE Fy rely Fara Eu RU LOE herbed rb ye Gap slg Fadel Magick Unibet aloo Renta Mos Bedale nek Pa bere e ' wee . - A por > Meds ernen Abe Aiba ea Ot Pe nd ek r on bh epeenpar

AH tial bh ie hd oe od 0 A ie oe ot ETP ONeY i ow t« ‘ —— - ‘ oy L - - ae WONG OU Mamas, beri, rats ee PRR gs peer - poe | mer . an * yore ' . - lan hid Wheat ol 2L02 outed aad ee CATT Tes Cae ee 0 oe pew RP Phen o6b eee 0 _— - “ Vet OPP EM Pe eben ety je . oes cory Oey Orem ag Ny et re 4 We petoe ys be ‘ ‘ , eh) tema a ee =e - en eevts be — ~s . : - ore y: - ’ . on rower - . - ee) oe Ob PRR eU mH er bry fev evrieg, b 1 ‘ ’ Payme . ee { oe wep ow ' ’ ‘ . rr -- —- » Faprastesucur coves PEF Heas Aoi pdneger a2 , " . ~ reps = wane babdeet : vee a) ne waned “ne, ne OREN S S uF ow ‘ a Gabet..+-toe tee a . 1 br oRe Deere bedee webs ieee Be, ere a usu rhs i rd ee she ney e) « rP td He Brew |