« at _ h “> ; voete i barty & * At ~- > qe! $ ! 4 “ natn a hehehe Oy begets etek gs 7 «wut ‘ . To J fey yet el ease. tA \ ' ‘ aa weelete peer hen Mrrrtiry otkn | ; : a re eer | ‘ . ey rie, hoe Vee heten hat tet de Melty hater Seder aul ah “ be i. ef 1 i vd Weta a Apo eM ater Ashe sig by 4: tq he lateaby'he te dade ln he hy po Renee rMatate tod co. aed deine? ” “6 ¢ Te hed oed A a ae Hy HN Were hla eo ee ce ae err nd or tate bets? & re | sh. la dee UA pty eettetntents be ttthe ds tet rtale ad Ly io » i : ty! ; eT Ah ‘q tytn te 4 - ’ tee gets ' ' 1 ee) ‘ ' : ot BY er deeds ty sete! este « mt , 7 4 , m ghebagt te tnte ese a a my} as! 1 ‘ q ‘ori at atl ' ] / ers le! Ba iebs Ca MeO, TF. Che als Meh Se i ew! . 4 Gee i Ae! as ; hee ' ‘ 4 } pal aety ie 4 Valet tag telly La, oC 40 Aeta tees Jet det te eae a ody eset el tet Peierls ‘ vats *s® 1 Se J ‘eh 1 4 ike Ge 9 ‘ th Pier At Mk BR RR DeTOn A Bonet bum MY Pee Way bt ted 1 be Hg te ed Hg tel Oy A te Oat alert a bre ols be tna be bathe’ “are . : Toy hs . tb. bee a ‘ i ‘ak , ST eae bel eee! PUPA en eA Pe | an 4 Sete we ' * +4 . 9. Abed i 1, ted toh ie 1.5, Web ady dete Nile te Deed dy fg Fale toty ia @ toh! ‘ ‘ l ue - et te.’ te) Oey OY i ta th i tated shed vetet 4. the Sa het pene ' + | * hsb ‘ wh ¢ (qs } ’ ; Sheehy CM clad ot aaltets Li tetateld gittia Ay Bt be Ala) oes + be tates & im! ‘ a b Pe “4 Peri» ' r ts RAP RRR RMT MEL QA bn Bette Loatuehs phan bral % Sala tek bb = : , in te be tered . Wa aie RPS | erratic na Bee Wee be tate Pe PY ts Netaeeh det : : e' id ‘ “e?,? A. ; aie ale P+) ; ; Med, peat a ayt ote Peer rnay Se aaa Nets pe ta ty tell gree ' . | ‘ : ar ] 4 t » ~ “4% 1 : : * hal ple A a telson tate bye te te dete Sets 4 Ms Get Made Oahy tals te tee ines tet He » ’ : r, M4 . y er" 4 re al ra 4 del eck ft de shat Ashes heda later Ut pa | heh ta ir wed detset oie ty ket hy 144: ‘ ‘ see ted ‘ piteteted r? ag se at Debihd tof bee toll ha be ly Lan) Lowe | toda ltte trieas® te dit , A) Woidy tal atnltntg Me tate het ** : Ae ‘ aivte ld a4 ( Mie) eae Mt Yas Pt Pe nd tn ate fete thts Th 4g pay hs Geb oitg Sete d ety hate tate” = Bel p . at take et }o3 Tp teat) Vay ha ta bets ig ete tytatete Lbeta ate Aye Ube ey Letig te Mea alt by sHglty bebe hay Gokg tptyte! q ° rer oh Wn 1 eOeB eM dat +O hee Oe or ath dant | had y Be ta ete Pee ee ff ' Lars a ’ 1 Ag hg be tgt Ast pe heats hake “A rere tats | aT ee Cat tety tote ae get ly Leh) Sag ty Spgs betet ety eng Ae le terte Sealy ata tade fe teh ® tothe behets lat whet 9,0, tg be be tet KARE 40 444 boy ty tele : eyiyis' bs he tet dels Soe ire “ lier teeees te esheed Wty tv's dg hte)! Pi datis he te 4 a dhs Mey Ba teate twit Co ‘ a? teks raha lita te het Hehalty ny Gh iP tide ete inh trhgte Ly twee shy is BP Ape tty) ete t 4 Vow. bbe dn by fe tetate Metal ede elie leMe® siete Dido ty Meer te ets Ma tg hy fete te Ge Ok a't ute state tte tede te tr tyQedehade a hetats lal) tg te Pitts teteaet et tee betel wet a. teva ae ptatetatatetetetal ath ‘ tetet abate ty tete te"e") ety tala lg te tet het ttehe i] steals ‘ stake da fede ‘ abe toy Any ft itl te ehehetetets 4 thee * beh bb % te Ley trtads ints h GO rirSats: atate este diets by a F Peek ore kb | ot oly gti 4 phel rha le tet be aerate el att! it be 1 he if ‘A teh beet! eu teeta civiatits istetghe ete bette Gi tetets abe. me) : aletetets te Dh late la! Tidl be MPT nie 7 ‘ Syeda eats Cre etree ts a «! nbn ih aisgits 44h N; ’ whet sthetar hi +8 4h rAte* iptet rtrh) a ee tee Padol ip be she tel ve : oe aay wie Ds py nt) MEd pte ae high (Ber PRR) ate seat ss tbe) aused ae acg Ud A linte sie atl ate) fap tabla Ae jad aay td + iah alas rat ey stay ‘uu Suotsts an aie tasty ‘ Nel hy ye ahh Webalt sibasst v : ' "Shh pR4e ead eriteee reat es eae ; tr Ayo tates sha ! si padease ss gedit: Pree phe (orein me bose tala ins erie yee evautade) wEois in br ane Gadete! Wats te bet fe? PURO rest » bape eee +r patos bth (a US hel db het 4te jabed ' 14 grata date te: io bette tra 4 sGstvoge Teatt { sirtaas? ' ites te eeeegenet! teow Pr etter Me) teres Bog erat om eral a (apr wey Cty Niel ef a, utetateretetetel oo bode oro hele abed Gage tar Ware rr eevee tr) take eh pes Lal WaT 4 PR: aheare bg tite budeta kal ol AaB he stomaretay NyPSd ba hot sags Cir 5 | Bey tga tote sehek gh stare degree ae iet apatdoioad Ora de t iat? ey ‘vat aataed va mane iQ ge it ohne Peet 3 weve Si finery Ghia te lis they) pwiateds Uiapas & He tore setahai gett , , ’ ‘ 6B gett AE oth 0 oS Saheb tate (ete * ethan « 1 shal Died tes ah ohatetnte fhoae yom Start or O58 Rte Bi ha Peat oboe SONG a ) ernis ie pier corks Poy ett a eet cae dette t Metwte) eres he alee ere ld ato « hah yoda Date eeer atyts Aah ete gobs Oy pa teh ere “are Aer) ee pit yebeb ees Dinah a het fend Obrat e wets (eqereten & chee dy iowa bpebetel yedettte picteletel ale libel sbeteted be papeiat GM, Seto! sreeeit h.Paueere wee by bad oh ater pare tage te rh) ‘ « b] ai arate Pete dy peat ” Abas mery toby * Shee br ete Per tie be ove y deashy ate! 5G ore phi daacten phmeey GREG det ite ated ody Me oP trae dete + Seas ru as nid ade y phate Wegee crite Wie isa Ga nrut gs st eran rps be WH) Lede pans ag eth ont shite hu patio e oes fares ae a thats Pe tes herr m ae gt the tet gabe ort Per etree Lh a Abe ete a giingat lee Aue leew ebeiciad “a “ 4 Rt better ren be mien dom bd ptt ered arte Pe hietereds babes Gapewete hers dpete hs te vie seeret Peet 7 sei ap tere a, Nay breve 4 cog a pene PATER ‘ wBsoaee eth wy Grebe pe taret atane ty taee elect Gy danas cetiguad Pate errr asad oded he eter ye are bd eretisnue ‘ TAG ' asters Piseth, Ot qt hla nities iis eel erayet bite ret ay Eilt feerrmertn er Creer Tteiks Dh reste a ht : ‘ mht Wate Th wien oot Ay Arter! Ra ra ‘ vo Sites 4 £58 OF bees “ er ea a eo a “ rs , an H te pee her perticber sheet 4 ei Gt ite ee 40 LP? ar cone raeanee Fat RETO tee | Sorvrerrnt ty | “veanet eat vanes rey Rt ape L4 Spee 909 wer drerghew ye ear eae eee Si ty so oraehte st ahite watt waned ates remo rag ehh eet wee » pan | Tue cAsraep peas ee enet + wre Prermrntitr Write CLs Ti a, Ebr iteasne: ody Ge Ql a earton Or onde tee & a deat) tee ide nah eae vtereryares #ibigad otasat os Lopevrevawe ri tive at ti Mey RO) POLS) Pereira tia dk eke) eres ' uteaas sea ape rari eet Oyrerny ewer wearer had fates ars ceree orp tp ae res oo sys 1 © ere mee uorsteaete Sea eet cre heegree hotest rene: 7 Trias tet |. 0s. see 302 O08 mg Wissahickon mica-gneiss with isoclinal dip.............. 303 “54 4 Crumpled Wissahickon mica-gneiss ...........0----seees 305_ eae ~ Granite-gneiss in Leeperville quarry............-...-..--. 309 OG aS Granite-gneiss in:‘Wards qualry..3... -4.'. eee 310 cae i Photomicrographs of hypersthene-gabbro..............+- 312 i OURS. a Photomicrographs: of hypersthene-gabbro eer yr. 313 Oo . Photomicrographs of meta-gabbro............... + aie eee 314 gee i) oc Lafayette soapstone quarry.........-..--+-+++++-ceee 5 a ee aa 5 | “ Specimens of steatite from Montgomery county, Pennsyl- VAMIA Oo. ie esis dete eh is dle sputee a ciate = Sheets lee 317 oe ie 4 Castle rock—a pyroxenite dike....... ee oe Se ee eee 318 a. 4G ae Dinbase dike: . 202.0006. 45 acest ecies e e 320 ‘* 64 ate Chester Valley limestone, with unconformable cover of Triassic shaleg......2 52.42 020. We atin te , 325 ‘© 65—Matuews and Mriuuer: Generalized sections across limestone area of Baltimore county, Maryland.......... 347 ** 66—FuLuER: Jacob sand and Montauk drift....... ......... Tae Sh eee 387 ‘* 67—Stanron and Martin: Triassic and Lower Jurassic formations in Alaskan. /s2.. cae’ say tyes eee sa ae OO: of 68 a . Enochkin and Naknek formations, Alaska.. 395 oa y a Enochkin and Naknek formations, Alaska.. 399 pe ee ie Exposures of Naknek formation, Alaska... 407 oe 71—Marsrers: Geological:map of Verniont. ....4... . os . ease ee 419 UE Geological maps of asbestos areas in Quebec and Ver- THOM bids. w win ice oe ha ales ein bo di she, ar sare Cheyer eer 421 gape 5" ‘t Belvidere mountain and Missisquoi valley ......... ..- 426 “74 “ Valley at foot of Belvidere mountain, looking southeast. 433 See hee ee Slip-blocks from the Tucker property. +. ote 434 ee TG 5 Serpentine block and a section of a cross- fiber Velo caee 435 ae By Photomicrographs of amphibolite .....2 ss: 2 eerneeee 45 pea ge: * Photomicrographs of amphibolite...... a 5a ay ee 438 sped (9) is Photomicrographs of serpentine>. .... .--...cessaeeeee 440 ie a0 ‘* -'~ Photomicrographs of amphibolite: 22. <02.2-. 0s see 440 ‘* 81—Marsters: Photomicrographs of serpentine (section 42)........... 441 ‘* -82—Cross and Howe: The Triassic unconformity at Ouray....... ..... 456 af es se “The Triassic unconformity at-Ouray .-)2..eeeeee 458 eB = as The Triassic unconformity at Ouray .... ....... 460 © ie i “« The Triassic unconformity at Ouray... ee 462 ‘© 86—Purrxins: Fossils of the lignite of Brandon, Vermont. ............. 515 oe eae os Fossils of the lignite of Brandon, Vermont ............ 516 de fe, i Sections indicating the development of Paleozoic strata in Nevada and the Wasatch uplift........ >. ~ = 2 eqneeeene 528 ** 89—Burkery: General map of northeastern Utah from Wasatch moun- tains to Green river..... ‘eeaue a ys eine = eo Ret eR 529 “* 90—ScuvucaErt: Portrait of Charles Ei Beecher’ ........ se. + ee eee ‘541 “- 91—Scorr: Portrait of John Gb) clatchet.s.« eee . ao nT e a eee 548 ‘* \92—Hovey: Summit of Mont Pelé 2.504. .eage ene aes ee ae 567 r 108 e Soufriére-of Saint Lucia: 3 o.s.veucs 02) aes Wee eee 569 “* 94 x Soufriére of Saint Lucia and boiling lake of Dominica...... 570 ILLUSTRATIONS vil FIGURES BRANNER: Page Figure 1—Bird’s-eye view of the region about Traicgéo and the mouth lee EUnet OT EOL eee nticetere aise wid) AGE ou. Eo Sm + ease r 2—Coastal lakes of the state of Alagéas, Brazil..... ......... 5 FAIRCHILD: Pwore i—Traneation of tributary valleys... we... cee ose eee eee eee 29 * 2—Contour on the Lockport Gaeniaie at the Niagara escarp- SRSAMNE PR Lae bette iaicier te cae «ass Pee kA oo ose bie’ * 3—Typical profiles of the Niagara escarpment .... . .......-. 53 Bs 4—Diagram showing vertical relation of Cayuga and Seneca valleys to Ontario.valley... <0. ...00. <2 tees eeen eee oi 5—Generalized cross-section profiles of Finger Lakes valleys, PeE NICU S ches ee Oh, k wos oe ach ah bawkeweee aed s ‘> 6—Map of Cayuga lake shores Sie Mion A primes: 5s ve ie aes 62 MERRILL: Figure 1—Veins on opposite sides of serpentine block... ... ........ 182 ¥ 2—Asbestiform veins in massive serpentine ............ ee 133 Hopsgs: Figure 1—Sketch map of Spuyten Duyvil creek........ ema tEs, avs oh 157 a 2 RIO ACTORS ELAPlICnY TIVET 0 6s. ca ek cael s Us dbo nw ececeen 158 i$: 3—Section across Harlem river........ .-. RE Teh ie OR Pe ee 159 “8 a ERIN ETORA TAME TOME TIVEE «2 citys 5 owe siainwnr posse va asescese 160 = 5—Section across Harlem river..... ..-..e- 1 i A Migic8 ne i A 161 - 6—Profile of rock beneath Harlem river..............-2-e0005. 161 A, Pectin SOTOSS PIATICRY FIVER. cc odes ev dae ds weep dieleciewes 162 sg S-——Sechion across Harlem river <......0+.- +. 2-5. +) eeeeee 163 7 9—Section across Harlem river at Park (Fourth) avenue... 164 i feb Across LATIGM TIVET. 5 6a: ots.) ede es cele dk hawle wees 164 Po) Si Seetion across Farlemrriver.. st). 2...' ca ko Lk ek be wl cee 165 fe ee ection -merods Harlem TIVEr.. 2. 6 26c deed. Bete ae de cen 165 ** — 138—Section across East river...... .. Ge ets te ee i an 166 See ee OL IGE MOTGRR EAED, FEV GPa. oj. wre waisnivic c's aes ge leew coves 168 pan lo ection ACrORS Fast TIVET.n. <5.) a i oc ven cos cee hlsbenenceas 170 a te erin ReTORs Maat, LIVER ss. cl. .svckae dduses ces weleue ey fp) Pees SEEtION GCTOSS TUARE TIVER. ocds.. 2.00 owas se oe vc oe Se pve eee ECE BehOSs faagh) FEVER 2). ek seasc ie dea ad os www oo wk 173 *« ~ 19—Location of drill holes under New York tower of East River Seger PRTPINLISE At Gay i era ie Sorat ark icc actos Oud ee s's 6 173 See enon Aarne: Neh PIVOT.) occ.\e. ses ass scene Rian ewes’ 174 ** 21—Rock surface beneath the ‘‘ J aa SESE re Sh SUH tee aa me . 175 | eae —weenon across Hndson river. 2.5 0 62S) dod eee 177 os 2s — elon Across Hudson Tivers 60 ie. ioc oe eek wees swe 177 ** 24—Section across Hudson river opposite Fifty- ninth street. .. 178 ARNOLD and Strone: ; Figure 1—Sketch map of southern California...............-...2--5- 184 2—Sketch map of the San Gabriel mountains................. 186 FENNEMAN: Figure 1—Recession more rapid than shifting............. .......00: 208 40 en PTMCERE CA ek ea ee ha kbs eee see esos. 210 * 3—Recession less rapid than shifting .......... oe ee tee 212 " 4—Generalized diagram of shore erosion...... ..........+005- 213 TARR: Figure 1—Diagram illustrating causes of unequal development of mo- Cy Sd 222 Vill BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA TARR: Page Figure 1—Cross-section of Seneca lake............ 2. ccccccccccccecs 231 a 2—Profile along Watkins Glen creek ............... aie Sea 231 73 3—Profile of Hector Falls creek, Seneca valley............ oes ee - MATHEWS: Figure 1—Distribution of formations in Maryland Piedmont and con- tiguoUs areas ...... 265.5, sseeascets ge oe 331 i 2—Sketch of axial lines for major structures of the Piedmont.. 342 Matuews and MILER: Figure 1—Map of vicinity of Warren................... , se elated eters 363 i 2—Map of vicinity of Rockland ..-.. ...4:.. sis 4.40eae see 365 FULLER: Figure 1—Index map of Fishers island. 2.2 2...2.223¢2— Sa 368 A eg Sad map of. Fishers island... 2. uc. c2seaeee eee 369 = 3—Section through hill three-quarters of a mile northeast of eastern end of Isabella island... ........ 5.2. aes) eee 370 ss 4—Generalized section of Fishers island................ ....-- o71 5—General relations of cretaceous beds and the Mannetto and JAIGCCOPTAVEIS o0 86 665k. ess soe we «nce eee 374 Js 6—North-south section across clay pit on Fishers island....... 376 on 7—Northeast-southwest. section along Isabella beach.......... 377 ee 8—North-south section through the bluffs at Isabella beach.... 378 f 9—Section at headland three-quarters of a mile northeast of north end of Isabella beach. ...-- 2.5... Sane 379 ‘* —10—Artificial section exposed in June, 1904, near ——— landing on west side of West Harbor ....1... 380 “* 1i—East-west-section in elay pit......2...: 2) cece see 382 Sranton and Martin: Figure 1—Map of Cook inlet and the Alaska peninsula .............. 394 S 2—Map of part of the west coast of Cook inlet»............... 395 Kemp: Figure 1—Northwest corner of the Port Henry, New York, Guna 413 i 2—One-ninth of a United States ee survey topographic sheet in latitude AA calm oe elect bee ices =e 415 MARSTERS : | Figure 1—Geological map of Belvidere mountain......... ....-.-.... 421 4s 2—Traverse map of Belvidere mountain...... ............... 423 5 3— Cross-section of .a slip-block: 2.730. 2= a... ase Ree 434 Cross and Howe: Figure 1—Explanation of plate 82....... lalb era, testes J 456 e 2—Explanation ‘of plate: $3.2..2. dic (hsetae corer eee Laas ae ‘ 3—Explanation of plate 84 . ......... ae wie mee Pose: re RE 460 aie 4—Explanation of plate 85....... pase a See ta ae Sn 462 PERKINS: Figure oe of the Brandon lignite area. ..04. <5)... sane eee 501 BERKEY: Figure 1—Outline map of the Uinta mountains and related districts.. 518 . 2—Map of region about the headwaters of Duchesne river. . 522 u 3—Generalized cross-section of the south flank of the western Uintas .. cceindin fe Sop train eereeheieeserers fewh'e S ycae 524 (94 plates; 74 figures) ~ PUBLICATIONS OF THE GEOLOGICAL SOCIETY OF AMERICA REGULAR PUBLICATIONS The Society issues a single serial octavo publication entitled BuLLETIN oF THE GroLocicaL Society or America. Thisserial is made up of proceedings and memoirs, the former embracing the records of meetings, with abstracts and short papers, list of Fellows, etcetera, and the latter embracing larger papers accepted for publica- tion. The matter is issued as rapidly as practicable, in covered brochures, which are at once distributed to Fellows and to such exchanges and subscribers as desire the brochure form of distribution. The brochures are arranged for binding in annual volumes, which are elaborately indexed. To this date sixteen, volumes have been published and an index to the first ten volumes. The Buuerin is sold to Fellows of the Society and to the public either in sepa- rate brochures or in complete (unbound) volumes. The prices are as follows: To libraries and to persons residing outside of North America, five dollars ($5.00) per volume; to persons in North America not Fellows of the Society, ten dollars ($10.00) per volume (the same amount as the annual dues of the Fellows) ; to Fel- lows of the Society, a variable amount, depending on the cost of publication. These prices cover cost of transmission to all parts of the globe. No reduction is made todealers. Subscribers should Specify whether they desire the brochures or the completed volume. Orders should be addressed to the Secretary, and drafts and money orders made payable to the Secretary of the Geological Society of America, Rochester, New York. DEscrIpTIoN OF THE PUBLISHED VoLuMES VoLuUMEs. PaGEs. Puates. Figures. Price to FELLOws SL Eel ae 093 + xii 13 51 $4.50 re OIE ee Uk 622 + xiv 23 63 4.50 (EAL eee as 541 + xi 17 72 4.00 (AAS a ns ee a 458 + xi 10 ay) 3.90 Og Ee - 655+ xii 21 43 4.00 LAL ee aaa 528 +x 27 40 4.00 Se a 558 + x 24 61 4.00 ee IhOG ey 446+ x ol 29 4.00 NE Reece sie 460 + x 29 49 4.00 ee MOOR, tose AS. 534 + xii 54 83 4.00 Index to first ten volumes......... 209 2.25 ES ce a 651 +- xii 58 37 4.50 Pee Re seuieew ys oa ee 538 + xii 45 28 4.00 \ ESOS ae ae ee ee 583 + xii 58 7 4.50 oo AE ies Ce ee Pye eee 609 + xii 65 43 4.50 Peis NONE oly oe igh ey 636+ x 59 16 4.50 GIS TG. POUL S Lert eis wy sci 636 + xii 94 74 4.50 ll (ix) x BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA BrocHuRES OF VOLUME 16 PriceEto PrRIcE To BROCHURES. PAGEs. Puatses. FIGURES. Feinows, sem Poet Stone reefs on the northeast coast of / Brasil: Ji CHBRENWER 7s) O22 aoe oe 1-12 I-11 1-2 $0.50 $0.75 Ice erosion theory a fallacy. H. L. ; IW ATR OH MED oie -c faa ee cee Cie ae es 13- 74 12-23 1-6 .90 L325 Hanging valleys. I. C. RusseLu...... 15> DOr eer te ine 10 18 Plumose diabase and palagonite from the Holyoke trapsheet. B.K. Emur- ON eAUROR Reh ©. SRE Bere en opiate A | .* O1-130 24-320 -70 1.05 On the origin of veins of asbestiform serpentine. G. P. MprRILL........; 131-1386 33-34 1-2 10 15 Bearing of some new paleontologic facts on nomenclature and classification of sedimentary formations. H.S. W1- GUA MSS b's) OUP Aer ache ig ens ae ete LSPS he 4, Sek 1 ee ee Origin of the channels surrounding Manhattan island, New York. W.H. ERORBS 3: 1.'S:A ae aeeenamieans steko eee 151-182 30 1-24 .40 .60 Some crystalline rocks of the San Ga- briel mountains, California. R. Ar- NOLD and, Ac Wa SPRONG J.505 sn. ces 1S3=204 - eo Le 2 .20 .30 Effect of cliff erosion on form of contact surfaces. N. M. FENNEMAN......... 205-214. .... 1-4 10 15 Moraines of the Seneca and Cayuga Lake yalléys.'< 2S PARBH Yoo ee ae): 215-228 36 1 20). oe ae Drainage features of central New York. BS) TARE. | Gee ee ine ca tee 229-242 37-42 1-3 .30 45 Pelé and the evolution of the Windward archipelago. nha Essen. eae 2103 243-288 43-47 .... .00 79 Piedmont district of Pennsylvania. F. BAGCOME:. Fe eis tal a eds cash Pe cee ae 289-328 48-64 .... 1.00 1.50 Correlation of Maryland and Pennsy]- vania Piedmont formations. E. B. 4 NG 2 BG Pa a Pon une oman ER Eee NE oct 329-346 ee 2 ee |) .30 Cockeysville marble. E. B. MarHrws aid We) 2 Me: ee ee weet 347-366 65 1- 2 30 45 Geology of Fishers island, New York. De IR UGB tena yee penn eae 367-390 66 1-11 .30 40 Mesozoic section on Cook inlet and Alaska Peninsula. T. W. Sranton and GC. “MUARrin ae con eae 391-410 67-70 1- 2 .35 .50 Geological book-keeping. J. F. Kemp. 411-418 .... 1-2 10 15 Petrography of the amphibolite, ser- pentine, and associated asbestos de- posits of Belvidere mountain, Ver- mont. V.-F. NeARsinRS) = fi ccs 419-446 71-81 1-38 .60 .90 Red beds of southwestern Colorado and their correlation. WuHitTmMaNn Cross and Hens Own... 05). Ps tae 447-498 82-85 Il1- Tertiary lignite of Brandon, Vermont, and its fossils. G. H. Perkrns...... 499-516 86-87 Stratigraphy of the Uinta mountains. ny RRKIRY oy 2 ).'s* oe Ue ake eee 517-630 88-89 1- Proceedings of the Seventeenth Annual Meeting, held at Philadelphia, Penn- sylvania, December 29, 30, and 31, 1904, including proceedings of the Sixth Annual Meeting of the Cordil- leran Section, held at Berkeley, Cali- fornia, December 30 and 31, 1904. H. L. Farrcuinp, Secretary .., ..... 531-636 90-94 ..., 1,20 1.80 (oe) loa a . bo . Or oo Or PUBLICATIONS Xl TRREGULAR PUBLICATIONS In the interest of exact bibliography, the Society takes cognizance of all publi- cations issued wholly or in part under its auspices. Each author of a Memoir receives 30 copies without cost, and is authorized to order any additional number at a slight advance on cost of paper and presswork; and these separate brochures are identical with those of the editions issued and distributed by the Society. Contributors to the proceedings are also authorized to order any number of separate copies of their papers at a slight advance on cost of paperand presswork; but such separates are bibliographically distinct from the brochures issued by the Society. The following separates of parts of volume 16 have been issued: Kditions uniform with the Brochures of the Society. Pages 1- 12, plates 1-11; 280 copies. January 31, 1905. fe te 4a ko 18-93. BOr” “ February 13, 1905. ce ma “FeO; aoe 27, 1905. BET aM aalbee be A a March 10, 1905. mba dag. 10 th 384"-) 23g) 2 6 és 10, 1905. 187-150, 13044 e 15, 1905. oy uift-182, plate’. 353.200).° April 12, 1905. ae Pe Oa 160 3 °¢ Fe 13, 1905. ei 305914, 130) as 14, 1905. Pa eit 267180" re 29, 1905. se -299-24?, plates 37-42; 180 “ ef 29, 1905. meee wag AR AT ep. May 12, 1905. a ene gon. | *. AR G4. 130) sy 27, 1905. « 329-346, 1 _ 29, 1905. ‘¢ 347-366, plate 65s 1a0. <<" n 31, 1905. 's67-390, .. * 66; 80 ‘* June 23, 1905. ‘¢ 391-410, plates 67-70; 230 “ - 24, 1905. eV ALI=418. 230* ‘< August 28, 1905. oe ate-44G60 7 ** 277-81 = 280" October 25, 1905. SR eo IS OS fo 2 December 15, 1905. ee aes -ane) 05 6687 100 5)" oe 18, 1905. Fe aly aee ta BS-89- 230 © ae 20, 1905. Special Editions.t Pages 532-539f 30 copies. February 10, 1906. “< 541-548, plate: 90; 80 ‘* nd 10, 1906. ‘¢ 548-555,“ St ab!" oY 10, 1906. ‘© 555-558, ao. ** “ 10, 1906. ‘¢ 858-560, - | ug 10, 1906. ** 561-562, 2 edie 10, 1906. Page 562, - a al i 10, 1906. pa ees au. te Es 10, 1906. ee * Without covers. + Bearing the imprint [From Bull. Geol. Soc. Am., vol. 16, 1904]. { Fractional pages are sometimes included. Xi BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Pages 564-565, 30 copies. February 10, 1906. ‘¢ 566-571, plates 92-94; 230 ‘ * 10, 1906. Page 572, Mt of 10, 1906. oh om a, OO aie a2 10, 1903. Pages 575-576, S0ih ts He 10, 1906. ‘¢ 877-578, BO). p78 a 10, 1906. Page 579, One re 10, 1906. 4 980, 5020 oS 10, 1906. Pages 581-582, 30 gs - 10, 1906. ‘* 583-585, LOO! Ua a 10, 1906. ‘¢ 586-589, SO: oni ‘G 10, 1906. Page 593, . BO hare be 10, 1906. Pages 595-604, Shae fc 10, 1906. ‘¢ 605-616, SO ae v 10, 1906. CORRECTIONS AND INSERTIONS All contributors to volume 16 have been invited to send corrections and inser- tions to be made in their papers, and the volume has been scanned with some care by the Editor. The following are such corrections and insertions as are deemed worthy of attention: Page 11, footnote; for ‘‘1851” read 1587 and omit ‘‘ 317 years ago.”’ ce 129, plate 30, figure 2, for ‘‘ dentrified” rcad devitrified. 181, in the title; for ‘‘in”’ read of. 133, foot-note, for ‘‘ Ellis” read Ells. 144, line 19 from top; for ‘‘ united by a hyphen” read on top of page 142. 146, line 16 from top; for ‘‘ of” read or. 149, line 14 from top; before the ‘‘:” insert an interrogation point. 157, title of figure 1, for ‘‘Sketch map of ” read Section across. 211, page heading ; for more read less. 287, line 16 from top; for ‘‘important’’ read impotent. 298, line 9 from bottom; for ‘‘ cambro-ordivician ’’ read cambro-ordovician. 299, page heading; for “ cambro-ordivician”’ read cambro-ordovician. 302, line 15 from bottom ; for ‘‘ gniess” read gneiss. 312, plate 57, figure 1; for ‘‘ vein” read rim. 3, plate 58, figure 1; for ‘‘ vein’’ read rim. 2, line 13 from top; for ‘‘ micro-ophitic ” read microphitic. 328, line 4 from bottom; for ‘“‘ pre Cambrian” read pre-Cambrian. ers by > ae ai, a a “ a — ue : ; i S 4 ~ * . e* : , > ' j oie OONEIWVN43ad LV 4d4dY SANOLS SHL L “Id *PO6L 91 “110A "WY ‘OOS "1039 ‘11nd BULLETIN OF THE GEOLOGIGAL SOCIETY OF AMERICA VOL. 16, PP. 1-12, PLS. 1-11 JANUARY 31, 1905 STONE REEFS ON THE NORTHEAST COAST OF BRAZIL *, ANNUAL ADDRESS BY THE PRESIDENT, JOHN CASPER BRANNER (Read before the Society December 29, 1904) CONTENTS Page nn ae CMe aes OP aes, le wees ew eet t Location, origin, and structural features of the reefs ......0 ............00-- 1 Factors in the formation of the reefs...........% 5 OLE PES eta 4 NINE cee Serre Len eg) Saini Bhs, 2 Wide ose 'e os’ SE ee See + Bammer Ce LINE COABL. - 55 cece e se eee Se ae BRR So ciaies hind Snaace 4 SNIET T TIIDETIOT: ic. oo as cle seg swe bee eds A ASEAN Rog eS ee te meniptetny Gf the sireams. 5... 2c. isso eee eee CARPAL ON sn), 6 IIT tr en ees Ot es os La ae bucks ele eee a we 6 IRN Ut be hn ai ig Binte nds oe od, 6X sod ae? Ret Mee cee ate a Chet ae | INERT he) We eid els nice cc kre eens cece ee wns i a EMINMINE ee St ee o L leta ee swe sn 2% telah Weak exe d 8 NEE rege Ne cio wale ec ses cusessece 8 ees cieis Sc winks cme ea eeies as Sei ae pe LE I oa aie ee Rae nea ee aia.” Uniform width of the reefs.......... PORE RoR ty OL i Peete avn eeeeen Oe Cause of the straightness of the reefs.. .... .............. cee Pete Mapes oa ees 0 Stone reef making still in progress....... TE Se SOR aa eae ia ES | Ee ee ae a Pt). Literature of the stone reefs..... ..... ... Sages RS EN ie a a peunrnmn mupoOriauce OF the reeis. el ce ccs tee eee eee mine F4 SENN IIE NITE IMERDO) 25s ese cs Doce ce hs eve ecb eeescoscewaness 2042 INTRODUCTION It is proposed to give very briefly in this paper the results of studies of the stone reefs on the northeast coast of Brazil, which were made in connection with a general geological and geographical investigation of that region. Location, ORIGIN, AND STRUCTURAL FEATURES OF THE REEFS The stone reefs of Brazil are mostly in an out-of-the-way part of the world. They lie aiong the northeast coast of that country, between *A detailed description of these reefs is published in the Bulletin of the Museum of Comparative Zoology, VII, Geological series. I—Butt, Gron, Soc, Am., Von. 16, 1904 (1) 9 j. GC. BRANNER—STONE REEFS ON COAST OF BRAZIL south latitudes 3 degrees 43 minutes and 16 degrees 30. minutes. Only one of these reefs is where it is readily accessible to trans-Atlantic steamers, and that is at the port of Pernambuco. ‘The others are scat- tered along a coast which has no commodious harbors, and are there- fore but little known outside of Brazil. These stone reefs are unique or almost unique geologic phenomena. Darwin, who saw the one at Pernambuco, says of it that he doubts ‘whether in the whole world any other natural structure has so artifi- cial appearance.” Wonder at these structures is greatly increased when one finds them repeated along the coast for more than a thousand miles. Perhaps the clearest idea of them will be conveyed in fewest words if it is stated at the outset that the reefs are lithified marine spits or beaches that have been encroached on from both sides until their edges are more or less broken and angular and their surfaces swept clean of the loose, unconsolidated portions. This theory of their origin appears to be sup- ~ ported by the great bulk of the data collected. Only a few of the more important features of the reefs, however, can be mentioned in this place. Where exposed at the surface the reef rock is sandstone, often almost as hard as a quartzite and ringing under the hammer like a clink-stone, but sometimes, and locally, it is only moderately hard. It contains abundant shells and other remains of animals and calcareous plants, apparently of the same species as those now living in the ocean along- side. The beach sands in the vicinity of the stone reefs invariably con- tain similar remains and abundant small fragments of shells, corals, and of other lime-secreting organisms. Except in the matter of hardness and compactness, the rock of the reef is therefore scarcely distinguishable from the present beaches. The cementing material of the hard rock is lime carbonate, which sometimes contains a little iron also. The rock has a remarkably fresh appearance and the-shells imbedded in it usually retain their bright colors. ; A section across a reef invariably shows the beds dipping toward the ocean at an angle Varying from 2 to 20 degrees ; the lower angles are the more common. In the gross structure no difference is apparent be- tween the beds on the ocean side and those on the landward side. In 1874 three drill holes were put down by the Brazilian government on the stone reef in front of the city of Pernambuco under the direction of the Knglish engineer, Sir John Hawkshaw. The records of these holes furnish the only data we have of the thickness of the hard sand rock and of the nature of the underlying strata. The deepest hole had a depth of 17 meters. | BULL. GEOL. SOC. AM. VOL 16) 190ts Plies SANDSTONE REEF AT RIO GRANDE DO NORTE As seen from a fort on the reef. he seaward dip of the beds is well shown by this reef jaol at} Jo do} uo jood Suo| v St ate J, QNV1 AHL SNIOF 1! SYHSHM 43434 ANOLS OHNILSOSV OLNVS S3dvd JO GN43 HLNOS HeldeevOote- Ob“ 1ONs - "WV “OOS "1035 *11Ng COMPOSITION OF PERNAMBUCO REEF 3 Record of a Boring on the Pernambuco Stone Reef Meters PIMP TGOP POCR. oo. oe wie cw en cs eG Ee ee De 2.95 White sand s. 0.2 fob pees: rR et 3, tte asts.s M22 Shells..... Fi ieee 2 ORE ia ee Ic OSs 84 A OW hg ae 1.10 Gray sand ..... is eat als PS tie Oars Bey it a she Be Arg 18 3) FARMS ET MSR N 503 4 FE a aa sui att ya A Sala hg oa = eee ae egret LE 1 Ue guna eaten tives 05) ga 28 Seats Eee Mae en 2.10 EB Sa PRINS Detar sae hin wie wom 3 1.80 IUCN Mie ee alee ban Xe arshabhese ea eins ose's Fe Patch able. 8 Bi 0.70 Raby Camb 26 A eS etoted ee ye Sy i tae tyne Boe 3.50 White sand...... Sharan SR 8 Bore MES Ee 6 2.20 In another bore the hard reef rock at the top was found to be about 4 meters thick; otherwise the records of the two other holes closely resemble this one. It is not known what is meant in the record by 7 sha BE iy, Wir SS:1 6. « i mad hh X Kea y : <2,° ZiT, 2a a Wa - SEZ , a a i EI TZ No Figure 1.—Bird’s-eye View of the Region about Traicao and the Mouth of Rio Mamanguape. Showing the relations of the stone reef to the shore. “broken rock.” In none of the holes was there any repetition, at greater depth, of the hard top rock; the underlying materials were either clays or they were loose and fragmental. The reefs are off the shore a little way and approximately parallel 4 J. C. BRANNER—STONE REEFS ON COAST OF BRAZIL with it. Usually they have one end connected with the land, at least during low tide, while the free end stands across the mouth of stream, embayment, or estuary; sometimes, however, they are not connected with the shore at all, at any stage of the tide. At high tide they are about flush with the surface of the water, while at low tide they are exposed like long, low, flat-topped walls or breakwaters. In width they vary from a few paces to 450 feet ; in length they are from a few hun- dred feet to 83 miles, the total length being concealed by sand, which covers one end. Commonly they are almost straight; when they curve at all the curves are gentle and only perceptible when one looks along them lengthwise or when they are carefully mapped. These remarkable natural walls of sandstone accompany the shores of northeastern Brazil, with many interruptions, from Ceara to Porto Se- guro, a distance of 1,250 miles. With unimportant exceptions, they do not occur beyond these limits. ) FactTorRs IN THE FORMATION OF THE REEFS IN GENERAL Having stated the broad theory of the origin of these stone reefs, I shall not weary the reader with the tedious process of elimination by which the accepted conclusions were reached, but shall invite attention directly to their history as finally worked out in a study of the problem which has been carried on with various interruptions for nearly thirty years. HISTORY OF THE COAST It will be necessary first to briefly outline the geologic and geographic history of this particular coast. The greater part of the rocks of the coast region are marine sedimentary beds, apparently of Hocene age. — During Miocene times the coast seems to have stood several hundred feet higher than it does at present, and narrow valleys were cut in the Kocene beds, mostly at right angles to the coastline. This period of erosion was followed by a depression, when many new bays were formed by the valleys which lie near the coast. There have been some changes of level since this period, but they have not been great, the evidence nowhere suggesting a movement exceeding 30 or 40 feet. In the mean- time the strong on-shore waves and the in-shore currents rapidly cut away the soft Eocene sediments of the headlands and threw them into the reentrant angles of the coast. This process continued until the mouths of the bays were completely or nearly closed and the entire coast made as nearly straight as it is possible for as long a coast to be. Streams flowing into these narrow bays silted them up from the upper ends: Marine erosion was so vigorous, however, that several such bays were almost completely closed long before the land sediments filled them up. BULL. GEOL. SOC. AM. VOL. 16, 1904. PL. 4 <> om & res Miro ap SS EN oty abe eT cm THE CUNHAHU STONE REEF AT LOW TIDE Jool OY} JO VOC] LOJNO UL ST OUL[ JANS OY, ‘Opt YSLY 4V Joot oY} PUTYOQ ToWVEIS B UOJ UOYR} MOT A 4a35Y4 OYNDSS OLYOd 1 he aii G “Id ‘PO6L ‘91 “1OA "WV ‘OOS ‘1035 "11nd COASTAL LAKES OF ALAGOAS 5 In this way were formed the coastal lakes of the state of Alagéas, which are brackish water lakes lying between watersheds from 200 to 500 feet high and separated from the ocean by low sand spits. THE : COASTAL LAKES OF | THE STATE OF : After the ees Hydrographic Charts j=. Seale: Nee miles SRE a a Se JOE sake sag ies Tava ects fied oo be be ar ne Figure 2.—Coastal Lakes of the State of Alagéas, Brazil. CLIMATE OF THE INTERIOR It is necessary at this point to direct attention to certain climatic and drainage peculiarities of this northeastern corner of Brazil which affect the problem of the stone reefs. If the reefs of northeast Brazil are but little known outside of that country, the climate of the interior along this coast is even less known. Everyone has heard of the great amount of rainfall about the mouth of the Amazon, and every writer on Brazil has something to say of the abundant rains of Rio de Janeiro. At one station of the edge of the plateau, south of Rio, the annual rainfall amounts to 11.7 feet. As the region of the stone reefs lies between Rio and the mouth of the Amazon 6 J. C. BRANNER—STONE REEFS ON COAST OF BRAZIL and is entirely within the tropics, it appears to the casual observer that it must also be a region of great rainfall. Asa matter of fact the country about cape Saint Roque and for hundreds of miles west, southwest, and northwest of there is a region of devastating drouths. PERIODICITY OF THE STREAMS Ordinarily the rains fall during two or three months of the year, but some years the annual rains do not come, and the dry season is drawn out to an entire year, totwo years, or even to five or six years. At such times all vegetation away from streams is parched, the streams disappear, cattle die of thirst, and sometimes the people themselves are compelled to leave the region. When the rains do fall they are frequently torren- tial, and the precipitation even in this region of drouth is in fact much larger than at Para, and many other notoriously wet places. The result of these conditions is that the streams of the region are strongly inter- mittent, some of them being big enough to float an ocean steamer at one time, and completely disappearing during the ordinary oe seasons, to say nothing of years of severe drouths. The Rio Sao Francisco is the only large perennial river along this entire strip of coast, and that river rises, not in the drouth region of north- eastern Brazil, but about latitude 21 degrees south, far to the west and south of the highlands of Minas Geraes and more than 1,000 miles from its mouth. On the maps many other rivers are shown, to be sure, but they are all more or less intermittent. Under such circumstances it fol- lows that the intermittent streams flow boldly into the ocean only during the season of rains, while during the dry season they become so enfeebled that the waves of the ocean throw their own silts and the beach sands back into their mouths, and in some cases complete barriers are thus built between the land water and the sea water. These processes have been in operation along the northeast coast of Brazil ever since the land took on its present approximate form. CLOSED STREAMS ‘My studies of the stone and coral reefs have been carried on partly in small boats and jangadas or rafts, but I have also walked several hun- dred miles along the beach for the purpose of examining them and of seeing their geographic and geologic surroundings. Naturally one tray- eling along the beach would expect to find difficulty in crossing the streams, but asa matter of fact during the longest trip made on foot and at the end of the rainy season only two streams were found that could not be waded in a distance of something over 200 miles. At the mouths of several rivers there was connection between the ocean and the streams only during high tide, and at several other places where streams were to evap % BULL. GEOL. SOC. AM. VOL. 16, 1994, PL. 6 Se esha SAT THE MAMANGUAPE STONE REEF = Taken at low tide from the port on the landward side Ties L ‘4d ‘OGL ‘9L “10A 435Y SANOLS AdVNONVAVAW SHL NO 30IL MOT “WV “OOS *1039D *11Nd CLOSED STREAMS 7 be expected, there were what are known in that country as “ rios tapa- dos” or shut up streams—that is, streams whose mouths are completely closed by bars across them. At one place a stream was waded that was cutting through asand bar and running from the ocean landward ; at two places gaps were found that had been cut in this way, but as they were passed at low tide, there was no water flowing through the breaks. The disposition of the sand, however, left no doubt about the water having flowed from the ocean toward the stream. VEGETATION Another important fact is that the region under consideration is in the tropics, and wherever there is fresh water, vegetation isrank. The banks of streams are everywhere overgrown with a dense jungle, bodies of fresh water are covered and filled with aquatic plants, while the densest of man- grove swamps cover the tide flats of the region of salt and brackish water. | DENSITY OF THE SEA WATER The other element of the problem is the density of the ocean water along the part of the Brazilian coast on which the stone reefs occur. In volume I of the Challenger reports, “ Physics and chemistry,” data have been brought together and a chart constructed showing the oceanic areas of different densities of the surface waters. This chart shows that the areas of the highest densities are in the Red sea and the Mediterranean sea; these, however, are land-locked basins. The highest densities in the open ocean form two areas in the Atlantic: one being near the mid- dle of the ocean, lying between northern Africa and the West Indies; the other being in the south Atlantic and hugging the coast of Brazil from cape Saint Roque to south of Rio de Janeiro. It is worthy of note that the area of high Atlantic density as repre- sented in the Challenger chart does not quite fit the area of the stone reefs. ‘The map seems to show the dense area to be too far to the south to suit the theory of the reefs here put forward. This is probably due in part to a lack of density data, especially for the dry season, along the coast of Brazil northwest of cape Saint Roque, and extending half way from there to Para. The southern limit of the high density lies south of Rio de Janeiro and far beyond the southern limit of the stone reefs. The absence of reefs in this direction is readily explained by the fact that rainfall on this part of the coast is very much larger than it is far- ther north, and the streams are therefore able to keep their mouths open. In connection with the subject of climatic conditions it should be noted that the long dry season which enfeebles the streams and permits the waves to dam them back must likewise be the season of the highest den- sity of the sea water. The South Atlantic equatorial currents flow west- 8 J.C. BRANNER—STONE REEFS ON COAST OF BRAZIL ward from near the coast of Africa and split on cape Saint Roque. In the season of drouths the surface density must be considerably increased, especially near the land and the shallow continental shelf. | Attention is called to the absence of reefs between Bahia and the mouth of Rio Sao Francisco. This is attributed to the influence of the large body of fresh water discharged by the Sao Francisco. The reefs begin a short way north of the mouth of that stream, but the currents set southward and shoreward along this part of the coast. This permits the formation of reefs north and prevents it south of the stream. CALCAREOUS SANDS One more factor is this: The ocean alongside is warm and teems with marine tropical life. There are many coral reefs along the coast, and everywhere are shell-bearing mollusca, ichinoderms, crinoids, crustacea, worms, corals, calcareous alge, and other lime-secreting organisms. All of these organisms contribute abundantly to the beach sands. THE CoursE oF EVENTS With this data in hand, we may now observe the course of events on the coast under consideration. We have an old coastline with long, nearly straight, sandy beaches. The sands are rather coarse and are commonly mixed with fragments of calcareous skeletons of the animals and plants living in the ocean. Across old embayments the sands have been thrown back landward by the waves until in many places the weak drainage is partly or entirely shut off from the ocean by banks of sand. In and about the pools, lakes, or sluggish streams thus formed, abundant aquatic and semi-aquatic plants live and die. ‘The fresh water is thus rendered acid by the presence of large quantities of carbon dioxide produced by organic decomposition. The acid water on the land side percolating through the embankment of sand at low tide attacks the calcareous matter in the sand and passes seaward with it in solution, but as it comes in contact with the dense sea-water on its way through the sand, the lime carbonate in solution is deposited in the interstices between the sand grains. In time the interstices are com- pletely filled, and the sand bank is hardened and so solidified that the water can no longer soak through it. ‘The process must then of a neces- sity come to a halt at that particular place, and percolating waters must either seek other loose sands or they must be turned aside by the now hardened and impervious spit and compelled to flow through the open channels. The percolation of fresh and salt waters would be very much the same whether the fresh waters to landward were completely damned in or were separated from the sea by a long spit, around which it had to flow at ebb tide. Thus the streams by the help of the sea build of BULL. GEOL. SOC. AM. VOL. 16. 1904, PL. 8 SAND SPIT AT TRAICAO ENCROACHED ON BY THE SEA There is a lake to the left of the spit | Aa ee, | An 45, ee . ‘ y ‘ . 7 i ’ d é * iPad ee eee a ee eo eee eT ea 33 eR ete ts ee Ler NS a ian tthe ohio wn tie Sew eens Sees 34 aL UERRNE gM, tid Ok Ne too doh sin + < hkie'n eleta ks Sew 30 IR De ear hohe EVEN NEN ie Geo cis oy ks Pale ae wine oe ens 39 NE See ere den ase od gos ol een S's, ien'sii ea Walaa ss Shae wee 41 SN UTRECRIESIONE im Se gh de eae st cca eek ones Lek ee ealc cs 47 Part II. Ice-sheet erosion in New York............. FEY Ber aS es Ee sls 48 es 5 CS eee a ee ee 48 Effects in Adirondack region ; ieee SESS i oe eee See ee as ee at 50 Work of the Ontarian lobe; western New York....... ..........0e208- 51 RM eM PISO 1G 5 cos re, uw nig apt ee mec one vege e's 51 RENEE EERE RAMEE og oo meas Vie ones we easel wanes ic 8 gl a Pah ee a 54 Effects in the Finger Lakes region; central New York....... Veettaenn sr 5d mesa MRUE CRE ETRUML ISIE, ho Siac d 20.46 bio ss wh teccke Oa Pay Welele a ole Sas IAS 342% 55 Adverse argument from observed phenomena................+.24-- 57 CSR RIM ARS OT ICO EROOTY oii fin einen etic ele rele oee SRE WY Stagnation of lower ice in the deep valleys...................... 57 Lobations of the ice front in the valleys................cceseeee 58 * The second part of this paper was présented in the abstract at the Saint Louis meeting. The theoretical discussion, part 1, was read in brief at the Philadelphia meeting. IlI—Butu. Geo. Soc. Am., Vou. 16, 1904 (13) 14 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY Page Effect of deep waters facing the ice................ ..- eae eee 59 No erosion in front of zone of deposition... ..2...-.syaaemere 59 Absence of moraines in the ‘basims...........5 ee see eee Pe. Small volume of the valley-heads moraine..... .............-8. 60 Idea of rock basins an assumption:.:.-....:. (> 2 he eee a ee Convexity of valley sides. )...000005.55% 42. Pm ei aS 2 61 Rock cliffs .....)03.cc.4 ons cae Regs casei cee ee 62 Islands and capes of rock in Cayuga basin... .......2.00=eeiee 63 Direction of ice flow due to the topography ............. bie ee 63 Transverse valleya.. .isca%. is ois. vive 6.08 vie deb. » ble ed 63 Adverse philosophy i: 2. ..nckie es ies oe ee Gk eee x) of Direct proofs-of monseresion i..5.1-45. «+ 229%. ao ee ee afr aad 64 History of the valleys... i.0...5. ede a ad 2 ae eed aia 66 PE rembacial COnagiwoOny 3 4.45% oa oh se claus ae ee oe eee ae ose ae Preglacial’drainage i... 00.502 cee eet oe se sm oe een 66 Tee invasions. ...° 202.5 yee Sie ls Oo we whee a0 o's ounce gia ee 66 Eifects of ice advanee:. 246008.) oa dose = oe 2 67 Conditions during maximum extent of the ice sheet.. .......... 67 Valley-heads moraine..cicgs0. 20a ees saa! 6,8 cn ahs Nee 68 Effects/of ice retreats tcc. oat. vs eel ener aw dala epee ci eee ~~ as Local glacial lakesi.. oo22 0. cs sve a nen g 7 ap om 2 oa ae 69 Drumlin-belt filling.: 605 0) a8. on ewe bes. 0) eee er 70 Buried walleye. cs ea o's 2 ects vicy wnle is ine aries ee ae Sieben 70 Greater elacial lakes. )..0 esos eiieas: ace ae vee: sce 71 Present Jakes. os oc5.ca. sede one oat Ciidses tent. 1 72 General .sammaary. . 2 Soscicn foe ee aE elec Pade «alas ee 73 Parr I—GENERAL J'HEORETICAL DISCUSSION INTRODUCTION Divergence of opinion.—Probably there is no subject in geology on which the divergence of opinion is so great while at the same time the observational material is so ample as that of glacial erosion. A few geologists believe that glaciers have excavated valleys and basins thou- sands of feet deep in solid rock, while others think that ice erosion has been inconsequential. There was a time when Joseph Le Conte and John Muir held the opinion that Yosemite valley had been excavated by Sierran glaciers, while J. D. Whitney was not convinced that glacier ice had ever even occupied the valley. ‘Today most students of living glaciers and glacial work deny that glaciers possess great erosive power, while a group of physiographers claim that the peculiar features of Nor- wegian, Alaskan, and other deep valleys, including those of the New York “Finger” lakes, are due to great deepening by ice excavation. With such an abundance of concrete evidence available, it would seem 2 : : PURPOSE OF THE PAPER 15 as if the extreme differences of opinion and the sudden and radical shifting of views were not creditable to the science. The work of recent glaciers in varied forms is conspicuous over vast territories, and living glaciers are available for comparison in large numbers and of various classes and conditions. This diversity and changing of views suggests a psychological phenomenon more interesting perhaps than the glacial. It implies either some unusual, peculiar, and indefinite character of the glacial phenomena, or failure in accurate observation, or faulty interpre- tation of the facts. : Purpose of the paper.—The original plan of this writing was to present the arguments and conclusive facts against great ice erosion in central and western New York. It was, however, found desirable to preface the discussion with a review of the general theoretic problem. The paper has therefore been expanded into a general consideration of the whole problem, and will try to aid in establishing a sound conclusion on this much-disputed subject. The truth is desirable, even though it destroys the pleasure of disputation. | Use of the term “‘ erosion.”’—One of the sources of disagreement in this study has been the failure to discriminate the several forms of ice erosion, to recognize its limitations, and to use a distinctive terminology. It will _ be necessary not merely to save circumlocution, but as aid in clear think- ing, to make distinctions and to explain the use of terms. It is conceded that glaciers have power to gather up and carry along loose material which they find in their way, and that they frequently, but not always, exercise that power. It is conceded that they can push away loosened blocks which obstruct their paths,and may thus to some extent cut down opposing cliffs or ridges; also that they can “ pluck” or pull away loosened blocks, under some conditions. It is also granted that glacial ice may lift and bear away layers of stratified rocks which previous weathering has detached or loosened. It may be conceded that at the feeding grounds of alpine glaciers the valley heads may be ex- panded into cirques and the walls steepened, by the aid of weathering, through the pull of the ice on the frost-loosened blocks, specially where facilitated by vertical jointing. It may even be conceded that the ice may rend or tear away the weathered material so as to leave an irregular surface with shallow rock-basins, specially in crystalline rocks; and as a further concession it may be admitted that alpine glaciers can possibly do some basin-making work at the foot of steep slopes or cascades of the ice in unobstructed valleys, analogous to the basins made by the plunge of cataracts. In brief, it is granted that glaciers may produce com- paratively small and shallow rock-basins in weathered crystallines, and rarely in mountain valleys, while it is generally recognized that glaciers 16 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY have the power, which they use capriciously, of clearing valley, slope, or plain of weathered or loose material. They change geest and alluvium into drift. But these effects are not ‘‘ erosion” in the full sense of the word as commonly used. If the advocates of “ glacial erosion ” limited the erosional work of ice to the operations mentioned above, there would be little occasion for any disagreement. It is evident with a very little thought that long and deep valleys and large lake basins are not pro- duced by any of these operations, but that they would require vast and deep excavation in solid or “live” rock by the ordinary flow of the gla- cier in its open course. Itmust be admitted that the production of lake basins like those of Chelan, or Cayuga, or valleys like the Norwegian fiords imply an erosion not only vastly different in degree from the work of glaciers conceded above, but essentially different in kind or quality. The failure to make this discrimination and to recognize the limitations of ice-work is the cause of much misconception in this matter. The term ‘‘ erosion ” will be used in this paper to signify the removal by the ice of firm, hard, unweathered bed-rock. For the admitted and competent work of ice other and self-explanatory terms will be used. Character of the argument for erosion.—The argument for extreme ice erosion is founded in analogy and is almost wholly inferential. It is mainly an assumption, resting on the argumentative method of exclu- sion. Some topographic forms, the precise genesis of which is not clearly explained, are assumed to be the product of ice erosion because ice was the last occupant of the area. The strength of this claim on the thought of students is due to their failure to discriminate the kinds of ice-work and to recognize its limitations. Geological text-books and treatises discuss the topic only in a general and indefinite way, and with more or less qualification seem to approve the conception, or to at least admit the possibility, of deep erosion by glaciers, while the leading text-books in physical geography positively affirm deep ice erosion. Many American geologists who are compelled by the concrete evidence to deny great erosional power to continental glaciers still have been will- ing te admit that alpine glaciers might produce deep cutting. This is partly a concession to the extravagant claims of the erosionists and partly due to the teaching of the text-books. Probably most geologists of the later decades began their work with an indefinite belief in the glacial origin of some lake basins, valleys, or fiords, and if they have come to ’ new and different opinions it has been through outgrowth of the old views. The writer belongs in this class. At the Minneapolis meeting of the American Association for the Advancement of Science, in 1888, there was a spirited colloquy on this topic between J. 8. Newberry and J.P. Lesley. The former held that even the basins of the Great lakes ve LITERATURE AND SUMMARy OF OPINION iy were partly, if not wholly (in the case of Erie and Ontario), the effect of glacial excavation. With a resounding blow on the table, Lesley said that “ice has no eroding power.” ‘There was a warm debate on a cold topic. To the writer Newberry was a demi-god and Lesley was a very impious mortal, but with the passing of time the conviction has come that Lesley was essentially right. Many readers of these lines can probably give similar experience. Titerature and summary of opinion.—The opinions and arguments relat- ing to this subject are scattered through an immense volume of glacial and physiographic literature, and it seems undesirable to cumber these pages with a host of references. An excellent digest of the literature on the subject was published by Culver in 1895.* In that paper it was _ shown that from the time of Ramsay, 1862, who was responsible for giv- ing currency to the idea of glacial origin of lake basins, up to 1895 very few glacialists of note were advocates for extreme glacial erosion. On the other hand, the vast majority of students of glaciers and glacial phe- nomena had been emphatic in denying effective erosion to glacial ice. Among these were Bonney, Judd, Neumayer, Heim, Forel, Whymper, and most American geologists who had spent much time on living gla- ciers. Culver’s quotation of opinions shows this striking fact, that the judgment adverse to ice erosion was almost entirely based on direct ob- servation and ftield-study of glacial phenomena, while the opinions favor- ing great erosion were abstract and theoretic, relying chiefly on physio- graphic features. The same division of opinion is true today, and the fact will be emphasized later. In 1900 Turner f summarized some of the opinions relating to ice ero- sion as bearing specially on the problem in the Sierras. In the same year Davis { published a review of previous writings on “ hanging ”’ val- leys with an appended bibliographic list. References to the literature bearing on the origin of the Great lakes and the valleys and lakes of New York up to 1900 may be found in two publications by Tarr.§ Some references to writings on special areas or particular localities will be given in the following pages. THEORETICAL DISCUSSION Burden of proof.—The argument for deep rock erosion by glacial ice is *G. E. Culver: “ The erosive action of ice.’’ Trans. Wis. Acad. Sci., Arts and Letters, vol. x, 1895, pp. 339-366. 7H. W. Turner: ‘The Pleistocene geoiogy of the south central Nevada with especial reference to the origin of Yosemite valley.” Proc. Cal. Acad. Sci., 3d ser., vol. i, no. 9, 1900, pp. 261-321. 1 W. M. Davis: ‘*Glacial erosion in France, Switzerland, and Norway.’’ Proc. Boston Soc. Nat. Hist., vol. 29, no. 14, July, 1900, pp. 273-322. ¢R.S. Tarr: ‘‘ Lake Cayuga a rock basin.’ Bull. Geol. Soc. Am., vol. 5, 1894, pp. 339-356. —— ‘The physical geography of New York state,” 1902, pp. 179, 227-234. 18 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY mainly based on analogy, observation showing that glaciers clear their channels, abrade their rock beds, and sometimes modify the form of their valleys. No limit to the amount of such erosion has been deter- mined, and hence an indefinite amount, equal to any requirement, is assumed. The burden of proof has thus been tacitly thrown on oppo- nents of the assumption, and the latter have timidly admitted a doubt or conceded the possibility. The mere admission of the possibility of deep ice erosion gives the argument from analogy full opportunity and places the objector on the defensive. The opponents of the conception have weakly rested in the attitude of defense instead of challenging the production of positive evidence. The phenomena of glaciation are so abundant that the advocates of wholesale glacial erosion should be asked to prove, first, the competency of the agent, and, second, the fact in the particular case. Argument for erosion. sa fs the absence of a complete theoretical state- © ment of the positive argument for deep glacial erosion the writer will try to supply one, as follows: (1) Glaciated rock surfaces are proof of abrasion by rock-armed ice. With sufficient intensity of the factors involved and alarge time element great effects could be produced. (2) It has been observed that the ice can “ pluck” or grasp and carry away masses of bed-rock. (3) Glacier ice behaves in a capricious manner, scoring hard rock in some places, overriding and leaving soft materials in other places. This suggests that intensified effects may locally be very great. (4) The analogy between rivers and Alpine glaciers suggests that glaciers may do excavating work comparable to that of rivers. (5) The conspicuous milky sediment of glacier waters not only is a proof of mechanical abrasion by the ice, but gives a volumetric measure of that abrasion, often of large amount. (6) The presence of a large constituent of calcium carbonate in the drift is evidence of large mechanical destruction of the rocks. (7) The enormous quantity of drift spread over the glaciated areas implies great erosion to supply the material. (8) The existence of numerous lakes only in glaciated areas proves their relationship to glaciation. (9) The attitude, form, and structure of many lake basins which he in the path of former glaciers (for example, lakes Cayuga, Seneca, Chelan) are more readily explained by and argue strongly for glacial excavation. (10) As some lake basins of the class noted above, under (9), have not been explained by the work of aqueous agencies, it is proper to appeal to ice action. SPECIFIC DISCUSSION 19 (11) The occurrence of rock-walled basins marked with ice-scorings, proves that ice does excavate basins. Cirques are admittedly a quarrying work of ice in combination with weathering. (12) The topographic features known as “ hanging” valleys are not recognized as normal product of weathering and stream work, but are well explained by deep ice erosion in the main valleys. Specific discussion.—To the layman, and even to many geologists, the points presented above would seem to establish a good case for the pos- sibility, if not the actual occurrence, of ice-excavated valleys and basins. They are certainly admirably adapted to serve as the foundation for general assertion and assumption, since their indefiniteness is good defense against direct attack. We will first consider the several points _ separately and then discuss the matter as a whole. (1) To produce important erosional effects by abrasion, or the rasping process, exhibited on glaciated rock surfaces, would require a vastly greater length of time than can be conceded. Probably millions of years would be required to accomplish any considerable amount of valley-cutting. The removal of rock by the slow process of glacial abra- sion is so ineffective that it is practically a negligible factor in ice erosion. The smoothing, polishing, or sandpapering of rock surfaces is rather an argument against deep erosion, as it is such a slow process that it is inconsistent with great excavation. Its impotency as a factor in valley- cutting is proven by the freshly exposed beds of alpine glaciers where the amount of erosion has been just about sufficient to prove that the ice has been there. And yet these were vigorous glaciers, which have been attacking their channels several thousands of years longer than those glaciers which are assumed to have cut valleys hundreds or thou- sands of feet deep. The glacial strize themselves supply one of the clearest proofs of the slow and ineffective character of abrasion. Cross-strise are very common phenomena, and may indicate different movements of the ice-body and not merely varying currents. This certainly proves the weakness of the later abrasion, for if general abrasion were such an effective process as to cut hundreds of feet into crystalline rocks during the Pleistocene period the rock should be removed so rapidly that double sets of strize would be rare phenomena. The deep groovings and cornice-like flutings, such as shown on Kel- leys Island, lake Erie, are by their rare occurrence good evidence of the weakness of glacial erosion. If abrasion were effective, such planing should be common. The curving channels evidently were not wholly made by the ice, and the straight, cornice-like flutings represent some rare combination of special and unusual conditions. It is illogical to 20 _#H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY use these phenomena as argument for great erosion, as their rarity is proof that the ice did not cut deeply or rapidly as a habit. Chatter-marks, the crescentric fractures crossing a line of gouging and concave downstream, are also evidence of the ineffectiveness of the gla- cial plane, not only by their rarity, but by proving that the ice did not hold its tools up stiffly to their work. Roche moutonnée forms prove that the glacier acts as a flexible rasp and notasa plane. The two factors, pressure and rapid flow, which should cooperate in order to produce ~ - rapid erosion, do not generally increase together. The reason for this lies in the peculiar mechanics of the glacier, which will be considered later (see page 25), along with a general discussion of the element of abrasion in its wider relation. (2) “ Plucking” is a term which has done much service in the cause of the erosion argument. Doubtless the ice can grasp and pull or push away projecting blocks that have become loosened by weathering or which are much exposed in saliences. The amphitheaters or cirques at the heads of glaciers are thought to be made by the pull of the ice on the frost-loosened blocks, specially in crystalline rocks with vertical jointing. Conditions somewhat similar may exist alongside the stream glacier, but the effect here is to widen and not to deepen the valley. The mechanical conditions at the bottom of the channel or beneath the glacier are entirely different. There the effect of the ice is to abrade away the saliences, and plucking could be only a part of the initial and temporary process of leveling. The idea that plucking has any part in the deepening of large valleys has no warrant in either observation or reason. An operation analogous to plucking, and which may be included under that term, occurs in the case of bedded rocks which have been weathered so as to produce open or weak horizontal joints. Such lifting of weath- ered strata is illustrated in plate 23. It can effectively occur only to the depth of weathering, and the discovery of it may be regarded as evidence that the ice has not eroded below the zone of weathered mate- rial. - This kind of work is more common in case of continental glaciers, and specially along the crests of escarpments or elevations. It is con- ceivakle that a stream glacier might bear such peculiar relation to the attitude of bedded rocks in its channel that, along with exceedingly slow action of water or weathering beneath the ice, it might slowly pluck at its channel bottom; but this is not important; and, as a mat- ter of fact, most of the deep valleys and fiords which have been attrib- uted to ice erosion are in crystalline rocks. (3) Capricious behavior of the ice within limits may be admitted ; but the admission that the ice acts differently under imperceptible or unknown SPECIFIC DISCUSSION at differences of condition does not justify any claim for extreme erosion. Such claim must be proved by direct evidence. (4) The correspondence between rivers and glaciers has been exagger- ated. It was a valuable idea in the early study of glaciers, while it was necessary to prove that alpine glaciers had motion. To the extent that the glacier has apparent viscosity and flow, the differential movement simulates that of streams; but there the correspondence ends, except that the glacier is in general an aqueous agent, conveying rock rubbish from higher to lower levels. In its manner of erosion, transportation, and deposition, the glacier is quite unlike ariver. Its transporting power seems to be equal to any imposed burden ; in other words, the glacier is never full-loaded. But the heavy loads of glaciers are not derived from erosion by the ice itself; but are contributed by weathering effects above the glacier, either at the valley head or alongside the valley by atmos- pheric action on the valley walls. To the extent that glacier-flow obeys the same law as river-flow, the rate of motion at sides and bottom is re- duced,and becomes so slow as to be an inconsequential element of erosion during the life of any glacier. In another respect the glacier is unlike the river. In the bed of the latter chemical decay is effective and the river bed is subject to disintegration ; and the coarse burden of the river is all carried at the bottom. In the bed of the glacier the disintegrating forces are ata minimum. The work of subglacial streams and the fur- ther general discussion will be found later. (5) The fine rock-dust which gives the milky color to the water from the glacier—the “ Gletschermilch ”—is admittedly the product of the glacier mill, but it is only partly derived from the abrasion of the bed- rock. It is also produced by the grinding of the transported rock rub- bish, as is shown by the worn and striated character of the boulders and pebbles in the till. It is optically conspicuous, not for its volume, but because it is so fine that it floats in the water. The argument for rapid glacial erosion, made by Helland, based on the gletschermilch, has been shown by Heim * to be fallacious, and that the large estimates are mis- leading. He shows that the amount of detritus washed annually from the alpine glaciers is equal to only a small fraction of the amount of detritus carried by the non-glacial streams of the same region. As com- pared with streams in equal time, glaciers are weak in erosive power. Properly interpreted, the milkiness of the glacier is a proof of its slight abrading effect. The estimates of the erosion by Muir glacier based on the amount of sediment held in the tidal waters of the inlet can be of little value? for * Albrecht Heim: ‘‘ Handbuch der Gletscherkunde.” 7 H. F. Reid: “Glacier bay and its glaciers,”’ Sixteenth Ann. Rep. U. S. Geol. Survey, 1896, pp. 154-458. G. F. Wright; ‘‘ The ice age in North America,” 4th ed., p. 64, IV—Butt. Geot, Soc. Am., Vou. 16, 1904 22 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY several reasons. The waters of Muir inlet have a depth of 300 feet and a tidal range of 23 feet; this, with the disturbance of the waters and of the bottom sediment, due to the calving of icebergs, makes the amount of sediment held in suspension at the front of the glacier no criterion for the sediment in the glacier drainage. Moreover, even if we had the true measure of the gletschermilch, it would be of no value as a measure of the Muir glacier erosion, since the lower part of the glacier for an unknown distance is overriding gravels. Besides all this is the fact that the sedi- ment washed from this or any other glacier is derived from the grinding of the stones within the glacier as well as the scratching on the bottom, and also from whatever rain and stream-wash there may be on the valley walls. | (6) The lime carbonate in the drift is doubtless due to mechanical action on calcareous rocks; but it is inconsiderate and unwarranted to attribute it entirely to abrasion of the channel walls. It represents in the main, probably, the trituration of the limestone fragments which the glacier was transporting, and which had been made ready for the ice-mill by the millions of years of pre-Pleistocene weathering. The great quan- tities of striated limestone fragments in the drift immediately southward of limestone outcrops show the main source of the carbonate in the till. (7) The volume of the glacial drift has been cited as the measure and proof of great glacial erosion. This is decidedly untrue and misleading. More justly it is to be regarded as merely the measure of the transporta- tion by the glacier. The mass of drift, no matter how great, is no proof whatever of any erosion at all of sound rock. The assumption fails to recognize the enormous supply of loose material which the ice found ready to its grasp. During the millions of years of the Mesozoic and Tertiary the northern lands were exposed to active weathering agencies under climatic conditions probably similar to those prevailing today in the middle and southern United States. We may well believe that the mantle of geest over all the glaciated areas of the northern hemisphere was fairly comparable to that which covers our southern lands today. This mantle was largely removed from the central areas of glaciation and was piled over the regions where the ice margins lingered. The amount of drift over any glaciated territory is very easily overestimated, and the mental bias of the writer must be noted. For example, the estimates of vast glacial erosion of Scandinavia based on the enormous amount of drift over the German lowlands and elsewhere, and all as- sumed to be derived from Scandinavia, is most certainly valueless. Morainal belts and valley fillings are the conspicuous. masses of the drift and strike the attention, but they are relatively local and should not be taken as typical of the general drift sheet. ee oe SPECIFIC DISCUSSION 23 It is very difficult to make any quantitative comparisons which would be more than mere guesses, but it seems probable that the whole volume of drift left by any ice body would fall far below a fair estimate of the amount of alluvium, geest, soil, and loosened rock which the ice body found on its territory. In any estimate of the volume of drift we should not fail to add that seized by the rivers and borne seaward, the loess mantle over great areas, and the matter carried away in solution. It will generally be found that the expert students of the drift make the more moderate claims for its volume. (8) The occurrence of tens of thousands of lakes in the glaciated areas is a striking fact when placed in comparison with their rarity in south- ern districts. There is no doubt of their being related to glaciation. The physiographic explanation is that they represent the infantile and youthful stages of the reconstructed drainage left by the interfering ice sheet. The geologic explanation is that they are chiefly morainal—that is, they occupy depressions in the irregularly piled drift or lie in valleys which are blocked by drift fillings. Some of the lakelets are in shallow rock-basins, which will be considered later. | Some of the larger lake-basins, like those of central New York, are complex in their form, structure, and relations and are not readily or positively explained as to their precise origin or genetic relationship without sufficient deep borings to show the preglacial topography. The slopes of some of these valleys have been smoothed by the glacial rubbing, and it is not surprising that they were once thought to be the preduct of glacial excavation. Naturally those valleys which lie in the direction of the glacial flow have been most modified by the ice action. That the ice has smoothed the valley slopes to some extent and swept away the talus accumulations and other loose materials is very prob- able, but to claim more than this for the ice-work is an assumption without any sufficient basis in geologic facts. The subject is discussed on pages 55-65. (9) and (10) Because some of the Swiss and Italian lakes, and the “ Finger” lakes of New York could not be at once fully explained in the causality of their attitude, depth, form, relation, etcetera, under non- glacial agencies, they have been regarded as illustrations of deep glacial erosion. Such lake-basins were, indeed, the first suggestion, and have been one main argument, for extreme ice erosion. Even recently some physiographers appeal to glacial action and postulate a thousand feet of cutting by ice because their rules and principles of topographic evolution do not immediately explain the peculiarity of the topography. Tacitly, it has been expected that one denying the theory of ice-cutting must disprove it; but the burden of proof belongs on the advocates of ero- 24 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY sion, who should first prove that it is even possible for a glacier to cuta valley or large basin somewhere, some time, or under some conditions. There are no facts of observation from any living glaciers, nor from any dead glaciers, which justify the idea that they can cut valleys or deep basins in live rock. The sufficient negative proof is given below. (11) There seems to be sufficient testimony to the occurrence of small or shallow rock-basins, in crystalline rocks, in the central areas of glacia- tion; also in cirques or valley heads; and perhaps in the course of mountain valleys at the foot of steeper slopes. Shallow basins in crys- talline rocks are to be expected, as a product of rock-weathering and glacial removal of the loosened material. This was recognized long ago by Pumpelly.* Because a rock-rimmed basin is found to bear glacial strize on its borders the conclusion does not necessarily follow, logically; that the basin was cut out of the solid rock by ice erosion, although this is commonly assumed. The striz merely prove that glacial ice has . passed over the rock. However, since weathering and stream action do not normally produce basing, it is a legitimate theory that ice really had some function in the forming of the basing; but since there is abundant proof from living glaciers that ice does not cut basins out of solid rock as a habit,it is much more reasonable and scientific to conclude that the ice has usually only developed a structure or form partly due to differen- tial weathering. The production of small or shallow rock-basins in mountain valleys ~ may be fully admitted without really touching the problem of large lake basins like Chelan or Cayuga. Both quantitatively and in principle the genetic process is entirely different. For example, a river can excavate a plunge-basin at the foot of a cataract, but this gives no warrant for assuming that the river can cut a great basin in its graded valley by its - ordinary flow. The same statement applies to cirques.{ These are a product of an exceptional process of quarrying, due to combination of frost-work and ice-pull at the bergschrund or along deep crevasses. The principle has no application to ice abrasion beneath the glacier in its ordinary flow. The forces involved in cirque-making may have some play alongside the *R. Pumpelly: ‘The relation of rock-disintegration to loess, glacial drift,and rock-basins.” Amer. Jour. Sci., vol. 17, 1879, pp. 133-144. +It is desirable that reliable observations should be directed to the determination of the question whether undoubtedly glaciated rock-basins occur in sedimentary rocks. { For the study of cirques consult specially the following : W. D. Johnson: On cirques. Science, new ser., vol. ix, pp. 112-1138. F. BE. Matthes: ‘‘Glacial sculpture of the Bighorn mountains, Wyoming.’ Twenty-first Ann. Rept. U.S. Geol. Survey, 1899-1900, pt. ii, p. 167, et seq. H. W. Turner: ‘‘ Pleistocene geology of the south central Sierra Nevada, with especial refer- ence to the origin of Yosemite valley.’’ Proc. Calif. Acad. Sci., 3d ser., Geol., vol. i, p. 289, et seq. A. ©. Lawson: ‘‘Geomorphogeny of the upper Kern basin.’’ Univ. of Calif. Pub., Bull. Dept. Geol., vol. 3, 1904, no. 15, pp, 291-376. (Particularly page 358.) SPECIFIC DISCUSSION 25 glacier in the process of valley widening, or in the production of the theoretical U-form of glaciated valleys. (12) “ Hanging” valleys have been the occasion in later years for pos- tulating deep valley-cutting by stream glaciers. The assumption is made of valley deepening in hard, crystalline rocks, of hundreds and even thousands of feet, simply to explain a discordance of valleys. The phy- siographic argument is based, in turn, on another assumption, that moun- tain valleys, as those of Norway and Alaska, the Alps and Sierras, should, under atmospheric and aqueous agencies, be graded to an accordant sys- tem. With the highest personal regard and admiration for the masters in physiographic science who have advanced the above ideas, the writer yet believes that their conclusions are wrong. The genesis of topographic forms involves a number of agencies which in their relative effects are indeterminate. The students of these very theoretical and complex problems should not expect to find the full explanation at once for all puzzling features; neither should they resort to short cuts which are mere assumptions. Discordant valleys will be found to have some rational explanation in non-glacial processes, probably as a natural stage of drain- age under exceptional conditions. The assumption of enormous ice excavation does not seem warranted. The further discussion of this in reference to certain mountain districts and to central New York topog- raphy will be found later in this paper. General negative argument.—Several elements or factors involved in the mechanics of glaciers are decidedly unfavorable to great erosion, and none are positively favorable. From the study of glacial phenomena for fifty years the following principles of glacier physics seem to be well established : (a) In the ordinary normal flow, the upper part of the glacier moves faster than the lower or bottom portions. * (b) The fluency of the ice diminishes in proportion to the amount of commingled rock-debris. + (c) Horizontal shearing occurs, the upper, more rapidly moving, layers sliding over the lower, basal, and laggard layers. f (d) The typical form of the. ice-eroded channel is U shape, as com- pared with the V form of youthful stream channels. Glaciers, therefore, widen their valleys. § (¢) The condition of load most favorable to abrasion is a light charge of coarse and hard rock fragments. || *H. F. Ried: “The mechanics of glaciers.’’ Jour. Geol., vol. iv, 1896, pp. 912-928. +I. C. Russell: ‘ The influence of debris on the flow of glaciers.” Jour. Geol., vol. iii, pp. 823-832. tH. F. Reid: Jour. Geol., vol. iv, pp. 920, 925; I.C. Russell: Jour. Geol., vol. iii, pp. 823-832; G. H. Barton, Tech. Quar., vol. x, p. 220; Chamberlin and Salisbury, Geology, I, p. 302. 2 W J McGee: ‘Glacial cafions.”? Jour. Geol., pp. 350-364. | I. C. Russell: Jour. Geol., vol. iii, p. 823-832; Chamberlin and Salisbury, Geology, I, pp. 271, 273. 26 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY It is evident that any deepening of glacial channels must be chiefly by abrasion, and that this must be proportionate to the two factors of velocity and pressure, other factors remaining uniform. With this in ~ mind, let us briefly consider the accepted principles enumerated above. The slower movement at the bottom of the glacier causes a corre- sponding reduction in one of the factors of abrasion. The reduced rate of flow may be due to either of the three causes noted above (a, 6, c,) or to all combined, and the reduction by a heavy load of debris and by shearing may amount to practical stagnation. We find here a most important principle in its bearing on deep erosion. Rapid corrasion by the ice is a self checking process. To the degree that the bottom ice is receiving rock debris, either by its own wear or otherwise, its flow is checked, while at the same time the subglacial rubbish serves as a shield and protection to the bed-rock. The product of abrasion is a rock-slime, which has in itself no helpfulness in erosion, but serves instead as a . lubricant to prevent erosion. It seems impossible for the rock-flour to be removed as rapidly as produced, if produced in large amount. Its production is inconsistent with free circulation of water beneath the ice, and the estimated rate of bottom melting, due to earth heat, amounting to one-fourth of an inch per year, can have no important effect in wash- ing away debris. The advocates of enormous erosion are requested to answer this question: If any glacier sawed down its bed “ thousands of feet’ during Pleistocene time, how did the saw clear itself for continuous cutting ? The checking of erosion by its own product implies that abrasion should be freer near the head of the glacier and diminishing toward the end as the subglacial load increases. This agrees with observation and contradicts the ice-erosion theory for the Swiss lakes and the Norwegian fiords, where it is claimed that the ultimate sections of the valley are the deeper. | Two more elements connected with the abrasion by the glacier are of interest and importance here. These are the apparent viscosity and the practical rigidity—two contradictory principles which seem, neverthe- less, to act together.* To the degree that the glacier moves as a rigid mass or bolt or plane its effectiveness is gone when its bed has been smoothed so as to offer little resistance to the ice plane. To the degree that viscosity is effective the abrasion is lessened by the failure of the ice to hold its cutting tools up rigidly and effectively to their work. Both elements are unfavorable to unlimited erosion. The work of the ice plane is rarely illustrated by straight groovings, * The latest discussion of this complex and much disputed problem is in Geology, I, by Cham- berlin and Salisbury, pp. 294-308. EVIDENCE FROM GLACIAL PHENOMENA 27 as discussed above on page 19. The viscous or plastic habit-is more commonly proven in glacial phenomena. The students of glaciers in Switzerland, Greenland, and Alaska have recorded many observations which show the yielding of the ice to obstructions by arching over them or bending around them. The production of roche moutonnée in the beds of glaciers is an illustration of its yielding habit and its adaptation to the irregularities of its bed. The rounding and smoothing of the rock bosses is proof of the polishing action of the glacier, but the existence of the bosses is proof that the abrasion was not sufficient to remove them or to level the bed. The ice could not remove one set of original irreg- ularities in its bed and then produce another set. If the ice could undo its own work, or act as a plane at one time and then in the same place scoop out irregularities and make mounds in its bed, we should find the roche moutonnée form as the common form of glaciated surfaces even in soft materials, whereas the form is characteristic of hard rocks and is doubtless produced by a moderate amount of scouring on weathered rocks. Other writers have recognized that the roche moutonnée form is not consonant with vigorous abrasion or planation by the ice, but, on the contrary, proves relatively slight abrasion. Drumlins illustrate on a vastly larger scale than roche moutonnée the same yielding habit of the ice. Whatever be the source of the drumlin material, their form is due to overriding by the ice and the rubbing and molding by ice flow. They prove not only failure of erosion at the local- ity of their occurrence, but failure of transporting power. They are large examples of the compliance of the marginal ice of the continental sheet. The question might be asked, Could not a very deep glacier, having great pressure on its bed, along with a steep gradient, giving high veloc- ity, rapidly abrade its bed? ‘The reply is, decidedly no! The postu- lated conditions can not occur together, except within ineffective limits. Extreme depth with high gradient is an impossibility in rivers and in glaciers. As a matter of fact, the glaciated valleys are not extremely steep for mountain valleys, and the upper and steeper sections often show little erosion (see description of Alpine phenomena, page 31). The great deepening of valleys has been assumed for the lower or ultimate sections (in case of the Finger Lake basins the cutting is assumed where the ice tongues were moving up the valleys). The physical conditions at the bottom of deep glaciers is not known inductively. It is believed that their bottom temperature is constantly at the melting point, which favors fluency and the yielding of the ice, while the slow melting tends to keep the rock debris beneath the ice, where it acts as a buffer. Moreover, in- creasing depth and pressure must tend, other conditions remaining the same, to diminish motion at the bottom. 28 H. lL. FAIRCHILD—ICE EROSION THEORY A FALLACY Shearing of the upper and more rapidly moving layers of the glacier ice over the lower, basal, and laggard layers seems to be established by observation. To whatever degree this factor is active it diminishes the velocity factor at the bottom of the ice and antagonizes erosion. It must be more effective where the bottom ice is obstructed. It argues specially against the flow of ice at the bottom of basins, and implies that the ice resting in a basin is likely to form a bridge over which the upper ice can travel. This is an important point with reference to the assumed glacial origin of lake basins. _ The mechanics of ice flow in basins has been well discussed by Culver (pages 363, 364 of paper cited on page 17). His conclusions, adverse to basining erosion, are certainly sound for the open, graded sections of valleys. The most effective abrasion must be on the rims of the basins, the tendency of which is to obliterate the basin rather than to make it, except, perhaps, at the foot of a steep slope. The U shape of ice-worn channels is generally accepted as of diag-’ nostic value. Like other physiographic features, it may be indefinite and liable to different interpretations by different observers, and the psychologic equation must not be forgotten. The U form belongs to stream valleys in a stage of their development. But such stream valley when scoured by ice is so much more striking, because of its uniformity and smoothness (just as a drumlin is more noticeable than a hill of the same general shape but of uneven surfaces), that it has been assumed as peculiar to ice erosion. Probably in many cases the ice has done noth- ing more than to clear the preglacial valley of its talus and other accu- mulations and to give the conspicuous smoothed surfaces. The accord- ance of the glaciated U-shaped sections of mountain valleys in their general proportions and their grade with their unglaciated extensions prove that the amount of ice erosion has not been great; but to what- ever extent ice erosion has occurred the U form implies that it has in- cluded widening of the valley. Some glacialists- have concluded from field study that glaciers tend to widen their valleys much faster than to deepen them, and the analysis of the mechanics of the glacial stream flow leads to the same conclusion. No critical study, so far asthe writer is aware, has reached any different conclusion. It is not believed possible for a glacier to deepen its valley to any appreciable extent without corresponding increase in width. The rea- sons for the more rapid widening of the valley are not difficult to find. It is not conceived as possible that ‘‘ plucking” or quarrying or any other mass-erosion can occur at the bottom of a glacier, under the great pressure, after the saliences are planed off and the rock surface has been smoothed. The only possible wear at the bottom is the slow process of EVIDENCE FROM GLACIAL PHENOMENA 29 abrasion. But alongside the glacier the forces of rock-destruction are greater in number and intensity. It has been shown that the combined resultant of the several erosive factors in stream glaciers—weight or press- ure, slope of channel or velocity of flow, and friction on the sides and bottom of the channel—is most effective along the sides of the glacier, and in addition to this we also recognize along the glacier edge the coopera- tion of weathering, and of frost- work, suggesting that at the bergschrund. The products of erosion are more or less removed from the plane of con- tact along the glacier edges and carried downward toward the bottom (to burden the bottom ice and there prevent erosion), while new tools of marginal erosion are constantly supplied by land-wash and from the lateral moraine. Probably the most important and conclusive paper in English relating to the mechanics of glaciers with reference to their erosional work on their channels was published in 1894 by W J McGee,* reasoning from studies in the Mono Lake region. His general conclusion is summed up in the last sentence of his paper : **Tt follows that these features do not necessarily imply extensive glacial exca- vation or indicate that glaciers are superlatively energetic engines of erosion.” Figure 1.—Truncation of Tributary Valleys. This theoretic diagram illustrates possible truncation of tributary valleys by glacial widening without deepening of the main valley. Some of the special and positive results of his analysis are stated in more positive terms, as follows: ‘Tt follows that the general tendency of glaciers must be to widen rather than to deepen the valleys they occupy, and to transform V to U canyons. It follows again that the characteristic glacial canyons must be only ahead stream-canyons” (page 359). He shows that one effect of changing a V-shaped valley to U shape is to cut off the lower or terminal part of the convex profile of the smaller tributary and ungraded valleys, and to thus produce what have been ; *W J McGee: “Glacial canyons.” Jour. Geol., vol. ii, 1894, pp. 350-364, V—Butt, Grou. Soc. Am., Von. 16, 1904 30 H. .. FAIRCHILD—ICE EROSION THEORY A FALLACY called “hanging” valleys. Figure 1 is his figure, somewhat modified. The words of his conclusion on this point are as follows: “Tt follows that the second feature (‘ hanging’ valleys) of the typical glacial canyons may naturally result from temporary occupation of water-cut canyons by ice, and that it does not necessarily argue profound glacial erosion’’ (page 360). | From this physical analysis he recognizes the possibility of some basining erosion: ape Thus the excavations of depressions by direct ice-action has a defi- nite though indeterminate limit, and can probably never exceed a moderate frac- tion of the depth of the ice . . .” (page 362). a The irregularities of gradient peculiar to such canyons are not greatly intensified, while glaciated rock basins are comparatively rare and of slight depth. It equally follows that the occupation was only temporary and the sum of glacial erosion relatively inconsiderable’’ (page 363). These conclusions of Doctor McGee adverse to ice erosion are spe- eially significant because they are based on merely the mechanics of the glacier as a plastic, moving body, taking into account weight, slope, friction, and the potential energy available for mass motion, or the down- stream impulse and weight at any point, constituting the intensity at that point. It should be clearly understood that this analysis does not take into account at all the important adverse factors due to the effect of the accumulating subglacial drift burden. The problem is treated as if the glacier had the power of constantly clearing its saw-cut. As a general conclusion it may safely be held that without glacial widening of a valley there can be little deepening. This is a very im- portant principle and can be effectively applied in special cases (see page 40). 3 From whatever side we attack the problem of glacial erosion the physical and geologic study leads to the conclusion that all the erosion factors are restricted or limited in their effectiveness. All the evidence from observation and direct study of living glaciers and all the theoretic conclusions derived by sound reasoning from known phenomena lead to the confident judgment that stream glaciers commonly act on their beds as flexible rasps. Rarely or locally they act as rigid planes. In either case their erosive effect is either limited or exceedingly slow. They scratch and polish their rocky beds, and to a limited extent they widen their channels. If the glaciers were given unlimited time of millions of years they might accomplish considerable cutting by the slow rasping process ; but their work would lag far behind that of rivers. Any claim that they can possibly cut faster than rivers is an untenable assumption. Their EVIDENCE FROM GLACIAL PHENOMENA ob transporting power is greater, mile for mile, but in equal time the river carries many times more material. Like the river, the glacier is not burdened by the product of its own corrasion, but by the material which other agents have given it. It is certainly true that rivers are the only valley makers. We have no evidence that any glacier has ever carved its own valley; nor have we any proof that any glacier has greatly deepened the valley it has occupied, notwithstanding the many asser- tions, even in the text books. The above positive and rather dogmatic statement is confirmed by all the observations on the phenomena of both living and extinct glaciers. The Pleistocene glaciers have recently abandoned their channels and have left us hundreds of examples of their work. Most living glaciers are waning rapidly and exposing their recent work. In no case has there been found any conclusive proof of valley making or.of basin making in rock; but, on the other hand, there is a wealth of positive and incontrovertible evidence that the ice has not seriously eroded its bed. The quite unanimous testimony of geologists who are most. familiar with living glaciers is to the effect that they do almost no erosional work, or scarcely more than sufficient to attest the fact of their presence. Yet the glaciers of the Alps were at work during all the life of the Pleistocene glaciers, which are assumed to have cut valleys “thousands of feet ” deep in granitic rocks, and they have continued to work for some thousands of years longer and should show proportion- ately better results. They show very little erosive effects. Plates 12-15 illustrate obstructions and irregularities in the beds of what have been vigorous glaciers of the Alps, and similar examples can be indefinitely multiplied,and doubtless from other fields. But wherein do the stronger glaciers of the Alps differ in size or principle or function or effect from those of the Sierras or Cascades or Alaska or Norway ? CONCRETE ILLUSTRATIONS Alps.—Apart from a group of eminent physiographers, the geologists and alpinists who are familiar with the glacial phenomena of central Europe are practically unanimous that glaciers are not effective agents of erosion. The argument was traversed by Albrecht Heim in his “Handbuch der Gletscherkunde.”’ The continued retreat of the Alpine glaciers is now exposing sections of their beds which have never been visible before and which are perti- nent evidence in this discussion. If any valleysin Switzerland, Norway, or elsewhere could have been greatly deepened by Pleistocene ice, it is very strange that active glaciers, working thousands of years longer, 32 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY - should have left obstructions and irregularities in their beds, as shown in plates 12-15. The end of the Unter Grindelwald glacier (plate 12) lies wedgingly in a chasm of rock cut by water—probably subglacial. It convincingly shows the failure of the glacier to effectively erode its bed, and all the rock forms at and below the end of the glacier give the same evidence of the glacial impotence. The walls of the valley have been merely sandpapered, and the same condition is practically true of all the glaci- ated valleys of the Alps. The accompanying plates give a few illustra- tions from several fields of the Alps. These facts are commonplace to Alpine geologists. It is an easy reply for the physiographic erosionists to say that the various conditions of erosion in the Alps were unlike those of Norway or the Sierras; but the burden of proof is on them to show the practical difference. The valleys of the Aare* and the Ticino ft nave rade been de-. scribed, from a physiographic standpoint, as showing great glacial ero- sion. The writer recently walked through the Aare valley above Meir- ingen (the Haslethal valley) and over the Grimsel, and passed by rail up the Ticino, and saw only evidences of the absence of great ice erosion.{ The Aare valley is described by Brigham as having a series of expan- sions and constrictions, which fact alone would seem to positively rule out great enlargement by a valley glacier. The cliffs at Grimsel are only smoothed by abrasion. In old photographs the rocks at the narrow gorge of the Aare, by the Grimsel hospice, appear to show glacial scor- ings passing up and over the cliffs. But this is deceptive, for on the ground it is seen that the strie are nearly horizontal and the abrasion was not sufficient to obliterate the irregularities and the surfaces due to jointing (plate 14). The jointing approaches the vertical, and the smoothed joint surfaces give at a distance the false appearance of nearly vertical ice scorings. The discussion appended to Brigham’s brief paper referred to above contains the gist of the whole contention, and the explanation and argu- ment by Turner adverse to ice erosion has not been answered, and is believed to be unanswerable. *A.P. Brigham: ‘Glacial erosion in the Aare valley.” Bull. Geol. Soc. Am., vol. 11, 1899, pp. 589-592. +W. M. Davis: “Glacial erosion in the valley of the Ticino.’’ Appalachia, vol. ix, 1900, pp. 136- 156; ‘*Glacial erosion in France, Switzerland, and Norway.’”’ Proc. Bos. Soc. Nat. Hist., vol. 29, 1900, pp. 273-322. t When the writer visited the Alpine valleys the descriptions referred to were not in mind, and the problem of glacial erosion was in mind only incidentally. The observations noted and the photographs taken were without any intention of discussion as to the origin of the valleys. This fact is given in justice tothe argument of this paper, and at the same time it illustrates how physiographic features are subject to very different and even opposite alte according to the mental attitude of the observer. BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 12 TERMINATION OF UNTER GRINDELWALD GLACIER The end of the glacier now hangs suspended in a narrow gorge, evidently the work of subglacial waters. The abandoned valley bottom is exceedingly irregular. August, 1902 BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 13 FIGure 1.—OBER GRINDELWALD GLACIER An angular mass of rock appears, by the waning of the glacier, in the middle of the valley. The chalets in the foreground are on an ancient moraine. August, 1903 Figure 2,—Summir or Grimset Pass anp TopTENSEE A heavy flow of ice once swept across this col, but the bosses of rock are merely rounded, August, 1903 GRIMSEL SUMMIT AND OBER GRINDENWALD GLACIER BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 14 er Figure 1.—CiLtirr At A SHARP TURN IN THE AARE VALLEY Where this cliff lies ice erosion should: have been most pronounced. The projecting masses are merely rounded. July, 1903 Figure 2.—NEARER VIEW OF CLIFF IN FIGURE 1 Showing horizontal abrasion. July, 1903 GLACIATED CLIFF NEAR GRIMSEL LAKE AND HOSPICE BP ts lg = ie BULL GEOL. SOC. AM. VOL. “6, 1904; PE. 15 RHONE GLACIER The recession of the ice reveals a *‘ hanging” valley in the main course of the glacier. July, 1908 EVIDENCE FROM NORWAY AND SCOTLAND 33 Norway.—To several of the geologists of Norway * the very striking topographic features of that country have seemed to require glacial erosion, and recently Professor Davis has favored the same conclusion. In the paper noted on page 32 he says: “. . . Hence glacial erosion must under this supposition be appealed to for the widening of preglacial canyons, steep-walled and narrow, into the existing fiord troughs, steep-walled and broad. At the middle of the fiord troughs the lateral erosion thus demanded would often measure thousands of feet, and that in the most massive and resistant crystalline rocks.”’ *. . . Hence to develop the existing discordant valley system from a mature preglacial valley system of normal river erosion requires a great deepening of the fiords by ice action, again to be measured by thousands of feet. Thus there seems to be no escape from the conclusion that glacial erosion has profoundly modified Norwegian topography.’’ (Page 290.) It should be noted that Professor Davis does not unequivocally assert that the fiords have been deepened thousands of feet, but the next maker of a text-book may use it as a positive and conclusive statement. Doubtless the fiords are difficult of immediate and precise explanation by stream work. They have recently been occupied by glaciers, and it was not unnatural to appeal to them as the cause of the deep valleys and the discordance. ‘The whole argument is based on the physiographic features, and implies glacial cutting of thousands of feet in the hardest crystallines. The conception certainly ignores the facts of observation, which show the impotency of ice-work. The Norwegian glaciers were of the alpine type, and there is no reason for attributing to them any special property or power. Scotland—The British geologists, even those who follow more or less the lead of Ramsay in the belief in ice erosion, have not appealed spe- cially to their own physiography. But a recent paper requires brief notice, because it has been quoted in this connection (page 51). This paper is a description or reference to features of topography and drainage in Scotland which seem to the author to be due to ice action. The paper is far from convincing. The statements are too general, too theoretical, too discursive, and too largely assertive. No maps or other illustrations of any kind are given. Apparently the author sees evidence of ice-cutting which expert glacialists have not seen, for if the features possessed the clearness and significance which he attaches to them they should have been thoroughly described long ago. The doubt concern- ing the matter of this paper pertains to much of the physiographic evi- * Most of the literature is foreign, but the facts and discussion may be found in Heim’s ‘‘ Hand- ouch der Gletscherkunde,” in the paper by Davis, noted on page 32, and in an article by Andr, M. Hansen, Jour. Geol., vol. 2, pp. 123-143. +J. G. Goodchild : “‘ Glacial furrows.’’ Geologists Magazine, vol. iy, pp. 1-7, June, 1896. \ 34 H. L. FAIRCHILD—ICVE EROSION THEORY A FALLACY dence. There is so much indefiniteness and genetic complexity in features of topography that interpretation has unusual latitude. The attitude of mind and the theory of the observer count for much more than in most other lines of study, and the arguments, conclusions, and assertions are often difficult to either verify or deny, although doubt may be very strong. Greenland.—During the past decade Greenland has been the Mecca of the glacialists, and among those who have visited the land and published their observations are Barton, Chamberlin, Heilprin, Nansen, Peary, Salisbury, Tarr, and Wright. The writings of all these men practically agree on the following points: That the West Greenland glaciers are carrying but little drift, and that mainly in the basal layers; that ero- sion is slight; that the exposed topographic features are not the product of ice erosion; that the lower layers of the ice show reduced movement, with some shearing of upper over subjacent layers; the overriding of boulders and other obstructions; the overriding by the edge of the ice of its own moraine deposits; and the comparative absence of clay or rock-flour. It will not be necessary to refer at length to the literature,* as fortu- nately there is no essential divergence of opinion over the fact of slight ice erosion. More than any other writer on the Greenland glaciers, Professor Tarr has referred to the erosional phenomena, and has particularly and em- phatically noted the failure of the ice in that respect. This is the more significant as Professor Tarr has been one of the American advocates of ice-eroded valleys. A few lines quoted from his writings will be per- tinent. ‘Although the hills of Turnavik (eastern margin of Labrador) are well rounded, and show signs of decided ice scouring, . . . itis evident that the glaciation did not succeed in destroying even the details of the preglacial topography. This evidence of slight scquring is in harmony with that found in Baffin land and Greenland, and alsoin parts of New England. It shows that preglacial decay was deep, and that the general ice scouring did not lower the surface far below the zone of decay in the weaker members of the rock series.’?’ (Am. Geol., xix, 192.) ‘‘ Very nearly the same conditions were present in Cumberland sound, though the details of the preglacial topography are perhaps even more perfectly preserved. The surface of the hills is extremely irregular, and valleys and ridges of minute size, distinctly of preglacialorigin, abound. . . : It wasastonishing to find how little effect in smoothing the surface was accomplished by the ice invasion of the * The writings of Professor Chamberlin are to be found in a series of articles in the Journal of R. D. Salisbury: Jour. Geol., vol. iii, pp. 875-902; vol. iv, pp. 769-810. G. H. Barton: Tech. Quart., vol. x, pp. 213-244. R. S. Tarr: Amer. Geol., vol. 19, pp. 131-136, 191-197, 262-267; vol. 20, pp. 139-156. Bull. Geol. Soc. Am., vol. 8, pp. 251-268. A. Heilprin: Pop. Sci. Monthly, vol. 46, pp. 1-14. EVIDENCE FROM GREENLAND AND THE SIERRAS 35 land ; but very nearly the same conditions exist in those parts of the Greenland coast, which were studied in detail. It is noticeable near Cumberland sound, as well asin Turnavik, and in Hudson’s strait, and indeed in Greenland, that there are many basins of small size, surrounded entirely by rock. While some of these have no doubt been scoured out by differential ice erosion, the position of many of them, along the line of weaker rocks, indicates that they represent differential pre- glacial weathering. The advance of the ice in these cases has served to remove the decayed rock, and perhaps to deepen the depressions formed by this action, though ice erosion would not in this case be the prime cause of the basin.” (Ibid., 195.) *‘In other words, the irregular surface (of rock) is made more regular by the nearly stagnant bottom ice, which fills the depressions. . . . The ice arches over the irregular land, making curves, whichare plainly seen by the arch of the debris layers, whose form is that of a generalized dome of a regular form. Some areas are scoured more than others, and there is a gradual reduction to a gen- eralized surface because the depressions are protected by more nearly stagnant ice. _while the projections are worn down toward the curve of average outline of the irregularity.’’ If there is any reason why the ice should have less erosive power in the Greenland region than elsewhere it has not been found. Possibly there is some explanation in latitude; greater rigidity of the ice, due to low temperature, might facilitate shearing and reduce plastic flow at the bottom. It is also possible that the Greenland ice cap is not very old and has not been very active; but, whatever explanation be forthcoming, the Greenland phenomena give only decidedly negative evidence on the question of deep ice erosion. Sierras.—It was formerly believed by a few geologists that the deep valleys in the Sierras, including the Yosemite, had been excavated by glaciers. At present probably no one holds such view; but opinions quite as radical and probably just as mistaken are still held for other regions; hence it will be pertinent to review the Sierran phenomena for comparison. Among many writings, we will note only a few which bear specially on the glaciation. In 1891 Becker * published a paper on the structure of the high Sierras between Truckee river and the south fork of the Stanislaus, in which his argument and conclusion were most emphatically opposed to great ice erosion. He says: “Some few geologists still believe that glaciers . . . vigorously erode solid rock. In my opinion, this theory is maintained in opposition to overwhelming evidence. Reference has already been made to some of the many facts indicating a trifling amount of erosion since the preglacial date in the higher part of the Sierra, and long before my examinations Professor Whitney reached the conclu- sion that the solid rock had been scoured rather than eroded by glaciers.” (Page 65.) *G. F. Becker: The structure of a portion of the Sierra Nevada of California. Bull. Geol. Soc. Am., vol. 2, 1891, pp, 49-74. 36 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY ‘‘Tce seems to have played a considerable part in clearing the canyons of frag- ments and in excavating shattered and decomposed patches, so that in a sense one must ascribe a large erosive effect to the glaciers; but the ice seems, nevertheless, to have been incapable of cutting into solid masses to any extent, or even into much fissured rock where little decomposition had preceded and where the blocks were tightly wedged together. ‘‘Tn many cases the glaciers have polished rock surfaces, the contours of which are so thoroughly characteristic of surface exfoliation, due to weathering, that no observer could doubt their character, and some of the surfaces are such that ice could not possibly have modeled them. Such evidence, together with that derived from the occurrence of glaciated lavas near the bottoms of the present canyons, in- dicates very clearly that the present system of canyons was established long before glaciation began, and probably during the warm and no doubt very wet Pliocene epoch.’”’ (Page 68.) From study of the canyons in the region about lake Mono, Russell * concluded that they were not the product of ice erosion, and Gilbert T agrees with all the other experts that the present drainage features of the Sierras were far advanced before the ice occupation. : The important theoretical paper on glacier mechanics, already quoted (page 29) as adverse to extreme erosion, published by McGee in 1894, was an outcome of his work in the Sierras. In 1900 a critical and important paper was published by Turner f{ on the south central Sierras. In this paper he gives an excellent review and discussion of the problem of ice erosion in application to the Sierras, and Yosemite in particular, and proves conclusively that the canyons, including Yosemite, were not made by ice, although he recognizes slight glacial action in valleys where it had not previously been noted. Regarding rock basins, Turner states that such are abundant in the glaciated Sierras, and ordinarily in granite, and says: ‘‘ Many of these rock basins are at the base of steep slopes, and it is possible in such cases that the great downward pressure of the ice excavated such basins, especially where the rock was much jointed. The location of other rock basins, however, although always where the ice mass appears to have been very thick at one time, is such as not to suggest the probability of there having been sufficient weight of ice to scoop out a basin in hard rock. . . . It is more than likely that in many such cases the basin was originally produced by unequal disintegration of the granite in preglacial time, this decomposed material being subsequently ex- cavated by the ice.’’ (Pages 286, 287.) Some of the rock basin lakes are more than a mile long and toward 100 feet deep. Lakes Washburn, Tenaya, and Johnson are mentioned. *I. C. Russell: ‘‘ Quaternary history of Mono valley.” Eighth Ann. Rep. U. S. Geol. Survey, pt. i, 1889, p. 352. 7G. K. Gilbert: ‘‘ Lake Bonneville.’*> Monograph i, U. S. Geol. Survey, 1890. tH. W. Turner: ‘‘The Pleistocene geology of the south central Sierra Nevada, with especial reference to the origin of Yosemite valley.” Proc. Calif. Acad. Sci., iii ser., vol. i, no. 9, 1900, pp. 261-321, sede ad a ee brass « EVIDENCE FROM THE SIERRAS 37 Professor J. C. Branner* has shown that the singular topography about the cataracts in the Yosemite valley was produced by the more rapid down-cutting by the streams draining the glaciers than by the ice itself. He says: ** The evidence of the falls at the mouths of the hanging valleys shows that the wearing done by the ice was trivial compared with the wearing done by the glacial streams. The subglacial streams also cut channels beneath the ice a great deal faster than the ice cut the broader floors on which it moved.” (Page 551.) The latest paper is by Lawson,f on the upper Kern valley, in the high Sierras, a very interesting description of a most remarkable valley and region : ** The canyon of the upper Kern is one of the great canyons of the Sierra Nevada, and in some respects it is the most remarkable of themall. . . . But while it is a feature essentially due to stream erosion its characters as such have been modi- fied by its having been occupied for a time by along trunk glacier, which extended down the canyon as far as the mouth of Coyote creek and which was fed by sev- eral tributary glaciers. . . . In consequence of this episode in its history the canyon above Coyote creek has the typical U shape in cross profile so characteristic of glaciated mountain valleys.’’ (Pages 328, 329.) | ‘The width of Kern canyon below the Kern-Kaweah river does not appear to have been appreciably increased by the occupation of the ice. This is practically proven ky the following consideration: The depth of the ice in the canyon aver- . aged probably not more than 1,200 feet. The walls of the canyon are about twice this height, and the lateral recession of this upper portion of the walls in conse- quence of glacial sapping or scour at their base could only have been by a process of shedding rock fragments upon the surface of the glacier. These fragments would accumulate at the terminal moraine; but the volume of the two terminal moraines together does not exceed 2,100,000 cubic yards. If we distribute this over the upper — 1,200 feet of both walis of the canyon for the distance of 14 miles from the Kern- Kaweah river to the main terminal moraine, it would make a layer 4 inches thick. If we consider that probably half of the moraine came from sources outside the canyon, the layer would be reduced to 2 inches. This estimate may be modified and corrected in a variety of ways, but it leaves a quantity for the glacial widening of the canyon which is negligible. The canyon, then, had practically the same width before its occupancy by ice that it has today.” (Pages 349, 350.) The perfectly straight canyon was occupied by a bolt of ice more than 20 miles in length, formed by the junction at the valley head of two strong glaciers, and joined along its course by six tributaries, each formed from several glaciers (see Lawson’s plate 81). So far as the drainage area (volume), valley conformation (freedom of flow), and latitude (tem- * J.C. Branner: “A Topographic feature of the hanging valleys of the Yosemite.’’ Jour. Geol., vol. 11, pp. 547-553, 1893, tA. J. Lawson: ‘* The geomorphogeny of the upper Kern basin.’’ Univ. of Calif. Pub., Bull. Dept. Geol , vol. 3, 1904, pp. 291-376. VI—Boutt. Geox. Soc. Am., Von. 16, 1904 38 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY perature) could.favor erosion, the Upper Kern glacier should have had great potency. An important fact to be noted in connection with the fact of slight glacial erosion in the Sierras is the existence of “ hanging ” valleys. ‘*. . . The apex of these cones is generally several hundred feet above the canyon floor. These cones are the product of streams which enter the canyon far above its bottom and which, in many cases, flow in glaciated trenches. These trenches are fine examples of hanging valleys.” (Page 329; italics not in the original.) Lawson divides the Upper Kern valley into two parts: ‘‘an upper part, in which the Kern has been engaged in vertical corrasion since the retreat of the ice, and a lower part in which it has been engaged in aggradation.” He thinks that the aggrading portion ‘‘represents an overdeepening of the canyon floor by glacial scour, a process which ~ has been exemplified in many glacial! troughs, such as in some of the lochs of Scotland and fiordsof Norway. . . . Both the degraded and aggraded portions of the glaciated canyon are U-shaped in profile, but the most perfect U-shaped profile is in the former, and is true of the rocky floor and sides of the canyon. The U-shaped effect of the aggraded portion is largely due to the talus slopes at the base of the canyon walls. If the rocky bottom of the canyon beneath the meadow floor be also U-shaped, as is very probable, then the depth of the aggra- dational accumulation may not be more than a hundred feet.’’ (Page 349.) The only doubtful matter would seem to be that of the conjectural basin. The fact of less perfect U form of the aggrading section, implying less total erosion, taken in connection with the small volume of morainal debris, and specially the absence in the latter of clayey material or product of abrasion, throws serious doubt on the suggestion of an erosional basin beneath the meadows. However, it is not an important matter, as a basin of 100 feet, or even greater depth, is inconsequential in the valley stretch of setae 8 or 9 miles. We find in the Sierras all the important features wien have been appealed to for proving enormous ice erosion, particularly U-shaped canyons and hanging valleys; but the geological experts are practically unanimous that the Sierran valleys are preglacial, and at the most have only been more or less modified by glaciers. But now, if these topo- graphic features can exist in the Sierras without great deepening of trunk valleys, then it seems illogical and unreasonable to require vast deepen- ing in Norway or Alaska to account for the same features. The only equivocal matter which we find in the Sierran study relates to the shallow rock-basins, but they can be explained without requiring any scouring of live rock; and in any case they are insufficient founda- tion for an assumption of thousands of feet of deepening. Rae ae. EVIDENCE FROM THE CASCADES 39 Cascades.—A very recent paper by Willis * revives the interesting ques- tion of the origin of lake Chelan, favoring the suggestiont of ice erosion. The problem of the Chelan basin is practically the same as that of other deep lakes, like the Swiss lakes, found in the tracks of former stream glaciers, { and it is well to discuss it here at sufficient length. The descriptions following are from Willis’ paper: ** Lake Chelan is a slender body of water, 65 miles long, whose southeastern end lies open to the sky between the grass-grown hills of the outer Columbia valley, while its northwestern lies in shadow between precipitous mountains in the heart of the Cascade range. . . . The canyon is that of the Stehekin-Chelan river, which rises in latitude 48° 30’ in glaciers of the Cascade range at altitudes of 5,000 to 8,000 feet. The headwaters descend very abruptly, 1,000 to 1,800 feet in the first mile below the glaciers, and combine in a U-shaped valley of gentler grade, the fall being 2,500 feet in 23 miles. This section is cut in rock bottom. For 12 _ miles farther down stream the valley is floored with boulders, coarse gravel, and sand, and the slope is but 20 feet to the mile, endingin the delta which the stream is building into lake Chelan. “The gravel-filled section of the valley is no doubt deeply cut in the solid rock» since but a short distance beyond the front of the delta the lake is more than 500 feet deep. For a distance of 35 miles the depth varies from 1,000 to 1,400 feet, 1,419 being the maximum yet sounded. As the water surface is but 1,079 feet above sea, the bottom of the lakeis fora short stretch 300 feet below sealevel, and an interesting question is raised as to how so deep a basin originated. Fifteen miles from its outlet the lake begins to shallow, and in its lower reach does not exceed 200 feet in depth.” (Page 58.) **Lake Chelan occupies a canyon in granite and gneissoid rocks. The waters are retained by adam of drift, but discharge through the gorge of Chelan river into the Columbia, whose bed is not far from solid rock. According to the alti- tudes above sea of rock in place in the Columbia, several miles below the junction, that of the lowest rock sill over which the waters of lake Chelan can have escaped is about 700 feet. . . . The maximum sounding was 1,419 feet—i. e., to 340 feet below sealevel, or 1,040 feet below the rock rim of the basin.’’ (Page 81.) ** Effects of glacial erosion are obvious throughout the Chelan-Stehekin system. They may be traced from the moraines of existing glaciers, in the cirques about the headwaters, down the grooved and rounded canyon walls, into the waters of lake Chelan, and out to the terminal moraine near the Columbia, in all a distance of 70 miles. Three miles above the terminus of the ancient glacier its thickness was 1,000 feet, and its surface stood about this amount above the present level of lake Chelan. Thirty miles above its lower end, that is, near Safety Harbor creek, the thickness of the ice was 4,500 feet, and toward the head of the lake it probably exceeded 5,000 feet.” (Page 82.) e The walls of the canyon are sloping, not precipitous. . . . The *Bailey Willis: Physiography and deformation of the Wenatchee-Chelan district, Cascade range. U.S. Geol. Survey, Profes. Paper No. 19, 1903, pp. 41-97. 7 Henry Gannett: Lake Chelan. Nat. Geog. Mag., vol. ix, pp. 417-428, 1898. t The New York or Finger lakes are an entirely different problem, as their valleys were never occupied by stream glaciers, but only by lobes of the continental glacier in valleys sloping toward the ice (see discussion, pages 55-73). 40 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY maximum inclination is usually near the water surface, and for a hundred feet or so may locally approach 60 degrees. Generally it is less than 40 degrees. And from this steepest facet the profiles rise in a curve which is convex upward, and pass into the nearly level spurs of the adjacent ranges.” (Page 59.) Questioning the amount of glaciation in the general region and on other valleys, the author admits that the valleys are preglacial and have only been modified, and that the Chelan valley is peculiar. Following are the words: ‘In the glaciated region of the higher Cascades, the canyons are developed on a grand scale. Russell has rightly described them as extending with channels at railroad grades into the very heart of the mountain mass, and they are from 1,000 to 3,000 feet in depth. They have been widened and deepened by ice-work, to - what degree is not to be determined, but apparently in some instances greatly. Nevertheless, when all reasonable allowance has been made for glaciation, the canyons which may be ascribed to corrasion during the Twisp stage (preglacial) were of great depth. They were relatively narrower than they now are, and their topographic development had the character of advanced youth.” (Page 81.) ‘*A further test of this conclusion (ice deepening) may be applied by contrasting it with other valleys in the same region, many of which have been occupied by glaciers, and none of which exhibit similar peculiarities.” (Page 83.) Plate 16, reproduced from Willis’ paper, shows the cross-profile and the proportions of the valley at ‘ The Narrows,” the narrowest part of the gorge, which is also the deepest part of the lake. This shows that the valley is decidedly V-shaped in the deep section, and the profile of the walls convex when the submerged section is included. These are the opposite of glacial characters. The gorge here shows no lateral erosion, although glacial widening is claimed for other valleys of the region. It is also admitted that no other valleys of the region exhibit the peculiar features of the Chelan gorge, the suggested explanation being the greater volume of the Chelan glacier. From the published facts, mostly quoted above, we are unable to find any evidence whatever of ice erosion in the Chelan gorge. All our knowledge of glacier work opposes the idea. Where are the concave profiles and U shape, recognized as special characters of valley erosion by glaciers, and where is the enormous mass of clayey moraine which should have been produced by such vast abrasion? No description is given of the drift. Itis inconceivable that a glacier should deepen a narrow gorge 1,000 feet and yet produce no perceptible cutting of the valley walls. The walls at the narrows were the obstruction which would have been cut away. The fact of the depth of the gorge being inversely proportioned to the width has been suggestively explained as an effect of increased velocity in the narrow section, citing the behavior of rivers. The comparison BULL. GEOL SOC. AM. VOL 16, 1904, PL. 16 NARROWS OF LAKE CHELAN After Willis EVIDENCE FROM THE CASCADES 41 seems faulty. The river cuts down faster in its narrowed section be- cause it saws only at the bottom (but even then it does not excavate deep basins), while the glacier cuts faster at the sides, and is prohibited from rapid bottom cutting by every known factor and principle of glacier mechanics. Ice erosion has been assumed as the cause of deep eke. basins which | lie in beds of former glaciers probably because it is the only near-by agent on which suspicion can fall. It has left evidences of its presence at the locality of the deed; but the circumstantial evidence is here mis- leading. We probably have in these lakes a much larger and more diffi- cult problem than has been supposed. ‘The idea of glacial erosion must certainly be abandoned, and the problem needs to be investigated along other lines. Following are a few suggestive facts which have a bearing on the subject. If the deep Chelan basin were due to ice erosion, it would seem as if lakes of similar character should be common in glaciated mountains. There is apparently no good reason for attributing any exceptional char- acter or power to the Chelan glacier. But lakes of the Chelan type are strikingly wanting in glaciated areas. No lakes occur, so far as atlases show, in the Caucasus, Pyrenees, Urals, or Carpathians, all of which moun- tains held strong glaciers, and only a few are shown in the Himalaya, mostly in the western districts, but no description of their character is available. TheSwiss and Italian lakes are most surely not produced by ice erosion; but the deeper ones, like Zug, 650 feet deep; Geneva, 1,100 feet deep,and Maggiore, 2,600 feet in depth (the bottom 1,900 feet below ocean), seem to belong in the Chelan category. The Dead sea, with its surface 1,300 feet below sealevel, or Assat, east of Abyssinia and far below sea, are types of basins which require explanation as truly as Chelan, but no one has yet ventured to suggest ice erosion. If orographic or subter- ranean movements or local subsidence can reasonably be invoked for these basins, possibly even so narrow a basin as Chelan may have simi- lar origin. The comparative absence of lakes in non-glacial regions is often over- emphasized. An inspection of a good atlas will show nymerous lakes scattered the whole length of the American cordillera from Colorado to Chile; also in other non-glaciated mountain districts. Itis probable that many of these intermontane lakes can be explained only by earth move- ments. An exhaustive investigation of the subject will possibly find that Chelan, Maggiore, and Dead sea belong in the same type. , Alaska.—There have been references to the glacial origin of valleys and fiords in Alaska, and statements to that effect; but, so far as the writer has found, no argument has been made nor proofs offered, except 42 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY the general assumption from fiords and hanging valleys. The very im- portant work of Russell over a large area in the northern lands, specially on the Malaspina and the Mount Saint Elias group, brings out no facts favoring erosion. The most studied glacier in Alaska is the Muir, which furnishes only indisputable evidence of its failure as an erosive agent. The Muir is now uncovering, or receding from, a thick deposit of gravel and forest ground which it had overridden during its latest advance, and did not remove. The fact is recognized by all students of the region, and the abundant literature * makes the extended description here unnecessary. Cushing writes as follows: ‘fA glacier of great thickness (over 2,000 feet) has advanced over these gravels, and done so for considerable time. . . . The influence of the ice upon it must have been more protective than anything else. . . . These deposits have for their floor an old land surface, with tree stumps still standing in the soil in which theygrew, . . .”’ (Page 220.) ‘‘At Muir glacier, in just the position where the greatest erosion would naturally be expected, soft gravels have been undisturbed by the ice.” (Page 230.) In discussing this behavior of Muir glacier, Russell says: ‘That glaciers of great thickness may overrun unconsolidated gravels, without disturbing them, is no longer open to question.” (Paper noted above, page 194.) He then refers to such deposits beneath the Malaspina glacier, and to uncovered gravels near Mono lake, California. More than other writers, Professor Cushing has noted the erosional phenomena of the Muir Glacier region, and his observations are interest- ing and important, and negative in their force. ‘On all the mountain slopes which Muir glacier has overrun, a tendency toward the production of a surface consisting of small, shallow valleys separated by low ridges is seen, both trending in the direction in which the ice has moved. The production of such surfaces in this region depends on the presence and distri- bution of the fissure planes. Weathering takes place along these planes to varying depths, resulting in the loosening of V-shaped blocks of varying sizes. After such decay has been in progress for a considerable length of time, a glacier riding over the ridge and removing loosened material will tend to leave a sur face composed of ridge-like projections with shallow depressions between. ‘* Lakes. —On the tops of the low mountains bordering Muir glacier, over which *S. P. Baldwin: ‘‘ Recent changes in the Muir glacier.” Am. Geol., vol. xi, 1893, pp. 366-375. H. P. Cushing: ‘“‘ Notes on the Muir glacier region, Alaska, and its geology.’’ Am. Geol., vol. Vili, 1891, pp. 207-230. H. F. Reid: ‘‘Glacier bay and its glaciers.’’ Sixteenth Ann. Rep. U.S. Geol. Survey, pt. i, 1896, pp. 415-461. i I. C. Russell: ‘Origin of the gravel deposits beneath Muir glacier, Alaska.” Am. Geol., vol. ix, 1892, pp. 190-197. G. F. Wright: ‘‘ Ice age in North America,” 1902, pp. 36-66. ee Cee EVIDENCE FROM ALASKA 43 the ice hasswept, diminutive lakes occur. . . . They occupy small depressions or basins on the tops of these ridges. . . . They are all very small, only a few yards in diameter and with no great depth. Some of them clearly occupy rock basins. . . . The conclusion can not be avoided that these hollows were the work of ice. In most cases the method of their formation seems clear.’ (Page 227.) ‘‘All these basins which I saw lie in small valleys on the ciiotin ibis tops, whose presence seemed to depend on the fissure systems and on the varying depths to which loosening of the blocks had taken place. They lie at the foot of slopes, down which the ice moved, impinging with unusual force at its base, where the greatest amount of polishing and striating has taken place.”’ (Page 230.) “ The rough edges have uniformly been somewhat smoothed, but the character of the surfaces seem to me to clearly show that the valleys and the basins have been formed by the removal of loosened blocks, leaving a rough, jagged surface, whose edges have been smoothed and polished.’’ (Page 227.) ** Old surface features not obliteraited.—On the mountains in Muir Glacier basin from which the ice has recently retreated, surface features are occasionally observ- able which seem incompatible with the theory that glaciers vigorously erode hard rock. . . . ‘That the glacier has done little more than to remove the loosened rock and polish the resulting surface is shown in a vast number of localities here by the character of that surface.’’ (Pages 228, 229.) The existence of numerous islands in Muir inlet and Glacier bay is conclusive proof that this fiord was not made by glacial erosion (see Cushing’s paper, page 227). His comments concerning the wash from the glacier confirm the statement made on page 21 of this paper to the effect that such estimates of the abrading power of the ice were value- less. He says: ‘* Estimates of the amount of material brought down by the glacier are difficult to obtain owing to the fact that the material is all carried into the sea; that the number of subglacial streams is not known; and that there is no evidence that those which issue from the ice directly into the water carry as much sediment as those which issue from the corners and flow through the gravels. I could find no evidence inconsistent with the supposition that the debris falling on the surface of theice yearly, together with the previously disintegrated material which the ice has removed and is removing, is amply sufficient to account for all the detritus depos- ited at the front of the glacier. The amount of material in sight on the surface of the glacier is enormous.” (Page 229.) Doctor Gilbert’s recent work on Alaskan glaciers* regards hanging valleys as proof of the glacial origin of the Alaskan fiords. The com- petency of Pleistocene glaciers to excavate one or two thousand feet in crystalline rock is assumed. If such explanation were correct, then it should harmonize with all other geologic facts of the region; but it is apparent that the phenomena are not in accord under this hypothesis, for with his characteristic fairness the author mentions inconsistencies *G. K. Gilbert: ‘‘ Harriman Alaska Expedition, III, Glaciers,” 1904, 44 H. L. FAIRCHILD—ICE EROSION THEORY A FALLAOY and difficulties, and other questions might be raised. A few of these points will be here mentioned. Gilbert agrees with Willis that the fiords of Puget Sound district are chiefly river work. ‘“‘ The system of troughs it (the ice sheet) left behind are regarded as pre-existent stream valleys, only moderately scoured and straightened by the ice which over- ran and occupied them.’’ (Page 135.) ““. . . Regarded as stream valleys, the channels tell of a preglacial baselevel at least 500 feet, and probably 1,000 feet or more, below the present sea surface.”’ (Page 136.) The question would naturally be asked why Alaska should not have stood as high above ocean during Tertiary time as the district imme- diately contiguous and possessing the same physiographic characters, or what the rivers of Alaska were doing while those of Washington were cutting the valleys that are now fiords? The only argument which Doctor: Gilbert makes against such land elevation as would permit the stream origin of the valleys now drowned is the following: ‘*. .. Under present climatic conditions, such a change would carry a very large area above snow-line, and would so promote the alimentation of glaciers as to flood the whole district with ice and abolish stream erosion. Stream erosion therefore could not have been carried, by lowering of baselevel, to the lowest parts of the channel system without the aid of important climatic variation. Without doubting the possibility of wide range in independent climatic factors, it seems easier to assume that the lowering of baselevel was comparatively mod- erate, and that a considerable part of the down-cutting of the channels was accom- plished by Pleistocene glaciers.” (Page 136.) The above argument seems to ignore the generally accepted fact of warm climate over arctic lands during the early and middle Tertiary, as shown by paleontologic evidence. Would it not seem easier to accept the evidence of warm Tertiary climate over northern lands and the stream origin of the Alaskan as well as the Washington fiord valleys, which gives a harmony of facts, than to assume that the Pleistocene glaciers could abrade the bottoms of their valleys 2,000 feet in crystal- line rocks? If Alaska had been during the Tertiary 1,000 to 2,000 feet lower than now, so that the fiords could not be the product of river work, there ought to be found conspicuous and indisputable remnants of an uplifted coastal plain, or at least lines of wave-work. The difficulty of accounting by ice erosion for the plexus or anasto- mosing system of deep valleys which characterize the Alaskan and Washington coast, or for even seriously deepening them, is recognized but not discussed further than to offer a suggestion of differences in rock structure. EVIDENCE FROM ALASKA 45 *‘ When it is considered that these fiords, being parallel to the coast, run athwart the general movement of the ice from land to sea, the fact that their depth is comparable with that of troughs lying in the direction of general movement is certainly remarkable.” (Page 147.) The absence of moraines or masses of drift over wide districts is noted as follows: “In the narrower parts of the inside passages we saw no accumulation of glacial drift.” (Page16l.) ‘The glacial deposits we encountered are of trifling magnitude collectively in comparison with the glacial erosion, of which we saw evidence, and it was therefore inferred that the principal regions of deposition lay outside the field ofourobservation, . . . and that its outer margin was beyond the present line of coasts.’’ (Page 162.) If the Alaskan glaciers could abrade their valleys during the Pleisto- -cene to enormous depths in the crystallines, then the erosional work was so rapid and effective that it could not have stopped suddenly, but con- spicuous recessional moraines should be left in even the higher valleys. However, if the ice first cleared the stream valleys of weathered ma- terials and subsequently simply slowly abraded the smoothed and firm rocks, as glaciers do today, then there should be no massive moraines in the higher valleys, and the facts are harmonious. Speaking of the ‘“‘inequality of glacial erosion,” the author frankly says: | “The great work which it has seemed reasonable to ascribe to ice in the deep- ening and widening of fiords and other troughs stands in striking contrast to the feebleness of ice erosion in other places, which permitted, for example, the preser- vation of the low peneplains of Annette island and the vicinity of Sitka. In the one case the depth of the erosion is measured by hundreds of feet; in the other by tens.’’ (Page 160.) The explanation then given of the supposed contradiction seems in- volved and inconclusive; but harmony would be secured, not merely in respect to this difficulty, but with other difficulties, by abandoning the idea of enormous ice-cutting anywhere in Alaska. An anomalous irregularity of supposedly glaciated surfaces is ex- plained by “plucking” (page .206); but this appeal to plucking to explain rough surfaces which are assumed to have been deeply eroded seems inconsistent with the following: “The work of rock sculpture accomplished by the middle and lower parts of a glacier is performed chiefly by the processes of abrasion and plucking. . . . If the plucked blocks have originally stood as projections they may be broken away, even if quite firm and flawless; otherwise it is probable that they can be removed only if originally separated by joints or other structural partings.”’ VII—Butt. Grou, Soc. Am., Vou. 16, 1904 46 H. lL. FAIRCHILD—ICE EROSION THEORY A FALLACY ae The prominences are therefore abraded more rapidly than the adja- cent hollows, and the profile is thus reduced to simple forms.” (Page 203.) ‘* Another factor on which rate of abrasion depends is pressure; the abrasion is more rapid as the pressure of the glacier against the bed-rock is greater. . . Thus in a second way there is a tendency to reduce the profile of the bed to simple forms.” (Pages 203-204.) | If plucking did occur in the beds of glaciers the scars should be found there; but plucking and abrasion are mutually opposing factors and can not long coexist. After abrasion has done its first work under the pressure of thick ice, how is it conceivable for plucking to occur in live rock? (See discussion, page 20.) Plucking can be an effective process only in superficial removal of loosened rocks. It is good evidence of lack of deep erosion. Mention of the islets in the fiords is only made in saying that in num- ber they are “uncounted,” but the existence of numerous islets in the fiords is not consistent with the origin or the great deepening of the latter by glaciers. The body of facts relating to Alaskan geology seem to be thrown into confusion by the hypothesis of ice erosion so intense as to deepen valleys 1,000 to 2,000 feet; but, under the view that the fiords are drowned stream valleys of the Tertiary uplift, modified by ice occupation in the Pleistocene, and with only proportionate glaciation over the general and intervalley areas, the phenomena will be harmonious. But the question may be asked, ‘“‘ What about the ‘hanging’ valleys?” The reply is that they can probably be satisfactorily explained when the effort | is made in an inductive way. They exist in mountain regions where the idea of deep ice-cutting will not hold and even in regions which are un- glaciated.* Discordant drainage features have been found about the Finger lakes, and the latter are positively not due to ice erosion. The most striking discordance of valleys which has so far been shown occur in young mountains of crystalline rocks. Their relation to glacia- tion is probably only incidental, except that they may have been made * Since this paper was written Professor I. C. Russell, in reviewing Doctor Gilbert’s book, in Science, vol. xix, May 20, 1904, page 785, questions the glacial origin of hanging valleys, andjwrites as follows: ‘‘Again, in well glaciated mountains, like the Cascades and Sierra Nevada, the great differences in level between a main valley and its tributary hanging valleys, amounting in some instances to 1,500 or 2,000 feet, and this where the main valley is short and has but a comparatively small gathering ground for snow, must needs make the conservative glacialist pause before accepting the conclusion that such discrepancies are solely due to differential ice erosion. Other consider- ations in this connection might be mentioned, such as the fact that a deep glaciated valley with hanging valleys along its sides not infrequently heads against a cliff, and its direct continuation above the cliff also has the characteristics of a hanging valley. Then, too, hanging valleys may be claimed to occur on the sides of steep mountains and on slopes overlooking the sea, where no evidence of a controlling ice body at a lower level is obtainable.”’ PHILOSOPHICAL CONCLUSION 47 - more conspicuous by such moderate widening of valleys and steepening and smoothing of the valley walls as the ice has possibly done. ' PHILOSOPHICAL CONCLUSION Even a small amount of ice wear may be conspicuous, while the actual amount of erosion will commonly be indeterminate. However, there are decided limitations to the possible degrees of erosion, and ignorance of the limit gives no warrant for excessive claims, even if it does give the opportunity. When along with the quantitative uncertainty of the amount of ice-work there is combined qualitative doubt in the diagnosis or interpretation of physiographic features, then the matter becomes chiefly a question of mental attitude and personal judgment. _ The claims for extreme glacial erosion have been almost entirely founded on physiographic characters—deep lake basins, fiords,and hang- ing valleys. This was true of Ramsay and his followers in England, of . Helland and his school in Norway, and of the group of eminent physiog- raphers in America. The assumption is made, without any attempt at proof except the physiographic argument, that Pleistocene glaciers could abrade thousands of feet in granitic rocks. The illogical argu- ment may be briefly stated as follows: Hanging valleys are common in glaciated regions ; they are not thought to bea normal product of stream work ; they may be explained by glacial deepening of the trunk valleys ; therefore the trunk valleys have been ice-deepened. But all the geologic evidence is to the effect that it was impossible in the time available for the glaciers to cut deep valleys. The advocates of ice erosion have never presented any proof from living or extinct glaciers that ice has made or is making or could possibly make a deep valley in hard, unweathered rock. The position is taken behind a bulwark of analogy and assump- tion. The burden of proof is properly on the advocates of unlimited ice erosion ; but it would seem wiser to essay the task of explaining anom- alous topography by the operation of ordinary and competent agencies. It is far more probable that some physiographic element has been over- looked, or that knowledge is deficient, or interpretation in error, than that Norwegian and Alaskan glaciers did a kind and amount of work which . glaciers in general have not done, are not doing, and apparently can not do. We do not yet know the erosional conditions and effects resulting from rapid uplift of high mountains with cores of crystalline rocks of hetero- geneous structure and under the probably warm and humid climate of the pre-Pleistocene. “ Hung up” valleys will probably be found a 48 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY normal product of atmospheric and stream work under some conditions of vigorous drainage, with or without rock displacement.* Part Il. Icr-sHeet Hrosion In NEw YORK GENERAL PRINCIPLES The previous writing (Part I) has treated specially of the work of alpine or stream glaciers. These did not exist in New York—at least in the part of the state which we will study. The phenomena here represent the work of a “ continental” ice body, and afford another critical test of the doctrine of deep ice erosion—particularly as they include the “Finger” Lake region. It needs to be clearly stated that any erosion of the basins of the New York lakes was not by alpine glaciers, but by mere lobations of the great Ontarian ice mass, as will be shown later. The work of continental ice sheets has not been so much in question. as that of stream glaciers; but the distinction often made between the erosive effects of alpine and continental glaciers is not well founded. Hrosion is dependent on the combination of the factors of velocity, press- ure, abrasive tools, and clearance, and any difference in the erosion by stream glaciers or continental glaciers is a matter of the intensity and reaction of these factors, the valid distinction being simply the variation of these factors, the same as between two alpine glaciers. However, it may be granted that the action of continental glaciers may involve con- ditions which can not be determined by study of stream glaciers. The more general effect of continenal glaciers has been regarded as that of leveling, by planing of elevations and rubbing of drift into the depres- sion. Yet, while the extravagant claims for the erosive effects of the “ Polar ice-cap ” made by the early glacialists have given place to more moderate views, the idea is still prevalent that the ice bodies had great excavating power. References are still made to “ deepening of the Great Lake basins” and to the “ glacial origin of the Finger lakes.” | The general failure of continental glaciers to effectively erode even soft deposits under their marginal portions is shown by the vertical suc- cession of glacialand interglacial beds left over wide areas of slight relief, as in the Mississippi area; but a theoretical distinction has been drawn ‘between peripheral zones of deposition and the central areas of erosion. Naturally the evidences of removal of material and abrasion of surfaces are pronounced in the centers of glaciation, and many observations are recorded which indicate superficial erosion, but not deep cutting of firm rock. * For an example of such features by faulting see ‘‘The hanging valleys of Georgetown, Colo- rado,” by W. O. Crosby, Tech. Quart., vol. xvi, March, 1903, pp. 41-50. ICE-SHEET EROSION IN NEW YORK 49 The Scandinavian geologists have claimed immense erosion of their highlands in order to account for the great volume of drift spread over the German lowlands and other areas marginal to the ice body. The probable exaggeration of the drift volume is accompanied by an under- estimate of the amount of rotted material in the crystalline regions, due to millions of years of preglacial decay. Such arguments are attractive in affording scope for’ the imagination, but they deal with uncertain elements and are very indefinite and inconclusive. The best illustration of the erosional effects of a great continental elacier is found in the broad area of Canada, and the facts are clearly stated in the following extract from a recent letter by Professor A. P. Coleman : _ “Tt seems to me probable that a vast ice sheet, like the great Labradorian glacier, would erode comparatively little near its center, much more powerfully midway from the central area toward the margin, and not at all at the margin. Your carefully studied New York region is too near the edge to show much erosion. The bottom of the sheet was clogged with coarse and fine debris by the time it reached you, and was incompetent to do much erosion. The same is true north of lake Ontario, as at Scarboro Heights, where the advancing Wisconsin ice pushed up over the stratified interglacial delta deposits of the Laurentian river with little effect, even over the upper stratified sand. “In the Sudbury district, however, much more scouring has been done, and very little moraine stuff and no preglacial weathered rock is to be found. All has been pared down to the fresh rock. The surface is, however, probably not very greatly different in relief from the preglacial one, since the harder bosses and ridges are still strongly marked hills, well rounded toward the northeast, rougher toward the southwest. Many of them have the moutonnée form, especially the granite hills. “The great morainic deposits of southern Ontario consist mainly of very fresh materials, even decomposable basic rocks, such as olivine diabases, coming out of the boulder clay in perfect freshness. Our boulder clay and morainic stuff do not show any signs of weathering or Oxidation of the geest. Perhaps all the surface material was swept farther south across the lake. ‘‘Though the amount of plucking and of erosion of live rock in this region, where I imagine ice erosion to have been most effective, is important, I do not estimate the average amount of cutting down of the surface as very great, probably considerably under 100 feet. We have numerous rock-basin lakes, but so far as I have seen all are smajl and not very deep.’’ The territory of New England and New York lies in the intermediate and debatable ground between central wear and marginal deposition. The practical absence of interglacial deposits in New York may be due, theoretically, either to greater erosive power of the later ice or to con- tinuity of the ice occupation. The high topographic relief of eastern New York is regarded as unfa- vorable to glacial erosion, but the great depressions followed by the ice 50 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY flow, the Hudson and the upper Saint Lawrence valleys, should display maximum effects.* A favorable area for ice erosion is the Ontario basin, and specially the plain and plateau border south of the lake, partly be- cause of the relation of the ice movement to the land surface and partly because the rocks are unusually soft and non-resistant. The Niagara plain was continuously overridden by the great Ontarian lobe of the Labradorian (Laurentian) ice body, while the Finger Lakes region re- ceived the full force of the southward flow of the spreading Ontarian mass. Western central New York thus offers a most excellent field for the study of glacial erosion, and it is a critical locality, since here more than for any other region in America the claim has been made for deep cutting by the ice. The writer has evidence that such claims are false, and the proofs will be given below in proper sequence. Another distinction between the work of continental and alpine gla- ciers must be mentioned here, for it seems to have been assumed that the supposed cutting in the Waeia of the Finger lakes was done by what were practically stream glaciers. To the degree that the margin of the ice body was ‘spate or had the flow concentrated along certain lines in the depressions of the land sur- face, the wear would approach that of stream glaciers; but there is an important theoretical distinction. The alpine glacier is supplied with rock rubbish and cutting tools by the weathering agencies at the valley head and sides. Like a river, the valley glacier is carrying a load of contributed detritus, which renews its cutting power along the margins. The ice mass which occupied the Ontario basin and the Finger Lake valleys had to do their work wholly by the slow abrasion process at the. bottom. The short lobes or tongues, which during advance and retreat of the ice sheet occupied the valleys of the Finger lakes, had very little resemblance to alpine glaciers in either origin, form, or effect, for it must be understood that they were pushing uphill. EFFECTS IN ADIRONDACK REGION ;. NORTHERN NEW YORK In the Adirondacks we have a highland area comparable in some ways with Scotland. -It was probably a center of local glaciation and ~was also overridden by the great ice body from the:north. While all students of the area find striking evidence of ice action, no suggestion of deep ice-cut valleys or basins have been made. Personal inquiries of two geologists who are very familiar with the field have elicited neither * A recent paper, ‘Trent River system and the Saint Lawrence outlet,” by A. W. G. Wilson, forming pages 211-242 of volume 15 of the Bulletin, furnishes conclusive evidence of slight ero- sive work of the Labradorian ice body north of lake Ontario and in the Saint Lawrence valley. The facts are presented on pages 221-224, with asummary on page 240, and are essentially the same class of evidence as given in the present writing for the south side of the Ontario basin. ICE-SHEET EROSION IN NEW YORK 51 facts nor opinions favoring deep localized erosion. Professor H. P. Cushing writes: . “My impression is that the northern slopes of the region have been consider- ably smoothed by ice action in its uphill climb there. There are cirques in the high peak district, and possibly a number of rock-basin lakes. Some of the passes may have been shaped by ice action. The rocks are very hard, but they are also much jointed, so that plucking might well go on along valley sides. There has been a tremendous amount of rock material, much of it very fresh, moved about by the ice, but it is of course impossible to say whether any amount of it was actually quarried by the ice or not. The joints would facilitate it if such action does take place. Certainly all rock in any way weathered was removed by the ice. The whole surface is rochemoutonneéd, especially on the north, where nearly all rocks are absolutely fresh. Even the diabase dikes, which are of pre- Cambrian age, often show the olivines perfectly fresh, and this when they cut _hard, resistant granitic gneisses, which are worn down to the same plane as the dikes. Iam sure there is no evidence in northern New York that the ice has cut any valleys.”’ . Professor C. H. Smyth, Jr., gives testimony to the same effect. The most definite observation directed to this subject has been placed on record by Doctor Gilbert * and it is negative. ‘This is the more sig- nificant, since it occurs in writing which favors the conception of ice erosion. After referring to the paper by Goodchild (see page 33) and expressing the opinion that the district of the Finger lakes “ owes more to ice work than to antecedent water work,” he writes as follows: ec ; , and in northern New York, where the rocks are comparable in hardness with those of the Scottish district, the ice seems to have accomplished comparatively little. Sandstones and limestones are not there so disposed as to afford good comparative data; but in a tract of crystalline schists lying northeast of Carthage and nearly bare of drift, the sculpture features are very different from those depicted by Goodchild. The principal structure of the schist is vertical, and its trend makes wide angles with the direction of ice motion. The ridges, which are at most only a few score feet in height, conform in trend with the strike of the foliation, and have been but slightly remodeled by sculpture on lines of ice motion. The bosses of the moutonnée pattern are measured by yards or rods.’’ The locality near Carthage, lying on the northwest flank of the Adi- rondack massif, should have experienced vigorous erosion, since the ice crowded past and around the highland mass. WORK OF THE ONTARIAN LOBE; WESTERN NEW YORK On the Niagara escarpment.—Fortunately for our study we have in the Niagara escarpment, which extends east and west, parallel to the south shore of lake Ontario (see figure 2),a broad feature of preglacial topogra- *G. K. Gilbert: ‘Glacial sculpture in western New York.’’ Bull. Geol. Soc. Am., vol. 10, 1899, pp. 121-130. 52 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY phy which has preserved not only evidences of ice work, but very satis- factory proof of the slight amount of work. Doctor Gilbert has drawn attention to the erosion phenomena in the short but interesting PARE already noticed and which we shall quote again. Against the Niagara escarpment the Ontarian ice body impinged in the most effective way, both in momentum and direction. The move-. ment was oblique to the cliff; and every mechanic learns that his cutting tool, chisel, rasp, or plane, hide its best work when the cutting edge is field oblique to the line of motion. . During all the life of the Laurentian ice in this locality, in all stages of its flow, the glacier struck this barrier at the very best advantage; and with what effect? The cutting by the: ice has been just about sufficient to prove that it was comparatively very’ slight. More or less cutting might have left the record equivocal. Doctor Gilbert has described (figure 2) how the ice cut oblique notches in the crest of the escarpment. A horizontal profile along the brow of LEWISTON | ot Oe rt e age PA KIN 10 2 4 : SMILES _Fieure 2.—Contour on the Lockport Limestone at the Niagara Escarpment. After Gilbert. The escarpment faces north. Arrows show observed directions of glacial striz. the scarp is serrate, with the serrations directed to the northeast, or against the ice-flow.. Many, if not all, of these notches were originated by preglacial agents, and the ice has rubbed them, sometimes enlarging them and changing their axial direction. Doctor Gilbert says: ‘‘All the more general features of the limestone belt thus seem to be preglacial.’’ After describing the limestone cliff in its form, relations, and erosion, he concludes as follows: ‘‘The configuration of the cliff seems to show that in the regions where the trend is southwest all minor salients have been pared away by the ice, and that where the trend is southeast minor irregularities of the face have been exaggerated and small reentrants drawn-into furrows; but the principal salients and reentrants of the topography are preserved, and ice modification is limited to minor details of form. ” “|. . It would appear that the ice sheet concentrated its work, so far as the Niagara limestone is concerned, on the crest of the escarpment, and that even there its results were of secondary rather than primary importance. Probably the . limestone at its escarpment lost on the average only 10 to 20 feet of thickness, and from the broad belt of outcrop the general loss may have been as small as 5 feet.’’ - With the above descriptions and conclusions the writer is in accord, and the estimate of removal on an average of 10 to 20 feet of rock from BULL. GEOL. SOC. AM. VOL. 16. 1904, PL 17 Froure 9.—Lockporr Limestone, Lockporr, N. Y. The limestone is at crest of Niagara escarpment. The derrick stands on perfectly preserved glaciation. At the left is seen higher and corroded beds Figure 2.—Lockpeorr LIMESTONE NEAR ROCHESTER Showing the deeply corroded dolomitie limestone FAILURE OF ICE EROSION ON LOCKPORT LIMESTONE BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 18 Figure 1.—AwN OvurLiecR CAPPED WITH ROTTED CORNIFEROUS LIMESTONE This exposure is one-half mile north of Honeoye Falls, New York Figure 2.—WaAvERLIME CEMENT QuarRRY, BUFFALO Corroded beds of Corniferous limestone appear at top FAILURE OF ICE EROSION ON CORNIFEROUS LIMESTONE Haas, tbe % BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 19 Figure 1.—CoRNELL QUARRY, HuULBERTON, NEw York Corroded sandstone layers disturbed but not removed Figure 2.—Horan quarry, Mepina, New York Corroded sandstone layers undisturbed FAILURE OF ICE EROSION ON MEDINAPLAIN BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 20 ~, f 5 he >. y tik | a 6 Figure 2.—‘**‘ Rock Criry”’ stRUCTURE IN ONKIDA GRIT This locality is 6 miles northeast of Oneida FAILURE OF ICE EROSION IN CENTRAL NEW YORK a an) ges bf ICE-SHEET EROSION IN NEW YORK 53 the crest of the cliff is acceptable when we include in the removal the weathered rock. No distinction was drawn between the weathered and loose rock and the firm rock, and this is an important point in discus- sion of ice erosion. It is certain that along this cliff almost no erosion of the live, unchanged rock occurred. At many localities a weathered layer may be seen in position. Sometimes the ice removed all the weathered rock down to a hard bed, while the latter is merely scratched. Plate 17, figure 1, shows an example; the derrick stands on live, striated rock, while at the left is the remnant of an upper and rotted layer. It is recognized, of course, that subterranean drainage and solution of the limestones probably extended to great depths; but the unequal re- sistance of the upper beds has frequently produced a very distinct plane between the severely weathered beds and an underlying bed that was slightly weathered (see plate 18, figure 2, plate 22, figure 2). All through western and central New York the preglacially weathered rock may be found in situ in numerous and critical localities. It would Ficure 3.—Typical Profiles of the Niagara Escarpment. After Gilbert. 1. Shale. 2. Lockport limestone, 3. Drift. seem to be conclusive evidence of lack of erosion, and will be referred to in discussing other localities. Plates 17-20, 22-23 area few illustrations, selected from many photographs, which show the preglacial weathering on sandstone, shale, and limestone. Brief descriptions are appended to the cuts. Along the Niagara escarpment the weathered rock fre- quently appears where the drift is removed. After looking at the figures showing the serrated cliff the reader might ask if the peculiar profile might not be the final effect of deep erosion, by lowering, or backward ice-cutting, of the cliff. Doctor Gilbert pre- sented conclusive topographic argument against this, and the above facts proving failure of erosion will place the matter beyond cavil. The slight erosion of the limestone cliff, and the failure to remove even the crest of shale where this was unprotected by a limestone cap, as shown by Gilbert, pages 123 and 124 of his article, and illustrated in figure 3, is satisfactory proof that here, at least, the ice had little erosive power. And why nothere? Every condition for vigorous erosion seems to have been fulfilled—an opposing cliff, not too high, composed of soft VIII—Bott. Geou. Soc. Am., Vor. 16, 1904 54 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY and jointed strata, trending at the very best angle for cutting, and the ice bottom freshly armed with Medina sands, gathered on the plain im- mediately northward, giving an abrasive much harder than the rock to be attacked. Now, it is no sufficient answer to say that ice behaves capriciously, and we should expect local variation in its work. We should expect it to get in its work where it had the best chance. If it were the great engine of erosion which plowed out the Finger Lakes valleys it should have exhibited some of its power in this locality. It is both illogical and unfair to claim that the ice was locally weak in those places where there is proof that it did not cut, and then to assume that it cut hun- dreds or thousands of feet in other places where there is no evidence of any cutting, but only a topographic difficulty. | On the Medina plain.—In the article already quoted Doctor Gilbert suggests some fluting by ice erosion in the Medina shales which underlie the glacial and lacustrine deposits along the smooth belt bordering lake Ontario, which we will call the Niagara-Genesee prairie. The present writer regards the low swells, lying in the direction of the ice movement, or northeast by southwest, as essentially drumlins. Near the Genesee river, in Monroe county, they are typical New York drumlins. West- ward in Orleans county they become lower and flatter, being only gentle swells, and they are finally almost imperceptible in Niagara county. Most of the western half of the prairie is perfectly flat to casual obser- vation. No rock exposures have been found by the writer on any ridge, although Doctor Gilbert refers to such, but the rock is frequently seen in the hollows or troughs in the stream beds. Along the lake shore the © waves of Ontario have dissected the plain, though not deeply, and the ridges and swells show only till. The singular parallelism of the stream directions, chiefly northeast- ward, was certainly determined by the surface form of the superficial drift, even if the ridges might sometimes have a rock core. However, the point is not worth any extended or detailed discussion here, for the matter in question is quantitatively unimportant. Doctor Gilbert sug- gests a furrowing of the underlying shales of only 40 or 50 feet, which is very moderate if the geest be included, as it properly should be; but even much more erosion than this in weathered Medina shales would be poor basis for claims of enormous cutting in granitic rocks. This paper by Doctor Gilbert is very significant in a negative way. It _should be expected that in a paper discussing and favoring ice erosion he would present the best evidence possible. He admits the absence of erosion on the schists near Carthage. On the Niagara cliff he con- cludes that the erosion was inconsiderable. The most he finds is some ICE-SHEET EROSION IN NEW YORK 55 removal of the Clinton along the escarpment and a furrowing of the soft Medina with “a general reduction of the surface to the extent of 40 or 50 feet, and the amount may have been considerably greater.” When we consider that whatever removal is admitted must include the weath- ered and loosened material, it makes a poor showing for ice erosion. Since Doctor Gilbert, with his long experience and wide observation in the field, with his special interest in glaciology, and with his command- ing ability, can give the erosionists no better support from direct evidence than this, they certainly have a very weak case. The statement is again pertinent, that no yood example of deep erosion has ever been proved. EFFECTS IN THE FINGER LAKES REGION; CENTRAL NEW YORK General description—The topography of central New York has long been known as striking in relief, peculiar in form, and puzzling in origin. A series of deep valleys, with a north and south direction and a north- ward pitch, holds a series of lakes having a digitate arrangement, of which Seneca and Cayuga are the central and larger members (see map, plate 21). These valleys are excavated in uppermost Silurian and De- vonian strata which are practically horizontal, but with a low southerly dip. Several of the valleys do not hold lakes. The valley walls are decidedly convex, and sometimes conspicuously smooth in general view, and southward toward the valley head are generally steep. The intervalley ridges are broad, rounded remnants of the north-facing slope of a dissected plateau (the Allegany cuesta), with an altitude at the north ends of about 500 to 800 feet above sealevel. Southward the ridges rise in 20 to 40 miles to about 2,000 feet, the higher strata being Portage-Chemung sandstones. The two larger valleys, holding Seneca and Cayuga lakes, have received most attention from writers and have been specially quoted as examples of deep ice erosion. But any sound theory for the genesis and history of these two valleys must also explain all the parallel valleys, from the © Genesee on the west to Chittenango on the east; they all belong in one category and have practically the same history. Another remarkable set of large valleys, which have not received much attention, are discordant in direction, and probably in altitude, with the north and south valleys. They have a general direction northeast and southwest, and intersect the north-south valleys. The most northerly example extends from the present head of the Onondaga valley, at Tully, northeast past Apulia and Fabius to the limestone valley near Delphi. Another one extends from the Cayuga valley at Ithaca northeast past Cortland and Truxton to the limestone valley at De Ruyter. Still another intersects, near their heads, the valleys of Canaseraga, Hemlock, Honeoye, 56 H. L. FAIRCHILD-—ICE EROSION THEORY A FALLACY and Canandaigua. These great transverse valleys have their bottoms obscured by drift, but the writer thinks that their rock bottoms are higher than those of the north-south valleys, and that they represent the work of an earlier drainage. The erosional history of the region from the time it was lifted out of the Devonian or Subcarboniferous sea, down to the time of the Pleisto- cene ice invasion, must have been eventful as well as long, and have in- volved great changes in attitude. We may believe that the primitive, consequent drainage from the Adirondack-Laurentian oldland was south and southwest across the new coastal plain to the Mississippian sea. The broad and high transverse valleys mentioned above may possibly be an inheritance from that earliest drainage. Eventually the broad Ontarian depression was produced as a subsequent valley along the outcrop and strike of the very thick and non-resistant strata of the Lower and Upper © Silurian, and along with the production of this great east and west trunk valley came the development of the (obsequent) drainage down the in- face or north slope of the plateau (the Allegany cuesta) which has left, as the finality of all the erosion, the north and south valleys that hold the Finger lakes. The above is merely a suggestion of the events in the long and obscure history ; but it seems probable that during the long exposure of the region to erosive agencies, during the later Paleozoic, Mesozoic, and Ceno- © zoic eras, many changes must have occurred and the normal development of the drainage have been modified by up and down land movements, with more or less tilting of the area.* The topographic forms and the valley relations have made this region an enigma to the physiographers. If all the rock topography as left by preglacial drainage were ex posed to view the expert translators of physio- graphic records might tell us most of the old history; but, unfortu- nately (or fortunately), the ice-sheet transgressed the entire region and rubbed its load of drift into the valleys in very irregular manner, and has badly obscured the preglacial topography. ‘The precise relationship of the valleys can not be known until borings determine the rock floors. The topography is anomalous, puzzling, and fascinating, and as it can not be immediately explained by supposed laws of stream work some students invoke the aid of ice and assume that glacial erosion is respon- sible for the unusual features. *Some suggestions of the history may be found in the following writings : A. W. Grabau: In New York State Museum Bull. No. 46, 1891. R. S. Tarr: ‘‘ The physical geography of New York state.”’ 1902. M. R. Campbell: Geographic development of northern Pennsylvania and southern New York.’’ Bull. Geol. Soe. Am., vol. 14, 1903, pp. 277-296. BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 21 Secale of Miles 10 20 P - > Ae ‘ i ° oe ‘ ones River ee SYRACUSE ® ~ 4)0, $ % on ¢ & ° \6 Senn NADICE £ A ~ x I A) = Q z 96 ay beste qneennenns” o~ Ss ~ 2 ‘a ost f) oteete"* ELMIRA {o ERIMINAL ) =. “ORR. M O\R ALN SEE PHYSIOGRAPHIC BELTS IN CENTRAL NEW YORK ICE-SHEET EROSION IN NEW YORK 57 Adverse argument from observed phenomena—Origin and basis of ice the- ory.—The suggestion of deep ice erosion as the cause of the Finger Lakes basins was definitely published by Lincoln * in 1892 and in 1894 by Tarr.t The argument is the discordance of the smaller tributary valleys in relation to their north and south trunk valleys. No other evidence of ice erosion has been given, and it is admitted that the broad inter- valley ridges have not been severely eroded, since transverse valleys occur both open and drift-filled, and that the drift is scanty in the wide belt which includes the several lakes. But it is assumed that the drift filling on the north is not sufficient to cause all the ponding of the waters, and that the valleys have been deepened and basined by ice erosion many hundreds of feet. The argument is essentially the same as for deep valley cutting in Norway or Alaska. As the writer views the matter, the advocates of the glacial origin of the Finger lakes should prove that ice did the work, or at least that it is possible for ice to do it, and not leave the proposition resting wholly on assumption. However, the writer will generously undertake to supply an argument that ice could not do the work, followed by positive proof that ice did not do it.t Stagnation of lower ice in the deep valleys.—The slopes of the val- leys and of the entire land surface in the region of the Finger lakes is northward, or against the ice-flow. When the district was buried under the larger ice body the flow of the ice must have been chiefly or entirely a flow of the upper layers down the sloping surface of the glacier. The ice in the depths of the larger valleys was probably inert or stagnant, and served as a bridge over which the upper ice traveled. When the ice was at its maximum, reaching the terminal moraine in Pennsylvania, the thickness of the mass over Seneca and Cayuga could hardly have been less than 4,000 or 5,000 feet. The high and irregular land: south and southwest of the valleys was an obstruction to the bottom flow. Another cause of resistance to flow of the deeper ice was due to the very important fact that the general ice movement during the greater expanse of the glacier was decidedly oblique to the valleys. ‘These combined conditions—the depth of ice, the opposition of the land surface, and the *D. F. Lincoln in Amer. Jour. Sci., vol. xliv, 1892, pp. 290-301. 7 R.S. Tarr in Bull. Geol. Soc. Am., vol. 5, 1894, pp. 339-356. The latter writing contains a bibli- ography on the subject. {Since these lines were written Professor Tarr has published an article, ‘“‘ Hanging valleys in the Finger Lake region of central New York,” Amer. Geol., vol. xxxiii, May, 1904, pp. 271-291, in which he argues against glacial erosion of Cayuga valley, using some of the facts and arguments presented in this paper. This isa reversal of the opinion expressed in ‘‘ The Physical Geography of New York State,” 1902, p. 180, where he says: ‘“‘The conclusions stated in my earlier paper have stood the test of much more extended studies, so that after seven years lam even more fully convinced that the two larger lakes owe their depth below the lake surface in large measure to ice erosion, and that they are in the nature of rock basins. Additional facts have been brought to light in support of this theory and none opposed to it.” 58 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY oblique push on the ice body (from the northeast)—must have produced stagnation of the ice in the bottoms of the valleys. The deficiency of drift filling in the deeper or more open sections of the Finger Lakes basins—in other words, the existence of the basins themselves—suggests comparative stagnation of the bottom ice in those sections. The great burden of subglacial drift was dropped along the drumlin zone (see plate —) at the north ends of the basins, where the transporting power of the ice lost efficiency. The low-lying but very heavy drift deposit north of the lakes, and which forms their northern barrier, was deposited previous to the retreat of the ice border along that belt, for it is crowned with drumlins which represent the molding by ice ~ of some depth and at some distance from the edge. It should be clearly understood that the drumlin belt filling is not terminal moraine, but subglacial or ground moraine, which the ice sheet rubbed into the valleys and subsequently overrode. If the deeper ice had possessed much move- ment it would have swept the drumlin belt drift filling farther southward into or entirely through the upper parts of the valleys. During both the earlier and the later stages, when the ice front was resting at the zone of the recessional moraine which now blocks the valleys and forms the present valley heads and water parting (see plate 21), the essential condition of the deeper ice was probably not unlike the former stage. The direction of ow was in line with the valleys, but the depth and pressure were less and the propulsion of the mass from the northward was not so forceful. It does not seem possible that any valley erosion could occur when the ice front was as far south as the valley-heads moraine. Any glacial wear on the bottoms of the Finger Lakes valleys must have been during other stages of the glacier. This suggestion of stagnation of the deepest ice applies in less degree to all the smaller Finger Lakes basins. Lobations of the ice front in the valleys—The only stage during which it is at all reasonable to suppose that the valleys could have suf- fered ice erosion is that phase of ice advance and retreat when the ice was thinner and the front formed lobations in the valleys. There never were any stream or valley glaciers in the Finger Jakes valleys. This important fact seems to have been overlooked. Valley glaciers are drain- age phenomena and do not flow uphill; neither do ice fields push glaciers uphill. When the ice front was south of the divide, on south- sloping surface, tongues from the ice front may have pushed forward down the valleys; but north of the divide the ice front had reentrants on the ridges and merely lobations in the valleys. Consideration of the mechanics of the glacier will show the necessity of this.: The valley lobes represent a small amount of concentration by the flow from the ICE-SHEET EROSION IN NEW YORK 59 higher ground either side of the valley, but they are chiefly due to the ~ relatively slower melting of the deeper portions of the front. Effect of deep waters facing the ice—Probably during the advance of the ice sheet, and certainly during its retreat, all the valley lobes were fronted by deep glacial lakes. It has been questioned whether the val- leys did not have reentrants in the ice front instead of lobations. In ‘any case, the water was unfavorable to elongation of the lobes, due to its melting effect, and unfavorable to ice erosion on account of ‘its buoyant effect. The earlier glacial waters had a depth in both Cayuga and Seneca of over 1,000 feet, while the Warren waters, during the withdrawal of the ice from the valleys, had a depth of over 800 feet in Cayuga and greater in Seneca. The detrital deposits left in the waters which faced the retreating ice give no evidence of having been formed along the margin of ice tongues, and the drainage lines — the valley walls show no diversion due to ice occupation. Whatever was the kind and amount of work by the ice in the Finger Lakes region, it was only that of a continental ice sheet and not that of stream or alpine glaciers. No erosion in front of zone of deposition. —The existence north of the lakes of the ground moraine valley-filling crowned with drumlins has been referred to above. Perhaps one might regard this deposit as an early terminal moraine overridden by later ice advance. Certainly it is not terminal drift of glacial recession. But, whatever the view as to its origin, it proves the transportational impotency of the ice along that zone, and it is unreasonable to suppose that the lower ice, on a rising slope, could effectively erode in front of this zone of deposition. It therefore seems impossible that the ice body which left the drumlin belt filling could have eroded the Finger Lakes basins. Of any earlier and more powerful invasion than that of the Wisconsin ice sheet we have no evidence whatever in this region. Absence of moraines in the basins.—There are no conspicuous mo- raines in the Seneca or Cayuga valleys between the heavy deposits left at the receding front, which now form the valley heads, and that which carries the drumlins and constitutes the blockade on the north. There are no pronounced valley moraines except at the valley heads. For some 50 miles, the stretch between the two deposits named, the val- ley walls are comparatively free from localized drift. This absence of recessional moraines is a surprising and important fact. Some accumu- lations may be covered by the deep lakes; but surveys do not reveal them, and the lakeless valleys and those with only shallow lakes are also characterized by general absence of valley drift and the smoothness of 60 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY the valley walls. This is inconsistent with the idea of erosion. The only explanation seems to be that the ice contained little drift in front of the drumlin belt zone of deposition, or that the valley ice was inac- tive and melted back without long pause. It is not thought possible that there could have been any serious erosion by valley lobes, saying nothing of ‘“ hundreds of feet’ of cutting, without leaving conspicuous piles of debris, even in the. presence of the glacial waters. Small volume of the valley-heads moraine.—The total volume of the valley-heads moraine, after making liberal allowance for the valley train drift south of that moraine and the fine material carried to the sea, is probably no more than should have been gathered by the ice from the supply of weathered material or geest which the ice found ready to its grasp in the district immediately northward. The valley-heads moraine is probably much less in amount than the mass of drift lying in the drumlin belt on the low ground between the Finger lakes and lake Ontario, and there are no heavy moraines south of the valley heads. Where is the immense volume of drift which would have been produced by excavation of the valleys to depths of hundreds of feet, by the Wis- consin or any other ice sheet? | | Idea of rock basins an assumption.—No real evidence has ever been presented to show that the lakes are not entirely due to barriers of drift ; or, in other words, that the rock bottoms of the valleys are not graded from their sources, south of the present valley heads, to the On- tario valley or to their trunk valleys. In figure 4 is shown the relation in vertical plane of the bottoms of Cayuga, Seneca, and Ontario valleys Ocean level * —54f Greatest depth of Cayuga ——m ~175f Greatest depth of Seneca ——aamme= Greatest depth of ontario =—492F. Figurs 4.—Diagram showing Vertical Relation of Cayuga and Seneca Valleys to Ontario Valley. Vertical scale about 100 times horizontal. at the present time. This shows a gradient from the deepest part of Seneca (somewhat south of the middle of the lake) to the abyss of On- tario of about 5 feet to the mile. Any filling of drift in the Seneca basin that would reduce this gradient for the rock valley may be offset by the probable drift in the Ontario basin. There seems to be nothing in our knowledge of the lake basins inconsistent with the idea of preglacial free northward drainage. Well borings will probably prove the fact of ° graded valleys of originally high gradient. ICE-SHEET EROSION IN NEW YORK 61 Convexity of valley sides.—Cross-sections of the Finger Lakes val- leys (figure 5) show convex valley slopes, not the concave profiles and U shape attributed to ice-worn valleys (see page 28). It would seem unreasonable to claim that the glacier cut hundreds of feet merely at - I I . 14% Canandaigua valley. I. Section at head of lake. Il. Section at Long Point. Keuka valley. Section at Grove Spring =| I I 0 W Seneca valley. I, Section 4 miles from head of lake. III. Seetion at Lodi. II. Section 1 mile south of North Hector. IV. Section 4 miles north of Willard. Cayuga valley. I. Section 2 miles from head of lake. II. Section 4 miles north of Trumansburg. Ill. Section at Aurora, ee ONS Owasco valley. Skaneateles valley. Section at Ensenore, 34% miles . Section at Glen Cove, about 4 miles from head of lake. from head of lake. Figure 5.—Generalized Cross-section Profiles of Finger Lakes Valleys, looking north. Plotted from the topographic sheets; the vertical scale 6.5 times the horizontal. the bottom, and not enough along the sides, in soft shales, to produce concave profiles of the valley walls. E The profiles beneath the lakes are apparently harmonious with the convex profiles of the exposed walls. Other valleys without lakes show the bottom contours, with some filling. As should be expected, these are IX—Butu. Grou. Soc. Am., Von. 16, 1904 62 H. L. FAIRCHILD—ICEK EROSION THEORY A FALLACY somewhat variable, but none of them are the typical U shape of the glaciated mountain valleys. 2 Rock cliffs.—A large part of the shores not only of Seneca and Cayuga, but of the smaller lakes as well, are nearly vertical rock cliffs, These may be regarded as the product of wave work on decomposable shales that are intersected with innumerable joints, but there is abundant testimony, of people living by the lakes and familiar with their features, — to the existence of submerged cliffs and ledges. These might be attrib- ROUTE i] | \ \ Hibiscus Pt ais NN ; EW SZ Figure 6.—Map of Cayuga Lake Shores at Union Springs. From the topographic map. uted to wave work when the lakes had a lower level, before the differ- ential northward uplift had lifted the outlets and so raised the water levels to the present planes. Moreover, there are cliffs, at ]east in the Seneca and Cayuga valleys, much above the present waters. These might be credited to the local high-level glacial waters which occupied the valleys during the ice retreat. ICE-SHEET EROSION IN NEW YORK 63 Certainly rock cliffs, with or without angles, are not probable features by glacial abrasion, though they might be produced along the edges of stream glaciers aided by weathering. However, there were no stream glaciers in the Finger Lakes valleys. These cliffs are either the work of preglacial streams or of postglacial waves; they are not ice work. Islands and capes of rock in Cayuga basin.—On the east side of Cayuga valley, at Union Springs, the shore has several sharp rock pro- jections and one island of limestone, Frontenac island, the latter a half mile from shore. These features are shown in figure 6. The boat pilots say that other submerged rocks lie offshore. Such features are emphat- ically inconsistent with the idea of glacial enlargement of the valley at this point, as any severe effect of the glacial plane would straighten the valley walls. They are normal products of the work of atmosphere and water. This point might properly be included under the next main heading as direct proof of non-erosion by ice. Direction of the ice flow due to the topography.—The radial or digital attitude of the Finger lakes has been regarded by erosionists as a possible effect of the spreading flow of the ice sheet. If the valleys were radiating on downward slope the conception would not be entirely un- reasonable, but they are spreading on upward slope. Undoubtedly the ice flow in the earlier and later stages may have been guided by the val- leys. During the waning of the Ontarian ice body the spreading flow of the ice was up the radiating valleys; but this direction of flow was an effect, and not a cause, of the valley topography. The ice favored the lowest ground, and the deeper ice and stronger flow was naturally along the depression of which Seneca and Cayuga are the broad axis. During the most forceful stage of glaciation the direction of flow was obliquely across the valleys. Transverse valleys.—These have already been mentioned (pages 55- 56). The transverse valleys are positive proof that the ice did not carve the more elevated topography. South of the divide, where the ice moved on downward slope, the transverse valleys and the irregular topography are proofs that the ice did not obliterate the relief features, even where it had greater fluency. In the belt of the Finger lakes the transverse valleys exist, although somewhat obscured by drift filling. Along the divide and southward the preglacial topography is conspicuous. It would seem unreasonable . to claim for a continental ice sheet the power of cutting trenches below some plane while above that plane the relief features are scarcely affected ; yet the claim of glacial origin of valley discordance in central New York requires essentially that assumption. 64 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY Adverse philosophy.—The considerations presented above bear di- rectly on the local problem, and they should alone be convincing of the fact that the Finger Lakes. valleys were never deepened by ice ero- sion. The general theoretic argument against the competency of ice to deepen valleys has been purposely waived to this moment in order to give force to the concrete phenomena, but it should now be emphatically stated that the theoretic objections outlined on pages 25-31 apply with special force to this case. Theoretically there never could have been any effective cutting by ice in the deep bottom sections of Seneca and Cayuga, and really there are no phenomena nor features of any sort which require or even strongly suggest it. . | Direct proofs of non-erosion.—Finally and conclusively we have in the valleys of the Finger lakes the same indisputable evidence of lack of ice erosion which has been described for the Niagara escarpment. In quarry exposures along the Seneca and Cayuga valleys and elsewhere. it may frequently be seen that the upper layers of rock are weathered and rotted far beyond what is possible since the ice retreat. Abraded surfaces of fresh rock of the same kind, in the same localities, even in the same exposures, give comparison for postglacial decay which for till-protected surfaces is practically nothing. Preglacially weathered rock in position may be recognized in a vast number of quarries or freshly cut rock sections throughout western- central New York. In the Finger Lakes district the rocks are chiefly shales, which are not favorable for preservation or discrimination of the old decomposition product. Fortunately there are numerous limestone quarries along the belt near the north ends of the valleys, and a few openings in the Tully limestone toward the upper end of at least the Cayuga valley; and limestone preserves the record in perfection. To one on the ground the evidence is perfectly clear, but the photo- graph does not give the differences in color and texture nor show the finer effects of solution and corrasion. However, the facts are well shown in plates 22 and 238, which are only a few examples from many that might be given. Plate 22 gives an example from the Onondaga valley similar to the plate 18, from Buffalo. This is a photograph of the east side of the Indian quarry, about 7 miles south of the center of Syracuse. The elevation is 580 feet and only about 120 feet above the bottom of the valley. The east wall of the valley rises 800 feet above the quarry. The preservation of the finest strie and polishing of the limestones beneath even a thin mantle of till proves the incompetency of postglacial weathering to pro- duce the degree of corrasion found in upper layers. In some parts of BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 22 Ficure 1.—Goopricu QuARRY, ONONDAGA LIMESTONE, Aupurn, New Yore Corroded beds in place at top FiGguRE 2.—INDIAN QUARRY, ONONDAGA LIMESTONE, 5 MILES Sourn or Syracuse, New York Corroded beds in place at top FAILURE OF ICE EROSION IN ONONDAGA AND OWASCO VALLEYS BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 23 Figure 1.—THOMAS QUARRY, ONONDAGA LIMESTONE, WATERLOO, NEw YorK Corroded bed glaciated but in place; buried in till Figure 2.—Porthanpd CEMENT CoMPANY's QUARRY, TULLLY LIMESTONE This quarry is 6 miles north of Ithaca, New York. Extremely corroded beds glaciated, but in place. These are removed from the glaciated foreground, but show in the background FAILURE OF ICE EROSION IN CAYUGA REGION ICE-SHEET EROSION IN NEW YORK ; 65 the Indian quarry the corraded layer shown in the view is removed and the till rests directly on the subjacent, firm, unweathered bed. Plate 22, figure 1, is from the Goodrich quarry, near Cottage street, in the city of Auburn, lying in the northward continuation of the Owasco valley. Plate 28, figure 1, is from the northside of the Thomas quarry, 1 mile south of Waterloo, in Corniferous limestone. About 2 feet of the base of the mound that looks like till, in the center of the view, is a bed of greatly corraded rock, but with the top of the weathered blocks well glaciated and with perfectly preserved polish. In the left of the view the same bed is shown quite unweathered, while on the west and south sides of the quarry the same bed, in perfect freshness, is buried under higher and weathered layers. In the Cayuga valley the proofs of non-erosion by the ice are abundant. At Union Springs the limestone quarries yield the usual good evidence, as do also the gypsum pits. A good example of preglacial weathering is found in the Portland Cement Company’s quarry, in Tully limestone, about 6 miles north of Ithaca, on the east side of the valley. Plate 23, figure 2, shows in the background remnants of a much corroded bed, about 4 feet thick, having the open joints filled with red, residual clay, Striz on the tops of the corraded blocks and elegant glaciation of the firmer bed in the foreground of the view prove the lack of postglacial weathering. The large, open joints and seams in the subjacent beds are also filled with residual clay, so characteristic that the quarrymen recognized the material as different from the till. This quarry is only about 200 feet above Cayuga lake. These facts of observation, which can be indefinitely multiplied and easily verified, prove beyond any reasonable doubt the failure of the glacial ice to remove even the superficial, weathered rock, even at low altitudes in the Finger Lakes valleys. It is plainly evident that the ice did not produce the valleys; it did not even enlarge them; and it would be unwarranted, in view of our knowledge of the mechanics of glaciers, to claim that the ice trenched below the present lake levels without effective cutting above those levels. The most that can be reasonably claimed for the ice-work is that it smoothed off the intervalley ridges and also the valley sides. The val— leys are stream valleys, like valleys everywhere, and only slightly modi- fied by ice action. Let us hope that assertions of the glacial origin or deepening of the Finger Lakes valleys (or any other valleys) will cease, and that former statements to that effect will be corrected. 66 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY History of the valleys—Preglacial conditions.—It does not seem proper to leave the subject of the Finger Lakes valleys with only a negative conclusion. Following is an outline of the sequence of geologic events, as the writer interprets the history. A brief statement has been given on page 56. During the immensely long pre-Pleistocene time the region was probably exposed to subaerial erosion, and many changes must have occurred in land movement and attitude of the region, which had effects on the development of the drainage. The final result was the north-flowing streams and the hard- | rock topography which we now find partially buried under glacial and lacustrine drift. A careful and patient investigation, without too much theorizing far in advance of facts, will probably discover some of the geologic events in the history, specially the later ones. Possibly there may be geologic factors involved of which we have not thought. The ereat desideratum for the study is data concerning the rock forms. Preglacial drainage.—The regularity and parallelism of the north- ward drainage is a striking fact, doubtless explained in part by the uniformity in structure and attitude of the comparatively soft strata. It is believed that at least the stronger streams, such as the Senecan and Cayugan rivers, were graded to the Ontarian river, but with a high gradient. The north-flowing streams were so closely spaced that the east-west streams were short and weak. Possibly in this may be found a hint of one cause of discordant drainage. Whatever may be the truth regard- ing this in other regions, in this area any serious discordance must be explained by the interaction of subaerial agents, with this qualification, that the very moderate amount of ice planing on the sides of the larger valleys in some places may possibly have accentuated the abrupt termi- nation of some high-level side valleys, but this widening of the valleys has not been an important effect. | Ice invasions.—We may not be positive regarding the number of ice invasions in central New York. At present we have evidence of only one, the Wisconsin; but the certainty of several glacial epochs in the Mississippi region and the accumulating evidence of more than one epoch in the New Jersey-Long Island district, with the evidence of inter- glacial epochs at Scarboro Heights, Ontario, should make us watchful for records of multiple invasions here. On first thought it might seem as if the occurrence of ice invasions previous to the Wisconsin would be favorable to the argument for ice - erosion, but such is not the case. If the invasions were separated by long interglacial epochs, giving opportunity for mature stream work, the ‘ eS oe - ICE-SHEET EROSION IN NEW YORK 67 argument from topography could safely apply only to the latest inva- sion, and if the invasions were simply cumulative in their effects, then all the facts and reasoning as given on pages 57-64 would apply in the same way as to only one invasion. It is safe to discuss the history of the region as involving only the Wisconsin glacial epoch, for no evidences of any earlier and more forceful or extended sheet have been found. If the Wisconsin sheet failed to seriously erode it is unreasonable to appeal to earlier invasions, that were certainly smaller and weaker. Effects of ice advance.—When the oncoming ice sheet transgressed our district, it might have had the same directions of flow and the same marginal form as during its retreat, which supposition is the most favor- able to erosion. The transportational and erosional effects were doubt- less the same in principle and general character, but there was certainly a difference in effect or degree. The ice found all the land surface, valley and hill, mantled in a thick sheet of residuum, which had to be cleared away before bed-rock could be affected. In addition to the local geest, the ice was already burdened with its enormous load of subglacial drift that it had gathered on its way; for it is not thought that any local glaciers could have affected this locality in advance of the continental sheet. With all its bottom load of debris, producing stagnation in the lower layers and acting as a buffer for the underlying rock, erosion must have been practically if not entirely nil. The advancing valley lobes were doubtless faced by lakes, as during their retreat. The upgrade which the ice had to climb is unknown, but it was certainly an uphill advance. The dropping of its basal drift at any line, as the drumlin- belt moraine, for example, would give no increased erosional power to the onmoving ice. 3 All considerations lead to the confident conclusion that the ice sheet during its advance did no erosional work. Its effect was to transport and distribute the debris which it had been forced to carry. Conditions during maximum extent of the ice sheet.—We may sup- pose that as the valleys become filled with ice during the slow advance of the great ice body, and the latter rolled its front on to the higher ground southward, the ice was less and less diverted by the topographic features until finally the general movement, at least of the upper layers, _was toward the southwest, and obliquely across the valleys of the Finger lakes, specially those east of Keuka. This conclusion regarding the direction of flow is derived from the trend of the great terminal and the recéssional moraines, which were normal to the ice movement. All our knowledge of the behavior of glaciers leads to the conclusion that the deeply buried ice in the valleys of the north-facing slope must 68 H. L. FAIRCHTLD—ICE EROSION THEORY A FALLACY have been comparatively stagnant and impotent. The thickness of ice over Seneca valley, when the ice front was at the terminal moraine in Pennsylvania, has been estimated by the writer as at least 4,000 feet.* Other estimates have made the depth much greater. Certainly the press- ure due to depth was very great, and it was partly resolved into a poten- tial pressure down the slope, or northward, and against the general movement. Three principles of ice movement were in play, namely, practical viscosity, rigidity, and shearing. The problem is too compli- cated to be here discussed, but the best conclusion is to the effect that during the burial of the region under the general ice sheet the deep ice was stagnant and erosion of the Finger Lakes valleys was impossible. The stagnating effect of the bottom debris must be taken into the account along with the mechanics of the clear ice. Valley-heads moraine.—With the waning of the ice body the front receded to the line which connects the divides and moraines at the present heads of the Finger Lakes valleys. There the ice front lin- gered some time and accumulated the valley moraines. The front of the ice was probably lobate in the valleys, and there the morainal drift was concentrated; but it should be distinctly understood that the morainal masses at the heads of the Finger Lakes valleys were not made by “valley” glaciers, but by lobations at the edge of the great Ontarian lobe of the Labradorian ice body. When the moraine was forming at the present head of Seneca valley the depth of ice over the valley must have been over 2,000 feet, and the lower ice was doubt- less inert. The moraines at the valley heads can not reasonably represent erosion from the bottoms of the valleys for reasons which have been given. This is an important fact which writers have overlooked, and specially so since these are the only large drift accumulations which the advocates of valley erosion can refer to. They are too small in vol- ume and too largely composed of far-traveled material to represent much valley cutting. Effects of ice retreat.—From the line of the valley-heads moraine to the drumlin-belt filling, a distance of about 50 miles on the Seneca Lake meridian, there are no morainal deposits in the valleys which have been thought worthy of notice. Careful search by experts will doubtless find lines of marginal drift, curving over the ridges, which will be useful in proving the form and amount of lobation of the valley ice in its ***Glacial geology of western New York.”’ Geol. Mag., Dec. iv, vol. 4, no. 402, December, 1897, p. 532, ICE-SHEET EROSION IN NEW YORK 69 retreat.* The drainage courses down the slopes give no indication that they were initially determined by oblique ice margins. It is apparent that the ice receded across the belt of the Finger lakes so steadily and rapidly as to leave 10 conspicuous marginal deposits. Surveys of both Seneca and Cayuga lakes prove that there are no large masses of drift beneath the water. None have been noted in any of the valleys, whether open or containing lakes. This absence of drift masses in the valleys north of the valley heads would seem to prove that no large amount of debris was produced by erosion after the ice border left the valley-heads moraine. In other words, there was no erosion of the valleys during the retreat of the ice, and consequently there was no ero- sion at any time. The positive proof of no serious erosion has already been given on page 64. Local glacial lakes—While receding northward over the north- sloping valleys the ice front acted as a dam to waters held in all the val- leys. These local glacial lakes have been named and described in other writings.— They are referred to here not only because their episode forms part of the history of the region, but also because they had some- thing to do with the valley forms. Below the highest level of these lakes, which was determined in each case by the elevation of the col across the moraine at the valley head, the slopes at the south end of the valleys felt the action of the laving waters. Some indefinite part of the smoothness of the slopes and the existence of the rock cliffs are due to the wave action. Along Seneca valley the waters of lakes Watkins and Newberry reached above 900 feet, or about 500 feet over the present lake. Lake Ithaca, in the Cayuga valley, southern end, had an elevation of over 1,000 feet, or about 600 feet above the present lake. The deltas built by land streams on the valley slopes in the high waters of the glacial lakes are conspicuous for their form rather than their volume, and are the most striking proofs of the existence of these lakes. The precise effects on the ice margins of the fresh waters of these deep Jakes is not known. They probably facilitated the melting of the ice, and so helped to prevent strong lobation, and by insinuation beneath the ice they probably had a buoyant effect and helped to reduce the pressure of the ice on the valley bottoms. * Professor Tarr has told the writer that he has been able to trace some faint lines of marginal drift left by the receding ice in the Cayuga valley, which show merely lobations of the ice front. +H. L. Fairchild: ‘Glacial lakes of western New York,” Bull. Geol. Soc. Am., vol. 6, 1895, pp. 353-374; ‘Glacial waters in the Finger Lakes region of New York,” ibid., vol. 10, 1899, pp. 27-68 ; “Glacial lakes Newberry, Warren, and Dana in central New York,’ Amer. Jour. Sci., vol. vii, 1899, pp. 249-263. X—Butt, Geox. Soc. Am., Vou, 16, 1904 70 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY Drumlin-belt filling —The heaviest deposit of drift in central New York is that belt lying north of the Finger lakes and which carries one of the most remarkable areas of drumlins in the world. It has not been recognized as morainal for the reason that it is so low-lying as to be buried or leveled by lake action. It forms the blockade to all the north and south valleys, both those with lakes and those without lakes today. The Flint and Bristol (Ganargua) valleys held shallow postglacial lakes which have been changed to swampy tracts. The belt of drift lies deepest over the east and west outcrop of the soft Salina shales and north of the outcrop of the Helderberg-Corniferous limestones. It was all beneath the glacial waters of lake Warren and the slowly falling waters succeeding the Warren (Hyper-Iroquois) ; and all the area of the Sodus-Cayuga depression was under lake Iroquois during all the life of the latter (see plate 21). It may be suggested that a large portion of this drift was rubbed into the valleys along the Salina outcrop during the early stages of the ice sheet, and thereby saved the more southerly part of the valleys (now occupied by the lakes) from great filling. The outcrop or low escarp- ment of the limestones was the most obdurate barrier which the ice en- countered on the Ontario plain. The drumlins were probably formed during the closing stages of the ice. The lakes have spread their silts over the lowlands and partially buried even the drumlins, while vegetal deposits have continued the filling, and form the vast peat marshes of the Montezuma swamps and along the Seneca river. Buried valleys—The writer is not able to fully prove that the val- leys are graded in rock and are merely dammed by drift, but all the available facts point to this, and it is believed that borings will so prove it. The assumption that any of the lakes have a rock barrier on the north is without any warrant in knowledge or in sound reasoning. From the foot of the Finger lakes northward to near the Ontario shore — the valleys are buried quite out of view beneath the drumlin-belt drift filling, but they partly reappear in the depressions along the Ontario shore. Ifthe reader will spread before himself in order the Macedon, Pultneyville, Sodus bay, and Oswego sheets of the state topographic map he will see that the Ontario shore from Irondequoit bay eastward to Sodus bay, a distance of about 25 miles, has no embayments. The Medina rocks are here visibly continuous, and no large streams ever traversed this belt of land. They were diverted either westward to the Genesee (Irondequoit gulf) or eastward to the Seneca-Sodus valley. At Sodus bay is a great break in the shoreline, and a broad valley intersects the rock strata. ‘This ancient valley is not less than 2 miles wide, but ICE-SHEET EROSION IN NEW YORK 71 its walls are buried under drift, while 6 or 7 miles south from the Ontario shore the entire valley is hidden from view. The Sodus valley lies in direct line with the axis of the Seneca valley, and it is not a forced as- sumption that the Senecan river flowed through the Sodus valley, and that the Keukan, Flintian, and Canandaiguan rivers, lying westward, were tributary to the Senecan, probably bending eastward along the strike of the less resistant strata (see plate 21). The stream in the Mud creek (Ganargua) valley may have turned westward to join the Genesee. Eastward from Sodus bay the maps show a series of bays or low ground. In order eastward these are: Port bay, Red creek, Black creek, Blind Sodus bay, and Fairhaven bay. The drift obscures the hard rock geology, but these sags probably indicate buried valleys. The precise correlation of these embayments with the Finger Lakes valleys is, of course, uncer- tain, but test borings will probably show that the Owascan, Skaneatelan, Otiscan,and Onondagan rivers occupied the buried valleys. The Cayu- gan river possibly united with the Senecan. Greater glacial lakes.—The story of the later glacial waters has been partially told in other writings.* The important fact here is that not only all the glacial drift left by the receding ice in the lower parts of all the Finger Lakes valleys was deposited under water, but that north of the lakes all the drift under about 880 feet elevation was so deposited, either in lake Warren or its successors. In consequence of this fact the glacial debris is much distributed and blended with or buried under the lake deposits, and such morainal topography as shows at all is much subdued. ‘The latter is chiefly kames or knolls of gravel. Lake Warren came to extinction by the draining to lower levels through channels in the neighborhood of Syracuse, which carried the water east- ward to the Mohawk valley. These falling waters have collectively been called Hyper-Iroquois. Finally, with the establishment of the outlet at Rome, came the long-lived lake Iroquois which has left the strong shore phenomena about the Ontario basin. When the ice blockade in the Saint Lawrence valley was finally lifted the region of the Thousand islands was about 40 feet below sealevel, and marine waters, the Ontario gulf. occupied the basin. As an effect of the northward uplifting of the entire region which has occurred since glacial time, and is still in progress,t the Thousand islands have been lifted toward 300 feet, and *In addition to references given above, consult papers in the Twentieth, Twenty-first, and Twenty-second Annual Reports of the New York State Geologist. A map of lake Iroqnois is in the twentieth report, and one of lake Warren in the twenty-second. +G. K. Gilbert: ‘‘Recent earth movement in the Great Lakes region.’ Eighteenth Annual Report U.S. Geol. Survey, p. 604. H. L. Fairchild; “‘ Land-warping in western New York.’’ Bull, Geol. Soc, Am,, vol. 10, p. 66, 72 H. L. FAIRCHILD——-ICE EROSION THEORY A FALLACY the possibly brackish waters of the Ontario gulf have been rinsed out to form lake Ontario, the latest representative of the series of water bodies in the Laurentian basin. . Present lakes.—Repetition is unnecessary here of the facts and con- clusions given in the preceding pages. The principal factor in the genesis of the Finger lakes is probably the drift barriers, along with differ- ential. northward uplift. The postglacial tilting seems to range from 2 or 3 feet per mile in the Seneca valley to 5 feet per mile east of lake Ontario. We do not yet know the changes in land attitude which occurred before ~ and during the Pleistocene. There may be some unrecognized factors in the problem, but the idea of “rock basins” has no firm basis in obser- vation, and ice erosion has had but small effect on the region. It is granted that the valley slopes were rubbed by the ice to a smoother contour; that some minor side valleys may have been obliterated by filling; that it is possible, though not probable, that the ultimate and. steep portions of some small side valleys might have been so effaced by ine, Pongal of the weathered walls of the main valley as to give a “ hung up ” appearance to the tributary valley. The following table of altitudes of the central New York valleys and lakes gives the quantitative data bearing on the subject: Vertical Relations, in Feet Above Tide, of the Central New York Valleys and Ontario Basin Height Greatest | Depth of above depth of |lakes below sealevel. lakes. sealevel. Conesus lake. 6.0% babe, Pan Gee UR NEE A hes rege Sa 818 Shallow. emilock jake: ooo 35 sah aes gine grees ates 896 86 Ganadice Wak@..<.2) ai. uae ache ee wie Se ee 1,092 86 Honeoye Waker 78 cee ecns oe oo ee ey 800 Shallow. iad creck. ds wikcs be eae ee Le PL Maree Pe en e 1,000-900 ae Canandaimna lake i. pissin Nien ead mane 686 258 HIM OEE acne chi. seme meh tek fee .| 1,000-900 ae Keuka lake eso: orth cht es Soa eae eee ee 709 179 te Seneca lake....... oie ie Mote onkt cone sa ciel nie ond 444 618 175 ONPARTO TAK be ui piss bot cake tage ee eee eee 246 738 492 Camara MAK so 's oe te tae winnie ate eae Raia act 381 435 54 Owasco Jake? oo: %..2e cases PRE a AM Bra RC 2 710 177 oe Siena Le bem Laces 6 cisco isuaeter ca coer a aereoene 867 294 PETER ACO. icon: aia is ok Sie eee Oe OER eRe bon ee 784 65? Qnondape Creeks: si Po eos swe ee cea 600-400 ae Butterwut creeks seb e Pet ha eee 900-700 LAeSSONE CLOG. < ck iee ee Siete ee 800 GENERAL SUMMARY 73 The altitude for creeks given in the above table is that of the valley section lying in the Finger Lakes belt. The figures are given on the map, plate 21. It will be seen that the western and eastern members of the lake series are comparatively shallow, and that drift damming is the reasonable explanation of their basins. This is equally true of Cayuga and Seneca when the depth of the buried section is considered. The Finger lakes are all properly classed as “ morainal.” GENERAL SUMMARY The first part of this paper discusses the question of glacial erosion theoretically, having special reference to alpine glaciers, with illustra- tions drawn from glaciated areas. The term erosion is restricted to signify removal of unweathered rock. After showing that the argument for erosion is inferential and chiefly based on uncertain topographic forms, an argument for erosion is formulated, which is then considered in detail and found inconclusive. A general negative argument based on recognized principles of glacier physics proves that bottom erosion of glaciers must be only by the slow process of abrasion, which is a self-checking process; that glaciers tend to widen rather than to deepen valleys; that glaciers can not deepen without widening their channels, and that all theory and observation lead to the conclusion that glaciers are ineffective agents in production of valleys or basins. The subject is then illustrated by examples drawn from areas of past and present gla- ciation in the northern hemisphere, with citation of observations and opinions by students of those areas showing the absence of any direct evidence of erosion, and that the chief claim for erosion is for Norwe- gian and Alaskan fiords and some Alpine valleys made by the physi- ographers on the ground of valley discordance. It is shown that such explanation does not harmonize the body of fact, but produces confu- sion. Lastly, it is shown that the ice-erosion argument for hanging valleys is illogical without proofs of the competency of glaciers, and that these features will undoubtedly be explained as normal product of atmospheric and stream work under conditions not yet understood. The second division of the paper treats of the erosional phenomena * in New York state as a critical area specially illustrating the work of a continental ice sheet. After showing that all observers agree on the slight erosional effect of ice on the Adirondack highland and in the ‘Saint Lawrence valley, the area of western New York is found to yield the same result. Central New York, or the district of the Finger lakes, 74 H. L. FAIRCHILD—ICE EROSION THEORY A FALLACY is then studied at length. It is shown that the valleys of the Finger lakes were never occupied by stream or alpine glaciers, but only by lobations of the ice-front during the advance and retreat of the Lauren- tian ice body; that the flow of the ice was against the slope, or uphill, and during the most forceful stage was oblique to the valleys; that the lower ice was probably stagnant; that erosion was unlikely in front of the zone of deposition on the north; that the convex profiles of the val- ley walls prove the lack of valley widening, and that other features help to a theoretical conclusion adverse to ice erosion. Direct positive proofs of non-erosion are then presented in the existence in the valleys of remnants of preglacially weathered rock. The history of the valley formations and their later hydrography is then briefly outlined. Fi otal BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 75-S0 FEBRUARY 27, 1905 HANGING VALLEYS BY ISRAEL C. RUSSELL (Read before the Society December 30, 1904) CONTENTS Page MERRILL" SIQMEUINE VAMON 5. bce ie ns oes 0ln 2d cose e gain gus eb Na ceers 75 Classification........ wna he REN SE HG a 1 a 8 a A er a OO A ae are 76 SUEDE TSE TaS tS anil ihc AS RR Rae eigen Pa ee 76 SnIrITIP nite Valleys. . 6 cack ik kt ees leet cede aed oad eee wees cit Seeeetormed banging valleys... 12.26). 5 hil Se ede le we eee es ORIEN O. 77 Diastrophic hanging valleys........ WA ee ee eee eee Fides cam hoe 78 Glacier-formed hanging valleys...... ..........--.05 RE 3 Sela Pee eet & 78 ereemumon ivpe Of hanging valleys... .<-...-- .s.sccess, vececscanssves 81 ETE STS TS NE ee Tv a a ete 81 IR MIRIRS ERT OR NERUA Se ha oe LS a Salsa acne isle eee aie aes WES Cee dl a o's 82 Detailed study of Kieger Creek canyon as a type..............-. 00: ee eee 83 Relation of pre-Glacial erosion to hanging valleys......-.............05 87 Further discussion of origin of glaciated hanging valleys..... .......... 87 Evidence derived from study of Bloody and Lundy canyons............. 88 Conclusions. ..... MLN od id oi eee ie 5 ies Sta crave aie Sighs Guts oy eR RAS 90 DEFINITION OF TERM “ HANGING VALLEY ” The topographic features to which attention is here chiefly invited have been defined by G. K. Gilbert* as follows: “A hanging valley is a small U-shaped tributary to a larger valley, the floor of the smaller being considerably higher at the junction than the floor of the larger. Many of them are short, high-grade troughs, heading in cirques; some are mere cirques, without troughs—spoon-bowl hollows, high on the walls of main valleys. They are associated with other evidences of glacial sculpture, and the elevation of their floors is believed to result, asa rule, from the unequal erosion of valleys by glaciers of unequal size.” As stated by T. C. Chamberlin and R. D. Salisbury,t ‘“‘ when the lower *G. K. Gilbert: ‘Glaciers and glaciation.”’ Harriman Alaska Expedition, vol. iii, New York, 1904, pp. 114-115. + Geology, vol. i, New York, 1904, p. 155. XI—Butt. Geon, Soo, Am., Von. 16, 1904 (75) 76 I. Cc. RUSSELL—HANGING VALLEYS end of a tributary valley is distinctly above the level of its main the former is called a “ hanging valley.” * The first of these definitions restricts the term Teh to the valleys of formerly glaciated regions, and ascribes to them a single mode of origin. As I hope to show in the following pages, a discordance in gradient between a main valley and its tributaries is not strictly con- fined to glaciated regions, and may result from one of several processes. For my present purpose, therefore, it is necessary to give a more com- prehensive meaning to the term hanging valley than is done in the first definition referred to. The second definition is more general than the one framed by Gilbert, and does not imply mode of origin, but can per- haps be amplified in order to make it more graphic. Provisionally and as used in this paper the generic term hanging valley may be understood to include any valley or valley-like depression the bottom of which is not in even adjustment with the bottom of the lower depression with which it unites and into which it drains, the passage from one to the other being by means of a slope of greater declivity than the gradient of the tributary valley, and in most instances precipitous—that is, hanging valleys are in striking contrast to the con- ditions normal to mature steam-eroded valley systems, in which, as was long since pointed out by Playfair, the branches of the system at their mouths are nicely adjusted to the level of the lower valley with which they unite. ‘The definition just presented takes account of present topographic conditions only, without reference to the manner in which ee ance in gradients referred to came abony, In attempting to define “species ” under the hanging valley “ genus” mode of origin may be taken as the leading criteria, and more or less conspicuous differences in hanging valleys produced by the same agency suggest the recognition of “ varieties.” CLASSIFICATION ENUMERATION OF SPECIES Our present knowledge of hanging valleys seems to warrant their pro- *The nature and mode of origin of hanging valleys has been discussed by W. M. Davis in an article entitled ‘‘ Glacial erosion in France, Switzerland, and Norway,’’ in Boston Society of Nat- ural History Proceedings, vol. 29, 1900, pp. 273-321. Tothe bibliography presented in this paper the following titles of subsequent publications treating of hanging valleys may be added : W. O. Crosby: “The hanging valleys of Georgetown, Colorado.’’ Technology Quarterly, vol. xvi, 1903, pp. 41-50. Ralph 8. Tarr: ‘‘ Hanging valleys in the Finger Lake region of central New York.’’ American Geologist, vol. xx xiii, 1904, pp. 271-291. q ee er CLASSIFICATION tl visional classification under four species, namely, stream-formed, ocean- - formed, diastrophic, and glacier-formed. STREAM-FORMED HANGING VALLEYS During the development of a river system a master stream may deepen its valley more rapidly than certain of its tributaries are able to keep pace, as has been explained by W. M. Davis and others, and the tribu- tary valleys become discordant in gradient in reference to the main valley with which they unite. The various ways in which such a result is produced are well understood and need not be reviewed at this time. An example of a hanging valley due to stream erosion was recently pointed out to me by M.S. W. Jefferson, near Ypsilanti, Michigan. In this instance the Huron river, in deepening its channel, left a long narrow point of land projecting from one side of its valley, and on the surface of this cape-like projection and transverse to it is a segment of an old chan- nel made by the Huron when flowing about 40 feet higher than now. This fragment of an abandoned stream channel is now a hanging valley, and at each of its ends there is a precipitous descent to the present flood- plain of the stream that eroded it. On the border of the Huron valley, about 5 miles east of Ypsilanti, as described by Isaiah Bowman,* there is a locality where the Huron river in broadening its valley cuts into the upper portion of one of its own tributaries. In this instance the portion of the tributary valley which was “captured ” and the neighboring portion of the abandoned segment of the same valley respectively, were left ‘‘ hanging ” in reference to the valley of the Huron, and illustrate a process by which a conspicuous discrepancy between the bottom of a main valley and the bottom of a tributary valley may be brought about. Each of the examples just cited may, as it seems, be logically classified as varieties of hanging valleys due to stream erosion. OCEAN-FORMED HANGING VALLEYS When a valley, and particularly one with a steep gradient, is tributary to the ocean, and a landward migration of the ocean shore occurs through the action of waves and currents, the distal end of the valley and its bordering uplands, may be removed, leaving its abbreviated up- stream portion opening in the face of a sea cliff and its draining stream, if one is present, discharging by means of a cataract into the sea. The portion of a valley thus truncated would have the characteristic topo- graphic features of a hanging valley, and might be of stream or glacial “A typical case of stream capture in Michigan.” Jour. Geol., vol. xii, 1904, pp. 326-334. 78 I. C. RUSSELL—HANGING VALLEYS origin or have resulted from the combined action of these two and pos- sibly also of other agencies. | Examples of valleys which have been abbreviated in the manner just referred to and left suspended above the ocean are present on Unalaska island and open to the sea several hundred feet above its level.* Other similar examples might be cited, but it is evident that whenever a sea cliff recedes at a more rapid rate than streams, or glaciers, on the adjacent land can deepen their channels and at their mouths maintain a position at sealevel, hanging valleys must result. DIASTROPHIC HANGING VALLEYS A fault may cut across a valley so as to produce a steep descent in its bed, and thus cause one of its segments to become a longitudinal hang- ing valley in reference to the next lower segment. An example in line with this suggestion has been described by W. O. Crosby 7 near George- - town Colorado. If a region with stream eroded or other valleys becomes broken by faults so as to produce block mountains, or when during periods of rest in the growth of such faults valleys are eroded, it is evident that exam- ples of discordant gradients of the nature of those under consideration may result. Examples of such diastrophic hanging valleys are abundant in the Great basin, which open into broad desert valleys or basins, high above their bottoms, although in many, and perhaps most, instances the steep descents at their mouths are concealed beneath alluvial deposits. It needs but a suggestion to make it clear that the upheavel of a bold coast in a state of mature topographic development or occupied by tidal glaciers might transform the valleys previously deepened approximately to baselevel into hanging valleys opening in the rim of an ocean basin. The three “species” of hanging valleys to which attention has just been directed would probably be considered as “imitative forms ” by persons who hold that the recognition of a glacial origin in the definition of a hanging valley is essential, or consider that hanging valleys are in themselves evidence of former glaciation. If, however, we assume that peculiarity of togographic form is the leading fact expressed by the term hanging valley, it is consistent and legitimate to recognize under that term the discordances in slope produced in the several ways just cited. GLACIER-FORMED HANGING VALLEYS As has been clearly shown by Gannett, Gilbert, Davis, and others, *@. K. Gilbert: “Glaciers and glaciation.” Harriman Alaska Expedition, vol. iii, New York, 1904, p. 185. +‘‘The hanging valleys of Georgetown, Colorado.” Technology Quar., vol. xvi, 1903, pp. 41-50. inet Cal mat icine nina ce ee Se fe GLACIER-FORMED HANGING VALLEYS 79 hanging valleys would result in case a main valley glacier eroded its bed more deeply than a tributary glacier deepens its bed. The differ- ence in the depth of such valleys, after the glaciers which shaped them had melted, would represent the excess of erosion of the receiving glacier over that of its tributary. This simple and one may say self- evident explanation of the origin of the numerous hanging valleys of glaciated mountains is seemingly complete and satisfactory if the large amount of glacial erosion it implies in numerous instances can from other evidence be proven to have occurred. This matter will be reverted to below. In the study of glaciated hanging valleys, however, the question arises: Are there not other processes by which glaciers can pro- duce similar topographic results? To aid search in this direction, at least six suggestions seem pertinent, and are embodied in the following paragraphs: Small glaciers sometimes originate on the sides of mountains or on the borders of deep valleys where, so far as can be ascertained, no pre- vious depressions existed. Examples of such “ mountain-side glaciers,” as they may be termed,* are known on mount Rainier, Washington, and on the Three Sister peaks and mount Jefferson, in Oregon, and elsewhere. It is evident from inspection that typical mountain-side glaciers are engaged in excavating depressions for themselves, which, at an early stage in the process, have the essential features of cirques, and at a later stage develop a valley of the typical U-shaped cross-profile, flat bottom, etcetera, with acirque atits head. These glaciers may be said to burrow into the mountain sides by headward extension, chiefly, as is judged, by “quarrying.” The lower limit to which they are enabled to excavate their beds, or the local baselevel, is determined by melting. A glacier of the type referred to deepens its bed to this level, and, given time enough, extends it headward until a flat-bottom trench is produced. Should the glacier then melt,a hanging valley would be left, the mouth of which would open out on the slope on which the glacier originated. Such valleys are ‘‘hung up” in reference to the country which they face, but differential. glacial erosion has no part in their production. A mountain-side or valley glacier sometimes ends at the summit of a precipice,where it breaks off and descends in avalanches, to be recemented at the base of the escarpment or there melt, according to climatic con- ditions. In such instances it is evident that the glacier before being broken might excavate a cirque or trough according to the length of its duration, and on melting leave a hanging valley. In this case, as in the *The class of glaciers referred to have, as it seems, at least in part, been previously termed “hanging glaciers.” 80 I. C. RUSSELL—HANGING VALLEYS one cited above, the resulting valley is hung up in reference to the de- . pression which it faces, and this depression may be a narrow or wide valley, a continental or an oceanic basin. The changes in topography produced by mountain-side glaciers in the ways just cited may be likened to the excavation of valleys by tidal glaciers, inasmuch as in each case a local baselevel is established above the bottom of a depression into which the glacier in each instance would have advanced if conditions had not intervened which led to the removal of its distal end. . The production of rock escarpments across young glacial valleys by the method of “ quarrying,” as outlined by Willard D. Johnson, so well illustrated in the glaciated canyon of the Sierra Nevada, Cascade moun- tains, and elsewhere, also results in breaking a valley or trough into parts, one of which is left hanging in reference to its next neighbor lower down. It may seem far-fetched to consider the examples just cited as. ~ hanging valleys, but in glaciated valleys like those of the Sierra Nevada, etcetera, the highest segment of a glaciated trough frequently leads to a cirque, and in many instances has essentiaily all the topographic features which pertain to normal hanging valleys on the sides of sim- ilar troughs. The only topographic difference between the two seems to be that the last “tread ” in the giant stairway a valley glacier makes, and the hanging valleys recognized by Gilbert and others, is that the tributary enters the main valley at its head and in a direct continuation of its course, instead of from the side. In case a glacier originates in the upper portion and is extended down the course of a previously stream-eroded valley, filling it, we will assume, to a depth of 1,000 feet for a distance of many miles, local glaciers may originate on the bordering valley slopes, either in previously stream-eroded gorges or on the precipitous valley side after the manner of mountain-side glaciers,and enlarge preexisting ravines or create new lateral troughs. Such lateral glaciers might become tributary to the main valley glacier, in which case they would have a baselevel of erosion determined by the level of the receiving glacier, less the thickness of the tributary. In this imaginary case, which, as it seems, could be paralleled in nature, the side glaciers would excavate valleys, which, if the ice should melt, would become hanging valleys, in reference to the main trough formerly occu- pied by the receiving glacier, but the discordance between the two would not be due entirely and possibly only to a small extent to differential glacial erosion. As has been pointed out by several writers, a large valley glacier which advances down a previously stream-eroded valley tends in an efficient way to straighten and broadenit. By this process lateral valleys are ab- = —— ne GLACIER-FORMED HANGING VALLEYS 81 breviated, and when the glacier melts a discordance in gradient between the main valley and its tributary valleys would become manifest. This result would follow under the process referred to, no matter whether the tributary valleys were occupied for a time by secondary ee or not. Still another manner in which glaciers may give origin to hanging valleys without the aid of differential ice erosion is illustrated about the northern border of the Malaspina glacier, where several alpine glaciers are tributary to a widely expanded ice sheet at the foot of the mountains from which they flow. The surface of the feeding ice streams near their mouths are not at present well adjusted to the surface level of the re- ceiving ice field, but make more or less sharp descents in order to join it. During a previous and higher stage of the glaciers, however, their sur- faces were more nearly in adjustment than at present. The point I wish to make is that a tributary to a piedmont glacier would have its base- level determined by the horizon of the receiving ice field less the thick- ness of the tributary, and, given time enough, would excavate down to that level. Should the glaciers meet after this result had been reached, hanging valleys would appear about the border of the basin or Ae formerly occupied by the piedmont glacier. To summarize: There appears to be at least six sets of conditions or processes each of which may produce glaciated hanging valleys without necessitating a conspicuously great measure of differential ice erosion. To generalize more widely, as I think is justified from the considera- tions presented in the preceding pages, there are at least five sets of con- ditions each of which may produce hanging valleys without the assistance of glaciers and at least six sets of conditions under which glaciated hang- ing valleys may originate. Most coMMoN TYPE oF HANGING VALLEYS CHIEF CHARACTERISTICS As is well known, hanging valleys are a characteristic feature of many formerly glaciated mountains, such as the Sierra Nevada, Cascades, the Alps, the Southern Alps, etcetera. In such mountains conspicuously straight or broadly curved trough-like valleys, U-shaped in cross-profile and several thousand feet deep, are of common occurrence and frequently have tributary glaciated valleys opening into them along their sides at elevations of perhaps 1,000 and in some instances 2,000 or more feet above their bottoms. In the most typical instances the gradients of both the receiving and tributary valleys are low, and the descent from the mouth of a tributary to the bottom of its main valley is precipitous 82 I. C. RUSSELL—HANGING VALLEYS and in alignment with the adjacent portions of the side of the main valley. Very commonly, too, the mouth of the tributary hanging valley is about on a level with the highest lateral moraine, or other evidence of glaciation, on the sides of the receiving valley, but this is not always the case. DISCUSSION AS TO ORIGIN It is chiefly in reference to the mode of origin of hanging valleys of the type just described that differences of opinion have arisen among students of glacial topography. Having in mind the several processes by which glaciers may produce hanging valleys, as enumerated above, it seems evident that those of the type just referred to must have been produced in one of two ways or by a combination of the two methods: Hither they have resulted from the differential ice erosion of the main valley and of its tributary, or a pre- Glacial drainage system has been modified by glaciers so as to have the characteristics referred to, or these two processes have been active at the same locality. If the first of these explanations is accepted, it follows that the rock removed to form the portion of a main valley lying at a lower level than the bottom of a tributary hanging valley is a minimum measure of the amount of ice erosion in the main valley. What the maximum amount of ice erosion in such valley may be is not apparent and, as it seems, has not received serious consideration. Unless pre-Glacial erosion is admitted, however, in the case of such mountains as the Sierra Nevada and Cas- cades, the hypothesis of differential ice erosion to account for hanging valleys implies that far more rock has been removed by ice than is ac- counted for by the differences in elevation between a main valley and its tributary hanging valleys. The hypothesis which essays to account for hanging valleys by the enlargement and modification of pre-Glacial stream valleys rests mainly for its support on the principle that a glacier developed in a main stream-eroded valley would not only tend to straighten and broaden it, and thus truncate the ends of tributary valleys, but establish a baselevel for lateral glaciers, and by determining the depth to which they would be able to erode lead to the origin of hanging valleys at the level of its surface, minus the thickness of the tributaries. Under this hypothesis also some differential ice erosion must be admitted, since other condi- tions being the same a thick glacier must be accredited with greater erosive power than a thinner one; but several qualifications would have to be considered in this connection in attempting a complete analysis. One of the hypotheses before us, then, demands a vast amount of —— ee oe ee, eee DISCUSSION AS TO ORIGIN 83 glacial erosion, while the other implies far less conspicuous results in this connection. In looking for criteria by which to test the two hy- potheses there are evidently two chief directions in which search may reasonably be expected to yield assistance. These are, first, the topo- graphic features due to removal of material in order to produce the valleys of glaciated mountains, and, second, the character, and espe- cially the amount of material still in sight and recognizable, which was removed by glaciers and deposited elsewhere. If the mountains in which hanging valleys are a common feature, owe the valleys—the excavations of which have given them their leading and characteristic features—wholly or toa conspicuously great degree to glacial erosion, or, on the other hand, have been shaped principally by stream erosion and only to a minor degree modified by glaciers, the study of the character, distribution, and interrelation of their valleys furnishes a basis for opinion. I must freely confess that Iam not in a position to thoroughly discuss this broad and far-reaching problem, but venture to direct attention to two localities which seem to furnish data bearing on the question im- mediately before us. Throughout the major portions of the Sierra Nevada and Cascade range there are conspicuous evidences of erosion both by water and ice; but to the direct question: Did pronounced stream erosion precede re work of the glaciers? It is difficult with the criteria at present avail- able to obtain a satisfactory answer. The conditions demanding con- sideration are complex, and the decidedly vigorous ice erosion which has occurred in many portions of the mountains referred to did much to erase the evidence pertaining to the nature of the pre-Glacial topography. Better localities for beginning the inquiry under consideration are, in my judgment, furnished by the mountains of the Great basin which bear records of a brief period of glaciation on their higher portions. For this purpose Stein mountain, in southeastern Oregon, is well suited. DETAILED STUDY OF KIEGER CREEK CANYON AS A TYPE Stein mountain* is a typical block mountain, with a bold eastern face, bordering a fault or a series of faults, and is inclined gently west- ward. Its elevation is about 9,000 feet, and it rises approximately 5,000 feet above the basin atits eastern base. Starting at the crest of the tilted block and leading westward are several deep canyons, one of which, Kieger Creek canyon, contains unmistakable evidence of glaciation in *The facts here presented in reference to Stein mountain are from Bull. no. 217, U. S. Geol, Survey, pp. 16-17. XII—Bvut. Geox. Soc. Am., Vou. 16, 1904 84 I. C. RUSSELL—HANGING VALLEYS the upper 4 or 5 miles of its course. The canyon is comparatively straight, and is simple in its topography. In its upper portion its bold northern border is a great precipitous wall, without lateral valleys or alcoves, but its southern border is conspicuously irregular and has four or five cirque-like hanging valleys arranged along it and about 1,000 feet above its bottom. These hanging valleys are less typical than in many other instances, and at their mouths the main valley widens and the steep descents leading from them to the receiving valley are within recesses, and do not form a part of the main wall of the larger valley. At the present time the snow melts in the main valley, even at its head, each summer, but lingers so as to form perennial banks in the cirque of each of the tributary hanging valleys on its southern wall. The question is: Was Kieger canyon in existence previous to the for- mation of glaciers on Stein mountain, and does the evidence point to deep stream dissection of the mountain before glaciers appeared, or was it excavated entirely by the glacier that occupied it, less the insignificant — amount of post-Glacial stream erosion that has occurred ? The facts most suggestive in this connection are: Kieger canyon is prolonged for some 20 miles below the locality where the lowest evidence of glaciation is discernible, and is not floored with coarse debris, such as occurs downstream from the extremities of existing valley glaciers. No alluvial terraces are present to show that conspicuous variations in the load of Kieger creek have occurred, such as are present in many valleys which have held glaciers in their higher tracts. There are no recogniz- able terminal moraines anywhere in the canyon and almost a complete absence of lateral moraines. The evidence of the former presence of a glacier in the upper end of the canyon is furnished principally by a noticeable increase of width in the portion formerly occupied by ice and a change from nearly vertical walls to an U-shaped cross profile. If we assume that Kieger canyon owes its depth below its hanging valleys to ice erosion, it must be remembered that its chief supply of ice came from the small glaciers on its southern side, as there is no well- defined cirque at its head and no other gathering ground for snow, except the nearly straight and unbroken surface of its precipitous northern wall—that is, the highest lateral glacier was enabled to excavate a val- ley some 3800 or 400 feet deep, but on turning at right angles to its upper course and being reinforced to a slight extent by the snow in Kieger canyon, above where the change in direction occurred, it was at once enabled to erode a canyon to a depth of 1,000 feet. Such an assumption, while called for by the hypothesis of differential ice erosion to account for hanging valleys, does violence to what is known to be the habit, so to speak, of glaciers and is manifestly untenable, : KIEGER CREEK CANYON AS A TYPE 85 The facts presented by Kieger canyon, on the other hand, are con- sistent with the idea that a deep water-cut trench existed before the small glaciers from the south entered it, and that it has experienced what may be termed a small degree of glacial erosion. This conclusion finds support also in the fact that, so far as has been discovered, none of the other canyons on the Stein mountains show evidence of glaciation. Kieger canyon heads in the highest portion of the upturned block in which it has been excavated, but the advantages thus assured in refer- ence to snow accumulation are not conspicuous. Seemingly the condi- tions were so delicately adjusted that a glacier was formed at the head of Kieger canyon, but not in neighboring canyons. This condition of delicate adjustment of elevation and topographic conditions to the cli- mate of the region is now manifest by the presence of perennial snow banks in the shelter of northward-facing cliffs on the border of Kieger canyon and their absence on the borders of the neighboring canyons. Considering all the evidence and endeavoring to make a just allowance for personal equation, it seems justifiable to conclude that Stein moun- tain was deeply dissected by streams previous to the Glacial epoch, and that during that epoch snow accumulated on its summit portion and gave origin to glaciers on the south side of Kieger canyon, near its source. These small glaciers descended into the canyon and formed a trunk glacier 4 or 5 miles long. With the amelioration of climate as the Glacial epoch drew to a close the ice melted out of the canyon, but still lingered in the form of small glaciers in the cirques under the shel- ter of its south wall and at the localities where perennial snow is still present. The small side glaciers high on the wall of the canyon exca- vated alcoves for the reason that in the earlier portion of their existence their ice during a certain stage in its advance gained the precipitous slope of the main canyon, there broke off, and descended as avalanches, thus having a lower limit of erosion analogous to a local baselevel estab- lished, and later, when Kieger canyon was occupied by ice, the side gla- ciers became adjusted to its surface level and had the downward limit to which they could excavate determined by this cause. Again, when the ice melted in Kieger canyon, the side glaciers continued their work in the same manner as during their earlier stage. The resulting topo- graphic change was mainly the production of high lateral alcoves or cirques on the side of the main canyon, which constitute, but are not typical examples of, hanging valleys. In the case of Kieger canyon, as I have attempted to describe, lateral glaciers entered a main valley which essentially had no cirque or other gathering ground for snow of its own, and that the side glaciers can not consistently be held accountable for any considerable share of the work 86 I. GC. RUSSELL—HANGING VALLEYS required to excavate it. This case does not stand alone, and there are at least some reasons for thinking it may prove to be typical rather than exceptional in the topographic history of glaciated mountains. Another example of topographic conditions similar to those just cited is furnished by the canyon of Rush creek, near Mono lake, California. In the case of Rush Creek canyon, as is well shown on an admirable map of the basin of Mono lake made by Willard D. Johnson,* a deep horse- shoe-shaped valley is present at the eastern base of the Sierra Nevada, which has four or five hanging valleys arranged along its precipitous outer margin. Two of the tributary valleys were formerly occupied by valley glaciers, and where they enter the main valley, at right angles to its course, there is a steep descent of about 1,600 feet, and two others of the tributary hanging valleys are little more than cirques, and enter the main valley at about the same horizon as the larger tributaries and also at about the level of the highest lateral moraines in the lower valley. The main valley had essentially no source of snow supply except the ~ contributions from its own hanging valleys, and it is inconsistent with the known behavior of glaciers to assume that the tributary glaciers from the lateral valleys abruptly changed their direction of flow and at the same time and at localities in definite alignment one with another deepened their troughs at least 1,600 feet in excess of their upstream portions. On the contrary, the facts just referred to are in harmony with the view that the horseshoe-shaped portion of Rush Creek canyon, and at least the larger of its tributary valleys, were in existence previous to the development of the glaciers of that region and gave direction to the glaciers which occupied them. In this instance, as in the case of Kieger canyon, pre-Glacial topographic conditions seem clearly to have given direction to the flow of the ice which occupied the canyon and to have been modified by ice abrasion to only a secondary degree. Returning to the consideration of Stein mountain, it is situated about 180 miles east of the Cascade range, is of about the same height as that — range, is composed of similar rocks, and, so far as can now be judged, was upraised at about the same time that volcanic eruptions built the principal part of the larger mountains. It may therefore reasonably be assumed to have experienced the same climatic changes as the Cascade range, but in a less intense degree. Its history may, as it seems, be taken as a simplified example of the more voluminous and more com- plex records of events inscribed on the greater mountain with which it is somewhat remotely associated. I venture to assert, in part from per- * Israel C. Russell: “Quaternary history of Mono valley, California.” Eighth Annual Report, part i, U. S. Geol. Survey, plate xvii. A description of Rush Creek canyon, etcetera, is contained in the same report. E FURTHER DISCUSSION AS TO ORIGIN 87 sonal observation, that certain other isolated mountains in the Great basin, as, for example, mount Newberry,* Oregon, and Jeff Davis peak, Utah, are similar to Stein mountain so far as their records of climatic changes are involved, and, as is well known, are similarly situated in reference to the Cascade range and the Sierra Nevada. RELATION OF PRE-GLACIAL EROSION TO HANGING VALLEYS It is exceedingly difficult to formulate and duly evaluate the evidence just considered, but when it is correlated with other evidence, such, for exampie, as the facts presented by Le Conte and others, showing the occurrence of a Sierrian epoch of stream erosion preceding the Glacial epoch, the only consistent conclusion seems to be that the Sierra Nevada and Cascade ranges were deeply dissected by streams previous to their occupancy by glaciers. In reference to hanging valleys, this conclusion favors the idea that they have been excavated on the sides of pre-Glacial canyons, chiefly because of the establishment of a baselevel of ice erosion in such canyons by large valley glaciers, which determined the lower limit to which the secondary glaciers on their sides could excavate, and does not favor the idea that differential ice erosion should be held accountable for the entire amount of discrepancy in depth between a receiving and its lateral hanging valleys. FURTHER DISCUSSION OF ORIGIN OF GLACIATED HANGING VALLEYS As stated on a previous page, two classes of evidence may reasonably be appealed to in seeking an explanation of the mode of origin of glaci- ated hanging valleys. One of them, namely, the topographic changes produced by excavation, has been briefly considered. The other, namely, the character and distribution of the material removed in order to make the excavation referred to and redeposited in recognizable form, remains to be reviewed. If the hanging valleys of such mountains as the Sierra Nevada and Cascade ranges are due to differential ice erosion, as claimed by several geologists and geographers, it implies a vast amount of ice work. For example, a conservative estimate of the volume of rock necessary to fill the valley of lake Chelan up to the level of the hanging valleys on its borders is in the neighborhood of 50 cubic miles. To perform a similar task in several other valleys of the Cascade range and Sierra Nevada would require the removal of similarly great quantities of material. Sev- eral of these trunk valleys with hanging valleys along their sides are 40 * Mount Newberry is described and named in U. S. Ge ological Survey Bull. no. 252. 88 : I. C. RUSSELL—-HANGING VALLEYS to 50 or more miles long and their tributary hanging valleys from 1,000 to 2,000 feet above their bottoms. If glaciers excavated even the portion of these valleys which are situated at a lower level than the floors of their tributary hanging valleys, not taking into account the portions of the valleys above such horizons, it is legitimate and logical to demand that evidence be produced as to the disposition the glaciers made of this material. | | In traversing such canyons as those occupied by Methow and Wenache rivers and the valley of lake Chelan, in Washington, or the Tuolumne and other similar canyons in California, one can not fail to be impressed with the fact that the amount of morainal material associated with them is insignificant in reference to the size of the excavations from which it was derived. In several typical instances the morainal material in sight or which may reasonably be assumed to be present, if spread over the glaciated surfaces of the canyons from which it was obtained, would, it is safe to say, form a layer only a few feet, and in most instances only a - few inches, thick. It can, perhaps, be claimed that the glaciers ground their grists so fine that the streams supplied by their melting carried it away and distrib- uted it so widely that it is no longer recognizable; but glaciers carry fine and coarse material alike, and the coarse material should, in the instances under consideration, still remain where the glaciers left it. While there is no known ratio between the coarse and fine debris of glaciers, we know that in many instances conspicuous deposits of till, boulders, etcetera, are laid down by them. For example, in the case of the continental glaciers that formerly occupied the northern half of North America the moraine and till sheets are far more conspicuous than the records of abrasion left by the same ice sheets. So, too, in the case of many alpine glaciers conspicuous lateral and terminal moraines remain as evidence of the work done. Such general considerations, as most students of glaciers will admit, no doubt are sufficient in them- selves to sustain the demand for ocular evidence as to what has become of the rock removed in case hanging valleys are due to differential ice erosion; but, as I think, a more specific argument in this connection can be presented. EVIDENCE DERIVED FROM STUDY OF BLOODY AND LUNDY CANYONS On the southwest border of the basin of Mono lake * there are several deep glaciated canyons, some of which haye typical hanging valleys on their sides. Two of these canyons, namely, Bloody canyon and *Tsrael C. Russell: 5‘ Quaternary history of Mono valley, California.’? Eighth Annual Report U. 8. Geol. Survey, part i, 1889, pp. 331-333, 337-340. EVIDENCE FROM BLOODY AND LUNDY CANYONS 89 Lundy canyon, present’ certain similarities and contrasts which are in- structive in the above connection. These two canyons are only about 5 miles apart, are situated on the same side of a mountain range, and hence must have been subjected to the same climatic changes. They have been excavated in rocks of essentially the same degree of resistance to mechanical erosion and were in each instance occupied by a glacier which extended beyond the lower limit of its canyon and entered Mono valley—the Lundy Canyon glacier reaching out half a mile and Bloody Canyon glacier between 3 and 4 miles into the main valley. Bloody canyon has a steep gradient, is about 2 miles long, approximately 2,000 to 2,500 feet deep, and a mile wide at an elevation of 2,000 feet above its bottom. Lundy canyon is 6 to 7 miles long, has a gentle gradient in its lower half, is steep in its upper portion, and about 3,000 feet deep, with a width of approximately 14 miles at an elevation of 2,000 feet above its bottom. The amount of material removed in order to produce Lundy canyon was certainly five times as great as the similar task in the case of Bloody canyon. If each canyon was excavated by a glacier, it is entirely consistent to conclude that the morainal material deposited by Lundy Canyon glacier should be at least five times as great as the similar material laid down by Bloody Canyon glacier. On the contrary, however, as shown by field observation and an inspection of carefully prepared maps, the recognizable morainal material in Lundy canyon and about its mouth is in volume only a small per cent of the volume of the similar material in and in front of Bloody canyon. The morainal em- bankments at the mouth of Lundy canyon are simple ridges about half a mile long and between 200 and 300 feet high. The similar embank- ments at the mouth of Bloody canyon are over 3 miles long, highly compound, and for a distance of over a mile near their source of supply are between 500 and 600 feet higher than the bottom of the trough between them, which itself has a thick accumulation of morainal and stream-deposited material beneath it. It is thus made clear that the amount of work the Bloody Canyon glacier performed was vastly greater than the work done by Lundy Canyon glacier. Why this discrepany in the case of two neighboring glaciers of approximately the same size and working on essentially the same average character of rock? The only explanation which seems to fit the case is that Lundy canyon had approximately its present size and shape previous to the origin of the glacier which occupied it, and that the glacier found the canyon, especially in its lower half, so well adapted for its use that but slight alterations were made in its contours, while Bloody Canyon glacier originated where only a comparatively small amount of previous excavation had been done, and it eroded its bed vigorously, 90 I. C. RUSSELL—-HANGING VALLEYS On the south side of Lundy canyon there is a.high lateral tributary valley, known as Lake canyon, which is a typical glaciated hanging valley. The floor of Lake canyon is about 1,000 feet above the bottom of the receiving canyon and approximately on a level with the highest evidence of glaciation on its sides. Bloody canyon has recesses in the higher portions of its walls, but none of them can be considered as even moderately well shaped hanging valleys. So far as these two canyons are concerned the one which has undergone the lesser amount of glacial alteration is the one which has the best defined hanging valley opening into it. Throughout the Sierra Nevada and the Cascade range it is the long, low-grade troughs similar to the lower half of Lundy canyon which have the best defined and most typical hanging valleys opening into them. It is in connection with these canyons, also, that there is a con- spicuous absence of evidence, such as is furnished by moraines, of glacial erosion. CoNCLUSIONS If I have read the story correctly, it would seem that both the destruc- tional and constructional topographic forms due to glaciers in the Pacific Cordilleras of the United States seem to favor two conclusions: First, that the mountains were deeply stream-sculptured before the Glacial epoch, and, second, that certain of the glaciated hanging valleys are not due, either wholly or in a large measure, to differential ice erosion. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 91-130, PLS. 24-32 MARCH 10, 1905 PLUMOSE DIABASE AND PALAGONITE FROM THE HOLYOKE TRAP SHEET* BY B. K. EMERSON (Presented before the Society December 30, 1904) CONTENTS ; Page Extent, thickness, and general structure of the trap sheet................... 92 Types of foreign inclusions m the diabase.:.... 2.1.2 cece cece cece ce seee 92 Meriden type: Blending of mud and lava at the base of the bed........ 92 Titans Pier type: Blending of mud and sand with the lava at the surface CM Te et Otrbe sb iet 1d.) seh seu Seles diel pls Sel ae Age suldwds eld 93 Holyoke Reservoir type: Blending of mud and lava in the ee portion MR Org Sart el na eenckhe SA y Sin 8 piled wplndiny aes wae alee 95 SEE Ds sis FM IRR RENEG Se OgSm Re e) eee e 95 PER SOE EMELINE cis a ta eee sy nue a'e ced og oe Mae ett 95 MERE Goo tr. hous. Soc he ea fencers cewe een es Piet ee Résumé of the structure of the drier Wiae AIGK 2 ede Li eh Lass 96 Abstract of theory of formation of plumose forms and of palagonite. 96 IETNNTRIREEIE oie 3 iy Saas oid hb es acto nldes sues Ts A SN chen inc iat 97 IN UIRTRR RRR ONC cc 2 9 So orp hn 5S ik ws spcreta 5 aS a Sxl Wiaia Or box! armies 97 Secemeranve Of analcite.:........ ... cee e ese DR ates Si ido x meee hahaa cdi area 98 SEE MeMErIDIIOn OF, FIG SCHHETEN. oo owe dn nn vee 0 cee nie srie neces 98 CerMe HEE PeMeLen TOCKH, 5.5500 ose seen css ces es ccecdepeceent 99 MENA EUERAB St 5 Si) a acaid od w Poin wipie Ke vo djs wd ywietelly Sad dws 2 99 Short plumose variety with remelted pyroxenes . ................. 100 Gise-hesring@ porpliyritic Giahase.....)...0.. cesses cece deter ewcce cs 101 eee ERODED Gt el cn a A Tica as ps, «2 cm wine & Re Ra he cotta as 102 ape ONE EPET RCT WEAREMONTIEUO oe Cin cok we nig ae niece vce u ely some coe» 103 Botryoidal glass with radiate fibrous devitrified ee Dy Lecce Le . 104 Spherulites, spheerocrystals, and lithophyse in the glass..... ........... 105 aS eemOrees GF HOLYORONE WAG... 5463 Joe ged sie balsa vie valde save dacaes 106 Wie holyokeite or, diabase-aplite dikes... .......2i000) sesslccetessuinas dene 107 2 OE eae eee eee Se Sc eee oe 108 Quaternary veins of quartz, of calcite, and of tuff..................0.05- 111 *This paper is published by permission of the Director of the United States Geological Survey. A friend called my attention to the interesting exposures made in clearing the surface of the trap for a new reservoir, and I asked my assistant, Mr Walter L. Allen, to examine the place. He discovered the interesting plumose trap and the glass described below and collected much of the material for the investigation, and has taken part of the photographs used in the paper. I owe the drawings to the skill of my daughter, Mrs Charlotte E. Hitchcock. XI[I—Butt. Geou. Soc. Am., Vou. 16, 1904 (91) 92 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE Page Chemical analyse oy... 26 cee oe ne oe bie een ee ae 112 Composition of the glass and inclosing lava...:.2)....-: - se: eee 115 Theoretical explanation of the formation of the holyokeite and palagonite and their inclusions <2. 0.0.0 ssc ea cone cee s cete nse ones aan oe 117 The general ‘process... 222.005.) ates «aim oe lee Ua ea thee 117 Formation of the holyokeite dikes.--- <)--)--) oe aee jos ice ee ea Formation of the glass with calcite spherulites.................24 eeeees 118 Formation of lithophysz with spherocrystals of ankerite and quartz ... 119 The palagonite of Seljadalr:: .. 22.0 ..00 se. Sede k eee oe ee eee 121 GSW MIG fie. Seal eee eco fe ble a's aie eialeteleits = ieatm aye en 124 Discussion by Alfred -R. Lane... 0.2.2 eee ee eee ren: 125 Explanation of plates... oi. 02. os 2 ows. oe ecm ee be ae ee 127 EXTENT, THICKNESS, AND GENERAL STRUCTURE OF THE TRAP SHEET The Holyoke trap sheet begins south of Amherst and extends 80 miles — across Massachusetts and Connecticut to Long Islandsound. Itis about 300 feet thick in the latitude of Holyoke and has a low dip eastward, resting conformably in the Triassic sandstones (see plate 24), and plainly was a submarine flow.* It and its neighbors are composed for the most part of a monotonous diabase ; but peculiar structures have been locally produced by the introduction in various ways of the mud, sand, and water of the sea bottom into the interior of the liquid and moving mass. TYPES OF FOREIGN INCLUSIONS IN THE DIABASE MERIDEN TYPE: BLENDING OF MUD AND LAVA AT THE BASE OF THE BED I have elsewhere described considerable areas where the trap, pro- tected below by a thin solid crust, has flowed over and rested on the muddy bottom, and then, the crust becoming ruptured and the liquid lava coming in contact with the mud below, many steam explosions have blown fragments of the under crust, with much mud, water, and filaments and tears of the lava, up into the still liquid mass. In one case a pipe- hole was blown clear through the sheet and a mud volcano formed on its surface. By the blending of the cold mud and the liquid lava much basic glass was formed, which differs much from the glass especially de- scribed in the present paper (see page 103). By theabove process minerals like eegerine-augite and structures simulating those of the crystalline schists have been produced. These basal explosions are not of limited *See description in geology of Old Hampshire county. Monograph xxix, U.S. Geol. Survey, 1898, p. 446, TYPES OF FOREIGN INCLUSIONS ~ 93 extent. I have described only the fine exposures at Greenfield and Meriden in a paper published in this journal.* TITANS PIER TYPE: BLENDING OF MUD AND SAND WITH THE LAVA AT THE SURFACE OF THE BED I have also described with some detail another extensive area within which the great Holyoke trap sheet, here about 400 feet thick, is over large patches both at its upper and its under surface contaminated by foreign sedimentary material, which was for the most part marly and is now consolidated into a dove-colored limestone, more rarely into a shale or sandstone. For 10-miles or more along the sheet from Titans pier on the Connecticut to the Holyoke-Westfield railroad (see plate 24) nearly every lower contact shows this contamination for 10 or 12 feet up, and it is present over large portions of the upper surface; almost every- where indeed where it seemed probable that the exact upper surface was exposed. The results are very different from the case described above. Every- thing is in accord with the hypothesis that the great highly heated trap sheet flowing beneath deep water caused ascending and consequent con- vection currents, by which the adjacent fine-grained marly or muddy sediments were quickly swept over the congealed surface of the trap, and by the disturbance of this thin crust (which disturbances sometimes went so far as to break it up into blocks which careened and sank into the liquid mass) much of the fine mud was quickly mingled with the liquid lava under’such pressure that the two were stirred together like two immiscible fluids. This sometimes produced a perfect emulsion, great swarms of drops of the marly mud being carried down into the lava and appearing now as dove-colored spheres of limestone. Sometimes each sphere is white for asmall segment at its upper surface which represents the cavity formed by the settling of the mud and which is now filled by the later infiltration of calcite. In these cases the cooling has been so rapid that no augite has formed in the adjacent trap, but all the iron has solidified as magnetite in a fine powder which is greatly concentrated in a band ¢ millimeter wide adjacent to the fragments and cavities, thus showing a distinct differentiation. Much more frequently the limestone and the trap are so intricately blended that both must have been en- tirely fluid at the same time and under such pressure that the moisture could not expand and make the mass scoriaceous. Atthe other extreme the trap, already solid and porous, has been shattered by small explosions and its angular fragments have been cemented Dy the mud. Of course all intermediate stages can be observed. * Diabase pitchstone and mud enclosures of the Triassic trap of New England. Bull. Geol. Soc, Am., vol. 7, 1898, pp. 59-86, plates 3-9. Geology of Old Hampshire county, 1898, pp. 419-439. 94 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE This deposit on the upper surface is repeated on the under surface. That which was deposited on the upper surface was carried forward and downward at the front of the sheet by under-rolling and found its place at last (inverted) beneath the sheet in contact with the subjacent beds. If one compares the base of the trap at the water’s edge north of Titans pier, or along the roadway blasted up onto mount Nonotuck above Mount Tom station, with the surface of the trap just south at the great section at Dibbles Crossing on the Westfield-Holyoke railroad, on the south line of Westfield, the identity of the two cases is clear. There may be seen the same blending of the trap with the dove-colored lime- stone which decreases upward. The mass is very scoriaceous at base, and this decreases upward, with the bubbles inverted. One marked structure appears both in the surface and the basal deposit which fur- nishes conclusive proof of this inversion by under-rolling, namely, the long tubular steamholes. So much water was carried into the trap that the latter was not only made very scoriaceous, but the steam escaped up- ward through the trap forming long parallel tubes and keeping them open until the trap was congealed (see plate 25, figure 1, a). The same long empty cavities appear inverted at the base of the trap as at Titans piazza running up six inches from the base (see plate 25, figure 1, 6). The sandstone beneath is soft and unbaked, so that the trap was solid when it reached this position, and it does not seem possible, even if the trap had been liquid, for steam generated in the sand on which it came to rest to bore a great number of parallel holes up into the trap while the layers of the soft sand directly beneath were quite undisturbed. This brings out strongly another important point in which this case is wholly unlike the former one—that is, the Meriden type—namely, the limestone which crowds the lower 12 to 15 feet of the bed can not have been driven up into the trap by explosions from below, because the rock below is a coarse arkose wholly unlike the material in the trap, while in the former case the material carried high up into the trap is of the same kind as that below and is continuous with it, and the bottom scoriaceous layer of the trap is shattered and its fragments are carried with the mud high up into the compact trap. In this second case the lower scoriaceous layer is continuous and grades regularly upward into the compact trap. The reason for assuming that the fine material was carried out over the trap by rapid convection currents of water is that so large an amount of this material was spread on the sheet during the short time of the outflowing and underrolling of the lava and before it came to rest, while the material deposited on it after it came to rest was much coarser. It is perhaps possible that the fine mud was spread over the trap by the ordinary process of sedimentation, because of an exceedingly slow ad- " ee TYPES OF FOREIGN INCLUSIONS 95 vance of the mass, but this would seem to involve a sufficient thickness of the crust to prevent the blending of this mud with the still lquid trap beneath. Unlike the first case, this process has in the cases previously described produced no glass and no metamorphic effect. We may call this the Titans Piertype. The mud has been mingled with the lava superficially in great quantity at the lowest possible temperature and pressure, and the heat has acted for the shortest time. This explains the difference in the results between this and the Meriden type. HOLYOKE RESERVOIR TYPE: BLENDING OF MUD AND LAVA IN THE CENTRAL PORTION OF THE SHEET General statement.—In the case now to be described the mud which in the immediate vicinity is spread over the surface of the trap seems to have been also drawn down more than a hundred feet into the interior of the mass, and thus, while subjected to increased pressure, time, and heat, to have produced physical and chemical changes as intense but very different from those of the Meriden type. Description of the area——The map (see plate 24 and section) shows a portion of the main trap sheet 53 miles long. At the south end, in the extreme southwest corner of Holyoke, at Dibbles Crossing, is shown the superficial area of the trap, which is filled with inclosures of drab lime- stone and shale, after the Titans Pier type. South of the Dibble house the railroad cut shows the structure perfectly. In the ridge just south- east and also in the brook cutting beneath the next house south * the sandstones can be seen resting on the diabase-limestone emulsion and the exposures are entirely satisfactory. A half mile northwest, out on the back of the deeply eroded trap ridge, is the area in question of the Holyoke reservoir type. It is about 1 mile long and 40 rods wide in its southern part. Topographically itis a quite deep depression in the trap, more so than the contour lines indicate, and was naturally chosen for a reservoir. North of the road the depression is continued across swampy and covered ground a mile farther north, where it becomes a deep notch in the trap in which the glass-bearing rocks of the reservoir again occur and have been blasted into for the electric road. The section shows that the area is about 150 feet below the upper surface of the trap sheet, though the possibility of faulting is not excluded. The normal diabase.—The trap in all this portion of the great sheet is the typical dark and fine grained Holyoke diabase. It is exceptionally fresh and compact, but is everywhere cut within the given area by schlieren or segregations of peculiar coarse-grained diabase types. *See figure 1 in ‘‘ Diabase pitchstone,’’ etc. Loc. cit., p. 62. 96 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE Résumé of the structure of the abnormal area.—Down the middle of the abnormal area runs a narrow band which contains rather sparingly frag- ments and filaments of a light pearl-gray clayey limestone exactly like much that is found in the upper surface of the sheet. The introduction of this mud caused much of the magma to crystallize in a coarse gabbroid diabase, greatly shattered the mass and seamed it with many quartz- ankerite veins. The fine trap on either side of this band is full of “schlieren ” of the strange, coarse, gabbro-like varieties, which will be de- scribed in detail. In them the pyroxene shoots out into branching and curving feathery plumes several inches long, with central suture and transverse parting, which heightens the resemblance to a feather ; or the feldspars occur in broad plates nearly an inch square, poikilitic with abundant pyroxene and palagonite grains in beautiful plumose arrange- ment; or the whole is crowded full of clots of black palagonite, which include spherulites of calcite and radiate blue quartz, and these glass clots are surrounded by broad acid,almost purely albitic,areas. Finally,and this seems of first importance for the explanation of the strange devel- opment, the coarse schlieren, which taper to nothing in the fine trap, have often a narrow central band of whitish aphanitic and very acid diabase, containing little or no pyroxene, of the same type as that sur- rounding the glass. This type may be associated with the white diabase which I have elsewhere described as ‘‘ holyokeite.’* Abstract of theory of formation of plumose forms and of palagonite—We can perhaps describe the multitude of exceptional and contradictory forms and phenomena crowded together here more lucidly and not less objectively if we arrange them in accord with the theory which seems best to bind them together and which we may state baldly as follows: (1) Along a narrow band parallel with the front of the sheet a great volume of the mud, like that found on the surface, was drawn down a hundred feet or more into the interior or the molten trap sheet by irreg- ular vortex motions of the flowing mass. (2) Under the high temperature and pressure the quartz and carbon- ates of the mud were abundantly dissolved. Strong internal motions carried sheets and filaments of the magma loaded with these solutions out into the adjacent area as great schlieren. A central band contains the remnants of mud which were not dissolved and carried out into the magma by convection or diffusion, and here explosions of the muddy water have shattered the newly solidified rock, and the calcite and quartz solutions have cemented it into a central breccia. The normal trap on either side is full of coarse schlieren, which represent the streaks * Holyokeite, a purely feldspathic diabase from the Trias of Massachusetts. Jour, Geol., vol. x, 1902, p. 508. TYPES OF FOREIGN INCLUSIONS 97 of the magma loaded with the dissolved material drawn out intoit. The superheated water and dissolved salts acted as ‘‘ mineralizers ” in a gen- eral way to produce the glass with spherulites, the plumose pyroxene, the coarse pegmatitic or poikilitic plagioclase, the large skeletonized magnetites, the micropegmatitic albite, and all the indications of rapid and abnormal crystallization and resorption. A remarkable differentia- tion has taken place in interstitial remnants of the magma into (1) a dark, much hydrated, glass (palagonite), in which beautiful calcite spher- ulites and spherocrystals of ankerite and blue quartz have formed, and (2) an acid quartz-albite ground, which surrounds the palagonite, and has also been extravasated in small dikes, which I have called holyokeite. DETAILED DESCRIPTIONS CENTRAL EXPLOSION BRECCIA The trap sheet is seen on the west to rest ona coarse buff sandstone or conglomerate, and there is no indication that any material has been carried up from below. It is thus not probable that this is a case com- ing under the Meriden type. On the contrary, at the nearest outcrops of the surface of the sheet areas of many acres are full of an indurated mud of exactly the same texture and color as that found here in the center of the sheet. It is therefore probable that the reservoir type isa downward extension of the Titans Pier type. Furthermore, the frag- ments of sandstone found in the breccia are sometimes perfectly stratified, showing that they had been sufficiently indurated at the surface to be transported into the mass without being wholly broken up, and this is also the case with the mud inclosures at the surface. Further, this in- cluded sandstone contains many scales of graphite, which is a widely disseniinated constituent in the Triassic sandstones, making it certain that they have a common derivation. Moreover, the abundant fractur- ing and the cementation are confined to the central band, where the fissuring is an original structure, and not one due to the upturning of the rocks, since large blocks of the trap were broken open in the blasting and isolated fissures found within them. The paragenesis of the pearl-gray vein fillings of these fissures is ex- ceedingly interesting for our purpose. On either wall of every fissure is a layer of a ferruginous carbonate, probably an ankerite.* Within these rusty carbonate layers are quartz layers, generally granular, but sometimes in perfect limpid crystals, one-third of an inch across, which *This is, under the microscope, often made up of perfect rhombohedra with convex faces, which are limonite-dusted in concentric bands. In basal and vertical sections of each rhom- bohedron there appears a perfect large black cross. The curved-face form is in effect a com- posite with many radiating axes. This produces the same result as a radiate-fibrous structure, 98 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE are at times amethystine. A third layer is a calcite in scalenohedra, R’, sometimes terminated by —? R, or in cleavable masses coating the other crystals or filling the cavity, with here and there a rich oil-green lustrous grain of apatitecompleting the filling. These veins are thought to have been formed immediately after solidification. It is interesting that there is no trace of datolite or zeolites in these veins, as both are abundant a few miles south and generally in the trap. They seem to occur only where large faults go through the trap dikes. In the north part of the reservoir area larger fragments of a gray flat-bedded calcareous sand- stone occur, included in the shattered trap, with many mica scales on the lamine and many scales of graphite imbedded. These sandstone fragments are quite large, and, in contrast to the small and rare fragments of the marly sandstone in the breccia farther south, they contain, indi- cating the violence of the explosions, small fragments about an inch across of the same coarse weathered gabbroid diabase. These occur just south of the sharp bend in the road between Hitchcock pond and Rock valley. The rock of the breccia varies in coarseness rapidly. Thestout _ plagioclase blades are often 8 to 10 millimeters long and are opaque white because they are uniformly changed to mats of coarse tufts of a white mica like the plumose muscovite of granite. ThisI take to bean original structure, produced by the continued activity of the heated waters and not a late product of subaerial influence, since it is uniform in the center of great blocks newly blasted out. It is metamorphism and not weathering. The ferromagnesian constituent is often changed to a deep green chlorite. The immediate influence of the abundant water is thus very strong, and is different from what it is in the schlieren rocks. All the minute mud fragments are surrounded by a broad band of crystalline quartz and calcite, showing that they have all shrunk and given up much water to the trap. OCCURRENCE OF ANALCITE In the small steam holes in a block of altered gabbro, near the center containing holyokeite dikes, the cavities are filled with a colorless non- polarizing mineral with cubical cleavage, whose index of refraction is below the balsam. This is doubtless analcite. GENERAL DESCRIPTION OF THE SCHLIEREN The schlieren appear in irregular bands or sheets from a fraction of an inch to several feet in the smallest dimension and tapering off to a thin edge in the midst of the fine-grained trap. Where these bands come in contact with the fine-grained trap the transition was not marked by a fissure as if one while hot had been injected in the other after its GENERAL DESCRIPTION OF THE SCHLIEREN 99 solidification, yet this transition was often very abrupt and was always made within aninch. The coarse variety generally seemed like “ schlie- ren” in the finer, but also seemed at other places to have been fluid for a slightly longer time, since small dike-like apophyses two or three inches wide branched off from the coarser and penetrated the finer rock for several feet, and fragments of the finer rock were sometimes slightly separated from the rest and floated off into the coarser. The long blades of the augite often rest by one end on the surface of the finer trap and project into the coarser for an inch or more. It was as if some foreign fluid had been introduced into the liquid lava when near solidification, and had been stirred into it in irregular streaks by the internal motion of the mass, which fluid made that part of the trap a little more easily fusible, so that it remained liquid for a slightly longer time than the rest and crystallized into coarser and peculiar types. There was thus a transition from the schlieren to the dike condition. It was repeatedly observed that several of these schlieren were placed one above another and all about parallel with the upper surface of the trap sheet, as if their shape was controlled by the advancing motion of the main trap sheet. VARIETIES OF THE SCHLIEREN ROCKS Long plumose diabase.—One of the most remarkable of the schlieren rocks, which I have called the long-plumose diabase, is found only in the immediate vicinity of the breccia band, and contains filaments of the brightly rusting ankerite derived therefrom. It is a coarse grained, jet black, fresh looking rock, in which the feather-like pyroxenes have shot out in flat, thin blades 3 or 4 inches long and nearly a fourth of an inch wide (see plate 26 and plate 27, figure 1), which radiate in plumes like a radiated actinolite. They branch at small angles and are bent grace- fully or sharply twisted, as if they had shot out rapidly into the liquid glass and had been swayed in its currents like a tuft of grass leaves in the wind. A twinning plane runs down the center of each blade, and close set basal partings run at right angles tothesame. These have the effect of the midrib and pinnule of a feather. The resemblance to grass is greatly heightened because the rock has been fissured across this band, and many of the pyroxenes have from weathering turned a bright green, or even straw color and white like dry grass. This is a change to talc. This variety appears in perfection only in a narrow, irregular band about 10 inches wide, traceable several feet in the ledge near the band of sand- stone inclusions. This growth is essentially spherulitic, although the sheaves form only a small portion of a sphere. There is a resemblance to the curved and branching feldspar blades in XIV—Butt. Geo. Soc. Am., Vou. 16, 1904 100 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE the spherulites of Obsidian cliff. Professor Iddings cites Lehmann’s explanation of unequal surface tension for the curved forms. The pyroxene is an almost colorless sahlite, which is slightly black- ened by refusion at surface and along certain cleavage planes, and this causes the biack color. The basal parting is very marked, and this causes the feathery appearance. Yhe central section is caused by twin- ning according to the usual law on (100), and the crystal is uniformly flattened on two of the prism faces (110), so that the twinning plane passes obliquely through the thin plate, causing the broad central suture, which completes the resemblance to a feather. The extinction is there- fore about 23 degrees obliquely to right and left, and an optical axis appears in the border of the field. The photograph (see plate 26) shows imperfectly the length and curved character of the blades, which are frequently notched along their sides by faces of the unit form. The figure (plate 27, figure 1) shows much better the feather-like texture. _ The associated feldspar is labradorite (ab* an‘), with extinction 30 de- grees on (100). This variety, because of its proximity to the breccia band, has, like it, its feldspars sometimes almost completely changed to a radiate tufted mica, probably paragonite, whose fibers are arranged in its cleavages. They are optically positive. Its pyroxenes are loaded with black gran- ules along the parting planes. | There is no glass in the brecciated band, and here glass particles are rare and completely altered. They are often minute, hollow, and beauti- fully botryoidal lithophyse, and the cavities are lined with a coarse radi- ating, brightly polarizing devitrification product, which is sometimes coated by a layer of secondary calcite. Short plumose variety with remelted pyroxenes.—A nother striking variant, which may be called the short plumose form, is very fresh in appearance, jet black, and coarse grained, and the abundant pyroxenes are in short blades, twinned like those described above, but only about an inch long. The peculiarity of this type of the trap is that all the large pyroxenes are more or less remelted while retaining, wholly or partly, their original positions and traces of original structure. This is shown in plate 27, figure 2. The common pale brown pyroxene (see A in the right central portion of the figure) has the strong prismatic cleavage (horizontal in the figure) and a trace of the nearly vertical basal parting. Below this a broad dark band g runs down athwart the figure; flanked on either side by a lighter brown band dg. These are glass from the melting of the pyrox- ene. In the zone of transition between the glass and the pyroxene the basal parting is developed into an almost micaceous cleavage, and this ee SHORT PLUMOSE VARIETY 101 ‘ and the prismatic cleavage can be traced through the whole glass. The central part is mostly non-polarizing, but small unchanged patches still polarize like the pyroxene above. The border of lighter color is a devit- rified, less ferruginous glass, and polarizes independently. This is seen to run out in a delicate fringe of glass filaments into the colorless ground, and this fringe is more or less filled with grains of magnetite, as in a common resorption rim. Outside this is a narrow band of the quartz- plagioclase ground, full of long feldspar microlites in fluidal arrangement. This is surrounded by a broad band of original calcite, partly in broad well cleaved untwinned grains, partly of grains made up of a matted fibrous mass, as if changed to aragonite. This contains several clots of the brown fibrous devitrified glass, and is outwardly interlaced with the coarse feldspar and pyroxene crystals. This variety lies next outside the long plumose variety, where the strong currents swept the suddenly formed blades out into a hotter part of the magma and the partial remelt- ing occurred, with the formation of a glass blacker and still more ferru- ginous than the normal palagonite described below. The type contains many large glass masses in which secondary altera- tion has developed many structures which are concealed in the fresher types. Their description is therefore deferred until after that of the simpler varieties (see page 118). Glass-bearing porphyritic diabase.—This is the most interesting variety of the trap. All the best specimens came from a great block near the south end of the reservoir clearing, but all the peculiarities were found on both sides of the central bands described above in the great schlieren everywhere, but not all concentrated so abundantly in single hand speci- mens. I do not think any one familiar with the Holyoke trap would think for a moment that the rock could come from that sheet. The black, perfectly fresh surface, with strong greasy luster, that suggest an ultrabasic rock rather than an exceptionally quartzose one, the large and abundant porphyritic and poikilitic feldspars, inconspicuous only because of their perfect freshness and limpidity, and the abundant clots of jet black glass with their spherulites and inclosures of calcite crystals and spherocrystals of ankerite and blue quartz, form a strong contrast to the monotonous Holyoke diabase. The large porphyritic feldspars are often almost an inch square. They are so fresh, glassy, and trans- parent that they are little distinguished from the mass of the rock. ‘They have straight triclinic striation produced by twinning according to sev- eral laws, which is often wanting over large areas and very unequally spaced. One large crystal cut on (001) extinguished at 12 degrees, indi- cating labradorite. They are poikilitic, with many rounded or elongate inclusions of the dull black pyroxene and of glass and spherulites. 102 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE These inclusions become elongate and lobate and produce a kind of micropegmatitic structure, or are in bands parallel to the outer bound- aries, or beautifully plumose, completely filling and skeletonizing the crystal (see plate 27, figure 3, lower part). The crystals are squarish, but often send out long lobes, as if formed very rapidly, and indeed a rapid growth is indicated by the abundant inclusions. They are often in large crossed forms like the cross-section of a staurolite or double arrow-head twin of gypsum (see plate 28, figure 6). The pyroxene is often very inconspicuous, and over considerable areas is found wholly in poikilitic arrangement inside the feldspars. Rarely a long twinned blade connects this type with the short plumose variety. Many large sections are almost wholly made up of cross-sections of the large fresh feldspars crowded full of these inclusions, often in beautiful dendritic arrangement and on so large a scale that it can be studied with a lens. The magnetite is also in unusual forms—in large perfect octa- hedra, the faces marked by strong cleavage lines or beautifully. skeleton- ized with inclosures of the colorless ground or of glass. When, at the surface of blocks etched by the marsh waters, the inclusions have been removed, and the magnetite rusted to a rutile color while retaining its brilliant luster, it looks like a sagenitic network of three dimensions (see plate 27, figure 3). . A surface from the large and best glass-bearing mass is covered by a fine striated slickensides made up of dark green antigorite with its prisms set at slight angle to the surface. The color of the mineral in a slide is green to brown. It is optically negative, with small optical angle.* Gabbroid diabase.—This most abundant variety of diabase in the schlie- ren is so coarse that all the constituents are visible to the eye. It is often entirely fresh, so that the lathe-shaped feldspar is- wholly trans- parent, and the coal black color is given by the magnetite and black pyroxene. The magnetite is often in large, sharp octahedra with stri- ated faces. The pyroxene is in short blades, showing the beginning of the feathery form described*above. Glass appears in small spots, hard to detect because of the freshness and dark color of the rock, and this glass appears even where the grain sinks to the fineness of the common trap. Only the granular calcite and the deep blue quartz appear promi- nently in this variety. Here alone the glass has opaque, black central portions from the great amount of the iron. This variety is especially | abundant in the great schlieren along the electric railroad. It may grow even coarser and the pyroxene occur in blades an inch long with the *In the Massachusetts state collection no. xvi is a specimen of the coarse porphyritic trap with long plumose pyroxenes, which manifestly came from the reservoir region. lt contains many quartz inclosures and cavities with dog-tooth spar crystals. The specimen is labeled ‘‘ Greenstone passing into syenite (boulder), West Springfield.” GABBROID DIABASE 103 feather-like character described above, but it does not then contain the glass in great quantity. It occurs also, but without glass (though much that we have called delessite may be devitrified glass), far north of the locality we are describing, at the surface of the ledge at Titans pier, on the Connecticut river. This type often forms also a transition to the last, in that the pyroxenes are often partly melted. A central unaltered nucleus is surrounded by a broad band of granular recrystallized pyrox- ene. Outside this is sometimes a partial band of a deep green chloritic mineral, and outside this a broad area of dark glass, doubtless derived from the entire melting of much of the pyroxene. In other cases the nucleus is surrounded by a very broad black soos band, largely made up of granular magnetite. These last varieties are generally perfectly fresh and are located far out on either side of the central band of impurities, and yet a trace of the bluish quartz appears here to show that the mineralizing fluid was carried to the outer border, but did not act to produce decomposition of the minerals formed. It promoted a widespread differentiation of the magma into a very hydrous glass, which, calculated in the anhydrous state,ismorethanathird iron. As the magnesia went with the iron into the glass the acid differentiate had almost the composition of an albite. We have thus to discuss (1) the basic glass or palagonite, (2) the spheru- litic inclosures in it, and (8) the acid halo surrounding the glass and forming a felsitic ground in all the coarser varieties, and also appearing as small aphanitic dikes. DESCRIPTION OF THE PALAGONITE The glass is in irregular portions, the longer from 1 to 30 millimeters in length, occurring often 20 to 30 in a square inch, and with a lens and the microscope many more swarms of minute patches and globules of the red brown glass appear. They are lobate and have formed in place. They are generally velvet black with highest glassy to resinous luster, more like common asphaltum than like obsidian. At times the larger grains are centrally of a slightly lighter shade of color, or two shades are concentrically interbanded as in agate. It is very soft and brittle; its hardness is 3, and its specific gravity is 1.91. Itis deep red brown by transmitted light with brownish white streak. It is red brown in thin- section, and sometimes a central portion is so deep colored as to be opaque. It is very rapidly dissolved in weak cold hydrochloric acid, and when a piece of the rock is put in strong acid all the grains of the glass soon show a cracked surface and a golden brown color like a dark resin; but the acid soon dissolves all the iron and leaves behind a white curd-like mass of silica which fills with cracks like drying starch. Often- 104 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE times the whole interior of the grain except a thin exterior film, which remains jet black, changes into a dull opaque greenish black amorphous mass. Its hardness is then 2.5-3 with the streak light green. The con-— trast between the thin continuous border of the velvet black glass and the dull greenish black slightly fibrous mass is marked and the transition line is sharp.* The shining black glass border is wholly non-polarizing. The dull black center shows in thin-section a shade of green and polar- izes very. dimly in several broad concentric bands. Each band is opaque at its outer edge and gradually lightens up more and more to the inner edge. The bands are faintly fibrous. When treated with acid the black exterior unalterated glass becomes crackled, and because of internal reflections looks brown, but soon changes into a white-fissured mass, while the weathered interior changes first to a ight green and then more slowly than the border becomes opaque white, but almost without crack- ing. It melts into a scoria with the blowpipe. There is probably a slight chemical difference between the first formed central portion of the glass and the exterior, in virtue of which the for- mer is less stable, and it is possible that iron is more protoxide in the green interior and more peroxide in the brown exterior. This glass is entirely different from that found at Meriden and Green- field mentioned above. That occurs in large masses, with isolated micro- scopic crystals of feldspar and augite, and its spherulites are microscopic drops of glass made up of concentric layers and mingled in the confused breccia and beautifully devitrified. Itis very hard, not affected by acids or alkalies,and is a liver-colored basic pitchstone, resembling the tachylite of Ostheim, in Hessen. It decomposes very slowly under weathering. This glass, on the contrary, is very soft and brittle, very easily dissolved by acid and slowly by caustic soda, and decomposes very easily into brown greasy masses, and is quickly removed from the cavities near the weath- ered surfaces. It is thus quite exactly like the palagonite of von Wal- — tershausen, and I have used this name to distinguish it from the normal tachylite from Meriden. BOTRYOIDAL GLASS WITH RADIATE FIBROUS DEVITRIFIED LAYERS In the short plumose diabase, with remelted feldspars (see page 100), the glass is more complex. Many large glass masses occur, which are generally centrally altered to a dull greenish black mass, with a thin rim of the unaltered fresh glass remaining. These glass clots are often beau- tifully botryoidal and show agate-like color banding in shades of black (see plate 30). The botryoidal cavities are sometimes empty or have a * The dull glass appears in the fresh coarse rock at the big schlieren on the electric railroad and in the central portion of the glass-bearing diabase at the reservoir. BOTRYOIDAL” GLASS 105 filling of calcite or blue quartz. Figure 2 on plate 30 shows such a botry- oidal cavity, where, within the thickness of the glass itself, is a distinct spherulite in which the few concentric bands are marked by a darker. parting and by a few original calcite crystals. The spherulite and adja- cent glass are wholly non-polarizing, but grade into a portion not visible in the slide which is beautifully banded like an agate in a few broad layers, concentric with the botryoidal interior and faintly fibrous and in this state polarizing distinctly but weakly. It is lined by a very regular double band of a lighter and completely fibrous substance in slightly - radiated tufts, often raveled out inwardly and polarizing more brilliantly than the banded glass. The fibers in both varieties of glass are positive and give a perfect coarse black cross. This brightly polarizing inner band may be thought to be an immediate alteration of the glass for a certain distance inward, produced by the absorption, before the deposition of the quartz, of the H,O whose expansion caused the cavity. This is pos- sible because both in this rock and the long plumose variety the magne- tite, which was the earliest to crystallize and which in fresh pieces incloses clots of glass, here incloses particles of the same size and shape of the green fibrous, brightly polarizing substance.* This lighter double band lines the cavity perfectly in all its windings to near one end, where it stops suddenly as if ruptured, and the remaining wall of the cavity is of the darker glass, but fractured and irregular. Where the section cuts the globular pro- jections it shows the black cross beautifully. Outside is a broad layer of the microlitic quartz-albite ground which I have below described as holyokeite. The whole is inclosed in the angular space between several of the large feldspar and pyroxene crystals, and is an isolated portion of that which makes up the whole ground of the rock in the porphyritic variety above described (see page 101) and in the holyokeite dikes. SPHERULITES, SPH ZROCRYSTALS, AND LITHOPHYSZ IN THE GLASS The glass may inclose (1) small perfect crystals or (2) crystal groups and spherocrystals of calcite, or (3) of calcite and ankerite, or (4) beauti- ful spherulites with alternating layers of calcite and the black glass, or (5) small masses of fine granular pale blue quartz, or (6) spheerocrystals of richest cobalt blue quartz, fibrous and eccentrically radiate, or (7) any combination of the above forms (see plates 28 and 29, with descrip- _ *Several delessites and chloropals occurring in basalts have a composition very close to that ot the glass. Much may be said in favor of the opinion that the so-called diabantite or delessite in the Triassic traps hereabout may be in large part altered palagonite. The diabantite always appears in the weathered rock just where the palagonite would be devitrified. On the other hand, the trap when as much or even more altered often fails to show a trace of the diabantite, which may be thought to be because the trap when fresh did not contain palagonite. The cavities in the Icelandic palagonites are also always lined by a fibrous layer which polarizes brightly (see Page 122), 106 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE tions). The glass many times molds perfectly the minute crystal ends of the calcite and the quartz and always incloses them entirely, and there can be no doubt that they have crystallized out from the liquid magma in quick succession, and the calcite has always crystallized first. The presence of calcite and quartz in solution together in the magma perhaps presents difficulties. It can only be said that the proof of the fact seems complete, and that at a pressure which would prevent the breaking uP of the CaCo, molecule, it and quartz may well coexist. The glass also contains minute lithophyse with crumpled inner sur- faces often filled with spherocrystals of ankerite and quartz. All the forms enumerated in this section are described in some detail, in connec- tion with the theoretical discussion as to their origin, on pages 117. THE LITHOIDAL OR HOLYOKEITE BASE There appears around the glass clots or groups of clots a broad bluish halo of very fine grained base, which seems at times purely quartzose or chalcedonic—at times felsitic—always compact and aphanitic. It is sometimes clear blue, but more often bluish black to black. The larger phenocrysts are often excluded from this halo for a considerable space. This base is found with the microscope to be extensively developed as a kind of mesostasis in small isolated angular patches in the plumose and coarse ophitic types, and making up the whole ground in the coarse porphyritic quartz diabase included in and between the great feathery — feldspars. It is an interesting case of micro-differentiation. Where the interstitial areas are bounded by feldspars the glass clots tend to be central. Where the boundary is partly pyroxene the glass is apt to be adjacent to it, as if a differentiation had set up in the original interstitial magma (caused by an influence of the already formed feld- spar surfaces like that which causes albite to coat the surface of micro- cline), by which the albite-quartz hyalopilitic ground grew outwardly from the feldspars, rejecting and crowding away the black basic iron- magnesium glass. Under the microscope it is a colorless base or is pale brown from a great quantity of brown dust, sometimes aggregated in fusiform shapes. With polarized light a beautiful and peculiar hyalopi- litic structure appears (see plate 30, figures 8 and 4). The delicate satiny needles of plagioclase radiate from central angular or shapeless albite grains, and are often arranged in very beautiful and characteristic loose anastomosing groups like hoar frost, the interspaces being filled with a mosaic of shapeless and blending grains. With common light these grains are indistinctly fine fibrous; with polarized light they seem to be albite filled with minute quartz rods, forming an exceedingly fine and close set micropegmatitic structure. This ground passes into radiate- LITHOIDAL OR HOLYOKEITE BASE 107 fibrous forms, showing abundant black crosses and the purity and bright polarization in whites and grays which characterize water-deposited quartz. The fibers are always positive, so that quartz and albite are hard to distinguish. It is generally full of small clots or beads of glass, forming an exceedingly fine brown dust. Sometimes the minute straight plagioclase microlites show a fluidal arrangement; sometimes many large and long crystals of apatite* and large square prisms of plagioclase occur quite abundantly, the latter with a central thread of glass like a minute chiastolite (see plate 30, figures 3 and 4). There are often many small irregular patches of calcite grains. It seems to be only when the glass is especially abundant that the aphanitic halo which surrounds it shows an excess of silica by the blue color and effervescence with soda. Generally there is only excess of silica enough to produce the micropeg- matitic structure. When such a bluish piece of the lithoidal rock is boiled for several hours in caustic soda a very considerable portion is dissolved, and the few feldspars stand out in relief,and an unexpectedly large number of minute pyrite cubes appear. The glass is, of course, also deeply dissolved and the remnant is cracked and brown. Whena fragment is fused with soda it effervesces abundantly and is mostly dissolved. THE HOLYOKEITE OR DIABASE-APLITE DIKES We pass by an easy transition from the isolated halos of the quartz- albite mixture which surround the glass and forms the groundmass of the coarsest gabbros described above to the small tertiary dikes of an aphanitic trap of light color which prove to be identical in character with the above groundmass (see plate 30, figures 3 and 4) and approach holyokeite so closely in composition that it will be convenient to apply the same name to both. They cut both the normal diabase and the secondary coarse schlieren of glass-bearing diabase and in many cases run down the middle of the coarse schlieren for long distances (see plate 25, figure 2). Under the microscope the dike material shows so many affinities to the glass-bearing quartz-diabase schlieren that the idea that they are more acid differentiates from the latter is strongly suggested. There is the same abundance of yellow glass, here always devitrified, into which long stout needles of plagioclase have penetrated, and the holyokeite ground is identical in both (see plate 80, figure 4). Calcite in small shapeless masses is often blended in the mass, and the fresh state of all * The apatites crystallizing in the albitic ground surrounding the glass would seem to add one more case to the short list of exceptions to the rule that the apatite crystallizes first. The later appearance is a proof of the rapid formation of the large poikilitic feldspars, XV—Butt. Geo. Soc. Am., Vou. 16, 1904 108 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE the constituents favors the conclusion that they were introduced together from without and are not products of decomposition. There is also very little magnetite, much pyrite, very little pyroxene, the latter in long isolated and uncorroded blades. The latter characters ally the rock more with holyokeite. The brown-dusted hyalopilitic ground is also in- dicative of the same alliance, but the rock is generally distinctly different from the first described holyokeite, though sometimes almost grading into it and very peculiar. While the normal holyokeite has the usual ophitic ground like the common diabase with the ferruginous constit- uents (except pyrite) absent, this is a peculiar variant of the hyalopilitic ground. Itsimulatesaganglionicstructure. Central feldspathic areas— sometimes regular phenocrysts, sometimes irregular angular areas—send out abundant shining white needles which sometimes branch or seem- ingly anastomose into pretty tufts like frost flowers, and the interspaces are filled by micropegmatitic albite. There is also a remarkably complete although irregular blending of this cryptocrystalline ground with fibrous © quartz, so that sometimes minute black crosses, indicative of the latter, can be seen everywhere in the field. The blue fibrous quartz thus asso- ciates most curiously on the one side with the feldspathic ground and on the other with the glass. In one slide many perfect quartz crystals just visible with the lens were scattered porphyritically in this ground- mass. The rock has thus affinities in several directions. It is plainly a trap in its prevailing characteristics, and is a differential from the main quartz diabase in the direction of the holyokeite, and a slightly greater abstrac- tion of iron would have made it a purely feldspathic trap like the orig- inal holyokeite. I have no doubt that this is the explanation of the latter rock. Wemay call the present case a quartz and glass-bearing holyokeite. On the other hand, the chalcedonic quartz is often deposited so abun- dantly in the mass that it forms a transition to the quartz-calcite vein- stones to be described in the next section but one. Indeed, I do not know of a more perfect instance of a transition from a purely igneous through igneoaqueous to purely aqueous solution than is here presented.* A COMPOSITE DIKE On plate 81 is shown a peculiar composite dike about four inches wide, which is exposed for several feet in the ledge. This dike is about * Holyokeite is also found at Greenfield as a cement of the glass and sand fragments which have been carried up into the sheet by explosions from below, and is there also a differential result of the action of the water on the normal diabase. A COMPOSITE DIKE 109 four inches thick, and is a composite of basic diabase aphanite, coarse plumose diabase, holyokeite, and quartz, and occurs near the center of the abnormal band. ‘The larger piece is deeply weathered, so that the two lighter bands stand out a half inch on the surface. The smaller piece shows a fresh fracture, and the central finer grained band is identical with the left band in the piece below. If it had been pushed a half inch more to the right the coincidence would have been more apparent. Starting at the right of the large block, the first portion (a) is the fine- grained diabase forming the ledge, presenting a weathered, uncharac- teristic surface. Next, a narrow vein of quartz (b) projects strongly. Third is a band of a fine-grained black diabase (c). Fourth, a broader band of the coarse, long-plumose diabase (d) which appears also on the upper fragment. Next isa band of the fine grained holyokeite (e), which bends over from left to right in the larger block, and is deeply etched and bleached, so that its outline does not show as clearly as it does through the center of the upper block. Lastly, a narrow band of the “ country rock,” the fine-grained diabase (a), appears in the lower rock and makes the left hand half of the upper one. This has a coarser texture just adjacent to the holyokeite (e) and grades into the common ground. This is a light gray rock, more granular than ophitic in texture. The difficulty of explaining this occurrence arises from the fact that apparently conclusive evidence is present to show that both the coarse diabase and the holyokeite must have solidified first. The inch wide holyokeite dike (e) was exposed for 7 feet, with constant characteristics, and it would seem that the diabase must have solidified and cracked to make a place for it. On the other hand, the coarse diabase has begun to crystallize on the opposite surfaces of the holyokeite dike, and its great crystals bristle out from it and diminish in size gradually as the coarse passes into the fine diabase, and it would seem that the surface of the solid holyokeite must have been there first. One may assume that the holyokeite was the result of a differentiation which took place in an adjacent area, and that, having the composition of albite, it would be highly viscid and would solidify at a higher tem- perature than the basic and highly ferruginous diabase. It is perhaps possible, but only remotely probable, that the holyokeite was carried by the violent motions as a broad, thin sheet of liquid ma- terial (a schlieren) into the still liquid diabase while so viscid as not to mingle with it, and then cooling first furnished the surface from which the large pyroxene and feldspar blades shot out, forming the coarse dia- base. The even thickness of the half-inch layer over at least two yards militates against this. 110 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE Again, it is perhaps conceivable that the diabase just become solid may have been cracked by the many explosions, and the adjacent holy- okeite, still liquid and thus at a much higher temperature than the barely solid diabase, may have filled the fissure and have imparted heat enough to the walls to cause a refusion and coarser crystallization adja- cent to itself. A further difficulty arises from the fact that while the slightly coarser diabase on one side of the holyokeite (e) grades outwardly into the normal diabase (a), on the other the coarsest plumose texture is con- tinued for 3 inches and abuts against the black diabase (c) without transition, the great needles rising from the surtace of the black diabase as from the surface of the holyokeite. The surface of contact on the black diabase (c) is a plane from which the great crystals spring, making the impression strongly that it was a solid surface when they were implanted on it. On the other hand, on contact with the holyokeite (e) the blades spring from the surface of the © latter commonly, but sometimes penetrate the holyokeite as if the two had been plastic together. The thin band of black diabase seems to be excessively rich in iron and to be the correlative differentiate of the white acid holyokeite, and how this comes to occupy its present position and to be separated from the country rock by a thin vein of quartz, which sends branches into the black rock, is hard to see. The preceding partial and tentative explanations may be taken rather as expositions tending to make clear the great complexity of the curious dike. The hypothesis that best correlates the facts would be as follows: The newly solidified trap (a) was fissured and the black extra basic trap (b), made basic by a slight differentiation in an immediately adja- cent portion of the magma, was injected and solidified. A later explo- sion reopened the fissure along its left-hand wall three inches wide and the magma filled it. The surface of the highly pyroxenic diabase (c) stimulated the growth of the pyroxenes in this magma, and they shot out nearly across the cavity, forming the long-plumose diabase (d) and thrusting, as it were, the more acid portion of the magma across to the opposite wall, where it solidified as the holyokeite. The definiteness of the plane between c and d agrees well with this, and the good degree of definiteness of the other boundary with exceptional blades of pyroxene . penetrating the holyokeite (e) would also comport well with this de- scription. The slight increase of texture of a, immediately adjacent to e, may be due to the reheating from the introduction of the new magma. The quartz (6) is an abundant last intrusion everywhere, and a last VEINS OF QUARTZ, CALCITE, AND TUFF 111 _ explosion may have reopened the fissure on the other side of ¢ and per- | mitted its entrance. This seems to harmonize with the explanation elsewhere advanced in this paper to explain the appearance of the holyokeite ground spo- radically in the coarse diabase, namely, abnormal conditions stimulate the overabundant growth of the pyroxene and magnetite, and the albitic magma remains as a sort of caput mortuuwm and crystallizes where it can find place either interstitially or squeezed out in small dikes. QUATERNARY VEINS OF QUARTZ, OF CALCITE, AND OF TUFF Small veins of quartz and calcite cut all the three types of diabase described above. They contain at the border much pyrite and a little calcite and are often distinctly bluish. Other similar veins are so loaded with the finest dust of the trap that they may be called tuff veins. In the midst of the coarse trap occur these apparent dikes of fine-grained trap, which prove to be made up of fine dust of the coarse trap of the type in which much of the diabase has been remelted. A slide was cut from a half-inch vein. This is bordered on each side by a band of the radiate granular quartz. This band is doubled on one side as if after the fissure had been injected full of the fine mud produced by the friction of the walls of the trap on each other the superheated waters had cemented the whole with quartz and then the cavity had widened, so as to separate the solidified mud from the walls (or the mud shrank on solidifying), and the narrow space on either side had filled with the pure quartz, and the fissure widening again on on one side a second comb of quartz had come in to fill the second cavity. The contents of this vein are curiously obliquely banded as if the crack had had a dip of 20 degrees, and layers of coarse and of fine stuff had been dusted in so as to lie horizontally, Others are full of finer dust and grains of the granular quartz and calcite cement. There is always a thin band of pure radiate quartz, separating each fragment of trap from the common ground, and there are also minute veins of pure dolomite or ankerite separating these tuff veins from the adjacent trap. The amount of the ferruginous carbonate is so great that the veins are always brown on the surface. This allies them with the many coarser veins in the central breccia zone. Pyrite in small cubes is common in these veins, and in one case beautiful octahedra of sphalerite occur, which are simple model-like twins after (111). 112 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE CHEMICAL ANALYSES Analbysig...\..... I. Ta. i, EL. IV. SO eee : Glass and Palagonite, Imade_ | Palagonite, | Palagonite, | quartz bear- Holyoke. | anhydrous. | Palagonia. | Seljadalr. | ingdiabase, Holyoke. SIO ise aus hacen 40.35 48.70 41.26 38.96 53.52 TAO. wales werkt 5.11* 6.17 8.60 11.62 9.70 Behe eee 24.99 30.16 25.32 14.75 8.06 EOE AL cee ee 3.55 BEDS PD a ned & le re ae ae oy ae ee 9.45 MeO ir. tia 5.48 6.61 4.84 6.29 Quan CaO. oa ee 132 1.59 5.59 ea be 5.64 WeeO cae Aa 18 0.22 1.06 68 2.24 TAO Sie ok ec be ah oa es oe eee gee a Oe oe ee Kae beste en 1.44 Lsv6 0.54 O72 1.50 H,O—below 100. SEE LO Se eg Pe ere eke ace Wee ’ 1.67 H,O—above 100. Soh Meee ane 12.79 17.85 2 AG / SiG Pee es See .20 Bi! ae (eee et ae ar | CU ee 1.98 POE Ce he EL Fig Ws ie alarerad = CONE he's Shader Gills lays he See ne a RST ee .03 CO ied el vices: Nou ee Nie ca LAE BRE eke ee. Bl heen neem 1.02 hack step ea rea acter tiie ae Meal cy ER ne a i eR None. | SED arr nme ub CLAM RER A Dee AN 3 Cede | Kanes og ce kL ana. ER Ge .36 ek Et ieic oer OF a At eet eel rial eo ALR Oe ile eee .10 Bo has where he dw ca esi eee tee eel auttie eo eke fel Se ota aes pater Ree io oe rr INGO 5 ee Ee oR NTE PA Rhee terse | RA CREA Saeed Oe None WhaOis ia 'see Lise es oo a) OGL As ek co Ge eae Ts are ee 26 Bae a5: Awake INGORE SE Sa illcee Sete hosel ocpdal CREE eae oterem oe eae ee None SO rencoe eres Ae kg EDR aay OTe ad] SARE UCR Aen ace ee None 99.86 99.08 100.00 100.00 100.21 . * With possible P2,Os. Analysis I. Palagonite in gabbroid quartz diabase, New Reservoir, Holyoke Massachusetts. Specific gravity, 1.91. By Mr George Steiger. Analysis Ia. I made anhydrous. Analysis II. Palagonite. Palagonia, Sicily. S. von Waltershausen, Ueber die Vulcanischen Gesteine. Analysis calculated to 100 after subtracting 10.99 insoluble residue. Analysis III. Palagonite, Seljadalr, Iceland. By Bunsen. Cited by von Walter- shausen, loc. cit. Insoluble residue (4.11) subtracted and the remainder cal- culated to 100. Analysis IV. Coarse gabbroid quartz-diabase, containing much palagonite and original quartz and calcite, from an average sample of the very fresh coarse porphyritic rock from the great block that furnished the best palagonite. By George Steiger. CHEMICAL ANALYSES 118 iets es dees Vv. VL VIL VIII. Analysis lV minus Holyokeite, Normal dia- Tachylite, Meriden. | glass and |mount Tom. base, mount Holyoke. carbonates. ere nro ey ls ale wie m 52.68 46.86 60.08 53.83 Nee ose wine Sek les 14.14 13.96 11.43 16.36 RO AT iw ce os dos 1.95 Bia 3.76 0.89 Me oes he mk 6 ese aw xis 9.75 4.67 MSO Fhe Foe ae ae 2 SS) aie 6.38 7.69 1.84 13 ava ng 9.38 9.42 6.21 9.81 (8 REGIS SEG eee a oa 2.56 1.85 2.43 7.89 Re oh cleo oats a! © has basins wis wnt ow oe NRCC oat ad CLS * Vid acte oe heats MI ee ee can cee .88 2.02 1.60 1.58 H,O—below 100............. Ig. 1.60 1 aes? lea «a Seo sae Merve: POU. ee ee. leek eed neces 51 al (a Zf .36 Ne aid dy [a's 8 oie wie sce L433 2.64 .86 MN. od wie és Hp URE ERR (RINE me Peg Pe ae ee ie .04 02 a EES SE ee eeriet Sareea anne tree te PP netee toe? ee 7.47 EE ater) In tak ard eR Ook Oe aka ata se we te amc [le.ocg soos Shows MEN otto nS. daly ays fs aera ued m3) 48 tt a aise rw ot Io he oe Giclee | Wis's wie 4 oo8ie = 13 17 er ne ee a a MCC fe ree Peer hake IAN an ae, ade a ee EI i a IE oe a a, Trace Oe PUL tt ek ee |. gS Ed 2 ANE ECU RES Lad ie .28 Little lost. et Soe a Bo a .03 None. None. econ Pt oe keds woh Trace. None. None. ae St SRE Pak De) Peon eens os er Ae 2 ins 14 99.80 Se er Senate 99.77 Analysis V. Compact diabase, mount Holyoke. Meanoftwoanalyses. By G. W. Hawes. American Journal of Science, 3d series, vol. ix, 1875, p. 186. Analysis VI. Diabase pitchstone from the base of the trap sheet at the ‘‘ash bed ”’ at Meriden, caused by the rising of muddy water up into the molten trap from below. By H. N. Stokes. Diabase pitchstone and mud enclosures of the Triassic trap of New England. Bull. Geol. Soc., vol. 8, 1897, p. 77. Analysis VII. Analysis LV of the glass, quartz, and calcite bearing diabase, with an amount of glass removed equivalent to the amount of water according to analysis I and of calcite and siderite equivalent to the CO,. The entire fresh- ness of the material and the visible amounts of glass and carbonates justify this. Analysis VIII. Holyokeite, east foot mount Tom, Hillebrand. MHolyokeite, a purely feldspathic diabase from the Trias of Massachusetts. Journal of Geol- ogy, vol. x, 1902, p. 508. 114 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE Analysis «2: <'0.. «+ es Be CX x, DEY pe Koei Holyokeite | Kerato- Albite, hve New (VIII) phyr (IX) | Danbury, P i. Reh minus H,O/ minus H,O| Connect- ; and calcite. | and calcite. icut. pee ee SiO, os tee ne Cee 60.13 65.37 64.77 65.73 NGL @ Plaines mre Acrteune > Sig erat oe Fe. 20.47 19,87 22.05 21.32 Bes, cnet 104° 4 oe pee 1.12 0.12 Fe Me een topmittteine seh 20 Ag! 72 108 JO ee MoO), 22 os Ie eee ee eee 1.15 16 16. ee CaO cccicre cost eeniactine © ieee Wont 2.59 .49 43 1.95 Na Oy sees Sent aan een 9.60 9.58 10.34 9.66 ThigO. 2 nia ib iin ba Bees CP veal Oba ee Ol) fc ee hee eee er sO ih MOE Nek re roe ees Beek Ooi Pe seca 1.06 1,92 3 0.95 H-O—below S00 iii. 6 sion hese Gla Bis Sate dle ke oles oe | eee 0.19 H,O—above 100... 6 ee etl eh Ga ake os Sac at neers a |r Oe o risies is doe ome ee oe een ee Trace. 1.04 Trace, Ghee DO ieee Worst Ravod Mees waeret cts ¢ ee 02° |e. oa. ects en ee OOM Gs cick oy 2 Pee ee relma Becsinin B44 eS eke Dae er PSR eee t ai Binks asmtractione cm bara Gp eee ere cr oe FAO er ee ck 0 steel iecare Gale nail ea aes Rae ee S13 an a as ee er Ne bie sa ieUede nude ue Reieke RIN ele asedisux Beano Nadas 21 a o-o-a:s o/s wie a0 al RNS ae INGO, os okie tebieldiaes wRare oa asalere @ allole vere ies ee sll ound he eget fete ces) Cate eee ae | DIO 3. tains Ce eee. eee Trace Alittle lost.) “Trace. 0 \230ceeeeeiee SAO). rane tess 6 Ries IE aE can eae iets eee None: cee e: orale eae SOs cians tecle Aha oe is ae a he a ae ee cee None. fic LA eben? fines Pee gees CUCU. cabene vip aaa hieten Gil eee eens V7 acess, aol OO 80 SE wees ae clo tal ae 3 ee eee 99.92 *Tgnition: Includes H,0. Analysis IX. Acid dike in the Trias at New Haven. By H.S. Washington. A relatively acid dike in the Connecticut Triassic area. By E.O. Hovey. Amer- ican Journal of Science, 4th series, vol. iii, 1897, p. 287. The rock is called Keratophyr. Analysis X. The Holyokeite analysis (VII) with the water and calcite equivalent to the CO, subtracted and the remainder calculated to 100 per cent. This is justifiable, since all the amygdules in the rock are of calcite. Analysis XI. The New Haven dike, analysis IX, recalculated in a way similar to that employed in analysis X. As there was not enough CaO to satisfy the CO,, it is assumed that about the same proportion of the ‘‘ ignition” is CO, as in the holyokeite analysis, and this is calculated to a mixture of ankerite and calcite, such as is common in the secondary carbonates in the trap. This is justifiable, since the rock analyzed effervesced freely with acid and since iron is ‘‘ conspicuous by its absence’’ in the slides examined by Mr Hovey. ‘This procedure perhaps overemphasizes the resemblance between the two analyses, since the New Haven rock is somewhat decomposed and contains some chlorite. Analysis XII. Albite from Danbury, Connecticut. By F. L. Sperry. American Journal of Science, vol. xxxiv, 1877, p. 8392. For comparison with the holy- okeite and the keratophyr of Hovey, analyses X and XI, DISCUSSION OF THE CHEMICAL ANALYSES 115 COMPOSITION OF THE GLASS AND INCLOSING LAVA The difficult separation of the glass was effected with great skill by Mr George Steiger. He found for the glass the composition given in analy- sis I, page 112. He remarks that as the mineral was dried at the water bath after making the ‘ Thoulet” separation some of the water coming off below 100 degrees was probably lost before the analysis was started. This would increase a little the close agreement with palagonite, which has a content of water varying from 12.79 to 20.67 per cent. The ma- terial dissolved easily and completely in hot HCl, leaving a sandy residue of SiO,. He found that the glass gave off H,O at 100 degrees = 8.57; at 150 degrees, 2.09; at 200 degrees, 3.12; at 250 degrees, 1.44; at 300 de- grees, .71; by blast lamp, 1.15—17.02. I place beside these the analysis of the palagonite of Palagonia (analysis IL) and of Seljadalr (analysis III) to show that they are as closely allied chemically as in all their physical characters, and also by way of contrast that of the diabase pitchstone or tachylite from the base of the diabase sheet at Meriden (analysis VI). It is noteworthy that if this latter were made anhydrous it is practically identical with the normal diabase (analysis V). There has been no differentiation. Analysis IV, by Mr Steiger, is made from a large sample of the glass-bearing trap and gives fully the average of the rock. For comparison with this analysis that of Mr G. W. Hawes (analysis V), of the trap of mount Holyoke, from the same sheet, a little farther north, is given. The recalculation of analysis IV, given under VIII, shows that the sample contained an unusual quantity of TiO,, which explains in part the large amount of FeO. It represents a mix- ture of holyokeite, normal diabase, and some glass. The H,O should have been somewhat greater. Attention is directed to analyses X and XI and the explanations given beneath them. The holyokeite is a fresh looking network of albite crystals with many steamholes filled with calcite. As this can not have come from the albite it may have been dissolved calcite as at the reservoir locality. ‘The practical identity of these rocks in fresh condition with albite is remarkable. The identity of the small holyokeite dikes and the lithoidal base surrounding the glass is also made clear by figures 3 and 4 of plate 30. Thus at many separate localities in the whole extent of the trap sheet this albitic leucocratic differentiate appears, generally in small amount and generally giving indication that water has been the efficient agent in its production. Only in this instance has the other differentiate, the highly basic melanocratic residuum, been preserved, unless some of the fibrous ferruginous material filling or lining cavities which we have called delessite may be this residuum in a devitritied condition, XVI—Butt. Geon. Soc. Am., Vou. 16, 1904 116 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE If we compare the two differentiates with the normal diabase we see that some CaO, MgO, and SiO, would be left over if the diabase split into the two extremes, so that it is perhaps not necessary to derive the quartz and oxides from without. It would suffice if water and carbon dioxide were brought in from without. While I have little doubt that in this case the foreign materials were brought in as I have described— that is, as a calcareous mud—since they are found in the immediately adjacent brecciated area, though not in the same slides with the glass, it would perhaps furnish a better general explanation of the presence of these two substances if we accept the remarkable conclusion of Professor Edward Suess * that there is an abundant and frequent present contribu- tion from the deep-seated molten magma of these and other volatile sub- _ stances. Then we may imagine that, the CO, taking part of the Mg and - Ca of the magma locally, the immediately surrounding portions of the magma separated into albite and basic glass, rejecting the excess of S1i0,, . while an inch away from the glass clots the normal conditions remained | unaffected and a normal diabase resulted. Indeed the enormous quantity of the highly hydrated palagonite pres- ent in many places where basic rocks are abundant, together with the great amount of water that is coming up from the interior through open craters, prompts the suggestion that the deep magma may be or may have been much more hydrated than we are accustomed to assume, and that this hydrated magma may have now and again been erupted, and iu by far the greater number of cases explosively erupted; hence the general ocurrence of the palagonite as a tuff (see page 121). In the quantitative classification all the analyses of the Triassic diabase are referred by Professor Iddings to class III, order 5, rang 4: Auvergase, subrang 3: Auvergnose. The diabase pitchstone is placed in the next higher subrang, 2 under Auvergnase, showing the slight chemical differ- ence between it and the normal diabase.t The diabase of Middlefield, Connecticut, is referred by Washington to rang 3: Camptonose, subrang 4: Camptonose ¢ under the same order. The palagonite, assuming it to — consolidate into an anhydrous rock, goes under the same class III into order 3: Atlantare and into an unoccupied place, rang 3, subrang 1. The quartz and glass-bearing diabase (analysis IV) goes into class II: Dosalane, order 4: Tonalose, and subrang 3: Hartzase. * Ueber heisse Quellen. Verhand. Gesell. Deutsche Naturforscher und Artze,”’ 1892, p. 3. + Chemical Composition of Igneous Rocks. Professional paper no. 18, U. 8. Geol. Survey, 1903, ee S. Washington: Chemical Analyses of Igneous Rocks. Professional paper no. 14, U. 8S. Geol. Survey, 1903, p.319. The same analysis is also cited under the next higher subrang as from Meriden, p. 317. DISCUSSION OF THE CHEMICAL ANALYSES 117 The holyokeite, freed from water and calcite (analysis X), finds place in class I: Persalane, order 5: Canadare, rang 1: Nordmarkose, sub- rang 4: Dosodic, and a comparison of this with the place of palagonite above gives the extent of the differentiation.* On plate 32 is given the graphic representation of the same differentiation. Ifthe anhydrous part of the palagonite had been calculated to 100 before using, the figure would have been enlarged, so that its base would have had nearly the same length as the silica line in the normal diabase. THEORETICAL EXPLANATION OF THE FORMATION OF THE HOLYOKEITE AND PALAGONITE AND THEIR INCLUSIONS THE GENERAL PROCESS The magma which had dissolved so much quartz and calcite under such temperature and pressure that they could not react on each other soon became unstable. When the point for the beginning of solidification was reached the great feathery pyroxenes in the one variety and the great poikilitic feldspars and skeletonized magnetites in the other were the first to form, shooting out into the very liquid magma, and, floating from place to place in it, were carried sometimes into hotter parts (the very change in the magma made it a better solvent), and then in whole orin part remelted. They thus sometimes inclosed large areas often an inch or more on a side wherein the differentiation and solidification took place in many separate interstitial spaces. The glass is peculiar in that more than a sixth of its mass is water ; more than a fourth iron. Nearly all the iron and magnesia are concentrated in the glass, while the other differentiate (the holyokeite) is over 92 per cent albite and anorthoclase (see plate 32). Wemust then assume a sudden differenti- ation in these interstitial residua. Albite needles bristled out from the cooler borders of these interstitial spaces, crystallization of the albite being perhaps promoted by the feldspar surface already formed as when albite forms on microline, and the quartz albite mixture filled the spaces between the albite needles and thus crowded the water, iron, and magne- sia toward the center to form a glass with the minimum of silica. FORMATION OF THE HOLYOKEITE DIKES In this way the differentiation into the holyokeite groundmass and the palagonite was effected. Where the development of the great pyroxenes * The keratophyr of Fair Haven freed from H20 and CaCO; in the same way finds the same place, Mr Washington calculates the analysis as it stands to the same class and order and to rang 2: Pulaskase and subrang 5: Persodia, a 118 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE and magnetites had been pushed to an extreme the residual and still liquid magma having about the composition of the holyokeite seems to have been sometimes squeezed out into minute fissures formed by shrink- age or explosions in the newly consolidated schlieren—fissures so small that the large crystals could not enter there. Thus were formed the nar- row dikes of almost normal holyokeite, and within them was completed a differentiation into palagonite and holyokeite like that in the coarse diabase ground, and the last of the long albite microlites thrust them- selves clear across the palagonite clots from all sides, and in their rapid - formation included a core of glass. FORMATION OF THE GLASS WITH CALCITE SPHERULITES In the supersaturated, unstable, and imperfectly blended magma the consolidation was extremely rapid, and antagonistic processes went on almost at the same time and place. The earlier formation of the holy- okeite base described above and its growth into the residual glass is only ~ a part of the process. The abstraction of the albite molecule made the glass supersaturated for the Ca and Mg carbonates, while the presence of the mineralizers abnormally prevented it from being supersaturated for iron. Another form of solidification in this glass (see plate 28, figure 1) then commenced with a minute crystal of calcite, always in the primary form, which sometimes grew in the magma to be 2 to 3 millimeters across, including threads of glass, or a small aggregate of such crystals formed and was surrounded by glass. By the absorption of heat in crys- tallizing or the abstraction of CaCO, from the adjacent magma, or both, the crystal becomes the cause of solidification for the latter, and a broad zone of glass uniformly surrounds the calcite. Again (see plate 28, figure 2), the first grain of calcite causes , as above, the solidification of a thin, spherical layer of glass around ‘tect ee the layer that has been changed and made less fusible by its own crys- tallization. Outside this layer the magma is in its original condition, and quickly a spherical shell of delicate calcite crystals surrounds the nucleus, and this is in turn the cause of the solidification of a second layer of glass, and this operation is repeated five or six times, forming beautiful spherulites as large as a small shot (8 to 5 millimeters). Some- times at the end a single crystal grows large and destroys the symmetry. In every case the calcites have sharp angles and lustrous faces, and the delicate feathery groups are so exactly enveloped in the fresh shining glass that even when they are not in spherulites the idea that the calcite has been swept in from some foreign source as solid crystal fragments can not be entertained. Both are ideally fresh and rest in ideally FORMATION OF THE VARIOUS INCLUSIONS 119 fresh trap, and soa later decomposition can not have furnished the calcite.* FORMATION OF LITHOPHYSZ WITH SPHZZROCRYSTALS OF ANKERITE AND QUARTZ (See plate 28, figures 3, 4, and plate 29.) In the cases thus far considered there was no excess of water or gas expelled in the formation of the glass, or the pressure prevented its expansion. In other cases the formation of the glass was accompanied by such expulsion, and the water or gas expanding forms a cavity, after the manner of lithophyse on a very small scale, in the midst of the glass, and thus probably further increased the amount of the glass by the ab- straction of heat. In these highly silicious patches, where silica and calcite have later crystallized out abundantly among the common con- stituents of the basic rock, conditions may have suddenly supervened (as, for example, a lower temperature) in the liquid magma which would permit the SiO, to expel CO, from the carbonate. _ Against the weight of 100 feet of the magma and its viscosity the gas or water vapor ex- panded to form cavities, sometimes a half inch across, and the walls of these cavities congealed on the instant (from»the heat abstracted in the expansion) into filmy spheres of glass like soap bubbles, and these (ex- panded by the explosion beyond a size they could maintain) shriveled, their plastic walls crumpling into complex wrinkles or buckling into the interior in sharp folds or, where more rigid, cracking and sliding past each other. In other cases the collapse was entire, and the shards of glass are seen by the microscope enveloped in the spherocrystals of cal- cite or quartz which follow or scattered abundantly in the surrounding crystalline ground (see the middle of the spherocrystal in plate 29). When the wall was more plastic it collapsed partly without fracture, and the continued development of the glass on the outside in the midst of the crystallizing magma gave a minute and perfect botryoidal inner surface to the glass cavities. One can see that these cavities were instantly occupied by solvent fluids—superheated steam, perhaps—because every point of the perfectly fresh fractured or bent glass projecting into the cavity bristles with radiate tufts of calcite or ankerite rhombs, while some of these tufts increase into the large perfect banded and radiate ankerite spherocrystals which pro- * The highly ferruginous glass produced by the refusion of the pyroxene (see page 100) is ex- cluded from this discussion. It is a deep red brown; this is yellow. It solidified in the imme- diate vicinity of the pyroxene without dissolving any other mineral; this is a residuum of the magma which abnormally held calcite and quartz in solution. 120 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE ject into the empty cavity. They are partial spheres because they rest on the irregular crumpled film of glass. They seem to derive their banded brown color from the slight oxidation of alternating layers of ankerite, between which are transparent layers of calcite. Treated with HCl under the microscope the lighter concentric layers effervesced abun- dantly as if calcite. The remaining darker layers effervesced slowly like ankerite. This continued halfan hour without complete solution. The slide was then heated and the remnant dissolved rapidly with efferves- cence. The interior of these cavities is filled by the fine, more or less radiate tufts of water-deposited quartz. More interesting (see plate 28, figures 4 and 5) is the case where the botryoidal cavities, often ¢ inch across, are filled with the radiating, deep cobalt blue quartz, whose radiations always start from a point or points on the botryoidal surface and are thus like the chondrites of meteorites. They have grown out as radiate fibrous balls until they have filled the cavities. In cross-section the globular projections of the glass penetrate into the quartz, and when the glass is dissolved the quartz grains present an exact cast of the botryoidal surface or of the spherocrystals of carbonate which have first formed on the walls of the cavity. These fibrous quartz balls have formed during the latter stages of consolidation while the extremely unstable glass and the calcite to which they fit were perfectly fresh. If the quartz had been brought in later by ordinary atmospheric waters, the calcite, ankerite, and glass would sometimes show trace of corrosion. There is still an- other form of quartz, found in the fresh rock, which is pale blue and fine granular and forms in larger and more abundant grains (see plate 28, figure 5), several quite large ones being often aggregated together. It differs from the other in that it has only a little glass associated with it. It often also fills veins which are of small extent, and in the solid and fresh lava run out to nothing in all directions. Fissures seem to have been formed in the just solidified and still greatly heated rock, and to have been filled by the quartz dissolved in the superheated water in the trap, but at a slightly later stadium than the blue quartz in the glass cavities. This quartz is often granular under the microscope; more often the whole field is made up of the most beautiful spheerocrystals, giving the black cross everywhere. The series is complete, and the succession of minerals is identical, from the glass cavities filled first with ankerite spherocrystals, followed by radiate quartz, to the great swarms of veins which cement the explosion breccia down the middle of the field, and are composed of bands of ankerite followed centrally by quartz often quite coarsely crystalline or amethystine (see page 97). The extreme purity of the calcite and quartz in the steam holes in the glass is note- worthy. PALAGONITE OF SELJADALR 121 I can conceive how, the small cavities being filled by a fluid (super- heated steam?) of different kind from the surrounding magma, the quartz can have been in part carried by osmosis through the film of glass which bordered the cavity and separated the two diverse fluids to form the pure radiate quartz, while the remaining portions of the quartz and calcite were compelled to solidify in the surrounding magma, in which needles, and even large crystals of the normal anhydrous con- stituents of the diabase, were forming in every direction. In this way all the variants of the quartz-albite ground were produced. THE PALAGONITE OF SELJADALR My mind goes back thirty-five years to the old house beside the labo- ratory of the Hofrath Wohler, in Géttingen, where that fine scholar, Professor Wolfgang Sartorius von Waltershausen, lectured on miner- alogy. I remember the detail with which he discussed the minerals of the Binnenthal and the theory of the feldspars, the series going up to krablite 1:3:24, and including the amorphous hydrated feldspars of the palagonite group. I little imagined then that a third of a century later I should find palagonite almost in sight of my own college at Am- herst and mix it up with sideromelan, but such was to be the case. The memory of those ancient lectures is not so clear as the memory of the genial peculiarities of the good teacher. So when I found a curious glass scattered in small portions in a peculiar facies of the Holyoke trap I turned to Rosenbusch,* who describes and figures the palagonite of Seljadalr as the type of an “Aschentuff,” wherein the darker grains with pitchy luster are the original lapilli, which he identifies with the nearly anhydrous sideromelan of von Waltershausen, while the lighter colored intervening bands are the very hydrous products of a secondary chemical decomposition, and form the “dull and earthy” cement of the clastic fragments. It was further stated that while this peculiar tuff was exten- sively developed in every part of the world the palagonite was found nowhere except in this clastic form. I remembered that von Walter- shausen connected the peculiar changes through which he believed the mass to have passed with a submarine deposition, and as the develop- ment of the glass in the case I was studying was connected with the unusual introduction of water into the mass I began to search for any description of a similar glass in situ, and soon I began to come on dis- crepancies. As my glass was soft, friable, and easily soluble in weak acid, l-had connected it with the palagonites immediately, and as it was jet black and very fresh I had called it sideromelan, supposing it to be * Gesteinslehre, p. 319; Mic. Phy., ii, 1896, p. 1040. 1:99 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE anhydrous and like the black parts of the rock from Seljadalr, although the latter was only brownish black. I was thus nonplussed when Mr George Steiger, who had separated the pure material from my Holyoke occurrence with the greatest skill and patience, announced that it con- tained 17 per cent of water. I then had slices cut from speimens of the “ palagonitfels ” of Seljadalr (see plate 30, figure 4), which came to the Amherst collection directly from Professor Steenstrup, of Copenhagen, and they seemed to me to be not clastic in any proper sense, but to be portions of a flow some- what brecciated by steam explosion in place, and an examination of the two good-sized hand specimens confirmed this opinion. Though quite friable they seem wholly homogeneous, and swarms of small pores con- tinue for long distances through the mass and fade away into the com- pacter portions, while the dark color in the same way fades away into the resin yellow parts. The whole seems to me not a tuff, but a much | cracked glass, containing a very few inclusions of a deep red glass and of an indeterminate basalt, and very rarely a minute crystal of augite or plagioclase. ‘The original glass was a fawn color and is still regularly scattered in remnants, often angular in the general mass, but more than half the mass has suffered a slight alteration toa bright yellow glass, sometimes slightly polarizing, and a similar layer is found around all cavities. There is also a thin layer of a brightly polarizing fibrous devit- rification substance lining each cavity inside the yellow layer. The peculiar irregular structure of the yellow glass is due to its being made up of collapsed steam holes. Threads of glass, the interior of these vari- ously collapsed and distorted steam holes, the interior of the unaltered steam holes, as well as the small solid spherulites, all show the same brightly polarizing fibrous layer, often several times repeated, and of closely the same thickness. This seems to me to bean original structure in the sense in which the fibrous structure of spherulites is original, and to have been produced by the influence of the H,O present at the instant: of cooling. This water hydrated the glass later for a further distance out from the cavities, changing the original fawn-colored glass to yellow. The repeated expanding and collapsing of the steam has often minutely shat- tered the glass, producing a micro-brecciated effect, but tracing from one end of the slide to the other the glass is essentially continuous, and I can detect no later microchemical cement, no true fragmental structure, nor any dull pulverulent interstitial matter like what one finds in an ordi- nary palagonite-bearing tuff, as, for instance, that from the Aschen Kuppel near Giessen, from which I have slides. This shows, even with the naked eye, its tuffaceous character, while the two fine large blocks of the Seljadalr rock seem with a strong lens a cracked but homogeneous PALAGONITE OF SELJADALR 198 resinous glass, which shows over considerable areas a finely ‘porous texture certainly original. A few cavities in the mass are filled with a colorless mineral of very low refraction and cubical cleavage sometimes slightly polarizing, which is doubtless analcite. The material resembles the Holyoke glass in the appearance of the collapsed steam holes and the fibrous inner devitrification layer, but in its hardness (4.5), color, and greater amount of lime it is more like the basal beds described by me from Greenfield and Meriden, and it seems probable that some part of the palagonite rock, including the specimens I have studied, had the same origin, namely, they were formed as a sub- marine flow over a bottom made up of volcanic ashes, and the water sometimes explosively penetrating from below the thin crust which had formed beneath the lava and rising up into and blending with the molten mass, caused the sudden formation of glass full of collapsed and distorted steam holes and carried up the foreign fragments of partly decomposed basalt, sideromelan (tachylite), augite, and olivine (and perhaps, also, the shells and infusoria sometimes found in the mass), into the interior, as can be seen distinctly to have happened in the walls of the trap ridges. at the localities at Meriden and Greenfield described above. On study- ing the matter further I came upon the following points confirming . this idea : Bunsen, who first mentioned the locality at Seljadalr, described it as a flow of glass 80 or 100 feet thick, in contradistinction to all the other localities of the palagonite in Iceland, which he calls tuff beds. Von Waltershausen describes it as almost wholly pure palagonite, “with few points of sideromelan,” and calls it palagonitfels, though he does not distinctly speak of it as a flow. This is in the same paper in which he describes the sideromelan,* and he describes the latter as black, and with a hardness of 6, and scarcely mentions it at Seljadalr; and as the black parts of the latter rock have a hardness of 4 or 4.5, it is not prob- able that the dark glass so abundant here (fawn-colored under the micro- scope) is the same which he analyzed and found anhydrous, nor is it possible that this dark glass isanhydrous. It is dark brown by reflected light and olive green when seen with the lens in the thin-section with transmitted light, and in this it agrees exactly with the ‘ diabase pitch- stone” from Meriden which I have described in the paper cited above and which contained 4.72 per cent H,O. The darker and original por- tions of the Icelandic glass may be not much more hydrous, while the yellow parts may have become much more hydrous by absorbing the water which frothed it up to produce the many collapsed steam holes. * Vulcanischer gesteine in Sicilien und Iceland, pp. 182, 202, XVII—Butt. Geot. Soc. Am., Vor. 16, 1904 124 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITH Indeed, the single deep red fragment of glass found in the slide, as men- tioned above, is more probably like the original sideromelan. It is plainly foreign and like the tachylite of Sasebtihl or Sordavala. Pro- fessor von Waltershausen figures and describes the palagonite as a great sheet overlying a soft and fresh tuff,* and finds all the contents of this tuff bed as scattered inclusions in the palagonitfels above. This is ex- actly as in the Meriden and Greenfield examples, except here, as the flow was over a rusty sand bed, the inclosures have a more foreign and peculiar aspect. Indeed, in an early paper upon the palagonites Profes- ‘sor Rosenbusch admits the possibility that the palagonitfels of Seljadalr may have been a lava flow,f and says that the characteristics of the | glass and its alteration can be equally well understood on either sup- position. : ; This palagonite has been the subject of most peculiar theories which one may recall briefly. Bunsen thought it due to the smelting of the normal basalt with 18 parts of CaO or K,O, and imitated it by this pro- cess. Von Waltershausen developed an elaborate theory of its forma- tion from a submarine tuff bed, where largely by the action of the sea water the adhydrous glass sideromelan was changed into a hydrated porodine cement which united the whole in a solid mass. This cement formed a series of hydrated feldspars. Professor Rosenbusch gave the correct explanation of the origin and alternation of the palagonite and its agglomeration in tuff beds, but seems to have chosen as a type of his ‘“‘Aschentuff” one of the few true and rare flows of the material. RESUME We may distinguish the following steps in the consolidation of the palagonite : 1. At the beginning of the process a minute globule of superheated H,O caused the formation around itself of a radiate-fibrous spherulite, brightly polarizing with black cross and positive sign, which may be also concentric structured by slight oscillations in its rapid growth. f{ 2. Thesame H;O may be able to expand, and then the radiate-fibrous layer is produced by the same cause around the inside of the cavity. The cavities may be spherical or drawn out by flow. The solid spheru- lites may be deformed or mutually impress each other or several be in- cluded in the continued growth, showing that they were formed in the still plastic magma. | * Loc cit., pp. 481, 483. + Neues Jahr. f. Min., 1872, p. 165. ft It is necessary to free glass from gas bubbles to prevent devitrification, RESUME ia 3. The cavities may be isolated or so abundant as to froth consider- able areas into a porous mass. 4, The expansion of the gas may be explosive and the collapse imme- diate, and the microscopic crystalline pellicle may be brittle, and its parts crack and slip on each other (plate 29, figures 1 and 2), or plastic and wrinkled into the shrunken cavity like a wet cloth (plate 28, figure 2), still very clearly showing its identity and differentness from the inclos- ing glass by the fibrous and brightly polarizing character. 5. The H,O is then slowly absorbed into the glass, and so far as it goes the pale brown glass is changed into a bright yellow slightly polarizing glass. 6. This change, extending outward from several adjacent centers, leaves angular, often concave-sided remnants of the brown glass which exactly simulate lapilli. It is also possible that the repeated small ex- plosions might occur when part of the glass had already solidified and it be shattered and the fragments moved and enveloped in the still liquid glass. 7. The H,O has expanded in a swarm of pores, which are elongated by flow and then lined by a perfectly even layer of the fibrous crystal- line material. This crystallization is thus plainly subsequent to the appearance of the H,O as a liquid and can not be called on to explain the isolation of the H,O by the formation of an anhydrous silicate in the hydrated glass after the manner of explanation of the lithophyse pro- posed by Professor Iddings. Indeed, when I examined the lithophysze in the Yellowstone their size and abundance seemed to me to demand some more abundant source than the glass itself for a part at least of the H,O. Here I should derive at least the larger part of the H,O from the sea bottom and think of it as introduced with the foreign tuff fragments found in the mass either simply by a picking up of the wet material as the lava flowed over it or by more local and explosive penetration of the thin solid bottom layer which would naturally form and generally pro- tect the liquid mass from the wet bottom. DiscussION By ALFRED R. LANE When Professor Emerson was kind enough to show to the Society at Washington his specimens of “ plumose diabase,” with the large arbo- rescent growths of augite, I was very much interested and examined them rather particularly, the more so because in one point they led my mind to an inference different from that which Professor Emerson 126 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE reached. He thought that the coarse outgrowths of augite showed extra rapid crystalline growth, produced by extra rapid cooling of the diabase by the inclosure. To me they seemed probably the result of extra slow growth, in accordance with the usual rule, “‘ the slower the cooling the coarser the grain.” As I conceive it, when inclosures were thus surrounded they cooled the surrounding lava sooner, not down to solidification, but down to the point where augite could begin to form, and the surrounding lava remained in the range of temperature of augite formation’ longer, only passing below when the rest of the flow there- about cooled below the temperature of augite formation and finished solidification. I do not mean to say, however, that chemical relations and the intro- duction of augite molecules may not also have been an important factor. There are a number of other cases of similar phenomena that I have seen and would explain in a similar way—“ plumose ” growths of augite along the contacts of gabbros and associated red rocks, and on the north side of Nahant and elsewhere contact lines between successive gushes or jets of igneous rock. | Mr Lane has been so kind as to send me a copy of the aban remarks, - which were called out by my explanation of the long-bladed pyroxene as a case of rapid crystallization ext by the cooling effect of the intruded mud. This was in a brief preliminary description of the occurrence, of which I have preserved no record, and I can not now remember how far I tried to enumerate all the factors concerned in the explanation of the peculiar long-bladed type. I certainly gave more weight to the cooling effect than I should now do, since the thin schlieren containing the long crystals have traveled so far from the central mud breccia and contain therefrom so little material in solution (H,O, CoCo,, SiO,) that the tem- perature differences seem to me to have been equalized long before they reached their present relation and approached crystallization. None of these peculiar effects occur in the immediate vicinity of the mud, There is here a somewhat coarser grain, for which I welcome the ingenious suggestion given in the last sentences of the above remarks; but for the narrow schlieren carried out from this central area, where the pyroxenes are a hundred times as long as in the adjacent trap, the prin- cipal cause for the difference in size seems to have been the chemical differences in the two magmas. The extensive development of skeleton crystals and micropegmatitic structures and the simultaneous crystal- lization of quartz and calcite, all in the midst of glass clots, speak for rapid crystallization. VOL 16, 1904, PL. 24 BULL. GEOL. SOC. AM. 2, oes "O,% y RSI ND SX OKA y/) G Plumose Diabase and Falagonite. OZ OE 2 Band of Steam holes in diabase. SORRY Ny SSK at * N11 ie Surface band of mud enclosures. Hol yoke Diabase Sheet Was /. -/- 7 AY 1° 7, fff. - Se SY ofe-B=. TrIaSSIC Sandstone. 15 aN SW ie aie SORE Cake eres ne Section through AB. MAP OF TRAP SHEET ACROSS HOLYOKE AND SECTION Seale: 1 mile to the inch ne — BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 25 Freupe 1.—Tusunar Cavirres (xX %) Figurb 2.—Bnock OF FRESHLY BLASTED ‘TRAP TUBULAR CAVITIES AND GROUP OF DIKES BULL. GEOL. SOC. AM VOL. 16, 1904, PL. 26 LONG PLUMOSE DIABASE (X 32) EXPLANATION OF PLATES aa The drifting of the long crystals into an adjacent portion of the magma, where they were remelted, would harmonize with Mr Lane’s suggestion of a slower crystallization in the schlieren, as would, perhaps, also the general development of an incipient differentiation. EXPLANATION OF PLATES Puate 24.—Map of Trup Sheet across Holyoke and Section Showing Holyoke diabase, plumose diabase, palagonite, and surface layer of mud inclosures. The section is along the line AB (see page 95). Scale 1 mile to the inch. Prats 25.—Tubular Cavities and Block of Trap Figure 1.—Tubular cavities (X 4). a. Tubular cavities formed in the superficial portion of the trap sheet by the escape of steam. 0b. The same found inverted at the base of the sheet. They illustrate the underrolling of the surface layer of the advancing sheet (see page 94). Figure 2.—Block of freshly blasted trap. This photograph of a block of freshly blasted trap (d) from the cen- tral brecciated band shows above the hammer the coarse gab- broid diabase (gd) and schlieren of the same. The oneunder the hammer faulted. Below isa branching dike of the coarse gab- broid diabase, having a thin central dike of holyokeite (h), which is also faulted. Below the hammer the holyokeite is broader and light colored. Itisjust above the h. At.the right it is a wavy white narrow band along the upper border ofthe lower branch of the coarse dike (see page 107). PLATE 26.—Long Plumose Diabase (xX 2) The pyroxene crystals have shot out into flat blades, which branch with small angle and are grouped in tufts all bent in a common direction by the motion of the lava, as grass leaves bend inthe wind. The central band bends forward and so does not show its curvature (see page 99). PLATE 27.—Pyroxene, Magnetite, and Plagioclase Figure 1,—Flat-bladed pyroxene (natural size). This pyroxene is from the central portion of plate 26, showing the central twinning suture, and the transverse basal parting (see page 99). 128 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE Figure 2.—Partially remelted pyroxene crystal (xX 40). The right central part is the pyroxene remnant (h) with the pris- matic cleavage horizontal. Inthe transition band the basal part- ing is developed (vertical in the figure) and continues both in the broad, dark-shaded band (g), which is non-polarizing glass, and in the broad, lighter shaded bands on each side, which are devitrified glass (dg), and which ends outwardly in a fringe of glass threads full of magnetite grains—the original resorption rim (rr) around the outside of the crystal. Outside this is the quartz ground full of feldspar microlites. Then comesa broad band of calcite grains (c), containing glass clots (g), which is intercrystallized with the normal pyroxene-plagioclase ground (see page 100). Figure 3.—Skeletonized magnetite and poikilitic plagioclase. If a diameter be drawn from the lower right to upper left the left half of the figure is a single plagioclase crystal, coarse poikilitic with augite on the upper and fine poikilitic on the lower half, The large black spots are clear yellow glass. The lines of mag- netite run across the slide like the strings of a harp. They seem to blend in broad black patches, because they inclose portions of the same yellow glass, which photographs black (see page 102). PLATE 28.—Glass Spherulites The first four drawings on this plate are of single particles of glass, which were surrounded by the bluish aphanitic holyokeite ground and in- closed in the coarse glass-bearing diabase. Figure 1.—Transparent calcite (X 6). The crystal is a rhombohedron of colorless transparent calcite, with lustrous faces, inclosed in jet black palagonite, which has pene- trated the crystal along the twinning plane —} R and also irregu- larly. The finer lines are cleavage (see page 118). Figure 2.—Calcite spherulite (x 15). A central grain of calcite, not seen in the drawing, is surrounded by a sphere of glass, which forms a nucleus around which a layer of calcite rhombs crystallize, followed by a second layer of glass, and soon. The outer layer of calcite is made of large and perfect crystals, always the primary rhombohedron (see page 118). Figure 3.—Calcite spherulite (x 4). The spherulite is broken across so as to show the drusy surface of a calcite spheerocrystal and the blue radiate quartz. The black glass shows a botryoidal interior from collapse (see pages 102, 119). Ficure 4.—Collapsed glass cavity (X 2). Showing on the black glass wall distinct crystals of calcite and spheerocrystals of calcite-ankerite and deep blue quartz radiate from the edges (see page 119). BULL. GEOL. SOC. AM. VOES Tee 19945 Play FIGURE 1.—FLAT-BLADED PYROXENE FIGURE 3.—SKELETONIZED MAGNETITE AND POIKILITIC PLAGLOCLASE PYROXENE, MAGNETITE, AND PLAGIOCLASE BULL. GEOL.“SOC. AM. VOL. 16, 1904, PL. 28 Figure 1.—TRANSPARENT CALCITE (X 6) Ficurk 2.—Cacite SPHERULITE (X 15) Figure 3,—CaLcrrE SPHERULITE (XX 4) fr” Figure 5.—CAVITY IN COARSE PORPHYRITIC Figure 6.—ARROWHEAD ‘TWIN OF LABRADORITE DIaBAsE (X 4) SPHERULITES IN GLASS gle i ae er a emt BULL. GEOL. SOC. AM. ‘VOL. 16, 1904, PL. 29 Figure 2.—CenrraL Porrion or Figure 1 (xX 32) GLASS-LINED CAVITY Lo, oe te 6 ~— co Fel Teg cs BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 30 Figure 1.—S8&LJADALR PALAGONITE (X 20) Figure 2.—BorryoipaL DENTRIFIED Grass Mass (X 32) FicgurRE 3.—HOLYOKEITE GROUND OF THE COARSE PORPHYRITIC Fraurr 4.—MaAreriAt FROM A HOLYOKEITE DIKE (X 32, D1aBaseE (X 32, polarized light) polarized light) PALAGONiTE, DEVITRIFIED GLASS, AND HOLYOKEITE MATERIAL « v, EXPLANATION OF PLATES 129 Figure 5.—Cavity in coarse porphyritic diabase (X 4). A large striated magnetite and a calcite project into the cavity, which is partly bordered by black glass. The cavity itself is fillec by pale blue fine granular quartz (see page 102). Figure 6.—An arrowheaded twin of labradorite, full of grains of pyroxene and glass (see page 102). Puate 29.—Glass-lined Cavity Figure 1.—Partly collapsed, glass-lined cavity (X 20). The cavity is in a coarse porphyritic glass-bearing diabase. It con- tains tufts and a syherocrystal of calcite and ankerite and is filled with fine granular quartz. The black border is a yellow-brown non-polarizing glass greatly wrinkled and broken by collapse, so that isolated fragments are found in the quartz, the spherocrystal, and the surrounding rock. It is missing in part only because the ‘ whole cavity is not preserved in the slide. Delicate tufts of brown ankerite crystals have formed on many projecting points, and one has developed into a beautiful spherocrystal having alternate layers of white calcite. The center is filled with limpid quartz, which polarizes in a multitude of black crosses and so reveals a minutely radiate texture. A part torn from the main bubble ap- pears isolated below. Figure 2.—Central portion of figure 1 (X 32). The ground below is the fine grained quartz-albite halo of holyokeite surrounding the glass (see page 119). PuaTE 30.—Palagonite, devitrified Glass, and Holyokeite Material Figure 1.—Seljadalr palagonite (X 20). The colorless spaces filled by secondary analcite are variously col- lapsed steam holes, into one of which a bent thread of glass pro- jects. They are bordered by yellow fibrous devitrified glass, and many similar bands appear in the fawn-colored glass, derived from wholly collapsed or shattered bubbles (see page 121). Figure 2.—Botryoidal devitrified glass mass (x 32). This mass is from the coarse diabase with resorbed pyroxenes. The upper third of the figure and the gray spot to the right of the center is the fine grained quartz albite holyokeite. The three large white spots are the finely radiate quartz-filling of the irregular cavity formed by the partial collapse of the steam hole. A few calcite grains mark the boundary of a perfect spherulite near thecenter. The outer boundary of this has suffered a slight fibrous devitrification. The center and the surrounding clear yellow glass is mostly non-polarizing except when it grades into an agate- like banded part which polarizes softly (not shown in slide). Where it borders on the cavity there is a double band of lighter color and even thickness, fibrous, brightly polarizing, and show- ing everywhere the black cross in perfection (see page 105), 130 B. K. EMERSON—PLUMOSE DIABASE AND PALAGONITE Figure 3.—Holyokeite ground of the coarse porphyritic diabase (x 32, polarized light). Showing its peculiar radiate hyalopilitic texture. The larger needles contain a dark central thread of glass (see page 106). Figure 4.—Material from a holyokeite dike (X 32, polarized light). Showing the same fine hyalopilitic texture asin figure 3. Many albite needles penetrate the large glass portion on the right and show a central thread of glass (see page 106). Puate 31.—Material from a Composite Dike (Xx 4) This material consists of basic diabase aphanite (c), plumose diabase (d), holy- okeite (e), and quartz (b) in normal diabase (a) (see page 108). PLATE 32.—Graphic Representations of the Constituents of the normal Holyoke Diabase. and the extreme differentiation Products; the Palagonite and Holyokeite The method employed is that of Brogger as modified by Hobbs (see page 117). hi \f ‘i e™ at de BULL. GEOL. SOC. AM. MATERIAL FROM A COMPOSITE DIKE VOL. 16, 1904, PL. 31 a” ae ‘ ” ¥i BULL. GEOL SOC. AM. VOL. 16, 1904, PL. 32 PALAGONITE Norman DIaBasE HoOLYOKEITE GRAPHIC REPRESENTATIONS OF THE CONSTITUENTS OF THE NORMAL HOLYOKE DIABASE AND THE EXTREME DIFFERENTIATION PRODUCTS ; THE PALAGONITE AND HOLYOKEITE + eg! Gr Pe ho > ae o * ' ~ es < * * . -_ Se . Pups \ + - j © = - . =. - £, ae . % B Py e ©. 34 " . * a. - iad ’ ” at at ie e x, 4 . ice a « ‘v ~ ? - . - “i, e ‘ f ° — y. b j ae 4 Pi * Pt > “se i Py ‘ ¢ ro eo * BULL. GEOL. SOC. AM. ; VOL. 16, 1904, PL. 33 BLOCK OF MASSIVE SERPENTINE of asbestos This block is 30 x 30 x 27 inches and is traversed by vein BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 131-136, PLS. 33-34 MARCH 10, 1905 ON THE ORIGIN OF VEINS IN ASBESTIFORM SERPENTINE BY GEORGE P. MERRILL (Read before the Society December 30, 1904) CONTENTS Page SES EUR SIA OP Ce 131 PEE Ol RII BDCCUMGN 6 )). siecle desu cds be lence cunanceaaens 131 Origin of the vein cavities......... fois pkecele wat ws ds 60 bre ragt ae aye ag! 6 thacsuts recat sees 133 ERIE WED CU VELIGG Soe ok. che ees Dl Fens cone sian anvaeeewronmans 136 INTRODUCTION As a result of a recent trip to Thetford mines, in Canada, the writer was able to secure for the National Museum, and through the courtesy of Mr B. J. Bennett, manager of the King Brothers mines, an unusually fine block of serpentine, with veins of the fibrous form, chrysotile (the so-called asbestos), which seems worthy of a special description, even though little, if anything, new is deduced relative to the origin of the material. DESCRIPTION OF THE SPECIMEN The block, as now on exhibition in the Museum (plate 33), is some 30 by 30 by 27 inches (76.2 by 76.2 by 68.6 centimeters) in dimensions. As seen in the plate, it is traversed from left to right by one large vein of the asbestos (chrysotile), with others not quite so wide, in an sees mately vertical direction. | Confining our attention first to the horizontal vein (some 40 millime- ters in maximum width), it will be noted that the fibers are not in all cases continuous from wall to wall, but that they are intercepted about midway by a narrow band of massive material, which shows only as a dark wavy line in the illustration. This is a very common occurrence. Though the vein is separated sharply from the massive material, there are numerous short, narrow veinlets, some almost microscopic, extending - into it from either side, above and below. These are not over 20 to 40 millimeters in length and 1 to 5 millimetersin width. Referring to the XVIII—Butt. Geox. Soc. Am., Vou. 16, 1904 (131) 132 G.P. MERRILL—ORIGIN OF VEINS IN ASBESTIFORM SERPENTINE approximately vertical veins (some 30 millimeters in maximum diamé ter), it is to be noted that they are often branched, and that, further, the branches are sometimes smallest at their point of attachment. At first glance, these vertical veins appear to be cutting across the horizontal. In reality such is not the case; with possibly one exception each and all fray out—that is, split up abruptly into a number of small veinlets, which terminate before reaching the larger. This feature is not readily apparent in the plate, owing to the half-tone process used in reproduc- tion, but is easily seen in the photograph, and better yet in the specimen. In asingle instance (see A in plate 33) the appearance is as if the vertical vein had at one time been continuous above and below the horizontal, but had been pinched out by some subsequent movement of the mass. Although not well shown in this particular specimen, these veins, as is i te ni i nie "t Mt ‘ih : i if iit Aaa ih lt my wi! Tar tls ' Teal it nooo: Figure 1.— Veins on opposite Sides of Serpentine Block. Shading represents veins, white portion serpentine. well known, are never continuous for any great distance, but pinch out to mere knife-like edges at the ends, or are variously forked and frayed, as indicated in the drawings in figures 1 and 2. A common and abrupt change in the form of the veins is shown in figure 1, representing a block of serpentine some 8 by 10 by 3 centimeters in thickness, and showing the same series of veins as they appear on opposite sides of the block. In all cases the vein material separates readily from the massive, in- dicating that the fibrous portions originated through crystallization in an open fissure, though it does not necessarily follow that the fissure was open to its present width when the filling process began. - Referring to the fibers themselves, it may be stated that they are soft, tough, and pliable, with a silk-like sheen or luster. In cross-section they are flattened or cylindrical, in this respect in strong contrast with the VEIN CAVITIES 133 fibers of the true asbestos (a variety of amphibole), which are polygonal. In diameter they vary down to .002 of a millimeter, and even less. ORIGIN OF THE VEIN CAVITIES The origin of these veins and the fibrous structure of the filling mate- rial are involved in some obscurity. The fact that the same mineral occurs in both massive and fibrous forms, so closely and intimately as- sociated, is certainly striking. That the vein material is of a later origin than the massive rock is self-evident, but why the later formed material should be always fibrous and of a distinct crystalline structure, while ij ‘lt un \ if! ays sai : berm my Wn ey ee ane cae Figure 2.—Asbestiform Veins-in massive Serpentine. the older is massive and nearly or quite amorphous, it is difficult to say, and existing literature is discreetly silent on the subject. The massive serpentine rock is itself described as an altered diorite or doleritic rock, rich in olivine.* The time that the present writer was on the ground did not permit a thorough investigation on this point, but blocks were noticed which, though highly altered, showed beneath the microscope structures more nearly that of massive enstatite rocks or pyroxenites. There is, however, apparently no doubt that they are altered highly magnesian igneous rock. Now, the process of hydration (serpentiniza- tion) in rock of this class must, provided there is no loss of material by * Asbestos and asbestic, by R. H. Jones, 1897, p. 108; also R. W. Ellis, Trans. Am. Inst. Min. Engs., vol. xviii, 1889, p. 322, 134 G.P. MERRILL—ORIGIN OF VEINS IN ASBESTIFORM SERPENTINE solution, result in expansion. T.S. Hunt showed * that the passage of olivine into serpentine under such conditions would result in an increase in bulk amounting to 33 per cent.t That some material is almost in- variably lost we have abundant proof. Nevertheless, expansion at some stage in the serpentinizing process usually results, and it is to the inci- dental readjustment of the rock mass that is commonly attributed the characteristic jointed condition and the slickensided surfaces.t These joint faces are often coated with platy and fibrous material, and bear out the author’s opinion, expressed elsewhere,§ to the effect that ordinary asbestiform structure among amphibolic and pyroxenic minerals is due to shearing stresses, the elongation taking place along the line of least resistance and being in some cases but an exaggeration of the normal cleavage property. Such structures are, however, quite different and independent of the veins we are now discussing. For the production of these last quite different causes need be evoked. Fritz Cirkel, in a paper on the occurrence and mining of asbestos in Canada,|| mentions the two prevailing theories. The one, that the fis- sures were formed in the serpentine magma as a result of cooling and contraction and subsequently slowly filled through a process of lateral secretion, and the other, seemingly ignoring the fissures, deriving the asbestos from the alteration of olivine and serpentine at high tempera- tures. No preference is indicated for either theory, nor does he express independent views of his own. In the course of a discussion with Professors T. N. Daleand J. F. Kemp the suggestion was advanced by the one that the fissures might be the result of tension or a stretching movement, and the other that they are due either to dynamic causes or produced by a shrinkage due to a loss in silica in the process of alteration (serpentinization). If due to ten- sion, it would seem to the writer that they should conform to one or more definite and determinable directions. Such, however, is apparently not the case, the veins being, as a rule, even more irregular in their direc- tion and distribution than shown in our plates and figures, and the opin- ion held by the present writer is more in harmony with the hypothesis of shrinkage.J * Mineral Physiology and Physiography, p. 506. + According to Professor Sollas but 30 per cent. See Geol. Mag., vol. ii, 1895, p. 259. t See *‘ On the serpentines of Montville, New Jersey.” Proc. U.S. Nat. Mus., vol. xi, 1888, p. 105. The Non-Metallic Minerals, p. 183. || Zeit. far praktische Geologie, vol. ii, 1903, p. 123. 4 The possibility of the vein cavities being produced by a torsional stress could be definitely decided could it be shown that the veins were themselves of more recent origin than the joints, sinee, under these conditions, any probable stress would find relief along the old lines of weakness rather than in the production of new rifts. Proof, if such shall be found, that the veins are older than the joints, while not in itself conclusive, would render the torsional theory less objection- able, though even then the lack of any definite direction in their trend would be wellnigh fatal to any such conclusion. ee VEIN CAVITIES 135 ‘Dr J. H. Pratt, on the other hand,* regards the vein cavities as pri- mary. Noting that such occur only in those portions of the serpentine in close contact with the country rock, it is argued that they were pro- duced by the more rapid cooling of these portions of the igneous magma, and have been “filled with serpentine deposited from aqueous solutions from their walls,” the filling and crystallization in fibrous form having taken place “some time before the complete alteration of the primary rock into serpentine.” With this view the present writer does not agree, since not only does he know of no instance in which an igneous rock has become cracked under the conditions described, but, further, these cracks and the fibrous serpentine are to be found among rocks which are not of igneous origin at all. In the case of the Montville, New Jersey, serpentine, for instance, the serpentine has been here shown f to originate through the hydration of a lime-magnesian pyroxene occurring in disseminated granules and nodules throughout a white crystalline granular dolomite. In this are reproduced all the characteristic phe- nomena of jointing, slickensided surfaces, platy and fibrous structures, but, as well, narrow veins of fibrous material (chrysotile), the veins being in a general way parallel to the surface of the original nodule of pyrox- ene. It is self-evident that in this case no cooling of an igneous mass can be considered as in any way effective, and almost equally evident that strains due to the movement of the material in mass also need no consideration. The writer’s own opinion, founded on the facts at present available, is that the crevices are due to shrinkage such as is incidental to the change of a highly hydrated, colloidal substance into a less hydrated * Mineral Resources of the United States, 1903. The exact wording is as follows: ““Tt can be conclusively shown in nearly all cases that the serpentine in which the chrysotile asbestos is found is of igneous origin. . . . The original rock in cooling would solidify first _along its contact with the rocks through which it had penetrated and where it was in contact with any included masses of the country rock that had been broken off during the intrusion of the molten magma. The outer portions of the molten rock would thus cool much more suddenly than the interior portion, and there would bea tendency for them to develop cracks and parting planes, In the alteration of these primary rocks to serpentine, through the agency of aqueous solution, vapors, etcetera, there would be perhaps, to some extent at least, a widening of these cracks; but in the end they would be filled with serpentine deposited from aqueous solutions from their walls, and the resulting fibrous structure of the serpentine filling these seams represents the nearest approach to a true crystallization that the mineral serpentine assumes except when it is found as a pseudomorph afteranother mineral. It is probable that this chrysotile asbestos may have been formed some time before the complete alteration of the primary rock into serpentine. “A study of the occurrence of the chrysotile asbestos in the field shows that wherever commer- cial quantities of this asbestos are found they are in the serpentine and close to its contact with the surrounding country rock or adjacent to masses of the country rock that have been included inthe serpentine. Farther away from these masses of gneiss or other country rock the chrysotile variety begins to give out. With very few exceptions, all the fibers of the asbestos are standing at nearly right angles to the sides of the seams, which would conclusively show that they were not formed by any shearing movements of the rocks.”’ +G. P. Merrill; Proc. U. 8, Nat. Mus., vol. xi, 1888, pp. 105-112. Ss 1386 G. P. MERRILL—ORIGIN OF VEINS IN ASBESTIFORM SERPENTINE and more solid form, and perhaps also to loss of silica, as suggested by Professor Kemp.. In other words, he would compare them with the shrinkage cracks which appear in clay on drying, or, better yet, those which result from the shrinkage of a gelatinous mass of iron carbonate, as in the so-called septarian nodules of clay-iron stone (see plate 34). Such cracks resemble those of the serpentine in a striking manner, their disposition to widen toward the center and pinch out toward the margin being, in the case of the septarian, due to the spherical or oval form of the body—a condition of affairs that does not exist in the serpentine. In putting forward this idea the writer realizes that he is complicat- ‘ing the commonly accepted ideas regarding serpentinization, since, as -here outlined, it involves a process of hydration and swelling sufficient to produce the jointing and slickensiding and a shrinkage sufficient to produce the cracks, and this latter, too, while the rock contained a suffi- cient amount of moisture to carry the new material into the veins, where -it crystallized. s pha FILLING OF THE VEIN CAVITIES It is perhaps a question if both the formation and refilling of the fis- sures were not contemporaneous with and incidental to this process of dehydration and shrinkage. The assumption ofa fibrous structure under quite similar conditions is sometimes seen in gypsum and more rarely in calcite. In the first named the crystallization apparently takes place _by a process of growth from one of the walls, considerable force being manifested—sufficient, it may be, to rupture the rock mass in which it is taking place.* Whether or no such a condition of affairs exists in the case of the asbestos veins is perhaps yet to be proved. It is noted, how- ever, that veins of any considerable width rarely show continuous fibers extending from side to side. In most cases the continuity is interrupted -by small fragments of the wall rock; or, again, where this is lacking there exists at some intermediate point between the walls a break or line of separation as though the crystal fibers had been pushed outward * from either wall until their extremities met. In many such cases the ‘growth has continued until the fibers are pushed past one another to a slight extent, the line of contact thus becoming jagged or saw-like. Again, there are other indications of pressure from the direction of the walls, manifesting itself most frequently in a crimping of the fibers. In brief, the views of the present writer are to the effect that the vein cavities are due to shrinkage and the vein filling to processes of crystal- lization, extending from either wall inward. *G. P. Merrill: The formation of gypsum in caves. Proc. U.S. Nat. Mus., vol. xvii, 1894, p. 81. BULL. GEOL SOC. AM. VOL. 16, 1904, PL, 34 SEPTARIAN NODULE OF CLAY-IRONSTONE The veins shown are shrinkage cracks filled with calcite BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 137-150 MARCH 15, 1905 BEARING OF SOME NEW PALEONTOLOGIC FACTS ON NOMENCLATURE AND CLASSIFICATION OF SEDIMENTARY FORMATIONS BY HENRY SHALER WILLIAMS - (Read before the Society December 29, 1904) CONTENTS Page -Summary of writer’s views previously expressed..............ceeceeee caeee 1a/ em Tat CEG UOYEY OL FIDO 14 cece ce eae apis cane weweeeie cis 138 IENUIBEE NEEL S5OUHIG, oc Suc ceo tk ope cence bwe: tebe ucsccsgoees 138 Illustrations of importance of stratigraphic rather than time Oasis wee bane Be 140 en ee la tein ee sew ew we Deana dine ewe dale werdh 140 Monterey, Romney, and nee beds of Virginia and West Virginia... 140 Ree HCA OF PONNEVIVAIIA.. . 2500s cccnc ss mccce er cee ine ces 140 Devonian section of Genesee valley, New York............... Race Sh ee 141 General discussion of the three typical illustrations...................... 141 INNER ERIN Gc) Yas) Ge bs ine) cae hie ven = 9f2 mp oye.d ocd yx BG Sitayee 143 MEME DIRE ETIC) CONCHOSIONS, «600: 025s ccc ck eens cece cect ece ts eees 144 EIEN OUR UDETEY ATID. sn ck ie wince cds wid can ens dacevea ventas 145 IT EMERGING 02/0 2), Sinise 6d (iS: Seite Wie wc os eobcle aes ciluvenaassy 147 SUMMARY OF WRITER’S VIEWS PREVIOUSLY EXPRESSED It was suggested by the writer, in a paper read before the Geological Society in 1893,* that duality of nomenclature was desirable in order to discriminate between the divisions of the time scale and those of the for- mation scale, and later (1902-’3), when the revised rules of nomenclature and classification were being prepared by the United States Geological Survey,y it was still thought that the two scales might be discussed sep- arately if only the criteria of discrimination and the nomenclature were kept strictly distinct. In 1903,f in a paper published in the Bulletin of this Society, the shifting of faunas during the continuance of their bio- logical integrity was shown to bea fact, and it was pointed out that consequent precaution was necessary in using fossils with precision in * Journal of Geology, vol. 2, pp. 145-160. +See Twenty-fourth Annual Report U.S. Geol. Survey, pp. 21-27. See vol. 14, pp. 177-190. XIX—BuLt. Grou, Soc, Am., Von. 16, 1904 (137) 138 H. 8S. WILLIAMS—NOMENCLATURE AND CLASSIFICATION classifying formations. In papers recently completed and now being published as Bulletin 244 of the United States Geological Survey the application of these principles and rules to certain specific cases is set forth in detail. SuGGESTED MODIFICATION OF RULE 14 In all these various papers the attempt was made to employ common. nomenclature, as far as practicable, in describing the intricate relation- ships existing between the fossil faunas and the geological formations under discussion. While revising the proofsheets of Bulletin 244, how- ever, the conviction became positive to the writer that one of the chief difficulties presented by this whole problem of classification and nomen- clature of geologic formations arises from the very vague and uncertain use of the word “ contemporaneity.” In the revised “ rules” referred to, although the word ‘‘contemporaneity ” is dispensed with, the idea is. 6b still perpetuated in the phrase “chronologic equivalences.” Rule 14 reads as follows: ‘‘The fundamental data of geologic history are (1) the local sequences of forma- tions and (2) the chronologic equivalences of formations in different provinces- Through correlation all formations are referred to a general time scale, of which the units are periods. The formations made during a period are collectively designated a system.” The purpose of the present paper is to raise the question as to whether “chronologic equivalences of formations’ are fundamental data of geo- logic history, and, if not, whether the “‘ fundamental data ” indicated by that expression are not in reality the similarities in the fossil faunas of formations of different provinces. In practice is it not also true that formations are not referred to a “general time scale,’”’ but to a strati- graphic scale, of which not “ periods,” but systems, are the units? Con- sidering these queries as answered in the affirmative, why should not rule 14 read as follows: The fundamental data of geologic history are (1) the local sequences of formations and (2) the similarity of the fossil faunas of the formations of dif- Jerent provinces. Through correlation all formations are referred to a standard stratigraphic scale, of which the units are systems. VIEWS OF HUXLEY AND GEIKIE Before discussing the facts of the particular case or bearing on this proposition, the exact point in issue may be emphasized by referring to the argument used by Huxley in 1862 for the substitution of ‘“‘ homataxis”’ for “ contemporaneity.” _<— OO VIEWS OF HUXLEY AND GEIKIE 1389 Huxley,* in his anniversary address before the Geological Society of London in 1862, said: ‘* Paleontology has established two laws ofinestimable importance: The first, that one and the same area of the earth’s surface has been successively occupied by very different kinds of living beings; the second, that the order of succession estab- lished in one locality holds good, approximately, inall. . . . Asaconsequence of the second law, it follows that a peculiar relation frequently subsists between series of strata, containing organic remains, in different localities. The series resemble one another, not only in virtue of a general resemblance of the organic remains in the two, but also in virtue of a resemblance in the order and character of the serial succession in each. There is a resemblance of arrangement; so that the separate terms of each series, as well as the whole series, exhibit a correspondence. Succes- sion implies time; the lower members of a series of sedimentary rocks are certainly older than the upper; and when the notion of age was once introduced as the equivalent of succession, it was no wonder that correspondence in succession came to be looked upon as correspondence in age, or ‘contemporaneity.’ And, indeed, so long as relative age only is spoken of, correspondence in succession is correspond- ence in age; itis relative contemporaneity. But it would have been better for geology if so loose and ambiguous a word as ‘ contemporaneous’ had been excluded from her terminology, and if, in its stead, some term expressing similarity of serial relation, and excluding the notion of time altogether, had been employed to denote correspondence in position in two or more series of strata. Inanatomy, where such correspondence of position has constantly to be spoken of, it is denoted by the word ‘ homology,’ and its derivatives; and for geology (which, after all, is only the anatomy and physiology of the earth), it might be well to invent some single word, such as ‘ homotaxis’ (similarity of order), in order to express an essentially similar idea.”’ Huxley further called attention to the fact that it is generally admitted by all the best authorities _ “that neither similarity of mineral composition, nor of physical character, nor even direct continuity of stratum, are absolute proofs of the synchronism of even approximated sedimentary strata; while for distant deposits, there seems to be no kind of physical evidence attainable of a nature competent to decide whether such deposits were formed simultaneously, or whether they possess any given difference of antiquity.t+ ; ‘* Edward Forbes was in the habit of insisting that the similarity of the organic contents of distant formations was prima facie evidence, not of their similarity, but of their difference of age.” t His conclusion was that ‘* There seems, then, no escape from the admission that neither physical geol- ogy nor paleontology possesses any method by which the absolute synchronism of two strata can be demonstrated. . . . For areas of moderate extent it is *Q. G. G. S., London, vol, xviii, p. xli, + Loe. cit., p. xliv. t Loe. cit., p, xlv, 140 H. S. WILLIAMS—NOMENCLATURE AND CLASSIFICATION doubtless true that no practical evil is likely to result from assuming the corre- sponding beds to be synchronous or strictly contemporaneous, and there are mul- titudes of accessory circumstances which may fully justify the assumption of such synchrony. But the moment the geologist has to deal with large areas or with completely separated deposits, then the mischief of confounding that ‘ homotaxis,’ or ‘similarity of arrangement,’ which can be demonstrated, with ‘synchrony,’ or ‘identity of date,’ for which there is not a shadow of proof, under the one common term of ‘contemporaneity ’ becomes incalculable and proves the constant source of gratuitous speculations.” * Discussing this same problem, Geikie ft says: ‘* Strict contemporaneity can not be asserted of any strata merely on the ground of similarity or identity of fossils.” ILLUSTRATIONS OF IMPORTANCE OF STRATIGRAPHIC RATHER THAN TIME BASIS IN GENERAL In the bulletin (number 244) referred to two or three particular cases are elaborated which exhibit the importance of using a purely strati- graphic basis in discussing the relations of faunas to formations and of freeing the definition of formations from all time designations. The geologist is referred to the bulletin for details. Only the main facts will be here mentioned. The three cases to which attention is called are as follows: MONTEREY, ROMNEY, AND JEN NINGS BEDS OF VIRGINIA AND WEST VIRGINIA First, the faunal combinations and successions in numerous sections cutting through the beds called Monterey, Romney, and Jennings, in ' Virginia and West Virginia, were analyzed. The facts developed show that at the base of the beds called Romney occasionally a few fossils of the Corniferous and Hamilton of New York, but thereafter the Marcel- lus, Genesee, and Portage (New York), faunas dominate, to be followed above by Chemung types in cases where these latter are not altogether wanting. In the cases where the Hamilton species appear they are associated with Corniferous forms and lie below the typical black shales called Romney (in Virginia), a formation reported to be the equivalent of the Hamilton and Marcellus. CATAWISSA SECTION OF PENNSYLVANIA The second case is that of the Catawissa section, in central Pennsyl- vania. After passing above beds which faunally and lithologically are * Loc. cit., p. xlvi. + A. Geikie: Text Book of Geology, second edition, 1885, p. 608. STRATIGRAPHIC VERSUS TIME BASIS : 141 correlated with the Genesee shales, the beds following, for 1,200 feet, are found to be dominated by an Ithaca fauna. This shows an extreme, so far met with, of the limit of expansion of that fauna. DEVONIAN SECTION OF GENESEE VALLEY, NEW YORK In the third case—the section of the Devonian in the Genesee valley, New York—the Ithaca fauna is entirely wanting, and is there replaced by the Buchiola fauna of the Cashaqua, Gardeau, and Nunda (“ Port- age”) formations. In the Seneca Lake valley a slight trace of the Ithaca fauna is seen above the Sherburne, but only slight. The Sherburne, with its Buchiola fauna, is followed by the Ithaca (400 feet), containing a rich and characteristic fauna, which is then followed by 600 feet of Enfield shales, in which the Buchiola fauna again returns with some modifications. GENERAL DISCUSSION OF THE THREE TYPICAL ILLUSTRATIONS The case of the Catawissa section, in which the Ithaca fauna occupies an interval of nearly 1,500 feet, wherever fossils occur, is a striking illus- tration of local difference in range of faunas, since the great thickness of sediments through which the Ithaca fauna ranges can not be interpreted as increase in thickness of sediments of that particular part of the section, for the sequence of faunules is in its normal order from Genesee to Chemung, but the Ithaca fauna (which is entirely wanting in the Gen- esee section), there dominates over all associated faunas from near the base to the top of the fossiliferous zone. The real problem before us may be presented by considering the first _ ease in detail. The chief facts are as follows: In New York there is a standard set of formations occupying a particular portion of the geolog- ical column, with which we are all familiar. The formations and their relative positions in the stratigraphic scale are Oriskany, Corniferous, Marcellus, Hamilton, Genesee, (“‘ Portage” or) Nunda, Chemung, and Catskill, together constituting the main part of the Devonian system of that province. The United States Geological Survey geologic folios for Virginia and West Virginia (take, as examples, the Stanton, Franklin, and Monterey quadrangles) present the same portion of the geological column, divided on a stratigraphic basis into the Monterey, Romney, Jennings, and Hamp- shire formations. In the text of the Stanton folio equivalence is implied by a table in which the divisions of the scale, with the names and symbols used in the folio, are (in an adjoining column) filled with names which have been used by various authors, as follows : 142 H. 8. WILLIAMS—NOMENCLATURE AND CLASSIFICATION hr: Hampshire formations...... Catskill. 15 ae ates Jennings formations........ Chemung. Brees Romney shale. ...3.5. 535% ¢ Hamilton. Me cies Monterey sandstone......... Oriskany. This folio was published in 1894. The Franklin folio, published two years later (1896), included the statement that “the implied correla- tions with other stratigraphic areas are not necessarily accepted,” and under Monterey, in the text, “‘ the fossil remains in this formation are in greater part those which are typical of the Oriskany formation of New York. Under Romney shale, in the same text, appears the statement “the Romney shale contains fossils, including species distinctive of the Hamilton group; those in the lowest beds comprise some species _char- acteristic of the Marcellus,” and under Jennings formation ‘“‘ fossils occur in various beds in the Jennings formation and represent the Chemung fauna.” In volume I of the Maryland survey (page 182) ‘‘four divisions are recognized in the sequence of Devonian deposits, known as the Monte- rey, Romney, Jennings, and Hampshire formations.” The inexact nature of this equivalence is indicated in the more de- tailed ‘‘ Report for Allegany County, Maryland,” in which the old name Oriskany is adopted for the first divisions, and the Romney formation is described as “corresponding in the main with the Marcellus and Hamilton formations farther north” (page 103). The Jennings forma- tion is described as ‘‘ closely related to the Chemung and Portage of the Pennsylvania and the New York Geological Surveys ” (page 106), and the Hampshire formation is said to be “‘ approximately equivalent to the Catskill of the north ”’ (see page 108). In the case of the Romney formations a twofold paleontological divi- sion of the formations is already claimed by Prosser.* It is evident that the classification and nomenclature adopted in the earlier studies of these formations is based on a general homotaxial equivalence and not upon either exact lithologic or paleontologic likeness of the forma- tions or their contents. A more minute examination of the fossil con- tents demonstrates a lack of parallelism, as is shown by Prosser’s paper on the Maryland section and by my Bulletin 244 for the sections farther south in Virginia, West Virginia, Kentucky, and Indiana. Weare therefore obliged to question the propriety of calling the forma- tions of the central Appalachian area, namely, Romney, Jennings, and Hampshire, as “ exact synonyms” of Hamilton, Chemung, and Catskill. In the case in question the facts seem to be established that a general: * Journal of Geology, vol. xii, 1904, pp. 361-372. {See Bulletin 191, U. S. Geol. Survey, pp. 351, 211, and 186, DISCUSSION OF TYPICAL ILLUSTRATIONS 143 similarity in the sequence of sediments * can be clearly recognized be- tween the sections of Maryland and Virginia and ee in the New York and Pennsylvania area of the Devonian. If lines are drawn across the sections in New York state corresponding to the more strongly marked lines of separation between different types of this sedimentation, they serve to divide off a set of standard forma- tions, and the fossils of these formations have been tabulated to consti- tute the faunas of the formations so established. Thus the Corniferous, Marcellus, Hamilton, etcetera, are established as standard formations. Passing a few hundred miles to the southward, to Virginia, similar lines may be drawn across the same general series of Devonian sediments to make three divisions corresponding in a general way with the more striking divisions of the New York section; but they do not agree in detail, in thickness, or in fossil contents. While, therefore, it may be convenient to speak of them as occupying the same general place in stratigraphic succession with the New York formations indicated as cor- related with them, the differences in all of their visible characters are sufficient to forbid calling them the same formations, or even chrono- logic equivalents. ANALYSIS OF THE FAUNAS The analysis of the faunules, as gathered in Bulletin 244, shows that the fossil faunas contained in the strata classified in the folios as Rom- ney, Jennings, and Hampshire are not the same as those of the New York formations with which they are compared, namely, the rocks be- longing to the part of the column called Romney, in central and southern _ Virginia, contain chiefly the faunas found in New York in the Marcel- lus, Genesee, and Nunda (“ Portage’’), with only traces of the Hamilton fauna near their base. The Jennings formation does, in many cases, hold a “ Chemung fauna;” but as it is followed southward along the Appalachian this latter fauna is lacking, and the succession is then di- rectly from the black Romney shales upward into a Mississippian, or lower Carboniferous, fauna, occurring in the shales after the black shale sediments ceased, while the fauna of the Genesee-Portage formations of New York chiefly fills the lower interval. The evidence indicates that when the specific Hamilton, Ithaca, and Chemung faunas are wanting in the sediments of the middle Appalachian region a fauna of the type (found intercalated between them in the New York area) seen in the Marcellus, Genesee, and Nunda (“ Portage”) of New York, dominates * Passing from limestone through black shales, fine and evenly laminated; then coarser, thin bedded, argillaceous shales and sandstones; next sandstones of thicker and more frequent occur- rence, running into coarser cross-bedded sandstone with red beds occasionally interspersed, and finally coarser sandstones with occasional conglomerates. 144 4.8. WILLIAMS—NOMENCLATURE AND CLASSIFICATION throughout that portion of the sections where the former faunas are expected. This interpretation of the facts is associated with other evidence point- ing to no unconformity or culling out of parts of the stratigraphic series, but rather to a continuous, uninterrupted sedimentation, in which the fossils so prominent at several particular horizons in New York are actually wanting, and their place in the stratigraphic series is occupied by other faunas, which in New York occupy intermediate positions. INTERPRETATIONS AND CONCLUSIONS Thus, if we interpret like faunas into equivalence of formation we are obliged to say that in central and southern Virginia the Genesee and Portage faunas range through the main part of the Romney and part of the Jennings formations, and the Jennings holds a pure Chemung fauna only in its upper or central part; but in no case do the fossil contents give a basis for saying that the Romney is the exact equivalent of the Hamilton or Marcellus, or both, nor that the Jennings is the exact equiv- alent of the Chemung, when by equivalence is meant the same fauna or fossil contents. Equivalence is therefore a correct term to apply to the formation names united by a hyphen only when the inference is drawn that the beds were deposited at the same period of time (contemporane- ity), since the lithology, stratigraphy, and fossils are all diverse for each couplet. Therefore it can not be claimed that there is either lithologic, stratigraphic, or paleontologic equivalence. Of course, it is to be ex- pected that at any particular epoch of geologic time in different regions of the earth formations. were being made which present no agree- ment in lithology or in fossils; but when we come to deal with such formations as geologic units, define them in scientific terms, and rep- resent their outcrops on geologic maps, it is all-important to restrict the terms of definition to observable facts, and to classify and name for- mations as the same only when the terms of their definition agree. In the case before us the terms of definition disagree. This fact alone is sufficient reason for applying different names to the stratigraphic divis- ions in Virginia (Romney, Jennings, and Hampshire) which are corre- lated in a general way with divisions called Marcellus, Hamilton, Che- mung, etcetera, in New York. It also follows from what has been said that equivalence, when used in a chronologic sense, may or may not mean equivalence of all or any of the criteria used in defining the for- mation. A formation may be said to be equivalent in the chronologic sense (that is, formed at the same time) when the lithology and paleon- tology are entirely discordant, and it may be true, but the evidence of the truth of the statement is complex and not discernible by simple . INTERPRETATIONS AND CONCLUSIONS 145 examination of the formations themselves. It is on account of this complexity of the proofs and the varying opinions as to both the value and the inferences regarding contemporaneity to be drawn from any or all the visible evidence supplied by the formations as geologic units that leads me to a conviction that we must dispense with any association of time with the definition of a stratigraphic formation, and use, in dis- criminating, defining, and classifying them, only those marks which are visible and can be measured, located, and described in scientific terms. In all such cases as in the one cited the facts are stated, when it is said that correlation with the Devonian system is recognized in the Vir- ginia formations, Monterey, Romney, Jennings, and Hampshire, that their division is recognized in the Virginia region on a basis of lithologic difference, and that the homotaxial relations of these formations, roughly speaking, correspond to the Oriskany, Hamilton, Chemung, and Catskill of the New York classification. It is misleading, however, to speak of the several pairs of formations as ‘‘ chronologic equivalents.” The evi- dence is not in hand to prove that they are or are not; their chronologic relations are still to be established. If any of the fossils occurring in the Hamilton formation, as defined in central New York, were strictly confined in their vertical range to the limits marking the base and the top of the portion of the section de- scribed as Hamilton, the case would be different. In such a case it would be possible to infer that the same fossils elsewhere could be inter- preted into contemporaneity of sedimentation. But the facts accumu- jated disprove such an assumption; and until the total stratigraphic range for fossils is ascertained for each area of distribution, and the - question as to whether that range is different in separate regions is estab- lished, fossils can not be used as proof of the contemporaneity of short sections of the stratigraphic column which happen locally to hold the same species. GEOLOGICAL USAGE oF TERM FAUNA A word may be added to explain the geological usage of the term fauna. , In literature there has grown up from the old conception of separate creations and the peopling of the earth with separate organism at the beginning of each geological period the idea that the fossils found in a formation are peculiar to the formation—are ‘leit fossilien.”” Thus the time when a formation was formed and the fossils contained in the formation have come to be regarded as correlative terms. The time scale is thus regarded as only another mode of indicating the strati- XX—BuoLu. Grou. Soc. Am., Vou. 16, 1904 146 H. S. WILLIAMS—NOMENCLATURE AND CLASSIFICATION graphic scale. This conception is fully elaborated in the early reports of the International Geological Congress, and the divisions of the time scale—‘‘ era,” “ period,” “‘ epoch,” “ age ”—are there regarded as strict equivalents of “ group,” “system,” “‘ series,” “stage,” so far as their ap- plication to the facts of stratigraphy is concerned. Most geologists, having been accustomed to use these terme as inter- changeable, may find difficulty in recognizing the bondage to old ideas which this usage enforces. Although the American geologists have adopted another system of nomenclature, the influence of this implication is still apparent in the © confusion of chronologic language with physical facts. To avoid this confusion the writer began several years ago to classify fossils into faunas, irrespective of the formational limits to which they were sup- posed to be restricted, and the fact has clearly developed that forma- tional limits do not by any means mark the range of either the fossils of a formation or of the integrity of associations of species into groups of faunas. Thus the fact has developed that the local formation which in a local section contains a diagnostic fauna is limited below and above, not by the beginning and ending of the life history of the particular fauna, but only by the beginning and ending of the fauna of the particu- lar locality where the sedimentation took place. In another locality, it may be not far distant, the same fauna may appear at a lower or higher stratigraphic horizon and in its integrity. It may also reappear in the same locality after having been entirely absent from the sediments for a period of time represented by hundreds or thousands of feet of sedi- ments, and in such cases the fauna is more apt to show disturbance of its contents than when the whole fauna has become shifted. This expla- nation will make it clear why the presence of a species of the Hamilton fauna in the Romney of Maryland, Virginia, Kentucky, or Indiana does not furnish proof of the Hamilton period, epoch, or even formation, as those terms are used in current literature. The fossils of the Hamilton formations of New York undoubtedly have a definite stratigraphic range, which we may hope to determine in the future, but that will not change the stratigraphic limits of that forma- tion. It will, however, enlarge the chronologic limits expressed by the fauna of the Hamilton formation. The facts already in hand show that the New York Marcellus below and the Nunda (“ Portage’) and part of the Chemung formations above are included in the time period through which the Hamilton fauna ranges. In order to distinguish the names of these two units the fauna of the Hamilton formation of New York is called the Tropidoleptus fauna. GEOLOGICAL USAGE OF TERM FAUNA 147 While it is practicable to establish homotaxial equivalence between a particular local fauna and the general fauna of some particular system (and perhaps to locate the fauna in its lower, middle, or upper portion), it does not follow that a closer degree of equivalence can be established between two local faunas by the same criteria. In the first case the general equivalence may be proven in a case in which few or none of the species are identical species, but even in case the identity of species in two formations is proven, the equivalence so established is only within the limits of the known range of the species of the fauna. This range in most known cases is at least as much as a third of the thickness of the system in which it belongs. We are therefore forced to the convic- tion that in the correlation of local formations the same species of fossils alone (when so much as 50 miles of distance separates their stations) can not be relied on for establishing more than a general homotaxial relation of the formations compared. The limit of range of every species is far greater, both above and below, than is indicated by any local formation in which it occurs. Geologists have already recognized the fact that uniform conditions of sedimentation are local as well as tem- porary, and the same principle must be applied to fossils. The ver- tical range of individual species, as well as that of their combination into faunules, varies greatly with the local conditions that prevailed during the life of the species, and thus their place in the vertical strati- graphic column varies with geographic distribution. SUMMARY AND CONCLUSIONS What has been said refers to the geologic formation as a particular -mass of stratified rocks occupying a particular position or horizon in the geological column and whose geographical extent may be determined. What I am urging is that greater clearness of description and accuracy of statement will be attained in describing such formations if all refer- ence to time relations be dispensed with. Let us speak of them as “homotaxial;” but when their lithologic or paleontologic characters differ, let us say so-and call the formations by different names and indi- cate their-general relations to the standard scale simply by bracketing them as Devonian, or, if the correlation be more definite, as Eodevonian or Mesodevonian, as the case may be, but not as the equivalent of a for- mation which we regard as formationally distinct. The next point I have to make is that in the definition, and particu- larly in the mapping, of formations it is important to make the defini- tion so that the mass not only can be located and recognized by the terms of the definition, but so that its limits can be distinctly recognized 148 H. 8S. WILLIAMS—NOMENCLATURE AND CLASSIFICATION in the field. Inthe examples selected the line between the Romney and Jennings and that between Jennings and Hampshire are frankly de- scribed in the folios as so indefinite as to be recognized with difficulty. Geologists are not always so frank on this point when they return from the field. The evidence brought forward by the analysis of the faunas shows that if the lithologic transitions are perplexing in the case in question the sequence of faunules is also unsatisfactory for establishing precise lines of separation between two contiguous formations. In the course of the present investigations the recurrence of faunules has become an established fact, and not only for a short vertical distance through the beds, but recurrences of faunules of the same fauna have been traced for a thickness of hundreds and in one region up to about 2,000 feet of sediments in which intercalation of entirely distinct faunas has taken place. With these facts in view, we are deceiving ourselves when we presume that half a dozen fossils of particular species occur- ring together determine the stratigraphic horizon, so that the local line below or above them may always serve for the limits of the formation. - The fossils do indicate a general portion of the geologic column, but it is only indefinitely, somewhere within one or two thousand feet of thick- ness of strata. They do not alone establish equivalence of formation, when by formation is meant a definite part of the stratigraphic column set off by definite boundaries below and above. It may also be stated that the sharper the paleontologic transition from one fauna to another (in ascending through a series of strata) the stronger is the certainty that the local limit thus assigned is not the stratigraphic equivalent of a similar sharp transition between the same ~ two faunas elsewhere. The very fact of the definiteness of the faunas is also sure evidence that they had lived a long time before the first trace of them in the section and lived a long time after the highest traces in the local section, and the sharpness of the transition from the one to the other is evidence that the superior fauna was not derived from the lower one, but that the local succession is a result of movement and replacement of the faunas themselves. The conclusion of the matter to which the facts force us is that not only lithologic but paleontologic facts are local. The fossil contents may completely change, often very rapidly and often in a few miles. The fossils undoubtedly are the means on which we chiefly rely for determining that kind of equivalence which is called contemporaneity and homotaxy ; but it must not be overlooked that the characters of fossils (that is, the marks by which species and genera are distinguished) are extremely longranging. Fossil species were not ephemeral things which changed every few feet of thickness of sediments. SUMMARY AND CONCLUSIONS 149 In the use of fossils, for determining the geologic horizon of the formations containing them, the essential fact open to investigation is the presence or absence of the fossils themselves. The presence of in- dividual fossils indicates, not some narrowly limited horizon, but a gen- eral portion of the stratigraphic scale represented by the system or by a large subdivision of it. By close discrimination characters of narrower vertical range can be detected, but up to the present time very few char- acters of fossils are known whose vertical range-limits are so narrow as to indicate an horizon of less than about a third of a standard system. Certain modifications of current usage are suggested by the facts here presented, which may be expressed by the following recommendations : In seeking to perfect the rules governing nomenclature and classifica- tion of sedimentary geologic formations should not the following prin- ciples be applied: 1. The abolition from the nomenclature, definition, and classification of geologic sedimentary formations of all reference to time or time rela- tions. The application of this rule would result in the rewording of rule 14 of the “ Revised Rules,” as suggested at the opening of this paper. 2. The adoption of lithologic characters, stratigraphic position, and paleontologic contents as three (at least) chief means for discriminating and defining sedimentary formations. 3. The revision of technical nomenclature in the following particulars : In the place of time scale use the term stratigraphic scale; in place of contemporaneity use homotaxy; in place of age, in its general sense, apply the term horizon'with the definite and technical meaning of posi- tion in the vertical stratigraphic scale; in place of period use the term system ; it has already become common practice to speak of group, series, and formation on this general principle; in the place of earlier or older adopt the terms inferior, lower, subjacent, or underlying; in place of younger or later adopt the terms superior, superjacent, overlying ; and in general in the selection of descriptive terms to apply to sedi- mentary formations, both in definition and correlation, employ physical characters (such as composition, texture, structure), special dimensions, and position in a vertical stratigraphic column in place of any chrono- logic terms, these latter, so far as formations are concerned, being infer- ential, not observable, and incapable of accurate application on account of the wide divergence of opinion as to the modes of their determination. In proposing these changes in usage it should be stated that it is not intended in any sense to exclude the consideration of time relations in geological discussions, but rather to remove from the definition and dis- crimination of formations any reference to their supposed time relations 150 H. S. WILLIAMS—NOMEMCLATURE AND CLASSIFICATION in order that actual history and time relations in geologic history may be studied with a greater precision and freed from that vague prejudgment which naturally arises from confusing chronologic with physical and spacial ideas. | It may be mentioned further that in the discussion of fossils, faunas, and faunules time relations must be considered, since heredity and evo- lution are time questions; but for such discussions the formations and their exact position in a stratigraphic scale must be first established, independently of the fossils they contain, before the historical relations of the fossil faunas can be accurately discriminated. In the same way that physical geography must stand ona basis of physical definition entirely independent of political boundaries or political characters, so must formational geology derive its definitions, terminology, and classifi- cation from the characters which are actually possessed by formations and which can be examined, measured, and defined before the history of the organisms preserved in them can be accurately determined and before questions of the absolute time relations of geologic events can be established on a firm basis of fact. cannes ; 4 ta) i mh S is or UPT PTET... Wy oo 4 24 i. Jt the ba j wer) BULL. if la Spuyren Duywi/ Causeway N 2 ca 13 13a Line of Gourdiirigs 10 Hoth | /4 IF 16 17 25 oF 25 26 27 28 29 JO F/ of2 33 J4 *s, Show *“Borings or Lire of 42nd St. GEOL. SOC. AM. Spuyten Duyvil Bridge : Spuy7e, 2 ayy Ae Kings Bridgé Dychrarss Cup : /\ Wasturgror Bridg é { Se HN hgh Bridgeé =) ve New New York Aqueduct 2 ; oh \ 8th Avene Bridge \ M Combs Dar (CertrallBrcge | NYCAR Brage over Cromwells creck Fapla Transit Terre! /45 1h Ft Briage. Madson Avenue BHAGE Fark Avenue RR BrHAGEe 4a Avenue Bridgé Ln Avenue BHAGE Willis Averrue Briage } lesth Street Fees. (22rd Strees Heef Proposed Hell Gate HR Briagée. Hell Care reefs. East Aiver Gas CosTurne/ Abandoned Farry Bridge fast iver Bridge No4 } Abandoried East Fiver Turire/ Projected East iver Turre! of Ferrhh L6th7 Street Heet ord S1reer Heer Last Hiver Bridge No2 Last Fiver Bridge NoJ Brooklyn Bridge Coerties Heer. api Trarstt luraeé/ Deep well or Governors Is/ard Darnond Freer heer Discovered 1902 g Abarrdoriéd Hudsor fiver lur77e/ 3 Pray ected /wasor liver Ture! of € Permmsylvariia Pre & Borings torock online of 59th St § (Gh) Scale or Ene imile LEGeN7a Sef C3 Made Lana -.—.— Yobable Faults Approximate Margins ot Depressed Serpentine (Outcrops doubly hachureda) Basalt underlaid by Gneiss Q — «ss Bridges or Tenne/s Areas OR wees Proposed Sridges or Tornels 33 734 Je it Greiss Horsts rear Harlete Alver Ww e Gneiss pes 1o o Limesfone ro) Sandstone ist = SKETCH MAP OF MANHATTAN ISLAND ing its hydrography, the location of some of the river sections, and the distribution of dikes (the last named after Julien). BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 151-182, PL. 35 APRIL 12, 1905 ORIGIN OF THE CHANNELS SURROUN DING MANHATTAN ISLAND, NEW YORK * BY WILLIAM HERBERT HOBBS (Read before the Society December 31, 1902) CONTENTS Page NN coir Neg sien) GS IO) nik wp %. 00.6.4 rk NG bb Ge Kare wwe 8 ws 152 Theories of origin of the New York water front........... ccc. cece eee eeee 153 Form of the rock floor beneath the New York water front. ................ 155 Generally accepted view ............ ny Nise ax suai sha ae oa iat ates te epee 155 IRENE DS ChE SED LLORES 223500) 500! 4 Seiden ts wins bos Bohn dele dal See A dee Be bo le . 156 Sebee CLTERERLRLIOY C8 TERUG .2 oeiceels vic whee ek ods w so.e be ase seat 156 mveet Dagyi bridge: Harlem river... -... ojo. de deie ee se eee Zhe kon ete eh VCRINGUS CUE. dels s/c ccas ceede fs sentrcecsnes es 158 Washington bridge, Harlem river............ . Bers nage bs She aes 158 Deeeet ite ItAtieml TIVET= = coc cca ck ea sa sv atapneccs Ean 159 New New York aqueduct siphon under Harlem river............ lec AOe Eighth Avenue bridge, Harlem river............... sea ne Mi paeae 160 eee) Pan, (Cetitral) DESC... sk Lek eee Mid bale waned aes . 160 Soundings to rock east of Central bridge..............-....222 .00 161 New York Central Railroad bridge across Cromwells creek ......... 161 One hundred and forty-fifth Street bridge....... ....... .. Ae ae ty. 162 eam ranean, (OAMEL 6. is le id pe ciee ee oe aces NA nat cat MEV pal dt 162 Madison Avenue bridge, Harlem river. ..................002-e0-++ 162 oo a EST Te 0) ag 4 7 a 163 We evenne eres Harlem: TIVEr. (0c! bele eee ee ee ee 163 ume DIN A CMU MREMCECE ATG 0) |. bist (sate nlc ews de eis elec eet tle ede ds 163 Mig ARON DRPEE 0s fake a di bee eee Pepaae He Steed 2253. See 164 One hundred and twenty-fifth and One hundred and twenty-second Sopa ROESARE) MIME ROME No re le ee ce wa win piv ald Mec 3 bin ts veil nye bs 164 Reopeercn: biol) Gate rstlroad PTIdge.. .< 5... ceesnen ses eee onde nce 164 apn RO Pe Aen Gen fas Shi xo eda sive ses ee eb ka ee 165 East River Gas Company’s tunnel (Blackwells Island tunnel)....... 166 Projected but abandoned Rainey bridge...............--.....seeees 167 East River bridge number 4 (Blackwells Island bridge).............. 168 JN SO 17 :) en 169 a eRe Ng. elects inital pid tir raiwinied duy.sk asia eis on, alc een © 169 Section on center line of Forty-second street produced.............. 169 *This paper was read December 31, 1902, at the Washington meeting, and is mentioned on page 543 of volume 14, under title “‘ Configuration of the rock floor of the vicinity of New York.”’ XXI—Buut. Geox. Soc. Am., Vor. 16, 1904 (151) Page Tunnels of the Pennsylvania Railroad Company..................6. LO: * Twenty-sixth Street and Third Street reefs.................. s.eee- Lyd East River bridge number 2 (Williamsburg bridge) .... ......... peta East River bridge number's. . 0.0... 02%) «<\s@ oe pele 171 Brooklyn bridge .. 0.05. 5024 sd eeeet nes wines Je ee 172 Coenties veer rt). th oe te eet ere wallid eile be oe hate dite cer 178 Rapid Transit tunnel under East river... J.:/2%, ....-<:. 2 eee 174 Deep well on north shore of Governors island... ........ ....+.--. 174 Diamond reef...°.13012. 3.0308. AAG CIO ie Reef off. Battery .:.. 0.20.05 2255 Sen bok Seis ee Oe ee 176 ‘Jersey flats) 5... sn Pee be chi tee oe eae ee alg ag Hudson River tunnel from Jersey City to New York (McAdoo tun- nely. eee: Pendens s 2 ie wieaiee AES AE Se oie oe 176 Projected Hudson River tunnels of the Pennsylvania Railroad Com- PAY oe SSRs ae Da lets ein enone are See ene aes ee 176 Proposed New York and New Jersey bridge.................--+0e8- 178 Conclusions respecting the origin of the channels.. ............. ......---. rg Harlem itivers i302 20504 G0 oy id ee Rae ite eo oy 179 Mast PIver.: .22%55 ocho ae Sabo Bo eee ete Ce on ee 180 Hudsom river i) 2.000.454.5024 2 Sie OE ites oe ee ee 180 Former hydrostaphy of Manhattan island .....2°).0.. 2.2202.) e eee 181 INTRODUCTION The engineering work recently completed and that now undertaken in and about the city of New York make it easily the focus for such enterprises on the planet—enterprises the cost of which must be esti- mated in the hundreds of millions. To mention but a few, there are the subway and the tunnels of the Rapid Transit railway, the East River bridges numbers 2, 8, and 4, the proposed tunnels of the Pennsylvania Railroad Company from Weehawken to Long Island City beneath the two rivers and Manhattan island, the completion of the old Hudson River tunnel from Jersey City to New York, and the dredging of the Buttermilk and Ambrose channels and of the Man-o’-War and Diamond reefs by the United States government. These are some of the larger of the recent enterprises. Of earlier ones, the government work in making the channel of Hell Gate by removing Flood and Mill rocks, the con- struction of the Croton and the new New York aqueducts, Dyckmans cut in the ship canal at Kings bridge, the East River Gas Company’s tunnel from New York to Long Island City beneath Blackwells island, and the numerous bridges spanning the Harlem river, have all facilitated the work of the geologist within the New York city area. Taken to- gether, they have furnished more than 95 sections crossing the rivers forming the water front of the island, many of them revealing the nature THEORIES OF ORIGIN 153 of the subjacent rock, and not a few giving practically complete sections across it. The present is, then, an especially favorable time to study the geology of the channels, and it is of great importance that observations be now made and recorded, lest the opportunity be forever lost. THEORIES OF ORIGIN OF THE NEw YorK WATER FRONT The unexcelled harbor facilities of New York city have furnished a fascinating subject for study, and writings of geologists on the structure of Manhattan island have devoted considerable space to it. As is well known, the water-courses surrounding the island form sharply incised tideways of quite exceptional depth. Stevens accounted for the location of the lower Hudson, the Harlem, and the East rivers both by the posi- tion of supposed belts of limestone and by longitudinal faults along the channels.* Dana ascribed their formation entirely to the presence of supposed limestone belts, and formulated the theory which has since formed the subject of many papers; all, so far as is known, in support of the theory, although comparatively little evidence has been adduced for it. Dana stated his views as follows : Tf “‘From the distribution of the limestone, as exhibited on the map, and the fact of its easy wear or erosion, we derive explanations of several topographic features of New York island and the adjoining region. For example, we learn— Why Harlem river has its present position and depth, and its north and south course; why there isan ‘ Eighth Avenue valley ;’ why the ‘ Inwood parade grounds’ are a broad rolling region from the Harlem to the Kings Bridge road ; why, south - of the Inwood Presbyterian church, there was a Kings Bridge road valley, to fix the position of that old highway; why Shermans creek bends around the Fort George heights; why Cromwells creek exists and the valley or ‘ Clove’ to the north ; why Fleetwood park is low and nearly flat, except its western side; why Third avenue in Harlem and the region east of it is low; why wide flats (with small exceptions) extend from East river more than two-thirds of the way across the island, just north of Central park, and, perhaps, why there is an East River channel.”’ An objection to the full acceptance of his theory led him to add: ‘‘The limestone lands that are not low may owe their height to the fact that erosion follows water-courses; but, besides, the rock when in nearly vertical beds— usually the fact in such places—is generally of a firmer kind, because the pressure which gave the beds this position, served to compact the rock and so favored closer and better consolidation.” * History of the geology of New York island Ann. Lyc. Nat. Hist. N. Y., vol. viii, 1865, pp. 108-120. + J. D. Dana: Geological relations of the limestone belts of Westchester county, New York. southern Westchester county, and northern New York island. Am. Jour. Sci. (3), vol. xxi, 1881, pp, 25-443; also ibid., vol. xxii, 1881, pp. 313-315. a 154 w. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND Newberry had earlier (1878)* offered a partial theory in assuming the East river to be the old course of the Housatonic, and the Harlem and Spuyten Duyvil Creek tributary streams. Such an explanation, how- ever, fails to cover the exact location of these courses, which was the point aimed at by Dana. Professor Kemp in his paper on the Geology of Man- hattan island t accepts Dana’s view, but adds that the course of the Harlem river east of the McCombs dam requires another explanation, and suggests that as its course below that point corresponds very closely to the direction of the glacial scratches, the river may have been directed by the glaciation. In a brief note on the Blackwells Island tunnel ¢ he has extended the theory to include the eastern channels, which was merely suggested by Dana, and has quoted Dr F. J. H. Merrill in support of this view. Later Merrill and Kemp have each elsewhere published the sameexplanation.§ Gratacap, || in his recent pamphlet on the Geology of New York, refers to the view of Stevens that the Hudson river is under--. laid by limestone, and adds that it is an extremely unlikely supposition. In the recently published New York city folio Willis { apparently ascribes the direction of the lower course of the Hudson to the location of weaker beds along its channel, though he adds that this eee: may be modified by the existence of faults. He says: ‘‘Tt [the Hudson river} had had a complex history, dating back to the Schooley plain (upon which it probably began its course across the highlands), involving adjustment to belts of weak rock, such as dolomite and arkose, and possibly in- cluding effects of that faulting which has been described as traversing the Newark rocks and of which there is some evidence in the physiographic relations of Staten island. All that early history of the river awaits elucidation.” Regarding the area east’of the Hudson, Willis says: ‘*Tt probably happened that originally the rivers of the area had other courses indifferently across the dolomite, schist, and gneiss, but such is the manner of growth of streams and valleys that, though they were once so situated, the larger ones must have become rearranged in precise adjustment to the lines of weak rock. . . . Adjustment of valleys to belts of easily eroded rocks is a condition which is reached only through prolonged competition among growing river sys- tems, and when it is so perfectly accomplished as it is east of the Hudson, it indicates that the streams have been long at work carving this intricate mosaic of rock masses.” *J.S. Newberry: Geological history of New York island and harbor. Pop. Sci. Month., vol. xiii, 1878, pp. 641-660. + Trans. New York Acad. Sci., vol. vii, 1887, p. 61. tJ. F. Kemp: The geological section of the East river at Seventieth street, New York. Trans. New York Acad. Sci., vol. xiv, 1895, pp. 273-276. 2F. J. H. Merrill: The geology of Greater New York. Trans. New York Acad. Sci., vol. xvi, 1897, p. 371. J. F. Kemp: Trans. Am. Soc. Civ. Eng., vol. xxxix, 1898, pp. 79-80. || L. P. Gratacap: Op. cit., p. 39. 4 New York city folio, p. 17, first and fourth columns. THEORIES OF ORIGIN 155 Merrill in the same folio * has carried the theory to its extreme. Not only are the windings of Spuyten Duyvil creek and the Harlem river supposed to follow the course of a belt of limestone, but even Long Island sound near New York is supposed to have the same origin. These assumptions are based largely on the “ white residuum ”’ obtained from dredgings by the dock department.f If we grant that this white material may be the residual product from the solution of dolomite, its presence at the bottom of a swift tideway like that of the Hast and Harlem rivers can have but little significance, when it is remembered that the Harlem flows between limestone walls for a considerable portion of its course. In a recent extended report on the Glacial and post-Glacial history of the Hudson valley, Peet { has favored the view that the Hudson water front was in post-Glacial time a lake impounded by a moraine, the inlet now existing through the moraine having been formed by the cutting down of the lake outlet so as to form the present Narrows. He says: “In conclusion, it may be stated that while no single argument seems to be fatal to the salt-water hypothesis accounting for the Hudson water body, unless those drawn from the phenomena on the outside of the moraine be such, it is likewise true that the facts are not fatal to the lake hypothesis, unless the sponge spicules reported from Croton represent salt-water species. Aside from these sponge spic- ules, the weight of the evidence seems to be in favor of the lake hypothesis.”’ Julien, in discussing the faults on the island, has called attention to the evidence of the transverse fracturing of the rock at Spuyten Duyvil creek : **At other localities, as, for example, at the huge pit on Spuyten Duyvil creek, portions of the hornblendic rock are traversed by innumerable veinlets of quartz or pegmatite, indicating a shattered or even brecciated mass.”’ ForM OF THE Rock FLOOR BENEATH THE NEw YorK WATER FRONT GENERALLY ACCEPTED VIEW From the above review it will appear how generally Dana’s theory has been accepted and also how little evidence for it has been adduced. It has come to represent, however, a body of opinion to some extent in- herited, and it probably owes its favorable acceptance largely to the fact that the principal channels about New York island trend approximately * Op. cit., p. 4, columns 3 and 4. + Merrill: Op. cit., p.4. See also F. J. H. Merrill and D. S. Martin: Note on the colored clays recently exposed in railroad cuttings near Morrisania, New York. Trans. New York Acad. Sci., vol. ix, 1889, pp. 45-46. t Charles Emerson Peet: Glacial and post-Glacial history of the Hudson and Champlain valleys. Jour. Geol., vol. xii, 1904, pp. 415-469, 617-660. Pee 1 et ee , 156 Ww. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND with the general strike of the rocks, as it does also to the known occur- rence of limestone at the bottom of the Harlem gorge. To the writer it has seemed that there is a general tendency among geologists to over- estimate the importance of limestone belts in conditioning the location of valleys and low areas in the topography. The exceptional oppor- tunity for testing this theory in the New York city area by reason of the numerous artificial sections across the river channels presented itself to him in the season of 1901, when considerable study was made of the area. It is perhaps true that nowhere in the world could the general problem be investigated under more favorable conditions. Theinquiry begun nearly four years ago has been prosecuted since as opportunity has offered, until data gathered are about as comprehensive as they can well be made now, though they are likely to be somewhat augmented in the future. As already stated, the generally accepted hypothesis assumes that the channels of the Hudson and East rivers and the sinuous course of the Spuyten Duyviland the Harlem, are alike underlaid by limestone, even where gneiss forms the wall rock on either side. This latter difficulty Merrill has sought to explain for the Spuyten Duyvil locality by assum- ing a sudden and steep pitch to the southward of the northern gneiss mass (Fordham gneiss) and the appearance of gneiss belonging to a higher horizon superior to the limestone (Hudson schist) on the south of the river. Hesays: ‘* High cliffs of Fordham gneiss border the north shore of Spuyten Duyvil creek. ’ The course of the creek is at any given point approximately parallel to the strike of the gneiss at that point, and the latter is everywhere seen to dip toward the creek. This definite relation between the windings of the creek and the variations in the strike of gneiss is due to the fact that the creek occupies the position of dolo- mite beds which formerly overlay the gneiss, but which have been almost entirely removed from view by solution and erosion. Several small outcrops of dolomite can still be seen at low water on the south shore of the creek dipping to the south along the cliffs of Hudson schist which line that shore” (loc. cit., p. 4, column 4). My own observations do not confirm the above statement in all par- ticulars, especially the uniform steep southerly pitch of the beds. In the other localities where gneiss appears, forming both walls of the river, an exceedingly sharp longitudinal syncline must be postulated in order to bring in the limestone. DETAILED STUDY OF SECTIONS Order of presentation of results——The data which have been secured bearing directly on the configuration of the surface of rock beneath the New York water fronts, refer both to the depth and to the character of _ DETAILED STUDY OF SECTIONS 157 the bed rock in the floors of those channels. They are considered in order, proceeding from the mouth of Spuyten Duyvil creek eastward about the island, so as to consider first Spuyten Duyvil creek, then the Harlem, East, and Hudson (North) rivers in succession. The sketch map of New York and vicinity shown in plate 35 will serve for location of sections. Inasmuch as the problem, so far as it is petrographical, is to locate the areas of limestone, it will be sufficient to refer to the harder rock as gneiss without attempting to refer it to one horizon or another. Spuyten Duyvii bridge, Harlem river.—This bridge is not founded on the rock bed, but rests on piles. Wash borings were, however, put down to the bed rock by Mr C. B. Brush, the depths varying from 94 to 124 feet below mean high water. No record has been preserved regard- ing the kind of rock at the bottom, and it was probably not known. Through the courtesy of the engineers of the New York Central rail- road, the writer was allowed to examine the samples taken from the bottom of these borings, but they afforded no clue regarding the char- acter of the rock.* eT ii | il | i ee t — BRONX Q so" too" et Ne Fri@ureE 1.—Sketch Map of Spuyten Duyvil Creek. Along the line of the swing-bridge. Both north and south of the creek at that point the gneiss rises abruptly to form bluffs. In the railroad cut north of the creek local northerly pitches of the beds were found, and the conclusion is drawn that no southerly pitch sufficiently steep to carry the overlying bed of limestone to the bed of the river across the strike is probable. The writer’s view is that the stream here flows along a cross-fracture, as stated by Stevens, and apparently concurred in by Gratacap.{ Under the causeway of the New York Central railroad across the northern edge of this gorge piles were driven to a depth of 35 feet without meet- ing rock.§ | *See, however, ‘“‘The Spuyten Duyvil swing-bridge, etcetera.” Engineering News, vol. xliii, 1900, pp. 397-398, figure 1. ¢ Loe. cit., p. 114. 1 L. P. Gratacap: Geology of the city of New York, 1901, p. 36. § Information furnished by Mr F. L. Chase, engineer of bridges, New York Central and Hudson River Railroad Company, 158 Ww. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND Kings bridge and Dyckmans cut.—In the vicinity of Kings bridge and along the line of the ship canal the limestone is much in evidence. The government work at Dyck- mans cut consisted in the excavation of a canal through a reef of white marble, and is always referred to by the Corps of Engi- neers, United States Army, as the unique instance of excavating or dredging in lime- stone in all their operations about New _ York island. Washington bridge, Harlem river.—This bridge has three piers—one on the west of the river, resting on gneiss, and two on the east side (see figure 2). The east pier rests on gneiss, while the middle pier is located in the alluvial material at the very edge of the water. This latter pier was sunk by caissons to “an irregular rock, partly gneiss, partly marble, with veins and pockets of very hard quartz.”* Mr William R. Hutton, the chief engineer in charge of the construction, has obtained for the writer a very interesting letter from Mr George Leighton, his assistant engineer, regarding the rock found beneath the middle pier. A portion of Mr Leighton’s letter follows: . 200° 100° fas Gnelss, Marule, and Fault Rock Figure 2.—Section across Harlem River. On line of Washington bridge. ‘‘ The gneiss was nearly all of granular character, and the upper portions of the ledge had disinte- grated. I donot recall that it was more micaceous than the ledge at and above Boscobe! avenue, but it was much inferior to that in structure and density. The gneiss immediately overlying the marble was much harder—better material. My recollection is that the marble filled a fault in the gneiss, in width something over half the width of the caisson, and extending nearly parallel with the sides of the caisson. There were quartz deposits in the marble—one of considerable size, which would strengthen the fault theory. I carried a large specimen of the quartz to the office.”’ HARLEM RIVER * William R. Hutton: The Washington bridge over the Harlem river at One hundred and eighty- first street, New York city, a description of its construction, 1890, p. 21, plate xxxii. DETAILED STUDY OF SECTIONS 159 It thus seems probable that the central pier of the Washington bridge is located on a line of faulting which follows the east bank of the river. High bridge, Harlem river.— All but two of the seven piers of this bridge which are in the river are on piles; the other two, however, are on rock.* Dana has fortunately preserved + a sectional view of this bridge, “ re- duced from a tracing obtained for me by Mr Benjamin S. Church at the engineer’s office in New York ” (see figure 3). Mr Church was resident engineer in charge of the Croton water works. From this interesting section it appears that the walls of the Harlem gorge in this vicinity are formed of gneiss, which plunges down beneath the silt and other river deposits along a steep incline; and, further, that in the middle of the rock gorge there rises, pedestal like, a reef of marble which has furnished a base for the three central piers of the bridge. 2888 dolled) NEW YORK ISLAND Nene jon aa: a ie ge ond Gnelss While ee : _~ ~—_— ‘ar Limestone ee WESTCHESTER CO. Figure 3.—Section across Harlem River. On line of High bridge. New New York aqueduct siphon under Harlem river.—The section of the aqueduct in this vicinity has furnished geological information of the first importance. The aqueduct crosses the river in an inverted siphon, the lowest arm of which is about 300 feet below the surface of the river. - Vertical shafts descend through gneiss on either side of the river to this depth, and a horizontal connecting arm penetrates approximately 800 feet of ‘‘ hard limestone or marble” beneath the river (see figure 4). The section shows the walls of this layer of limestone to be roughly par- allel and to dip at very steep angles to the east. The first intention of the engineers was to locate the horizontal arm of the siphon at a more moderate depth (see figure 4), but on running the drift to the east a “ crevice” was encountered with crushed stone at the boundary of the limestone. The report says: **Under Harlem river the gneiss and limestone were found to meet on a con- spicuous diagonal line without any visible solution of continuity, the particles of the gray and of the white rock being so closely intermingled at the surface of contact that the exact line of separation could not be traced.”’ t The report and the section are therefore in accord in showing the *F. B. Tower: Illustrations of Croton aqueduct, New York, 1843, p. 110. 7 Loc. cit., pp. 435-436. t Report New York Aqueduct Commissioners, 1887 to 1895, p. 88 and sheet 32. XXII—Buut. Geou. Soc. Am., Von. 16, 1904 160 w.H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND western boundary of the limestone beneath the river to be along a line of dislocation diagonal to the bed- ding. It is highly probable that the east- ern boundary is also located on a parallel fault, but the evidence from this section is not conclusive. Eighth Avenue bridge, Harlem river. — This 1" = 200" Horizontal {° =500° Belov A= Lah ia 7 ” Vertical (tc Grade. _ _ aT os Ter {A sf ae Seales yn - - = Crotan Datum: a5 5. Gress Froth All efforts to learn the depth to which they were carried have been futile. McCombs Dam (Cen- tral) bridge.—AI piers of this bridge rest on a reef of gneiss which runs from one bank of Harlem river to the other at depths vary- ing from 24 feet unde, the central pier to 3( feet at the west bulk- head line and 27 feet at the east bulkhead line. Mr Martin Gay, assistant engineer of the department of bridges, New York city, who has fur- nished me with this information, adds that “on the line of the bridge to the east- ward the tees drops to an unknown depth in the swamp and comes to the surface again at One hundred and sixty-first street” (see figures : Hard Lime Stone of Marble Figure 4.—Section across Harlem River. Along line of new New York aqueduct. 8 HARLEM FIVER Hard Ciels§ 2 | j | { } | ! ! 1 | { t 1 ! | t t { | l t I ' 1 { I i { ! t | t ‘ 1 { ! G =< ZLIID aa 2 ‘ i paetaeieae aos | ra Ry = MANHATTAN bridge rests on piles.. DETAILED STUDY OF SECTIONS 161 5and 6). Mr Gay was in all the caissons himself,and thus was able to examine the rock personally. This reef crossing the Harlem is clearly the continuation of the Fordham Heights ridge, which was also once ex- tended southward by a range of gneiss outcrops now largely removed.* Soundings to rock east of Central bridge—The data used in constructing this profile (see figure 6) were furnished by the department of public parks to Professor I. C. Russell, who, in his valuable paper on the Geology “ae ik RA ten ode tee -<——. . —erere TX Oo. io0' 290' Figure 5.—Section across the Harlem River. Along the line of McCombs Dam (Central) bridge. of Hudson county, New Jersey,f gives the depth of the underlying gneiss at all points in the section. This section not only covers the bottom of the river, but extends some 2,000 feet north of it into the lowland of Cromwells creek. It is interesting in showing how faintly the present channel is indicated in the configuration of the surface of the bed rock. New York Central Railroad bridge across Cromuells creek.—This bridge, as will be seen from the map, is almost a continuation to the eastward of the HARLEM RIVER ” i; : y; eee /f/- ie MANHATTAN. BRONX SASS p 2 5 ‘ Op 4, L re ‘ 0 "50 00 30d ~- S00" AMT 2 ae, : : Gneiss 7 S44 Ficure 6.—Profile of Rock beneath Harlem River. On line 400 feet east of Central bridge. McCombs Dam bridge, and data which it furnishes should therefore be considered in relation to those of the last-mentioned section. The bridge rests on piles which go down 120 feet, but do not reach to bed rock.t It thus appears that the reef of gneiss which comes so near to the surface _ of the Harlem river at McCombs dam plunges off to great depths but a short distance farther to the east, the line of this declivity corresponding in direction with the extension of Cromwells creek northward or parallel *See Viele’s map and a paper by Ries in Transactions of the New York Academy of Science, vol. x, 1891, pp. 113-114. + Loe. cit., p. 75. t Information furnished by Mr F. L. Chase, engineer in charge of bridges, New York Central and Hudson River Railroad Company. 162 w.H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND with the faults believed to have occurred in the western gorge of the Harlem river. If they there exist, Fordham heights would appear to be a long “horst”’ or “ briicke”’ lying between parallel faults and agreeing closely, both in size and orientation, with the “horst” of Washington heights, a little farther west. One hundred and forty-fifth Street bridge-—This bridge has five piers: four of which rest on whitemarble and the fifth on “ extreme hard-pan.”’* Crystalline limestone is also exposed at the surface near the New York end of the bridge. The section along this bridge reproduced in figure 7, like that at certain other bridges, affords no evidence that the river here flows in a rock gorge. rMud _ .. ” +FUING = eocwes se py ps ‘sedan, , Lime“Srone/ ; evel and Boulders: - K -75.08 =o ae fe, i + Resa HAVA Fan: Figure 7.— Section across the Harlem River. Along line of One hundred and forty-fifth Street bridge. Rapid Transit tunnel—This tunnel connects Lenox avenue and One hundred and forty-first street, on Manhattan, with Westchester avenue, in the Bronx. The approximate surface of the rock in the vicinity of Harlem river along the line of this tunnel is brought out in the profile of figure 8. The rock penetrated is limestone. Madison Avenue bridge, Harlem river.—This bridge rests on piles, which at no point reach to bed rock.{ Russell gives the data for a section across the river at this point.§ *Information furnished by Mr F. W. Allen, engineer for the contractor, who also kindly sup- plied data for the section. + Information and profile furnished by Mr George 8S. Rice, deputy chief engineer, Rapid Transit Railroad commissioners. {Information furnished by Mr A. P. Boller, consulting engineer, and Mr McLean, engineer of the comptroller’s office. 21. C. Russell; Geology of Hudson county, New Jersey. Ann, N. Y. Acad. Sci., vol. ii, 1882, p. 75. DETAILED STUDY OF SECTIONS Park Avenue Railroad bridge.-—The approx- imate surface of the rock in this section is brought out in the sketch profile which has been furnished by Mr Alfred P. Boller, con- sulting engineer, who was the engineer in charge of construction (see figure 9). Gneiss is found near the present grade beneath the piers on both banks of the river. The north bank of the river is formed by a nearly ver- tical wall of gneiss, which was followed to a depth of 70 feet by the piles beneath one of the piers. The strike of this wall is, accord- ing to Mr Boller, about at right angles to the line of the bridge. This wall would appear to represent a line of dislocation. Beneath the central pier of the bridge a drill was put down 104 feet to either a ledge or a boulder of limestone. Mr Boller says: ‘‘We found some pieces of marble, which is probably the same as the Tuckahoe marble in Westchester county.’’ Third Avenue bridge, Harlem river.—In con- structing this bridge the caisson of the south- west pier, which is 70 feet from the Manhat- tan shore, rests on a ledge of gneiss at a depth of 52 feet. The bottom of the caisson was 24 feet by 110 feet, and an unbroken area of gneiss extended the entire width of the cais- son and was uncovered for 30 feet of its length.* The other piers of the bridge rested on boulders (see figure 10). Second Avenue bridge-—The north and south abutments and the-four piers of this bridge rest alike on the detrital material of the river. Borings were sunk beneath the piers to depths ranging from 40 to 482 feet without encount- ering rock of any kind. ‘The profile of figure 11 has been kindly furnished by Mr J. J. R. Croes, who was chief engineer in charge of construction. ‘(jouun) yIsuRI, pldvy) xuoIg ‘onuea 1ozsoyoysaM 07 ‘URZBYUL ‘ONUEAY XOUOT pUv 490198 4S.1Y-AZ1OJ PUB PeIpUNyY UO Wo1y ‘ay Wa,LMET ssouay uoijzvag—"y AAAOTYT NVLLIVHNVW oor = oS = =< FUO{S JUNT ete 1a QUO {S FLT, * Information furnished by Mr Edward A. Byrne, assistant city engineer, New York. 164 Ww. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND Willis Avenue bridge. —About 200 feet out from the Manhattan shore the center pier of this bridge, at a depth of 80 feet below mean high water, rests in part on a ledge of marble and partly on boulders. The other piers rest on boulders (see figure 12). Mr H. A. Byrne, assistant city engineer, has preserved a sample of the marble, which was submitted to the writer for examination. It is a white coarsely crystalline rock like that quarried at Tuckahoe. BRONX SIDE (NE) MNESTOME- ‘ or boulder" 150° a ion’ 200" Figure 9.—Section across Harlem River at Park (Fourth) Avenue. One hundred and twenty-fifth and one hundred and twenty-second Street reefs, East river.—The first mentioned of these is in the mid-channel of the East river off One hundred and twenty-fifth street, and the second about 300 feet off the foot of One hundred and twenty-second street. Both reefs are of gneiss and have been reduced in height by the Corps of Engineers, United States Army.* South Fest Pier ase #0 ovr of Caisson .MANHAT TAN, ; : /One155. Boulders Soulders 0. 50°: 190" 200° Ficgurk 10.—Section across Harlem River. Along line of Third Avenue bridge. Projected Hell Gate railroad bridge—The New York Connecting Rail- road Company, of which Mr A. P. Boller, is chief engineer, has pro- jected a double-track railroad bridge across the East river at Hell Gate * Report of the Chief of Engineers, United States Army, 1896, part i, p. 106. The writer is in- debted to Captain Edgar Jadwin, Corps of Engineers, United States Army, for information regard- ing the composition of these reefs. DETAILED STUDY OF SECTIONS 165 from Long Island City to the Bronx. * Diamond drill borings have been made for piers at the crossing of Hell Gate.t Gneiss was found on the Long Island side at a depth of about 100 feet below mean high water. On the Wards Island side of this crossing the rock is hornblende gneiss. Biss oe 100" . 200° (Se ee Figure 11.—Section across Harlem River. Along line of Second Avenue bridge. Between Wards and Randalls islands the gneiss surface was found at a depth of only a few feet below tide. ¢ Hell Gate reefs—The government works at the entrance to this channel Center Prer Caisson North Slate, MANHATTAN | Boulders Figure 12.—Section across Harlem River. Along line of Willis Avenue bridge. included the removal of large masses of reef a considerable distance out from Hallets point, a shelving ledge extending into the ship channel from Mill rock, and the removal of the cap of Flood rock in the middle *See Engineering Record, vol. xli, 1900, p. 453. 7 See plate xix for location of bridge. t Information furnished by Mr A. P. Boller, chief engineer. 166 Ww. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND one of the channel, the latter the greatest work of the | | op) kind ever attempted. All of these obstructions | 4 i Wy ee are of gneiss, and similar rock is to be found ex- posed on the south shore of Wards island on either side of the small bay which indents that shore. Together these exposures of gneiss make almost a complete section across the river. , East River Gas Company’s tunnel (Blackwells Island tunnel).—There is considerable literature treating of the construction of this tunnel. A shaft was sunk on the New York side to a depth of 139 feet, from which point a drift was carried on a descend- ing grade of 0.5 to 100 beneath both channels of the East river and Blackwells island to a vertical shaft on the Ravenswood, Long Island,side. The roof grades of the tunnel are on the New York side 104 feet below the level of mean low water and on the Ravenswood side 116.6 feet below the same datum plane. The writer is indebted to Mr J. Vipond Davies, one of the engineers. in charge of the construction, for much information of scien- tific value obtained during the construction of this tunnel. Mr Davies states that in sinking drills they were in no place able to penetrate more than 5 or 6 feet into what appeared to be hard gravel, sand, and a great quantity of very large boulders lying on the bed of the river. It is probable that the strong tides of Hell Gate have scoured out all the finer material, leaving only such as was too large to be transported. It was thus found im- possible by means of borings to determine in this vicinity the depth of bed rock below the river sur- face, but it was considered by the engineers in their work that the line of the bottom of the river was practically the surface ofrock. The roofgrade of the tunnel is approximately 32 feet below the deepest point of the river bed in the west channel and about 65 feet in the east channel. The accom- panying profile (figure 18) has been slightly modi- fied from one published by Aims in the Journal of the Association of Engineering Societies. The section is of special interest to geologists, particularly as regards the composition and also the struct- Figure 13.—Section across East River. Along gas company’s East River tunnel. Se ee ee eee DETAILED STUDY OF SECTIONS 167 ural planes of the rock penetrated by the tunnel. Much attention has been given to the rock composition by the engineers who have written on this tunnel, and their descriptions have been supplemented by the petro- graphic studies of Professor Kemp. For the greater part of the distance the rock is a gneiss of varying hardness, in this respect reaching a maxi- mum in the hornblende gneiss encountered under the east side of Black- wells island. Under the bottom of the west channel a considerable thickness of soft decomposed rock was found, which Kemp has shown to be, in part at least, the residue from the alteration of one large and a number of smaller pegmatite veins. Beneath the east channel a thick- ness of about 350 feet of hard white marble was found enclosed between soft decomposed bands, which appear to correspond to nearly vertical fissure planes. The loose material filling these zones about the fissures is in part decomposed pegmatite and in part altered mica schist. Ina personal letter Mr Davies has added some valuable notes concerning the nature and direction of the fissures both in the east and the west chan- nels. He says: “*Under the New York end of the tunnel is highly micaceous gneiss rock. Just outside of the pier line it intersected a fissure. . . . Under the east channel is a seam of about 350 feet of dolomite, bounded on both sides by fissures of com- pletely decomposed (and soapstone-like) mica schists. . . . The dip of these fissure faces is about 22 degrees off the vertical.* The strike is slightly west of north. “*Curiously, when we were excavating the foundations for the Rainey bridge, which was to be built at Sixty-fourth street, New York, we found at the site of _ the pier on the west side of Blackwells island a fissure exactly corresponding in the direction and strike to the one tunneled through opposite Seventieth street. In this last case, however, although it appeared 15 feet wide at the top surface of rock (some 10 feet below mean low water), we excavated it entirely out to a sharp V, at a depth to the point of the V about 25 to 30 feet below mean low water.” Fissures found along the line of this tunnel seem, therefore, to be in- cluded in a parallel series, the strike of which is a few degrees west of north and the hade of which varies from 78 degrees east to a very steep angle west. One of the faults discovered under the pier of the Rainey bridge belongs to a different series following the course of the channels in this vicinity. Projected but abandoned Rainey bridge.—The location of this bridge was to have been from the north side of Sixty-fourth street, in Manhat- tan, to the foot of what is named on the maps Harsell street, Long Island City, though as a matter of fact no such street exists. The work * This appears to refer to the fissures in the western channel only, for the sections of Aims and Jacobs show that those of the east channel are quite close to the vertical.—Eb. XXITI—Butt. Geo. Soc. Am., Vor. 16, 1904 168 W. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND QUEENS MANHAT TAN 400 Figure 14.—Section across East River. On line of East River bridge number 4 (Blackwells Island bridge). on this bridge is of scientific interest, chiefly be- - cause of the intersection of faults belonging in two different series, as described in the last para- graph. East River bridge number 4 (Blackwells Island bridge).—This bridge is to connect Fifty-ninth street, Manhattan, with a point between Rogers” and Charles-streets, Queens. Diamond drills have been put down along the proposed line of this bridge beneath the piers on the Manhattan side of the river, on both shores of Blackwells island, and on the shore of Long Island City. The sur- face of the rock is in all cases but a few feet be- low the surface of the ground. The rock on the east side of Manhattan island at the water’s edge. slopes very abruptly toward the channel (nearly 45 degrees). On the west and east sides of Black- wells island the rock surface is at about the level of high water and is badly decomposed for a dis- tance of from 2 to 5 feet below the surface. On the Long Island shore the rock surface is also at about the level of high water and is found more decom- posed than upon the shores of Blackwells island, it having been necessary to remove about 15 feet of this material before suitable foundations could be secured. Through the courtesy of Mr R.S. Buck, former chief engineer in charge of this bridge, the writer was enabled to examine cores from all the drill borings and found them to represent a granitic type ofgneiss.* From Mr H. A. La Chicotte, since engineer in charge, the section was obtained which has been reproduced in figure 14. It is quite likely that the decomposition of the eneiss at the shores along this section is to be explained in the same way as is the decomposed rock along the fissures in the Blackwells Island tunnel ; namely,through the local fracturing of the rock and the consequent introduction of water from the river. During the construction of the Black- wells Island tunnel the operations were greatly * Information regarding East River bridge number 4 was obtained from Mr R. 8S. Buck, former -chief engineer, and Mr H. A. La Chicotte, who succeeded him in charge of this bridge. DETAILED STUDY OF SECTIONS 169 hampered by the oozing into the tunnel of water and decomposed rock in the form ofa thick mud. In the construction of the new New York aqueduct the same difficulty was encountered at the boundaries of the limestone (see figure 14). Abandoned East River tunnel.—The only work that was done on this tunnel was the excavation of a shaft in section 10 feet by 8 feet and 80 feet in depth to the grade of the proposed tunnel. The shaft was sunk 10 feet in the earth to the underlying gneiss, into which it was carried 70 feet. ‘There occurred a very serious explosion when this stage of the work had been reached, much damage to property being caused and a number of persons seriously injured. The conditions are now favorable to an early resumption of the work and the ultimate completion of the tunnel. Man-o’-war reef—The Man-o’-war reef forms a continuation of Black- wells island to the southwestward and has the form of an elongated prow. Considerable work has been done by the engineers of the United States army in removing the upper portions of this reef. Through the courtesy of Captain Edgar Jadwin, of the Corps of Engineers, United States Army, the writer secured a number of specimens from this reef. These speci- mens represent respectively a granitic type of gneiss, a hornblende gneiss of undoubtedly igneous origin, and a dense basalt like that so character- istic of the Newark areas of the Atlantic border. It seems likely that this latter rock may be from a portion of a narrow dike within the series of crystallines. Section on center line of Forty-second street produced.—Through the courtesy of Mr August Belmont and his chief engineer, Mr Allan A. Robbins, the data from the series of core drillings made in the bottom of the Kast _ river have been placed at my disposal. The revelations concerning the topography of the river bottom and of the nature and depth of the bed rock beneath are set forth in figure 15, which has been drawn from the blue prints furnished by Mr Robbins. The line of the section passes near the Man-o’-War reef, and those cores which were obtained near the banks of the river and near the Man-o’-War reef are all hard gneiss. Near the middle of the western channel drills (number 2 A) brought up a core of compact dolomite. All others, save those already described as from the vicinity of the reef and shores, were of decomposed rock material, which in some instances the drillers were unable to separate from the sand and silt immediately overlying. The samples from these drillings have, however, been placed at the writer’s disposal and sub- jected to a careful microscopic examination. The material is throughout partially disintegrated gneiss, consisting of feldspar, quartz, and biotite, with smaller amounts of muscovite, magnetite, and garnet, Consider- 170 w.H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND able differences in coarseness oo eR esRgesRs es of texture are noted, and one = specimen was even labeled clay. This specimen differed, however, in no respect from the others except in its greater fineness. Tunnels of the Pennsylvania Railroad Company.—These ; proposed tunnels are to run ‘ | from the area between Thirty- second and Thirty-third streets, Manhattan, to that between Flushing and Bor- den streets, Queens. Core drill borings on both shores of the East river along the line of these proposed tun- nels have been made under the direction of Mr A. Noble, chief engineer of the com- pany for the East River sec- tion. Between First avenue, New York city, and Front street, Long Island City, no less than 35 core borings and 112 wash borings have been made. The core borings, which are all on the land- ward side of the bulkhead lines, are in gneiss exclu- sively. The wash borings in et e the river give with much de- em tail the surface of the rock ee nelss (After Robbins.) Paw EG. 600° EAST PIVER o = Mean High Wafer Gress. Fiaurk 15.—Section across East River. On center line of Forty-second street produced. Dolor te bottom of the river over a belt about 600 feet wide on FSS the Long Island shore, but E they, of course, fail to reveal "" pesGHAg- TO SUSIE ZL VIZ Rye the nature of the rock at the 7 bottom. The depths in the section of figure 16 are averaged from the values for approximately equal distances from the shores. Quite remarkable and sudden changes DETAILED STUDY OF SECTIONS Via in level are, however, revealed by the indi- vidual borings. ‘To bring out in some meas- ure these differences in the profiles along the northern and southern margins of the belt it’ is only necessary to study the individual sections. Twenty-sixth Street and Third Street reefs.— The first mentioned of these is located in the channel of East river off the foot of Twenty- sixth street, New York, at about one-third the distance to the Brooklyn shore. The Third Street reef is in mid-channel opposite the foot of Third street. From information received from Captain Jadwin, of the Corps of Engineers, United States Army, both reefs are of gneiss. East River bridge number 2 (Williamsburg bridge).—This bridge, now approaching com- pletion, connects Delancy street, Manhattan, with the space between the South Fifth Street and South Sixth Street piers, Queens. For the New York tower, anchorage, and ap- proach, and for the Williamsburg tower, a large number of drill holes were sunk, those under the towers and anchorages with core ‘drills. The rock, so far as encountered, is everywhere gneiss and the surface much more regular than in most other sections. Figure 17 is made from the extensive draw- ings kindly furnished by Mr H. A. La Chicotte, of the New York department of bridges. Through the courtesy of Mr R.S. Buck, the former chief engineer, the writer was permitted to examine the cores from the foundations of the towers and anchorages. East River bridge number 8.—Diamond drill borings were put down at a number of points under each of the piers and anchorages of this bridge, and carried about 10 feet into the rock in every case. Through the cour- tesy of Mr R. S. Buck, chief engineer, the cores from all one borings were opened up for hie writer’s inspection. In all cases but one the “s _ ais) : Y2O'H SSIBUD,. . if ALID MYOA MAN YINY LSVF ‘LaUay ISD Sso1soy UoYIeEG—"9T LUNAS Pua YaADAD,.": eesSu eRe Ne ‘AuvdwopH pvol[lvey puss] Suo7y pus y1OX Mon ‘vruva[Asuueg 9y} Jo sjouun, pesodo.d Jo oul, uO 4202 S1OUD Ail? GNV1SI ONOT Cin ayy pin janes ‘puns ing buyply.E ‘S Vernon Ave.. 172 w.H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND : JOVHOHINY \N NATHOOUE us 4suld + YIMOL WEN NA THOOU Mear Hight Water. _ YIMOL WHLLYHNEW LE LSv2 LS NOSAHOHL : TY Pepe geen entre ZOVHOKINK SISOS NVLLVHNEN SSOXRSXRRWS peace ee ~ AS NIINGW £5 #24309 ZS SIMIT L$ NONNWD Je OE YS viawni2 Fie} £5 AHATHS V0 PB Figure 17.—Section across Hast River. Along line of East River bridge number 2 (Williamsburg bridge). rock is of the granitic type of gneiss, which contains frequently small gar- nets. The exception is from hole number 2 near the center line, outer side of the New York tower, the rock from this core being a beautiful white dolomite. Figure 18 is a profile of the river along the line of this bridge, which shows the lo- cations of both towers and anchor- ages. Figure 19 shows the location of drill hole number 2 under the outer edge of the New York tower. The position of this drill hole between eneiss in holes on either side of it at the corners of the tower makes it likely that this occurrence of limestone is infaulted between walls of gneiss—mortised into gneiss—as was the case in Vosburg hill, Toms hill, and elsewhere in the Sheffield valley, or has been pinched sharply in a fold. The depth of the rock surface below mean high water is, under the New York anchorage, from 72 to 84 feet, under the New York tower from 108 to 183 feet, under the Brook- lyn tower 90 to 97 feet, and under the Brooklyn anchorage 68 to 74 feet. Brooklyn bridge—The New York and Brooklyn piers were carried down to the bed rock. In the re- port of the chief engineer the follow- ing statements are made in refer- ence to the foundations beneath the New York tower: ‘*The projecting peaksof bed rock which already made their appearance at 75 feet were blasted down for some distance ee ne a oe ee one rT] _ ra DETAILED STUDY OF SECTIONS 173 under the shoe. . . . The rock is the ordinary gneiss found on Manhattan island with a dip almost vertical.’ * Russell states, on the authority of information furnished by the depart- ment of docks, that the borings for foundations of the New York tower, made September 10 to 23, 1877, were carried to a depth of 107.6 feet, BROOKLYN: Figure 18.—Section across East River. Along line of East River bridge number 8. and penetrated mica schist to a depth of 6.4 feet. He quotes Mr F. Col- lingwood, chief engineer, as follows: “The rock on which the pier foundations rest is eneiss, with a very irregular surface; in a place 172 by 100 feet the depth below high tide varied from 75 to 94 feet. The caisson was stopped at 78 feet.” Bulfhead Line IN N Limestone SI i x “4 . | loGn t! Bad ae = oGr. +108 ue oGnelss | ° 50' 100° 200' Figure 19.—Location of Drill Holes under New York Tower of East River Bridge Number 3. At the Brooklyn tower mica schist was found at a depth of 88 feet.T Coenties reef—This reef of gneiss is located in the East River channel about one-third of the distance from the Battery to Joralemon street, in * Reports of Executive Committee, Chief Engineer, and General Superintendent of the New York Bridge Company, Brooklyn, 1872, p. 26. + Russell, quoting F. Collingwood, loc. cit. 174 w. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND NHATTAN Cc MA GHEIES Gne/ss Figure 20.—Section across East River. Along line of New York-Brooklyn tunnel (Rapid Transit tunnel). Brooklyn. Itis therefore quite near the line of the Whitehall Street tunnel, now being constructed. Rapid Transit tunnel junder East river.— The course of this tunnel is from the foot of Whitehall street, Manhattan, to the foot of Joralemon street, Brooklyn. The data from the preliminary wash and core drill borings for this tunnel are set forth in fig- ure 20, which is reproduced from supple- mentary drawing 2-D-24 of the Rapid Transit Railroad commissioners. On au- thority of Mr George 8S. Rice, deputy chief engineer for the commissioners, the rock of all the cores is gneiss, no limestone being anywhere found in the section. The section is nearly complete across the river, there being, however, two deep channels separated by a sharp reef about a quarter of a mile from the Brooklyn shore, the bottom of which was not found by the drills at the depth of about 100 feet. The quite remark- able and abrupt changes of level shown by the surface of bed rock is in all respects of the type revealed by the borings‘of the same commission along the line of Broadway be- tween the Battery and Union square. Deep well on north shore of Governors island.— This well entered the rock at a depth of 69 feet 10 inches. It had, when last reported to the writer, been carried to a depth a little in excess of 1,800 feet, and had penetrated a micaceous type of gneiss for the entire dis- tance.* The same type of rock is said to occur on the south shore of the island. Diamond reef.—This obstruction to naviga- tion, which is now being removed,is located in mid-channela little north of Governors island in the direction of Coenties reef. Like the latter,it consists of the common type of gneiss. * Information furnished by John Sampson, chief quartermaster, Governors island, New York. — a = 175 STUDY OF SECTIONS DETAILED Figure 21.—Rock Surface beneath the ‘‘Jersey Flats.” From borings made by the Corps of Engineers, United States Army. Lillis [sland XXIV—But. Geon. Suc, Am., Vou. 16, 104 176 w. H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND Reef off Battery.—A small reef has been found recently in the North river about a quarter of a mile west-southwest from pier A, with least depth of 28 feet-at mean low water. The rock is gneiss, as are all the other reefs in the channels about New York island. Jersey flats—The area of the Jersey flats to the west and southwest of Ellis and Bedloes islands has been investigated through the medium of borings made under the direction of the Corps of Engineers, United States Army,* to determine the practicability of constructing a new chan- nel in that vicinity. The rock contours and the depths to rock measured from mean low water are given on figure 21. The rock encountered by the drills was in all cases gneiss “ nearly vertically stratified.” For the data entered on the map of the New Jersey flats I am indebted to Lieutenant Colonel C. W. Raymond, of the Corps of Engineers, United States Army. Data are not available concerning the depth of bed rock beneath Ellis island. For the foundation work of the new Ellis Island - hospital penetration tests were made, using a #-inch iron rod. On the south side of Ellis island piles were driven into hard bottom at depths ranging from 16 to 23 feet at mean low water. No rock was, however, encountered. Buttermilk channel has been dredged to a clear depth of 26 feet below mean low water without uncovering rock in place. Hudson River tunnel from Jersey City to New York (McAdoo tunnel).— The section of this tunnel (figure 22) published by Spielmann and Brush f shows that a reef of rock with a steep western and a gradual eastern slope rises near the east bank of the river along the line of the tunnel. The highest point of this reef, along the line of the tunnel, was at a depth of 89 feet below mean high water. The tunnel was first aban- doned because of the difficulty of passing from the air locks to the rock, but the enterprise was again taken up, and has since been successfully carried through to completion as the McAdoo tunnel, of which Mr Charles M. Jacobs has been the chief engineer (see figure 22). Projected Hudson River tunnels of the Pennsylvania Railroad Company.— These tunnels, six in number, are to run from Shippen street, Wee-. hawken, to West Thirty-fourth street, New York. Core-drill borings have been made upon the Weehawken shore, extending out to a point 700 feet from the Weehawken shore, from which wash borings were taken at intervals of 500 feet or less to the Manhattan bulkhead line, from which latter point wash and core drills were put down across the * Survey of a point between Ellis island and the docks of the New Jersey Central railroad toa point between Reef light and Constable hook in waters of New York bay, New Jersey. Report of Major G. L. Gillespie, Corps of Engineers, United States Army, Appendix E18. Annual report of Chief of Engineers for 1882, pp. 719-724. +Arthur Spielmann and Charles B. Brush: The Hudson River tunnel. Trans. Am. Soc. Civ. Eng., vol. ix, 1880, pp. 259-277, plate viii. Gs 2s, a ae ls 177 DETAILED STUDY OF SECTIONS om 30° JERSEY all e/isage Ave CITY HUDSON RIVER NEW YORK Be ee ES a he Figure 22.—Section across Hudson River. On line of Jersey City-New York tunnel. line ” 2 thAve- N “ 3 I a Beippead Line Prerhead line HUDSON RIVER NEW JERSEY e nee pte : ~~ airtand Rock (Undetermined) Figure 23.—Section across Hudson River. On line of proposed tunnels of the Pennsylvania Railroad Company. 178 W.H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND “aavasll Foo island to connect with the East 18 River section (see page 170). qs = The results obtained from the 4 S8Sa9 s borings in the Hudson River Utaass 2 section are displayed in figure | “ 23, which is reduced from a drawing furnished the writer by Mr Charles M. Jacobs, chief engineer of the North River sec- tion. Onthe Weehawkenshore the core drills were driven to different depths, the maxi- mum depth of 237.8 feet: be- ing reached in the hole sunk 700 feet out from the shore. . In all of these borings only Newark formations were en- countered, and in all save one only red and white varieties of sandstone. In drill hole num- Frock Or Bou/sder. d & Ss ————— == HUDSON RIVER Mean High Water ug a ie ber 18, which is nearest the Jf MSS foot of the Palisades, the sec- ae &8 tion penetrated by the drill ane DRA showed baked shale between : a ARS the depths of 20 and 68 feet, | and below that point, to a depth of about 80 feet, Newark basalt. The sandstone slopes | away from the shore towards the bluffs along a low angle. On the Manhattan shore the nature of the slopes toward the channel indicate that the gneiss continues to the bot- tom, though the borings are all wash borings. | Proposed New York and New Jersey bridge.—This project con- | templated at first a cantilever E33 ee ee ©~—SObridgeand later a stiffened sus- pension bridge across Hudson river at such place between Fifty-ninth and Sixty-ninth streets as should be approved by the Secretary of War. BT Mog AbdA SS Figure 24.—Section across Hudson River Opposite Fifty-ninth Street. Bulls Ferry Road ° ° n CONCLUSIONS CONCERNING HARLEM RIVER 179 A line of soundings to rock was made across the river, starting from a point between Fifty-ninth and Sixtieth streets, with results which are set forth in figure 24.* Mr Charles Macdonald, who made the borings for this section, informed the writer that the apparatus used was not sufficiently accurate to determine the profile all the way across the river. The figure, which discloses the results of wash borings, records ‘ rock or boulder” at the bottom, inasmuch as the drill did not enter the obstruction. Mr Macdonald has furnished the additional information to that afforded by the section, that at a point 2,000 feet east of the bulkhead line on the New Jersey shore, or near the middle of the river, rock was found at a depth of 300 feet below mean tide. Before reach- ing the rock the boring tool passed through 240 feet of silt and sand. By reason of the very meager information concerning the bottom of the ‘Hudson river, this section is of very considerable interest. CONCLUSIONS RESPECTING THE ORIGIN OF THE CHANNELS HARLEM RIVER From the above it appears that little correspondence between the di- rections of belts of limestone or dolomite and of the New York water front can be established, except within the stretch between Kings bridge and McCombs Dam bridge, where the observed facts point to the occur- rence of a narrow strip of limestone dropped down between nearly ver- tical faults. In other portions of its course the Harlem river flows over limestone or gneiss, but in all instances in a direction transverse to the strike and to the probable longer axes of the rock areas. The two reefs of gneiss over which the stream flows are located at McCombs dam and between Third and Fourth avenues. At the first-mentioned locality the backbone of the reef lies at a very moderate depth below the surface, from which in both directions the slope plunges away to a very consid- ' erable depth; as shown by the fact that the piles beneath the bridge of the New York Central and Hudson River Railroad Company over Crom- wells creek were driven to a depth of 120 feet without meeting rock. The Harlem River sections, which are furnished by the numerous bridges across it, show clearly that it is not a simple erosion valley result- ing from the cutting of the stream. In the north-south and north-north- west south-southeast stretches of the river the rock banks are generally not in evidence, and even when they are they do not always correspond in position with the present banks. The bed of the stream appears to be not a uniform decline in a single direction, or made up of slopes in * Specifications for a suspension bridge over the Hudson river at New York. Engineering News, vol. xxxiii, March 7, 1895, pp. 159-160. (Figure.) 180 w.H. HOBBS—CHANNELS SURROUNDING MANHATTAN ISLAND two directions from an intermediate point, but the floor is marked by sudden changes of level, particularly, however, where the reefs of gneiss cross it at McCombs dam and at Third avenue, thus connecting ridges of gneiss on the north with their extensions on the south. EAST RIVER Under East river limestone has been found at but three localities— under the eastern channel at Blackwells island, in the western channel on the Forty-second Street section, and in one of the drill holes beneath the Manhattan pier of East River bridge number 3.* The limestone east of Blackwells island is probably enclosed between parallel fault walls, and would appear to have been dropped down along them. These fault walls are, however, not parallel to the gorge of the river, and the lime- stone belt is, with much probability, soon cut off by faults which follow the direction of the gorge.t The multiplied observations of gneiss, and. gneiss only, beneath bridge piers and in tunnels in many sections of East river, the numerous reefs of the same rock in midchannel, and, more than all, the nearly complete section of gneiss across the river at the Battery, at Forty-second street, and the nearly complete section at Hell Gate— these multiplied observations leave little room for doubt that the rock bed of this river is mainly of the harder rock (see plate 35). HUDSON RIVER Regarding the bed rock beneath North river, comparatively little is known. The great depth of this river in this vicinity and its thick floor of drift and silt make it unlikely that very much will be learned about its rock bed in the near future. There is no reason to suppose that lime- stone may not underlie certain portions of it,though there is little reason to assume that it does. The rock of the eastern shore is largely gneiss and that of the western shore to the southward of the Palisade border (south of the principal bend at Weehawken) is also either gneiss or serpentine (see plate 35). The origin of the North River channel is sufficiently accounted for, however, by its position along the contact of the Newark beds with the crystallines. That this border is a fault border seems to be abundantly proven, not only by its markedly rectilinear extension, by the great scarp of basalt, and by the inferior position of the newer terrane as revealed by surface development,t but especially by the new borings along the line of the proposed tunnels of the Penn- sylvania Railroad Company (see figure 22 and page 176). * Limestone is mentioned either in ledges or blocks at Corlears hook by some of the earlier writers, but from recent borings it seems certain that boulders are referred to. 7 See ante, p. 167. Cf. Bull. Geol. Soc. Am., vol. 13, 1902, p. 143. FORMER HYDROGRAPHY OF MANHATTAN ISLAND 18] As to the origin of the channels of Spuyten Duyvil creek and of the Harlem and East rivers, in the opinion of the writer their directions have been largely determined by lines of jointing and displacement. Regard- ing the Spuyten Duyvil stretch, however, the facts are meager, and the problem is not free from doubtful indications. _ FoRMER HypROGRAPHY OF MANHATTAN ISLAND Owing to the early importance of Manhattan island and to the fact that the gridiron of streets and avenues was laid out by surveys before * any considerable change in topography had been wrought by man, we are in possession of unusually high-grade maps for the period in which they weremade, Randall’s map of the island was published in sections ‘between the years 1811 and 1891. This map, which is now preserved in the office of the commissioner of public works, comprises four volumes of 92 sheetseach. The individual maps are 25 by 37 inches, and are on a scale of 100 feet to the inch. Randall gives the precise location of all the old farms in their relation to water-courses and topography. General Viele’s map, published in 1874,* is on a scale of 1,000 feet to the inch, and is based on Randall’s map. It shows the original shoreline of the island, the made land, the drainage system, the topography, and the location of each rock exposure south of Manhattanville, all superim- posed on the gridiron of streets and avenues. The accuracy of this map has been abundantly attested by engineers, real-estate men, and others who habitually use it, and, so far as exposures of rocks are con- cerned, it has been attested by the writer, both by comparison with the early reports of geologists on Manhattan island and by examination in the field. The made land and the hydrography have been reproduced in plate 35. It will be noted in how largea degree there is accord between the drainage directions and the directions of streets and avenues. As an indication that this orientation of the drainage has been to a large extent determined by planes of fracture, it is interesting to consult the recent map by Julien, showing the location and orientation of the principal dikes on the island. These dikes quite generally run along the avenues (see plate 35). Julien has shown also that cross-fracturing is a common feature of the rocks of the Manhattan uplands, and has given instances of definite cross-faults directed nearly at right angles to the avenues or along the cross-streets. Thus the fissure planes which become occupied by the dikes and the perpendicular series described by Julien (often occupied by quartz lenses and pegmatite) correspond very closely in direction with the two series making up the main drainage * Topographical atlas of the city of New York, by Egbert L. Viele. system. The writer’s observations show that me nent joint planes in the rocks of the island have the cross-fissures—north 60 degrees west.* ( may ‘The role of the dolomite in fixing the locations of th nels would thus appear to have been a subordinate o far as the direction of its boundaries has ae determi planes. vey . * Of, Bull, Geol. Soe. Am., vol. 15, 1904, p. 556. \ BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 183-204 APRIL 18, 1905 SOME CRYSTALLINE ROCKS OF THE SAN GABRIEL MOUN- TAINS, CALIFORNIA * BY RALPH ARNOLD AND A. M. STRONG (Presented before the Society December 30, 1904) CONTENTS Page Rar fe Te ee LD eb wivid.si ams Mnis{aloveyhls a mw ef ere oe 183 Peer Lae oan Gabriel MmOUNtAUES. 6. foi... cc cco se eae nerecrcersees 184 NN eee se Bs oc ce aye Acetuers DE ne ee ee 185 PeeeMOe GeTALUNe ©). 6c ee es ce ee BR ES oD ye PAT a OEE ON eg Oy 2 i 187 Age of the San Gabriel mountains........... ee ION ay I SES eee 188 General character of the rocks....... Brepe aites se ae Siac pe oie Se kote 188 Detailed petrography.................06. Sees he Ga a hah aes bpp en a es, Races 189 SNMIENRROMN oe Ft akg hw aks. be os aD NS eee fe ee wins Chg iek we 189 PME PIG TACLOTIMEVOR «oop aieai ccs sine o's sn clea hin ciwed « whens Reon ages 8 189 0 ETE Sa OEE SS sn a 9 ae a 189 au MSINTNDAIIR EEN ec) Moi ant a ea oP Mate geil «5 vc weaide we Be rad 191 oe ee PEAR ee gree ene et sate 2 oka dls 191 IP RIRBE ECP era a et ee LT re ie ac play «a eae b 197 IRE ERE ee Ses Aah Se ne ee ee he odd wees 198 Occurrence limited ......... LN oer Vi gens he ek ae eg ee BET, ee 198 a BERN So a lps Se RRR A eA dhe A UP MINMREINMNE et Se aR ans it) OA i ie ween ace elk sere iye'ae mio 198 Quartz-hornblende-porphyrite........... Bee ae aaNet. tte Saale eae 199 SEES ONENESS. ow a woe Rage selene iehd ape) wai ome Pid cease 200 Metamorphic rocks...... Hoe “Gul Bek ae ERTS A Se Se aS Dugeeeegee othe 200 UT A se a ee ref in Kp Seay Cena 9 En, Rana 200 REID EMIC-IOFILC-SNEIOR . . «is os sgt soca Ae eals wk vee o claasceendeen 200 IGE PTANIEG-SOHCIRB.. 2... . 22sec de sane SOR ors ae 5d. ania our oe eS et Ma ee ee AR cae OL) 203 RET aE EPC pMAERS BRTSPLS 268 = 05 012,79 ton Oe ee ae er 203 0 LS nea LG erry ne 203 INTRODUCTION The following paper embodies the results of some studies which were begun by the writers in 1897, while students in the department of geology * Published by permission of the Director of the U. 8. Geological Survey. XXV—BuLu. Geox. Soc. Am., Von. 16, 1904 (183) 184 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL -at Leland Stanford Junior University, and which have been continued intermittently ever since. The analyses were made by the junior author in the laboratories of the university. The writers are indebted to Dr James Perrin Smith, of Stanford University, and to Dr Waldemar Lind- _gren, of the United States Geological Survey, for helpful suggestions relating to the preparation of the paper. LOCATION OF THE SAN GABRIEL MOUNTAINS The San Gabriel mountains are a group of more or less intimately associated ridges and peaks occupying the territory between Cajon pass, V, WY Y y (ae eS GIN INN = p= T3ete SIU MNNTE SSS Mes Ze AW RPG Ni, thy anta Tnes Prertiiy Ui, ty\' my na S\UYy BH 4 // ee = a \ 1 ggg anliego PLUG, 2 \ shins ke .--— ee Te me Fieure 1.—Sketch Map of Southern California. Showing relation of the San Gabriel to the other mountain chains. in San Bernardino county, and the Santa Clara river, in Los Angeles county, their greater portion lying inthe latter. They comprise an area of about 50 by 25 miles, nearly all of which lies within the confines of the San Gabriel timber reserve. The chain is exceptional in that its axis lies oblique to (and in some instances nearly perpendicular to) those of most of the other prominent ranges of California. What genetic rela- tion exists betweer. this group and the Sierra Nevada, Mount Diablo, and other ranges north of it is yet to be determined, but much light will LOCATION AND TOPOGRAPHY 185 doubtless be thrown on the subject by a study of the region at the southern end of the San Joaquin valley, where the axes of several of the ranges appear to focus. ToPOGRAPHY The San Gabriel mountains are divided into two more or less distinct ranges by streams which flow in easterly or westerly directions approxi- mately parallel to the length of the chain. The most prominent of these streams is the San Gabriel river, which cuts nearly through the center of the group from south to north, and whose east and west branches separate the major portion of the southern or Sierra Madre range from the rest of the mountain area. The Middle Fork of Lytle creek, immediately west of Cajon pass, and Tujunga canyon, north of La Canada, nearly complete the separation of this southern border range from the northern mass. The Sierra Madre range is long and narrow, averaging only about 6 milesin width. Its sides are precipitous and are cut by short, deep, narrow canyons. Thesteep southern face of the range probably represents an ancient and bold fault scarp, which, however, has been considerably altered by erosion since its origin. Theaverage height of the Sierra Madre is something over 6,000 feet, its highest point, Cuca- monga peak, however, reaching an elevation of 8,911 feet. Great detrital fans, varying in area and depth with the size of their parent streams, stretch out from the southern base of the Sierra Madre over the San Gabriel valley, which borders the mountains on the south. North of the Sierra Madre is an elongated and complex range of ridges and peaks which cul- minates near the eastern end of the group in San Antonio peak (“ Old Baldy ”’), elevation 10,080 feet. The slopes in this northern area are more gentle and the relief less sharp than in the southern, the outlying hills of the former grading off into the buried valleys of the Mojave desert, which bounds the San Gabriel mountains on the north. The Verdugo mountains and the San Rafael hills are isolated portions of the San Gabriel mountains lying south of and separated from the latter by La Canada, a valley averaging nearly 2 miles in width, which runs for several miles parallel to the western end of the Sierra Madre » range. At the eastern end of the San Rafael hills and La Canada lies Pasadena, while 9 miles southwest of the latter is Los Angeles, the chief city of southern California. The rocks described in this paper were collected for the most part during reconnaissance trips over the following territory: Mount Lowe railroad to summit of mount Lowe, Rubio canyon, Eaton canyon, Big Santa Anita canyon, Mount Wilson toll-road, Sierra Madre-Mount Wilson CRYSTALLINE ROCKS OF SAN GABRIEL 186 ARNOLD AND STRONG race Tee Sia ijva “hos” ‘ See wee Ory 0188490 YAON ore be PL ,Pfee! Hd OMpopuy ui VINYOS UR WYIHLNOS ave £37AuNeG WoIIDOTKZD BN mony ——_—_—_—— s2i'w 31v9S “Sap pPreae Dues qo Rewpunog ayeinrooddy Syrway 0N3931 i ae eens Winosto ; toeen . > voan meds) MOT ; Opurusdg UBS TOPOGRAPHY 187 . trail, Sturtevants Camp trail, Mount Wilson-West Fork-Pine Flats trail, ~ Chillao canyon, mount Waterman, Alder creek, upper Tujunga canyon, and, in the region west of Pasadena, Devils gate and the southern por- tion of the San Rafael hills. As no topographic maps of the region were available at the time of the inauguration of the work, no effort was made to map the different rock areas, and as a consequence not as much information concerning the relations existing between the different types was obtained as would have been the case had the work included detailed field differentiation and areal mapping. PREvious LITERATURE For an interesting mountainous region of over 1,200 square miles in extent, located in as important a section of country as that in the vicinity of Los Angeles, the San Gabriel chain has certainly received very little notice from geologists. Doctor Trask, first state geologist of California, in 1855, referring to the geology of the San Bernardino mountains (in which he includes the San Gabriel and all other ranges from the San Jacinto to the Santa Inez), says: * ““These mountains are made up for the most part of the primitive rocks, and consist chiefly of the granitic series. They form by far the most of all the higher ridges and more elevated peaks belonging to the chain.”’ ' In 1857 Blake f in his Pacific Railroad Survey report mentions com- pact and gneissose granite, talcose slates traversed by quartz veins, and trappean rock as occurring in the Cajon pass. Slate, hornblende rock, and gneiss were found by Whitney,{ state geologist from 1860 to 1874, as float in the mouth of the San Gabriel canyon. In 1900 Claypole§ presented a paper on the “Sierra Madre near Pasadena ” before the Cordilleran Section of the Geological Society of America, in which he described that range in a very general way, dwell- ing mostly on its relation to the region south of it. Hershey || in 1902 mapped the western end of the San Gabriel moun- tains as*“ plutonic” and the eastern end as “ granite.” He also men- *J. B. Trask: Report on the geology of the Coast ranges. Cal. Sen. Doc., no. 14, 1855, p. 20. ~ Wm. P. Blake: Pac. R. R. Report, vol. v, 1857, p. 88. tJ. D. Whitney: Geol. Survey of California. Geol., vol. i, 1865, p. 172. ¢ E. W. Claypole: Bull. Geol. Soc. Am., vol. 12, 1901, p. 494, | O. H. Hershey: The Quaternary of southern California. Bull. Dept. Geol. Univ. of California, vol. iii, 1902, no. 1. 188 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL tions the probable formation of a great fault along the northern front of the chain at about the opening of the Quaternary era. AGE OF THE SAN GABRIEL MOUNTAINS There has been great divergence of opinion regarding the age of the: San Gabriel mountains and their correlation with the other prominent ranges of California. Trask considered the granitic rocks of the chain as primitive, while Whitney placed their elevation as post-Cretaceous. Fairbanks, than whom no one is better qualified to speak, has this to say regarding the age of the plutonics and metamorphics of southern California : * ‘*T believe that the great convulsion which upheaved and metamorphosed the older rocks and intruded granite into them took place as it did in central and northern California between the Cretaceous and the Jurassic. There is no break - in the line of granites and crystalline schists the whole length of California.” It appears quite probable, however, in the light of some recently ob- tained evidence, that at least the greater part of the elevation of the San Gabriel mountains took place during either the late Eocene or Oligocene period. ‘Tilted strata of sandstone and shale of lower Eocene age, found at an elevation of over 5,000 feet in the vicinity of Rock creek, on the northern face of the range, show that the chain has been elevated at least 5,000 feet since the deposition of the lower Eocene.f It has also recently been discovered that the conformable series of conglomerates, sandstones, and shales which flank the San Rafael hills on the south and underlie the southern portion of the city of Pasadena are of Mio- cene age.{ As the conglomerates of this formation (for which the name Pasadena is here proposed) rest on and are composed of the San Gabriel plutonics and metamorphics, it is evident that the chain is certainly pre- Miocene, although quite a little elevation has taken place since the depo- sition of this Miocene formation. GENERAL CHARACTER OF THE Rocks The following rocks have been found by the writers in the San Gabriel mountains and are described in this paper: Biotite-granite, quartz-mon- *H. W. Fairbanks: Geology of San Diego county; also a portion of Orange and San Bernardino counties. Eleventh Ann. Report of the California State Mineralogist, 1903, p. 119. + Fossils from Rock creek recently sent by Dr W. C. Mendenhall to Doctor Dall for determina- tion proved to be of lower Eocene (probably Martinez) age. ¢ The senior author and his father, Delos Arnold, have recently obtained a good series of fossils from Raymond hill, Pasadena, which proves the age of this formation ta be either lower or middle Miocene. . <= Fag, ~~ = “« ‘ GENERAL CHARACTER 189 zonite, granodiorite, hornblendite, aplite, micropegmatite, quartz-horn- blende-porphyrite, diabase-porphyry, hornblende-diorite-gneiss, biotite- granite-gneiss, hornblende-schist, and garnetiferous schist. The Sierra Madre range consists essentially of granodiorites and gneisses, with more acid areas in which the country rock is quartz- monzonite. Large dikes or included masses of hornblendite are present at several localities, notably on the south slopes of mount Lowe, while at other places smaller dikes of quartz-hornblende-porphyrite and diabase porphyry cut the country rock. Aplite dikes and quartz veins are of common occurrence, some of the latter yielding traces of gold and silver. Garnet-bearing and hornblende schists are also found in the southern range. The character of the rocks of the mouptain area north of the Sierra _ Madre is considerably different from that of the latter. True biotite- granite and rather coarse grained granodiorite, decidedly different in physical appearance from that of the south range, are found in the northern mass. Aplite and micropegmatite are also found in the latter region. Taking the San Gabriel mountains as a unit, therefore, it is found that the southern border is composed principally of fine grained granodiorites and gneisses, while the central portion (the extreme northern border of the mountains was not examined by the writers) is composed of rela- tively coarser grained rocks, which are probably somewhat more acid in average composition than the border range types. DETAILED PETROGRAPHY PLUTONIC ROCKS General characteristics—The plutonic rocks found include biotite-gran- ite, quartz-monzonite, granodiorite, and hornblendite. With the excep- tion of the latter, all of the rocks of this class are gray colored and fine grained, and all, without exception, are prone to fall an easy prey to the destructive forces of weathering. The outcrops as a rule show rounded outlines, and in many instances huge detached boulders, the products of weathering, are found on the nude surfaces of the country rock. On account of the susceptibility of these plutonics to the weatherwing process, fresh specimens were hard to obtain, and the determination of the rock, even when examined in thin sections, was often more or less uncertain. Biotite-granite.—The paucity of true granites in the territory under dis- cussion is rather remarkable when the size of the area and the general 190 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL character of its rocks is considered. Similar conditions, however, prevail in large areas of the Sierra Nevada, so that possibly this is the normal condition for the ranges of this part of the continent. The most char- acteristic granite area is in the vicinity of mount Waterman, and three rocks of this type from that locality will be described. A mass of biotite-granite occurs in Chillao canyon, several miles west — of mount Waterman. Megascopically this rock (A. M. 8. number 15) is light brown, fine grained, and is distinguishable from most of the rocks of the region by the paucity of the ferromagnesian minerals. Its pecu- liar rusty color is caused by the weathering of small amounts of the biotite. Although the rock is unusually fresh, thin sections show it to be much crushed. It is hypidiomorphic granular in structure, and con- sists essentially of orthoclase and quartz, although well developed crystals of plagioclase are not uncommon. Some of the orthoclase crystals are much broken and show weathering slightly along the joint planes. Biotite is found sparingly throughout the rock and a little muscovite is also present as a secondary product in some of the orthoclases. Biotite-granite, much richer in biotite than the one just described, is found at Fern camp, in Buckhorn canyon, near mount Waterman. The rock at this camp appears to be finer grained and less weathered than in most other places along the canyon, but in other respects is quite similar to the common facies. Hand specimens of this granite (A. M. S. num- ber 19) are dark gray in appearance and show feldspar, quartz, and much biotite. Microscopically the rock is interesting on account of the micro- cline which is characteristically developed init. Intergrowths of this - mineral and quartz were noted. The orthoclase and microcline together are in excess of the plagioclase, although there is nearly as much plagio- clase as orthoclase alone. One crystal of plagioclase shows every alternate albite twin completely kaolinized,while the intervening ones are unaltered. Epidote is the principal weathering product of the feldspars. The biotite, which is present in considerable quantities, shows brown to greenish pleochroism. Magnetite also occurs sparingly. Another large mass of granite is found on the northeastern face of mount Waterman. This rock (A. M.S. number 22) is medium grained, gray, and shows feldspar, quartz, biotite, and hornblende. In thin sec- tions the rock is seen to be hypidiomorphic granular. Orthoclase, apparently the dominant mineral, occurs in prominent xenomorphic crystals, through which pass parallel microscopic veinlets of kaolin and possibly muscovite. Zonal structure is common, the weathering follow- ing the zonal lines and producing kaolin, muscovite, and epidote. Occa- sional crystals of microcline are also present. The plagioclases are next PETROGRAPHY OF PLUTONIC ROCKS 191 in abundance to the orthoclase, and were determined to be oligoclase in most instances. Quartz is abundant as small grains. Some brown biotite is present, but is often partially weathered into chlorite, epidote, or muscovite. Green hornblende in spindles and a small number of magnetite grains complete the composition of the rock. Quartz-monzonite—Among the specimens of plutonic rocks collected in the Sierra Madre range are three which appear to be quartz-mon- zonites. Although these specimens are considerably weathered and the determination of their feldspars is not as satisfactory as one would wish it to be, still the bulk of the evidence favors their classification as above. One specimen (A. M. 8. number 7) comes from the Henniger flats— Eaton canyon trail, about half a mile above the canyon. It is fine grained, dark gray in color, and shows feldspar, quartz, hornblende, and . biotite. In thin sections the structure is seen to be hypidiomorphic granular, and the orthoclase appears to be slightly in excess of the plagi- oclase. Zonal structure and zones of minute inclusions occur in some of the orthoclase crystals. A few of the plagioclases show pericline twin- ning. Brown biotite, a little green hornblende, and numerous small particles of magnetite are the other constituents. This rock is charac- terized by the nearly equal amounts of orthoclase and plagioclase and by the xenomorphic occurrence of its minerals, very few crystal faces being developed except in the plagioclase feldspars. A. M.S. number 30, from the flanks of mount Lowe, below Alpine tavern, is another of the quartz-monzonites. It is coarser grained than A. M.S. number 7, and contains some apatite. The plagioclase, which is slightly less abundant than the orthoclase, appears to be about equally divided between andesine and oligoclase. Specimens of this rock were compared with a granodiorite from lake Tahoe, Sierra Nevada mountains, and found to be similar in general appearance, although the Tahoe rock was somewhat coarser grained. A microscopic comparison of the two showed the Tahoe specimen to contain proportionately less biotite than A. M.S. number 30, and also showed the former to contain much more plagioclase than orthoclase. A quartz-monzonite from the south side of mount Lowe near the sum- mit (A. M.S.number 31) is similar to number 30, except that it contains no hornblende, less quartz, and a little microcline and muscovite. The orthoclases in number 31 are much jointed and the cracks filled with a dark colored opaque substance. Granodiorite.—Granodiorite is by far the commonest rock in the San Gabriel mountains, at least in the region north of Pasadena. It and its associated gneisses constitute most of the Sierra Madre and are also found XXVI—BuLt. Geox. Soc. Am., Vou. 16, 1904 192 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL at many other localities throughout the mountain area under discussion. The rock consists of quartz, plagioclase (either oligoclase or andesine, or both), orthoclase, hornblende, and biotite, with titanite, zircon, magnet- ite, and apatite as accessories. The quartz is present in relatively small quantities, so small, in fact, as often to place these particular cases very near the true diorites. The plagioclase is always in excess of the ortho- clase, but in some areas gradations toward quartz-monzonite show almost as much orthoclase as plagioclase. Hornblende is the principal ferro- magnesian mineral, although biotite is often present in considerable quantities. The character of the rock varies from medium to very fine grained. In color the granodiorites range from light to dark gray, de- pending on the amount of the ferromagnesian minerals present. In some localities the feldspars are pink and give the rock a reddish cast when viewed from a distance: This latter condition is particularly noticeable in the granodiorites exposed along the line of the Mount Lowe railroad, in Grand canyon, and in the west wall of Eaton canyon. ‘Two facies of the granodiorite are recognizable in these mountains, a fine, almost uni- formly grained type being characteristic of the Sierre Madre or southern | border range, while a somewhat coarser grained form containing large erystals of orthoclase occurs in the central mountain mass. A typical example of the first class is the granodiorite in the vicinity of Strains camp, on the northern slope of mount Wilson. On the surface this rock is usually badly weathered and decomposed, but in some of the stream beds it is possible to find comparatively fresh exposures. The specimen here described (A. M.S. number 10) is from the lower spring near thecamp. Inthe hand specimen the rock is medium grained and rather dark gray in color, and shows feldspar, quartz, a consider- able amount of hornblende, and a few flakes of biotite. Some of the hornblende is altered to chlorite. Examined in thin sections, the plagio- clases were seen to be in excess of the orthoclase, and, though consider- ably weathered to kaolin and occasionally to epidote, were determined to be mostly oligoclase. The orthoclase occurs in xenomorphic grains, which are somewhat larger than those of the plagioclase, and often show zonal structure and occasionally inclusions. Muscovite and kaolin are its common alteration products. The hornblende is abundant in auto- morphic crystals, which show dark green to brown pleochroism. Quartz occurs only sparingly in small grains. Fibrous aggregates of colorless zeolites, microscopic veins and small grains of epidote, and occasional particles of magnetite are also present in the rock. Titanite was noticed in a section of a similar granodiorite (R. A. number 7), which was found near A. M.S. number 10, but was not present in any of the thin sections of the latter. PETROGRAPHY OF PLUTONIC ROCKS 193 A chemical analysis of A. M.S. number 10 follows: Analysis of granodiorite (A. M. 8S. number 10) from mount Wilson, A. M. Strong, analyst “0 Se ESR ae ae a ae Ne ERTS A Near 61.38 TE hes ck? ress uae pea aria ag 14.33 ae OC Be ao wae ne set vt ve eee Aa eee 7.64 * MEM RTS yo cn ce alain t's Aa cio ree eG areuatetrene ae 102 * TOR Pa te P at) wrk th alert walolye eva ewes etd wate Trace Ee See, rere te tiny: w n't a Sonik eh tees | Soe che hare 5.42 TDS e tear ate aie tbh os cicihes Gin waite es SA sero bewios 4S 2.98 PR ee ret oo Aah s ant eiat Sad scree & oot revo 2.58 aR ees ass RAIS Gg een Nay” wv a lh Syn 4.71 ERD a aah eR eR APLC LS rand alely ais Saal eGNN Wien dab 0.13 100.19 Granodiorites of the same general type as A. M.S. number 10, but still having minor differences, occur at several places within the territory under discussion. One of these (R. A. number 7) may be taken as char- acteristic of the granodiorites of the crest of the Sierra Madre range. Megascopically this rock is very light gray, fine grained, and shows the clear glassy quartz and the glossy cleavage surfaces of the feldspars. Small particles of biotite are scattered quite thickly throughout the mass, but the light colored constituents so predominate as to give it almost the appearance of aplite. A few small crystals of quartz have a greenish yellow cast, and, near the weathered surfaces, small amounts of green chlorite may be detected. When viewed microscopically the structure is seen to be hypidiomorphic granular. Plagioclase, which was deter- mined to be oligoclase and andesine in about equal quantities, is the most abundant constituent. In this mineral the faces parallel to the twinning planes are generally perfect, while the ends of the prisms are irregular in outline. Albite twinning is common, and, in one instance, twinning according to the pericline law was seen. Inclusions of mus- covite, biotite, and magnetite occur in some of the plagioclases. Ortho- clase, next in order of abundance to the plagioclase, occurs usually in irregular grains, but also occasionally in automorphic crystals. Zonal structure is common, and twinning according to the Carlsbad law was noted in several crystals. One orthoclase shows inclusions of biotite, *There was undoubtedly an error in the determination of the iron, the percentage of ferric iron being altogether too high for a rock containing as much hornblende as A. M.S. number 10. Taking the total iron as 8.66 (this also being too high, as the oxidation of the ferrous iron would add weight), the calculated percentages for each oxide would be about as follows: 194. ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL orthoclase, and plagioclase. Quartz is a somewhat less important con- stituent than the orthoclase, is present in small grains, and is free from inclusions so far as noted. Biotite occurs quite commonly in small grains, which show the characteristic fine cleavage lines and brown to green pleochroism. Muscovite is also present in small amounts, but probably only as a secondary product. A few grains of magnetite, occa- sional small green hornblendes, and some chlorite are also present. A characteristic diamond-shaped crystal of titanite occurs in one of the sections. The San Rafael hills, west of Pasadena, are largely composed of biotite- granodiorite, which in most places is much jointed and weathered. Specimens of this rock (R. A. numbers 8 and 31) appear dark gray in the hand, and are seen in thin sections to be composed principally of plagioclase, biotite, orthoclase, and quartz, named in order of relative abundance. 3 It is not unusual in the granodiorite areas of the Sierra Madre range to find portions of the rock mass in which the ferromagnesian minerals are relatively more abundant than in the typical facies. A specimen (R. A. number 24) from one of these segregations at the summit of mount Wilson shows biotite, plagioclase, hornblende, orthoclase,. quartz, and secondary chlorite, occurring in relative abundance in the order named. The minerals are nearly all xenomorphic and their characters similar to those of A. M. 8. number 10. Still another granodiorite similar to A. M.S. number 10 and R. A. number 7, but containing a large percent- age of hornblende, is found at the summit of mount Lowe. The min- erals composing this rock are, in order of relative abundance, plagioclase, hornblende, biotite, orthoclase, and quartz, with considerable quantities of chlorite and a little muscovite as secondary products. An example of the second or coarser grained facies of the granodiorite is found at Pine flats, southwest of mount Waterman. It forms a band of rock about a mile in width, which has a northwesterly trend across the divide from the West fork of the San Gabriel to Alder creek. The rock is more resistant to weathering than the adjacent plutonics and forms a terrace across the face of the mountain. Hand specimens of this granodiorite (A. M. 8. number 12) are gray in color and are charac- terized by phenocrysts of orthoclase 5 to 18 millimeters in length, which are usually twinned according to the Carlsbad law. The groundmass consists of feldspar, quartz, hornblende, and some biotite, the last some- times being altered to chlorite. Microscopically the rock is seen to be made up principally of xenomorphic crystals of feldspar, hornblende, and quartz. The plagioclases, which appear to be about equally divided PETROGRAPHY OF PLUTONIC ROCKS 195 between oligoclase and andesine, are the dominating minerals. Peri- cline twinning is common in the oligoclase. The orthoclase, as previously mentioned, occurs principally as large phenocrysts, twinned according to the Carlsbad law. Considerable quantities of quartz are present in xenomorphic grains of various sizes. The hornblende is in irregular masses, shows green to yellowish pleochroism, extinction angles of about 20 degrees, and is often altered to epidote around the periphery. Titan- ite is present in small quantities in wedge-shaped crystals. Zircon, apatite, magnetite, and hematite in scattered crystals complete the mineral composition of the rock. The following is an analysis of this granodiorite : . Analysis of granodiorite (A. M. S.number 12) from Pine flats, A. M. Strong, analyst Sere Soe mer S . S85 ier oa Ba soa Gots Pee op os 64.45 Nae ean as 2 ain Bho cla ol Wane ww sed « dip. eats sine Trace PURE EE ere a We tnt eM, Se itis exis dindoigy «dese 17.18 RIED =P SPUN) ene cE ES he eile. oa cb de we Fee 3.02" Mg hic hr eer on, shires Se ee ae Soe od a den 0.60 * MNES iON cts ah re ia iwhe a hot trek Gee a Caae 1.62 ee ae ee a eer ee OE ea eet eee 4.31 Per SR NE Mate Yas Lo) I hae HE Ss Adal Whistles Meh ace 0.75 2 A SE COR a 2 Ae a eran ee 2.98 RR ree ee See Ee. i) AEE AE NG thes Sy cate aah ad lace win: eae 4,24 I Se cree as a wna ke aw kek wh ales 0.59 De CR PE eS th stenc s aia Soni on. Slalvc nies wheat > aie ee ee Trace 100.94 The granodiorites of the San Gabriel mountains appear on the average to be finer grained and to contain less quartz, titanite, and zircon than those from the Sierra Nevada range of central California, but otherwise the rocks of this class from the two regions appear to be quite similar. The granodiorites near the center of the southern California chain are more nearly like those of. the Sierra Nevada than are the rocks of the same class from the outer or Sierra Madre range. This statement holds true not only of the grain and color of the rocks, but also of their chem- ical composition. | ; | A comparison of the analyses given in the following table will show the chemical relations of the granodiorites from the San Gabriel and Sierra Nevada mountains. *See note under previous analysis concerning error in determination of iron oxides, In this ease the Fe,0; should be about 1.31 and the FeO 2,61, 196 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL Table of analyses of granodiorites i. af, tae IV. Ne VI. Drinleser es A, NLS. Latics coe W.L., | Limits | Average aeSRAE number 12, bred 7 Sierra of composi- Wilson. Pine flats. ede Nevada. | variation. tion. Per cent Per cent SIO eat ee ete 61.38 64.45 63.43 65.54 59-69 HO, once SF ee eee as Trace 0.73 0.39 |. 2.7) ACU ee Als. 36 ocr 14.33 17.18 14.20 16.52 14-17 16 J HeOs. ee 2.50* 131%) 154 1.40 14-2} 1.50 BeO ese ee 6.16 * 2.61 * 4.56 2.49 13-44 3 MnO see Trace 1.62 0.03 0.06 |vsicseeeeeloeee eae BAO lees, ARs nes 2 eee Son ee 0:06 ]...0. dN eek ee ‘ CIO oe ees 5.42 4.31 DOL 4.88 3-63 5 NE Oas oct, Sed 2.98 0.75 2.35 22, 1-24 2 Re eee ice 2.58 2.98 ZAG 1.95 1-32 2:20 NaN oo ae 4.71 4.24 3.49 4.09 | 23-42 3.50 BACON a ee 0.13 0.59 1.65 0.59 ||... ee 1 6 Baile Wao! Sats Be mar : Trace 0.11 0.18 (Remainder 1.75) 100.19 100.04 99.85 100.61 100.00 * See notes under same analyses on previous pages. I..A. M. S. number 10, mount Wilson, San Gabriel mountains, Los Angeles county. A. M. Strong, analyst. II. A. M. S. number 12, Pine flats, San Gabriel mountains, Los Angeles county. A. M. Strong, analyst. III. H. W. T. number 17, Smartsville area, 2 miles northeast of Bangor, Sierra Nevada mountains. W. F. Hillebrand, analyst. (H. W. Turner, 17th Annual Report U. S. Geol. Survey, pt. i, 1896, p. 724.) IV. W. L., Sierra Nevada mountains, Lincoln, Placer county, Sacremento folio. W. F. Hillebrand, analyst. (W. Lindgren, American Journal of Science, vol. ix, April 1900, p. 273.) V. Limits of variation for granodiorite. (W. Lindgren, loc. cit., p. 272.) VI. Average composition of granodiorite. (W. Lindgren, ib. cit.) Comparing the analysis of A. M.S. number 10 with Lindgren’s limits of variations for granodiorite, it is seen that the Mount Wilson rock is rather low in SiO,, but high (to excess, according to the limits given in the table) in Fe,O,, FeO, MgO, and Na,O. _Disregarding the latter two, which are only slightly above the limit, it is evident that the iron shows the only great discrepancy. This is accounted for partially by error in the analysis (see previous notes), but mostly, however, by the large amount of hornblende in number 10. The hornblende also accounts for the high percentage of magnesium and soda. No titanium was dis- covered in the analysis of number 10, and no titanite was seen in its PETROGRAPHY OF PLUTONIC ROCKS 197 slides, but this mineral was present in a section from another specimen of a similar granodiorite found not far from the locality of number 10, thus showing the presence of titanium in the magma from which num- ber 10 was probably crystallized. A. M.S. number 12, from Pine flats, appears to be a normal grano- diorite, although it appears to be high in MnO and lowin MgO. It contains traces of titanium and phosphoric acid, although not in as large quantities as do most of the Sierra Nevada rocks of the same class. Hornblendite—Hornblende-gabbro is found quite abundantly through- out the range, occurring generally as large irregular dikes or masses associated with the more acid rocks. A typical occurrence is in Rubio canyon, at the foot of the famous Mount Lowe railroad incline. The rock at this locality is nearly pure hornblende. Another characteristic hornblendite is found on the slopes of mount Waterman, on the divide between the east branch of Buckhorn canyon and Bear canyon. Speci- mens (A. M.S. number 24) from this area show slight alteration, are rather fine grained and greenish hued, and consist principally of horn- blende. Plagioclase feldspar, in crystals much smaller than those of the hornblende, also occurs sparingly in the rock. In thin sections the most important constituent is seen to be the large crystals of hornblende, which show bluish green to yellowish green pleochroism. Calcite is the principal alteration product of the hornblende, and often replaces a large percentage of the original mineral in some of the crystals. Small num- bers of considerably altered plagioclases fill the interstices between the hornblende crystals, and, in some instances, form inclusions in the latter. Epidote in small grains is an occasional constituent. A chemical anal- ysis of this rock (A. M. S. number 24) from mount Waterman, by A. M. Strong, gives the following results: ee ec ie nm gard Dake So 49.68 OSA ACRES AL ET enero iam ey aes are ee 2 42. 07 ip eee ees ok ON sxe u:to.e ares as Sits Sg) vel eet erat o ees oo 10.57 * oP a leps se hee i) OR Sat I PR LE So ots A a Trace RPE Le ee ae FL 5 LIALAN ? Sel okie Oates eRe es late 13.85 Perera ate rere ae Bees gS tous Oe ite arcade tats os 9.02 Re ere aut e LIN ie 56) pence ads chal sleek © ASR wets BE hes 1.15 ee a 2 a eae a eee ieee tk 3.31 Re eS ML Ne ike o's nok Be ve oS MRO Mal dA a Us .56 100.21 A piece of water-worn hornblendite from Tujunga canyon below the mouth of Alder creek is almost black in color, and in the hand specimens shows nothing but large crystals of hornblende. In thin-sections this * FeO determined with Fe203. 198 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL rock is seen to be made up of large pseudomorphs of brown hornblende, each being composed of, numerous very small columnar crystals, all of which are oriented with their long axes parallel to the long axis of the pseudomorph. Between the more or less continuous rows of these small crystals are strings of quartz grains. Calcite occurs abundantly through- out the rock, and the quartz and calcite are- doubtless the alteration products of the mineral of. which the pseudomorph is but the skeleton. Magnetite ocuurs quite abundantly scattered throughout the mass, and an occasional zoisite is seen. Originally there were probably some feld- spars in this rock, but if so all traces of them have been lost in the pro- cess of alteration. DIKE ROCKS Occurrence limited.—With the exception of aplite, dike rocks are not of very common occurrence in the San Gabriel mountains. The basic dikes, though not the most abundant, are still the most conspicuous, for, as a rule, their dark color contrasts strongly with the lighter colored . granodiorites, gneisses, etcetera, which are intruded by the dikes. Aplite, micropegmatite, quartz-porphyrite, and diabase porphyry are the dike rocks recognized by the writers in the territory examined. Aplite.—A plite occurs abundantly, usually in small dikes cutting the granodiorites and other plutonics. It is easily recognizable by its light color, and is generally finer grained and less decomposed than the intruded rocks. ° | A typical aplite dike cuts the granodiorite about a quarter of a mile below Devils gate, northwest of Pasadena. It varies from 6 inches to 2 feet in thickness, is light gray in color, and microcrystalline in structure. In thin sections the rock (R. A. number 26) is seen to be made up largely - of quartz, which occurs in small, clear, xenomorphic chrystals, without inclusions. Next in importance is the orthoclase, which is generally found in clouded xenomorphic grains, but which occasionally occurs in crystals showing one or more well developed faces. Inclusions of quartz and orthoclase particles were noted in some of the larger orthoclase crys- tals. A small amount of plagioclase, much weathered, but showing albite twinning, is also present in the rock. This constituent appears to have crystallized earlier than the orthoclase or quartz, as its crystal faces are much better developed than those of the two latter minerals. A little biotite, showing fine cleavage and green to brown pleochroism, together with some secondary epidote and chlorite, complete the com- position of the rock. Micropegmatite.—Along the borders of the Pine Flats granodiorite area, at the head of the main branch of Pine Flats canyon, are several small PETROGRAPHY OF DIKE ROCKS 199 dikes of micropegmatite, but the fractured condition of these dikes and of the adjacent rock area made it impossible to study their relations satisfactorily in the field. Hand specimens of the rock (A. M.S. number 13) are light colored, very fine grained, and appear quite fresh. Witha pocket lens quartz, feldspar, and muscovite are distinguishable. The section shows a decided micropegmatitic structure, the intercrystalliza- tion of quartz and ortlroclase in parallel positions producing the typical graphic appearance. These pegmatitic crystals are imbedded in aggre- gates of xenomorphic grains of quartz, orthoclase, and a small amount of plagioclase, none of which show the pegmatitic structure. Small amounts of muscovite and biotite, with occasional grains of magnetite, are also found in the groundmass. The biotite is generally altered to a brown, non-pleochroic mineral, but in some instances is changed to mus- covite. Some of the latter mineral, however, may be primary in origin. Quartz-hornblende-porphyrite—A dike of quartz-hornblende-porphyrite is found in the Big Santa Anita canyon, about 300 yards below the main falls. It (A. M.S. number 2) is a grayish rock showing small feldspar erystals imbedded in a groundmass of feldspar and hornblende. Micro- scopically plagioclase is seen to be the dominant mineral. It occurs principally as rather broad, lath-shaped crystals, considerably weath ered, but appearing to have numerous small inclusions. Alteration begins in the center of each crystal, and kaolin appears to be the princi- pal resulting product, although muscovite is occasionally present. Next in importance to the plagioclase is orthoclase, which is generally much altered to kaolin. It occurs both in automorphic and xenomorphic crys- tals and usually contains numerous small inclusions. The principal basic mineral is a green, weakly pleochroic pyroxene, which has under- gone a process of uralitization. It occurs in rather small crystals be- tween the larger feldspars. Hypersthene is common in small grains. The quartz occurs in small xenomorphic grains scattered throughout the groundmass. Small, slender, lath-shaped crystals of a bluish min- eral are also present. Magnetite is common, and pyrite and ilmenite are also present in lesser quantities. Apatite occurs sparingly in long, narrow, colorless crystals showing parallel extinction, while epidote is found in small grains. Quartz-hornblende-porphyrite also occurs as a dike which crosses the old Sturtevant trail southeast of mount Wilson. Microscopically this rock (R. A. number 382) is gray in color and rather fine-grained in text- ure. It is much jointed, and fresh surfaces are hard to obtain, but where they are examined carefully they usually disclose feldspar crys- tals large enough to be seen by the unaided eye. In thin sections pla- gioclase is seen to be the most abundant mineral. It occurs in small, XXVII—Buut. Grou. Soc. Am., Vou. 16, 1904 200 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL elongated prisms, between which are xenomorphic grains of orthoclase, hornblende, and, rarely, quartz. Inclusions are common in the plagio- clase crystals, and twinning according to the Carlsbad law was often noticeable. The orthoclase grains show zonal structure quite commonly. Hornblende is abundant in small aggregates, showing yellowish green to light green pleochroism. Chlorite is a prominent constituent, being a secondary product of the hornblende. A little secondary muscovite is also present. Diabase porphyry.—A dike of diabase porphyry 4 feet wide breaks through the light colored granite rock which forms the cascades in Big Santa Anita canyon near Sturtevants camp. The porphyry is more re- sistant to weathering than the intruded granitic mass, and, as a conse- quence, protrudes slightly from the bed of the stream and from the sides of the canyon. The rock appears to be affected by a series of invisible joint planes which become apparent when the rock is subjected to shock. Fracturing along these cracks takes place easily, but otherwise the rock is very hard and resistant. Hand specimens of the porphyry show it to be very dark colored, micro-crystalline, and with a few phenocrysts of feldspar scattered about in it. Fresh fractured surfaces exhibit the lus- trous faces of the phenocrysts. Sections viewed microscopically show a holocrystalline groundmass of small lath-shaped crystals and irregular grains or pseudomorphs of chlorite, the latter being the alteration pro- duct of pyroxenes. Phenocrysts of orthoclase, plagioclase, and rarely of pyroxenes are scattered throughout the groundmass. Inclusions are common in the feldspar phenocrysts, one case in particular being noted where a single orthoclase contained a C-shaped mass of chlorite and an isolated mass of unaltered enstatite. METAMORPHIC ROCKS In general.—As previously mentioned, metamorphics play an impor- tant part in the composition of the Sierra Madre or border range of the San Gabriel mountains. The gneisses are by far the commonest type in this class of rocks, but schists are also present, though in less abundance than the gneisses. The following metamorphics were obtained in the territory : Hornblende-diorite-gneiss, biotite-granite-gneiss, -hornblende- schist, and garnetiferous schist. Hornblende- diorite-gneiss—Gneisses of this type are probably the com-. monest rock in the southern range, with the exception of the granodiorite. They are intimately associated with the granitic facies and are probably derived from the latter by shearing stresses. They are generally fine grained, distinctly banded in light and dark, and with the banding irregu- larly and intricately contorted. A typical example of hornblende-diorite- PETROGRAPHY OF METAMORPHIC ROCKS 201 gneiss is found in the west wall at the mouth of the Big Santa Anita canyon. Megascopically the rock (A. M. S. number 4) is fine grained, distinctly colored in light and dark bands, and shows quartz, feldspar and hornblende. The bands are not continuous, but are in strips vary- ing from one-fourth to one-half inch (5 millimeters to 10 millimeters) in width and from 2 to 12 inches (5 centimeters to 30 centimeters) in length. The dark coloration is caused by hornblende. Under the microscope the gneiss is seen to consist of fine, xenomorphic crystals of plagioclase, orthoclase, and quartz in relative abundance in the order named. The feldspars are more or less fractured and weathered, calcite resulting from the alteration of the plagioclase and traversing the rock in small veins. The hornblende occurs in varying proportions, being quite abundant in the dark bands. Itis the green variety, and is present both as grains and in spindles having their long axes parallel to the banding planes. The fracturing so apparent in the feldspars is not present inthe hornblende. Chlorite results from the alteration of the hornblende. Magnetite in grains is also abundant, being arranged in rows in more or less regular fashion parallel to the banding. A chemical analysis of this hornblende-diorite-gneiss (A. M. S. number 4) from Big Santa Anita canyon, by A. M. Strong, gives the following results: EE ergs RN Sa a wid ae a As ne Re ne oA S hg n'e 62.41 Oe rete ats etre es ade vg cake to ee wwe ea aca 13.91 ra ea Br ater Ge cess Ue Plate alee goa bes ore G:87 * no oP AE he Sone pe es ea ea eee e ae Re 3.15 es ike dah a a sa aes tye iets! Y vast hn, 4 Siolemie tin Wile Siete Di aca Doon, De a tet oh eek eth ts Sha an et Uae Away: asia thew Ge 3.19 OMT ered an i A ee a aia land WRG dies win see BIC eS 3.34 RUE re ett ni oe 2 ole ithe Ses a cite wie beet Bian oS ole 2.57 100.66 A transition from hornblende-diorite-gneiss to quartz-monzonite takes place along the Henniger Flats—Eaton Canyon trail. A gneiss very simi- lar to A. M. S. number 4 occurs in the bed of the canyon at the foot of the trail, while 209 yards up the trail, and covering the top of a small ridge over which the trail passes, is another gneiss (A. M.S. number 8). This latter differs from that in the canyon by having much larger crys- tals of hornblende, all of which are oriented with their longest axes in a certain direction, but which show no regularity of arrangement (or banding) in the direction perpendicular to the long axes. This irregu- lar distribution of the oriented crystals gives the rock a very peculiar *The iron was all determined as Fe.O03, no separation of FeO being made. This makes the total iron given a little higher than the true amount. 202 ARNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL and characteristic spotted appearance. Still farther up the trail from No. 8, and grading imperceptibly into it, is the quartz-monzonite de- scribed as A. M. S. number 8. Transitions similar to the one just described are rather common throughout the range, and often take tie within relatively short distances. A hornblende-gneiss (R. A. number 9), having about equal amounts ; of orthoclase, plagioclase, and hornblende, makes up a large area on the Sierra Madre—Mount Wilson trail, a short distance above the Half-way house. The orthoclase in this rock shows zonal structure and, in sev- eral cases, twinning according to the Carlsbad law. Quartz, epidote, chlorite, magnetite, and hematite are also present in relative abundance in the order named. Nearly all of the minerals, especially the feldspars, are clouded by rows of minute inclusions. Biotite-granite-gneiss.—The walls of the canyon leading back from Al- pine tavern toward the summit of mount Lowe are composed of an intri- . cately contorted banded gneiss. Megascopically this rock (R. A. number 4) is rather dark colored, the darker bands having a somewhat glossy appearance along the planes of schistosity, caused by the minute flakes of biotite, which megascopically appear to be the dominant mineral. Under the microscope, however, the principal constituent of. the rock is seen to be orthoclase feldspar instead of biotite. This feldspar occurs in xenomorphic grains, some of which are quite large. Rows of inclusions parallel to the lines of schistosity pass through the feldspars and occa- sionally through the quartzes.. No twinning is noticeable in the ortho- clase, but occasional examples of zonal structure are seen. Biotite is next in order of abundance to the orthoclase, although, as previously mentioned, it appears in the hand specimen to be the dominant min- eral. It occurs in small reddish brown flakes scattered among the feld- spars and shows pleochroism from yellow to green. One case of twin- ning was noticeable in the biotite. Following the biotite in relative abundance is quartz, which occurs in small xenomorphic grains. A small amount of plagioclase (oligoclase), a few elongated crystals of hornblende containing rows of inclusions, some epidote, magnetite, zir- con, and a little chlorite are also present. The sections are crossed by minute cracks approximately parallel to each other, but perpendicular to the banding. The feldspars are noticeably more kaolinized along these cracks, but the latter appear to have no relation to the cleavage, as the same crack may extend through a feldspar and an adjacent quartz or hornblende crystal. These cracks are probably due to stresses which acted perpendicular in direction to those producing the original banding of the gneiss, and which took place after the crystals had become com: pletely formed. «J all r | oe! PETROGRAPHY OF METAMORPHIC ROCKS 203 Hornblende-schist—H ornblende-schist makes up a considerable area at the foot of the Mount Wilson trail,at Sierra Madre. It occurs in thick layers and is dark green in color. Hand specimens of the rock (R. A. number 28) show it to be fine grained, but with no apparent lines of schistosity, although the lines are easily distinguishable in large masses in place. The schist is weathered on the surface to chlorite, which also coats the faces of the numerous joint cracks; but small crystals of feld- spar are abundant enough in places to give it a decidedly speckled ap- pearance. Under the microscope hornblende is seen to be the prin- cipal constituent. This mineral is present in small grains, elongated fibrous blades, and elongated prisms with frayed ends (shown in sece- tions parallel to the long axis). Pleochroism is from light green to blue, and light yellowish brown to blue and green. Most of the crys- tals show the characteristic cleavage lines, but a few were noticed'which showed no cleavage whatever. The usual alteration products are pres- ent in abundance. Orthoclase is not uncommon, occurring in small xenomorphic grains, which are often clouded by minute inclusions. Crystals of plagioclase are also occasionally present. Garnetiferous schist.—A large boulder of a dark-gray garnet-bearing schist was found in the Eaton Canyon wash about a mile below the mouth of the canyon. The garnets, which were abundant throughout the rock, ranged in size from minute dots to five-eighths of an inch (16 millimeters) in diameter. No sections of this rock were made. Garnets ‘an inch in diameter are reported as having been found in rock in place in Eaton canyon, but, with the exception of the boulder noted above, no garnetiferous rocks were found in this territory by the writers. SUMMARY The San Gabriel mountains, comprising an area of about 1,200 square miles, extend for approximately 60 miles in a west-northwesterly direc- tion from Cajon pass,in San Bernardino county, to the Santa Clara river, in Los Angeles county. Considerable divergence of opinion re- garding the age of the chain has prevailed among previous writers, but it is probable that it received at least the greater part of its elevation during late Eocene or Oligocene time. The southern range of the chain, the Sierra Madre, is composed prin- cipally of granodiorite and gneiss, with some associated quartz-monzonite and gabbro and intruded aplite, quartz-hornblende-porphyrite and dia- base porphyry. The central portion of the mountains consists of some- what coarser grained granites and granodiorites, with intruded aplite, micropegmatite, etcetera. 204 ABNOLD AND STRONG—CRYSTALLINE ROCKS OF SAN GABRIEL The granites described are of the biotite variety and are found in the central part of the chain. The granodiorites consist of two facies, a fine grained hornblende-bearing variety from the Sierra Madre or southern border range and a somewhat coarser grained variety, containing por- phyritic orthoclase, from the central mass. These granodiorites differ from those found in the Sierra Nevada of central California by being, on the average, finer grained and having less quartz, titanite, and zircon. Hornblendite, consisting mostly of hornblende, but also containing a little plagioclase, is found throughout the whole area. Aplite is found over the whole region, while micropegmatite was found only in the vicinity of Pine flats. Quartz-hornblende-porphyrite and diabase por- | phyry occur in dikes in the southern range. Of the metamorphic rocks hornblende-diorite-gneiss is by far the commonest. Together with some biotite-granite-gneiss, it is associated with the granodiorites and quartz- monzonites of the Sierra Madre. Hornblende-schist and garnetiferous - schist, found by the writers only in the southern range, complete the list of crystalline rocks described. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 205-214 . APRIL 14, 1905 EFFECT OF CLIFF EROSION ON FORM OF CONTACT SURFACES BY N. M. FENNEMAN 3 (Read before the Society December 30, 1904) CONTENTS Page INR aes ar trey act eihircla coe wince tis (t bs 6 Swe Sw: ob dels i biew Sp as 205 Observations in the field ................. SC Pe ee eee beet pikes Om. £42 205 In POOR ISIN TOPOPTADINY. foi bi ive lak tlacie elestnd ele oe sewinien dace cwwees 206 Reena conditions of the problem... 2.2... 0. .scecensadecsseesorecacaes 206 Illustrative cases of sta ies CPCS. cd ody ed ve Sy val ales os yes 208 gra es Le i's Go N ic veh x ay che nde de bs ea ek die cowed wee be 208 Case I. Shore recession more rapid than shifting by submergence... -- 208 Case II. Recession progressively retarded until its rate equals that of RE eee ee Rete Ne ate ade stalh ona 4c ole dint Wau biel a win eed gles Mee aie eae’ 209 Application of case II to the area studied.................0.5008. die es 211 Case III. Recession less rapid than shifting by submergence....... ..... 211 RMU ee CE 218 oo A gw woo Sep A ies isin ia pie mcye Shh op cca 8 212 THe AREA 8TUDIED The occasion of this study was an attempt to determine the relations of the Wyoming Red beds to the underlying granite in the Front range of the Rocky mountains in northern Colorado. The southern half of the Boulder and the northern part of the Denver quadrangles represent the area of investigation. OBSERVATIONS IN THE FIELD It was evident from the nature of the variations in thickness of the sandstones that the sediments were laid down on a submerging land sur- face of considerable relief. It was also apparent that the details of this relief were totally unlike the forms produced by stream erosion. The rocks immediately below the contact surface are as free from weathering as those which lie hundreds of feet lower. It is therefore evident that during the progress of the submergence of the granite land surface, shore XXVIII—Boutt. Geou. Soc. Am., Vou. 16, 1904 (205) oe a nee 206 N.M. FENNEMAN—EFFECT OF CLIFF EROSION ON FORM erosion by the advancing waters has removed a part of the original land surface and altered, or rather erased, the details of relief. It was for the more exact interpretation of the forms as seen at this place that the following elementary formulation of certain general principles was undertaken. EFFECT OF PREEXISTING TOPOGRAPHY When an eroded land surface is submerged and covered by later sedi- ments the surface of contact between the two formations may be expected to have some dependence on the topography of the preexisting land. Other factors being equal, a topography of steep hills and valleys will show itself after burial in a corresponding unevenness of the surface of unconformity. The surface of contact is, of course, not visible, but a section of it appears when the formations at the line of contact are up- lifted and eroded. If the outcrop be on a plane surface and the line of contact straight, the surface of contact is supposably a plane. There is, then, a suggestion, at least, that the land surface of the lower formation before submergence and burial was one of faint relief. The more the line of contact deviates from straightness (ignoring subsequent folding) the greater the presumption in favor of strong relief before submergence. These inferences very properly take into account the effect of the topog- raphy of a submerging land on the form of the surface of contact with later formations, but they leave out of consideration the important change which may accompany the advance of the sea. Facrors AND CONDITIONS OF THE PROBLEM There are two factors which determine the deviation of the surface of contact from a plane. These factors are (1) the form of the preexisting land surface, and (2) the strength of shore erosion during submergence. These elements may combine in any proportion. Hypothetically, two extreme suppositions may be made. The extreme of rapid subsidence and small wave erosion would result in leaving the original subaerial topography unaltered and retaining it as the surface of unconformable contact. At the other extreme is the complete dominance of shore ero- sion, whose tendency is to plane off the hills and to make the contact with the newer sediments a horizontal plane. It should be stated here, though somewhat parenthetically, that shore erosion may, under certain circumstances, increase instead of destroy the relief of a submerging land surface, but such conditions are excep- tional. This is apparent when it is remembered that the shoreline is a contour line; that whatever straightens the contours is bringing the FACTORS AND CONDITIONS OF THE PROBLEM 207 surface nearer to a plane, and that the general tendency of shorelines is to straighten themselves; hence to reduce relief. The possibility of the opposite effect depends on unequal strength of the shore rocks, and can occur only when wave erosion is cutting out bays in the weaker rocks of a rugged coast, thereby making the coastline more irregular. Even under these conditions of unequal strength of rocks, increase of relief is not universal. In the following discussion this phase of marine denu- dation will not be considered. There is one further exceptional and unusual case which will not be considered in the paragraphs to follow. That is the case in which no erosive work whatever is performed by the oncoming waters. This may be exemplified by a preexisting land surface of such gentle seaward slope as to require the construction of offshore barriers in order to steepen the beach profile. But even in this case observation may sometimes detect a small degree of cliff-cutting along the new shoreline before the barrier appears. The usual case, and the only one needing considera- tion here, is that in which submergence is attended by the paring down of hilltops. Subsidence alone causes the shore to shift landward at a rate deter- mined by the slope of the land and the rapidity of crustal movement. This change of position of the shoreline will be spoken of as shifting. But the actual migration of the shoreline will depend partly on this hori- zontal shifting and partly on the rate of the attendant cliff-cutting. The actual rate at which the cliffs recede and the shoreline migrates may be greater or less than this shifting. The total movement of the shoreline, regardless of causes, will be spoken of as migration. If the position of the water-level be newly assumed against an irregular land surface, the first cliff erosion causes recession which is more rapid than the mere shifting which could be caused by any possible rate of subsidence. In other words, no supposition is allowable which would engulf an island so promptly that the waves could not etch its descend- ing slopes. This forging ahead by cliff-cutting becomes less and less rapid as the cliffs become higher, the amount of detritus yielded by each foot of recession becomes greater, and the disposing of the waste becomes an increasingly difficult task. By continual cliff-cutting against a land sloping shoreward, the cliffs will eventually become so high and their rate of recession so slow that the latter will just equal the rate at which the shore would shift landward by submergence alone. When high cliffs have once been developed it may happen, as ex- plained below, that their recession over a given space will be slower than mere shifting of the shoreline over the same space would have been had no cliffs been cut. The extreme of this condition is found where a 208 N. M. FENNEMAN—EFFECT OF CLIFF EROSION ON FORM broad horizontal surface is reached by the receding cliff. Over such a surface shifting would, of course, be instantaneous if the sealevel were as high as the level of the plain. When there is a cliff separating these two levels the rate of migration of the shoreline is, of course, that of cliff recession and no more. ILLUSTRATIVE CASES OF SUBMERGENCE CONDITIONS IN GENERAL The above conditions of submergence are more conveniently discussed by assuming certain cases and indicating them diagrammatically. No one case can represent the conditions during the whole history of a sub- mergence, but the entire history and the forms resulting may best be described in terms of these cases. CASE I. SHORE RECESSION MORE RAPID THAN SHiFTING BY SUBMERGENCE — Assume the rate of cliff recession to be constant and greater than the shifting due to subsidence. In figure 1 let B A represent the original A D . G Figure 1.—Recession more Rapid than Shifting. land slope. B Cis a certain assumed amount of vertical sinking. CA is then the amount of horizontal shifting of the shoreline corresponding to the assumed amount of sinking. D A is the actual amount of migra- tion of the shoreline, due partly to cliff-cutting. The tendency of the waves is to cut away the higher land and reduce it to a plane just as though there were no subsidence. In this case, however, the plane would not be horizontal, but inclined at an angle depending on the ratio of A D and D E—that is, the relative rates of cliff recession and subsidence. The water’s edge, which was at the outset at A, advances toward #, and in the meantime occupies all positions along the line A EL. This line, therefore, represents the sloping surface of a denuded bench having a gentler seaward inclination than that of the former land surface. This result is due to the fact that, on account a.) ee * Rate. a ee ae RECESSION MORE RAPID THAN SHIFTING 209 of active cliff-cutting, the migration of the shoreline has been more rapid than shifting due to subsidence alone would have been. The diagram represents only an initial, broad, unbroken slope of the land. If this slope be diversified by valleys and ridges, the latter will be truncated to the level of A EZ. The dotted line parallel to B A in the diagram, may be taken to represent the depth of the valley bottoms. As the cliff recedes it necessarily increases in height. The new-made sea bottom approaches continually nearer to the valley bottoms which at first were beneath its level in bays. ‘The bays become smaller and fewer, the cliff more continuous; the maturity of the shore is advancing. Finally, with the extension of the plane inland, its level may fall below that of the deepest valleys, the cliff will be continuous, and the subaque- ous cut terrace will have no depressions to be filled with the sediments _ of the newer formation. From this line landward the surface of uncon- formable contact between the older and newer formations will be approxi- mately a plane regardless of initial topography. This stage, in which denudation cuts below the valley bottoms, may or may not be reached. Cliff recession is accomplished against accumu- lating difficulties. Not only does the increasing height of the cliff yield an increasing measure of detritus for each foot of recession, but each cliff is becoming longer ; the bays in which the shore drift was formerly stored - in the form of spits, bars, and wave-built terraces, become fewer and smaller, and the material won from the cliffs must be disposed of by the slower process of dragging offshore. At any point of the line A FE the waters may prove unequal to the task of cutting a higher cliff. They must then be content to push the shore landward at the same rate at which subsidence would shift it without cliff-cutting. CASE II. RECESSION PROGRESSIVELY RETARDED UNTIL ITS RATE EQUALS THAT OF SHIFTING Assume as a second case that the rate of cliff recession is progressively retarded to a minimum. If the initial ratios be taken as in case _ I—that is, the migration of the shoreline greater than the shifting, as defined above—then the surface of marine denudation will be begun, as before. with a lower slope than that of the land surface. However, as the rate of cliff recession is diminished it will approach that of the shifting, which it will finally equal. The surface of denudation in the meantime curves upward, as represented by the line A Hin the diagram. When the rate of recession has become equal to that of shifting, the slope of the plane of marine denudation will equal that of the land surface. The two will be parallel and the receding cliff will have a constant height. This height will be such as to allow recession at the same rate at which the shoreline would move landward by submergence alone. 210 N.M. FENNEMAN—EFFECT OF ULIFF EROSION ON FORM The land surface is thus pared down to a uniform depth. If this depth is greater than the former relief, the resulting cut surface will be a plane or a broadly rolling surface, as described below. This condition is represented by taking the upper of the two dotted lines in the figure to represent the depth of valley bottoms. If the depth of paring be less than that of the valley bottoms (which may then be represented by the lower dotted line), the valley bottoms will escape the paring process and their lower parts will continue to indent the denuded surface. The dimi- nution of relief will, of course, bg equal to the constant cliff height. In general, if denudation cuts below the valley bottoms, the entire surface of the cut terrace will be of fresh rock. If erosion fail to cut so low, the valleys will be filled, and the weathered rocks of the valley bottoms will thus be preserved on the lower side of the contact. Figure 2.— Recession retarded. Under the conditions of case II the maturity of the shoreline will at first be advanced, but a limit of development will be reached, at which the rejuvenating effects of submergence will balance the advancement in the cycle dueto erosion. If the down-cutting does not extend to the valley bottoms, there will be on the site of each headland a cut terrace whose surface will, in the main, be a plane; but the sites of the valleys, and therefore of the bays, will show depressions in the older formation, filled by sediments of the newer. If marine denudation cuts below the valley bottoms, cliffs along the shoreline will be continuous, and (with one important modification, given below) the old land surface is planed down to a level. : | | The modification necessary to the general statement that when cliffs are entirely continuous along a sinking coast the cut terrace thus pro- duced has a plane surface, may be stated thus: In the submergence of a surface of hills and valleys the coastline can not well be straight. Low cliffs recede more rapidly than high ones; hence where the land is low the shore is pushed more rapidly landward. Reentrant curves are thus produced, whose possible depth is limited by several factors. In general, — RECESSION MORE RAPID THAN SHIFTING ya the advancing waters have an advantage on the headlands because of their exposure, and a corresponding advantage in the bay heads because the cliff there is lower. The shape of the shore tends to adapt itself by a suitable curvature, so that these two advantages shall be in equilibrium. Within limits, therefore, cliff-cutting may be entirely continuous along a sinuous shoreline. Since the form of the shoreline at any one time is a type of the contour lines on the surface of the cut terrace, in the case in hand the terrace has a fluted surface with low ridges running seaward along the line of retreat of each headland. It is especially to be noted that the rock surface of such a cut terrace will be unweathered. The sediments laid on it, being derived partly from the cliffs and partly from active streams, may be of the coarsest sort and may be quite fresh even when feldspathic. Both these condi- tions clearly distinguish this case from the advance of the sea over a peneplain with its deeply disintegrated rocks. In the latter case the sediments can be coarse only when the mantle rock of the peneplain contains peculiarly resistant pekbles. APPLICATION OF CASE II TO THE AREA STUDIED In the area defined in the opening paragraph the contact of the Ar- chean granite and Wyoming sandstone presents the features appropriate to that phase of case I[,in which denudation by cliff-cutting has cut beneath the valley bottoms of the submerging land. Ignoring the sub- sequent folding which made the Boulder arch, and supposing the east- ward tilting due to mountain-making to be undone, the surface of the Archean-Wyoming contact has a maximum relief of perhaps 500 feet. Its undulations are of the smoothest sort, showing no slopes less than 3 miles long, with a maximum steepness of about 150 feet to the mile. These long slopes are entirely unbroken by valleys in the Archean floor, and the rocks of the latter are quite as fresh at the contact as at any greater depth. These features indicate that wave erosion by the ad- vancing Wyoming sea. pared the entire surface, even the valley bottoms, to a depth at least equal to the thickness of the zone of weathering ; that in so doing the minor features of relief were erased and the major features considerably reduced. Cliff-cutting was therefore substantially continuous along the whole of the curved shoreline, and the sinking of the land was slow enough to admit of a mature shoreline topography during the stage recorded in the present outcrops. CASE III. RECESSION LESS RAPID THAN SHIFTING BY SUBMERGENCE Assume, as a third case, that the rate of migration is less than shifting by submergence alone would produce. On this supposition the surface 212 N.M. FENNEMAN—EFFECT OF CLIFF EROSION ON FORM of denudation will have a steeper slope than that of the land. This is impossible unless an initial cliff be assumed. This case therefore be- comes important only as a later stage of a process begun under other conditions. In the diagram, BC represent a cliff whose recession has become less than the simple shifting of the sinking shore would be without a cliff. The recession is equal to CD during a vertical subsidence equal to D E. During the same subsidence the shifting of the shore with no cliffs would have carried the water’s edge from A to E through a horizontal distance equal to D A. | Here, again, cliff-cutting is assumed to continue at a uniform rate; hence the cut slope is a plane, and the cliff height has been reduced to zero at H. In actual occurrence the rate is accelerated as the height is £ D G = Fieure 3.—Recession less Rapid than Shifting. reduced, and the line CE becomes a curve, convex upward and never actually intersecting the line A B so long as any land remains above water. Assuming the conditions of this case after cliffs have once been developed, the surface of the cut terrace will rise landward, intersecting first the deeper valleys and then the shallower, resulting in the greater separation and subdivision of headlands and the isolation of islands. The rejuvenating effects of sinking are more rapid than advancement in the cycle of erosion, and the features of maturity give way again to those of youth. GENERALIZATION Abandoning straight lines, but using the principles stated above, a truer picture of the forms actually resulting from submergence may be obtained by assuming land forms of familiar type. Assume, for exam- ple, a horizontal land, but having a seaward slope near the shore. Its surface may have something of the compound curve shown in figure 4. . ke ee a GENERALIZATION 9138 The land is assumed to have considerable relief, and the submergence is assumed to be complete. Taking the process as a whole, the migra- tion of the shoreline is equal to what it would be by shifting alone; but at different stages the actual migration and the hypothetical shifting have varying ratios. If we assume the process to begin with relatively low initial cliffs (a common case), migration at first exceeds mere shifting by reason of excessive cliff recession (case I); the two curves representing the land surface and that of marine erosion separate; the latter is concave up- ward. With the decreasing rate of cliff-cutting there comes a stage where it merely equals the rate at which shifting alone would carry the shore landward (case II); the curves are about parallel and have their B ms Figure 4.—Generalized Diagram of Shore Erosion. The upper curve represents the original land surface. The lower represents the surface of marine denudation. maximum separation. Then migration becomes less than the theoretical shifting by reason of the small seaward slope of the remaining land (case III); the lower curve becomes convex upward and approaches the upper. Interpreted in terms of shoreline topography, the erosive work of the sea at first advances the shoreline in its cycle, smoothing its contour, pushing back its cliffs rapidly, and increasing them in height and length. As this process grows slower by its own development, the effects of sink- ing begin to keep pace with those of erosion. The shore forms remain stationary in their cycle while the shoreline recedes landward. The cliffs have at this time their maximum height. With the diminution of slope of the land surface, the surface of marine denudation approaches it, the cliffs become lower, the rate of their recession is augmented, and the features of youth return. Continued sinking will at length cause the surface of the cut terrace to intersect that of the land, and submergence will be complete. 914 N. M. FENNEMAN—EFFECT OF CLIFF EROSION ON FORM If the last land to be submerged were an undissected plain, it would be narrowed by cliff-cutting at an increasingly rapid rate as the height — of the cliffs was lost. If the last remnant be cut by valleys, continued submergence must subdivide it into islands, the story of each one of which will reproduce, in a manner, that of the entire land mass. In actual experience it may be expected that a large part of all cases — of submergence will be found to begin with well developed cliffs cut while the land was standing still. The height of these may not be greatly increased during the migration of the coast. In such cases the initial stages of the process represented by figure 4 are omitted. : : 7 = "y ‘ i 4 4 , ‘ ve BIL see ‘ 7 » a ey ; y t > a ‘ y ‘ > . \ VOL. 16, 1904, PL. 36 BULL. GEOL. SOC. AM. DIAGRAMMATIC REPRESENTATION OF THE DISTRIBUTION OF MORAINE AND OUTWASH GRAVEL (DOTTED) ON THE NORTHERN HALF OF THE WATKINS GLEN QUADRANGLE BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 215-228, PL. 36 APRIL 29, 1905 MORAINES OF THE SENECA AND CAYUGA LAKE VALLEYS* BY RALPH 8. TARR (Read before the Society December 31, 1904) CONTENTS Page i en Sn Gea: ie aa ccigeees ee aii) var sda eke edks pam Lnednns 215 Ecce et a WU SG, Gin! Loca Wohew ela Rie Galen oe sla 38 216 EEN EER EMAENTOMN Sy ea c'alivarn a's as wkwees dcop beige ards veecene acces 216 MEE TAINED DE 0 OCCUPALIOD.-. . 2.6... s esse ccc ke ced ccclet ce eceedceces 217 Semueny Sembured or the MOraines 2. 1. o etl ee sce vedecas cesees 218 SremmenttEOm JAGEFA] MIOFAINGS, f. osb 1 he ok a cle claws hee da ccuclvcecasdeasctces 218 IS RES 2 Tid awa akieisie ac padteid tle d objet bene dese cs 219 INR CMON a esc win ge see 0 G.0.d bv sd c'elenld oS a nyenins'e's 220 colar Sw co's fai aha a aloye oio.c. overs @isim-e'e's e:hp pre es dv dvees 221 Unequal development of moraines of valley lobes...........00ee.ceeeee eens 2271 Morainic complex between Cayuga and Seneca valleys..........0-.0eecee eee 228 IEEE SEIS Sc) ae ee bei de cewek eeda cde \ccadieeslcaeds 224 IGE Es oe PEL) Gore i iol asic - cyheb owidd CA valeaeiasesdalies cleadaec 224 I eye ee Mesh 8 seo on bos badd aleldabedervceddeus 224 Morainic complex in the upper Capes MG ORBEA VRLESY Ss. «cic, i ote ois r'0, ren 6 225 Pa EREUNCIEND OF GPG TIOTAINIGS. «4 occa sep ne con cle ce enc vce vee veescvesren 227 MIE MEIOOELTIOW CHATIFIGIS: 65 SF ary et eee 233 Ea ed PRS ES eee ten eer ee cee ee eee eee 233 Seneca and Cayuga valleys................ Seba avstt tS kod fo Seles ab eh 234 The Chemung near Himira...............50- actaigie CAS ae tee 234 II tao) re Rey a iy, ww due) evs 6 od wie win gim SSL !a ae (eter eee 234 ENR hac rere ss ae cin als cig Gs a ww sik wie wih oo x!) eieyeien se © 235 Willseyville valley ......... ea he fe iel aca teks a (ore ata wlio RE, br elds sy) & ca oe Ee WRMRINE DME REM ODNS et Sas oh h Sly oy Lisle xafGAl sick a.e' ee is ‘ if = 4, t \ i 8 N ? x \ F ifirz, - \ ip ® ITHACA TOPOGRAPHIC SHEET ied ph of a model (by William Stranahan) of northeastern portion of plate 37 =) Photogra VOL. 16, 1904, PL. 39 BULL. GEOL. SOC. AM. VER nNioGaA RI TrouGH ERSION OF THE Div u Figure 1.—Tu GA LAKE VALLEY AYU OPE OF C STEEPENED SL 2.—'THE ay E 2 GUR Fi DIVERSION OF TIOUGHNIOGA RIVER AND STEEPENED SLOPE OF CAYUGA LAKE HANGING VALLEYS 25k in which the streams tumble precipitously down the steepened slope as a series of rapids, cascades, and falls (figures 2and 3and plate 38). The maturity of the hanging valleys proves that a long time was required for their development at a baselevel not far from the 900-foot level. There Seneca Lake Figure 1.—Cross-section of Seneca Lake. Three miles north of Watkins. Column of figures refers to elevation with reference to sealevel Vertical scale exaggerated about five times. are two ages of gorges cut in the steepened slope, one distinctly post- Glacial, the other, being both broader and deeper and partly filled with deposits made by the Wisconsin ice sheet, evidently antedating the advance of this ice sheet. 7 400 we 4 1080 Fr ~700 no hock. Figure 2.—Profile along Watkins Glen Creek. Scale same as figure 1. The headwaters of the hanging valleys are far less mature than the lower reaches, for from the headwaters the valleys broaden downstream, while the valley walls become less steep. A gorge condition exists in many of the headwater tributaries, and the valley walls are almost uni- formly steep, indicating™that active erosion was in progress here when 232 R.S. TARR—DRAINAGE FEATURES OF CENTRAL NEW YORK the new cycle was introduced by which the main valley bottoms were lowered. Hanging valleys in the Tiougnioga trough.—The Tiougnioga valley, which extends southeastward from near Cortland across the Cortland and Harford topographic sheets, carries southward to the Susquehanna the headwaters of the Fall Creek drainage area, normally tributary to Cayuga lake* (see plate 39, figure 1). From near Cortland to a point about 3 miles south of Blodgett Mills the valley narrows, becoming there a deep, steep-sided, gorge-like valley, below which there is again a broadening. This narrow section is without doubt an old divide region across which the upper Fall creek has been diverted. WwW Seneca Lake 500 —————— 00 Saree eases ene ascent 00 Hae rE a See 200 —— a? /00 Wee C ontin uation -/00 = Seed of valley to ~200 “it depth bp ae: Loring ~300 1 Qf ‘Wat Ins -400 /080 feer. -~500 -600 -700 Figure 3.—Profile of Hector Falls Creek, Seneca Valley. Two and one-half miles north of Watkins. Scale same as figure 1. On either side of this valley, both above and below the narrow gorge, the tributaries occupy gorges cut in hanging valleys. The upper por- | tions of these tributaries are in broad, mature valleys, while the lower portions are in rock-walled gorges 50 to 100 feet in depth, resembling in appearance the buried gorges in the Cayuga and Seneca troughs. In each case the gorge is cut in a rock bench extending across the mouths of hanging valleys. Viewed from the opposite side of the Tiougnioga val- ley, these tributary valleys are plainly seen to be hanging above the main valley, with the gorges cut in their bottoms. That these gorges are not post-Glacial is proved by the presence in them of drift deposits and, in at least two cases, of buried sections so completely filled with dritt that the present stream has been turned * Carney: Journal of Geography, vol, ii, 1903, pp. 115-124, HANGING VALLEYS 233 aside and forced to cut a narrower gorge apparently entirely of post- Glacial age. While this condition of hanging valleys is present in a number of the tributaries, it is most typically illustrated in East Virgil ereek, west of Messengerville on the Harford sheet, and in an unnamed creek on the east side of the valley about 2 miles south of Blodgett Mills on the Cortland sheet. It is in these two valleys that the buried gorge sections are found. Other hanging valleys—In a number of places on the Watkins Glen quadrangle there are hanging valleys, though in most cases not as typical as those described above. More or less completely drift-filled gorges cut in rock benches across the valley mouths, and changing upstream to broad, mature valleys arecommon. This condition is illustrated very clearly in a number of the tributaries to Cayuta creek between Van Ktten and _ Waverly. In the work on this folio it has not been possible to carefully investigate each of the suspected hanging valleys, and therefore it can not be stated just how many such valleys there are; but enough have been proved to exist to demonstrate that the hanging valley condition is widespread in this region. Far the most perfect examples are those of the Cayuga and Seneca troughs, but the instances in-the Tiougnioga and Cayuta valleys are just as certainly instances of hanging valleys. Extension of these studies to other areas and correlation of the results obtained will, it is hoped, warrant interpretation not now possible. LOWERED DIVIDES General conditions.—One of the most striking features in the topography of the divide region between the Saint Lawrence and Susquehanna drainage is the marked absence of well defined divides between the larger streams which head in this region. Along a number of valleys it is possible to pass from one drainage system to the other through open valleys in which the present divides are determined not by rock but by drift deposits. A similar condition is found between the headwaters of the larger tributaries on each side of the main divide; and even in the case of the smaller. tributaries there is frequently a condition of lowered divides. In discussing this paper at the Geological Society meeting in Philadelphia Professor Davis applied the very descriptive name of “through valleys” to this condition of valleys connected across lowered divides. Accompanying this condition there has been much diversion of drain- age across the lowered divides, so that it is very frequently the case that the present divide does not coincide with an earlier position. ‘This is proyed by the fact that streams frequently head on drift deposits in a 234 KR. S. TARR—DRAINAGE FEATURES OF CENTRAL NEW YORK broadly open valley, flow toward a narrowing section of the valley, pass through a gorge with rock walls and drift floor, and thence on into another broader section. These conditions may best be understood by describing several specific instances. Seneca and Cayuga valleys—In both these valleys (plate 37) there is an open trough southward across the present divide, which is in drift; and in both cases one side, and in some cases both sides of the valley have pre- cipitous sides. A canal and a railway have passed across the Seneca divide with moderate grade, and a railway with a somewhat steeper grade passes through the Cayuga trough. In these two valleys the topography does not demonstrate the exact location of the earlier divide The Chemung near Elmira.—From Corning to Elmira there is a broad valley swinging northward past Horseheads, and followed by two rail- roads (see plate 40). The Chemung river leaves this valley near Big Flats and makes a cross-cut to Elmira behind a high mass of hills. — This section of the valley is much narrower than the abandoned portion — past Horseheads, and it flares both ways from a narrow gorge section, evidently the site of an old divide. No rock is encountered by the stream, and the valley is partly filled and clogged with glacial drift, in-- cluding well defined moraines. Near this region there are four Satiee hills, with valleys behind them flaring both ways from a central divide, which bear a close resemblance to the valley followed by the Chemung. Two of these lie about 2 miles southwest of Elmira (see plate 41, figure 1), one about 2 miles northeast, and one about 2 miles northwest of Horseheads (see plate 40). If the plain of outwash gravels over which the Chemung flows had been built 100 feet higher, there would be in these cases an almost exact duplica- tion of the conditions in the Chemung valley, only on a smaller scale (see plate 40). Cayuta valley—Next to the Chemung river, Cayuta creek is the longest stream on the Watkins Glen quadrangle. It is, however, joined by no large tributaries, and for most of the distance its divide is within from 2 to 4 miles of the creek. Its valley has two narrow, gorge-like sections, in each case with a broadening of the valley in both directions from the narrow portion (see plate 37). One of these narrow sections is about midway between Cayuta and Rodbourn and one between Van Etten and Waverly... The topography clearly indicates that the narrow sections were the sites of earlier divides, and that parts of three stream systems have been united to form a single creek. One portion was tributary to the Seneca valley, one flowed eastward past Spencer, and the third, as now, southward to the Chemung river at Waverly. ‘This valley is fol- BULL. GEOL SOC. AM. ¥ flats. +f oi SSN ~ +> SS , (> ee Mother f * > - B eae Pine fOity P a Pegg = wih y DIVERSION OF THE CHEMUNG RIVER NEAR ELMIRA, NEW YORK VOL. 16, 1904, PL. 40 SS ee aan [Eh pent Base, a xy ee: ge : hyo uthport WA oy Bw LORFESS . ag VCarohkne oe é , a = = pores AYRE] mete PI 2 chen a Tee ae aes ACR Figure 2.—THrE WILLSEYVILLE VALLEY LOWERED DIVIDES SOUTHWEST OF ELMIRA AND WILLSEYViLLE VALLEY LOWERED DIVIDES 935 lowed by the main line of the Lehigh Valley railroad, and toward the east, past Van Etten, by a branch of this railroad. The present divides are low drift masses, and any old divides that may have existed have been completely obliterated, so that no rock is encountered in the stream course. Tiougnioga valley —Carney * has shown that the headwaters of Fall creek, instead of following the broadening valley toward Ithaca, turn southward at Cortland into a narrowing valley, through a gorge section, evidently the site of an old divide, and thence into a broadening valley. As has been stated above, this doubly flaring valley has tributaries hang- ing well above the present valley bottom on both sides of the narrow sec- tion. Drift fills the valley, till and moraine occur on its sides, and no rock is encountered in its bottom, even in the narrowest part (see plate 39, figure 1). Willseyville valley — Between the headwaters of Six Mile and Willseyville creeks, on the Dryden sheet, there is a deep, gorge-like valley, with walls so steep that they are not cleared of forest (see plate 41, figure 2). The valley bottom is occupied by extensive moraine accumulations, and no rock appears in it, nor any well defined divide, the waters at present part- ing in the moraine. Toward the narrow gorge portion the valley narrows from each side, indicating the location of a former divide. At present two railroads pass with easy grade through this peculiar valley. Tributary valleys, with bottoms below the level of the moraine filling of the main valley, enter it in the narrow gorge portion. Other instances.—It would be possible to multiply the instances of this condition from the southern central portion of New York. For example, confining attention to the Watkins Glen quadrangle, Post creek, on the western side of the quadrangle southwest of Watkins (see plate 37), flows from a broad into a narrowing valley. There is no definite divide in the Burdett-Reynoldsville valley northeast of Watkins, nor between the upper Taghanic and Cayuta Lake valleys, nor the Pony Hollow and Butternut valleys (see plates 37 and 38). It may be said, in fact, that there is no single case of a well defined rock divide between the head- waters of the main branches of any of the larger streams on the Watkins Glen quadrangle. Texas hollow.—This valley is one of the most peculiar in the Watkins Glen quadrangle (see plate 42, figure 1). It is a deep valley with sides so steep that for the greater part of the distance the forest remains, and no road ascends to the upland; and it is so narrow at the bottom that, excepting at the ends, there are only a few small, poor farms. The * Journal of Geography, vol. ii, 1903, pp. 115-124. ale, Sig? 936 R.S. TARR—DRAINAGE FEATURES OF CENTRAL NEW YORK valley broadens both ways from a narrow central portion, in which there is a low rock divide near the present divide, which is in drift. This “hol- low ” is therefore a double valley, with both a sloping bottom and flaring walls on each side of the divide. Its form is gorge-like, its walls re- markably straight and smooth, and its tributaries confined to small streams, which head on the very edge of the “‘ hollow.” Such steep, smooth, straight valley walls, though nowhere on the Wat- kins Glen quadrangle as well developed as in Texas hollow, are, never- theless, a common type of topography in this region (see plates 37 and 08). They are found, for example, south of both Watkins and Ithaca ; in the Cayuta valley; near Elmira; in the valley southwest of Mecklen- — burg; and in many other places. Such valley walls are most perfectly developed in the neighborhood of lowered divides, and they occur in both north-south and east-west valleys. Divides of smaller streams.—Many of the smaller headwaters are located. in broad, cirque-like, upland valleys, whose divide crest commonly reaches elevations of from 1,500 to 1,800 feet. These upland valleys, which are decidedly mature in form, frequently have gorges cut in their bottoms, and in some cases are breached by gorges. The best and most typical instance of this so far observed is on the very edge of the Wat- kins sheet, almost due west of Watkins. A very perfect cirque-like valley faces eastward, and the smooth, regular divide is cut completely across by the gorge of a west-flowing stream. From this condition there is every stage to such cases of complete obliteration of divides as those described above. Southwest of Van Etten, for instance, the divide between Baker and Wyncoop creeks (see plate 42, figure 2), is a low, flat-bottomed valley, bounded by smooth, straight-sided, precipitous walls. Just south of Breesport, at the head of Baldwin Creek valley, the divide is so low that if the gravel plain that was built at the closing stages of the ice occupation could be removed Newtown creek could be easily diverted across it. These instances may serve as types which could be added to if it were necessary ; and itshould be noted that both in the large and small valleys there are instances of lowered divides extending in all directions. Present stream courses.—The facts stated above prove conclusively that there has been a widespread condition of divide lowering in the region under consideration. Exactly the extent to which this lowering has been carried in the major valleys is not certain because of the drift fill- ing; but in some of the smaller valleys the divide lowering has not gone so far that the glacial deposits have obliterated the rock divide. Whatever the cause for the lowering of the divides, the present stream VOL. 16, 1904, PL. 42 BULL. GEOL. SOC. AM. TROUGH HoLiow Figure 1.—T'rexas Figure 2.—DIvVIDE BETWEEN BAKER AND WyNcoop CREEKS TEXAS HOLLOW TROUGH AND DIVIDE BETWEEN BAKER AND WYNCOOP CREEKS LOWERED DIVIDES yy courses, which so often lead across the sites of ancient divides, are due to the fact that glacial deposits have so graded the valleys as to make streams of two systems unite in a single course. For instance, in both Post Creek and Pony Hollow valleys, moraines form the present divides, and outwash gravel plains supply a grade down which the streams flow across ancient divide sites. The same is true of Cayuta creek. The course of the Chemung west of Elmira is also determined by outwash gravels, which spread fan-shaped from the neighborhood of Horseheads, both toward Elmira and Big Flats, making the part of the valley near Horseheads higher than those parts farther away from the moraines, during the building of which the outwash gravels were supplied. By these morainic and outwash gravel deposits the present divides have to a large degree been determined, and, as a result of the changes thus brought about, the drainage of this divide region has been pro- foundly altered; but without a previous lowering of the divides these decided changes in drainage would not have been possible. INTERPRETATION STATEMENT OF HYPOTHESES Three different hypotheses suggest themselves as possible explanations of the drainage peculiarities described above: (1) Ice erosion; (2) ero- sion by ice-born streams; and (38) headwater erosion, lowering divides and capturing opposing headwaters. The ice-erosion hypothesis must include not only the last ice advance, but also earlier ice advances, of which, however, there is no direct evidence so far discovered in this region. The same is true of the hypothesis of ice born streams; and, in addition, the possibility of water erosion during both the advance and retreat of each ice sheet must be considered. These hypotheses will be considered separately. ICE-EROSION HYPOTHESIS A number of facts are opposed to this hypothesis. In the first place, as has been stated elsewhere,* there is in this region definite evidence of weak ice erosion during the last ice advance. Residually decayed rock abounds in the southern half of the quadrangle, and some is present even on the margin of Cayuga lake below the edge of the steepened slope. Moreover, gorges cut in the steepened slope have not been erased nor distinctly modified in form by ice erosion. If it is true that the last ice advance in a main valley like the Cayuga trough did not perform marked *R.S. Tarr: American Geologist, vol. xxxiii, 1904, p. 286; Journal of Geology, vol. xiii, 1905, p. 160. XXXII—Bu.tt. Geou. Soc. Am., Vou. 16, 1904 238 R.S. TARR—DRAINAGE FEATURES OF CENTRAL NEW YORK erosion, it would seem certain that ice erosion in less favorably situated valleys could not have been great. Furthermore, some of the best instances of “through valleys” are south of the zone of most active ice erosion and in a region where residually decayed rock abounds. The great number of instances of lowered divides and steepened slopes, in valleys of all sizes and extending in all directions, would demand an altogether remarkable irregularity of ice movement. This may be illus- trated by the case of the Chemung west of Elmira (see plate 40). To account for its lowered divide, ice would have to move either east or west ; but at the end of its narrow section, 2 miles southwest of Elmira, are two small hills with valleys similar to that of the Chemung, one extending north and south, the other east and west (see plate 41, figure 1). These three valleys are apparently of the same origin, for, barring size, their characteristics are essentially the same. To account for them by ice erosion would require ice movement in two valleys at right angles and — all within an area of a square mile. The flaring of the valleys in two directions from a narrow central por- tion hardly seems a probable result of ice erosion. In Texas hollow, for example, the same characteristics of valley wall form extend both north and south of the narrowest part. It is a question whether, in the case of ice passing through a narrow divide gap, its movement would not be checked by the narrowing valley and a tendency toward stagnant conditions be brought about. Especially would this seem to be the case where the valley extended at an angle to the general ice motion, as was the case in a large number of instances in this region, including Texas hollow. There are some cases where ice erosion by the Wisconsin ice sheet is entirely out of the question. This is best illustrated in the Tiougnioga valley, where, on each side of the narrow gorge portion, there is a condi- tion of hanging valleys with gorges in their bottoms, which were devel- oped before the Wisconsin iceadvanced. In the Willseyville and Cayuta valleys also there are tributary valleys developed after the lowered divide conditions were brought into existence, and therefore, since they are drift filled, before the last ice advance. Since this has been proved to be true of several valleys, it seems probable that it is true also of others, concerning which such definite proof has not yet been found. Altogether the facts weaken the ice-erosion hypothesis, and in some of the valleys the evidence is definite that their condition is not due to ice erosion during the Wisconsin ice advance. Facts have not been dis- covered which disprove the hypothesis that these valley conditions are due to an earlier ice advance; and, although no evidence of such earlier INTERPRETATION 239 ice advances in this region has been discovered so far, that hypothesis must still‘be held as a working hypothesis in future studies. ICE-BORN STREAM-EROSION HYPOTHESIS Wherever positive evidence could be discovered in this area the ero- sion of streams associated with the melting of the Wisconsin ice sheet was found to be very slight, in most cases amounting merely to a notch- ing of the drift deposits. The ice-born streams were depositing rather than eroding profoundly. That the peculiarities of the valleys described . above are not in any important degree due to erosion by streams flow- ing at the closing stages of the last ice advance is demonstrated by the fact that practically all the valleys contain undisturbed morainic deposits _which could not have escaped destruction if there had been marked erosion by ice-born streams. This form of evidence does not, however, eliminate the possibility of water erosion connected with the advance of the Wisconsin ice; but in certain cases, like the Tiougnioga, Cayuta, and Willseyville valleys, the presence of older buried tributaries does eliminate this hypothesis for these par ticular cages, and, since they are analogous to the others, weakens the hypothesis that they are due to water erosion during the advance of the Wisconsin ice sheet. The great number of instances, extending in all directions, and of all sizes, would call for a vast amount of water erosion under very variable conditions. I find it exceedingly difficult, for example, to postulate any conceivable set of conditions by which ice-born stream erosion could possibly cut the complex of channels near Elmira and Horseheads, de- scribed above. Moreover, it does not seem possible that water erosion of overflow or marginal type could form the doubly flaring condition of Texas Hollow valley, nor the Chemung valley west of Elmira, nor a score of other similar valleys. Even if several ice advances have been involved, and a large number of ice front positions are postulated to account for these peculiar valleys, the efficiency of glacially supplied streams to form such topographic features is questionable. While the ice was advancing and receding from this hilly region its front could not at any time have long stood on the divides, and while it did, judging from the history recorded at the closing stages of the last ice advance, deposition would seem to have been the rule, not erosion. When the ice fronts stood north of the divides there were marginal lake conditions, and the inefficiency of such a river as the Niagara or the Saint Lawrence to form a rock gorge does not lend much support to the hypothesis that the overflows of the Re bes 240 R.S. TARR—DRAINAGE FEATURES OF CENTRAL NEW YORK - glacial lakes were capable of such profound erosion as that necessary to — account for the many deep, long cuts in the divide region of southern New York. From the facts, as indicated in the above outline, it is evident that the peculiar valley conditions of this area are not the result of the action of streams supplied by the melting of the Wisconsin ice sheet, and that in several instances, at least, the conditions were brought about long before the advent of this ice sheet. The hypothesis that these conditions were brought into existence by possible earlier ice advances is weakened by several facts aside from the fact that studies in this region have not yet revealed any evidence of the presence of former ice sheets. It must, nevertheless, remain as a working hypothesis until either some other explanation is established or this one eliminated. HEADWATER-EROSION HYPOTHESIS On this hypothesis the assumption is that streams gnawing at their headwaters have in many cases so lowered the divides as to permit com- plete diversion of streams across these divides when the valleys have been graded up by glacial deposits. By this hypothesis it is further assumed that the gnawing back at the headwaters often caused an en- croachment of one stream system on the opposing system and the con- sequent capture of the headwaters of rival streams in favorable cases. At the present stage of the study it does not seem wise to enter into the consideration of why this headwater erosion may have been in progress nor exactly how it was operating. Further study over a wider area may make this consideration desirable at a later time. The difficulties in the way of accepting either ice erosion or ice-born stream erosion as an explanation of the peculiar valley conditions of this area may be considered as favorable to the head water-erosion hypothesis. Moreover, many of the facts opposing the rival hypotheses directly sup- port headwater erosion. The doubly flaring condition of the valleys is a type of form that such headwater erosion would be expected to pro- duce. If two streams heading in the narrow part of the Chemung valley west of Elmira were lowering their divides, just such a condition as exists west of Elmira would be produced. This process of divide lowering seems to have been in progress behind the small hills 2 miles southwest of Elmira and behind those northeast and northwest of Horseheads. In fact, in many places there are lowered divides in all stages of lowering, from simple notching of the divides to such complete reduction as to permit burial beneath glacial deposits. INTERPRETATION AND CONCLUSIONS 241 That before the glacial period stream erosion was vigorous throughout this divide region is evident from the steep slopes of the valley sides both in the small and the large valleys and from the presence of gorges in many of thevalleys. Neither glacial erosion nor ice-born stream erosion, even with the maximum possible number of ice advances, can possibly be appealed to in explanation of all these steep valley slopes, and since they harmonize in form with the steep slopes of the lowered divides they lend support to the hypothesis of headwater erosion for the lowered divides. On the other hand, as has been suggested to the author by both Doctor Gilbert and Professor Davis, such a condition of ‘ through valleys,” while not uncommon in regions of faulting or in regions of inclined and varied strata, has not been described in regions of horizontal strata outside the glacial belt. I have examined several hundred sheets of the United States Geological Survey topographic map in regions outside the glaciated area without finding even an approach to the conditions described above. To this difficulty must be added that of understanding just how divide lowering, operating excessively in some valleys, has produced no effect whatsoever in neighboring headwaters. 3 DISCORDANCE OF THE CAYUGA AND SENECA VALLEYS The lowering of the divides is in no way related to rejuvenation result- ing from the deepening of the Cayuga and Seneca valleys, for it extends outside of their drainage areas, and, even where developed in the head- waters of tributaries to these two valleys, occurs in valleys hanging high above the bottom of the main lake troughs. The upper parts of these hanging valleys are broad, deep, and steep-sided, with the main divides obliterated by erosion and glacial deposit, but with the lower ends ter- minating on the edge of a steepened slope notched only by small gorges. This fact favors the theory of glacial erosion for the basins of these two lakes, although other facts, stated in another paper, oppose it. CONCLUSIONS The one conclusive result of the studies so far made toward the solu- tion of this drainage problem is that, in several typical instances, the phenomenon of lowered divides antedates the last or Wisconsin ice in- vasion. The establishment of this fact is the excuse for the present paper. In all other respects this paper must stand as an unfinished study, in which certain observations are recorded and difficulties opposing each of three hypotheses pointed out. 242 R.S. TARR—DRAINAGE FEATURES OF CENTRAL NEW YORK The complete solution of the problem has not yet been reached, but some progress has been made; and, since others as well as the writer are at work on this complex field, this report of progress is made in the hope that it may serve as a slight contribution toward the ultimate un- raveling of the events which have caused the peculiarities of drainage in this region. BULL. GEOL. SOC. AM. VOL. 6, 1904, PL. 43 18° ALTA a a 100 Montserrat gy £500 Gallaiit e / /| d / Ss 1 S 9 - aris ! ' o ' S SN 9007 — 22 DBoychiila * ae Los Roques }’ oa \ ae Testigos \ \ : 6 | } . 2500 oe ee eee Saar EOS Lp 700 Ny ee a eam ae S = Ps te 12 Pan 229 al Se MO Ri (oe | ee r ny a 0 AS ears AB Tortuga ddd SUBMARINE CONFIGURATION OF THE WINDWARD ARCHIPELAGO Figures are in fathoms. Dotted lines, owing to lack of soundings, are conjectural. Compiled for Professor Alexander Agassiz from all known data by Robert Tl’. Hill. * BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 243-288, PLS. 43-47 May 12, 1905 PELE AND THE EVOLUTION OF THE WINDWARD ARCHIPELAGO BY ROBERT T. HILL (Presented before the Society December 31, 1904) CONTENTS : Page IO ett eh io aS sips yu dene ak nin less ac cb eae aly oe 244 Volcanic eruptions of 1902 and how they were ahidmie. ECOL AoC pee 244 I ie Ela eine Ses ada tele Coan sd Feb e s Pa eae re a. oe ebas Stas 245 en Wee IEON HOEY OL CIC or. oe Se eae eee Seep ee sicereca 245 AC Ber a re or Ss oo ae en Oa ee ey 246 Summary of tine recent eruption necessary........ 2.6 cee cece ee eee 246 Mecunmiomn and Work of the volcano. . 2... ose. ees ieee eect see awees 246 EIRENE CPINLTOWISS Ens ole Wc. aja ayo duo ogo h clea, cue s.ee eee Skee eet 250 Rock products...... eh tant oe, SiMe alos 2 a acid aa esas Pe Saat, 6 ay ores wih al aac ala ase a 250 EIRENE pecs, at ee ay rai a wy lied eB ope a Wan ved bes b's me 250 PEMRMTEMU Een eh otis ect cia od ess aes rere e ona.c Sac ene 251 Pelé’s contribution to the earth’s surface.............. cee cece eset eens 254 IIS SERA Pots k 5 a5 bois dod cay eid Sh Winks f lonsys! o F:GHa bi bold 2 o.d bidiclé wlelw 08 do 255 I MM GMITIEY OL PONG... ora a vec. cid ees nileeecus wandeaneemey 255 Pelé a type of the volcanic battery of the Windward archipelago........ 256 new see datvrer story May be studled....0.0...c0ce sec cee ce weeusss arene 257 Vague theories concerning origin of the islands of the Windward archi- ee tere eee ea) och hae Sa Scie wit Held os v aise ea nals fae ake'd . 258 Geographic differentiation of the Windward islands from the other shouts . Of the West Indian archipelago.............. Lud dusdcterradd Se peelNS «biay s 259 Configuration of the islands................... Ly Rjashanl eet recente caw F6rre 260 Sang heiebis gad Oceanic depths... ... 2. nw neces ccn des menesscucnens 260 eeu MELINA CEOS 20s o's sw se hn es ib ee ve o's baheMeerae poles o's ae .. 261 Submarine configuration........ BR I chin PE a Sa ee PR eG oes 4 62 262 Era Oy WN PER PRURE tee Soa erc dese. wwinides sed isda des ¥e 264 Volcanic origin and arrangement of the material of the ielands. Ani And 265 Antiquity of the volcanoes as attested by the age of the rock material... 267 How the ocean tears away as the volcanoes build up...... Wey ease R cipats bs 269 OT 0 ee 273 a ahah cc ye chin Gd alte ttc canines eee e ges t= ae 274 Specter We ORME OL COP VOICANIOCS. 25... 5 005 c et eee cee nc ceeasseces 277 Deductions from the Windward volcanoes...........0..00ceeceeeeeennes 277 Inadequateness of crustal theories in explaining eruptions of Pelé....... 279 XXXII —Butt. Geon. Soc. Am., Vou. 16, 1904 (243) 244 . he, L—PELE AND THE wie ARCHIPELAGO ib big! blast ote ee a elle pc a a Interior theory y of vuleaniénr..:...-c..00% «++ i.ss. onc 281 Recent views on the condition of the earths interior... ..':, of solid matter is wafted away from the island piles (and this estimate is small), then it is evident that the volcanic rocks and ashes piled up around the vents and constituting the present islands, without any con- sideration of the wear and tear of surface erosion, can represent only roses Of the actual matter which came up through these vents from the earth’s interior. Hence these relatively large islands represent only an infinite fraction of the total solid matter discharged by these old vents during long eons of time. But even this little fraction must be again subdivided, for destructive forces long in operation have been tearing away almost as fast as the volcanoes could build them up. The trade winds, blowing with full severity against the eastern shore, create an enormous surf, which is constantly and continuously eating -away the shore line and cutting down below sealevel the voleano-built islands. This destructive work and its combative attacks on the grow- ing areas of land is a most instructive feature of the region. In fact, this destructive process is secondary only to that of the constructive process of volcanic pile-up, and the present aspects of the islands are chiefly a result of the endless battle between these agencies. The power of the sea to plane off in horizontal directions obstacles projecting above it is everywhere seen. Huge cubes of rock which have fallen into the sea near Bathsheba, on the Windward coast of Barbados, have been completely cut in two. The processes of the sea in cutting through these rocks by attacking them at surf line is but an illustra- tion of how larger islands have been completely reduced to submarine banks. A superb example of sea destruction is the bight known as Porte * Text-book, edition 1903, p. 266. 27? R. T. HILL—PELE AND THE WINDWARD ARCHIPELAGO D’Enfer, on the north coast of Grande Terre, Guadeloupe, where the surf of the ocean is undermining cliffs 200 feet in height (see plate 47), The eastern border of Martinique is cut into hundreds of shallow bays and inlets by the strong action of the heavy surf, which everywhere on that side prevents free navigation, and the contour and topography of the adjacent sea border shows that this land once extended to the outer margin of the islands which now border it. On the leeward shore of the islands similar work is going on all the time, although the tremendous biting effect of the surf is not so great on . the leeward shore. It is, nevertheless, sufficient to produce a well de- fined horizontal indentation into the shore, which results in the under- mining of the cliffs. The lee shore of all the islands exhibits the result of this process. If islets, such as these constituted by the Barbadian rocks, can be thus destroyed, it is not unreasonable to assume that the same processes may . reduce larger areas of land to low benches below sealevel, and if there is no further elevation or addition to the masses of the islands now pro- jecting above land these same processes can ultimately plane them all down to a level below the surf and wave action of the sea. On the windward side of all the islands to a large degree and the lee side to a smaller one there are low shallow benches and banks, which have clearly been made by this process. It is not unreasonable to be- lieve that great areas of submarine banks without any land surface, like those of the Saba banks or the great banks which extend as a submarine shoulder for 100 miles north and east of Martinique, may have once been land which has entirely disappeared largely by this process of cutting and planing by the sea. In fact, if no other. process were in operation than this one of de- structive surf planation, it alone would account for the dissection of the various archipelagoes of the Windward group arising from a common bank, such as the Virgin islands, the Grenadines, Guadeloupe, and Antigua. : That the planation has continued for a long time past as an important factor in producing the present configuration of many islands is directly shown by several examples. All the islands of the Grande Terre, Guadeloupe bank, represent an old volcano pile larger than the present island of Martinique, which was planed down to and below sealevel in Tertiary time, veneered with organic material, and afterward elevated to its present position. Such, probably, has been the case with Barbuda, Anguilla, and many other of the small calcareous islands. He who indulges in speculation would find no difficulty in conceiving that the great Saba bank owes its present configuration to the same causes, aided OSCILLATIONS OF OCEAN’S BOTTOM 273 by slight changes of level. Even the Bahamas may be similar remnants of old volcanic piles.* OSCILLATIONS OF THE OCEAN’S BOTTOM There has been still a third and less understood process at work in producing the present configuration of the Windward islands. This is the great process of regional movement (uplift and probably subsidence) of the sea bottom, together with all the islands resting thereon—veritable heavings of the bosom of the earth. | These movements are clearly and unmistakably recorded in the con- figuration of elevated reefs, benches, and terraces already mentioned. By these processes banks which once existed only below the water have been elevated into islands, and islands which showed but slightly above _ the water have been lifted still higher, and all of the islands, of whatever composition and structure, both volcanic and sedimentary, have partici- pated in these movements. Plains and terraces which originated by marine erosion at or below sealevel are now lifted into terraces and island summits, in many cases covered by growths of coral reef or shell debris. Old baselevels of ero- sion, which represented land planed down to sealevel, are now similarly lifted and again undergoing destructive degradation. Colonies of sea life, which inhabited the marginal waters of the ocean at the shore of the volcanic piles, are now found at heights of from 100 to 900 feet above the sea, resting on the great masses of the volcanic piles which have been lifted with them. Concerning the elevations, the evidence is sharp and clear, the subsi- dences are only inferential, but nevertheless probable. Owing to the slowness of these movements, it is impossible to measure them or to state with positiveness whether they are in operation at pres- ent, but that such is the case may be inferred from the recent character of the upraised benches and the unmistakable records of their long con- tinuity in the past. In the writer’s work on Jamaica, in part V, ‘‘ Changes of level in the West Indies,” attention is called to the evidence of uplift and subsidence recorded in the Greater Antilles. Darwin and others have long since described similar phenomena on the Pacific coast of South America. Writers have also shown that elevated benches and terraces conspicu- ously mark the periphery of the European Mediterranean. * Moreau de Jones, in 1816, discovered that the calcareous outer islands of the Caribbee chain rested on igneous formations. He showed that these islands were mostly situated to the wind- ward of the volcanic group, and that even in the volcanic islands, where calcareous formations were also foiind, the latter were mostly on the Atlantic side. In fact, there is evidence that the vents of volcanic activity have migrated westward slowly during geologic periods, as the sea planed away the Windward piles. 274 R. 1. HILL—PELE AND THE WINDWARD ARCHIPELAGO That these movements have been regional and not orogenic is testified by the harmonious evidence that all the islands participated in them. At present the cause of these great uplifts and subsidences can not be interpreted. They can in no way be ascribed to sudden and disastrous catastrophies accompanying the violent disturbances popularly supposed to be associated with volcanic eruptions, although we can not say that they are not gradual after effects. Associated with these phenomena are the problems of the origin and meaning of the triple submarine Windward ridges—the Aves ridge to the west, the Barbadian to the east, and the deep troughs separating them from the great central ridge from which the present Caribbee volcanoes apparently arise. | | Who can explain the mystery of these troughs and ridges and their part in the history of the Windward islands? It is a legitimate inquiry to ask if their present condition and relations, like the great changes of level noted, may not be isostatic results of vulcanism—that is, if they could have been the result of adjustment of the crust to interior vacuities created by the extrusion of the great volcanic piles. The cross-section of this tri-peninsulate configuration might be the basis of a hypothesis that the present conspicuous Caribbee ridges are piles on the axis of a subsiding dynamic valley or line of weakness between the older ridges. Such speculations present enchanting fields for reflection, but they carry no supporting data or convincing conclusion. 3 | Some writers have even gone so far as to suggest that these changes in the level have been. those of the sea water rather than of the land and bottoms. It may be thatthe great oceanic waters have changes in volume, but as yet there is no tangible evidence of it. SUMMARY It must be apparent now that the conceptions as to the continental origin of these islands, mostly created to uphold some preconceived theory, without being based on adequate geologic research and which have confused the pages of scientific literature, are erroneous. Analysis of the geologic history of these islands fails to support one of the many complicated theories which have appeared in print concerning their con- tinental origin. If they were once a part of the Andean and Antillean mountain sys- tems, a great fold or a continental bridge, now destroyed by subsidence, some evidence of these facts would have been found by investigation undertaken without preconceived prejudice or theory. A great subsidence of 6,000 feet, leaving the present islands as the tips of asubmerged continent, must be proved to have occurred in the Wind- SUMMARY 275 ward islands in late geologic time in order to give the least support to the continental theory with its mysterious drowned ridges. No evidence of such an extensive subsidence would have been found in the submarine topography. Considering all the lands and bottoms within the 1 000. fathom con- tour, it is impossible to find any evidence of former existence of a united body of land which has been dissected into fragments, as has been sug- gested. All the evidence concerning the Anegada channel, one of these alleged rivers which separates the Windward Island group from the Virgins by the 500, 1,000, and 2,500 fathom contours, from Porto Rico and the Vir- gins, is that it is one of the orogenic troughs of the Antillean system. The Saint Lucia and Martinique channels, on the other hand, occur- ring between great extrusive volcanic piles instead of being drowned rivers, are constructional valleys of coalescence between slopes of adja- cent volcanoes. It is impossible to find any evidence that a subsidence of 6,000 feet, which would have been necessary to dismember the alleged Antillean continent, has taken place or that the sea bottoms ever stood 6,000 feet higher than now, as they must have stood to make these im- aginary connections. 1 The narrowing, steep, leeward submerged profiles of the volcanic Caribbees south of Guadeloupe are constructional piles exactly similar to those of their continuation above the water and not those of old con- tinental river valleys, while the banks extending to the windward are clearly the work of marine planation proceeding in the past as it is today. The elaborate tectonic theory maintaining a structural connection between the Venezuelan, Antillean, and Windward ridges falls before an analysis of geological facts. It is impossible to connect the east and west Antillean and Venezuelan folds and faults with the main Caribbee ridge, and all the evidence shows positively that they did not connect. The only hypothetical connection which could be established between the Venezuelan, Antillean, and Windward groups would be via the Bar- badian ridge, se such a oe would require a most imaginative mind and entire disregard of facts. _ Proofs of great rock folds postulated by Milne and Anderson as a part of his “‘ macroseismic ” theory are also absolutely lacking. Instead, the Caribbee volcanic ridge apparently rises from a trough between two more ancient ridges. There is no evidence of folded structure at all except the easternmost of these, the Barbadian. It is utterly impossible to connect the volcanoes with crustal movements resulting from any sedimentary loads. XXXVII—But. Gro. Soc, Au., Vor. 16, 1904 276 R. T. HILL—PELE AND THE WINDWARD ARCHIPELAGO = These theories all disappear before the indisputable proof that the islands are simple constructional volcanic piles which have grown above the ocean floor around persistent volcanic vents, oceanic in origin, and which have never been united as a whole with each other, much less with the North American continent. It is better now to abandon these unfounded continental theories and to concentrate our attention on what the islands really are—merely vol- canic piles, as oceanic (non-continental) in origin and relations as the islands-of the Pacific. These volcanoes are the most ancient feature of the region and typify a persistent field of vulcanism which has existed as far back in geologic time as we can see in the locality, and which existed prior to the formation of any of the secondary local sedimentary rocks which now veneer the volcanic piles, and which would not have accu- mulated had it not been for the volcanic piles in the oceans. The simple fact that all the islands, excepting Barbados, are or we been old volcanic piles born of the ocean, as most yoleauees are, is the basis of the entire history of the Windward islands. Volcanic eruption, marine planation, secular upheaval, and lime-making oceanic life —these are the simple factors which have built up these islands, planed some of them down below sealevel, and lifted up the lime-covered banks into calcareous veneered islands again. Each of the Windward islands represents a chimney, past or present, of the great volcanic mechanism of the earth’s interior, where dynamic substances, usually banked in below, have now and then, through the long years of geologic time, occasionally broken forth and added strata of ashes to the preexisting surface. Hence it is that the present islands, projecting and subterranean, ever diminishing by atmospheric and ma- rine erosion, represent only infinitesimal fractions of the total quantities of material which has been erupted. In comparison with the vast work of the greater mechanism to which the volcanoes belong and the totality of the matter transferred from the earth’s interior to its exterior, the lessons inland and continent making, and the conflict between the constructive work of the volcanoes and the destructive work of the atmosphere and ocean, one can not but consider. the human events of the recent eruptions as indeed a trivial incident of the larger stcry. Here isa great amphitheater where the competitive battle of natural forces of construction and destruction have and are being fought—where vulcanism in its broadest sense is bringing up ma- terial from the earth to add to the land of its surface, to beattacked and redistributed by the exterior processes. In fine, we have in these islands the spectacle of a battery of voleanic vents, pin-holes, in the ocean’s bottom, leading up from the earth’s inte- = SUMMARY 277 rior, out of which for countless years the interior magma has been pour- ing its volumes of matter, adding material to the earth’s crust, its atmos- phere, and its oceans. In view of the great stretches of history behind them and the relative insignificance of the seemingly large products of the recent eruptions of Pelé and Soufriere, we are also naturally led to doubt those explana- tions which have ascribed the recent eruptions to ephemeral and super- ficial causes, and to inquire if there is not a deeper-seated and more permanent source of vulcanism than many geologists have been prone to acknowledge. | Beyond the roar of the land-destroying ocean surf, in the unknown interior, beneath the veneering of the ash-made islands, in the great gas- sodden atmosphere above us, back of all the phenomena described, there is always presented for solution that still greater problem, What made the volcanoes ? DiscussIoN AS TO ORIGIN OF THE VOLCANOES DEDUCTIONS FROM THE WINDWARD VOLCANOES The facts presented in the previous pages show that the recent erup- tions of Pelé were not sudden nor produced by accidental superficial causes, such as the letting in of oceanic waters through fissures, but, on the other hand, they were a gradual emanation from an ancient, per- sistent deep-seated volcanic center which has existed beneath the vicinity since remote geologic periods. The facts have also shown that these volcanoes, from a world-making standpoint, are not destructive engines, but, on the other hand, are great constructive agents building islands in the sea, adding mass to the pre- existing lands and volume to the atmospheric waters. Inasmuch as these favts do not accord with the current popular ideas of volcanoes or those in most text-books, we feel that this paper would not be complete without a résumé of current theories of vulcanism and the presentation and correlation of some recent and rather revolutionary views thereof, which have been presented by various eminent men of science. These views, while not pretending finality, are at least much more in accordance with geological facts than many that have collectively ap- peared hitherto, and will explain more satisfactorily the events of the recent eruptions of Pelé than the theories which are more popular. A volcano is the terminal exhaust pipe of a concealed underground mechanism. Our geological studies of volcanoes are limited to the cones and craters which represent the outer rims of the exhaust pipes of these 278 R. T. HILL—-PELE AND THE WINDWARD ARCHIPELAGO mechanisms leading up from the unknown interior of the earth, and even these studies must be carried on at a respectable distance when the vol- cano is at work.* The chief result of studies under such unfavorable conditions has been to ascertain that the vent holes discharge at the earth’s surface nearly every known chemical element in various combinations and forms— solids, liquids, and gases. These materials in the condition in which they reach the surface—for they convey but little idea of the conditions within the interior from which they have evolved—represent exhausted and completed products of nature’s workshop within. The concentration of our studies on these erupted products, together with the security in which we inhabit the earth’s surface, has led many to regard the world as a completed or finished object, while in fact it is a living mechanism containing within itself forces and material capable of great chemical work like that so obvious in the sun and some other’ stellar bodies. Thusitis that purely geological studies of voleanoes have resulted chiefly in the presentation of numerous and confused systems of classification of the apparently infinite varieties of igneous rocks, which all students now admit to be but the varied manifestations of the one great material substance of the interior of the earth. It is at least dis- couraging to learn, after twenty-five years of collecting, slicing, and micro- scopically studying volcanic rocks, that petrographers have recently de- creed that “ there are no well defined chemical groups of rocks, but rather a continuous series with no natural divisions,” t thus admitting that the various forms of crystalline rocks as we see them in their cooled condi- tion are but the differentiated products of a mother substance from which they have all evolved. The geologist, notwithstanding all his studies of the earth’s crust, its marvelous history and rearrangement of material, likewise admits that the great changes of level, such as the upheaving and subsidence of land and continents, are explicable by no known superficial agency, and real- *The field and laboratory geologists’ opportunities for interpreting voleanic phenomena have many limitations. Their observation of the vital phenomena are as restricted as those of a man who endeavors to ascertain the great reactions which take place within a roaring furnace by dis- tantly observing the escaping smoke from its cupola and studying the cold slag of its dump pile. They can only see the superficial and relatively secondary phenomena of voleanoes—the top of the chimney—for nature has never exposed the deeper internal mechanism to view, or, at most a shallow depth of what was formerly the underground portion of dead voleanoes, frequently exposed by erosion, showing a cast, so to speak, of some of the old rock material. Traces may also be seen of what are called the ‘expiring after effects of vuleanism,” such as dikes which push upward into the old cone or overburden as the cooling vents clogged, or deposits of copper, silver, gold, sulphur, and other metals which have either evolved from the great proto- magma during the expiring after moments, and filled veins and fissures, or which were differen- tiated like the other elements of the rocks from the cooling magma into distinguishable ore segre- gations (pegmatitic) before being still further collected into more concentrated ore bodies by the subsequent circulation of atmospheric waters. +Iddings, 1904, oie DISCUSSION AS TO ORIGIN 279 izes that there are planetary processes within the earth’s interior beyond his power of solution. Geikie* confesses that “‘ not even a satisfactory solution of the problems of the upheavals and depressions of the land has been given. When we consider the wide tracts over which terres- trial movements are now taking place or have occurred in past times the explana- tion of them must manifestly be sought in some far more widespread and generally effective force in geological dynamics. The causes of upheaval and depression of land must again be traced back mainly to consequences of the internal heat of the earth.” _ Thus it is that the results of the purely geological study of the outer phenomena of vulcanism, the examination of volcanic craters and rocks, notwithstanding their long years of research and the great value from other points of view, so far as their contributions to knowledge of the conditions of the earth’s interior are concerned, are incomplete and unsatisfactory, and science must turn to the physicist, the mathema- tician, and the astronomer for aid in investigating the earth’s interior. INADEQUATENESS OF CRUSTAL THEORIES IN EXPLAINING ERUPTIONS OF PELE Owing to the fact that the geologist’s observations are limited to the phenomena which he sees on the earth’s surface, there has been a tend- ency to explain all the phenomena of vulcanism by the obvious ex- terior processes. These theories, with their multiplicity of variations, all try to associate volcanic phenomena with movements of the earth’s crust, and some of them limit the causes and phenomena of vulcanism to a narrow zone of the earth’s outer diameter. All hypotheses of this class may be grouped under the head of crustal theories. The recent West Indian eruptions resulted in the attempted applica- tion of many of these crustal theories to the incidents in that region, and it was the writer’s studies of the geology of the islands and the im- possibility of fitting the facts to the theories that has led to the writing of this paper. The various crustal theories differ in detail. The most popular and generally accepted of them is that volcanoes represent extrusions of gases and molten rock from the earth’s subcrustal layers. Surface fis- sures are supposed to permit the entrance of surfical waters to the melted rock, creating an expansive force, which extrudes the material. Some advocates even maintain that the causes of the molten condition them- selves are crustal, involving the melting of the rock through crustal movements. Some maintain that the source of heat may be the load of the crust; others admit that it is the general heat of the earth, but require the in- Jetting of surface waters to produce extrusion and explosion. Nearly all * Text-book of Geology. 280 R. T. HILL—PELE AND THE WINDWARD ARCHIPELAGO maintain that predetermined fissures are essential for volcanic extrusion and inletting of water, and all crustalists postulate that the water of erup- tions is derived from crustal sources. Mallet maintained that all the present manifestations of hypogene action are due directly to the more rapid contraction of the hotter in- ternal mass of the earth and the consequent crushing in of the outer coolershell. “ Thesecular cooling of the globe,” he remarked, “ is always going on, though in a very slowly descending ratio. Contraction is there- fore constantly providing a store of energy to be expanded in crushing parts of the crust, and through that providing for the voleanic heat.” Professor N.S. Shaler has also advanced a crustal ‘‘ blanketing theory,” in which he accounts for the volcanic heat by continuous accretions of sediments or load. The theory of the “ fissurists,” who maintain that volcanic action itself is due to the letting in of water through fissures to the hot magma, has been expressed by Dana, who states : * ‘‘ For expansive eruptions, water in large quantities must gain sudden access to the interior of a magma conduit, for the projectile force of the abruptly generated vapors is enormous.”’ Russell,f in his excellent work on North American volcanoes, while admitting that the initial expansion of the magma forced it up into the subterranean fissures of the crust, postulated that the actual eruptions at the surface required the inletting of surface waters. An exposition of the crustal theory, in its extreme development and application to the West Indian eruption, was recently set forth by Pro- fessor Milne in the Journal of the Royal Geographical Society of London for January, 1903. | Briefly stated, Milne’s theory is that dormant volcanoes in a state of volcanic strain may be brought into activity by a mass displacement of a fold from which they rise. This displacement may occur a thousand miles or more from the site of the volcano. For instance, he alleges that the eruption of mont Pelé in 1851 was preceded by a great earthquake in Chile, fully 2,000 miles distant. According to this author (but not so in nature), nearly all active vol- canoes occur along the ridges of rock folds which are in proximity to oceanic waters. By the percolation of this water to the foundations of these folds, where it comes in contact with a heated magma, extraordi- nary pressures are developed, the sudden relief of which results in a vol- canic outburst. * The term ‘‘magma”’ is used by geologists for the molten igneous volcanic fluid which differ- entiates into various kinds and forms of crystalline rock under physical conditions of pressure, cooling, etcetera. +1. C, Russell; Voleanoes of North America, p. 308, Se DISCUSSION AS TO ORIGIN 281 . . . A good illustration of this relationship between sudden movements of rock folds and displays of volcanic activity is to be found in the history of the vol- canic eruptions in the West Indies and the large earthquakes which have occurred in the West Indies or in adjacent countries. As soon as a fracture occurs pressure is relieved, the deep hot rocks become fluid and are forced up the fissures by the weight of the crust.” Unfortunately for this theory, there is absolutely no indorsement in nature to its basic proposition that “nearly all active volcanoes occur along the ridges or rock folds which are in the vicinity of ocean waters.” In this connection it is interesting to note that M. de Montessus de Ballare,* from careful field studies reaches a conclusion opposite to those of Milne. He says: ‘It is very remarkable that the distribution of seismic instability in all degrees of intensity presented, in all possible combinations with the presence or absence of volcanoes, in their activity or their extinction, affirms at once a most complete independence in time and space of the two orders of phenomena.”’ The fissure theory has also been directly applied to Pelé by Jaggar ft as follows: 7 ‘A slip of some sort liberated a steam column; the cause of the fracture or the source of the steam is one step too far back into the theory to venture to treat it here. Release once started followed old vents, water holes, and these vents were Soufriere and Pelé. The explosion that followed release of pressure tore away the walls of the fissure and its violence ground the material topowder. The material came from a depth where the rocks were hot, and it was heated further by friction.” INTERIOR THEORY OF VULCANISM Recent views on the condition of the earth’s interior.—Of late years broader thinkers have realized more and more that the problems of the earth’s interior could not be interpreted merely by geological study of its sur- face conditions, and that they required the assistance of the laws of chemistry, physics, and mathematics, by which we obtain our knowledge of the sun and other kindred heavenly bodies, of which our planet is one.{ Great physicists, like Sir William Thompson, now Lord Kelvin; George Darwin, Newcomb, and others, by mathematical processes grad- ually destroyed the earlier hypotheses concerning a fluid condition of * Investigations on the earthquakes in the region of the equatorial Andes. Academie des Sciences, January 11, 1904. + T. A. Jaggar: Popular Science Monthly, August, 1902, vol. Ixi, p. 365. ft Major J. W. Powell, while Director of the United States Geological Survey, saw the importance of physical research in connection with the interpretation of the earth’s interior, and at one time secured from Congress an appropriation to conduct experiments with high-pressure temperatures under the direction of Dr Carl Barus, now professor of physics at Brown University, but the in- vestigation was too deep and far réaching for the appropriation committee to apprehend. The little work that Professor Barus did, however, before his allotment was expended remains as practically the only American attempt to rationally study the condition of the earth’s interior. 282 R. tT. HILL —PELE AND THE WINDWARD ARCHIPELAGO the earth’s interior. Asa result, and opposed to the crustal theories of origin of volcanoes,* these new ideas are focusing around the belief that volcanoes and their products are manifestations of forces and material existing within the interior of the earth itself, held by some to be neither liquid nor solid, but gaseous, and which by expansion has the power to ascend through the earth’s crust and to produce by differentiation under. ' different physical conditions of pressure and cooling all the known sub- stances accompanying volcanic phenomena—rocks, metals, gases, and water. . These theories have recently been strengthened by four distinct lines of research, to wit, physical studies bearing upon the conditions of the earth’s interior; practical researches in deep mines by various eminent geologists in relations to the origin of ore deposits; Professor Suess’ conclu- sions as to the interior origin of atmospheric waters; Sir William Ramsey’s investigation of gases exhaled by the earth. The writer’s own personal unpublished investigations of the geologic history of the West Indian and Central American and Mexican volcanoes have at least led him to believe that the interior theories are much nearer the true explanation of vulcanism than those of the crustalists. eel In America interest as to the nature of the earth’s interior has been principally kept alive by a practical, rather than the merely theoretical, line of research. Mining methods have advanced beyond the old axiom of “‘ follow the ore” to the scientific stage where the modern miner em- ploys geologists to explain and “hunt the ore” by studying laws of origin and occurrence of mineral deposits. The widespread scientific study of the origin of ore deposits has resulted in views concerning the earth’s interior far in advance of those maintained by the purely aca- demic geologist. Men like Kemp, Lindgren, Weed, familiar with deep mines as well as theoretical geology, whose researches the writer could largely supplement by his personal observations of the great mines, realized the intimate relation of metallic ore bodies to the great interior processes of the earth. Arrhenius’ theory of a gaseous center.—In a volume published by the Institute of Mining Engineers on the subject of ore deposits a few years ago Vogtt quoted some paragraphs from the writings of Professor Arr- henius, the eminent Swedish physicist, who presented for the first time in this country an intelligible hypothesis of the conditions of the earth’s interior and its contributions to the crust, which alike satisfied the re- * Even the crustalists like Dana admit that ‘igneous action has its origin almost exclusively within the earth’s interior,” p. 556, while it is generally acknowledged that deep-seated rocks on account of their high temperature are in a potentially plastic or even gaseous condition, which may become plastic on the relief of pressure. + Transactions American Institute of Mining Engineers, vol. 18. a °° eee Rnd . ne eg a ra ae = i : ae _ . i it i i i i, fi ig a ~~ Se DISCUSSION AS TO ORIGIN 283 quirements of the mathematical physicists and explained far better, at least, than the heretofore superficial phenomena of vulcanism. Professor Arrhenius’ conclusions thus cited were that the crust of the earth is solid to a depth of about 40 kilometers where there is a temper- ature of about 1,200 degrees centigrade and the pressure of about 10,840 atmospheres—that is to say, at this depth begins a liquid molten con- dition. Beyond that, 300 kilometers, the temperature must, without doubt, exceed the critical temperature of all known substances, and at this point the liquid magma passes gradually to a gaseous magma sub- ject to extremely high pressure. The viscosity and lack of compress- ibility may be greater than those of the liquid magma. ‘These liquids and gases possess a viscosity and incompressibility such as to permit them to be regarded as solid bodies.* This great interior, which is the foundation of Arrhenius’ theories, with its celestial temperatures, and. which its author considers as gaseous, is explained as being “ something wholly different from what we ordinarily understand as gases.” + By experiment and deduction we know that all the rocks of the earth’s crust can be melted into liquids and at still higher temperatures converted into gases. It is also known that hot gases can cool into molten liquids, which in turn change into solids on further cooling. It is not illogical to suppose that these processes are repeated, and that from a greater and more primitive gaseous protomagma, as postulated by Arrhenius, all the material of the earth’s crust, which is all secondary matter, with its variety of forms and conditions, have been evolved. Granting that matter in the earth’s interior does exist in a gaseous or potentially gaseous condition, and remembering that these gases are com- posed of all known elements of the earth’s substances, the mind can also conceive that on escaping to the cooler surface these gases as they ap- proach the outer crust and atmosphere will be gradually and success- ively converted into all known primary forms of minerals, water, and gases as they exist on the crust today, first condensing into liquids and then into solids, producing exactly the conditions seen in the workings of a volcano, which, as deeply as we can see into it or its roots as ex- posed by erosion, is merely the cooling superficial crustal terminus or conduit of a cooling gas column leading from the greater invisible depths to the surface. *“ Zur Physik des Vuleanismus.”’ Geol. Foren i Stockholm Forhandl., xxii (1900), pp. 895-419. + At this point we may be pardoned for suggesting a new term which will assist in the discus- sion of the interior theory. In Arrhenius’ statement it will be noticed that he recognizes three zones of condition—the outer crust, the molten inner layers, and the great gaseous centrum, Theterm “crust” isan accepted one. The liquid massis the magma. For the matter of the great interior centrum from which the liquid magma and all the known substances and the superfice have theoretically differentiated no name is given. For the want of a better term this great protoplasm of the inorganic world might be called “‘ protomagma.”’ XXXVIII—Butt. Grou. Soc. Am., Von. 16, 1904 e 284 R. T. HILL—PELE AND THE WINDWARD ARCHIPELAGO This gaseous theory of Arrhenius, so revolutionary to previous concep- tions of the earth’s interior, and which hitherto has been recognized in America only by the students of mining geology, has lately received the -approval of Sir Archibald Geikie. This recognized leader of geological thought, in the newest edition of his ‘‘ Text-book of Geology,” 1903, re- ceived while the present paper is being written, generously states that ‘‘For some of the latest views regarding the nature and origin of volcanic action we are indebted to Professor Arrhenius, of Stockholm, whose observations on the probable condition of the earth’sinterior have been already cited, and who, bring- ing the results of modern physical and chemical research to a consideration of the subject, confirms what has been the growing belief on the part of geologists in regard to this part of their science. . . . The aspect thus presented of the probable constitution of the interior of our planet appears to accord well with the geological requirements. Not only does it furnish an explanation of the charac- teristics of earthquake movements, but, as Professer Arrhenius cogently shows, it helps us to understand some of the more difficult problems of volcanic action.” | Thus Arrhenius gives science the first tangible hypothesis with which to combat the assertion of the crustalists that the heat and materials of vulcanism are generated within the subcutaneous layers of the earth by crustal load and movements. We can now see that within the earth’s centrum is contained matter so intensely hot that its temperature may be classed as celestial. This matter contains all the elements of the crust metals, gases, and rocks; possesses the potentiality to escape from the interior through the crust to the surface, and the power to assume an infinite variety of forms, combinations, and conditions as it ap-— proaches and reaches the surface. Source of the water of vulcanism.—W hile Arrhenius’ theory has given us a more reasonable working hypothesis than any hitherto possessed con- cerning the interior, the source of water of vulcanism still remained the bone of contention. The crustalist views largely centered around the belief that the water of volcanoes is superficial and admitted to the magma from above, thereby creating the steam of expansion and explosion. Thoughtful investigators have also lately been finding many reasons for disbelieving that the inletting waters of the oceans have been the exciting cause of vulcanism, and have boldly suggested that the water of volcanoes, instead of being contributed by the oceans, is derived from the gases of the earth’s interior. They have even inquired if the oceans— the great aqueous envelope of the globe—have not been made by con- tributions from the condensation of volcanic gases from its great cooling, shrinking interior protomagma. An objection long since pointed out by Geikie to “ the constant influx of water from the surface is the difficulty of conceiving that water should descend at all against the expansive force within. Experience in deep SOURCE OF WATER OF VULCANISM 285 mines, however, rather goes to show that the permeation of water through pores of rocks gets feeble as we descend.” Still another argument against the water of vulcanism being derived from the surface has been the fact that the volcanic rocks were largely composed of water-making gases, which entered into combinations far below the surface and under conditions of temperature where surface water could not exist. It is well known that the crystals of deep-seated igneous rocks contain gas and liquid-filled cavities, due to the presence of gas or steam in the crystal at the time of consolidation. The usual gas is hydrogen, with traces of oxygen and carbon-dioxide. Sometimes it is entirely carbon- dioxide or hydrogen and hydrocarbons; the liquid cavities are usually filled with water in which carbon-dioxide may be present. In most of _ these cases the liquid inclusions are to be referred to the conditions in which the mineral crystallized out of the original magma. The fact that granite, a deep-seated magma, when heated to 1,000 de- grees centigrade was found by Gautier to give off more than 20 times its own volume of gases and 89 times its volume of steam as vapors shows that under the conditions of original solidification these apes were present in the deep-seated interior. Professor Kemp quotes him as follows: ‘If we give due weight to the expansive power which these rock gases must develop whenever the pressure upon the heated rock in the interior of the earth permits, we see that the old theory of the production of volcanic outbreaks by the introduction of water isno longer necessary. By astill strongerignition the volume of the emitted gases appreciably increases. . . . When one realizes the ex- plosive power which this implies, one may dismiss the introduction of surface waters into the glowing reservoirs of rock from the theories of volcanic action.” * **The abundant occlusion of hydrogen in meteorites and the capacity of many terrestrial substances, notably melted metals, to absorb large quantities of gases and yapors without chemical combination and to emit them on cooling with eruptive phenomena, not unlike those of volcanoes, have also led some observers to conclude that the gaseous ejections at volcanic vents are portions of the original constitution of the magma of the globe, and that to their escape the activity of volcanic vents is due.” Thinkers who took the view that the water as well as the rocks were protomagmatic did not make much impression against the crustal vul- canologists, however (judging by the way in which crustalists continued to pour in the ocean into the roots of the recently active West Indian volcanoes). until recently when they found a new champion in Professor Suess, of Vienna, who, with Geikie, stands preeminent among the world’s greatest geologists. * Zeitschrift fur Practische Geologie, October, 1901, p. 303: cited by J. F. Kemp, Trans. Amer. Inst. M. E., October, 1903. 286 R. T. HILL—PELE AND THE WINDWARD ARCHIPELAGO Breaking away from the old traditions, this eminent leader of geolog- ical thought during the past year has boldly advocated that the sources of the waters of hot springs and volcanoes are internal and not superfi- cial, and that the waters of these are distinct contributions from the earth’s interior to its exterior. He says:* **The steam of the volcano cannot be derived from surface infiltration ; for, if it is, whence the carbonic acid? Both must come from the deeper regions of the earth ; they are the outward sign of the process of giving off gases which began when the earth first solidified, and which, today, although restricted to certain points and lines, has not yet come toa finalend. It is in this manner that the oceans and the whole surface hydrosphere have been separated from the solid crust. Volcanoes are not fed by infiltration of the sea, but the waters of the sea are increased by every eruption.” THEORY OF LINEAR ARRANGEMENT ALONG FISSURE LINES Another position of the crustal theorists has been the assertions as to the linear arrangement of volcanoes along certain great lines of imagi- nary fissures in proximity to the sea, the actual known occurrence of some volcanoes along fissures, and the presumption that the volcanoes could not have within themselves the power to reach the surface with- out the preexistence of these fissures through which the water was let in- It is now found that even the data on which these deductions are based are not conclusive. Volcanoes do and have occurred both on the land far away from the sea and in the sea far away from the land. So far as the occurrence of volcanoes adjacent to the ocean margin is concerned, it may justly be argued that this arrangement is in part the natural sequence of occurrence along lines of least resistance and partially due to the fact that these marginal land volcanoes are merely the fringe of the greater area of marine volcanoes concealed by the ocean. Indeed, increasing evidence is accumulating that the greater number of volcanoes, past and present, are marine and not accompaniments of the land areas, but really makers of the land areas, the continental interiors representing areas where in times past the crust has already been so thickened by vol- canic extrusions as to prevent egress of volcanic gases in those localities. While admitting that volcanic protrusions naturally may follow pre- existing lines of weakness, such as faults and fissures, the fissuring also usually follows the volcano. Instances are even found in the San Fran- cisco and Mount Taylor regions of volcanoes far distant from oceanic waters without a trace of preexisting fissures where the magma has forced itself up through thousands of feet of sedimentaries. * Professor Eduard Suess, Royal Geographical Journal, vol. xx, November, 1902, p. 520. +In this connection see the excellent chapter on submarine volcanoes in the new edition of Geikie, vol. i, pp. 332-342, and his proof of the conclusion that “ volcanic activity is displayed over a wider region of the ocean’s floor than on the surface of the land and on a more gigantie scale.” LINEAR ARRANGEMENT AND MAGMATIC HYPOTHESIS 287 In 1897 Geikie showed that the volcanic rocks of central Scotland “ had been blown out without any trace of their coincidence with lines of fault.” Bronco, in Schwabia, and Boese, in Mexico, have shown similar condi- tions. As we write these lines we read* that the sedimentary rocks through which the remarkable row of 19 old volcanic necks penetrate on the cape of Good Hope “ never show the least shattering or plication.” It is much more reasonable to assume now that volcanic forces have within themselves the power to penetrate to the surface. If the interior of the globe, as Arrhenius and other physicists hold and as cited by Vogt,t is vast masses of highly superheated and compressed gases, these gases must ever be seeking release through any porosity or even by their own solvent powers. Ascending through the subcrust either from long cool- ing and condensation, the contact with the vadose circulation may have - converted into ascending springs, the hot water of which possesses the powers of solution and erosion sufficient to create a vent. To assume that the conditions of the highly heated interior of the earth are impor- tant is by no means a secure conclusion. THE MAGMATIC HYPOTHESIS The magmatic hypothesis not only conforms more nearly with our geological facts than the crustal, but with the greater astronomical his- tory of our globe, which, when considered, invariably leads back to the hypothesis that the earth at one time in its astronomic history was an undifferentiated magma of hot gases, like the sun. The present interior of the earth, as a sequence of its history as a cooling globe, must still inherit some of its ancestral conditions and all of the present materials of the crust must be the ultimate products of a great mother magma, the remnant of which, outwardly encrusted, still constitutes the interior, with inherited celestial temperatures. Since the year 1900 there have been important additional contribu- tions from separate lines of research, which collectively give courage to those who have never been able to accept the crustal theories. These contributions all tend to uphold the theory that the interior of our globe is a living field of potential activity, from which matter is con- stantly forcing its way in a gaseous state through the outer crust, ocean and atmosphere. In fact, the dead-planet view dissolves on considering some of these great processes going on within, and there is far more rea- son to believe that it is still a live, heated mechanism, containing be- neath the outer crust or shell a great interior magma, with all the mighty forces and activities of uncombined chemical elements. * Report Geological Commission, Cape of Good Hope, 1900-’02. + Zur Physik des Vulkanismus (Geol, Foren, Forh.), Stockholm,*1900. 288 Rk. T. HILL—PELE AND THE WINDWARD ARCHIPELAGO Evidence is rapidly accumulating that the earth is constantly exud- ing its interior substance as gases into space, and that its interior is a great reservoir of material and forces from which invisible matter is escaping. Sir William Ramsey has just announced that the supply of helium, an element only recently discovered by Raleigh in the atmos- phere and which is known to exist in the sun, is constantly passing from the earth to the atmosphere. Itnot only comes up with the hot springs, but presumably oozes from the soil, and the quantity thus escaping is from 3,000 to 6,000 times more than can be accounted for as a return to the atmosphere of helium washed down by rain. In the light of the interior theory, vulcanism in its simplest concep- tion may be theoretically considered as the transfer of matter from the earth’s interior to its exterior in the proportion of about 1 part of solids to the crust and 99 parts of gases to the atmosphere and ocean. The enormous masses of crystalline rocks now found on or near the surface of the earth (and we have no language to express its weight in tons or dimensions in cubic feet) represent secondary material differentiated from the interior gaseous protomagma. These rock masses which now appear to us as solids may be merely gases which have been locked into conditions of stability by chemical and physical combinations. Reduced to the simplest statement, the interior hypothesis of voleanoes is that they are gas vents, and volcanic rocks are their by-products. In view of the facts presented, may we not weigh with consideration the words of the venerable Professor Suess, of Vienna, one of the world’s greatest geological thinkers, who has lately stated that ‘‘voleanoes are not fed by the infiltration of ocean water, but ocean water received additions to its volume by every (volcanic) eruption. . . . The hottest dry fumaroles, forming deposits of the ore by sublimation, the rain of hydrochloric acid from Vesuvius and the salt of the Altenzalza mines, the hot vapors which recently burned the bodies of the unfortunate victims at Martinique without set- ting fire to their clothes, and the warm, healing waters which rise up here before our eyes are members of one individual series of phenomena. The earth is still giving off gases in a manner which may be compared to what we observe in the spots on the sun or on every large mass of cooling steel.’’ * The traditions of geology have held the majority of its students to the crustal theories, but beyond the rank and file there are a few larger minds who see greater causes than the simple testimony of the cold, dead crust. These constitute the nucleus of the modern school of geology, which, with the aid of the physicist, the chemist, and the mathematical astrono- mer, will ultimately solve the great problem of the physical nature of the basic primitive matter of the earth’s interior, from which all known crustal substances have been evolved through volcanic action. *E. Suess: Hot springs and voleanic phenomena. The Geog. Jour., London, vol. xx, 1902, p. 522, BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 289-328, PLS. 48-64 MAY 27, 1905 PIEDMONT DISTRICT OF PENNSYLVANIA * BY F. BASCOM (Read before the Society December 30, 1904) CONTENTS Page Geography of the Piedmont... .... ..-........ Lah eT ee E bath) edn santa woe 290 EE ARUN see One Sera Te | Ene arene oe a 290 SERIO he SoS ug le a tie Gy lid lee, cis eee cave dee eee «ae 320 Pesmeatitiess: 61 ivae tiede So See Seas Was po eud va amcwdtes Soe 322 Geologic history of the Piedmont district of Pennsylvania............ ...:- oan PRESUMES Sos a occ Anand seb en oo awa ca tex epud ae mle a 2 ales Oe ee 327 AOR as ninis mye ood wine a's we berate arelely & mais his Sk moe ek oe 327 GEOGRAPHY OF THE PIEDMONT GENERAL RELATIONS _ Of the three physiographic provinces into which the Atlantic border region is divisible—the Appalachian district, the Piedmont plateau, and the Coastal plain—the Piedmont plateau preserves the record of the longest and most varied geologic and physiographic history. The plateau lies at the southeastern foot of the Appalachian system and is separated from the Atlantic ocean by a belt of coastal plain of variable width and from the edge of the continental plateau by a belt of coastal province possessing a uniform width of 200 miles. The Piedmont district extends north and south from Maine to Ala- bama, with an average width of 50 miles. Its western limit is defined by the eastern slopes of the Blue ridge; its eastern boundary is defined by an equally conspicuous change in topography, the abrupt transition from a diversified upland to an undiversified lowland. Eastward from this boundary the navigable streams, opening into tidal estuaries, afford good shipping facilities. Westward the streams cease to be navigable and occupy rocky channels. At the head of navigation and on the boundary between plateau and plain are situated many of the large cities of the Atlantic states. In general the plateau is a level upland of moderate elev ation, sloping GEOGRAPHIC RELATIONS 291 east and southeast, with shallow valleys and scattered residual emi- nences. South of the New England states these eminences do not rise -above a height of 1,600 feet, while the plateau varies from 200 feet in the east to 1,200 feet on its western edge. The level lines of the plateau are unrelated to the underlying rock structure. The larger streams, which have cut into the plateau, converting it into a diversified upland, have maintained courses which are independent of the structure and character of the rock floor. The tributary or subsequent streams, on the other hand, are adjusted to the rock floor, and by means of them the heterogeneity of rock constitution and complexity of rock structure is finding expression. The upland is covered by a thick mantle of fertile soil, comparatively free from stones and exposing few rock ledges. LOCAL RELATIONS The Piedmont district of Pennsylvania, forming with a width of 65 miles the southeast portion of the state, is an important part of the Atlantic Piedmont. In geology and physiography it is an epitome of the larger district. Geologically the Piedmont district is a complex of highly metamor- phosed sedimentary and intrusive igneous materials of pre-Paleozoic and Paleozoic age largely concealed beneath a cover of unmetamor- phosed sandstones and shales and unconsolidated gravels, clays, sands, and marls of Mesozoic and Cenozoic age. The pre-Paleozoic and Paleo- zoic crystallines are exposed on the southeastern and northwestern bor- ders of the Piedmont district and are covered in the central portion of the region by the Triassic sandstone and shale series. It is the purpose of this paper to discuss the formations and structures of the southeastern exposure of crystallines. This belt trends northeast and southwest, with a width, in the region of Philadelphia, of 20 miles. It widens southwestward and pinches out northeastward, where it disappears under a cover of ‘Triassic, Cretaceous, Tertiary, and Quaternary material and emerges again at the surface in northeastern New Jersey, southeastern New York, and in New England. The valley of the Delaware is the southeast boundary of this belt. From this valley, which is less than 20 feet above sealevel, the upland gradually rises, reaching, at a height of 160 to 180 feet, the base of a well defined escarpment. This escarpment, which is marked by the 180 and 200 foot contour lines, extends from Somerton on the northeast south- westward to Gordon heights. Fox Chase summit, Green Lane reservoir, Swarthmore College buildings, and the Chester reservoir are located on its crest, from which the relatively flat slope to the Delaware can be surveyed. 292 F. BASCOM—PIEDMONT DISTRICT OF PENNSYLVANIA Northwest of the escarpment a more rugged topography prevails. The upland rises more rapidly, and 10 to 14 miles northwest of the Delaware elevations of 400, 460, and 500 feet are reached. These elevations mark- the level tops of hills, which trend northeast and southwest, and which constitute a more or less well defined topographic feature known as Buck ridge. | In the central portion of the belt Buck ridge is separated by a shallow valley (“Cream valley’) from well.defined hills rising to a height, at Paoli and Devon, of 520 feet. These hills are known as the south Ches- ter Valley hills. Beyond them lie Chester valley, with a width of 2 miles, and the north Chester Valley hills, also trending northeast and southwest and rising to a height of 600 feet. The contours of these hills and of Chester valley are controlled by the underlying Paleozoic formations. To the northward these materials pass under a cover of Triassic shales. In the northeast Buck ridge and the south Chester Valley hills are merged ; Chester valley is first reduced in width and then disappears and Buck ridge itself passes under the Triassic cover. With the presence of the Trias there appears a marked change in to- pography; the marked northeast-southwest trend of ridge and valley disappears; well defined ridges do not exist and the range of elevation. decreases. The more open, level character of the district underlain by Triassic formations is conspicuous from the summit of the north Chester Valley hills, which separate the country of the Paleozoics from that of the Triassic formation and command both districts. In the southwest the belt of pre-Paleozoic and Paleozoic materials widens, passing into Delaware and Maryland with a width of more than 80 miles and preserving the southwest trend of ridge and valley. The south Chester Valley hills and Buck ridge again merge. Chester valley maintains its integrity until within 8 miles of the Susquehanna river, where the topography abruptly alters by reason of underlying harder formations coming to the surface. GEOLOGY OF THE PIEDMONT GENERAL GEOLOGY In the Piedmont of western New England, Pumpelly, Dale, and Wolff* have determined the following succession : Silurian: Berkshire schist. Cambro-Silurian: Stockbridge limestone. Cambrian: Vermont quartzite. Pre-Cambrian: Stamford gneiss. * Monograph xxiii, U. S. Geol. Survey, p. 13. GEOLOGICAL SUCCESSION 293 For the Piedmont of southeastern New York, Merrill* gives the fol- lowing succession : Silurian: Hudson schist. Cambro-Silurian: Stockbridge dolomite. Cambrian: Poughquag quartzite. Pre-Cambrian: Fordham gneiss. In the Maryland Piedmont, Mathews Tf has determined the following series : Silurian: Peach-bottom slates, Cardiff quartz BOs cept Wissahickon mica-schist, including an upper phyllite member. Cambro-Ordovician: Cockeysville marble. Cambrian: Setters quartzite. Pre-Cambrian: Baltimore gneiss. In the District of Columbia, in Virginia, North Carolina, and Ten- nessee, Keith { includes the mica-schist, mica-gneiss, and granite-gneiss under the division: Pre-Cambrian or Archean Carolina gneiss. The Pennsylvania Piedmont shows a similar series of arenaceous, cal- careous, and argillaceous sediments. SEDIMENTARY SERIES OF THE SOUTHEASTERN BELT This belt contains the full series of sedimentary formations found to the westward in the Pennsylvania Piedmont. The original character and structures of these formations, however, have been rendered more obscure than is the case with the western belt by greater metamorphism, due both to dynamic forces and to the injection of great masses of acid and basic igneous materials. The sedimentary series has been completely recrystallized and indu- rated ; the argillaceous sediments have been converted into gneisses and schists; the arenaceous sediments into quartzite and quartz-schists ; the calcareous material into marble. The dynamic forces which have metamorphosed the sediments were generated by tangential thrust in southeast-northwest direction, which, combined with gravity, has also produced longitudinal folds, striking northeast, cleavage normal to the compressive force, and fissility both normal and diagonal to the force. * New York City Folio, no. 83, U. S. Geol. Survey. +The American Journal of Science, vol. xvii, February, 1904, p. 143. t The Washington Folio, no. 70, U. S. Geol. Survey. The Asheville Folio, no. 116, U. S. Geol. Survey. 294 Fs BASCOM—PIEDMONT DISTRICT OF PENNSYLVANIA The igneous intrusives cut across the belts of sedimentary material more or less irregularly, but in general show a northeast-southwest trend. The sedimentary series of the Piedmont complex fall into four groups: Ordovician: Wissahickon mica-schist and mica-gneiss. Cambro-Ordovician: Chester Valley limestone. Cambrian: Chickies quartzite. Pre-Cambrian: Baltimore gneiss. PRE-CAMBRIAN ROCKS: BALTIMORE GNEISS Distribution.—This formation appears in the central portion of the belt and traverses it obliquely to the northeast. Itis very thoroughly injected by gabbro, which, with the Baltimore gneiss, constitutes the flat-topped highland known as Buck ridge. North of Chestnut hill the gneiss belt becomes very narrow, though still persistent and expanding again to the northeast. It is bounded by faults throughout most of its extent. The formation is best exposed on the east bank of the Schuylkill between Lafayette and Spring Mills. In Cecil county, Maryland, the Baltimore gneiss occurs in a narrow, wedge-shaped area (1 mile by 3%) on the Susquehanna river 2 miles northwest of Port Deposit. It expands southwestward in Harford county. Character of the formation and stratigraphic relations—The Baltimore - gneiss is a medium grained, thoroughly crystalline aggregate of quartz, feldspar, and biotite, and is characterized by a pronounced banding, which may be very fine and intensely plicated. A gritty feel and a pseudo-porphyritic texture further characterize the formation. To the alternation of layers: of biotite with quartz layers or quartz-feldspar layers is due the finely gneissic character of the rock. Biotite occurs in minute plates and is never developed in such dimensions or in such excess as to render the formation schistose. Associated with the biotitic layers are hornblende, epidote, titanite, garnets, and more rarely stauro- lite or augite. Hornblende is sometimes as prominent a constituent as — biotite, and, like it, arranged in layers. Rounded apatites are also present. The feldspar is microline, orthoclase, and acid plagioclase of about the composition of oligoclase. The porphyritic texture is due to the presence of lenticular areas of quartz and of feldspar irregularly inter- spersed along layers. These lenses are without crystal boundary and distinctly pebble-like in character. They are sometimes a marked feature of the rock and indicate an original conglomeratic character... The fresh character of the crystallization and absence of pressure effects on the crystals, indicating that strain was relieved by recrystal- — lization, the rounded apatites, the {quartz and feldspar pebbles, the sort- UOLOES ABATY [[P Autos § JO yqNOS tu Sutad [Asuueg ‘AyUno0d Aro WOSyuUOW ‘|| TUBA v uop.ey ‘gq ‘a Aq poydvasojoyg SSIAND SYOWILIVE GSALYOLNOO “WY "OOS *1039 “11Nd 8b “Id ‘PO6L ‘91 “10AN udplVAy ‘gq “a Aq pouduasojzot ‘eluvalAsuuog ‘Ajunoo Asewoszuow; ‘{]1ut Sutidg Jo yaNOS UOoT4OeS ABATI < uJ if 4 C A] . e Dd e bin ¢ ° SSISND SYOWILIVE GALYOLNOD "Id ‘pO61 ‘OL "WY "90S "103D ‘11nd PRE-CAMBRIAN ROCKS 295 ing of the mineral constituents, are microscopic evidences of an aqueous origin. In the field the gneiss has the appearance of a stratified forma- tion. In the Schuylkill section the formation is heavily bedded and shows fan folding, with crumpling on the periphery (see plates 48 and 49). In the center the folds are gentle and open. The formation also contains considerable disseminated graphite, which has been mined in Montgomery and Chester counties. A chemical analysis of assorted samples of the formation shows an acid rock whose composition, as is to be expected in the case of assorted samples, is of a granitic character and not decisively that of a sediment. The two analyses, which repre- - sent each a single locality, more distinctly indicate the sedimentary character of the formation. Chemical analyses of Baltimore gneiss i. IL III. On SU as 70.21 72.99 63.93 1 8 See 13.95 10.90 12.02 Ot eae 1.05 0.55 2.40 PM Se clas coke 3.08 2.50 2.95 | 1.26 7 £O7 2.44 Be Aalst. 3.10 1.88 4.00 ee 3.97 - 3.34 3.15 metho, 2.69 1.20 0.84 ee Oh 1.47 1.20 SO ie aaa 0.11 ene 1.13 Not det. Meek CLS 0.52 0.84 1.04 Ce Trace Sy Fe Cen PETRIE Sat “Oo, ; 051 (5) See 0.10 0.18 Not det RNa are Not estimated. Bis earSieen tre elt EP Peres Not estimated. FeS,..... 1.61 Fe,S, 4.04 Bee Se atos.cs 0.09 eee ba ibe Widens Me bis dt Faint trace. Trace MORE be 6 iss 0.11. a at Trace Ce ae 0.09 i.«. CuO Trace. 2) ES 2. 7 hae i PSR TS Serene ave Bea a ena Be CO RENO eon ad ce LODE Ey JACEE PMI a A ae ee 100.30 99.66 98.62 Analysis I was made by W. F. Hillebrand, of the United States Geological Sur- vey, from six samples representing six localities. Analyses II and III were made by F. A. Genth, Jr., volume C 6, Second Geolog- ical Survey of Pennsylvania, pp. 116, 105. Material of II was from Johnson’s farm, 1 mile east of Feasterville and north of road to Somerton. Material of III was from west side of Neshaminy creek, on the property of Phineas Paxon. 296 FE. BASCOM—PIEDMONT DISTRICT OF PENNSYLVANIA In the neighborhood of the intrusive gabbro mass the Baltimore gneiss loses its stratified character and becomes massiveand darker colored. The change in color is due to the development of hornblende, or more rarely augite, and of garnets, which develop in great profusion in the contact zone. The central body of gneiss is very completely penetrated by the gabbro. The banded structure is therefore much more conspicuous on the flanks of Buck ridge. A massive granitic character is considered an indication of the proximity of gabbro, although the gabbro may not always be exposed at the surface. Thickness, correlation, and name.—There is no means of estimating the thickness of this formation, which is the floor upon which the other - members of the sedimentary series were laid down. This gneiss under- lies material known to be of Cambrian age and presumably Ge: rgian, It is therefore pre-Cambrian and is correlated with the pre-Cambrian Stamford gneiss of western New England, with the Fordham g* eiss of New York state, with the Baltimore gneiss of Maryland, and the Garolina gneiss of the District of Columbia and Virginia. While the pre-Cambrian gneiss of Pennsylvania is not stratigraphically continuous with the Baltimore gneiss of Maryland, similar stratigraphic relations, like litho- logic character, and proximity of the two formations have found recog- nition inacommon name. The pre-Cambrian gneiss of Maryland has been called the Baltimore gneiss because of a fine exposure of it on Jones falls, in the city of Baltimore. This name is also given to the pre-Cam- brian gneiss of Pennsylvania. The Baltimore gneiss includes H. D. Rogers’ “ primal lower slate” and a part of his northern or “ third gneiss belt.” The major part of the “third gneiss belt” is gabbro. Both the Baltimore gneiss and the gabbro are included by the Second Geological Survey of Pennsylvania under the term “ Laurentian gneiss.” CAMBRIAN ROCKS: CHICKIES QUARTZITE Distribution—This hard resistant formation constitutes the north Chester Valley hills, the highland at Hickorytown and Coldpoint, the hills west of Whitemarsh, Fort hill, Camp hill, the highland about Wil- low Grove, and the long ridge known as Edge hill and Lafayette hill; also the north Huntington Valley hills. Character of the formation.—It possesses a conglomeratic lower mem- ber, which is largely composed of elongated pebbles of the blue quartz which characterizes the Baltimore gneiss. This member is brought to the surface in the nose of a pitching syncline. It may be found 1 mile south of Morganville and an equal distance east of Willow Grove. It is also brought to the surface on the middle limb of a fold at the base of the north Chester Valley hills, near the dam at Valley Forge. This CAMBRIAN ROCKS 997 conglomerate passes upward into a gray compact crystalline quartzite, which in turn grades into a silicious slate or into a sericitic quartz- schist, or is altogether replaced by the quartz-schist. The sericitic quartz-schist is typically exposed in quarries one-half mile northeast of Somerton station and in the Edge Hill quarries. It is a thin bedded formation here. Bedding and cleavage coincide and dip steeply southeast. The quartz schist is of a light buff to white color and always contains feldspar. The feldspar is for the most part orthoclase, more rarely microcline, and is usually more or less kaolinized. Tour- maline, apatite, zircon, magnetite, and staurolite are accessory constitu- ents. Stretched tourmalines are very characteristic of the formation. The schist readily cleaves into slabs and, by means of fissility developed in shear planes, into flattened rhombohedrons. Locally the quartzite may be quite geodiferous. The geodes are lined with quartz crystals. Quartzite of this character occurs at a locality known as “ Diamond rock,” on the southeast flank of the north Chester Valley hills. A chemical analysis shows the rock to be highly silicious, with sufficient alumina and potassa for sericite and lime-soda-feldspar. Analyses III and IV are of very sericitic facies of the quartz-schist. Chemical analyses of Chickies quarizite K LF.* III. EY. Sn SR ide Dee aes 87.87 84.59 58.97 56.35 Rs. $4 Bl eh ds 6.61 phat 22.61 22.28 POM TE a rei b Sulit a 2 39 aide 5.67 S72t tS Se ee ree ger Trace. err 0.25 1.40 Lt RR SRS eae 0.24 ae 0.08 0.19 LoS? SAE es ae ae 0.19 0.19 0.32 0.38 Me wre es ee Lis 2.79 7.34 12.63 Set ae 1.20 1.66 Fa 2.89 Mae er kok sk 0.38 pes LAL 0.82 Eres O iels na key ish 0.06 ae 0.07 0.16 Sy Se ae el ae 0.13 Faint trace. .... Trace. (UR 4 We eRe BY ae re tiscoky dhe eee Strong reaction. 100.80 soa A005 §6100.31 I. ‘‘ Itacolumite,’’ 13 miles southeast of Vanartsdalen’s, near Neshaminy creek. Il. ‘‘ Itacolumite,” east bank Neshaminy creek. III. ‘‘Itacolumite” from quarry northeast of Somerton. IV. One-half mile south of Willow Grove station, Montgomery county, Penn- sylvania. Analyses made by F. A. Genth, Jr., volume C 6, Second Geological Survey of Pennsylvania, pp. 107, 117, 121. * Partial analysis. XL—Butt. Grou. Soc. Am., Von. 16, 1904 298 F. BASCOM——-PIEDMONT DISTRICT OF PENNSYLVANIA Stratigraphic relations—The Chickies quartzite lies in an overturned synclinorium on the northwest flank of the Baltimore gneiss anticli- norium. It immediately overlies the Baltimore gneiss. There is no appearance of unconformity between the two formations, save as uncon- formity is indicated by the presence of the conglomerate member of the quartzite formation. Thickness.—The thickness of the formation does not exceed and may be less than 1,300 feet, although the isoclinal folding gives the appear- ance of much greater thickness. Overturned folding with stratification and cleavage dips to the southeast are the prevailing structures. The average strike is north 50 to 70 degrees east and the dips 45 to 80 degrees southeast. Faulting explains the disappearance of the quartzite on the south limb of the anticline. Correlation and name.—The name of the formation is taken from the locality of its finest exposure and greatest thickness on the Susquehanna river north of Columbia. At this locality Scolithus linearis has been ‘found, which occurs also abundantly in the north Chester Valley hills, and the quartzite of this locality underlies quartzite in which olenellus fragments have been found by Mr Walcott,* thus fixing its age as Georgian or lower Cambrian. The quartzite of the southeastern belt, lying farther _ to the east than this typical exposure of Georgian quartzite on the Sus- quehanna river, may have been deposited in an encroaching sea, and thus represent a later horizon in the Cambrian than the Georgian. No forms of life save Scolithus linearis have been found in it; hence it can — not positively be stated to be of Georgian age. It can safely be stated to be Cambrian, and is to be correlated with the Vermont quartzite of New England, the’Poughquag quartzite of New York, and Setters quartzite of Maryland. It isthe “ primal sandstone” of H. D. Rogers and “ for- mation number 1, Chickies sandstone,” of the Second Geological Survey of Pennsylvania. CAMBRO-ORDIVICIAN ROCKS: CHESTER VALLEY LIMESTONE Distribution.—This is heavily bedded crystalline, white or blue magne- sian limestone. Its surface exposure is largely confined to Chester valley, with a few scattered outcrops along the Huntington and Cream valley fault lines, and to the southeast in Chester county and Delaware. The presence of the limestone in Huntington valley along the course of Meadow brook is indicated by the character of the well water. The rock actually outcrops, however, only in the cellar of a wagon-house a quarter of a mile northeast of Meadow Brook station. In Cream valley, *C. D. Walcott: The Cambrian rocks of Pennsylvania. Bulletin U. S. Geol. Survey, no. 134, 1 i Er : : ™4] eu" i et VIOvALAsuded ‘{jUNOD ATAWOSJ UO “IAAT [[IY[ANYoOS “dip ysvoyynos wlojran Surmoys SNOLSSWIT ASTIVA Y3SLSSHO . ¥ 3 itt 0S ‘Id ‘yO6L ‘9L “1OA “WV "90s *1039 *11Nd 4 RS - r CAMBRO-ORDIVICIAN ROCKS 299 which lies on the west side of the Schuylkill and follows the southeast flank of the south Chester Valley hill, there are three exposures of the limestone—in West Conshohocken, at the head of the Gulf ravine, and in the bed of Gulf creek 14 miles southwest—while a series of sink holes in line with the strike of these outcrops attest the presence of the lime- stone near the surface. ‘This line of outcrops is continued southwest in Chester county, where limestone comes to the surface at some seven locaiities and is farther exposed to the south of this series of outcrops at nine isolated localities. Character of the formation.—The limestone is highly silicious and mag- nesian, but there are no analyses of it which give a sufficiently high percentage of magnesium carbonate to warrant calling the formation a dolomite. Limestones from Mogeetown to Conshohocken . BA be OS 3 4 5 ad 7 8 9 10 11 12 13 MENTOR sadicn neice 60.18 55.09 41.59 39.95 48.04 63.36 91.62 53.27 58.02 61.43 93.32 85.00 54.11 FegC Og. .....--c000e momacecrmileds~ 255 Notdetor eis letkS FPA acces | cesose — canane dnoyge Insoluble resi- Not GG n facani. w-- Get. 2.59 26,48 46.10 37.60 2.65 7.85 9.83 5.89 6.60 5.63 4.79 3.15 West Conshohocken Pee eg Re a eS eS Io ee ky a fs s'elel aces 40.27 SR RE ra Bie eis t ie Suysel v's 2 kc 6 wes 31.24 EEN erat his ree as atti pia. |. banier oie ps .aie reine 6 24.23 So ee er 1.12 Ey EE Bes Sa a ierheay gue «'s iio h do mse b 8-0 1.06 PEE RM ee Lene ow Said me-s/diis as sw oe 0.11 oe To LE RE a pk died ae ener gne ae 0.55 PUM er tae hits Pa nea a held Age td oie wee «vt 1.42 100.00 Analyses made by F. A. Genth, volume C6, Second Geological Survey of Penn- sylvania, pp. 126, 127. The limestone is always crystalline and increasingly so from west to east. Associated with increasing crystallinity is a lighter color, though blue and white marble may occur in the same quarry. It is sometimes quite micaceous and is always so in the neighborhood of the overlying mica-schist. The beds immediately underlying the mica-schist are silicious, micaceous, and schistose, and are to be char- acterized as calcareous schist. It is abundantly traversed by calcite and quartz veins. Quartz, feldspar, phlogopite, graphite, pyrite, siderite, and limonite are accessory constituents. Stratigraphic relations.—The limestone lies above the Cambrian quartz- ite in an overturned synclinorium. The prevailing strike is north 60 to 300 F. BASCOM—PIEDMONT DISTRICT OF PENNSYLVANIA 90 degrees east and the dips vary from 35 to 85 degrees southeast. There is a gradual change in the average strike and dips around the end of the synclinal trough in the northeastern end of Chester valley. Asin the case of the quartzite on the limbs of the overturned isoclinal folds, strati- fication and cleavage dip are coincident when the stratification dip is to the southeast. The prevailing structure is isoclinal (see plate 50). In an abandoned quarry at Rennyson, 13 miles northeast of Berwyn, a compressed overturned syncline may be seen, with cleavage and bed- ding coincident and dipping steeply southeast on the limb of the syn- cline. An overturned isoclinal anticline is exposed in a rock cut on the north bank of Valley creek one-half mile north of Howellville (see plate 51). A similar overturned anticline is to be seen in the cut made by the Washington branch of the Pennsylvania railroad near Arlingham, 13 miles southeast of Fort Washington. These secondary folds illustrate the type of the primary folding of the limestone. The limestone of Cream valley and the limestone appearing in scat- tered exposures to the southwest, are brought to the surface by means of erosion on the crest of low anticlines, or fill the troughs of overturned synclines. 3 7 | Thickness—With the interpretation of the structure given above, the thickness of the formation must be much less than the width of its out- crop. It is not perfectly determinable, but probably is not greater than 1,500 feet. 3 Correlation and name.—Fossils of the Chazy, Calciferous, and. Trenton ages have been found in the limestone occurring to the west of Chester valley and stratigraphically continuous with the Chester Valley lime- stone. Fossils have also been found in Chester valley in somewhat ambiguous material. This material is a geodiferous drusy quartzite, which is found in place south of Bridgeport near the Trenton branch of the Philadelphia railway, and which follows the contact of the mica- schist and the limestone, showing as loose boulders on the surface of the ground. It is interpreted as a secondary replacement of the uppermost calcareous beds of the Chester Valley limestone. At Henderson station loose material of this character lies on top of the limestone, and in this material have been found gastropod and cephalopod fossils. The following determinations have been made: Raphistoma, two species; Maclurea, Lituites, and Cyrtoceras. These are Ordovician forms, and indicate a horizon in the upper half of the Canadian series. The limestone overlies conformably Cambrian quartz- ite and is Cambro-Ordovician in age. It is correlated with the Stock- bridge limestone of New England and New York, with the Cockeysville limestone of Maryland, and the Knox dolomite of Virginia, BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 51 COMPRESSED ANTICLINE IN CHESTER VALLEY LIMESTONE Chester valley, Chester county, Pennsylvania Pi ats mA SPORE eee gy ori Pa ba tt ER ee 27.84 DOMINATE Mech hs Ue a a! Ti adele s Ke ned tisha 8.90 12:23 Meee ek tty coo ric sc wnee oh a pep aed 23.58 23.06 Anorthite ....... Bey mt Abe t erage y Wiel: ty he 22.24 Diopside...... Beet ee ead ew Peo s Wey joes eo 1.36 UE ae eae eae cAicaisle Ohtue 8.80 Magnetite. ...... “oo SLO ORE GRE SP at 1.39 2.32 MMRMORNN eet ol LE rade oo oni bln! xa 'o Swe bie we seni 0.91 Th SA ee er Oe Ee Saas 0.34 NE hi tiers al hs aig ese «sia vo Petey hy ry eee 1.06 1.25 99.76 100.46 The more acid facies of the Port Deposit granite is a biotite-grano- susquehannose* (class I, order 3, rang 3, subrang 4). This means that the salic constituents preponderate; that of the salic constituents quartz and feldspar are present in nearly equal amounts ; that of the feldspars the alkali molecules are equal to the lime molecules; that of the alkalies soda is dominant, and, finally, that the texture is megascopically hypautomorphic granular, and that the only abnormative mineral present ‘in the rock is biotite. : The more basic facies falls in class II, order 1, rang 3, subrang 4. Like the acid type, it is dosodic and alkalicalcic, but differs from the - other facies in the fact that the salic minerals are merely dominant, but not preponderating, and feldspar is dominant over quartz. The texture of the rock is hypautomorphic granular, biotite is the only abnormative mineral present, and is a critical mineral. The name of the rock is there- fore a biotite-grano-tonalose. GABBRO (GABBRO, HYPERSTHENE-GABBRO, NORITE, META-GABBRO) Distribution.—This is a great igneous body which, with many ramifi- cations and varying petrographic facies, intrudes into the Atlantic belt of crystallines from Virginia to New York. * The name susquehannosée is suggested by the writer for subrang 4 of rang 3, order 3, class I. This subrang, in the absence of any strong claims for preeminence on the part of the species falling in it, was left unnamed by the authors of the quantitative system of nomenclature. The appropriateness of susquehannose lies not only in the fact that large quarries in this rock species have long been worked on the Susquehanna river, but also that calcic granites of this general type are especially abundant in the region drained by that stream. The attention of the writer has been called in this connection by Dr H. S. Washington to the fact that there is a zone of calcic rocks striking parallel with the Atlantic coast and showing in Labrador, eastern Canada, Adirondacks, Pennsylvania, and Maryland. East of this belt is a zone of sodic rocks showing in Ontario, New Hampshire, eastern Massachusetts, New Jersey (Beemerville), and farther south in Georgia, Texas, and Tamaulipas, Mexico. The high calcie content of the other igneous types of the Pennsylvania district will be made evident in their discussion, 312 F. BASCOM—PIEDMONT DISTRICT OF PENNSYLVANIA It is an important formation in Maryland, Delaware, and southeastern Pennsylvania, where it intrudes alike into pre-Cambrian Baltimore. eneiss and into Paleozoic limestone and mica-gneiss. In the Pennsylvania belt its maximum. development is in the acaths east. From the Susquehanna river and from the region about Wilming- | ton, Delaware, it extends northeast, forming the main mass of Buck ridge, where it shows itself at the surface in exceedingly irregular areas. It is intimately associated with the Baltimore gneiss, and has so affected this formation along contacts as to produce an appearance of gradation from banded gneiss to massive gabbro. There are many instructive exposures of ealnbie and Baltimore gneiss. They all suggest the intercalation, along the periphery of the gabbro’ mags, of the former rock between the beds of the latter and the subse- quent folding of both materials. The penetration of the gneiss by the gabbro is quite irregular. Thin sheets swell into large masses, which, on exposure, weather into spheroidal boulders. Because of this pecul- iar injection of the gneiss by gabbro, gabbro boulders may appear spo- radically in areas where gneiss is the prevailing formation at the surface, and vice versa. At Glen Mills, a few rods south of a large quarry in gabbro, an aban- doned quarry shove eneiss and gabbro interbedded, yet the gneiss does not appear as a surface formation. Under these conditions it is impossible, in agile the boundaries of these two formations, not to include some gneiss within the gabbro area, while gabbro boulders may be seen within the gneiss area and gabbro injections appear in rock cuts, though not prevailing at the surface. This is particularly true of the areas northwest of Media. The Baltimore gneiss and the gabbro have not before been separated in the Pennsylvania Piedmont. They alike give rise to a relatively elevated rolling country, with irregular, rounded eminences. Dark-colored boulders of disintegration, with rusty exteriors, strew the fields and afford almost the only indication of the underlying rock. The gabbro boulders are exceedingly tough, except along contacts with the gneiss, where the development of hornblende and mica renders them more schistose and more easily attacked mechanically. Character—The gabbro is a medium-grained massive rock,* possessing either a bronzy-gray or a greenish-gray color, depending on the character of the ferromagnesian constituent. Quartz, pyroxene, and feldspar.may usually be determined in the hand specimen. The gabbro is a typical hypersthene-augite-plagioclase- * For full petrographic description of the Cecil County gabbro and norite, see Bascom, op. cit., pp. 121-132. BULL. GEOL. SOC. AM. VOL. 116,°1904, PL. 57 hw a4 Ne : Tomar So, et ty, . See 3.9 aw w, : Rae Figure 1.—HYPERSY'HENE WITH GARNEY VEIN. PPL 67 Figure2.— THE SAME wirH Crossep NIcots. 61 PHOTOMICROGRAPHS OF HYPERSTHENE-GABBRO Wayne, Delaware County, Pennsylvania. Photographed by F. J. Keeley i eth an al BULL. GEOL SOC. AM. VOL. 16, 1904, PL. 58 Figure 1.—HyYPERSTHENE WITH GARNET VEIN. PPL. X 67 Figure 2.—THrE Same wirH Crossep NIcOLs AND Gypsum PiatE. X 65 PHOTOMICROGRAPHS OF HYPERSTHENE-GABBRO Two miles southeast of Berwyn, Chester county, Pennsylvania. Photographed F. J. Keeley IGNEOUS ROCKS 318 rock, with accessory quartz, biotite, hornblende, magnetite, apatite, titan- ite, pyrite, pyrrhotite, garnet, and orthoclase. Exclusively decomposi- tion minerals are actinolite, chlorite, and serpentine. Quartz is very variable in amount, ranging from 0 to 30 per cent. The pyroxenic constituent may be exclusively hypersthene, chiefly hypersthene, or less frequently, chiefly augite or exclusively augite or diallage. It constitutes from 10 to 40 per cent of the rock. The plagioclastic constituent is labradorite or labradorite-bytownite: and varies from 5 to 50 per cent of the rock. Between the pyroxene, whether hypersthene or augite, and the labradorite there occur reaction- ary peripheral bands of garnets. Between the garnets and the pyroxene there is a narrow zone of quartz and hornblende. Hypersthene and the anorthite molecule will produce hornblende, quartz, and common garnet (an isomorphous mixture of grossular, almadine, and pyrope). In the case of the monoclinic pyroxene the feldspar molecule is not needed and lime is liberated. These garnet rims are the most striking petro- graphic features of the gabbro (see plates 57, 58, and 59). Wherever the gabbro has been subjected to pressure, as along the periphery of the intrusive mass, pyroxene is replaced, chiefly by green hornblende and subordinately by biotite. Such a margin of hornblende- gabbro is so marked a feature of the Cecil County gabbro mass that it can be mapped as a separate formation. A more or less schistose struct- ure accompanies the development of these two minerals. Along the con- tacts hornblende also accompanies biotite as a constituent of the gneiss. This contact phenomenon increases the difficulty of separating the two formations. The gabbro is associated with and, through decrease in feldspar, grades into pyroxenite or, with the addition of olivine, into peridotite. The youngest material into which the gabbro intrudes is the Ordovician mica-schist. The gabbro is therefore post-Ordovician in age. Its rela- tions to the granite are such as to indicate that the granite, which is also post-Ordovician, is the earlier intrusive. By “Sg previous surveys it has been included ante the “ Laurentian gneiss ” or syenite. XLII—Butt. Geox. Soc. Am., Von. 16, 1904 514 F. BASCOM—-PIEDMONT DISTRICT OF PENNSYLVANIA Determination of the rock species represented by the gabbro mass i - S103 cy SBF A Opts SAG LO | Ble On2300 E2iso FeOn oes MaQe5 eh 2i33 CROs. Ueiseao NeO oe. ell: MRaO ee F101 H,O+..- 1.27 H,O—.. 0.21 CO... 5.72 None. PIOM ee aed ZtQOy on O09 P,Osi se ae Chives None. ) SATE Not est. FeS,.... Trace. Sie titers Trace. Cr, .'a> a None: NiOCoO None. MnO. 2... 0.18 BaO.... Trace. SrO..... Trace. 11,0,.... Trace. V5Og-. =2:- > O12 C018) Total. 100.07 if Watia Vereen: 21.78 Orthoclase. .... 6.12 Al bIbe. is Seekers 17.82 Anorthite...2.2-3° toll TAX COMMS s4ie Gd speays “Diopside.s: ces. 2.32 Hypersthene.... 11.11 Magnetite....... 4.18 Timenite: COCKEYSVILLE MARBLE 335 sharp and generally easily recognized by the quarrymen, who are assisted by the fact that the dolomitic marble averages finer grained and richer in magnesium-mica than the better-burning, magnesium-poor rocks, Attempts have been made by acid tests in the field to recognize some stratigraphic distribution of the magnesium and calcium-rich rocks, but these have failed. On thecontrary, it has been found that there are rapid sharp alternations of the two types in a way which strongly suggests that whatever dolomitization occurred must have taken place prior to emer- gence from the sea and probably contemporaneously with the formation of the deposit as has recently been described by Professor Branner as occurring on the coast of Brazil. No fossils have been found in the marbles, and as they are highly crystalline it is very doubtful if any will be found. The deposit, how- ever, underlies the Wissahickon schist in exactly the same way as the Chester Valley marbles underlie the typical Wissahickon schists, which have been traced continuously into the Maryland areas. WISSAHICKON FORMATION The Wissahickon formation in Maryland occurs as a broad band wrapping about the older Baltimore gneiss and limestone areas and - occupying practically all the territory between the Blue Ridge and the Coastal Plain deposits except those portions consisting of the sediment- ary rocks already described, the overlapping Triassic sandstones of Car- roll, Frederick, and Montgomery counties and the igneous rocks. This wide belt of rocks may be roughly divided into the Wissahickon schists and the Wissahickon phyllites with included limestone lenses. The line of separation between these two divisions is not marked stratigraphically by any pronounced change in lithologic character, and it is not entirely clear that the division is a stratigraphic one, although it seems probable that the phyllites in all instances are of as high or higher horizons than the more crystalline Wissahickon schists. It has been clearly shown by recent field work that the more crystalline Wissahickon schists pass by gradations into the less crystalline phyllites and sericite schists. The more crystalline aspects of the Wissahickon formation lie in the eastern part of the Maryland Piedmont, where they are in association with the older, more crystalline sedimentaries and the large masses of intruded gabbros, granites, and other igneous rocks. The areal distribution of the Wissahickon formation may be discussed . in four parts. The area of Wissahickon schists lying east of the broad phyllite band, which extends southward from the Susquehanna river to southeastern Carroll county, represents the area of highest crystallinity Pe 7" “ a44.* 336 5. B. MATHEWS—MARYLAND AND PENNSYLVANIA PIEDMONT and closest similarity with the Wissahickon schists of the type locality near Philadelphia. The formation in this part of its development broadens from a narrow band at the Susquehanna river by increased folding about the anticlines of Baltimore gneiss and synclines of Cockeys- ville marble into a belt 10 to 15 miles broad as it crosses the Northern Central railroad. From this point it narrows somewhat to the south- westward, and the area is occupied in large part by the large mass of granite passing from Sykesville southwestward to Washington and ex- tending thence many miles southward into Virginia. North of the phyllite syncline occurs a corresponding mass of more crystalline Wissahickon schist. When, however, this is compared with the rocks on the southern limb of the synclinorium it is found that these rocks average slightly less crystalline and less metamorphosed than the corresponding rocks on the south. That they represent the same horizon seems to be well established by the areal distribution of the various masses, although it has been found impossible to carry the mapping of individual beds more than a few miles along the strike, and hence it has seemed inadvisable to attempt detailed cartographic representation. The Wissahickon schists on the west side of the syncline of phyllites passes southwesterly across the state, narrowing considerably in the southern portion of Carroll county and widening somewhat in passing southward to the Potomacriver. The areal mapping of this region has been carried out in considerable detail by Mr Keith, of the United States Geological Survey, who has shown that the line of separation between the eastern and western portions of the Piedmont plateau, as laid down by Williams,* is in reality a sharply crenulate line, due to the folded structure of the area, which brings the contact between the phyllites and the more crys- talline schists repeatedly to the surface. The band of phyllite, forming a synclinal trough extending from the Susquehanna southward, enters the state from York county, Pennsyl- vania, continues as a belt,varying from 5 miles in thickness at the Susque- hanna to about a mile at Whitehall, on the Northern Central railroad, whence it gradually widens southward to an average breadth of 3 miles in the southern part of Carrollcounty. The areal distribution indicates that we havea synclinal trough of considerable extent and well defined character, which is buckled at its center, and plunges northeastward and southwestward, reaching its maximum depth in the vicinity of Delta, Pennsylvania, where the Cardiff quartzite-conglomerate and Peachbot- tom slates are found folded within it. The southern termination of this phyllite belt has not been mapped in detail, and the limits here given * Bull. Geol. Soc. Am., vol. 2, 1892. WISSAHICKON FORMATION 337. are based on reconnaissance work on the part of Doctor Williams and the writer. Throughout its entire extent no limestone has been found up to the present time. It is, however, quite possible that small masses of this material have been overlooked, as the area traversed is a portion of the less dissected part of the Piedmont plateau, where the easy dis- integration of the less resistant phyllites offers few exposures. The areal mapping, however, is easily carried on from a study of the fragments, which occur plentifully in the soil. The rocks constituting the phyllite portion of the Wissahickon forma- tion are essentially sericitic, chloritic, and occasionally talcose schists, which clearly show their sedimentary origin, and have been less meta- morphosed than the Wissahickon schists already described. ‘Two views are held regarding their relations to the contiguous formation. They may be regarded as an infolded younger series, as held by the late Pro- fessor Williams, or they may represent a less metamorphosed portion of the Wissahickon formation. It seems probable that there is truth in both views, and during recent years the impression has developed that they represent the upper portion of the Wissahickon formation, which has been less metamorphosed, but that they are not separated by any great interval from the more crystalline Wissahickon schists which border them on either side, and from which they cannot be separated by any sharp line. When crossing the boundary between the two for- mations one may recognize within comparatively short distances that a ‘boundary has been passed, but up to the present no contacts between the two portions of the Wissahickon formation have been found. The interpretation of the phyllites here discussed does not necessarily or even probably apply to the more extensively developed phyllites lying a little farther west, though some of the latter may be the equivalent of the more eastern phyllite. The western phyllitic rocks composing the area between Parrs ridge on the east and the sedimentaries lying at the base of Catoctin moun- tain occupy a broad belt which enters the state of Maryland from Penn- sylvania and crossess the state into Virginia. Detailed work on this territory has not yet been completed, but the facts at hand would seem | to indicate that they rest on the Shenandoah limestone of the Frederick valley, and are in turn overlain by the Triassic sandstones and shales. Passing westward from the more crystalline rocks east of Westminster, there appears to be a constantly decreasing amount of recrystallization and metamorphism until along their western margin they present char- acters similar to those of ordinary shales. Within this broad belt of phyllites occur numerous narrow valleys of limestone or crystalline marble and lenticular areas of chloritic and sericitic schists, which are XLV—Bu.t. Geo. Soc. Am., Vou. 16, 1904 49 338 E. B. MATHEWS-—-MARYLAND AND PENNSYLVANIA PIEDMONT manifestly the metamorphosed representatives of surface volcanoes. The western limits of the slightly altered phyllitic rocks and their rela- tion to the sedimentary rocks of known age at the base of Catoctin moun- tain have thus been described by Professor Williams: * : ‘* Tf we follow the succession of strata eastward from Catoctin mountain, which bounds the Piedmont plateau on the west, we find above the hard Cambrian sand- stone a little limestone, which, however, is immediately buried beneath the over- lap of red sandstone. This blue limestone, whose fossils show it to be of lowest Silurian (Trenton) age, soon, however, emerges from beneath the overlying and unconformable Triassic (Newark) sandstones as a series of considerably folded beds, which are succeeded on the east and apparently overlain by carbonaceous and hardly altered shales. These are like those which occupy a similar position above the same limestone farther westward [Martinsburg shales]. Still beyond there follow with the earterly dip the thick beds of sandstone which compose Sugarloaf mountain. . . . The Sugarloaf sandstone passes on its eastern side upward by a gradual transition to shaly layers into sandy slates, and these again into the succession of sericite and chloritic schists, which compose the mass of the semi-crystalline area.” CARDIFF QUARTZITE The quartzite and quartzose conglomerate which occur in the north- east part of Harford county form a narrow band apparently resting on the phyllite and underlying the Peachbottom slate, wrapping around the latter and extending beyond its southwestern limits to the valley of Broad creek. The formation is of small thickness and of slight extent. PEACHBOTTOM SLATES The Peachbottom slates extend as a narrow strip within the limits of the Cardiff quartzite and pass beyond it across the Susquehanna river into Lancaster county, Pennsylvania. The formation is composed en- tirely of characteristic blue-black slates, similar to the material put on the market, and the homogeneity of the formation is now so complete that it is impossible to perceive within it any succession of sedimentary beds. It is usually considered, however, that the central portion of the ridge differs somewhat from the sides, and that the formation represents the uppermost member in a tightly pinched syncline. The age of these Peachbottom slates has been somewhat questionably determined on doubtful fossil evidence to belong to the Hudson River horizon of the Ordovician. If this correlation is correct, we have an upper limit fixed for the age of the deposits of the eastern Piedmont in Maryland, and an apparent confirmation of the position assumed that * Maryland, Its Resources, Industries, and Institutions. Baltimore, 1893, p. 66. AGE OF THE FORMATIONS 339 the Wissahickon is itself of Ordovician age, since there is no well defined and extensive line of separation between the Wissahickon and these locally developed slates. AGE OF THE FORMATIONS Within the Maryland Piedmont, so far as studied in detail, there are no means of locating accurately the position in the stratigraphic column of the four formations—Baltimore gneiss, Setters quartzite, Cockeysville marble, and Wissahickon schists and shales—but on either end of a con- tinuous extension of these formations we have a sequence into sediment- ary rocks of known horizon. On the east Doctor Bascom has shown that the Baltimore gneiss is pre-Cambrian, the quartzite Cambrian, the limestone Cambro-Ordovician, and the Wissahickon Ordovician. On the west the phyllites are apparently in the same relation to the known Cambrian and Cambro-Silurian deposits, although in this area the geo- logical mapping has not been conducted in detail on account of the lack of satisfactory topographic maps. Theconclusion from the facts at hand would therefore seem to warrant the correlation shown on page 340. Opposed to this conclusion is the fact that directly on the strike of the Wissahickon schist are a series of more quartzose and more compact metamorphosed sediments, which have been regarded by Mr Keith, in his discussion of the geology of the Washington quadrangle, as Carolina gneiss, which, as defined by him, is pre-Cambrian. The grounds for the assigning of such an age to rocks of this region involve the areal work of Mr Keith in the region to the north and west of Washington, which has not yet been published on account of a lack of a proper topographic base.* STRUCTURE AND STRUCTURAL RELATIONS OF THE PIEDMONT FORMATIONS For a proper understanding of the structural relations of the Piedmont deposits of Maryland it is necessary to recognize the character of the major structures along the eastern Atlantic coast from New Jersey south- ward and the position of the Maryland deposits with respect to these structures. The facts which are given below are familiar to students of American geology, but it seems desirable to restate them in relation to the Piedmont rocks under discussion. Among the more striking features of the continental structure along *A conference has been proposed for the workers in the Piedmont of Virginia, Maryland, Penn- sylvania, and New Jersey, and an attempt will be made to reach a mutually satisfactory under- standing as tothe age correlations. Untilthis is done it is better to regard the age of the Maryland deposits as unsettled, with the evidence favoring the position taken in the preceding discussion, ‘UBLIQUIB) -O1 **SSIous OIOUIIYV -- =gsrqug a10uINeg * “SSIOUS O10UII}[B +999 ‘g}SIUO8 ‘soJIUBIQH "TT UBLIQUIB)-81 “UOT}BULIOF UNOpNOT | ( ‘eUOJSPUBS TO}LOAD MA | -requiet 1eddn | fo as See as “UBLIQUIBD Snosdv[[Is1e UII ‘ojeys siodie py | ‘uBLIquisy ‘eyzyvnbseryory " eae. oar ce gest a41zZ41 Bub 8109999 | | -quojspuvs wivjoryuy J “9U0}SOUWUT “UBLIN[IG-0.1q W1B/) AoyeA t9yseqyQ cr oe eae +s) -QIQIBUTE][IASAOYDOH *9T10}SOUTT yeopusueyg *“UBIOTAOPAQ-O1Q UB) "UOT}BUL “"u0T} ( ‘o[Bys SInqQsulzls yA “IOATY, UOSPNY -IOJ UOYOIYVSSI AA -VUILOJ UOFOTYVSST 4 “S{SIYOS UOZOIY *9U0}8 ‘oqizyIvnb PIpVy -BSsiA\ Pus ayaqd L-pues ueynuvsseyy UBIOIAOPIO. "9}BIS U0}JOQ TIVE ‘UloIsey —“ULEYNOG ‘“U1eysvy ‘[B.1]UID "U19]89 MA CS SSS —— ee ‘LNONGHIG VINVATASNNG ‘LNONGHIg GNVIAYV IA 340 5. B. MATHEWS—MARYLAND AND PENNSYLVANIA PIEDMONT STRUCTURE AND STRUCTURAL RELATIONS 341 the eastern coast of the continent is the generally northeasterly trend of the folds constituting the Appalachian system. This structure holds for most of the territory from Alabama to Maryland and from the Canadian boundary to New York city. In the region of Pennsylvania, however, there is a marked deflection of these parallel folds, on account of which they are found to strike almost due east and west across the state from the Potomac to the Delaware. Beyond the Delaware the formations gradually resume their northerly trend. The change in direction of the folds of the Appalachians is readily seen in the geological map issued by the state of Pennsylvania and in smaller geological maps of the United States issued in the standard text-books. For the present purpose the structure is clearly brought out by a consideration of the areal distribu- tion of the limestone of Ordovician age, which are known by various -names—Shenandoah, Lancaster, Chester Valley, Trenton, etcetera— throughout the long course of its development. A study of a map (fig- ure 1) showing the geology of the Virginias, Maryland, and Pennsylvania permits one to trace the extent of this formation from its northeast course along the Shenandoah valley on the west side of the Blue Ridge to the north end of South mountain in the region between Chambersburg and Carlisle, Pennsylvania, where the trend changes from northeast to east. - This easterly trend is pursued to Harrisburg and the vicinity of Read- ing, when a series of folds causes the areal distribution to extend some- what farther north, although the structure lines still remain approxi- mately east and west. If one disregards the overlaps of Triassic in this portion of Pennsylyania, it may be seen that the Cambro-Ordovician limestone widens by a series of folds along the Susquehanna and Schuy]l- kill rivers and extends back southward along the eastern flank of the South mountain anticline as far as the Potomac river. Within this broad region of limestone may be noticed several anticlines of gneiss bordered by Cambrian quartzite and surrounded by the Cambro-Ordovician lime- stone, which in few instances appears overlain by shales and schists of Ordovician (Hudson River) age. The southernmost portion of the Cam- bro-Ordovician limestone appears in many cases, as at Frederick, near Strassburg, at Downingtown, and on the Schuylkill, to be on the south side of anticlines, and thereby representing the northern limb of syn- clinoria, which possess the general axial features noticed in the Ordo- vician limestone. Overlying the limestone along its southern and eastern borders are the phyllites and less crystalline representatives of the sup- posed very ancient members of the Piedmont plateau. Disregarding the possibility of extensive thrust faults, for which the facts in the area rarely indicate any probability, the most natural inference would be that these shaly phyllites are Ordovician deposits lying in an eastern geo- 342 FE. B. MATHEWS—MARYLAND AND PENNSYLVANIA PIEDMONT syncline whose axis conforms to the general character of the continental structure. Such a geo-syncline would be between the Blue Ridge and Great valley on the north and west and the accumulation of metamor- phosed sediments and igneous rocks on the east. This inference is, moreover, corroborated by the character of the folding where it has been deciphered in eastern and central Maryland. The map still further shows that, while the forces at work to produce the Appalachian structure along the Atlantic coast were exerted north- SYNCLINAL AXEs, ANTICLINAL AXES FAYLTS, Figure 2.—Sketch of axial Lines for major Structures of the Piedmont. west and southeast, the distribution of force in the Maryland Piedmont was very much more complex, since the center of the major curve of the continental structure lies approximately coincident with the city of Baltimore. ‘This fact would seem to indicate that whatever forces were at work in the Maryland Piedmont during the formation of Appalachian structures must have worked on lines radiating from a central point, and that if any difference of force along these lines occurred there must have been developed more or less torsional stress, and that such torsional STRUCTURE AND STRUCTURAL RELATIONS 343 action would be likely to produce folds of shorter major and greater minor axes than are the rule in the long-drawn-out ellipsoids of Appa- lachian folding. The effect of these forces must thus have a direct ex- pression in the areal distribution of the various deposits found in the Piedmont of Maryland, and, conversely, if such results are found there is developed a reasonable probability that the present major structures of the Maryland Piedmont were produced by the same forces and at the same time as the Appalachian folds. Starting on the west with the faulted and sharply folded anticlines of the Blue Ridge bordered on either side by Cambrian rocks and associated igneous masses (compare figures 1 and 2), one may pass successively eastward through Maryland across the gently eastward sloping lime- stones of the Frederick valley, which in turn, as above described, appear _ to dip under the so-called semi-crystallines or phyllites of the Piedmont. While the detailed structure is not fully known, it seems probable that ° there exists in this part of the state a very open general structure by which the beds lie almost horizontal in their major folds, with a much compressed subordinate structure, which, because of the numerous minor folds, give to the rocks an appearance of highly inclined and complicated folding. East of Parrs ridge the rocks are more crystalline and the fold- ing a little more pronounced in its general features, with a change in strike of the axes of the major folds in conformity with the change of direction in the continental folding previously described. Between the area of open folding just described on the northwest and the cover of Coastal Plain deposits on the southeast one may readily recognize in the Maryland area the broad synclinal trough of the eastern phyllite belt and that of the Cockeysville marble, separated by the dome-like anti- cline of the Baltimore gneiss already described. Still.farther east, sepa- rated from the Cockeysville synclinorium in part by asouthern anticlinal border of Baltimore gneiss,.is a broad zone of igneous rocks composed of gabbros, granites, and other plutonic types which occupy most of the eastern border of the Piedmont between Wilmington, Delaware, and Laurel, Maryland. Minor igneous masses are found with the same general trend, and these are seen to be rather closely associated with the structure lines of the region, occupying as they generally do anticlinal axes. This relation to the structure lines is particularly shown in case of the long belt of serpentines extending from Lancaster county, Pennsylvania, across the Susquehanna river almost to the nose of the northern anticline of the Baltimore gneiss. Farther to the southwest, almost on the strike of this anticlinal axis, begins a long and somewhat narrow body of granite ex- tending from Sykesville, on the Baltimore and Ohio railroad, southward Rae 5° > Weer h 344 &. B. MATHEWS—MARYLAND AND PENNSYLVANIA PIEDMONT across the Washington quadrangle and thence continuing probably as far as the Fredericksburg region. EASTERN PIfDMONT OF VIRGINIA Our knowledge of the Piedmont formations in Virginia is by no means equal to that of Pennsylvania and Maryland, since little or no work has been done in the area between the Blue Ridge on the west and the Coastal Plain deposits on the east since the days of the Rogers. There have ap- peared, however, occasional local descriptions of limited areas, and these the present writer has attempted to correlate with the facts given in the early reports of the Virginia survey, and these correlations have in a measure been checked by personal reconnaissance as far south as the James river. It should not be thought, however, that the following statements are regarded as any more than a working hypothesis, which may aid in the ultimate interpretation of the Virginia Piedmont. A study of the Piedmont geology of Virginia from the literature im- presses the student at once with several generalizations, the strongest of which is that in all of the previous work little or no attempt has been made by the different students to recognize and map the different forma- tions within the limits of the Piedmont east of the mountains. A second impression from the descriptions published is that formations similar to those recognized in Maryland extend across the state of Virginia with a trend similar to that of the Blue Ridge and the Shenandoah valley farther west, with the exception, however, that the limestones and marbles are apparently absent in any extensive development comparable to the areas of Cockeysville marble in the Maryland region. On the other hand, the little lenses of marble which have been recognized and but little studied in the western part of the Maryland Piedmont may be traced, where the older rocks are not obscured by the Triassic sandstones, all along the western side of the Virginia Piedmont from the Rapidan to the Staunton river and probably to the North Carolina line. Again, it may be noticed that the igneous rocks occupy similar posi- tions in Virginia to those in Maryland, with the exception that they are relatively less abundant, and, so far as the facts at hand indicate, are usually granite and not gabbro, although the writer has found hornblende schist in the valley of the James which probably represents metamor- phosed gabbros, and the lithologic descriptions suggest that gabbroic rocks, more or less metamorphosed, may be found in several parts of the state. The principal masses of granite so far recognized are (1) the south- ward continuation of the granite of the Catoctin belt recognized by Keith,* * Fourteenth Ann. Rept. U.S. Geol. Survey, pp. 285-395. EASTERN PIEDMONT OF MARYLAND 345 which evidently extends southward into Albemarle county; (2) the southward continuation of the Sykesville-Washington granite area, which apparently is more or less continuously developed from Washing- ton to Fredericksburg, and possibly includes the granitic masses at Columbia, on the James river, which is approximately on the strike, according to the general trend of the formations ; (8) the granitic mass in the area about Richmond, which conforms in position to the more easterly granite of the Maryland area. When an attempt is made to decipher from the remarks and lithologic descriptions-a general clue regarding the structure of the Virginia Pied- mont one may be led to the following suggestions: On the west is a well defined anticline of the Blue Ridge with Cambrian and Ordovician rocks on the west and the rocks of the Piedmont on the east. Throughout most of the distance from the Potomac to the James the analogues of the Blue Ridge and Catoctin mountains of Maryland may be traced as com- panion topographic features. It is, however, to be noted that in passing southward these parallel ridges become separated and the easternmost less prominent and broken up, suggesting the southward dying out of the eastern part of the double-topped anticline exhibited in the Maryland area. With the Blue Ridge on the west and a southward continuation of Catoctin mountain, represented by Bull Run, Southwest, Carters, Green, and Finley mountains, on the east are associated amygdaloidal chloritic rocks representing surface flows, which have been termed by Mr Keith “Catoctin schist”? and occasional meta-rhyolites. East of the more mountainous part of the Piedmont little is known of the structure in the northern part of Virginia south of the Washington sheet, but it seems reasonably probable that we have here a broad synclinal area compara- ble to that of the western Piedmontin Maryland. The rocks are in large degree similar and the areal distribution and position of the region with respect to the Maryland territory support this assumption. South of the James river, if one may judge at all from the meager de- scriptions published, including the sections of the Virginia survey, there are a series of more sharply compressed folds, which bring to the surface more crystalline gneisses comparable in character to the Baltimore gneisses of Maryland. Associated with these are less crystalline schists similar to the Wissahickon formation of Pennsylvania and Maryland, but differing apparently by a somewhat less pronounced degree of metamor- phism. It isin this area that Darton * found fossils indicating the Pale- ozoic age of these deposits. Southward from Amelia court-house is the suggestion of a large synclinorium plunging to the southward along a line from Amelia, Virginia, to Warrington, North Carolina, which would * Amer. Jour. Sci., 3d ser., vol. xliy, 1892, pp. 50-52, XLVI—Butt. Grou. Soc. Am., Von. 16, 1904 346 EE. B. MATHEWS—MARYLAND AND PENNSYLVANIA PIEDMONT explain the relative broadening of the Piedmont along the Virginia-Caro- lina boundary, where the structures of the west are trending southwest- ward and those on the east in a more southerly direction. CoNCLUSIONS - The preceding survey of the present state of knowledge regarding the formations and structure of the Piedmont in Maryland would seem to warrant the following conclusions: | (1) The four formations recognized by Doctor Bascom in the Phila-_ delphia area may be traced across the state of Maryland, and they prob- ably constitute the bulk of the rocks forming the Piedmont of Virginia, which have heretofore been mapped as a unit. (2) The structural character of the Piedmont from Trenton southward to southern Virginia is similar throughout to the Appalachian structure — of the less metamorphosed Paleozoics to the westward—that is, one may recognize within the Piedmont a series of long and narrow folds and arches trending parallel to the trend of the Appalachians. An excep- ion to this general rule is noticeable in the central Maryland area, which lies toward the center of a local curve in the Appalachian structure, where the structural forms are more nearly circular than is the general rule in the Appalachians. (3) The age of the different formations in the Piedmont is still un- settled, with the weight of opinion so far presented in favor of an early Paleozoic age for the formations immediately overlying the Baltimore oneiss. (4) The conclusions reached, so far as they relate to Maryland, are based on a detailed mapping or intimate knowledge of practically all of the Piedmont, but those of Virginia must be regarded as merely sug- gestive and warranting credence only in so far as the principles of inter- pretation established by detailed work for part of the province permit of application to contiguous areas. Pa M7 Oe Oo pt yt) SAND r mses a re ent i le eeadneaeae Vag : = oF ee 4 yi a 2 oF / ff Treat oe, t WE Air : / oak OMe ; - wan ~, \ ‘ 3 . hit ~ gr. vs ~ “* ? = ye a Rate & :) oft nn ee ides he 4 tr a F ‘ ‘ , : : , ’ i. y ‘ San giigS AtAy 7 - A) ook yt ih; be ae Reh a 4 f hr tii Pa : Ye Wie #f iq Pes ght m4, Wo b Lit jinh? othe o i/ 7 - 2 * > j < ; ee = >. oe 2 fs “ . 9 pl ae aS : 4 < y - = oo Pied ad Sie? pee = y a ee oy OT a ; a! ett neS, tt, Pri” : ; As ? ; 7c: Sa en © le eae Bie ey ee Cie dag Ms Lee na, f ‘ | P { — . . j a5 Ade 7 a Cpe + Ps ke : ae : Fs ea : * eee OFM P / f : . we : ie “ae ons a etie “of > al - rica - a ete ~ ; ; . i ks Sato thas Lea dilehigv ici PS @ ae z AeA Moe PP 4 * 5 t zs * : a ( - « - —. 2 a b. Ds te) eal & ie § ute: ~ ° - ait Ona " ” ~ et hent aN =O my S ~ ; SJ agg . Baey sedi” BNSC. EE eh eraamiata ” a xy Se “2. hon Oe “Se ‘ew Bo tia Gb pnt cicate (8 Pad 5 Q wnt ° bes | + < : Ces Stati Ld be é =~ : 7 q . Pe 4 al Mr <2¢ > a ~ . . mm Pod. St .6° € 59. Sue : * Mette ae cae > ~ <8 aS oe | fo eK * q A > ore ees ; ~ oO 2 Q ee a ee ee a ‘ ‘f rites. C = =— * re ; él: 2 8] Dee Se Le = ak ‘wo 1 2 Op artT) Q f +3 e ces & a. oe Ve... ee m aves cya ae 2 gina ot a = gis a4 OPS -. <-¥ .. “2 . ‘ Oo. a yt DiS Pe . A He Fz. - “= i a oe Ri. Saeae ia mates 0 Be Ss £5 * 25508 | : A345 a BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 347-366, PL. 65 MAY 31, 1905 COCKEYSVILLE MARBLE BY EDWARD B. MATHEWS AND W. J. MILLER (Read before the Society December 30, 1904) CONTENTS Page ENE Paes Sis kao er odie os bebtealaless Malse and bis ee nae hrcathes Sate gv eG 347 Rocks of the region........... ae Loe Unset Few ents anle Wh ae ota s 348 STONING i BR eR, or Rn one 348 BHOPE: SNCISS. ©... 25 accede ce cso: FREE 2 aie eee ner ap ae Oe EEOC Tree ae 348 MIR ele rete on nen cit cla nine ee asain eu “Baie OA sid Sea eee alae 349 SMESEIOMENIC! TERUG sc oye fe ce cp ce vs ske ses ees- Lacs s puter me ecty (ae bie oe . 349 EP METEL SE ett, le tes i Shera eles eee ies welcdeeecavaneseacs 351 Igneous rocks.......... adalat ee PEEL Seba North shia y tiny eee tee a ek owt whale 502 Seerteiribution Of the rocks. .:...... 60-66 eieeees Es tae ene eS: Eola 352 I asf PR eae ES bla cde ‘vy eta peia'in VL 8ed a KAe wid Sie OTe Me Re 352 DEE AETY OP SAILIINIOTS PREING.. 2. cree sec cc cae cc cee ccccberestawecas 352 ERE CAME TE cat CON RNa y e Ee ee ae wi hee chews 352 RM re ete ee ee ysis tomepiel dn mainte elk Pl easy «wee 353 NE PME CRE SE CUCTO CUIATEZIGE. oc gine eee le ve els wc baeere ces ccapocsecees 354 General characteristics of the exposures............-.. 000-0 e ee eeee 354 EE EMRE MM eee cee eS aL! wh alah alee wee ea bee kd we oada oes 304 MMR ENMC TN SP RSENS ey Ae fee are Cael Syn es SREB N's ah dla h dlers ocd d vty the 355 Runnin or Cockeysville;marble «.......5....060). cseesese wccceebleesue 357 I BT 8 ois bam nan, initia eons, 6) gain pion Sa ite Gecalhatata 307 Greenspring valley ..... ee Ne Bi eRe 2 oe Ee gO reg ae 357 MMMM Ce ats ie ic 4. Pot alec aed o wee esto eede beaks wh lgsceee 358 farcumeton valley—Butler ares. ..... 4.0... bo ee eee ce cee eee ees 359 menripution of Wissahickon formation..... ... 00.00. cescecccacecceccee: 360 gE ie co. s GS on bls yn eine nee dab che adeeh cesses Rito, 361 MINCE ROLUENE, arc cisra)s oa. aie mi Ae a> ole 6S Ble bsg aw aes wi we did file 361 OSD eee Ee ee ee eee re ee rk ee 362 Faulting...... RRS oats sans @S,0 A oa e100 = Nght OP Re ae 363 INTRODUCTION In earlier papers * the senior author has presented an interpretation of the structure of the Piedmont region as exhibited in Maryland, and * Amer, Jour. Sci., 4th ser., vol. xvii, pp. 141-159, 1904; see also paper in this volume, pp. 329-346, XLVI[—Bott. Geo. Soc. Am., Vor. 16, 1904 (347) VOL. 16, 1904, PL. 65 BULL. GEOL. SOC. AM, 7 a a Se a inane Pec ee Oe Tn re ee oe x4, QI f j, <4 5 2 t é kad e4, tA, ah LS, =, \ py +N aN ie CAS, i, (rs re STRUCTURE SHEET. ‘S a > 2 My, oy ¢ 4, Sig BY EDWARD B. MATHEWS awo WILLIAM J MILLER. MILES. 1905 SCALES D HORIZONTAL EQUAL, WISSAHICKON SCHIST. CRYSTALLINE LIMESTONE. SCTTER's QUARTZITE. s z & z o re co 4 MARYLAND. . ry < 3 < a cs MORE COUNTY, gd ee As &), Hy AREA OF BALTI— weg: é ih SERPENTINIZED GABBAD, [EES] ecemarire. GACBRO—DIONITE. ESS enveerre CROSS THE LIMESTONE GENERALIZED SECTIONS A— SECTION ‘SYMBOL oe ihe. VERTICAL AN AO, Hayy AY e \ \\ \ AKA) Ny ANN \ LY AN MUN Y WM MANNY SRA SQA roth hy \ ANT ANN A we ‘ aN Lith fy YOM » RN OT 324] Vat gy Tyr yt Dee Ss Ss / ga tapes ih } GENERALIZED SECTIONS ACROSS LIMESTONE AREA OF BALTIMORE COUNTY, MARYLAND < 348 EE. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE suggested that it is possible to recognize in the highly crystalline and much metamorphosed rocks of the area lines of bedding which indicate that the crystalline rocks of the Piedmont possess a general structure comparable to that of the Appalachians lying to the westward. In the present paper it is proposed to give a somewhat more detailed discussion of a local area, the special problem of the junior author, in which the type of folding is well exhibited and more easily seen because of the sharp differences, lithological and topographical, between the limestone or marble and the adjacent rocks. The area under discussion occupies a tract of approximately 300 square miles, represented on the southern portions of the Belair, Parkton, and Westminster sheets and the northern part of the Gunpowder, Baltimore, and Ellicott sheets of the United States Geological Survey, or, in other words, between 76 degrees 25 minutes and 76 degrees 50 minutes west longitude and 39 degrees 20 minutes and 39 degrees 40 minutes north latitude. The Northern Central railroad from Baltimore to Harrisburg passes directly across the region, while the Western Maryland frem Balti- more to Hanover skirts its western limits. 7 Rocks oF THE REGION LIST OF THE FORMATIONS Within the limits of the region under discussion are exposed the rocks of the four formations, namely, (1) Baltimore gneiss,* (2) Setters f quartz- ite, (3) Cockeysville f marble, (4) Wissahickon * mica-schist, mica-gneiss, and phyllite, described by the senior author in the discussions already referred to. The lithological characters of these different formations vary somewhat from place to place, but as exposed in the vicinity of the Cock- eysville marble may be described as follows : BALTIMORE GNEISS This is a highly crystalline gneiss, composed of quartz, feldspar, and mica, or hornblende, with accessory minerals so distributed as to pro- duce a well marked gray banded gneiss, the individual bands of which vary from a fraction of an inch upward. The average thickness, low- ever, is quite small. Some of these beds are highly quartzose, resembling a micaceous quartzite or less frequently a vitreous quartzite; others are rich in biotite or hornblende, producing dark to black rocks indistin- guishable in a hand specimen from the mica and hornblende schists and gneisses derived from igneous rocks by metamorphism. Through these banded gneisses are intruded pegmatite and aplitic dikes more or less * Accepted by the Committee on Geologic Names of the U. S. Geological Survey. + Old terms used provisionally until an agreement on names is reached. “Lh c x= ROCKS OF THE REGION 349 parallel to the regular banding of the gneiss. Many of the broader bands composed usually of hornblende-schists are probably of igneous origin, as has recently been shown for similar rocks in New York by Professor Julien. The banding of these gneisses has generally been regarded as secondary, but the fact that the strike and dip of the bands coincide with the strike and dip of the sedimentary rocks around the nose of folds and on either flank makes it seem probable that they are very often, if not always, indicative of original variations in sedimentation. SETTERS QUARTZITE This formation as developed in its type locality, Setters ridge, in the southwest corner of the area under discussion, is a fine grained, some- what saccharoidal, thin-bedded quartzite of white, cream, or light-brown color. The beds are usually separated by thin films of sericite in tiny » glistening flakes. On the surface of these mica-covered planes fre- quently occur black tourmaline crystals, which, as first shown by Wil- liams, give evidence of movement along these planes. In other portions of the area, especially in the vicinity of Butler and along the Gunpowder river west of Glencoe and east of Phoenix, the rock becomes more vitreous and less clearly bedded. ‘This is especially true in the cutting of the Gunpowder just below Warren, where the quartzite is ex posed in a double- _ topped anticline of moderate size. Resting on top of the more homogeneous and quartzose members occa- sionally occur quartzite bands interbedded with micaceous layers, which in the hand specimen appear to be garnet schists, practically indistin- guishable from the Wissahickon schist, which occurs higher in the column. At times the quartzitic layers become insignificant and the whole mass looks like a Wissahickon schist. There are, however, minor features which can not be put into words, but which may be recognized during the progress of continuous mapping, which render the aspect of the exposure and the characterof the rock valuable as aids in mapping. It was at first thought that this portion of the quartzite formation was in reality a faulted-in representative of the Wissahickon, but detailed mapping of critical areas, where the formations possess more marked minor folding, show that a fault can not explain the development of this local member of the quartzite formation. The Setters quartzite is generally found dipping at a steep angle, and because of its resistance to weathering agencies it is often a topographic feature which aids in determining its limits. COCKEYSVILLE MARBLE The carbonate rocks, which because of their more extensive exploita- tion and peculiarities in weathering have been of especial service in 350 E. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE deciphering the structure of the Maryland area, consist of calcitic and dolomitic varieties. The main mass of the rock is magnesian and would be classed as a crystalline dolomite. This is not a mapable unit as op- posed to the marbles, which are of pure calcium carbonate, and the investigations of the junior author, so far as they have progressed, do not warrant any statement indicating a stratigraphic difference in position between the lime and magnesian rich rocks. The intimacy of association may be judged from the following detailed section made by the junior author : Section showing alternations of calcitic and dolomitic marble Feet Inches Medium eraimed, caleitieis.s.3) <5 260-24 ne case see eee 5 Rather coarse grained, clear, white, calcitic.............. if 2 Coarse grained, bluish, pyrite, calcitic.............. .... se 4.5 Very fine grained, friable, dolomitic. ........ .......... 1.5 Very. pure, coarse erained; ‘ealeitie..... .¢2) suendied Bes 6 Fine grained, gray, micaceous, dolomitic ................ sith aes Fine grained, grayish brown, impure, calcitic............ Fine erained, pure, dolomitve.../: 25. v jucans oss oe eee , Medium orained, blué,ealeitie. 070.00 200 sk ee 1 Medium to coarse grained, white, calcitic........... .... Fine,erained; pure, dolontitie. (0... 0.0. os. 2 4ie ae ee 8 Medium grained, blue, calcitic. -..%.2. 2.42. ahs ete ae 1 Kine grammed, white, dolomite. 5...260 5 oc skeen oe Bee Medium to coarse grained, brown, calcitic....... ....... 1 Medium grained, impure, bluish, calcitic................ 1 6 Fine grained, micaceous, gray, dolomitic................ 2 Medium grained, pale blue, impure, calcitic..... tae Coarse grammed, white, caleities. 0. .5 2.000402. cae eee 2 1 1 — AInNnNtonaret oO wr Fine erained, brown, dolomities:. so. o9n ok ce eae Medium to coarse grained, calcitic........ hc. dick eee Very coarse grained, white, calcitic 7... cide. 2. e's. ee ee Fine grained, brown, dolomitic. [25..)..2)-05.. = tesge coe Medium grained, blue brown, calcitic................08. 1 Fine grained, micaceous, white, dolomitic............... Coarse’prained:; pure; ‘caleitie. (008) Skee nose etek eed Fine grained, micaceous, dolomitic..... 2.0.22... eins lees Coarse eramed, white: calettves. =. iWoc- cd a vs tee ee ee Hine grained, micaceous;, dolomities, <0 Jise.eue ssn ata ee Coarsegraimned, pure, white, ’caleitic.........0s5.%.2-5 see 1 Fine erained, pure, dolomitie.. oo fe eo. in vn se ee ee 1 2 Fine grained, micaceous, dolomitic................ Siae 6 Fine grained, brown, dolomitic.: +2. et.s. 0. aos ae) ees oe ad 7 Coarse grained, pure, friable, calcitic.............. ... oe Fine prained, brown; dolomitic. 2 ..'...52 ss een ames ee 1 Coarse grained, pure, calcitic.. ..... arate lod eevee kee naee i 1.5 Fine grained, brown, micaceous, dolomitic. .............. tO me eH bo Ot bo & bo ROCKS OF THE REGION 351 Feet Inches Bette Stained, PUTE CAICILIC.. <9 02. sac cence ns cere annie ns aie SO Fine grained, micaceous, dolomitic...............5. see 2 Faulted, beds disturbed, calcite veins..................4. 10 me Prmiticn. GOIOMILIC DEGS.... 5... - 0s cee wen decceue pe 12 Medium grained, light blue, calcitic..................... 4 ER ae Cem el Ds cae ald sn wate eww lew Fe wine ate 73 The different layers within the formation vary widely in coarseness of grain from the fine, almost statuary, marble obtained at the Beaver Dam quarries at Cockeysville to the coarse so-called alum stone, in which the invividual grains may reach a diameter of one-half, three-fourths, or even 1? inches, as is found in many parts of the region. So far as any general- izations can be drawn regarding the variations of texture, it appears that the dolomitic or magnesian-rich varieties are finer grained and more com- pact as compared with the purer lime carbonate rocks,'which are generally coarsely crystalline. Another feature of these rocks is the presence of im- purities along fairly well defined lines which have for the most part been recrystallized into magnesium silicates. These lines ofimpurity may sep- arate the different lime-rich beds from those rich in magnesium, or, what is more commonly the case, dolomitic layers from each other.. It seems to be rather generally the rule that the impurities are more intimately asso- ciated with the layers rich in magnesium than with those rich in lime, but frequent exceptions may be found to this statement. Among the most common accessory minerals found in this formation are phlogopite, biotite, muscovite, and iron pyrites; tremolite, quartz, and occasional tourmaline are, however, more frequently found within the limits of in- dividual beds. The general rule employed in field work, based upon innumerable acid tests, has been that the coarser grained beds are calcitic and the fine grained beds are dolomitic; each type is white, cream, brown, or some- what dirty in color. WISSAHICKON SCHIST This formation consists of a series of highly micaceous, very schistose, and often crinkled aggregates of quartz, more or less chloritized biotite and garnet, with accessory orthoclase, cyanite, staurolite, etcetera. With the increase of feldspar the rock passes into a gneiss. This, however, is less distinctly banded than the gneisses of the Baltimore formation. The individual beds in this type are only indistinctly marked, and their sepa- ration from the well-defined lines of foliation is often attended with con- siderable difficulty. ‘The dip of the foliation varies somewhat, but is usually about 45 degrees, while the effective dip of the individual beds, - Ratt | od van aa , 352 E. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE which themselves are highly crinkled, may vary from 0 to 90 degrees, representing in their general structure open folding, but in their intimate structure the folds are frequently much compressed and often overturned and slightly faulted. IGNEOUS ROCKS Through these several sedimentary formations have been intruded igneous rocks ranging in composition from granites to peridotites. The main development of the igneous rocks, however, lies outside of the field selected, only the extreme members being represented, in the granite between Warren and Cockeysville and in the serpentinized peridotite of the Bare hills. These igneous rocks were intruded either during the folding of the rocks or subsequently. There are also several lines of diabase boulders representing aie which have been correlated with the great Triassic intrusions farther north. AREAL DISTRIBUTION OF THE Rocks IN GENERAL The rocks of the region, as shown by the accompanying sketch map (figure 1), may be roughly grouped into three broad belts extending parallel to the strike of the formations from northeast to southwest. On the south are two anticlinal areas of the Baltimore gneiss, separated from each other along the strike by a fault and sharp folding, while on the north is an oval anticlinal area of Baltimore gneiss about 15 miles in length and 5 miles in breadth in its widest point. Between these two areas in a broad synclinal trough occur in regular succession the Setters quartzite, Cockeysville marble, and Wissahickon schist, the latter occu- pying the larger portions of the area and connecting on the north, east, and west with the larger body of Wissahickon, which extends diagonally across Maryland from the southwestward continuation of the type Wissa- hickon of Pennsylvania to the Potomac river on the west. DISTRIBUTION OF BALTIMORE GNEISS Southern area.—The southern development of the Baltimore gneiss on the south side of the Greenspring and Mine Branch valleys falls into two distinct areas, separated by the limestone and Wissahickon valley of lake Roland and possibly by a fault. On the west is a lenticular anticlinal mass entirely surrounded by the Setters Ridge quartzite, wherein the bands of Baltimore gness stand at a high angle, ranging from 40 to 75 degrees at the center, and strike parallel to the major axis of the ellipse except near the ends, where their strike follows the general contour of DISTRIBUTION OF ROCKS OF THE REGION 353 the areal distribution, and the dip lessens to between 10 and 30 degrees. The exposures within this area, which rises to the level of the Piedmont plateau, are not good, and it is not possible to make a detailed correla- tion of any of the beds found within it. The facts observed, however, clearly indicate that we have here, as is so often the case, an anticlinal zone which plunges downward at either end. The best exposures of the formation are those along the northern part of Park Heights avenue between Eccleston and the road from Pikesville to Rockland, which is known locally as the “ Old court road.” The eastern half of the Baltimore gneisses on the southern borders of the area extends from the north and south fault along the Northern Central railroad between Hollins and Sherwood eastward to Towson, and thence across the Gunpowder river to the vicinity of Glenarm, where it terminates in a steep, tightly pinched anticline. On the south the limits of the Baltimore gneiss have not been entirely worked out, but the detailed mapping of the late Professor Williams would indicate that it is bordered by the quartzite which extends from the Northern Central railroad near Mount Washington to lake Montebello, within the north- eastern limits of the city of Baltimore, where the crystalline rocks are covered by the later unconsolidated deposits of the Coastal plain. On the east this formation is bordered at first by the quartzite and lime- stone, but these successively pinch out within a mile or two of the eastern nose of the anticline, and do not appear along the deeply cut trench of the Big Gunpowder river. It is probable that the southern limit is a a strike fault or the contact with the large gabbro mass which extends in a northeasterly direction across Maryland from the eastern limits of Baltimore city to the Susquehanna river near Darlington. Throughout this region of Baltimore gneiss the exposures are poor, due to the high state of cultivation of the land, its plateau-like character, and the presence of numerous well kept country estates. There are, however, numerous exposures along the Gunpowder and in some of the other streams, but the gneisses at this point are intricately penetrated by nu- merous granitic and grabbroic intrusions. Northern area.—The northern area of Baltimore gneiss is broadly an ellipsoidal mass, representing a large anticlinal dome, which, like the smaller one of the south, plunges at either end. This plunging of the anticline to the westward brings the overlying formation down to the surface of the country, and thereby causes the surface exposure of the Baltimore gneiss to narrow rapidly to the westward of the Northern Cen- tral railroad. The marked differences in character between the quartzites, limestones, and Wissahickon schists allow the working out of the struct- ure in greater detail than is possible in the eastern half of this anticlinal 354 5. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE dome; but the observations within the Baltimore gneiss in the eastern half of the lenticular area, to the eastward of Monkton and Phcenix, are entirely in accord with those of the western portion, and indicate that the gneisses occur in numerous tightly pinched folds, possessing a com- mon strike parallel to the major axis of the dome and steep dips, some of the time to the north and some of the time to the south. On the whole, the northerly dips predominate, indicating that the anticline is somewhat overturned to the south. Numerous exposures of the gneiss are found along the Big and Little Gunpowder rivers and their tributaries, but they are few and unsatis- factory, except for areal mapping, in the plateau portions of the region which is a prosperous farming community under a good state of cultiva- tion. The exposures encountered show that the Baltimore gneiss is penetrated in the northern area by igneous intrusions, now much meta- morphosed, of granitic and gabbroic materials. The former are altered to granite-gneisses and the latter to hornblende-schists. DISTRIBUTION OF SETTERS QUARTZITE General characteristics of the exposures.—This quartzite formation, with its local variations toward a garnet-schist, occurs on the borders of the Baltimore gneiss area, usually as long, narrow ridges, rising steeply from the level of the limestone valleys to the surface of the plateau or upland, and usually may be found outcropping where the streams have - cut across the strike. Its exposures are highly characteristic, though generally poor, on account of the small rhomboidal form of the frag- ments into which it breaks when fine bedded and somewhat micaceous. When highly quartzose and more compact it is likely to be confused with the Baltimore gneiss, and when very micaceous and carrying gar- nets it is likely to be confused with the Wissahickon schists which overlie the limestones. Southern area.—The ocurrence of the quartzite about the Baltimore | gneiss dome in the southwestern part of the area may be traced almost continuously, by means of fragments, around the entire zone, and ex- posures in which it is possible to obtain the dip and strike of the beds may be found frequently. In this area, the typical area for this forma- tion, the rock is characteristically thin-bedded, the beds being separated by films of sericitic mica. Along the northern border of the anticline from Rockland to Red Run the rock dips uniformly northward under- neath the marble at an inclination of about 70 degrees. On the south the dips are less uniform, being sometimes to the north, but generally to the south, indicating a minor folding in the beds and some overturn- DISTRIBUTION OF ROCKS OF THE REGION 355 ing, as is brought out more clearly by the small stringers of marble found infolded with them. On the end of the anticline to the west the beds change rapidly in strike from east-northeast to northeast, through north and northwest to west-northwest. On the eastern end the strikes of the beds change similarly from east-northeast to northwest, through north to north 45 degrees east, the dips in all cases being away from the under- lying Baltimore gneiss and beneath the overlying marble or Wissahickon schists, as the case may be. ‘The eastern exposures of quartzite forming the ridge on the south side of the limestone valley extending from Sherwood to Glenarm are less satisfactory than those just described, but wherever observed indicate a northern dip of somewhat less inclination until near the eastern end of the anticlinal fold, where the beds rise steeply on the north with a ‘northerly dip of 75 degrees and an easterly dip on the south of about 40 degrees. The anticlinal character of the structure at this point is usuaily well brought out for Piedmont conditions by the formation of a triangular hill of quartzite produced by the nose of the fold, which is cut through by Long Green creek about a mile from its apex. The - strikes of the quartzite may be traced at this point along the top of the ridge, where they show progressive changes in position through all azimuths from north 45 degrees east, through west and north, to north 30 degrees east, and even reach north 60 degrees west near the crossing of the Harford turnpike. The character of the quartzite in this part of the fold is not that typical of Setters ridge, but shows the development of more mica with accessory garnets and occasional cyanite. Throughout this entire southern region the quartzite shows the average thickness of about 500 feet. Northern area.—The areal distribution of the quartzite about the north- ern dome of Baltimore gneiss is much less constant than is the case about the southern areas, and there are many evidences of a marked erosional conformity and a few strike faults in this part of the region. The occurrence of the quartzite may be traced almost continuonsly from a few miles west of the Northern Central railroad near Cockeysville along the southern side of the road to the northeastern nose of the fold on the road between Taylor and Jarrettsville, where a well defined V-shaped hill is formed by the upturned beds of the quartzite, and thence westerly on the northern side of thedome. The quartzite is not distinguished in the cut of the Northern Central railroad north of Monkton, but the typical tourmaline-bearing mica-schist is found outcropping on the hills across the river, and thence may be traced along the westerly side of the dome in several bands to the southwestern nose of the anticline, where it may XLVIII—Butt. Geo. Soc. Am., Von. 16, 1904 356 E. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE be found in a few poorly developed exposures. The exposures along the northern boundary east of the Northern Central railroad are very poor, and the presence of the quartzite is only actually established at a few points, owing to the cover of soil and the close similarity of the Balti- more gneiss at this point. Itis quite possible that the quartzite is cut out locally, as it can not be recognized in the well exposed cuts along the railroads just north of Monkton. If this is so, the lack of quartzite only occurs for a short distance along the strike, as it is exposed in the road from Monkton to Hereford, just west of the Gunpowder river, and from this point may be traced without interruption through Pine Hill to the sharply folded region between there and Glyndon. Along the southwestern limits of the Baltimore gneiss the quartzite ig seldom found between it and the overlying marble, and wherever so found it is usually very thin and poorly developed. It has been noted along the northern edge of the Worthington valley near Councilmans run and a little farther north above Slades run, but is apparently lack- ing along the contact between the limestone and the Baltimore gneiss as exposed along Westernrun. Such a rapid thinning of the quartzite from a thousand feet or more in the ridges between Butler and Stringtown to zero in Western run, a distance of less than 2 miles across the strike of | the folds, is quite unusual, but no facts were found indicating a fault, and many observations point to a rapid thinning, due apparently to an erosional unconformity. The unusual and rather peculiar development of garnet and garnet-mica-schists interbedded with the quartzite and apparently con- stituting an upper member of that formation is best exhibited in this northern region, especially between Pine hill and a point i mile northwest of Butler, in the Stringtown valley. The quartzitic layers at this point are not well developed, and the outcrops along the sides of the hill strongly suggest the Wissahickon schist. The true position of this bed is, however, shown by the folding of the quartzite-garnet rock and limestone northwest of Belfast along Buffalo creek, where the areal dis- tribution and structural observations seem to exclude the possibility of a fault and demand the interpolation of an upper member in the quartz- ite formation. This same conclusion is the most satisfactory deduction from the observations made in the quartzite formation along the gorge of the Gunpowder between Warren and Royston branch, where the gar- netiferous rock interbedded with quartzitic layers is found resting on more quartzose beds, and beneath the marble the whole conforming with the structural relations shown by numerous exposures of marble in the valley of Royston branch. The detailed character of the structure at idle. DISTRIBUTION 357 this latter point will be more fully discussed in a later portion of the paper. DISTRIBUTION OF COCKEYSVILLE MARBLE Areas in general.—The Cockeysville marble lies in a synclinal trough between the southern and northern Baltimore gneiss areas and in the synclinal folds on the northwest side of the northern anticline and out- crops frequently within the limits thus outlined wherever the formation is not covered by the Wissahickon schist. The occurrence of the car- bonate rocks at the surface is always marked topographically by the occurrence of limestone valleys, most prominent among which are the Greenspring and Mine Bank valleys, lying to the north of the southern Baltimore gneiss, between it and the overlying Wissahickon. Between these two valleys and the corresponding valleys farther north the Wissahickon gneiss has been removed, giving a very low divide in the drainage system underlain by crystalline limestones extending from Lutherville to Cockeysville. This, together with the narrow portion of the Worthington valley and that of Green run, which border on the southern flank of the northern anticline, represent crudely a recumbent letter H. The limestone also extends northeastward from Lutherville, forming the Dulany valley, which in turn has a small offshoot of the — limestone (the complementary flank of a small anticline) which runs up the valley of the Gunpowder to the mouth of Royston branch, where the limestone leaves the Gunpowder valley and occurs in a gentle anticline in the valley of the smaller stream. From the western end of the Worthington valley the limestone wraps around the narrow nose of the northern anticline and outcrops in a series of narrow parallel valleys, separated by anticlinal ridges of quartz- ite and gneiss or synclinal areas of Wissahickon schist. Beside these larger areas, which may be traced as one continuous mass, there are three smaller areas, separated from the larger. The largest of these is that forming the Long Green valley, which apparently is only a portion of the Dulany valley and Glenarm bodies, from which it is sep- arated on the higher land by the overlying Wissahickon schist. The second area lies south of Taylor, and is apparently separated from the Green Run arm of the main mass by a strike fault. The third area is* represented by a single cutcrop of very small extent, occurring beside the road just east of Glencoe station. Greenspring valley—The marbles of the Greenspring valley extend from west of the Reisterstown turnpike and the Western Maryland rail- road eastward to the Northern Central railroad,where the valley broadens, reaching out into the various valleys already described. The marbles 358 E. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE in this region are found in small outcrops, where they strike parallel with the axis of the valley and dip northward at a decreasing angle as one passes from the south and west to the limits of the overlying Wissa- hickon formation. The dip is, however, generally steeper than 45 degrees. As the limestone circles round the southwestern anticline the strikes change in conformity with the contour of the major fold, and the dip is uniformly away from the quartzite and beneath the Wissahickon schist lying to the west of Lake Roland. Between the eastern and western ends of the anticline the limestone is compressed within minor folds in the quartzite at Mount Wilson and in several places along Moores branch. It is lacking, however, in its normal position (between the quartzite and the Wissahickon) from a point 1 mile east of Cockeysville to the western limit of the fold, with the exception of the single exposure, already re- ferred to, occurring at Mount Wilson. In the valley above Lake Roland the limestone is only exposed once or twice, as at the junction of Jones falls and Rolandrun. Such structural observations as can be made are in accord with the synclinal structure of this small southern offshoot of the Greenspring valley. The limits of the limestone south of the Green- spring valley are determined on the east by a north and south fault passing from Sherwood through Ruxton to Lake Roland and thence into: the lowland above Mount Washington. The exposures of marble from Sherwood to Glenarm show relatively simple monoclinal dips to the northward, except in the region southwest of the Wissahickon, where this marble unites with that of the Dulany valley, and in the vicinity of Glenarm, where the limestones fold sharply around the upturned anti- cline of quartzite already described. At each of these points there is minor folding, and the local structure is much confused in its detail, although harmonizing well with the broader structure as here outlined. The exposures for the most part are poor and occur almost exclusively in small private openings, where the stone has been extracted for lime. Dulany valley—This valley, which extends for 5 miles northeasterly of Lutherville, with an average width of from 1 to 2 miles, shows numer- ous exposures west of the Gunpowder river, but is almost entirely lack- ing in the same from the Gunpowder to its easternmost limit. Enough observations, however, have been made to show that the limestone is here very flat, with several minor crests and folds extending parallel to the longer diameter of the valley, the limestone dipping beneath the Wissahickon on the north at a varying angle. North of Merediths bridge, where the Jarrettsville turnpike rises from the limestone valley to the level of the plateau, the strike of the lime- stone changes rapidly through west and northwest to a little east of north, following the course of the Gunpowder river. From the vicinity of | DISTRIBUTION 359 Overshot run the strikes again change to west of north, following the course of the Gunpowder and Royston branch, the latter curving with the limestone as it wraps around the small anticline near Warren. The dip on the east side of Gunpowder valley is northeasterly under the Wissahickon, in all instances observed, but along the west side there seems to be some slight overturning, the limestones dipping at times to the westward beneath the quartzite, which normally lies below the lime- stone. The exposures in this offshoot from the main limestone valley are rather better than usual, and show with unusual clearness for Pied- mont conditions the shifting stripes and dips due to the folding of the region. Thisis particularly true in the area northeast of Warren, which is discussed more fully at another place. Worthington valley—Butler area.—The marbles which are so well devel- oped in the. valley between Lutherville and Cockeysville extend west- ward from the latter point along the south side of the northern anticlinal dome, widening about 5 miles west of Cockeysville into the Worthington valley, which is in reality the southwestern nose of the northern anti- cline, as already described. The exposures of limestone in the narrower portions of the valley are rather unsatisfactory, but show an east-and- west strike and a dip of 40 to 60 degrees away from the Baltimore gneiss and beneath the Wissahickon schists on the south. In the wider por- tion of the valley the dips become much less, reaching as low as 5 and 10 degrees. The strike also changes, as the limestone of the valley wraps around the anticlinal axis, from east to west and northwest and north to east of north. In this part of Worthington valley the natural expos- ures are few, but the solid rock has been exposed in many places by small quarries made for the extracting of limestone for agricultural purposes. Extending northeastward from Worthington valley the limestone is found in three well defined bands forming narrow valleys. These dif- ferent bands are representatives of the same formation brought to the surface again and again by the folding of the beds along the level of the present surface of the country. The dips are often steep and sometimes overturned, but the bands unite to form a continuous valley in their southwestern limits. The strike in every instance appears to be parallel to the valleys. The more southerly bands are separated by a synclinal of the overlying Wissahickon, and are in turn separated from the north- ernmost bands by a tightly compressed anticline exposing the Cambrian - quartzite and the underlying Baltimore gneiss. These limestone areas do not extend east of the Northern Central railroad, and in only one instance are they found east of the Baltimore and York turnpike. The relations existing along their eastern limit are not determined with eg Pe! ey os 7 d 560 E. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE entire satisfaction. The areal distribution, dips, and strikes suggest — that we have here the emergence of a syncline above the surface of the country, but the underlying quartzite which one would naturally expect to find bordering the limestone is absent, and the cause for this absence is not entirely evident. It is easily recognized in the field that the quartzite formation is thinning rapidly as one passes across the strike from the broad exposures in the hills south of Stringtown to the thinner development bordering the Butler-Belfast valley. Moreover, to the southward the limestone rests immediately on the Baltimore gneiss. These facts would seem to indicate that the quartzite did not extend over the entire area of the Baltimore gneiss beneath the marble at the time when the limestone was laid down. If this inference is true, it is easy to explain the non-occurrence of the quartzite on the eastern bor- der of the limestone, and possibly to define the limits of deposition of the quartzite in this local area. Unfortunately, as is so often the case in the southern Piedmont, the exposures along contacts are very poor and frequently wanting at the critical point. DISTRIBUTION OF WISSAHICKON FORMATION The Wissahickon formation, which overlies the marble, occupies the remainder of the region, occurring in broad areas between the different limestone valleys already described. The schists and gneisses of the formation extend entirely around the northern anticline and occupy very much of the region between it and the southern anticline. The removal of the Wissahickon across the axis of the synclinorium along the course of the Northern Central railroad between Lutherville and Cockeysville separates the Wissahickon, however, into an eastern and western portion. The western representative of the Wissahickon, lying between the northern and southern anticlines and the Western Mary- land and Northern Central railroads, forms a series of well rounded hills, which reach to the level of the plateau along their summits. The ex- posures throughout this region are poor, the material of the Wissahickon formation yielding a good soil and breaking down easily to a protecting mantle over the readily disintegrating garnet mica-schists. It is possi- ble, however, to recognize that in this general basin are one or two minor folds, giving an anticline across the area a little south of the center and two minor synclinal axes just within the limits of the Wissahickon- Cockeysville marble contact. The more eastern area of Wissahickon lying on either side of Dulany valley and extending thence northeast appears to be somewhat more complex. The rocks in the area about Loch Raven appear more gneissic and even approach the Baltimore gneiss in appearance, while the schists GENERAL CHARACTERISTICS 361 on the north side of Dulany valley in the vicinity of Blenheim differ somewhat from those exposed farther west. The change is due in part to the presence of a small area of intrusive meta-gabbro. The dips and strikes, as is customary in the Wissahickon formation, are rather variable; but in their cumulative effect they show a syncline overturned to the southward in the vicinity of Loch Raven and a gentle normal syncline forming the ridge in the vicinity of the Jarrettsville turnpike. Farther westward in the high land between Royston branch and Green run the structure is more complex, and will be described in more detail when discussing the structure in the vicinity of Warren. STRUCTURE GENERAL CHARACTERISTICS The structure of the Cockeysville marble area is not thoroughly un- derstood without a consideration of its relations to the general structure of the eastern part of the continent.* As is well known, the general tec- — tonic lines extend northeast from the south across Virginia until they reach the limits of Maryland, when the strike of the various formations is deflected to a more easterly position, sometimes even becoming east and west. This general easterly trend of the formations passes in a broad band, reaching from central Pennsylvania to the Atlantic, and gradually returns in New Jersey and New York to its original north- easterly trend. The Maryland area,and particularly that portion north of Baltimore, is in the concave side of this major fold, a fact which ex- plains certain of the structures found within this area, which appear to be somewhat unlike those described from other parts of the Piedmont and which doubtless led the late Professor Williams to an accentuation of the oval-shaped figures for his different formations. The occurrence of the various forces involved in the curvature of the general structure have produced locally within the Maryland area conditions favorable to torsional deformation, since the lines of distribution of the forces are not parallel, but slightly inclined to each other. It is for this reason, in part, at least, that the various folds, which are usually very long and narrow throughout the Appalachian region, are here rather short and dome-like, with intervening areas of less compressed folding. The effects of these differences in conditions will be brought out more fully in dis- cussing the faulting. The structural character of the Cockeysville region has already been given in describing the areal distribution of the various formations, but it is well to recall the fact that it consists essentially of anticlinal domes * These are more fully discussed by the senior author in the preceding paper, pp. 334-335, 3862 5. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE separated on their sides by synclinoria, the axes of which run parallel to the general structure of the larger folds, which are found farther to the north and west, where they partake of the structural characteristics of the eastern portion of the continent. FOLDING The folding of the rocks within the region, as opposed to the minor textual features, such as crinkling, cleavage, and schistosity, may be broadly characterized as consisting of a series of open folds of simple character, the individual portions of which are marked by numerous sharply compressed smaller folds, with axes parallel to those of the gen- eral structure. The character of the folding is most readily seen by an examination of a few exceptional localities and the areal distribution of the various formations involved. Its not easy, usually, to recognize the structural - characteristics from given exposures, though these may generally be de- tected within the individual formations when the general structural lines have been determined. This fact has rendered most attempts to work from the more detailed to the more general barren of structural results. The simple open character of the major folding is well indicated in the structural sections across the face of the accompanying map, but the detailed complexity of this exceedingly intricate region is only shown diagrammatically. A feature of the folding which should not be overlooked is a tendency toward unsymmetrical folds and often to overturned folds. The unsym- metrical folding seems to be a property characteristic for the entire re- gions, though never very sharply brought out, while the overturned fold is a feature of local development, as along the south border of the south- west anticlinal dome, where there is an intertonguing of the quartzite and marble, with dips indicating overturned folds. This overturning is also shown in many minor folds of the northern anticlinal dome, but the similarity of the beds and the poor exposures in this area render it difficult to do more than decipher the local structure here and there, The intervening well cultivated fields or soil-clad forested areas make it impossible to carry minor structures across the folds. FAULTING The faulting in the area, so far as it is seen, introduces one of the most interesting structural features noticed and brings out the influence of the continental structures on this region. Although many small faults of slight throw may be detected during the course of field study, only three faults of considerable magnitude have been noted, and these are 9 FOLDING AND FAULTING 363 not very clearly shown except by the areal distribution and contiguous structural details. These faults, as shown on the general map, lie in a curved line passing from the Little Gunpowder near Hess to the valley A=-GRANITE, 6=GABBRO i CRYSTALLINE LIMESTONE, PSEUDO—WiISSA— HICKQN SCHIST. SET TER'Ss QUARTZITE, BALTIMORE vA = Ow SSeS aes SEAS eV PRR ae SOR. mals N Z aN Wa 2 oS SG oye X23 cs a seceere 266 Tope ay BES era uae . STRIKE, Ae Ne v MILE. SECTION ALONG LINE AB. SSS Sere [TLL Ficure 1.—Map of Vicinity of Warren. of Jones falls in the vicinity of Mount Washington. They are all in the nature of thrust faults, but the almost complete absence of anything approaching contacts renders it difficult to demonstrate this fact. There XLIX—Butt. Geox. Soc. Am., Vou. 16, 1904 064 5. B. MATHEWS AND W. J. MILLER—UOCKEYSVILLE MARBLE is, moreover, a torsional feature in these faults which is of especial inter- est. The thrust at the northern end of the fault is slightly to the east- ward, as in the vicinity southwest of Hess, where the Setters quartzite may be found above the Wissahickon and limestone if the latter is present. Along the central and southern portions of the fault line, how- ever, the thrust is westward. The relationships may be most thoroughly studied in the valley of the Gunpowder at Warren, where it is quite clearly shown that we have a somewhat complex anticline plunging to the northward with a compressed fold in a line passing through th town of Warren. The anticlinal character of the folding is evident (figure 2) east of Warren near the mouth of Royston branch, where the Gunpowder flows in a gorge cut through the Setters quartzite and ‘‘ pseudo Wissahickon ” (the garnetiferous upper portion of the quartzite formation) and Cockeys- ville marble to the contact between the latter and the underlying Wissa- hickon, which it follows southward to Dulany valley. The strikes and dips as shown in the valley of Royston branch indicate at this point a northward plunging anticline, the west limb of which is replaced by the fault under discussion. The Wissahickon schists may be traced con- tinuously around the anticline to the Gunpowder river immediately west of Warren bridge, where they are found striking to the south and dipping to the west or east, as the case may be. Near the quartzite on the south side of the Gunpowder the strike is southwesterly, and the dip is toward the east as a result, apparently of the overriding of the quartz- ite at this point. In the stream bottom beside the road leading from Warren to Cockeysville on the line of the fault is a recemented breccia, which indicates a portion, at least, of the fault zone. The structural features of this locality indicate that the forces at work were northwest and southeast, and that the thrust, which here is slight as compared with that farther south, carried the Baltimore gneiss, quartz- ite, and marble across the Wissahickon formation. The structural features along the southward continuation of this fault are much obscured by the intrusion of a granite mass which forms the eastern boundary of the Wissahickon and probably occupies the eastern side of the fault, where one would naturally expect the Baltimore gneiss if there had been no granitic intrusion. The southern fault, which extends from near Lutherville to south of Mount Washington, is similar in character to that already described about Warren, but much more pronounced. | The accompanying sketch map * (figure 2) of the valley of Jones falls *The relative position of the detailed maps may be indicated approximately by placing the southwest corner of the neat line of figure 1 over the northeast corner of figure 2. Both figures oriented north and south, a a | § FAULTING - 365 between Rockland and Hollins, two stations on the Northern Central railroad, shows in detail the complexity of the structure in this area as well as the observations on which the present interpretation is based. re °: Jiggs: eo eo OfOML eo? @ ee — CRYSTALLINE Li MESTONE, Sec hionm ALONG KLINE CD. Figure 2.—Map af Vicinity of Rockland. On the west is the eastern end of the southwestern anticlinal dome alread y described, and on the east the southerly extension of the fault under dis- cussion, Earlier work in the region led Doctor Williams to interpret the 2 Soe Mele a ee 366 5. B. MATHEWS AND W. J. MILLER—COCKEYSVILLE MARBLE eastern termination of the western anticline as a fault complementary to the fault recognized by him and the authors on the eastern side of the valley.. It was supposed by Professor Williams that this valley repre- sented a faulted block of limestone. That such is not the case is well shown by observations recorded in the accompanying map, where the strikes and dips may be traced with constantly changing azimuth about the nose of the anticline from Rogers station to the southwestern corner of the sketch map. The structural details in the center of the valley in the limestone indicating a southward plunging syncline also corroborate the later interpretation. The original view evidently arose from the fail- ure to recognize the difference between the overlying Wissahickon and the underlying Baltimore gneiss. Although no exposures are found along the actual fault line between Sherwood and Brightside, the sharp divergence in strike and dip as well as the difference in character of the rock leaves no doubt as to the occur- rence of a fault at this point. The ends of the various beds of the Balti- more gneiss strike northwesterly against the limestone and Wissahickon, which have a more southerly trend and a dip to the westward. The westward thrust of the eastern anticline widens very perceptibly (see general map, plate 65), the distance between the quartzite areas representing the limbs of the anticline. ‘They are fully twice as far apart as in the corresponding anticlinal dome on the western side of the val- ley. If the foregoing parts of what appears to be a single fault zone be regarded as separate faults, they may be characterized as follows: The northernmost fault differs from each of the others in some particulars and in other particulars is ike them. Like the Warren fault, it occurs parallel to the strike of the major anticlinal dome, but, unlike both the Warren and Ruxton faults, the thrust is toward the southeast. This dif- ference in the direction of the thrust within the distance of a few miles would be a serious matter to explain without a knowledge of the conti- nental structure to the westward. With these facts at hand the relation- ships become clearer. All three of the faults show the older, more crystal- line, and more competent Baltimore gneiss thrust over the younger and more yielding marbles and Wissahickon schist. In the faulting the quartzite is associated with the underlying gneiss, with which it is litho- logically very similar. The difference in thrust at the different points may be produced by the slightly divergent lines of force which have pro- duced compression and local glancing blows resulting in a small amount of contortion. Thus on the north the Baltimore gneiss is shoved south- ward and on the south it is shoved northward, while in between is a less marked faulting, which partakes of the westerly thrust from the south, but is here not as strongly marked. BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 66 IsLAND, New York Figure 1.—Jacop SAND OVERTURNED AND RECUMBENT ON HEROD GRAVEL; GARDI Gardiner clay at right of part of fold shown Figure 2.—SemMI-strRaviriep ‘ype or Monravk Drirr; Monrauk Yornr, New York JACOB SAND AND MONTAUK DRIFT BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 367-390, PL. 66 JUNE 23, 1905 GEOLOGY OF FISHERS ISLAND, NEW YORK* BY MYRON L. FULLER (Presented before the Society December 31, 1904) CONTENTS Page Location and general relations............ SUdaiel weds cthaias 5 ds £34105 S45 USS 368 MONI 5 i ook. ods ic 5 u'peciss Sees daeess sme Ghdwit siaekd 9 98 aePeasou a sas 369 DME iis hie ck css ess vee St a aes fy ERE PE Oe See 371 oo eee 1 SER Ep ie faeeng 1 Sean agers eae 371 ner and mantery Of tie Geposits. ...2...... 02. c ee cence ce eeceeenns 372 Ce foeciie te St soe 2 a aw swish cos aed wok nce ee noes 372 ENING Sa ihe a hc an bam k Se wine Sp si jows Ooe eee antic a atece 372 eUREREGG GAD SAINIECO PTAVEIE. ©. 2. oc. coe coe ce cece be sccesecces 373 EE ee ee ee See pe Ore: 375 Re Ons or ido t ies a tai Sas ag uals cna sess 375 ENE ARMM EMIING choy has icia sss a oles been oer et esac ncee 375 SS “Sa oe 376 ST RE OCCURPENIFE =D so bce nce ace dteces Rearend wid ae 377 eS SENET Dai, | in Uae ee ee, oe re ae 378 NN, MESPEREIE ME th Ser, Ske Sil eesini Cas n's'co eee sé Ba s00° aie 378 epPUE READIN UP OMMIWIRITIGM. «occ 's 5 wee a's anit Sees anes dloascs ne . 378 NM EMP RMEPEREEAII Soe iene cit ow c'e tan eis sels ewb.d oe wie sae 379 I SEAN UE eg i a nv oe dhis's o'xicjn's © diem vw ah om eai ens 380 Se ae ee ee ea tae lave APRH eS: x SS es 381 ee aC a aa oe a 381 EC aa a en ee 381 a ee a ee 381 NN RIE, AIEEE ee an rhe Due 2ida oie PAS «om one sds chee 382 Montauk drift..... Te fla a co Sadan OMe « eee Ree nna i ae Pas € « 382 RRM ae oe) ay nice aia VE er nes owe we eee 382 DE RECMNA AME, CRCIISL IONE a oss Se ees oo no PPM te as BE SaaS 383 Character and occurrence of the Montauk drift................. 383 nNNae ON SEMPEIIEER OWONICD. Sy ce ce a sn pe edeeceldcpcveneeseescesee 384 NE NE Ga tects aro i'w ow mip b e'e 0 sine oie Sadelvs ee wees 385 ee eee eth k Sin SU LeS Sut celt eee a ess bees 386 ee a alls cinale) ainpicig dip gscccmeoosspaeneacce- 386 Relation of Gardiner clay, Jacob sand, and Herod gravel to “ Sankaty beds” (Woodworth) of Marthas Vineyard and Nantucket.......-..... 386 * Published by permission of the Director of the United States Geological Survey. L—Butt. Grou. Soc. Am., Vo. 16, 1904 (367) 368 M. L. FULLER—GEOLOGY OF FISHERS ISLAND Page Correlation with the drifts of Pennsylvania, Mississippi valley, and Canada. 387 Correlation table .. 0.0.0.5. 060. bs oe cee «Ben cle sienna ss one rr 389 Distribution of pre-Wisconsin formations represented on Fishers island. . 390 LOCATION AND GENERAL RELATIONS Fishers island, although belonging to New York state, is situated off the coast of the eastern portion of Connecticut, the western end being 7 miles southeast of New London and the east extremity within 2 miles of Napatree point, at the extreme southwest corner of Rhode Island (figure 1). The island is a little less than 7 miles in length and has an average width of somewhat over half a mile, although near Clay point and west eo FISHERS |. 9 € PLUMS eatee Figure 1.—Index Map of Fishers Island. Showing its location and relation to the principal moraines of the Wisconsin stage. of West harbor its width is upward of a mile (figure 2). It lies in the prolongation of the structural axis of Orient point, Long island, which axis, after passing through Plum and Fishers islands, is recognized along the ridge extending eastward from Watch hill, Rhode Island. Its posi- 9 LOCATION AND TOPOGRAPHY 369 tion is masked throughout by a mantle of drift of the inner Wisconsin moraine. The island has the general sinuosity of coastline and inequal- ity of topography which would be expected to result from the partial submergence of an irregular ridge. Such, in fact, has doubtless been its origin. ToPoGRAPHY The topography of Fishers island, like its coastline, is highly irreg- ular, its eminences being without apparent system other than the linear arrangement due to the shape of theisland. Some ofthe hills, like those along Isabella beach, have a well marked northeast-southwest trend. At other points, as southwest of Chocomount cove, the hills occur as isolated knobs, or again, as between Hay and West harbor, they occur as somewhat general but broken elevations. vile -~ eve cs : e 7 North ee 2 Suth Dumpling D q Flat Hammock Figure 2.—Topographic Map of Fishers Island. Seale about 2 miles to an iuch; contour interval, 20 feet. _ The general aspect of the topography is decidedly morainal—an appearance which is heightened by the presence of a considerable num- ber of large erratic boulders. An analysis of the topographic features in the field, however, shows that, while there are many minor knobs and numerous basin-like depressions which are sometimes of considerable size and contain ponds many acres in extent, the morainal features are, on the whole, mainly superficial, the broader lineaments of the landscape belonging clearly to an older topography, which in part appears to have been due to subareal erosion, especially at the western end of the island and near Clay point, Ree? 370 M. L. FULLER—GEOLOGY OF FISHERS ISLAND At other localities, as at the Isabella Beach hill and the hill forming the headland three-quarters of a mile to the northeast, the topography is more puzzling, and only after the unraveling of the geology was it determined that these hills were primarily constructional in form, repre- senting in reality great anticlines in the Pleistocene sediments. ; Some of the depressions occupied by ponds are evidently true kettles, representing the position of detached ice masses, which later become sur- rounded or buried by glacial outwash, but other basins appear to be due to the obstruction of older valleys by loose materials shoved up by the last ice-sheet, or to the excavation of shallow basins by its erosive action. Many of the more or less inclosed coves which occur along the coast of the island represent similar features which have been partly submerged. The morainal topography is not confined to any one part of the island, but is most marked along the north shore, where the knobs and kettles Figure 3.—Section through Hill Three-quarters of a Mile Northeast of eastern End of Isabella Beach. Showing anticlinal structure of hill and the extent of marine erosion. @ are not only more conspicuously developed, but where erratic boulders are present in the greatest numbers. On the south side, while morainal _ features are not lacking, the topography commonly shows smoother and more rounded outlines. The difference is evidently due to the fact that the ice during the deposition of the morainal materials, which belong to a retreated stage of the Wisconsin, reached in general only part way across the island, the south shore topography showing to a greater or less extent pre-Wisconsin forms slightly modified by the overriding ice dur- ing the maximum advance when the ice margin stood at Block island or beyond. | Topographic forms due to wave action are of relatively slight impor- tance. In general the bluffs are low, and erosion is only active in times of especially severe storms. Usually they are covered by talus slopes, which effectually mantle the outcropping strata. At Isabella beach and again at the point three-quarters of a mile to the north the bluffs are steep and clean. ‘The exposure at the latter point is of special interest, TOPOGRAPHY Be as it affords a basis for calculating the amount of erosion which has taken place since the sea reached the original southern base, the position of which, owing to the constructional origin of the hill, can be readily de- termined. In figure 3 is given a section through the hill showing its anti- clinal structure and former extent. A few small spits have been built out at a number of points on the coast of the island, and barrier beaches have been formed by the waves across some of the small reentrants, giving rise to certain of the ponds, but at other points, as in the vicinity of Silver Eel pond, the barriers are composed of boulders and appear to be in part, at least, the work of glacial ice. GEOLOGY GENERAL CONDITIONS Fishers island falls in line with the inner or later of the two Wiscon- sin moraines (figure 1), and as it exhibits a kettle and knob topography and has a surface of till, or at least a sprinkling of large erratic boulders over considerable areas, it has generally been regarded as a morainal island of late Wisconsin age. A careful study of the composition and structure of the deposits, however, brought out the fact, which was already suspected from the topography of the island, that the Wisconsin drift forms a thin superficial mantle resting on folded and eroded Pleis- tocene beds of a much earlier stage. Ficure 4.—Generalized Section of Fishers Island. a, Wisconsin till; b, Montauk drift; c, Herod gravel; d, Jacob sand; e, Gardiner clay; f, Jameco gravel; g, Mannetto gravel (?); h, Cretaceous clay ; i, light-gray granite. In working out the geology of the island, cliff sections furnished the greater part of the information, although the big clay pit on the east _ side of West harbor and an artificial section near the steamboat landing on the west side of West harbor afforded important data. In the field work all prominent bluff exposures along the coast were visited, the highways traversed, and such artificial sections as could be found examined. ) The field work brought out the fact that not only are the greater part of the deposits of pre- Wisconsin age, but that in the upbuilding of the 372 M. L. FULLER—GEOLOGY OF FISHERS ISLAND island a considerable number of stages, including both those of deposi- tion and erosion, have been represented. Of these at least three may be recognized in the portion of the deposits above sealevel, the first being marked by thick deposits of dark-colored clay, the second by the depo- sition of glacial gravels, followed by the deposition of till or by a period of folding, and the third by a long period of erosion. These events are recorded by the succession of strata and unconformities, as illustrated by the accompanying section (figure 4), and will be described in detail in the following pages, beginning with the lower or oldest bed. CHARACTER AND HISTORY OF THE DEPOSITS Granite foundation.—The formations exposed above sealevel on Fishers island are all of Pleistocene age, but in the Ferguson well, drilled by C. L. Grant, certain older materials were encountered. Record of Ferguson well Feet Gravel, boulders, and sand .............. 2.5 eS == NS 5477 wwe 4 27 Ficure 11.—East-west Section in Clay Pit. Showing details of folding in the sandy phase of the Herod gravels; clay at base. Except at the localities described, where the gravels are seen resting on the Jacob sands, the Herod beds can not be recognized with absolute certainty, although there is every reason to believe that the yellow and buff sands and gravels, which are seen at a great number of points be- - neath the thin coating of Wisconsin till, belong to this formation, form- ing, in fact, by far the greater part of the island. They probably reach their greatest development either in the high hills south of West harbor, already mentioned, or in the similar hill southeast of Chocomount cove, which rises to over 120 feet above sealevel and is probably composed mainly of Herod gravels. Montauk drift—Use of term.—The designation “ Montauk drift” is ap- plied to a sheet of drift from 20 to 50 feet or more in thickness, which nor- mally overlies conformably the Herod gravels. Owing to the fact that it was laid down in more or less direct connection with an ice-sheet which alternately eroded and deposited, local unconformities, due to contempo- raneous erosion and sometimes cutting as deep as the Gardiner clay, are notuncommon. The material is ordinarily partly stratified, but retains a distinct till-like appearance, owing to the big, probably berg-dropped, MONTAUK DRIFT 883 boulders present. It takes its name from Montauk point, Long island, where it is typically developed (plate 66, figure 2). Conditions of deposition.—So far as can be judged from the character of the Herod deposits, the accumulation was entirelysubmarine. Ante- dating the beginning of Montauk deposition, however, there was an up- lift, bringing the land in the vicinity of Fishers island somewhat nearer to the present level, although, as indicated by the occurrence of the Mon- tauk deposits below sealevel, except when brought up by folding, the level was still lower than that at present existing. The ice-front, which had been remote at the beginning of the deposition of the Jacob sands, had approached nearer during the deposition of the Herod gravels, and at the beginning of the Montauk substage actually invaded the area, passing in its maximum advance over Fishers island and southward to a point beyond Block island. The accumulation seems to have taken place beneath the ice-sheet itself in case of the more till-like deposits, where stratification is absent, while in the case of the semi-stratified materials it probably occurred outside but immediately in front of the margin. The thin beds of true gravels, which are occasionally inter- bedded with the more till-like deposits, probably represent temporary retreats of the ice-margin to points at some distance from the localities of accumulation. Character and occurrence of the Montauk drift.—The general features of the Montauk drift have already been outlined and the normal type illustrated ine plate 66, figure 2. On Fishers island it is of somewhat different character, although the exposures are typical of certain well defined phases near the type locality. The drift has only been recog- nized above sealevel at two points: (1) At the West Harbor exposures and (2) in the bluff three-quarters of a mile northeast of the north end of Isabella beach. The “ boulders” reported in the Ferguson well probably belong to this formation. At the West Harbor exposure (figure 10) the Montauk drift is proba- bly represented by the heavy bed of till resting unconformably on the upturned and eroded edges of the Jacob sands and Herod gravels. This unconformity is believed to be the result of essentially contemporaneous erosion by the same ice-sheet that laid down the drift. On Long island, where many more exposures are to be seen, there are similar evidences of slight deposition of till at this time, but the main epoch of folding was of a later date. At the second of the localities mentioned the exposures show a series of beds of clayey sand alternating with beds of pebbles about the size of a hen’s egg occurring in a matrix of clay orclayey sand. Both sand and gravel show the peculiar mixture of coarse grains or pebbles with fine LII—Bui. Grou. Soc. Am., Vou. 16, 1904 WN" 3 Pree Pe te hs i Low" 4 5 7 384 M. L. FULLER—GEOLOGY OF FISHERS ISLAND silts, which is characteristic of rapid and only partially assorted deposits such as accumulate at the margin of glaciers. It unquestionably belongs to the Montauk stage, representing certain of the more aqueous phases of deposition. A little farther to the north, although the contact is not seen, the semi-stratified deposits appear to be replaced by till. Where well exposed, it is usually not difficult to distinguish the Mon- | tauk from the Wisconsin drift, the latter, which is relatively sandy and of a buff color, contrasting quite sharply with the older bluish and more clayey drift. Inthe ordinary imperfect exposures, however, the two can not readily be distinguished, and although certain parts of the till exposed at the surface and at numerous points in the low bluffs of the island may belong to the Montauk, this can not be established with certainty ; in fact, it seems probable that most of the surface till is Wisconsin. Later pre- Wisconsin events.—The accumulation of the Montauk drift on Long island was followed by the deposition of a conformable series. of marine gravels, which are found at many points both in the eastern and western parts of the island. In the latter locality they are probably represented by the lower part of the Manhasset gravels of J. B. Wood- worth,* or that portion below the boulder bed so well exposed along Hempstead harbor. Following the deposition of these beds the ice, which appears to have temporarily retreated for some distance, readvanced over the whole of eastern Long island and the islands to the east, fold- ing and planing the older beds, but, except in western Long island, where the upper Manhasset was laid down, it deposited very-little of the ‘materials now to be seen above sealevel. It is at this time that the severe folding on Fishers island such as shown at the clay pit (figure 6) and at other points (figures 3,7, and 8) appears to huve been pro- duced. It is believed to be contemporaneous with the folding on Block island, at Gay head, etcetera. During the period commencing with the deposition of the Herod gravel and continuing through the Montauk stage and through the sub- sequent stage of deposition and folding a gradual subsidence appears to have been going on, until on western Long island, as indicated by the upper level of the marine glacial deposits, the land was 250 feet lower than at present. Fishers island, if not beneath the ice, must have been covered by the sea if a similar depression existed in that region. Following the retreat of the ice an uplift took place, the land gradu- ally rising from a position 250 feet lower than at present until it stood considerably higher. The uplift was not sudden, there being, on the contrary, several periods of halt, the most marked being at the present * Pleistocene geology of portions of Nassau county and the borough of Queens. New York State Mus., Bull. 48, pp, 618-670, PRE-WISCONSIN EVENTS 385 100 and 40 foot levels. At points on the coasts of Long island, Marthas Vineyard, Buzzards bay, and cape Cod there are indications of terraces at both these levels; but if ever present on Fishers island they were so altered by the Wisconsin ice invasion as to be practically unrecognizable. The only suggestion of such a terrace was near the south end of West harbor, where an imperfect 40-foot level was observed. - During this period not only was marine erosion at work, but streams were actively deepening the valleys. On Long island quite a complete drainage system with channels 40 to 80 or more feet deep was developed, being in marked contrast to the post-Wisconsin erosion, which has only cut a few insignificant gullies. In fact, the length of the erosion interval known as the vineyard interval, seems to have been very many, per- haps fifty, times as long as the post-Wisconsin period. In the central portion of the country the erosion between the Wiscon- sin and the Iowan stages of glaciation was very slight, indicating a period of rather limited length and one not in harmony with the great length of the Vineyard interval. It seems likely, therefore, that the latter in- terval covers more than the brief Peorian stage, representing rather the combined Sangamon, Iowan, and Peorian stages. The fact that the Iowan ice did not reach very far south in the Mississippi valley suggests that it may likewise have fallen short of reaching the Long Island region. If _ the Montauk stage is to be regarded as the equivalent of the Illinoian, there is certainly no evidence of the Iowan on the island. Wisconsin deposits—The Wisconsin drift, as already indicated in con- nection with the descriptions of several of the older deposits, is commonly very thin, being usually represented by a sheet only a few feet in thick- ness. The ice during its maximum advance completely covered the Fishers Island region, reaching as far as Block island, 10 to 15 miles farther to the south. Allowing a gradient of 40 feet to the mile of the ice surface, which is one frequently assumed, there would have been from 400 to 600 feet of ice over Fishers island. Notwithstanding this considerable thickness, the ice seems to have accomplished very little in the way of erosion. It is doubtless true that there was a further round- ing of the pre-Wisconsin hills, although the main sculpturing, if due at all to ice-action, was accomplished by the more powerful Montauk sheet. In fact, beyond a moderate rounding of some of the hills, the scooping out of a few basins in the soft sands, the shoving up of low barriers of drift across old valleys, and the deposition of a thin drift mantle, the Wisconsin ice appears to have accomplished but little. The absence of evidence of pronounced erosion and the slight thickness of the deposits both point to a relatively slow and weak movement of the ice, and it is probable that the period of occupation was not a long one. ~— te) ‘\# ee uh i ey : el” ~~ + y “4 386 M. L. FULLER—GEOLOGY OF FISHERS ISLAND The maximum advance occurred in the early part of the stage, the ice reaching as far south as Block island, but later it drew back to Fishers island, or perhaps beyond, where it halted during the building of the morainal deposits. The recession was a relatively rapid one, the inner moraine and its attendant outwash plain being built up, as shown in eastern Long island, before the ice-blocks left by the retreat had time to melt. . The deposits of the Wisconsin stage on Fishers island consist mainly of till, which varies somewhat in character, according to the derivation of the material. When from the island it is buff and sandy, with few boulders, representing, in fact, simply the reworked parts of the under- lying materials, especially the Herod gravels. At other points the till is composed largely of granitic materials brought from the mainland. In such till clay is relatively abundant, and, together with the granitic frag- ments, gives the till its grayish color. Boulders are numerous and often of considerable size. The best development of till is in the western part of the island, especially west of West harbor. In this region the low bluffs at many points show a till very full of boulders, while the kettle-pitted uplands and the morainic knobs point to more than the ordinary amounts of Wisconsin drift. In the eastern part of the island the Wisconsin drift is thinner, the sands and gravels of the Herod formation being but thinly covered. Sometimes the Wisconsin drift is limited to a few scattered boulders. No stratified materials of this stage are recognized, and, while a few may have been overlooked, it is certain there is no extensive devel- opment of such deposits on the island. Recent history.—There is little record on the island of any post-Wis- consin events. There has been nostream erosion, and the cutting of the bluffs has probably not been extensive. Some small spits have been built, and there has been a slight accumulation of marsh and swamp deposits in spots; but, in the main, there has been little change in the island since the ice left it, other than a possible submergence of a few feet. Even of this, however, no local evidence was found, although from studies on Long island and at points to the east of Fishers island it appears that a subsidence of perhaps 10 feet has occurred since the retreat of the ice. CORRELATIONS RELATION OF GARDINER CLAY, JACOB SAND, AND HEROD GRAVEL TO “SANKATY BEDS” (WOODWORTH) OF MARTHAS VINEYARD AND NANTUCKET The term “Sankaty beds” was proposed by J. B. Woodworth f * The diversity of the glacial period on Long island. Jour. of Geol., vol. xi, 1903, pp. 763-776. + Glacial brick clays of Rhode Island and southeastern Massachusetts. U.S. Geol. Survey, Sev- enteenth Ann, Rept., pt. i, p. 977. : CORRELATION 387 for a series of marine sands [and gravels] with occasional minor clayey layers, locally carrying a Pleistocene fauna, which are involved in the folding which affected all the New England islands. The name is from Sankaty head, Nantucket, at which point the fossils were first observed. No true clay was exposed at the type locality except a thin fossiliferous bed only a few feet in thickness, although a series of brownish clayey sands incorrectly designated as clay were exposed to a depth of 20 feet at the base of the cliff some 50 years ago.* The remaining por- tions of the bluff were made up mainly of sands, gravels, and semi-till. Later Mr A. C. Veatch applied the same namef on the basis of its similar fauna to the blue, gray, or red Pleistocene clays of Long and Gardiner islands, in the former of which they reach, according to the evidence of wells, a maximum thickness of about 100 feet. The writer was originally inclined to the use of this term,{ but subsequent field work, during which the shores of the greater part of Long, Gardiner, Plum, Fishers, and Block islands were traversed and the bluff sections of Marthas Vineyard, Nantucket, and cape Cod visited, brought out the fact that the true clays in each of the localities mentioned are strati- graphically below the lowest of the deposits at Sankaty head, and hence should not be included in the term. Moreover, the Sankaty beds of Woodworth are separable into two distinct formations, the Jacob sands and the Herod gravels (see pages 378 and 381), each of which has been traced by the writer from western Long island to Nantucket. The true clays, to which the term “ Gardiner” is given, and the Jacob sands are interglacial deposits, while the Herod gravels are in part, at least, of glacial origin. ‘The term “ Sankaty,” as used by Woodworth, therefore, includes materials not only of unlike character, but also of diverse origin, while as used by Veatch it includes only a part of the interglacial series. If the term is to be retained, as it possibly will be (because of the well known fauna of Sankaty head), it should, in the writer's opinion, be limited to a single class of deposits, namely, the interglacial deposits here separately described under the headings of Gardiner clay and Jacob sands. The glacially derived Herod gravels should be excluded. © CORRELATION WITH THE DRIFTS OF PENNSYLAANIA, MISSISSIPPI VALLEY, AND CANADA The geologic successions on Fishers island seem to be essentially sim- ilar to those existing on Long island, which may be considered as the *E. Desor and E. C. Cabot: On the Tertiary and more recent deposits in the island of Nantucket. Geol. Soc. Quart. Jour., vol. v, 1849, pp. 340-344. 7 The diversity of the glacial period on Long island. Jour, of Geol., vol. xi, 1903, pp. 762-776. } Probable pre-Kansan and Iowan deposits of Long island, New York. Am. Geol., vol. 32, p. 311. 388 M. L. FULLER—GEOLOGY OF FISHERS ISLAND type locality of Pleistocene Coastal Plain formations in the northeastern United States. Ifthe interpretations outlined in this paper are correct, four glacial and three interglacial stages have been represented in the upbuilding of both islands. The question of whether the succession of ice invasions in this region was the same as in the central portion of the country and in Canada is difficult to determine. The lobate character of the margins of the earlier ice-sheets in the Mississippi valley is such as to suggest that they might be, to a certain extent, confined to that region. On the other hand, the Wisconsin invasion is known to have reached from the Rocky mountains to the Atlantic coast, reaching nearly as far south in this region as at points farther west. The most important evidences pointing to the similarity of succession in the Atlantic and Mississippi regions are the relative weathering of the drift, the erosion features of the interglacial stages, and the evidence of elevation or depression of the land in the two regions. For instance, the old extra morainic drift of Pennsylvania (the oldest drift on the main- land east of the Mississippi valley) was deposited when the land was low - and the streams relatively sluggish, and on Long island the earliest inva- sion was associated with a marked eae ession, the land standing 300 feet below its present level. In each case the period of low level was suc- ceeded by a pronounced uplift, which on Long island resulted in the nearly complete removal of the Mannetto gravels and in Pennsylvania in the very considerable deepening of the Allegheny, Monongahela, and other river valleys. This strongly suggests that the sequence of these early events was not only the same in each case, but that they were contemporaneous. As to the absolute age e of the Mannetto gravel of Long island and the older drift of Pennsylvania, there is little direct evidence except the state of weathering of the pebbles, which is materially greater than in the Kansan deposits of the Mississippi valley; so much so, in fact, that it seems as if an interval of time, perhaps as great as all subsequent Pleis- tocene time, must have intervened between the deposition of the Man- netto and the Jameco. Itis believed that these two drifts, together with the intervening interglacial stage, can be correlated with a fair degree of certainty with the Albertan, Aftonian, and Kansan stages of Canada or the Mississippi valley. The Jameco gravel is followed conformably by the Gardiner clay and the Jacob sand, both of which are interglacial deposits, and would nat- urally be referred to the Yarmouth interglacial stage. The Jacob sand is normally followed, in turn, on Fishers island by the Herod gravel and Montauk drift,and on Long island by still other gravels. Between each CORRELATION 389 of these there are local unconformities, such as might be reproduced by readvances of the ice-sheet with which the deposition was associated. Throughout the whole of the series, however, there are no evidences of interglacial soils or of periods of stream erosion ; and while it is certain that the ice invaded the region at least twice, it is believed that the advances simply mark substages of the same invasion, which is regarded as Illinoian. CORRELATION TABLE Fishers island. Glacial. Interglacial. Deposition of Wiscon- sin drift. Erosion. Erosion. Erosion. Deposition of Mon- tauk drift and Herod gravel. Deposition of Jacob sand and Gardi- ner clay. Deposition of Jameco gravel. Erosion. Deposition of Man- netto gravel. Mississippi valley or Canada. Glacial. Interglacial. Deposition of Wis- consin drift. Peorian soils, etcetera. Deposition of lowan drift. Sangamon soils, etcetera. Deposition of Illi- noian drift. Yarmouth soils, etcetera. Deposition of Kan- san drift. Aftonian soils, etcetera. Deposition of Al- bertan drift. es ‘ : - « ; ’ f 390 M. L. FULLER—GEOLOGY OF FISHERS ISLAND Following the deposition of this series of deposits, with its somewhat complicated history, there was a long period of erosion, during which all of the older materials were extensively eroded, both by marine and stream erosion, before the Wisconsin ice invaded the region. From the extent of this erosion it seems probable that the interval was an exceed- ingly long one, being many times as long as the post-Wisconsin time. This is what would naturally be expected if the period included the Sangamon, Iowan, and Peorian stages of the Mississippi valley, and the correlation with these stages fits in well with the known shortness of the Peorian period in the Mississippi valley, which can not form, judging from the amount of erosion, but a small part of the interval between the Wisconsin and the next preceding invasion in the Fishers Island — region. DISTRIBUTION OF PRE-WISCONSIN FORMATIONS REPRESENTED ON FISHERS ‘ Long island. ‘Gardiner island. Plum island. Fishers island. Long island. Gardiner island. Plum island. Fishers island. Long island. Gardiner island. Plum island. Long island. Gardiner island. Plum island. Fishers island. Long island. Gardiner island. Fishers island ) (below Plum island sea). ISLAND MONTAUK DRIFT Block island. Elizabeth islands. Nantucket. Barnstable. HEROD GRAVEL Block island. Marthas Vineyard. Nantucket. JACOB SAND Fishers island. Block island. Marthas Vineyard. GARDINER CLAY Block island. Elizabeth islands. Marthas Vineyard. Nantucket (below sea). JAMECO Block island. Marthas Vineyard. Barnstable (in well). Cape Cod. | Indian hill (Plymouth). Scituate. | Boston and vicinity. Truro. Scituate. Boston and vicinity. Nantucket. Cape Cod. — Boston and vicinity (?). Truro. Barnstable. Indian hill (Plymouth). Boston harbor (?). Indian hill (Plymouth). Truro. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 391-410, PLS. 67-70 JUNE 24, 1905 MESOZOIC SECTION ON COOK INLET AND ALASKA PENINSULA* BY T. W. STANTON AND G. C. MARTIN (Presented before the Society December 31, 1904) CONTENTS Page EMRE ECAD Git hoc uel ack e uc vias ak owiehe ee uous ae. COI fe as hte OBE General geology of the region........ aA anerionts See aaah, AONE ane wel 392 EET AMIS ATI PATITIAGI i) Os wale OOS on ee alee eee tlo ed ble lewctilaae 393 NNRINA RRR tyre he) cathe uta eharge ee wiv Gi a ayar Me Gie Siig He sind ie p's ee wD 393 Lower Jurassic. ...... OE, ROMER GARE ne Cae CE eR LET Pa eee OLE mn Ieee 396 ee arasstc— PnOChKIn, fOTMALION. « é..0'.05.06 e/ocas'- + cece taweesoence 397 peer arasmnc—- Naknek fOrmation wo 060 ccc cece cate a seta sew eeeces 402 Lower Cretaceous..... ........ Se Mee ere dl nattac a Wee ene Ga om 407 DUMMEPTELECOOMS <6. 05s cee cece vances Bees wens ois Bere ee on wee ote 408 Bésamé .:....... Bee aks Ss OTE Pe pry “88 eat ee TSI sak ata atetp 409 INTRODUCTION The localities discussed in this paper are situated in the southwestern part of Alaska, on the shores of Cook inlet and on the Alaska peninsula. Most of them are on the strip of coast forming the southern half of the shore of Cook inlet and the eastern half of the south shore of the Alaska peninsula. The facts here presented include some of the results of a trip made by the authors in small boats during the summer of 1904 from Snug harbor, on Cook inlet, to Cold bay, on Shelikof strait. Isolated localities both to the eastward and to the westward were also visited. There are various references to the Mesozoic rocks and fossils of parts of this region. Many of the early references t have been summarized * Published by permission of the Director of the U. S. Geological Survey. 7C. Grewingk: Beitrag zur Kenntniss der orographischen und geognostischen Beschatfenheit der Nord-West Kuste Amerikas mit den anliegenden Inseln. Verhandl. Russ. Kais. Mineralog. Ges, zu St Petersburg, Jahrg. 1848 und °49 (1850), pp. 121, 344-347. E. von Eichwald: Geognostisch-Paleontologische Bemerkungen uber die Halbinsel Mangi- schlak und die Aleutischen Inseln. St. Petersburg (1871), pp. 88-200. J. Marcou: Explication d’une seconde edition de la carte geologique de laterre, Zurich (1875), pp. 138-140. LIII[—Buut. Geox. Soc. Am., Vo. 16, 1904 (391) 392 STANTON AND MARTIN—MESOZOIC SECTION ON COOK INLET by Dall,* who made some valuable original observations on the stratig- raphy and -paleontology at Tuxedni harbor (Snug harbor), Cold bay, Kialagvik bay, and Wood island, near Kodiak. The fossils collected by Doctor Dall were examined by Professor Alpheus Hyatt, who contrib-. uted a brief statement concerning them.} — Professor J. F. Pompeckj{ has examined and described some of the Jurassic collections obtained by the Russians more than half a century ~ ago and has briefly discussed the age and correlation of the beds. Dr K. O. Ulrich § has described the fossils of the Yakutat series, which he referred to the Lias. This formation occurs on Kodiak and adjacent islands and on Yakutat bay, but has not been seen on the mainland of the Alaska peninsula. Mr J. E. Spurr crossed the Alaska peninsula at Naknek lake and Katmai in 1898, and has described || the Mesozoic sec- tion exposed in that region. The junior author visited a few localities on these shores in the summer of 1903, and has described] part of the Mesozoic stratigraphy, but no connected account of the entire Mesozoic section giving any adequate idea of the thickness and stratigraphic relations of the different forma- tions has been previously published. GENERAL GEOLOGY OF THE REGION The most prominent topographic features of this region are three dis- tinct mountain ranges, with two intervening valley regions. These ex- tend roughly parallelin an approximately northeast-southwest direction. The Chigmit mountains are at the northwest and extend from mount Redoubt, near the northern end of Cook inlet, to the southeast corner of Iliamna lake. The Aleutian mountains occupy the axis of the Aleutian Paul Fischer: Sur quelques fossiles de ]’Alaska, in voy. a la céte N. W. de l’Am. 1870-72, par Alphonse Pinart. Paris, 1875, pp. 33-36, pl. A. Charles A. White: On a small collection of Mesozoic fossils obtained in Alaska by Mr W. H. Dall. Bull. U.S. Geol. Survey, no. 4, 1884, pp. 10-16, pl. vi. M. Neumayr: Die geographische Verbreitung der Jura-formation. Denkschr. der Wiener Akademie, Bd. 50 (1885), pp. 93, 94. C. A. White: Mesozoic mollusea of the southern coast of the Alaskan peninsula, Bull. U.S. Geol. Survey, no. 51 (1889), pp. 64-70. * Report on the coal and lignite of Alaska. Seventeenth Ann. Rept. U.S. Geol. Survey, pt. i, pp. 865-869. + Report on the Mesozoic fossils: Report on the coal and lignite of Alaska. Appendix iii, Sey- enteenth Ann. Rept. U.S. Geol. Survey, pp. 907-908. { Jura-Fossilien aus Alaska. Verhandl. Kais. Russ. Min. Gesellschaft., 2d ser., vol. xx xviii, pp. 239-282, pls. v-vil. St. Petersburg, 1900. ? Harriman Alaskan Expedition, vol. iv, Geology, pp. 125-146. || Reconnaissance in southwest Alaska. Twentieth Ann. Rept. U.S. Geol. Survey, pt. vii. q Petroleum fields of Alaska and the Bering River coal fields. Bull. 225, U. 8. Geol. Survey, pp. 376-381. The petroleum fields of the Pacific coast of Alaska, with an account of the Bering River coal deposits. Bull. 250, U. S. Geol. Survey. Werk oes Sos al GENERAL GEOLOGY 393 islands and the Alaska peninsula, having their eastern termination on cape Douglas, at the mouth of Cook inlet. The Kenai mountains occupy the southeastern half of the peninsula of that name, and have their west- ern termination at cape Elizabeth, with a possible detached extension in the lower summits of Afognak and Kodiak islands. The Alaska peninsula contains a coarse crystalline core of granite or of similar rocks, flanked on the eastern side by Mesozoic sediments and on the western side by late Tertiary or post-Tertiary beds. The Meso- zoic beds are overlain in places by early Tertiary formations. Both the Mesozoic and the Tertiary beds are cut by andesite and basalt. The intrusion and volcanic outflow has continued from late Jurassic time until the present, the region containing several active volcanoes. The structure of the region is varied. The west shore of Cook inlet has its general position outlined by a number of great overthrusts, by which the Triassic rocks have been brought in contact with the Upper Jurassic. The Alaska peninsula is a region of open folding, with the folds cross-cut by an irregular series of faults. The general distribution of the Mesozoic rocks throughout the region discussed is shown in the sketch map, figure 1, and a somewhat more detailed map of part of the coast of Cook inlet is given in figure 2. Mesozoic FoRMATIONS AND FAUNAS UPPER TRIASSIC Upper Triassic rocks have been seen on the north shore of Bear cove and on Bear bay, both on the west shore of Cook inlet, and on the Alaska peninsula at the entrance to Cold bay and extending several miles east- ward. They are also probably present at numerous localities on the south shore of Kachemak bay, as at Halibut cove and Seldovia, where there is a great development of thin bedded and contorted cherts, with some silicious limestone and igneous rocks. This series is tentatively correlated with the fossiliferous Trias on the west shore of Cook inlet on account of lithologic and structural resemblances and because of its association with the Lower Jurassic. The Triassic rocks of Cold bay, Bear cove, and Bear bay, whose age has been definitely determined by fossils, consist of thin bedded chert, limestone, and shale of varied colors. The chert and limestone are usually dark-black, green, or dark-red when fresh, but weather to lighter shades. No measurement of the thickness has been made, but it is esti- mated to be at least 2,000 feet in the exposures seen by us, in which the base was always cut off by intruded igneous rocks. These rocks are always closely folded and are frequently crumpled. They are usually cut by numerous dikes of diverse character and com- 8394 STANTON AND MARTIN—MESOZOIC SECTION ON COOK INLET position, varying from granite to andesite and basalt. The more acid dikes are apparently characteristic of the Triassic rocks, and were in- truded soon after the folding which must have closely followed Triassic time, for they do not cut the younger rocks. A characteristic exposure of these folded beds is shown in plate 67, figure 1, and their relation to the Jurassic is indicated by the comparatively undisturbed condition of LEGEND = iccmeas JURASSIC I es (I) MESOZOIC REPORTED Fiaure 1.—Map of Cook Inlet and the Alaska Peninsula. Showing the distribution of the Mesozoic rocks. the Upper Jurassic sandstones less than one-fourth mile distant, as seen in plate 68, figure 2. _ The Triassic fauna of the region as now known is almost limited to the single species Pseudomonotis subcircularis Gabb, which is very abundant in certain layers of shale and limestone at Bear cove and Cold bay. Specimens from the latter locality were described and figured by Fischer as Monotis salinaria, which it resembles in its general features, but a com- parison of a large series of specimens from Alaska with a similar series BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 67 Figure 1.—FoLDED AND FAULTED TRIAssic LIMESTONE AND Currrs, BEAR Bay, Cook INLET Figure 2.—Lower Jurassic TUFFS AND CHERTY LIMESTONE, Port GRAHAM TRIASSIC AND LOWER JURASSIC FORMATIONS IN ALASKA i Per bet “YuSV IV ‘SNOILLVWY LLIN] M009 ‘AVG Uvag ‘SHNOLSGNVG MINUVN—'Z FUNDY GAIN, MOOD ‘AVG TIO dO AGIG ISVA NO SNOILVWHOY MANMVN GNV NIMHOONGY—'[ FUNDY 89 “Id ‘vO6L ‘9L “1OA "WY “00S "1035 "11nd MAP OF PART OF COOK INLET 395 LEGEND Upper Jurassic Naknek formation "Middle Jurassic = Enochkin formation Triassic 10 15 miles Ficure 2.—Map of Part of the West Coast of Cook Inlet. Showing the distribution of the Mesozoic formations. 396 STANTON AND MARTIN—MESOZOIU’ SECTION ON COOK INLET of Gabb’s California species shows that they are not separable. The species belongs to the group of Psewdomonotis ochotica, which is character- istic of the Upper Triassic of Siberia and the boreal regions generally. In California Pseudomonotis subcircularis is confined to the Swearinger _ slates, which form the uppermost Triassic formation of that region. Sim- ilar beds with the same fossil occur in Vancouver and Queen Charlotte islands and on the mainland of British Columbia, and they cover con- siderable areas in the Copper River region of Alaska. At Bear bay the much folded Triassic limestone yielded imperfect specimens of a Halobia, and at Cold bay a few imperfect Ammonites not generically determined were obtained in beds overlying the Pseudo- monotis layers, but possibly still within the Triassic. One other area of probable Triassic has been reported by Hyatt * from cape Thompson, northwest Alaska, on account of the supposed occur- rence of Halobia or Daonella there, but subsequent collections from that region have proved the rocks to be Paleozoic. : LOWER JURASSIC In the neighborhood of Seldovia there is a considerable thickness of sedimentary and associated igneous rocks that are believed to be Lower Jurassic. They lie west of the cherts previously mentioned as of prob- able Triassic age, and are not so much disturbed and metamorphosed as the cherts, from which they are doubtless separated by an unconformity, though the dips, which vary in both direction and amount, average per- haps 40 or 50 degrees and in some exposures are as much as 70 degrees. The rocks are apparently composed almost exclusively of fragmental ‘igneous material, and should be classified as tuffs rather than sandstones and conglomerates. The fine grained beds are dark greenish gray when fresh, but weather to lighter shades, while some of the coarser beds, made up of more or less angular fragments, are red, greenish, and variegated. This series was seen on the west shore of Seldovia bay at the entrance to the harbor and for 3 miles westward along the shore of Cook inlet; also on the east shore of port Graham south of the coal-bearing Kenai exposures. At the latter locality the section includes a conspicuous bed of light colored cherty limestone (see plate 67, figure 2). It is probable that the ‘‘ dense Neocomian limestone” near English bay, from which Kichwald cited Arcomya crassissima and a Janira, belongs in this series. Fossils are not very abundant nor well preserved, but those obtained on Seldovia bay and along the coast for 2 miles to the westward indicate a horizon low in the Jurassic. They include Pentacrinus, a Trigonia of the section Glabree, Cardinia (?), Myophoria (?), Gryphea, three or more *Seventeenth Ann. Rept. U. S. Geol. Survey, p. 907. a ee ee ei ee ee, a ee a Piiiew LOWER JURASSIC 397 species of Pecten, Pinna, and two or three species of Ammonites repre- sented by fragmentary specimens, one of which seems to be an Arietites or a closely related form. The Myophoria (?) is a small shell with four strong radiating ribs, and if the genus is correctly determined it would suggest an age as old as the Rhetic, but the other fossils give stronger evidence of Jurassic age. None of the fossils in this little assemblage is known elsewhere in Alaska, and we are unable to make any definite cor- relation with other American formations, although rocks of the same age doubtless occur in the California section, It is probable that Lower Jurassic rocks of a different character occur on the Alaska peninsula. At Cold bay there are nearly 4,000 feet of shales and sandstones lying between the determined Triassic rocks and the Cadoceras-bearing Enochkin formation, but their exact position in the Jurassic or Triassic was not determined by paleontologic evidence. In the Cold Bay section the unconformity between the Triassic and the - Jurassic is not so obvious as at some other localities, and the boundary between the two systems was not definitely fixed by our very limited examination, the relations being obscured by several important faults and by the absence of characteristic fossils in the critical part of the section. It is more than probable, however, that there is an unconform- ity above the Triassic. The Ammonites from Kialagvik or Wide bay _ described by White are referred by Pompeckj to the Upper Lias, and Hyatt referred other collections from the same bay to the base of the Lower Oolite, stating that they are related to faunas that are called Upper Lias by German geologists. These are in unaltered sandstones, and whatever may be the final decision as to their age, they are evidently older than any part of the Enochkin formation. Hyatt also tentatively referred certain specimens of Lytoceras and Phylloceras from Kamishak bay to the Upper Lias, but we now know that they really come from the Upper Jurassic Aucella beds of the Naknek formation. The Yakutat formation, on Kodiak island, across Shelikof strait from the Alaska peninsula, which was referred to the Lower Jurassic by Ulrich, differs completely in both lithologic and paleontologic features from all the Jurassic formations known in the region, and as the fossils it has yielded are all new forms of doubtful affinities, its exact age is still uncertain. MIDDLE JURASSIC—ENOCHKIN FORMATION The Enochkin formation is typically exposed on the east shore of Enochkin bay, from which it was named.* It extends from here in a * The petroleum fields of the Pacific coast of Alaska, with an account of the Bering River coal deposits. Bull. 250, U. S. Geol. Survey, 1904, pp. 37-55. 398 STANTON AND MARTIN—MESOZOIC SECTION ON COOK INLET continuous belt west of and parallel to the shore of Cook inlet at least as far northward as Snug harbor. It also covers a large area on the Alaska peninsula in the vicinity of Cold bay. The occurrence of Cado- ceras wossnessenski reported “ near Katmai” by Grewingk indicates an- other possible locality for the rocks of this formation in the Cold Bay area. The following section shows the character of the formation at the type locality, essentially as described in the paper just cited: | Compiled detatled section of the Enochkin formation . Zone D: Feet Drab shale with numerous bands of limestone concretions filled with well preserved specimens of various ammonoids, mostly Cadoceras and Belemnites, with occasional sticks of fossilized wood............ 146 Concealed by talus (all shale as above), thickness computed........... 77 Shales as above with same concretions and WoOd ..........ceeseeceees 196 Iimestone...c. so1ee eee ele 6 ne worm nieialg: #66 0.0m ara eosieheie kee 1 Shales'as above... ses cece sce es vec seus sag eas see tee 363 Shales (partly concealed by talus), paicieness comiputed .. : 2. eke eeeee 300 Shales as above with Cadoceras doroschini and a few other fossils....... 200 Concealed oo. eck bipec sre wie Va 0 eine sie eae a els ane ae et er 25 Zone C: : Sandy shales with many Belemnites and other fossils................- 50 Gom@ealed epics v's cers etew' erat ms 3 eran’. + se Rue eedate oiciale ace eae ey cnrtrn. Meee) Soft SBS. en aleiabre aia bie seed 44 ele wuetene ea 6 Ue aie obo 3 eae 20 Dark drab shale with scattered fossils... . 2... .0. 2+. «32 0 ee 33 Hard calcareous shale full of fossils, principally Inoceramus, Pleuromya, and ‘other pelecypods.0 i. sie. es sje ule e eee | le ole he See 2 Black sand... cic 5 acc eo ss plage nels tele 28 isin we ot = aries ia ale ee 1 Darkishale . 2 cc0e cs e's ede 6p -5e> eos eine pi loUe Bik Glee ate ee vee) ciel ee 5 Bllavekk Sain ..265 5 id lav sie 2/2 pri eee aay Sie es Sb ORS he eda wig) leer re Dark shale with many fens wns ton pee se. 9d oe 8 oO > 2 12 Reddish limestone. 0200/06 ec) eve so bot ee 6 os be tle, s0 ge ek $ Dark shale with many fossils: 405 0.5 es eis ee eee | 14 Dark shale with scattered fossils. ..-.42.%. <2 00020. tess pee Dien ee 62 Dark soft sandstone with streaks of conglomerate ..............+-+00- 10 Comeeale de c22siee: Wie sce BOO oie Ge oy ee eee . jo an habe ane Zone B: Shale with several fossil bands containing Trigonia doroschini, etcetera. 50 COmeGa led. eisai im ape, 8a ceahelepe. in Ge kom Gosohan ciate» QE ie ee aloita aye ate ee ee Zone A: Shale with one or more bands packed with fossils—Trigonia doroschini, T. devexa, etcetera—each fossil bed 10 to 25 inches thick ............ 30 Ooncealed |... sic aves. oewin weed waecqeeaan eee aoe us es a Shales: cb. eee es bai An ee llega I Be ne 12 Coarse conglomerate 2. f.%). os. sa ssc aes alee pa a cetp alee ate a 20 UmcComformni bys ie ones see Andesite 22 0.62 oi o's cago Bm SERS es etertn bese Sonia = 0 lowe seve aie Concealed (for 50 feet at 20 degrees dip)..2......-..5......4 2. 2s INTVGESILE 0 oc Sica wl 14 a ols wie ww soc apevetie rm Oey eialade ie ie a tee sake ete ene Concealed (for 130 feet at 20 degrees dip)................. Pe a IN GESILG sore cn Tianiean wa ha @ Tyson baie wie Belo Misa & eR oe sikinle = See A PPIOMMEV ALE: 5'—. > 5 sais es one eo as wns a lainiy ous adae ese apes eae Arkose with shaly bands ..2<...065 06 cieie cs once ceewe nels se ae) Andesite with agglomerate at base. . 2... 0. c.5...0520+050005- pd bln db ce) wa casvesveuewes 150 Enochkin shales (see section on page 400) Pere ose hse yi weve bens unos ecccacuse cave del 5,139 A06 STANTON AND MARTIN—MESOZOIC SECTION ON COOK INLET This section is given in detail on account of the very great apparent thickness of the rocks referred to the Naknek formation. It is possible that a part of this thickness may be due to repetition of beds by faulting, though we were unable to detect any repetition and the faults seen did not seem to be important. Fossils are not abundant, but the character- istic Aucella resembling A. pallast was found sparingly both near the top and near the bottom of the formation. Section of Naknek formation on east shore of Oil bay, Alaska Naknek formation: Feet Arkose, andesite, sandstone, conglomerate, and shale..............-<.- 2,000 Sandy shale with Aucella near base. ©... ...00...<2 6520 ues oo eee - eine ch Shalewith fossns: 2.3 5 sec aes ve bee owed bs Cee ee ee + 1b Cee Coarse sandstone. «21.6... ssi lens cals 85 necks ciel 7 bee per 3 Shale with Cardioceras, Astarte, etcetera.........cccceecccctecccee coe 165 Concealed... . oi .cadadaciemcadaad ls Sede tg aeteas oe eae ee a Sandstone and sandy shale with Lytoceras, Phylloceras, and plant im-— pressions..... nian a 2 ee filiwim/a weja-way acpiare a's) dle 4s sw leita 60 fo "gays 10) ey ete 310 Naknek formation (?): Agglomerate with an abundance of small pebbles, one-twelfth to one twenty-fifth of an inch in length, and with | numerous poorly preserved plant IMPTressiONs..5<.0 as Ws eed ein eae ees d se sns oot pe eee 7 Sandy shale and sandstone.. ...........-. aa tend. we ean ee ee ee 85 Agglomerate with pebbles, as above. ......-. ccc ececeeavsees die aad ie 3 SECs cs wines ear eer choke meek nee I GA Oe ae cues « SGA aE eee aoe be 1 Fine agglomerate of same Hobbies as above... .fssv. .viweek dena 7 Fine agglomerate of same pebbles, as above, but interbedded with shale. 14 Olive shale with an abundance of small pebbles, as above...........++ 30 Shale (Enochkin formation) 2s. 6s. sed io. tisiee wea 6 on ee eee tire WON MDa ie eeu ciate tee sislastlascherscichs docs sc es: 2ee =n 3,645 These exposures are shown in plate 68, figure 1. Section of Naknek formation on east shore of Enochkin bay Feet Sandstone, arkose, shale, andesite flows, and agglomerate.........+ -..+- 270+ . Dark sandy shale, with Aucella in upper part... ...c0..ccee cee cee eeeeens 583 Coarse agglomerate iis. seis. sce ss cee coos esd sasa ner ceessd cue eerie 290 Shale (Enochkin formation): ..65..0c00ssstucseeccvewaus e's 2 wees eee Ota ass fG c,d w bS sine b win ee bk ee Seems Ryo a a = als ener a ai a 1,148 The coarse agglomerate forming the lower 290 feet of the last section is regarded asa local lens. It was also seen in the same position on Chisik island. The fine pebble agglomerate extending through a local thickness of 147 feet in the Oil Bay section is regarded as probably the representative of the coarser beds seen elsewhere. This section forms the upper two-thirds of the cliffs shown in plate 69, figure 1. BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 70 Figure 1.—UNcONFORMITY 1N THE NAKNEK Formation, Dovetas River, Cook INLEY Figure 2.—ConGLoMERATE Ciirrs, NAKNEK ForMATION, CoLtp Bay, ALASKA EXPOSURES OF NAKNEK FORMATION, ALASKA FAUNA OF NAKNEK FORMATION 407 In Kamishak bay the exposures of the Naknek formation are usually very fossiliferous sandstones, with low dip, and representing only a small part of the whole formation. One such exposure on Rocky bay is shown in plate 69, figure 2. On Douglas river, which empties into Kamishak bay west of Shaw island, there is an interesting local unconformity within the Naknek formation, showing an unevenly eroded surface, which was traced in the cliffs for abouta quarter of a mile. The photograph (plate 70, figure 1) shows a detail of this exposure. The fact that the same fauna is found both above and below this unconformity is evidence that the erosion interval was geologically brief, and it probably did not affect a wide area. * The fauna of the Naknek formation is especially characterized by the presence of Aucella belonging to species very closely related if not iden- tical with A. pallasi and A. bronni of the Russian Volga beds. These fossils are at some localities very abundant, completely filling thick beds. At other places they are so rare that they may be easily over- looked, but a careful search will find them in almost every section. As- sociated with the Aucella there are usually two or threespecies of Belem- nites, frequently a large Lytoceras and a Phylloceras, and occasionally a few gastropods, Trigonia, and other pelecypods. At Oil bay the lower part of the formation yielded two species of Cardioceras related to C. al- ternans and C. cordatus which aid greatly in making more definite corre- lations with both American and European horizons. It is clear that the Naknek formation is of about the same age as the Mariposa beds of Cal- ifornia with Aucella erringtoni and Cardioceras cf. alternans,* and it also includes the horizon of the marine Jurassic with Cardioceras cordiforme in the Black hills, where, however, the Aucella element is lacking from the fauna, and probably only the horizon of the basal portion of the Naknek is represented. A similar fauna occurs in Russia in the Volgian beds, and it is widespread in the boreal region, occurring on Spitzbergen, Nova Zembla, and elsewhere. In Alaska, Aucella of the same type occur in the Kennicott formation of the Copper River region and at many places on the Alaska peninsula as far west as Herendeen bay. © LOWER CRETACEOUS The presence of the Russian type of Lower Cretaceous on the Alaska peninsula is suggested by the occurrence of Aucella related to A. crasst- collis at port Méller and Herendeen bay, where Jurassic types of Aucella also occur in other beds, but the details of the stratigraphy are unknown. *Two names, C. whitneyi Smith and C. dubiwm Hyatt, have been proposed for probably the same species in this formation. LV—Butt. Grou. Soc. Am., Vou. 16, 1904 408 . STANTON AND MARTIN—MESOZOIC SECTION ON COOK INLET These beds, with Aucella cf. crassicollis, comparable with the Upper Knox- | ville beds of California, are widespread in Alaska, though we found no indication of them at any localities studied by us. The species was ob- tained by Mendenhall on Bubb creek, a branch of the Taxlina, in the Matanuska series; by Spurr in the Oklune series on the Kanektok near the mouth of the Kuskokwim; by Schrader in the Koyukuk and Anak- tuvuk series in northern Alaska, and by Wright on Admiralty island, southeast Alaska, where Aucella piochi, another Lower Cretaceous species, was found at a neighboring locality. It also occurs in the Mission Creek series on the Yukon. The geographic distribution of these Cre- taceous Aucella beds is thus seen to differ radically from the distri- bution of the Naknek formation with its Jurassic Aucella, and this is regarded as evidence of a probable unconformity between them. UPPER CRETACEOUS On Chignik bay Upper Cretaceous rocks, closely associated with plant- bearing Kenai beds, occur on the lagoon 1 to 2 miles northeast of the Alaska Packers Association cannery and on Whalers creek, 5 miles west of the same place, where they apparently include a workable coal bed. The exposures at the first locality consist of several hundred feet of shales and sandstones with some thick beds of coarse conglomerate in the middle portion. The beds beneath the conglomerate contain great numbers of fossil plants, a collection of which yielded the following Upper Cretaceous species identified by Doctor Knowlton: Osmunda arctica. Zamites sp. Sequoia reichenbachi. Myrica sp. Sequoia rigida. — Querous johnstrupt. Taxodium sp. ‘ Quercus Nn. sp. Torreya brevifolia. Ziziphus sp. Anomozamites schmidti. The shales above the conglomerates yielded Inoceramus fragments, Pecten, Nucula (Acila) cf. truncata, Corbula, Thracia, Dentalium, Cinulia, and some undetermined ammonoid fragments. At the locality on Whalers creek Mr R. W. Stone obtained an Inoce- ramus, related to I. digitatus,a Trigonia of the type of T. leana, and Anomia. These fossils indicate correlation with a horizon in the Chico as de- veloped in California and Vancouver island, which includes practically all of the Upper Cretaceous, but the beds at Chignik are probably not older than basal Senonian. Beds of possibly the same age occur on the north shore of the small hay north of Aievak or Douglas village, where there is a series of shales UPPER CRETACEOUS 409 and sandstones with an estimated thickness of 2,000 feet, from which fragmentary specimens of a large Inoceramus and a Desmoceras (?) were obtained. Pompeckj, who restudied the old collection of Wosnessenski from near Katmai, reports a specimen of Belemnitella labeled as coming from that place, and on that account infers the presence of Upper Cretaceous there, but this lacks confirmation. The Cretaceous beds at Chignik may be directly correlated with those on the Yukon near Nulato, and less certainly with those on the Anak- tuvuk, in northern Alaska, which are the only occurrences of Upper Cretaceous hitherto reported in the territory. It is probable that more detailed study will reveal considerable areas of Upper Cretaceous rocks on the Alaska peninsula, but their recogni- tion in rapid reconnaissance is made difficult by the fact that litholog- ically they resemble the Eocene Kenai beds of the same region, which, like the Upper Cretaceous, are coal-bearing and contain fossil plants of similar general types. The relations are further obscured by frequent faults, so that paleontologic evidence is necessary for the identification of the formations in each area; but fortunately both the plants and the animals of the Cretaceous are easily distinguished from those of the Kenai when sufficient collections are obtained. There are doubtless unconformities both below and above the Upper Cretaceous of this region, since all of the Lower Cretaceous is lacking at most localities, and at some places the Kenai rests directly on the Jurassic. RESUME The Mesozoic section of southwest Alaska includes representatives of the Upper Trias, Lower, Middle, and Upper Jurassic, Upper Cretaceous, and probably Lower Cretaceous. The Jurassic shows the greatest development, both stratigraphically and faunally, and is probably unequaled in these respects elsewhere on the American continent. The total thickness can not be much less than 10,000 feet, and the areas covered by the upper half of the Jurassic are large. The faunal type is essentially Russian—that is, boreal, though it differs in the common occurrence of Phylloceras and Lytoceras at sev- eral horizons. The succession of the faunas from the Callovian to the top of the Jurassic is the same as in Russia, but the vertical thickness of beds through which each ranges appears to be very much greater in Alaska. The Cretaceous and Triassic rocks so far as now known occur only in 410 STANTON AND MARTIN—MESOZOIC SECTION ON COOK INLET small scattered areas, and their faunas may be directly correlated with those of formations in California and elsewhere on the Pacific coast. The general relations of the formations may be epitomized in the fol- lowing section: Tertiary—Kenai formation. Shales, sandstones, and conglomerates with several beds of coal. The entire formation non-marine and characterized by a large flora. Thickness, + 2,000 feet. Unconformity. Upper Cretaceous.—Lithologically similar to the Kenai, but including some marine shales and sandstones with an Upper Cretaceous fauna. Thickness, + 1,000 feet. Unconformity. Lower Cretaceous.—(Not seen within the area studied.) Shales and sand- stones with Aucella crassicollis. Unconformity ? Upper Jurassic—Naknek formation. Conglomerate, arkose, sandstone, and shale with interstratified andesite flows. Thickness, about 5,000 feet. Middle Jurassic—Enochkin formation. Shales and sandstones with some conglomerate beds. Thickness, 1,500 to 2,500 feet. Unconformity ? (Possibly conformable on Lower Jurassic when that is present.) Lower Jurassic—Tuffs and sandstones. Thickness, + 1,000 feet. Unconformity. Upper Triassic—Thin bedded cherts, limestones, and shales usually much folded and contorted and with many intrusive masses. Thickness, + 2,000 feet. Base not seen. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 411-418 AUGUST 28, 1905 GEOLOGICAL BOOK-KEEPING BY J. F. KEMP (Read before the Society December 29, 1904) CONTENTS Page RUIN EE, WIR TAA Eg UGS Si oN tapi an pemiy hed din l-¥ ais Alotos Stu Mpa ® W PEA 411 ETI STACEE LIEGE o oo 2 ahs oh ce Sd ty Sia vid vide hme eed Jie Rew eT ee es 412 IR ETENEC EC IG. ha hy ate Se ope ohh eh cindex alae aie RVG BA 'Win 9 eee wi Ripe nae 416 eee om which the system is based... ....... 2... 6. sees etme sees ac cence 417 INTRODUCTION So long as a geologist is in the field for a short and continuous period and in a restricted and comparatively simple district, no more elaborate ~ and systematic system of notes is needed than is afforded by a map and a book. But when the district becomes complex and difficult, as in many metamorphic areas, and when the field work extends through sev- eral seasons, often with intervals of a year or two between, some method of indexing and compiling becomes absolutely necessary. The best of memories can not carry the endless details involved, and during office work time can not be spared for the discouraging search through a num- ber of note books. A system that will work almost automatically is as essential to the geologist as is double-entry book-keeping to a large com- mercial house, and it fills very much the same field. If, moreover, a teacher has to give field instruction to rather large parties in structural and areal work, it is almost as essential to have some uniform method in accordance with which all will record their observations and codrdinate them one with another. From experience of both these kinds the writer has been led to evolve the plan here set forth and has found that it answers satisfactorily and adds greatly to efficiency. Itis also quite evident that were this or something like it in general use by state surveys, it would, as years go by, save much useless repetition of field observations. A later worker might then in- herit and understand the observations of one or more predecessors. The first essential of satisfactory field work is a map, and now that the topographic sheets of the United States Geological Survey are be- LVI—Butt. Grow. Soc, Am., Von. 16, 1904 (411) 412 J. F. KEMP—GEOLOGICAL BOOK-KEEPING . coming so numerous and widely distributed they can be taken as the basis of work in all parts of the country. They will therefore be as- sumed as the foundation of the system, although it will apply to any map. The subject may be best set forth under two heads: The field map and note book, and The compilation book. THE Fretp Map anp Note Book The usual methods of employing maps and locating observations, so far as the writer can learn, are the following: The note book being ruled in squares, a suitably sized piece of the map is pasted upon a page so that locations can be made, as in atlases, by a row of letters across the top and numbers down the sides. Between intersecting horizontal and vertical lines any special point may be noted. With numerous and closely set observations this system does not admit of great accuracy, © and it is difficult to locate oneself accurately when away from the sides and top of the sheet. According to another method, the observer, when setting out from a base or center of operations, draws a line on the map showing his course, and upon this marks consecutively numbered stations where specimens or notes are taken. This is very accurate, but in detailed work it leads to a confusing multiplicity of lines, and does not work up to systematic compilation as an end-result. | In the method here set forth the usual topographic sheet is lightly ruled with waterproof india ink in squares of a mile or two miles on a side according as the scale is gzkgp Ol geztaay- This gives useful areas in estimating size and distance. Each second vertical and horizontal line is drawn double, so that the double lines enclose four squares of single lines. At the intersection of the single lines numbers are written, also in waterproof india ink, and in this invariable order: Beginning in the upper left-hand corner the numbers 1, 2, 3, and so on to 7, are written at the intersection of the single lines, as shown in figure 1, so far as it goes across. Below 1, 11 is put on the next row, and so on across, 12 running under 2, 13 under 8, etcetera. The next row begins with 21, and is all in the twenties, and so on to 81 and other eighties along the bottom row. This makes an invariable decimal system. The four squares around each number are now themselves numbered in an invariable order,-as shown in figure 1, but these numbers are not placed on the map, since it requires no effort to recall that the northwest square is 1, the northeast 2, the southwest 3, and the southeast 4. We thus have a system by which each square has its own number, as 1, 2; 43.3; 61.4, etcetera. Each square is now subdivided into ninths, and the ninths are enumerated, as shown in square 12.1 of figure 1. They are not put down on the map, however, as it is easy to estimate them, FIELD MAP 413 Lf: yi f sitams [ MG \ > lo) ASS \Za t S\N (Moe OK PON AZ ANN AD} } . : 4 | (AS Figure 1.—Northwest Corner of the Port Henry, New York, Quadrangle. Illustrating method of recording notes. In a full U.S. Geological Survey topographic sheet the numbers would read 7, 17, 27, etcetera, in the right vertical column, and 81, 82, etcetera, to 87 below, in the lowest horizontal row, de el? yf be a um 414 J. F. KEMP—GEOLOGICAL BOOK-KEEPING The use of ninths also keeps us within the limits of single digits. For greater accuracy each ninth may now be considered to be divided into other ninths, as shown in 12, 1.6, and beyond this it is almost never necessary to go. The star (*) on the map, figure 1, is in 2,869 (read, two, three, six, nine). The plus sign (+) is in 11,827; the multiplica- tion sign (X) in 22,455. In the note book one would merely record 11,327 and the observation, after placing a dot or little cross on the map: For field use the maps, after being drawn with squares, should be trimmed of the borders to the quadrangular lines and cut in thirds along the parallels of latitude. Each strip is then doubled back on itself once, map side out, and folded forward in quarters, map side in, so as to make a little accordion-like book, which can be tied in the note book and turned over like the leaves of a book. On the back of each should be written the name of the quadrangle, and “ north third,” “ middle third,” and ‘“‘south third,” as the case may be, and the year or years in which the maps were used in the field. When ruled in squares of this size there will always remain except in a few fortunate latitudes a strip at the right side of the map which does not make an even inch. Latitudes 41 degrees 30 minutes to 42 degrees, ‘for example, are almost exactly divisible, but those to. the north and the south are not. There will also always remain at the bottom of the map an incomplete lower row about half a square high, and in the eighties. These fractions, however, cause no difficulty, because the proper numbers of the parts remaining are perfectly apparent and the others simply drop out. In cutting up the maps into three strips along the parallels of latitude other squares are cut into fractional parts, but this also occasions no serious trouble, because the order of enumeration being invariable, if we have the numbers of one row before us, those of all the others adjacent os ‘id : are at once apparent. Thus in figure 1, in the fractional strip on the ~ right, we can locate ourselves just as well from the column of the squares, 2, 12, and 22, as if we actually had the appropriate numbers, 5, 13, and 23, which are cut off. As a variation and perhaps as an improvement upon the above method of drawing the squares, it is, of course, possible to place the double lines at intervals of three squares, so that each square of double lines encloses nine of single lines, then number in the same way and go down to any desired accuracy by subdivisions of ninths. Or itis possible, as the sheets are all divided into nine parts by the intersections of the meridians and parallels, to make the rectangles, thus afforded, the basis of the primary numbering and then go down by ninths, never needing double figures for © the large unit area, but this latter plan throws us out of the mile-on-a-side squares which are very useful scales of distance. ENLARGED SECTION OF FIELD MAP 415 Figure 2 illustrates one-ninth of the same quadrangle which is used for figure 1. Its northwest corner is at the intersection of latitude 44 degrees 15 minutes, with longitude 73 degrees 80 minutes. As the quad- - Fiaure 2.—One-ninth of a United States Geological Survey Topographic Sheet in Latitude 44. The ruling is designed to show the coordinate method by ninths, 416 J. EF. KEMP—GEOLOGICAL BOOK-KEEPING rangles are situated farther south, the east-and-west dimensions widen appreciably, and for those farther north they become more narrow. With the increasing width the ninth of the fourth dimension becomes larger and it may be desirable to add a fifth ninth. This, however, goes automatically, and it is slight trouble to use the additional digit. The variability of the east-and-west spaces makes it necessary to refer to the scale of each sheet in order to use the subdivisions as scales of actual dis- tance, but should the method receive any extended adoption it would be a great convenience to field workers if the United States Geological Survey would print upon the map for those who wished light surcharges. of rectangular ninths of the second dimension, the 5 minutes of longitude and latitude in the scale 7545, forming the ninth of the first dimension. For surcharges of this sort a series of squares, such as a mile on a side and having an invariable size, would be much less expensive and more convenient than one whose size varied with every 15 minutes of latitude, and therefore the writer favors the use of those a mile on a side and in blocks of four. THE COMPILATION Book For the compilation book a blank book is selected of eonvenient size, and preferably a standard one, which can be always obtained of the dealers. It is then paged in a series corresponding to the squares, usually 1.1; 1.2; 13; 14; 24; 2.2; 2.8; 2.4; 3.1, and sosqmetauaee for the first method of squares, or up “to 9.9.9 for the last one. Now, taking the field book day by day, the notes are copied into the compi- lation book, each on the page corresponding to its square. At the top of page 1.1 is put the observations in 1.1.1; andif 1.1.9 is first met in transcribing the field notes, its record is placed at the foot. Intermediate ones are interpolated at intervening points as nearly as the compiler will estimate their serial location. Note books should all be numbered in series and paged. Two compiled notes may then read as follows: 1.1.7.5. B.26,117, Gabbro-gneiss N. 70 E.45 N. Spec. 261. Becomes massive 100 paces north. Glacial scratches N. 60 E. and N. 40 E., former older. Trap dike, 4 ft. wide. N.45 W. Spec. 262. 1.1.8.9. B.45, 23, Beekmantown 1. s. 40 ft. exposed. N. 50 W. 10 E. Silicious, no fossils. Glaciation!! N. 60 E. From the first we know that in this particular square, as recorded in note book number 26, page 117, gabbro-gneiss occurs, with the strike and dip given, and with the associated phenomena subsequently recorded. The significance of the record for the next square is apparent. When the field work is completed in a quadrangle or any part of it, all the observations ever taken on any square, no matter if years have inter- ted COMPILATION BOOK 417 vened, are recorded upon one or two pages, and are available for use in drawing a map. In the office, therefore, and with a fresh map ruled in squares, the geologist begins with 1, 1.1, and plots in the geology square by square: Having a series of empty drawers at hand, the specimens are assembled as the work progresses, and grouped by narrow localities in an easily intelligible way. Should he desire to refer to the original note book and field- map, he does so instantly. If, moreover, one merely turns the leaves of a compilation book, it is possible to note by the blank, or sparsely written pages, where observations fail or are few. Revision or amplification follow, if necessary, as a matter of course. Should the observations be sufficiently detailed, they are available for sections at any points and in any direction. It is also advantageous to select a compilation book with sufficiently abundant pages, so that at the back petrographical or other office notes may also be included, and every- thing be thus kept together. For convenience of reference a fresh map is ruled, folded, and fastened in the compilation book, the field maps being left in the note book. Finally, from the first rough map the fin- ished one, which is to go with the final report, may be drawn with col- ored inks and a right line pen. ? Where large classes of rather numerous squads of two each are taught in the field, the compilation book can be passed from squad to squad, each pair writing in their notes. Finally, the record being complete, each squad can use the compilation book in preparing a final report and drawing a map of the whole district. With a party of ten or fifteen pairs located in a quadrangle which has been mapped on the scale of ;z4qq and which is suitably provided with means of transportation, within a week or ten days and within the limits of skill of younger workers, the whole area will be overrun in the greatest detail and almost every outcrop duly mapped and recorded. Areas are assigned square by square, reported on systematically to the instructor in the evening, and colored in on a compilation map. The students see the results gradually assuming definite shape, and great interest is usually aroused in the final result. When such compilation books are kept on file in an institution or placed with a state geologist, they would be of much permanent value. PRINCIPLES ON WHICH THE SysTEM Is BASED The above system is based on the principle of locality for the orderly entry of observations. Reflection on the subject by any experienced geologist will soon develop three possibilities: First, one may not com- pile at all, depending on the memory as a guide to the notes distributed through several field books. The field books may be esteemed of purely 418 j. F. KEMP—GEOLOGICAL BOOK-KEEPING ephemeral value, to be discarded after the final map and report are prepared. It may be and usually is assumed that they will be of no further use and will never be consulted by another or subsequent ob- server. Itis true that field books are often only intelligible to those who write them, and, as usually recorded, a later reader without the personal guidance of the original recorder would have difficulty in in- telligibly following a traverse with numerous observations; yet the field book is necessarily the most detailed of all the records, and provided it is intelligible, it contains what is of greatest, value to a subsequent worker. Its weak point is the frequent lack of a systematic, simple, elastic, and uniform method of recording locations, since, if the writing can be read at all, the uncertainty alone arises from the location of the observation. Over this method the system here set forth certainly has advantages both in the field and in the office. The second principle on which observations may be ayatemnahieaie grouped is that of time. An observer who makes no index or compiled summary usually, when trying to recall what he has seen in a partic- ular locality or region, does so by running back in his mind over his movernents in various years and months until in this way he can estab- lish the particular note book which is sought, yet as note books accu- mulate this process is slow and tedious and often elusive. As compared with it a method based on locality is much to be preferred. If note books are indexed or provided with a table of contents, the search is simplified, but even then, should a searcher wish to know everything which he or his colleagues or predecessors had observed in a certain square mile of a certain quadrangle, it would be a slow process to find this information by any scheme, however well indexed, if based on the time principle. The third principle is that of locality, and is the one suggested in this paper. So far as the accurate recording of observations is concerned, other methods may be as good as this one, perhaps even better, but none lead up to the compilation book, which is the great saver of time and effort. So long as a spot or a square mile or any larger area can be located on a standard map one can turn in a moment to a compilation book and find the condensed records of all observations, and, if the plan is systematically carried out, not only of one, but of many observers. The detail and completeness of the work is at once apparent and the progress of knowledge much facilitated. The principle of locality has, finally, incomparable advantages, be- cause, so far as areal geology is concerned, observations are alone of value when accurately tied up with a dente place. BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 71 VY VV VV VY VN YV VwYv Mv VV VA Vv SS = NS SAR SSS SX onsrs NS SS SA WN AQ SSRN NN NANA NAS Ws NA “Wann < WIV N Sys WOARRE SSG if LEGEND 4 GRANITE. SYENITE. NN NN Y S SOQOUWVYL BF 57 (SRA ey SSSA SERPENTINE. = == NE RAN J : BT RRR: VA, ow USS ANN SY YZ GNEISS. 6 NO SING NES Ne, = A\\ sx WA le HORNBLENDE SCHIST. a= SS XN ANS aN SO SS): TALCOSE SCHIST. SS SARIS SZ IS, ay SESS YON SERPENTINE BANDs. — Oo SY ys INNS sy = MICA scuist AS Y NN N SZ ISS CALCIF EROUSMICA-SCHIST. ¢ pI 6 77, (14) NN a) CLAY-SLATE. t A TALCOSE CONGLOMERATE. UPPER HELDERBERG. [@] tac 23> GEOLOGICAL MAP VERMONT GEOLOGICAL SURVEY 1861 GEOLOGICAL MAP OF VERMONT ie ae Sie Panne : - " <= BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 419-446, PLS. 71-81 OCTOBER 25, 1905 PETROGRAPHY OF THE AMPHIBOLITE, SERPENTINE, AND ASSOCIATED ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN, VERMONT * BY VERNON FREEMAN MARSTERS (Presented before the Society December 30, 1904) CONTENTS 2 Page Early views on the stratigraphy of Vermont.................000.0ees 3 aes 420 Semen or belvidereand Lowell areas £2... 2.002500 nsaed--cgeascepeveees 420 EEOC EEEPINTEATIC® OF SEPHEMEING: dio. 55 oc. a eee ee ce ce epee dee newescnss 422 IEE GE ICLVAGGTS BOMIOT. | oo givin. lok eed oss e Guo es tans dauas sens 422 a ol gees COREE GT A 4 LO RUN ie at eae De ee 424 SE EDL ST RS et Se ee se oe ie on be 424 I re ter pee Pert nee ee aac oneswedgeyccun 424 Peet Ate eIGN OF COMIPOSILION, 0... <2 26.2. .c es tee ese: Pacis sta eee NEE Yt ei nec iy eet ata to tae tel be awa As en RS SERGE SERS Gg RIS BS ae SRS ey CTO ats means be SO ee eee 426 Areal distribution....... ... .... See Sed ls Bhai ES tig Ua ahd a pds hab ies 426 Macroscopic characters eee Mineral COMPARLION ©. «6 «nj 5 nin se nlnges . 426 I Et ei re es See Sle gi OR sea y adap sane es Le ee 497 | Macroscopic characters and mineralogy ..... .........-¢. ae a 427 UNUIENNS SERIES. PNG ee tee os hye Owens « Stic rete Rapes teem aes 428 EIR tr iri sheteci ts Pac ee PN ae wee J. 496 arty knowledge of asbestos and ifs uses. -......0.5...-6. ccc ees tee enee 430 Development of Canadian industry (Thetford)............. 002.02 ee cee eee 431 RE IMR ca cia cl gn A fog itch s we pie vie. bo 4. 4a pve vik Ag es oe 432 I ied lye ado ¢.c) eee 439 Micro-structure of Belvidere serpentine..............0e see ee eee eee 440 TYPES... Sik alan hs Po dale Palak wc wate ene cee er . 440 Fibrous serpentine: ....\\ci0 5 Je, ste toda bad sees ee yee 440 Lamellar and bastite serpentine..... 0.5 22... weeds: ee eee eee 441 Discussion-asg to oricim. >). i542, 6.224 b peeiter Jala es whe oa\y,+ ie alate cdi ee 441 Contact’ phenomena... 6... 5.200028 int decline a 441 Applications of HEmerson’s VIOWS.; : <))52 50> 24s ene eee pia eee 442 Analyses of type rocks and fiber....... 6.00000. 5.02005 0 0a too ee 443 Conclusions.) 5 cicck en's eck eel sew Oo ee ee GE eee as S45 EARLY VIEWS ON THE STRATIGRAPHY OF VERMONT The occurrence of serpentine and associated minerals was early known to the pioneer geologists of Vermont. In the report prepared by Pro- fessor Edward Hitchcock and his co-workers (1861) frequent mention is made of them, and it is stated that in many instances more or less as- bestos and talc, in several varietal forms, are to be found. While con- siderable preliminary prospecting was carried on in the early seventies, no industry of any moment was established until a comparatively late date. According to the observations of Professor Hitchcock, the serpentine formations are very largely confined to a broad band of so-called talcose schists which enters Vermont on the north in Orleans county. At the northern boundary of the state the schist belt has a maximum width of some 15 miles, with its eastern limit near the western shore of lake Memphramagog. This series of metamorphics is shown on his map as extending the entire length of the state and occupying portions of Orleans, Lamoille, Washington, Addison, Orange, Windsor, and Windham coun- ties and having a minimum width at the southern boundary of the state of some 23 miles. While some ten occurrences of serpentine are reported to exist in the southern half of the talcose belt, by far the largest of these deposits is located in Orleans and Lamoille counties. Itis very probable that this serpentine and the associated schists belong to the same group as those of the Thetford region. The schists are regarded by the Canadian Geological Survey as Cambrian in age. LOCATION OF BELVIDERE AND LOWELL AREAS The small area under consideration in this paper lies within the adjoin- ng townships of Eden, in Lamoille, and Lowell, in Orleans county. a4 a 4 ey ce Nee = YOR . aa . ie te ee FORSS AKGALAT | TS 4 i VOL. 16, 1904, PL. 72 UU" = | VA WSS ANS < ~ ; SSE [TET SS SS GHARH AD mre TH MLO 10 “Tey! “NT ita Oe CEOLOCICAL MAP ASBESTOS AREA. ==] TALCOSE SCHIST : VERMONT . || il CLAY SLATE. PREPARED from MAP | "of GEOLOGICAL SURVEY » UPPER HELOQERBERG 1861 : LIMESTON=. ; : CALCIFEROUS MICA SCHIST. | IN QUEBEC AND VERMONT ee ee * BELVIDERE AND LOWELL AREAS 421 According to the Hitchcock map, the serpentine bands, of which two are reported, terminate near the central part of Lowell township; the west one, however, is represented as extending a little south of the village of Lowell. In Eden township no serpentine is represented. There LEGEND AMPH! B OLITE == Y, serpents GZ GARNET ZONE I Y hy, $ a dy Ficure 1.—Geological Map of Belvidere Mountain. is, however, in Eden, and continuing in Lowell, a most interesting area of these magnesian rocks, forming the southern slope of Belvidere moun- tain, and a plateau-like projection to the south and east of the mountain crest. This may be designated as the Belvidere area (see map, figure BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 72 TOT 7 I Tati FESS) cwenoacveian 5 = Meares wert. KS CEOLOCICAL MAP PROVINCE of QUEBEC SHOWING ASBESTOS AREA PAEPARED FROM MAPS OF GEOLOGICAL MAP ASBESTOS AREA . VERMONT . PREPARED from MAP "ef GEOLOGICAL SURVEY GNEISS. GEOLOGICAL MAPS OF ASBESTOS AREAS IN QUEBEC AND VERMONT 4922 Vv. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN 1). So far as the data at hand bear on the question as to the limits of the area, it is to be said that all field facts strongly suggest that it is not connected with the belt passing through Lowell village. Economic IMPORTANCE OF SERPENTINE Apart from the scientific interest in the origin and development of the serpentines, they are also of special economic importance, inasmuch as a considerable amount of asbestos is known to occur within the limits of the region under consideration. Small lenses of talc, too, are not un-. common within the central part of the Belvidere area, as well as in the Lowell belt, but the asbestos is by far the more important economic product of the two. It should be added, however, that it does not follow that tale is always of secondary importance in serpentinous rocks. On the contrary, in the region of Moretown, Washington county, large tale deposits are being opened and a mining plant is in process of construc- tion. In this case, so far as can be determined from the preliminary prospecting, asbestos-like minerals are quite secondary to the develop- ment of tale. The talc appears to occur near the contact of serpentinous rock, with talcose and micaceous schists or as small lenses within the serpentine. It is not improbable that careful prospecting within the limits of the Moretown area may bring to light other tale deposits of sufficient economic importance to form the basis of a small but profitable industry. TOPOGRAPHY OF BELVIDERE REGION Topographically this region consists of a series of valleys and ridges having a general northeast-southwest trend. Near the west corner of Lowell lies Belvidere mountain, a sharp-crested ridge, with its highest point at its southern extremity and within a few rods of the Lowell-— Eden township line. The eastern side of Belvidere is shown in the accom- panying photograph. The altitude of the crest gradually decreases to the north. A steep-sloped valley, locally known as Hazens notch, in the southwest corner of Westfield, separates Belvidere from the ridge to the north. Hadley mountain is separated from Belvidere by the west branch of the Missisquoi river. From its northern flank a low spur extends to and beyond Lowell village, where it is crossed by Johns branch, a tribu- tary to the Missisquoi. A part of this spur is locally known as the Leland hills. In the southwest corner of Lowell are the so-called Lowell moun- tains, with the same general trend as Belvidere. The greatest altitude of Belvidere is approximately 2,100 feet above Eden corners and some 1,200 feet above the office of the New England TRAVERSE MAP OF BELVIDERE MOUNTAIN A498 Mining and Development Company. The upper half of the ridge pre- sents exceedingly steep cliffs, occasionally alternating with talus slopes and “slides” containing enormous blocks of rock, the form of these masses being due to the development of two well defined sets of joints. To the north, however, the steepness of the slopes gradually decreases, and a cover of waste extends well up the flanks. At the south end of Belvidere a somewhat crescent-shaped plateau rims the south and south- east sides. This topographic element is largely composed of serpentine, @rerorteo. ©),not COLLECTED. (O).THIN-SECTION NO. SW.SERPENTINE IN WASTE. Sw.SCHIST IN WASTE Figure 2.—Traverse Map of Belvidere Mountain. the level portion being comparatively free from drift or talus, except along its upper edge and again at its foot, where it is in part covered up by sand or gravel deposits, forming terraces on the bottom and along the lower part of the Missisquoi valley. A typical view of the outer slope of the serpentine plateau, as well as the valley at its foot and the eastern ridges of the Green Mountain series, is shown in the accompany- ing photograph. The gradual extension of the glacial deposits up the slopes to the north and their coalition with the debris from above, the whole being covered with a dense forest, renders the lithologic relation- 424 v. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN ships of the area between the plateau and Hazens notch very difficult to ascertain. Rock Typrs GENERAL DISTRIBUTION Within the area under consideration three widely different types of rock are to be found, namely, schist, amphibolite, and serpentine. The first occupies the valley floors and lower slopes of the Missisquoi; the second forms the uppermost 1,000 feet of Belvidere mountain, and the third occupies the area between the schist and the amphibolite. SCHISTS Brief discussion of composition.—It will be seen from an inspection of the map showing the distribution of a part of the formations of the state that the Belvidere area lies entirely within the so-called talcose schist’ as mapped by Hitchcock and published in the report for 1861.* On account of the lithological characters and mineralogical associa- tions the schists of Vermont were early regarded by Hitchcock and others as a part of the great terrane extending from southeastern Canada (Quebec) to Alabama. Among the first scientific men to devote atten- tion to these peculiar rocks was T. Sterry Hunt, who carried on a series of chemical investigations with the hope of gaining some knowledge of the origin of this rather prominent series of metamorphics as they appear within Canadian territory. His studies led to the conclusion that they were derived from slates. His chemical results, moreover, brought out the fact that the name “ talcose ” as applied to the schists was a misno- mer. This fact is evident from an inspection of his analysis, which is appended : Site a Wee a Sw Oak eee eu ote eee Ce 66.70 Aamir ons tee SG 25 ees) Se ESE eee ee 16.20 Peroxyd Of 1f0n ie ido eee done tee ee ee eee 6.90 LEV Cc OV ere a3 iA ERLE EL Sa 2 ON ee ee Bate Mit .67 Mae mesial s wie are. ae ails et meumhatiaase haeu etree ieee ase eee 2.75 Alkalies (by difference)... Glee kek s cee ae eee 3.68 Water i 1o bok euieckl. ai amte het he ceenchs Pick ned, hs ee 3.10 100.00 Thus it is seen that less than 3 per cent of magnesia is to be found in this sample (Saint Marie), and even in the slates with which the tal- cose schists are closely associated no$ more than 8 per cent. It is * Edward Hitchcock : Geology of Vermont, vol. 1, p. 504. Map, vol. 2. COMPOSITION AND AGE OF SCHISTS 425 thus evident that, in so far as the Canadian representative is concerned, the above term is not at all applicable. The announcement of these results stimulated further study of this same problem within the limits of Vermont. The investigations of Mr G. H. Barker under the directions of the state survey produced similar results. The specimens analyzed came from Roxbury, Pownal, and Middlesex respectively. The name “‘ talcose” as applied to the Vermont schists is also an un- warranted term, although it is to be said that tale and talcose lenses are not uncommon within the limits of the schist formation. Age of the schists—As to the age of the schists, various interpretations have been offered. If I am correct in correlating the schists, with which the asbestos-bearing serpentines are associated in Quebec, with the “occurrence under consideration, the Vermont series would be regarded by the Canadian geologists as belonging somewhere in the Quebec group. The term “Quebec” is applied to a large series of metamorphics, including slates, schists, and serpentine deposits. It is probable that it, in reality, included both Cambrian and pre-Cambrian rocks. More- over, the occurrence of graptolites within the limits of the Quebec group also shows that it includes rocks of Ordovicic age as well. Where in the series the schists and the associated serpentines belong is not yet clear from the evidence at hand. It has been stated by the Canadian Survey that the schists are in all probability Cambric in age. The southern extension of the schists, as represented in Old Hampshire county, Massachusetts, and Windham county, Vermont, has been studied in very great detail by Professor Emerson. Using the Vermont geolog- ical map as the basis for comparison, the talcose schists of Vermont are continuous with the Goshen schists of Franklin and Old Hampshire counties. This terrane is considered as Siluric by Emerson. Inasmuch, however, as the United States Geological Survey did not adopt the term “ Ordovicic,” as suggested by Lapworth, for Lower Silurian in the no- menclature of geologic folios until this year, it should be noted that the term “Silurian ” as used by Emerson probably includes rocks of Ordo- vicic or possibly earlier age. If this interpretation be correct, it is clear that the Quebec group of the Canadian Survey and the Silurian as used by Emerson may overlap or include one or more formations of one and the same age—that is, the upper members of the former may be syn- chronous with the lower members of the latter; but to what extent they may be chronologically parts of the same formation can not be deter- mined with the data at hand. Such evidence as can be obtained from 496 V.F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN the literature suggests that the Vermont series may be the late Cambric or early Ordovicic in age. AMPHIBO LITES Areal distribution.—An inspection of the map of the Belvidere region will show that the main area of amphibolite is confined to the uppermost 1,000 feet of the mountain crest at the southern terminus of the ridge. This rock probably makes up the tip of the ridge as far as Hazens notch ; how far it may extend beyond has not been determined. Small but important exposures of the same rock occur at the lower edge of the serpentine and in prospects opened up by Judge Tucker. There are thus three areas; but the two smaller ones are particularly important, inasmuch as they provide certain data which it is believed _ will make clear with reasonable certainty the true lithological relation- ships of the amphibolite and the serpentine and the origin of the latter. At the south end of Belvidere the amphibolite forms the steep sloped portion above the plateau-like terrace, as seen in the accompanying illustration (plate 78, figure 1). Macroscopic characters and mineral composition. pile very large portion of the rock is composed of dark-green hornblende, so arranged with respect to the cleavage of the individual components that a gneissic and, at times, a decided schistose structure is fairly well defined. At the base of the hornblendic mass, or just above the level of the rimming plateau referred to, the amphibolite presents an additional phase in becoming highly garnetiferous. The relation of the garnet-bearing zone to the underlying serpentine is not easily determined, as the probable contact is covered by waste from the cliffs above. Ata few points in the rear of the plant of the New England company garnet was recognized as high as 70 feet above the foot of the talus slope. So abundant is the garnet at various points along the base of the cliff that the rock assumes a marked reddish color, and thus, at first sight, might be regarded as quite distinct from the normal amphibolite. From this point upward the garnet gradually disappears. Above the 70-foot mark (barometric determination) it could not be detected in any of the hand specimens with the unaided eye. The microscope, however, reveals the fact that garnet does occur in small amount above this penhike The main mass is nevertheless confined to the base of the amphibolite, and probably represents a local contact phenomenon. Its extent, so far as known, is indicated on the map of the Belvidere region. The texture of the normal amphibolite is quite variable. At the top of the ridge it is a fine grained mass, with occasional bands of feldspathic an BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 73 Figure 1.—BeLvVIDERE Mounrain FROM THE NORTHEAST? Figure 2.—Missisquor VALLEY LOOKING EAST FROM HIGHER SLOPES OF BELVIDERE Mountain BELVIDERE MOUNTAIN AND MISSISQUO!] VALLEY ete. wae LJ _ : — {THE AMPHIBOLITES 427 minerals and a tendency to split in planes parallel to the minute mineral constituents, 90 per cent of which is hornblende. Half way down the slopes on either side of the highest point the rock assumes a finely fibrous texture. Thisseems to bea local condition, and does not extend through any great thickness in the central portion of the formation. Again, along the fault on the southwest side of the mountain the amphibolite takes on a fibrous and, at times, a lamellar structure. This also is quite local, it being induced by the crushing and shearing consequent on the fault- ing. The main body of the formation is moderately coarse in grain, showing cleavage planes 6 or 7 millimeters in length and 3 or 4 in width. It is not uncommon to find in cross-sections of individual hornblende crystals a marked bending or curvature, best shown in sections along the longer axis. Thecleavage conforms very closely to one direction, so that the rock splits very easily into plates with an apparent schistose structure. Both pyrite and magnetite grains have been detected with the unaided eye, but in most cases they are in very small amount. The remaining exposures of amphibolite do not differ in any respect from the typical representative of the coarse grained type, already de- scribed. In both exposures there is clear evidence of marked crushing and shearing, particularly in the exposure on the brook as shown on the map. Unfortunately these areas are very largely covered beneath gravel and waste from the slopes, so that it is impossible to determine the extent and relationship of these two small but important exposures. The details of the contact of the amphibolite with the serpentine, as shown in the Tucker quarry to the north of the brook section, will be discussed later. It is sufficient to state here that this contact is believed to furnish sufficiently clear and unmistakable proof of the intrusive origin of the rock from which the serpentine has been derived. SERPENTINE Macroscopic characters andjmineralogu.—The serpentine is remarkably uniform in texture throughout the entirearea. It is rather fine grained, grayish green to dark oily green in color and splintery or hackly in fracture. On freshly broken surfaces it is not uncommou to discern a tendency to assume a lamellar or more often a fibrous texture. The fibrous tendency is invariably associated with zones of shearing and possible thrust. In the upper part of the serpentine belt this texture is most evident along a zone passing through the open cuts of the New England Mining Company. When examined under a hand glass it will be seen that much of the fibrous content of the rock is in reality confined to minute seams and shearing planes and stretched out LVIII—Butt. Grou. Soc. Am., Vou. 16, 1904 428 v.F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN along the line of movement. It would thus seem that the fibrous con- tent, inasmuch as it fills cracks and seams, even of microscopic size, must be secondary in origin. The stretching of the so-called fiber is due to subsequent slipping after the deposition of the mineral within the orig- inal fracture. The same conditions obtain at various points. It is very well shown in the prospects of the United States company (Blake prop- erty) 250 feet below the mill of the New England company, as well as in many of the openings made by Tucker, Stone, and Farington along the lower edge of the serpentine deposit. Occasionally on the fresh fractures minute lamellar structure may be seen. Under the hand lens they appear to be small silvery to light-green scales, but too small for specific determination. On the basis of structure it is reasonable to con- clude that these minute individuals may be some form of the lamellar serpentine, to which the name Antigorite has been applied. It is to be regretted that sufficient material is not at hand to determine this point. In the Tucker prospects, however, considerable variation in color and an appreciable range in texture is noticeable. The color varies from light grayish green to oily dark green, and brilliant dark green when wet. Scattered through the mass are many patches of magnetite. with which is associated a very small amount of chromite. It is in the Tucker quarry that the best showing of cross-fiber in this serpentine area can be seen. The dark bands of serpentine are intimately associated with the fiber-bearing veins. They form the adjacent walls, and vary in width with the size of the vein from one to several inches. *When the veins form a minute network the whole mass may assume the dark oil green color and then form large bands and blotches 2 or 3 feet in diameter. This peculiar association, so far as observed, is confined to the prospects of the Tucker property. Local tule lenses—The main part of the serpentine deposit is a fine grained light grayish green rock, sometimes exhibiting a tendency to become talcose, and wherever this occurs it is accompanied with a slight schistose structure. In fact, on the line of the cross-section, at a point about 150 feet below the edge of the plateau, a number of talc-bearing lenses may be seen. They contain moderately pure talc in the centers, but grade out into the normal serpentine on either side with loss of the schistuse structure. From an economic point of view, they are too small to invite any investigation. STRUCTURE OF THE REGION The structure of the region is indicated in the accompanying cross- section. Amphibolites form the upper 1,000 feet of Belvidere, and Min te ee ee ee ee ee ee es oe STRUCTURE OF THE REGION 429 include the steep slopes extending upward from the crescent-shaped plateau, the latter being composed of serpentine. At the juncture of the two rock types is a band of garnetiferous amphibolite. An impor- tant fault occurs on the west side of Belvidere. This was first recognized by Professor J. F. Kemp. How extensive the faulting may be was not determined. It is to be said, however, that certain features in the topog- raphy of the western slopes suggest that series of faults may occur to the north and may have been a determinative factor in the minuter topography of this unique and picturesque ridge. On the eastern slopes no faults were detected. Much local crushing and local shearing is evi- dent within the limits of theserpentine. Such shear zones are intimately associated with the so-called asbestos. On the Tucker property, at the foot of the platcau, are to be found certain structural facts which at first ‘suggested the occurrence of a zone of crushing and shearing. At this point a contact between the serpentine and amphibolite is unmistak- ably clear. This was at first regarded, on the basis of a field observation, as a fault, the crushed zone of serpentine forming a very favorable depos- itory for the cross-fiber. It will be shown, however, that the contact of the serpentine with the amphibolite does not represent a fault plane, but an igneous contact. The development of a crushed zone adjacent to the contact is necessarily a subsequent phenomenon. The remaining in- stance of a possible contact relationship is to be found a few hundred yards to the south of the preceding case. Its location is shown on the accompanying map. Here crushed amphibolite was found within a few feet of the serpentine. Whether, as in the preceding case, an igneous contact or a simple fault zone may exist here could not be determined with any degree of certainty, but it is believed the former is the case. Unfortunately the greater part of the rim or foot of the plateau is buried beneath a coat of waste and till, which renders further discovery of structural relationships well-nigh impossible. Protruding through the till may be seen excellent exposures of the surrounding schists, which form the valley floors and subordinate ridges in the immediate vicinity. It maintains a very steep dip, with a strike varying from nearly true north to north 20 degrees east. A consider- able variation in the lithologic characteristics of the schists obtain. The variation may be marked in a comparatively short cross-section. Bands of micaceous rock, which at times are apparently talcose to a slight de- gree, are followed by sandy schists, including small lenses of quartz, with an appreciable amount of pyrite. No limestones were found in the vicinity of Belvidere. The positions of a few of the outcrops noted are indicated on the map. The structural features, in conjunction with the 430 Vv. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN mineralogy of this formation, favor the view that at least a large part of this terrane is sedimentary in origin. EARLY KNOWLEDGE OF ASBESTOS AND-.Its Usks It is a well known fact that the early Greeks and Romans were ac- quainted with the economic uses of asbestos fiber. The idea of an or- ganic origin has been attributed to Herodotus, who considered it “a kind of mineral wool” which grew on trees. According to the same writer, cremation cloth was made of this material, it being used to completely cover the body, hold and keep separate the remains from the fuel used in the funeral pyre. Pliny alsospeaks of it asa rare and costly cloth, the funeral cloth of kings. Assuming its vegetable origin, he calls it “ linum vivum,” the difficulty of weaving which, he says, was very great on ac- count of the shortness of the fiber, which is a somewhat singular reason to give, seeing that he was not speaking of chrysotile, to which the re- mark might with ‘some justice have been applied, but to the Italian mineral amianthus.* It is known that asbestos was made use of in connection with the sa- cred fires in the temples of the gods. Itis also stated by some writers that this material was employed in the preparation of wicks for the lamps of the Vestal virgins. Strabo and Plutarch both speak of these lamps, calling them dofeora (perpetual), because the wicks maintained a perpetual flame without being consumed.f Asbestos!cloth was probably used as well for the preparation of napkins and certain forms of dress. It was not, however, until the beginning of the last century that the application of this mineral product began to be recognized in the me- - chanical arts. Moreover, its uses as known in the olden times may still be seen in Greenland and Labrador, where the natives are known to twist it into lampwicks. While it is more than probable that asbestos paper and cloth were manufactured as early as 1700, in Norway the modern history of asbestos as a useful commodity dates from the beginning of the last century. Early in this period we find Madame Perpenti of Cone engaged not only in the manufacture of paper, but also cloth from an Italian product. Its first use as a means of protection to firemen is attributed to Chevalier Aldini, 1850, and a little later we find one Guiseppe della Corona, a cult- ured Florentine priest, engaged in the manufacture of millboard; and, later still, Signor Albonico, having given some attention to this product of the mountains of his native province, joined himself with a priest, Corona, and the Marquis di Baviero, a distinguished Roman nobleman, * Jagnaux, Traité de Min.; also Quenstedt, + Quenstedt, Handb. der Min. EARLY KNOWLEDGE OF ASBESTOS AND ITS USES 431 in the manufacture not only of asbestos cloth, but asbestos paper as well. Having to some extent succeeded in this, they were unsuccessful in their endeavors to secure a contract with the Italian government for the sup- ply of paper for the use of bank notes and other securities, any prospects _ they might have had in other directions being effectually destroyed by the outbreak of the Franco-German war of 1870. - DEVELOPMENT OF CANADIAN INnDustTRY (THETFORD) In this country and Canada chrysotile was not known to exist in any quantity until-the seventies. It was discovered in Canada long before this date, for in 1862 specimens of Canadian chrysotile were exhibited at the International Exhibition in London, England. It was then regard- ed asa mineralogical freak rather than a product of great economic impor- tance. Ina report of the Canadian Survey for 1847-1848 reference is made to its occurrence in serpentine rock in the region of Bolton. But, while the extension of the belt of serpentine rocks in which this mineral is known to occur had been traced with some care from the Vermont boundary in the township of Potton to and beyond the Chaudiere river, the deposits of asbestos observed were comparatively limited. In the United States veins generally of short and harsh fiber were found at several points, and a considerable quantity of a tremolitic variety was mined, which, while ill adapted for the purpose to which asbestos is now generally applied, was used for the manufacture of fireproof paints, cements, etcetera. The chief source of supply for fibrous asbestos was the mines of Italy, where deposits of irregular extent occur, the mineral often possessing a long and silky fiber, which well adapted it to spinning, and from this source the material for fireproof curtains and similar manu- factures were obtained. In 1877-1878 asbestos was discovered in the serpentine hills of Thetford and Coleraine. ‘The size of the veins, often several inches thick, led to the expectation that deposits of value might exist there, though their true importance was not ascertained for several years. The credit of the discovery in this locality is claimed by Mr Robert Ward, though by others it is stated that the first find was made by a Frenchman named Fecteau. Following closely upon the discovery several parties secured areas both at Thetford and Black Lake, in Cole- raine township, on the line of the Quebec Central railway, which for some milesruns . . . between high ridges of serpentine, in which, the timber having been burned off, the veins were observed at the surface by the weathering and felting of the mineral on the surface of the bare rock. From this time (1878) on the industry developed with the demand for the product. Early in the nineties the region became the leading producer of the world, and holds that position at the present time. In 432 Vv. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN 1903 the United States produced but 2.4 per cent of the value of the imported product. This condition should stimulate the most strenuous search for so valuable a product. GEOLOGY OF THE CANADIAN AREA Concerning the lithology of the Canadian area, Mr Ells says: ‘*All the mining locations are situated on areas of serpentine, which is associated with green, gray, black, or reddish slates and quartzose sandstones and conglom- erates. The serpentine is more particularly related to considerable masses of diorite and whitish granitic rock, and is apparently due to the alteration of por- tions of these masses. . . . These slates and associated rocks are supposed to belong to the Cambrian system, though the serpentine is sometimes connected with areas of older rocks, such as chloritic and talcose schists and considerable masses of soapstone. In the serpentines which are found with the older rocks the asbestos appears to be in very limited quantity, and no attempt has been made to work any such deposit.” # North of the Saint Lawrence river, in Ottawa county, are a series of serpentinous limestones, of probable Laurentian age, which contain a small amount of fiber, but its grade is too low to be of any considerable value as compared with the products from the Thetford region. Ser- pentine areas are known, too, to exist in a line extending to the northeast in the direction of the Gaspé peninsula, but little is yet known about these occurrences as possible producers in the near future. VERMONT ASBESTOS OCCURRENCE The discovery of asbestos in this region is to be accredited to Judge M. E. Tucker, of Hardwick, Vermont. On November 9, 1899, he recog- nized the possibility of its occurrence in sufficient quantities to warrant careful prospecting. A considerable area lying along the township line between Lowell and Eden, as well as a portion of the belt passing through Lowell village to the north, was examined with considerable care. At a later date portions of the Lowell area were investigated by Mr Silsbey, of Lowell. During the two succeeding years the Belvidere area attracted the attention of some of the more prominent miners of asbestos in the United States and Canada. In 1900 Mr B. B. Blake likewise discovered fiber in the ledges on the southeast slopes of Belvidere, in Eden town- ship. These finds at once increased local interest, and later led to the formation of a number of companies, but only one proceeded beyond the prospecting stage. In 1901 the New England Asbestos Mining and Milling Company erected an elaborate plant, equipped with the most Sa) et : . ed 4 er oo , BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 74 VALLEY AT FOOT OF BELVIDERE MOUNTAIN, LOOKING SOUTHEAST Showing plant of New England Company VERMONT ASBESTOS 433 modern machinery for the treatment of asbestos-bearing rock. Active mining operations were begun in May, 1902, but in October of the same year the plant closed its doors. No official statement has been obtained concerning the amount or the grade of the fiber produced or its value in the markets at that time. That seen by the writer, which was said to have been the product of the New England mines, while too short for purposes requiring tensile strength, should fill the standard require- ments in the manufacture of all asbestos goods in which non-conductivity of heat is the only essential quality desired. A view of the New England plant is shown in plate 74. | KINDS—SLIP-FIBER AND CROSS-FIBER While it is universally true that chrysotile is confined to serpentine, it is by no means true that all serpentine deposits contain it. Although the Belvidere area is a difficult one to investigate as regards struc- tural details, sufficient data have been obtained to demonstrate with reasonable certainty that the fiber contents are very largely restricted to certain belts, the localization of which is due to structural feat- ures confined to the serpentine deposits. In the open cuts and pros- pects made, especially by the New England Asbestos Mining and Mill- ing Company and others near the upper contact with the amphibolite, the fiber, when in sufficient quantity to be easily detected, shows that it is largely confined to a shattered and sheared zone of rather limited extent. In the central part of the zone are large masses with slickensided surfaces. Wherever cross-sections of these could be found careful examination revealed the fact that they in turn were also sheared, and could with a little careful manipulation be separated into a series of smaller wedge-shaped masses, each with smoothed surfaces. It is along these planes that the fiber has attained its maximum development. It occurs in two forms, and has been so recognized by the local prospectors. In a large part of the zone of shearing the fiber has stretched or pulled out along the slipping planes, and hence has been called “slip-fiber.” In certain parts of the area, however, there has been a maximum devel- opment of fracture with minimum shearing. Insuch fractures the fiber has assumed a transverse position, and is locally known as “ cross-fiber.”’ The same phenomena may be seen in thin-sections prepared from smaller blocks. With regard, therefore, to the structural details bearing on the areal distribution of the fiber the facts as they now appear force the conclu- sion that the fiber will be limited to the zones of fracture and shearing. How many of these exist within the serpentine zone under consideration is a difficult matter to determine. Enough data are available, however, 434 Vv. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN to indicate that a zone of fracture with marked shearing crosses the prop- erty of the New England company. It is probable, too, that a smaller one crosses the property of the National company. A third belt of frae- ture with minimum shearing occurs on the property of Judge Tucker. This case is adjacent to the igneous contact to be discussed in another section. The fiber is largely of the “ cross-fiber ” type. It has the color and luster of true chrysotile. Under the blowpipe it behaves in all respects like the fibrous form of chrysotile. It yields considerable water and is nearly infusible. There is a tendency on continuous appli- cation of the flame for very fine films or strands of the fiber to become noticeably brittle. The cross-fiber was, however, quite uniform in its — behavior before the blowpipe. Still one or two samples did not yield as much water as should be expected in true chrysotile. How preva- © lent this may be has not been determined. The “ slip-fiber ” differs from the “‘ cross-fiber”” in having a duller and waxy luster, less flexibility, and a tendency to develop a coarser but longer strand, sometimes reaching 38 inches in length, while the latter rarely attains 1 inch in length. Under blowpipe tests the slip-fiber yields much less water and fuses, with some difficulty, to a white enamel. The color test for calcium was also evident. ‘These reactions strongly suggest that the slip-fiber is not true chrysotile, but a fibrous form of amphibole. This variety, however, is on the market as asbestos, but its market values are much less and its uses not so great in range. Figure 3.—Cross-section of a Slip-block. Showing the minute fiber veins in a position where the block would be subjected to the greatest strain, and hence offer the best lines along which the fiber might develop. At the lower contact of the serpentine with the amphibolite on the Tucker property is the best showing of cross-fiber seen in the entire belt. Itis only just to say, however, that it does occur as well on the land of the New England company, but in very small amount. While cross-fiber is the more prominent of the two sorts, shearing, which seems to be a constant associate with the slip-fiber, is by no means absent in the Tucker exposures. It is confined, however, to very narrow zones. On the other hand, the cross-fiber is very largely confined to the borders of the thrust zone, and, so far as could be seen from the new exposures, BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 75 Figure 1.—S.Lip-BLOCKS FROM THE TUCKER PROPERTY Showing rounded character of edges on flatter sides and film of fiber on surface Figure 2.—Sipe View or Brock NuMBER 4 Showing sharp edge of the slip-block and the slicken-sided and film-covered surface SLIP-BLOCKS FROM THE TUCKER PROPERTY BULL. GEOL. SOC. AM. VOL. 16. 1904, PL. 76 7 Figure 1.—BLOcK FROM THE SERPENTINE-AMPHIBOLITE CONTACY ON THE TUCKER PROPERTY Light colored or left-hand side of the block is serpentine; dark portion amphibolite. Sharpness of line of contact best shown at upper and lower edges FIGURE 2.—SECYION OF CROSS-FIBER VEIN Fiber at intervals is separated by films and bands of magnetite. Ore-bands may be central or lateral, but are rarely well developed adjacent to the serpentine. They sometimes fill what appears to be a fractured wall or edge of the fiber-band, which in some respects suggests a replacement of the serpentine by the fiber along a line of weakness, and not necessarily the filling of an actual crevice in the rock. SERPENTINE BLOCK AND A SECTION OF A CROSS-FIBER VEIN VERMONT ASBESTOS 435 occupied a greater breadth or area than the zone of thrust. Nowhere in the field are small slip-blocks better developed than at the Tucker contact. The accompanying illustration (plate 75, figure 1) gives some idea of the form and character of the surface. It will be seen that the surface has heen polished and sometimes covered with a fibrous film of hornblende, which has been stretched or pulled out in the direction of movement. In consequence of the shearing the trituration of the blocks upon each other has produced alternate sharp angles around the greatest periphery, with alternating obtuse edges roughly in a plane at right angles to the plane of the former and along its greatest diameter, or as many as three obtuse angles may occur on each flat surface of the block. In cross-section minute partings may be seen lying roughly parallel to the trend of the surfaces and filled with glistening films of fiber—that is to say, the strain to which these masses were subjected was so great that they in turn were also fractured aud subsequently knitted together by the deposition of silky fiber. These facts are shown in the sketch (figure 3) of the cross-section of one of these blocks. STRUCTURE OF CROSS-FIBER VEINS A cursory examination of any block containing typical cross-fiber veins shows that they usually occur in groups of bifurcating and rejoin- ing members, usually associated with one or two larger members not exceeding 1 inch in diameter. ‘This feature is shown in the following sketch. In thin-section of even the microscopic veins magnetite and probably a small amount of chromite are found to be constant associates. Pyrite was also recognized, but in very small amount. In the hand specimen it would appear that the magnetite was confined very largely to the cen- tral portion of the individual vein. In thin-sections of the smaller veins this relationship is not so evident. On the contrary, sections may be selected in which there seems to be no localization of the metallic con- tents, which are scattered throughout the entire width of the fiber band and extend well into the body of the serpentine. This, however, is not the rule, but rather the exception. From the material at hand I think we may safely conclude that the ores have a decided tendency to ac- cumulate within the central zone of the fractures, while the deposition of the fiber favors the walls. DISCUSSION OF PROBABLE MODES OF GROWTH The method of growth of the fiber and its relation to the ores is a vexed problem. Whether its deposition begins on the walls and gradually pushes its growth toward the central space and finally coalesces along a LIX—Butt. Grou. Soc. Am., Vou. 16, 1904 436 v. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN central line or whether the fiber develops simultaneously along its en- tire length is not easily demonstrated. If the latter were true and the deposition of the ores were cotemporaneous with the growth of the fiber we should not expect to find a tendency toward central localization of the metallic contents, for the chances for uniform deposition at all points within the crevice would be equally good. It is true that such condi- tions can be found, but do not conform to-the general plan observed. On the basis of the first hypothesis, if we assume that the fiber began its growth on the walls of the fracture and pushed its way toward the center, then we should expect an increasing amount of ore deposition in the central portion, with the infringement of the fiber from the sides. As a matter of fact the ore does, in the majority of cases seen, favor a central position, as if the circulating waters which carried the ore were compelled to move in a central zone. It follows, then, as a consequence of this sort of growth that a coalition and knitting together of the-oppos- ing ends as they came in touch should result. Moreover, it would seem that the line of juncture should be detected in cross thin-sections, for it would be most remarkable that the growth of opposite individuals or strands should have the same optical orientation and unite in such a way, where not interfered with by the deposition of iron ore, as to appear to be one and the same individual. In the few sections at hand [ have not been able to detect this relationship, but it is believed that further study of the problem may show that this interpretation is in the main the correct one. It might be said that the deposition of the ores was subsequent to the _ growth of fiber and along secondary fractures within the body of the latter. It is true that the fiber-bearing veins have been sheared, for it is this process that has produced the slip-fiber. This supposition may be true, but it still remains that veins which do not show any evidence of shearing or fracture also contain much magnetite in central partings or as rods and bunches parallel the strands; hence in such cases there was _ no opportunity for the subsequent deposition of the ore. While it must — be admitted that sufficient material and data are not at hand to solve the problem, it is believed that such facts as we have bream the interpre- tation as expressed in the above hypothesis. AMPHIBOLITE MICRO-STRUCTURE AND MINERALOGY In the predominant type of amphibolite hornblende is far in excess of all the other constituents combined. It makes up from 70 to 90 per cent of the entire mass. In thin-section the hornblende appears as large in- ” na * vScansret— | ‘Get aie is not sally psi be tratec ~~ deposition of thé ores were cotempora we should not expect to eary pend the metallie contents, for Me th: within the erevice would “tions can be found, but Gg NOt ke On the basis of teddy ) its growth on the wa — center, then we should ex us, © aro the central portion, with’ jhe ire APomet pot ta Age matter of fact he pears wa im eee aye ‘ Sanates! position, as if _ eavpelled to move in @ ter ms ors Z the anrt of prow th that a Mein “ats sre ae. toey’ Cae PA tog ey aA a are : 408 Tate ey $F as at. eee Fae Bhar 4 , Yg “a follest detail, bat so Tar as Sai te datariinee 1 ‘ A gine, | Paar he ei A ALPE QOLTSCCIOTIR 1a : piwitive altnine the Seg A Oeeure Binge i mY v gia} : 2 the Ne v “ie an ain ane thef 2on of win ich ‘is indioaeae n Tie nian ig a see ya ad by fe i . the plant of pene ‘ ‘ oy 2 Vy Gxt Site eye } } T24) te 4 mw unainie i ‘ as aes | > Mi Prt “rounded by ius ee are bea se yk t's oh eeeeie:, With it atS associated arnall amount OUGErY QUT? Ana ‘Faleera es which perme: ule the crac ka of the rariet mai meee OTS re west wet the pr “ap hery of the individauah ihe alteration may he ao comPplate that no trace of the original gi -t i 4 a } <3 the. ia 5 snby the ene ry al outline. In such « if Hortions are chiefl composed Ot ‘ chlorite, while the outer: in \ \ BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 78 FigurRE 1.—Secrion (87) OF TYPICAL AMPHIBOLITE FROM THE LOWER CONTACY ON THE TUCKER PROPERTY Section shows crushed epidote in a mass of partly altered hornblende Figure 2.—AMPHIBOLITE (SECTION 45) CONTAINING LARGE AMOUNT OF GRANULATED EPIpotTE EMBEDDED IN CRUSHED HORNBLENDE PHOTOMICROGRAPHS OF AMPHIBOLITE VARIATIONS OF THE AMPHIBOLITE 489 are occupied by the quartz and feldspar. In nearly all cases the chlorite belongs to the variety penninite. It should be added that garnet is not confined to the zone indicated on the map. Several sections from points well up the western slope, as well as near the top on the eastern side, show garnet changed to penninite. The amount is very small and probably local in development. Development of anthophyllite——Along the fault line already referred to there are some modifications of the normal amphibolite worthy of de- tailed study. While the fault is sharply defined—about midway of the line as indicated on the map—it can nevertheless be traced a short dis- tance above and below by a breccia zone. The amphibolite along this zone has taken on a schistose and somewhat fibrous texture. The orig- inal hornblende has been crushed and granulated to such an extent that the primary structure has been largely obliterated and the mass now in part altered to chlorite. Kmbedded in the chloritic mass are numerous fibrous individuals, with ragged terminations and fairly well defined prismatic boundaries. No pleochroism is evident. A transverse and longitudinal cleavage is easily detected. The former is the more promi- nent of the two. The relief is quite strong, and hence the interference colors are correspondingly high. The lack of extinction angle, the high interference, and absence of pleochroism places this mineral with the orthorhombic amphiboles. It is regarded as anthophyllite. The sec- tion (45) containing this mineral was found in line with the extension of the fault. Its position is shown on the traverse map. Embedded in the rock are also many grains of garnet, now altered to penninite. It is not improbable that the garnetiferous zone may extend to and beyond this point, and that the anthophyllite is consequent upon the shearing, and hence entirely local in distribution. The remaining amphibolite localities occur at the Tucker quarry and again on the little brook a few hundred yards to the south (see map, figure 1). In both cases the rock belongs to the normal type. In the latter, however, there is abundant evidence of crushing, thus suggesting that a sheared zone or line of faulting might be near at hand. No clean contact could be found. Typical serpentine occurs within a few feet of the amphibolite exposure, but owing to the density of the forest and the heavy covering of loose materials the interval between these two points could not be seen. SERPENTINE CRITERIA FOR DETERMINING ORIGIN In the previous investigations of serpentine and its probable origin it has been found that by far the greater number of cases on record are 440 v. F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN regarded as having been derived from some form of basic eruptive. The interpretation rests on the possible preservation and recognition of rem- nants of mineral constituents belonging to the original mass, or in cases of complete alteration, at least, the retention of some of the characteristic and determinative structures belonging to the essential minerals of the original rock. | MICRO-STRUCTURE OF BELVIDERE SERPENTINE Types.—On the above basis the following types of structure have been established as proof of the kind of rock from which the serpentine was derived : : : (1) Mesh structure, in serpentine derived from olivine. (2) Bastite structure, in serpentine derived from enstatite or bronzite. (3) Lattice structure, in serpentine derived from non-aluminous horn- blende. | . (4) Knitted structure, in serpentine derived from non-aluminous augite. In the Belvidere serpentine, however, serpentization is so far advanced that the recognition of original structures belonging to minerals of the original rock is rendered well-nigh impossible in the majority of cases. There are, however, a few sections which shed some light on this most interesting question. In sections of the normal and fine grained serpentine two distinct types of micro-structure can be seen—one, a very complete fibration of the rock, the other assuming a noticeable lamellar phase or possibly scaly habit. Ifstructure alone is the only essential element necessary upon which to recognize and classify serpentines, the latter should be regarded as lamellar or antigorite serpentine. Fibrous serpentine.-—In the fibrous group at least two distinct arrange- ments of the fibers may be seen, with all gradations between; in one the fibers are arranged in parallel series andin bundles; in the other they show a marked tendency to diverge at the ends or to radiate from centers, and sometimes even to interlace. More than one phase may appear in a single section. According to the interpretations of A. Lacroix and others, the variations in structure may be and are used as a legitimate basis for the establishment of distinct mineral varieties. If these va- rietal forms, however, are crystallographically and chemically one and the same thing, it would seem to be undesirable to give specific names because of variation in form alone. 7 The greater part of the Belvidere area shows a markedly fibrous struct- ure. No particular arrangement is characteristic of any limited part of the area. Coarse and fine textures may be found in the same prospect. BULL. GEOL. SOC. AM. VOL. 16, 1804, PL. 79 FiGurE 1.—SEcTION OF SERPENTINE, U.S Company, Pir NumBer 3 Showing parallel fibrous structure Figure 2.—SERPENTINE AND AsBestos VEIN (SECTION 90) Showing the sharp line of contact and a tendency to shearing at the extremity of the . fiber band PHOTOMICROGRAPHS OF SERPENTINE eo halleN i BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 80 Figure 1.—PHOTOMICROGRAPH OF SECTION 89 Showing the cross-fiber and bifurcation of the amphibolite veins. The relation of the parts is such as to strongly suggest fracturing by shear-pressure Figure 2.—Secrion (40A) oF 1yPICAL COARSE AMPHIBOLITE Showing the curved cleavage partings in a large hornblende PHOTOMICROGRAPHS OF AMPHIBOLITE BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 81 PHOTOMICROGRAPH OF SERPENTINE (SECTION 42) Showing the divergent arrangement of fibrous structure of the serpentine iam ae BELVIDERE SERPENTINE 44] The parallel and radiate arrangement may be found in the same sec- tion. According to Rosenbusch, the parallel arrangement is regarded as the chrysotile variety, while the radiate is characteristic of picrolite and the divergent phase of the metaxite variety. The parallel and diver- gent structure is fairly well shown in sections numbers 7 and 57, from pit number 2, United States company. Section 42 shows a very good case of the divergent and interlaced arrangement, while the lamellar or antigorite variety is well illustrated in sections 43 and 58. Lamellar and bastite serpentine—While most of the sections from the serpentine area show the characteristic fibrous aggregates common to thoroughly serpentinized rock, there are in certain sections some addi- tional features, which are believed to throw some light on the probable origin of the original mass. Some of these features are shown in sections 51 and 69. It isnot uncommon to find a series of parallel lamine mak- ing up an individual crystal. In some cases grains or stringers of mag- netite may be distributed along the partings. In all cases parallel and perpendicular extinction is obtained. These individuals appear like secondary masses derived from an orthorhombic pyroxene, probably some such mineral as enstatite. In other words, the secondary mass is bastite, a pseudomorphous product which is so common in members of the gabbro family. It would thus seem probable that the original rock from which the serpentine was derived was a massive igneous rock in which an orthorhombic pyroxene formed an essential member (see also section 57). DIscUSssION AS TO ORIGIN CONTACT PHENOMENA The locality yielding the most important facts bearing upon the prob- able origin of the rock from which the serpentine was]derived is the con- tact on the Tucker property. The contact is clear and well defined, as shown in the accompanying photograph of a block (plate 76, figure 1), showing both the serpentine and undoubted amphibolite and the marked line between them. As seen in thin-section, it is very clearly defined. Near the line of junction the serpentine is almost isotropic. When ex- amined, however, in parallel polarized light and under a sensitive tint there appear minute fibrous and block-like individuals, which suggest an actual crystalline structure. At the immediate contact the serpentine assumes, under cross-nicols, a deep ultra-blue color. This area includes a part of the absorbed edge of the amphibolite, for the individual com- ponents of the amphibolite may be traced by gradations into the blue areas, where they finally lose their optical characteristics. In other 442 v.F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN words, these facts show that the amphibolite has been absorbed by an intruded igneous rock. The somewhat slaty and baked character of the amphibolite on the immediate contact bears out this conclusion. In sections of the serpentine at a short distance from the contact the fibrous agoregate structure appears again as in the normal type from other parts of the area. It should be stated here that the above contact was at first regarded as a simple fault.* That crushing of the serpentine, as well as the am- phibolite, has taken place is admitted ; but if the contact is of the nature of a fault alone, brecciation should form a prominent factor in the thin- section. This does not appear, but the phenomena of absorption and infusion of the amphibolite are the prominent features. The fracturing of the serpentine and the adjacent hornblendic rock is to be attributed to a subsequent movement, with which was associated the deposition of the asbestos. APPLICATIONS OF EMERSON’S VIEWS An alternative is to be considered in discussing the probable origin of the serpentine. It has been shown by Professor Emerson that ortho- rhombic pyroxenes develop in limestones by processes of metamorphism and also along contacts of igneous or intrusive masses with lime-mag- nesia rocks. The question therefore arises with regard to the Belvidere serpentine. Might not the bastite (enstatite) in this case represent a transi- tion stage in a rock, originally sedimentary, high in lime and magnesia and containing a variable percentage of iron? It has been shown by Professor Emerson f that the above relationships occur in Old Hampshire county, Massachusetts. Moreover, it should be stated that this area lies in the same geological province with the Belvidere area. Emerson also demonstrates that not only the serpentine is derived from the pyroxenic calcareous rocks, but that the amphibole rock—amphibolite—is also a metamorphic product derived from the same series. In other words, both the serpentine and the amphibolite are, according to his interpreta- tion, secondary products resulting from processes of metamorphism of magnesia-bearing limestones, and not of igneous or intrusive masses. The succession of changes from the original sedimentary rock to the final secondary products, as suggested by Emerson, are shown in the following tabulated scheme : * The crushed and broken character of the serpentine observed in the region of the contact on the Tucker property was regarded as sufficient evidence of a true fault, and was so stated in the Preliminary Report of the Vermont Geological Survey, 1903-1904, page 96. Further detailed study of thin-sections makes it reasonably certain that the crushing of the rock was a subsequent phenomenon, which provided favorable conditions for the depositions of the ‘‘cross-fiber.”’ + Monograph xxix, U.S. Geol. Survey, pp. 78-117. Oe en ee Original rock. Ferruginous and Argillaceous limestones. — EMERSON’S SCHEME OF METAMORPHISM First stage. Pyroxenite limestone. Enstatite limestone. Amphibole limestone. Tremolite limestone. Second stage. Sahlite serpen- tine (tremolite). Enstatite serpen- tine (actinolite, bastite, tremo- lite). Amphibolite (epidote). Olivine serpen- tine (tremolite). Third stage. Serpentine (steatite). Serpentine. Serpentine. Serpentine (asbestos). 443 Tale. Tale. Tale. Tale. Tale. At the present stage of investigation of the Belvidere region it can only be said that neither the amphibolite nor the serpentine has been found in association with limestones or highly calcareous rocks of any sort which would in any respect show an analogy with the occurrences in Old Hampshire county, Massachusetts. ANALYSES OF Type Rocks AND FIBER An inspection of the following analyses and comparison with those of chrysotile from the Canadian area bring out some significant facts. Number SiO, Al,0, Fe,0Q; FeO MgO H,O CaO Total ae i 2.23 41.99 14.28 sane | L00:54- ES 40.57 0.90 BSE) 41.50) IS500, se ee/. ° 99.38 1 Ae 40.52 - 2.10 wat Of ~AZ On ABI46e, -2i..; | 10050 re 39.97 Tonk 40.78 12.51 0.50 101.03 BP oic5's 44.10 43.00 12.90 100.00 Numbers 45, 48, and 49 are analyses of chrysotile from Shipton, Quebec, Brough- ton, and Templeton, and are selected from a series in Dana’s Systematic Mineral- ogy, page 673. . Number 1. Belvidere chrysotile (cross-fiber) from the Tucker prospect; analysis by Mr C. H. Jones, Burlington, Vermont. Both the oxides of iron and the alumi- num are included under the 7.27 per cent. Number 2. Theoretical composition of chrysotile, as given in Dana’s Systematic Mineralogy, page 671. In the Belvidere fiber number 1 it will be noted that the silica runs somewhat lower than is indicated in the Canadian cases and considerably lower than the theoretical limit, while the total alumina and iron is, on the other hand, much higher than in numbers 45, 48,and 49. The iron is probably somewhat high, for the reason that the fiber contains micro- scopic grains of magnetite tucked away between the films to such an LX—Butt. Geot. Soc. Am., Vou. 16, 1904 444 v.F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN extent that its removal by the use of a magnet may not have been complete. The mechanical inclusion of the ore would necessarily lessen or lower the percentage of the other constituents. The analyses of the serpentine, in which the asbestos is included, require brief consideration. To this series is also added the theoretical composition of serpentine as given by Professor Kemp in his “ Hand- book of Rocks,” and two analyses of the amphibolite which forms the crest of Belvidere mountain. Number SiO, Al,0, Fe,0; FeO MgO CaO H,O Total : Eiacet ore): AANA ok ache ye dn 4007 12 eee | CTEM ee 40.21 5.73 .... 40.98 0.82 12.68 100.42 igaspheng iy 42.93 32.77 loca B96 ° 14.29. “W276 Weare Doe UY 44.36 98.85 Joc. S74 10490) Wes Number 1. Theoretical composition of serpentine. — p’s Handbook of Rocks, page 140. Number 90. Analysis of serpentine from foot of the serpentine slope, Belvidere mountain. Number 35. Analysis of serpentine from the upper slope and taken from the property of the United States company. Number 5. Analysis of amphibolite from the base of the formation COQ,. Number 23. Analysis of amphibolite from the top of Belvidere present but not determined. It accounts for the low total. Boe ae 40.82 7.63 .... 8840 1.87 12.41 100.63 : An inspection of the analyses of the serpentine, when considered in the light of the theoretical composition, brings to light some variations needing a word of explanation. It will be seen that the silica runs noticeably below the theoretical limit, and in fact below that usually found in many serpentines derived from pyroxene or olivine rocks. This can be accounted for by the fact that the rock almost invariably shows a large amount of iron ore, and probably by far the greater part of the total estimate of the alumina and iron is composed of iron oxide. This, as in the case of the fiber, would lessen the relative amount of silica in the analysis. The presence of a small amount of Sian 2 is to be accounted for as well. It is already a recognized fact that alumina cannot be considered as an introduced product, but, to the contrary, must have formed a part of the original rock from which the serpentine was derived. Now, the facts gathered from the thin-sections suggest as the original a rock containing an orthorhombic pyroxene. The very large amount of magnetite precludes a derivation of the pyroxene by metamorphism. of limestones or highly calcareous deposits, If this interpretation be CONCLUSIONS 445, correct, we are forced to conclude that the original from which the serpentine was derived must have been some member of the gabbro family. The alumina, therefore, could be derived from the presence of an aluminous pyroxene, or more probably from the feldspar. The calcium could likewise be derived from the same sources. The thin sections suggest that a very large part of the calcium is now present as carbonate. It is not, however, universally present in all sections. CoNCLUSIONS It is believed that the petrographical facts, as obtained from the thin sections of the serpentine-amphibolite contact on the Tucker property, are sufficiently evident to render the application of Professor Emerson’s explanation of similar deposits impossible. It must be admitted that a number of the thin-sections from the serpentine area show an apprecia- ble amount of calcite. But an excess of calcium carbonate is by no means a proof of a derivation from a limestone. It is quite as easily accounted for as a secondary product from pyroxenic contents of an eruptive or igneous mass. } It is also significant in this connection to note that petrographic and chemical investigations of serpentine deposits, both in the United States and other countries, have almost invariably proven that such rocks are derived from some member of the basic igneous series. The accompany- ing tabulation of explicit statements, culled from the literature, contains but two cases of a derivation from a pyroxenic limestone out of a total number of twenty-six occurrences. While this statement contains noth- ing of the nature of proof with reference to the area under consideration, it nevertheless reveals the significant fact that the conclusions reached with reference to the Belvidere serpentine are quite in accordance with the predominant views held by the vast majority of petrographers. Concerning the origin of the amphibolites, I can only say that so far as lithological relationships in the field and petrographical characters are known no data have been found which even offers a suggestion as to the nature of the original rock from which the amphibolite was derived. It is not, however, associated with any argillaceous limestone, such as has been recognized by Professor Emerson, in the Old Hampshire county occurrences. Locality. Analysis derived from. Colfax folio, California......... Peridotite. Downieville folio, California ... Peridotite, pyroxenite. Jackson folio, California. ..... Peridotite, gabbro, pyroxenite. Mother Lode folio, California.. Intrusive, original not determinable. Placerville folio, California..... Peridotite, pyroxenite. 446 v.¥F. MARSTERS—ASBESTOS DEPOSITS OF BELVIDERE MOUNTAIN Locality. : Analysis derived from. Sonora folio, California........ Basic gabbro or ultra basic rocks of the peridotite family. Moriah, New Yorke gncci cei. Pyroxeue. Aqueduct, New York.:....:... Montville, New Jersey......... Saxonite. Grenville, California........... Pyroxene rock. Mount Diablo, California. ..... Peridotite. Blandford, Connecticut........ Sahlite in limestone (Monograph 29, U.S. G. 8.) Granville, Connecticut......... Enstatite in limestone (Monograph 29, U.S. G.S.). Kamloops, British Columbia... Basic volcanics. Ishpeming, Michigan.......... Roseburg folio...... BEGt APTS Ay Saxonite and olivine gabbro. Lassen: peak. jp. cicip ds ba-atesie Dunite. Holyoke? folio. occ 'sf-t seo osee Associated with limestones. Mosquito range, Colorado.. ... Pyroxene and amphibole. Rossa, 2320. Beh 5 ae eee Olivine and diallage rock. Southwest Minnesota.......... Saxonite. Roxbury, Vermont. ....:2. 5% < Manhattan island, New York.. Actinolite, tremolite. Rainy Lake region ............ Olivine eruptives. Nevada City and Grass valley, California. Sierra Nevada.) olen yeas oe Gabbro (dike) peridotite. Coast range, California........ Peridotite (see 177a). Marquette region............., Elkhorn district, Montana..... Alteration product from diorite. Baltimore, Maryland.......... Enstatite olivine rocks. Mg | his a BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 447-498, PLS. 82-85 DECEMBER 15, 1905 RED BEDS OF SOUTHWESTERN COLORADO AND THEIR CORRELATION * BY WHITMAN CROSS AND ERNEST HOWE _ (Read before the Society December 29, 1904) CONTENTS Page Part I. The Red beds of southwestern Colorado, by Whitman Cross and 8 ER eC A a neal eg ae feeds 448 EMRE ae tects fC eo digo teu dale aie so oe Niele shoe hea a Pee 448 Eaperature concerning San Juan “ Red beds”... .... 2.2.02... 2.0 ceicccees 449 eee Renn OREN CR UOTUETOMAE, circ ke Sec pe eS esse cage snes wsces 449 Work of the United States Geological Survey ...-.................. 451 Ar TMEMNINLONTERSRY 202 0u ros eds les clk eee eale Sack eee. eee 453 Description of formations.................. ca Rea Sie i Ii pa 459 ERS S SATEEN pita cath Sle Re a a 459 Me NMMRM PUR IMIR Ee Bloor ard gn 2p |. ws, bho Sdea!s od aye w SKE ws sinie ew ad 459 Sees ad Ns MRED EMRE MMR ge als oan 350) = > p Rigen ’= aac oi ow aid d wle vv dla en 461 Perera the) LUGINOIMIG CUSETACLEE 5 5 oe nae cen wes ccuescues =o op eee Typical section........... vil tO oe ONE Pc Ae ea 463 ye a a a }vEURIYdUTY EY} pIBMo, Yoo1D UOAUBD UMOP ZurHoo] MoTA G8 ALVId 10 NOILVNVIdXQY—T GTUASIy a ng 11970 FS, Ci 7 Ye : Ho ly 4 Tite 0 eck eee ee MLS O7OT Pays ——— 7 AYO" eee 456 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO AVYUNO LV ALINYOASNOONNA OISSVINL SHL Z8 "Id ‘POGL ‘9L “IOAN "wv "OOS "103D "11Nd a OURAY UNCONFORMITY 457 This discordance in dip continues through the cliff face almost to the _ Blowout creek, as is clear from plate 83. | In plate 84 is reproduced a view of the unconformity in the cliffs either side of Cascade creek, as seen from the cliffs above Ouray on the west _ side of the river and north of Oak creek. The point of view is but little below the plane of unconformity and the discordant, greater dip of the Carboniferous strata is very evident. In following the unconformity from the Amphitheater northwest it is found that the strata below the Triassic conglomerate belong to the Her- | mosa formation nearly to Cascade creek. Just before reaching that stream a massive pale reddish sandstone appears, the highest fossilifer- ous bed of the Hermosa occurring directly under it. This stratum of sandstone is approximately 100 feet thick, exceptionally massive, and as a similar bed appears likewise on the west side of the Uncompahgre, just above the uppermost fossil-bearing limestone, the base of this red sandstone is assumed as limiting the Hermosa formation. The heavy sandstone just mentioned underlies a wooded bench on the shoulder adjacent to the Blowout, and it is there separated by only a few yards from the Triassic conglomerate above it. Beyond this point, on the east side of the river, the relations are obscured for some distance through the Blowout porphyry intrusion and a great amount of land- slide and other debris, and we now turn to the exposures on the west side of the Uncompahgre as best illustrating the relations of the Tri- assic to the Red beds proper of this district. On the northwest of the town of Ouray rise cliffs even more precipi, tous than those to the east, but not quite so high. The lower half of these cliffs exhibits Hermosa strata, but the upper part contains many sandstones and shales of stronger red tones than can be found in the Cascade Creek section. No unconformity appears in these cliffs as seen from the east. On searching for the Triassic horizon, it is found a little above the main cliffs and below a pronounced bench of glacial molding. This bench is carved at about the Triassic horizon, descending gradually northward with the dip of the strata. Its floor is uneven in detail, more or less debris-covered and timbered, but the white La Plata sandstone outcrops in many places just back of it, and the Triassic conglomerate appears below it, within a few yards in most places. Where this bench begins, on the north side of Oak creek, the Triassic conglomerate occurs - at about 8,900 feet, the same level it occupies adjacent to the Blowout on the east; but fossiliferous strata of the Hermosa do not underlie it here, ~ and in fact 400 to 500 feet of grits, conglomerates, and deep red sandy and micaceous shale intervene before the Hermosa limestones appear. The Hermosa and overlying Red beds in the cliffs dip 15 to 20 degrees 458 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO uoj{s0d oy} Surmoys SUO[}BUIIOJ 10]3nO pus BsoulleH ey} PU’ SeLO[O EY} eed 4eq AzIWIJOJUODUN Jo eustd ey jo qnomolg eu pus ieyeouyydwy OY} WEEM7EQ SYI[O OY} PAvAro} ‘AvINO 4¥ Ao]|¥A OISYBdWOOUL eY4 sso10e8 ButHoo, MatA €8 JIVIg XO NOILVNVIaxq—'Z TANI gq ra=ysow aan AINL NONE NYS P. ~_ YALVIHLIHANY. aH : 3 AVYNO LY ALINYOSNOONN OISSVINL AHL €8 ‘ld ‘vO6L ‘9L “IOAN "WY ‘00S ‘1039 "11NS OURAY UNCONFORMITY 459 almost due north, and it is plain that a monoclinal’fold is necessary to bring the top of the Hermosa to the horizon it occupies beneath the Tri- assic conglomerate east of the river. This fold, of general north-south axis, must have been situated where the present canyon of the river is cut, and hence is not visible. The greater fold, soon to be described, probably obscures this canyon plication in its northward extension. The northward dip of the strata below the Trias is disturbed by a sharp monoclinal fold with northwest-southeast axis and northerly pitch, which crosses the Uncompahgre canyon shortly below Ouray. The visible por- tion of this fold is mainly within the unfossiliferous Red beds, and as the Triassic and overlying strata are not affected by it the relative age of the movement is plain. The fold brings 1,000 feet or more of beds of deep red color to light between the Triassic and the Hermosa and shows _ the magnitude of the unconformity in striking manner. On account of local topographic features the unconformity is less evident above this fold than in the cliffs near the amphitheater. Plate 85 shows the fold on the west side of the river back of the Amer- ican-Nettie mill. The diagram opposite shows where the line of the un- conformity runs, and an observer stationed at about the Triassic horizon on the eastern side of the river, as, for instance, on the bench where the workings of the Bright Diamond mine are situated, can readily perceive that the Dakota and La Plata limestones do not take part in the sharp fold, and at some points the actual unconformity can be clearly seen. From the cliff adjacent to Oak creek to Corbett gulch the Mesozoic strata dip at about 6 degrees. To the north of Corbett gulch (see plate 85) the Paleozoic and Triassic beds come into such close approximation to conformity that careful obser- vation is required to detect discrepancies in strike and dip which actually occur. There are minor, gentle, pre-Triassic folds in the Paleozoic beds, but other similar structures are evident in which the Mesozoic strata take part and the unconformity is thus easily overlooked. At about 7 miles north of Ouray the Triassic and lower Jurassic (La Plata) beds dip beneath the alluvial plain and are not again exposed in the Uncompahgre valley. Very good outcrops of both these formations occur on the eastern bank, - and near where a diabase dike cuts them, but little above the floodplain of the valley, the basal conglomerate of the Trias contains bone fragments and some belodont teeth. This is at about 100 feet below the La Plata sandstone. Their presence at this point was first noted by H.S. Gane in 1895. DESCRIPTION OF FORMATIONS Cutler formation—Selection of name.—The preceding account of the Ouray unconformity will have made plain that no further reason 460 cROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO ~~ FIGuRE 3.—ExXPLANATION OF PLATE 84 View looking across the Uncompahgre canyon at Ouray, toward the. cliffs either side of Cascade creek. ‘The point of view is nearly in the projected plane of the unconformity 16, 1954, PL. 84 VOL. BULL. GEOL. SOC. AM. THE TRIASSIC UNCONFORMITY AT OURAY DESCRIPTION OF FORMATIONS 461 exists for including within the Dolores formation the red sandstones, shales, conglomerates, etcetera, occurring below the well marked hori- zon of the “Saurian conglomerate.” The contingency provided for in the original definition of the Dolores (page 451) has arisen and the subjacent portion of the Red beds, as far as the uppermost Rico or Hermosa horizon, must be distinguished as a separate formation. It seems appropriate to name this stratigraphic unit from the district where its true relations to the Dolores are most clearly exhibited. Since the terms Ouray and Uncompahgre have already been applied to other formations, we have named these lower Red beds after Cutler creek, an eastern tributary of the Uncompahgre, entering it some 4 miles below Ouray, at the old settlement of Portland. The bright red sandstones and shales of the Cutler formation are very prominent beneath the Tri- as3ic cunglomerate on the north side of Cutler creek and the angular unconformity is there very plain, though less pronounced than at the localities described. Near the mouth of the creek several of the heavy conglomerate bands are well exposed. The name Cutler formation has already been used in the Silverton and Needle Mountains folios, now in press. As a groundwork for the remarks on correlation, which appear in the second part of this paper, it seems desirable to review the |characters exhibited by the Cutler, Dolores, La Plata, and McElmo formations, as observed in the San Juan region. The last two do not strictly belong to the ‘‘ Red beds,” but assume reddish colors in some places to be referred to, and in the discussion of correlation it will be necessary to note the probable equivalents of the La Plata and McElmo formations in certain districts. General characterization.—The Cutler formation embraces somewhat more than the lower half of the Red beds section of southwestern Colo- rado. Its strata are variably red in color and include sandstone, arkose grit, conglomerate, shale, and limestone. The maximum observed thick- ness is about 1,600 feet. The formation seems conformable with the underlying Pennsylvanian beds, but above it occurs a stratigraphic break with at least local uncon- formity. The base of the formation is indicated by the Pennsylvanian fossils of the Hermosa or Rico formations and in a broad way by the color line. No fossils have been found in the Cutler beds. Details of lithologic character.—Great variability in lithologic consti- tution, both vertical and lateral, is one of the most striking features of the Cutler formation. The sandstones are sometimes fine grained and massive, but bedding is ordinarily distinct and few homogeneous beds exceed 10 or 15 feet in thickness. All strata are calcareous, and the finer 462 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO Bi *ORTry Diamond mine, on the east side of the river Fiaur8 4,—EXPLaNATION oF PLaTE 85 _ The Triassic unconformity in the cliffs on the west side of the Uncompahgre, above the American-Nettie mill, as seen from near the Bright AVYNO LY ALINYOANOONN OISSVIYL AHL GS “1d “bO6L '9L “TOA “WY “90S *10359 *11Nd DESCRIPTION. OF FORMATIONS 463 grained sandstones grade into calcareous shales and impure mars or into sandy limestones. These rocks are naturally more or less friable and crumbling. The finer grained strata are of the strongest red color, which is due to a ferritic pigment, and they are also commonly characterized by abun- dant bronze or rusty mica, which renders them fissile. Clay beds are rare, aS is massive limestone. Commonly the more calcareous strata are nodular or gnarly and grade into calcareous sandstones. Greenish and grayish tints are locally found in the nodular limestones and a mottling with red iscommon. Some of the nodular limestones appear to be intra- formational conglomerates. | The sandstones frequently grade into arkose grits and these into con- glomerate. With increasing coarseness of grain the red changes to pink, and locally-beds of coarse grit are gray or almost white. In other cases the finer matrix of grits and conglomerates is dark red. The cement of: the strata is calcite, and most of the conglomerate and arkose beds are comparatively resistant to weathering and form prominent ledge outcrops on all steep slopes. The grit beds often reach 35 feet in thickness. They are variably mas- sive, being in some places almost homogeneous from top to bottom, while more frequently divided by several thin shale or sandstone layers. Cross. bedding is almost universal. Sporadic pebbles are present in all grits, - and with their increase the stratum becomes a conglomerate. The sandstones are mainly quartzose, the grits contain much feldspar, mica, and small pebbles like the larger ones of the conglomerates. The latter contain pebbles of granite, gneiss, and various schists, of quartzite and limestone, of greenstone and porphyry, and many of red, pink, smoky, or white quartz, part of which may come from veins. The pebbles are in general larger near the San Juan mountains. Boulders a foot in diameter are occasionally present, but most pebbles are only a few inches in diameter. The relative abundance of different rocks among the pebbles varies according to locality.. The greenstone schists, metadiorite, and green porphyry are most abundant in the Uncom- pahgre valley exposures ; granites and quartzites are always prominent. Taking the formation as a whole, the grits and conglomerates com- prise about one-third or less of its total thickness in the quadrangles sur- veyed, and they are distributed throughout the section. It may be assumed that as distance from the source of the pebbles increases, the formation becomes more and more a series of fine-grained sandstones and shales, with subordinate grits and conglomerates. Typical section.—No opportunity to measure a complete section of the Cutler beds has been found, but the various partial sections exposed in LXII—Buiu. Grou. Soc. Am., Von. 16, 1904 464 cROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO Dolores valley, a few miles below Rico, when combined may serve to: illustrate the character of the formation. This combined section is given: below. The upper 1,118 feet (numbers 32 to 60) of this section was measured by EH. B. Mathews at several places on the northwest side of the river above Priest gulch, while the lower 390 feet (numbers 1 to 31¥ was measured by Whitman Cross on the north side of Scotch creek, an eastern tributary of the Dolores. The hiatus between the sections, if there be any, is small. The Rico beds occur immediately beneath the Scotch Creek section and the Dolores Triassic beds rest on the upBRes most stratum of the exposures near Priest gulch. _ Dolores River section of the Cutler formation Top Feet 60. Coarse sandstone or grit, cross-bedded, locally conglomeratic, rather purplish-in tome. .s.0 60st Se eho See 100. 59. Calcareous sandstones, sandy shales, Beau micaceous and fissile, either red or mottled red and greén in color... -.. J... 1.57.7 oe 120 58. Grit-conglomerate, of variable texture, forming ledge outcrops......-.. 10 57. Fine grained calcareous sandstones, sandy shales, with occasional thin layers of harder sandstones in red or variegated red and green...... : 80 56. Sandy shales, with thin sandstones at intervals; variegated or mottled dark red and light green; have peculiar nodules ............... -.- 55 pe. Grit-conclomiersitie ci mae scons Soou, us bis dolore eras a OED ee oie Ree See eee 20 o4. Calcareous sandstone orsandy shale... 2-2... .0:)0.0.. 5-0-5 gee 100 53. Grit-conglomerate, very similar to number 60....... bn ee) 30 o2. Friable sandstone. r: 2.0. sno 9.. Set a em, Coe ek ale 35 51. Grit-conglomerate, hard and forming a prominent ledge adic Siete 15 50. Sandy shales and crumbling sandstone, strong red color, partly calca- TOOWS «jose: 84 Rb bith tere Be Sloe BRUM hai Fl, oe 110 49. Caleareous shaliesia” <[: snes cont eset ie a PWR Sel Kae Ae oe eee oeets 20 48. Coarse arkose sandstone................ » Weetmaaqeberenee PP 30 47. Calcareous shales......... sj deduo Stghece genie + due he Weheccte iy clare en 30 46. Sandstone variegated.. eo... 0s. od. oe een ins oe ee 20 45. ‘Caleareous shales 2.0.0 se. oe cies wo os eG wo a eee tee 2 15 44. Coarse sandstone; friable, variegated... .0...005. )..0 0.21.5 eee 15 43. Hard sandstone, with few small pebbles of quartzite ......... ........ 15 42. Dark red sandy shales; in places calcareous and then massive; in other parts. micaceous and then fissiles: ui... ci 26 sigs oe ev oe ieee 90 41. Compact arkose sandstone, with few pebbles; cross-bedded; friable sandstones NEAT LOP. 2.0.6. s vee eae ee Ss ee aeae eo ee oe ee ee 15 40. Calcareous clay, or marl, reddish, with spots of various shades......... 33 39. Friable sandstone, largely quartzitic, variegated in color............... 15 38. Conglomerate and arkose grit; most conglomeraticin center; agrit with some pebbles in upper and lower portions; pebbles of greenish slates . and schists, quartzite, granite, and greenish porphyry......... aie 30 37. Calcareous shales, fine-grained, variegated...... ......... a dace oleh eae 30 36. Fine-erained, reddish ,shale. . 3. 0655 6 lin Dae Uo wlelee oe He os See eee 15 DOLORES RIVER SECTION Dolores River section of the Cutler formation—Continued . Massive sandstone, coarse grained, cross-bedded, streaked with light- @norea nyers: some shaly partings. ... 20... ese. ces slelemil oe da slsces . Friable sandstone, purplish and grayish layers alternating cefecalaaly4 . Sandstones, finely laminated, compact, cross-bedded, variegated....... . Calcareous clay shales, graduating into more massive sandy shales and a mnnmae LOMSMENY CMEMATACMNAINEN Oy tha d! Aia es So wink Suds Rees Cale cee aonatone . Sandstone, micaceous, red or green, carrying a thin layer of coarse con- Ppnnmmrncemn cir tories) 20020 ol os leer ee Sas aoe . Arkose sandstone, cross-bedded, mae Tote tee are ae loid Bibel aiige Sak ee atone. MicaCeOUs, TED ANU STEEN ...... 6.26. ccc cer cnsceceomsswen . Arkose sandstone, coarse grained, white................:+ececcencsces . Sandstone, compact, micaceous, salmon color to red..............0200. Peareons Ahales, green and DrOWM 5 oc... 6.02. ere ccc ens she cess ewes . Sandstone, micaceous and compact, containing layers of gnarly lime- Penance tle Dnehes Hiiieke fe). cee hb el atk abla ae debans ve dhe . Calcareous sandstone, generally red, but mottled ; contains thin layers Pa ME WUMERESSEOIO 92202 orto! Shana Ly 4 0d oO cual Sis via fd’ MG martieed Cle aude His) Dae . Arkose sandstone, cross-bedded and conglomeratic in part; this layer is purple at the base, and this color alternates with greenish-yellow and red bands; in the upper part a white and cream color prevails...... 22. Sandstone, rather flaggy, containing a layer of gnarly limestone at the EER RA orate has olin ee ela bal ae ol mss cis SG alee wivin et Aale wee aed ale 21. Calcareous sandstone, shaly in the lower part and containing gnarly faestone near the top: color, bright red. 2... é..cc0 0666846. ds eee ees 20. Sandstone, micaceous, massive, becoming flaggy at tie top, where it Sonics limestone nodules; color, red... .. 2.260. cee es cate ee cease 19. Calcareous sandstone, containing small limestone nodules near the top; EN ee ak Sia Sa cies big cians ws gE Cal d kb meee weet 18. Caicareous sandstone and red arkose, containing limestone pebbles; purple and white in the upper part; red below...............00000- NSS LS RO Oe ee ne Peg eee oe Pea eer eee 16. Arkose sandstone, variegated, red, white, and purple.................. IMnCMIG TAIOACC ONS. TAM ON oo. ae io eos anh oda ca cliedanendaredee 14. Arkose sandstone, flagzy, becoming finer grained and micaceous in the upper part; color, pink, with narrow white bands.................. 13. Arkose sandstone, somewhat conglomeratic in the upper part, and with thin shaly bands near the center and at the top; color, white and I eS Te ES Ge PP oe eee oe 12. Sandstone, micaceous, flaggy, very thin bedded at top; colorred...... 11. Calcareous sandstone; a few thin layers of nodular limestone; color, red. 10. Sandstone, micaceous, flaggy; color, red........... Wein tfa alae! setae oa ie 9. Sandstone, micaceous, flaggy, becoming more compact in upper part... 8. Arkose sandstone, white, banded with red...... ..........ceeeeees eee 7. Calcareous sandstone, rather poorly exposed in the upper part, but flaggy and somewhat shaly, becoming more shaly in the lower part, and con- taining nodules of limestone; color, red...........00.-.cee0e oe pent, Pp oArROse samastane, red above, while below..........ccccccacccccessvece 465 Page 15 20 25 10 18 10 ale ; ' 466 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO Dolores River section of the Cutler formation—Continued Top Feet 5. Shale, probably calcareous; contains nodules of limestone; color, red. . 5 4. Sandstone, micaceous; color, dark purplish red............. ....eeee Det 3. Calcareous sandstone, irregularly nodular and flaggy; contains gray nodules of limestone, but no well defined limestone; color, red...... 25 2. Arkose sandstone, micaceous, thin conglomeratic, cross-bedded; near the base thereis a thin black shale which is quite variable; color, © white, with bluish and’ green Zones, . .c 6. 400 ) dca os 2 15 1. Calcareous sandstone, with a gnarly gray limestone at the top; color, red. 10 Total. eS. b ee 1,508 Age of Cutler beds.—The Cutler beds are manifestly the uppermost portion of the Paleozoic section now preserved in southwestern Colorado. ‘Since they rest on beds which contain a fauna intermediate in character between those of the Pennsylvanian and the so-called Permian of the Mississippi valley, though more closely related to the former, it is natural to refer the Cutler provisionally to the Permian series. If it does not belong to thé Permian it is transitional (Permo-Pennsylvanian) or be- longs to the uppermost Pennsylvanian. Such is the simple reasoning when the relations of the Hermosa and Rico faunas to the eastern Car- boniferous are considered. But the question is not so simple when the relations of the Colorado formations to those of the Plateau province are reviewed. The Hermosa and Aubrey faunas are both regarded as Penn- sylvanian, but Mr Girty informs us that the Hermosa has no species in common with the typical Aubrey of the Grand Canyon section, as far as known. Mr Girty further states that the lower Aubrey fauna from beds at the junction of the Grand and Green rivers, comprising a large part of the Aubrey fauna described by White in Powell’s Uinta report (36), is markedly different from the Aubrey of the Grand canyon, as it also is from the Hermosa fauna of Colorado. These faunal differences must seemingly be explained in one of three ways: (1) By a rapid gradation of forms within a comparatively narrow zone; (2) by the as- sumption of an effective barrier between the Aubrey and Hermosa seas, extending for hundreds of miles from eastern New Mexico west and northwest across New Mexico and Utah; or (8) by assuming one of the formations to be younger than the other, and that the Pennsylvanian sec- . tion is incomplete both in Colorado and in the Plateau country. Studies of the fossiliferous Carboniferous about the La Sal mountains, which are in contemplation, should throw light on this subject, and further discus- sion seems unnecessary at this time. | If the Aubrey and Hermosa are practically equivalent, as the strati- graphic relations suggest, the Cutler beds occupy a position correspond- ing to that of the Permian of the Kanab valley in Utah and the formation DESCRIPTION OF FORMATIONS 467 of the Zuni plateau referred to the Permian by Dutton, as will be brought out in the discussion of correlation. Dolores formation.—General characterization.—The Dolores formation embraces the Triassic portion of the Red beds of southwestern Colorado, - several hundred feet in thickness. It is limited, both above and below, by planes of unconformity. In most places it rests on the Cutler for- mation and is overlain by the La Plata sandstone. It consists of sand- stones, shales, and fine-grained conglomerates, all more or less calcareous. There are two divisions of the formation: the lower embraces variable sandstones, conglomerates, and shales, partly greenish or gray in color and persistently fossiliferous at several horizons; the upper portion isa .very fine and evenly-grained sandstone and shale of strong red color. Lower fossiliferous portion.—The lower portion of the Dolores forma- tion consists of an alternation of reddish sandstones, more or less shaly, with conglomerate, consisting chiefly of very small limestone pebbles. These conglomerates characterize several bands within the lower 300 or 400 feet of the Dolores, and, owing to their fossil content and notable lithologic characteristics, these beds are the most diagnostic of the whole formation. The conglomerate beds are very variable in number. In some places a ledge 20 feet thick may be seen to consist chiefly of con- glomerate with numerous sandy partings and common cross-bedding. A few yards distant the same strata may be composed chiefly of sand- stone with a number of thin layers of conglomerate. Some of the conglomerate bands are more persistent than others, but it is probable that no stratum of the conglomerate is continuous for any great distance. The limestone of the pebbles in the conglomerate is usually very fine grained, and does not appear to have been derived from the older limestones of the Hermosa formation, since no fossil-bearing pebbles have been found. The pebbles are commonly very small, and sometimes they are of such minute, shot-like appearance as to suggest that they are pisolitic. Pebbles several inches in diameter are sometimes found, and these are usually unsymmetrical. It is believed the con- glomerate is, at least partially, derived from the breaking up of limestone deposits of the Dolores sedimentation. The conglomerates are commonly associated with thin bedded gray sandstones or greenish-gray sandy shales. A complex of such alternat- ing strata 50 to 75 feet in thickness can be traced for long distances on the flanks of the San Juan mountains. Carbonaceous material is com- mon in the shaly beds, but determinable leaves have not been observed. The lowest stratum of the Dolores is generally more or less conglom- eratic, and in some localities is a harder stratum, forming distinct ledge outcrops. The thickness of the lower division of the Dolores seems to 468 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO vary considerably, increasing from the mountains west and southwest toward the Plateau country. ' Upper red sandstone member.—The upper member of the Dolores is commonly a very even, fine grained, reddish sandstone, free from con- glomerate, but variably shaly in different places. The shales are, how- ever, always very sandy and there are seldom pronounced division planes of great lateral extent. The bands of parallel, massive sandstone are often 20 feet or more in thickness. The material is mainly quartz sand, with a calcareous cement. In color this upper sandstone is usually bright brick-red or vermilion, shading sometimes into purplish above and a duller, darker red below. In texture this red sandstone is very much like the overlying La Plata sandstone, and where the latter is highly. colored the two formations seem sometimes inseparable, but the La Plata is commonly orange or yellow in color when it departs from the normal gray or white. The upper member of the Dolores is absent in the Ouray and parts of the Telluride quadrangles, but presents an increasing thickness south- ward through the Rico and La Plata quadrangles. Nearly 500 feet of quite uniform sandstones were observed on the eastern flanks of the La Platas. The variation in thickness and the disappearance of the red sandstones to the north is due chiefly to the post-Dolores erosion. Fossils.—The limestone conglomerates of the lower Dolores contain scanty but very widely distributed fragmentsof bones and teeth belonging to vertebrate animals. Much less frequently invertebrate and plant re- mains of identifiable character occur in the same strata. The vertebrate remains have never been found as connected skeletons, or even closely associated bones belonging to oneindividual. They are usually worn and often broken. The most common fossils are the teeth of crocodiles and dinosaurs. Material collected from many localities has been studied by F. A. Lucas, The greater number of the remains belong to the belodont crocodiles, while less numerous are those belonging to a megalosauroid dinosaur, perhaps Palzoctonus of Cope. It is probable that the belodont remains belong to the genus described by Lucas (28) under the name Heterodontosuchus ganei, or to allied forms. Poorly preserved outlines of Unio have been seen in several places and a small gasteropod shell has been found in the Rico and La Plata quadrangles. According to Stanton, the latter belongs to the genus Viviparus or to a closely allied form. The fossils first announced by Hills (20) from the Dolores beds were found in the Telluride quadrangle, on the north side of the San Miguel river, 14 miles below the present site of the town of Telluride. Appar- ently, the locality is unique in the number and comparatively excellent :tate opreservation of the remains found. These fossils were submitted DESCRIPTION OF FORMATIONS 469 to experts for determination, but unfortunately they were lost before being fully identified. Mr Hills regarded the abundant teeth found in one stratum as belonging to a crocodile near Belodon priscus, and certain fish remains as representing a ganoid similar to Catopterus gracilis. A small gasteropod shell and eleven or twelve apparently determinable species of plants were found. The Belodon and Catopterous remains were found in the limestone conglomerate, occurring here about 50 feet below the La Plata sandstone, and the fossil leaves in thin bedded, red- dish, micaceous sandstones not far below the conglomerate. _ La: Plata formation—The strata of assumed Jurassic age occurring below the Dakota Cretaceous in the Elk mountains of central western Colorado were grouped together and called the Gunnison formation by Eldridge in the Anthracite-Crested Butte folio. The formation is de- scribed by Eldridge in these terms: ‘At its base is a heavy white quartzite, 50 to 100 feet thick, usually in a single bed. Above it, in some cases succeeded by other sandstone layers, is a blue lime- stone containing abundant fresh-water shells of the genera Limnea, Valvata, and Cypris. The remainder of the formation consists of gray, drab, pink and purple clays and marls, through which run thin intermittent beds of drab limestone.” _ In the San Juan region and the adjacent country it is desirable to sub- divide the Gunnison, grouping the two lower sandstones with the inter- vening limestone and associated shales as the La Plata formation, and the upper marls, clays, and sandstones as the McElmo formation. ‘This has been done in the Telluride and La Plata folios (8 and 4). The La Plata sandstones are commonly not indurated as in the Elk mountains; instead they are rather friable and crumbling, although of homogeneous texture. Cross-bedding is a marked feature, and not infre- quently a massive ledge as much as 100 feet in thickness has no promi- nent division planes. | Of the two sandstone members the lower is commonly thicker and much more massive than the upper. The latter is in fact occasionally thin bedded and shaly and may be inconspicuous. The calcareous member is very variable in character. On the San Miguel river, in the Telluride quadrangle, it is in some places a pure massive, blue-gray limestone in several beds and with almost no shale. Usually dark calcareous and bituminous shales and thin bedded sand- stones, with more or less of massive limestone, occur between the two main sandstones and sometimes reach a thickness of nearly 100 feet. The total thickness of the La Plata formation varies, in the area we have examined, from about 100 feet in the Ouray and Telluride quad- rangles to 500 or more in the La Plata mountains, and it is known that to the west all members increase still further in thickness. 470 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO = , The sandstones are almost wholly quartzose, and their normal color adjacent to the San Juan mountains is white or gray. But in several localities the lower sandstone is more or less strongly orange or yellow- ish in color, and even red tints have been observed. This coloration is most pronounced to the west of the San Juan, in the Dolores valley. The possibility that the La Plata may be of strong red or orange color is the principal ground for bringing the formation into the present dis- cussion. ‘che | The McElmo formation.—This formation includes the upper part of the Gunnison formation of Eldridge, described by him as consisting of vari- ously colored clays and marls with subordinaate limestone. About the San Juan mountains sandstones or sandy shales become prominent, al- though the marls are everywhere a very characteristic feature. Eldridge pointed out that the Gunnison beds seemed lithologically and strati- graphically equivalent to the Morrison formation of the eastern border of the mountains. The discovery by Riggs (387) of vertebrate remains of Morrison types in the Grand River valley indicates clearly that cer- tain portions of the McKlmo and Morrison are to be correlated, but that the formations are coextensive remains to be proven. The McElmo beds range from 300 to 900 feet in thickness in different parts of the San Juan area. Where thickest, they consist largely of sandstone. The formation is ordinarily not highly colored except in its marl and clay strata, but in certain localities the sandstones are dull red. Parr II. CoRRELATION OF THE FORMATIONS INTRODUCTION The Red beds of the Rocky Mountain province have proved to be perhaps the most difficult portion of the Paleozoic and Mesozoic section. to analyze and reduce to definite groups and formations. In the first place, the red color has been found unreliable as a guide in correlation. It extends at some places into the fossiliferous Pennsylvanian rocks and may penetrate upward into Jurassic beds. Fossils have been discovered in Red bed strata in many different localities—here of Permian, there of Triassic affinities. Unfortunately the two kinds of fossil evidence have seldom been found in one section, affording a means of division. Triassic fossils occur in some places very near the top of the Red beds and in others near their base, in proximity to known Pennsylvanian strata. This ap- parent range of Triassic fossils has been a prominent factor in forming what seems to be the general opinion that the larger part of the Red beds is to be assigned to the Trias. CORRELATION OF FORMATIONS 471 As far as I know, no one has presented evidence of a stratigraphic break within the Red beds or suggested that the conflicting evidence as to age might be explained by such a break. The Ouray unconformity, however, establishes for southwestern Colorado a definite line between Triassic and pre-Triassic portions of the Red beds, which can be traced on a lithologic basis where no visible unconformity exists. From evi- dence which will be presented in following pages it seems certain that the Dolores formation may be positively recognized through its fossils and lithologic character in Utah and Arizona and very possibly in New Mexico. Asastep toward a better understanding of the Red beds, I desire to make certain suggestions as to the correlation of the Cutler and Dolores formations with strata in other parts of Colorado, in Wyoming, and particularly in the Plateau province of Utah, Arizona, and New Mexico. The discussion is based on a study of the literature, on recent published discoveries and on unpublished information. Natu- rally much of the correlation suggested is tentative. Four principal factors have had influence in this tentative correlation : a. The lithologic character of formations compared.—This is the most im- portant on the whole, since it is the only one applicable in all places. b. Fossil evidence—This guide is becoming more and more important as new discoveries aremade. Unfortunately,collectors of fossils,and even paleontologists are not always alive to the importance of careful strati- graphic studies and accurate measurement and description of sections in the localities where fossils are obtained. Their work is therefore often shorn of much of the value that should attach to it. c. The influence of post-Triassic elevation and erosion.—The stratigraphic break at the base of the La Plata sandstone is one of great magnitude throughout the western portion of Colorado. In several places the La Plata is seen in angular unconformity with the entire Paleozoic and older Mesozoic section. It is possible that synchronous elevation, in gentle folds or low domes, occurred in the Plateau district, but have not been recognized. | d. The influence of the pre- Triassic elevation and erosion.—W hile it appears probable that the pre-Dolores uplift was by no means so extensive as the post-Dolores disturbance, it is plain that its influence must be looked for in the Plateau country as well as in the Rocky mountains. RELATIONS OF COLORADO FORMATIONS TO THOSE OF THE PLATEAU PROVINCE The Red beds and adjacent formations have been described in fore- going pages in the development they possess on the eastern side of the zone between the geologic provinces of the Rocky mountains and of the Plateau country. Even in the small area studied in detail these forma- 472 cROSS AND HOWE—RED BEDS OF SOUTHWESTERN. COLORADO tions illustrate the general principle that in the mountain belt of repeated disturbances, of rapid erosion, of land areas, and shore lines, the strati- graphic section will exhibit much greater variation in composition and thickness of its members than in the adjacent region, which has been one prevailingly of deposition. The general conditions under which correlation of the San Juan for- mations with those of the Plateau section must be made are as follows: Adjacent to the mountains there is a broad zone of gentle westward slope in which Cretaceous beds occur. The main streams flowing west and south cut valleys into and in some places through the Cretaceous into. underlying formations. Nearer the canyon of the Colorado the valleys widen and broad platforms and terraces of Jurassic and Triassic beds appear, the Cretaceous being restricted to the divides and isolated mesas. The Paleozoic formations appear at first only in isolated exposures in the deeper canyons, but far to the southwest rise to form the broad plain called the Colorado plateau, on the south side of the Grand canyon. Thus the older the formation the greater are the gaps between districts of good exposures, and the greater the likelihood that in the covered tracts unsuspected complications have entered into the problem. The discussion following will first take up the evidence afforded by recent observations in certain districts and later consider the correlation from a more general standpoint. FORMATIONS OF DOLORES VALLEY AND UNCOMPAHGRE PLATEAU The region in general.—It is natural to begin this discussion of correla- tion with a consideration of the formations occurring in the Uncom- pahgre plateau and the bounding valleys of the Dolores and Grand rivers, a district traversed by A. C. Peale, of the Hayden Survey, in 1875 (35). The heart of the plateau is about 80 miles northwest of Ouray, and Cretaceous formations occupy the greater part of the intervening country. The San Miguel river, however, affords a section extending somewhat below the Dakota sandstone for the entire distance from the Telluride quadrangle to its union with the Dolores river, and the Dolores itself penetrates locally through Jurassic and Triassic into Carboniferous formations. Observations of A. C. Spencer.—In 1899 Dr A. C. Spencer, who had been my assistant in western San Juan work for 3 years, made a reconnais- sance trip to Paradox and Sinbad valleys, which are tributary to the Dolores. His route of travel was from Placerville, on the San Miguel river, following down that stream for some 15 miles, and thence across to the East Paradox valley. This route allowed Spencer to trace the Dakota, McElmo, and La Plata formations from the Telluride quadrangle CORRELATION OF FORMATIONS 473 to the lower Dolores valley, with but two gaps of a few miles each for the two formations below the Dakota. The following general statement of Spencer’s observations is made with his permission. - The La Plata formation was found to increase in the thickness of its sandstone members going down the San Miguel valley, while the calca- reous medial member disappears as such, but seems to be replaced by sandstones of thin bedding. In Paradox valley the La Plata formation has three parts. The upper is 400 to 450 feet in thickness, white in its upper portion, and stained red or orange below. It is mainly massive, but is often friable and crumbling in its lower portion. The middle member is reddish sandstone, shaly at the top with a few chert bands. Thickness, 250 feet. The lower sandstone is from 300 to 350 feet thick, of the typical massive and cross-bedded texture common in this part of the formation. In some places this sandstone is white, but it is more frequently reddish or orange-colored through weathering, while nearly white within. The La Plata thus becomes a formation of about 1,000 feet in thickness, and its massive portions form abrupt cliffs of red, orange, yellow, gray, or white colors in different places. Above the La Plata Spencer found the McElmo formation to be about 500 feet in thickness, with variable relations of sandstone and shale. It has a decided red color locally. Below the La Plata, distinguished by Spencer, there occurs a much deeper red sandstone formation, vermilion or brick-red in hue, and cor- responding to the upper division of the Dolores as it has been described in this paper. Apparently the lower part, characterized near the San Juan mountains by the variable strata carrying fossiliferous limestone conglomerate, is not, on the Dolores, markedly different from the upper in texture orcolor. Spencer found limestone conglomerate rich in bone fragments in several places, and on La Sal creek this horizon was but about 100 feet below the La Plata sandstone. The total thickness of the nearly uniform Dolores formation is about 1,000 feet in the neighbor- hood of Sinbad valley, while much less in other localities. Below the Dolores beds Spencer found coarser Red beds, often con- glomeratic, with pebbles 3 inches or more in diameter, and several hun- dred feet of such strata were noted. No opportunity was found to measure a section showing the full thickness of these coarser Red beds, but, as observed by Peale, they are underlain by fossiliferous Pennsyl- vanian Carboniferous in Sinbad valley, where there is also much struc- tural complexity obscuring the relations. - Spencer’s observations seem to show that the section of the lower Dolores valley embraces strata to be correlated with the Cutler, Dolores, La Plata, and McElmo formations of the San Juan region. 474 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO Comparison with Peale’s section.—The Hayden map of the Uncompahgre Plateau distinguishes ‘“‘ Upper Carboniferous,” Trias, Jura, ‘“‘ Lower Da- kota,’’ and ‘“ Upper Dakota ” as the cartographic units of the sedimentary section. The “‘ Upper Dakota” is clearly the Dakota as commonly defined. The “ Lower Dakota” embraces a group of strata approxi- mately, or perhaps closely, corresponding to the McKlmo, and the “‘ Jura” must include the upper part of the La Plata with, perhaps, the lower part of the McElmo. The “ Trias” is represented by the map as the main formation of the region and as resting throughout the Plateau on granite, gneiss, or schist. Near the Dolores the Carboniferous rocks intervene. According to the descriptions of Peale, this Triassic formation clearly embraces the greater part of the strata referred by Spencer to the La . Plata formation and at least the upper part of the Dolores. His general description is as follows (35, page 80): ‘‘A massive yellow, white, or pink sandstone forms the top of the series. Toward the western part of our district (i. e., the area under discussion) this sandstone is calcareous. In many places the sandstones are markedly cross-stratified. The color is subject to much change locally, passing from white, through orange and pink, into deep red. Below the massive sandstone are blood-red shales, followed in most places by massive brick-red sandstone.” This description seems obviously to apply to a section embracing the lower La Plata and upper Dolores beds. Below the Triassic of Peale occur ‘‘ shales and blood-red sandstones, which on the Dolores, change gradually into gypsiferous shales and sandstones. The latter, I have considered as belonging to the Permian. It is difficult to draw any line between the Trias and Permian, and I have been obliged to do so arbitrarily.” (35, page 80.) The strata referred to by Peale as probably Permian are called ‘‘ Upper Carboniferous” on the published map. It is apparent that the Cutler formation, if present in the lower Dolores valley, must embrace the beds called Permian in Peale’s report. In view of the fact that an examination of the Uncompahgre plateau during the coming season is contemplated, it seems undesirable to sug- gest at the present time any further correlation. It was Peale’s belief that the absence of the Carboniferous and the varying thickness of Mesozoic beds between the gneisses and the Dakota Cretaceous in the Uncompahgre plateau and the adjacent Gunnison valley to the east was evidence of an island never covered by sediments until the Dakota epoch. It seems probable, however, that the erosion of the pre-Dolores interval removed the Paleozoic beds, and that the pre-La Plata erosion in turn removed some, if not all, of the Dolores strata in this district. Peale’s overlap may prove to be wholly explainable in this way. Faulting and CORRELATION OF FORMATIONS 475 folding, presumably of several epochs, undoubtedly adds much to the complexity of the relations. Newberry’s section.—The Red beds of the Grand River valley and can- yon, for some distance below the mouth of the Dolores, may be assumed to correspond in general to the development possessed in the Dolores _ valley. It is only from 25 to 50 miles west from Paradox valley to the region where Newberry made, in 1859, asection in the ‘‘ Canyon Colorado,” which enters Grand river from the east a few miles above its junction with Green river. In Newberry’s often cited ‘‘Generalized section of the valley of the Colorado ” (83, page 99) (in which it should be noted he does not refer to the “‘ Rio Colorado Grande,” as the Colorado river is termed on his map) some 1,726 feet of beds are referred to the Trias. Below this section some fossiliferous Carboniferous (Pennsylvanian) strata and above them Jurassic beds with large saurian bones and fossil wood, which latter undoubtedly belong to the McElmo formation. It seems to me that the three upper divisions of Newberry’s Trias, aggregating 970 feet in thickness, belong to the Jurassic, and possibly represent the three divisions of the La Plata formation, corresponding to those occurring in the Dolores valley. Nothing appears in Newberry’s section which can be definitely correlated with the ‘‘ saurian conglomer- ate” zone of the Dolores formation, but the suspicion may be entertained that the equivalent of the latter is described in number 11 of his section, embracing 92 feet of “ greenish-gray micaceous conglomerate and gray sandstone, separated by red or purple shales.” If this is correct, there are 514 feet of red sandstones below, which may represent the Cutler forma- tien. FORMATIONS OF THE LOWER SAN JUAN VALLEY In general.—From the summits of the La Plata or Rico mountains one may look westward far out over the plain, the monotony of which is relieved here and there by isolated mesas, remnants of older plateaus. In the early morning or near sunset one can often get a glimpse of the precipitous scarps and the gigantic towers and pinnacles of San Juan valley, 100 miles away, which stand out with a rosy gleam or in con- trasting dark shadow, and give a faint suggestion of the wonderland lying just beyond the limit of vision. Going toward this distant valley of remarkable erosional forms one passes over the plain called the Monte- zuma valley, underlain by the Dakota sandstone, and at about 25 miles from the La Plata slopes enters the drainage of McElmo creek. This stream unites with the San Juan river where it turns westward to the Colorado after its long detour in New Mexico and Arizona. The can- yons of the various branches of the McElmo are cut in the Jurassic beds and the name has been given to the upper formation described in pre- Sie. 476 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO ceding pages. Huge fossil bones are commonly reported to occur in these beds and they are supposed to represent the saurian fauna else+ where known in these strata. The San Juan valley was partially explored by Holmes (23) and New- berry (383), but there is little profit in trying to recognize the Dolores and. Cutler formations in the sections they refer more or less provisionally to the Trias. No fossil-bearing horizon corresponding to the Dolores ‘“‘saurian conglomerate ” was observed by these explorers, although both correlate certain beds with the horizonat which Newberry obtained dino- saur remains farther north, and both refer to this bed as “ Triassic (?) ” (23, general section, page 244); (33, page 105). In his generalized sec- tion Newberry himself calls this fossil-bearing horizon “ Jurassic (?),” a reference which seems more reasonable. Observations by H. S. Gane.-—Important evidence as to the formations of the San Juan valley in Utah has been secured by Dr Henry 8. Gane, who in 1897 went from Mancos, Colorado, to the canyon of the Colorado, passing down the San Juan valley on its northern side. Doctor Gane had been my assistant in the Geological Survey during the study of the Telluride and La Plata quadrangles, and was thus well equipped to make correlation studies in the Plateau district. With his permission, I make certain statements of his observations. In the San Juan valley the La Plata formation is stated by Gane to have a general development similar to that which it possesses in the lower Dolores valley, according to Spencer. It thickens greatly and its upper and lower sandstone members are very massive, and exhibit red, orange, and yellow colors, though they are often white or gray. The great cliffs and remarkable towers and pinnacles of the valley are mainly caused by the massive La Plata sandstones, and not by Triassic Red beds, as assumed by Holmes and Newberry. Opposite the ‘“ Water Pocket fold” the canyon of the Colorado is, according to Gane, mainly cut in the T.a Plata beds, 1,000 feet or more deep. That Gane’s correlation is correct, as to the La Plata, is indicated by the presence of the Dolores formation with its fossiliferous limestone- conglomerates below the La Plata. In this conglomerate Gane obtained a large portion of a crocodile jaw, at Clay hill, some 20 miles east of the Colorado and 10 miles north of the San Juan river. This specimen has been described by F. A. Lucas as representing a new form called by him Heterodontosuchus ganei and stated to have marked Triassic affinities (28). It is one of the belodonts, and teeth of these animals are perhaps the most common fossil throughout the limestone conglomerates of the Dolores. Lucas identifies the same species in the Trias of Arizona, as will be brought out later on. \ CORRELATION OF FORMATIONS ATT Limestone conglomerate was found by Gane widely distributed in the San Juan valley, carrying saurian and crocodilean teeth and bone frag- ments, fossil wood, and ill- preserved shells of Unio. He made no meas- urement of the thickness of the Trias and had no opportunity to study the underlying formation. While Gane observed no unconformity between the Dolores and is Plata formations, it is noteworthy that the fossiliferous beds of the Dolores occur near the top of the formation, and that there is here, appar- ently, no massive, vermilion-colored sandstone, like that which is the normal upper member. It is not improbable that this sandstone was largely or wholly eroded before the La Plata deposition. THE ZUNI PLATEAU, NEW MEXICO The development of the Triassic and Jurassic formations to’the south from the Colorado area can not be traced so satisfactorily as to the west in the Dolores and San Juan valleys. For more than 100 miles south from the’ La Plata mountains the Red beds and the overlying Jurassic formations are concealed by Cretaceous strata. They reappear, however, in the Zuni plateau, on the western border of New Mexico, and form the surface of a large part of eastern Arizona, both regions being within the Plateau province. — Newberry traversed this district with the Ives expedition in 1858, and in his report distinguished two groups of strata between the fossiliferous Carboniferous (Pennsylvanian) and the Cretaceous, namely, a lower group variably designated as the “ Red sandstones,” “ Saliferous sand- stones,” or the “Salt group,” and the overlying “variegated marls ” (33). This simple division Newberry applies all the way from the Little Colorado to Santa Fe, remarking repeatedly that he sees no reason for further subdivision. A much more instructive analysis and description of the section of the Zuni plateau was given by Dutton in 1886 (10). He refers the “ varie- gated marls” of Newberry to the “ Jurassic ” and calls them the “ Zuni sandstones.” The ‘ Saliferous sandstones ” of Newberry’s report Dutton divides into three formations. At the base is a formation 450 feet thick, mainly “ sandy shales, containing gypsum and selenite in abundance, with here and there thin bands of limestone.” At some unspecified hori- zon in this formation Dutton found “ several specimens of Bakewellia and an attenuated form of Myalina.” On this ground he correlates these beds with the Permian of the Kanab Canyon district, where Walcott had dis- covered a more extensive fauna. ‘“‘ The Permian beds are distinguished for their dense and highly variegated colors—chocolate, maroon, dark brownish reds alternating with pale, ashy gray, or lavender colors” (10). 478 CROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO The Permian strata thus described are overlain by “‘a very coarse, almost conglomeratic sandstone,” some 50 feet in thickness, which Dut- ton correlates unhesitatingly with the “‘Shinarump conglomerate” (a particular conglomerate within the Shinarump group), referring to the fact that it is persistent and uniform in aspect wherever it appears throughout the plateau country of Utah and Arizona. He does not further describe it in this report. Following Walcott, the conglomerate thus identified as the Shinai in the Zuni plateau is considered by Dutton as the basal stratum of the Trias. Above it occur some 1,600 feet of ‘“‘ sandy shales,” also referred to the “ Lower Trias.” The lower 650 feet of these beds are almost ‘exactly like the Permian below, while the upper portion is of lighter color. Succeeding the “‘ Lower Trias ” shales comes the “‘ Wingate sandstone ” formation, 450 feet in thickness and bright red in color. Dutton de- scribés it as “in reality a group or subgroup of sandstone in which the lines of bedding are generally, but not always, effaced. Sometimes, however, the partings are almost obliterated, so that the edge of the entire subgroup is presented as a single indivisible member. Some- times a portion of the partings is effaced, and a part is so presented. Sometimes partings are seen to divide the whole of it into a series of beds varying in thickness from a yard or two to 20 feet. Most fre- quently there will be at least 250 feet presented without subdivision, as a vertical wall.” Of importance to the present discussion is the apparent absence of cross-bedding. As to correlation, Dutton says: ‘‘This formation is without much doubt the equivalent of the Vermilion Cliff series in southern Utah,’’ where it attains a thickness of more than 1,000 feet. He also remarks that ‘‘ out of it have been carved the most striking and typical features of those marvelous plateau landscapes, which will be subjects of wonder and delight to all coming generations of men. The most superb canyons of the neighboring region, the Canyon de Chelly and the del Muerto, the lofty pinnacles- and towers of the San Juan country, the finest walls in the great upper chasms of the Colorado, are the vertical edges of this red sandstone.”’ Above the Wingate sandstone comes the formation generally charac- terized by Newberry as the “ variegated marls,” but described by Dutton as prevailingly arenaceous and named the “ Zuni sandstones.” This formation is 800 to 1,300 feet in thickness. ‘*It is wonderfully banded and variegated in color. Many pages might be written descriptive of the changes of color which it’ presents, not only as between different beds in the same section, but as between the same beds in different sec- tions.” CORRELATION OF FORMATIONS 479 In some places the bedding is inconspicuous and several hundred feet of sandstone may appear as one unit of massive rock. Cross-bedding is often very pronounced, as in the striking remnant of erosion called the “ Navajo church.” The lower portion is the more massive, the upper part more friable, and gypsiferous shales are common in the latter. Above the Zuni sandstones comes the Cretaceous section, beginning with the undoubted equivalent of the Dakota. Comparing the formations of the Zuni plateau, as described by Dutton, with those of southwestern Colorado, it seems probable that the Zuni sandstones represent the Gunnison group. Dutton’s section is not suffi- cienly detailed to permit a suggestion as to the exact equivalents of the La Plata and McElmo formations, but it is difficult for me to suppose that the Navajo church is constituted of anything but one of the La Plata sandstone members. The Wingate sandstone corresponds in position and character to the upper, vermilion-colored sandstone of the Dolores formation. The ab- sence or subordination of cross-bedding and the constancy of the red color both tend to support such a correlation. Ifthe Wingate is upper Dolores, it would appear probable that the “lower Triassic ” of Dutton is at least approximately the equivalent of the lower Dolores, and it may well be that the basal conglomerate called the Shinarump by Dutton is actually the same as the basal conglomerate of the Dolores. Dutton does not speak of limestone conglomerates in the Trias, nor did he find fossil remains in it, except the abundant fossil wood. If the Wingate sandstone belongs to the Dolores, Dutton was probably wrong in saying that the striking towers and pinnacles of the lower San Juan valley are made of that sandstone, for the observations of Gane indicate that the lower La Plata sandstone is the one producing the prominent erosion forms of the San Juan, being there of pronounced red color. Beneath the Trias in the Zuni district is Dutton’s Permian, lithologic- ally atypical Red bed formation. That would correspond to the Cutler formation in stratigraphic position; but other factors enter into the prob- lem at this point and render any definite suggestion of such a correlation premature. Below the Zuni Permian comes the Aubrey group, and beneath the Cutler occurs the Hermosa formation (ignoring the uncer- tain Rico beds), both Aubrey and Hermosa carrying Pennsylvanian invertebrate faunas. Dr G. H. Girty informs me, however, that these faunas are not known to have a single species in common, and the equiva- lence of the two fossiliferous formations is therefore by no means to be assumed, though their stratigraphic position seems to be the same. It is hoped that data bearing on this question may be obtained in the near LXIII—Buut. Grou. Soc. Am., Von. 16, 1904 480 cROSS AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO future, and further discussion is therefore postponed. Dutton’s descrip- tion of the Permian and Aubrey formations shows that the term Red beds applies to them. NORTHEASTERN ARIZONA From the Zuni plateau a wide tract of upland stretches for 150 miles or more northwest to the brink of the Colorado canyon. On the north- east is the San Juan valley and on the southwest is the Little Colorado. It isthe land of the Moqui and the Navajo Indians. Beneath the Eocene and Cretaceous strata of the higher central plateau appear the Jurassic and Triassic beds, the character of which on the San Juan side has already been considered. ‘That the same systems are represented continuously from the Little Colorado to the Zuni plateau was long ago ascertained by Newberry (82), though definite evidence of the Triassic age of any par- ticular strata has been but recently brought to light. In 1899 Lester F. Ward explored the celebrated “ Petrified Forest ” OF Arizona, which occurs on the northeast side of the Little Colorado, in the Triassic beds (42). The fossil wood collected by Ward has not been described, but in association with the wood he discovered fragmentary vertebrate remains, of which F. A. Lucas announced the general character in a preliminary note (29). Subsequently further collections were made by Barnum Brown for the United States National Museum, and a part of the fauna has been described by Lucas (80). The greater part of the material found by both Ward and Brown is referred by Lucas to two belodont crocodiles, one being the species described by him as Heterodontosuchus ganet, the type of which|was found by Gane in the Dolores beds of the San Juan valley; the other called Episcoposaurus Cope. . Associated with these remains were found bones of three other vertebrates. One of these, a large labyrinthodontamphibian, is described as Metoposaurus fraast, n. sp., the generic identity having been concurred in by Dr E. Fraas, the authority on Triassic vertebrates of Kurope. Another form is described as Placerias hesternus, a Cotylo- gaurian, both genus and species being new. Still other remains are referred by Lucas to the dinosaur Palzoctonus Cope. After referring to the discovery of a similar vertebrate fauna near Lander, Wyoming, which will be discussed on a later page of this paper, I.ucas states: ‘Aside from the interest attached to the finding of this new species is the more important fact, pointed out by Doctor Fraas, that the genus Metoposaurus is char- acteristic of the Keuper of Europe, and that we have in these Triassic beds of Arizona, Utah, and Wyoming the same combination of belodont and labyrintho- dont as in the Keuper’’ (30). CORRELATION OF FORMATIONS 481 It is now manifestly important to determine the stratigraphic relations of the horizon or zone in which this well marked vertebrate fauna occurs, and upon this point Ward (42) made very valuable studies, although it is still difficult to correlate his data with the more general statements of earlier observers concerning other localities. The vertebrate fauna discovered by Ward occurs near the middle of a section some 3,500 feet in thickness, all of which is assigned to-the Trias. Ward divides this section into three parts. At the base are the “Moencopie beds,” 700 feet in thickness, consisting chiefly of dark reddish-brown, soft, laminated argillaceous shales, nearly destitute of silica [quartz], highly charged with salt and gypsum.” Some calcareous beds grade into white impure limestone. No fossils were found in the Moencopie beds, and Ward states that “ the whole series, wherever the contact can be found, always rests in marked unconformity upon the underlying Paleozoic rock (Upper Aubrey).” Upon these facts and the observed transition into overlying Shinarump beds, Ward argues that the Moencopie beds are not upper Paleozoic, as believed by “ certain geologists” not named. This seems to mean that they can not, in his opinion, be considered equivalent to the fossiliferous ““ Permian beds ” found by Walcott in Kanab valley, or to the “ Permian” of the Zuni plateau. } Succeeding the Moencopie come 1,600 feet of variable strata called by Ward the Shinarump. Within this he distinguishes 2 formations, each 800 feet thick, the lower being the ‘“‘ Shinarump conglomerate ” and the upper the “ Le Roux beds.” | The Shinarump conglomerate of Ward is by no means all conglomer- atic. His concise characterization is as follows: “Conglomerates and coarse, cross-bedded sandstones, with clay lenses interstratified with gray argillaceous shales and variegated marls.” In fact, the marls become locally most prominent in zones which are elsewhere strongly conglomeratic. The Le Roux beds are principally variegated marls, argillaceous and calcareous, followed upward by sandstone, limestone, with flint frag~ ments, and at the top more calcareous marls. Fossil wood occurs all through the Shinarump group and none is found beyond it. The petrified forests occur within the Le Roux beds and the vertebrate remains were only found in these strata. Bones and fossil wood were found together in many places. The “Painted Desert beds” of Ward follow the Le Roux beds and consist of sandstones. ‘The lowest stratum, 100 feet thick, is soft, friable, highly argillaceous, and of orange color. Above this come 800 feet of variegated sandstones, regularly stratified and brilliantly colored. Brown, bas EE he: hn Ben 482 cROSS AND HOWE——RED BEDS OF SOUTHWESTERN COLORADO cross-bedded sandstones (200 feet) and white massive sandstones (100 feet) form the top of the section studied. No doubt the Shinarump group of Ward is in large degree equivalent to the Shinarump of Powell, Dutton, and others, as developed in the Plateau country of Utah, and the vertebrate fauna of the Le Roux beds serves to correlate them with the lower Dolores strata containing the same fauna; but, until the unconformity reported by Ward at the base of the Moencopie beds has been traced sufiiciently to demonstrate the importance of the stratigraphic break it indicates, further correlation of the Moencopie strata is difficult. It is reasonable to suppose that the base of the Moencopie is actually the base of the Triassic section, and that deposition began in Arizona much earlier than in southwestern Colorado. On the other hand, it may be that the belodont fauna will be found at horizons lower than the base of the Le Roux beds. If the Moencopie beds are Triassic the stratigraphic break below them accounts for the absence of strata equivalent to the Permian of Kanab valley found by Walcott and to be referred to in the next section of this discussion, The lower 900 feet of the Painted Desert beds of Ward may plausibly be referred to the Vermilion Cliff, or upper Dolores sandstone, while the brown and white sandstones above probably represent a part of the White Cliff, or the La Plata sandstone. THE PLATEAU SECTION OF SOUTHERN UTAH Sections of other investigators.—Having practically traced the Red beds and overlying formations of the Jura from the San Juan mountains to the Grand and Colorado rivers and having tentatively correlated them with formations of the Zuni plateau. and the Little Colorado valley, we will now cross the Grand canyon and compare the original sections of the Plateau country, described by Powell, Gilbert, and Dutton, with those which have been studied more in detail in outlying portions of this great stratigraphic province. Divisions established by Powell_—The formations with which we are now dealing occur in most wonderful development as the steps between the Tertiary high plateaus and the Carboniferous plain through which the Grand canyon has been cut. This succession of terraces and bounding cliffs, often compared to an open book of geological history, exhibit the major divisions of the stratigraphic section in such perfection that Powell naturally drew upon the descriptive terms of this region for the names he was the first to confer. Powell called one great group of strata ‘‘ the Shinarump,” the name meaning “ the weapons of Shinay,” the wolf-god, and referring to the huge petrified tree trunks which are so common in the formation. To the succeeding divisions the terms ‘“‘ Vermilion Cliff” and “‘ White Cliff” were given from the great scarps which face the Grand CORRELATION OF FORMATIONS 483 canyon and can be traced for hundreds of miles with scarcely a break. These terms are also used by Gilbert and Dutton, save that they substi- tute the name Gray Cliff for White Cliff. The reconnaissance explorers of this vast labyrinth of canyon, scarp, and terrace necessarily strove to grasp first the major elements of the problems before them, and it seems clear that in their broader correla- tions they were in the main correct. The great cliff-makers, the Ver- milion and the White Cliff or Gray Cliff sandstones, are traced by Powell to northern Utah and the adjacent portion of Colorado (36), and by Dutton across the Grand canyon to the Echo cliffs above the section examined by Ward. But eastward, toward the San Juan mountains, the change in character of these sandstones caused Dutton to hesitate. He indeed says distinctly that “the Jurassic White sandstone (White Cliff) seems to be peculiar to the northern and western portions of the Plateau province. In southern Colorado and western New Mexico no strati- graphic member has yet been found which can be identified with it ” (8). But this positive statement is at once followed by the suggestion, now known to be correct, that through a thinning and assumption of red color, the Jurassic in this eastern district may be wrongly assigned to the Triassic, together with the also diminished Vermilion cliff (8). It seems peculiar, however, that in the report on the Zuni plateau (10) Dutton did not recognize that the white cross-bedded strata of the Navajo church might be the representative of the White Cliff sandstone. It will be recalled that he did correlate the underlying Wingate sandstone with the Vermilion cliff. The great cliff-forming sandstones of the Plateau province are of sim- ilar texture and not yet greatly indurated. They seem to grade one into the other at many points, and the line between them is generally arbi- trary as drawn by Powell, Dutton, and others. Powell, indeed, includes the White Cliff in the Triassic because of the apparent transition below and the sharp line above, where a limestone bearing a marine Jurassic fauna appears. But this seeming gradation also appears in Colorado, where it is perfectly plain that a part of the Dolores sandstone has been removed, and near localities where the magnitude of the stratigraphic break between these formations is very evident. The soft Dolores sand- stones were no doubt broken up and redeposited, forming a large part of the lowest La Plata beds and causing the apparent transition. j Observations by C. D. Walcott.—The general observations of Powell, Dutton, and Gilbert concerning the Plateau formations were supple- mented by Walcott in 1879 through detailed studies of the excellent section displayed in the Kanab valley, and again in 1882 by special examination of the Paleozoic formations. A portion of the results ob- 484 cCROss AND HOWE—RED BEDS OF SOUTHWESTERN COLORADO tained respecting the Paleozoic section and certain structural features have been published. Of special interest to the present discussion are the discovery of Permian fossils and of an unconformity between the Permian and the Triassic beds. In his published section of the Paleozcic formations Walcott (89) refers to the Permian some 855 feet of reddish brown, chocolate, lavender or drab, gypsiferous and arenaceous shales and marls, with impure fossiliferous limestone. There are two divisions- with erosional unconformities above and below each one. The uncon- formity above the Permianis the most important and has been specially described (40). While this unconformity is not everywhere visible, the Shinarump conglomerate was found abutting against a cliff of Permian beds on the east side of the Kaibab fold near Paria, Utah. To the west the shales above the conglomerate overlap the latter and come in appar- ent conformable relations with the Permian strata. The sections of Mesozoic formations made by Walcott in Kanab valley. have never been published, but through his courtesy I am permitted to give a typical one of the Jurassic and Triassic beds. Section of Jurassic and Triassic formations in the Kanab valley, Utah Made by C. D. Walcott in 1879 Top : Feet Cretaceous strata occur above this section. 1. White friable sandstone, stained yellow locally.................---06- 75 2. Cream-colored arenaceous, friable rock, somewhat gypsiferous.. ...... 275 3. Red arenaceous shales.................. + wiis'a 8) pa Walee ee 6 ene se 150 4, Conglomerate of quartz pebbles and fragments of sandstone and lime- stone with calcareous cement. 2.2. 652.0220. + 4 ae cee ve ee oe 50 5. White massive beds of gypsum with irregular partings of white marl. . 30 6. Red marl and conglomerate, the latter very variable in thickness...... 115 7. Red arenaceous and gypsiferous shale and marl. .............ecesee eee 50 8. Cream-colored magnesian limestone, with many small cavities in upper portion and fossil-bearing at 6 feet below top................-seee0: 25 Fossils: Myalina sp.? Camptonectes bellastriatus, C. extenuatus,? C. stigius, Pecten n. sp., Myophoria ambilineata, Astarle? sp.? Trigonia ? sp. ? Ostrea strigilicula, Solarium? sp.? 9. Sandy shale with few indurated layers; contains many fossils......... 75 10, Taimestonesy 4 chs Favs (este see reec Bebe Sonam mee renews a Sein 10 T1h Sandy shales. v7 22 woes yas B So ace eon, Stararer des lope orders ae Mala Ne 65 12. Buff and cream colored, fine-grained magnesian limestone in layers ’ from +: inéh to 2 feet thick) 320-0 0 ee eed ee ek 40 13. White Cliff sandstone, massive, cross-hedded, light-gray, broken into five principal belts by horizontal lines of bedding...............-.-. 585 14. Vermilion sandstone; cross-bedded, friable, readily disintegrating, form- ing the foothills and slope to the more compact sandstones at the northern end of Vermilion Cliff canyon..).... 56.0226. 2. oc. te eee 650 15. Gray and reddish-brown, cross-bedded sandstone. Horizontal beds of varying thickness divide the mass into bands of from 25 to 100 feet in POICKTCNS : 62 ic ee nate aes ign als LN OMe Atak Ate Crain Perse eae 300 KANAB VALLEY SECTION 485 Section of Jurassic and Triassic formations in the Kanab valley, Utah—Continued Top 16. 17. 18. 21. 22. 24. 25. 26. 28. Evenly bedded red sandstones. Upper portion an indurated, dark red- dish brown stratum. Indurated layers alternate with more friable IEE ENERO TIOTICH GIN 5. osc 6.5 w sae Wier k 0-8 we avin des se eee memo ie Massive gray sandstone, cross-bedded. Upper portion is a light-gray massive friable bed. ‘The entire mass is subdivided into six principal beds by subhorizontal lines of bedding ofa dark, more indurated sand- stone. The beds are from 20 to 80 feet in thickness, and may be seen on many steep escarpments along the canyon...............ee.eeeee Solid, partially cross-bedded sandstone, vs i from gray to various ERMNEERINDEL ey FSIS ys in ene ak ae: Bia ioe fans nya g 6 Fsrcwigye es, one de . Evenly bedded, light-red sandstone with a hin layer of intercalated gray sandstone ..... Fev at epee oleate ep tacnathed LRA Ld ds Oke a Nie 56h . Dark-red sandstone; massive layers alternating with shale, which dis- integrates and forms a sloping talus to the gray sandstone beneath. . IN I CRIEUING gill his eects AE wll kia ald olka bs ay iva emardrins de Bedded sandstone of various panies of redand gray. The ee of sand- stone and their shaly partings are irregular in thickness. Scolithus borings occur in great numbers in a friable yellow sandstone. Frag- ments of vegetable matter and carbonized wood also were seen...... . Thin layers of sandstone, alternating with bands of fine argillaceous eninge mish TEeLM and BCLS . di. cise ete hls cos lle Hawkee ee Massive, light-brown sandstone, broken up into thick layers........... Alternating layers of sandstone and fine argillaceous shales with fish Mo opal Oe anaes ay ened aid eis a eA +, 2 = 5 — ~ : Oy ~ > ~. : as - . ROSH 1h ie a : ee - ; RE = tes Pte pra Ste - era a». : Saat am eer 102 * = eae tet ie ted ee ea ree = SIS 1) We ik i Shh oS Oe = ayy 5 Caste’ ark ils oP cot ae NRG AOD Pasi ato ee) eS “ie ie 5 +e ae i a. v ¥ 2 ea f way TSEC Ficure 2.—Map of Region about the Headwaters of Du Chesne River. Showing approximate boundaries of the formations involved in the fault and flexure of the southern flanks of the western Uintas. 8. The Mesozoic formations of the Triassic, Jurassic, and Cretaceous, of great thickness, follow in apparent conformity. 9. On these formations in turn lie the Tertiary strata in a typical progressive, overlap unconformity, which culminates in conglomerate oe “FAULTING 523 cappings of the upturned, eroded edges of the older formations in many isolated patches.. These relationships may be seen in the accompanying map (figure 2). FAULTING ON THE SOUTHERN FLANK The most persistent and important fault of the south flank of the Uintas, as noted above, crosses the east fork of the Du Chesne at the first western tributary—a small creek locally called Iron creek. The Tron Creek fault strikes east and west, and is strictly, in all this imme- diate district, the dividing line between the great basal quartzite of the Uinta mountains Geaek “ Weber’) and the shales, sandstones, and limestones of the later Paleozoic. The basal quartzite is not in sight below the shales in ascending the stream to the fault zone, then sud- denly it appears as the only rock even to the heights of the adjacent mountain divides. A throw of at least 3,000 feet is therefore measur- able at this point, since the difference in altitude between the gorges and the ridges either side is at least that amount. On the plateau between the Du Chesne and Rock creek, locally called “‘ The Old man,” this dis- placement of the strata presents the phenomenon of bringing two promi- nent quartzite beds together—that is, the basal (first) quartzite north of the fault and the conglomeratic (second) quartzite lying immediately below the gray limestone beds. On this plateau one may easily overlook the break entirely, as the formation seems to be continuous, but in the adjacent gorges the fault shows plainly. It seems possible that this rather unusual juxtaposition* of the quartzites may account for the neglect of 3,000 feet of shaly members of the series lying in this V-shaped area in the river valleys. ~ Toward the east the major fault line cuts higher up in the series, at least above the carboniferous. ‘There are other smaller faults and shat- tered zones along which there has been vertical displacement. One case of faulting noted by Powellt on the north side of the range showed a dis- placement of 20,000 feet. It is estimated that the folding and associated faulting has resulted ina total elevation of these sedimentary rocks, form- ing the central portion of the Uintas, to approximately 30,000 feet above *On page 146 of volume i and page 313 of volume ii, Fortieth Parallel Reports, several fossils gathered from the drab limestone of ‘‘ Rhodes’ spur”’ are noted in the following words: ‘*‘ From the base of the formation, not far above the Weber beds, were obtained Chonetes granulifera, Mar- tinia lineata, Syringopora, multattenuata, Zaphrentis, Lithostrotion, Enomphalus.’’ When it is pointed out that this is the very region where more than 3,000 feet of strata are in sight below the “drab limestone,’ and that their contact with the so-called ‘‘ Weber” in turn is a fault line of unknown displacement the phrase “ not far above the Weber’’ would seem to insist too strongly on the unity of these formations, +J. W. Powell: Types of orographic structure. Am, Jour. Sci., vol. xii, p. 414. U.S, Geologica | and Geographical Survey, vol. vii. 524 c. Pp. BERKEY—STRATIGRAPHY OF THE UINTA MOUNTAINS their former level. Later erosion has removed 15,000 to 20,000 feet of this and redeposited it along the flanks and in adjacent basins forming all the later rocks. UNCONFORMITY An erosion unconformity occurs in the upper car- boniferous beds in this western Uinta region. Evi- dence is not readily detected along the usual lines of travel, but in no less than four somewhat out- of-the-way places the break is well marked. It may be seen in the vicinity of the forks of the Du Chesne river, just above the massive and cherty limestone member, and is succeeded by the last heavy third quartzite formation of the Paleozoic series. The conglomerate was noted on Farm creek, Rock creek, and Rhodes plateau. Where best ex- posed, at the headwaters of Farm creek, the upper margin of the gray limestone is very uneven and carries a great quantity of chert—often more than 50 per cent of the whole rock. In addition to the unevenness of this formation, whose billowy out- line is not followed by the overlying laminations, the two are not everywhere conformable in angle. There is discrepancy in dip and strike of the two beds, the limestone being more steeply inclined and less uniform. The base of the overlying formation, chiefly quartzite, is a true basal conglomerate. There are abundant fragments and pebbles and boulders from the cherty limestone bed immediately below, and in some places the finer cementing or filling matter is calcareous rock flour (calcilutyte) and granular limestone (calcarenyte) and chert (silicarenyte). Fossils are very abundant below the break, but rare above itin this area. From the above it is cer- tain that there is an erosion unconformity in the Upper Carboniferous of the Uintas that marks a moderate readjustment of levels, so that the strata are not perfectly conformable in angle, although the later folding of the range has been so much more profound that this is lost sight of except along the p o- re) 2 4 o 2) ra o "SpqUIQ ULazSaM 2Y7 [0 YUDIWT YINOS 2Y7 {0 U01J99S-sSOU) pPaZnU2eUEH—'e TUOOI Heures guo “AJLUIOJUOOUN PUB 4[NVJ JOIYO 9Y} PUB SUOIBULIOJ JO UOIssoooNs SuLMOYS UNCONFORMITY 525 immediate break. It marks a time interval also of sufficient length for the consolidation of the rock and the development of a cherty facies, thus converting the cherts and other rock of the floor into suitable materials for succeeding accumulations. S. F. Emmons * describes a conglomerate of uncertain relationship in the region just north of La Motte peak, a locality on the north flank of the Uinta range directly opposite that studied by the writer. It was described as apparently conformable to the underlying Upper Carbonif- erous limestone which dips away at an angle of 52 degrees. The occur- rence, as the only one observed in the Upper Coal Measure group, was explained as possibly a relic of the Tertiaries or the Wyoming con- glomerate and therefore belonging to the latest formations of the region. The writer has not seen this case, but he is of the opinion that if the conglomerate is essentially conformable to the limestone and tilted at such an angle, there is much doubt about it being Wyoming conglom- erate for the reason that this rock, being the latest formation of the area, is seldom tilted at all. The discovery, therefore, of a Carboniferous con- glomerate on the southern flank would lead one to expect the same relationship in the La Motte Peak occurrence. DIscussIoNn Only at one point do the pre-Cambrian rocks project up into the great quartzite formation of the Uintas enough to be seen. This was noted by Powellt and by King and S. F. Emmons,{t who used the name “ Red Creek quartzite’ for the underlying formation and regarded it as Algon- kian in age. The contact is considered the mark of a great erosion un- conformity. There are no traces at any point of limestones or shales below the great quartzite similar to the Silurian, Devonian, and Lower Carboniferous for- mations, such as lie belowthe true “‘ Weber” in the typical Wasatch section. If the great quartzite is ‘“‘ Weber,” then clearly there is no pre-Weber in that particular part of the Uintas, and a break more profound than even Major Powell claimed would have to be admitted ; for in that case the 12,000 feet of Cambrian, the 2,000 feet of Silurian, the 1,000 feet of Devo- nian, and the 7,000 feet of Lower Carboniferous strata so well marked in the Wasatch must be considered as wanting in the eastern Uintas. On the other hand if the hiatus is to be regarded as strictly local in the vicinity of this baraboo of Red Creek quartzite, then one is to expect, as King did, that all these enumerated formations do lie buried beneath the great quartzite. * U.S. Geological Exploration of the Fortieth Parallel, vol. ii, p. 524. t+ Geology of the Eastern Uinta Mountains, 1876, p. 145. t U. S. Geological Exploration of the Fortieth Parallel, vol. ii, pp. 198, 268-269. \ 71. = ie ee aS aD , - 526 «cc. P. BERKEY—STRATIGRAPHY OF THE UINTA MOUNTAINS Again, if the Uinta basal member is the true “‘ Weber,” then all known Paleozoic beds have greatly thickened in passing eastward from the Wasatch as a standard. The 6,000-foot Weber has become from 10,000 to 14,000 feet thick. The 2,000 feet of Carboniferous post-Weber shales and limestones have become 5,000 to 6,000 feet of shales, quartzites, con- glomerates, limestones, and sandstones. While if the pre-Weber forma- tions may be assumed to have maintained nearly their normal thickness, and may be counted as present underneath the exposed beds, then we should be confronted with a probable thickness of 20,000 to 25,000 feet of Carboniferous, including one erosion interval in the Uinta mountains— avery extraordinary formation, to say the least. Besides, it would make still more difficult an explanation of the “‘ Red Creek ” baraboo by enor- mously increasing its abruptness and its great height above the old floor, a range of 20,000 to 25,000 feet. It would seem more promising to try the evidence along a different line. Instead of thickening formations, all persisting eastward, may they not rather be thinning and pinching out under ordinary overlap conditions? Such behavior, in view of the comparatively small development of the whole Paleozoic still farther east in Colorado and Wyoming, as low as 1,200 feet, and its entire absence in certain areas farther southeastward, is consistent with much local evidence. It is worth noting in this connection that the pre-Carboniferous strata in central Colorado * amount to only from 300 to 800 feet. The thickest bed of quartzite in the Wasatch is given as 12,000 feet, and belongs clearly to the Cambrian, at least in its upper part. A sim- ilar thickness of the same formation might be expected in the Uintas, and if the great basal quartzite is placed here the comparison is not only satis- factory as to thickness, but also as to general character of rock. Both are dense, hard quartzites, with occasional shale layers—sometimes red in color and often striped, usually showing massive bedding structures. This similarity was noted by Emmons,f who says ‘‘the lower beds of this group (the Uinta quartzites) resemble perhaps the Cambrian rather than the Weber quartzite of the Wasatch.” Later, in a footnote, recognizing the unconformity described by Powell, the Cambrian age of these quartz- ites is further emphasized. In a later articlet Mr Emmons even suggests the Algonkian age of the Uinta sandstones, basing this conclusion upon the investigations of Mr Walcott in the Big Cottonwood region. *U. S. Geological Survey Folio 48, Ten-mile quadrangle, Colorado; U. S. Geological Survey Folio 9, Anthracite-crested Butte quadrangle, Colorado. 7+ U. S. Geological Exploration of the Fortieth Parallel, vol. ii, p. 199. tEmmons: Orographiec movements of the Rocky mountains. Bull. Geol. Soc. Am., vol. i, p. 258. - DISCUSSION 527 Another point in the evidence is the existence of the above-described unconformity in the midst of the Carboniferous. Its presence, whether marking a great interval or a comparatively short one, nevertheless lengthens Carboniferous time by just somuch. The facts as already stated seem to the writer to argue a considerable time break. ‘he effect is to increase the improbability of such immense thickness of Carbon- iferous strata. It is recalled here that Powell considered the break in the eastern Uintas one of considerable time value, and notes that the break increases toward the southeast. There is no evidence either for or against an uncomformity between the basal quartzite and succeeding formations on the south flank in the western Uintas, since the contact there is marked by a fault whose throw of probably several thousand feet brings later shales and quartzites squarely against the basal member. F oy te Emmons’ remarks of the Fortieth Parallel region that both limestones and shales become increasingly silicious toward the east.* This is con- sistent with the idea that a land area lay eastward, and one should expect a rather complete change in the nature of corresponding beds in that direction, with possible thinning and occasional actual breaks in the succession. 3 If the pre-Paleozoic floor may be assumed to rise toward the east and the Paleozoic sea has encroacbed on it from the west or northwest, then we should expect equivalent beds of somewhat unlike lithologic char- acter in widely separated areas. Limestones of the Wasatch might cor- respond to shales in the Uintas, sandstones to conglomerates, shales to sandstones, and in local oscillations some beds might be entirely missing. In the Wasatch the “ Weber quartzite” is both preceded and followed by limestones that are highly fossiliferous and upper Carboniferous in age. In the Uintas there are no Paleozoic sediments found. preceding the basal quartzite, and the description of the later formations varies for different parts of the region. Powell gives 4,000 feet of limestones and shales and sandstones partly fossiliferous, but with the lowest member of his series (the Ladose group) missing in the southeast. Emmons gives gray limestone, calcareous sandstone, and cherty limestone as the general succession. The writer has found in the western Uintas 3,000 feet of shales and sandstones and as much more of quartzites and limestones above the basal member—only the upper members being fossiliferous. The Weber of the Wasatch is the last great quartzite of the Paleozoic series. In the western Uintas there are two strongly developed quartzites * U.S. Geological Exploration of the Fortieth Parallel, vol. ii, p. 199, 528 co. P. BERKEY—STRATIGRAPHY OF THE UINTA MOUNTAINS above the so-called Weber. Barring discrepancies in thickness and noting only the succession, the uppermost one of these would appear to corre- spond fairly well to the true “ Weber.” As the erosion break occurs here at its base, a sufficient cause for its limited development is at hand; and in point of association with the fossil-bearing strata such correlation seems to the writer more satisfactory, since it avoids the introduction of 12,000 to 16,000 feet of unfossiliferous strata in the midst of the Upper Carbon- iferous series. , In the Wasatch the Carboniferous strata, with the exception of the Weber, are very fossiliferous, especially the upper members. In the Uintas there are unfossiliferous beds to the thickness of from 3,000 to 4,000 feet overlying the so-called Weber before coming to the marked fossil-bearing belt. In this connection there is a very suggestive statement made by Emmons in his discussion of the western Uintas. After giving the list of fossils found on Rhodes plateau, in the immediate region under discussion, attention is called to the fact that out of the seven species enumerated two seem to indicate Lower rather than Upper Coal Measure group *—that is, elsewhere they are found only below the Weber. It would then be all the more remarkable to find them here 15,000 feet above their usual horizon. Mr Boutwell + also reports the finding of three lots of fossils from this limestone. The fossils were determined by Doctor Girty to belong to the lower Carboniferous (Mississippian). Although Mr Boutwell does not mention the fault or the intervening beds between the limestones and the quartzite, he says “it would appear that the great sandstone ~ series is earlier than the lower Carboniferous.” The quartzite body has furnished no fossils. Three loose fragments f found by the Fortieth Parallel survey were assumed to come from that formation, but as has been pointed out, there are three other quartzites in the Uintas either of which could as well carry fossils. The writer found a spirifer in a loose fragment of quartzite clearly from the uppermost (third) quartzite above the unconformity. Itis his conviction, after see- ing these different quartzites many times, that residuary fragments of the upper formations might, in spite of all the erosional activity of the region, still lie far within their present limits on the broad anticline of older. rocks, and that there is not necessarily enough lithologic difference in *U. S. Geological Exploration of the Fortieth Parallel, vol. ii, p. 313. + U.S. Geological Survey, Bull. 225, p. 224, 1904. t U. S. Geological Exploration of the Fortieth Parallel, vol. ii, p. 290. U. S. Geological Exploration of the Fortieth Parallel, vol. i, p. 152. 3 ; J -BULL. GEOL. SOC AM. VOL. 16, 1904, PL. 89 Greak Salt Late D Lon Pr a " : Trounsay9 > We ay yen mil Pty ~ =iae “ZL Y Yor SyI0u" ) LIHM & } ADN30V ’ F] 1 ANS3AHING] wa GENERAL MAP OF NORTHEASTERN UTAH FROM WASATCH MOUNTAINS TO GREEN RIVER Showing location of areas involved. of “a s iat re Tag pig te a F ph ee ; Nate? he ~ “engae : 4 =i Maa f | i —_ a 7" strromolano 2 i : il - Jsurstal noieere . ns, fe 1 + — <_ t Veh ee eer nen _—— 4 F ee sate nS ie ery pooner ee A en ae —— — ami =a Picea. —~ " + et nd rat’ = ‘ ea me — seule. ™ ra aaa —. id, fe ~r 7 ee my iP * “ ” a a mag. ee > At cya JS - Svixstrxe wh r : ’ relpaoaH Pihbase dene ol tineog wa) tee we edema: ' |. 16, 1904, PL. 88 BULL. GEOL. SOC. AM. “Pex mo- Gar bonike Syr< Upper Coal Xe asure . ; “3 eve* Quer ‘jper Qubrey wer Qubrey 9 ue eS walt Group oSoce Group ear Ervesi on neon JormiAy inta rmartrion., (of Semele ) Gambdrian x Be wae. + = aller oth Parallel as Powel) The connecting lines are inte ee SUGGESTED CORRELATION 529 small fragments of the different quartzites to satisfactorily trace the origin of a loose piece. SUGGESTED CORRELATION In accord with this view of the stratigraphic relations, the following correlation is suggested : _ The basal quartzite (Cambrian) of the Wasatch is still a basal quartzite in the Uintas. The ‘ Ute limestone” (Silurian) appears eastward as shales—the py- ritiferous, black, and other shales of Iron creek—Iron Creek shales. The “ Ogden quartzite” (Devonian) is represented by a quartzite of precisely the type described in the Wasatch, even to the intermixture of rounded and polished pebbles. _ The “ Wasatch limestone” is much reduced and is represented by heavy limestone at the base, a series of shales in the middle, and cherty limestone followed by an erosion interval at the top. The “ Weber” is also greatly reduced. In part it is represented by the erosion unconformity and its upper portion by 1,000 to 1,500 feet of quartzite. A generalized diagram suggesting this relationship is attempted in plate 88. } _ With this interpretation of relations the apparent discrepancy in strati- graphic position of the unconformities noted in the two districts (Powell’s eastern Uinta and the writer’s western Uinta) is conceivable as one and the same; for it is believed that the basal sandstone rises in the geo- logic|scale eastward, so that its upper margin may not represent the Cam- brian there as it does in the Wasatch, but may be much later. Allow- ing then for a withdrawal of the sea slowly westward to and beyond the area studied by the writer and a more rapid readvance to its former boundaries, it is conceivable that the missing interval in the Green River region may be equivalent to not only the interval itself, 50 miles farther west, but also to some thickness of sediments both above and below. The accompanying chart of geologic sections for the Wasatch, western Uintas, and eastern Uintas, drawn approximately to scale, exhibits more concisely the views outlined in this article. In conclusion, one is obliged to regret that not only in the first at- tempt, but in every subsequent one, made at correlation of these forma- tions, there has been insufficient organic evidence to close the argument from that additional side; but this is sure to be found sooner or later. The loss of a considerable collection of fossils from the formations under discussion in the common accidents and limitations of field transporta- tion is a great handicap to the writer. Without the material as evidence, BULL. GEOL. SOC. 4M. Per rmo- Gar bonierous Sholes Upper CoalIMte asure List BTe v NOK == a\ter Holb Paralteh ) +> " Quexd?*** “Mee* “ ” on” nee oo “ayo? Silurian | Devonian Garbonife yous Western Conglomerate Erosion Interval VOL. 16, 1904, PL. 3g Upper Audrey Lower GQubrey Red walt Group Lofore Group ASS SECO Uncon formidy Uinta Quartzite 12000’ Uinta Region SECTIONS INDICATING THE DEVELOPMENT OF PALEOZOIC STRATA IN NEVADA AND THE WASATCH UPLIFT Wimio 740007 f Yormartion (of Power) Pre Gambrian Eastern Wintas (alter Powell) The connecting lines are intended to represent the correlation of formations presented in this paper for the western Uintas, and suggest the possible eastward-lying equivalents and overlap conditions 530 co. P. BERKEY—STRATIGRAPHY OF THE UINTA MOUNTAINS it is useless to take up that line of argument. The writer feels justified on the other grounds to present the above tabulated correlation as ex- hibiting the best explanation of the known facts of stratigraphy in the western Uinta mountains. It is to be regretted that the use of the term Uinta for Tertiary beds in the same region makes its adoption objec- tionable. 7 The existence of two well marked beds of quartzite above the great basal quartzite member, the barrenness in fossils of the first 3,000 feet above it, the existence of an erosion interval and unconformity in the Carboniferous itself, and, after making allowance for the hiatus, the close correspondence lithologically between the Uinta strata and those of the Wasatch from bottom to top, with no radical departure from the succes- sion of the formations, all point to the same conclusion. Atleast no dis- cussion of Uinta stratigraphy can afford to neglect the above structural features. The fact that some of them have been overlooked or under- valued is the writer’s excuse for raising so large a question upon exam- ination of so very limited field. The basal quartzite of the western Uintas is surely not “ Weber.” Itis apparently late Cambrian and possibly in part post-Cambrian lapping up against the margins of the Paleozoic continent toward the east. It should have been known by the name that Powell gave it—that is, the “‘ Uinta formation” or Uinta quartzite—in preference to any other. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 16, PP. 531-636, PLS. 90-94 FEBRUARY 10, 1906 PROCEEDINGS OF THE SEVENTEENTH ANNUAL MERTING, HELD AT PHILADELPHIA, PENNSYLVANIA, DECEMBER 29, 30, AND 31, 1994, INCLUDING PROCEEDINGS OF THE SIXTH ANNUAL MEETING OF THE CORDILLERAN SECTION, HELD AT BERKELEY, CALIFORNIA, DECEMBER 30 AND 381, 1904 Herman Le Roy Farrcuixp, Secretary CONTENTS Page eat auatireday., December 29... i. oo bias wine slet ed os le due cee serdnoee 532 ERE SRE RRIAETN Goo Sooo xc open Wve ies yeh Od ow Sa alee aay Oa a8 pes ear 532 EE AMEN a I cok ace iS akisid ote oI wily Cqal'slape Ob.g m adteld A olay 533 NUMRECS BEEN Nee ics find ain 'p a plate bused i= bse a a> Amb) sid saw ope aable s 535 ERNE. nr as sg nadie wih ein be av eee eg ath t Bam ats Sd hegs's 538 ESTE 72 Fy SNR pee Oe a BO a en es ee 539 I HETRR hy ei mar Te. an sliw igs od pee os Cue oa a pla 'nielere Sta oe aoe 540 CI III SS a aris ics C2 tly clay Sis cium did nil oh Supe lb, cia’e va oles eles 540 Memoir of Charles Emerson Beecher [with bibliography]; by Charles MN rig atte thE cath etl ao cls aRilaas ciusligh sé can wie Sid enti wie wiehiw Sule 541 Memoir of John B. Hatcher [with bibliography]; by W. B. Scott....... 548 Memoir of Henry McCalley [with bibliography]; by Eugene A. Smith.. 555 Memoir of William Henry Pettee; by Israel C. Russell..... Ch ine ge oF 558 Memoir of Charles Schaeffer; by Angelo Heilprin....... .............. 561 Development and morphology of Fenestella [abstract]; by Edgar R. ESE iM ieee he at series Sede Mees wisi he’s v + vin Soa bias vl datSleed CPO e wes 562 Origin of the caves of Put-in-bay, lake Erie [abstract]; by E. H. Kraus. 563 Zuni salt lake [abstract]; by N. H. Darton..............0..ecececee oe 564 Experimental investigation of the compressibility and iste deformation of certain rocks [abstract]; by Frank D. Adams and E. G. Coker...... 564 Session of Thursday evening, December 29..... RSet Ey k fores Megha obese apr eeee iy 565 NN aNEE METI SIOCOTI ET D> oi. sa nls one pds lols nie dare vis's cp aueuies ty ations as 566 Auditing committee’s report.................. Pee ert tr tab ee ts aa tarde wet 566 Present condition of Mont Pelé [abstract]; by Edmund Otis Hovey..... 566 Soufriére of Saint Lucia [abstract]; by Edmund Otis Hovey............ 569 Boiling lake of Dominica [abstract]; by Edmund Otis Hovey..... -.... 570 Nitrogen gas well at Dexter, Kansas [abstract]; by Erasmus Haworth... 572 SSID MIRMENS Sh ais LIS coca wale oo we ou aie a ena v'e nla ous emd y's 573 Occurrence and distribution of celestite-bearing rocks [abstract]; by RRNA EE CERO i 5s wc hieiaiein Wena wo ein cepa wind vate biceecinee dectenaceds 574 Suggestion as to origin of riebeckite rocks; by G. M. Murgoci........... 575 IIIT SPPPECIAINED oy cic ais a ores ons vs cle cnn scuccesevced dees mevns 576 New York drumlins [abstract]; by H. L. Fairchild..................... 576 LXVIII—Butt. Geox. Soc. Am., Vou. 16, 1904 (531) 532 PROCEEDINGS OF THE PHILADELPHIA MEETING Page Drumlins in the Grand Traverse region of Michigan [abstract] ; by Frank Leverett. «00 25 sino cece ees deeb sa mele Gels a oe ee ee 577 Drumlin areas in northern Michigan [abstract]; by Israel C. Russell.... 577 Classification of the upper Cretaceous formations of New Jersey [abstract] ; by Stuart Wellef =. 00-200. 0.22 cd es co meets oe eee 579 Fauna of the Cliffwood clays [abstract]; by Stuart Weller.............. 580 | Pre-Cambrian rocks in the vicinity of lake Temaskaming, Ontario [ab- stract]; by Willett G. Miller. ... 2.0... ccees cso sen eee eee 581 Relative ages of the Oneida and Shawangunk conglomerates [abstract] ; by A.W. Grabau. . o.oo 2. ccs cb eae ee c pee oes eal 582 Notes on the Ontaric, or Siluric, section of eastern New York [abstract]; by C. Ay Hartnagel. oo. bo. 2 vcs cae s cae eee - eee 582 Rocks of Mount Desert island, Maine bared by Persifor Frazer... 583 Determination of bruciteas a rock constituent [abstract]; by A. A. J Sica 586 Shifting of the continental divide at Butte, Montana [abstract]; by W. H. Weed on oes civic ates sas wsiels dees Sate Oo) yele sate ge 587 Relation of lake Whittlesey to the ces beaches [abstract]; by Frank BS NOR Soon an a 6 nwa ss wlelb oo eG oe nes saleee'e Gk eee ea 587 The loess and associated interglacial deposits [abstract]; by B. Shimek.. 589 - Fifteenth annual report of the Committee on Photographs ............. 590 Resolution of thanks. :. 1.1.5. .%..25 5.5 k Ses oss ae es aoe 250 Register of the Philadelphia meeting, 1904............. cece ee cee cee cecees 590 Session of the Cordilleran Section., Friday, December 30, 1904 ............. 592 A detail of the great fault zone of the Sierra Nevada [abstract] ; by J. A. RED. Ae Lee SR Oe. Sa ae a oe oe 593 Register of the meeting of the Cordilleran section: ........... ao: sen eee 594 Accessions to the Library from July, 1904, to July, 1905; by H. P. Cushing, PACT OPIN So sic 2s POP. Sa SBS AE ee Sia ale, be dato Its Oe te eee 595 Officers and Fellows of the Geological Society of America........ .....--4-- 605 Index -to volume: IG 2. 23... 0005 ee ewes tee oan cee ee seen at ace 617 SESSION OF THURSDAY, DECEMBER 29 The Society was called to order by the President, John C. Branner, at 9.30 o’clock a m,in room 116, the Geological Museum and Lecture room, of College Hall, University of Pennsylvania, where all of the sessions of the meeting were held except the evening session of this day. The report of the Council was called for and was presented by the Secretary, in print, as follows: REPORT OF THE COUNCIL To the Geological Society of America, in Seventeenth Angie Meeting Assembled : The Council has held only one meeting during the year, in conjunc- tion with the meeting of the Society at Saint Louis. Some business has been transacted by correspondence. SECRETARY’S REPORT 533 The rules relating to publications have been modified, the most im- portant change being reduction of the price of Bulletin brochures. The reduction amounts to 20 per cent of the former prices to Fellows and 40 per cent of the prices to the public. The amended rules and the Constitution and By-Laws with changes to date are published as the closing pages of Bulletin, volume 15. In all respects the affairs of the Society continue in a highly satis- factory condition, as shown in the following reports of officers, which give the details of administration for the sixteenth year in the life of the Society. 7 SECRETARY’S REPORT To the Council of the Geological Society of America : Meetings.—The record of the Saint Louis meeting will be found in the clusing brochure of Bulletin, volume 15. With the change by the American Association for the Advancement of Science from summer to winter meetings, it becomes desirable to re- move our consitutional requirement which compels summer meetings in conjunction with the Association. i Membership.—Since the last printing of the List of Fellows the names of four Fellows have been taken from the list by death—C. E. Beecher, J.B. Hatcher, Henry McCalley, and W. H. Pettee. The names of eleven new Fellows have been added to the list and one removed by resigna- tion. This makes the present enrollment 259, or six more than at the last printing. Fifteen nominations are now before the Society, and sev- eral candidates are awaiting action by the Council. Distribution of Bulletin. —At this date 448 pages of volume 15 of the Bulletin have been distributed, and the remaining brochures are in hand, awaiting the completion of the Proceedings brochure for final mailing. The irregular distribution of the Bulletin during the past year has been as follows: Complete volumes sold to the public, 15; sold to Fellows, 2. Brochures sent to supply deficiencies, 32; sold to the public, 10; sold to Fellows, 15; sent in exchange for other brochures, 32. One copy of volume 14 has been donated and three copies bound for use of the officers and the Library. The sale during the past year of back volumes to the Fellows has been very small. Asa complete set of the published volumes, including the 10-volume index, now costs the Fellows $64.75, newly elected Fellows can not be expected to purchase them. Only one complete set has been sold the past year to libraries. Future demands for sets will be largely supplied by sets offered for sale from libraries of deceased Fellows. Our future sale of the Bulletin will be chiefly of current volumes. 584 PROCEEDINGS OF THE PHILADELPHIA Bulletin sales.—Receipts from sale of the Bulletin during the past year: appear in the following table: MEETING Receipts from Sale of Bulletin, December 1, 1903, to December 1, 1904 Grand total Total. $1 60 | $11 10 40 14 90 45 10 45 yaaa 5 00 20 5:20 ieee 5 00 A By 9 25 30 5 30 2 65 7 65 60 5 60 70 5a 5 25.) 36025 40 185 40- pert 25 00 $26 39 | $575 39 oh ges 295 $26 39 | $577 89 $577 89 7,577 08 $8,154 97 D2 aa ‘$8,187 29 Complete volumes. Brochures. Public. | Fellows.| Total. Public. | Fellows. Volume 1. $5 00 $4 50 $9 50 Oo) GOs a as Volume 2. 10 00 4 50 TA GO ae $0 40 Volume 3. TO OOF e oo). 10 OO had 25a 45 Volume 4.. 5 00 BOO" HL Se. a eee Volume 5.. HOW eek os ace DOO Ait ete oak 20 Volume 6.. OO ae Mace B00 i Oe ee ee ee Volume 7.. Fy SOD eee ae 5 00 3 00 125 Volume 8.. 5 U0 dice Spied os tae 30 Volume 9.. Oil ese are SUOOM it: bas ke 2 69 Volume 10.. HOM eo See 5 00 604s ee Volumell.. OY Pan ak es 5 00 GO ek eee Veolia rive 12s coy a ee OS eee ae REVO Te CLARE se dace che ee Volume 13.. SOO r i ate 5 00 2 80 6 79 Volume 14.. OBOE A ee oes 260 00 3 90 Pigs Volume 15.. ASOD 4 eet oats 185 00 AQ 4 aoe eee Volume 16.. DOs Ae Sater, op Dy OO Hie 2.28 G alee ocean $540 00 $9 00 | $549 00 $13 00 | $13 39 Index! 2.20: BBO W os2 Sse: . Be oe eo! We et Gee $542 50 $9 00 | $551 50 $13 00 | $13 39 Beceigis for the fiscal yeahs: Oe 4 se es oe wee Se een ee Previous receipts, to November SO ISOS se. hrs verse Be ees ‘Total'feceipts: te date: .o...veuk ce eee te cee eee Charged: and uncollected 22f 5.005 ea ee a Total Bulletin. sales to date 2c. 8 es eee eee Bills have not been sent to regular subscribers for volume 15. Exchanges.—The exchange list includes one more than last year, and the Committee on Exchanges will recommend a few additions and omissions. Expenses.—The following table gives the cost of administration and Bulletin distribution from the Secretary’s office during the past year: EXPENDITURE OF SECRETARY'S OFFICE DURING THE FISCAL YEAR ENDING NOVEMBER 30, 1904 Account of Administration Postage-and :telesramise. i043 F Arete Oe ae ee eee = TOx pressae 35 ceisae (isc Saws Oe desig tole appr gavin es beptee Wb aa a eects Printing (including stationery) Meetings (not includeéd/ip printing): 72. 24/60, ie ah de ee sere wee wee eer ewe eee evr eres ees eee er eee A — we | hea tabinstete ee ee a a : F TREASURER’S REPORT 585 Account of Bulletin EPA SS ache he inde bled den d Wil a'e wv dale yw blew Wd wlalaig me's $126 50 SS GS ee a ee See ae Lahisiti es Oe Wrapping material ferivalopes, TR ets shai ee = Gee ena ee ee 1 05 TE oe! SSS ap oe parame om ee ctu ers. 1 94 Binding three copies TENET TO MR IN ener cee 3 00 Purchase of brochures to fill deficiencies...... ea Rh PORE? 5 3 00 IDEN Ue es ea oe wh bs len acewane sabe bis 4 09 ER COs OMIA. Oe wm is Viv cee ald pw dk Cvs emia Ay coe bn Polen $193 02 rE MRO FF TA YORE Eg Lio cise 5 )c9 Siva tm yididsaWins das! 12 ob Abbie $353 76 Respectfully submitted. H. L. Farrcnixp, Rocuester, N. Y., December 10, 1904. Secretary. TREASURER’S REPORT To the Council of the Geological Society of America: The Treasurer herewith submits his annual report for the fiscal year ending December 1, 1904. Ten (10) Fellows remain delinquent for two years, while eee (26) are delinquent for this year. Five (5) Fellows—G. D. Harris, R. R. Hice, R.S. Tarr, E. O. Ulrich and F. E. Wright—have enrolled for life by the payment of the one- hundred-dollar fee, thus increasing the total number of Life Commuta- tions to sixty-eight (68). The $1,000 bond of Tioga township, Neosha county, Kansas, was called for redemption, without any knowledge of the Treasurer, in 1903, and hence, when the interest coupon due February 1, 1904, was forwarded to New York for collection, the fiscal agency of the state of Kansas de- clined to pay the interest and replied that the bond would be paid on presentation. As there was no provision for redemption on the face of the bond prior to maturity in 1916, and as Treasurer Williams (H. S.) had purchased the same at a considerable premium, in the belief that it could not be redeemed before the date mentioned on the face of the bond, the Treasurer declined to accept the terms offered. The attorney for Tioga township claimed the legal right to redeem the bond under a general statute of the state providing for such redemptions before ma- turity, and after the Treasurer had contested his views in a lengthy cor- respondence (which is submitted to Council herewith), a compromise was effected, in which the township paid the deferred interest coupon and redeemed the bond at $1,081.80 on April 1, 1904. The terms of this compromise were submitted to Dr H. S. Williams, First Vice- President of the Society, the acting President during the absence of Doctor 5386 PROCEEDINGS OF THE PHILADELPHIA MEETING Branner in Europe, and he fully sustained the views of the Treasurer in advising the acceptance of the same rather than a resort to the courts, in which the costs, even if successful, would have much exceeded the amount involved. With the receipts from this redemption and several fees for life com- mutations the Treasurer had more funds than seemed advisable to retain ‘in bank at only 4 percent interest; so, in concurrence with Doctor Wal- cott and Doctor Emmons, the two other members of the Committee on Investments, the Treasurer purchased on April 11, 1904, three (3) $1,000 second mortgage 5 per cent bonds of the United States Steel Corporation at 78% net, or $2,366.25, which yield an annual interest return of $150 in two payments (May 1 and November 1), or a little more than 62 per cent ontheamount invested. These bonds are now (December 20, 1904) quoted at 92%, and if the Council does not approve of the investment in this class of securities, which yield the higher rate of interest, the bonds can be sold at a profit to the Society and the proceeds reinvested in any other securities which Council may prefer. At the time of purchase these were the only bonds available which appeared to be reasonably safe and would yield 6 per cent interest on the investment. — The interest item from all sources ($576.98) has thus been consider- ably increased over that ($328) for last year by this increase in the in- vested funds from $6,300 to $8,300. On Account of Publication Fund The securities now owned by the Society (all of which are deposited in the fire and burglar proof vaults of the Bank of the Mone Valley at Morgantown, West Virginia) are as follows: March 17 and 25, 1898, two Texas Pacific Railroad first mortgage 5 per cent bonds, cost $1, 976. 25 BAG og Bde ah) ctiz ye eho Pauate! wee Whee ore apace ee $2,000 February 6, 1901, 10 shares of the capital stock of the Iowa Apartment House Company; Washington, (D.C. 6 wk cei 8 2 8 eee ean 1,000 April 1, 1903, 20 shares of the capital stock of the Ontario Apartment House Company, cost $2,000..4 . oso le ecb. te ee De 2,000 May 5 and September 27, 1895, 3 first mortgage 6 per cent bonds of the Kingwood, Tunnelton and Fairchance railroad, cost'$304.....<5ae eee 300 April 11, 1904, 3 second mortgage 5 per cent bonds, United States Steel Cor- poration, cost $2 0) sy 4) eg EIR rr CREA REM OR i cS 3,000 Total cost, $7,646.50; total par value....... eer se sc. $8,300 The Texas Pacific and United States Steel bonds are quoted on the New York Exchange at 119} and 924, respectively, on December 20, 1904. The general financial condition of the Society, as shown by the receipts and disbursements for the past year, is exhibited in the following tabular statement. Pitta ant 537 RECEIPTS. Balance in treasury December 1, 1903........ Fellowship fees 1901 (1).. ate farce LO OO MOR at .cestenat an vs ee O0 = = ol DORE (24)... scorer UL ee SED Janes vo xtetelre dk yO70s OO « SOD iticats sacs. 0 00 Initiation fees (11).......... ee . at iC eC ik bn CRC fa. lade comitiutations (5). c. 660... 5... esas © Interest on investments: = Iowa Apartment House Co. stock. $60 00 4 Tunnelton, Kingwood and Fair- Pe chance Railroad bonds ...... 18 00 ae Tioga Township, Kansas, bonds.. 35 00 x Ontario Apartment House Co. a Bit cree cosas eal Svcs on Hive 120 00 =] Texas and Pacific R. R. bonds.... 100 00 4 U. S. Steel Corporation bonds . 150 00 = Interest on deposita with Security a EFOSEACOMIDAMY. «0/55 0240 2 is 93 98 Redemption of bonds: Tioga Township, Kansas, bond and pre- mium, “a ee RIGS Of MU GNCHIOUM,ncubap pares ves ap ness eas oseeaeeeee Statement of Receipts and Expenditures Total receipts brought forward.. .............. $7,778 46 een oe mo alate EXPENDITURES. ; Secretary’s office: ACUMINISETALIOM oss. Oe vs. eae 6 $160 74 Bulletin st Mite eiae wae i 193 02 Allowance (traveling and clerical expenses)....... fiat kere se e OUO OD a ODD 10 $1,900 00 Treasurer’s office : 110 00 ROB Orh wa claw ee a, ee oivee cplo-O0 500 00 Expressage. ee Rare a 1 88 a 16 88 ibranian’s Ofice vay ccss.cs essa ss cote Pat inane 7 66 PhiotOpra pire aCeOUNtr., cosa as ees oes ke es 15 00 Publication of Bulletin: PPOVINEIN Seon aac tari ain cee 2c ey test 4 $1,851 79 Engraving fasyvnna Malad Van Hes Sao a4 Editor’s allowance, » personal and . Office GX PENSeS. oes Sos in os wis 250 00 — 2,545 23 TAVORTRION TAs: So manta nese aces ee Cian ieee ye OU sen 576 98 Total expenditures to December 1, 1904........ 5,804 78 Balance in treasury December 1, 1904............ $1,973 68 1,081 80 577 89 —_—— 4,746 67 Total amount of receipts (carried forward)........ $7,778 46 Respectfully submitted. Morgantown, W. Va., December 22, 1904. I. C. Warts, Treasurer. 538 PROCEEDINGS OF THE PHILADELPHIA MEETING Epitor’s Report To the Council of the Geological Society of America: The Editor is particularly glad to be able to state that through the cooperation of the members it has been possible to complete volume 15 before the winter meeting of the Society. This desirable result can always be accomplished if contributors will be prompt in forwarding their papers and ‘in returning proof. Though volume 15 is unusually large and surpasses its predecessor in size, it is less voluminously illustrated. It contains 636 pages of text and has 59 half-tone plates and 16 line drawings. There has been a noticeable tendency to increase the size and the illustrative material of later volumes. The average number of pages of the first ten volumes was 544 and of plates 26, while volumes 11 to 15, inclusive, have an average of 603 pages and 57 plates. There has been in consequence an increase in cost of publication, but the average per page has been very sight. The cost of volume 15 is given below and a tabulated comparison made with the average cost of the first ten volumes and the individual cost of volumes 11, 12, 18, and 14. Average. ; Vols. 1210.| Vol 1. Vol. 12. Vol. 13. Vol. 14. | Vol. 15. —_———————— | | CN Ce ___ | pp. 544. pp. 651. pp. 538. pp. 583. pp. 609. | pp. 636. ; pls. 26. pls. 58. pls. 45. pls. 58. pls. 65. pls 59. Letter-press.....2000 cscs es $1,465 14 | $1,815 56 | $1,445 73 | $1,647 12 | $1,657 50 | $1,661 21 Stra GIO MIS) se sueessnesece costes terse: 200 40 373 68 414 80 ATT 27 431 21 457 76 $1,860 53 | $2,124 39 | $2,088 71 $2,118 97 AVELAZe Per PAZe.........-sereeeee $3 23 $3 36 | $3 45 $3 64 $3 43 $3 33 The following is a reasonably correct analysis of the contents of vol- umes 7 to 15, inclusive: Vol. 7, Vol. 8, Vol. 9, Vol. 10, Vol.11, Vol. 12, Vol. 18, Vol. 14, Vol. 15, Divisions. Pages. Pages. Pages. Pages. Pages. Pages. Pages. Pages. Pages. Areal geology........... 38 34 2 35 69° 199 12s 48 115 Dynamic geology...... - 3 24 85 24 110 23 17 47 2 Economic geology....... 4 14 16 28 7 5 4 1 3 Glacial geology.......... 5 98); 188 96 21 55 1 48 48 Historical Oso sees s<'.he me ae 5 16 46 By 24 1 52 Weemoirs dc.) see cme ats 28 8 12 27 60 2 32 14 18 Official matters: *2.2.-4 .h 56 69 54 72 59 58 153 68 63 Paleontology... .... wating: 123 58 64 68 188 5 42 22 1 Petiology Hee re: ain ae 40 43 44 59 54 24 28 8=6183 rs Pysiographic geology.... 53 5 AF 37 10 53 24 59 54 Geology and pedagogy... 12 a me sig Re a cn es oe Rock decomposition..... 74 26 17 9 bs 16 a8 he 5 Stratigraphic geology.... 21 67 28 62 31 98: 116 TS. 266 Terminology. }.. sce... 3% i =e ay 1 aca There 5 bo a ere ee EDITOR’S REPORT 539 Long experience has convinced the Council that the prices at which the Society’s publications have been offered to its Fellows and the public have been too high, and the rates for the brochures of volume 15 have been reduced to cost plus 100 per cent to members and cost plus 200 per cent to the public. The prices are given in each volume in the pre- liminary matter. In volume 15 they are on pages vii and viii. Owing to the fact that summer meetings have been abandoned prac- tically, volume 15 begins with the presidential address instead of the proceedings of the summer meeting, as heretofore. Respectfully submitted. JOSEPH STANLEY-BRown, New York, December 10, 1904. Editor. LIBRARIAN’S REPORT To the Council of the Geological Society of America : The accessions to the library for the past year have been duly cata- logued and acknowledged as received, and the list of accessions up to July 1 has been made out and transmitted for publication in the Bul- letin. The library now comprises some 2,600 numbers, of which 1,400 are bound volumes. Of these 1,200 belong to the continuing sets of publi- cations received as exchanges, the remainder being scattering and indi- vidual publications. ‘Some 200 of these volumes were received bound, while the remainder have been bound by the Case Library authorities. Pamphlets and individual separates make up the remaining numbers, although there are quite a number of maps. None of these have been bound ; yet they should be the better to preserve them, and their proper disposal in this respect is a perplexing problem, concerning which the Librarian would be glad to receive suggestions. This is especially the case with the maps. Cleveland is such a smoky city that it is by no means an ideal place in which to keep even bound books, and unbound materials tend to rapid deterioration. The expenses of this office for the past year are as follows: PERT CRITE: fs iccern Se Pek Sens eas Saath Pern ye $1 50 SOME EARAER IRE nc Febe Ee IY oe wa oats bed, ) scald idle ave 8 aoe oy woes Aen ees 1 16 Seen MME htt) Fates kee use oa be Pov auto luk ae 6 5 00 $7 66 Respectfully submitted. H. P. CusHine, CLEVELAND, OHI0, December 10, 1904. Labrarian. On motion of the Secretary, it was voted to defer consideration of the Council report until the following day. LXIX—Buut. Gro. Soc. Am., Vor. 16, 1904 540 PROCEEDINGS OF THE PHILADELPHIA MEETING As the Auditing Committee to examine the accounts of the Treasurer, the Society elected J. S. Diller and E. O. Hovey. ELECTION OF OFFICERS The result of the balloting for officers for 1905, as canvassed by the Council, was announced by the President, and the officers were declared elected as follows: : President : RAPHAEL PUMPELLY, Dublin, N. H. First Vice-President : SAMUEL CALVIN, lowa City, Iowa. Second Vice-President : W. M. Davis, Cambridge, Mass. Secretary : H. L. Farrcuiip, Rochester, N. Y. Treasurer : I. C. Wuitr, Morgantown, W. Va. Editor : ; J. STANLEY-Brown, Cold Spring Harbor, Long Island. Inbrarian: H. P. Cusnine, Cleveland, Ohio. Councillors : H. M. Ami, Ottawa, Canada. J. F. Kemp, New York city. ELECTION OF FELLOWS The Secretary announced that the candidates for fellowship had re- ceived a nearly unanimous vote of the ballots sent, and that Fellows were elected as follows: RaLpPu ARNOLD, Ph. D., Washington, D.C. Geologic Aid, U.S. Geological Survey. JoHn ADAMS Bownocker, D. Sc., Columbus, Ohio. Professor of Inorganic Geology, Ohio State University. REGINALD Water Brock, M. A., Ottawa, Canada. Geologist, Geological Survey Department; Professor of Geology, School of Mining, Kingston. Nevin MenancrHon FenneMANn, Ph. D., Madison, Wis. Professor of Geology, University of Wisconsin. CHARLES Newton Gou.p, B.S., A. M., Norman, Okla. Professor of Geology, Uni- versity of Oklahoma. Mark S.W. Jerrrerson, A. M., Ypsilanti, Mich. Professor of Geography, Michigan State Normal College. OO ee a eS ee ae ae oe mth 82 wecaagapsich isenanis! Se ee ee ee FP ee 7, ih hed ; D borhan eta ee A i : 3 : Toh ae eh ta) " ib eed hae a f Vi as * P § f ¥ 7 vs ~ . d h 4 Z * ; iJ , La. , a oe Sets! se va * pied . oe 2 P Mod ae ery , Pile te 4 “ae ata : ‘| Fad Cees Se iF ae : a 7 vs ek oY ea roa } wad: a . ’ ; ¥. } ' LY nAS 2 r ; * I - 1 ‘ \ hee pe , Beatin ? ’ a i" S ‘ ) d r Pat tate : ‘ ; } : sees t a vy “ : n . j Fj . ‘se - } ‘ * LT 4 “ i 7 vey . : © f \! 1 , % ae . , > ae q * a * * § ~ oy + + hee. mt ar? mat Aas te Se l in ss Fine -_ ' ews 7 j " 1 ‘ it ‘ " : , Y ) s ‘ , a0 Le r = ‘ BULL. GEOL. SOC. AM. VOL. 16, 1904, PL 90 MEMOIR OF CHARLES EMERSON BEECHER 541 Hiram Dryer McOasxey, B. S., Manilla, P. I. Chief of the Mining Bureau of Manilla. BengamMin Le Roy Mixusr, Ph. D., Bryn Mawr, Pa. Associate in Geology, Bryn Mawr College. Henry Montcomery, Ph. D., Toronto, Canada. Professor of Geology and Biology _ in Trinity University. CiropHas CIsNEY O’ Harra, Ph. D., Rapid City, S. Dak. Professor of Mineralogy and Geology, South Dakots School of Mines. Abert Homer Purpusg, B. A., Fayetteville, Ark. Professor of Geology, University of Arkansas. ArTHUR Epmunp Seaman, B. S., Houghton, Miss. Professor of Mineralogy and Geology, Michigan College of Mines. SoLon SHEepv, A. B., Pullman, Wash. Professor of Geology and Mineralogy, Washington Agricultural College. Bouumit Suimek, C. E., M.S.,lowa City, lowa. Professor of Physiological Botany, Iowa State University. . GILBERT VAN INGEN, Princeton, N. J. Curator of Invertebrate Paleontology and Assistant in Geology, Princeton University. No new business was presented. The President called for the necrology, and the following memoirs of deceased Fellows were presented. In the absence of the author, the first memoir was eae by H. E. Gregory: MEMOIR OF CHARLES EMERSON BEECHER* BY CHARLES SCHUCHERT One of America’s leading paleontologists, and a Fellow of this Society since 1889, in the fullness of intellectual power, suddenly passed away on February 14,1904. Few men were better prepared for great results and more promising of them for the next twenty years than Charles E. Beecher. Dall has said: ‘“There is no doubt that in the death of Professor Beecher not only has Yale sustained a serious loss and paleontology a severe blow, but the ranks of those capable of bringing to the study of fossils keen insight and a philosophical spirit of enquiry, guided by principles whose value can hardly be exaggerated, are dimin- ished by one whom science could ill afford to lose.” Like most successful students of organic life, Beecher was a born natu- ralist. As a boy of twelve years he began to make a collection of recent shells and fossils, continuing to add to this for the next thirty years ; so that, in 1899, he was able to present to Yale University, “‘ uncondition- *Sketches of Beecher have appeared as follows: Yale Alumni Weekly, March 2, 1904, by Bush, Chittenden, Schuchert, and ‘‘a graduate student;’’ Science, March 18, 1904, by Dall; Amer. Natu- ralist, June, 1904, by Jackson; Amer. Geologist, July, 1904, by Clarke; Museums Jour., London, April, 1904; Geol. Mag., London, June, 1904, by Woodward. 542 PROCEEDINGS OF THE PHILADELPHIA MEETING ally,” upward of 100,000 fossils. To the Albany Museum he gave his entire collection of land and fresh-water shells, some 40,000 specimens. In the field few excelled Beecher as a collector. To him more than to any other we owe the present methods of washing clay for immature invertebrates as well as of etching silicious fossils from limestone. The Yale collections are rich in such delicate and well-preserved material. Clarke, who often collected with him, stated that “‘ he was the most dis- criminating acquirer of the unusual, the exceptional, and the fine that it was my fortune to know.” As a paleontologist he was trained in stratigraphy and in the descrip- tion of species and genera, but latterly he took almost no direct interest in this kind of work. Often he told me that he wished all our fossils were named. This is all the more remarkable because of his long association with Hall and Marsh. The explanation seems to lie in the fact that his philosophic bent did not come to full fruition until he had personally met the philosophic American paleontologist, Alpheus Hyatt. From that time his mind was absorbed in working out the ontogenetic stages in fossil species and in tracing their genetic sequence through the geolog- ical formations. To Beecher we owe the first natural classification of the Brachiopoda and the Trilobita, based on the law of recapitulation and on chronogenesis. He also gavea very philosophic account as to the origin and significance of spines in plants and animals. On these works his reputation in days to come will chiefly rest. | Beecher was not only a born naturalist, but also had much mechanical ability. Nothing pleased him more than to free fossils from the sur- rounding matrix, and his unexcelled talent in this direction is shown in the preparations of Triarthrus and Trinucleus in the Yale University museum. More than 500 specimens have been prepared by him, and this work has required peculiar skill, patience, ingenuity, and a great deal of time. It is very unfortunate that he did not live to finish his studies on the trilobites, but he left all the better specimens completely worked out, and of most of them he had made photographs and drawings. Charles Emerson Beecher, son of Moses and Emily D. Beecher, was born in Dunkirk, New York, October 9, 1856. Not long after this date his parents removed to Warren, Pennsylvania, where he prepared for college at the high school, and was graduated from the University of Michigan, receiving the degree of B.S. in 1878. The ten succeeding years he served as an assistant to Professor James Hall. In 1888 he was invited by Professor Marsh to remove to New Haven and to take charge of the collections of invertebrate fossils in the Peabody Museum. His career as a teacher of geology began in 1891, when for two years he took charge | a i ’ MEMOIR OF CHARLES EMERSON BEECHER 543 of Dana’s classes at Yale, and in 1892 he was made Assistant Professor of Historical Geology in the Sheffield Scientific School, serving in this capacity until 1897, when he became Professor of Historical Geology and a member of the governing board in the Sheffield Scientific School. In 1899 he succeeded the late Professor Marsh as curator of the geological collections, and was made a member of and secretary to the board of trustees of the museum. In 1902 his title was changed to that of Univer- sity Professor of Paleontology. He was eminently successful as a teacher, both with undergraduates and with advanced students, his enthusiasm and kindliness of character at once arousing their interest and devotion. Beecher received the degree of Ph. D. from Yale in 1889, his thesis be- ing a memoir on the Ordovician Brachiospongide. In 1899 he was elected a member of the National Academy of Sciences and a foreign correspond- ent of the Geological Society of London. In 1900 he was elected Presi- dent of the Connecticut Academy of Arts and Sciences, and filled this office for two years. He was also a member of the American Association of Conchologists, Geological Society of Washington, Boston Society of Natural History, and Malacological Society of London. Some time before Beecher was graduated from the University of Mich- igan, the desire of his youth to follow as his life’s work the study of fossils becameaconviction. The year before his graduation he is seen worship- ping at the shrine at Albany, where many another paleontologist had preceded him on the same errand. Clarke describes Beecher’s introduc- tion at Albany in the following interesting way : ‘*On a hot summer day in 1877, pale with weariness, he staggered with pack on back into the laboratory of Professor James Hall at Albany. He had sought what to him had seemed the fountainhead of knowledge of his fossils. It had been the goal of many a youthful dream to show to the author of the Paleontology of New York the treasures he had found. The great and keen-eyed Hall ever had an appreciative reception for such endeavor. With the most friendly concern he refreshed and nursed this acolyte, and, when strength had returned, expressed a lively interest in his efforts and his ambitions. On going away Beecher had promised to come back to Albany when his college course was done and join Hall’s corps of workers on paleontology. So, in the summer of 1878, the year of his grad- uation, he became assistant to Professor Hall, entered upon his work, and was received with genuine enthusiasm.’’ Beginning with the summer of 1880 and contimuing into 1883, he read, according to a list still extant, more than 18,000 pages of standard liter- ature. During the 10 years with Hall he assisted very largely in the preparation of the Paleontology of New York, treating of the Lamelli- branchiata, Gasteropoda, Cephalopoda, and Bryozoa; and to a less extent on the volumes pertaining to the Pteropoda and corals. These were great days of preparation and they bore most valuable fruit later on. 544 PROCEEDINGS OF THE PHILADELPHIA MEETING As to his methods of investigation, Clarke says: ‘‘A part of Mr Beecher’s fine natural equipment for scientific research was his indomitable patience necessary to establish broad premises. His conclusions were never hasty nor ever stated on merely one aspect of the evidence. All the more far-reaching and striking of his deductions in his later work, when his mind had turned chiefly to problems of biogenesis, are known to his friends to be the result of tireless acquisitions of material and the focussing of light from every source. In some quarters, his methods. unknown, their results were not accepted; they were regarded as startling, as iconoclastic, and even unreliable.” During his bachelor days at New Haven he lived in “the atic, a. series of rooms fitted up in Bohemian style in old Sheffield Hall, with Penfield, Pirsson, and Wells, all of whom are now full professors. After ce the day’s work the “attic philosophers” met here in delightful inter- course, social and scientific, and it was here that during the late ’80’s and early ’90’s many pleasant acquaintances and recollections were ac- quired with the young scientific men of this and other countries. Beecher’s first paleontologic paper was published by the Geological Survey of Pennsylvania in 1884, when he was 28 years old. It treated of new genera and species of Phyllocarida from the Devonian, a group of rare Crustacea, most of which he found about his home. He was always on the lookout for these rare fossils, and after securing many hun- dred additional specimens he again returned to the subject, and in 1902, in a paper published by the Geological Society of London, embodied all that is known of the Upper Devonian Phyllocarida of Pennsylvania. Beecher’s first turn from stratigraphic paleontology to pure paleo- biology and correlation had its origin in the brachiopods. Hall had as- sembled some tons of the Silurian fossils occurring at Waldron, Indiana. This collection contained many slabs, and as much loose clay adhered to them, Beecher and Clarke night after night for an entire winter washed this material ; eventually they together obtained about 50,000 specimens of young brachiopods, among which were included every stage of develop- ment of these shells. Their results were published in 1889 in a well- illustrated paper entitled ‘“‘ Development of some Silurian Brachiopods.”’ From a study of the nature of the pedicle opening, these authors con- cluded that the “ phylogenetic development tended in two main chan- nels,” and this arrangement foreshadowed two orders of brachiopods for which Beecher later proposed the names Neotremata and Telotremata. My acquaintance with Beecher began in 1889, and at that time it was evident that the paper just referred to was being considered with a better understanding of what Hyatt’s principles meant when applied to Brachi- opoda. The very fact that nearly all the Waldron, Indiana, brachiopods began with smooth shells having a subcircular outline led him to look : ' & MEMOIR OK CHARLES EMERSON BEECHER 545 for this early stage in other genera, but as no other young shells were at hand, he resorted to a study of the beaks in well-preserved examples of mature shells. In the spring of 1891 he announced that he had seen the initial shell in 15 families representing 40 genera. A study of the stages of growth in many brachiopods, from the Cam- brian to the living forms, enabled Beecher to show that the old classifica- . tions were not expressive of genetic relationship. He demonstrated that on the basis of types of pedicle openings all brachiopods are naturally grouped into four orders, of which two are without, and two possess hinge teeth. The most primitive order (Lingula, etcetera) he named Atremata, and this gave rise directly to the Telotremata (Rhynchonella, Terebratula, etcetera). The Neotremata (Crania, Discina, etcetera) also originated in the Atremata, and from the former descended the Protremata (Stro- _phomena, Productus, etcetera). In 18938 there was discovered in the Utica formation near Rome, New York, a thin band in which nearly all the trilobites occur as pseudo- morphs in iron pyrite and retain antenne and legs. Trilobites with legs had been known before in two specimens and in four genera. Walcott determined the presence of legs by slicing enrolled individuals. Antenne, however, had not been clearly made out until 1893, when their presence was announced by Matthew in the August number of the American Journal of Science. This discovery was of great value and promised much toward a better understanding of the ventral anatomy of trilobites and their systematic position among the Crustacea. Beecher was thus led to visit the locality in 1893, when he took out several tons of shale; since then he has published fifteen papers on trilobites. Of these, three are devoted to the larval stages, seven to the ventral anatomy, and five to the classification and systematic position of these forms. Beecher showed that in Triarthrus the entire series of thoracic legs are biramous, one being set-bearing and used for swimming and the other without setz and used for crawling. The limbs of the pygidium overlap each other, are much crowded,and are adapted for swimming or guiding the animal, although they may also have served as ege-carriers. ‘The head has five pairs of appendages, four pairs of which are biramous and closely resemble the thoracic legs. He also observed that in the first or unsegmented stage of the most primitive trilobites there are neither dorsal free cheeks nor eyes, but that in some of the later forms both the eyes and free cheeks have migrated to the anterior margin or may even have progressed a little posteriorly down the dorsal side of the first or unsegmented stage. This led him to undertake a study of all trilobite genera, more than two hundred in number, and it was seen that these could be arranged in three groups or 546 PROCEEDINGS OF THE PHILADELPHIA MEETING orders on the basis of the nature and position of the free cheeks. These orders he named Hypoparia, Opisthoparia, and Proparia. In 1892 he became greatly interested in the significance of spines, ac- cumulating data until 1898, when he presented his studies in a paper entitled “ The origin and significance of spines.” This paper he regarded as his best and most philosophic work. He found that all kinds of spines in plants and animals can be arranged into eleven distinct cate- gories. Further, that two generalizations result, as follows: ‘‘That spinosity represents the limit of morphological variation, and, second, that it indicates the decline or paracme of vitality.” . . . ‘‘ Finally it is evi- dent that, after attaining the limit of spine differentiation, spinose organisms leave no descendants, and also that out of spinose types no new types are developed.” Beecher’s standing among biologists and paleontologists was high ; he was a leader among students of Brachiopoda and Trilebita, and Jack- son has said that he “ became the leader of the Hyatt school.” He had the artist’s gift, nearly all the drawings illustrating his various papers being made by himself and exhibiting a high order of merit. He was a slow and very careful worker. Those who knew him well saw in him an enthusiast, but his exuberance was always held in check by his judi- cial qualities, which also made him an excellent counselor. He was orderly in his work, and, as he had the “ museum instinct” well devel- oped, ne made one of the best of curators. In 1894 Beecher married Mary Salome Galligan, of Warren Pennsyl-— vania, who, with two daughters, survives him. He died very andere: of angina pectoris, at his home, shortly after 1 o’clock on Sunday after- noon, February 14, 1904. Up to about 11 o’clock of the same day, he was in his usual health. He lies in Grove Street-cemetery, in the shadow of the Sheffield Scientific School. BIBLIOGRAPHY The following bibliography includes Beecher’s more important papers :* 1876. List of land and fresh-water shells found within a circuit of 4 miles about Ann Arbor, Mich. [Walker and Beecher.] Proc. Ann Arbor Sci. Assoc., pp. 43-46. 1884. Ceratiocaride from the Chemung and Waverly groups of Pennsylvania. Second Geol. Survey of Pennsylvania Report, pp. 1-22, pls. 1, il. _ 1884. Some abnormal and pathologic forms of fresh-water shells from the vicinity _ of Albany, N. Y. Thirty-sixth Ann. Report N. Y. State Mus. Nat. Hist., pp. 51-55, pls. i, ii. 1886. A spiral bivalve shell from the Waverly group of Pennsylvania. Thirty- ninth Ann. Report N. Y. State Mus. Nat. Hist., pp. 161-164, pl. xii. * For a complete bibliography see Jackson, in the American Naturalist, June, 1904, pages 418-426, where 108 titles are cited. 1888. 1889, 1889. 1890. 1890. 1891. 1891. 1891. 1892. 1895. 1898. 1893. 18953. 1893. 1894. 1894. 1895. 1898. 1899. 1896. 1896. 1897. 1897. 1897. 1898. BIBLIOGRAPHY OF CHARLES EMERSON BEECHER Bae Method of preparing for microscopical study the radule of small species of Gasteropoda. Jour. New York Mic. Soc., pp. 7-11. Brachiospongidze: A memoir on a group of Silurian sponges. Mem. Peabody Mus., Yale Univ., vol. ii, pp. 1-28, pls. i-iv. The development of some Silurian Brachiopoda (with eight plates). [Beecher and Clarke.] Mem. N. Y. State Mus., vol. i, pp. 1-95, pls. i-vili. On the development of the shell in the genus Tornoceras Hyatt. Am. Jour. Sci. (3), vol. xl, pp. 71-75, pl. i. Koninckina and related genera. Jbid., vol. xl, pp. 211-219, pl. ii. The development of a paleozoic poriferous coral. Trans. Conn. Acad. Sci., vol. vili, pp. 207-214, pls. ix—xili. Symmetrical cell development in the Favositide. Jbid., pp. 215-220, pls. EIvV; XV. Development of the Brachiopoda. I. Introduction.. Am. Jour. Sci. (3), vol. xli, pp. 348-457, pl. xvii. Development of the Brachiopoda. II. Classification of the stages of growth and decline. Jbid., vol. xliv, pp. 183-155, pl. i. Revision of the families of loop-bearing Brachiopoda. Trans. Conn. Acad. Sci., vol. ix, pp. 376-391, pls. i, ii. The development of Terebratalia obsoleta Dall. Jbid., vol. ix, pp. 392-399, Geeta enent of the brachial supports in Dielasma and Zygospira. [Beecher and Schuchert.] Proc. Biol. Soc. Washington, vol. viii, pp. 71-78, pl. x. Larval forms of Trilobites from the Lower Helderberg group. Am. Jour. Sci. (3), vol. xlvi, pp. 142-147, pl. ii. | On the thoracic legs of Triarthrus. Jbid., vol. xlvi, pp. 367-370. On the mode of occurrence and the structure and development of Triarthrus Becki. American Geologist, vol. xiii, pp. 38-48, pl. iii. The appendages of the pygidium of Triarthrus. Am. Jour. Sci. (8), vol. xlvii, pp. 298-300, pl. vii. Further observations on the ventral structure of Triarthrus. American Greolo- gist, vol. xv, pp. 91-100, pls. iv, v. Structure and appendages of Trinucleus. Am. Jour. Sci. (3), vol. xlix, pp. 307-311, pl. iii. The larval stages of Trilobites. American Geologist, vol. xvi, pp. 166-197) pls. vii-x. James Dwight Dana. Jbid., vol. xvii, pp. 1-16, portrait, pl. i. The morphology of Triarthrus. Am. Jour. Sci. (4), vol. i, pp. 251-256, pl. viii ; reprinted in Geological Magazine (London), dec. IV, vol. iii, pp. 193-197, pl. ix. Outline of a natural classification of the Trilobites. Am. Jour. Sct. (4), vol. ili, pp. 86-106, 181-207, pl. iii. The systematic position of the Trilobites. [Kingsleyand Beecher.] Ameri- can Geologist, vol. xx, pp. 33-40. Development of the Brachiopoda. III. Morphology ofthe Brachia. Bulletin 87, U. S. Geol. Survey, chapter iv, pp. 105-112. Origin and significance of spines. Am. Jour. Sci. (4), vol. vi, pp. 1-20, 125-136, 249-268, 329-359, pl. i. 548 — PROCEEDINGS OF THE PHILADELPHIA MEETING 1899. Othniel Charles Marsh. Jbid., vol. vii, pp. 403-428; the same, abridged, with alterations. Bull. Geol. Soc. Am., vol. 11, pp. 521-537, and American Geologist, vol. xxiv, pp. 135-157. 1900. Trilobita. In Textbook of Paleontology, by Karl A. von Zittel. Trans- lated and edited by Charles R. Eastman. Vol. J, pp. 607-638. 1901. Studies in evolution: mainly reprints of occasional papers selected from the publications of the Laboratory of Invertebrate Paleontology, Peabody Museum, Yale University. Pp. xxiii and 638, 34 plates. New York. ©1901. Discovery of Eurypterid remains in the Cambrian of Missouri. Am. Jour. Sct. (4), vol. xii, pp. 364-366, pl. vii. 1902. The ventral integument of Trilobites. Jbid., vol. xiii, pp. 165-174, pls. ii-v. 1902. Paleozoic Phyllocarida from Pennsylvania. Quart. Jour. Geol. Soc. London, vol. lvili, pp. 441-449, pls. xvii-xix. 1903. Observations on the genus Romingeria. Am. Jour. Sci. (4), vol. xvi, pp. 1-11, pls. i-v. ; 1904. Extinction of species. Hncyclopedia Americana, vol. xv, 4 pp. MEMOIR OF JOHN B. HATCHER* BY W. B. SCOTT A full account of Mr Hatcher’s varied and eventful life would be a fascinating story of adventtre, of daring and unconquerable energy, of self-sacrificing devotion to duty, and of single-minded love of science. This story can never be adequately written, for only the hero could have done that; what he might have made of it is indicated by the “ Narra- tive” of his expeditions to Patagonia. In reviewing this work for Science, Doctor Dall has said: | ‘“‘At times wrapped in gloomy fogs or swept by tempests of incredible violence ; fronting the towering Atlantic surges with unshaken cliffs and serrate talus, look- ing out to shifting bars of sand, the terror of the navigator; a vast cemetery for ghostly herds, upon the like of which alive no man has ever looked ; it is astrange, silent, bitter, lonely land. ‘How our author went out into it, what he met, and how he fared are told in modest yet most interesting fashion in this stately quarto. His story is so inter- esting and the unpretentious courage of the narrator so evident, the spirit of the land and its mysterious fascination so fully expressed, that few will close the book without a regret that it can not reach a wider audience. It is really too good to be reserved for the readers of quartos. ‘‘The volume is so full of scientific meat that it is difficult to make a satisfactory abstract and impossible to condense it within the limits of such a review as this. There is something for every taste. The life of bird and beast; the phases and contrasts of vegetation ; the life of the Tehuelche Indians and the waifs who have cast civilization aside like a garment at the call of the wild; the topography and geology; and mingled with it all a flavor of real North American character, to which something in each reader’s soul will leap with sympathy and admiration.” * This memoir was not read on account of the absence of the author, but is inserted here in its proper place. ee ee ee » 4 BULL. GEOL. SOC, AM. VOL. 16, 1904, PL. 91 MEMOIR OF JOHN B. HATCHER 549 While it is now impossible to tell this remarkable story in detail, its principal events may be compressed into a brief statement. John Bell Hatcher, the son of John and Margaret Hatcher, was born at Cooperstown, Brown county, Illinois, on October 11, 1861, and at a very early age removed with his family to Greene county, lowa, where he remained till his twentieth year. Asa child, he was so weak that his parents could scarcely hope to see him grow to manhood, and his early education was given him by his father, who was a teacher as well as a farmer. Gradually his strength increased, and as he grew older there matured within him a determination to become a thoroughly educated man—a determination which he followed with characteristically persistent energy and scorn of obstacles, however seemingly insuperable. ‘Tosecure the funds necessary for his education, he became a coal miner, and, being a born observer, the miner’s life soon awakened in his mind a great and ever growing interest in geology and in the fossils which he saw around him. It was to follow this early bent that in 1880, after a short stay at Grinnell College, Iowa, he went to Yale University, whither he was espe- cially attracted by the fame of Professor J. D. Dana, whose books he had zealously studied. At New Haven he devoted himself to the natural sciences, more particularly to geology and botany. A collection of Car- boniferous fossils, which he had made in his coal-mining days, was the means of introducing him to Professor Marsh, who sent him to the West as a collector immediately after his graduation. In this connection I may perhaps be permitted to quote what I have elsewhere written : “Thus began a career which was unrivaled of its kind, for Hatcher had a positive genius for that particular work, as is well known to all who have had the privilege _of accompanying him in the field. Marvelous powers of vision, at once telescopic and microscopic, a dauntless energy and fertility of resource that laughed all ob- stacles to scorn, and an enthusiastic devotion to his work combined to secure for him a thoroughly well earned success and a high reputation. He may be said to have fairly revolutionized the methods of collecting vertebrate fossils, a work which before his time had been almost wholly in the hands of untrained and unskilled men, but which he converted into a fine art. The exquisitely preserved fossils in American museums, which awaken the admiring envy of European paleontologists, are to a large extent directly or indirectly due to Hatcher’s energy and skill and to the large-minded help and advice as to methods and localities which were always at the service of any one who chose to ask for them. **Hatcher’s uprightness and sincerity of character, no less than his remarkable energy and persistence, attracted to him the admiration of many western men, by whom frequent tempting offers were made him to leave the unremunerative paths of science for the material rewards of business; but in vain. He would not seri- ously consider the abandonment of his chosen work for any reward whatever, and he died in harness.” 550 PROCEEDINGS OF THE PHILADELPHIA MEETING Of his wonderful activity and success as a collector, Schuchert has remarked : ‘From 1884 to 1892 he sent in nearly 900 boxes of vertebrate material. Asa rule, these boxes were of large size, and one exceeded 3 tonsin weight. This huge box (about 10 feet long, 5 feet wide, and 6 feet deep), containing the largest known skull of Triceratops, had to be lifted out of a ravine 50 feet deep and hauled to the railroad over a trackless country and through streams for more than 40 miles. It is no exaggeration to state that during the 20 years of Hatcher’s paleontological activity, he, with the assistance of a few field helpers, sent to the United States National Museum, and to Yale, Princeton, and Carnegie museums not less than 1,500 boxes of fossils. This isa record that will stand unequaled—a work that Hatcher loved—resulting in material part of which he hoped it would be his lot to study. After Marsh’s death the uncompleted Ceratopsia volume was assigned to Hatcher by the United States Geological Survey. This gave him great gratifica- tion, for he was thus enabled to associate his name, not only as a collector but aiso as a student, with these great and curious beasts, all of which he had discovered and taken up.”’ . Dr W. J. Holland, director of the Carnegie museum in Pittsburg, has borne similar testimony to Hatcher’s work as a collector: ‘‘Mr Hatcher’s position as a paleontologist was unique. He is universally ad- mitted by those most competent to pass judgment to have been the best and most successful paleontological collector whom America has ever produced. In saying this it may at once be admitted that he was in all probability the most successful collector in his chosen domain who has ever lived. Professor Hatcher and those associated with him under his control during the years of his activity in the field assembled more important vertebrate fossils than have been assembied by any other one man whose name is known in the records of paleontology. The larger proportion of the choicest vertebrate fossils now in the Peabody museum at Yale University, in the collection of the United States Geological Survey, in the museum of Princeton University, and in the museum of the Carnegie Institute at Pittsburg were collected by him. Toa very large extent the American methods of collecting such remains, which are now universally admitted to be the best methods known, were the product of his experience in the field and of his careful thought. Ina letter just received by the writer from Professor Henry Fairfield Osborn, the Paleon- tologist of the United States Geological Survey, he says, alluding to the death of Professor Hatcher: ‘I can hardly tell you how shocked and grieved Iam. I had often thought of the probability of Hatcher’s death in the field when taking great risks and entirely away from medical and surgical attendance, but of his death at home I had not thought a moment. In his intense enthusiasm for science, and the promotion of geology and paleontology, and the tremendous sacrifices he was prepared to make, and had made, he was a truly rare and noble spirit—the sort of man that is vastly appreciated in England and in Germany, but I fear very little appreciated in America. Huis work as a collector was magnificent—probably the greatest on record.’ ” While thus at work as a collector over an enormously wide range of country and through almost the whole geological column from the Per- MEMOIR OF JOHN B. HATCHER 551 mian to the Pleistocene, Hatcher was continually studying the stratigra- phy of the beds in which he worked and determining their faunistic divisions and subdivisions. Nor did he neglect the dynamical and structural problems involved in the formation of the successive beds- Not only was he an unusually keen and accurate observer, but he pos- sessed a singularly original and independent mind. He was utterly - impatient of authority in science, and to him every theory must be sup- ported by convincing evidence and not merely buttressed by the weight of greatnames. His untimely death has robbed the world of a rich store of geological knowledge, the publication of which had not fairly begun. In 1893 Hatcher accepted a call to Princeton University as assistant in geology and curator of vertebrate paleontology in the museum, and with unabated zeal continued his work along much the same lines as before. During tbe three seasons 1893-1895 he worked in the Uinta, White River,and Loup Fork and Sheridan beds, gathering great quanti- ties of priceless material. On all of these trips he was accompanied by field parties of students, who became his fast friends and ardent ad- mirers; his skill, energy, persistence in the face of difficulty and peril appealed most strongly to their imaginations. On the other hand, he took the warmest interest in his students, especially in those whose edu- cation must be gained through their own efforts ; with admirable delicacy " and tact, he was fertile in devices to enable them to help themselves and thus continue their studies unweighted by any humiliating sense of being the objects of charity. The most important work that Hatcher undertook during his connec- tion with Princeton, and perhaps the most important enterprise of his whole career, was his exploration of Patagonia in the three expeditions of 1896-1899. The plan and execution of this great work were his own; from his former students he obtained a large part of the necessary funds, to which he generously contributed himself. Indeed, proportionately to his means he was the largest subscriber to the fund. The expeditions, the main object of which was to secure representative collections illus- trating the geology and paleontology of Patagonia, were brilliantly suc- cessful, and their scope was gradually extended so as to include as well the botany and zoology of the region. ‘To his assistants, Messrs Peterson and Colburn, much credit for the success of the work is due; but the leading spirit was Hatcher’s throughout. In his “ Narrative,” already mentioned, may be found the extremely well written and interesting account of his Patagonian journeys; but, interesting as it is, this book gives to the reader a very imperfect conception of his achievement. Only those who heard his intimate talk and had the pleasure of reading his fascinating letters from the field can understand what were the diffi- ae +, 552 PROCEEDINGS OF THE PHILADELPHIA MEETING culties and dangers that opposed his advance or can rightly estimate the dauntless courage and unrelenting energy which triumphed over the most formidable obstacles, both material and moral. Wounds and sick- ness, long weeks of helpless and agonizing pain, hardships of every kind, represent but a few of the difficulties that he met and conquered. In the long and glorious history of scientific exploration there are but few chapters that tell of truer heroism and finer achievement than Hatcher’s life in Patagonia. When all the collections had been brought to Princeton and it was seen what a mass of new and valuable material had been secured, Hatcher conceived the plan of publishing the whole in a uniform series of reports by the ablest specialists who could be induced to cooperate. The only alternative would have been a crowd of more or less fragment- ary and uncorrelated papers scattered through many technical journals and proceedings of societies. The liberality of J. Pierpont Morgan, Esq., has made possible the realization of this plan, and the “ stately quartos ” of the Reports of the Princeton University Expeditions to Patagonia will form a lasting monument to the memory of Hatcher, whose labors and sacrifices they record. In February, 1900, Hatcher became curator of vertebrate paleontology in the museum of the Carnegie Institute, Pittsburg, a position in which ~ he remained till his death, on July 3, 1904. The same qualities that had distinguished his earlier career continued to make him equal to his new and larger duties and responsibilities. To him especially is due the extremely rapid growth of the Pittsburg collections and their remarkably high quality. Aside from the desolated household and the circle of bereaved friends, the essential tragedy of Hatcher’s early death lies in the fact that his work had just begun. Though best known as a collector and the most - skillful and successful of collectors, he was very much more than that. For 20 years he had been giving himself the most thorough training and acquiring an experience of such magnitude and variety as falls to the lot of few geologists: All his previous life had been but the seed time, and. just as the harvest was ripe for the sickle, the reaper was stricken down. Owing to a modest self-distrust, his productive period began relatively late, and his first paper was published after his removal to Princeton ; but he gradually gained confidence with experience, and had his life been spared he would surely have enriched science with a series of notable contributions. For enthusiastic, self-sacrificing devotion, unconquerable determination, and high achievement, we shall not soon look upon his like again. 1893. 1893. 1894. 1894. 1894. 1895. 1896. 1896. 1897. 1897. 1897. 1897. 1899. ~ 1899. 1899. 1900. 1900. 1900. 1900. 1901. 1901. BIBLIOGRAPHY OF JOHN B. HATCHER 553 BIBLIOGRAPHY * The Ceratops beds of Converse county, Wyoming. Am. Jour. Sct. (3), vol. xlv, pp. 135-144. The Titanotherium beds. American Naturalist, vol. xxvii, pp. 204-221, text figs. 1-3. A Median horned Rhinoceros from the Loup Fork bedsof Nebraska. Amer- can Geologist, vol. xiii, pp. 149, 150. On a small collection of vertebrate fossils from the Loup Fork beds of north- western Nebraska; with note on the geology of the region. Anerican . Naturalist, vol. xxviii, pp. 236-248, text figs. 1, 2, pls. i, ii. Discovery of Diceratherium, the two-horned rhinoceros, in the White River beds of South Dakota. American Geologist, vol. xiii, pp. 360, 361. On a new species of Diplacodon, with a discussion of the relations of that genus to Telmatotherium. American Naturalist, vol. xxix, pp. 1084-1090, text figs. 1, 2, pls. xxxviil, xxxix. Discovery, in the Oligocene of South Dakota, of Eusmilus, a genus of saber- toothed cats new to North America. Jbid., pp. 1091-1093, pl. xl. Some localities for Laramie mammalsand horned Dinosaurs. Jbid., vol. xxx, pp. 112-120, pl. iii. Recent and fossil tapirs. Am. Jour. Sci. (4), vol. ?, pp. 161-180, text figs. 1, 2, pls. ii-v. The Cape Fairweather beds; a new marine Tertiary horizon in southern Patagonia. TIbid., vol. iv, pp. 246-248, 1 text fig. On the geology of southern Patagonia. Jbid., pp. 327-354, text figs. 1-11, and sketch map. Diceratherium proavitum. American Geologist, vol. xx, pp. 313-316, pl. xix. Patagonia. Nat. Geog. Mag., vol. viii, pp. 305-319, 2 text figs.and map, pls. 35-37. The Third Princeton Expedition to Patagonia. Science, n.s., vol. x, pp. 580, 581. [Unsigned article. ] Explorationsin Patagonia. Scientific American, vol. 1xxxi, pp. 328, 329, 9 figs. The mysterious mammal of Patagonia, Grypotherium domesticum. By Rudolph Hauthal, Santiago Roth, and Robert Lehmann Nitsche. Revista del Museo de La Plata, vol. ix, pp. 409-474. Review. Science, n.s., vol. x, pp. 814, 815. Sedimentary rocks of southern Putian Am. Jour. Sci. (4), vol. ix, pp. 85-108, pl. i. Some geographic features of southern Patagonia ; with a discussion of their origin. Nat. Geog. Mag., vol. xi, pp. 41-55, 3 text figs., pl. 2. The Carnegie Museum Pulepatslosionl Expeditions of 1900. Science, n. vol. xii, pp. 718-720. Vertebral formula of Diplodocus (Marsh). Jbid., pp. 828-830. The Indian tribes of southern Patagonia, Tierra del Fuego, and adjoining islands. Nat. Geog. Mag., vol. xii, pp. 12-22, 4 text figs. The lake systems of southern Patagonia. Bull. Geog. Soc. Philadelphia, vol. ii, pp. 189-145, map; and American Geologist, vol. xxvii, pp. 167-173, pl. xvi. * From Schuchert, American Geologist, 1905, page 139. 1903. PROCEEDINGS OF THE PHILADELPHIA MEETING . Some new and little known fossil vertebrates. Ann. Carnegie Mus., vol. i, pp. 128-144, text fig. 1, pls. i-iv. On the cranial elements and the deciduous and permanent dentitions of Titanotherium. Jbid., pp. 256-262, text fig. 1, pls. vii, viii. Sabal rigida; a new species of palm from the Laramie. Jbid., pp. 263, 264, text fig. 1. The Jurassic Dinosaur deposits near Canyon City, Colorado. JIbid., pp. 327- 341, text figs. 1-5. . Diplodocus Marsh: Its osteology, taxonomy, and probable habits, with a restoration of the skeleton. Mem. Carnegie Mus., vol. i, pp. 1-63, text figs. 1-24, pls. i-xiii. . On the structure of the manus in Brontosaurus. Science, n. s., vol. xiv, pp. 1015-1017. . A mounted skeleton of Titanotherium dispar Marsh. Ann. Carnegie Mus., vol. i, pp. 347-355, pls. xvi-xviil. . Structure of the fore limb and manus of Brontosaurus. JIbid., pp. 356-376, text figs. 1-14, pls. xix, xx. . The genera and species of the Trachodontidee (Hadrosauride, Claosauride), Marsh. Ibid., pp. 377-386. . Origin of the Oligocene and Miocene deposits of the Great plains. Proc. Am. Phil. Soc., vol. xli, pp. 113-1381. . Oligocene Canide. Mem. Carnegie Mus., vol. i, pp. 65-108, text figs. 1-7, pls. Xiv-xx. . Discovery of a musk ox skull (Ovibos cavifrons Leidy) in West Virginia, near Steubenville, Ohio. Science, n. s., vol. xvi, pp. 707-709, 1 text fig. . Field work in vertebrate paleontology at the Carnegie Museum for 1902. Ibid., p. 752. . A correction of Professor Osborn’s note, entitled ‘‘ New Vertebrates of the mid-Cretaceous.’’ Jbid., pp. 831, 832. . A new sauropod Dinosaur from the Jurassic of Colorado. Proc. Biol. Soc. Washington, vol. xvi, pp. 1, 2. . The Judith River beds. - Science, n. s., vol. xvii, pp. 471, 472. . L’ Age des Formations Sédimentaires de Patagonie ; by Florentino Ameghino. Annals Soc. Cientif. Argentina, pp. 3-231, Buenos Aires, 1903. Criticism, Am. Jour. Sci. (4), vol. xv, pp. 483-486. . Discovery of remains of Astrodon (Pleurocoelus) in the Atlantosaurus beds of Wyoming. Ann. Carnegie Mus., vol. ii, pp. 9-14, text figs. 1-6. . Relative age of the Lance Creek (Ceratops) beds of Converse county, Wyoming, the Judith River beds of Montana, and the Belly River beds of Canada. Am. Geol., vol. xxxi, pp. 369-375. . The stratigraphic position of the Judith River beds and their correlation with the Belly River beds. Science, n. s., vol. xvili, pp. 211, 212. . Vertebrate paleontology at the Carnegie Museum. JTJbid., pp. 569, 570. . Osteology of Haplocanthosaurus, with description ofa new species and remarks on the probable habits of the Sauropoda and the age and origin of the Atlantosaurus beds. Mem. Carnegie Mus., vol. ii, pp. 1-72, text figs. 1-28, pls. i-v. Additional remarks on Diplodocus. Jbid., pp. 72-75, text figs. 1, 2, pl. vi. 4 MEMOIR OF HENRY MCCALLEY 555 1903. Narrative and geography. Reports of the Princeton University expeditions to Patagonia, 1896-1899, vol. i, pp. Xvi + 314, plates and maps. v 1904. A new name for the Dinosaur Haplocanthus, Hatcher. Proc. Biol. Soc. Washington, xvi, p. 100. ’ An attempt to correlate the marine with the non-marine formations of the middle West. Proc. Am. Philosophical Soc., vol. xliii, no. 178, p. 341. MEMOIR OF HENRY MCCALLEY BY EUGENE A. SMITH Henry McCalley, who died of pneumonia in Huntsville, Alabama, on November 21, 1904, was born in that city February 11, 1852. He was the son of Thomas Sanford McCalley, of Spottsylvania county, Virginia, and Caroline, daughter of Robert Landford, who built the second house in Huntsville. Mr McCalley was one of a family of nine children who reached adult age. He lived at his home, 2 miles west of the court-house in Huntsville, from his birth tomanhood. His school career was begun under the care of Mrs McKay, then considered the most excellent teacher for young children. From Mrs McKay he went to Dr J. M. Bannister, rector of the Church of the Nativity, and afterward to the noted Mr Charles Shepard, who is still living and engaged in teaching. At the well known school of Dr Carlos G. Smith he was prepared for college, soon after the end of the civil war. As the University of Alabama was at that time in the hands of the “ carpet-baggers? and without students, he went to the University of Virginia, from which institution he was graduated in 1876 with the degrees of Civil Engineer and Mechanical Engineer. At the university he applied himself very closely to his studies, gaining the highest esteem of both professors and students, but sacrificing his health. On his return home after graduation he spent one year on the farm with a view to restoring his health. With the strong recommendation of the faculty of the University of Virginia, he took charge of a school at Demopolis, Alabama, where he remained one year and part of another. In the summer vacation of 1877 he came to the Geological Survey of Alabamaas a volunteer assistant and traveled with the writer through a part of the Warrior coal field and the valley of the Tennessee. The following year, 1878, Mr McCalley gave up his school and came to the University of Alabama as assistant in the department of chemistry, then in charge of the writer of these lines. This position he held until 1883, at the same time also serving as volun- teer assistant on the Geological Survey, for during the first ten years of the existence of this second survey the annual appropriation was only $500, none of which went for salaries. LXX—Butv. Geon. Soc. Am., Vou. 16, 1904 During the summer months of 1879 we had charge of a survey of the Warrior river for the Engineer Office of the War Department, under Major Damrell, the object of which survey was primarily to ascertain the nature and extent of the obstructions to navigation and to obtain estimates of the cost of removing or overcoming the same, and, secondly, to collect statistics of the natural resources of the country lying adjacent to the river. The levelings and soundings along the river were made under the di- rection of Mr McCalley, while the geological data were collected by Mr Joseph Squire and myself. Our joint report was published ina Report of Progress for 1879-1880. Later in the year Mr McCalley spent some time in the Tennessee valley, and the results of his labors were given in the same report of progress. In 1888 the legislature made an appropriation for the Geological Survey which made it possible to employ salaried assistants, and Mr McCalley then received the appointment as assistant state geologist, which position he held until his death, a period of 21 years. His first work in this capacity was in the Warrior basin, on which his first report was published in 1886. This was the first cotprelianeive statement of the characters and succession of the coal seams of this great field and it gave great help to those who were engaged in the development of the state. Next he took up the study of the plateau portion of the Warrior field, in northeastern Alabama, and his report on the coal meas- ures of this section was. published in 1881. His next work was in the Paleozoic formations of the Tennessee, Coosa, and other great valleys-in which occur the limestones, iron ores, and bauxites of the state, and the results of several years work in this section were published in 1896 and 1897 under the title ‘‘ The Valley Regions of Alabama,” part I being devoted to the valley of the Tennessee and part II to the Coosa and other anticlinal valleys, Cahaba, Wills, Jones, and Blount Springs valleys. | The great activity in coal mining during the 10 years following the publication of the report of 1886 on the Warrior basin rendered neces- sary a reexamination and more thorough study of the field, and Mr Mce- Calley spent much time in going again over the ground with Mr George N. Brewer as an assistant, and in 1890 appeared his Be on the War- rior basin, with a large map. Since 1900 his work has fae in the region of the igneous and meta- morphic rocks, upon which he was engaged at the time of his death. Unfortunately his notes on this region were not written up, though quite full and comprehensive. This will make it impossible to get the full © benefit of his work. 556 PROCEEDINGS OF THE PHILADELPHIA MEETING nk ie MEMOIR OF HENRY MCCALLEY | 557 In personal character Mr McCalley was modest and somewhat retir- ing, but no one could be more firm and decided than he in the defense of a friend and in the defense of his own opinions on geological matters after he had formed them from his own extended observations. In his scientific work he was careful and painstaking to an extraordinary degree, and his conclusions were rarely hastily formed, and they were in conse- quence generally correct. He was one of the most truthful of men, and he could be relied upon to do to the best of his ability whatever work was assigned to him. When called upon to give his views he did it with the utmost frankness, swerving neither to the right nor the left from the straight path of truth. In his connection with the Survey he was sey- eral times called upon to expose frauds, and, after he had become con- vinced on thorough examination of the facts, he left those concerned in ho doubt as to his conclusions and convictions. He was a consistent and active member of the Episcopal church and very generous in his support of it, responding cheerfully to all the calls made upon his time and means. It is said of him that asa child and a young boy he had never re- ceived a correction from either parent or teacher and he never neglected a task set before him; these characteristics followed him through life. Mr McCalley was a member of the American Association for the Ad- vancement of Science; Fellow of the Geological Society of America ; member of the American Institute of Mining Engineers; secretary of the Alabama Industrial and Scientific Society, etcetera. The following bibliography takes account of all his papers and reports and, it is believed, the greater part of his minor publications : BIBLIOGRAPHY 1878. Letters to the Huntsville Democrat, descriptive of the physical, geological, and economic features of northern Alabama. 1879. Chemical report to the State Geologist of Alabama. Reportof the State Ge- ologist of Alabama for 1877-1878. 1880. Study of chemistry. Article in the Alabama University Monthly. 1881. Geological report of that part of Alabama that lies north of the Tennessee river. State Geologist’s report for 1879-1880. 1881. Review in the Tuscaloosa Gazette of the Geological Report of 1879-1880. 1883. Letters to the Tuscaloosa Gazette, descriptive of the physical and geologic feat- ures and economic wealth of the Warrior coal field. 1883. Review in the Birmingham Daily Age of the Report of the State Geologist for 1881-1882. 1883. Report in the Mountain Eagle, Jasper, Alabama, on the economic wealth of Walker county. 1886. North Alabama or the mountain, manufacturing, and mineral region of Ala- bama. Chapter ii in the Mineral Wealth of Alabama and Birmingham IIlus- trated, published in 1886. 558 1886. 1886. 1886. 1887. 1888. 1890. 1891. 1891. 1892. 1894. 1894. 1890. 1890. 1896. 1896. 1896. 1897. 1897. 1897. 1897. 1899. 1899. 1901. PROCEEDINGS OF THE PHILADELPHIA MEETING | The coal fields of Alabama and their economic wealth. Article in Dizie, Atlanta, Georgia. Report of the Warrior coal fieldsof Alabama. Geological Survey of Alabama Report. Review in the Montgomery Advertiser of the Report of the Geological Survey of Alabama for 1882-’83, 1883-84. Topography, geology, and natural resources of Jefferson county, Alabama, chapter i, in the History of Jefferson County and Birmingham, Alabama. Published by Steeple and Smith, 1887. Mineral resources of Tuscaloosa county. Paper presented to committee for selection of site for manufacture of heavy ordnance. The coal fields of Alabama. Scientific American, Supplement. Report on the Plateau region of Alabama. Geological Survey of Alabama. Natural gas and petroleum in north Alabama. Alabama Industrial and Scientific Society, vol. i, no. 1. Alabama bauxite. Alabama Industrial and Scientific Society, vol. ii. Bauxite mining. Science, vol. xxili, no. 572. : Bauxite. Mineral industry for 1893, vol. iii. Alabama barite or heavy spar. Alabama Industrial and Scientific Society, vol. v. Report on so-called gold deposit of Sauta creek, Marshall county, Alabama. Valley regionsof Alabama. Part i, on the Tennessee Valley region. Report of the Geological Survey of Alabama. Limonites of Alabama, geologically considered. Engineering and Mining Journal, vol. xliii, no. 25. Hematites of Alabama, geologically considered. Engineering and Mining Journal, vol. xliii, no. 2. Review inthe Montgomery Advertiser of the Geological Survey Report for 1896. Fluxing rocks of Alabama, geologically considered. Engineering and Mining Journal, vol. xlviii, no. 5. Report on the Valley regionsof Alabama. Partii.On the Coosa Valley region. Geological Survey of Alabama. Manganese ores of Alabama. Mineral resources of the United States, 1896. Map of the Warrior coal basin, with columnar sections. Geological Survey of Alabama. Review of the map of the Warrior coal basin in different state papers. Coal fields of Alabama. Minesand minerals. [May, 1901.] Vol. xxi, no. 10. MEMOIR OF WILLIAM HENRY PETTEE BY ISRAEL C. RUSSELL William Henry Pettee was born in Newton Upper Falls, Massachusetts, January 18, 1886, of representative New England parentage. His father was a manufacturer of cotton fabrics and of mill machinery. In boy- hood his studious tastes had to be restrained and his college preparation delayed out of regard to his somewhat slender bodily frame. He entered Harvard College at nineteen years of age, took high rank in the required classical course of that period, was selected to deliver a Latin oration MEMOIR OF WILLIAM HENRY PETTEE 559 in his junior year, and graduated with distinction in the class of 1861. He continued in graduate work in the same university for over three years, receiving the degree of Master of Arts in 1864, studying at first in the engineering department of the Lawrence Scientific School and later in the college, where at the same time he was an assistant. ; From 1865 to 1869 he traveled and studied in Europe, his main work being in the Royal Mining Academy of Saxony, at Freiberg, with vaca- tions in the mining regions of Germany. In 1868 Mr Pettee returned to Harvard University as a teacher in the School of Mining and Practical Geology, then established, under the direction of Josiah D. Whitney. His appointment in 1869 was that of instructor in mining, but in 1871 he was advanced to the rank of assist- ant professor in the same branch and provision made for work upon geological surveys to be carried on under the auspices of the Harvard School of Mining. In the summer. of 1869 Professor Pettee made a geological and topo- graphical survey of South Park, Colorado, and during the year 1870-1871, having been granted a leave of absence from Harvard, he became con- nected with the California State Geological Survey. Besides making a study of gold-bearing gravels of California, he undertook systematic work in correction of the determination of altitudes by means of the barometer. Some of the results of this investigation, collected from the detailed re- ports of the survey, were published by authority of the California state legislature in 1874, entitled ‘“‘ Contributions to Barometric Hypsometry, with Tables for use in California,” to which a supplement was added in 1878. Professor Whitney’s estimate of the onerous labor, the accuracy, and perseverance of Professor Pettee’s work in this undertaking appears in the prefatory note to the volume above mentioned. From 1871 to 1875, in addition to other duties, Professor Pettee gave in- struction to an elective section of undergraduates in physical geography, geology, and meteorology at Harvard; but before 1875 the conditions of the gift supporting a school of mining at that institution were altered and provision for a special instructor in these subjects was withdrawn. In 1875 Professor Pettee was appointed to a professorship of mining engineering and related subjects in the University of Michigan, a position which he held with various changes of title until his death. In the first semester of 1879-1880 Professor Pettee was granted leave of absence from the University of Michigan to continue his investigation of the auriferous gravels of California. His report on that work was pub- lished as an appendix to the first volume of Whitney’s “ Contributions to American Geology.” It has been adjudged to show that careful exam- ination of phenomena, weighing of evidence, and painstaking accuracy, . ° 7? : A 560 PROCEEDINGS OF THE PHILADELPHIA MEETING which those best acquainted with Professor Pettee always expect in papers prepared by his hand. 7 | : The annual Transactions of the American Institute of Mining Hngineers have been submitted to Professor Pettee for many years for critical proof- reading and correction. Of that society he was a life member, his election dating from 1871. For many years he was a coworker with its secretary, Rossiter W. Raymond, who, in a recent memorial published in the Trans- actions of the Institute, expressed high appreciation of Professor Pettee’s ability as a literary critic. He was one of the oriyinal fellows of the Geo- logical Society of America, a fellow of the American Agsociation for the Advancement of Science, in which he was general secretary in 1887; a member of the American Academy of Arts and Sciences during his resi- dence in Massachusetts, and a member of the American Philosophical Society of Philadelphia. For Professor Pettee the members of the faculty of the University of Michigan, of which he was a member for twenty-nine years, hold only memories of the highest respect and the warmest friendship, These cher- ished sentiments are but a reflection of his own genial and loving nature, left by him as an inheritance to all with whom he came in contact. His more pronounced characteristics as revealed in his intercourse with his colleagues, whether personally and socially or in connection with official duties, and equally conspicuous to the students who received his instruc- tion, were a kindly and loving nature, patience under difficulties, pains- taking accuracy in all of his work, love of truth, and unswerving upright- ness of character. With these high ideals were coupled an abundant and ever accessible knowledge of the history and traditions of education in Michigan and a love for the branches of science to which he devoted his time and energy. . Professor Pettee died at his home in mean Arbor, Michigan, on May 26, 1904. While his chief work during life was that a a teacher, his few con- tributions to geology and kindred subjects show that he was a painstaking and accurate observer. BIBLIOGRAPHY 1874. Contributions to barometic hypsometry, with tables for use in California. Geological Survey of California, J. D. Whitney, State geologist. Pub- lished by authority of the legislature, 1874. To this report a supple- mentary chapter consisting of pages 89-112 was added in 1878, which was included in subsequent editions of the same volume. 1879. Report on an examination of portions of the gravel mining regions of California, in Placer, Nevada, Yuba, Sierra, Plumas, and Butte counties, made in 1879. Forms appendix A, pp. 379-487, of The auriferous gravels of the Sierra Nevada of California, vol. 1, Contributions to American geology, by Whitney, Harvard College, Museum of Comparative Zoology, memoirs, vol. 6, Cambridge, 1880. | —— ‘a MEMOIR OF CHARLES SCHAEFFER 56] In the absence of the author the following memoir was presented in abstract by Professor A. P. Brown, University of Pennsylvania : MEMOIR OF CHARLES SCHAEFFER BY ANGELO HEILPRIN In the death of Dr Charles Schaeffer, which took place on November 24, 1908, the friends of science have lost a genial and most lovable associate, and one who, while not professionally nor professedly a naturalist, was to that extent interested in the beauties and workings of nature as to consider the field of inquiry of the naturalist as his own by association. A love for flowers, a discriminating eye for minerals, and an unusually skillful hand in the manipulation of the camera, brought to Doctor Schaeffer a many-sided nature, which years of pleasurable travel, with a loving wife for helpmate, broadened out into what are still to many “pastures new.” Most of Doctor Schaeffer’s observations were given by word of mouth to his associates, and only exceptionally were they published in paper form. He was a member of numerous scientific societies: American Philosophical Society, Academy of Natural Sciences of Philadelphia, Geological Society of America, Franklin Institute, Geographical Society of Philadelphia, American Association for the Advancement of Science, Pennsylvania State Medical Society, Philadelphia County Medical Society, College of Physicians of Philadelphia, Historical Society of Pennsylvania, Photographic Society of Philadelphia, Horticultural So- ciety of Pennsylvania, etcetera. His closest association was with the Academy of Natural Sciences of Philadelphia, to whose body of mem- bers he was elected in 1861, and in which he served for long terms of years on the Council and as Recorder to the Botanical, the Biological, and the Mineralogical and Geological sections. For many of the later years of his life Doctor Schaeffer made the Canadian Rocky mountains and the Selkirks his summer abode, drawing from them a wealth of new infor- mation and imparting to others through his collections charming les- sons of the region which he could almost justly call hisown. The little colony of Philadelphians who year after year return to Glacier House or thereabout to make observations on glacial phenomena or mountain structures, and on the habits, colors, and distribution of the Canadian sub-Alpine floras, have been pioneered to their camps by Doctor Schaeffer, who sought in the mountain air the new vigor of life. Doctor Schaeffer’s name will be fittingly perpetuated in asumptuously illustrated volume on the flowering plants of the Canadian Rocky moun- tains—the work, artistically and otherwise, principally of Mrs Scheeffer— now in course of preparation. 562 PROCEEDINGS OF THE PHILADELPHIA MEETING Following the reading of the memoirs, the Secretary made announce- ments concerning matters of business and the program, and Mr N. H. Darton, Committee on Photographs, announced that the collection of photographs was displayed in an adjoining room. The President declared the scientific program in order. The first paper presented was the following : DEVELOPMENT AND MORPHOLOGY OF FENESTELLA BY EDGAR R. CUMINGS [ Abstract] Thin sections and serial sections of exceptionally well preserved bases of Fenestella - (semicoscinium of authors) from the Hamilton of Thedford, Ontario, show the exact size and shape of the primary zocecium (Protecium) and the morphology and orientation of the primary buds. The protcecium consists of an elongate tubular zocecium with a large basal disk. It is without hemisepta. Morphologically, it is strictly comparable to the protcecium of Cyclostomata (of Tubulipora, Lichenopora, etcetera). The two primary buds arise from the dorsal face of the protcecium, usually just above the basal disk, and are very symmetrically orientated with ref- erence to the dorso-ventral plane. They are of about the same shape as the pro- toeecium and somewhat smaller. Each of these buds produces a single bud in the first tier, and an additional bud arising from one of the latter completes the first tier of buds—six zoccia, including the proteecium. Zocecia of the shape charac- teristic of the adult Fenestella colony do not appear till the colony begins to branch. Hemisepta have not been seen in any of the earlier zocecia. These studies seem to definitely relate Fenestella genetically to the Cyclostomata. The Cyclostomata are therefore the ancestors of the Cryptostomata and through them of the Chilostomata. Remarks on the paper were made by H.M. Ami. It is printed in full in the American Journal of Science, volume xx, pages 169-177. The second paper was BEARING OF SOME NEW PALEONTOLOGIC FACTS ON NOMENCLATURE AND CLASSIFICATION OF SEDIMENTARY FORMATIONS BY HENRY SHALER WILLIAMS The paper was discussed by the President, H. M. Ami, I. C. White, N. H. Darton, A. C. Lane, D. W. Langton, W. B. Clark, A. W. Grabau, W. N. Rice, and the author. It is printed as pages 137-150 of this volume. The last paper of the morning session was the following : GEOLOGICAL BOOKKEEPING BY JAMES F. KEMP The subject was discussed by H.S. Williams, A. P. Coleman, and the author. The paper is published as pages 411-418 of this volume. ORIGIN OF CAVES OF PUT-IN-BAY ISLAND 563 Following the discussion of Professor Kemp’s paper the Society adjourned for the noon recess. At 2.20 o’clock p m the Society reconvened, and the first paper read was ROCK CLEAVAGE BY Cc. K. LEITH This paper is published as Bulletin number 239 of the U. 8S. Geological Survey. | _ The second paper was ORIGIN OF THE CAVES OF THE ISLAND OF PUT-IN-BAY, LAKE ERIE BY EDWARD H. KRAUS [ Abstract] The rocks of the Lower Helderberg are locally very much disturbed. The largest of the four caves open to the public is Perry’s. The ceiling of this cave shows the folding very clearly and also some interesting irregularities, in that all the strata do not extend entirely across the cave. This gives the ceiling an appear- ance similar, to some extent, to an inverted steps. The floor of the cave conforms to the unevenness of the ceiling, for where there is a depression in the former there is a corresponding projection downward in the latter and vice versa. This phe- nomenon is, no doubt, the result of folding and subsequent collapse. In all prob- ability the folding was caused by the hydration of anhydrite, for large deposits of gypsum have been encountered in the sinking of wells in the immediate vicinity. Cores from these wells show a large amount of brecciation. Inasmuch as the increase in volume caused by the hydration of anhydrite may be as high as 60 per cent, and since there is a large supply of water present to bring about the solution of the gvpsum thus formed—the level of the lake is reached at a depth of about 40 feet—we have given the probable causes for the folding, leaching, and subsequent collapse. Remarks were made by A. C. Lane, G. K. Gilbert, E. H. Kraus, M. L. Fuller, H. P. Cushing, and the President. The paper is published in the American Geologist, volume xxxv, pages 167-171, March, 1905. The third paper was MOUNTAIN GROWTH AND MOUNTAIN STRUCTURE BY BAILEY WILLIS Remarks were made by A. C. Coleman, with reply by the author. Announcement of details relating to the annual dinner and the even- ing session were made by the Secretary. 564 PROCEEDINGS OF THK PHILADELPHIA MEETING The’ President called S. F. Emmons to the chair, and the next paper was OVERLAP RELATIONS ALONG THE ROCKY MOUNTAIN FRONT RANGE BY N. H. DARTON The following paper was presented by the same author : ZUNI SALT LAKE BY N. H. DARTON [Abstract] Forty miles south of the Indian pueblo of Zufii, there is, in the plain, a circular depression about a mile in diameter, containing a salt lake and two cinder cones. The depth of the depression is about 200 feet, and its walls are Cretaceous sand- stone, capped on one side by a lava flow. All around the rim is a wide low ridge of water-laid volcanic ejecta. The history of this remarkable feature is not clear, but a hypothesis will be offered as to its origin. : The paper was illustrated with Jantern slides. Itis printed in the. Journal of Geology, volume xiii, April-May, 1905. pages 185-193. The last paper of the day was EXPERIMENTAL INVESTIGATION OF THE COMPRESSIBILITY AND PLASTIC , DEFORMATION OF CERTAIN ROCKS BY FRANK D. ADAMS AND E. G. COKER [ Abstract ] The paper presents the results of an investigation into the cubic compressibility of rocks and certain phases of rock flow, carried out at McGill University by the aid of a grant from the Carnegie Institution. The apparatus employed and the methods adopted are first described. The cubic compression of rocks is then con- sidered. Seventeen (17) typical rocks have been selected and their cubic com- pressibility determined. In these determinations an indirect method was em- ployed which is believed to give as accurate results as any direct method known. The following are the rocks examined: Carrara marble, Vermont marble, Tennessee marble (pink), Belgian marble (‘‘ Noir Fin’), Montreal limestone (Trenton), Baveno granite, Westerly granite, Quincy granite (gray), Lily Lake granite, Peter- head granite, Stanstead granite, Montreal nepheline syenite, Mount Johnson essexite, New Glasgow gabbro, New Glasgow anorthosite, Sudbury diabase, Cleve- land sandstone. From the standpoint of their compressibility as well as of their composition, these rocks fall into three main groups, as follows: GTBUTUCS cscs cdawercedeskunscermmumcareneerss vacoanee Value of D averages about 4,400,000 MAIO GAs» sescan odes to prsnhgoher aeutnierccdeche nen OS SE ASS yy ‘© 6,300,000 BASIC -PlIUCOWICSapee.s c2.2ce ere eeeethoae oeeen toms ES lee s ‘© 8,800,000 The sandstone, being porous, has a much higher compressibility. D isthe modu- lus of cubic compression, and is represented by a value the reciprocal of which is the decrease in volume of a cubic inch of the rock for a pressure of one pound per COMPRESSIBILITY AND DEFORMATION OF CERTAIN ROCKS 565 -square inch. The question of the amount of pressure required to deform rocks was then investigated. This investigation was directed to obtain an answer to a problem stated by Mr G. K. Gilbert as follows: ‘‘It has been thought that great pressure breaks down the structure called solidity, and so reduces viscosity that very little differential stress is necessary to produce flow. It is thought that the strength of rocks is practically unaffected by pressure. It is certainly conceivable also that the strength of rocks is increased by pressure, so that the production of flow requires differential stress greater than the crushing stress as conditioned by the temperature.” ‘The rocks investigated were Cockeysville (Maryland) dolomite, Carrara marble Belgian marble ( “ Noir Fin”), and Baveno granite. The investigation shows that whena rock is submitted to differential stress after the elastic limit has been ex- ceeded and the texture of the rock has been broken down, owing to the internal, friction of the mass, the differential stress required to produce movement in the deformed rock is much greater than the pressure required to start deformation by breaking down the original texture] of the rock. Certain figures are presented in the case of each of the rocks and their mathematical significance described. It is shown that this work opens up a very extended field for investigation. The results of a series of experiments on the deformation of rock-making min- erals is then described. The deformation of Carrara marble under much higher pressures than those em- ployed in a former investigation on the flow of marble was then studied, the rock being inclosed in tubes of nickel steel, for which the authors are indebted to the Bethlehem Steel Company. Under the very high pressures employed a perfect plastic flow was developed in the marble at the ordinary temperature. Carrara marble was also deformed very quickly under a high pressure and a comparison made between the deformed marble and overstrained metals in respect of the in- fluence of time and of heat in increasing the strength of the deformed rock. As in the case of steel, it was found to become stronger by simple lapse of time. The influence of rate of deformation upon the strength and character of the marble was also studied. The work was then extended to the deformation of limestones containing various impurities, such as sand, clay, bituminous matter, etcetera, some of them being highly fossiliferous. The paper concludes by presenting the results of the study of the deformation of series of dolomites that from the Beaver Dam quarries, Cockeysville, Maryland, being most thoroughly studied, and with a description of some preliminary expe- riments on the deformation.of granite. The paper was discussed by J. F. Kemp, C. K. Leith, and A. P. Coleman. Session oF THurspay Eveninc, DeceMBER 29 At 6.80 o’clock the annual dinner was served at the Hotel Walton, with ladies and other guests present. At 8.30 o’clock the Society met in formal session in a parlor of the 566 PROCEEDINGS OF THE PHILADELPHIA MEETING Hotel Walton, and the President of the Society, John C. Branner, deliy- ered an address entitled STONE REEFS ON THE NORTHEAST COAST OF BRAZIL The address is printed as pages 1-12 of this volume. Following the presidential address a social reunion was held in the same room. SESsION OF Fripay, DECEMBER 30. The Society met at 10 o’clock a m, President Branner in the chair. The Council report was taken from the table and adopted without debate. AUDITING COMMITTEE’S REPORT The Auditing Committee reported that all the accounts of the Treas- urer had been found correct, and the report was adopted. Announcements relating to the program and administrative details were made by the Secretary. An invitation was read from the Directors of the Zoological Society of Philadelphia to visit the Zoological Garden, inclosing cards of ad- mission. The scientific program was declared in order, and the first three papers were presented by the same author, as follows: PRESENT CONDITION OF MONT PELE* BY EDMUND OTIS HOVEY [ Abstract ] The condition of Mont Pelé during the year 1904 was described as being one of continued activity, with occasional comparatively heavy outbursts. The form of the dome surmounting the volcano has suffered much change from time to time, several large needles having at times formed the profile, sometimes with and some- times without a predominating needle on the site of the great spine of 1902 and 1903. During September the main mass of the dome rose, while the altitude of the extreme summit remained stationary through compensating losses from the top. Early in January the remains of the obelisk, or spine, of the spring of 1903, which was destroyed in. the summer and fall of that year, were reported to be rising again with reference to the other portions of the dome. Jn October, however, the dome lost about 100 meters of its altitude, which is 40 meters more than the gains of the preceding ten months. This loss was followed on October 10 by a loss of about 10 meters from the top of the terminal tooth of the dome. * December, 1904. BULL. GEOL. SOC. AM. , VOL. 16, 1904, PL. 92 Figure 1.—Summitr or Mont Pert Looking southwest from the basin of the Lac des Palmistes. October 18,1904. - The central summit is the residue of the great spine of March, 1903. This view was taken from nearly the same spot as that published by the author in the American Journal of Science, October, 1903. Figure 2.—Summit or Mont PELf From the northern rim of the ancient crater. October 18, 1904. This view is at an angle of about 45 degrees from the preceding, and shows the serrate character of the summit of the new dome. Photographs by L. Guinoiseau. SUMMIT OF MONT PELE PRESENT CONDITION OF MONT PELE 567 In November the great spine was observed to be slowly rising again, a move- ment which was checked in the latter part of the month. The official reports, as published in the “‘ Journal Officiel de la Martinique,’’ from which the data printed in this notice were obtained, are somewhat obscure, but it seems that there has been a net loss of about 80 meters during the year. The valley between the new cone and the remains of the ancient crater, which was estimated to be about 200 feet deep in March, 1903, has been greatly reduced in depth by the debris thrown into t by the spasmodic partial demolition of the rising dome. Frequently luminous points were observed in the dome at night, and the almost constant gentle discharge of steam was varied at irregular intervals by explosions of reddish or whitish dust-laden steam clouds. Some of these clouds descended into the upper valley of the Riviére Blanche, others toward Précheur, and a few expended their force toward the east, over the site of the Lac des Palmistes. On April 26 the ascending column of steam sometimes attained an elevation of 3,000 meters. An equal altitude was attained by the steam clouds in the latter part of September and again in November and December. The author discussed briefly the origin of the dome and spine, referring to the three theories that have been advanced to account for it. He dismissed the idea that it had been formed through the piling up of plastic fragments,* by calling attention to (1) The shape of the mass and its component parts as not being those of an exogenous cone; (2) To the appearance of the whole, which was and is that of a relatively homo- geneous though much rifted rock-mass ; (8) To the corrugated northeastern eas of the spine of 1902-1903, which pre- served vertical striations probably due to friction against the wall of the conduit ; (4) To the fact that the scores of major and minor eruptions which had been observed had uniformly thrown their ejecta beyond the rim of the great crater surrounding the doine or down the debris slopes into the valleys of the Blanche, Claire, La Mare, and Précheur rivers. The author considers untenable the second theory also, which is that the dome and spine are the elevated plug of ancient lava supposed to have closed the con- - duit before the activity of 1902 began.t The volcano of mont Pelé is a composite affair, made up of beds of solid lava and fragmental debris (tuff), the latter by far predominating. It is entirely probable that some of the previous eruptions have been of the character of the present outbreak and that some of the massive lava beds now in evidence originated as massive solid exudations. The causes of the inception and cessation of volcanic activity are not sufficiently well known to assert that a plug of solid lava necessarily fills the conduit of a volcano at the quiescence of a vent. The total elevation of dome and spine, had no losses occurred, would probably have amounted to not less than 6,000 feet above the bottom of the old crater. This, according to the ‘‘ elevated plug” theory, would predicate the presence of a stopper of cold or, at any rate, solid lava at least 6,000 feet in vertical measuremeut. The *T.A. Jaggar, Jr.: The initial stages of the spine on Pelée. Am. Jour. Sci., iv, xvii, 39. Jan. 1904. +A Heilprin: The tower of Pelée. 1904. J. B. Lippincott Co. 568 PROCEEDINGS OF THE PHILADELPHIA MEETING probable cross-section of the conduit or system of conduits of eruption was about 1,500 feet in diameter. Such a vast massof rigid lava certainly would offer greater resistance to ascending eruptive forces than would the surrounding fragmental beds, and it would not be lifted bodily into the air. A new vent would more probably be opened beside the old plug or the rising heat would melt its way through the plug. A violent eruption like that of Mont Pelé on May 8, 1902, would certainly have cleared the conduit of the volcano or at any rate have given a practically free course to the rising eruptive material. The clearing of the conduit had been going on gradually, but with increasing violence, for some weeks before the catastrophic explosion occurred which overwhelmed Saint Pierre. | If the ancient conduit were occupied with congealed lava prior to 1902, the re- vived voleanic agents found their way through the plug or ejected it in fragments into the air before the massive dome and spine began to form. Such action might or might not leave a tubular ring of solid lava around the new conduit. If such a tube were left between the conduit and the main tuff beds of the mountain, it hardly seems possible that the eruptive forces could get under it or between the tube and the matrix in such manner as to lift the shell bodily into the air. If no tube of solid lava were left, there would be no shell of ancient lava to be elevated as even the remnants ofa plug. The positive arguments, however, in favor of the third theory are more convincing than the negative arguments which have been advanced against the second hypothesis. } The third theory is that the dome and spine consisted of lava which solidified and became rigid as it was extruded or else in the upper part of the conduit, at least to such an extent that no flow of lava resulted. This is the theory which the author has advocated* and which accords with that advanced by Professor Lacroix,t who first recognized the fact that the new cone of eruption was mainly massive and not fragmental in character. More recently the theory has been elaborately and ably defended by Israel C. Russell.t- The lava of Pelé is andesitic in type and therefore rather difficult to fuse, The result is a viscous stream at the ordinary temperature and under the usual condi- tions of eruptions. In some of the preceding eruptions of Pelé streams of molten lava have issued from the vents and flowed short distances; in most, however, there has been sufficient excess of water vapor present to produce entirely explosive erup- ‘tions. The presence of expanding gas and vapor lowers the temperature of the containing fluid, the effect increasing with increase in the proportionate amount of gas and vapor present. The sudden diminution of pressure at the vent of the volcano produced corresponding expansion of the occluded gases and rapid lower- ing of temperature, with resulting greater or less solidification.§ At Pelé, after the outburst of May 8, 1902, the balance between forces was so nicely adjusted that the lava welling up out of the conduit congealed as it rose and formed the famous dome and spine instead of flowing down the mountain in a stream. The lava near the walls-of the conduit would naturally be cooler and therefore more viscous than that near the center of the rising column. The result would be the freer rise of gases and vapor through the center than through the sides of the *C, R. IX Congr. Geol. int. de Vienne, 1903, pp. 724, 734. + Comptes Rendus, 27 Oct., 1902. t Science, n.s., vol. xxi, p. 924. 2See A. C. Lane, in Science, n. s., vol 18, p. 760, and G. K. Gilbert, in Science, n. s., vol. 19, p. 927. “4 7 Spey yt 42 az IS - @Y} JO OpIs ey} WOIJ [BI1e}8UI Opl|s-pus] Jo dn opew eq 0} SWeeS SseU OYY JO4IBG “B19}00}0 ‘oplydjns snotioy ‘uiNn|e ‘InYyd[ns YIM poyeudeidu (901 pesodwodap) PNW poyIpl[os Jo Zuystsuoo ‘ySty 1997 Og ‘punoul oY} SI 4Y SII OY} OF, “4JO] 94} 0} OLB SojOUBUNY puB ssutids Surjlog jedioulig “S061 “LI YoIRW “4YseMysioU Saryoo'y VIONT LNIVS 4O 3uYuaIYINOS 86 “Id ‘bO6L ‘91 “IOA “WY “90S *1035 *11Na PRESENT CONDITION OF MONT PELE 569 mass, whether there were a constantly open vent or vents there or not. The activ- ity was not central within the old crater, but was most intense in the northwestern quarter. This eccentricity allowed the lava of the southern portion of the conduit to exude rather continuously and solidify as a very steep-sided cone which was disturbed perhaps only during the heavier eruptions. The eccentricity of the maximum activity seems also to account for the northern portion of the dome rising faster than the southern. The relief of pressure at the orifice of the general conduit permitted expansion of the occluded gases and vapor, which increased the cross-section of the dome with reference to the conduit and maintained it in posi- tion, even when there was little pressure from below or rise of material. In the extreme northern part of the crater there is an area of new rock which, like the southern portion, has not risen as rapidly or as much as the subcentral part bear- ing the spine. The spine-bearing portion of the dome seems to have been com- paratively independent of the other two rock portions of the dome and probably. was separated from them by actual or potential fracture zones. SOUFRIERE OF SAINT LUCIA* BY EDMUND OTIS HOVEY [ Abstract] The island of Saint Lucia is entirely volcanic in origin. The rocks and tuffs of the island have suffered extreme subaereal alteration; erosion is in an advanced stage, and volcariic eruptions seem to have ceased on Saint Lucia long before they did on any other of the inner line of the Caribbees. No semblance of a volcanic cone or crater has been reported from any part of the island, but the northern end of the island might repay search for recognizable remains of strato-volcanoes, as indeed has been suggested by Sapper. Warm springs are known to occur in sev- eral places on the island, but the only important locality is one called the Soufriére, which lies about a mile east of the northern of the two remarkable pinnacles of volcanic rock, which rise precipitously from the seaabout 13 miles south of Castries and are known as the Gros and the Petit Piton. This fumarole area covers 2 or 3 acres of the Ventine estate near the head of one branch of the valley which dis- charges at the little town ef Soufriére on the west coast a mile or more north of the Petit Piton. At the time of the author’s visit, March 16-18, 1903, there were within the area eight distinct boiling springs, one vigorous blow hole, which was driving out with much force and noise the water flowing into it from a little stream, many less vig- orous vents, and numberless quiet fumaroles. The water was thoroughly boiling in the large springs and was scalding hot in the smaller ones. During the rainy season there is much more water in the springs than there was at the time of the author’s visit, the temperature of the springs is lower, and the activity less violent. The pools vary in size up to 30 feet in diameter. The vapor rising from these vents is steam charged with varying quantities of hydrogen sulphide. Crystalline sulphur is forming around most of the fumaroles and in the cavities of the porous deposits from the springs which are permeated by the volcanic gases. Ferrous sulphide blackens the mud of many of the springs, particularly the quieter ones. Alum forms in places from the discharges and min- * Based upon field observations made for the American Museum of Natural History. 570 PROCEEDINGS OF THE PHILADELPHIA MEETING gles with the sulphur. Kaolin results from the decomposition of the feldspars of the original rocks. . The sulphur springs area lies on the northaadt flank of the Soufriére mountain, and seems not tobe the remains of or connected in any way with an ancient crater. It is not always easy, or perhaps even probable, to determine the exact position of beds which are so deeply decomposed and so heavily covered with vege- tation as these of Saint Lucia, but the convex character of the ridge on which the sulphur springs area is located and the general relations of the mountains to one another seem to preclude the idea of the existence of a crater at this point. Some effort has been made to mine the sulphur deposited here, but the enterprise has not been a commercial success. _ The paper was illustrated with lantern slides. BOILING LAKE OF DOMINICA * BY EDMUND OTIS HOVEY [ Abstract | The island of Dominica is entirely volcanic in origin, but high-level marine beaches and benches, coral beds, and river terraces indicate comparatively recent elevation amounting to not less than 500 feet. Dominica is one of the largest and highest of the Lesser Antilles. The altitude of the land mass produces copious pre- cipitation, and the surface consequently is well dissected. The rivers for the most part are still deepening their V-shaped channels, and only the larger streams, like the Roseau, the Layou, and some others, have flood-plains. Marine erosion has advanced more rapidly than stream erosion and bluffs form the major portion of the coast. Many warm and hot springs aad fumaroles occur in different parts of the island, and residents report the existence of volcanic craters in Morne Diablotin, the cul- minating point of the island (4,747 feet), and elsewhere in the mountains. The best known and most important center of volcanic activity is the Grande Soufriére, about 6 miles east-northeast of the town of Roseau. Here is situated the famous Boiling lake of Dominica. The Grande Soufriére was in feeble eruption for a brief period in 1880, ashes being thrown as far as Roseau. The whole fumarole area of the Grande Soufriére is nearly a mile in longest diameter, from southwest to northeast, and occupies a vast amphitheatrical depres- sion in the northeast flank of Watt mountain. The amphitheater is breached to the base toward the southeast, and resembles in form the present craters of Mont- serrat and Nevis and that of Mont Pelé before the eruptions of 1902 began. Within the great crater-like depression there are many hot springs and steam vents, most of which arrange themselves into four groups. The principal of these springs or groups is the Builing lake, lying in the extreme northeastern portion of the great amphitheater. The Boiling lake is a circular pool of boiling water 50 or 60 yards across, in the bottom of a pot-like basin, which may be 100 yards in diameter at top. The sur- face of the lake is abuut 560 meters (1,837 feet) above the sea by uncorrected aneroid measurement. Two beaches, about 2 and 5 feet above the water at the * Based upon field observations made for the American Museum of National History. BULL. GEOL. SOC. AM. VOL. 16, 1904, PL. 94 Figure 1.—GRANDE SOUFRIERE OF DOMINICA Near view of one of the largest of the boiling springs. March 17, 1903. This is a true spring the bowl of which is about 30 feet in diameter. No surface water was flowing into the bowl at the time of the author’s visit, and there was a streamlet flowing out of it. > ". et «$e 7 ss Figure 2.—Boining LAKE OF DomINiIca Southern portion of basin beside outlet. Shows indistinct terraces and the composition ol the wall. Photographs for the American Museum of Natural History by E, O. Hovey SOUFRIERE AND BOILING LAKE OF DOMINICA BOILING LAKE OF DOMINICA 571 time of the author’s visit (April 5, 1903), show that the lake stands at higher levels during rainy seasons, with corresponding increase in diameter. Twosmall streams of cool water flow into the basin from the north-northwest, while a single stream discharges the overflow of the pool toward the south-southeast. There are two centers of ebullition, one of gentle action in the northeastern quar- ter of the pool and one of great vigor in the western third. Over the latter the water rises in a dome about 2 feet higher than the general level of the lake, and occasional jets throw water 5 or 6 feet into the air. The quantity of water ordi- narily rising through the two conduits indicated by the centers of ebullition seems to be rather small, judging from a comparison between volumes of the inflowing and outflowing streams. It probably varies with the seasons. The gas rising from the Boiling lake is mostly water vapor (steam), which must have a temperature of fully 100 degrees Centigrade at the vents, since the temper- ature of the brook leaving the pool is 87 degrees to 88 degrees Centigrade, as determined by Sapper. A strong odor of sulphur gas (seemingly H,S) pervades the steam. The water of the lake is bluish milky white fram the particles of pre- cipitated sulphur which it carries in suspension. Carbon dioxide also seems to be present in the discharge from the vents, to judge from the death by asphyxiation of the Englishman, Mr Clive, and his porter on the border of the pool in Decem- ber, 1901. Sometimes, according to Matson Rolle, the guide, when the lake is very full, the strong boiling ceases and bubbles of gas and steam rise copiously from all parts of the lake. On days when such conditions prevail the lake is particularly dangerous to approach. The basin has been formed by the action of the fumaroles and spring water, in conjunction with the surface water, undermining and removing the slide rock and debris from the mountain. Mud is thrown out occasionally from the lake, but the walls of the basin seem to have been built up by ejected materials. Matson Rolle and other residents of the colony state that during exceptionally dry seasons the streams disappear and the basin is empty, except for periodic inflows through the fumarolic vents in its bottom. The basin, therefore, is shallow, a fact that was. indicated also by the shape of that part of the bottom visible at the time of the: author’s visit. The walls of the basin rise about 15 feet above the water at the: outlet, about 60 feet on the west side of the pool, and somewhat higher on the east side. They are vertical or nearly vertical, but the borders of the pool can be easily attained along either the inflowing or outflowing stream. The intermittent rising and falling of water in the conduits during times when the basin is dry, observed by Mr Bell, administrator of the colony, and reported by Sapper, points toward geyser action, but the fountain-like discharge of typical geysers is absent. The paper was illustrated by means of lantern slides. The three papers by Doctor Hovey were discussed as one by Angelo Heilprin, I. C. Russell, and the author. LX XI—BuLL. Grou. Soc. Am., Vor. 16, 1904 572 PROCEEDINGS OF THE PHILADELPHIA MEETING The fourth paper was presented in abstract by the Secretary, the author being absent, as follows: NITROGEN GAS WELL AT DEXTER, KANSAS BY ERASMUS HAWORTH [ Abstract] The well, 400 feet deep, is in Cowley county, southern Kansas, the top being in Permian strata and the gas-bearing sandstone lying near the division between the Lower Permian and the Upper Coal Measures. The static pressure of the gas is 120 pounds and the flowage capacity is 5,000,000 cubic feet daily. An amount of gas equal at atmospheric pressure to 125,000,000 cuble feet has been allowed to flow from the well. The following table of its composition was made after the gas had been allowed to freely flow for eleven days through an 8t-inch pipe and for three days more through a 3-inch pipe. CAL HOM AIO KEAS... 53k. cecadakowis tehcoene sts tate scetseuo cea ceesaactsldsedacdaseocs desne bate resents 0.00 Carbon monoxide......... 0.00 CRY POM eis cs cacoccccurcccvacecectancetesuoss@e'cacoshenclawecduanbswebacseeubecee hemes cess ce estas teats 0.20 FRY GTO ROB Me ls clc deeb scuba Gacceundece sels deenneen cocteehdeddueed Greeeceaetah ttsavee oe eeeee veeeeeenee 0.80 Methane (CH 4) seiissccsccuse ito osesadenctonscuccdaceneses tae der ee digdnesusecenauscecteskteseeetee ee eee 15.02 INDET OREM e; leveastnnees cctaesieadaetecessch fete gankhe ros qudeade danwedcsiigweuarataneanasta= ose 71.89 INET MOSTAR WO cc secs crcoind:ase'se te sedenecctes ceasvaceresensnesce aon we gatever ad. dete euee ee eee eae 12.09 Mota se 2ccciak erates ae dation tobias oevcaed cesar case cnawbecs gutenasenaser tee AO Ate eee 100.00 The full paper is published in Science, volume xxi, February 3, 1905, pages 191-193. The fifth paper was PIEDMONT DISTRICT OF PENNSYLVANIA BY F. BASCOM The paper was discussed by B. K. Emerson, Persifor Frazer, C. D. Walcott, and the author. It is printed as pages 289-328 of this volume. The sixth paper was then read, as follows: CORRELATION OF MARYLAND AND PENNSYLVANIA PIEDMONT FORMATIONS BY EDWARD.B. MATHEWS The paper is printed as pages 329-346 of this volume. The President called Vice-President H. 8. Williams to the chair. The seventh and last paper of the morning session was read: COCKEYSVILLE MARBLE BY EDWARD B. MATHEWS AND W. J. MILLER This paper is printed as pages 347-366 of this volume. - SECTION OF PETROGRAPHY 573 The Society adjourned for the noon recess. On reassembling at 2.10 o’clock, Vice-President H. 8S. Williams in the chair, it was announced that the papers in petrography would be presented in a separate section, to meet in room 104 of the same building. The general program was continued in room 116 and the first paper was EFFECT OF CLIFF EROSION ON FORM OF CONTACT SURFACES BY N. M. FENNEMAN The paper is printed as pages 205-214 of this volume. The second paper was DRAINAGE FEATURES OF CENTRAL NEW YORK BY RALPH 8S. TARR Remarks on the subject of the paper were made by G. K. Gilbert, W.M. Davis, and the author. The paper is printed as pages 229-242 of this volume. The third paper was HANGING VALLEYS BY ISRAEL C. RUSSELL This paper is printed as pages 75-90 of this volume. The fourth and last paper of the day was the following : ICE EROSION THEORY A FALLACY BY H. L. FAIRCHILD The paper was discussed by J. W. Spencer, A.C. Lane, A. P. Coleman, I. C. Russell, G. K. Gilbert, W. M. Davis, and the author. The paper is printed as pages 13-74 of this volume. The Society adjourned. No evening session was held. SECTION OF PETROGRAPHY During Friday afternoon, while the papers described above were pre- sented in the general session the papers classified on the printed program as petrographic were presented in room 104. The temporary section was called to order by J. P. Iddings, and by viva voce vote B. K. Em- erson was made Chairman and E. O. Hovey Secretary. 574 PROCEEDINGS OF THE PHILADELPHIA MEETING The first paper presented was GARNET CONTACT ZONES AND ASSOCIATED COPPER ORES AT SAN JOSE, TAMAULIPAS, MEXICO BY J. F. KEMP The paper was discussed by J. P. Iddings, H.S. Washington, Whitman Cross, and A. C. Lane. It will be published in the transactions of the American Institute of Mining Engineers. The second paper was | OCCURRENCE AND DISTRIBUTION OF CELESTITE-BEARING ROCKS BY EDWARD H. KRAUS [ Abstract ] A study of the rocks—shales and dolomitic limestones—of the upper portion of the Salina epoch in central New York shows that celestite occurs quite widely dis-’ seminated throughout them (1) in the form of well developed crystals and (2) in small circular spots. The celestite was no doubt deposited simultaneously with the rock material. The rocks on the island of Put-in-Bay, lake Erie, and in southern Michigan, especially those at the Maybee quarry, Monroe county, show a similar occurrence of celestite. When celestite-bearing rocks are leached by the action of circulating water the celestite passes quite readily into solution, and the rock then assumes a porous character; in this manner the so-called ‘‘ vermicular limestones ”’ of New York, and also the ‘‘gashed” and ‘‘acicular’”’ dolomites of Michigan, may beexplained. Such celestite-bearing rocks are also the source of the large deposits of celestite which are so abundant on the islands of lake Erie—especially Put-in-Bay— and at the Maybee quarry, Monroe county, Michigan. The paper was discussed by A. C. Lane, A. P. Coleman, E. H. Kraus, G. P. Merrill, and the Chairman. It is published in full in American Journal of Science, volume xix, pages 286-293, April, 1905. The third paper was LIMESTONES OF CENTRAL AND SOUTHERN CALIFORNIA BY T. C. HOPKINS Remarks were made by J. S. Diller. ; The fourth paper was ORIGIN OF VEINS IN ASBESTIFORM SERPENTINE BY GEORGE P. MERRILL ~The paper was discussed by A. P. Coleman, J. P. Iddings, F. D. Adams, J. F. Kemp, and the Chairman. It is printed as pages 131-136 of this volume. ee Coli, Ey, SO. wih: KiX, NO: w: nouvelle serie, livr. 5-9. oe i ‘¢ livr. 10-18. Bulletin, vol. 22. Region aurifére d’lénissei, feuilles K 7-8, L 6, 8-9 and texts. oa = de Lena, feuille II 2, with text. c¢ ce ¢¢ RUSSISCH KAIS. MINERALOGISCHEN GESELL- SCHAFT, ST PETERSBURG Verhandlungen, band xli, lief. 1-2. Materialen zur Geologie Russlands, band xxi, lief. 2. 2 3 e ui ex er of GEOLOGISKA BYRAN, ‘STOCKHOLM Sveriges Geologiska Undersokning, Series C, nos. 195-6. ‘ ‘ ef ‘¢ Ac, nos. 5 and 8. a cs ‘i ‘tA, nos, LISn eat, ok a cf ‘« Aa, nos. 124, 127-8. ee a Ps ‘* Al, no. al, with map. GEOLOGISKA FORENINGENS, STOCKHOLM Porous band xxv, hafte 6, 1903. ‘* xxvi, hafte 5-7, 1904. < “xxvii, hafte 1-3, 1905. NEUES JAHRBUCH FUR MINERALOGIE, STUTTGART nlenes Jahrbuch, 1904, band i, heft 3. zy, 1904, band ii, heft 1-3. ae ea 1905, band i, heft 1-2 Centralblatt, 1904, nos. 9-24. Centralblatt, 1905, nos. 1-9. KAISERLICH-KONIGLICHEN GEOLOGISCHEN REICHANSTALT, VIENNA Jahrbuch, band liii, 1904, and band liv, 1905. KAISERLICH-KONIGLICHEN NATURHISTORISCHEN HOFMUSEUMS, VIENNA Annalen, band xix, Nr. 1-3, 1904. 2137. 2334. 2633. 2634. 1515. 2527. 2630. 2317. 2032. 2344. 2651. 2463. 2138. 2697. 2603. 2044. 2706. 2713. 2714. 2702. 2492. 2597. 2674. 2677. ACCESSIONS TO LIBRARY 601 (c) ASIA GEOLOGICAL SURVEY OF INDIA, CALCUTTA Memoirs, vol. xxxii, part 4. - ce vol. xxxv, part 3. vol. xxxvi, part l. Records, vol. xxxi, parts 1-3. “e IMPERIAL GEOLOGICAL SURVEY, TOKYO (d) AUSTRALASIA GEOLOGICAL DEPARTMENT OF SOUTH AUSTRALIA, ADELAIDE Record of the Mines, Supplementary Report. Review of Mining Operations in S. A. during 1904. GEOLOGICAL SURVEY OF QUEENSLAND, BRISBANE Reports 193-195. CANTERBURY MUSEUM, CHRISTCHURCH DEPARTMENT OF MINES OF VICTORIA, MELBOURNE Annual Report of the Secretary of Mines for 1903. Bulletins, nos. 13-17. Records, vol. i, part 3. Diagram showing yield of gold per annum, etcetera. GEOLOGICAL DEPARTMENT OF WESTERN AUSTRALIA, PERTH Bulletins, nos. {1-14. Annual Progress Reports for 1901 and 1903. Bulletins, nos. 16-17. GEOLOGICAL SURVEY OF NEW SOUTH WALES, SYDNEY Annual Report of the Department of Mines for 1903. Records of the Geological Survey, vol. vii, part 4. Memoirs, Paleontology, no. 13, part 1. Records, vol. viii, part 1. Annual Report of the Department of Mines for 1904. ROYAL SOCIETY OF NEW SOUTH WALES, SYDNEY Journal and Proceedings, vol. xxxvii, 1903. (e) AFRICA GEOLOGICAL COMMISSION, CAPE TOWN Annals of the South African Museum, vol. iv, parts 4-6. Annual Report for 1900. Index to Annual Reports, 1896-1903. GEOLOGICAL SOCIETY OF SOUTH AFRICA, JOHANNESBURG Transactions, vol. vii, parts 1-3. 602 2500. 2637. 2667. 2675. 2612. 2611. 2719. 2720. 2555. 2203. 2338. -2531. 2721. 2616. 2617. PROCEEDINGS OF THE PHILADELPHIA MEETING GEOLOGICAL SURVEY OF THE TRANSVAAL, . Report for the year 1903. GEOLOGICAL SURVEY OF NATAL AND ZULU- LAND, - PIETERMARITZBURG . Second Report of the Geological Survey, 1904. (f) HAWAIIAN ISLANDS HAWAIIAN GOVERNMENT SURVEY, HONOLULU (B) From StatE GEOLOGICAL SURVEYS AND MINING Bone UNIVERSITY OF TEXAS MINERAL SURVEY, + AUSTIN Bulletins nos. 8-9. GEOLOGICAL SURVEY OF OHIO, COLUMBUS | Bulletin no. 2, 4th series. Report on the Ohio Cooperative Paecer ine Survey. Bulletin no. 3, 4th series. DEPARTMENT OF THE INTERIOR, OTTAWA Topographic Maps; Windsor sheet, Ontario; map showing Stations of Mounted Police in Northwest Canada; map showing ‘Railways in Manitoba, Saskatchewan, Alberta, etcetera. NEW JERSEY GEOLOGICAL SURVEY, TRENTON Final Report, vol. vi, 1904; Clay Industry. GEOLOGICAL SURVEY OF ALABAMA, UNIVERSITY Index to Mineral Resources of Alabama. (C) From Screntiric SocleTIES AND INSTITUTIONS © (a2) AMERICA BROOKLYN INSTITUTE OF ARTS AND SCIENCES, BROOKLYN Science Bulletin, vol. i, no. 4. COLORADO COLLEGE, COLORADO SPRINGS Science series, Colorado College Studies, vol. xi, nos. 36-38. SOCIEDAD CIENTIFICA “ANTONIO ALZATE,”’ MEXICO Memorias y Revista, tomo xiii, num. 7-8. a be -* xix, num. 8-12. a re: ‘ek aM 0-12; PROVINCE OF ONTARIO, TORONTO First Report of the Bureau of Archives, 1903. (b) EUROPE SCHLESISCHE GESELLSCHAFT FUR VATERLANDISCHE CULTUR, BRESLAU 8lst Jahresbericht. Die Hundertjahrfeier, etcetera. PRETORIA ; ’ of ’ ' 2699. 2722. 2723. 2724. 2725. 2726. ACCESSIONS TO LIBRARY 603 OBSERVATOIRE ROYALE DE BELGIQUE, BRUSSELS Annuaire Astronomique pour 1906. NATURFORSCHER-VEREINS ZU RIGA, RIGA Korrespondenzblatt, xlvii, 1904. R. ISPETTORATO DELLE MINIERE, ROMA Catalogo della Mostra fatta dal Corpo Realle delle Miniere all’ Espo- sizione Universale di Saint Louis nel 1904. SECTION GEOLOGIQUE DU CABINET DE SA MAJESTE, ST PETERSBURG Travaux, vol. vi, livr. 1. (c) ASIA TOKYO GEOGRAPHICAL SOCIETY, TOKYO Journal of Geography, vol. xvi, 1904. IMPERIAL UNIVERSITY OF TOKYO, TOKYO Journal of the College of Science, vol. xx, articles 2 and 5. (D) From FELiows oF THE GEoLogicaL SoctrTy oF AMERICA (PERSONAL 2727. 2728. 2729. 2730. 2731. 2732. 2733. 2734. 2738, 2736. 2737. 2738. 2739. 2740. PUBLICATIONS) H. L. FAIRCHILD Glacial Waters from Oneida to Little Falls. Geology under the Planetesimal Hypothesis of Earth-origin. C. H. GORDON On the Origin and Classification of Gneisses. On the Paramorphic Alteration of Pyroxene to common Hornblende. Pyroxenites of the Grenville Series in Ottawa county, Canada. C. H. HITCHCOCK Glaciation of the Green Mountains. New Studies in the Ammonoosuc District of New Hampshire. R. G. McCONNELL Report on the great Landslide at Frank, Alta. JOSEPH HYDE PRATT Ten Separates on the Production of Minerals in 1903. ISRAEL C. RUSSELL The Topographic Survey of Michigan. Cooperation among American Geographical Societies. Research in State Universities. Physiographic Problems of Today. SAMUEL WEIDMAN Baraboo Iron-bearing District. Bull. xiii, Wisconsin Geological Survey. LXXIII—Butt. Grou. Soc. Am., Vor. 16, 1904 (£) From Misce,tangous Sources — G. SIMOENS, 2741. Réponse aux Critiques formulées par M. Emm. de M la Bibliographica Geologica. MICHEL MOURLON, : FLORENTINO AMEGHINO, _ 2743. Paleontologia Argentina, no. 2. ALEXANDER V. KALECSINSZKY, 2744. Uber die Akkumulation a> Sp in verschiedenen ‘ta keiten. ‘Soh FLORENTINO AMEGHINO, 2743. Paleontologia Argentina, no. 2. . MINING MAGAZINE fers 2745-2746. Mining Magazine, vol. x, nos. 1 and 5; vol. xi, no. 3. G.-F. DOLLFUS ET G. RAMOND, 2747. Etudes géologiques dans Paris et sa Banlieue, iv. WIRT TASSIN, 2748. The Persimmon Creek and Mount Vernon Meteorites. OFFICERS AND FELLOWS OF THE GEOLOGICAL SOCIETY OF AMERICA OFFICERS FOR 1905 President RAPHAEL PuMPELLY, Dublin, N. H. Vice-Presidents SAMUEL CALVIN, Iowa City, Iowa W. M. Davis, Cambridge, Mass. Secretary H. L. Farrcninp, Rochester, N. Y. Treasurer I. C. Waitr, Morgantown, W. Va. Editor J. STANLEY-Brown, Washington, D. C. Tibrarian H. P. Cusnine, Cleveland, Ohio Councillors (Term expires 1905) R. D. Sauispury, Chicago, Ill. J. E. Woirr, Cambridge, Mass. * (Term expires 1906) JoHN M. CriarkE, Albany, N. Y. GrorcE P. Merritt, Washington, D. C. (Term expires 1907) H. M. Amt, Ottawa, Canada J. F. Kemp, New York city (605) 606 PROCEEDINGS OF THE PHILADELPHIA MEETING | FELLOWS IN DECEMBER, 1905 *Indicates Original Fellow (see article III of Constitution) CLEVELAND ABBE, JR., Ph. D., 1441 Florida Ave. N. W., Washington, D. C. Au- gust, 1899. FRANK Dawson ApdAmMs, Ph. D., Montreal, Canada; Professor of Geology in McGill University. December, 1889. GrEorRGE I. ADAMS, Se. D., Corps of Mining Engineers, Lima, Peru. December, 1902. JoS& GUADALUPE AGUILERA, Esquela N. de Ingeneiros, City of Mexico, Mexico; Director del Instituto Geologico de Mexico. August, 1896. TRUMAN H. AtpricuH, M. E., 1739 P St. N. W., Washington, D. C. May, 1889. Henry M. Ami, A. M., Geological Survey Office, Ottawa, Canada; Assistant Paleontologist on Geological and Natural History Survey of Canada. De- cember, 1889. FRANK M. Anverson, B. A., M. S., 2604 Attna Street, Berkeley, Cal. In Cali- fornia State Mining Bureau. June, 1902. Puitie ARGALL, 728 Majestic Building, Denver, Colo.; Mining Engineer. August, 1896. RatepH ARNOLD, Ph. D., Washington, D. C.; Geologic Aid U. S. Geological Sur- vey. December, 1904. | GEORGE Hatt ASHLEY, M. E., Ph. D., Washington, D. C., U. S. Geological Sur- vey. August, 1895. Harry Foster BAIN, M. S., Champaign, I11., State Geological Survey. Decem- ber, 1895. RuFuS MATHER Baae, Ph. D., Socorro, N. Mex.; Professor of Mineralogy and Petrography, State School of Mines. December, 1896. S. PRENTISS BALDWIN, 736 Prospect St., Cleveland, Ohio. August, 1895. ERWIN HINCKLEY Bargour, Ph. D., Lincoln, Neb. ; Professor of Geology, Univer- sity of Nebraska, and Acting State Geologist. December, 1896. JOSEPH BARRELL, Ph. D., New Haven, Conn.; Asistant Professor of Geology, Yale University. December, 1902. ‘ 4 GEORGE H. Barton, B. S., Boston, Mass.; Curator, Boston Society of Natural History. August, 1890. FLORENCE Bascom, Ph. D., Bryn Mawr, Pa.; Instructor in Geology, Petrography, — and Mineralogy in Bryn Mawr College. August, 1894. WILLIAM S. Baytey, Ph. D., South Bethlehem, Pa. December, 1888. *GEORGE I’. BECKER, Ph. D., Washington, D. C., U. S. Geological Survey. JOSHUA W. BEEDE, Ph. D., Bloomington, Ind.; Instructor in Geology, Indiana University. December, 1902. RosBert BeEwu, C. H., M. D., LL. D., Ottawa, Canada; Acting Director of the Geological and Natural History Survey of Canada. May, 1889. : CHARLES P. BERKEY, Ph. D., New York city; Columbia University. August, 1901. SAMUEL WALKER BEYER, Ph. D., Ames, Iowa; Assistant Professor in Geology, Iowa Agricultural College. December, 1896. . ARTHUR BiBsBins, Ph. B., Baltimore, Md.; Instructor in Geology, Woman’s Col- lege. December, 1903, FELLOWS OF THE SOCIETY 607 ALBERT S. BIckMORE, Ph. D., American Museum of Natural History, New York ; Professor in charge of Department of Public Instruction. December, 1889. Irvine P. BisHop, 109 Norwood Ave., Buffalo, N. Y.; Professor of Natural Science, State Normal and Training School. December, 1899. JOHN ADAMS BownocKer, D. Sc., Columbus, Ohio.; Professor of Inorganic _ Geology, Ohio State University. December, 1904. *JOHN C. BRANNER, Ph. D., Stanford University, Cal.; Professor of Geology in Leland Stanford, Jr., University. ALBERT PERRY BRIGHAM, A. B., A. M., Hamilton, N. Y.; Professor of Geology and Natural History, Colgate University. December, 1893. REGINALD W. Brock, M. A., Ottawa, Canada, Geologist, Geological and Natural History Survey of Canada ; Professor of Geology, School of Mining, King- ston. December, 1904. ALFRED HULSE Brooks, B. S., Washington, D. C.; Assistant Geologist, U. S. Geo- logical Survey. August, 1899. ERNEST ROBERTSON BUCKLEY, Ph. D., Rolla, Mo.; State Geologist and Director of Bureau of Geology and Mines. June, 1902. *SAMUEL CALVIN, Iowa City, lowa; Professor of Geology and Zoology in the State University of Iowa. HENRY DONALD CAMPBELL, Ph. D., Lexington, Va.; Professor of Geology and Biology in Washington and Lee University. May, 1889. Marius R. CAMPBELL, U. S. Geological Survey, Washington, D. C. August, 1892. FRANKLIN R. CARPENTER, Ph. D., 1420 Josephine St., Denver, Colo.; Mining Engineer. May, 1889. ERMINE C. CASE, Ph. D., Milwaukee, Wis.; Instructor in State Normal School. December, 1901. *T. C. CHAMBERLIN, LL. D., Chicago, Ill.; Head Professor of Geology, Univer- sity of Chicago. CLARENCE RAYMOND CLAGHORN, B. S., M. E., Tacoma, Wash. August, 1891. *WILLIAM BULLOCK CLARK, Ph. D., Baltimore, Md.; Professor of Geology in Johns Hopkins University ; State Geologist. JOHN MASON CLARKE, A. M., Albany, N. Y.; State Paleontologist. December, 1897. J. MORGAN CLEMENTS, Ph. D., 11 William St., New York city. December, 1894. COLLIER Coss, A. B., A. M., Chapel Hill, N. C.; Professor of Geology in Univer- sity of North Carolina. December, 1894. ARTHUR P. CoLEMAN, Ph. D., Toronto, Canada; Professor of Geology, Toronto University, and Geologist of Bureau of Mines of Ontario. December, 1896. GEORGE L. CoLuiz, Ph. D., Beloit, Wis.; Professor of Geology in Beloit College. December, 1897. ARTHUR J. CoLLier, A. M., S. B., Washington, D. C.; Assistant Geologist, U. S. Geological Survey. June, 1902. *THEODORE B. Comstock, Sc. D., Los Angeles, Cal.; Mining Engineer. *WPRANCIS W. CRAGIN, Ph. D., Colorado Springs, Colo.; Professor of Geology in Colorado College. ALJA ROBINSON Crook, Ph. D., Evanston, Ill.; Professor of Mineralogy and Economie Geology in Northwestern University. December, 1898. *WILLIAM O. Crossy, B. S., Boston Society of Natural History, Boston, Mass. ; Assistant Professor of Mineralogy and Lithology in Massachusetts Insti- tute of Technology. 608 PROCEEDINGS OF THE PHILADELPHIA MEETING WHITMAN Cross, Ph. D., U. S. Geological Survey, Washington, D. C. May, 1889. GARRY HE. CuLvEr, A. M. 1104 Wisconsin St., Stevens Point, Wis. December, 1891. EpeaR R. CuUMINGS, Ph. D., Bloomington, Ind.; Assistant Professor of Geology, Indiana University. August, 1901. *Henry P. CusHine, M. S., Adelbert College, Cleveland, Ohio; Professor of Geology, Western Reserve University. *NELSON H. Darton, United States Geological Survey, Washington, D. C. *WILLIAM M. Davis, S. B., M. E., Cambridge, Mass.; Sturgis-Hooper Professor of Geology in Harvard University. Davip T. Day, Ph. D., U. S. Geol. Survey, Washington, D. C. August, 1891. ORVILLE A. DerRBy, M. S., Sao Paulo, Brazil; No. 80 Rua Visconde do Rio Branco. December, 1890. * JOSEPH S. DILLER, B. S., U. S. Geological Survey, Washington, D. C. EDWARD V. D’INVILLIERS, E. M., 506 Walnut St., Philadelphia, Pa. Dec., 1888. RicHARD EH. Dover, A. M., Teachers’ College, West 120th St., New York city; Professor of Geography in the Teachers’ College. August, 1897. NoaH FIELDS DRAKE, Ph. D., Tientsin, China; Professor of Geology in Imperial Tientsin University. December, 1898. CHARLES R. Dryer, M. A., M. D., Terre Haute, Ind.; Professor of Geoseagin Indiana State Normal School. August, 1897. *HDWIN T. DUMBLE, Austin, Texas; State Geologist. *WILLIAM B. Dwicut, Ph. B., Poughkeepsie, N.Y.; Professor of Natural History in Vassar College. ARTHUR S. EHAKLE, Ph. D., Berkeley, Cal.; Instructor in Mineralogy, University of California. December, 1899. CHARLES R. EASTMAN, A. M., Ph. D., Cambridge, Mass. ; In Charge of Vetebrate Paleontology, Museum of Comparative Zoology, Harvard University. De- cember, 1895. ARgTHuUR H. EvrrmMan, Ph. D., 706 Globe Building, Minneapolis, Minn. Decem- ber, 1898. *BENJAMIN K. EMERSON, Ph. D., Amherst, Mass. ; Professor in Amherst College. *SAMUEL EF’. EMMowns, A. M., E. M., U. S. Geological Survey, Washington, D. C. JOHN HYERMAN, F. Z. S., Oakhurst, Easton, Pa. August, 1891. HAROLD W. FAIRBANKS, B. S., Berkeley, Cal.; Geologist State Mining Bureau. August, 1892. *HERMAN L. FAIRCHILD, B. 8., Rochester, N. Y.; Professor of Geology in Uni- versity of Rochester. J. C. FAes, Danville, Ky. ; Professor in Centre College. December, 1888. OLIVER C. FARRINGTON, Ph. D., Chicago, Ill.; In charge of Department of Geol- -ogy, Field Columbian Museum. December, 1895. Nevin M. FENNEMAN, Ph. D., Madison, Wis.; Professor of Geology, one of Wisconsin. December, 1904. Aucust F. Forrsts,; Ph. D., 417 Grand Ave., Dayton, Ohio; Teacher of Sciences. December, 1899. WILLIAM M. FontTaAINgE, A. M., University of Virginia, Va.; Professor of Natural History and Geology in University of Virginia. December, 1888. *PERSIFOR FRAZER, D. Sc., 1042 Drexel Building, Philadelphia, Pa.; Professor of Chemistry in Horticultural Society of Pennsylvania. *HoMER T. FULLER, Ph. D., Fredonia, N. Y. FELLOWS OF THE SOCIETY 609 Myron LESLIE Futter, S. B., U. S. Geological Survey, Washington, D. C. De- cember, 1898. Henry STEWART GANE, Ph. D., Santa Barbara, Cal. December, 1896. Henry GANNETT, S. B., A. Met. B., U. S. Geological Survey, Washington, D. C. December, 1891. *GROVE K. GILBERT, A. M., LL. D., U. S. Geological Survey, Washington, D. C. ADAM CAPEN GILL, Ph. D., Ithaca, N. Y.; Assistant Professor of Mineralogy and Petrography in Cornell University. December, 1888. L. C. GLENN, Ph. D., Nashville, Tenn. ; Professor of Geology in Vanderbilt Uni- versity. June, 1900. CHARLES H. Gorpon, Ph. D., University Station, Seattle, Wash. August, 1893. CHARLES NEWTON GOULD, A. M., Norman, Okla.; Professor of Geology, Univer- sity of Oklahoma. December, 1904. AMADEUS W. GRABAU, S. M., S. D., Columbia University, New York city; Pro- fessor of Paleontology. December, 1898. _ULysses SHERMAN GRANT, Ph. D., Evanston, Ill.; Professor of Geology, North- western University. December, 1890. HERBERT E. GREGORY, Ph. D., New Haven, Conn.; Assitant Professor of Physi- ography, Yale University. August, 1901. GEORGE P. GRIMSLEY, Ph. D., Morgantown, W. Va.; Assistant State Geologist, Geological Survey of West Virginia. August, 1895. - Leon S. GRISWOLD, A. B., 238 Boston St., Dorchester, Mass. August, 1902. FREDERIC P. GULLIVER, Ph. D., Norwichtown, Conn. August, 1895. ARNOLD HaGuk, Ph. B., U. 8. Geological Survey, Washington, D.C. May, 1889. *CHRISTOPHER W. Hatt, A. M., 803 University Ave., Minneapolis, Minn.; Pro- fessor of Geology and Mineralogy in University of Minnesota. GILBERT D. Harris, Ph. B., Ithaca, N. Y.; Assistant Professor of Paleontology and Stratigraphic Geology, Cornell University. December, 1903. JOHN BURCHMORE HARRISON, M. A., F. I. C., F. G. S., Georgetown, British Guiana; Government Geologist. June, 1902. JOHN B. Hastines, M. E., 1480 High St., Denver, Colo. May, 1889. *ERASMUS HAworTH, Ph. D., Lawrence, Kans.; Professor of Geology, Univer- sity of Kansas. C. WILLARD HAYES, Ph. D., U. S. Geological Survey, Washington, D.C. May, 1889. *ANGELO HEILPRIN, Academy of Natural Sciences, Philadelphia, Pa.; Professor of Paleontology in the Academy of Natural Sciences. RicHArRD R. Hick, B: S., Beaver, Pa. December, 1903. *EHUGENE W. HILGARD, Ph. D., LL. D.; Berkeley, Cal.; Professor of Agriculture in University of California. FRANK A. HILL, Roanoke, Va. May, 1889. *RoBERT T. Hitt, B. S., U. S. Geological Survey, Washington, D. C. RicHarpD C. HiILys, Mining Engineer, Denver, Colo. August, 1894. _*CHARLES H. HitcHcock, Ph. D., LL. D., Hanover, N. H.; Professor of Geology in Dartmouth College. WILLIAM HersBert Hopes, Ph. D., Madison, Wis.; Professor of Mineralogy and Petrology, University of Wisconsin; Assistant Geologist, U. S. Geological Survey. August, 1891. *LEvI Horprook, A. M., P. O. Box 536, New York city. ARTHUR Ho.tick, Ph. B., N. Y. Botanical Garden, Bronx Park, New York; In- structor in Geology, Columbia University. August, 1893. = 610 PROCEEDINGS OF THE PHILADELPHIA MEETING we *JOSEPH A. HoLMEs, 6017 Cabanne Ave., Saint Louis, Mo.; State Geologist of North Carolina; In charge of investigation of fuels and structural ma- terials, U. S. Geological Survey. THOMAS C. HopxKIns, Ph. D., Syracuse, N. Y.; Professor of Geology, Syracuse University. December, 1894. *KDMUND OTIS Hovey, Ph. D., American Museum of Natural History, New York city; Associate Curator of Geology. *Horace C. Hovey, D. D., Newburyport, Mass. ERNEST Howe, Ph. D., Washington, D. C.; Assistant Geologist, U. 8S. Geological Survey. December, 1903. *KDWIN E. Howe .t, A. M., 612 Seventeenth St. N. W., Washington, D. C. Lucius L. HuspsBarpD, Ph. D., LL. D., Houghton, Mich. December, 1894. JOSEPH P. IppiInes, Ph. B., Professor of Petrographiec Geology, University of Chicago, Chicago, Ill. May, 1889. A. WENDELL JACKSON, Ph. B., 432 Saint Nicholas Ave., New York city. Decem- ber, 1888. | ROBERT T. JACKSON, S. D., 9 Fayerweather St., Cambridge, Mass.; Assistant Professor in Paleontology in Harvard University. August, 1894. THOMAS M. JAcKSON, C. E., S. D., Clarksburg, W. Va. May, 1889. Mark S. W. JEFFERSON, A. M., Ypsilanti, Mich.; Professor of Geography, Michi- gan State Normal School. December, 1904. ALEXIS A. JULIEN, Ph. D., Columbia College, New York city; Instructor in Co- lumbia College. May, 1889. ARTHUR KEITH, A. M., U. S. Geological Survey, Washington, D. C. May, 1889. *JAMES F. Kemp, A. B., EH. M., Columbia University, New York city ; Professor of Geology. | CHARLES ROLLIN KEYES, Ph. D., Socorro, N. Mex.; President State School of Mines. August, 1890. FRANK H. KNOw ttTon, M. S., Washington, D. C.; Assistant Paleontologist, U. S. Geological Survey. May, 1889. EDWARD HENRY KRAUS, Ph. D., Ann Arbor, Mich.; Assistant Professor of Min- eralogy, University of Michigan. June, 1902. Henry B. KuUMMEL, Ph. D., Trenton, N. J.; State Geologist. December, 1895. *GEORGE EF. Kunz, A. M. (Hon.), Ph. D. (Hon.), care of Tiffany & Co., 15 Union Square, New York city. GEORGE EpGaR LAppD, Ph. D., Rolla, Mo.; Director School of Mines. August, 1891. J. C. K. LAFLAMME, M. A., D. D., Quebec, Canada; Professor of Mineralogy and Geology in University Laval, Quebec. August, 1890. ALFRED C. LANE, Ph. D., Lansing, Mich.; State Geologist of Michigan. Decem- -ber, 1889. DANIEL W. LANGTON, Ph. D., Fuller Building, New York city; Mining Engineer. December, 1889. ANDREW C. LAwSOoN, Ph. D., Berkeley, Cal.; Professor of Geology and Miner- alogy in the University of California. May, 1889. . WILLIS Tuomas Lez, M. S., Washington, D. C.; Assistant Geologist, U. S. Geological Survey. December, 1903. CHARLES K. LEITH, Ph. D., Madison, Wis.; Professor of Geology, University of Wisconsin; Assistant Geologist, U. S. Geological Survey. December, 1902. ARTHUR G. LEONARD, Ph. D., Grand Forks, N. Dak.; Professor of Geology and State Geologist, State University of North Dakota. December, 1901. FELLOWS OF THE SOCIETY 611 Frank Leverett, B. S., Ann Arbor, Mich.; Geologist, U. S. Geological Survey. August, 1890. WILLIAM Lipsey, Se. D., Princeton, N. J.; Professor of Physical Geography in Princeton University. August, 1899. WALDEMAR LINDGREN, M. E., U. S. Geological Survey, Washington, D.C. August, 1890. GEORGE Davis LOUDERBACK, Ph. D., Reno, Nev.; Professor of Geology, Univer- sity of Nevada. June, 1902. Rosert H. LoueuripeGr, Ph. D., Berkeley, Cal.; Assistant Professor of Agricult- ural Chemistry in University of California. May, 1889. Tuomas H. Macsripe, A. M., Iowa City, Iowa; Professor of Botany in the State University of Iowa. May, 1889. HiraAM Dreyer McCaskey, B. S., Manila, P. I.; Chief of Mining Bureau of Manila. December, 1904. RIcHARD G. McConne tL, A. B., Geological Survey Office, Ottawa, Canada ; Geol- ogist on Geological and Natural History Survey of Canada. May, 1889. JAMES RIEMAN MACFARLANE, A. B., 100 Diamond St., Pittsburg, Pa. August 1891. *W J McGesr, LL. D., Director Public Museum, Saint Louis, Mo. WittiaAM McInnrs, A. B., Geological Survey Office, Ottawa, Canada; Geolozist, Geological and Natural History Survey of Canada. May, 1889. PETER McKELLAR, Fort William, Ontario, Canada. August, 1890. Curtis F. Marsut, A. M., State University, Columbia, Mo.; Instructor in Geology and Assistant on Missouri Geological Survey. August, 1897. VERNON F.. Marsters, A. M., Bloomington, Ind.; Professor of Geology in In- diana State University. August, 1892. GEORGE CuRTIS Martin, Ph. D., Washington, D. C.; U. S. Geological Survey. June, 1902. EpwarpD B. MATHEWS, Ph. D., Baltimore, Md.; Instructor in Petrography in Johns Hopkins University. August, 1895. WILLIAM D. MATTHEW, Ph. D., New York City; Associate Curator in Vertebrate Paleontology, American Museum of Natural History. December, 1903. P. H. MELL, M. E., Ph. D., Clemson College, S. C.; President of Clemson College. December, 1888. WARREN C. MENDENHALI, B. §., 1108 Braly Building, Los Angeles, Cal.; Geol- ogist U. S. Geological Survey. June, 1902. JOHN C. MERRIAM, Ph. D., Berkeley, Cal.; Instructor in Paleontology in Uni- versity of California. August, 1895. *FWREDERICK J. H. MERRILL, Ph. D., 225 West End Ave., New York city ; Consult- ing Geologist. GEORGE P. MERRILL, M. S., U. S. National Museum, Washington, D. C.; Curator of Department of Lithology and Physical Geology. December, 1888. ARTHUR M. MILER, A. M., Lexington, Ky.; Professor of Geology, State Uni- versity of Kentucky. December, 1897. BENJAMIN L. MILLER, Ph. D., Bryn Mawr, Pa.; Associate in Geology, Bryn Mawr College. December, 1904. WILLET G. MILLER, M. A., Toronto, Canada; Provincial Geologist of Ontario. December, 1902. HENRY MontcomMery, Ph. D., Toronto, Canada; Professor of Geology and Biology, Trinity University. December, 1904. LXXIV—Butt. Geow. Soc. Am., Vou. 16, 1904 612 PROCEEDINGS OF THE PHILADELPHIA MEETING *WRANK L. NASON, A. B., West Haven, Conn. JOHN F. Newsom, A. M., Stanford University, Cal.; Associate Professor of Metallurgy and Mining. December, 1899. WILLIAM H. NILES, Ph. B., M. A., Boston, Mass.; Professor, Emeritus, of Geology, Massachusetts Institute of Technology; Professor of Geology, Wellesley College. August, 1891. WILLIAM II. Norton, M. A., Mount Vernon, Iowa; Professor of Geology in Cor- nell College. December, 1895. CHARLES J. Norwoop, Lexington, Ky.; Professor of Mining, State College of Kentucky. August, 1894. CLEOPHAS C. O’HaRRA, Ph. D., Rapid City, S. Dak.; Professor of Mineralogy and Geology, South Dakota School of Mines. December, 1904. EZEQUIEL ORDONEZ, Esquela N. de Ingeneiros, City of Mexico, Mexico; Geologist > del Instituto Geologico de Mexico. August, 1896. *AMOS O. OSBORN, Waterville, Oneida county, N. Y. Henry F. Osporn, Se. D., Columbia University, New York city; Professor of Zoology, Columbia University. August, 1894. CHARLES PALACHE, B. 8., University Museum, Cambridge, Mass.; Instructor in Mineralogy, Harvard University. August, 1897. *HoRACE B. Patron, Ph. D., Golden, Colo.; Professor of Geology and Mineral- ogy in Colorado School of Mines. FREDERICK B. PEcK, Ph. D., Easton, Pa.; Professor of Geology and Mineralogy, Lafayette College. August, 1901. SAMUEL L. PENFIELD, Ph. B., M. A., New Haven, Conn.; Professor of Mineral- ogy, Sheffield Scientific School of Yale University. December, 1899. Ricuargp A. F. PENROSE, JR., Ph. D., 1331 Spruce St., Philadelphia, Pa. May, 1889. GEORGE H. PERKINS, Ph. D., Burlington, Vt.; State Geologist. Professor of Geol- ogy, University of Vermont. June, 1902. JOSEPH H. Perry, 276 Highland St., Worcester, Mass. December, 1888. Louts V. Pirsson, Ph. D., New Haven, Conn.; Professor of Physical Geology, Sheffield Scientific School of Yale University. August, 1894. * JULIUS POHLMAN, M. D., University of Buffalo, Buffalo, N. Y. JOHN BONSALL PorRTER, E. M., Ph. D., Montreal, Canada; Professor of Mining, McGill University. December, 1896. JOSEPH Hybr Pratt, Ph. D., Chapel Hill, N. C.; Mineralogist, North Carolina Geological Survey. December, 1898. *CHARLES S. PROSSER, M. S., Columbus, foes Professor of Geology in Ohio State University. *RAPHAEL PUMPELLY, U. S. Geological Survey, Dublin, N. H. ALBERT HOMER PERDUE, B. A., Fayetteville, Ark.; Professor of Geology, Univer- sity of Arkansas. December, 1904. FREDERICK LESLIE RANSOME, Ph. D., Washington, D. C.; Assistant Geologist, U. S. Geological Survey. August, 1895. HARRY FIELDING REID, Ph. D., Johns Hopkins University, Baltimore, Md. De- cember, 1892. WILLIAM NorTH Rick, Ph. D., LL. D., Middletown, Conn.; Professor of Geology in Wesleyan University. August, 1890. CHARLES H. RICHARDSON, Ph. D., 884 Elm St., Manchester,.N. H. December, 1899. a FELLOWS OF THE SOCIETY 613 Hernricu Ries, Ph. D., Cornell University, Ithaca, N. Y.; Assistant Professor in Economic Geology. December, 1893. *ISRAEL C. RusSELL, LL. D., Ann Arbor, Mich.; Professor of Geology in Uni- versity of Michigan. *JaAMES M. SAFForD, M. D., LL. D., Dallas, Texas. ORESTES H. St. JoHN, Raton, N. Mex. May, 1889. . *RoLLIn D. Sarispury, A. M., Chicago, Ill.; Professor of General and Geo- graphic Geology in University of Chicago. FREDERICK W. Sarpreson, Ph. D., Instructor in Paleontology, University of Minnesota, Minneapolis, Minn. December, 1892. FRANK C. ScHRADER, M. S., A. M., U. S. Geological Survey, Washington, D. C. August, 1901. CHARLES SCHUCHERT, New Haven, Conn.; Curator, Geological Department, Yale University. August, 1895. WiuuiAM B. Scort, Ph. D., 56 Bayard Ave., Princeton, N. J.; Blair Professor of Geology in College of New Jersey. August, 1892. ARTHUR EDMUND SEAMAN, B. S., Houghton, Mich.; Professor of Mineralogy and Geology, Michigan College of Mines. December, 1904. Henry M. SeEety, M. D., Middlebury, Vt.; Professor of Geology in Middlebury College. May, 1899. *NATHANIEL S. SHALER, LL. D., Cambridge, Mass.; Professor of Geology in Harvard University. GEORGE BURBANK SHATTUCK, Ph. D., Baltimore, Md.; Associate Professor in Physiographie Geology, Johns Hopkins University. August, 1899. Soton SHEppD, A. B., Pullman, Wash.; Professor of Geology and Mineralogy, Washington Agricultural College. December, 1904. EpwWARD M. SHEPARD, A. M., Springfield, Mo.; Professor of Geology, Drury Col- lege. August, 1901. WiLL H. SHERZER, M. S., Yysilanti, Mich.; Professor in State Normal School. December, 1890. BoHUMIL SHIMEK, C. E., M. S., Iowa City, Iowa; Professor of Physiological Botany, University of Iowa. December, 1904. *FREDERICK W. SIMONDS, Ph. D., Austin, Texas; Professor of Geology in Uni- versity of Texas. *EUGENE A. SMITH, Ph. D., University, Tuscaloosa county, Ala.; State Geol- ogist and Professor of Chemistry and Geology in University of Alabama. FRANK CLEMES SMITH, B. S., Harrisburg, Arizona; Mining Engineer. Decem- ber, 1898. GrorGE OTIS SmitH, Ph. D., Washington, D. C.; Assistant Geologist, U. S. Geological Survey. August, 1897. WILLIAM S§. T. SmitTH, Ph. D., Box 30, Los Gatos, Cal. June, 1902. *JoHN C. Smock, Ph. D., Trenton, N. J.; State Geologist. CHARLES H. SmyTuH, Jr., Ph. D., Clinton, N. Y.; Professor of Geology in Ham- ilton College. August, 1892. Henry L. Smytu, A. B., Cambridge, Mass.; Professor of Mining and Metal- lurgy in Harvard University. August, 1894. ARTHUR CoE Spencer, B. S., Ph. D., Washington, D. C.; Assistant Geologist, U. S. Geological Survey. December, 1896. *J. W. SPENCER, Ph. D., 2019 Hillyer Place, Washington, D. C. 614 PROCEEDINGS OF THE PHILADELPHIA MEETING JosAH E. Spurr, A. B., A. M., U. S. Geological Survey, Washington, D. C. De- cember, 1894. JosEPH STANLEY-Brown, Cold Spring Harbor, Long Island, N. Y. August, 1892. TIMOTHY WILLIAM STANTON, B. S., U. S. National Museum, Washington, D. C.; Assistant Paleontologist, U. S. Geological Survey. August, 1891. *JOHN J. STEVENSON, Ph. D., LL. D., New York University; Professor of Geology in the New York University. WILLIAM J. Sutton, B. S., E. M., Victoria, B. C.; Geologist to E. and N. Rail- way Co. August, 1901. JosrpH A. Tarr, B. S., Washington, D. C.; Assistant Geologist, U. S. Geolog- ical Survey. August, 1895. JAMES E. TautmMace, Ph. D., Salt Lake City, Utah; Professor of Geology in University of Utah. December, 1897. RatepH S. Tarr, Cornell University, Ithaca, N. Y.; Professor of Dynamic Geology and Physical Geography. August, 1890. FRANK B. Taytor, Fort Wayne, Ind. December, 1895. WILLIAM G. TieguTt, M. S., Albuquerque, N. Mex.; President and Professor of Geology, University of New Mexico. August, 1897. *JAMES HE. Topp, A. M., Vermilion, S. Dak.; Assistant Geologist, U. S. Geolog- ical Survey *HENRY W. TuRNER, B. S., 508 California St., San Francisco, Cal. . JOSEPH B. TYRRELL, M. A., B. Se., Dawson, Y. T., Canada. May, 1889. JoHAN A. Uppen, A. M., Rock Island, Ill.; Professor of Geology and Natural History in Augustana College. August, 1897. EDWARD O. ULRicH, D. Se., Washington, D. C.; Assistant Geologist, U. 8. Geo- : logical Survey. December, 1903. *WARREN UPHAM, A. M., Librarian Minnesota Historical Society, Saint Paul, * Minn. *CHARLES R. VAN HIsE, M. S., Ph. D., Madison, Wis.; President University of _ Wisconsin; Geologist, U. S. Geological Survey. FRANK ROBERTSON VAN Florn, Ph. D., Cleveland, Ohio; Professor of Geology and Mineralogy, Case School of Applied Science. December, 1898. GILBERT VANINGEN, Princeton, N. J.; Curator of Invertebrate Paleontology and Assistant in Geology, Princeton University. December, 1904. THOMAS WAYLAND VAUGHN, B. S., A. M., Washington, D. C.; Assistant Geol- ogist, U. S. Geological Survey. August, 1896. *ANTHONY W. VopcEs, San Diego, Cal.; Captain Fifth Artillery, U. S. Army. *MARSHMAN EH. WADSWORTH, Ph. D., State College, Pa.; Professor of Mining and Geology, Pennsylvania State College. *CHARLES D. Watcott, LL. D., Washington, D. C.; Director U. S. Geological Survey. THomaAsS L. WALKER, Ph. D., Toronto, Canada; Professor of Mineralogy and Petrograhpy, University of Toronto. December, 1903. CHARLES H. WARREN, Ph. D., Boston, Mass.; Instructor in Geology, Massachu- setts Institute of Technology. December, 1901. HENRY STEPHENS WASHINGTON, Ph. D., Locust, Monmouth Co., N. J.; August, 1896. THOMAS L. Watson, Ph. D., Blacksburg, Va.; Professor of Geology in Virginia Polytechnic Institute. June, 1900, FELLOWS OF THE SOCIETY 615 WaLter H. WEED, M. E., U. S. Geological Survey, Washington, D. C. May, 1889. Frep. BoucHton WEEKS, Washington, D. C.; Assistant Geologist, U. S. Geolog- ical Survey. December, 1903. SAMUEL WEIDMAN. Ph. D., Madison, Wis.; Geologist, Wisconsin Geological and Natural History Survey. December, 1903. Sruart WELLER, B. S., Chicago, Ill.; Instructor in University of Chicago. June, 1900. Lewis G. WESTGATE, Ph. D., Delaware, Ohio; Professor of Geology, Ohio Wesleyan University. Tuomas C. WESTON, 591 Saint John St., Quebec, Canada. August, 1893. Davin Wuirte, B. S., U. S. National Museum, Washington, D. C.; Assistant Paleontologist, U. S. Geological Survey, Washington, D. C. May, 1889. *ISRAEL C. WHITE, Ph. B.. Morgantown, W. Va. *Rospert P. WHITFIELD, Ph. D., American Museum of Natural History, 78th St. and Highth Ave., New York city; Curator of Geology and Paleontology. *EDWARD H. WILLIAMS, JR., A. C., E. M., Andover, Mass. *Henry S. WILLIAMS, Ph. D., Ithaca, N. Y.; Professor of Geology and Head of Geological Department, Cornell University. BAILEY WIL LIs, U. S. Geological Survey, Washington, D. C. December, 1889. SAMUEL W. WILLISTON, Ph. D., M. D., Chicago, Ill.; Professor of Paleontology, University of Chicago. December, 1889. ARTHUR B. WiLuMorTT, M. A., Sault Ste. Marie, Ontario, Canada. December, 1899. ALFRED W. G. WILSON, Ph. D., Montreal, Ont., Canada; Demonstrator in Geol- - ogy, McGill University. June, 1902. ALEXANDER N. WINCHELL, Doct. U. Paris, Butte, Mont.; Professor of Geology and Mineralogy, Montana State School of Mines. August, 1901. *HORACE VAUGHN WINCHELL, Butte, Montana; Geologist of the Anaconda Copper Mining Company. *NEWTON H. WINCHELL, A. M., Minneapolis, Minn.; editor American Geologist. *ARTHUR WINSLOW, B. S., 84 State St., Boston, Mass. JOHN E. Wo.rr, Ph. D., Harvard University, Cambridge, Mass.; Professor of Petrography and Mineralogy in Harvard University and Curator of the Mineralogical Museum. December, 1889. RoBerT S. Woopwarp, C. E., Columbia University, New York city; Professor of Mechanics and Mathematical Physics, Columbia University. May, 1889 JAY B. WoopwortH, B. §8., 24 Langdon St., Cambridge, Mass.; Assistant Pro- fessor of Geology, Harvard University. December, 1895. FREDERIC E. WricHT, Ph. D., U. S. Geological Survey, Washington, D. C. De- cember, 1903. *G. FREDERICK WRIGHT, D. D., Oberlin, Ohio; Professor in Oberlin Theological Seminary WILLIAM S. YeEaTEs, A. B., A. M., Atlanta, Ga.; State Geologist of Georgia. August, 1894 FELLOWS DECEASED. *Indicates Original Fellow (see article III of Constitution) *CHARLES A. ASHBURNER, M. S., C. E. Died December 24, 1889. CHARLES E. BEECHER, Ph. D. Died February 14, 1904. 616 PROCEEDINGS OF THE PHILADELPHIA MEETING Amos Bowman. Died June 18, 1894. *J. H. CHapin, Ph. D. Died March 14, 1892. *EXDWARD W. CLAYPOLE, D. Sc. Died August 17, 1901. GrEoRGE H. Coox, Ph. D., LL. D. Died September 22, 1889. *EDWARD D. Corr, Ph. D. Died April 12, 1897. a ANTONIO DEL CASTILLO. Died October 28, 1895. *JAMES D. Dana, LL. D. Died April 14, 1895. GEORGE M. Dawson, D. Sc. Died March 2, 1901. Sir J. WILLIAM Dawson, LL. D. Died November 19, 1899. *GEORGE H. Evpripcr, A. B. Died June 29, 1905. * ALBERT EX. Foote. Died October 10, 1895. N. J. Giroux, C. E. Died November 30, 1890. *JAMES Haut, LL. D. Died August 7, 1898. JoHN B. HatcHer, Ph. B. Died July 3, 1904. *ROBERT Hay. Died December 14, 1895. Davip HonNEYMAN, D. C. L. Died October 17, 1889. THOMAS StTERRY Hunt, D. Se, LL. D. Died February 12, 1892. *ALPHEUS Hyatt, B. S. Died January 15, 1902. *JOSEPH F. JAMEs, M. S. Died March 29, 1897. WILBUR C. KnicHT, B. S., A. M. Died July 28, 1903. RaLpH D. Lacozr. Died February 5, 1901. * JOSEPH LE ContTE, M. D., LL. D. Died July 6, 1901. *J. PETER LESLEY, LL. D. Died June 2, 1903. Henry McCarey, A. M., C. E. Died November 20, 1904. OLIVER Marcy, LL. D. Died March 19, 1899. OTHNIEL C. MarsH, Ph. D., LL. D. Died March 18, 1899. JAMES BW. MILus, B. 8S. Died July 25, 1901. *Hrenry B. Nason, M. D., Ph. D., LL. D. Died January 17, 1895. *PETER NEFF, M. A. Died May 11, 1903. *JOHN S. NEWBERRY, M. D., LL.D. Died December 7, 1892. *EDWARD ORTON, Ph. D., LL. D. Died October 16, 1899. *RICHARD OWEN, LL. D. Died March 24, 1890. *HRANKLIN PLATT. Died July 24, 1900. *WILLIAM H. PreTrer, A. M. Died May 26, 1904. * JOHN WESLEY POWELL, LL. D. Died September 23, 1902. *CHARLES SCHAEFFER, M. D. Died November 23, 1903. CHARLES WACHSMUTH. Died February 7, 1896. THEODORE G. WHITE, Ph. D. Died July 7, 1901. *GEORGE H. WILLIAMS, Ph. D. Died July 12, 1894. *J. FRANCIS WILLIAMS, Ph. D. Died November 9, 1891. *ALEXANDER WINCHELL, LL. D. Died February 19, 1891. Ei ALBERT A. WRIGHT, Ph. D. Died April 2, 1905. Summary Ovizinal Fellows’ 22 see! oa ek oe eee 65 Milected. MeO wis: sowie nus clots ete eee oe orton en ae ae 206 Menrbersiip coc 3.2 oc otineeee tic ndng evaneie eereie ee 271 Deceased Fellows .......... Say ehgtand 5 Sie aha teette 4+ INDEX TO VOLUME 16 ws Page ~AARE valley, Cliff in, View of......... oe ——, Glacial erosion in.............. 32 ABIQUIU, New Mexico, Triassic plant Remains found NEA... occ dale 494 AccESsIons to Library from July, 1904, to July, 1905; H. P. C ing, Librarian 595-604 ADAMS, F D., Record of remarks by... 574 — and EB. G. Coker; Experimental in- vestigation of the compressibility and plastic deformation of certain rocks [abstract] ADIRONDACK region, Glacial erosion in. 50-51 AGASSIZ, ALEXANDER, cited on West In- i By Rete fais at Sh fe lois hin Silaa.o3 21a, wholes 259 —, Map of Windward archipelago com- BUMP ATMERUEIO IT Se etc te al siaions) abs. wie S- &, 8 243 - AIMS and JaAcors, Sections of rocks at New York city prepared by, Refer- MMM RRIAD Sete clo Oktay la, «atic speteta ea tiiec eek 167 ALAGOAS, Brazil, Coastal lakes of, Chart LDS LE a ae ee ee 5 ALASKA, Glacial erosion in.......... 41-47 — peninsula and Cook inlet, Cretaceous rocks and fossils on.......... 407-409 — — — — — , General geology of.. 392-393 — — -— , Jurassic rocks and fos-. DLS EIST Phe ae a ee ae eee cl 396-407 — -— — — — SOE OE ee ete Sapa Sven OO — —— — — — , Mesozoic section on. 391-410 — — — — — , Triassic rocks and fos- oe Oe doce so duste’as 393-396 Peer IYSIS Of .. .. 2.2 ss oe ee ce 114 ALBONICO, , Asbestos cloth and pa- per manufactured by......... 430-431 ALDINI, CHEVALIER, Asbestos clothing BEMICUIECE “DY ~.. . cs civ oes e's cc 430 ALLEN, F W., Aid rendered by........ 162 ALLEN, W. L., Aid rendered by........ 91 ALPINE lakes, Implications of, as to SS hs Ba Se oe ae 23-24 Aups, Glaciers of, Erosion by....... 31-32 AMI, H. M., elected Councillor........ 540 —, Record of remarks by...... 562, 580, 582 AMPHIBOLITE, Analyses of............ 444 —, Micro-structure of ......... 436-439 ——, Mineralogy of .. 2.6 c cence cae 436-439 NNN OE i cla dos boas Cb ee ae 445 —, Photomicrographs of ......... 436, 438 —, Serpentine, and associated asbestos deposits of Belvidere mountain, Vermont; V. F. Marsters..... 419-446 AMPHIBOLILES of Belvidere mountain, Vermont, Distribution and charac- ORME ee cia ithoce aps ot AT ees 426-427 AMPHITHEATER (The), Uncompahgre valley, Colorado, Sedimentary beds BR eis OBI M19 oieiel aw, Wl we OY a 454-457 —— -—— —--—, Views of........ 456, 458, 460 See names of sub- stances analyzed. ANDERSON, F. M., Title of paper by.... 592 ANDERSON, TEMPEST, and FLert, John S., cited on eruption of Mont 5 MN On VRS Oa ey See 249-250 —-—-—, Study of West Indian volea- ROCESS Coat strc.) cis. coh Weeakcrak ct. 244 ANDESITE, Composition of............ 250 ANEGADA channel, West Indies, Charac- AG eg Bsc ah de via wick 0.0 275 ANORTHOSITE, Compressibility and plas- tic deformation of, Experiments ORY Salen etal Seen O\ae wo HOD are le, oie bs 564-565 564-565 } Page ANTIGORITH, Occurrence Of........<.+. 428 Apeibobsis gaudinii, Description of.... 508 MP VILG: (OL. a, oincese so Sistelcvaile Doone ae elate 515 APLITtTE from San Gabriel mountains, California, Character Of... «s< sab 1 ARGON, Presence of, in volcanic MVPD cs sv Ssak ois ovelinca asus eealeaua ets 252, 253 Aristolochites elegans, Description of... 508 eee NPE 8 OL. cies: s se oo eyes vid eis eee onare 515 — majus, Description of.............. 508 eet TEC OL fine ese oo op'svloi a! cin fonaits fo MacoteMe 515 — sulcatus, Description of............ 508 eh ee TOU TOL COE ca, ss ge W9)0l 8i6 Uo fur w) Wve thie adage 515 ARIZONA, Northeastern, Geologic feat- EPO OE: sacri 5 Bele sper ebb) oso Seeee ohonlece -482 ARKANSAS sandstones, Distribution of.. 491 ARKONA beaches, Character and occur- TETIGOVOL Wah chesd os bole tes salty oak aah eee 588 ——, Relation of lake Whittlesey EOE adelante deere ove teautns way Aca atts 587-589 ARNOLD.) ONGOS), AiG DY xis ave. ss. wletele a 188 ARNOLD, RALPH, elected Fellow........ 540 —and A. M. Srrone; Some crystalline rocks of the San Gabriel mountains, Califormiate. csc stated teas to vine te woes 183-204 ——, Title. of paper by.............. 51D ARRHENIUS, S. A., cited on gaseous cen- KET (Of AEMGs CAPD ace atevcts iucrete tel « 282-283 ASBESTIFORM serpentine, Origin of veins in, Paper on, by G. P- Merrill.. 131-136 ASBESTOS, Cross-fiber form of.......433-435 — — — — — , Figures showing ....... 435 —, Deposits of, in Vermont....... 419-446 Se , Map showing area of... 421 —, Early knowledge of............... 430 = PRAM OSY Op ao) biel ote tao leserxtohene. opauere is 433-435 es OUP GLI) ope clave salah olshe isn afele, orm 435-436 —, Photomicrographs of ............. 440 ==, PL -GDEL TORDL: OL < simi a bidie pinnae 433-435 —_— — — — , Figures showing ....... 434 ey VISES OL sherk oretecal lesa’ avate! sven eieve tenets 430-431 —, Veins of, in block of massive ser- pentine, Plate showing........... 131 AUBREY group, New Mexico, Age of.... 479 AUBURN, New York, Onondaga limestone exposed at, View showing........ 64 AUDITING committee, Appointment of.. 540 a RE DOLE ME Neo nlipic ic Sieleia s\o /s0 as levele 565 Bapito formation, Occurrence of...... 491 BAKER creek, New York, Lowered divide i HeHO Oba nie oe oie onl «ctl s wile a piss seve, vate) «Bintan oes 29 —-, Views showing contortion of.294, 295 (617) 618 Page BARBADOS, Geology and geologic his- LORY OE Sc cca oe cae eek eee eee 260, 269 ——.. Marine erosion: ON. ciai..ceiiese = eto ee BARTON, G. H., cited on glacial erosion. 34 BARUS, CARL, Experiments in high press- ure temperatures made by........ 281 Bascom, F., cited on crystalline rocks of Cecil county, Maryland... 302, 312 ——— granite - gneiss of Piedmont Pennisylvaniasre ses 2s ciee a toec ne 309 — — — Wissahickon mica-gneiss....... 303 —;Piedmont’ district of Pennsyl- vania 28 —, Record of remarks by......5...... 572 ==, Tithe Of paper Dya5 ks acme & cree e eee 52, BASINS (Rock), Taplications of, as to LCESCLOSION, OA cota tesa cid hese te aie Sie eae 24 BATTERY (The), New York city, Reef in Hast river off, Rock composing. 176 BAVIERO, Marquis DE, Asbestos cloth and paper manufactured by... 480-431 BEAR bay, Cook inlet, Alaska, Triassic limestone at, View showing A once at BEARING of some new paleontologic facts on nomenclature and classi- fication of sedimentary formations; HS Ss Witlanige. Unio a 137- 150 BECKER, G. F., quoted on glacial erosion in the High SIETEAGS Se aac 3 BEECHER, C. E., List of papers by.. 546- 548 —, Memoir of, by Charles Schuchert . 541-548 BELL, ROBERT, Title.of paper Py: =. o- 576 BELMONT, AUGUST, Aid rendered by. 169 BELVIDERE mountain, Vermont, Altitude CE ee Rie ace. Te cca BOM oe) cee 422 —-——-—, Amphibolites of......... 426-427 — — — — — (HOLY SeS#Ofien. ae ae ea 4: ——-+, Amphibolite, serpentine, and 3 asbestos depostts70f. 5005 6. <: 419-446 —-——, Chrysotile from, Analyses of.. 443 —— =~, Geological map of. ...¢.. 2.2. 421 a ROCK LYIDES) abe apie wie = nn 424-428 —-_ | SCMISUSHOL, (ACE sOhs.2 6) oes 425-426 , Character of ....... 424-426 —_—— —, serpentine from, Analyses of.. 444 — — — — — , Character Ole ear i 427-428 — — — — — ” Microstructure of .. 440-441 —— —, Structure OP Ste amie terete 428-430 —-— —,, Topography of region near. 422-424 a EAVEESE ila DO. see te cete ee 423 —— View: from. lh ee ee ee he 426 = i VIOW Ol ee Sin he es Ae ee 426 =, VIEWS “Neat 5. wikis. ee tee 433 BENNETT, B. J., Acknowledgments ane 131 nie ta ‘CHARLES ee Acknowledgments — pula veraphy of the Uinta moun- PAREN GE 5 a Ret co ae cre eee te eee 517-530 <=, Birtles of papews Wygrec colic was we tones 582 BIBBINS, ARTHOR, Title of paver wy: 589 BIBLIOGRAPHY : C. E. Beecher. 546-548 ¢ hohner. Wathen. i. we eee Oe 553-555 —: Henry *MieCalley he ne ie ee sds 557-558 s William’ Henry Pettee,.-2. 22%. ..2 560 Bicarpellites knowltoni, Description of. 510 IMU e NOL oie ke Bone hae ee 515 — rugosus, Deseription- Of - soc soe. & es 510 == = HHI SURO OL sass eso coe ee eee ay Ls Big SANTA ANITA canyon, San Gabriel mountains, California, biotite- oe DLOCLTOM sl tare ie Sears nee 189-191 — biotite - granite - PTIESS) LLOMM Gere. cede ec ane. alee 02 — , Diabase por- PHY Py Ero aes ec che See noe 200 — , Hornblende-dio- Fite sSneiss* frome 2 <4) ees eee 201 —_ , Quartz - horn - blende porphyrite from........... 199 BIOTITE-GRANITE from San _ Gabriel mountains, Character of...... 189-191 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Page BIOTITE-GRANITE-GNEISS, from San Ga- briel mountains, California, Char- acter Of) 06 is... see 202 siere lake, Canada, Asbestos deposits BE ans bbe os alehe ane tecel See Ronen een BLACKWELL’S Island bridge, New York city, Rocks beneath. s225- 7.55 168-169 , Section across Hast river at. ...0b cs See Oe 168 —-—tunnel, New York city, Rocks dis- closed by J. oS. 7 eee 66-167 showing 66.050. 45 ee oe eee BLAKE, B. B., Acknowledgments to.... 419 —, Discovery of asbestos in Vermont |) Geren Se en BLAKE, W. P., cited on geology of San Bernardino mountains 2. eee 187 BLoopy canyon, California, Hanging val- leys along ....... 20 ae,5 80 eee 88-90 — BOESE, , cited on volcanoes mr fault lines, lack of relation of. 287 BOILING lake of Dominica [abstract] ; a B.-'Q. (Hovey...) se eee ee 570-571 + = =- =, View of «8. eee eee 570 BouueEr, A. Pe Aid rendered by.162, 163, 165 —, quoted on rock beneath Harlem PIVED cei sb es oe eee ee eee 163 Bonps of Tioga township, Neosho . county, Kansas, Redemption of... 535 BONNEY, T. G., cited on ice erosion.... 17 BoOTRYOIDAL glass, Occurrence of, in Holyoke trap sheet.) 4420-2 104-105 BOUTWELL, J. M., Carboniferous fossils found in Uinta mountains by..... 528 —,cited on geology of Uinta moun- . tains —, quoted on geology of the Uinta mountains BOWMAN, ISAIAH, cited on hanging stream-cut valley in Michigan..... BOWNOCKER, JOHN ADAMS, elected Fel- VOW | fete eve eave ci cs 00 Oe 540 BRANDON, — — — — — » APO OF oicn eS bs eter ee ere — — — — — , Beds associated with. 501-502 — — — — — , Bibliography OL aries 514 — — — — — , KOSSIIS Of eee "504-514 Brandonia globulus, Description of.... 513 , Figure of BRANNER, J. C., quoted on ice erosion in the Yosemite valley........... —, Record of remarks by.(3. 3.22 oon 563 —; Stone reefs on the northeast coast of Brazil; annual address by the Pres- ident =, Title of: paper, Dy’.< #0. csveeneeeee . 566 Brazvin, Climate of part Of... 5-6, 12 —icoast of, History Of... -. eee 4-5 —, Stone reefs on northeast coast of, Address of President on.......... 1-12 ee Commennn — — — , Features of.. — — — — — — , Location Of ~oc. ees — — — » Origin (of Seine 2 — — — — — — — , Structure of ..... 2-4 — — — — — — — . Views ‘OL. ts salen eee —, streams of, Periodicity of......... 6 BricHaM, A. P., cited on glacial erosion. 32 Brock, REGINALD WALTER, elected Fel- VOW osc ec ead bois ke ele eke eee 540 BRONCO, , cited on voleanoes and faultelines <2. fcc sk ota eee 287 BROOKLYN bridge over Hast river, New York, Rocks disclosed by work om os iae Rar ear sce - 528 . | . INDEX TO Page oo A. H., Alaskan fossils collected y Brown, BARNUM, Fossils collected in PENSE S ES ING ce aon cirov's vc mls joe elaleucsw he's 80 Brown, N. H., Fossiliferous Triassic beds in Wyoming discovered by... 489 i alamet WILLIAM, Chemical analysis MUR fics cys) ae. Stel v Subse, «aris oomsn’e BRUCITH as a rock constituent, Deter- BRM PMPRER ESTER OT hate ie ao 'evece is ec areca esis e 586 BRYN MAwRr gravel, Age of........... 326 Buck, R S., Aid rendered by...... 168, 171 BucKHORN canyon, San Gabriel moun- tains, California, Biotite-granite of. 190 BuFFALO, New York, Corniferous lime- MiP ER DONCU Sb. \c iu cise ace cs cw © 52 Boermcrny, Distribution of............ 533 see TDL ES ee eee 534 akan R. W., Analysis of palagonite ae eee Ae BI eR Eee vs —, cited on palagonite of Seljadalr, Ice- PMG E ete ore a cei gvave Cacecs 23, 124 Burter, Montana, Shifting of continental BUIBIESE Aes raver nt cre coe s Sieyotio aie o°é 3/0 a 587 BYRND, EH. A., Aid rendered by........ 163 CagoT, E. C., and Desor, E., cited on geology of Nantucket............ 387 CALCITE, Transparent crystal of, Figure showing “Cc eae Etc 2 CALCITE spherulite, Figure showing. 128 CALVIN, SAMUBL, elected First Vice- NIM Tonto hgh ai ale oie » 540 CAMBRIAN rocks of Piedmont Pennsyl- vania, Features of... 296-298, 306-308 and pre-Cambrian formations of Piedmont Pennsylvania, and Mary- land, Character, distribution, rela- tions, and thickness of....... 294-298, 330-334, 348-349, 352-354 — » Correlation Oe ee ee 40 CAMBRO-ORDOVICIAN formations of Pied- mont Pennsylvania and Maryland, Character, distribution, relations, RUEMCICHICSNU OL! b)< vs.c as arecea de 298-301, 334-335, 357-360 , Correlation of.. 340 , Features of 298-301 306-308 CAMPBELL, M. R., cited on erosional his- tory of central New York......... —, Reference to work of............ CANADA, Asbestos deposits in, Geology of 432 — — industry Me eee Pata ewe vas rao as 431-432 ——ieiaeiat CrOSION IN... 20.666. esse 3 49 CANANDAIGUA valley, Cross-section pro- OSS Coe RR ree eee ear es 61 CANYON City, Colorado, Triassic beds PRTC ee Valais) arc ere. sists ie ancl « 493-494 CANYON creek, Colorado, Sedimentary beds along 2M i RAS Be a ener 454 ANCHOR ss oh). 3.6 va acaje s'sce.8 we 456 CARBON gases, Presence of, in volcanic TOADS se coils ey ee oo ae 252 CARDIFF quartzite, Age and _ strati- PTPPHIC. DIAC OL. 5.626. ties a 330, 340 tS PISEFIDUTION OF oso ess cb eset e 338 CARIBBEAN islands, Volcanic eruptions in, Students and studies of.. 244-245 CARNEY, FRANK, cited on altered course of Fall creek, WORE che. icine 6 200 _-_— Tioughnioga MARES eM ctet a) avails) 6. x's 232 Carpolithes brandoniana var. elongata, TEs eg heu GV Cr) Big ee eee eee gi lal ————, Wigure of..............-. 516 — — — obtusa, Deserintion of... .°..,%:. 511 ————, Figure of................ 516 — elongatus, Description of.......... 511 LXXV—BuLL. Geox. Soc. Am., Vou. 16, 1904 VOLUME 16 619 Page Carpolithes elongatus, Figure of....... 516 — fissilis, Description Tap al ia Ss. Oks aa ne NI OUITS HON: soe susse ates cuateaels aiats ae 516 —= obtusus, Description. Oke cutacscte oes 5 ee cin 4) 0 a 0) SOROS Rare eceey CNR MC I Oe 516 CASCADE creek, Colorado, Sedimentary beds’ alone... sie cee eee 455-457 ———, View Near.......cscesceees 460 CASCADE mountains, Ice erosion in. 39-41 CATAWISSA section, Pennsylvania, Fos- SUISHTOE so a5: coc eke oo sheeaaet atereiye waoue 140-141 Caves of island of Put-in-bay, lake Brie; “Origin cOtss sc crate aera 563 CAYUGA lake, Map of shores of, at Union Springs, Ne@w MOrK «aia sew pie 62 EER valley, Cross-section profiles of 61 —w—, Discordance of, with valley of Seneda. Takers cue cee ee eee eae 241 —w—, Failure of ice erosion in, Plate SHOW LE iro cq ete aeekel aetian sr ereieneey one & 64 sa, General FEATURES OF... tse ocsreta 216 ——. ” Glacial ice motion in, Direction of 216-217 ——, Hanging valleys in........ 230-233 —w—, Ice-dammed lakes in.......... 221 , ice occupation of, History of. 217-218 , Map showing portion of........ 231 —-—, moraine loops in, Terminal. 224 , Moraines in lateral valleys of 220-223 —w—, moraines of, Frayed........ 219-220 — — — — , General features Of... < s.5.. 218 ————, Lateral ............. 218-219 — — — — Marginal and outflow chan- nels associated with. >... ........ 228 ————,, Origin of............ 227-228 ——, Morainic FPanetinte ds: 0 eee eis 224 —-—, Nunatak moraines in....... 224-225 ——, Outwash gravels in............ 228 ——, Vertical relation of, to Ontario lake basin, Diagram showing...... — and Seneca lake valleys, Moraines of, Paper on, by R WATTERS versie 6 215-228 CayuTA creek, Hanging valleys along.. 233 — —, Moraine-formed divide at the head a —-—, Old divides cut by..... 234-235, 238 CELESTITE-BEARING rocks, ANG” GIStEIDMTMOM Of 2). cic ss, coe octane 5 CENTRAL bridge, New York city, Sections at CEARLENGAE, report of the, Citation OU cance ache Stare Oey ahaa fe de dedee (dato tage CHAMA river, New Mexico, Fossil plants TL EDN sortie, caer motets etter ct or ataret evens ta, wits etal ae CHAMBERLIN, T. C., cited on glacial ero- SUT ia ore ick: cae ee ors arn tench s Larecetene —-—- moraines of western New D405 3 ally Boies ohne Ae hin PEO ee ree 215 — and Salisbury, R. D., cited on glacial CPORIOMG ayes adie aie fe ane aise eee aha oe sl ehtun « Z — — — —_ — hanging valleys ...... 75-76 —— — “Hanging valley” defined by.. 75-76 CHASE, F. L., Acknowledgments to 157, 161 CHELAN lake, Glacial erosion at...... 39-41 —w—, View of CHEMUNG river, New York, Course of, Outwash gravels determining..... PAY —-—, Diversion of, near PHlmira.. 234, 238 a , Map showing........ 234 —-—, Lowered divide on............. 234 ——— — —, Causes of........ 238, 239 CHEMICAL analyses. See names of sub- stances analyzed. CHESTER VALLEY limestone, Age and COLTEISTIOU Oty casnsscren mieter sooears 300-301 — ——, Character of .... 2. .s2es sere 299 ——w—, Contact of, with ‘Triassic shales, View showing. eee ine ibe eanue ED PISELIDWOLOM: OL. es «salen 298-299 ———, Stratigraphic relations of 299-300 620 Page eee VALLHY limestone, Structure i] ee eee Pas ree. ie icge eae src aA Sian = SONNE CKNESS 20 fiancis tocaneucueicrsteun nite te 300 —— —, Views of exposures of.... 299-300 etre quartzite, Age and correlation es OF EE Seema coh ee ee —w—, Character of.............. 296-297 ——, Chemical analyses of.......... 297 ——-=— Distribution of, in Piedmont Penn sSylWaniar pian vsentece ake epee amen tes —w—, Stratigraphic relations of...... 298 =, SUPUCHIBG Ob ea ico, acs leyanele a Shatrate 307 ee ENT CKNESS) Ole.) sas, cusnshenc le tyee ds acute 298 CHILLAO canyon, San Gabriel moun- tains, California, biotite-granite of 190 CHINITNA bay, Alaska, Enochkin forma- TION ats = SECTION (OL. sas eleusiereress 400-401 ———, Section at ............. 403-405 CHRYSOTILE, Analyses of....... is natcans 443 —, Occurrence of, in Canada.......... 431 —, Block of serpentine containing veins 1 eaten Mad eM a en An See itty ci 131-136 —_— — — — — — ~ Plate Showine. . ac. slok CHUGWATER formation, Occurrence is oe CHurcH, B. S., cited on character of. rock beneath Harlem river........ 159 CIRCEL, FRITZ, cited on origin of asbes- LOS Peake eee tote nal nue a ee eee Baereeenaors 134 CIRQUES, Implications of, as to ice ero- STONE Hoh eae toc ape Cinnamomum lignitum, Description of. 514 ees NL UITO ON. oo auc) eve wae seue fests euorerede me ons 516 CLARK, W. B., Record of remarks by... .562 CLARKE, J. M., Record of remarks by. 582 CLASSIFICATION of the Upper Creta- ceous formations of New Jersey [abstract]: Stuart Weller........ 579 CLAYPOLE, E. W., cited on Sierra Madre TANTILEL, SOb ec eee eee ee ee le eens CLIFF erosion, Effect of, on form of con- tact surfaces, Paper on, by N. M. THEnNNMeMANies. ee ees ele ee a eis 205-214 CLIFFWOOD clays, Fauna of, Abstract of DAPPER OM hes see cae ee eaten otek 580 COCKEYSVILLE marble, Age and strati- graphie place of.... 330, 335, 339, 340 —-—, Area of, in Maryland, Faulting MTD ON sca idshee ope cee ey chs mmeton hoys aes 362-366 — — — — — — (ie OLGIME IN. ches ee OO — — — — — — Structure of .... 361-366 — —, Character of...... 334- 335, 349-351 ey OES EPI EION GE Sentence dian. 334, 357-360 —-—; H. B. Mathews and W. J. Mil- 1K SA ONE ee Fatt A at Re rg Ault ene 347-366 COENTIES reef, East river, New York, IROCKAOE Seth ete toe Monae eee 173-174 Coker, FE. G., Frank D. Adams and; Experimental investigation of the compressibility and plastic deforma- tion of certain rocks [abstract] 564-565 CoLpD bay, Cook inlet, Alaska, Naknek formation at, View showing cepa 407 CoLEMAN, A. P., ‘quoted on glacial ero- SLOWS si cate eee be meee Anne De 49 —— RECOM OM LEMATKS DY. s oo cole eee 562, HO, 96D). Dilla: 574 COLERAINE, Canada, Asbestos at...... 431 COLLING Woop, d Phe cited on rock beneath Hast river, New MOE UCILY fires oe eiics COLORADO, Cliff erosion in, Studies of 205-214 -——, Pre-Carboniferous strata in, Thick- TOSS. Ole iccetane kbencuca ner ten cae Ee 52 —, Red beds of, Character, distribution, and correlation OFS so Ga. toro t4 Como beds, Equivalents of .......... 496 COMPILATION book, Use of, in Beglosica Held work oe CMe e eae ae eee 16-417 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Page COMPRESSIBILITY and plastic deforma- tion of rocks, Experimental investi- gation: Of ...... .s.3eeeeee 564-565 CONTINENTAL divide at Butte, Montana, Shifting of .% ..:555 eee 587 Cook inlet, Alaska, Enochkin formation at, Sections of — — — — — — , Views showing.. 395, 399 ———, Map of part of west coast of. 395 ——-—.,, Naknek formation at, Sections — OB) 0 Sa iand Ale cana ci eee 403-406 —-— —, Naknek formation at, Views SHOWING Ws.2 25 sae 395, 399 —-——., Triassic limestone at, View SHowiN?. ss, 2/5 <-: ora begelle/aneteneeeea 394 — wand Alaska peninsula, Cretaceous rocks and fossilS on.........~ 407-499 — —- — — — , General geology of.. 392-393 — — — — — , surassic rocks and fos- SHS Os. in552.5 heen eee 396-407 — — — — — , Map .0f..i0. .412.e i a — — — — — , Mesozoic section on. 391-410 — — — — — , Triassic rocks and fos- SilS (Of cn aoa Ge eee 393-396 Copz, E. D., Triassic vertebrate fossiis found in New Mexico by..... 494-495 CorBETT gulch, Uncompahgre valley, Colorado, Sedimentary strata in.. 459 CORDILLERAN Section, Election of officers i MP Sere cs ea ao ( —-—, Proceedings of............ 592-594 —-—, Register Of. .% 2... 3 m= eee 594 CORNELL quarry, Hulberton, N. Y., View és BE Sie eee e aude o he ste ee CoRNIFEROUS limestone, failure of ice erosion On, ViIeCwS (Ofc. :ei eee 52 CORONA, GIUSEPPE DELLA, Asbestos mill- board manufactured DY. 2.2 scare 430 CORRELATION of Maryland and Pennsyl- vania Piedmont formations; EH. B. Mathews < si u.(e ee eee 829-346 CouNcIL, Report of...) >... eee 532-533 | ——-=— adopted 60.35 cers 565 COUNCILLORS, Election Of \..2- 3-46 eee 540 CREAMY sandstone, Age and geologic equivalents of ..J4..-5 ee 491, 492 CRETACEOUS rocks and fossils, Occur- rence of, in Alaska...... 408, 409-410 CRETACEOUS (Upper) formations of New Jersey, Classification of ......... 579 Cross, J. J. Aid rendered by...... 163 CROMWELLS creek, New York city, New York Central Railroad bridge across, Borings: at. 22 ose eee 161 CROSBY, W. Ox “cited on geology of Mount Desert island | score thot eee 583 ——_-—-— hanging valleys .........2. 48, 76 —. Diastrophic hanging valley described DY es) o's ovevs one eae ote Ue eee Cross, WHITMAN, Record of remarks by 574 —, Section of Cutler formation meas- ured Dy .!.j..s s+ duc ee eee epee ee 464 — and Sees Howe, Title of paper by 589 —-_—-—., Paper by, on Red beds of southwestern Colorado and their correlation 2... eee eee 447-498 — and A. C. Spencer, Work of, on Red beds of southwestern Colorado. 451-453 CRYSTALLINE rocks (Some) of the San Gabriel Mountains, California ; Ralph Arnold and A. M. Strong 183-204 CuLvER, G. E., cited on the literature of ice erosion CUMINGS, EpGAR R.; Development and morphology of Fenestella [ab- Stract] . oo. esse «sees eee 562 CUNHANHU Stone reef, Brazil, View show- 1: ere en CUSHING, H. P.; Accessions to Librar from July, 1904, to July, 1905. 595- 604 oe eee ere e eee eee eee eee 398-401 1 J : : a INDEX TO VOLUME 16 CuSsHING, H. P., elected Librarian..... —, quoted on erosional phenomena of PIM INCIOT eas oe ccs ee oe 42, 43 —-— — glacial erosion in the Adiron- dacks. [a AB A SR eee ee ——= Record of remarks by............ —, Report by, as ibrarian. fv... . CuTLeR creek, Colorado, Sedimentary EE IRF ie 081 oa? oreo? ssh ek & woe’ 1 — formation, Colorado, Age of 466-467, 496 —w—, General features of 461 51 563 ——, Lithologic features of...... 461-463 ae IRININE OF. ese ee we ee 459-460 cee EO CIOM” Of ices et ve eee ees 463-466 —w—, Stratigraphic place of ........ 461 Pe EC MICKTICSS OF 6 oicivie vcs ccvesivese 461 DALE, T. N., cited-on origin of chryso- tile veins in serpentine —, Map of lignite deposit at Brandon, Vermont, made by 500, 501 DALL, W. ay cited on Plame geology. 392 —, Eocene fossils from California iden- e108 «© ea = 8 © 6 "eee eee eee MME PMEIAUS DoSccs soe. 2 otk. ait syeie ie Yeoeie gee «or 188 Dana, J. D., cited on Brandon, Vermont, PPE ett aig chee a stinaie eiesk ec wiels os 502 — —— causes of volcanic action..... 280 —-—w— origin of watercourses sur- rounding Manhattan island....... 153 —-— — volcanic exhalations......... 251 DarTON, N. H., Chugwater formation es- SMS FIM ed Sos G'S mcs! ss ane 0 0 6 — cited on Creamy sandstone......... 492 ——w— equivalents of the Creamy 2 EAL GS SSR i ig ee ree 49 —— — Fountain formation ...... 491-492 ———-— _mumnekanta formation ...... _ 492 —-—— Opeche formation .......... 492 — —— Red bed fossils ............. 493 —-— — Red beds in New Mexico..... 493 — — — Spearfish formation ......... 492 —-—-— Tensleep sandstone ......... 492 ——w-— Wyoming formation ........ 492 —, Correlation of formations in Black ERO PEPION DY oc shes oe es vie aie 491-492 —, Fossils found by, indicating Paleo- zoic age of rocks in Virginia...... 345 —, Record of remarks by......... 562, 579 —, Report of Photograph Committee MEROMNCENDY. (eile otal eissc e oo es a ate fcc 590 Pe, PIE VOL PAVEE WY) fo si0s os icicle ce 0ls «ee 564 Pauni salt lake fabstract].......... 564 DAVIES, J. V., Aid rendered by........ 166 — cited on rocks beneath Hast river, Pees OT Nc hate SL etn eo seers ee Davis, W. M., cited on geology of Mount Desert Ce ROS 5 hie Meee aS eae 58 — —— glacial erosion ............ 32, 33 —-—— hanging valleys ............-. — — — literature of ice erosion....... 17 —elected Second Vice-President ...... 540 — quoted on glacial erosion .......... 33 —, Record of remarks by..... 573, 578, 579 —, Term “through valleys” used by.. 233 DARWIN, CHARLES, cited on stone reef at Pernambuco, Brazile oc sas ee a en IP DESIRADE island, Features of ..... 268-269 Dersor, E., and Caport, E. C., cited on geology of Nantucket ........... 387 DETAIL (A) of the great fault zone of the Sierra Nevada [abstract] ; John UE RCOM Ta east ocd ei a caie. owes oie Woe 593 DETERMINATION of brucite a4 a rock constitutent [abstract]; A. A. Ju- ER 5 Nal ed ees She we ae Dae oa 586 DEVELOPMENT and morphology of Fe- nestella [abstract] ; Edgar R. Cum- SE eee eee ey eck Son's 5 DOS 621 Page DEVILS gate, Pasadena, California, ap- lite dike near, Character of....... 199 DexrprR, Kansas, Nitrogen of gas well at Sb. -8 hoes SRR ERR hi aetna Sead 72 DI4BaASDy ANSlYSES? OL s sce 5 oe T2113 a, Compressibility and plastic deforma- tion of, Experiments on ...... 564-565 —, Dike oh View Gb 5 « 270.5 ba 6 Ue dca Wie se elec ere 244 DoOLoRES formation, Age of ....... 451-452 meet CON AMMACTOT OL o's sic cc + os shure sl winters 473 —— =, DCHNIMON OL oss. 6 8 eee dees 451-452 —-—, Equivalents of ....... 493, 495, 496 —-w—, Fossiliferous horizon of, Geo- Srapiicak “Extent OL... 6. . cise is alate 495 ——, Fossils of ............ 468-469, ane —-—, General features of ............ —-—, Lack of equivalents of, in Front SoM OGM TESTO: tocicie oreo on x ne cers. 4,< 3.5.0 492 ——, Lithologic character of..... 467-468 — —, Members of ........22eee0. 467-468 —-—, Occurrence of 452, 454 — —, Unconformity between Hermosa formation and 453-459 DouoreEs valley, Colorado, Red beds in, ISOECELA TIO OL 6 5 6. 5 00a se) Cee soar 472-475 DoMINICA, Boiling lake of, Paper on, wae Oe ee eR le by E. O. PEO VEE treed oa whens ele 570-571 ao ik ae ae, Ee 570 DoucLAs river, Alaska, Naknek forma- tion at, View SHOWS ci2)k ween 407 DRAINAGE features of central New ey a inl FS ete he ppd Br is bee teary cele 229-242 Drirt, Glacial, Evidence of, as to gla- pub Gromenie'. 6.0 34 eae 22-23 DRUMLIN areas ve northern Michigan [abstract]; I. C. Russell. . DTT-578 DRUMLINS in the Grand Traverse re- gion of Michigan [abstract] ; Frank Leverett —of New York, Abstract of paper on.. 576 Drupa rhabdosperma, Description of. y RABULE OL mayo veined se nee © ee i 622 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Page DUCHESNE river, Geologic studies at headwaters of —-—, Geology of region at headwaters of 520-530 —-—, Map of region at headwaters of.. 522 DULANY valley, Maryland, Cockeysville oe @ @ = oe ns ise 8 © 8, © ©lalsis 6 2 e816 MAEDICM TN a. Anant ate cretion 358-359 Dutton, C. E., cited on geology of SouthernyWita: (Wises ciclo «ee rie 483 ——— geology of Uinta mountains.. 519 — quoted on geology of Zuni plateau, IN@ WwW Memi@ 0) is ira chs ous cue sce an ovens 477-479 pide ai cut, New York city, ae BEE eiesciidhe wlameveasceue eines ieee ie aavenewenaas Aj EAGLE river, Colorado, Red beds of.... 490 HAKnM, A. S2-Titlevof paper Dy, « -)-le ieee FAUNA, Geologic use of terms. oe 145-147 Stuart’ Weller, .. 0.3.2.0. aoe 580 FECTEAU, , Discovery of asbestos in Canada * DY «os ocd crete ee FELLOWS, Hlection of <3. .cse eee 540-541 =. List Of 93-505) os .sias ae 606-616 FENESTELLA, Development and morphol- OY Of L2c...s seis s sa en 562 FENNEMAN, N. M.; Effect of cliff erosion on form of contact surfaces.. 205-214 -—¢élected Mellow «..../.. «00 sae 540 —, Title of paper by .......2. soe 573 FERGUSON well, Fishers island, New York, Record of ......+... 00 372 FERN camp, San Gabriel mountains, Cal- ifornia, Biotite-granite from. 190 FIELD map in geological work, Methods of employing ......-+ 4. 412-416 — work in geological surveys, Methods of performing .....«-...2 411-418 FINANCIAL Statement ........« < sscleatenenenenee 537 FINGER lakes of New York, cross-section profiles of valleys of ......+++-. 574, 576, 578, 582 : * i + INDEX TO VOLUME 16 Page FINGER lakes of New® York, Heights of, above sealevel and greatest depths ROLE (SHOWIN isis cc wc ss ve ts 72 — — — —- — , Implications of, glacial erosion , Map showing valleys of. 56 - - , Origin of valleys of. 66-73 —-—- region, New York, Ice erosion in ale Byer es a Gy 4s a 6.66) 4 0) 8 0,6 ESET COE oh oc o's: aive ine) aledst'o'e vce) seis ' FISCHER PAUL, cited on Alaskan paleon- UNED? gee AISI enee SIAR oe eee FISHERS Pear New York, Formations on, raphic distribution of. 390 - a Geiser at RPP E res ttctenev aac 367-390 —-—-— —, Location of ............ 368 — .~Map showing ...... 368 —— ——,, Map of .................. 369 —— ——,, Sections on. 377, 378, 379, 380, 382 —-- — —. Topography a ae 369-371 FISSURE lines, Theory of arrangement of volcanoes along, Discussion of. 286-287 FLAMING Gorge group, Age of......... 487 — — —, Equivalents of ............. 496 . ———, Section of ................. 486 FLEIT, JOHN S., and ANDERSON, TEM- PEST, cited on eruption of Mont NESS Sealers aia) ahercints Gide ed ave 249-250 ———, Study of West Indian volca- noes ‘i172 = Stn ere ae ee ae 244, 249 FOREL, F. a cited on ice erosion...... 17 ForTIETH Parallel Survey, Citations from reports of.... 517, 519, 520, 521, Had p25, 026; 521,328 See also King, ‘Clarence. FouNTAIN formation, Age of ..... 491, 492 sae Se TSE 91 —-—, Character and thickness of..... 491 FRAAS, B., Vertebrate fossil from Ari- A MOeCSCTINECE DY so ciscceisis ole esis 480 FRANKLIN, Virginia, folio of Geologic Atlas of the United States, Paleon- 0 eee eee 142-145 ey PrEeRSIFOR, Record of remarks a we eke of Mount Desert island, 1D ei i a ee 583-585 FRONT range of Rocky mountains, wee ET SLATS! ss SU mere cure 490-494 FULLER, Myron L.; Geology of Fishers pena NeW VOLK 720255. wis 3 367-390 —, Name “morainic fan” suggested by. 224 5 -—, Hecord of remarks by ............ 63 —;Title CREEP DEN SD Y. is cbr e bo cit ce seeing ore 589 FURLONG, E. L., Title of paper by...... 592 GaBBro, Compressibility and plastic de- formation of, Experiments on. 564-565 — of Piedmont district of Pennsy ster PBOEMCFON= OF bie piec Wis bce 8 12-316 —_—— Distribution of.. aL 312 —, Photomicrographs ot BR 312, 313, 314 GALLINAS mountains, New Mexico, Tri- assic vertebrates found in.... 494, 495 GANE, H. S., Fossils of Dolores forma- tion Midepvere by. ois es es ee 476 —, Observations by, on formations in San Juan Vp Re eee eee 476- 477 GARDEN of the Gods, ‘Formations in; ...491 GARDINER clay, Fishers island, New York, Age and correlation of.. 388, 389 — — — — —_ — , Composition of. 376-377 ——-— —~——.,, Conditions of deposi- PARTIDO rls. ch eR ae bios, Ue ie hy ac « 375-376 — — — — — , Correlation of 386-387 So Ley eedetapnic §8 86 distribu- pee es Se a a a a 390 _——_ Name of, source of 375 .370, 371, 374, 376 623 q Page GARDINER clay, Fishers island, New York, Occurrence of 377-378 — — — — , Sections showing.. ok, 374, 376, 377 — island, New York, View showing Ja- cob sand and Herod gravel on..... 367 GARNETIFEROUS schist from San Ga- briel mountains, California, Char- SGCLER (OL) ees are eaves eo es we GAS well at Dexter, Kansas, Nitrogen of 572 GAUTIER, , cited on causes of vol- ecanic action GAY, MARTIN, cited on rocks beneath ee ee Os OV Ow 2 s S18 wie) w 6 we Harlem river at McCombs Dam DEIASE\ enaSicl Sta feeaad es ee sresekana eens 161 GEIKIE, ARCHIBALD, cited on exhala- tions of volcanoes and volcanic MARS) sb e-shieke eves ithe a iovereeouchalond ae —-—w—egas contents of clouds above aetive Volcanoes = ass sais aera oan — —— geological contemporanity of formations as indicated by their POSSULS 5 1c. 315. ag) a dee eke ee One rae — -—— independence of volcanoes of PATE LIMOS YS a eats eh pciee Hee pneeeners 287 — —-— influx of water to earth’s in- CEDIORN “wiser ache aie si aias wi eterna 284-285 — — — lack of satisfactory solution of problems of upheaval and _ subsi- GENEEOE TAI es arc sr iatale ret aleve abte es 279 —-— —on theory of Arrhenius con- cerning gaseous center of the earth ..05 oe Pte eee 284 — -— — volcanic exhalations ......... 251 —— — — volcanoes and fault lines, Lack OLA DEIDIIOR OE, ss leave shame Ae reel 287 GENESEE valley, New York, Devonian SECCIONAOF, TOSSING: OF 145-602 er cc klecerc GENTH, F. A., Chemical ee ia bys «:. 295, 297, 299, 304, 321 GEOLOGICAL book-keeping, Paper by J. ERSINGIN [D9 OTM tote sheaevckepat eloterieia ios a 411-418 GEOLOGY of Fishers island, New York; Myron: “min Mellen’s sic 2.2 we 367-390 GEORGETOWN, Colorado, Diastrophic hangines. valley cnear os 6 ec sederee w 18 GILBERT, G. K., cited on age of Foun- (ile torsnation..s 3. >. cde 491 — —— earth-movements in Great PIKES: NEMO tevevene ic chs, uelars arareueee —-—W— glacial erosion along Niagara escarpment, New York.:........ a bic —-— — —— in Alaska ............ 43-46 —_— — — — — on the Medina plain, New WOT Scitere hte eee aera eh akate 54, 5b —-—— glacial origin of Alaskan TROLS Svea ha oes te ERS aay crelate be 43 —-—— hanging valleys ............ 63) — —— lava of Pelé................ 568 — —— ocean-formed hanging valleys. 78 —-—-— origin of drainage features of the “High SICrras eo. «, oecho oe aie “Hanging valley” defined by. : 75 WES Ea on fiords of Puget Sound dis- CHIE fag, Seah catenin aye Tae he RU ee ee 4 —-—w— glacial erosion......... 44, 45, 46 563, 573 GILLESPIE, G. L., cited on rocks beneath Jersey flats, near New York city.. Girry, G. H., Carboniferous fossils from Uinta mountains identified by.... —, cited on age of fossils of Cutler for- mation, COlOradg. 54 naa nas aeeoe 466 — —-— Carboniferous fossils of Colo- rado 452-453 —-— — faunas of Aubrey and Her- mosa formations My ©) i.e)? Behe woe ow Rh) 6 aoe Oe Ree m = ss . ol ; i 4 q 624 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA ‘ 7 Page 7 GLACIAL deposits in Seneca and Cayaee GRIMSLEY, G. P., Chemical analyses by. “10 4 Lake valleys, Study of....... 215-218 GRINDELWALD glaciers, View (Of ioe oe 32 . — drift, Evidence of, as to glacial ero- GUADELOUPE, Geologie features of). 055 267, ; sion ja tee an erate ae 22-23 268, 269 p —erosion, arguments against, Sum- ==, Map Of voce kao wae eee 267 4 VATA OR U0 tals id yet cirepenaiec at at cee heen 25-31. -=, Physiography of °\<' 2 ee alee 267-268 ‘ —-—, arguments for, Summary of... 18-19 —, Views on............... 268, a 270 ; a =, Malldicy (Of LM Ory sOF ait liar tc 13-74 GULLIvER, F. P., Title of paper bys 587 GLACIATION in Seneca and Cayuga Lake GUNNISON group, ASE OL. 5..acte eer 496 Willeys, Study ofesc we were. 215-218 - -=——, WMeatures) of 2). 43205) wee. 469 GLACInRS, Action of, compared to that ——-—,, ‘Subdivisions of)... pee eee 469 OF sPIVERS): Rev cgcicnccanenarte tac ae tahoe ae 21 ——. See also La Plata formation and GLASS, Devitrified, in diabase, Figure McElmo formation. showing Mirae hele de tetas Vaya iene rape cetep ine : —spherulites, Figures showing....... 128 GLASS-lined cavity in diabase, Figure HAMPSHIRE formation of Virginia and SIMO WAIN Seibee care ceo eusnee aioe are eee West Virginia, Paleontology and GLETSCHERMILCH, Implications of, as to geologic correlation of. 140, ia 145 eroding power of glaciers....... 21-22 HANGING valleys, Definition ‘of term. 75-76 GLOSSOCARPELITES, Description of..... 510 --———, Diastrophic, Mxamplesietone or 78 4 — elongatus, Description o Seer cami, sie 511 Bey ey G)acier-formed, Features and ex- = a SUITE OR Ue eeeshoreene. see ert e eee te anne 516 amples’ of. (0.0/3.2 «-. se eee 78-81 — obtusus, Deseriptiom: wii. sso ce nee . BIL, 3 LUC. Russell. .% S22 eee 75-90 — —, Figure (Oy Seen AU MeMUaletacys Seneeh epic Poca 516 —w—, Implications of, as to glacial ero- — parvus, Description of.............% 510 STOW: 2) she 2 eae ee 25, 29-30 a MOUS VOR ah lee eo ecccr te egas) seo steitene fee 515 ——, Occurrence of, in Watkins Glen Goopar, G. L., Acknowledgments to... 506 quadrangle, New York........ 230-233 GoopcHILD, Je yew cited on glacial ero- —-—, Ocean formed, Features and ex- SHON eis SS Gah ee Ie IC cect ee gee age 33 AMIpPles OF ‘ie Vs. swe nae eee 78 GoopRIcH. quarry, Auburn, New York, ——, Origin of ............. 82-83, 87-88 Onondaga limestone exposed at, —-—, Stream formed, Examples of.... T77 View (Showings: Sse oene eee 64 HANSEN, ANDR. M., cited on glacial ero- GOULD, CHARLES N., elected Fellow.... 540 SLOW 6635 355 5's bla eo widvc ae ee ; GOVERNORS island, New York, rocks be- HARLEM river, New York, Origin of 1 0V242 110) ONG Ue RCE aire Gea aU BRE 174 channel of Sic. tee ee eee 179-180, 181 GRABAU, A W., cited on erosional his- — ———,, Rocks beneath ....... 158-164 tory of central New York........ SQ 0 === == Sections across.) ..2aamee LO —, Record of remarks by..... .562, 582 158, 159, 160, 161, 162. 163, 164, 165 ) —: Relative ages of the Oneida and =e eS Wisure showane eee 158 Shawangunk conglomerates [ab- Harris, GILBERT D., enrolled as life Sibel Chl) Ose Woes male ee eee ene ate ale 582 member...) :. 0. ne ee 535 “a = Titleot PAaperibyn esos we oo ek 582 HARTNAGEL, C. A.; Note on the Ontaric GRAND canyon, San Gabriel mountains, or Siluric section of eastern New granodiorite from, Character of... 192 York =jabstract} ose se. Res ss oo 582 GRAND river, Colorado, Red beds on.... 475, Hartt, C. F., Paper by, on stone reefs j 488, 490 of Brazil, Reference to........... : GRAND TERRE island, Geologic history HASLETHAL, Glacial erosion in........ 32 ; OSES Ne ant vas onal g yee ake ne 268, 269 HatTcHER, JOHN B., List of papers ae St MR Olt oc ane, 2 eee ake ore eee 267 DD? sie aulde lel wate 4: cteine lavalane de eee 53-555 > —— —, Physiography of ........ 267-268 — Memon of, by W. B. Scott..5-. 548-555 3 —— === Wiews! OM i... oss Yow «i 269 Hawes, G. W., Analysis of diabase by. 113 F GRAND TRAVERSE region, Michigan Hawxksuaw, J. C., Paper by, on stone DE WHUMINS, Nes. ee lke one eee CTE reefs of Brazil, Reference to...... 12 GRANITE-GNEISS of Piedmont Pennsyl- HAwnN brothers, Red beds of Colorado ; wania, Character’ of) nm 0500. 309-311 observed) DY” )). =. 2c. «meetin 449 : 308-309 — — — — — , exposures of, Views SOWING aioe sh a densears cosa cet 309, 310 GRANILIES, Compressibility and plastic deformation of, Hxperiments on 564-565 GRANODIORITE, Analysis OF sce Low O Dal OG —from San Gabriel mountains, Char- ACCC MOL Si Latin ete ean hee NR oe 191-197 GRATACAP, L. P., cited on bed of Hud- SOM BEVET rs Mutat reenter eee cle ec eee 154 —-—- origin of course of Spuyten Duyvil GREEK is rans teases ushers ie eras GRAY Cliff sandstones of Utah, Distribu- CLOM AOE Poy eee Ai tree rec oneal Rue Mean arene 483 GREEN river, Colorado, Mesozoic forma- TORS COE) Latah ae ees 487-488 GREENLAND, Glacial erosion in....... 34-35 GREENSPRING valley, Maryland, Cockeys- Walle “ma mble: aliments. sce wele 357-358 CePA. Island of, Geologie features ae GREWINGK, C., cited on Alaskan geol- ORY. ieee ss te) alias eae is ee a 391, 398 GRIMSBL pass and Todtensee, View of. De —— lake, Glaciated cliff near, ‘Views of. 32 HAWorTH, HWRASMUS; Nitrogen of gas well at Dexter, Kansas [abstract]. 572 HAYDEN SURVEY, Colorado sedimentary beds mapped DY. | 4:5. 4idew i ee 474 —-—, Explorations of Red beds of southwestern Colorado by... ——, Observations of members of, in eastern Uinta mountains ........ 487 —-—, Stratigraphic breaks in Rocky Mountain province noted by...... 488 —-—, Work of, in Red beds of Front range of Rocky mountains........ 490 HEADWATER erosion, Lowering of divides by, in western New York..... 239-240 Hector FALLS creek, New York, Profile OE ad 2S he a 3 eee, Seu ae 2 HBILPRIN, ANGELO, cited on carbon diox- ide in lava of Mont Pelé......... 252 — — — glacial erosion .............. 34 — -— — Mont Pelé and Martinique.... 259 — — — the tower of Pelé............ 567 —; Memoir of Charles Schaeffer...... 561 —, Record of remarks by............- 571 —, Study of West Indian volcanoes by. 244 HeEIM, A., cited on glacial erosion... 21, 33 = INDEX TO VOLUME 16 Page HELLAND, A., cited on glacial erosion... 21 HELL GATE railroad bridge, New York city, Borings for, Rocks disclosed > EN ie i ee Re eae 164-165 —w— reefs, Rocks at, Character of. 165-166 HENNIGER flats, Sierra Madre range, Quartz-monzonite from .......... 191 Hermosa formation, Age of...... 452, 479 ee NOES OF, oo ase. v.05 © « tie shale ala sis 452 a PPCUEPENCE Of. i.e. sales see 455 — —, Stratigraphic position of ....... 496 ——, Unconformity between Dolores OPT UIOM ANG |. os 0.0 a v-s,0,01¢ «2 453-459 Hprop gravel, Age and correlation Pa Ge sc) cpa eia Wa ware en sas-a wie s 8, 389 PN ORAPACLEE OL . sc:c cen ee se e's 381-382 — —, Conditions of deposition of..... 381 —w—, Correlation of ............. 386-387 ——~, Geographic distribution of...... 390 ——, Name of, origin of............. 381 ——, Sections showing............... 371, 376, 377, 380, 382 eo PMCKNIOSS OF 2. ce ee ais eee eee we 381 HERODOTUS, cited on asbestos.......... 430 HersHey, O. H., cited on geology of San MOLICL MIOUNCAINS 6.6.5 sere & ose 187-188 Hicoria biacuminata, Description of... 512 SSS SSS) ce rr 516 Hicoroipses, Description of .......... 513 — angulata, Description of .......... 513 OM MPRITE OL cle cys. viwn 6 torn a tale ena6 513 — ellipsoidea, Description of ......... 513 Se Sey ES OR ee 513 Hice, R. R., enrolled as life member... 535 Hien bridge, Harlem river, New York, PR RETICAL 65.5 hcic.0, oocha's Sib clever as 159 — ,» Section at, 5 DTTP EMS 5) Sa As a eae es Sen 159 196, 295, 304, 310, 314, 315, 319 HI.u, R. T.. cited on changes of level in oo SET ee ee Dike —— — West Indian archipelago...... 259 —} Pelé and the evolution of the Wind- WareG archipelago .... 2.660 243-288 meee Or Daner Dy... ..6s06c0cs- sc 589 HILLEBRAND, W. F., Chemical analyses y —, cited on fossils of the Dolores for- Peemiom, Colorado: < oc odcis sss 468-469 —-—-— Stratigraphy of Uncompahgre OREM ROUSE UEND: cece me 6 ahd orem e's os 454 —, Fossiliferous Red beds on Grand BEver GUSCEVCd DY. ios .n sce ees eis a os 488 —, Fossils of Red beds of southwestern Colorado found bv 450-451 mveenroem, C,H. Aid by... ...-.<.e- 91 Hitrcucock, C. H., Title of paper by.. 587 Hitcucock, Epwarp, cited on geology of Belvidere mountain, Vermont.. 424 —--—lignite deposit at Brandon, Vermont. ..... 499, 500, 501, 502, 503 —-——Serpentine deposits of Ver- Bieaais ett aes fc ae ahah « 420, 421 mon Alopes, W. H.; Origin of the channels surrounding Manhattan island, New A aes aie i eta 2's vaio aca Sew id' 151-182 2 ei ee eae CPE eae ad 80 HOLMES, W. H., Explorations in Colo- OW ey a sas asic oa do ale as eee A 476 —, Red beds of southwestern Colorado observed WN ee etsten ee hc-a Sots) «ods 5 —-—-—,, Diabase in ............. 101-103 Sa aT sy SR a 92 ——-—, Gabbroid diabase in ..... 102-103 625 Page IloLYOKB trap sheet, Glass bearing por- phyritic diabase In........... 101-102 — — —., Holyokeite base in .......... 106 — — —, Inclusions in, Types of..... 92-97 — — —, Lithophyse in .......... 105-106 ——-—, Map and section showing.... 126 —— ———- ——, _Palagonite In). «0. ls saws 103-104 — — —, Plumose diabase in........ 99-101 — — — — — and palagonite from, Pa- per on, by B. K. Emerson...... 91-130 —-——, Rock specimens from, View LES Ee oi oe daraare atic! oho eeehe pokemons 126 — —-—,, Schlieren in ............. 98-103 —— —, Spherocrystals in : ——-——, spuerulites in botryoidal ae MIRB Ey tere Sc oretick ers. hers sc daigre seen deals -106 HOLYOKPITE, Analyses of ........ 113, 114 —, Diagram showing > mineral constit- MGM GS eet ene ce ere eens wicv eral ueciaete 130 —, Dikes of, in Holyoke trap sheet. 107-103 —, Formation of, Theory of....... 117-121 ——, Rock associated with ...........-. 96 —, Thin sections of, Figures showing. 129 HoMOTAXIS; Use of term... ....7:- 138-140 HoNnEOYE FALLS, New York, Cornifer- ous limestone exposed at......... 52 HoPKINS, T. C., Title of paper by.... 574 HORAN quarry, VTE We brea o Site oie aisha: onal poem e cane 2 HORNBLENDE-DIORIIE-GNEISS from San Gabriel mountains, Character of 200-202 HORNBLENDE-GABBRO from San Gabriel mountains, California, Character OL STAN wtierets bras 55 GEE Ie 197-198 HORNBLENDE-GNEISS of P*edmont Penn- sylvania, Occurrence and charac- ter of 319-320 HORNBLENDE-SCHIST from San Gabriel mountains, California, Character of 203 HORNBLENDITH, Analyses of Hovey, E. O., appointed on Auditing Medina, New York, 2 ee 0 wi phe le -« ¢ « Wee «so 6 sp ew COMMITEE BO ee tel ava ease wel ee 540 —, Bibliography of literature of Mont Peleshy; Reference: tov. ss ict Al4 SO eee caterer eee . 158 626 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Page Page Huxuey, T. H., cited on term homo- JENNINGS formation of Virginia and CARAS Oe Cee ve eee eter 138-140 West Virginia, Paleo ae and Hyart, ALPHEUS, cited on Alaskan pa- leontology ni aac 392, 396, 397, 401, 402 HYDROCHLORIC acid, Presence of, in vol- CADIC ME VASMS Selo e euke aren ana eum ene 253 HYPERSTHENE-GABBRO of Piedmont dis- trict of Pennsylvania, Character CEG oe Gar Favetnle ehcp emit eke case 312-316 — — — — — — — , Distribution of. 311-312 — —, Photomicrographs Obeie ee ee Oleee mealies IcB-BORN streams, Lowering of divides by, in western New York..... 239-240 —erosion, arguments against, Sum- mary 2 Be COE SRIN A DEON idlaea Be Bink are 25-31 HIE, (SULTRY @E fobs ease oe 18-19 —-—in western New York, Character bile Ah cee cp Me, a fe an cc NN wet cat cA aaa 237-239 —— theory a fallacy ; 2 EE ae, Hain OU ee lodenen aeabe co Mieale ecoreneiaah ont eae 13-74 IpDINGS, J. P., cited on lack of natural division lines among igneous rocks. 278 —, Classification of diabase by....... 116 ~— CCOLG Ok aLeManksy DV ose eck ein 574 INDIAN quarry near Syracuse, New York, Onondaga limestone exposed in, Wiew SHOWANS ha ciciepeus « -aerelalste 64 INVESTMENTS, Report JOM) ic lemcel ses) ous 536 TRON creek, Utah, Fault om........... 523 — — — — — , Section showing ....... 524 ————, Stratviorapnmy OM oy... Seve. ee Eyal —w— shales, Stratigraphic equivalents On iere One LR RAG Srna naree be ase ccd cue ne 529 ISABELLA beach, Fishers island, New York, Sections of.. 370. 377, 378, 379 ITACOLUMITH, Chemical analyses of.... 297 ITHACA quadrangle, New York, Repro- duction of photograph of model OES ais eee Shae Pace el es aL ele oe Kee ane omeeeale 230 IvES expedition, Observations on geol- ogy of Zuni plateau made by...... 477 JACOB sand, Age and correlation of 388, 389 ——, Character of .............. 379-380 — —, Conditions of deposition of... 378-3879 =— =— |Correlation: OF... 25666). aes 386-387 — —, Exposure of, View showing...... 367 ——, Geographic distribution of ...... 390 Se INDIRA OL ics os anlar eta sy iro le meena Serie ee ats === OCCURTCNCEAOf, Koo ster cee. Hehe 380-381 -— —, Sections showing.. = IMC K MESS OR ye — cys cae eee ee ate JACOBS and AIMS, Sections of rocks at New York city prepared by, Refer- CINCO MUOM rie cel br estes ME we eders raise ale Pe ae 167 JACKSON, Dr. C. T., cited on Mesozoic FOSS Gerona Wy Pa ae ees cists o coe cee JADWIN, Captain HpGar, Acknowledg- IMOMES “CO! Oey stats Weraeierw ashe le le acaliat ene 164 =—- AIG: TEMGEKEM JPY iy. . o\hve clause Scien 171 JAFFA, Asia Minor, Stone reefs at, Ref- CECMCO: GO! asus ie) atalenepeh diene take Gis c,h JAGGAR, T. A., quoted on cause of erup- tion of Mont AFOUC RI eas Ae lan eb iawn 281 — cited on the spine of Pelé.......... 567 —, Study of West Indian volcanoes by, ReELeren Cento seein 2 Oe ee ers Shean 244 JAGNAUX , cited on asbestos...... 430 JAMECO ’ gravel, Age and correlation OE Sete ee ere ined ie age eect 388, 389 —-—, Character and occurrence of. 373-375 —-—, Geographic distribution of...... 380 —-—, Sections showing .......... 371, 374 =~ ==, (PRICKNESS).Ghelekns aeons 373 JEFFERSON, MARE §S. W., elected Fellow. 540 —, Hanging valley due to stream ero- Sion pointed out by ....... Cecisie arte geologic correlation of. 140, 141-145 JERSEY flats, Rocks penéath’ =: seaause ut —-—, Map showing JOHNSON, W. D., cited on cirques .... 24 JONES, C. H., Analyses of chrysotile by 443 JONES, MOREAU DE, cited on geology of oe eee eee wee the Windward islands .......... PALE —-— —lost Antillean continent...... 258 JONES, R. H., cited on serpentine rock of Canada: is. vk ool eee hos JUDD, J. W., cited on ice erosion....... Ly Juglans brandonianus, Description of ae JULIEN, eat A., Acknowledgments to... 419 — cited on fracturing of rocks at Spuy- ten Duy vil) ‘ereek) .)2 oa 155 —; Determination of brucite as a rock constituent [abstract %22. ~ same 586 —, Map of New York city prepared by, Features Of: (002)... ee Se cree eee 181 JURASSIC rocks and fossils, Occurrence of, sin’ Alaskaci2. ise 396-407, 409, 410 KANAB valley, Utah, Sections of Meso- Zoe Strata: dn. 2. eee eee 484-486 KEITH, ARTHUR, Aid) by eee coe eee 329 — cited on age of probable equivalents of Wissahickon schist —-— — geology of Piedmont district of southern States: ...\o/. cue siareneree eae 293 —, Field conference with, Results of 327-328 —, References to work of ........ 333, 344 Kemp, J. F., Acknowledgments to..... 419 — cited on composition. Of serpentine.. 444 e mountain, Vermont (2.9. ace See eee 429 ——— origin of chrysotile veins in serpentine: 0.3 Ue eee 134, 136 — — — — — Wakernaas around Man- . hattan island)... 20... 154 — elected Councillor .. 2a .4e ee eee 540 —, Paper by, on Geological bookkeeping 411-418 —, Record of remarks by .... 565, 574, 576 —, Reference to work of ............ 282 ==, Titles: of papers by, 2a seme 562, 574 —, Translation from Gautier by...... 285 KrnAI formation, Character and thick- ness Of) o. 0h) oe ee eee i KBERATOPHYR, Analysis of............. 114 Kaen wales California, Glacial cre Tene CREEK canyon, Oregon, aire valleys along |)... soe eee ee 83-87 KING, CLARENCE, cited on geology of Uinta mounitaims= aoe. 487, 517, 525 — quoted on geology of Uinta moun- tains KING BROTHERS mines, Thetford, Can- ada, Chrysotile-veined block of serpentine “from 555.4 meee 131-136 KINGS bridge, New York city, Limestone ee BIG. oe wis see edie e's eae ee KNIGHT, W. C., quoted on Red beds of the Laramie basin .......... 489-490 —, Resolution i death: Of. 3. die) oy Cee eee 592 Knopr, A., and P. THELEN, Title of pa- Per PY. Fl es we: ke eed 593 KNOWLTON, F. H., cited on Alaskan fos- Sil) plants. ik he eae eee 401, 408 — — — fossil Vermont. 6 ov ioc hints dieeeue: Gee ee eee 510 —, Lignite deposit at Brandon, Ver- mont, and contained fossils de- SCLIDER "DY: Vaiss. arte revo Neves AOS os 500 INDEX TO Page KNowLtTon, F. H., quoted on age of lig- nite deposit at Brandon, Ver- OU SS ee eters ae 502, 504 Kraus, EDWArD H.; Occurrence and dis- tribution of celestite-bearing rocks [abstract] —; Origin of the caves of the island of Put-in-Bay, Lake Erie [abstract]. 563 —, Record of remarks by .... 563, 574, 575 KUMMEL, H. B., Record of remarks by. a BS, Cre eee MN) « OO) ey a Be eee. Kunz, GEORGE F., Title of paper by... 583 LABRADORITE, arrowhead twin of, Fig- MNTERROR ETI ek ae Saei are ch elule 0 128 LA CuHicotTnp, H. A., Aid rendered by.. es LAcrRoIx, A., cited on serpentine...... 440 —, cited on the spine of Pelé......... 568 —-— — zirconium in riebeckite rocks. 576 —, Observations by, on gaseous exhala- tions of Mont Pelé.......... 52, 203 —-———,on temperature of lavas of LJ EG See eee 253 —, Study of West Indian volcanoes by. aye LAFAYETTE, Pennsylvania, Soapstone quarry near, View showing....... LAKES, Glacial, Implications of, as to eroding power of ice .......... 23-24 LA Morte peak, Uinta mountains, Con- Pee HORT nce. Soy desc Sok «8 525 LANDER, Wyoming, Fossiliferous Tri- MENOMINEE AI! Sci 25 seid. dee See ee 489 LANDES, H., cited on carburetted hydro- gen gas in ashes from Mont Pelé.. 252 LAND, A. C., cited on the lava of Pelé. . —, Record of remarks by........ 562, 563 573, 574, 576 —, Resolution of thanks offered by.... 590 LANE, A. R., Discussion by, on plumose SEIT Pets iS )e a selec cs Gie @'e' « 125-126 LANGTON, D. W., Record of remarks by. 562 La Puata formation, Character of. 469-470, 473 ———~ Mauivalents of ....... 00005... 496 i ICRTIGSS, OF sw )t 2 neice sis cele 469, 473 — quadrangle, Colorado, Geologic work UE op Sa ne eae eee eka 451-452 LARAMIE basin, Red beds of...... 489-490 Lavas, Sulphurous acid exhaled by. 251-252 LAWSON, A. C., cited on cirques ...... 24 — elected Secretary of Cordilleran Sec- Soa Or cores, a uc Saha e debwee dia ie o's 592 — quoted on canyon of the upper Kern PREM ee ajstia iakvatersls oletciete , 38 eee GE PANer DY 6... sce eee el 594 Lawson, A. J., cited on glacial erosion my Raerra “Neyada’ ...5...035..%- LE CoNnTE, JOSEPH, Reference to opinion of, concerning ice erosion of Yo- TE OS a ee 14 Len, W. T., Red beds traced into New (PES SE UN Aa PS eae es a 493 LEIGHTON, GEORGE, cited. on rock be- neath Harlem river, New York... 158 LEIPERVILLE quarry, Pennsylvania, Granite-gneiss exposed in, View SEAMED pects oie, Shave GN chahdes « 309 LEITH, C. K., Record of remarks by... 565 —, Title of paper by ...... ee eee 563 Le Roux beds of Arizona, Character SMES PCR IORN Gin ord 5 685.6 alas bce ks 481 ——, Geologic equivalents of......... 482 Les CHENEAUX islands, Drumlins on... 577 LesLey, J. P., cited on eroding power EGO Siren ane cndia eal cic hs! ont 17-18 LESQUEREUX, LEO, Fossil plants from Brandon, Vermont, described by 499-500 LXXVI—BuLt. Geox. Soc. Am., Vou. 16, 1904 VOLUME 16 627 Page LESQUEREUX, LEO, quoted on fossil plants from lignite deposits at Brandon; “Vermont Visi aan eee tues! 502 —, Permian plants from Colorado iden- tified’ Dy nS Oue ae inal he ae ee 90 LEVERETT, FRANK; Drumlins in the Grand Traverse region, Michigan _ PHDSELACEI Ge As Gree eA oer 577 LIAIS Paper by, on stone reefs of Brazil, Reference: to! vi.ici5).% ste. 12 LIBRARY, Accessions to 595-604 LIBRARIAN, Hlection of H. P. Cushing BSE See Sa fas vg ael alee ABE lane oder duane esa erates 540 == REPOLGs Of: is silos eis IOs ae eee ee 539 LIGNITE deposit at Brandon, Vermont, BCH OLY) dig th of ceed tn aL Sohne aera 502 — — — — — , Bibliography of........ 514 wee Character Of... 10.4 eDO2-o0e — , Deposits associated with 501-502 — , DOSSHS OL O44 hose Oa oe — — — — — , Location and extent of 500-501 —, Tertiary, of Brandon, Vermont, and Sts; LOSSES Fe oO Pe iS 499-516 LIMESTONE, Compressibility and plas- tic deformation of, Experiments GIT Te repens Fa Cee ny Se ene eee 564-565 LINCOLN, D. F., cited on origin of the Hinger Takes vaek.& sere epentsteieke orn Se ti LINDGREN, WALDEMAR, Acknowledgments Lie od ao ee a hatat ee Rear el au atere — -—, Reference to work of.......... 282 TSH OL WeEMOWS Ae ose Sid 2 states Serale ene 606-616 a OTMCET SE. Sch n tal a so Cette sa ele e Glace aes 605 LITHOPHYSZ, Formation of, in Holyoke trap sheet, Theory of......... 119-121 —, Occurrence of, in Holyoke trap SHGEE esc de oe ctawe rahe eee cl ouehare 105-106 LOcKPoRT limestone, contour of, at Ni- agara escarpment, Diagram show- LIRR Seed othe Sloe cua be merle wie eae 52 —w—,, Failure of ice erosion on, Views SHOWIN ichere-cuelne 6 hele ehanete eer ead 52 Loess and associated interglacial de- posits [abstract]; B. Shimek....... 589 LOUDERBACK, GEORGE D., elected Coun- cillor of Cordilleran Section....... 592 == PIE (Ob Paper, Dyes is ieee ee Soils 593 Lows, Mount, Biotite granite-gneiss EV OMUN IHS Seto PCa so arb eho ch aeehte me Oe -—— Granodiorite from ............. 194 —— Quartz-monzonite from ........ 191 Lucas, F. A., cited on fossils of Dolores Tppmation; Colorado ic). ss sae as 468 —, Triassic fossil described by........ 476 —, vertebrate fossil identified by....... 493 LUNDY canyon, California, Hanging val- LOVS ALOIS GP atcyye eiaeink o/ciehe) wa cewek 89-90 McCAauuby, Henry, List of papers by 557-558 aot ROMER iE of ou eo Uae a erased a eis! PRs 555-558 McCaskby, HIRAM Dryer, elected Fel- Sf Seca abate ahah okacite alae kPa aia) ee aces ERIS 5¢ wee. 6 et « 8 ee oe eee Clee SS MACDONALD, CHARLES, Aid rendered by 179 McELMOoO creek, Colorado, Formations OEM aig Carakete shies, erect ed oeeatel waa reaeee — formation, Colorado, Character thickness of — —, Equivalents of McGern, W. J., cited on glacial can- MOS: = kop aan Ci are Od eee 25, 29 — quoted on glacial canyons ...... 29, 30 MAGMATIC hypothesis, Statement of 287-288 MAGNETITE, Skeletonized, Figure show- DS ae Sd ha ae eke eae ee 128 MALASPINA glacier, Hanging valleys at 81 M Gre -8 oh 6 8 Chere ee a 628 Page MALLET, J. W., cited on causes of vol- CANIC.AGHONG @ a cied.cc cae tee 280 MAMANGUAPE river, Brazil, Bird’s-eye view Of rezion “about f..2.00% aeoee — stone reef, Views showing ..-..... Goat MANHATIAN island, New York, Chan- nels surrounding, Origin of.... 151-182 ——— —, Former hydrography of...- 181-182 491 389 —-—, Character and occurrence of. 373-375 ——, Section showing 371, 374 =e TERTEKHESS. HOfe) Jee e heehee ae Ske MAN-0O7-WAR reef, East river, New York, MANITOU region, Red beds in...... 490, ale aie ‘gravel, Age and correlation ey eee eee eens ROCK Si Ola Leer reas wusutolen ta ahaa 169-170 MARBLES, Compressibility and plastic deformation of, Experiments on 564-565 MARCOU, JULES, cited on Alaskan geol- OBE. Se hiwlersre © S05 eibeed Oe ae ele SU ae 391 ——— occurrence of. Triassic strata IGN ING W-ViCRICO “Ss nteieins o seni cece MARIE GALANTE island, Features of.. 269 MARINE erosion on Windward islands, Hxam ples sors s55 ts pee. auele bas 271-278 MarsuH, O. C., cited on geology of Uinta MOUNTAINS ad seius isles eee eae eae ee 519 —, Mesozoic fossil from Colorado de- SCLIDEO DYES vesac cole ees ee Oe ee 494 MARSTERS, V. F.; Petrography of the amphibolite, serpentine, and asso- ciated asbestos deposits of Belvi- dere mountain, Vermont...... 419-446 MartTIN, G. C., Record of remarks by.. 578 —T. W. STANTON and; Mesozoic sec- tion on Cook inlet and Alaska pe- ENTHS WHA ey te gs . Seaseit busi oleae 391-410 ———, Title of paper by .......... 580 MARTINIQUE, Geologic features of .... 269 ——; Marine erosion (OM: & osske. wee ek ie PA (Ps SN OPICIM (Qf c ccvenele cron eaee Ake ae 256 MARYLAND, Piedmont deposits in, Cor- TALON GF fs) toe eee 340 ——-——, Map showing ............ 33 — — ——,, Structure and _ structural Felasions- Okay {Sse eee te 339-344 —and Pennsylvania Piedmont forma- tions, Correlation of, paper on, by BS NRA WiSs ik eee ena lete 329-346 MASSANUTTEN sandstone, Age and cor- relation Ohne ses. i. cts eee ea ee 340 MATHEWS, E. B., cited on geology of Piedmont district of Maryland. 293 ; Correlation of Maryland and Penn- sylvania Piedmont formations 329-346 —, Field conference with, results of 327-328 —, Section of Cutler formation meas- ~ TOGO neta ee atom Ge oie Bite 464 =, PileioL paper py a. coe: eee tees 572 —'and W. J. Maer Cockeysville mar- Dle iain hile ne dake ane ea es 347-366 —— —, Title of paper by .......... 572 MATTHES, F. E., cited on cirques...... 24 MEDINA, New York, View in ea at 52 se plain, New York, Ice erosion on. 54-55 — sandstone, failure of ice erosion | on, Views showing ren hc hteia. trate Cr ne 62 MEMOIR of Charles Emerson Beecher : CharlessSchuchert “s-)-G 541- 548 —-— John B. Hatcher; W. B. Scott 548-555 — — Henry McCalley ; Eugene A. Sait Sie acne eae ete eee 555-558 —— William Henry Pettee; Israel C. RUSSell oes ae ee eee 58-560 —- Charles Schaeffer; Angelo Heil- DE. rss". Sete ee ee 561 MENOMINEE region, Michigan, Drumlins OF ORAS SUG See, ceteea ee eee eae ae 578 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Page MERIDEN type of inclusions in Holyoke trap sheet, Features of ........ 92-93 MeRRIAM, JOHN C., Titles of papers by 592 MerRRILL, FEF. J. HH. cited on origin of course of Spuyten Duyvil creek. 156 — — — geology of Piedmont district of New York .......--+-++-+teee: 293 —-—w— origin of waterways around Manhattan island y\ciseee seat 154, 155 MERRILL, G. P., cited on origin of fibrous gypsum in Caves .../12asn eee 136 —_— — — — — fibrous serpentine...... E35 —;On the origin of veins in asbesti- form “serpentine, 2 .s- ene 131-136 —, Record of remarks by ............ 574 —, Title of paper by ........-.-+-+-: 574 Mmsozorc section on Cook inlet and Alaska peninsula; T. W. Stanton ' and. G— ¢.) Martine 2) 391-410 METAGABBRO of Piedmont Pennsylvania, Character: Of) 235 256 elon ee ore —— —— — Distribumon sO lar 318-319 MICROPEGMATITE from San _ Gabriel mountains, California, Character OF | ais ene haere re 198-199 MIGRATION of shoreline, Definition of.. MILLER, BENJAMIN LE Roy, elected CHOW oo. soe eee MILLER, WILLET G.; Pre-Cambrian rocks in the vicinity of Lake Te- miskaming, Ontario [abstract] 581-582 —,E. B. MATTHEWS and, Title of paper 1?) Ga isis oc 2 NETRIGER EV a erelicee testo MATHEWS and ; Cockeysville marble 347-366 MILNE, JOHN, cited on causes of — ? eanic action 280-281 ——-— earthquakes in Saint ve preceding eruption of Mont Pelé. — — — Windward Islands MOENCOPIB beds, Age of.............. — —, Character and thickness of..... oop b. ale ee LD oe oe elim a, oe) mapa) Oe eee MOoISSAN, , cited on gases in lava of Mont ‘Pelé. ..... 52 eee 252 Mono LAKE basin, California, Hanging valleys im <2. s..\. see “88-90 Monocar pelts gibbosus, Description of 512 ———— Bicure of) 5...) eee 516 — sulcatus, Description (Of Ge aeeaeer 512 ——, Figure Of 2. 56a eee eee 516 MontTAUK drift, Age and _ correlation Of. ORES iihag SP oe eee 8, 389 — -—, Character and occurrence of. 383-384 — -—, Conditions of deposition of..... 383 — —, Definition of .............. 382-383 — —, Distribution of —3. 2o.ae ae eee 390 —- —, Thickness ‘Of <. 2. . cage eee 382 ——, Section showing .......... 371, 380 ——, View showing exposure of ..... 36 _— Point, New York, View showing Mon- tauk drift at MoNTEREY formation of Virginia and West Virginia, Paleontology and geologic correlation of. 140, 141-145 MONTEZUMA valley, Colorado, Forma- tion: Im: 02 eho eee 475 MONTGOMERY, Henry, elected Fellow.. 541 Mont PELE, Present condition of, Pa- per on, by H. O. Hovey...-:-. 566-569 ——, Views OF... sas 2 ais ye hee MoRAINES of the Seneca and Carey lake valleys; R.-S.. Tarro. 0. 215-228 MoraAInic fans, Occurrence of, in Cay- uga and Seneca Lake valleys, New Vork) to2.. 06.52 eae eee 224 Morrison formation, Equivalents of.. ——, Geologic horizon of Mount DESERT island, Maine, Rocks ie by Persifor Frazer: 2. oe sen 83-585 INDEX TO VOLUME 16 Page M¢ecnt Lowe, Biotite granite-gneiss from Grenoediorite: FLOM «6 oj. sss occa ss — —, Quartz-monzonite from......... 191 — —railroad, Granodiorite along.... 192 MounT WATERMAN, Biotite-granite from 190 —-—, Granodiorite near ......... 194-195 — -—, Hornblendite from Mount WILSON, Granodiorite from 192° 193 porphyrite —_ — ? 9 6 BP 6 ieee) ee & ——, Quartz - hornblende from 199-200 — w— trail. Hornblende-gneiss from ... —-—-—, Hornblende schist from...... 203 Muir, JoHN, Reference to opinion of, concerning ice erosion of Yosemite valley Murr glacier, erosion |i es CA gh = aa a Murcoc!i, G. M.; Suggestion as to the origin of riebeckite rocks..... 575-576 SpeTRi me BO. ee ‘eee ye oe b. 85.6 Be 0 Se See) Se ie ae 6 we a eine Ce ere © -NAKYEK formation, Age, character, ex- jent, and thickness of.... 402: 407, 410 —-— Distribution of, on “Cook inlet, Map Ca Ee ee eer ee 395 Peeve Coe are aah cee 407 ——, ’ Sections MBL eR hae eA Sian, oars 403-406 ——, Views showing ....... 395, 399, 407 NANSINN, F., cited on glacial action in Ee Os a he NEPHI LINE - SYENITE, Compressibility aiid plastic deformation of, meat? LES Se A A ee 564-565 NEUMAYR, M., cited on Alaskan paleon- oo | ee Set eee 92, 401 — —-—-ice erosion ................. aif NEVADA, Paleozoic strata in, Columnar ee eeens SHOWIME . 0... soe ce eas owe NEWBELRY, J. S., cited on formations of San Juan valley, Colorado..... 476 —-—-—-geology of Arizona and New _ 2 LAS Ee oe ae eee 480 — — — ice SEOSOB ns ne gah isle ee wes 16-17 —-—— origin of waterways around MaMHALIAN, IBIANG.... 6c oe Sb e5c 8 154 Red beds of Colorado... 449, 475 —, Explorations in Colorado*by....... 476 —, Observations by, on formations of Zuni plateau, New Mexico......... 477 —, Red beds of Colorado observed by. —, Triassic fossils found in New Mex- IIONL aE ES chsh a Regn i Si aa Mining and Milling Company, Asbestos deposits Pre. 820. e 6 6 eo a Ole ie ive ve he ee o 0.6 = 5 ACW OD Ac 2s. NEW JERSEY, oe Cretaceous forma- tions of, Classification of ........ NEW Mexico, Triassic beds of .... 494-495 —, Zuni plateau in, formations of. 477-480 NEw YORK, Central, Drainage features MIRREN bic he LAS ete tiv s5 wren cass 229-242 ——, Failure of ice erosion in, Views illustrating ea ey ee Oe 2 — —, Physiographic belts in, Map show- ing aS) i Re eee ee eee 5 — Eastern, Ontarie or Siluric section of, Note OM eM attakscbaiaie-ci/s les ckcte ee 582 4 Oe sheet erosion. In ..........<. 48-73 — city, Channels surrounding, ign re Se ee ee 151-182 —-—, Water front of, Origin of, Theo- ries RONCOR IN ey obec ela erat 153-155 — — — — — , Rock floor beneath, Form ee ee ee ee 155-179 — drumlins [abstract]; H. L. Fairchild 576 —and New Jersey bridge, Rocks dis- closed by borings for ........ 178-179 —_— , Section of river at 178 14. 42-43 Page NIAGARA escarpment, Contour of Lock- port .limestone alone oi)... 2 05. ss —w—, Diagram showing profiles of.... 53 —= ==, [ce work: GlOmg) 2k wcimesieice se 51-52 NITROGEN of gas well at Dexter, Kansas [abstract] ; Erasmus Haworth.... 572 NORITE, -ANBlyBIS GE erie bre erecta ats 314, 315 NorruH river, New York city, channel of, Origin: oF. SkAS i sane hee 180-181 Norway, Glacial. érosion..tn °....5. « «ss 33 Notr on the Ontariec or Silurie section of eastern New York [abstract] ; Ci: As artna@erel:s siis-c ites er ccaretons 582 NOTE-BOOK, Geological, 2-416 Nyssa crassicostata, Description of. 509 Sa HISTO! OL, Tihaj ane al our omevanee once ie keke 515 — jonesii, Description Of: ode. es 509 = TES Obs -oticwsegaeyeruononeiate terete econ 515 — lamellosa, Description of ......... 509 ems NTE EE. OL, oe ah ss iene ere eee eae 515 — lescurii, Description of ........... 509 ray BASUTS “OL, vies Sve cet ascre vate ai eines 515 Oak creek, Uncompahgre valley, Colo- rado, Sedimentary beds along. 457-459 OBER GRINDELWALD glacier, View of. 32 OccURRENCE and distribution of celes- tite-bearing rocks [abstract]; Ed- RSE CEI UG PATI ova eee tales ecaretcca 22 574 ORES, vB CUOMN OE och eles eerie. ocs.058 540 —, List Sa oes ee ee She 605 —'of Cordilleran Section, Election of.. 592 OGDEN quartzite, Age, character, and thickness of ...... hy Oise rue Teas oe 520 — —, Stratigraphic equivalents of 529 O {ARRA, CLEOPHAS CISNEY, elected IETLOWES ae ose A auckutsv ecco Mids. 2. ache) diets 41 OIL bay, Cook inlet, Alaska, Enochkin formation at, Section of 400 , Naknek Section ERE ES. Rs Toe RAE EOS 406 OLD BALDy, Sierra Madre, Elevation of 185 OLFERS, » VON, cited on Pernam- Guba teeth cae ts | ene 11 OLIVINE, Occurrence of, in serpentine, WIE WSU SRO WHO oo. ss. aa bee Stee 317 ONEIDA and Shawangunk conglomerates, Relative ages of — grit, Exposure of, View showing.. 52 ONONDAGA limestone, Exposure of in quarry at Auburn, New York, View SSA OWES Ss og sesh Sud eet Gis ol a arora ee — valley, Failure of ice erosion in, View showing ONTARIC section of eastern New York, INGER OTE Hoes ahr Macare. eG) shal nce ec wlen 582, ONTARIO, Lake, Height of surface of, ea sealevel, and greatest depth a NUS rf eb cha witcha to wh ct alte wicalen ob ation dad! Ghagetens a ORDOVICIAN rocks of Piedmont Pennsyl- vania and Maryland, character, dis- tribution, relations, and thickness of.. 3801-306, 335-338, 349-351, 354-357 — , Correlation of . 840 — , Features of... 301-308 ORIGIN of the caves of the island of Put-in-Bay, lake Erie [abstract] ; yQdward H. Kraus —c_ the channels surrounding Manhat- tan island, New York; W. H. RUORDS? (paisa eta le nate ae 151-182 ———veins in asbestiform serpentine; Gio Px Merrill ss oan ute teek eee 131-136 OSMONT, VANCE C., Title of paper by.. 594 OuRAY, Colorado, Sedimentary beds ex- POSCGETNOATY co. %:3 a iSid oud v steed — —, Unconformity in Red MOST S26 os hie. ee he ea 453- 459, 495 ib | 456, 458, 460, 462 oo» acs ee Se. Si Npehe@ & oem Age Of >). Mi.) Utes PEO-4o5 — — — — — , Character of ..... 448, 452 — — , Correlation of, with beds of Rocky. Mountain province. . 488-495 — — — — — Formations composing, Correlation “Of an bere fe we ee 470-496 — — , Description of 459-470 ————— oo , Literature concerning 449-453 —— — — —, Thickness of. 448, 449, 452 RED CREEK quartzite, Occurrence of... 525 Rep DirT creek, Colorado, Fossiliferous Red beds ont. Noe eee 488-489 REEFS, Stone, on northeast coast of | 55) A | POS DEORE SRS & ee tenis hye! bkeacic -12 REGISTER of Cordilleran Section....... 594 — -— Philadelphia meeting........ 590-591 Reip, H. F., cited on erosion by Muir SIACIOM 3a hwdecc Ghatake Wanted eee ee ee — — —flow of glaciers.............. 25 REID, JOHN A.; A detail of the great fault zone of the Sierra Nevada Pastries ected ede tae ernie eye 593 RELATION of lake Whittlesey to the Ar- kona beaches-[abstract] ; Frank B. NAVIOTE ser ee ayer sitee cc fe oes 587-589 RELATIVE ages of the Oneida and Sha- wangunk conglomerates [abstract] ; PA WV BEAUTY) eye etsce tens let aevice eahar 582 REporT of Auditing Committee. 565 —— Council Ue ee thane ofan Ck ee ae 532-533 Pe ctaierer es aacee cand a se Bieierekaats 5388-539 — — Librarian Wis PAPA Plc alah acres eee ora 539 —-— Photograph Committee ......... 590 et OCT OEME Ys | alighcvatats a scate ene athe ere 533-535 = PT CASUEEI ii sis .cl ataie: vs ceeeleis ss 535-537 WESOLU DION Of, Thanks. 22)... <2 esos 590 RuHopES plateau, Uinta mountain re- S10, Geolosy -Ofiss . se 523, 524 KHON) Sacer; View (Obs o's 3 cere n chia 32 Rieu, G: S:, Aid rendered by... 2. i>. * 162 RICE, W. N., Record of remarks by. 562 Rico’ formation, Age, horizon, and fos- SUS Ode eps wee eee Sade GS ee ee meas 451, 452 — -—, Character and thickness of...... 452 RIEBECKITE rocks, Origin of, Suggestion ESE LOM Tso aetedevale tater mrcnccateleeyietcts. 5 575-576 Ries, HetInricu, cited on geology of Fishers island, New VOLK. . sans e iL i ( ——-— rock beneath Harlem river... 161 Rio GRANDE DO Norte, Sandstone reef Lie VIM Olea oer devers coc devecie yy ce Bee RIo MAMANGUAPH, Section near mouth OL bird s-eye: View Of J. des. ee. Rivers, Action of, compared to that of glaciers ORS aso nay egadeie dB aOR eT 21 Rogpsins, A. A., Aid rendered Dye ater 169 ROBINSON, B. Fi, Acknowledgments to.. 506 Rock basins, Implications of, as to ice CVOSIONS Ss sNas cee toy th tea ale cle ee 24 Rock creek, Uinta mountains, Fault TIRE Nes cretets. ate aie 8 ais Oops es aie ee a aaa 523 524 ROCKLAND, Maryland, Map of vicinity OEM sich A Sere! oa IS aSet ak he eran at otto 365 Rocks of Mount Desert island, Maine; PeLsifor \PLAZeL ys ssc ael ee eee 583-585 Rocky mountains, Front range of, Red PEGS dies iat), lakes cuek eres 490-494 — mountain province, Red beds of, Cor- FOUPLIOM (OF. «ss. \chu steve utara, tls 488-495 632 Page Rocers, H. D., cited on geology of Piedmont district of Pennsylvania. 296, 301, 306 ROMNEY beds of Virginia and West Vir- einia; HMossils Obs te ante eeateliel mesh 140 — — — — — — , Paleontology and geo- logic correlation of....... 140, 141-145 ROSENBUSCH, H., cited on palagonite of Sel PAC AE Ne 250 .e edlomake hiesentiers 121, 124 = SOL DOMAIN GY <5. csi tres siosdeetagisn wanes 441 Royau Society oF LONDON, Commission appointed by, to study volcanic eruptions in “West Indies, Work of. ae RUBIO canyon, San Gabriel mountains, Hornblendite from, Character of 197-198 RuUFFNER, Capt. H. A., Red beds of Colo- TAO! OOSEH VEO Diy. Bhi ce lesaeteueeaels 449 RusH CREEK canyon, California, Hang- ine. valleys a lOmey 22s iy a eran iciene 86 RUSSELL, I. C., Bibliography of litera- ture of eruption of mont Pelé by, RGfETENnGCenGO! eens ews come ie este tees 245 —, cited on amount of solid matter dis- charged by non Pelé during erup- GION Of WOOD src pestcs erence oe suede aie enebals 254 — — — causes of volcanic action..... 280 —-———eerosional phenomena of Muir HACC as WA tees. Seale, rare taniny eye 42 — — — flow of glaciers ............. 25 — —-— Mono valley, California. 86, 88 —-—— origin of canyons about ‘lake INO TN Oy i sd So Yossi witeikepal aad er amemera leben sativwenn ions —-—-—rock beneath Hast river, New CONSE iia ledes si tats te MRS abet: ah spate anna ral eee eRe 1738 —— —section across Harlem river, ING Wi MORK = crypts oe er stare ee att —-——the spine of Pelé............ 568 —-—-— underlying rocks near New MOrk .Ciby. tiartiera waia tere sien beetle « 161 —,; Drumlin areas in northern Mich- igan: \Pabstracel). ace, H. ceva. 577-578 -vblansing Valleys: oc cece obs cectaerelts 75-90 ; Memoir of William Henry Pettee 558-560 —, —, quoted on glacial: erosion. = acco 46 —, Study of West Indian volcanoes by. 244 —, Record of remarks by......... HUME ale ==, DEtLe SOLA PAPEL sDVce serene oe aiden eae tee e 573 SABA IslandoV lew tOhci.c sk uid oa see 269 SAINT CHRISTOPHER’ island, Geologic FEACUTES Ol c aae hae leeeds tdsany awake eis SAINT HuSLATIUS island, Geologic feat- ULES se OL te ecenaterre che at at ne eGR eneree se he 269 SAINT LAWRENCE-SUSQUEHANNA dnvite, Lowered character of ........ 233-237 Saint Lucia island, Geologic features 0) —-—, Soufriere of, Paper on, by E. O. Hovey PRL Or oe ee 569-570 ——-— —, Views of ............ 569, 570 SALIFEROUS sandstones of New Mexico, POSSI OR aia g issn omen mec rene renee ATT SALINA~ shales, exposure of, at Hast Syracuse, ‘New MOM rah ta so ecekerowsietene 52 SALISBURY, R. D., cited on glacial ac- bron ins Greenland 2. acc oo eee 34 SAN ANTONIO peak, elevation of....... 185 SANDSTONE, Compressibility and plastic deformation of, Experiments on 564-565 — burrowed by sea urchins, View of... 10 SAN GABRIEL mountains, Age of... 188, 203 ———,, California, Crystalline rocks of, Paper on, by Ralph Arnold and JA; MsiStroney. Misi wo etieroae 183-204 —— —., Dike rocks of............ 198-200 — — —, Location and general features (0) at a Pa eM.) Aertel, Paces tank 184-185, 203 — — —, Metamorphic rocks of.... 200-203 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA = Page SAN GABRIEL mountains, Rocks of, Pe- trogztaphy (Of |... 2 aecbeee eee 189-204 ———, Sketch map of.............. 186 — —-—,, Topography of .......... 185-187 SAN a uAN tuffs, Colorado, Occurrence ae OF: on 0 ol lena wileces OW MO eee — w— region, Red beds of, Correlation of, with Red beds of Plateau _ re- PION: 2 aideecees oes s ell eee 471-472 —-——, Rico formation in....... 452-453 —-—-—,, Triassic unconformity at Ou- ray. in) a4. J sd da eee 453-459 — —yalley, Formations in. . 495-477 SANKATY beds, Equivalents of, on Fish- ers island, New Yorks. 7 aeee 386-387 SAN MIGUEL valley, Colorado, La Plata formation in ...... see 469, 473 — SAN RaragE. hills, Biotite-granodiorite from, Character of —— —, Location’ of. .i:.4aeeeaee 185 Sanro AGOSTINHO stone reef, View of.. 3 Sao Francisco, Rio, Regimen of...... 6 Sapindoides americanus, Description of. 508 ==, Figure: of «5.0. eae Hee ae U5 — medius, Description of............. 507 — —-+, Figure of 2.5.3.5) 22m een eee ras — varius, Description of ............. 508 — =; Figure Of 0 sc.) s, Serene eee 515 SAPPER, Karu, Study of West Indian — volcanoes. by 2.22 ¢, ahleeniciee ieee 244 SCHAEFFER, CHARLES, Memoir of, by An- gelo Heilprin ........5.4 ahora 561 ScHisT, Analysis Of. 355 eee 424 SCHRADER, KF. C., Alaskan fossils cok lected DY (2s. 0's 20d sOhe eee 402 SCHUCHERT, CHARLES; Memoir of Charles Emerson Beecher..... 541-548 —, Record of remarks by.............. 582 SCOTLAND, Glacial erosion in...... re ebearey Scorr,. W. B:3 »Memoixs fore wionners: Hatcher > sacs oe ee 548-555 SEAMAN, ARTHUR EDMUND, elected Fel- TOW ooo. ee oS Oe ee 5A SEA urchins, Sandstone burrowed by, View Of © sssps sd «ci s,e) ciel eee eee j — water, Density of, Notes on........ 7-8 SECRETARY, Election of H. L. Fair- Child: AS & ..'s occ. cee eee ee 540 ==) Report of :,...../... .2,ompaeeeeeee 533-535 SECURITIES owned by Society, Report (0) 0 ee Ene dis cos os 0 8 SEDIMENTARY formations, nomenclature and classification of, bearing on new paleontologic facts on, Paper by HB: S!) Williams one ee 137-150 SENECA LAKE basin, vertical relation of, to Ontario lake basin, Diagram SHOWING, i055 4). ba. ss a ee eee 6 — — —, Cross-sections of . 61, 231 — — —, Discordance of, with valley of Cayuga lake? sss 05. 2 2S eee 241 —'——. —, General features’ Of... 02%. see 216 — ——, Glacial ice motion in, Direc- tion Of . Ssicint hr AG Ree 216-217 — -— —, Hanging valleys in...... 230-233 — — —, Ice-dammed lakes in ........ 220. ———,Ice occupation of, History Of 20s oS ee eee 217-218 — —-—, moraines of, Frayed ..... 219-220 ae Bee , Gener al features of. 218 — = =. "in lateral valleys.... “990-223 — — — — — , Lateral ..\..0....065 een — — , Marginal and _ outflow channels associated with ........ 228 — — — -— — Origin of. ......... ue == —— -——, Moraine sfans il. 0-5-0 224 ——-—-— loops in, Terminal........ 224 — Nunatak, moraines ihre 224-225 —-— —— Outwash) sravels ins.) eree 228 —and Cayuga Lake valleys, moraines of, Paper on, by R. 8. Tarr. 215-228 INDEX TO VOLUME 16 SERPENTINN, Analyses of ............ —asbestiform, Origin of veins in, Pa- per on, by G. P: Merrill...... 131-136 —, Block of, Plate showing.......... —, Cross-fiber blocks of, Views. show- MM ede ues ate A Fh la tea ace 6 clase 435 —, Economic importance of .......... 422 —, Fibrous, Features of .......... —, Massive, block of, Plate showing.. 131 —, Microphotographs of ......... 0, 441 SS——IMAGrOSEFUCTULE OF ws. cece 440-441 cen Du 2 435-4386 —, Slip-fiber blocks of, Views showing. 434 —of Belvidere mountain, Vermont, Pa- per by V. FEF. Marsters on...... 419-446 SERPENTINES of Piedmont Pennsylva- BPM HAFACTET 2 OL ©. i6ie. 4 Sc. eie ore!» 316-318 =——-— — —,, Distribution of .......... 316 SETTERS quartzite, Age and_ strati- 330, 339, 340 SE cs 333-334, 349 332-333, 354-357 graphic place of.<.:...; ——, Character of — —, Distribution of .... SEVENTEENTH Annual Meeting, Pro- EMMIS CLIN: fois is etoiv wd slecO ia Se wl ese 531-616 SHALER, N. S., cited on causes of vol- ae 8 6 ws wile w 6) wae eee ss SHAWANGUNK and Oneida conglomer- Stes, belative ages: Of... 0.28... 2 > SHEDD, SOLON, elected Fellow......... SHENANDOAH limestone, Distribution of, in Maryland and Pennsylvania, Map showing SHIFTING of the continental divide at abi Mentana [abstract]; W. H. Be. v @ Cw BB db) '6 eye we _e).e. ee SEL SOS ae eae 587 —of shoreline, Definition of ......... 207 SHIMEK, BoHUMIL, elected Fellow..... 541 —; The loess and associated inter- glacial deposits [abstract]........ 589 SHIMER, , cited on Mesozoic fossil MIRED ions os ov eye's wee ve uo e.e'c SHINARUMP group, Character and thick- MNLF aoc tela we sié, alah artis 478, 481 ee CATER MEICTIES, OF -. 6 ce 6 wcrc ety 482, 496 So rr 482 SES SS a 486 SHORE erosion, Effect of, on form of BeBe (SUMIACES.. 6. ci oo ae ec 205-214 —recession, Diagrams showing rate ~ eS eeee 208, 210; 212, 2138 ———-—— eypical: ‘cases: Of......600%. 208-214 SIERRA MADRE, Elevation of .......... 185 pa eMC SOL ac. oi «sso Sass \ASH,- 203 ——, Granodiorite from, Character RNR ore enn Sie hare. Sele ede e damed 2 193 — —, Metamorphic rocks of, Character IS PRN Oa ai ethic chs bc ee daca wralees 200-203 et BOGIES OL) oth «oS Gus sce oe leg od 189, 203 SIERRA NEVADA, Fault zone of........ 593 Sek. RET ee 35-38 ——, Quartz-monzonite from, Charac- ter of SILSBEY, SINBAD valley, Colorado, Red beds in.. pies ar on valley, Cross-section pro- eo Se eee eis Oo 8 we we ee eee oe sion in the Adirondacks ......... 51 -—of ankerite and quartz, 633 Page SnuG harbor, Cook inlet, Alaska, Enoch- kin formation at, Section of... 399-400 SoLLaAs, W. J., cited on increase of bulk incident to passage of olivine into SEM OMEN oi. hells, < 0/0 C8 aiorel ate Pereae tel 134 SOUFRIERD (LA), Guadeloupe, Features ee RMR a hase Fs. os sacwh ors:

enrolled as lite smem perc. tai oie cle 535 ; Moraines of the Seneca and Cayuga ” lake VOMey.Sie es on eo aeons 215-228 — quoted on glacial erosion......... 384-35 —, Record of remarks by..... Doe LM OnanS —, Titles of papers by PR ene kit Be Hilios oo TAYLOR, FRANK B.; Relation of lake Whittlesey to the Arkona beaches abstract] Sear eis eae mee 587-589 TELLURIDE quadrangle, Colorado, Geo- Logie s.5> 6 sheer sian ee teens 529 ete OUATTAITG AN. 2 \¢ alatw ina wae ae eyeneners 526 _——, Strata in, Correlation of strata in Uinta mountains with ..... 529-530 — —, Stratigraphic columns in... 520, 529 WASHINGTON, H. S., Analysis of ker- DCODN YT DY to vee) e Ss oosnerene see Crees 14 —, Classification of diabase by. 116, 117 —, Record of remarks by..... 574, 575, 576 WASHINGTON bridge, Harlem river, New Vork, Rocks, benen thes... sen hee 158 — — — , section at, SHOWING 2 dis Mater susie craeateraeevarcorenees WATER of vulcanism, Source of. 284- 286 WATERLOO, New York, Onondaga. lime- stone ‘exposed in quarry near, View STUO WTS. cate atavatien shaper enemas sree ores =. 64 WATKINS GLEN creek, Profile along.. Del — — quadrangle, Glacial ice motion in GUIPCCLIOMMOE” 3hteie tere: ei es 5 aneve nate 216-217 — ——., Hanging valleys in ...... 230-233 — — —-, Ice-dammed lakes ale eee a ehE 2a — — — —erosion in, Character of 237-239 — -- — — occupation of, History of 217-218 —— — —, Lowered divides in 233-237 — — —, Map showing drainage features oa, ere) he. Of MOPthern partyok . 2c. 229-242 a moraines and outwash gravels on northern part of...... 215 —-— -—, Morainic fans in ........... 224 — ——— loops in, Terminal ........ 224 — —-—, Moraines of, General features Ol. Pasar eee eats a ee Bie epee 218 — — — — — in lateral valleys of. 220-223 aa adi SM ic 0 | eee ee ee tin gd LS 8 — — — — — Saterall~. ss... foo. Pee a , Marginal, and outflow channels associated with......... 228 — OTIS (OP ie cece. sc Meee — — — — — Mes OF 12.5. .).° ALa-aas —-—w—, Nunatak moraines in.... 224-225 — ——, Outwash gravels in ......... 228 ———,, Topography of ......... 216, 230 WEAVER, CHARLES H., Title of paper by 594 WEBER quartzite, Age of..... 518, 520, 627 = MAT AGLED OL \' opal cre e)n dae Sip wis. oem 520 — —, Occurrence of 525, & ———, Stratigraphic equivalents of..... — —, Thickness of Wenp, W. H., Reference to work of.... . 282 ; Shifting of the continental divide at Butte, Montana [abstract] .... 5 WELLER, STUART; Classification of the Upper Cretaceous formations of New Jersey [abstract] ; Fauna of the Cliffwood clays [ab- stract] 5 , Record of remarks by West INDIES, Map showing submarine configuration of Windward archipel- POOL educa sei er'e ere nies a eaters 243 Wuirn, C. A., cited on Alaskan paleon- tology adie Xap athe tt et taratagaten re Sanat te 392, 397 — —— geology of Uinta mountains. é — -—— geology of eastern Uinta moun- tains 487-488 ane a we oe ae eee ewww ewe © Oa 6 Os) eS CR We ee wa Oe Oh we LXXVitByix. Grou. Soc. Am., Vor. 16, 1904 636 BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Page WHitTks, I. C., elected Treasurer ...... 540 —, Record of remarks by ............ 562 —, Report by, as Treasurer ..... 535-537 WHITEAVES, J. F., cited on Alaskan pa- LEONCOLORY eine sheets ee ees ot eee oreo 402 WHITE CLIFF sandstone, Age of ...... 487 —-——, Distribution of ............ 483 ———, Equivalents of ............. 496 2 (Na@MIN GE OF hs ei hiss ale ters wee were 482 WHITNEY, J. D., cited on erosion of Yosemite’; valley. | ei. cies orenetote sels 14 ——— geology of San Bernardino MOUMEAING Soe eras oie ee ecdhe hae he 187, 188 WHITTLESEY, Lake, Beaches of........ 588 ——, Relation of, to the Arkona WEAGCHES I tive Aer stot cto sees. bie ore 587-589 boat FREDERICK, cited on ice ero- Weide G. H., Acknowledgments to. 329 — cited on eruptive rocks of Piedmont PLACA: Asis shewale te eee ee ees 305 WILLIAMS, H. S.; Bearing of some new paleontologic facts on nomencla- ture and classification of sedimen- tary, formations: .9.).. 45 see 137-150 —cited on dual nomenclature in geol- OL Ys laeeitea'y seakel alien on ebee ead cc nia ered eles 137 — —-— shifting of faunas....... MBI(G DESte" —, Record of remarks by. 562, 579, 580, 582 —, References to work of. 353, 361, 365-366 =) Title of paper DY. sai; csc cue lee ee ee 562 WILLIAMSBURG bridge across Hast river, New York, Rocks disclosed by work OWS spss: Spe ee exes eee ereie Reap anensrione ers ib¢(al WILLIS, BAILEY, cited on origin of wa- terways around Manhattan island. 154 — quoted on lake Chelan .......... 39-40 —, Record of remarks by ........... 563 = DIE OF DADCE SDY s od tecsevessctva etss ache 563 WILLIS AVENUE bridge, Harlem river, New York, sections at....... 64, 165 WILLISTON, S. W., cited on age of Hal- LODUSS ei SR. AO Se te ee 494 -—— —-— Triassic fauna from Wyoming 489 =. PILE. Of SPAPeL | DYx mialev ie euserore wets) ees 589 WILLSEYVILLE creek, New York, Low- ered divide at headwaters of.. 235, 238 — — — —, Map showing valley of.... 235 WILSON, A. W. G., cited on glacial ero- SLOTS cre 2s a eeohenorans el ata eee eens 50 WINDWARD archipelago, Submarine con- HAUCATION OL ote ichoe Glenlece es 262-265 — — — — — Map" Showin? <..i5.- 5 een 240 — islands, Changes of level in .... 273-274 — — , Configuration Oi Page is 260-265 — —, Marine erosion in ......... 269-273 ——, Origin of ......... 258-259, 274-277 —-—, Paper on Pelé and the evolution of, yesh Ne MEETS ee awe ee care 243-288 ——~, Sedimentary formations of, Age Eg aaa EC IN SAPO UES RA OS ct 269 Page WINDWARD islands, Theories concern- ING fru es hea eee . 258-259 — —, Uplift and subsidence in .... 273-274 — —., Voleanic, Origin and arrangement of material forming ;..2 5. pee 265-267 — — volcanoes of, Antiquity of .. 267-269 2 ore the ee arta 277-288 WINGATE sandstone, Correlation of 478, 479 = Meatures. Of 1). 20 2 see eee 478 — —, Stratigraphic equivalents of..... 479 WISCONSIN deposits, Fishers island, New York, Features of ...... 385-386 — — — — — — , Sections showing .... 371, 74, 376 WISSAHICKON formation, Age and strat- igraphie place ‘of... 22. ee 306, 330, 335, 338, 339, 340 — —, Character and distribution of 302: 306, 335-338, 360- 361 — mica-gneiss, Age and stratigraphic — place.Of (0....2.G 15 See — —, Distribution of ............ 302, 303 — —, Exposure of, Views showing. 303, 805 — —, Thickness Of sis 305 — mica- -schist, Age and_ stratigraphic place of... 3806, 330, 335, 338, 339, 340 ——, Character and stratigraphic rela-. tions of . 801-302, 303-305, 351-352 —-—, Distribution of ............... 301 —-——-—,, Map showing ........... 331 ——, Exposure of, View showing .... 302 ———, Thickness ‘of ~.o) Slee eee 302 Wo.Lrr, J. E., cited on geology of Pied- mont district of New England.. 292 WoopwortH, J. B., Manhasset gravels of, Reference to eee ions ic —, Sankaty beds named by........ 386- 387 WORTHINGTON valley, Maryland, Cock- eysville marble in ........... 359-360 WRIGHT, F. E., enrolled as life member 535 WRIGHT, G. F., cited on glacial ero- STON \orecchcne? Dheuel ss choco eee ee 21, 34 WyYNCOOP creek, New York, lowered di- — vide at head Of (oo... Ae 36 — — — , Map showing 236 WYOMING, Red beds in....../.:... 488-490 — formation, Colorado, Divisions of 491-492 — —, Occurrence Of 23h eee 52D -YaMpa plateau, Utah, Mesozoic forma- TIONS OF « oes)... bet oe 487-488 YOSEMITE valley, Glacial erosion in.. 35-38 YPSILANTI, Michigan, Stream-formed hanging valley near .vc2.4s- cen CF, eee plateau, New Mexico, formations AT ea Pio ee 477-480 aes: lake [abstract]; N.-H. Darton. 564 — sandstones of New Mexico, has OP Rie ae ates eh ee wees. 478-479 4 e ARS §, 5 Mo SOC i ete eh wa tea SNe Rs ian | 3 9088 01309 1855