ate ene ed ogame ihe Date, baued He 9% ») et peye hoard eh at! are Vath itso oot» (erent stedeneday rte Aghoyet eee unhe Oper erie pe ty ee ee {tem Ds terete Code tel Matty 4 2 Hn he 4 ‘ie tii eae] ahnieey aahaadi isacigatinn aifzs=1\ witty ost ataete eheectae a aaetene wtetite ene ratias Ve hehe Aye ett FFP pa host Del ange ett: ORC Pearse ah Bort ha i eae . ares genta wig * “gical aes ria att voit a : hu er ysp te pF ata aegis clit anes tobemla rs ae - te Ea yee Fethettg ed & A Reacts Pease iy Ldheck ated lytcay hott dea oteh Hatrtaht anita rgb: etter betad ste he tele 4 leita preity Nadeg : rekent Cgepte te tenga. Dpthet 134° oil hoes pared bigd wee yeaag ‘ ee bat or hee ree tise get) ast oe et} heat ey apthwe 18h od M Arrant Taw! a fs 4 ney ‘ ented . " satu ” 7 rs ve weg NS4RY ST ME Lerner ar) PPererevirestrt ot: buat thd Sued Ned 4% ep ities yee heb tad ieee he te heasaa’ cit we reiegs i aged! ite reach inde de dhs hase ty eiinae ester Pout i By 4 Leh yt ps rear ph vi 4 at Hes Pidgte steers te Preure ia he ete Sele ast ates IMs, jt deaeie| ri aan ¢ hens ah ine (gs qe Cit5 ptt * vs “7 ie at re Suey Sid raga ge as cave fi tte hie! He Bes (arse a aie yom gait a fea ae ae Stel oe ae tees ett: ra) Hi fa he sbtes oe sat tient sot abe ae a ae aae tik autre * arte esr iets BeBe ae ‘a ; sarees * > areas iterod ABiwty me rabhes) Hs bad m ets oe fi oA ” steals bt eterety sree Sh EA AU hdd A oe ie : te # Pty sew faestaninamrdy Aa rae ‘ eeu cur a. AR Aleeh Libatand Hein 721 aie ohorteery Tatu haute aw ve pa ier iran) ehens od oe 4 4 ebeil eqs, omen a Sete ip TAMe Terie : VB yt Ree ‘ whiaie Jieigtiter sass. 1M eee tt shou 19 98 pet tale HOU eS opieipty eh Aj Hu soe ca payee eihect vd FON UK uSt a ith ad i Hed $ Jiwbyta pve: S335 sao teaieey " +H 8 Ap we ere Vee , dep atte ed eGets “a yrotut eee ais Fem meyea Tey 6 Fgh vn ee ee arian ADiee di reer ‘ i ai ie doy age ig? toy WrarKeu ed eryitoad eye iay ore a ome ae soho tt anwene Pee eee ce easy iT A eee ee setae pei at ouse Leu >! AA feet ig We haa “ a ‘ oy te Pah el year Nabe ddbcs eis usd 9i9 184 watts aie yyaqonew Tey : Spb toy “ Patt sUem watery Te has ater et ote a teae rae ite . aeueeet 0 ae wanes wis he wee Ms steers etary etary eae beet ety Palen Hd ve ae a 49% 4 ewe ties vise argrnes Guat ae i pie “eet ae I aac ht Rint ae : ae a 429 hth oe Be iFuewe, sides) patty aaa uae tise seedee Sr is! ahat Suites Paes isan, soaveanyets abate 8 eae aaa erie, rere eae pices hp sata side yates sony . ey eened a? Seite bop arheH oes seinen roeqsaaaet aaa A aonpeee prlnate \ os Pihesks Bid es aeteet Sarat aed sel “4 L ih nhhahtatnas is t we eée oe i bret bt be cate aoe ee oi LD ehh Mediate ates 1] oieda etal wore evga sweets Leis ” Peep sissy rm pd ap if sdhine urd eerd Haugan urate? it coy phates had ate ar Lt H pr 4 pan iay! wanda , 4: ie Benet mabe ererhe ard age Me) ’ rae * abe: ius “ 9" ° yreeny ashy saieyetewegine whl gin? ited ny ans Ul sans 4}. Bi hc sn ra pity cia Pca : sid suey /aaeteat in li BULELEGIN. OF THE EOChe i: SOCLE T ¥ OF Pe VERO 2 Ansenian Ins LAN Asti, ~ ; FEM tit Lit ~~ i Geol o * W,,. VOL 6 Monat Muse JOSEPH STANLEY-BROWN, Editor ROCHESTER PUBLISHED BY THE SOCIETY 1895 ¥ COUNCIL FOR 1895 N. S. SHaer, President JOSEPH LE ite Vice-Presidents CPE imcrncocr, H. L. Farrcump, Secretary ' I. C. Waiter, Treasurer Class of 1897 ) R. W. ELLs, C. R. Van Hisx, Class of 1896 .f. D. ADAMs, Members-at-large Ce hiss, | Class of 1895 | K. A. Smita, J C. Dy Wiacors, ~ PRINTERS Jupp & DrtTwWEILER, WASHINGTON, -D. C. ENGRAVERS Tut Maurice Joyce ENGRAVING Co., WASHINGTON, D. C. b] b CONTENTS Page Proceedings of the Sixth Summer Meeting, held at Brooklyn, New York, August 14 and 15, 1894; Herman Lr Roy Fartrcuixp, Secretary.......- .... it SesstOn, OL, MuUESdayaPAUCUSE VARS iis dosh oa a wee seen ey mie ees ee wie ee 1 Hse etme LCE Fe IMO Siniey yt ohe cel tuatraynichis Usa vn os S95. sieve Sc gay oes Mhcual saben oa etaw ste 2 The nickel mine at Lancaster Gap, iBemnseaceaiey, and the pyrrhotite deposit at Anthonys Nose, on the Hudson [abstract]; by J. F. Kmmpe 3 A connection between the chemical and optical properties of am- phiboles: abstract); by, AmmREpsO. WANE: 5) 02.2416. ye sy.e «eles Dale es 3 On a basic rock derived from granite [abstract]; by C. H. Smiru, Jr. 4 Dislocations in certain portions of the Atlantic coastal plain strata and their probable causes [abstract ; discussion by N. S. Sralien | ; Dy AUR ENUF O Ls TTOK - uiian (ae ystal neh gee eee ema cic a Sle ak oo oar oleate athe 5 Glacial origin of channels on drumlins; by Grorcre H. Barton.... 8 SeccionuoVecdmesdaya AUCUSE Noe hii le sce telah. ahs ato slousvehers arehole ets 13 Report of the Mount Rainier Pacific Forest Reserve Committee.... 138 Eroposedeamendments to theconstitutiom) ss2.05.-..6--.205e88-- 2 LO Rropesedaamendment to the by-laws. «0... -aeo0-*+«.o00- eS. 15 Tertiary and early Quaternary baseleveling in Minnesota, Manitoba and northwestward [abstract]; by Warren UpHAM............. 7/ Departure of the ice-sheet from the Laurentian lakes [abstract]; by WN ACUI ENR IPUETANVse ees St 00s 19h oes) os mira SPARS mateo a Shade aaa aeipare tabs ea 21 ‘Note on Florentino Ameghino’s latest paper on Patagonian paleon- OO OWE EOWA INN Elle, SCOMD Ys acs Siete'eod «6s a's Sele ean we atehdeie eek eaares 28 ecister or une; DEookdym Summer Meeting... ....os,002-s2+ 02.0 28 Kansas River section of the Permo-Carboniferous and Permian rocks of Kansas ; SRM SMES SER Fone cot hi cet ant, wie lrg aie wie ble a are Beles aeoouaatuis 29 The extension of uniformitarianism to deformation; by W J McGerr......... 5d Review of our knowledge of the geology of the California Coast ranges; by oP PRBLD AUD ONIKCSt me ratte cited pclae orate htt 2 nado Sie a Salie o giateee' eo Sawn ete 71 Reconstruction of the Antillean continent; by J. W. SPENCER................ 103 myidences as to change of sealevel; by N. S. SHALER............0.2.0-...... 141 The Magnesian series of the northwestern states; by C. W. Hau and F. W. “SL EDISON owen tas MOS eB ce een a ec eee epee ae 167 Recent glacial studies in Greenland; Annual address by the President, T. C. \, TUAA BILTON 6. 8.5 a8 Sk Se eh ee UR ae 199 Characteristic features of California gold-quartz veins; by W. Linparen...... 221 Crystalline limestones, ophicalcites and associated schists of the eastern Adi- BME AG LM NNN SKE UAGEONEP a0 0) Qe Hae nck aia \araes td Stare tole Mtme Grd, Seated pater tela stew 3 241 Crystalline limestones and associated rocks of the northwestern Adirondack ieee SUEDE MONS a LS MENVAIEN: HIER cp 2 0 f.e osei ex ¢,d.0 2 oni oa We yee DeIeAnS Sues artd Rin Wears ore: « 263 Faults of Chazy township, Clinton county, New York; by H. P. Cusnine.... 285 Honeycombed limestones in lake Huron; by Ropert BELL.................- 297 lV BULL. GEOL. SOC. AM., VOL. 6. Page The Pottsville series along New river, West Virginia; by Davin Wuite...... 305 Disintegration of the granitic rocks of the District of Columbia; by G. P. IMR TRI acc 2k os ey ee etd ak 2 ot. ee 321 Lepee buttes; by.G. K. Ginpmrr and F. P. GuiLiver.. (22 2 32 ee eee 339 Discrimination of glacial accumulation and invasion; by Warren UpHam.... 343 Glacial lakes of western New York; by H. L. Farrcuinp....... Se ae B09 Cretaceous of western Texas and Coahuila, Mexico; by E. T. DuMBLE....... 31D Highwood mountains of Montana; by W. H. Weep and L. V. Pirsson....... 389 Proceedings of the Seventh Annual Meeting, held at Baltimore, December 27, 28, and 29, 1894; H. L. Farrcniup, Secrefary.............1... 00 423 Session of Thursday, December 27.......... 24.025 «000 «0 00 er 424 Report of the Council... ......2...5...00+02 5.1000 oe 424 Secretary's report... 0. 05 ss cesses ees ose yee ar Lee 425 Treasurers reports... 65 S08 kes oe ne a ep el ee 429 Editor's: report... . . 2... cee es eee: wet ne © on tine 429 Election of officers..... wn ds Pace bbe S'e4 cle 6 ae oe rr 431 Election of Fellows... ....5...¢006 se gh. fo « omen oe ee 431 Ameéndments to the constitution. .2.....-..2.. 5... «seen 431 Amendments to the by-laws: ...-... 2052.4 .55 1 ese ee 432 Memorial of George H. Williams [with bibliography]; by W. B. CLARKS... 6... cabibb sce be nies oe Gee nee eee 432 Memorial of Amos Bowman [with bibliography]; by H. M. Amr.. 441 Session of Friday, December 28: 2... 2.6.0.6. 0) eae ee 445 Report of auditing committee. 2°. ..: 262 oa, ee 445 Fifth annual report of committee on photographs.................. 445 Report of committee on Royal Society catalogue................+.- 457 High-level gravels in New England [abstract; discussion by J. W. Spencer]; by C. HAH mencock 0.0. 0.0. see eer 460 Variations of glaciers [abstract]; by H. F. Rem: - 2322 461 Lake Newberry the probable successor of lake Warren [abstract ; discussion by G. K. Gilbert, W J McGee, Warren Upham and J. W. Spencer]; by H. L: FArmcsILD .... 7... ...eeeee 462 Notes on the glaciation of Newfoundland [abstract]; by T. C. CHAMBERLIN... 5. .05..s50 cate ae soe soe Sede ole 467 Organization of the temporary petrographic section... ........... 469 Crystallized slags from copper smelting [abstract]; by A. C. Lanr. 469 The granites of Pikes peak, Colorado; by E. B. MAarHews......... 471 Illustrations of peculiar mineral transformations [abstract]; by B. KK. JEMERSON... 0 ene sae 4 2 bd sige «eee 473 Spherulitic volcanics at North Haven, Maine; by W. 8S. Bayiny... 474 A new intrusive rock near Syracuse [abstract]; by N. H. Darron and J. FE. Kamp. .. 03... coe ke wed ee ees ene 477 Session of Friday evening, December 28.............. J... se eee 478 Session of Saturday, December 29....... 0... s..4.. 6. . <5 479 Cretaceous deposits of the northern halfof the Atlantic coastal plain ; by. W SB OAR 6. eee stn Gowen 6 Nace aie Meena dake ae 479 Surface formations of southern New Jersey; by R. D. Sarispury.. 483 Register of the Baltimore meeting, 1894... ............s.55) se ee 490 ILLUSTRATIONS. Vv ; Page Officers and Fellows of the Geological Society of America........ ..:...-..-. 491— imtcecsions tolibrary to January, 189)..........-...5,..6-.+% Sarat As aden: 501 Mndiex tor yolume 1G... .....0s0... 1 PRE ee OL Pee ema outa oe TOR ee aie Ree on TE Se 517 ILLUSTRATIONS. Plate 1—Spencer: Drowned valleys of the Antillean lands................. 103 «¢ —6- 2—Hatt and Sarpeson: Map showing distribution of the Magnesian series in the northwestern states .........- 167. > _-OMAMBERDIN: Bryant glacier (2) Heures)... 2... sie eee een es 202 cee A x Bryant and Gable glaciers (2 figures).......-........ 204 a dey, “ Bowdornm elacien (Zaioures) Qos...) ee ee ns 205 2 Sathaaal hi Mnkcoo lacie (2 OWMEOS) <2 ..5 25%. ciste Se wis. o' syeeinue sone #8 206 meric es Gainleralacier (Qnnoutes): tse. 5 eee oleae od hig acta 207 nathan Co * ELMO CHET nO INES ik tate teicy a ele Lk ele ena cos gira ees es 208 cae) is East glacier and edge of ice-cap (2 figures)..........- 214 NG) ey Dalrymple and Carey islands (2 figures).... ........ 219 ei LinbGRENn = The central gold belt of California..................0... 221 ‘« 12—Cusuine: Geologic map of Chazy township, Clinton county, New BVA ete was ott | a cee ene oe RRR ORC AA Want ie Ay ea tesa 4 .. 285 Pale -bkit,; oneycombed dolomite (2 figures)... .0..25..-- 2. oe 0. eee. 298 ‘e 14 ‘¢ Pitted limestone of the Black River formation (2 figures).... 299 pS ‘¢ Glaciated and honeycombed dolomite of the Guelph forma- OO) Ag oA gene tt UAE ¢) unm Sear 17 a ga at mee 000 “* 16—-Merriuui: Disintegrating granitic rocks in the District of Columbia... 322 2 /—Gineert and GULLIVER: Tepee buttes (2 figures)......-....-.--. ooo ‘« 18—FarrcHiLp: Hydrography of western New York................+. 300 Gay taal 8) S5 Naplesnailesn(?, tlomphes)) a's, i.¢.0 Sty eel BORN A ores 362 He 20) = Wat kets lauken: to tmes eo dnt stcc sae eens eye: mechan oeclend 366 oe al eS Watkansslaken(Qneunes) i. cisieac ions is Sekwae eb olan oe 368 eee a Outlet channel of West Danby lake (2 figures).......... 370 Ay oe ne Hii aGa lake) (2, MeUnes) eee see ct iets ate asedc weiss Caneibedehe ats o12 a — WEED and. Presson: Square butte (2 figures). ......0.2.+0s.eeeer. es: 402 oS) oe . Squareduiter(2 Meures) ios ac. eee. 2 wae se 403 i, 20 es of Square outie(2momres) ss ..fas si es Stine 404 7 Oran : Portrait of George H. Williams..s.... ..¢o.02sse:sha.0--. 432 SPENCER : Figure 1—Section across Lookout-Wills valley, Alabama, at the col..... 106 ‘¢ 2—Section from Pigeon to Little Sand mountain, Georgia....... 106 ‘* 8—Cross-section representing a baselevel valley in Trinidad AUC CML EVIN Ge OUD A. 2 a eren tiieeernts Wee Atco eh ae eure Seanad duet ue 107 we -Secromnor tie Iroquois! beach no. 50) skis sd ce saw meine ods 108 ‘¢ 5—Section from South Carolina to the Mississippi, showing de- OrMmMahion, Ofte, Latayerte Tormation. 2.00.4 .eo0see.) soo: 108 ‘* 6 —Map of Honduras and Rosalind banks, showing fiords....... 116 ‘< 7—Oscillations of the Antillean continent................. Beye Wiel Kemp: . Bacires!— Mapr.ot, Essex county, New York. .¢ 0222.22 hse)... ee65. 2 3k. 247 ‘¢ 2—Group of inclusions in crystalline limestone... ............. 248 v1 BULL: GHOL.WSOC! AML.,; VOL.) 6. KEMP: Page Figure 3—Cross-section at ophicalcite quarry near Port Henry, New Yorke ccs ect eee pele peed 22S ee rr 250 ‘« 4—Cross-section one mile north of cross-section represented i in FOUTS! Bee ace disiw od epale elie. b aves 6 a 6 het ee 250 ‘* . 6—Cross-section at.Cheever mine... .....% 4... eee eee 251 ‘* — 6—Cross-section in Moriah township, east of Spragues Corners.. 251 ‘¢ — 7—Cross-section near Weston mine, Keene..................:- yi SMYTH: Figure 1—Granite cutting laminated eneiss.............. 22a 267 ‘¢ _ 2—Granite veins parallel to lamination of gneiss............... 268 ‘¢ 3—Basic gahbro cutting limestone and gneiss................ . 268 ‘¢ 4—Gabbro-limestone contact. Lake Bonaparte ..... ......... 275 ** -5—Inclusions of eneiss in gabbro.:........- + 3. eee 276 | ‘* 6—Banding of gabbro parallel to side of gneiss inclusion........ 276 Wiitr: Figure 1~-Piney Creek seetion. 02:5 jc. ou. dae cele Cae rr 308 +) -2—Nuttall and Hawks Nest'section’...: 7. 22 33s. eue GILBERT and GULLIVER: Ficure 1—Group of tepee buttes.. 222... sees toe = a cee) aoe ‘¢ _ 2—Map of the tepee zone northeast of Pueblo, Colorado ....... B00 | “~-3=Ideal section through tepee core. =... ...5..26- eee eee 306 ‘* 4—Chif on the Attawapishkat river....... ip. .3 .,. bcc. c2e% oma ce oe oe ce 341 ‘¢ {—Ideal section of butte represented in figure 6............... 341 FAIRCHILD : Figure 1—Outlet of the Watkins lake ...:....5.....>. 25 066 ‘<. 2—Delta terraces in Cayuga Inlet valley ...:...-. 23.22 372 ‘¢ 8—Approximate height of terraces in Cayuga Inlet valley..... 372 Weep and Prirsson: Figure 1—Geological map of the Highwood mountains, Montana...... 393 ¢ 2—Cross-section through Highwood mountains................ 396 ‘¢ 8—View of Square butte from the slopes of Palisade butte...... 401 ‘* 4—Geological map of Square butte... .:... . 2.5. eee 402 ‘f 5—The white: band... fickle 300 ae 406 “ 6—Cross-section of Square butte. ......... 0.208) eee 407 “ (—TLwinned pyroxene:crystal.....<. 0.22... 5 23) - ee .. 410 ‘¢ — 8—Micro-drawing of shonkinite multiplied 14 dinmereea oe (27 plates; 40 figures.) PUBLICATIONS OF THE GEOLOGICAL SOCIETY OF AMERICA REGULAR PUBLICATIONS The Society issues a single serial publication entitled BuLLETIN oF THE GEOLOG- IcAL Society or AmeErRica. This serial 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 the larger papers accepted for publication The matter is issued as soon as possible after acceptance, in covered brochures, which are at once distributed to Fellows and exchanges. The brochures are arranged for binding in annual volumes, which are elaborately indexed. The Butietw is sold to Fellows and the public either in full volumes or in sepa- rate brochures. The volume prices are, to Fellows, a variable amount, depending on the cost of publication ; and to libraries and to the public, the fixed amounts given below. The brochure prices for volumes 1 and 2 are given on pages 1x—xi of volume 2; the prices for the brochures of volume 3 are given on pages vili-ix of that volume. Volume 1, covering the work of the Society from the organization, in 1888, to the end of 1889, comprises 593 + xii pages, 13 plates and 51 cuts. Price to Fellows, $4.50; to libraries, $5.00; to the public, $10.00. Volume 2, covering the work of the Society for 1890, comprises 662 + xiv pages, 23 plates and 63 cuts. Price to Fellows, $4.50; to libraries, $5.00; to the public, $10.00. Volume 3, covering the work of the Society for 1891, comprises 541 + xi pages, 17 places and 72 figures. Price to Fellows, $4.00; to libraries, $5.00; to the public, $10.00. ~ Volume 4, covering the work of the Society for 1832, comprises 458 + xi pages, 10 plates and 55 figures. Price to Fellows, $3.50; to libraries, $5.00; to the public, $10.00. Volume 5, covering the work of the Society for 1893, comprises 665 + xii pages, 21 plates and 43 figures. Price to Fellows, $4.00; to libraries, $5.00; to the public, $10.00. Volume 6, covering the work of the Society for 1894, is now complete, and com- prises 528 + x pages, 27 plates and 40 figures. Price to Fellows, $4.00; to libra- ries, $5.00; to the public, $10.00. The volume is made up of 21 brochures, as follows: ; BRoOcHURE. Pacers. Puatres. Ficures. PRICE To Price To FrLiows. tHE Pusuic.- Proceedings of the Sixth Summer Meet- ing, held at Brooklyn, New York, August 14 and 15, 1894. H. L. Farr- BERUNI SCOLCUATY 22 ats a Rc clsapers aoe Sb 8% 128 Wee eee) Aa OPO) 50.60 Kansas River section of the Permo Car- boniferous and Permian rocks of Kan- Ramee oe ROSSER (csc feels cle eso Sate yA) Page Ae By) 70 The extension of uniformitarianism to deformation. W J McGems........... DLO) = cern Sates MNS .30 (vii) vill BULL: GEOL. SOC. AM., VOL.. 6. Price to Prick To ‘HURDE. GES. L, s. Fia Brocuure PAGE PLATE URES: “ninsows. | ane Review of our knowledge of the geology of the California Coast ranges. H. W. LA TRIBANMES? 5.45. bc. Gee ee ieee PaSTOD. ost ese .30 .60 Reconstruction of the Antillean conti- Hens dl. Ws SPENCMR Ss oo... dae 103-140 0-1 1-7 . .45 .90 Evidences as to change of sealevel. N. SS SORLAMMIR's dee bo wn oF lismhickness and oeneral characteristics..<: 2...) t2)eet pads. es wlewan 37 Poe Ole Meek Andi ENAM CMe cantatas. teed suaiAuan's clticla: a chaare s Mua 40's. ele 37 PSHM NITIES TOME? SOly WOW ANLON, sa alate nett x ce sw cies sale W aie asians a6 6 Sve 6,4 38 Fauna of the shales above the Manhattan stone..................... 38 Haumearor the Cottonwood slialey...2252..0..5se.s.h0: «- Aa eee, BABS Bo) Correlation of the Manhattan with the Cottonwood stone........... 40 ipromumcnViambatbamesStOMe. ai a o.kc sigh dd oe eect He eae ee. keane 4] noe Mnlit@reek-ceolocie section. si. J aeiee one cee Le bc eb AEP ean ark 4] PHM ORVLOMUMORSeCOMUAINGs LUNA Gest eaters Gs 4 veld! Gee dis eos Scales 41 McFarland section and comparison of its fauna with that of Buffalo Mound. 42 AUIS) GSGTICOIS Ae Mes oop ev eaEe yea Tene a7 te eee eg ae aN or ae ee +4 eS te ine Keres Sy ame het Nee chee te bkets ata Badin we Mah ode Yolen 8 mastin hacia vieya Shem +t Worrelation of the Alma with the Manhattan stone.................. 44 Kinsey dime pedeabove the Alma StOMe 2s. 6. Sane oes eae eo 45 The geologic section of the upper Kansas river....... Dice ARAL RRR MS tle RAS 45 eMermcinanieeerishies camienniae pron iat Pat ek Lal wl Take ee 45 * Published by permission of the Director of the United States Geological Survey. V —Butt. Geox. Soc. Am., Vor. 6, 1894. (29) 30 Cc. S. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. Page Lamellibranch fauna above the Manhattan stone......... ..........-0-- 46 First flint bed above the Manhattan stone.:......:.../:.02: 32... e sees 47 Its character and position .......5. 20... Gc. ..22..¢290104 6 oe Bed 18 of Meek and Hayden... 0 .0.56....0822 ce er 47 ‘ Fifth cherty limestone” of Swallow....... Wee ES "5 hw She oleae 47 . ““Wreford dimestone ’? of ay 2... 00. ..05 hae eee «2 eee 47 Fort Riley section: 25). 6} oes os os ee de eee ne ee 48 Professor Hay’s investigations........ ee ah ; | 2) 48 Second flint bed and its correlatives............. tape a ee 48 Fort Riley limestone and its correlatives......,.......7.-.s0e eee 48 Review of the geologic correlation of the Kansas river section................ 49 Chart: of SCCbiOIS' ojo... 5% Lie elo Be lh 4 oa aia vce ee eyeeeepe severe ae eae 51 Comelusion: 2.5.05 ce eee ee Seceye ts ale ees ee 2G ee ee 54 INTRODUCTION. Along the upper course of the Kansas or Kaw river, in the northern part of eastern Kansas, are good exposures of the rocks belonging to the Permo-Carboniferous and Permian systems. The early students of the geology of the state—Meek, Hayden, Swallow, Hawn, and later St. John— clearly recognized the importance of this section, although there was a decided difference of opinion among them regarding the structure and correlation of the rocks. The strongest feeling that has ever existed in reference to any question concerning the geology of Kansas was developed by this controversy, and now that most of the participants in the discussion have passed away, it is interesting to review the arguments and compare them with our present knowledge of these formations. Later writers apparently have not considered to any extent the descriptions of this section, and on this account it also seems advisable to call attention to this early work. REVIEW OF PREviIous WoRK. MEEK AND HAYDEN’S EXPLORATION OF 1858. In the summer of 1858 Mr F. B. Meek and Dr F. V. Hayden studied the Paleozoic rocks of northeastern Kansas, and in January, 1859, pub- lished an interesting account of their observations.* Their route was from Leavenworth, first to the southwest, reaching the Kansas river valley near the mouth of Soldier creek and North Topeka; then up the north side of the Kansas and Smoky Hill rivers to the mouth * Proce. Acad. Nat. Sci., Philadelphia, vol. xi, pp. 8-30. EXPLORATIONS OF MEEK AND HAYDEN. on of Solomon river; then they crossed to the south side of the Smoky Hill river, followed it to a point near the western boundary of McPherson county, thence east to the head of the Cottonwood valley, which was fol- lowed nearly as far as Cottonwood Falls; then across the divide through Council Grove and Lost Springs to the Smoky Hill river near the mouth of Solomon river, and finally the course was down the south side of » the Smoky Hill and Kansas rivers to Lawrence, where they crossed the Kansas river and returned to Leavenworth. Between Leavenworth and Manhattan a number of local sections are described and the thickness of the different layers indicated, with lists of the characteristic fossils, but no continuous general section is given until the mouth of the Big Blue river is reached at Manhattan. From this point the authors state: ““As our examinations along the Kansas and Smoky Hill rivers . . . were made in more detail, where the outcrops were more frequent and continuous we have, as we believe, been able to trace out the connections and order of succession of the various strata with considerable accuracy.’’ * Then follows what is called a ‘‘ General section of the rocks of Kansas valley from the Cretaceous down, so as to include portions of the upper Coal Measures,” which is composed of forty beds.- Number one is the Dakota sandstone on the summit of the Smoky hills, and the order is de- scending until number 40, composed of Carboniferous shales, is reached, opposite the mouth of the Big Blue river. A brief description of the geologic characters is given; also lists of common fossils and thickness and location of the different beds. SWALLOW’S REPORT. In 1866 Professor G. C. Swallow, state geologist of Kansas, published a “ Preliminary Report of the Geological Survey of Kansas, which con- tains a section of the rocks in eastern Kansas.” { This section begins with the Quaternary, which is called system I. The base is the lower Carboniferous or formation C' of the Carboniferous, which is system VI. The Permian rocks constitute system V, which is divided into the upper and lower Permian,§ and includes numbers 12 to 84 of the general sec- tion. The statement is made that the base of the lower Permian is Nibids ps 1p: 7 Ibid., pp. 16-18. This section is quoted by Dr Newberry in Report upon the Colorado river of the West, expedition in 185758 of Lieutenant Joseph C. Ives, pt. iii, Geol. Rep. by J. S. Newberry, 1861, pp. 112-114. ft Pp. 9-29. This section was also published in Proe. Am. Asso. Adv. Science, vol. 15, 1867, pp. 57-75. 2 This division is made in the part termed the “ Geology of Kansas,” pp. 42, 43, although in the section the “ Upper” is called “the Permian strata,” pp. 11, 12, numbers 12 to 30, and then follows the ‘‘ Lower Permian,” pp. 12-16, numbers 381 to 84. 32 C.S. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROUKS. shown at Manhattan and on Mill creek, some twenty miles southeast of Manhattan. Professor Swallow insisted that the Mill Creek section showed an unconformity, and that numbers 84 to 95 of the sections near Manhattan were not found on Mill creek.* Henry Engelmann, who was geologist of the exploring expedition of Captain J. H. Simpson across the Great plains and Great basin in 1859, gave some information in reference to the geology of this region. The details, however, refer mainly to the country some forty-five miles farther north, which was crossed by the expedition.t Tor MANHATTAN GroLocic SECTION. BROADHEAD’S SECTION. Before attempting to determine the position of the beds described by Meek and Hayden, and Swallow, it will be well to give a general descrip- tion of the geologic section at Manhattan. Some ten years ago Professor G. C. Broadhead published a section of the rocks at Manhattan, which, in a condensed form, is as follows: Feet. Feet. 1, (Drab limestone? 2 ccaddy Boa Bg, SAPO els he ae ee ee 43 = 208 2, HAlY SlOPO./.ac. 62 s..,ee ate eee Seed ees ett oe ee he ee re 30. 2035 3: Drab, compact, tine orained lumestome. 4.2.1... 4. oe eee eee x == Wise 4. Chiefly shales to base of hill; a bed of red shale half way down.. 170 = 170t PROSSER S SECTION. Near Manhattan are steep bluffs rising abruptly from the river to an elevation of more than 200 feet above the river level. Two of them stand out prominently—one, called Blue mount, just north of the city, and the other, to which Professor Broadhead’s section refers, called mount Prospect, south of the Kansas river. An accurate section of the rocks of the region may be constructed from the outcrops on these two hills, although all of the layers are not well exposed on either mount. Blue mount rises very sharply from the Big Blue river, its summit being com- posed of the massive limestone, quarried so extensively about the city, which is called the Manhattan stone. On top of Blue mount is the city reservoir, the coping of which is 215 feet above low water in the Big Blue river and 10 feet higher than the top of the Manhattan stone. The follow- ing section was made on the east face of Blue mount: * Op. cit., p. 44. ; + Report Exploration across the Great Basin of the Territory of Utah in 1859 by Captain J. H. Simpson. Appendix I, Report on Geology of country between Fort Leavenworth, Kansas Terri- tory, and the Sierra Nevada. Section I, Northeast Kansas and Southeast Nebraska, pp. 251-259. t Trans. St. Louis Acad. Science, vol. iv, pt. iii, p. 491; read Nov. 6, 1882, and published in 1883 or 1884. bo . Yellowish, bluish and blackish shales, with thin layers of argillaceous 64 PROSSER’S SECTION. gk Reservoir level : Feet. Feet. . Covered slope on Blue mount. Atthis horizon yellow shales contain- 10 = 216 ing plenty of fossils are exposed on mount Prospect, in the Uhlrich quarries up Wild Cat creek and at numerous other places about Manhattan. . Manhattan stone—a light y ailowich gray, massive limestone contain- 5 — 205 ing a considerable amount of chert, and in the upper part great numbers of Fusulina cylindrica, Fischer. In the quarry at the top of mount Prospect it is 5 feet 4 inches in thickness. . Covered slope. On mount Prospect are shales, with some beds of 40 = 200 laminated limestones about a foot in thickness. . At the top a drab to bluish limestone of irregular texture which 64 = 160 weathers yery unevenly. This layer is between 23 and 3 feet in thickness on mount Prospect. On Blue mount it forms the first marked ridge, and the slope beiow the outcrop of this ledge is cov- ered to the top of the road cut. | 96 limestone (6 inches to 1 foot in thickness). The limestone in the cut near railroad level is somewhat bluish and contains fossils. The blackish shales near the top of the railroad cut at its southern end contain numerous fossils. meovercd slope tolevel of Bie Blue river: .... 24... ¢ ences ene ge aes cee aa == (32 COMPARISON OF FAUNAS. Fauna at Foot of Blue Mount.—The railroad cut, especially the blackish and yellowish shales in its upper part, afforded the following species: Productus cora, d’Orbigny.* (a) This species is quite common in the yellowish shales. Productus longispinus, Sowerby. (c¢) Productus nebrascensis, Owen. (1) Productus semireticulatus (Martin) de Koninek. (rr) Spirifer cameratus, Morton. (c) Spirifer (Martinia) planoconvexus, Shumard. (aa) This is a common species of the yellowish shales. Rhynchonella uta (Marcou), Meek. (ce) Hustedia mormon (Marcou), Hall and Clarke.f (a) Athyris subtilita (Hall), Newb. (ce) Chonetes granulifera, Owen. (c) . Chonetes glabra, Geinitz. (rr) Discina manhattanensis, M. and H. (rr) This species is not figured, but the specimen agrees with Meek and Hay- den’s description, and the original specimens came from the vicinity of this horizon at Manhattan. *The relative abundance of the sp2cies is indicated in the following manner: a= abundant ; aa = very abundant; ¢ = common; r = rare; rr = very rare, but one or two specimens found. 7 The generic name Hustedia has recently been proposed by Hall and Clarke for the shell called Retzia mormonii (Marcou), Meek and Hayden (Pal. N. Y., vol. viii, pt. 2, fascicle i, July, 1893, p. 120). 34 ©. $8. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. 13. Lingula mytiloides, Sowerby, or L. umbonata, Cox. (7) These species are difficult to distinguish. 14. Derbya crassa (M. and H.), H. and C. (rr) 15. Meekella striato-costata (Cox), White and St. John. (rr) 16. Syntrilasma hemiplicata (Hall), M. and W. (rr) 17, 0Cranmia.ap. (rr) 18. Allorisma subcuneata, M.and H. (rr) 19. Aviculopecten occidentalis (Shum.), M. and W. (?). (rr) 20. Nuculana bellistriata, Stevens, var. attenuata, Meek (?). (rr) 21. Dawsonella meeki, Bradley (?). (rr) 22. Lophophyllum proliferum (McChesney), Meek. (rr) 3. Synocladia biserialis, Swallow. (rr) 4. Fistulipora nodulifera, Meek. (c) 25. Rhombopora lepidodendroides, Meek. (rr) 26. Fusulina cylindrica, Fischer. (aa) 27. Fusulina cylindrica, Fischer, var. ventricosa, M..and H. (c) 28. Archxocidaris sp., spines and plate. (7) 29. Archxocidaris sp., very large spine. (rr) 30. Chetetes sp. (rr) 31. Phillipsia major, Shumard (?). (rr) The glabella of the specimen. Although I have seen neither the figure nor the description of the glabella of P. major, it seems probable from the size and some other characteristics that it is this species. 32. Crinoid, fragments of stem. (rr) Fauna of Bed 37 of Meek and Hayden’s Section—Meek and Hayden re- ported at 562 feet above high-water mark of the Kansas river, opposite the mouth of the [ Big] Blue river— ‘‘Alternations of dark gray and blue soft decomposing argillaceous limestone, with dark laminated clays or soft shale containing great quantities of Fusulina cylindrica, F. cylindrica, var. ventricosa, Discina manhattanensis, Chetetes, and fragments of Crinoids; also Chonetes verneuiliana, C. mucronata, Productus splendens (?), Retzia mormonti, Rhynchonella uta, Spirigera subtilita, Spirifer cameratus, S. planoconvexa, Euomphalus near E. rugosus, and Synocladia biserialis ; also Cladodus occidentalis,’’* This was number 37 of their section. Since the above was written changes in synonymy have referred Cho- netes mucronata, M. and H., to C. granulifera, Owen. Productus splendens, Norwood and Pratten, has been referred by Professor Meek and Dr White to P. longispinus, Sowerby. Dr Waagen has taken it for the type of the eroup which he called Marginifera, and in the latest work of Professors Hall and Clarke it is called P. (Marginifera) splendens, N. and P., and Spirigera subtilita (Hall), M. and. H., has been changed generically to Athyris. * Proc. Acad. Nat. Sci., Philadelphia, vol. xi, p. 18. + Eleventh Ann. Rep. State Geol. [New York], 1892, p. 228. FAUNAS OF BEDS 37 AND 34. 35 The list of fossils which I have given above is the result of two hours’ collecting in the railroad cut, and the number of species would undoubt- edly be increased by further search. Meek and Hayden reported Cho- netes verneuiliana, N. and P.; Euomphalus cf. rugosus, Hall, and Cladodus occidentalis, Leidy, which I did not see, while my list contains the fol- lowing additional species, viz: Productus cora, d’Orbigny ; P. nebrascensis, Owen; P. semireticulatus (Martin), de Konineck; Chonetes glabra, Geinitz; Lingula mytiloides, Sowerby (?); Derbya crassa (M. and H.), H. and C.; Meekella striato-costata (Cox), White and St. John; Syntri- lasma hemiplicata (Hall), M. and W.; Crania sp.; Allorisma subcuneata, M. and H.; Aviculopecten occidentalis (Shum.), M. and W. (?); Nuculana bellistriata, Stevens, var. attenuata, Meek (?); Dawsonella meeki, Bradley (?); Lophophyllum proliferum (McChesney), Meek; Fistulipora nodulifera, Meek; Rhombopora lepidodendroides, Meek; Archoxcidaris, two forms, and Phillipsia major, Shumard (?). The list of Meek and Hayden contains 16 species and my list 19 addi- tional species, making the total number 35. I think there can be no doubt but that the shales and argillaceous limestones in the upper part of the railroad cut near the foot of Blue mount represent Meek and Hayden’s bed called number 37. ‘There are also fossils in the dark gray to bluish limestone near the railroad level which probably represent their number 39. Iam not confident of this horizon in Swallow’s section, although I am inclined to think that it is the Pusulina shales—number 96—of his upper coal series, which were described as ‘‘ dark blue marly shale,” 12 feet thick, containing “ numerous Carboniferous Brachiopoda,” exposed at Manhattan, Cottonwood and Mill creek. MOUNT PROSPECT EXPOSURE OF MEEK AND HAYDEN'S BED 44. This bed is described as composed of “alternations of bluish, purple and ash-colored calcareous clays, passing at places into clay stones and containing, in a thin bed near the middle, Spirifer planoconvexa, Spirigera subtilita, Productus splendens (?), Rhynchonella uta,” etcetera,* the base of which, according to their section, is 1262 feet above the Kansas river. On the steep western slope of mount Prospect, about 120 feet (barometri- cally) above the river, is a bluish shale between two calcareous layers, which contains a great many specimens of two or threespecies. The list is as follows: 1. Spirifer (Martinia) planoconvexus, Shumard. (a) 2. Rhynchonella uta (Marcou), Meek. (c) 3. Athyris subtilita (Hall), Newb. (c) * Proc. Acad. Nat. Sci., Phila., vol. xi, p. 18. 36 ©. 8. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. 4. Hustedia mormoni (Marcou), H. and C. (rr) 5. Productus longispinus, Sowerby. (77) 6. Cladodus mortifer, Newberry and Worthen. (77) i. (Orthoceras “sp. (ar) This fossiliferous shale undoubtedly represents bed 34 of Meek and Hayden. Below this horizon and near the base of the quarry on mount Prospect is a yellowish calcareous shale, probably in bed 37 of Meek and Hayden, which contains Spirifer (Martinia) planoconvexus, Shum.; Pro- ductus cora, dOrbigny; Athyris subtilita (Hall), Newberry, and other species. BASE OF SWALLOW'S PERMIAN. Bed 27 of Meek and Hayden.—The lowest prominent terrace of the bluffs in the vicinity of Manhattan is composed of the drab, hard lime- stone which is mentioned as forming the top of number 3 of my section. It is well shown at the northern end of mount Prospect, as well as on Blue mount. This stratum is probably number 27 of Meek and Hayden’s section, which is described as a “ gray limestone, often fragmentary, with much clay above; lower part hard and more or less cellular in middle,” exposed near Ogden Ferry and Manhattan; and, taking into considera- tion the thickness which they give for the intervening beds, about 58 feet above the Rhynchonella uta (Marcou), Meek, zone.t “Dry Bone Limestone” of Swallow.—This stratum is an important one in Professor Swallow’s section, as it forms the base of the rocks which he called the lower Permian. The beds of this part of his section are well described and may be readily identified in the region about Manhattan. This drab limestone was called the “ Dry bone limestone, brown, con- cretionary and cancellated limestone, 5 feet, Synocladia biserialis, Spirifer planoconvexa,” etcetera, at Manhattan and Mill creek. Next above this stratum Swallow gave one foot of bluish brown marls, and then bed number 82: ‘*Cotton rock, 5 feet; a light cream-colored argillo-magnesian limestone; some- times in thin beds, with shale partings.’’ t This stratum is well exposed at numerous places near Manhattan, as at the northern end of mount Prospect, along the Manhattan-Ogden road west of Wild Cat creek, etcetera, and is quarried to some extent for * This specimen is somewhat similar to figure 5, plate 30, vol. 5, Geol. Sury. of Illinois, which is not identified specifically. + My barometric section of mount Prospect gave me about 45 feet; but, on account of the rapid changes in the barometer that day, Iam not confident of the approximate accuracy of this read- ing Consequently I am quite willing to admit the thickness of these beds to be greater than 45 feet, although I fancy 58 feet is an overestimate of their actual thickness as shown in the vertical section near the top of mount Prospect. t Prel. Rep. Geol. Survey Kansas, p. 16. “DRY BONE LIMESTONE” AND MANHATTAN STONE. on building stone. The layers vary from about 6 inches to 1 foot in thick- ness. This is probably number 3 of Professor Broadhead’s Manhattan section, which he described as a “rather uniformly fine-grained lime- stone” 33 feet in thickness,* and it is also Meek and Hayden’s bed 26, a light gray, argillaceous limestone showing on weathered surfaces a some- what laminated structure; contains large spines of Archxocidaris,t exposed near Ogden Ferry and Manhattan, and 9 feet in thickness. Professor Swallow stated that these three beds, numbers 82 to 84, “ are sometimes represented by a bluish gray and buff, porous magnesian limestone,” which is exposed on the Cottonwood. This stratum is well shown on the north side of the Cottonwood river east of Strong City and 52 miles west of the city or on the south side of the river toward Elmdale. Ina paper by Professor Erasmus Haworth and Mr M. Z. Kirk on the Cotton- wood River section this limestone stratum was called number 12 of their section.{ Above the thin limestones are shales and marls, with thin limestone layers, which Swallow described as “ blue, brown, purple and green” in color, 31 feet in thickness, while Meek and Hayden assign a thickness of about 36 feet to this bed. THE MANHATTAN STONE. Its Thickness and general Characteristics.—Capping the shales just re- ferred to is a massive limestone stratum, number 5 of my section, the base of which Professor Swallow gave as 37 feet above the “dry bone ”’ or irregular limestone, Meek and Hayden as 45 feet, and the writer about 40 feet on mount Prospect. This stratum is a massive yellowish to light gray limestone, 5 feet thick, containing a considerable amount of chert and in the upper part large numbers of Fusulina cylindrica, Fischer, and is known as the Manhattan stone, being the most important economic as well as stratigraphic horizon in the Manhattan section. The rock, which is very valuable for building and abutment stone, is quarried extensively in the vicinity of Manhattan and forms a well marked stratum, which, when taken in connection with the yellowish, fossiliferous shale on top, is the most distinctive and readily traced formation yet seen in the upper Paieozoic rocks of Kansas. In another unpublished paper I dwell upon this fact in connection with the Cottonwood River section, and here I wish to call attention to the same fact in reference to the Manhattan limestone and shale in order to show that these formations are one and the same. Bed 24 of Meck and Hayden.—The Manhattan limestone is bed 24 of Meek and Hayden's section, which was described as a “ hard, very light * Trans. St. Louis Acad. Science, vol. iy, p. 491. 7 Proc. Acad. Nat. Sci., Phila., vol. xi, p. 17. { Kansas Uniy. Quarterly, vol. ii, p. 113, and pl. iv, fig. 3, VI—Buwtt. Geo. Soc. Am., Von. 6, 1894. 38 ©. 8S. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. yellowish gray magnesian limestone, with Fusulina and spines of Archxo- cidaris,” 6 feet in thickness, and forming a marked horizon about 10 miles below Fort Riley.* The sum of the thickness of all the beds below the Manhattan stone down to the level of the Kansas river at Manhattan according to their section is 242 feet, which, from the position of the limestone on Blue mount and from the determination of its elevation by an exact survey, we know to be overestimated about 40 feet. It is true that Meek and Hayden did not report this stratum at Manhattan, but stated that it formed a marked horizon 10 miles below Fort Riley. The present highway from Manhattan to Fort Riley crosses such a ledge four miles southwest of the city, on the bluff east of Hureka lake, and 10 miles from Fort Riley. This is undoubtedly the horizon noted by Meek and Hayden, and there is no question but that it is the Manhattan stone, for the stratum may be readily traced from the hills about Manhattan to this locality. The Manhattan stone is probably bed number 1 of Broadhead’s section at Manhattan, which he gave as 42 feet thick.t “ Fusulina Limestone” of Swallow—This stratum forms bed number 80 of Swallow’s section, which he named the Fusulina limestone, a ** buff, porous and magnesian ” limestone 6 feet thick, exposed at Manhattan, Cottonwood Falls and Mill creek. It is important in reference to the locality to observe that Professor Swallow noted the occurrence of this limestone not only in the Kansas valley, but in the Cottonwood valley at Cottonwood Falls. ! | Fauna of the Shales above the Manhattan Stone.—Immediately above the Manhattan stone are yellowish shales containing abundant fossils. From several exposures of these shales about Manhattan, particularly on mount Prospect; in the Uhlrich Brothers’ quarry, 23 miles southwest of Man- hattan ; and farther west, by the side of the Manhattan and Ogden road, on the hill between Wild Cat creek and Eureka lake, the following species were obtained : . Chonetes granulifera, Owen. (aa) . Athyris subtilita (Hall), Newb. (ce) . Productus semireticulatus (Martin), de Koninck. (c) . Derbya crassa (M. and H.), H. and C. (c) Also some large forms like D. keokuk (Hall), H. and C., and D. robusta (Hall), H. and C. 5. Fusulina cylindrica, Fischer. (c) 6. Synocladia biserialis, Swallow. (c) 7. Rhombopora lepidodendroides, Meek. (c) 8. Archxocidaris, spines and plates. — (c) 9. Straparollus (Huomphalus) subrugosus, M. and W. (7) Hm OO b> Et * Proe Acad. Nat. Sci., Phila., vol. xi, p. 17. + Trans. St. Louis Acad. Sei., vol. iv, pt. iii, p 491. FAUNA OF THE COTTONWOOD SHALE. . 39 10. Aviculopecten maccoyi, M. and H. (rr) 11. Meekella striato-costata (Cox), White and St. John. (c) 12. Chextetes cf. carbonarius, Worthen. (c) 13. Fistulipora nodulifera, Meek. (r) 14. Crania sp. (rr) 15. Crinoid, segments of stem and plates. (c) Fawna of the Cottonwood Shale-—The shale referred to above is very similar in lithologic appearance to the yellowish shale overlying the Cot- tonwood limestone in the Cottonwood valley, which I have called the “Cottonwood shale.” The Cottonwood shale is abundantly fossiliferous, and the following species have been found in the vicinity of Cottonwood Falls and Strong City: . Chonetes granulifera, Owen. (aa) . Derbya crassa (M. and H.), Hall and Clarke. (aa) . Athyris subtilita (Hall), Newb. (a) . Productus semireticulatus (Martin), de Koninck. (a) . Meekella striato-costata (Cox), White and St. John. (c) . Productus nebrascensis, Owen. (7) . Aviculopecten maccoyi, M. and H. (7) . Straparollus (Euomphalus) subrugosus, Meek and Worthen. (rr) . Synocladia biserialis, Swallow. (77) *10. Rhombopora lepidodendroides, Meek. (rr) 11. Lophophyllum proliferum, McChesney. (rr) *12. Fusulina cylindrica, Fischer. (r) 13. Aviculopecten occidentalis (Shum.), M. and W. (rr) 14. Terebratula bovidens, Morton. (rr) *15. Chetetes sp. *16. Crinoid, segment of stem and plates. *17. Archeocidaris, plates and spines. 18. Phillipsia scitula, M. and W. (?). (rr) 19. Glauconome sp. (r) * kK OK OK OCOONDOarRWNW FH From the above list it will be seen that all the abundant and really characteristic species are common to the yellowish shales, both of Cot- tonwood Falls and Manhattan. As far as the present collections are con- cerned, Productus nebrascensis, Owen ; Lophophyllum proliferwm, McChes- ney ; Aviculopecten occidentalis (Shum.) M. and W.; Terebratula bovidens, Morton; Phillipsia scitula, M. and W.(?); and Glawconome sp. have been found only in the Cottonwood shale near Cottonwood Falls, and Crania sp. only at Manhattan. Undoubtedly careful search would increase the number of species at each locality and probably show a larger number common to both regions. Manhattan is 65 miles north of Cottonwood Falls, and there are as many identical species from two localities belong- * indicates that the species is common to the Manhattan and Cottonwood shales. 40) c.S. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. ing to the same formation as could be expected when we take into con- sideration the distance between them. Meek and Hayden reported some of the fossils of this shale from the “bluish, ight gray and brown clays, with occasional layers of magne- sian limestone,” which they described as 35 feet thick on top of the Man- hattan stone and exposed at the same locality 10 miles below Fort Riley. They mentioned Chonetes mucronata, Orthisina umbraculum (?) [probably specimens of Derbya], Monotis, Fusulina, etcetera.* Swallow gives 38 feet of “blue, brown and purple marls, some very much cancellated, and a few beds of thin hmestone ” exposed on Cotton- wood and Clarke’s creek, in which are “‘Chonetes, Productus costatoides, Orthisina missouriensis [ Meekella striato-costata] and umbraculum (?), Syno- cladia, Archxocidaris and Euomphalus.” T Correlation of the Manhattan with the Cottonwood Stone-—Neither Meek and Hayden nor Swallow called attention to the marked stratigraphic and paleontologic characteristics of the Manhattan stone and shale, which are the same for the Cottonwood limestone and shale in the Cot- tonwood valley. I have described the importance of this horizon in the Cottonwood valley and proposed the name Cettonwood formation for the limestone and overlying fossiliferous shales. There seems no question but that the same formation (Cottonwood) is represented by the Man- hattan stone and shales, to which, therefore, the same name should be applied.t As far as observed, this is the most distinctly marked formation in the upper Paleozoic rocks of Kansas, and probably defines a horizon which may be readily traced across the state, and will be of great assistance in dividing this series of rocks for the purpose of mapping. Near the summit of hills or where there is a gentle slope the shale is often eroded and the massive hmestone simply remains, but usually a little careful search will reveal the yellowish, fossiliferous shale in some run or cut with specimens of Chonetes granulifera, Owen; Athyris subtilita (Aall), Newb.; Productus semireticulatus (Martin), de Koninck, etcetera. Locally this limestone has been known for a long time as the Manhattan limestone,§ and if we bear in mind the fact that it belongs in the Cotton- wood formation, this name may be used for that region. * Proc. Acad. Nat. Sci., Phila., vol. xi, p. 17. + Prelim. Rept. Geol. Surv. Kansas, pp. 15, 16. {Since the preceding part of this paper was written I have traced the Cottonwood formation (the limestone and shale) across the country from Cottonwood Falls to Manhattan. The formation reaches the Neosho valley near Dunlap, follows the river toward Council Grove, from there extends northeasterly across the high ground near Bushong and Eskridge to the Mill creek valley above Alma, and then northwesterly to Manhattan. 2Since the above was written Mr Charles D. Walcott has informed me that Manhattan could not be used as a name of a formation, since it is preoccupied by the “ Manhattan gneiss” near New York city. BUFFALO MOUND SECTION. Al Dip of the Manhattan Stone—The Manhattan stone may be easily traced from Manhattan up the Kansas river to Seven-mile creek, where there is a good exposure on the creek bank by the highway just east of Ogden, with the yellow fossiliferous shale on top. Another exposure occurs a short distance east of the Manhattan and Ogden highway, near the Union Pacific railroad, 15 feet higher than the track. This locality is nine miles southwest of Blue mount, at Manhattan, and there is an approximate difterence of 110 feet in the altitude of the Manhattan stone, which gives a dip of 12 feet to the mile.* Meek and Hayden found the dip of a higher stratum extending from the vicinity of Ogden to Chapman’s creek, 28 miles southwest, to be a little less than 14 feet to the mile.t | | The greatest dip is supposed to be to the northwest, and the Manhat- tan stone traced in that direction along the Chicago and Rock Island railroad up Wildcat creek would probably give a greater dip, possibly 20 feet to the mile, as suggested by Meek and Hayden. THe Mint CREEK GEOLOGIC SECTION. BUFFALO MOUND SECTION AND FAUNA. Professor Swallow, as well as Meek and Hayden, frequently referred to the exposures of rocks along Mill creek, a tributary of the Kansas river southeast of Manhattan. The rocks fora considerable distance below the Manhattan stone, and above it extending into the “ flintseries,” are well exposed in bluffs along the creek. On the south side of the creek, three miles southwest of Maple Hill, is a prominent hill called Buffalo mound. The top of the hill is between 350 and 360 feet above the level of Mill creek. Meek and Hayden gave a detailed section of the mound.t At present it is not possible to clearly determine all the beds described by Meek and Hayden in the lower part of the section, although the most important strata are fairly well defined. About 160 feet below the top of the mound is a ledge of bluish gray limestone, which is exposed on the road southeast of the mound and forms a conspicuous ledge for some distance on both sides of the road. This stratum is quite fossiliferous and is probably bed number 2 of Meek and Hayden’s section, although it may belong in the lower part of their bed number 1. The following species were collected from the exposure on the highway: * Gannett (Bull. U.S. Geol. Surv., No. 76) gives the elevation of Manhattan as 1,014 feet. The Manhattan stone on Blue mount is approximately 173 feet above the railroad level, or 1,187 feet above tide. Ogden is given as 1,062 feet, making the approximate elevation of the Manhattan stone 1,077 feet above tide, which would give a difference of 110 feet in the elevation of the Man- hattan stone at the two localities. + Proc. Acad. Nat. Sci., Phila., vol. xi, p. 22. tibid:, pp: 12; 13. 42 ©. S. PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. . Productus longispinus, Sowb. (a) . Productus semireticulatus (Martin), de Koninck, (rr) . Athyris subtilita (Hall), Newb. (7) thynchonella uta (Marcou), Meek. (7) . Spirifer (Martinia) planoconvexus, Shum. (rr) . Chonetes granulifera, Owen. (rr) . Derbya crassa (M. and H.), H. and C. (?). (rr) . Meekella striato-costata (Cox), White and St. John. (rr) . Spirifer cameratus, Morton. (rr) . Chonetes verneuiliana, N. and P. (rr) . Pinna peracuta, Shum. (rr) é / 12. Allorisma geimitzii, Meek (?).. (rr) . Phillipsia scitula, M. and W.(?). (rr) Simply a fragment of the buckler. . Campophyllum torquium (Owen), Meek (?). (rr) . Fusulina cylindrica, Fischer. em Co bo re a = OO WM eI mS i ie) a Ot Abundant in the shaly limestone on top of the massive stratum, but also in the massive rock. 16. Crinoid stems. (r) McFARLAND SECTION AND COMAPRISON OF ITS FAUNA WITH THAT OF BUFFALO MOUND. The bluff on the south bank of Mill creek, opposite the railroad station at McFarland, nine miles west of Buffalo mound, affords a good section. The following beds are exposed in descending order: Feet 10. Yellowish, somewhat porous rock, capped by gray shaly limestones to the summit of the bluff. Feet 9. Light gray shaly limestone near the top of bluff, containing Pseudo- 1155 monotis hawni (M. and H.) and Aviculopecten occidentalis (Shum.), M. and H. 8. Slope mgastly- covered ; few exposures:. ....-::... §-:.+ sen cee Bes = i Shaly licht eray limestone; with Muswlina: 0222.2 4-4. eens == eee } “Geng G: Slope-covered:. -o ces. @eiie soe, (aac ae oe eee gen el OR ets sl= 83 5. Bluish gray massive limestone, containing fossils.................- 1 ee 4. Slope mostly covered: s4.0./\05) See igs eee Qe 6 aes eee 42 = wal 3. Greenish, blue and yellowish shale, alternating with light grayand 7= 9 yellowish limestone, containing plenty of fossils. 2. Coal, 3 inches in thickness. 1. Blue shale, reachine the level or Mill creek”... 5. 3. 2) yas soe a oe The lower part. of number 38 contains numerous fossils, and the follow- ing species were obtained : Chonetes granulifera, Owen. (aa) Chonetes glabra, Geinitz. (rr) Productus longispinus, Sowb. (ce) Productus semireticulatus (Martin), de Koninck. (c) bo ce i FAUNA OF THE McEARLAND SECTION. 43 5. Productus cora, d’Orbigny. (c) Part of the specimens are larger than the ordinary specimens of this species and possibly belong to the P. equicostatus of Shumard;* but Dr White states ¢t that P. cora sometimes reaches a large size. 6. Spirifer (Martinia) planoconvexus, Shum. (c) 7. Productus nebrascensis, Owen. (7) _ 8. Productus symmetricus, McChesney. (77) 9. Spiriferina kentuckensis, Shum. (rr) 10. Hustedia mormonii (Marcou), H. and GC. (r) 11. Meekella striato-costata (Cox), White and St. John. (rr) 12. Syntrilasma hemiplicata (Hall), M. and W.” (rr) 13. Spirifer cameratus, Morton. (rr) . 14. Terebratula bovidens, Morton. (rr) 15. Derbya crassa (M. and H.), H. and C. (rr) 16. Phillipsia scitula, Meek and Worthen. (rr) 17. Synocladia biserialis, Swallow. (r) 18. Rhombopora lepidodendroides, Meek. (7) 19. Chextetes sp. (rr) - 20. Crinoid, stems and pilates. (7) 21. Fusulina cylindrica, Fischer. (c) From the second limestone—stratum omamben & of the section—Mr Warren Finney, who assisted me in the work on Mill creek, obtained the following species : . Productus longispinus, Sowb. (a) . Spirifer (Martinia) planoconvexus, Shum. (c) bp Fe 3. Hustedia mormoni (Marcon), H. and C. (r) 4. Athyris subtilita (Hall), Newb. (r) 5. Spirifer cameratus, Morton. (1) 6. Rhynchonella uta (Marcou), Meek. (r) 7. Derbya crassa (M. and H.), H. and C. (?). (7) Not clearly preserved. 8. Chonetes granulifera, Owen. (rr) 9. Productus nebrascensis, Owen (?). (rr) 10. Productus cora, d’Orbigny (?). (rr) 11. Spiriferina kentuckensis, Shum. (rr) 12. Discina nitida (Phillips), Meek and Worthen. (rr) 13. Meekella striato-costata (Cox), White and St. John. (rr) 14. Pinna peracuta, Shum. (r) 15. Myalina subquadrata, Shum. (rr) 16. Allorisma subcuneata, M. and H. (rr) 17. Aviculopecten occidentalis (Shum.), Meek and Worthen. (rr) 18. Straparollus (Huomphalus) subrugosus, Meek and Worth. (rr) 19. Phillipsia scitula, Meek and Worthen. (7) 20. Fusulina cylindrica, Fischer. (c) * Geol. Surv. Mo., 1855, part 2, p. 201. 7 Thirteenth Report Indiana Geol. Survey, p. 126. 44 ©. 8, PROSSER—PERMO-CARBONIFEROUS AND PERMIAN ROCKS. It seems probable that the above stratum is the same as the bluish gray, fossiliferous limestone described on Buffalo mound, although the stratum has not yet been carefully traced from the mound to McFarland. ALMA SECTION. Its Thickness and Fauna.—At Alma, 4 miles southwest of McFarland and 18 miles southeast of Manhattan, on the east bank of Mill creek, is a section of the same rocks as those exposed in the bluffs of the Kansas river near Manhattan. The following section refers to that part of the bluff on which is the quarry of the “ Sunflower cement works,” capped by an exposure of the Alma massive limestone: Inches. Feet. 12. Thin limestone but a few inches beneath the soil.......... ..... 0) == et Feet. Feet, 11. Yellowish shale, with concretions in the upper part and abundant 10 = 1803 fossils in the lower portion. 10. Light yellowish gray, massive limestone, locally called the ‘‘Alma 53 = 1703- stone.”’ Qo Coyered, Slope. wie 2 sak sap e ses ae ooo sae Ate oe 40 =165 8. Argillaceous, thin bedded limestone, with a hard, irregular lime- 10 = 125 stone at the base. 7. Slope showing outcrops of shaly limestone and shales, but mainly 18 =115 covered. In the shaly limestone or calcareous shales are speci- mens of Fusulina cylindrica, Fischer. 6. Massive grayish limestone at top of cement quarry .............. 2 St OPO livessialer ss 5 kavaeeiveey pee ee eee ene ae pls UO A EN = 4. Drab limestones, with one layer of drab shale. Thelimestonesare 53—= 88 used for cement and known as numbers | to 3 of the cement quarry. 3. Yellowish, very friable chalk-like limestone, number 4 of the 5 = 824 cement quarry. 2. Bluish and yellowish white shale, the lower containing concretions 73—= 774 I. ‘Covered slope to: Malliercek inn). aes eso on eine ee ee _ a Omen From the lower part of the yellow shales, number 11 of the section, the following species were obtained : 1. Chonetes granulifera, Owen. (aa) 2. Athyris subtilita (Hall) Newb. (a) 3. Productus semircticulatus (Martin) de Koninck. (c) 4. Mecekella striato-costata (Cox) White and St. John. (rr) 5. Derbya crassa (M. and H.) H. and C. (?). (rr) 6. Rhombopora. lepidodendroides, Meek. (rr) 7. Fusulina cylindrica, Fischer. (rr) 8. Archxocidaris sp. (ce) 9. Crinoid stems. (rr) Correlation of the Alma with the Manhattan Stone-—The massive lime- stone (number 10) below the yellowish shales, called the Alma stone, is FAUNA OF FLINT BED ABOVE THE ALMA STONE. A5 quarried to a considerable extent near Alma, and is simply another ex- posure of the Manhattan or Cottonwood limestone. Therefore the mas- sive limestone, with the overlying yellow, fossiliferous shale, near Alma represents the rocks which I have called the Cottonwood formation. Professor Swallow mentioned-the occurrence on Mill creek* of his Fusulina limestone, number 80 of his section (which we have shown is the Manhattan and Cottonwood limestone) ; his ‘‘ Cotton rock,” number 82 (the argillaceous limestones at the top of our number 8), and his “ Dry bone limestone,” number 84 (the hard irregular limestone which with the overlying argillaceous limestones we have called number 8 of the Alma section). | This lower limestone Professor Swallow called the base of the Permian, and he stated that there was an unconformity on Mill creek, where this ‘‘ Dry bone limestone” (number 84) rests on what he called the * Fusulina shales” (number 96), while beds numbers 85 to 95, which represent strata 86 feet in thickness exposed at Manhattan, are missing on Mill creek.t First Flint Bed above the Alma Stone-—Near the top of the high hill east of the Cement and Alma stone quarries, and two miles east of Alma, is another limestone quarry. The stratum is not so massive as that of the Alma stone, and above it is a shale which is capped by layers of limestone alternating with chert and covered by a few inches of soil. This chert or flint bed is barometrically about 145 feet above the top of the Alma stone, or between 310 and 315 feet above the level of Mill creek. On the summit is an excavation in which the alternating layers of gray limestone and chert are nicely shown. ‘This shaly limestone is somewhat fossiliferous. A few fossils were also found in the chert, and the following species were obtained at this locality : . Syntrilasma hemiplicata (Hall), Meek and Worthen. (c) . Athyris subtilita (Hall), Newb. (c) . Chonetes granulifera, Owen. (rr) . Derbya crassa (M. and H.), H. and C. (?) (rr) Productus nebrascensis, Owen. (rr) . Pseudomonotis hawnt (M. and H.). (rr) . Bryozoan sp. (77) NOOR WD He THE GEOLOGIC SECTION OF THE UPPER KANSAS RIVER. GENERAL CHARACTERISTICS. Above the Manhattan stone on the Kansas river is a series of blue, drab and chocolate colored shales alternating with thin buff, drab and * Prelim. Rept. Geol. Surv. Kansas, p. 16. + Ibid., p. 44; also, see Trans. Acad. Science, St. Louis, vol. ii, 1868, p. 521. VII—Butt. Geou. Soc. Am., Vou. 6, 1894. 46 ©.S. PROSSER—PERMO-CAKBONIFEROUS AND PERMIAN ROCKS. bluish limestones. 0)...4- des. 508 gee oe 91 Correlation of the quartz-veims | ic2. oss. ¢ 25 oe tye pew oe see ee 92 Age-of the sedimentary series). 025.0206. 00.945 +4 a) Weaelelece en 93 Paleontologic evidence... +2 gk Oi 0.4 ons eae a et ae keener 93 Stratigraphicand Mthologie evidence! x. -2.0.5.% eet +e en 94 Cretaceous of the Coastrerammests %. ssivosdnch ediane ee gone eee 2 gia alsa seas steelers 131 Bilagic bearing, of the physical ‘changes of level... i..0-.. Joe. 22 ce eee es we 134 Sete PC TA can Ney Oe AS a eon iat) earn sintcenge tea Gere Lae a bees 139 INTRODUCTION. For decades a vague impression has existed that some kind of conti- nental extension formerly united the West Indies to the mainland of one or the other of the continents, or to both, and that the exclusion of the At- lantic currents and the adinission of the Pacific waters into the Mexican eulf and Caribbean sea have at some time prevailed. Suggestions have been made as to the necessity of a foreign source for the mechanical sedi- ments found in the islands which do not possess the geological quarries for the supply of such materials. Wallace* has illustrated how the Antil- *“Tsland Life,” by Alfred Russell Wallace, second edition, 1892, p. 263. XV—But1. Geox. Soc. Am., Von. 6, 1894. (103) 104 J. Ww. SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. lean lands would be extended if the region were uniformly raised 6,000 feet. The charts of Agassiz * and others show the positions of the sub- merged basins. Geologists have demonstrated the great changes of level to which the Caribbean region has been subjected, but I am not aware that any geologist has hitherto + attempted to restore the topography of the submerged continent and set forth the geomorphic evidence that the drowned valleys are those of former lands, now depressed beneath the sea. The later changes of level have had far reaching effects, not only on the geographic forms, but on the climate of, perhaps, the whole At- lantic basin. These changes have also placed obstacles in the way of the distribution and development of life, or cut off and exterminated such forms as reached the Antillean lands, and have given rise to the modern distribution of Atlantic and Pacific types in the waters of the West Indies. The present contribution is primarily based on the physical geology of the Antillean region. The phenomena of first importance are the fjords, revealed by the now numerous soundings, in connection with the land valleys of the continents and islands. Some of the yalleys are now wholly submerged, even to their sources. The formation of the valleys at various elevations, depending on baselevel of erosion; the continua- tion. of the valleys into fjords; the deformation of deserted water-mar- gins by warping or terrestrial undulations, as recorded in the -elevated terraces, beaches and other shore phenomena; the successive cycles of. erosion and filling of the valleys, with the consequent unconformity of the strata, and the distribution and elevation of the later formations, with the subsequent erosion—all of these phenomena are made use of in the study set forth in this paper. Some difficulties yet remain, but to geo- morphy we may hopefully look for their removal. The dynamic studies explain many biologic phenomena, and these in turn support the con- clusions regarding the physical history of the continent. This investigation is the sequence to years of inquiry into the geologic history of the Great lakes. Their depth indicates a former elevation of the continent considerably above the present altitude In looking for proofs of this greater altitude it was found that the fjords of the coast in higher latitudes are only submerged valleys.{ At that time the writer hesitated to admit continental changes of level greater than 3,000 feet, but with increasing knowledge of deep submerged valleys elsewhere, no- tably the much deeper fjords of the Antillean region, and with the find- *“ Three Cruises of the Blake,’ by A. Agassiz, vol. 1, 1888, figs. 56 and 57. + An advanced notice of some of the drowned valleys appeared in Bull. Geol Soe. Am.. vol. v, 1893, p. 19, under the title of “Terrestrial Submergence Southeast of the American Continent,” by the writer. {‘* High Continental Elevation preceding the Pleistocene Period,” by J. W. Spencer, Bull. Geol, Soe. Am., vol. 1, 1889, p. 65. INTRODUCTION. 105 ing of abyssmal deposits, described by Messrs Jukes-Browne and Harri- son,* and showing great changes of level in one direction, he was led to the conclusion that the great depth of the drowned valleys is not inconsist- ent with their fluvial origin, and this opinion has been sustained by more recent discoveries regarding their connections with existing and buried river channels of the continents andislands. It has been found that these fjords are not exceptional or scattered, but numerous and apparently connected into systems of drowned valleys. In order to extend our knowledge of the connection between the Antilles and the continent, and to obtain evidence as to when the connection existed the field-work was carried from the continent to the Greater Antilles with success beyond expectation. The later geology of that region proved that with minor modifications the terrestrial movements of the coastal plain of the con- tinent, so admirably set forth by Mr W J McGee,f extended to the West Indies. The geographic history of Jamaica was not explained by the government geologists, but their excellent data, as also those of Dr W. M. Gabb§ in San Domingo and Costa Rica and the more scanty surveys in Central America, all contain considerable material for extending gen- eralizations based on discoveries in Cuba and the southern states (where the writer has done much work) over the whole Antillean region. These generalizations are further justified by the distribution of hfe in the waters of the West Indies. GEOMORPHY AS EXEMPLIFIED IN VALLEYS. | In northern regions the geologic broom of the Ice age swept over the hills and filled the valleys so as to greatly obscure the topography pro- duced by meteoric agencies. Farther southward the atmospheric forces have left their record deeply engraved in the surface rocks. From the study of land features it would appear that the forms of the valleys are largely independent of the geologic undulations of the region, and that open valleys (and sometimes those closed into basins) are to a great ex- tent the direct result of atmospheric erosion. For example, over the plains of both the coast and interior, valleys from a few rods to many miles in width may either follow the strike of the rock strata of the country or may cross their strike at any angles. In the southern Appalachians numerous valleys follow the direction of the ridges, which: is generally that of the strike, and at first sight appear to occupy mountain folds. * “Geology of Barbadoes,” Quar. Jour. Geol. Soe. London, vol. xlvii, 1891, pp. 197-250, and vol. xlviii, 1892, pp. 170-226. 7‘ The Latayette Formation,” 12th Ann. Rep. U.S. Geol. Survey, 1892, pp. 347-521. t “Geology of Jamaica,” by J. G. Sawkins, 1869. 2 ‘* Geology of San Domingo,” Trans. Am. Phil. Soc., vol. xxv, 1873. “Geology of Costa Rica,’ Am Jour. Sci., vol. viii, 1874, p. 388, and vol. ix, 1875, pp. 198, 320; also manuscript in archives of the U.S. Geol. Survey 106 J.w. SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. This, however, is rarely the case, for if the denuded strata were restored the position of the valleys would occupy the crests of the undulations, as shown in the accompanying actual sections, where the broad valleys occupy the surfaces of both anticlinals and synclinals. Another char- acteristic of old valleys is that the streams flowing through them are in- significant compared with their magnitude in both depth and width. The limestones have been carried away in solution and the finer mechan- Figure 1.—Section across Lookout- Wills’ Valley, Alabama, at the Col. The incision in the tableland of Carboniferous sandstone here is about 4 miles wide and 500 feet deep. From this section, for many miles downward in both directions, the general slope of the floor of the united valley is about ten feet per mile, so that the tableland presents bold esearp- ments more than 1,000 feet above the lower reaches of the rivers. It is an anticlinal valley. ical muds have been washed by the rains into the larger streams and, suspended in the waters, they have been transported out of the valleys and deposited on flood-plains or in the sea. Occasionally the streams undermine the banks and obtain extraordinary cargoes, but the principal widening agents are the rains and rills that everywhere wash away the surface and undermine the mountain sides, which action is intermittently retarded by the temporary protection of the unremoved materials of the Pigeon Mt.--“"_ ~ . - - - . - . - . Scale ee MS 729. Vertical a 20> 10°° Feet Figure 2.—Section from Pigeon to Little Sand Mountain, Georgia. This section represents a valley of about a dozen miles in width, with the complex geologic base shown in the figure. It illustrates well how the valleys in the southern Appalachians are produced by atmospheric denudation and not by mountain folding. landslides. While the tablelands are high, the streams are constantly deepening their channels, but when the bottoms of the ravines are re- duced to the baselevel of erosion, then the streams almost cease to corrade and become geologic carriers of the surface washings of the valley. In the recently elevated mountains of Cuba and Jamaica, as in the older Appalachian chain, the valley-making forces have overcome the physical structure, so that the valleys are more or less independent of it. GEOMORPHY AS EXEMPLIFIED IN VALLEYS. 107 When the depressions in the rock surfaces grow wider and deeper and are kept open in their descent, one can onlv conclude that they represent the molding of old land surfaces by running water, no matter whether the valleys be now buried by drift.or submerged beneath the sea. The eeomorphic studies especially consider narrow canyons with steep walls, made by rapidly descending streams; broad valleys with commonly sloping sides, arising from rain-washes upon the hillsides when the drainage of the depression was reduced to the baselevel of erosion; the burial and the reéxcavation of the valleys; their terracing; the tilting of the land surfaces and the change of direction of the drainage or the closing of the basins into lakes. From all of these phenomena we can learn something of the geologic history of the land, while the rock formations indicate the contemporaneous history of the sea. To repeat, when we Figure 3.— Cross-section (v) representing a baselevel Valley in Trinidad Mountains, Cuba. This district was recently elevated, anda stream at c is engaged in cutting back a canyon 600 feet in depth. find systems of valleys beneath the sea, unless there are other local causes, we are led to conclude that they are the remains of land features now submerged, or, in other words, that they are evidence of former conti- nental elevation. Under this interpretation the writer has correlated the extension of the great rivers into their fjords cut through the conti- nental shelves along the coast, which have been made known by the numerous soundings.* DEFORMATION OF LAND SURFACES. The gentle but broad undulations in the earth’s crust which change the; relation of the land and sea and raise up barriers across valleys, so as to form basins, or divert the drainage of the land without producing crumpling and folding of the strata or other great distortions, such as in mountain uplifts, have been by Mr Gilbert denominated epeirogenic (continent-making) in contrast with orogenic (or mountain-making) movements. Such undulations are well known in the gentle rising and sinking of coasts. In the deserted beaches of the region of the Great * See U.S. Hydrographie charts nos. 31-36, 21, 21a, 1007, 1411, Coast Survey charts C, D, and numer- ous harbor charts. The Hydrographie charts have on them many soundings taken from the British Admiralty and other surveys adjacent to the islands and coast of Florida not given on the Coast Survey charts. These British Admiralty charts have also been studied. 108 > 3. Ww. SPENCER—KECONSTRUCTION OF ANTILLEAN CONTINENT. lakes we have been able to measure the recent deformation of the earth’s crust, since the abandoned shores rise from zero to three, six or even more feet per mile toward the northeast, as shown in figure 4. The deforma- tion is also recorded in the variable altitudes of the recent geologic formations shown in figure 5. Everywhere the movement of elevation ° so 100 MILES 5 1200 FEET ; be I a SS SS 3 ta : co Fiaure 4.—Section of the Iroquois Beach. A raised water-line from the head of lake Ontario for about 400 miles to south of Malone in the Adirondacks. The rise increases from an elevation of 363 feet above tide near the lake to 1,482 feet in the mountains, where the greatest deformation occurs. appears to be greater in the mountain regions than on the plains, and it is noteworthy that the rate of depression along the coastal margins seems larger than farther inland. The epeirogenic movement does not gener- ally deface the topographic features, although it sometimes changes the direction of the drainage and turns the valleys into basins; and in this 100 MILES O 4 8 12 I6KUNDRED Ce FEET. * (Se Figure 5.—Section from South Carolina to the Mississippi, showing Deformation of the Lafayette Formation. At Columbia, South Carolina, the elevation of the Lafayette formation is about 800 feet above tide; in the mountains farther westward it has double that altitude, and in Arkansas less than 300 feet. respect it has been an important factor, inclosing the Mexican, Honduras and Caribbean valleys into sea basins, though orogenic and volcanic forces also combined with the gentler terrestrial undulations in producing these abysses. SUBMARINE VALLEYS AND FJORDS OF THE CONTINENTAL AND ANTILLEAN REGIONS. THE CONTINENTAL SHELF. In passing southward from New Jersey the continental shelf narrows to only about 15 miles wide off cape Hatteras (section / F”’ on accom- panying map), but from that point it widens to nearly 300 miles east of Florida. Although crossed by the straits of Florida and the GulfStream, the plateau embraces the Bahamas, east of which its broken remains rise in banks north of Haiti and beyond it has been largely swept away. This shelf has been mostly removed by denudation, only fragments, of it remaining east of the Windward islands, but off the coast of South ° THE CONTINENTAL SHELF he) s 109 America it reappears, having a width of from 150 to 500 miles. An elevation of only 50 feet would turn the Bahama banks into a broad land area, and a rise of 100 to 300 feet would extend the continent to the edge of the upper shelf in front of the southern states, coastwise of which the lower terrace, or Blake plateau (so named by Professor A. Agassiz), is depressed from 2,500 to 8,500 feet. A considerable portion of this drowned plain has an average depression of 2,700 feet. Hast of the continental shelf (both the normal and the Blake plateau) the continental margin descends rapidly to a depth of 12,000 feet or more. Glancing at the accompanying map (section B B’), the drowned flats, seaward of the coastal plain of the southern states, behind which are the Appalachian mountains, and the Bahama banks in front of the plains and mountains of Cuba and Haiti, may be seen without any epeirogenic break and with only the interruptions of a few great valleys, the ana- logues of which occur on the continent, indicating that the whole conti- nental shelf is a geologic unit and must be so treated. The Blake plateau may be somewhat modified by the Gulf Stream, the bottom of which in passing over the highest ridges of the straits of Florida is only about 2,100 feet below the surface, or several hundred feet above the mean depth of the water over the Blake plateau; but it is to be remembered that the force of the current, which diminishes with its depth, is very much re- duced except on the more elevated surfaces of the submerged terrace. It might be noted here that in such places the current keeps the surface of the rocks free from loose deposits (Agassiz). The depressed plateau has all the appearances of the modern coastal plain of the continent, as if formed in the same way, which would indicate a long continued ele- vation of the region as great as is the present submergence. This shelf is traversed by great valleys and fjords, and its margin is indented by cor- responding embayments. The western side of the Florida mass shows the continental shelf sub- merged to a depth of about 3800 feet, or somewhat more, beyond which there is a steep descent to 10,000 feet. The northern slope of the Mexi- can basin is not so precipitous as that of the Florida shelf (section A A’ on map). Moreover, in the vicinity of the Mississippi fjord there is a terrace like the Blake plateau submerged to about 4,000 feet. North of the Yucatan banks there is a precipitous descent (section A A’ on map) from the shelf (which is there submerged from 50 to 300 feet) to the floor of the Gulf basin, about 12,000 feet below the surface. This trough con- tinues to the isthmus of Tehuantepec. The continental shelves also occur in the Caribbean sea and the sea of Honduras. These basins are separated by the Honduras and Rosalind banks and Jamaica. On the northern side the sea of Honduras is sepa- 110 J. W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. rated from the gulf of Mexico by the Yucatan banks and Cuba. Owing to the demarkation just given and the great depths of the central basin, the necessity for a distinctive name in future discussion has caused me to give the basin the appellation of the sea of Honduras. The imperfect demarkation of the submerged shelves or terraces in the two seas named may partly arise from fragmentary knowledge, due to scanty soundings in some portions of the basins. The deeply submerged Cayman ridge, terminated by the Misteriosa banks, the insular plateau in the eastern part of the Caribbean sea, and the extended plateaus off the Honduras banks, about which we know very little beyond their general submerg- ence to depths of 4,000 or 5,000 feet, suggest old coastal plains, now be- neath those seas, and consequent pauses in the terrestrial oscillations at those stages. It should be noted that the gulf of Mexico and Caribbean sea are gen- erally broad valleys (sections A A’and C C” of map) and have their coun- terparts on the Pacific side of the continent; but the sea of Honduras is different, being composed of two very deep channels, almost land-locked on both sides, but situated between the mountain ridges of Jamaica, Haiti and Cuba, rising from 7,000 to 9,000 feet above the sea. The sea of Honduras reaches a depth of 20,000 feet, while the Caribbean basin de- scends to 15,000, and the gulf of Mexico to only about 12,000 feet. DROWNED VALLEYS OR FJORDS. The submerged valleys may be divided into three classes—the notable embayments in the continental plateaus, the valleys crossing the sub- merged shelves, generally at right angles to the mountain ranges of the land, and the valleys or fjords parallel to the Antillean chains of moun- tains. | The embayments are everywhere notable at the terminations of the fjords, and where several such end near together they are broad and conspicuous. As the fjords and embayments coalesce and are more or less apparent through all of the detailed contours, it may be well to de- scribe them at the same time. Lindenkohl* has deciphered the records of the canyon of the Hudson, now submerged to 2,832 feet, where the plateau is depressed to only 600 feet below the surface of the sea. He hasalso found in the same plateau another fjord with a depth of 2,534 feet. This is a continuation of the united valleys of Great Heg Harbor and Little Ege Harbor rivers. The Delaware extension 1s indicated on the map, but not extended for want of the necessary soundings. Along this section of the coast the conti- nental border descends precipitously from about 400 to 8,000 feet. * American Journal of Science, vol. xli, 1891, p. 490. DROWNED VALLEYS OR FJORDS. 111 From the Susquehanna the special study of the writer begins. Its old channel is now buried and passes under the peninsula of Maryland, where the deposits of sand have extended the surface of the land. The coastal currents have done much to obscure the old channels by deposit- ing sand in them, for the water is deeper than 200 or 300 feet only on approaching the margins of the continental shelves, and there, as well as on the lower slopes of the plateaus, the drowned valleys reappear. Thus the Susquehanna fjord is 6,420 feet where the plateau rises 1,800 feet above the channel. Again, where the depth reaches 9,846 feet, the rocky boundary of the canyon is still 1,500 feet above its drowned floor. Even at 12,000 feet the fjord opens into an embayment in the edge of the plateau, and has a width of 40 or 50 miles, or about that of the lower Mississippi flood-plain. The fjords of Blake plateau are especially remarkable. In spite of the tendency of the Gulf Stream to silt up the transverse channels and the filling produced by the coastwise drifting sands, and in spite of the sometimes incomplete soundings, several important drowned valleys have been discovered. The Santee and Pee Dee rivers formerly joined and cut a deep channel across Blake plateau, the soundings indicating channels still open to the depth of several hundred feet. The old Sa- vannah valley has cut its way through the whole remnant of Miocene formation, as shown by well-borings submitted to Mr Louis Woolman, who has found in the higher limestones the characteristic microscopic Upper Eocene shells, and is now buried to a depth of 250 feet by more recent sandy deposits. We do not know that this is the greatest depth, since the wells may not be in the center of the buried valley, which is several miles wide. Crossing Blake plateau, a submerged valley in line with the Savannah embayment is still left open to a depth of over 1,650 feet, with the plateau depressed to 1,950 feet more below the surface of the sea. After traversing Blake plateau for a distance of 250 miles, this fjord enters the deep embayment in the edge of the continental shelf. Another remarkable submarine valley is that in line with Altamaha river.* Ata point where the plateau is submerged 2,500 feet this canyon- lke depression reaches a depth of more than 7,800 feet beneath the sea- level, and 300 miles farther off shore the depth is 18,560 feet, with the outer embayment considerably deeper. This valley is comparable to the canyon of the Colorado river of the west. The depression which the writer has called the Bahaman fjord also crosses the same plateau just north of the group of small insular remains of the coastal plain which * Hastward of this region the soundings are not numerous, but at the edge of the plateau evidence appears to show that the canyon is much deeper. Although this depression at 7,800 feet is not con- firmed by a chain of soundings, its absence would not affect any other portion of the argument, as it was an after observation. XVI—Bott. Geor. Soc. Am., Von. 6, 1894. 112 J.W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. now constitutes the Bahama islands and banks. ‘The configuration here indicates that formerly the Bahaman valley drained the northern side of what is now the ridge underneath the Gulf Stream where it emerges from the straits of Florida. ‘This channel is entirely submerged to depths of from 2,064* to 10,314 feet in a distance of 350 miles. Near its mouth the fjord is bounded on one side by the plateau, rising 6,800 feet above its floor, and on the other by the watershed between it and the Altamahan fjord, which has also the great elevation of over 5,000 feet, although so deeply submerged. The Abacan fjord (named from the adjacent island) differs from the last three in that it passes between low islands, though crossing the same subcoastal plateau. It extends eastward from a point west of the Bahamas, where the depth is nearly 2,400 feet, with the trans- verse ridges of the straits of Florida considerably higher. Thirty miles eastward it is 5,688 feet below sealevel, and increases to a depth of over 10,000 feet south of Great Abaco island, where it joins another canyon- like depression no less clearly indicated by the soundings. The Abacan fjord is of particular interest as showing that the Gulf Stream has taken possession of three different valleys. Its presence also indicates that while the Gulf Stream has probably deepened the divide between the northern and southern valley (deepest sounding, 2,064 feet), yet the val- leys existed before the Gulf current, for, while eroding the higher cols, the current tends to fill up the deeper channels, into the quieter waters of which the sediments settle; but the Abacan channel, not receiving the Gulf Stream, except where crossed at its head, has not been filled by the débris of the current-scour even in its upper reaches, thus suggesting that this marine current has not made, though it may have modified, the valley of the Floridian strait. The Androsan fjord is 4,500 feet deep at the head of the tongue of the ocean and 8,940 feet near the mouth of the Abacan, below which the two valleys are united and enter the oceanic embayment with a depth of 14,178 feet. Here the soundings indicate an interesting feature. Into this embayment, which is 120 miles wide, there enter four great fjords (Altamahan, Bahaman, Abacan and Androsan). The embayment is bounded on the northern side by an island, now submerged to a depth of about 4,000 feet, and in front of it there is another, rising to 5,000 feet above the floor of this arm of the Atlantic basin. All ofthe submerged channels crossing the Blake plateau have directions at right angles to the mountain ranges of the continent, and apparently represent continuations of the existing rivers. There are also many short, drowned canyons among the Bahamas and banks, but the soundings are * In latitude 2734°, longitude 794° W., the channel has reached a depth of 3,882 feet, as shown in an unpublished sounding of the Hydrographie Office. FJORDS OF THE GULF OF MEXICO. 113 not everywhere sufficiently numerous to work out the details. Among the Bahamas the Exuman depression increases to a depth of over 6,000 feet, and may belong to the same group of drowned valleys as those already described, or possibly to those parallel to the mountain folds of the Great Antilles; still it is 250 miles distant from any mountains. In the eulf of Mexico the similar submerged valleys exist, and their interpretation has been confirmed by recent studies of the writer, in which evidence has been found of several old valleys, now somewhat filled in their lower reaches. ‘This has led to investigation of the char- acter of the depressions, which in turn has bound together in indissolv- able union the existing land valleys and the submerged fjords. Thus the drainage of southern Florida flowed by Key West, and the channel is recognizable near that key at a depth of 2,400 feet, where the adjacent submerged land is only 3840 feet below tide. With rocky banks of 600 feet in height, it is still apparent at a depth of 5,600 feet, and just beyond it joins the great Floridian fjord. On the western side of the Floridian mass the Tampan, Suwaneean, Apalachicolan and other fjords all in- cise the rapidly descending continental margin to a depth of over 10,000 feet. The Apalachicolan, Escambian, Mobilan and Mississippian fjords all unite to form the great Mississippi embayment of 9,000 feet in depth, and where deeper the basin becomes a channel-like depression, increas- ing to 12,000 feet, or about the full depth of the Mexican gulf. The Apa- lachicola represents a broad valley extending far up into the Appalachian mountains, and so do the Escambia and Mobile rivers—all the valleys being several miles wide. Even in its upper portion the Mobilan fjord has a depth of 700 feet, and the shelf which it crosses is itself submerged to 1,500 feet. The Escambian fjord shows a depth of 1,800 feet beneath its walls, which are 1,200 feet below tide level. All these drowned valleys are recognizable in the soundings to a depth of over 10,000 feet. West of the Mississippi there are other fjords which are apparently not con- nected with the greater drainage of the modern rivers. One of these indicates that during one of the periods of elevation the Mississippi made another channel west of the present one. The Red River canyon was also independent before it united with the Mississippi on its present flood-plain, which is from 30 to 80 miles wide. Again, the rivers enter- ing the northwestern region of the gulf have continued seaward, to form fjords and other embayments. - Thus in front of the Rio Grande drowned valleys are recognizable, descending from about 600 feet to 8,280 feet, where the sides of the valley rise 3,000 feet above the bottom of the de- pression. Although the soundings are sometimes scattered, there are enough to indicate these depressions with clearness, and the occurrence 114 J.W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. of the great depressions in lines with the existing embayments seems fully to warrant the inference of formerly continuous drainage systems: On the northern side of Yucatan the descent from a depth of 300 feet to 12,000 feet is generally precipitous (section A A’ on map); but cross- ing the submerged terraces one great valley is known—the Yucatan; which is apparent from 3,500 feet below the surface to nearly 12,000 feet, with a short but deep tributary from the west. Its landward connections have not yet been recognized. The Floridian valley is drowned for a length of about 400 miles, and descends from a depression of 2,064 below the surface to 6,000 feet oppo- site Havana, and to a depth of 12,000 feet after passing the line of the Florida and Cubashelves. Several confluents tributary-like have already been revealed by the soundings, as those from Havana, Matanzas, Cabanos (3,075 feet, with the adjacent shelf only 1,200 feet) and two large branches from between the Bahamas and Cuba, with cols not over 1,500 feet below the surface. Longitudinal and transverse sections of the Floridian fjord are shown on the accompanying map. The slow deep- ening and increase in size of this Floridian valley is much like that of a river. Part of this valley joins the Abacan fjord. The fjords of the sea of Honduras are remarkable. Of the type thus far treated, that of the gulf of Cazones is most notable. Near the head of the gulf the submarine valley is 2,250 feet deep, and it has tributaries from Cochinos and Xagua bays. Still enclosed between the land and the keys for a distance of 70 miles, the fjord of Cazones increases in depth to 7,500 feet before joining the outer valley. The Cayman repre- sents another depressed channel uniting with that from the gulf of Cazones. Fjords parallel to the mountain folds are best represented by the nar- row channel north of Haiti (the Haitian, shown in section CC’ on the map). Its source north of Cuba is depressed 1,500 feet below tide, but it deepens, so that at its mouth it is 13,746 feet below the same level. While this drowned valley cuts through the continental shelf, it is par- allel to the mountain ranges and part of this depth may have been due to folding by mountain movements, such, however, as did not close the submerged valley and forma basin. There are other notable valleys ex- tending westward from Haiti which are also parallel with the mountain folds. These connect with Bartlett deep south of Cuba. This last de- pression forms a narrow trough extending westward for 600 miles and strikingly resembles a land valley ; but it reaches to the enormous depth of over 20,000 feet, while its western end is closed by Honduras and entral America. It lies between the great mountain folds of Cuba, Haiti and Jamaica, but these mountain disturbances grow weak before reach- FJORDS OF THE CARIBBEAN SEA. 115 ing the western end of the deep. Interpreting it as originally a level valley, Bartlett deep may not indicate a general elevation of the adjacent lands to the amount of the subsidence, which here was probably ampli- fied in the foldings of this mountain region. Of the same character is the depression between the Virgin islands and the nearest of the Wind- ward ridge (Santa Cruz), where the depth is 15,000 feet, although the outlet to the basin is not known to exceed 10,000 feet. It seems probable that both these depressions are submerged valleys, and that the epeiro- genic movements of the region have not been obliterated by the orogenic, since all of these remarkable deeps have great tributary fjords, of the ordinary type, which are known to descend to considerable depths. The Caribbean sea is essentially a basin, but it receives numerous short tributary canyons. That from the gulf of Paria (a mouth of the Orinoco) is very noticeable, extending as it does to a depth of 12,000 feet. Between South America and Granada the Windward ridge is depressed to 2,526 feet. North of the Grenadines the sea is less than 1,600 feet, with a westward bound channel recognizable in the present soundings to 3,600. Between Saint Vincent and Saint Lucia the sea is reduced to a depth of less than 3,000 feet, with apparent fjords increasing to over 6,000 feet towards the west. North of Saint Lucia the submergence has de- pressed the ridge to 3,500 feet, but with drowned valleys noted to 6,000 feet, westward bound. North of Martinique the ridge is 4,000 feet below the surface, with westward opening valleys shown to a depth of 6,000 feet.. North of Dominica the depth is 3,300 feet, with a westward drainage. Beyond Guadeloupe the ridge is submerged to 2,400 feet, with drainage in both directions. Beyond these small islands there is the great deep of the Virgin group noted before. It is apparent in the hydrography, from South America, that the Windward ridge extends and forms a sep- arating barrier between the Atlantic and Caribbean sea, nowhere lower than between 1,600 and 4,000 feet below tide-level, with the valleys gen- erally declining westward. The eastern side of the ridge descends rap- idly to the Atlantic, with relatively short valleys strongly suggesting subaérial sculpture. The recent submergence is further marked by the depth of the Orinoco, which reaches to 360 feet below tide level at points 400 feet from its mouth. South of Haiti and of Jamaica the fjords are apparent in the bays. Thus Morant bay, the Saint Lucia and the Mosquito cove deepen rap- idly to 600 feet or more, while the depressed coast is not flooded toa greater depth than 60 feet. The valley of Savannah la Mar is nearly 2,000 feet deep, yet the coastal shelf is not more than 600 feet below the surface; nor do we need to go to land-locked bays for drowned chan- 116 J.W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. nels. On the Honduras banks, where the water is about 175 feet deep, there are several drowned canyons as shown in figure 6. On the northern side of this submerged plain the channel forms a canyon trace- able to 690 feet and on the southern side to 940 feet. The passage be- tween Rosalind and Pedro banks, with a depth of 4,400 feet and a width of 80 miles, connecting the Honduras and Caribbean seas, is only the union of the valleys on the opposite sides of the ridge, apparently broad- ened and deepened by marine currents established when the submer- gence was half or less that of the present. The channel between Pedro bank and Jamaica, with a depth of about 3,000 feet and width of 40 miles is of similar character. That between the sea of Honduras and the gulf of Mexico reaches a depth of 7,000 feet and a width between the banks of 100 miles, or 140 miles from land to land. This is of like 1) ee - ene RC 1000 foe) ee ; ya RELAY Pago \ 2 = ren 499 FAD, ng 97,183 Gi B23 “37 38%, 4550 2 as a ee 561 28 «36.2 im pes a is 2, po 12 + x05! Bajo Nuevo eesor New Bore aa 120 Ficure 6.—Map of Honduras.and Rosalind Banks, showing Fjords. (From Hydrographic Office Chart No. 21.) character to the other channels and has been broadened out by the ma rine currents when the submergence was less than the present amount. From the structure the writer concludes that a greater amount of depres- sion occurred here than in the region of the Rosalind banks, or else that the submergence was earlier. The physical character of both these pas- sages is modern, in the region of least orogenic disturbances, and their depths, inferior to those of the fjords (the Yucatan being in juxtaposi- tion to them), show that the outlets of the present basins were not across these ridges. One other fjord must close the list. This has a depth of 7,000 feet within the limits of the lower terrace off the Honduras coast. It is probably a continuation of Segovia river. On the Pacific coast the somewhat scanty soundings prevent exhaust- ive study, but it seems that the shelf depressed off the coast of Centra 7 LAND VALLEYS. hey America to 600 feet or less is much narrower than on the Atlantic side of the Antilles, or is wanting (the basin of Panama gulf, which is like a submerged terrace, is an exception); but deep water often approaches near the land, and in places fjords may be seen. A quite noticeable feature occurs south of the isthmus of Tehuantepec and a similar one south of the gulf of Panama, where broad, deep basins of the Pacific ocean extend landward as if they were once continuations of the Mexi- can gulf and Caribbean sea. This array of data in the geomorphy of the vast and only partly sur- veyed Antillean region, although scanty in proportion to the area, sug- gests physical problems which should no longer be overlooked. Inter- pretations may differ, but the facts are of undoubted significance and seem to the author strongly to indicate vertical oscillations of great am- plitude during the course of development of this and neighboring dis- tricts. CHARACTERISTICS OF THE LOWER REACHES OF THE LAND VALLEYS. As the correlation of the valleys and the submerged canyons will fol- low, the general characteristics of the depressions of the country traversed by the lower reaches of the rivers may be noted, especially as the descrip- tions of their forms are not readily available. After leaving the older formations and entering the less coherent Cretaceous and Tertiary strata, the rivers pass over the gently sloping coastal plains, which may have a width of 200 miles or more and a descent of 400 or 600 feet before reaching the existing coast. But where the plains are not over 250 feet above tide, the rivers occupy broad troughs. Even at 200 miles from the sea the valleys may be from two to four miles wide, and where only partly filled by the deposits of later date the flats are characterized by flood-plains and swampy areas. Farther down their courses the valleys widen and are delimited by bluffs rising perhaps 50 or 100 feet above the streams which touch them at only occasional points. The drainage of the swampy reaches is often retarded by the necessity of the streams crossing durable rocks which have become exposed, owing to the gentle deformation of the surface, during terrestrial undulations, or by a change in the course of the stream. Nearer the sea the form of the valley becomes obscure, owing to the filling with sand or alluvium during the more recent epochs and to modern sedimentation. Here the shallow valleys are apt to be swampy, but limited by hills rising from 25 to 75 feet, more or less modified, owing to interruptions on account of the entrances of great lateral branches. As all the features are low, but on a broad scale, with the depressions from five to ten miles wide, on reach- ing the coast the true characteristics of the valleys are best appreciated 118 J. W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. in leveled sections. Near the coast the broad valleys are commonly deeply buried, even to hundreds of feet. In the case of the Mississippi, at New Orleans, a well to a depth of 900 feet below the surface did not reach the floor of the old valley of erosion. To give a list of examples would include all of the valleys of existing rivers and others now com- pletely buried. These buried valleys are discovered by well borings, and in other cases inferred by the forms of the valleys themseives, whose outlines are not completely obliterated. ANALOGY BETWEEN THE SUBMERGED VALLEYS OR FJORDS AND THE LAND VALLEYS AND CANYONS, WITH INFERENCES AS TO FORMER CONTINENTAL DEVELOPMENT. When the valleys of the southern Appalachian mountains, whether two, four, twenty or forty miles wide, are compared with the sub- merged Antillean depressions the resemblance is complete. Even the embayments into the continental plateaus, characteristic of the expanded mouths of the fjords, are no greater than the present embouchures of many rivers or their flood-plains. The only difficulty in accepting these drowned valleys as those of the former land depressions is the great depth to which they reach. Where the soundings are numerous, as they are in some localities, the submarine contours show the drowned valleys continuously from where the surface coastal currents cease to act to the greatest depths. Again, these flooded valleys terminate in broad embay- ments, which are greatest where several fjords leave the plateau together, just as the divides between the neighboring rivers are gradually reduced beneath the general level of the plateau, owing to the double denudation on both sides of the ridges. The submerged valleys have their tributaries coming from different directions, as is the case with rivers. In some places the fjords have steep walls, while again they are V-shaped or broader valleys, a few miles wide. The direction in relation to the mountains is at every angle, but the prevailing systems are at right angles to the mountain ranges of the land, and accordingly the depressions are not mountain folds. Two or three fjords are parallel to the mountain folds, and may have been deep- ened by the movements of an orogenic nature. The valleys commonly cross coastal plains of undisturbed strata, which largely belong to the later geologic formations. All of the drowned valleys are connections or continuations of the rivers of the continents or islands. The fjords are recognizable for distances from 50 to 250 miles, and in one case for 600 miles.* * See foot-note, p. 140. ANALOGY BETWEEN LAND VALLEYS ‘AND FJORDS. 119 From all of these considerations the writer has been led to conclude that the fjords are land valleys greatly depressed. Once in the progress of our science it was supposed that fissures were formed by plutonic forces and left open. This vision of great open fissures belongs to the past. The drowned valleys diverge in all directions, and the valleys generally are not from orographic folds, except perhaps two or three in part; nor havethe epeirogenic movements defaced their character so as to obscure their resemblance to land forms. What inductions are we to make? Can we deny that these systems are old rivers because they are depressed two miles or more? Must we not accept the physical evidence of the great submergence of the land as here recorded, just as we accept the evidence of the great changes of level registered in the older geologic formations ? The gulf of Mexico appears to have been a plain, with the fjords and embayments reaching nearly to its greatest depths. Onasmaller scale, it resembles the Mississippi valley, or the country between the Appalach- jans and the Rocky mountains. Such being the case, its floor was ele- vated somewhat more than 12,000 feet, and over it drained the Antillean rivers, except the short streams entering the Atlantic basin. It was this precipitous drainage that has removed most of the coastal plains in front of the Windward islands. Caribbean sea was also another basin which apparently was once a plain, as indicated by the deep fjords, although the explorations are less complete than those of the Mexican gulf. The Windward ridge has been sufficiently investigated to show that the drainage was mostly to the west ; but between the Virgin islands and Santa Cruz the natural land valleys appear to have been partially deformed by orogenic movements. The sea of Honduras is unlike the other two Antillean depressions, as it is not basin-like, but composed of two valleys, now of great depth. These are parallel to the mountain folds. While the land valleys run into these channels, as in the other cases, yet their greater depth, reach- ing to 20,000 feet, impresses one as being so excessive, and their occurrence between the mountain ranges of from 7,000 to 9,000 feet, elevated in re- cent times, is so suggestive of orogenicaction that I am inclined to attrib- ute part of the subsidence to an abnormal depression of an orogenic fold, though of such a character as not to obliterate the form of the valley. This hypothesis would remove the necessity of supposing that the Great Antilles stood 20,000 feet higher than at present. Such unequal depres- sion is in accord with continental movements already described, but only part of the enormous sinking of the floor of the sea of Honduras could be assigned to orogenic movement, as is shown by the deep lateral fjords. XVII—Butt. Grou. Soc. Am., Von. 6, 1894. 120 J.W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. Through the physical study, the writer infers that the Antillean con- tinent lately existed somewhat as shown by the drainage, but in the vari- ous oscillations of level Central America was warped upwards and cut off the western drainage, which appears to have extended into the Pacific ocean, as shown by the basins opposite to the gulf of Mexico and Carib- bean sea on the southern side of the isthmuses of Tehuantepec and Panama. While the barrier of Central America was being raised. as illustrated by measured examples already cited on page 108, the Antil- lean basins sank. On the surface of the land a good example of similar movement may be seen in the Jordan-Akabah valley, which has sunk bodily for thousands of feet to 2,000 feet below sealevel, yet without obliterating the topography of the Jordan valley, even where obstructed by the transverse barrier between it and the gulf of Akabah. As to the dates of the deformation of the Antillean continent we shall inquire later. Accepting the foregoing inferences as to the meaning of the submerged valleys and plains, it would appear that the magnitude of the conti- nental elevation varied; that after making allowances for foldings and amplified marginal depressions the northern side of the Mexican gulf has suffered a depression of not less than 8,000 feet and perhaps some- what more; that Yucatan has gone down 12,000 feet; that the Greater Antilles have been depressed 10,000 or 12,000 feet and the southeastern margin of the continent nearly the same amount. While the Caribbean basin has in part become depressed to about 15,000 feet, yet it is hardly likely that any part of the surrounding lands except mountain ridges ever stood at that elevation above their present surfaces. The two Amer- icas were united and the Atlantic currents were deflected eastward. While the Antillean lands of that day were greatly elevated, the plains of the now depressed basins were at no great elevations above the Atlantic, between which and the Pacific some elevated insular masses were appar- ently being slowly pushed up. While the Antillean elevation lasted long enough for canyons to be cut back to the depths given, yet the time was too short to allow of the dissection of the interior of Florida and the border islands by deep canyons. The climate of the elevated continent with the attendant meteoric conditions must have been quite different from those prevailing today. Mip-TERTIARY SUBSIDENCE OF THE REGION OF THE WeEs't INDIES. The earlier terrestrial condition of the Antilles will be passed over after stating that in the Cretaceous period, or before extensive accumu- lations of mechanical deposits were formed, the sediments were mostly DEFORMATION OF THE WEST INDIAN REGION. 1A derived from localities now depressed beneath the sea or buried by later accumulations; so also from the Cretaceous days, or before, to modern times great volcanic activity in one locality or another has been added to the geologic forces of the region of the West Indies. During the earlier part of the Eocene period a portion of the West In- dies was elevated, but this elevation does not seem to have extended to the adjacent continental area. During the later Eocene and most of the Miocene period only a few islands appear to have existed in the seas of the West Indies and Central America, and the accumulations of strata reached extensive proportions. As the Miocene often succeed the HKocene strata without a break, they form a physical unit. In Cuba their united thickness, actually observed, is 1,400 feet, with a faulted structure, which indicates a total development amounting to 2,000 feet. Along Chattahoochee river the Kocene is 1,400 feet thick. At Jacksonville, Florida, the upper Kocene, beneath 400 feet of overlying accumulations, extends to a depth of 1,500 feet without reaching the bottom beds. At Galveston 2,000 feet of upper Miocene (Dall) alone have been revealed ina well. At Savannah Miocene strata have been denuded to a depth of 250 feet below tide in the buried valley, and in southern Florida, be- neath the late Pliocene basin, only four feet of upper Miocene strata re- main (Dall). In Jamaica the Kocene and Miocene strata aggregate 5,000 feet (Sawkins), and there the deposits have been raised to 3,000 feet above the sea.* The Miocene formation of San Domingo is 2,000 feet thick and raised to an altitude of 3,855 feet (Gabb). The greatest elevation of the Miocene in Cuba appears to be 2,300 feet south of the Sierra Maestra.f Similar limestones form the divide of the isthmus of Tehuantepec, and on adjacent hills rise to 1,000 feet,f with the maximum height unknown. At Panama the Miocene strata occur to a height of 500 feet on some hills rising out of the harbor (Maack ),§ and in the neighboring parts of Costa Rica they rise to 3,000 feet (Gabb). Similar strata also form the divides between the valley of, Atrato and the Pacific ocean, with elevations of 763 feet and higher, but the maximum elevation is not given. Thus in the Miocene times, so far as can at present be determined, only a few small islands of Cretaceous, with some plutonic, rocks could have risen above the common surface of the Atlantic and Pacific oceans, and all of the to- pographic and hydrographic features are post-Miocene. In the Miocene period there appears to have been a great subsidence of many portions of the Antillean and continental regions. Along the coast * See reports on Jamaica and San Domingo, cited before. 7 J. P. Kimball, Am. Jour. Se., Dec., 1884. t Report on “Isthmus of Tehuantepec,’ by J.G. Barnard and J. J. Williams, Appleton’s, 1852 (see geological section). 2‘* Isthmus of Darien Ship Canal,” U.S. Navy Department, 1874. 122 J. W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. of the continent diatomaceous and foraminiferal earths occur in the mid- dle Miocene beds. This is true also in Jamaica, and apparently of Cuba, where radiolarian deposits at the eastern end of the island appear to be long to the same date. In Barbadoes Messrs Jukes-Browne and Harri- son* have described great deposits of radiolarian earths. They assign the oceanic deposits provisionally to the Phocene, but do not object to the earlier age, as they had not the data for settling the question, but they established the succession of insular strata and the wonderful amount of subsidence. The related rocks in the island do not form a series for close comparison of the age of the abysmal earths. They lie on the greatly eroded surfaces of what may be Cretaceous deposits (judging from dy- namic conditions) and unconformably underlie limestones of probably the latest Pliocene epoch, if correctly correlated with the rocks of the Greater Antilles. Under these conditions it may not be straining the evidence to place the Barbadian earths in the Miocene system, under which the geomorphic changes of the whole region would be in harmony. Accordingly, in the Miocene period there appears to have been a great subsidence, extending from the West Indies to New Jersey (apparently commencing:a little earher in Barbadoes and later in the north), the be- ginning of the stupendous oscillations that culminated in the continental Antilles and ended with the modern depression of the region of the West Indies. Tot ANTILLEAN CONTINENTAL EXTENSION IN THE PLIOCENE PERIOD. Throughout most of the Pliocene period there was an extensive ele- vation and development of the Antillean region. In part, this eleva- tion may have commenced in the later Miocene; for, according to the paleontologic studies of Mr Robert Etheridge y+ on the fossils of An- tigua (Hocene, according to Jukes-Browne), Anguilla and part of Trin- idad, the upper Miocene rocks are absent, either from not having been deposited or from subsequent denudation. Littoral Miocene beds are also wanting in Barbadoes and probably in other regions. In Cuba, San Domingo, Jamaica and other places, as in Florida, the upper Miocene beds are found. In places the earliest Pliocene beds appear to be also found as a stratigraphic unit with the Miocene in Florida (Dall) ;{ con- sequently the great continental elevation in some regions seems to have commenced at the end of the Miocene and in other places in the early Pliocene period; but the Tertiary seas were being gradually re- stricted, from the earlier Eocene times, along the continental margin. * The Geology of Barbadoes, by A. J. Jukes-Browne and J. B. Harrison, Quar. Jour. Geol. Soe. London, vol. xlvii, 1891, pp. 197-250, and vol. xvili. 1892, pp. 170-226. if See the reports on San Domingo and Jamaica. { Bull. U.S. Geol. Surv., no. 84, cited before. EXTENSION OF THE ANTILLEAN CONTINENT. 123 Turning to the land features, it appears that the more or less upturned Miocene beds were being extensively eroded into broad valleys, with the fjords creeping inland. Even the little valley of the Yumuri of Cuba was excavated to a width of three miles, and the same is true of several valleys in Jamaica, San Domingo and Costa Rica, which are exca- vated out of Miocene limestones and other strata’ The valley of Atrato, in Colombia, which is 40 or 50 miles wide, is more recent than the Miocene period. The elevated valleys, with bases 1,500 feet above the sea, in the Trinidad mountains of Cuba, and similar valleys in Jamaica, up to 3,000 feet, have been elevated much more recently than even the Pliocene erosion which molded their forms to a great extent. Such valleys are being now produced by widening of the rapidly growing canyons. In short, at that time the Antillean mountains were not rela- tively so high above sealevel as now. The Pliocene drainage reduced the valleys to the lowest level, and these were miles in width. It was then that the modern topography was first well established. The dura- tion of the epoch of erosion was long, and the formations which were degraded were those that formed the surface of the country which was largely covered by Miocene deposits. Thus the Matanzas fjord was first entirely excavated to a depth of 1,500 feet before joining the outer fjord. The Pliocene deposits of southern Florida occupy a broad, shallow basin, with the older Miocene formations rising on both sides.* In short, throughout most of the Phocene period the continental elevation con- tinued with the degradation of the surface into canyons extending far inland, but not so far as to carve deep valleys into the interior of the then elevated tablelands which now constitute the coastal plains, except such as are now buried to the depth of a few hundred feet. The geologic development of Central America is yet somewhat hypo- thetical. That the drainage was toward the Pacific is highly probable: if not certain, since the characteristics of the adjacent portions of the ocean bed indicate a continuation of the Gulf and Caribbean valleys and plains; but in the great oscillations of the land from abyssmal depths to continental elevations of 8,000 or 12,000 feet some insular masses doubt- less rose into prominence. Such heights would refer mostly to the region of the Greater Antilles and the adjacent continents, for the Gulf and Caribbean plains must have been low. The former tablelands are in part illustrated by the modern great plateau basin of Mexico and the tablelands of Guatamala, which rise from 6,500 to 8,000 feet above the sea, or by the still higher tablelands of Asia. During the Plocene elevation there was at least one volcano in Jamaica,f and some of the volcanoes of Central America appear to have * Bull. U.S. Geol. Surv., no. 81, “ Neocene Correlation Paper,” by W. H. Dall, map, page 156. + Geology of Jamaica, p. 120. 124 J.Ww.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. ‘then been active (Gabb), as well as several volcanic cones in the Wind- ward group. The surveys of the sea of Honduras are much less complete than those of the gulf of Mexico, but they are sufficient to indicate that a great por- tion of that sea was shrunken to narrow limits, if not entirely drained. Still, of this we have no proof at present, as the regular continuity of the tributary fjords to the greatest depths is not shown by the incomplete soundings so far made. In the oscillations of recent geologic times the deeper portion of the Honduras basin may have remained a sea and formed a retreat for such antique types of life as may be found in the deeper Antillean waters. DROWNING OF THE PLIOCENE LANDS AND BURIAL BENEATH MARINE ACCUMULATIONS. In Cuba, Jamaica and San Domingo, resting upon the upturned edges and denuded surfaces of Miocene and earlier formations, there is a de- posit of soft, earthy, white or creamy limestone, made out of the mechan- ical residue of older limestones, with some small masses of corals and shells. All the observed species in Cuba* are the same as the living ones.t To this formation the writer has given the name of the Matanzas series. Owing to the modern facies of the organic remains, Salterain { has included it in the post-Pliocene, but states that it may be Pliocene. The accumulation occurs somewhat bedded, and has a thickness of about 150 feet. The beds generally he at low angles, dipping (2° to 8°) toward the coast or are nearly horizontal. This chalky limestone is soft and can easily be cut, but soon hardens on exposure. It can be used for build- ing purposes or as road metal. The lower bed, in which there are some limestone pebbles from the older formations, has been seen to rise to nearly 400 feet in altitude. It sometimes forms the barriers in front of the modern bays, and these are then apt to be incised by recently formed canyons, of which the outlet of the harbor of Cienfuegos is an example. In geomorphic position the Matanzas formation corresponds with the Lafayette of Mr W J McGee, and also with some marine deposits of south- ern Florida. Professor A. Heilprin has also found the same formation on the northern plains of Yucatan.§ From all the evidence before the writer, he has placed the deposition of the formation at the close of the Pliocene period, though in fact it may extend somewhat later. During the * Geographical Evolution of Cuba,” by the writer, in preparation. +Dr W. H. Dall and Mr Charles T. Simpson kindly determined the fossils for me. t{‘‘Apuntes para una Descripcion Fisico-Geologica de las Jurisdicciones de Habana,” Madrid» 1880, p. 20. 2‘*Geological Researches in Yucatan,” by A. Heilprin, Proc. Acad. Nat. Sci. Phila. for 1891, pp. 136-158. LAND SUBMERGENCE IN THE PLIOCENE. 125 Matanzas depression, Cuba and the West Indies were reduced to small islands without surface enough to furnish the red residual loams and quartz gravel, such as make up the Lafayette of the northern continent. The Matanzas formation is widespread throughout the Antilles. In San Domingo Gabb * describes this low lying formation as post-Pliocene, on account of the modern aspect of the fossils, and calls it the “ coast formation,” a name somewhat confusing, as it is also given to the modern coral reef formations. Its thickness is about 200 feet. In Jamaica Sawkins* and other geologists describe the “ white lime- stone’ as Miocene, but by some circumstance have tabulated the forma- tion as post-Pliocene. It is a very much disturbed formation 2,000 feet thick and elevated to 8,000 feet above tide, altogether unlike the later deposit in the Antilles. This has led to errors in correlations. In his summary,Sawkins describes the surface of the “ white hmestones” as a “white marl” derived from the limestones; but in the excellent detailed local descriptions in numerous places he shows that the “ white marls” rest unconformably upon the “ white limestones” or older surfaces, the marls having a thickness of not more than 200 feet. In Cuba, where un- conformity or other criteria are not apparent, it is somewhat difficult to distinguish the Matanzas limestone from the older Tertiary rocks from which it is largely derived. From the descriptions and also the map of Mr Sawkins, it is apparent that the older Miocene surfaces were enor- mously eroded before the deposition of the marls, as the latter lie in val- leys hundreds of feet deep and three or five miles wide. The fossils found are not abundant, but mostly belong to living species. A similar so-called post-Pliocene formation with modern fossils has been noted on the isthmus of Panama and on the Atlantic side of Costa Rica respectively by Dr G. A. Maack and Dr W. M. Gabb f lying uncon- formably on Miocene strata. It occurs up to an elevation of at least 150 feet above the sea and constitutes the eroded hills of the low coastal plain. Dr J. Crawford? notes the occurrence of recent oyster-bearing beds in Nicaragua. He informs me that they reach an elevation of about 500 feet and unconformably succeed Miocene strata. In position the Matan- zas limestones are represented on the Atlantic side of the isthmus of Tehuantepec by Mr J. J. Williams in his geologic section. § Most important are the observations of Professor A. Heilprin. where he shows the occurrence of this soft limestone with some extinct fossils over the extensive low plains of northern Yucatan. There, too, the sur- faces are eroded, and the geomorphy is the same as that of the Antilles. * Cited before. 7 “Isthmus of Darien Ship Canal,” cited before, and Gabb’s Costa Rica, cited before. ft Report of the British Association for 1890, p. 812. 2‘“*Isthmus of Tehuantepec,” cited before. 126 J. W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. In Barbadoes there is an extensive capping of “raised reefs or coral rocks,” rising in terraces to 1,100 feet, with a thickness of from 150 to 260 feet, as given by Messrs Jukes-Browne and Harrison.* The contained fossils are modern, except perhaps some of the corals. Although the formation is elevated somewhat higher than the same deposits farther westward, yet the geomorphic position, the fossils, and the magnitude and character of the deposits would lead me to correlate it with the Matanzas formation or the latest Pliocene; still the epeirogenic movement possibly began a little earlier on one side of the basin than the other. In Guadeloupe, Anagade, and several of the northeastern Windward islands fragments of the Matanzas limestone appear to exist, but they have not been separated from the Miocene strata. The nucleus of the Windward mass is Cretaceous or igneous, succeeded by Eocene and Mio- cene strata, most of which has been removed by the stupendous denuda- tion of the region during recent geologic times.t In Trinidad there is no corresponding calcareous formation ; but rest- ing on certain deposits referred to the Miocene and unconformable to it there are the Moruga sands, and possibly some of these beds may be the equivalent of the Matanzas limestone. Turning now to the continent, Dr Dall has mapped a large shallow basin opening southward, containing a few feet of Pliocene beds. In other localities only two or four feet of Miocene deposits have escaped denudation. Geomorphically the upper marls with recent shells occupy the same position as the Matanzas beds. The basin is such as would have been formed during the Pliocene period of erosion, as already de- scribed. If the writer be correct in the interpretation of the Antillean phenom- ena the equivaient of the Matanzas formation is found in the Lafayette formation of Mr W J McGee, with which the writer is familiar, over a wide extent of country. The materials of the continent are essentially red or yellowish loams, sands and water-worn gravels, which last occur adjacent to the old waterways. On the higher lands the thickness is about 20 feet, butin the valleys the writer has seenit 120 feet thick, and in the Mississippi channel itisa much heavieraccumulation. The formation often showed no stratification where the gravelis absent. When present the gravel gen- erally forms the lower part of the deposit. The materials were primarily derived from the residuum of the rock decay, somewhat varied accord- ing to the source, whether it was obtained from the surface remains of * Cited before. +See Transactions of the Royal Academy of Sweden, T. ix, no. 12,1871, where Professor P. T. Cleve gives a summary of the geology of the northeastern islands of the West Indies. From his paper and other information received from unpublished sources, I should expect to find fragments of both the Matanzas and Zapata formations on those islands, although perhaps the materials would not be of the same constitution as elsewhere, LAND SUBMERGENCE IN THE PLIOCENE. 27 the metamorphic rocks, of Paleozoic limestones or of the impurities of Ter- tiary limestones. The old land surfaces furnished an abundance of such material to the exclusion of calcareous organisms, for in the formation no marine life has been found. The Lafayette formation was deposited on the eroded surfaces of all such formations as occur along the coast of the continent of geologic date from the Archean time to the later Mio- cene period. In the West Indies the physical conditions were different from those of the continent, for there were few islands to furnish sedi- ments and so the Lafayette loams were replaced by the Matanzas lime- stones. That the lands in the Antilles would have supplied such mate- rials if they had been more elevated is proved by later events in the geology of that region. Thus ina visit to the West Indies the writer was not prepared for the identification of accumulations so dissimilar, but on the discovery of the key it was found that such differences should have occurred. The Antillean region may be too great an area to bring within the scope of the gentle epeirogenic movements, but beyond the limit of the orographic disturbances the deformation of the earth’s crust over the vast region from New Jersey to Mexico shows undulations in the coastal plain of hardly a thousand feet (from 100 feet, above tide, near Cape Hatteras, to 800 feet in South Carolina, 250 feet in Arkansas and 1,000 feet on the Rio Grande). Only on approaching the vicinity of the isth- mus of Tehuantepec do the undulations become involved in the recent and great mountain movements. ‘The elevations of the Matanzas lime- stones from the Windward islands to Central America seem to be only affected by gentle undulations until reaching the zone of transverse but recent mountain uplifts. Herein he some difficult and unsolved prob- lems. Hxcept in the region of the Pacific barriers and one or two other localities the evidences of moderate terrestrial undulations is markedly shown in the character of the submerged valleys. From the various considerations set forth the conclusion is reached that the Matanzas epoch (about equivalent to the Lafayette) represented a general submergence below the present altitude, not only of the costal plain to. from 100 to 1,000 feet, but that the Antillean lands at the end of the Pliocene period were depressed so that only a few islands remained at altitudes from 100 to 1,100 feet lower than today. But at that time there was also an- other variation in the topography, for the mountains had not their axes so highly elevated above their flanks as they now are, as pointed out on page 125, and as demonstrated by the character of the modern erosion of the recently elevated bases of the mountain valleys. The epeirogenic movements may not have been quite synchronous ; perhaps beginning a little later in the north than farther south and also XVI[I—Bun1, Grou. Soc, Am., Von, 6, 1894, 128 J.W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. 7 ending later in one region than in another, but the general undulations belonged to the same system of changes of level. THE EARLIER PLEISTOCENE ANTILLEAN CONTINENT AND ITS DEGRADATION. After the deposition of the Matanzas limestones and the Lafayette loams, the continent rose to a great elevation, as is recorded in the amount of succeeding erosion. The enormous degradation is one of the physical problems which McGee so strongly emphasizes in his researches in the Lafayette.* He considers it greater than that of the Pliocene elevation. For the Antillean region, the writer is not fully satisfied with his conclusion in this first study, although the filling of many of the old Pliocene valleys was almost entirely removed, and in many places the channels were further enlarged. At present, the writer thinks that the degradation in the pre-Matanzas and in the post-Matanzas epochs was of about the same magnitude, but a longer duration of erosion in the earlier Pliocene period would explain the inferior elevation of that time. In both periods the valleys were excavated so as to leave depressions several miles in width, not merely along the great rivers of the continent, but also along the shorter streams of the West Indies. Thus the Yumuri valley in Cuba was reéxcavated to a width of three miles, and Xagua bay to a greater breadth, for this is only a recently submerged valley. The same is illustrated on the south side of Jamaica (in Vere and West- moreland parishes) and in Haiti. In Costa Rica the effects of this epoch of erosion are seen in the rounded hills rising out of the low plains and in the broad valleys on the western side of the continent. Everywhere the amount of denudation would indicate slopes corresponding to those of the earlier Pliocene elevation. The fjord of Matansas bay, which has a depth of 1,500 feet within its land boundary, is cut through this latest Phocene limestone. While the Pliocene valleys were more or less refilled in the Matanzas epoch, it is certain that they were reopened in the earlier part of the Pleistocene period. It would appear that the present lands of the West Indies and the adjacent parts of the continent stood quite as high, if not higher, than during the Pliocene elevation, so the amount of erosion equaled or exceeded that preceding the Matanzas epoch. At any rate, the fjords are open to the great depths already described. The character of the drowned valleys, involving the Matanzas lime- stones and the Lafayette loams, and their physical relations to the succeed- ing deposits, point to the conclusion that the American continents were united by the Antillean bridge with an altitude as great as that of the * Cited before. WEST INDIAN SUBSIDENCE IN LATER PLEISTOCENE. 129 Pliocene period or greater, or varying from 8,000 to 12,000 feet or more, and, subordinately, that almost all of the drainage flowed into the Pacific ocean. While most of the canyons did incise the frontal margins of the plateaus and receded to great distances in them, yet the elevation did not last long enough for the deep valleys to be completely cut back and leave great depressions in the central portions of what are now the coastal plains. Whether the elevation was great enough to completely drain the sea of Honduras (as the Caribbean sea) cannot be told at present. The great altitude of the Antillean land is no longer a question. The climate of the high lands may have been more or less arid in some local- ities, like the plateau-valleys of modern Mexico and Guatemala, or even parts of San Domingo. SUBSIDENCE OF THE West INDIES IN THE LATER PLEISTOCENE PERIOD. The subsidence which followed the earlier Pleistocene elevation is marked by some terraces rising in Cuba to an elevation of 1,000 feet and lower altitudes. This terrace problem needs careful revision before the Pleistocene and later made shorelines can be distinguished over widely separated areas. But both subsidences affected and depressed all of the ereater Antilles, Central America, and the coastal margins of the conti- nent from about 25 to 500 or 700 feet lower than now. ‘This depression ereatly reduced the size of the larger West Indies and Central America ; it also made the coast of the northern continent recede 100 or 150 miles, and drowned most of Florida. The accumulations in Cuba and the other ereater Antilles, and also in parts of Central America, consisted of red- dish loams and gravels (in the vicinity of the streams), which are now seen at an elevation of 200 feet or moreinsome rezions. ‘To this forma- tion the writer has given the name of the Zapata in his forthcoming paper on the geographical evolution of Cuba. The Zapata occurs in Jamaica, San Domingo, apparently in Trinidad and widely over Central America.* In Yucatan it appears that the upper post-Pliocene marls of Heilprin belong to this epoch. The sediments were principally derived from the residual loams and gravels left by the solution of the Miocene and other limestones, as there were then suffici- ent land surfaces to furnish such materials. Turning now to the continent, the Zapata formation is of about the same age as -McGee’s older Columbia series, which covers 150,000 square miles of the coastal plain. In Carolina it reaches an altitude of 650 feet, in southern Alabama only 25 feet and along the Rio Grande from 100 to 200 feet above the sea. Physically, it is of the same character as the La- * The authorities here mentioned have been cited before. 130 J.Ww.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. fayette formation, which supplied materials for the newer deposit, which is like the Zapata. The fossiliferous sands and the coquino found in the wells of Saint Augustine, and some marls in southern Florida, are prob- ably of Columbia age, distantly removed from the source of the non- fossiliferous red loams, which were rapidly laid down nearer the sources of mechanical materials. In the Antilles geologists have not hitherto (Geographical Evolution of Cuba) differentiated the Zapata formation. Gabb includes it with his post-Pliocene “ coast limestone” of San Domingo. In Jamaica it is sim- ply called the “alluvium” or the “older alluvium,” although in both islands itis from 200 to 300 feet thick. Its position and specialization on the continent were largely the result of the classic labors of Mr. W J McGee, who has surveyed it from New Jersey to Mexico. The geologic forces acting on this widespread formation have progressed quietly over an enormous area of the continent and the Greater Antilles, and in the Lesser Antilles it may, perhaps, be found represented by limestones in places and by clastic deposits in others. The absence of great elevation (probably nowhere exceeding 700 feet or 800 feet) further convinces the writer that the continental oscillations were becoming moderately uni- form over a very great area, and this reduces the geology to simplicity ; but in the mid-Pleistocene subsidence which lowered the Antillean continent, the ridge of Central America became prominent—an undu- lation of less than one degree, including orogenic movements, being sufficient. Whether the Zapata formation extends over the divides in Central America or not is an important question, for that settles the date of the final separation of the Pacific waters from the Antillean seas, but the eravels filling the old valleys in Nicaragua, according to Mr J. Crawford, occur up to altitudes of 500 feet, and it is not improbable that they are of the same age as the Zapata deposits. Under any circumstances, the con- tinent was lost from the date of the Zapata subsidence, which was in the mid-Pleistocene epoch. REELEVATION OF THE LANDS AT THE CLOSE OF THE PLEISTOCENE PERIOD. From the Zapata subsidence the Antilles rose from 150 to 200 feet above the modern altitude. Then the streams cut out canyons to the depths named, and made many new outlets to the bays, excavated in part out of the Matanzas limestones, but closed by the Zapata loams and gravels. This elevation somewhat enlarged the land area, and increased it to about the proportions shown on the shaded portions of the accompanying map. The Bahamas formed two or three large islands, but neither these nor PLEISTOCENE REELEVATION OF THE LANDS. ASHE the Greater Antilles had any continental connections. The post-Zapata erosion did not exceed from one-fifteenth to one-fiftieth that of the earlier epoch of Pleistocene elevation. Minor depressions of 100 or 200 feet, or perhaps more in places, fol- lowed the post-Zapata elevation, as recorded by the modern terraces, which have not been differentiated from those of earlier date except at one or two places. The elevation of the terraces was not uniform, put accompanied by a slight deformation of the beaches. The recent movements have been slight and very uniform, as shown by the extensive submerged plains, now constituting banks, by the slight elevation of the modern reefs to a height of 10 or 25 feet, more or less, and the non-deformation of the drowned valleys. In some places the coasts appear to be sinking, as the eastern side of Florida and the Bahamas, while other localities appear to be rising, as the southern side of Cuba. Figure 7.—Oscillations of the Antillean Continent. Horizontal line represents sealevel; dotted iine, the oscillations; Pe = pre-Cretaceous ; C= Cre- taceous; H = Eocene; M= Miocene (variable in different localities); P= Pliocene; Pl = Pleisto- eene; M = modern. The changes of level in the West Indies are graphically represented in figure 7. This hypothetical diagram could not be drawn to accurate scale, either for length of time, represented by the horizontal measure- ment or for the’changes of level, shown in the vertical scale, but in a rude way it illustrates the oscillations of the Greater Antilles in the more -recent geologic times. THE SEPARATION OF THE ANTILLEAN BASINS FROM THE PACIFIC OCEAN AND THEIR CONNECTION WITH THE ATLANTIC. It has been shown that the drowned valleys are newer than the Miocene. The partial filling of the Pliocene channels was accomplished at the end of that period, and these accumulations were subsequently removed by the denudation of the earlier Pleistocene period. The fjords trend west- ward and are traceable to near the floor of the Antillean seas, leaving the inference that those basins were low lands extending to the Pacific side of Central America; but it is to be remembered that this inference is tenta- tive only, and that even if later researches show that the drainage did 1382 J.W.SPENCER—RECONSTRUCTION OF ANTILLEAN CONTINENT. not cross the line of Central America the conclusion as to Antillean oscillations willremain. I=fthis inference be true, then the modern islands of the West Indies formed an elevated plateau bridge between the two Americas during the two epochs of elevation, namely, in the Pliocene and in the Pleistocene periods. This conclusion is supported by all the geomorphic structure except perhaps the lower depths of the sea of Honduras, lying between the mountain folds of Cuba, Haiti, and Jamaica, These mountains rise to from 7,000 to 9,000 feet above the modern sea- level; also on the west this basin is bounded by plateaus with ridges and peaks rising to from 6,000 to 14,000 feet above tide, with valleys as low as 2,956 feet (railway survey). The higher peaks are mostly modern or late volcanic, but there are older crystalline rocks, which have been elevated to 12,000 feet since the deposits of Miocene age, which have been forced up by the granite masses to an elevation of 3,000 feet and now form the oldest sedimentary accumulations in Costa Rica.* Dr Gabb concluded that the Miocene formations extended several thousand feet higher than now on the flanks of the granite masses, but have since been degraded. These observations of the deceased geologist are highly suggestive, as his other work has been. It seems to be the key to the orogenic obstructions crossing the Antillean basins, which have been also partly modified by the epeirogenic movements. The latter movements per se are not rudely deformatory, but along with them there have been some modern faults and tilting of even the later strata, yet there has not been enough disturbance to obliterate the geomorphic features. desea Se ey SN Ay SI se 184 PONG Sec nae Eh OtSey Ie ors 4 ria is Ne ais tioeigncela alana Paenr'n Ss wee ards 184 SILMCHOUSICe RCO AON Sas ace Geis seals Ao us CLR nee lio dhe ad glkA 185 SLAG DIE: CTUMTISNS LEMS A Pi AA aes 4 0 i A hei nS 185 isl Omer C OMA ONTET AGC 2 oles ol seas oe So Sas dare ern ns bee ees | male da wale os 186 XXIV—Butt. Geox. Soc, Am., Von. 6, 1894, (167) 168 HALL AND SARDESON—THE MAGNESIAN SERIES. Page The Senesis: OF TMewerles vo nares tae het eee ai sites 2 eee 187 ‘Lhe ‘sandstones 3.05 Waited ey ts Se eee ee Stee alone Whe shales... 5 ga fess o.d steele Ph tens slain 56 Mic hele esac ee et 188 The dolomities ; 2: 64.0 basen be ia sss cea eee eee 189 Historie oublime sss (0 sce «2 ule ne Cee ee 20 hee 189 Sediment-buildimg ‘2... pr Pt o.oo. inh e a ee ne 191 Contents of seawater... 2. ci0. . Gee os so ctessiels ul press oie 191 Composition of modern sea-deposits:.....: 00. 22. sa. 192 Composition of spring and other witers............+-sho. eee 194 Analyses of travertine and impure corali2.\. 2.24. 22+ 4ee eee 194 Geographic:considerations.........:..2:05-- ess. 6: er 195 Continental movements. .. occ: s.dge.seee seas et. eee 195 Removal of calcium carbonate and its effects .......:...... Ys. neueee 196 Basis:of the discussion . cciicies os. ects as os eee oe ee 197 SUMIMALY sooo ee. as caine ition e heels tole alee bee Mee ce ROE ee ee 197 GEOGRAPHIC POSITION OF THE SERIES. The Magnesian series, which it is proposed to discuss in this paper, consists of several alternating beds of dolomite, dolomitic shale and sand- stone. The series lies between the so-called Potsdam sandstone below and the Saint Peter sandstone above. It occurs in widely distributed localities throughout southern Wisconsin, southeastern Minnesota and northeastern lowa. Nowhere in the states named has either one of the delimiting formations been found absent, save in eastern Wisconsin, where, at a single locality, Chamberlin found an arch of the billowy lower Magnesian rising into actual contact with the Trenton limestone and thinning out the Saint Peter sandstone to zero.* Hisroric RESUME. A résumé of earlier investigations of the extent and contents of the Magnesian series of Minnesota has already been given by the writers. To that résumé these notes can be added: In Wisconsin the geologists of the last geological survey very carefully explored the lower Magnesian beds as they occur in that state.[ They, however, relegated the Mendota limestone (Saint Lawrence) and the Madison sandstone (Jordan sandstone) to the underlying Potsdam. The lithologic and structural characters of the series are very fully discussed * Geology of Wisconsin, vol. ii, 1877, p. 272. + Paleozoic Formations of Southeastern Minnesota: C. W. Hall and F. W. Sardeson, Bull. Geol. Soe. Am., vol. 3, pp. 331-368 and plates 10-12. {See the Geology of Wisconsin, vol. i, 1883, pp. 138-144; vol. ii, 1877, pp. 268-285, 547-555, 577-607, 671-675; vol. iii, 1880, pp. 397, 398; vol. iv, 1882, pp. 64-81, 123-129, 194-204, 248, 249, 511-518. HISTORIC RESUME. 169 on the various pages cited. Paleontologic notes also found there state that ‘“‘some seaweeds, a few mollusks, an occasional fragment of a trilo- bite and a few obscure forms make up the meager list of fossils” which could with certainty be referred to this, the Mendota, limestone.* In Iowa the lower Magnesian limestone has been noted in the north- eastern portion of the State. It is most conspicuous along the upper Iowa river and in the valleys of Paint creek and Yellow river.t The thickest beds were 250 feet; the variations in texture and color were re- ported as considerable; the chemical composition varied but httle from that of a pure dolomite; its brecciated and concretionary character was noted as a principal feature, and no fossils were found in the formation within the limits of Lowa. Twelve years later, however, Dr C. A. White stated that a few traces of the stems of crinoids had been found near McGregor. They were so fragmentary and indistinct that no identifica- tion of them could be made; also some traces of possible fucoids were found.t Recently W J McGee has very pointedly discussed the nomenclature of this series for lowa § and for the broader northwest. || In the past year Professor Calvin, state geologist of Iowa, has noted the discovery of a fauna which leaves little doubt of the exact equivalence of the lower Magnesian limestone of Iowa and the Calciferous series of northeastern New York.4j During the last two seasons, having gone over the ground in Minne- sota more thoroughly and critically and having extended their studies into the adjoining states of Wisconsin and Iowa, the authors desire to discuss certain phases of the Magnesian series which were barely touched upon in their former paper. Those phases are partly stratigraphic and paleontologic and partly lithologic and genetic. MEMBERS OF THE SERIES. ORDER OF OCCURRENCE AND THICKNESS. The Magnesian series consists of five formations, namely, three doio- mites and dolomitic shales and two silicious sandstones. Enumerated in ascending order, they are: 1, the Saint Lawrence dolomites and shales ; 2, the Jordan sandstone; 3, the Oneota dolomite; 4, the New Richmond *Tbid., vol. i, p. 141. t+ Geology of Iowa, James Hall, state geologist, vol. i, 1858, part 1, p. 332. t Geology of Iowa, Charles A. White, state geologist, vol. i, 1870, p. 174. 2W J McGee: The Pleistocene History of Northeastern Iowa, Eleventh Ann. Rep. U.S. Geol. Survey, 1889-90, pp. 187-577. || Bull. Geol. Soe. Am., vol. 3, 1892, p. 464. { American Geologist, Minneapolis, vol, x, 1892, pp. 144-148. ie) HALL AND SARDESON—THE MAGNESIAN SERIES. sandstone; 5, the Shakopee dolomite. Of these five formations the first three are considerable in vertical extent, reaching a thickness of 100 to 200 feet. Next above them the New Richmond sandstone is inconsid- erable in thickness, seldom appearing more than a few feet and nowhere more than 20 feet. Lastly, the Shakopee, paleontologically important, since it carries a unique and well defined fauna, seldom reaches a thick- ness of 50 feet. The following table shows the position of the series, with the maxi- mum thickness of each division as determined for the states under consideration : Ordovician (Lower Silurian).... Saint Peter sandstone. Faunal break. ( (?) Shakopee dolomite, 65 feet. | a Faunal break. ie ( wn | New Richmond sandstone, 20 feet. s S| 2 é 4 Wiper Camarines aise 4 Oneota dolomite, 175 feet. Ss | ae Jordan sandstone, 200 feet. = Saint Lawrence dolomite and sandy { | shale, 213 feet. Faunal break. Middle Cambrian.............. Dresbach sandstone. LITHOLOGIC CHARACTERISTICS. The considerations for grouping the five formations named under one term, “the Magnesian series,” are partly hthologic and partly paleonto- logic. The lithologic are apparent when we consider the dolomitic com- position of formations one, three and five; their uniform color, texture and structure; the secondary nature of their crystalline habit; the pres- ence of intercalated odlite in them all; the regular occurrence of brec- ciated phases in each; and, finally, the entire absence of any known physical character by which the geologist could distinguish the special traits of each at any of its outcrops throughout the entire field studied. The sandstones, formations two and four, are equally identical with FAUNA OF THE SERIES. 1h each other and bear similar relations to the dolomites with which they are interbedded. PALEONTOLOGIC CHARACTERISTICS. The paleontologic characteristics may be summarized as follows: In the Magnesian series there are two faunas not united by a single species common to both; neither is the lower of these faunas united to the preceding nor the upper to the succeeding fauna by any more of a bond. ‘This observation is the result of a comparison of all the brachio- poda, gasteropoda and cephalopoda at hand. It is believed that the same result will come from a comparison of the species of trilobita when the obstruction in identifying and locating geologically the large number of described species in this class of fossils has been overcome. Between the Obolella polita horizon, which, according to Charles D. Walcott, belongs to the Middle Cambrian,* and the next or Dicellocephalus minnesotensis (Saint Lawrence) beds of the Upper Cambrian occurs a very marked faunal change. This latter—that is, the Saint Lawrence (Mendota) formation—contains besides Dicellocephalus minnesotensis, Owen, Lingula aurora, Hall; L. mosia, Hall; L. winona, Hall; Orthis (Billingsella) pepina, Hall. Of these last named species a variety of the first, viz., Iingula stoneana, Whitfield, is found in the Jordan sandstone, and a form not yet distinguished from LZ. mosia, Hall, has been found in the Oneota (Shak- opee B) formation. Orthis pepina is found in the Saint Lawrence at the typical locality, Saint Lawrence, Minnesota, and is common to the Saint Lawrence, Jordan and Oneota formations. Again, Raphistoma minnesotensis, Owen, and Murchisonia, n. sp., are found in the Jordan sandstone, and are also very common in the Oneota. Mollusca have not been found in the Saint Lawrence (Mendota), unless those from Barraboo, Wisconsin, described by R. P. Whitfield,t belong here. Meloptoma barabuensis, Whitfield, has been found in the Jordan sandstone at Osceola, Wisconsin. In the New Richmond (Elevator B) sandstone there are no fossils known to the writers, but stratigraphically it seems to belong to the Oneota (Shakopee B). In the Shakopee (Shakopee A) several species of mollusca are widely distributed geographically, since they have been found at Burkharts Mills and Argyle, Wisconsin; Cannon Falls, Utica, Shakopee and other places in Minnesota. They are specifically distinct from the fauna of the Oneota below, and they do not at all coincide specifically with any of the score and more of species from the Saint Peter sandstone above. * Am. Jour. Sci., 3d series, vol. 44, 1892, p. 56. { Geology of Wisconsin, vol. iv, 1882, pp. 194-199. 7 2 HALL AND SARDESON—THE MAGNESIAN SERIES. A correlation of the Shakopee (Shakopee A) with the upper Calciferous of New York and the Oneota (Shakopee B) with the lower Calciferous seems to encounter few objections, although there is no present intention to set forth a decided opinion. Thus it will be seen that the Magnesian series as understood by the authors includes all of the Upper Cambrian in the northwest and a part of the Lower Silurian (Ordovician), provided the Shakopee (Shakopee A) be correlated with the upper Calciferous and both referred to the Lower Silurian.* Tue Saint LAWRENCE DOLOMITES AND SHALES. LOCALITIES. In Minnesota many exposures occur between Redwing and Lake City ; at Rollingstone creek ; the railway cut above Stockton ; at Winona, Dres- bach, Hokah, Stillwater, section 28 Saint Lawrence township; Jordan, section 30 Blakely township, and at Judson. In Iowa they are to be found at several quarries around Lansing. In Wisconsin they occur at Spring Green, Lone Rock, Lodi, McBride’s point lake Mendota, Hudson and vicinity, Trempealeau, La Crosse and Osceola. At Saint Lawrence, Minnesota, is found the type exposure, the rock in view representing the lower half of the formation. The Mendota lime- stone at the type locality described by Irving} represents the upper portion of the same formation. Were a division advisable into two parts the names Saint Lawrence and Mendota would not be synonyms, but would stand for different beds in what is now considered a single formation. SPECIAL FEATURES. At section 28 Saint Lawrence, Scott county, Minnesota, about one mile south of Saint Lawrence landing, on the Minnesota river, there is exposed in one of the quarries six feet of a buff colored dolomite. Irreg- ular laminee of green shaly material are scattered through the entire rock exposed. the plane of contact; ice much veined. GABLE GLACIER, INTRUSION OF EARTHY MATERIAL. 207 to disclose the intimate nature of their mass. Definite planes of parting were developed between some of the layers. This was often true even when the layers were not separated by any earthy filament, the ice on both sides being white and pure. The plane of parting between the layers was often slightly gaping at the surface ; sometimes the two layers seemed to be peeling apart. I found it easy by a moderate stroke of the spike of my alpenstock to split blocks along these partings. The sepa- rated blocks presented smooth surfaces, which seemed to leave no ques- tion of their analogy to slickensides. The layers on either side were made up of coarse granules of ice, intimately interlocked, so that an attempt to cleave the mass at other points resulted in a fracture of the most ragged and irregular sort. But the best evidence of the verity of shearing between the ice-plates hes in the intrusion of the earthy material itself. Iwas fortunate enough, unless I misinterpret, to observe the actual process of intrusion. The best illustration was found on the north side of a short lobe of the great ice-cap designated the Gable glacier. Just back of the point of observa- tion there was a large embossment of rock, which expressed itself at the surface of the ice by a beautiful halfdome, like the Halfdome of the Yo- semite. The other half of the dome was cut away, revealing the opera- tions at the base within. Here it was observed that trains of debris, apparently rubbed from the surface of the embossment, were being car- ried out almost horizontally into the ice in its lee. Some of these were short, while others extended several rods into the ice. They were some- what inclined downward, but the slope of the glacier being greater, they passed out into the body instead of following the base of theice. At one point the overthrust reached such a degree as to carry the earthy layers obliquely almost across the thickness of the glacier, producing a pronounced unconformity. The illustrations will show these phenom- ena with an accuracy and vividness quite beyond the power of a verbal description. (Figures 9 and 10, plate 7.) On the East Branch glacier a similar phenomenon was observed below a cataract of the ordinary type. Here tongues of debris, having their origin in the bowlder-clay below the glacier, were seen to reach out into the basal portion of the ice as though they were being introduced into it by the differential movement of the layers upon each other. The mode of operation seems to be this: When the ice is forced over a prominence it settles down a little in its lee, and is then protected somewhat from the thrust of the ice behind ; the next ice that passes over, being prevented by the former portion from settling down at once, is thrust forward over it. To some extent this is accomplished by the bending and doubling of the layers and to some extent by distinct shearing. At length, however, XXIX—Butt. Geot, Soc. Am., Vor, 6, 1894. % 208 T. C. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. the first layer is compelled by the general friction to move somewhat for- ward, and in time to join the common moving mass, carrying the over- thrust layer of debris between it and the ice-layer above. The way is then opened for a repetition of the process. This picture of the behavior of the ice is quite radically different from that entertained by the viscous hypothesis, in which the ice is supposed to flow down the lee side of a prominence, as if it were liquid. The motive power here seems not so much gravitation pulling a fluent body forward as the thrust of a rigid body by a force in the rear. | Behavior of the Ice in passing over low Prominences.—Several excellent opportunities for observing the behavior of ice in passing over low em- bossments were offered. From the front of the embossment there origi- nated laminze which extended backward with a graceful, arching curve, much like the profile of adrumlin. A portion of the ice remained be- tween these curving laminations and the upper and rear portion of the embossment. After reaching a point in the rear of the embossment, the lamine curved downward with increasing rapidity until well in the lee, when they turned about at a more or less sharp curve, or even angle, and ran backward to some point not far in the rear of the embossment, where they ended. The higher laminz made the longest curves and had the sharpest angles in the lee of the embossment. It appears obvious that the ice in the lee of the embossment moved more slowly than that above ; hence the doubling of the laminz upon themselves. It appeared upon close inspection that some of the inthrust layers described above consist in reality of very sharply reduplicated lamine. It seems, therefore, that this phenomenon grades insensibly into the preceding. A study of lamine: not associated with embossments showed many signs of doub- ling upon themselves in a similar way. It appears, then, that there is a' gradation from lamine that simply suffered doubling up to layers that obviously sheared upon each other and produced manifest uncon- formity by pronounced overthrust. (Figures 11 and 12, plate 8; also figure 9, plate 7.) Development of blue Bands.—Some of the laminz observed to originate on the brow of embossments of rock were simply blue bands. They were even seen on bowlders underlying the ice. So far as observed, the blue bands started at some little projection or rugosity in the brow of the embossment. From this point they extended rearward, usually curving a little upward and free from the embossment, following a drumloidal curve until they had passed its lee, when they turned downward and sometimes returned as described above. Now, it is interesting to note that within the curved loop in the lee of the embossment I observed in one instance several nearly vertical blue bands, standing parallel to each BULL. GEOL. SOC. AM. VOE. 6, 1894, PL. 8. Figure 11.—Porvrion or East SIDE oF FAn GLACIER. Showing behavior of the ice in passing over a low embossment of rock and drift. This figure shows only the upper portion, the next only the lower portion; the central part, of about equal length, not shown. Motion from right to left. Ficure 12.—Lower Porrion oF ABOVE VIEW OF FAN GLACIER. Showing the curving down and bending back of the lamine in the lee of the embossment. The prominent dark line in the center turns back with a sharp curve a short distance beyond the limits of the view, and is apparently continuous with one of the bands shown very obscurely in the lower left-hand portion of the picture. FAN GLACIER, VD att Pao BLUE BANDS. | 209 other and stretching part way across the space embraced in the loop. Here it would seem that the blue bands are produced by the exceptional pressure of the ice in moving over rugosities on the brow of the emboss- ment, and that their position in the ice is parallel to the ice-movement, while at the same time blue bands may also be developed nearly at right angles, after the analogy of slaty cleavage. | It has already been remarked that the most solid ice was usually observed in immediate association with the lamine of earthy matter. The inference is, therefore, that the agencies which introduced the earthy material by the same act developed solid ice. Independently of either of these forms, it appeared to be beyond serious question that solidified layers of ice were developed out of the crusts of the original snow, and hence that a variety of the bands is a direct derivative of the original stratification. In so far as the general shearing of the strata upon each other takes place independently of the special process by which earth was introduced, to that extent, I judge, the faces which moved over each other developed greater solidity than the adjacent parts and approached the more perfect ice of the blue bands. I am not sure that these observa- tions traverse in any serious way the current doctrine, which we owe to Tyndall, that the blue bands are chiefly the product of pressure in con- stricted portions of the glaciers or in the descent of cataracts, but they do suggest that this doctrine needs limitation and qualification. | Summary.—In a word of summary, therefore, it would appear that stratification originated in the inequalities of deposition, emphasized by — intercurrent winds, rains and surface meltings; that the incipient strati- fication may have been intensified by the ordinary processes of consolida- tion ; that shearing of the strata upon each other still further emphasized the stratification and developed new horizons under favorable conditions ; that basal inequalities introduced new planes of stratification, accom- panied by earthy debris, and that this process extended itself so far as even to form very minute lamine. Discussion OF CAUSES OF MOVEMENTS OF GLACIERS. Individuality of Ice-layers—There is involved in the foregoing concep- tions the idea of an ice-layer acting as a unit of movement, or at least individuality of movement in the layer is recognized, an idea that, if cor- rectly entertained, is one of some importance, I think, in the physics of elaciers. This view involves the idea of rigidity rather than viscosity. It will not have escaped attention that the explanation heretofore given of the introduction of earthy material into the ice-layers involves the idea of thrust rather than pull. The picture is not that of gravitation 210 T. C. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. pulling a thick, stiff liquid down the lee side of an embossment, but of a rigid body thrusting itself over its crest. It is not easy to escape the influence of these observations if we push inquiry back to the cause of movement. Competency to thrust and _measurable ability to individualize itself in layers seem to be requisites. A general force might perhaps so individualize itself, but the phenomena naturally lead us to seek an agency acting within the layers. The limits of this address will not permit me to enter far upon the mooted question of the cause of glacial motion, but the hypothesis that has come to be dominant as the result of the summer’s observations may be briefly indi- cated. Granulation.—Back of the stratification of ice lies the phenomenon of granulation. A glacier starts with snow-crystals or snow-pellets; thence there is a growth into shot-hke granules, and thence into larger and larger accretions. Drygalski places the limit at the size of walnuts. So far as macroscopic study goes, this progressive growth of granules con- stitutes the most essential change through which the ice passes. This invites the inquiry whether the essence of glacial movement does not lie in the changes which the granules undergo. If for a moment we enter- tain the quite erroneous supposition that all the granules of a given horizon grew from the smallest to the largest dimensions, it would appear that the expansion of the layer and the movement at its free end would be very great. Every doubling of the diameters of the granules would push the foot of the layer forward a distance equal to the whole length of the glacier. The western slope of the Greenlandic ice-field in the northern tract is probably 500 miles. Three or four doublings of the constituent granules would push its foot, if unmelted, over into Alaska ; but it is obvious that each original grain does not develop into a larger one—some are sacrificed for the growth of others. The hypothetical case is introduced to emphasize this and to illustrate the possibilities of motion involved in changes of the constituent granules. It would be exceedingly helpful if we knew the laws which determine the destruction of some granules and the growth of others. The process of progressive granulation obviously involves the melting of some particles and the freez- ing of the water in new relations. The vital question is: At what points do melting and freezing respectively take place and what are the results? We owe to James Thompson the law that pressure lowers the melting point of ice. Whatever incompetency this may have as a sole agency, it may be an extremely efficient factor in determining the precise points at which melting will take place when both heat-energy and differential pressure are present. I suppose the converse of Thompson’s law also holds true. We owe to Faraday and Tyndall the observation that ice GRANULATION. Atal melted under pressure promptly freezes again when free from pressure. We owe to the same investigators the law that freezing is facilitated by the presence of frozen surfaces in close juxtaposition. Weowe to Tyndall the doctrine that isolated particles or points of ice melt more freely than others from lack of support on either hand. Here, therefore, is a group of agencies which favor melting under certain conditions and freezing under other conditions. Now, all of these conditions affect the individual eranule as it occurs in the mass ofa glacier. It has its points of contact and pressure, its points of free surface, and its capillary interspaces. It is subject to pressures, torsions and tensions, according to the stresses imposed upon it by neighboring granules. It is always under the influ- ence of gravity acting directly upon it, and also indirectly through sur- rounding granules. The combined effect is a resultant pressure urging motion down the slope, but with every yielding of the granules, by melt- ing or otherwise, there is a new adjustment_of pressures, torsions and tensions, and hence new susceptibilities to melting and freezing. Now these being the conditions of the granules, it seems only necessary that there pass over them an agency capable of acting upon the different sus- ceptibilities of their different parts to produce loss here and gain there, and hence to determine the growth of some parts of each granule and the decadence of other parts. In other words, a granule may continually change its form by partial melting and freezing, by loss in one part and gain in another, and through this may either move itself or permit motion in its neighboring granules, or both. Now, every warm day sends down into the glacier a wave of heat- energy. This enters the upper surface as sensible temperature, but for - the most part it is soon changed to potential heat-energy in the form of melted ice. We should not fail to see that the sheet of melted ice that creeps down between the granules of the glacier as the result of a day’s sun-action is as truly a wave of heat-energy as if it remained in the form of sensible temperature. With what freedom the day’s heat is conveyed below by the melted product is not accurately known, but there is good evidence that it is large. Upon the Igloodahomyne glacier we observed “ at midday that the dust-wells were covered with thin films of ice, from which the water below had shrunk away to an average distance of per- haps two inches. The suggestion was that this was the amount of absorp- tion of water which had taken place since the freezing of the film during the preceding night, or, in other words, the absorption of perhaps twelve hours. Circumstances did not permit the careful watching of an indi- vidual well, and this inference was not verified, but it is certain that wells of the largest sizes become entirely emptied of their water within a few days after cold weather cuts off their supply. The moisture which, 212 T. C. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. according to the testimony of all observers, pervades the interior and basal portions of glaciers, has, with little doubt, mainly descended from above. We seem, therefore, altogether safe in repeating that every warm day sends down into the glacier a wave of heat-energy, sensible or potential, and that every night sends after it a wave of reverse nature. These waves follow each other indefinitely, until by intercurrent agencies they become vanishing quantities. Hach season sends through the mass a greater and more complex wave. The problem, therefore, in simplified form, postu- lates a mass of ice-granules predisposed to melt at certain points and to freeze or to promote freezing at others, acted upon by the ever present but differential force of gravity and swept by successive waves of heat- energy competent to cause melting where predisposition to melting exists and to cause growth by freezing where predisposition to freezing exists. Out of this it would seem that localized freezings and thawings, growths and decadences, innumerable and constantly changing, must result, and with them motion of the granules themselves and of the common mass. This statement lacks very much in completeness and qualification, and I can only ask you to accept it as indicating the line of thought to which the observations of the summer have led. If the truth lies along this line, it is obvious that these evolutions would proceed with different rapidity in different portions, and that they might affect an individual layer in a degree different from its neighbor layer, or they might affect the common mass to a nearly equal degree, and that therefore differential movements, alike with common move- ments, would be possible under suitable conditions, and that gravity would control the whole mass much as if it were a liquid. Viscosity.*—My observations seem to be adverse to anything which can be properly termed viscous fluency. On two or three of the glaciers it was observed that the surface rises in the direction of the movement of the ice, so that the surface streams flow backward. Possibly this may be explained on the basis of a viscous flowage of the mass, but it seems much more consonant with the view that the ice-mass was pushed for- ward by its own internal molecular changes, and that it rode up over the inequalities of its bottom as any flexible but relatively rigid sheet would do. *The term “ viscosity ’’ unfortunately has two senses which are nearly contradictory. Both are derived from the original use of ‘‘viscous”’ to signify a sticky, gelatinous, tenacious, semifluid substance, such as the exudation or extract from the sap of the genus viscum. In one ease atten- tion is fastened on the plianey or semifluidity ; in the other, on the adhesiveness or tenacity. In the first case viscosity becomes opposed to rigidity and implies an element of fluidity; in the other it only needs to be indefinitely increased to become identical with rigidity, infinite viscosity being perfect rigidity. The term is commonly used in glacial discussions to signify a degree of fluidity, while in physical investigations it more commonly means a degree of tenacity. OBJECTIONS TO THEORY OF VISCOSITY. 213 The extreme fragility of the ice is difficult to harmonize with the idea _ of viscosity. It was noticeable that whenever the ice passed over an undulation of even moderate dimensions it was abundantly crevassed. The movement of the ice in most such instances was obviously exceed- ingly slow, so that the tension brought to bear upon the surface by the small curve was relatively slight and came into action with exceeding slowness. If the property of stretching were possessed in any but the very slightest degree it would seem that crevassing would be avoided. This objection, which was long since forcefully urged by Tyndall, be- comes intensified when applied to the broad, slowly moving ice-sheets of the far north. : Lieutenant Peary called my attention to a glacier on the south side of Inglefield gulf which breaks entirely in two in passing a steep descent, and reunites below and moves on. Similar phenomena are well known, but they become more emphatic in this northern region. I saw no indication that bowlders descend through the ice as heavy substances descend through viscous bodies. As already remarked, the laminee on approaching a bowlder usually divide and a part curves under and a part curves over it. Nowhere was seen any indication that the bowlders had carried the lamine down, as the superior specific gravity of the bowlder might be expected to do in a viscous body. Everywhere the aspect of the ice was that of rigidity rather than viscous fluency. The rigidity, to be sure, did not prevent-contortions and fold- ings of the laminations, such as take place in crystalline rocks, but fault- ing and vein structures also occur, and there seems no more occasion to assume viscosity in the one case than in the other. Even if a certain measure of viscosity be admitted, it does not follow that viscosity was an essential agency of motion. There is a theoretic objection to the assumption of viscous flowage in the very fact of crystallization itself. The property of viscous flowage rests upon the relative indifference of a particle as to its special point of adhesion to its neighbor particle. The property of crystallization rests upon the strongest preferences respecting such relationship. Particles of ‘ water in their fluent condition lie against and cohere to each other in- differently. When they take on a crystalline form they arrange them- selves in specific relationships by the exercise of a force of the highest order. In the presence of this very forceful disposition of the particles to retain fixed relationships to each other, it would seem little less than a contradiction of terms to attribute to them viscous flowage. The crys- talline body may readily be made to change its form by the removal of particles from one portion by melting and their attachment at other 214 T. C. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. points by congelation, but not, I think, by the flowing of crystallized particles over each other while in their crystalline condition. ~ RELATION OF THE GLACIERS TO THEIR DEBRIS. The northern glaciers afford little that is new respecting lateral and medial moraines, and they may be neglected. It has already been seen that much basal material is carried in the lower layers of the ice. It— was also a matter of frequent observation that debris lies under the ice. Apparently the ice sometimes pushes this along and sometimes slides over it. At the end of the glacier the debris within the ice is freed by melting and accumulates as a talus-slope. This sometimes protects the basal layers from melting, and they become at length incorporated in the growing accumulation. Their subsequent melting gives rise to one form of kettle-holes, but only one form. It appeared from the stages presented by the several glaciers that where a glacier is slowly advancing the talus-slope gradually grows forward and constitutes an embankment upon which the glacier advances. It thereby grades up its own path- way in advance. On seeing this process one is at no loss to understand how ice can advance over fields of sand or soil without in any way dis- rupting them. It buries them before it advances upon them. A large number of the glaciers of the Inglefield region rest upon embankments or pedestals of this kind. Some, which have retreated, have left these exposed to observation. (Figure 13, plate 9.) Where the frontal material accumulates in a large mass it opposes such a degree of resistance to the ice that its layers are curved upward on the inner slope, and if the glacier subsequently advances the ice rides up over the moraine. Several such instances were observed, but none was seen where the ice showed any competency to push even its own debris, in notable quantity, in front of it. The ice is weaker than the moraine as a whole. WIND-DRIFT BORDER. Not only is the ice of the north Greenland glaciers weak when tested by the resistance of its own frontal moraine, but it is even weak when compared with the wind-drift accumulations of snow on its front. There is a very notable wind-drift phenomenon connected with the border of the great ice-field of north Greenland to which Lieutenant Peary was the first, I think, to call attention. The winds of the great ice-cap flow chiefly down its slopes, as though by direct control of gravity. They carry great quantities of snow, and this lodges in the lee of the terminal moraine. The border-drift thus formed has a breadth of from 1,000 to BULL. GEOL. SOC. AM. VOL. 6, 1894 PL. 9. FIGgureE 13.—Porv1I0oN OF LATERAL Face or East GracieR. Showing the perching of the glacier on its own debris, and the gneiss-like contortion of the laminz, due tothe resistance of the mound of debris in front. The overjutting of certain of the upper layers is ulso shown. Bowdoin bay seen in the distance. Figure 14.—Portion oF EpGr oF THE ICE-CAP. Showing upward curving of basal debris-bearing layers, due to the resistance of accumulations in front. EAST GLACIER AND EDGE OF ICE-CAP. ABSENCE OF ESKERS AND KAMES. 215 3,000 feet, and its slope rises from 100 to 250 feet, though a portion of this elevation is doubtless due to a slope of the earth’s surface below. This snow remains from year to year and becomes solidified after the fashion of a glacier; indeed, it is little short of a peripheral ribbon-lixe glacier, skirting the border of the great ice-cap. Between this and the ice-cap, as a narrow line of division, hes the terminal moraine. The three or four sections across this which were open to observation made it apparent that the moraine was formed by the basal layers of the ice-cap curving upward on encountering the resistance of the wind-drift border in front. The upward movement may have been initiated by a con- cealed moraine below, but superficially, at least, it would appear that even solidified snow in great mass is sufficient to deflect the advancing layers of ice, paradoxical as this certainly seems. ESKERS AND KAMES. No eskers or kames were seen in process of formation except in min- iature type. Nothing of the kind was seen upon the backs of glaciers; because, with trivial exceptions, there was no material there from which to form them. ‘The basal drainage of the glaciers was found to be chiefly accomplished by streams running along the sides of the glacial lobes. The central tunnels which most alpine glaciers possess were generally absent. The lateral streams frequently tunneled under the glacier or were bridged by snow-drifts, and doubtless when the ice has vanished there will be found terraces and side ridges of gravel analogous to one of the forms of eskers of-our drift, but nothing distinct or typical was seen. The radical reason lies, I suppose, in the fact that the total drainage is too small and too narrowly confined to the summer months. So also, in regard to the kames, it was observed that the drainage from the terminal slope of the ice-cap usually followed the inner side of the terminal moraine for greater or less distances until a low spot was found - across which it made its way. These transverse channels doubtless afford an illustration of the manner in which the gravel accumulates on the inner side of a moraine during its growth, and is subject all the time to disturbance by the movement of the ice; but here again only illustrative phenomena were seen. My observations, therefore, seem to have but one important bearing upon our theories as to kames and eskers. The debris of the great ice- sheet is confined to its lower portion, with trivial exceptions. It is almost absolutely wanting in the upper portion and on the surface. The heights at which it is found in the lower portion are not greater than the heights of kames and eskers; therefore, unless we resort to the violent hypothesis XXX—Bourn. Geox. Soc. Am., Vou. 6, 1894. 216 T. Cc. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. of supposing that the material was borne from lower to higher altitudes by the streams that formed the kames and eskers, only to be let down again, we are compelled to locate their origin at the bottom of the ice- sheet. This appears to confine hypotheses to the question whether accumulation took place in tunnels under the ice or in channels cut back from its edge. DRUMLINS. No drumlins were seen in process of formation, nor were any seen in the abandoned territory, unless we force interpretation in a few doubtful cases. The observations, however, seem to have some important bearings upon the elucidation of drumlins. The limitation of the debris to the basal layers of the ice limits the horizon of drumlin-making, as in the case of eskers and kames. The observations which showed the weakness of the marginal ice in comparison with the resistance of its own debris furnish ground for comprehending the accumulation of masses of drift beneath the edge of the ice. In describing the behavior of ice in passing over embossments of rock it may be recalled that the lamin were found to start on the frontal side of the embossment and to curve gradually upward and backward and at length downward in the lee, the trend of this curve being quite similar to the profile of a drumlin. I suspect that this is the true drumloidal curve, and that it represents the balance or the accommodation between the force of onthrust on the part of the overriding ice on the one side and the friction and resistance of the ice and debris against the emboss- ment on the other. I suspect that the progressive tendency in such a case is toward the accumulation of debris below this drumloidal line, which was apparantly a line of shearing, and that the result of such an accumulation would bea drumlin. Why this particular curve should be assumed is a problem the precise mechanics of which I do not profess to understand, but seeing the curve developed repeatedly I infer that it must be in conformity with the dynamics of ice-motion under these con- ditions, and that nothing remains requisite to the formation of a drumlin but the lodgment of drift below the drumloidal curve. (Figures 11 and 12, plate 8.) RATE OF MOVEMENT OF THE ICE. Lieutenant Peary has commenced a series of observations upon the movements of glaciers of the Inglefield gulf region, both by instruments and by photographs taken at intervals. He found the daily movement of the Bowdoin glacier, the most active in the immediate vicinity of his headquarters, during the month of July to be four-tenths of a foot at the RATE OF MOVEMENT OF THE ICE. Pile slowest point, near the east border, and 2.78 feet at the fastest point, near the center, with an average of 1.89 feet for the whole. The movement of the majority of the glaciers in that region is very much slower; indeed, in most cases it is obviously exceedingly slow. Many of the ordinary signs of movement are absent. In front of the Fan glacier there are cones of granular ice brought down by the surface streams, and also embankments of old snow, soiled, granulated, and half solidified into ice, as though at least a year old, all of which lie banked against the terminal face of the glacier without any indication of move- ment on its part since their formation. As these lean against the face to heights of 30 or 40 feet at least, it is obvious that there had been no melt- ing of the base of the extremity to counteract the effects of advance. Phenomena of similar import were observed in several other glaciers. The very firm impression was given by such physical signs that the average rate of movement of the glaciers of the region is very slow. At the head of the gulf are a few glaciers which produce large icebergs and which must be notable exceptions to the prevailing slowness of motion. GLACIAL DRIFT ON ABANDONED TERRITORY. ‘The amount of drift on the territory once occupied but now free from ice is notable rather for its scantiness than its abundance. On Disco island it was found to be very limited, except along the immediate fronts of the present glaciers. In the Inglefield Gulf region there are at some points very considerable accumulations of drift within a mile or two of the present ice-front, but at the same time much of the territory between the ice-front and the sea bears a very scant covering of drift. No great moraines were seen, nor any thick mantles of drift. The valleys in front of the glaciers are well floored with glacial wash, but even here the rock occasionally appears. Considerable delta-fans project into the gulf, but none of them exceed half a mile in depth. Consonant with this scantiness of drift, the topography of the border- land shows only moderate evidence of glacial subjugation. It is mildly rounded, but not greatly molded. ARE THE GLACIERS ADVANCING OR RETREATING ? Several glaciers on Herbert and Northumberland islands showed evi- dences of retreat; the terrace-like pedestals which they had formerly built were in part abandoned. Three other glaciers showed by the pres- ence of old moraines immediately in front that in the past they had been more extended than at present. These moraines may be a few hundred years old, but they offered no evidence of very great antiquity. One 218 T. Cc. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. glacier was seen overriding its terminal moraine in one portion and re- treating within it at another. This, taken in connection with the massive- ness of the moraine, probably indicates that it has stood practically stationary for a considerable period. The most remarkable evidence relative to former extension is furnished by a driftless area on the east side of Bowdoin bay immediately adjoin- ing the present great ice-cap. It is obvious that at this point the ice is as far advanced as it has ever been in the recent geologic history of Greenland. The verity of this driftless area is attested not only by the absence of transported material upon it, but by the exceedingly angular, ragged disintegration of the harder terranes of rock embraced in the com- plex gneissic series and by the deep disintegration of the gneiss itself. The gullies and ravines reveal the fact that the gneiss is deeply decom- posed to a soft, rotten mass, which is not only easily crumbled, but is even pliant under the fingers. It was possible to descend steep slopes by thrusting the heels deep into the softened mass. The combined weight of all this evidence puts beyond serious question the verity of the drift- lessness of this region. The area is small, not exceeding three or four miles in maximum diameter, and lies between the ice-edge and Bowdoin bay on ground whose average altitude is less than that of the glacier, so that its immunity from glaciation has not been due to elevation. It is clear, therefore, that the ice-border was stayed at this point by agencies concerned in its own development and not by any topographic barrier. Immediately at the south of this small driftless area there lies in front of the Gable glacier (which is but a short tongue of the main sheet) a stout old moraine, the surface of which has been notably weathered and has become covered with vegetation in the scant fashion of the region. There is nothing in the nature of this moraine to indicate an antiquity beyond perhaps a few hundred years, but its presence at this point seems _ to indicate that the ice has stood in the vicinity for a considerable period, and therefore that it is probably, on the average, neither much advancing nor much retreating. ForMER EXTENT OF GLACIATION. It is evident that the occurrence of even a small driftless area on a border of the widest stretch of the Greenland ice-sheet is extremely sig- nificant respecting its former extension. The general scantiness of the drift over the territory immediately outside of the present ice seems also to raise doubt as to any great former extension. There are two other lines of important evidence that bear upon this question. Dalrymple island is a mass of hornblendie gneiss rising from the water's edge to a height of perhaps 100 feet, with steep slopes and ragged surfaces. It is BULL. GEOL. SOC. AM. VOL. 6, 1894, PL. 10. Ficure 15.—DatryMeLr ISLAND, NEAR THE GREENLAND COAST. Showing unglaciated profile. Figure 16.—SourHEASTERN CAREY ISLAND. This island is about thirty miles west-northwest of Dalrymple island. The view shows glacial contour produced by movement from the north; not from adjacent coast of Green- land. The geologic structure of Dalrymple and Carey islands is almost identical, the differ- ence in contours being apparently due wholly to glaciation. DALRYMPLE AND CAREY ISLANDS. - FORMER EXTENT OF GLACIATION. 219 a famous nesting-place of the eider duck, which finds it suitable to its purpose because of this raggedness of surface. The island bears no sign of glacial abrasion. It stands at the mouth of Wolstenholme sound, on the west coast, in about latitude 76° 50’. In other words, it is just off the border of one of the broadest stretches of Greenland’s ice-field. Thirty or forty miles distant to the west-northwest le the Cary islands, which are formed of almost identical rock. They are very notably abraded by glacial action coming from the north. Striee are still preserved upon them at heights of 500 feet above the sea. There also occur upon them erratics of limestone, sandstone, shale and quartzite wholly unlike any- thing that occurs in the islands themselves. So far as I know, no rock of similar kind occurs in Greenland to the eastward. These erratics appear to have come from the region beyond Smiths sound to the north, either from Grinnell land or from the northwestern coast of Greenland—more likely the former than the latter. It appears, therefore, that while a very notable southerly movement from the far north took place down the valley and reached at least to the Cary islands, there was no correspond- ' ing movement from the east. (Figures 15 and 16, plate 10.) At the very first glimpse of the coastal mountains of southern Green- land I was impressed by their pronounced angularity and the absence, unless it were in the lower valleys, of any notable signs of the horizontal rasping which must have resulted had the inland ice ever pushed across them into Baffins bay. Subsequently I saw approximately a thousand miles of coastline, and an effort was made to discriminate the portions once overridden by ice from those which had not been. Tracts of angular, unsubdued topography were found alternating with tracts of rounded, flowing contours. About one-half of the coast seemed to belong to each type. The inference was drawn that the ice formerly so extended itself as to reach the present coast for about half of its extent, while in the re- maining portion the ice fell short. Combining this topographic evidence with the specific data furnished by a comparison of Dalrymple island with the Cary islands 'and with the still more stubborn facts offered by the driftless area of Bowdoin bay, the inference seems unavoidable that the ice of Greenland, on its western side, at least, has never advanced very greatly beyond its present border in recent geologic times. This carries with it the dismissal of the hy- pothesis that the glaciation of our mainland had its source in Greenland. ‘FORMER ALTITUDE OF GREENLAND. There is no ground to question the former elevation of Greenland. Its plateaus, like its valleys, indicate this; but glacialists are especially con- cerned to know whether the former elevation of Greenland was coincident 220 T. C. CHAMBERLIN—GLACIAL STUDIES IN GREENLAND. with its glaciation or not. Aside from the contours of the plateaus and valleys, which seem to indicate a fashioning rather by meteoric agencies than by pronounced glaciation, the driftless area appears to afford the most specific ground for induction. Bearing in mind that this is a small area between the present edge of the ice and sealevel, which would be overridden easily and completely by an advance of the ice-edge of less than five miles, it seems necessary to conclude that at the time of the former greater elevation the climatic agencies of glaciation could not — have been what they are now, for the increased elevation would have caused an extension sufficient to overwhelm the little driftless area. If it is safe to conclude that elevation favors glaciation, then it is necessary to conclude that during any period of previous glaciation there was here no elevation sufficient to cause an advance, unless accompanied by coun- teracting adverse climatic conditions. The raggedness of Dalrymple island bears similar testimony. The general angularity of the coastal mountains of south Greenland throw the weight of their evidence in the same direction. It would appear, therefore, that the former elevation of Greenland was not coincident with conditions favoring glaciation. BULL. GEOL. SOC. AM. THE “CEN TAAL GOLD BELT OF CALIFORNIA. 7S. O7UI U2 2979 ® Gold Quartz Veins (/ Y 5S SacramHpr yop ~ a "UY; 80 \ y 6 ol Poe AD ‘ Yd So Q . GOLD QUARTZ VEINS OF \. CALIFORNIA. 2 x ¢ ) fetes VOL. 6,-11894, PIE 44. — = g Lae 1c © Gold Quartz bvecns ULL Pleistocene of the Grea valley, ty, Le 22rd grern BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 221-240, PL. 11 MARCH 5, 1895 CHARACTERISTIC FEATURES OF CALIFORNIA GOLD-QUARTZ VEINS * BY WALDEMAR LINDGREN (Read before the Society December 29, 1894) CONTENTS Page AIBA MICRAISTA SS foo Thos a Sa ER ew hes a ee ase Ate ees Messen aal (PRE LUE TCLD ES CSU aN ODN 8 Ue Aa es cs Pr eg a 222, “2 BLL DLTY PELE OES S RS Bs hc Be esr mn Sec mM gma ar OE 222 I aN oe Ser cok di oi ce falas SN (o roe 2S tha Sed aiid w 8 wth waned Seca ana 225 Differing types of gold deposits............ NOS PALS cope n See gi Maen acta 226 Sunmerural relations......:2.:.--- eR Ore ere ee ene Wha NS TE aha on A 226 PERU OOM Meh VeINS 1m eet een Aes kee ele cee a dae cave baeed Pensa Ursa 229 Association of minerals in gangue goad ONES cacti adh UN MinEne Cee Bea Miele Obie wail) Miareion Ot fuerwold in the ves. .. 0.46.62. 46- cee se ens et ee ees 231 The alteration of the country-rock ........ et Seger ne ak sewed 2 Ua oy < oh. SeNe oek 232 ae ae ee CR COMANCE UREA OIA Sy loon 5 ie od PhS) 2 Sah Miciasdie si Rais os dees a Maw woe Ys sae 236 Comparison with quicksilver deposits............. High AP haan Se ae An ee ea Za Origin of the gold......-.... Pe a eee ene Se tte tind. ct VU Ive 239 Summary...... GAN LRC ae PARP aA AS ey A hem ADAAD aati tA e Shs mer A Pern SMe LE EER 240 INTRODUCTION. The gold-quartz veins of California, in spite of many local variations, form a remarkably well defined type of mineral deposits, the salient characteristics of which it 1s intended to portray in this paper. The results, indicated in brief outlines, have been obtained during general and detailed mapping for the United States Geological Survey in the gold-bearing region of California. Referring to the map of the distribution of veins, it should be stated that Inyo county, as well as the central and eastern part of San Ber- nardino, contains many gold deposits which have not been indicated. In many cases they carry both silver and gold, like the Comstock mines, * Published by permission of the Director of the U. 8. Geological Survey. XXXI—Butt. Grot. Soc. Am., Von. 6, 1894. (221) 29> W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. or differ in other respects from the normal gold-quartz veins, though the latter are not without representatives. For many notes and valuable suggestions I am under obligation to Messrs G. F. Becker, H. W. Turner and J.S. Diller. The reports of the state mineralogist of California have also been frequently consulted in the preparation of the maps. GEOGRAPHIC DISTRIBUTION. The general map of California accompanying this paper indicates the extent and distribution of the gold-quartz veins. Beginning in the penin- sular range south of the Mexican boundary, the deposits continue in scattered form and with many intermissions up to Fresno county, a few of them also occurring at isolated points along the coast ranges south of San Francisco. In Fresno county they become more abundant, and in Mariposa county the auriferous belt rapidly widens. From here north- ward to the point where they are covered by the great lava fields of northeastern California the maximum development is obtained. In latitude 40° the gold deposits extend from the great valley on the west to the summits of the Sierra Nevada on the east. In a northwesterly direction the continuation of the gold-bearing area is found in Shasta, Trinity, Siskiyou and Del Norte counties in California, and its northerly end occupies the counties of Jackson, Josephine and Curry in south- western Oregon. Volcanic flows and more recent superjacent formations cover the gold-bearing area toward the east and north. A smaller auriferous belt of less importance runs along the eastern slope of the Sierra Nevada, beginning in Alpine county and continuing southward through Mono, Inyo and San Bernardino counties. Most of the deposits along this line differ more or less from the normal type of the western slope. GEOLOGIC RELATIONS. In the northern part of the Mexican peninsula and in San Diego county granitic rocks prevail, but in them are imbedded numerous more or less contact-metamorphosed areas of slates and schists of uncertain age. The gold-quartz veins usually occur in, or at least close to, these areas. ‘The principal mining districts in San Diego county are Julian and 3anner, in the central part, and Pinacate, near the northern boundary.* Granitic rocks, with smaller schist areas, continue through San Ber- nardino and Los Angeles counties. Placer deposits and smaller veins * Mr W. H. Storms has described interesting lenticular veins from the former locality, which, according *to his explanation, doubtless correct, are only modifications of normal fissure veins. Eleventh Ann. Rep. State Mineralogist of California. GEOLOGIC RELATIONS. oS are found around San Bernardino mountain, as well as at several places in clay-slate near the summit of the range,* in the central and northern part of Los Angeles county. Very scattered and isolated deposits occur in Ventura, Santa Barbara, San Luis Obispo, Monterey and Santa Cruz counties. In Monterey paying veins have been found near the coast at Los Burros, sandstone being mentioned. as the country-rock. A short distance north of Santa Cruz a few gold-quartz veins are said to occur in unaltered sedimentary formations. In Kern county there is a lne of paying veins with a northeasterly strike, extending from Kernville to Tehachipi pass. Granitic rocks predominate, but contain a number of smaller schist areas, with which the gold deposits appear to be associ- ated. The locality is of interest on account of the number of hot springs occurring near the veins. Tulare county contains but few quartz-veins, but placer diggings are found along several of the rivers. In Fresno county, again, several streaks and smaller areas of schists and slates occur in the main granitic mass; again, the quartz-veins, which here attain greater importance, are closely associated with the former, though not exclusively occurring in them. Continuing north- ward for about fifteen miles to Mariposa, these belts of schists and slates suddenly widen, and at the same time begin to contain numerous and rich quartz veins. Between this region and the lava fields of the north le the most productive gold-mining regions of California. The western slope of the Sierra Nevada is from here northward occu- pied by a gradually widening belt of rocks, to which the name “ meta- ‘morphic series” is usually given. It attains its maximum width in Butte and Plumas counties and continues across northwestern California and southwestern Oregon to the Pacific ocean. The eastern part and the summit of the Sierra Nevada are still occupied by the continuation of the southern granitic area, bordering upon the ‘‘ metamorphic series,” with an irregular and jagged contact-line, along which evidence of the later origin and intrusive character of the granite may be frequently observed. ‘This contact-line is indicated on the map. The “ metamor- phic series,” sometimes also referred to collectively as the “ auriferous slates,” is a very complex mass of rocks. It consists largely of more or less altered and highly compressed sediments, of an age ranging from early Paleozoic to late Jurassic, and bearing evidence of having been subjected to several mountain-building disturbances. Associated with these sediments are igneous masses —augite-porphyrite, diabase, serpen- tine, etcetera—also ranging in age from Paleozoic to late Mesozoic, though the greater mass of them appear to date from late Jurassic or early Cre- taceous time. To a considerable extent these igneous rocks have been * Acton mining district. 224 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. acted:on by the dynamo-metamorphic processes which also affected the sedimentary rocks, and are largely converted into crystalline schists. It may be said in general that the sedimentary rocks prevail in the eastern part of the metamorphic belt, while along the great valley basic, igneous rocks are found in the greatest abundance. The granitic rocks of the high Sierra Nevada are to a large extent granodiorites, the name adopted on the survey maps: for a quartz-mica-diorite containing more or less orthoclase. In the metamorphic series there are many smaller masses of the same rock—the latest intrusions—which are usually but little affected by dynamo-metamorphic processes. The intimate connection of the gold deposits with the metamorphic series or the auriferous slates has been recognized for a long time, and Professor Whitney emphasizes it repeatedly in his works. The auriferous region, indeed, corresponds closely with the extent of the metamorphic series. Even in the south, where the granitic rocks predominate, it has been shown that the gold deposits are usually connected with the scattered schist areas. Few gold-quartz veins are found in the granitic area, and then usually near the contact. Within the typical gold-bearing region the veins are distributed with remarkable impartiality, and occur in almost any of the great variety of rocks which make up the metamorphic series. They are found in granite, diorite, granodiorite, gabbro and ser- pentine ; in quartz-porphyrite, augite or hornblende-porphyrite and dia- base; in amphibolite and other dynamo-metamorphosed rocks; in sedi- mentary, more or less altered slates, sandstones and limestones. In Tertiary volcanic rocks gold deposits are only found on the eastern slope of the range. It is apparently impossible to formulate any law as to their lithologic occurrence or to say that they prevail in any one kind of rock in the metamorphic series. - Regarding the quartz-veins of California F. von Richthof en has made a frequently quoted statement which in a certain sense may be correct, but which unless qualified is apt to lead to grave errors. Itis as follows:* The auriferous quartz veins ‘‘ have in their occurrence clearly discernible con- nection with the extension of the granite. They are crowded closely at its contact with the metamorphic rocks, and occur here partly in the former, partly in the latter. The greater the distance from the granite, the rarer they become in the metamorphic rocks, and only occur as an exception where the influence of the out- cropping granite would not be expected on account of its distance. In the same way they become less frequent in the granitic regions as the distance from the con- tact increases, and are, as a rule, entirely lacking in the interior of the large granite masses.”’ This statement cannot be accepted for the main granitic contact, which, on the contrary, except near Sonora, is remarkably barren of important * Zeitschrift der deutschen Geol. Gesell., B. xxi, 1869, p. 727. GEOLOGIC RELATIONS AND AGE. 225 deposits. In the larger part of the gold region a wide belt of Paleozoic slates comparatively poor in gold deposits separates this contact from the principal gold-producing districts. In very many places, however, the contact clearly marks the abrupt beginning of auriferous deposits, though perhaps poor and of small extent. The sudden change of recent and Tertiary river-beds from barren to auriferous when cutting across the contact is often very noticeable. Though not applicable to the main granitic contact, the statement quoted is to a certain degree true of the smaller masses of granodiorite scattered through the metamorphic series, for it is very common to find the gold-quartz veins clustered near their contacts in the manner indi- cated. It is not so general, however, as to be called a rule or a law, for there are many included granitic masses the contacts of which are in no way remarkable for abundant deposits. Dr W. Moericke, who has recently published several very interesting papers on the gold deposits of Chile, has come to the conclusion that they are closely associated with acid, igneous rocks, and drawn a comparison between the occurrences of that country and California.* In view of this, it may be well to emphasize the fact that the gold-quartz veins of Cali- fornia do not in their surface relation show any remarkable dependence on acid, igneous rocks. The great mother-lode, for instance, is in loca- tion and occurrence of its ores in no way related to such rocks, they being, on the contrary, as a rule, distant from it. . Normal gold-quartz veins in diabase and augite-porphyrite sometimes occur far away from other rocks, although the larger areas of the former are, on the whole, rather barren. AGE. Before beginning the discussion of the characteristics of the deposits, their age may be briefly touched upon. It has long been apparent and insisted upon by Whitney, von Richthofen and others that the quartz- veins of California are of late Jurassic or early Cretaceous age, and the same authors have suggested that they probably owe their origin to thermal action following the granitic intrusion. For the larger number of the quartz-veins this is undoubtedly true. It is certain that the majority were formed subsequent to the latest dynamo-metamorphism of the sedimentary and old eruptive rocks of the Sierra Nevada, subse- quent also to the granitic intrusion. It is, however, also certain that some deposits antedate this period, for in the latest sedimentary member of the bed-rock series there are conglomerates containing quartz pebbles and free gold,j which appears to have been concentrated as placer gold * Zeitschrift fur prakt. Geol. Jahrgang, 1894, p. 28. 7 W. Lindgren: Am. Jour. Sci., October, 1894. 226 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. at the time the conglomerates were formed. It does not appear easy to separate the earlher deposits from the later, but it is probable that they were neither very numerous nor very rich. Again, the eruptive activity of late Tertiary time which was centered along the summit and on the eastern slope of the Sierra Nevada was followed by another period of thermal activity, and another line of gold deposits was formed. This intermittently recurring action confirms von Richthofen’s generalization that a region once metalliferous is always metalliferous. Successive eruptions in such vicinity produce successive mineral deposits, whiie other eruptive centers are wholly barren of them. DIFFERING TyPpEsS oF GoLD DEPosIts. It is desirable to eliminate a few deposits of a different type from the prevailing one. Most important among them are the impregnations,* of which several examples occur in the Sierra Nevada and which may be of two types: First, zones containing grains of iron pyrites disseminated in fresh dynamo-metamorphic amphibolitic schists. These zones are seldom strongly auriferous, but may enrich quartz-veins passing through them, and are apparently similar to the so-called “ fahlbands” in erys- talline schists. These deposits are distinctly older than the principal quartz-veins and contemporaneous with the dynamo-metamorphism which produced the schists from the diabases and other rocks. Second, impregnations of later date forming irregular zones, in which the massive rocks or schists have been decomposed and filled with secondary aurifer- ous sulphides. These deposits are probably contemporaneous with the principal period of vein-filling and only a phase of it, in which the solu- tions, instead of following distinct fissures, permeated whole masses of rocks. The first of these types of impregnation is not of great economic importance, but the second sometimes affords large masses of low grade ores. STRUCTURAL RELATIONS. Regarding the structural relations of the normal gold-quartz veins it should first be stated that they are fissure veins, and emphatically not so-called segregated $ veins or “ lenticular masses ” in the auriferous slates. * This word is here used in its general sense, and not confined to the filling of interstitial spaces in porous rocks. + See later, under “ The alteration of the country-rock,” page 235, line 4 from bottom. { The term “‘segregated vein” is not quite clear and has been variously interpreted. A. Phillips evidently considered the only criterion of a segregated vein to be in its parallelism with inclosing slaty or schistose rocks, admitting motion along the walls and filling by foreign material, while R.S. Tarr, in a recently published volume, regards a segregated vein as the result of dynamo- metamorphism and a concentration of material from surrounding rocks, preéxisting cavities not being necessary. I have used it as meaning more or less lenticular openings in the mass of slates and schists, parallel to strike and dip, produced by longitudinal compression and filled by a sort of lateral secretion or exudation from the surrounding rock. STRUCTURAL RELATIONS. 227 It is everywhere plain and evident that the fissures have been broken open subsequently to the metamorphism of the rocks. These post- Jurassic and post-granitic quartz-veins form the latest chapter in the Mesozoic revolution in the Sierra Nevada. Neither Whitney nor von Richthofen commit themselves to an expres- sion of the “segregated ” nature of the veins. A. Phillips, in his book on mineral deposits, mentions their affinity to fissure veins, although classing them as “segregated veins.” All these writers, however, state that the veins nearly always conform in strike and dip to the inclosing slates. This has evidently led the authors of recent text books to class the Cali- fornia veins as “segregated.” Thus Professor J. F. Kemp, in his ‘“ Ore deposits of the United States,” classes them as such with some doubt, while Professor R.S. Tarr, in his “ Economic geology of the United States,” thinks that “in spite of the recent observations (by H. W. Fairbanks) it still seems as though these quartz-veins must be of segregation origin.” Quartz-veins like those Professor Tarr has in mind, formed by a sort of dynamo-metamorphic process, | am quite sure do not exist in the gold-belt. The somewhat auriferous “ fahlbands ” in certain amphibolites approach nearest his conception. Iam by no means prepared to deny, however, that there may be some minor ore-bodies deposited in openings in the slate from silicious solutions derived from the immediately sur- rounding rocks, but if they occur, they are surely exceptions to the general rule. In altered quartzose slates nodules and lenses of quartz seemingly of such origin frequently occur on a small scale. This rule of “ parallelism with inclosing slates”? must unquestionably be rejected in a general description of the veins. It should first be pointed out that a very large number of veins, especially in the northern part of the gold-belt, from Placer to Butte county, do not occur in slates or schists, but in massive rocks, such as diabase, granodiorite or gabbro, and among these a predominating direction of dip and strike does not exist. In slates and schists the veins often strike about parallel to the slaty cleavaee—that is, northerly or northwesterly—but other directions are nearly as common. Only very exceptionally is there a strict paral- lelism in both strike and dip. The great mother-lode, for instance, is parallel to the strike of inclosing rocks, but differs not inconsiderably from them in dip. Its character of fissure vein is clear and unquestion- able, and has been justly insisted upon by H. W. Fairbanks.* All direc- tions and all dips are in fact represented among the California quartz- veins, only dips below 20° and above 70° are comparatively rare. A general rule for strike and dip cannot be given; different laws guide them in different mining districts. The quartz-veins are the expression * Tenth Ann. Rep. State Mineralogist of California, 1890, p. 86. 228 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. of the greater and minor strains to which the Sierra Nevada has been subjected, and a study of the former will, to a considerable degree, illus- trate the latter, which have certainly varied in intensity and direction from point to point. Thus, to pick out a few illustrating examples, the veins of Ophir, Placer county, consist of two principal systems, one set of veins running west-northwest and dipping south, while the other has a west-southwest strike and southerly dip, both cutting the surrounding schists obliquely to their strike and dip. At Grass Valley and Nevada City there is one system with a general northerly direction and dipping either east or west; another system courses east and west and dips north or south at varying angles. The surrounding rocks are here mostly massive. The veins in the vicinity of Sierra Buttes, Sierra county, show the greatest divergencies in strike and dip. Equally variable are the veins about Sonora, Tuolumne county. The force producing these fissures appears in most cases to have been a compressive stress acting at an angle more or less oblique to the hori- zontal. In some cases this force produced one large and prominent fracture, but far more commonly one or several series of fractures, or a sheeting * of the country-rock along which the auriferous solutions could circulate. Along the larger fissures considerable movement has taken place, but when the country-rock has been sheeted the motion along the individual joints has probably not been very great. In many cases, when the direction of the movement could be proved, it has been found that a relative upward movement of the hanging wall has taken place. The force did not produce a single, sudden and catastrophic movement ; on the contrary, it continued for long time, resulting in repeated dislo- cations, as proved by the reopening and refilling of some veins and by a sheeting of some veins, producing what is usually described as “ ribbon rock.” Recemented quartz-breccias are also of common occurrence. I should here like to mention one misleading circumstance relating to parallelism of vein and country-rock. When larger fissures are opened in massive rocks it is not at all uncommon to find the immediately ad- joining wall-rock converted entirely locally into schists parallel to the fissure, under the influence of the enormous shearing stress to which it has been subjected. Such veins would have the appearance of cropping in preéxisting schist-masses, and of parallelism in strike and dip with these. The conclusion to be derived from the relation of the veins to the larger, regionally metamorphosed schist-masses is that the schistose structure antedates the formation of the vein fissures; and that the forces * The relation of the forces and the sheeting has been discussed by Mr G. F. Becker: Bull. Geol. Soc. Am., vol. 4, p. 18. See G. F. Becker, “Geology of the Comstock Lode,” Mon. III, U. S. Geol. Survey, p. 182, and 8. F. Emmons, ‘‘ Structural Relations of Ore Deposits,” Trans. Am. Inst, Min. Eng., vol. xvi, p. 814. CHARACTER AND FILLING OF THE VEINS. 229 to which these fissures are due, while bearing a general similarity to those manifested in the cleavage, often differed from them in direction to a sensible extent. Different rocks influence the character of the fissures to some extent. In massive rocks they are apt to be straight, clear cut and well defined ; in slates and serpentines there is often a tendency to splinter into a net- work of cracks and fissures, extremely small, but often very rich. In such cases the whole mass, country-rock and vein, may be extracted and milled. Linked veins are common and chambered veins* sometimes occur. Very long and continuous veins are not common, and in this respect the mother-lode is rather an exception. Only rarely can a quartz-vein be traced more than a few miles, and many important veins crop out only for a short distance.f THE FILLING OF THE VEINS. The typical gold-quartz veins cannot be considered as anything but fissures and fractures filled with quartz, accompanied by small amounts of native gold and metallic sulphides. Replacement proper of the min- erals of the country-rock along the fissure by quartz I have never been able to observe, and cases supposed to be of such nature have always proved to be due to the shattering of the country-rock and the filling of it by silica along narrow cracks. The clean quartz usually forming the vein I cannot account for in any other way than by filling of cavities, as it does not seem possible that a replacement of the ferro-magnesian sili- eates and other minerals could occur without leaving chloritic stains or other signs in the resulting mass. In all quartz-veins of this type it seems unavoidable to admit the existence of open spaces along the vein, supported at frequent intervals by the contact of the two walls or by rock fragments. Even the heaviest veins show in the underground workings frequent places where the walls “shut down.” Such fillings of clean quartz may vary in width from a few inches to several feet. ‘“ Horses,” *G. F. Becker: Quicksilver Deposits of the Pacific Slope, p. 409. + It is not true that every fissure vein holds out to indefinite depth, though it is probable that most of the larger veins of the gold-belt will continue to a depth exceeding the limit of practica- ble exploitation. As arule, the probable permanence of a fissure vein will be in direct proportion to its traceable length and to its width. In regions where strong sheeting of the rocks has taken place it is quite probable that many of the smaller fissures and joint-planes will pinch out and disappear indepth. Fissuring, after all, is most intense near the surface, and probably compara- tively few of the fissures reach down to deep seated regions; when the rocks become plastic by pressure and heat, or by a suitable relation between the applied stresses, such as must prevail below a certain level, all fissures must cease to exist. The smaller fissures probably received their quartz-filling by communication with the larger ones, which must be regarded as the princi- pal conduits for the solutions. XXXII—Butt.. Gron. Soc. Am., Von. 6, 1894. 230 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. of course, frequently appear in the larger veins, separating them in two or more parts. The heaviest veins appear to be found along the mother- lode in Tuolumne and Mariposa counties, where the clean quartz often reaches a width of 10 to 15 feet, and in isolated cases even more. ‘This extreme thickness seldom continues for any great distance. It may probably be safely assumed that gold-quartz veins of this type cannot be formed at any extreme depths below the surface, probably not below 10,000 feet, for at such depth open spaces could hardly exist. These very large fissure veins are, however, not very abundant; a moderate width of one to three feet is far more common. In many cases, indeed, there are no large open spaces at all, but a network of smaller cracks and fractures, in which the solutions deposited their contents. ASSOCIATION OF MINERALS IN GANGUE AND OREs. In the predominating milky white quartz of the veins but few other gangue minerals are found. Calcite, more rarely magnesium carbonate, or a mixture of both, occur occasionally, but always in subordinate quantities and usually concentrated near the walls. The quartz is ordi- narily massive, but excellent examples of comb-structure may be found. Barite* and fluorite are conspicuously absent. A white mica with pearly luster is sometimes found in the quartz at some of the mines along the mother-lode, and a green potassium-mica, colored by chromium, and which Professor Silliman has called mariposite,t occurs abundantly in places, though usually not in the quartz itself. Roscoelite,a vanadium- potassium-mica, has been found in one place, and albite occurs in isolated cases. Rhodonite has been found in Plumas county. Titanium min- erals, such as titanite, ilmenite and anatase, occur occasionally. The native gold is distributed through the quartz-gangue in an irreeular manner. The particles may be of microscopic size, or coarser, and visible to the naked eye as scales, threads and smaller masses. Occasionally large pieces, of all weights up to fifty pounds or more, will be found. In the ores from the larger mines it is, however, rare to find the gold visible to the naked eye. The gold always contains a little silver, in rarer cases as much as 30 per cent. A variable but always comparatively small quantity of metallic min- erals accompanies the gold. It varies from a fraction of 1 per cent to 5 or 6 per cent, but ordinarily makes up from 2 to 3 per cent of the mass * For rare occurrences of barite see W. Lindgren, Am. Jour. Sci., vol. xliv, 1892, p. 92, and H. W. Turner, Am. Jour. Sci., vol. xlvii, 1895. + For analyses see H. W. Turner: “ Further notes of the gold ores of California,” Am. Jour. Sci., vol. xlvii, 1895. {See the interesting paper by H. W. Turner: Am. Jour. Sci., May, 1894, vol. xlvii, p. 467. ASSOCIATED MINERALS IN THE VEINS. ZS) of the quartz. Sulphides are most common, but compounds of arsenic, antimony and tellurium also occur. A list of the associated minerals in the quartz-veins would include the following species: Iron-pyrites (universally present). Tetrahedrite. Pyrrhotite (not common). ~ Antimonal lead sulphides (rare). Copper-pyrites (common). Cinnabar (rare). Zinkblende (common). Tellurium minerals—Hessite, Altaite, Galena (common). Calaverite, Sylvanite, Petzite, Melon- Molybdenite. ite (frequent, in small quantities). Arsenical pyrites (common). Nickel and cobalt minerals (very rare).* Marcasite is noticeably absent from gold deposits as noted by Mr Louis. I have once seen it, however, from a mine at Grass Valley. These metallic minerals, usually referred to as “sulphurets,” contain more or less gold and silver and are frequently very rich, the concentrates ranging from thirty to several hundred dollars per ton. Bismuth and cadmium have been found in small quantities in the concentrates from the Nevada City mines, the former also in Shasta county. Compounds of tin, wolframium, uranium, boron,} phosphorus and fluorine appear to be entirely absent. Cuprite, bornite and chalco- cite are also lacking. Cobalt and nickel minerals are occasionally present. Titanium occurs sparingly. ‘A slight influence of the wall-rock upon the character of the mineral association cannot be denied. It appears to be a fact that veins in grano- diorite contain more sulphurets than those in other rocks. Pyrrhotite appears to be entirely confined to veins in this rock. Itis known only from the vicinity of Washington, in Nevada county, Sonora, in Tuolumne county, and Westpoint, in Calaveras county. Veins in black sedimentary slate or on the contact between greenstone and slate seldom contain much besides iron-pyrites, and perhaps arsenical pyrites; neither are veins in augite-porphyrite or diabase usually rich in sulphurets. Veins in gabbro often contain copper. These are no strict rules, however, and the influence of the wall-rock may, on the whole, be considered as re- markably smali. ; DISTRIBUTION OF THE GOLD IN THE VEINS. Gold is universally distributed in the quartz-veins of California. The definition of what is ore, or quartz paying for exploitation and metallurgic treatment, will necessarily vary at different times and in different places. * Compare a paper by Henry Louis in Min. Magazine, vol. x, 1893, p. 241, on the minerals asso- ciated with gold deposits in general. +R. Pearce, in Trans. Am. Inst. Min. Eng., vol. xviii, p. 447. {Tourmaline has been found in the abnormal veins of Meadow Lake, Nevada county. Am. Jour. Sei., vol. xlvi, 1893, article 30. Doz W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. Under exceptional circumstances rock containing as little as one or two dollars of gold to the ton will pay. In the deep mines the tenor of the extracted ore is usually from five to twenty dollars. In wider veins a small streak near one of the walls will sometimes con- tain the pay, while the rest is comparatively barren. Equal distribution of value in cross-section is, however, common enough. Considered in projection on the plane of the vein, there is rarely an equal distribution of the gold over large surfaces. The richer ore is concentrated in bodies and masses, which sometimes may be wholly irregular, but which usually show more or less regular outlines. These richer masses are called chutes or chimneys, and appear on the plane of the vein in long-drawn linear or elliptic form, with a dip which usually is above 45 degrees. Flat ore- chutes occur, however, as, for instance, in the Idaho mine, Grass Valley. Their width ranges from a few feet up to several hundred, and their length may exceed 2,000 feet. It is not uncommon to find one of these ore-chutes give out in depth, but another chute will then probably be found in some place below it, if a thorough exploration is carried out. It is a practical rule in many districts, and one which holds good ina remarkable number of cases, that the chutes dip to the left when one is standing on the apex and looking down along the dip. The explanation of the ore-chutes is difficult. They may,as F. Posepny and others have suggested, simply indicate the direction of least resistance for the gold- bearing solutions. This explanation is not entirely satisfactory, for in the intervals between the chutes it is by no means the rule to find the walls shut down tight. On the contrary, it is common to find the barren vein between them as wide if not wider than the rich vein in them. An increase in the quantity of the sulphurets always accompanies the in- crease of gold in the ore-chutes. | No gradual decrease in the tenor of the ore takes place with increasing depth: on the whole, the character remains constant. Individual ore- chutes may be exhausted, but others, as a rule, are found below them. Certain veins show no. large bodies of milling ore at all, but coarse gold concentrated at certain points; such deposits are called “ pocket veins.” Small seams may sometimes carry a surprisingly large amount of gold. Intersection of seams or veins often, but by no means always, produce pockets or ore-chutes. Tur ALTERATION OF THE COUNTRY-ROCK. The study of the changes and alterations which the rocks adjoining the fissures have undergone is a subject of the highest importance, for in this way a closer insight into the genetic processes of the vein may often be obtained. ALTERATION OF THE COUNTRY-ROCK. Doar It would at first glance seem more likely that the rock in the vicinity of the quartz-filled veins would have undergone a silicification. Such is not the case. Instead of a silicification there is, as a rule, a most marked carbonatization, or a conversion of the country-rock to carbonates. Most intense next to the vein, the alteration gradually decreases at a distance from it, the width of the altered zone varying according to the width of the vein. The carbonate zone, surrounding the quartz-vein on both sides, may often be studied to great advantage in small veinlets cutting through hand specimens. , This action upon the adjoining country-rock is in itself, to my mind, the strongest possible evidence against the application of lateral secretion in its narrower sense to these veins. It appears to completely refute the theory of the veins being formed by percolating surface waters, and prove the existence of an agency active in the fissures and gradually extending outward. The solutions circulating in the fissures acted with different intensity on different rocks. Nearly all igneous rocks, acid or basic, are profoundly altered, the latter more than the former, and serpentine more than any other. Only extremely silicious rocks, and especially certain carbona- ceous slates, appear to successfully withstand the action of these solu- tions. ‘The process of carbonatization has not in all cases been carried out to its full extent; in some veins it is more marked than in others; occa- sionally fresh rock may lie close up to the vein.* Crushing of the rock next to the vein facilitates the process and increases the width of the altered zone, which may vary from a few inches up to twenty feet, and even more in exceptional cases. With all variations, there is no doubt that the process is a general one, and characteristic for the type.t The result of the process, when it has been thoroughly carried out, is the conversion of the country-rock by replacement to a mixture of car- bonates, white potassium-micas (sericite), a small amount of chloritic minerals and residuary quartz; besides, there is always a large amount of iron pyrites.{ usually more than in the vein; arsenical pyrites{ is also frequently present, but never, as far as I know, any other sulphides in noticeable amounts. Calcium carbonate usually prevails, but the car- bonates of magnesium, iron and manganese are also present. Accord- ing to numerous analyses, calcium is always added, while nearly all of the sodium is carried away. The potassium of the orthoclase re- mains transferred to the sericite. As abundant potassic micas are often * Such cases are perhaps due to layers of impervious clay-like detritus on the wall. +A good instance has been described by the author in the Fourteenth Ann. Rep. U.S. Geological Survey, in a paper entitled “ The Gold-silver Veins of Ophir, California,” now in press t Both occur as small but extremely sharp crystals, while the sulphurets in the quartz are usually massive. 234 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. found in wall-rocks originally very poor in this metal, it is probable that some potassium was also added. In silicious rocks the quartz is often attacked, but never completely carried away. The iron ores and partly also the bisilicates of the original rock appear to have been converted into pyrites,* while the titanium in it was transformed to leucoxene.t In the case of clean-cut fissures, with well defined quartz-veins, it is usual to find by far the largest amount of gold in the quartz and in the sulphides associated with it. The altered country-rock is not entirely | barren, but it does not often contain native gold, and its sulphides are much poorer than those in the quartz.t This is not entirely without ex- ceptions, for in several places, usually adjoining rich chutes, the altered country-rock will pay for milling, and may, in isolated cases, go as high as $12 per ton. And again, there are cases in which the altered country- rock is traversed by a great number of minute quartz seams, in which the gold is concentrated. Such a case is the Rawhide mine, Tuolumne county, in which this altered and fissured country-rock is far richer than the main quartz-vein. At the same place the gold sometimes also pene- trates and coats the cleavage faces of the adjoining talcose or serpentinoid schistose rock. One frequently hears of native gold in tale, slate or other rocks. I have always found such occurrences to be more or less altered rocks from the immediate vicinity of some vein. The gold occurs on minute, sometimes hardly visible, seams traversing them. Indeed, many fissures are absolutely microscopic. It has been stated above that serpentine § is peculiarly liable to altera- tion by the auriferous solutions. The zones of altered rock are in this case often very large and always very characteristic. They may be twenty or thirty feet wide, or in the case of branching veins a whole area, several hundred feet across, may be more or less completely altered. The serpentine is converted into a mixture of magnesic and calcic car- bonates, a green micaceous mineral containing potassium and colored by chromium, to which the name of mariposite has been given by Professor Silliman, together with more or less iron pyrites. The altered mass is frequently shattered and traversed by seams of mixed quartz and car- bonates. It has a rather coarse, crystalline structure, and a bright green color from the disseminated mariposite. The carbonates referred to as * A similar alteration has been shown to have taken place in the country-rock of the Comstock lode by G. F. Becker, Monograph III, U. 8S. Geol. Survey, p. 210. + The alteration and replacement of the wall-rocks has been emphasized by S. F. Emmons in regard to the fissure-veins of Colorado and Montana, and he points out that, especially where ex- tensive sheeting has taken place, the fillings of open spaces are often small comparéd with the alteration and impregnation of the adjoining country-rock. “Structural relations of ore-deposits,” Trans. Am. Inst. Min. Eng., xvi, p. 808. t This fact, as well as many others, of course, speaks strongly against the derivation of the gold in the vein from the decomposed zone adjoining it. 2 As well as tale-schist and other slaty magnesian rocks derived from serpentine. ALTERATION OF THE COUNTRY-ROCK. 9385 ankerite by Professor Silliman are in reality, as indicated by H. W. Fair- banks,* a mixture of varying composition, ranging from calcite to mag- nesite, and often containing considerable iron. Magnesic carbonate, on the whole, predominates. The mineral mariposite is, as Silliman observes, only associated with magnesian and chloritic rocks. Fairbanks + states that it is particularly characteristic of the mother-lode. This is not correct. It is, however, eminently characteristic of all quartz-veins in or at the contact of serpentine, though occasionally occurring in very small quantities in diabase and other basic rocks. The writer has noticed the ‘same characteristic mixture of carbonates and mariposite from a great many places in the gold-belt besides the mother-lode; thus, for instance, at the Phoenix and Red Chief mines in Sierra county, and also near Washington, Nevada county. It appears at the mother-lode wherever that great quartz-vein breaks through serpentine. Quartz mountain, Tuolumne county, is an excellent place to study it. Along the mother-lode the altered serpentine has been variously inter- preted. Whitney inclined to the belief that the vein represented a stratum of silicified dolomite, a theory that has not been supported by more detailed investigation. Fairbanks, who some years ago carefully examined the mother-lode.{ considered it at first as vein-matter deposited in open fissures, but regarded it subsequently (as the needed, once open space would manifestly have been too large, in places several hundred feet) as an altered, coarsely crystalline basic rock. The latter theory, while nearer the truth, is unnecessary. A careful investigation will not fail to disclose the fact that the mixture of carbonates and mariposite is nothing but an altered serpentine, and abundant transitions may be found to prove this. A locality showing this plainly is the App mine at Quartz mountain, Tuolumne county. This conversion is not astonishing when the facility is considered with which the serpentine is decomposed by carbonated waters into magnesite and chalcedonic quartz. Hxperi- ments by C. Doelter$ show that while at ordinary temperature and pressure water containing carbon-dioxide will, with simultaneous decom- position and formation of carbonates, dissolve 0.3 per cent orthoclase and 0.5 per cent oligoclase, serpentine will be dissolved at the rate of 1.24 per cent. The large bodies of decomposed rock referred to on page 226 as con- taining impregnations of auriferous pyrites and rarely free gold are in many respects interesting. In the ferruginous outcrops the iron-pyrites is usually converted into ferric hydroxide and the gold set free ; the whole * Tenth Ann. Rep. State Mineralogist, p. 85. j Loc. cit. { Loe. cit. 2 Allgemine chemische geologie, Leipzig, 1890, p. 190. 236 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. mass can then sometimes be profitably mined and milled, though it is of very low grade. Veins and seams of quartz are often entirely absent in these impregnated zones. In the cases which have come under my observation the action on the rock is much the same as in the decom- posed wall-rocks of the veins—that is, there is an abundance of carbon- ates and iron-pyrites in sharp edged, little crystals. While there is abundant evidence of replacement by carbonates, [ have not yet seen anything proving a replacement by quartz, though the possibility of such a process cannot be denied. However, in these deposits the action of the solution on the rock-forming minerals must have produced much free silica in soiution and probably also much sodie silicate ; in fact, there are in these deposits occasional masses of granular, grayish quartz very different from the ordinary vein-quartz and probably partly chalcedonic. This quartz often contains iron-pyrites in small scattered crystals, and appears to represent in part residual masses from leaching, in part depo- sition from the supersaturated silicious waters. H. W. Fairbanks has recently described two deposits in El Dorado county, the Big Canyon and the Shaw mines,* as showing in marked degree a replacement of the rocks by silica. Though the latter mine was not worked during my ex- amination of the Placerville sheet, I have, through the kindness of Mr H. W. Turner, had occasion to examine an excellent suite of specimens, lately collected. The vein is partly in black slate, partly in a feldspathic dike. Both rocks contain an abundance of stringers and seams of quartz and calcite, but I fail to see any evidence of replacement of the wall- rock by the former mineral. On the contrary, the porphyritic dike is to a very marked degree converted into carbonates in the vicinity of the veins. Regarding the Big Canyon mine, I have seen only two specimens of greenstone impregnated by pyrite from this mine, and collected by Mr H. W. Turner. These specimens show carbonatization to a consid- erable extent, but no evidence of replacement by silica. It is not in- tended to deny that such a process may take place, but only to point out that it is something requiring more and more detailed investigation. Calcite is found pseudomorphic after an enormously large number of minerals, while pseudomorphs of quartz after other minerals are much less common. GENETIC CONCLUSIONS. The country-rock altered to carbonates, standing in strong contrast to the vein filled nearly exclusively by quartz, affords a much-needed key to the genetic processes of the deposits. It shows, first, that besides silica, the water circulating in the fissures contained large amounts of * Twelfth Ann. Rep. State Mineralogist, 1894, pp. 103 and 114. ORIGIN OF THE VEINS. Del, carbon-dioxide, as well as dissolved calcic carbonate. It certainly con- tained sodium as carbonate taken up from the feldspars of the adjoining rocks, probably also as silicate and chloride. It further contained sul- phur, in what form is not certain, but most probably as sulphuretted hydrogen or as sulpho-salts. The presence of large quantities of sul- phates does not appear probable. Waters of this composition, containing abundant carbon-dioxide, are only known in nature as ascending, usually hot springs. The process of deposition took place as foilows: At first the carbonated waters began to act with great energy on the soluble minerals in the wall-rocks of the fissures, converting them more or less completely into a mixture of carbonates, potassium-micas and pyrites, adding calcium-carbonate and sulphur, probably also potassium, to them, and abstracting sodium. Finally, this process being completed, and the walls usually coated with crystals of carbonates, the formation of the latter ceased, and in this surrounding of carbonates the silica now began to be deposited, and with it the gold and une rest of the metallic sulphides. A most interesting question in connection with this subject is, why the walls should, to such a large extent, act as a separating barrier for the gold and most of the sulphides. Mr G. F. Becker, in discussing the quick- silver deposits of the Pacific coast,* has suggested that this may be due to an osmotic action, transmitting through the septum only the chem- ically active spol ula Admitting that the gold-quartz veins were deposited by such mineral waters, the next question is, in what form the gold and other metals were in solution. While not intending to enter into a detailed discussion of the difficult problems associated with the question,t I would like to call attention to a few general facts connected with them. Gold is soluble at 200° centigrade in a 10 per cent solution of carbonate of sodium to the extent of 1.23 per cent (Doelter), while silver is hardly attacked. Silicates of alcalies dissolve gold at 250° centigrade to the smaller ex- tent of 0.101 (Doelter and Liversidge). Besides, gold is more or less solu- ble in a great many other salts (T. Egleston). G. F. Becker has shown the solubility of gold in alkaline sulphides, and the solubility of the sul- phides of Hg, Fe, Cu and Zn in either sodic sulphide, sodic sulph-hydrate or sodic carbonate, partly saturated with sulphuretted hydrogen. Sil- cate of gold, the existence of which was first suggested by G. Bischof, has been frequently mentioned as probably contained in mineral waters; * Mineral Resources of the United States, 1892, p. 21. + Mr A. Liversidge has recently given an interesting historic résumé of the experiments regard- ing the solubility of gold, as well as many original experiments, in the Proc. Roy. Soc., New South Wales, vol. xxvii, 1893, p. 303. XXXIII—Buuu. Grou. Soc. Am., Vou. 6, 1894. 238 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. but it should be borne in mind that the existence of this salt has never been proved. It appears that the mentioned facts are sufficient to show that the mineral waters, once circulating in the quartz veins of California, may easily have held gold in solution.* It seems of questionable use to speculate on the particular combination in which the gold is contained in the water, for, according to the views of modern chemistry, watery solu- tions, when sufficiently diluted, contain the solids in a state of dissocia- tion, so that it is uncertain whether salts of gold could exist as such in the always much diluted natural solutions. The precipitation of the solids contained in the solution could have been brought about by many means, such as diminution of pressure, dilu- tion, etcetera. The reducing influence of carbonaceous slates, so often maintained as the probable cause of the precipitation of the gold, appears of questionable importance. Veins entirely in massive rocks and far away from any sedimentary areas show too much similarity with those in such areas to attribute a paramount weight to this argument. COMPARISON WITH QUICKSILVER DEPosITs. There are certain interesting analogies between the gold-quartz veins of the Sierra Nevada and the quicksilver deposits of the Coast ranges. In the Sierra Nevada the association of minerals is native gold with predominating quartzose gangue; carbonates in the wall-rocks; next in importance, iron-pyrites with smaller quantities of the minerals of cop- per, lead, zinc, arsenic, and antimony; quicksilver-ores are occasionally present. In the Coast ranges we have quicksilver in predominating quartzose, and to some extent carbonate gangue; next in importance, * These reactions are of course by no means the only ones which are likely to take place. Itis thus very likely that the reactions established by C. Newbery (Trans. Roy. Soc. Victoria, vol. ix, p. 754) have taken place. According to him, the iron is contained as ferrous carbonate with sulphates ; chloride of gold ean be held in such very diluted solutions in presence of alkaline carbonates and excess of COs. “This is true of chloride of gold, and if the sulphide is required in solution, it is only necessary to charge the solu@on with an excess of H.S. In this manner both sulphides may be retained in the same solution, depositing gradually with the escape of the carbonic acid.” It does not seem probable, however, that sulphates have played a very important part in the chem- istry of the gold veins. The explanation of Phillips for the contemporaneous deposition of gold and pyrites (Proc. Roy. Soe. London, vol. xvi, 1868, p. 294) was that as gold is soluble to some extent in ferric sulphate, solution of this salt containing gold was transformed by a reducing agency into pyrites, the gold at the same time being reduced to the metallic state. The presence of a ferric salt in deep-seated waters would bea very unusual occurrence. The presence of ferrous sulphate, on the other hand, in solution carrying gold does not appear possible, for the latter would be imme- diately precipitated. The fact that in the gold-quartz veins silver occupies such a subordinate position would seem to lend strength to the view that the solutions once circulating in them were not adapted for the dissolving of silver compounds. While thus G. F. Becker found that PbS and AgoS were insoluble in sodie sulphide, sodic sulphydrate or in solution of sodie carbonate partly saturated with hydrogen sulphide, these salts or metallic silver may be soluble toa very slight degree when in combination with other compounds. An alloy of much gold with slight amount of silver may thus be soluble. I do not know of any experiments on this subject. Doelter, referring to the dissolving action of sodic silicate and carbonate on silver, remarks that it is “ hardly” attacked, thus implying some action, COMPARISON WITH QUICKSILVER DEPOSITS. 239 iron-pyrites with smaller quantities of copper, antimony, arsenic, and nickel; gold is very commonly present. Regarding the rocks adjoining the deposits, Mr. Becker says* they— “have in many cases been greatly modified. Metamorphic rocks often appear to have been converted into or replaced by more or less dolomitic carbonates by the action of solutions. . . . Both serpentine and the metamorphic rocks seem to be subject to this conversion.” Containing a similar association of metals, similar gangue and similar altered country-rock, it seems justifiable to express the conviction that similar mineral solutions have circulated in both classes of deposits; and, in fact, the still abundant thermal waters found in intimate connection with the ore deposits in the quicksilver region closely correspond in character to the inferred composition of the once existing hot springs of the gold-belt. They all show free carbonic acid, as well as abundant carbonates (sodic, calcic and magnesic) ; silica is always present; usually also sulphuretted hydrogen or alkaline sulphides. Carbonatization of the wall-rocks of fissure veins has neither been de- scribed by A. v. Groddeck nor by F. v. Sandberger in their researches, though the former has found abundant sericitein many. The wali-rocks of the Comstock lode, according to Mr. Becker, are rich in iron-pyrites, but do not contain much carbonates. J. H. L. Vogt has shown that along certain veins of Norway the granite and gneiss are altered to pro- ducts resembling the “ greisen ” of the tin deposits. ORIGIN OF THE GOLD. Regarding the origin of the hot, auriferous solutions which have pro- duced the gold-quartz veins it is best, at this stage of our knowledge, to speak with great reserve. Even the results of assays or analyses of country-rock must be received with the greatest caution, to make sure that the percentage discovered is primary constituent and not later im- pregnation. It is not to be denied that many reasons speak strongly against a derivation from the surrounding rocks. Thus, for instance, the diorites of Nevada City and Grass Valley contain an appreciable amount of barium, and still there is no trace of barite in the veins of those locali- ties. In another instance the diabasic rocks of the same region contain copper, and yet the gold veins passing through these rocks are remark- ably poor in copper minerals. In discussing this difficult question there are several broad facts which must be borne in mind: First, that the gold-quartz veins throughout the state of California are closely connected in extent with the above-described metamorphic series, * Monograph XIII, U.S. Geol. Survey, p. 392. 240 W. LINDGREN—CALIFORNIA GOLD-QUARTZ VEINS. and that the large granite areas are almost wholly void of veins, though fissures and fractures are not absent from them. : Second, that in the metamorphic series the gold-quartz veins occur in almost any kind of rock, and that if the country-rock exerts an influence on the contents of the veins, it is, at best, very slight. Third, that the principal contact of the metamorphic series and the granitic rocks is in no particular way distinguished by rich or frequent deposits. It is further apparent that gold deposits have been formed at different periods, though by far most abundantly in later Mesozoic times. Some of these later veins may have been locally enriched by passing through earlier impregnations in schist or old concentrations in the sandstones and conglomerates of the metamorphic series, the gold contents of which have, however, only been proved in isolated cases. These considerations, though involving many most difficult questions, strengthen the belief that the origin of the gold must be sought below the rocks which now make up the surface of the Sierra Nevada, possibly in granitic masses. underlying the metamorphic series.* SUMMARY. The auriferous deposits extend through the state of California from north to south in an irregular and broken line. | The gold-quartz veins occur predominantly in the metamorphic series, while the large granitic areas are nearly barren. The contact of the two formations is not distinguished by rich or frequent deposits. The gold-quartz veins are fissure-veins, largely filled by silica along open spaces, and may dip or strike in any direction. The gangue is quartz, with a smaller amount of calcite; the ores are native gold and small amounts of metallic sulphides. Adjoining the veins the wall-rock is usually altered to carbonates and potassium-micas by metasomatic processes. The veins are independent of the character of the country-rock, and have been filled by ascending thermal waters, charged with silica, car- bonates and carbon-dioxide. é Most of the veins have been formed subsequent to the regional meta- morphism which affected the auriferous slates and the older igneous rocks associated with them, and also subsequent to the granitic intru- sions which closed the Mesozoic igneous activity in the Sierra Nevada. * Mr Becker, reasoning from analogy, has some time ago suggested such a derivation: ‘* Quick- silver Deposits,” p. 449.. Militating against this view is the general absence of compounds of boron and fluorine so often occurring in ore deposits in granitic rocks. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 241-262 MARCH 21, 1895 CRYSTALLINE LIMESTONES, OPHICALCITES AND ASSOCI- ATED SCHISTS OF THE EASTERN ADIRONDACKS * BY J. F. KEMP (Read before the Society, December 28, 1894) CONTENTS Page Eemapsicvof authors previous! investigations. ./20). 0.52... se ee ce ee cee eee 241 SHOES Oe OVEN Oe 9 ole) Ge ea Hie a ious See Aisi tee a os ie eg Seeman 242 Geographic distribution of the rocks.......... Sea tee Ne gk ep alee 242 Previous and contemporaneous work in the region...... BR Me, oer ee a2 243 MireMiMtesi anes Olathe FCOION 0h qen ohare aie Sak a awe ee es eet ged ed aks 245 eerie cot GUIS EM OUNGLONs =a Meta te eS aL Ne Bical. Amel tuare lhe eos iad mags alts 245 Mode of occurrence and associated rocks.............. RS Ac nae bol 245 ee MMeSFOneS OF MSSOX: COMMU ne)a00)5 6 leas iri ae so ee se ne fa a Bielees spn isuar'e soe 246 Location, extent, relations and mineral constituents..................... 246 MHeEEOcHElemry by pe locality...i4 tes 2ny eo ie lees Ye PL 249 GemeralvdeseriptiOmis i) .//2hiaess< ie os BD CWE Be Fae SWI Sis SVG SO WR 249 WKOSS-SECHIONGS sh ...08f0240 02. IES er eet Seeds ares ke Uy cae Tf 249 SMO liy, ANG: MUIMCKALO GY. s:. 2c Guys «ea aie Saisie Be wh wee age 8 Des eal 252 The limestone......... te spanned pe Seta By i aM Ape ae 252 athe opinicaleites!..)st24 1.0) Sig. so Seem B a SSO pe Ws ae o % 6 o_o do 2 > ho 1 MILES. CLINTON COUNTY, NEW YORK. Me ie | ( whee, Oe L \ fi ie = Re ha NOERS Tea aiane. BULL. GEOL. SOC. AM. [eneln eo fs ayo pe e( offs »} Tofeote eo > = ole ole «fo Faults. Pleistocene, THROWN. HEAVED —— Chazy. a4 Trenton, E=>3 calei E> Calciterous. = Pots dam, cue ae sre Secels =) ae oste cy I —— ste 2 Le WN SHIP. GEoLoGic MAP OF CHAZY TO SLINTON COUNTY, NEW york. ae Pe a) | A We ea i i ‘ape ‘a BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 285-296, PL. 12 MARCH 23, 1895 FAULTS OF CHAZY TOW NSHIP. CLINTON COUNTY, NEW YORK BY HENRY P. CUSHING (Read before the Society December 27, 1894) CONTENTS Page Situation and general features of the district............... OC TUS: ERAT Re eae 285 eMsHeR Tams Tepresented: <0... oka lwce. oe Reb na ace Seles smaee ete h ease 286 ei TELUS ge A a ei ae CHER Sa aN NAN, Lease ede eae 287 Their prevalence in the region............ Pee ee cd gee Rv Reet ue ha eee es ee 287 Hewes ale classitication of the faults: 5,052 <6. es Seles hee see eke. 288 erence OMe maT Si OlASGEs Sel coynr ey nce ne alee Uncen agais cits abe aleu lau ead eco 8 288 TP -SUTUG Jeno ees a ave ih Ae Se ee Ri Oa alge nes Oa 288 Tegel CeO ive iais Ceeneteene Or Peneiciay Oh ad estrone ie ey aE, ee Peg po Pe 289 SE alglapin ny ae eR nee ye PRU eee ee yl th, hl A SG ea geha 290 ae arnamary-On faults of the first class.) s4e1.<00 o2. 3. cece oe ose 290 PM srOmt he SCCONG CLASS) i... cere 0c oR8 Hs Sei evs Oos8 i icicFa PMO ne te ROLE a 291 Lenellvs Te dCs ste Ssh cet a ae ae a Os 291 Se Pen aa es epee ar ee em eine cc elt ems ie iste asian wig eter e ars, d'a)are Myo attieu om 291 22 SEN LDEID). glee Se AAS aes al atari Bc eR ee cA oncaeid tempest 291 SPRUCE NEN ate ee det Saree Meee ke te ak 292 Summary of faults of the second Hes EROR Re re acte thay iat Sth SINR UM 292 ET SS DIE WOES Te OTC BCCI SISTA petites Nee esse str OES Rick ens aN oa a 292 LST, LBs] Oy 02 iaiae se ee pl es Sekai ee an en Be 292 Mae tar ema Sb eres Pegg ane cee A teehee delle ad rast Whe GSP oad ay coaicn/e ote Sosid foe aie Hales 292 Mae AM nr Camerata tr ere oe Ae ines 2 ts Ie aca al ave g seal aie Sal's ee dues 293 2 TORING Jo i=l al erased coceee a8 so cc aes Bees Ne ta. gee: eee rE 293 Os ell, 6. 5 loea Shor atits ga case wee RD ae ae oleae Ici oe eg a eC a 293 RSIS eT eck art eck 9 cab Gai nine Siac EEE, 0c CG An ae re ae 294 PeUMTEMACEN AOU at Me TAUNES sates cae 2h iec.5 vated) sysia 2a alacctee les Hews ade BORE Y ole . 294 DULLED’ DIRG! OBL OWS) TEE ASIN SPR "Stet Sate cc ne eae 295 Non-appearance of the Calciferous at Chazy village..........0.. 0.0000. c es 295 oy) SEBO IDS ws loans Gell Gia a Aa RR a ee Ue gedh de219) 8) SITUATION AND GENERAL FEATURES OF THE DISTRICT. Chazy township is situated in the northeastern part of Clinton county, bordering on lake Champlain, with Champlain township intervening between it and the Canada line. The county is separable topographically XL—But.. Grou. Soc. Am., Vou. 6, 1894. (285) 286 H. P. CUSHING—FAULTS OF CHAZY TOWNSHIP, NEW YORK. into three divisions—a hilly region in the southwest, a high plain sloping north from the hills, and a strip of lowland of varying width along the ‘lake-shore, whose rise to the level of the high plain is quite abrupt. Chazy township belongs, for the most part, in this latter subdivision, but in its western portion the level rises rapidly to that of the high plain to the west. The minor topographic features of the township are dependent to a surprising degree on the faults to be described. These features are more or less obscured by drift, but in the vicinity of Chazy village this cover- ing is so much less marked than elsewhere along the low strip that the relations can be readily worked out. The dip here is frequently pronounced, so that the strata outcrop in linear ridges whose crests are the cut-off edges of the various more resistant layers. When followed along the strike, these ridges are found to be sharply intersected at vari- ous intervals by low marshy tracts, beyond which other ridges appear in like manner, but do not correspond with their predecessors. Many of the lines of faulting to be described are occupied by marshes of this char- acter. Furthermore, the main streams occupy fault-lines to a very great extent. THE FORMATIONS REPRESENTED. Leaving out of consideration the Pleistocene deposits, the geologic formations exposed at the surface in the township may be tabulated as in the adjacent table. Of these formations the Chazy and the Black River limestones are well exposed, are everywhere fossiliferous and were therefore very serviceable in working out the stratigraphy. This was true in a marked degree of the Black River limestone, which, due to its slight thickness, furnished an especially valuable datum point wherever it appeared. On the other hand, the Trenton and Caliciferous are but meagerly and unsatisfactorily exposed, while the frequent outcrops of the Potsdam are of little serv- ice, owing to the lthologic similarity of the mass, the scarcity of fossils and the lack of exact knowledge of its total thickness. This latter must itself be very variable in the region, as the Potsdam is a shore deposit laid down on an uneven floor. One exception must be noted to this general statement concerning the Potsdam, namely, that the passage- beds, separating it from the Calciferous are lithologically distinct from either, and it is believed furnish a recognizable stratigraphic horizon. Mention should be made here of the admirable stratigraphic work done on the Chazy in the vicinity of Chazy village by Messrs Brainard and Seeley, which served as a starting point for this investigation, and to * Am. Geologist, November, 1888, pp. 323-330. Formation. Trenton ..<.. . TABULATION OF THE FORMATIONS. 287 Formations exposed in Chazy Township. Character. 8 Black, slaty limestone, highly fossiliferous at certain hori- zOns. Black River... Black, slaty limestone, more massive than the overlying Trenton; sparingly fossilifer- ous, but with a characteristic fauna. Massive beds of limestone, the whole separable into a lower, a middle and an upper divis- ion, each well marked litho- logically and _ paleontologic- Massive, gray, siliceous lime- stones and dolomites, which Thickness. Unknown in the Lake Cham- plain region, but clearly sev- eral hundred feet at least. From 30 to 50 feet. About 750 feet. The lower, middle and upper divisions have a thickness of 350, 200 and 150 feet respectively.* Not known for this vicinity. At the head of jake Champlain the formation has a measured thickness of 1,800 feet. t Chay cis... ally.* Calciferous.... are sparingly fossiliferous. Potsdam, 2... Massive beds of sandstone, of red, white or yellow-brown color and varying degree of induration, with conglomer- ate and arkose at the base, and passage-beds to the Cal- ciferous at the summit. THe FAULTS. Unknown, but at least several hundred feet. Mr Walcott has measured sections of 250 and 350 feet respectively at Cha- teaugay chasm and Ausable chasm, which represent mere- ly unknown portions of the whole. THEIR PREVALENCE IN THE REGION. Wherever any bit of the region bordering on lake Champlain has been mapped in detail one or more faults have been disclosed. Sometimes the fault-line itself is shown in section, but more commonly its presence is indicated by the structure merely. The number of faults already mentioned or mapped in the region must be 20 or 30, yet they probably represent only a small proportion of those which exist. While faults * Brainard and Seeley: Am. Geologist, November, 1888, p. 324. } Brainard and Seeley: Bull. Am. Mus. Nat. Hist., vol. iii, no. 1, p. 2. tC. D. Walcott: Bull. U. S. Geol. Survey, no. 81, pp. 348, 344, 288 H. P. CUSHING—FAULTS OF CHAZY TOWNSHIP, NEW YORK. abound, in general folds are absent. This is true in a marked degree of the region at the lower end of the lake. At the upper end of the lake in Vermont, as the line of the great thrust fault is neared, the rocks are found to be folded as well as faulted.* Away from this vicinity, how- ever, folds do not appear, or else are extremely gentle. In each fault- block the dip is steadi!y and persistently in one direction, and the vari- ous strata follow one another in regular order with no repetition. FEATURES AND CLASSIFICATION OF THE FAULTS. In none of the faults occurring in Chazy township is the fault-plane itself open to inspection. Elsewhere on lake Champlain, however, fault- planes are visible, and when this is the case they are seen to be nearly or quite vertical. Attempts to determine the hade of the Chazy faults geometrically give very unsatisfactory results, because data of sufficient precision are not to be obtained, but an approach to verticality is sug- gested by them without, however, indicating whether such hade as exists is to the up or the down throw. There is at the present time no evi- dence at hand to show that in respect to hade and throw there is more than one class of faults in the region. In going from end to end of lake Champlain a succession of faults produces a frequent repetition, in whole or in part, of the lower Paleozoic series. Locally many faults have been mapped, but no attempt has as yet been made to trace the faults from one place to another in order to determine their extent. It is indeed questionable whether this is possi- ble in any considerable measure. In that portion of the region directly under consideration the faults may be conveniently grouped in three classes, but there is as yet no evidence that this grouping may be applied to the region as a whole. For the purpose of description, they may be referred to as of the first, second and third classes, and the considera- tion of their differences will more appropriately come in after the faults themselves have been described.t FAULTS OF THE FIRST CLASS. Fault A-A.—The most pronounced structural feature in Chazy town- ship is the great fault whose position corresponds closely with the line of Tracy brook from a point one mile north of West Chazy village, where * See Brainard and Seeley: Bull. Am. Mus. Nat. Hist., vol. iii, no. 1, pp. 8, 9. + Some references to faults in other parts of the Lake Champlain region are appended. The list makes no pretense of being complete. kK. Emmons: Geol. of New York, vol. ii, p. 274. Brainard and Seeley: Am. Geologist, November, 1888, p. 326. Brainard and Seeley: Bull. Am. Mus. Nat. Hist., vol. iii, no. 1, pp. 5, 8, 11, 15, 18, 21. C. D. Walcott: Bull. 81, U.S. Geol. Survey, p. 344. T.G. White; Trans. New York Acad. Sci., vol. xiii, p. 225. FAULT A-A. 289 the brook turns at an abrupt angle into the fault line, to Chazy village, where the brook joins the Little Chazy river. Beyond Chazy village the river leaves the fault-line, but the railroad closely follows it to the town- ship line. The fault can be first recognized north of West Chazy, where the road crosses Tracy brook, and here ledges of Potsdam sandstone are exposed on the west side of the stream, while only a few rods away to the east, with very different strike, are beds well down in the lower division of the Chazy. To the south the further extension of the fault is concealed by a heavy covering of drift. To the north, however, it is readily traceable throughout the township. The heaved block on the west brings Potsdam sandstone to the surface in all outcrops. On the east the thrown block is much broken, especially in the vicinity of Chazy village, so that ledges of varying age abut against the fault-line, the range being from the lower Chazy up to the Black River limestone. Every vestige of the Calciferous is faulted out. This fault is traceable, with a great degree of probability more than half way across Champlain township next north of Chazy, giving it a known length of from eleven to twelve miles. In that township the Calciferous comes in, conformably overlying the Potsdam, on the west side of the fault. The vertical throw of this fault cannot be accurately determined, owing to lack of knowledge concerning the thickness of the Calciferous here. Brainard and Seeley have shown that at the upper end of lake Champlain the Calciferous has a thickness of 1,800 feet, and that, while toward the lower end of the lake no complete exposures are found, the different members of the formation hold their thickness pretty persistently.* If this is assumed to be its thickness here, the fault has a throw of about 2,000 feet at Chazy village. The character of the rock exposed there on the west side of the fault is that of the passage-beds between the Potsdam and Calciferous, while on the east side are beds of the lower division of the Chazy, so that the throw of the fault is not much in excess of the thickness of the Calciferous. In their work at Chazy village Brainard and Seeley recognized this fault and fully realized its importance.t Though they did not attempt to plot it or trace it for any distance, their map indicates its course for a mile and a half southwest from the village. The fault was also recog- nized by Mr Walcott when there, though he seems to have regarded its throw as of small amount and accounts otherwise for the non-appear- ance of the Calciferous.t Fault C-C_—At a varying distance of from one to two miles east of fault A-A is another great north-and-south fault, roughly parallel with it. * Bull. Am. Mus. Nat. Hist., vol. iii, no 1, pp. 1-23. + Am. Geologist, November, 1888, p. 327, and Bull. Am. Mus, Nat. His., vol. iii, no. 1, p. 13. t Bull. 30, U.S. Geol. Surv., p. 24. 290 H. P. CUSHING—FAULTS OF CHAZY TOWNSHIP, NEW YORK. It has been traced for a distance of two miles and a half, but as it then passes at both ends into country heavily drift-covered, further tracing is impossible. Between it and A-A lies the zone of much faulting. To the east of it is a tolerabiy regular and nearly continuous section rang- ing from the upper Calciferous to the Black River hmestone, with a dip and strike which correspond with those in the Potsdam west of A-A, while the varied dips and strikes in the shattered zone between the two faults bear no apparent relation to either. If, however, these two faults are considered as genetically connected—that is, as forming practically one fault, with the zone between regarded as merely an unusually wide, crushed strip—the apparent lack of relationship is at once explained. Both faults are dip-faults and the combined throw is about that esti- mated for the fault A-A, or 2,000 feet. The Little Chazy river follows the fault-line C-C for two miles, then, near Chazy village, passes from it to fault H-H, which it follows to A-A, where Tracy brook joins it. Fault B-B.—This is a third great north-and-south dislocation, which is first discernible at a point about a mile and a half east of Chazy village, north of which point outcrops are too few and meager to permit of fol- lowing it further in that direction. Its trend, however, suggests a possible junction with the fault A-A somewhere to the north. Where first recog- nizable the higher beds of the middle division of the Chazy are exposed on the west side of the fault-line, and outcrops of an horizon somewhere in the Trenton on the east side, these last lying at a level 100 feet lower than that of the beds across the fault-line. Outcrops on the east side of the line are few and far between, but the fault can be traced with tolerable certainty to the shore at Monty’s bay, where it passes beneath the lake. It reappears on the opposite side of the bay and is traceable all the way across Beekmantown township, next south, into Plattsburgh township, a distance of between ten and eleven miles. Throughout Beekmantown township the entire Chazy is faulted out along this line. In north Beek- mantown an horizon low down in the Trenton is exposed on the east side of the fault, and an unknown horizon in the Calciferous on the west side. Assuming that the thickness of the Chazy here is the same as in Chazy and Plattsburgh townships—that is, in the neighborhood of 750 feet—the throw of the fault here is an as yet unknown amount in excess of that. At the Plattsburgh-Beekmantown line the fault seems to divide, enclos- ing a much faulted block of Chazy and Black River limestone between the Calciferous and Trenton. Still farther south, at Bluff point in Platts- burgh, the Trenton is faulted down against the Chazy on the prolonga- tion of this fault-line. If this prove to be the same fault its known length is increased an additional five miles. Summary of Faults of the first Class—The three faults just described FAULTS L-L, M-M AND D-D. PASM are the main dislocations of the vicinity. They differ from the remain- ing faults of the region, so far as the evidence goes, merely in magnitude. They are the main lines of displacement; the other faults are but minor breaks in the faulted blocks into which the great breaks divide the region. In this vicinity these great faults have a roughly north-and-south direc- tion, but in the region as a whole there is no probability that this will be found to be the case. FAULTS OF THE SECOND CLASS. Fault L-L.—Two sets of facts indicate a break along the line L-L. First, the beds of middle Chazy age on both sides of the line have a much greater width of surface outcrop than their thickness entitles them to at the measured angle of dip; second, the angle of dip is abruptly increased along that line, otherwise being quite persistent on either side. It is probable that there is also repetition of a part of the series, but more detailed work than has yet been possible is necessary in order to furnish the proof. The fault is a strike-fault with downthrow to the south, and a throw certainly less than the thickness of the middle Chazy—250 feet, and probably considerably less. Fault M-M—North of L-L is another fault of the same character, bear-. ing to the northeast instead of the southeast, and with the thrown block on the north instead of the south side. At the west end of the fault Black River limestone is exposed on the thrown side; beds of the middle division of the Chazy on the heaved side. The missing strata are at least 200 feet thick, and the dip and strike on the opposite sides of the fault- plane are quite unlike. At the eastern end of the fault middle Chazy beds on the south are brought up against Trenton on the north, the exact horizon in the latter being unknown. ‘The Trenton limestone along lake Champlain has not yet been carefully studied. That the throw of the fault increases in amount from west to east is, however, clear from an inspection of the dip and strike on the two sides. Fault D-D.—While it is uncertain whether this supposed fault belongs to the second or third group, as herein classified, its description may be conveniently given now. The line of the fault is occupied by a wide marsh, so that an interval of half a mile separates the first outcrops on one side from those on the other. North of the marsh are beds of lower and middle Chazy age in normal relations to one another and with a strike normal to the line of the marsh. On the other hand, south of the marsh are found beds of lower Chazy age alone, with different strike and dip. Itshould be stated that in the triangular section of country enclosed between the faults A-A and D-D the relations are obscure and not yet fully worked out. Out- 292 H. P. CUSHING—FAULTS OF CHAZY TOWNSHIP, NEW YORK. crops are tolerably plentiful, disclosing no rocks of other age than lower Chazy. Accurate determinations of dip and strike are attended with much difficulty, but the data at hand suggest that faults are present. In the state of our knowledge the relations north and south of the marsh are best explained by the presence of a break of some sort somewhere beneath the marsh, with a downthrow to the north. Supposed Fault N-N.—If the beds of lower Chazy age exposed just east of the southern end of fault A-A were prolonged southwestward in the direction of their strike they would pass to the west of the Calciferous exposures at West Chazy, which have an easterly dip. These Chazy beds, however, are tolerably near the fault-line of A-A, so that their in- clination and strike may be merely local. A fault along the line N-N would make the relations between the two clear, but the evidence of its presence is slight. Summary of Faults of the second Class—Faults like L-L and M-M seem to be distinguishable from those of the first class in their inferior magni- tude and extent. The evidence at hand indicates that they are subor- dinate to the main faults, and represent breaks in the blocks lying be- tweenthem. If this conception be the true one, these faults of the second class should be limited each to a single main fault-block, should be cut off at the great faults and not pass across them. Proof that they are really so cut off is difficult to obtain, though the negative evidence all points in that direction. FAULTS OF THE THIRD CLASS. Fault E-E.—The thrown block of the great fault A-A has suffered much minor dislocation, being separated into a series of small blocks by a number of faults, which are roughly normal to A-A. Near the fault- line, however, the confusion is so great as to preclude detailed mapping. As the township is entered from the north along A-A, fault E-K, the first of the series, is met. North and south of the fault are ridges of middle and upper Chazy beds followed by the Black River limestone. The fault is a dip-fault with a heave of about 400 yards, the beds on the south side of the fault lying about that amount further east than the corresponding beds to the north. The amount of heave, taken in con- nection with the angle of dip, indicates a vertical throw for the fault of from 200 to 250 feet, with the downthrow on the north. Fault F-F—The ridges of rock south of fault E-E extend but a short distance southward, and are then abruptly cut off along the strike at the line F-F. Just south of this line outcrops are absent, except for a con- siderable knoll of Black River limestone which lies close to the railroad and very near fault-line A-A. It lies a full mile to the west of the Black FAULTS G-G, H-H AND I-I. 293 River limestone exposures north of F-F, so that a fault of great throw is indicated if this outcrop is taken as indicative of the normal relations along the fault-line. There are, however, at least two other outcrops of Black River limestone, and perhaps more, which occur at points farther south along fault A-A, between the Potsdam west of the fault and the Chazy east of it, and it may be that this outcrop under consideration should be classed with them as an extremely aberrant mass merely in- dicative of the great shattering which has taken place along A-A. If this be accepted as the true explanation, the only evidence we have of a fault at F-F is the sudden disappearance at that line of the ridges of Chazy limestone which lie north of it. | Fault G-G.—To the south of this knoll of Black River limestone no outcrops have been noted for a distance of half a mile, when the section exposed along the river at Chazy village, and in the village itself. is reached—a section showing the middle and upper Chazy followed by the Black River. The absence of outcrops makes the presence of a fault here purely conjectural, and the data at hand could be equally well ex- plained by the presence of a synclinal trough running from fault F-F to the outcrops at Chazy village. The fault is preferred as the explanation merely because no other fold of the kind is known in the township. Fault H-H.—Brainard and Seeley’s detailed map of the vicinity of Chazy village commences on the north at the river section just men- tioned, and extends thence for a mile to the south and a mile and a half to the southwest, showing faults H-H and I-I.* The map is exact in all respects, and may be profitably consulted for details. The fault-line H-H is occupied by the river for a portion of its length. Beyond the point where the river leaves it, its course is clearly indicated by the abrupt cutting off of the ridges of rock on the opposite sides of the line and by the heave + as well as by the change in amount of dip. The Black River limestone south of the fault lies 150 yards west of the same stratum north of the fault, as measured normal to the strike, or 250 yards distant along the fault-line. The south is therefore the thrown block, and the vertical displacement is not far from 200 feet. Fault I-.—The strike swerves somewhat toward the west as this fault is approached, and the fault itself has a more nearly north-and-south trend than the others of its class. The Black River limestone west of the fault is heaved 250 yards to the south, a greater lateral distance than the heave of H-H, but the dip is correspondingly less. A throw of about the same amount as that of H-H is thereby indicated, but in the reverse ~ direction, the northeast being the thrown block. In other words, the * Am. Geologist, November, 1888, p. 326. + Geikie: Text-book of Geology, third edition, p. 553. XLI— But, Grou. Soc. Am., Von. 6, 1894. 294 H. P. CUSHING—FAULTS OF CHAZY TOWNSHIP, NEW YORK. block between H-H and I-I is a wedge thrown down between the adja- cent blocks to the north and south, these last being in substantial accord. Their continuity has not been affected by the faults, which have simply thrown down the intervening block. Fault K-K.—At the extreme southeastern limit of the area shown on their map, Brainard and Seeley met and noted evidence of this disloca- tion. The testimony, as to its reality and position is of the same un- equivocal character as that for the other faults, and, as the Black River limestone furnishes in every case the most convenient horizon for use in defining the fault, the corroborative evidence of other horizons may be omitted from the discussion. The Black River, south of the fault, lies 85 yards to the east of the same bed north, measured normal to the strike. This, together with the low dip, indicates only a slight throw for the fault—40 feet as a maximum, with the throw to the south. SUMMARY OF THE FAULTS. From the preceding descriptions it may be seen that the assemblage of faults E to K, inclusive, are peculiar in certain respects. The zone lying between the great faults A and C is greatly shattered, not only absolutely, but also when compared with the rest of the area under discussion. The zone between faults B and C will answer best for comparsion. Though the work on that zone is somewhat incomplete, it is evident that it has suffered far less from faulting than the other. As has already been indi- cated, the disturbance around Chazy village is greater than can be dis- played on a map of small scale. Witness the outcrops of Black River limestone along fault A-A between the Potsdam and the Chazy.* Two possible explanations of the disturbed zone at Chazy village sug- gest themselves to the writer. One has already been hinted at, namely, that faults A-A and C-C form a sort of double fault, or,in other words, may be considered as one fault with a crushed zone of unusual width.f The second explanation would perhaps be a more natural one. If the three faults A-A, B-B and C-C are prolonged northward, holding approx- imately the same trends, they may perhaps come together—that is, the three faults may have been formed by the subdivision of one fault. What seems a similar case is exhibited north of Plattsburgh, where the fault B-B separates into two branches, bringing up a wedge of Chazy between the Calciferous and Trenton.{ This wedgeis much shattered and broken in a manner quite similar to that in the region around Chazy. Brainard and Seeley’s map of Providence island and the neighboring portion of South Hero seems to illustrate a similar case.§$ * See ante, p. 293. + See ante, p. 290. t See ante, p. 290. 2 Bull. Am. Mus, Nat, Hist., vol. iii, no. 1, pp. 18, 21. STATUS OF THE INVESTIGATIONS. 295 The faults of the third class may be characterized as minor breaks produced abundantly in narrow zones, either between two branches of a fault of the first class or else between two faults of that class which ap- proach rather closely and seem related to each other. They are confined to the block between the two faults, being in that respect hke the faults of the second class. They may perhaps be better regarded as a mere phase of the faults of that class, produced in unusual number under local circumstances. OTHER PROBABLE FAULTS. The facts here set forth have been incidentally noted by the writer while engaged as assistant to Professor J. F. Kemp in mapping the areal geology of the region. The results are therefore incomplete, portions of the area requiring more detailed work than it has yet been found possi- ble to bestow on them. There are indications of the presence of. other faults than those here noted, this being more especially true of the south- ern part of the township, where the apparent great extent of the Calcif- erous and the lack of other formations implies a considerable amount of faulting. Some of the faults already mapped require further study for their proper elucidation. They are sufficiently well worked out, however, to answer the purpose for which the paper was written. NoON-APPEARANCE OF THE CALCIFEROUS AT CHAZY VILLAGE. The theory has been advanced that the Calciferous is lacking at Chazy village because of non-deposition.* Asa result of their work in the vicinity, Brainard and Seeley maintained that the Calciferous might be thrown out by a fault along Tracy brook, that such a fault existed, and that the great disturbance of the rocks at Chazy village was the result of said faulting.f With the latter explanation the writer fully agrees and would urge in its favor that the present relations existing there be- tween the Potsdam and Chazy are clearly the result of disturbance and give no clue whatever to their possible relations prior to the disturbance; that a great fault (A-A) separates the two, and is in itself sufficient to account for the absence of the Calciferous; that at the only locality in the township where the structure permits of the outcropping of the beds which he beneath the Chazy the Calciferous appears; that both to the north and south within from two to four miles the Calciferous is present in such force that the warping necessary to produce the supposed ces- sation of deposition must have been local in the extreme, and finally *C. D. Waleott: Bull. 30, U. S. Geol. Survey, p. 22. + Am. Geologist, November, 1888, p. 327. 296 H. P. CUSHING—FAULTS OF CHAZY TOWNSHIP, NEW YORK. that in the township next south a precisely similar fault (B-B) hides the entire Chazy limestone from view throughout the township, though it reappears again in the next town beyond in the same strength which it exhibits in Chazy township. CONCLUSIONS. In the rocks of Ordovician age occurring along lake Champlain the detection of faults is not a matter of great difficulty, but back from the lake in New York the criteria which avail for their determination in the Ordovician rocks are not furnished by the older Cambrian and _ pre- Cambrian strata.* Yet that both are much faulted is certain. It has just been shown that a large proportion of the contacts between rocks of different age in Chazy township are fault-contacts, and the same may be shown with equal readiness in most other townships on either side of lake Champlain. It is believed that this will be found to be true also in the Adirondacks themselves, but their discrimination from ordinary deposition-contacts will be extremely difficult. As an illustra- tion: Wherever in Clinton county the writer has found demonstrable deposition-contacts between the Potsdam and the older rocks a basal conglomerate or an arkose or both are shown. Where these are absent the relations are often such as to strongly suggest contact by faulting. At other times no indications whatever of the character of the contact are afforded. That the crystalline rocks of the Adirondacks are also faulted can be often shown by means of the numerous dikes and of the beds of iron ore. The topography is often such as to strongly suggest faulting. Certainly the possible presence of faults must be constantly kept in mind when endeavoring to interpret the stratigraphy of that re- gion, and the main purpose of this paper is to emphasize this fact, in view of the work now being prosecuted there. * The term pre-Cambrian is preferred at present for these rocks, as it is purely non-committal. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 297-304, PLS. 13-15 MARCH 25, 1895 HONEYCOMBED LIMESTONES IN LAKE HURON BY ROBERT BELL (Read before the Society December 29, 1894) CONTENTS Page Pea GMOLCHrrence ANG CONGIMONS, 5.02% 25.20 0es so de Shido eter eases cules 297 iwciedwenaracceristics Of the eroded TOCKS. ../... v.06 65005 ole oe eo ee hee we 298 mec amalaitiiude of the eroded rocks. . 02. 80.. bele Vanes naet fee eee ees Nena 209 Bowen forms in relation to variety of rock ..... 22.2.0: spe ea ne ad seed tested 299 eceble Or oanis OF GO CVOSIOM: 66:5 a 0d 5,0 ps mes © be bbls HE GrE HOLA ee + tate we OE 300 Hhornes of mollusks...) 4<0%.% OP eee Mice Sih Uh Lt) ooh aaa aug eee 300 PPE GMO SPOUCES ANG AGE: Gas .c 6 ayia cs od bs 'gs a erap bo ep Reda alae 4 Bis ee 301 aM CHMONKOL. DE MDIES 32 cece ax oi4 ie beso odie’ «j50 BEMAA we 48 2d yee oes eles wb 301 Neia water the probable cause of the erosion”..............08.-.ceedetseeees 302 Sulphuric acid in the water and its source ...............2.. Peas ahes WON UC 303 REE MOTIS ES, heed 8a a Neh hs kre Sige tb yield bl alee edo dudes dls enl sl ie OOS AREA OF OCCURRENCE AND CONDITIONS. ~ The limestones in the bottom of a certain portion of lake Huron are undergoing a peculiar kind of erosion, which, from want of better terms to describe the process, may be called honeycombing and pitting. The portion of the lake in which this phenomenon is most prevalent is that around Grand Manitoulin island, the Indian peninsula and in Big gap, which lies between them and connects the main body of the lake with Georgian bay. This curious form of erosion appears to be progress- ing most rapidly under a considerable depth of water, say, from 50 or 100 feet, down to greater depths, but it may also be going on in shallower water. The existence of the honeycombed limestone all over the bottom of this part of lake Huron is well known to every one living in the vicinity, and especially to the fishermen, to whom it is a great source of annoyance from their nets catching upon it. Visitors carry away speci_ mens of it every summer as curiosities, but, so far as the writer is aware, no one has yet described its occurrence, attempted to explain its cause, or reported a similar condition elsewhere. The phenomenon therefore appears to be rare, if not unique, and worthy of a description from a geologic standpoint. XLII—Bort, Geot, Soc. Am., Vou. 6, 1894, (297) 298 Rk. BELL—HONEYCOMBED LIMESTONES IN LAKE HURON. Grand Manitoulin island is 80 miles in length. Along its southern side the undisturbed Silurian limestones slope very gently southward under the lake, so that shallow water extends a considerable distance from shore. On surfaces which have been long exposed to the weather or to the wearing action of the waves the pitting is more or less effaced, but wherever they have been covered by the water or otherwise sheltered it is still plainly visible. In the autumn the whitefish come into shallow water to spawn, and the fishermen say that the rough limestone bottom, here described, is a favorite resort for this purpose. As its inequalities protect the eggs from destructive currents and predaceous fishes, it may be said to serve an economic purpose. PHYSICAL CHARACTERISTICS OF THE ERODED Rocks. Various kinds of limestones have been acted upon by this peculiar form.of erosion. The unaltered varieties have been completely riddled with cavities, varying from very small holes up to three inches or more in diameter, the average being between half an inch and an inch, but in altered or crystalline limestones or dolomites the pits are mostly larger and on. In unaltered dolomites one form assumed by the cavities is globular © to pear-shaped, and in the process of enlargement they encroach upon one another until only thin walls remain between them, while others» coalesce, and ultimately the whole mass becomes separated into a highly eroded skeleton, as shown in figures 1 and 2, plate 13, taken from photo- graphs. Even in this stage, the solid angles between adjacent cavities become perforated by smaller holes, and at length the rock crumbles to fragments, with deeply hollowed concave surfaces. The removal of such a wasted exterior exposes a pitted surface on the next lower layer of lime- stone, the contiguous cup-shaped hollows usually occupying the whole area. A completely sculptured surface resembling this may also be pro- duced by the direct solvent action of the water without the intervention of the globular honeycombing.. Large areas showing surfaces of this kind, which had been eroded at a considerable depth, may now be seen in the clear, shallow water or exposed to the air, owing to the lowering of the level of the lake, while wide borders of nearly horizontal limestone beds, similarly eroded, are exposed in some localities between the mar- gin of the water and the wooded shore. These extremely eroded sur- faces have a striking appearance, and the sharp and pointed edges of the pits would be painful to walk on without thick soled boots. Another form which these cavities take is finger-shaped, crowded to- gether and deeply indented, or of such a shape as would. be completely filled by inserting.the sharp end ofa cigar. As in the other forms, the cavities at their tops adjoin each other, leaving no space between them. BULL. GEOL. SOC. AM. VOENG loO4 PEs tee FIGURE 1.— VIEW AL A RIGHT ANGLE TO THE BED-PLANE. Figure 2.—VIEw Or SAME SPECIMEN AT AN ANGLE OF 45° 1o THE BED-PLANE. HONEYCOMBED DOLOMITE. From South Bay Mouth, Manatoulin Island, Ontario, Canada. en a : re ay Wig th 7 eh a thet ice a hen ¥ f BULL. GEOL. SOC. AM. VOL. 6, 1894, PL. 14. Figure 1.—VIEW OF UNDER SURFACE At A RIGHT ANGLE T0 tHE BED-PLAN Fraurkt 2.—VIEW OF SAME SPECIMEN At AN ANGLE OF 45° 10 THE BED-PLANE. PITTED LIMESTONE OF THE BLACK RIVER FORMATION. From Little Cloche Island, Ontario, Canada. CHARACTERISTICS, AGE AND ATTITUDE. 299 This variety of deeply pitted surface resembles limestone which has been thoroughly riddled by the burrows of Saxicava. In a well marked case, to be again noticed, where a submerged shelf or hanging ledge was eroded on the under side, the pitting took the form here described. It is well shown in figures 1 and 2, plate 14. AGE AND ATTITUDE OF THE ERODED Rocks. The rocks of Grand Manitoulin and the adjacent islands embrace eight different formations, from the Chazy, in the northern part of the La Cloche island, to the Guelph, on Fitzwilliam island and on the southeastern part of the main island, while dolomites of the Huronian system are met with on many of the islands of the channel between the Manitoulin chain and the main north shore of lake Huron. The Chazy is represented by brownish red and green marls and fine grained white sandstone; the Black River consists of pure limestone and yellow weathering dolomites ; the Trenton principally of bluish gray limestone, with earthy beds; the Utica of black bituminous shale; the Hudson River of marls, with thin beds of limestone and sandstone; the Clinton of dolomite, with bright: red and green marls at the base, while the Niagara and Guelph forma- tions are composed almost entirely of dolomites. The dip is uniformly to the south, at a very low angle, and the naked beds of the higher forma- tions above enumerated slope gently under the lake along the southern sides of all the islands of the Manitoulin group. Erosion ForMs IN RELATION TO VARIETY OF Rock. The largest and most irregular cavities are in the magnesian lime- stones of the Guelph formation. Their appearance in situ is shown in plate 15, which is from a photograph. The globular and pear-shaped varieties, shown in figures 1 and 2, plate 18, are excavated in somewhat argillaceous dolomites of the Niagara and Clinton formations. The cup- shaped hollows are commonest in the pure limestones of the Hudson River, Trenton and Black River and in the dolomites of the last men- tioned formation. The finger-shaped honeycombed structure was found principally in the pure limestone of the Black River formation, while the smoother and rather larger excavations are characteristic of the Huronian dolomites. The shallower varieties of this form of pitting resemble that of a well eroded aérolite. The occurrence of these various forms of honeycombing and pitting in such a variety of limestones and dolomites in this portion of lake Huron proves that the phenomenon is not due to. anything unusual in the general composition or to any chemical pecu- liarity of a particular variety of rock, but to some outside cause. The various forms which the erosion takes, however; show a slightly unequal solubility connected in some way with the internal structure of the rock; 300 R. BELL—HONEYCOMBED LIMESTONES IN LAKE HURON. otherwise it would be of a more uniform character, since it is due to some common external cause operating alike upon them all. The dissolving of the unaltered limestones or dolomites goes on at right angles to the bedding or directly downward or upward, as the rocks are practically horizontal, never at an oblique angle nor horizontally, which, in the absence of some inhibiting cause, mignt easily take place in loose masses which lie at all angles on the bottom of the lake. On isolated blocks, where the sides have been as freely exposed to the water as the top, the solvent process appears to prefer to work downward from the ‘quarry bed” and not to eat inward from the sides. Experiments made by Professor Goodwin, of Queen’s University, Kingston, on the action of solvents on slowly soluble substances seem to show that the tendency of solution is strongest directly upward and downward. This, together with a faint concretionary structure, to be noticed further on, may help to account for these forms of erosion. Allusion has been made to a well marked case, where the pitting has progressed directly upward from the under surfaces of beds of pure lime- stone forming the roof of an overhanging ledge or shelf which had been submerged when the lake was a little higher than it is at present. This occurs along the east side of The Narrows between Little Cloche island and the southern part of Cloche peninsula. The rock, which belongs to the Black River formation, consists of a soft bluish grey limestone con- taining a little argillaceous matter. The tapering finger-shaped pits are closely crowded together and they penetrate upwards from two to four inches from the general outline of the roof. Silicified fossils project from the walls into these pits, and when the rock is broken across a dis- coloration of iron oxide is seen to extend a short distance all round the wall of each one. Plate 14 illustrates these pits. PossIBLE ORIGINS OF THE EROSION. BORINGS OF MOLLUSKS. The writer may not have arrived at a correct explanation of these curious forms of erosion, but he will endeavor to state the suggestions which have occurred to him in regard to their origin, along with a de- scription of the circumstances connected therewith. Many theories to account for these phenomena may present themselves both to those who have and those who have not seen them on the ground, and it may be as well here to notice briefly the more obvious ones. Cavities resembling some of those above described are made in rocks by boring mollusks, such as Sawicava, Pholas, Petricola, Callista, Tapes, Venerupis and Lithodomus.. The case of the zones on the marble pillars of the temple of Jupiter Serapis, at Puzzuoli, eroded by lithodomi, BULL. GEOL. SOC. AM. VOES Oy ISO Pen tio: GLACIATED AND HONEYCOMBED DOLOMITE OF THE GUELPH FORMATION. South Bay Mouth, Manitoulin Island, Ontario, Canada. Re ye a ey re en ae, Aes a oa ade a. vi POSSIBLE ORIGINS OF THE EROSION. 301 which has been rendered classic by Lyell in his “ Principles of Geol- ogy,” will occur to every geologist. The cavities formed by boring mollusks are, however, burrows which are of pretty uniform calibre, or increase in diameter with the depth, owing to the growth of the animal as it proceeds. They are generally many times deeper than wide, whereas those under consideration are either globular or short by comparison, and they contract instead of enlarge toward the bottom. ACTION OF SPONGES AND ALG. A small sponge, Cliona celata, makes burrows in the shells of oysters, and this fact led to the supposition that possibly some species of fresh- water sponge might have aided in deepening or enlarging the cavities under discussion, or in determining their form by the production of small quantities of an organic acid, either during life or upon decompo- sition, but no evidence could be found in support of this idea. The re- mains of branching fresh-water algee may be seen in some of the cavities, but they do not appear to have exercised the least influence in their production. The lower or jelly-like alge of fresh water do, however, possess the power of dissolving limestone. The journal of the Royal Microscopical Society for October, 1894, says, on page 597, that— ‘“Professor F. Cohn points out the important part played by very lowly organ- ized algee—Phycochromacez and Cyanophycese—in the formation of calcareous and silicious rocks. Many beds of marble and travertine have been formed in this way. He further enumerates the alge that are known to destroy calcareous rocks-by erosion. In all fixed alge there appears to be this contrast between the _ basal cells and the rest of the filament; that the former excrete an acid which dissolves lime, while the latter has the power of depositing a soluble lime-salt be- tween the filaments, but within the mucilage which is excreted from the sheath.” WEARING ACTION OF PEBBLES. A popular notion current among many of the residents of the localities where the pitted surfaces occur is that the cavities have been worn by the whirling of pebbles and sand in a manner analogous to the forma- tion of potholes at rapids and falls. That they have not been thus formed is obvious from the following reasons, selected from among others: Their shapes do not correspond with this mode of formation. They occupy the whole surface of the rock, whereas potholes occur at irregular intervals. The walls of the cavities are generally uneven or rough, and delicate silicified fossils often project from them or stretch completely across the cavities, whereas the wearing action of rotating pebbles and sand would have produced smooth and cylindrical walls. No pebbles are ever found in them except such as can be shown to have been placed there subsequent to their excavation. If these cavities 302 R. BELL—HONEYCOMBED LIMESTONES IN LAKE HURON. had been due to this cause, their occurrence would not be confined to the limestones of lake Huron, but would be a general phenomenon in con- nection with similar rocks all over the world. On the contrary, the pebble theory is not sustained by facts anywhere. Pebbles washed by currents passing over rocks or grated against them by the waves, not only do not pit or honeycomb rocks, but they have the opposite effect and wear them smooth. It is true that gravel may now be seen scattered over local pitted sur- faces which have been laid bare by the recession of the lake, but on examining cases of this kind it is always manifest that the gravel was placed upon such surfaces long after the completion of the pitting. The occurrence of, the cavities equally on the tops of loose blocks of limestone and on the solid beds is another objection to this theory. - A final objection is the fact that some of the most beautiful examples of deep pitting are found on the under surfaces of overhanging beds, the eroding agent having worked upward in a free body of water. Acip WATER THE PROBABLE CAUSE OF THE EROSION. Having eliminated all the possible causes which have suggested them- selves as being unlikely to have produced the erosion under discussion, the question arises: To what must we attributeit? Itappears most prob- able that the cause will be found in the differential solubility of the rock in the water of the lake itself. The slight difference in the solubility of those parts of the rock which give rise to the cavities is likely due to its internal structure, which originated, in the first place, at the time of the formation of the limestones or dolomites themselves. The eroded beds are not generally those which are largely made up of organic remains, but oftener those which have been due to chemical precipitation. There are, however, exceptions to this general rule which add to the difficulty of accounting for this singular phenomenon. In the process of consoli- dation of chemically formed beds in which lime, magnesia and argilla- ceous matters were present, there would naturally be more or less tend- ency to concretionary action around certain points or centers, thus giving rise to slight differences in composition. Even if such concretionary structure were too obscure to be readily noticed, the probability of its existence in the rocks of the above composition will be readily admitted ; but in some cases, especially where oxide of iron is present, this structure may be detected in the form of obscure concentric lines. The globular shape of the cavities is a fact which points to this origin. But, if this be true, it may be asked, why are similar limestones not always eroded in this manner when they are covered by fresh water? The explanation of the difference is probably to be found in a sufficiently acid condition of the water of this part of the lake to slowly dissolve the limestones. ACID WATER AND ITS EFFECTS. 303 The solvent action of the slightly acidulated water is, no doubt, aided by certain conditions favorable to it in the present case, but which might be absent elsewhere, even if the requisite amount of acidity existed. The water of this part of lake Huron is perfectly limpid and free from sus- pended impurities, so that there is nothing to check the progress of the dissolving process, no matter how slowly it may be going on. That it has been exceedingly slow in its operation is shown by the pitted or honey- combed surfaces retaining their glaciated forms. As the basin of lake Huron has no doubt been filled with water since the disappearance of the ice-sheet, the time required for the erosion under consideration must have been many thousands of years, but the proportion of acid in the water probably increased gradually, as the explanation to follow will show. SuLpHuRIc ACID IN THE WATER AND ITS SOURCE. Although the water of this part of lake Huron has not been analyzed, it appears to contain a notable proportion of sulphuric acid. It has the property of slowly reddening vegetable matter, purple or blue. Its cor- roding action on tin pails and pans is a source of annoyance and loss to housekeepers and campers who use the water for domestic purposes. A new tin pail, if kept filled with the lake-water, will in a few days show rusty excrescences in the bottom, which increase rapidly and soon per- forate the vessel. The water of the portion of the lake where honeycomb- ing of limestones is most obvious is distinctly harder than elsewhere, probably owing to the pressure of sulphate of lime. The source of the sulphuric acid may be looked for in the Huronian _ rocks lying to the northward of the lake. It is remarkable that the por- tion of the lake in which this form of erosion of limestone takes place lies directly in front of that part of the north shore occupied by the Huronian rocks, and receives several considerable rivers which drain an immense area of rocks of the same age. They are mostly of volcanic origin and are rich in sulphides, whereas all the other rocks around the lake are comparatively free from them. Sulphides of several metals, but especially of iron, are disseminated through most of the Huronian rocks. Pyrites and pyrrhotite are par- ticularly abundant in the greenstones, and these form a considerable proportion of the series. The two sulphides referred to occur both as disseminated grains and in the form of numerous masses, some quite large, such as those around Sudbury, a few of which are being worked for nickel. The decomposition of these sulphides upon the surface of the rocks and along their fissures and joints produces iron-sulphate, which is carried away by the streams, where its presence is frequently shown by the precipitation of a part of the iron. In the swamp at the Murray mine near Sudbury, where a large mass of pyrrhotite and chal- 804 R. BELL—HONEYCOMBED LIMESTONES IN LAKE HURON. copyrite comes to the surface, the presence of acid sulphate could be detected by the taste of the water, and when the latter, containing as it does vegetable matter also, was used experimentally for the boilers, it gave off a most offensive smell and had a very corrosive effect. The streams of the whole of this Huronian region no doubt receive many contribu- tions of acidulated water from similar sources. As the decomposition of the greenstones and other pyritiferous rocks, and also the oxidation of the drift materials derived from them, proceeds, the quantity of acid derived from them and carried into the northern part of lake Huron will increase. At The Narrows between Cloche peninsula and Little Cloche island, as already mentioned, the pits in the soft argillaceous grey limestone are surrounded by a zone stained by iron oxide. Thisis just what might be expected to result if such a rock were being slowly dissolved by iron-sulphate and, therefore, this fact helps to support the present hypothesis. The fresh rock-surfaces and unoxidized drift left at the close of the Glacial epoch would produce a much smaller proportion of sulphuric acid in a given time, and we therefore suppose that the erosion of the limestones in the bottom of the northern part of lake Huron went on even more slowly then than now. Another example may be cited here of water containing sulphuric acid which has been derived apparently from volcanic rocks. A sample from the shallow fresh-water in the estuary of the Nelson river, Hudson bay, was taken by the writer to the late Professor William Dittmar, the well known authority on water analyses, and he found it to contain no less than 4.73 grains of sulphuric acid to the imperial gallon.* The source of this acid appeared to be the drift-material which had come from the yolcanic rocks of the central or eastern part of the bay. CoNCLUSIONS. The conditions which have contributed to the production of the pecu- liar forms of erosion above described appear to have been: 1. The internal structure of the limestone itself. 2. A small quantity of acid in the water acting for a great length of time. 3. A considerable depth of water, the hydrostatic pressure seeming to promote the dissolving of the rock. 4. Freedom from sediment during the long time required. 5. The rock must be exposed to the open or free action of the water. 6. Shifting currents in the water would also appear to assist the process. *See Appendix V, Report C, of the Canadian Geological Survey for 1879-’80, p. 78, BULLETIN OF THE GEOLOGICAL SOCIETY QF AMERICA VOL. 6, PP. 305-320 MARCH 27, 1895 THE POTTSVILLE SERIES ALONG NEW RIVER, WEST VIRGINIA BY DAVID WHITE (Read before the Society December 28, 1894) CONTENTS Page eeriemaT eNO Ie nob acc cyah ah A Seren SET UM ha Sind Mente A ots) a anea ee ee eee 305 Use of term ‘‘ Pottsville” and formations included in ‘‘ Pottsville series” ..../ 306 TAPES CLE AMG ei SUGONG ig RANA Oy eee Oe a eRe Me a a Were err eh ae Pep Ee 306 Rapid differentiation of and change in floras during Pottsville time.......... 307 Somme ROMAU MO ISETICS | co. a5 ot, hottie ee aici Dee ooh oo cee apaleraiata Ea wieaheneeke s 0 308 Generalstatement .........-6+% Orch iceel eters te Leek ne UO gh cee ee tg OMEN anes. 0. Conglomerates of Piney Creek eovlion MS Cal te nen er Es tear sey cae Oey ag Nee 308 Completion of series at Nuttalland Hawks Nest .................. 2.2... 509 Number and .position of coals mined on New river............ 226.2... 40: 310 Pimearmimnromt rare Creek OTIZOMS...:. 6.2 eip.is ogee els ec we Gos ape es ave ual see sere 310 ese MRO EE Well COM ony arith s ks oe eek oy! dis ae aw Wao sok de tw ne nea ae SMe oll Pay RODD CTU TREE COS ee ae Aa, am ee nga aa alZ General statement +. ....5...06...6... SJR oho Cara ROR Ss are 8 312 eer e OU SEMONUZOMS! altro uae GN he ls we el Mahe RR Nee the wicsie he eats ws OY SEZ Fossil plants near the lower conglomerate on Piney creek ...... ........ als Reece ACME SACOM ON ING WatIVeNes o.2 86! 5 cig ci Toler: litaces ejohs ess ae die ecdle ae ale ale os 314 Flora of the Quinnimont-Fire Creek stage and its aifinileles eee ee 314 Rionarot-tne Sewell Coal ama WS TANGE:. [oie ays ace cco acs coc ele om Soc kslcager’ > 316 Hosein plantsvot the.” Homewood sandstone’... 6. 6.600. se bee eo eo ee 318 Paleontologic relation of New River section to Sharon coal in Ohio...... 319 Manguaience ol Cottsville; Sections: sa. 2.502 te ee eee ce dees betes 319 Evomemyot correlation of. base of Pottsville series 00.0.6...) die c eee tees 320 INTRODUCTION. This paper is intended to present certain general conclusions reached in the course of a preliminary study of the stratigraphy and paleon- tology of the Pottsville series along New river in West Virginia. While concerned mainly with paleontologic correlations, it touches incidentally XLIII—Butt. Gror, Soc. Am., Vou. 6, 1894. (305) 306 D. WHITE—POTTSVILLE SERIES ALONG NEW RIVER, W. VA. on the stratigraphic position of the well-known New River coals mined at numerous points from Quinnimont down to Hawks Nest. Usk oF TERM “S PoTTSVILLE”’ AND FORMATIONS INCLUDED IN “ PoTTSVILLE SERIES.” The Pottsville series, or the “ Conglomerate series,” as it is perhaps better known in West Virginia, embraces the group of sandstones, con- glomerates, sandy shales and coals lying between the green and red cal- careous Mauch Chunk shales below, and the softer, more argillaceous terranes of the ‘“‘ Lower Productive Coal Measures” above. It is essen- tially a sandstone series, though it includes some of the most valuable soft coals of the Appalachian region. Its massive ledges, rising often abruptly at short intervals from one another, support a more or less clearly defined terrace plateau, across which for more than 80 miles the river has cut its celebrated gorge. The outcrop of the series is conform- able to the general Appalachian trend. It is fringed on the east by cer- tain high knobs northeast of Hinton, and it descends westward to the falls of the Kanawha, below which it passes under water level. The general characters of the group in southern West Virginia have been well described by Professors W. M. Fontaine* and I. C. White.y To the New River section the former gave the name ‘ Conglomerate series,” so defining it as to include the lowest and the uppermost con- elomerates in the Piney Creek section, he supposing the uppermost to represent the Kanawha Falls sandstone. Fontaine's Conglomerate series was in large part referred by Professor White to the “ Pottsville,” both authors regarding it as the equivalent of the Pottsville conglomerate, the lower boundary being drawn by the latter author at the top of the red and green shales. The reference of the entire series to the Pottsville has been made, so far as I know, almost wholly on the basis of the strati- graphic evidence, the facts being, first, that it is more or less distinctly conglomeratic, and, second, that it occupies the interval between the Lower Carboniferous marine beds and the true Lower Productive Coal Measures (XIII of the Pennsylvania geologists) of the Appalachian basin. PURPOSE OF THIS STUDY. The present paper will necessarily be limited to a few somewhat gen- eralized conclusions, resulting from a preliminary paleontologic and * The Great Conglomerate on New River, West Virginia. Am. Jour. Sci., third series, vol. vii, 1874, pp. 459, 573. The Conglomerate Series of West Virginia. Am. Jour. Sci., third series, vol. xi, 1876, pp. 276, 374. + Bull. U.S. Geological Survey, no. 65, 1888, p. 179 et seq. FLORAL CHANGES IN POTTSVILLE TIME. 307 stratigraphic examination of the New River section, made in order to ascertain the general relations of the major divisions of the series there, _ together with the contained coals, to the Pottsville series in other por- tions of the eastern Carboniferous basins, as well as to establish a paleon- tologic section for comparison of local floras in the central portion of the Appalachian region. Although the work on the section is not yet com- pleted, it is thoughtethe facts ascertained are of sufficient interest to justify a preliminary publication. So far as the discussion concerns other regions, the evidence considered will be mostly paleontologic and therefore Somehow. “RAPID DIFFERENTIATION OF AND CHANGE IN FLORAS DURING POTTSVILLE TIME. It should be remarked at the outset that wherever plants have been gathered from several horizons in thick sections of the Pottsville in dif- ferent portions of the Appalachian belt, a careful scrutiny of the speci- mens shows a distinctly notable difference between the floras gathered at intervals of several hundred feet in the same section. Indeed, the period of change in conditions of environment attending the transition from Lower Carboniferous marine to true Coal Measures formation was marked by an extraordinarily rapid development and modification of vegetable species. Within a relatively short period the meager flora of the De- vonian and Pocono is multiplied to the inexhaustibly fecund and highly diversified flora of the Carboniferous, a development scarcely possible except in this division of organic life, which is the most sensitive to climatic change or environment, excepting perhaps the higher verte- brates. In the lower part of the Pottsville series many species show a relation to the floras of the Vespertine or Calciferous Sandstone series ; in the middle portion many of the forms are unique, while in thickly developed sections it is only near the top of the series that we see occa- sional Coal Measures forms creeping in. These modifications and differentiations of forms have been found to be fairly consistent and generally constant in their relative position in the various sections thus far examined. This is true even of those very dis- tant, but because the modification of a plant from one stage to another, though representing a definite phase or form, is often not sufficient to constitute a distinct species, and because these stratigraphic modifica- tions of species have received little or no attention in our American literature on Paleozoic plants, I shall frequently be obliged to refer to them as forms, designating them by the name of some locality or well 308 D. WHITE—POTTSVILLE SERIES ALONG NEW RIVER, .W. VA. established horizon in which they have so far been characteristically 00’, Dark Shale 1000 + =| Dark Shale’ G 900’ sae aS 3 S| Black Shale 800° Be Coa/- 4lack Shale = Dark Shale €oas® 700 600 += $00 4= 400° 300° fled andGreen Shales 200° F282 ESS Red andGreen Shales Red and Green La ROU RIESE Engen) Dark Shale Red andGreen Shales 100°. UN ele Oe ORET a. Shine eid RedandGreen Shales /40' to #ai/roed Ficure 1.—Piney Creek Section. (Quinn mont or predominantly found. STRATIGRAPHY OF THE SERIES. GENERAL STATEMENT. Before introducing any paleontologic facts it will be necessary to present their proper strati- graphic setting. Accordingly, while it is not my province to discuss the stratigraphic equivalents of the individual beds in other Appalachian sections, I give here two sections which show the series as a whole, the position of the mined coals and the beds from which plants were obtained. CONGLOMERATES OF PINEY CREEK SECTION. The first section (see figure 1), that along the road up Piney creek, about two and one-half miles below Princes station on the Chesapeake and Ohio railroad (chosen because it is the type section described by Professor Fontaine), is one of the most complete along the river, though badly weathered, and consequently appearing much less arenaceous than other sections up the walls of the gorge. The two conspicuous bench marks of the sec- tions along this portion of the river are the mas- sive conglomeratic sandstones at the top and near the base of the declivity. The lower one, frequently more or less calcareous, 1s a conspicu- ous feature of the gorge of New river in the vicinity of Mill creek, Quinnimont and Princes, where enormous blocks, which at first sight re- semble those so abundant in the neighborhood of Nuttall and Fayette, have fallen into the river. Although the calcareous red and green shales extend nearly 200 feet higher up, this lower conglomerate was made the base of the ‘Conglomerate series” as originally defined by Professor Fontaine. Though almost without partings in the region of Piney creek, it loses much of its massive conglomeratic individuality in passing down the river, where it becomes hardly distinguishable from GEOLOGIC SECTIONS. other more or less conglomeratic sandstones at various horizons higher in the series. Largely to this fact, as well as to the irregu- larity and instability of the lower sandstones of the various sections and to a slight undula- tion of the strata, are probably due the differ- ences of opinion still current respecting the number and equivalence of the coals worked along New river. ; The upper member of the Piney Creek sec- tion, another massive conglomeratic sand- stone, is now known to be quite distinct from the remarkable formation at the top of the Nuttall section correlated with the Home- wood sandstone of Pennsylvania by Professor I.C. White. With the exception of the latter, this top Piney Creek sandstone is the most regular and persistent member of the entire series in this region, though its conglomeratic habit is somewhat variable. I have traced it quite clearly in more than twenty-five sec- tions from Crow, about 9 miles southwest of Quinnimont, to near water level at Hawks Nest, where it forms the foundations for the railroad bridge across the river. Its massive conglomeratic ledge is the “table rock” at Table Rock post-office and defines the brow of the terrace plateau and river gorge for most of the distance down to Fire Creek. From its crest above Fire Creek mine there opens a- superb vista of the gorge and terrace, extend- ing to the northward, in which near Keeneys creek and Nuttall the ‘‘ Homewood,” about 400 feet higher, is seen to descend to complete the wall of Pottsville rocks which gradually declines to the falls of the Kanawha. COMPLETION OF SERIES AT NUTTALL AND HAWKS NEST. - The continuation of the Piney Creek section up to the base of the ‘“ Homewood sandstone ” is given (see figure 2)* from the Nuttall sec- “Homewood” Geel Sandstone Coal Bloom Coal K Black ShaleA Heteey == Slack Sha/e #1 Black Shale = Bre Creek Coa/ 500 += = Figure 2.—Nuttall and Hawks Nest Section. * The accompanying sections are platted from the barometric readings at the specified localities. 310 D. WHITE—POTTSVILLE SERIES ALONG NEW RIVER, W. VA. tion, which I have chosen because the series in this vicinity has been selected for description by Professor I. C. White.* ; The features of the “ Homewood,” which completes the Pottsville series, and the superimposed basal portion of the Lower Productive Coal Measures I have taken from the section at Hawks Nest. — | The examination of a number of measurements indicates the thickness of the Pottsville series along New river to be approximately 1,600 feet, if we measure from the base of the lower conglomerate on Piney creek, the base of Professor Fontaine’s ‘‘ Conglomerate series,” to the top of the Homewood, or about 1,300 feet if we follow Professor White in measuring from the top of the red and green shales, a result which agrees in the main with that published by the latter author. : NUMBER AND POSITION OF COALS MINED ON NEW RIVER. A comparison of the sections, which were made at nearly every mine above Sewell, shows almost conclusively that along New river only two seams are mined in the Pottsville series. In fact, instead of finding that the operations cover two or three veins below the. upper Piney Creek conglomerate, as has commonly been supposed, all the openings appear to be at the same horizon and in the same stratigraphic sequence—the Quinnimont coal being the same as the Fire Creek coal. — Although my sections of the series are barometric and were sometimes made under unfavorable conditions, they cover so many localities within a distance of about thirty miles along the river, and they are so remark- ably harmonious in showing that the workable coals fall so closely within the same limits, as to establish a probability, so strong as to justify the assumption as a working hypothesis at least, that all the mines below the upper Piney Creek conglomerate are in the Quinnimont-Fire Creek seam, the mines between that conglomerate and the Homewood being confined to the Sewell coal.T QUINNIMONT-FIRE CREEK HORIZON. In the Piney Creek section (see figure 1) I have marked as “ Quinni- mont’ a coal having precisely the same position and local characters as that found in the mines on the Quinnimont seam in that vicinity. Its distance from the top of the upper Piney Creek conglomerate in this case is the same as that measured on Mill creek, at Quinnimont and Alaska, the stratigraphic association being exactly that at the Royal mine, near * Bull. U. S. Geol. Survey, no. 65, p. 197. + So close, if not identical, are the Quinnimont and Fire Creek coals that a number of leveled sections will be required to fully establish their relations. Considering the circumstances of ex- posure and existing developments it seems improbable that two workable coals should be so close together in one section of this region without the discovery of both in the same section. QUINNIMONT-FIRE CREEK AND SEWELL COALS. al Princes, two and one-half miles above. If my barometric sections are not erroneous, the same seam is worked at Beechwood, Stone Cliff (lower opening), Dimmock, Rush Run, Red Ash, Beurys and Fire Creek. Thus at the last named mines, which admittedly work the Fire Creek coal, the interval from the mine mouth tothe top of the upper Piney Creek con- glomerate falls within the same limits, approximately 295 feet, while the stratigraphic environment is the same as at Quinnimont, Princes and other mines unquestionably working the Quinnimont coal. Among New River coals, as well as among Pottsville coals in other regions, there is much variation in the roof and in the thickness of the seams themselves. At Quinnimont and Mill creek good fossils are ex- tremely rare; at Princes, Alaska, Fire Creek and Beurys a few fragments were obtained, while Stone Cliff and Red Ash approach the rich flora found at Dimmock and Rush Run. The increasing richness of the flora toward the apex of the long bend of the river near Thurmond suggests a better preservation of the plant remains toward the northwest or down the dip, though the latter circumstance may be merely coincident. POSITION OF SEWELL COAL. Measuring again from the top of that valuable bench mark, the upper Piney Creek conglomerate, my barometric readings show the mine mouths at Thurmond, Brooklyn opposite East Sewell, Cunard opposite Sewell, Nuttall, Fayette, Elmo and Hawks Nest to fall within a distance of from 55 to 85 feet, or approximately 75 feet. There is little room for doubt that the mines at these points are all in the same seam, best known as the ‘Sewell ” or “ Nuttall” coal. This coal (see J of figure 2) is also exposed at Rush Run, about 70 feet above the upper Piney Creek con- glomerate, or 365 feet above the opening on the Quinnimont-Fire Creek coal. The same seam is reached by tram in the knob back of Beurys, and again by the same method farther up the river, at Stone Cliff. Although no sections were made at the following points, it appears very probable that the mines at Slaters and Central are in the Quinnimont-Fire Creek coal, the operations at Caperton, Keeneys Creek, Gaymont and Sunnyside being in the Sewell coal. Dr D. W. Langdon, whose geologic interpretations are well known to be reliable and who is especially familiar with the New River series, kindly informs me that the Loup Creek coal mines operate in the Sewell coal, a correlation with which the evidence of the fossils is quite harmonious. From what has been stated above it appears that all the New River (Pottsville) coals now mined in this region come from two horizons— the Quinnimont-Fire Creek horizon and the Sewell coal. As indicating in a general way the direction of the strike, it may be 312 D. WHITE—POTTSVILLE SERIES ALONG NEW RIVER, W. VA. noted that the coal outcrop and the upper Piney Creek conglomerate are respectively at nearly the same distance, barometric readings, above the Chesapeake and Ohio Railroad at Stone Cliff and at a point a little to the east of Beurys; so also at Rush Run and Cunard; or at the Thur- mond mine and a point probably between Nuttall and Keeneys creek. PALEONTOLOGIC RELATIONS. GENERAL STATEMENT. The general affinity of the plants collected by Professor Fontaine from the New River coals with those in the Sharon coal of Ohio has already been stated by Professors Fontaine and White. A portion of the mate- rial listed below comes from one or two of the former author’s localities. A preliminary examination of a collection recently made in the typi- cal section of the Pottsville in the southern anthracite field of Pennsyl- vania and a comparison of it with that from New river shows that the floras are essentially the same, they being largely identical in the corre- sponding portions of the sections. In other words, the plants found in the greater portion of the “Conglomerate series ” on New river are Potts- ville plants and belong to stages represented in the Pottsville section of Pennsylvania. I have found the same to be true of the Great Flat Top Mountain and Tug River sections farther south. FOSSILIFEROUS HORIZONS. Before proceeding further it is best to pass briefly over the localities from which plants were obtained, referring at the same time to the strati- graphic position of the fossiliferous beds in the accompanying figures. The lowest beds from which plants were gathered are the strata imme- diately above and below the Piney Creek conglomerate, comprising the base of Professor Fontaine’s “‘ Conglomerate series” (see A and B, figure 1). Scanty material was gathered from these beds on Piney creek. Numerous fragments, poor in species, were gathered from or near an horizon about 370 feet above his conglomerate on Piney creek, at the mouth of Ar- buckle creek, and near Rush Run (see C, figure 1). Specimens were col- lected at from 60 to 100 feet below the Quinnimont-Fire Creek coal on Mill creek, Piney creek (see D and #, figure 1), and at Nuttall (see Dand E, figure 2). The Quinnimont-Fire Creek coal plants came from Quinni- mont, Princes, Alaska, Beechwood, Stone Cliff, Dimmock, Rush Run, Beurys, Red Ash, Fire Creek (see F, figure 1), and Nuttall (see F, figure 2). Fossils were gathered at an horizon, about 100 feet higher, at Crow post-office, on Mill creek, and on Loup creek (see G, figure 1), and at Nuttall (see G, figure 2). Above the upper Piney Creek conglomerate FOSSILIFEROUS HORIZONS AND FOSSIL FLORA. Slay and below the Sewell coal fossils were found at Turkey Knob and Hawks Nest. From the Sewell coal plants were collected at Stone Cliff, Turkey Knob, MacDonald, Thurmond, Brooklyn, Cunard, Nuttall and Hawks Nest (see J, figure 2). Plants were found at several horizons between the Sewell or Nuttall coal and the base of the “‘ Homewood sandstone” at Nuttall and Hawks Nest (see Jand K, figure 2). At the last named locality a little material was dug from a parting in the Homewood itself (see L, figure 2). These were the highest plants collected from the Potts- ville series. Some material was obtained from a coal in the Lower Pro- ductive Coal Measures, or ‘‘Alleghany series” of I. C. White, a short distance above the Henesoed sandstone. FOSSIL PLANTS NEAR THE LOWER CONGLOMERATE ON PINEY CREEK. As it is not my purpose in this paper to attempt any local or detailed paleontologic correlations, I shall consider only the plants obtained from a few of the richer or more interesting horizons. The identifications are preliminary, and many of the names, for reasons stated at the beginning, are, pending revision or description, merely tenta- tive and subject to change. The shales immediately above and below the lower Piney Creek con- glomerate, which, like portions of the conglomerate itself, are more or less Pe arcane. are peor in plants. From those at the base (see A, figure 1) I obtained : Sphenopteris subgeniculata, (Stur.) Schiitze (?). Sphenopteris cf. decomposita, Kidst. Asterophyllites, sp. indet. Sphenopteris subgeniculata is one of the Kuropean Culm species, while S. decomposita is found in the Calciferous Sandstone series of Scotland. From immediately above this conglomerate (see B, figure 1) were obtained : Adiantites, sp. smaller than antiquus, (Ett.) Stur. Sphenopteris, sp. extremely lax. Sphenopteris distans, Stb. Asterophyllites cf. minutus, Andr. Carpolithes, small. Rhabdocarpus, n. sp. Here again those acquainted with Paleozoic fossil plants will recognize a general Lower Carboniferous cast, though the forms are few. Sphenop- teris distans is a true Culm species, being one of the characteristic plants of the Hainichen-Ebersdorf Culm and the roofing-slates of Moravian Silesia. About 150 feet of largely calcareous red and green shale and sand- XLIV—Butt. Grou. Soc. Am., Vou. 6, 1894. 314 D. WHITE—POTTSVILLE SERIES ALONG NEW RIVER, W. VA. stones overlie the lower conglomerate, all of which are excluded from the “ Pottsville” as restricted on New river by Professor I. C. White. In connection with this fact it may be noted that, while on New river the transition from marine to coal measures sedimentation is very much more gradual than in the Pottsville basin in Pennsylvania, it is marked by a much stronger contrast and evidence of change than is apparent in the Great Flat Top Mountain section. This section, while nearly desti- tute of conglomeratic material, presents an essentially arenaceous and quite frequently phytiferous series, with occasional coaly layers, as far down perhaps, if there is no unconformability, as the horizon of this lower Piney Creek conglomerate. This circumstance will be referred to later in relation to certain Appalachian evidence tending to show that the base of the Pottsville series (lithological) diagonals in time. POCAHONTAS COAL ON NEW RIVER. One of the most interesting stages in the New River section is the next higher level at which plants were found. In shales associated with a thin coal (see C, figure 1) nearly 700 feet below the top of the upper Piney Creek conglomerate, or about 400 feet below the Quinnimont coal, a few species are common at Piney creek, at the mouth of Arbuckle creek, and near Rush Run. They are the following : Sphenopteris, n. sp., Pocahontas form. Neuropteris smithsi, Lx., Pocahontas form. Rhabdocarpus, sp., Pocahontas form. Alethopteris, sp. The fact that the first three of these are predominant in and character- istic of the Pocahontas coal in Great Flat Top mountain and have not been found to extend far above or below that horizon led me to regard this stage, from paleontologic evidence, as equivalent or near to the Poca- hontas coal, an opinion which has since been corroborated on the strati- graphic side by Mr M. R. Campbell, who has traced the strata from Tug river, about 60 miles away, across to New river. FLORA OF THE QUINNIMONT-FIRE CREEK STAGE AND ITS AFFINITIES. Without stopping at this time to discuss the paleontologic details of other intermediate horizons we will pass to the consideration of the general affinities of the fossils from the important coals. To concentrate the data as much as possible the species obtained at various localities from the Quinnimont-Fire Creek coal* are tabulated in one list. The * While the equivalence of the Quinnimont and Fire Creek coals is, as I have said above, not conclusively proven, they are certainly so near together, if not the same, that ina broad considera- tion, dealing with groups, they may be treated as at one stage. FLORA OF THE QUINNIMONT-FIRE CREEK COAL. Plants of Quinnimont-Fire Creek Coal (see F, figures 1 and 2). Species. Adiantites cf. tenuifolius, (Goepp.) Schimp........... Eremopteris cf. elegans, (Ett.) Schimp. * NUVCTOPIMNIUG, VaixXe\\2)iceceene se. Sphenopteris hoeninghausii, Brongn................... osaee ne dicksonioides, (Goepp.) Schutze, form............ 2 = 00000004 90000000552 800 000004 DOO ECOSS4 OH OES SEED DOSES: FOR 009008 COCO CooeHH Oe cero eS eOCrEs osoeeeses DOL SOEES Le MVC OCLID Cam Wi Reece wesnets ea eta eans coe icles eae so uses o Gas Toss abou eee Cows alors capteneomeeiws iPeiedies worst eesonteers oe QLUCTICA COLO t UCSC MOU eth us aga eceawete tons vot neceeirocaueveaboeSedeeysded doves ae duGesiwes seek i FOOMACASSUTG, (IB iis, )) exelmutraay oy, TAKEO (0X8) 9). .caconnaanonoccaoonnosencooenacn oseeooneceeacnobo0000 08 66 ef. goepperti, (Ett.) Schimp. non Dunk., nec (Miinst.) Gein.............. cee eeeeee Pseudopecopteris DUOC WEG a CS TO Mets) ele, OLIN sc08, axe ehossectecsaeticstiaesecccieoseeststeroecca ceuseonecnts 6c (74 66 SOOO OOOO KS SSeS OLED COO CE EEE DEED SD DO EESe Oo EEE EOED OEE EES Eeeene oecves :: Olig CHOI OIG NURS ppcsmeo:iccendsso0ereaIanOu Stage or group. ET ORS@ WE Ml wo.casscSotssewecoseeecoveeeneeouassteos cictewnanieeteee Chesrens (UMMMOs)ptoriiahessecmesseveeereeescreeces seceee POP eee e COO Es OOO EHEEES COS OOOSED ATESEE OOOOH OOEE HEH EES HOES EE EEE eeD Alabama. Horsepen..... ..... WS Saeneten dasene seers scutes HOLE MOMs essaceactavausenven teusr oon eaes MME MIMNCSSC Ceess.cvsrvetiors spy verdvestestocecuceeares. SGQ8C5IK FROTS CWC MG eeesascavceese noah dbneszas overcesesxteckew ees eaen COSC coe oe sees es SOOO SEOOO EEE eEe eeecee Coe ooereeceeses os eseses oovees recess: ELONSePen——GMMIb MS) StOneCss pevesscieeecesteessssrseren Pilot. Dade...... Haccucciicdhe ccossescoesstcrsicccuerscosoreess FIOUS Op Clitieancaccvassseuesaceeen weuteah vee eaeeeiesceton does latifolia, (Brongn.) Lx., form.......... Tpageaeianss ogatuavaleaMetasscna siamese, AMM eU bo BITNAeeaadetecas eases Seowherus tenes necwarpase tewsaenthe Megalopteris sewellensis, Fout. (2)...s..es008 TcLe concep reer Saeed ost deciiiee sk cece cinsslenaaearrae wl | deolevosbaeleseae adeanot vxth ty acebeee eaulieas dovesebacen owe aeac 20005 Se re GIO MUGO7t0, HUAN ceresaservisecatscsancoweesrsy teers Ts ata SSS TUCO Ree ae SASS ce TSS CCS el SERIES See TRG oe CONG See i ENO CIEUIG WORE SUIGTROE sles Soon NU vin walsSoame Neuropteris smithsii, Lx., original...... sahsahiones Fale ere nioae oem ci eeaer rea a Rena poco necosnodonncas || JelOpys@jexsin, ANTE ONIN Sospcossscenocon ease Teaaesees ivires be 4 SoeeLOTGUM eesterte VAL CHILO DECILES MS) cteeemnce tices ae toon enee Comune a sch aisle’ Gos os seein eee cdeaececcbenens Gace heliocaeriesates Sodistire: OOOCRCE Bornia radiata, (Brongn.) Schimp....... COLOTLULES SED imac scete han eectin see eee essa facets stveteshoostes Asterophyllites minutus, ANY. (?)......c:ccceseeeeseee Hauisetites ct. oe cidentalis, TiX......cce.0escc0ee..0.. 0000 CULANIOSE CCU SIUC COLOIG, Vix, LOVIN. c2..c1s40+covasseusesacesbaotere. coos. ssncacsttecarccescaiiodiasesseubersersevecete AT ULC T LORE URN ANLOS Qs @NVIGLG Gee nadieis chivas sasiedowsesaslegeas oeae agunsccdsercesedeed stessune IE Ineo Mien eee tesei ee sansa nonattr aie Coaaeanauia est etnece vaca sree ws eoPeus icon ace tan seoa var aedens eseuei ces Sphenophyllum, n. sp........ BOCiECd0d0 CoOL EE NACHOSEECONCS adencme cues cuincae caus oveceeerstuseetene LARD OOO: 1: SAO AUB Got ce ces SPR PEER PTT co eeaeCC CEEOL EEE Eee Te ere Lepidodendron sternbergti, Brongn., form..... POP COCS ooo cess oe roe SO EES EE OEEES soe eeecesene Sees Peo sesso svees eeecesoeeses Bo sceoss ete SSEseEsEeeese sucess eeccscoeccece POO OC eo Oo C OHH sasene wesoeasreses CESSES SOE ETE DSS SOETOH OESees EEoee see eceet seen woos eseces eoecssscsceccsccocse se seecccces 66 ECDL Dey LULATERTUERS Devesteese= setiue clas cn restated se ances naesasnvs ovassabouvassdevanwaguvesedinetes ca. aeesuuteees Bondodsa0 a COU CULIOIVUTI Ys) ekire Rex's stiles weave wattans wags wivasiaesce sacs seins ea resunachsssnaine ewereieceoere Ulodendron, sp............ BELEE OPUS pen ae easter ct oe aR eas Soke ao Neun dood van os ==) ROLE Tae TAD Agee ihc aie eters) ere < ies etiam ds AOS a Oe tey fe: NO aire = ete Oo yee ale ANALYSES. By) The hydrochloric acid extract (4) yielded : SO) atet tea ats cer ctcceuete Me uetar a8 ..... 0545 grams == 0.109 per cent. SUN OV tege eee ue es ae an Np 3 -66020) 3 eek ARG ORs s Seats oyster toe Be 30s et PEO OMe ay eri oo a CAO Satine oe ae eee ote are .024 a= = OLOLS wv (F OR SA aa es oe apy aie bases She HOSSie i ade ODL AA uy UNE Ore ppicam teas x's astege an bees ee OGIS) 5°.) OLAS) oe esol i == a OOON* FE ANALYSES OF MATERIAL MECHANICALLY SEPARATED. In order to make more clear the change in physical conditions which the rock had undergone, 400 grams of the pulverulent material, free from roots and other recognizable organic debris, were submitted to mechan- ical separation by passing through sieves of varying degrees of fineness. The 17 grams tabulated below as “ silt” were obtained by washing the 43 grams of material which passed through fine bolting cloth of 120 meshes to the lineal inch, and represents the impalpable mud which remained for some time in suspension, while the 26 grams of “ fine sand” sank in the course of a few moments to the bottom of the beaker. The results of this mechanical separation are as follows: SUN sc OSU eee 17 grams: largest grains 0.1 millimeters in diameter. iT. Gen 26 es ae 66 0.18 66 66 66 S20! cc eer 45 ** % te 0.25 tg Ge «6 Sree: SC a 15 oe ce 6“ 0.65 ‘6 AG iG Si, ee re APE ges os COS AE OO r “ & Sand 5 SiS e eas ie ed eae 94 os ve 66 els iT: és aa Maamse Sand... se se ess dl Saat’ He “ 900 ce & « (SOS ee Agent te DCO AERDG) ée e ‘< Rotel ce. peal ar, 400 The coarser of these particles, like the gravel and the coarse sand, are of a compound nature, being aggregates of quartz and feldspar, with small amounts of mica and other minerals. In the finer material, on the other hand, each particle represents but a single mineral, the process of disaggregation having quite freed it from its associates, excepting, of course, in the case of microscopic inclusions, which could be liberated only by a complete disintegration of the host itself. These particles as seen under the microscope are all sharply angular and in many cases surprisingly fresh and undecomposed. The mica shows the greatest amount of alteration, the change consisting mainly in an oxidation of its ferruginous constituent, whereby the folia become stained and re- * This silica is that taken up in acid solution only. A much larger amount would have been ob- tained by treatment of the residue with carbonate of soda solution (see p. 326). 326 G. P. MERRILL—DISINTEGRATION OF GRANITIC ROCKS. duced to yellowish brown shreds. The feldspars are in some cases opaque through kaolinization, but in others are still fresh and unchanged even in the smallest particles. The finest silt, when treated with a diluted acid to remove the iron stains, shows the remaining granules of quartz, feldspar and epidote beautifully fresh and with sharp, angular borders, the mica being, however, almost completely decolorized and resembling sericite more than biotite. An analysis of this silt yielded the results given further on.* Column I shows the actual results obtained, and column II the same recalculated to a water-free basis. In columns III and IV are given the attempts to determine the soluble and insoluble portions of the same silt. The soluble portion was that obtained by digestion, without fur- ther pulverization, for two hours in hydrochloric acid of one-fourth nor- mal strength, the insoluble residue being treated for a like period with carbonate of soda solution in order to extract the gelatinous silica set free by the acid. This insoluble residue was in the form of a beautiful fine, white sand made up of very sharply angular particles of quartz, fewer feldspars, an occasional epidote and a considerable sprinkling of almost amorphous material, in part kaolin and in part a gum-like sub- stance, evidently representing a transitional stage of the feldspathic alteration into kaolin. The analysis of the soluble portion is unfortu- nately incomplete, owing to the cracking of a beaker and consequent loss of a portion of the material. The insoluble residue from the two grams treated amounted to 1.206 grams, or 60.3 per cent, and the soluble portion by difference to 0.7949 grams, or 39.7 per cent. Analysis of Silt. is THe 1G 8 TV. Actual Re Soluble portion gig analysis. | water-free. (30.7 7). (60.3 %). lonition 4242: S120 | ODOR Seer yanits cole ah) eka 1.61 uF soe Extracted in HCl . 2.83 x S70) ran eee 49.39 | 58.74 Betmcted i Na.0O., Bene 61.85 MNS, Wee” 23.84 OMe 1k ee ON ARRON AS _ 23.21 22.21 TEVO Nene ae 3.69 Ain Ae ter ees hd < scd eet ae 11.26 1.36 ONC Re a 4.41 4.79 |) (| _ 4.80 o ) 5 | ae eo oe eae os oe + Undetermined: . 9... 22.5 { ene Agr «eee ween 2) ° | | . OY, arse Se: 2.49 Dec ane | eh 99.90 99.84 98.16 * Unless otherwise stated, all analyses here given were made by the writer of this paper. CONDITIONS AFFECTING RESULTS OF ANALYSES. oma From these analyses it would appear that of the 17 grams of silt, rep- resenting 4 per cent of the total disintegrated material, only 39.7 per cent is soluble; and, further, that a very considerable proportion of the insoluble residue, as indicated by the high percentages of alkalies and lime, still consists of unaltered-soda lime and potash feldspars, the iron and magnesia alone having been largely removed. CONDITIONS AFFECTING THE RESULTS. These results are not quite what one would be led to expect from a perusal of the literature bearing upon the subject of rock decomposition. As long since noted by J. G. Forchhammer, G. Bischof, T. Sterry Hunt and others, the ordinary processes of decay in siliceous rocks containing ferruginous protoxides and alkalies consists in the higher oxidation and separation of the protoxides in the form of hydrous sesquioxides and a general hydration of the alkaline silicates, accompanied by the formation of alkaline carbonates, which being readily soluble are taken away nearly as fast as formed. More or less silica is also removed, according to the amount of carbonic acid present—a portion of the alkalies forming solu- ble alkaline silicates when the supply of the acid is insufficient to take them all up in the form of carbonates. The apparent anomaly here shown is partially explained by examination of the various separations with the microscope. Thus the low percentage of silica is found to be in large part due to the fact that the residual quartz granules are in many cases too large to pass the 120-mesh sieve, or, if passing, have been largely separated in the process of washing. Further, it is found that the sifting has served to concentrate the small epidotes in the fine sand, and a por- tion of them have even come over with the silt. The presence of this epidote also explains in part the high percentage of lime shown, since the mineral itself carries some 20 to 24 per cent. of this material. The large percentages of magnesia, soda and potash cannot, however, be thus accounted for, and we are led to infer that either these elements are there combined in minute amorphous zeolitic compounds, unrecognizable as such under the microscope, or, as seems to me more probable, the feld- spathic constituents to which the alkalies are to be originally referred have undergone a mechanical splitting up rather than a chemical de- composition. This view is to a certain extent borne out by microscopic studies, but it is difficult to measure by the eye the relative abundance of these constituents with sufficient accuracy to enable one to form any satisfactory conclusion. The magnesia must come from the shreds of mica, many of which, from their small size and almost flocculent nature when decomposed, would naturally be found in the silt obtained as stated. 328 G. P. MERRILL—DISINTEGRATION OF GRANITIC ROCKS. It is to be noted that the magnesia, together with the iron, exists almost wholly in a soluble form. ANALYSES OF MATERIAL FROM OTHER LOCALITIES. Not wishing to attach too much importance to analyses of samples from a single locality, others were obtained from along the same belt. In I of the columns below, is shown the composition of a soil from the road-cut west of Pierce’s mill, and in II and III material from the deeper cut where this road crosses Connecticut avenue extended, num- ber II being from some 3 feet beneath the surface where it was overlaid by a thin layer of the Potomac gravel, and III from the bottom of the cut some 20 feet below the present surface. The last sample, though sufficiently soft to be readily removed with the fingers, showed scarcely any of the oxidation which discolors the superficial portions, thus indi- cating that oxidation itself is not an essential part of the disintegrating process, but merely incidental to it. In column IV is given an average of the three analyses, and in V the same calculated on a water-free basis. For purposes of comparison the results given in column III, et 020, are here repeated in column VI. if TL? dt ve onmninon Ass yee Oe 5.51 3.87 Ot) 4A Via i ee SiO meer eeiae eur ekce 64.25 64.87 63.42 64.15 67.13 69.10 se pitti Rie ae en 19.97 21.32 23.08 21.26 23.29 20.99 e,0, IEC AMOR nah 312 301 2.69 294 2.07 2.77 Gapreiee Sho 20th. 318 290 3.01 3.03 3.17 2.76 [CCR NRT eae 217; 239 215 294 230 e om een ems 1.55. 186 1.77 172 We0ueeee 99.75 100.22 100.09 99.79 99.80 99.95 It should be stated that in all these cases special care was exercised in securing samples from areas which had never been under cultivation in order that there might be no possible contamination or acceleration of decay through the action of fertilizers or of plowing. Equal care was taken to obtain material in place and where it had undergone only the leaching of surface waters percolating downward from above. The re- sults, though showing a somewhat more advanced condition of decay, agree even more closely than could be expected from samples collected from widely separated localities. TIME-LIMIT OF DISINTEGRATION. A possible time-limit to the beginning of this disintegration is furnished by the Potomac (Cretaceous) and more recent deposits of the region. CRITERIA AS TO TIME-LIMIT OF DISINTEGRATION. | 329 While in the first case described the disintegrated granitic material forms the present surface soil, there are abundant street and road cuttings in the northwestern part of the District where the unconsolidated sands and gravels of the Potomac and Lafayette formations as described by Messrs McGee* and Darton are to be found overlying it at this same or greater altitudes and in beds of no inconsiderable thickness. In all such cases the line of demarkation between the two is well defined and there is no apparent admixture of materials. Although both the Potomac and Lafayette gravels contain materials undoubtedly derived from these older crystalline rocks, yet we do not find along the line of contact anything to indicate that they were laid down on surfaces such as now exist or were other than fresh and hard. There are included in the lower part of the gravel none of the large angular masses of quartz from the veins, such as now so commonly dot the surface, nor natural joint-blocks of the granite. On the supposition that the beginning of the present decomposition antedates the laying down of these gravels, we must assume a submergence and deposition in waters so quiet as not to disturb the rotted materials. That such a condi- tion is impossible becomes apparent when we consider the character of the deposits themselves. As described, they consist of quartzite pebbles derived evidently from the axial quartzites of the Blue ridge, quartz pebbles identical with the vein-quartz of the region and from which they were evidently derived, and a loosely consolidated arkose made up of angular grains of quartz and of feldspar or flakes of kaolin, scales of mica, etcetera. To this list I would add for the region about Washing- ton an abundant sprinkling of well rounded pebbles of a felsitic quartz- porphyry, which, like the quartzite, was evidently derived from the Blue ridge. The character of the accumulations, as‘Mr McGee states,f are— ‘¢ Just such as would be formed by the assortment and deposition of the different materials by ‘ powerful currents’ (author’s italics), but the quantity of coarse ma- terial is greater than would result from simple admixture of the disintegrated gneiss of the Piedmont zone and such proportion of the Blue Ridge quartzite, vein-quartz, etcetera, as appear to be mingled with it, suggesting that the portions of the formation now exposed were littoral, and that the finer materials were swept into deeper, offshore waters.”’ The pebbles of this formation, it should be stated, are almost invariably well rounded by water-action and occur of all weights up to 200 and more pounds. It seems safe to assume that these somewhat sporadic, larger forms are due to drifting ice and for our present purposes may be left out of consideration. * Am. Jour. Sci., February, March, April and May, 1888. + Op. cit., February, 1888, p. 139. XLVI—Butt. Grou. Soc."Am., Von. 6, 1894. 830 Gc. P. MERRILL—DISINTEGRATION OF GRANITIC ROCKS. Aside from these, an abundant sprinkling of well rounded pebbles of from one to 5 or 6 pounds weight each form one of the most character- istic features of the gravels. It seems impossible that such material could have been brought to its present position except by the aid of currents or wave-action so energetic as to erode the then existing decom- posed granitic material which the lithologic character of the Potomac formation, as above given, tends to prove existed. The point which I now wish to make is, however, that all such ma- terial was removed from its position in situ prior to the deposition of these gravels. The fact that everywhere along the lower part of the de- posits there is a notable lack of the angular quartz fragments and jointed blocks of granite such as now form so conspicuous a feature leads, as it seems to me, irresistibly to the conclusion that prior to their deposition all loose and partially decomposed matter was eroded away and the later deposition made upon hard and comparatively fresh surfaces. Hence the disintegration as we now find it, extending in some cases to a depth of 50 or more feet, is almost wholly post-Cretaceous. That this apparently rapid rate of decomposition is not anomalous is well illustrated in a large dike of diabase at Medford, Massachusetts, the petrographic nature of which has been made known by Dr Hobbs.* Portions of this dike are in an advanced stage of disintegration, which is undoubtedly postglacial. The writer hopes todescribe the changes which have here taken place in another paper. As a matter of passing interest and as bearing upon the same general subject, I may mention the fact that the pebbles of felsitic rock noted as -occurring in the Potomac gravels are, as a rule, in a condition of such complete decomposition (kaolinization) as to fall to pieces except when handled with the greatest care. Indeed, wherever exposed through the cutting of streets, they fall away quickly to loose sand. Nevertheless, the outlines of these pebbles are sharply oval and the surfaces smooth and almost polished. They are beyond question water-worn pebbles, and as such could only have assumed their rounded form when their materials were in an entirely fresh and undecomposed condition—that is to say, their decomposition was posterior to their deposition, or at least to the time of their becoming water-worn. This particular occurrence I regard of interest as showing, first, the great depth to which disintegration can be carried without excessive decomposition, and, secondly, the relative rapidity of the process. I should add that in areas examined farther to the west and south, beyond the limit of the Cretaceous submergence, I find similar rocks in a state * Bull. Museum of Comparative Zodlogy, vol. xvi, no. 1, 1888. AGENCIES PROMOTING DISINTEGRATION. Soll of much more advanced decomposition, being in most cases at the im- mediate surface reduced to the condition of residual clays. CAUSES OF DISINTEGRATION. It is evident from what has gone before that the changes which have taken place in the mass of rock are as much in the nature of disintegra- tion as decomposition. The question, then, very promptly arises, what are the agencies which have been instrumental in bringing about a disin- tegration which in extreme cases extends to a depth of 50 feet and upward. It is customary to divide the forces ll ae active in promoting rock-weathering into two groups—physical and chemical. Of the phys- ical agencies, temperature changes alone need be considered in this con- nection ;* of the chemical agencies, oxidation, hydration and solution. It has been abundantly demonstrated in the work of the various ex- periment stations that at a depth of a few inches beneath the surface the daily variation in temperature is very slight, and we may safely assume that at depths of a few feet both the annual and daily variations are also so small as to be practically inoperative. The purely physical agencies may be therefore omitted from further consideration. Of the chemical agencies, it is evident that the process of solution has not been sufficiently active to carry away more than an extremely small proportion of the material, but has contented itself with bringing a frac- tional part of the elements into a new state of combination. These facts, would seem to render it very doubtful if bacterial agencies, as suggested by A. Muntz and others,f have operated to any appreciable extent. Oxidation has manifested itself in the superficial portions in the par- tial destruction of the protoxide silicates, but even this action to a large extent ceases at a depth of 20 feet below the surface. Of all the agencies enumerated, hydration seems most eee and most nearly universal. Now, bonne hom in a rock-mass, without loss of any constituent, necessitates expansion, and as the various minerals undergoing this process will expand unequally, a tendency toward dis- integration is manifested, even when the process has stopped short of the complete kaolinization of the feldspars. This fact was impressed upon me in a very striking manner some years ago, and inasmuch as I do not * In the discussion which followed the reading of this paper before the Geological Society of Washington in January, 1895, the question was raised as to the possible efficacy of capillarity in promoting disintegration. The writer can only say that he is unabie to conceive of the direct physical action of capillarity as being other than neutral. As a secondary factor in promoting hydration, it is undoubtedly of importance. + Comptes Rendus de |’Academie des Sciences, vol. ex, 1890, p. 1370. 300 G. P. MERRILL—DISINTEGRATION OF GRANITIC ROCKS. find reference to ike phenomena in existing literature I may be excused for describing it somewhat in detail. While excavating in the tunnel for the water-works extension in Wash- ington, sharply angular natural joint-blocks of granitic and dioritic rocks with smooth, even faces, were brought to the surface from varying depths up to a hundred and some odd feet. Much of the material was perfectly fresh and sound, and has been utilized for road-making and building purposes. Much, on the other hand, while apparently fresh and show- ing on casual inspection no signs of decomposition, gave forth only a dull sound when struck with a hammer and showed a lusterless fracture. Blocks of this last type nearly always rapidly disintegrated into coarse sand after short exposure, though manifesting no other sign of mineral- ogic change than a whitening of the feldspars. So marked was this feat- ure that even the workmen noticed it, and on more than one occasion samples of this or the sound rock were brought me by builders who questioned its durability, inasmuch as some of the material “slacked like lime,” as they expressed it, on exposure. My explanation has always been that the various minerals composing the rock (with the exception of the quartz) underwent a partial hydra- tion from percolating waters, but, held in the vise-like grip of the sur- rounding rocks, were unable to expand to the extent of loss of cohesion and consequent disintegration. As soon as freed from compression ex- pansion and presumably further hydration took place, the mass became spongy, and, freely absorbing water, fell into sand and gravel. This idea led me to make a few experiments toward ascertaining the actual amount of expansion the rock undergoes during this transforma- tion. Barring the error due to loss of material by solution, it is evident that a fair approximation may be gained by a comparison of the weight, bulk for bulk, of the fresh and decomposed material. It being obvious that in order to fulfill existing conditions no great refinement of methods was essential, I contented myself with taking a quantity of the air-dried material and measuring it in straight, cylindrical glass vessels, bringing it to the approximate condition of the soil by tamping with water, and afterward drying and weighing. By comparing the weights per cubic centimeter thus obtained with the weight of a cubic centimeter of the fresh rock, as shown by its specific gravity, I was able from an aver- age of several determinations to obtain an approximation of 1.88, which represents with a fair degree of accuracy the average amount of expansion which the rock has here undergone in passing from its fresh condition into that of undisturbed soil a foot beneath the surface. = i each . ” se 2 oe. Veen Sse A ay eaters Pee sre weet =; BULL. GEOL. SOC. AM. VOL..6; 1894," Pier Figure 1.—ViEew or Butte; CorE NOT EXPOSED. Figure 2.—SUMMIT OF LARGEST Burtre. MEPEE BUimiES: BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 333-342, PL. 17 MARCH 30, 1895 TEPEE BUTTES BY G. K. GILBERT AND F. P. GULLIVER (Read before the Society December 27, 1894) CONTENTS Page <1) TREE IR ED TSS oo RCE Sin eS eer oR rar Gee ttt ae a a 333 DEsermpiian Of THE COLES i. 6. 62s es aso vee os Bie ea de eer See er ne re ee 333 SVDUDIE NG THRE OTC) ie 8) Seger a au re a oe ne rg 333 TOS SPIOTUBIOTE, 5.0 BE REIR O SION eee Ses COSI OTIC Re SE nS oa et 334 SST BIEN, SEBS UURS She paige age ahd heehee ea NO i te aM EP aS ONCE US oo 339 Me TENCE TOCK 6s. ses on ess Rea SoS As reik aa Re SS eee te eum, Se dete 336 Ease ee ROME ea Meh i Pete Gy eh cre cya apd nated. visa a B uBoEme NIE MRee ash iw''a Saver 337 Eee acm OMmenaihn CAMad ar. chcac oak ycwimeien cm Sasa eget che ga wed oe tae dee 338 Meee storie OMe Olsthe: COFES-...2!asiels oa tinct sakes cat oka be bakes lees 338 Wem eplOnMmG EOIN ced emgiae ci heii. ye he My ogtets csc 8 4 ores buallsid'y Ph toe GaMelag 8s 338 Spring theory... 526.6 OES Ae eR Re AM aah Ree aay cg RC UR ers ah re 339 alent MC ONpy eek, Seva ty eta tea tave Seem eins Sr ule sicaue ola eae aw ordis siete ade. ibs Shei aye 839 eisOUpiMeOlya. cote ea weutise rade. rhode Zt WHALEY, PA OMe Sie Gaon 340 eM CMISION ase ac ee Wesel os as A AS Deas pee Cie anc te a eit aa Oa eae 340 Be re Mee en OP ec inet Byars sk = RNS BIE CUTE ls Se CR eaasds lA ERR SAY, LOCKPORT } ors } eece=esF re Se aS ork t. : ROCHESTER ® e, ce, * Tona me Lat) wanda Cc on % *eanenus 439 % is: BATAVIA GO K i (ater) cree y ry > Ly a x gS an , . S x 2 2 Zz ATTICAQ s cal a La ~ s N ? “ Ni ' e > N “ a ee - ° oe She's Sil LS (Seer a Oo WARSAW G = S BS ae MTOMBRRIS < S e : axe) [522 ® SILVER £ CS ; ) 8 a 2 + « ew > a @ = x ® 4 = = a . > 2 ry au oo “2 8 a rs = ry v DANSVILLE\S z wo £ < = ° a a ~ Re] a ic} 2 oO! . > s s a : 1 @ ° e * » s % s ‘. e * 619, Teng, A LJ 7 f Allegheny RS [ie aes \ 9° ' alee HYDROGRAPHY Water-partings between drainage systems sh Figures indicate altitudes in feet above seale The figures placed transverse to broken line VOL. 6, 1894, PL. 18. 7\6° \ 7|5° Scale of Miles. S SS ~\) ROMEO aa —— \ ——- SS xX 4 6 f ONEIDA L fey \ 268 “ay yh, UT/£AO > AN =, ‘Ve fh, {? RON ON DAGA & So LYONS Le, —.__B> NSS Q EIS Nenecs rive ye et: 6YRACUSE © 9 j HERAIMER Lat. 439 f wy oe s [ onl2il? e : ; se a) @ez, 46 AN AUBUR (sen||87e2an0?” EN oe Saaam ae et XY ats SCHUYLER L ))) OWASCO \ —— Le = ea ew mt) JES $40 CAYUGA y\ 378 \ \ \ Yo \ os Ne) STH ACAKSS————_ vy Zs) Se one” Fi : & . ¢ r ° 8 ° o vy * sae 8 Bee WATKINS * . ° ‘8 . 8 4! ° e Se, Os - @512033— a . a ®na- s o° ~ ~) 'e a o> “ue ° zy, H © a j & c /o jis / { Os Pe ee e 5 % se ° . ser *s = 3 % g oot s o OR 2 a oe a 7 x ei 8 rs) 66,2 aoe c=) EINGHAMTON Res ee ee ew a ee we on oes oe —<— = 7\6° STERN NEW YORK. y heavy broken lines. cate jocation and altitude of glacial lake outlets. He rahe "4 tz i (righ a ( o] q V , > : i x oe igi f ’ oi ; a ; . v — = ' ? ; yi ; s vs ie 2s f= , é ( , } ae (Aa J ea eee eet *: (aan by j ie fal MI i s] y if { Wty \ HA ; t x Wa! } c ah vay . di \ A i f tn 4 ' ' i, j I¥ly ull f M . J ay [ BULL. GEOL. SOC. AM. - 6, 1894, PL. 18, ONTARIO. Scale of Miles. = LOCKPORT ee TT te tae D 4 ROCHESTER S dp i meee ner ATAVIA Ey (nis) ATTICA OwWAsco L 703 HEMLOCK L 2, "* a : APIA Seay Lae) e : . ’ yy 2 = « a . 2 2 *. . . 2 . G ? ott eoee INGHAMTON Bro ge rypRoGRAPHY OF WESTERN NEW york, < shown bi rainage systems shown by heavy ies ae above sealevel: : proken !ines indieate joo, Water-partings between d Figures indicate altitudes in feet iver euyesiplaced\ transverse Ho ation and altitude of glacial lake outlets. en 4 en Pec Say 4 | r é aa» el i SUMMARY OF DATA RELATING TO THE LAKES. Sou TABULATION OF DATA. ' Altitude above [Estimated dimen- sealevel. sions. 1 Names of extinct lakes. | Present lake or stream.| BP = Mm z = 6 5 RN. — zS) as} a re Sp = S ma S, = i S S) © ® <= eins Stes Feet. |Miles. |Miles. EE MEEICA, (55 eos 5 woes sRomanwealnGay Cree Kean [Moet ac liek > Srelisie oe oe ine lle oe 25 Engst) ei rr OpikarGreeki, <. dase le oe oe lee yg ete ke era enc Poa: SCFETICSER: ol ven s ces Wipper Gemesce FIVER aE Wee eale eet wc se ease Sek. ee : a Dansville. =... 2.5 +... Canaseraga creek..... 1,200} 1,250+-| 500 | 24) 2 Se Scoutsoure. 1... ..: Conmesus lakes.) 2..2.- 900+] 920+] 150] 12] 2 6. Springwater........ Henniloeke lakers oiler nae oe Slee ee hs |e rene | SeheesSlbiaess Petalaciacanadice.... .| Canadice lake, ... o..ccbee ad closs euesl he tibia [nce Peeieia ortoncoye... | rroneoye (Ake... 66s). de ce lene ce lee ee elec es ele eee Co, LEIS) Ei ern IMIATGIOCNES Keys hoo 2th, tes (Menke rere te I oh Re LOS NGC) (2) Canandaigua lake... |1,340+] (?) 800 | 18 | 2 ile LELIG i ris MIEN ORE Kotperatiocrs orn Se lho een! ol Berdicrer: eee nee allie aro 12. Hammondsport..... Keuka lakes. 322. oS. 1,125 |1,158 600 | 24] 2 War ALKANS. 5... .6< 642-5 Senecawlake sa. ls ne: 900 961 |1,000 |. 30| 5 WPPMACAN n= oi)e vis ce G58 Gayien Taken. 565: Die, 10205 KOON Sa) LG fe GCOUOM 5.55 2. Onwasconlalcercte. peel he wl Ui ate Res Eee Wel ears io Glacial-Skaneateles..| Skaneateles lake ..:..| -:/. theese dvileeeceste.. i. gpaey as: lly. Glacial Otisco. ..2... AOS OINe ees eee Aer illo 0s he Keres, ore Ne ev eunlie cate. 6 Obes lea iS lully valley 0... ©. Onendacs creeks ale se cael watts s welmee a epren ALTITUDES. The figures for altitudes given in the table are mostly from personally conducted spirit-level measurements, using railroad or lake altitudes as datum-points. The determination of accurate heights was the most diffi- cult part of the work. The doubtful figures are marked + or —. They are discussed in the paper and, at the worst, are not far wrong. Allow- ance must be made for the indefiniteness of the shore phenomena. Those depended upon in this study are chiefly the terraces of deltas accumu- lated where streams poured into the extinct lakes at high levels, and they represent planes at an uncertain and variable number of feet above the water-surface. DESCRIPTION OF THE MAP. For the accompanying map (plate 18), the United States post-route map of New York has been used as the base. The water-partings between the several hydrographic basins and the smaller streams heading upon the divides have been carefully represented. The numerals all indicate altitude above ocean level, and are the result of much study and correlation of data. In several cases they supersede formerly recognized altitudes. The altitudes on the divides are in nearly 358 H. L. FAIRCHILD—GLACIAL LAKES OF WESTERN NEW YORK. all cases taken from verified railroad levels, and usually represent the lowest places or notches in the divide. In no instance do they indicate hilltops or extremely high points. . The glacial lake outlets across the divide, which are referred to in this paper, are indicated by placing the numerals transverse to the line of water-parting. These figures give the height of the present bottom of the stream-channel on the col, or the present height of the “ waste-weir.” Obviously this is considerably below the lake surface. THE DANSVILLE LAKE. CANASERAGA VALLEY. The Canaseraga creek flows into the Genesee river at Mount Morris. The lower and main part of the valley extends from Mount Morris south to Dansville, a distance of 15 miles. This valley has a width at the bottom of more than one mile, an altitude above sealevel of about 690 feet, and the slopes rise steeply on either side an added height of from 800 to 1,000 feet. About three miles south of the present village of Dansville the valley is interrupted and deeply filled with drift. Two mature post- glacial valleys originally united here. One of them leads southwest toward Hornellsville and is now occupied by the middle portion of the Canaseraga creek. The other, of less definite character, opens southeast toward Wayland and holds Whiteman and Perkinsville creeks. A fourth, and postglacial stream, Stony brook, flows down from the tableland southward, and has produced one of the finest glens in the state. These four streams join near the village. The preglacial valleys are choked with glacial drift, and south of the moraine-fillings the valleys are half buried under the gravel overwash and the stream deposits. The valley of Conesus lake is connected with the Dansville valley by a cut or trans- verse valley about 200 feet higher than Dansville. DIVIDES. The Delaware, Lackawanna and Western railroad climbs the east side of the valley, being 335 feet above Dansville village, and winding south- east over the moraine to the Cohocton valley, gives us by its summit level the height of the col in the southeast tributary valley. The highest point on the railroad is three-quarters of a mile west of Wayland station, with an altitude of 1,864 feet. This is only five or six miles from Dans- ville, and the valley bottom here is a comparatively smooth plain. The col in the southwest or middle Canaseraga valley is near Burns station on the Hornellsville branch of the ‘‘ Erie” railroad, about ten miles from Dansville. The drift is here smoothed off into a plain, upon which the railroad lies, and the altitude of the railroad station is given FEATURES OF THE DANSVILLE LAKE. 309 on old profiles as 1,203 feet. This is not at the head of the Canaseraga creek, which has its source some miles further northwest beyond the vil- lage of that name. The ground at and beyond the cols has not been studied by the writer with reference to this subject, and the stream channels of the lake outlets cannot be here described. DELTAS AND WATER-LEVELS. The successive levels held by the Dansville lake may be determined not only by the remnants of delta terraces at the head of the main valley, but by those made by numerous brooks pouring down the steep sides of the valley all the way to Mount Morris. The slopes are too steep to pre- serve any beaches. The principal level is found at about 1,250 feet. This forms a gravel plateau either side of Stony Brook glen, and the railroad station of that name on the Central New York and Western railroad is located on it. This level is prominent at Conesus village, on the east side of the Conesus basin, and may be seen all about the valley. The Canaseraga-Stony Brook delta has been cut by both streams to a depth of 200 feet and has been almost destroyed. Between the two streams, however, is left a strip, which toward the delta front is a quarter of a mile wide and a mile long, and gives good lower terraces. The higher levels are too much eroded to be clear, and at one point the strip is only a ‘“‘hog-back.” At 915 feet it is a broad, cultivated plateau, and the terminus is similar, with an altitude of 894 feet (Clark terrace). The altitudes were taken by spirit-level, using as datum- pout the railroad levels, which are subject to some revision. The 894-foot level is found in a well marked terrace on the east side of Stony brook, and can be located at other points. In the side of the strip of delta which is on the property of Anson Whiting, on the west side of Stony brook, is a shelf about 500 feet wide, a quarter of a mile long and at an altitude of 849 feet. A delta-point at Culbertson glen, on the Dela- . ware, Lackawanna and Western railroad, midway between Dansville and Groveland stations, is at 853 feet, and other terraces visible along the steep slopes of the valley seem to be at about this level, which is the lowest well marked lakelevel observed. Lower levels are seen as stream flood- plains. The two conspicuous levels in and about the valley are, one at about 1,250 feet, the other ranging above and below 900 feet. LAKE HISTORY. Perkinsville Lake-—For a brief time a smaller lake must have occu- pied some part of the southeast branch, with outlet near Wayland into the Cohocton over the col 1,364 feet high. As it covered the site of 360 H. L. FAIRCHILD—GLACIAL LAKES OF WESTERN NEW YORK. Perkinsville, it may be called the Perkinsville lake. The evidences of this highest level have not been studied. The main Lake-—Before the ice had melted back as far as the mouth of the present Stony Brook glen, the Canaseraga tributary valley was the basin of a small lake with its overflow near Burns into the Canisteo creek, which remained the lowest outlet of the enlarged or Dansville lake. This earlier episode of the lake may be called the Poags Hole episode, using the local name applied to the narrow, picturesque middle © part of the Canaseraga valley. The lake remained at this level while the ice-dam was melting back at least to Mount Morris, 18 miles from the head of the main valley, and perhaps much longer, or even after it reached the Genesee valley. During this time the other streams above named brought down a great amount of detritus and built large deltas at the south end of the broad valley. These have been spread out over the north slope of the moraine, and being principally fine and incoherent material they have been eroded into forms which resemble at first glance the hummocky, morainic drift. Indeed the larger part of the deltas has been removed, and the remnants are so eroded that the terraces or water- levels are not conspicuous, although clear upon examination. A fine view of the head of the valley is obtained from either the Delaware, Lackawanna and Western railroad or the Central New York and West- ern railroad, the latter connecting with the former at Wayland. The altitude of the valley-bottom at Dansville is given by the Mount Morris and Dansville railroad as 691 feet. Hence at its full height the lake was more than 500 feet deep, and this lasted for a time sufficiently long to allow the ice-lobe to recede at least 20 miles. When the ice-barrier was so far removed as to permit a lower outlet of the waters northward, probably to the northwest by Caledonia and Le Roy, then the middle Canaseraga creek came into existence and the reversed drainage began to fill the valley and eventually joined forces with Stony brook in the building of the delta southwest of Dansville. When the Dansville lake was much lowered the Canaseraga was com- pelled to reéxcavate its middle valley and has produced the rough and peculiar topography of the narrow, deep gorge, six miles long, which has been locally called Poags Hole. In cutting down near the present mouth of the gorge the stream fell upon an angle of rock projecting from the west side of the great valley, and was compelled to make a rock-cutting which gives a singular postglacial exit to an old preglacial valley. THE ScorrspurRG LAKE. CONESUS VALLEY. Conesus lake is eight miles long and about one mile wide. The valley extends south as a swampy area for more than two miles, and then rises FEATURES OF THE SCOTTSBURG LAKE. 361 gradually for nearly two miles more to the village of Scottsburg, which has an estimated altitude of about 900 feet. A small stream, called simply the Inlet, drains the higher, narrow valley south of the village and flows on ina sinuous course to the lake. The only considerable tributary is Conesus creek, from the east, which joins near the lake. DIVIDE. The bounding walls of the valley are unbroken except at one point. A remarkable depression or postglacial transverse valley severs the western wall at the village of Scottsburg and leads to the Canaseraga (Dansville) valley, about three miles away. This divide is at nearly the same alti- tude as the village, or about 900 feet, and gave free connection with the Dansville lake at the higher levels of the latter. DELTAS AND WATER-LEVELS. There are two conspicuous levels in the Scottsburg valley. The higher one is at about 1,250 feet and corresponds to the Dansville summit levels. The finest example seen of this level is at the village of Conesus on the Erie railroad, three miles east of the head of the lake, where the Conesus creek debouched into the expanded waters held up to the level of the Dansville lake. ‘The lower level is seen at many points along the valley sides where streams have built deltas at a height estimated at 80 to 100 feet above the lake, which has an altitude of 819 feet. The village of Scottsburg is located on the delta of Inlet creek. The large lower delta of Conesus creek is, in its highest part, apparently something over 100 feet above the present lake. LAKE HISTORY. The valley of Conesus extends so far north of the parallel of Mount Morris that the north end was probably closed by the ice after the Dans- ville lake had been lowered into the vast Warren water which then buried all the ice-uncovered area of western New York and the Great lakes. In this event it would have overflowed by the gap at Scottsburg, but the fall could not have been many feet, and it is possible that this col was near the level of the Warren lake. Further observation is required to determine the full history of the Scottsburg water. Toe NAPLES LAKE. CANANDAIGUA VALLEY. The present lake is about 15 miles long. The village of Naples lies about 4 miles south of the head of the lake, and it is 4 miles more to the col. The valley is from one to two miles wide, but at Naples is narrowed to only about half a mile. The moraine is remarkably de- veloped and occupies 4 miles of the head of the valley, from near Naples LII—Butt, Gro, Soc, Am., Von. 6, 1894. 362 H.L. FAIRCHILD—GLACIAL LAKES OF WESTERN NEW YORK. to near Atlanta. The elevation of the lake is 687 feet. Naples lies about 100 feet higher. The walls of the valley here rise directly 800 or 900 feet, and the surrounding tableland is more than 2,000 feet above tide. DRAINAGE. A numbe: of streams pour into the valley near Naples. From the northwest comes West Hollow creek. This is joined from the west, not far from the village, by Springstead brook. From the southwest comes - Naples brook with several tributary brooks, the main one being the Garlinghouse, from the west. From the south comes Olney brook; from the southeast Tannery Glen brook. All these minor streams united make Naples creek, which, near its mouth, three and a half miles north of the village, is joined by West river, the latter heading near Rushville, between Canandaigua and Seneca lakes, and flowing southwest. DIVIDE AND CHANNEL. These streams all occupy deep valleys cut out of the tableland, and head near streams flowing into other drainage systems. The lowest of them all is West river; but this, coming from the northeast, was under the ice while the southern divides were uncovered. The next lowest divide is the col to the west, between the Springstead brook and the inlet of Honeoye lake, but the northern outlet by Honeoye lake was also in the earlier life of the lake dammed by ice. The lowest southward outlet was over the col between Naples brook and the Cohocton creek, near Atlanta (formerly Bloods), on the Erie and the Delaware. Lackawanna and Western railroads. ‘This divide has an altitude of about 1,340 feet. The outlet channel of the Naples lake is a good example of an aban- doned river-bed. It is something over a mile long, 20 to 25 rods wide, with banks 15 to 20 feet high and a flood-plain of varying width. It heads at the divide among hills of drift, and pursuing a nearly straight course it opens into the Cohocton valley about one mile northwest of Atlanta station. The highway leading northwest from Atlanta into the Naples valley crosses the channel by the house of William Rowe, at which point the direction of the channel changes from south to west of south (see figure 1, plate 19). No stream of consequence has occupied this channel since it was abandoned by the overflow of the glacial lake, and the pavement of cobbles and bowlders in the bottom of the channel is still well shown through the vegetal accumulation. DELTAS AND WATER LEVELS. Fine deltas have been produced by all the side streams entering the valley and debouching into the deep lake. They are fairly well pre- served and the terraces are conspicuous (see figure 2, plate 19). BUIEESGEOL. SOG: AM. VOTO, 1694 PE ato: Figure 1.—OvurLter CHANNEL. View looking north, or upstream, from near the mouth of channel.

¥ : | | : ; = U es a. * > : j : t f ( | H ay 2 : ch ai. f : ; + ? >» f i 3 ; Date, a C oe RELATION OF WATKINS LAKE TO OTHER WATERS. 369 before the water could escape over the plateaus either side of the north end of the lake at a level under 900 feet. After the lake had attained by the northward recession of the ice-dam a length of about twenty to twenty-five miles from Pine valley, it received the overflow of the Hammondsport lake, and soon afterward it probably received the waters of the Naples lake. If the receding ice-front held an east-and-west trend across the region of these lakes, there must have been a stage when the waters of the Watkins lake coalesced with the waters filling the Canandaigua and Keuka valleys upon the west, and the Ithaca and perhaps other valleys on the east, up to the 900-foot level. } To this more widely expanded water at the Horseheads level, uniting probably several of the local lakes, it seems desirable to give a non-geo- graphic name, and the author proposes a name honored in American geology and prominently identified with the glaciology of the Lauren- tian lakes. Let it be known as lake Newberry. Possible Relation to Warren Waters.—The Horseheads channel is at present the lowest and best developed of all the passes over the divide east of lake Erie. At the time we are considering it was probably, by the de- pression of central New York, the lowest outlet east of Chicago. Conse- quently, if the vast Warren waters, into which all these local glacial lakes were drained or with which they were blended, found, by the depression of the region, any southward escape lower than Chicago while yet the Mohawk valley was ice-covered, then the Horseheads pass was such out- let. In this case lake Newberry retained its level as a part of the great lake succeeding lake Warren. “Tue Irmaca LAKE. CAYUGA VALLEY. The geography of this valley and its glacial history is similar in gen- eral to that of the Dansville valley and lake. Its proportions are on a larger scale, but there is the bifurcation southward and the double lakes in the beginning blending later into one. ‘The lower outlet in the Ithaca lake is, however, the eastern one. Cayuga lake has a length of 38 miles and a breadth of one and a half to three miles. Its altitude is 378 feet above tide; its depth more than 400 feet. The main valley reaches about two miles farther south, to South hill, in Ithaca, where it divides. The main branch, or Cayuga * The topography, drainage and altitudes are well described in “The Cayuga Flora,” by Pro- fessor W. R. Dudley, to whom the writer is indebted for many facts in the description of this basin. Some of the figures of altitudes are changed to agree with later determinations. LIlI—Burt. Gror. Soc. Am., Vor. 6, 1894. 370 H. L. FAIRCHILD—GLACIAL LAKES OF WESTERN NEW YORK. Inlet valley, continues southward, with the divide 12 miles from Ithaca, at Spencer Summit, and leading over the col to the Susquehanna at Waverly. The lesser branch is directed southeast, forming the valley of Six Mile creek, with the divide ten miles from Ithaca, and leading over to the Susquehanna at Owego. There are several streams debouching into the head of the main valley (see figure 2, page 372). Cayuga Inlet creek has several tributaries from the south, Buttermilk creek and Coy Glen brook joining within two miles from Ithaca, Butternut (Enfield) creek two miles farther up the valley, and the West Branch (Newfield) creek and Lick brook six miles from Ithaca. Six Mile creek, draining the southeast valley, joins Inlet creek in the city, two miles from the lake. Two important streams come in from the east—Fall creek at the north of Cornell University grounds, and Cascadilla through the University grounds and the city. Both of these creeks have cut deep postglacial ravines. Many smaller streams have cut similar ravines all along the lake-shores. Taughannock and Tru- mansburgh creeks on the west and Ludlowville creek on the east are the most important. DIVIDES AND CHANNELS. The moraine in the Cayuga Inlet valley extends from Newfield station, on the Lehigh Valley railroad, south to Spencer Summit, a distance of 8 miles. The abrupt upper or south end of the moraine forms the col. From this point southward is an open valley. The railroad finds a low pass near the west wall of the valley, with a summit elevation of 1,065 feet, one-fourth of a mile north of Spencer Summit station. The old stream channel, which once carried the overflow of the glacial waters from the lake north of the divide, is, however, upon the extreme eastern side of the valley, upon the farm of Mr A. Signor. The head of the out- let is at the north border of the only primitive pine forest in the Cayuga basin, at which point the morainic hills fall off steeply to the deep valley northward. The channel runs southeast through the pine forest about 80 rods, then some 40 rods through cleared fields (see figure 1, plate 22), and then bends abruptly to the east, and in 20 rods reaches the Cattatonk creek (see figure 2, plate 22), which enters the valley from the north and at this point flows in a rock bed at the very base of the eastern rock- wall of the valley. Just before reaching Cattatonk creek the channel crosses the highway close to the house of MrS. D. Turk. At this point it is about 12 feet lower than at its head in the north edge of the forest. This figure is upon the authority of Mr Signor, who relates that during a flood of the creek in June, 1855, the waters set back up the extinct channel until some water actually fell northward toward Cayuga lake, the flood standing 12 feet high on the buildings by the highway. The BULL. GEOL. SOC. AM. VOlESG; 1694 Pio. Figure 1.—MippLe Portion oF CHANNEL. View looking south, and downstream, from edge of pine forest. In the left,background the channel turns abruptly to the left. Figure 2.—Moutn or CHANNEL. View looking east, showing sudden termination of channel in Cattatonk creek, which flows at the base of the rock wall. Point of ridge at the right is rock. OUTLET CHANNEL OF WEST DANBY LAKE. n iW i iN i t aa y Lew 1 oe f ) m iw wat n fv Seth ( FEATURES OF THE ITHACA LAKE. oll channel is 10 to 15 rods wide, and is the smallest and shortest of the ancient lake outlets so far seen by the writer.* The altitude of this outlet is not accurately determined, but is esti- mated, by comparison with the ce across the valley, at about 1,040 feet. Southeast of Ithaca, in the Sais of Six Mile creek, the moraine is not so well developed as in Inlet valley, and has suffered much erosion. It ends north of the divide, the heavy deposit reaching only to Caroline, and the col being a mile further south in a long stretch of open valley. The divide is at the Bell schoolhouse, one mile south of Caroline sta- tion, where the valley is nearly one mile wide between the rock walls, and the swampy bottom is one-fourth of a mile wide. Including the divide, and for considerable distance north and south, the valley bottom is nearly level. The altitude of the Delaware, Lackawanna and Western railroad at the crossing of the divide is 985 feet, which is perhaps 10 feet above the low ground. The Beaver and Wilseyville creeks emerge from ravines in the western rock-wall of the valley, and after nearly meeting, part company, the former flowing north to join Six Mile creek, the latter south to join the Cattatonk below Wilseyville. The head of the definite stream channel is about a mile Dee the divide, at White Church station, with a width of about one-eighth of a mile and with walls perhaps 20 feet high. The tracks of the Delaware, Lackawanna and Western and the Elmira, Cortland and Northern rail- roads lie in the channel from White Church to below Wilseyville. At Wilseyville the valley widens, and the channel is bordered by extensive flood-plains (see figure 1, plate 23). DELTAS AND WATER-LEVELS. Location and Height.—The deltas are not large compared with those of the other glacial lakes and with the size of the valley, both branches of which were lake outlets for a time—the detritus being borne south to the Susquehanna basin. When the fall of the lakes reversed the drainage, the inpouring streams from the cols spread their load over the many miles through which the moraine stretches; but the side streams have built definite deltas. The most conspicuous near Ithaca are on the steep slopes at Coy Glen (see figure 2, plate 23), Buttermilk, Enfield, and Newfield ravines. The deposits of Fall and Cascadilla creeks are spread out on the plateau above the campus of Cornell University. Other deltas are seen at the mouths of streams down the valley or northward. In the valley of Six Mile creek are conspicuous terraces of considerable extent. There are two of these levels. The upper is above the White * The author is indebted to Mr G. K. Gilbert for indicating the location of this channel, 372 H. L. FAIRCHILD—GLACIAL LAKES OF WESTERN NEW YORK. Church divide about 35 feet, as determined by hand level, or at an alti- tude of 1,020 feet. The same terrace farther north, as measured by the engineering department of Cornell University, is from 1,014 to 1,028 feet. The lower terrace has not been measured. Figure 2.—Delta Terraces in Cayuga Inlet Valley. Comparison of Terrace Levels—tIn the Inlet valley the deltas of all the streams show conspicuous but not large terraces. For their illustration the writer is permitted by the courtesy of the United States Geological Survey to use the unpublished Ithaca sheet which covers the Inlet valley, SouTH EAST SIDE STREAMS. WEST SIDE STREAMS. WORTH: STRATTON. cicw - SUTTERMILA WEST BRANCH BUTTERNUT. COY GLEN. o © 6 8 <== =] ——— —— ——s = o £9 nn LEVEL OF LOWER INLET VALLEY — Se 3B Liv or Caran Lar —-—_—_-- 78 Figure 3.—Approximate Height of Terraces in Cayuga Inlet Valley. but not the Six Mile creek valley. Upon each of the four principal deltas, namely, Coy Glen, Butternut, and West Branch upon the west slope and Buttermilk on the east slope, the best developed terrace is shown in every case at the delta-summit, and is bounded at the lower edge, with BULL. GEOL. SOC. AM. VOE. 6,.13894, Ply 23: Figure 1.—Ovurter CHANNEL AT WILSEYVILLE. View from Wilseyville station, looking north. At this point the channel widens. Heavy flood-plain is seen in distance, on west bank. DELTA. Figure 2.—Coy Gute» View from middle of Inlet valley; looking northwest. ITHACA LAKE. TERRACES OF THE ITHACA LAKE. ale only one exception, by either the 1,000 or 1,020-foot contour. The sloping surface of the terraces could not be below these levels, but might be a few feet higher. It will be observed that this height agrees closely with the higher terrace in the valley of Six Mile creek, and evidently indicates the water-level of the Ithaca lake while overflowing by the White Church outlet. The manuscript map does not definitely show by the contours any terraces marking the higher level of the earher water, which must have been restricted to this western valley with the overflow by the col at Spencer Summit. Such evidences of water surface should be sought at an elevation of about 1,060 or 1,070 feet. | Several small terraces lower than those above described are, however, found on these deltas. Three deltas show terraces at 940 feet, four at 820 to 840, two at 700 to 720, and four at 600 to 640. Two deltas also show small terraces at 500 to 520 feet, only about 100 feet above the valley bottom, and 122 feet above Cayuga lake. In the sketch the terraces are indicated by the dotted areas, and the height of the contour-line bounding the lower edge of each area is given in the comparative table. In the latter an attempt is made to represent graphically the relations of the several terraces. The vertical lines indi- cate the ravines which bisect each delta deposit. As the sketch is taken from the Geological Survey map, the contours are 20 feet apart, and on - that account may not fully represent the correspondence in height of the terraces. Closer contouring would not lower the higher terraces of any set, but might raise the lower ones so as to produce a much nearer ap- proximation to one level. Variation in Height of Terraces.—It is evident that the subsiding waters were able to produce minor terraces upon the steep slopes of the inco- herent stream deposits during relatively brief halts, but the interaction of lake and stream, with varying conditions of wind, slope of shores, and amount and character of material, caused the delta terraces to vary con- siderably in height within short distances or even in the two sides of the same terrace. However, notwithstanding the variation in height, the sketch and table show a substantial relationship between the terraces of all the streams in the Cayuga Inlet valley. LAKE HISTORY. West Danby Lake——In the early stages of the ice-retreat there was a minor lake in each of the two southern forks of the valley, these separate lakes being held up to the height of their respective divides, and this condition continued until the ice-retreat had uncovered the point of South hill. The lake in the Inlet valley requires a separate name, and may be 374 H. L. FAIRCHILD—GLACIAL LAKES OF WESTERN NEW YORK. called the West Danby lake. When the ice-front had passed the north point of South hill, in the city of Ithaca, the West Danby lake fell only about 50 feet to the level of the lower lake in the southeast valley, and then began the major stage of the Ithaca lake. The main Lake.—This level must have been preserved during the con- siderable time which was required for the front of the glacier to retreat more than half way down the Cayuga valley, or until the crest of the table- land east or west of the valley was uncovered at an altitude less than the White Church outlet. The waters then probably fell about 85 feet to the level of the Watkins lake, thus forming a part of lake Newberry, and subsequently to any lower levels of the great lake which covered all of northwestern New York until the ice was removed from the Mohawk valley. The Ithaca lake was the largest and deepest of the local glacial lakes of western New York. At its maximum it was probably 35 miles long and perhaps from 5 to 10 miles wide. Its depth was more than 600 feet over the present level of Cayuga, and at the north end the total maximum depth was over 1,100 feet. This would seem to be sufficient proof of the competency of glacial ice to act asa barrier retaining a great depth of water. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 375-388 APRIL 13, 1895 CRETACEOUS OF WESTERN TEXAS AND COAHUILA, MEXICO BY E. T. DUMBLE (Presented before the Society December 29, 1894) CONTENTS Page RUE POEs he Nees oer oe hc rae eke wh event pan wale ava wav clhy we Eas 376 Localities of occurrence, character and relations of the rock................4- 376 PUM RET AN SECINON:S Weis o/s ahaa Swe) Waa nid ede ee Mie RUMEN a Moteur Sotais- sie islarreced ais 376 Peaocaion-and, the character of the country... .. 2.42.6). cages seers oe’ 376 ROMO MEGUMI oxalate ces yp Sis rahec eas we caic caine cious SATII anual ale S reatha minis old Bower Cretaceous ........... ROR ee me ee PT aS eee Is acausb oes syd held 378 net SE G1 Ste REI IE EO A EE es eg a 378 wold TEE SVP UDG ISS SUG ates ER re a cig ng ogee aa Re 378 ROE CMOT tee tees a A tea OT asi feleatectw cae tees yeas 378 LCA TMS e090) |e Wee aaa at ae BOG. ets ms Ol ARR REE Ere 378 Exposures in the Arboles and Burras mountains .................6. 379 PmiMenes OL une POSHUC PEMOUiot yates. o 6 405-4 xian caus stm sree vow eve 379 SE MMe CE IUOMCIVISIOM se Pe Are oon too 2h ais so Sears, Lrayy Geteba no ths 379 SG Re eC OS MIVETs fio Se hres eer os hice Gaee aia surne @ aleve poe 5b. 379 Womans Alte MOLENZO SCCUIOM cesta e ee heat ket a guly whe ad hess sees 379 Caprina crassifibra as a criterion of Fredericksburg age............ .. 380 LES SLUG En PERT OLT) 1 a gee gn ERS TP Pert eh ener Oe Cer aes 380 a aE SN OM 5 12 econ’ apopern. Ske ste «lacahe sos Seeia De Pree anal lee Grad ci AD kasd 380 mpm Ea TIn OCU ed ARCA ae ce cee e cll meis eaten Alona gle sie Wie whee a vole a Sieve 380 ee ne oe RO Mert ED Wry ate swath ay) ees rns Dine Sipiges acpi Wie wus vend wae PE Date 380 La EOTULSI Ee Poy 55) 205 105 0 Qe cy PL go a a 381 PPeRRES SAO VEMZO BECIIOM se soe eke sce are etig 45 pote sae os ews Dee 381 ihe Pinleysiacle: Mountains iseetion\y io2 coe. Ps eee ee ee ee ws kd on 382 Ccommpunsons with ober localities. ic... aeedie neice peck eons ate 382 PRETENCE ONS 5. NG heh ced osiviiecr Wcins> pica se cys sein OWie obi Wedel So kas 0g lee 383 Pe rte SUCCESSION ANG COTTCIA LONG Whe shee eis do eye ca G Bvele ee, 1 oe ese 395 Rr eernet MC MEEACLCT Ona At nett hay! cr ie aoe sean nwo sed unehngonethls, ala 8 395 Sop ean bare wWyOOd CONES). 22s Stack wais las eis n eee Glos ae Sa ae eas 396 ARCO LC Sry ter ate Put bays feet thy ici gos ysis g Baths bee nie eee + eieie Same oust 396 SSO MAT COCR ete Mee ee ii ee a cer igs ap Ai unis Se sie ch ed IAN SST SaaS cha ase 397 BAM ARCO TUXEC OTE ee ay eer aoe 0d aun See er aes IS Coa a se alabat ete muse ssheuebegate) § 398 eT ULC yr herein a hae aes SEN dR Ua E Ny ableton elk aera aces 398 Pep MOMTa SAO LACCONMEC:. coo icc ks ase wie le hse Ges aa bow come a US aces me Sk 399 (PELL DENG LDNSIOIAIR AI Ue jie ie es SRE a A ea Pe ani ee 499 Part IJ.—Square butte and its remarkable differentiation zone............... 400 Mia eaA LING ROTI A See ge See RE bn i ale Dw echt cys ie SERS gysiwid aleve's) s.c0e a Bat lots ee seke 400 PSUS EC EEG es So Ra SR yet I ie Gn ey aera ae Oe er VU Te 400 Sewer GEserip Uo Ol SQUATe OULLC. ces poe cic sce dine wee stele, ape ale ohedas Gis 400 era CONIC OMRON ee micyers «eit Pah oo 65 OL ies Sialtale delsyw ols 2 she See digs aol atch te 401 Lower zone of dark Thanloor pi ay BE dad ae Re PRS a peer rach es Ag 402 Mim mee mumcloleawitite TOCK yt... Oak net sales aie oc. 2's 2 toe eye Sage Hee wees Os 404 Orin Ghine platy Parting 0.6666. 0. 2... 2 os ee ale eA eae cient ke a 405 DADE, THLNTNYS OPTIC aS pense tm een ae te ane i en oe ae rns WE 405 Wigeramimnatte sechlioniot Square WUbtle. . 1c .une sds els or ee eee we Selene ee 407 Summary of field results ...... Bn esae ae praeeeie araar hag 2,18 Bom. Lau th nee 407 eros py Ol SQUALS WOULC Ls... 6 5 <6 lease eeciseos 2 da tac cles eee et gees 407 Characteristics and minerals of the dark rock...........:+0..0:0.+0% 407 Meca scopic anesmMICrOSCOPIG. © s/c. ota s sie dsiae. vo te oe et weenie 407 Aves aie ey eee PAP ert the 2 8 od «is Hots Seeroum.ctns aay tees aieieraiers sco Slaleote 408 COMSAT OA: 1 2a 5 Je RS ea ee eo ASP SPP ran CEP eget Shs 3 409 TRG Gk 8 os 0 Sg Po Oar BLM ee OCR Pens Cle 409 ay CMU Me are so) Saas ahs) 5 pss 4.4 Hic amie eee sheveka wievnun a ao eye tus legals 410 ADpi te ease cenete ee Cs ah atl oo Ge ecalesec ga witieidd OSs eters, a ove seca olee eas 411 * Published by permission of the Director of the United States Geological Survey. LVI—Butt. Geox. Soc. Amu., Vou. 6, 1894. (389) 390 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. Page Plazioclasen.: orf vac Sec, Ae he Bee Fie ate ele er 412 Nepireline $e paused wos ates Bs cle etal we enue eee Pep scs' 412 CamOriniite? 2: isan ee sean ain eRe oa ea Goh dak sensed 413 SOAS onc 25.s Fie bays wd os a Alec a «eevee eae ie nate ee we. 413 Natrolite oo) cctv bien 3 ov nie » mola aisles ous ence ee 4138 Chemical compogitgOn..: 0.02 2.0 2-2 oa deen ae ost ee 418 Structure and classification .....-..2.. re fe 414 Name shonkinit®,.... sc... oa ea ve wee os ane ee er 415 White rock or sodalite-syemite -....... 2... ..s5.--2.5- se 416 General petrology of Square butte......2..2... 7.04.52 2 =f eee 417 SUMAMATY ekd cist eee eet eee ia bei slele a gale « ses ayes ee ee ae een 422 Part I.—GErEoLOoGY oF THE HigHwoop MovunrtTAINs. SITUATION. The group of mountains of which a brief account is given in this paper form one of the detached mountain groups lying east of the Rocky Moun- tain Cordilleras and rising abruptly from the level plains of central Mon- tana. =~ d 4 + ‘ bs sy " e d r % 4 a oe oe ” - ty ” BULL. GEOL. SOC. AM. VOL. 6, 1894, PL. 25. Figure 1 —Hoopoos or SHONKINITE SHOWING PLATY PARTING AND Dip. FIGURE 2 .—In ZonE OF DARK Hoopoos OF SHONKINITE ON SOUTH SIDE OF Bure. Showing the disk-like structure, SQUARE BUTTE. HOODOO ZONE OF SQUARE BUTTE. 403 Above this is seen the intensely white color of the naked upper slopes, masses and precipitous walls of rock which rise above the bristling fringes of hoodoos below, a contrast rendered all the more intense by the white- ness of the former and the black color of the latter. These peculiarities are shown by figure 1, plate 24, which is a view of the butte from the lower slope on the southeast side, and by figure 2, plate 24, which shows the south side of the prolongation on the southeastern side of the moun- tain and which has been previously referred to. It must be said, how- ever, that the photographs present only ina feeble way what is most striking in nature. As one approaches still nearer and enters the region of black mono- liths, it is found to be a labyrinthine maze of small glens, separated by towering masses and pinnacles of black rock. The hoodoos attain in many places a height of from 100 to 150 feet, and from that sink in size down to examples but a few feet in height. The attention is immediately arrested by a peculiar and regular arrangement of a platy structure they possess. They are built of a series of inclined disks, each a few inches in thickness and oval to subangular in shape, and with rounded edges which accentuate the disk-like form. Generally the disks decrease in size from bottom to top, but there are many exceptions to this rule, and in these cases strange and weird figures are produced, resembling colossal statues, sarcophagi, etcetera. Occasionally the disks are not flat, but slightly dished ; the hoodoo then resembles a pile of huge inverted watch glasses. The plane or hade of the disks is not horizontal, but slopes to the outside in all directions around the mountain approximately parallel to the prevailing slope, which, indeed, is conditioned by this platy parting. The disposition is precisely ike the dip and strike of sediments in a domed anticline, and the resemblance at times to sedimentary strata is quite striking, as may be seen in figure 1, plate 25. The hoodoos are apt to be disposed in radial trains around the moun- tain slopes, each train growing consecutively smaller as it ascends. Between them are the small wooded glens previously mentioned (see figure 2, plate 25). The rock forming these strange pinnacles is uniformly in all cases a rather friable granular one, composed chiefly of a basaltic augite, to which their black color is due. The origin of these spire-like masses is partly explained by the frequent presence of large, often huge, spheroidal bowlders of white syenite rest- ing upon their points and often balanced in almost incredibly delicate positions. As these masses which descended from the upper slopes are hard, tough and feldspathic, they have resisted weathering and erosion much better than the crumbly, easily altered dark augitic variety upon which they fell, and have thus conditioned the construction of the pin- 404 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. nacles, just as rocks condition the formation of ice-tables and pinnacles on glaciers (see figure 1, plate 26). In many other cases the white syenite bowlders are to be seen lying at the foot of the columns from which they have fallen. We may remark here that in all our experience we have never seen a more weird and curious labyrinth of pillared rocks than this which sur- rounds the lower base of Square butte. In singular scenery it equals, if it does not surpass, the famous Hoodoo country, on the northeast border of the Yellowstone National Park, in northwest Wyoming. UPPER ZONE OF WHITE ROCK. Ascending through the zone of hoodoos which measures perhaps a mile along the slope, they are found to diminish in size and a horizon is reached where the character of the rock changes abruptly from the dark, nearly black augitic phase to the white syenite described by Lindgren. In many places the hoodoos continue, but now they are made of the white rock. They are smaller in size, but possess the same remarkable disk like, platy structure, and the disks are perfectly parallel to those of the black variety below. The white hoodoos are rarely pointed, and are more apt to be flat topped and of the character seen in figure 2, plate 26. The transition line between the two rock varieties is extremely abrupt, but it is not of the nature of a contact. The rock continues of even grain throughout, but in the space of a few inches or a foot or so the black augite begins to diminish and finally disappears, the rock assumes a more feldspathic character, hornblende occurs, and it rapidly passes into the syenite described by Lindgren and which constitutes the main inner mass of the mountain. There is thus a narrow mottled zone between the black and the white rock. The hornblende is present in so small an amount in the syenite that the rock, especially when seen in full sun- light, has the whiteness of marble. The monoliths which lie near the transition zone are sometimes seen to be of black disks resting in place on a pediment of the white rock be- low; sometimes the transition band passes through them and they are of black disks resting on white ones, or it passes through the disks at a nearly vertical angle so that one part of each disk is white and the other black; at other times a white hoodoo is but a few feet distant and above a black one, both resting in place on a continuous exposure of the white rock. These facts are illustrated in figure 2, plate 26, which shows ‘a black hoodoo resting on white syenite, with a white hoodoo above and to the left. The facts just presented are to be carefully noted, because they show beyond the possibility of a doubt that however much the two varieties of rock may differ and however abrupt may be the change from one into BULL. GEOL. SOC. AM. VOIR Ow SO4 a P26: ING O Figure 1.—Bow per OF SYENITE REST PILLAR OF DARK SHONKINITE. Tn the zone of dark hoodoos, south side of butte. s ee a : ay ernie ee = a ‘ te Figure 2.—Top OF THE DARK ZONE. White syenite above and to the left; to the right below dark shonkinite resting in place on white syenite. SQUARE BUTTE. i j 4) ne t - i ' < % r aa 2 lle gene Sellen ee eel ; in e A} : nibs ; i? : | | . y “55 ; 4 _ ks ; f H ; ‘ - : | : [ cs , ie it a : ae 5 . f < a : . , = j nt A lj ( iE ek F ‘ ; , Fr: i ny 2, ‘ (a Mie 4) ie ; \ 1 : Hf t ; f } ad) . (one a F} f * ; 2 Rt 7 i - 7 i ‘ } f : m ‘ ‘ f es ; ! ’ i y | ¥ | ‘ u u oO } ; i in aint al 1 A Waa ee ‘* ecu 1 eit eens Crease arn. ae en | }} \, ie si es : se ; | ; i fi i : ‘ ~ WHITE ROCK ZONE OF SQUARE BUTTE. 405 the other, they were not formed by two separate intrusions, but that, on the contrary, they are a geologic unit, and that the mass as a whole was intruded at one and the same time and cooled and crystallized under the same conditions, and that the explanation of the peculiarities which it presents must be sought in another way—one which has an important bearing on theoretic petrology. As one approaches nearer the top no more black rock is seen; the re- mainder of the mass is of pure white or pinkish syenite and it presents everywhere the same even grain. The same platy structure continues and at times there are no talus slopes, vegetation, herbage, or even soil, only vast smooth white surfaces of naked rock on whose almost polished slopes it is impossible to climb. Towards the top the average thickness of the plates increases somewhat and their dip gradually becomes less until eventually they are horizontal. Itis by their breaking off when horizontal that the enormous ring-shaped, mural precipice which forms the top has been made. ‘The regularity of this platy jointing, together with the even rounding of the corners through weathering where the joint planes cross, gives a most remarkable likeness to colossal masonry in the upper walls. The mural front seen from the plains below has so close a resemblance to bedded and jointed sedimentary strata that only close inspection shows it isnot. On the top the slabs produced by joint- ing are often of great size, forming large tables of stone about 2 feet thick by 15 or 20 feet long and half as broad. ORIGIN OF THE PLATY PARTING, From what has been already said in regard to the platy parting which forms so marked a feature of Square butte, it will be seen that it bears the same relation to the mass as a whole as do the enfolding leaves of an onion to the bulb cut in half by a horizontal plane. We believe that they therefore represent parting surfaces parallel to the former covering of the laccolite from which the isothermal planes of cooling descended into the mass. We can conceive of no other hypoth- esis which would give a reasonable explanation of their arrangement and disposition, and since Square butte is unquestionably an intrusive mass we regard them as one of the strongest proofs of its laccolitic nature. Such an arrangement of the parting planes of a cooling igneous mass is by no means unknown, however, as it frequently occurs in the great phonolite domes of central Kurope. THE WHITE BAND. There still remains an interesting feature of the butte to be described. This is the presence on the south side of a band of white rock which LVIII—Butt. Gron. Soc. Am., Vou. 6, 1894. 406 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. passes through a number of the dark hoodoos near the lower portion of the dark lower zone. Looking down from the top, it is seen passing through successive hoodoos from right to left on the same horizon, and swinging, perhaps, one quarter of the way around the mountain. Its dip or hade is down and out, similar to that of the platy parting, but at a much more nearly vertical angle, so that the platy parting passes through it (see figure 5). Its shape as a whole, then, is like a segment of the surface of a truncated cone and it enwraps the mountain in a_ partial way as a bulb is partly enfolded by one of its leaves. The thickness of the band varies from one to two feet, averaging about 18 inches. Cutting across the dark hoodoos, it forms a striking and con- — ‘y Figure 5.—The white Band. spicuous feature,as may be seen in the sketch shown in figure 5. It was at first supposed to be a dike, but a study of it showed that this is not the case. It was found that there was no sign of contact between it and the dark rock through which it passes. The grain continues all through the same, but at a certain line the augite ceases, the feldspathic con- stituents increase and make up almost the whole mass of the rock. More convincing yet, as shown in the sketch in figure 5, the platy part- ing, with the remarkable disk-like structure, passes through both rocks alike, and hence it cannot be a dike, but must be an integral portion of the original liquid mass before it crystallized and cooled. It thus repeats on a smaller scale what has already been observed at the transition zone. CROSS-SECTION OF SQUARE BUTTE. 407 Its petrographic character is similar to the sodalite-syenite described later on. DIAGRAMMATIC SECTION OF SQUARE BUTTE. The facts which have been detailed in the foregoing pages may now be briefly recapitulated and summarized in the diagrammatic section shown in figure 6. If we were to pass a vertical axis through the center of the above sec- tion and revolve it upon this axis the figure of revolution which would be generated would represent quite correctly the structure of Square butte and the disposition of its several parts. Observe also in this connection the map on page 402. DS ce VERTICAL = HORIZONTAL SCALE MILES Fieure 6.—Cross-section of Square Butte. a= white syenite ; b = dark basic rock; ¢ = dark hoodoos: d= restored laccolitic cover ; e = up- turned Cretaceous sandstones; f= protruding sheet or edge of laccolite ; h = white band; 7 = tran- sition zone from white to dark rock, actual and imagined. SUMMARY OF FIELD RESULTS. From the facts thus shown we believe that Square butte is a laccolite consisting of two kinds of rock, an inner mass of an acid feldspathic variety surrounded by a zone of a basic augitic one. That it is nota case of one intrusion occurring on top of another is clearly shown by the facts already presented, and by the further ones that the relations of the light rock to the dark one are in nowise determined by the varying topography, as must have been the case were the black one a lower in- trusive sheet, and by the inclined circular plane of the transition zone, which has approximately the form of the surface of a truncated cone. Basic peripheral zones in connection with intruded masses of igneous rock, caused by the local concentration of dark colored ferro-magnesian minerals, are known and have been described by several authors, but, so far as we have been able to discover, no example has ever been seen or described before which illustrates them with such striking completeness of process and such perfection of erosiye dissection as Square butte. The significance of the facts and their bearing on theoretic petrology will be discussed in the latter portion of this paper. PETROGRAPHY OF SQUARE BUTTE. Characteristics and Minerals of the dark Rock—Megascopic and Micro- scopic.—The dark rock, seen at a distance, appears of a grayish black or 408 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. dark stone color, like many basic diorites. In the hand specimen, how- ever, it is found to be so coarse grained that the distinction between the dark colored ferro-magnesian components and the light colored feld- spathic ones becomes strongly accentuated, the contrast giving the rock a mottled appearance. Thus by inspection of the specimen one readily distinguishes the chief components. They are augite in well formed, often rather slender, idio- morphic crystals, of a greenish black color, attaining at times a length of one centimeter, but not averaging perhaps more than a quarter of that length, and biotite, of a bronzy brown color, whose occasional cleavage surfaces attain a breadth of from one to two centimeters, but whose out- lines are not clear and idiomorphic, but irreguler, dying away among the other components in shapeless patches. These biotites are, more- over, extremely poikilitic, inclosing the other components. With the lens these broad cleavages are seen to be made up of great numbers of smaller biotite individuals in parallel growths, but including the other minerals. They are thus, as one might say, spongy, skeleton crystals on a large scale. Filling the interspaces between these dark minerals is a white feld- spathic material, from which one obtains occasionally the reflection of a good feldspar cleavage. With the lens one detects greenish grains of olivine in addition. An inspection of the rock shows at once that its predominant character is the great abundance of the augite, which must form at least one-half of the mass by volume and a greater proportion by weight. With this large amount of augite, it is clear that if it were a dense fine grained rock instead of being so coarse grained as it actually is, a pronounced basaltic appearance would characterize it. In texture the rock is rather friable and crumbly, and blows of the hammer will frequently cause a specimen to fall into a coarse gravel. This is not due necessarily to alteration, but to the great number of pyroxene prisms and their idiomorphic character, there being little adhe- sion between their polished faces and the white feldspar material which fills their interspaces. A single heavy blow will often loosen these prisms so that the rock will crumble under the fingers. In thin-sections under the microscope the following minerals are found to be present: Apatite, iron ore, olivine, biotite, augite, albite, anorthoclase, orthoclase, sodalite, nephelite (?), cancrinite (?) and zeolites. Apatite.—This is the oldest mineral appearing in idiomorphie outlines even in or abutting into the iron ore. It is in short, stout prisms which often attain a length of 0.5 millimeter. Though commonly colorless, it is at times filled with excessively fine, dusty particles, and then becomes MINERALS OF THE SQUARE BUTTE DARK ROCK. 409 pleochroic: «= pale steel blue; » = pale leather brown. This dusty pig- ment is very apt to be confined to an inner core, which is surrounded by a clear colorless zone. Sometimes the apatites are of a pale red-violet brown and nonpleochroic. The crystals are bounded by the unit prism and several pyramids, but they were too small to determine the planes on material separated by the heavy liquids. The basal parting is common. Cases of twinning like that mentioned by Washington * were not observed. As shown by the analysis, the mineral is present in considerable amount. Olivine.—This mineral presents the usual type, but is at times of a very pale yellowish color in the section and then shows a faint but clearly perceptible pleochroism in tones of yellow and white. It is generally quite fresh, but sometimes has borders and patches of alteration into a reddish ferruginous material. Biotite.—The large cleavage surfaces of this mineral, made up of com- posite individuals,jhave been described above. It is strongly pleochroic, the colors varying between a very pale brownish orange and a deep umber brown. Cleavage plates appear uniaxial, but in the section, where very thin edges may be found, there is enough of an opening to the arms of the cross in convergent light to establish it as meroxene, the usual variety. The twinning and inclined extinction sometimes seen in the biotites of nephelinite and theralite rocks were not observed. Besides this [brown variety of biotite, there is present also in much smaller amount a pure deep green kind, which, from its method of occur- rence, we infer has been formed from the brown one. All gradations are found between them, but in such cases the brown forms an inner core which changes to green on the outer edges. This green kind is partic- ularly to be seen around the olivines, and especially where they come in contact with orthoclase. The appearance of these colorless olivines surrounded by this deep green jmantle is very striking. This variety shows very little change of pleochroism or absorption; it is uniaxial, and its double refraction is equally strong with that of the brown. It is quite irregular in outline. The intermediate position that biotite, in respect to its icmanenl nature, holds between olivine and feldspar has been noted by Iddings ft and is shown in the analysis of its formula. Thus if we consider typical biotite as (HK), (MgFe), Al, 5i,0,,, this separates into (MgFe), SiO, + (HK), O + ALO, + 2 $10,, thus furnishing olivine and the oxide molecules necessary for orthoclase. It is possible that this intimate relation may condition the appearance of secondary biotite where olivine and ortho- clase are contiguous. * Jour. Geol. Chicago, vol. ili, 1895, p. 25. 7 Origin Igneous Rocks: Bull. Phil. Soc. Washington, vol. xii, 1892, pp. 165, 166. 410 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. Pyroxene.—Of all the ferro-magnesian minerals this is by far the most important, determining with the orthoclase the essential character of the rock. Owing to the ease with which it may be de- tached from the matrix, excellent specimens may be obtained for crystallographic study. In general they present the common form of augite bounded by the planes a (100), b (010), m (110) and s (111), and somewhat tabular on a (100). The form 0(221) has also been observed. Twinning on a (100) oc- curs, and a crystal of this type having the form Ficure 7.—Twinned Pyrowene g (221) in addition is shown in figure 7. This Crystal. : 2 crystal was measured on the reflecting goniometer with the following results : Theory. Measured. aam (100 110)...... 46° 25’ 46° 27’, 46° 23’, 46° 42, 46° 51/ SN SICREPN ELD) ss Sete: Hoa og) 48 MUA OCLLOTA 220) ne 85 29 35 40 SAS (111 AIL twin)... 26 52 26 24 26 309 The reflections of the signal were only moderately good, and the meas- ured angles are therefore of value only in determining the faces. As this variety of augite is very common and persistent, not alone at Square butte but generally throughout the Highwood rocks, at times, however, passing into varieties which have a narrow mantle of material rich in the aegirite molecule, as, for example, aegirite-augite, it has been deemed important to investigate it chemically, especially since the Square Butte rock presents such excellent material. The analysis yielded the following results : Analysis of Pyroxene.* Oxygen ratios. SRR i tak Ss A tere AGAD so. 4) 18286 ae Olt et ee Be ts “O08? f -8303 PTA ai gine sue) A 1 Te ay 0415 10 TRS en aries a Sigg sc (0178 } ODER EO cede ihe ttle ello ee 0772 ETO NG ahh es ee AO cu. 0014 aera Moore te enews 13.58 ...... .3895) (1.04) | 8176 Cana G Rio tate (te heel ee eae "3995-3995 (1.00) NOM isu cee ACOA. oo oe. 0167 Trad aa aR aie ORE SoG ert 0040 f +9207 TAYO (Gib aU) al). 1 ae 09 PGE Ee st ea ere 100.21 * By L. V. Pirsson. ANALYSIS OF SQUARE BUTTE PYROXENE. All In the foregoing analysis the rock was crushed, sifted, and the result- ing powder washed and then separated by the use of Retgers’* silver- thallium-nitrate fluid in the apparatus devised by Professor Penfieid,t and by this means, aided by the magnet, material of exceptional purity was obtained. The comparison of the ratios in ne analysis shows that CaO to (feMg)O is as 1 to 1, and that the diopside molecule is thus chiefly present. The presence us the alumina suggests that Tschermak’s molecule RAI, SiO, must also be present. If we subtract from the sum of the RO molecules enough to make the number of the R,O equal to that of the R,O, and take out the same number of $10, molecules, the following table shows the composition of the augite : RO = .0386 PeOe—— 1190) SION ClO): IO. ae ae 0593 : R,O, = .0593 : SiO, = .0593::1:1:1. The very striking agreement of these ratios with the theory must cer- tainly be held to add another very strong proof to the correctness of Tschermak’s assumed molecule. The augite then has almost exactly the following composition: 13 Ca (MgFe) Si,O,+2 (Na,R) (AlFe), SiO,. Since the qualitative analysis of the feldspars has shown the absence of lime, if we deduct enough from the amount found by the mass analysis of the rock to turn the phosphoric anhydride into apatite, a comparison of the remaining amount, 11 per cent, with the 22 per cent of lime de- manded by the pyroxene, shows this mineral forms one-half of the rock by weight, a fact which agrees with the appearance of the hand specimen and the study of thin-sections. Orthoclase-—The predominant feldspar is orthoclase. This is shown by the study of thin-sections, by the separation of the feldspathic con- stituents by heavy liquids, and may also be inferred from the chemical analysis of the rock where potash is seen to greatly predominate over soda. The mineral is quite fresh and wholly allotriomorphie, its shape being determined by the angular interspaces between the pyroxene in which it is found. Sometimes it assumes rude lath-shaped forms. It is apt to be filled with fine interpositions whose exact nature can- not be told. They commonly possess the form of their host and their longer axis coincides with that of the crystal, and, so far as can be deter- mined, they are arranged in planes parallel to prism faces. They do not contain bubbles, the reflection band surrounding them is narrow and * Jahrbuch fir Min. 1893, vol. i, p.90. This most happy discovery of Professor W. Retgers has placed all working mineralogists and petrographers deeply in his debt. + We desire to express our thanks to Professor S. L. Penfield for kindly aid in making the sepa- ration in apparatus recently devised by him for the special use of the Retgers’ fluid, and by means of which the operation may be carried on with nearly the same ease and with all the certainty of the usual heavy liquids. 412 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. they do not act on polarized hight. From these facts we believe them to be of glass. Sometimes the orthoclase is colored a pale brownish tone by a fine dusty pigment. It shows in some places a slight tendency to kaoliniza- tion and in some others is discolored by the alteration of its interpositions, but usually it is quite fresh. The angle of the optic axes is variable, generally small and sometimes nearly zero. It sometimes shows intergrown patches of a feldspar which has a higher index of double refraction and is believed to be anorthoclase. In a few cases a tendency for orthoclase laths to group themselves in radial spherulitic forms starting from a common center were observed ; since the laths are broad and coarse it does not present a striking feature. Again, in other places the patches of orthoclase filling adjoining areas between augites and olivines have the same optical orientation over some distance, thus presenting a rude poikilitic effect. Plagioclase.—A triclinic striated feldspar is also present, but in no con- siderable amount. When the rock powder is placed in the mercuric- iodide solution, and the ferro-magnesian minerals, the magnetite and apatite have fallen out, no feldspathic materials are deposited until a specific gravity of 2.60-2.61 is reached. At this point a very smali pre- cipitate is obtained of a feldspar insoluble in HCl. Subjected to qualita- tive analysis it is found to be free from lime and gives abundant reaction for soda. It is therefore albite, which agrees with the optical character of the minerial in thin-sections, the extinction on either side of the albite twinning plane reaching a maximum of about 15 degrees. The study of this striated feldspar has shown that certain crystals possess remark- able properties. Thus the twinning lamelle, which are very narrow, can be seen in many cases very distinctly in ordinary light without using the analyzer, some of them possessing a higher refraction than others. Between crossed nicols it is seen that crystals possessing this peculiarity have no position of equal illumination, but the lamelle can be seen in all positions. It must be, therefore, that these lamelle possess a different chemical composition from those adjoining them, and since lime is excluded they must represent intergrowths of albite and anor- thoclase of varying composition, joined after this singular manner. Recently Federoff* has called attention to similar intergrowths of twin lamellee of different composition in the lime-soda feldspars and the same phenomenon had been studied and noted previously by Michel- Levy.t | Nepheline.—The presence of this mineral is only indicated by the fact that the powders falling between the specific gravities of 2.55 and 2.60 * Zeit. fur Kryst: vol. 24, Heft 1 and 2, 1894, p. 130. + Mineraux des Roches: Paris, 1888, p. 84. MINERALS OF SQUARE BUTTE DARK ROCK. 413 dissolve slightly in HCl, give a small amount of gelatinous silica, with reactions for Na and none for Cl, H,O or CO,. It must be present in the rock only as a rare accessory mineral, and the recognition in the thin- sections of an occasional patch is rendered difficult by the practically uniaxial character of some of the orthoclase. Cancrinite.—This is indicated by the fact that the rock powder obtained at a specific gravity of 2.47 dissolved in HCl with gelatinization, and in dissolving slowly and continuously gave off CO,, while carbonates, which would have been thrown down at a higher specific gravity, are absent in the rock, as seen in thin-sections. It can be present only in very small amount, and the certainty of recognizing an occasional piece in the sec- tion is diminished by the common occurrence of natrolite. The two minerals are alike in their appearance in fibers with parallel extinction. The cancrinite has, it is true,a higher double refraction, but sections may be as low as natrolite, and only by establishing the uniaxial character ean the cancrinite be definitely determined. This we have not been able to do, and its presence is therefore only inferential. Sodalite.—This also occurs as an accessory component. The rock powder separated below a specific gravity of 2.40 consists partly of this mineral, together with some zeolites. It dissolves readily in HCl ane. HNO,, the solution in the latter yielding a precipitate with AgNO, and none with BaCl,, thus showing the presence of sodalite and absence of hauyn or nosean. In thin-section it is very clear and limpid, but con- tains little interpositions somewhat like the feldspars. The actual amount of sodalite in the rock is very small, and this is shown also by the small amount of chlorine obtained in the analysis, part of which belongs to the apatite present. Natrolite—The presence of zeolites is indicated by the water obtained in the analysis. Some analcite may occur, but the chief zeolite is natro- lite, which is present in considerable amount. It is recognized by its parallel extinction and positive character, by the small angle of the optic axes, and by the strength of its double refraction, which compared with the feldspars, rises to .010-.012. It occurs in characteristic bundles of fibers, and is in part secondary after sodalite and in part after albite and anorthoclase. The fibers are plainly seen eating their way into the feldspar, and in a given crystal they do this according to a definite oriented direction, as the different patches in the:crystal always have the same orientation. Chemical Composition.—The chemical composition of the rock is shown in the following analysis. In it the minute trace of CO, due to a little possible cancrinite is not determined, nor is the amount of rarer LIX— Butt. Gro. Soc. Am., Vor. 6, 1894. 414 WEED AND PIRSSON—HIGHWUOD MOUNTAINS OF MONTANA. . elements which could not influence its results. The very large amount of P,O, is noticeable, and proves what the microscope reveals, the large amount of apatite present. The amount was fixed by two closely agree- ing determinations. Analysis of Shonkinite.* STOR haem Stir teen ne ct a en ee en 46.75 PiOk ey ache PS eRe cae eee ee ee eae 78 AD OB paideaie- lg SMh cna Sante Geet SER AR LA ee 10.05 HE; Oe: e tvers cacttian cise idiots ieeee om Oe RO ene 3.03 L e( O arene AA Ce I Che PRAM re me. APRON ToS 8.20 Hi) Eval @ areneeyetet sino are em Pe Nee SE er are Mubecree Coe GeO gh ETT 28 NEO) oe cena a ee gat ye cits aia ety ae 9.68 GO ae ie tee SM AR ne tg ER eee Mun Sa nant do le), 2. Ie ire4 On, ei Pa Te i Oe Siena ie ene 1.81 1 GL G Se ee vy Pie NRC Fane, Pema eed (8 NG | 3. 1G Ps Olnss cl: ete Sate bode Se eualercs She ee ee ee 1.24 Bes Ose sean Ry whats pndualw Wis ara Baas Sacto oe See 1.51 (0) BARC Gt ey ee RETR RAC R ME Cee ey 18 100.97 OS@k as faeces A Boe ee oe ee 04 4) WO (2) De SR ae ie a ae ae ra Ce pear re mnleertys Jay) -6.$ 100.93 To be noted here is the low silica and very high magnesia, iron and lime. It is evident that although the feldspar raises the silica percentage it is not in sufficient amount to counteract the olivine, iron ore, biotite, apatite and other mineralS which tend to lower it. The water comes in part from zeolites. Structure and Classification.—The minerals in the order of their erys- tallization are, first, apatite, then iron ore, olivine, biotite and augite. The period of the last two overlaps. Then followed the feldspathic com- ponents, whose succession is quite doubtful as regarding one another, except that on the whole the albite-anorthoclase group appears to be among the earliest. These minerals lie unoriented, forming a holocrystalline, rather coarse granular hypidiomorpic structure. It resembles in many respects the coarser grained theralites of the Crazy mountains; in others certain coarse- vrained dolerites. The structure is illustrated on the next page by figure 8, which indicates also the prominent position the augite plays in the composition of the rock. From what has been given in the foregoing description it is evident * By L. V. Pirsson. STRUCTURE OF THE SQUARE BUTTE DARK ROCK. 415 that in this dark rock of Square butte we have a granular, plutonic rock, composed essentially of augite and orthoclase, with smaller amounts of olivine and iron ore and with accessory apatite, sodalite, nepheline, etcetera. In its chemical composition it stands very close to certain vogesites and minettes—basic rocks of the syenitic group. It differs from them essentially, first, in its mineral composition and, second, in its structure. Fora rock of its character there seems to be no position in any of our present schemes of classification. It would be manifestly improper to term such a rock an augite-syenite, as its chemical compo- sition removes it very far from syenites. It bears indeed such a relation to augite-syenite as vog- esite does to hornblende-syenite ; that minette or, perhaps better, the Durbachite of Sauer * does to mica-syenite. It stands generally related in- deed to rocks of the basic class— low in Si0,, high in MgO, CaO and FeO, and thereby related to rocks of the lamprophyre family. Moreover, this type is found in the Highwoods not only in the outer mantle of Square butte, al- though constituting there an im- MeMeewiiass, OMt at Many obher 4 = aueite: O— olivine: B= biotite: Or—sor points forming great intrusive thoclase; An = anorthoclase; A = apatite. Actual stocks. As briefly noted by Lind- *'*™™ eren,t the variability of the augite and orthoclase in the Highwood rocks is very great. Asin the gabbro family we have every range from anor- thosite at one end to peridotites at the other, with the gabbros standing in an intermediate position, so in the Highwoods variation extends from syenites practically devoid of ferro magnesian minerals to those in which augite becomes the chief constituent, though the basic extreme entirely devoid of feldspar has not been observed by us. Name Shonkinite.—For this type of rock, then, we propose the name of shonkinite, from shonkin, the Indian name of the Highwood range, by which name, indeed, it is still called by many, and shonkinite we define as a granular: plutonic rock consisting of essential augite and orthoclase, Ficure 8.— Micro-drawing of Shonkinite multiplied 14 Diameters. * Mitt. d. Bad. geol. Landesanstalt, ii Bd., p. 247. 7 Proc. California Acad. Sci., ser. 2, vol. iii, p. 47. ‘Tenth Census, vol. xv, p. 725. 416 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. and thereby related to the syenite family. It may be with or without olivine, and accessory nepheline, sodalite, etcetera, may be present in small quantities. The Square butte rock is thus olivine-shonkinite, with these accessory minerals. The fine grained dense porphyritic forms which bear the same relation to shonkinite that trachyte does to syenite are dark to black heavy basalts. They are, in fact, orthoclase basalts, a type which although so far as we know has not yet been described from European localities, is by no means rare in western America. Besides its occurrence in the Highwoods, and also in other localities in Montana alluded to by Lindgren,* its presence in the Absaroka range and Yellowstone National Park has been men- tioned by Iddings.—| Somewhat similar rocks have been also mentioned by Zirkel,f who does not, however, discuss this type of basalts in the recent edition of his great work on petrography, so far as we have been able to discover in the absence of complete indexing. White Rock or Sodalite-syenite—The petrography of the light colored inner core of the denuded laccolite has been so completely investigated by Lindgren and Melville § that a further examination enables us ‘to add but very little to their comprehensive description. The rock is shown to be a sodalite-syenite, and for purposes of convenience we briefly sum- marize their results, referring to the original paper for fuller information. Megascopically the rock when very fresh is nearly pure white, often with a brownish to pin‘xish tinge, consisting mainly of feldspar, which often reaches. 5 millimeters in diameter. Through this are scattered slender, glittering black hornblende prisms which attain at times the same length. It is scarcely sufficient in amount to detract at a distance from the general whiteness of the rock. Small grains of a salmon to brown colored sodalite are also present. The rock is thus rather coarsely gran- ular, and in fact of the same size grain as the shonkinite, with which it is so intimately connected. The microscope shows the following minerals present in the order of their formation: Apatite, hornblende, orthoclase (with some albite), sodal- ite and analcite. The hornblende is in slender prisms bounded by m, 110 and 6, 010, terminations wanting, frequently twinned on a (100). It is strongly pleochroic ¢ and 6, deep brown a, yellowish brown and ab- sorption very great } —¢ >a. An outer mantle often shows a greenish color (from change into the arfvedsonite molecule ?—L. V. P.). Angle c /\¢ = 18 degrees; is idiomorphic against the feldspathic constituents. * Loe. cit., p. 50; also, Am. Jour. Sci, vol. 45, 1893, p. 289. + Bull. Phil. Soc. Washington, vol. 12, 1892, p. 169. { Mic. Petrog. Fortieth Par., 1876, p. 225. 2 Loe. cit. WHITE ROCK OF SQUARE BUTTE. A417 It is closely related to barkevikite, as shown by the analysis quoted later in this article. The orthoclase occurs in lath-shaped forms and in irregular grains. Those abutting against sodalite show crystal faces. Associated with the orthoclase is a triclinic feldspar referred to albite. The sodalite is found in irregular grains between the feldspars, allotriomorphic in regard to them, idiomorphic against analcite. The latter, which is in considerable amount, was along with the sodalite separated and analyzed. The anal- cite is thought to be derived from the albite. The rock is calculated from the analysis (given later in this paper) to consist of 66 parts of feldspar, 23 of hornblende, 8 of sodalite and 3 of analcite. In addition to these facts we have only to add that in the additional material studied by us we have detected a small amount of nephelite, which is being changed by borders, bays and tongues of analcite eating into it and thus suggesting an additional origin for the analcite; also considerable natrolite is sometimes present. Its fibrous masses are sec- ondary after sodalite and at times it completely replaces it. GENERAL PETROLOGY OF SQUARE BUTTE. The facts which have already been given in regard to Square butte show it to be one of the most remarkable and interesting occurrences of an igneous rock that has been described and from a petrologic point of view one of the most important; for while the differentiation of a molten magma as a factor in the formation of igneous rocks is now regarded by the majority of petrologists as an established fact, it is also true that the theory has been founded almost entirely upon inferential proof and by the exclusion of other hypotheses. The direct proofs which have come under observation have not been all that could be desired, and some of them indeed, as in the case of mixed dikes, have had more than one interpretation. In the case of Square butte, however, the proof of differentiation is unequivocal and direct, for in no other rational way, we believe, would it be possible to explain the disposition of the rock masses, the cone-in- cone arrangement of the two differing masses of intruded igneous rock, so unlike in chemical and mineral composition, yet geologically a unit and absolutely homogeneous in granularity and texture and so perfect in continuity of structure and platy parting. It is therefore a matter of interest to compare the chemical and min- eral composition of these two rocks, the syenite and shonkinite, with one another and see, so far as possible, how and under what conditions the differentiation has taken place. For this purpose the analyses of the two rocks are here compared : 418 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. Rock analyses. Chief oxides to 100. Molecules. A B Al B A? EP SiO, ... 56.45 46.73 DIOy.. .. pies 48.36 65.61 49.27 IGS ao cae 18 ALLO; 205B7 10.40 13.62 3a AVVO. .~' 20.08 10.05 ReO:, 2. Quant 11-78 5.39 10.02 HejOs. 2 isi 3.580! MoO), . e264 «SAL 1.10. = 15.98 FeO.... 4.39 8.20 GaO: 6222.19 13.68 2.65 14.84 MnO... 7:5 09 .28 Na.O.-. ) OD 1.88 6.33 1.82 MaQey- an. 26x 9.68 KEO" 2 aad 3.89 5.30 2.50 Ca@). 2 65) 2Al4 13.22 - —— NasOW.. o:6l 1.81 100.00 100.00 100.00 100.00 KOs. 8 eS 3.76 Ons. We 1.24 Pesce.) ES 1.51 (Ol ee 48 18 100.45 100.97 O= Cl. : AQ <1 104 Total.. 100.35 100.93 In the above table the analysis of the syenite by Melville is given under A; that of the shonkinite by Pirsson is repeated under B. For purposes of more easy comparison they are repeated under A' and B', with the non-essential elements omitted, the ferric iron reduced to ferrous, and the whole brought to 100. This at once brings out the most important chemical characteristics of the shonkinite. its very high iron, lime and magnesia, properties which show its difference from the typical syenites and its approach to the basaltic and lamprophyre groups. In A? and B? are given the percentages of molecules in the rocks derived from the oxygen ratios. The percentages by molecules gives in general a much clearer idea of the chemical composition of a rock than that by weight, because it shows more correctly its capacity for forming minerals. From the above table it is seen at once that the magnesia shows the ereatest differentiation, then the lime, and then iron. The relative pro- portion of the alkalies to each other and to alumina is about the same in each; they vary some, it is true, but the variation is insignificant com- pared with that of the bivalent oxides. The tendency of variation, then, has been for the lime, iron and magnesia molecules toward the outer cooling surface, while the alkalies and alumina have remained a constant, or if we imagine the silica to remain a constant, they have moved in- wardly. It is also clear that the bivalent oxides have not kept a nearly constant ratio, for magnesia is much more concentrated than iron. Of course, this implies that the molten mass before intrusion into the laccolite cavity was of uniform composition; that one liquid mass of one kind was not succeeded by another of different composition. The GENERAL PETROLOGY OF SQUARE BUTTE. 419 very regular and symmetric arrangements of the parts, the absence of all inclusions or “schlieren,” the cleanness of the zonal edge, together with the common properties already pointed out, utterly preclude this idea. There are, indeed, places in the Highwoods where intruded masses show further movements after differentiation has taken place, with the result of remarkably banded and streaked rocks, whose very occurrence shows that such was not the case at Square butte. We are, indeed, forced to conclude at every step that the mass was originally homogeneous, and that differentiation took place by the dif- fusion of the bivalent oxides toward the outer surfaces. It would add greatly to the value of the results here presented if we could know or could obtain the composition of the original magma in which the differentiation took place. This, however, cannot be done by comparing the masses of the two rocks, because, although it is probable that the amount of syenite now present represents pretty nearly the original one—that is, that there has been only a small erosion of that rock—the case is quite different with the shonkinite,a very large part of which has been carried away; hence, not knowing the relation of the two masses involved, we cannot estimate the composition of the original magma. It is evident, however, that it must have been between the syenite and shonkinite. Shonkinite, however, occurs in large bodies in the neighborhood of Square butte and elsewhere throughout the Highwood range, while rocks closely related to it in chemical and mineral composition are found in the form of dikes, extruded lavas and breccias. Throughout the district what may be called acid or highly feldspathic rocks play but a subordi- nate rdle. In view of these facts, we are inclined to believe that the com- position of the original magma approximated more closely to shonkinite tnan to the syenite. It will be seen, therefore, that Square butte presents in a demonstrative way the same idea that Brégger inferentially deduced and presented as the explanation of the processes of differentiation by which the varied rocks of the region of south Norway have been formed.* Recently Harker7 has described an interesting occurrence of a gabbro massif, which grows steadily more basic or richer in the ferro-magnesian minerals as the outer boundary is approached. Harker explains this occurrence by pointing out that the order of concentration of the min- erals is the same as the order of their crystallization, and hence accounts for the differentiation as a process of crystallization. Square butte is algo more basic as we approach the outer boundary, but the transition occurs abruptly, so to speak, or within such a narrow zone that it practically * Zeit. fur Kryst., vol. xvi, 1890, p. 85. + Quart. Jour. Geol. Soc., vol. 1, 1894, p. 311. 420 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. does. It is evident, however, that differentiation did not take place at Square butte as a process of crystallization, but in a liquid magma before any crystallization occurred. This is rendered quite evident, since none of the ferro-magnesian minerals of the shonkinite are found in the syenite. The only one, indeed, which is found in the syenite is the barkevikite-like hornblende, while in the shonkinite are found iron ore, biotite, olivine and pyroxene. Thus Square butte affords a striking con- firmation of the ideas recently expressed by Brogger in his remarkable work on the basic rocks of Gran.* ‘ It is a matter of some interest here to compare the comipeninad of the augite of the shonkinite, by far its most prominent constituent, and the hornblende of the syenite from Melville’s analysis. Barkevikite. Augite. SiO hei Aue! same ie Dalal Ue ea ai 38.41. 49.42 SIO) Beis’, ce page tink oo ph Ee ee eRe 1.26 55 AO epee | eden tele BAe Ware eH 16.39 4.28 Bes Oya vi hematin ea UC OL ee eee 3.75 2.86 REO kactaw eee Ty laseaenaians tage Sc eee 21.75 5.56 Winn, each see SA po eae 15 10 MeOs 8.5 e vA tsi le Mah een ery ee 2.54 13.58 CHO: ear Mgrs 106 Oink Ree store oak 10.52 22.85 NEO Mem et olen etn rene TOE) Dot ee 2.95 1.04 TKN ails aN eee sy RN 2/0 pee 1.95 38 TET) ARR aoe tas Ua ae me cote ad tea 24 09 99-9 100.21 The result of the increase of magnesia and lime shows itself in the change in composition of the dark mineral. The iron shows a moye- ment in the opposite direction ; in the syenite it is all found in the horn- blende; in the shonkinite large quantities had been used for the iron ore and olivine, and to some extent for the biotite before the augite began crystallizing; hence it is not so prominent as in the barkevikite. In general, however, the difference is of like kind with that shown by the mass analyses of the rocks and shows clearly how the composition of the prominent dark mineral is a function of the magma in which it is formed. ‘That minerals indeed are so often conditioned by the magma in which they are formed is without doubt the fact that has given to some the idea that definite mineral molecules individualized as such can exist in the molten magma. Recently Johnston-Lavis + has formulated a theory for the different composition of igneous rocks occurring at the same eruptive center by supposing that the body of molten magma which gave them birth was * Quart. Jour. Geol. Soc., vol. 1, 1894, p. 15. + Natural Science, vol. iv, February, 1894. GENERAL PETROLOGY OF SQUARE BUTTE. 421 originally homogeneous, but became of different composition on its outer margin by fusion and absorption of the country rocks with which it came in contact. Whether this is ever so or not is fairly a matter for argument. That such a process cannot, however, be appealed to as a general explanation is clearly shown at Square butte, where the outer margin, as already shown, is much more basic than the interior, and yet the magma has been intruded into sandstones—that is, rocks much more acid than the original magma. The singular white band which has been previously described as oc- curring on the south side of Square butte presents on a small scale the same process of differentiation between the syenite and shonkinite. We believe that it represents what may be called a residual differentiation— that is, that after the main process had already taken place and the outer margins of the laccolitic cavity were filled with that magma which was later going to cool and crystallize into shonkinite this further differentia- tion took place in the shonkinite fluid. The latter, probably owing to increasing viscosity, was not able to permit the white band fluid to pass in by diffusion to the main body of the syenite and rt therefore remained parallel to the transition zone of the two principal masses. , It will be noticed that a section passing from the center to the south of Square butte passes twice through white feldspathic and twice through dark augitic rock, if we take the white band into consideration. Fur- ther, that these various layers have a concentric arrangement with respect to each other, and hence one sees that Square butte presents on a huge scale a rude parallel to those spheroidal masses which sometimes occur in granites and diorites, and which are often remarkable for the regular concentric arrangement of spherical shells of varying composition. Backstrom * has sought to explain certain cases of such spheroidal masses as portions of a partial magma separated out in the liquid state from a mother liquor, in which, by sinking temperature, they are no longer soluble. Backstrom has expanded this idea and sought a general explanation f for the differentiation of igneous magmas in a process of “ liquation,” by which is meant that an originally homogeneous magma by sinking tem- perature becomes unstable and separates into two or more fluids which are insoluble in each other—that is, non-miscible. It seems to us that the concentric arrangement of parts and the clear and sharp line of divis- ion between them at Square butte point very favorably to this view as * Geo]. Foren. Férh., Stockholm, Bd. 16, 1894, p. 128. 7 Jour. of Geol., Chicago, vol. i, 1893, p. 773. LX—Butt. Geo. Soc. Am., Vou. 6, 1894. 422 WEED AND PIRSSON—HIGHWOOD MOUNTAINS OF MONTANA. an explanation. Backstrom, however, expresses himself as strongly against the idea of “ diffusion,” by which we suppose is meant the dif- fusion of the basic oxides toward the outer cooling surfaces. That such diffusion, however, can take place is clearly shown at Square butte, where it has. In any case a diffusion of some kind must take place or the magma would remain homogeneous. We do not see indeed that Backstrém has advanced any reason which would prove that these two ideas, diffusion and liquation, necessarily exclude each other. We do not see in fact why both may not be operative. As a matter of fact, the more that the differentiation of igneous rocks is studied the more evident it becomes that no one simple process will explain all cases, but that to produce such results a variety of factors must be included, any one or all of which may operate to produce a given phenomenon. Such, for example, may be pressure, change of tempera- ture, convection currents (which are shown by the “ flow structure ” and parallel arrangements of phenocrysts on the margins of intruded masses), diffusion of certain oxide molecules toward cooling surfaces, liquation and crystallization. The operation of these on molten silicate magmas is as yet but little understood and much more must be done and learned before any generally satisfactory theory for differentiation can be ad- vanced. Whatever may have been the causes at work at Square butte, two things at least are evident, that the basic oxides concentrated toward the outer edges and that the changes which produced this took place very slowly and with extreme regularity, allowing the differentiation to be very complete and thorough. SUMMARY. Square butte is a laccolite which has been intruded in Cretaceous sandstones. After the intrusion differentiation took place in the liquid mass, the iron, magnesian and lime molecules being greatly concentrated in a broad exterior zone, leaving an inner kernel of material richer in alumina, alkalies, and silica. This crystallized into a sodalite-syenite, while the outer mass formed a basic granular rock composed essentially of augite and orthoclase, to which the name of shonkinite has been given. After solidification the cooling developed a fine platy structure throughout the mass parallel to the form of the laccolitic cover. Since then erosion has removed the cover, laying bare the laccolite and dis- secting it so that its structure is clearly brought out. Owing to the erosion and the platy parting the broad marginal zone of shonkinite has been carved into a wide band of singular monoliths which extends around the mountain on its lower slopes. BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 6, PP. 423-528, PL. 27 APRIL 27, 1895 PROCEEDINGS OF THE SEVENTH ANNUAL MEETING, HELD AT BALTIMORE, DECEMBER 27, 28 AND 29, 1894 Herman LeRoy Fatrcntixp, Secretary CONTENTS Page PE momar y, December Qi. fects. 2 dune sew cous ov dde erlee een ek 424 emonivol the COUNCIL |. sic oe.ce chan core ves a Sia Maen Me ree tae eee RAED 2 ce ct 424 ‘SSE? EI PASS TSS OOH Fee a cag ete ret Ora OE 425 “SPER ISU RETE SSE OOH ahaa eG a ng uae RN nl le Ve Oe ol 429 JEVCUIGOIP Ss SRE OOF C, Ee eR Ee ee Med ne Ak eevee ae a en eS 429 LUE OIC LECIOGUCE) dc) See ee ee See Ey creer hre MA nee ae 431 Baueemasct oy tay epee ILC idl cei fis wi oe a e/g Sais abst eh d haa sca eboncann Wd byctagio melas 431 PImemamentsrto the: COMStibUObiON, o£ fic. fel nee bc dba be pe ae ee ve bos 431 PMREMCMEeMiLOMUNe DY AWS. Yc. og vac fac vee lak Gs Gbes*. pees aoures 432 Memorial of George H. Williams [with bibliography]; by W. B. Clark.. 432 Memorial of Amos Bowman [with bibliography]; by H. M. Ami........ 441 Breer memeOumlTiacy. I CCCMMDET 28) os lcci oo st sak hake e sew edele gee deale bee be bee 445 ep mmaOn ALCULIN, COMMMIELCE.s,. A saci ws ee fs ol 28 5 Pb ole see aed acdGied «ie dl .. 445 Fifth annual report of committee on photographs....................... 445 Report of committee on Royal Society catalogue... .................000. 457 High-level gravels in New England [abstract ; discussion by J. W. Spen- SET] 8. LON Qa BIB Es, Bela iol av eroxe) ake ile Re erie are retin De 460 Varanions of elaciers: [abstract]; Joy H. F. Reid. .........25 6.025.000 0006 461 Lake Newberry, the probable successor of lake Warren [abstract ; discus- sion by G. K. Gilbert, W J McGee, Warren Upham and J. W. Spencer] ; “On? EL, Tbe" Je niee a CIE eC sete a erage ge ee Te 462 Notes on the glaciation of Newfoundland [abstract]; by T. C. Chamberlin. 467 Oreanization of the temporary petrographic section............-..:..--. 469 Crystallized slags from copper smelting [abstract]; by A. C. Lane....... 469 The granites of Pikes peak, Colorado; by E. B. Mathews............... 471 Illustrations of peculiar mineral transformations [abstract]; by B. K. SEO -BIES OS gah os eet ema Res a Pe eh Re a Re PE 473 Spherulitic volcanics at North Haven, Maine; he Mes “Bayleyes- aces 474 A new intrusive rock near Syracuse eybeteactle by N. H. Darton and J. F. US GTTOY OGG Maeva el Oa na eee ce TRS eR LP RSIAY PA Meee coe Rae OL 477 I aommonm rmday eyenine, Wecember 28.6.5). Secs ee tog eee ee eee eee ee de dee 475 LXI—But1. Gron. Soc. Am., Vou. 6, 1894. (423) 424 PROCEEDINGS OF BALTIMORE MEETING. Page Session-of Saturday; Decemberi29). iii... els kine cine ae er 479 Cretaceous deposits of the northern half of the Atlantic coastal plain; by W.: Bo Clarke os ik ie Soares x pet eo eee aici ake ener S32 es 479 Surface formations of southern New Jersey; by R. D. Salisbury......... 483 Register of the Baltimore meeting, 1804. ..0.0..e 0 2. hcl eee 490 Officers and Fellows of the Geological Society of America.................-- 491 Aceessions to library to January, 1895... 2.0...) ieee seule oe ee 501 Index: to volume ’'6 2.0.6 225 sign coe © nis sche genels ct epee elm el ee 517 SESSION OF THURSDAY, DrcEMBER 27 The Society was called to order by the President, Professor T. C. Chamberlin, at 10 o’clock a m, in the geological laboratory of Johns Hopkins University, in which room all the sessions of the meeting were held. The President introduced Dr Daniel C. Gilman, the President of the University, who welcomed the Society in a cordial and graceful aa- dress, referring particularly to the geological equipment of the University and the recent opening of the building devoted wholly to geological science, in which this meeting was held, the occasion being in a sense an auspicious dedication of the building. He spoke with feeling of the loss to geology, the University and the Society by the death of Professor George H. Williams. President Chamberlin responded in a few words of thanks to President Gilman and the University. The report of the Council was called as the first item of business and was submitted by the Secretary in print and distributed to the Fellows. REPORT OF THE COUNCIL To the Geological Society of America, in Seventh Annual Meeting assembled : With this meeting begins the seventh year in the life of the Society. The Council congratulates the Fellows upon the eminent success it has achieved, and rejoices with them in the outlook for future prosperity and usefulness. The influence of the Society has been marked in the direction of more sympathetic codperation and harmonious working among the geologists of the continent. ‘Twelve meetings have been held, and the social profit of those gatherings has been even greater, perhaps, than the scientific. The five handsome volumes of the Bulletin are evi- dence of a working Fellowship and an active organization. Notwithstand- ing the great cost of the Bulletin and the expenses of administration, due REPORT OF THE COUNCIL. 425 to a scattered membership, it has been possible thus far, by careful man- agement and economy, to carry on the publication without abridgment. It has, however, been necessary to make a choice from the material offered for publication, and it will probably be necessary to make even more strict selection in the future. To the officers upon whom has fallen the burden of administration the success of the Society has been a great satisfaction. During the past year the Council has held two well attended meetings, in conjunction with the Boston and Brooklyn meetings of the Society, each consisting of several sessions. The details of the administration are shown in the following reports of the officers: SECRETARY’S REPORT To the Council of the Geological Society of America: Membership.—For the second time the Society has lost an officer by death. Second Vice-President George H. Williams died on July 12. Mr Amos Bowman died June 18. The last printed roll of membership bears the names of 220 living and nine deceased Fellows. At the Brooklyn meeting eleven persons were elected, and all have qualified, as follows: Miss Florence Bascom, R. C. Hills, E. D. Ingall, R. T. Jackson, D. F. Lincoln, C. J. Norwood, C. Pa- lache, L. V. Pirsson, H. L. Smyth, L. G. Westgate, W.S. Yeates. Five Fellows have been dropped from the roll for non-payment of dues and seven others are now so in arrears that they are liable to be dropped. Five candidates for membership are now before the Society. The Fellowship of the Society is at this date distributed over the con- tinent as follows: District of Columbia, 34; New York, 27; Canada, 28; Pennsylvania, 17; Massachusetts, 17; California, 12; Ohio, 12; Illinois, 10; Connecticut, 7; Iowa, 7; Minnesota, 6; Michigan, 5; New Jersey, 5; Kentucky, 4; Missouri, 4; Alabama, Colorado, Kansas, Texas, Vir- ginia, Wisconsin, 3 each; Maryland, South Dakota, Vermont, West Virginia, 2 each; Arizona, Georgia, Idaho, Indiana, Maine, Mississippi, North Carolina, New Hampshire, Rhode Island, Tennessee, 1 each, and 1 each. in Brazil, Burma and Mexico. Total, 229. Distribution of Bulletin—The Secretary calls attention to the matter under this head in his report of last year (volume 5, page 610), which need not be repeated here, except to state that the edition of volume 1 was only 500 copies, and that the first two volumes were distributed to the Fellows direct from the printers. A comparison of the following re- port with last year’s report will give the details for the past year: 426 PROCEEDINGS OF BALTIMORE MEETING. DISTRIBUTION OF BULLETIN FROM THE SECRETARY’S OFFICE DURING 1891-1894 Complete Volumes Voli, Vol.2. se 2S eines atta ceet se ieee wine ahs ieee aa ee N 1 66 APU RAD ROS SAGE he oes twine Ale, tae Ste eds ens fogs sete cal 4 cat ede lovee ea err 4 94 Stationery.-and meconds)) 4.52) fe ose ce ee wate bole che oop. eee eee 8 44 PriMtime- wneludine staheoneny 22 ss se sone eae ee wees 156 93 IMGeLIMe SEE? AS hic eye eet sere cee Lee. -« Sei. ee RWS eer 2 00 DAN Oy 2 08 1 ees geet ce emt ne ne Re ET US oo vee 14 96 PROG So be cig evans cee ow lve aie ak tock igs oe alle ane aie dre $229 11 Account of Bulletin ROStAG et testis PV cap OMe at OREN TEMES GA Rabanne es Shee Lo a $100 50 MCC CPAMIS ci os Ac kee Motes ee ole wile see eee evans Mn begetele ceeake es eee 70 EKA ESSAC! st. Lae alas oes aoe aie there ate ee ne eta eee eae ee 68 66 Wiesner At elas ieee tice, cots Sy atieee te eee Ne eae nt rr 1 90 Taio in Tay eer fee Sa Bees ae ere “arate ae Pic ne Cae seh Pt' f2 eae da ea 2.2) alae I TAA Os eyo ae aE Oe aes, Aik Aedes ees eG: er 2 50 Collechonvot-clecks steer eine es < sau ok Ager 2 25 Labor correction in volunmensia: ace tie erie ee 14 00 Mo tall phe dicis< center sacle yall De SERS tnt apc ate, eee ,< svelte eine eae $204 26 MoO talWexpem cure.’ + were ence reenter sets laa ue dh hep ee $433 3 All of which is respectfully submitted. ! H. L, FAarreuimp, Secretary. RocHEstER, New Yorx, December 21, 1894. * This list is printed as the closing matter of this volume. TREASURER’S AND EDITOR’S REPORT. 'TREASURER’S REPORT To the Council of the Geological Society of America : 429 In accordance with the By-Laws, a condensed statement of the oper- ations of the Treasury for the year ending November 30, 1894, is hereby submitted : RECEIPTS Balance in Treasury November 30, 1893........-.: 0 ...220...: $633 09 REPT MCONS MTL LOCS aciatocs vra)6 0 Scie igs Seas wl ee eles a ace sage 2a 1,880 00 MeamPGRIM Pe COTAET ECS oocee yates o/h cine ea wale F toa gs dee ame e woeias edhe 150 00 Meme MOMMMIbAMON CE yo liiedy ore cc cok ye cade Secale e oe dae ee 100 00 “UD EIMESE CLL, TRUEST COVE 01 ar ane 158 96 BAe OMMUNNICUGIONMS 612 2 GILBERT The paper is published in the Journal of Geology, volume iii, 1895, pages 121-127. ORGANIZATION OF THE TEMPORARY PETROGRAPHIC SECTION This division of the Society met at 11.80 o'clock a m in the “ Williams room ” of the geological building. Professor B. K. Emerson was made chairman and Mr Whitman Cross secretary. The first paper read was— THE RELATION OF GRAIN TO DISTANCE FROM MARGIN IN CERTAIN ROCKS BY ALFRED C. LANE The paper elicited discussion, in which the chairman and E. O. Hovey, J. F. Kemp, J. P. Iddings, G. P. Merrill, F. D. Adams and Whitman Cross participated. The second paper was by the same author: CRYSTALLIZED SLAGS FROM COPPER SMELTING BY ALFRED C. LANE [ Abstract ] The specimens of slags exhibited are from the smelting works at Dollar bay, and on Torch lake, in the copper country of upper Michigan. The copper as it comes from the mines, if not in masses suitable to be directly smelted into bars, is stamped and washed, and the resulting concentrates or ‘‘mineral’’ is melted into ingots. Most of the copper is found native, therefore, and remains so, but in the process of melting down a small part is oxidized. The waste and scraps of the direct process must therefore be reduced in a cupola furnace. Kelly Island limestone from lake Erie is used as a flux with the coal. It has been the custom at times to let the slag from the cupola furnace run into hemispherical iron pots, which are two feet in inner diameter and one foot deep and mounted on a carriage. The slag is then allowed to cool naturally and the solid contents dumped. In the interior of these pots, within the crust formed by the first cooling, cavities formed, and from these came some of the crystalsexhibited. These slags show in general a strong tendency to be crystalline, and are very often devitrified to within two centimeters from the outer surtace. 1. The most interesting crystals are the very large ones of melilite, one or two centi- meters square and up to one centimeter thick. The general form is that of square tablets, but the faces are rounded, exactly as if they were on‘ the point of being 470 PROCEEDINGS OF BALTIMORE MEETING. remelted and in a viscous condition. Carefully examined, they show a reticulated surface. re xxiv Kotet, 1894 (except pp. 261-352). LXXI—Buwtt. Geot. Soc. Am., Vou. 6, 1894. 504 PROCEEDINGS OF BALTIMORE MEETING. NORGES GEOLOGISKE UNDERSOGELSE, CHRISTIANIA 432. Selbu, 1890, pp. 1-39. 433. Aarbog for 1891, pp. 1-100. 434. Salten og Ranen, 1890, pp. 1-232. ACADEMIE ROYALE DES SCIENCES ET DES LETTRES DE DANEMARK, COPENHAGEN 428-431. Oversigt i Aaret, 1891-’94. 432. Fortegnelse, I Tidsrummet, 1742-1894, pp. x + 1-135. NATURWISSENSCHAFTLICHEN GESELLSCHAFT ISIS, DRESDEN 310. Festschrift, May, 1885, pp. 1-178. 311. Sitzungsberichte und Abhandlungen, Jahr 1892, pp. 1-51 + 1-124. SNe Pe ates a ‘¢ 1893, pp. 1-53 + 1-186. 313. + “4 sf ** 1894, pp. 1-20 + 1-82. ROYAL SOCIETY OF EDINBURGH, EDINBURGH 274-275. Proceedings, vols. xviii, 1890-xix, 1892. 276. Transactions, vol. xxxvi, 1889-’91, 3 parts (Nos. 1-23), pp. 1-786. PHM he i vol. xxvii, 1891-’93 (Nos. 1-24), pp. 1-528. NATURFORSCHENDEN GESELLSCHAFT, FREIBURG I. B. 314-316. Ben Band v, 1890-vii, 1893. 317. “viii, Jan., 1894, pp. 1-209. GEOLOGICAL SOCIETY OF GLASGOW, GLASGOW 278. Transactions, vol. ix, 1888-’90, pt. 1, pp. 1-240. PETERMANN’S GEOGRAPHISCHE MITTEILUNGEN, GOTHA Abdruck, 1892, heft xii. cs 1894, hefte vili-ix. KSL. LEOP.—CAROL. DEUTSCHEN AKADEMIE DER NATURFORSCHER, HALLE 326. Nova Acta, Band xl, 1878, Nos. 4, 8, 9. Bog) in SB Nog. al Ie: O28 E ‘f xliu, 1881, No.4. BP Bae o . xlv, 1882, No. 2. Sa0 merce - ‘¢ xlviti, 1885, No. 3. Soler = ty ie ct li, 1886, No. 1. eae te _ i hii, 1888, No. 1. S355 | 5 7 lvi, 1891, No. 3. SO4ee) te ¥ ss lvii, 1891, Nos. 3, 4, 5. GEOLOGISKA UNDERSOKNING, HELSINGFORS 367. Beskrifning a Kartbladet, 1879-1882, Nos. 1-21; 19 maps. 368. a ¥ 1894, Nos. 22-26; 4 maps. GEOLOGISCH-MINERALOGISCH MUSEUM, LEIDEN 318. 319-320. 321. 322. 323, 324. 125. 424-426. 427. 339-337. 435. 436. 279. 280. 281. 282. 283. 288-292. 293-296. 284. 285. 286. 287. ACCESSIONS TO LIBRARY. 505 KONIGLICH-SACHSISCHE GESELLSCHAFT DER WISSENSCHAFTEN, LEIPSIC Berichte ber die Verhandlungen Mathematisch-Physische Classe, 1891, ili-v, pp. 271-678. Berichte tiber die Verhandlungen Mathematisch-Physische Classe, 1892- OS : Berichte tiber die Verhandlungen Mathematisch-Physische Classe, - 1894, i, pp. 1-184. Abhandlungen der Mathematische-Physische Classe, Bande xviii, 1892, Nos. 2-8, pp. 66-492. Abhandlungen der Mathematische-Physische Classe, Bande xix, 1893, pp. 1-167. Abhandlungen der Mathematische-Physische Classe, Bande xx, 1893, Nos. 1-4, pp. 1-551. Abhandlungen der Mathematische-Physische Classe, Bande xxi, 1894, Nos. 1, 2, pp. 1-42. SOCIETE GEOLOGIQUE DE BELGIQUE. LIEGE Annales, Tomes xviii, 1891; xx, 1898. = ‘¢ xxi, 1893-94, Liv. 1, 2, pp. 1-28 + xevi + 168. SOCIETE GEOLIQUE DU NORD, LILLE Annales, xix, 1891—xxi, 1893. COMMISSAO DOS TRABALHOS GEOLOGICOS DE PORTUGAL, LISBON Communicacoes, Tom. I. e *¢ II, Fase. ii, 1892, pp. 129-287 + xxxii. BRITISH MUSEUM (NATURAL HISTORY), LONDON Catalogue of the Fossil Birds in the British Museum, 1891, pp. 1-368 + 17. Catalogue of the Mesozoic Plants in the British Museum, pt. 1, 1894, pp. 1-179 + 18. Guide to the Collection of Fossil Fishes, 1888, pp. 1-51. Guide to the Exhibition Galleries, Dept. of Geol. and Pal., pt. 1, 1890, pp. 1-98. Guide to the Exhibition Galleries, Dept. of Geol. and Pal., pt. ii, 1890, pp. 1-109. GEOLOGICAL RECORD, LONDON GEOLOGICAL SOCIETY, LONDON Quarterly Journal, vols. xlvi, 1890—], 1894. Lists of the Geological Society, 1891-1894. GEOLOGICAL SURVEY, LONDON GEOLOGISTS ASSOCIATION, LONDON Proceedings, vol. xii, parts 6-10, 1892, pp. 225-415. os vol. xili, parts 1-10, 18938-1894, pp. 1-415 + xii. List of Members, Nov., 1892, pp. 1-35. py Feb., 1894, pp. 1-36. 506 297-298. 344-349. 338-341. 342. 343. 350. 351. 352. 353. 354. 355. 356. 403-408. 385-387. 388. 389. 390. 391, . Bollettino, vol. PROCEEDINGS OF BALTIMORE MEETING. COMISION DEL MAPA GEOLOGICA DE ESPANA, MADRID SOCIETA ITALIANA DI SCIENZE NATURALI, MiLAN . Atti, vol. xxxiv, Fasc. i, 1892, pp. 1-135. SOCIEFTE IMPERIALE DES NATURALISTES DE MOSCOU, MOSCOW K. BAYERISCHE AKADEMIE DER WISSENSCHAFTEN, MUNICH RADCLIFFLE LIBRARY, OXFORD UNIVERSITY MUSEUM, OXFORD Catalogues of books added to the Library in 1892, 1893. ANNALES DES MINES, PARIS Annales, Tomes I, 1892-VI, 1894. COMPTOIR GEOLOGIQUE DE PARIS, PARIS Hennigire peclee due Universel, Tomes vi, 1889—ix, 1892. cs x, Fase. 1, ppal-tor Bulletin Trimestriel, No. 1, Jan., 1893; Catalogue Général, géologie, minéralogie, paléontologie, pp. 1-134. SOCIETE GEOLOGIQUE DE FRANCE, PARIS Bulletin, 3d se Tome xviii, 1890, Nos. 66 éé 1-9, pp. 1-979. xix, 1891, Nos. 1-18, pp. 1-1252. xx, 1892, Nos. 1-8, pp. 1-566. xxi, 1893, Nos. 1-8, pp. 1-707. xxii, 1894, Nos. 1-8, pp. 1-528. Comte-Rendu des Seances, 1893, 3d series, Tome xxi, Nos. 1-18, pp. exlviii. “ a 1894, 3d series, Tome xxii, Nos, 1-18, pp. exci. ée (5 (5 6c (35 (73 Ce ce ce REALE COMITATO GEOLOGICO D’ITALIA, ROME Bollettino, vols. xx, 1889—xxv, 1894. SOCIETA GEOLOGICA ITALIANA. ROME x, 1891, Fase. 2-5, pp. 99-1023. xi, 1892 Fase. 1-3, pp: 1-70 xii, 1893, Fasc. 1-4, pp. 1-891. xili, 1894, Fasc. 1, 2, pp. 1-202. 66 (79 . Indice dei primi dieci volumi, 1882-1891. . Statuto, Regolamento, etc., 1888. RASSEGNA DELLE SCIENZE GEOLOGISCHE IN ITALIA, ROME . Anno II, Fasc. 1-8, 1898, pp. 1-192. ACADEMIE IMPERIALE DES SCIENCES, ST PETERSBURG Bulletin, Nouvelle Serie, i (xxxiii)-ili (xxxv), 4to. 5 os 1 iy (xx vi), NOs dl, 2, pp. =auc. Memoires, Tome xli, No. 5, 1898, pp. 1-124. “ © xIii, No. 3, 1894, pp. 1-242. xlii, No. 5, 1894, pp. 1-93. 66 66 369-371. 372-375. 376. 377. 378. 379. 380. 381. 082. 383. 384. 392-398. 399-400. 448-470. 471. 444-447, 303-308. 309. 360. 361-362. 362-369. 066. 472. 473. 474-477. 478. ACCESSIONS TO LIBRARY. 507 COMITE GEOLOGIQUE DE LA RUSSIE, ST PETERSBURG Bulletin, 1891-18938, vols. x—xil. Supplement, vols. ix, 1889-xii, 1892. Carte Geologique de la Russie d’ Europe, 1893 (6 maps). Memoires, vou iv, Nos. € 3. - v, Nos. 1-5. viii, No. 2. en INOS 2 x; Nos iy 2: xe INOiz: xi Nox 2! Selb Owl. RUSSICH-KAISERLICHEN MINERALOGISCHEN GESELLSCHAFT, ST PETEBSBURG Verhandlungen, Zweite Serie, Band 24 (1888)—Band 30 (1893), 4to. Materialien zur Geologie Russlands, Band xiv (1890)—xvi (1893). GEOLOGISKA BRYAN, STOCKHOLM Sveriges Geologiska Undersdkning, Ser. C, Nos. 112-134. Systematisk Forteckning, 1862-1893, pp. 1-14. GEOLOGISKA FORENINGENS, STOCKHOLM Foérhandlingar, 1891, Band 13—1894, Band 16 (Nos. 134-161). 7 NEUES JAHRBUCH FUR MINERALOGIE, GEOLOGIE UND PALEONTOLOGIE, STUTTGART Jahrgang, 1892-1894 (6 volumes). Bielage—Band viii, 1-2 Heft, pp. 1-417. KAISERLICH-KONIGLICHEN GEOLOGISCHEN REICHSANSTALT, VIENNA Jahrbuch, 1892, Band xlii, 1, 2 Heft, pp. 1-386. a 1893 (Band xliii) ; 1894 (Band xliv). KAISERLICH-KONIGLICHEN NATURHISTORISCHEN HOFMUSEUMS, VIENNA ete Band vi, 1891—viti, 1893. ‘¢ ix, 1894, Nos. 1, 2, pp. 1-247 + 51. DIE BIBLIOTHEK DES HiDG. POLYTECHNIKUMS, ZURICH (c) ASIA. GEOLOGICAL SURVEY OF INDIA, CALCUTTA Contents and Index of the first twenty volumes of Memoirs, 1859-1883, pp xi. Memoirs, vol. xxiii, pp. x + 1-232 + xix. Records, vols. xxiv, 1891-xxvii, 1894. A Manual of the Geology of India, 2d ed., 18938, pp. 548. 484. 485. 486. 487-490 491 492. 493. 494-497. 498-499. 506-508. PROCEEDINGS OF BALTIMORE MELTING. GEOLOGICAL SOCIETY OF TOKYO, ; TOKYO . Journal of the College of Science, Imperial University, vol. iv, 1891, pt. li, pp. 239-366. . Journal of the College of Science, Imperial University, vol. v, 1892, pt. i-iv, pp. 1-353. . Journal of the College of Science, Imperial University, vol. vi, 1893, pt. i-Iv, pp. 1-885. . Journal of the College of Science, Imperial University, vol. vii, 1894, pt. i-ili, pp. 1-248. . Journal of the College of Science, Imperial University, vol. viii, 1894, pt. i, pp. 1-278. IMPERIAL GEOLOGICAL SURVEY, TOKYO (d) AUSTRALASIA GOVERNMENT GEOLOGIST, . ADELAIDE Annual Report for year ended June 30, 1894, pp. 1-26, 4to (with maps). Catalogue of South Australian Minerals, etc., by H. Y. L. Brown, Goy- ernment Geologist, 1893, pp. 1-34. On additional Silurian and Mesozoic Fossils from Central Australia, by R. Etheridge, Jr., 1893, pp. 1-8, 4to. | GEOLOGICAL SURVEY OF QUEENSLAND, BRISBANE Annual Progress Reports of the Geological Survey for the years 1890-1898. Ato. Twenty-six reports and papers, 1890-1894, 4to, with map of Chartre Tower Goldfield. From R. L. Jack, government geologist. Geology and Paleeontology of Queensland and New Guinea, by R. L. Jack and Robert Etheridge, Jr., 768 pp., with 68 plates and maps, 1892. The Mount Morgan Gold Mine, Queensland, by F. W. Sykes, 88 pp., 1893. From Mr. R. L. Jack. CANTERBURY MUSEUM, | CHRISTCHURCH GEOLOGICAL DEPARTMENT OF WESTERN AUSTRALIA, PERTH DEPARTMENT OF MINES, SYDNEY Annual Reports for the years 1889-1893, 4to. Records of the Geological Survey of New South Wales, vol. ii, 1890-1892 ; vol. ili, 1892-938. . Records of the Geological Survey of New South Wales, vol. iv, 1894, parts 1, 2, pp. 1-113. 2. Memoirs, Paleontology, No. 5, parts i, ii, pp. ix + 1-181, 4to. ce ep No. 8, parts i, pp. vii + 1-49, 4to. “ Geology, No. 5, pp. ix + 1-149, 4to. . Two geological maps of New South Wales, 1898, 4to. ROYAL SOCIETY OF NEW SOUTH WALES, SYDNEY Journal and Proceedings, vols. xxv, 1891—xxvii, 189 38. 509. 510. 511. 512. 518. 514. 515. 516. 517-520. 521. 522. 523. 524-528. 529. 530-531. 532-534. 539. 536. 537. 538. 539-546. 547. 548-556. 597-598. 559-566. ACCESSIONS TO LIBRARY. 509 (e) AFRICA GEOLOGICAL AND IRRIGATION BRANCH, CAPE TOWN (f) HAWAIIAN ISLANDS HAWAIIAN GOVERNMENT SURVEY, HONOLULU (B) From Stare GeotocicaL Surveys AND Minine Bureaus GEOLOGICAL SURVEY OF ALABAMA Report of Progress for 1876. E. A. Smith, State Geologist, pp. 1-100. Report on the Cahaba Coal Field, by Joseph Squire, 1880, pp. 1-189, with map. Report on the Coal Measures of the Platea Region of Alabama, by Henry McCalley, 1891, pp. 1-238. Report on the Geological Structure of Murphree’s Valley, by A. M. Gibson, 18938, pp. 1-132. Bulletin No. 1, 1886, pp. 1-85. ag No. 4, 1892, pp. 1-85. GEOLOGICAL SURVEY OF ARKANSAS | Report of Joint Committee to the General Assembly of Arkansas, 1889, pp. 1-19. Report of Joint Committee to the General Assembly of Arkansas, 1891, pp. 1-25. peal Report for 1888, vols. i-iv, John C. Branner, State Geologist. a ‘* 1889, vol. 11, Crowley’s Ridge, pp. 1-283 + xix. a ae “¢ 1890, vol. i, Manganese, pp. 1-642 + xxvii. CALIFORNIA STATE MINING BUREAU Annual Report of the State Mineralogist, Henry G. Hanks, 1880, pp. 1-43 Reports of the State Mineralogist, second, 1882—sixth, 1886. Reports of the State Mineralogist, eighth, 1888, William Irelan, Jr., pp. 1-948. Reports of the State Mineralogist, eleventh, 1893; twelfth, 1894, J. J. Crawford. Bulletin, Nos. 2-4. First Annual Catalogue of the State Museum of California, 1852, pp. 1-350. Catalogue of the State Museum of California, vol. 2, 1885, pp. 1-220. Catalogue of Books, Maps, etc., in Library, 1884, pp. 1-19. GEOLOGICAL SURVEY OF GEORGIA The Paleozoic Group, by J. W. Spencer, State Geologist, 1893, pp. 1-406. GEOLOGICAL SURVEY OF ILLINOIS Volumes i, 1866, to viii, 1890, A. H. Worthen. Volumes viii, plates, 1890. GEOLOGICAL AND NATURAL HISTORY SURVEY OF MINNESOTA Annual Reports: N. H. Winchell, State Geologist, 1 (1873), 2, 11, 14, 16, Ag 19, 20; 21 (1892). Geology of Minnesota, Final Report, vols. i (1872-82), ii (1882-’85). Bulletin, Nos. 1, 2, 4-8, 10. 510 567. 568. 569-573. 574-575. 576-577. 578. 579. 580. 581-588. 589-590. 591-705. 706-708. 709. 710. (AL: PROCEEDINGS OF BALTIMORE MEETING. GEOLOGICAL SURVEY OF MISSOURI First and Second Annual Reports: G. C. Swallow, 1855, pp. 1-240. Report of 1873-74: G. C. Broadhead, 1874, pp. 1-734 + xlix + 4, with atlas. Bulletins, Nos. 1-5: Arthur Winslow, State Geologist. GEOLOGICAL SURVEY OF NEW HAMPSHIRE First (1869) and Second (1870) Annual Reports upon the Geology and Mineralogy: Charles H. Hitchcock. Reports of Progress, 1871, 1872. Geology of New Hampshire, vol. ii, 1887, C. H. Hitchcock, pt. ii, pp. 1-684. Geology of New Hampshire, vol. iii, 1878, C. H. Hitchcock, pts.. iii-v. GEOLOGICAL SURVEY OF OHIO Rericht ueber den Fortgang in 1870: J. S. Newberry, 1872, pp. 1-561. Reports, vols. i, 1873-vi, 1888 (some in 2 parts): J. S. Newberry, State Geologist. SECOND GEOLOGICAL SURVEY OF PENNSYLVANIA Reports of the Board of Commissioners for 1875, 1876. Reports and Maps: J. P. Lesley, State Geologist. (A complete set of the publications to date.) . GEOLOGICAL SURVEY OF TEXAS First (1889), Second (1891), and Fourth (1893) Annual Reports: E. T. Dumble, State Geologist. (C) From Screntiric SocrETIES AND INSTITUTIONS (a) AMERICA TEXAS ACADEMY OF SCIENCE, AUSTIN Transactions, vol. i, Nos. 1, 2, Nov., 1892—Nov., 1893, pp. iv + 1-102. LOUISIANA STATE EXPERIMENTAL STATION, BATON ROUGE Geology and Agriculture: Prelim. Report upon the Hills of Louisiana, by’ Otto Lerch, 1893, pp. 1-158. FIELD COLUMBIAN MUSEUM, CHICAGO Pub. 1, vol. 1, No. 1, History and Desc. Account of the Museum, 1894, pp. 1-90. KANSAS UNIVERSITY QUARTERLY, LAWRENCE . Vol. ii, Nos. 3, 4, 1894, pp. 99-290. Vol. iii, Nos. 1, 2, 1894, pp. 1-163. UNIVERSITY OF WISCONSIN, MADISON . Bulletin, Engineering Series, vol. i, Nos. 1, 2, 1894, pp. 1-40. AMERICAN GEOGRAPHICAL SOCIETY, NEW YORK . Bulletin, vols. xxiv, 1892-xxvi, 1894. 727-730. 740. fc ACCESSIONS TO LIBRARY. 511 ROCHESTER ACADEMY OF SCIENCE, ROCHESTER . Proceedings, vol. i, Broch. 1, 2, 1890-91, pp. 1-216. a vol. ii, Broch. 1-8, 1892-94, pp. 1-228. TECHNICAL SOCIETY OF THE PACIFIC COAST, SAN FRANCISCO . Transactions, 1893, vol. x, Nos. 3-10, 12, pp. 27-314 (except 273-291). DIRECCION DE OBRAS PUBLICAS DE CHILE, SANTIAGO . Revista, Ano i, Num. 1, 1890, pp. 1-115. NATIONAL ACADEMY OF SCIENCES, | WASHINGTON . Report for the year 1891, pp. 1-39. NATIONAL GEOGRAPHIC SOCIETY, WASHINGTON . National Geographic Magazine, vol. 1, Nos. 3-4, pp. 183-335. 66 ce ce 66 ii, Nos. 1-5, pp. 1-285. ef a - ** 10, pp. 1-30, 53-204. ‘ os ay ‘fv, pp. 96-256, 257-263. oe “ a < -Vi, pp: 1-238: (b) EUROPE GEOGRAPHISCHEN GESELLSCHAFT, BERNE Jahresbericht ix, 1888—xii, 1893. FACULTE DES SCIENCES DE CAEN, CAEN . Bulletin du Laboratoire de Geologie 17° Année Nos. 1-7, 1890-92, pp. 1-278. SOCIETE DE GEOGRAPHIE DE FINLANDE, HELSINGFORS Bulletin, Fennia 5, 1892 (various separate papers). a ‘« 9, 1894 (various separate papers). ‘¢ - 11, 1894, pp. 1-225. INDUSTRISTYRELSEN I FINLAND, HELSINGFORS . Meddelanden, Fjerde Hiftet, 1887, pp. 1-100. Attonde Hiiftet, 1888, pp. 1-94. CONGRES INTERNATIONAL D’ANTHROPOLOGIE ET D’ARCHEOLOGIE PREHISTORIQUES, LISBON . Compte Rendu de la Neuvieme Session a Lisbonne, 1880, pp. xlix + 1-728. CONGRES GEOLOGIQUE INTERNATIONAL, LONDON . Compte Rendu de la 4™° Session, Londres, 1888. GEOLOGICAL INSTITUTION OF THE UNIVERSITY OF UPSALA, UPSALA Bulletin, vol. 1, No, 1, 1892, pp. 1-98. LX XII—Butt, Gror. Soc. Am., Vou. 6, 1894. 512 PROCEEDINGS OF BALTIMORE MEETING. (D) From Frettows oF THE GEoLoGIcAL Socrery oF AMERICA (PERSONAL PUBLICA- TIONS) FRANK D. ADAMS 741. Seven pamphlets. W.S. BAYLEY 742. Six pamphlets. J. C. BRANNER 743. Fifteen pamphlets. G. C. BROADHEAD 744, Five pamphlets. H. P. H. BRUMELL 745. One pamphlet. J. H. CHAPIN 746. The Creation and the Early Development of Society, 1885, pp. 1-276. 747. From Japan to Granada, 1889, pp. 1-325. WHITMAN CROSS 748. Twenty-two pamphlets. G. KE. CULVER 749. Three pamphlets. H. P. CUSHING 790. Three pamphlets. T. NELSON DALE r) 751. Seven pamphlets. W. M. DAVIS 752. One pamphlet. J. WILLIAM DAWSON 753. Acadian Geology, 3d edition, London, 1878. 754. Twenty-five pamphlets. A. DEL CASTILLO 755. Geological Maps of Mexico, 12 sheets. W. B. DWIGHT 756. Eight pamphlets. H, W. FAIRBANKS 57. Five pamphlets. H. L. FAIRCHILD 758. Seven pamphlets and books. P. MAX FOSHAY 759. Three pamphlets. PERSIFOR FRAZER 760. Three pamphlets. 770. 776. Mh Ce 778. 779. 780. ACCESSIONS TO LIBRARY. A. C. GILL . One pamphlet. U. S. GRANT . Ten pamphlets. W. F. E. GURLEY . Three pamphlets. H. G. HANKS . Twelve pamphlets. ROBERT HAY . One volume of bound pamphlets. C. H. HITCHCOCK . Fifty-two pamphlets. MARY E. HOLMES . One pamphlet. E. O. HOVEY . One pamphlet. H. C. HOVEY . Celebrated American Caverns, 1882, pp. 1-228. . Four pamphlets, J. F. JAMES . Thirty-two pamphlets. E. JUSSEN . Two pamphlets. R. D. LACOE . One pamphlet. FRANK LEVERETT . Five pamphlets. JOSUA LINDAHL One pamphlet. W J McGEE Seven pamphlets. P, McKELLAR Threé pamphlets. G. P. MERRILL Twenty-five pamphlets. R. A. F,. PENROSE One volume. C. A. PROSSER Three pamphlets. O13 514 781. 782. 789-790. Tone 795. 796. PROCEEDINGS OF BALTIMORE MEETING. H. M. SEELY One volume. E. A. SMITH Four pamphlets. Ue Jee SOM at . One pamphlet. R. S. TARR . Fourteen pamphlets. J. KE. TODD . Two pamphlets. H. W. TURNER . Two pamphlets. WARREN UPHAM . Four pamphlets. C. WACHSMUTH . Fourteen pamphlets. C. A. WHITE Two volumes. J. F. WHITEAVES Eight pamphlets. G. H. WILLIAMS . Geological Map of Baltimore, with text. N. H. WINCHELL . Twenty-nine pamphlets. ARTHUR WINSLOW . Six pamphlets. G. F. WRIGHT Two pamphlets. L. G. YATES Six pamphlets. (£) From MiscELLANEOUS SOURCES MICHIGAN MINING SCHOOL, HOUGHTON . Three pamphlets. MINING BULLETIN, STATE COLLEGE, PENNSYLVANIA . Vol. i, Nos. 1-5, 1894, pp. 1-100. NORTHWEST MINING REVIEW, SPOKANE . Vol. 1, Nos, 238, 24, 1893: Vol. ii, Nos. 1-3, 12, 17-20, 189394. Vol. iii, Nos. 1, 4, 7-11, 189495. 800. 801. 802. 803. 804. 805. 806. 807. 808. 809. 810. S11. 812. 813. 814. 815. 816. \ ACCESSIONS TO LIBRARY. 515 PARIS EXPOSITION, 1878, PARIS Catalogue of Minerals, Ores, etc., in the Pacific Coast Exhibit, pp. 1-99. UNITED STATES EXPLORATIONS, WASHINGTON Report of Exploration of a Route for the Pacific Railroad from Red River to the Rio Grande, by Captain John Pope (H. Doc. 129), 1854, pp. 1-324. Exploration and Survey of the Great Salt Lake of Utah, by Captain Howard Stansbury, 1853, pp. 1-495. Report of Secretary of War on Pacific Railroad Explorations (H. Doc. 129), 1855, pp. 1-43. Report to Illustrate Map of the Hydrographical Basin of the Mississippi River, by I. N. Nicollet (Sen. Doc. 237), 1848, pp. 1-170. JOSE G. AGUILERA Y EZEQUIEL ORDONEZ, TACUBAYA Datos para la Geologia de Mexico, 1893, pp. 1-87. FREDERICK H. CHAPIN Mountaineering in Colorado, 2d ed., 1890, pp. 1-168. JOHN CRAWFORD, LEON, NICARAGUA Papers on the Geology of Central America. EUG. DUBOIS, TULUNG-AGUNG, JAVA Pithecanthorpus erectus, eine Menschenachenliche Uebergangsform aus Java, 2 plates, 4to, pp. 1-39, Batavia, 1894. E J. DUNN, LONDON Geological Sketch Map of South Africa, 1887. WILLIAM FRASER HUME, LONDON Chemical and Micro-mineralogical Researches on the Upper Cretaceous Zones of the South of England, pp. 1-108, 1893. F. W. HUTTON, CANTERBURY, NEW ZEALAND Thirteen pamphlets on the geology and paleontology of New Zealand. J. FELIX AND H. LENK, LEIPSIC Beitrage zur Geologie und Palaontologie der Republic Mexico, ii Theil, 1 Heft, pp. 1-54 + lv, 5 plates, 4to, 1893. DANIEL W. MEAD, , ROCKFORD, ILLINOIS Geological Map and Table of Economic Resources of Illinois. Notes on the Hydro-Geology of Illinois in relation to its Water Supplies, pp. 1-24. ROMOLI MELI, ROME Eight papers on the geology and paleontology of Italy. M. A. MICHEL-LEVY, PARIS Notes sur la Chaine des Puys le Mont-Dore et les Eruptions de la Limogue, pp. 688-952 (Ext. Bull. Geol. Soc. France), 7 plates, 1891. 516 817-818 819-821 823. 824, . Geology of Knox County, Ohio, pp. 1-28, 1878. PROCEEDINGS OF BALTIMORE MEETING. MICHEL MOURLON, BRUSSELS Geologie de la Belgique, Tomes i, 1880; ii, 1881. Four papers on the geology of Belgium. CARL OCHSENIUS, MARBURG . Bedeutung des orographischen Elementes ‘‘ Barre” in Hinsicht, etc., pp. 189-253, 1893. ALEXIS PAVLOW, MOSCOW Four papers on the geology of Russia. FERNAND PRIEM, : PARIS ‘* La Terre,”’ pp. 1-192. A. PENCK, EDWARD BRUCKNER, LEON DU PASQUIER, NEUCHATEL . Le Systeme Glaciare des Alpes, pp. 1-86, 1894. EDMUND C. QUEREAU, AURORA, ILL. . Die Klippenregion von Iberg in Osten des Vierwaldstatter-Sees, pp. 1-54, Ato, 1893. | M. C. READ, HUDSON, OHIO HANS REUSCH, | CHRISTIANA . Geologiska iagttajelser fra Trondjems stift (with summary in English), 1891, pp. 1-60. XAVIER STAINER, BRUSSELS . Eighteen papers (in French), personal publications. FRAU GERHARD VOM RATH . Gerhard vom Rath, eine Lebensskizze. By H. Laspeyres, Bonn, 1888, pp. 1-58. . Sachs-und Orts-Verzeichnis zu den Mineralogischen und Geologischen Arbeiten von Gerhard vom Rath: W. Bruhns und K. Busz. Leipzig, 1893, pp. 1-197. GUSTAV RITTER v. WEX 32. Periodische Meeresauschwellungen au den Polenund am Aequator, Wien, 1891, pp. vii + 1-59, 4 tafeln. INDEX TO VOLUME 6. Page ACCESSIONS to the library........cce:scseeee epeoaces 501 A CONNECTION between the cheniical and optical properties of amphiboles; A. C. PAIL no eeece aosis's caesccecs sper devtasecsdusssscessuseseas 3 ADAMS, F. D.. cited on Canada limestones.. 255 Se SEACINS OF PAPCH DYq-.cac-coe esecenescrceencee==s 471 —, Reference to discussion by.............. ... 8, 469 —, Remarks on Kemp’s paper by................ 3 MELLON PAPEL, DY 22 -ovnennesowesedennetersrindeeacnse 468 ADEHEMAR, Jes Gieed on glacial accumula- THONG Meret ees chonarcurees ves tiasuac eres weneevdoat eosiste 145 ADIRONDACK region, Limestones of the VO LUMUVVESUCTI, 5.002 ossecesndcerecessteesacceces esses 263 —-—, Crystalline limestones, ophicalcites and associated schists of the................ 2AT AFRICA, Evidences as tochanges of levelin. 162 AGASSIZ, A., Reference to charts of....:........ 104 —, Term “ Blake plateau ’’ first used by...... 1a9 AGAssiz, L., cited on deep-sea inverte- OTA CSueinc ci sceresiinaccos tacoseoarsttwaeeieescooattel asesae 133 — — — forms of life in Central American NVEMUGIS sooccccecacsscssceccoacsdvancnsuscacceaedccssensves 134 ALABAMA, Geologic Section im..............0.sce00 106 —_— , Zapata formation the equivalent of Co- Maia Talo fe-ess-ceccee see PE ee scseldaechcickeainleesezs 129 ALASKAN coast as an evidence of subsi- NNEC Gites secsccsccencncncavstecssccassasossesneascsescessss 160 ALGONKIAN age of Pikes Peak granite........ 471 AOL ECRAG Hr cccece.scese. 5.5. ceeecsssenocesnes Been eee eeae 376 ALGONQUIN, Glacial lake.. seeeaiee 25 AMEGHINO’S, FLORENTINO, ‘latest paper on Patagonian paleontology, Note on1......... 28 AMENDMENTS to the By-Law................ 15, 432 — — — Constitution 20. el. eee eeees 15, 431 AMERICAN ASSOCIATION FOR ADVANCF- MENT OF SCIENCE cited on Mount Rai- TIC TPES CLV Csteac san ccnsoNeoee aaseccaie fee teee ball oeeeee 14 —, Resolution thanking local committee of. 27 AMERICAN GEOLOGIST, Publication of UE: ham’s paperin .... 20 —, Reference to Fairbanks’ publication i in. 72 — — — map published by H. W. Fairbanks DMs coer. co sc occtc coves ddecavetencoes seniesetcassace 83 Amt, H. M.; Memorial of Amos Bowman.. . 441 _, Reading OM MAPS Dyn -e-ceseeeseccaseanseentsnce 489 —, Reference to discussions by.............. 443, 488 Amos BOWMAN, Metnorial off:....-.........-:...: 441 AMPHIBOLES, Connection between the chemical and optical properties of......... 3 ANALYSES: artesian well-water from Iowa.. 194 AMIE C tiene cons teee. eee secon oa sces 420 BAT REVUKACC vac scscecccssencestetncuess: 420 COG UIMA case cpaer taccecvar cus tcaitentonsescs 192 coquina gravels from Florida... 193 CORAM SAM Mies. cevetssceesececceeeecssexcsss 192 Geadicorall ayes eseeecey in neces 193 gabbros from the Adiron dacks. 274 glauconite... =appnccos0aTe . 185 Hawiian chalk.. Gee sccoueseeessieae cusses 192 Hawiian CORAUTOCIS feecarsaenasan: 193 lakes and rivers of Minnesota.. 194 MTTMESLO Meee eer cece ca ecccncctass 258 AIVSMU TTS Rear eee eres coe le sccseeess 470 mineral spring water of Wis- consin and Minnesota........... 194 Nelson! River water...) sss. 304 ING@WANOLKaSTaMite ss. vcs. scesses 4 peigoutc.: Necbtoesasccessccccasecossevcssee 478 PVEOXEME! HOSSUIS Oa Wes: 1 ee ae pea ee 377, 378 — IA aAMCOMILes Of NEW J ClSCYce-c:cc.s-0..<0scvecees 185 — northwestern plains, Baseleveling of the. 17 ==> OIF INGA IES aan eee eee F535 4188 te PACIIC COASE, FANLES. sap ciccteccececssesd ceosees 95 — — the Highwood mountain.......... ..... 390-353 — — western Texas and Coahuila. Mexico ; ES eM lelscacccsscsseseses AOS DOOOLEO RSE DEeCOSCER 375 —— period of the Wiest T11dies:.).cces.-ocseoe- 2-2. 120 — strata, Glacial deformation of................6. — 2 a yong island, Deformation of........... — — — Marthas Vineyard, Deformation of. 5, 6, 2 ——— Staten island, Deformation of......... 5 — —— the Crazy MI OMMMEANTIS ee le oes Be) CRYSTALLINE limestones and associated tocks of the northwestern Adirondack TESTO, Cele SUL tlie Ih, plete stares aseeesces 263 CRYSTALLINE limestones, ophicalcites and associated schists of the eastern Adiron- GAGES Fi lin EO Iem psy Uke kes anesdaleendiess » 241 CRYSTALLIZED slags from copper smelting JeNa MOG OBITS ede bane tot Ec tos poche IAB EE BEET CRE nent . 469 CROSS, WHITMAN, cited on Pikes Peak rock. 471 eee SPST INT SS sesso vad vtacclkc oa leccvedenc 476 — made secretary of Petrographic section.. 469 —, Reference to discussion by... 468, 469, 473, 476 LX XIII—Butt. Grou. Soc. Am., Vou. 6, 1894, 519 Page CuBA, Application of geologic discoveries 50 WA Reena HOSE SEL, REM ie mE E Rn A a ee nec 10 —, Cross-section in valley of Trinidad moun- TET) feet Ora eee A aoc Se) Sn a 107 _, Description Ole fossils sesesereseee eee eee 124 — during Matanzas GE pressiotlsescesescsseueeaes 125 _—, Erosion TO OR Ste eg Sia Rn Aer eae ners RAP oo 128 —, Existing mammals Of............:ecceeeeees 138, 139 Nin the Pleistocene. 0. /..csll sscdicsedoged sexe 133 — Miocene and Kocene, Thickness of......... 121 = DOU SiOlieanesecsstaee ten scctcreskenea coseaasiwanseeeeecae 122 == OLIMALIONS) Off cese ce fienes soon cccsenoneneersess 132 == 1IFTV CSCO ME Oli esse scc ce oetesccsecdecerescneqeecensses 124 —, Modern orogenic movements in............ 131 _, Radiolarian GePOSIES UME eieeveeeesees eS =- 5 WDZ —, Reference to elevation of mountains of.. 106 Sao eT OUMMCAITS Ole erie site esac tee ence 109 —, Relation of, to adjacent Seas.............cssee 110 CULM of Europe, Reference to the........ 313, 320 CUMMINS, W. F., Reference to collections DV Aa ceeteeesiasaec slp oeeccaecrcasvevenscsaweceoese Bean caer ae 281 — — — Texas section made by ................008+ 386 CUSHING, H. P., cited on distribution of an- OREM OSICES ey seek eae bce seep cds cess eeu eeeconewaeeres 242 —; Faults of Chazy township, Clinton OHO INE Sod, aaah oebepaboauonedscuadunoonecs 385 Stl GOL) PApPed Di saresessse-cerssseensecesssesreseseee 443 DALL, W. H., cited on Alachua clays of lO rd Gay eae ee csarees cong eee iey, actead neers 136 — — — Miocene and Pliocene of Florida..122, 123 — — — Pleistocene fossils .............cc00 coe eecees 138 — — — Pliocene and Miocene deposits........ 126 — — — thickness of Florida Miocene.......... 121 —-— —— — MEAS WUIOEEIAS scacoasnoccosecoooccone EAI — determines fossils from Cuba ........... 124 —, Paleontologic assistance rendered J. W. GYOENESIP 19817) acon cacecodcosecodeobodeondeancocecoonoaad 136 Dakota formation, Reference to the.......... 18 IDAUSVA, fs 1a, Analysis of coral sand by ease 192 i Hawaiian Chalk Wyte ccscccessnees nen esaeees 192 — cited on coral from Howlands island...... 195 — — — glacial ice-sheet............ coc, SyU7/ DARTON, N. H., and J. F. KEMP; “A new in- trusive rock neat SATEICUIGS caacanaocadsodcsodse 477 — cited on Potomac and Lafayette forma- CLOTS SS aerrce rec ee a soso eect ea ea Coe eer bee anatase 329 — — — Severn formations..............cseceeeeeeees 480 —, List of photographs Dy ... .......s......eceeees 450 _— ” Reference LOVAISCUSSTOM)Dyzeeceneseeeeeseea ese: 482 — ae MOIS OIE POBYOESES LOY, cannoosesecionogsa access 16, 17, 482 DARWIN, CHARLES, Naas of dead coral LDN sa ccocccoodocnansascqondeocddondaddens nooucn neo snodansAcedG 193 —, Influences of, on methods of classifica- CLOM esos Ws eos sea tear eect wot eeets es betenwenecs 65 Davis, W. M., cited on baseleveling in Pennsylvania ang NewaiiGGs Gyarseseneee 19 — — — erosion of Crazy mountains ........... 19 — — — Highwood mountain............... 390, 392 = i IMMOMAGCMOCK SG) Aree secseerssnicserscadseresse sLAQ — on Photograph Committee fae nascueeeeeseeess 445 Dawson, G. M., Acknowledgment to......... 442 — , Reference (ca ae eats era areas ee 108 fy 441 DAWSON, Tho Vion IRGTETHEINEE UO ssanoccesen nesaseosna06 458 IDWS, 10)o AP on excused from Mount Rainier Girma b hee tke esa sn ek NE LA 14 DEFORMATION, Extension of uniformitari- AMUSE ON J. Seid sius cacealevats tecessetecsveatceccduire ces —of the Atlantic coastal plain and Antil- Lami RESTO tls oe sre eaten Ac cces nonsense ieee teeta 1277; DEFORMING agencies affecting shorelines.. 146— 149 DELAWARE bay, Changes in shores of........ 155 —, Cretaceous deposits ON A479 DELTAS of glacial lakes of western New Ot ksiwcsse Mack aseeseseciissivcecenescecectstevesctererne 5 DENMARK, Reference to glacial phenomena Didar rde Eas peste Ba gcabanseeoavecedekiaseueoeletoniawance DEPARTURE of the ice-sheet from the Lau- rentian lakes ; Warren Upham............. 2 520 Page DILLER, J. S., approves accounts of Treas- UDR e ye i conoscnnccoceos cangog Wak: se tiscrlonnonodooa 7 SonascoG 445 — cited on age of Sierra Nevada rocke......... gi — — — continuity of Cretaceous sediments.. 97 — — — origin of California serpentine..... 98, 99 — — — Pacific coast TANGES.......csesecceeeee oe 74, 76 — — — thickness of Cretaceous and Eocene SERA cesercecadec wasters ede weacsnaasessacaeants — — — unconformity of rocks of Klamath WIOMUMEATIG Pacnscstacsuaccacetessosneeececc seen eae ete 89 — obtained Carboniferous fossils from Cali- IVC ONE bos dndeccon pececacericoc => terracesjor BatrbadOSiit-.cressreccseteoseeee 126 HAWAIIAN coral, Analyses Of........ 00... ..eseeeee 193 HAWES, G. W., cited on twinning of feld- spar... atte 257 HAwn, B. “cited on “geology ‘of Kansas.. . 30, 50 HAWOoRT H, ERASMUS, Reference to Cotton- wood River section TAGS Dyyeeseesensere neers 37 ars | ati of paper by wigiewmae eereretesrrrrerererrsers teeta 16 522 Page Hay, ROBERT, cited on limestones of Kan- CES Ganconacsooct: not csd00so jendgnecccsssosce sna) Eoagn 48-50 HAYDEN, F.V. cited on geology and paleon- tology of Kansas... 30, 32, 34-38, 40, 41, 50, 51 HAYEsS,.C. W., Reference to discussion by... 443 HEILPRIN, Ay, cited on! clacirersices. tesecoce nese 202 — — — Matanzas formation of Yucatan... 124, 125 ——— — Vucatai marls.............ccsee-csceresesves 129 HERSCHEL, J., cited on displacement hy- 10].0) 8) ESTES Ese ae concer aceead Mea secichonceacludorge

; S. F. Emmons and Bailey Wilhist2ce & 9. eae eee oeec nner 13 === SECKEUAI: Jt osss seceiotsces| 242, 244 — — — discussion DY.........ce.cceee ceeees 443, 460, 468 ah ADE OPE VOEH OYE ONY AoooosccuasennhdBadsea esces 443, 468 WALLACE, AV Re nGibedhom AmbilleSr cesses 103 — — — relations of land areas........cccsseeesees 161 WARD, L. F.. cited on Potomac flora........... 480 W ARR Ne GilacralWMlalcetaeerectscesecme sciatic cess 25 WASHINGTON, H. S., cited on twinning...... 409 WASHINGTON, Mount Rainier Reserve in CUI US) (0) Pocmearnas. se sebke oc Goneeu sedusc.cduedasnoscanco¢soud. 14 Watts, W. L., cited on jasper and sand- stone of California and Oregon....... usa 82 WEED, W. H., and L. V. PirRSsoN; High- wood mountains of Montana.seeseeesce. 389 — Ie ISt Of PMOLOStA PMS) DWArr-nc. scecee-t ech eetecees 449 SS) IBIS OPE ON OVENE ION, cacocecbonscne HodooansusudoocoonNeG 444 WEIDMAN, SAMUEL, Title of paper by......... 488 WESTGATE, Ds, Election of 21.28 eos: 2 —, Reference to election Of.......0.cssseceeeees 425 | West INDIAN REGION, Faunal life of the... 135 WEST INDIES during Matanzas depres - SOM ene onccon cee cvee ech ocaanek eeciesslecssanannecsnecars 125 —-- the Cretaceous, Hocene and Miocene... 121 ==: JH EOSIOM) Il seaesccceoehsceccescvaccneacc se coseiete es 128 —, Existing mammals Of .............esscccssesereree 139 — formerly united to Florida..................206- 135 — Fossil siromithierr cee pov se see sccaerestestors 138 —, Miocene subsidence in the..................06. 122 —, Modern orogenic movements in............. 131 —, Reference to continental relations of the 103 SP OUDSIGESM CekOfseteess teases tee iersewescenenensrace ooo AS) WESTERN Superior glacial laleen ange woes 24 WEST VIRGINIA, Fossil plants from. 313, 318 —, Pottsville series along New river............ 305 WHITE, C. A., cited on Iowa fossils.............- 169 — cited on New Jersey Cretaceous.............. 188 — — — fossils of San Miguel beds......... 383, 384 — makes rock collection at Square butte..... 400 WHITE, IDA), ANUS Ge RYOXSP LON coocacodsacsas 468 — - The Pottsville series along Nee river, ” “West Wattoairnlay eaten saenaiien ee east cia cos sccees 305 WuiTrrE, I.C., cited on Nuttal section.... 309, 310 —-—— plants of the New River coals......... 312 — — — Pottsville Series..............csesecceeees 306, 314 ——— Productus cora ... Gaagdne Zig — — — Virginia- Kentucky coal field .....s.c.s. 319 —, Determination of fossils Dy... 0--..cc-sccsccos 34 —'elected Treasurer. ...cecscsscessescstcecseerere Boobede 431 — makes report as TreaSure.........:...cssecscores 429 == (Oya! Intopeeneny opanbaolrht YES, canon anccoddboonsonoosoncsoree 427 —, Reference to discussion TDA ea ceesssaee AA 40S — — — the “Alleghany series ”’ Ghent eee Bag WHITE, JAMES, Acknowledgment to.. ........ 442 WHITE, T.G., cited on Adirondack apatite. 260 — — — faults of Clinton county, New York.. 288 =a MElChenCe to GiISCUSslOmbyAnses-cciesenesnte 476 Wuit FIELD, R. P., Record of Gieuenon byes ‘Title of paper by BSod ToS ee HEROD COME Dtanosee, dio: 488 WHITING, ANSON, Reference to delta on property OR Aieedince te ade iaiehdens, aes eenaae eee 359 WHITNEY, J. D., cited on auriferous slates.. 224, 225, 227 — --— Cretaceous and Tertiary ABE of Pa- cific Coast ranges.. deecceetasmseeds 76 —~ -- — Inoceramus of California.......c., 93 —— asp eLnsioty Calitoimtlial gancsenve voscecces scree 84 Shae = PACING (SOASL MAMSES i cores eassisyecease 74 ott LIAL UZ VELILG ss aeicepeseucraccsseescuecencnaseces 235 ROLE TEN COMO sayi ears tes maier te aucaddeee caste WILL! IAMS, ABIGAIL (DOOLITTLE) ; mother OLIG WEL BVWallitais mec. san seers ae eve atone 433 uncle of G. H. Williams.. 433 , Announcement of WILLIAMS, F. W.; WILLIAMS, GEORGE H. C7 1) Oke SER NOR RCE I —, Bibliogr raphy of.. : Bean nt anee cee ~ A877. —'cited on Maine volcanics ...... 474 475 O39, BULL. GEOL. SOC. AM. Page WILLIAMS, GEORGE H., cited on origin of Adirondack limestoneS...ceesesetesss sees: 244 —- -- — polysynthetic tie Pe Areas: cr 280 — — — SYTacuse AiKE ....0. 00. socsscssecnsessecsn sae TO --, Memorial of.......:....... Vauewbedeespnasnoessomreene 432 424, 425 -—’-- — reconnoissance in the Adirondacks aes Besavecenad hey) WILLIAMS, ‘i. a appointed on “committee. 2 — cited on Permian beds of Kansas........ 50, 51 —, Reading: of paper Dwen.p-cc,y-seeseeess ener 17 —, Reference to discussion by.....13, 444, oa - _, ’ Remarks uy-on Hovey’s paper byes peered —- reports on Royal Society catalogue......... pe Title of paper: DYsaerrensearcdyece eens 13, 468 WILLIAMS, J. J., cited on isthmus of Te- huantepec solvwtenecnante(onces- bededegnrseee tee aeaamee 121 —_---- Matanzas limestones of Tehuan- tepec.. eer L25 WILLIAMS, Ge Ss. “father of G. ue ‘Williams.. 433 WILLIAMS, S. W.; uncle of G. H. Williams.. 433 WILLIS, BAILEY, offers resolution of thanks. 489 --, Reference to discussion Dye: ‘The Mechanics of Appalachian Structure” iyi. c-sccsce