dncetdiamcetirsba EPP EES TASTE TSS TS: MEMOIR No. | 38 Se CLASES, ATEN OS Se ners NET SAS TR wR NORTH AMERICAN CORDILLERA ~FORTY-NINTH PARALLEL BY REGINALD ALDWORTH DALY SRoR REE aaoee PART | GEOLOGICAL SURVEY DEPARTMENT OF MINES OTTAWA 4912 BR ’ ' FROM THE LIBRARY O GEORGE P MBRTMILL Us: USEU X NATION AT, M ene itself.” ilton Mi ills reason k 1 BX LIBRIS good book © ?) D 12) eS SS ad WY) vo xe} 0 = iV) LS ae) =) Q x x 0) iG =) pw) © AV) = Oo ) aq © iS O m7 cw Vv cL © wo = S © iS © Mw = fe) i= S ee 5 14 Wieser J4651 U. S. GOVERNMENT FRINTING OFFICE: 1930 GEORGE P. MERRIL, FEB 13 1915 DEP'T OF GEOLOGY Frontispiece. Puate 1. Terminal Boundary Monument set by the first International Commission 3 te at the Pacific Shore. __, 25a—vol. 11. CANADA DEPARTMENT OF MINES GEOLOGICAL SURVEY Hon. Rosert Rogers, Minister; A. P. Low, Deputy MInNIsTER; R. W. Brock, Director. MEMOIR No. 38 GEOLOGY OF THE NORTH AMERICAN CORDILLERA AT THE FORTY-NINTH PARALLEL BY Reginald Aldworth Daly. IN THREE PARTS joa Rd Od bs One AUNN EX GOVERNMENT PRINTING BUREAU 1912 $364—aA No. 1203 INTRODUCTORY. Through the courtesy of W. F. King, O.M.G., LL.D., B.A., D.T.S., Chief Astronomer, Department of the Interior, the Geological Survey is enabled to publish this Memoir. The field work was done under, and at the expense of, the International Boundary Commission, and appears as an appendix to the report of Mr. King, the Canadian Commissioner. As the report constitutes a most important contribution to the geology of western Canada, and as in its ‘Blue Book’ form it would not be available for many libraries and individuals that would have use for it, Dr. King kindly consented to allow the Geological Survey to print it as a Geological Survey Memoir, and thus secure for it adequate distribution in geological quarters. The Geological Survey is pleased to be able to add to its list of Memoirs this work that deals with the geology of such a long and important section through the Western Cordillera. It must be referred to constantly in future work dealing with the geology of British Columbia, and were it not available in the publications of the Survey great loss and inconvenience would result. (Signed) R. W. Brock, Director. GEOLOGICAL SURVEY, Ottawa, October 21, 1912. 2 GEORGE V, “SESSIONAL PAPER No. 25a A. 1912 APPENDIX 6. REPORT OF THE CHIEF ASTRONOMER, 1910 GEOLOGY OF THE NORTH AMERICAN CORDILLERA AT THE FORTY-NINTH PARALLEL, BY REGINALD ALDWORTH DALY. IN THREE PARTS PART I 8364—B 2 GEORGE V. SESSIONAL PAPER No. 25a : A. 1912 LETTER OF TRANSMITTAL. MASSACHUSETTS INSTITUTE OF TECHNOLOGY, Boston, Mass., April 30, 1910. W. F. Kine, Esq., C.M.G., B.A., LL.D., Commissioner for Canada, International Boundary Surveys, Ottawa. Sir,—I have the honour to submit the following report on the Geology of the mountains crossed by the international boundary at the Forty-ninth Parallel. The report is based on field-work carried on during the seasons of. 1901 to 1906, inclusive. To yourself, under whose direction the whole work has been done and from whom I have received help in many ways, I-beg to tender my sincere thanks. I have the honour to be, sir, Your obedient servant, REGINALD A. DALY. 8364—Bt 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 TABLE OF CONTENTS. PART I. CHAPTER I. Page. Introduction. . Area covered. . Conditions of “eH § in ‘ihe GGL. Acknowledgments.. ...... Collections. . Previous Ee hcat ons! be ihe ater on aie Bort: minelt ‘Parallel geology Earlier work on the geology of the Forty-ninth Parallel. . Continuation of the Forty-ninth Parallel section. General sketch of the subject matier.. .. .. . MO WOH HH CHAPTER II. SyNOpsiseotatherreEponurey cise aie vie ssa ene eee cs eee anes colts Ceara gd Vata taroanette 4) CHAPTER III. Nomenclature of the mountain ranges crossed by the Forty-ninth Parallel. . Bs yy ane CMR ies Real A a LA Ud MOUs og De Me EM i Introduction cal ectlinese BORK An AO ney rants Pah Riso teh IAREP nie a eiR ee gis oat LY Different nomenclatures in use.. ....... detonate nega Diverse naming of the Heseien. mountain ean as a aolee. ROME Minny wae ko} Diverse naming of ranges crossed by the Forty-ninth Parallel... .. .. 22 Adopted principle of nomenclature for the Boundary mountains.. .. 23 Trenches and greater valleys.. .. .. Aes veeey ONAIIC Migr Maida AS cae edhe SOUR MOE Subdivision of Rocky Mountain system. . ONE ir be kangen, Alas pea) Sate Reo gna enon aE Purcell mountain system and its subdivision.. .. .............. 80 Selkirk mountain system and its subdivision... .. ............ 84 Columbia mountain system and its subdivision.. .. ............ 37 Belt of Interior Plateaus... .. . PAE cv des Sehr On tee aele eA () Cascade mountain system and ate subdivivion: RA bed Maa Se Mirae: poked 8 Gane 322K 0) Summary.... EA le Rcaniaar Td tae en SRD LI, DEH Nae ate N eR CANS oe ake aes Z 8) ‘Lawns wetter. Shi idlegs abe Vase Sk Warn ee Mee roelsiue hn oediameee Lian DRA er eS Peal AS CHAPTER IV. Stratigraphy and structure of the Clarke range.. .. .........-..... 47 Rocky Mountain Sue PYISINA tala eee eta es elcusrod bens Neeisitecena came ET Lewis series.. .... SiGe ve RGR TEI AE UE asec mais eer cathe cher perenne O. viii DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Pace. WViaitorcom’ LOTMIALIONS 2 cis 52 ccc Sass siglpares Gn ese ce Dasa oe PAT EVM AROPIMAVION ss fs oa. voces ce ace clude cle se ceceiealon wee et ean Generali;description: 6654. ccGe elec ate a ens es ee MEOW . Areanwestior Oalmoneriviens cet seo. en eee eee Correlation. . Summary on the stiches Bf ihe Nelean ange. CHAPTER XII. Intrusive rocks of the Selkirk mountain system. . Metamorphosed basic intrusives in the Priest River eerie Abnormal granite intrusive into the Kitchener quartzite. . Rykert granite batholith. . Baia Bayonne batholith and its eafellices Petrography of the batholith. . Contact metamorphism... .... .. Saal Satellitic stocks on the divide... ......... Petrography. . ae Contact metamorphism. . Quartz-diorite apophyses. . Relation of the stocks to the Baro: onne Sha tholieh ay Gem OPE aN: Pera Wak Lost Creek granite body. . Bunker Hill stock.. .. .. . Salmon River monzonite.. .. .. Lamprophyric dikes and sills. . Porphyritic mica minette.. .. Augite minette. Hornblende- eeaite nineties Olivine-augite minette. a , Comparison of the siinaitics rites the Portia -average. Xk PAGE. 257 257 258 259 260 261 264 264 265 267 268 268. 270 Q71 xii DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Pace. WXersantites. sche see aha eat cg EUR > o-4) Go) Ge Gamiptonite ish sai ua sas eye hws eee as ales Vals Saat era hee eS em Odimite: =~. ee Rie cle te ae RUSE a eee Aplitie and acid Soak eaall dikes?! ait Prarie ett te ile: Dike phases of the Rossland and Beaver Moumeain Golenmesne igettares, soul Relative ages of the eruptive bodies... . 2 - .1.j2 0c). ba ciee neoe CHAPTER XIII. Formations of the Rossland mountain group.. .. w. 9.1.5.3 .. «. as ones OO Paleozoic formations.... . Loca ey hly eters ate et LO) Carboniferous beds in Tittle Sheep antes Palte io ip Aint Maar CaO Carboniferous limestone in the Rossland mining camp.. ........ 321 Sutherland ‘schistose complex: 2%... 4). (Sc) acs chen canoe get epee eee Summary... .. og i lavel Sahay PARES ee Oy eel Mesozoic dente at Taide: ehees pad. 1a) sa us le, Yor Sith alan 0a eee oe nee een Rossland voleanic Sroup: << « © ss «sw wee eivalece Sieve) terete entee ye meee Tene General description.. .. . 3 (Rpels Beets CR Soe eee Petrography of the lavas ee Pp ocleeticee Lecce a ne ieee aeeaeeietee Roe Augite latite. oo. wie ase: vowel 6 dik sieve chlet ere! hee ok ee ev be Seale ee ee Aucite-biotite latitesi oe isos ob aa Yo. vv bok Mae ee eS 2G Augite-olivine datite. <6 5.0. sods ts, cc8) ope ee eeu ree eS Hornblende-augite: latite:: osu; Seas) us 0. a te eee be Hornblende-biotite; latite. a). 2... 5. sccm ere Biotite latite... cs 2. scsu Sais lends bce yore el bee ee el Femic augite latite.. .. Loom Comparison with Sierra Nevada latte aad arith average mon- ZOMALG 2 is: oso 's 0 6 we te pal sogh oo, yh pausa te ee BOOM RET Bae eee ES il Augite aadlesiton Pre a AMET aS 5 Grind om Toad OOS Basaltsec vic: 5 Se bids lend, BB lae telah ened okie te RCRA ame Flow of liparitic anobienS, rea pee er re MES Tee S eae: Gate) Tufts ‘and -agglomerates./.:.ci Ys See lone eee «eer ulna ae eee Dunites cutting the Rossland volcanos eas 8 Sef oe pg a ee ee Dunite on McRae creek. . SSE ee! cabelas set obi esl dle Mind eee ee oe Porphyritic harzburgite (picrite?). . : sceieagh Gist ats ae Gabbros and peridotites near Gheisine lakee seller's 2, eRe ies See oun ema Rossland monzonite.. ..... ee ere ee ea) 8a) Basic monzonite and Waciblenditen on Beer eeaalel bile, Oig he sega ee Shonkinitic type at Bitter creel. .°2. 00-220... 34s ee ee Granite stock éast of Cascade.’s 2.0.8 e. ois. Wicks ee ee ee Lrail-batholith.)s .% <...3 G6 RS ee ee Definitions 5.5 ...05 2) See en Petrographiys.:. = ls Gs. Soa ee Sn eee Differentiation in places. i.e oo Secs.) oats eee eee hatter-belts ce secs. 4l6 Su ee ie eee Oe eee REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a Conglomerate formations. . Conglomerate at Lake SiON. Conglomerate at Sophie aN Conglomerate area at Monument 172. . Conglomerate area at Monument 169.... . BR ais any Th ee Correlationy and O1ieinives cscs ee Cesena eeuir heals tn eee: Beaver Mountain group.... . Generalideseriptions: cy0.0 208 < OGM eTIES igre eaten acliaya ge Olav ailis saree aece Wiol ean ese ueutets arse ietts ares eal ong canted noite Sheppard granite.. . Bae Porphyritic olivine syenite.. .. Coryellgsyvenitosbatholithyers: cca sre chek cs orev sue os ene peokeaie acauseoele, oo poles AD) OTFMMEATTE PPM ASOetesnetpac\enics spears ecu a Gis. sees ocala hike, Canes ales apnoea lam eran Basicnpmaser at, COMtACt a0: s6 sisi ie sac, ci Apophyses. . Contact acamornhisnt: tar aieake Syenite and oe eS parolee to fais Coryell batholithe Saecr Chonolith. . SECU epiteu ia TR Ree an MeN aa pa Wikece reese tess ks aise: eine Missourite dike. . Various other dikes. . Summary of structural Polarionel in ne Rossland mountains ee rsa ime prel stron Sraesss evel «su sevedvcecehescay'+s Observed facts..... BAL SR a SOS A RN NS eGR iBrobableprelationseicsnctct ce sroncr theron sass tiene La ieabane wis eure ceteas Correlation. 3.052. CHAPTER XIV. Formations in the mountains between Christina lake and Midway (Middle part of Columbia, mountain: system). . eo. tees eso. j Generaledeseniptronisrse- cerca) vay veters cen sans siete eis ies ccna Grand Forks schists... .... . Cascade gneissic ‘yeaah. General description. . : Nature and origin of bandine! 5 Smelter granite stock. . Attwood series .. .. LEM ake A pos Nea ey Et er RR CD ee Mgr RD Chlorite cine bosmlbilgeale Cane Rene en mr tiay GIs SRE Oe ee TR Phenix voleanie group.. DS Bie Nelo Rett acon eae DOR PCIMEIM Care es Mie ee ctr ye tah eran oeen ein: Lace aoe Renna rer ene wet aie Granodiorite.. ...... Correlation.. .. . xiii Pace. 350 350 350 B51 352 352 352 352 353 354 354 356 358 359 360 362 362 362 363 365 366 369 370 372 372 375 376 377 377 378 379 379 380 381 382 383 383 385 386 387 xiv DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 CHAPTER XV. Page. Formations of the five-mile belt between Midway and eye, lake Sisal mountains and Anarchist mountain-plateau).. oe 389 MAVETOGUCHIOM Sse ee este ele. eee eee Ge ans ee eat net en age eg ee aL Anarchist SOViCSs ss 66 es Re SA Ne en ee General description... .. . DoS ea wan ote rene te oN he ar ee Neciee of tho metsnmorshien eam MAN iy ambien en... Haye Rock:‘Creek plutonic bodies?3. 5) Bese es wes a ee ed IDiOritessiees ae. See lore Se Tics ccient omg Stee Death mer oe Ne gn a Granodiorite. ©. Pe ne i ee 393 Munites: fo 2. Ula rare he soy erm atts SEMAINE talib a SU Reotieliiver- formation, . A arin RN gaara Mh Se ae tcacrmdliiiiy Ae tawl ehnit Aer it oy’ BADE General ‘descriptions. 68 8 a er ee i a Oe Geological age...... Mae Say AE IRA AMC ire ssaton, «ITC Midway voleanic group Ga aoe Wa rveles cetece te, ONG AU ee oO General description.. . ene MEMRAM mau. B\iNe) Petrography of the subalalins lane See aiae salt wer age histo se cat Sane ae RS Rock Creek chonolith. . PAu coat MP eA Rh mee eMt Ss rus ea iri G, cKO Seevomimalrelations 0 a Oe ae 401 Dominant Tock (ty Pea’. ove ek Ses ece se eee) ole ase ceo, Wale: Selec el cone eaeeetmeR AOL General. deseription 3. 206 0 Sat we, oo eh aie oO Rhomb-feldspar. e.s6 ccs she ce eck ole fo ela ce eteoe deen io coo pen) Other constituents... .. ae ee eae Oe Chemical Aomisenition Ang Glassifidation ‘of ans ese ea, ecacte Bor eLOD: Contact: phase. of tthe chonolith.. 2.0 5.0.0.2. lee ec ek ee eS Other intrusions of Shombicorsuaee sented ane clatatcuepeel as Stee ta Mugen gate eee OG) Extrusive phase of the rhomb-porphyry.. .......... ++ .. ep «« +» 410 Analcitic rhomb-porphyry Soar ra were ected ie COLL North: of Rock reek: 5c) cic ens) ecccurpere ale” co. 0 Renin ene me Other oceurrences.. .. . Paneer cima rate Wer”) AGU Intrusive rocks cutting Kettle Rives erate Ciclo Sod atts 1 Pag BAoetoe eee tama IPOTPHYTIGES «2, <)e sess. kee ose lash Metedn artes, Sevapeigs-_« Sea Wedel ctetrencd bee gare meta Pulaskite porphyry.. .. SNPS Evens sais get ale eRe et gead: aa aun ATE Order of eruption of the Miasact lavaele Srearens 420: Structural relations of the Columbia mountain Sega ahs a Konencnne VAG es re Se es ee iwe eo ie aie he mbeMeete Ie rai. esa ale) Raond p ae ea O GOrrelatiaMs arecuis crycie. wis seco. volta lui Nee Beet eR copay sic: ca cine a el eee CHAPTER XVI. Formations of the Okanagan range and of Kruger Mountain plateau.. .. 425. General description of the batholithic area.. .. .. .. 1. 1. ee we ee ee 495 Roof-pendants.. .. . Ca mae EE A a are ge SEAS Unity of the ceminoniic! batholithe Bene oh GAN eet ott Sedimentary rocks and associated basic onlonmice wd i TR ee Bg REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a Tertiary (2) rocks at Osoyoos lake.. . Petrography,of the composite! batholith...../. 2: °.0.es% 0.6 56 es Richter: Mountam hornblenditesic . 4 <2 eanwaia ok ecules see eel Chopaka basic intrusives.. . Ashnola gabbro. . Basic Complex... .. .. Nodule-bearing Boriderite Ghia, Vesicular andesite dikes. . Osoyoos batholith. . ; Original granodioritic noe Dynamic and ee aoe metamorphisn| oe ANG eraniodionite 441 Remmel batholith. . Western phase.. Eastern phase. . . 6 Interpretations of the ‘hin anes. Kruger alkaline body. . General description. ; Nien Augite-biotite malignite.. .. .... Femic nephelite syenite.. .. Nephelite syenite. . DUMMATY seme ee: Meta OFDMISIN ego yen eor es aitonb..e Similkameen batholith.. .. .... General character. . Basic phase at contact.. .. Comparison with Reece Alealine, bode Dikes cutting the Similkameen batholith.. .. Cathedral batholith.. .. . Older phase. . Younger phase.. .. Relation to Giileamesn patholite: Dikes cutting the Cathedral Parolihen Park granite stock. . ere a Geological relations and mente A cang uneasy _ Résumé of the geological history. . Sequence of the eruptive rocks.. .. .. Method of intrusion.. .. . General summary.. ...... CHAPTER XVII. Hagrmations of the Elozomeen range: .° cus cis cee ee ie ioe oeeateth esinelen os General description.. .. Pasayten series.. .. MG TOAUCEION rags) fous oe ee hoes Stratigraphy... .. .. Fossils collected... .. KV xvi DEPARTMENT OF T'HE INTERIOR 2 GEORGE V, A. 1912 Pace. ee ee ees A89 Lightning Creek diorite.. .. .... ih SESE Oe 496 Other basic intrusives cutting the ncayient fomantione! MA PEL eee Om Castle Peak stock. . wlioaiel cv laoi@iwn lareh Week ele) seh Rane etna Tea eke AY ae Sr rene eiene idl amportanes.. ef ee ee 499 Dominant. Phase dishes ice cos wove wale ote: vast eas leds Ue ey es OSTA Re eer Basic: contact Phase ericie hve: diaisnsh det MAE Oe Paes a Ren ate gee Structuralrelations. 2% j..c. cc vb veo, Ve cso BEERS SET ROT M tis TENN tp a Intrusion: of syenite porphyry. cs.) a. odes s tes wie wide otaid ee ee OO Petrogra play ic. sis) 2s Bia aha No Bebe gar on eta le EG Peete HOME Ca Oe, had Sere meneame @orrelation:s S05. os Tee eerie, UR eo ce vale eee te eee ee OU) FTOZOMECTWSCLICS ws 4)s.5506.) 0 £2 ere soko eg Sie ie oo Ge co 0?) Gin ne ne TER em OU me SCE UOH NG is a ee 500 Correlation.. .... PERN Js hla: ent | OK Structural relations in hoe TANGLES. Voicbna sie HE eee aia ee chnehe es ates err cumimay Oe Correlation.. .. .. ay aus earn MET Taine state. isch cue GUE: Summary of eoolocical eter waece ahinsd eles, fa lal cg so ONES OLA is Bh eran tereeeamn OLE CHAPTER XVIII. Formations ot the Skagit mountain range. : .. J. 8. 0. 2c Oe Generalt'statement oa. 3s ye eae 6 Se i ee Oe Stratified sorma trons! 26%.) yee te aay aac) oe ee eS Hozomeensseriess cee Sank eee LO eR Ee eee ee Chilliwack series.. ... Se Sin A i EAs AL DE A eee aS General character and Giccmation se antes ah ls ethos com Gir OS Detailed sections and the forsiliferous Hovieoust Sener tenn a) 5) (07 General icolumnarsectionas ss. 22 0 ee ee ee Geological-age ‘ofthe series: © 4455) 24a ae came es Non te cine Cultus formation.. .... Ree ee ee ran Gaiety eb ier an Cie 5715) Siraticnaphy and structimemem me ee ee 516 FVOSSU Soke W550 Fo. Site. be ne ate Ue cr Pann eal B oe RIN ancl Genie a is oe Me oie, DOA eg SS REE Sea een eee ae enn EE) ES Enamel oaormationt: Pee MM Ln a MMM e LEME Ane! URNA Se) Ly] figneous-rock formations. «062 i ee ee eh eater Chilliwack volcanic formatiane sib dia te hee cn See, ae erm OND Weddet ercenstoné:. ;.../. 8. sc iee ess ee 522 Custer: granite-eneiss. . <2 eis hae Ue Le oe Original rock-type. « ..ci9 5a Wises ese le Ge oa pete ee eee Banded structure.) 2% }os Sees Se ak. oe ee Sumiasi granite and diorites.) 220006 . i aicldit eee eee ae ee Granite dhs ek oR A ee GE eee Dronites ie fsck .s: ohn apne gn eldinn ite than Do wieee te Leas co ee rere Sine coldaniedormationy Mca. ow, 528 Skeeutyharzburgites. fae bu) ek eA See oe REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a Slesse:dioritess. ice cies sie ees Retrography sah aria repo near ee ae Contact metamorphism... .. ....... Chilliwack granodiorite batholith.. .... . Petrography. . Contact metamorphism. . : Intrusives cutting the Skagit woleanitae Monzonite stock. . Wikeswaive coves Dikes cutting fe) @hilliwack batholine Acid dikes cutting the Chilliwack series. Basic dikes and greenstones in the ila. 5 SETICS tains asta GEE Structural relations. . @orrelatiome. <0 ss ere PART II CHAPTER XIX. Correlation in the Western Geosynclinal belt.. . Principles used in correlation. . Fa Correlation among formations at the foes intl aralley Correlation within the Western Geosynclinal belt. . General features of the Western Geosynclinal belt.. . CHAPTER XxX. Summary of geological history and note on orogenic theory. . Geological history of the Cordillera at the Forty-ninth Paraltelse : ‘ : ; Observations bearing on the theory of mountain-building. . CHAPTER XXI. Glaciation of the Cordillera at the eee aoe Introduction. . BAW ea cat WANs Relea ohco eM Ren ete Clarke range. ois Shots Nature and oat ae elaciel erosion. : Galton-MacDonald mountain group.. ...... Purcell mountain system. . Selkirk mountain system. . Columbia mountain system and the Teen Pintenee Okanagan range. Hozomeen range. . Skagit range.. .. Summary.. XVIE Pace. 532 532 534 534 535 540 540 541 541 54D 543 543 544 545, 547 547 550 555 565 567 567 572 517 577 579 580 584 586 588 589 591 593 594 597 XVili DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 CHAPTER XXII. Physiographic notes on the Forty-ninth Parallel section.. .. . Origin of the master valleys. . Individual mountain ranges as wlieelonatan rutin Front range syncline.. . ae Galton-MacDonald horst. . Question of a Tertiary danguletiad in ithe Redes Mountain bein. Purcell compound horst. . Nelson range monocline. . : Bonnington-Rossland mountain prone Christina range and Boundary Creek dee ‘Midway volcanic district. . eer ES a mteriorJelateaus: = ace Genoese aires Okanacantranve:. Si Ge ein ee eos Hozomeen range... ..... Skagit range. : Question of a eonerel iNertiaey ‘peneplain in iene Gascadol mountains Development of accordance of summit in alpine mountains. Explanations by inheritance. . ; Spontaneous ee of summit- level accordance Summary. . ; General conclusions on othe phoniepreakic eee ae ‘le “Cordillera at the Forty-ninth Parallel. . CHAPTER XXIII. First calcareous fossils and the origin of the an limestones.. .. Introductory; abstract of chapter... Explanations of the unfossiliferous character of “he pre- a@ammbeian ged ments. wae ae eraineae of the Helen eeolie Naesee nom oe foseile remains. Brooks hypothesis. . Suggested hypothesis. . : Precipitation of lime salts firouee the Gecompesition of dead aoninieme Duration of the nearly limeless sea. Becks Effects of the Huronian orogenic Seaeintnorilc Analyses of the Ottawa river. as Be aes Comparison of the Ottawa aad athens TIVETS. 2) css : Chemical contrast of pre-Cambrian and later river ‘ene tee Variations in the calcium supply during and after the pre- ‘Gamba First calcereous fossils . is Ma ae ale Tests of the suggested eothece Bt Corroborative experiments.. .. .. Observations on the Black Sea. . Pre-Cambrian sedimentary deposits. . ; Origin of dolomite and of other magnesian ‘sadinmenteae 656 656 661 661 REPORT OF THE CHIEF ASTRONOMER xix SESSIONAL PAPER No. 25a Pace. Average ratio of calcium to magnesium in the limestones of the different periods. . Pye Sal hal olde taienl stele eA OOA: Origin of certain iron ores, Elects! andl Paepera! San rea el 669 Origin of the petroleum and natural gas emanating from lise Cambrian sediments.. .. . 669 Direct evidence of the ehemical tee miter bf ths ans ser in the Priest river and Belt-Cambrian terranes... .. ........ 670 SSULITR TING Tayi) IT UU RR cs) EL TELAT IOLA Cae ges ea aoe Tice ESTETMISCS ARN NS ee re eee RM PLETE ANGI RUAN LATS UN SO Wane rate gi ent, die, GOMELUSTONS OU PEON Ss eke ee aN AEN eae; nme RNG Ht ashain AU aM resins PRO Mie uma Teed CHAPTER XXIV. introductions to) the theory, of igneous rocks.) j222y.yile, | ylucls)) «16, ele lievlven ele Glial Wlassiticatronmor theme meouswLOCkSacne iss. vedi lractvesplye casi ce euniet cn tales ioe Henan RaN GSE Averazencompositions of leading, typesiaiisien.. ai) seauccc ciaiieeMevenieetenen ere mL OIGD Average specific gravities of certain types.. .. ................ 696 Source of magmatic heat.. .. ... Sashes a AO OG Composition of the substratum; ihe general cons ae saline rato Primary acid shell of the earth. . AOR GIANT AA A AM NCR alm mUPN CN Neu ci 7K 0), AMbyssalyiny eChlon: Of MALTA Mia stuish. taec it svenseoy us) ay suet ete te Siipstany Aveo naar AOD Oriein of voleanicnaction sewer oie aes ohewiale PN neni Mie ean TN RIRET AH Ei Ola UY OUP CHAPTER XXV. @lassiicationy of AeneousmntrusiveDOGLeS. Hy ra; eye ae eee ese eine ae een cay Introduction... ... CORA NY ave Tas SLSNIAN VOOR Na NR PAUL aa UL fea Gag) ee nicnuer eleseincation: SPEC ATEC etic Lon CMPD OE MEINE Tey CU MN i GLE THF ECECUDOCTES car cven crete ge ie conse Mae ed Wale liley eT UROL QU Et been can MC Nadal nae imme Ga Cp AD) tebeenaere eorart os eet Pek ec ich cars fetdiay Meakice cia MTA PORE Ue UCT eR FMLA yo ede Bana LIG MNTERUISIVEXSNEC Ee isl cline ste cis cial oes to ECE ENGR EN esc OFS oS EE alle oa ey eT GAC COME eee sense red ae Fetot te teh Cal ek RE Sa Te OIE ROG. ARTE Une MEH a ean gh orl Gh EAT CO bab eps siete fete Sear iy fener Santas ong GaN EUA SDM MU US CO aie aM Ea ane vanes ne RIA Sy STMT ala themes woven eet alte Nat a eae Neo) Der Pa I oa REC nL att ante RLU he eau aT) NVA C AMDT CMTC C ey cee Woks ede Hace MeN AR etn E EEUU es Sura eeaN Cea onan tMl ects On (a1) Goro ie erase tel cen an aR ek er eT OE LMC UT Meters Mire naan gue agT Toei saa) als, i OAae tenet PA ie Ae aie eae VA Maiaiaioee Su SOU HAN a MUON) a Ada ean aM 7 O28) SMUT ACETULANOMTESe ys Meine eh ta euiane bate reas esi ea UM OR Mann TRU Mears ews Ulead) TB ZOP Hsieh ST SNe AMA gre Ua eae Seg eT AEE TR ee aaEy 1) SiOCks ey eR eM elec es! Uy. tiara, fay Menem aU SU Rei DEE NM yea pa) tr dale ene at loluthvasee sn stants eek slain usa ban ap amen mavee OMe NE Vy ui pay mir Aes Mle Me On SrOVOSEALCILASSINTCALIOM Men eri iie ky aati weary (out cate) iat aN Cun cs Ne Mahone nail OO 8364—C XX DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 CHAPTER XXVI. PAGE. Mechanism of batholithic intrusion. (202.00 a ee ee Biel relations.) 2 Gh sersg ey eke ij ROR! diner iii OAs RDN INS i eda HE Mimmeyrel abroris’ Ph) iasty Hes aN Nn AAR COUN ERE Ri Cin Na a ne caleelenone) sya GE GSTS AR De nye | B20 AU eG ee a Theories of batholithic aap raavonl Sg) AMT BU CRN SON UH OR A ea net ‘Laccolithic’ hypothesis.. .. .. OSU ee ry EO) a Te ‘Marginal assimilation’ hypothesis. . we pelet he DO ai es cal) a Toe Hypothesis of ‘magmatic stoping’.. . RENT e cohee Meriens (i312 © Magmatic shattering by diferent eran enamine Pye ed (339) Relative! densities of ‘magma’ and’ xenolith: : 20... 347 22 a Influence of plutonic pressures on rock density... .. ........ 744 Sinking of the shattered) blocks:iiy.snefy eo) 0) 2 eo Rise of magma through! stoped) 2: sah). ere Pestimony, of laccolitlis s/o 0 ava Mea IN UR sa se Problem of the cover..... 748 Supply of the eee heats manna ieapenent nnd Me. causes 750 Capacity of superheated, plutonte magma for melting and dissolv- ing KeNOMtHSse) sae 752 Objection founded on ee of tevidenees ‘of lacumalation| at Bb: served \wall-rocksy UF eae aah MU aEY ON AACN Tea ge Abyssal assimilation.. .. . HAMMER AA Ns at 4700515 Existence of basic Snel aad ibatholidiel™ Gide. ER CENTS TAN CE alae co a Differentiation’ of the syntectic magma. 22) 4.57. Sag Origin of granite; the Cancer ce wwlalie igha eck tUN a Hen ee mee OO Eruptive sequence.. .. . Big disetey Dp aives iiss) Veale yp Socal MLUNO Vatu ta om ang Origin of magmatic watet Ande FASES I A Me era) Se ei eye eas Ce General remarks on the stoping Re neeheeee allstl aie sista on cio ana ama CHAPTER XXVII. Masrmatiej;ditterentiatrome ec anise ae ee RN Ree te Preliminary Motes seuss) sens) sg lets el oll wltee etal aibecahes bea ose sho oeL Tne aT We eT oan TRSEAN Gb Gaiaiell oan wie) baitiee ie a gen dens yetnrtBolked crab Sle vay een ells tS edt am ERO Marmited! rmaserorlatye ye sale eles sie tavlep et LM a asten te ale Mitek AGA PUURIN Mecano aC Gravitative. differentiation; 2... VC os Te Origin of basic contact-shells.. ...... Wr eh Chemical contrast of plutonic and Car cestmaline. afrasiee ra “ieee Expulsion (of residual: magma. 2/2 ei) oe seb eee Eitect\ of solution. of Loreien) TOCK. 62.230) oes eee CHAPTER XXVIII. General theory of the igneous rocks and its application.. .. ........ 717 @ondensed ‘statement of a general ‘theory. 2.2.) se ee ee REPORT OF THE CHIEF ASTRONOMER xxi SESSIONAL PAPER No. 25a PAGE. Genetic classification of magmas.. .. NE ac wep OT) Application of the Lac to the ore ene Parallel abe SS AN TAGS) Introduction...... SEAN Aa e en arate me Ur ar() Evidence of a afin aed fee nals Ss Seta ia obs Se A ore Me 113) Hividence ota basaltic Substratums sy steve iy pie Sli Lea ieee ne afi) SVMPCCEUCS ered icc iste osha Han mua a can AS UM AAU aL ADVIS MMAR Tadd Re a RURAL Meera 1beseciceyeiaced sg) Mr ateaeindaeat MoMA Sia onal fuel on SN MA Da aM Ana SUR aa ATS ee divi Pesos ose hal: DAA ty Oc MRSA BE WIE PHU MER MAGAVANE ARR A bl? va rg SY The diorites and acid Bndetea Bi EDN NDS CMF MY ASI59 The complementary dikes and sheets, nd a ea, hea PAB STO MHevAbNOrmall apse acc siyie sy consis aves Nisan ae epee cs ALU Mall gully Hina ay Nee TES Gp Ae allcalrmeyROGK seni sy maa: (cs Mi EL eC eNMr nN MTU ada heal |b Ne able Es AA aM ELE APPENDIX ‘ A’ Mablesotuchemicalaamaly seis, sus usus cds lc leclves (vaieuy renal Mele AUN sieht aie AAD Loan Fe Wau Reg) sp APPENDIX ‘B’ Report on fossil plants, by Professor D. P. Penhallow, D.Sc., F.G.S.A... 800 Oo OO -y ILLUSTRATIONS. PLATES. . Frontispiece—Terminal Boundary monument set by the first International Commission at the Pacific shore. . Profile sections showing relative reliefs of the Alpine chain, the Himalayan chain, and the part of the Cordillera of North America between the Gulf of Georgia and the Great Plains. . Map illustrating proposed subdivision of the Cordillera in the vicinity of the Forty-ninth Parallel. . Boundary slash across the Rocky Mountain Trench at Gateway. . Looking east across the Purcell Trench, from western edge of Kootenay river delta near Corn creek. . Belt of the Interior Plateaus; looking north from near Park mountain, Okanagan range, over Ashnola river valley. . Looking down Kintla Lakes valley. . Cameron Falls on Oil Creek at low-water season. . Mount Thompson, seen across Upper Kintla Lake. . A.—Sheared phase of Siyeh limestone, Clarke range. B.—Sheared phase of dolomitic lense (weathered) in Kitchener formation, at Yahk river. xxii DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 . Casts of sallt-crystals in Kintla argillite. . Looking east across Flathead Valley fault-trough to Clarke range. . Head of Lower Kintla Lake. . A.—Cliff in Siyeh limestone, showing molar-tooth structure; at cascade in Phillips Creek, eastern edge of Tobacco Plains. B.—Concretion in dolomite; lower part of Gateway formation, Galton range. . A.—Limonitized, simple and twinned crystals of pyrite, from Gateway for- mation at summit of McGillivray range. B.—Similar pyrite crystals in metargillitic matrix. . Exposure of the massive Irene conglomerate in head-wall of glacial cirque. . A.—Ripple-marks in Ripple quartzite; positives. B.—Ripple-marks in Ripple quartzite; negatives (casts). . Negatives of rippie-marks in quartzite. Summit of Mt. Ripple. . Mount Ripple and summit ridge of the Selkirk mountain system. . Columnar sections of the Summit, Purcell, Galton, and Lewis series. . Diagrammatic east-west section of the Rocky Mountain Geosynclinal at the Forty-ninth Parallel. . A—Molar-tooth structure in Siyeh limestone (weathered), Clarke range. B.—Molar-tooth structure in Castle Mountain dolomite (unweathered) on main line of Canadian Pacific railway. . A.—Porphyritic phase of the Purcell Lava; from summit of McGillivray range. B.—Quartz amygdule in the Purcell Lava. . Secondary granite of a Moyie sill, fifty feet from upper contact. . Phases of the Moyie sill: specimens one-half natural size. . Looking eastward over the heavily wooded mountains composed of the Priest river terrane, Nelson range. . A—Contrast of normal sericite schist of Monk formation (left) and contact- metamorphosed equivalent in aureole of summit granite stock, a coarse- grained, glittering muscovite schist (right). B.—Spangled, garnetiferous schist characteristic of Belt E of Priest River terrane. ; . Typical view of Bonnington-Pend d’Oreille mountains of the Selkirk system. . Percussion marks on quartzite boulder in bed of Pend d’Oreille river. . A.—Sheared phase of the Rykert granite, showing concentration of the femic elements of the rock (middle zone). B.—Massive phase of the Rykert granite, showing large phenocrysts of alka- line feldspar. . Tourmaline rosettes on joint-plane of quartzite; from contact aureole of summit granite stock, Nelson range. . Felsenmeer composed of Rossland volcanics, Record Mountain ridge, west of Rossland. . Two views of shatter-belt about the Trail batholith, Columbia river. . Sheared Cascade granodiorite, showing banded structure. . Park land on Anarchist plateau east of Osoyoos lake. . Fossil plants in the Kettle River sandstone. REPORT OF THE CHIEF ASTRONOMER xxiii SESSIONAL PAPER No. 25a 37. 38. 39. Specimens of nodule-bearing peridotite from forty-foot dike cutting schis- tose rocks of Basic Complex. Western slope of Anarchist mountain-plateau, viewed from west side of Osoyoos lake. Types from the Kruger alkaline body: A.—Porphyritic alkaline syenite. B.—Nephelite syenite (salic variety). C.—Malignite. . View looking southwest from slope of Mt. Chopaka. . Typical view in higher part of the Okanagan range. . A.—View of cirque head-wall composed of massive Cathedral granite. B.—Felsenmeer on Similkameen batholith. . Looking southeast along summit of Skagit range from ridge north of Depot creek. . A.—Carboniferous limestone, summit of McGuire mountain. B.—Rugged topography at the Boundary, east of Chilliwack lake, and north of Glacier Peak. C.—Horn topography between Tamihy and Slesse creeks. D.—Horn topography on ridge between Slesse and Middle creeks. . Western edge of Skagit range, viewed from alluvial plain of the Fraser Valley at Chilliwack. . Summit of the Skagit range. . Typical view of granitic mountains (Chilliwack batholith). . Mount Baker, taken from prairie at Sumas lake, Fraser valley, . Profile cross-section of the Cordillera at the Forty-ninth Parallel, showing vertical limits of Pleistocene glaciation, ete. . Glaciated valley of Starvation creek, Clarke range. . Head-wall of glacial cirque, summit of Clarke range. . Winged-out moraine at mouth of Starvation creek canyon, in Flathead valley. . A.—Hanging valley of Phillips creek, cascading into Kootenay river valley near Gateway. B.—Drumloidal deposit and water-filled glacial kettle in thick drift of the Rocky Mountain Trench (Tobacco Plains). . Tandem cirque-lakes near summit of Nelson range, seven miles north of Boundary line. . Looking east across the Columbia river to Boundary Town, lying in the old gravel-floored bed of the Pend d’Oreille river. . Abandoned channel of the Pend d’Oreille river at Boundary Town. . Winter-talus ridge on southern wall of glacial cirque, Okanagan range. . Wooded boulder-moraine forming dam at lower end of Chilliwack lake. . Looking up Chilliwack lake from point near its outlet. . Looking up Chilliwack lake over forested morainal dam of the lake. . A—View of the gravel plateau representing the late Pleistocene delta of the Fraser river. B.—Detailed section in the sands and gravels of the Pleistocene deposit represented in A. XxiV DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 62. A.—Photograph showing relatively rapid erosive effects of glacierlets with very small accumulators (snow-fields). B.—Small glacier deepening cirque, about seven thousand feet above sea. 63. Lower Okanagan valley and Osoyoos lake. 64. Looking southeast across Starvation creek canyon. 65. Compound alluvial cone at Midway. 66. Accordant summit levels in the Selkirk range. 67. Plateau-like surface of Remmel batholith. 68. Plateau-like surface of unroofed Similkameen batholith. 69. Meadow and park near tree-line about six thousand feet above sea-level, Bonnington range. 70. A.—Coarse felsenmeer in massive grit of the Wolf formation. B.—Coarse felsenmeer in quartzite of the Ripple formation. 71. A.—Looking south along ridge between Middle and Slesse creeks, Skagit range. B.—Southern slope of Mount Ripple, Selkirk range. 72. (Sheet No. 18), in Part III (with maps). A.—Typical view in Clarke range. B.—Summit of the Nelson range (Selkirk system). O.—Nelson range, looking west from summit ridge north of Dewdney trail. 73. (Sheet No. 19), in Part III (with maps‘. A.—Columbia River terrace and the Pend d’Oreille mountains (Selkirk system). B.—Typical view in the Midway mountains. O.—Typical view in the Skagit range. FIGURES. 1. Diagrammatic map showing subdivision of the Rocky Mountain system at the Forty-ninth Parallel. 2. Diagrammatic map showing subdivision of the Purcell mountain system at the Forty-ninth Parallel. 3. Diagrammatic map showing subdivision of the Selkirk mountain system at the Forty-ninth Parallel. 4. Diagrammatic map showing subdivision of the Columbia mountain system at the Forty-ninth Parallel. 5. Diagrammatic map showing position of the structure-section east of the Rocky Mountain summit. 6. Structure section across the strike, along the ridge southeast of Oil creek, eastern slope of the Clarke range. 7. Diagrammatic drawing from thin section of Waterton dolomite. 8. Diagrammatic drawing from thin section of typical sandy dolomite of the Altyn formation. 9. Section showing common phase of the molar-tooth structure in the Siyeh formation. 10. Diagrammatic drawing to scale, from thin section of amygdaloidal basalt in the Sheppard formation, Clarke range. REPORT OF THE CHIEF ASTRONOMER XXV SESSIONAL PAPER No. 25a aie 12. 16. i 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. Drawing from thin section of metamorphosed argillaceous sandstone, Wolf formation. Diagrammatic map showing approximate position of the Rocky Mountain geosynclinal prism in its older phase. . Locality map of the Moyie sills. . Section of Moyie mountain and the Moyie sills, along the International boundary line. . Diagram showing the petrographic nature of each of the Moyie sills and its stratigraphic position in the quartzites. Diagram illustrating the hypothesis that the partially differentiated syn- tectie magma of a thick sill may break through the roof and form, at stratigraphically higher horizons, several thinner sills differing in com- position among themselves. East-west section on ridge north of Lost Creek, Nelson range. Diagram showing stage of development of the thrust illustrated in Figure iG North-south section illustrating probable explanation of the great intensity and extent of the contact metamorphism at Summit creek. Diagrammatic map of summit granite stocks with wide aureole of contact metamorphism. Section along line A-B of Figure 20. Diagrammatic section showing relation of the summit stocks of Nelson range to the Bayonne batholith. Map showing relations of Pend d’Oreille argillite, aplitic granite, and two dikes of minette. Section of syenite porphyry chonolith satellitic to Coryell batholith. Section northeast of bridge over Kettle river, six miles above Midway. Partly diagrammatic drawing from thin section of ground-mass of ‘ shacka- nite.’ Section of the Okanagan composite batholith. Map showing relations of the Osoyoos, Similkameen, and Kruger igneous bodies and the invaded Paleozoic formations. Plunging contact surface between the Similkameen batholith and the Chopaka roof-pendant. Outcrop of the same intrusive contact surface shown in Figure 29. Map of the Similkameen and Cathedral batholiths and the Chopaka intru- sive body, as shown in the Boundary belt. Map showing relations of the Cathedral and Remmel batholiths and the Ashnola gabbro. Map showing relations of the Remmel batholith, Park granite, and Basic Complex. Columnar section of the Pasayten series, including the Pasayten Volcanic formation (member A.) Map showing relations of the Castle Peak stock to the deformed Pasayten formation. Contact surface between the Castle Peak granodiorite and tilted Creta- ceous sandstones and argillites. xxvi DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 37. Plunging contact surface between intrusive granodiorite of Castle Peak stock and Cretaceous argillites and sandstones of Pasayten series. 38. Plunging contact surface between intrusive granodiorite and Pasayten for- mation. 39. Plunging contact surface, Castle Peak stock, south side. 40. Intrusive contact between granodiorite and nearly vertical Pasayten ar- gillite. 41. Diagrammatic section showing origin of a ‘ winter-talus ridge.’ 42. Illustrating two methods by which basic contact-shells in a stock or a dike might be formed. ie TABLES Correlation of the Rocky Mountain Geosynclinal rocks.. .. II. Showing general lithological character of the four standard 100k sections in the Rocky ‘Mountain Geosynclinal. . : Showing composition of equivalent formations (Kipehewer: Siyeh).. . Showing compeatnon be aiwalent eormationsh Onesto Altyn).. . Densities be foemetionee in athe WRoucy Monntainl Geosynelinal . Correlations in the Rocky Mountain Geosynclinal . . Walecott’s correlations in the Belt terrane. 4 ie . Correlation with Canadian Pacific Resiway eeeoue ee . Weight percentages of minerals in rocks of Moyie eten ne . Chemical analyses of phases of the Moyie sills. . . Columnar section through the Moyie sills.. .. .. .. . Analyses of sill granite and invaded sediments, Movie aillaiel . (Annulled in press.) . Analyses of Rykert granite and related rock. . . Analyses of minettes.. ...... : . Analyses of augite latites.. . 3 . Analyses of augite-biotite latites.. .. . Analyses of hornblende-augite latites. . ik . Comparisons among latites and monzonite.. .. .. ...... . Chemical relations of Rossland monzonite.. .. .. . . Comparison of basic syenite and average minette.. .. . Analyses of missourite. . Analyses of rhomb- nomahenes and aclated! roche . Analyses of pulaskite porphyry and related rock. . . Analyses of malignites and nephelite syenites. . . Showing chemical relation of Similkameen and @athedenl batholiths. . . Analyses of members of WOlanaoan’ connec pachelithee . Correlations among members of Okanagan composite batholith PaGeE. 161 167 169 169 173 178 182 194 235 236 237 241 287 312 325 327 329 332 344 357 368 406 419 454 463 472 474 REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a XXXII. XXXII. XXXIV. XXXV. XXXVI. XXXVI. XXXVIUTI. XXXIX. XL. XLI.-XLII. XLII. XLIV. XLV. Analyses of Castle Peak and Similkameen Sy amie Analyses of granodiorites.. .. Correlation at the Hora ninth Parallel: ‘ Correlations within the Western Geenclinal Pelee Geological events in provinces of the Western oan Belt. . Principal events ean in ihe Western Geosynclina Bolte as a whole.. Calcium and magnesium i in Bohemian: rivers. Analyses of Ottawa river. watts ss Calcium and magnesium in various rivers. ees Experimental results with ammonium carhonatoe , Calcium and magnesium in limestones of the ponlocical periods. . Ghee Average gomboeaens clonal foe ihe pene ieneous-rock types.. Comparison of Oe erate anelnes cE pranitel Nasal Aiorite, and andesite. . . Comparison of average Bralvees tof sranite end fern mass of augite andesite. . . Showing rates of thermal iiiinciiy 4 in Pec . Specific gravities of rocks and glasses. . att . Specific gravities of crystals and AIERES. Srvahehity . Decrease in density, rock to glass at 20°C.. . Specific gravities of rocks and melts.. ae ts . Change of density of rocks with change of (ame ON, - : Showing quantities of volatile matter in sediments. . . Water in igneous rocks. . sacred . Comparison of plutonic and eineive! Poke APPENDIX ‘A,’ Table of chemical analyses made for the report.. .. .. .. 2.2 2+ oe oe ee PARTIII XXVii 793 Containing seventeen geological maps, with structure sections (sheets 1 to 17), and two sheets of photographic panoramas (sheets 18 and 19). 8364—D Nuts 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER I. INTRODUCTION, Area Covered.—In 1901 the writer was commissioned by the Canadian Minister of the Interior to undertake the geological examination of the mountains crossed by the Boundary Line between Canada and the United States, at the Forty-ninth Parallel. Field work was begun in July of that year and continued through the different summer seasons to and including that of 1906. During the summer of 1901 reconnaissance surveys on the American side of the same line were led by Messrs. Bailey Willis, F. Leslie Ransome, and George Otis Smith, members of the United States Geological Survey. No further geological work in connection with the Boundary survey was carried on by the United States Government, and the map sheets prepared by the United States topo- eraphers were placed at the disposal of the writer as geologist (for Canada) to the International Boundary Commission. The present report represents the principal results of the study made during the six field seasons. The geological examination covered a belt along the Forty-ninth Parallel, from the Strait of Georgia to the Great Plains. The belt is 400 miles long and varies from 5 to 10 miles in width, with a total area of about 2,500 square miles. Its width was controlled, in part, by that of the map sheets prepared by the topo- graphers of the Commission parties; in part, by the necessity of depending on the trails which those parties built into the Boundary belt. As a rule, this moun- tainous belt is heavily wooded and, without trails, is almost inaccessible to pack-animals. During the first three seasons accurate topographic maps on the required scale were not available, and in 1902 and 1903 the writer used, as topographic base, an enlarged copy of the West Kootenay sheet of the Canadian Geological Survey. In that relatively accessible part of the Boundary belt (from Grand Forks to Porthill, eighty miles to the eastward) it was found possible to cover a zone ten miles in width. Conditions of Work in the Field—No geologically trained assistant was employed in any part of the field. The work was, therefore, slow. Each traverse generally meant a more or less taxing mountain climb through brush or brulé. The geology could not be worked out in the detail which this moun- dain belt deserves. For long stretches the rock exposures were found to be poor. Such was the case for the heavily drift-covered mountains between Osoyoos lake and Christina lake, and, again, for nearly all of the 60-mile sec- tion between the two crossings of the Kootenay river, at Gateway and Porthill. Some confidence is felt in the maps aud structure sections of the Rocky Mountains proper (from Waterton lake to Gateway), of part of the Selkirk 25a—Vol. ii—1 2 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 mcuntain system (from the summit to the Columbia river), and of the Okana- gan and Hozomeen ranges of the Cascade mountain system. Elsewhere, the maps and sections, in their rigid lithography, suggest more certainty as to the run of contacts and as to underground structures than the writer actually feels. As a whole, the results lie half-way between those of a reconnaissance survey end those of a detailed survey. One of the leading difficulties felt by all workers in this part of the Cordil- lera is the remarkable lack of fossils in the sedimentary rocks. The writer has been able to discover but few fossiliferous horizons additional to the small number already known to Canadian and United States geologists. Many of the correlations offered in the following report are to be regarded as strictly tenta- tive and should not be quoted without reference to the many qualifications noted in the running text. Some large proportion of the inaccuracy in maps and sections is due to the fact that for half of the field seasons the writer was provided either with no topographic map or merely with the four-miles-to-one-inch West Kootenay reconnaissance sheet of the Canadian Geological Survey. This sheet is excellent fur its purpose, but was manifestly not intended for the use of the structural gcologist, whose topographic-map scale should be at least one mile to one inch in the Selkirk and Columbia mountain systems. First in 1904, the writer was able to use copies of the manuscript Boundary Commission plane-table maps, on the scale of 1: 63,360. Sheets 1, 2, 3, 4, 5, 12, 18, and 14 were constructed on that basis and are superior in accuracy of geological information to the cther sheets. Between the Gulf of Georgia and the western limit of sheet 17 the Boundary line crosses a continuous thick deposit of Pleistocene gravels and sands. No other formation is there exposed in the five-mile belt and the broad plain is not represented in the maps. Acknowledgments.—The writer was efficiently aided in the physical work of carrying on the survey, during five seasons, by Mr. Fred. Nelmes of Chilliwack, British Columbia. His faithfulness in many a tedious place was worthy of his sterling work as a mountaineer. During the season of 1903 the writer was similarly assisted in able manner by Mr. A. G. Lang, of Waneta, British Col- umbia. In the field many courtesies and much help were extended by Mr. J. J. McArthur, chief topographer for the Canadian branch of the Boundary Commission; by Mr. E. C. Barnard, chief topographer for the United States branch, and by his colleagues. In the office work the writer was aided by many members of the Geological Survey of Canada, and owes much to the personal encouragement of Honour- able Clifford Sifton, Minister of the Interior, during the progress of the survey. In numberless ways the work was forwarded by the able and most generous help of the Canadian Commissioner, Dr. W. F. King, to whom the writer owes the greatest debt of acknowledgment. Professor D. P. Penhallow of McGill University has made thorough study of the collections of fossil plants. The collections of fossil animal remains were, with much generosity, carefully REPORT OF THE CHIEF ASTRONOMER ane SESSIONAL PAPER No. 25a studied and determined by Drs. T. W. Stanton, G. H. Girty, and C. D. Walcoté of the United States Geological Survey, and by Dr. H. M. Ami, of the Geologivat Survey of Canada. Professor M. Dittrich, of Heidelberg, Germany, and Mr. M. F. Connor, of the Canadian Department of Mines, performed valued service in making the large number of chemical analyses noted in the report. The dzaughting has been performed with zeal and care by Mr. Louis Gauthier, of the Chief Astronomer’s office at Ottawa, and by O. O. Senécal and his assistants of the Geological Survey. A number of professional geologists have discussed theoretical matters and thus markedly assisted in the composition of the report. To each of these gentlemen the writer tenders his thanks for all their kind and efficient help. Equally sincere thanks are due to the president and corporation of the Massachusetts Institute of Technology, who for more than two years have granted every available facility for the preparation of this report. Special mention at this place may also be made of the fact that chapter XIV is largely a direct quotation from Mr. R. W. Brock’s report on the Geology of the Boundary Creek Mining District. The corresponding part of sheet No. 10 has been compiled from the map accompanying Mr. Brock’s report. In view of the care spent on this part of the Boundary belt by this able investi- gator, it seemed inadvisable to spend much of the limited time allotted to the transmontane section on the Boundary Creek district. Accordingly, the present writer made no more than a couple of rapid east-west traverses across the district, corroborating, so far, the accuracy of Mr. Brock’s mapping in a par- ticularly difficult terrane. Professor Penhallow’s paper on the collection of fossil plants forms an appendix to the present report. Collections—During the survey 1,525 numbered specimens, with many duplicates, were collected. Each of the localities, whence the specimens chemi- cally analysed were taken, is noted on the map sheets with a small cross and the collection number. Some 960 thin sections of the rocks were prepared and studied. Sixty rock analyses and one feldspar analysis were made for the report. Thirteen hundred photographs were taken by the writer, besides which many kundreds of others were taken by the photo-topographic parties operating for the Canadian branch of the Boundary Commission. Previous Publications by the writer on the Forty-ninth Parallel Geology.— After each field season a brief account of the ground covered was published either in the summary report of the Director of the Geological Survey of Canada or. in the annual report of the Chief Astronomer of Canada. As the work pro- gressed it was thought advisable to publish separate papers on certain general and theoretical problems, which had arisen during the survey of the Boundary belt. The list of these papers, some of which, in more or less amplified form; form parts of this report, is as follows :— tR.W. Brock, Annual Report, Geological Survey of Canada, Vol. 15, 1902-3, Part A, pp. 98 to 105. — 25a—Vol. 1i—14 4 DEPARTMENT OF THE INTERIOR ’ 2 GEORGE V., A. 1912 1. The Geology of the Region adjoining the Western Part of the Inter- netional Boundary: Summary Report of Geological Survey Department of Canada for 1901, in Annual Report, vol. 14, 1902, Part A, pp. 39-51. 2. Geology of the Western Part of the International Boundary (49th Parallel): Summary Report Geological Survey Department of Canada, for the year 1902, in Annual Report, vol. 15, Ottawa, 1903, Part A, pp. 136-147. 3. The Mechanics of Igneous Intrusion: Amer. Jour. Science, vol. 15, 1903 pp. 269-298. ; 4. The Mechanics of Igneous Intrusion (Second Paper): ibid., vol. 16, 1903, pp 107-126. 5. Geology of the International Boundary: Summary Rep. Geol. Survey Department of Canada for 1903, in Annual Report, vol. 16, Ottawa, 1904, Part A, pp. 91-100. 6. Geology of the International Boundary: Summary Rep. Geol. Survey Bepartment of Canada for 1904, Ottawa, 1905, pp. 91-100. 7. The Accordance of Summit Levels among Alpine Mountains; the Fact and its Significance: Jour. Geology, vol. 13, 1905, pp. 105-125. 8. The Secondary Origin of Certain Granites: Amer. Jour. Science, vol. 20. 1905, pp. 185-216. 9. The Classification of Igneous Intrusive Bodies: Jour. Geology, vol. 13, 1905, pp. 485-508. 10: Report on Field Operations in the Geology of the Mountains crossed by the International Boundary (49th Parallel): (1) in Report of Chief Astro- nomer for Canada for 1905, Ottawa, 1906, pp. 278-283; (2) in Rep. of Chief Astronomer for Canada for 1906, Ottawa, 1907, pp. 133-135. 11. The Nomenclature of the North American Cordillera between the 47th and 58rd Parallels of Latitude: Geographical Journal, vol. 27, June, 1906, pp. 536-606. 12. Abyssal Igneous Injection as a Causa] Condition and as an Effect of Mountain Building: Amer, Jour. Science, vol. 22, Sept., 1906, pp. 195-216. 13. The Differentiation of a Secondary Magma through Gravitative Ad- justment: Festschrift zum siebzigsten Geburtstage von Harry Rosenbusch, Stutt- gart, Germany, 1906, pp. 203-233. 14. The Okanagan Composite Batholith of the Cascade Mountain System: Rull. Geol. Soc. America, vol. 17, 1906, pp. 329-376. 15. The Limeless Ocean of pre-Cambrian Time: Amer. Jour. Science, vol. 23, Feb., 1907, pp. 93-115. 16. The Mechanics of .Igneous Intrusion (Third Paper): Amer. Jour. Science, vol. 26, July, 1908, pp. 17-50. 17. The Origin of Augite Andesite and of Related Ultra-basic Rocks: Jour. Geology, vol. 16, 1908, pp. 401-420. 18. First Caleareous Fossils and the Evolution of the Limestones: Bull. Geol. Soc. America, vol. 20, 1909, pp. 153-170. 19. Average Chemical Composition of Igneous-rock Types: Proceedings Amer. Acad. Arts and Sciences, vol. 45, January, 1910, pp. 211-240. REPORT OF THE CHIEF ASTRONOMER 5 SESSIONAL PAPER No. 25a 20. Origin of the Alkaline Rocks: Bull. Geol. Soc. America, vol. 21, 1910, pp. 87-118. Earlier Work on the Geology of the Forty-ninth Parallel—The British and United States governments attached geologists to the parties of the first Inter- rational Boundary Commission appointed (1857-61) to mark the Forty-ninth Parallel across the Cordillera. The geologist, George Gibbs, traversed the Beundary belt for the United States government and published his results in the third and fourth volumes of the Journal of the American Geographical Society, New York, 1873-74. The late Hilary Bauerman was the geologist for ihe British government. His brief report was not published until 1884, when, at the suggestion of George M. Dawson, it appeared as a part of the Report of Progress of the Geological and Natural History Survey of Canada for 1882-3-4, Fart B, Ottawa, 1884. Dawson himself entered the same transmontane belt at its eastern end during his work as geologist to the British North American Boundary Commis- éion. His report published in 1875, at Montreal, bears the title ‘ Report on the geology and resources of the region in the vicinity of the forty-ninth parallel from the Lake of the Woods to the Rocky Mountains.’ Since then Dawson continued his memorable reconnaissance of British Columbia and, in the Boun- dery belt, was accompanied or followed by McConnell, McEvoy, Brock, Leach, Young, LeRoy, Camsell, and other members of the Geological Survey of Canada. On the United States side of the line many other workers have similarly added to our knowledge of the formations crozsed by the Forty-ninth Parallel, though comparatively few of them, other than those already mentioned, have actually reached the Boundary line in their detailed work. Reference to the publications on British Columbia, Alberta, Montana, Idaho, and Washington geology can be readily found in the bibliographic bulletins of the United States Geological Survey and in the general index to the reports of the Geological Survey of Canada (published in 1908). Special note should be made of the papers published by the American geologists attached to the present Boundary Commission, namely :— Stratigraphy and structure, Lewis and Livingston ranges, Montana: by Bailey Willis: Bull. Geol. Soc. America, Vol. 13, 1902, pp. 305-352, A geological reconnaissance across the Cascade range near the Forty-ninth Parallel: by George Otis Smith and Frank C. Calkins, Bull. 235, U.S. Geol. Survey, 1904. Continuation of the Forty-ninth Parallel Section—Mr. Charles H. Clapp is now employed by the Canadian Geological Survey on a structural study of Vancouver island, and it is hoped that materials will soon be in hand for a continuation of the Forty-ninth Parallel section across to the open Pacific. General Sketch of the Subject Matter—The 400-mile section crosses the grain of the Cordillera and accordingly includes a high proportion of all the Cordilleran formations to be encountered in these latitudes. The structural 6 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 eomplexity, like the stratigraphic complexity, is near its maximum for the given area in such a straight cross-section. A preliminary sketch of the different geological provinces traversed by the Boundary belt will aid the reader in under stauding and grouping the mass of observations to be detailed. In the first approach, the Cordillera at the Forty-ninth Parallel may be regarded as divisible into great zones. These are called the Eastern Geo- synclinal Belt and the Western Geosynclinal Belt. The two overlap in the vicinity of the Columbia river. From the summit of the Selkirk range, just east of that river, to the Great Plains, sedimentary formations are dominant and are almost entirely included in one huge structure, hereafter named the Rocky Mountain Geosynclinal Prism (or simply Geosynclinal). The prism extends from Alaska, through the Great Basin, to Arizona. These rocks are sy nearly unfossiliferous that their correlation with the standard systems is a matter of difficulty. Reasons will be shown for the belief that the whole con- formable group ranges in age from the Mississippian to a great unconformity at the base of the Belt terrane, or Beltian system, as recently named by Wal- cott.* Near the western crossing of the Kootenay river, the prism rests on an clder group of metamorphic rocks, here called the Priest River terrane. Tho basement on which the Rocky Mountain Geosynclinal rests is nowhere else exposed on the Forty-ninth Parallel. Younger and much more local geosynclinal prisms, of Cretaceous and Ter- tisry dates, have been laid down on the Rocky Mountain Geosynclinal along its eastern border, in Alberta, Montana, and farther south. None of these younger prisms of great thickness is represented in the section at the Inter- national Boundary, but it is convenient to refer to the whole compound belt of heavy sedimentation under the one name, the Eastern Geosynclinal Belt. Similarly, the dominant sedimentaries west of the Columbia river, of Pennsylvanian, Triassic, and Cretaceous age, have been accumulated in great thicknesses. The Pennsylvanian strata have been recognized at many points, from Alaska to Southern California, and it appears probable that late Paleozoic sedimentation on a geosynclinal seale took place throughout that long stretch. More local Mesozoic and Tertiary geosynclinals were imposed upon the Pacific border of the prism developed in the Pennsylvanian period. Rocks apparently representing this older group of deposits crop out at intervals all the way from the Columbia river to the Gulf of Georgia. A part of one of these Mesozoic prisms was found in an enormously thick mass of Cretaceous strata largely composing the Pasayten mountain range between the Pasayten and Skagit rivers. A thick Triassic series is known on Vancouver island and forms part of the western slope of the Skagit mountain range, which lies between ‘the Skagit river and the Strait of Georgia. The edge of the Tertiary geosynclinal composed of the Puget beds is, apparently, represented in the Fraser valley. To the entire composite mass of post-Mississippian sediments occurring in the western half of the Cordillera, the name Western Geosynclinal Belt may be given. *C.D. Walcott, Smithsonian Miscellaneous Collections, vol. 53, No. 5, 1908, p. 169. REPORT OF THE CHILE ASTRONOMER Tf SESSIONAL PAPER No. 25a The Eastern Geosynclinal Belt is characterized by open folds, fault-blocks, and overthrusts, with but moderate regional metamorphism and quite sub- - ordinate igneous action. The Western Belt is characterized by close folding, mashing, strong regional metamorphism, and by both batholithic intrusion and veleanic action on a grand scale. From the western crossing of the Kootenay river to the Kettle river at Grand Forks the section crosses the West Kootenay Batholithic province, which is partly overlapped by the Rossland Volcanic provinee. West of Grand Forks is the Midway Volcanic province. The Okana- gan (eastern) division of the Cascade mountain system is composed of the kanagan Composite Batholith, and the heart of the Skagit range is made up of the Skagit Composite Batholith. These various geological provinces are treated in the order of succession as they are encountered in passing from east to west. Numberless problems have arisen during the progress of the work. Special studies have been made on the relations and origin of the igneous rocks, which occur in the section on a scale not often surpassed in other mountain chains. A chapter on the nomenclature of the Cordilleran ranges at the Interna- ticnal Boundary illustrates the need of a systematic attack on the difficult problem of names. The long discussed but ever new question as to the origin of limestone and dolomite, coupled with that as to the cause of the rarity of fossils in pre-Silurian sediments, has prompted a theoretical chapter which, like the chapter on nomenclature, has already in largest part been published. Other subjects, including glaciation and physiography, were more inevitably to be considered and need no special introduction at this place. ‘on sana 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER II. SYNOPSIS OF THE REPORT. For the convenience of the reader a brief abstract of each of the following chapters is here offered. As a rule, the petrography, which forms a large part of each systematic section dealing with the rock formations, is not summar- ized. CuHapTer III.—The necessity of subdividing the western mountain chain of North America into relatively small orographic units is felt by the naturalist who covers any large section of these mountains and then attempts to describe the results of his observations. Such subdivision for a belt lying between the Forty-seventh and Fifty-third parallels of latitude is suggested. For scientific writing the well recognized name ‘Cordillera of North America,’ with the alternative, ‘Pacific Mountain System of North America,’ is preferred for the chain as a whole. The many other alternative names for the chain are listed. The existing nomenclature for the ranges crossed by the Forty-ninth Paral- lel is inadequate and to some extent in confusion. An amplified nomenclature, based as far as seemed possible on prevailing usage, is offered. The main principle adopted is that of existing topographic relations, largely irrespective of the genetic history or rock composition of the different ranges. Specifically, the lines of delineation are the axes of the greater valleys and ‘trenches’ in the mountain complex. The Rocky Mountain Trench, the Purcell Trench, the Selkirk Valley, and the Lower Okanagan Valley represent partial boundaries of the Rocky Mountain system, and the Purcell, Selkirk, Columbia, and Cascade Mountain systems at the Forty-ninth Parallel. The suggested subdivisions of these systems for the region adjacent to the International Boundary include the Lewis, Clarke, MacDonald, Galton, Flathead, McGillivray, Yahk, Moyie, Cabinet, Nelson, Slocan, Bonnington, Valhalla, Pend D’Oreille, Priest, Kaniksu, Rossland, Christina, Midway, Colville, Sans Poil, Okanagan, Hozomeen, and Skagit ranges or groups. A small area of the Belt of Interior Plateaus is also represented in the Boundary line section. In the preceding list the names in italics are proposed Ly the present writer. The others date from the expeditions of Palliser, Daw- son, Willis, Smith, Calkins, and MacDonald. The subdivision is illustrated with sketch maps. CuapPTteR IV.—The geological description begins with an account of the Rocky Mountain Geosynclinal Prism, of which nearly all the mountains between the Great Plains and the summit of the Selkirks are composed. Chap- 9 10 DEPARTMENT OF THE INTERIOR 2 GEORGE V.,, A. 1912 ter IV. discusses the stratigraphy and structure of the Clarke range (Livings- tou range of Willis), the most easterly of the mountains covered by the survey. Willis’ results on the succession of formations were confirmed by detailed study. The oldest formations in this part of the Rocky Mountain system are the Altyn and Waterton magnesian limestones and dolomites. The former is believed to be considerably thicker than the minimum estimate given by Willis, in whose traverse the lower part of the Altyn formation was not visible. The base of the Waterton formation is concealed. At Waterton lake a boring has located the plane of the Lewis overthrust at a depth of about 1,500 feet below the lake-level. At that level the bit of the machine entered soft shaly rocks assigned to the Cretaceous. The fossil Belttna danai was found in the Altyn formation. No other determinable fossils were found in this range, the sediments of which were assigned by Willis to the Belt terrane of the Algonkian. They are here alluded to as the Lewis series. At the Flathead river, a local fresh-water, fossiliferous deposit of clays and sands—the Kishenehn formation—occurs; it is assigned to the Miocene. The Clarke range forms a dissected broad syncline, which is accidented with a few faults and secondary warps. The valley of the North Fork of the Flathead river is an eroded graben or fault-trough. The range has been moved eastward at least eight miles along the great Lewis thrust. The writer favours the view that this thrust, as well as nearly all the other deformation represented in the range, dates from the close of the Laramie, but this has not been finally preved. CuHaPTER V.—Continuing westward, the older members of the geosynclinal, all unfossiliferous, were found to make up the greater part of the MacDonald and Galton ranges. The lithology has, however, changed and in some cases new names are given to the constituent formations. The whole conformable group, corresponding to the Lewis series, is called the Galton series. On the east and west sides of the Galton-MacDonald mountain group down- faulted blocks of fossiliferous limestone, upper Devonian to Mississippian in age, make contact with some of the lowest members of the much older Galton series. The dominant structural unit of the twin ranges is the fault-block. CHAPTER VI.—West of the Rocky Mountain Trench the geosynclinal rapidly assumes a lithological character markedly different from that found in the four ranges just mentioned. The Purcell system is largely composed of massive quartzites and metargillites, forming the Purcell series, which is the more silicious equivalent of the dominantly argillaceous and calcareous or dolomitic sediments of the Lewis and Galton series. The Purcell series is of much more Lomogeneous composition than the other two series. An interbedded volcanic formation, of the fissure-eruption type, has been followed from the Great Plains to the summit of the McGillivray range, where the lava is thicke:t. It is named the Purcell Lava. A special feature of the REPORT OF THE CHIEF ASTRONOMER av AL SESSIONAL PAPER No. 25a Purcell system is the presence of thick sills of a peculiar hornblende gabbro. These eruptive formations are described in later chapters. The Purcell system is also characterized by numerous examples of block- faulting, though the McGillivray range shows a broad anticline and syncline. CHapter VII.—At the Purcell Trench the continuity of the geosynclinal mass is effectively broken. From the alluviated floor of the trench to a line about sixteen miles farther west the rocks chiefly belong to the older Priest River terrane, on which the geosynelinal was deposited. At the summit of the Nelson range the nearly entire thickness of the geosynclinal is exposed, the prism having here been upturned to a vertical position. Its sedimentary mem- bers are heterogeneous, including conglomerates, grits, coarse and fine sand- stone (quartzites), and metargillites. A very thick volcanic formation, older than the Purcell Lava, is interbedded. A great unconformity at the base of the geosynclinal is exposed. The name Summit series is given to the whole con- formable group of formations, from the basal uneonformity to the horizon corresponding to the youngest member of the Purcell series. West of the great menocline the Summit series makes an apparently conformable contact with a younger metamorphosed mass of sediments named the Pend D’Oreille group. West of that contact the Rocky Mountain Geosynclinal rocks do not reappear in the Boundary section. Cuapter VIII.—In this chapter the detailed description of the Selkirk geology is interrupted, and the correlation of the Lewis, Galton, Purcell, and Summit series is discuzsed. The systematic variation in the lithology of the geosynclinal, as it is crossed from east to west, is noted in some detail, and the conclusion is drawn that the source of the clastic materials lay to the westward, probably not far from the present location of the Columbia river. Notes on the metamorphism of the prism and on its average specific gravity are entered. The lithological correlation of. the geosynclinal with the Cambrian forma- tiens described by McConnell and Walcott on the Canadian Pacifie railway is then discussed. The result is to point to the probability that the geosynclinal at the International Boundary is largely Cambrian in age, though its basal members belong to pre-Olenellus horizons (Beltian of Walcott). Similar correlation with sections described in Montana and Idaho suggests a similar conclusion as to the age of the sediments in the four Boundary series, and it is held that a considerable thickness of the ‘ Belt terrane’ is possibly, if not probably, of Middle and Lower Cambrian age. The chapter closes with an outline of the argument that the eastern half of the Cordillera, from Alaska to Arizona and including the Great Basin of the United States, has been the scene of specially heavy sedimentation during the Beltian, Lower Cambrian, and Middle Cambrian periods. The lower part of the Rocky Mountain Geosynelinal, as defined, has an axial trend faithfully parallel to the main Cordilleran axis of the present day. This geosynclinal suffered a local deformation during an early Middle Cambrian period, and, at the Middle Cembrian Flathead stage, was generally depressed. The area of sedimentation 12 ; DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 was thus enlarged and deposition was generally continuous throughout the Cor- dilleran belt until near the close of the Mississippian. Upon the Paleozoic beds tnick Cretaceous and Tertiary prisms of sediment were locally laid down. Those local geosynclinals and the master Rocky Mountain Geosynclinal compose the Eastern Geosynclinal Belt of the Cordillera. Cuaprer I[X.—Returning to the systematic description of the rocks, the important Purcell Lava formation is here considered. Its characters in the McGillivray, Galton, Clarke, and Lewis ranges are outlined. Certain associated dikes and sills are described and the relation of this fissure-eruption to the thick sills of the Yahk and Moyie ranges is discussed. CHAPTER X.—The intrusive gabbro sills of the Purcell mountain system have already been described in preliminary papers. The matter of these pub- lications, together with some new material, is presented in chapter X. It is largely petrographic. A group of the most important intrusive bodies dis- covered has been given the name, Moyie sills. It illustrates the ability of some very thick magmatic sheets to assimilate their country-rocks—quartzites in the ease of the Moyie sills. The proofs of this are discussed in detail, and similar cares are briefly compared. The principle of gravitative differentiation of magmas is evident in all the cases. CHAPTER XJ.—The sedimentary rocks of the Nelson range, other than those of the Summit series and some others intercalated in the Beaver Mountain voleanie formation, include: the Kitchener quartzite, a small outlier of which seems to be represented along the western edge of the Purcell Trench; the Priest River terrane; and the Pend D’Oreille group. The pre-Cambrian Priest River terrane, the oldest rock-group identified in the Boundary section, is composed of micaceous schists, quartzites, quartz schists, dolomites, and metamorphosed greenstones, arranged in meridional bands, but so complex in structure as to defy all attempts at deciphering their true relation to one another. The petrography of the different bands is des- cribed, and a note is added on the correlation of the terrane with others found in the Cordillera north and south of the Boundary line. The Pend D’Oreille group is divided into the Pend D’Oreille limestone and the Pend D’Oreille schist. As exposed in the Boundary belt, these rocks occur in the batholithic province of West Kootenay, a fact which helps to explain the heavy metamorphism of this group. The limestone is locally unfossiliferous and, with some doubt, is correlated with the similar marbles of definitely Car- boniferous age at Rossland. The schistose division includes phyllite, sheared quartzite, amphibolite, and massive greenstone, which are intimately associated with the limestone and are therefore tentatively referred to the upper Paleozoic. Then follows a brief analysis of the structure of the Nelson range (Selkirk - system). The Purcell Trench is located at the Forty-ninth Parallel on a fault trough representing great vertical displacement. Horizontal shifts and a power- ful overthrust, with rotation of the thrust-plane, are among the more important structural elements in this area of strong deformation. REPORT OF THE CHIE# ASTRONOMER 18 SESSIONAL PAPER No. 25a CuapreR XII.—The Priest River terrane, the formations of the Summit series, and the Pend D’Oreille group are cut by batholiths, stocks, and dikes of igneous rocks of varied nature. Chapter XII is largely devoted to the petro- graphy of these eruptive bodies as exposed in the Selkirk range. In addition, certain minette dikes of the Rossland mountains, being closely related to others occurring in the Selkirks, are described in this chapter. A thoroughly abnormal ‘granite,’ probably a hybrid rock, cuts the Kitchener quartzite at the edge of the Kootenay river alluvium. A tentative correlation of all the formations composing the Selkirk mountains within the Boundary belt is given in tabular form. Cuaprer XIIT.—Though the Rossland mountain group is a small sub- division, the ten-mile belt crossing it shows an extensive variety of formations, chiefly igneous. Fossiliferous Carboniferous limestones, and Cretaceous (?) shales occur near Rossland; and conglomerate bearing fossil leaves (Cretaceous or Tertiary in age) was found on Sophie mountain. The areally important formations include the Rossland and Beaver Mountain groups, (latites, ande- sites, and basalts), the Trail batholith (granodiorite), the Coryell batholith (syenite), the Rossland monzonite, and the stocks of Sheppard granite. A peculiar ‘olivine syenite,’ occurring also in the Bonnington range, and a dike of the rare petrographic type, missourite, are described. The structural and time relations of the formations are discussed. CHAPTER XIV.—Between Christina lake and Midway the bed-rocks form a complex, which is very similar to that in the Rossland mountains. The Chris- tina range is chiefly composed of plutonic rocks, which include the gneissic granite (sheared granodiorite) of the Cascade batholith and the aplitie granite of the Smeiter stock. The origin of the banding in the batholith is briefly dis- cussed and a lateral-secretion hypothesis favoured. Across the north fork of the Kettle river the formations have been studied in detail by Mr. R. W. Brock, from whose report liberal quotations are made. For purposes of convenience in later correlations the present writer gives special names to two of the formations described by Mr. Brock. These new names are: Attwood series, and Phenix Volcanic group. The usual tentative correlation tuble is appended. CuaprerR XV.—Just east of Midway the section enters the Midway volcanic province, representing thick Tertiary lavas and pyroclastic deposits. West of the volcanic mass is a broad band of metamorphosed Paleozoic sediments extend- ing to the Osoyoos batholith. This chapter describes the two provinces, the Midway province demanding the greater detail of statement. The Paleozoic sediments, with included greenstone and basic schists of igneous origin, are riamed the Anarchist series. This series is unfossiliferous, but on lithological grounds, is correlated with the Cache Creek series and other upper Paleozoic groups described north and south of the Boundary. Unconformable upon it is the fossiliferous (Oligocene) Kettle River formation, composed of conglomerates, eendstones, and shales. These sediments are conformably overlain by thick 14 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 masses of basalts and andesites. Younger than any of these formations is an alkaline suite of extrusive and intrusive masses. These intrusions include dikes and irregular injected bodies (chonoliths). Rhomb-porphyry and pulaskite por- phyry are the intrusive types. A less erystalline rhomb-porphyry, alkaline trachyte, and ‘shackanite,’ an analcitic lava, are the extrusive types. Various pre-Tertiary intrusives are also described. : CuHapterR XVI.—The Okanagan Composite Batholith extends from the eastern slope of Osoyoos Lake valley to the Pasayten river, where it is covered by unconformable Cretaceous rocks forming the Pasayten series. The com- ponent batholiths and stocks, with their country-rocks, are described. The whole body is by far the largest continuous mass of plutonic rock in the Evundary belt. The petrographic types represented have wide range of com- pesition, and the dates of eruption vary from late Paleozoic to the late Tertiary or the Pleistocene. A general idea of the order of eruption and nature of the different bodies may be quickly obtained by an inspection of the general table of contents under ‘Chapter XVI’. The reader is also directed to the summary at the end of the chapter itself. Cuapter XVIT.—The Hozomeen range forms a distinct geological provincee,. being principally made up of an extremely thick geosynclinal mass, the Pasayten scriex. Its arkose, conglomerate, sandstone, and shale were deposited in a local, rapidly deepening downwarp of Cretaceous date. An important deposit of andesitic breccia forms the basal member of the series and lies on the eroded surface of the Remmel batholith at the Pasayten river. The Cretaceous rocks ere fossiliferous at various horizons. They compose a faulted and otherwise deformed monocline with westward dips steepening toward the west. At Light- ning creek canyon a great fault brings the youngest Cretaceous beds into contact with the much older Hozomeen series, which is tentatively correlated with the Anarchist, Attwood, Pend D’Oreille, and Cache Creek series. Intrusive bodies with the relations of stocks, dikes, and chonoliths cut the Pasayten series. Special attention is paid to the Castle Peak granodiorite stock, since its structural relations are clearer than those of any other great intrusive mass in the Boundary belt. The evidences of its downward enlargement and of its having replaced or absorbed the Cretaceous sediments are believed to be clear. Ouapter XVIII.—West of the Skagit river, which is located on another master fault, the Hozomeen series is again represented in small patches. On the Pacific slope of the Skagit range a thick body of argillite, sandstone, and limestone, with a heavy mass of interbedded volcanics, is fossiliferous (upper Carboniferous) and under the name Chilliwack series is correlated with the Hozomeen series. So far as known, these are the oldest rocks locally developed in the Skagit range. Fossiliferous Triassic argillite, included in the Cultus formation, was found near Cultus lake. A thick mass of sandstone, etc. to the scuthward is called the Tamihy series. Jt is unfossiliferous as yet, but on lith- ological grounds is correlated with the Cretaceous Pasayten series farther east. REPORT OF THE CHIEF ASTRONOMER 15 SESSIONAL PAPER No. 25a On Sumas mountain coal-bearing, obscurely fossiliferous sandstones and con- elomerates are included in the Huntingdon formation, which is probably equival- ent to the Eocene Puget beds of Washington. A very thick volcanic pile (Oligocene?) occurs on the eastern slope and is called the Skagit Volcanic formation. : Recks assigned to the Hozomeen series are cut by the Custer gneissic bath- clith (sheared granodiorite), outcropping at the summit of the Skagit range. It is possibly of Jurassic age. It is cut by the Tertiary Chilliwack batholith of granodiorite, which is genetically connected with a batholithic mass named the Slesse diorite. Other intrusive masses are also described. The chapter closes with notes on correlation and on the structure of the Skagit range. CHAPTER XIX.—Deals with the correlation of all the bed-rock formations encountered in the Boundary section between the Purcell Trench and the Strait of Georgia, the approximate limits of the Western Geosynclinal Belt at the Boundary line. The correlation of the Forty-ninth Parallel rocks with those described in sections ranging from Alaska to California is then briefly discussed and thrown into tabular form. A summary history of the Western Belt of the Cordillera closes the chapter. CHapTeR XX.—Having described the many formations in the Boundary section, an attempt is here made to summarize the geological history of the Cordillera of the Forty-ninth Parallel. That necessarily brief statement is followed by a note on the theory of mountain-building. CHAPTER XXIJI.—This chapter gives a sketch of the observations made on the glacial geology of the section. The limits of the great Cordilleran ice-cap at the Forty-ninth Parallel, as to ground-plan and depth, are noted. The two double rows of valley glaciers draining the Rocky Mountains and the Cascade system during the Pleistocene are described. The glaciation of each range is then considered, beginning with the Clarke range on the east. The résumé of the chapter is to be found at its closing page. CHAPTER X XIJ.—Discusses certain of the physiographic problems connect- ed with the section. A note on the origin of the master valleys is followed by a division of the Boundary zone into physiographic provinces, listed in a table. A running account of the morphology of the successive provinces, beginning with the Front range synclinal area, is accompanied by a theoretical discussion of the question as to Tertiary peneplanation of the Front ranges and of the Cascade mountains. The cause for the accordance of summit levels in alpine mountains (large extracts from a preliminary paper on that subject) is con- sidered. The chapter closes with a statement of general conclusions on the physiographic development of the Cordillera of the Forty-ninth Parallel. CuapTeR XXIII.—Is a theoretical chapter dealing primarily with the explanation of the fact that fossils are relatively rare in pre-Ordovician forma- tions, and of the related fact that the great majority of those fossils are not caleareous like most of the post-Cambrian fossil remains. The favoured explana- 16 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 tion was given in two preliminary papers and the argument as a whole is here presented for the first time. A summary on this highly complex subject is given in the chapter. The origin of the thousands of feet of limestone and dolomite found in the Rocky Mountain geosynclinal and in the Priest River terrane is attributed to direct chemical precipitation on the floor of the open ocean. Statis- tics show that the limestones of the earlier geological periods were originally more magnesian than those of the later periods. This evolution of the limestones is paralleled with a chemical evolution of the ocean. “CuapTeR XXIV.—Is an introduction to a general theory of the igneous rocks, the statement of which occupies the rest of the report. The Mode classifica- tion is preferred and a table showing the average chemical composition of each rock type is inserted. Magmatic heat in the earth is believed to be chiefly a primitive inheritance, though some of it is due to radioactivity. The argument for a general basaltic magma (perhaps highly rigid at the depth of the sub- stratum) is presented, and is followed by the argument for a primary acid shell at the earth’s surface. All igneous action is preceded by abyssal injection, whereby the basalt of the substratum mechanically displaces the lower part of the earth’s crust and rises to an average level which is at moderate depth below the surface. A note on the essential mechanism of central-eruption voleanoes as distinct from fissure-eruption volcanoes closes the chapter. CHAPTER XX V.—Discuszes the classification of igneous intrusive bodies. The favoured primary division is into injected and subjacent bodies, the former group being largely satellitic to the subjacent masses, which are incomparably the more important as to volume. CuapTeR XX VI.—The genetic problem of the eruptive rocks is, at its heart, also the problem of the batholith. This chapter discusses the processes by which batholiths are believed to have been formed. Their typical field and chemical relations are sketched. The older hypotheses as to the methods of intrusion are compared with the stoping-abyssal injection hypothesis. Abyssal assimila- tion of sunken roof-blocks is a prominent element in the favoured explanation cf batholithic magmas. The chapter is largely a reprint of three preliminary papers, the matter of which is here systematically assembled. CuapteR XXVII.—Considers briefly certain points in the wide subject. of magmatic differentiation. The dominating control of gravity is emphasized. CuaPpteR XXVIII.—The principles stated in the last four chapters are here applied to a genetic classification of magmas, and then to rocks actually found in the Forty-ninth Parallel section. The rock families specially discussed are the granites, granodiorites, diorites, andesites, gabbros, basalts, comple- mentary dikes, pegmatites, and the alkaline types, including the syenites. ‘O[VOS VUUBS OF UMBAIP SUOTIOIS VaLY4 [PV “SULC[ 4va.try oy puew "18.100 4) jo jpne oY} Wooagoq volloury YON JO VAtoT[IPLog oyy Jo qaud oyy puv ‘uLeyo ULALTeTUTPT oy} ‘ULeyo ourdpy ez JO Sfolfot OATZLV[OI SUIMOYS SUOL4DOS O[YOrg aa . ov ououe OG) fory 000 —————EEE— | [DI'P4A9A 7 JO4fuOZN0OY c= = T Say'W OO! . 0S S2 (0) NIVHD NVAVTIVWIH AHL JO NOILITS SSOUD ~ = S x Be) y 5 8 S : : . s S < : N | S aS S _ g 8 : : & a S| S SdTv 3HL JO NOILDIS SSOUD § eg S : Saas 8 OUD] OWW?721 Dye B22 3S ISSO LPC , x TaTIVYVd HININ ALYOS 3HL LY WYSTNGYOD 3HL 4O NOILO3S SSOUD : SNIVLNNOW AM9O0H SFONVY T1FOYNd . 2°Y MYDS gSONVY VIGNININD «5 SNIVLNNOW Zavosvo S RU N 3 d o y W BN A N 8 x : S s S ‘ : : 5 '% HLV Ig YTRANS SUL SUI] Greig DILWI2) 40 1?PVS 25a—vol. ii—p. 16. 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER III. NOMENCLATURE OF THE MOUNTAIN RANGES CROSSED BY THE FORTY-NINTH PARALLEL. INTRODUCTION AND OUTLINE. Although the section covered by the Boundary Commission does not extend to Vancouver island, it is about as long as the longest line of cross-section of the entire Himalayan group of ranges from peninsular India to the Tibetan plateau. If the whole of Vancouver island were included in the Forty-ninth Parallel section, it would be nearly one hundred miles longer than any section-line crossing the Himalayan complex at right angles. Plate 2 shows the reliefs of Himalayas and Alps at their broadest as com- pared with the partial section of the North American chain covered by the Boundary Commission. The great size of the North American chain is further indicated by a comparison of areas. The chain of the Himalayas, using that term in its larger sense, covers about 300,000 square miles; the Alps of Europe from Nice to Vienna, not more than 70,000 square miles; the Andes, about - 1,000,000 square miles; and the western chain of North America, over 2,300,000 square miles. (See also Plate 3.) The vast mountain region crossed by the International Boundary between Canada and the United States has always been very sparsely inhabited. In the eregraphic features it is generally complicated, often to the uttermost. Its exploration is only well begun. There are thus excellent reasons why the moun- ain units of this region are so inadequately named and systematized in geogra- phical works, whether issued as official Government reports, as educational text- beoks or atlases, or as popular records of travel. Yet, whether he will or not, the explorer responsible for a report on any part of this region must confront the question of names. He returns from his rugged field, and, to tell of his findings, must use common nouns to indicate what kinds of land-relief he has found, and proper names to aid in individualizing and locating those reliefs in the huge backbone of the continent. This duty has fallen to the writer in the task of reporting on the geology of the mountains crossed by the Forty-ninth Parallel. Though the same trans- montane section has been described by the geologists attached to the 1857-61 Commission, though it occurs along the most thickly settled part of British Cclumbia, and though it is nowhere very far from the lines of two transcon- tinental railroads, a complete and systematic grouping of the mountains on the Beundary has never been made. The difficulty of supplying the lack was felt by 25a—Vol. ii—2 a 4 18 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 the writer in the first of the six seasons devoted to the geology of the Boundary, but the difficulty was more fully realized as the confusion of the nomenclature already in vogue became apparent. It is manifest that any attempt to develop a constructive view of the | Boundary mountains should be founded as far as possible upon established units already understood and named. The literature has, therefore, been carefully searched to furnish this required foundation. The result has shown a truly surprising variety of usages in names and in concepts of the topography. The course of compilation inevitably led to the study of the nomenclature of western ranges even far away from the Forty-ninth Parallel of latitude. Examples of the differences of usage are recorded in the first part of this chapter. The record may serve in some degree to illustrate the need of a consistent scheme of nomenclature, possibly to suggest partial grounds on which uniformity may some day be established. The second part of this chapter is concerned with a brief account of the nomenclature that seems most appropriate for the ranges crossed by the International Boundary. DIFFERENT NOMENCLATURES IN USE. The search for the variations of nomenclature was made _ both among authorities responsible on the ground of priority and among authorities influential as standard compilers from original sources. For the present purpose of indicating the lack of uniformity and the confusion into which the great mass of the people may be led by consulting existing works of reference, it is not sufficient to record names found only in Government map or careful scientific monograph. Perhaps more important still in this connection is the record to be made from standard atlases, from school geographies, and from standard influential guide-books. In reality, it has required the examin- aticn of but a very limited number of each kind of authoritative works to point the moral. With few exceptions the only works consulted were printed in the English language. DIVERSE NAMING OF THE WESTERN MOUNTAIN REGION AS A WHOLE. The question of the best general title for the western mountains may be con- sidered as trite by those who do not feel the immediate need of its solution in their professional work. The writer by no means believes it to be trite, as he now comple- tely realizes the wide latitude in naming among the recent influential publications dealing with North American geography. It is scarcely to the credit of our geographical societies and alpine clubs that they will publish at length the state- ment of one traveller, that he found mosquitoes in Newfoundland, of- another that his hotel accommodation in Manila was bad, and leave undiscussed the suggestive paper of Prof. Russell and his correspondents on the names of the larger geographical features of North America.* There would be no advantage * Bull. Geog. Society of Philadelphia, 1899, Vol. 2, p. 55. REPORT OF THE CHIEF ASTRONOMER 19 SESSIONAL PAPER No. 25a to the European geographers if the Alps masqueraded under a dozen different general titles dependent on the personal tastes of individual writers on those mountains. t is well known that one of the first designations of the entire mountain group lying between the Pacific and the Great Plains was due to Humboldt. His ‘Cordilleras of the Andes’ extended from Cape Horn to the mouth of the Mackenzie river. Humboldt occasionally used the singular form ‘ Cordillera of the Andes’ for the same concept. In view of the general restriction of the term ‘ Andes’ to the mountains of South America, Whitney, in 1868, proposed that the name ‘ Cordilleras,’ with variants, ‘ Cordilleran System’ and ‘ Cordil- l-ran Region,’ be retained to designate the North American equivalent of the Andes. This name was adopted in the United States census reports for 1870 and 1880, and by a great number of expert geologists and geographers since 1868. in process of time, however, the singular form, ‘Cordillera’ and variants, became used in the same sense. In one of these forms the Humboldt root word with Whitney’s definition has entered many atlases. It appears on numberless pages oi high-class Government reports, geographical, geological, and natural history memoirs, and of such works as Baedeker’s ‘ Guide-book to the United States,’ Stanford’s ‘Compendium of Geography,’ ete. The time-honoured, erroneous, similarly inclusive name ‘ Rocky Mountain,’ with variants, ‘Rocky Mountain System,’ ‘Rocky Mountain Belt,’ etc., has, however, held the dominant place in the popular usage. Its inappropriateness for the heavily wooded Canadian mountains west of the Front ranges is abun- dantly evident. For the United States, Clarence King wrote a generation azo: ‘The greatest looseness prevails in regard to the nomenclature of all the general divisions of the western mountains. For the very system itself there is as yet only a partial acceptance of that general name Cordilleras, which Humboldt applied to the whole series of chains that border the Pacific front of the two Americas. In current literature, geology being no exception, there is an unfortunate tendency to apply the name Rocky moun- tains to the system at large. So loose and meaningless a name is bad enough when restricted to its legitimate region, the eastern bordering chain of the system, but when spread westward over the Great Basin and the Sierra Nevada, it is simply abominable.”* The following table summarizes the above-mentioned variants along with others more recently introduced, and still other general names now only of historical interest. The names of prominent authorities and the leading dates when they have published the respective titles are also entered in the table. The authority for some of the older names is Whitney’s work on the United States, published in Boston, 1889. Mountains of the Bright Stones General use, end of eighteenth century. Shining Mountains Morse, Universal Geography, 1802. *U.S. Geol. Exploration, 40th Parallel, Systematic Geology, 1878, p. 5. 25a—Vol. ii—24 20 Sicney or Stony Mountains Columbians (sic) Mountains Chippewayan Mountains The Cordilleras of the Andes (in part) The Cordillera of the Andes (in part) The Cordilleras The Cordillera The Western Cordillera of North America The Cordilleras of North America The Cordilleran Region The Cordilleran System The Cordillera System The Cordillera Belt The Pacific Cordillera The Cordilleran Plateau The Cordillera of the Rocky Mountains The Rocky Mountain System The Rocky Mountain Region The Rocky Mountain Belt The Rocky Mountains The Pacific Mountains The Western Highland The Rocky Mountain Highland The Western Plateau In most technical writings, DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Arrowsmith, 1795; President Jefferson. Tardieu, 1820. Hinton, 1834. Humboldt, 1808. Humbo!dt, 1808. Whitney, 1868; many authors since. G. M. Dawson, 1884; Gannett, 1898; Rand-MecNally, 1905. J. D. Dana, 1874, 1880. Hayden, 1883; Leconte, 1892. Whitney, 1868; Hayden, 1883; Shaler, 1891. Whitney, 1868; King, 1878; Baedeker, 1893. Hayden, 1883. G. M. Dawson, 1879; Rand-MeNally, 1902. Rrssell, 1899, 1904. Hayden, 1883. J. D. Dana, 1895. Leconte, 1892; Heilprin, 1899; many others. Powell, 1875; G. M. Dawson, 1890; Gannett, 1899. Rand-MecNally, 1902. Lewis and Clarke; popular. Russell, 1899, 1904; Powell, 1899. Raedeker, 1893; Keith Johnston Atlas, 1896; Davis, 1899. T'rye, 1895, 1904. English Imperial Atlas, 1892. of both governmental and private origin, the suggestion of Whitney has been followed with varying fidelity during the last forty-four years. It is clear that the inherent connotation of ‘ Cordilleras’ is different from that of ‘Cordillera.’ The one emphasizes the compound nature of the orographic unit; the other, the singular form of the word, emphasizes the organic union of members. Hayden used both forms of the word. In recent years there has been a rather widespread adoption of the term in the singular number. In 1874, J. D. Dana proposed that the great mountain systems of North America be referred to as the ‘ Western Cordillera’ and the ‘ Eastern Cordillera,’ the latter thus synonymous with what is now commonly called the Appalachian system. Russell, in 1899, proposed ‘Pacific Cordillera’ and ‘Atlantic Cordillera’ with respectively the same significance. Usage has, REPORT OF THE CHIEF ASTRONOMER 21 SESSIONAL PAPER No. 25a however, declared that there is but one Cordillera in North America. The expression ‘ Pacific Cordillera’ is, according to such established usage, redun- dant. ~ The Cordillera of North America,’ ‘The Cordilleran system, ‘ The Cor- dilleran Region,’ or, with the proper context, simply ‘The Cordillera,’ seem to be to-day the best variants on the Humboldt root-word. The fine, dignified quality of the word, convenient in adjective form as in noun form, its unequivocal meaning and its really widespread use in atlas and monograph make ‘Cordillera’ incomparably the best term for technical and even for the more serious popular works. In fact, there seems to be no good reason why the name should not be entered in elementary school atlases. The ebjection that the word is likely to be mispronounced by teacher or scholar would equally exclude ‘Himalaya’ and ‘Appalachian’ from school-books. In teaching or learning what is meant by ‘the Cordillera, the teacher or scholar would incidentally learn so much Spanish. If, in the future, this should be deemed an intolerable nuisance, speakers in English could, in their licensed way, throw the accent back to the second syllable and avoid the unscholarly danger. The second objection that a cordillera is hereby made to include the extensive plateaus of Utah and Arizona or the great intermontane basins of the United States is more serious. It will, however, hardly displace the word from its present technical use as designating a single earth-feature ruggedly moun- tainous as a whole, but bearing subordinate local details of form and structure net truly mountainous. If this objection be regarded as invalid by advanced scientific workers, it will have still less weight for popular or educational use. The ordinary connotation of the term ‘highland’ makes it unsuitable as part of the name indicating the world’s vastest mountain group. Like Powell’s name ‘Stony Mountains,’ suggested for the majestic Front ranges north of the Union Pacific Railroad, ‘highland’ is ‘belittling.’ To most readers it would inevitably suggest Scotland’s relief. If the word be raised to the dignity proposed in ‘ Western Highland’ or ‘Rocky Mountain Highland,’ the writer on the natural features of the Cordillera runs the risk of ambiguity in employ- ing the indispensable common noun ‘highland, while dealing with local pro- blems of geology, geography, or natural history. For popular use, the best title alternative with ‘Cordillera’ is, in the writer’s opinion, ‘ The Pacific Mountain System.’ It is suggested by Russell’s ‘The Pacific Mountains. The addition of the world ‘system’ ..ems advisable ax stating the unity of the whole group. The proposal of J. D. Dana to restrict the common noun ‘system’ to mean merely the group of ranges formed in a sit.gle geosyncline has to face overwhelming objections. The usage of genera- tions is against it; the difficulty of actually applying it in nature is, perhaps, yet more surely fatal to the idea. The restriction of the titles ‘Pacific Ranges’ (Hayden), ‘ Pacific Moun- tains’ (Powell in his earlier use of that term; he later applied it to the whole Cordillera), and ‘ Pacific Mountain System’ (A. ©. Spencer and A. I. Brooks) to the relatively narrow mountain belt lying between the ocean and the so-ealled ‘Interior Plateau’ of the Cordillera, seems particularly unfortunate. 22 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 {i there is one grand generalization poszible about the entire Cordillera, it is that the Cordillera is, both genetically and geographically, a Pacific feature of the globe. The Rocky Mountain ranges proper, the Selkirks, and the Bitterroots bear the marks of interaction of Pacific basin and continental plateau as plainly as do the Sierra Nevada, the Coast ranges, or the St. Elias range. The large view of the Cordillera assuredly claims the word ‘ Pacific’ for its own, and eannot allow in logic that ‘Pacific Mountain System’ shall mean anything less than the entire group of mountains. The artificial nature of the narrower definition would be equally manifest if it were applied to a topographic or genetic unit forming a relatively small part of the Andes along the immediate shcre-line of South America. The Andes mountains form the Pacific mountain system of South America as the whole North American Cordillera forms the true Pacific mountain system of North America. Yet the term ‘system’ is itself so elastic that it is fitly applied to a sub- division of the Cordillera. For example, the Rocky Mountain System expresses an unusually convenient grouping of the northern ranges in Alaska, and of the eastern ranges of the Cordillera in Canada and the United States. Popular, as well as scientific, usage has. once for all recognized the propriety of there being in name, as well as in fact, system within system in the grouping of mountain:. DIVERSE NAMING OF RANGES CROSSED BY THE FORTY-NINTH PARALLEL. There is a double difficulty in describing the mountains along the Inter- national Boundary. The same range may bear different names with different authorities, or may be differently delimited by different authorities. Some examples chosen from recent atlases and texts will illustrate this point. 1. Cascade range (also called Cascade chain or Cascade mountain chain), according to different authorities :— (a) Extends from Mount Shasta into the Yukon territory; (6) Extends from Mount Shasta to the British Columbia boundary; (c) Extends from Mount Shasta to the Fraser river, and east of it to the Thompson river; (d) Forms the extreme northern part of the British Columbia Coast range north of Lynn canal, the real Cascades being mapped as the ‘Coast Range’ (Johnson’s Cyclopedia). ; 2. Coast range of British Columbia, also called the ‘ Alpes de Colombie’ (Atlas Vidal-Lablache) and ‘See Alpen’ (Stieler’s Handatlas, which con- tinues the ‘ Caseaden Kette’ across the Fraser-river). See also usages under ‘Cascade Range.’ 3. Selkirk mountains, according to different authorities :— (a) Lie west of Kootenay lake, entirely in Canada, or extending into the United States; (b) Lie west of Kootenay lake, and entirely in Canada, or extending into the United States; REPORT OF THE CHIEF ASTRONOMER 23 SESSIONAL PAPER No. 25a (c) Extend on both sides of Kootenay lake, but entirely in Canada; (d) Do not extend south of the northern extremity of Kootenay lake; (e) Contrary to all of the above-mentioned usages, extend across the - Columbia river north-westward to Quesnel lake in 53° N. lat. (Brownlee’s Map, 1893). 4, Purcell range, according to different authorities— (a) A local rangelet in the West Kootenay district, British Columbia; (6) Includes all the mountains between Kootenay lake and the Rocky Mountains proper, entirely in Canada; (c) Includes the same mountains as under (b), but extends into the United States as far as the great loop of the Kootenay river. 5. Bitterroot mountains (also spelled ‘ Bitter Root’) used— (a) In the larger sense of most maps; or (b) In a much narrower sense, a small range overlooking the Bitterroot river (Lindgren). 6. Rocky Mountains or Rocky Mountain system, also called the Front range, and Laramide range; often alternative for ‘ Cordillera.’ 7. Gold range of British Columbia, a name applied to a local range crossed by the main line of the Canadian Pacific railway, and west of the Columbia river; also applied to a much greater group, including the Selkirk, Purcell, Columbia, Cariboo, and Omineca ranges (Gold ranges, an alternative furm of the title in this latter meaning). The confusion of the nomenclature is aggravated in the case of certain atlases, which in different map-sheets give different titles to the same range. Thus, on one map of the new Rand-McNally ‘ Indexed Atlas of the World,’ the western mainland member of the Cordillera in British Columbia is correctly named the Coast range and, on another skeet, incorrectly named the Cascade range. The same indefensible carelessness even appears in certain Canadian school atlases. In the Rand-McNally map of British Columbia, the Selkirks are represented as ending on the south at the head of Kootenay lake, and are continued to the eastward of that lake by the ‘ Dog Tooth Moun- tains, the latter name being little familiar to the people of British Columbia. In the genéral map of the United States published in the same Atlas, the Selkirks are represented as quite defined to the westward of Kootenay lake. The area thus inconsistently mapped has a width equal to the average width of the Alps. ADOPTED PRINCIPLE OF NOMENCLATURE FOR THE BOUNDARY MOUNTAINS. On the line of the Forty-ninth Parallel, the Cordillera has already assumed what may be called its British Columbia habit as contrasted with its Fortieth Paral- lel habit. The division of the whole into orographic units is relatively simple in Colorado, Utah, Nevada, and California, where the building and erosion of the Cordillera have resulted in a comparatively clear-cut separation of the 24 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 component ranges by broad intermontane plains of mountain waste, or of lava filling vast structural troughs or basins. Nothing quite comparable is to be seen anywhere in the Canadian portion of the Cordillera. Near the latitude of Spokane, the whole mighty group of ranges is marshalled into a solid phalanx of closely set mountains which sweep on in substantial unity north-westward through Yukon territory into Alaska. The area of British Columbia alone would enclose twenty-four Switzerlands. For purposes of exposition this moun- tain sea must be divided and subdivided. How shall it be done? The remarkable insight and generalizing power of the pioneer in British Columbia geology, G. M. Dawson, early supplied what seem to be the only fruitful principles. His classification applies chiefly to southern British Col- umbia, but it is probable that its principles must be extended throughout the Canadian Cordillera. In 1879 Dawson announced the possibility of a natural division of the mountains between the Forty-ninth and Fifty-fifth parallels into three broad belts paralleling the coast. The middle belt, the ‘ Interior Plateau,’ afterwards described in some detail, has the special style of topography characteristic of closely folded mountains once reduced by denudation to mere rolling hills or an imperfect plain, since uplifted and cut to pieces by streams. In other words, the Interior Plateau is, by Dawson’s definition, an uplifted, dissected peneplain, a region of plateaus and hills remnant from the old surface of denudation. Yet Dawson himself concluded that, while many of these tabular reliefs may be correlated into the ancient facet of denudation, other similar reliefs in the belt are structural, and due, namely, to the erosion of wide, flat-lying lava flows that flooded the country after the peneplanation. Another and simpler explanation of the topo- graphy makes the lava flooding anterior to peneplanation. Still, a third history may, on further investigation, turn out to be the true one. At the present time it is impossible to decide between the rival views. A safer definition of the region is purely topographic; it may thus be called the Belt of Interior Plateaus, or, briefly, the Interior Plateaus. (Plate 3.) This slight change in Dawson’s name lays stress upon the individual tabular reliefs so characteristic of the region. These reliefs are facts; the peneplain and the involved assumption that the myriad individual reliefs belong to a physio- graphic unit, a single uplifted peneplain, are matters of theory. The pluralizing of the word ‘plateau’ in the title not only changes the emphasis, but, in so doing, restores the term to its more advisable definition of a tabular relief bounded by strong downward slopes. The Interior Plateau as defined by Daw- son is bounded on all sides by the strong upward slopes of the enveloping mountain ranges. The belt of Interior Plateaus having thus been differentiated on special grounds, we may pass to the subdivision of the remaining two parts of the British Columbia complex. Those two belts separated by the plateau belt are rugged, often alpine, and, as a rule, do not show tabular reliefs. Present knowledge of the vast field cannot provide a rational treatment of these moun- tains rigidly on the basis of either rock composition or structural axes or ” “Frage PACIFIC OCEAN “COLUMBIA /LAVA P au Sketch map showing subdivision of the Cordillera at the Forty-ninth Parallel. HEPART MEN: Pee ae ry Wa tory REPORT OF THE CHIEF ASTRONOMER 25 SESSIONAL PAPER No. 25a geological history. It is possible, if not indeed probable, that the ranges imme- diately bounding the belt of Interior Plateaus have had a common history with it; they certainly include the same rock formations as occur in the interior plateaus. Jf the peneplain theory be finally accepted for the latter, it may ultimately prove best to treat the Coast range and other ranges in terms of the same theory. The only feasible scheme of subdivision at the present day must be based on topography only. Mere hypsometry will not serve alone; the ranges of summit altitudes is too slight, their variation too unsystematie, Bigs that. Dawson found that continuity of crest-lines and the position of the greater erosion valleys formed the most available basis of classification. As field work progresses in British Columbia, it becomes more and more certain that this double principle is the best that could be devised for present use. Many of the larger valleys are undoubtedly located on structural breaks, but it is evident that the strength of most of the valleys is the more direct result of fluviatile and glacial erosion. Owing to the peculiarly complicated rearrangements in the drainage of the Cordillera, whether due to glacial, voleanic, or crustal disturbances, or to spontaneous river adjustments, the valleys of British Columbia are in size very often quite out of relation to their respective streams. For example, the longest depression in the whole Cordillera is occupied by relatively small streams, the headwaters of the Kootenay, Columbia, Fraser, ete. Each of the rivers named, in its powerful lower course, flows through narrow canyons. Erosion-troughs rather than rivers have, therefore, been selected by Dawson and other explorers as the natural lines of demarcation between most of the constituent ranges of the Cordillera in these latitudes. The procedure is not new, but it is noteworthy as the most wholesale application of the principle on record. It stands in contrast to the more structural treatment, not only possible, but enforced by the orographic conditions in the United States. In the course of his own work, the writer has become convinced of the permanent value of Dawson’s early and consistently held general view of the British Columbia mountains. But there has arisen the necessity of extending it to cover the Boundary mountains which, for the most part unvisited, were left unnamed by Dawson. The task of systematizing them is simple only in the stretch from the Great Plains to the Kootenay river at Tobacco Plains, a width of about seventy-five miles. The remainder, or five-sixths, of the Cor- dillera on the international line is not generally grouped into organic units at all; or, where so grouped, the names of the groups are not universally accepted. In attempting to supply this lack of system, the writer’s aim has been to develop a system of grouping and nomenclature largely founded on names and concepts already in use, but not generally applied to the mountains so far south as the Boundary. TRENCHES AND GREATER VALLEYS. A point of departure is readily found. Within the Cordillera on the Forty-ninth Parallel, there are four principal longitudinal valleys which serve as convenient lateral boundaries for leading members of the system. 26 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 (Plate 3.) The whole valley occupied at the Boundary by the Kootenay river is the easternmost and much the longest. It is a part of a single Cordilleran feature easily the most useful in delimiting the Canadian ranges. From Flathead lake to the Liard river, a distance of about 800 miles, this feature has the form of a narrow, wonderfully straight depression lying between the Rocky Mountain system and all the rest of the Cordillera. Unique among all the mountain-features of the globe for its remarkable persistence, this depression is in turn occupied by the headwaters of the Flathead, Kootenay, Columbia,.Canoe, Fraser, Parsnip, Finlay, and Kachika rivers, and is therefore not fairly to be called a valley. It may for present purposes be referred to as the ‘Rocky Mountain Trench. The term ‘trench’ throughout this report means a long, narrow, intermontane depression occupied by two or more streams (whether expanded into lakes or not) alternately draining the depression in opposite directions. An analogy is found in a military trench run through a hilly country. (See Plate 4.) The first-rank valley next in order on the west is also occupied at the Boundary by the Kootenay river, returning into Canada from its great bend at Jennings, Montana. This valley begins on the south near Bonner’s Ferry, Idaho, and is continued north of Kootenay lake by the valley of the Duncan river. Recently, Wheeler has shown that the singular 40-mile trough occupied by Beaver river, which enters the Columbia river at the Canadian Pacific rail- way, is precisely en axe with the Duncan river valley.* The whole string of valleys from Bonner’s Ferry to the mouth of the Beaver, a distance of approxi- mately 200 miles, forms a topographic unit that may be called the ‘ Purcell Trench.’ (Plate 5.) The third of the first-rank valleys is drained southward by the Columbia river, expanded upstream to form the long Arrow lakes. At its northern extrem- ity near the Fifty-second Parallel of latitude, this valley is confluent with the Rocky Mountain Trench. The southern termination of the valley regarded as a primary limit for these mountain ranges occurs about sixty miles south of the Forty-ninth Parallel, where the Columbia enters the vast lava plain of Wash- ington. To distinguish this orographiec part of the whole Columbia valley between the points just defined, it may be called the ‘ Selkirk Valley.’ A glance at the map will show that the two primary trenches and the Selkirk Valley are in simple mnemonic relation to three principal mountain divisions of the Cordillera. They lie respectively to the westward of the Rocky Mountain system, the Purcell range, and the Selkirk system. The fourth of the first-rank valleys carries the south-flowing Okanagan river, with its various upstream expansions, including Osoyoos and Okanagan lakes. The latter lies wholly within the belt of Interior Plateaus, a primary Cordilleran division. Important as Okanagan lake is, no one has yet suggested that the plateau belt itself be subdivided into named portions separated by the lake. It appears, on the other hand, wiser to recognize in the nomenclature * A. O. Wheeler, The Selkirk Range, Gov’t Printing Burean, Ottawa, map in Vol. 2, 1905. + YSvlSs ‘p ALVIg ‘punoasyovq ut osuey Ava TDW OY} UL O[QISIA deATy ABUojyooy “ABMoyex) Ye YOU Ip, ulequnoyy,2 AYooY 9} ssoroe yseys A AI® P 164 nog .126. . 1i—p 25a—vol ‘IDALI AY} JO [OAST OAOGV ‘YSIY Jeoy PUVSNOY] IMO] 07 doIYy WA, pUY YULASTP SoTL OIIT ynoqe SL YIO UY] “YootD usloH avou vypop tary Avusgooy jo ospo W19jsaM WoT] ‘OUST, [[ooIN Yq Sso10V 4Sva SULOOTT 'G ALVIG Fg an es REPORT OF THE CHIEF ASTRONOMER 27 SESSIONAL PAPER No. 25a the essential unity of the belt. The southern portion of the Okanagan valley stretching from the mouth of the Similkameen river to the confluence with the Columbia, has, however, a decided function in separating the Cascade range from the very different mountains east of the Okanagan river. This portion may be called the Lower Okanagan valley. SUBDIVISION OF THE ROCKY MOUNTAIN SYSTEM. The Rocky Mountain system, where it crosses the Forty-ninth Parallel, is very definitely bounded: on the east, by the great plains; on the west, by the Rocky Mountain Trench. (Plate 4.) This great element of the Cordillera is itself so vast that, for the purpose of presenting the facts of its stratigraphy and general geological history, the system must be subdivided into convenient units. By a kind of international co-operation this is being accomplished. In Dawson’s reconnaissance map of the Rockies, published in 1886, he designates as the ‘ Livingstone Range’ the long Front range stretching from the Highwood river at 50° 25’ N. Lat. southerly to the North Kootenay Pass at 49° 35’ N. Lat. On the west it is bounded, for many miles, by the straight valley of the Livingstone river and, in general, by the low mountainous area covered by the Crowsnest Cretaceous trough. The name had appeared in Arrowsmith’s map of 1862, and in Palliser’s of 1863, but Dawson gave the range its first definition.* Sixteen years later Willis made his admirable recon- naissance of a part of northwestern Montana and proposed that the ‘ Living- ston Range’ be considered as extending across the International Boundary as far as Lake McDonald.t There are, however, certain oblecuions to making this change of definition. These may be briefly stated. The crests of the Livingstone range, as delimited by Darou are composed almost entirely of Devono-Carboniferous limestones. Midway in the range- axis these rocks are interrupted, for a distance of about two miles, by a trans- verse band of Cretaceous beds, but this local variation in geologic structure inyolyes no marked break in the line of crests. On the other hand, Dawson’s map and accompanying text indicate clearly that the range unit ends a few miles north of the North Kootenay Pass. At that point a broad area uf Cre- taceous rocks squarely truncates the Devono-Carboniferous limestone and forms comparatively low mountains of the foothills type. The independent rangelet of which Turtle mountain is a part, is also composed of the Devono- Carboniferous limestone and is in a similar manner cut off by the zone of Cretaceous hills. The zone is fully twelve miles broad on the line of the axis of the Livingstone range as mapped by Dawson. On the south of the zone, lofty mountains of the Front range type are again to be found and these con- tinue in strength to and beyond the International Boundary. The rocks com- posing these mountains south of the broad, transverse Cretaceous belt are, however, not of Devono-Carboniferous age but belong to a much older Cambrian series underlain by conformable pre-Cambrian strata. * Annual Report, Geol. Survey of Canada, 1885, one B, p. 80. + Bull. Geol. Soc. America, Vol. 13, 1902, p- 312 28 DEPARTMENT OF THE INTERIOR 2° GEORGE, V., An tote It is thus seen that the Cretaceous zone at the North Kootenay Pass makes a complete structural and topographic break in the Front range of the Rockies. To the north of the zone Dawson’s Livingstone range forms a well-defined unit, its summits being composed of the later Paleozoic limestone. To the south of the zone the Front range, also rugged and in strong topographic contrast to the Cretaceous hills, is essentially composed of quartzites, argillites, and magnesian limestones of pre-Cambrian and earliest Paleozoic age. It seems, therefore, inappropriate to extend the Livingstone range any farther south than the North Kootenay Pass. From that Pass south to McDonald lake in Montana the great range lying between the Flathead and the Great Plains in Canada and between the Flathead and the Lewis range on the American side of the International Boundary, needs a special name. Such a name has not hitherto been suggested. To supply the need the title ‘Clarke range’ may be proposed. The name is taken from that of the colleague of Meriwether Lewis who led the famous Lewis and Clarke expedition into the region in 1806. This splendid range is worthy of the able explorer and his memory is worthy of the range. The new designation for these mountains is in simple mnemonic relation to the name of the adjacent Lewis range, a name which is officially recognized by the United States Geological Survey.* (See Figure 1.) : After a review of the topographic and geologic relations Mr. Willis has expressed his own belief that the proposed change of nomenclature is advisable. In a letter to the writer he states: ‘T took the name of Livingstone range from a Canadian map with- out particularly investigating the topography north of the boundary. It sufficed for my study at the time to know that there was a range in the United States which was in ik with one called the Livingstone range in Canada. Hane ‘Your proposition to give a distinct name to the range in the United States is, I think, fully justified, and the one you select is a most happy counterpart to the name Lewis. I should be glad to have you publish the nomenclature as you suggest, namely, giving to the range west of the Lewis range, from McDonald lake northward to the Kootenay Pass, the name of Clarke range.’ On the Canadian side of the Boundary for a distance of thirty miles the Clarke range is the Front range. Just north of the Boundary line it runs behind, to the westward of. the equally important member of the system, called the ‘Lewis range’ by Willis. At the Forty-ninth Parallel the wide valley occupied by Waterton lake and its affluent, Little Kootna creek, forms a definite boundary between the Clarke and Lewis ranges, which, further south, are separated by the head-waters of McDonald creek. According to Willis, the Lewis range extends southeastward almost to latitude 46° 45’. On the north it ends in Sheep mountain, a couple of miles beyond the spaleea iro! * See Chief Mountain sheet of the Mopopraphic Atlas, U.S. Geol. Survey. M4 /15* REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a Yakin chak Creer Lo iN 98°40 Diagrammatic map showing subdivision of the Rocky Mountain System at the Forty-ninth Pavallel. Figure 1. 30 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 line. Dawson gave the name ‘ Wilson range’ to a limited group of mountains in which Sheep mountain occurs. However, the title ‘Lewis range’ is to be a permanent feature of geographical nomenclature in Montana and must include the Wilson range, which is but a part of a whole first recognized in the scientific exploration of Montana. (Figure 1.) West of the Flathead river and east of the Kootenay river, Dawson (follow- ing Palliser) recognized two distinct ranges as including the mountains along the Forty-ninth Parallel. On his 1886 map the more easterly range bears the name ‘MacDonald range’, the other bearing the name ‘ Galton range.’ These ranges are separated, for a part of their length, by the straight valley of Wig- wam river. Willis appreciated the undoubted fact that the Galton range continues, with relatively unbroken crest-continuity far to the south of the Boundary line. In his 1902 map of northwest Montana, this range is repre- sented as extending to the main Flathead river at Columbia Falls, the south- western and western limit being fixed at the valleys of Stillwater creek, Tobacco river, and Kootenay river; and the northeastern limit in Montana being fixed at the valley of the North Fork of the Flathead river. Between the North Fork and the Wigwam the mountains are not named on Willis’ map, but, apparently, were considered by him to belong to Dawson’s ‘MacDonald range.’ In this view the MacDonald range is limited on the south by the strong trans- verse valley of Yakinikak creek. According to Dawson’s map the northern limit of the Galton range seems to have been fixed at the Elk river and the northern limit of the MacDonald range at the Cretaceous area along the North Kootenay Pass. Combining the views of Dawson and Willis we have a convenient subdivi- sion of the western half of the Rocky Mountain system at the Forty-ninth Parallel into the two ranges, the Galton and the MacDonald, each of which, according to the law of crest-continuity, is a fairly distinct unit. The sketch-map of Figure 1 illustrates the conclusions reached by the writer as to the most desirable topographic subdivision of this part of the Rocky Mountain system. It is very possible that further mapping of the region may show the necessity of modifying this orographic scheme. In its present form it will be found useful for the purposes of this report and seems to have the advantage of meeting the views of the few trained observers who have pene- trated these mountains. PURCELL MOUNTAIN SYSTEM AND ITS SUBDIVISION. Westward from Tobacco Plains, on the Forty-ninth Parallel, we cross, in the air-line, sixty miles of ridges belonging to a range unit which is almost as systematic as the great group on the east. (Plate 4 and Figure 2.) The crests of this second group are in unbroken continuity from the wide southern loop of the Kootenay river at Jennings to the angle where the Purcell Trench is confluent with the Rocky Mountain Trench. Throughout this area the drainage is quite evenly divided by the easterly and westerly facing slopes of the unit- REPORT OF THE CHIEF ASTRONOMER 31 SESSIONAL PAPER No. 25a Tle as View U5° SELL AIAPH RANG FLATHEHED PANGE 48% FicuRE 2.—Diagrammatic map showing subdivision of the Purcell Mountain System at the Forty-ninth Parallel. relief. This strong and extensive range has, in its northern part, been gene- rally regarded as part of the Selkirk Mountain group. The middle and southern part, though broader and including most of the area, has, as a whole, never been authoritatively placed in the Selkirk system. Palliser gave the name ‘Purcell Range’ to a single component of the unit, namely, the group of summits lying between Findlay creek and St. Mary river. Dawson extended the name to cover all the mountains between Kootenay lake and the Rocky 420 32 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Mountain Trench, these mountains forming the ‘ Purcell division of the Selkirk system’; but he did not fix either a northern or a southern limit to the group so named. The same usage appears in the maps and texts of most geographers pub- lishing during the last twenty-five years. It was officially adopted by the Canadian Geological Survey, and by the British Columbia Government (1897). It appears in the general geological map of the Dominion, edited by Selwyn and Dawson, and issued by the survey in 1884. The name was accordingly entered in most of the American and European atlases of the world. For some unknown reason, the second edition of the official geological map of the Domin- ion (1901) represents the Purcells as constituting merely Palliser’s original small group of summits, and this tradition has been followed in the new general map of the Dominion issued by the Canadian Department of the Interior (1902). Both official and general previous usages conflict with this quite recent official return to Palliser’s mapping. In reality, the Palliser usage is not familiar to the people of British Columbia; it is subject to the criticism that the rangelet mapped by Palliser is not defined on the west by natural limits. The lack of definition in Palliser’s exploratory sketch-mapping is such that it may even be doubted that Dawson really broke the law of priority in giving ‘Purcell range’ its broader meaning. The name is practically useless if it be not so extended. The long-established tradition of the influential atlases following the lead of Dawson makes it expedient to use the title in the broader meaning. The question remains as to the northern and southern limits of the Purcell range. As a result of compiling all the available information, the writer has concluded that the range has no natural boundary to the northward, short of the confluence of the Purcell and Rocky Mountain trenches. The conclusion has been strikingly corroborated by the detailed studies of Wheeler along Beaver river. There is, similarly, no natural boundary on the south, short of the great bend of the Kootenay river in Montana. However vaguely supported by definite knowledge of the field, the latter conclusion has been anticipated by the editors of the Century Dictionary Atlas (map of Montana), of the Eneyclopedia Americana (maps of British Columbia, Montana, and Canada), of Bartholomew’s English Imperial Atlas, of Keith Johnson’s Royal Atlas, and of Stieler’s Handatlas. Maps occurring in all of these works represent the Purcell range as continuing southward into the United States as far as the Kootenay river. So far as known to the writer, there is no popular or official designation for the mountains lying between that river and the Canadian Boundary. The Cabinet mountains lie entirely south of the Kootenay river. The first attempt on the part of the United States Geological Survey to name, in published form, the natural subdivisions of this extensive group of mountains was made in 1906. In Bulletin No. 285, published in that year and bearing the title ‘ Contributions to Economie Geology, 1905,’ an outline man of northern Idaho and northwestern Montana was issued in connection with Mr. D. F. MacDonald’s report on mineral resources of the district (page 42). On this | REPORT OF THE CHIEF ASTRONOMER 33 SESSIONAL PAPER No. 25a map all the area enclosed between the International line and the Kootenay river as it swings through the great bend between Gateway and Porthill is shown as occupied by the ‘Loop mountains.’ That subdivision lying to the west of the Moyie river is mapped as the ‘ Moyie range’; a middle subdivision lying between the Moyie river and the Yahk river is mapped as the ‘Yaak range’; an eastern division, lying to the eastward of the Yahk river is mapped as the ‘Purcell range.’ No discussion of this scheme of nomenclature is given in MacDonald’s paper, which was apparently written about the time when the preliminary paper of the present writer was in preparation. The name for the eastern subdivision of the Loop mountains was evidently given in the belief that the local Purcell range, as mapped by Palliser and Dawson, should be extended southward across the Boundary. A serious objection to this proposal is that the unit mapped by Palliser as the ‘ Purcell range’ is, at the south, cut sharply off by the strong transverse valley of St. Mary river and by the wide plains about Cranbrook, nearly forty miles north of the Boundary line. If, then, it were thought expedient to limit the name ‘ Purcell’ to an elementary range unit, as suggested though not enforced in Palliser’s map, it is hardly possible to carry the Purcell range south of St. Mary river. On the other hand, we lave seen that some official usage and the usage of several influential atlases have familiarized us with the idea of giving the old name ‘ Purcell range’ to the entire mountain group occupying the area between the Rocky Mountain Trench and the Purcell Trench. This view implies that the rangelet limited on the east by the local mountain wall seen by Palliser as he looked across the Rocky Mountain Trench and mapped as belonging to the ‘ Pureell range,’ should receive a special definition and a special name as soon as its extent as an orographic individual is known through actual mapping. The general name ‘ Loop mountains’ was presumably suggested by the loop of the Kootenay river, which bounds the whole group on the south. This great bend in the river is so remarkable a feature that the name is certainly appropriate on the United States side of the Boundary line. It is, however, true that four-fifths of the area and five-sixths of the length of the orographic unit involved, lie to the north of the Boundary and in no immediate relation to the bend of the Kootenay. For the greater part of the unit the name ‘ Loop mountains’ is not appropriate. It is clear that the political boundary should, ideally, have no influence in fixing the general name. Systematic orography, supplemented by priority of usage, seem to declare for the older general name “Purcell range’ for the mountains considered, whether north or south of the line.* In summary, then, the great range unit here called the Purcell range is bounded by the Rocky Mountain Trench, the Purcell Trench, and the portion ef the Kootenay valley stretching from Jennings, Montana, to Bonner’s Ferry, Idaho. * Since the last paragraphs were written, Calkins has published Bulletin 384 of the United States Geological Survey, in which (Plate I) the “Loop mountains” are re-named the “Purcell mountains.” 25a—3 34 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 In the present report MacDonald’s name ‘ Moyie range’ will be used to include all the mountains bounded by the Purcell Trench and by the strong valleys of the Moyie and Goat rivers. Similarly the name ‘ Yahk range’ will be used with limits as follows: on the west and north by the Moyie river; on the south, by the Kootenay; on the east, by the Yahk river from source to mouth. The largest subdivision, the eastern one, will here be called the Mc- Gillivray range, a title taken from one of the earliest names of the Kootenay river.* This range is bounded on the east by the Rocky Mountain Trench; on the south, by the loop of the Kootenay river; on the west, by the Yahk river and the Moyie lakes; on the north, by the Cranbrook plains. (Plate 6.) This three-membered part of the Purcell system is there marked off by two huge trenches and by deep and wide transverse notches, faithfully followed by the two transmontane railroads, the Canadian Pacific and the Great Northern. (Figure 2.) SELKIRK MOUNTAIN SYSTEM AND ITS SUBDIVISION. The Selkirk Mountain system next on the west likewise forms a range unit considerably longer than the area generally ascribed to the Selkirk group. (Plate 3.) On principles similar to those adopted for the Purcell range, the Selkirk system may be defined as bounded on the east by the Purcell Trench; on the north and northeast by a portion of the Rocky Mountain Trench; on the west by the Selkirk Valley; on the south by the Columbia lava plain, Pend D’Oreille lake, and a short unnamed trench extending from that lake to the Purcell Trench at Bonner’s Ferry. For a short stretch the Selkirk system is apparently confluent with the Coeur D’Alene mountains, though a short trench followed by the Great Northern railway may separate them. This extension of the Selkirks across the Boundary has already been indicated on maps of the Encyclopedia Americana, Stieler’s Handatlas, and the Vidal-Lablache atlas. The whole mountain complex embracing the Purcell range and Selkirk system, as just defined, may be viewed in another way. The Purcell range is thereby considered as part of the Selkirk system, and that division of the whole lying to the westward of the Purcell Trench, might be called the Selkirk range. The Selkirk system would thus inelude the Selkirk range and the Purcell range. As already noted, Dawson seems to have adopted this alternative view. An objection to it is the chance for confusion in using ‘ Selkirk’ to mean now a component range, now the inclusive system. In favour of Dawson’s view is the fact that in rock composition, structural axes, and geological history, the mountains lying between the Rocky Mountain Trench and the Selkirk Valley form part of a natural unit. On the other hand, the Selkirk range is, struc- turally and lithologically, as closely allied to the Columbia system as to the Purcell range; the Purcell range is, lithologically and historically, as closely allied to the Rocky Mountain system as to the Selkirk range. The practicable orographic classification, being based upon erosion troughs, recognizes the *In J. Arrowsmith’s map of British Columbia in British Government Sessional Papers relative to the affairs of British Columbia 1859. REPORT OF THE CHIEF ASTRONOMER 35 SESSIONAL PAPER No. 25a dominant importance of the Purcell Trench. That superb feature of the Cor- Gillera cleaves the mountains in so thoroughgoing a manner that a logical gTouping must regard the Purcell range as a member co-ordinate with the Selkirk range. In the map the latter division is called the Selkirk system, because it includes subordinate ranges. If, for purposes of exposition, this compre- hensive character is not fixed for emphasis, the same Cordilleran member may be called the ‘Selkirk range.’ Similarly, when the Purcell range is, in the future, subdivided into its orographic units, it may bear the name ‘ Purcell system.’ ‘Cascade range’ and ‘Cascade system,’ ‘Coast range’ and ‘ Coast system,’ for example, may be profitably employed with the same distinctions. In all these cases it is a matter of emphasis. The value of this distinction in common nouns, the great orographic significance of the Purcell Trench, and the weight of much authority in previous usages have caused the writer to suggest that the whole Purcell range be con- sidered as co-ordinate with, and not part of, the Selkirk system. No systematic subdivision of the system has ever been attempted. In discussing the geology of the system at the Boundary line there will be found to be much advantage in recognizing its subdivision into units of more con- venient size. A tentative scheme will therefore be proposed. Just north of the Forty-ninth Parallel a strong, though subordinate trench runs meridionally along the middle part of the system. This trench is occupied by the main Salmon river and by Cottonwood creek, which enters the West Arm of Kootenay lake at Nelson. It divides the system into two broad ranges, both of which are cut off on the north by the transverse valley enclosing the West Arm and the outletting Kootenay river. The eastern range, for which the name ‘ Nelson range’ is proposed (from the name of the chief town of the district), is bounded on the east by the Purcell Trench and on the south by a trench occupied by Boundary creek, Monk creek, and the South Fork of the Salmon river. The western range may be called the ‘ Bonnington range,’ from the well-known falls which break the current of the Kootenay river. The southern limit of this range is the Pend D’Oreille valley; the western limit, the Selkirk Valley. (Figure 3.) In the preliminary paper the Slocan mountain group was stated to be ‘separated off definitely by the Slocan Trench, which is a longitudinal depres- sion occupied by Slocan river, Slocan lake, and the creek valley mouthing at Nakusp, on Arrow lake.’ The definition was framed partly on the ground that this mountain group includes the valley of Little Slocan river. On maturer consideration the writer wishes to recall this definition and to propose the name, ‘Slocan mountains’ for the group east of Slocan river and Slocan lake. The group west of the Slocan valley should probably have the name ‘ Valhalla mountains,’ which was entered by Dawson, in 1890, on his ‘ Reconnaissance map of a portion of the West Kootanie District, British Columbia,’ as the title for the complex of high peaks west of Slocan lake. 25a—34 36 DEPARTMENT OF THE INTERIOR 1 GEORGE V, A. 1911 BONNINGTON FALLS | S re) 18°00! 117°30! 117%00° 11630 Figure 3.—Diagrammatic map showing subdivision of the Selkirk Mountain Syst.m at the - Forty-ninth Parallel. REPORT OF THE CHIEF ASTRONOMER 37 SESSIONAL PAPER No. 25a The mountain group lying southwest of the Pend D’Oreille river was called in the preliminary paper, the Pend D’Oreille mountains. It may further be proposed that the two groups separated by Priest River valley be named the Kaniksu range (on the west) and the Priest range (on the east). ‘Kaniksu’ is the old Indian name for Priest lake. Though these names may not prove finally satisfactory, the writer believes that the naming of these groups in an authori- tative and systematic manner would be a geographic gain. In passing, the question may be raised as to the advisability of regarding the mountains lying between Priest river, Pend D’Oreille lake, and the Kootenay river, as part of the Cabinet mountain range. The bulk of the Cabinet range, as now generally recognized, lies to the southeast of the strong trench running from Bonner’s Ferry to Sandpoint. To the writer it seems both easy and expedient to consider this trench as bounding the Cabinets on the northwest and the distinct range, hitherto unnamed, on the southeast. The limits of the latter range are: Bound- ary creek on the north, Priest river valley on the west, the Purcell Trench on the east, and the Bonner’s Ferry-Pend D’Oreille trench on the south and southeast. COLUMBIA MOUNTAIN SYSTEM AND ITS SUBDIVISION. The principal range unit adjoining the Selkirk system on the west is here called the Columbia system. (Plate 3.) It is definitely limited on the east by the Selkirk Valley and by a part of the Rocky Mountain Trench, the latter truncating the northern end of the Columbia system as it does the Selkirk and Purcell groups. On the south the Columbia system is limited by the Columbia lava plain. On the west the limit is determined by the lower Okanagan valley, and, to the northward, less well by the eastern edge of the belt of Interior Plateaus. That edge may be located for about thirty miles in the line of the main Kettle river valley. North of the main line of the Canadian Pacific rail- way, the belt of Interior Plateaus seems to reach, but not cross, Adams lake and Adams river. Still farther north, the western limit of the Columbia system is fixed by a trench occupied by the headwaters of the North Thompson river, and by an affluent of the Canoe river. Northwest of this trench begins the great system including the Cariboo mountains. Apparently the first official (Governmental) name for the mountains explored on the Canadian Pacific railway line west of the Columbia river was “Gold Range.* The group so named extends from the latitude of Shuswap lake to the narrows between the Arrow lakes. This usage has been adhered to by the Government of British Columbia.t In 1874, the Dominion Depart- ment of Railways and Canals introduced the name ‘Columbia range’ for the much larger mountain group including the ‘Gold range,’ and extending from * Map of British Columbia, compi'‘led under the direction of the Hon. J. W. Trutch, Chief Comznissioner of Lands and Works and Surveyor-General. Victoria, 1871. + Map of the province of British Columbia, compiled by J. H. Brownlee by direc- tion of the Chief Commissioner of Lands and Works. Victoria, 1893. 38 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 the headwaters of the North Thompson river southward to Lower Arrow lake.t This usage was confirmed by Selwyn and Dawson, each in turn Director of the Geological Survey of Canada.§ Nevertheless, the new general map of Canada issued by the Department of the Interior at Ottawa (1902) gives the name ‘Gold range’ to this larger group. The extension of the limits of the Gold range is a departure from the official tradition of both the provincial and Dominion governments. It appears best to hold the name ‘Gold range’ to its original designation of a local mountain group, and retain the title ‘ Columbia range’ with a broader meaning. For the immense Cordilleran unit stretching from end to end of the Selkirk Valley, and bounded on the east by the Columbia river, there is no question that the name ‘ Columbia range’ is more significant and appropriate than the name ‘Gold range.’ The latter name has a special disadvantage worthy of note. Although Dawson, in his later writings, used the name ‘Gold range’ in its original sense of a local mountain group, he as often used ‘ Gold range’ or ‘Gold ranges’ to include the Selkirk, Purcell, ‘Columbia’, Cariboo, and Omineca ranges. This inconsistent usage rebs the title ‘Gold range’ of even that modicum of value which it has as an alternative for the more significant title. As already stated, the name ‘Columbia range,’ with its comprehensive meaning, has the priority. The extension of the apposite title, ‘Columbia range’ (with variant “Columbia system’), to cover the larger area described in the foregoing para- graphs is, it is true, not according to tradition, but, as in the case of the Selkirk system and Purcell range, the widening of the meaning is justified by the lack of definition as to the true areal extent of the ‘Columbia range’ in its original use, and is enforced by the fact of crest continuity within a fairly well delimited belt of the Cordillera. The southern third of the Columbia system is characterized by compara- tively low mountains, which in rock composition are allied both to the northern part of the system-and to the belt of the Interior Plateaus. These southern mountains commonly show uniformity in summit levels; yet there are no remnant plateaus or very few of them, and it is advisable to regard these mountains as forming a group distinct from the Interior Plateaus. A con- venient name for part of the group, ‘ Colville mountains,’ was given as early as 1859-60 by the members of the Palliser expedition. In the preliminary paper it was proposed that the Colville group should inelude the mountains lying between the two forks of the Kettle river as well as all the part of the Col- umbia system south of the river. Further study and the test of actual convenience in description have since suggested the expediency of recognizing the moun- tains between the two forks of the Kettle river as forming an independent subdivision, and to them the name ‘ Midway mountains’ is given. Further- tS. Fleming, Exploratory Survey, Canadian Pacific Railway report, Ottawa, 1874, Map-sheet, No. 8. § Forest Map of British Columbia, published by G. M. Dawson in Report of Progress, Geol. Surv. of Canada, 1879-80. REPORT OF THE CHIEF ASTRONOMER 39 SESSIONAL PAPER No. 25a ° /20 OSSLAND [4 7. LF 5) GE | (1S TIA Ch. AN aN © NTAINS COLWLLE /70U, 48° Ficvure 4.—Diagrammatic map showing subdivision of the Columbia Mountain System at the Forty-ninth Parallel. more, the north-south trench occupied by the Sans Poil river, Curlew lake, and Curlew creek, divides the mountains south of the line into two distinct parts. To the eastern part, which is that nearer the site of old Fort Colville, the name ‘Colville mountains’ may be restricted; while the western division, bounded by the Kettle river, the Sans Poil-Curlew trench, the Columbia river, and the Okanagan river, may be called the ‘Sans Poil mountains.’ (Figure 4.) 40 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Another important, though small, natural subdivision of the system is limited on the north, east, and south by the Selkirk Valley; on the west, by the lower Kettle valley, and by a short trench running from Lower Arrow lake to Christina lake and the Kettle river at Cascade. This group may be called the Ross!and mountains. (Plate 3.) Again for convenience, the mountains occurring between Christina lake end the North Fork of Kettle river will be referred to as the Christina range. (Figure 4.) The more northerly part of the Columbia system is yet too imperfectly known to permit of subdivision in a systematic way. BELT OF INTERIOR PLATEAUS. As we have seen, the belt of Interior Plateaus is of primary importance in the systematic orography of the Cordillera. (Plate 6.) It is difficult of delimitation. On nearly all of its boundaries the belt fades gradually into the loftier, more rugged ranges encircling it. Its limits have been compiled and drawn on the map (Plate 3.) after a study of Dawson’s numerous reports of exploration. The limits are to be regarded as only approximate. The plateau character is obscure at the VForty-ninth Parallel, but the roughly tabular form and considerable area of Anarchist mountain, imme- diately east of Osoyoos lake, seem to warrant the slight extension of the belt across the International line. The southernmost limit of the belt is an irregular line following—(1) the main Kettle river valley; (2) a quite subordinate trench occupied by Myer’s creek and Antoine creek, in the state of Washington; (3) a part of the lower Okanagan valley; and (4) the Similkameen-Tulameen valley. CASCADE MOUNTAIN SYSTEM AND ITS SUBDIVISION. Usage, both official and popular, has gone far toward finally establishing the nomenclature for the immense ranges lying west of the Columbia lava plain, Midway mountains, and belt of Interior Plateaus. The Cascade range is now defined on the principle of continuity of crests. not on the basis of rock-composition. At the cascading rapids of the Columbia river the range is a warped lava plateau; in northern Washington it is an alpine complex of schists, sediments, granites, ete. In British Columbia, Dawson adopted the name ‘ Coast range’ to enforce the view that the granite-schist British Columbia moun- tains on the seaboard should be distinguished from the lava-built Cascades, as originally named, at the Columbia river. It has, however, become more and more evident, as the study of the Cordillera progresses, that rock-composition can never rival crest continuity as a primary principle in grouping the western mountains. Meanwhile, the name ‘ Coast range’ has survived, and is, in fact, the only name officially approved by the Geographic board of Canada for any principal division of the Cordillera “AT[VA LOATY VOUySsW 9 ALWTd I9AO ‘QSuURY UVsvUYG ‘ULeyUNOU IVgq Iw au WO IJ YQLOU SULOOT Ssnvajye[q LOMMaZUT 9yy Jo 4[oq een p. 40. - lL vol 25a, REPORT OF THE CHIEF ASTRONOMER 41 SESSIONAL PAPER No. 25a Dawson did not fix a southern limit for the Coast range. General usage has not fixed the nortiern limit of the Cascade range. The solution of the problem is obvious if the principle of limiting units by master valleys and trenches be applied. The Fraser river valley clearly supplies the required boundary between the two ranges. There seems to be no other simple adjust- ment of the two usages, which undoubtedly sprang up because of the existence of a political boundary at the Forty-ninth Parallel. It is important to note that the delimitation here advocated is not new, since it appears on two of the earliest official maps of British Columbia—those accompanying the 1859 British Blue Books, entitled ‘ Papers relative to the affairs of British Columbia.’ The remaining boundaries of the Cascade and Coast ranges, as well as the boundaries of the Olympics and of the Vancouver range, are at once derived from the map, and need no verbal description. These natural boundaries seem in large part to be located along structural depressions, and belong, therefore, to a type unusual in the Canadian Cordillera. The subdivision of the system where it crosses the Forty-ninth Parallel has already been recognized by Bauerman and, more in detail, by Smith and Calkins.* With these authors the present writer is in accord on the matter and a quotation from the report of Smith and Calkins will suffice to indicate such subdivision as seems necessary for the present report. ‘In northern Washington, where the Cascade mountains are so prom- inently developed, the range is apparently a complex one and should be subdivided. This was recognized by Gibbs, who described the range as forking and the main portion or ‘ true Cascades’ crossing the Skagit where that river turns west, while the ‘eastern Cascades’ lie to the east. Bauer- man, geologist to the British commission, recognized three divisions, and as his subdivision is evidently based upon the general features of the relief it will be adopted here. To the eastern portion of the Cascades, extending from mount Chopaka to the valley of Pasayten river, the name of Okanagan mountains is given, following Bauerman. To the middle portion, includ- ing the main divide between the Pasayten, which belongs to the Columbia drainage, and the Skagit, which flows into Puget sound, Bauerman gave the name Hozomeen range, taken from the high peak near the boundary. For the western division the name Skagit mountains is proposed, from the river which drains a large portion of this mountain mass, and also cuts across its southern continuation. It will be noted that the north-south valleys of the Pasayten and the Skagit form the division lines between these three subranges, which farther south coalesce somewhat so as to make subdivision les; necessary. ‘The Okanagan mountains form the divide between the streams flow- ing north into the Similkameen and thence into the Okanagan and those flowing south into the Methow drainage. In detail this divide is exceedingly irregular, but the range has a general northeast-southwest trend, joining *G.O. Smith and F. C. Calkins, Bull. 235, U.S. Geol: Survey, 1904, p. 14. 42 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 the main divide of the Cascades in the vicinity of Barron. The highest peaks, such as Chopaka, Cathedral, Remmel, and Bighorn, have a nearly uniform elevation of 8,000 to 8,500 feet and commonly are extremely rugged. Over the larger portion of this area the heights are above 7,000 feet, and below this are the deeply cut valleys.’ The respective east and west limits of the three ranges are clearly and definitely fixed by the longitudinal valleys of the Similkameen, Pasayten, and Skagit rivers, and by the partially filled depression of the Strait of Georgia. The northern and southern limits cannot at present be determined; that further step may be made when, in the future, the cartography of the rugged system is completed. (Plate 3.) SUMMARY. e The writer keenly feels the responsibility of suggesting many of the changes and additions proposed in the cartography of this large section of the Cordillera. The attempt to describe the geology of the Boundary belt without some kind of systematic orography on which to hang the many facts of spatial relation, is truly the making of bricks without straw. The scheme outlined above has thus developed out of a clear necessity. The orography of the International Boundary cannot profitably be treated without reference to longitudinal Cordilleran elements, often running many hundreds of miles to northward and southward of the Boundary. For this reason the accompanying map is made to cover all of the Cordillera lying between the forty-seventh and fifty-third parallels of latitude. (Plate 3.) The terms ‘range’ and ‘system’ are used in their common elastic mean- ings, with ‘system’ more comprehensive than ‘range. The Cordilleran system, or Cordillera, includes the Rocky Mountain system, the Selkirk system, etc. The Cascade range includes the Okanagan range, Skagit range, etc. A system may include among its subdivisions a mountain group without a decid- edly elongated ground-plan; thus the Columbia system includes the Rossland mountains. But both ‘range’ and ‘system,’ used with their respective broader or narrower meanings, involve the elongation of ground-plan and a correspond- ing alignment of mountain crests. The great weight of popular and official usage seems to render it inadvisable to attempt any more systematic organiza- tion of the common nouns in this case. It has been found almost, if not quite, as difficult to organize the proper names in an ideal manner. The basis of mountain grouping is purely topographical, and is, in the main, founded on established usage. A primary grouping recognizes within the Cordilleran body two relatively low areas, characterized by tabular reliefs, accompanied by rounded reliefs, generally accordant in altitude with the plateaus. These two areas are the belt of Interior Plateaus in British Columbia and the Columbia lava plain of the United States. The remainder of the Cordillera—ridged, peaked, often alpine—is divided into systems, ranges, and more equiaxial groups, either by ‘ trenches,’ by master valleys, or, excep- tionally, by structural depressions. REPORT OF THE CHIEF ASTRONOMER 43 SESSIONAL PAPER No. 25a The Caseade range, the Olympic mountains, the Vancouver range, and the Coast range of British Columbia, with their continuations north and south, compose what may be called the Coastal system. All the ranges east of the Rocky Mountain Trench, with their orographic continuations north and south, constitute the Rocky Mountain system. The Columbia lava plain and the belt of Interior Plateaus form the third and fourth subdivisions. A fifth more or less natural group, yet lacking a name, includes the Bitterroot, Clearwater, Ceur D’Alene, Cabinet, Flathead, Mission, and Purcell ranges, the composite Selkirk system, and the composite Columbia system, with the unnamed system including the Cariboo mountains. LEADING REFERENCES. Texts— Baedeker’s Guide-Book to the United States. 1893. BavERMaN, H.—Report on the Geology of the Country near the 49th parallel of latitude. Geol. Survey of Canada, Report of Progress, 1882-3-4, section and page 8B. British Government Blue-Books.—Papers relative to the explorations by Captain Palliser in British North America, 1859. Papers relative to the affairs of British Columbia, 1859. Further papers relative to the Palliser exploration, 1860. Further papers relative to the affairs of British Columbia, 1860. Map to accompany above-mentioned papers: Stanford, London, 1863. Brooks, A. H.—Professional Paper No. 1, United States Geological Survey, 1902, pp. 14, 15. National Geographic Magazine, vol. 15, 1904, pp. 217, 218. Proce. Eighth International Geographical Congress, St. Louis, 1904, p. 204. CANADIAN PaciFic RamLwAy.—Pamphlets. Century Dictionary, article ‘ Cordillera.’ Dana, J. D.—Manual of Geology, 2nd edit., 1874, and 8rd edit., 1880, pp. 15, 16, in each edition; 4th edit., 1895, pp. 389, 390. Davis, W. M.—In Mill’s International Geography, 664,668, 671, London, 1990. Dawson, G. M—Quart. Jour. Geol. Soc., London, vol. 34, 1877, p. 89. Report of Progress, Geol. Survey of Canada, 1877-78, part B, pp. 5-8. Canadian Naturalist, new series. vol. 9, 1879, p. 33. Report of Progress, Geol. Survey of Canada, 1879-80, part B, pp. 2, 3, and map. Report, British Assoc. Advancement of Science, vol. 50, 1880, p. 588. Geological Magazine, new series, vol. 8, 1881, pp. 157 and 225. Ann. Rep. Geol. Survey of Canada, new series, vol. 1, 1885, part B, pp. 15, 17, 22; vol. 3, 1888, part B, p. 12 and map, and part R, pp. 7-11; vol. 4, 1889, part B, pp. 6, 7; vol. 8, 1894, part B, pp. 3, 4 and map. 44 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Transactions, Royal Society of Canada, vol. 8, sect. iv., 1890, pp. 3, 4. The Physical Geography and Geology of Canada,’ reprinted from the Handbook of Canada, issued by the publication committee.of the local executive of the British Association, p. 7 and p. 40ff. Toronto, 1897. Bulletin Geol. Society of America, vol. 12, 1901, pp. 60, 61. Dawson, Sir J. W.—Handbook of Geology, pp. 202-204. Montreal, 1889. Dawson, S. E.—In Stanford’s Compendium of Geography: North America, vol. 1, 1898, map and text. De Lapparent, A.—Lecons de Géographie Physique, pp. 629-632. Paris, 1898. Fieminc, §.—Exploratory Survey, Canadian Pacific Railway Report, map- sheets 8 and 12. Ottawa, 1874. Report on Surveys on preliminary operations of the Canadian Pacific Railway up to January, 1877, map-sheets 1, 6, and 7. Ottawa, 1877. Frye, A.—Complete Geography, Boston, 1895, and later editions, various maps and diagrams. Grammar School Geography, map of North America and relief map, Boston, 1904. GaNNeET, H.—In Stanford’s Compendium of Geography: North America, 1898, vol. 2, p. 26, and maps of United States and North America. Geographic Board of Canada, Annual Report, 1904. GOSNELL, R. E.—Year-Book of British Columbia, p. 5. Victoria, 1903. Hayven, F. V.—-In Stanford’s Compendium of Geography: North America, 1883, pp. 40, 57, 58, 110, 111 and map. Hopkins, J. C.—Canada: An Encyclopedia of the Country, vol. 1, map, Toronto, 1898. JOHNSTON, KritH.— Geography,’ p. 426. Stanford: London, 1896. Kine, CLarENCE.—United States Geol. Exploration, 40th parallel, System- atic Geology, 1878, pp. 1, 5, and 15. Leconte, J.—Elements of Geology, p. 250. New York, 1892. Lixpcren, W.—Professional Paper, U.S. Geol. Survey, No. 27, 1904, p. 13. Réctus, E.—Nouvelle Géographie Universelle: Amérique Boréale, 1890, pp. 260, 264, 281. RussEtut, I. C.—Bull. Geog. Society of Philadelphia, vol. 2, 1899, p. 55 (with letters from Davidson, Davis, Dawson, Heilprin, and Powell). North America, pp. 60, 61, 120-125, 147, 165-168. Appleton: New York, 1904. Setwyn, A. R. C.—In Stanford’s Compendium of Geography, North America, 1883; map of British Columbia. (Setwyn, A. R. ©., and) G. M. Dawson.—Descriptive Sketch of the Physical Geography and Geology of the Dominion of Canada, pp. 33, 34. Montreal, 1884. REPORT OF THE CHIEF ASTRONOMER 45 SESSIONAL PAPER No. 25a SwirH, G. O., and Cankins, F. C.—Bull, U.S. Geol. Survey, No. 23%, 1904. pp. 12-14. . Suater, N. S.—Nature and Man in America, pp. 250-256. New York, 1891. Wueever, A. O.—The Selkirk Range, 2 vols. Government Printing Bureau: Ottawa, 1905. Wueeter, G. M.—U.S. Survey west of the 100th Meridian, Geographical Report, 1889, p. 11. Wuirtney, J. D—The United States, pp. 22-30, 68, 79, 122, and refer- ences. Boston, 1889. General Atlases and Encyclopedias— Atlas Universelle; Hachette: Paris, 1904. Atlas Vidal-Lablache; Paris, 1904. Century Atlas; New York. Encyclopedia Americana, 1904. Encyclopedia Britannica Atlas. English Imperial Atlas; Bartholo- mew: London, 1892. Home Knowledge Atlas; Toronto, 1890. John- son’s Universal Cyclopedia; edited by Guyot, 1891. Keith Johnston’s Royal Atlas; Edinburgh, 1885. Rand-McNally Indexed Atlas of the World, 1902. Rand-McNally Indexed Atlas-of the Dominion of Canada, 1905. Stanford’s London Atlas, folio edit., 1904. Stieler’s Handatlas, 1897. Times Atlas; London, 1900. Universal Atlas; Cassell: London, 1893. Official Maps— Map of the Dominion of Canada, geologically coloured from surveys made by the Geological Corps, 1842-1882; two sheets, scale 45 miles to one inch. Geol. Survey Department, 1884. (Topography based on wall- maps issued by the Department of Railways and Canals.) Geological Map of Dominion of Canada, western sheet; scale, 50 miles io one inch. Geol. Survey Department, 1901. Map of the Dominion of Canada and Newfoundland, James White, F.R.«.s., geographer; scale, 35 miles to one inch. Department of the Interior, Ottawa, 1902. Map of British Columbia to the 56th parallel N. lat., compiled and drawn under the direction of the Hon. J. W. Trutch, Chief Commissioner of Lands and Works and Surveyor-General. Victoria, 1871. Map of the province of British Columbia, compiled by J. H. Brownlee, by direction of the Hon. F. G. Vernon, Chief Commissioner of Lands and Works. Victoria, 1893. Map of the west division of Kootenay District and: a portion of Lillooet, Yale and East Kootenay, B.C. Compiled by direction of the Hon. G. B. Martin, Chief Commissioner of Lands and Works. Victoria, B.C., 1897. ba) yr Pigg eee ee Ae . (uk hee Sesie SSeS ees ‘ 4J9] uO ‘punossyouq UL aBury preuoqovp, pue AaTea pvayyerg § 4yS8ta uo vaery pooIng jo daeos ‘euojsoul] YaATG Jo pesodwoo ‘uosdmoyy, yuNoy § punors e_pprm ul axe, taddq AQ][VA Saye] VpULyT, UMO nH [s Sutyoor] 5a—vol. ii—p. 47. 9° 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER IV. STRATIGRAPHY AND STRUCTURE OF THE CLARKE RANGE. ROCKY MOUNTAIN GEOSYNCLINAL PRISM. One of the least expected results of the Boundary survey consists in the discovery that almost all of the mountains traversed by the Commission map between the Great Plains and the summit of the Selkirk range—an air-line distance of one hundred and fifty miles—are composed of a single group of eonformable strata. These rocks are as yet largely unfossiliferous but all of them are believed to be of pre-Devonian age. For the most part they are water- laid, well-bedded sediments but contain one important sheet of extrusive lava which extends quite across the whole Rocky Mountain system and the eastern part of the Purcell system. Though the sedimentary group is a unit, it has been found that noteworthy lithological differences appear in the rocks as they are followed along the Boundary line from the Front ranges westward. These differences are due to gradual changes of composition and no two complete sections taken five miles apart on an east-west line would be identical. Never- theless it has been found possible to relate all the essential features of these varying strata to four standard or type sections. The most easterly type section was made in the Clarke range. It agrees very closely with the section already described by Willis from the Lewis range at localities lying on the tectonic strike from the localities specially studied in the Clarke range by the present writer. The rocks thus found to compose both the Lewis and Clarke ranges belong to what may be called the Lewis series. The type section constructed from traverses made in the Galton and Mac- Donald ranges include strata which are here grouped as the Galton series. The equivalents of the same series compose the entire Purcell mountain system at the Forty-ninth Parallel and belong to a sedimentary group which may be called the Purcell series. The fourth type section was constructed from magnificent exposures occurring in the eastern half of the Selkirk mountain system. This assemblage of beds will be referred to as the Summit series. The name is taken from Summit creek along which a great part of the series is exposed; the creek was itself named from the fact that it heads on the water-divide of the Selkirks. Analogy with the other three series names suggests ‘Selkirk series’ for this fourth group of strata, but that designation has already been used by Dawson for the related but lithologically distinct group described in his traverse on the main line of the Canadian Pacific railway. 47 48 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 The retention of there four series names implies some slight tax on the’ memory but that drawback is much more than offset by the ease of grasping and systematizing the many petrographic and stratigraphic facts which must be reviewed before the constitution of the great geosynclinal prism is under- stood. In view of the general lack of fossils throughout the belt, the differ- entiation and correlation of the beds must be based on lithological properties. The following description of each series includes a statement of the facts on which is founded the writer’s belief in the integrity of the whole sedimentary field, one huge sedimentary prism constituting the staple rocks in the eastern third of the Cordillera at the Forty-ninth Parallel. The summary of the individual facts, as they are clustered in describing the four series, will further well illustrate the systematic variation in the geosynclinal prism as it is crossed from east to west. In each type section the formations will be considered in their natural order, beginning with the oldest. The description will, in each case, be made concisely and will be shorn of many items of fact which do not appear of importance in the larger stratigraphic problem. The Purcell Lava formation will ke treated in chapter IX. The description of the four type series will be found nearly to cover the stratigraphy of the different ranges from the Lewis on the east to the Yahk on the west. In the Galton and MacDonald ranges there are bodies of fossilifer- ous Devonian and Mississippian limestone which are properly parts of the prism, but, having generally been eroded away, now form only quite subordinate masses within the Boundary belt. These will be described in connection with the account of the Galton series. The only other bed-rock sedimentary forma- tion occurring between the Great Plains at Waterton lake and the Purcell Trench at Porthill is a thick but local deposit of Tertiary fresh-water clays and sands flooring the Flathead valley. This occurrence will be noted in con- rection with the stratigraphic description of the Clarke and Lewis ranges. The stratigraphy of the Selkirk system is much more highly composite than any of the eastern ranges; its description will, therefore, be detailed only so far as the Summit series and the underlying terrane are concerned, and will then be interrupted by a chapter giving the results of correlating study on this gigantic stratified unit, the Rocky Mountain Geosynclinal.* As an aid to clearness it may be noted, in anticipation of a later chapter, that the Rocky Mountain Geosynclinal includes all the sedimen- tary formations from the base of the Belt (pre-Olenellus) terrane up to and including the Mississippian formation, as these beds are developed in the eastern half of the Cordillera. The Lewis, Galton, and Purcell series represent only a part of the whole prism, in each case the youngest exposed bed being * Following Dana (Manual of Geology, 4th edition, p. 380) the writer distinguishes the geosyncline, the large-scale down-warp of the earth’s surface, from the load of sediments which may accumulate on the down-warped area. In the present report the load of sediments will be referred to as a “geosynclinal prism” or, more briefly, as a “geosynclinal.” REPORT OF THE CHIEF ASTRONOMER 49 SESSIONAL PAPER No. 25a not far from an Upper Cambrian horizon, and the oldest exposed bed being located well above the base of the Belt terrane. The Summit series includes the entire Belt terrane and a vast thickness of conformably overlying strata which may represent the whole Paleozoic succession up to and including the Silurian. Overlying the Summit series, apparently conformably, is a very thick and massive limestone which is probably Carboniferous but may, in its lower part, belong to the Devonian. In other words, it seems possible that the complete gecsynclinal prism is represented in the exposures of the Boundary belt where it passes through the southern Selkirks. The name ‘Summit series’ refers only to the unfossiliferous formation making up the lower and greater part of the prism in this mountain system. LEWIS SERIES. The writer has carefully studied the Lewis series only within the limits of the Clarke range. Since the Commission map extends but a mile or two to the eastward of the summit monument, a close mapping of the different formations between that monument and Waterton lake was not feasible. In this stretch of fifteen miles the field work was confined to the measurement of a few sections. These, however, occurred in areas of unusually complete rock- exposure and much light on the composition of the lower one-third of the series was derived from their examination. In the Lewis range the writer had no opportunity for close work and his experience there was limited to rapid traverses from Waterton lake to Chief mountain and thence, by way of Altyn and the Swift Current Pass, to Belton, Montana. Limited as that opportunity was, it sufficed to corroborate the belief— already reached after reading Willis’ paper on the ‘Lewis and Livingston Ranges ’—that the stratified sequence in the Lewis range is essentially identical with that in the Clarke range. It will, in fact, appear in the following account that the columnar section constructed by the writer from data obtained wholly within the Clarke range, matches well, member for member, with Willis’ columnar section derived almost entirely from observations in the Lewis range. Partly in order to emphasize this identity the name ‘ Lewis series’ has been selected to cover the whole group of strata in the Clarke range—the group now to be described. (See Plate 7.) Beginning at the top the formations included in the Lewis series have been listed in the order of the following table: formation, Thickness in feet. Deminant rocks. Top, erosion suriace. IKGintlasiw esc, cere 860+ Argillite. Sheppacda weer reeneta. 600 Silicious dolomite. Parcel puiavarreee sera 260 Altered basalt. SLVOM eile watch oe ee ete nc 4,100 Magnesian limestone and metargillite. Grinnell c Gilesh ae ees 1,690 Metargillite. Appelcunmys. ws) =. ei bs 2,660 Metargillite. MANGUM Veet Sanit ele ea ee 3,500 Silicious dolomite. Wiatertones io 5. cause 200+ Silicious dolomite. 13,720 Base concealed. 25a—4 50 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Excepting the Purcell Lava these rocks will be described in the present chapter. The volcanic formations of the range are described in chapter IX. WATERTON FORMATION. The lowest member of the Lewis series as exposed within the Boundary belt was seen at only one locality—at the cliff over which the waters of Oil creek (Cameron Falls brook of the older maps) tumble from the hanging valley of Oil creek into Waterton lake (Plate 8). This member may be ealled the CLARKE RANGE Ol Prospect ( Out es ) 0 2 4 Miles Ficure 5.—Diagrammatic map showing position of the Structure-section (See Fig. 6) east of the Rocky Mountain Summit. Waterton dolomite. At the cascade it is seen to be conformably overlain by the Altyn limestone. From the sharp bend below Oil City the creek faithfully follows the axis of a strong anticline which pitches gently northward. As one descends the creek he also descends in the stratified rock-series, and at Cameron Falls walks upon the Waterton dolomite, the visible core of the anticline. The speq Jo drip pur aqitojop u0y °2 ALVI Ia}e AA JO ‘OUT[OIZUR Y9eIH [IQ JO CUT] JSaMTYQNOS UL LITOVLCYO SoPBlASU[TL MOL A, “UOSBIS 19JCM 7 ID TIO UO sTpBip uO LITU . li—p. 50. vol 25a "Gg “SLT 9S UOTZVOOT 10,7 ‘qsvayqnos SUTYOO'T ‘asuvy oxIV[O 9yy Jo edo[s ussysva “YaeIH [IG Jo JsveyyNos ospld oY) Suo[e ‘ay114s 94 sso1oV UOTZO9S OINJONIYG—'9 AMNOLT] $3ajo0r¢ ]O2']49Q pyo [Oyuoz40L, Sa}! & SS =| IW 1 = [pAaq Das MS “3'N z SNOSIVLAYO tied hy SLE se Q N : 2S S 8 2 Nin Ss S s Sy SS ~s 8 Nd = S N N S ~ < S 2 S cS y N : ae US Sees < S Sa 8 << SS S s S 25a—44 52 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 section of the dolomite is incomplete; with all its typical characters it dis- appears, to the eastward, beneath the stream gravels and glacial deposits surrounding the lake. (Figures 5 and 6.) Throughout its whole observed thickness of 200 feet the formation consists of an exceptionally strong and massive, dark gray carbonate rock, weathering dark gray to brownish gray and sometimes buff. In the field the rock has a most deceptive resemblance to a homogeneous, thick-bedded argillite. It effer- vesces but slightly in cold dilute acid, and the essential carbonate character SAH Ee FEED st om XY Sk Oh Figure 7.—Diagrammatic drawing from thin section of Waterton dolo- mite, showing middle part of a lense of orthoclase in interlocking granules. Rhombohedra of dolomite are embedded in the anhedral dolomite which forms most of the rock. The cleavages shown in the carbonate are diagrammatic only and in reality are seldom visible. The black spots represent carbonaceous matter. Highly magnified ; diameter of circle 0°15 mm. was not suspected until the more careful laboratory study was put on the rock. The thin section showed immediately that it is largely composed of carbonate grains. Their size is very small, the diameters steadily averaging about 0-02 mm., with a few grains reaching twice or thrice that diameter. These grains are sometimes knit together in a thorough, interlocking manner but more often REPORT OF THE CHIEF ASTRONOMER 53. SESSIONAL PAPER No. 25a show a tendency to assume the rhombohedral form, the habit characteristic of the grains in true dolomites. In many of the laminae of the rock (0-2 mm. to 1 mm. in thickness) the minute rhombohedra are embedded in a compound, colourless to pale-brownish base. It is composed in part of very minute, anhedral grains of glass-clear substance. ‘These range in diameter from 0-01 mm. to 0-05 mm. A few of them are undoubtedly quartz; the great majority have the single and double refraction of orthoclase. In addition to the rhombohedra of carbonate, the base is charged with abundant black, opaque dust. The particles of the dust average under 0-01 mm. in diameter. Since the rock decolourizes before the blow-pipe it seems clear that the dust is largely carbon, though hematite and probably magnetite are also represented in some amount. Some laminae of the rock are seen to be specially charged with roundish clumps and lenses of minute orthoclase crystals. (Figure 7.) These are inter- locked and in all of three thin sections made from two different hand-speci- mens, show no trace of a clastic origin. They give the writer the impression of having been introduced and erystallized from solution, or at least segregated in their present positions from the general mass of the rock. The few quartz grains interlock with the orthoclase and are just as clearly not of clastic origin. Professor M. Dittrich analyzed a typical specimen of the rock, (No. 1338) with the result shown in Col. 1 of the following table. The extraordinary abundance of potash prompted a second determination of the alkalies in the same rock-fragment; this time the potash showed 6-12 per cent and the soda, 0-25 per cent. A different fragment of the same large hand-specimen gave Mr. M. F. Connor 5-54 (also 5-71) per cent of potash and 0-24 (also 0-18) per cent of soda. The average of all four determinations is entered in Col. 2, the other oxides being given in the amounts shown in Professor Dittrich’s total analysis. Col. 3 shows the molecular proportions corresponding to Col. 2. Analysis of the Waterton dolomite. 1b 2. 3. Mol SiO, 30-46 30-46 508 A1,0, 6-86 6-86 088 2,0; 4.53 4-53 -028 eO.. 1-89 1-89 026 MgO.. 10-07 10-07 252 Care: 16-02 16-02 286 Na0-: Rates oe +87 38 006 K,O Bis Jot iniGe COMO ORDEED OR UO Lom oe 5-71 5-77 062 H.O, ‘at 110°C... Meats Pein combate tu att amet “11 “11 H.O, above 110°C. Rie oeetae ros cplte rs tatarg, cre etek is ee meke 1-31 1-31 Sie wie CO ea ete Bo en Hit ay ete ae ae Bo ale Ota 513 100-38 99-95 SJORIPARG A vaatiSe lee Ge tOale ONAO MEOMAGEDon Sond onda eae 2-749 Insol. in hydrochloric acid.. Se ECE Ua IaG PAR Ga E TCE 42.80% Soluble in Bee CTIG acid: Fe,0... ce ee cee ert 0 Br 3 eve rakes ene Mm? BY ALO: slelp elo sielpaeloinielel) sje steie\r ele) | ele) /s7e tele: COMDIOOO DO soo Goscoao oO 2-03 CaQla =: ; 16-23 MeO tet i Se os. savin.) hase Geant ara ourtete ahs lb 9369 5O4 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 The average analysis of Col. 2 has been calculated, on the assumption that the alkalies are referable to the orthoclase and albite molecules and the iron oxides to magnetite. All of the lime is referred to the carbonate. The result, noted below, is probably not far. from representing the actual composition of the rock. Onthoclasen. nese 2 oye mec eee Oe RR ea ee 34-47 ATID ICO yes So tiega, alee tered ana Zeaeee Norske Ue LAE RL No es ae 3-14 Qantas eee re Mca weechauss joie pois ose POSSE Pas EERO AR en EER ane 6-00 Miaronetite ison cieee Miva oSuilisin sede he Ua ee pele ci aileeent Sinncn mee IOS 6-26 Macnesiumicarbonatens shin sents nee mete e shen BAe eer 21-17 Calcimmycarbomated sce Ae asics aa: eranccds saat ern fol orem ays apelin cre atens ge 28-60 99-64. Half of the rock is composed of the two carbonates, in which the ratio of Ca to Mg is 1-87: 1, indicating but a very slight excess of calcium over that found in normal dolomite. The remainder is chiefly silicious, especially feldspathic matter. The rock is apparently unique among analyzed dolomitie sediments in showing such a high percentage of potash. This alkali is without doubt contained in the orthoclase, which is probably somewhat sodiferous. The concentration of so much of this feldspar in a dolomitic sediment is hard to understand. If the microscopic relations permitted the view that the orthoclase, like the feldspars of the Altyn beds, were of clastic origin and derived from a granitic terrane, one would still be at a loss to understand the relative poverty in quartz. The suggestion due to optical study, that the feldspathic material has really been introduced in solution offers obvious difficulties but seems to be a more pro- mising hypothesis to explain the presence of most of the feldspar. This more probable view itself suffers from the doubt arising from the fact that the rock shows no evidence of having been reerystallized or notably metamorphosed, as we might expect if it had been penetrated by solutions to the extent demanded. A third hypothesis, that, under special conditions, the potash was introduced into the original carbonate mud in the form of the soluble aluminate of potassium, which during burial and lithifaction, reacted with dissolved silica in the mud-water to form orthoclase, is perhaps worthy of mention; but it faces the obvious objection that no conditions in nature are known by which the aluminate is formed from the potassium salts in sea-water. Another sugges- tion may be drawn from the fact that isomorphous mixtures of calcium and potassium carbonates can be prepared in the laboratory. If such isomorphic mixture were thrown down from the sea-water of the Waterton time, the one constituent of the orthoclase would be added to the mud but the presence of alumina in its exact proportion to potash (and soda) would be hard to explain. Finally, as suggested to the writer by Professor C. H. Warren, the presence of so much alkali may possibly be due to the original precipitation of glauconite in the mud, in which the feldspar was formed by recrystallization under peculiar conditions. In view of its obvious difficulties the problem of this extraordinary rock must be left unsolved. REPORT OF THE CHIEF ASTRONOMER 55 SESSIONAL PAPER No. 25a The carbon dioxide shows some deficiency if, as seems necessary, practical- ly all of the magnesia and lime are to be referred to the normal carbonate forms. From the fact that a similar deficiency is found in all of the analyzed carbonate rocks from the overlying Altyn, Siyeh, and Sheppard formations, it is reasonable to suppose that it is not due to the necessary errors of analysis. In all these cases the deficiency may be hypothetically explained by the presence of small amounts of hydromagnesite, (MgCO,), Mg (OH), + 3 H,O. The large proportion of water expelled above 110° C. might also be referred in large part to the basic carbonate. It is thus possible to conceive that 'from five to seven per cent of the rock is made up of that substance. The specifie gravities of three type specimens of the impure dolomite were found to be respectively, 2-749, 2-777, and 2-782; the average is 2-769. These values show that magnesia must be high in all three specimens. Though the dolomite occurs in massive plates from six to eight feet thick, aml though it is highly homogeneous from top to bottom of the section at Cameron Falls, yet a close inspection of the ledges shows that the rock is made up of a vast number of thin, often paper-thin, beds. Scores or hundreds of such laminae can be counted in a single hand-specimen of the massive dolo- mite. Their surfaces are generally parallel, and cross-bedding, ripple-marks, or other evidences of shallow-water deposition are absent. The character of the rock, on the other hand, indicates that the carbonate was deposited quietly, persistently, on a sea-floor not agitated by waves or strong currents nor recelv- ing coarse detritus from the lands. The minute bedding and the exceeding fineness of grain, point to an origin in chemical precipitation. The presence of the carbonaceous dust suggests that the precipitation took place in the presence of decaying animal matter and that the dolomite is thus analogous to the chemically precipitated, powdery limestone now forming in the deeper parts of the Black Sea. The theoretical questions regarding this and the other carbonate rocks of the geosynclinal prism will be discussed in chapter X XIII. The Western Coal and Oil Company have made a boring a few hundred yards from Cameron Falls and in the middle of the Oil creek anticline. The log shows that the bore-hole penetrates 1,500 feet of hard limestones inter- stratified with subordinate beds of quartzite and silicious argillite (metargillite). All these rocks are fine-grained and, so far as one may judge from the drillings, many are similar to common phases of the Waterton formation. The beds all seem to underlie the visible Waterton dolomite conformably. We have, there- fore, in addition to the exposed members of the Lewis series, at least 1,500 feet of still older beds which should be considered as belonging to the series. Until these strata are actually studied at surface outcrops: they cannot be described adequately and for the present report, the Lewis series is considered as extending downward only to the bottom bed of the Waterton formation where it crops out at the cascade. At the depth of about 1,600 feet the bit of the boring machine passed from the hard limestones into soft shales which persisted to the lower end of the bore- bole about 2,000 feet from the surface. These shales are referred to thse 56 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Cretaceous. In other words, the cldest sedimentary beds visible in the Rocky mountains at the Forty-ninth Parallel here overlie one of the youngest forma- tions of the region. The relation is plainly one of overthrusting, which will be discussed in the section devoted to the structure of the Clarke range. ALTYN FORMATION. General Description—The Altyn formation, immediately. overlying the Waterton dolomite. was so named by Willis, who described it from a typical section near the village of Altyn, Montana, fifteen miles south of the Boundary line. The Altyn is not exposed within that part of the Clarke range which is. covered by the Commission map. The writer studied the formation chiefly in a fine section on Oil creek and thus on the Atlantic side of the Great Divide. The exposures are there excellent for the greater thickness of the formation. In this section the eroded edges of 3,000 feet of Altyn strata can be seen on the long ridge running southwestward from the bend in Oil creek two miles below Oil City (Figure 6). At least 500 feet of additional, basal beds are exposed along the lower course of the creek and it is these which have been referred to as conformably overlying the Waterton dolomite. Calcium and magnesium carbonate are the dominant constituents of the formation. With these are mixed grains of quartz and feldspar in highly varia- ble proportion. The rock types thus include arenaceous magnesian limestones, dolomitic sandstones, dolomitic grits, and pure dolomites, named in the order of relative importance. The character of the bedding and the colours of the rocks were often found to vary in sympathy with the rock composition. On this threefold basis the thick formation as exposed along Oil creek, has been sub- divided, though only approximately, as follows: Columnar section of the Altyn formation, showing thicknesses. Top, conformable base of the Appekunny formation. a 300 feet.—Medium-bedded, light gray, sandy, magnesian limestone, weathering gen- erally Fale buff or, more rarely, strong brownish buff; a few inter-. beds of magnesian limestone. lay ETN) Thin-bedded, light gruy and greenish gray magnesian limestone, weather-. ing buff; subordinate interbeds of sandy limestone. Oeil) = 2 Massive, homogeneous, light gray, sandy limestone; weathering yellowish white; in some horizons bearing cherty nodules and large, irre- gularlvy concentric silicious concretions. al BI) Thin-bedded, bufi-weathering magnesian limestone. Massive, highly arenaceous or gritty, gray magnesian limestone, weather- ing white or very pale buff HiG50)) + Thin-bedded, relatively friable, gray or greenish gray magnesian lime- stone, weathering buff or yellowish white. ge2o0- Light gray, thick-bedded, sandy and gritty magnesian limestone, weather- ing pale buff; cccasional thin intercalations of thin-bedded mag- nesian limestones bearing cherty nodules and silicious concentric concretions. 3,500+ feet. Base, conformable top of Waterton formation. REPORT OF THE CHIEF ASTRONOMER 57 SESSIONAL PAPER No. 25a The total thickness of the Altyn as shown in this Oil creek section is much greater than that seen farther south by Willis (1,400 feet). The difference is not to be explained by overfolding or overthrusting. Individual beds and groups of beds aré, it is true, considerably crumpled, especially in the lower part of the section; but the average southwesterly dip of about 30° is preserved throughout. The evidence of original conformity from top to bottom seems as clear. Lithologically, the sediments here differ from those described by Willis in carrying a notable proportion of rounded grains of quartz and feldspar. The sandy and gritty strata occur chiefly in the middle of the section, there totalling nearly 1,000 feet in thickness. It is thus convenient to recognize a tripartite division of the Altyn as exposed along the International Boundary :—- ' An upper member (@ and b) of thin-bedded, silicious dolomite 1,250 feet thick; a middle member of thick-bedded, massive arenaceous dolomite and ealcareo-magnesian sandstones (c,d and e), 1,350 feet thick; and a lowest mem- ber of generally thin-bedded, silicious dolomite (f and g), at least 900 feet thick, containing sandy beds toward the baze. Nowhere in the formation were there found sun-cracks, rill-marks, ripple-marks or any other indication that the sediments were laid down in very shallow water or on a bottom laid bare between tides. A visit was paid to Chief mountain and to the original locality at Altyn, Montana, where the rocks were found to correspond to Willis’ description except in being often distinctly arenaceous. Willis’ brief summary of the facts ob-erved by him reads as follows:— - ‘Limestone of which two members are distinguished; an upper mem- ber of argillaceous, ferruginous limestone, yellow, terra-cotta, brown, and garnet red, very thin-bedded; thickness, about 600 feet; well exposed in summit of Chief mountain; and a lower. member of massive limestone, grayish blue, heavy-belded, somewhat silicious, with many flattened con- eretions, rarely but definitely fossiliferous; thickness, about 800 feet; type locality, basal cliffs of Appekunny mountains, north of Altyn, Swift Current valley.’* . As the formation is followed southeastward the uppermo:t member shows a decided darkening of tint—to terra-cotta, red, and brown of various deep shades, which then dominate the lighter buff colour characteristic of that mem- ber at the Boundary. It seems clear that the whole of the lowest member and part of the middle member of the Altyn at the Boundary are not exposed in the sections studied by Willis. On the whole the field relations in the Oil creek section are more favourable to giving one an accurate idea of the whole Altyn formation than are the field relations at either Chief mountain or at Altyn itself. As already noted, this great formation is heterogeneous but every bed of it seems to carry a notable percentage of carbonates. The cement of even the * Bull. Geol. Soc. America, Vol. 13, 1902, p. 317. 58 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 most sandy and gritty layers is dolomitic. Hand-specimens representing the principal phases were collected; each of them has a cement soluble in hot hydrochloric acid. The weathered surfaces of the arenaceous beds are always roughened by the clastic grains of quartz and feldspar standing out above the carbonate, the constituent more soluble in rain-water and soil-water. None of the many specimens collected shows other than the feeblest effervescence with cold, dilute acid. The specific gravity of thirteen specimens ranges from 2-688 in the most silicious phase, to 2-814 in the least silicious phase. The average for al] thirteen is 2-763. These facts, together with the character- istic buff tint of the beds on weathered surfaces, of themselves indicated that the formation is throughout highly magnesian. That conclusion has been greatly strengthened by the chemical analysis of three specimens which. respectively represent the staple rock-types in the lower, middle, and upper members of the formation. The analyses will be described in connection with the microscopic petrography of the three members. Lower Division—Thin sections from the dominant rock of the lowest members, a very homogeneous, compact, thin-bedded limestone, show that the carbonate occurs in the form of an exceedingly fine-textured aggregate of closely packed, anhedral, colourless grains averaging from 0-01 mm. or less to 0-02 mm. in diameter. The largest of the grains may run up to 0-03 mm. in diameter. A very few minute, angular grains of quartz and unstriated feldspar, and some dust-like, black particles (probably both magnetite and carbon): are embedded in the mass. The bedding is well marked in ledge or hand-specimen but is yet more conspicuous under the microscope. The lamin are bounded by sensibly plane surfaces, affording in section parallel lines often only 0-2 mm. apart. This bedding lamination is brought out rather by small differences of grain among the layers than by admixture of material ~ other than carbonate. The specimen chemically analyzed has the microscopic characters just outlined. It was collected at the 5,050-foot contour on the spur running southwestward from the right-angled bend in Oil creek on the south side of the ercek and about one mile below Oil City. The analysis made by Professor Dittrich (specimen No. 1322) showed weight percentages as follows :— Analysis of type specimen, lower Altyn formation. Mol SiO, 13-46 +224 Al,O, 1-56 015 Fe,0, 1-05 006 FeO -48 007 MgO 17-81 445 CaO 25-08 448 INCL ORS Sacuaamea secre 28 005 NG Oe rvatserticis Rasatoon 1-08 012 Te EC Ge Li hed Oa ae eee Se OR Re Pereira ne ie artis a Ie eR 04 Rass He Okahove 110°C so s.5 see cee eee ee 1-23 -070 Oe eee osc aeh lb SERIE, A eee irae nar ie Oe OS +865 S105 eS cider Oth Ree Ine Mn ar ComarmratmEn RT aS Mia alge 2-805 REPORT OF THE CHIEF ASTRONOMER 59 SESSIONAL PAPER No. 25a A second analysis gave the proportions of the oxides entering into solution in hydrochloric acid and also the percentage of insoluble matter, as follows :— insoluble imehydnochloricn acide ty eerer ut cmaocemeieica een rie nr 16-02% Soluble in hydrochloric acid: EO OSS Sa oad sadd eres cs 1-70 Al,O3.. “37 CaO.. 25-16 MgO.. 16-83 In the soluble portion CaO: MgO =25-16: 16-83=1-495: 1, a ratio only very slightly higher than the ratio- for true dolomite, namely, 1-4:1. The carbon dioxide required to satisfy those bases is 38-28 per cent, which is close to the percentage actually found. Considering that the alkalies belong to the feldspars and the iron oxides to magnetite, the proportions of the various constituents have been calcu- fated to be:— Calermmecarhonates..= 4 seas ae ee 44.9 Magnesium carbonate.. .. :....-. 35-3 UATE er iee cimio cis seston tar ates Mec elem 8-0 Orthoclase molecule... .. ... 5-6 Albite molecule... .. .. .. 2-6 Magnetite.. .. . 1-4 Remainder.. .. . 2-2 ay = ne a) As in the case of the Waterton dolomite it is difficult to understand the high proportion of combined water. It may occur with the silica alone or it may occur in a hydrous silicate of magnesium. About 80 per cent of the rock is composed of carbonates in the form of true dolomite. Middle Division—A specimen characteristic of the middle member (zone e), though not of its most sandy part, was collected at the low cliffs four hundred yards east of the derrick at Oil City. This rock on the fresh fracture has the typical pale gray colour of the forma- tion and weathers whitish to pale buff. On the weathered surface the glassy wind- worn or water-worn, rounded to subangular quartz and feldspar grains stand out like white currants in a flour paste. The grains are of varying size up to 0-3 mm. in diameter, averaging about 0-2 mm. The quartz grains are the more abundant. The feldspar is chiefly a fresh and characteristic microper- thite, with orthoclase in more subordinate amount. No soda-lime feldspar could be demonstrated. In this analyzed specimen as in the majority of the thin sections from all three members of the formation, round grains of chalcedonic or cherty silica, with diameters also averaging 0-2 mm., occur in considerable number. These small bodies are probably of clastic origin. Oolite grains with poorly developed concentric and radial structure are likewise rather abundant in both the analyzed specimens and others. Dr. H. M. Ami has noted that some of these grains have a certain resemblance to radiolaria, but regards their inorganic, concretionary origin as more probable. 60 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 All of these various bodies are embedded in a carbonate base which in all essential respects is similar to that found in the dolomite of the lower member. The carbonate again forms a compact aggregate of anhedral grains, varying from 0-01 mm. or less to 0-025 mm. in diameter, averaging about 0-015 mm. The total analysis of the specimen (No. 1320) afforded Professor Dittrich the following result :— Analysis of type specimen, middle Altyn formation. Mol. SiO slaty Sct cale Persea ckaces, Boise Se ekia At ea Se ets ee 18-89 315 DINU Cech read coche Ms te otk atta MES ange tag ROAR EN tS Fe 0-49 -005 GS Oe pte orek ae en ty Pte sate aici cons is ean te UR ey me ect 0-72 004 ECO aieiiciel bays eee See eR eh a CIR ea Caen O trate ON (yeq © esha trae Pete nie cc Ua perh a Bee ters PON UO accent IN) AN ce 16-79 420 CTO Miro ES UR TSS AN Fed BO SOR ae RMU gs ily Seca oar ear 23-86 426 HO at 110°C.. SEP LAAS ole taahellsattactattemmane aren Nevehesee eae ea create hastened 0-18 Rae H.O above THe CRTC Sealine UNG Peruana ant hun SAgieTs 1:57 087 CONS SCN ENGR VNB MED Aik treme NORE dee cg ON NONE TEO 36-89 +838 100-43 Sip Babee erecta ie wate arte aetna oes eens tao Se ceteris 2-802 A second, partial analysis of the same specimen gave the following data :-— Insolublovin yhydrochloriciacid@ess mess ceh en in ea itocmcenae 21-13 Soluble in ae ee acid: Bee A105; iter earatih Sec eN ls treCee e seen ween. hap aieea ece al ete Gantee epee naaS 0-58 MgoO.. iar nea nee tae ar i SPetionl Ameo SINC. SVG The table ae eisai proportions shows that the alumina is too low to match the alkalies of the feldspars actually present. Another determination of alumina and iron oxides of a part of the same specimen gave AI,O,, 1-22 per cent; Fe,O,, 1-01 per cent; and FeO, 0-33 per cent. The ratio of CaO to MgO in the soluble portion is 1-48:1, closely approxi- mating the ratio in true dolomite. Calculation gives the following mineral proportions in the rock :— Calcium carbonate jo cc sues eteb reel eee Ue car ob oe 42-6 Magnesium ear bonbites «oi i'5) Weide extn aes Oke gh a 30:3 Quartz and chert.. ties Lote Moher eve) Male naelenetclomtarounets Wrere abs covey are teers 14-2 Orthoclase Tholecule. |. kia ke nce te Reman ok aes eee Sraeige 3-3 Albite molecule... 3-7 Magnetite.. .... 5 Remainder... .. 4, 100-0 The rock is plainly an essentially normal dolomite rendered impure by the simple admixture of clastic grains of quartz and feldspar and by the presence of some silica and iron oxide, both of which may be of chemical origin. Upper Division—A specimen typically representing the chief phase of the upper member of the Altyn was collected at the 7,300-foot contour on the back of the ridge south of Oil City. The bed was situated about 100 feet vertically below the top of zone 6 of the cclumnar section. REPORT OF THE CHIEF ASTRONOMER 61 SESSIONAL PAPER No. 25a Microscopically this phase is like the one just described but is yet more strongly charged with water-worn, eminently clastic grains of quartz and feld- spar. The latter stand out conspicuously on the weathered surface; they vary from 0-25 mm. to 1-0 mm. in diameter. In this section the feldspars were determined as microperthite, microcline, orthoclase, and very rare plagioclase. One grain of the latter, showing both Carlsbad and albite twinning was found to be probably andesine, near Ab,An,. A few oolite-like grains of carbonate 0-5 mm. or less in diameter, a number of round grains of chert, and a very few small specks of magnetite complete the list of materials other than the general carbonate base. In grain and structure the base is practically identical with that of the specimens above described. Professor Dittrich has analyzed the rock (specimen No. 1326) with result as here noted :— Analysis of type specimen, upper Altyn formation. Mol. SO eres Nm ee cree ein area Ue Ne ta nr pa UN AM rer Sat Gh 25-50 437 JADE O55 4.) Gitbyh Cache ierCc hae US ae GEA A OC Es nA NE 2-25 023 e.03 Oy od CHO OO oD Blelvejell evel) (ele)) \elels ie) er ijehe ele peheliele ee 62 04 FeO . 38 -005 MgO.. 14-77 °369 CaO.. 21-65 °387 INGE OR pos sa heeihorc tors 86 014 KEOe PUERCO ea Coin CUA SPC) is ryemees elias Atos cee Slee Mee 1-27 013 H.O at 110°C... ; aia Le rele re Cala ole PUTS) Site es Vaewen a vc nhcter tarene an ees 12 eG H.O above 110°C. . BCC TIC Re CTE ORIEA CHE TOR ete 42 023 CO oe. BRC aC SCPE Mars MIME NE SURE K esas RUE Rls 32-03 729 99-87 SP a Te ietee eta clin mictcate anlar Verena na te een i eee be th Ak 2-768 Insolubtenin hydrochloric Aeidiy ise vicn veces eves seuiiet ale 29-21% Soluble in muatogtleric Ce FeO... oe ey nehe Pelelt efenineelMeletaieey lice! ees sels .elew ele! lee °95 Ou Se, ie oir eicleoreey Pes geceer BV at aur sic Gucte erste hate tans ee etaais'e Mpauemne tenets “19 CHORE ONS ei ere re Ee I AER a re esse Mot. go IV OR spares cree ieee stal ine evaisth ace eta Ment ere NOEL ee NGS Mala 14-29 The fact that sensibly all the lime is soluble shows that basic plagioclase can be present only in extremely small amount. The ratio of CaO to MgO in the soluble portion is 1-534:1, showing that here too the calcium carbonate is but slightly in excess of the proportion required for true dolomite. The approximate mineral composition of the rock has been calculated with the following result :— CALCIUM CATE DOMALC Mee or rainy eee ace ere SA oh Ore aT an aie NESE HERE ah 39-1 Wie ranGeaha) CAO NRT oo as Go Be Hole GK an AN SoLoG AomoGlaalageue 30-0 Ouartzgardgcher eps yaaess te a Me Hse ei HA us i era tiles IAW eNBa EAE aoa ico Ran el 14-7 Orthoclasemoleculors aie eee cee Sai et ee aS on en ae 7-2 AND EEGRNTOTEC We ig ct eee cee eH ae ae nent Sig BUINC een Cea mes cana TCO 7-3 Magnetite.. .. .. .. : 9 Remaind charset 8 62 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Comparison and Conclusions—Comparing the three analyzed specimens with all the others collected from the Altyn, as well as with the rock-ledges encountered during the different traverses, it appears probable that the average rock of the whole 3,500 feet of beds is composed of about 75 per cent of pure dolomite, about 4 per cent of free calcium carbonate, about 10 per cent of quartz and chert, and about 10 per cent of microperthite, orthoclase (with microcline), plagioclase (only a trace), magnetite, kaolin, and carbonaceous dust, with, possibly, a small proportion of hydrous magnesium silicate. No beds of pure calcium carbonate could be found anywhere in the section, nor any beds of ideally pure dolomite. Even in the most compact specimens examined a notable percentage of clastic quartz and feldspar never failed. Though these rocks are thus impure, like the Waterton formation, they may - conveniently be referred to as dolomites. One of the most noteworthy facts concerning all these beds, including both the Altyn and Waterton, is the constant size of grain.in the dolomitic base. (Figure 8.) The minute anhedra of carbonate everywhere range from 0-005 mm. to 0-03 mm. in diameter, with an average diameter a little under 0:02 mm. This is true, no matter what may be the size of the clastic quartz or feldspar. The quartz grains vary from scarcely discernible specks to small pebbles 5-0 mm. or more in diameter. As regards the relative amounts and individual size of these silicious materials, the Altyn formation is quite variable in composition. But its essential base of dolomite is remarkably uniform in grain and in composition. This contrast between carbonate base and enclosed clastic materials is worthy of close attention. The quartz grains in different phases of the forma- tion vary from those as small as the average grain of dolomite to those several million times greater in volume, while through all the thousands of feet of strata, the grain of the dolomite itself is rigidly held below an extremely low limit. In most of the slides the quartz and feldspar fragments are thus gigantic, compared to the granular elements of the base and, in most cases, there are very few silicious grains giving the full transition in size between the sand grains and the carbonate grains. Such transitions are to be seen but, as a rule, most of the silicious grains are enormously bigger than the carbonate grains. This steady contrast of size suggests very strongly that the mode of deposition of the quartz and feldspar grains was quite different from that of the enclosing carbonate. The former were unquestionably rolled and rounded by wind-action or under water and were then deposited from water-currents as mechanical precipitates. The purity and homogeneity of the carbonate base, its remarkably fine grain, and its perfectly regular microzcopic lamination of bedding all point to an origin in chemical precipitation. The sea-water must have been free. from mud, the shores furnishing pure sand to the undertow and marine cur- rents of the time. If the carbonate were the result of the mechanical breaking up of shells, coral reefs or older limestones, we should inevitably expect the detrital grains of carbonate to be much larger, or at least much more variable in REPORT OF THE CHIEF ASTRONOMER 63 SESSIONAL PAPER No. 25a size than the actual particles. There must have been changes of sea level or of depth of water, or changes in both during the accumulation of these 3,500 feet of sediment. It is virtually inconceivable that, throughout such changes, the size of carbonate particles broken off from either shells or bed-rock and brought hither by currents, should always average from 0-01 to 0-02 mm. in Ficure 8.—Diagrammatic drawing from thin section of typical sandy dolomite of the Altyn formation. Round (wind-blown?) grains of quartz (clear white) and much less abundant microperthitic feldspar (transverse lines) drawn to scale. The dots represent, on the same scale, the size of the extremely minute carbonate granules com- posing the matrix of the rock. Diameter of circle 4°5 mm. diameter and never reach diameters above 0-05 mm. or thereabouts. If the carbonate base were of detrital origin one should expect to find variations in its grain as he approached or receded from the source of detrital supply. Such variation is not to be found at any of the sections yet studied in the Altyn formation. So far as the pre-Altyn rocks are known there seems to have been in the adjacent Cordilleran region no magnesian limestones of anything like the volume required to furnish, from their mechanical disintegration, the material for the thousands of cubic miles of carbonate represented in the Altyn. The same may be said of the pre-Altyn formations underlying the Great 64 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Plains, for we doubtless have in the cores of the Belt mountains and Black Hills uplifts, average samples of such rocks as were washed by the sea waves during Altyn time. In some thin sections the carbonate is often balled up in spherical or spheroidal bodies, averaging from 0-25 mm. to 0-5 mm. in diameter. These sometimes have an obscure concentric structure, recalling oolite grains. More rarely an imperfect radial arrangement of the minute granules making up each spheroid, is discernible. In ali of the observed cases these granules accord in size with thoz:e making up the general base of the rock. Certain of the spheroidal bodies recall the ‘ coccoliths’ such as are precipitated by the action of decomposing albumen on the calcium sulphate of sea water.* There is no evidence that they are foraminiferal tests. Often associated with these carbonate concretions are fairly abundant spheroids of cherty matter, averaging about 0-25 mm. in diameter. These may be due to the silicification (replacement) of the carbonate spheroids, or they may also be due to direct chemical precipitation from sea-water. The former interpretation seems the more probable, therewith correlating these microscopic bodies to the manifestly secondary, large nodules of chert found at many horizons in the formation. The field and laboratory studies of the Altyn rocks seem, thus, to show that the dolomites and the carbonate base of the subordinate sandy beds are alike the product of chemical precipitation from sea-water. The same may be said of the massive, underlying Waterton dolomite as above described. The cause of the precipitation will be more fully discussed in chapter XXIII. on the theory of limestones. For the present it suffices to state, that the cause may possibly be found in the bacterial decay of animal remains on the sea-bottom. The ammonium carbonate generated during such decay reacts on the calcium and magnesium salts dissolved in sea-water, throwing down calcium carbonate and magnesium carbonate. The strong content of carbon- aceous matter in the Waterton dolomite and its occasional occurrence in certain phases of the Altyn formation, may represent the residue of animal carcases. Special note should be taken of the nature of the clastic feldspar. It is always remarkably fresh and is mostly a microperthite with typical characters. In view of the unusual nature of this dominant feldspar, it may be used as a sort of fossil in correlating the formation. On stratigraphic grounds it was concluded in the field that the Altyn formation is the equivalent of a part of the Creston formation in the Purcell range and of the Wolf grit and associated members of the Summit series in the Selkirks. The discovery that this special and far from common feldspar is an abundant constituent in all these forma- tions to a degree corroborates this correlation. The wonderful freshness of the feldspars suggests that these clastic frag- ments were derived from a terrane undergoing mechanical rather than chemical disintegration. One naturally thinks of an arid climate as supplying the *G. Steinmann, Berichte der Naturforschenden Gesellschaft, Freiburg i.B., Band 4, 1889, p. 288. REPORT OF THE CHIEF ASTRONOMER 65 SESSIONAL PAPER No. 25a necessary condition and recalls the observation of McGee, who deseribes parts of the Gulf of California as being floored with quartz and fresh feldspar sand washed into the Gulf, during the cloud-burst seasons, from the adjacent arid land.* In no eace is there any evidence of pronounced metamorphism of the sedi- ment. The tendeney of metamorphism would be rapidly to increase the grain of the rock and to obliterate the delicate structure of bedding. The persistence of the extremely fine grain and of thin bedding seems to show that we have the sediment searcely more changed from its original state than was necessi- tated in the act of consolidation. Fossils —Very abundant chitinous or caleareo-chitinous plates or films of highly irregular forms were found at a horizon about 975 feet below the top of the Altyn formation. These were seen at only one locality, namely, on the back of the ridge south of Oil City, at a level barometrically determined to be 6,875 feet above sea. In the well exposed ledges at that point thousands of the fragments can be readily laid bare by splitting the thin-bedded, silicious dolo- mite in which they occur. At least 200 feet of the series is, at intervals, characterized by the fragments. In spite of the formless nature of the frag- ments they were at once suspected to be of organic origin and to belong to the pre-Cambrian genus, Beltina, described by Walcott as occurring in the Greyson shales of the Belt mountains in Montana.t A collection of the fragments was sent to Dr. Walcott, who kindly determined them to have the essential features of Beltina danai. The resemblance of the material to that collected at Deep creek in the Belt mountains extended even to the character of the rock. No other species were discovered among the fragments nor did the formation prove fossiliferous elsewhere. At several horizons but particularly in the lowest member of the Altyn, large concentric concretions suggesting Cryptozoon were found but no evidence of their being of organic origin has been forthcoming. The Beltina horizon at Oil City must be close to that which had been found by Weller at Appekunny mountain, near Altyn. Reporting on his col- lection Dr. Walcott wrote:— ‘The mode of occurrence of the material is similar to that found in the Greyson shales of the Algonkian in the Belt mountains, Montana. Hun- dreds of broken fragments of the carapace of the crustaceans are distributed unevenly through the rock. Occasionally a segment or fragment of what appears to be one of the appendages is sufficiently well preserved to identify meee The repeated occurrence of the Beltina bed at three widely separated locali- ties shows their very considerable importance as a horizon-marker in this little known part of the Cordillera. The fossils themselves have intrinsic interest in representing one of the oldest species yet described. *W. J. McGee, Science, Vol. 4, 1896, p. 962. +C. D. Walcott, Bull. Geol. Soc. America, Vol. 10, 1899, pp. 201 and 235. = Bull. Geol. Soc. America, Vol. 13, 1992, p. 317. 25a—5 66 DEPARTMENT OF THE INTERIOR 2 GEORGE V,,A. 1912 APPEKUNNY FORMATION. The formation immediately and conformably overlying the Altyn limestone has been named the ‘ Appekunny argillite’ by Willis. His original description applies to these rocks as they crop out in the Boundary belt and it may be quoted in full:— ‘The Appekunny argillite is a mass of highly silicious argillaceous sediment approximately 2,000 feet in thickness. Being in general of a dark-gray colour, it is very distinct between the yellow limestones below and the red argillites above. The mass is very thin-bedded, the layers varying from a quarter of an inch to two feet in thickness. Variation is fre- quent from greenish-black argillaceous beds to those which are reddish and whitish. There are several definite horizons of whitish quartzite from 15 to 20 feet thick. The strata are frequently ripple-marked, and occasionally coarse-grained, but nowhere conglomeratic. An excellent section of these gray beds is exposed in the northeastern spur of Appekunny mountain, from which the name is taken, but the strata are so generally bared in the eliffs throughout the Lewis and Livingston [Clarke] ranges that they may be examined with equal advantage almost anywhere in the mountains. ‘The Appekunny argillite occurs everywhere above the Altyn lime- stone along the eastern front of the Lewis range from Saint Mary lakes to Waterton lake and beyond both northward and southward. Jt also appears at the western base of the Livingston range above Flathead valley | and is there the lowest member of the series seen from Kintla lakes south- ward to McDonald lake.’* At the eastern end of the South Kootenay pass the lower part of the Appe- kunny includes a 75-foot band of thin-bedded magnesian limestone which is identical with the staple rock of the Upper Altyn. Several other bands, each a few feet in thickness, are dolomitic sandstones and grits, quite similar to the beds of the Middle Altyn. The two great formations are thus transitional into each other. On the other hand, the top of the Appekunny is rather sharply marked off from the overlying Grinnell red beds. The formation as a whole is not exposed in any one section within the belt covered by the Commission map. The best exposures studied occur on the southern slope of King Edward peak and on the mountain slopes north and - south of Lower Kintla lake. A complete section was found on the ridge south of Oil City, and thus outside the area of the Commission map. The total thickness in these sections was estimated to be 2,600 feet. The dominant rock of the Appekunny is gray or greenish-gray and silicious, weathering lighter gray or more rarely light greenish-gray or light rusty brown. The content of silica is often so great that the rock might well be called an impure quartzite. As noted in Willis’ description the thin-bedded ‘ argillite’ is often interleaved with more massive strata of gray, whitish, and rusty-weathering *B. Willis, Bull. Geol. Soc. America, Vol. 13, 1902, p. 322. REPORT OF THE CHIEF ASTRONOMER 67 SESSIONAL PAPER No. 25a quartzitic sandstone. All these rocks are very hard, and were it not for the fissility incident to thin bedding, the formation would be exceptionally resistant to the forces of weathering. The alternation of the quartzites and more _ argillaceous beds is so common and the graduation of the one rock-type into the other is throughout so persistent that it has proved impossible to make a useful minute subdivision of the formation. In the section at King Edward peak the uppermost 200 feet are very thick-bedded and are composed chiefly of typical quartzite. In the sections between Oil City and the north end of Waterton lake at least 100 feet of blackish, red, and reddish gray shaly beds are inter- bedded with the magnesian limestones and sandstones at the base of the forma- tion. None of these types was noted in sections farther west, and, in its lower part at least, the formation seems to become more dolomitic or more ferruginous as it is followed eastward. Sun-cracks and ripple-marks, especially the former, were seen at many horizons from top to bottom of the Appekunny. No fossils have yet been found in it. Collecting all the information derived from the Boundary belt, a composite columnar section of the formation as exposed in the Clarke range, has been constructed and may be described in the form of the following table :— Columnar section of Appekunny formation. Top, conformable base of the Grinnell formation. 200 feet. egg poaded quartzite with subordinate interbeds of gray and rusty metar- gillite 2,025 “ Light gray to rather dark gray (dominant) silicious metargillite and quart- zite, weathering gray and rusty-gray, thin-bedded; many relatively mas- sive beds of whitish and rusty quartzite occur among the staple thin beds of the rapidly alternating metargillite and quartzite) sun-eracks and ripple-marks common. 75 “* Thin to medium-bedded, buff-weathering silicious dolomite. S00Ri Highly variegated, gray, green, reddish, and black metargillite and quart- zite, weathering in tones of brown, red, and gray; a few interbeds of buff dolomitic sandstone and grit; sun-cracks and ripple-marks. 2,600 feet. Base, conformable top of the Altyn formation. Thin sections of typical phases have been examined microscopically. They revealed an even higher percentage of free quartz than was in the field sus- pected to characterize the rock. This mineral occurs in very minute angular individuals from 0-005 mm. or less to 0-03 mm. in diameter, with an average diameter of about 0-01 mm. No certain trace of an originally clastic form was anywhere observed. The quartz is intimately intermixed and interlocked with a nearly colourless to pale-greenish mineral, which, on account of the extremely small dimensions of its individuals, is difficult to determine. The single refrac- tion is notably higher than that of quartz; the birefringence is apparently low but, in reality, may be high, the common low polarization tints being due either to the section’s passing across the optic axes or to superposition of differently Pha—5i alee ¢ 68 DEPARTMENT OF THE INTERIOR 2 GEORGE V,, A. 1912 orientated crystals in the exceedingly fine-grained rock. Many fine shreds and thin scale¢ of similar material with needle-like cross-sections, have all the optical characteristics of sericite, and it seems highly probable that it is this mineral which forms the base of the dominant rock. A whole thin section may, then, be made up of a homogeneous, intimate mixture of quartz and sericite with accessory grains of iron oxide; or, as is more commonly the case, the slide shows a well-defined banding representing original bedding. In the latter type of section the bedding is marked by alter- nation of more quartzose and more sericitic material or, yet more clearly, by long lines of limonitic and carbonaceous particles. A few grains of ilmenite or magnetite seem never to fail, but no other accessories, such as feldspars, have been observed. The specific gravity of four specimens, taken to represent the average types of these quartz-sericite rocks and thus the greater volume of the whole formation, varies from 2-708 to 2-760, with a mean value of 2-740. This comparatively high density shows that the sericite is fairly abundant. The almost complete recrystallization of the original rock is evidenced in the intimate interlocking of the quartz and micaceous mineral and in the entire absence of amorphous argillaceous matter. The sericite is slightly more developed in the bedding plane than elsewhere, thus somewhat aiding the fis- sility of the rock in those planes. On the other hand, many scales of this mineral have their longer diameters developed at high angles to the bedding planes. Rarely is there a marked sheen on any surface of a hand specimen, nor has true schistosity been developed except in a few very local areas. The recrystallization of the typical rock is clearly the result of slow molecular rearrangement incident to age-long deep burial without true dynamic meta- morphism. Although evident only after microscopic study, this change is so pronounced that it is scarcely correct to speak of the normal phase of the Appekunny as an argillite at all. It is as much a crystalline rock as is a granitoid gneiss. We shall see that the same difficulty of nomenclature adheres to the description of many thousands of feet of beds, in each of the Lewis, Galton, and Summit series. It is convenient to have a term to represent these argillites, recrystal- lized, yet neither hornfelses (due to contact metamorphism), nor true mica slates or schists (due to dynamic metamorphism with development of notable cleavage or schistosity, generally cutting across bedding planes). For such once- argillaceous rocks, recrystallized merely by deep burial the name ‘ metargillite’ will be used in the present report. The meaning of the term may be made clear through a comparison with the names now in general use for pelitic rocks. An unconsolidated pelite is a clay or mud. If consolidated but not extensively recrystallized, it is an argillite. A thin-bedded, unaltered argillite which readily splits along the bed- ding planes because those are planes of original weakness, is a shale. If recrys- tallized during dynamic metamcrphism only to the extent that easy cleavage following the planes of similarly orientated microscopic mica plates is developed, the rock is a slate. If an argillite becomes phanerocrystalline and foliated REPORT OF THE CHIEF ASTRONOMER 69 SESSIONAL PAPER No. 25a through dynamic metamorphism, it is a quartz-mica schist or a quartz-feldspar schist or gneiss. If an argillite has been more or less completely recrystallized by thermal action on igneous contacts it is a hornfels. If, finally, an argillite retaining the bedding structure has been essentially recrystallized by deep burial and without being affected by direct magmatic influence or by the notable development of cleavage or schistosity, it may be called metargillite. At no point within the Boundary belt was true argillite (shale or slate) found in the Appekunny formation. Everywhere the once-pelitic phases belong to the metargillite type as just defined. On the mountain slopes running up eastward from the Flathcad valley both metargillite and quartzite have been sheared and cleaved, the former giving local phases of slaty metargillite. The erystallinity never rises to the degree of true mica schist. GRINNELL FORMATION. The Grinnell formation was named by Willis and described by him thus:— “A mass of red rocks of predominantly shaly argillaceous character is termed the Grinnell argillite from its characteristic occurrence with a thickness of about 1,800 feet in mount Grinnell. These beds are generally ripple-marked, exhibit mud-cracks and the irregular surfaces of shallow water deposits. They.appear to vary considerably in thickness, the max- imum measurement having been obtained in the typical locality, while elsewhere to the north and northwest not more than 1,000 feet were found. It is possible that more detailed stratigraphic study may develop the fact that the Grinnell and Appekunny argillites are really phases of one great formation, and that the line of distinction between them is one diagonal to the stratification. The physical characters of the rocks closely resemble those of the Chemung and Catskill of New York, and it is desirable initially to recognize the possibility of their having similar interrelations. “The Grinnell argillite outcrops continuously along the eastern side of Lewis range and its spurs, occurring above the Appekunny argillite and dipping under the crest of the range at the heads of the great amphi- theaters tributary to Swift Current valley. About the sources of the Kennedy creeks it forms the ridge which divides them from Belly river. Mount Robertson is a characteristic pyramidal summit composed of these red argillites. The formation occurs in its proper stratigraphic position between the forks of Belly river and west of that stream in the Mount Wilson range of the Canadian geologists, the northernmost extremity of the Lewis range; and it dips westward under the valley of Little Kootna creek and Waterton lake. On the western side of Livingston [Clarke] range the Grinnell argillite was recognized as a more silicious, less con- spicuously red or shaly division of the system, occurring about Upper Kintla lake.’* +B. Willis, Bull. Geol. Soc. America, Vol. 18, 1902, p. 322. 70 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 The 4&# mation is admirably exposed on the southwest side of King Edward Peak, which overlooks the Flathead valley about three miles north of the Boundary line. In this section the total thickness is 1,600 feet, distributed as follows: Columnar section of Grinnell formation. Top, conformable base of the Siyeh formation. 355 feet.—Thin-bedded, red metargillite with intercalations of red, quartzitic sand- stone. 20s Flow of basic, amygdaloidal lava. TOU? Thin-bedded, red metargillite. 100 “ Thick-bedded, gray metargillitic quartzite, weathering light rusty brown. 1,050 “ Thin-bedded, red to reddish gray metargillite and quartzitic sandstone. 1,600 feet. Base, conformable top of the: Appekunny formation. Sun-cracks and ripple-marks are common in all the sedimentary members. In a section on Oil creek the ripples at one horizon measured from four to twelve inches from crest to crest, indicating currents of great power, such as the heavy tidal rips occurring in the Bay of Fundy and other estuaries of the present day. If these ripples were caused by wind-wave current, the waves must have been of very large dimensions. Under the microscope the sandstone specimens are seen to be composed essentially of rounded quartz grains, averaging 0-25 mm. in diameter. Many of the grains are secondarily enlarged and to such an extent that the rock has the fracture and the strength of true quartzite. A small amount of amorphous, apparently argillaceous matter, tinted with the red oxide of iron, forms the rest of the cement. No feldspar was seen in this section. The metargillites are made up of a very compact mass of sericite, quartz, chlorite, and abundant iron oxide. In one thin section minute crystals of a pale brownish carbonate, probably dolomite, are distributed through the mass. The carbonate seem also to be an original constituent and may have been chemically precipitated along with the mechanically deposited mud, the domin- ant original component of these beds. The specific gravity of a type specimen of the red quartzite is 2-678. The specific gravities of two specimens of the metargillite are 2-740 and 2-757. The average for the whole formation is about 2-725. The amygdaloid is a dark green-gray, compact rock, which both macroscopic- ally and microscopically, is similar to non-porphyritie phases of the overlying Purcell lava (described in chapter IX). It is much altered, but minute, thin tabular crystals of labradorite with the same abundance and mutual arrange- ment which this essential mineral has in the Purcell formation amygdaloid, still represent the original microphenocrysts. The base was once glassy but is now mainly composed of the usual secondary chlorite, quartz, and calcite. Numerous small erystals of original ilmenite are now represented only by pseudomorphs of REPORT OF THE CHIEF ASTRONOMER Fal SESSIONAL PAPER No. 25a yellowish leucoxene. The pores of the rock seem to be entirely filled with deep green chlorite. The rock is too greatly altered to afford a useful analysis but it is evidently a common type of basaltic lava. SIvYEH FORMATION. In Willis’ original description of the ‘ Algonkian’ rocks of the Lewis range, the following concise account of the Siyeh formation occurs: ‘Next above the Grinnell argillite is a conspicuous formation, the Siyeh limestone, which rests upon the red shales with a sharp plane of distinction, but apparently conformably. The Siyeh is in general an exceedingly massive limestone, heavily bedded in courses 2 to 6 feet thick like masonry. Occasionally it assumes slabby forms and contains argillace- ous layers. It is dark blue or grayish, weathering buff, and is so jointed as to develop large rectangular blocks and cliffs of extraordinary height and steepness. Its thickness, as determined in the nearly vertical cliff of mount Siyeh, is about 4,000 feet. ‘This limestone offers certain phases of internal structure which may be interpreted as results of conditions of sedimentation or as effects of much later deformation. Some layers exhibit caleareous parts separated by thin argillaceous bands, which wind up and down across the general bedding and along it in a manner suggestive of the architectural ornament known as a fret. It is conceived that the effect might be due to concretionary growths in the limestone, either during or after deposition, or to horizontal compression of the stratum in which the forms occur. Other strata con- sist of fragments of calcareous rock from minute bits up to a few inches in diameter, but always thin, constituting a breccia in a crystalline limy cement. Again, other strata consist of alternating flattish masses of ealeareous and ferruginous composition, which rest one upon another like eards inclined at angles of 30 to 45 degrees to the major bedding. At times the lamination is so minute as to yield a kind of limestone schist. These internal structures suggest much compression, but the apparent effects are limited by undisturbed bedding planes, and it is possible that the peculiar- ities are due to development of concretions and to breaking up of a super- ficial hard layer on the limestone ooze during deposition of the beds. Walcott has described similar structure as intraformational conglomerates. ‘The Siyeh limestone forms the mass of Mount Siyeh, at the head of Canyon creek, a tributary which enters Swift Current at Altyn from the south. It constitutes the upper part of all the principal summits of Lewis range north of Mount Siyeh, including Mounts Gould, Wilbur, Merritt, and Cleveland. It extends beyond Waterton lake westward into the Livingston [Clarke] range and forms the massive peaks between Waterton and North Fork drainage lines. Above Upper Kintla lake it is sculptured in the splendid heights of Kintla peak and the Boundary mountains.’* *B. Willis, Bull. Geol. Soc. America, Vol. 13, 1902, p. 323. 72 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 The Siyeh is the great cliff-maker of the Front ranges. Quite apart from the fact that it is capped by the resistant Purcell Lava, the limestone is itself strong enough to stand up in precipices thousands of feet in height. (Plate 9.) Among the many admirable exposures one of the best within the Boundary belt is that at the head of Starvation creek canyon; this section is typical of the formation as it occurs in both the Clarke and Lewis ranges. The formation is notably homogeneous for hundreds of feet together; yet,. as shown in the columnar section, it is divided into five zones of contrasted lithological character. Columnar section of Siyeh formation. Top, base of the Purcell Lava. 150 feet—Medium to thin-bedded, reddish metargillite with subordinate, thin inter- beds of buff-weathering, magnesian metargillites; cun-cracks, rain-prints-. and ripple-marks abundant. 950 “ Gray and greenish, thin-bedded, often calcareo-magnesian metargillites; weathering buff (dominant), fawn, and gray; sun-cracks, rain-prints and ripple-marks common. 100 “ Gray, concretionary, silicious and non-magnesian limestone weathering light gray, with a five-foot band of buff-weathering dolomite at thirty feet from the top. 2,000 “‘ Massive, thick-bedded, dark gray or dark bluish gray, impure magnesian limestone, weathering buff; thin interbeds of dolomitic metargillite, weathering buff, and a few thin beds of gray sandstone. Molar-tooth structure characteristic of limestone; metargillites bear sun-cracks and, rarely, obscure ripple-marks. 900 “* Thin to thick-bedded, light to dark gray and greenish gray calcareo-mag- nesian metargillites and quartzites, weathering fawn and buif, with a few imnterbeds of buff-weathering dolomite without molar-tooth structure. 4,100 feet. Base, top of the Grinnell formation, Notwithstanding their lithological variations, the strata form a natural unit; the peculiar buff tint of the weathered surface contrasting with the deep browns and blacks of the Purcell Lava and with the strong purplish red of the Grinnell formation below, is a common feature for most of the strata in the Siyeh. The uppermost beds weather reddish but they are so intimately inter- leaved with buff-weathering strata that a clean-cut separation of the red beds is impossible. It has thus appeared best to follow Willis in including all the strata between the Grinnell and the Purcell Lava under the one formation name. It is, on the whole, a single magnesian group. A second general charac- teristic is the massiveness of the beds. This is most prominent in the 2,000 feet of dolomite and limestone composing the middle part of the formation. The table and Willis’ statement afford a sufficient general description of the strata in this section. There are, however, certain structures in the buff magnesian limestone and the thick band of gray limestone which merit special notice. At the outcrop of the buff rock the observer’s eye is struck with a repeated colour variation in the rock. The cause is speedily apparent. The caleareous constituent of the rock is seen to be segregated, sometimes irregul- PLATE 9. Mount Thompson, seen across Upper Kintla Lake ; summit 5,500 feet above lake. Illusttates massive character of Siyeh formation. 25a—vol. ii—p. 72. REPORT OF THE CHIEF ASTRONOMER 73 SESSIONAL PAPER No. 25a arly, sometimes systematically. The buff-tinted (weathered) general surface of the ledge is thus variegated with many small masses of pure light gray to bluish gray limestone. These masses are in the form of roundish nodules and pencils, flat lenses, or irregular stringers of no definite shape. They are essen- tially composed of pure calcite; they effervesce violently with cold dilute acid. As in so many dolomites the calcareous segregation is often quite unsystematic, Kicure 9.--Section showing common phase of the molar-tooth structure in the Siyeh formation. The calcitic segregations are lenticular and stand perpendicular to the plane of stratification, as shown by the metargillitic interbeds (M). The ae layer of limestone is two feet thick. Locality, north fork of the Yahk iver. but, on the average, it is definitely related to the two master planes of structure in the limestone. Where the rock is uncleaved, the bedding-plane has been selected as the favoured locus of growth of the segregation. The stratification may thus be marked by many small, independent lenses of lime carbonate com- pletely surrounded by the magnesian matrix. There is transition between such isolated lenses and entire, uninterrupted beds of gray limestone conformably intercalated in the buff magnesian rock. Such beds may in many cases be due to original sedimentation. The conclusion that the lime-carbonate lenses, pencils, and irregular bodies, and even some of the continuous bed-like masses, are due to secondary segrega- tion within the dolomite, is clearly upheld by the relation of another kind of lime-carbonate partings. These were long ago observed by Bauerman and later by Willis. In localities where the dolomite has been specially nipped and 74 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 squeezed or somewhat sheared, a cleavage was developed. Since the dip of the bedding is generally low, this cleavage runs at high angles to the plane of stratification. The cleavage planes have permitted easy passage to circulating waters. Owing to their activity, there has been a wholesale segregation of the more soluble lime-carbonate in the cleavages which have been thereby healed so as to restore much of the rock’s original strength. In regions of formerly strong lateral pressure the rock is now converted into a laminated rock composed of thin, alternating, more or less continuous layers of pure lime-carbonate. These layers are highly inclined to the stratification planes and often run parallel to cleavage planes in argillites above and below the limestone and, like those planes, may be crumpled. (Plate 10 and Figure 9.) That the gray calcareous partings are due to secondary chemical deposition is shown also by the fact that where the original rock was argillaceous or sandy, these impurities remain entirely within the magnesian parts of the rock. On a weatherea surface the latter may be quite gritty to the feel while the lime-carbonate partings are smooth and marble-like. Bauerman described this rock as ‘an impure limestone, in which the carbonate of lime is intermingled with argillaceous patches in folds resembling the markings in the molar tooth of an elephant.’* This appearance is most striking on the weathered ledges, the stringers of the more soluble, gray calcite locating numerous channels and pits which are separated by the brownish, pro- jecting ribs of the more resistant magnesian and silicious parts. Using Bauerman’s simile, the structure may be called the ‘ molar-tooth’ structure, whereby will be understood, in general, the internal modification of the original limestone by the secondary segregation of the calcium-carbonate. The term will also be used for the very common case where the weathered surfaces do not show the chance imitation of a worn molar-tooth; the last is best shown in the cleaved phases. The name is thus conveniently generalized as it may then be applied to the concretionary limestone even when cleavage has not been developed. The structure is of importance as an aid in the recognition of the Siyeh formation over great distances. Under the microscope the contrast between the calcitic and magnesian parts of the molar-tooth rock is marked. The calcite, light gray in the hand- specimen, is colourless in thin section. It forms a compact aggregate of poly- gonal, sometimes interlocking grains varying in diameter from 0-005 mm. to 0-02 mm. and averaging about 0-01 mm. Very seldom, if ever, do these grains show the rhombohedral or other crystal form. A few minute cubes of pyrite are embedded in the mosaic, but, otherwise, the lenses and stringers are made up of practically pure carbonate. The buff-weathering, main part of the rock is sharply distinguishable under the microscope. It has a decided, pale yellowish-gray colour and a mixed composition. Anhedra and rhombohedra of carbonate, which is doubt- less high enough in magnesia to be called dolomite, form more than half of the * Report of Progress, Geol. Nat. Hist. Survey of Canada, for the years 1882-3-4, Pt. B, p. 26. PLaTE 10. Sheared phase of Siyeh limestone, Clarke Range. Light parts highly magnesian ; dark parts nearly pure calcium carbonate. Three-fourths natural size. Sheared phase of dolomitic lense (weathered) in Kitchener formation, at Yahk River. Illustrates molar-tooth structure ; sunken parts, calcium carbonate, and projecting ribs, silicious magnesian limestone. Two-thirds natural size. 25a—vol. ii—p. 74. REPORT OF THE CHIEF ASTRONOMER 75 SESSIONAL PAPER No. 25a volume. The rhombohedral and subrhombohedral crystals average about 0-01 mm. in diameter. They are embedded in a base which is partly composed of numerous anhedral granules of the same carbonate but of much smaller size, with diameters varying from 0-001 mm. or less to 0-005 mm. A few minute angular grains of quartz and feldspar and a few dust-like particles of carbon and pyrite are associated with the carbonate. The remainder of the base is a colourless, amorphous to suberystalline cement whose diagnosis is extremely difi- cult. Its single refraction is low and its double refraction either nil or extremely faint. These properties are like those noted for the silicious base of the Altyn dolomite. The chemical analysis shows a content of silica and alumina con- siderably in excess of the amounts required for the little clastic quartz and feldspar in the rock. It thus appears that the cement carries both silica and alumina, and it must carry the combined water. True argillaceous matter may be present, as well as amorphous and chalcedonic silica. In none of the thin sections could sericite be detected. The specific gravity of six type specimens varies from 2-657 to 2-760, with an average of 2-702. These low values indicate the impurity of the carbonate. The lightening of the rock must, in largest amount, be attributed to the cement of the magnesian part. Owing to the extensive and highly irregular rearrangement of the car- bonates in the molar-tooth rock it is not easy to secure a specimen which shall faithfully represent its average composition. A specimen approximating to this ideal was taken from the cliffs of Sawtooth ridge, 1-5 miles east of Lower Kintla lake. Material from this specimen was so selected as to contain magnesian base and calcitic segregation in about their average proportions in nature, and the powdered mixture (specimen No. 1306) was analyzed by Pro- fessor Dittrich. The result is as follows: Analysis of Siyeh impure limestone. Mol SiO, 35-58 593 Al,O, 3-40 033 Fe 1-56 ‘ aor 87 012 MzgO 10-09 +252 CaO 19-72 +352 Na.O 51 008 AOA aera teiclie Skate ara s clea eecttinee Masa cenaetcier nee plage nobararats 1-21 013 IEEO)-cKe UM Cpe ie Oe eamery learners Soar seekers rubrtc RECA rok ien 17 senses TETAO) Fal sah oed OS Sete eee eminence ania te Sete co eey bce 2-93 163 C Ore et are ean ta Saat cars tgses nick Gyre Vas tet che ea uaa a Cone edad a eS OH Ligh 23-80 +541 99-87 Sp. gr. 2-741 Portion imsoluble im™hydrochlorice acide. + 5-0 ecu. seta cen ees 40-69% Portion soluble in hydrochloric acid: IDE Oe ee ed no Ps Lobe Oa) oe Monte TOC ana POC e son Mato mire rier 2-17 CEOS Thee Ge laa b dao oe oo ao On MHps tone Pha Bneie elie. «ttn Veto a ciemsetes ice 19-76 76 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 This analysis cannot be calculated quite as readily as those of the Altyn dolomites; the alkalies are here not assignable with certainty to definite feld- spars. For the purpose of comparison, however, the same method of calculation has been applied here, giving a ‘norm’ wherein the soda is assigned to the albite molecule and the potash to the orthoclase molecule, just as the alkalies are assigned in calculating the ‘norms’ of igneous rocks. The carbonates have been calculated directly from the analysis of the soluble portion. The results are given in the following table, which shows the ‘mode’ for the rock as far as the car- bonates are concerned, the other constituents being more arbitrarily treated: Calcium carbonate.. . 35 Magnesium) Carbonate’ ss isic cis ccst cha aise coun, PALMS eee esi hed < ORO 18 SUT Cae a aL Nhe Ne in hh aol eu yesics aieaa bate, ACN ay Hea) ch gn aS 28. Orthoclase molecules: .22) 5" -2) opus ee 7 Albite molecule.. .. ; 4 Magnetite. . 2 Remainder.. . 3 The proportions of the carbonates correspond to 41-2 per cent of normal dolomite and 12-9 per cent of free calcium carbonate. This excess of calcium carbonate is probably not due to its having been introduced into the molar-tooth rock from other beds. The magnesian portions of the molar-tooth rock effervesce somewhat with cold, dilute acid; it seems simplest to believe that the calcium carbonate was there originally in excess and dates from the time of the deposition of the sediment. The two carbonates together are seen to make up about 54 per cent of the rock. The high percentage of water (above 110°C) is of interest as showing that this metamorphic agent, even at the present time, is enclosed in sufficient amount to explain the solutional effects illustrated in the molar-tooth structure. The content of carbon, low as it is, is partly responsible for the normally dark tint of the fresh rock. Different as the Siyeh dolomite limestone and the Altyn dolomite are in field-habit, the two types are yet similar in several important respects. In each the carbonate base has a remarkably fine and homogeneous grain, with no suggestion in either case that the carbonate is of clastic origin. In each case the dolomitic grains tend to assume the rhombohedral form. Their average diameter is sensibly identical in size with that of the average calcite grains composing the lenses, pencils, and stringers of the molar-tooth rock. This average diameter is also practically equivalent to that characterizing the gran- ules which compose each of the egg-like bodies forming occasional thin beds of oolite in the Siyeh and neighbouring formations. There is no doubt of the chemical origin of the latter, nor can there be doubt that the calcitic partings of the molar-tooth limestone were gradually crystallized out from water solu- tions. It would seem next to incredible that these three associated rocks, characterized by the same average size of constituent carbonate particles, could have in two eases an origin in chemical precipitation and. in the third, an REPORT OF THE CHIEF ASTRONOMER TS SESSIONAL PAPER No. 25a origin in the deposition of land detritus on the sea-floor. A study of all the available facts has, thus, forced the writer to the belief that the huge Siyeh and Altyn formations are chiefly the product of long continued throwing down of ealecium and magnesium carbonate from sea-water, from which there was a likewise slow deposition of silicious muddy matter brought from the lands. The molar-tooth structure of the Siyeh is secondary and was developed after burial. SHEPPARD FORMATION. ° General Description—Conformably overlying the Purcell Lava in the Clarke and Lewis ranges is a group of strata which has been named the ‘Sheppard quartzite’ by Willis. He speaks of it as belonging to a ‘distinctly sandy phase of deposition. . , .a@ quartzite which is very roughly estimated to have a flielcnese oi 700 feet, It forms the crest of Lewis range in the vicinity of Mount Cleveland and Sheppard Glacier between Belly river and Flattop mountain [type locality]. It has not been studied in detail but is recognized as a distinct division of the series.’* The lithological character of the beds oceurring in the Boundary sections and equivalent to the strata at Willis’ type locality differs somewhat from the character stated in his brief description. This lack of accordance may possibly be explained through actual differences in the beds as they are encountered at different points along the axis of the Clarke range. It may be noted, how- ever, that the present writer, during a rapid traverse across the Lewis range via the Swift Current Pass, found that there the beds of the Sheppard forma- tion are extremely like those studied in the Boundary belt. The staple rock of the Sheppard is not easy to diagnose in the field. It was only after micro- scopic study that one could be sure of the true nature of the sediment. Its colour, compactness, and general habit are those of an impure, flaggy quartzite. The thin section shows that the rock is largely composed of carbonate (dolom- ite) and that quartz occurs as minute grains rather evenly distributed through the mass of carbonate. The staple rock of the Sheppard is, thus, in the Boundary belt and probably also farther south, a silicious dolomite or dolomitic quartzite. More typical quartzite occurs as a subordinate constituent of the formation, as shown in the following columnar section of the formation where exposed just north of the Boundary monument on the Great Divide:— Columnar section of Sheppard formation. Top, conformable base of Kintla formation. 580 feet—Thin-bedded, light gray, highly silicious dolomites, weathering bufi—a homogeneous member occasionally concretionary; some of the more silicious beds 2pproximating magnesian quirtzice. 20° Reddish, interbedded quartzite and silicious argillite. 600 feet. Base, conformable top of Purcell Lava. *B. Willis, Bull. Geol. Soc. America, Vol. 13, 1902, p. 324. 78 DEPARTMENT OF THE INTERIOR 2 GEORGE V,, A. 1912 * At the head of Starvation canyon the section is slightly different :— Top, conformable base of Kintla formation. oe feet —The staple, thin-bedded, buff-weathering silicious dolomite. Basic, amygdaloidal lava. 2 — Medium to thin-bedded reddish and gray, interbedded sandstone and argillites, with thin intercalations of buff-weathering dolomitic rock. 585 feet. Base, conformable top of Purcell Lava. Of the two columnar sections the former is to be regarded as the more typical for the Boundary belt. The intercalated bed of lava represents a quite local outflow, not found in sections a few miles to the eastward, nor anywhere in the Galton range. The basal red beds of the formation are similar in character to the upper- most strata of the Siyeh formation and are like common phases of the Kintla. Ripple-marks and especially sun-cracks are common here as in all other strata below the base of the Kintla formation. Under the microscope the most common rock of the Sheppard is seen to be a highly impure, very finely granular, homogeneous mass of carbonate. It occurs in the form of pale brownish grains, varying from 0-005 mm. or less to 0-03 mm. in diameter and averaging about 0-02 mm. The larger grains often show rhombohedral outlines. These are enclosed in a fine-grained base of anhedral carbonate grains, quartz, sericite, and probably feldspar fragments of minute size. Along with these fairly determinable constituents the base carries a con- siderable amount of colourless, nearly isotropic material, which seems to be identical with the cement found in the Siyeh limestone and certain phases of the Altyn dolomite. As in the latter rocks, this material must carry much of the combined water, which here forms nearly two per cent of the whole rock. The specific gravity of the dominant phase, as represented in three fresh specimens, ranges from 2-695 to 2-785. A type-specimen (No. 1301) collected at the head of Starvation creek, has been analyzed by Professor Dittrich, with result as follows: Analysis of Sheppard wmpure dolomite. Mol SiO, 24-61 410 Al,O, 6-84 067 FeO, +58 004 FeO 2-01 028 MgO 13-34 334 CaO 19-14. +342 Na,0 Rte ae -62 -010 HO at 110°C... UNA Ne Ree op SALES Side ae orn aa pc 24 ae H.O above 110°C... Be Oe ND eyes inten feu eae SEN ee oM EERE enero es 1:76 -098 COs. NRE See aE xr ioe vio aa A tea ig os Ge oe 28-89 656 100-10 SS PoSg Oars tarecerge resist lsiate vaybapne sc kereu rc Roa heier archer ulstclans meant ameynuaale ages 2-779 REPORT OF THE CHIEF ASTRONOMER 72 SESSIONAL PAPER No. 25a Portion insoluble in hydrochloric acid.. .. .. .. .. .. .. .. .... 32-23% Portion soluble in hydrochloric acid: CO OF iae re eee dS Setar eh ees 2-36 CAOR Ss ee eran eleal ihe ticbios. Stabler tre Wad Maks od WE 18-86 Nr eae ees A Sates tor te cg Rede SO pS ENC St, 4-08 The carbonates could be rather closely calculated if it were known how much of the ferrous iron is present in the sideritic molecule. Since the carbon dioxide is no more than sufficient to satisfy the lime and magnesia of the soluble portion, it is probable that iron carbonate is present in but very small amount. A definite assignment of the alkalies is here impossible. The soda is arbitrarily assigned to the albite molecule, although it is possible that paragonite is present. A partial calculation yields the following result: Waleiumpcanbonatesca ccs B o =I REPORT OF THE CHIEF ASTRONOMER 10 SESSIONAL PAPER No. 25a These figures cannot, of course, represent the actwal composition or ‘ mode’ or the rock excepting as regards the carbonates. The calculation has some value, however, in facilitating the chemical comparison of the molar-tooth rock in the Galton series with that in the Lewis series. A reference to the table. col. 2, showing the ‘norm’ of the latter rock, indicates how nearly equivalent the two rocks are in chemical composition. In both cases we are dealing with a silicious, strongly dolomitic limestone of peculiar history and structure. In analysis B an appreciable amount of carbon was determined; here again the carbonaceous matter largely controls the dark tint of the fresh molar- tooth rock, which decolourizes before the blow-pipe. The specific gravity of three specimens of the molar-tooth rock varies from 2-670 to 2-748; that of four specimens of the metargillites in the formation, from 2-620 to 2-739. The average of all seven specimens is 2-700. GATEWAY FORMATION. A striking difference in the lthological character of the Lewis and Galton series is to be found in the nature of the beds conformably overlying the Purcell Lava in the respective ranges. We have seen that, in the Clarke and ‘Lewis ranges, the Sheppard formation, occupying this position, is a homogene- ous silicious dolomite and that it is overlain by the red beds of the Kintla. In the Galton range the beds intervening between the Purcell Lava and the red beds, equivalent to the Kintla, have a much greater total thickness than the Sheppard and a quite different composition. These strata are well exposed on the heights east of Gateway and overlooking Tobacco Plains; they may be grouped under the name, Gateway formation. Its total thickness was found to be about 2,025 feet. It includes two members of unequal strength. The lower member resting immediately upon the Purcell Lava contains beds which at once suggest possible identity of origin with the Sheppard. This correlation is so important that a specially detailed columnar section of the member is here noted. It was made on the basis of field sections along good exposures north of Phillips creek. Columnar section of Gateway formation (lower part). Top, base of 1,850-foot member. 5 feet—Massive, light gray dolomite, weathering buff and brown. AEE Massive, light gray quartzite. Gis Light gray magnesian and ferruginous limestone, weathering rusty brown. il Tua ea leak light gray quartzite. Grey Highly silicious, gray metargillite. snc Thin-bedded, gray dolomite weathering buff. D0 orcs Thick-bedded, hard, light gray, often cross-bedded and _ ripple-marked quartzitic sandstone. 20 eee din bedeed, concretionary, light gray dolomite, weathering strong buff aad Town. fl) Massive, dark gray, coarse, feldspathic sandstone, bearing locally lenses of grit and fine conglomerate one to two feet thick. 125 feat, Base, conformable top of Purcell Lava. 108 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Apart from the concretions found in certain layers, the limestones are in many respects similar to the staple phase of the Sheppard formation. The high specific gravity of some specimens, 2-826 to 2-871, shows that they are very high in magnesia or iron and probably approximate ideal, though some- what ferruginous dolomite. In any case all the carbonate bands are rich in magnesia. The concretionary structure noted in the thickest dolomitic stratum is a constant feature but is not always typically developed. Though the con- cretionary masses strongly resemble type specimens of Cryptozoon, there seems to be no reason to regard them as of other than inorganic, metamorphic origin. They are spheroids or ellipsoids composed of dolomite in concentric layers separated by thin laminae of cherty silica. The diameter of the bodies varies from a few inches to a foot or more. (Plate 14, B.) Similar, though smaller con- cretions were found in the basal beds of the Sheppard formation in the Clarke range. The upper member was estimated to be 1850 feet thick. It is a fairly homogeneous mass of thin-bedded, highly silicious metargillite, interstratified with subordinate, more or less sericitic metasandstone. On a fresh fracture both rock types are generally light gray or greenish gray, the metargillite naturally being of somewhat darker tint. The weathered surface may be gray, brownish gray. rarely red or reddish brown. The member is more ferruginous toward the top. Ripple-marks, rill-marks, sun-cracks, and casts of salt-crystals up to 2 em. or more in diameter, are all exceedingly common throughout this member. The salt-crystal casts were not found in the lower member. Under the microscope these rocks show great similarity to the chief phases of the MacDonald formation. Im all the slides, though especially in the more quartzitie types, feldspar is seen to be present. Orthoclase, microperthite, and plagioclase (probably andesine) form a considerable percentage of the clastic grains. A few broken zircons and tourmaline crystals were observed. Sericite, chlorite, and secondary quartz have replaced the original argillaceous matter. The specific gravity of seven type specimens varies from 2-643 to 2-701, with an average of 2-676. The average for the formation as a whole is about 2-680. The stratigraphic position, chemical composition, and occasional concre- tionary structure of the lower, dolomitic member are features directly correlat- ing that member with the Sheppard formation of the eastern ranges; the Sheppard thus thins rather rapidly to the westward. The thick upper member of the Gateway carrying abundant salt-crystal casts, is almost certainly of contemporaneous origin with the lower part of the Kintla and, like the Kintla, was doubtless deposited as a continental deposit in an arid climate. PHILLIPS FORMATION. The Gateway beds are specially ferruginous toward the top, where they gradually pass into a still more ferruginous mass of sediments. From its occurrence on two summits about two miles north of Phillips creek, this assem- REPORT OF THE CHIEF ASTRONOMER 109 SESSIONAL PAPER No. 25a blage of strata may be called the Phillips formation. The exposures are not extensive and the formation crops out nowhere else in the Boundary belt. Two different traverses covered the formation; on both occasions, because of bad weather, the writer was not able to make a thorough examination of these beds. The essential facts of the lithology were obtained but it is not known whether these strata or those of the overlying Roosville formation are fossili- ferous. The Phillips consists, for the most part, of about 550 feet of dark, purplish or brownish red, fine-grained to compact metargillite and metasandstone in alternating thin beds. At the base three massive beds of gray quartzitic sand- stone, respectively four, ten, and twenty feet thick, are intercalated. Sun-cracks and ripple-marks are again plentiful. No salt crystal casts were found, though they might, on more prolonged search, be found. Under the microscope, speci- mens of the red rocks proved to be always highly silicious. Small subangular to angular grains of quartz, orthoclase, microperthite, plagioclase, and cherty silica lie embedded in a variable base of sericitic mica and fine grains of magnetite and hematite. The mica is, as usual in the series, abundantly developed in the planes of bedding. According to the abundance of the once- argillaceous material, the rock may be classed as a metargillite or metasandstone. The total thickness of the formation is about equally divided between these two rock-types. The specific gravities of three specimens were found to be 2)-652, 2-674, and 2-721. Their average, 2-683, is about the average for the formation as a whole. The general composition, colour, and field relations of the Phillips are so similar to those of the upper part of the Kintla formation that one can hardly doubt that the two are in the main, stratigraphic equivalents. The chief litho- logical difference is shat the Phillips appears to be slightly the more silicious and coarser grained of the two. It may be noted that neither in the Gateway nor in Phillips was any contemporaneous lava discovered. Roosvi~tLE FORMATION. The Phillips formation is conformably overlain by the Roosville, the highest recognized member of the Galton series. The name is derived from the post office recently opened on Phillips creek. The Roosville outcrops at only one point within the area covered by the Commission map. It there forms the summit of a peak lying three miles east-northeast of Phillips creek cascade at the junction of the creek canyon with the great Kootenay trough. Erosion has removed the upper part of the formation, of which only about 600 feet of beds now remain. How much greater the total thickness may be is not known. The formation as exposed at this one locality is essentially made up of thin-bedded, light green, light gray, and greenish gray silicious metargillite bearing thin, more quartzitic, interbeds. The colours of weathering are light gray or brownish gray. Sun-cracks and ripple-marks are common. In field 110 DEPARTMENT OF THE INTERIOR 2.GEORGE V., A. 1912 habit and most lithological details the dominant phase of the Roosville is very similar to that dominant phase of the Gateway formation. It seems, however, that casts of salt-erystals are wanting in the younger formation. The metar- gillite is composed of angular quartz and feldspar grains (averaging only 0-02 mm. in diameter) in an abundant matrix of sericite, chlorite, iron ore, and possibly, in some beds, a little of the original argillaceous matter. The feldspars again include orthoclase, microperthite, and plagioclase. Bedding planes are well marked by glinting sericite in the form of innumerable minute foils and shreds. Thus, at the top of the Galton series as at the bottom, static metamor- phism has effectually changed the original clayey sediments into nearly or quite holoerystalline rocks. The mica foils developed in the Hefty or MacDonald strata are, at many horizons, larger than the micas characteristic of the Roos- ville, Phillips, or Gateway, and the top members of the series may have retained a greater quantity of original argillaceous matter. In these two respects the older formations have, through deeper burial, suffered a slightly more advanced metamorphism than the beds lying seven to ten thousand feet higher in the series. Nevertheless, the evidence is clear that the Roosville formation, like the Kintla of the Lewis series, has been buried beneath many thousands of feet of still younger strata, doubtless including the heavy Devonian and Car- boniferous limestones; to that ancient burial the development of the metargilli- iic facies of the Roosville beds is due. The specific gravity of a type specimen from the metargillite is 2-730. A somewhat weathered hand-specimen gave 2-675. The average for the forma- tion is probably about 2-710. The Roosville has yielded no fossils. The formation appears to be younger than any beds belonging to the Lewis series as above described. It may prove to be equivalent to an upper division of the Kintla which is not exposed in the Boundary belt, or may represent the westward extension of a distinet forma- tion. DEVONIAN FORMATION IN THE GALTON RANGE. DESCRIPTION. At the eastern edge of the drift-covered Tobacco Plains (115° 3’ W. Long.), a block of fossiliferous Devonian limestone has been faulted down into contact with the Gateway formation. On the west and south the limestone is covered by drift and alluvium. The main fault which limits the block on the east can be rather sharply located, the strikes of the limestone and Gateway metargillite being nearly at right angles to each other. This fault is marked on the map sheet, where it will be seen to run roughly parallel to other faults that are responsible for the local graben character of the Rocky Mountain Trench. The limestone is itself affected by numerous minor slips, so that it is impossible to be certain of the thickness. In general, the block REPORT OF THE CHIEF ASTRONOMER 1il SESSIONAL PAPER No. 25a is monoclinal, with an average northeasterly dip of about 45 degrees. The apparent thickness of all the strata is approximately 1,600 feet. Of this total 300 feet represents dolomitiec quartzite, occurring at the base of the section. The quartzite is white to cream-coloured on the fresh fracture, weathering yellowish or buff. Its beds are generally thick and massive. It bears no observed fossils other than a few markings like annelide borings. Conformably overlying the quartzite is the very massive limestone, which rarely shows bedding planes. This rock is usually fetid or bituminous on the fresh fracture. It weathers from the normal dark gray tint to a much lighter one. Cherty nodules up to three or four inches in diameter, are common in certain horizons. FOSSILS. Just above the contact with the underlying quartzite a collection of fossils, bearing the station number 1217 on its labels, was made. These were determined by Dr. H. M. Ami, whoze notes are here entered in full :—* ‘ Station No. 1217.—Boundary monument at eastern edge of the Tobacco Plains. In a dark gray, impure crinoidal and at times semi-crystalline limestone. Age: Upper Devonian. Formation: Jefferson limestone. Genera and species: . Crinoidal columns. . Productella subaculeata. . Schizophoria striatula. . Athyris vittata. . Athyris vittata, a narrower and more tumid form. . Athyris vittata, fimbriate form. . Athyris parvula, Whiteaves, or allied species. . Athyris aff. coloradoensis, probably a new species. . Athyris. . Trematospira (?) sp. No species of this genus has as yet been obtained from these limestones in Montana. 11. Pugnax pugnus, a small diminutive form. 12. Spirifer whitneyi, compare Hall’s Spirifer whitneyi (Spirifer animascensis). 13. Spirifer disjunctus, var. avimascensis. A specimen with high area and fine plication on the costae, high and twisted beak. Resembles a form from S. W. Colorado. © 00 -y & OL PR OD PO * Both the writer and Dr. Ami are under special obligation to Dr. G. H. Girty and to Dr. E. M. Kindle of the United States Geological Survey for valuable aid in deter- mining the material of these collections; also for excellent opportunities for Dr. Ami to compare the Canadian forms with specimens from various localities south of the International Boundary. 112 DEPARTMENT OF THE INTERIOR 2 GEORGE V.,, A. 1912 14. Spirifer utahensis. Shows a plication in the sinus. Ventral valve 15. 16 17. 18. 19. About station No. with twisted beak. This is the only specimen found of Spirifer utahensis in Dr. Daly’s collections. This is eminently character- istic and abundant in the Jefferson limestone of the United States. Camarotoechia, sp. . Pleurotomaria, (probably a Liospira or allied form), flat-valved. Gasteropod. Aviculoid (%) shell, too imperfect for identification. Orthoceras, sp. Portion of shell representing some twenty septa of a test rapidly increasing toward the aperture.’ 800 feet higher up in the apparent monocline other fossils, taken at 1218, are named by Dr. Ami as follows :— “Station No. 1218.—150 yards north of Boundary line, eastern edge of the Tobacco. Pl ains. A dark gray coralline limestone; very similar to the characteristic rock of the Jefferson limestone of Montana. The identical association of forms and the general physical properties of the limestones of British Columbia and Montana are remarkable and leave no doubt as to the identity of the horizons. Age: Upper Devonian. Forma tion: Jefferson limestone. Genera and species: 18 2. (J) On 92) The q horizons ar Stromatoporoids. Exhibit concentric lamine, cf. A. variolare; large and small masses. Favosites, sp. A form very close to, if not identical with, FP. limitaris, Rominger. . Favosites, sp. A form consisting of much larger fronds and smaller corallites than those of last species. (New species?) . Favosites, sp. Cf. F. limitaris, Rominger. . Brachiopod, ribbed; too imperfect. for identification. . Athyris, sp. Cf. A. parvula, very obscure example of what appears to be this species. . Athyris. Small species resembling A. parvula W. . Spirifer englemant. The same form occurs also. in Montana. . Spirifer. Cf. S. argentarius. The radiating lines which are pro- minent on the fold constitute a rather distinctive feature in this species.’ uartzite is tentatively assigned to the Devonian. Mississippian e possibly represented in the fault-block, for the collection of fossils has been by no means exhaustive. The greater part of the limestone is to be correlated with the Jefferson limestone of Montana. . REPORT OF THH CHIEF ASTRONOMER 1138 SESSIONAL PAPER No. 25a PALEOZOIC LIMESTONES OF THE MACDONALD RANGE. DESCRIPTION. With reference to the fault troughs respectively occupied by the Kootenay river (at Gateway) and the Flathead river, the Galton-MacDonald mountain system is a compound horst. We have seen that the Devonian limestone at Tobacco Plains now stands at a common level with strata as old as the base of the Siyeh. Much greater displacements at the western side of the Flathead trough have dropped Devonian and Mississippian limestones down into contact with the oldest members of the Galton series, including the Altyn formation. The result of this faulting is peculiar, since a long, slab-like block of Altyn, Hefty, and MacDonald beds is bounded on both sides by Mississippian limestone. The younger, fossiliferous limestones form two masses separated by the slab and may be referred to as the Western and Eastern blocks. The western block is well exposed only at comparatively few points; else- where it is covered by heavy forest. The bounding faults are, therefore, mapped only approximately. This limestone is dark bluish-gray, weathering light gray to whitish. It is massive, rarely showing stratification planes; fetid under the hammer; semi-crystalline, with the larger calcite crystals blackened by films of bituminous matter. In one shear-zone the normal colour is changed to yellowish gray or brown. At other outcrops the fresh limestone is erystal- line and white; the bituminous matter has there been distilled out. The rarity of visible bedding-planes makes it impossible to make certain az to the attitudes assumed by the limestone throughout the block. The best exposures along the Commission trail, where it threads the canyon at the Boundary line, show a horizontal position, but farther to the northwest probable dips of about 30° to the southwest were observed. It is likely that the western block is compound and bears numerous local faults and shear-zones. A few fossils were found at a cascade just south of the Commission trail at 114° 38’ W. Long., and 400 yards south of the Boundary slash (Station No. 1278). Dr. Ami identified these as including a species of Menophyllum and an Athyroid form. The rock elsewhere bears crinoid stems. The horizon could not be determined but it is ‘ presumably upper Mississippian.’ The eastern block i; composed of both Devonian and Mississippian lime- stones which are greatly broken by step-faults. In the field no lithological distinction could be made between the two limestones. Wherever fossils occurred the rock was massive, crinoidal, bituminous, and gray, corresponding in all respects to staple phases of the western block and of the limestone at Tobacco Plains. The proved Devonian beds, however, are specially rich in cherty nodules and are often mottled with irregular magnesian and dolomitic parts. : At the edge of the Flathead valley drift cover, 5,000-foot contour, and 1,090 yards north of the Boundary line, some 500 feet of unfossiliferous, pinkish- gray, sandy beds were noted. These are generally magnesian and include thin lenses of grit containing small, black pebbles of argillite. Cross-bedding was 25a—vol. 11—8 114 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 seen in the more quartzose layers. As a rule these beds, like the main lime- stone, were fetid under the hammer. A strong cleavage affects them, with ~ strike N. 12° W. and dip, 75—80° E.; it suggests local faulting parallel to the trend of. the Flathead valley. On the slope immediately above the reddish zone the normal gray, massive limestone begins and continues westward to the great fault where the limestone and the MacDonald metargillite make contact. In that traverse the dip gradually steepens to a maximum of about 40° or more, tothe south-westward. At the 6,500-foot contour and one-half mile north of the Boundary line, the dip abruptly changes to 55° S.W., with strike N. 55° W. The change of dip takes place at a meridional belt of intense shearing, where, for fifty feet across the belt, the limestone is a white, brecciated marble. East of this shear-belt the fossils collected are Mississippian in age; west of it the fossils are Devonian. The shear-belt seems, thus, to mark the outcrop of a strong fault along which the Mississippian limestone has been dropped down, relatively to the Devonian on the west. That traverse is probably typical of a number which might be made across the eastern block. The relations are those of step-faulted blocks with down- throw to the east. Further remarks made on page 117 as to the local structures should be added to this brief account. Since the distribution and throws of the various faults are unknown, it is not possible to state the true thicknesses of the fossiliferous limestones. Either the Devonian or the Mississippian lime- stone is certainly many hundreds of feet in thickness; their combined thickness must be well over 1,000 feet. Neither top nor bottom of the series has been discovered in the Boundary belt. In the Yakinikak valley, about five miles south of the Boundary line, in this same mountain range, Willis found a small mass of limestone carrying numerous fossils of the Saint Louis horizon of the Mississippian. He writes that the limestone— ‘Is without upper stratigraphic limit, but rests conformably on a quartzite, which is unconformable on Algonkian strata. The quartzite is about 25 feet thick, and it and the limestone lie in a nearly horizontal position. The name Yakinikak is here applied to the limestone, exclusive of the quartzite, which may elsewhere develop independent importance. . Its [the limestone’s] occurrence on Yakinikak creek is apparently due to down-faulting, as it lies at a comparatively low level among mountains eomposed of the Algonkian argillites. Its presence in this locality, taken in connection with other occurrences north and south, may be considered evidence of the former extension of the upper Mississippian limestone over the entire region. The absence of earlier Mississippian strata is significant of an unusual overlap.’* : . Tf this Yakinikak limestone were deposited unconformably upon the Galton (‘ Algonkian’) series, there must have been strong deformation and extensive *B. Willis, Bull. Geol. Soc. America, Vol. 18, 1902, p. 325. REPORT OF THE CHIEF ASTRONOMER 115 SESSIONAL PAPER No. 25a local erosion beforehand, for the Devonian limestone is represented in great thickness only three or four miles to the north. Many observations indicate with some certainty that such orogenic movements in this part of the Cordilleran region have not intervened between the deposition of the Jefferson limestone and the upper Mississippian limestone. Some other than Willis’ interpretation of the Yakinikak contact seems legitimate, if not necessary. The writer knows of no facts which inyolve any notable erosion unconformity between the Devonian or Carboniferous sediments and the Cambrian-Beltian series of southeastern British Columbia or of Montana northwest of the Belt mountains and the Helena district. FOSSILS. The Devonian fossils were all found along the western edge of the eastern block. The exact localities and the faunal lists prepared by Dr. H. M. Ami, are here given. Station No. 1276—At 114° 38’ W. Long.; seven miles west of the Flathead river, and two miles and a half north of Boundary (6,400-foot contour); close to great fault mapped. In a dark gray, impure, dislocated limestone, weathering peppery-gray, yellowish-gray, or buff; fractured and recemented, more or less altered by pressure. Surface marked by pitted structure in uniformly shallow rounded depressions, or cavities, in which a layer of caleareous (?) matter appears, as a thin lining on the inner wall. Age: Devonian. Formation: Jefferson limestone in the upper part of the Devonian system. Genera and species: 1. Chonetes (7) sp. An imperfectly preserved specimen not recogniz- able. Crushed valve showing punctate structure. . Atrypa aspera. One specimen and two small fragments of this species characterize these limestones. 3. Spirifer englemani. Exfoliated specimen in a block of limestone. Another individual, partially exfoliated, represents one of the mucronate types of Spirifer englemani with a high hinge area, and resembles very closely the Spirifer englemani from the Jefferson limestone of Utah, Montana, and Nevada as repre- sented in the collections of the U. S. Geological Survey obtained by Dr. E. M. Kindle. Station No. 1277——Five hundred yards southeast of station No. 1276, and ‘gee to great fault. The fossils occur in a dark, fractured and recemented limestone, weather- ing peppery-gray; calcite veins prevalent. Age: Upper Devonian. Formation: Jefferson limestone. 25a—vol. i1—84 bo 116 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Genera and species: 1. Sponge-like organism. Long, slender cylindrical stem-like rods or spicules (?) Hyalostelta (2). . 9. Favosites sp. Compare Favosites limitaris, Rominger. 3. Cladopora, one specimen. 4, Atrypa aspera, Schlothurm. Station No. 1290.—Near meridian of 114° 33’ W. Long.; four miles west of the Flathead river, and one-half mile north of the Boundary line; 6,500-foot contour. A limestone weathering rusty or brownish-yellow; specimens of fossils silicified. Age: Upper Devonian. Formation: Jefferson limestone. Genera and species: 1. Zaphrentis (2) sp. 2. Atrypa reticularis, Linneus. 3. Atrypa aspera, Schlothurm. The Mississippian fossils of the eastern block are listed, with locality indications, by Dr. Ami, as follows :— Station No, 1285.—At 114° 32’ 30” W. Long.; two and a half miles west of Flathead river, and one-half mile north of the Boundary line; 5,500-foot contour. Age: Upper Mississippian. Formation: Madison limestone. Genera and species: . Lithostrothon, sp. . Syringopora, sp. . Zaphrentis, sp. . Menophyllum, sp. Stenopora. . Monticuloporoid. . Composita, sp., aff. Composita trinuclea. . (2) Reticularia, sp. . Producta cora. . Productus, sp.; compare P. giganteus. Exhibits sculpture similar to that in English specimens. 11. Camarotechia, sp. 12. Spirifer, sp.; aff. S. Keokuk. 13. Spirifer leidyi, or a very closely allied species, agreeing with Nor- wood and Pratten’s description in nearly every detail. Middle rib and sinus does not always extend to beak. Sometimes three ribs of sinus equal; at other times one larger and two smaller, ete. SCO OANTAAPWNYH ra REPORT OF THE CHIE ASTRONOMER ilabey SESSIONAL FAPER No. 25% 14. Cliothyridina hirsuta. Specimens large and lamellose. Ragged edge of lamelle like spines in previously described specimens. See also Athyris hirsuta, Hall; figs. 18-21 (Spergen Hill), and Pl. 6, Vol. 1, Bull. No. 3, Amer. Museum of Natural History. Station No. 1287—Three hundred yards west of station No. 1285; 5,800- foot contour. Age: Probably upper Mississippian. Genera and species: 1. Lithostrotion, sp. 2. Syringopora, sp. 3. Menophyllum, sp. STRUCTURE OF THE GALTON- MACDONALD MOUNTAIN SYSTEM. The geosynclinal rocks between the Flathead and the Rocky Mountain Trench at the Forty-ninth Parallel are very much more deformed than are those of the Clarke range or the Lewis range. The exceedingly inflexible nature of the rocks has prevented the development of systematic folds; the structure all across the Galton-MacDonald system is almost entirely determined by faulting. At least twelve major fault-blocks are represented in the map sheets as occurring in the five-mile belt where it crosses the two ranges. Within the belt the dips range from 0° to 90°, averaging about 30°. The most easterly and the most westerly blocks contain, respectively, Car- boniferous and Devonian limestones which, excepting the Kishenehn lake beds, are the youngest bed-rock formations in the Rocky Mountain system at this latitude. These particular blocks are of special interest since they clearly show the magnitude of the displacements to which the Flathead trough and the Rocky Mountain Trench owe their origin. The Carboniferous limestone on the west side of the Flathead is on the same level with strata on the east side, belonging to the lower Appekunny. One may fairly estimate that a net displacement of at least 15,000 feet or possibly 20,000 feet is here indicated. The western part of the Clarke range has been lifted nearly or quite three miles higher than the most easterly block of the MacDonald range. The latter block is downthrown by an even greater amount with respect to the block next on the west. The Carboniferous limestone at the Flathead valley is, in fact, the visible upper portion of a broad block or series of parallel blocks which have been dropped a minimum of about three miles below the adjacent blocks of the Clarke and’ MacDonald ranges. The Flathead trough is thus structurally a typical fault-trough or ‘graben.’ It is also highly probable that the depression has always been a graben in a topographic sense. It has been partially filled with lake beds and has been deformed by the folding of those beds but there is no evidence that the initial trough form was ever quite destroyed. It will be observed from the map sheet that the Carboniferous limestone of the MacDonald range occurs in two different fault-blocks separated by a 118 fie DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 narrow, slab-like block of strata ranging in age from the Altyn to the Mac- Donald inclusive. It is not easy to understand the conditions under which this narrow slab, composed of the oldest sediments in the range, can make contact on each side with the youngest formation of the range. Two hypotheses are conceivable. According to the first the relations were established by simple normal faulting whereby the Carboniferous blocks were dropped down. According to the second hypothesis one may postulate a local overthrust of the older rocks upon the Carboniferous, followed by normal faulting which dropped the Altyn-MacDonald block down into the Carboniferous limestone. This second view would naturally correlate with the speculation that the whole Clarke range has been thrust over Carboniferous or Cretaceous formations. Extreme as this idea may be, the known facts do not exclude it and the two hypothetical alternatives are still open. It may be noted that the outcrops of the fault planes on the east side of the Altyn-MacDonald slab have been drawn with considerable confidence. The fault line on the western side was not so readily plotted in the field, but it is believed to be mapped with approximate accuracy. ; In four or more leading cases the faul. tines have been shown as following stream courses among these mountains. The local valley of Wigwam river has been determined in position by a break which has strongly affected the dips of the MacDonald formation on either side of the river. The five most westerly blocks of the Galton range show a progressive down- dropping of blocks from east to west, wita the result that the MacDonald, Siyeh, Purcell, Gateway, and Devonian-limestone formations are successively in lateral contact. The equivalence of level between the Devonian limestone at Tobacco Plains and the lower MacDonald beds along the Wigwam shows that the net relative displacement of the two blocks has been at least 10,000 feet and may have been several thousand feet greater. We are therefore pre- pared to find that the Rocky Mountain Trench at the Forty-ninth Parallel has been located on a zone of strong faulting. This conclusion will be noted again, in. the next chapter, on the Purcell mountain system. The relation of the Devonian and Carboniferous formations to the older geosynclinal prism has been discussed in connection with the stratigraphy of the younger limestone:. The Boundary belt has furnished very little inforftna- tion on this subject. 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER VI. STRATIGRAPHY AND STRUCTURE OF THE PURCELL MOUNTAIN SYSTEM. PURCELL SERIES. As one leaves the Rocky Mountain system and crosses the wide master trench to study the composition of the Purcell system along the Forty-ninth Parallel, he enters a much more difficult field. Between Gateway and Porthill the mountains seldom rise above tree-line and the forest cap is, throughout the stretch, of unusual density and continuity. Notwithstanding the steepness of the mountain slopes the timber generally stands thick upon them. Beneath the trees a heavy growth of brush and generally, a discouragingly thick layer of moss and humus, form an impenetrable cover over most of the bed-rock on the Boundary belt. During many traverses made during the season of 1904 outcrops absolutely failed for a mile, or even for several miles, at a time. Field work was further rendered unsatisfactory during that extraordinarily dry season on account of the thick smoke which hung over the mountains. For one period of seven weeks the smoke was dense enough to interfere seriously with the work of discovering outcrops. In the Pureells the stratigraphic conclusions were rendered all the more delicate because of the remarkable uniformity of the sedimentary formations. It was found that much the greater part of the belt is underlain by the strati- graphic equivalent of the Galton and Lewis series. This equivalent has been named the Purcell series. Very seldom is there represented among its members anything like the lively contrasts existing, for example, between the Kintla and Sheppard formations, between the Siyeh and Grinnell, between the Appe- kunny and Altyn, or between the respective pairs of formations in the Galton series. For thousands of feet together the strata of the Purcell series exhibit a homogeneity that is bound to excite wonder in the mind of the geologist. In the Moyie and Yahk ranges not a single stratum of marked individuality has been discovered which is proved to persist throughout the ranges. In none of the three ranges has any formation yielded fossils. This failure of well defined horizon-markers in a region of considerable structural complexity is, perhaps, the greatest of the difficulties that confront the geologist in the Purcells. For these reasons the writer has not felt justified in attempting to des- eribe the Purcell series in the detail which is warranted in the case of the formations composing the Galton or Lewis series. It has seemed safer to express the stratigraphy of the Purcell mountains in terms of three very thick, 119 120 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 conformable sedimentary formations, each of which, on account of its homo- geneity, as yet defies profitable systematic analysis into subdivisions of more usual thickness. Even these grand divisions of the thick series, called the Creston, Kitchener, and Moyie formations, are not always with ease separable from one another in the field. All are highly silicious in character; all are fine-grained to compact in texture; all show phases which are indistinguishable in the hand specimen or in the ledge. The three formations are, in fact, separated on the ground of comparatively subordinate lithological differences, such as colours of fresh fracture and weathered surface. Microscopically and chemically the immensely thick Creston and Kitchener formations are proved to be almost identical in constitution. A prevailing and clearly minor difference between them, consisting in the fact that the Kitchener is the more ferruginous of the two formations, has been used as a principal means of distinguishing these two parts of the series in the field. In addition, the Kitchener is thinner-bedded than the Creston. With such criteria merely, it is clear that the mapping of these formations in the fault-riven mountain masses is a delicate matter. The geological boundaries as shown on the map sheets are thus to be considered as drawn, in many instances, with more doubt than is the case with the sheets located east of Gateway. Two of the sedimentary formations have been named after stations on the Canadian Pacific railway; the third, the Moyie formation is so called after the river of that name. Their estimated thickness and general composi- tion are noted in the following table: Formation. Thickness in feet. Dominant rocks. Top, erosion surface. Moyie..=.. .. tay, Gis: MoraPrane 3,400+ Metargillite. Purcell Lava.. 4. .. .. «. ~ 465 Altered basalt. Katehenenin we. is in icrouiee 7,400 Quartzite. CWT EStON se) sete iicie eis eis eie 9,500+ Quartzite. 20,765+ Base concealed. CRESTON FORMATION. General description—The lowest member of the Purcell series and the old- est formation seen in the Boundary belt within the entire Purcell mountain system has been named the Creston formation. Its best exposures include the one in and east of the lofty McKim cliff four miles from Porthill; a less com- plete one on the slope immediately east of the Moyie river; and, finally, the most favourable one of all, on the two sides of the wide valley occupied by the east fork of the Yahk river. In each of these exposures the formation pre- serves nearly constant characters to the lowest bed visible; it is thus highly probable that this gigantic sedimentary formation is, as a whole, yet thicker than the total mass actually measured in the field. In different sections among the fault-blocks characteristic of the Purcell mountain system, estimates of from 6,000 to 9,909 feet were obtained for the REPORT OF THE CHIEF ASTRONOMER 121 SESSIONAL PAPER No. 25a whole thickness locally observed. The highest figure refers to the remarkably extensive outcrop of the Creston rocks at the Yahk river. High as the estimate appears, a minimum thickness of 9,500 feet is assigned to the formation. The estimate is the result of two complete traverses run across the great monocline at this locality. It cannot be denied that there may be some duplication in this particular section, but, on tae other hand, the writer, after careful study in the field, found not the least hint of duplication. Similarly, in each of a half-dozen other sections in as many different fault-blocks, as much as 5,000 to 7,000 feet of the upturned Creston quartzite were measured without any clue to repetition of the beds. In several fault-blocks the strata stand nearly vertical and errors of mensuration were reduced to a minimum. At McKim cliff, about 3,000 feet of nearly horizontal, typical Creston are exposed to one sweep of the eye, with neither the summit or base of the formation to be found at that locality. A further indication that the Purcell series, of which the Creston makes up nearly one-half, is enormously thick, is derivable from McEvoy’s recon- naissance map of the East Kootenay District.* The map shows that at least 3,000 square miles of the Purcell mountain system north of the Boundary is- almost continuously underlain by a silicious series evidently equivalent to that cropping out at the Forty-ninth Parallel. The continuity of the colour repre- senting the series on the map is broken only by patches of gabbroid intrusions doubtless similar to the intrusions so plentifully found in the Boundary belt. When it is remembered that the rocks of the large area in East Kootenay are much faulted and otherwise disturbed so as to present all angles of dip even to verticality, we see certain proof that these conformable strata must have very great total thickness. This conclusion may be corroborated by information won from even the fleeting glance one can give to the rocks that are visible from the railway train on the stretch from Cranbrook to Kootenay Landing. In minor degree the estimated thickness of the formation may vary according to the somewhat arbitrary po:ition assigned at each exposure to the upper limit of the Creston. In every case the formation gradually becomes more ferrugi- nous and thus passes slowly into the overlying Kitchener. The doubtful inter- mediate band of strata often totals several hundred feet in thickness. The top of the Creston has been generally fixed within the band where the thinner bedding as well as the rusty character of the Kitchener becomes pronounced in the quartzitic strata. In conclusion, then, the writer believes it to be best to trust the minimum estimate of 9,500 feet for the Creston as embodying the net balance of pro- babilities derived from the field study. It may be addéd that, in the opinion of the writer, this vast thickness for a single formation is not to be explained as only the apparent thickness of beds deposited in fore-set bedding as a sub- marine delta. The recent emphasis ot geologists on this source of error in -measuring the actual thickness of a ciastic formation is certainly justified. * Accompanying Part A, Ann. Rep. Geol. Surv., Canada, Vol. 12, 1899. OOS DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 — Ip the present case, however, the criteria of inclined fore-set bedding, in con- trast to practically horizontal bedding on a _ subsiding, flat sea-floor, do not seem to be matched by the facts. The prevalence of sun-cracks, ripple- marks and other shallow-water markings in the perfectly conformable Kitchener and Moyie formations, as well as in the Creston formation in less degree, appears to show that the sea bottom and the bedding planes of the sands and muds were nearly level throughout the deposition of the Purcell series. The Creston formation is no more extraordinary for immense thickness than it is for its wonderful homogeneity in any one section. There is a signal absence of well-marked lithological horizon-markers. The nearest parallel to this homogeneity among the Boundary formations is that afforded by the basal arkose member of the Cretaceous section at the Pasayten river. The lack of strong horizon-markers is not to be explained by the lack of sufficient out- erops; the frequent recurrence of the Creston rocks among the fault-blocks, coupled with the excellence of exposure for portions of the formation in each large outcrop, render it improbable that important bands of rock other than the staple quartzite have been overlooked in the Boundary belt. The forest eap interferes much more with determinations of total thickness and of the larger structural features such as faults and folds, than with the study of the details of composition. Neither in slide rock nor in gravels of the canyon- sireams of the areas mapped, as underlain by the Creston, was any other rock discovered in large amount than those which are the dominant components of the Creston quartzite as hereafter described. | While relative homogeneity characterizes the formation from top to bottom at any one exposure, noteworthy changes in its constitution were observed as the Boundary belt was traversed from west to east. The Creston as outcropping in the Moyie range and western half of the Yahk range thus stands in a certain lithological contrast to the same formation where it crops out farther east. Por the understanding of this important fact it is convenient to recognize two different phases of the formation in the Purcell mountain system—a western and an eastern phase. 5 Western Phase.—At McKim cliff and in the outcrops immediately east of the crest the material was largely gathered for the following description of the Creston formation in a typical section representing the western phase. In the cliff itself the staple rock is a very hard and tough quartzite, break- ing with a sonorous, almost metallic ring. The individual beds vary from a few inches to twenty-five feet or more in thickness, averaging perhaps three feet. Very often the more massive plates are seamed with thin dark-gray laminae of once-argillaceous quartzite or metargillite, but true shale or slate was never seen in this part of the section. For 2,500 feet measured vertically up the cliff the quartzite, which dips 3°-10° eastward, is specially massive, giving the effect of superb cyclopean masonry, broken horizontally by widely spaced bedding-planes and broken vertically only by master joints. ‘Toward the top of the cliff the rock is somewhat thinner-bedded, but is still a strong, typical quartzite. REPORT OF THE CHIEF ASTRONOMER 123 SESSIONAL PAPER No. 25a The dominant colour on fresh fractures is throughout gray or greenish- gray, weathering to a somewhat lighter tint of nearly pure gray. A few white or grayish-white beds occur irregularly through the formation and, also rarely greenish-gray beds weathered light rusty-brown or reddish-brown, so as to resemble typical Kitchener quartzite. Heavier beds characterize the formation where its upper part crops out just east of the cliff. That seems to te the rule for the quartzite generally as it is exposed in the Purcell range; the bedding is thick and massive in the top and bottom divisions and thinner-bedded in the middle division of the strata. The exposures, however, are nowhere continuous enough to allow of a trust worthy estimate of tke relative strength of these three divisions. The sediment sometimes, though quite rarely, shows cross-bedding. Sun-cracks, rill-marks, ripple-marks, and annelide burrows were not identified in a single case among the strata exposed on McKim cliff. Elsewhere within the Boundary belt these markings were found; rarely in the dominant quartzite, but more particularly in the metargillitic horizons. Already in the hand-specimens numerous glints of light from non-mica- ceous particles suggest that the rock is highly feldspathic. At the-same time it is seen that the general greenish tint of the quartzite is due to disseminated minute plates and shreddy foils of mica. These observations are confirmed by microscopic examination. Interlocking quartz, feldspar, and mica are seen to be the essential constituents. Each of these minerals is glass-clear in the fresh specimens. Orthoclase, microcline, microperthite, oligoclase, and probably albite make up the list of feldspars. Of these ovthoclase and. microperthite are the most abundant, though it is not certain that, in any specimen, the other feldspars of the list are absent. The mica includes both highly pleochroic biotite and muscovite, the latter being either well developed in plates or in _the typical shreds of sericite. In some specimens the biotite is the more abund- ant of the two micas but in others it kecomes subordinate vo muscovite and may disappear altogether. Other constituents are very subordinate; they include rare anhédra of titanite, titaniferous magnetite, pyrite, epidote and zoisite. The quartz and feldspar grains vary from 0-02 min. to 0-2 mm. in diameter, averaging perhaps 0-06 or 0-08 mm. The lengths of the mica scales are usually not much greater. Though few direct traces of clastic form are left among the minerals, it is probable that these dimensions represent approximately the size of the original grains. The texture of the quartzite is thus, quite fine in the type specimens as, indeed, throughout all the exposures; in all the thousands of feet of thickness no conglomeratic, gritty, or even very coarse sandy bed was seen. It is an open question, perhaps, whether this rock should be ealled a quartzite if by that term we mean an indurated sandstone. The average quartz grain is much too small to have formed originally a true sand. In fact the average grain of the rock is not more than one one-thousandth as large as the average grain of typical beach sand. The name ‘ quartzite’, adopted 124 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 by McEvoy, Dawson, and others for these rocks, has been retained because of the chemical composition and tough, massive field habit of the beds selected as the types of the formation. To the writer a distinct genetic problem remains. One cannot easily understand the conditions under which such an immense accumulation of fine quartz and feldspar particles has been made. The purely argillaceous material must have been quite subordinate through thousands of feet of the Creston formation. The question arises as to the mechanism by which residual clay has been thus separated from the more silicious matter. Such separation is very rare, if not unknown, in the muds now accumulating on the ocean-floor. The writer has failed to find in the ‘ Challenger’ report on the deep-sea deposits an account of any mud which chemically or minerale- gically matches the Creston type of deposit. The ‘Blue Muds’ of the report furnish the nearest parallels and yet show vital contrasts. This problem of genesis applies also to the rock forming the type of the Kitchener quartzite* The micas and the accessories are chiefly the result of the crystallization of a small original admixture of micaceous, argillaceous, and ferruginous material in the sandy sediment. It is probable that most of the quartz and of the feldspars represent clastic material cemented together by secondary growths of the original crystal fragments. One of the plagioclases, referred with some doubt to albite, may be of metamorphic origin. The metamorphism which led to the crystallization or recrystallization was, almost without doubt, not dynamic but static in nature. As in the case of the metargillites of the Lewis and Galton series, these effects have resulted from deep burial with consequent increase of temperature and pressure. , Professor Dittrich’s analysis of a typical specimen (No. 1125) of the homo- geneous quartzite from McKim cliff gave the following result— * While this chapter was going through the press the writer had opportunity to study the Shuswap terrane, from which the clastic materials of the Creston, Kitchener, and other formations composing the Rocky Mountain Geosynclinal were derived. Great thicknesses of phyllites, chlorite schists, green schists, greenstones, and fine-grained mica schists were found in this pre-Beltian terrane as exposed at the Shuswap lakes. In general these rocks are abundantly charged with secondary quartz developed in minute individual crystals (anhedra). During the secular weathering of such rocks the more soluble micas, chlorite, tale, uralite, etc., would be leached out and the more resistant quartz and alkaline feldspar would be ‘washed out to sea. The writer is inclined to credit this explanation of the silicious muds which have been consolidated to form the thick, very dense quartzites of the Cambrian and Beltian formations. The rich content of microcline and microperthite repeatedly emphasized in the descrip- tions of the latter can be explained as due to the weathering and washing of the millions of aplite and pegmatite dikes and sills cutting the Shuswap sediments and green schists. These injections are associated with large batholiths of likewise pre- Beltian granite: its débris is also represented in the Rorky Mountain Geosynclinal. Oi REPORT OF THE CHIEF ASTRONOMER 12 SESSIONAL PAPER No. 25a Analysis of Creston quartzite, Western Phase. Mol. STUDS. casks: HSSmea Suns Hale GGL MA AON GI CREME | Pina nae 82-10 1-368 TiOse: SES RETA G ac) Oe REST aG OT SD ARLE eEna: ares ae 40 005 Al,O3.. 8-86 087 ¥e,0,.. 49 003 FeO.. 1-38 019 MnO.. 03 MgO.. 56 -O14 CaoOe +82 014 Na,O A 2-51 -040 i (Mena hey Pach tec weve Montes TO SINS SL hed esa eh ele SE SO reas 2-41 026 H,.O at 110°C... Ba whe ce ors Ore RR eee ae? es 05 aes H.O above 110°C... Spee PURE Meth ta Sn Ig (WE Me at Reh aareote 37 -029 POs: Says Ma os Wao WG readied eons ts eee eg WS ae 04 eG 100-02 Sifs HPPSS cd. 86 56 00-60 00 Hoo Go 165 oo GO OG. H0 focro Wolbes | ChiteEL Assigning all of the soda to the albite molecule, one half of the lime to the anorthite, the other half of it to titanite, apatite, epidote, and zoisite, the weight percentages of the constituent minerals have been roughly calculated as follows :— Quartz... 58 Albite molecule. . Sa Og Ce OSreE coacoiet 21 Orthoclase molecule. SED EO SO DE SOG GLO ciican One once 9 AMOLEthitewmMolecule’ ce sy, caeec)omcishastein cis) lomeiel (acineiss e's 2 MICAS te CAchias ae aan Sec set ese rf IACCESSOTIESSs 2s ee eae 3 = 100 It is quite possible that a considerable fraction of the soda should be assigned to the sericite and even that paragonite itself is present. However, from microscopic evidence it is probable that the soda-feldspar molecule is the principal source of this alkali. The mineral percentages are, therefore, believed to be nearly enough accurate to give a fair idea of the composition of the average quartzite. It is clearly a quite highly feldspathic sediment. The only notable variation from this average composition of the typical quartzite is found in the thin, darker, more micaceous and ferruginous laminae which often interrupt the dominant light gray quartzite. These laminz, vary- ing from a centimetre or less to several centimetres in thickness, often have the habit of metargillite, but usually they are so acid as to rank among the impure quartzites. The specific gravity of the analyzed specimen, 2. 681, is near the average for the staple, light gray rock. The average for ten specimens typical of*the whole western phaze is 2-698. The monotony of the western phase is seldom broken by the appearance ef any lithological novelties. At a few horizons the micaceous material of the rock is segregated into flattened spheroidal or more irregularly shaped, concretionary masses of all sizes up to a foot or more in width. The greatest diameter of the concretion 126 DEPsaRTMENT OF THE INTERIOR -2 GEORGE V., A. 1912 almost invariably les in the plane of bedding. The heart of each segregation is especially rich in biotite and sericitic muscovite. Their main mass is com- posed of a grayish-white, granular base of interlocking grains of quartz and subordinate feldspar,-in which are embedded abundant, conspicuous foils of black biotite 5 mm. or more in diameter, and many highly poikilitic red garnets, with large anhedra of titanite. The material of the dark-coloured minerals has plainly migrated inward from te surrounding rock-mass, for each segregation is enclosed in a white, decolourized shell of quartzite, consist- ing of nearly pure quartz and feldspar. In each of the larger segregations the mica and garnet are not regularly distributed with reference to the periphery but occur in numerous small clumpy aggregates within the main body of the segregation. In some of the smaller segregations the micas are more evenly distributed, in a manner similar. to that observed in concretions in the Kitchener quartzite. At a few other horizons the feldspathic quartzite is spangled with large biotite foils up to 1 em. in diameter. These cut acros; the bedding plane at all angles. The cause of their growth and of their restriction to a very limited number of strata in the great, apparently homogeneous series is not under- stood. Neither special dynamic metamorphism nor the thermal metamorphism of igneous intrusives were feasible explanations for the spangled quartzite at the localities where it: was actually discovered. As a rule the Creston quartzite is not cleaved, but in the fault-block at the Moyie river there is a distinct cleavage crossing the bedding planes at relatively low angles. In this case sericite is developed in the secondary planes as well as along the bedding. Eastern Phase-—The western phase just described characterizes the forma- tion as it crops out in the Boundary belt between Porthill and the Moyie river. Eastward of the river the Creston gradually assumes the features which are normal to the eastern phase. The latter is typically developed at the Yahk river, where the formation finally disappears beneath younger rocks. The most important lthological contrasts with the western phase consist in :—-first, a decided decrease in the average thickness of the beds, often leading to a fine lamination at many horizons; secondly, a pronounced increase in the amount of argillaceous matter which here forms many distinct beds and also occurs “as a notable impurity in the still dominant quartzite; and thirdly, the appearance of calcium and magnesium carbonates as subordinate elements in both the quartzite and the more argillaceous strata. The increase of the carbonate manifests itself in the rock-ledges, which, on account of the special solubility of the carbonates, present, to sight and touch, a characteristic rough- ness on weathered surfaces. In general, the calcium carbonate seems to be in some excess over the magnesian carbonate, as shown by a certain amount of effervescence with cold dilute acid. In order to obtain a definite idea as to the composition of the eastern phase, type specimens were collected at the Yahk river section and have been studied microscopically. Oue of these specimens, taken from a large outcrop REPORT OF THE CHIEF ASTRONOMER 127 SESSIONAL PAPER No. 25a on the Commission trail, about one thousand yards west of the main fork of the river, has been chemically analyzed. Its description will serve to show the general nature of the typical calcareous part of the formation. On the fresh fracture the rock is light gray, compact, and thin-bedded, thougk platy because of the cementation of many laminae of varying composition. The weathered surface is generally of a still paler gray colour, but for a depth of one or two millimetres below the surface there is usually a shell of altered rock of a brown or buff colour. The decolourization at the surface is doubtless an effect of leaching by vegetable acids. Under the microscope the rock is seen to be composed of carbonates, quartz, feldspar, sericitic mica, a little green biotite, and small grains of _limonitized iron ore. These constituents are named in the order of decreasing abundance. The carbonate grains vary from 0-01 mm. or less to 0-03 mm. in diameter and average about 0-02 mm. ‘They never appear to have rhom- bohedral development. The quartz and feldspar grains which are, doubtless, in largest part of clastic origin, vary from 0-02 mm. to 0-1 mm. or more in diameter, averaging about 0-06 mm. The dominant mica, sericite, is not dis- tributed uniformly but is most abundant in rather sharply defined lamine of specially fine grain. Such lamine were evidently more purely argillaceous than the remainder of the rock. No true argillaceous material can be dis- cerned in thin section; the sediment has been very largely recrystallized and its insoluble base is a metargillite. The percentages in Professor Dittrich’s chemical analysis (specimen No. 1179) are not very different from those roughly deduced from microscopic study :— Analysis of type specimen, Creston formation, Eastern Phase. Mol. SO a ras ee ty art are ere eet MER RUNNY es Ne oY 2 as ete 51-65 861 BA ee Ba Siasrcyate she kero aton oisiowe eens lovV SsTIL Gedo oi Sie Tae olen eyelet nee 7-85 077 TBE eras WWE aed os Ue arg tae a ny er ON AC Beste 1-74 O11 IDO GS. Se nieee ierayay Dnt LA renter a Ate repeat, ya eg Gan mee aL aa a ae ie 98 014 Wed OS 8: SAY Ga VET OS Do CIO ICE Po IaOIS TO RIS aes Sane ees 3-67 092 CORK O ig ie oe eee Saree Rep re ie ae (eae i i ee ep 15-02 268 INAS OR ene et a RIN A hl crct nier eR aaa cae ania 2-69 044 EG Oey eater RT STORE Sie ORE Se ees aT CaS ts 1:38 015 EEO ate OSC ae eee sia: So sii | Seat eo otal oe ain 09 ee EE ORADO VERE OC Cie pepaet er ceitae ee cient renee Siero nee arts 1-81 100 COM eae eae Snes. S Uae ea trea a antes ee bet 13-05 297 99.93 Semele eet cicened ret Grete ma cacartn ote Wl amn ciel ale ss ea eeaei te 2-654 imsoluplepnenydrochlorieracid’. sain treo ese cn ere oa ee enaee 66-21% Soluble in “hydrochloric acid: Fe.OQ,.. OO OG (C'O 00 700 OO) O0lmcd=o0, CO OO) OO FOO) DON dO Cosco DO 6OO SOO 1-92 IANNE (Separate comeurtr enna ex Eade Eret 2. otk ct NON eS 2-02 CORN OS EASE SES es Gee tiel ao Ge en re TSS RSA St a we oe 12-88 IN IGS O)eg ids ACS oT OG Ot UC OEP SE ne? DCL cok ts REO ar hy cme 2-4] Assigning the soluble lime and magnesia to the carbonates, the remainder of the lime to the anorthite molecule, the soda to the albite molecule, the 128 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 potash to the orthoclase molecule, the iron oxides to magnetite, and the residual silica to quartz, the following ‘norm’ has been calculated for the rock:— Quartz... sl Wacee anal ode testi): Hever ialee eyed dutele tare Datapziolat beega state elutes 25-5 Albite molecule.. its, saeyavati asa cueis iene avedee Rie 6) Bisvall case linusiineretaneie ko nrewr are Meio ae 23-0 Orthoclase molecule.. Er Ar CENA aries RN ese kat ae tis a ONS 5 8-3 Anorthite: molecule sh; eye kramer yee elas deren ate ee ee eats 10-5 Magnetite.. .. . Saf p stoic evsisben Aeaaeiacel teteene ce iis eval exes pick a Msenutere ibe Petanee 2-5 Calcium carbonate. . Si eee taalan ciceaais ae cody he, eat: Luge Gi nee ere 23-0 Magnesium carbonate. . See Sete Bat ee EC ES ra ee aan anes Ar eas 5-0 Remainder.. ST ORs eG Se a hon ea rp A meray DRY Peper u id Vee aut ts 2-2 100-0 In general this ‘norm’ is not far from representing the actual mineralo- gical composition of the rock. The unexpected abundance of the soda again raises the suspicion, here as in the study of the analyzed western phase, that paragonite is really present; how far the ‘norm’ deviates from the ‘mode’ in this respect cannot be declared. Chemically and mineralogically the rock has certain features of each of the three different types:—a feldspathic quartzite like the type of the western phase of the Creston formation; a metargillite like that dominant in the Mac- Donald or Appekunny formations; and a magnesian limestone. The size of grain of the carbonate is close to that characterizing the Altyn dolomite and other carbonate-bearing members of the Galton and Lewis series. This eastern phase may thus be a rock-type transitional between the western phase and the rocks composing the Waterton, Altyn, and Appekunny formations. KXITCHENER FORMATION. At all the localities where the two formations have been seen in contact, the Creston passes quite gradually into the conformably overlying Kitchener formation. The change from one to the other is so gradual, and the lithological differences between the two are of so low an order that, as already noted, the mapping of these formations offered considerable difficulty at many points in the Boundary belt. Much additional field work and the discovery of more favourable sections will be necessary before the Kitchener formation can be described in detail. It is convenient and instructive to group the facts known about the Kitchener into a statement regarding both a western and an eastern phase. Where outcropping in the Moyie and Yahk ranges, the dominant rock belongs to the western phase; the western slope of the McGillivray range bears thick masses of strata belonging to the eastern phase. Finally, on the eastern slope of the McGillivray range, the eastern phase of the Kitchener was found to be so far changed as to be, for hundreds of feet together, indistinguishable from the Siyeh formation. So, in fact, the rocks of the McGillivray range, have been mapped with the express recognition of the stratigraphic equivalence between the Siyeh and the main mass of the Kitchener. (See map sheets Nos. 3 and 4.) REPORT OF THE CHISF ASTRONOMER 129 SESSIONAL PAPER No. 25a Western Phase.—The thickness of the formation as exposed in the Moyie and Yahk ranges was roughly measured at two sections nearly along the Bound- ary slash on the two sides of the Moyie river. One measurement. gave approxi- mately 8,000 feet; the other, 7,400 feet. In neither case was the base or top of the formation actually visible. In a section still farther west both base and top can be found in a nearly complete section of the Purcell series, but the poor exposures in the dense forest cap there conspire with the difficulties of ‘mensuration, in an area of variable dips, to prevent a trustworthy measure- ment of total thickness. This third section has, however, offered sufficient data to render it probable that in the two former sections we have nearly the whole thickness represented. The smaller of the two estimates, 7,400 feet, was won from the structurally very favourable section in the fault-block bearing the great Moyie sills at the Boundary line and immediately west of those sills. It is possible, however, that even this lower estimate is too high and that the true ‘thickness might be more accurately placed at 7,000 feet. It appears certain only that the Kitchener in this area cannot be less than 6,500 feet thick or ‘much more than 8,000 feet thick. For the present the original estimate of 7.400 feet may be accepted with the understanding that it may be several ‘hundreds of feet too great. The dominant rock of the western phase is to be classed as a notably uniform quartzite. The bedding is, on the average, considerably thinner than in the typical Creston quartzite. Individual strata range from a minute fraction of an inch to six feet or more in thickness. pect wes sta cx.d5 OSpeenon nea ks oc 2-708 IPULCcellceriesee scar ic he oa eradiee ae oe Oe ENTS eee 2.702 SUMAN PESCLI ES eer ee mene Rone roe cette ry atu ley Map tNpa on mt 2-705 174 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912: Table V. gives actual densities in a part of the Cordilleran region and may possibly be of some value in discussions of pendulum observations or of other geophysical problems as they may be concerned with this region in the’ future. The averages of Table VI. express the range of densities in a typical, thoroughly consolidated (statically metamorphosed) geosynclinal prism. Table VII. indicates the approximate density relations of the Galton series, the least completely exposed series from the prism, to those of the equivalent. strata of the other three series. The average densities of the two western equivalents are sensibly the same as that of the Galton series. On account of the extensive development of dolomite in the Lewis series, its average density is, as was to be expected, considerably higher than that of any other of the series. CorRELATION OF THE Four BounpAry SERIES WITH THE CASTLE Mountatn-Bow River (CAMBRIAN) GROUP. During the course of the field work in 1905 it gradually became suspected that the as yet unfossiliferous Siyeh limestone is the stratigraphic equivalent of the Cambrian Castle Mountain limestone of McConnell’s well-known section on the main line of the Canadian Pacific railway. This suspicion was strengthened in the course of a brief examination of the rocks at and east of Mt. Stephen in the autumn of that year. The importance of the correlation prompted a second and longer field-study which might, to some extent, supple- ment McConnell’s all-too-brief report on the great section. ‘Toward the close of the season of 1906 the writer accordingly spent five days in working over the type sections on the northeast and southwest sides of the Bow river valley: The time available was too limited to secure a detailed columnar section of the group; yet the field evidence was clearly in favour of the correlation of the Siyeh and Castle Mountain formations. The principal information was obtained from two partial sections, the one running northeastward from Eldon station to the 9,800-foot, unnamed summit. northwest of Castle mountain; the second, running westward from Lake Louise chalet to the base of Popes Peak. Combining the results of the two traverses;. the following succession was established: Top, erosion surface. 3,500 feet.—Impure magnesian limestone with thin interbeds of shaly metargillite.. 10500)% 35 Quartzite in thin to thick beds. 1,200+ ** Fine-grained conglomerate, grit and quartzitic sandstone. Base concealed. Reference to the published report and, afterwards, personal consultation with Mr. McConnell, gave assurance that the limestone typically represented the Castle Mountain formation in its lower part, while the quartzites, conglom- erate, and grit as typically represented the Bow River formation im its upper part. PLATE 22. Molar-tooth structuce in Siyeh limestone (weathered), Clarke Range. Sunken parts caicium carbonate ; remainder dolomitic. Two-thirds natural size. Molar-tooth structure in Castle Mountain dolomite (unweathered) on main line of Canadian Pacific Railway. Light parts dolomitic ; dark parts calcium carbonate. Two-thirds natural size. 25a—-vol. i—p. 174. REPORT OF THE CHIEF ASTRONOMER 175. SESSIONAL PAPER No. 25a In the section northeast of Eldon the Bow River rocks are not well exposed but the limestone shows about 3,500 feet of its thickness. The dip averages 25° northeast. At its base the limestone is rather massive and is composed of firmly knit, thin beds of alternating, gray-weathering and light brown to buff weathering, impure carbonate. The gray layers carry little magnesium car- bonate, but the brown-buff layers are dolomitic. The thickness of these layers runs from a fraction of an inch to two inches. About 1,000 feet above the base a band, 100 feet or more in thickness, is exceptional in being thin-bedded and easily cleaved but it preserves the buff weather-tint and a dolomitic composi- tion. Similar thin zones occur both above and below this band. In general, however, the limestone is not only thick-platy in structure as in the coursing of heavy masonry, but, like the Siyeh limestone, shows a rather uniform, buff to brown weathering tint and high content of magnesium carbonate. Both phases of the fresh limestone are normally rather dark-gray or bluish-gray. The rock is often arenaceous or argillaceous; the weathered surface is rough- ened very often, through the projection of the sand-grains. The unequal dis- tribution of carbonate and impurity renders the surface characteristically pitted. In a specially argillaceous bed at the top of the 100-foot, thin-bedded lime- stone band, fossils, chiefly trilobite fragments of apparently Middle Cambrian age were discovered. Middle Cambrian fossils were also found at the base of the whole limestone formation. The likewise excellent section above Laggan, thirteen miles to the north- westward and across Bow river valley, disclosed practically identical features in the limestone. The lower beds, which were found to crop out at Lake Agnes afforded fragments of indeterminable trilobites and crustacean tracks. Perhaps the most significant fact derived from the section is that the lime- stone very often possesses the typical molar-tooth structure so characteristic of:- the Siyeh limestone. (Plate 22). This structure is not well developed in any part of the Eldon section but it was again seen in the limestone at Mt. Stephen and at several other points along the railroad, where the line cuts across outcrops of the Castle Mountain formation. As in the Siyeh limestone the molar-tooth structure seems to be best developed where the rock has been locally cleaved or cracked by orogenic stress. The thick quartzite underlying the limestone is well exposed at the Laggan section. It is a thick-bedded formation, heavy plates of quartzite alternating with subordinate, thin, fissile intercalations of silicious metargillite. The general colour of the quartzite on a fresh fracture is pale reddish to reddish- gray; the colours of the weathered rock are in general, rusty-brown and red but vary through white, pale gray, pale red, pink, brown, purple, and. toward the top of the formation, deep maroon-red. The argillaceous interbeds are dark-gray or greenish, weathering greenish or grayish-brown. Cross- bedding, ripple-marks, annelide trails, and borings are all common. The litho- logical similarity of the lighter tinted and thicker beds to the Ripple quartzite 176 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 of the Summit series was marked, and even more striking was the likeness of the reddish beds to standard phases of the Wigwam formation of the Galton series and the Grinnell formation of the Lewis series. No fossils were dis- covered in the quartzite. The Bow River conglomerate and grit were also best seen in the Laggan section. Only 1,200 feet of these rocks appear, the base being here hidden beneath the Glacial gravels of the Bow valley. All stages of transition are represented between the conglomerate with well-rounded pebbles an inch or less in diameter, to a quartzitic sandstone of medium grain. The abundant and heavy beds of grit represent the rock of intermediate grain. These three sedimentary types occur in alternation through the whole 1,200 feet, though the conglomerate lenses seem most common toward the top. All the strata belong to one great lithological individual of heterogeneous grain but rather constant chemical composition. The conglomerate is made up of glassy, white, or bluish, often opalescent quartz pebbles with subordinate, large rounded grains of feldspar. These fragments are all cemented in a silicious matrix, itself feldspathic to some extent. The grits and sandstones are but finer grained phases of the same silicious, sedimentary material. In composition and the gray and greenish colours of fresh and weathered surfaces, the conglomerate and grit can hardly be distinguished from staple phases of the Wolf grit of the southern Selkirks. The similarity even extends to such a detail as the changeable tints of the opalescent quartz pebbles and grains. No fossils were found in this division of the Bow River formation, though it was apparently within this subdivision that Dawson found Lower Cambrian fossils at Ver- milion Pass. It would be highly desirable to have studied in, the field the Bow River beds below the conglomerate-grit member and also the upper part of the Castle Mountain formation, but sufficient time for this could not be spared out of the field season. Yet it is believed—and Mr. McConnell, to whom the field data and typical specimens were submitted, agrees in the belief—that sufficient evidence has already been secured to suggest the stratigraphic relation of the Castle Mountain-Bow River group to the old sedimentary prism traversed at the Forty-ninth Parallel. The suggested correlation is as follows. The lower 4,000 feet or more of the Castle Mountain limestone is stratigraphically equivalent to the Siyeh formation and thus to the larger part of the Kitchener quartzite, and, again, to the larger part of the Beehive quartzite. The 1,500-foot quartzite immediately under- lying the Castle Mountain limestone is the equivalent of the Grinnell and Wigwam formations, of the lowest beds of the Kitchener quartzite, and of the Ripple quartzite. The Bow River conglomerate-grit member at the base of the Laggan section is equivalent to the Dewdney quartzite and upper part of the Wolf grit formation in the southern Selkirks. Since the foregoing paragraphs were written, Walcott has made a detailed study of the Castle Mountain group. His results corroborate MeConnell’s stratigraphy and show yet more precisely the range of the Upper, Middle, and REPORT OF THE CHIEF ASTRONOMER WT SESSIONAL PAPER No. 25a Lower Cambrian horizons in the great series.* Fossils were obtained in sufii- cient abundance to show that the base of the massive dolomitic limestone is the plane separating the Middle Cambrian from the Lower Cambrian in the region. McConnell’s early view of the correlation is, therefore, finally estab- lished. It follows that, if the Siyeh formation and Castle Mountain dolomite are synchronous deposits, a critical horizon in the Forty-ninth Parallel series has been somewhat definitely fixed. Walcott also found Lower Cambrian fossils in the Lake Louise formation, at a zone about 3,000 feet below the top of the Bow River group. Summary.—The evidences for the correlation may be restated in summary form. In composition, in colours of fresh and weathered surfaces, in character of bedding and general influence on mountain forms, the Siyeh and Castle Mountain limestones are almost identical. The similarity is specially marked in the occurrence of the highly peculiar molar-tooth structure in both lime- stones. The correlation on these grounds is strengthened through the strong improbability that two magnesian Hmestones of such immense thickness and -of similar characters should have been deposited so near together as these Bow River and Boundary line sections and yet be of widely different dates of forma- tion. The discovery of fossiliferous Castle Mountain limestone in large devel- opment at Nyack creek, only ten or fifteen miles from Siyeh mountain itself, renders this improbability all the more convincing.§ - The Siyeh limestone in the Galton, Clarke, and Lewis ranges is underlain by red quartzitic sandstones which correspond in essential features to the quartzite at the top of the Bow River formation. Certain whitish and massive beds in this quartzite also strongly recall the Ripple quartzite of the Summit series, a formation which, on independent grounds, has been correlated with the Wigwam and Grinnell formations. Finally, the Bow River conglomerate is as strikingly similar to the Monk grit of the Summit series as the Castle Mountain limestone is like the Siyeh. Also on independent grounds the Wolf grit has been correlated with the Appe- kunny quartzite-metargillite which underlies the Grinnell and thus belongs to a stratigraphic horizon below the 1,500-foot Bow River quartzite at Laggan and Eldon. Not only are there close similarities of lithological detail between the northern and southern rock-groups; the succession of formations is alike. The differences between the successive members of the two sections is due simply to the expected differences subsisting between contemporaneous sediments laid down in the one sea-basin. The Castle Mountain-Bow River group of strata is, in a sense, a composite of the entire pre-Silurian geosynclinal as exposed at the Forty-ninth Parallel. The Castle Mountain formation has its nearest lithological equivalent in the extreme eastern, Lewis series at the Boundary *C. D. Walcott, Smithsonian Miscellaneous Collections, Vol. 53, No. 1804, 1908, p. 1 and No. 1812, 1908, p. 167. § Cf. C. D. Walcott, Bull. Geol. Soc., America, Vol. 17, 1906, p. 18. 25a—vol. ii—12 178 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 line. The Bow River grit and conglomerate, where examined, have features: identical with those of the Wolf formation in. the extreme western, Summit series at the Boundary. The Bow River quartzite, where examined, has features like both the Ripple quartzite of the Selkirks and the equivalent Wiewars sand- stone of the Galton range. The systematic position of the Sheppard and Kintla formations, as of their respective equivalents, the Gateway, Phillips and Roosville formations of the Galton series, the Moyie formation of the Purcell series, and the Lone Star formation of the Summit series, is not apparent from their lithological comparison with the rocks of McConnell’s section. All of these Forty-ninth Parallel formations are unfossiliferous and their conditions of deposition (chiefly subaerial or in shallow water) were markedly different from those under which the upper beds of the Castle Mountain series. (dolomitic limestones) were laid down. A further clue to the correlation has been found in the fact that, in the: Belt mountains and to the westward, the equivalent of the Siyeh formation (Marsh-Helena beds) is, over large areas, conformably overlain by the fossil- iferous, Middle Cambrian Flathead sandstone. This sandstone is often coarse, little metamorphosed, and clearly shows its origin as a sandy deposit on the floor of a transgressing sea. This genetic feature seems to be well matched in the character of the massive, coarse sandstone beds occurring at the base of the Gateway formation, immediately above the (Purcell) lava-cap of the Siyeh.. The lithological resemblance, coupled with the similar stratigraphic relations to the common (Siyeh, Marsh-Helena) horizon, suggests that the lower beds of the Lone Star, Moyie, Gateway, and Sheppard formations are, respectively, equivalents of the Flathead sandstone and are thus of Middle Cambrian age. Since the Flathead horizon is well below the recognized top of the Middle Cambrian, since the succeeding Middle Cambrian time was long enough for the deposition of at least 1,500 feet of limestone in the area of the Canadian Pacific section, and since there is perfect conformity in all four of the Forty- ninth Parallel series above the Siyeh or its equivalent, it seems probable that. the argillites and dolomites overlying the Siyeh or its equivalent are all or nearly all of Middle Cambrian age. The foregoing tentative correlations are expressed in Cols. 1, 2, 7, 8, 10 and 12 of Table VIII. §C. D. Walcott, Bull. Geol. Soc., America, Vol. 10, 1899, p. 209. i - to] ‘ aa TABLE VIII. CORRELATIONS IN THE ROCKY MOUNTAIN GEOSYNCLINAL. 1 Sumaip Series, SELKIRK Ranoe, 49° N. Lar. 2 Porcer. Serres, Porcert Rance, 49° N. Lar. 3 Caur p’ALENE Skrrigs (*) 4 SeRTES IN CABINET Rance (*) 5 SeRIES IN PHILIPSBURG Districr (2) Mission 3) 7 Garon SERIES, GALLON Raney, 49° N. Lat. 8 Lewis Series, Cr, AND Lewis Rane 49° N. Lav. 9 Serres in Berr Mounatrns (4) 10 Castie Mounratn-Bow River SERTES (5) 11 Series av BraoksMITH Fork, Urax (5) 12 System Conformity with upper Pa- Erosion surface. Erosion surface. Conformity with upper Pa- Erosion surface. Erosion surface, Conformity with upper Pa- Conformity with upper Pa- Conformity with upper leozoic ? leozoic. leozoic. leozoic. Paleozoic. i | Sherbrooke. Liniest one ;|St. Charles. Limestone ; gray. 1,375 feet. with sandstone layers ; ‘Paget. Gray limestone.| gray. 1,227 feet. 360+ feet. UPPER CAMBRIAN. Bosworth. Gray limestone and shale, 1,855+feet. » targillite,|Mfoyie. Metargillite, shale,|Striped Pcak. Shales and| Striped Peak. Shales and|Meagher. Limestone, grit-|Mathead. Sandstone.|foosville. Metargillite|Kintla. Argillite, quart-|@allatin. Limestone. __ enone TE Sal ecerre tar various atone red and) shaly sandstones, pre-| ty; gray colours. 400 feet.| Lhickness? and quartzite; gray| zitic and dolomitic in- Flathead. Quartzitic dark tints of gray, green,| tints of (deep) gray, red, green. 1,000+fect. vailingly dark red ; rip-| Wolsey. Shale, calcareous|Camp Creek. Sundstones,| andgreenish. 600-+feet.| terbeds; red. 820 feet.) sandstone. etc. 2,000 + feet. green, ete.; shallow wa- ter features. 3,400 + feet. ple marks, ete. 2,000+ feet. y black. 100—300 feet. to reddish. 50—300 feet. in upper part ; green to Flathead. Quartzite ; white 11,700 (?) feet. shales, and limestones. Phillips. | Metargillite and quartzite ; red. 550 feet. Gateway, upper part. Metargillite with quart- zite; gray when fresh, gray and brown weath- ered. 1,850 feet. | i Sheppard, upper part. Dolomite; gray, wea- thering buff. 500 feet. Beehive. Quartzite and me- Kitchener, upper part. Wallace. Shales, more or Blackfoot (called Newland Camp Creek. Red shale, Eldon. Stephen. Cathedral. Limestone ; gray, weathering buff and gray. 2,728 feev. Limestone and shale; gray. 640 feet. Dolomitic| Nounan. Gray limestone! often sandy. 1,041 feet. Bloomington. Gray and greenish limestone and shale. 1,320 feet. Blacksmith. Gray lime- stone. 570 feet. Ute. Gray limestone with shaly layers. 729 feet. Spence. Greenish shale. interbeds of conglomerate and metargillite; weat- hering gray and pale brown. 2,000 feet. Wolf, upper part. Grit, sandstone, and fine con- glomerate; gray, fresh and weathered. 1,000+ feet. Wolf, lower part. Grit, con- glomerate, and sandstone; gray. 1,900+ feet. Monk. Quartzite, metar- gillite and conglomerate ; gray tones. 5,500 feet. Irene Volcanic formation. 6,000+ feet. Trene Conglomerate. 5,000 +-+ feet. Total—32,050+ feet. Unconformity. Priest River Terrane. light brown. 1,400+feet. Creston, upper part. Quartzite, in thick and thin beds; gray and greenish, weathering gray. 3,000+feet. prevailingly gray - green. 2,000 feet. Prichard, upper part. Metargillite and quart- zitic sandstones, with shallow water features ; gray to black. 1,500+ ‘eet. 5,000+ feet. Ravalli, lower part ; litho- logy as above. 3,000+ feet. Prichard, upper part. Metargillites, quartzites, and sandstones; dark bluish, weathering brown. 2,000+ feet. metargillite; gray and greenish, weathering gray and brown. 2,350, feet. a Hefty. Sandstones, with some shaly beds, some- times calcareous; red) and reddish gray. 775) feet. | lite; gray or greenish gray, weathering gray or brownish. 2,600 feet. feet. Lake Louise, Shale; gray. 105 feet. Fairview. Gray quartzite, weathering 600 +feet. brownish. Blackfoot. Thin-beddedli-|Gateway, lower part.|Sheppard, lower part.|Marsh. Redshale. 800 feet.| limestone; gray, weath- 30 feet. : targillite; gray and pale| Quartzite and metargil-| less calcareous, with) by Calkins). Lime-| etc. 0—5,000 feet. mestone with layers of] Quartzite, dolomite,| Buff-weathering dolo-|Helenz. Limestone with faee buffand gray. 1,595|Zangston. Gray lime- greenish, weathering] lite ; greenish and gray,| thin limestone inter-| stones, thin-bedded, si-| Blackfoot. Calcareous ar-| metargillite; buff wea-| andsandstone; various) mite and reddish sand-| some shale; gray, wea- eet. | stone. 498 feet. brown of different tints.| weathering to browns.) beds. Limestones and| licious and ferruginous,| gillites and impure lime-| thering. 4,805 feet. colours. 125 feet. stone. 100 feet. thering buff. 2,400 feet. Brigham. Gray and green- 7,000 feet. 6,000+ feet. calcareous shales wea-| interbedded with more] stones; dark gray to Siyeh. Dolomitic lime-|Siyek. Dolomitic lime-|Empire. Argillite; green- ish quartzite, weather- ther buff. 4,000 feet.| orless calcareous shale.| greenish gray when fresh, stone with much metar-| stone with metargillite} ish gray. 600 feet. ing brown; upper part. St. Regis. Shales and} 5,000+feet. weathering buff and (lime- gillite; gray and green-| interbeds; coloursas in sandstones, purple and stones) gray. 4,000 feet. ish, weathering buff.}| Galton series. 4,100 green. 1,000 feet. 4,000 feet. feet. Ripple. Quartzite; white,|Kitchener, lower part.|Revett. Quartzite ; white,|Ravalli, upper part.|Ravalli. Quartzite bassing Ravalli, upper part. Quart-| Wigwam. Sandstonesand|Grinnell. Metargillite|Spokane. Argillite; deep|Mownt Whyte. Graylime-| Brigham. Lower part. pale pink and yellow;| Quartzite, relatively| massive. 1,200 feet.| Quartzitesand metargil-| (upwards) into purplish) zite; grayish purple and) metargillites;redcolors| and sandstones, gene-| red. 1,500 feet. stoneand shale. 390 feet.| Total thickness, 1,232 very massive. 1,650 feet. | massive; gray, green,|Burke. Silicious metar-| lites; gray, greenish,] and greenish gray argilli-| gray. 4,550 feet. dominant. 1,200 feet.| rally red. 1,600 feet. |Greyson. Argilliteand)|St. Piran, Greenish and] +feet. Dewdney. Quartzite, with) and white, weathering] gillite, with quartzite,) white, and purplish.| te. 2,000 feet. MacDonald. Chiefly|Appekunny. Metargil-| quartzite; gray. 2,000+| gray sandstone.2,705 feet. CHIEFLY MIDDLE CAMBRIAN. MIDDLE CAM- BRIAN. LOWER CAMBRIAN. Creston, lower part. Quartzite, withinterbeds of metargillite ; gray co- lours. 6,500-+feet. Base concealed. Total—20,300 +feet. Prichard, lower part. Me- targillite and quartzite; gray to black. 6,500+ eet. Base concealed, Total—17,200+ feet. Prichard. Metargillite ; dark bluish, banded. 2,000 feet. Sandstones ; thick-bedded to shaly ; gray. 10,000+feet. ‘Base concealed. Total—27,000 + feet, Neihart. Prichard, lower part. Des- cription same as that of upper part. 3,000+feet. Quartzite; light! coloured. 1,000+ feet.; Base concealed. Total—12, 550-18, 000 feet. 4,000 + feet. Base concealed. Total—25,055 feet. Ravalli, lower part. Quart-| zite; greenish gray. Altyn. Silicious dolomite, thin-bedded ; gray and greenish-gray, weather- ing buff. 650 feet. Base concealed. Total—12,100 feet. Altyn. Silicious, often sandy dolomites ; light’ gray, weathering buff. 3,500+ feet. Waterton. Dolomite; gray. 200+feet. Base concealed. Total—13,420 feet. Greyson, lower part. Argil- lite; gray. 1,000+feet. Newland. Silicious_ lime- stone; gray, weathering buff. 2,200 feet. Chamberlain. Argillite ; ais gray toblack. 1,500) eet. Weihart. Gray and green- ish quartzite. 700 feet. Total—14,000+feet. a Unconformity. Continuation of Bow Ri- ver argillites, etc. Base concealed. Total—12, 353 + feet. Cherry Creek Beds. . Unconformity. Base concealed. Total—6, 647 + feet. BELTIAN. ) F. C. Calkins, Bull. 384, U.S. Geol. Survey, 1909, p. 40. al Oe 25a—vol. ii—p. 178 | “Archean”. C. Calkins, Bulls. 384 and 315, U.S. Geol. Survey, 1909 (p. 40) and 1907 (p. 33), 3) CO. ( G D, Walcott, Bull. Geol. Soc. America, Vol. 17, 1906, p. 2. . D. Walcott, Bull. Geol. Soc. America, Vol. 10, 1899, p. 201, and references therein, to Peale and Weed. 178a (°) C, D. Walcott, Smithsonian Mise. Collections, Vol. 638, No. 1812, 1908. ; { a Sib bi sar ao erential oa sel REPORT OF TRE CHI#HFEF ASTRONOMER 179: SESSIONAL PAPER No. 25a CORRELATION WITH THE BELT TERRANE.* Earlier Views on the Belt Terrane.—To one acquainted with the geological literature of the Cordillera, the writer’s foregoing correlation is obviously quite different from that adopted by the United States Geological Survey for these old formations as exposed in the United States. By the present official views of that survey, (1) the whole of the Belt Mountains series which underlies the Flathead sandstone (the Belt terrane) is referred to the pre-Cambrian (pre- Olenellus) or latest ‘ Algonkian’; and (2) the whole of the Lewis series as described by Willis is considered as belonging to the same terrane and to the same age. A brief history of the explorations on which these conclusions are based,. is given by Waleott in his paper on ‘ Pre-Cambrian Fossiliferous Formations.’+ To facilitate the present discussion it is advisable to review the history at least so far as to show the divergence of views on the correlation. In 1875 Dawson had incorrectly referred the Siyeh formation to the Carboniferous, but later he stated the possibility that the Siyeh limestone is the equivalent of MecConnell’s Castle Mountain limestone. He referred all the underlying formations as far as the upper Altyn, to the Cambrian.t In 1883 Davis made several sections through the Belt terrane rocks and referred them provisionally to the Lower Cambrian, though without fossil evidence.§ The next year Peale sectioned 2,300 feet of the terrane, then called the East Gallatin group, and, referred the whole series to the Cambrian.** In 1893 he published a second account of the terrane, calling it the ‘ Belt formation.’ A significant paragraph may be quoted: ‘There is no doubt that after the Belt formation was deposited there was an orographic movement by which the Archean area of nearly the entire region represented on our map south of the Gallatin and Three Forks was submerged just prior to the beginning of the Cambrian, before the Flathead quartzite was deposited. Whether this movement occurred immediately after the laying down of the Belt beds or after an interval is of course the question to be decided, and the decision cannot be posi- tively reached with the meagre data now at hand. I am inclined to think that the subsidence of the Archean continent (or possibly islands) began with the first accumulation of the sediments that formed the lower portion of these beds and was coincident with their deposition throughout the * Recently Walcott has used the adjective “Beltian,”’ a systemic form, to designate the “ Belt terrane” of earlier publications. (C.D. Walcott, Smithsonian Misc. Col- lections, Vol. 53, 1908, p. 169). +C. D. Walcott, Bull. Geol. Soc. America. Vol. 10, 1899, p. 201. ~G. M. Dawson, Report on the Geology and Resources of the Region in the vicinity of the Forty-ninth Parallel, 1875, p. 74; Bull. Geol. Soc. America, Vol. 12, 1901, p. 68; cf. Ann. Report, Geol. Surv. of Canada for 1885, p. 39B and 50-51B. re § W. M. Davis, Tenth Census report, Vol. 15 on Mining Industries, 1886, pp. 697- 702. ** A.C. Peale, 6th Ann. Report, U.S. Geol. Surv., 1885, p. 50. 25a—vol. 1i—124 180 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 entire pertod. It may have been succeeded by an emergence of the land area for a brief period, but the probability is that the interruption to the downward movement, if it occurred, was slight. Next, the widespread pre- Cambrian subsidence preceding the formation of the Flathead quartzite took place, and the Cambrian sea covered large areas that had hitherto been above the sea level. There is a marked difference in the character of the beds of the two groups. Little, if any, induration is seen in the Flathead formation, while the Belt beds are so altered in most cases as to resemble the metamorphic rocks which underlie them, and from the break- ing down of which they were derived. Notwithstanding the metamorphism, there is no mistaking their sedimentary character.’ In 1896 Peale summarized his net conclusion regarding the correlation in the following words: ‘It is possible that further investigation may result in the reference of this formation to the lower part of the Cambrian. At present, however, it is referred provisionally to the Algonkian.’* Weed, Iddings, and Pirsson in several publications issued between the years 1894 and 1899, refer the terrane to the Algonkian, though Weed and Pirsson, after close study of the Castle Mountain (Montana) mining district, wrote: ‘Both the character of the sediments and their position beneath the beds of Middle Cambrian age indicate their similarity to the Bow River beds of the Canadian geologists, in which Lower Cambrian fossils are found. It has, however, been decided to class the beds as Algonkian.’§ In 1898 Walcott made a general study of the terrane as exposed in the Big Belt and Little Belt mountains and in the Helena district. He writes :— ‘The results of my investigation were the discovery of a great strati- graphic unconformity between the Cambrian and the Belt formations; that the Belt terrane was divisible into several formations, and that fossils occurred in the Greyson shales nearly 7,000 feet beneath the highest beds - of the Belt terrane.’t Walcott’s columnar section of the terrane is that given in Col. 9, of Table VIII. In 1902 Willis stated in the following words his correlation of the Lewis series as exposed in the Clarke and Lewis ranges: ‘The oldest formation of the series, the Altyn limestone, is assigned to the Algonkian period on the basis of fossils discovered by Weller in its characteristic occurrence at the foot of Appekunny mountain near Altyn, Montana. These fossils are fragments of very thin shells of crustaceans [chiefly Beltina danai]..............The fossiliferous strata of the Belt +A. C. Peale, Bull. U.S. Geol. Survey, No. 110, 1893, p. 19. * Three Forks Folio, U.S. Geol. Surv., 1896, p. 2, § Bull. U.S. Geol. Surv., No. 139, 1896, p. 139. tc. D. Walcott, Bull. Geol. Soc. America, Vol. 10, 1899. p. 204. REPORT OF THE CHIEF ASTRONOMER 181 SESSIONAL PAPER No. 25a formation in the Belt range are separated from the Cambrian by 7,700 feet of sediments and an extensive unconformity. In the Front range of the Rockies 10,700 feet of apparently conformable strata overlie the fossili- ferous bed, and it is possible that the plane of division between the Algon- kian and Cambrian as determined by paleontologic evidence will be found in this great series. In the upper part of the Siyeh limestone near the _ head of Mineral creek, Weller found some indistinct forms which he con- siders as possibly to be parts of crustaceans. Walcott expresses a similar view, saying: ‘“ Mr. Weller’s suggestion that the fragments possibly represent crusta- cean remains appears to be the most plausible. If from a Devonian horizon they would suggest the genus Licas, or some of its subgenera. It is a case where more material is needed in order to arrive at any definite conclu- SIOH.. In his paper on ‘ Algonkian Formations of Northwestern Montana’ Wal- cott refers the entire Lewis series and its equivalents to the pre-Cambrian (Algonkian) system.f In this conclusion he has been followed by Calkins, Ransome, MacDonald, and Lindgren, all working on the western phase of the Belt terrane in Idaho.g The same view has governed the compiling of the geological map of North America which was prepared for the session of the International Geological Congress held at Mexico City in 1906; the large area of ‘ Neo-Algonkian’ shown in the States of Montana and Idaho represents the Belt terrane. As one of Walcott’s last (1906) papers on this subject shows the trend of opinion among the United States geologists, the more important parts of his table of equivalents has been reproduced in Table IX. of the present report. B. Willis, Bull. Geol. Soc. America, Vol. 13, 1902, p. 317. C. D. Walcott, Bull. Geol. Soc. America, Vol. 17, 1906, p. 17. § W. Lindgren, U.S. Geol. Surv., Prof. Paper, No. 27, 1904, p. 16. F. L. Ransome, U.S. Geol. Surv., Bull. No. 260, 1904, p. 277. D. F. MacDonald, U.S. Geol. Surv., Bull. 285, 1906, p. 43. F, C, Calkins, U.S. Geol. Surv. Bull. 384, 1909, p. 27 182 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 TaBLE [X.—Showing Walcott’s correlations im the ‘Belt Terrane’ ———__— Beit Mrs., Montana. Cambrian, (Flathead ss) Unconformity. Marsh, 800’ Helena, calcareous, 2,400’ Empire, 600'+ Spokane, 1,500’ + Greyson, 3,000’ + Arenaceous, 5,100’ Newiand, ‘Calcareous, 2,200! + Chamberlain, Silicious, 1,500’ Neihart, 700’ Unconformity. Archean. Total, 12,000’ LEWIS AND CLARKE RANGES, MonTAna. No superjacent strata. Kintla, Sheppard, Quartzites, 1,200’ Siyeh, Limestone, 4,000’ Grinnell, Appekunny, Silicious, 3,800’ Altyn, Caleareous and silicious, 700’ Base concealed. Total, 9,700’ Camp CREEK, MISSION RANGE, Montana. Cambrian, Unconformity. la, arenaceous zsray, 1,762’ 2a, calcareous and arenaceous, 1,560’ da. to 3g. Arenaceous, mostly reddish, 4,491’ 4, to Te. Arenaceous, red and gray,. 3,887’ (198’ of li. near Ca@ur D’ALENE District, IpaHo. No superjacent strata. Striped Peak, 2,000’ summit. ) Blackfoot, Wallace, Calcareous and! Calcareous and silicious, 4,805’ silicious, 5,000’ + Ravalli, — silicious and arenaceous, purple, greenish and gray beds. 8,255’ Burke—St. Regis, Silicious and are- naceous, purple, greenish an gray beds, 8,000’ Base concealed. Total, 24,770’ Prichard, Banded, dark blue gray, blue black and gray, silicious series. 10,000’ |Base concealed. Total, 25,000’ PURCELL RANGE, IpanHo-Brir. CoLUMBIA. No superjacent strata. Moyte, Metargillite and quartzite, 3,500’ Kitchener, Quartzite, 7,400’ Creston, Quartzite, 9,500’+ Base concealed. Total, 20,400’ REPORT OF THE CHIEF ASTRONOMER 183 SESSIONAL PAPER No. 25a A more recent table of correlation has been published by Calkins, who ~has similarly equated the Kitchener with the Ravalli and Burke horizons; Wallace and Blackfoot with Newland and Altyn.* The present writer has not been able to agree with these correlations. As stated in his summary report for 1905, the Kitchener quartzite passes into, and is the equivalent of the Siyeh limestone; and part of the Creston quartzite is equivalent to the Appekunny metargillite. In 1907 the writer came to suspect that the Blackfoot limestone is the equivalent of the Siyeh limestone, and the next year Walcott proved this to be the case.t It seems necessary, therefore, to make significant alter- ations in Calkins’ table. His ‘ Newland’ limestone in the Philipsburg and Cabinet Range districts, if the equivalent of the Blackfoot limestone, as he states, must belong to the Siyeh horizon. The other members of each strati- graphic column must be correspondingly shifted. These changes are noted in the general table VIII. Though it may be at fault in details, that table is believed to express the main relations subsisting among the different sections in the ‘ Belt terrane.’ Walcott’s discovery of the equivalence of the Blackfoot and Siyeh lmestones has clearly simplified the whole stratigraphic situation in Montana and Idaho. Evidence of Fossils—The principal difficulty in the correlation with recognized systems is, of course, the rarity of fossils which can in any sense determine horizons. The only well characterized fossil horizons yet found in the Belt terrane as defined by Walcott, occur in the lower part of the Greyson shales of the Belt mountains and in the upper part of the Altyn formation in the Lewis range. These two horizons may well be practically contemporaneous, as they occur in similar stratigraphic relations and both carry the abundant species, Beltina danai. Neither that species nor any of the associated obscure organisms can directly date the horizon, which has never been found in undoubted association with the Olenellus zone or other general horizon of the Cambrian. So far as purely paleontological evidence is concerned, it is quite within the bounds of possibility that the Beltina horizon is really a lower phase of the Lower Cambrian, Olenellus zone, or is but slightly older than that zone. No organic remains giving decisive indications of age have been found in the overlying Spokane, Empire, Helena, and Marsh formations of the Belt mountains or in the equivalents of these, either at the Forty-ninth Parallel or in the thick deposits of Idaho and western Montana. Walcott recognizes the Cambrian-Ordovician equivalent of McConnell’s Castle Mountain group as occurring near Belton, Montana, and at Nyack creek, Montana.§ At these localities, massive bluish and greenish limestones bearing. a species of Raphistoma and a Stromatoporoid form, were found in great development. As shown by Plate~6 of Walcott’s paper, the field-habit of these limestones is extremely similar to that of the Siyeh limestone at Mt. Siyeh, *F. C. Calkins, Bull. No. 384 and Professional Paper No. 62, U.S. Geological Survey, 1909 and 1908. + Information supplied by letter. § Bull. Geol. Soc. America, Vol. 17, 1906, pp. 12, 19, 22. 184 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 which is less than 15 miles distant from the Nyack creek locality. It is difficult to avoid the suspicion that these Castle Mountain limestones are, in truth, identical with the Siyeh limestone, in which, therefore, Middle Cambrian fossils may at some future time be discovered. Walcott has, however, come to a quite different view. He writes: “The series of limestones at the head of Nyack creek, illustrated by plate 6, are of Cambrian or Ordovician age, as indicated by fragments of fossils that I found in them. I do not think the Siyeh limestone is to be correlated with them, nor with the Castle Mountain limestones of McConnell.’ Walcott’s latest correlation paper for this region contains a section on the Dearborn river, which carries both Middle and Lower Cambrian fossils.t+ The fossiliferous rocks are quite conformable and show a thickness of 2,205 feet; they are chiefly massive or thin-bedded limestones (deseribed as weathering yellow to buff at some horizons), with interbedded shales. The description of the rocks is of essentially the same quality as that which must be applied to the Siyeh formation. Only a few miles away another section of somewhat similar beds, with, however, dominant argillites, had been measured by Walcott, and the whole referred to the Algonkian as part of the ‘Belt terrane.* No statement is given as to the precise relation of these two sections except the following (page 208 of the 1908 paper): _ ‘Beneath the Cambrian sandstone the Empire shales of the Belt Terrane of the Algonkian occur with apparently the same strike and diy as the base of the sandstone. Traced on the strike, however, they appear to- be unconformably beneath the sandstone.’ If the apparent unconformity should be explained in the manner suggested (on a later page) by the present writer, it follows from these studies at Dear- born river that the Empire shale is either Lower Cambrian or is not signifi- eantly older than the Lower Cambrian. In favour of the writer’s view is the fact that these Dearborn river sections occur at points more than 100 miles distant from the old shore-line zone of the Belt mountains, that is, at such a part of the geosynclinal downwarp where the Lower Cambrian beds should be: expected to appear in fairly full development. We have, therefore, at Nyack creek, Belton, and the Dearborn river, three localities where fossiliferous Cambrian formations lie, respectively, side by side with typical members of the ‘ Belt terrane.’ At one or more of these points some geologist may be fortunate enough to find the paleontological evidence which will, at no distant day, fix the position of the Belt terrane among the- standard geological systems.t TBull. Geol. Soc. America, Vol. 17, 1906, p. 19, +7 C. D. Walcott, Smithsonian Miscellaneous Collections, Vol. 53, 1908, p. 200. *C. D. Walcott, Bull. Geol. Soc. America, Vol. 17, 1906, p. 8. t+ In this connection it is of interest to note that, as reported by Wood in 1892, “in the vicinity of Missoula, a few fossils were obtained in the silicious limestone (dolo-- mite) and identified by Mr. Charles Schuchert as Obolella.” The relation of this for- mation to the Belt terrane is not stated. Herbert Wood, Amer. Jour., Sci., 3rd ser.,. Vol. 44, 1892, p. 404. REPORT OF THE CHIEF ASTRONOMER 185 SESSIONAL PAPER No. 25a The argument that the Lewis, Galton, and equivalent series should be referred to the pre-Cambrian because they are almost or quite unfossiliferous is a dangerous one. Most of the known Cambrian strata of the world are quite unfossiliferous, as far as present knowledge goes. They have been assigned to this system because a few, generally very thin interbeds bear determinable index species. Even in the Mount Bosworth section of British Columbia, one of the most highly fossiliferous among Cambrian series, Walcott found no traces of life in 1,855 feet of continuous strata.* The famous Ogygopsis shale at Mt. Stephen is clearly a lens. It peters out rather rapidly and is not repre- sented in either the Mt. Bosworth section or at Castle mountain. Except for a few worm borings, a massive dolomitic limestone totalling 1,680 feet in thickness, at Mt. Stephen, is unfossiliferous.t A thousand feet of the calcareous Nounan formation of Walcott’s Blacksmith Fork section in Utah is as poor in organic remains. Why interbeds similar to the Ogygopsis shale fail to appear in the Forty-ninth Parallel section is not apparent. Barrell has suggested a continental origin for much of the Belt terrane sediment, but we have seen that this is true of probably but a small part of the series. That chitinous fossils are relatively abundant at Mount Stephen and very rare in contempor- aneous marine sediments one hundred miles away is not more difficult to under- stand than that chitinous fossils often occur in the Cambrian and generally do not occur in equally unmetamorphosed pre-Olenellus strata. The one contrast means conditions different in space; the other, conditions different in time. In each case explanation is needed. While awaiting complete explana- tion we must regard this negative character of the ‘ Belt terrane’ as of little direct value in correlation. Relative Induration and Metamorphism of the Belt Terrane and Flathead Formation.—One of Peale’s arguments for the Algonkian age of the Belt terrane is noted in the first of the foregoing quotations from his writings. The point consists in the recognition of a much greater degree of metamorphism in the Belt terrane rocks. as compared with the ‘little, if any, induration’ of the Flathead sandstone. The weight of this argument is considerably lessened by reason of the fact that the Belt terrane where exposed in other regions, is often little folded or sheared and is scarcely at all affected by dynamic metamorphism. Its rocks have truly been well indurated and largely reerystallized under deep-burial conditions, but such alteration by static metamorphism is not of itself evidence of great difference of age between older underlying beds and the younger beds of a rock group. McConnell and Walcott have proved that the Cambrian period was long enough for the accumulation of 11,500 feet of strata, chiefly the slowly deposited limestone, in the Mt. Bosworth district of British Columbia.t In the same period of time, shales, sandstones, and subordinate limestones might have elsewhere accumulated to even greater thicknesses. It is reasonable *C. D. Walcott, Smithsonian Misc. Collections, Vol. 53, No. 1812, 1908, p. 208. +C. D. Walcott, Canadian Alpine Journal, Vol. 1, 1908, p. 232. tCf. C. D. Walcott, Smithsonian Misc. Collections, Vol. 53, No. 1804, 1908, p. 2. 186 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 to expect that the lower part of such a colossal deposit would show more induration than the upper part. Microscopie examination of many specimens has, in fact, showed that in each of the Forty-ninth Parallel series, static metamorphism has operated much more strongly in the older formations than in the younger, quite conformable ones. This rule is, however, not absolute. Many of the Grinnell beds still largely preserve their original clastic structure, while the overlying argillaceous beds of the Siyeh are now typical metargillite. In the field these Grinnell strata look as young as the Carboniferous shales in the mountains farther north. The writer was much struck with the relatively slight induration of the sandstones at the pase of the Moyie and Gateway formations. Yet there can be little doubt that they are, respectively, thoroughly conformable to. the Kitchener and Siyeh formations and, with these, make a mass of continuous sedimentation. These particular sandstones sre just those which the writer has, on other grounds, correlated with the Flathead sandstone. In all these cases the relative lack of metamorphism is to be attributed more to the peculiar nature of the sandstone than to any great difference of age between each sand- stone lens and the immediately underlying beds. It appears fair to conclude. that the criterion of relative induration does not imply a great erosion-gap between the Belt terrane and the Middle Cambrian sandstone. Evidence of Unconformity.—The one controlling principle used in referring the Belt terrane to the pre-Cambrian consists in postulating a strong uncon- formity between the Middle Cambrian Flathead sandstone and the entire series below the top of the Marsh shale, the uppermost member of the terrane at the original localities. The unconformity is believed by Walcott and his colleagues to be similar to that found between the Middle Cambrian Tonto sandstone and the tilted Chuar series in the Grand canyon of the Colorado. In his original announcement of the westward extension of the unconformity beyond the region where Peale had first suspected its existence, Walcott wrote as follows :— ‘The unconformity now known proves that in late Algonkian time an orographic movement raised the indurated sediments of the Belt terrane above sealevel, that folding of the Belt rocks formed ridges of considerable elevation, and that areal (sic) erosion and the Cambrian sea cut away in places from 3,000 to 4,000 feet of the upper formations of the Belt terrane before the sands that now form the middle Cambrian sandstones were deposited.’* In one of his later publications Walcott states that :— ‘One hundred miles farther north the section appears to be conformable from the Ordovician down through the Middle Cambrian and the Lower Cambrian of the Bow River series, and not to reach down to the Algonkian *C. D Walcott, Bull. Geol. Soc. America, Vol. 10, 1899, p. 218. REPORT OF THE CHIEF ASTRONOMER 187 SESSIONAL PAPER No. 25a as it occurs in Montana, the Bow River series being the sediment deposited, in part, at least, in the erosion interval between the Algonkian and the Middle Cambrian.’+ In another place he writes:— ‘Absence of Lower Cambrian rocks and fauna is accounted for by the fact that that portion of the continent now covered by the Belt and associated middle and upper Cambrian rocks was a land surface during ~ lower Cambrian time.’t The detailed work of Weed and Pirsson resulted in the definite conclusion that during the deposition of the Belt terrane, there was a land area covering the region north of Neihart in the Little Belt mountain district. -While the Belt beds were being laid down the pre-Belt rocks were reduced to a nearly level plain. In Flathead time there was a submergence of the old peneplained surface, with a resulting overlap of the sandstone and shale upon the pre-Belt formations.* Similarly, during most or all of the Belt terrane period there seems to have been land in the southern and eastern parts of the Livingston folio area in Montana (see folio); in the area covered by the Yellowstone National Park**; in the Absaroka quadrangle (see folio); in the Black Hills area of South Dakota and Wyomingtt; in the Bighorn mountains and vicinity.§ There thus seems to be little doubt that the Belt terrane sediments were in part supplied by the erosion of a large land area covering South Dakota, Wyoming, and eastern Montana. In part they were supplied from the moun- tainous pre-Cambrian land of western Idaho, Washington, and Oregon. In other words, the rocks of the Belt terrane were laid down in the relation of a typical geosynclinal prism elongated in a meridional sense between the two land areas. The unconformity postulated by Walcott and his colleagues has been deduced from a study of the eastern shore-zone of the ancient gulf or sea. The irregularities of such a coast line, coupled, it may be, with minor oscil- lations of level, would necessarily involve maximum variations of thickness in the different sedimentary lenses of the geosynclinal. The lenses must thin to nothing either at the actual shore-line, on the rims of off-shore depressions, or at the outer edge of the coastal shelf which was swept by waves and currents during a long period of stationary sealevel. The resulting irregularities of deposition are homologous to those observed in the section of a river delta which has grown out into sea or lake; those irregularities are necessarily most pronounced near the shore. The failure of individual members of the Belt terrane to appear beneath the Flathead sandstones cannot, therefore, be directly 7 Bull. Geol. Soc. America, Vol. 17, 1906, p. 16. t Bull. Geol. Soc. America, Vol. 10, 1899, p. 210. *U.S. Geol. Surv., Little Belt Mountains folio, 1899, and Fort Benton folio, 1899. ** See Yellowstone Park folio and U.S. Geol. Surv., Monograph 32, part 2, 1899. 7t See the various United States government publications on the Black Hills. § See the Hartville, Aladdin, and Sundance folios (1903-5) of the U.S. Geol. Sur- vey: also N. H. Darton’s Geology of the Bighorn Mountains, Prof. Paper, No. 51, 1906. 188 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 taken to mean an erosion unconformity or structural unconformity of the Belt and Flathead beds by the amount of missing strata at any one or more localities. The section which, according to Walcott, most clearly shows the extent of the unconformity is that running eastward through the Spokane Hills. In the diagram illustrating the relationships there, the Middle Cambrian beds are . represented by Walcott as conformably overlying the Helena formation both at Helena and near White’s canyon in the Belt mountains uplift.* These two localities are twenty-four miles apart. Nearly midway between them, at the Spokane Hills, the Middle Cambrian ig represented as again conformable on the Belt beds but this time resting on the Spokane shales. Thus fully 3,000 feet of strata, the thickness of the Helena and Empire formations, are con- sidered as lacking beneath the Middle Cambrian at the Spokane Hills. The text accompanying the diagram does not state whether the fossiliferous Cambrian at the Spokane Hills is the stratigraphic equivalent of the Flathead sandstone. Mr. Walcott has, by letter, very kindly informed the writer that the Middle Cambrian beds at the Spokane Hills are not only faunally but litho- logically the equivalent of the Flathead. There thus seems to be an actual failure of at least 3,000 feet of Belt beds at the Spokane Hills. Whether the failure is due to a lack of original deposition or to the erosion of a local upwarp of the Belt beds is apparently not an easy question to decide. Walcott has taken the latter view for this locality. On the other hand, he himself writes, concerning somewhat similar relations about the town of Neihart :— ‘Whether the shore-line conditions, which are known to have existed near Neihart during the period when the Belt terrane was formed, causing a wedging out of the beds to the north, so that the Cambrian rests on different horizons at this locality, or whether pre-Cambrian erosion was extensive enough to pare down the exposed edges of the beds, is not certain from the evidence, though the latter view seems improbable.’ All workers on these Montana rocks have observed that, wherever the Flathead sandstone is seen in contact with Belt formations, no important angular discordance of dip can be seen. Such slight discordances as have been described and figured by Walcott in his 1899 paper, can be explained either by slight, perhaps submarine, erosion of the older surface, or by local and very gentle warping of the surface just before the Flathead subsidence. As early pointed out by Peale, Flathead time saw a general, rapid, but not very pronounced subsidence of the western Montana region. A large supply of quartzose debris was thus brought from the drowning land and depos- ited alike over Archean schists and the various lenses of the geosynclinal. In this way a fairly homogeneous formation was spread over a sedimentary mass which, in the nature of the case, must have been composed of many and varied lenses all of which petered out toward mainland, island, or shallow. *See C. D. Walcott, Bull. Geol. Soc. America, Vol. 10, 1899, p. 211. + Ibid, p. 210. REPORT OF THE CHIEF ASTRONOMER 189 SESSIONAL PAPER No. 25a In brief, it seems to the writer that the facts so far recorded do imply a sedimentary overlap of the Flathead but not a great structural unconformity, or even erosion unconformity, which is general at the base of the sandstone. Considering the size of the area, the observed minute discordances of dip can- mot be used safely as positive evidence. The observed failure of beds at certain points can be explained by original wedging-out or by the quite moderate erosion of local upwarps just before the Flathead subsidence. It must be remembered that Middle Cambrian time was exceedingly long. It sufficed for the deposition of 5,000 feet of limestone in the Canadian Rocky mountains at the localities recently studied by Walcott.* During the deposition of such a slowly accumulated sediment, there was evidently plenty of time for local upheavals, considerable erosion, and renewed subsidence along the border of the Cambrian sea. It took only a portion of Pliocene time (next to the Pleis- tocene, probably the shortest of all the major divisions of geological time) to form 5,800 feet of sediments represented in the Merced series of California, a series which itself rests on a Pliocene land-surface.t Summary of Conclusions.—In view of the foregoing conclusions the writer does not believe that the pre-Cambrian age of the upper part of the Belt terrane as defined by Walcott, is proved; and regards the Helena-Siyeh formation as probably Middle Cambrian, somewhat older than the Flathead sandstone. On the supposition that the Lewis series and the original Belt terrane have been correctly correlated, that terrane as far down as the upper part of the Greyson shale is tentatively considered to be of Middle and Lower Cambrian age. From the lower part of the Greyson shale to the base of the Neihart formation the beds are correlated as pre-Cambrian (pre-Olenellus) but conformable to beds equivalent to the Olenellus zone elsewhere. The name ‘Belt terrane’ (or Walcott’s ‘ Beltian’), for the remainder of this report, is restricted to this pre-Cambrian portion of the great geosynclinal prism. Table VIII. presents a résumé of the writer’s tentative correlation of the Forty-ninth Parallel series with the formations described to the south of the Boundary line. The Ccur D’Alene series has been tied on to the Purcell and Summit series through lithological resemblances. Calkins has traced the Prichard formation northward, where he found it to pass into the Creston quartzite; we have seen that the Creston is the off-shore equivalent of the Wolf and Monk formations. The special white colour and massive appearance of the Revett quartzite are duplicated in the Ripple quartzite, the two doubtless repre- senting another definite common horizon for the two series. Both series are, in the table, tied on to the Lewis series and thus indirectly to the more fossili- ferous series on the line of the Canadian Pacific railway. The table embodies, with some modifieations, the correlation of the (Lewis series, Belt series, and. the Camp Creek-Blackfoot-Ravalli group, as suggested *C. D. Walcott, Smithsonian Misc. Collections, Vol. 53, No. 1804, 1908, p. 2. a Cia C. Lawson, Bull. Department of Geology, University of California, Vol. 1, 1893, p. 142. j 190 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 by Walcott.* The whole Purcell series as shown in Table IX. is certainly placed by Walcott much too low in the geological scale. His reason for making that particular correlation was probably in part due to the following statement of the present writer’s in his summary report for 1904 (p. 97) :— ‘The nearest relatives of the Creston and Kitchener quartzites in the Rockies are respectively the two thick members of the Altyn limestone delimited by Mr. Bailey Willis, who, in the year 1901, carried out a recon- naissance survey of the Boundary belt on the Montana side.’ The expression ‘the nearest relatives’ should have been ‘the nearest lithological relatives, as the intention was to note the lithological relations of the Purcell and Lewis series, rather than to imply equivalence of age among individual members. As a matter of fact the Altyn formation is believed to be the stratigraphic equivalent of a part of the Creston quartzite. On the.other hand, we have seen that the facts point towards the correlation of the Kitchener quartzite with the Siyeh and Grinnell formations of the Lewis series. The somewhat elaborate correlation Table VIII. is intended to illus- trate a suggestion rather than a proof. The unfossiliferous rocks of the Forty-ninth Parallel were approached by Dawson, McEvoy and others from the north, where lithologically similar formations bear Cambrian fossils; and, somewhat naturally, regarded the thick quartzites, etc., to the south as probably Cambrian. The United States geologists have as naturally refused to place the nearly unfossiliferous Belt terrane in the same part of the geological column as the formations of Utah and Nevada, where Cambrian fossils are not rare. _ The present writer has had to rely chiefly on lithological characters in making correlation and his tentative conclusion may be ultimately proved to illustrate once again the danger of using this criterion. It is certain, however, that the pre-Cambrian age of the Belt terrane is not proved, and we are yet at the stage where all reasonable correlations should be fully stated and carefully examined. By the writer’s suggested view the Eastern half of the Cordillera carries a simple Paleozoic-Beltian geosynclinal prism which is only locally interrupted by uneonformities. The pre-Ordovician thickness of this prism has an observed maximum of about 30,000 feet. According to the view of Walcott and his former colleagues in the United States Geological Survey, the Eastern belt of the Cordillera carries what may be called a compound geosynclinal prism, made up of a pre-Cambrian series reaching observed thickness of about 30,000 feet, separated by a strong erosion unconformity from a Cambrian series reach- ing a maximum observed thickness of at least 20,000 feet. The pre-Ordovician sedimentaries, excluding such huge series as those represented in the Priest River terrane, the Cherry Creek beds of the Belt mountains, the Red Creek quartzite of the Uinta mountains, ete., are thus credited with some 50,000 feet of maximum thickness. By the writer’s view the Eastern Cordilleran belt (including the Great Basin), from the Yukon boundary to northern Arizona, was the scene of * Bull. Geol. Soc. America, Vol. 17, 1906, p. 18. REPORT OF THE CHIEF ASTRONOMER 191 SESSIONAL PAPER No. 25a generally uninterrupted) sedimentation through Cambrian time. Walcott’s correlation involves the conclusion that a very large area included within southern British Columbia and Alberta, much of Idaho and of western Montana, represents more or less continuous land (Belt terrane), separating the Cambrian basin of the Canadian Rockies from the Cambrian basin of Utah and Nevada. These fundamentally different conceptions are important not merely in strati- graphy; they should be in the mind of anyune who attempts to decipher the conditions under which orogenic forces built the ranges of the Great Basin and the Front ranges of Montana and Alberta. “¢ CORRELATION WITH Dawson’s SELKIRK AND ADAMS LAKE SERIES. Shortly before his death George M. Dawson read before the Geological Society of America a paper summarizing his views regarding the geology of the Canadian Cordillera.* It is fortunate for the science that he was enabled to complete this able review of his discoveries during a quarter of a century of nearly continuous exploration in the mountains. In the delicate and principal matter of correlation no other person could have so authoritatively digested Dawson’s numerous reports along with the others published before the year 1900. The reader of his summary will note how Dawson used his accus- tomed scientific caution in making correlations among the older rocks of British Columbia. In so brief a review of a vast area it was inevitable that all of his doubts and qualifications could not be expressed. Still more in his original government reports he shows how other interpretations might be deduced as field work progressed. Somewhat different correlations are, in fact, suggested by the field data at the Forty-ninth Parallel. The present writer believes that the lithology of those sections as described, is sufficiently similar to that of the Forty-ninth Parallel formations to warrant certain tentative correlations within the Selkirk mountain system. Where all is so difficult in the study of these unfossiliferous groups of strata, it is well to entertain all possible views of the relations until accumulating facts shall narrow down the alternatives. Dawson had found in the Selkirk range and the ‘Gold ranges’ (the Columbia system of the present report) three thick groups of rocks, which he named the ‘ Nisconlith series,’ the ‘Selkirk series, and the ‘Adams Lake series. All three were referred to the Cambrian and each series was regarded by Dawson as a stratigraphic equivalent of some part of the Castle Mountain- Bow River group of the Rocky Mountain range. For the purposes of the present discussion no briefer, more accurate way of presenting Dawson’s salient ecnelusions concerning these series can be devised than to: quote his own summary in full. He wrote:— ‘Passing now to the next mountain system, to the southwest of the Laramide range and parallel with it—the Gold ranges—we find in the *G. M. Dawson, Bull. Geol. Soc. America, Vol. 12, 1901, pp. 57-92. 192 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Selkirk mountains a great thickness of rocks that have not yet yielded any fossils, but appear to represent, more or less exactly, the Cambrian of our typical section. Resting on the Archean rocks of the Shuswap series is an estimated volume of 15,000 feet of dark gray or blackish argillite schists or phyllites, usually calcareous, and toward the base with one or more beds of nearly pure limestone and a considerable thickness of gray flaggy quartzites. To these, where first defined in the vicinity of the Shuswap lakes, the name Nisconlith series has been applied. The rocks vary a good deal in different areas, and on Great Shuswap lake are often locally represented by a considerable thickness of blackish flaggy limestone. I other portions of their extent dark-gray quartzites or gray- wackes are notably abundant. Their colour is almost everywhere due to carbonaceous matter, probably often graphitic, and the abundance of carbon in them must be regarded as a somewhat notable and characteristic feature. These beds have also been recognized in the southern part of the West Kootenay district and in the western portion of the Interior plateau of British Columbia. ; ‘The Nisconlith series is believed, from its stratigraphical position and because of its lithological similarity, to represent in a general way the Bow River series of the adjacent and parallel Laramide range, but there is reason to think that its upper limit is somewhat below that assigned on lithological grounds to the Bow Kiver series. ‘Conformably overlying the Nisconlith in the Selkirk mountains, and blending with it at the junction to some extent, is the Selkirk series, with an estimated thickness of 25,000 feet, consisting, where not rendered micaceous by pressure, of gray and greenish-gray schists and quartzites, sometimes with conglomerates and occasional intercalations of blackish argillites like those of the Nisconlith. These rocks are evidently in the main equivalent to the Castle Mountain group, representing that group as affected by the further and nearly complete substitution of clastic materials for the limestones of its eastern development. “In the vicinity of Shuswap lakes and on the western border of the Interior plateau, the beds overlying the Nisconlith and there occupying the place of the Selkirk series are found to still further change their character. These rocks have been named the Adams Lake series. They consist chiefly of green and gray chloritic, feldspathic, sericitic, and some- times nacreous schists, greenish colours preponderating in the lower and gray in the upper,parts of the section. Silicious conglomerates are but rarely seen, and on following the series beyond the flexures of the mountain region it is found to be represented by voleanic agglomerates and ash-beds, with diabases and other effusive rocks, into which the passage may be traced by easy gradations. The best sections are found where these materials have been almost completely foliated and much altered by dyna- mice metamorphism, but the approximate thickness of this series is again about 25,000 feet. REPORT OF THE CHIEF ASTRONOMER 1938 SESSIONAL PAPER No. 25a ‘The upper part of the Cambrian system, above the Bow River and Nisconlith series, may thus be said to be represented chiefly by limestones in the eastern part of the Laramide range, cale-schists in the western part of the same range, quartzites, graywackes, and conglomerates in the Selkirk mountains, and by voleanic materials still further to the west. It is believed that a gradual passage exists from one to another of these zones, and that the finer ashy materials of volcanic origin have extended in appreciable quantity eastward to what is now the continental watershed in the Laramide range. No contemporaneous voleanic materials have, however, been observed in the underlying Bow River or Nisconlith series.’** The writer has studied Dawson’s original reports with a view to understand the grounds of the correlations mentioned in the foregoing quotation. Unfor- tunately the arduous and rapid nature of his reconnaissance surveys prevented Dawson from constructing columnar sections in detail sufficient to make inten- sive lithological comparisons possible. Nevertheless, the more detailed facts certainly seem to warrant the belief that the Selkirk series is, in the main, equivalent to the Summit series of the Forty-ninth Parallel section and to the Castle Mountain-Bow River group of McConnell’s section. On the other hand, any satisfactory conclusion as to the relation of the Nisconlith-Adams Lake terrane to the formations mapped at the International Boundary could not be reached without further field-work. Since the forward- ing of the original manuscript of this report for publication, the writer has spent a season in the principal area, along the main line of the Canadian Pacific railway, where Dawson studied these old rocks. At the time of the present writing (November, 1911), the results of that season’s work are not fully compiled, but certain of them, bearing on the question of correlation, are: already in shape for definite statement. The writer has been forced to differ from Dawson in several moriant conclusions. The evidences in each case are necessarily too detailed to be stated in the present report, wherein the writer’s relevant conclusions only will be briefly noted, as follows :— 1. The ‘ Nisconlith’ series of the Selkirks, as sectioned by Dawson between Albert Canyon station and Glacier House, represents the northern continuation of the Beltian (Belt terrane) rocks at the Forty-ninth Parallel, and conformably underlies the thick quartzites of the Selkirk series, which are probably of Cambrian age. The writer believes that these ‘ Nisconlith’ rocks of the Selkirk mountains should logically be included in the Selkirk series. 2. The ‘ Nisconlith’ series of the Shuswap lakes area is an _ entirely different, pre-Cambrian and pre-Beltian, group of sediments, which underlie the ‘ Nisconlith’ of the Selkirks unconformably. *Tbid. pp. 66-7. In the second volume of the same bulletins (1891), p. 165, Daw- son treats the Selkirk section at greater length, giving a structure-section and table of correlations. He attributes nearly 40,000 feet of thickness to the Cambrian alone. 25a—vol. 11—13 194 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 3. The Adams Lake volcanic series conformably overlies the thick lime- stones of the ‘ Nisconlith’ series in the Shuswap lakes area and is likewise of pre-Beltian age. 4, The ‘Shuswap series’ of the Shuswap lakes region is not a distinct gneissic group unconformably underlying the ‘ Nisconlith’ series, but repre- sents a facies of the ‘ Nisconlith’ series of the same region, where the latter has been specially metamorphosed. This metamorphism is thermal and is largely due to batholithic intrusion. The batholiths are pre-Beltian in age. 5. In many essential respects the lithology of the Priest River terrane corresponds with that of the ‘ Nisconlith’-Adams Lake group of the Shuswap lakes. In a general way those two pre-Beltian groups may be tentatively correlated. The correlations suggested by the new facts are summarized in Table X. TaBLE X.—Correlation with Canadian Pacific Railway Section. WESTERN PART | won ats Ep E 2 SELKIRK RANGE, 49TH) OF COLUMBIA | SELKIRK BANGE ae pide on re PACT: RANGE; INTERIOR| Ee RANGE; Bow RIVER AGE. PLATEAUS. | AN. FAC. TY. SECTION. Summit Series: | | WONCUS GAT A cs cso eels ate eee | Selkirk Series, upper|Castle Mountain|Middle Cam- BS COLULVE reteverste ch ovece- ne ltove ore aa tisssonvnstts J\ part. Series, lower part. brian. set ee ee el — (eee ne RTP PLE eee sioleeerccs fis) seseajorceane cracuereiols Selkirk Series, middle Bow River Series,\Lower Cam- IME waney ee os astisiee| (saci wilae mene: | part. | upper part. brian. Wrolftupper part): ss|i.e>a In specimens which appear to have been slightly altered, the hornblende is still compact but the colours are considerably paler, so as to give the mineral the look of actinolite. A further stage of alteration is represented in a fibrous phase of the amphibole, suggesting uralite in colour and other essential respects. This fibrous amphibole is so common in the slides that it was at first believed that it might be secondary after a pyroxene. A close study of a Jarge number of thin sections has, however, led to the conclusion that the fibrous amphibole is really secondary after the compact form. All stages of transition can‘ be found between the two, and the fibrous type has demonstrably grown at the REPORT OF THE CHIEF ASTRONOMER 223 SESSIONAL PAPER No. 25a expense of the other, which has been simultaneously decolourized. No trace of any pyroxene or of pseudomorphs of pyroxene has been discovered in any slide. Many of the sliced specimens are so fresh, as shown by the preservation of the essential minerals as well as of biotite, that the pyroxene must be dis- coverable if it had ever entered into the composition of the rock at the time of erystallization from the magma. Another hypothesis, that some of the fibrous hornblende has resulted from the speedy alteration of originally crystal- lized pyroxene, through the influence of magmatic vapours which acted long before the rock was exposed to ordinary weathering, cannot be excluded. So far, however, the positive microscopic evidence declares in favour of the first view. Similar cases of the derivation of fibrous amphibole from compact amphi- bole through metasomatie changes are described by Zirkel.* The hornblende is, in the prismatic zone, idiomorphic against the feldspar; it fails to show good terminal planes. The ends of the crystals characteristically run out into narrow forked blades. The extinction on (010) averages about 138° 30’; that on (110), about 14°. In phases of the rock where quartz is an abundant accessory, the amphibole is often highly poikilitic, the prisms being charged with swarms of minute droplets of quartz. For the purpose of finding the optical orientation the attempt was made to produce etch-figures on the more likely looking specimens of the amphibole but, on account of the poikilitic and blady character of the amphibole, the attempt was not successful. From the chemical and quantitative analyses of the type rock, a rough calculation of the chemical composition of the hornblende gave the following proportions :— SOs ease ta 49-8 WALZ ORe CHE CLO. OO NO CANCiO MiG.O. SOO” OO Me oO 1G 5-2 TERGE(O Yeo Gree eae eek iarc ih Sc See as rane Sve rg Me ea 5-2 DEQ EE Piste Sino ae GRU Amiens NES he ena ete Ett RR ili 12-1 MGT OR eee seas eteneicrs. Wfeikaed ash eiee eiepyrenl ateee eeapied 2 99-7 The estimate is crude but it shows that the amphibole is a common horn- blende high in silica, iron, magnesia, and lime, but low in alumina. The feldspar is plagioclase, always well twinned on the albite law and often on the Carlsbad law. Many individuals extinguish with angles referring them to labradorite, Ab, An,; some have the extinction angles appropriate to basic bytownite; a very few others are zoned, with anorthite in the cores and andesine in the outermost shell. The average composition of the plagioclase in the normal rock is near that of the basic labradorite, Ab, An,. Magnetite, titanite, pyrrhotite, and apatite are all present but are strik- ingly rare in most of the slides. Their forms and relations are those normal *F. Zirkel, Lehrbuch der Petrographie, Vol. 1, 1893, p. 325.” 224 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 to gabbros. The quartz often bears many fluid inclusions. Chlorite, epidote, leucoxene, and a little calcite are rare secondary minerals. Professor Dittrich has analyzed a specimen of the fresh sill-rock from a point situated about nine miles east of the Moyie river and 1-5 miles north of the Boundary line. This specimen (No. 1153) represents the principal rock type of most of the sills. The analysis resulted as follows :— Analysis of dominant gabbroid type in the Purcell sills. Mol Si0,.. 51-92 865 TOs. °83 010 Al,O,. 14-13 137 Fe,O,. ° 2-97 019 FeO.. 6-92 096 MnO.. 14 001 MgO.. 8-22 205 CaO 11-53 205 Nias Oana: 1-38 023 OTe a oid Reece Runge dabe can GmeeU ae MCae ee UeiN Ue ahind -aauuee 4.7 005 ROR AG TOS Shs Ce EARS inal TREO DETER NEHA TE Nn ar, 10 stephe H.O above 110°C., Ref MISE Apacs tes Ete eee ines 1:07 Prater PEO eee ver cesclce) ies 204 COS es 06 99-78 Soe hese te ae oes oer (Commected valie) 2.990 . A fairly accurate optical determination of the weight percentages among the principal mineral constituents (Rosiwal method) gave the result: EPOrmb ened ei: Wis glee pees a Swe, ere bra oa otabeen orauiuats uy aR HG Rtn real Me A lobar eG 58-7 GAT ACOTIGCE eat ren ele aoe ales aees peter, are Webs aks oie Tesayn sta teen Frere Meets Gelayene ener 34:8 Quarto. ea We cel Gea bee we eel te ea eae Lahti ui Ae See a ae 4-0 Titanite and magnetite... .. .. . 1-4 Biotite.. . ; 9 Apatite. . 2 100-0 The comparative poverty in alumina and the high acidity are evidently related to the composition of the hornblende, which has been estimated as above. In some respects the analysis recalls the diorites but both the magnesia and lime, as well as the amount of femic material in the rock, are too high for that class. It seems best, for the present, to place this type among the horn- blende gabbros, although it is to be regarded as an abnormal variety in that class. The standard mineral composition or ‘norm’ of the Norm classification was calculated to be:— REPORT OF THE CHIEF ASTRONOMER 225 SESSIONAL PAPER No. 25a QUaTtZ eae crepe erence ela cas, nowrap avon ie lean oe Mace vere), mt aan ieee tae 6-78 OTE OCT ASS ere ta ae lees teh atom tice ESE too ea Teh avemiteve, “Tare Metst teven Wave 278 MAUD 1. Ceeaiprrc tay erste ste Tee eletey cara alee oe ae elie SiMe so epeiar en) eisit ate 11-53 PACT OTS EMT GO area rate cosy raves i cociols rie ei pata sleet fe lssysh role Gestsiofs, UMS eieulcsian ence Ten jee 30-86 TO PSIG Orta res oe eeion ict ns ehece, Geis Plan a cee Fess Retesiotonimrs cheer sudeiemnteteeete tole 21-07 Fy MOESEN ONC ce hese eis kee ie ete ias eISe caleba ala aeie re ateieate nels ci tresauitese tte a luere 19-44 FOR Gr caot ne Cte MR anED OR ERC Er a ne AG. coimtmored Blac omoOtOr 1-52 IW IB Vee ah 25 ae eee Ine Gen OIG OT Om CODEC OmiC EG GBS o OGG) OpMmOnErG aco 4.4] H,0 and CO... ee e0 e8 ©8 ©8 ee 28 ©8 2©8 28 #8 88 28% eo ee of oe 1-23 99-62 Accordingly, in this method of classification, the type belongs to the pre- sodic subrang of the percalcic rang, in the order, vaalare, of the salfemane class. The ratio of Q to F in the norm is very close to that which would place the rock in the order, gallare. VARIATIONS FROM THE NORMAL COMPOSITION. Variations from this gabbro type are very common in most of the sills. These generally consist in an increase of quartz and biotite, along with the appearance of orthoclase, which is crystallized either independently or in the form of micrographic intergrowth with quartz. As these constituents increase in amount, the hornblende seems to preserve its usual characters, but the plagioclase shows a strong tendency toward assuming the zoned structure; the cores average basic labradorite, Ab, An,, and the outermost shells average andesine, near Ab, An, When the quartz “and micropegmatite become especially abundant, the plagioclase averages acid andesine or basic oligoclase. In several thin sections the plagioclase is seen to be mostly replaced by ortho- clase and quartz, which, with the still dominant hornblendes, form the essential substance of the rock. These changes in composition, indicating that the sill-rock has become more acid, are always most notable along the contacts and especially along the upper contacts. A good illustration of the acidification along the upper con- tact occurs in a well exposed sill outcropping in the band that runs south from the Boundary line at a point nine miles east of the Moyie river. This sill is about 500 feet thick. A specimen (No. 1) taken twenty feet from the lower contact is unusually rich in hornblende but bears much quartz and orthoclase along with the subordinate essential, acid andesine. It carries no biotite nor micropegmatite but orthoclase dominates over the plagioclase. Specimen No. 2, taken seventy-five feet from the lower contact, is a very similar rock in which the plagioclase is an unzoned labradorite somewhat subordinate to the ortho- clase in amount. Specimen No. 3, taken fifteen feet from the upper contact, is gabbroid in look, though lighter in colour than either No. 1 or No. 2. It is essentially composed of hornblende, quartz, orthoclase, and basic andesine, named in the order of decreasing abundance. The accessories include micro- pegmatite and much biotite, the latter in small, disseminated foils. The essentials are all poikilitic with mutual interpenetrations and enclosures. The structure is quite confused. 25a—vol. ii—15 226 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 All three specimens are very fresh and their densities clearly indicate the acidification along the upper contact. The respective specific gravities are :— INOw 3; . 15 feet below: upper “contact: .'-.<1).s) a) nee eee one 2-983 INO 2en425 cee e SEE AGS Nie Vial nbd eer ee 3-001 No. 1, 480 “ oe * sf Fen SR Slee Set 3-072 Lower contact, 500 feet below upper contact. MOYIE SILLS. Of all the intrusions those outcropping on the isolated ‘ Moyie Mountain,’ immediately west of the Moyie river at the Boundary line, show the most remarkable variations in composition. (Figures 13 and 14; Plate 25.) Some years ago the writer published two papers detailing the petro- graphy of the more important phases of these sills.* The description was based on field work during only a few days in the _ season of 1904. The importance of this particular section was not fully appar- ent until the field season was over and the rock collection had been micro- scopically studied. If time could have been spared during the continued recon- naissance of the Boundary belt, the writer would have early made a second visit to the Moyie sills to test the conclusions of the 1904 season regarding field relations. Unfortunately, no such opportunity for additional personal field work became available. In 1905 an untrained assistant was sent to the locality, and he collected new petrographic material at points along the Boundary slash, as designated by the writer. The character of the specimens thus added to the material in hand seemed to corroborate the general conclusions of the writer and the two publications above mentioned were issued. Thus, in 1905 and 1906, the writer believed that the intrusive ‘rocks oceur- ring on the western slope of Moyie mountain together form a single sill about 2,600 feet in thickness. Such was his belief at the time when the present report was sent in for publication. In 1910, Mr. Stuart J. Schofield was commissioned by the Director of the Dominion Geological Survey to make a geological study of the Purcell range. At the writer’s request, Mr. Schofield examined the section of Moyie mountain at the Boundary slash. He found that the igneous rocks of the western slope really compose three sills, separated by Kitchener quartzite. He also found two thinner sills on the eastern slope of the mountain and in the same Boundary-line section, an area which the writer was not able to traverse in 1904. In 1911 Mr. Schofield guided the writer to his various contacts, all of which were seen to be correctly located in his profile of the mountain. Recent forest fires had cleared the exposures somewhat since 1904 and there can now be little doubt as to the structural relations here- after described. The writer’s sincere thanks are due to Mr. Schofield for his eareful, efficient field-work on this problem. The relations are, therefore, more complex than was formerly believed by the writer. However, it may be stated well in advance that the theoretical con- * American Journal of Science, Vol. 20, 1905, p. 185; Festschrift zum siebzigsten Geburtstage von Harry Rosenbusch, Stuttgart, 1906, p. 203. REPORT OF THE CHIEF ASTRONOMER i 227 SESSIONAL PAPER No. 25a clusions published in 1905 and 1906 as a result of a study of the ‘ Moyie Sill’ are essentially strengthened by the new facts of structure. Gravitative differ- entiation is illustrated not once but thrice, that is, in each of the sills occurring on the western slope of the mountain. It is illustrated a fourth time in the more important of the two sills on the eastern slope. Figure 13. Locality map of the Moyie sills, showing in solid black the parts best exposed. The straight line in the middle of the map represents the Boundary line; the other straight lines represent the approximate outcrops of major faults. The block between the faults includes Moyie mountain. Contour interval is 500 feet, Scale, 1 : 68,000. For convenience, the five sills of Moyie mountain will be distinguished by the letters, A, B, C, D, and E, named in stratigraphic order, with A the highest, E the lowest in the Series. (See Figures 13, 14 and 15). Of these sills, C, D, and E correspond to the whole ‘ Moyie sill’ of the 1905 and 1906 papers. Of the five sills, B is the only one with a sensibly homogeneous composi- tion. Each of the other four presents phasal variations of notable character. 25a—vol. ii—154 228 : DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 With the addition of one type, the list of rock varieties recognised in the early publications will serve for all the bodies to be described. The description of the individual sills may be anticipated by an account of all the phases, MOVIE MOUNTAIN XB A Moyie River ‘alley Figure 14.—Section of Moyie mountain and the Moyie sills, along the International Boundary line. Sills in solid black. Bedding-planes of the quartzite and fault- planes shown. beginning with the most acid one, a granite, and ending with the most basic and ferromagnesian one, a metagabbro or abnormal hornblende gabbro. The following account of petrography and theory will largely consist of a revised edition of that contained in the 1905 and 1906 papers. ABNORMAL BIOTITE GRANITE. In sills A, C, and D the intrusive rock forms distinctly acid zones. The chief constituent is a biotite granite. This is a gray rock, much lighter in tint than the deep green gabbro (Plate 24). The grain varies from quite fine to medium. Very often roundish grains of bluish, opalescent quartz interrupt the e¢ontinuity of the rock. These are considered to be of exotic origin as they were seen to graduate in size into larger blocks of quartzite (xenoliths) shattered from the sill-contacts. To show the average composition of the granite, and the approximate limits of its lithological variation, fresh specimens, taken from sill C at three points in the section following the wagon-road, west of the mountain, will be described. They were collected at respectively 15, 40, and 50 feet from the upper contact with the quartzite. The specimen taken at a point 15 feet from the contact, and representing what may be called Phase 1, has the macroscopic appearance of a finely granular gray granite. In thin section it is seen to be a micropegmatite with a hypidio- morphic granular structure sporadically developed in many parts of the section. The crystallization is confused and does not show the regular sequence of true granites. The essential constituents are quartz, mieropegmatite, microperthite, orthoclase, oligoclase-andesine and biotite; the accessories include titaniferous magnetite, a little titanite, and minute acicular crystals of apatite and rarer zireons. The characters of all these minerals are those normally belonging to PLATE 24. Secondary granite of the Moyie sill C, fifty feet from upper contact. Natural size 25a——vol. ii— Pp. 228. REPORT OF THE CHIEF ASTRONOMER 229 SESSIONAL PAPER No. 25a common granites. The chemical analysis of the rock shows the mica to be magnesian. A striking feature of this, as of the other phases of the acid rock, is the advanced alteration of the feldspars which are usually filled with dust-like aggregates of epidote, kaolin and muscovite. This alteration is believed to be due to magmatic after-action, probably the result of the expulsion of vapours during the solidification of the underlying gabbro. The calculation of the quantitative mineralogical composition of the rock has been attempted by the Rosiwal method. In the process the secondary pro- ducts were neglected and the feldspars were arbitrarily regarded as fresh. The inaccuracy of the result is manifest but it does not affect the value of the com- parison among all the phases of the sill. Especially between the gabbro and the acid zone the contrasts of quality emerge with the same clearness and certainty as characterize the related contrasts established in the chemical analyses. The total chemical analysis by Prof. Dittrich of this Phase 1 (specimen No. 1137) is here given: Analysis of Granite (Phase 1) of Moyie Siils. Mol. Sid. ae 71-69 1-195 TiO. Rae +59 -005- Al.O, aa. 13-29 +130 Fe.0; Ae +83 005 FeO 5 86 A 4.23 058 MnO ov 2 09 001 MgO Ws 1.28 032 CaO 1-66 030 INGE Oya doesomer rs 2-48 -040 LE ORT CU Oe ot ee a Re el a ae ee eg eae wre 14 Shite H.O above 10°C. Papa Nena Weel Satan S als cia atcha etek. etece siewtals 1-31 P.O. . ee ee ef Of WOO 9 OOF 100 “O00 500 OOS OO BOCE HOC OO OO 07 CO CA DO8 DONC GO LCO MOC OO rr CO. C0" INDO. CORO MOOR DO RC 10x 13 100-16 Sytp fara coe oe a .. (corrected value). 2-733 This wack is eleaaleas an masa iene of biotite granite. The most evident peculiarity is the low total for the alkalies; it accords with the relatively small proportion of feldspar present. Notwithstanding the abundance of free quartz, the silica percentage is kept low by the comparative richness in biotite and by the magmatic alteration of the rock. The estimate of the mineralogical com- position gave the seas result in weight percentages :— — He Lt com Oi wooo WM og Quartz.. Sodiferous orthoclase. Bioti Bats Tela ereriare cyclo eie) reer hine els Meee ROBO OO OC aT Microperthite.. Oligoclase.. een ete Magnetite. . Seis Apatite.. isl | | 230 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 In the Norm classification the rock enters the-sodipotassic subrang, teha- mose, of the domalkalic rang, alsbachase, of the order, columbare, and the persalane class. The norm has been calculated as follows :— OTA TEA icteric let oretittna can peveieinte ope iivsie cl siatine ai ine oat orennecclar stoners mete smeiote 40-14 Orthoclase. (reer a oe ae eae eee teeta) INIBbGG Bch ed PeLees hts oy, GE eo ee mm OULOG Anorthite.. RecA Rods NVA aa oid ach, suet aie ealane tans star Rese? eign reve elie Ur erenmeverhtatetinets 7-23 COPUNGUMs 6: oe se Ria SR a alee a elas tine Mere er totes 3°98 Hn eens SERIA a oe iath Maes ial PLE SAC GA Ta SORE EIN eats he GRU e Racretpeate 9.27 Magnetite.. ORs oer Pay LHe MC | deri, Pua Seveitaratn Rarare alate ta Paiahta saute rents: eyrape 1-16 Ilmenite.. .. rire wether cate WEN Aen UR. cee Ay Sa UM ent ate Ml i tecae pet eee a aoe OPs 1-21 H.O and CO... fone’ lei’) ere. Nofel Miele iece | cates vere: |Neisi je elu ejewterem ole} Keleuro jen ieretpeveunoye 1-58 99-43 The second analyzed specimen of the biotite granite, Phase 2, is that collected at the point 40 feet from the upper contact of sill C. It is closely allied in composition to the phase just described and is chiefly distinguished from the latter by a coarser grain and a different structure. Microscopic examination shows this rock to be eugranitic (hypidiomorphic-granular), with small isolated areas of the micrographic intergrowth of quartz and feldspar. The constituents are nearly the same as in Phase 1. Here, however, muscovite is an accessory so rare as not to enter the table of quantitative mineral propor- tions. True soda-orthoclase replaces nearly all the microperthite of the micro- pegmatitic facies. Calcite enters the list of constituents; it may be in part of primary origin. : The chemical analysis by Prof. Dittrich of Phase 2 (specimen No. 1138) is as follows :— Analysis of Granite (Phase 2) of Moyie Sills. Mol SiO, 72-42 1-207 TiO, +68 009 Al,O, 10-47 103 CoV, 83 005 FeO 5-50 076 MnO +16 002 MgO 41 010 CaO 2-53 045 Na,O _ 1-93 | 031 Darn tshsri latches’ veiws veel" 21s Violet rave tate moveslemtate laleuee e isroueenelone Te 2-94 031 ZO Pate Ce od i aa eet on arn groin a tciemear sapere eee 06 H.O above 10°C... SH ROUEDOICO DO0t0b- Dd abe Ob nonans an ad 1-11 Serete POR a0 il SRerceeess beac al Faceted ciara tate NeVeinnetee falcata ce aerate roirers more 11 001 (OF 0 ee Se ie aR PAN eAUER A ree ereniiest TAC ege Meare ee TAR verteabe pase a 61 014 99-76 The corresponding mineral composition in weight percentages was roughly determined by optical means, thus :— REPORT OF THE CHIEF ASTRONOMER 231 SESSIONAL PAPER No. 25a UAT EA ore a ae, es eves oT oI cee Seah Oeratnne wp litaipM olabe uel ad lon veser “reve 46-0 Ribeiro Gass od secbeebo.u wo. bo cos 60 08 nod 165,.BO G0 .G doL00N00 29-1 TBO LAL in ee Ss ab. ag ON eee SeOr mG SCOT CON GONG c. SCD iO creo ocreca 22-0 Oligoclase.. . : eee RIN cue Cr dd aRbnt concrete | Remand Abie ante Aida 1-5 Magnetite.. .. Sh ROB 7) WAM ALILOME sys ere cee ook we 5 Calcite.. <2 -. “4 100-0 The high proportion of quartz, the very low percentages of the alkalies, yet lower than in Phase 1, and the low percentage of alumina indicate that we have here again, as in Phase 1, a quite abnormal! kind of granite. In the Norm classification this rock belongs to the sodipotassic subrang of the domalkalic rang of the order, hispanare, in the dosalane class, with the following norm :— UATE Zeiss eee ee aoe aaah Lats veh ae rE Ta ale iSre” wiel aie safe atately fos 42-30 OEE OCTASS Sars. eyoes oe Let ee pceeraos os satetec Ae SU acl ods She Moran Oic Ra eRtene erohtese ees 17-24 JI STUD Se ce OER CORD: LUGO nO OC COTO RET CAMS ORAO acer Gr See 16-24 PANO RGU Crepes eterno ere Nee cro oararell cloths ersiureten e svoiareleueroietlieretaeter revsh dolesicar era ere hviere 7°78 (Coe ataXs bi bane ry la ra eel eels aye avr mos amen ane Font Im Anas Betis rage te Li 1-30 IER ODES M45 Ob 60 Moo Go u0l.Gd Go ODN eO G6 b0MOD\oo 40 ‘dun Canon pp 9-45 Mlirme rite seseey cet dates ecclie eosa cive cavertiacaheceia alee ar sie eievaiea)y aNey qeiel eles nieles davatfiots 1:36 Mitctorne Gite cectarren cysmraaimennt cate erat tare nara ately, eratimaraim ate) aie feje svete Nolel teleinets 1-16 IATOER BLS GB Gete. ob HOO OROMID DIO hod Son) oo. ADE EOD Domi Os ODE tae od 31 Cal Cite eee ee NS a ea Sine Oe re sem Stoo epee len jofentelen ale 1-40 H,O Orde FORO OOD OO Oty. GbdsOO GOT OO OO COMICIO= 3:08 15:0) 5 O09 O18 ONCe ONO CROTON 1:17 99-71 Neither rang nor subrang has yet received a distinct name in the system. Phase 3, collected at the point 50 feet from the upper sill-contact, is unusually quartzose. It has nearly the same qualitative composition as Phase 2 but the structure is more like that of Phase 1, being essentially that of a rather coarse-grained micropegmatite. The optical method gave the following weight percentages for the different constituents :— ao a nmin go oS ores OI DIO co Owantzesge te ang ease n: Sodiferous orthoclase.. .. . ISSO Kayes oo AE eee MUSCOVALC Seach aveinisleitere water ees Calcite are ig: tala lolarreten-teree fede sais ist Gols caret toaye : IMianonetitomteianaawsre stew Gs katie wien Seu died see auc aee Oligoclase sii sisccx2 sis ness Us 3 | It is clear that there is notable variation in the composition of the biotite- granite zone as represented in Phases 1, 2, and 3. The apparently regular increase in acidity in the zone from above downwards is fortuitous. The zone is in reality irregularly streaked in many such phases, carrying variable pro- portions of the mineral and oxide constituents. Whatever the cause, the 232 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 magma was not homogeneous at the time of its solidification. To that fact is doubtless to be related the confused, rapid crystallization of the essential mineral constituents. ABNORMAL HORNBLENDE-BIOTITE GRANITE. The biotite granite of sill C graduates downward into a rock of similar habit, with hornblende added to the list of essential constituents (Phase 4). The amphibole resembles that of the unaltered gabbro of the Purcell sills. The structure of this hornblende-biotite granite changes rapidly and appar- ently irregularly from the micrographic to the hypidiomorphic-granular. The top zone of sill D is composed of the same rock type. No chemical analysis has been made of it. The specific gravity of a specimen from sill C is 2-765, being slightly greater than the average for the overlying biotite granite. INTERMEDIATE ROCK TYPE. Underlying the hornblende-biotite granite in both sill C and sill D, and underlying the biotite granite in sill A, are zones composed of a rock which combines features of granite and gabbro (Phase 5). It is, in fact, a rock directly transitional into the dominant gabbro of the Purcell sills. A specimen illustrating this intermediate rock was collected at a point 200 feet below the upper contact of sill C, and has been analyzed. Macroscopically this phase is much like the usual gabbro of the sills. It is a dark, greenish-gray, granular rock of basic habit. Its essential minerals are hornblende, biotite, and andesine; the accessories, quartz, orthoclase, titanite, titaniferous magnetite and apatite. The secondary minerals are zoisite, kaolin and epidote. The structure of the rock is in general the hypidiomorphic-granular, but local areas of micropegmatite are common in the section. The total analysis of this phase (specimen No. 1140) by Prof. Dittrich gave the following result :— Analysis of Intermediate Rock (Phase 5) of Moyie Sills. Mol. SiO... 52-63 877 TiO... V -008 Al,0,. , 16-76 165 Fe,0, 2-86 018 FeO. . 10-74 149 MnO ee 38 006 MgO 4-33 108 CaO e ° ee ° 6-17 110 Na,O 6b Go oe 1-41 -023 UGOWeMas uNebe daroalda dea: HGlLae sdelsOuddltocU nooo Oo 2-29 024 H,O at 110°C... @(e! wterey o1e) tee), viele) (ele) bi elehuiele) se) 01) ole: mele, nee) ye leuiece 12 coe HS Osabove MOCO so 2. ence mero mesiislon ory nc: si=1ri0) whence 1-17 “oer PiOrs ec ce ec8 ec cf ef ©8 e8 © #8 #©6 ©8 ©8 © © e8 288 Ge 33 002 CO...: ee 08 ee ee © ee ec ee ef e868 ©8 0e0 ©0 00 08 08 se se 10 ecoe Spaig tay: cos suits (aay lossy cout wees) «6 (Corrected value). 2-954 REPORT OF THE CHIEF ASTRONOMER 233 SESSIONAL PAPER No. 25a The quantitative mineral composition by weight percentages was deter- mined (orthoclase not separately estimated but included in the andesine) thus :-— Ja Kan ai) 6) Key 1X0 K veep ae ec RIT EN oe aa Pu) aa he ata rena ae aa 49.4; SLOG EGS chee ts arte ay © oteaih ree Wet ON cyt een Tepe er al gen edge URNS eo 22-0 PAIL ESTM G55.) 595.4 Sior cotse tei Same ee See eee Gwe BLE Mon nents 16:5: OU arte eee oe Sra i Ses cree inten tells, srs ear eae ct ey 2 Ae ean ta 11-7 PANO AUTCG eit ios: ak ares shay cae cree eNO aI ec Te a Tektite Bets 3 IM OTE ET Ces sea yen m a sinned Sane ree) cheep ccalige ds Charen Attar ve oa arene 1 100-0 The abundant biotite and quartz go far to explain the differences between: the chemical analysis here and that of the normal gabbro. It also appears from the analysis that the hornblende is here unusually aluminous. Chemically considered this intermediate rock has its nearest relatives among the diorites; yet the low feldspar content forbids our placing this rock variety in that family. Like both the gabbro and the granite it is an anomalous type. In the Norm classification the intermediate rock appears in the as yet unnamed sodipotassic subrang of bandase, the docalcic rang of the dosalane order, austrare, with the following norm :— OUI AGS es, Gor OBOE ROR Bn Mine Uonec ni HB ciRMOG ware HOLE Cine) ing mn ONICEN ei 9-72: Orthoclasercy ec ca asa cial eco aa ee Mauee ea ee Sle ees 13-34 JNITOTRE EE ce as einen cepa se BABS SRT cs on IN HEA Gc AUC aU RT ic ASH N 12-05 PATVOT ELEC str certlere crear crc ee Gee ee ei oie SURO IEEE Ss 28-63: COL eb rs Wott Gee Gas IUCr eRe PTET Tren ea ee rae 1-53 1a ROSEANNE 8) Ginetie sre (Oey Ae ER Is ROTI OSE EG e un Ges Genta 26°51 DUCE RES aa aps is se Sa ae Re ee era ey UR Re 1-22 IMT C IES eM eee er ey rol See aN iene eae ie etic aan eva tan area cage a umes 4-18 ANID ETE G em ease reste crcmpees aire ne eT cto ace Cc Mlb egret FNC OR as 62 H.O and CO,.. Ss ee Ca eee ania! 1-39 99-19 At the perpendicular distance of 200 feet from the lower contact of sill E,. another specimen of the intermediate rock was collected. It gave the following weight percentages (mode) :— Hornblende.. .. . eae 42-9 Quartz ore ce ae 22-8 Andesine.. .. . 18-5 JB AVGRG UY iti berries ee IEE PPE ae UE Coe Nai meaner iE MEeS SEAM Anon eg Cea 6-6 Sodiferous orthoclases: sun se ce tee oe) oe Pew tie eateries 5-5 Titanite.. oe 08 02 » -@ ©8 ee @e © 280 ©0 ©0 08 08 of se 3°7 100-0 | ABNORMAL HORNBLENDE GABBRO. The whole of sill B, and the lower part of each of sills OC, D, and E are all constituted of dark, heavy gabbro (Phase 6), which is either sensibly like the usual gabbro of the thinner Purcell sills, or differs from it in unessential 234 DEPARTMENT OF THE INTERIOR 2 GEORGE V.,, A. 1912 details. The foregoing description of the usual gabbro will suffice, also, for most of the femic rock in these Moyie sills. Yet microscopic and chemical study of the lower internal zone of contact of sill E, shows that here the rock is not quite the same as the usual gabbro. This Phase 7 was collected at a point 80 feet perpendicularly from the lower surface of contact. In macroscopic appearance and internal structure it is not markedly different from the usual gabbro. The essential minerals are horn- blende and labradorite; the accessories, quartz, potash feldspar, titanite, mag- netite. Zoisite, kaolin, and much chlorite are the secondary constituents. Chemical analysis of Phase 7 (specimen No. 1148) by Prof. Dittrich gave the following result :— Analysis of Gabbro (Phase 7) of Moye Sills. Mol Sid... 52-94 882 TiO,. 73 009 Al,O. 14.29 +139 Fe,O,.. 2-08 018 eO.. 8-11 113 Mn0O.. 39 005 MgO.. 6-99 175 CaO 10-92 °195 Na,O 1-40 023 KROME oraneaa eo bet 49 005 Oval ATOl sc yn. ea sits ee ae Tae prey coe mm a 12 Aaa HeOm above: 110°C oh sas eas tee oh en See tae nee ae 1-56 ate PAG) RMI Sate 88 fa aris Sean pla ate eves ey | EARL ee dia nae eee 08 001 99-99 NS] NGS rien HA en PRB ME ce RAS NUSa iin cart kos masta CaCO haa 2-980 The corresponding mineral composition in weight percentages is roughly as follows: Hornblende.. . iatiits 54-8 Wabradorites. 2. «es. ve ke 25-6 Chiorite.. ca... ; 11-0 @ulartzues.. 6:3 Tit amibess score tees cers) wor eale eye Mase 2:0 Maen tite yencict wae otis. x ee eisai 3 100-0 | On account of some alteration in the rock, it was found difficult to dis tinguish with certainty the small amount of alkaline feldspar; which has, accordingly, been entered in the total for labradorite. GZ LLY TD “Cy T[IS Jo gowquoo szaddn utoay qooy AQyy oj1uess °C) [[iS Jo gouquos zoddn urcayz qooy U9aqyy oqlueid olqqes Jo asveyd OLutey : 4JoT Jamory OAAGVS VSVIDAV : ¥JoT odd ‘AZIS [VANYBVU J[VY-9UO suetpooeds : ) hus : 44811 LaMoryT > 4ysit tadd yy) atAOTW OY} JO sasvy p. 234. il = vol. 25a REPORT OF THE CHIEF ASTRONOMER 235 SESSIONAL PAPER No. 25a The caleulated norm is: QTR asso N wae SAB Reins emOR Te ry ck Ania eG” BMIEMNN ce Weaaate Urs tet 8-40 Mrthoclasey ve ese eee ee ee Me adel ee hes eh 2-78 Albite.. eco ee ©2 ¢©2 © 809 ©8 e8 ©2 ee o8 le 0©e0 @©0 089 80 ©0 68 #8 e8 12-05 PATO TN TGS so sisih coq cee Cor ase ee eeTor ett Si STE. ee OUSIDE ris apeclia naa Tete ee 30-86 Eley MeETStHOMot es patente estes ea ee ea saree eerie are a seal Shek Nee 21-18 WIODSIer es sce: sae eee Ge ee min, es Loe en sees cae BAD Malai e bite Saaiein succinct nat ce nunerletiwmy! Siamese ec EUR Rata 3°02 MUTT T ON is aie sok Sn is Sette pee ead popu etn anen org aaah mr gih muvee 1:36 ANTTPPR TLE, Ue epee hg OO ee We ae ci hace ers I ne pean Ea rie err CL 31 \WEMESS so d0-G0 66 oo 0G ob 0066 40 OO Ho dal ee Gob. 0B a0 on 6s 1-68 100-04 This rock belongs to the presodic subrang of the as yet unnamed docalcic rang of the salfemane order, vaalare. RESUME OF PETROGRAPHY. As a convenient summary, the mineralogical and chemical analyses of the different phases have been assembled in Tables XI and XII. TaBLeE XI.— Weight percentages of minerals as determined by the Rosiwal method. ! oy a a \ong PR» py ep. coir oe | pons |ee 8 Sy od ee Sees ee es ae as Raa gat fe 5 Bo.8 8.8 |on A ree Bl ea | 2 leis lees ieee | B20 | sac | £20 Oa os? Se Se a sie) Se | Seer se Se aa | 5S | Eee] Fes Seen) 292 | Soa | fou BS | 25 | #25 | $28 |5&9'a| 88 | 62s | se =) Dime 4 4 a0) ia aa al Hornblende. . .. 58°7 54°8 49°4 42°9 a REL al ee tee haem allies Se aa ee BiOtiterc sets 6 fea leek OT Ee reac: 22°0 6°6 17°3 89 22-0 15°2 abradorite; 4..—.. >. gall oes dae oy Wal a Bikes Sy eiecetwet Seer Lacey eens | abana PAE ah Sareea ane PAI ESIN Ching aiaie stccs ois se eho eee eee | 16°5 Steal cope rae patellisg sta ees ol nce arse crue oe eanarts Olicoclasewe rr sate ele n sews MS a ll adap alevade ei leeserceel| ca tavall Besreteg sea Salat) bo) 1°0 Soda-bearinerorthoclases sna alee o es eciitinoe oe ee 55 27°8 24°9 29°1 BPI) Microperthiterns haa seem lencecr manne elec antonlee acer ei dar tial eget oly aaes aks 3°9 HALAS ee ee ae +0 Gea} |). blew 22°8 37°2 57°1 46°0 41°6 VIMISCOVAGEseipie iN ay camer [Mir rac. poate [Pepegcseeeseesoetl | ee oa rentoent unease Sade Pan eee 4°6 Apart bess oo: veal SOAS ene asa | i lesen ces | Mae pers cae Ro EG 5 z MICaMIGes AA Hee cs astosc tert 1°4 PARDO RE areca ste 3°7 SIO ai eta eote, oy waayeel| are Sven ons Nese cone Magnetite or ilmenite.....|........ 3 SET alec epee es 1122) 19) a) 1°0 Chioriten rs ee a elem nen TO ie sere Al enti aen NR iesa ra ceili das bebe Ica aac ted 2 | Rent as entra Calciteree rs. sone lbecae wales va: ep eesr oer ierotaierets cll Nenana 2°5 el eae Frc The total is 100°0 in each case. Specific gravity........... | 2°990 | 2-980 | 2°954 | 2°942 | Serie eae | 2°754 | 2°728 «=| 2°733 236 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 TaBLE XIT.—Chemical analyses of Phases of the Moyie Sills. ! ¢ VS) By pr Pv ac) ie) os 12 & So Soha rai ape me 2 | 22 ‘38 £8 23 Srieme Zt eae eapoue ace aoe SB S eee ae = 50 SO On | | oS 8 SOR as SO) eye aaa os co zs o BR o Bre 3a lee wte ys aes sa 2) #Q2 i ed ee B25 6.2.8 O38 =) pe 5 = = | S(O isla canoc nou su aa ane nono nae: 51°92 | 52°94 52°63 72°42 71°69 FUL OVS ena tedm ny cye cetera peaews wanesetanse 83 | ‘73 62 8 59 AME eevee crass ee, ciatesaree eerie 14°13 14°22 16°76 10°47 13°29 MCR O srr iacamctrapiend “sine cca. ou 2°08 2°86 83 .83 COE erate aye aeiiens nal crotnee oreo am 6°92 811 10°74 5°50 4°23 Min OC va sesnae a slants otiates otets 14 35 38 16 09 IMA 0) SoBe etree cern Sayer ryan aise 8°22 6°99 4°33 41 1°28 CaO Wee te eon Or eee hat Mee ae 11 53 10°92 6°17 2°53 1°66 IN ais Ortaca aihod aloes Mash Sat ee 1°38 1°40 1°41 1°93 2°48 Oe tet heearcaio atiasae eiaiioe 47 “49 2°29 2°94 2°37 ENS ORatwllOl Cee ease tere ‘10 "12 12 06 14 HG OFabuve lo C: ese sack. neni 1°07 | 1°56 nese teal 1°31 TET Oe 4 el Ra ae a i | 04 cae ‘08 33 11 07 (OLE ais aaah rich sane aR peter Ae 06 ltomcacacoroe 10 61 13 99°78 | 99399 99°91 99°76 100°16 The two tables illustrate the abnormal character of every one of the rock types occurring in these sills. The tables also show the great range of rock variation. The changes in mineralogical and chemical composition and in density are clearly systematic in the series from gabbro, through intermediate rock, to hornblende-biotite granite, and then to biotite granite. It now remains to indicate that the same serial arrangement characterizes the rock-zones in each of four of the sills; and that there is an analogous series in passing’ upward from sill E, through sill D to the top of sill C. ESSENTIAL FEATURES OF THE DIFFERENT SILLS. The reader will readily seize the situation by a glance over the following stratigraphic column (Table XIII) and the corresponding diagram (Figure 15). At the top of the column is the recently discovered sill A with its cap of quartzite; at the bottom is the quartzite underlying sill E at the valley floor west of Moyie mountain. It should be noted that thicknesses of sills and zones, and the positions of type specimens have been determined only with approxi- . mate accuracy. The section described occurs almost exactly in the line of the Boundary slash. REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a TaBLE XIII.—Showing Columnar Section through the Moyie Sills. ‘Sills, thick- nesses in feet. E, 200+ Rock zones, thicknesses in feet. 2 950 + |Gabbro. | 750— 50+ 150+ | eA tae aos _|Quartzite of great thickness Usual gabbro of Purcell sills Character of rock. Quartzite of great thickness. . Acidified gabbro, 2 specimens, sp. gr. 2°89 and 2°97...... Biotite granite Slightly acidified gabbro Quartzite Biotite granite, 5 specimens, sp. gr. 2°72—2°754............ Hornblende-biotite granite, 3 specimens, sp. gr. 2°74—2°84.. Intermediate rock, 2 specimens, sp. gr. 2°95 and 3 00 .... Gabbrometpocncns siete toes Netratey olen scterepaySicveutesinis Del seleMenelieletoluieife) oivis™ ioqeletelclove rel ejeiel>iejeiel cpist, Icke Eornblendesbiotiteranitess. «- ems nce ee cece nine ee imtermediate rocks wvwsaene esc ceimeie soy ASIN epee says Quartzite COO OUUDOOOOOUOU MUDD ODDO OODIDODOUOONUOOO GO WO in Wc Imbermediaterrocketyascerseens ete eh ar erty ee Cee tt ie te Gabbroi(somewhataweathered) 4.05.) cles ci eiac Sieh elefs,ie/eeleiie) Jeielels ie lee) ej.) elieivtelele)aheelw 237 Average specific gravities of igneous rocks. ee ees Sill A is well exposed on the eastern slope of the mountain at the contour 300 feet lower than Monument No. 214. The overlying quartzite dips 54° in a northeasterly direction. gabbro. The uppermost 25-foot zone of this sill is acidified That rock shows rapid transition into an underlying, 80-foot zone of biotite granite, which similarly graduates with some rapidity into a nearly quartz-free gabbro approximating the usual rock of the thinner Purcell sills. This is the only one of the sills which has been seen to have a gabbroid zone overlying a granitic one. The upper gabbroid zone seems to represent a layer of magma which was rapidly chilled against the cool roof of quartzite. The lower part of the mass had a longer period of fluidity and became stratified through gravitative differentiation. Between sills A and B is a 100-foot layer of quartzite, with strike N. 25° W. and dip 65° E.N.E. Sill B, 30 feet thick, is a fine-grained gabbro of the usual type in the Purcell sills; it is apparently of quite homogeneous.composition throughout. 238 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Below sill B is a 550-foot band of quartzite with strike N. 25° W. and dip 65° E.N.E. Sill C, about 580 feet thick, is well exposed in the Boundary-line section as well as in that on the wagon road north-northwest of the summit of the mountain, where three of the analyzed specimens were collected. This body is the most striking of all in its evidence of gravitative differentiation. The 80- foot zone of biotite granite at the top passes gradually into the underlying 110-foot zone of hornblende-biotite granite, which, in turn, merges into the 60-foot zone of intermediate rock overlying the 280-foot zone of the usual Purcell gabbro at the bottom of the sill. Between sills C and D comes a band of quartzite with strike N. 20° W., dip 60° E.N.E. Sill D is poorly exposed but seems to be largely composed of the usual gabbro, overlain by successive zones of intermediate rock and _ hornblende- biotite granite. The outcrops do not suffice to show the exact thickness of any of these zones, but it seemed clear in the field that the total thickness of the two more acid zones was little more than 100 feet. Between sills D and E is a band of quartzite, estimated as about 750 feet in thickness; its strike is N. 80° W. and dip 60° E.N.E. Sill EH. The lower part of sill E is well exposed a few hundred yards south of the Boundary slash, but its contact with the overlying quartzite was nowhere discovered. As already noted, the existence of that layer of quartzite was not even suspected in 1904, as it was entirely covered by talus along the line of traverse then followed by the writer. It then seemed most probable that the gabbro masses exposed at the top and bottom of the great talus slope formed parts of a single sill. For 100 feet or more from its lower contact the rock of sill E is practically the usual gabbro of the Purcell sills. That zone is overlain by a zone of intermediate rock, the top of which has not been discovered. The two zones show a gradual transition into each other. ORIGIN OF THE ACID PHASES. Preferred Hxplanation—Among all the conceived hypotheses as to the origin of the acid zones, the writer has been forced to retain one as the best qualified to elucidate the facts concerning the Moyie sills. More important still, this hypothesis, better than any of the others, affords a coherent, fruitful, and, it seems, satisfactory explanation of similar occurrences in other parts of the world. It will be presented in some detail, since it is believed that these sills, and similar ones in Minnesota and Ontario represent gigantic natural experiments bearing on the genetic problem of granites and allied rocks in general. The view adopted includes what has been ealled ‘ the assimilation- differentiation theory.’ The acid zone is thereby conceived as due to the diges- tion and assimilation of the acid sediments, together with the segregation of most of the assimilated material along the upper contact. REPORT OF THE CHIEF ASTRONOMER 239 SESSIONAL PAPER No. 25a Flat Position of Quartzite at Epoch of Intrusion.—Since the granophyre- granite zone of sill C is known to have a tolerably constant thickness through- out an exposure of at least three miles along the outcrop, the hypothesis involves the assumption that that sill and the adjacent ones lay much more nearly horizontal at the time of intrusion than they do now. This assumption is favoured by all the pertinent facts determined during field work, though it cannot be claimed that they furnish absolute proof. In the first place, it is probable that the majority of the faulting and upturning suffered by the Purcell sedimentary series was brought about at one orogenic period. The intrusive sills are themselves profoundly faulted and their outcrops are repeated by faulting in such a way as to indicate throws of thousands of feet. If this extensive disturbance of the sills had followed their intrusion, which itself followed earlier important dislocations of the intruded sediments, we might reasonably expect that the detailed structures of the twice- faulted sediments would show some evidence of the history. As a matter of fact, the cleavage often fully developed in the quartzites apparently belongs to one orogenic period and to one only. It was developed after the intrusion of the sills, for the gabbro itself is occasionally cleaved with its planes of cleavage parallel to those in the adjacent quartzites. The sediments must, of course, have been slightly disturbed as the intrusive bodies were injected, but true mountain-building seems to have been postponed until long after the solidifica- tion of the magmas. The repetition of sill outcrops by faulting is most easily anderstood if it be believed that the dips have been greatly increased by the relatively late disturbance. If the strata had been well faulted, tilted, and cleaved before the intrusions took place, the injected bodies shou!d. show much greater irregularity of form than they now actually show: most of them would be in the relation of dikes or chonoliths (injected bodies of irregular form). following faults and other secondary planes of weakness, rather than in the relation of sills following bedding-planes. A second argument is to be derived from the fact that sills and dikes of hornblendic gabbro, mineralogically and chemically very similar to these sills of the Yahk and Moyie ranges, cut the Kitchener formation and equivalent Siyeh formation of the McGillivray, Clarke, and Lewis ranges at horizons immediately below the Purcell Lava, and, almost without question, represent feeders or offshoots of the magma represented in the widespread lava flood. That these eastern dikes and sills do thus represent the contemporaneous intru- sive facies of the lava is suggested, as above remarked, not only by the litholo- gical consanguinity, but also by the fact that none of the formations overlying the Purcell Lava horizon, 1.e., the Sheppard, Kintla, Gateway, Phillips, Roos- ville, or Moyie, is known to have been cut by dikes or sills which are younger than the older beds of the Kintla formation. Granting, further, the contem- poraneity of all these Purcell mountain-system sills with the (Middle Cam- brian?) Purcell Lava, which is rigidly conformable to the geosynclinal sedi- ments, it follows that the intrusions took place when the strata lay flat and, in the eastern ranges, were covered at the end of Kitchener (Siyeh) time, by the great flows of the extrusive, post-Siyeh lava. 240 ~ DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Evidently neither of these two arguments is quite conclusive, but the ‘balance of probabilities is certainly on the side of the belief that the strata cut ‘by the Purcell sills lay nearly horizontal as the thick bodies were injected. In -view of the perfect conformity of the Moyie and Kitchener formations (both laid down in shallow water) it appears probable that the surface of the Kitch- -ener formation was not elevated through the full 2,000 feet represented in the total thickness of the Moyie sills; it seems more likely that the beds underlying the sill-horizon were down-warped nearly or quite 2,000 feet, so as to make room for the sill magma. Superfuston of Sill Magma.—The hypothesis carries the second assumption that the gabbroid magma was, at the time of intrusion, hot enough and fluid enough to permit of the solution of a considerable body of quartzite and the diffusion of the dissolved material to the upper contact. The assumption is supported by the discovery of the great horizontal extent and uniform thick- ness of the intrusive bodies; if the magmatic viscosity had been high, each ‘body would have probably assumed .the true cushion shape of the typical laccolith. The extreme fluidity of the Purcell Lava is proved by the great distances to which its flows ran before solidifying. If the Moyie and other sills -were but the contemporaneous intrusive facies of the same lava, the intrusive magma must have been highly fluid. Its temperature was at least slightly higher than that of the extrusive and therefore somewhat chilled lava; and, -secondly, the pressure of the few thousand feet of overlying Kitchener beds could raise the solidifying point only to an insignificant extent (probably less than 5° ©). From a study of the grain in the Moyie sills, Lane has calculated that the magma, when injected, must have been considerably superheated, and therefore quite fluid.* Finally, whatever theory of the acid zones be adopted—whether that of pure differentiation, of assimilation, or of both—the fact is clear, from the foregoing lithological description, that the diffusion of silicious material through the gabbro actually occurred on a large scale and that this diffusion eould not have taken place unless the original magma were possessed of a high degree of fluidity. Chemical Comparison of Granite and Intruded Sediment.—A third, even more clearly indispensable condition of the- hypothesis relates to the composi- tion of the invaded sediments. One of the most noteworthy features of the huge series of conformable strata in the Creston-Kitchener series in this particular district is the marvellous homogeneity of the whole group. As already indicated, even the division into the two great subgroups, Creston and Kitchener, is founded on merely subordinate details of composition. Hence it is that the study of comparatively few type specimens can give a very tolerable idea of the average constitution of the quartzites. For convenience a brief description of both Creston and Kitchener specimens analyzed will be here repeated. Single beds typical of the Creston occur interleaved in the * A.C. Lana, Jour. Canadian Mining Institute, Vol. 9, 1906. ° REPORT OF THE CHIEF ASTRONOMER 241 SESSIONAL PAPER No. 25a Kitchener and occasionally rusty beds are intercalated in the Creston series. In both series the average rock is a quartzite, always micaceous and often ‘decidedly feldspathic. Many of the strata above and below the Moyie sills have a composition essentially identical with that of typical Creston quartzite. Hence the chemical analysis of this latter rock partly shows the constitution of the sedimentary group invaded by the gabbro. From Mr. Schofield’s descrip- tion, the underlying Aldridge quartzite seems to be like the Kitchener. Professor Dittrich has analyzed such a type specimen collected several miles to the westward of the Moyie river. It is very hard, light gray, fine-grained to compact, and breaks with a subconchoidal fracture and sonor- us ring under the hammer. The hand-specimen shows glints of light reflected from the cleavage-faces of minute feldspars scattered through the dominant quartz. A faint greenish hue is given to the rock by the disseminated mica. This rock occurs in great thick-platy outcrops, the individual beds running from a metre to three metres or more in thickness. Occasionally a notable increase in dark mica and iron ore is seen in thin, darker-coloured intercala- tions of silicious metargillite. In thin section this characteristic Creston quartzite is found to be chiefly composed of quartz, feldspar, and mica, all interlocking in the manner usual with such old sandstones. The clastic form of the mineral grains has been largely lost through static metamorphism. The feldspars are orthoclase, micro- cline, microperthite, oligoclase, and probably albite. The mica is biotite and muscovite (possibly paragonite), the latter either well developed in plates or occurring with shreddy, sericitic habit. The biotite is the more abundant of the two micas. Subordinate constituents are titanite in anhedra, with less abundant titaniferous magnetite and a few grains of epidote and zoisite. The chemical analysis (Table XIV., Col. 1) shows a notably high propor- tion of alkalies, and therewith the importance of the feldspathic constituents, especially of the albite molecule, which alone holds about 15 per cent of the silica in combination. TaBLE XIV.—Analyses of Sill Granite and Invaded Sediments. 1. 2. 3. 4. 5. SiO... 82-10 76-90 74.23 79-50 72-05 Oe 40 °35 : 38 Al1,0, 8-86 11-25 13-23 10-13 11-88 Fe,O, 49 -69 84 “59 83 Fe : 1-38 3:04 2-65 2-21 4-87 MnO 03 -02 07 02 12 MgO 56 1-01 1-02 78 85 CaO.. 82 °88 1-13 85 2-10 SroO.. Bron LARS Nias Ozer oer 2-51 3-28 2-78 2-89 2-20 KE OR aria cert tia eio ree me 2-41 1-36 2-66 1-89 2-66 Jat0) eye BUA Oe ooo) on ico 05 20 08 12 10 H,O above 110°C... ... .. 37 1-20 81 °78 1-21 CORES Pana ee cael venues de tr. 08 ons 37 PE Oe Ape mran laiteant wee ete Tes 04 15 scala 09 09 100-02 100-33 100-16 100-23 99-96 Sp. gr. (corrected) .. .. 2-681 2-680 2-722 2-680 2-730 25a—vol. 1i—16 242 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 1. Type specimen of Creston quartzite. Analyst: Prof. Dittrich. 2. Type specimen of Kitchener quartzite. Analyst: Mr. Connor. . Specimen of Kitchener quartzite from contact ‘zone, Moyie sill C. Analyst: Prof. Dittrich. 4, Average of analyses 1 and 2. 5. Average of two analyses of biotite granite in the Moyie sills. > ft Mr. Connor has analyzed a specimen collected as a type of the Kitchener quartzite itself. It was taken from a point about 400 feet measured perpen- dicularly from the upper contact of the Moyie sill C, and this specimen repre- sents what appears to be the average quartzite above, below, and between the sills. It is .rather thin-bedded, the thin individual strata being grouped in strong, thick plates sometimes rivalling in massiveness the beds of the Creston quartzite. The thin section discloses a fine-grained interlocking aggregate of quartz grains cemented with abundant grains of feldspar and mica. The feldspar is so far altered to kaolin and other secondary products that it is most difficult of accurate determination. Only one or two small grains exhibit polysynthetie twinning and the preliminary study referred practically all the feldspar to the potash group. Mr. Connor’s analysis shows conclusively, however, that soda feldspar is really dominant. The analysis was most carefully performed, the second complete determination of the alkalies agreeing very closely with the first. Supplementary optical study of the rock hag pointed to the probability that pure albite, as well as highly sodiferous orthoclase, is present. Quartz makes up, by weight, 50 to 60 per cent of the rock, and feldspar from 25 to 40 per cent. Biotite both fresh and chloritized is the chief mica; sericite is here quite rare. Colourless epidote is the principal accessory; titanite, magne- tite, apatite, a few zircons and pyrite crystals are the remaining constituents. The analysis is given in Table XIV, Col. 2. Column 4 of the same table shows the average of Cols. 1 and 2 and may be taken as nearly representing the average chemical composition of the quartzite invaded by the Moyie sills. This average is to be compared with that of the two analyses of the biotite granite of the sills, represented in Col. 5. The general similarity of the two averages is manifest. There is clear chemical proof that the greater propor- tion of the elements in the granite could have been derived directly by fusion of the quartzite. The conviction as to such a secondary origin for the granite has been enforced by an examination of the exomorphic contact-zone at the upper limit of sill C. For the perpendicular distance of at least 60 feet from the upper surface of contact, the quartzite has been intensely metamorphosed. The rock is here vitreous, lightened in colour-tint, and exceedingly hard. Under the microscope the clastic structure is seen to have totally disappeared. Recrys- tallization is the rule. It takes the form of poikilitic or micrographie inter- growth of quartz with various feldspars, along with the development of abund- ant well crystallized biotite and (less) muscovite. The feldspar is chiefly micro- perthite and orthoclase, the latter often, perhaps always. sodiferous. Albite in REPORT OF THE CHIEF ASTRONOMER 243 SESSIONAL PAPER No. 25a independent, twinned grains of small size seems certainly determined by vari- ous optical tests. Innumerable, minute grains of zoisite and epidote occur as dust clouding the feldspars, micropegmatitic intergrowths, and even the quartz. Seattered anhedra of magnetite and small crystals of anatase and apatite are rather rare constituents. The chemical analysis of this highly metamorphosed quartzite is entered in Col. 3, Table XIV. In the preliminary study of the sill it was considered as probable that the quartzite had been somewhat feldspathized during the metamorphism, but the critical analyses seem hardly to bear out any certain conclusion on that point. The analysis shows that in several respects the metamorphosed rock is intermediate in composition between the granite of the sill and the unaltered quartzite. However, there is a perfectly sharp line of contact between the granite and this metamorphosed zone of the quartzite. The former rock has been in complete fusion; the latter rock still preserves its bedded structure. The net result of the foregoing mineralogical and chemical comparisons affords good grounds for believing that the striking similarity of granite and quartzite is really due to a kind of consanguinity; that the igneous rock is due to the fusion of the sediment. Comparison with Other Sills in the Purcell Range.——The assimilation theory assumes sufficient heat to perform the work of fusion. It is, hence, an indication of great value that there is some acidification of the respective upper-contact zones in all of eight different sill-outcrops optically studied in the 60-mile stretch from Porthill to Gateway; yet that this acidification is, in general, in a direct pro- portion to the thickness of the sills. The closely associated Moyie sills A, B, C, and D together have about three times the thickness of any other of the intrusive bodies. Presumably, therefore, the total store of heat in the Moyie group was a local maximum and the capacity for energetic contact-action was there much the largest. As a matter of fact, the Moyie sills are the only sills bearing the truly granitic phase. The other sills are also somewhat more acid at their upper contacts than at their respective lower contacts, but the rock throughout is of gabbroid habit. The acidification in these cases has, as we have seen, led to the development of abundant interstitial and poikilitic quartz, abundant biotite, and less abundant alkaline feldspar in the hornblende-plagio- clase rock. The rock of the acidified zones is here very similar to, if not identical with, the intermediate rock of the Moyie sills. The acidification is relatively slight because these sills have been more rapidly chilled than the huge Moyie complex. This point is based on deductive reasoning but it is no less positively in favour of the assimilation theory than the testimony of chemical comparison between the acid zone and the sediments. Evidence of Xenoliths—There is, finally, direct field evidence that the eabbro has actually digested some of the quartzite. Along both the lower and upper contacts and, less often, within the main mass of the sill, fragments of the quartzite are to be found. These rocks have, as a rule, sharp contacts with 25a—vol. 11—164 244 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 the gabbro, but, none the less, they have the appearance of having suffered loss of volume through the solvent action of the magma. In some of the other sills the blocks are yet more numerous and many of them are surrounded by shells of mixed material such as would result from the. solution of the quartzite in the basic magma. Since the blocks were suspended in a magma of different density and since the product of solution was not diffused away, the viscosity of the magma must have been high. Under these special circumstances, it is not surprising that a limited amount of solution was possible, even though the viscosity of the pure gabbro was relatively high. On the one hand, the very strong contrast in the ionic composition of solvent and: substance dissolved, implies a specially great lowering of the melting point. On the other hand, the original water of the sedimentary rock would facilitate solution even at the comparatively low temperature of 1000° C. or less, at which the nearly anhydrous gabbro became toughly viscous in cooling. Hybrid Rock.—A special instance was studied, optically and chemically, in connection with material collected in one of the sills at a point on the main Commission trail which is six miles up stream from the Boundary slash on the west fork of the Yahk river and two miles north of the Boundary line. The sill rock there forms low knobs on each side of the trail. Scores of gray, angular quartzite blocks, surrounded by the gabbro, can be seen on the glaciated ledges. One of these, measuring perhaps 100 cubic feet in volume, is enclosed in a shell a foot or two thick, composed of the solutional mixture. The quartzite has been completely recrystallized, with the development of large, poikilitic quartzes, as in the case of the quartzite metamorphosed on the main contact of the Moyie sill C (analysis in Col. 3, Table XIV.). Abundant, minute granules of epidote were also developed. Recrystallized orthoclase (probably sodiferous) and a little oligoclase are accessory constituents; no biotite could be found. The original sediment must have been composed of nearly pure quartz and ‘seems to have been far less feldspathic than the strata cut by the Moyie sills. A certain amount of osmotic action has taken place, for the quartzite is shot through with narrow, greenish-black prisms of hornblende, 10 to 20 mm. in length. This exotic hornblende has the optical properties of that in the normal gabbro. It is specially abundant near the surface of the block which is, however, sharply marked off from the shell of mixed material. Titanite and apatite in notable amounts have also been introduced into the body of the ‘inclusion. The shell of mixed material consists of a coarse aggregate of deep green hornblende in prisms 10 to 40 mm. long, and poikilitic quartz, which encloses much granular epidote, titanite, apatite, a little ilmenite, and abundant, minute prisms of the amphibole. No feldspar whatever is apparent in thin section. An analysis of this mixed material (specimen No. 1164) gave Professor Dittrich the following result :— REPORT OF THE OHIEF ASTRONOMER 245 SESSIONAL PAPER No. 25a Analysis of hybrid rock in gabbro sill. SiO) oo oo LOG SDOR GOL OECD NMG EMA Beanie tcouCol ba: fdhc.c 54-02 TiO... ec ee ec 0©28 +e 60 88 coe ce ee ee 1-95 INH O hoe bl 66. Ga ado, oBle pic : ourste 12-08 Fe,0,.. ° e ° e ee ee o e ee 6°85: hae e ee e eo e ae 5-61 nO.. : re ers -09 MgO.. e e e eo 08 ce ° ee eo e e ee 2-82 sees ° ee ° ° e ° eo ee ef eo ° ° « 14-63 TU... . ee ° eo eo ec ee ° © ee e6 oe tr Na,O.. ec oe e e e e Z 60 K,O.. ec ec ce ee ef @©8 280 #62 ©8 2 08 @8 60 ©8 08 82 @o #8 6e 06 of 14 MEO arta UO LG mrsererscrecers orca cranes tarcearebicist, core vate; colePictetlh acl 2ie.e: | evols tre) jeyewhets 06 HOtabovecllOs@ sie cceac oc ce isisyceleelsused te ty els, sien mee isis, eed sia een ore 62 PFOre ec 88 @¢ ef 0©8 © 00 ef ef ©0 ©8 ©8 @©0 ©e © 00 ©8 08 © ce ef 21 Co... ec ec ee ef #@8 e0 e@@ 28 © @8 © ee © 82 ©0 ©0 80 00 oF 08 we 19 99-87 SDM BR aera ated Sub ctl sihitalbele eed Sum cena memnacnetacoey iets GBLAE The alkalies appear to belong, wholly or in largest part, to the hornblende: the alumina, ferric iron, and lime to the epidote and hornblende. The epidote has all the appearance of a primary mineral. In any case it has not been derived through ordinary weathering, for the rock is strikingly fresh. The composition of the shell is evidently anomalous and represents a double effect. On the one side, the abundant quartz and probably part of the alkaline constituent -in the hornblende represent material dissolved from the block; on the other side, the special abundance of the amphibole, to the appar- ently entire exclusion of soda-lime feldspar, shows that the block formed a centre around which the amphibole, as one of the earliest minerals in the magma to erystallize, segregated. As the amphibole substance was osmotically trans- ferred into the quartzite block, so the quartz substance was diffused outward into the magma. The shell has clearly not the composition expected through the mere solution of the quartzite; the actual composition has also been con- trolled by the concentration of the basic hornblendic material around a foreign body. The latter may have acted after the manner of the crystal introduced by the chemist into a saturated solution so as to produce crystallization through ‘ inoculation.’ In other words, magmatic assimilation and differentiation are both illus- trated in the history of this shell of mixed material about the quartzite block. It is, nevertheless, certain that the sill magma as a whole was acidified by the solution of this block and still more by the solution of others now invisible because completely dissolved. ? The phenomenon of the partial digestion of xenoliths is quite familiar at intrusive contacts; its significance is only properly appreciated if one remem- bers that the visible effects of digestion have but a small ratio to the total solvent effects wrought by the magma in its earlier, more energetic, because hotter, condition. It is not a violent assumption to consider that many quartzite blocks have thus been completely digested in the original gabbro-: BAG “~~ "" DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 magma. The product of this digestion is not now evenly disseminated through the crystallized gabbro, which, except near its upper and lower contacts, is very nearly identical in composition with the unacidified gabbro occurring elsewhere “in the district. No conclusion seems more probable than that the material of the dissolved blocks is now for the most part resident in the acid zone at the upper contact. The same view holds for the perhaps much more voluminous material dissolved by the magma at the main contacts themselves. The excess of acid material at the lower contact was held there because of the viscosity of the magma in its final, cooling stage. For the greater bulk of the digested material there. has been, it appears, a vertical transfer upwards, a continuous cleansing of the foreign material from the basic magma. Assimilation at Deeper Levels—Another cause of acidification is to be sought in the conditions of sill-injection. In the Purcell range, as generally throughout the world, channels (dike-fissures) through which the magmas have been forced into the greater sill chambers, are relatively narrow as compared with the thickness of the respective sills. In most cases the feeding fissures seem also to be few in number for each sill. The magma must pass through such a fissure during a considerable period of time in order to form the enormous bulk of a first-class sill. At! that stage the magma is at its hottest and it is being moved rapidly past the country rock. The effect is analogous to that of stirring a mixture of salt and water; solution is stimulated by the movement. The original magma is thus converted into’a syntectic magma, with greater or less chemical contrast to the original. Such a case may be represented in the great sill. in New: Jersey, which outcrops for a distance of more than a hundred miles. Lewis has shown that its rock is chiefly a quartz diabase.* Since the sill-rock shows chilled contacts, it appears probable that its magma after reaching the sill-chamber, was too cool to accomplish much solution on roof or floor, though some xenolithic material (sandstone) may have been dissolved. The special composition and structure of the New Jersey sill can be explained as that of a syntectic of primary basaltic magma, which dissolved a small proportion of the acid rocks (sand- stones and pre-Cambrian ecrystallines) forming the walls of the feeding fissures. Though the temperature of this sill was too low for much evident solution of the sill’s floor and roof, it was high enough, and the sill magma therefore fluid enough, to permit of the remarkable gravitative differentiation deseribed by Lewis. Some acidification of the Purcell gabbro may. thus, have occurred in its long passage through the thick lower quartzites and other sediments of the Pureell series. Nevertheless, the great chemical similarity between the biotite granite and the average quartzite strongly suggests that the assimilation in the Moyie-sill magma chiefly occurred in the quartzite formation, and not in the underlying pre-Cambrian formations of differing composition. *J. V. Lewis, Annual Report, State Geologist of New Jersey, 1907, p. 99. —_ REPORT OF THE CHIEF ASTRONOMER 247 SESSIONAL PAPER No. 25a Assimilation through Magmatic Vapours——Again, the influence of mag- matic water and other vapours must be given due weight. The quartzites to-day are not entirely dry rocks. They must have been moister in that early time when the intrusions occurred. From heated roof and floor of each sill, and from each heated xenolith, water vapour must have been injected and forced into the sill magma. The volatile matter contained in assimilated sediment must similarly enter the magma. A large part of such vapour would rise to the roof, and there aid in the solution of the quartzite. Such resurgent vapour must not only lower the solution-point (of temperature) for the roof-rock; it must also specially metamorphose the sediment outside of the magma chamber. Tt is a fact that the quartzite above each sill seems to be more thoroughly erystallized than the quartzite below the sill. Summary of the Arguments for Assimilation.—The facts and deductions bearing on the subject are so numerous that it will be convenient to review -them in brief statement. The writer’s belief in the principle of assimilation as a partial explanation for the acid zones in the Moyie sills is founded on the following considerations :— 1. The strong mineralogical and chemical similarity between the biotite granite and the invaded quartzite. 2. The existence of solution aureoles about the visible xenoliths of quartzite. 3. The field evidences of superfusion in the sill gabbro. 4. The relation between sill-thickness (heat supply) and degree of acidifi- cation. 5. The necessary recognition of various loci of solution in the sill, namely, at root and floor, at xenolith contacts, and in the feeding channels below the sills. Resurgent and juvenile vapours, collected at the roof, must tend to hasten solution in that place specially. 6. The fact that differentiation may partially mask the direct evidence of assimilation. 7. The existence of many other sills and sill-like intrusions showing similar or analogous relations of gabbroid magma to sediments. Some of these cases will be listed after the nature of the differentiating process at the Moyie sills has been sketched. 8. The inadequacy of the hypothesis that the various phases of the sills are due only to the pure and simple differentiation of a primary earth-magma. This point is implied in the foregoing argument; it was briefly discussed in the writer’s 1905 paper. Gravitative Differemtiation—Inspection of Table XIII. and of Figure 15 will lead to the conviction that, in sills OC, D, and E, the igneous rock is strati- fied. In each of these instances the specific gravity increases from top to bottom of the sill. The same is true of sill A, with the exception of the shell of gab- broid rock next the roof and overlying the biotite granite. An explanation for this exceptional arrangement of zones has been given in the preceding descrip- 248 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 | Acidified gabbra d Granite Prttrtriitriyitirrtitt) Gabbro QUARTZITE 30° Bo Seer Gabbro 738’ Al QUARTZITE Biotite grartite Hornblende -biotite granite Intermediale Tock : BREREEREEEEE 530°C tl Gabbro T Hornblende- biotite granite intermediate Tock. 7050'D Gabbro QUARTZITE Figure 15. Diagram showing the petrographic nature of each of the- Moyie sills and its stratigraphic position in the quartzites. The sills are distinguished by letters; the approximate thick-- ness of each sill is indicated, in feet. REPORT OF THE CHIEF ASTRONOMER 249 SESSIONAL PAPER No. 25a tion of sill A. The zonal character of the four individual sills is clearly due to gTavitative adjustment. The same principle has probably controlled the rough system implied in the succession of average rock-densities of sills D, C, and A, as illustrated in the following table:— Sill. Approximate mean densities. The layer of quartzite separating sills C and D is only 75 feet thick. It is entirely possible that this layer is wedge-shaped, or else was penetrated by one or more connecting dikes. By such means B and C might have been in magmatic communication. We may imagine a partial differentiation within this larger chamber, whereby sill C became more acid, on the average, than sill D. Continued differentiation by gravity, within the partially separated masses © and D, led to the observable stratification of each. It is not impos- sible that all four sills were similarly connected in a common (sill-like) magma chamber, from which each visible sill was a kind of great, flat apophysis. On the other hand, these sills may not have been of exactly contemporaneous intrusion. A large part of the magma now represented in the rock of sill A may Eave formerly rested in the chamber of sill B. Since then, after partial differentiation, that part of the magma may have broken through the roof of chamber B to the new horizon now occupied by sill A, where continued differ- entiation produced the actual zonal arrangement. Similarly, sills A, B and C may have been apophysal from the great sill D. One cause for such successive injections may be found in the enormous gas-pressure generated by the assimi- lation of moist quartzite—a tension amply sufficient, under certain conditions, to fissure the roof of the slightly older chamber and cause the rise of the magma to the higher horizons of the existing upper sills. (See Figure 16.) In spite of the relative complexity of the whole system, we may conclude that gravitative differentiation is clearly the dominant process in developing the zonal structures of the Moyie sill group. It is hardly necessary to dwell on the chemical side of the differentiation. It was probably founded on the limited miscibility of gabbro and secondary magma at the low temperature immediately preceding crystallization. The magma was not quite the chemical equivalent of the invaded sediments. Each of the two granite types contains more ferrous iron and lime than the average quartzite. The hornblende-bearing granite is clearly more ferromagnesian and calcic than the sediment. However, the total volume of the gabbro in the sill system is so great that its average original composition was not essentially affected through the transfer of the extra lime, iron oxides, and magnesia to the granite zones. Sinular and Analogous Cases——The writer’s explanation of the Moyie sills has been greatly strengthened by the discovery of similar features in other 250 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 thick basic sills cutting silicious sediments. More indirect corroboration is offered in certain cases where large basic injections have become differentiated by gravity, apparently after the absorption of considerable limestone. poneanee SOE QUARTZITE Qu UAI R T Z1 TE QUARTZITE eee Gece pees HERREUESE ZEEE Sal i UARTZITE Be Peed Bes Eo) Eeael eam | | QUART ZITE Acidified sgn TTT Toon Figure 16. Diagram illustrating the hypothesis that the partially differentiated syntectic magma of a thick sill may break through the roof and form, at stratigraphically higher horizons, several thinner sills differmg in composition among themselves. Some later differentiation in the derived sills is assumed. The original sill is shown in the drawing on the left : the derived sills and the remnant of the original sill are shown in the drawing on the right. The channels (dikes) connecting the sills are not indicated . Direct parallels to the Moyie sills have already been noted as occurring in the Purcell range to the east of the Moyie river. Schofield has found many other stratified sills in the same range north of the Boundary belt.* At Sud- bury, Ontario; at Pigeon Point, at Governor’s Island, and at Spar, Jarvis, and Victoria islands, and other localities in Minnesota, and among the Logan sills on Lake Superior the same general association of gabbroid-granitic magmas and quartzose sediments occurs. These instances were cited in the writer’s 1905 paper, where a rather full summary of the facts concerning Pigeon Point and Sudbury intrusives was given. There, and still better in the original memoirs of Bayley, Barlow, and Coleman, the reader will find evidence of the extreme similarity of these cases to that of the Moyie sills. Many other examples have been described in late years, but it is hardly appropriate or necessary to note them individually in the present report. * 8S. J. Schofield, Summary Report of the Director, Geol. Survey of Canada, 1909, p. 186, and 1910, p. 1381. REPORT OF THE CHIEF ASTRONOMER Qi SESSIONAL PAPER No. 25a Likewise significant is the analogy of the Moyie sills to several igneous masses which have cut thick limestones and have then undergone differentiation by gravity. The well known laccoliths of Square Butte and Shonkin Sag, in Montana, have been ably described by Weed and Pirsson.* In a later, independent publi- cation, Pirsson has described the differentiation as due to the combined effect of fractional crystallization, convection currents, and gravity.t In the present writer’s opinion, thermal convection must be of infinitesimal strength in such bodies and he cannot find adequate explanation of the shonkinite and other basic phases of these sills in fractional crystallization. On the other hand, the writer finds most satisfaction in the view that the leucite-basalt porphyry of Shonkin Sag, occurring at top and bottom ofthe laceolith, represents the quickly chilled magma originally injected into the chamber. The syenite and the shonkinite are the two poles of a gravitative differentiation of the remaining leucite-basalt magma, which, in the heart of the mass, remained fluid long enough for splitting. The segregation of the two polar magmas is of a kind suggesting limited miscibility between them. The leucite-basalt can be explained as itself a differentiate from basaltic magma which had dissolved a moderate amount of the thick pre-Tertiary limestones traversed by the mag- matic feeder of this laccolith. A similar explanation may be applied to Square Butte. Tyrrell has described another noteworthy analogy in the Lugar sill of Scotland, where the alkaline pole is teschenite overlying the femic pole, a picrite. This sill is injected into the Millstone Grit; its feeder doubtless tra- versed the underlying Carboniferous limestones and perhaps absorbed them in some measure. ‘Tyrrell explains the differentiation of the Lugar sill in essentially the same way as that outlined by the present writer for the Shonkin Sag laccolith.t Shand has recently described a large laceolith near Loch Borolan, Scotland, in which quartz syenite (specific gravity 2-635) overlies quartz-free syenite (specific gravity 2-65), which in turn overlies nephelite syenite (specific gravity 2-67), and ‘ledmorite’ (specific gravity 2-74— 2-78). This mass clearly cuts thick Cambrian limestone and other sediments. Shand attri- butes the layered condition of the laccolith to differentiation under gravity.$ He makes no statement as to the origin of the magma thus differentiated. On account of its ‘ desilicated’ character, the present writer is inclined to sus- pect its derivation from a basalt-limestone syntectic. Finally, the thick sill described by Noble as cutting the shales in the Colorado canyon is worthy of special emphasis in the present connection.** The *W.H. Weed and L. V. Pirsson, American Journal of Science, Vol. 11, 1901, p. 1. and Fort Benton Folio, U.S. Geological Survey, 1899. +L. V. Pirsson, Bulletin 237, U. S. Geological Survey, 1905, p. 42. tG. W. Tyrrell, Transactions of the Geol. Society of Glasgow, Vol. 13, part 3. 1909, p. 298. $S. J. Shand, Transactions of the Geol. Society of Edinburgh, Vol. 9, 1910, p. 376. #*7,. F. Noble, American Journal of Science, Vol. 29, 1910, p. 517, 252 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 main mass of this intrusive is olivine diabase. Towards its upper contact the diabase appears to grade into a pink rock, which proved to be a hornblende syenite. The syenite makes a sharp contact with the argillites at the roof of the sill. The floor rock is also an argillite. The feeding channel or channels must have traversed the calcareous shale, limestone, and jasper of the lower Unkar series. Noble does not discuss the genesis of the syenite. Is this rock not a gravity differentiate from a syntectic of the various sediments dissolved in diabase magma? Since the argillites are dominant, one must expect for the differentiate a degree of acidity intermediate between that of the lighter differentiate in a Moyie sill which has assimilated quartzite, and that of the lighter differentiate in a sill which assimilated limestone. This expectation seems to be matched by fact: In these and other cases which might be cited, the chemical composition of each lighter, salic pole in differentiated sills varies with the chemical com- position of the invaded sediments. Herein rests a powerful argument favour- ing the secondary origin of the respective salic or alkalic magmas. As indicated in each instance, the intrusive mass is stratified in‘ the way demanded by the theory of gravitative differentiation in a syntectic. GENERAL CONCLUSION AND APPLICATION. The remaining paragraphs of this chapter are devoted to the broader bearing of the main conclusions regarding the Moyie sills. The statement is almost identical with that already published in the writer’s 1905 and 1906 papers, but a few changes have been made in the form of presentation. Sooner or later experience must teach every careful field student of igneous rocks the truth of the principle of magmatic differentiation. That principle is, indeed, so generally accepted by petrologists that it may be considered as a permanent acquisition in the theory of their science. Yet it is a long step from the recognition of the doctrine to its application to the origin of igneous rocks as actually found in the earth’s crust. The principle becomes really fruitful, in fact becomes first completely understood and realized, when certain chief problems have been solved. Among those problems there are naturally three that are fundacewen Only after they are solved has petrology done that which it has set out to do, namely, determine, under the difficult conditions of earth study, the true nature and genesis of rocks. The first insistent question is, in every case, what was the magmatic mixture or matrix from which the material of the existing rock-mass or rock-masses was produced through differentiation? The second question is, how far did the differentiating process operate? The third insistent question is, what was the process of differentiation itself? All three problems are interdependent and involve a study in structural geology. They cannot be solved simply by acquiring even the fullest informa- tion to be derived from single plutonic contacts, nor, as a rule, from such as may be derived from entire ground-plan contact lines. On the other hand, REPORT OF THE CHIEF ASTRONOMER 253 SESSIONAL PAPER No. 25a it is necessary that, more or less completely, the petrologist shall know his magma chamber as the chemist or metallurgist knows his crucible. No student of fused slags can obtain safe results from the profoundest examination of merely one surface or one section of the fused product. He must think: in three dimensions. In the same way, the petrologist attempting to unravel the complex history of a magma chamber, should, ideally, know its general shape, size, and contents, as well as the method by which the chamber has been opened within the earth’s crust. Until these conditions are fulfilled his problem of rock-genesis through magmatic differentiation remains wholly or in _ part unsolved. The geologist knows how hard those conditions are. He is dependent upon erosion’s rendering his contact accessible; yet erosion destroys surfaces of contact. He can find no bottom to the chamber of stock or of batholith, though large-scale differentiation is most commonly evinced in stocks and batholiths. It is not to be wondered at that, notwithstanding the great number of des- cribed instances of magmatic differentiation, the phenomenon itself is so little understood or that the origin of the igneous rocks is still shrouded in the mists of hypothesis. In view of the difficulties surrounding the study, the discovery of single cases where the requisite field conditions are tolerably well fulfilled, merits special statement. Descriptions of bodies differentiated in chambers of known form are in the highest degree rare. Nevertheless, it is precisely in the light of these rare cases that the laws of differentiation can be most intelli- gently discussed. Such instances are discussed in this chapter, in which have been described exceptionally clear examples of differentiation within magmatic chambers, the erystallized contents of which can now be examined from top to bottom. The form and geological relations of the chambers are sufficiently well determined to serve for the discussion of the magmatic problem. The general nature of the magma whence differentiation has evolved the existing igneous rocks is believed to be deducible from the field and chemical relations in each case. The compound magmas were themselves derived, owing their composition to the digestion or solution of acid sedimentary rocks in original gabbro magmas. Finally, the facts seem indisputable as to the nature of the method by which the differentiation took place. The actual segregation of the sub-magmas appears to have been directed by gravity, producing simple stratification in the chambers. In each sill the less dense sub-magma of splitting overlies the denser sub-magma of splitting. In almost every case the opponents of the assimilation theory have treated of the assimilation as essentially a static phenomenon. Each interpretation of field facts has been phrased in terms of magmatic differentiation versus magmatic assimilation as explaining the eruptive rocks actually seen on the contacts discussed. Nothing seems more probable, however, than that such rocks are often to be referred to the compound process of assimilation accom- panied and followed by differentiation. The chemical composition of an intru- sive rock at a contact of magmatic assimilation is thus not simply the direct 254 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 product of digestion. It is a net result of rearrangements brought about in the compound magma of assimilation. In the magma, intrusion currents and the currents set up by the sinking or rising of xenoliths must take a part in destroying any simple relation between the chemical constitutions of the intru- sive and invaded formation. Still more effective may be the laws of differen- tiation in a magma made heterogeneous by the absorption of foreign material which is itself generally heterogeneous. The formation of eutectic mixtures, - the development of density stratification, and other causes for the chemical and physical resorting of materials in the new magma ought certainly to be regarded as of powerful effect in the same sense. A second fundamental principle has, as a rule, been disregarded in the discussions on magmatic assimilation. If differentiation of the compound magma has taken place so as to produce within the magma chamber layers of magma of different density, the lightest at the top, the heaviest at the bottom, the actual chemical composition of the resulting rock at any contact will depend ‘directly on the magmatic stratum rather than on the composition of the adjacent country-rocks. In the foregoing discussion the secondary origin of some granites has been deduced from the study of intrusive sills or sheets; but it is evidently by no means necessary that the igneous rock body should have the sill form. The wider and more important question is immediately at hand—does the assimilation-differentiation theory apply to truly abyssal contacts? Do the granites of stocks and batholiths sometimes originate in a manner similar or analogous to that just outlined for the sills? General reasons affording affirmative answers to these questions are noted in chapter XXVI. Gabbro and granophyre are often characteristically associated at various loealities in the British Islands as in other parts of the world.t The field rela- tions are there not so simple as in the case of the Moyie sills, for example, but otherwise the recurrence of many common features among all these rock- associations suggests the possibility of extending the assimilation-differentia- tion theory to all the granophyres. Harker’s excellent memoir on the gabbro and granophyre of the Carrockfell District, England, shows remarkable parallels between his ‘ laceolite’ rocks and those of Minnesota and Ontariot At Carrock Fell there is again a commonly occurring transition from the granophyre to true granite, and again the granophyre is a peripheral phase. Still larger bodies of gabbro, digesting acid sediments yet more energetically than in the intrusive sheets, and at still greater depth, would yield a thoroughly granular acid rock as the product of that absorption with the consequent differentiation. The difficulty of discussing these questions is largely owing to the absence ot accessible lower contacts in the average granite body. All the more valuable +See A. Geikie, Ancient Volcanoes of Great Britain, 1897. t Quart. Journal Geol. Soc., Vol. 50, 1894, p. 311 and Vol. 51, 1895, p. 125. REPORT OF THE CHIEF ASTRONOMER 255 SESSIONAL PAPER No. 25a must be the information derived from intrusive sills. The comparative rarity of such rock-relations as are described in this chapter does not at all indicate the exceptional nature of the petrogenic events signalized in the Moyie, Pigeon Point, or Sudbury intrusives. It is manifest that extensive assimilation and differentiation can only take place in sills when the sills are thick, well buried, and originally of high temperature. All these conditions apply to each case cited in this chapter. The phenomena described are relatively rare largely because thick basic sills cutting acid sediments are comparatively rare. On the other hand, there are good reasons for believing that a subcrustal gabbroid magma, actually or potentially fluid, is general all around the earth; and secondly, that the overlying solid rocks are, on the average, gneisses and other crystalline schists, and sediments more acid than gabbro. Through local, though widespread and profound, assimilation of those acid terranes by the gabbro, accompanied and followed by differentiation, the batholithic granites may in large part have been derived. True batholiths of gabbro are rare, perhaps because batholithic intrusion is always dependent on assimilation. The argument necessarily extends still farther. It is not logical to restrict the assimilation-differentiation theory to the granites. For example, the prepar- ation of the magmas from which the alkaline rocks have erystallized, may have been similarly affected by the local assimilation of special rock-formations. See chapter XXVIII. The officers of the Minnesota Geological Survey have shown that the same magma represented in the soda granite and granophyre of Pigeon Point forms both dikes and amygdaloidal surface flows.* The assimilation-differentiation theory is evidently as applicable to lavas as to intrusive bodies. But demon- stration of the truth or error of the theory will doubtless be found in the study of intrusive igneous bodies rather than in the study of voleanoes either ancient. or modern. Finally, the fact of ‘consanguinity’ among the igneous rocks of a petro- graphical province may be due as much to assimilation as to differentiation. *N. H. Winchell, Final Rep. Minn. Geol. Surv., Vol. 4, 1899, pp. 519-22. The Duluth gabbro and the broad fringe of red rock (partly extrusive) on the southeast, together seem to form a gigantic replica of the Pigeon Point intrusive! 2 GEORGE V- SESSIONAL PAPER No. 25a A. 1912 CHAPTER XI. STRATIGRAPHY AND STRUCTURE OF THE SELKIRK MOUNTAIN SYSTEM (RESUMED). Between the Purcell Trench and the Selkirk Valley (Columbia river) the ten-mile belt includes stratified rocks belonging to four groups in addition to those forming the Summit series. (Maps No. 6, 7, and 8). These other groups have been named the Priest River terrane, the Pend D’Oreille group, the Kitchener quartzite, and the Beaver Mountain group. The first two groups rival the Summit series in areal importance within the Boundary belt. The Kitchener quartzite and the Beaver Mountain group cover but small patches and their description can be given in few words. The Beaver Mountain sedi- ments are intimately associated with basic voleanic rocks which in turn are involved with the Rossland Voleanic group. Their description is best post- poned to chapter XIII, in which the igneous rocks of the Rossland mountains are discussed. KITCHENER FORMATION. Along the western edge of the Kootenay river alluvium and north of the Rykert granite opposite Porthill, the foot-hills are composed of unfossiliferous quartzite and interbedded metargillite, which in lithological characters are essentially like the Kitchener strata across the river. These beds are appar- ently not metamorphosed in any sense different from that which is true of the unfolded Kitchener quartzite of the Purcell mountains; that is, one misses in them tke evidences of great dynamic metamorphism, intense mashing, and recrystallization observed in the neighbouring Priest River terrane and the evidences of likewise intense contact metamorphism which has affected the Priest River rocks in the batholithic aureoles farther west. The relative lack of dynamic metamorphism is quite striking and largely on that account the writer has separated these rocks from the Priest River terrane, postulating a master fault of great throw on the west side of the Purcell Trench. This fault is thus considered as bringing into contact a very old member of the Priest ‘River terrane (Belt G) and the quartzite which is tentatively correlated with the Kitchener formation. The down-throw is on the east (see map), and may be as much as 30,000 feet. On account of the great structural importance of this correlation a detailed study of the sediments west of the alluvial flat of the Kootenay is imperative. While in the field the writer was not entirely conscious of the importance of the Jithological comparison, for at that time the existence of the Kitchener forma- 25a—vol. i1—17 257 258 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 tion itself was unknown and was not determined until the camp had been moved many miles to the eastward. Since then no favourable opportunity has arisen by which the study of this quartzite could be eantinued in the field. It is now only known that, throughout most of the meridional belt of the Kitch- ener quartzite as mapped on the west side of the Kootenay, the rocks are indis- tinguishable from types of the Kitchener strata collected at the Moyie river. The staple rock is a greenish gray quartzite, weathering brownish. Under the microscope the dominant quartz is seen to be regularly associated with small grains of microperthite and orthoclase, with generally a little plagioclase, a few zircons, and pyrite crystals. There is always mica present, generally colourless and sericitic, though minute biotites are seemingly never absent. Where the quartzite is cleaved, as it is at certain points north of Corn creek, the micas are specially developed in the cleavage planes. The metargillitic interbeds have not been microscopically examined but they appear to be com- posed of the same materials as the metargillites of the Kitchener formation. It is equally true that this local quartzite-metargillite series is lithologically similar to the Beehive formation as developed on the summit of the range. This is, of course, natural if the writer is correct in correlating the Kitchener and Beehive quartzites. At Summit creek and north of it for a half-mile the quartzite is extremely massive and of a gray colour when fresh, and very often grayish to light brown- ish-gray when weathered; it is possible that here we have a large outcrop of the Creston formation underlying the Kitchener. There is so little certainty of this, however, that the colour representing the Kitchener on the map has been extended northward across Summit creek. North of Summit creek the strike averages about N. 16° E., and the dip is about vertical. The same strike (dip observed at 60° E.) is preserved fairly well for a couple of miles south of the creek when it abruptly changes to N. 22° W. then to N. 90° E., becoming highly variable in a locality of structural turmoil. A half-mile farther south the strike is N. 45° E., and the average dip about 50° S. E. This general attitude of the beds was observed at several points south of Corn creek. On the whole it must be said that the strike of the caarieie is distinctly transverse to the trend of the Purcell Trench. The western limit of the quartzite is shown on the map only apoeexinee For the reason already noted, the amount of structural and areal work done in the field was insufficient to show that limit and therewith the exact place of the postulated master fault. Few points in the structure section along the Forty-ninth Parallel are more important than this one and it is especially here that further and more detailed work is needed. Priest RIveR TERRANE. Tt has already been noted that the basal conglomerate of the Summit series. rests uneonformably on older rocks outcropping at, and to the eastward of, the bead-waters of Priest river. The name ‘ Priest River terrane’ may be appro- Ieee 2 Looking eastward over the heavily wooded mountains composed of the Priest River Terrane; Nelson Range. Glazial lakes (rock-basins) in Irene Conglomerate formation. 25a—vol. 1i—p. 255. REPORT OF THE CHIEF ASTRONOMER 259 SESSIONAL PAPER No. 25a priately given to this whole group as exposed in the southern Selkirks at the Boundary. It appears to be the oldest series anywhere exposed on the Forty- ninth Parallel. The group is of sedimentary origin but has been largely recry- stallized. It is as yet entirely unfossiliferous. Its stratigraphic relation to the Summit series leaves no room for doubt that the Priest River terrane is both pre-Cambrian and pre-Beltian in age. Exposures and Conditions of Study.—Within the 10-mile Boundary belt where it crosses the Selkirk range, this old terrane eovers about one hundred square miles. Such an area would seem sufficient to afford leading data as to the composition and structure of the series. Yet a comparatively long and certainly arduous field attack on the area has been execedingly unsatisfactory im its results. The difficulties of geological exploration in this area are unsur- passed in the entire Boundary section. The intense metamorphism of the series in almost every part, and its structural complexity would alone render the solution of the main geological problems as difficult as in most typical Archean terranes. The strong relief of the country and, above all, the heavy and con- tinuous forest cap add special physical troubles in a field where the geologist’s mental troubles in interpretation are already of the first order. (Plate 26.) With wearisome repetition outcrops failed at critical loealities. For a mile or two together the sections were often left quite blank where fallen timber, deep moss, or humus effectually covered the rock ledges; so complete was this cover of vegetation that even the ‘wash,’ frost-riven from the ledges, was invisible for long stretches. Under these conditions it has proved impossible to treat the Priest River terrane in anything like as satisfactory a manner as would be desirable. Though its rocks are almost entirely of clearly sedimentary origin, not the slightest clue was discovered as to the succession of beds. Neither top nor bottom, nor certain indication of relative ages among individual members has yet been determined. Four, more or less complete, traverses, besides several shorter ones, were run across the area, and a tolerable idea of the lithological nature of the series was obtained. The map and section as well as the following des- cription of the series, indicate that the characters of the roeck-members and the attitudes of the beds are not favourable to the discernment of stratigraphic sequence. It has thus seemed best to map the series on a purely lithological basis. Compiling the data won from the several traverses it appears that the rocks of the terrane may be grouped into seven irregular belts which will be hence- forth referred to by the letters A to G. In general they run meridionally and follow, more or less faithfully, the strike of the bedding planes, which appear usually to lie parallel to the planes of schistosity. Belts A, B, C, D, and E have been most fully investigated. The relative inaccessibility of the area covered by belts F and G has caused the information concerning them to be very scant. Along the northern edge of the Boundary belt all the belts exposed show specially complicated features as a result of the intrusion of the great 25a—vol. 1i—174 260 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Bayonne batholith. Peripheral schistosity and cleavage and a very intense degree of recrystallization have been developed about that batholith. Belts F and G are also much disturbed and altered in the vicinity of the Rykert granite batholith in the southeastern corner of the area. The eastern limit of belt G occurs at a master fault, along which quartzites referred to the Kitchener formation have been dropped down into contact with the pre-Cam- brian schists. Petrography of Belt A-—South of Summit creek the Irene conglomerate directly overlies belt A. This is a heterogeneous group of rocks, including biotite, chlorite, and sericite schists; sheared, compact quartzites; and dolo- mites. The micaceous sehists occupy most of the belt; sericitic quartzites are next most abundant; the dolomites occur as thin bands intercalated in schist and quartzite. The schists vary in colour from light to dark greenish gray, according to the nature and abundance of the essential micaceous mineral, sericite, biotite, or chlorite. They are often interrupted by veinlets of quartz and of dolomite lying in the sehistosity planes. In certain phases crystals of ‘dolomite occur in individuals or groups disseminated through the schist. Rock types transi- tional between the true schists and impure dolomite are found. On the west side of the trail at Copper Camp the dark phyllitie schist is abundantly charged with single crystals and small clumps of a light brown ferruginous carbonate which is probably ankerite. The rock has, in consequence, a pseudo-porphy- ritic appearance. The quartzitic bands sometimes run over a hundred feet in thickness. They are always sheared, with an abundant development of sericite in the shearing planes. At several localities the quartzites, like the schists, are magnesian to some extent. They thus pass over into the dolomites which have the habit of compact, more or less silicious, marbles. On fresh fractures the dolomites range in colour from white to a delicate pinkish-brown, weathering to a light though decided buff tint. The exposures of the dolomites in belt A are very poor but it appears that no one bed measures much over fifty feet in thickness. Throughout most of the belt the strike of both bedding and schistosity averages a few degrees west of north and seems to cut the plane of uncon- formity with the Summit series at angles varying from 10° to 25°. The dip is generally nearly vertical but angles of 75° to 80° to the eastward are not uncommon. About one mile south of the Dewdney trail the belt is broken by a strong transverse fault along which, as shown in the map, the block to the south has been displaced westward with respect to the block on the north. Within the northern block the belt rapidly narrows down as if there it had been cut away during the erosion preceding the deposition of the Irene con- glomerate. In this short tongue of belt A the strike averages about N. 30° E.; the dip about 75° northwest. Large quartz veins, usually lying in the planes of schistosity are common in the schists. One of these veins, from 15 to 20 feet in thickness, and well REPORT OF THE OHIEF ASTRONOMER 261 SESSIONAL PAPER No. 25a exposed in a high cliff occurs at a meadow on the divide between Priest river and a small fork of Summit creek. Fifty feet to the eastward of this vein are two narrow sill-like injections of minette. This association of vein and erup- tive prompted the assay of the quartz for values in the precious metals. The result was negative. The dolomites of the belt characteristically bear isolated crystals and small pockets of galena and chalcopyrite, and some active prospecting of these rocks bas taken place at the forks of Priest river. The sulphides are reported to carry both silver and gold, but so far no workable lode has been discovered. The pockets of galena form the principal ‘ ore’ of the prospect-dumps but the small size and rarity of the pockets—clumps of crystals only a few inches in diameter at most—have led to the abandonment of the claims, which certainly seem to have no commercial value. The intrusive rocks occurring in belt A will be described in the section on the igneous bodies of the Selkirks. Petrography of Belt B.—The next belt to the east is, so far as lithological types are concerned, very similar to belt A; the chief contrast between the two lies in the different proportions of these types in the belts. Belt B bears thick and persistent bands of dolomite alternating with quartzites and phyllitie and coarser mica schists. The best exposures were seen on the divide between Priest river and Summit creek, to the northwestward of North Star mountain. A tolerably complete section of the belt was there made. At the northwest end of this section the western limit of belt B occurs at a bed of silicious dolomite, one hundred feet in thickness. This dolomite is white to bluish white on the fresh fracture but weathers buff-yellow. Though generally massive, it is greatly cracked and shattered, the cracks being filled with vein-quartz which ramifies in all directions through the rock. The strike is N. 9° E.; the dip is practically vertical. That limestone is followed on the east by 110 feet of biotite schist, which in turn is sueceeded by about 800 feet of thinly laminated, schistose silicious dolomite of colour and composition like the first limestone. This rock too is highly charged with narrow, irregular veinlets of white quartz. The strike is here north and south; the dip, about 65° E. This second limestone is succeeded on the southeast by a 150-foot band of dark, glossy biotite-sericite schist with _ its planes of schistosity striking north and south and dipping 70° E. It is followed by 95 feet of white dolomitic quartzite (weathering yellowish) with conformable attitude. The quartzite is succeeded by a thick band of light to dark greenish gray phyllitic mica schist. The observed width of this band was 1,400 feet across the strike, which runs N. 10° E. The dip is 85° E. On its eastern limit this schist is in contact with a band of dolomitic quartzite of which the thickness measured 340 feet. Here too this rock type is white on the fresh fracture and weathers buff-yellow. The staple dolomitie quartzite is interlaminated with thin beds of nearly pure dolomite and others of nearly pure white-weathering quartzite. The strike is N. 5° E.; the dip, 85° E. Next to that band, on the east, comes a conformable band of phyllite, followed 262 DBPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 by another band of dolomite, which is very similar to the first dolomite occur- ring at the western end of the section. The dolomite there is about 450 feet thick. The specimen of this dolomite (No. 886) is fairly typical not only of the whole band but also of the whole group of carbonate bands occurring in the Priest River terrane. It has, accordingly, been selected for chemical analysis. On the fresh fracture the rock varies in colour from white to pale blue and weathers rather uniformly brownish yellow or buff. It is transected by numer- ous veinlets of white quartz and by others of very compact dolomite. Other- wise the rock is a very homogeneous, fine-grained, marble-like mass of carbonate, which in the ledge shows no appreciable impurity. The specifie gravity is 2.822, corresponding to normal dolomite pretty closely. The analysis by Professor Dittrich, afforded the following result :-— Analysis of dolomite, Priest River terrane. Mol Si0, eo 5-84 097 : Al,O, : 80 008 Fe,0, oe “79 005 e eke 16 002 MgO A 19-38 °485 CaO pads 28-31 506 Na,O Aegean vale 027 004 MGSO Weer acho etait ital have trols. ve. sean tie ENON ores bicte nese linarouicars Mes 09 001 FR Ocat t10oCs ae ee ty cr a a eel Mae an 08 HeOvrabove 10CO 5. taht or edase Rive wed silos amas 63 035 CO pane it neh ar Pacan dvb een ett ty oe Forays Nats etoicetct reset ten sve 43-55 990 99.85 See ies hres acd eas eels se PUY ru cleaner yy cly gine Ute: Ole Portion insoluble in hydrochloric acid, 5-96% Under the microscope the carbonate is seen to occur in the form of a granular aggregate, the grains being of rather uniform size and averaging about 0-08 mm. in diameter. They never show the rhombohedral outlines so common in the dolomites of the Lewis and Galton series. This difference may be easily explained by the fact that all of the Priest River dolomites have been thoroughly recrystallized and now have the structure of true marble, while the younger dolomites seem to have preserved their original sedimentary structure more or less perfectly. The granular dolomite of the thin section is inter- rupted by a few small grains of glass-clear quartz and feldspar. The visible quantity of these impurities matches well the portion of the rock found to be insoluble in hydrochloric acid. About 94 per cent of the rock by weight is made up of the carbonate, which. as shown by the ratio, CaO: MgO (1-46:1), is almost ideal dolomite. It happens that a smail veinlet of carbonate, cross-cutting the main mass of the rock, appears in the thin section studied. This veinlet is about 1 mm. ¢ in diameter. Throughout its visible extent its grains average about 0-02 mm. REPORT OF THE CHIEF ASTRONOWER 263 SESSIONAL PAPER No. 25a in diameter or sensibly equal to the average diameter of the grains in the Waterton, Altyn, Siyeh, and Sheppard dolomites. Here as there we have a steady persistence in the size of grain which characterizes the chemically pre- cipitated carbonate. The strike and dip of the 450-foot band of dolomite was, on account of the massiveness of the rock, not readily determined but, as usual in the zone, the former was a few degrees east of north, while the dip seemed to be nearly vertical. East of the analyzed dolomite, outcrops were few for about 400 feet of cross-section but that stretch seems to be underlain by dolomitie chlorite schist and phyllitic mica schist. Immediately to the eastward and just at the western base of North Star peak, a 200-foot, nearly vertical, band of sheared dirty- white dolomite, weathering yellow, forms the most easterly part of belt BD. The strike of the band and of its schistosity planes is about N. 5° K.; the dip averages 80° E. A review of the field-notes suggests that belt B may constitute a closely appressed fold, the erosion of which has produced a duplication of the three dolo- mitie bands on the two sides of the belt. However, the very considerable differ- ences of thickness between the respective bands thus supposed. to be duplicated, are so great that one cannot be sure of the postulated repetition. In any case, there is no evidence in this section as to whether the fold is an anticline or a syncline. In no other part of belt B could this point be settled. In the general structure section, therefore, no attempt is made to show the true relations in the great monocline. It has seemed better to illustrate simply the empirical facts of field observation rather than to attempt the projection of folds which, under the circumstances, could be nothing else than fanciful. Belt B is, thus, composed of both mica schists and dolomites. In that belt the carbonate rocks are relatively more abundant than in any other belt in the Priest River terrane. The persistence of the dolomites along the strike, their nearly vertical dip, the notable straightness of each bed and of the entire belt across North Star mountain, and the general parallelism of belts A and B to the band of Irene conglomerate, would, at first sight, suggest that at least the upper part of the Priest River series is conformable to the overlying Summit series. It is believed, however, as noted elsewhere, that this general parallelism of belts A and B to the band of Irene conglomerate is partly an incidental result of the strong upturning and mashing which have forced the two unconformable series into positions of apparent conformity. Six miles north of the Boundary, belt B has been broken by the same fault which offset belt A a mile or more south of Summit creek. At the creek itself the dolomites and schists of belt B are entirely cut off by a second fault (see map) so that these rocks are replaced, north of the Dewdney trail, by the sheared quart- zites characteristic of belt C. The dolomitic bands of belt B, like those of belt A, carry small bunches of galena and occasional crystals of copper pyrites. Neither of these ores where they have been actually prospected, as at the claims of ‘Copper Camp,’ 264 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 occurs in masses of workable size. No reliable information was obtained on the ground as to the values found in assayed specimens of the sulphidés, but the material collected from the prospect-dumps nowhere suggested the possi- bility of a high-grade property. On the other hand, the small size and com- parative rarity of the bunches of ore shows that no known claim in the ‘ camp’ can prove successful as a low-grade mine. a Petrography of Belt C.—In width, length, and axial trend, belt C is very similar to belts A and B. In composition C is, in some respects, like A but does not seem to bear any dolomitic bands. The most complete section across belt C was made at the summit of North Star mountain. Elsewhere in the belt, exposures are very poor and it is very possible that the boundary lines, especially that on the eastern side, are drawn too straight. ‘This third belt is composed essentially of well and thinly foliated phyllites, chlorite-sericite schists, and phyllitic biotite-sericite schists, all tending toward a dark greenish gray colour. Within these staple rocks there occur strong bands of a very dark gray intensely sheared quartzite. The quartzite bears abundant little ‘foils of sericite and biotite, disseminated in planes of schistosity. The inter- locking, metamorphic quartz grains are full of opaque black dust which may be driven off before the blow-pipe and is probably carbon in graphitic or other form. This carbonaceous matter is abundant and explains the dark colour of the rock in ledge or hand-specimen. A few sheared quartz pebbles were found in the phyllite on North Star mountain near the western limit of the belt. Wherever outcrops were found in the belt the attitude of the planes of schistosity corresponds well to the average attitude in belts A and B. Through most of the belt the strike varies from N. 7° E. to N. 10° W.; the dip averages about 75° E. At one locality near the summit of North Star mountain, the dip of the schistosity plane was 75° EK. Such discordance appears, however, to be local and, in general, the planes of bedding and schistosity may be nearly coincident. The schists do not extend beyond the Dewdney trail and seem to be cut off by the same transverse fault which has been postulated to explain the failure of belt B north of the trail and so marked on the map. Petrography of Belt D.—The fourth belt is dominantly quartzitic. The quartzite is normally more or less sheared. Both biotite and sericite are largely - developed, in fact never failing entirely in this metamorphosed sedimentary. Within the quartzite beds are numerous, though thin intercalations of sericitic and echloritic schists along with beds of dolomite. The quartzites are of com- pact texture and vary in colour from white to pale greenish-gray, weathering white or buff. They are often charged with accessory grains of carbonate, which qualitative analysis shows to be probably typical dolomite. The same mineral is also an abundant accessory in the chlorite and biotitic schists, The study of thin sections seems to show that much, perhaps all, of the chlorite found in the schists is secondary after biotite and after the rather rare garnets which sometimes appear among the accessories. REPORT OF THE CHIEF ASTRONOMER 265 SESSIONAL PAPER No. 25a At Summit creek and north of it, the rocks of the belt have been pro- foundly metamorphosed by the Bayonne granodiorite intrusion. ‘The effects are most notable in the schistose bands. In them the small shreds and foils of sericite, chlorite and biotite are replaced by felted aggregates of large biotite and muscovite foils. The resulting coarse-grained mica schist bears a most striking contrast to the more plyllitic schists far from the batholithic contact. Though the recrystallization by contact-action is so pronounced, the original banding or bedding is as fully marked as in the staple phases of these old sedimentaries. The thin bands of coarse schist are sharply marked off from the enclosing quartzite, which, though it bears disseminated plates of biotite and muscovite of relatively large size, is still a true hard quartzite. Occasion- ally minute, deeply coloured tourmalines are seen under the microscope to be distributed through the quartzose matrix. Feldspars are characteristically absent, or at least, are indeterminable in the normal schist and quartzite, but both plagioclase and orthoclase are recognizable in considerable amounts in the schists and quartzites of the thermally metamorphosed part of the belt. It is not possible to attribute their presence with certainty to feldspathization by the granitic magma, however probable it may seem from the field relations ‘of the feldspar-bearing phase. Strong contact-metamorphism is visible for at least two miles from the granite contact. The great width of the metamorphic collar as illustrated in belts D, FH and G indicate the probability that the contact-surface of the granite body plunges under the rocks at and south of Summit creek. The vertical dis- tanee between the granite and the rocks exposed in the depths of the canyon at the creek is probably less than two miles. (See Figure 19.) The best exposures of the belt were found on the ridge running southeast from North Star mountain. There the strike of the bedding, the planes of which usually coincide with the schistosity planes, varied from N. 25° W. to N.—and—S., the dip varies from 75° E. to 70° W., with the average about 85° E. The nearly vertical dip and meridional strike persist for a distance of some six miles north of the Boundary line; but along Summit creek the strike has swung around, so as to run, on the average, about N. 40° E. near belt C and gradually approaching N. 65° E. as the eastern limit of quartzites on the Dewdney trail is approached. Throughout the whole width of the belt on the Dewdney trail, the strike thus follows very closely the general contact-line of the Bayonne granite; the relation affords an excellent illustration of the development of peripheral cleavage about a batholith. The dip of the banding (bedding) in the schist-quartzite along this contact collar seems to coincide generally with the ‘dip of the schistosity. It averages about 60° to the north- west, but is highly variable, as expected in a belt of rocks energetically dis- placed and mashed during batholithic intrusion. Petrography of Belt E.—Belt Eis composed of a group of acid sediments even more intensely metamorphosed than those of belt D. The dominant type is a highly sericitic schist in which large biotite foils have been extensively 266 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 © developed along the planes of schistosity. (Plate 27, Fig. B.) The general ground-mass of the rock is, as a rule, a light to medium-tinted greenish-gray, silvery, glittering felt of quartz and abundant sericite. Sprinkled through the felt are the round or hexagonal biotite plates, which range from 1 mm. to 3 mm. in diameter. The biotite is highly lustrous, and, on account of its darker colour, stands out prominently on the surface of the rock. This special pseudo- phenocrystic development of biotite is characteristic of the whole belt and, while occasionally seen in narrow bands of belt D, is not an essential feature ef any other than belt H. For this reason it may be called the belt of ‘ spangled schists.’ Along with the biotite spangles there are often many pale-reddish anhedral garnets also developed in the planes of schistosity. The sericite is commonly replaced by well characterized muscovite of the ordinary type, though it never takes on the size of the biotite spangles. Around the large biotites and the garnets the small shreds of sericite and quartz grains are often seen under the microscope to be arranged in concentric layers; this rekation is the familiar one to be observed so often in garnetiferous schists. A little magnetite, a few zirecons and needles of rutile form the accessories of the schist. There are all stages of transition between the typical spangled schist and sheared quartzite, which is always sericitic and commonly speckled with minute dots of dark biotite. These quartzites are similar to those characteristic of both belt D and belt PF. Near the divide between Summit creek and the north fork of Corn creels the spangled schists enclose a band of common amphibolite about one hundred feet in apparent thickness. This is evidently a sheared and highly metamor- phosed basic igneous rock, probably of intrusive origin. A second sill-like intrusion of much altered basic rock (now a hornblende-chlorite schist), with an exposed width of ten feet, was found on the ridge about a mile and a half E. N. E. of North Star mountain. With these exceptions belt H is a fairly homogeneous body of acid, sedimentary rock wholly metamorphosed. The schistosity planes usually strike parallel to the boundaries of the belt as laid down on the map; the dip is always very high, varying from 90° to 75° E. It is apparently more characteristic of this belt than of any of the others that the attitude of the bedding is highly discordant with that of the schistosity. The two planes were often seen, in the same ledge, to cut each other at angles of from 60 to 80 degrees. Unfortunately the exposures were not sufficiently numerous to enable the writer to determine even the main facts concerning the true position of the bedding planes throughout the belt. It is only known that these rocks are often greatly crumpled and that the folds and crinkles are crossed indifferently by the master-structure. Considering the intense metamorphism, the bedding is well preserved and is represented by good contrasts of colour between the lighter tinted, more quartzitic layers and the darker, more micaceous layers once rich in argillaceous material. (Plate OTB): The spangled schists were followed from the Boundary line to the ridge PATE 27. A. Contrast of normal sericite schist of Monk formation (left) and contact-metamorphosed equivalent in aureole of summit granite stock. a coarse-grained, glittering muscovite schist (right). The sericite schist specimen shows dark patches of surface stain. One-half natural size. B. Spangled, garnetiferous schist characteristic of Belt E of Priest River Terrane. Banding represents original belding of a silicious argillite. Three-fourths natural size. 2da—vol. ii—p. 266. REPORT OF THE CHIEF ASTRONOHER 267 SESSIONAL PAPER No. 25a just south of Summit creek. There belt H has already passed into the collar of contact metamorphism belonging to the Bayonne batholith. On the Dewdney trail all trace of the normal spangled schist is lost and the rocks which appear to represent it are relatively very coarse-grained, crinkled muscovite-biotite schist, alternating with micaceous quartzite. So complete is the recrystalliza- tion that it has proved impossible to separate the contact-metamorphosed part of belt # from the similarly altered schists of belt G. For this reason belt # is, in the map, represented as ending in an arbitrary line drawn to indicate the northernmost limit of the schist which actually shows the spangling with biotite. From near the Kootenay river flat to a point four miles up Summit ereek, the coarse, glittering mica schists with their quartzitic intercalations represent the utmost crystallinity and a very striking parallel to typical mica schists in the great pre-Cambrian field of eastern Canada. This spectacular exomorphic collar is more than two miles wide as measured outward from the Bayonne granite. Within the collar the schists are powerfully crumpled and the strike of both schistosity and bedding has been forced around so as to be sensibly parallel to the contact-line of the Bayonne granite. The dip averaged mrO about 75° to the north but is quite variable. Petrography of Belt F.—¥ast of the zone of spangled schist good outcrops are specially rare for several miles. These sections were run across belt F' but, on account of the heavy forest cover, the information was but meagre. The net result of these traverses went to show that the zone is, like belt D, com- posed of sheared quartzite with subordinate interbeds of mica schist. The quartzite is here usually much more schistose than that in belt D and is chiefly a true quartz schist. Sericite or well developed muscovite, biotite, and chlorite, all in minute foils giving by reflexion point-like scintles of light from the planes of schistosity, are the micaceous minerals formed by the dynamic metamorph- ism. The intercalated mica schists are much like those of belts A to D, but almost never show the biotite spangles characteristic of the rocks of belt £. In both the quartz schists and the mica schists there is, close to the Rykert granite, an increase in the size of the mica foils and usually some development of reddish garnets. These features are regarded as due to special contact- metamorphism. A band of garnet-bearing amphibolite, 125 feet wide, and apparently following the bedding-planes of the schistose quartzite near the Rykert granite, is another example of greatly metamorphosed basic intrusives in the Priest River terrane. Peripheral schistosity was developed in the belt by the Rykert granite mtrusion. On the north slope of Boundary creek near the contact, the strike of bedding and schistosity was observed to run from N. 25° E. to N. 45° E., with dips varying from 30° to 45° N. W. On the top of the ridge and a mile from the contact the average strike is about N. 30° W. and the dip varies from 75° E. N. E. to 75° W. S. W. Farther west the dip is northerly and flattens to 20° or less. Toward the western limit of belt #, on the same ridge, the strike is about N. 25° E. and the dip nearly vertical. In all these cases the strike and dip refer to the banding of the 268 ' DEPARTMENT OF THE INTERIOR 2 GEORGE V.,, A. 1912 qaartz schist-mica schist series; this banding seems undoubtedly to represent original bedding. The schistosity is for the most part apparently coincident with it It looks as if the rocks of this belt lying to the west of the Rykert granite form an appressed and greatly crumpled syncline but, in view of the scanty rield data, no great confidence can be felt in this interpretation. Petrography of Belt G—The most easterly of the seven belts is even more obscure as to its detailed structure than the other belts. Belt G lies between the Bayonne and Rykert granite batholiths which have conspired to perfect the metamorphism begun by the crush of earlier mountain-building pressures. Half-way between the two batholiths and from four to five miles from either, the rocks have the peculiar habit of micaceous contact-hornfelses. Intense crumpling of the sedimentaries in the zone has been brought about by a combina- tion of the strong orogenic pressure which has affected all the belts, and of the outward pressures exerted during the forceful intrusion of the batholiths. The structural problem of the belt is further rendered difficult by the rarity of good bed-rock exposures. The belt is essentially composed of glittering coarse to medium-grained mica schists. These vary in colour from light to dark greenish-gray and dark rusty brown. The average phase is distinctly more ferruginous than the staple schists of any of the other six belts. As a rule the schists are well banded, much after the fashion of the spangled schists of belt H. It is believed that the bands represent the true bedding. The original sediments were doubtless chiefly argillites more or less rich in silica, with subordinate thin interbeds of sandstone. Their existing metamorphic equiva- lents are muscovite-biotite schists carrying variable and often important amounts of red garnet, yellow epidote, and tourmaline. The muscovite is some- times sericite but generally occurs in the form of the usual foils of relatively large size. : As already noted, the northern part of the belt along Summit creek includes schists which form the probable extension of belt # into the exomorphie collar of the Bayonne batholith. All across belt G, at the creek, the strike is a little north of east and thus roughly parallel to the contact-line of the batholith. It is possible that similar peripheral schistosity was developed in belt G north of the Rykert granite but this point could not be determined in the time that could be allotted to the area. Elsewhere in belt G the average strike of the banding varies from N. 25° E. to N. 45° E. The dips are exceedingly variable, those observed ranging from 70° N.W., through verticality, to 50° S.E. ~ Thicknesses and Structure in the Priest River Terrane.—With the exception of a few relatively unimportant bands of amphibolite, the whole of the. Priest River terrane is composed of originally sedimentary rocks. The list of these include argillites, argillaceous sandstones, highly silicious sandstones, dolomitie sandstones and argillites, and dolomites. All these rocks are tremendously sheared and metamorphosed, so that not a single ledge observed in the field . REPORT OF THE CHIEF ASTRONOMER 269 SESSIONAL PAPER No. 25a nor a single one of about one hundred specimens, more closely studied in the laboratory, is without abundant signs of crushing or, at least, recrystallization. The carbonate rocks occur in belts A, B, and C, but are chiefly concentrated in belt B. In the section crossing that belt, northwest of North Star mountain, the six great beds of the dolomitic marbles aggregate about 1,500 feet in thick- ness. If the three beds outcropping on the western side of the belt are but duplications of the three beds outcropping on the eastern side, and, if half the mean of the thickness be assumed as indicating the real thickness of the three beds, this would total 750 feet. In belts A and C there must be at least 250 feet of highly magnesian rock additional. The writer believes, in fact, that 1,000 feet represents the minimum thickness of the total dolomitic rock as exposed m this area of the Priest River terrane. Most of belts A and C and a large part of B are composed of rather homo- geneous mica schists, including great’ masses of phyllite and chloritic schist. It is possible that belt B represents a duplication of A; with this assumption a very rough estimate of the minimum total thickness of the argillaceous strata corresponding to these schists is 5,000 feet. The thicknesses of the dominant quartzites of belts D and F’, which are. lithologically very similar, are extremely difficult to estimate but it is believed that at least 6,000 feet of different beds must be represented. The total appar- — ent thickness of the spangled schist is over 6,000 feet. and an estimate of 3,000 feet, based on the possibility that belt H coincides with a simple closed vertical fold, seems to be a safe minimum estimate for the thickness of the spangled schist. Belt G consists of mica schists which in several respects are very similar to the schists of belts A, B and C, yet no dolomites have been found in belt G and it would be unsafe to correlate the strata of belt G with those of any of the western zones. In any case, it appears that at least 3,000 feet of recrystallized strata, not appearing in any of the members estimated above, must be added to complete the total of strata exposed in the area. Tt thus seems likely that this total is, at the minimum, 18,000 feet. Even that estimate is large in absolute measure but the total number of feet is but a relatively small fraction of the apparent thickness of the whole series. Rough as the estimate is, it indicates the fact, amply demonstrated by the field obser- vations, that this old series is of great thickness even when compared with the more certain minimum totals for the neighbouring Summit and Purcell series. At one stage in the work of interpreting the terrane it was postulated that at least part of belts D,'H,F,and G really form part of the Cambrian-Beltian series, being thus equivalent to the quartzitic and argillaceous phases of the Summit and Purcell groups. A careful study of the field data and of the col- ~ lection of specimens has, however, led the writer to believe that this supposition is inadmissible. Quite apart from the thermal action of the Rykert and Bayonne batholiths, the whole Priest River terrane is intensely metamorphosed, to a degree never seen in the Purcell series and only rarely, and then but locally, observed in the Summit series. Moreover, the detailed composition of none of the belts agrees with any similarly thick portion of the Summit or Purcell 270 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 series. Lastly, it may be noted, as good evidence, that the huge basal conglo- merate of the Summit series contains myriads of pebbles manifestly derived from ledges quite similar in composition to those of belts A, B, OC, D, F, and G. Since the strikes and dips of the Irene conglomerate are nearly or quite parallel to those in belt A of the older terrane, one might doubt the existence of the unconformity at the base of the conglomerate, were it not especially for the similarity of the dolomitic pebbles in the conglomerate to the dolomitic bands in the Priest River series. Largely for this reason a pre-Beltian age is ascribed to all the schistose rocks (not intrusive) situated, within the Boundary belt, between the Irene conglomerate and the down-faulted Kitchener quartzite at the western edge of the Kootenay River alluvium. This great Priest River group presents a structural problem as yet quite unsolved. Correlation.—lt is, of course, too early to attempt a fixed correlation of the Priest River terrane with the other pre-Cambrian terranes of the Cordillera, but it is not without interest to observe that in various regions there are thick masses of ancient sedimentaries which appear to correspond both lithologically and in stratigraphic relations to the Priest River terrane as exposed along the Boundary line. A few references to typical sections in the Belt mountains of Montana, the Black Hills of South Dakota and adjoining portions of Wyom- ing, the Fortieth Parallel region, and the sections worked out by Dawson on the main line of the Canadian Pacific railway, may be useful as showing the places where possible equivalents of the Priest River terrane may be sought. In the Three Forks, Montana, folio of the United States Geological Survey (1896) Peale describes the ‘Cherry Creek beds’ as a series of mica-schists, quartzites, gneisses, and marbles or crystalline limestones. These beds are highly inclined, apparently conformable to one another, and, notwithstanding the obscurity of the folding, are known to total thousands of feet in thickness (at least 7,000 feet shown in columnar section). The series is lying ‘ probably’ unconformably upon ‘ Archean gneisses’ and is unconformably underlain by the Belt terrane, 7.e., by equivalents of the lower members 0i the Rocky Mountain Geosynclinal prism as just described in this report. In the Hartville, Wyoming, folio (1903), W. S. T. Smith and N. H. Darton describe, under the name of the ‘Whalen group’ a series of schists, gneisses, quartzites, and limestones, which are said to resemble closely the ‘ Algonkian’ rocks of the Black Hills. These rocks have high. or even vertical dips. They appear to resemble also the pre-Cambrian schists of the area covered by the Sundance, Wyoming, folio, which are unconformably overlain by the Middle Cambrian Deadwood formation. The Algonkian rocks of the Black Hills have not been adequately described but include garnetiferous and*other mica schists, graphitic schist, ferruginous quartzite and amphibolite.* These meta- morphosed rocks, with high dips, lie unconformably beneath the Middle Cam- brian overlapping strata. *T. A. Jaggar, jr., Prof. Paper No. 26, U.S. Geol. Survey, 1904, 34. See also Newton and Jenney’s Report on the Geology and Rescuress of the Black Hills of Dakota, Washington, 1880, p. i ee REPORT OF THE CHIEF ASTRONOMER al SESSIONAL PAPER No. 25a MacDonald mentions an important group of metamorphosed and highly crystalline sediments, now schists, outcropping along the west shore of Cour d’Alene lake.g This locality is about 120 miles due south of the area of the Priest River terrane as mapped for the present report. It seems possible that the one terrane is a continuation of the other. King recognized a greatly deformed series of slates, quartzites, limestones. dolomites, mica schists, and hornblende schists in the ‘Archean’ division of the rocks encountered during the Fortieth Parallel survey.t Farther south the quartzites and micaceous schists of the Vishnu group in the Grand Canyon section represent other pre-Cambrian sediments which have suffered, apparently, about the same measure of deformation and metamorphism as those characterizing the Priest River terrane. In British Columbia, north of the Boundary belt, it is fully as difficult as in the cases already noted, to correlate with confidence. Among the des- eribed rock-groups, the nearest. approach, lithologically, to the Priest River terrane is the Nisconlith series of Dawson, as exposed around the Shuswap lakes.t This series is made up of calcareous or graphitic mica schists, flaggy. often dark-coloured limestones, gray and blackish quartzites in apparent con- formity. The series appears to lie conformably beneath the Adams Lake series and both are placed in the Cambrian, the Nisconlith overlying the truly Archean Shuswap series of gneisses, ete. All three series are quite unfossili- ferous and the present writer suspects that the correlation of the Nisconlith with the Priest River terrane is at least as justifiable as that with the Cam- brian of the Front ranges. The foregoing brief statement of the constitution and relations of the various groups indicates lines of thought in the future correlation of the ancient formations of the Cordillera, rather than any definite view as to the correlation. One thing is certain, however; the Cordillera is at many points underlain by very thick and important groups of sediments which are not only pre-Cambrian but also pre-Beltian in age. It is possible if not, indeed, probable that the total thicknesses of these stratified rocks rival those of the pre-Cambrian terranes in the Great Lakes region of Canada and the United States, as wel] as those of the vast formations of Finland. PEND D’OREILLE GROUP. General Description.—Between the western limit of the Summit series monocline and the southeastern edge of the great central volcanic field, an area of about sixty square miles of the ten-mile belt is underlain by a thick group of unfossiliferous, heavily metamorphosed sediments. A considerable §D. F. MacDonald, Bull. 285, U.S. Geol. Survey, 1906, p. 42. + Report, Vol. 1, Systematic Geology, 1878, p. 532 tG. M. Dawson, Explanatory notes to Shuswap sheet, Geological Survey of Can- ada, 1898. For further references see Bull. Geol. Soc. America, Vol. 12, 1901, p. 66. Since this report went to press, the writer has proved the pre-Beltian age of the Niscon- lith of the Shuswap district. 272 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 part of the season of 1902 was spent in their study but the results were, in many essential respects, very meagre. These rocks occur in one of the Cordil- Jeran zones of maximum orogenic shearing and mashing, with complete recrysta!lization. Numberless crumplings, overturnings and _ faultings characterize the region, which, as already noted, has been the scene of repeated igneous injections in the form of dikes, sills, stocks, and batholiths. Again and again the region has been buried deeply in volcanic ejectamenta. In spite of prolonged erosion these volcanies still cover hundreds of square miles and conceal many desired facts concerning the sedimentary rocks. To such p» .- cipal difficulties in-analyzing the complex assemblage of strata along the Pend D’Oreille river there was added that common disadvantage of the geologist — on the Forty-ninth Parallel, the dense evergreen forest with its deep mat of brush and fa!len timber, (Plate 28.) At the time of the writer’s exploration, the Commission trail on the south side of the Pend D’Oreille river, had not been cut. The crossing of this dangerous river akove Waneta was effected only once; hence relatively little is known of the rocks and structures on the left bank of the river. In that part of the belt, outcrops of rock are relatively few. Attention was therefore concentrated on the strip of altered sediments lying between the river and the Rossland-Voleanic terrane to the north. The ancient Priest River rocks themselves are scarcely more baffling in structural analysis than are these much younger schists along the Pend D’Oreille. Their clean-cut mapping, their order of superposition and the determination of thickness could not be thoroughly worked out. Nearly all that is possible, as a result of the reconnaissance in 1902, is to give a general qualitative description of the metamorphosed sediments. They are conveniently referred to in the present report, under the name, Pend D’Or.Alle Group; the wild canyon of the Pend D’Oreille river in the lower twenty miles of its course has been excavated in the rocks of this group. Their distribution in the Boundary belt is shown on the map though not with entire accuracy, for it is extremely difficult if not impossible with existing exposures, to separate, in several areas, the rocks of the group from the younger members of the Summit series or from the old, schistose phases of the Rossland voleanics, The group may be divided into two parts, the Pend D’Oreille schists (including greenstone and amphibolite, as well as phyllite and quartzite), and the Pend D’Oreille marbles. They are primarily not strati- graphic subdivisions so much as purely lithological ones. It was “7nd imprac- ticable to use the limestones as definite horizon-markers and equally impossible to be sure of the relative ages of the limestones and their non-calcareous associates. Several of the larger bodies of limestone seem to form gigantic, isolated pods which have been squeezed, like a truly plastic substance, through the schists, to accumulate locally and with exaggerated thickness in these great masses. On this view the limestone pods are, in part, exotic—they might be ealled non-igneous intrusives—with reference to the enclosing schists. In any case, it has appeared unsafe to use the few legible records of original MALY 9[[L9IO,. pues 9G} PABMOZ YoITP) O[LUM- W994 FT uMOp SULOO'T 97 DLV UAySAS YALYLIS oF Jo sSULvIUNOTAT I [12919,.q purd— uojSUIUUOd ul 9 Mota [wodAT, REPORT OF THE CHIEF ASTRONOMER 273 SESSIONAL- PAPER No. 25a bedding in schist or marble, as indicating ithe real stratigraphic relations in any detail. A columnar section is as yet impossible. The brief description of a few typical traverses may suffice to show the general characters of the rocks. Area East of Salmon River.—One of the most continuous exposures of the group was found on the top of the broad ridge running westward from Lone Star mountain. East of that peak the schist is in contact with the Lone Star formation, but, as indicated above, the relation between the two formations is very cbscure. A special reason for the uncertainty as to the true relation is found in the existence of the wide break in the section, caused by the intrusion of the Lost Creek granite. ; From the bottom of the col between Beehive and Lone Star mountains westward to the Salmon river, the dominant rock is a typical carbonaceous phyllite. It is a very dark gray or greenish-gray to black rock, highly schistose, and generally with few certain traces of the original bedding. For hundreds of yards together along the ridge this greatly crumpled schist shows marked homogeneity, but, in places, it passes into an abundant schistose, likewise car- bonaceous quartzite. Both these phases may be caleareous and carry accessory tremolite and epidote as metamorphic products, along with the quartz, sericite, and carbon dust. The schists are often pyritized to some extent and in many parts, bear numerous veins of mineralized quartz. Biotite is very often developed as an abundant accessory constituent of the schists. Strain-slip schistosity with the resulting crinkly rock-surface is well developed at many points. On the top of Lone Star mountain a pod of banded, white and bluish marble is intercalated in the phyllite-quartzite. The limestone is enormously crumpled and mashed, so that it is impossible to determine its thickness. Its average dip seems to be 30° to the east. A mile west of Lone Star peak a much larger intercalation of banded, dark gray and bluish-white marble crosses the ridge. It can be followed continuously on a band of fairly constant width from Lost creek to, and beyond, Sheep creek on the north. The continuity suggests that this band represents a sedimentary member which retains nearly its original thickness and has not been seriously thinned or thickened by orogenic shearing. In this view the thickness must be at least 2,000 feet, for the true dip is 70° and is against the mountain slope, while the width of the band is nearly half a mile. This limestone, like all the others found in the area now described, is a true marble, fine-grained and completely recrystallized. None of the limestones of the group seems to be magnesian to any great extent; all the specimens col- lected effervesce violently with cold, dilute acid. Occasionally flattened grains of quartz appear in thin sections and, more rarely, minute crystals of basic plagioclase, probably anorthite, lie scattered through the thin section. . Chert nodules or beds were never seen in any of the marbles. In one bed on the south side of Salmon river, concretions of finely granular quartz of the size of large peas, are embedded in calcite. Excepting these accidental ingredients the marbles are to be regarded as composed of notably pure cx'cium carbonate. 25a—vol. ii—18 274 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 A strong cataclastic structure was microscopically observed to be a general feature of the marble. It is impossible effectively to distinguish the true stratigraphic positions of all the marble bands in the area, or to be sure of their correlation among themselves. They are, therefore, mapped under the common name, ‘ Pend D’Oreille limestone.’ Between the mapped monzonite stock and the Salmon river flat the quartz- ite-phyllite bears one or more strong intercalations of amphibolite, composed of dark olive-green hornblende, quartz, and highly granulated residual individuals ot basic plagioclase. Other interealations of amphibolite and hornblende schist were observed on Lost creek just above its confluence with Lime creek, and on the Salmon river below Roseleaf creek. Throughout the four-mile section the dips of the schistosity planes are generally high (40° to 75°) to the eastward, though they are, of course, often reversed in the numerous crumples affecting the schists. The banding of the limestones and their planes of contact with the schist were usually seen to dip eastward at similarly high angles. The attitude of the bedding-planes cannot be taken as directly indicating the succession from older to younger in this sedimentary monocline; there is every possibility that the whole group has been overturned along with the apparently conformable Lone Star and Beehive. members of the Summit series. In favour of this conception is the fact that a massive limestone of great thickness, of similar lithological characters, and lying nearly flat, overlies a thick series of phyllitic and quartzitic rocks between Roseleaf creek and the Pend D’Oreille river. This limestone covers at least five square miles and dips from 10° to 30° south; it is highly improbable that so large a mass has been overturned. The underlying schists are in the main like those exposed in Lone Star mountain. The tentative conclusion has thus been reached that the schistose rocks composing the Lone Star section from the western contact of the Lone Star schist to the eastern contact of the great limestone band all underlie that limestone and, with it, have been over- thrown so as now apparently to overlie the limestone. On the same tentative basis these older schistose sediments may be set down as totalling at least 3,500 feet in thickness. The large body of marble situated at the confluence of Lest creek and the south fork of the Salmon river is probably the down faulted equivalent of the 2,.000-foot. band of limestone above described. If so, the phyllites and quartzites lying to the westward of that band may be wholly or in part of the same age as the schists lying to the eastward of the band. In this Lone Star-Salmon river section, therefore, one cannot be sure that there are any sediments younger than the great limestone. Unfortunately, no other area in the Boun- dary belt has afforded any more certain help in carrying the stratigraphic suecession higher or completing the columnar section for this region. It is probable that the micaceous schists exposed in Sheep Creek valley for three miles from its intersection with Salmon river, are younger than the great limestone, but the exposures are much too imperfect to warrant a definite conclusion on the point. REPORT OF THE CHIEF ASTRONOMER 275 SESSIONAL PAPER No. 25a Area West of Salmon River—Dark greenish, or dark gray to black phyllite, alternating with blackish quartzite, is the dominant rock on both banks of the Pend D’Oreille, from its confluence with the Salmon to its mouth at Waneta. The schists enclose lenses of white to gray marble, varying from ten feet or less to 200 feet or more in thickness. Near the Columbia and especially on the west side of that river, the phyllites and quartzites are associated with very abundant, thick masses of greenstone and altered basic breccias, so that it there becomes very difficult to separate the Pend D’Oreille group from the younger Rossland Voleanie group. Lithologically, there is a great similarity between the respectively dominant rock types on both sides of the Salmon but it is also clearly impossible to develop a useful columnar section of these metamorphosed sediments along the lower Pend D’Oreille. The great limestone is not represented. It is, however, probable that most of the phyllite and quartzite is the equivalent of the rocks tentatively regarded as stratigraphically underlying the great limestone on the Lone Star ridge. They are unconformably overlain by the Rossland lavas as developed east of Sayward. Between Nine-Mile and Twelve-Mile creeks a strong and’ persistent band of silicious limestone is intercalated in the phyl- lites; it is truncated at each end by the overlapping lavas in such a way as to illustrate the unconformity. (See map.) The structure of this area is fully as complex as that east of the Salmon. The schists are well exposed for miles in the canyon of the Pend D’Oreille, where the dips and strikes of the schistosity planes were seen to shift every few hundred feet. On the average the strike runs a little north of east, so that the river section is not favourable to the discernment of the field relations. Numerous acid and basic intrusions have also affected the structural relations. As a negative result of the field work among the schists it may be stated that the leading problems regarding their age, their subdivision into recogniz- able members, and their thickness must apparently be solved outside the ten- mile belt. It is most probable that, if ever found, the key to these secrets will be disclosed on the United States side of the Boundary line. In the present report the whole assemblage of phyllites, quartzites, traps, and limestones is included under the name, Pend D’Oreille group. Its minimum thickness is believed to be 5,500 feet. Correlation.—The marbles and schists themselves carry no fossils, so far as known, but a hint as to their age is found in the fact that lithologically similar marbles bearing a Carboniferous species were found by McConnell and -by the writer in the Rossland district.* In central Idaho, eighty to one hundred miles to the southeast of the Boundary section at the Pend D’Oreille, Lindgren has found closely allied rocks at several, rather widely separated localities, and at most of them some of the rocks bear Carboniferous fossils.t His description * Cf. R. G. McConnell. Explanatory notes to Trail sheet issued by the Geological Survey of Canada, 1897. 7 W. Lindgren, 20th. Annual Report, U.S. Geol. Survey, Part 3, pp. 86-96, 1900. 25a—vol. 1i—184 276 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 of the Wood River series in his report will be found to match fairly well with the account of the Pend D’Oreille group just given. About one hundred miles to the northward and north-northwestward of the Boundary section at the Pend D’Oreille are considerable areas of stratified rocks referred by Brock to the Cache Creek series or to the Slocan series which he regards as probably equivalent to the Cache Creek.t In that region the Cache Creek series is made up of ‘dark argillites, greywackes, quartzites and limestone, with some eruptive material’; the description of the Slocan series is in similar terms. In the Kamloops district, still farther northwestward, the Cache Creek beds are well exposed and there they have been studied in some detail by Dawson. His summary statement of their succession is as follows :— ‘The lower division consists of argillites, generally as slates or schists, cherty quartzites or hornstones, volcanic materials with serpentine and interstratified limestones. The volcanic materials are most abundant in the upper part of this division, largely constituting it. The minimum volume of the strata of this division is about 6,500 feet. The upper division, or Marble Canyon limestones, consists almost entirely of massive limestones, but with occasional intercalations of rocks similar to those characterizing the lower part. Its volume is about 3,000 feet. ‘The total thickness of the group in this region would therefore be about 9,500 feet, and this is regarded as a minimum. The argillites are generally dark, often black, and the so-called cherty quartzites are probably often silicified argillites. The voleanic members are usually much decom- posed diabases or diabase-porphyrites, both effusive and fragmental, and have frequently been rendered more or less schistose by pressure.’* Much of the Cache Creek series is fossiliferous and definitely Carboni- ferous (Pennsylvanian), but Dawson points out that the lower beds may include formations somewhat older than the Carboniferous. He emphasizes, after many years of experience, the great constancy of the series from the Yukon boundary of British Columbia southward throughout the length of British Columbia. In view of these various facts it seems to be the best working hypothesis that these greatly metamorphosed rocks of the Pend D’Oreille group roughly correspond to the Cache Creek series and that they are in large part of Car- boniferous age. The lower schists may include sediments of any age from the Carboniferous to the Silurian inclusive. There is no evidence of unconformity with the Summit series; the Pend D’Oreille schists seem, on the other hand, to pass gradually into the underlying Lone Star schists. Because of the special local intensity of thermal and dynamic metamorphism it must be long before the correlation of the Pend D’Oreille group is anything other than hypothesis. Yet, as in so many cases, the tentative correlation seems to be better than none, IR. _W. Brock, Explanatory’ notes to West Kootenay sheet, issued by the Geo- logical Survey of Canada, 1902. *“@G@. M. Dawson, Bull. Geol. Soc. America, Vol. 12, 1901, p. 70. “OUOT JOOF OMY JHOQV Lop[NO 63 MLV IG OALY, OT[L9IO,q pues Jo poq Ul Japlnog aqizjzawnb uo syavur Woissnoaeg, iO, .li—p. 27 vol 25a REPORT OF THE CHIEF ASTRONOMER 2 277 SESSIONAL PAPER No. 25a for even in its upsetting, the future observer’s eye will be sharpened for the essential facts. SUMMARY ON THE STRUCTURE OF THE NELSON RANGE. The structural geology of the Nelson range where crossed by the ten-mile belt naturally involves a study of three different types of areal geology, corres- ponding respectively to the Priest River terrane, the rocks of the Summit series, and the large bodies of batholithic granite. The obscurity of relations among the old sediments of the Priest River terrane has been described in the account of the different zones (belts) of the terrane. Schistosity and bedding often coincide. Both sets of planes are highly inclined, with dips averaging about 75° to the eastward. Quite vertical dips are very common in the southern half of the belt. In the northern half the Priest River rocks have been intensely crumpled by the intrusion of the Bayonne batholith, giving loeal dips at all angles and in all directions, with average northwesterly to north-northwesterly dips of about 70°. The original dips due to tangential pressure have likewise: been greatly modified by the intrusion of the Rykert granite batholith. The failure to find recognizable folds in the terrane has already been sufficiently noted. South of Summit creek, zones A, B and C have been affected by a strong horizontal shift (a fault in which there has been horizontal movement of one block past the other). At the creek the three zones appear to be cut off entirely by a fault which is entered on the map. A less important break cuts off zone B near the Boundary line. With these exceptions the writer has failed to find structural elements which can be definitely mapped. On Map No. 6 a long band of Kitchener quartzite is shown along the western edge of the Kootenay river delta between the Rykert mountain granite and the mouth of Summit Creek canyon. The quartzite is referred to the Kitchener formation on lithological grounds and there are many points of resemblance to the Beehive quartzite. The microscope shows that microper- thite is a relatively abundant constituent of all three quartzites while the felds- pathic material of the quartzites belonging to the Priest River terrane is quite different. In other respects also this quartzite along the river alluvium corresponds well with the Kitchener formation in essential features. Though the brushy slopes to the westward have not been thoroughly explored it appears safe to postulate a great north-northwest fault on which these Kitchener beds have been dropped down into contact with the Priest River terrane. This fault is shown on the map. Its exact course is represented only approximately ; further field-work is imperative before greater precision may be attained. The fact remains, however, that this quartzite, which has thus been correlated with the Kitchener and the equivalent Beehive quartzite, has been downthrown through a vertical distance equal to the whole thickness of the Summit series below the Beehive formation plus an unknown thickness of the Priest River terrane. The downthrow may measure 20,000 to 20,000 -feet. 278 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 THRUST 0 2 Miles Horizontal and Vertical Scales. Figure 17.—Kast-west section on ridge north of Lost Creek, Nelson Range ; showing duplication of beds by thrust, the plane of which has been rotated. Figure 18.—Diagram showing stage of development of the thrust illustrated in Figure 17. REPORT OF THE CHIEF ASTRONOMER 279 SESSIONAL PAPER No. 25a On both sides of the Purcell Trench, therefore, we have evidence of huge displacements which have given this part of the Kootenay valley the character of a fault-trough. The down-faulted block or blocks have, of course, lost much substance through erosion but it seems most probable that the trench was located in a constructional depression due to faulting. Another of the primary structural features of the Nelson range is the unconformity at the base of the Irene conglomerate. The existence of the unconformity is not conspicuously shown by contrasts of attitude between the conglomerate and the older sediments. In fact, as above noted, the strike and dip of the conglomerate and of zone A are often closely similar. The evidence is more fully derived from (1) the much stronger metamorphism of the Priest River terrane; (2) the abundant pebbles of Priest River rocks in the conglo- merate; and (3) the truncation of zones A, B and C by the lower-contact plane of the conglomerate. The Nelson range covers the only part of the Boundary belt where the Rocky Mountain Geosynelinal is sounded to its full depth. Within the great Summit series monocline itself one of the most conspi- cuous structural complications is the horizontal shift mapped as crossing the divide between Monk creek and the south fork of the Salmon river. In the field the effects of the shift are spectacularly clear. The almost vertical forma- tions have been dislocated by a movement of about a mile along the vertical west-northwest plane of shifting. The relative displacement is that which would have been produced if the southern block had moved westward through that distance. The outcrop of the shift-plane could be readily followed for four miles; its continuation westward across the southern slope of Lost moun- tain is less evident in the field but seems competent to explain the relations of the Pend D’Oreille limestones and schists to the quartzites on Lost mountain. A second horizontal shift, not so evident, is mapped just.south of Summit creek. About three miles west of the main divide of the range the upper beds of the Summit series are duplicated for a great thickness by a powerful thrust. This thrust is among the most remarkable elements in the anatomy of the range. (Figures 17 and 18.) The plane of the thrust is stratigraphically located in or very near the 225-foot band of conglomerate in the Dewdney formation. The conglomerate has apparently acted as a local plane of weakness. Along that plane the entire overlying part of the Summit series has been driven eastward and has then been pushed up on the back of the Lone Star schists. Either during the thrusting, or, less probably, afterwards, the overridden and overriding blocks together with the thrust-plane have been rotated so as now to stand almost perfectly vertical or to show a slight overturning to the westward. As a result the observer traversing the ridges on either side of Summit creek will, on going westward, pass over the Dewdney beds, then the Ripple, Beehive, and Lone Star in regular order, and will then, after crossing the thrust-plane, pass over the upper Dewdney, the Ripple, Beehive, and Lone Star formations once more. These relations are illustrated in Map 7 and in Figure 17. They are specially clear on the high, nearly treeless ridges north and south of Summit creek. The extreme northern and southern extensions of the thrust-plane are not so well exposed and the mapping is there somewhat tentative. 280 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 It is scarcely necessary to remark that the straightness of the bands of colour corresponding on the map to the Summit series formations, is controlled by a structural necessity, namely, the nearly or quite vertical dip which is general throughout the greater part of the monocline. The thrust-plane just described must similarly be nearly vertical. Deep as the canyons are, the out- crops of the different formations deviate but little from the straight line where the bands cross the canyons. West of Beehive mountain the Pend D’Oreille series is so thoroughly disordered that the structure section in this part could be represented only in a schematic way. The same procedure is necessary for the continuation of these rocks across the Salmon river. Finally, in the Nelson range section we find the outposts of the army of granitic intrusives which cut the stratified rocks of the Cordillera at intervals all the way from the Purcell Trench to the Pacific ocean. In general the sediments of that greater part of the Boundary belt are much younger than the rocks of the Rocky Mountain Geosynclinal; but, because of the inherent weakness of the younger sediments, because of the intrusion of many batholiths, and probably also because of a greater intensity of the orogenic forces in the western half of the Cordillera, these sediments are generally more metamorphosed than those of the older prism. Just east of the Salmon river the Summit series plunges under the Pend D’Oreille group of schists and marbles and the still younger Rossland volcanics, never to reappear in the sections farther west. The Nelson range is the greatly worn product of the mightiest crustal upturning on the Forty-ninth Parallel; beside the range is the Purcell Trench, the eroded representative of one of the deepest structural depressions of the Cordillera. 2 GEORGE V, SESSIONAL PAPER No. 25a A. 1912 CHAPTER XII. INTRUSIVE ROCKS OF THE SELKIRK MOUNTAIN SYSTEM. From the Great Plains to the Purcell Trench the igneous-rock geology of the Boundary section shows relative simplicity. It has centred principally around the discussion of the Purcell Lava formation and the basie sills and dikes of the Purcell mountain system. Crossing the trench westward we enter a region where igneous rocks become areally important and, because of their petrographical variety and complicated relations, deserve considerable attention. All the rest of the Boundary belt, from the Purcell Trench to the Fraser flats at the Pacific may be described as an igneous-rock field. It is not always possible to treat of the many intrusive and extrusive bodies in groups corresponding to the various mountain ranges crossed by that long belt. In somes cases the igneous-rock bodies are crossed by the master valleys which have been taken as the convenient lines of separation between the ranges. This is true cf several of the igneous- rock units which, in part, make up the Selkirk system at the Forty-ninth Parallel. It happens that the larger areas covered by these bodies occur in the adjacent Rossland mountain group of the Columbia system, and it is appropriate to discuss such areas in the following chapter devoted to the geology of the Rossland moun- tains. In that chapter will, then, be described the formations which have been mapped under the names ‘ Rossland Voleanic group,’ ‘ Beaver Mountain Volcanic group, ‘Trail batholith” ‘Sheppard granite, and ‘ Porphyritic olivine syenite.’ (Maps No. 7 and 8.) In the present chapter there will be noted in some detail two granitic bodies, named the Rykert and Bayonne batholiths; several stocks which appear to be satellitic to the Bayonne batholith; a sill or dike of very abnormal hornblende granite which cuts the Kitchener quartzite near Corn creek; sills and dikes of metamorphosed basic intrusives cutting the Priest River terrane; numerous lamprophyric dikes and sills and other basic intrusions, together with a few acid dikes and sills cutting the younger sedimentary formations as well as the Priest River terrane; and a boss of monzonite near the main fork of the Salmon river. No attempt will be made to describe these bodies rigidly in their order of age or geographical grrangement, though the usual procedure of taking them up in the order from east to west will be followed. The difficult problem of their succes- sion in geological time will be discussed in a following section. The Irene Voleanic formation has already been described in its natural place as a member of the Summit series. Further reference to it is unnecessary except in the general summary on the igneous rocks of the Selkirks. 281 282 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 METAMORPHOSED Basic INTRUSIVES IN THE Prizst River TERRANE. Various belts of the Priest River sedimentavies enclose narrow dikes and sills of basic igneous rocks and one or two basic bodies of larger size. With few exceptions these are poorly exposed and the intrusives are enormously altered. It is, therefore, impossible to give a satisfactory account of the intrusives either as to the field relations of several of the bodies or as to the original nature of the magmas whence they have been derived. The largest of the bodies outcrops for a distance of several hundred yards on the trail running from Boundary lake to Summit creek and at a distance of about 2,000 yards west of the top of North Star mountain. This body is at least 500 feet broad. Whether it is a great dike or sill or an irregular intrusion could not be determined. The rock is a dark green, fine-grained, highly schistose trap. Labradorite in small broken individuals is apparently the only primary mineral remaining after the profound metamorphism that the rock has undergone. Most of it is now a mass of chlorite, uralite, epidote, secondary quartz, leucoxene, and pyrite. The original structure seems to have been the hypidiomorphic-granular. The rock was doubtless a gabbro, now: altered to a chlorite-uralite-labradorite schist or greenstone. A ten-foot sill-like intrusion of a somewhat similar rock was found in belt FE where it crosses the main fork of Corn creek. Just below the lower contact of the Irene conglomerate on Summit creek, belt A of the Priest River terrane is cut by a relatively uncrushed hornbliendite, occurring as a sill three feet in thickness. The essential amphibole has nearly _ the same optical properties as the hornblende of the Purcell sill gabbros. Feld- spar is absent. Magnetite and apatite are the observed accessories. Chlorite, quartz, and a little carbonate are secondary products. A larger sill-like body, at least 100 feet thick, cuts the quartzites of belt D at the junction of the North Fork and main fork of Summit creek. This rock bears much quartz, orthoclase, end some indeterminable plagioclase, along with the dominant green hornblende. Its composition and habit recall the acidified hornblende gabbro of the Purcell sills. A quarter of a mile from the Rykert granite contact the schists of belt F’ are cut by a 125-foot sill of originally basic igneous rock which is now a dark green amphibolite, composed essentially of green hornblende, quartz, and basic plagio- clase (labradorite to bytownite) along with much accessory orthoclase and red garnet. This sill has been squeezed to a highly schistose condition and thor- oughly metamorphosed during the intrusion of the ‘Rykert granite. Beyond the fact that these intrusives cut the Priest River sedimentaries, there is little direct evidence as to their age. The thoroughness of their dynamic metamorphism indicates a pre-Tertiary age, while the lithological similarity of the gabbroid bodies to the gabbro of the Purcell sills suggests the possibility that the former may also be as old as the Middle Cambrian, to which the Purcell sills have been tentatively referred. Some of these intrusives may represent the deep- seated phase of the yet older Irene volcanics. In any case the impression won in the field was that the chlorite-uralite schist, the amphibolite, and the sheared hornblende gabbro are of much older date than any other of the igneous bodies REPORT OF THE CHIEF ASTRONOMER 283 SESSIONAL PAPER No. 25a occurring in this part of the Selkirks. The isolated peridotitic sill, hornblendite, may be of the same general date as the schistose derivatives of the gabbro or may be younger. ABNORMAL GRANITE INTRUSIVE INTO THE KITCHENER QUARTZITE. At the edge of the Kootenay river alluvial flat and 2,000 yards south of Corn creek, the down-faulted Kitchener quartzite is cut by a peculiar granular rack exposed in the form of a band about 600 feet wide and elongated in the strike of the invaded quartzite. The igneous mass seems to be in sill-relation to the sedimentaries, although the exposures are not sufficient to cause certainty on that point. The dip of the adjacent quartzite is 60° to the southeastward ; if the intrusive body is a sill its thickness is nearly 500 feet. The igneous rock is dark bluish-gray, medtum-grained, and has the abit of a quartz diorite. In the hand-specimen idiomorphic, lustrous black prisms of hornblende up to 5 mm. in length are very abundant; these are often arranged with a rough fluidal alignment. Quartz is easily recognized as a dominant constituent; feldspar is as clearly subordinate. Under the microscope the rock is seen to be very fresh, though slightly strained, with possibly some granulation in places. The observed amount of deformation is not sufficient to explain the rough parallelism of the hornblende prisms, which is apparently a primary feature established during the crystal- lization of the magma. The essential and accessory constituents are here listed in their order of quantitative importance (by weight) as determined by the Rosiwal method :— Quartz ac eee 41-3 Hornblende.. .. sa 33-4. Orthoclase sane eeee 19-2 Garnetiee- ene 2-8 Magnetite.. ... 2-1 ibpidotess .) 6 Apatite.. .. Z “5 TATCOM eRe il 100-0 The hornblende is highly pleochroic, with unusually beautiful tints :— a—Light yellowish green. b—Very deep olive green. c—Bottle green with pronounced bluish tinge. Absorption very strong: b>c>a. The extinction on (010) is about 11° 15’; that on (110), about 13°, as average of eight measurements on cleavage pieces. Etch-figures on (110) show that ¢ lies in the obtuse angle 8 in Tschermak’s orientation of amphibole, and also that the hornblende is rich in alumina. The hornblende is quite idiomor- phic in the prismatic zone but the prisms are seldom, if ever, terminated by crystal faces. They lie in a mesostasis of quartz and feldspar and have suffered 284 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 somewhat by resorption carried on by this acid matrix in the late mag- matic period. The hornblende seems also to be truly poikilitic, through the inclusion of minute droplets or microlites of quartz ahd feldspar. The feldspar is either a sodiferous orthoclase or its chemical equivalent, a poorly developed microperthite. Not a certain trace of soda-lime feldspar could be seen. The surprisingly abundant quartz occurs as glassy-clear, granular aggregates. The garnet is, in thin section, of a very pale pink colour and is probably a common iron-lime variety. The garnet is idiomorphic against the hornblende. The order of erystal- lization seems to be: zircon, apatite, and magnetite; followed by garnet; then, in order, hornblende, orthoclase-microperthite, and quartz. Calculation shows that the rock must carry about 68 per cent of silica, not more than 8 or 9 per cent of alumina, and not more than about 5 per cent of alkalies. The specific gravity of a typical hand-specimen is 2-894. The presence of essential quartz and orthoclase would place this rock among the granites, but it is clearly an aberrant type in that family. It may be questioned that it is advisable to risk overweighting the granite family by including this rock within it, but no other place is offered to it in the prevail- ing Mode classification. Its abnormal composition may be due to some assim- ilation of the quartzite. There are many points of similarity between this rock and certain phases of the Purcell sills ‘across the river and it is quite possible that the abnormal granite is the result of the solution of the quartzite in an original hornblende gabbro magma. The quartzite is here very poor in feldspathic and micaceous constituents; hence, possibly, the absence of biotite, which is so universal a constituent in the acidified phases of the Purcell sills. This abnormal hornblende granite is tentatively correlated with the Purcell sills. Though little more crushed than those sills, it may also be possible to credit a correlation with the sheared basic intrusives found in the Priest River terrane; for the deformation of the latter must have taken place at a depth several miles greater than that at which the intrusives cutting the much younger Kitchener formation began to feel the post-Paleozoic orogenic stresses. The higher temperatures and pressures of the more deeply buried massive rocks at the time of deformation would seem to be amply sufficient to explain such differential metamorphic effect. RYKERT GRANITE BATHOLITH. This granite, as shown on the map, covers some fifteen square miles of the Boundary belt north of the line; it extends in a broad band southward for an unknown distance into Washington and Idaho and the whole body is, doubt- less, of batholithic size. It has intrusive relations to the Priest River terrane, as shown by numerous apophyses, and by the development of a metamorphic aureole about the granite. On the eastern side of the batholith the Kootenay river alluvium conceals the bed-rock relations, but the granite is probably there PLATE 20. Sheared phase of the Rykert granite, showing concentration of the femic elements of tle rock (middle zone). Natural size. Massive phase of the Rykert granite, showing large phenocrysts of alkaline feldspar.) 25a—vol. ii—p. 284, REPORT OF THE CHIEF ASTRONOMER 285 SESSIONAL PAPER No. 25a in contact with the Kitchener (7?) quartzite which have been faulted down against it. This faulting is believed to be of later date than the intrusion of the granite; no apophyses were found in the quartzite. Lithologically and structurally the batholith is unique in the whole Boundary belt, although in both respects this granite is paralleled by many intru- sive bodies both in Idaho and in British Columbia. The rock is distinguished by a very coarse grain and commonly by an unusually perfect gneissic struc- ture due to crush-metamorphism. (Plate 30.) The colour is a light gray to a light pinkish-gray. In the ledge and hand specimen the most conspicuous ele- ments are large phenocrysts of alkaline feldspar, and, less commonly, of acid plagioclase; these are embedded in a coarse matrix of quartz, feldspar, and biotite. The phenocrysts range from 2 cm. to 8 cm. in length. In the less erushed rock they are subidiomorphic and lie with their longer axes parallel, recalling a true fluidal structure. Such phenocrysts lie sensibly parallel to the planes of crush-schistosity. Generally, however, the crushing has been so intense that the phenocrysts are now lenticular and more or less rounded. In this case they stand out as ‘ eyes’ and, while the core of each crystal still holds its glassy lustre and recognizable cleavages, the outer shell of the crystal, for a depth of one to two millimetres, is opaque-white and lustreless, owing to the peri- pheral granulation of the phenocrysts. A third and very common phase consists of zones from a few inches to fifty feet or more wide, in which the crushing has developed a medium to coarse grained, equigranular biotite-gneiss or muscovite- biotite gneiss. This gneiss is devoid of phenocrysts, probably for the reason that these have been completely merged with their ground-mass through excessive granulation in zones of maximum shear. Of the three phases the augengneiss is the most abundant. The planes of schistosity of the granite have a fairly constant attitude with a strike varying from N. 30° W. to N. 10° W., and dips varying from. 60° W. to 75° E. The average attitude is about: strike, N. 15° W., and dip, 80° W. The eneissic bands are very seldom, if ever, crumpled, but continue nearly vertical through thousands of feet of depth in the mass. The apophyses of the batholith are often coarsely pegmatitic. They are often greatly faulted, distorted or pulled out into discontinuous pods, showing that the country-rock about the intrusive has shared in the energetic deformation of the batholithiec body. It is possible that the deforming stresses were at work before the granite had thoroughly solidified, thus explaining the apparent flow-structure in certain phases of the batholith; but most of the deformation must have followed the crystallization of the ground-mass, the minerals of which are so greatly strained or granulated. Under the microscope the phenocrysts are seen to be chiefly orthoclase or microcline, more rarely acid oligoclase, near Ab, An,. The ground-mass is com- posed of quartz, oxthoclase, oligoclase, microcline, microperthite, biotite, and muscovite with a little accessory magnetite, apatite and titanite. All of these minerals are more or less bent or fractured. The crushing has been so intense that it is now impossible to state the original diameters of the ground-ma:s essentials, though the average for the quartz and _ feldspars 286 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 must have been several millimetres. It is likewise difficult to be certain of the exact nature of the original feldspars. Microcline, microperthite, and musco- vite are all more abundant in the phase of greatest crushing, and are probably in the main of metamorphic origin. Some part of their volume may thus represent the product of changes wrought in the somewhat sodiferous orthoclase. The soda-potash intergrowths of the microperthite have not, as a rule, the regularity of form characteristic of this feldspar when erystallized directly from an alkaline magma. In the present case the albitic material has been segregated in irregular lenticules and stringers which seem to represent fractures in the original ortho- clase. A little of the muscovite may be a primary accessory of the rock, for it then occurs in parallel intergrowths with the undoubtedly primary biotite. It may be noted that the accessories, apatite, magnetite, and titanite, are either entirely absent or exceedingly rare in the zones of specially intense shear- ing and crushing; their removal seems to be one of the results of the meta- morphism. In several thin sections prisms of allanite were noted and, in one slide, a little fluorite; these minerals should, probably, be added to the list of primary accessories. All phases are generally very fresh and the secondary products, kaolin, chlorite, and sericite, are unimportant. The observation was made in the field that the rock of the zones of maximum shearing and crushing is very consider- ably tougher under the hammer than the coarser porphyritie granite and augen- eneiss alongside. In the bed of Boundary creek the former rock stands out in long ridges or riffles, between which the softer granite has been eroded by the sluicing waters of the creek. This contrast of strength shows that the batholith lay deeply buried at the time of its shearing so that the crush-zones underwent cementation, which made them actually stronger than the rock more closely resembling the original granite. The specific gravities of typical specimens from the batholith vary from 2.640 to 2-677, with an average for five specimens of 2-658. A large type specimen, collected at a point on the Boundary creek wagon- road, about two miles from the ferry at the eastern end of the road, has been analyzed by Mr. Connor. The large phenocrysts are here generally micro- cline, although a few, twinned on the Carlsbad and albite laws, are acid oligo- elase near Ab, An, The essentials of the coarse ground-mass are quartz, microcline, orthoclase (sometimes obscurely microperthitic), oligoclase averag- ing apparently Ab, An,, muscovite and biotite. The eccessories are the same as those noted in the foregoing description of the average rock. e>a. In sections parallel to (010) the extine- tion is 16° 30’; in sections parallel to (110), 20° 15’. These values show that the optical angle is unusually small and near 50°.* The hornblende has proper- ties somewhat similar to those of the variety ‘ philipstadite.’+ The biotite is deep brown with powerful pleochroism; it is sensibly uniaxial. The diopsidic augite is colourless to pale greenish in thin section and is not noticeably pleochroic. It is quantitatively subordinate to the biotite but in all the specimens collected must be ranked among the essentials. The other constituents need no special note. Though the rock is unusually strong and fresh, a little secondary kaolin and yellow epidote may occasionally be seen. The specific gravity of the rock varies from 2-743 to 2-785; the average for five fresh specimens is 2-757. ; Mr. Connor has analyzed a typical specimen (No. 858) from the vicinity of the Bayonne mine, with the following result :— *Cf R. A. Daly, Proc. Amer. Academy of Arts and Science, Vol. 34, 1899, p. 311. + Proc. Amer. Academy of Arts and Science, Vol. 34, 1899, p. 433. REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a Analysis of basic granodiorite, Bayonne batholith. oO.. HO at 110°C. 25s SDP agin The caleulated norm is: Quartze. 15 23 20 00% 100-01 2-785 13:32 19-46 24.63 24019 717 5-90 3°48 1.21 31 38 100-05 19-5 17-4, 3-7 23-8 16-2 11-8 44 6 oy] 5) 1-2 100-0 291 3 ee 2 har- In It is quite possible that the batholith has more acid phases in the region north of the Boundary belt and thus nearer the centre of the mass. 25a—vol. 11—194 292 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Basie segregations, in the form of deep green to black ellipsoids from five centimetres or less to ten or fifteen centimetres in diameter, are quite common. These small bodies are of two classes. In the one class the essential components are hornblende, labradorite (Ab, An,), biotite, and augite, named in the order of decreasing abundance. A little quartz and orthoclase, with much crystallized titanite, magnetite, and apatite are accessory. Microperthite and microcline seem to be entirely absent. The specific gravity of a typical specimen is 2-924. In the other class of segregations the colour is yet deeper and is explained by a complete lack of feldspar. The essentials are hornblende, biotite, and augite, also named in the order of decreasing importance. Quartz is accessory but is considerably more abundant than in the first-mentioned class of segregations. The other accessories are titanite, apatite, and specially abundant magnetite in erystals and rounded grains. The specific gravity of a typical sample is 3-214. There can be little doubt that all these bodies are indigenous and that the segregation of the material, if not its actual crystallization, took place in the early stage of the magma’s solidification. The granodiorite is generally massive and uncrushed. Straining and granulation through pressure were not observed in any of seven thin sections cut from the specimens collected. Sometimes, though rarely, thin partings in the granodiorite carry much biotite, which is arranged with its lustrous foils lying in the planes of parting, as if there developed as a result of shearing-in the crystallized batholith. At the Bayonne mine the rock is sheeted and locally sheared. On the whole, however, the batholith is notably free from evidences of dynamic disturbances and appears never to have suffered the stresses incid- ental to an important orogenic movement in the region. As regards its influence on the intruded formations the Bayonne grano- diorite has typical batholithie relations. A glance at the map suffices to con- vince one that this huge mass is a cross-cutting body. Four of the thickest members of the Summit series are sharply truncated by the main southern contact. For distances ranging from one to two miles from that contact the rocks of the Wolf, Monk, Irene Volcanic, and Irene Conglomerate formations. are greatly crushed, fractured, and metamorphosed by the energetic intrusion. Farther to the eastward, for a distance of ten miles down the Dewdney trail, the schists and interbedded quartzites of the Priest River terrane, though likewise truncated, have been almost completely driven out of their regional strike and a well developed schistosity peripheral to the batholith has been found in these recrystallized rocks. For the lower twelve miles the east-west Summit creek canyon has been excavated along the strike of the schists, which have been forced out of their originally meridional trends by the force of the intrusion. Abundant apophyses of the batholith sometimes 300 or 400 yards in width, cut these various invaded formations. The main contact is sinuous but clean-cut. Inclusions of the invaded rocks are not common in the bath- olithic mass as studied in the ten-mile belt. REPORT OF THE CHIEF ASTRONOMER 293 SESSIONAL PAPER No. 25a CONTACT METAMORPHISM. The reerystallization of the rocks of the Priest River terrane through the influence of the intruded magma, is most conspicuously shown along the Dewdney trail. This trail threads the floor of the deep Summit creek canyon as it rises from the 2,000-foot level near the Kootenay river to the 3,000-foot level, about nine miles farther up Summit creek. The main contact of the TERRANE PRIEST RIVER Summit Creek Sea level 0 { 2 Miles es ere Re oe et Se} Vertical and Horizontal Scale Figure 19. North-south section illustrating probable explanation of the great intensity and extent of the contact metamorphism at Summit Creek. Aureole of contact metamorphism shown by cross-lining. Folds shown in Priest River Terrane purely diagrammatic. batholith runs nearly parallel to trail and creek and at an average distance of about 2,200 yards from both. The aureole of contact metamorphism is here two to three miles wide. The metamorphic effects seen along the trail are, however, greater than they would be at the same distance from the exposed igneous contact and on the same level as the nearest contact. The line of contact runs generally from 2,000 to 3,200 feet higher than the trail at the bottom of the canyon. The extraordinary intensity of the metamorphism along the trail is, thus, in part explained by the depth to which Summit creek has excavated its canyon in the sloping roof of the batholith. In other words, the strength of the metamorphism suggests that the contact-surface of the batholith is not vertical but dips under the creek bed; and that, on this south- ern extremity at least, the batholith has the section of a body enlarging down- wardly. (Figure 19.) 294 DEPARTMENT OF THE INTERIVUR 2 GEORGE V, A. 1912 In these Priest River rocks the thermal metamorphism has not developed new types of minerals to any notable extent. The changes in the quartzitic beds con- sist chiefly in their becoming micaceous, with the liberal generation of both muscovite and biotite. The phyllites, metargillites, and quartz-sericite schists, interbedded with the quartzites, have been converted into coarse, glittering mica schists, in which the individual mica-plates average scores of times the size of the original micaceous elements in the equivalent bands farther south and not thermally metamorphosed. These metamorphosed schists are regularly composed ot dominant quartz, muscovite, and biotite in variable proportion, giving mus- covite-quartz schist, muscovite-biotite-quartz schist, and biotite schist. Grains of plagioclase and orthoclase are accessory in variable amount. Here and there prisms of tourmaline are developed in abundance. In general, the metamorphic effects along the trail are of a nature leading to a higher crystallinity and ecoarser granularity in the ancient sediments rather than to the generation of new minerals. This effect is manifest for distances as great as three miles from the main contact of the granite. Since the exomorphic collar was not thoroughly studied in the part lying north of the Dewdney trail, it is possible that many variations on the described simple scheme of metamorphism would be discovered by one exploring the inner edge of the collar. On the other hand, the mineralogical changes in the Summit series of rocks are often very marked. This is the case even at long distances from the grano- diorite contact. One of the most remarkable instances is shown in the band of basal Irene conglomerate. At the Dewdney trail, nearly two miles south of the batholithic contact, this rock is exposed on a large scale. As usual it is intensely sheared, with its quartzite, carbonate, and slate pebbles rolled out into flat lenses and ribbons. The thermal metamorphic effects are most pronounced in the cement, which is often abundant. In ledge and hand-specimen the cement is of a dark green colour and of silky lustre, evidently due to abundant biotite and mus- eovite crystallized in minute individuals. In the less metamorphosed beds the microscope shows that grains of quartz and carbonate are the other essential constituents. There is considerable effervescence with cold dilute acid, showing: that the disseminated grains of carbonate are, in part, calcitic. The numerous pebbles of carbonate are true dolomite. Through their mashing the cement has become mechanically impregnated with grains and small, granular aggregates of dolomite. The calcite may be, in part at least, of secondary origin and, in any ease, is subordinate to the magnesian carbonate. On the whole, the composition of these few, relatively unaffected bands of the conglomerate is like that des- eribed for the standard sections of the Irene formation. For hundreds of feet of thickness the cement has been very notably altered through contact action. The chief effect consists in the extremely abundant generation of dark green, actinolitic amphibole, forming long straight or curved prisms. These often shoot irregularly through the quartz-mica ground-mass or form beautifully developed sheaves and rosettes, which are specially well exhibi- ted on fractures parallel to the schistosity. The individual prisms run from 1 em. or less to 3 or 4 em. in length, with widths usually under 1 mm. The amphibole has the optical properties of actinolite. REPORT OF THE CHIEF ASTRONOMER 295 SESSIONAL PAPER No. 25a The study of several thin sections has convinced the writer that the amphi- bole has been generated at the expense of the dolomitie grains disseminated through the cement, thus illustrating a familiar phase of the metamorphism of carbonate-bearing rocks. When the carbonate was abundant in the cement, the actinolite now forms as much as a third or a half of the rock. ‘Considerable epidote and basic plagioclase were also formed in some beds. Such metamorphic effects are noteworthy in view of the distance of these outcrops from the main batholithic contact,—about 3,000 yards. A partial explanation of the metamor- phie intensity is again to be found in the probable fact that the granodiorite lies beneath these outcrops and at a distance of considerably less than 3,000 yards downward. Two specimens of the Irene lavas were collected at the Dewdney trail. These seem to be typical of the lavas of the exomorphie zone where, as a rule, they have been completely changed to fine-grained or medium-grained, highly fissile hornblende schists. Green hornblende and quartz are the principal com- ponents; grains of carbonate, apparently dolomite, and a little basic plagio- - clase are present in both thin sections. The phyllitic schists of the Monk formation have been signally metamor- phosed by the batholithic intrusion. For a distance of a half mile or more cutward from the granite, these rocks have been converted into a schistose hornfels composed of quartz, muscovite, biotite, sillimanite, and red garnet, along with much untwinned feldspar, apparently all orthoclase. The muscovite foils either lie in the plane of schistosity or oceur with random orientations through the rock. In the latter case they are spangles measuring from 0-5 mm. to 1-5 mm. in diameter and are in phenocrystic relation to other consti- tuents. The sillimanite has the usual development in needles which are often aggregated in tufts or sheaves very conspicuous under the microscope. The orthoclase grains show a tendency to aggregate along with some grains of quartz in lenses 1 mm. to 2 mm. long, these lenses lying in the plane of schistosity. The abundance of the orthoclase in some of the specimens suggests that its substance has keen introduced from the magma, but this is not ‘certain. The garnet is pale reddish to nearly colourless in thin section and has the usual habit of the mineral in contact-zones. On the top of the 6,600-foot ridge which overlooks Summit creek on the north and runs eastward directly from the peaks at the western head of the ereek, a thick series of ferruginous schists are exposed for a distance of a mile measured along the ridge. These schists dip under the Wolf grit and overlie the 200-foot bed of breccia-conglomerate at the top of the Irene voleanic forma- tion. There is little doubt that these ferruginous schists are the much meta- morphosed equivalents of the rocks of the Monk formation. Four type spect- mens were collected at points about 1-5 miles from the contact of the Bayonne granite. All of them have been microscopically examined and prove to belong to the one species of staurolite-schist. The staurolites form subidiomorphic erystals and anhedra of all sizes up to 15 mm. in length. In transmitted light they are usually of a strong yellow colour. As usual, quartz inclusions are 296 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 very numerous, so that hundreds of minute clear lenses or droplets of that mineral are contained in a single crystal of the staurolite. The inclusions are almost invariably arranged with their longer axes’ parallel to each other and, at the same time, parallel to the plane of schistosity of the rock. This orienta- tion of the inclusions appears to indicate that they are residuals of the quartz grains composing the schist before it was thermally metamorphosed; the stauro- lite crystals grew quietly in the rock without causing mechanical disturbance of the pre-existing, schistose structure. Sericitic muscovite, biotite, and quartz form the matrix in which the abundant staurolite lies. These relations of the staurolites to the ground mass find full analogy in the rocks illustrated in figures 88 and 89 of Rosenbusch’s Elemente der Gesteinslehre, 1898, p. 498. Abundant twinned crystals of disthene, which do not show inclusions of the ground-mass often accompany the staurolites. Even from the foregoing brief account of the contact action of the Bayonne batholith, it is clear that the exomorphic collar is unusually broad and that the - action was correspondingly powerful. To the future geologist who plans to make a thorough study of the collar, interesting results may be promised. The different beds which have been altered should be identified and followed, so as to determine the whole gamut of changes involved in the metamorphism of each, and to find the relation of these changes to distance from the grano- diorite. This work would entail the expenditure of much more time than could be devoted to the study during the Boundary belt survey. The mountains. are very rough; the work must, in any case, be time-consuming and arduous, but the result would amply repay the effort. SATELLITIC STOCKS ON THE DIVIDE. On the main water-parting of the range and just south of the Dewdney trail a granite stock, cutting the middle members of the upturned Summit series, is well exposed. In ground-plan this body is an ellipse with a north- south major axis of 2-5 miles and a width of one mile. One-half mile west of this stock there occurs a small intrusive mass of the same granite which sends a long dike-like tongue northeastward across the Dewdney trail, where the rock is easily studied. Petrography.—This granite is medium-grained, of a light pinkish-gray tint, and is noticeably poor in dark-coloured constituents. Quartz, microperthite, orthoclase, a little microcline, and considerable oligoclase, Ab, An,, with a quite subordinate amount of biotite are the essentials; titanite, magnetite, apatite, and zircon are sparingly present. Primary muscovite is accessory and is often regularly intergrown with the biotite. Along the western contact of the larger stock the muscovite becomes so important that the rock may be called a two-mica granite; its structure in this contact zone tends to the panidio- morphic. The average specific gravity of four fresh specimens of the granite fs 2-628. REPORT OF THE CHIEF ASTRONOMER 297 SESSIONAL PAPER No. 25a A typical specimen from this stock was studied quantitatively according to the Rosiwal method and the following weight percentages of the different constituents were found: (COENEN Ae =e eae ache a PRE ae eR RAH BLOTS ON no prs A rl he fh 24-9 Micropenthiterreenecmscicts nn el eee ae 56 31-3 Sodiferous orthoclase.. 15-9 Oligoclase.. .. .. » 12-1 Biotitescs 7 BAYONNE BATHOLITH 2 4 5 Miles Horizontal and Vertical Scales. Figure 22.—Diagrammatic section showing relation of the summit stocks of Nelson Range to the Bayonne batholith. differentiation of the magma through gravitative adjustment. On the other hand, the smaller bodies may owe their lower density to their having been specially acidified by the solution of the invaded quartzites; or, thirdly, the more salic character of the satellitic stocks may be due to special concentration of magmatic fluids in the smaller chambers, facilitating more extreme differentia- tion in them than in the main Bayonne batholith. Probably .all three causes have operated. LOST CREEK GRANITE BODY. The peculiarly shaped mass of granite over which Lost creek flows is, mineralogically, chemically, and genetically, akin to the granite of the Summit stocks. The staple rock is alkaline, with microperthite and orthoclase as the dominant feldspars. Oligoclase is the subordinate feldspar, biotite the only femic essential; primary muscovite, magnetite, apatite, and zircon are the accessories. In places the muscovite has the rank of a subordinate essential, so that the rock varies from biotite granite to biotite-muscovite granite. The average specific gravity is 2-617. Along all observed contacts this granite, for a distance of several score of feet inward, is aplitic and poor in mica. The apophyses are generally composed REPORT OF THE CHIEF ASTRONOMER 303 SESSIONAL PAPER No. 25a of the same aplitic phase. At certain points numerous blocks and shreddy frag- ments of quartzite and schist were observed in the granite. These xenoliths, especially the schists, have undergone much metamorphism, with the generation of abundant andalusite in stout prisms, broad leaves of muscovite, and biotite in aggregates which mottle the rock in striking fashion. Again large amounts of microperthite are disseminated through the altered schist, as if introduced from the magma. The northern arm of the body has the form of a huge irregular dike or sill which follows the strike of the invaded schists. The exposures do not favour the decision as to whether or not the mass here follows planes of bedding or schistosity. The other and larger arm of the mass is clearly in cross-cutting relations. The width of this band is doubtless the greater because of the excava- tion of the deep canyon of Lost creek. If erosion should remove a few thousand feet more of the sedimentary cover at the head of the creek, the Lost creek body and the summit stocks would doubtless be found to form one continuous batholithic mass. A small intrusion of the Lost creek granite occurs_on the divide between Sheep and Lost creeks and 1-5 miles east of Salmon river. It cuts schists and limestone probably of Carboniferous age, and the youngest bedrock forma- tions with which this whole group of granites, including the Bayonne batholith and its satellites, is known to make contact. The date of these intrusions will be further discussed in a following summary. BUNKER HILL STOCK, Within the ten-mile belt an igneous body which appears to be the most wes- terly satellite of the Bayonne batholith is a stock covering about eighteen square miles and lying almost wholly on the western side of the Salmon river. This stock is composed of a medium to rather coarse, alkaline biotite-granite (specific gravity, 2-610) which, in all essential respects, is identical with the granite forming the small summit stocks and the Lost Creek body. The Bunker Hill mine (now shut down) is situated in the metamorphic aureole of this stock and it may, for convenience be referred to as the Bunker Hill granite stock. Being generally more weathered, this granite has a more reddish tint than the Lost creek and Summit granites. The stronger weathering effect may be partly due to the fact that the Bunker Hill granite has been much more strained and crushed than the more easterly bodies. A distinct schistosity has been thus produced at many points in the stock. The gneissic structure is most pronounced near the southeastern contact, at the confluence of Lost creek and Salmon river. For a distance of 500 feet or more from the contact the granite is specially basic and consists essentially of quartz, biotite, plagioclase (abradorite Ab, An, to basic oligoclase, Ab, An,), with very subordinate ortho- clase, and abundant muscovite foils. The plates of the white mica lie in the planes of schistosity and are of metamorphic origin. This basic phase recalls the muscovite-bearing quartz diorite which forms the many apophyses in the shatter-zone about the summit stocks. 304 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 This granite also has thermally metamorphosed its country rock, in this ease the Pend D’Oreille schists. The metamorphic aureole is nowhere as wide as those about the summit stocks or the Bayonne batholith; it is thus probable that the contact surface of this stock dips under the invaded rocks at higher angles than those characteristic of the contacts in the eastern bodies. The Bunker Hill aureole has not been systematically studied with the microscope. Thin sections of two specimens collected, one at the southwestern contact, the other at a point about 1,000 feet from the contact, both showed the abundant generation of andalusite prisms in the characteristic micaceous hornfels. At Bunker Hill mine the andalusite schist is enormously crumpled and is cut by - veins of gold-bearing quartz. On one of the veins the mine shaft has been sunk for free-milling ore. SALMON RIveR MONZONITE. Halfway between Sheep creek and Lost creek, and a mile east of the Salmon river, the Pend D’Oreille schist is cut by a small stock of plutonie rock, which, in chemical and mineralogical composition, is unique among the known intru- sives of the Selkirks within the ten-mile belt. The stock has the subcircular ground plan of a typical granitic boss, measuring 700 yards in diameter. The rock is relatively prone to disintegration and it has weathered freely into huge bouldery masses, whose forms have been produced by exfoliation and con- centric weathering on joint blocks. By the energetic intrusion the schists round about have been crumpled, hardened, and converted into hornfelsy, massive rock. This contact aureole is a few hundred feet in width; it has not been studied microscopically. The igneous rock is dark greenish-gray and rather coarse-grained. It is massive and quite uncrushed. With the unaided eye, augite, biotite, and feld- spar can be readily identified as the essential constituents. The first named mineral forms highly idiomorphic, stout prisms of varying lengths up to that of 7 mm. or 8 mm. The biotite occurs in lustrous black, often idiomorphic foils which may be 2 mm. or more in diameter but average about 0-6 mm. Between these femic essentials the feldspar forms a kind of mesostasis, numer- ous individuals approaching 5 mm. in diameter. Many of the larger crystals schillerize in vivid sky-blue colours which are specially brilliant when the rock is wetted. Under the microscope the augite shows the cleavages, the very pale green almost colourless tint, double refraction, and extinction angles of a diopside. One crystal in a thin section showed a narrow interrupted mantle of green hornblende about the pyroxene. The biotite is sensibly uniaxial and has power- ful absorption. The feldspar belongs to the alkaline and soda-lime groups, which are represented in nearly equal proportions. The larger, schillerizing individuals have the optical properties of soda-orthoclase and microperthite. The same crystal often has the homogeneous structure of soda-orthoclase in One part and the familiar microperthitic intergrowth irregularly developed REPORT OF THE CHIEF ASTRONOMER 305 SESSIONAL PAPER No. 25a in other parts; these two feldspar varieties are here clearly transitional into each other. The extinction-angle of the soda-orthoclase is 10° 30’ on (010), showing a high content of soda. Its double refraction is markedly low. It is possible that some of this homogeneous feldspar is anorthoclase, but the extiné tion of flakes cleaved parallel to (001) was found, in three cases, to be parallel and thus corresponding to the monoclinic isomorphic mixture. The schilleriz- ing effect, like the chemical composition, relates this feldspar to the dominant feldspar of Brégger’s original laurvikite. The alkaline feldspar often encloses poikilitically idiomorphic to sub- idiomorphie plagioclase, which occurs always in relatively small crystals, averaging about 0-5 mm. in length. These are commonly twinned according to the Carlsbad and albite laws and are often irregularly zoned. The average plagioclase is labradorite, near Ab, An,. Moderate amounts of apatite and magnetite are accessory, while very rare, interstitial grains of quartz are also found. The structure is the hypidiomorphic-granular. The order of erystalliza- tion appears to be: apatite and magnetite; augite; biotite; plagioclase; soda- orthoclase (and microperthite) ; quartz. Mr. Connor’s analysis of a fresh specimen (No. 671) gave the result :— Analysis of Salmon River Monzonite. Mol SiO, 50-66 844, TiO, 1-32 016 Al.O, 16-91 166 e0,.. 1-71 011 FeO... 6-17 086 Mn0O.. 16 002 MgO.. 5-50 138 CaO.. 8-26 147 Sro.. -08 01 BaO.. 23 001 Na.O. 2.89 047 K.,O 4-45 047 EP Or aed OOO ee er mk aman al he tae HEOZaboviertl OL Me tery ciate ors chera ane nero feta eet oe ent Ce 1-06 IOs. 91 006 100-45 See iysrne rahe aro Pom ores eh. casera he Dose epee tient ekoetares 2.843 The calculated norm is: Orthoclase-me serie see rela eae eee ees a cena eee oe 26-13 LMM Tiers eeeiic Nia UAE see Sete Pie MEER Bate oer Gel ee ABPRMELEE Lae APL re Cob s othe Demled Kiyargrt INepneliteneaerery ce stened Sete cane DY @ eae iceman ale ome RC ume Gey Pyne cere te rn ATR hE Meer Ree ERS 1-05 007 100-00 SSO ate easca ws owneves iSonic is ties kere Ao on re enantio meets ura 2-723 The calculated norm is:— @MEthOClaSesy tcc ce cccacion catycle Feiee oar e eta Sen a fotanetn Pee 25-58 IAUIDIGG Sse fain ater pa eines! Wekeri ole heamucved eels ache Reelin Sussterto ure 32°49 INGO SS oa Go Ob Go Garoo so od. oC 85 Anorthite.. . eis Se, RR Soa CIS ete asians 15-29 DIOPSIAC <5 vis csiel vee eles note! eleruinsteee use. 3 8-47 Olah ne ang ba GOmonumOnsoGw ilo od 9.29 MMO WO ceo as Core sevens: 6 1-67 MER aN ae ade oom: GuLtoo. Goode coun bo. doo 06 1-86 IN OER OKs Seereo HOG AO OGNOO ot. uorao mb noe 2-17 IWiatODi et 230 Last Aes Bi EE OFabovie OSC car gc merch 5 oe saps face wae ecioe Weed picay eee Sete 2 OOM errr TPL O jes say re Mites a a le ba 54 O04 Cor. Giddens 100-66 SDenslyiereme ee eeaers wears mea sent eh ient ts) (Soe. SW ee ae, Shas 2-740 *Cf. L. V. Pirsson. 20th Annual Report, U.S. Geol. Survey, part 3, 1899, p. 532. 314 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 The calculated norm is :— Orthoclalsewd: (383s ye cral SAR heres Besa Ns re cae 24-46 PANU are es Siete ciate Wiese ies” aie ata: phere os levels We oN Ok ep ama am 15-20 INGDHCLICO es pete eich Gas Ruemivie es bell en Mertelen eee oe hE Tae 3-69 PAN OTEHICG Ss eiF ose EEL bio Fae Ta aa EOS A ees ee 18-90 DFO PST oi25 alae wee eel tes ee Masons eee ave lea Toc Jove ove de ops UaR cane a oe 16-68 QU WAM Onis ear caen a, Scebed oP tne Miers lumi Carat tbel Ux tera styrene hee ea 5-97 MTOM Ube See eae NSS orm ree os fois fcersuna se aye ene Ghee See Tee cS ao 1-36 Miaanebiteniaeic rs. ce. iocey aan ot idectmagelerebl ys shuee Pe shia tene sabe eames 3-94 ON EL EO ee Tae Soa he RG | a a NU a Rah SN 1-24 Water tandyGOsicn secret ecitsne res mete ist tere creep aes eae ocd eee Renee aaneee es 9-14 100-58 According to the Norm classification the rock enters the sodipotassic sub- rang, shoshonose, of the alkalicaleic rang, andase, in the dosalane order, ger- manare. Chemically it is nearer minette than a typical kersantite, but, by the older classification, the character of the feldspar places the rock in the kersantites. Camptonite—Only one occurrence of camptonite is known as a result of field study in the Boundary belt across the Selkirks. This rock, which microscopic study shows to conform well with the type camptonite, forms a wide but very poorly exposed dike cutting the Pend D’Oreille phyllite on the south side of the Pend D’Oreille river about 1,900 yards east of Waneta. Odinite-—A half mile farther up the river and on the same bank, the phy]l- lite is cut by a six-inch dike of a rock which appears to represent another occurrence of typical odinite as described ‘by Rosenbusch in his last edition of the Mikroskopische Physiographie der Massigen Gesteine. This lamprophyre is a dark greenish-gray, compact rock with conspicuous though small phenocrysts of augite and others of labradorite. The microscope shows these to be embedded in a microcrystalline ground-mass composed essentially of very many minute prisms of hornblende, feldspar microlites, and less abundant granules of augite. A detailed description of this one thin dike, though composed of a relatively rare species of lamprophyre, is scarcely warranted in the present report. APLITIC AND AcID APOPHYSAL DIKES. Practically all of the granitic bodies in the Selkirks where crossed by the Boundary belt have sent tongues or apophyses into their respective country- rocks. These dikes show the familiar variation from quartz-feldspar aplites to the aschistic porphyries corresponding to the different types of plutonics. Other sills and dikes occur at distances too great to be regarded as necessarily apophyses from any visible stock or batholith, and in some cases it is not possible to determine whether these detached acid eruptives represent distinct periods of eruption. None of the bodies seems to demand special description. One of the dikes is cut by augite minette and by the analyzed hornblende- augite minette which occur on the western bank of the Columbia river about. REPORT OF THE CHIEF ASTRONOMER 315 SESSIONAL PAPER No.: 25a 300 yards south of the Boundary slash. The acid dike is a typical biotite granite porphyry. It is between 200 and 300 feet wide and is paralleled by other great dikes of similar material outcropping at low water in the islets of the river channel. They may be acid apophyses from the extensive Trail batholith toward which they strike; they are, however, noted here because their relation to the younger minettes is very clear. (See Figure 23.) A white aplitie sill cutting the’ Pend D’Oreille phyllitic schist on the right bank slope of the South Fork of the Salmon, about 2-5 miles S. 30° W. of the summit of Lost mountain, may be mentioned on account of the unusual struc- ture of the rock. It is slightly porphyritic with phenocrysts of quartz and sodiferous orthoclase. The ground-mass is partly the common panidiomorphic aggregate of quartz and alkaline feldspar (much sericitized) but contains quite numerous, small spherulites of alkaline feldspar which is developed in rosettes. A few grains of magnetite represent the only other constituent. The relations of this sill to the other granitic rocks of the range are unknown. DIKE PHASES OF THE ROSSLAND AND BEAVER MOUNTAIN VOLCANICS. The formations older than the Rossland and Beaver Mountain lavas are, naturally, cut by dikes which indicate vents for the lavas or the fillings of fissures connected with those vents. A few of these dikes have been found in localities where erosion has stripped away the voleanic cover and some of them have been microscopically examined. Among these, four types may be listed but it should be understood that the list does not exhaust the different varieties of the dikes genetically connected with the volcanics. Just east of the large boss of Sheppard granite mapped on the Pend D’Oreille river, the schists are traversed by a fifty-foot, nearly vertical, north- south dike of porphyritic monzonite. The phenocrysts are stout prisms of augite up to 8 mm. in length. The essentials of the hypidiomorphic-granular ground-mass are orthoclase, microperthite, labradorite (Ab, An,), augite and biotite; the essentials are magnetite, apatite, zircon and a little interstitial quartz. The plagioclase crystals are characteristically clumped in the ortho- clase mesostasis. About three-quarters of a mile north of Old Fort Sheppard, where the mountain-spur projects through the terrace sands and gravels to the Columbia river, there are large outcrops of slaty and quartzitic rock which have been mapped as part of the Pend D’Oreille group. The crumpled and mashed slate is here cut by a 25-foot vertical dike of dark-gray hornblende-biotite monzonite striking N. 8° E. (visible at low water). Some 300 yards south of the Boundary slash on the same side of the river and at the water’s edge, three dikes from ten to thirty-five feet wide and of macroscopic appearance somewhat similar to the monzonite were found to consist of hornblende-augite gabbro. In this type the feldspar is basic labradorite (Ab, An,), and alkaline feldspar is entirely absent; a few foils of biotite are accessory. 316 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Finally, a three-foot, north-south, vertical dike of highly amygdaloidal basalt, cutting the Pend D’Oreille phyllite about fifty yards west of the mouth of Twelve-mile creek, may be noted. RELATIVE AGES OF THE ERUPTIVE Bopies. The entire lack of paleontological evidence within the ten-mile belt makes it impossible to form a full chronological column for the formations occurring in this part of the Selkirks. It may be recalled that the Priest River terrane unconformably underlies the great Surhmit series, with a part of which (the Beehive formation) the Kitchener quartzite is believed to be equivalent. The Pend D’Oreille group overlies, with apparent conformity, the Summit series and, as will be further indicated in the next chapter, unconformably underlies the Rossland and Beaver Mountain groups of sediments and voleanics. The relative ages of the igneous rocks can be partly indicated through their relations to these sedimentary groups as well as through their relations to each other. The observed facts may be briefly summarized. The intensely crushed Rykert granite batholith cuts the Priest River terrane, including bodies of metamorphosed hornblende gabbro which themselves cut the schists and quartzites of the terrane. The uncrushed and very rarely sheared Bayonne batholith cuts formations belonging to the Priest River terrane and Summit series respectively. The satellitic stocks believed to be contemporaneous © with the Bayonne batholith cut the Pend D’Oreille group and one of them—the Bunker Hill stock—seems to cut the older members of the Rossland volcanic group. The Salmon river monzonite jstock cuts the Pend D’Oreille schists and limestone. The abnormal hornblende granite at Corn creek cuts the Kitchener quartzite and is tentatively correlated with the Purcell sills. The minettes, kersantites, camptonites, and odinites cut the Pend D’Oreille schists or limestones and probably also cut the Rossland volcanics, since similar lamprophyres cut the Rossland monzonite stock which is almost certainly of the same general age as many of the Rossland lava flows. The peculiar por- phyritic olivine syenite cuts the Rossland voleanics; in the next chapter the correlation of this syenite with the minettes will be indicated. The Sheppard granite cuts the Trail granodiorite which itself cuts the older members of the Rossland voleanics. Since a half dozen of the principal formations in these Selkirk mountains are more directly associated with fossiliferous sediments in the mountains across the Columbia river, the discussion of the final correlation of the Selkirk rocks will be postponed to the chapter dealing with. the geology of the Rossland mountains. At this point it will be sufficient to anticipate that discussion by tabulating the Selkirk formations in their probable order of age:— REPORT OF THE CHIEF ASTRONOMER 317 SESSIONAL PAPER No. 25a Salmon River monzonite stock 0.2... . 6... sweeten cence Bayonne batholith and its satellitic stocks................ 0... Sheppard granite stocks and dikes...............-...2-e.e0eee0++- Lamprophyres, minettes, kersantites, odinite and camptonite . AN RIKSO GHGS aca daigosaeoba woubeo ures snecauDCOOUmedeanoods efvele Trail granodiorite batholithacpee scar cnc acces ane ae ReaversMountaintcroupy. eeeree cre eee eee eee oer Rossland volcanic group, with interbedded sediments (in par t).. Monzonite, gabbro, and basaltic dikes cutting Pend d’Or eille group UNCONFORMITY. Rvkertigranitelbatholith spre puch racic aol ye aoe cress Renasd.Oreille rou pis eret clerics stave cere ae nieve cisiacneierest Abnormal hornblende granite sill cutting Kitchener quartzite at Wornvereeksekern vay Sete ceia Cataract aise cate aloud eeaue wiseueee eae Metamorphosed gabbro sills and dikes cutting Priest River terrane KG TCHEN ere OLMA LION seeks cero see eT ae Te aE ine } Post-Eocene (Miocene 7). Post-Laramie (Eocene ?). | ani (Cretaceous 7). Late Jurassic ? Carboniferous ? (and older ?). ja tddle Cambrian ? Middle Cambrian ? Cambrian and ‘Beltian. Pre-Cambrian and pre-Beltian. \ 2 GEORGE V, SESSIONAL PAPER No. 25a A. 1912 CHAPTER XIII. FORMATIONS OF THE ROSSLAND MOUNTAIN GROUP. It will be recalled that, in the chapter on the nomenclature of the mountain ranges, the Rossland mountain group where crossed by the ten-mile Boundary Belt, is bounded on the east by the Selkirk Valley (Columbia river) and on the west by the meridional valley occupied by Christina lake and the lower Kettle river. On the east the formations of the Rossland mountain group in several instances extend over into the Selkirk system. Of these the Pend D’Oreille series has already been described, as well as a few of the dikes cutting that series along the western bank of the Columbia river. The Trail batholith, Sheppard granite, Rossland and Beaver Mountain voleanic groups, and small bodies of a peculiar porphyritic olivine syenite are represented on both sides of the Columbia and will be described in the present chapter. The western topographic limit of the Rossland mountain group is also, within the limits of the Boundary belt, a clean-cut and convenient line of division between the geological formations of the Rossland and Midway-Christina mountain groups. (See Maps No. 8 and 9.) From the Columbia to Christina lake igneous-rock formations dominate very greatly. Sedimentary rocks appear only in small patches, and are nearly always much deformed and metamorphosed. Though there are good reasons for believing that these rocks are chiefly if not altogether late Paleozoic or post- Paleozoic in age, fossils are almost as rare as they are in the formations of the Rocky Mountain Geosynclinal. The writer has been able to add but little to the stratigraphic information secured by McConnell, Brock, and others who have made studies in the region. However, the interpretation given the few scattered facts in hand differs somewhat from that adopted by these observers. The older sedimentary formations will be described first. They include, besides the small area of the Pend D’Oreille slates, phyllites, quartzite, and lime- stone near the Columbia, a small’ patch of obscurely fossiliferous limestone asso- ciated with chert in Little Sheep creek valley: fossiliferous limestone occurring with the older traps north of Rossland; an intensely deformed series of lime- stones, quartzites, and schists sectioned by the railway line east of Christina lake and named, for convenience, the Sutherland schistose complex; and a few small outerops of old-looking quartzite and argillitic rocks intimately associated with the Rossland voleanics. A very limited exposure of fossiliferous (plant-bearing) argillite, probably of Mesozoic age, will then be described. The voungest sedimentaries observed in this part of the Boundary belt are conglomerates and sandstones which, again from very imperfect fossil evidence, seem to be of early Tertiary or mid-Tertiary age; these beds form four small patches at or near the Boundary line. 319 320 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 The igneous formations to be treated include those which have been named by MeConnell and Brock the Rossland and Beaver Mountain voleanic groups; and those which are referred to by the present writer as the Trail batholith; the Sheppard granite (stocks and dikes); the Coryell syenite batholith with its satellitic dikes, and a satellitic chonolith of syenite porphyry; the Rossland monzonite; several bodies of gabbroid and ultra-basic intrusives; and certain of the numerous dikes which have certain special petrographic interest. At the time when the writer made his examination of the Rossland moun- tain group it was understood that the Geological Survey of Canada was planning a detailed study of the Rossland camp and its vicinity. Accordingly, very little work was done in the region of the town and, in fact, no attempt was made to plan an exhaustive report for the region between Sophie mountain and the Columbia. Specimens of the rocks were collected, but many of the field rela- tions could not be decided in the limited time which it seemed advisable to devote to this part of the Boundary belt. Nearly all of McConnell’s contacts, as pub- lished in the Trail sheet, were followed up and verified. For the rest the present chapter can claim to be no more than a report of progress on the geology of these unusually complicated mountains. PALEOZOIC FORMATIONS. Carboniferous Beds in Little Sheep Creek Valley.—In the bottom flat of Little Sheep creek valley, about 1,000 yards north of the Boundary line and on the west side of the creek, there is a low hill of limestone surrounded on all sides by alluvium. The limestone is of blue-gray to white colour and is much brecciated and highly crystalline. It contains cherty and quartz lenses and true quartz veins. The attitude of the bedding is obscure, observed strikes ranging from N. 55° E. to N. 80° E., with an average northerly dip of about 60°. The limestone contains numerous, poorly preserved crinoid stems which are of some value as pointing to the probability that the limestone is of Paleozoic age. Across the creek there are several large outcrops of cherty quartzite also greatly deformed, with average strike, N. 35° E. and dip, 90°. That rock extends 200 feet vertically up the steep eastern slope of the valley, where it is unconformably overlain by a coarse breccia (probably a voleanic explosion- breccia) containing fragments of the. same obscurely fossiliferous limestone and chert as that just described. The breccia is part of the Rossland voleanic formation, which has here an average strike, N-S. and dip, 385° E. From the composition of the breccia and from the stratigraphic relations, the Rossland voleanies as represented are clearly unconformable to the Paleozoic strata. The latter seem, in fact, to be part of the foundation on which the voleanic mass was spread. During his mapping of the Trail sheet McConnell found in the similar breccia outcropping on the opposite side of this valley, fragments of marble bearing the fossil remains of a species of Lonsdalia and the marble was referred by Dr. Whiteaves to the Carboniferous. It would seem simplest directly to REPORT OF THE CHIEF ASTRONOMER 321 SESSIONAL PAPER No. 25a correlate the limestone in place with the limestone fragments in the breccia on each side of the valley, and the formation, including the limestone and chert, is tentatively placed in the Carboniferous system. Carboniferous Limestone in the Rossland Mining Camp.—tIn 1905 Brock discovered in a limestone band interbedded with andesitic greenstone at the O.K. mine, four miles north of the last mentioned locality, certain fossils which have been referred to Carboniferous species. Sutherland Schistose Complex.—A group of metamorphic rocks, exposed in the railway cuttings between Cascade and Coryell stations, were sectioned during the season of 1902. Although nearly a week was spent on the section, the results of the structural study were meagre. The oldest rocks of the section consist of highly crystalline schists of sedimentary origin. With these are associated many irregular bands of eneissic, gabbroid rocks and amphibolites, and sheared hornblende porphyrites, all of which represent greatly altered basic intrusives. The metamorphosed sedimentary rocks are now represented by -garnetiferous schist, sericite schist or phyllite, biotite-epidote schist, actinolite- biotite schist and andalusite-biotite schist. Massive, often brecciated, greenish ‘quartzite and at least two large pods of white to light gray marble are inter- bedded with the schists. | Structurally the complex is characterized by utter confusion. Neither bedding-planes nor planes of schistosity preserve a steady attitude for more than a few score or hundreds of feet together. The section is located in a zone of maximum dislocation, a zone now followed by the deep trough of Christina lake. The immense alteration of these formations is further due to the intru- sion of numerous large bodies of acid and basic igneous rock, including various gabbros and peridotites as well as the great Coryell syenite batholith. No trace of a fossil was found in the sedimentaries and it is still impos- sible to correlate them with known horizons. The quartzite and limestone associated with the schists are, in general, similar to the quartzite and crinoidal limestone of Little Sheep creek valley and to staple phases of the Pend D’Oreille group. All of them are possibly of Carboniferous age. The gabbroid and peridotitic masses cutting the schists are evidently of more recent dates; some of them show neither crushing nor even appreciable straining under the microscope. Three of these basic intrusive bodies will be briefly described below; a microscopic description of the schists themselves is scarcely warranted by any special petrographic interest they possess. Summary.—In conclusion, it may be noted that some at least of these old- looking metamorphosed sediments are almost certainly of Carboniferous age. Others may be either pre-Carboniferous or else Triassic, if not as late as Jurassic. For the present the writer follows the tradition of McConnell and Brock in placing all of these formations in the Paleozoic. Whatever the age of the sedi- ments, some of them seem to be contemporaneous with thick, massive greenstones and metamorphosed ash-beds of andezitie sort, and it is highly probable that the 25a—vol. 1i—21 322 DEPARTMENT OF THE INTERIOR 2 GEURGE V, A. 1912 greenstones occurring in the Pend D’Oreille group (especially those near the Columbia river) are of the same age. The quartzites and slaty rocks of the Pend D’Oreille group are almost if not quite indistinguishable both in composition and in degree of metamorphism from the quartzites and slates interbedded with the greenstones of the Rossland mountains. The Pend D’Oreille marbles are lithologically identical with the obscurely fossiliferous limestones just described. As the best working hypothesis, therefore, the writer is inclined to believe that the western slope of the Selkirk range and the eastern half of the Columbia system are underlain by residuals ot a very thick upper Paleozoic, probably Carboniferous, series which represents the oldest sedimentary rocks of those parts of the Boundary belt. It will be seen that the same series probably has similar fundamental relations in the Midway and more westerly mountain groups. Mesozoic SEDIMENTS AT LittnE SHEEP CREEK. At Monument 175 in Little Sheep creek valley, erosion has laid bare a con- siderable thickness of stratified rocks which are evidently much younger than the marbles and quartzites farther up the valley. The exposures are not good but, since these younger rocks are also obscurely fossiliferous, the field observa- tions so far made may be detailed. At the Boundary monument the steep slope of Malde mountain is seen to be largely underlain by black and red argillite, enclosing thin beds of gray sandstone and of angular conglomerate, as well as a number of layers of sandstone which is described in the field notes as hard black quartzite. The quartzite is sulphide-bearing. These beds are greatly deformed, the argillite specially showing frequent changes of strike and dip in short distances both up the slope and along its foot. The more rigid sandstone beds tend to have a fairly steady strike of N. 0°-10° E., with an average dip of from 35° E. to 90°. The series, chiefly argillitic, continues eastward to a contour about 600 feet above Little Sheep creek, and there it appears to dip under the voleanie breccias of Malde mountain. This general eastward dip appears to characterize the series throughout its extent of 600 yards up the valley from the Boundary slash. The exposures south of the line did not promise useful results and the beds were not followed in that direction. The exposures are likewise very poor on the west side of the creek, but the shale-sandstone series seems to extend on the Sophie mountain slope at least 500 feet above the creek. The argillite is there greatly crumpled, but probably strikes in the average direction, N. 65° E., with dip high to the northwest. The series seems thus to be at least 600 feet thick and to have the attitude of a broken and mashed anticline plunging to the north, carrying the sediments beneath the Malde mountain and Sophie mountain breccias and lavas. The field relations are, however, so obscure that this conception must be regarded as only suggestive and by no means proved to be correct. At the rock-bluffs along the railway track and on each side of the Boundary slash, a number of very poorly preserved remains of plants were found in the shales. These fossils were submitted to Professor D. P. Penhallow, who iden- tified them ‘as the rachises of a fern, in all probability of Gleichenia (gilbert- *punorsyorq UL WIeyUNOTL ALOT) PIE CE ALVTT ‘si B[SS asodu0d aaULUSTA 5 ‘purfssoy Jo 4samM ‘osplt UleqUNOPY plooayY, ‘sormVd[OA puL[SsoY Jo p PA p. 322. vol. ii 25a REPORT OF THE CHIEF ASTRONOMER 323 SESSIONAL PAPER No. 25a thompsont),’ and tentatively correlates the beds with the lower Cretaceous Gleichenia-bearing strata on the Pasayten river.* ‘The only other information in hand on this question of age is that based on the condition of the stratified series. It is, apparently, too greatly deformed to be placed in a post-Kocene period, while, on the other hand, the degree of metamorphism is too low to warrant our referring the series to the Paleozoic. Either a Mesozoic or Eocene date would be preferable to either of those alternatives. For the present, it seems best to consider the beds broadly as of Mesozoic age. ROSSLAND VOLCANIC GROUP. GENERAL DESCRIPTION. From the Saimon river to the Kettle river at Cascade, a distance of forty miles, the ten-mile Boundary belt contains an irregular though continuous band of basic volcanic rocks. This band covers about 150 square miles of the belt and is part of a voleanic area in the West Kootenay district of British Columbia aggregating 500 square miles. West of the Columbia river the voleanics are developed on the United States side of the Boundary but how extensively is not known. (See Plate 32.) The entire volcanic area is highly accidented by basic and acid plutonic masses which, in general, are younger than the volcanics and cut them. Long continued erosion has revealed many of the dikes, stocks, and batholiths, so that the mapped contact-lines of the effusive rocks are extremely sinuous- Owing to severe orogenic stresses the lava flows, ash-beds and’ breccias usually have high dips and complicated structures. Most of these rocks are altered by erush-metamorphism and contact-metamorphism. They are often involved most obscurely with the Paleozoic sediments just described and also with younger strata which are generally unfossiliferous. The differentiation of the lavas on the ground of geological age cannot as yet be carried out systematically. It is certain that the volcanics were erupted in at least two different periods. The oldest lavas, ash-beds, and. agglomerates seem to have been extruded contemporaneously with the Carboniferous limestones, cherts, and slaty rocks, and have since, through regional metamorphism, been converted into massive and schistose greenstones which often keep their porphyritic structure more or less plainly preserved. No chemical study has been made of these older voleanics, and microscopic analysis is generally helpless in the attempt to refer them to definite types of lava. From their general habit and from the nature of the alteration and metamorphic products it appears probable that the whole series of Carboniferous extrusives should be classed with the common augite andesites and basalts. In his reconnaissance of the region during the preparation of material for the Trail sheet, McConnell recognized the Carboniferous age of these rocks and called the more massive, porphyritic *D. P. Penhallow, Transactions, Royal Society of Canada, ser. iii, Vol. 1, pp. 290 and 329, 1908. } 25a—vol. 11—214 324 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 phase ‘ augite porphyrite.’ One of the chief difficulties in mapping these rocks lies in the fact that the distinct and much younger augite latites are extremely difficult to distinguish in the field from the older augite andesites. There are, moreover, true augite andesites and basalts belonging to the younger series of lavas and the problem of differentiating them from the Carboniferous lavas is in many cases not to be solved. Since, therefore, most of the voleanic belt has defied clear-cut division on the map, the writer has followed McConnell and Brock in colouring under one legend, the ‘ Rossland Voleanie Group,’ most of the voleanie formations occurring in the Boundary belt between the Salmon river and Christina lake. Between the Columbia river and Christina lake the larger part of the voleanic masses have been found to belong to the family of latites, although there are some flows of true basalt and augite andesite associated with them. In the Beaver Mountain region there is a considerable area of relatively unaltered lavas and tuffs which nowhere seem to have any latitic phase. Chiefly because of their relatively fresh and recent appearance, Brock has already separated this series of voleanics, and he has given the series the name, ‘ Beaver Moun- tain Group.’ The petrographic distinction just noted further justifies our fol- lowing Brock in his mapping, and this part of the whole volcanic area will be separately described, as well as separately mapped in the accompan'ying sheet. Tf, in the future, the Rossland voleanic group can be analyzed with sufficient accuracy to permit of its subdivision on the map, it would be appropriate to veserve the name ‘ Rossland Volcanic Group’ for the latitic lavas and asso- ciated pyroclastics, for these seem to be the dominant extrusives of the area. PETROGRAPHY OF THE LAVAS AND PYROCLASTICS. The writer has collected about one hundred specimens of the freshest and most typical rocks of the volcanic belt and from them about eighty-five thin sections were cut. It was not until these had been microscopically examined that the lithological diversity of the lavas became fully apparent. Seven varieties of latite, olivine basalt, olivine-free basalt, augite andesite, and possibly picrite (corresponding to harzburgite among the plutonic rocks and described among the latter) have been recognized among the less altered lavas. The most abundant types are probably the augite latite and biotite-augite latite. These are respectively transitional into olivine-augite latite and biotite latite. Hornblende-biotite latite and hornblende (-augite) latite and a specially femic augite latite are of more local occurrence. The true basalts are far less common than one would suspect in the field, since so many of the latites have basaltic habit. True augite andesite is probably more abundant than the basalts. Augite Latite—Massive lava belonging to this variety was found at widely. spaced localities, among which are specially noted the area between Castle moun- REPORT OF THE CHIEF ASTRONOMER 325 SESSIONAL PAPER No. 25a tain (southeast slope) and Record mountain ridge, the divide between Malde and Little Sheep creeks, and the bluffs on the west side of the Columbia river about four miles north of the line. The following brief description of a typical, relatively unaltered phase relates to one of the younger flows occurring on the unnamed conical peak west of the Murphy creek-Gladstone trail and about two: miles north of Stony creek. The voleanic rocks are there exceptionally well exposed above tree-line, where thick sheets of highly porphyritic latite alternate: with more basaltic sheets and with coarse agglomerates composed of these lavas. The latite when fresh is a deep greenish-gray to almost black rock bearing abundant phenocrysts of tabular plagioclase up to 3 mm. in greatest diameters: and of smaller, stout prisms of greenish-black pyroxene. Microscopie examination shows that the rock is uncrushed, the phenocrysts: being unstrained and almost perfectly unaltered. The plagioclase is the more: abundant. On (010) and in the zone of symmetrical extinctions for simul- taneous Carlsbad-albite twins, individual erystals give extinction angles appro- priate to the series from labradorite, Ab, An,, to bytownite, Ab, An,. Occasionally one of these basic individuals is surrounded with a narrow rim of orthoclase. The average plagioclase phenocryst has about the composition of labradorite, Ab, An,. The pyroxene is a common, non-pleochroic, pale greenish augite of diopsidic habit. The ground-mass has been somewhat altered, with the generation of uralite in small needles, zoisite in rather rare granules, chlorite, abundant biotite, and more sericitic mica in minute foils and shreds. Orthoclase was not certainly detected in the ground-mass, which was originally hyalopilitic, with plagioclase microlites embedded in glass. Magnetite and apatite occur in the usual well- formed crystals. A specimen collected at this locality (No. 543) and answering to the fore- going description has been analyzed by Mr. Connor, with result as follows. (Table XIX., Col. 1.) :— Table XIX.—Analyses of augite latites, Rossland district and Sierra Nevada. 1. La. 2. Mol SiO, 54-54 -909. 56°19 TiO, 96 012 69 Al,O, 18-10 177 16-76 FeO, 1-14 007 3°05 FeO 4-63 064 418 s MnO 10 -001 10 MgO 4-56 114 3:79 CaO 5-85 104 6-53 SrO 15 001 tr BaO 21 -001 19 Na,O ; 3°38 055 2.53 LCL O te ti: Gl Sale IR Coa PRI ces de Rar ea mie eee nD 544 058 44.6 OR AESTID IO Shp stern, eee ctec te en baich Hote a alters whe 10 aad 34 EE OkabovewllOSC ay cr cel) Meee ls sey Teel ile +50 oss “66 21 0 ete OMe oC OSCE SAC TR RO COM Seat MeL A eibee 46 004 55 100-12 100-02 SPs or eye ran aia arya stciteys Soh rere eka tea htaneare rey cee 2.745 326 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 The caleulated norm is:— Orthoclases is. (ee see we 32-25 Albitexedoe) ice). 26-20 Nephelite.. .. 1-42 Anorthite.. .. ... 17-79 Diopside.. .. .. .. .. . 6-87 Olivanesy eo cre 10-18 Ilmenite.. .. ..... 1-82 Magnetite.. .. .. . 1-62 Apatite.. . 1-24 eG Water.. . -60 ‘ 99-99 According to the Norm classification the rock enters the sodipotassie sub- rang, monzonose, of the domalkalic rang, monzonase, in the dosalane order, germanare. The mineralogical and chemical composition and structure all perfectly match the typical augite latite of Table mountain, California, as originally described by Ransome.* The analysis of the more basic phase of the Table mountain flow is entered in Col. 2 of the foregoing table. From the fresh rock just described all transitions to profoundly altered phases are represented in the area. The latite has often been transformed into a dark green, massive rock, still showing its porphyritic character by the presence of broken and altered feldspar phenocrysts or of uralitic pseudomorphs after the augite. For the rest the completely changed rock is, in thin section, seen to be a confused mass of epidote, calcite, quartz, chalcedony, chlorite, biotite, uralitic and actinolitic amphibole, zoisite, pyrite, etc., in ever varying proportion. Some- times, though not often, an amygdaloidal structure is preserved. This is not so much because it has been obliterated by metamorphism as because these lavas were largely non-vesicular when first consolidated. Augite-biotite Latite—This type of massive lava is at least as important in the area as the augite latite. As above noted, the two varieties grade into each other, and the only noteworthy persistent difference is the absence or presence of biotite among the original phenocrysts. Biotite also often occurs in minute, shreddy foils in the ground-mass but it appears to be generally of secondary origin. The phenocrystic biotite is of a deep, rich brown colour and has powerful absorption; its optical angle is probably under 2°. The other phenocrysts, the accessories, and the ground-mass have characters essentially identical with those of the augite latite. No perfectly fresh specimen of the augite-biotite latite was secured. One of the least altered ones, collected on the ridge joining Record and Sophie moun- tains, at a point two miles north of the Dewdney trail (No. 456), has been analyzed by Mr. Connor. It is a compact, deep greenish-gray rock with numerous small phenocrysts of labradorite (averaging about Ab, An,), biotite, and urali- tized augite. These minerals are embedded in an abundant, originally hyalopilitic, greenish base. The latter is chiefly devitrified glass. Its advanced *F. L. Ransome, American Journal of Science, Ser. iv. Vol. 5, 1898, p. 359. REPORT OF THE CHIEF ASTRONOMER SESSIONAL PAPER No. 25a 3a7 alteration has led to the formation of kaolin, uralite, sericite, epidote, zoisite, Orthoclase was apparently never indi- chlorite, carbonate, and a little quartz. vidualized. Mr. Connor’s analysis resulted a follows (Table XX., Col. 1) :— Table XX.—Analyses Fe Opa tell OSC se ovecieteies teiactone HeOvabovertl02@ sees ie eee ees BAS Ja nee Posie CGE Se SDseei vec. of augite-biotite latite. 1. 99-32 2-796 la. w searster ott tauren on Coy lemags SRLS GBRSBSSRRS In the Norm classification the rock enters she sodipotassic subrang, shos- honose, of the alkalicaleic rang, andase, in the dosalane order germanare. The norm is as follows:— Quartzoate tee Orthoclase.. .. . ADItCraaetem oc. oie Anorthite.. .. .. Hypersthene.. . Diopside.. .. .. eet MimOenitereins. os. ccces Miagnetiter.in. m0 Apatitesc iss. << EeORandsCOx. - In the older classification this variety is clearly a biotite-augite latite. In Col. 2 of Table XX, the analysis of one of Ransome’s types, that from near Clover Meadow, California, is entered. The alkalies are a little lower. in the British Columbia rock, but the respective differences are too small to cause doubt as to the classification. 328 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Augite-olivine Latite-——This type has been identified at only two localities in the Boundary belt. On Record mountain ridge it is interbedded with the chemically analyzed biotite-augite latite; it also occurs on the top of the broad ridge west of Malde ridge at a point about a mile and a half north of the Boundary line. The specimens collected at these places are comparatively fresh and are uncrushed. Macroscopically, there is little to distinguish these rocks from the more: common augite latite. The colour, grain, and general habit is the same. ‘T'he phenocrysts are augite, olivine, and labradorite (averaging Ab, An,). The ground-mass may be cryptocrystalline, devitrified-glassy, or microcrystalline, with greater or less development of microlitic augite and labradorite. The accessories and secondary products are the same as those in the augite latite, except that a little phenocrystic biotite is developed in the specimen from: Record mountain ridge. That specimen (No. 465) has been analyzed by Mr. Connor. The micro- scope showed that the augite is here somewhat uralitized and the olivine partly serpentinized, while the plagioclase is very fresh. The hyalopilitie base bears microlites of labradorite, magnetite, apatite, and possibly orthoclase; most of the ground-mass is, however, a glass which is turbid through the very abundant generation of sericitic mica and other secondary products. The specific gravity of three specimens from this locality varies from 2-700 to 2-751; the higher value is the more reliable since it refers to the freshest specimen. From the chemical analysis it is clear that this latite verges on augite andesite. Analysis of ‘augite-olivine latite. Mol SiO... 58-67 -978 TiO. 1-00 013 Al,O3.. 15-67 154 Fe,0... 2-85 -018 FeO... 3-28 -046 MnO.. ell -001 MgO.. 3:86 097 CaO.. 5-33 095 SroO.. -09 -001 BaO.. “11 001 Na,O.. 4:77 077 10 ee nce 3-08 -033 A OF att Ole 2) oS tec ee oul eaiere -02 leer H:OVabove 110°C... .. =... ane 54 foe BO Kstomicis’ ccisthieisluineiaifomiicion eis ar cioe jack ele Lrg 16 001 99-54 tS) E17 SO Ree TLE Set Ga MUR CIM Me AN SPreRA SIRS EE ein S' or 2-751 In the Norm: classification the rock enters the dosodic subrang, alerose, of the domalkalic rang, monzonase, in the dosalane order, germanare. “The norm is as follows:— REPORT OF THE CHIEF ASTRONOMER 329 SESSIONAL PAPER No. 25a (ON EES Br Reh: ey OEE He) MO CEE i a ER EIR Pie Pe er acm NRSC Ca ere tae aN Boe 3-90 OTEHOCTASO Re raise Siecle, vere Cae NES alpaca aneh Hercll OM amet MAU U RENE re Natal tol 18-35 INO ak Gor SAL Oe AO REE RR Galore Rist OUC CAL, SANE a adal sical 40-35 BATT O TG DEO icon ais rare Nps ara eal AeA RETO ee lalameet Ve tetal apaeme Me orci takes, ket fever recs 12-23 ITO PSIG Gre ers as ren omtenhs ahaa iMails dp euavobe rei tie vah irope el cler SVeMmmctoelc Cone eveyncel| apaNeless 11-02 MEY PETSEMEM Os ce ee ee alae eek ME TRSTOT VERS OWP in bias ak ae cae Sexe via ce Scatiee 6-78 IM IGM ra oY} a Bee aes yh eam Ena con ecreg beerd genera ea StS GO Ms aU) pie eseye | deat 5 4-18 UNDA HER | EMEA ena eu suey egos BRIO REN aR MM NS Rea AMranN DIC alE | SN 1-98 PASSED Re EWE Esa Al Se DS SUSIE LN ade MTS AN 81 2 eo ef oe sheiytere ee ee oe ee ee ee ee evelunie-a:| fetal) sefiel ie re)) evemtepe: leven e)e. ele 56 99-66 Hornblende-augite Latite—A fourth type was collected at the 3,100-foot- contour on the slope due east of Sayward railway station. It is a dark gray rock with conspicuous, lustrous, black prisms of phenocrystie hornblende in a eray-tinted ground-mass. The acicular hornblendes vary from 1 mm. to 4 mm. in Jength and are arranged in roughly fluidal fashion. They are accom- panied by a subordinate number of idiomorphic augite prisms, also pheno- erystic but first discovered in thin section. The ground-mass is a rather confused, microcrystalline aggregate of the same bisilicates and feldspar. In this ease there can be no question that orthoclase forms a large proportion of the ground-mass feldspar microlites, which for the rest are probably labradorite. Magnetite, pyrite, pyrrhotite, and a little titanite are ‘accessory minerals; calcite, chlorite, epidote, kaolin, and sericite are secondary products. This specimen (No. 557) is comparatively fresh. Its analysis, by Mr. Connor, resulted in the form shown in Table XXTI., Col. 1. Table XXI.—Analyses of hornblende-augite latite. 1. as 2a. Mol. SiO,.. 52-17 52°17 -870 Ti@s.: 80 010 Al,O.. 16-59 16-59 163 Fe,0; 8-32 1-86 012 e not det. 3°74 052 MnO “11 11 001 MgO 3-87 3°87 097 CaO 8-25 8-25 147 SrO 05 05 BaO 15 15 001 Na,O 3°91 3-91 063 20.. Oo cle tet crete Sterile EMS “ey AUREL ith 4-00 4-00 043 HO at 110°C.. 5 RTM a Seaae Svoniienen, Sei Wake lato upe ee 13 13 H,O above 110°C... SAMs ene og a nto moat 1-17 1-17 Ate Pony eo wo 6 On OO, ODE G0 NO. DOO Oe DOT OOO ant 'OlD 24 24. 001 CO... OC COMO Om OO O0n OO OO 00! io OO cron root Ot.) OL0 56 °56 econ FeS, and Fe,8,.. oe eo al et Hetetreseyiiere tl lee eer dleet sien’ eeese 2.31 101-69 99-91 SS Pass LTieeey sere a ee ste fegeh ei rates coldest lougeredisredtereciee 2-852 On account of the presence of pyrrhotite the ferrous oxide could not be directly determined. The proportion of this oxide was estimated, as shown in 8380 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Col. 2. First, an amount of Fe,O, representing suflicient Fe to satisfy the sulphur present, was apportioned. The sulphides of iron were arbitrarily considered as half pyrite and half pyrrhotite. The remaining Fe,O, was calculated to repre- sent the FeO and Fe,O, of this rock by assuming that these oxides occur in the average proportions which they have in other analyzed Rossland lavas (Nos. 456 and 543). The analysis, so recalculated, is entered in Col. 2; the corresponding molecular proportions are shown in Col. 2a. The norm was calculated from the values given in Cols. 2 and 2a, with result as follows :— Orthoclase see Ae aac eins leek tis uae eell seas ecta harsh chen ee PS Ae ure 23-91 MANDIGO re tories wish save Lace genta gall SG revcrat kitcar St Ul s aenkan os cmER y Seagate 23-06 yrs} 0) a2) be pk A a SRE VR Ee RGM aR a SS SA Mice 5.40 PAMVOTENICS Ms in uitveec vel pe este ee sake AU EU RAN OSes CSU Or, SRCHIULER TK ca eat 15-85 WD TOPSIAS Sie. bes Ulan crete iverat Marc. abies ina nskeciaracrou dest anata’ inal etoucha veloute ten Necp F cotuutens 19-71 Olive a OE TU AIA cl elie Pree AAG NS Be OL Se dO ety be Re 3-09 MT MONTES ye Ae VPN WR AV V nc Sk at Hee eh ea oy Zita Car er eas 1-52; Magnetite.. Sidi aber Ss. caved Peravaitve acts ee tae vu te ays Maen tea eae S SGD ae IN Susp ete 2-78 Apatite.. .. SsUMNe NaN Ee ata e hee Se ere Niels Tera eee So uherstie cere ae 31 Pyrite and pyrthotite.. Base UG Lape Aa toatl CaN UA tha JO recA RAN Dares EROR RICE Usa 2-31 H.O and Cco,. ee e ejere tele) leave iejye, sels) s) ot ele! yee fete) s-,0-0 / ese) /'e'10) Ve le:) ,e7ey 18.6) 1-86 99-80 According to the Norm classification the rock enters the sodipotassic sub- rang, monzonose, of the domalkalic rang, monzonase, in the dosalane order, germanare. According to the older classification the rock is both cnineralgerelne aaa chemically a hornblende-augite latite. It was nowhere seen to be vesicular but, on account of its persistent fine grain, it is believed to belong to a massive flow: rather than to an intrusive body. A somewhat similar porphyritic rock, perhaps intrusive, crops out on the Dewdney trail where it crosses the low ridge between Sophie mountain and (the western) Sheep creek. Orthoclase is very abundant in the ground-mass of this rock. Hornblende-biotite Latite-—A fifth type of latite was collected on the moun- tain spur running up from Bitter creek southeastward at a point four miles due east of the railroad station at Cascade. The rock crops out at the 3,300-foot contour as a massive, gray to greenish gray, porphyritic, non-vesicular trap, and seems to extend uninterruptedly along the ridge to the 4,300-foot contour, where it is interbedded with hard bands of fine basic ash. Continuing southeastward to the Boundary line, the same lava is seen interbedded with coarse quartz con- glomerate. This type of lava was not identified at any other locality. The phenocrysts are brown biotite and green hornblende, the former pre- dominating. The determinable feldspar, averaging labradorite, Ab, An,, is confined to the ground-mass where it forms minute, tabular, twinned crystals in great number. A green shreddy biotite of low absorptive power and evidently of different composition from the phenocrystic mica, is extensively developed among the plagioclase microlites. This green biotite also forms complete pseudo- morphs after the hornblende phenocrysts, so that it is doubtful that any of the mica of the ground-mass is original. Ortheclase was not observed and it is REPORT OF THE CHIEF ASTRONOMER 331 SESSIONAL PAPER No. 25a practically impossible to determine the character of the original ground-mass, so great has been the alteration of the rock. The other secondary products, as well as the accessory minerals are like those in the augite-biotite latite, to which the hornblende-biotite latite must be chemically quite similar. Biotite Latite.—A sixth type represents a lava which is macroscopically like a mica andesite, but under the microscope shows features relating it to the latites just described. In its present condition it is a greenish-gray rock of decidedly lighter tint than the great majority of the Rossland lavas. Biotite and labra- dorite (averaging about Ab, An,) are the only phenocrysts. The base was probably once largely glass in which microlites of labradorite and more irregular ones of (probably) orthoclase were embedded. The ground-mass is now abundantly charged with secondary sericitic mica and some quartz which is doubtless also of secondary origin. This rock has not been chemically analyzed but the analysis of a fresh speci- men would correspond to many mica andesites which are rich in potash. In view of the intimate association of this type with the undoubted, analyzed latites, it seems best to regard the rock as a salic latite rather than a true andesite, though it must be on the border-line between the two species. Femic Augite Latite—Finally, an altered lava which seems to represent an opposite pole in the differentiation of the latitic magma, was found on the eastern slope of the hog-back ridge in (the western) Sheep creek valley. A second but more doubtful occurrence was noted on the ridge between Boundary monument No. 170 and the Coryell batholith. The chief mineralogical differ- ence between this type and the analyzed augite latite consists in a great increase in the number of augite phenocrysts, a corresponding decrease in the abundance of labradorite phenocrysts (which may entirely fail in the thin section), and apparently a decrease in the relative amount of the ground-mass. The accessory and secondary minerals are the same as those noted for the augite latite; orthoclase was not observed in the ground-mass, which in all the collected specimens has largely gone over to green biotite and sericitic mica. Comparison with Sierra Nevada Latite and with Average Monzonite— Before noting the other types of lava in the Rossland group it will be instruc- tive to review their classification in terms of the chemical constitution of the original latites as defined by Ransome. In Table XXII., Col. 1, the average of the Rossland latites is given, and in Col. 2 the average of six typical latites from California. Col. 3 shows the average of all ten latites and Col. 4, the average of the twelve monzonites recorded in Osann’s compilation of chemical analyses throughout the world. The last three averages have been reduced to 100 per cent. In making the average of the Rossland latites the augite latite and augite-biotite latite were considered as of equal weight and their average was weighted as four against the average of the analyses of the hornblende- augite latite and olivine-augite latite which together were weighted as unity. This weighting corresponds approximately to the relative volumetric impor- tance of the different types in the Rossland district. 332 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 Table XXII.—Comparisons of latites and monzonite. 1 2 3 | 4 ae at ; Sierra Nevada | Average of | “verage of Rossland latites. type /latites: land 2 world- | | monzecnite. yg bi ANN pad Sn lich eM acca ing a la mee ea SIONS MIN srt noei Se yo t 56°52 58°70 57°85 5525 IT Oa Aseria Speedin st eels! sialon: | 1°00 1°06 1°02 “60 AU ORM aimee erie Frage et eth eel: 16°96 16°75 16°76 16°53 HesO acne Wey (ON | 1°10 3°00 2°44 3°03 FeO. 4°51 | 3°03 3°48 4°37 NEMO ee cite Durr tect we 14 | “06 | “09 “15 IVT oO) en aon tere eee 4°01 2°50 3°09 4°20 CaO Ms sate Aiea catia 5°93 5°02 5°53 7°19 POLO) eur area een tat ctor fees 13 “02 | 10D (yy (04 Se eee Ba Ole erence ee ucenarl 16 9 1D hohe otters Nia OR Rare alah ea ca A RA 3°36 Su 3°61 3°48 a BERNA Meal kT Le po eAln BER Aca ae 4°58 | 4°42 4°11 Bi Sees Tra arial AMA a a aR sala *30 1513} { : TE (0) ae I es ae RE 48 a) | 74 J Pe ee “31 46 | 38 | 43 DONE IAR TRE NESE uM “34 eset east idee 14 cae ge HeSptandliMerasie Mi fect. Coie SEE Te eae Oa eecee Apbetae S: | Sen EON eee MS RrG er oA ba 04.0 99°75 160°00 100° 00 | 100-00 The close correspondence of the Rossland and Sierra Nevada averages shows an essential identity of the magmas from which the respective lavas erystallized; the justice of correlating the Rossland rocks with the latites is clearly demonstrated. That latite should, as pointed out by Ransome, be considered as the extrusive form of monzonite is indicated in the comparison of Cols. 3 and 4. The two are not strikingly divergent at any point, yet there are differences which together form the exact analogue of the difference between the world’s average syenite and trachyte, or the difference between the world’s average granite and rhyolite, or, in fact, between the world’s averages of any of the principal plutonic types and its generally recognized effusive equivalent. In all these cases (as proved by the writer through actual calculation; see chapter X-XIV.), the effusive rock is the more salic and some- what more alkalic; magnesia, lime, and iron oxides are characteristically lower in the surface lava than in the corresponding plutonic. In all these eases it would seem as if magmatic differentiation tends to be more perfect when magma approaches and reaches the earth’s surface, the more salic pole naturally developing at the top of the voleanic vents where it may be erupted as true surface lava. Without further discussing this theore- tical point we may conclude that petrography will gain by accepting fully Ransome’s highly useful conception of the latites as forming a group as important among lavas as the monzonites are important among the plutonic types. REPORT OF THE CHIEF ASTRONOMER 333 ‘SSESSIONAL PAPER No. 25a Augite Andesite—At two localities in the volcanic area, lavas belonging to the common species, augite andesite, have been identified. This rock may occur at many other points but its macroscopic similarity to the augite latite makes its discovery very uncertain. As already noted, the writer believes that this type as well as the true basalts are subordinate to the latites in the region covered by the Boundary survey map. A specimen belonging to what seems to be a massive flow was collected near the Coryell syenite contact on the ridge running northward from Monu- ment 171. It will be observed that this ridge is just west of the body of enstatite-olivine rock which is mapped as harzburgite but may represent a picrite, v.e., an extrusive form of the harzburgite magma. The augite andesite has, notwithstanding its altered character, all the ear-marks of this species of lava. The phenocrysts of augite and labradorite are embedded in a much altered ground-mass in which mierclites of those minerals can be detected as the essentials. The alteration products are uralite, chlorite, and quartz and thus differ essentially from those which are so characteristic of the latites. The evidence is quite clear that the ground-mass is not rich in potash. True augite andesite was also found to compose most of the blocks in a yery coarse agglomerate capping the ridge lying between Monument 172 and the confluence of Santa Rosa creek and (the western) Sheep creek. The larger blocks are there from three to four feet in diameter. A few fragments in the breccia are exceptional for this voleanic series in ‘being of acid composition, a biotite-quartz porphyry. Basalts—A typical olivine basalt was discovered on Mt. Tamarac, the broad divide between Malde creek and Little Sheep creek. It seems to form there a very thick and massive flow interbedded.in specially voluminous basic breccias. The phenocrysts are labradorite, augite, and olivine. The ground- mass is the usual holocrystalline aggregate of augite and feldspar. The feldspar is here much more altered than the femic minerals; this is just the con- trary of the rule with the latites, in which the feldspars are almost always not so badly altered as the augite, hornblende, or olivine. In the col on the trail southwest of Lake mountain an equally typical olivine-free basalt forms at least two thick flows separated by a two-foot layer of basic tuff. The contact-planes show here a strike of N. 10° E. and a dip of 75° to the westward; the series has evidently been greatly deformed at this point. The distinction of this basalt from the augite latite is easily made, for the fairly fresh ground-mass is the typical diabasic. A very similar rock, though distinctly vesicular, was collected at the edge of the volcanic area on the west side of Twelve-mile creek; it may, however, easily belong to the series of lavas included in the Beaver Mountain group. Flow of Liparitic Obsidian?—Throughout the whole area covered by the Boundary belt in the Rossland mountains, acid lavas are extremely rare. Frag- ments of biotite-quartz porphyry, probably a liparite of extrusive origin, are. as we have seen, enclosed in the coarse agglomerate at one point. The enly other 334 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1912 possible occurrence of acid lava observed by the writer was noted as a 30-foot intercalation in the Cretaceous (?%) argillite at the crossing of the Boundary line and Little Sheep creek. This is a white, aphanitic, massive rock showing a fairly distinct banding which in the field was taken for bedding. The rock weathers yellow to brown. Under the microscope the one thin section made from the rock showed no certain proof of the origin of the rock but the general appearance was that of a devitrified, partially spherulitic obsidian. The very small spherules seem to be poorly developed radial aggregates of quartz and feldspar and their matrix is a very fine-grained granophyric intergrowth of the same minerals. No other minerals have been certainly determined in the thin section. The banding may be a flow-structure. Since its characters are obscure and largely negative the writer must regard his reference of this rock to the acid obsidians as tentative. Tuffs and Agglomerates—Most of the area occupied in the Boundary belt by the Rossland voleanic group is underlain by massive flows of the latites, andesites, and basalts. A considerable tract, estimated as covering at least fifteen square miles, is, however, underlain by a thick, more or less continuous mass of coarse volcanic agglomerate. This pyroclastic composes the majority of the outcrops between Lake mountain on the east and the top of Sophie mountain on the west, besides extending for several miles along Record moun- tain ridge, northward, from the Boundary line. The constituent fragments are angular to subangular, ranging in size from dust-particles to blocks four feet in diameter. The deposit is usually . without stratification but consists of a tumultuous, massive aggregation of fragments which were evidently never sorted by water-action. Most of them are composed of augite latite, biotite-augite latite, or, to a less extent, of basalt and augite andesite. Besides these, abundant angular blocks of fossiliferous white crystalline limestone occur in the breccia throughout the whole eastern slope of Sophie mountain, and are likewise conspicuous in the breccia on the eastern side of Little Sheep creek valley. A basie agglomerate macroscopically similar to the Sophie mountain type but lacking the limestone blocks, crops out three miles to the westward, between the Boundary line and Santa Rosa ereek. As already noted, microscopic examination of the fragments showed them to be chiefly augite andesite, with a notable proportion of blocks of dark coloured biotite-quartz porphyry. These breccias bear numerous intercalations of the massive lava flows, thin basic ash-beds and a few, thin beds of black, carbonaceous shale. In a few localities the dip could be taken; in general it is high and ranges from 70° to 90°, showing that the whole group has been heavily mountain-built. DunITES CUTTING THE ROSSLAND VOLCANICS. At various points the andesites encircling the Coryell batholith within the ten-mile belt are cut by dikes and irregular masses of dunite, now partly serpentinized. The largest body occurs on Record mountain ridge, one mile REPORT OF THE CHIEF ASTRONOMER 335 SESSIONAL PAPER No. 25a north of the Dewdney trail. It extends over the ridge downward into Little Sheep creek valley. The fresher specimens show the presence of much olivine and some undoubted chromite, but the rock has largely gone to serpentine, tale, and magnetite. A similar irregular mass occurs on the Red mountain railway west of Rossland. A large dike of rather thoroughly serpentinized dunite cuts the andesitic greenstones south of Castle mountain summit, and a five-foot dike of the same rock cuts the small stock of crushed granite immediately to the southward. Sinee this rock is very apt to escape detection among the old voleanics, it is fair to suppose that only a portion of the whole number of occurrences has been discovered. The region has evidently been the scene of fairly numerous intrusions of this very basic type. From the various local relations the dunite has, in part at least, been injected at a relatively late date, possibly as late as the Cretaceous or Tertiary, when it cut the breccias and traps of Record moun- tain ridge. DUNITE ON McRae CREEK. On McRae creek about three miles above its mouth, the section along the railway crosses 350 yards of a massive, ark, greenish-gray homogeneous intru- sive which proved, on microscopic examination, to be a dunite. It cuts biotite schist and a tough, old-looking andesitiec breccia. The body probably has the pod form. The olivine occurs in a fairly fresh anhedra varying from 0-4 mm. to 2 mm. in greatest diameter. The alteration products are tale, tremolite, magnetite, and a little carbonate, probably dolomite. No chromite could be recognized in thin section. An analysis of a relatively fresh specimen (No. 528) gave Mr. Connor the following result :— Analysis of McRae Oneek dunite. SiO,.. 41-36 TOD. none Al,O,.. 1-21 HesOrs: 9-18 FeO... not det MnO.. 10 MgO.. 42-90 CaO.. 1-34 SrO.. none BaO. none K,O.. eo ee 8¢ «8 ©0 86° 88 80 ©0 #8 089 2©0 0©8 #8 @8 © 8 © be ee ce 04 TAO PA ERI OSG eeceiey epee ea Nee oon eee ean tar sania tukater cna tata lesa traeud eituempantes 16 HE OVADOV El TIOSC Hate, fone ch, ane eta clearer nace UMS Clas RNS Los Mes ty 1-94 2 5e2 22 ee elstmrelaie s/obticlsapelallfeteilicle) Mekoutelaniisleiielisiisalai cele! 1ia'elsielelieset ie 04 Cco,.. 91091916!) 56/0)" oe), 0 le) 10-6) sea), © 8°, U0.6) (8 6) 010)! 8:6. (0:6) (ele;i )8.0, (ele) seve), jane) cose). evel ele 1-40 CrOr: os ae 00 6s 08 08 60 08 88 68 08 oe we te ee oe oe we 8 ew oe oe “15 Ss em elone eirinieleatin ellie. see eels iericdie) ele seie) (eat yielei- (eles lelenisexe, ele: Week) efemlerel celen iets 50 100-51 SDP Bye eae be etal a oeitete 12 ae oeree orate Ne veal lacavitakoe Heveil onal Cinec ebay ang atetegreniuat micah 3-160 336 DEPARTMENT OF THE INTERIOR 2 GEORGE V, A. 1942 The presence of sulphur interferes with the determination of the relative proportions of the iron oxides in this rock. The analysis clearly corroborates the microscopic evidence that we here have a common type of dunite. PoRPHYRITIC HARZBURGITE (PICRITE?). At the Dewdney trail south of the head-waters of Santa Rosa creek, the older andesitic traps of the Rossland voleanic group enclose a mass coloured on the map as harzburgite. It is a massive, deep green rock, bearing on its surfiace abundant cleavage-faces of idiomorphic enstatite, which is embedded in a compact base of olivine and its derivative, serpentine. Many outcrops are characterized by spheroidal weathering, and the rock has assumed a strong brown colour. Here and there it is sheared and thus locally converted into nearly pure serpentine. The enstatite phenocrysts measure 1 cm. or more in length by 1 to 2 mm. in diameter. Besides olivine the only other constituents are chromite and magnetite; the latter may be entirely secondary from the altered olivine. The enstatite is generally fresh but has yielded some secondary taleose material. The olivine occurs in unusually small grains, which vary from 0.02 mm. to 0-6 mm. in greatest diameter, with an average diameter of probably not more than 0-1 mm. This fine texture of the olivine ground-mass suggests that the mass did not erystallize under a heavy cover. In the field the mass was taken for a thick flow and it is quite possible that it does represent the lava corres- ponding to a peridotite. A second visit to the locality might solve this inter- esting problem of relations; meanwhile the rock may be called a harzburgite, and is described among the intrusives. Mr. Connor has analyzed the fresh specimen (No. 392) collected, with result as follows :— Analysis of porphyritic harzburgite (effusive?). Mol SiO,.. 42-99 -716 Oss. tr asic INO yee 1-11 011 MerOse. 1:87 012 FeO.. 5-91 082 MnO.. 05 Ha MgO.. 43:14 1-079 CaO. -10 -002 srO.. none : BaO.. none ate. AN ate Rae. FN ate Bi TINS 6 AS DO ta aime ee WT ee RCN eek a 29 005 LEG OE eal UACU Act Ot at Mem CRE HERES Gia: Ne (De aE grea: ee MEME Nyy I 13 001 Oa OSE aie sice Cee a RN wae ee wae esas Car a ae “51 BAS HEORabovecldlOcG eRe kn as See nega won nt cetera eames 4-00 229 QU 526 «8 oe ee ee oe ee te ee lle lle et lle tt lle hee le le 04. ecce 100-29 NS Dr yy Nace heave ra CMU ro hats fe MUI GUN St WR a Ge aC ge 3-075 REPORT OF THE CHIEF ASTRONOMER 337 SESSIONAL PAPER No. 25a The calculated norm is :— ° Oxthoc&lasonsress Soe es ee eee ere ne “56 JA ORS Rots tea Re ase Ua hE eRe LG BRN G Ree iets SER he RY Lc lau Aber Bees Foxe 2-62 /NXGTEE RG Aes arenes SUE COST REE ame Pen ISEELS G LOE roenicae SON ao eat 5D .. 63-30 1-055 Ose +50 0 Al,O... 17-64 173 Fe,0.. 1-58 010 FeO.. 3-08 0438 Mn0O.. 4.7 007 MgO.. i 1-23 031 CaO.. Sica 5-03 089 SrO.. Re ote none. BaO.. at .05 Na.O. sted Batouite A 4-56 074 K,O.. seia Riglahu teres Uavorbnaverina teu Metsulcten rata eecen Menem me eases 1-16 013 H.O at 105°C.. ao TER ae ah wei CIS) 518 RRO aA AST Reg co 14 H.0 above 105°C... ie cal does 2S LR de RAR nam Nap area ey enon a 51 seis sO S lcshsimr clea mean Repaae deca Satis pak rei ese Meese tole 27 002 99-52 Sp. gr. 2-721 The norm was calculated to be:— Qa aTi6Z FALE, EO ARIE ERE ae eae IS Tok te 2) Ae) ER Le RA 18-24 Orthoclase. . oY. ye. Shsunsd) be adbe wo, SARA VE ace buds SER Lee gnenp pS arnt Aan 7-23 PATI EG Hg es tere gaics Releste tiee BSvocuh. Leche. maker ale iat Bene eae RTE tae el 38-78 Anorthite.. .. SCan cae ene mI Pa ge et eas ot AD ices 22-80 Corundums ties. Sie Bo Nay ys eS es ae cre oe ean 40 FLV DELS ONG y7. whe less id Seas eases aa US tae Payee ee ee een 7:59 Migiomobite se isc ta oy oie eu eal cee EE, Hel Rina etree aE ene 2-32 Mir enites Ay 3) tess he, a A Oat ey BLE, 2 CONES ae a ee En “79 AAT TO ia cop) Gust sha Ah Salbayos. Pevsie Re SN Gan ROR ue cote RTS Ca eV Ae cP a 62 AW aE O Tu. Sys uscesee-a7e" tote ns sot eye nic Oise sad ke es se eaoaeTe Eee eds rues meee 65 99-42 The mode (Rosiwal method) is approximately :— Quarta eo Pes aa ait ave aye oe calla ee ater tet eos PN es ees ae me 27. Orthoclase. . REL AIM med Nee I REMC Scr io a a 7 ATHUOSING 25, sass Ne ion Kleclois asercla Macho Ne eae a ee ee 50- I BYLOLAR A= Spine Onn es MOR a A eh ot ea crtetANy Hei mete ao eat ete b Aig. 5. Hornblende.. BE iri set Erm tar eT IRE TE Be Hist IA ch ocE tema 4 Magnetite.. 3 Titanite. . Apatite.. 3 Epidote and zircon. REPORT OF THE CHIEF ASTRONOMER 445 SESSIONAL PAPER No. 25a In the Norm classification the rock is the dosodie yellowstonose of- the alkalicalecic rang, coloradase, in the persalane order, britannare. According to the older classification the rock enters the class of quartz-mica diorites but verges on typical granodiorite. Seven other specimens of the batholith as exposed to the westward of the Ashnola gabbro were studied microscopically. They were found to include yet more basic diorites and also types which belong to the biotite granites rich in plagioclase. The specific gravities of the seven specimens range from 2-644 to 2.775, averaging 2-706. Where strong shear-zones occur in the Western phase they are occupied by dark greenish-gray, fine-grained, fissile hornblende gneiss very rich in horn- blende and similar to the metamorphic filling of shear zones in the Osoyoos eranodiorite. Between these narrow shear zones the more normal rock usually shows mechanical granulation and fracture rather than extensive recrystalliza- tion. Roughly estimating the relative volume of each type, the writer has con- cluded that the Western phase is, on the average, a granodiorite which is very lose to a quartz diorite. At the western side of the exposed batholith where it disappears beneath Cretaceous sediments, the granitic rock is relatively un- erushed, poor in orthoclase and rather abundantly charged with phenocrystic biotite and with hornblende. Toward Park mountain the zones of intense shearing become more and more numerous. The rock then loses its porphyritic appearance and tends to be a gneissic biotite granite, in which hornblende is wanting and orthoclase has increased at the expense of the soda-lime feldspar. Near the long band of Ashnola gabbro the Western phase carries bands of rushed rock which is indistinguishable from the staple rock of the Eastern phase. : Eastern Phase.—FKast of the roof-pendant of Ashnola gabbro the batholith ' shows evidence of having undergone its maximum shearing and metamorphism. Tt there consists of narrow bands of highly micaceous gneiss alternating with parallel, much broader bands of less micaceous gneiss. These bands are generally more acid than the typical rock of the Western phase. A specimen fairly representing the average of the Eastern phase was collect- ed at a ledge 1-8 miles south of the Boundary line and in the middle of the zone ot the batholith composed of this phase (Figure 32). The rock is in macro- scopic appearance a light gray, medium-grained, somewhat gneissic granite, weathering light brown. Quartz, biotite, orthoclase, and plagioclase (probably andesine, near Ab, An,) are the essential components. Rare apatite, zircon, and magnetite grains are the accessories. A few reddish garnets are occasion- ally developed. There is seldom any indication of straining or crushing of the minerals constituting the band whence the specimen was taken. Microscopic study leaves the impression that the material of this and similar bands has been wholly recrystallized. The structure is now the hypidiomorphic-granular. This specimen (No. 1398) was analyzed by Mr. Connor with result as follows: 446 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Analysis of Remmel batholith, Eastern phase. Mol Si0,.. po R Wea 70-91 1-182 TiO... eis 20 003 Al,O,. . ee 16-18 159 Fe,0;.. Mois 51 003 eOe ae. AG 1-09 015 Mn0O.. eats 04 MgO.. ae Ys 009 CaO Sera 2-92 052 BaO 10 001 Nias OMe aes 1-33 021 KeOs. ELE ES itnae ROR AG Sit bk OES eRe cee TIO 5-53 059 H.O at 105°C... 5 SOON Hoe OO ue POOR Dn Os Boe OON On 03 H,O above 105°C... ER EO CN Man eet Weed MU, wai Maar ey her efi heaters 12 ee P.0,. . ae Pe Sein ens ERS ae Crm T SS Trp arts hve cy eR LN 11 001 99-44 SD eee ek eee Since Gaetvale ceceeileie cau eie: tole reremtie meron: 2-654 The norm was calculated to be:— MATER eh ore aed he sda cick aeiciw oe Lecter ta eles ince be okeeeia en Eee 34-68 Orthoclase.. Ye ate: twa’ slau, Ioetaeiet, eMC ale MRtE oEsaje Tia ny ee one On ISOM aTE ee 32-80 AM ai te et ae I IST 5 Setar ee ae aT Gh» ea 11-00: Anorthite.. sar Ue hove Grater vce ver mateclderty sel a elo laveren husks’ Met Den pasier eee ee eee 14-73 Coruna wm ss rte recay eet eager ee cee oe ao ae eee 2-75 Fy PELStHeNie ese. chee AN ete eee ee ny ita Se eee Re ae 2-09 Magne titer oiiiu oitetepecs ici asa aeioteo hele lel e aces bie ace eT ee eee 70: I MMON TE Gas cick face asec salto e cet OL ee eS es OE ahs 46 AD ALLO roo werk ot nomcerets 1 sve rere © Benne ee. Pelee Tyee la Ri taomaee 31 Wiaterace sua ol Str SRG Pe ETS ie eta ros a be emer 15 99.67 The mode (Rosiwal method) is roughly :— QUALEZ eS ie wie ae eae) wie Sie, SiS ood ee kae KCL Ln cme eae ieee 34-3 OTthoclase sks ei. esa ce es, Sate nore aie Rael lean? Coe eae 37-1 IANGOSING HS Fix isre Foe hase es cio cla eel eee ieee eee 2565 Biotite.. 2:3 Magnetite. . A 5 Apatite, zircon, “and epidote.. 3 100-0 In the Norm classification this rock is transitional between the, as yet, unnamed dopotassic subrang of the alkalicalecic rang, coloradase, in the per- salane order, britannare, and the corresponding, likewise unnamed subrang of the order, columbare. In the older classification the rock has the chemical and mineralogical composition of a common biotite granite. It is, however, impro- bable that this type is an original product of crystallization from the batholithic magma. The specifie gravities of four fresh specimens of the less micaceous bands of the Eastern phase vary from 2-644 to 2-654, averaging 2-651. These narrow limits of variation agree with the microscopic study of the same specimens in showing that the lighter bands are relatively uniform in composition, REPORT OF THE CHIEF ASTRONOMER 447 SESSIONAL PAPER No. 25a The darker bands have not been systematically examined with the micro- scope but their field habit is that of common mica gneiss, often passing over into feldspathic mica schist; they never seem to carry any hornblende. They occupy probably no more than five per cent of the area covered by the Eastern phase. These zones were regarded in the field as located along planes of maximum shearing. They accord very faithfully in attitude with a strike of N. 2° to 25° W. and a dip nearly vertical, but sometimes 75° or more to the east-northeast— structural elements induced by regional orogenic»>movements in the Cordillera. It is improbable that the banding represents peripheral schistosity about the Cathedral batholith. The chief reason for excluding this view is that peripheral schistosity is lacking in the great Similkameen batholith, which is also cut by the Cathedral granite. It appears, on the other hand, that the Remmel bath- olith was already crushed and its banding produced before either the Simiil- kameen or Cathedral granite was intruded. Interpretations of the Two Phases.—Three interpretations of the two phases are conceivable. They may be supposed to be distinct intrusions of two differ- ent magmas; or, secondly, original local differentiation products in the one batholith; or, thirdly, distinguished in their present compositions because of the unequal dynamic metamorphism of a once homogeneous magma. Against the first view is the fact that the two phases, where in contact, seem to pass insensibly into each other. Jn favour of the third view are several facts which do not square with the second hypothesis, and the writer has tentatively come to the conclusion that the third hypothesis is the correct one. Among those facts: are the following: 1. The Eastern phase covers that part of the Remmel body which has suffered the greatest amount of dynamic stresses exhibited either in the Remmel or in any other of the larger components of the Okanagan composite batholith.. It has been seen that the less intense though still notable dynamic metamor- phism of the Osoyoos granodiorite led to the special excretion of most or all of the hornblende, apatite, magnetite, and titanite from that rock and the secre- tion of those leached-out compounds in the free spaces of the shear zones. The biotite was similarly segregated, but its mobility was found to be considerably less than that of the hornblende. If the metamorphism had been yet more energetic in the Osoyoos body, the more soluble compounds would have been carried away completely and the whole would have crystallized in the form of acid biotitic gneiss banded with especially micaceous schists in the zones of maximum shear. Such appears to the writer to be the best explanation of the Eastern phase of the Remmel batholith. 2. The composition of the rock and the fact that, as above mentioned, it seems to have been thoroughly recrystallized into a strong, well knit, banded gneiss without cataclastic structure agree with this view. 3. The conclusion is substantiated in the study of more moderate shearing in the Western phase itself. There the strongly granulated and not recrystal- 448 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 lized granodiorite shows impoverishment in the more mobile hornblende and accessories, which are segregated into intercalated recrystallized bands. Thus hornblende-free, crushed rock indistinguishable in composition from the rock of the Eastern phase occurs sporadically in many local areas within the normal crushed granodiorite of the Western phase. In summary, then, the Remmel granodiorite, gneissic biotite granite, biotite gneiss, biotite-quartz diorite, and hornblende gneiss appear to belong to a single hatholithic intrusion. The mean of the two chemical analyses corresponds to the analysis of a fairly typical granodiorite. In view of the greater volume of the Western phase it appears that the average original rock of the whole bath- olith was a granodiorite quite close in its composition to a quartz-hornblende- biotite diorite. This mass has been dynamically and hydrothermally metamorphosed with, iutense shearing in zones trending N. 20° to 25° W. Over most of the batholith so far investigated these zones of physical and chemical alteration are not so well developed as to obscure the essential nature of the primary rmagma (Western phase). The shearing and transformation are much more pro- found in a wide belt elongated in the general structural direction N. 25° W. Here the rocks are well banded biotite gneisses, the material of which is residual -aiter the deep seated, wholesale leaching of the more basic mineral matter from the crushed granodiorite (Eastern phase). Krucer ALKALINE Bopy. General Description—aAll the way from the Great plains to the Pacifie waters nepheline rocks are extremely rare on the Forty-ninth Parallel. The Boundary section is now so far completed that it can be stated that in the entire section the Kruger body is the only plutonic mass bearing essential nepheline; it is likewise the most alkaline plutonic mass. One of its principal characteristics is gréat lithological variability. It varies signally in grain, in structure, and*above all in composition. (Plate 39). All the varietal rock types carry essential feldspars of high alkalinity—miero- perthite, microcline, soda-orthoclase, and orthoclase. Nephelite, biotite, olive- green hornblende, a pyroxene of the sgerite-augite series, and melanite complete the general list of essentials. Titanite, titaniferous magnetite or ilmenite, rutile, apatite, and acid andesine, Ab, An, (the last entirely absent in most of ‘the rock phases), form the staple accessories, though any one or more of the coloured silicates may be only accessory in certain phases. Muscovite, hydronephelite, kaolin, calcite, epidote, and chlorite are secondary, but on account of the notable freshness of the rocks are believed to be due to erush- metamorphism more than to weathering. According to the relative proportions of the essential minerals, at least ten different varieties of alkaline rock have been found in the body. These are:— Angite-nephelite malignite, Hornblende-nephelite syenite, Augite-biotite-nephelite malignite, Biotite-melanite-nephelite syenite, Augite-biotite-melanite malignite, Augite-biotite-nephelite syenite, Hornbleude-augite malignite, Porphyritic augite syenite, Augite-nephelite syenite, Porphyritie alkaline biotite syenite. Pare 39. TYPES FROM THE KruGER ALKALINE Bopy. A.—Porphyritic alkaline syenite ; one-half natural size. B.—Nephelite syenite (salic variety) ; two-thirds natural] size. C.—Malignite ; two-thirds natural size. 25a—vol. ii—p. 448. REPORT OF THE CHIEF ASTRONOMER 449 SESSIONAL PAPER No. 25a There is a question as to how far this list of varieties actually represents the original magmatic variation within the body. The evidence is good that the augite and hornblende and a part of the biotite, along with the feldspars and nephelite, crystallized from the magma. It is not certain in the case of melanite which, in the Ontario malignite, as described by Lawson, appears to be a primary essential.* Microscopic study shows that much of the melanite in the Kruger rocks is of magmatic origin, but that perhaps much more of it has replaced the pyroxene during dynamic metamorphism. In such cases the pyroxene, where still in part remaining, is very ragged, with granular aggregates of the garnet occupying the irregular embayments in the attacked mineral. A further stage consists in the complete replacement of the augite by the melanite aggregates which are shot through with metamorphic biotite. These peculiar reactions between the pyroxene and the other components of the rock are wide- spread in both syenite and malignite. All the phases so far studied in this natural museum of alkaline types can be grouped in three classes—granular malignites, granular nephelite syenites, and coarsely porphyritic alkaline syenites. The malignitic varieties are always basic in look, dark greenish-gray in colour, and medium to coarse in grain (specific gravity, 2-757 to 2-967). The nephelite syenites are rather light bluish-grey in tint, medium- to fine-grained, and break with the sonorous ring characteristic of phonolite (specific gravity, .2-606 to 2-719). The-third class of rocks is much less important as to volume; they are always coarse in grain, of gray colour, and charged with abundant tabular phenocrysts of microperthite which range from 2 to 5 centimetres in length. These phenocrysts as well as the alkaline feldspars of the coarse groundmass are usually twinned, following the Carlsbad law—a characteristic very seldom observed in the malignites or nephelite syenites. (Plate 39, A) The nephelite syenites often send strong apophysal offshoots into the malignites, but such tongues are highly irregular and intimately welded with the adjacent basic rock as if the latter were still hot when the nephelite syenites were intruded. Moreover, there are all stages of transition in a single broad outcrop between typical malignite and more leucocratic rock indistinguishable from the nephelite syenite of the apophyses. Similarly, even with tolerably good exposures, no sharp contacts could be discovered between the coarse, porphyritic syenites and the other phases. The porphyritie rocks almost invariably showed strong and unmistakable flow structure, evidenced in the parallel arrangement of undeformed phenocrysts; these generally lie parallel to the contact walls of the body as a whole. The phasal variety of the Kruger body and the field relations of the different types seem best explained on the hypothesis that the phases are all nearly or quite contemporaneous—the product of rapid magmatic differentiation accompanied by strong movements of the magma. These move- ments continued into the viscous stage immediately preceding crystallization. (Plate 39, A). Three specimens representing as many principal types were submitted to Professor Dittrich for analysis. cate Gy, Lawson, Bulletin, Dept. of Geology, University of California, vol. 1, 190. 25a—vol. ii—29 450 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Augite-biotite Malignite-——The first specimen was collected at a ledge about 50 yards west of the contact with the older Kruger-mountain schists and 1,200 yards west of the small lake on the top of the mountain-plateau. This rock is dark-coloured, medium to fairly coarse-grained, and of gabbroid habit. (See Plate 39, Figure C). The essential minerals are augite, (with rare outer shells of olive-green hornblende), biotite, microperthite, microcline, nephelite, and prob- ably soda orthoclase. Apatite, a little titaniferous magnetite or ilmenite, and titanite are original accesories. Melanite is also an abundant accessory but in this case all of the garnet may have been derived from the pyroxene through crush-metamorphism. A little hydronephelite and more abundant muscovite, which seems to replace nephelite, are present as secondary products, but on the whole the rock is to be described as fresh. The order of erystallization among the original minerals is: apatite; iron ore; titanite; augite; feldspars; nephelite. The order is unusual in that the nephelite follows the feldspars. The chemical analysis of this specimen (No. 1100), by Professor Dittrich, resulted as follows :— Analysis of malignite, Kruger alkaline body. Mol. RSI Qian, ape Meee teste, tee Pie BL wn hres tt a ge a ocean ao Ate en ee 50-49 842 ETT G)BPMR Heat, | rere Mentaes See IGA NUE = the W)C Re RL AUS UNCC Cah Pk adie 92 O11 PANE Ose citettae Sais tale’ ta. ceatetneiatng colt Reve Seats nk noaiale aes ote sts 15-83 155 RO OT Notte sietk auc thoeet Malay Carcnliorel pomagerehy oiieteteotiiciomen ee EM Omer 6-11 -038 120) a eee eran acta oD RE ae Ra a Ar Re Soe DS ae ha 3:04 +042 IMO as Sink ee azaleas & dotnet, CEE OOF Peas SU 2a Re eee 11 001 IMG OMS pec ee i.) Bair tc tae gt nelle mean Tae aun Atel Ried 3-38 084 OFA 0 erie coe ote oe eee Ne ea CREE Oa - oe OI Creer nS BRNO 7-99 +143 TIN als One Moree alter Se tau aes calictete re re reg Marner al hana Se ye ht gece martes 3-12 050 ONE ES Stl Eras Salat te eas SE EN Lar, REO er beomies 6-86 073 ERE OS aC ett lC sec ssh amue iy matt Reowaciouaelolecreaels aoa chattel mereenree 29 BSE HE Onapovie: WOO. tpesieestuice [ous eer ate waste Recetas 1-20 agar 99-83 Seung, unger ks aL Wd me ee ep aa RU aS ENG phy cab car Age ea 2.849 The calculated norm is :— Orthoclage sk Lek eI ET eG Le ASO oe Ta ee Neen 40-59 | yh ee REE rea ee Bees Perma Pe nem men er er ms Guerre Fe i) Ba 7-34 ING PHCLTGE Ss Sragiis.«: xcs hesee FW desrseee ete Te Gs BRIM ae STOOD ae eae 10-22 UATOTEG FE GOs hee soe ic eae ta Ran hs ROLY a ee a 8-90 DiOpsidee 8. i see eee 2c ee ey I ee ee eee 18-14 Wo Weis tomuiite ses fc ia patted ce sree al ieice Rtg ie ial ceco Scene rie a Lae es coe a euan 1:97 MTC ETE ie scare: Cave eee te Sere eee a a CSS Oe re a ON ane ST te ae 7-42 Mmmrennters. Sisk Area. et AR ie tit anal. tea ee aia ea oy ene es 1-67 1 8 Ida hal] POR CSR Renae ete Ene be an oa tar Men beenK eaubME Ss A Ke GL Fg 96 PAUSE AERIS he APA DS ST ae OLE Ie RRS MRE EONS sour ees as Ob 93 REPORT OF THE CHIEF ASTRONOMER 451 SESSIONAL PAPER No. 25a The mode (Rosiwal method) is approximately :— Microperthite.. . aoa | Wacrocluness 5 fe ee 55 be SIRE Soda orthoclase.. .. . ser] INO DITO Soca rota one fs 36:5 IBiotitessass. access : 11-0 Melanite.. =. .. 9-5 Nephelite.. 54 AMatiter. coos eas. lies ae 1-0 Magnetite and titanite.. .. ... 3 100-0 In the Norm classification the rock enters the sodipotassic subrang, boro- lanose, of the domalkalic rang, essexase, in the dosalane order, norgare. According to the principles of the older classification the nearest relatives to this rock are found in the malignites of Ontario, as described by Lawson (see Table XXVIII, Cols. 5, 6 and 7). This Kruger mountain rock differs from the Ontario types chiefly in the fact that here potash greatly predominates over the soda. Though in this respect the rock is an extreme member of the group named by Lawson, it is, apparently, best classified as an augite-biotite malignite. Femic Nephelite Syenite-—The second analyzed specimen was collected near the contact with the Kruger Mountain schists at a point about 1,000 yards northwest of the locality where the first specimen was found. This second rock is a bluish-gray, medium-grained, somewhat porphyritic type. The phenocrysts are tabular crystals of microperthite, reaching 1 em. or more in length. Qualita- tively the mineralogical composition is like that of the specimen just described. Here, however, the femic constituents are decidedly less abundant, while the feldspars and nephelite have notably increased. The order of crystallization and the decomposition-products are, respectively, the same as in the first speci- men. In the thin section of the second specimen it was observed that the garnet and biotite interpenetrate so intimately as to suggest a primary origin for the former, though decisive proof of that has not been found. Like the first speci- men, this one has been somewhat crushed, so that a metamorphic origin of the garnet is quite possible. The rock powder gelatinized strongly on heating with acid, showing that nephelite is abundant. Optical tests seemed to show that some free albite here accompanies the other feldspars. Professor Dittrich’s analysis (specimen No. 1110) gave:— Analysis of femic nephelite syenite, Kruger alkaline body. Mol SiO. 52.53 “875 TiO. -07 001 Al,O, 19-05 186 Fe.0, 4-77 030 FeO.. 2-10 029 MnO.. 13 001 MgO.. 1-99 050 BaO.. -09 001 CaO. 5-75 103 SEO ee tree eet on epee nee AUN ee ts 19 002 25a—vol. 11—294 . 452 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Analysis of femic nephehte syenite, Kruger alkaline body—Continued. Mol. INICIO sees aIsth Riomelt ee autem iain PPL aU MS Isa RRS leh A it tale 4.03 065 TK Oey eae VaR aren Can Len Cea ean rIIR AL NAA MioAS en Mae Ws stay Up 7-30 078 PAE OAC STA OS Geren NE oS RA ee sa pnend a Sc A ane re eae 13 Brie EM Oe abo V6; AOC Oras ceotiee cases ice ole onaticeioa tr cee eater 1-49 ware P.O,;.. aie t 10 \@) siete, / ea, tieles Hele, feleh ieisr) telex tee! Weiss \velelimilel ele) ples tele Mreueluvene H +28 002 100-17 SDEUE Eisai sae oer oH a eae eltoaten le Haeobe acs NoMa 2-719 From the microscopic study it is very probable that the titanic oxide is notably higher than is shown in the foregoing table. Otherwise the chemical analysis corresponds well with the optical analysis. Rough calculation has shown that the garnet must be low in alumina and high in lime and iron, and is thus, as already suggested by its colour, a true melanite. The appreciable amounts of barium and strontium oxides suggest that some of the feldspar mixtures may in complexity rival the phenocrysts of the Rock Creek rhomb- porphyry. We have here one more illustration of the rule that these two oxides tend to occur in relatively high proporticn in the highly alkaline rocks. The norm ealeulated from the analysis is:— Ornthoclasore cee eee ie eo te sto eee Mee SUN ONE LMS or AEST UO RENN Soar 4337 Jl] cil ai As UR ey cate ane a RAG PR AU USTSONLiR ecm Oo ASS enemas mer Miata Mela con 11:00 no rbhite nn vce yilstmeic soul aiay women eigen tment 8 MAM sain eu a 11-95 NED Sh Ga sscc.ePeik vark city She eI oe Ire caret Lone REE TS 12-50 JO NK oy cai (- Peer neo te MR Am SAA er An Wet Mia ee Ap AMO oles ciel CAR Aa Bee UGE a 10-80 SWholllaisGommitientscencicy ctrtns ce yee baer ahaa rane eon ecie ttecok end Ser perc ool ARSED “70 Mia onetiten: sce celine cme: cacuksn orn Gee iecicee ca lmeecOee oe pacman oer era 6-78 ETO TEA eee se os ET ase a cea orga ae em ay 2 ke ce SF tee) ETE EEO alte hy a ernceD 15 MLS MIA EIbe ae Hester oS ck. oc. ckanesehoen ven weliehvede say io Seis ta aR eee oR 16 DASE COS ae casey oo cuticles oh go boly heii reBs Meee tcaty ci wince toraste Rated aA GaN ee nae pO 62 VViAI GOT aie sehretares Seat ane mere ee cots CE ORM Pe eae ae et Tk ah antes Enna te 1-62 é 99-60 The mode (Rosiwal method) is approximately :— 1 D1SHU0 ES) of es neti Oa PP RERE Seen yeas cra y aO Me Ptyeg yen SRI Crea 9 63-9 UNISDIVOLUG SS esate tis cic cte eles cree eT MEE tig: nae TELE Anibal oe ian cn tae 15-1 Bio tite tice fhe ae ee IRR Saati, Ca CPE UL pene ea Be a nena a 11-1 Mie Tamitesss .c%e. ck ascertis ed alvetors sible ete Wake eoS ocr) ees et ee eee I aioe 8-8 74) 0511 Pe car ca a Al a Te ee Te EpRenE ays Jac tau iiNet “6 PCANICO. iwc sacs he ate een lacs 5 100-0 In the Norm classification this rock must be classified with the first speci- men as borolanose. In the older classification it may be best named a biotite- melanite-nephelite syenite, transitional to malignite. Nephehite Syenite—The third specimen was taken from a ledge 2,300 yards due west of the southern end of the lake on the plateau and 1-5 miles north of the Boundary line. It represents a specially large, relatively homogeneous mass a mile long and 400 yards wide, which crowns the 4,200-foot summit west of the lakelet. This mass is made up of the leucocratic phase of the alkaline body. (Plate 39, B). REPORT OF THRE CHIEF ASTRONOMER 453 SESSIONAL PAPER No. 25a The rock is a light bluish-gray, rather fine-grained syenite, breaking with a sonorous ring. In the hand-specimen it shows a weak parallel structure, pro- bably due to flow in the late magmatic period. A few small hornblendes and many small feldspars twinned on the Carlsbad law, are arranged parallel to the planes of the flow-structure. Minute biotites can also be detected macroscopi- eally. Under the microscope the fairly abundant hornblende is seen to be a strongly pleochroic, olive-green variety of great absorptive power. The biotite is scarcely more than accessory. Nephelite, orthoclase, microperthite, microcline, and probably soda-orthoclase [extinction of 8° on (010)] are the light coloured essentials. The list of accessories includes melanite, apatite, and titanite. Iron oxides are absent or are present in but the barest traces. The rock is very fresh, even the nephelite showing little alteration. In this case the relations of the melanite point to its being a primary mineral. The rock has been little, if at all, crushed since it crystallized. The garnet is generally poikilitic, enclosing feldspar granules, and seems to have been the last product of crystallization. Or Go > & CO 100-0 The Younger phase approaches an aplitic relation to the Older. The con- tacts between the two were seen at several points; they are sharp, yet the two rocks are closely welded together, and it seems probable that the coarser granite was still hot when the younger granite was injected. Relation to Similkameen Batholith.—The Cathedral granite is unquestion- ably consanguineous with the Similkameen granodiorite. Apart from their obviously close association both in the field relations and in the geological 462 DEPARTMENT OF THE INTERIOR 2 GEORGE V.; Ay 1972 chronology, a near magmatic relationship for the two batholiths is indicated by the essential similarities in the optical properties of the respective minerals. These likenesses are observable in the quartz, microperthite, microcline, ortho- clase, biotite, and the accessories, as well as in the hornblende which, as we have. seen, is very rare in the Cathedral batholith. It would be a matter of the highest importance if one could-demonstrate the cause of this blood-relationship between the two batholiths. To say that they are magmatic differentiates is only to restate the petrogenic problem. The profitable questions are: What was differentiated in the two intrusive periods; and, what was the actual differentiating process ? These questions cannot be answered with assurance. All that seems possible now is to indicate the lines on which future investigation is needed. To do even that would anticipate part of chapters XXVJ. and X XVII. and the writer will here offer only one conjecture as to the relation between the bodies. The guess is based on the proved efficiency of density differences to explain splitting in a heterogeneous magma, like that which composed the Moyie sills; secondly, on the view that a mediosilicic magma tends to separate into the antagonistic gabbroid (basaltic) and granitic magmas, this separation taking place with special ‘rapidity just before solidification of the original magma could take place. Let us assume that part of the Similkameen granodiorite long remained molten or was, by whatever means, partly remelted, and then gradually cooled. It is conceivable that during the cooling the basic elements corresponding in total composition to a gabbro, would settle down, leaving a persilicic residue in the upper part of the magma chamber. To develop the hypothesis still further the basic differentiate is assumed to have the composition of the local Ashnola gabbro. Finally, it is assumed that just one-fifth by weight of the remelted granodiorite settles out, this particular proportion being that which would give a residue with silica very nearly equal to that in the Cathedral granite. The residue has thus been calculated and found to be fairly close in composition to the Cathedral granite in al] the other essential oxides. The result of the cal- culation is shown in Col. 3 of Table XXIX. Cols. 1, 2, and 4 respectively state the analyses of the Ashnola gabbro, the Similkameen granodiorite, and the Cathedral granite. PLATE 42. View of cirque head-wall composed of massive Cathedral granite. Scale given by man on the less jointed cliff. Felsenmeer on Similkameen batholith, about seven thousand feet above sea- = level, Okanagan Range. 25a—vol. ii--p. 462. a ! REPORT OF THE CHIEF ASTRONOMER 463 SESSIONAL PAPER No. 25a TABLE X XIX. Showing chemical relation of Similkameen and Cathedral batholiths. aaa= i 2 3 4 SVs 5 SAIS Se i Riera GB i 47°76 66°55 71°41 (BRPAL COE eho Aaron ae ths on i eer ae eS 2°20 *40 ‘00 16 TNA OAs tek acre name re pict an tren a tains PRE eee 18°58 16°21 15°65 15°38 TRA O ain cers helt Hee ea nO os eee re 2.19 1°98 1°92 25 TREO). ate ee tie SY meee UCU A 9°39 1°20 “00 ta WET O sc Racer cere Rte OMe nein rents SR eIab “29 "12 ‘O07 06 IVT Oeste eros coer SenSiy eye oie te tae alte ote aia 4°15 132, OL 33 CAO eae eae ee er eR gS ae ne haee Ml 9°39 3°86 2°47 WG3T SEOR A eee aise ender dies ie ‘03 ‘01 “00 None BES 2 0) eee a epee ie er Na toners ee ties ‘02 °03 °04 ‘09 Irie Osaaacds cae cee ciel AGIs NSD SECA ES 3°61 4°07 4°19 4°28 oO} 3 tiles a a ame See he Bem ea SE ey eae 47 2°84 3°44 4°85 lal (0) = vegas cones Gare ee Coe EE De rnrme nee ! “12 01 “00 “02 1B (O)A5: tice Ra en ee a SEE rae NPI SUP ei “53 24 "16 43 GAO) sch BOIS D berey ceca an Min a OE ee ah 15 “00 05 DEVS VVV An T CL S Teee ence ee a TT CR Se ic tet | AE eR ey SN Rea TNO HOB ahi ves ane accoreeers 99°51 99°59 100°00 99°95 1. Analysis of Ashnola gabbro. 2. Analysis of Similkameen granodiorite. 3. Result of subtracting one-fifth part of each ode shown in Col. 1 from the amount of the corresponding oxide in Col. 2, and recalculating to 100 per cent. 4. Analysis of Cathedral’ granite. The divergence of the oxide proportions in the calculated residue from those in the Cathedral granite is inconsiderable except in the case of potash and lime and even those differences are no greater than those often observed in two analyses from any one batholith in other regions. It may fairly be claimed that the grayitative separation of non-silicic and subsilicie constituents (gab- broid mixture) making up about one-fifth by weight of the Similkameen grano- diorite, would leave, in the upper part of the magma-chamber, a more silicious magma quite like that of the Cathedral granite. The composition of the less dense residue would be the same whether the separation took place through fractional erystallization or through true magmatic splitting. Obviously, little stress ean be laid on the actual figures resulting from the ealeulation just described. It has rather been intended as offering a concrete illustration of the hypothesis. On the other hand, the general principles under- lying the hypothesis are, in the writer’s belief, worthy of attention, for they seem to be among the most promising among all the principles of modern petrology.. The calculation shows that it is not unreasonable to retain the con- ception that the Cathedral granite is a gravitative differentiate from the Simil- kameen granodiorite magma, and that a magma allied to gabbro or diabase and thus matching the basaltic and other dikes actually cutting the Cathedral 464 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 granite, is the other pole of the differentiation. The chief difficulty of discuss- ing this view, as of all its competitors, lies in the limited nature of the data from the structural geology of the range. Herein lies the importance of a com- parison with the magmatic history of the Purcell sills and analogous injections of which the structural relations are well understood. Such comparison will be noted in the theoretical chapter X XVII. Dikes Cutting the Cathedral Batholith—Near the highest peak on-Bauer- man ridge the coarse Cathedral granite is cut by a small dike of typical olivine basalt. The dike is exposed for sixty feet, in which distance it varies in width from four feet near the middle of the exposure to less than two feet at each end. The basalt thus forms a lenticular mass, standing practically vertical. The strike of the dike is N. 35° E. and in the same quadrant as the average strike of the andesite dikes cutting the Basic Complex. The basalt is even more vesicular than the andesite mentioned. The middle of the dike is abund- antly charged with gas-pores one to three millimetres or more in diameter. These are commonly elongated parallel to the walls of the dike. For five or ten centimetres from each wall the pores are very rare and the rock is compact, as if by chilling. The basalt carries xenoliths of the adjacent granite and ot large quartz and feldspar crystal fragments also torn from the walls. The microscope shows that the basalt is exceedingly fresh, not even the olivine being essentially affected by weathering. In view of this freshness it is noteworthy that the vesicles carry no trace of calcitic or other filling. It looks as if they had never been filled with mineral matter. These facts together with the vesicular character of the lava, suggest that the basalt was injected near the surface and is therefore of later date than the unroofing of the batholith. In any case it is the youngest eruptive known to occur within the Okanagan composite batholith. The phenocrysts are greenish augite and colourless olivine, both of which sre abundant. The ground-mass consists of bytownite laths and augite granules, with a mesostasis of brown glass. Two small, parallel, lamprophyric dikes of pod-like form and less than three feet in maximum width, cut the Cathedral granite on the ridge 1,200 yards northeast of Cathedral Peak. These dikes, in contrast with the basalt, are much altered and it is difficult to diagnose them. The original constituents seem to have been plagioclase, green hornblende, diopsidic augite, and possibly some biotite. The grain is fine; the structure, panidiomorphie to eugranitic. The rock may be a greatly altered camptonite or else hornblende diabase. Park GRANITE STOCK. The Park granite stock measures 4 miles in length by 23 miles in width (Figure 33). This granite is coarse, unsqueezed, and in almost all respects resembles macroscopically the Older phase of the Cathedral batholith, of which the Park granite seems to be a satellite. Under the microscope the rock differs from the coarser Cathedral granite chiefly in the entire replacement of micro- REPORT OF THE CHIEF ASTRONOMER 465 SESSIONAL PAPER No. 25a perthite by orthoclase; so that this granite is a normal biotite granite rather than a soda granite. The greater homogeneity of the dominant feldspar may explain the fact that the Park granite is somewhat more resistant to the weather than the Older phase of the Cathedral batholith. A few prisms of dark green hcrnblende are accessory in much the same proportion as in the Younger phase oVVVV VV VV VV VY Oe PaHeArS ee OF RENT SOME MME NERD NE SENET m (op) yl m 20 < < < < < & < Bes << < < < < < < iV VV VY YY YOY VV OS n NYREN/IE N/E N/a AN/in NHN SEN/ ENS N/UNYieN/ N/E NZ H RPEMMEL Bip Avepa rls Oi cips tie Ye a Va We May Me NE On PAV/ WW AVP SAV SV AV) WAV ZS WV AVIV SV AV AY L\VVVVVVV VV VV VV VV | / Vv i ff ny PAR K Gis Rav Ag NG leilien Ber is AW WHEW WY (ote WP WRAY WP AY 4 av VV \ | Ve eV . v ) x00-SP GOS -§ | x00. LP FL0-6 { "BOLLS JO | *AQtAwty asvqueoiag | oytoedg esvioAy | osvisay GLO.G EdvIA ‘soylUEhG FE8.G MoAw ‘soqrusipeyAy “089.6-Ga9.6 “" "686 -6-G69-3 8 SEBO OOTY tere "89-99, Dh oe ee te e296 (C2056). BOE. “ S1T.¢ ‘oqtund 696.% “oaqqes) “APIABLY) OYLOadg F ul WOlYVIIVA poaaresq() "sox Ip eqtaAydi0g “77 **-sexTp oyIsopuR oULSN VW see wae "-*'8ax{Ip ¥[BSeq OUTATTO “eseyd Jeplg “YgTOYU VG opULIZ UVEUIEY]LULIG “"Apoq ourpexle tesn1y7 ‘oqisopur uayAuseg ‘Wstydiowejour 03 oup saseyd Tedrourrd omy £ yZTpOUZeq oy1101pouRAs JoUULEYT CEC Pa OC EO FORO HOm so Bo aekices ss (DOSOUG -LOULBYJOUL) = YAOYIVq eyltorpouras sookos(Q RN ee ecg ae ae (pasoydazoureqout) xepduiog ose sfiela! }).010-s)\e}0 eats O1gqes vlouysy “8 eqipuerqusoy Joy Ory SOAISNIZUL OIS¥q VyVdoyO “Apoq jo oure yy “UOIs “nd gut jo 234g cretssessssss) ag9048Ie[q) Sieve ves otter etelloaede. FTaceveehoewarene * 4 9U99038I9T J) % ix} is} oa ae rises ceeee sees “fymrsaog |S S = 8 i gS =. iss} ausctaehe aaee Se.) 11 oS oO 1a, 10 ‘AturereyT ayy Jo eso[g "77s *5"* gnoaoeqerg 1eMory) ‘ neeeeesess o1gggan ep weale (‘oIssetay, A[qissod ‘{SNOLOJLUOIBD) d10Z0a[Vq 99v'T "SIDMIG OUSNLIW WSL ‘ad VW [Bols0[0aK) “YRYOYZ0G Apsodmor upbounyO 24} fo saquaw Buowy suosr.uvdwos pup su0iyyja400o9—TX XX 21905 REPORT OF THE CHIEF ASTRONOMER 475 SESSIONAL PAPER No. 25a The theoretical bearing of this double law underlying the evolution of the Okanagan composite batholith will be discussed more fully in chapter X XVI. At present it may only be pointed out that the proved facts regarding the changes of acidity and density in the batholith are readily correlated with the view that post-Cambrian granitic magmas are of secondary origin and have been differentiated primarily through density stratification. On this view the basic (gabbroid or basaltic) magma is the original carrier of the heat, and the granites as a class have resulted from the interaction of the superheated basic magma on acid gneisses, schists, and sediments or on pre-existing granitic terranes. The Osoyoos-Remmel granodiorite is the product of the assimilation of acid Paleozoic and pre-Paleozoic terranes by invading basic magma. The Similka- meen batholith is largely the product of the refusion of the Remmel and Osoyoos batholiths. The Cathedral granite is a later differentiate of the magma which had partly crystallized as the Similkameen granite, or was a differentiate from the Similkameen granite when partly remelted. Among other purposes Table XXXI. will serve to show the correlation of the different formations described in the present chapter, excepting that the oldest of all, the Anarchist series (Carboniferous?), is not entered; nor is the Tertiary (%) conglomerate at Osoyoos lake noted, for its relations are not important to a treatment of the composite batholith. The principal cause of differentiation has been sought in gravitative adjust- ment, stratifying the magmatic couche according to the law of upwardly decreas- ing density (meaning, in general, increasing content of silica from below upward in the magmatic strata). Some authors hold that large-scale differentiation may develop basified contact zones by the diffusion of basic materials to the sur- faces of cooling. In chapter X XVII., an alternative and preferable explanation of the thicker basic contact-shells is outlined, again with primary emphasis laid on gravitative differentiation. It has not proved possible to demonstrate a law of increase of density with depth in the Similkameen granite. A series of fifteen fresh specimens of the rock were collected at altitudes varying from 1,200 to 8,050 feet above sea, and their specific gravities were carefully determined. The difference between the densities of specimens taken near or at the two extremes of vertical distance was found too small to allow of a definite conclusion, though the difference, small as it is, favours the law of density stratification. It must be remembered, how- ever, that the concentration of volatile matter, such as water vapour dissolved in the magma but largely expelled during crystallization, would possibly be greatest at the roof. The specific gravities of the crystallized rocks may there- fore not afford direct values for the total density stratification during the fluid state of the magma. Then, too, the observed relative uniformity of the Simil- kameen granite is a function of the scale of the subcrustal magma couche. It was unquestionably very thick; strong density differences are probably not, on any hypothesis, to be expected'in a vertical section less than several miles in - depth. ss The whole petrogenic cycle had already closed and the Cathedral batholith was solidified when the dikes of vesicular basalt and andesite were injected into 476 DEPARTMENT OF THE INTERIOR 2 (GEORGE VevAse one the Cathedral granite and the Basic Complex. These dikes represent essentially the same common basic type which forms the Ashnola gabbro and other of the oldest intrusives of the range. As the plutonic energies became exhausted in the formation of the Cathedral granite, the original heat-carrier has alone survived in the molten state and is capable of injection on the small, dike scale. In this feature the history of the composite batholith is similar to that of many other batholithic provinces, where the latest granite is diked by common basalt. or by its hypabyssal, chemical equivalent. Finally, it will be noted that the conditions of crystallization underwent a decided change during the long interval between the intrusion of the Osoyoos- Remmel granodiorite and the Kruger alkaline body. Magmatic stages la to 3b inclusive, afforded non-alkaline rocks rich in hornblende and carrying plagio- clase, either basic or of medium acidity, as the dominant feldspar. These bodies may be regarded as belonging to one consanguineous series. Magmatic stages 4 to 7 inclusive, afforded alkaline rocks bearing nephelite in the most basic phases and microperthite (orthoclase in 6b and 7) as the dominant feldspar throughout the series except in certain basified contact-zones. This group belongs to a second consanguineous series. The youngest of all the intrusives, the basalt and andesite dikes, belong to a third consanguineous series, closely allied in mineralogical and chemical composition with the earlier members of the first series. The first and third series each began with a magmatic type which is chemically equivalent to the commonest of extrusive lavas (basalt). The second series began with a basic magma which may have been a peculiar differentiate of the same original basaltic cowche or, as seems more probable, of that couche locally modified and controlled in its differentiation by some al.soroiion of sedimentary terranes into which the Kruger body was injezted. METHOD OF INTRUSION. Year by year the conviction has been growing ever stronger in the minds of many able geologists that such a batholith as any one of those here described has assumed its present size and position by actually replacing an equal or approximately equal mass of the older, solid rock. The Okanagan composite batholith repeatedly illustrates this truth. The writer is unable to conceive that the huge Cathedral batholith, for example, could have been formed by any process of simple injection, without leaving abundant traces of prodigious rend- ing and general disorder in the granites alongside. We have seen, on the contrary, that the Similkameen granite on the east is notably free from such records of orogenic turmoil, while the shear zones of the Remmel batholith on the west most probably antedate the Cathedral granite intrusion. The very scale of these great bodies is suggestive of bodily replacement; it is hard to visualize an earth’s crust which would so part as to permit of the laccolithic or chonolithie injection of a mass as great as a batholith. The problem will be discussed at length in chapter XXVI.; in which the many facts won from the study of the Boundary section will be correlated with the essential facts of the field in other parts of the world. . REPORT OF THE CHIEF ASTRONOMER 477 SESSIONAL PAPER No. 25a GENERAL SUMMARY. 1. At the Forty-ninth Parallel of latitude the Okanagan mountains and a part of the belt of the Interior Plateaus (the Interior Plateau of Dawson) have been carved by erosion out of an assemblage of plutonic igneous rocks which, in spite of the diverse lithological character of the rocks, should be regarded as an enormous single member of the Cordilleran structure. This plutonic group is named the Okanagan Composite Batholith. The details of its constitution are given in a foregoing résumé of its geological history. 2. This composite batholith was of slow development, beginning with small intrusions in late Paleozoic (or possibly Triassic) time, increased by great batho- lithic irruptions of granodiorite during the Jurassic, and completed by likewise immense irruptions of alkaline hornblende-biotite granite and biotite granite batholiths of Tertiary age, possibly as late as the Upper Miocene or the Pliocene. The satellitic Tertiary stock of Castle Peak in the Hozomeen range (see next chapter), is composed of normal granodiorite. 3. The local intrusion of a small, composite body of malignites and nephe- lite syenites; the regular basification along the batholith and stock contacts, giving collars of monzonites and diorites; and the sporadic appearance of certain peridotites (hornblendites and dunites) are probably all incidents of magmatic differentiation and do not directly represent the compositions of general sub- crustal magmas. 4, The composite batholith offers striking testimony to the probable truth of the assimilation-differentiation theory of granitic rocks. 5. The composite batholith includes two consanguineous series of intrusions. The older one is non-alkaline; the younger, alkaline. They are separated in time by the whole Cretaceous period, at least. 6. The two consanguineous series nevertheless appear to belong to one com- pound petrogenie cycle. Throughout the cycle batholithic intrusion has followed the usual law of decrease in magmatic density and increase of magmatic acidity with the progress of time. 7. Exposures of contact surfaces in the Similkameen batholith illustrate with remarkable clearness the downward enlargement of such bodies with depth. 8. The Similkameen granite bears three roof-pendants. Their distribution suggests that the present erosion surface of this batholith west of the Similka- meen river is not far from coinciding with the constructional, subterranean surface of the batholith. 9. The Osoyoos and Remmel granodiorites have been extensively metamor- phosed by orogenic crushing and its attendant processes. The metamorphism was both dynamic and hydrothermal. The granodiorites have been locally, though on a large scale, transformed into banded gneisses and schists. These changes have been brought about through the hydrous solution and migration of the original mineral substance of the granodiorites, especially the more basic minerals. The dissolved material has been leached out from the granulated rock 478 DEPARTMENT OF T HE INTERIOR 29GEORGE We Are oile and has recrystallized in strong shear zones to which the solutions have slowly travelled. The shearing and metamorphism probably began at a time when the Remmel batholith was buried beneath at least 30,090 feet of Cretaceous strata. 10. The intensity of this metamorphism and the development of the great Tertiary batholiths agree with other facts to show that post-Jurassic mountain building at the Forty-ninth Parallel was caused by much more powerful com- pression than that which is shown in the broader Cordilleran zone passing through California; there the Jurassic batholiths are relatively uncrushed and Tertiary batholiths seem to be lacking. 11. The problems of the Okanagan composite batholith illustrate once again, and on a large scale, the utmost dependence of a sound petrology upon structural geology. A suggested chief problem involves the relation of mountain- building to the repeated development cf large bodies of superheated magma only a few miles beneath the surface of the mountain range. The fact of this association ig apparent; its explanation is not here attempted. (See Chapters XXIV to XXVIII). SF ULV d *‘yootg yodaq jo Yydou ospta WoT] OSURY YISVYG Jo JlULIUNS SUOTe YsvoqyNos SuLfoor'T p- 478. a —vol. il 5 2 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER XVII. FORMATIONS OF THE HOZOMEEN RANGE. GENERAL DESCRIPTION. As the section is carried westward across the Pasayten river, we enter a new and more or less distinct geological province. One natural western limit of this province occurs at Lightning creek, but it is convenient to describe in the same connection the formations extending a few miles still farther west, so as to group within this chapter the various facts known about the geology of the Hozomeen range. At the Skagit river there is another abrupt change of formations. The Hozomeen range at the Forty-ninth Parallel is, in fact, an unusually well defined mountain group both in its topographic and its structural relations. (Maps No. 14 and 15). Within the limits of the Boundary belt the range is composed of a dominant sedimentary group of rocks, here called the ‘Pasayten series’; a more sub- ordinate, older group of sediments and greenstones, here named the ‘ Hozomeen series’; a volcanic member of the Pasayten series, here named the ‘ Pasayten Volcanic formation’; two small stock-like bodies of ‘ Lightning Creek diorite, which cuts the Pasayten series; a larger, typical stock of ‘Castle Peak grano- diorite,’ also cutting the Pasayten series;. a chonolith of syenite porphyry, cut- ting the Pasayten series and probably satellitic from the larger stock; and a few sills and dikes of porphyrite, cutting the Pasayten series and perhaps satel- litiec from the Lightning Creek diorite. West of the Pasayten river a small area of the Remmel batholith enters the five-mile belt. This plutonic mass is the local, unconformable basement of the great Pasayten series of rocks. The geographical order of the formations as they are encountered in carry- ing the section westward, will be roughly followed in the brief descriptions of ° this chapter. The oldest rocks, those of the Hozomeen series, crop out only in the ridge of Mount Hozomeen itself and will be considered last of all. PASAYTEN SERIES. Introduction—From the Pasayten river to Lightning creek at the eastern foot of Mount Hozomeen—a distance of twenty miles—the Boundary belt is underlain by an extraordinarily thick group of sedimentary rocks, here and there punctured by small bodies of intrusive igneous material. ‘These sediments form a large area which was traversed by Russell and by Smith and Calkins during their respective reconnaissances in the state of Washington. During his journey along the Boundary in the years 1859-61, Bauerman crossed an area of stratified rocks which doubtless represents the northern continuation of the 479 480 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 strata now to be described.* G. M. Dawson made a traverse up the northeastern headwater of the Skagit river and described the same body of rocks in greater detail.+ The area where crossed by Dawson is about fifteen miles north of the Boundary line. Though he measured one section over 4,400 feet thick and, from paleontological evidence, proved the ‘newer Mesozoic’ age of the series of beds—later referring to them as Cretaceous—, Dawson did not give a special name to the series. His brief description will be found to correspond quite closely to the following account of the sediments. The present. writer adopts the name ‘ Pasayten series,’ thus modifying somewhat the title given to this great group by Smith and Calkins. The change from the original name, ‘ Pasayten formation,’ seems to make one more appropriate to an extremely thick assemblage of strata which range in age from Lower to Upper Cretaceous. Stratigraphy—On the whole the Pasayten beds are tolerably well exposed, so that the succession can be made out with fair accuracy. : At the Forty-ninth Parallel they compose a gigantic monocline with its base at the Pasayten river and its uppermost beds forming the steep ridges north, south, and west of Castle Peak. Across the strike the monocline measures at least sixteen miles in width. West of Castle Peak the youngest exposed member of the series, a thick mass of argillite, is strongly folded and faulted, giving steep dips. The lack of well marked horizons in this folded belt has rendered it as yet impossible to state its exact structure. Consequently there is much uncertainty as to the precise nature of the general columnar section in its upper part. At Lightning creek the argillite is cut off by a profound fault which brings it into sharp, more or less vertical contact with the Paleozoic rocks of the Hozomeen series. Ju the Boundary section, therefore, the top of the Pasayten series is not visible and the youngest exposed bed seems to be truncated by an erosion surface. No other area of the series has been examined in detail and the columnar section ean be stated only in terms of observations made in the five-mile belt; such observations are necessarily incomplete. As the writer carried his traverses from the basal unconformity at the Pasayten river westward, he became truly embarrassed by the colossal thickness which characterized the successive members. The cumulative thickness in a plainly conformable and comparatively young formation seemed almost incre- dible. For this reason special care was exercised in the field to note any possible hints of duplication of strata in the great monocline. It was found, however, that such duplication could have taken place only to a quite limited extent. The upper two-thirds of the series is charged with conspicuous horizon-markers; these would inevitably be repeated visibly among the fine exposures of the rocky ridges, if important duplication through normal faulting or other means had taken place. With a conviction which increased greatly as the field work and then the office study progressed, the writer has concluded that the series must *H. Bauerman, Report of Progress, Geol. Survey of Canada, for 1882-3-4, Part B, p. 14. +G.M. Dawson, Report of Progress, Geol. Surv..Canada, for 1877-8, Part B; p. 105. REPORT OF THE CHIEF ASTRONOMER 481 SESSIONAL PAPER No. 25a total at least 30,000 feet in thickness. This is a minimum estimate, for the field sections as plotted show a total thickness of over 40,000 feet. The chief uncertainty resides in the determinations for the top and bottom members. As noted in the columnar section their respective strengths, namely, 3,000 and 10,000 feet, are estimated as the lowest possible minima. (Figure 34). The whole succession is shown in the following table :— Columnar section, Pasayten series. Member. T’ ee on LIithological Character. |Top, erosion surface. 3,000 |Gray te black argillite, bearing plant-stems and impressions of ammonite shells. 7,100 |Gray and green feldspathic sandstones with interbeds of black argillite ‘and thin lenses of conglomerate ; fossil plants and animal remains. 1,400 (Coarse conglomerate. 300 | Black argillite. | 3,500 Green feldspathic sandstone with rare argillitic interbeds; fossil plants and shells about 200 feet from the top. Fairly coarse conglomerate. 1,500 (Gray and green, feldspathic sandstones. 100 (Conglomerate. 1,100 (Gray and green, feldspathic sandstone. 600 Red argillite and sandstone. 10,000 | Very massive, medium-grained arkose sandstone ; fossil plants at about 900 feet from the top and also about 3,500 feet from the base. 1,400 | Volcanic agglomerate conformable to sandstone B. P RSH BAS RS i) S 30,200 |Base, unconformable contact with older Remmel batholith. * The voleanic agglomerate was crossed in four different traverses. Sufficient information was obtained to indicate its relations to the neighbouring forma- tions. The agglomerate forms a remarkably straight and clearly continuous band of nearly even width, crossing the whole five-mile belt in a northwesterly direction and thus parallel to the strike of the adjacent sandstone. Though the breccia at every observed outcrop is quite devoid of bedding-planes, there can be little doubt that it is everywhere conformably underlying the sandstone. It is regarded as practically contemporaneous with the lowest beds of member B. Within the Boundary belt the agglomerate rests on the eroded surface of the Remmel batholith. The petrographic character of the agglomerate will be described in a special section of this chapter. For a distance from the agglomerate the granodiorite is thoroughly decolour- ized and has the look of having undergone secular disintegration before the breccia was deposited. The depth of this shell of ancient weathering was measured near the Pasayten river and found to be about 400 feet. Such a depth means that the pre-voleanic surface was characterized by low slopes on which the rotted rock could lie and slowly increase at the expense of the fresh grano- diorite beneath. The straightness of the line showing the contact between 25a—vol. 1i—31 Exvostor Surface: aes J pet! |4325-4 H QoUmMnNDa 1430 a hs - ee ee pee0oe ec 2 eB M2 oe eee 2 w~—rccee = = = ETE | 4? g 18) 4000 8000 (2k Sees 7 ea ee a Scale in feet Figure 34.—Columnar section of the Pasayten Series, including the Pasayten Volcanic formation (member A). Approximate horizons of fossils indicated by collection numbers. REPORT OF THE CHIEF ASTRONOMER 483 SESSIONAL PAPER No. 25a agglomerate and granodiorite is a direct indication that the land surface was flat when the voleanic activity began. With the vulcanism or closely following it there was a strong down-warp- ing of the region. Great changes of slope must have occurred, for the agglo- merate is overlain directly by member B, a very thick sandstone essentially made up of the decomposition-products of the granodiorite which were now swept into the down-warp from an uprising area on the east. The resulting accumulation of arkose and feldspathic sands was immense and of long dura- tion. Striking characteristics of member B are its massiveness and uniformity of grain and substance. The massiveness is so great that even in large outcrops representing strata fifty feet or more thick, it is often difficult to find the bed- ding-plane at all. In such eases the writer was at first in doubt as to whether the rock were really detrital, so much did it simulate a decolourized granite. Careful search, however, always showed the presence of true bedding which was: best displayed in thin partings of dark shale in the sandstone. These shales and sometimes the sandstone itself were found to carry fossil plants; no further question was possible as to the nature of the whole formation. A few ripple- marks were discovered in the upper beds of the member. On both sides of the Boundary line the numerous readings of strike and dip showed close accordance all across the sandstone through the six miles from the Pasayten river to Chuchuwanten creek. The strike averaged. about N. 30° W.; the dip, about 48° S.W. The apparent thickness of member B is at least 15,000 feet. It is possible, however, that the strata have been in part repeated by a northwest-southeast normal fault running along the valley just east of Monument 81, and it has appeared safer to estimate the thickness from the simple monoclinal element between that valley and the band of agglomerate three miles to the eastward. Even this estimate gives 10,000 feet as the mini- mum. At the strong, canyon-like valley of Chuchuwanten creek there is such change of dip (though almost no change in strike) that another normal fault with up- throw on the northeast has been postulated and marked on the map. It is possible that some of the youngest beds of member B are represented only on the southwest side of that fault but they are neglected in estimating the mini- mum thickness of the member as given in the general columnar section. The sandstone is normally a light-gray, medium to rather fine-grained rock ; seldom showing the bedding-planes in the hand-specimen. It weathers gray to brownish-gray, rarely whitish. Excepting for the rare and thin interbeds of argillite already noted there are almost no variations from the monotonous character of the sediment; no conglomerate was found in this member. The sandstone is well consolidated and is often quite tough before the hammer. A typical specimen was sliced and examined microscopically. As expected from the macroscopic appearance the rock was found to be very rich in feldspar fragments. A rough estimate of the weight percentages credits about 30 per cent to quartz, 30 per cent to orthoclase. 35 per cent to plagioclase (andesine to 25a—vol. 1i—314 484 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 labradorite) and 5 per cent to biotite, titanite, epidote, and limonite. All but the epidote and limonite are of detrital origin. The feldspars are greatly kaolin- ized and were doubtless nearly as much altered before the fragments found their places in the bed. The biotite occurs in thin, ragged and crinkled flakes, quite like those which may be seen in micaceous sands of the present day. The specific gravity of the specimen is 2-625. The mineralogical composition of the sandstone is like that of the secularly weathered shell of the Remmel granodiorite below the agglomerate, member A. In both cases hornblende fails to appear, as if it had been leached out completely during the ancient weathering. Otherwise the important constituents of the Remmel batholith are all represented in the sandstone. There can remain no doubt that the sandstone has resulted from the destruction of thé batholith. It is probable that the various sandstones overlying member B have had a like origin, but, from the lack of microscopic analysis, the proof ot this has not yet been completed. Member C is exposed in but a small area, occurring at the northern limit of the Boundary belt on the eastern slope of the Chuchuwanten valley. Farther south it is faulted out of sight by the Chuchuwanten fault. This member is the most highly variegated portion of the Pasayten series. It consists of a group of rapidly alternating red argillaceous sandstones and grits; gray, feldspathic, often pebbly sandstone and grit; with red, gray and green conglomerate. The beds range from an inch or less to twenty feet in thickness. The pebbles of the con- glomerates are composed of hard, gray quartzite, chert, and hard, red and gray slate. Some of the larger, always well rounded boulders are as much as two feet in diameter. sa apes These beds of C have variable attitudes; the rapid changes are probably connected with the adjacent fault. A mile or more east of Chuchuwanten creek the member dips rather steadily about 20 degrees to the north and visibly over- lies member B. It is itself there overlain by 400 feet of member D, which on the north gradually approaches a horizontal position and is terminated above by an erosion-surface. The measurement of the thickness of member C, 600 feet, was made at this locality. Member D is lithologically like the great basal sandstone but tends to assume a dominant green colour. Member E is well exposed near the Boundary slash in a cliff overlooking Chuchuwanten creek on its west side. The conglomerate is of medium coarse- ness and seems to contain few pebbles not composed of gray quartzite or vein quartz. It is overlain by member F’, some 1,500 feet of green and gray felds- pathic sandstone of rather dark tints but essentially like the sandstone below the conglomerate LH. Member G is a 200-foot bed of conglomerate recalling EH in its general character but abundantly charged with pebbles of an andesitic nature—the only known occurrence of such material in the conglomerates of the series. Member H is not well exposed; so far as seen, it is a homogeneous, green, feldspathic sandstone. The estimate of the thickness, 3,500 feet, though so REPORT OF THE CHIEF ASTRONOMER ; 485 SESSIONAL PAPER No. 25a great, is believed to be a minimum. This member is best seen on the trail up Castle creek; there abundant, though not well preserved fossils, both shells and plants, were found at a horizon about 200 feet below the top of the member. These fossils will be described on succeeding pages. The overlying black argillite member I, is, so far as known, unfossiliferous. It very clearly overlies member H and underlies the conglomerate of member J. On account of its coarseness and great thickness—1,400 feet—member J is a very conspicuous element of the series. It was traced continuously from the summit north of Castle creek to a point well south of the Boundary line. Throughout that distance the conglomerate preserves a strike averaging about N. 22° W., and a dip of from 60° to 65° to the west-southwest. This steady behaviour of so prominent a member tended to make the structural study of the formation west of the Chuchuwanten comparatively easy. Its occurrence only once in the wide monocline has been a principal reason for believing that pro- nounced duplication of strata by strike-faults has not taken place. This conglomerate is usually coarse; the pebbles reach eighteen inches or more in diameter. They are highly diverse in character. The list of different materials is long, including: gray, banded quartzite; blackish quartzite; hard, black and gray argillite; gneissic and massive hornblende granite; white aplitic granite; biotite granite; syenite porphyry; amphibolite; fine-grained diorite; coarse hornblende gabbro; greenstone schist; sericite schist; aphanitie quartz porphyry; and a breccia composed of white quartz fragments with jaspery cement. This list shows that most of the staple pre-Cretaceous formations of the region are represented. Among the granitic pebbles were a considerable number having the composition of typical, fresh Remmel granodiorite or quartz diorite in its Western phase. Both the hornblende and the large, lustrous-black erystals of biotite in the Remmel are to be seen in these pebbles. The latter must have been derived from fresh, little weathered ledges. The cement of the conglomerate is a green feldspathic sand essentially like the green sandstones overlying and underlying this great conglomerate member. The cement yields rather readily to the weather, so that long talus-slopes of the weathered-out pebbles fringe the many cliffs where the conglomerate crops out. Towards the top the conglomerate grows finer-grained and merges into the very thick sandstone member K. This sandstone forms a continuous band cross- ing the Boundary belt. The band is a little over two miles wide and the average dip is about 45° to the west-southwest. The calculated thickness—7,100 feet— ig again enormous but it is a minimum. Three cross-sections of the wide band were traversed. In none of them was there any sign of repetition of beds nor any serious departure from the average strike and dip. The uniformity in the width of the band is another indication of the EbgINe 2 of strong faulting within the area covered by this member. The sandstone of K is a hard, green to gray, ee feldspathic 1ock much like those in members D, F and H. It is interrupted by numerous beds of black to rusty argillite and argillaceous sandstone and is itself often more argillaceous than the average sandstone of the older members. Thin lenses 486 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 of fine-grained conglomerate also occur’ at intervals. One of these, about 2,000 feet below the top of the member, carries fossil shells. A few feet away from the locality where the large shells were found, the sandstone encloses plant- remains. From the eastern end of the Castle Peak stock of granodiorite to the Lightning creek fault—a distance of nearly ten miles—the greater part of the Boundary belt is underlain by the black argillite of member L. The strata here show dips varying from 25° to 90°. As already noted it has not proved feasible to work out the folds and faults with entire confidence; in consequence, the thickness of the argillite is in doubt. It is known, however, that it must be at least 3,000 feet and may, as estimated in the field, be more than 5,000 feet. At its base it grades rapidly into the conformable sandstone member K. Member L is a rather homogeneous, hard, black or dark-gray shale, in which thin, green and gray sandstone beds are intercalated. The shale weathers gray and brown in varying tints. At three horizons,—one found opposite the mouth of Pass creek, another 700 yards south of the 7,860-foot summit overlooking the ereek, and the third on the ridge 1,000 yards east of Frosty Peak,—the shale earries fossil plants and ammonite impressions. Granitic intrusions have to some extent metamorphosed the argillite. The metamorphic effect is apparently most pronounced about the Castle Peak stock, - though the effects are nowhere very striking. The shale inclusions in the stock have been converted into hornfels of common type. Fossils Collected—It has been seen that the monotonous chain of failures in the many efforts to discover fossil remains along the Forty-ninth Parallel was seldom broken. The decided novelty of finding them at several horizons within the Pasayten series was specially welcomed, as these discoveries bade fair to clear up many points in the dynamic history of the eastern half of the Cascade mountain system and incidentally to throw light on the history and relations of unfossiliferous formations in the broad Columbia system as well. Many of the correlations noted in preceding chapters have, in fact, been made in the light of the analogies which may be traced between the structure and stratigraphy of the more easterly ranges and the more closely determined struc- ture and stratigraphy of the Hozomeen range. The conditions of field work during the Boundary survey did not permit of exhaustive collections at any point. As a guide to the future paleontological study of the Pasayten series the exact localities of the different collections of plant and animal remains will be noted. Each locality will be referred to by the corresponding specimen number. In connection with each the stratigraphic and paleontological details will be added. No. 1428. At the 6,750-foot contour 400 yards southeast of the 6,920-foot peak situated two miles north of the Boundary line and about three miles west of the Pasayten river. Stratigraphic position: about 3,500 feet above the base of member B. Sand- stone with shaly interbeds. REPORT OF THE CHIEF ASTRONOMER 487 SESSIONAL PAPER No. 25a Fossils: plants only; determined by Professor Penhallow as: Gleichenia gilbert-thompsoni Font. Glyptostrobus sp. Pinus sp. Salix sp. Horizon: Cretaceous of Shasta series; see Appendix B. No. 1430. At 4,200-foot contour, east side of Chuchuwanten creek canyon, about 400 yards north of Boundary slash. Stratigraphic position: about 900 feet below top of member B. Shale bands in sandstone. ‘Fossils: plants only; determined by Protessor Penhallow as: Cladophlebis skagitensis, n. sp. Gleichenia sp. Aspidium fredericksburgense, Font. Nilsonmia pasaytensis, n. sp. Cyaadites unjiga, Dn. Populus cyclophylla, Heer. Myrica serrata, n. sp. Quercus flexuosa, Newb (7) Quercus coriacea, Newb. Sassafras cretaceum, Newb. Dorstenia (2) sp. Horizon: Professor Penhallow writes: ‘Reviewing this evidence, we observe that there are eleven species of plants from locality 1480. Of these Dorstenia (?), which is of questionable character, and Pinus (sic), which is chiefly represented by seeds and may indicate any one of several horizons, need to be eliminated because not specifically defined. This leaves nine well-defined species, of which three are definitely Lower Cretaceous and six as definitely Upper Cretaceous.’ He concludes that this flora shows two well defined horizons within the Shasta-Chico series. See Appendix B. Nos. 1432-33-84. 4,700-foot contour, north side of Castle creek valley, four miles down stream from crossing of that stream and the Boundary line; just east of conspicuous band of thick conglomerate, member J. Stratigraphic position: about 300 feet below top of member H. Fossils: plants, determined by Professor Penhallow; animals, determined by Dr. T. W. Stanton. Plants: ‘The only specimen under number 1433 showed on one side, two small fragments of leaves which, from their obviously parallel venation, are to be regarded as belonging to some endogenous plant, the nature of which could not be determined. On the opposite side of 1433 is a single leaf of a pine.’ See Appendix B. 488 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Animal remains: “1482: Pecten operculiformis Gabb. Trigonia sp. Fragmentary imprint. Eriphyla ? sp. Small casts. Pleuromya ? sp. Fragmentary imprint. Rissoa? sp. A small obscure gasteropod with the general form and sculpture of this gerus. 1434: . Serpula ? sp. Pecten operculiformis Gabb. Trigonia sp. Related to I’. equicostata Gabb, and T. mauden- sis Whiteaves. Eriphyla ? sp. Small casts. Pleuromya papyracea Gabb. Ancycloceras remondi Gabb.? Fragment. Ancycloceras ? sp. Hamites ? sp. é Lytoceras batesi (Trask) ? Fragmentary small specimen. Beleninites impressus Gabb. ?. Fragmentary imprint. Horizon: Regarding the animal remains, Dr. Stanton writes: ‘The two lots from Castle creek, numbered 1432 and 1434, evidently belong to the same fauna. The horizon is clearly Cretaceous and apparently within the limits of the Horsetown formation.’ Nos. 1485-386. %,000-foot contour, 350 yards east of 7,622-foot peak five miles nearly due east of Castle Peak. Stratigraphic position: 2,800 feet below top of member K. Fossils: plants and (1485) one fossil marine shell. Professor Penhallow found that the plant remains of 1486 consist, apparently, of fragments of the rachises of ferns which remain indetermin- able, although he is inclined to consider them as derived from the one species Gleichenia gilbert-thompsoni, thus relating this horizon to that of 1428. Dr. Stanton writes: ‘The specimen numbered 14385, which according to your section comes from a much higher horizon than 1482 and 1484, has not been identified, but it is suggestive of Tertiary rather than Cretaceous. It is a marine shell.’ He described the shell thus: ‘Lucina? sp. \ EN ON NEON Se MSN EON IME MeN MVE ONE IN, Vas SOR go Baia Van Vana Vanw a Van Gy ROAMING Ils hkl exveev VEN NEA, 255 ree EAC BY AV vvvuvuvVv vi \ \ Shoes gy vvvvvvVV Wve vv vee pV VV Nw SPI Vine ta Vi ONT ENS / yVuvVV VV VY v See PATS AYN WP VA IA / % v 8 vvuvvyvyv Vv \ v vvvvy VIVAL A VanWae WAND ANZ Wredie Fictre 35.—Map showing relations of the Castle Peak stock to the deformed Pasayten formation. Strike and dip (in degrees) shown; faults in broken lines. Seale :—1 : 115,000. sensitive to such dislocation. The,faulting actually displayed in the Cretaceous beds is strike faulting and was completed before the granodiorite was intruded. (Figure 35.) The igneous body is thus neither a bysmalith nor a chonolith. The magma entered the tilted sediments, quietly replacing cubic mile after cubic mile until its energies failed and it froze in situ. Not only so; the superb exposures seen at many points in the deep can- yons trenching the granodiorite illustrate with quite spectacular effect the downward enlargement of the intrusive body. At both ends and on both sides of the granodiorite body the steep mountain cliffs exhibit the intrusive contact ~ 496 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 ttt + PF ttt ett tt tt tee ttt ett a ee ee ear a coe ea Ficure 36.—Contact surface between the Castle Peak grano- diorite and tilted Cretaceous sandstones and argillites. Section in wall of glacial cirque at eastern end of the stock, the point marked ‘‘ A” in Figure 35. Scale :—one inch to 185 feet. FicurE 37.—Plunging contact surface between intrusive granodiorite of Castle Peak stock and Cretaceous argillites and sandstones of Pasayten series. Drawn from photo- graph, looking south. Contact shown by heavy line in middle of view. Granodiorite on left ; sediments on right. _The vertical distance between the two ends of the contact line as drawn is 1,500 feet. Castle Peak on the left. RHPORT OF THE CHIEF ASTRONOMER 497 SESSIONAL PAPER No. 25a surface through vertical depths of from 300 to 2,200 feet. In every case the contact surface dips away from the granodiorite, plunging under sandstone or argillite and truncating the beds. The angle of this dip varies from less than 20° to 80° or 85° (Figures 86 to 40). On the north side of the granodiorite a a I Ficure 38.—Plunging contact surface between intrusive granodiorite (on the right) and Pasayten formation (on the left). Drawn from a photograph taken on the north side of the Castle Peak stock, near the point ‘‘C”, Figure 35. View looking east. Con- tact shown by heavy line in middle of view. Granodiorite on right of the line, which represents 1,700 feet of depth at nearer ridge. Contact also located in the background, with broken line. section of the domed roof of the magma chamber still remains (Figure 40). It is noteworthy that a well developed system of rifts or master joints in the granodiorite seems, with its low dip, to be arranged parallel to the north sloping roof, as if due to the contraction of the igneous rock on losing heat upward by conduction. 25a—vol. ii—32 ) 498 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 This fact of downward enlargement makes it still more surely impossible to conceive that the granodiorite was injected into the sediments by filling a cavity opened by orogenic energy. A visible section even 2,200 feet deep does not prove the continuance of downward enlargement with depth; yet there is no logical reason to doubt that at least the steeper observed dips of the igneous contact surface are but samples of its dips for several miles beneath the present Figure 39.—Plunging contact surface, Castle Peak stock, south side, near point “‘D”, Figure 35. View logking east. Granodiorite on left of line showing contact. The ver- tical distance between the two ends of this line as drawn is 800 feet. The highest summit is Castle Peak. land surface. Moreover, if the granodiorite made its own way through the stratified rocks and was not an injected body, passively yielding to ordinary orogenic pressures, there must have been free communication between the now visible upper part of the magma chamber and the hot interior of the earth. Downward enlargement is not only proved in visible cliff sections; it is demanded as a necessary condition of heat supply during spontaneous intrusion. REPORT OF THE CHIEF ASTRONOMER 499 SESSIONAL PAPER No. 25a The Castle Peak plutonic body thus appears to be a typical stock, an intrusive mass (a) without a true floor, (b) downwardly broadening in cross- section, and (¢) intruded in the form of fluid magma, actively, though gradually, replacing the sedimentary rocks with its own substance. It is the most ideally exposed stock of which the writer has any record. F'IcurE 40.—Intrusive contact between granodiorite and nearly vertical Pasayten argillite. Sketched in the field, on the north side of the Castle Peak stock, near point marked ‘‘ E” in Figure 35. Granodiorite on the right. Figures show elevations in feet and dips of contact surface. INTRUSION OF SYENITE PoRPHYRY. At Monument 80 the Boundary slash crosses a small mass of hornblende- biotite syenite porphyry. It is intrusive into the Pasayten sandstones. The area of the body as exposed on the present erosion-surface is about one-half of a square mile. The contacts are very poorly displayed and it was found impracticable to determine the structural relation of the porphyry. In ground- plan the body is elliptical, with its longer axis directed N.W.-S.E. This ele- ment of form suggests that the underground relations are those of a true stock, but the steady persistence of a strongly porphyritie structure at all points in the body tends to show that it is an injected, rather than a subjacent mass. If this second view is correct the body must be classed among the chonoliths rather than among the laccoliths, for it cuts across the edges of the sandstone beds along the whole length of the intrusive body. 25a—vol. ii—324 ~ 500 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Petrography.—In the field the porphyry has great uniformity of colour, texture and grain. It is a light-gray, fine-grained, strong rock, breaking with a sonorous ring. Nowhere could evidences of crushing be discerned. The phenocrysts are orthoclase, andesine, brown biotite, and green hornblende. The ground-mass is a microcrystalline, granular aggregate of orthoclase, plagioclase, and quartz, with accessory magnetite and apatite. The feldspars are generally more or less altered. The hornblende is usually represented only by pseudo- morphs of carbonate and chlorite. The biotite is much less thoroughly altered. The specific gravities of the two freshest specimens collected are 2-623 and 2-617. Chemically and mineralogically, though not structurally, this porphyry is in many essential respects like the younger phase of the Similkameen batholith. As the younger and older phases of the batholith are of common magmatic origin and nearly of contemporaneous age, and since we have grounds for correlating the older phase with the Castle Peak stock, it seems simplest to regard the syenite porphyry as a satellite of the Castle Peak stock and both of those bodies as satellites of the Similkameen batholith. Correlation.—A tentative correlation of the Similkameen batholith, Castle Peak stock, Lightning creek diorite bodies, and syenite porphyry chonolith may be expressed in the following form :— Okanagan range— Hozomeen range— Similkameen batholith. Castle Peak stock. Older phase. Principal phase. Basic contact-shell. Basie contact-shell. Younger phase. Syenite porphyry chonolith. HozoMEEN SERIES. General Description.—The ridge culminating in the remarkable double summit of Mt. Hozomeen is wholly composed of pre-Cretaceous rocks to which the name ‘ Hozomeen series’ may, for convenience, be given. These rocks extend from the major fault at Lightning creek, to the alluvium of the Skagit river. Another area of what appears to be the same series is mapped in the Skagit range. In both areas the rocks are enormously crushed, so much so that all efforts to define the original succession or structures have so far failed. The difficulty of discovering the relations of the series is enhanced by the fact that the eastern or Hozomeen mass is cut off on both sides by faults and the western area is cut off on the west by an intrusive granite batholith, on the south by a master- fault and on the east probably by another great fault. The eastern area was studied at the close of the season of 1905; the western area during the season of 1906. On neither occasion did the plan of the Boundary survey permit of the study of areas more than three or four miles distant from the Boundary slash. REPORT OF THE CHIEF ASTRONOMER 501 SESSIONAL PAPER No. 25a Either to north or to south of the Boundary belt the field-conditions may favour the discovery of the essential geological features of the series; within the belt they are distinctly unfavourable and the writer’s results are largely negative. The Hozomeen ridge is composed of a group of massive greenstones, cherty quartzites, and rare intercalations of white to pale gray limestone. Of these the greenstone is dominant wherever outcrops occur, and appears to make up the nearly or quite inaccessible horn of Mt. Hozomeen as well as the higher though accessible summit just to the north. The greenstone has everywhere been crushed and altered, in both respects so profoundly that the writer was unable to secure a single specimen which in character even approached the original material. Minute jointing is extraordinarily developed, making it almost impossible to trim a specimen to standard size or shape; usually the very freshest rock crumbled to small polygonal pieces under the hammer. The rock has the normal dark gray-green colour and almost aphanitic, massive character of greenstone. Occasionally it shows a brecciated structure which simulates that of a pyroclastic, yet no distinct beds of agglomerate oz tuff could be discerned. Everywhere irregular, discontinuous and innumerable planes of fracture, generally heavily slickened, cut the rock in all directions. At one or two points suggestions of an original vesicular structure were encountered but they were too obscure to make the effusive origin of the green- stone perfectly clear. Nevertheless, the writer believes that this rock does represent the altered equivalent of basaltic or basic-andesite flows of great aggregate thickness. Four typical specimens were examined under the microscope. They were all found to be essentially made up of secondary material,—the usual mat of vralite or actinolitic hornblende, epidote, chlorite, saussurite, omphacite, calcite, zoisite, and quartz, with here and there a granulated, altered plagioclase feldspar. Original crystal forms have been obliterated in all four thin sections. Tf all of the greenstone exposed in the Hozomeen ridge is of extrusive origin, its total thickness must be great; 2,000 feet is a safe minimum. The cherty quartzite is the next most important member of the series. If its whole strength were known it might prove to have a greater thickness than the greenstone. Some colour is lent to this idea by the discovery of other thick sections of the quartzite as it is followed along the Skagit valley trail towards Hope. In that stretch a large quantity of phyllite was found to be interbedded with the quartzite. In the Boundary sections phyllitic phases were comparatively rare; the silicious sediments there are pretty generally gray to greenish gray, compact, cherty rocks. They are thin to thick-bedded, breaking often with subconchoidal fracture. Like the greenstones they are heavily jointed, crushed and veinleted with white quartz. Under the microscope the rock is seen to. have the common cryptocrystalline to microcrystalline structure of chert in which minute grains of apparently clastic quartz are embedded. The limestone beds intercalated in the greenstone were observed only on the long spur running north-northeast from Mt. Hozomeen. Wherever seen, the beds are never continuous for more than a few hundred feet but occur as pods or 502 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 lenses from thirty to forty feet thick in the middle and tapering off to nothing at each end. This form of limestone body is that often assumed when the rock- series in which it occurs has been subjected to powerful squeezing and rolling- out. The carbonate acted as if it were plastic, thinning here, thickening there, according as the lines of force were directed. The material pinched out at one point became accumulated in pods elsewhere. The relations are thus parallel to those found in the Pend D’Oreille series of the Selkirk range. It is little wonder that the traces of bedding have here disappeared. The rock is now a fine-grained, white to light bluish-gray marble, charged with concretions of chert. These concretions are most irregularly distributed, giving no indication of original bedding-planes. The pods are always vertical or nearly so and strike rather faithfully in the direction N. 20° FE. Other masses of limestone may occur on the slope down to the Skagit river but the thick brush of the slope prevented their discovery, though the out- crops sufficed to show the predominance of the greenstone and quartzite all the way to the river-fiat. It is not possible to state the relative ages of the different members with confidence. From the analogy with the less disturbed and probably contempor- aneous Chilliwack series on the west slope of the Skagit range (see next chapter), it seems best to believe, as a working hypothesis, that the Hozomeen greenstone and limestone are younger than the principal quartzite (phyllite) group and overlie the latter conformably. Correlation.—Since the series is so far quite unfossiliferous, the search for its equivalents among the determined formations of the Cordillera is aided only by the analogies of stratigraphic relations and of lithological resemblances. On these grounds the provisional correlation of the series, at least in part, with the Carboniferous Cache Creek series has been made. Rocks which are clearly much like those at the Skagit river have been found by Smith and Calkins in their reconnaissance of the Boundary belt and have similarly been tentatively referred by them to the Cache Creek division. Their description may be quoted at length :— : ‘The supposed Cache Creek series, as represented in this district (upper Okanagan valley), comprises both sedimentary and voleanic rocks. Its lower portion consists chiefly of clay slates and graywacke slates, usually of gray or greenish colour, together with some moderately coarse metamorphic sand- stones and fine conglomerates, but comprises no coarse conglomerates. Occa- sionally the arenaceous portions of the series take on the character of fairly pure quartzite. Material of this sort becomes especially abundant near Mount Chopaka. In the upper portion of the series, as developed at Loomis, there are at least two beds of light-gray limestone, whose areal distribution is indicated roughly in the geologic map. ‘Farther south, to the west and northwest of Riverside, this rock plays a more important role. The western wall of the coulee north of that place is a cliff perhaps 200 or 300 feet high and composed mainly of limestone. REPORT OF THE CHIEF ASTRONOMER 5038 SESSIONAL PAPER No. 25a ‘The upper part of the series comprises large volumes of volcanic material, which it was not found practicable to separate, on the preliminary map, from the slaty rocks. These old voleanics are for the most part extensively developed on the southern end of Palmer Mountain, in the basin southeast of that point, to the west of Blue Lake, and on the hill southwest of Palmer Lake. Lithologically, they were roughly classified in the field as greenstones. In broad, distant views the dark brownish hues of the weathered surfaces and their rugged erosion forms give them a resemblance to basaltic rocks. In hand specimens their original character is found to be obscured by de- composition, but the porphyritic texture is occasionally noted, as well as amygdaloidal structure and brecciated structure suggestive of pyroclastic origin. Microsecpic study of these rocks is productive of no very satisfac- tory results, owing to the advanced decomposition which they have univer- sally suffered, the original materials being almost always completely re- placed. The character of the resulting secondary minerals, however, as well as the textural features, confirms the field diagnosis of the rocks. They are basic extrusives, probably for the most part basaltic, though perhaps includ- ing some basic andesite. Pyroclasties appear to be fully as abundant as the massive lavas. | ‘The rocks tentatively referred to the Cache Creek series have suffered various degrees of metamorphism. The sedimentary portions have in general an indurated slaty character in the localities removed from granitic intrusions. In the vicinity of the several intrusive granite contacts, how- ever, much more advanced alteration has taken place, the slates being more or less completely converted to mica-schist. Interesting changes have been produced also in the basic eruptives by the granitic intrusions, the description of which will be deferred to the chapter on petrography. ‘The upper part of Jacks Mountain or Mount Nokomokeen, is carved from a series which has the aspect of being much older than the Cretaceous rocks farther east and may be equivalent to the supposed Cache Creek of Okanagan Valley. It comprises both sedimentary and voleanic rocks. ‘Most prominent of the sedimentaries are quartzites and bedded cherts. The latter are generally of a light-gray or drab tint and are cut by innumer- able veinlets of quartz. Their bedding is their most noteworthy feature. They are built up of distinct laminae, about an inch in average thickness, readily separable from one another. The similarity of their structure with that of the red cherts of the Franciscan series in California is striking. As in the supposed Carboniferous of Okanagan Valley, there are beds of lime- stone (which are, however, rather thin and lenticular), and_ the highest portions of the mountain reached was built up largely of altered vol- eanic rocks, among which amygdaloids were observed. Although obscured greatly by alteration, the constitution and texture of these rocks as observed under the microscope indicate that these old lavas are basaltic. . ‘Old schists, slates, cherts, and quartzites are also the’ arid wall cbiinites: rocks in the valley of the Skagit above Ruby Creek, as far north as Jackass 504 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Point, but some ‘ greenstone’ (basalt or basic andesite) also occurs. North of Jackass Point the country rock is mainly granitic, though interrupted by a belt of slate. The impression of the observer was that these sedimen- tary and volcanic rocks were plainly older than the Cretaceous and might in part be correlative with the Cache Oreek series.’* The results of Messrs. Smith and Calkins are seen to be essentially similar to those of the present writer in his study of the Anarchist and Hozomeen series. The discovery of Upper Carboniferous fossils in the very thick sedimentary series cut by the Chilliwack river canyon, a series which corres- ponds well lithologically with the Anarchist series and with some of the rocks in Mt. Hozomeen, is further significant. (See chapter XVIII.) These sedi- ments on the Chilliwack river are only about twenty-five miles west of Mt. Hiozomeen. It seems probable, therefore, that the Hozomeen series is to be correlated with the Anarchist series, and both of them with Dawson’s Cache Creek series as well as with the likewise fossiliferous Chilliwack River series. There is nothing, however, to prove that some part of the Hozomeen series, if not a part of the rocks grouped under the name Anarchist series, is not of Triassic or even early Jurassic age. Yet it should be noted that the fossiliferous Triassic: rocks of the lower Chilliwack valley are lithologically unlike any rocks observed in the Hozomeen ridge or in the area of cherty rocks across the Skagit. Finally, this matter of correlation can not be fully understood without reference to Dawson’s several descriptions of the original Cache Creek series; to his papers the reader is referred for fuller information.t : STRUCTURAL RELATIONS IN THE RANGH. In the Hozomeen range, for the first time since leaving the summit of the Selkirks 180 miles to the eastward, we enter a comparatively broad belt where stratified rocks afford horizons which permit of the discovery of the usual mountain structures, folds and faults. The structures are relatively simple and are illustrated in the map and section. The fundamental feature in the stratigraphy of the Hozomeen range is the erosion unconformity at the base of the Pasayten series, where it rests on the Remmel batholith. Above that horizon all the members of the series seem to be quite conformable. The only pre-Cretaceous sediments are those in the Hozomeen series which are also clearly in unconformable relation to the Lower Cretaceous beds. ; As already noted the Cretaceous series forms a great monocline complicated at its top by secondary crumples. The arch-and-trough structure is seen locally on the heights east of Lightning creek. (See general profile-section on map sheet). Elsewhere and thus generally throughout the range, faults are much more important structural features than folds. Profound normal faulting took place *@G. O. Smith and F. C. Calkins, Bull. 285, U.S. Geol. Survey, 1904, p. 22. +See specially G. M. Dawson, Bull. Geol. Soc. America, Vol. 12, 1901, p. 70, where further references. REPORT OF THE CHIEF ASTRONOMER 505 SESSIONAL PAPER No. 25a in the line of Lightning creek valley and, perhaps simultaneously, along the trough excavated by the Skagit river. In each case the faulting was probably normal, with throws as shown in the profile-section. Less important faults are postulated and mapped at Chuchuwanten creek and in the axis of the anti- cline traversed by the Castle Peak stock. CORRELATION, The fossiliferous character of the Pasayten series renders possible its definite correlation with the Shasta-Chico series of California. On account of the fact that the youngest members of the series are upturned to verticality and otherwise show evidences of deformation much more intense than that usually seen as a result of Tertiary orogenic movements in the Cordillera, it seems in high degree probable that no Tertiary strata are represented in the series. Impressions of ammonites have, in fact, been found well above the base of member L. The upturning of the series is believed to have been largely com- pleted during the post-Laramie orogenic revolution. There is no question that the Hozomeen series is in unconformable relation to the overlying Pasayten series. We have concluded that the former series probably represents the Carboniferous Cache Creek series of western British Columbia. The pyroclastic beds of the Pasayten voleanie formation bear no fossils but the structural relation of this member to member B of the sedimentary series suggests the advisability of dating the volcanic outburst in the Lower Cretaceous. Tt is hardly likely that members A and B would show such apparent strict conformity if the volcanics were of Triassic or Jurassic age and a Paleozoic age is almost certainly excluded. If the Remmel granodiorite is truly of late Jurassic age, the Pasayten volcanics cannot be other than post-Jurassic; among other obvious reasons for this conclusion is the fact that the Pasayten agglo- merate also occurs in pipe-like form within the Remmel batholith in such relations as to suggest strongly a true volcanic neck. - The intrusion of the Castle Peak stock has been assigned to the Miocene: The argument for that reference is much the same as the one outlined for the dating of the Similkameen batholith (page 469). The stock is certainly post- Cretaceous. It shows no sign of such straining as would be expected if it had undergone the squeezing incidental to the late Miocene mountain-building which has so generally affected this part of the Cordillera. The lithological similarity of this stoeck-with the proved Miocene granodiorite of Snoqualmie Pass is some further indication that the stock should be referred to a geological date so relatively recent. If the writer is correct in considering the syenite-porphyry chonolith as a satellite from the Castle Peak stock, the chonolith should be dated the same, namely, in the Miocene. The Lightning Creek diorite and the apparently satellitic sills and dikes of porphyrite have been subjected to the correlation already briefly discussed. 506 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 They may have been intruded during the Eocene, Oligocene, or early Miocene, preferably during the Miocene; in any case they seem to antedate the Castle Peak stock. The following table indicates the probable correlations for the Hozomeen range :— INASOSINIS Koko Cons Oe kau eS Glacial and Recent deposits. { Syenite-porphyry chonolith. \ Castle Peak eranodiorite stock. Lightning Creek diorite stocks. Porphyrite sills and dikes cutting Pasayten series. MIGOCEN Ce rsh iesih eo aele posi IY OOOANO REA ROG CaO eA { Cretaceous (Shasta-Chico) ...Pasayten series, members B to L. Cretaceous, near or at base of : SLASEIOROUD aes arte cae Pasayten volcanic formation, agglomerate beds and voleanic neck (?) Unconformity. AYRES O33 aD auto Chernin Remmel granodiorite batholith. Carboniferous (Cache Creek) Hozomeen series, quartzite, chert, limestone and dominant greenstone. SUMMARY OF GEOLOGICAL HISTORY. The Hozomeen formation represents a part of the Paleozoic formation which was intensely mashed and metamorphosed in Mesozoic, doubtless Jurassic, time. That crustal revolution was immediately followed by the invasion of the Remmel batholith from below. Rapid erosion followed, during which the cover of the batholith was partly removed. The region subsided just after the erosion-surface had been deeply covered by the mantle of Pasayten pyroclastics. The subsidence continued during the formation of a typical geosynclinal depression. Keeping pace with the sinking, an enormous thickness of (partly marine) Cretaceous strata was piled on the geosynclinal surface. This great body of strata was deformed in post-Cretaceous time and, on account of the intensity of the action, it seems best to attribute this upturning to the well-established post-Laramie, early Eocene orogenic revolution. The penetration of the Cretaceous beds by porphyrite sills and dikes, by diorite stock-like masses and by the Castle Peak stocks with its satellites, probably all occurred in later Tertiary time, with the Miocene assumed as the best date for the largest stock. The great faults about Mt. Hozomeen may date from the early Eocene or from a later, pre-Pliocene time. The possibility is thus recog- nized that they may be somewhat younger than the folds in the upper strata of the Pasayten series. 2 GEORGE V. SESSIONAL PAPER No. 25a A. 1912 CHAPTER XVIIL FORMATIONS OF THE SKAGIT MOUNTAIN. RANGE. GENERAL STATEMENT. From the Skagit river to the great gravel plain traversed by the lower Fraser river, the Boundary belt crosses a large number of distinct geological , formations which range in age from the Miocene to the Carboniferous, if not to the pre-Cambrian. The oldest fossiliferous sediments so far discovered date from the Upper Carboniferous; these belong to a thick group of rocks (named the Chilliwack series) most of which are believed to be Carboniferous. The as yet unfossiliferous Hozomeen series crops out in the area east of Chilli- wack lake; these rocks are probably contemporaneous with certain phases of the Chilliwack series. A very thick andesitic group forms the upper part of the Chilliwack series as exposed near Tamihy creek and will bear the special name, Chilliwack Voleanic formation. A peculiar antrusive, dike-like mass of highly altered gabbroid rock, forming the western part of Vedder Mountain ridge, may be called the Vedder greenstone. Triassic argillites showing great thickness in the region east of Cultus lake have been grouped under the name, Cultus formation. Southwest of Tamihy creek canyon a group of conglomer- ates and green, massive sandstones, to which the name Tamihy series is given, seems to represent the equivalent of the Pasayten series farther east. On Sumas mountain, north of Huntingdon railway station, fossiliferous sandstones and conglomerates, named the Huntingdon formation, seem to represent the Eocene Puget group. It overlies unconformably a body of intrusive diorite cut by a biotite granite, which will bear the respective names, Sumas diorite and Sumas granite. These intrusives may be contemporaneous with a batholithiec mass of ereatly sheared granite occurring on and near Custer ridge at the main divide of the range; this body will be referred to as the Custer granite-gneiss. It seems to cut the Hozomeen series and is provisionally assigned to a Jurassic date of intrusion, but this truly old-looking rock may really represent a pre- Cambrian terrane. The eastern slope of the range at the Boundary line forms ar. area where, possibly in early Tertiary time, vigorous voleanic action built a thick local accumulation of andesitic breccias, associated with flows and with more acid lava; to the whole group the name, Skagit Volcanic formation, may be given. The remaining, specially named bodies in the range are the Slesse diorite and the Chilliwack granodiorite, both of which are in batholithic, intrusive relation to the Chilliwack series. The former occurs on Slesse creek; the latter forms the bed of Chilliwack lake and spreads far out on all sides. Both bodies are believed to be of mid-Tertiary age. (See Maps No. 15 and 16.) 507 508 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 The invention of these many new formation names is intended to facilitate correlation along the Boundary; it is hoped that they may be of service to geologists who, in the future, need to correlate with any of the rock-groups cropping out along the Forty-ninth Parallel. STRATIFIED FORMATIONS. HozoMEEN SERIES. A group of rocks believed to belong to the Hozomeen series covers three or more square miles of the Boundary belt north of Glacier Peak and just east uf the main divide of the Skagit range. The area presented no geological features of special novelty and its description may.ebe given in few words. The cherty quartzite is here the prevailing rock, occurring generally in thin, flaggy beds from one inch or less to three inches in thickness. Phyllitic interbeds are commoner here than at Mt. Hozomeen. Near the Custer batho- lith, the quartzites are micaceous and the once-argillaceous beds are now mica schists. Occasional bands of probably conformable and extrusive greenston2s are intercalated, but greenstone nowhere in this area assumes the importance it has east of the Skagit river. No limestone was observed in the main area; a patch of intensely metamorphosed schist and quartzite with included limestone pods occurs on the ridge-summit north of Depot creek, where the older Custer granite makes contact with the Chilliwack granodiorite. This stratified mass formed part of the roof of the older batholith and then a second time under- went metamorphism as it was invaded by the Chilliwack batholith. The lime- stone will be described in the section dealing with the contact-aureole about the latter intrusive. At all the outcrops in the western areas the quartzite-phyllite series has steep dips, ranging from 70° to 90°. In the larger area the beds are intensely crumpled but the strike averages about N. 35° W,; the dip is generally about vertical. It is probable that several thousand feet of the sedimentary beds alone are represented in this area but it has proved so far impossible to seeure either top or bottom for the series. CHILLIWACK SERIES. General Character and Distribution.—From the western limit of the Chilli- wack granodiorite batholith to a point about two miles below the confluence of Tamihy creek,—a distance of sixteen miles in an air-line—, the Chilliwack river flows over a great thickness of sedimentary rocks to which the name, Chilliwack series, has been given. These rocks cover the whole width of the Boundary belt (as mapped) throughout most of the distance and extend far to north and south of the belt. They were examined by Bauerman in his reconnaissance of 1859, when he estimated the total thickness of the sediments exposed along REPORT OF THE CHIEF ASTRONOMER 509 SESSIONAL PAPER No. 25a the river as about 24,000 feet.* While he did not allow for duplication by fault and fold, his belief that the series is very thick was certainly justified. The Paleozoic section along the Chilliwack river is, indeed, one of the most complete of all those so far recorded on the western slope of the Skagit range, and besides the definitely Paleozoic strata of this section, there is another important group of Mesozoic beds occurring along the Chilliwack river. To the former group only, and particularly to the Carboniferous portion of it, the name Chilliwack series is intended to apply. For the first half-dozen miles westward of the Chilliwack batholith there are heavy masses of old-looking sediments which are so far unfossiliferous and may in part belong to the pre- Carboniferous terranes. From the mouth of Slesse creek to a point about ten miles due westward, and from the river southward to the Boundary line, the Chilliwack series is typically represented. and is fossiliferous at so many points that little doubt remains as to the Carboniferous age of practically all the sediments occurring in these sixty square miles. The eastern limit of the large area of Chilliwack sediments is, within the Boundary belt mapped, fixed by the intrusive contacts of the Slesse diorite and the Chilliwack granodiorite. The western limit is exceedingly difficult to place but is provisionally placed at the outcrop of an assumed master-fault mapped as crossing the belt a few miles west of Tamihy creek. The northern and southern limits of the sedimentary mass have not been determined. From the fault just mentioned to another assumed fault running along the axis of Cultus lake valley, the Mesozoic (probably Triassic) formation separates the main body of the Chilliwack rocks from a smaller one which forms much of the long ridge known as Vedder mountain. No fossils have been found in this ridge but it seems most probable that its rocks form the lewer part of the Chilliwack series and may be, therefore, all of Carboniferous age. On this view the intervening block of Mesozoic strata have been faulted down into lateral contact with the Carboniferous Chilliwack series. Fossiliferous limestones associated with some shale and with a heavy body of contemporaneous andesite make contact with the Mesozoic formation along a line running nearly parallel to, and just south of the Boundary line. The former group represent a part of the Carboniferous series which has, apparently, been here thrust up over the Triassic rocks. The thrust-plane dips south at an unknown angle. Finally, the Chilliwack series may be represented in some small areas of poorly exposed quartzites and slaty rocks uneonformably underlying the Eocene (2?) beds on Sumas mountain. Notwithstanding a very considerable amount of arduous climbing distri- buted through part of each season in 1901 and 1906, not sufficient data are in hand to afford a complete idea of the succession of rocks included in the Chilli- wack series. The density of the vegetation in these mountains, unparalleled as it is on the whole Boundary section elsewhere, will always stand in the way ee Bauerman, Report of Progress, Geol. Surv. of Canada for 1882-3-4, Part B, p. 82. : 510 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 of the full discovery of the facts needed for the stratigraphy of the series and the structural geology of its rocks in these areas. It is to be understood that the following statements should be subject to careful revision through future field work. Detailed Sections and the Fossiliferous Horizons.—Neither lase nor top has been found for the series. Partial sections have been roughly measured and these will be described in brief form. On the basis of these as well as a multi- tude of details, isolated facts entered in the field note-books, a provisional columnar section embracing the rocks actually observed east of Cultus lake, has been constructed. Section TJ. About one mile west-southwest of Monument 48,beds which are believed to be the youngest exposed members of the Chilliwack series are unconformably overlain by grits and conglomerates belonging to the Tamihy Cretaceous (7?) formation. From that point to the ridge of Church mountain two miles north of the Boundary line the exposures are unusually good for this region and a partial section of the series has there been made with some degree of confidence. The order is as follows :— Top, unconformable contact with Tamihy formation. a. 50 (or more) feet.—Quartzitic sandstone. been 20 « Dark gray argillite. z @. 5 “Light gray limestone, bearing fossils with numbers 1506, 1509-10. d. 60 (estimated) “ Gray calcareous quartzite and dark gray, calcareous argillite. e. 2,000+ « Andesitie flows, tufis and agglomerates. f. 200 ““ Gray and brownish shale and sandstone; thin conglomerate bands; crumbling, thin-bedded; highly fossiliferous. Col- lection Nos. 1512, 1514. g. 600 (estimated) “ Light gray, generally crystalline limestone, with fossils, No. os 2,980 feet. Base concealed. For the determination of these as well as of the other -collections made in the Chilliwack rocks, the writer is indebted to the great kindness of Dr. George If. Girty, and Dr. R. S. Bassler. The latter determined the bryozoa; the other genera were determined by Dr. Girty. The results may be quoted from Dr. Girty’s letter, in terms of the collection numbers :— No. 1506. About 900 yards south of the Boundary slash and 1,500 yards southwest of Monument 48. Fossils: crinoidal fragments. No. 1509. 100+ yards southwest of Monument 48. Fossils: Zaphrentis sp. Campophyllum sp. Euomphalus sp. REPORT OF THE CHIEF ASTRONOMER 511 SESSIONAL PAPER No. 25a Nos. 1510-11. Same locality as 1509. Fossils: Fucoidal markings. No. 1512. On top of ridge 1,500 yards northwest of Monument 48. Fossils: Plant fragments. Clistophyllum sp. Crinoidal fragments. Fenestella sp. Rhombopera sp. Cystedictya sp. Productus senireticulatus Martin. Productus aff. jakovlevi Tschern. Spirifer aff. cameratus Morton. Reticularia lineata Martin (?) Spirtferina aft. campestris White. Martinia (2) sp. Seminula (2) sp. Terebratuloid (2?) Myalina aff. M. squiameosa Sowerby. Aviculipecten sp. leurophorus (?) sp. Orthoceras (?) sp. No. 1518. On same ridge as 1512, 1,000 yards farther north. Fossils: Thonsdaleia sp. Campophyllum (7%) sp. Crinoidal fragments. Fistulipora sp. No. 1514. About 1,200 yards west of summit of Church mountain; top of ridge. Fossils: Fenestella sp. Pinnatopora sp. Rhipidomella aff. nevadensis Meek. Chonetes sp. Productus semireticulatus Martin. Productus-_ aff. wallacei Derby. Productus aft. jakovlevi Tschern. Spirifer aff. cameratus Morton. Spirtfer aff. lyra Kut. Spirtfer sp. Reticularia lineata Martin (2) Martinia (?%) sp. > 512 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Fossils—Continued. Spurferina aff. billingsi Shumard. Chothyridina pectinifera Sow (2) Hustedia aff. compressa Meek. Hustedia aff. meekana Shumard. Pugnazx aff. wtah Marcou. Dielasma (2): sp. Camarophoria sp. Aviculipecten aff. coxanus M. and W. Parallelodon aff. tenwstriatus Meek and Worthen. Parallelodon sp. Sanguinolites sp. Naticopsis sp. Orthoceras sp. Section II, On the Commission trail running along the Boundary line eastward from the Cultus lake valley and about 1,200 yards southwest of Monument 45, a massive limestone with a fifty-foot interbed of dark gray shale was found to carry fossils (No. 1500). The species were identified by Dr. Girty as:— Fusulina elongata Shumard. Rhombopora sp. Productus (%) sp. This limestone appears to correspond to member g of Section I. It dips under the great voleanic member in apparent conformity. The exposures at this point are too poor to make the section of very great value. SECTION TT. One of the most useful sections in the series is one traversed, in 1901, along the west slope of McGuire mountain where it steeply plunges to the bed of Tamihy creek, 6,000 feet below its summit. This mountain is crowned by a very ragged and broken syncline of massive limestone equivalent to that on Church mountain across Tamihy creek and to member g of Section I, (Plate 42, A). An infold of the shale overlying the limestone seems to correspond to the shale of member f of that section, but the volcanic member seems to have been here entirely destroyed by erosion. Below the massive limestone is a great thickness of sediments which are fairly well exposed in the gulches leading down to Tamihy creek. Measurements are very difficult to make on account of frequent faults and crumples in the bed. An approximate idea of the succession ean be obtained from the following table:— REPORT OF THE CHIEF ASTRONOMER 513 SESSIONAL PAPER No. 25a Top, erosion surface. a. 200 feet (rough estimate) Shale and sandstone. b. 600 Massive light gray to whitish, crystalline limestone with numerous crinoidal fragments in places. Cay 90S Shale, sandstone and grit. ds 110=-“ Massive light gray limestone with large crinoid stems and same fossils as member e. €. 300=— Dark gray shale with fossils, No. 104. ee 100 ee Massive, hard sandstone. g- 1,400 “ (rough etimete) .—Hard sandstone, red and black shale, grit and thin bands of fine conglomerate. h. 800 “ 59 5 Hard massive gray sandstone with gritty layers. Base hidden under talus of Tamihy creek canyon. Dr. Girty found the fossils of No. 104 to belong to the following species :— Pentremites sp. Platycrinus sp. Fenestella aff. perminuta Ulrich. Fenestella sp. Pinnatopora sp. Polypora cf. submarginata Meek. Chonetes sp. Productus semireticulatus Martin. Spiriferina sp. Hustedia aff. compressa Meek. SEcTION IV. At Thurston’s ranch nearly opposite the mouth of Slesse creek, from there northward up the mountain-side, and also westward along the Chilliwack river, a very rough section has been run through the dense brush. No great confidence is felt in the result, for there is a possibility that strike faults’ or other unsuspected structural complications have repeated members of the series, or, on the other hand, have faulted some of them out of sight. Attention was first called to this particular part of the river section by the discovery of- abundant crinoid stems in a heavy limestone cropping out just north of the ranch. This limestone seems to be at least 400 feet thick, though its base is concealed; it probably corresponds to member e in Section I. Northward from this outcrop the succession was crudely determined to be:— Top of section, not well exposed. a. 125+4feet.—Dark gray and black shale. De) ASO Ee Coarse agglomerate composed of dark andesitic or basaltic fragments of large size together with other large fragments of limestone. c 90+ “* Typical pillow-lava, basaltic; pillows round, up to three feet in diameter, with the spaces between them filled with cherty matter. d. 7 ““ Brown and gray shale. e. 300 “ Light gray, massive limestone. f: he an Brownish shale. g. 150 “ Coarse feldspathic sandstone with conglomerate lenses. h. 400+ “ Light gray. crystalline, massive limestone with large crinoidal stems quite abundant. 1,340+ feet. Base concealed. 25a—vol. 11—33 614 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 West of Thurston’s ranch the crinoidal limestone seems to be repeated by a strike-fault, the beds retaining their general northeasterly dips of from 20° to. 60° or more. This attitude is fairly well preserved in the outcrops along the river trail all the way to the mouth of Tamihy creek. The welter of forest, brush, and moss, as well as a heavy mass of Glacial drift on the river- valley floor, prevent any accurate conception of the nature of the beds crossed in this seven-mile traverse down the river. It is probable that the rocks corres- ponding to members e, f, g and h of Section III. are represented in this section or have been faulted out of it and that the very thick, phyllitic argillite seen along the north bank of the river at and just above the confluence of Tamihy creek with the river, is an older member of the series than any of those so far mentioned. Nothing better than a guess as to the thickness of this member is possible but 1,000 feet is apparently a very safe minimum. The pillow-lava and agglomerate of Section IV. seem to represent the lower _ part of the great voleanic member e, of Section J. The adjacent rocks match the respective members of Section I. in a rough way; considering that contin- uous exposures were not to be found at either locality, an exact correspondence should not be expected. General Columnar Section—Combining the facts determined in these four sections with the many scattered observations made elsewhere, the following table may be made to express the writer’s tentative conclusion as to the anatomy of the Chilliwack series :— Top, erosion surface at plane cf unconformity with the Tamihy (Cretaceous?) formation. 50 + feet —Quartzitic sandstone. 20 Dark gray argillite. 50 «Light gray limestone; fossils, Nos. 1506, 1509, 1510. Gray calcareous quarizite and argillite. 2,000+ ‘ Andesitie flows, tufis, and agglomerates (pillow-lava i baie in this. member where locally developed). This member may for conven- ience be referred to as the Chilliwack Volcanic formation. 6. 200 “Gray and brownish shale and sandstone, with thin conglomerate bands; shales crumbling and thin-bedded; highly fossiliferous. Fossils Nos. 1,512 and 1,514. : 7. 600+ “ Light gray, massive, generally crystalline limestone, often crinoidal; with fossils No. 1513 (crinoidal fragments also represented in Nos. 69, 70, 71, 72, 98, 129). 8. 90 «Shale, sandstone and grit. 9. 110+ “ Massive light gray limestone, with large crinoid stems and fossils as No. 104 (not collected here). 10. 300+ ‘“ Dark gray and brown shales, with fossils, No. 104. 11. 100 «« Massive, hard sandstone. 12. 1,400+- ‘“ Hard sandstones and black and red shales with bands of grit and thin beds of conglomerate; thickness very roughly estimated. 13. 800+ “ ##Hard, massive sandstone with gritty layers. 14. 1,000 + “© Dark gray to black, often phyllitic argillite with quartzitic bands. 6,780+ feet. SU Co De Base concealed. Geological Age of the Series.—As already indicated, the lower members. ot the Chilliwack series may belong to one or more systems older than that. PLATE 44. Looking north. ; b. Rugged topography at the Boundary, east of Chilliwack Lake and north of Glacier A.-—-Carboniferous limestone, summit of McGuire Mountain. Peak. Mountains composed of Skagit volcanic formation. Looking southwest. C.—Horn topography between 'l'amihy and Slesse creeks. Peaks composed of metamor- phosed members cf the Chilliwack series. Looking southeast from McGuire Mountain. D.—Horn topography on ridge between Slesse and. Middle creeks. Massive crags of Chilliwack granodiorite. Looking east. 25a —vol. ii—p. 514. REPORT OF THE OHIEF ASTRONOMER 515 SESSIONAL PAPER No. 25a one represented in the fossiliferous horizons. On that question there is at present absolutely no light. It remains to note in Dr. Girty’s general summary of the status of the fossiliferous beds themselves. He writes :— ‘In the way of explanation I may state that, owing to the imperfect Inowledge of most of our western Carboniferous faunas and to the poor state of preservation in which the fossils occurred, it was not possible to make positive identifications in most eases. ‘Faunally, I would be disposed to arrange these collections into several groups. Lots 1512 and 1514 are closely related and represent, perhaps, the only strongly marked fauna in the collection. Lot 1500 is also rather diagnostic. Lot 104 is moderately extensive, but is not strongly character- istic. It seems to differ considerably from either of the two faunas just mentioned. Of the remaining collections, which are faunally very limited, two groups can possibly be made. One of these comprises such lots as consist only of very abundant and very large crinoid stems (lots 69, 70, 71, 72, 129, 1506 and possibly 98), or crinoid stems and cup corals (lots 1498 and 1513), or cup corals alone (lot 1509). The other group shows only fucoidal markings (lots 1510 and 1511). ‘The most natural geologic section with which- to compare these faunas is that of northern California. The sequence of the Carboniferous formations there consists, in ascending order, of the Baird shales, the McCloud limestone, and the Nosoni formation (formerly the MeCloud shales). The Baird shales have usually been regarded as of Lower Carboni- ferous age and the McCloud and Nosoni as Upper Carboniferous. Al] three have extensive and characteristic faunas. ‘There is nothing among your collections which suggests the Baird or McCloud. The most strongly characterized of your faunas (lots 1512, 1514, and 1500), however, have much that is similar to the Nosoni. At present I am disposed to correlate the two horizons. Lot 104 is less certain, but possibly belongs to the same group. The lots furnishing only corals and crinoids differ widely from 1512 and 1514, but they might readily come from a specialized bed in the same formation. Nothing positive can, however, be stated about them. As to the three remaining lots, the data do not warrant suggesting any- thing whatever. On the whole, from the little that I understand of the stratigraphic relations and from the relationship manifested by the most marked of your faunas with that of the Nosoni formation, I am disposed to correlate all your beds in a general way with the latter. They may contain measures younger or older than the Nosoni, but from the absence of the well-marked Baird and McCloud facies it seems probable that none of the horizons from which your collections came is as old as the McCloud.’ In conclusion, it may be stated that the great voleanic member, the Chilli- zack Volcanic formation, which will be specially described, is of distinctly Jpper Carboniferous age. Just above and just below this member are con- rmable limestone beds containing samples of the fauna discussed by Dr. 25a—vol. 1i—334 516 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Girty. The estimated thickness given above for the sediments of the series— 6,780 feet—is the minimum thickness of the Upper Carboniferous sediments in this region. CuLtus ForMATION. Stratigraphy and Structure—In 1859 Bauerman recognized the strong lithological contrasts between the rocks on the two sides of Cultus lake and remarked that the distinctly more metamorphosed sedimentaries on the west side looked geologically older than the shales and. sandstones of the eastern — shore. The writer is inclined to share Bauerman’s view and, as noted above, tentatively maps the rocks of Cultus ridge, as well as the large area of (fossili- ferous) argillite to the southeast of that ridge as Triassic, while the beds occurring in Vedder mountain are mapped as Carboniferous. The name, Cultus formation, may be advantageously given to the younger group of strata. It may be defined as the local series of sediments which belong to the same geologi- cal system as the thick argillite bearing the Mesozoic fossils of lot No. 1502, hereafter described. The dominant rock of the formation is a dark gray to blackish argillite, often bituminous in moderate degree. With it there are generally associated thin or thick bands of gray or greenish-gray sandstone and grit, and, more rarely, interbeds of fine conglomerate. The gritty beds are characteristically charged, very often, with small angular fragments of black argillite. All the coarser-grained types tend to be decidedly feldspathic, sometimes suggesting an arkose. ‘These rocks are invariably deformed, with dips running up to 70° or 80°, though those of 30° or 35° are the commonest. The strike is highly variable in many places but the average direction is that parallel to the Cultus Lake valley; the average dip of, say 30°, is to the southeast all across the area where the formation is mapped. The argillites are very often heavily slickened by local faults but the formation as a whole cannot be described as much meta- morphosed. Phyllitic phases, for example, were not discovered. This relative lack of metamorphism was one of the criteria by which the formation was separated from the argillaceous phases of the Chilliwack series. As Dawson found in Vancouver island, the difficulty of distinguishing the Paleozoic and Mesozoic beds is greatly enhanced by the fact that in both, argillaceous types of great similarity in’ their original composition are the dominant types. Needless to say, the future worker in the geology of the lower Chilliwack valley will not take the accompanying map too seriously but will regard it as simply the first rough approximation in mapping. Incidentally, the present writer anticipates with great sympathy the struggle of such future worker with the jungle beneath which the truth is here hidden. Two great normal faults and a no less important over-thrust are entered on the map as explaining the lateral relations of the block of Cultus sediments with the surrounding Chilliwack formations. These suggestions will need special scrutiny. ‘gouvISIP OY} UL SULRIUNOP WeayD yoemTpyO ye Aop[eA toese1y oyy jo ureyd [eianye wos pomata ‘asuey yideyg jo espa uszaysa Ay ‘Gh HLVIg . 516. vol. 1i—p 25a REPORT OF THE CHIEF ASTRONOUER 517 SESSIONAL PAPER No. 25a The question of the thickness of Triassic beds actually exposed in the Boundary belt cannot be fully answered. A safe minimum is 1,000 feet but there is reason to believe that it is much greater. The great monocline of Cultus ridge alone seems to carry between 6,000 and 7,000 feet of beds, chiefly argillite with stbordinate sandy layers. The possibility of duplication in this section makes it advisable to place the minimum thickness at no more than half the apparent thickness, say, 3,000 feet. The exposures both in this mono- clinal section and elsewhere are too poor to permit of even an approximate columnar section for the Triassic formation. Fossils—The only fossils found in the Cultus formation were discovered in 1906 at a point about 500 yards south of the Boundary and 900 yards west-south- west of Monument 47. Here the. staple black to dark-gray argillite is very homogeneous and carries few lenses of sandstone. Near the bottom of the 800- foot section, where the Boundary Commission trail crosses the creek, the fossils were discovered. Throughout the section the strike averages N. 65° E., and the dip, 45° S.S.E. There is considerable evidence of local slippings, with some brecciation and slickening of the argillite. The fossils are usually much dis- terted; all of them were found in a thin band close to a plane of slipping. The writer owes the determination of the fossils, so far as that was possible, tc the kindness of Dr. T. W. Stanton, of Washington. He writes :— “Lot No. 1502 contains: Arniotites vancouverensis Whiteayes? Numerous, more or less dis- torted specimens apparently belonging to this species. Aulacoceras ? sp. A single fragment of a belemnoid which resembles A. carlottense Whiteaves. “The lot numbered 1502, consists almost entirely of ammonites which seem to be identical with some described from the Triassic of Vancouver and Queen Charlotte Islands. Like the original types with which they are compared, they are not well enough preserved to show the septa and other features that are needed for their accurate classification.’ Combining this paleontological evidence with a comparison of the lithology of the Cultus formation and the Triassic rocks of Vancouver island, the writer has come to the view that little doubt need be entertained as to the Triassic ege of the Cultus beds. Neither limestone nor contemporaneous volcanic matter have been found in association with the Cultus argillites, but this failure, by which we recognize an important difference from the Triassic sec- tions of Dawson, can be readily explained on the view that these formations, if present, are faulted out of sight in the Cultus lake region. It is, of course, possible, though not probable, that these important members hore apparently Missing, were never laid down in the Chilliwack region. 518 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 Taminy SERIES. For about two miles from the Boundary line down Tamihy creek, the southwestern slope of the deep valley is underlain by an important series of rocks which have not been discovered with certainty at any other point in the Boundary belt this side of the Skagit river. This group of rocks was first seen in 1901 and given the provisional name, ‘Green Quartzite Series.’ It was unfossiliferous but it was thought to be older than the Chilliwack series.* During the season of 1906 much better exposures were found on the heights west of the creek and especially on the south side of the Boundary line. The relations are such as to enforce the belief that this new series lies unconform- ably upon the Carboniferous limestone, quartzite, and greenstone, and is very probably younger than the Triassic Cultus formation as well. Instead of the name chosen in 1901 the writer prefers to use the localizing name, Tamihy series, for this younger group of sediments. It is not the intention to name it as if it had been thoroughly analyzed and become stratigraphically under- stood; the name is chosen for convenience in the present report only, though possibly it may be of service in the hands of the geologist whose duty it will be to investigate this important mass of strata. The relation of the Tamihy series to the Carboniferous rocks was determined with a fair degree of finality on the summit southwest of Monument 48 of the International line. At that point the quartzite noted as member 1 of the general columnar section of the Chilliwack series, is overlain by a well exposed body of strata, of which some 400 feet are clearly visible near tree-line. It is a heterogeneous mass of gray conglomerate, black quartzitic sandstone, dark-gray paper shales, gray grit, and green sandstones in rapid alternation. The pebbles of the conglomerates include quartzite, vein (?) quartz, chert, and argillite, with almost certainly a few of greenstone like that of the Chilliwack Volcanic forma- tion in the immediate vicinity. The chert pebbles are apparently of the same material as that so commonly found in the chert nodules of the underlying limestone. A few obscure plant-remains were found in the sandstone. The atti- tude of these beds is variable but the local dip averages about 30° to the south- southwest. What appears to be the same series of rocks was followed for a mile south of the Boundary line as far as the top of a long ridge which runs eastward to Tamihy creek. That ridge is composed of a thick, very massive group of green and gray sandstones, grits and conglomerates like that just described, with little or no argillite. The dominant sandstone is extremely thick-bedded, so much so that it is rarely possible to obtain indications of strike and dip. Where these could be read, ‘as south of Monument 48, the dips were 25°-30° to the southeast. On that traverse, as along the. Tamihy creek section, the minimum thick- ness is estimated to be 2,500 feet, but there is an indefinite addition to be made to this estimate when the area south of the Boundary line is investigated. No *See Summary Report for 1901, page 51. REPORT OF THE CHIEF ASTRONOMER 519 SESSIONAL PAPER No. 25a hint of a top to the series was anywhere visible and the total thickness may be several times 2,500 feet. _ In the field the writer was much struck with the extreme similarity of the dominant (highly feldspathic), characteristically green sandstone with the sandstones which make up so much of the great Pasayten series farther east Messrs. Smith and Calkins also note a similarity between ‘ Mesozoic’ rocks occurring at Austin Pass some twenty miles southeast of the locality now being described, and the sandstones of the Pasayten.* In 1898 Dr. Stanton reported on some fossils collected by Mr. W. H. Fuller on Cowap creek which lies imme- diately south of the Tamihy creek area. Dr. Stanton wrote :— ‘The fossils are evidently all from one horizon, which I believe to be upper Jurassic, this opinion being based chiefly on the distinctly striated form of Aucella, identified with A. erringtoni (Gabb) of the California upper Jurassic Mariposa beds. This species was collected at both localities. The collection from Canyon (Cowap) creek includes asiso a fragmentary Pleuromya and the impression of a small belemnite. ‘The collection from the divide between Canyon creek and the waters of the Fraser river contains the Aucella erringtont, a fragment of an ammonite apparently belonging to the genus Stephanoceras, a smal] slender belemnite like that from the last-mentioned locality, and the phragmacone of a large robust belemnite.’ ¢ It is possible that these fossils were collected from beds which belong in the Tamihy series as here defined, or they may have been taken from beds more directly associated with the Cultus Triassic formation. Though the text and map of the report of Messrs. Smith and Calkins imply that those authors regard the green sandstones as of probably Jurassic age, their statement of the litho- jogical resemblance of the sandstone with that of the Pasayten series suggest also that the Austin Pass rocks are really Cretaceous. The present writer is inclined to take the view that the Tamihy series, as represented on the Forty- ninth Parallel, should be referred to the Shasta division of the Cretaceous. This tentative reference has naturally little value; it invites criticism as a result of much additional field work in the region. No other occurrences of the Tamihy series have been proved in the Boundary belt, but in the floor of Cultus lake valley above the lake, and, again, on the top of Pyramid ridge near the contact with the Chilliwack granodiorite batholith, certain econglomeratie and sandy beds have strong similarity with certain phases of the Tamihy series. Huntinepon ForMATION. The southern end of Sumas mountain is underlain by a cover of stratified rocks which pretty clearly belong to a period much later than any other group of consolidated sediments so far seen in the Boundary belt west of Osoyoos lake. *G. O. Smith and F.C. Calkins, Bull. 285, U.S. Geol. Surv., 1904, p. 27. } From Bull. 235, U.S. Geol. Survey, 1904, p. 27. 520 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 In 1901 a brief examination of these beds was made. The exposures are gener- ally small and poor, so that a complete treatment of the series cannot be given. It consists of heavy masses of medium-grained, gray-tinted conglomerate alter- nating with sandstone and shale, the conglomerate being apparently most abundant at the base of the group. Unlike the clastic rocks of the Tamihy series, these are friable to a notable extent. Thin and seemingly unworkable beds of tolerable coal have been found in the upper part of the formation, and it is reported that borings have declared the presence of a valuable bed of fire-clay which was found beneath the ledges cropping out at the edge of the Fraser river alluvium southeast of the main sedimentary area. The conglomerates contain pebbles manifestly derived from the (Chilliwack series ?) quartzite and from the Sumas granite, with both of which the sedi- ments make unconformable contact. The sandstones are feldspathic and arkose- like. The shales are sometimes carbonaceous, and at a point about 800 feet above the prairie and near the northern edge of the formation as mapped, Messrs. D. G. Gray and M. McArdle, who were in charge of the boring operations, dis- covered some fossil leaves in the shales. The fossils were found in the vicinity of a thin coal-bed which is consider- ably broken, and seems not to afford a body large enough for economical working. The plant specimens were submitted to Mr. F. H. Knowlton, who reported that the collection was of little diagnostic value. He writes that the material has somewhat the appearance of species regularly found in the Laramie group, but states that much weight should not be given to this impression won from the study of the very poor material. Mr. Knowlton ventures on no specific names for the fossil forms submitted to him. The general relations of the deposit, its degree of induration, and the evi- dence of the fossil plants, slender as it is, suggest that the formation should be equated with the Puget group, and thus belongs in the Eocene. The dips are not often to be read, but they seem to be always rather low, with 30° the observed maximum and 5° to 12° common readings. ‘The strike is highly variable. We seem, therefore, to have here a relatively little disturbed cap of strata laid down at a date distinctly later than that of the post-Laramie orogenic revolution, which so signally deformed the Cretaceous rocks of this general region of the Cordillera. The thickness of the visible beds totals probably about 1,000 feet. To this group of sediments the name, Huntingdon formation (from the name of the neighbouring village), may be given. Rocks of apparently the same nature and age have been long known as coal-bearing in the Hamilton and lower Nooksak Valley districts south of the Boundary line. * Near the southern end of Wade’s Trail over Sumas mountain a bench of the Huntingdon conglomerate and sandstone is cut by thin sheets of a greatly weathered porphyry, apparently a syenite porphyry. The relations are obscure; the porphyry may occur as one or more sills, or as dikes. It is the only eruptive rock known to cut the Eocene formation. The porphyry has not been examined under the microscope. * See G. O. Smith and F. C. Calkins, op. cit., page 34. REPORT OF THE CHIEF ASTRONOMER . 521 SESSIONAL PAPER No. 25a IGNEOUS ROCK FORMATIONS, CHILLIWACK VOLCANIC FORMATION. On Tamihy creek about three miles below the Boundary-line crossing, a large body of altered basic lava was discovered during the season of 1901. It was followed southwestwardly to the top of a very rugged ridge where the lava was found in close association with strong beds of obscurely fossiliferous lime- stone dipping under the lava. It was, however, not until the Commission trail west of the creek was opened up that the writer was (in 1906) able to secure definite evidence as to the age of the lava and as to its relation to the sedimen- taries. A few hundred yards south of Monument 48, the southern limit of the lava was found. It there makes direct contact with a fifty-foot bed of fossili- ferous limestone similar in habit to that at the lower contact of the lava. From the fossils collected in the upper limestone Dr. Girty has concluded that this limestone is certainly Paleozoic and in all probability upper Carboniferous in age. The limestone at the base of the lava formation is likewise apparently upper Carboniferous. Since the lava is conformably intercalated between the two limestones, it must also be referred to the upper Carboniferous. From its position in the limestone one may fairly conclude that the eruptions were wholly or in part submarine. Westward from Tamihy creek the band of old lavas was followed for a distance of some ten miles. Throughout most of that stretch the northern con- tact of the lava lies only a few hundred yards south of the Boundary slash. The best exposures are on the ridge southwest of Tamihy creek. The formation may be here called the ‘ Chilliwack volcanics.’ The total thickness can be only roughly estimated but it must be at least 2,000 feet. The formation consists mainly of thick, massive flows, which are so welded into one another and so altered. as to make the individual flows very hard to distinguish in the field. Among the flows a notable, though subordinate amount of ash-bed material is intercalated. At no point, however, were the conditions favourable for working out a detailed columnar section of the formation. Although every effort was made to secure the freshest material for study, it was found that all of the twenty type-specimens collected were greatly altered. The exact petrographic nature of the different flows is therefore obscure. The net result of the microscopic study of nearly all of the twenty specimens went to show that two rock-species are represented—augite andesite and hornblende andesite. The former is probably the more abundant. The lavas are often amygdaloidal, with calcite generally filling the pores. Very often the andesites have been altered into typical greenstones, or, where the shearing has been particularly intense, into green schists. In a few speci- inens, the augite has the relations and abundance observed in olivine-free basalts. Olivine was not found in any thin section, but its absence may be due to the profound alteration of the lavas. Some of the specimens carry quartz in the ground-mass but it may all be of secondary origin. 522 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 At Tamihy creek about five miles from the Chilliwack river, a 100-foot bluff of rhyolite was discovered. The density of the forest-cover in the vicinity rendered it impossible to determine the relation of this rock to the Paleozoic sediments or to the adjacent Chilliwack andesites. The writer conjectures that the rhyolite is a lava flow occurring at or near the base of the great andesitic series and the rhyolite is tentatively included in the Chilliwack Volcanic for- mation. The rhyolite seems to be about 100 feet thick or even more but it could not be followed far in any direction. No similar acid rock was found in the sections of the formation farther west. The rhyolite is peculiar in being coloured almost black by an abundant material rather uniformly distributed through the ground-mass. This substance is quite opaque, dead-black and amorphous and has a dust-like appearance in the thin section. It is certainly not an iron-oxide. The rock decolourizes before the blow-pipe and it seems almost assured that the black dust is carbon. The oceurrence of this element in a rhyolitic lava is unusual and the writer can find no record of its having been found in rhyolite or porphyry elsewhere, in anything like the abundance observed in this case at Tamihy creek. X VEDDER GREENSTONE. The northwestern slope of Vedder mountain ridge is underlain by an altered, basic igneous rock which seems to be intrusive into the Paleozoic argillites and sandstones of the ridge. As exposed the body forms a remarkably long and straight band, running from the head of the Chilliwack river alluvial fan to the International line south of Sumas lake. The body was not followed farther to the southward. As shown on the map the known length of the mass is more than ten miles. On the northwest, for most of its length it is covered by alluvium, so that the exact shape and relations of the body cannot be deter- mined. At the point nearest to Sumas lake the igneous rock is bounded on the northwest by a narrow belt of dark-gray argillite, cropping out at intervals for about 700 yards along the wagon-road. Here the argillite seems to dip south- eastward and thus under the intrusive rock, at an average angle of 65°, while-at the southeastern contact on the summit of the ridge, the dip of the argillite is about 40° to the east-southeast. At this point, therefore, the intrusive appears to have the relation of a great sill, injected into a bedding-plane of the sedi- ments. The width of the outcropping igneous mass is about 1,000 yards. Elsewhere in the ridge the dips and strikes of the strata, always highly variable, show no such simple relation to the intrusive. The singular straight- ness of the southeastern contact suggests that the body is a gigantic dike, and this view is tentatively adopted. An intrusive character is inferred more from the petrographic nature of the mass and from its position in the sedimentary terrane than from the usual criteria of apophyses, inclusions, and contact-meta- morphism. Owing to the dense brush and heavy mat of moss and humus, not a single, actual contact was discovered. ‘OF SLVIG ‘oury Alvpunog eyy Jo Yytou saTlut Omg ynoqe yulod & WOA} ysaMY NOS BuLyOOrT ‘vSUvY YlsevyQ aya Jo yIMWMNG Se p- 522 vol. u— 35a — REPORT OF THE CHIEF ASTRONOMER 523 SESSIONAL PAPER No. 25a From end to end of the body the igneous rock is profoundly altered; hence it is almost impossible to ascertain its precise original composition. It is now chiefly a mass of secondary minerals, including serpentine, tale, epi- dote, zoisite, kaolin, chlorite, and quartz. A pale green actinolitic amphibole never fails among the essential constituents; it also is probably secondary. A darker-tinted green hornblende is commonly present, and may represent a product of original crystallization from the magma. With this hornblende an original plagioclase, probably labradorite, is usually associated. The feldspar is always altered in high degree. A. little magnetite, pyrite, and avait and much titanite are accessories. The original rock was probably a basic diorite or a gabbro. It has been ‘greatly sheared and mashed and has degenerated into several secondary types. The commonest of these is a massive greenstone bearing a fair amount of the skeletal plagioclase and dark green hornblende which are regarded as primary in origin. This rock is often intimately sheared and slickened, but is scarcely a true schist. Toward the International Boundary the mass becomes distinctly gneissic, with the field-habit of a medium- to coarse-grained hornblende-diorite gneiss; under the microscope, however, this type was seen to be a hornblende- zoisite-quartz schist, the amphibole being of actinolitic appearance. In certain zones of specially intense shearing the rock has been converted into a garneti- ferous tale-quartz schist. The amount of shearing and alteration undergone by this gabbroid intrusive is of the same order as that seen in the Chilliwack volcanics, which have been referred to the upper Carboniferous. It is possible that this great Vedder moun- tain ‘ dike’ represents the intrusive phase of the same eruptions which gave rise to the surface flows of the Chilliwack formation. In any case the greenstone is certainly pre-Kocene and probably pre-Jurassic in age.* CUSTER GRANITE-GNEISS. On Custer ridge, which locally forms the main divide of the Skagit range, the Boundary belt crosses a considerable mass of crushed and now banded, intru- sive granite. Its exposed area is known to be at least twenty square miles, but it may be found to be much greater as the body is followed northward and south- ward from the Boundary belt. The western limit of the banded granite so far as mapped is fixed by the intrusive contact of the younger Chilliwack grano- diorite. The eastern limit is fixed in part by a band of the Hozomeen sedi- mentary series, into which the banded granite is intrusive; in other part, by the very thick blanket of Skagit volcanics, which are clearly younger than the eneissic granite. From its occurrence on Custer ridge this batholithic body may be ellie the Custer granite-gneiss. “During the preliminary examination of this district, in 1901, the relations of this schistose intrusive were not understood and the body was regarded as part of a basal crystalline series. In the Summary Report for 1901 (page 51) this series was given the provisional name ‘ Vedder Mountain gneisses.’ The writer wishes to with- draw this name which should obviously not be “perpetuated in the literature. 524 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 If the Hozomeen series is of Carboniferous age, the Custer batholith must have been intruded in late Carboniferous or post-Carboniferous time. The general relations and metamorphosed condition of the batholith point to a pre- Tertiary date of intrusion. The similarity in these respects to the Remmel and Osoyoos batholiths has led the writer to place the date tentatively in the Jurassic, thus making all three batholiths essentially contemporaneous. It is obvious, however, that such correlations among the older batholiths must be held with a very open mind, for they are founded largely on simple conjectures as to the ages of the metamorphic rocks cut by these batholiths. Until fossils are actually found in the Hozomeen series there is nothing to compel the view that the Custer batholith is of late Paleozoic age; it may, indeed, be an uplifted frag- ment of an old pre-Cambrian terrane. In the field the batholith has all the appearance of a typical pre-Cambrian gneiss. It is seldom quite massive, and at no known point has it escaped more or less powerful crushing and shearing. As a rule, the original rock has been converted into a well-banded gneiss, very similar to that produced in the meta- morphosed Cascade and Remmel batholiths farther east. Original Rock Type—The original -rock seems to have been a grano- diorite. Because of the intense metamorphism of the whole body, it is not possible to distinguish the primary phases into which the batholithic magma erystallized. Indeed, there are few places where the crushing and chemical rearrangement of the mass were slight enough to leave remnants of the original granite. One such locality was found at the head of Depot creek and about one mile north of the Boundary line. The rock there is crushed and somewhat eneissic, but it is not banded. It is of a darkish gray colour and of medium grain. The hand-specimen shows the presence of much hornblende, less biotite, and little quartz. From the persistently white to gray tint of the dominant feldspar one would suspect, from the macroscopic appearance, that one were dealing with a plagioclase-rock. That conclusion is corroborated by the study of thin sections. The essential minerals, named in the order of decreasing abundance, are: plagioclase, varying from basic andesine, near Ab, An, to basic oligoclase near Ab, An,; dark green hornblende; orthoclase and microcline; quartz, and biotite. The usual acces- sories, magnetite, apatite, and titanite, are present. The essential minerals all show straining. The plagioclase lamelle are often bent or broken, and some of the orthoclase has been converted, by pressure, into microcline. Nevertheless, there is no doubt that the specimen just described represents a common phase, and probably the dominant phase, of the original batholith. With this grano- diorite type the materials making up the bands which form the staple rock of the batholith at present, are in striking contrast. Banded Structure—As in the case of the Remmel, Osoyoos, and Cascade batholiths, the bands are here often of stratiform regularity. They may be - divided into two classes: one acid-aplitic, the other basic in composition. The acid bands are light-gray to whitish or very pale pink in colour. The erain varies from rather fine to quite coarsely pegmatitic. In the latter case it REPORT OF THE CHIEF ASTRONOMER 525 SESSIONAL PAPER No. 25a is sometimes not easy to distinguish such bands from the younger pegmatitic dikes (off-shoots from the Chilliwack batholith?) cutting the banded granite. In most cases, however, the pegmatitic habit of the light bands is apparently due to some recrystallization of the original rock of the Custer batholith itself. In thin section these light bands were seen to consist of dominant quartz, microcline, and orthoclase, with subordinate oligoclase (generally untwinned or poorly twinned), and biotite. A few, small, pink garnets, rare crystals of zircon and apatite, and small anhedra of 'titanite are accessory. Both hornblende and free iron oxide seem to be entirely absent. These bands have, thus, the composi- tion of many acid, aplitic granites poor in biotite. The component minerals are generally strained and the cataclastie structure is usual. The specific gravities of two fresh specimens of the light bands were, respectively, 2.655 and 2-641. The dark bands are of three kinds, according to the character and propor- tions of the constituents. The commonest kind is a dark greenish-gray, foliated, medium-grained, highly biotitic rock, composed of dominant plagioclase (basic andesine), biotite and quartz, with rare orthoclase. A few grains of garnet, some magnetite, and apatite are accessory. One specimen showed a specific gravity of 2-732, but many bands, yet rieher in biotite, would be heavier. Only one thin section of this type—a biotite-diorite gneiss—was studied. It showed neither granulation nor pronounced straining of the component minerals, and it seems necessary to believe that the material of these dark bands was erystallized in its present form during the metamorphism of the batholith and has not since been subjected to extraordinary orogenic stress. Dark bands of the second kind differ from those of the first in carrying essential hornblende as well as biotite in large amount. No special study has been made of these, but they doubtless have the same principal features as the biotite-diorite gneiss, excepting for the entrance of essential hornblende. Bands of this class have the composition of basic hornblende-biotite diorite gneiss. Basie bands without essential biotite are uncommon but were noted at several points. In these green hornblende is the only important femic mineral. Basic labradorite is the only other essential constituent. Much apatite, very abundant, well crystallized titanite, and some pyrite are the observed accessories. The specific gravity of a somewhat altered specimen is 2-888. Bands of this third class seem to range in composition from amphibolite to hornblende-diorite gneiss. A few of them may possibly be sheared basic dikes cutting the batholith, but the majority, like the other dark bands, must be regarded as forming meta- morphic phases of the sheared batholithie rock. The Custer batholith thus includes the following species of rocks :— Original type: granodiorite. Secondary, metamorphic types: Biotite-aplite gneiss; Basic biotite-diorite gneiss; Basie hornblende-biotite-diorite gneiss; Basic hornblende-diorite gneiss; Amphibolite. 526 5 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 The explanation of the banding is here the same as that offered for the banding of the Osoyoos, Remmel, and other batholiths farther east. The light bands represent the intensely granulated diorite from which the hornblende, biotite, basic plagioclase, and accessories have been slowly leached during the shearing of the batholith. The dark bands represent the shear-zones in which the same basic materials were recrystallized. In many cases there has also been some recrystallization of the light bands with the development of new quartz, biotite, feldspar, and some garnets. SuMAS GRANITE AND DIORITE. Rather more than one-half of Sumas mountain is composed of plutonic igneous rock. Nine square miles of the central and northeastern parts of the mountain are underlain by a biotite granite which may be called, for conven- ience, the Sumas granite. An area of about three square miles is underlain by a plutonic breccia. This breccia consists of a vast multitude of blocks of a dioritic rock cemented by the Sumas granite; the whole forms a peripheral intrusion-breccia on a large scale. The diorite is evidently the older of these two rocks and may be called the Sumas diorite. On the north and east the plutonic masses disappear beneath the Fraser valley alluvium. On the northwest the granite makes contact with a hard, massive quartzite, into which it is intrusive. On the southwest the granite is unconformably overlain by the nearly flat Eocene (?%) beds. Granite-—The granite is pre-Eocene in age. The date of intrusion cannot yet be more closely fixed with certainty. The rock is nowhere crushed in any notable way. It seems therefore doubtful that it was intruded before the great orogenic revolution of the Jurassic and the date may be tentatively fixed as later Jurassic, or (less probably) Cretaceous. The diorite of the intrusion- breccia is also massive and unsheared and may belong to the same general period of igneous action, though of course, being older than the granite. The granite is a light pinkish-gray, fine, to medium-grained rock, poor in _ dark constituents. The composition and structure are both those characteristic of mica granites. Quartz, orthoclase (sometimes slightly microperthitic), basic andesine, averaging Ab, An,, and biotite are the esseniials. A pale green horn- blende, magnetite, apatite, titanite, and rare zircons are accessory. In all of the four specimens collected, the rock is seen to be considerably altered. The alteration is so marked even on, well glaciated ledges that one may possibly refer it in largest part to the secular weathering which preceded the deposition of the Eocene beds. The feldspar is often much kaolinized and the biotite is generally chloritized to some extent. The freshest specimen (No. 201) has been analyzed by Professor Dittrich with the following result :— REPORT OF THE CHIEF ASTRONOMER 527 SESSIONAL PAPER No. 25a Analysis of Sumas granite. Mol S10: : 71-24 1-187 1 Qee ° +4 -005 Al,O,.. 14-11 -138 203. 1-75 011 FeO.. 1-23 -017 Mn0O.. ‘ tr ess MgO.. 1-07 027 CaO 2-87 051 BaO 09 001 oun Severe niee ts ay 038 H.O at "710°C... oe ae ae aisha CRM aeee CeeM RE VaMENe melee toe Nuc? tale “11 hiniste HOt abone. 110°C... he) Tae en ed Oa Sy aeRO REE Sr ae eee re 59 ate 2 A330 ee . ee ee ee ee ee eo-ce eo e080 ©8 88 88 #8 17 6001 100-27 Spare eare ein orem ereaiclon note asics. sie aan stellleieieie: (eicaishewh ete a sireien ricsOOL The calculated norm is:— QA b Zee ae elaterah oe lehire Raa zal earaeie oe bscen datemdated reteukinlay lst econsevan nerees epeu sere 34-86 OTE MNO CLAS Ehespee eter ote rcnsUarets nce. lovetursicbe: sieteatar slbsisled baveultemnunerey icin ciel tare 23-35 TANI TG Pa ois orceteye eaten ae Sn SiSM SSR Te eave, ain cose evan iste saisusvalen terme eher deren eye 19-91 UNS OTE Obert rdes re EE Rte ke Cette acre oisdiayes Gok aie Mites oiel cieieieve s 13-62 MOOT UTI ena eee ai cet ee rere eneeL chew ial: Santarete sich ordi selelwetennaee asks 92 ERSTE FUSING SS: Sol oo. aoe Wood bo odeoo: Ot.wo Ion sdueberctloomonouncd 2.83 WBA ReRod GoGo Go" So Wo. 05 66. dG UOn>o ido noosdocadedouoomdd. on oo 2-55 IDTANR ERAS G6 lob coe: oboe 2bo a moo NOmecal OosbOMo OWiGG Base Gncco00 io “7 Apatite... .. AS ePr ce Ra GTS ion sess ho Ouse Niciolneloticven ore eisle Wietemnatineeediete 31 Water and COM tet NTR Caren rots O ee nC TRIS AGU oe Riad ed rate gtotecaines 98 100-09 In the Norm classification the rock enters the sodipotassic subrang, riese- nose, of the alkalicalcic rang, riesenase, in the persalane order, columbare; but it is very close to amiatose, the corresponding subrang of the order britannare. In the older classification it is obviously a common type of biotite granite. The specific gravities of two of the freshest specimens are 2-651 and 2-653. On account of the alteration of the rock a useful determination of the actual mineralogical composition by the Rosiwal method is practically impossible. The mode is in this case not very different from the norm. Diorite-—The dioritic rock of the intrusion-breccia is also considerably altered, as if by weathering. Its essential minerals are green hornblende and plagioclase, averaging basic andesine. Orthoclase is an abundant accessory and some interstitial quartz is always present. Much epidote with some chlorite and kaolin are the secondary minerals. The smaller diorite xenoliths in the granite have been more or less com- pletely recrystallized and modified in composition by the granitic magma. The hornblende there characteristically occurs in long idiomorphic blades shot through the feldspar and other constituents. In one thin section of a xenolith, potash feldspar and quartz are so abundant as to cause the rock to simulate a granodiorite; in that case it seems probable that the original diorite has been 528 - DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 affected in its composition by the introduction of material from the granite. The metamorphic effects are analogous to those observed about the xenoliths in the Moyie sills of the Purcell range. In other xenoliths which show in their rounded outlines the corrosive effects of the acid magma, a large number of peculiar round bodies have been developed. These are of the size and shape of a small pea and, because of their special hardness, they project above the general weathered surface of the xenolith. Under the microscope each of these small bodies is seen to be composed of pure quartz, generally as a single crystal, but sometimes in the form of a coarse- grained aggregate. The quartz nodules are perfectly clear and bear no inclu- sions of the dioritic material. Between the diorite matrix and the quartz there is usually a narrow aureole of idiomorphiec orthoclase and plagioclase crystals. These project into the quartz much as similar crystals project into vugs and miaroles of other rocks. The origin of these silicious nodules is not clear. They can hardly be regarded as filled amygdules of the ordinary type, but seem rather to represent phenocrystic growths in the xenolith after the latter had been softened by the granite magma and been impregnated with silicious material from that source. Sxacir Vouicanic ForMATION. From the first summit west of the Skagit river to the summit of Custer ridge (the main divide of the Skagit range), the Boundary line crosses a very thick group of voleanic rocks which may be called the Skagit voleanie formation. These rocks extend over at least twenty square miles in the belt and continue unknown distances in the mountains to north and to south. The volcanic rocks characteristically weather into jagged peaks, knife-edge ridges, and forbidding precipices, forming the highest and most rugged moun- tains in this part of the Skagit range. (Plate 44, B.) Glacier Peak and its neighbours are, indeed, among the most inaccessible summits in the whole Boundary belt west of the Flathead river. Small but numerous glaciers and a succession of impassable breaks in the ridges render the study of the volcanic formation difficult even where outcrops are plentiful. Below tree-line it has so far proved quite impossible to find a sufficient number of actual contacts or to work out the succession of the many members of the group. The results of the exploration are, therefore, far from being satisfactory. It is known that the formation is exceedingly thick—apparently at least 5,000 feet thick at the Boundary line— but the writer has been bafiled in the attempt to construct a ‘detailed and final columnar section. The great thickness of the volcanic accumulation and the abundance and coarseness of the agglomerates suggest that the major eruptions actually took place in the area of the Boundary belt. It is quite possible that a vast cone of Mount Baker or Mount Rainier pro- portions was situated over the present site of Glacier Peak. The lower and greater part of the formation, probably 4,000 feet or more in thickness, is composed of massive breccias and ash-beds, with one layer of coarse conglomerate and with many interbedded flows of compact and vesicular lava. REPORT OF THE CHIEF ASTRONOMER 529 SESSIONAL PAPER No. 25a All of the purely voleanic constituents are andesitic. Conformably overlying these rocks and underlain by other andesitic flows and breccias, comes a widely extended layer of white to pale-gray trachytic or rhyolitic tuff, aggregating perhaps 200 feet in thickness. The top of the whole group is not exposed in the Boundary belt and the series remains incomplete. The following table gives an extremely crude idea of the general relations and thicknesses as estimated in the field :— Top, erosion-surface. 1,000+-feet.— Andesite flows and breccias. 2004- Liparitic (?) tuff. 900-- * Andesite flows, ash and breccias. UCN Conglomerate. 3,000—- ** Andesite breccias, flows and ash-beds. 5,200--feet. Base, unconformable contact with Custer batholith and Hozomeen sediments. The andesitic members are always very massive. It is seldom possible to distinguish the contacts between different flows, and even the contacts between flow and breccia are generally obscure. The individual flows seem to be usually very thick; cliffy slopes as much as 300 or 400 feet high do not disclose undoubted breaks in the massive lava. The more basic material of the breccias, ash-beds, and flows has great uni- formity of composition. Nine typical specimens were collected in different parts of the area and at various horizons from near the base upward. ‘Thin sections of all the specimens were studied. Though not crushed they are all more or less altered. Without exception, the flows and lava-fragments of the agglomerates seem to belong to the one common type, augite andesite. The phenocrysts are regularly labradorite, averaging Ab, An,, and augite. The latter is generally uralitized pretty thoroughly. Neither primary hornblende nor olivine was detected, though in some cases the former may have accompanied the augite as a subordinate phenocryst. The ground-mass is more altered than the pheno- erysts and is a mass of chlorite, uralite, plagioclase microlites, and indetermin- able material, perhaps derived from glass, which was apparently a very-abundant, staple constituent of the ground-mass. The beds of agglomerate are usually thick, individual ones measuring more than 200 feet in thickness. The blocks are of all sizes up to those a foot or more in diameter. At many points angular fragments of cherty quartzite and slate, identical in look with the dominant rocks of the Hozomeen series, were found in the breccias. At one section in the deep valley northwest of Monument 68, and about 1,200 yards from the monument, there occurs a layer of breccia wholly or almost wholly made up of fragments of cherty quartzite and serpen- tine; this bed is at least 75 feet thick. The fragments are angular, ranging in size from sand-grains to blocks six inches or more in diameter. There can be little doubt that these fragments were derived from the Hozomeen series. The bed shows no sign of water-action; from its position in the midst of manifest volcanic agglomerates, it may best be regarded as a special product 25a—vol. ii—34 ; 5380 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 of gas-explosion which operated in this vicinity and blew out a large quantity of the foundation rock. The matrix of this bed was not examined microscopi- cally; it may be a fine andesitic ash. Sometimes, though rarely, granitic and gneissic blocks appear in the staple breccias; most of those observed seem to have been derived from the underlying Custer batholith. a About one mile northwest of Monument 68 a bed of coarse conglomerate, 100 feet or more in thickness, interrupts the succession of breccias and flows. The pebbles are very well rounded and were unquestionably long rolled by waves or currents. They vary in size but few are over a foot in diameter. They consist of altered andesite (dominant kind), quartzite, chert, slate, and, rarely weathered granite. The matrix is sandy. The bed dips about 16° to the eastward and seems to be quite conformable to the yet more massive volcanic members above and below. Above the conglomerate the tuffaceous rocks carry several thin conformable lenses of gray argillite, which also appear to have been laid down under water. The acid tuff was seen at two localities. It crops out on the summit of the rugged ridge 1-5 miles south of Monument 68 and on a much greater scale upon the long ridge running eastward from Monument 69. .This tuff covers the latter ridge for one mile of its length and from its white colour is very conspicuous in the landscape. The tuff is extremely jointed, so that it is difficult to secure a hand-specimen of standard size. Some of the rock is vesi- cular and it is possible that thin flows are represented in the middle part of the 200-foot band. The whole composite mass overlies the andesites, dipping at angles of from 10° to 15° to the north. On the higher ridge on the west the ucid tuff seems to be overlain by younger andesites roughly estimated to be 1,000 feet thick. The acid tuff is nearly pure white to pale-gray when fresh, weathering white to pale-yellow. Macroscopically it is quite aphanitic for the most part, with only the rarest suggestion of a small feldspar phenocryst. _The rock reminds one ct porcelain viewed on a broken edge. Under the microscope the phenocrysts of the angular fragments are seen to be few in number and to have the proper- ties of sanidine or orthoclase. The ground-mass is a cryptocrystalline aggregate of quartz and feldspar, with the character of a devitrified obsidian. The matrix of the tuff is optically like the ground-mass of the fragments. The mass has clearly the composition of an acid obsidian and is perhaps nearer trachyte than rhyolite. The age of the formation has not been determined by direct fossil evidence. The lava-flows, ash-beds, breccias, and interbedded conglomerates are not crushed. The dips are generally low, running from 5° to 30° as the observed maximum. The breccias and conglomeratic beds contain many fragments and pebbles of quartzite, slate, and granite which were without much doubt derived from the eroded Hozomeen series and the Custer gneissic batholith. It seems reasonably certain, therefore, that the vuleanism dates from a period much later than the intrusion of the batholith and, 4 fortiori, than the folding of the Hozo- REPORT OF THE CHIEF ASTRONOUER 531 SESSIONAL PAPER No. 25a meen series of sediments. If the latter are of Carboniferous age and the granite is Jurassic, the Skagit volcanics rest upon a late Jurassic or post-Jurassie erosion- surface. The relation is somewhat similar to that between the voleanie breccia at the base of the Pasayten formation and the underlying, probably Jurassie Remmel batholith. There is something, therefore, to be said for the hypothesis that the Skagit volcanics are of Lower Cretaceous age and contemporaneous with the Pasayten voleanics. If, however, the Custer batholith was intruded in the late Jurassic and sheared and metamorphosed during the orogenic revolution at the close of | the Laramie period, it would seem certain that the Skagit voleanics must belong in the Tertiary. This follows from the fact that the volcanic rocks are com- paratively little disturbed and are nowhere sheared in anything like the measure shown in the Custer gneissic batholith. It would seem impossible that the basement could be so profoundly affected while the thick cover should escape the deformation. That the Skagit volcanic formation is not younger than the Miocene is probably indicated by the fact that it is cut by a stock of quartz- bearing monzonite, which shows evidence of being essentially eontemporaneous with the Castle Peak stock (late Miocene). At present the dating of the vol- canics cannot be made any closer with definiteness. In the correlation tables the writer will postulate an Oligocene date for them, thus equating the Skagit andesite with the proved Oligocene andesite in thé Midway district. The Skagit andesite may, on the other hand, be Eocene or, possibly, Cretaceous. SKAGIT HaRZBURGITE. On the ridge 2,500 yards north-northwest of Monument 67, at the 6,600- foot contour, the Custer gneiss is cut by a large pod-like intrusion of coarse peridotite. This mass is 150 feet or more in width and can be followed along its longer, north-south axis about 900 feet. It appears to taper off toward each end. It is probably an irregular dike injected into a schistosity-plane of the gneiss. From wall to wall the peridotite is very coarse, showing olivines often reaching 2 em. in diameter and an abundant pyroxene of similar dimensions. At the ledge the rock is seen to be somewhat altered, but it shows no sign of erushing. The general colour of the rock is a deep, almost blackish, green. Feldspar is entirely lacking. In thin section the composition and structure are seen to be that of a typical, partly altered harzburgite. The only primary minerals are clivine and enstatite, both colourless in thin section. About fifty per cent of the rock is made up of secondary minerals, including serpentine, tremolite, iddingsite, tale, chlorite, much sulphide (probably pyrite), and con- siderable limonite. Minute inclusions of picotite or chromite could be discerned in the olivine. The iddingsite noted has most of the features described by Lawson for the type material at Carmelo bay, but the optical angle is very small, 2V being well under 5°. The specific gravity of the rock described is 3-083. The date of this intrusion is apparent only in relation to the period when the Custer batholith was sheared; the shearing seems to have been completed 25a—vol. ii—344 532 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 before the peridotite was injected. The proximity of the Skagit voleanics over- lying the gneiss leads one to suspect that these basic rocks belong to the same eruptive period, and that the harzburgite and andesites are genetically connected. The relation is conceivably the same as that connecting the peridotites of the Columbia range and Midway mountains with the basalts and andesites of those regions. : SLESSE DiorITE. The walls of Middle creek canyon in its lower part are composed of diorite, which extends over the divide past Slesse mountain to Slesse creek. The diorite forms a stock-like mass covering about nine square miles on the Canadian side of the line; it was not mapped to the southward, but it is known to extend several miles into Washington. ‘The diorite body once undoubtedly stretched farther eastward, but it has there been replaced by the younger-Chilliwack batho- lith of granodiorite. The diorite is very clearly intrusive into the slates on the west and north. These argillites are highly altered, but, as they enclose lenses of crinoidal Car- boniferous limestone, it seems most probable that the date of intrusion is post- Carboniferous. The diorite is not crushed or greatly strained except in the immediate vicinity of the great Chilliwack batholith, where such effects might naturally be expected. Elsewhere there are no evidences that the diorite has undergone the severe pressures involved in the post-Cretaceous. mountain-building of the Cascade range; it is therefore probable, though not proved, that the diorite was intruded in post-Laramie time. The contacts of the body are so imperfectly exposed in this densely forested area that its structural relations have not been fully worked out. The diorite certainly cross-cuts the sedimentaries and has metamorphosed them in the thorough way characteristic of most stocks. The intensity of the metamorphism is of a higher order than that usually observed about a laccolith or chonolith, and it seems safer to regard the mass as a true stock or batholith, that is, a sub- jacent, downwardly enlarging body. Fhe diorite is in places richly charged with large, slab-like inclusions of crumpled black slate; these often attain lengths of 50 to 100 feet or more. A large number of them, forming a veritable breccia on a great scale, may be seen on both slopes of Middle creek canyon, especially at points about four miles from the confluence of the creek with the Chilliwack river. Petrography.—tThe diorite is a dark brownish to greenish gray, fresh rock of normal habit. It appears to have a rather uniform chemical composition. The chief variations are those of grain. At its own intrusive contacts the stock is fine-grained as if by chilling; elsewhere the grain is generally of medium size. Where the diorite contacts with the younger granodiorite the grain is still medium, but the more basic rock has been metamorphosed along a narrow zone. Basie segregations were not observed in the diorite. The list of essential minerals in the diorite includes acid labradorite, near Ab, An,, hornblende, and biotite, named in the order of decreasing abundance. REPORT OF THE CHIEF ASTRONOMER 533 SESSIONAL PAPER No. 25a The hornblende sometimes, though rarely, encloses small cores of nearly colour- less augite; the latter mineral also occurs in small independent anhedra, but is clearly only an accessory constituent. The other accessories are quartz, magne- tite, pyrite, apatite, and titanite. The structure is the usual hypidiomorphic- granular. The order of crystallization is not very clear, but much of the plagio- clase seems to antedate the biotite and hornblende. A type specimen (No. 54), with the mineralogical composition just described, was collected on Middle creek in the heart of the main mass. It has been analyzed by Professor Dittrich, with the following result :— Analysis of Slesse diorite. Mol aie : 000 ae 1035.) . 1 INNO Ra 18-17 178 Fe.,Q;.. 1-23 008 FeO.. 5-88 082 Mn0O.. 21 003 MgO.. 4-36 109 CaO g3 6-51 116 SrO : 18 002 Wa20 3-23 052 1-57 017 al at 110°C... PSE Ged DA OCI Mor nee GIS SET Ca ARE ROL TENS 12 HL.0 above 1 110°C... SN AICP ST at baa ove ae IEE Mag 77 Pp P : Oe ae Salk Oras dss eigeten eee eaaiy Merle! 08 100-15 S]Bo Gikoc. a5. 09) 00 104 BO! 00 60 06 bO08 06 “0008 Soolmoo! on 2-793 The ealculated norm is:— QUUENAWo5-00 6o.00100 00. 00. do) 0b Ua 0000 G0 ‘90 Oo od) Gu on) Go Ae) 66 8-04 OTE OCTASS ess Sree ae oe Orel eis oko LO RCS TRAM S ON aa apa Chee emia ks 9.45 PAU TER te ees oe eee el wi eicten eter eie a creltiavs Moreh or mereteapanac abe aren neonates col ure 27-25 AS rae ra end GG ey tae cs cel vite a) rarae ie she rotetin ia ela even iaved chs lapaseleyalobers eae atch Darou vekaoheiat, 1a aces 30-30 IEEV DETSENSTCricn com reres elecdieen testes eee eI ake caves Meleleeerore (ore teva ereiareca galt clas 19-08 TOP ST Ce Pe ere ie tees Nae eatin ai ake ye ssa ast rcuie ar aaa ts Mureep ics etutiaes bey cur ieee nel 1-36 Maire ibe aveue yur sayoiirctetee ous oi cis stale autalenvoladueis Hisboe tee Bon ladntarbatalaveretics 1-85 STilivTe tte cer eve: crass, ious houmetome connie os by ela eee ISG. Sem calets ih uaiole trey suecs 1-52 Apatite.. .. SO eater rear a) oh cf ee ata oe ee AAMT ol inte orecmaelon nexeuaraes 31 Water and CO: Sree ar Met e eainane tires eldrteracitero LPS G aes elk ela paris eans Mavala erates eal 98 100-14 The mode (Rosiwal method) is approximately :— Quartze. 2. Se are ca ar alot acecbnres oe Ien Veta) Wap ate aha: SRE Mg ea ea 9. Labradorite.. Or OOo OE POOTIIG, POO OR Oey a ey ary OIE Tt Neate ae eae 58: Hornblende .. A ois Se Be ONO eNOS OG BBE OT Re a eee 12 SLOG COM es Dee rete Shp eens ese Nar hs etiel a burs’ Se umenea an Migrar rare Tash Stat 12.- PAST D1 CO MAGE lire tare rrsuatavcuaaeune ca careua ones rei sah aishn ie tel sTet colony abe MeavPuUrol eveaiecam Suartavou ete 4 MAYTAG evs vere theta: aor. slon oan cron oie utero liie: -slofereeveives 1 MOM TUCO si ey) eee cian Moreton Tt MEL ek eau are Naeicus Apatite.. SUNT Bese aera eer Ca Rar ae Titanite ‘and ‘zircon. Anees atreltetd Velen olen ts se tack eran einicrdianwock = i] 884 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 In the Norm classification the rock enters the dosodic subrang, andose, of the alkalicalcic rang, andase, in the dosalane order, germanare; but is near the corresponding subrang of the docalcic rang, hessase. In the older classification the rock is a hornblende-biotite diorite. Miner. alogically and chemically it is almost identical with a California diorite des- cribed by Turner.* The specific gravities of six fresh specimens vary from 2-786 to 2-863, averaging 2-806. The apophyses of the body are chemically similar but have the structure of hornblende-biotite diorite porphyrite. Contact Metamorphism.—The Paleozoic sedimentaries cut by the diorite have been decidedly metamorphosed. The effects were noticeable at all of the observed contacts, but were specially studied on Pierce mountain which forms part of the rugged divide running southward between Slesse and Middle creeks, and again along the contacts on Middle creek. The belt of altered rock seems to average at least 600 feet and may be 1,000 feet or more. (Plate 44, C). Mineralogically the metamorphism shows nothing very unusual. The sandstones have been converted into tough, vitreous quartzites. Some of the argillites have been changed into dark greenish-gray hornfelses or schists, richly charged with metamorphic biotite and sericite. Other argillaceous beds have been recrystallized, with the generation of abundant cordierite, that mineral forming, as normally, large, interlocking individuals which are filled with inclusions of quartz, biotite and magnetite. A few thin lenses of pale green, felted tremolite and of more granular tremolite and epidote probably represent completely altered beds of limestone; other limestone bands have been changed to white marble. With the limestones much chalcedonic silica is often asso- ciated. The contact-belt is often traversed by small-quartz-veins, some of which form fairly high grade, free-milling gold ore. The Pierce claim on Pierce mountain is located on one of these veins, close to the main contact of the diorite. Like all the others seen in the vicinity this vein is quite variable in width, pinching out,irregularly from its maximum width of a few feet. At the time of the writer’s visit to the claim, in 1901, not enough development work had been done to show the amount or average value of the gold-bearing quartz. A similar, though narrower vein, nine to twenty-one inches wide, cuts the diorite at a point about 700 feet above Middle creek and 3,000 feet or more below the main claim on Pierce mountain; the two veins may be connected, and both were being opened up by Mr. Pierce in 1901. From the writer’s experience the veins occurring along this contact must be very high grade if they are to pay for their development; they are much too small and irregular to give hope of profitable low-grade ore. CHILLIWACK GRANODIORITE *BATHOLITH. Some of the wildest and most rugged mountains in the Skagit range are composed of a massive granodiorite which forms the largest intrusive area in - . *See Bulletin 228, U. S. Geol. Survey, 1904, p. 234. OSUBY JLSVyG JO giuauUANsS AvoU “(YPPOYYVE xouvmipLyH) sur~gunour og1uwas Jo morta peord Ay, ‘Oey OVA TLS JO Jsvo Qt UUs WOT 4S JOY INOS SUIYOOT p. 534. vol. ii 25a REPORT OF THE CHIEF ASTRONOMER 535 SESSIONAL PAPER No. 25a the Boundary belt west of the Remmel batholith. The basin of Chilliwack lake has been excavated in this rock, for which the name, Chilliwack granodiorite, has been selected. The body has the size and field-relations of a typical batholith. (Plates 44, D, 47, and 62, A). On the Canadian side of the Boundary line the granodiorite underlies at least 100 square miles of mountains. The formation stretches an unknown distance to the northward of the Boundary belt and also continues a few miles on the United States side. A couple of miles north of the Boundary line, and a like distance west of the Skagit river, a small granitie stock, of composition probably similar to the more salic phase of the Chilliwack batholith, cuts the Hozomeen series. Owing to bad weather and to other causes, the writer was unable to examine this western slope of the Skagit valley. At his request, Mr. Charles Camsell, of the Dominion Geological Survey, mapped the formations on this slope, and special thanks are due him for this service. He discovered the small stock and has referred its date of intrusion to the Tertiary. As yet the rock has not been studied with the microscope. The date of the intrusion of the main batholith can be fixed within certain limits. The granodiorite clearly cuts the greatly deformed sediments on the long ridge northwest of the lake. In that region the strata are unfossiliferous but appear to belong to the same group as the definitely Carboniferous beds west of Middle creek canyon. The granodiorite cuts the Slesse diorite, forming a wide belt of intrusion-breccia with the latter where the main contact crosses Middle creek. The diorite just as clearly cuts fossiliferous Carboniferous slates and limestones. It follows that the granodiorite is of post-Carboniferous date. At no point does it show evidence of crushing or of pronounced straining; as in the case of the older diorite, there can be little doubt as to the relatively late date of the intrusion. In field-habit, as in many essential microscopic details, this granodiorite is like that of the post-Cretaceous Castle Peak stock. There are, thus, some grounds for the belief that the Chilliwack granodiorite was, like the granodiorite at Snoqualmie Pass to the southward,* intruded at a date as recent as the Miocene. In the field the batholith preserves great uniformity of colour, grain, and massiveness. It was only after microscopic examination that its actual varia- tion in composition became apparent. Three main phases were recognized from the thin sections. Petrography.—The most basic phase of the three is a quartz diorite rather than a true granodiorite.. It occurs along main contacts and also at points two or more miles from any visible contact; so that it is apparently not the product of simple contact-basification. A type-specimen was collected at the Boundary line in the lower of the two cirques occurring in the mountains southwest of the upper end of Chilliwack lake. This point is at least two miles from any lateral contact and probably at least a mile from the original roof-contact. The rock is exposed on a great scale-on the steep, 4,000-foot slope to the westward of the * See page 469. 536 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 lake and upper river. The walls of the tandem-cirques seem to be composed throughout of the quartz-diorite phase. The typical specimen is a light-gray, medium to moderately coarse-grained, granitic rock poor in quartz and speckled with abundant, brilliant prisms of hornblende and black foils of biotite. The microscope shows that the dominant constituent is an unzoned plagioclase, averaging labradorite, Ab, An,. Ortho- clase appears to be entirely lacking. The amphibole is a common hornblende with an extinction of about 17° on (010) and colour scheme as follows :— a — pale yellowish green with olive tinge. b = strong olive-green. c = olive-green. b> ¢ >a: The mica is a common brown biotite with the usual strong absorption. Quartz is interstitial and in relatively small amount. Magnetite, apatite, and rare zircon crystals are the accessories. The order of crystallization is: the accessories ; een plagioclase, followed in order by biotite, hornblende, and quartz. The structure is the eugranitie. Professor Dittrich has analyzed this phase (specimen No. 7), with the following result :— Analysis of quartz diorite, Chilliwack batholith (phase 1). Mol. Gree hie a sea cater cle aac, ae op miele commas, alert e Bean eta venercee tain abe 60-36 1-006 THOse: ; SAUER pu pe a hb ci lal ace eis oak MN Nay -70 009 Al,O,.. 17-23 169 Fe,O;.. 1-93 012 eO.. 3-74 052 MnO.. 14 002 MgO.. 3-66 093 CaO.. 6:07 109 Na.O.. 3°58 058 K,O.. Se cert meray GIN ra ere ce rae erin Rte en 1:74 -018 HO at. 110°C.. fis ark avaess saa ae ee Gea invey Ve eee eh OEE REN 06 Ses H.O above 110°C... SLES 122 Re ng erepe etnias as an ae tae came 55 Bes P,0;.. si Ds ire Uae ea ache ena ene ae: 11 001 He Igri 08 sates 99-95 SAO LE cacy cbiticn) wisbieist coven sel iete occu nyepate ees crake eae: reer eee ene 2-757 The calculated norm is:— (OTE oh qe Aare NE Ae ae RE Reena aL e aaah as el ee ane Mba ae heli 0 A 13-56 Oxthoclasessk = 05 Se A aia crore ce ea eo ogee ee 10-01 JOST eee Ree ener OR a ee SME MAEM MGT pan lan’ Eat ui “alg 30-39 PNeMOT ENTE eye cie! rence wars verse tereoc Nas Meret NO nares Tot Lisa Sacer: ae ee 25-85 ey.persthe4ne fotos ee Reine oan fs alitiorcss tal ne aie en ee Woe 12-16 ATO DSTO eetes hess AF ays oe te tee mapastapegs alae pctv aokst rec: es cag ca ae 2-90 IW TK21 ate ern heen er OMe Ene are Cem nO Ne ReOR eM SoM y a6 Ga: 2-78 TETTON TEC ai ak ee a ee ie eae rahe aE Ete ee eo en ee 1-37 Apatite... .. Sen aetna EES OG ea GU Mat lis las 31 Water and CO... b Shvtaelbtee aeieng wie aeGble iol olbaetan arene Osl aici) eee aaa RCT 69 REPORT OF THE CHIEF ASTRONOMER 537 SESSIONAL PAPER No. 25a The mode (Rosiwal method) is approximately :— QMORHA.S ed Ss Bo dog 66-66 66 6006" Gb 05 0d 06 0b co dolpo 1n6 on 19-1 Onmehoclase ee eee ee Oe Eo Mea D ON eHt eM aNy seme icmvee aye 2-0 GAOT OTIC ee ty ee ek a otnigeee Meccan eae cetera 55:7 Hormblend esse. ce eee ae ee ae Ciel aE EC Rie Cn iat oisl Vays Gas 11-1 TB TO GUO Se eee eee ne Reeds ae ae Pon aS venta caa cetntetehonlen eerie” faite Daphne 10-3 Magnetite.. 5 1-3 Apatite and zircon. 3) 100-0 In the Norm classification the rock enters the dosodic subrang, tonalose, ct the alkalicaleic rang, tonalase, of the dosalane order, austrare. In the older classification it is a typical quartz-biotite-hornblende diorite. A rock which appears to be a second phase of the batholith, was Cale at the western wall of the canyon of Silver creek where it debouches on the valley-flat one mile north-northwest of the lower end of Chilliwack lake. This type is a fresh, light pinkish-gray granite with abundant quartz and _ biotite, but with no hornblende. Orthoclase is an essential plainly visible as such to the naked eye. The dominant feldspar is again plagioclase. It is often zoned, the outer rims reaching the acidity of the mixture, Ab, An, The average mix- ture seems to be near Ab, An,. The orthoclase is often somewhat microperthitic. Magnetite, apatite, zircon, and a little titanite are the accessories. Professor Dittrich’s analysis of this phase (specimen No. 30) resulted as follows :— Analysis of soda granite, Chilliwack batholith (phase 2). S10. . 71-41 1-190 i0, bd 004 Al.O... 14-38 0141 HesOn: 1-33 008 eO.. 1:17 016 MnO 04 ieee MgO 1-13 028 CaO. 2-51 045 BaO. 03 Sous Na,O 4:12 066 K, 2:97 032 HO AE ALOCC os ee 09 pee H,O slove 110°C... Ba RA ire IC SCRE Eero tener ne ali -80 Sate IPO he 6c SAA Na Neko elt Ee Oates Fae Seen Res Gee ene NER 13 001 100-07 TS Dt REASONS TC eee ar 2-653 538 DEPARTMENT OF THE INTHRIOR 2 GEORGE V., A. 1912 The calculated norm is:— QU ATE Ze ee ee are ic anninedoe Oar ay (agen) ee nena ie alec SL Ree Gea 29-16 _ Orthoclase.. 0 NG euiod! Go-b0 G0 -d900"G0 60 4b) G0 60 00 00 00 oc JOR CMOS Ge Aan Hane Wek ner ot aero mE GleC SL EUR ad oa eee Uren SUS 34-58 PAO ETUC ciel alecetnarer oe Vaart coven tate ithe tele ree Oe tcl Siao nee NSS eee TS UES EEO 11-68 (OFT era Kin None ones AS Par EN Ah ci ee Ae nua Rulers resales ania mr iaicher sions s5 10 METVPELSUMOT Cio airyeiescccarse ceo ahs ene LUO lo esc Lil E Gre Oe OLS ORO RRS Re ie estes 3°33 MEA OMMGEDE OT ote seta me caug cel aidi en ale: els Tateae es rareceredmetaro olen ereraniema eens 1-86 AMMO MIGO oe aren ime oe co be ecto ete, ie oe ee erictal « Leaeare aura deere nsamray 61 J: OER GY Fie GesO TREO MRO CRE ETE MSH ER ELDON On Wee SG aap Aa nat “31 Wrater and) | COs ccc oe cca ilene, tone slaacclel alethelacleves ere aloua cetare eats 51 99.93 The mode (Rosiwal method) is approximately :— QM aTEZE ctercuisiet cers aie ate cassthae oh aes seveltial Pele k aleubel oukiiee eunmheloetcu creer 34-1 Orthoclase and microperthite.. .. SAGA Rate vncyorabot bes ceaeee trate kates 25-7 Olipoclase ss ota Ses rows aesiietee cies vercuelcle. Merecsinter aver telele sei leh ome areuncainers 30-2 IBiOtitere eis. (eee) ot roew seals Sone eS tee ele mL Seta Una repae dis oie a rate aera theta ate 8-1 Mia enetites': ct. Soe evel vale ecoecee ele tele stale oi aes evn Pelee eee eras 1-1 BTCA MTG Oe ie etree ee icra ares nora nliacer Mele eave elcuucots Morey i) a chemtoreiaielice age stenrere 4 PAD ATIEO CS Vai ceis. tare: vetol revel avery ieaietarel Pete t Sich cig rere vane raremem one Wave enet anee rene 3 ZAP COT 0 Taal Boley ioe aha See esate e ie PRIN AiR RTE a Shree SHE ae hea ace nang Rays 1 In the Norm classification the rock enters the dosodic subrang, lassenose, of the domalkalic rang, toscanase, in the persalane order, britannare. In the older classification it is a biotite granite with dominant oligoclase—a soda granite. So far as observed, this rock occurs only on the north side of Chilliwack valley and north of the lake. It may conceivably represent a distinct intrusive body, bearing the same relation to the hornblende-labradorite phase of the main batholith as the Cathedral granite of the Okanagan range bears to the Similka- meen granodiorite. Yet no sharp contact between the granite and diorite phases was found, and the writer has concluded that both probably belong to the one batholith. It is of some interest to note that the arithmetical mean of the two analyses is almost the exact equivalent of the analysis of the average grano- diorite in the Cordillera. The latter average appears in column 4 of the follow- ing Table XX XIII, and represents nine analyses from California types, one Oregon type, and two Washington types, all of these being taken from Bulletin No. 228 of the United States Geological Survey. REPORT OF THE CHIEF ASTRONOMER 539 SESSIONAL PAPER No. 25a Table XXXIII.—Analyses of granodiorites. 1 | 2 3 4 a |Phase 1 of Chilli- Phase 2 of Chilli- Average Average wack batholith. | wack batholith. | of Two Phases. of Twelve Types. SIO sae Saunenematene oeeeres 60°36 71°41 65°88 65°10 UbTG ete aan a aes rent ‘70 “34 52 “54 ANS Os. fools is haan este: 17°23 14°38 15°80 15°82 Her OO en See 1°93 1°33 1°63 1°64 He OS rete sees. ioe 3°74 Weis 2°46 2°66 Min OS ahve tnt ye omen 14 “O04 09 “05 |S OY Se eee 3°66 1°18 2°40 2°17 CAOR ae. Poe ite eae 6°07 2°51 4°29 4°66 Nas OMe iss ene ee Al 3°58 4°12 3°85 3°82 RO Oe ee ieee Serge nin om. | 1°74 2°97 2°35 2°29 TEL) wins NOs eae eon onpee “06 “09 “08 “16 HeOvabove 110°C: 22. =. s.) 55 “30 > £3 "93 Fe O oe te Sereieteicekes rae “aii “1183 "12 16 COR ere oe ane ee | 08 12 fel LO lees (re Ae eeene 5c 99895 160° 04 100°00 | 100° 90 The third major phase of the batholith is probably the most abundant of the three. Macroscopically it is almost indistinguishable from the phase first described. The microscope shows, however, that orthoclase is here an essential constituent. In the order of decreasing abundance the essentials are plagioclase, near Ab, An,; quartz; orthoclase; hornblende; biotite. The two femic minerals are in about equal amount. In optical properties all these minerals are identical with those of the first phase. The accessories are also the same as there but a few grains of allanite are associated. This phase occurs at many points in the batholith. The type specimen was collected at the mouth of Depot creek on the east side of Chilliwack lake, and thus in the heart of the batholith. The specific gravity of this specimen is 2-678. It is a normal granodiorite. It has not been analyzed, but its analysis would probably be close to the mean of the two analyses of the other phases (column 3). We may conclude that the average rock of thts batholith is a true grano- diorite tending towards the composition of a quartz diorite. The specific gravities of six fresh, type specimens from the batholith vary from 2-626 to 2-757, averaging 2-693. Nodular basic inclusions occur at various points in the mass. They are seldom very numerous and, so far as observed, never of large size; diameters exceeding 10 cm. are very uncommon. All of these dark-coloured nodules are probably indigenous bodies. They are of two kinds, both of which occur in the staple granodiorite phase. The one kind has some similarity to the Slesse diorite. It is a rather dark greenish-gray, fine-grained rock, composed of labradorite, green hornblende, and less important biotite as the chief essentials, with magnetite, apatite, titanite, and zircon, a little quartz, and a very little orthoclase as accessories. The struc- 540 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 ture is peculiar in being remarkably poikilitic. The larger individuals of each essential mineral contain smaller individuals of each of the other essentials as well as erystals of the accessories except quartz and orthoclase. Those two minerals are as usual the youngest of all. A few of the hornblende crystals contain small cores of colourless augite. The specific gravity of a typical nodule of this kind is 2-791, which is near the average for the Slesse diorite (2-806). The nodule evidently has the composition of a basic hornblende-biotite diorite. Its special structure could be explained on the hypothesis that it is simply an inclusion of the Slesse diorite which has been heated and largely recrystallized in the younger granodiorite magma. Since, however, these nodules occur in parts of the batholith far removed from the diorite contact, and since they show perfect interlocking with their host, it seems at least equally probable that they are true basic segregations. If this second hypothesis could be proved we should have one more illustration of the obvious consanguinity of the two batholiths. The other kind of inclusion is of a much darker green-gray colour and is also fine-grained. The essential components are a nearly colourless diopsidic augite (very abundant), pale green hornblende, and labradorite, Ab, An,. The accessories are magnetite, apatite, titanite, a very little biotite and quartz, and, possibly, a little orthoclase. The structure is the hypidiomorphic-granular, but at various places in the thin section suggests the diabasie structure. The specific gravity of a typical nodule of this class is 2-908. It has the composition of a gabbro or of a basic augite-hornblende diorite. Contact Metamorphism.—The thermal metamorphism of the Carboniferous sediments on the divide between Slesse and Middle creeks is intense, and is- essentially like that noted as due to the intrusion of the Slesse diorite. A new type of metamorphic product was found on the ridge north of Chilliwack river, about four miles from the lake. This is a hornfels richly charged with pheno- eryst-like prisms of andalusite, which are shot through a mat of green mica and quartz—a rock clearly derived from a silicious argillite. At the main contact of the granodiorite, on the ridge north of Depot creek, a small patch of intensely metamorphosed limestone is cut by basic diorite dikes, by the Custer gneiss-granite, as well as by the Chilliwack granodiorite. It is probable that all three kinds of intrusive rock, especially the more acid ones, have produced the observed recrystallization of the limestone. That rock has the appearance of a typical pre-Cambrian crystalline limestone of Ontario or Quebec. It is a white coarse-grained mass of calcite, bearing numerous scales of graphite, epidote, and zoisite in rounded grains, cubes of pyrite and anhedra of grossularite. INTRUSIVES CUTTING THE VOLCANICS. Besides the occasional andesitic and basaltic dikes which have evidently originated in the same magma as the surface lavas, the Skagit formation is cut by a small stock and by several wide dikes of quite different materials.* The * One highly vesicular, basaltic dike cutting the intercalated conglomerate may be of distinctly later date than the Skagit voleanic formation. REPORT OF THE CHIEF ASTRONOMER 541 SESSIONAL PAPER No. 25a stock and most of ihese dikes have the composition and structure of a quartz- bearing monzonite verging on granodiorite. One great dike has the properties of a typical hornblende-diorite porphyry. Monzonite Stock.—The stock is intrusive into the volcanics at their fault- contact with the Hozomeen quartzites a short distance north of Monument 69. This stock as exposed has an elliptical ground-plan, measuring about 1,200 yards in its greatest diameter and 800 yards along the minor axis. Like the agglomerate it is devoid of any crush-schistosity and the intrusion appears to have occurred later than the post-Laramie epoch of intense crushing. The intrusion may have been genetically connected with the faulting by which the voleanies were dropped down into their present lateral contact with the old quartzites. The material of the stock seems to be rather uniform, with the habit of a fresh, light gray, medium-grained syenite. The essential constituents are, in the order of decreasing importance: plagioclase, orthoclase, hornblende, quartz, biotite, augite. The plagioclase is often zoned, with Ab, An, in the cores and oligoclase in the outer shells; the average mixture is an andesine near Ab, An.,. The hornblende is green in about the same tones as those of the amphibole in the Chilliwack granodiorite. The characters of the other essential minerals and of the accessories are also the same as in that batholith. The structure is the eugranitie. ‘ The rock clearly belongs among the quartz-bearing monzonites and chemi- cally would show the composition also allied to that of a basic granodiorite. The Chilliwack batholith is only five miles distant and it is highly probable ihat this monzonite stock is its satellite. Dikes.—On the rugged, glacier-covered ridge south of Monument 68, the Skagit voleanics are cut by two or more great, north-and-south dikes of mon- zonite, similar to the staple material of the stock but relatively richer in plagioclase and quartz and poorer in biotite. These dikes, which range from 100 to 300 feet or more in width, are doubtless giant apophyses from the magma- chamber of which the Chilliwack batholith was a part. : A ‘half-mile west of these dikes and running nearly parallel to them is a third dike over 100 feet in width. It is composed of a dark gray to greenish- gray, medium to fine-grained, somewhat porphyritic rock of different habit from the monzonite. The phenocrysts are green hornblende which is often in parallel intergrowth with augite; and basie labradorite. The ground-mass is a hypi- diomorphic-granular aggregate of labradorite and interstitial quartz. Magnetite, apatite, and titanite are the accessories. Orthoclase seems to be absent. The rock is somewhat altered and is charged with a considerable amount of uralite evidently derived from augite. A small amount of chlorite may represent original biotite, but none of this mineral was discovered in the thin section. - The rock is to be classed as a hornblende-diorite porphyrite. The dike is uncrushed. It has the habit and nearly the composition of the finer-grained phases of the Slesse diorite. The similarity is so great that one 542 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 may believe that the porphyrite is an off-shoot of the same magma as the diorite. That relation would be parallel to the one just postulated for the neighbouring monzonite dikes and the Chilliwack granodiorite. In fact, it seems simplest to suppose, first, that all four rock-types belong to one eruptive period, the more basic intrusions antedating the acid intrusions by only a short interval of time; and, secondly, that all four rocks were differentiates from one great magma-chamber. DIKES CUTTING THE CuiILLiwack BaTHOLITH. Two different kinds of acid dikes cut the Chilliwack granodiorite. One of these kinds seems to be merely a later expression of the same magma from which most of the batholith itself was crystallized. Such dikes are not common and were never found far from the main batholithic contacts. This fact suggests that the batholithic magma first solidified along the contacts and that this early formed shell was injected by dikes from the still molten interior of the mass. Four of these dikes were observed on the ridge north of Depot creek. They are all composed of light gray granodiorite porphyry, somewhat more acid than the staple quartz-diorite of the contact-shell into which they have been intruded. Acid dikes of the second kind also specially affect the borders of the batho- lith but occur in the interior as well. They are not numerous and rarely attain widths greater than four feet. They are light pinkish-gray to whitish, fine-. grained granites of aplitic habit. The essential constituents are quartz, micro- perthite, orthoclase, andesine (Ab, An,), and biotite: titanite, magnetite and apatite are accessory. The structure is the hypidiomorphic-granular. The rock is an alkaline biotite granite, verging on biotite aplite. Its relation to the granodiorite recalls the similar succession of granites—acid, alkaline and micro- perthite-bearing biotite granite succeeding granodiorites—in the Okanagan and Selkirk ranges, as so often in other granitic provinees of the Cordillera. Two classes of basic. dikes cut the granodiorite. So far as known, the one class is represented only in one 10-foot, nearly vertical dike at about the 5,000- foot contour on the southern slope cf Pyramid mountain (the high conical peak ~ northwest of the outlet of Chilliwack lake). This rock is fine-grained, dark ‘ greenish-gray, and of lamprophyric habit. Under the microscope it is seen to - have the composition and structure of an acid camptonite. The other kind of basic dikes has been recognized at several poms but always in bodies of small size; no one of them is known to be wider than two feet. Four of these dikes were found at a point on the same southern slope of Pyramid mountain at about the 4,100-foot contour. A fifth was encountered in the gulch running eastward from the southern end of Chilliwack lake and at a point about 2,200 feet above the lake. Al! of the dikes are greatly altered and their diagnosis is not easy. One of the thin sections showed, however, some residual augite intersertally placed in a web of basic plagioclase, the only other primary essential. The quantities and relations of the minerals as well as their alteration phenomena show pretty clearly that these dikes are normal diabase. REPORT OF THE CHIEF ASTRONOMER 5438 SESSIONAL PAPER No. 25a The basic and acid dikes were nowhere found in contact. Judging from analogy the diabase dikes would be regarded as younger than the camptonite or than either of the acid kinds of dikes. : Acip DIKES GUTTING THE CHILLIWACK SERIES. Apart from the somewhat numerous dikes which are plainly apophysal from the Chilliwack granodiorite (granodiorite porphyry), there are relatively few acid dikes cutting the Paleozoic sediments of the region. The Glacial drift of the valley carries a considerable number of boulders of a porphyritie rock which, judging from the distribution of the erratics, should be in place at several points in the Chilliwack river basin between the lake and Tamihy creek. This rock was actually found in place as a dike or sheet at the 2,400-foot contour on the slope north of the confluence of Middle creek and the river. The exposure is poor and neither the exact relation nor the thickness of the body could be determined. This dike-rock is of a light-gray colour, weathering a pale brown, with conspicuous white phenocrysts of oligoclase standing out of the coloured matrix. The phenocrysts measure from 0-5 em. to 1-5 em. in length. Much smaller, likewise idiomorphic crystals of quartz and orthoclase can also be seen with the unaided eye. No ferromagnesian mineral could be found either in the hand-specimen or in the thin section. The ground-mass is a finely granular aggregate of quartz, orthoc'ase, and oligoclase. The rock is a granite porphyry of aplitic composition; it is, however, a rock of very different habit from the aplitic dikes cutting the Chilliwack granodiorite and there is no evidence that the granite porphyry has any direct genetic connection with that or any other of the visible batholiths of the region. Basic DIKES AND GREENSTONES IN THE CHILLIWACK SERIES. At a few points the great argillite-sandstone series of the Chilliwack river valley and the adjacent region is charged with small bodies of basic and ultra- basic igneous material, all of which is probably intrusive. One of the bodies has the form and relations of a much faulted dike about twenty feet in width; it cuts the sediments close beside the diorite contact on Pierce mountain. The dike has been squeezed and rolled out into a number of more or less perfectly. disconnected lenses. Its compact, dark greenish material proved, on microscopic examination, to be a mass of serpentine, original olivine, and magnetite. The rock was doubtless originally a dunite. At this locality considerable masses of tremolite occur in the sediments and may in part at least have been derived from the serpentine through the metamorphic action of the intrusive Slesse diorite. Close beside this dike of altered dunite, the crumpled argillites are inex- tricably mixed with similarly mashed, dike-like bodies of amphibolite, which ig transitional in a few places into a fairly coarse-grained gabbro. In the vabbro the bisilicate has all gone over to an amphibole of actinolitic habit. 544 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 A mass of composition similar to that of the just mentioned gabbro occurs as a sill or great dike, cutting the Paleozoic strata at the high cliff.facing the — mouth of Middle creek on the north side of the Chilliwack river and about 2,000 feet above the river. In general relations, chemical composition, and degree of metasomatic alteration all of these smaller bodies are much like the Vedder greenstone and they may be tentatively correlated with it both in age and origin. STRUCTURAL RELATIONS. In structure and composition the Skagit range is, in many essential respects, analogous to the Columbia mountain system. Here, however, the Paleozoic rocks are less intensely crumpled and metamorphosed. The Skagit range is structurally divisible into two parts. From the Skagit river to Middle creek it is chiefly composed of intrusive granites or allied rocks, which occur in such large numbers and so differing in age that we may fitly eall the whole plutonic group the Skagit composite batholith. The oldest member of the batholith is unconformably overlain by the Skagit volcanic group. A remnant of the Hozomeen formation appears as a second rock-body which is not part of the composite batholith but is a part of its country-rock terrane. i) West of the Slesse diorite the mountains are made of dominant sedimentary rocks. The Paleozoics (Chilliwack series) are very thick, the suggested mini- mum of about 6,800 feet of strata being, perhaps, much below the real thickness of the rocks actually exposed. An unknown additional thickness of conformable strata underlies those beds; thus no base is known to the Paleozoic (largely Upper Carboniferous) sediments of the west slope. The heavy mass of basic (andesitic) lavas and pyroclastics, named the Chilliwack Volcanic formation, is plainly contemporaneous with the fossiliferous uppermost beds of the series. The Triassic argillites and sandstones of the Cultus formation are not well exposed, but they seem also to be of imposing thickness: It is not known whether they are conformable with the Paleozoics, but an uneonformity is suspected. Very little of the terrane called the Tamihy series and tentatively equated with the Pasayten series, occurs within the Boundary belt, and it has not been’specially - studied. It is unconformable upon the Upper Carboniferous and probably upon the Cultus Triassic beds as well. The Eocene (?) Huntingdon formation forms only a small patch on Sumas mountain; it is unconformable upon the Paleozoic quartzite and also upon the Sumas granite, provisionally assigned to the Upper Jurassic. Throughout the whole width of the range simple folds are extremely rare. A much broken syncline, pitching gently eastward from the summit of McGuire mountain, is one of the very few decipherable structures in the mountains of the Boundary belt. The Chilliwack river, between Slesse and Tamihy creeks, seems to be flowing on the axis of a broken anticline, the east-west axis of which pitches eastward at a low angle. The southern limb of this arch is also the REPORT OF THE CHIEF ASTRONOMER 545 SESSIONAL PAPER No. 25a northern limb of the McGuire mountain syncline. The east-west direction of these axes may possibly be connected genetically with the east-west course of the wide Fraser valley to the north. Elsewhere the only observed structures in the stratified rocks are local crumples, faults, and small thrusts. Of these, normal faults seem to be most important in explaining the actual distribution of the rocks now exposed. As noted long ago by Bauerman, the section up the Chilliwack river seems to be that of a gigantic monocline, showing an almost. incredible thickness of Paleo- zoic rocks. This is probably a deceitful appearance. East of the nose of the supposed anticlinal near Slesse creek a heavy, crinoidal limestone with moderate northeasterly dip appears four times in the river section, besides appearing in the northern limb of the anticline. The writer is inclined to regard this lime- stone as representing the same horizon throughout; if so, it is best to suppose that the repetition of the limestone, with the associated shales and sandstones, is due to normal faulting. ‘The faults are probably strike-faults, running in a general northwesterly direction; the downthrow being on the southwest in each of the four displacements postulated. It should be added that the exposures are so poor that this partial explanation of the great thickness of beds outcropping along the Chilliwack river is in high degree still hypothetical. Somewhat more certain is the necessity of mapping the faults bounding the Triassic Cultus formation on east and west. The one on the west seems proved rather clearly; the other is not proved as to its actual location, but has been entered on the map to explain in this case the relation of fossiliferous Mesozoic and Paleozoic strata in lateral contact. The faults limiting the Skagit voleanics on north and west as well as at the Skagit river, have already been mentioned; little doubt is felt as to the exis- tence of all three of these master displacements. Nothing need be added to the descriptions of the structural relations of the granitic bodies, as already given in the respective sections of the present chapter. The cardinal fact of magmatic replacement of the huge Paleozoic geosynclinal prism as well as the pre-Cambrian basement terrane by the Chilliwack batholith, and also by the Custer batholith if it is of Jurassic date, seems to the writer quite obvious in the field. The relations are precisely the same as those stated for the vast Coast range batholith, described by Dawson, Lawson, and the geolo- gists working in Alaska, except that the Chilliwack batholith is probably younger than its great neighbour. Nearly all of these observers agree as to the fact of the replacement for the Coast range batholith. The significance of their agree- ment is great, for they have studied the world’s greatest post-Cambrian batho- lith invading one of the world’s greatest geosynclinals. CORRELATION, The geological dating of the various formations observed in the Boundary belt where it crosses the Skagit range, has already been discussed in connection with the description of the more important roeck-bodies. Many doubts remain as to the exact order in which they should be arranged in the geological time- 25a—vol. ii—35 546 DEPARTMENT OF THE INTERIOR 2 GEORGE V., A. 1912 scale. However, as implied so often in the preceding chapters, the writer believes that a tentative correlation made by the geologist who has actually observed the rocks in the field is better than no correlation at all and in most eases will give safer results than the correlations which would be made by systematists who have no personal knowledge of the ground. For the Skagit range we have the advantage of knowing that there are certain definitely fossili- ferous bands in the different stratified series; the chances for serious error are not nearly so great as in some of the eastern ranges. Among the more impor- tant unsolved problems are those relating to the age of the Custer granite-gneiss, of the older members of the Chilliwack series, of the Skagit volcanics and of the Tamihy series. If the Hozomeen series is really Carboniferous the Custer. batholith is almost certainly of Mesozoic, and presumably late vurassic, date. But it is conceivable that the country-rocks of this batholith are all of much older date and that this gneissic body may be a small fragment of the pre- Cambrian terrane whence the materials of the Rocky Mountain geosynclinal prism were derived. With the exception of this one body it seems likely that we have no other exposure of those ancient rocks west of the Priest River terrane in the eastern Selkirks. Tempting as it seems to regard the granite- gneiss as a part of the missing pre-Cambrian, the writer is inclined to dismiss that hypothesis and to adhere to the correlation given in the foregoing text, with which the table of preferred correlations should be read :— Correlation in the Skagit Range. IPlewstocene sen ncinecn acl Recent and Glacial (including the gravel plateau of the lower Fraser river). Miocene or Post-Miocenc 2. Diabase dikes cutting the Chilliwack batholith. Camptonite dikes cutting the Chilliwack batholith. Syenite-porphyry dikes cutting the Chilliwack series. ‘ Mapone? Syenite-porphyry (?) dikes (?) cutting the Huntingdon formation. SM eG yh aa aan memes Monzonite stock cutting the Skagit volcanics. | hittiwack granodiorite batholith. Slesse diorite stock (?) {Stasi volcanic formation. ve Oligocene (??).....-.0.00 Skagit harzburgite intrusion. Dunite dikes and gabbro dikes (in part) cutting Chilliwack series. JUNOT ON SSO AGOOR OTS O SaaS Huntingdon formation ; unconformable on Chilliwack series, quartzite and granite. Unconformity. Cretaceous (2). 0.00. re cews Tamihy series. Unconformity. Sumas diorite. {Suina granite. Custer granite-gneiss (possibly pre-Cambrian). EH UASSUCIA se lovate ciate sieleisisiel atone Cultus formation. Unconformity. ay f Vedder greenstone (altered gabbroid rock), Upper Carboniferous .... \ Chilliwack Volcanic formation. Upper Carboniferous (and { Chilliwack series. OLLI Tetonlciacieteras cele s | Hozomeen series. Vertes hhided eats LN AA v. 1 Geolo pay ae North American cor