A 7^ ENGR QE 195 .D27 ^ 1898 i^¥. ■ --C,_t Mi -t*.**,^:^. 3 1924 072 684^95^ ENGR.LIBR. AUG 121996 All books are subject to recall after two weeks. Engineering Library 1 DATE DUE 1 I 1 ; 1 GAYLOHD PRINTED IN U S A Cornell University Library The original of tiiis book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924072684495 GEOLOGICAL SUEYET OF OAlsTADA G. M. DAWSON, C.M.G., LL.D., F.R.S., Dieectoe EEPORT ON AN EXPLORATION IN THE YUKON DISTEIOT, JST.W.T. AXD ADJACENT NORTHERN PORTION OF BRITISH COLUMBIA 1887 BY QEORGE M. DAWSON With Extraeta relating to the Yukon District from Report on an Exploration in the Yukon and Mackenzie BasiTU, 1887-88 BY R. G. McCONNELL OTTAWA PRINTED BY S. B. DAWSON, PRINTER TO THE QUEEN'S MOST EXCELLENT MAJESTY 1898 Price Thirty cents. Geological iSihvey of Canada. Plate I. G. M. D.— Photo., May 29, 1887. STIKINE RIVER, LOOKING DOWN FROM TELEGRAPH CREEK. Showing the two principal terraces. G. M. D.-Photo., July 8, 1887. PRANCES LAKE, LOOKING WEST, POUR MILES PROM OUTLET. GEOLOGICAL ^UEVET OF CAXADA G. ^[. DAWSOX. C.M.G., LLD., F.R.S.. Director EEPORT ox A>r EXPLORATION IX THE TrKOX DISTRICT, X.AV.T. ADJACEXT XORTHERSf POETIOX OF BRITISH COLUMBIA BY GEORG^E XI. i:>AT\'SOX Witk ExTi-a::> rdattMg to t*« i'u.co . JXstrid from Report oh on SxplcmtioA in '.'m Tnbm attd Maekeasie Basins, 18S7-S8 R. G. McCOXXELL OTTAWA PRDTTED BY S. E. DAWSOX. PRDfTER TO THE QUEEX^ MOST EXCELLEXT MAJESTY , ^^ 1898 1 ' -*^ "- <\ - %^ f ^ y- Note. — The bearings throughout this report are given with refer- ence to the true meridian, unless otherwise specially noted. PREFACE TO REPRINT. As a result of the late remarkable discoveries of gold in the Klondike region, much general attention has been awakened in the entire Yukon District of Canada, and so gr^at a demand has arisen for the reports and maps of the Yukon Expedition of 1887-88, that the supply of these is now practically exhausted. In order to meet the inquiries for these reports and maps which still continue and appear likely to increase in number, it has been decided to reprint the writer's report of 1887, together with the parts of Mr. R. G. McConnell's report 'of 1888 that relate to the Yukon district. The present publication thus includes in a convenient form all the information directly obtained by the Geological Survey of Canada, comprising practically all the geological data available for the Yukon district; constituting a basis of fact upon which further investigai- tion may proceed, and one likely to be of use to the prospector and miner by placing him in possession of the main structural features of the country. The " Report on the Yukon District and adjacent Northern Portion of British Columbia," is reprinted without change, but the appendices have been omitted. The titles of these appendices are as follows : — T. Notes on the distribution of trees and of certain shrubs. II. Notes on the Indian tribes. III. List of plants collected. IV. Zoology. V. Notes on the lithological character of some of the rocks collected, VI. Meteorological observations. VII. Summary of astronomical observations. All that part of Mr. McConnell's report which relates to the Liard River below Fort Liard, near the eastern boundary of British Columbia, together with that dealing with the Mackenzie valley proper, are not reproduced ; but that relating to the Porcupine and Yukon Rivers is here reprinted. IV GEOLOGICAL SURVEY OP CANADA. Although no work later than that of 1887 and 1888 has been done by the Geological Survey in the Yukon District, much additional information respecting certain parts of the region has since been obtained, particularly as the result of surveys and examinations by Mr. W. Ogilvie, D.L.S. This has been published by the Department ol the Interior, and a further publication of the same kind is soon to appear. Some later geological notes on a part of the region are also given in a paper by Mr. 0. W. Hayes, entitled "An Expedition Through the Yukon District," published in the National Geographic Magazine, Washington, May 15, 1892. The map, in three sheets, which will accompany the present reprint, altho'T;^'_i in great part reproduced from the original engraving, has been brought up to date by the addition of the latest available surveys of the region which it includes, according to authorities noted on the map itself. It will thus present a more complete geographical view of the region than that which was possible at the time the reports here reprinted were written. G. M. D. January, 1898. INDEX. PAGE. Abies suhalpina 114 Aga-zi-za River, Frances River 103 Agriculture in Yukon district. . .... 20 on Stikine River 20, 58 Dease Lake 75 Uease River 86,90 Liard River I!t5 Agricultural resources, Fort Liard 207 Alaska coast, glaciation 37 Anvil Mountain, Dease River fi6 Argillites, Wrangel 30 Stikine River 53 Asbestos, Stewart River 24 note on occurrence in Yukon district 244 Atlin Lake 169 Basalt in Yukon district 35 on Stikine River 54 Tahltan River 62, 69 Upper Liard River 99 Pelly River. 120, 130 Porcupine River 223 Lewes River 144 Yukon River 229-230 Miles Canon, Lewes River 161 Beady Mountain, Dease Lake 75 Beaver River, Liard River 204 Bed-rock Bar, Liard River 191 Bell River, Porcupine River 211, 214 Bennett Lake, Lewes River 166, 171 Big Salmon River, Lewes River 15, 150, 151 Black River, Liard River, gold on 79, 206 Blake, P.W., quoted 50 Blue Fish River, Porcupine River 218 Blue River, Dease River 89 Boswell, T., Teslintoo River. , 154 Botany. See Plants, Trees. Boulder clay. See Glacial deposits. Boulders, Frances Lake 105 British Columbia, geological features of, continued in Yukon district 29, 31 Cambro-Silurian 31, 92 Campbell, Robt., explorations bj; 10, 79, 135 Campbell Mountains, Frances River 105 Canon, Great, Stikine River 44 Grand, Liard River 199-200 Little, Stikine River 47 Kloochman, Stikine River 47 Lower, Upper Liard River 96 Middle, Frances River 101 False do 103 Upper do 104 Hoole, Pelly River 118, 120 Granite do 129 Miles, Lewes River 160 Canvas boat, used on Liard River 199 Carboniferous rocks 31 McDame Creek 91 Frances River 102 Lewes River 170 PAGE. Cassiar Bar, Lewes River, gold 152, 181 Gold district 23, 76-84, 182 Range 11, 84 eastern slope of 88, 100 seen from Liard. 190 Cassiar trail 62-71 Chilkat Pass 18 Chilkoot Indians destroy Fort Selkirk. . . 138 Cliilkoot on Taiya Pass 174 Clearwater Rivt-r, Stikine River 4S Climate, Yukon district 18, 27 Stikine Ri ver 56 Dease Lake 73 Dease River 89 E. of Cassiar Range 88 Lewes River 173 soil frozen 89, 119, 123, 18? Coal on Lewes River (analysis) 147 in Yukon district 244 seam reported near Yukon River .... 232 Coal River, Liard Rivsr 194 Coast Ranges, B.C 8, 29, 46 eastern edge of 170 Conglomerate, Lewes River 145, 149 Copper, native, Thibert Creek 80 ore in Yukon district 244 Cordillera metalliferous belt 24 Cottonwood Creek, Dease River -SO, 90 Cranberry Rapids, Liard River 191-192 Cretaceous rocks on Bell River, Porcupine River 214 Lake Labarge, fossils 158 Lewes River, fossils 145 Lower Liard 203 Mackenzie River 239 Peel River Portage 213 Porcupine River 2i6-218, 231 Stikine River 55 Yukon River 231, 240 Crystalline schists on Yukon 238 D'Abbadie or Big Salmon River, Lewes River 150 Dall, W. H., referred to 12, 19 Dawson, Sir J. Wm., determination of fossils, by 124, 145, 148 Dease Creek, Dease Lake 76 gold mining on 74, 78 Dease Lake 73, 84 gold on 74 ancient outlet of 67 Dease River 81, 84-90 geology of 90-94 Defot Creek, Caflon Creek, Dease River . 78 Devil's Portage, Liard River 198 Devonian rocks on Porcupine Ri ver 226 Dip of rocks on Pelly- Yukon 239 Eagle River, Dease River gfi Economic minerals 23-25, 242-244 Fauna. See Zoology 22 VI GEOLOGICAL SURVEY OP CjlNADA. li'inlayson Lake I'inlaysou Kiver Fishes, Yulcon district Lewes River Flora. Sec Plants, Trees. Fort Halkett, Liard K'iver . . . Fort Liard. Fort Nelson, Liard Hi\ er .... Fort Reliance, copper ore at. Fort Silkirk, rocks at Fort Yukon . PAGE. 11 :i, 115 no, 114 '.'.'.'.'.'. 162 .... 195 207 205 244 238 226, 227 Forty-mile Greek 181-182, 232-234, 243, 244 Fossils, P<}St-Tertiary 39 Cretaceous, Lewes River 145 Lake Labarge 168 Triassic, Hell Gate, Liard Kiver. ... 202 Stikiue River 54 Carboniferous 31 Ordovician, Dease River 93 plants, Pelly River 124 Le^ves River 148 woods, Lewes River 145 Frances, Fort, Frances Lake 136 Frances Lake 300-112 gold on Ill Frances River 9S-10B Lignite on 101 French Creek, Dease River 89 Furs, Yukon district 25 Galena, argentiferous, McDame Creek. . . 80 on Forty-Mile Creek 233 Garnets, Rothsay Point 52 Geographical results of Yukon Expedition 6 Geology, i^eneral, of, Yukon district. 28-40, 238-244 Glacial'drift 10,38 Cassiar Trail 71 Dease River 94 Finlayson River 117 Pelly River 119, 124, 129, 131 Lewes River 148, 153 action in Yukon basin 241-242 Strife, Pelly River 129 Lewes River 153, 171 Glaciation, Lake Labarge 158 Lake Marsh 163 Glaciers of Stikine valley 49 recession of 50 Glenlyon Mountains 125, 127 Glenlyon River, Pelly River 125 Glenora, Stikine River 49, 58 Gold districts, Yukon and Cassiar 23 in Yukon district, references to 242-244 granite, connection of 32 and glacial drift, connection of 40 on Dease River 74 on laard at McCullough's Bar . . . 195, 242 at mouth of Nelson River 205 on Upper Liard River 99 Forty -mile Creek 232-233 Stewart River 236 Stikine River 60, 77 Mining, Cassiar district 76 on Upper Liard Kiver 99 Frances Lake Ill Pelly River 133 Big Salmon River 151 Lewes River 152, 178 Grand Cafion, Liard River 199-200 PAGE. Granite, Coast Ranges 29 Stikine Kiver 52 Cassiar Ranges.' 90 Upper Frances Kiver 104, 111 Glenlyon Range 127 Granite Canon, Upper Pelly River. . 129 Lewes River 144, 171 ChilkootPass 17li Granitic intrusions, Yukon district 235 Granitic axes, Yukon district 9, 32 Graptolites, Dease River 92 Great Glacier, Stikine River 50 Highland River 18S Hell Gate, Liard Kiver 201 Hoffmann, G.C., analysis of coal by 148 Holt, George, discovers gold on Lewes K. 179 Hoo-chee-koo Bluff, Lewes Kiver 144 Hoole Kiver, Pelly River 118 Hotalinqu River, see Teslintoo 13,109 Howling-dog Rock, Porcupine River .... 225 Hudson's Bay Co., enterprise of 59, 81, 139 Il-es-too-a River, Frances Lake 107 Indians on Liard River 187 Porcupine River 224 Indian farms on Liard River 206 tribes of Yukon district 178 Iskoot Pjver, Stikine River 48 Island Lake, Big Salmon River 151 Jade on Lewes River 36, 146 Jubilee Mountain, Lake Marsh 163 Ketsa River, Pelly Kiver 121 Krause, Dr. Arthur, exploration of 180 Kwikhpak or Yukon River 11 Labarge Lake, Lewes River 155 Lake basins, in valley of Yukon 241 Laketon, Cassiar 83 Lansdowne Mountain, Lake Marsh 162 Lapie River, Pelly River 121 Lapierre House 213 Lapworth, Prof. Ohas. on Graptolites, Dease River 92 Laramie rocks, Yukon district 34, 124, 148 Lewes River 13, 15, 140-155, 160 section of volcanic ash and drift on. . 33 jade on 36, 146 coal (analysis) 147 gold on 152, 178 general notes on 173 Liard River, Upper 94-1 l)H gold and lignite J9 dangers of, as route of travel 140 Lower 186-207 notes on gold on 242 Lignite on Lower Li.ird River 194 on Upper Liard River 99 Frances Pdver 101 Limesto' e, Cambro-Silurian, Yukon dis- trict 30 Carboniferous, Yukon district 31 Stikine River 63, 08 McDame Creek 91 Frances River 101 Lake Labarge 156 Lewes River 170 Lindeman Lake, Lewes River 172 Little Cafion, Liard Kiver ]88 Little Salmon Kiver, Lewes River 146 Lome Mountain, Lake Marsh 162 INDEX. Vll PAGE. Lower Ramparts, Porcupine River 226 Lower Post, Dease River 90 Mackenzie River and Yukon River com- pared 13 McMiUan River. 14, 126 Macpherson Lake, near Frances Lake . . . 109 Maps accompanying; 6 previous, of Stikine River 45, 61 of Yukon district 11, 178 Marsh Lake, Lewes River 162 MoConnell, R. G., work by 2 quoted 45 description of Lower Liard by 186-207 Rat River 207-209 Peel River Portage 209-215 Porcupine River 215-226 Yukon River 227-238 note on geology of Yukon district by. 238-244 McCuUough s Ijar, Liard River 82 gold foun 195 McDame Creek, Dease River, gold min- ing 79,87 silver in galena from 80 MoEvoy Lake, Finlayson River 114 McLeod, J., explorations by 81, 135 McLeod Mountain, Dease River 75 Minerals of Yukon district, notes on 23-25, 242-244 Miner's Range, Lake Labarge 156 Mining and minerals, Yukon district .... 23, 182 Mioliie Mountain, Lake Marsh. 162 Moose,, abundance of on Liard River .... 198 Mount Prudence, Liard River 203 Mountain Portage Rapids, Liard River. . 193 Mountain sheep, variety on Peel River Portage 210 Mud or Marsh Lake, Lewes River 163 Mud River, Liard River 192 Nares Lake, Tagish Lake 164, 169 Navigable waters, Yukon district 16 Stikine River 44 Pelly River 132 Lewes River 164, 173 Nelson River, Liard River. 205 Newberry or Teslintoo River, Lewes River 153 Ni-sutlin-hi-ni River 154 Nordenskiold River, Lewes River 146 Nourse River, Taiya Inlet 175, 177 Ogilvie, W., work by 2 notes on rocks of upper part of Por- cupine River 231 Ogilvie Valley, Lake Labarge 156 Old Grow Mountains.... 219, 220 Old Crow River, Porcupine River 218 Orchay River, Pelly River 121 Ordovician fossils, Dease River 93 Orography of Yukon district 8 Pas-ka River. 105 Peel River Portage 209-213 terraces in 240 Pelly Banks, PeUy River 13,136 Pelly Range 118, 122 Pelly River, upper course 13, 117-132 general notes on 132 gold on 133 confluence of, with Lewes River 134 exploration of 135 and Yukon, identity of 137 Plants, Stikine River 67 Tahltan River 63 Plants — Contirmed. PAGE. E. of Oassiar Range 88 Upper Liard River 98 Frances River 101 Finlayson Lake 116 Pelly River 131 Lewes River... 162,168 Platinum, Yukon district , 25 Porcupine Bar, Liard River 190 Porcupine River 215, 226 tributaries 226 terraces 241 Portage BrQle, Liard River 194 Quartz veins, Yukon district 33, 244 Frances River Ill, 117 Pelly River 133 Quartzites, Yukon district 31 Stikine River 53 Lewes River 171 Quiet Lake, Big Salmon River 151 Rabbit River, Liard River 193 Rampart House 224 Ramparts, Porcupine River 222 Rapid River, Dease River 88, 222 Rapids of the Drovnied, Liard River .... 200 Rat River 207, 209 Rink Rapid, Lewes River 143 River Channel, ancient, Dease Lake, 67 gravels and salt, Dease Lake , . . , 74 Frances River 116 Pelly River 129 Lewes River 146, 150 River valley, formerly an inlet, Stikine River 47 Rivers of Yukon District 7, 10 compared with other rivers 15 Riviere des Vents, Liard River. 196 Rocky Mountains, N. W, extremity 8 on Liard River 196 on Lower Liard 206 Peel River Portage 212 Rat River 208 Ross River, Pelly River 14, 120 Rothsay Point, Stikine River 52 Routes of travel, Yukon District 16, 183 CassiarTrail 62,72 Liard River 140, 186, 207 Lewes River 173 ChilkootPass 174 White Pass 177 Peace River to Liard via Nelson River 205 see also Navigable waters. Sandstones, Lewes River 144 Sayyea Creek, Liard River, gold mining., 79, 98 Scenery, Frances River 107 Lewes River 165 Schistose rocks of Yukon District 235 Schists, crystalline, on the Yukon. , . . 238 Yukon District 30 Dease River 92 Lower Canon, Liard River 97 Pelly River 123, 129, 131 Lewes River 170 Schwatka, Lieut., surveys by 12, 142 Scud River, Stikine River 48 Sections, volcanic ash and drift, Lewea River 43 basalt overlying gravels, Stikine River 70 VIU GEOLOGICAL SURVEY OF CANADA. I'AGE, Selkirk, Fort, Felly River 134, 137 distances from , 13 Seuiinow Mountains, Lewes River 149, lod Serpentine 76,116,123,244 Shales, Dease Ri\ e- =1 Felly River 1^4 Sheep mountain, Dease River 88 Sheep Mountains 231 Silver in galena, MoDame Creek 80 on Forty-mile Creek 243 Simpson Lake, Frances River 102 Simpson Mountain, Frances River 101 Sixty -mile Creek 235 Skree Range, Dease River 86, 90 Smith River, Liard River 195 Stewart River 9, 137, 236 asbestus 24 gold mining 182, 243 Stikine River 15, 44-62 gold on, discovery of 60, 77 sections of basalts and gravels 24 " Stone House," Chilkoot Pass 175, 177 Sucker River, Porcupine River 225 Sullivan Mountain, Dease Lake 75 Superficial deposits, Yukon basm. .. 35-43 240-242 Sylvester, Rufus, trading post on Liard.. 192 Sylvester's Landing, Dease River 87 Sylvester Peak, Dease River 87 Tagish Lake 164, 169 Tank-heena River, Lewes River 159 Tahltan River, Stikine River 62 gold on 78 Taiya on Chilkoot Pass 174 Taiya River, Taiya Inlet 176 Tako Arm of Tagish Lake 168 Taku River, Taku Inlet 18 Tanzilla River, Stikine River. 66 Tatonduc River, notes by Ogilvie on rocks of 231 Tatshun River, Lewes River 143 Telegaph Creek, Stikme River , .49, 55, 58 trail from, to Dease Lake 62 Telegraph exploration, Stikine River. ... 60 Tent Peak, Too-tsho Range 104 Terraces, Stikine River 56, 71 Tanzilla River 66 Dease River, 86. 89, 94 Frances River 107 Finlayson River -. 116 PeUy River 130 Lewes River 143 Peel River Portage 240 Yukon valley 241 Tertiary rocks, Yukon district 35 Dease River 93 Frances River 101 Porcupine River 220, 221, 225 Mackenzie River, alluded to 240 Liard River 189 plateau, Upper Liard River. 100 TertiaiT lavas, Yukon 238 Teslin Lake 154 PAGE. Tes-lin-too River, Lewes River 13, 152 Thibert, Henry, discovers gold, 1872 .... 82, 109 Thibert Creek, Cassiar, gold mining 78 native copper, &c 80 Thomas River, Frances Lake 109 Til-e-i-tsho Lake, Thomas River 109 Timber (see Trees) 20 Timber line 114 Toad River post, Liard River 203 I'oo-tshi-too-a River, Frances River 103 Too-tshoo Range, Frances River 104, 111 Tooya River, Stikine River 64 Trees, Tahltan River 62 east of Cassiar Range 88 Lower Dease River. 89 Upper Liard River 97 Frances River 101 Frances Lake 110 Finlayson River 114 Pelly River 119,123,133 Chilkoot Pass 177 Triassic fossils. Hell Gate, Liard River . 202 Triassic, Stikine River 31, 54 Trout River, Liard River 197 Taes-i-uh Range, Liard River 100 Tumagain or Black River, Liard River, gold 79 Tus-tles-to Lake, Thomas River 109 Views, TanziUa River 65 Too-tsho Range 108 Bennett Lake 167 Volcanic ash deposits 41 rocks, altered, Stikine River 53 Lewes River , 171 Vologda, province of Russia, compared with Yukon district 28 Watershed between Arctic and Pacific Oceans 7 Whirlpool Canon, Liard River 193 Whirlpools, Liard River 193 White Horse Rapids, Lewes River 160- White Mountain, Lake Marsh 163 WhitePass ... 177 White River 9, 237 Whiteaves, J. F., determination of fos- sils, by 145, 158, 202 Whymper, F., referred to 11, 179 Winds, summer and winter 19 Wrangel Island 20, 62 Yukon District, area of 1 former explorations 10 routes to 17 climate ' 19 gold mining 23 economic importance 27 fort 137, 139 river, confused nomenclature of trib- utaries of 14 and Pelly River, identity of 137 description by R. G. McConnell 227, 238 note on gold of 243 Yua-sez-uh River, Frances Lake 109 REPORT ON AN EXPLORATION IN THE YUKON DISTRICT. N.W.T. AND ADJACENT NORTHERN PORTION OF BRITISH COLUnVIBIA. BY GEORGE M. DAWSON. Introductory. The Yukon expedition, to which the present report relates, was Yukon undertaken for the purpose of gaining information on the vast and Expedition, hitherto almost unknown tract of country which forms the extreme north-westerly portion of the North-west Territory. This tract is bounded in the south by the northern line of the Province of British Columbia (Lat. 60°), to the west by the eastern line of the United States territory of Alaska, to the east by the Rooky Mountain Ranges and 136th meridian, and to the north by the Arctic Ocean. The region Yukon dis- thus generally defined is referred to as the Yukon di.strict, this name *"°'- being rendered appropriate from the fact that the greater part of its area lies within the drainage-basin of the river of that name. The Yukon district, as above defined, has a total area of approx. Areaof Yukou, imately 192,000 square miles, of which, according to the most recent district. information, 150,768 square miles is included in the watershed of the Yukon. The superficial extent of the district may perhaps best be realized when it is stated that it is nearly equal to that of Prance greater than the United Kingdom by 71,100 square miles, ten times the area of the province of Nova Scotia, or nearly three times that of the New England States. It is unnecessary to add that the present report must be considered merely as a first contribution to our know- ledge of this wide country. The immediate necessity for the exploratory and surveying work Purpose of the Tin:dertaken by the expedition, arose from the fact that somewhat im- ^^P'' enable him to make a micrometer survey of the Stikine from the end of the line measured by Mr. J. Hunter in 1877, to Telegraph Creek, while I proceeded up the river by the first steamer of the season to Telegraph Creek, the head of navigation. From this place, goods are carried by pack animals to Dease Lake, the centre of the Cassiar min- ing district ; and here again a delay of several days occurred, as the animals had not been brought in from their range or shod for the Journey to season's work at the date of our arrival. Finally, on June 5th, we Dease Lake. j.gj^g}jg(j tjjg j^p^d of Dease Lake, and found the greater part of the lake still covered with ice. It was not till the 9th that we were able to reach the point on the shore near Laketon at which two men, Buildingboats previously sent on in advance with an Indian packer, were savi'ing lumber for boats. Seven days were here busily employed in this work and in constructing three boats for the purposes of the expedition. On the evening of the 16th, a strong wind having broken up the remain- ing barrier of ice, we reached Laketon with our boats, Mr. McConnell, with a crew of five Coast Indians intended for my work on the Upper- Departure Liard, having meanwhile joined us. On the 18th, having completed Lake. ^^"^ supplies and outfit at Laketon, we left that place, and on the 23rd reached the " Lower Post " at the confluence of the Dease and Liard rivers. Here Mr. McConnell, with one boat and two men, separated from us for the purpose of surveying and geologically examining the Lower Liard. Personnel of On leaving the confluence of the Dease and Liard, my own party par IBS. included, besides myself, Mr. McEvoy, Messrs. L. Lewis and D. Johnson, engaged at Victoria, two Tshimsian and three Stikine (Thlinkit) Indians, all good boatmen. Two local Indians hired as guides and to help in portaging, deserted a day or two after engaging, and from the " Lower Post " to near the confluence of the Pelly and Lewes, for an interval of more than six weeks, we met neither whites nor Indians. Journey to The ascent of the Liard and Frances rivers to Frances Lake proved ■ unexpectedly difficult and tedious, the rivers being swift throughout and three bad canons having to be passed through. Frances Lake was reached on the 8th of July, and after spending a few days in examining and mapping the lake, making the observations necessary to fix its position, and in the endeavour to find some Indian trail by which we might travel across to the Pelly, we began the wprk of portaging on the 17th. Journey over- As we had been unable to discover any route now in use by the R?ver.° ^ ^ Indians, and no trace whatever remained of the trail employed by the YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 5 Hudson's Bay Company in former years ; and further, as no local Indians could be found to act as guides or to assist in carrying our stuff, it was evident that the crossing of this portage (which had been estimated by Mr. Campbell at about 70 miles in length) would be a difficult matter, and that we might indeed find it impossible to carry over a sufficient supply of provisions for work on the Pelly. We therefore constructed a strong log cdche on the shore of Frances Lake, and left there, to be taken to Dease Lake by the Indians when they returned, everything we could possibly dispense with. Had we been unable to effect the portage, there was in our c&che a sufficient supply of provisions to en- able the whole party to return to the "Lower Post." After a very toilsome journey, we were," however, so fortunate as to reach the bank of the Upper Pelly on the 29th of July, with still nearly a month's provisions for four persons, our instruments and a small camping out- fit, a canvas cover from which a canoe might be constructed, and the tools and nails for building a wooden boat, should that prove to be necessary. Our Indians, who had for a long time been very uneasy because of their distance from the coast and the unknown character of the country into which they had been taken, were here paid off, and to their great delight allowed to turn back. As a dangerous rapid was reported to exist on the upper part of Descent of the the Pelly, it was decided to construct a canvas canoe in preference to ^PP®"^ PeUy- building a boat, which it might prove impossible to portage past the rapid. Having completed the canoe, we descended the Pelly, making a portage of half a mile past Hoole Rapid or Canon, and arrived at the confluence of the Lewes branch with the Upper Pelly on the 11th of August. We had now reached the line of route which is used by Prooeedingsat the miners, and expected to find, at the mouth of the Lewes, a pre- ™°"*h of arranged memorandum from Mr. Ogilvie, from whom we had separated in May. As we could not find any such notice, and as Mr. Ogilvie had not been seen on the lower river by a party of miners whom we met here on their way up the Lewes, we were forced to conclude that he had not yet reached this point. The same party informed us that there had been few miners during the summer on the Stewart River, where most of the work had been carried on in 1886, but that in con- sequence of the discovery of " coarse " gold on Forty-mile Creek, over 200 miles further down the river, most of the men had gone there. We were also told that Harper's trading post, where I had hoped to be able to get an additional supply of provisions should we fail to connect with Mr. Ogilvie, had been moved from the mouth of the Stewart to Forty-mile Creek. From the place where we now were we still had a journey of nearly 400 miles to the coast, with the swift 6 GEOLOGICAL SURVEY OF CANADA. waters of the Lewes to contend against for the greater part of the distance. If therefore it should have become necessary to go down stream 200 miles to Forty-mile Creek for provisions, so much would have been added to our up-stream journey that it would become doubt- ful whether we should be able to afford time for geological work on the Lewes, and yet reach the coast before the smaller lakes near the Building our mountains were frozen over. I therefore decided to set about the building of another boat, suitable for the ascent of the Lewes, and on the second day after we had begun work, Mr. Ogilvie very oppor- tunely appeared. After having completed our boat and obtained Mr. Ogilvie's preliminary report and map-sheets, together with the neces- sary provisions, we began the ascent of the Lewes, from the head- Beach Lynn waters of which we crossed the mountains by the Chilkoot Pass and Canal. reached the coast at the head of Lynn Canal on the 20th September, In addition to the physical obstacles to be encountered on the long Eeporte'd In- route above outlined, some anxiety was caused by reported Indian troubles on the Yukon. We heard a most circumstantial account of these from a couple of miners who followed us in to Dease Lake, the report being that the hostile Indians had retreated up the Pelly. As it was impossible either to confirm or refute these reports without practically abandoning the scheme of work, it was determined to pro- ceed according to the original arrangement. On reaching the mouth of the Lewes we ascertained that the story was entirely false, but it had none the less kept us in a state of watchfulness during a great part of the summer. Main Geographical Results. Geographical The main geographical results of the Yukon expedition, in so far as ata obtained, ^jj^gg ^^^ covered by the present report, are best shown by the accom- panying maps, which may be compared with previous maps of the same region. Mr. Ogilvie's instrumentally measured line from the head of Lynn Conal to the intersection of the Yukon or Pelly by the 141st meridian, will form a sufficiently accurate base for further sur- veys. In addition to this we now have an instrumental survey of the Stikine from its mouth to the head of navigation (Telegraph Creek), which is connected with Dease Lake by a carefully paced traverse. This is continued by a detailed running- or track-survey following the lines of the Dease, Upper Liard and Pelly rivers, and connecting with Mr. Ogilvie's line at the mouth of the Lewes, the total distance from the mouth of the Stikine to this point, by the route travelled Length of being about 944 miles. Adding to this the distance from the mouth of route tra- j.i t i i , versed. ^^^ liewes back to the coast at the head of Lynn Canal (377 miles). YUKOK DISTRICT AND NORTHERN BRITISH COLUMBIA. 7 the entire distance travelled by us during the exploration amounts to 1322 miles. This, taken in connection with the coast-line between the Stikine and Lynn Canal, circumscribes an area of about 63,200 square miles, the interior of which is still, but for the accounts of a few pros- pectors and reports of Indians, a terra incognita. The same descrip- tion with little qualification, applies to the whole surrounding region out- side the surveyed circuit, but much general information concerning the country has been obtained, which will facilitate further explorations. Along the routes thus travelled numerous points have been carefully Positions fixed in latitude by sextant observations, and a sufficient number of chronometer longitudes have been obtained by which to lay the whole down within small limits of error. Special attention was paid to the sketching and fixing of mountain topography in sight from the line of travel, and the approximate altitudes of a number of the more prominent peaks was ascertained. 'So reference is made here to the further work carried out by Messrs. Ogilvie and McConnell in 1888, which wUl, as above stated, be separately reported on. Orography and General Features. The region traversed by the routes just mentioned, including the Drainage extreme northern part of British Columbia and the southern part of ^^^ ^'"' the Yukon district (as previously defined), is drained by three great river systems, its waters reaching the Pacific by the Stikine, the Mac- kenzie, (and eventually the Arctic Ocean), by the Liard, and Behring Sea, by the Yukon. The south-eastern part of the region is divided between the two first-named rivers, whose tributary streams interlock, the Stikine making its way completely through the Coast Ranges in a south-westerly direction, while the Liard, on a north-easterly bear- ing, cuts across the Rocky Mountains to the Mackenzie valley. The watershed separating these rivers near Dease Lake has a height of 2730 feet, and both streams may be generally characterized as very rapid. To the north-westward, branches of the Stikine and Liard again Watershed, interlock with the head waters of several tributaries of the Yukon, which here unwater the entire great area inclosed on one side by the Coast Ranges, on the other by the Rocky Mountains. As the general direction of this line of watershed is transverse to that of the main orographic ridges of the country, it will probably be found, when traced in detail, to be very sinuous. The actual watershed between the Liard and Pelly, on our line of route, was found to have an elevation GEOLOGICAL SURVEY OP CANADA. Courses of rivers. Relief of the region. Trend of ranges. The Coast Ranges. of 3150 feet, but it is, no doubt, much lower in the central portion of the region between the Rocky Mountains and Coast Ranges. To the north of the Stikine, at least one other river, the Taku, cuts like it completely across the Coast Ranges, but its basin is compara- tively restricted and httle is yet known of it. It will be noticed, that while the several branches of the Yukon con- form in a general way to the main orographic axes, the Stikine and Liard appear to be to a large degree independent of these, and to flow counter to the direction of three mountain ranges. The region as a, whole, being a portion of the Cordillera belt of the west coast, is naturally mountainous in general character, but it com- prises as well important areas of merely hilly or gently rolling coun- try, besides many wide, flat-bottomed river-valleys. It is, moreover, more mountainous and higher in its south-eastern part— that drained by the Stikine and Liard,— and subsides gradually, and apparently uniformly, to the north-westward ; the mountains at the same time becoming more isolated and being separated by broader tracts of low land. The general base-level, or height of the main valleys, within the Coast Ranges, thus declines from about 2500 feet, to nearly 1500 feet at the confluence of the Lewes and Pelly rivers, and the average base- level of the entire region may be stated as being a little over 2000 feet. Disregarding minor irregularities, it is found that the trend of the main mountain ridges and ranges shows throughout the entire region here described a general parallelism to the outline of the coast. In the south-eastern and more rugged tract, the bearing of such ranges as are well defined is north-west by south-east, while beyond a line which may be drawn between the head of Lynn Canal and Frances Lake the trend gradually changes to west-north-west. The Coast Ranges, with an aggregate average width of about eighty miles, the whole of which is closely set with high, rounded or rugged mountains, constitutes the most important orographic uplift in the entire region, and here reproduces geographically and geologically the features characteristic of it in the more southern portion of British Columbia. Beyond the vicinity of Lynn Canal, this mountain axis runs behind the St. Elias Alps, ceasing to be the continental border, and may be said to be entirely unknown, as any indications of moun- tains which have appeared on this part of the map are purely conjec- tural. Notwithstanding the great width of the Coast Ranges, it is not known that any of their constituent mountains attain very notable altitudes, but it is probable that a great number of the peaks exceed a height of 8000 feet. These ranges are composed of very numerous mountain ridges, which are not always uniform in direction, and, so YUKON DISTRICT AND NORTHERN BEITISH COLUMBIA. 9 far as has been observed, there is no single culminating or dominant range which can be traced for any considerable distance. The mountain axis next in importance to that of the Coast Ranges, Epoky , TT T ■, I Mountain IS that which forms the water-parting between the Upper Liard and Ranges. Yukon, on one side, and the feeders of the main Mackenzie River on the other. This represents the north-western continuation of the Rocky Mountains proper. Its eastern ridges were touched on during the present exploration in the vicinity of Frances Lake and the head waters of the Pelly River, and are there designated on the map as the Too-tsho Range. This forms, so far as has been ascertained, the culmin- ating range of a number of more or less exactly parallel ridges, and summits in it attain heights of from 7000 to 9000 feet. It has, how- ever, been traced to a comparatively limited distance only, and it appears probable that a very complicated mountain system remains to be worked out in this portion of the region. A third notable mountain axis, which I have designated on the map Cassiar as the Cassiar Range, is cut through by the Dease River in its upper course, and further to the north-westward appears to form the line of water-parting between the tributaries of the Upper Liard and those of the branches of the Yukon. Peaks near the Dease, in this range, exceed 7000 feet, but it is probable that none much exceed 8000 feet, and that the range in a general way becomes lower to the north-westward. In the north-western and less elevated moiety of the region, the mountain ranges and ridges are in general lower and become discon- tinuous and irregular, or while retaining a general parallelism, assume an overlapping or echelon like ari'angement. In each of these mountain chains above described granitic rocks Granitic appear in greater or less force, as more fully noted on succeeding pages. '■°°'^^- In the intervening and subordinate mountain systems of the south-east, granitic axes are not found and do not exist as pr'Ominent features. Scarcely anything is known of the character of the country drained Orography o£ by the Macmillan, Stewart and White rivers, but it is probable that reg?o''ns°"""^ the basins of the two first-named streams closely resemble that of the Upper Pelly, which is described in following pages. Miners who have ascended the Stewart for a hundred miles or more, report the existence of a continuous range of mountains of considerable height, which runs parallel to the river on the north, from a point about fifty miles from its mouth onward. The absence of tributaries of any size along the south-west side of the Lewes below the Tahk-heena, with the general appearance of the country in that direction, so far it has been over- lookedj seems to show that the basin of the upper portion of the White River must be comparatively low, and situated as it is within the St. 10 GEOLOGICAL SURVEY OP CANADA. Elias Alps, this country must possess most remarkable features, both geographically and from a climatic point of view, and well deserves exploration. It would further appear to be nearly certain that the sources of the Tanana River are to be looked for in this district, well to the east of the 141st meridian. Effects of the The topographical features of the entire region here described have ■ been considerably modified by the events of the glacial period, and the changes produced at that time have more particularly afifected the drainage-basins and the courses of the various streams. The valleys and lower tracts of country are now more or less completely filled or covered by extensive deposits of boulder-clay, gravel, sand and silt laid down during that period. To these deposits are due the flat floors of the larger valleys, and also to a great extent the appearance which the more irregular mountain regions present of being partly submerged in level or rolling plains. Many changes in direction of flow in river- valleys, have doubtless also been produced during this period, though most of these yet remain to be worked out. The general result has been to produce systems of " inconsequent " drainage wherever the natural slopes of the country are easy and the limiting ranges irregular. Most of the rivers at the present day have done little more than cut out new channels in the glacial debris, touching only here and there upon the subjacent rocky floor. Sources of the Yukon and Nomenclature. nomlldlture ^"°^ particulars as have been ascertained relative to the various rivers examined in the course of the exploration, are given in a sub- sequent part of this report. As, however, some confusion has arisen in respect to the nomenclature of the Yukon and its tributaries, and erroneous statements have been made as to the "source" of the river it may be appropriate here briefly to note the facts in the case in so far as I have been able to ascertain them. Further details of the ex- ploration of the river are given on a later page. tionoffukon" '^^^ estuary of the Yukon appears to have been first explored by " °°' the Russian, Glasunoff, in 1835 to 1838, and the river was then named by the Russians the Kwikhpak, which name, according to Mr. W. H. Dall, is in reality that of one of the channels by which it issues to the sea. The lower part of the river, however, continued to be known under this name for a number of years, and it is so called on the (Russian) map of Lieut. Zagoskin, made from reconnaissance sur- veys which, in 1842-43, he carried up as far as Nowikakat. The mouth of the river is shown on Arrowsmith's map of 1850, but is there nameless. YUKON DISTRICT AND NORTHEEN BRITISH COLUMBIA. 11 The name Yukon was first applied in 1846 by Mr. J. Bell, of the Origin of the Hudson's Bay Company, who reached the main river by descending the Porcupine, and called it by what he understood to be its Indian appel- lation. The head waters of one of the main tributaries of the Yukon had previously been attained by Mr. R. Campbell (also an officer of the Hudson's Bay Company) in ] 840, and in 1850 he descended the river as far as the mouth of the Porcupine, naming the whole river thus traversed the Pelly, and naming also the Lewes, White and Stewart rivers, as well as numerous smaller tributaries. Campbell's nomencla- Connection of ture and his sketch of the river appear on Arrowsmith's map published k; khmk in April, 1854, and Campbell practically established by his journey the identity of his Pelly River with the Kwikhpak of the Russians. The connection between the two is given by a sketch (shown in broken lines) on the map just cited, on which also the mouth of the Tanana River (under the name Mountain-men River), is shown, and other de- tails represented with reasonable accuracy. The sketch of the river below the mouth of the Porcupine appears to have been due to the Hudson's Bay Company's traders, who, before Campbell had communi- cated his geographical information in London in (1853), had already met the Russian traders at the mouth of the Tanana. Much later, in 1863, I. S. Lukeen of the Russian Trading Company, ascended the river to the Hudson's Bay post. Fort Yukon, at the mouth of the Porcupine.* The name Yukon does not appear at all on Arrowsmith's map of tJsage on 1854, that of the Pelly standing for the whole length of the river subsequent ' J p n maps. explored by Campbell, but since that date the term Yukon has gradually become applied to the main river. The next map in order of publica- tion in which original data are employed is, I believe, that accompany- ing Mr. Whymper's paper of 1868, in the Journal of the Royal Geographical Society, t which is also reproduced in his book, Travels in Alaska and on the Yukon (1869). His general map gives the name Pelly for the whole river above the mouth of the Porcupine, " Kwich- Pak or Yukon," for the lower part. In his large-scale map, on the * By a singular oversight, Mr. W. H. Dall states in the first part of his work, Alaska and its Resources 1870 (p. 4) that " the identity of the Yukon [Pelly] River with the so-called Kwikhpak of the Russians " remained to be established when the explorations of the Telegraph Survey commenced on the river in ISOS, while giving the credit of this achievement to Lukeen on a subsequent page (p. 277). Raymond repeats this error in his Report of a Reconnaissance of the Yukon River (1871). Mr. Dall's work above cited constitutes a veritable mine of information on the subjects of which it treats, and is frequently referred to in the sequel. Where, therefore, as in the above case, criticisms are offered, it is in no spirit of detraction. + Vol. xxxviii. 12 GEOLOGICAL SURVEY OF CANADA. same sheet, the river above the Porcupine is named the Yukon or Pelly. Whymper refers to the river as the " Yukon (or Pelly) as it has long been called on our maps."* In the United States Coast Sur-* vey map dated 1869f the main river between the Porcupine and Lewes is definitely named the Yukon ; but in the map accompanying Ray- mond's official report (1871) this name is again confined to the river below the Porcupine, and the statement is made in the report (p. 21) that from Lake Labarge to Fort Yukon the river is called the Lewes. Further Much later, Lieut. Schwatka, in the maps accompanying the official nomenclature, report of his explorations of 1883 and in other maps elsewhere pub- lished, in defiance of the fact that the name of the Lewes had a pub- lished priority of thirty years, erased it completely, extending the name Yukon so as to include under that designatipn the Lewes River. This extension of the name Yukon appears to be justified by Lieut. Schwatka on the ground that the Lewes is the larger branch at its confluence with the Upper Pelly. As elsewhere stated, this is no doubt true, but from what is now known of the Upper Pelly, that river is almost certainly the longer, its sources are furthest removed from the mouth of the Yukon and its course is more directly' in con- tinuation of its main direction than is the case with the Lewes. Granting, however, that the Lewes excelled in all these particulars, it would still, I believe, be unjustifiable to alter an old established name for the sole purpose of giving to a river a single name from its mouth to its source. In any case it is incorrect to state that the Yukon (f.ewes) rises in Lake Lindeman, or streams flowing into it, as is done by Schwatka, for by far the greater part of the water of the river enters by the Taku arm of Tagish Lake. With respect to the substitution of the name Yukon for that of Pelly on the portion of the river between the Porcupine and Lewes, it is simply a question of well established priority versus use. It is pos- sibly a matter of small importance which shall be employed in future, but no valid excuse can be offered for the attempt to substitute any new name for that either of the Lewes or Pelly above the site of old Port Selkirk. Tru^ourcesof' From the point of view of the physical geographer, and apart from the question of nomenclature, the position of the furthest source of the great Yukon River is, however, an interesting subject of inquiry; though it may yet be som e years before we are in possession of suffi- * Op. cit., p. 223. " ~ t In United States Coast Survey Report for 1867 the same nomenclature is adopted, as it is also in the map accompanying Mr. Ball's Alaska (1870), in the Ethnological map of Alaska by the same author (1875), and in most later maps. These, however, do not embody any original data for this region. YUKON DISTRICT AND NORTHEBN BRITISH COLUMBIA. 13 cient information to settle this question definitely. It may be con- fidently assumed that this point is to be found by following up either the Pelly or the Lewes from their confluence at the site of old Fort Selkirk. As already stated, the Lewes there carries the greater volume of water, but draining as it does a considerable length of the humid Coast Ranges, which bear throughout the year great reserves of snow and numerous glaciers, it does not compare on terms of equality with the Upper Pelly, which unwaters a region relatively dry. Whether reckoned by size or distance from its mouth, the source of the Lewes must be placed at the head waters of the Hotilinqu River,* explored by Byrnes, of the Telegraph Survey, in approximate latitude 59° 10', longitude 132° 40'. In regard to the Pelly, it is not yet absolutely certain that the Pelly proper rises further from the common point at Fort Selkirk than its great branches, the Macmillan and the Ross rivers, but it is highly probable that it will be found to do so. With the above facts premised, we may compare the respective dis- Comparison of tances of assumed or probable sources of the Yukon as below, the tributaries, distances being in each case measured in a straight line from the common point at Fort Selkirk : — Fort Selkirk to summit of Chilkoot Pass, source according to Sohwatka (position fixed) 224 miles. " " to head of Hotalinqu River of Telegraph Survey (position approximate) 294 " " " to " Pelly Banks " (position fixed) 213 " to head of Pelly Lakes (above "Pelly Banks," according to Campbell's sketch) 276 " The upper lake on the Pelly must be fed by a considerable stream or streams, the addition of the length of which, if known, would add considerably to the last of the above distances. I must confess to having been somewhat disappointed in the size of size of the the Pelly or Yukon where we saw it below the confluence of the Lewes. ''"''"''■ The river is there, when undivided by islands, about 1700 feet only in width, with a maximum depth scarcely exceeding ten feet when at a stage which may be considered as its approximate mean. It appeared to me to be about equal in size and velocity to the Peace River at Dun- vegan and Mr. Ogilvie, who is also familiar with the Peace, concurred in this estimate. Below this place the river, of course, receives a num- Comparison ber of important tributaries, but at any fairly comparable point on the wit'j^Macken- two rivers I believe that the Mackenzie must far exceed the Yukon in volume. Numerical data on this point are unfortunately still almost * The Tes-lin-too occupies the main orographic valley above its confluence with the Lewes, but is smaller than the Lewes, and besides doubles back on its course, as is rfiown on the map. 14 GEOLOGICAL SUEVEY OF CANADA. Discharge of rivers. entirely wanting, but the comparison of the drainage-areas of the two rivers, according to the latest available information, strongly bears out the statement just made, that of the Yukon being 330,912 square miles as against 677,400 square miles in the case of the Mackenzie.* In other words, the drainage area of the Mackenzie is more than double that of the Yukon, while nothing is known to show that the mean annual precipitation over the two areas, as a whole, differs very greatly. Exaggerated statements which have been made, to the effect that the Yukon discharges a volume of water approaching that of the Missis- sippi, appear to carry their own refutation in the fact that the basin of the latter river has an area of no less than 1,226,000 square miles. Some attempt was made by us to gauge the flow of several of the rivers in the Yukon basin, as more fully stated further on. The dis- charge of the several rivers above the site of the old Fort Selkirk may be roughly summarized as follows. The iigures given in heavy type are derived from measurements more or less approximate, the others are based merely on comparisons made by eye and without any attempt to cross-section the streams. The scheme may, however, serve to give a general idea of relative dimensions. All the figures represent cubic feet per second and refer to the latter part of the summer, when the rivers may be assumed to be at their mean height. In common with all the streams of the interior region these are highest in the early summer and lowest toward the end of the winter. Upper Pelly and Lewes compared. Pelly River, at " Pelly Banks " 4,898 Ross River, at mouth 4,898 Maomillan River, at mouth : 9,7961 Tes-lin-too, at mouth 11,436 Lewes River, above confluence with last 18,664 Big Salmon 2,726 Total thus accounted for .52,418 Pelly or Yukon at site of Fort Selkirk 66,955 Difference from above total, not accounted for. . 14,537 By adding two-thirds of the water thus not accounted for to the Pelly and one-third to the Lewes, to represent the flow of numerous smaller tributaries not enumerated, the discharge of these two rivers above Fort Selkirk will stand thus : — * Of the area drained by the Yukon, about 150,768 square miles are in Canadian, 180,144 square miles in United States territory. + The discharge of the Ross is taken as equal to that of the Pelly at " Pelly Banks," that of the Maomillan as equal to the combined waters of the Pelly and Ross. * YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 15 Lewis 37,672 Pelly 129,283 Total as above 66,955* Navigable Waters and Routes of Travel. The numerous large and important rivers by which the Yukon district and the adjacent northern portion of British Columbia is intersected, constitute the principal routes of travel, and during the summer months render inter-communication comparatively easy. The Stikine is navigable by stern-wheel steamers for a distance of 138 miles, as more fully stated in a subsequent part of this report, where also details respecting the connecting trail to Dease Lake are given. This constitutes the travelled route to the Oassiar mining district. A trail was, at one time, opened from Eraser Lake overland to Dease Lake by which cattle were driven through, but of late no travel has ocoiirred on it. The Dease River can scarcely be considered as navigable for steamers, though constituting Navigable a fairly good boat route. The Upper Liard and Frances rivers, above the mouth of the Dease, are also passable for large boats, with occasional portages, but not so for steamers. The difficulties of the Lower Liard, however, are such as to render it an undesirable route, even for boats, and scarcely suitable as an avenue of trade between Oassiar and the Mackenzie. Numerous tributary streams in this district may also be ascended by boat or canoe for considerable distances, though with many interruptions from rapids and bad water. Communication may easily be established by railway from the mouth Possible of the Stikine to the centre of the Cassiar district and beyond, when * For the purpose of comparison, the following extract from a table of various rivers, contained in the General Report of the Minister of Public Works for the fifteen years from 1867 to 1882, is quoted : — NAMES. Mississippi St. Lawrence Ganges Nile Thames Rhone Rhine Ottawa (Grenville) Area of drainage in square miles. 1,226,000 565,000 432,000 520,200 5,000 38,000 80,000 80,000 Length in miles. 4,000 2,600 1,680 2,240 215 560 700 700 Discharge in cubic feet per second. Low Water. 447,200 36,300 23,100 1,330 7.000 13,400 35,000 Mean. High Water. 900,000 207,000 220,000 21,066' 33,700 85,000 1,270,000 494,200 7,900 204,000 161,000 150,000 16 GEOLOGICAL SDRVEY OF CANADA. Taku River and trails. Tributaries of Yukon. Navigable lengths of rivers. Aggregate length of navigable waters. this shall be called for, and it is probable also that this district might;, without difficulty, be connected by rail with the more southern por-, tions of British Columbia by one or more routes of which the main outlines can already be indicated. Following the river valleys, by a route practicable for a railway, from Rothsay Point at the mouth of the Stikine to the mouth of the Dease, the distance is found to be 330 miles. Thence to Fort Simpson on the Mackenzie, is a further dis- tance of 390 miles, making the total distance by this route, from the Pacific to the navigable waters of the Mackenzie, about 720 miles only. Little is yet known of the Taku River, but the Indians ascend it in canoes to a point at a distance of about eighty miles from the head of Taku Inlet, and Indian trails lead south-eastward from this vicinity to the Tahl-tan, eastward to Tes-lin Lake and north-eastward to the lakes near the head of the Lewes. From what has been ascertained of these, it is probable that it would not be difficult to construct a trail suitable for pack-animals, if not a wagon road, from the vicinity of the head of navigation on the Stikine to these lakes connecting with the navigable waters of the Lewes. The rivers draining the Upper Yukon basin, have in general lower grades, and affiDrd better navigable water than chose above referred to, and are in consequence likely to prove of greater importance in con- nection with the exploration and development of the country. The distance to which they may be respectively ascended by boat or canoe has as yet been determined in only a few cases. It may, however, be stated that the Yukon is continuously navigable for small steamers from its mouth, on Behring Sea and following the Lewes branch, to Miles Canon. Thence, after an interruption of about three miles, to the head of Bennett Lake and to an additional con- siderable — though not precisely determined distance — by the waters extending south-eastward from Tagish Lake. The Tes-lin-too is probably navigable for stern-wheel steamers for a hundred and fifty miles or more from its mouth while the Tahk-heena and Big Salmon rivers may probably both be ascended by steamers of the same class for some distance. From the site of old Fort Selkirk, again, the Pelly might be navigated by small steamers of good power to within about fifty miles of the site of old Fort Pelly Banks, and the Macmillan branch is also navigable for a considerable, though not ascertained distance. The same may be said of the Stewart River, but White River is, so far as known, very swift and shoal. The total length of the waters which may be utilized for navigation by light stern- wheel steamers on the main river and its branches to the YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 17 east of the 141st meridian or Alaskan boundary, measured in straight lengths of fifty miles, is therefore at least 1000 miles, and following the sinuosities of the various streams would be very much greater. This does not include the Porcupine River, and with the exception of the single break above referred to on the Lewes, forms a connected system, all parts of which lie to the east of the above meridian. If the upper portion of these rivers, above the first obstacles to such navigation, were included, the total here given would doubtless be greatly added to. At the present moment but three routes of access to the Yukon dis- Means of trict are employed. (1). That of the portage by the Chilkoot Pass y "^^^j^ ° from the head of Lynn Canal to the navigable waters of the Lewes. District. (2). That from Peel River, near its confluence with the Mackenzie by portage to La Pierre's House on a branch of the Porcupine. (3). That from Behring Sea by the main river. The first is that almost exclusively used by the miners, the second is employed only by the Hudson's Bay Company, and the last is that of the Alaskan traders. There are now three small stern-wheel steamers on the lower river, which ascend each year as far as the trading post at Forty-mile Creek, bringing the greater part of the goods used in trade with the Indians and for the supply of miners. It is not possible, however, for miners to reach the scene of their operations by this route in time to make a season's work, and the chances of reaching or leaving the Yukon mouth are few and precarious. Particulars relating to the Chilkoot Pass and Lewes River will be found on later pages. The character of the pass is such that it would scarcely be possible to construct a useful trail across it for pack-animals, but the White Pass appears to offer a better opportunity for making a trail or road which, if constructed, would render the entire region much more easy of access. Another route, also leading from the head of Lynn Canal to navigable water connecting with the Lewes, is that by the Chilkat Pass. This was formerly much employed by the Indians, but entails a much longer land carriage, one which is said to occupy the Indians for twelve days when carrying packs, as against two days of packing by the Chilkoot Pass.* The Indians inhabiting the region to the south and east of the site Indian routes, of old Fort Selkirk are poor boatmen and follow the various rivers in the course of their periodic journeys to a very limited extent. Most of their travelled routes appear, indeed, to run nearly at right angles to the direction of drainage, the rivers being crossed in summer on * [All the above remarks respecting routes, of course refer to the date of publication of the Eeport.] 2 GEOLOGICAL SURVEY OF CANADA. General character of climate. Regions of heavy and light rainfall. Progress of spring. rafts, the remains of which may frequently be observed. In travelling thus they carry their entire small camping outfit on their backs. Climate, Agriculture and Flora. While the available information as to the climate of the northern portion of British Columbia and the Yukon district is necessarily as yet very imperfect, its general features are sufficiently obvious, repeat- ing as they do those met with in the similarly circumstanced region to the south, with such modification as is produced by their higher latitude. The coast and coastward slopes of the Coast Ranges consti- tute a belt of excessive humidity and great precipitation, with some- what equable temperatures, while the interior region to the eastward of these ranges is relatively dry, with a temperature of extremes.* In the interior, however, the climate is largely influenced by the altitude of each particular district, and in consequence of the general lowering of the country beyond the 60th parallel (constituting the north line of British Columbia), it is certain that the climatic conditions are there much more favourable than in the Cassiar district. The mean annual temperature of the coast region is considerably higher than that of the interior ; yet, in consequence of the great depth of the snow fall and persistently clouded character of the skies, the Coast Ranges are found to support numerous and massive gla- ciers, while these are almost or altogether absent in the Cassiar Mountains, in the mountains about Frances Lake and in the other ranges seen by us in the interior. The heavy accumulation of snow upon the Coast Mountains and in their valleys, retards the progress of spring, as is very clearly evidenced on the Stikine, and explained more fully elsewhere. The depth of snow in winter continues to be inconsiderable or moderate, at least as far down the Pelly (Yukon) as the mouth of Stewart River and Forty-mile Creek, while at Nulato, on the lower river and in a similar latitude, but 500 miles further west, the depth of snow from April to November is said to average eight feet and often to reach twelve feet.f Mr. Dall also writes : " The valley of the Lower Yukon is somewhat foggy in the latter part of summer ; but as we ascend the river the climate improves, and the short season at Fort Yukon is dry, but pleasant, only varied by an occasional shower." Relatively to the country of the Upper Yukon basin, the advent of spring is much retarded in the country to the west, and it is stated that on the river below Nulato * A mean of the total annual precipitation for Fort Tongass, Wrangell and Sitka gives a general mean for the coast of 86"84 inches. f Alaska and its Resources, W. H. Dall, 1870, p. 437. YUKON DISTRICT AND NOETHBRN BRITISH COLUMBIA. 19 alder buds were found just opening and tender leaves beginning to appear on the 4th of June. These and other facts seem to show- conclusively, that in the absence of a continuous mountain barrier in that region, the humid winds of the Pacific are enabled to push eastward a long way up the Yukon valley, carrying with them the belt of heavy snowfall, which ceases to be conterminous with the Coast Mountains, as it is to the south-eastward. As in the more southern parts of British Columbia, the dryest Dry belts. country is found to occur in a belt bordering the eastern or lee side of the Coast Ranges, and this phenomenon recurs, though in a less marked degree, in connection with each of the well defined mountain ranges of the interior. Thus a region of greater humidity is found near Dease Lake, on the western side of the Cassiar Mountains, with a dry belt on the east side of the range ; while humid conditions, with recurrent showers in summer, characterize the district in the vicinity of Frances and Finlayson lakes. Further illustrations of this fact, with other climatic observations, will be found in the body of this report. A noteworthy circumstance in connection with the Stikine valley, Summer and the passes leading from the head of Lynn Canal, and doubtless in all '^™^^^ winds, the low gaps in the Coast Ranges, is the change in direction as be- tween the summer and winter winds. During the summer strong winds blowing up these valleys inland, are of very frequent occurrence and they commonly freshen in the afternoon and die away toward night. In the winter months the conditions are precisely reversed, the strongest winds blowing seaward. The summer winds are doubtless homologous with the sea breezes observed in many other regions, while the direction of the winter winds probably depends on the existence of a persistent anti-cyclonic area in the interior during that season. The temperature of Wrangell, just off the mouth of the Stikine, may Temperatures probably be taken as fairly representative of that of the coast in these ^terior latitudes. For the interior region, here particularly treated of, we are unfortunately without a series of thermometer readings extending even over a single year, but some idea of its climate may be formed from that of Fort Yukon, which is, however, situated far to the north, almost exactly on the arctic circle. The mean seasonal temperatures for these two stations may be compared as below.*—- * From the United States Coast Pilot, Alaska, Part I., 1883, p. 269. 2* 20 GEOLOGICAL SURVEY OP CANADA. Wrangell. Fort Yukon. Spring 40-4 14-6 Summer 57-1 56-7 Autumn 430 17-4 Winter 28-3 —23-8 Year 42-2 16-8 Comparison with Russia. Crops and cultivation. In the central provinces of European Russia, the thermometer descends to — 22° and — 31°, and occasionally even to — 54°, in the winter months, but rises at times to 104° and even to 109° in summer. The rainfall is small, varying from sixteen to twenty-eight inches, the maximum precipitation taking place during the summer months, and not, as in western Europe, in the winter, while the months of advanced spring are warmer than the corresponding months of autumn, t So far as our information goes, the above statement might almost be adopted as characterizing the climate of the southern half of the Yukon district. At Telegraph Creek and in its vicinity on the Stikine, to the east of the Coast Ranges (lat. 58°), wheat, barley and potatoes are successfully grown with the aid of irrigation. Their cultivation has so far been attempted on a limited scale only, on account of the want of any mar- ket, and wheat has been grown only experimentally, as it cannot, like bai'ley, be employed for feeding pack-animals. None of these crops can be successfully grown or ripened on the coastward side of the mountains. At Fort Yukon (situated, as above noted, on the arctic circle) Mr. Dall states he was informed that barley had once or twice been tried in small patches and had succeeded in maturing the grain, though the straw was very short. J A few cattle were also kept here at one time, when the post was in the possession of the Hudson's Bay Company. PetrofT, in his Census Report on Alaska, endeavours to discredit Ball's statement as to the growth of barley at this northern point, but I am fortunately in possession of independent evidence as to its accuracy, the late Mr. James Anderson, of the Hud- son's Bay Company, having noted in an official report on the district that both potatoes and barley have been grown at the fort. Conclusions as Taking into consideration all the facts which I have been able to- agriculture, obtain, as well as those to be derived from an examination of the natural flora of the country, and the observed advance of vegetation, which, in the absence of actual experiments, are capable of aflbrding valuable data, I feel no hesitation in stating my belief that such hardy t Encyclopedia Britannica, vol. xxi., p. 67. J Op. cit., p. 441. YUKON DISTRICT AND NOETHEEN BEITISH COLUMBIA. 21 crops as barley, rye, turnips and flax can be successfully cultivated in the Yukon district as far north as the former position of Fort Selkirk, near the 63rd parallel, or in other words about 1000 miles north of Victoria. Taken in conjunction with the physical features of the region, this means, that chiefly withia the drainage area of the Yukon, and for the most part to the north of the 60th parallel, there exists an area of about 60,000 square miles, of which a large proportion may — and doubtless in the future will — be utilized for the cultivation of such crops, and in which cattle and horses might be maintained in sufiicient number for local purposes, without undue labour, as excellent summer grazing is generally to be found along the river-valleys and natural hay-meadows are frequent. I do not maintain that the country is suitable for immediate occupation by a large, self-supporting agricul- tural community, but hold that agriculture may before many years be successfully prosecuted, in conjunction with the natural develop- ment of the other resources of this great country, of which by far the most valuable portion lies to the east of the line of' the Alaskan boundary. A note on the distribution of the various species of trees and on that Trees, of some of the herbaceous plants forms a separate section of this report (Appendix I.), while in Appendix III., Prof. Macoun gives a list of the plants collected.* Remarks on the quantity and quality of timber along the various Timber, routes are given under the local headings. It may suffice here to state, in this connection, that the country is generally wooded,! and that in all portions of it, in valleys and on low lands, there is an abundance of white spruce, of fair to good quality, well suited for purposes of con- struction. The other species of trees present are of inferior economic importance. Fauna. The fauna of the region traversel by us, does not difi'er notably from Larger ani- that of other parts of the northern country which are already mod- '""■ ^ °°''™' erately well known. There are, no doubt, many interesting points yet to be determined in respect to distribution, but our opportunities for obtaining information of this kind were very limited. The smaller black-tailed deer {Cariacus Columbianus) occurs on the islands of the southern portion of Alaska and the adjacent mainland coast, but is nowhere found on the inland side of the Coast Ranges. The mountain * [These are not here reprinted.] t No areas of timd/ra or frozen moraBS, such as are stated to be characteristic of the country of the Lower Yukon, were found in the region here reported on. 22 GEOLOGICAL SURVEY OF CANADA. Smaller fur-bearing animals. Salmon. goat is moderately abundant in the Coast Ranges, and is also found in the mountainous inland regions, probably throughout. The big-horn or mountain sheep occurs, together with the last mentioned animal, on the mountains about the head of the Lewes and other parts of the inland spurs of the Coast Ranges, but does not inhabit the seaward portions of these ranges. It is also found generally in the mountains of the interior, including the Rocky Mountains. The moose is more or less abundant throughout the entire inland region, and together with the caribou, which is similarly ubiquitous, constitutes a great part of the food of the Indians. We found the moose particularly plentiful along the Upper Liard River, and it is stated that the country drained by the White River is noted among the Indians as a moose and beaver region. The caribou is everywhere common, but is scarcely seen in the valleys or' lower country during the summer, when it ranges over the high, alpine moors and open slopes of the mountains. The black and grizzly bears roam over the entire region and are often seen along the banks of the rivers in the latter part of the sum- mer when dead or dying salmon are to be obtained with ease. Wolves are not particularly abundant, but the cross, black and silver-fox are more than usually common. The smaller fur-bearing animals, being similar to those found gene- rally in the northern parts of the continent, do not require separate enumeration. The entire Upper Yukon basin, however, yields furs of exceptionally high grade. Some notes as to the quantity annually obtained from the region will be found in a subsequent paragraph (p. 25 b). Among a few skins brought back by us, is that of a mouse which Dr. 0. H. Merriam has found to be a new species, and has described under the name of Evotomys Dawsoni* The salmon ascend tne Lewes River as far as the lower end of Lake Marsh, where they were seen in considerable numbers early in September. They also, according to the Indians, run almost to the head-waters of the streams tributary to the Lewes on the east side. Salmon also run up the Pelly for a considerable distance above the mouth of the Lewes, but their precise limit on this river was not ascertained. The lakes and rivers generally throughout the country are well supplied with fish, and a small party on any of the larger lakes would run little risk of starvation during the winter, if provided with a couple of good gill-nets and able to devote themselves to laying in a stock of fish in the late autumn. *American Naturalist, July, 1888. YDKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 23 As might be anticipated from the interlocking of streams tri- other fishes, butary to the Mackenzie and Yukon in this region, the fishes in both drainage-areas appear to be identical, so far as I was able to observe, with the exception of the salmon, which, is of course, confined to the Yukon tributaries. The principal fishes noticed are white-fish (Core- (jonus Nelsoni), lake trout (Salvelinus Namaycush), grayling {Thy- mallus signifer), pike (Usox lucius), and sucker {Catostomus catos- tomus). The names above given are on the authority of Dr. T. H. Bean, of the U. S. Commission of Fish and Fisheries, who has very obligingly examined for me the photographs of fishes which were taken. No photograph, unfortunately, was obtained of the salmon seen on the Lewes, etc., but Dr. Bean informs me, from my description of its size, that he has little doubt it was the king salmon, Oncorhynchus chuicha. Appendix IV. includes a list by Mr. James Fletcher, F.R.S.O., of Insects, the species of insects collected.* Mining and Minerals. Mining has so far been confined, within the Capsiar district and in Placer gold- ihe Upper Yukon basin, to the working of gold placers, and in the "i™™?- latter, almost entirely to river-bar mining, the inception of which indeed dates only from 1880. Particulars with reference to the rich creeks of Cassiar will be found elsewhere, and facts relating to the rivers tributary to the Yukon on a later page. Almost all the large streams which have been prospected in the Yukon basin, have been found to yield placer gold in greater or less quantity and the aggregate length of the rivers thus already proved to afford gold is very great, but little has been done toward the examination of their innumerable smaller feeders. Similar river-bar mining on the Stikine and Liard rivers preceded the discovery of the smaller creeks in which the richer de- posits of " heavy " gold were oljtained, and a few miles in length each of Dease, Thibert and McDame creeks produced the greater portion of the $2,000,000 worth of gold credited to Cassiar in 1874 and 1875. Discoveries similar to these may be expected to occur at any time in the Yukon district, the generally auriferous area of which already proved is very much greater than that of Cassiar. Scarcely anything Metalliferous has been done as yet, even in the Cassiar district, toward the search ^^"^^• for or proving of metalliferous veins, and practically nothing in the Yukon district, but. there can be no reasonable doubt that such deposits * [Not here reprinted.] 24 GEOLOGICAL SURVEY OP CANADA. Importance of metalliferous belt. Bearings of climate on mining. Statistics of gold. Asbestus. exist.* The present activity in mining enterprise in the southern part of British Columbia will, before long, spread to this northern region also, and then, if not before, its valuable character as a portion of the metalliferous belt of the continent will be realized. The Yukon district with the northern part of British Columbia, measured from the vicinity of Dease Lake to the intersection of the Pelly (Yukon) with the 141st meridian comprises a length of over 500 miles of the Cordillera belt of the west, which, wherever it has been examined, has been found rich in minerals and particularly in the deposits of the precious metals. The width of this particular part of the Cordillera belt is also great, as it appears, so far as our explorations have gone, to extend from the coast to the eastern ranges of the Rocky Mountains in the vicinity of the Mackenzie River. This portion of the Cordillera region, together with that of the more southern part of British Columbia, gives an aggregate length of between 1200 and 1300 miles, almost exactly equal to the length of the same metalliferous belt contained by the United States, and in all probability susceptible of an eventual mining development equally great. In the northern districts here reported on, it is true that the winter climate is a severe one, rendering the working season for ordinary placer-mines short and likely also to present some special difficulties in the way of " quartz mining." There is, however, on the other hand, an abundance of wood and water, matters of great importance in connection with mining, and means of communication once provided, mining operations should be carried on here at less cost than in dry and woodless regions such as are great portions of Arizona. Statistics of the former and present gold production of Cassiar are given in connection with that district, on a later page. It is difficult if not impossible to arrive at even an approximate statement of the total amount of gold which has been so far afforded by the Yukon district, but from such inquiry as I was able to make in 1887, I estimated the value of gold obtained in that year at a minimum of $60,000 ; the number of men engaged in mining at 250. A specimen of asbestus (chrysotile) being part of a small vein of that material about an inch in thickness, has been brought from the Stewart River, and the occurrence of serpentine in large mass elsewhere, tends to show that workable asbestus deposits may yet be found in that region. " A specimen of galena obtained from MoDame Creek, Cassiar, was found to con- tain 75 ounces of silver to the ton and of seven specimens of veinstuff collected by us on the tipper Pelly and Lewes, five proved to contain distinct traces of gold on assay. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 25 Platinum is found in small quantities along all or nearly all the platinum, tributaries of the Yukon, in association with the gold. It has also been observed in the Cassiar district. Fur Trade. Gold and furs are at present the only articles of value derived froHl j; „ f the great region here referred to as the Yukon district. It is im- of furs, possible to secure accurate information as to the value of furs annually obtained, but sufficient is known to show that it must be very con- siderable. Petroff, in his report, states that the total annual value of the furs shipped by the Yukon probably does not exceed $75,000,* and it is known that a great, if not the greater, portion of this total is derived from the region lying east of the 141st meridian. Dall states that, at the date of his visit (1867), the value of furs annually obtained at Fort Yukon, then maintained by the Hudson's Bay Com- pany, was not less than $50,000. Captain Raymond notes that the total number of skins collected in 1869, at this place, was stated at 10,000, but adds that he believes this estimate to be excessive.! Prac- tically the whole of these may be regarded as having been brought by Indians from the region east of the Alaskan line. An approximate estimate of the furs derived from Canadian territory and taken down the Yukon, obtained from Mr. Frangois Mercier, who spent many years trading on the river, places the annual value at about $27,000. The annual catch is made up, according to the same authority, about as follows ; — Beaver 1200 to 1500 skins. Cros.s fox 100 Black fox 100 Red fox 300 Bear 300 Marten 4000 Otter 200 Mink 2000 Lynx 600 Wolverine 150 Wolf 100 9350 In addition, however, to the furs taken from the Yukon district by this route, the Hudson's Bay Company obtains a large quantity of skins export of furs, from their posts on the Porcupine, which reach the market by the * Report on the Population, Industries and Resources of Alaska, p. 5, IT. S. 10th Census, vol. viii. t Op. eit. p. 115. 26 GEOLOGICAL SURVEY OF CANADA. Mackenzie River route. A certain number of skins derived from the country north of British Columbia is, further, annually traded at the little post at the mouth of Dease River, and taken out by the Stikine. A considerable quantity of furs also each year finds its way by the Chilkoot and Chilkat passes to the head of Lynn Canal, and some are brought down by the Taku River to the coast, though the greater part of these last is probably derived from the north-westei'n corner of the province of British Columbia. Information obtained on the spot indi- cates that the value of the furs reaching Lynn Canal from the interior is from $12,000 to $15,000 annually. Economic Importance of the Region. Value of the Yukon district. Isothermal lines. Without including the northern part of British Columbia, respecting which more has already been made known, but restricting ourselves to the great area of 192,000 square miles situated to the north of the 60th parallel and west of the Rocky Mountains, which I have referred to as the Yukon district, it may be said that the information now obtained is sufEcient to warrant a confident belief in its great value. Very much yet remains to be learned respecting it, but it is known to be rich in furs, well supplied with timber, and it is traversed by a great length of navigable rivers. It is already yielding a considerable yearly product in gold, and presents every indication of a country rich as well in other metals, and including deposits of coal. In its southern portion, situated between the 60th and 65th degrees of latitude, is comprised an area of probably not less than 30,000 square miles, suitable for eventual agricultural occupation, and presenting none of the characters of a sub-arctic region, which have, in advance of its exploration, been attributed to it by some writers. In each of these particulars and in climate it is greatly superior to the corresponding inland portion of the territory of Alaska. It may, in fact, be affirmed with little room for doubt, that the region here spoken of as the Yukon district surpasses in material resources the whole remaining northern interior portion of the continent between the same parallels of latitude. The winter climate of the whole of this great region is known to be a severe one, and its northern extremity lies within the arctic circle, but it must be remembered that the climatic conditions on the western and eastern sides of the continent are by no means comparable, and that the isothermal lines representing the mean annual temperature, trend not westward but north-westward from the Manitoba recion. The lines, in particular, which would represent the mean summer tern- YUKON DISTRICT AND NOETHEEN BRITISH COLUMBIA. 27 perature would assume, in the far north-west, a proximate parallelism with the Pacific coast, instead of tending to follow lines of latitude. It is needless here to recapitulate the well known causes which produce this remarkable difference in climate, but the lines as already approxi- mately drawn upon the maps, represent in a generalized form the aggregate of influences which, working together, produce at the site of old Fort Selkirk on the 63rd parallel of latitude in the Upper Yukon basin, an attractive landscape, decked with well-grown forests and with intervening slopes of smiling meadow, while in the same latitude in Hudson Strait we find, even at midsummer, merely a barren waste of rocks and ice. To instance a region which reproduces the general conditions of the Comparison Yukon district and adjacent northern portions of British Columbia, ^''y pi'ci™oe we must turn to the inland provinces of Russia, to which allusion has already been made in connection with climatic features. The province of Vologda, in European Russia, appears to offer the nearest parallel. It is circumstanced relatively to the western shores of Europe, as is this district to the western shores of the North American con- tinent. Its area is 155,498 square miles, situated between the 58th and 65th degrees of latitude. The climate in both cases is a continental one, in which severe winters alternate with warm summers, and the actual degrees of cold and heat, so far as our information goes, are not dissimilar. There is no very heavy rainfall in either region, such as we find near the western coasts bordering on the Atlantic and on the Pacific respectively. The agricultural products from the province of Vologda are oats, rye, barley, hemp, flax and pulse. The mineral products comprise salt, copper, iron and marble, but the precious metals do not appear to be important, as in the Yukon district. Horses and cattle are reared, and the skins of various wild animals, as well as pitch and turpentine, are among the experts. The population of the province is stated at 1,161,000. While the Yukon district and the northern portion of British Colum- Ultimate bia are at present far beyond the limits of ordinary settlement, we ass™red™^° may be prepared at any time to hear of the discovery of important mineral deposits, which will afford the necessary impetus, and may result, in the course of a few years, in the introduction of a consider- able population into even its most distant fastnesses. To-day it may well be characterized by the term which has been employed in connec- tion with the Mackenzie basin, a portion of " Canada's Great Reserve." It appears meanwhile eminently desirable that we should, encourage and facilitate, in so far as may be possible, the efforts of the miners and others who constitute our true pioneers in the region, and to 28 GEOLOGICAL SURVEY OP CANADA. whom, in conjunction with the fur companies and traders, the peace^ ful conquest of the whole of our Great "West has been due. In the future, there is every reason to look forward to the time when this country will support a large and hardy population, attached to the soil and making the utmost of its resources. General Geology. Difficulties of In a reconnaissance carried out along a single line, in which the the field. greater part of one's time is necessarily occupied in overcoming the diffi- culties of the route and in securing the necessary geographical data, it is difficult to obtain any very complete knowledge of a region geologically complicated. In the present case this difficulty is increased by the circumstances that the geology of the corresponding portion of the Cordillera belt in the southern part of British Columbia, is as yet very imperfectly understood, though considerable attention has been devoted to it ; while with respect to the older rooks of the analogous region in the western part of the United States very little published informa- tion of a systematic kind is available. General Speaking broadly, however, and with reference to the general fea- with British tures of the region, the rock-series represented are evidently similar to Columbia. those found in 1 he southern portion of British Columbia between the Rocky Mountains and the coast, and an important general result of the work here reported on, is the further demonstration of the great constancy in lithological characters of the several formations, when followed in the direction of the main north-west and south-east axes of uplift — a constancy which contrasts markedly with the diversity found when comparisons are made as between localities situated at right angles to this direction. Rooks of the The Coast Ranges, where traversed by the valley of the Stikine, and Coast Ranges. ^^^:^^ where crossed still further north by the Chilkoot Pass, are found to consist, for the most part, of granite and granitoid rocks, almost invariably of gray colour and frequently rich in hornblende. With these are occasionally included stratified or stratiform masses of mica- and hornblende-schists, and both these and the granites are frequently traversed by pegmatite veins, diabase dykes and intrusive masses of coarse diorite. The schistose portions of these ranges may possibly represent the still recognizable remnants of rocks of Archaean age, or may be merely portions of much newer series which have suflered extreme alteration No demonstration of the date of the origin of the granitic rocks of the Coast Ranges was obtained in this region, but there is every reason YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. '29 to believe that it is comparatively recent, and due to a time lying be- tween the Triassic and the Cretaceous, as has been found to be the case with their continuation to the south, near the northern part of Vancouver Island.* The argillites of Wrangell, together with those met with near RqoIjs of the Juneau, and at Sitka, on the Alaskan coast, and also in various places ''°'^?*; . Sir c n 1 p6 idig o I along the east side of Lynn Canal, together with the altered volcanic rooks found in association with these on Lynn Canal and elsewhere (examined by me particularly in the vicinity of Seduction Point), closely resemble rocks of the same class composing the Vancouver group of the Queen Charlotte and Vancouver islands. Though no fossils were obtained at these northern localities, the rocks may, like those just referred to, be provisionally classed as Triassic, with the reservation, (as made in the case of the similar series of the Queen Charlotte and Vancouver islands), that Palaeozoic strata may also be represented. The width of the belt of granitoid rocks composing the Coast Ranges General is, on the Stikine, about sixty -five miles, measured from their sea- goj^gt^eit, border inland at right angles to the main direction of the mountains. It is somewhat less in the latitude of the Chilkoot Pass, but may be assumed to occupy a border of the mainland about fifty miles in width along the whole of this part of the coast. Broadly viewed, however, the coast archipelago in reality represents a partly submerged margin of the Coast Ranges, and granitic rocks are largely represented in it also. The examination of these two northern cross-sections of the Coast Ranges, serves, with observations previously made, to demon- strate the practical identity in geological character of this great orographic axis, from the vicinity of the Fraser River to the 60th parallel of north latitude — a length, in all, of about 900 miles. East and north-east of the Coast Ranges, the interior region tra- Formations of versed is, for the most part, floored by Palaeozoic rocks of very varied region. appearance, and probably referable to several of the main subdivisions of the geological scale. In so far as the information obtained in the region here in question enables conclusions on the subject to be formed, the lowest part of the rocks, (1) consists of greenish and gray schists generally felspathic or hornblendic, but often quartzose and including distinctly micaceous and talcose schists, with some bands of limestone ; the lithological character of this subdivision being exceedingly varied. Apparently overlying these are, (2) gray and blackish, often lustrous and sometimes more or less micaceous oalc-schists and quartzites, in- cluding beds of limestone of moderate thickness, which are often more *See Annual Report Geol. surv. Can., 1886. 30 GEOLOGICAL SURVEY OF CANADA. Fusuline limestones Altered volcanic materials. Relations of therockseries. or less dolomitic. These are associated with, or pass up into, (3) black- argillites or argillite-schists, also containing thin beds of limestone, which at one locality on the Dease, have afforded a small number of graptolites of Cambro-Silurian age. Next above these is a series (4) consisting chiefly of massive limestones, generally of gray or blue-gray colour where unaltered, but often locally changed into white or variegated crystalline marbles. These are closely associated with quartzites which usually show the peculiar fine-grained cherty character of those of the typical Cache Creek series on the Fraser and Thompson rivers. The thickness of this subdivision cannot (any more than that of those previously mentioned) be stated with precision, but that of the limestones alone must be several thousand feet in some places. On the Dease, on the Frances, and again on Tagish Lake, fossils of Carboniferous age, including more particularly a species of Fusulina, have been detected in some beds of this limestone series, probably belonging to its upper portion. Forms of the genus Fusulina are characteristic in certain zones of the Carboniferous limestone in Cali- fornia. They have been fpund by the writer in a number of places in British Columbia, which, with the discoveries here reported on, occur at intervals along a belt of country to the north-east of the Coast Ranges for a distance of over 800 miles. The limestone last-mentioned appears to be conformably followed or even in part interbedded with (5) a great mass of more or less evidently stratified rocks of volcanic origin, comprising amygdaloids, agglomerates, and other more massive ma- terials which apparently represent old lava-flows. All these are highly altered, so much so that in some cases their original physical character is scarcely demonstrable, while they have suffered changes also in con- stitution, having been converted for the most part into diabases. Analogy with the southern portions of British Columbia which I have examined, leads me to believe that the greater part of these volcanic materials are also to be classed as of Carboniferous age, but it is quite probable that here, as to the south, they comprise as well rocks of similar appearance which are of Triassic age, but which we are at present unable to separate from them. This is further rendered probable by the occurrence in certain black argillites at Glenora, on the Stikine, of Triassic fossils (p. 53) and by the discovery by Mr- McConnell of fossils of this age on the Lower Liard River, some dis- tance to the east of the region covered by this report.* No unconformity has been proved to occur throughout the whole of the above Palseozoic series, but the examinations made were scarcely of a sufficiently detailed nature for the detection of any stratigraphical * See Summary Report of the Operations of the G-eologioal Survey for 1887, p. 11. YUKON DISTRICT AND NOKTHEBN BRITISH COLUMBIA. 31 break unless of a very obvious character. Respecting the first- mentioned of the above subdivisions, I feel some doubt as to whether it really "constitutes a lower member of the series or whether it may re- present some of the other members — particularly the rocks of volcanic origin — in a highly altered state, as seems, from late observations, to be the case with rocks of similar appearance in southern British Columbia. The proximity of the rocks classed under the first subdivision to certain granitic axes is equally explicable on either hypothesis. It must also be added that there appears to be a recurrence of rock materials originally volcanic, in greater or less force in several parts of the series, and that important beds of serpentine occur at one or more horizons. For the purpose of assisting future more complete inquiry, and in ggolosioal view of the tentative character of the classification here offered, the notes on map. more important details observed are noted on the face of the map accompanying this report, for which it would, be premature to attempt a geological colouring. The preponderantly Palseozoic floor of the region east of the granites ., , . of the Coast Ranges, is broken through on two main lines by granite granitic axes, axes. The first of these is cut across by the Dease Rivgr a short dis- tance below Dease Lake, and was again met with — over 300 miles north- westward — on the Pelly near the mouth of the Maomillan. Though re- ferred to as a single granitic axis, this uplift probably consists rather of a series of alternating and more or less irregularly shaped granitic masses, which, however, preserve a general alignment. There are on the Upper Pelly in fact three separate granitic ridges in place of the single one met with on the Dease. In close association with these granites are some gneissic rocks and holocrystalline mica and hornblende-schists, which have not been referred to in previous paragraphs as they are regarded as probably Archsean, rather than as representing highly altered Palaeozoic rocks. A small tongue of granite occurs on the Lewes a few miles above the mouth of the Little Salmon, which may be con- nected with the south-western side of this granitic axis, but with this exception its continuity between the Dease and Pelly is indicated merely by the statement of Mr. J. McCormick that granites and mica-schists occur on the south-west side of Quiet Lake and near the Big Salmon River, below that lake. Its further extension in a north-westerly bearing is, however, proved by the occurrence of a great preponder- ance of rocks of the same character in the collection made by Mr. Ogilvie* on the lower Pelly or Yukon, between the mouth of the Lewes and Forty-mile Creek. *Sent out by him in charge of the latest party of miners in the autumn of 1887. 32 GEOLOGICAL SURVEY OF CANADA. Granites of Too-tsho Range. Connection of On comparing the position of this irregular granitic axis and its gold with the surrounding altered rooks (in part referable to several of the Palaeozoic rock series. ^ . subdivisions previously described) with that of the richer deposits of placer gold so far discovered and worked, it will be found that they are closely associated. The chief placers and river-bars are, in fact, scattered along this line or belt, and extend, like it, all the way from Dease Lake and McDame Creek to Forty-mile Creek. Evidence was moreover found on the Pelly, to show that the development of quartz veins in the Palseozoic rooks had occurred contemporaneously with the upheaval of the granites, and probably by some action superinduced by the granite masses themselves while still in a formative condition. While cutting the stratified rocks, the quartz-veins seldom or never cut the granite masses in this district. These observations should afford an important clue to the further search for auriferous ground, as well as for the lodes from which the placer gold has itself been derived. Of the second granitic axis of the interior region very little is yet known, but it is probable that it is still less regular in character than the last. It occurs in the mountainous region to the east of Frances Lake and River, and probably also in the vicinity of the Pelly Lakes. Its lithological characters and those of the rocks in its neighbourhood are similar to those of the last described, and here again in its vicinity, on Frances Lake and on the Liard paying gold placers have been found. The district is, however, so difficult of access that it can scarcely as yet be said to have been at all prospected. I am inclined to believe that the two granitic axes of the interior region above described are of much greater age than that of the Coast Ranges. The reasons for assigning a comparatively late date to the latter have already been alluded to. It is found, too, that while the stratified rocks usually conform to an ascending order in receding from these granitic axes, there is evidence along the north-eastern flanks of the Coast Ranges of an irregular line of junction, and though on the Stikine the Palaeozoic rocks appear to rest upon the granites of the Coast Ranges, the supposed lower members of the series are not seen, while on the lakes near the head of the Lewes some of the upper por- tions of the Palaeozoic are directly in contact with, and have apparently bean broken through by the granites. The granitoid rocks of the interior region are, moreover, different in general appearance from those of the Coast Ranges, and resemble more closely the probably Archaean granites of the Gold Ranges in southern British Columbia. Lithologically the granites and granitoid rocks of the Coast Ranges are generally fresh and unaltered in appearance, gray in colour and Age of granites. Lithological character. YUKON DISTRICT AND NORTHEEN BRITISH COLUMBIA. 33 tiot often distinctly foliated, while those of the ranges of the interior show evidence of considerable alteration subsequent to their formation, are more highly quartzose and often reddish in tint. Some particulars respecting a few of the granites of the region which have been micro- scopically examined by Mr. F. D. Adams are given elsewhere. Besides the Triassic rocks previously referred to, .the Mesozoic Cretaceous i, . and Laramie period is represented also by strata of Cretaceous and Laramie age. rooks. These rocks are distinctively more recent in appearance than, and rest quite unconformably on all the older formations, though they have since been to some extent involved in their flexures. On the lower part of the Lewes, below the mouth of the Little Salmon, these rocks are cut across by the river for a distance of at least thirty -five miles. Some fossil mol- luscs and plants have been obtained from this area, from which it would appear to include beds referable to the Middle or Lower Cre- taceous and to the Laramie period, and it is not improbable that the series is a consecutive one between these limits, as the total thickness represented must be very great. The strike of these beds varies much in direction, and the angles of dip are so irregular that no even proximate estimate of thickness could be formed, and it is impos- sible to arrive at any definite conclusion with respect to the trend of the basin in which they lie. The rocks comprise, in their lower por- tion, coarse conglomerates, grauwacke-sand stones, yellowish and gray quartzose sandstones and dark calcareous slates. The upper portion, in which Laramie plants are found, consists chiefly of rather soft sand- stones, shales and clays, generally of pale colours. Evidence of con- temperaneous volcanic action is observable in both parts of the series, and the higher beds include lignite-coal of good quality. Some miles further up the Lewes, midway between the Little and Big Salmon rivers, peculiar green, grauwacke-sandstones and green, highly calcareous conglomerates occur, which are also provisionally referred, though with some doubt, to the Cretaceous. They are at least newer than the Palseozoic rocks, being composed of fragments of these and of the granites. Conglomerates and sandstones similar to the last are again found Cretaceous of near the lower end of Lake Labarge, on the east side, and are asso- ciated with black calcareous slates, which recur in several places along the same side of the lake, further up, and from which a few fossils have been obtained. These seem to show that the beds are on or near the horizon of Series C, of the Queen Charlotte Islands, which is of Middle Cretaceous age, approximately equivalent to the Gault. On the Upper Pelly River, forty-three miles below Hoole Cafion, a Cretaceous of single low outcrop of hard, dark shales, containing fossil plants of 34 GEOLOGICAL SURVEY OF CANADA. I Cretaceous or Laramie age was found, but in the absence of further exposures along the river in that vicinity, nothing can be said of the\ extent of this area, except that it must be quite limited in width. Again, on the Stikine River, between Glenora and Telegraph Creek, there are local occurrences of conglomerates and soft sandstones which may be regarded as probably Cretaceous, though no palseontological evidence is forthcoming. Relations of The position of these last-noted areas, as well as that of those along the Cretaceous ^j^g Lewes River, occurring as they do in a zone of country immediately within the line of the Coast Ranges, is analogous to that held by Creta- ceous rocks on the Skeena and in other localities still further south- ward in British Columbia. Further investigation will probably show that rocks of this age occur in many additional places, and occupy somewhat extensive areas in this belt of country. In the vicinity of the Lewes, particularly, it is noted that the plane of the original base of the Cretaceous, now thrown into a number of folds, is about that of the present surface of the country, and these rocks may therefore be expected to recur frequently in the form of troughs or basins, more or less strictly limited and only to be discovered in detail by 'thorough examination. The loose material brought down by the Big Salmon River, appears to indicate the existence of a considerable development of these rock? not far up the valley of that stream. Tertiary rooks No wide-spread Tertiary areas like those of the southern interior portion of Briti4i Columbia appear to occur in the region here described. The most important occurrence of beds of this age met with, is that which occupies the wide valley of the Upper Liard, but its extent to the north-west and south-east was not ascertained. The rocks are soft . shales, sandstones and clays, generally of pale color, and holding beds of lignite in some places. Flows of basalt either cap these rocks or are included in their upper portion, and from the considerable angles of dip observed, the formation would appear to have suffered some flexure subsequent to its deposition. Basalts. In the Stikine valley, east of the Coast Ranges, important local basalt-flows are met with, overlying old river and valley-gravels and the lignite reported to exist some miles up the Tahl-tan is, doubtless, also of Tertiary age and inferior in poution to the basalts. Basalt efi'usioas of a sporadic character may be frequent in other places in the region, as such were actually noted in three other widely separated localities, viz., above Hoole Cafion on the Pelly, at Miles Canon on the Lewes, and again at the confluence of this river with the Pelly. Ageofbasalts. The basalts are at least pre-glacial in age, and though no character- istic fossils were observed in the associated bedded deposits, both may YUKON DISTRICT AND NORTHEEN BRITISH COLUMBIA. 35 be provisionally classed from their analogy with similar deposits in the more southern portion of British Columbia, as Miocene. I Occurrence of Jade on the Lewes. Having become interested in the question of the origin of nephrite Occurrence of or jade, on account of its former extensive employment by the natives J*"*^ pebbles. of the West Coast for the manufacture of implements,* I kept a close watch for this mineral along our route, and ultimately succeeded in iinding several rolled pieces of it in gravel-bars along the Lewes. Of the pebbles collected by us, at least five have the specific gravity a,nd other physical characters of jade, though they have not yet been subjected to chemical or microscopical analysis. Several of these are evidently, however, pure and typical jade, of which the finest and most characteristic was found by Mr. W. Ogilvie, near Miles Canon. This specimen is a pale green translucent to sub- transparent variety weighing a pound and three-quarters, after a piece, probably equal to about one-fourth of the original mass, had been broken ofi' and unfor- tunately lost. Some of the specimens collected, but not referred to in the above remarks, appear to show the passage, by admixture of other materials, of the pure jades into various altered rocks of volcanic origin, as described in the publication above referred to. So far as I have been able to ascertain, the discovery of jade here noted is, with one exception, the first actually direct one made in the region of the Pacific slope. The exception above alluded to is that of jade found at the Kwicbpak mouth of the Yukon during Captain Jacobson's stay in that vicinity and which was obtained by him and taken to Berlin, t Glaciation and Surface Deposits. Such details as appear to be of interest respecting glaciation, and the superficial deposits, are gii^en in the subsequent descriptive portion of this report. The general bearings of these are here merely sum- marized in the briefest possible manner. J Previous observations in British ColumbiaS have shown that at one -„ . . . Previous stage in the Glacial period — that of the maximum glaciation — a great observations confluent ice-mass has occupied the region which may be named the Columbia. * See Canadian Record of Science, vol. II., No. 6, April, 1887. f See paper by Prof. A. B. Meyer, Jahresberioht des Vereins fur Erdkunde zu Dresden, 1884. X The substance of this summary has been published in advance in the Geological Mag.%zine. Decade III., vol. V., p. 347 (Aug., 1888). § Quart. Journ. Geol. Soc, vol. XXXI. p. 89. vol. XXXIV. p. 272. Canadian Naturalist, vol. VIII. •3i 36 GEOLOGICAL SURVEY OF CANADA. Interior PLateau, between the Coast Ranges and the Gold and Rocky- Mountain ranges. Fr./iii the .JoUi to the 49th parallel, this grey,t glacier has left traces of its general .southward or south-eastward move- ment, which are distinct from those of subsequent local glaciers. The southern extensions or terminations of this confluent glacier, in Washington and Idaho Territories, have quite recently. been examined by Mr. Bailley Willis and Prof. T. C. Chamberlin of the U. S. Geolo- gical Survey, * and their observations tend to confirm the views above outlined, which had previously been stated by the writer. There is, further, evidence to show that this inland-ice flowed also, by transverse valleys and gaps, across the Coast Ranges, and that the fiords of the coast were thus deeply filled with glacier-ice, which, supplemented by that originating on the Coast Ranges themselves, buried the entire great valley which separates Vancouver Island from the mainland, and discharged seaward round both ends of the island. Further north, the glacier extending from the mainland coast touched the northern shores of the Queen Charlotte Islands. The observed facts on which these general statements are based have been fully detailed in the publications already referred to, and it is not here necessary to review former work in the region, further than to enume- rate the main features developed by it, and to connect these with the observations made during the summer of 1887, in the more northern region described in the present report. Glaciation of The littoral of the south-eastern part or " coast strip " of Alaska, presents features identical with those of the previously examined coast of British Columbia, at least as far north as lat. 59°, beyond which 1 have not seen it. The coast archipelago has evidently been involved in the border of a confluent glacier which spread from the mainland and \\ as subject to minor variations in direction of flow dependent on surface irregularities, in the manner described in my report on the northern part of Vancouver Island, t No conclusive evidence was here found, however, in the valley of the Stikine River or in the pass leading inland from the head of Lynn Canal, to show that the inland- ice moved seaward across the Coast Ranges, though analogy with the coast to the south favours the belief that it may have done so. The front of the glacier must have passed the outer border of the archi- pelago, as at ytika, well-marked glaciation is found pointing toward the open Pacific J (average direction about S. 81° W. astr.). the coast. * Bulletin U. S. Geol. Survey, No. 40, 18S7. t Annual Report Geol. Surv. CanaJa, i&85, p. 100 B. t Mr. G. F. Wright has already given similar general statements with regard to- this part of the Coast of Alaska, American Naturalist, March, 1887. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 37 It is, however, in the interior region, explored and examined by us Giaoiation of in 1887, between the Coast Ranges and the Rocky Mountains proper, the interior and extending northward to lat. 63°, that the most interesting facts have come to light respecting the direction of movement of the Cor- dilleran glacier. Here, in the valleys of the Upper Pelly and Lewes, traces were found of the movement of heavy glacier-ice in a northerly direction. Rock-surfaces thus glaciated were observed down the Pelly to the point at which it crosses the 136th meridian and on the Lewes as far north as lat. 61° 40', the main direction in the first-named valley being north-west, in the second north-north-west. The points referred to are not, however, spoken of as limiting ones, for rock exposures suitable for the preservation of glaciation are rather infre- quent on the lower portions of both rivers, and more extended exami- nation may result in carrying evidence of the same kind further toward the less elevated plains of the Lower Yukon, as elsewhere detailed. Neither the Pelly valley nor that of the Lewes is hemmed in by high mountainous country except toward the sources, and while local variations in direction are met with, the glaciation is not suscep- tible of explanation by merely local agents, but implies the passage of a confluent or more or less connected glacier over the region. On Lake Labarge, in the Lewes vaUey, both the sides and summits of rocky hills 300 feet above the water were found to be heavily glaciated, the direction on the summit being that of tlie main (north- north-west) orographic valleys, while that at lower levels in the same vicinity followed more nearly the immediate valley of the river, which here turns locally to the east of north. Glaciation was also noted in several places in the more mountainous country to the south of the Yukon basin, in the Dease and Liard valleys, but the direction of movement of the ice could not be deter- mined satisfactorily, and the influence of local action is there less certainly eliminated. While the greater part of the area traversed is more or less com- Deposits of the pletely mantled with glacial deposits, it will be observed, in referring S^^^^^ penod. to subsequent pages, that true boulder-clay was found in certain parts only of the southern and more mountainous portion of the region, while it spreads over almost the entire length of the Upper Pelly and Lewes valleys, though not found expo-ed quite to their confluence. The boulder-clay generally passes upward into, and is covered by, important silty beds, analogous to the silts of the jSTeohacco basin, further south in British Columbia, and to those of the Peace River country to the east of the Rocky Mountains. It may be stated also that the country is generally terraced to a height of 4000 feet or 38 GEOLOGICAL SURVEY OP CANADA. Mammoth remains. General facta established. Other instances of northward glaciation. more, while on an isolated mountain-top near the height-of-land between the Liard and the Pelly rivers (Pacific-Arctic watershed) rolled gravel of varied origin was found at a height of 4.300 feet, a height exceeding that of the actual watershed by over 1000 feet. No remains of mastodon or mammoth were observed in the country traversed by us, but according to Campbell such remains occur not far from the site of Fort Selkirk, and chey are known to be moderately abundant at points further down tVie river." Sir J. Richardson speaks of a tibia of Elephas primigenius sent to England by Roderic (Robert) Campbell from this region.* Reverting to the statements made as to the direction of the general glaciation, the examination of this northern region may now be con- sidered to have established that the main gathering-ground or n^v4 of what I have called the great Cordilleran glacier or confluent glacier-mass of the West Coast, was included between the 55th and 59th parallels of latitude, a region which, so far as explored, has proved to be of an exceptionally mountainous character. It would further appear that this great glacier extended, between the Coast Ranges and the Rocky Mountains, south-eastward nearly to lat. 48° and north-westward to lat. 63°, or beyond, while sending also smaller streams to the Pacific Coast. In connection with the northerly direction of ice-flow here ascertained, it is interesting to recall the observations which I have collected in a recently published report of the Geological Survey, relating to the northern portion of the continent east of the Mackenzie River. f It is there stated that for the Arctic coast of the continent, and the islands of the archipelago off it, there is a considerable volume of evidence to show that the main direction of movement of erratics was northivard. The most striking facts are those derived from Prof. S. Haughton's Appendix to M'Clintock's Voyage, where the occurrence is described of boulders and pebbles from North Somerset, at localities 100 and 135 miles north-eastward and north-westward from their supposed points of origin. Prof. Haughton also states that the east side of King William's Land is strewn with boulders of gneiss like that of Montreal Island, to the southward, and points out the general northward ice-movement thus indicated, referring the carriage of the boulders to floating-ice of the glacial period. The copper said to be picked up in large masses by the Eskimo, near Princess Royal Island, in Prince of Wales Strait, as well as on Prince of * Am. Journ. Sci. and Arts, vol. XIX., 18h5, p. 132. t Notes to accompany a Geological Map of the Northern Portion of the Dominion of Canada East of the Rocky Mountains, p. 57R., Annual Report Geol. Surv. Can., 1886. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 39 "Wales Island, * has likewise, in all probability been derived from the copper-bearing rocks of the Coppermine River region to the south, as this metal can scarcely be supposed to occur in place in the region of horizontal limestone where it is found. Dr. A. Armstrong, surgeon and naturalist to the Investigator, notes the occurrence of granite and other crystalline rocks not only on the south shore of Baring Land, but also on the hills at some distance from the shore. These, from what is known of the region, must be supposed to have come from the continental land to the south- ward. Dr. Bessels, again, remarks on the abundance of boulders on the shore of Smith's Sound in lat. 81° 30', which are manifestly derived from known localities on the Greenland coast much further southward, and adds : " Drawing a conclusion from such observations, it becomes e-\idcat that the main line of the drift, indicating the direction of its motion, runs from south to north." t It may further be mentioned that Dr. R. Bell, has found evidence of a northward or north-eastward movement of glacier ice in the northern part of Hudson Bay, with distinct indications of eastward glaciation in Hudson Strait. \ For the northern part of the great Mackenzie valley we are as yet without any definite published information, but Sir J. Richardson note.s that Laurentian boulders are scattered west- ward over the nearly horizontal limestones of the district. Taken in conjunction with the facts for the more southern portion Two great of the continent, already pretty well known, the observations here out- 8^^'^'i^'^" in£lSS6S> lined would appear to indicate a general movement of ice outward, in all directions, from the great Laurentian axis or plateau which extends from Labrador round the southern extremity of Hudson Bay to the Arctic Sea ; while a second, smaller, though still very important region of dispersion — the Cordilleran glacier-mass — occupied the Rocky Mountain region on the west, with the northern and southern limits above approximately given, and a length, in a nort]j:-west and soutJi-east direction, of at least 1200 miles. It is inexpedient at the present moment to enter into any detailed discussion of the glaciation of the extreme north-west, as Mr. Mc- Connell's observations, made in the prosecution of his portion of the work of the expedition, are likely to add much to our store of facts bearing on the subject. * De Ranee, in Nature, vol. XI., p. 492. + Nature, vol. IX. } Annual Report Geo], Surv. Canada, 1885, p. 14 DD. and Report of Progress, 1882-84, p. 36 DD. 40 GEOLOGICAL SURVEY OF CANADA. Economic I* ^^7 ^^ added, that while the study of the phenomena of the glacial importance of period is One not without its bearings on economic problems even in the eastern part of the continent, it has, in British Columbia and- the Yukon district, a direct value in its connection with the distribution of the placer gold deposits and on the existence and position of the buried channels of rivers and streams, in which some of the richest of those deposits are often found to occur. Thus the greater part of the "fine" gold found along the river-bars and banks of the larger streams in the Yukon district, is doubtless proximately derived from the gravels and other superficial deposits in which these streams have re-excavated their beds since the period of glaciation. By the general dispersion and intermixture of these materials, composed of the debris of the older rook formations, it is even possible that the existence of a few com- paratively limited areas of great richness might account for the wide- spread auriferous character of the alluviums of the Upper Yukon basin. In the former direction of ice-movement, and consequently that of its transport of material, we obtain an important clue as to the source of the finer gold which may now be found in any particular area. This subject is too wide in its ramifications to be followed out here, but it is one to which considerable attention has been devoted, and to which I hope to return at an early date in greater detail. Volcanic Ash Deposit. Extent of the -^ circumstance of some interest in connection with the later super- deposit, ficial depjsits of that part of the Upper Yukon basin drained by the Lewes and Pelly rivers, is the occurrence of a widespread layer of volcanic ash or pumiceous sand. The existence of a peculiar white line or band in the upper parts of scarped banks along the river, was first remarked not many miles below the point at which we reached and embarked on tlje Pelly. As its character was not at first under- stood, I omitted to note the precise point at which it was first seen but ajn of opinion that it probably extends to the east of the place where we reached the river. After recognizing its character and im- portance, however, it was looked for and noticed almost continuously along the whole course of the Pelly, as far down as the mouth of the Macraillan, beyond which, to the site of Fort Selkirk at the mouth of the Lewes it was not distinctly recognized, but according to Mr. McConnell (1888) it extends down the river for about ten miles below Fort Selkirk. It is likewise seen along nearly the whole course of the YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 41 Lewes, being last noted at the Narrows between Lake Nares and Bennett Lake, known as Caribou Crossing.* This ash deposit appears to be entirely due to a single period of origin of tbe eruption. It is homogeneous in character wherever seen, forming a deposit, single layer not divided by intercalations of other material, and has been spread everywhere over the entire area characterized by it. It is much more recent in date than the white silt deposits, which are the last of those properly referable to the glacial series, having been de- posited after the river-valleys were excavated in the glacial materials, and at a time when the rivers had cut down nearly or quite to their present levels, a fact rendered evident by the circumstance that it over- lies the deposits of river and valley-gravels and sands in all cases, ex- cept in those of some low river-flats, where these deposits sometimes cover it to a depth of several feet. In most places it is overlain merely by the surface soil with a depth of six inches to two feet, and in a few instances it was noted as constituting the actual surface of terraces of moderate height, the present forest being rooted in it. The ash appears to have fallen tranquilly, much in the manner of Mude of snow deposited from a calm atmosphere. The examination of scarped 'I'^P'j'sitionand banks along the two rivers showed it to occur near the surface of ter- races about 200 feet in height, as well as on lower terraces and river- fiats down to within about ten feet of the actual river-level in August and September. It was also detected in some places on the sloping fronts of terraces. The thickness of the layer was no doubt originally pretty uniform, and it still retains this uniformity where it rests upon wide flat terraces. Its average normal thickness for the Pelly, as a whole, was estimated at about five inches, but this is somewhat ex- ceeded along the part of the river immediately above the Maomillan. On che Lewes, below Rink Rapid, its normal thickness is about a foot, but above this point it joecomes much less and where last seen, at Ca- ribou Crossing, is not over half an inch thick, and only to be re- cognized when carefully looked for. In addition to these dififerences in normal thickness, however, and Local acoumu- much more striking than them, is an irregularity due to local circum- stances. Thus in hollows, and particularly when these occur at the foot of steep slopes, the material has evidently been washed together by rains occurring shortly after its deposit, and sometimes attains a thickness of as much as three feet. In correspondence with this it has been completely removed from some sloping or exposed surfaces. The *I found subsequently that Schwatka had observed this peculiar layer along the Lewes and correctly characterized it as a volcanic ash. Along Alaska's Great River p. 196. 42 GEOLOGICAL SURVEY OF CANADA. Source of the ash deposit. same local circumstances explain the varying depth in different locali- ties of the soil or ordinary sand which overlies the ash deposit. Where the ash deposit rests undisturbed upon the original surfac^j this appears very generally to be a yellowish or reddish quartzose\ sand. There are, in some cases, remains of burnt trees at the base of ' the layer, and traces of similar forest fires are found as well in the sand or soil overlying it. So far as the observations I was able to make go, the volcanic ash is thicker on the lower part of the Lewes than elsewhere, and the thickest part of the deposit on the Pelly lies nearly due east of the portion of the Lewes just referred to. The greater mass of the deposit in that direction, seems to show that it was derived from the westward, and a line drawn across the portions of the Pelly and Lewes above defined, lies between the 62nd and 63rd parallels of latitude, with a nearly east-and west bearing, so that if produced to the westward it would pass, at a distance of about 200 miles, through the mountain region near the Copper River, of Alaska, which includes Mount Wrangell. Mount Wrangell is the nearest known volcano,* and this or one of the neighbouring mountains in the same group, may not improbably have been the source of the material M-hich has been so widely spread over the Upper Yukon basin. It should be stated, however, that the Indians report the existence of a burning mountain near the head-waters of the White River, and that it is uncertain whether this report refers to Mount Wrangell or to some still unknown mountain which may be even nearer to the district here described. Fig. 1. EBLATIONS OF ASH-BED ON LEWES KIVER. a. Soil. 6. Ash-bed. u. Stratified sands and gravfls. d. Buried drift-wood, e. Stratified gravels. Date of the Respecting the date of the eruption to which the ash-bed is due, very eruption. little can be said with certainty. As already noted, the rivers have not * See Lieut. H. T. Allen's Reconnaissance in Alaska, Washington Govern- ment, 1887. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 43 certainly cut their beds perceptibly deeper since the deposit occurred on their flood-flats, so that the period to which it belongs cannot be an exceedingly remote one. It was further observed in one place, on the Lewes, to rest upon stratified sands a few feet thick, which in turn overlie a mass of drift logs still quite sound and undecayed. This fact, with the general appearance and mode of occurrence of the de- posit, leads me to believe, that while the eruption must have happened at least several hundreds of years ago, it can scarcely be supposed to have taken place more than a thousand years before the present time. Dall, in his work on Alaska, gives a list of volcanic eruptions (derived from Grewingk which have occurred in the Aleutian Islands and along the western part of the Alaskan Coast from the year 1690. f While it is quite improbable that any of these was connected with the formation here described, it is interesting to note that great quantities of ash were observed to fall on several occasions, notably in 182-5, when the whole peninsula of Alaska was covered with black ashes. That the eruption of which the occurrence is marked by the ash-bed Volume of ash of the Lewes and Pelly, was on a great scale, is sufficiently evident ''J®°''^'*- from the extent of the deposit ; which must necessarily be very much greater than the area to which the present observations refer. By drawing a line such as to include the outer limits of the observed ex- tent of the ash, a roughly triangular area of about 25,000 square miles is outlined, and if we assume the average depth of the layer over this area alone to be three inches, the mass represented would be equivalent to a prism one mile square, with a height of 6240 feet or (making allowance for interspaces in the comminuted material) equal to nearly a cubic mile of rock. It has not been considered necessary to make a complete examination Lithological of the character of this ash. In appearance it is a fine, white, sandy character, material, with a harsh feeling when rubbed between the fingers. Microscopically it is found to consist chiefly of volcanic glass, part being merely frothy and pumiceous, but of which the greater portion has been drawn out into elongated shreds, frequently resembling the sub- stance known as " Pele's hair," in which the inclosed vesicles become more or less completely tubular. In addition to this glass, fragments and small perfect examples of sanadine felspar crystals occur together with portions of minute crystals of hornblende and probably of other minerals. t Alaska and its Resources, 1870, p. 466. u GEOLOGICAL SUEVEY OF CANADA. The Stikinb Riveh. General Features. Kiver cuts across the Coast Ranges. Route to the interior. Information obtained. Since the year 1873, when the placer gold mines of Cassiar were first developed, the Stikine River has become a somewhat important avenue of communication from the coast to the interior of the northern part of British Columbia. Like the Fraser, the Skeena, the Nass and several other smaller streams, it rises to the east of the broad belt of mountains which constitutes the Coast Ranges, and cuts completely through this belt with a nearly uniform gradient. In size and general character the Stikine closely resembles the Skeena, which reaches the coast 200 miles further south. It is navigable for stern-wheel steamers of light draught and good power, to Glenora, 126 miles from Rothsay Point, at its mouth, and under favoi^rable circumstance.s to Telegraph Creek, twelve miles farther. Above Telegraph Creek is the " Great Canon " which extends for many miles and is quite impassable either for steamers or boats, though traversed by miners in winter on the • ice. The head-waters of the Stikine are unknown, but lie for the most part to the south of the 58th parallel of north latitude, in a coun- try said to be very mountainous, interlocking there with northern branches of the Nass and western feeders of the Black or Turnagain River, a tributary of the Laird. From Telegraph Creek, the head of navigation, a pack-trail sixty-two miles and a half in length, constructed by the British Columbian Government, follows the valley of the Stikine, generally at no great distance from the river, and eventually crosses from the Tanzilla or Third North Fork to the head of Dease Lake, which may be regarded as the centre of the Cassiar mining district. This route has long been known to the Indians, the Stikine having been to them from time immemorial an important avenue of trade, by which, as by the Skeena, the coast tribes penetrated a con- siderable distance inland. My personal acquaintance with the Stikine, as far as Telegraph Creek, was such only as could be made from the deck of the little steamer in which we ascended the river to that point, and merely enabled me to note the main features of the valley. This was supplemented, however, by the observations of Mr. McConnell, who remained behind for the purpose of making a micrometer survey of the river from the furthest point reached by Mr. Hunter's survey of 1877 to Telegraph Cfeek. Mr. McConnell's notes and map with specimens collected by him have been consulted in the following sketch of the river, and are drawn upon particularly in respect to its geological features. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 45 As the result of Mr. McConnell's survey, taken in conjunction with jf^p of the that of Mr. Hunter, we are now for the first time in possession of a Z'^®'^- correct map of the river to the head of navigation. The best general map of the river and route to Dease Lake previously in existence, was a sketch made by Mr. G. B. Wright and published in the report of the Minister of Mines of British Columbia in 1875. This map also includes Dease Lake and part of the Dease Eiver, and I may take this oppor- tunity of stating that much credit is due to Mr. Wright for its general accuracy, taking into consideration the circumstances under which it was made. The general trend of the Stikine valley for twenty miles from the Trend of the sea, is east-and-west, corresponding in direction to Bradfield Canal, '"'^"^ ^^ ^^' which penetrates the coast thirty miles to the south, and also to part of the northern portion of Behm Canal and Burroughs' Bay, still further south. At this distance from the coast the river bends through a quadrant of arc, and assumes a nearly due north direction, which it maintains for about sixty-six miles, beyond which the valley is con- tinued in a nearly direct north-eastward course to the vicinity of Dease Lake, but in its upper portion is occupied, not by the main river, but by the Tanzilla or Third North Fork, the main river (•uter- ing this continuous valley from the southward. The particular range of the Coast Mountains, which locally assumes The Coast a culminating or axial character on the Stikine, is that which is traversed by the river-valley near the great bend above alluded to. As seen from the sea, at some distance off shore, it is notably higher and rougher to the north of the river-valley than to the south of it, and is surmounted by sharp, jagged, rocky pinnacles in some places. The highest summits in this range here probably average about 8000 feet. It carries much snow throughout the year, and in it are the sources of the principal glaciers which debouch along the north-and-south part of the river above the great bend. The inland border of the Coast ^Mountains may be said, on the Stikine, to be near Glenora, giving a transverse width, from the coast, for this rugged belt of country, of nearly eighty miles. The current of the navigable portion of the Stikine is swift through- Current of the out, but there are no rapids properly so-called, though the Little Canon "^®''' (fifty-three miles above the great bend) forms a serious impediment to navigation when the river is at its highest stage in June or July, in consequence of the great velocity of the current in this narrow and rocky though deep gorge. Near the mouth of the river the current scarcely surpasses two miles an hour, but it increases as the river is ascended, till it attains a rate of six to seven miles in many places 4G GEOLOGICAL SURVEY. OP CANADA. between the great bend and Telegraph Creek, the swifter water being chiefly met with above the Little Canon. The average rate of flow of the navigable portion of the river must be about five miles an hour, The width of the Stikine immediately opposite Telegraph Creek was found on May 29th to be 480 feet only, but it is here deep, and had a velocity of 6-08 miles per hour, as determined from several observa- tions. A few days later it was rising fast, and the velocity was con- siderablj' greater. Navigation. Stern-wheel steamers for the navigation of the river should have good engine-power, and should draw not more than four feet of water when loaded. The height of the river above sea-level at Telegraph Creek, as deduced from simultaneous barometric observations at the mouth and at this place, is 540 feet, giving an average fall of over four feet to the mile by the course of the stream. The actual fall on the upper part of this length of the river must, however, considerably exceed this figure, while that of the lower portion is inconsiderable. Under ordinary circumstances the ascent of the river to Telegraph Creek, with a suitable steamer, occupies about three days, and it is generally neces- sary to carry a line ashore at a few places. The extensive flats near the mouth of the river render it necessary to enter it about high-tide. Mr. Hunter ascertained that the channel across these flats has from one to two feet only of water at low ebb. A consider- able proportion of the traffic is carried on by Indians with canoes, and the Stikine Indians are very expert in all the necessary operations of tracking and poling in swift water. Notes on the dates of opening and closing of the river will be found on a later page. Lower part of The entrance to the Stikine from the sea is not distinguishable in its river at one j^j^in orographic features from that of many of the salt-water inlets by which this part of the West Coast is dissected. The lower portion of this river-valley may, in fact, be regarded, like that of the correspond- ing part of the Skeena, as an inlet which has become filled with detri- tus in consequence of the great size and sediment-carrying capacity of the river which has emptied into it. Unlike the Skeena, however, the dSbris brought down has in this case been projected seaward so as to completely block the wide channel between the mainland coast and Mitkoflf Island with shallow tide-flats and bars, above which several smaller, high, rocky islands project. The mountains immediately bor- dering the valley of the Stikine at its seaward entrance are from 2000 to 3000 feet in height, and rise abruptly from the wide alluvial flats. YUKON DISTEICT AND NORTHERN BRITISH COLUMBIA. 47 through which the river there winds, often without even touching the \cfwer spurs of the hills. The flats are generally covered with fine groves of cottonwood, Character of mingled with spruce and other trees, and are often cut through cfi™'"^^**^^ by sloughs and channels, which become so numerous in some places as to render it difficult to decide which is entitled to rank as the main stream. The valley-bottom maintains an average width of from two to three miles as far up as the Little Canon, which place may be regarded as nearly marking the head of the old salt-water inlet which has been silted up by the river. The canon is about three-fifths of a mile long, and in places not more than fifty yards wide. It is bordered by massive granite cliffs, 200 to .300 feet in height, above which, on the west side, rugged mountain slopes rise. On the east, are low rocky hills representing part of a former spur of the mountain, through which the canon has been cut. A tract of low land separates these hills from the eastern side of the main valley, aud it is difficult to explain under what circumstances the river has taken its present course. For some distance above the Little Carion the Stikine valley appears Valley above to cut very obliquely through a series of somewhat irregularly parallel J-'''"^ Canon, ranges. Eight miles further up is the " Kloochman Canon," which to some extent repeats the features of the last, but it is nearly 300 feet in width and ofifers no impediment to navigation. At four miles above the "Kloochman Canon" is the so-called "Grand Rapid,'' which, in con- sequence of recent changes in the river, is now by no means formid- able, though the water is still particularly swift and the river wide and shallow. Here the valley begins very markedly to open out, the mountains retiring further from the river and decreasing in altitude, while irregular basaltic hills, of no great height, appear between the river and the bases of the mountains. This, taken in conjunction with the dry climate which characterizes the country to the east of the mountains, and the fact that most of the slopes have been bared of timber by fire, gives an entirely different aspect to the landscape. The Stikine is joined by some important tributaries in the part of Iskoot River, its course above described, though none of these have yet been examined in detail. The Iskoot or Skoot, which flows in from the eastward about thirty-five miles from the mouth, or just within the locally cul- minating range of the Coast Mountains, is known to be navigable for some distance by canoes, and one branch is said to head not far from the extremity of Portland Canal, to the southward. By following this river to its head and there making a portage, the Indians are reported to be able to reach the Nass River without difficulty. The Iskoot has been prospected by a few miners, but apparently without 48 GEOLOGICAL SURVEY OF CANADA. Scud River. Clearwater River. Fir.st South Fork. Telegraph Creek and Glenora. any notable result, though the Indians report the occurrence of coal. The northern branch of the Iskoot, to the east of the Coast Mountains, was traversed in 1867 by P. J. Leech, of the Western Union Telegraph Exploration Survey, who crossed from it to the head of the First South Fork of the Stikine. The valley is there reported to be from 2500 to 3700 feet above sea-level, generally timbered, but with some open, grassy slopes. About seven miles below the Little Cafion, the valley of the Scud River opens to the east, but the exact position of the mouth of the stream has not been fixed on the map. Some gold has been found by prospectors on this stream, but no workable placer deposits. It is said to head in a low country behind the Coast Mountains, and if this be correct, must nearly inosculate with branches of the Iskoot and First South Fork of the Stikine. Six miles above " Kloochman Canon,'' the Clearwater River enters the Stikine on the west side, by several mouths. This is a stream of considerable size, and is navigable for canoes for some distance. It is said to head near the sources of one branch of the Taku River, and is noted by the Indians on account of the great number of salmon which ascend it. The First South Fork joins the Stikine about a mile and a-half below Telegraph Creek. It is a large turbid stream, and for a number of miles from the main river, flows in a rough narrow gorge, between high hills and mountains. Further up, according to the Telegraph Exploration sketch, it is bordered by level, partly timbered terraces or " benches." The summit between its head-waters and those of the Iskoot, on the route followed by Mr. Leech, is given on his authority at 5000 feet. Salmon do not ascend this stream. Telegraph Creek is an inconsiderable stream, which falls rapidly to the river through a narrow rocky cleft in the bordering hills of the right or north-west bank of the Stikine. Its name is due to the fact that here the Western Union Telegraph line was intended to cross the Stikine. The little town of Telegraph Creek occupies the narrow delta of the stream and the lower terraces bordering it on both sides, its site being identical with that of "Fort Mumford" of the older maps. Glenora, twelve miles below Telegraph Creek and on the same side of the Stikine, consists of a single row of houses built along the edge of the river at the foot of a steep bank. Both places were at one time busy little towns, but are at present very much reduced in importance, though I believe it will probably noi; be long before further mining developments in the Cassiar district will lead to the renewal of their activity. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 49 Glaciers. The glaciers constitute one of the most remarkable features of this Previous notes on the glaciers. part of the Stikine valley. There are a number of these on both sides °" of the river, in its lower part ; but four only of special importance, all of which are situated to the west of the river, and all but the first on the eastern slopes of the most massive central ranges of the mountainous region. The only detailed previous notice of these glaciers is that given in a report by W. P. Blake.* Mr. Blake's account of the glaciers is transcribed in the Fifth Annual Report of the United States Geological Survey, where it was placed under the some- what misleading title of Glaciers of " Alaska." Two of the glaciers are illustrated in the last-mentioned volume by reproductions of photo- graphs taken under the direction of Dr. I. W. Powell, Saperintendent of Indian Affairs in British Columbia. That named the " Orlebar Glacier" represents part of the front of the Great Glacier of the miners and of Mr. Blake's map. The " Bernard Glacier " I am unable to identify with any certainty, but the illustrations evidently repre- sent part of either the Flood or Dirt Glacier. Mr. John Muir, who spent some time on the Stikine in 1879, gives an interesting popular description of its glaciers in a letter dated from Sitka in December of that year, and published in the San Francisco Bulletin. Mr. Muir informs me that no more systematic account of his observations in this region has yet been made public. The glaciers are also noticed at some length in an account of a trip on the Stikine by Mr. W. H. Bell, in Scribner's Monthly, 1879, Vol. XVII. The accompanying illustrations, though striking and artistic, have been idealized so far as to be scarcely recognizable. The glacier known as the First or Little Glacier by the miners Little Glacier. (named the Popoff Glacier by Blake) fills a high valley on the north side of the river, about ten miles from its mouth. As seen from a dis- tance it offers no features of particular interest, resembling many other minor glaciers of the Coast Mountains. The next and most important glacier is that universally Great Glacier. known on the river as the Great Glacier, and so named also by Mr. Blake, who gives an excellent description of its main features. The high snow-fields from which this glacier must take its rise are not seen from the river, the glacier entering the wide valley of the Stikine nearly at right angles, through a break in the mountains 'Geographical Notes upon Russian America and the Stikeen River. W. P. Blake, Washington, Government. 1868, A 50 GEOLOGICAL SURVEY OF CANADA. Recession of the glacier. Dirt Glacier. two to three miles distant from the river bank. Before entering the Stikine valley, the glacier has a width estimated at from on,e- half to three-quarters of a mile, but upon freeing itself from ihe borv dering mountains immediately expands in a fan-like manner, its actual front upon the river being from three to three and a-half miles in width. The slope of the surface of the glacier where it issues from the mountains was estimated — as seen at right angles — at above five degrees. Beyond this point it flattens out, and portions of the surface become extremely rugged, breaking off near the front in series of descending steps, as described by Mr. Blake. When seen by us, on the 20th of May, much of the surface was still covered by the new snow of the preceding winter, but notwithstanding this, a great quan- tity of rooky debris was visible, giving a gray tint to portions of the ice. The front of the glacier appears to be quite close to the edge of the river, but is actually about a third of a mile distant at the nearest points. This interval is occupied by moraines and marshy pools, the outer tier of moraines, or that nearest to the river, forming wooded hills about one hundred and fifty feet high. The newer moraines were partly covered and overridden by the front of the decaying ice. Large streams issue from beneath the ice, the position of outflow frequently changing from year to year. Next to its size, the most remarkable feature about this glacier is the regularity of the fan-like form in which it terminates. It resembles in this respect the Davidson Glacier on Lynn Canal. The miners state that, during the few years which they have known the Stikine, the Great Glacier has steadily and notably receded, though the total amount of such recession can evidently not have been more than the distance from the wooded bordering-moraine to the present ice-front. The Indians relate as a tradition, that at a former period the glacier stretched completely across the valley, the Stikine passing lieneath the ice through a tunnel-like opening. It is, however, im- possible to determine whether this is a remembered fact or a fancied inference. Curiously enough, a copious hot spring is situated imme- diately opposite the glacier on the east side of the Stikine valley. Ten miles above the Great Glacier, and also on the west side of the valley, is the Dirt Glacier, so named by the miners because of the great quantity of rocky ddhris with which its surface is covered.* This is much smaller than the last, having a width estimated at a quarter of a mile but possibly greater than this. Like the Great Glacier, it comes quite down on the river-flats. * Also so named on sketch map in Report on Customs District, Public Service and Resources of Alaska Territory, by W. G. Morris, 1879. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 51 The last important glacier, sixteen miles further up the river, is ^Elood Glacier. the Flood Glacier. This also comes down to the level of the river-flats, but does not closely approach the river. From the valley of this glacier a great rush of water occurs almost every year towards the end of the summer. This, no doubt, arises from the blocking by the glacier of the mouth of some lateral valley in which a lake is formed and from time to time breaks through the glacier dam. The quantity of water thus liberated is so great as to raise the river from a low stage to half-flood level for a short time. There is a large quantity of ddbris also on this glacier, though less than on the last. Geological Notes on the Stikine. The only information as to the geology of the Stikine, up to the t,, , , present time, has been that embodied in notes in Mr. Blake's report, observations. already referred to,* and these include the lower portion of the river alone, as his furthest point was a few miles above the Little CafioQ. The result of Mr. Blake's examination are by himself summed up in the following terms : — " The mountains of the Stikine valley, from the Little Canon down to near the coast, are formed of syenite and granite, with some metamorphic beds at intervals. The walls of the Little Canon are granite. At the mouth of the river, and below the Indian villages, the rocks are quite different, being formed of the great sandstone and shale formation already described. The direction of uplift of these strata is about N. 80° W. magnetic. This formatit)n is some thousands of feet thick, and resembles the rocks of San Franciscoi but is more changed by metamorphic action. They are probably of the secondary period. It appears to pass into mica-slate just above the site of an old stockade or fort of the Hudson's Bay Company, where I found a locality of garnets like those of Monroe, in Connecticut."! The rocks seen along the west shore of Wrangell Island, in the . Rooks of vicinity of the town and harbour, are chiefly black, flaggy argillites, Wrangell remarkably uniform and regular in their bedding and with a westward ^ ^ dip. They are considerably indurated and contain small staurolite crystals in some layers, while on the surface of others crystals of mica have been developed. Similar rocks are found on other parts of the coast, both to the north and south, and from a. lithological point of view, they much resemble the Triassic argillites of the Queen Charlotte Islands, though no fossils were found at this place. The ridge behind * Mr, Blake's notes on the geology of the river are also given in Petermanns Mitteilungen, vol. X, 1864. f This does not appear to be the locality near Rothsay Point, subsequently men- tioned. 4i 52 r.EOLOGICAL SURVEY OF CANADA. the town of Wrangell is chiefly composed of rather fine-f;rained gray granite, which is probably intrusive and may have been the cause of the incipient crystallization observed in the argillites. The north point of the island is formed of similar granite, probably a continua- tion of the same mass. Rocks at On crossing to the mainland, to the eastward, mica-schists and river. granites are met with, and beyond Rothsay Point (which may be regarded as the entrance to the Stikine River) granitic rocks only were seen for some miles. Near Rothsay Point, at a short distance Garnets. from the shore, is the locality from which are obtained fine claret- coloured garnet crystals, sometimes an inch or more in diameter. The matrix of these, as seen in hand specimens, is dark, highly crystalline mica-schist, but the locality was not visited. The general strike of the rocks west of Rothsay Point is about S. 35° E., with prevailing west- ward dips. Rothsay I'oiut From Rothsay Point to the Great Glacier, the mountains bordering to Great ,1. i.n 11. ■ t ..,, Glacier. the river are chieny composed or granites and granitoid rocks. These alternate with gneissic and schistose rocks of similar composition, including mica-schists ; but massive granites probably form much the greater part of the whole. The granites are usually gray in colour, and conta.in both hornblende and mica, with white felspars, which are often porphyritic, giving the rook a spotted appearance. Some varigties become granitoid diorite, while others are highly quartzose, and contain little or no hornblende. The series as a whole closely resembles that cut across by the Prater River in its lower course, and generally characteristic of the Coast Ranges of British Columbia.* Rooksbrought The Great Glacier, rising many miles back in the higher ranges gkcier.'*' * "* °^ *^^ mountains, in the material which it has brought down and deposited in its moraine, affords a mode of ascertaining the gen- eral composition of the central ranges. The material was found by Mr. McConnell to consist almost entirely of gray granite of medium grain, composed of felspar, quartz and hornblende in nearly equal pro- portions, but holding also a little mica and occasional crystals of sphene. Diorites and mica-schists occur in smaller quantity, together with coarse pegmatite, which is evidently derived from veins intersecting the granite. Klooohman Similar granitoid rocks, with occasional schistose areas, constitute Grand Rapid. *^® whole of the rock-exposures seen along the river to the so-called Grand Rapid, about four miles above " Kloochman Canon." The mountains bordering the valley also appear to be entirely, or almost * Compare Annual Report, Geol. Surv. Can., 1886, p. 11 e. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 53 entirely, composed of the same materials, though at one place, (nearly opposite the site of the Hudson's Bay Company's old post) the upper portion of a mountain seems to show a massive bedding, recalling that of the Cretaceous quartzites of Tatlayoco Lake, which occur there in a similar position relatively to the Coast Ranges.* No further evi- dence was, however, met with tending to show that rocks of this age occur here. A short distance below the " Grand Rapid," distinctly stratified Stratified rocks of dark colour are seen capping some of the mountains and rest- Rapfd' ing upon the granites. These beds have a dip of N. 70° E. < 30°, which brings them down to the level of the river near the rapid. They are there found to consist of hard argillites and grauwacke-quartzites, interbedded with shaly, gray and brownish impure limestones, the whole being considerably disturbed and cut near the granites by coarse gray prophyritic dykes of that rock. The argillites were not observed to hold staurolite, mica, or other crystalline minerals like those of Wrangell, and otherwise differ somewhat in appearance from these, though their relation to the granitic rocks appears to be similar. They are followed in ascending order by a massive gray-blue sub-crystalline limestone of considerable though undetermined thickness, which can be traced in the mountains for some distance on both sides of the valley. These limestones are believed to represent those afterwards noted on the Dease and there referred to the Carboniferous period. About two miles and a half above " Grand Rapid," near the mouth AJtered vol- of the Clearwater, the limestone is followed — apparently still in ascend- ^^'^^° ""^ ing order — by a series of altered volcanic rocks which are, for the most part, of gray and greenish colours. These are apparently chiefly dia- bases, but include also porphyrite-like rocks. The rocks are gener- ally rather fine grained, and would require microscopic examination before they can be named in detail. f Though clearly forming a stratified series, evidence of bedding can seldom be detected in the exposures, in consequence of their homogeneous composition and shat- tered state. They seem to be identical with those forming a part of the Cache Creek group, in the southern interior of British Columbia^ and though no fossils were found in the limestones previously men- tioned, they, and possibly also the argillites beneath them as well, may be referred with considerable probability to the same Upper Palaeozoic age. * Report of Progress, Geol. Surv. Can., 1875-76, p. 253. t One of these which has been microscopically examined by Mr. V. D. Adams, is described as a diabase-porphyrite (Stikine No. 16. ) 54 GEOLOGICAL SURVEY OP CANADA. limestones. Triassio fossils. Tertiary vol- canic rocks. ArgiUites and Altered volcanic rocks only, like those above noted, were seen aloi^g the river for about twelve miles above the Clearwater, but there is reason to believe that outliers of Tertiary basalt also occur in this part of the valley. At the distance just mentioned above the Clearwater, and about six miles and ,a-half below Glenora, exposures are found of slaty argillites and dark shaly rocks, containing some impure lime- stone, all very much broken and disturbed, and associated with altered volcanic materials. Some beds of these shaly limestones prove on microscopical examination to consist chiefly of organic fragments which are not, however, sufficiently distinctive for the reference of the beds. Dark shaly rocks occur near Glenora which were not specially examined. It is probably from these that some specimens containing Triassic fossils, which were given to me some years ago by Mr. J. W. McKay, were derived. The form represented is a species of Holohia probably a finely sculptured variety of H. Lommeli. From this point to Telegraph Creek, basaltic and other comparatively modern volcanic rocks become prominent features, the basalts appear- ing as remnants of horizontal flows, the broken edges of which form scarped cliffs. These rocks are due to a period antecedent to that of the glacial deposits, and are of Tertiary age. Analogy with neigh- bouring parts of British Columbia indicates that they may be assigned with probability to the Miocene. The bas.alts have evidently flowed along and partially filled the old river- valley, and uiicouformably over- lie the old altered volcanic rocks previously alluded to, as well as all the other rock series. About two miles below Glenora, the basaltic rocks were noticed in one place to have filled the old river-bed, coxifoi-ming in their lower layers to the slopes of its sides, and to have been subsequently cut across obliquely by the present river. Other examples df this charac- ter are mentioned on following pages and are of special interest in connection with the occurrence of placer deposits of gold. Between Glenora and Telegraph Creek, the rocks seen below the basalts include at least two distinct series. The first and oldest of these is represented by a number of occurrences of altered volcanic rocks, like those previously referred to, as well as by considerable exposures (beginning about a mile above Glenora) of gray and blackish, rather cherty quartzites, often nearly on edge. The second consists of slightly indurated conglomerates, sandstones and shales, the conglomerates being often very coarse and containing pebbles both of the older volcanic series and of the granites and granitoid rocks. These lie at comparatively moderate angles of inclination. No fossils were ob- served in them, but in their lithological character as well as in their Palaeozoic and Cretaceous rocks. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 55 position relatively to the Coast Ranges, they resemble rocks of Cretaceous age met with in other parts of British Columbia, both to the south and north of the Stikine, and may be provisionally referred to that period. In the immediate vicinity of Telegraph Creek, the prevalent rock is Rocks near a gray-green, speckled, altered volcanic material, which proves to be Tele^aph a fine-grained diabase-tuff.* The high hill immediately opposite Telegraph Creek, on the other side of the river, is composed of similar old volcanic rocks, comprising compact diabase and a massive diabase- agglomerate. About two miles below Telegraph Creek, on the right bank of the Basalt flows. river, a portion of the basaltic filling of the old valley forms a range of columnar cliffs about 200 feet above the present water-level. A second similar remnant occurs just above Telegraph Creek, on the same side, and a portion of it extends up Telegraph Creek itself for a mile or more. Basaltic dykes, which may have served as sources of supply of molten material at the time of eruption, are found cutting the older rocks. Though in some cases simulating the appearance of terraces, the basal- tic shelves along the sides of the valley are quite distinct from, and of earlier date than these. Notes on the various rocks met with will be found on the face of the Xotesonmap. map accompanying this report. The country to the east of the granitic rocks of the Coast Ranges would require much time and attention before its somewhat complicated geological structure could be properly defined. In the gorge of Telegraph Creek, a large boulder of gray sub-crys- Fossiliferous talline limestone was found, closely resembling in character and degree li™<*'**°"«- of alteration that seen near the " Grand Rapid," but in this case con- taining large branching corals and numerous Funulinm, indicating its Carboniferous age. It is of course impossible to state with certainty whence this boulder was derived, but it may very probably have come from the mountains to the north within the drainage-area of Telegraph Creek. The portion of the Alaskan coast which I have seen, viz., that to Terraces, the south of the 59th parallel, shows the same general absence of terrace deposits which has already been noted and commented on in the case of the British Columbian coast. In the vicinity of the mouth of the Stikine, terraces fifteen to twenty feet in height are found, resembling the wooded flats met with further up the river, but as they are here upon tide-water, indicating, doubtless, an elevation of the ^ * (Stikine No. 25.) 56 GEOLOGICAL SURVEY OF CANADA. Gold. Two distinct climatic regions. coast-line to that amount. Further up the river, the first appearance of high-level terraces is at about two miles be low the Great Glacier. Those here seen are quite narrow, and were estimate d to be 500 and 700 feet, respectively, above the river. The river, for the first time, shows bordering terraces of from thirty to fifty feet in height, about six miles below the Little Canon, and similar terraces are frequently seen above this point. On the mountain above Glenora a distinct but small terrace was seen from a distance at an estimated height of 1500 feet above the river. At Telegraph Greek the two principal terraces are 90 and 200 feet respectively above the river-level. (See Plate I.) The mode of occurrence of gold on the Stikine, and the placer mining which has occurred along the river, are described on a subse- quent page, in connection with facts on gold mining in the Cassiar region generally. Climate. The traverse of the Coast Ranges by the Stikine River, from its mouth to Telegraph Creek, affords an excellent illustration of the difference between the coast and inland climates, repeating to a great extent the phenomena met with in making a similar traverse of the same ranges in the southern part of British Columbia. It is here, however, all the more remarkable, as so great a difference between these climates would scarcely be anticipated in this northern latitude. Some records of observations in Appendix VI.* may be referred to for details, but it may be stated here, as showing the broad general contrast, that while the annual precipitation at Wrangell, at the mouth of the Stikine, is over sixty incliesf, that in the vicinity of Telegraph Creek on the inland side of tlie mountains, is so small that it is necessary to irrigate cultivated land. Nor does this comparison of rainfall sufficiently mark the great diversity which actually obtains between the two climates, the pre- valence of clouded skies in the coast region being accompanied by a saturated state of the atmosphere, while precisely opposite conditions are found on the eastern side of the mountain belt, at not more than eighty miles inland from the general line of the coast. The coast climate is, of course, much more temperate than that of the interior, which, even no further off than Telegraph Creek, becomes one of extremes. It is probable that the total annual precipitation is even * [Not reprinted.] tXJ. S. Coast Pilot, Alaska, Parti., 1883, p. 271. The precipitation at Wrangell is moreover much less than that at more exposed parts of the Coast, for at Tongass and elsewhere it exceeds 100 inches annually. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 57 greater in the vicinity of the culminating and central ranges of the Bearin" of Coast Mountains than at Wrangell, and as a large proportion of this snowfall on 6XlStj6IIlC6 01 occurs as snow, it suflBciently accounts for the existence of the import- glaciers, ant glaciers and the heavily snow-covered appearance of the mountains till late in the summer. Miners state that the snow accumulates on the river-flats of the lower part of the Stikine, within the mountains, to a depth of from eight to ten feet, while at Telegraph Creek and on the Tahl-tan River it seldom exceeds eighteen inches, and at the latter places horses and mules have been wintering out for a number of years. The great depth of snow retard the advance of spring all along the portion of the river where it occurs, and thus by a cumula- tive effect conserves the already large quantity of snow for the supply of the glaciers, which are consequently due rather to the extremely heavy snowfall than to the actual latitude of the region. When we left the coast, on the 19th of May, the hills near the sea Observations were generally denuded of snow to a height of several hundred feet, ^'" tentrionalis, amongst other plants, were in flower, and butterflies and humming-birds were abundant. The change in species of plants met with in ascending the river is also clearly indicative of that from a very moist to a dry climate, as a reference to the lists in an appendix to this report will show. The devil's club (Fatsia horrida) extended only a few miles above " Kloochman Canon," while Elceagnus argentea and other forms char- acteristic of a dry region was first seen at Telegraph Creek. The 58 GEOLOGICAL SURVEY OK CANADA. Local climatic differences. Notes at Telegraph Greek. state of progress of the season at this place appeared to be nearly, if not quite, equal to that found at a similar date in the vicinity of Ottawa or Montreal. The local differences of climate are, however, quite important. Thus Glenora, though about twelve miles only from Telegraph Creek, is said to experience much greater cold in winter, and the snow fall is also greater, beine estimated at three feet and a-half. Less snow falls on the Tahl-tan than elsewhere, the amount increasing both to the east and west of that place. Strong winds blowing up stream or inland are prevalent in the Stikine valley in summer, but occur in the reverse direction, as a rule, in winter. Further observations on the winter climate of the Stikine are given in Appendix VI.* During the few days spent at Telegraph Creek, in the latter part of May, the wind generally blew up the river and was often strong. The high distant ranges of the Coast Mountains to the west, were usually enveloped in clouds, and heavy showers were there evidently of constant occurrence. The sky at Telegraph Creek was also as a rule largely obscured, but after passing over the Coa << g !> be a < &D H e hi] '& a oya- jg about six miles. For about half this distance, to Ward's house, (now, like other places of call along this route abandoned) the trail runs near the Stikine River, the immediate valley of which still con- tinues to be occupied by basaltic flows. Above these, however, the sides of the valley are generally formed of very regular :ind high terraces, composed of horizontally stratified sands, gravels and earthy deposits, which though generally very fine, are rather silts than true clays. The gravels frequently include large boulders. At Ward's, the trail turns away from the river and cuts across a high point to the Tooya, the highest terrace-level crossed being about 1000 feet above the river. On these terraces the vegetation was perceptibly less ad- vanced than in the lower parts of the valley. Swampy spots are fre- quent, and the country, as we recede from the vicinity of the Coast Mountains, has evidently a somewhat more humid climate and is more subject to summer frosts. Potatoes and other crops are successfully grown at Ward's, situated on one of the lower terraces, but irrigation is there necessary. Tooya valley. The Tooya valley, where it is crossed by the trail, is a great gorge, about 600 feet in depth, cut out through the terrace deposits. The river, which is spanned by a small bridge, is a wild torrent — almost a series of cascades. Its scarped banks show a section of about 400 feet of the terrace deposits, which are of the character above noted, but include rough, bouldery and gravelly layers, and a number of large granite boulders occur in the bottom of the valley, resembling in their lithological character the granites of the Coast Ranges. Sources of Scarcely any authentic information is available regarding the head- iooya. waters of the Tooya, though these have been reached by prospectors from Dease Lake. A lake of considerable size is reported to exist on its upper part, as indicated in the accompanying map, and the volume of water in the river is such as to lead to the belief that it must drain a large area to the south of the Yukon watershed. Plateau About a mile beyond the Tooya, on the summit of a wide, undulatinar beyond Tooya. . „,.., f , ^^ . , , , terrace, is Wilsons house. Here turnips and potatoes have been grown, but the potatoes do not fully mature. From Wilson's to Cari- YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 65 bou Camp, about twelve miles, the trail crosses an extensive high ter- race qr plateau, with a nearly level or slightly undulating surface, which is generally wooded with aspen, black pine and white spruce of fair growth. A few very small streams, which flow toward the main valley, are crossed, but the river is generally some miles distant and scarcely visible from the trail. The Tooya valley is here snid to run nearly parallel with the main valley of the Stikine and at no great Head of distance from it, but is invisible from the trail. No mountains were here seen to the north-eastward, but high, rounded mountains, with broad, bare summits, continue to border the south-east side of the Stikine valley. About mid way betw een Wilson's and Caribou Camp, the Stikine, or Too-dessa of the Tahl-tan Indians, coming from the southward, enters the main valley, cutting through the bordering mountains in a narrow canon, which the Indians report impassable. Their route to the upper waters of the river crosses the mountains to the west of this canon. They state that after again reaching the Stikine, above the canon, they can ascend it in canoes without difiiculty for a long di--tance. No rock exposures were seen along this part of the trail, and only occasional groups of boulders. The soil appears to be excellent, but the altitude is probably too great for the successful cultivation of any but the most hardy crops. The vegetation and appearance of the country afford evidence that the climate is still a rather dry one. The trail reaches the edge of the valley of the Tanzilla, or Third Tanzilla. North Fork, about a mile south-westward from Caribou Camp. This valley is cut out to a depth of 450 feet below the level ef the plateau, and is about a mile in width from rim to rim. The sides show evi- dence of extensive landslips, both old and recent. The river is a com- paratively small though swift and muddy stream, with an estimated width of 180 feet and depth of about three feet. No rocks are exposed Silt and clay in the valley, the entire depth of which appears to be excavated in bedded clays and silts, which weather to gray, earthy slopes. No true boulder-clay was seen, but the occasional presence of large granitic boulders, with the singularly contorted character of some of the clay- beds, appear to indicate the existence of floating ice in the body of water in which the deposits were originally laid down. The clays and silts are evidently the same with those seen in the Tooya, but here, so far as observed, want the rough bouldery and gravelly beds which are there intercalated, and which are doubtless connected with the entry of the former representative of the Tooya into the lake in which the silty deposits have been formed. Flat or lenticular calcareous nodules are abundant at certain horizons in both places, and are also reported 5 66 GEOLOGICAL SURVEY OP CANADA. Tanzilla valley. Valley connecting Tanzilla with Dease Lake. Origin of the valley. to occur at some distance up the Tahl-tan, where the same silty deposit is probably continued. The level country based on these deposits. has a width of several miles on both sides of this part of the Tanzilla. From Caribou Camp to the vicinity of Dease Lake, or for about twenty-six miles, the trail runs along the north-west side of the Tan- zilla. The valley of the stream gradually loses its depth, owing to the fact that, while the grade of the stream is considerable, the ter- races at its sides continue at about the same level. These consist, so far as can be seen, of similar silty and clayey materials, but the edges of the terraces are less marked, and they show a tendency to merge into slopes, which rest upon the bases of the mountains bordering the valley. The mountains that extend to the south-east of the river here become higher than before, and take the form of a well-marked range, which is known to the Indians as Ilo-tai' -luh* Swampy spots l)ecome frequent and the vegetation more alpine in character, with evi- dence of a considerably greater rainfall. The white spruce is relatively more abundant, and Betula glandulosa and Ledum latifolium were her'e met with for the first time. A great part of the forest all along this portion of the valley has been destroyed by fire. Eock is seen in place only on approaching the bases of the mountains. Opposite the head of Dease Lake, the Tanzilla turns off abruptly, and is seen to take its rise in a high range of mountains, holding much snow and running in a north-east and south- west bearing, at a'distance of ses'en or eight miles. The main valley, which has heretofore been (icoupied by the river, turns northward, through a right angle, and becomes continuous with that of Dease Lake. The distance from the head of the lake to the Tanzilla, at the nearest point, is about three miles, the level of the Tanzilla being somewhat lower than that of the lake. The height of land is about seventy feet above the lake or 2730 feet above the sea, and constitutes the watershed between the Arctic and Pacific slopes. The part of the valley which connects the Tanzilla with Dease Lake is floored by terrace deposits, and is without doubt very deeply filled with such materia], as no solid rock is seen in it. It has evidently been part of a through river-course of very ancient date, but in which direc- tion the stream which originated the valley flowed, it is now difficult to surmise. It has, however, been again occupied by a river in compara- tively recent post-glacial times, subsequently to the formation of the terrace deposits, as it is traversed by a well-marked river-bed, filled with rolled stones and gravel. This old channel appears to rise * The name.s of geograjihioal features which have not been previously recognized, whether Indian names or names applied by himself, are throughout the descriptive portion of this report printed in italics, on the occasion of their first occurrence. YUKON DISTRICT AND NORTHEKN BRITISH COLUMBIA. 67 slightly towards Dease Lake, and there can be little doubt that the stream by which it was formed flowed out of the lake. Geological Notes on the Cassiar trail. Respecting the older rocks which characterize the greater part of Rocks chiefly the country between Telegraph Creek and Dease Lake, few details were noted, and no approach to a general section was obtained, as they are not usually exposed except along the bases of the mountains, which are, as a rule, at some distance from the route of travel. They may be described as consisting of gray and greenish-gray quartzites and grauwackes, with a large proportion of altered volcanic materials, generally felspatbic, but passing into diabases and becoming in some cases more or less schistose. Rocks originally of volcanic origin notably preponderate in the vicinity of Telegraph Creek, while near Dease Lake they are less abundant, and at about two miles from the lake, on the trail, massive gray fine-grained limestone occurs, in Limestone, exposures which are nearly continuous for about a mile. None of the mountains in sight on either side of the valley are distinctly granitic, and rocks of this character were observed only in one locality, where they occupy a relatively small area. At about two miles along the trail to the south-west of the Tahl-tan, Homblende- a dark, blackish-green, highly crystalline hornblende-rock occurs in Lanite. considerable mass, and is much broken and shattered by a gray por- phyritic and hornblendic granite, which appears to be of later date, and which may have a width of about two miles on the trail. In the bed of the Tooya River rocks differing in appearance from any seen elsewhere on this trail were found. They are reddish and purplish in colour, fine-grained and in some beds slightly porphyritic, and Purple felsites appear to be chiefly felspathic in composition.* One of these is identical with a rock met with in the lower part of the bedded series, a short distance above "Grand Rapid," on the Stikine. No fossils were found in the limestones above alluded to, and the rocks, as a whole, can at present only be classed as Palaeozoic, though showing many points in common with those of the Cache Creek group of southern British Columbia, which is believed to be, in great part at least, of Carboni- ferous age. The pre-glacial age of the basaltic rocks is shown, as already noted. Basalts filling by their relation to the terraces of the valley, and also by the occur- valley!^'^' rence upon them of large granitic boulders, the transport of which * (Cassiar Trail No. 4.) 68 GEOLOGICAL SURVEY OF CANADA. Basalts near Tahl-tan. Basalts of several dates. must be attributed to glacial action. This is seen particularly in some places between Telegraph Creek and the Tahl-tan. The basaltic rocks, at the period of their eruption, have filled the old river-valley, and may very probably have at one time done so continuously from below Glenora to the Tooya, or perhaps considerably further. There is no reason to suppose that the basalts were erupted from a single volcanic centre, and indeed the existence of basaltic dykes cutting the older rocks at Telegraph Creek would appear to lead to an opposite conclu- sion. Subsequent to the period of basaltic eruption, the river, still flowing in the same great valley, has cut down through the basalts in several places, exposing sections of the gravel deposits of the ancient river. The new channel thus formed is not, however, coincident with the old, but cuts across it at several points, and above Telegraph Creek, the excavation of the new bed has been carried to a depth estimated at from forty to seventy feet below the earlier one. A few miles below Glenora, where the basalt filling of the old valley has been cut across, it seems, however, that the old river-bed is below the present water-level, indicating, in connection with the previous observation, that the grade of the original river was greater than that of the present. Directly opposite the mouth of Tahl-tan River, on the left bank of the Stikine, a good section of the old river-bed is exposed in the truncated end of a point which forms a spur of the plateau to the south, the basalts filling it like a great ingot, and resting, at the bottom, on the old gravels; at the sides, directly on the rocky banks of the old channel. The angle between the Tahl-tan and the Stikine, on the upper side, has already been referred to in connection with the peculiarly disturbed character of the basah layer by which it is capped. Beneath the basalt at this place is a great thickness (apparently not much less than one hundred feet) of well-rounded gravel and boulders. It is probable that this deposit does not reach to the water-level, but , its disinte- grated material has formed a slope which conceals any basis of old rocks which may be beneath it. The eruption of basalt has, moreover, not been confined to a single period, but must have occurred at several difierent times separated by rather wide intervals. The occurence in some places of three or more superposed flows, shows this to have been the case, but a still more striking proof of the same fact is found in a section, observed from a distance, on the left bank of the Stikine above the Tahl-tan. At this place a thick and apparently extensive deposit of gravels has been covered by three superposed basaltic flows. Through these, a narrow vertical-sided canon has been cut by some tributary stream, which has even excavated a portion of the gravels beneath YUKO,\ DlSriUCT AND NOUTIIKUX BKITISH COLUMBIA. 69 FIG. 2.— SECTION SHOWINS OLD EIVEE-OHANNEL CAPPED BT BASALTS. EAST BANK STIKINE ElVEE BELOW TAHL-TAH. FIG. 3. — SECTION SHOWING OLD KIVEK-OHANNBL FILLED WITH BASALT. MOUTH OF TAHLtTAN. !1 i_\ I I '''K.;^a|!i'iI'i.'Ii..I.ii l! ■ lil.O^^iMlJllI'l'llMjliiLLLiL ' "' .i:<^t.^'^'i|iiM II"; frfm; 1 1 , 1 1 , ' 1 1 )i > :i^^r^^r^^^>m ^z^ u. FIG. 4. — SECTION SHOWING RELATIONS OP BASALTS AND 6KAVBLS. STIKINE EIVER, EAST BANK, ABOVE TAHL-TAX. a. Old basal rocks. b. Old gravels. c. Superposed basalt flows. d. Basaltic filling of a later gorge. 70 GEOLOGICAL SUEVEY OF CANADA. Old channel and gold pla- cers. the lowest basalt. A fourth basaltic flow has then occurred, which has completely filled the caflon and partly overflowed on the surface of the higliest of the three earlier basaltic layers. Though the basalts of Tertiary age actually seen by me are confined to the Stikine valley, it is highly probable that further explorations will prove their occurrence in other valleys, and possibly also the existence of similar rocks, in the form of plateaux of some size, in the region east of the Coast Ranges. The basaltic formation of this part of the Stikine has been described in some detail, on account of the importance which it possesses in respect to the distribution of gold. The gold along the Stikine was said by the miners to be "spotted," or irregular, in its occurrence, but the greater part of the "heavy " gold was found just along that portion of the stream now characterized by the basalts, and it appears even possible to trace a connection between the richer bars which have been worked and those places in which the present river has cut through or followed the old basalt-protected channel. This being the case, it seems very desirable that the old channel should bo fully prospected, which I cannot learn has ever been attempted. If gold should be found in it in paying quantity, it might easily be worked, and would give rise to a considerable renewal of ' activity in mining. It is not known to what extent similar conditions may occur up the Tahl-tan valley, where also remunerative bars were worked some years ago. Superficial Deposits and Terraces. Boulder-clay not observed. Thick silty and clayey deposits. No true boulder-clay was recognized either on the Lower Stikine or in the country between Telegraph Creek and Dease Lake ; neither were any striated rock surfaces actually seen. The last-mentioned fact is to be attributed to the want of suitable localities for obsei'vation in ths vicinity of the route followed and to the necessarily cursory character of the examination, as evidence of glaciation of a general character, shown in the rounding of rocky hillocks and the transport of large boulders, is abundant. The most characteristic later forma, tion of the country between Telegraph Creek and Dease Lake, is the silty and clayey deposit which has already been referred to in several places. The whole of the great valley has evidently in later-glacial times been filled with this deposit, which must have been laid down in a comparatively tranquil lake-like body of water, into which coarser material was in some places washed by entering torrents, as in the case of the Tooya. It appears to me possible that this body of water TCKON DISTRICT AND NORTHERST BRITISH COLUMBIA. 71 was held in by means of glacier-ice accumulated on the Coast Ranges on one side and those of the Cassiar Mountains on the other, and the increased height of the terraces in the vicinity of Dease Lake, as com- pared with those near Telegraph Creek, may show that the terrace- deposits have been laid down near the front of a retreating glacier- mass, the water-level of the lake being reduced pari passu, with its recession. The highest terrace-level observed near the Tahl-tan, is at an approximate elevation of 1700 feet above the sea, while half way Heights of between the Tooya and Dease Lake the terraces run up to a height of about 2800 feet. At the head of the lake a well-marked terrace-edge was obser\ed at 520 feet above the lake, or 3180 feet above the sea. The irregular surface of the same terrace sloped upward to a further height of about 100 feet, and granite boulders were found on the sum- mit of a hmestone hill 1000 feet above the lake, or 3660 feet above the sea. If the supposition of the considerable inland extension of the glaciers of the Coast Mountains at one epoch of the glacial period be correct (and it is strictly paralleled by similar circumstances in the more southern part of British Columbia), the greater part of the granitic erratics met with may probably have been derived from the Coast Ranges, though the Cassiar Mountains, and possibly other ranges in the region, are characterized by similar rocks. Route for Wagon-road or Railway. Dease Lake is the central point of the Cassiar district, and thou^, Districtnatur- as shown by statistics subsequently quoted, the yield of gold has access. greatly fallen off since the palmy days of its first discovery, it is very probable that further placer mines of value may yet be found in this region, (oi which a great part still remains to be carefully prospected) and there is every reason to believe that quartz mining and other industries will before long be developed on a considerable scale. Even at the present moment this district is more easily accessible than that of Cariboo, and when a wagon-road shall have been built from the head of navigation on the Stikine to Dease Lake, it should be easy to lay down goods at the latter point at very reasonable rates. The construction of a wagon-road, with moderately favourable Route for , wagon road, rrades, between Telegraph Creek and Dease Lake, would not be very difficult or expensive. The first ascent from Telegraph Creek is steep, but might easily be overcome. Between eight and ten miles from Telegraph Creek, or for a distance of about two miles, the road would have to follow a rough hill-side above the canon, where some blasting and grading would be required. The descent to the Tahl-tan would entail some heavy side-hill cutting in rock and earth and a bridge 72 GEOLOGICAL. SURVEY OP CANADA. would be necessary. The ascent and crossing of the '-lava bed" would entail about a mile of rough work on the opposite side of the Tahl-tan, and should the line of the present trail be followed, a long and steep ascent, with grading in gravel and clay, would be required at Ward's, and again in descending to and ascending from the Tooya valley, but no rock-work would be necessary. It seems quite probable, however, that a better route might be found for a road, at a lower level, from Ward's to the mouth of the Tooya, in following along the side of the main valley. In either case a good bridge would be required at the Tooya. Beyond this, all the way to Dease Lake, no further serious obstacle is met with. Portions of the route are clayey and swampy, and to render these easily passable, from eight to ten miles of corduroy in all would be required, for which suitable material could be obtained near by in all oases. Railway route Should the construction of a railway be contemplated, the difficul- ties to be surmounted would be greater in proportion, particularly between Telegraph Creek and the Tahl-tan, where the line would have to follow the side of the caiion, which is very rough and rocky. Beyond this point, so far as the valley could be seen from the trail, it presents no very serious impediments. Below Telegraph Creek, to Glenora, or a, little further, a railway would involve some moderately heavy side- hill work ; but further down the Stikine, to the sea, it might follow the river-flats at a nearly uniform level. The greatest difficulty to be apprehended on this part of the line would be that likely to arise in winter from the very heavy snow-fall on that part of the river below the Little Canon. It may be pointed out, in this connection, that the survey of the Stikine and of the valley leading by the Tanzilla to Dease Lake, shows that the route is an exceedingly direct one to Dease Lake, and that, taken in conjunction with the valleys of the Dease and Liard rivers, it affiards almost an air-line from the Pacific Coast to the great Mackenzie River. (See p. 19 B.) Freight rates. The present [1887] rates for goods, from Wrangell to Dease Lake are about as follows : — Wrangell to Telegraph Creek by steamer, 2| cents per lb. Thence to Dease Lake by pack animals, 6 cents. Thence by lake to Laketon, f to 1 cent. Total, about 9^ cents per lb., or $195 per ton. The result of such high prices is to discourage pros- pecting in the district and seriously to retard its further development. Dease Lake. Proceedings We reached the head of Dease Lake on June 5th, and eventually atDeaseLake. j^fj. ^j^g ^^^^ ^^ ^-^^ morning of June 19th, spending thus thirteen days YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 73 in all upon the lake. At the date of our arrival, the lake, with the exception of a small area at its head, was still covered with the decayed but unbroken ice of the previous winter, and this did not finally break up and disappear till the 16th. Meanwhile, almost all our time and a,ttention was devoted to sawing out boards and building three boats. It would have been impossible to have left sooner, even if we had had boats ready at the outlet of the lake, which had long been open, in consequence of the entire exhaustion of supplies in the district, from which it was necessary that we should depart provisioned for the greater part of the summer's work. As it was, our boats were finished a few hours before the final disruption of the ice, which occurred in the end with extraordinary rapidity, under the influence of a strong wind. It will easily be understood that we had but little time or opportunity for the examination of the surrounding country, which is nevertheless of considerable interest on account of the rich gold-pro- ducing character of some of the streams. The disappearance of the ice must always be late in this lake, in Opening and consequence of its high altitude, the want of any large entering " "^'"^ ° streams and its contracted outlet. It was, however, in 1887, later than ever before known since mining operations began. The following dates, obtained from Mr. Robert Reid, of Laketon, are those of the opening and closing of the lake for the past few years ; — Year. lake opened. Lake closed, 1882 June 9 December 5 or 6 1883 May 30 December 5 (Clear from end to end.) 1884 June 2 . December 2 1885 June 3 December 1 (Frozen completely across.) 188G June 5 December 16 (Crossing on 17th.) 1887 June 16 Further particulars respecting the climate of Dease Lake will be found in Appendix VI. [Not reprinted.] Dease Lake has an elevation of 2660 feet above the sea, and lies Dease LaKe. nearly due north-and-south on the 130th meridian. It has a total length of twenty-four and a third miles, with an average width of rather less than one mile, being somewhat narrower at the northern than at the southern end. Dease Creek, on the delta of which is situated Laketon, the chief place of the Cassiar district, enters on the west side at sixteen miles and three-quarters from the head of the lake, and is the largest tributary stream. It is also the most important, as beint' that on which the richest of the gold deposits were discovered, 74 GEOLOGICAL SURVEY OP CANADA. Country sur- rounding the lake. Mountains. Mining camps and on which gold is still worked to a limited extent. A certain amount of business is still carried on here, and it is the head-quarters of the present Gold Commissioner, Mr. Crimp. At the south end, or head of the lake, there are a few buildings, now virtually abandoned, and at Porter's Landing, on the west side of the lake near its north end, goods are landed for Thibert's Creek. The old Hudson's Bay post was situated about two miles from the lower end of the lake, on the east side. A small steamer was put upon the lake when the mines were in a flourishing condition, and is still employed in making occasional trips up or down the lake with supplies. The country about the lake is everywhere wooded, though trees large enough for lumber are found only in sheltered valleys or on low land. It is not roughly mountainous, though several prominent summits occur. The most conspicuous of these lies four miles back from the lake, about half-way between the head of the lake and Laketon. As I could not ascertain that this is recognized by any name, I propose naming it McLeod Mountain, in honour of the discover of the lake. Its height is about 6300 feet. Between McLeod Mountain and Dease Creek is Mount Sullivan* not so elevated as the last, but nearer the lake-shore and very conspicuous from it. On the opposite, or east side of the lake, a couple of miles back from the shore, is a group of rounded and wooded mountains, somewhat exceeding 1000 feet in height above the lake, or about 3800 above the sea. Beady Mountain, another notable landmark, is also on the east side of the lake, about three miles from its lower end and near the creek of the same name. Its height was not determined, but is less than that of McLeod j\Ioun- tain. With the exception of -these and some other nameless moun- tains, the country near the lake is merely hilly, or rises in long, light slopes from the shores to undulating wooded uplands, a few hundred feet only above it, which coalesce with the bases of the mountains. Only near the northern end of the lake do the mountains begin to crowd down more closely to the water's edge. The lake is shallow and marshy at both ends, but is elsewhere evidently very deep, though no soundings have been made in it. Rook-exposures are infrequent along the margin of the lake, which, when scarped, generally shows only stratified,, sandy, clayey and gravelly terrace-deposits, like those seen on the trail to the south- eastward. None of these were recognized as true boulder-clay. The lake is probably held in at its northern end by the accumulated delta deposit of Thibert Creek. f It is much constricted, further up, by the * So named for Mr. J. H. Sullivan, first Gold Commissioner of the district, lost in the wreck of the Pacific in 1875. t A stream about fifty feet wide. Superficial deposits. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 75 similar deposit of Dease Creek, and has narrowly escaped being divided at this place into two lakes. It is rimmed round at its head by somewhat irregular terrace deposits, which have already been alluded to as filling the ancient valley which communicates with that of the Stikine. ' The vegetation gives evidence of a greater rainfall and conditions more alpine and less favourable than those met with on the trail to the south-eastward, and sharply contrasting with that of Telegraph Creek Vegetation and the Tahl-tan. The effect of the ice upon the lake in spring, in tion. retarding the vegetation in its immediate vicinity, was extremely apparent. Agriculture can scarcely be regarded as practicable in this region, and the results of gardening, however carefully conducted, are small. Potatoes can be grown, but in some years they are much injured by frost, and carrots, lettuce, cabbage, cauliflowers and turnips may be made to afford a fair return. Such rock-exposures as could be reached near the shores of the lake Kooks. were inspected, and the material brought down from the hills by ser- eral streams was examined, the evidence afforded being in favour of the belief that the whole country is underlain by Palaeozoic strata resem- ' bling those described to the south-eastward. In addition to the lime- stone already noted as occurring at the head of the lake, there are gray and greenish rocks, representing altered materials of volcanic origin, associated with leek -green serpentine, in which] some minute veins of chrysotile or asbestos were noted. Besides these, and probably pre- dominant as a whole in the valley of the lake, are argillite-schists, which vary from a black plumbaginous, to a gray, finely micaceous, character, and are often lustrous and not unfrequently highly calcareous. The rocks, as a whole, closely resemble those of parts of the gold bearing series of Cariboo district. Dease Creek is said to be about twelve miles in length and to rise Bease Creek. in a lake about five miles long. It has cut a deep, narrow \/"Shaped valley through a series of terraces, which have evidently been formed at its mouth when the lake stood at various levels higher than the present. The ancient pre-glacial valley has, at the same later-glacial period, been filled with clayey and gravelly deposits, among which large and often glaciated boulders are common. These deposits fre- quently resemble boulder-clay, and are possibly entitled to be so called. The present valley has been cut down through them, and often to a considerable depth into the rock beneath them. The mining has occurred chiefly in the bed of the stream, along the surface of the solid rock, in the sides of the valley, and in various places in the gravel deposits which still remain ; also at the head of the flat on which Lake- 76 GEOLOGICAL SURVEY OP CANADA. ton stands, where the stream issues from the narrow recent valley. Much quartz occurs in the wash of the stream, and the gold, being "coarse," is evidently of local origin and has been liberated by the disintegration of the rocks in the immediate vicinity of, if not entirely %7ithin, the actual drainage-area of the stream. Gold Mining in Oassiar District. Yield of gold. The following table, based on the reports of the Minister of Mines of British Columbia, clearly illustrates the sulden rise and gradual decadence of the gold yield of Cassiar district. — Estiriiated value of Gold produced by Cassiar District, from 1S74 to 1SS7. 1873 Not known. 1874 S1,000,000 1.S7.5 830,000 1870 550,474 1877 499,830 1878 519,720 1879 405,200 1880 297,850 1881 198,900 1882 382,800 _ 1883 119,000 ' 1884 101,600 1885 50,600 1886 63,610 1887 60,485 Total s-4,8S0,069 No estimate has been formed for the yield of the mines in the first year of their operation (1873), but as that for the following year appears probably to be overstated, it may, for the purpose of arriving at a general estimate of the whole, be assumed that the sum of one million includes both years. The value of the gold may be stated as from $16 to $17 per ounce, though that of Dease Creek is usually priced at about $15.50 only. In the Report of Progress of the Geological Survey for 1886-87, I Sources of in- was enabled to give a general note on the various creeks worked for mining. gold in Cassiar and on the Stikine.* The information there given was chiefly furnished by Mr. G. B. Wright. I am now able to add to this, particulars as to the actual condition of the workings in 1887. These were largely obtained through the kindness of Mr. J. S. Crimp, the present Gold Commissioner for Cassiar district, though facts were also * Op. cit., pp. 138140. ' YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. / / gatjiered from several old miners who were among the first to enter the country. As explained on a previous page, my opportunities of personally investigating the Cassiar district were restricted by the necessity we were under of pushing on to our main field of explora- tion. Chiefly from the sources above mentioned the following sum- mary account of the different localities is derived. Summary of Facts relating to Gold on the Stikine and various Creek:) in the Cassiar District (1887). Stikine River. — Gold discovered, 1861. Very fine gold can be found Localities of on almost all parts of the river, but very little profitable work was ^ ever done below the mouth of the Clearwater. The rich ground may be said to have begun about nine miles below Glenora, and io have extended thence to the Grand Canon, above Telegraph Creek. Here Sheck's or Shake's Bar, and Carpenter's, Fiddler's and Buck':; Bars were situated, the richest being between Glenora and Telegraph Creek, though gold was also worked in a few places in the Grant] Canon. With the exception of a few spots in the lower part of thi' canon, below the Tahl-tan, and one nearly opposite Wilson's, all the cold was very fine. Coarse gold was also found on the lower part of the Tahl-tan, which proved quite profitable, and bars were worked foi' a distance of ten or fifteen miles up the river. Pellets supposed to bi; of silver but probably of arquerite or silver-amalgam were also found on the Tahl-tan. The bars on the Stikine at first averaged S3 to $10 a day to the hand, and as much as two to three ounces was sometimes; obtained, but not more than $1 to $3 can now be got, and work has practically ceased. It is stated that none of the higher benches so far prospected will pay for hydraulic work, but it is doubtful whether these have been 'examined with sufiicient care, as the area of such benches is very considerable. Dease Creek. — The bed of this creek has been gone over several times, and is now nearly worked out. It formerly yielded $8 to $50 a day to the hand, and paid well from the head of the flat, at its mouth, for six miles up. Above this a few isolated good claims were found, particularly the Cariboo Company's claim, eight miles up, from which much heavy gold was obtained. This claim has been worked over four times. The best remaining claims are bench-claims on the south side of the creek, some of these being upon an old high channel which yields well in places. Some hydraulic work on a small scale is being carried on. In 1886 there were sixteen whites and thirty -five Chinese at work, and the total amount produced was about $15,000. The 78 GEOLOGICAL SURVEY OF CANADA. Localities of gold mining, continued. gold is generally well water- worn and somewhat mixed in character, varying in value from $15.50 to $16 per ounce. Thibert Greek. — The bed of this stream is also worked out. It paid for about six miles up from the mouth, yielding at about the same rate as the last. Bench claims are now being worked, two by the hydraulic method, the rest by tunnelling. An old high channel had also been found on the south side of this creek, upon which two claims are being worked, one paying very well. Yield in 1S86, nearly the same with Dease Creek, about twenty-two whites and twenty-five Chinese being employed. Gold valued at $16 per ounce. On a tributary named Mosquito Creek, very good prospects have lately been obtained, — as much as $40 to a six-foot set of timbers. "Work is now going on here. Defot Creek. — A tributary of Canon Creek, on the same (west) side of Dease River with the last. It. rises on a plateau high above the river, where great numbei's of quartz reefs occur, and the gold found is quite rough and full of quartz. Large nuggets have been obtained, including one of fourteen ounces in weight. Some work is still in progress, though the creek-bed is worked out. Gold worth $17 per ounce. Canon Creek. — No paying deposits found. Cottonwood Creek. — This large stream heads in the same mountains with the last, but no paying deposits have been found upon it. Beady Creek. — A little mining was done here in 1874 and 1875, but nothing of importance ever found. Eagle River. — No mining ever developed. McDame Creek. — Discovered 1874. The highest average daily yield varied from $6 to $100 to the hand when mining was at its best. Most of the gold was obtained in what appeared to be an old high-level channel, which crossed points of terraces or benches on both sides of the present stream. A very small proportion of the yield was from the gtream-bed. Four or five whites and forty Chinese are now at work here, the greater number of the Chinese being employed on wide flats, which occur about nine miles up the creek. Bench claims run for about seven miles up the creek or to Holloway's Bar. Gold worth from $17.75 to $18 per ounce. Snow Creek, a tributary of the last. — The richest claim found in Cas- siar was near the mouth of this creek, yielding for a week 300 ounces for six to eight men. Only two men now at work. Quartz Creek, a branch of Trout Creek, which is also a tributary of McDame Creek. Good claims were worked here, yielding rough gold full of quartz. Much quartz in the vicinity. Two miners now at work. YUKON DISTRICT AND NOETHEKN BRITISH COLUMBIA. 79 Rosella Patterson and Dennis Creeks. — Yielded moderate amounts of Localities of . *^ 1 -k-r 1 T 1 Sol"^ mining, gold, paying " wages, say, at f o a day. Now abandoned. continued. The remaining creeks mentioned in the report cited, viz.. Gold Greek, Slate Creek, Somer's Greek or First North Fork of McDame, Third North Fork of McDame, Spring Greek and Fall Greek, are now abandoned, though several of them yielded a considerable amount of gold at one time. Sayyea Creek. — Near the head-waters of the Upper Liard, yielded excellent prospects, but has never been properly examined. The gold obtained was found in the benches, and some of it was very coarse . The creek yielded at the rate of $10.90 a day to the hand for a short time, to three miners who discovered it. Walker Creek. — Said to be distant about seventy miles in an easterly direction from the mouth of McDame Creek. Some work has been done here, but no great quantity of gold obtained. Black, Turnagain or ^' Muddy" River. — Reached by trail running easterly from a point opposite the mouth of McDame Creek, and said to be ninety miles distant. Fine gold stated to have been obtained to the value of .520 per day to the hand, and it is generally believed that coarse gold may occur on its head-waters. In 1874 prospectors found streams about seventy miles south-east of Dease Lake, which are supposed to be tributaries of this river, and yielded $6 a day in coarse gold, but at the time this was considered too poor to work. Considerable difficulties were experienced in mining operations in Frozen ground some parts of the Cassiar district on account of frozen ground, often met with below the wooded and mossy surface. It is on record that on Dease Creek, the ground continued to be frozen to the end of a tunnel driven in one hundred and fifty feet from the slope of the hill, and at a depth of forty feet from the surface. After the woods and moss had, however, been burnt off, little further complaint was heard of frozen ground. Very little has yet been done in the way of prospecting for metal- Metalliferous liferous veins in this district, but from what I have been able to learn it would well repay a thorough examination, and the comparative ease with which it may be reached from the coast, together with the facil- ity it affords for the construction of a good road to the very centre of the district, should not be forgotten. A specimen of galena, holding a little copper- and iron-pyrites, from the "Acadia Claim," South Fork of McDame Greek, was given to me some years ago by Mr. J. "W. McKay. This has since been assayed by Mr. C. Hoffmann, and proves 'to contain 75 ounces of silver to the ton of 2000 lbs. A piece of native copper, fifteen pounds in weight, was at one time found in Boulder Gulch, Thibert Creek. 80 GEOLOGICAL SURVEY OF CANADA. Prospects of further placer mining. Taking into consideration the great extent of generally auriferous country included in the Cassiar district, it must be conceded, that apart from the immediate vicinity of the well known productive camps, it has been very imperfectly prospected. A great part of the district has in fact merely been run over in search of rich diggings, the simplest and cheapest methods of prospecting only having been employed in the quest. It is not improbable that additional rich creeks like those of the vicinity of Dease Lake, may yet be discovered elsewhere and it may be considered certain that there are great areas of poorer deposits ■which will pay to work with improved methods, and will eventually be utilized. It is also to be anticipated that " quartz mining " will ere long be inaugurated and will afford a more permanent basis of prosperity than alluvial mining, however rich. DlSCOVEEY AND EXPLOEATION OP CaSSIAR DISTRICT. Exploration by McLeod. Discovery. The Cassiar district of the northern interior of British Columbia may be said to have been twice discovered, first by officers and em- ployees of the Hudson's Bay Company, and again, after a considerable interval, b}' the gold miners. The unsuccessful attempt made by the Hudson's Bay Company in 1834, to reach the trade of the interior country west of the Rocky Mountains from the mouth of the Stikine, has already been noted, (p. 59 b). Efforts were at the same time being made to open up routes from the eastward. In the summer of 18.34: Mr. J. McLeod, chief trader, was employed exploring the Liard River above Fort Halkett, and in endeavouring to discover some stream flowing to the westward. He found and named Dease Lake,* crossed to the head- waters of the Stikine, which he proposed to name the " Pelly River," and travelled westward in the valley apparently as far as the Tooya or Second North Fork. The Indian bridge (afterwards named Terror Bridge by Mr. Pv,. Campbell), by which this river was crossed at the foot of " Thomas' Fall," was of such a character that neither McLeod nor any of his eight men dared to attempt it, and from this point he and his party retraced their steps. McLeod's The geographical information obtained by McLeod is incorporated routeandmap. j^^ Arrowsmith's map of 1850, on which, however, the upper part of the Stikine, which McLeod had proposed to call the Pelly, is named " Frances River," and is placed much too far north and is not con- nected with the Stikine. The name Frances is still retained as aq * Dease Lake and River were so named by McLeod after Peter Warren Dease, the Arctic explorer. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 81 alternative one to "Stikeen" on the edition of 1854, though it has since fallen entirely into disuse. McLeod's route from the head of Dease Lake, as shown on these maps, crossed the Tanzilla within a few miles of the lake, and followed its left bank, recrossing before the main Stikine enters the valley, probably by an Indian suspension bridge, which is reported still to exist, within a mile or two of this point. On careful consideration of the facts there can scarcely be any doubt that the Tooya River was McLeod's furthest point, and the Indian bridge probably crossed it near the position of the present bridge, though it may have been at some point further up the stream which has not yet been mapped. In 1836, McLeod's successor at Fort Halkett was instructed to Second establish a post across the mountains and to extend the trade down ditiou. the Stikine, or " Pelly,'' as it was then called from McLeod's naming. For this purpose he left Fort Halkett early in June, with a party of men and two large canoes, but the expedition entirely miscarried. The appearance, or reported appearance, of a large force of hostile Indians at Portage BruW, ten miles above Fort Halkett, so alarmed the party that they turned back in great haste, abandoning their goods, and lost no time in running down stream to Fort Simpson. For most 6t the above particulars I am indebted to Mr. Robert Exploration Campbell, who was at the time of the return of the last-mentioned ^ '''"'^ party in temporary charge of Fort Simpson. The news brought back by these expeditions was of a character to discourage further enter- prises in the region, the extremely difficult and dangerous navigation of the Liard River, which constituted the avenue of approach from the Mackenzie, being an additional deterrent. In 1838, however, Mr. R. Campbell volunteered his services to establish a trading post at Dease Lake, and in the spring of that year succeeded in doing so. He was accompanied by a half breed and two Indian lads only. After ascertaining that the " Pelly " of McLeod was identical with the Stikine, he returned to Dease Lake, where, to employ his own words, " we passed a winter of constant danger from the savage Russian [coast] Indian.{ n, party of miners who spent the winter of 1874-75 in its vicinity foiii- died of scurvy. Of the other tributaries of tho LianI, which must be numer- ous, I have been unable to ascertain anything authentic. Respecting the names of the Liard and Fi'anees rivi^r's, it should be mentioned that Campbell called that which is now known as the Liard the "Bell River," after Mr. J. Bell of the Hudson's l!ay Company. Under this designation it appears on Arrowsmith's map of lH.'j'l, the name Liard being applied to the branch now known as the Frances. Usage has, however, changed the first nomejiclatup'e, and it is undesir- able to attempt to reviTt to the original names, iis, irrespective of the (luestion of relative size, the physical characteristics of the Liard below the confluence are undoubtedly continued on tlu^ west rather than on the east Vjranch above that point. The Indian name of the l''rat]C(\s is identical with that of the l><'aKe, being Too-tsho-tooa', or "big lake river." Rocks like those of the flower Cafion are seen at intervals for about two miles above its liead, beyond which, for alioiit three milc^s, .sl.ra,tified gravels and sands only appear in the banks. Mix miles from the CMfion Tertiary clays of whitish and gray colours, and assoe.iaUul with impure lignite, are first met \\il,h, and lliesi' continue; to appear h(!r(iand there along the river as far as ohe I'Vances. Tho thickest bed of lignite observed was about three feet, four miles below the P'rances. Tho Names of Liavd and Frances ri vcfH. Rfjokn alievi the Gauon. YUKON D13TEICT AND NORTHEKN BRITISH COLUMBIA. 99 lignite is generally impure and often very distinctly laminated. It Lignite, resembles in character the lignites of the Miocene of British Columbia, and the associated clays and soft shales are similar in character to those of that formation. Numerous boulders of basalt are found along Basalt, this part of the river, and the basalt was observed to form a mural cliif, at a height of about 30,0 feet above the river, at a place just below the mouth of the Rancheria River. This rock evidently over- lies the lignite-bearing beds. The shaly clays and lignites show evi- dence of considerable disturbance, and dip in some places at rather high angles. This may be due to the action of old land-slides along the banks of the river, but appears to be rather too constant to be satisfactorily aocouted for in this way. Near the mouth of the Frances the white silts again become a pro- White silts. minent feature, though scarcely seen lower down the z-iver. They overlie the Tertiary rocks and hold concretions of various forms here and there. They are capped by the usual stratified sands and gravels, which generally have a yellowish or rusty colour. The gravel bars and the shores of this part of the Liard are almost *^uld. half composed of rolled quartz pebbles, which have evidently been derived from veins traversing relatively soft schistose rocks like those of the caflon. The great quantity of such vein material present in this district may be regarded as a favourable indication in respect to mineral development. Some small bars have paid to work along this part of the river, and gold is also found in some layers of the gravel deposit which overlies the older rocks along the canon and above it, where " wa^es " at $4 a day can be made. The amount of cover which it soon becomes necessary to remove in following the paying layers, has prevented extensive mining, but probably these gravels might be advantageously worked as a whole, by sluicing or by the hydraulic method. No general view of the country can be obtained from the river. General ap- owing to woods and the depth of the valley, but from high points of the country, banks above the river, near the Frances, a large area may be over- looked. Thus seen, the country is found to be a wide, rolling plateau, with an average elevation of about 500 feet above the river, or say 2700 feet above the sea. It rises here and there, however, in broad, rounded swells, or flat-topped higher plateaux with steep edges, anda considerable part of this higher ground is at an elevation of about 1000 feet above the river. The plateau is everywhere wooded, except where intersected by grassy or mossy swamps of small area. There is a large triangular region of country of this kind between the Upper Liard and Dease, bounded to the westward by the front of the 100 GEOLOGICAL SURVEY OP CANADA. Cassiar Range Cassiar Range, the sharp, rocky peaks of which carry a considerable quantity of snow and run along the horizon line for many miles, but which, with the exception of a few outlying summits, is at a minimum distance of about twenty-five miles from the mouth of the Frances. The same low country runs in a north-northwestward direction, without apparent limit, forming the upper part of the Liard valley. To the Mountains in eastward it is bounded, at a distance of about ten miles, by a compara- tively low range of rounded mountains and hills, which, from the Indian name of one of its salient points to the northward, may be called the Tses-l-uh Range. The part of this range nearest to the confluence of the Frances and Liard has an elevation estimated at about 3500 feet above the sea. Like other main features of the country, it runs in a north-northwest by south-southeast direction, but dies out com- pletely before reaching the Upper Liard River, which, if continued, it would do at the Lower Caiion. It is probable, indeed, that the cation is produced by the river cutting across the extension of the same ridge Tertiary of rocks which produces these hills. The general uniformity of the plateau appears to be largely due to the Tertiary rocks, which doubtless underlie almost its entire area. The appearance of some of the flat- topped hills above alluded to is suggestive of the occurrence of sheets of basalt overlying the softer rocks. The forms of the Tses-i-uh Mountains show that they are not thus composed, but Tertiary rocks may again occur beyond them, in the valley of the McPherson or Highland River, which is not far off. Frances Riveb. Mouth of The general direction of the Prances, for nine miles from its mouth, MklcUe'cauon is north-northwest. It then bends to the north-eastward, and in four miles the lower end of the Middle Canon is reached. For the first few miles above its mouth the Frances is extremely tortuous, so much so that the distance following the actual course of the river to the foot of the canon is twenty -two miles. This river, like the Liard, was at a me- dium stage near the end of June, 1887. Marks along the banks showed that it had been about six feet higher in the spring, and that it had since been falling. Its average width in this part is about 600 feet, and the rate of the current, at the medium stage above referred to, about four miles and a-half an hour. Terrace. The highest land immediately bordering on this part of the river is a terrace at a height of about 160 feet above it, the surface of which is in some places composed of almost pure sand, upon which open woods of Pinus Murrayana grow. Larch was observed to be moderately YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 101 abundant in damp, shady localities and the banks were in some places diversified with flowers, of which Potentilla fruticosa and Primula mistassinica were specially noted. Numerous small exposures of Tertiary shales and clay, of gray. Exposures in blackish and yellowish tints, occur along this part of the river. Lig- ^^'^ '"*°''°' nite is strewn in great quantities over some of the bars, and though thin seams occur in places in the banks, it is probable that thicker ones exist in the bed of the river. The lignite often holds drops of fossil resin or amber. The Tertiary rocks are very generally covered by silts, like those already several times alluded to. In a bank near the mouth of the river, which gives a complete section from the top of the terrace above described, the lowest deposit consists of roughly stratified gravel and clay, with some glaciated boulders. This is fol- lowed by the silts, which are again in turn covered by stratified sands and gravels. The lowest deposit probably represents the boulder-clay, and resembles that of the Upper Pelly River, subsequently described. Quartz is not so abundant a constituent of the gravel of the river- bars on this part of the Frances as it is on the Liard below, and no basalt blocks or boulders were observed here. The Middle Canon, as it may be called for the purpose of distin- Middle guishing it, is about three miles in length, the river being hemmed in by broken, rocky cliffs of 200 to 300 feet in height for the greater part of this distance. We took our boats up along the south-east bank, making four short portages of part of the stuff, and two of both boats and load, across narrow, rocky points. One portage of greater length, on the opposite bank, would overcome all the really bad water, but the banks on that side are rougher, and the whole force of the current sets against the cliff in one place in a dangerous manner. The total fall in the canon is estimated at about thirty feet. This canon is evidently produced by the southern extremity of a second range of mountains parallel to the Tses-i-uh Mountains, but to the west of that range. This southern spur, though submerged in the Tertiary and drift materials by which the general surface of the coun- try is covered, nevertheless presents a rocky barrier to the passage of the river, and in this place constitutes the eastern margin of the Ter- tiary basin, no beds referable to which are seen further up the Frances. The range of mountains here referred to may, for convenience, be named the Simpson Mountains. The exposures in the lower part of this canon show limestones, some Books of the of which are moderately pure, but others are nearly half composed of *^*2°"- small, more or less rounded fragments of siliceous and schistose rocks. These weather to rough surfaces, and have a very peculiar aspect, as 102 GEOLOGICAL SUEVEY OP CANADA. Fusulina. an intense pressure appears to have flattened the contained fragments parallel to a single direction. These limestones are associated with several varieties of calc-schists, with hard, grayish-blue, cherty quartzite, with schistose breccia, which appears to have been originally of volcanic origin, and with some pretty evident volcanic agglomerate. Further up, the rocks have been completely shattered and variously changed in appearance by solfataric or some similar action, some parts being bleached, while others are reddened by the deposition of iron, forming cliflFs of a remarkably varied appearance. The upper gate of thf; canon is composed of white cliffs of marble and quartzite, all much shattered. The marble and limestones are in places associated with red shales, resembling those found in a similar association on the Dease, near the eastern edge of the Cassiar Range, and the rooks of this part of the section are probably like those, of Carboniferous age. Specimens of Fusulina, with polyzoa, etc., are found in some of the limestones on microscopical examination. Stratigraphically, the rocks seen in the sections are all much confused and broken, and the angles of dip are very varied, though the strike may, in a general way, be given as north-northwest. The Simpson Mountains which are upon this line of strike are doubtless composed of rocks of the same character. Above the Middle CaiLon, the general course of the river is again north-northwestward for about twelve miles. It is here usually bor- dered by quite low land on both sides, and the valley between the southern end of the Simpson Mountains and northern part of the Tses-i-uh Range is about three miles in width. The wide, uniform plateau country is now, however, left behind, and we enter a generally mountainous region, though the highest summits in this immediate vicinity scarcely exceed 3000 feet above the river. Their forms are rather rounded and flowing, and the slopes of those on the east bank are nearly bare of trees, while the opposite range is generally wooded, but evidently with trees of small growth. The river itself is wide and deep with a rather slack current. A single exposure only, composed of gray-green, silvery schist, was seen along this part of the river. Simpson Lake Near the end of this reach of the river, two considerable streams enter on the west side, and on one or other of these, at no great dis- tance from the river, Simpson Lake of McLeod and Campbell is situ- ated. As the Indians who had accompanied us up from the mouth of the Dease had deserted before we reached this place, I was unable to ascertain any definite particulars respecting the lake, though it is reported to be a good one for fish. The position of Simpson Lake, as indicated by broken lines on the map, must therefore be regarded as quite uncertain. The same doubt applies to the Indian names of Beginning of mountainous country. YUKON DISTRICT AND NOETHEKN BRITISH COLUMBIA. 103 several ri-s'srs tributary to the Frances above this point, for although Tributaries of one of the local Indians had made for us, upon a sheet of canvas used as a boat cover, an elaborate charcoal drawing of the whole svstem, it proved to be extremely difficult to recognize the features repre- sented. The Indian map, such as it is, serves to show that the streams tributary to the Frances River rise in a number of lakes, some of which are reported to be of considerable size, and offer a most attractive field for further exploration. We were told, however, that none of the lakes in this region are equal in size to Frances Lake, for which we were heading, a statement borne out by the circumstance that both this and Dease Lake are known in their respective districts as Too-tsho, or " big lake," while the Frances and Dease rivers are, as already mentioned, both similarly named Too-tsho-tooa', or " big lake river." From the point just noted, the direction of the river changes to north-east, cutting across the direction of the Tses-i-uh range, which terminates at the edge of the river in low, wooded hills. The current is moderately swift throughout, and in one place the river is bordered PkIsp Cafion. on both sides by low, rocky banks, but no rapids are met with. This we named the False Gallon. One or possibly two streams enter from a valley which runs to the east of the range just mentioned, but they are not of large size. Grayish-green, quartzose mica-schist and greenish silvery schists were seen in one or two places, and in the low rocky banks above alluded to, blackish argillites and gray quartzites, of a less altered appearance than usual, but from which no fossils were obtained, occur. From the end of this reach the general course of the stream again be- River above comes north-northwest for about thirteen miles, running for the greater ^^^'^'^ Cauon. part of this distance parallel to, and a mile or two miles from the base of a mountain range, which comes in to the east of Tses-i-uh Range. The country to the west of the river is here either flat or characterized merely by low, rounded and wooded hills for many miles back, the eye ranging across this country to the continuation of the Simpson Mountains, which, with generally rounded forms and no striking sum- mits, reach elevations of 6000 to 6500 feet. These mountains do not form a strictly connected range, but appear rather as a series of moun- tainous areas, separated by wide, low passes. The Indian map above referred to shows three or four lakes in this region, supplying a Lakes, stream named Too-tshi-too-a, which flows into the Frances, reaching it probably just above the Upper Canon. None of these lakes were visible from any point reached by us. On the opposite side, one stream of considerable size joins the Frances. This is supposed to be the Aga-zi-za of the Indians, and, if so, is represented as rising in 104 GEOLOGICAL SUEVBi' OP CANADA. a chain of small lakes, some of which drain in an easterly direction to the Macpherson (Hgis-e-too'-a) River. The valley occupied by these lakes is a travelled route employed by the Indians. Granite rocks. The current is swifter in the upper than in the lower portion of this part of the Frances, and there are numerous islands in the river, but no rock-exposures occur. The mountains to the east of the river are high,Aiut have blunt, rounded forms. Much bare rock shows in their sides, but there is no appearance of stratification, and this, -with their form and colour, and the great abundance of that material found in the streams in this vicinity, renders it nearly certain that they are com- posed of granite. Character of The mountains so far met with in the vicinity of the Frances form rather isolated ranges or masses, which rise somewhat abruptly from generally low country, or are separated by wide valleys, the appear- ance being that of a mountain system partly buried in later deposits ; though no Tertiary rocks, either in place or as loose fragments, are met with above the Middle Canon. The granitic mountains last referred to form an outlying spur or buttress of the most important ranae of the district, the axis of which is here about twelve miles east of the river. This it is proposed to designate the Too-tsho Range.''' The southernmost high summit observed was named, from its form, Tent Peak. It is situated in latitude 60° 52' 45", and has an altitude of 7860 feet abjve the sea. Upper Oaiion. The river next makes an abrupt turn to the west for four miles, a mile and a quarter of this distance being occupied by a series of rapids, which are rocky and rather strong, and have a total fall of about thirty feet. The banks rise steeply from the river to heights of 100 to 200 feet, though the rooky cliffs along the water are of inconsiderable height, scarcely anywhere exceeding fifty feet. This place may be named the Upper Canon, and is the last serious impediment to the navigation of the river. We found it necessary to make several short porta,<;es, but with a large boat and at a good stage of the river, it is probable that one portage of about 1000 feet in length, on the south bank, would overcome all the dangerous water, while the boat might be ti-acked up light. A stream, with moderate current at the mouth and about fifty feet wide, enters a ^hort distance below the cufion, cominT from the mountains to the north of Tent Peak. The rocks of Rocks of the Upper Can m comprise black, glossy oalo-schists, black quartzite pper anon. ^^ g^jg^j-,^ bluish limestone, and some green-gray, silvery schist. Simi- lar rocks are seen ai;ain a couple of miles up the river, above the caELon, * From the native name of Frances Lake. I was unable to ascertain the Indian name of the range, if indeed it has any such. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 105 ■where a rapid occurs. The dips arc all low, and so far as observed, uniformly in a northerly direction. Some of the schists are highly silicitied by action subsequent to their deposition, and parts of all the rocks, including the limestone, are reticulated with narrow quartz seams. Near the upper part of the canon some hard conglomerates occur, holding schistose fragments, as well as limestone pebbles in which crinoidal joints are observable. It is not improbable that two unconformable series of rocks occur here, but I was unable to find means of distinguishing them in the sections.* From the Upper Canon to Frances Lake, a distance of twenty-one TJpperCauon n , i» . . , ,. . . . to Frances miles and a-hali in a straight line, the river maintains a northerly Lake. direction with considerable uniformity. It is deep, with a moderate current, for about eight miles, or to Moose Island, above which for ten miles the current is again swift, averaging from four and a-half miles to five miles an hour. It again becomes slack for a short distance below the lake. Some portions of this part of the river are inuch broken up by islands and gravel bars. The river-valley averages about ten miles in width, extending to the slopes of Too-tsho Range on the east, and being bounded to the westward by a series of rounded mountains, which diverge to the northward from the direction of the first-men- tioned range, and run to the west of Frances Lake. These I propose to name the Camphell Mountains, for Mr. R. Campbell, whose initial exploratory trip through this country for the Hudson's Bay Company is elsewhere referred to at length. The valley is partly occupied by terrace-flats, and partly by wooded hills or ridges with rounded or flowing forms, but which rise in some places to heights of several hundred feet. TheToo-tsho Range is here very Too-tsho compact and regular, and runs due north-and-south, the higher summits ^°^^' reaching 6500 to 7000 feet, and carrying some snow, though nothing like true glacier-ice was anywhere seen. The lower slopes of the range toward the river-valley are singularly unifornj and light. Two or three streams of some size enter the river from the eastward in this part of its course. All that could be gathered respecting these is shown on the map. The furthest north was named Tyer's River by Campbell, and is known to the Indians as Pas-ka . The exact position of its mouth was not definitely ascertained. On approaching the lake, low ridges and irregular mounds begin to Moraimo appear in the vicinity of the river, projecting through the lower ter- " ^^^' races and suggesting the existence of morainio deposits. Large bould- ers also become abundant in the river-bed. No rock-exposures what- ever were seen along this part of the river. *If so the rocks noted may represent the Cretaceous to which they are lithologically similar. lOG CEOLOCKJAL SURVEY OF 'CANADA Travelling Qur actual working time on Frances River, from its uiouth to .the Frances Eiver lake, was sixty-seven hours and a-half. The difference of level between Frances Lake and the mouth of the Dease is 477 feet. By assigning ninety feet of this to the fall in the three canons, and dividing the remainder by the total length of the river (less the aggregate length of the cafi'jns), we obtain an average rate of descent very slightly exceed- ing three feet to the mile, which is about what might be anticipated from the current met with in the river, as compared to that of other streams in the district. Frances Lake and Vicinity. Frances Lake. The elevation of Frances Lake above the sea, as d'?termined by a series of barometer observations extending from the 8th to the 16th of July, is 2577 feet. Three miles from its lower end, the lake bifurcates, forming two approximately equal and nearly parallel arms, with lengths of about thirty miles. The two arms are about eight miles apart, and are separated by a group of low, rounded mountains, the culminating point of which, with an elevation of 5230 feet, was named Simpson's Tower by Campbell, the lake itself receiving its name at the same time in honour of Lady Simpson. The eastern side of its east arm is bordered by the Too-tsho Range or hills attached to it, while the country to the west of the west arm rises more gradually to the bases of the Campbell Mountains, some miles distant. Though so far referred to as a single lake, this body of water is in reality entitled to be considered as a group of lakes. The upper end of the west arm, with a length of five miles and a-half, is separated from the main por- tion of the same arm by a river-like constriction over a mile in length, while the east arm is entered by a narrow and inconspicuous opening with a strong current flowing out, and the real extent of this arm is seen only after passing, for about seven miles, through a series of irre- gular basins and small lake-like expansions connected by narrows, in which a very perceptible current is found. It appears best, however, to retain Campbell's original name for the whole body of water, rather than to multiply names for which there is no immediate call. Character of Except in its bifid form, for which there are several parallels, arid 1 3 va ey. which results merely from the convergence of two valleys of similar characters, Frances Lake closely resembles a large number of lakes in the mountainous regions of British Columbia, and has the long narrow parallel-sided outline characteristic of lakes which occupy old valley- excavations, the drainage of which has become interrupted in \arious ways. In this case, as in a number of others, there can be little doubt, YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 107 that the lake is held in by morainic accumulations. The great number of boulders near its outlet has already been referred to, and the lumpy irregular mounds and ridges composed of detrital matter, on the lower seven miles of the east arm, as well as on the corresponding portion of the west arm, appear without doubt to represent moraines in a more or less degraded condition. The average width of the upper portions of both arms, above these interruptions, is pretty uniformly maintained at about a mile and a-half, such irregularities as occur being produced by gravelly and sandy flats a few feet above the water-level, which have been formed at the mouths of entering streams. The extent and uniformity of these points, is such as to show that the lake must have remained, during a long period, approximately at its present level. The most important of these low points is that which separates the upper part of the west arm, already referred to, and is due to the Pinlayson River.* The next is produced by a second stream which enters on the same side five miles and a-half below. This stream is somewhat smaller than the Finlayson, and is known to the Indians as the Il-es-tod -a. Except along the upper part of the eastern side of the east arm, Country the mountains do not slope down abruptly to the shores of the lake, surrounding Elsewhere, the lake is almost continuously bordered by a terrace-like plateau, which is widest to the west, and has an average elevation of about 300 feet. This resembles the low country found about Dease Lake, though even more uniform and less sloping in character, and is not far from the same actual elevation above the sea in both cases. The streams entering the lake generally cut down through the edges of this plateau-like margin in deep narrow gorges, the sections in which show that it is composed largely of rock, though levelled up to some extent by the addition of superficial gravelly deposits. This fact Origin of appears to indicate that the lake, or the drainage level of the country terraces, was maintained for a very prolonged period at or near the height of this plateau before the base-level of erosion was so lowered as to permit of the excavation of the lower valleys in which the lake now lies. There is, in addition to this, a second lower terrace, not so well marked, and not often of great width, at an elevation of ninety feet above the lake. This is seen on both arms, and is composed of ijravel and other detrital deposits. It is specially conspicuous about the mouths of the entering streams, and marks a later stage at which the water of the lake stood, for a relatively short time, subsequent to the glacial period. * So named by Campbell after Chief Factor Duncan Finlayson, afterward.s a mem- ber of the H. B. Co. Board of Directors. 108 GEOLOGICAL SURVKY OF CANADA. Picturesque Pew lakes which I have seen surpass Frances Lake in natural beauty, and the scenery of the east arm, bordered on the east by the rugged masses of the Too-tsho Range, is singularly striking. The mountains of this range are very varied in form, and a number of points surpass 7000 feet in height, while one was found to attain an elevation of about 9000 feet. This is named Mount Logan, for the late Sir W. E. Logan. Heavy masses of snow rest in some of the valleys, but no true glaciers are produced, a fact indicating a comparatively small snowfall. The broken outlines of this range contrast strongly with the rounded forms of the mountains to the west of the west arm, with which Simpson's Tower and associated hills, separating the two arms, also conform. Terminations The west arm terminates in a nearly circular basin about a quarter of a mile in diameter, at one side of which a fair-sized river, easily navig- able for boats, flows in. The east arm was not followed to its head, tliDUgh its termination in low land was seen. Here also, according to Campbell's sketch, a considerable river, which he has named Thomas River, enters. Ui)per parts The two valleys, the lower parts of which are occupied by the east s-alltVs. ^"*^ west arms of the lake, run on far beyond the heads of these arms. Each of the rivers flowing in these valleys (according to the Indian sketch already referred to) eventually bifurcates, and all four streams thus formed rise in lakes. The river flowing into the head of the west arm is named Yus-sez'-uh, and the lake on its western branch is known as Us-tas'-a-tsho. No name was obtained for the lake on its eastern branch, which is evidently, however, Macpherson Lake of Campbell. Henry Thibert, who made a prospecting expedition up the west arm of Frances Lake and far up the river in question, some years ago, has kindly supplied me with some notes on it. He estimated the river, from the head of the arm to Macpherson Lake, to be about thirty-six miles in length, while Macpherson Lake is ten miles long. The first distance was, however, intended to include the windings of the stream. He did not visit TJs-tas'-a-tsho, which, however, from the termination tsho (^ big) is probably large. Us-tas is the name of the mythical culture-hero of the Tinne. The Indian name of Thomas River is Too-tlas', the lake in which its eastern branch rises being named Tus- tles-to, the source of its western branch Til-e-i-lsho. These lakes and connecting rivers are indicated by broken lines on the map, from Thibert's account and the Indian sketch of the country, combined with the observed positions of several of the mountain ranges. Mountains to The mountains to the north in which these rivers rise, were too dis- tant to enable us to fix them with any great accuracy from points oc- cupied by us on Frances Lake, but the whole country in that direction. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 109 •a oS'ss 'n; Eh b O ■a oS'K 'n: •a oQ^St 'N O 5> 8 -o •HoS'S'N (^ O ■3 oS'8 'H O 2 a: o o O '^ •A oS'G 'n: 'M o6 'n: o Q o •A\ ol'T -H 110 GEOLOGICAL SURVEY OP CANADA. from such views as were obtained of it, appeared to be rugged and high. It was with great regret that we were obliged to abandon the idea of exploring these rivers further, but the summer was already so far ad- vanced, that this was impracticable, in view of the journey still before us. The water of Frances Lake is clear and of a pale brownish tint, and the lake is evidently very deep in its upper portions, though rather Rock ex- shallow where encumbered by the morainic accumulations already posure . alluded to. Rock exposures are, as a rule, quite unfrequent along its shores, which are generally composed of gravel and sand. At the time of our visit it was about six feet below its extreme high-water level, and it does not appear to be subject to very great fluctuations. Drift- wood is very abundant along some parts of the shores, particularly in the west arm, and it is probable that much of this is brought down by the river entering at the head of this arm. Lake-trout, whitefish, pike and suckers were found in the lake in considerable abundance. Old Hudson's The site of the old Hudson's Bay post is just above the narrow en- ay pos . trance to the east arm, on the edge of the bank, facing westward. Though Mr. Campbell had given me an accurate description of its posi- tion, it was so completely overgrown with bushes and small trees, that it was discovered with difficulty. The outline of the old stockade, with bastions at the corners, is still visible, though all traces of the struc- ture itself has disappeared. This post nas been abandoned since 1851. Forest. AH the lower country about Frances Lake is well wooded, and the mountains are also covered with forest, save where exceptionally steep and rocky, to a height of at least 1500 feet above the lake, while trees of smaller growth extend in the valleys considerably higher. The most abundant tree, here as elsewhere in the region, is the white spruce (Picea alba). It frequently attains a diameter of two feet, growing tall and straight on low ground and in sheltered places. The black spruce (Picea nigra) is also abundant. The larch (Larix Ameri- cana) is characteristic of damp, cool, northern slopes, and birch {Beiula papyrifera) is moderately abundant, though not large. The shores, and particularly the delta-flats at the mouths of streams, are characterized by groves of cottonwood (probably all referable here to Populus hal- samifera) and black pine (Pinus Murrayana). Burnt country Large tracts of country have been burnt over, many years ago, and extensive recent fires have swept the western side of the upper part of jhe east arm. Where a second growth has had time to spring up, it consists generally of mixed spruce, aspen and birch. Alders are com- mon, but scarcely arboreal, along the borders of the lake. In the middle of July thickets of wild roses were seen in many places in full bloom. YUKON DISTRICT AND NOETHEEN BRITISH COLUMBIA. Ill Taken as a whole, the growth of the forest and appearance of the General aspect country is remarkably pleasing, considering the high and northern posi- "" ' ' tion of the lake. The only characteristic difference of the woods here, as compared with those of the interior of British Columbia about the 54th parallel, is the great abundance and depth of the soft, mossy and lichenous floor, which is everywhere found in them. The trees are also often well bearded with moss, affording evidence of a continuously moist atmosphere, to be accounted for by the almost daily occurrence of light showers and the great prevalence of clouded skies, which was found throughout this part of the country. As before noted, however, the snowfall cannot be great, nor is there any indication that the total annual precipitation is very considerable. The infrequency of rock exposures along the shores rendered it dif- ffeology of ficult to obtain any connected idea of the geology of the lake, both ^ '^"*^' arms of which appear, however, to occupy valleys excavated along the strike of comparatively soft black and grayish glossy schists, which are often calcareous and frequently interbedded with bluish limestone. In lithological appearance these rocks resemble those of the Cambro- Silurian, in which graptolites were found on Dease River, though no fossils were obtained here. The strike is parallel, so far as observed, to the directions of the arms, the prevailing dips being westward on the west arm, eastward on the east arm. The character of the harder and doubtless older rooks which occupy the centre of the anticline thus formed and compose Simpson's Tower and connected hills, was not ascertained. The high rugged central parts of the Too-tsho Range are composed Quartz veins largely or entirely of gray granite, pebbles and boulders of which are everywhere abundant, and particularly so along the beaches of the east arm. There is, too, a notable abundance of quartz along all the beaches of the lake, this material being derived from innumerable veins which traverse the schists in all directions, though most often found parallel to the bedding-planes, and generally assuming forms more or less lenticular. The largest of these are often several feet in width, and those seen in the caiion of the Finlayson, near its mouth, are of workable dimensions, if only moderately rich in gold. Specimens of quartz veins, containing some iron and copper pyrites, from the east side of the east arm about midway up it, were found to contain traces of gold on assay by Mr. Hoffmann. In general appearance the rocks of Frances Lake very closely Prospects for resemble those from which the rich placer gold deposits of Dease Lake ^° ™ii"ng- are derived, and they are probably of about the same age. Several ." colours '' to the pan were obtained from surface gravel at the mouth numerous, 112 GEOLOGICAL SURVEY OP CANADA. of Finlayson River, which struck me as specially promising in aspect, and there seems to be no reason why some of the streams flowing' across the schistose rocks into the lake or in its vicinity should nob prove to be richly auriferous. ^ This entire district well deserves care- ful prospecting. After my return to the coast, in the autumn, T ascertained from Charles Monroe that he and some other miners had actually done some prospecting in the vicinity of the lake at the time when the Cassiar mines were yielding largely, and the more enter- prising men were scouring the country in search of new fields. He reached the lake from Cassiar by the same route we had followed. On comparing notes we found that he had worked for a short time at the mouth of the Finlayson, where he found the gravel to pay at the rate of from |8 to $9 a day. Country between Frances Lake and Pelly River. Proceedings at We reached Frances Lake on the morning of the 8th of July, and jjranoes a e. ^^^ ^^ been able to find any local Indians to serve as guides and assist in carrying over our stuff, we should have proceeded at once to the best point for that purpose and continued our journey overland toward the Pelly. As it was, it became our first object to endeavour to find the trail used many years previously by the Hudson's Bay Company, of which a general description had been furnished by Mr. Campbell. This necessitated a careful examination of the west shore of the west arm to its head, which enabled us to identify, with toler- able certainty, the stream which Campbell had named the Finlayson. It was supposed that the Indians might have employed the same route in the periodical journeys which they were known to make from the Pelly down the Frances to the little trading post at the mouth of the Dease ; but though the remains of an old log cache of the Hudson's Bay Company were eventually found, together with the nails and iron work of a large boat which had evidently been burnt on the beach near it, no sign of a trail could be discovered. It thus appeared very doubtful whether we should be able to make our way across to the Pelly, with sufiicient provisions and the necessary instruments for the continuation of our survey in the Yukon basin. Search for In order to exhaust the possibility of obtaining further assistance before* making the attempt, I made a light trip in one of our boats ound into the east arm, which was known to exist from Campbell's I'eport, but the narrow entrance to which had not even been observed on our way up the lake. This also enabled me to sketch the east arm; but no Indians were found, in fact, we discovered traces of only a' Indians. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 113 single camp which had been made during the same summer, most of the Indian signs being two or more years old. All that now remained to be done was to make the best of our own Arrangements resources. We, therefore, went carefully over all our stuff, separating ^^jj°"™«y*° out everything which was not absolutely essential, and making up the remainder in packs, together with as much food as could be carried. This done, we stowed a great part of our camp equipage, together with some provisions, in a strong log cdche, which was constructed for the purpose in the bay immediately south of the mouth of the Pinlayson, and moved round to the north side of the delta of that stream, to what we believed to be the best point from which to start. We then hauled out our two boats, and on the 17th and 18th of July carried our remaining stuff to a point some miles up the Finlayson and above the canon and cascades, which render its lower part utterly impassable. Here we set up the Osgood canvas boat, which we had also carried Difficult land over. Into this a portion of our stuff was put, and two of our Coast ''^''"^se. Indians were instructed to endeavour to track it up the shallow and winding stream, while the rest of the party found their way as best they could along the valley, with heavy packs. The walking was extremely fatiguing on account of the deep moss, alternating with brush and swamp, and as in addition the weather was very warm and the mosquitoes innumerable, our rate of progress was slow. On arriving at the forks of the stream we unfortunately took the wrong branch for several miles, leading to some loss of time, but we event- ually reached a lake which we recognized as Finlayson Lake, on July 24th. The canvas boat did not arrive till the evening of the next day, as great difficulty was met with in getting it up the shallow stream, which was badly blocked with fallen trees. In the meantime, obser- vations for latitude and time were taken, and a raft was constructed on which the stuff might be floated to the head of the lake, which lay in the general direction of our route. The lake proved to be nine miles and a-half in length, and near its Reach Finlay- head we again found the ruins of a Hudson's Bay cdche, but no appear- ^°" T.'^ke. ance of a trail. Having selected the most promising looking place from which to continue our journey, we took out the raft-sticks, in order that they might remain dry and serviceable for our Indians on their return, and made a second small cdche of provisions. The Osgood boat being almost worn out by its hard usage on the Finlayson, and being besides quite too heavy to carry overland in addition to our other stuff, was also drawn up and abandoned. Soon after leaving the lake we fell upon small streams which evi- Reach the dently drained toward the west, and about noon on the 29th of July ^PP®'' ^'''•y- 8 114 GEOLOGICAL SURVEY OP CANADA. Probable better route. Mountains. General character of the country. we had the satisfaction of reaching the bank of the Pelly River. From this place our five Coast Indians were sent back, with instructions to take back to Mr. Reed, at Dease Lake, the articles left in the c&ohe on Prances Lake, and this duty, we subsequently learned, they faithfully performed. Having constructed a canoe from the canvas brought over for that purpose, we began the descent of the river on the 1st of August. The above is a summary of our proceedings from Prances Lake to the Pelly. I now feel convinced, that if we had had Indian guides, we might in all probability have shortened the land carriage and possibly have found a travelled Indian trail, by following up the waters tributary to the west arm of Prances Lake. It remains to give some notes of the country actually traversed. Though the region between Prances Lake and the Pelly may be described as a mountainous one, no very high summits were seen, the elevations being, as a rule, rounded and regular in outline, and form- ing broad, plateau-like areas above the timber-line in some places. The Too-tsho Mountains, which run along the east arm of Prances Lake nearly due north, appear to turn more to the westward beyond the head of the lake, but the line of travel followed toward the Pelly still diverged rapidly from this, the culminating range of the district, and the general direction of the principal ridges in the vicinity of the Finlayson River is not far from east-and-west. It is probable that the general character of the country here met with fairly represents that of a wide belt to the west of the Prances River and north of the Liard, including the Campbell and Simpson Mountains and their vicinity. The mountains are about equal in alti- tude to those last mentioned, averaging from 5000 to 6000 feet. The country is traversed by wide, wooded valleys, of which that occupied by the Finlayson is here the principal. The climate evidently becomes less moist as Frances Lake and the vicinity of the Too-tsho Mountains is left, and dry, gravelly terrace-flats, with Pinus Murrayana, are not uncommon on the upper part of the Finlayson. Larch was observed in places for about half the distance up the river toward Finlayson Lake, but was not seen further west. White spruce is still the most abundant tree, and grows as large as along Prances Lake. The black spruce also, however, occurs, and Abies subalpina becomes the common and characteristic tree near the upward limit of » forest growth on the mountains, which here occurs at about 4200 feet. Grassy swamps are found in a number of places, and a good growth of grass is also met with where areas have been denuded of forest by successive fires, so that should it ever become desirable to use horses on this portage, they might be maintained without difiiculty. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 115 The lower part of the Fialayson, for about four miles near its Character of mouth, forms a series of rapids and small cascades in a narrow, River, rocky gorge, making in this distance a total descent of 300 feet to the lake. Above this caflon it is rapid for several miles, with gravelly bars, and quite shallow, but further up it becomes a narrow and often deep stream, flowing between muddy and sandy banks. At twenty-two miles from its mouth, it divides into two nearly equal branches, the northern of which comes from McEvoy Lake, the south- ern from Finlayson Lake. Each of these streams, at their confluence, is from twenty-five to thirty feet in average width and about two feet deep. The northern branch, however, soon becomes shallow, rapid and stony, while that coming from Finlayson Lake is extremely crooked, winding in all directions in a flat valley about a mile in width, and is besides, as already mentioned, very badly blocked by fallen trees. From the summit of an isolated, bare- topped hill, which occupies the View from a angle between the two streams, and rises about 1 200 feet above the val- leys, an excellent view of the surrounding country was obtained. To the north-eastward, broken masses of high, rugged mountains, patched with snow, limit the horizon. These are evidently connected with the Too-tsho Range, but appear to form an irregular western spur, which is not continued to the west of north. The intervening district, as well as the whole country to the south, is occupied by bare-topped, rounded mountains and ridges of less elevation. Amid these hills the round or flat-bottomed valleys of the streams extend, showing here and there the paler green of a patch of meadow. The most remarkable feature is, however, a tract of low, level country, in which McEvoy Lake and the head of Finlayson Lake lie. This runs nearly due east- and-west, and appears to continue in the former direction till it inter- sects the northern continuation of the valley of the west arm of Frances Lake, and is also observable to the west of the head of Fin- layson Lake. It is intersected throughout by numerous small lakes or ponds and by swampy meadows. Finlayson Lake {Tle-tlan -a-tsoots of the Indians) is nine miles and Character of a-half in length and rather irregular in form. Its elevation above sea- Lake, level is 3105 feet, as determined by our barometer readings, and it may be regarded as occupying the summit of the watershed between the Mackenzie and the Yukon, as no stream of any importance enters it. The country about it is all rather low, but is diversified, to some extent, by wooded ridges and hills, which rise highest near its upper end. The water is apparently shallow throughout, and had, in con- sequence, a much higher temperature than that of Frances Lake. It 8i 116 GEOLOGICAL SUEVEY OP CANADA. The water- shed. Country west of Finlayson Lake. Geological notes. is well stocked with white-fislj and lake trout, and also, no doubt, with the other species found in Frances Lake. A dead amia, eighteen inches long, was also found on the shore. The immediate shores of the lake are generally quite low and often svifampy, and the country is covered with small, poor timber, much of which has been killed by fire. A pretty well marked terrace, at 100 to 150 feet above the water, runs nearly all round the lake, and at the head are irregular, low, lumpy ridges and islands, which evidently represent moraine accumulations. No rock-exposures were anywhere seen along the lake-shore. The distance from the head of the lake to the nearest point on the Pelly, in a straight line, is about fifteen miles, but the low tract of country already referred to, runs some miles to the south of such a line for the greater part of the way. The actual watershed in this low country is probably not fifty feet above the lake, but there is no evi- dence that the lake ever discharges toward the Pelly. Its height above sea-level is about 3150 feet. Small streams rising to the west of thu lake, flow together to form a respectable brook about half way across. This occupies a wide, terraced valley, the ridges bordering which gradually diverge as the Pelly is approached, and the river itself is bordered by undulating terrace-flats several miles in width. On ridges west of the head of Pinlayson Lake Abies subalpina becomes moderately abundant, but the white and black spruce are still the characteristic trees, and the former is well grown in sheltered valleys. No larch or black pine were seen on this part of the portage. The western yellow pond-lily (JVuphar polysepalum) was observed in great abundance in a small inlet of Finlayson Lake, and on the hills beyond Polymonewm pulchellum and Mertensia paniculata, with Po- tentilla fruticosa, were noted as specially abundant flowering plants. The vegetation in the vicinity of the Pelly was much further advanced than any we had yet seen, and the climate of the valley is evidently more favourable than that of the watershed region. The soil of the river-terraces is a fine, silty material, which, judging from the luxuriance of plant growth, must be very fertile. In consequence of the width of the valleys and the mantle of drift deposits, few rock-exposures were met with along the whole route from Frances Lake to the Pelly, and those examined consisted wholly of schists or schistose argillites, associated with cherty quartzites in some places, and generally of blackish or gray colours. The gravel wash in the various smaller streams and the rock debris met with on slopes of hills and elsewhere, consisted also predominantly of similar materials,, and it would appear that the whole of this country is underlain by YUKON DISTRICT AND NORTHEKN BRITISH COLUMBIA. 117 rocks resembling those described on Frances Lake and part of the river of the same name. A reddish argillite was noted as locally abundant in some streams entering the Finlayson on the south. Pebbles and small boulders of gray granite are not uncommon, but all appear to have been transported from a distance. In addition to the Rocks of Fin- rocks above described, the gravel beaches on Finlayson Lake show numerous fragments of gray-blue limestone, some of which contain obscure fossils. Green serpentine, like that of the vicinity of Dease Lake, and exhibiting minute veinlets of serpentine-asbestos, with a few pebbles of reddish serpentine, were also noted here, and fragments of a peculiar white quartz-porphyry are not uncommon. Large pieces of the same stone were afterwards found on the hills to the west of the lake, but it was not anywhere seen in place. Quartz vein-stuff is everywhere very abundant, and on the terrace overlooking the Finlay- Massive son, on the north side, three miles below the lake, a large mass of quartz occurs in place. The extent of this mass of quartz could not be ascertained, as it protruded from the soil only in isolated spots over an area several hundred feet in length and breadth. A few specimens were collected, but on assay proved to contain neither gold nor silver. One of the most notable features of this watershed region is the Superficial de- , . , . , , ., . , 1 • 1 1 , , . posits of the great quantity of detrital material or dritt with which the whole is watershed. covered. On the lower part of the Finlayson, irregularly bedded, clayey, gravel depo.sits, containing large glaciated stones and resem- bling boulder-clay, were seen ; but elsewhere stratified gravels and sands are generally shown in any scarped banks which occur. Well rounded gravel was found scattered over the very summit of the iso- lated mountain at the forks of the Finlayson, above referred to, at a height of 4300 feet above the sea, the material being of varied origin and including granite. No distinct terraces were found on this moun- tain, but terraces were noted further down the valley on the slopes of higher hills, at a height equal to, or greater than, that of this place. The evidence appears to be conclusive that a body of water, in later glacial times, extended quite across the Pacific- Arctic watershed in this region, standing at a level more than 1000 feet above it. Glacial striation, in a bearing parallel to that of the valley, was noted on the surface of the quartz mass previously alluded to, but the direction of motion of the ice could not be determined. Upper Pelly River. Our first camp on the Pelly was situated in lat. 61° 48' 52", long, tipper Pelly 131° 01' 06", the height of the river being at this place, as approxi- 118 GEOLOGICAL SURVEY OP CANADA. mately determined from the mean of a number of barometer observa- tions, 2965 feet. The river is here 326 feet wide, with a current slightly exceeding two miles and a-half an hour, and a middle depth of seven feet. An approximate estimate of the discharge placed it at 4898 cubic feet per second. The river had evidently fallen very considerably since the early summer and was probably at or below its mean stage. The water is here nearly clear, with a light brownish tinge. From explorations made at the time of the existence of the Hudson's Bay post, as well as from Indian report, the river is known to be navigable by boats for a considerable distance above this point, and to rise in two lakes, the position of which is approximately indicated on the map, according to Mr. Campbell's sketch.* Our camp was about two miles above the mouth of the stream which has already been mentioned as rising on the portage near Finlayson Lake, at the ajigle between which and the Pelly the old post named "Pelly Banks" was situated. We saw no trace of the buildings which formerly existed, though the old site might, no doubt, have been determined by a little search, had we thought it worth while to devote the necessary time to it. " Pelly Banhs " to Hoole Canon. Prom our initial point, above meationed, to Hoole Canon and Appearance of ^ , .' -i - • , i- the country. Oafion, is a distance of tbircy-one miles m a straiglit line, the direction of which is a few degrees north of west. The river, liowever, forms a wide curve to the south of this line, and is besides veiy tortuous in detail, the actual distance, following its course, being fifty miles. The main orographic river-valley is here not confined, as is so often the case, between parallel ranges of mountains. There is on the contrary a wide tract of irregularly hilly country, which is bounded to the south by a well-defined mountain range at a distance of ten to twelve miles. This range is crowned by a series of square-outlined pyramidal peaks, which are probably composed of stratified rocks. It is proposed to dis- tinguish it as the Pelly Range. To the northward, no definite bound- ary to the low hilly region can be seen. The actual trough in which the river meanders, however, is scarcely more than a mile in average width, and is generally bordered by terraces a hundred feet or more in height. „. Thirty- three miles, by the course of the river, below our starting point, a tributary comes in from the mountains to the southward, about fifty feet wide by one deep, and very rapid. This is identified [*Correoted on the mapnow published by Mr. Warburton Pike's later observations.] YUKON DISTRICT AND NOBTHEEN BRITISH COLUMBIA. 119 as Hoole River* Its water is bluish in tint, and clearer than that of the Upper Pelly, which by this time has become slightly turbid from material derived from its soft, silty banks. The river, between our first camp and Hoole River, has a rather moderate current, scarcely exceeding four miles and a-half an hour, though with several little " riffles " or small rapids. Just below the mouth of Hoole River is a rapid about 600 feet Rapid, long, with a total fall estimated at about ten feet. There is an easy portage on the right or north bank, but a fair-sized boat might run through without danger at most stages of the water. As a precaution- ary measure, we lightened our canvas canoe, of the behaviour of which in bad water we had had at the time no experience. From this rapid to Hoole Canon the water is swift, and there are several little rapids. The banks and beaches of the Pelly above Hoole River, are gen- erally silty or muddy, though the strength of the current is sufficient to produce well-washed gravel-bars in mid-stream. Below that point the banks and beaches are also as a rule gravelly, in conformity with the swifter flow of the stream. The banks along the south side of this part of the river, are for the Vegetation, most part densely wooded, and where shady and damp the growth of timber is small and scrubby, with much black spruce. The banks on the opposite side are very different in appearance. Those above Hoole River show numerous open, grassy patches, and below that place grassy slopes preponderate over the wooded area, the grass having the characteristic growth and dry, tufted appearance of ''bunch- grass." The trees are similar to those found along the rivers previously described, except that Pinus Murrayana and larch do not occur, and but a single white birch was noted, near the mouth of Hoole River. Groves of cotton wood of medium size cover some of the flats. In one Frozen soil, or two banks into which the river was cutting, and where the surface was covered with a dense, mossy growth, frozen soil was observed. The depth to which it extended could not be ascertained, as it went below the water-level of the stream. Above the mouth of Hoole River the rock-exposures are few and Rooks inconsiderable. Near the mouth of Campbell Creek, a yellowish- ''^P'^^^"*^"' weathering irregularly silicified dolomite was observed in several low outcrops, together with green-gray, slightly lustrous schists, and similar schists were seen in one or two other places on this part * So named by Mr. Campbell after his interpreter. Mr. Campbell, on his original sketch and on Arrowsmith's map, in the construction of which it was used, has named a number of the tributaries of the Pelly. It has been found difiBcult to identify some of these, but all have been placed on the accompanying map, after a careful comparison. 120 GEOLOGICAL SURVEY OF CANADA. of the river. The composition of may be accepted as indicating in a the gravel of the river-bars general way the character 'if the formations of the drainage-area of the upper river. These include fragments of schists, quartzites and argillites of the same kind as before described on the Frances, with gray, fine-grained limestone, some pieces of coarse white marble, and occasional small pebbles of the same peculiar quartz-porphyry seen on Finlayson Lake. There are also represented several varieties of granitoid rocks, the most abundant of which has a coarse gneissic lamination, with whitish or greenish mica and large, white, porphyritic felspar crystals, round which the layers of the rock bend. This peculiar granite or gneiss is probably derived from the head-waters of the main river, and often occurs in lar^e boulders, which can scarcely have been carried by the stream, and have probably been transported during the glacial period. Basalticaroa. At the rapid at the mouth of Hoole River, low bluffs of dark-brown basalt appear, and the same rock crops out in several places further down the river, extending probably to within two miles of Hoole Canon, though the lower exposures have rather a dioritio appearance, somewhat different from those first noted. The horizontal extent of this local development of basalt is approximately indicated on the map, but as the country to the north is here all low, its limit in that direction is entirely hypothetical. This is probably the first occurrence of basalt on the river, as no basalt pebbles were seen above this place. The basalt is often amygdaloidal or vesicular and contains chalcedony, calcite and some fibrous zeolite, but is scarcely at all columnar. Scarped banks Above the mouth of Hoole River, frequent scarped banks exhibit white or gray bedded silts, associated with or underlain by stratified gravels, with a dark-gray, earthy matrix. These latter deposits often hold largre boulders, and observations made further down the river appear to show that they represent the boulder-clay. Between Hoole River and the canon, the silts are scarcely seen in the banks of the Pelly, being replaced by gravelly and coarse sandy deposits. Hoole Canon to Boss River. Hoole Canon. At Hoole Canon, the river makes a knee-like bend to the north- eastward, and is constricted between rocky banks and cliffs about a hundred feet in height (see Plate IV). These render it impracticable to use the line, and as the water is very rough and dangerous, it was found necessary to carry not only all our stuff, but the canoe as well, to the lower end of the canon. The distance by the river is about three-quarters of a mile, by the portage half a mile, the highest point Geological Sukvet of Canada. Plate IV. 6. M. D.— Photo , Aug. 3, 1887. ENTRANCE TO HOOLE'S CANON, PELLY RIVER. - w- n ni ' ' G. M. D.— Photo,, Aug. 4,1889. LOOKING DOWN PELLY RIVER, FROM GRAVEL BARS AT MOUTH OF LAPIE RIVER. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 121 being one hundred feet above the river. The portage is on the south side of the river, and we found traces on it of skids which had been laid by the Hudson's Bay Company many years ago, but no sign of its having been employed by the Indians, who in all this district generally travel by land, making rafts when they are obliged to cross any of the larger rivers. Sixteen miles and a-half below the canon in a straight line, or Koss River, twenty-three miles by the course of the Pelly, is the mouth of a river which is identified as the Ross River of Campbell.* This stream, which comes from the north-eastward, is to all appearance as nearly as possible equal in volume to tlie Pelly, having a width of 290 feet, with a current of four miles and a-half an hour. Its water is tur- bid and milky, and colder than that of the Pelly, leading to the belief that it is not derived from lakes like that stream, or that if lakes do occur on its upper waters, they are much less in area than those of the Character of it". county. Pelly. Its size would indicate that it may rise in the same distant range with the Pelly, but no long view was obtained up its valley, which is narrow and bordered by high, steep hills at its mouth. Mid- way between the canon and the Ross, Ketza River,f a swift stream, about forty feet by one foot, joins from the south. Like other streams from that direction, it carries clear, blue, mountain water, and brings down quartzites, argillites and schists of the usual character, together with a great abundance of quartz-gravel. The Pelly, between the canon and Ross River, is swift throughout, Eooks. with numsrous little rapids. To the south of the river there is still a wide extent of low, wooded country between it and the continuous range before referred to as the Pelly Mountains. To the north the view is more limited, particularly near the mouth of the Ross River, owing to the existence of a long, steep ridge, parallel to the course of the Pelly, and from 600 to 800 feet in height above it. The southern face of this ridge, which is cut through by the Ross River, is more than half, open grass land, and would aiford excellent pasturage. The rocks of Hoole Canon and its vicinity are chiefly white marble, associated and interbedded with gray and black cherty-looking quartz- ites which are often thin-bedded and sometimes rather schistose, and precisely resemble the Cache Creek quartzites of southern British Columbia. Further down the river are occasional exposures of green- ish and gray schists and schistose argillites. The rocks along this part of the Pelly strike nearly parallel to its main direction, or about north-west by south-east, and are either nearly vertical or have dips to *So named after Chief Factor Donald Boss. + So named by Campbell, after one of his faithful Indian companions ; not Kelzas, s shown on Arrowsmith's map of 1854. 122 GEOLOGICAL SUKVEY OF CANADA. the south-westward. Stratified gravels are seen in a few places in scarped banks, but silts are not here specially characteristic. Eoss River to Glenlyon River. From the mouth of Ross River to the Glenlyon River, the general course of the Pelly is almost direct, on a bearing of N. 50° W., the dis- tance being sixty-four miles. In consequence of the number of minor flexures in the stream, this is, however, increased by the river to eighty- two miles. Ten miles below the Ross, following the river, a stream, Lapie Lapie River. River* sixty feet wide by one foot deep, and resembling in its general character and colour of water Hoole and Ketza rivers, comes in from the south. Twenty-three miles from the same point a smaller tri- Orchay River, butary joins from the north, which is supposed to be the Orchay of Campbell. This is about twenty feet wide by six inches deep, and carries clear, brownish water, doubtless derived from a lake or lakes. Ten miles further, on the same side, is a valley running to the north, which probably brings in a moderate sized stream. But this falls into a slough, behind islands, and was not seen. Featuresofthe All the way from the Ross to the Glenlyon the Pelly is closely bor- dered on the north by ridges and hills of considerable height, which become mountains of 4000 to over 5000 feet before the last-mentioned stream is reached. These entirely preclude any outlook over the country on that side. To the south, the important and well-marked Pelly Range. Pelly Range is continued to a point opposite the Orchay River, where it appears to terminate in a group of mountains lower than those of its eastern part, but still from 5000 to 6000 feet in height. These are situated at a distance of about six miles back from the river, the inter- vening country being occupied by lower, wooded hills and broken coun- try. The forms of the mountains are rather bold, consisting of steep crests and ridges, with intervening narrow, V -shaped gorges, and they appear to be grassed or covered with low, herbaceous growth, giving them a general greenish tint. There are few bare, rocky summits, and the whole appearance is that of a range shaped by normal pro- cesses of denudation from schistose or other crumbling rocks of a stratified character and nearly uniform hardness. They still carried a few patches of old snow on the 4th of August. The higher portions of this range to the eastward, present a rougher and almost serried outline, but there are not even there any exceptionally high points, and the slopes of the peaks are extremely uniform. Beyond the west- * This stream waa not named by Campbell. I call it Lapie River, after one of his Indians, he having given the name of the other (Ketza) to a neighbouring tributary. C5 Ph > M D rt iS H hH . Jj ^ •n <; O) J & ^ f=^ o o ^ © W s H ,c t) 4J O ^ S 0) ^ ho ^ OJ p^ « H 1 > -f3 1— 1 m « £? |x +3 hJ (S J 03 Ph ^ g P5 'S H 13 CM 0) > Ph o P ^ ^ Ni" ^ E^ s, |3 S" iJ i-l PP "1 H 1" fH 13 P5 Q 2 3 O ^ iC p s YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 123 em termination of this range, for a distance of about twenty miles, no Gap to the mountains were seen to the southward from the river-valley. From the fact that evidences of a more humid climate were found along the corresponding length of the river, it is highly probable that a some- what important gap occurs in this direction, of sufficient width to admit the entrance of the moisture-bearing winds. The greater humid- ity of this part of the valley is particularly marked by the densely wooded character of the slopes on the north side of the river. Beyond the gap above referred to, a low mountain-range appears on the south side of the river, not in line with that last described, but quite close to the rivei", which here, for a number of miles, flows in a valley between two mountain axes. The Pelly, for rather more than half the distance between the Eoss Notes on the riv6r. and Gleniyon, continues to be pretty swift, and is much divided among islands and gravel-bars ; the remaining part is comparatively tranquil, with the exception of the rapids in the immediate vicinity of the Glen- iyon. The forest growth throughout is much like that previously described, save that the birch is now moderately abundant, and the Black Pine, black pine {P. Mv,rrayana) appears, coming in lirst on dry northern slopes thirteen miles eastward from the Gleniyon. Cottonwood, aspen, alder, spruce and willows are the prevailing trees on the river-iiats, which are usually about ten feet above low-water level. Frozen soil was again seen in several places along the shady side of this part of Frozen soil, the river, extending from about eighteen inches below a mossy and peaty sod to the water-level, with a depth of ten feet or more. Some of these banks were being rapidly undercut by the water, which thaws the soil wherever it comes in contact with it, and causes large masses, with the superincumbent sod and trees, to fall into the stream. The rapids above alluded to as near the Gleniyon are two in number. Rapids. The first occurs in an S-shaped bend about two miles east of the Glen- iyon ; the second just below the mouth of that stream. The upper rapid is wide and rather shallow, with some rocky impediments. It is easily run with a canoe, but at low stages of the river, doubtfully passable for a steamer, unless of light draught. The current in the second rapid strikes full on the face of a rocky bank on the right of the river, and forms a heavy confused wash in consequence, but is otherwise unimpeded and deep. The rocks seen along the Pelly between the Ross and Gleniyon, Geological while resembling in a general way those previously described, differ in their greater alteration and in the evident importance in their com- position of products originally of volcanic origin. The most abundant are blackish-gray and greenish quartzites and schists, often more or 124 GEOLOGICAL SUHVEY OF CANADA. Laramie or Cretaceous rooks. less micaceous, and in places passing into true mica-schists. Three and a-half miles west of the Ross River, on the left bank, are exposures of massive, dark leek-green serpentine, associated with green serpen- tinous and quartzose schists, and a spotted white and green rock, which may be a much decomposed diabase or diorite. One portion of the bank shows at least a hundred feet in thickness of nearly pure serpen- tine, but no asbestos veins were seen here. Thirty-six miles west of the Ross a very peculiar purplish felsite was noted. It is schistose and slightly micaceous in the division planes, besides being porphyritic with irregular white felspar masses. Four miles east of the Glenlyon, a close-grained, banded, white-and-gray felsite occurs, in a, much shat- tered exposure, and is probably a fine, altered volcanic ash. All these rocks evidently form members of a single series, and though with numerous local irregularities, strike in a general way parallel to the course of the river. The circumstances tend to show that they are on the whole stratigraphically higher than those seen further up the river, but it is not certain that the dips are normal. The most interesting fact developed on this part of the Pelly is, however, the occurrence of rocks of Laramie or Cretaceous age. These were noted in a single low exposure on the south side of the river, twenty-seven miles and a-half west of the mouth of Ross River. They consist of black carbonaceous or possibly plumbaginous shales, rather hard, and interbedded with gray-brown sandstones, the whole dipping nearly due south at an angle of forty-five degrees. But this single occurrence of rocks of this character was found, and no rocks are seen for several miles up or down the stream, so that the area characterized by the formation to which they belong is uncertain. Sir William Fossil plants. Dawson writes as follows of the fossils obtained here: — "The few specimens examined are full of impressions of dicotyledonous leaves, much crushed and imperfect. One has the venation of Corylus Mao Quarrii (Forbes). Another seems to be a Juglans, near to J. acuminata (Braun). Both of these species are said by Heer to occur at English Bay, Alaska, and also on Mackenzie River. The determinations can- not, however, be considered as a certain." Fine sections of the drift deposits frequently occur along this part of the Pelly River, forming scarped banks from eighty to a hundred and fifty feet in height, and at times a quarter of a mile or more in length. The stratified gravels, with gray or brownish earthy matrix, which have previously been described, were seen just above the mouth of Ross River to be interbedded with and overlain by silts of the usual character. A few miles below the Ross they were first seen in associa- tion with indubitable boulder-clay, and thence down stream, the earthy Superficial deposits. YUKON DISTEICT AND NORTHERN BRITISH COLUMBIA. 125 or clayey sravels and boulder-clay form the lower portion of most of Peculiarities of boulder-clay the scarps, being capped by bedded silts, which in some places are fifty feet in thickness. The stratified earthy gravels andboulder-clay are interchangeable and may be considered as constituting a single formation. These gravels are sometimes interbedded with rude layers of boulder-clay, while in other sections they occur in the bottom of the bank, with massive-looking boulder-clay above them, and in yet other instances these conditions are reversed, and the gravels pass above into the bedded silts, which everywhere constitute the upper member of the glacial series. When the matrix of the gravels becomes clayey, they closely resemble boulder-clay in composi- tion, but do not often hold such large stones as the latter occasionally does. The stones of both the gravel and the boulder-clay are more or less completely water-rounded, and striated stones were seldom seen along the river. The gravels above referred to are quite distinct from the old river-valley gravels, which also occur at various levels, some- times as the highest member of the section in scarped banks, in other cases forming the substratum of river-flats. With the appearance of the boulder-clay the scarped banks begin to assume a characteristically castellated appearance, standing often in series of nearly vertical but- tresses and pillars, with deep intervening gullies. Glenlyon River to Macmillan River. For the purpose of description, the next convenient length of the Upper Pelly which may be taken, is that extending from the Glenlyon to the Macmillan River.* This again naturally subdivides itself into two parts, the first, with a general bearing of N. 53° W., twenty-eight miles and a-half, in continuation of the general course of tlie upper river, the second about N. 77° W. thirty-three miles. The total dis- tance, following the course of the river, from the Glenlyon to the Macmillan, is ninety-one miles. The tributary streams in this distance, Tributary again measuring by the course of the Pelly, down stream, are as ^*'^^™°' follows : — Glenlyon River, forty feet by one foot. Eight miles and a-half below, stream on the north, sixty feet by six inches. Twelve miles, tributary from the south, twenty feet by three inches. Seven- teen miles, on the south, dry torrent bed. Twenty miles and a-half, on the south, ten feet by three inches. Thirty miles on the north, thirty feet by six inches ; the Earn River of Campbell. Forty-seven miles on the south, thirty feet by three inches ; probably the Tummel River of Campbell. From this point to the Macmillan no tributary * Of Campbell, named for Chief Factor Macmillan. 126 GEOLOGICAL SURVEY OP CANADA. Glenlyon Mountains. The Detour. Approaoli to Macmillan River. streams were observed, the country to the northward evidently drain- ing toward the last-named stream, and that to the south, at no great distance, being in all probability within the drainage-basin of the Lewes. The above estimates of the sizes of the streams refer to the date at which we saw them, in August, when most appeared to be at or near their lowest sta.ges. For about twenty miles below the Glenlyon River, the Pelly is more than usually free from abrupt bends, and few islands are met with. It is bordered to the south by a range of mountains, that may be named the Glenlyon Mountains, the highest points of which somewhat exceed five thousand feet. Lower irregular hills border the north bank, and these, as usual, show extensive grassy slopes on the southern exposures. At the distance just mentioned from the Glenlyon, the river turns abruptly to the northward, making an S-shaped bend, and cutting completely through the ridge which has previously bounded it on that side. After a sinuous course of about fifteen miles to the north of the ridge, (about midway in which it receives the Earn River), it turns again with equal abruptness to the southward, rounding the west point of the ridge, which here dies away. This peculiar flexure is distinguished on the map as The Detour. To the south of the ridge is a wide valley, that lies in the general direction of the river, and which, though now apparently floored by drift deposits, doubtless represents a pre-glacial valley of the Pelly. The distance from bend to bend of the river, through this disused valley, is eight miles and a half, and the height of its floor above the water-level was estimated at about two hundred feet. As far as the lower end of The Detour the current is rather swift and there are a number of little riffles, some of which might be called rapids, though none are of a character to impede navigation. Prom The Detour to the Macmillan, the Pelly has a current averaging not more than three miles an hour, though attaining four miles in a few places. The country begins to open out to a greater extent than found anywhere on the upper river, and is diversified only by comparatively low and irregular hills. About fifteen miles east- ward of the Macmillan is a wide low gap opening to the north, by which the Indians evidently cross over to the valley of that river, as several large rafts were seen here on the north bank. Between this place and the mouth of the Macmillan, the river becomes again rather strictly confined between ranges of hills, but just below its confluence it turns sharply to the north and is broken up into several channels among a number of low, wooded islands. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 127 The Macmillan and the Pelly valleys coalesce at an acute angle at Confluence of the rivers. the western point of the range of hills which alone has separated them for some distance, and the two streams must run nearly parallel for many miles above their junction. The Macmillan is bordered to the north by a well defined range of low mountains, which continues to the westward for about ten miles as the bordering range of the united streams. At the confluence, the Pelly appeared evidently to be some- what the larger river at the time of our visit, and it is probably so at all properly comparable stages of water. The inequality in size can not, however, be very great. The width of the Macmillan, just above its mouth, was found to be 455 feet, the rate of the current, which is greater than that of the Pelly at the same place, was 2 '7 miles per hour. The Macmillan water is much more turbid than that of the Pelly, and of a yellowish colour. The temperature of both rivers was identical on the ninth of August, being 54° F. It may probably be assumed from this circumstance that the origin of the rivers is similar, and that the Macmillan, like the Pelly, rises in or flows through considerable lakes, in which the water is warmed to a like extent. The suspended matter of the Macmillan may be entirely due to the washing away of silty banks, which is the usual cause of the turbidity of streams in this district. The upper part of the Mac- Macmillan millan has never been explored, but its size would intiicate that it may rise as far to the eastward as the Pelly, and probably, like it, in mountains representing the western ranges of the Rocky Mountains. We do not, however, know to what extent this river shares with the Stewart the drainage of the comparatively low country to the north- ward. I afterwards met a couple of miners (Messrs. Monroe ' _ ° St nation. tion and ice-rounded rock-surfaces met with on the Pelly was noted, and below this place similar evidences were found in a. number of localities. The direction of striation in ali cases closely accords with that of the main depression of the river-valley, though cutting across it obliquely where it turns north towards the Macmillan. The form of the surfaces is such as to show that the sense of the movement of ice was westward or down stream. Macmillan River to Lewes River. From the mouth of the Macmillan to the confluence of the Upper Pelly Tortuous part and Lewes rivers is a distance, in a straight line with a general bearing ° a few degrees south of west, of forty-six miles. A considerable portion of this part of the river is, however, extremely tortuous, and in con- sequence no doubt of the generally low character of the country, its flexures are not merely short, sharp bends from side to side of a trough- like valley, like those usually met with further up, but lax irregular curves of greater dimensions. The distance from the Macmillan to the mouth of the Lewes measured along the course of the stream is seventy-four miles. Four miles below the mouth of the Macmillan, on the north bank, is a small log- cabin, the first sign of habitation we had seen. We after- wards ascertained that two miners had lived here during the winter of 1886-87. At five miles and a-half below the Macmillan the Pelly was found to be 754 feet in width, with a current of 2-3 miles per hour ; a few miles below this the river turns south-westward and then nearly due south, entering Granite Canon at thirteen miles from Granite — — — . Canon. * (Upper Pelly, No. 57.) 9 130 GEOLOGICAL SUBVEV OF CAXADA. Tatlmain Lako. liOW country the Macmillan. The rocks met with in the cafion have ah-eady been alluded to. The cailon is about four miles iu length, with steep, rocky, scarped banks and cliffs, 200 to 250 feet in height. In the cafion are several little ivipids, bub the water is deep, and with the exception of some isolated rocks, the navigation would be quite safe for steamers, even at a low stage of water. As the river is much confined, howe\er, it is probable that pretty rough water may be found here during floods. Just beyond the canon, or sixteen miles and a-half below the Macmill!i.n, a small stream, about ten feet by three inches, enters from the south-eastward. The bed is wide, and it appears at seasons ot flood to become a foi-midable torrent. At thirty-six miles from the Macmillan another small stream was observ- ed on the south side, but with this exception, the river receives no fur- ther tributaries before meeting the Lewes. .Judging from Canapbell's sketch, this otherwise insignilicant stream must be that flowing from Tatlmain Lake, which was probably a source of supply of fish for Fort Selkirk while that post was maintained. After passing the ridge which is cut through by Granite Canon, the country on both sides of the river, for about fifteen miles, is quite low. No mountains or high hills are in sight on any bearing to the west- ward, and wide terraces run far back from the river at heights of 150 to 200 feet above it. These are often lightly wooded, largely with aspen, and are clothed with a good growth of grass presenting a very attractive appearance. The soil is good, and at the time of our Appearance of visit the country was very dry. Anemone patens was seen here for the first time since reaching the Yukon basin, and Eloeagnus argentea was also noted, both species characteristic of a rather dry climate. For the remaining distance to the mouth of the Lewes, the river is more closely bordered by low hills and ridges, which sel- dom exceed a height of 400 feet. At one place the stream is confined between high and somewhat rocky banks, but no rapid is met with. The southern slopes of the hills are generally open and grassy, and would afford excellent pasturage. The northern exposures are still pretty thickly wooded. Just above its confluence with the Lewes, the Pelly makes an abrupt turn to the south, and runs for several miles along the eastern base of a scarped cliif of basalt. From Granite Canon to the confluence, the current scarcely exceeds two miles and a-half an hour. For a number of miles below Granite Cafion, no rocks are met with along the river, but lower down there are frequent exposures, extending for some miles, of highly crystalline schists, which strike nearly east- and-west, parallel to the main direction of the river, with general the country. Rock exposurea, YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 131 high northward dips.. These rocks differ coasiderably from any before seen, including dark, wholly-crystalline schists, holding horn- blende or hornblende and mica. With these are associated consider- able exposui-es of a pecuUar dark-green chlorite rock, containing large mica crystals. This rock is interbedded with white, coarsely crystalline marble, in some places, and is quite evidently the alteration-product of some stratified material which has doubtless, however, been of volcanic origin. Xine miles above the confluence, by the course of the river, a great mass of impure serpentine comes out on the bank, and six miles Serpentine. and a-half above the same place, gray granite of the usual character is again met with,* and appears to constitute the hills to the east of the river for the remaining few miles of its course. The basalt, which forms a plateau from 200 to 350 feet in height in Basalt bluffs, the angle to the west of the Pelly at its junction with the Lewes, con- sists of several superposed flows, not always distinctly separable at a distance. It forms a mural cliff, with a long talus of angular frag- ments sloping down to the river-level, and though not very distinctly columnar where it fronts on the Pelly, becomes so below the con- tluence. The rock is brown in colour and often vesicular. The bedded silts were seen above Granite CaSon, but are scarcely Superficial represented below it. Boulder-clay was noted in one place below the ^^^^ cnnm, and the old stratified gravels in several places. The scarped banks along this lower part of the river are, however, chiefly composed of well rolled valley- or river-gravels, which become much more im- portant than before, and often constitute entire terraces up to forty feet in height. A remarkable layer of fine white volcanic ash, which overlies the Voloanio ash. •;laoial deposits all along the Upper Pelly valley, as well as in that of the ILiewes, has already been described on page 40. General- yoftg on the Upper Pelly River. The total lentrth of the Upper Pelly, following the course of the river, J'fPStl',™'' from the point where we first re.iched it at the west end or Oampbell s river. Portage to its confluence with the Lewes, is 320 miles. The elevation at the first-mentioned point isabout 2965 feet, that at the confluence 1555 feet, oiving a total fall of 1410 feet or 4-4 feet to the mile, a considerable portion of which, however, occurs in the numerous little rapids and riffles of its course. In Hoole Canon the estimated fall is about twenty feet. Two hundred and eighteen islands were counted in the ri\er, ' (Upper Pellv. No. 61.) 9* 132 GEOLOGICAL SURVEY OF CANADA. Navigability. Tributaries from the south Timber. without including such gravel bars as are submerged at high water, and are consequently without vegetation. The general course of the river is remarkably direct, and it embraces two main directions, the first bearing N. 55° W., the second, N. 87° W. These are parallel to the principal orographic features respectively of the upper and lower parts of the country traversed, and appear to indicate the main slopes of the general surface of the region. With the exception of Granite Canon, where warping might have to be resorted to at one place, the river would be easily navigable for stern-wheel steamers as far up as the mouth of the Macmillan, and the latter stream is also navigable for a considerable though unknown dis- tance. Above the Macmillan, I believe, no serious difficulty would be met with in taking a small stern-wheel steamer of good power up to the mouth of Ross River, and possibly as far as the foot of Hoole Canon. A line might have to be carried ashore at a few of the stronger rapids, but the chief difficulty to be encountered would be from shoal water at low stages. Where the river is widely spread and swift, a depth of three feet could scarcely be found across some of the gra^ elly bars. The Ross River is a navigable stream at its mouth, but its upper part is quite unknown. Hoole Canon is, of course, quite impassable for a steamer of any kind, and the rapid met with seventeen miles east of it, at the mouth of Hoole River, might prove to be a difficult one to surmount by warping, its fall being estimated at about eight feet. Above this point, the river is again, however, an easily navigable one for small steamers to the furthest point seen by us, and possibly as far as the lakes.* All the streams and small rivers flowing into the Pelly from the south and rising in or beyond the Pelly and Glenlyon mountains, are notably swift, and most of them are evidently subject to heavy freshets. Some notes have already been given respecting the character of the timber along the Upper Pelly. On the lower part of the river there are numerous groves on or not far from the banks, with good spruce up to two feet in diameter. Spruce of the same size is found also on the whole upper part of the river, but is relatively less abundant there. * Dall, in hia Alaska and its Resources, 1870, p. 278, gives a hypothetical itinerary of the Pelly River, making it head in Frances Lake. The Pelly, as shown on the map which accompanies the work, is evidently in part after Arrowsmith, and where it differs, (as in regard to the connection of Trances and Pinlayson lakes with the Pelly), becomes misleading. It is difficult to understand why this part of Arrowsmith's map, bearing intrinsic evidence of proximate accuracy, should not have been followed throughout in the compilation of later maps, particularly as the Pelly had not been further explored since Campbell's time. YUr:0\ DISTRICT AND NOKTHERN BRITISH COLUMBIA. 133 As in the case of the Upper Liard and Frances rivers, quartz, derived Quartz. from veins, is an abundant constituent of the gravel-bars of the Pelly, and numerous small quartz veins were observed in the rocks in many places. Where the granites are approached, the veins are found to cut all the rocks except these, and it appears that the development of the quartz veins is due to the same period of disturbance which has given rise to the uplift of the granite axes or their extrusion. Small " colours " of Gold gold may be found in almost any suitable locality along the river, and " heavy colours," in considerable number, were found by us as far up as the mouth of Hoole River, in the bottom of a gravel-bed there resting on the basalt. The river has been prospected to some extent by a few miners, but no mining of importance has yet been done on it. Thomas Bos well, whom we met on the Lewes, informed me that he had found and worked for a short time, a bar which paid at the rate of $18 per diem. This was on a tributary which, from his description, is probably identified as the Ross River. Two miners only, Messrs. Mon- roe and Langtry, were at work on this river in 1887, and their opera- tions were confined to the part below Granite Canon, where they made, on a couple of bars, from $10 to $20 per diem to the hand. The head- waters of the Macmillan and Ross, and those of the Pelly itself yet re- main unprospected, as well as the very numerous tributary streams of these rivers, in some of which "coarse" gold may yet be found. From observations subsequently made on the Lewes, coal may be Coal. looked for in the Laramie or Cretaceous rocks near the river at the locality described on page 124. Confluence or the Lewes and Upper Pelly. The country about the confluence, is generally speaking low, with Comparison of extensive terrace-flats running back to the bases of rounded hills and rido'es, of which none in sight probably exceed 1000 feet above the river. The moderate current which has been described ai characteristic of the Upper Pelly for some distance above the confluence, continues toits mouth, but the Lewes is much swifter, and though at the point of junc- tion divided among wooded islands, is evidently the larger stream, carrying a volume of water considerably greater than that of the Pelly, though probably less than twice as great (p. 15). It does not, however, necessarily follow from this that the Lewes is to be considered the principal head stream or continuation of the Yukon. The question of nomenclature and that of the position of the furthest sources of the Yukon, have, however, been referred to at greater length in another portion of this report, (p. 12). 134 GEOLOGICAL SURVEY OP CANADA. Volmne of united rivers. The water of the Lewes is of a bluish, slightly milky cast and is easily distinguished from the brownish muddy colour by which the Pelly is characterized below its junction with the Macmillan. The Temperature, temperature of the water in both rivers was found to be practically identical, on the 17th and 18th of August, at 7 p.m., being 59° F. The river below the confluence of the Pelly and Lewes averages about a quarter of a mile in width, and though its appearance is placid and there is no rough water, it is uniformly swift. An approximate cross-section made by Mr. Ogilvie, showed the river to be, at the point in question, 25-76 chains wide, from bank to bank. Of this width about two-thirds had an average depth of ten feet, with a surface velocity of four miles and three-quarters an hour. The remaining third was occupied by shallow bars and slack water. Using the same formula as before (see p. 95) the volume of the flow is found to be approximately 66,955 cubic feet per second. Traces of the passage of flood-water in the preceding spring were found on the river-flat, about ten feet above the low-water level, which, with a velocity no greater than that above stated and assuming the banks to be vertical, would give a flood discharge of at least 167,400 cubic feet per second. It may be, however, that the water-marks observed were caused by the damming back of the river by an ice-gorge. The ruins of Fort Selkirk, formerly a post of the Hudson's Bay Com- pany, stand on a partly open flat, on the south side, at a short distance back from the river, and about a mile and a-half below the confluence of the Pelly and Lewes. One chimney, built of basalt blocks which must have been brought across the river, and cemented with clay which has been baked almost into brick by the combustion of the ruins of the fort, still stands erect and uninjured. The lower part of a second is near it, and the fragments of several others strew the ground, which is partly overgrown by small aspens. These, and the traces of a couple of ex- cavations which have probably been cellars, are all that now remain to mark the site of the buildings which were pillaged by Indians from the coast in 1852. Euins of Fort Selkirk. DiSCOVEEY AND EXPLORATION OP THE LlARD AND PelLY. Hudson's Bay Fort Selkirk, of which the ruins alone now exist, was at one time exriorers' *^^ most important post of the Hudson's Bay Company to the west of the Rooky Mountains in the far north, and with the exception of Port Yukon, it was the farthest permanent post ever maintained by the Company to the north-west. On previous pages of this report, frequent mention has been made of the former establishments of the YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 135 (Jompany in the Liard and Yukon region, and as Fort Selkirk is the ]ast of these with which our route brought us in contact, a note as to the operations of the Hudson's Bay Company may appropriately be given here. When we call to mind that our knowledge of the geo- graphy of the region described in preceding pages has been up to the present time almost wholly due to the explorations carried out by the officers of the Company in connection with these establishments, it becomes evident that the history of these explorations and the facts respecting the several posts are well worthy of record. The circumstances attending the discovery of the Dease and Upper McLeod. Stikine have already been given on page 80. Mr. J. McLeod, whose explorations in that quarter are there referred to, appears in the same year in which he reached the Stikine (1834), to have ascended the Liard as far as Simpson Lake, and to have brought back the informa- tion according to which the river was represented on Arrowsmith's map of 1850. It was to the energy of Mr. Robert Campbell, however, that Campbell, the exploration of the Upper Liard and Yukon is almost entirely due. The only published account of Mr. Campbell's work, so far as I know, is that which appeared in the Royal Reader, Fifth Book, Toronto, 1883, p. 435, and which was reprinted, with slight alterations at Winnipeg in 1885, as a small pamphlet entitled " Discovery and Exploration of the Youcon (Pelly) River.'' From this source and from additional facts furnished by Mr. Campbell in answer to questions addressed to him, as well as from allusions in the unpublished journals of Chief Factor James Anderson, which have kindly been placed at my disposal by his son, the following brief account is drawn up.^ — After the abandonment of Dease Lake post in 1839, Mr. Campbell was, in the spring of 1840, commissioned by Sir George Simpson to explore the " north branch " of the Liard to its source, and to cross the height-of-land in search of any river flowing to the westward, especially the head waters of the Colville, the mouth of which on the Arctic Ocean had recently been discovered by Messrs. Dease and Simpson. Mr. Campbell writes : — " In pursuance of these instructions, I left Expedition to Fort Halkett [on the Lower Liard | in May, with a canoe and seven ^ ^ y- men, among them my trusty Indians, Lapie and Kitza, and the inter- preter, Hoole. After ascending the stream some hundreds of miles, far into the mountains, we entered a beautiful lake, which I named Frances Lakej in honour of Lady Simpson. * * * Leaving the canoe and part of the crew near the south-west [sic] extremity cf this fthe west] branch of the lake, I set out with three Indians and the inter- preter. Shouldering our blankets and guns, we ascended the valley of it river', which we traced to its source in a lake ten miles long, which, 136 GEOLOGICAL SURVEY OF CANADA. Establish- ment of Fort Frances. Exploration of the Pelly. wi(;h the river, I named Finlayson's Lake and River." From this point, Mr. Campbell struck across to the Pelly, which he then named in honour of Sir H. Pelly, a Governor of the Company. " After reaching the actual bank of the river, we constructed a raft, on which we embarked and drifted down a few miles on the bosom of the stream, and at parting we cast in a, sealed tin can, with memo- randa of our discovery, the date, etc." During Campbell's absence the remainder of the party built a house at the point between the two arms of the lake, which was then named " Glenlyon House," but was afterwards known as Frances Lake House or Fort Frances. Returning down the river, they met a trading outfit which had been despatched for them, at Fort Halkett, and turned back with it to Frances Lake, after sending out a report of their proceed- ings. The Company now resolved to follow up those western discoveries, and in 1842, birch bark, for the construction of a large canoe to be used in exploring the Pelly, was sent up from Fort Liard. In the same year Fort Pelly Banks was constructed, or its construction begun, and early in June, 1843, Campbell left that place in the canoe which had been made, accompanied by Hoole, two French Canadians and three Indians. They saw only one family of Indians ("Knife Indians") till they readied the mouth of the river which Campbell called the Lewes. Here was a large camp of " Wood Indians," and these, nfter recovering from their surprise at the sight of the party, so discouraged Campbell's men by their stories of the number and ferocity of the people on the lower river, that he was obliged to turn back. 'Pelly Banks'' For some years afterwards the operations of the Company did not extend beyond "Pelly Banks," though during the summer, hunting parties were sent down the Pelly to collect provisions, and in that way information was received respecting the river and the Indians inhabit- ing its vicinity. In the winter of 1847-48 boats were built at Pelly Banks, and early in June following Campbell set out to establish a fort at the confluence of the Pelly and Lewes rivers. This was named Fort Selkirk, and was at first situated on the extreme point of land between the two rivers, but this point being found subject to floods during the disrup- tion of the ice, the post was in the spring of 1852 moved to a site a short way below the mouth of the Lewes, on the left bank. The inner work of the new buildings was still unfinished at the time of the Indian raid, noticed further on. Establish- ment of Fort Selkirk. YUKON DISTRICT AND NOETHEEN BRITISH COLUMBIA. 137 ■Meanwhile an entry was being made into the Yukon basin from Explorationof another direction. Mr. J. Bell had already, in 184:2, reached the Porcu- R°ver.'^™ pine or Rat River, and had descended it for three days' Journey. He was in 1846 in charge of the Hudson's Bay post on Peel River, near the mouth of the Mackenzie, and was instructed again to cross the mountains and to further explore the Porcupine River. In pursuance of these instructions, he in that year reached the mouth of the Porcu- pine and saw the great river into which it flows, which the Indians informed him was named the Yukon. In 1847 Fort Yukon was estab- lished at the mouth of the Porcupine by Mr. A. H. Murray. It still lemained, however, for Campbell, in 18-50, to prove that the Identity of Pelly and Yukon were identical. This he did by descending the river Yukon from Fort Selkirk, to Fort Yukon, after which he ascended the Porcu- ^^°™- pine, crossed the mountain portage, and returned to Fort Simpson by the Mackenzie.* One result of this journey was to show that the route from Fort Selkirk by way of the Porcupine River to the Mackenzie was preferable to that originally discovered. The navigation of the Liard was both arduous and dangerous and several lives had been lost in boating on that stream. Added to this was the length of the land transport from Frances Lake to the Upper Pelly and the fact that great diificulty had been found in maintaining the posts in that district. In 1849, the post at Pelly Banks, with the exception of the men's Abandonment house, was accidently burnt. In 1850 it was finally abandoned, and °f P™*^- in the spring of 1851 Fort Frances was likewise abandoned, t The abandonment of these posts was not due to any hostility of the natives, who were on the contrary most friendly, but in consequence of the circumstances above noted, and the fact that while these establishments were very expensive to maintain, they merely bought furs which would otherwise have been carried by the Indians them- * ilr. Campbell states that when again on his way down the river from Fort Sel- kirk to Fort Yukon, in 1S51, he found a great number of the Indians had been carried off during the previous winter by some virulent disease. He has further informed me, in answer to my inquiries on the subject, that the Stewart River was so named after his " dear and gallant friend and assistant-clerk, James G-. Stewart, son of the late Hon. John Stewart, of Quebec." Stewart was sent out in the winter of 1849 to follow the Indian hunters in quest of meat. He found them some distance north of this river, which he ci-ossed on the ice. White River, Mr. Campbell named on account of its milky colour. Of the other streams entering between Forts Selkirk and Yukon he says, " Antoine River " was named after the interpreter at one time at Fort Yukon, a son of his interpreter Hoole ; "Forcier River," after his guide, Baptiste Forcier; "Lolique River," for Forcier's wife ; and " Ayonie's River," below the White River, was named after the natives of that quarter. t Forts Frances and Pelly Banks are erroneously stated in Call's Alaska and its Resources, to have been burnt and pillaged, p. 115, foot-note and p. 508. 138 GEOLOGICAL SURVEY OF CANADA. selves to other posts, if these particular, and to them more conve- nient ones, had not been in existence. J""E''°" °f The several ruined chimneys of Fort Selkirk still to be seen, with other traces on the ground, are in themselves evidence of the impor- tant dimensions and careful construction of this post. The establish- ment consisted, I believe, in 1852, of one senior and one junior clerk and eight men. The existence of this post in tlie centre of the inland or "Wood Indian" country had, however, very seriously interfered with a lucrative and usurious trade which the Chilkoot and Chilkat Indians of Lynn Canal, on the coast, had long been accustomed to carry on with these people ; acting as intermediaries between them and the white traders on the Pacific and holding the passes at the head-waters of the Lewes with all the spirit of robber barons of old. In 1852, rumours were current that these people meditated a raid upon the post, in consequence of which the friendly local Indians staid by it nearly all summer, of their own accord. It so happened, how- ever, that they absented themselves for a couple of days and at that unlucky moment the Coast Indians arrived. The post was unguarded by a stockade and yielding to sheer force of numbers the occupants were expelled and the pi ice was pillaged, on the 21st August. Two days afterward Campbell, having found the local Indians, returned with them and surrounded the post, but the robbers had flown. Being now without means of support for the winter, Campbell set oif down stream to meet Mr. Stewart and the men who were on the way back from Fort Yukon. He met them at the mouth of White River, and after turning them back with instructions to arrange for wintering at Fort Yukon, set out himself in a small canoe up the Pelly River, crossed to Frances Lake, descended the Liard and arrived at Fort Simpson with the tidings of the disaster, amid drifting ice, on the 21st of October. Being anxious to obtain Sir George Simpson's permission to re- establish Fort Selkirk, Campbell waited only till the river froze, when he left Fort Simpson on snow-shoes and travelled overland to Crow Wing in Minnesota, where he arrived on the 13th of March. On the 18th of April he reached Lojidon but was unable to obtain from the directors of the company the perrnission he desired. A short account of this remarkable journey appeared in the Perthshire Advertiser and Inverness Courier, but I do not know the precise date of the publication. In the autumn of 1853, one of Campbell's hunters arrived at Fort Halkett on the Lower Liard by way of the Pelly and Frances. This is the last traverse of Campbell's Portage of which I can iind any Campbell".^ iourney of 1852. News from Fort Selkirli YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 139 record, though it may doubtless have been used by the Indians subse- quently. Prom this man it was learnt that the buildings at Fort Selkirk had been all but demolished by the local Indians for the pur- pose of getting the iron-work and the nails. He also stated that the Chilkats, being unable to carry away all their plunder in the pre- ceding year, had taken merely the guns, powder and tobacco. They had cached the heavier goods, which were afterwards found and appro- priated by the local or wood Indians. At a later date the ruins of the post must have been burnt, as their present appearance indicates.* Fort Yukon, at the mouth of the Porcupine, was continuously main- The United tained till 1869, when the Hudson's Bay Company was expelled by the lodf YTkon. United States Government as represented by Capt. Charles W. Ray- mond, Corps of Engineers, U. S. Army ; he having ascertained by astronomical observations that the post was situated to the west of the 141st meridian. He describes his proceedings as follows. — "On the 9th of August, at 12 m., I notified the representative of the Hudson's Bay Company that the station is in the territory of the United States ; that the introduction of trading goods, or any trade by foreigners with the natives, is illegal, and must cease ; and that the Hudson's Bay Company must vacate the buildings as soon as practi- cable. I then took possession of the buildings and raised the flag of the United States over the fort." f The fort was afterwards aban- doned and allowed to go to ruin. The utmost credit must be accorded to the pioneers of the Hudson's Pioneers of the Bay Company for the enterprise displayed by them in cariying their Company. "^ trade into the Yukon basin in the face of difficulties so great and at such an immense distance from their base of supplies. To explorations of this kind, performed in the ser\ice of commerce, unostentatiously and as matters of simple duty by such men as Mackenzie, Fraser, Thompson and Campbell, we owe the discovery of our great north-west country. Their journeys were not marked by incidents of conflict or bloodshed, but v.'ere accomplished on the contrary with the friendly assistance and co-operation of the natives. Less resolute men would scarcely have entertained the idea of utilizing, as an avenue of trade, a river so perilous of navigation as the Liard had proved to be when explored. So long, however, as this appeared to be the most practicable route to the country beyond the mountains, its abandonment was not even contemplated. JS either distance nor danger * Of Reid House, shown on Arrowsmith's map of 1854, near the Stewart River and to the north of Fort Selkirk, I have been unable to learn anything. Mr. Campbell never heard of it, and if it had any existence it was probably a temporary outpost of Fort Yukon. t Report of a Reconnaissance of the Yukon River, 1871, p. 16. 140 GEOLOGICAL SURVEY OP CANADA. appear to have been taken into account, and in spite of every obstacle a way was opened and a series of posts established, extending from Fort Simpson on the Mackenzie to Fort Yukon. Fort Simpson may itself be regarded, even at the present day, as a post very far removed from the borders of civilization, but this further route, which nearly half a century ago became familiar to the Company's voyageurs, stretched out beyond it for over a thousand miles. Mr. James Dangers of the Anderson, in 1853, writes thus of the Liard River : " You can hardly conceive the intense horror the men have to go up to Frances Lake. They invariably on re-hiring endeavour to be exempted from the West Branch [Liard], The number of deaths which have occurred there is fourteen, viz., three in connection with Dease Lake and eleven in con- nection with Frances Lake and Pelly Banks, of these last, three died from starvation and eight from drowning." The"returns.'' At the time of the establishment of Forts Yukon and Selkirk, and for many years afterwards, the " returns '' from these furthest stations reached the market only after seven years, the course of trade being as follows: Goods. — 1st year, reach York Factory; 2nd year, Norway House ; 3rd year. Peel River, and were hauled during the winter across the mountains to La Pierre's House, 4th year, reach Fort Yukon. Retarns. — 5th year, reach La Pierre's House and are hauled across to Peel River ; 6th year, reach depot at Fort Simpson ; 7th year reach market. Lewes River. Moetinp with Mr. Ogi) vie. Our proceedings at the mouth of Lewes River and our meeting there with Mr. Ogilvie have already been noted. We finally left the conflu- ence and began the ascent of the Lewes on the 18th of August, and arrived at Lake Lindeman, where the portage to the coast begins, on September 16th. We were during this time on the one travelled route of the country, and every few days fell in with small parties of miners, generally on their way out, up the river. A few men were still found working on bars, and six or eight passed down stream with the pur- pose of wintering at or near Forty-mile Creek. Though my own observations did not extend below the month of the Lewes, Mr. W. Ogilvie sent out by the last party of miners met by him on their way to the coast, a small collection of rocks from places further down the river, as far as Forty-mile Creek. The general character of these has already been noted in the introductory part of this report (p. 32). They are not further described here, as they will be referred to by Mr. McConntll in connection with his geological examination of the portion of the river from which they were derived. YUKON DISTRICT AND NOETHERN BRITISH COLUMBIA. 141 The Lewes River* was discovered and named by Mr. Campbell in Discovery of 1842, as already stated. It is indicated in an approximate manner, according to information supplied by this gentleman, on Arrowsmith's map of 1854. Mr. Campbell informs me that he was well aware of the existence at its head of a portage to the sea by which the Chilkat Indians came inland to trade. This route he had the intention of exploring, but the question of ..tion of Lake Marsh. Lake Marsh, so named by Schwatka, in honour of Prof. 0. 0. Marsh, is known to the miners as " Mud Lake." It is twenty- miles in length, with an average width of about two miles, pretty uniformly maintained. The valley of which this lake occupies the centre, is notably wide, and YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 163 the country in the immediate vicinity of the lake in quite low, consist- ing of terrace-flats, or low rounded or wooded hills and ridges. Con- spicuous mountain summits, however, occur at a distance of some miles Mountains. inland on both sides of the lake. A moderately well defined range, of which Michie Mountain* 5540 feet in height is the most elevated point, bounds the view on the east side of the lake, from which it diverges in a south-easterly direction. To the west is an irregular and broken mass of mountains in which several notable gaps occur and which occupy the countrj- between Lake ]\Iarsh and the Watson valley, previously referred to. The highest points of these, Mounts Lome and Lansdovune, were ascertained to have approximate elevations of 6400 and 6140 feet respectively. The diversified forms of the mountain.s in view from this lake render it particularly picturesque, and at the time of our visit, on the 10th and 11th of September, the autumn tints of the aspens and other deciduous trees and shrubs, mingled with the sombre greens of the spruces and pines, added to its beauty. The shores of the lake are generally rather shoal, and in some places The lake and the beach-gravel was found to rest on mud or clay, but these peculiari- ties are not so striking in themselves as to warrant the application of the name " Mud Lake.'' The mouth of the M'Clintock River, which enters the lower end of the lake from an important valley, was not examined closely. It appeared, however, to be a small stream, and the greater part of the country to the east of the lake probably drains to the Tes-lin-too. A second stream joins the lake at the south-east angle at its head. As in the case of Lake Labarge, Mr. Ogilvie's measured line was Track-survey, here carried along the west shore, and in order to complete the outline of the lake, I travelled up the opposite side, making a track-survey of it, which has been embodied on the map. The rocks seen at the lower end of Lake Marsh, and on the island, Rocks. may be taken as representing the composition of the range which borders the Lewes on the north-east for some distance below, They consist of greenish and purplish altered volcanic materials, probably all diabase in composition, and are bedded, the strike being approxim- ately parallel to the run of the range just mentioned. Further up the lake on the same side, similar rocks were noted in several places, together with some which might be taken to represent the grauwacke-sandstones of the newer series described on Lake Labarge. The higher mountains off to the eastward appear to be, at least in part, composed of limestone. At the head of the lake are considerable exposures of black and dark- * So named by Schwatka. 164 GEOLOGICAL SUEVKY OF CANADA. gray, hard, slaty argillilcs, finely cleaved and tra\'ersod by mimorous small quartz veins. The strike of the cleavage is S. 30" E., with a high easterly dip. Glaoiation. The rock surfaces along the lake are heavily glaciated, the directinn being, in general, parallel to that of the main valley. The evidence is such as to show, however, that at least the subjacent part of the ito of the former great glacier, bifurcated at the north end uf the lake, one branch taking the valley of the Lewes, the other that uf the M'Glintock River. In the same way, at the south end of the lake, a great jiart of the ice has been delivered through the wide valley which conies from the south-east between Jubilee Mountain* and Mount White, j Tagish, Bennett and Lindeman Lakes. Navigable waters. Conneoting The upper end of Lake Marsh is connected with Tagish Lake by a. wide tranquil reach of river five miles in length. The current is here very slack, and the depth, according to Ogilvie, from six to twelve feet. The river is bordered by low terraces, which are particularly wide on the west side, and are covered with open woods, chiefly consisting of white spruce and cottonwood. To the east, the long irregular ridges and slopes which culminate in Jubilee Mountain begin to rise u, short Indian houses, distance back from the river. A mile above Lake Marsh, on the east bank of the river, are two roughly built houses belonging to the Tagish Indians. These are the only permanent houses seen along the whole course of the Lewes, and here the Tagish people who roam over this part of the country, reside during the winter months. From the description just given, it will be seen that the navigation, by steamers, from the head of the caflon through Lake Marsh and to Tagish Lake would offer no difficulties, while the tranquil character of the connecting river between the two lakes last mentioned, is such as to practically render Lake Marsh the lower portion of an extensive system of still-water navigation which includes not only Tagish Lake but also Lake Nares, Lake Hennett and possibly other connected waters and which will prove of the greatest utility at no distant date in faci- litating the opening up and development of the mineral resources of the tract of country in their vicinity. The through distance, from the lower end of Lake Marsh to the head of Lake Bennett, measured along the central line of the various * 1887 being the year of Her Majesty's jubilee. t So named in honour of the late Hon. Thos. White, to whose initiative the despatch of the expedition to which this report refers was largely due. X X < X K ^ << » 5 <; 03 YUKON DISTRICT ASD NORTHERN BRITISH COLUMBIA. 165 lakes and connecting waters, is about seventy miles, made up as follows. — Marsh Lake 20 miles. River 5 " Tagish Lake ICi-G " Lake Nares 27 " Bennett Lake 25'8 " Total 70'1 The still-water navigation, however, includes also the West Arm of Bennett Lake, about twelve miles ; Windy Arm of Tagish Lake, eleven miles ; Tako Arm of same lake, (of which some notes are given below) at least twenty miles ; making a grand total of at least one hundred and thirteen miles. Taken as a whole, these lakes constitute a singularly picturesque ggenery region, abounding in striking points of view and in landscapes pleasing in their variety, or grand and impressive in their combination of rugged mountain forms. The elevation of this remarkable system of lakes above the level of Height above the sea, as approximately ascertained by the mean of about eighteen ^'^'*"1®'^^'' barometric observations extending over several days, (all taken in triplicate on as many different aneroids,) is 2150 feet. The mean of the barometer readings was compared with that of those taken during the same days at Haines' mission, on Lynn Canal, and the resulting height as above given, is probably a near approach to the truth. The inner or north-eastern edge of the Coast Ranges is not here p •!• very well defined, but Tagish and Bennett lakes, with their several relatively to arms, may be described as lying upon this border and as in part pene- trating the outskirts of the range. The lower part of Tagish Lake occupies the continuation of the same wide valley in which Lake Marsh lieSj and the valley of the Tako Arm may also be included as * part of the same depression. To the west of this, the upper part of Tagish Lake and Bennett Lake must be considered as lying among the mountains of the Coast Ranges, and the height as well as the abrupt and rugged character of the mountains increase in that direction, their slopes and summits holding large areas of permanent snow, even late in the summer. The lower part of Tagish Lake is very generally bordered by terrace- q -..(.jn- _« flats or by low land of the nature of terraces, and the valley runs the lakes, through to the lower end of Bennett Lake, with a nearly uniform width, as measured between the bases of the mountains, though the lake is somewhat constricted between Tako and Windy arms by the extension of the low land from the north side. A similar projection 166 GEOLOGICAL SURVEY OF CANADA. of low alluvial land separates Lake Nares from the west end of Tagish Lake, a river-like current being distinctly observable in the narrows. The same occurs at the narrows known as the Caribou Crossing which separates Nares from Bennett Lake, but here with the addition of a number of irregular sand-hills, with which the low land in question is covered, and which also extend round a considerable part of the north end of Bennett Lake. Anns of the The Windy Arm of Tagish Lake, together with the upper part of ^^' Bennett Lake and its West Arm are comparatively narrow mountain- walled inlets, with all the characters of true fiords. The Windy Arm terminates in low, hilly land, from which a couple of small brooks come, but no important stream ; and it would appear that the drainage which might be expected to enter here is cut off by a transverse valley which holds a lake and flows out towards the Tako Arm. The West Arm of Bennett Lake was not explored to the head, but terminates in a mass of wild, rugged and bare mountains, of which the outline sketch given on the opposite page may give some idea. The upper part of the main lake lies, as a narrow water-way, between beetling granite ranges which rise almost perpendicularly to heights of 3000 and 4000 feet above it. Many of the summits beyond the heads of these fiords and in the vicinity, attain heights of 6000 to 7000 feet above the sea, and the region is in every sense an alpine one, though no dominating peaks of great altitude occur. Ulimate and In consequence of the position of this country, in the lee of the higher ' crests of the Coast Ranges, and notwithstanding its considerable altitude, the climate appears to be equally dry with that met with about the side of old Fort Selkirk, and no very striking difference is observed in the character of the vegetation. The southward- facing slopes of some of the mountains, to a height of a thousand feet or more above the lake, are grassy and open, a circumstance particularly observable on the north side of the west part of Tagish Lake and on Lake Nares. The Amelanchier was seen on the lower part pf Tagish Lake, this being the most northern station observed, and Anemone patens continues abundant, locally. Somewhat more alpine conditions are, however, indicated by the occurrence of Picea suhalpina, which wajS noted as abundant at the water-level on Windy Arm, and on the upper part of Bennett Lake forms entire groves, growing to a con- siderable diameter, but tapering rapidly and very rough and knotty. Several places were, however, noticed where miners had built boats on Bennett Lake, and it was here that Mr. Ogilvie constructed a large scow for the transport of his provisions and bulky freight down the river. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 167 /('■ \'' mil ■AV.08-K 0^ Eh O o ■JA oS-Si -S Hi: §^ •AV =.£9 'S •AVoS-9S'S /1/,-ruf r iJ s. Eh O S -to ■A\ 08^ 'S ■JA 08S 'S /S" %^ll •^UIOj, ^3(S[a!A8[3Jt difficulty which presented itself. The natives along this part of the Indians along Laird are very inferior canoemen, and are afraid to venture on the "^'"^' river except in its smoothest portions. They seldom ascend the river with canoes, as they prefer carrying their outfit along the shore to tracking a boat against the rapid current, and in descending use small spruce bark canoes, which they can build in a couple of hours and abandon without much loss. They belong to the Kas-ka branch of the great Tinneh family, and are commonly referred to as the Grand Lake Indians. '■ Bulletin de la Societe de Geographie, Paris, 1875. u GEOLOGICAL SURVEY OP CANADA. Bordering forests. Hyland River. Width of river The Liard River below the mouth of the Dease has a general width and valley. ^f fj-Q^j ^wo hundred and fifty to four hundred yards, but widens out in places to over half a mile, and a current of four miles and a-half an hour. It separates in places into a number of channels inclosing low alluvial islands usually well wooded. Its valley is from two to three miles wide, and is shallow with rolling banks sloping easily up to the general level. The country is everywhere well wooded, but the trees are usually small, seldom exceeding a foot in diameter. The principal varieties observed were the black pine (Pinus Murrayana), the White spruce, {Picea alba), and the smooth and rough barked poplars, (Populus tremuloides and halsamifera. ) Besides these, groves of larch were occasionally seen, and some varieties of willow and alder. Twelve miles below the mouth of the Dease the Liard receives a large tribu- tary from the north. This stream is over a hundred yards wide, and is called the Highland [Hyland] River, after a prospector who ascended it for some distance. On the older maps it is named the McPherson River. In the direction from which it comes are some low hills at a distance of four to five miles, behind which appear snowy mountains. From the mouth of Dease River, the Liard runs in a general E.S.E. direction for eighteen miles, and then making a sharp bend to the north, runs for about the same distance in a direction nearly at right angles to its former course. Below the bend the I'iver, for twelve miles is wide and filled with islands, after which it gradually decreases in width, and this, with the steeper slopes of the valley and the increas- ing strength of the current, which occasionally breaking into small riffles now hurries along at the rate of seven miles an hour, all afibrd Signs of our approach to the Little CaiLon, of the dangerous character of which we had been previously warned, and which soon comes into view. The Little Canon is about half a mile long, and in its narrowest place about two hundred feet wide. It is easily navigable in low water, but is dangerous for small boats during flood, as the channel is very crooked, and the current striking with great violence against the right hand bank is thrown forcibly back, with the production of a number of breakers running nearly lengthwise with the direction of the channel, and large enough to swamp any ordinary river-boat which is drawn among them. A number of Chinamen were drowned at this point some years ago. This cufion can be run with safety by entering it nearly in the middle of the stream, which is as close to the left hand bank as the lines of reefs and isolated rocks running out from that side will allow, and once past these making all haste to the left so as to Little Canon. YUKON DISTRICT AND NORTHERN BRITISH COLUMlilA. 189 clear the breakers below. In high water the rapid can be avoided by making a portage of about half a mile along the right bank. Rock-exposures are infrequent in the valley of the Liard between Rooks between the mouth of the Dease and the Little Canon, but those observed will -LftUe ^Cauori be described in order. Three miles below the mouth of the Dease is a small exposure of somewhat soft dark shales associated with friable sandstones and conglomerates. A second exposure of the same beds was observed about a mile farther down the river, where they dip N. 60° W. < 20°, beyond which they disappear. These rocks are unfossiliferous, but from their lithological characters, and the fact that they overlie unconformably the hard quartzites, slates and limestones, of the neighbourhood, were referred to the Tertiary. At the mouth of Highland River, on a small island, is an exposure of hard whitish sandstone, passing into quartzite. This rock weathers yellow, and dips N. 50° E. < 50°. Six miles farther down, at a bend which the river makes to the north, is a cut-bank showing unconsolidated sands, sandy clays and gravels, and holding some small beds of impure lig- nite. Below this, with the exception of rolled ri\er gravels, no further exposures were seen until near tlie Little Cation, when black shales appear in a couple of places. The rocks in the Little Canon consist of dark and sometimes cleaved Rocks in Little shales, holding large flattened ironstone nodules, hard sandstones and CI^iJ""- quartzites, and some beds of fine-grained, hard siliceous conglomerate. They are closely folded together and strike N. 35° W. No fossils were found in any of these beds, nor any definite evidence of their age obtained, beyond the fact that the shales have a close lithological resem- blance, both in appearance and composition, with those on Dease River, from which Dr. Dawson obtained graptolites of Utica-Trenton age. Below the Little Canon the river widens out to over half a mile, and character of the steep, rocky banks are replaced by easier slopes of gravel and sand. £J't^t^e'^c°,^o„. These continue for three miles, after which shaUs and sandstones reappear in the bank, and their confining influence is immediately seen in the rapid contraction of the stream and the formation of a second nar- rows. These sudden dilatations and contractions constitute one of the most characteristic features of the Liard, and are an indication of the hetero"eneity of the formations through which it cuts. Through the de- file just mentioned the stream rushes wi(h great velocity, but with an even current until near its foot, where it is forced between two points of rock scarcely a hundred feet apart, which project into the stream from either bank and determine the formation of two rapidly gyrating and dangerous-looking whirlpools. These can be avoided, if necessary, by 190 GEOLOGICAL SURVEY OF CANADA. Shales, sand- stones, etc., replaced by limestones. Picturesque valley. Porcupine Bar. Appearance of country. making a portage of a few feet across one of the points. In ordinary stages of the water, however, they can be run without difficulty. At this point the shales, sandstones and conglomerates which have been exposed at intervals all the way from the mouth of the Dease, are replaced by shaly limestone and soon afterwards by more massive varieties of the same rock. Beyond the narrows, the river at once resumes its ordinary dimen- sions, and rushing rapidly around a short bend, enters one of the most picturesque portions of the valley of the Liard. The river here aver- ages about three hundred yards in width, and glides along with a strong even current of about five miles an hour. It is narrowly confined by sloping banks, which follow closely all the bendings of the stream, without any intervening flats, or, except at low water, any disfiguring bars and beaches. The valley is everywhere densely wooded with ever- greens, aspens, birch and alder, the changing greens of which are agree- ably relieved at intervals by gray limestone cliffs, which rise steeply from the water's edge, and ruffle the surface of the otherwise glassy streaiu. Eight miles below the entrance to this portion of the river is situated Porcupine Bar, once the scene of active mining operations, but now worked out and abandoned. Opposite to it is a range of low hills, at the base of which I camped somewhat early on the afternoon of the 28th, for the purpose of ascending them and so obtaining a view of the surrounding country. These hills extend in an irregular manner for some miles along the left bank of the river, but appear to have no definite trend. They have the rounded outlines that characterize all glaciated districts, but no striae were anywhere observed. They are composed of lime- stone and have an altitude of 1500 feet above the river, pr one thousand feet above the general plateau-level. From their base stretches, in all directions, an irregular rolling plateau, broken here and there by ranges of low hills and dotted with innumerable small lakes and marshes. To the south the horizon is broken by the serrated crests and jagged summits of the Cassiar Range, one prominent peak bearing S. 20° W. In a direction N. 25° W., at a distance of twelve to fifteen miles, are some low hills still covered with streaks of snow, while a range of partially snow-clad hills was also seen at S. 26° E. The plateau is everywhere densely wooded, the principal trees observed being the white spruce, the black pine, the larch, the rough and smooth barked poplars, the birch, and species of alder and willow. Of these the spruce, which attains here a diameter of fifteen to twenty inches, is by far the most abundant and valuable. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 191 The limestone of which the hills are formed is usually grayish in character of colour and rather compact, but passes in many places into a whitish limestone. highly crystalline variety without distinct bedding. It has a general strike of N. 15° W. It is destitute of determinable fossils, but holds fragments of crinoid stems, and traces of brachiopods and trilobites. Prom Porcupine Bar the river runs S. S. B. for some miles, and then i^yer below bending more to the south, preserves a general southerly direction for Porcupine ten or twelve miles, when it is closed in by a nameless canon. In this reach it presents much the same features as those noted above, but its valley is somewhat wider, and is now bottomed by long narrow well wooded flats. Also the bed of the river becomes considerably enlarged in places and divides around a number of alluvial islands. The banks of the valley have an average height of five hundred feet and show frequent exposures of whitish coarsely crystalline, grayish, fine- grained limestone, striking a few degrees west of north and dipping at all angles. Four miles and a-half below Porcupine Bar, is Bed-rock Bar, now like the former deserted. Eight miles farther down is another abandoned miner's camp now represented by a single log hut. Passing this, the river bends more to the north and divides around a rocky island, on either side of which is a short rifiie, then runs with an even current of about four miles an hour to the canon mentioned above. This canon is scarcely a hundred yards in length, and is bounded by precipitous limestone cliffs about one hundred and fifty feet apart. It presents no obstacle to navigation. Immediately below the canon the ' river dilates for some distance into a large island-filled basin, beyond which it contracts again to its ordinary width of three or four hundred yards, and runs with a swift even current in an easterly direction for five miles, when its course is interrupted Ify the Cranberry Rapids. The limestones seen along this part of the river are often coarsely character of crystalline and cut up by white calc-spar veins. Other varieties show limestone, wavy lines projecting from weathered surfaces, due to alternating mag- nesian and calcareous layers, and closely resemble in this respect the limestones of the Castle Mountain group as developed along the Bow River Pass. In some places the limestone becomes shaly and impure, and is altered into an imperfectly developed schist. Two miles above Cranberry Rapids, the limestones are replaced by LimestoneB shales, sandstones and conglomerates, and a change is immediately replaced by- noticeable in the character of the stream. The declivity is greatly increased, and for the next fifty miles rapids are of constant occur- rence. The rough water at Cranberry portage has a total length of a mile Rapids at and a-half, but there is a reach of comparatively undisturbed water Ymlagr^ 192 GEOLOGICAL SURVEY OP CANADA. Kocks at Cranberry Portage. Mud Riv«r. about half way down. The upper part of the rapid is exceedingly- wild, as the bed of the river is filled with huge angular masses of rocks against which the current breaks with frightful violence. No part of the channel is clear, and a glance at the forbidding array of foaming breakers and whirling eddies showed at once the utter hopelessness of any attempt to run it with our small, heavily laden boat. We passed it by portaging our outfit about half a mile along the right bank, and then dropping the empty boat with a rope, and at the worst places dragging it across points of rock. The lower part of the rapid is not so rough, and we managed to work our boat through this without unloading. The rocks at the Cranberry Rapids, consist of shales, limestones and conglomerates, closely resembling those observed at the Little Canon and evidently of the same age. The shales are dark, finely laminated, and occasionally show well marked cleavage lines. They are inter- stratified in places with the sandstones, and often inclose flattened ironstone concretions, a number of which were broken in a resultless search for fossils. The sandstones, lighter coloured than the shales, are hard and often pass into quartzites. The conglomerates are very fine-grained, and consist principally of white, well rounded quartz pebbles imbedded in a siliceous matrix. The shales and accompanying beds are broken through and altered to some extent by a series of dykes. They have a general easterly strike, and lie at all angles from horizontal to vertical. On the opposite side of the river, an exposure of soft shales and conglomerates, resting unconformably on the beds just described, and evidently of Tertiary age, was observed, but was not closely examined, owing to the impossibility of getting across to it. Below the rapids the river hurries on with a smoother surface but with scarcely diminished velocity, its strength being shown by the way it foams around the occasional rocky obstructions in its course. It is bordered for some distance by large eddies, between which and the downward current are long lines of dangerous looking whirlpools. Further down is a long but easily navigable riffle, beyond which, with the exception of an occasional rock, the channel remains comparatively clear until the stream variously known as Black, Mud, or Turnagain River is reached. This stream, although one of the principal affluents of the Liard, is at present almost unknown. It originates near the Finlay branch of the Peace River, and joins the Liard after a course of about two hundred and fifty, miles. At its mouth it is over one hundred and twenty yards wide. Eighty miles above its mouth is situated a small trading post, built some years ago by Mr. Rufus Sylvester, but now YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 193 m the possession of the Hudson's Bay Co. This post is connected with the central post at the mouth of McDame Creek on Dease River by a pack trail about seventy miles long. From Mud River the Liard bends more to the north, and, still run- Mountain ning with great rapidity and breaking into occasional riffles, reaches, ^apS^* in a couple of miles, the Mountain Portage Rapids, one of the worst rapids met with on the trip. The river here falls over a band of Shales irregularly hardened by a system of dykes and worn into a succession of ridges and hollows, and the roughened surface thus pro- duced throws the hurrying waters into an indescribable turmoil. We landed at the head of the rapids on the right bank, and were forced to spend a day in making a difficult portage of about half a mile with both boat and outfit. I afterwards learned from W. Lupine, an old Hudson's Bay voyageur, that we might have avoided this portage if we had landed on the left hand side, as what we supposed was the mainland was really an island, behind which a small channel existed which can be run with comparative safety. Half a mile below Mountain Portage, Rabbit River comes in from Rabbit River, the south. This stream is about two hundred feet wide, and brings in a large volume of clear water. At its mouth is a large auriferous bar, Gold mining, which has evidently been worked by placer miners to a considerable extent. I was, however, unable to obtain any information in regard to it. Below Rabbit River the channel is clear for a couple of miles, and then dancing white-caps on ahead indicate the presence of another rapid. In the next mile the river alternately narrows in and expands three times, and falls over short but strong riffles at each constriction, all of which can easily be avoided, if necessary, by making portages a few yards in length. The behaviour of the water in the dilated basins between the narrows is somewhat peculiar, as it seems, viewed from the bank, to be running in all directions and to be split into a network of cross currents. At the lower narrows three ugly looking whirlpools are formed by the rapidly contracting stream endeavouring Whirlpools to crowd its way through its narrowed channel, while the water sucked down by the whirlpools is thrown up a little farther down in huge boils and with a sound resembling the rumbling of distant thunder. The whirlpools occur near the left side, and can be passed in safety by keeping to the right bank. From Cranberry Portage to Whirlpool Canon, the point now reached Rocksbetween in the description of the river, numerous exposures of the same shale ^™berry sandstone and conglomerate series previously described are everywhere Whirlpool present, and to the heterogeneity of this formation are mostly due the numerous rapids which occur on this portion of the river, the 13 194 GEOLOGICAL SURVEY OP CANADA. harder bands narrowing in and damming back the stream, while the softer and more easily eroded parts have acquired a more uniform slope. The shales and associated beds are everywhere greatly disturbed and usually dip at high angles. They have a general south-easterly strike. ^^*/^^ d b -^^ Whirlpool Caiion the shales disappear and are replaced Toy a shaly limestone. variety of limestone. From Whirlpool Canon the river flows swiftly around a sharp bend, at the extremity of which it receives Coal River, and after a clear course of less than four miles, plunges over the rapids at Portage Brul^. Coal River. Coal River is a small, clear stream about a hundred feet wide, and is interesting on account of the quantity of lignite which it brings Lignite at down. At the time of our visit a bar at its mouth was thickly strewn River with large angular and apparently little-travelled blocks of this min- eral. The fresh appearance of the lignit'e induced me to spend part of a day, while the men were packing across Portage Brul^, in explor- ing for the bed from which it originated, but a walk of several miles up the stream failed to reveal its presence in situ, although an abund- ance of drift fragments was everywhere noticed. The lignite is of inferior quality. It is soft and shows a well-marked woody structure. The banks of Coal River, as far as my examination extended, are low, and consist of uncemented sands, clays and gravel, like those holding the lignite beds above the Little Canon. This formation is of irregular thickness, but of wide distribution, as it was observed filling depres- sions of the older rocks all the way from the mouth of the Dease to the passage of the Rockies. PortageBrCile. Portage Brule is nearly two miles long, and leads across a nearly level, well wooded flat, which, at the upper end of the portage, is only elevated a few feet above the surface of the river, but at the lower end is terminated by a sharp descent of over two hundred feet. A good track was cut across this portage when mining was being pro- secuted on the Liard, and a windlass built at the east end for the pur- pose of hoisting boats up the steep bank, both of which are still in good condition. It was at the lower end of this portage, in the year 1836, that a party of Hudson's Bay voyageurs, bound on a trading expedition to the Stikine, after carrying their packs up the hill, were seized with a panic caused by the supposed approach of a band of hos- tile Indians, and, abandoning their outfit, fled for safety down the river. In the succeeding year Mr. Robert Campbell found the goods in the same position in which they had been left. Rapids at The rapids at Portage Brul4 do not look so formidable as those at Portage Brftle. ^jjg Mountain Portage, and if I had examined them before making the YUKON DISTEICT AND NOETHBRN BEITISH COLUMBIA. 195 portage I wbuld have been strongly tempted to try and run them with the empty boat. They are about two miles long, and are caused by numerous limestone blocks and small islands obstructing the channel. At the lower end the river is narrowly confined by high vertical cliffs. The rocks observed along the river from Whirlpool Canon to the jio^ks lower end of Portage Brul4 consist altogether of different varieties of between limestone. This occurs in some places in massive beds, ranging in tex- Cafion^and ture from compact to moderately crystalline. In other places it be- ^'oi'tageBrtUe. comes very impure and shaly, and often passes into imperfectly devel- oped calc-schists. JSTo fossils were obtained from it, but it has a close i lithological resemblance to the limestone occurring above Cranberry Rapids, and to the Castle Mountains group of the Bow River section and is probably of Cambro-Silurian age. After freeing ourselves from the rapids at Portage Brul^ no farther j,- obstacles to navigation were encountered until the Devil's Portage was PortageBrdl^. reached. The river is wide and filled with low islands and bars, some of which are auriferous, McCullough's Bar, on which gold in paying MeCulIough's quantities was first discovered on the Liard, occurs in this vicinity, but ■^^^' I was unable to identify it. The river valley is now lined with rows of terraces rising up to a height of several hundred feet, and clothed Terraces along in unwooded portions by as luxuriant a growth of grasses and vetches "^®'' as I have ever seen in any part of the country. Behind the terraces is a gently undulating region, occasionally swelling into elevations of from 1200 to 1500 feet in height, and everywhere densely forested, chiefly with white spruce. To the eastward the elevations increase in height and frequency until they merge into the range of the Rocky Mountains, the dim outline of which can now be seen along the eastern horizon. This part of the country, judging from the luxuriance of the vegetation Agricultural and the character of the soil, seems well adapted for agricultural pur- P^^P*"^ ^■ poses, but the complete absence of climatic statistics render any posi- tive statements in this connection premature. On the present trip at the Little Canon, large snow banks were observed in sheltered places alonff the banks of the river as late as the 28th of June. It must be borne in mind, however, that the spring of 1887 in this part of the country was an exceptionally late one. Ten miles below Portage Brul4 Smith River comes in from the north. This is a small stream about one hundred feet wide, and appears to originate in a north-westerly spur from the Rocky Moun- tains, visible in the distance. At its mouth was situated fort Halkett, Fort H.ilkett. a Hudson's Bay trading-post which has been abandoned since 1865. I found no traces of the post. 13J 196 GEOLOGICAL SURVEY OP CANADA. Shales at Near the mouth of Smith River, a number of exposures <^i darlc mouth of shales were observed. These shales are much softer than those seen Smith Kiver. farther up the river, and are probably of Mesozoic age. They were traced down the river for several miles, and are then replaced by lime- stones. No fossils were found in them. River below From Fort Halkett, the mountains appear quite close and the river °^ * " • rmig swiftly in an E. S. E. direction straight towards a narrow gap which now appears in their ranks. Before entering this we pass, on the right-hand side, the mouth of Riviere des Vents. This river comes from a large lake a few miles south of Fort Halkett, from which the fish supply of the post was obtained. It cuts off from the main range a steep sided, massive looking mountain, which I named Mt. Read in honour of Mr. Reid, the Hudson's Bay officer at present in charge of Fort Providence and an old traveller in these regions. From Riviere des Vents we approached the gap cautiously, on the lookout for the Devil's portage and rapids, which were reported to exist in this neighbourhood, but, much to our surprise, passed through without hindrance, and in a few minutes found ourselves among the rolling foot-hills on the eastern side of the range. The mountains here are narrowed to a single range, and even this, a few miles north of the river, is so reduced in height that it can scarcely be distinguished from the ordinary ridges of the district. To the south the range is much wider and the river seems to skirt the northern extremity of what may be considered the main division of the Rocky Mountain system. Rooky Moun- The Rocky Mountains, regarded as forming the eastern mountain tains. system of the Cordillera, are consequently interrupted in this part of their length. The range of which the northern extremity is here found has a length of over 1000 miles. It extends uninterruptedly south- ward to the International Boundary, and is still further continued into Montana to about lat. 46°. The width of this persistent mountain range probably averages throughout about fifty miles ; and its main physicial and geological features are almost identical in all parts of its length. Where the particular line of crumpling and upheaval of the earth's crust to which this range is due dies away at the Liard, another similar line begins, nearly in the same latitude, but about eighty miles farther to the east. The mountain range produced by this new line of disturbance extends northward nearly parallel to the general course of the Mackenzie to the Arctic Ocean. South of the Liard, the bare limestone ridges are ranged in parallel lines, and are surmounted by large zigzag knife-edges, or jagged Rocks in serrated crests. The ridges have a general strike of N. 30° W. The mountains. spur of the mountains which crosses the river consists of a grayish and YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 197 moderately compact limestone. The beds are at first nearly horizontal, but are soon thrown into almost vertical attitudes, and have the appearance of a sharp anticline. Several of the exposures were examined unsuccessfully for fossils, and the only specimens collected here consist of some fragments of corals which were found loose at the mouth of Riviere des Vents, and resemble those occurring in the Intermediate limestone of the Bow River section. The limestones are exposed along the river for six miles, and are then covered with dark shales similar to those from which Triassic fossils were afterwards Hmestones obtained. It is noteworthy that in the Liard section dark shales of repla'oed by S hSil API Mesozoic age are found resting on both jianks of the mountain, and are not confined to the eastern slope, as in the case farther south. East of the passage of the Rockies, the mountains gradually recede Rounded hilla from the river towards the south, and are replaced by high rounded 1^°^ moun- and well wooded hills and ridges built of dark shales, numerous ex- posures of which occur all along the banks of the river. The shales undulate at all angles, and do not appear to have any predominating dip in towards the mountains such as characterizes them in other localities. They are interstratified in places with beds of quartzite, and are, so far as I could learn, completely unfossiliferous. Since leaving Portage Brul6 the river had remained wonderfully River smooth, smooth, and we had the pleasure of passing one night away from the roar of a rapid, and without the usual prospect of having the next morning either to run a rapid or make a portage. The river has here an average width of four hundred yards, and a steady current of about four miles and a-half an hour. It is bordered in places by long gravel and sand beaches, and incloses occasionally wooded islands. Ten miles east of the gap. Trout River joins the Liard from the south. This is a frout River swift, clear mountain stream about a hundred and fifty feet wide, which seems to cut back into and drain the central ranges. Below Trout River the Liard bends abruptly to the north for some distance, and then, turning to the east, continues on with an ever increasing current, between banks which gradually become steeper and higher, until they develop into a wide canon. "We had been on the lookout for the Devil's Rapids ever since leaving the Riviere des Vents, and as the threatening appearance of river and valley indicated that we Threatening were approaching them at last, we dropped down cautiously along appearance of the right bank, watching carefully all the time for signs of the old portage. We failed to discover any, but landed at what seemed to be the last break in the almost vertical cliffs, with which the river was now bordered, and just at the head of a long easy rifile. Looking down the river ominous streaks of white could be seen in a couple 198 GEOLOGICAL SURVEY OF CANADA. of places, stretching from bank to bank, while the familiar roar of clashing water was plainly audible. On landing we found, after a short search, traces of the old portage-track, and the next few days were spent in carrying our outfit across. The river at this point makes a great bend to the north-east, all around which is a succession of rapids and canons. At the elbow of Fall reported, the bend a large fall is reported. At the lower end of the bend the river is reduced to a mere thread, as it is scarcely a hundred and fifty feet wide, and as fully a third of this is occupied by shore eddies, its bed must be eroded to an enormous depth. Immediately below the contracted part is a large eddy, and the river expands at once to over half a mile in width. The Devil's Portage to Hell Gaet. Devil's Portage. Abundance moose. The portage across the bend proved, greatly to our satisfaction, to be less than four miles in length instead of twelve, as we had been informed and expected. It is, however, very difficult, as it passes over a ridge fully a thousand feet high, on both sides of which the slopes are exceedingly steep. The old portage-track was easily followed among the heavy timber for some distance after leaving the upper end, but going east it became gradually overgrown with brushwood and at last disappeared, and we were obliged to cut out a new one for ourselves. This track was cleared out by Messrs. McCullough and Thibert in 1871 for the purpose of hauling their boat across, and it speaks somewhat favourably for the activity of forest growth in this region that it should now be covered with shrubs and small trees several inches in diameter. E In crossing the portage we started several moose, and it may he mentioned here that the country we have been passing through and as far on as Hell Gate is probably the best moose country in North America. Everywhere we landed, fresh tracks in abundance were observed. We killed one at the mouth of Riviere des Vents and another farther down near Crow River, and could have shot a number of others if so desired. At the "Rapids of the Drowned" we scared three into the river, but these unfortunately attempted to swim the rapids and were drowned. They were found afterwards some distance below lodged in a drift-pile. Their abundance is due to the fact that the country is practically uninhabited. After leaving the Little Canon we saw no Indians nor any traces of them, such as old camps, abandoned canoes, old cuttings, etc., anywhere along the river. The country in question forms a kind of neutral ground between the YUKON DISTRICT AND NORTHEUN BRITISH COLUMBIA. 199 Indians trading east and west of the mountains, and is also difficult of access on account of the danger in navigating the river. The absence of Indians and the consequent immunity from hunting enjoyed by the moose, since the abandonment of old Fort Halkett, has resulted in a great natural increase in their numbers. Besides, as they are persistently hunted in the adjoining country on either side of the mountains, by the bands of Indians trading at Fort Liard and at the mouth of Dease River respectively, many of them must escape into this district as into "a city of refuge." The beaver is also abundant, and, like the moose, appears to have thriven in the absence of their hereditary enemies. Grizzly bears were reported to be especially common on the Devil's Portage, but we did not meet with any. We spent altogether six days on the Devil's Portage crossing our outfit and framing a boat. We found that it would be impossible to cut out a track and haul our heavy wooden boat up the steep hills on the portage without wasting more time than we could well spare, and decided, somewhat reluctantly, to abandon it. To meet such an emergency I had provided myself, before leaving Ottawa, with a p- canvas roll of stout canvas sewn up in the shape of a boat, and this we at ^^o^i- once proceeded to put into shape. It was stretched on a stout plank hewn out of a small pine tree. Spruce poles, to which the canvas was firmly sewn, were used as gunwales, and willow withes for ribs, while slips to lay between the ribs and the canvas were easily cut. We painted the canvas with half a gallon of oil, which had been brought for the purpose, but this did not prevent it from leaking badly, and we were obliged to give it a second coat, made up of everything oleaginous which we still possessed. This mixture consisted of sperm candles, gun oil and bacon grease, stirred up with spruce gum, and proved effective in keeping out the water. Our new bo8.t was not well adapted for running heavy rapids, especially where sudden turns had to be made to avoid rocks, but was quite serviceable in ordinary water and on easy riffles, and was, besides, light and easily portaged. Below the Devil's Portage for thirty or forty miles the river flows Grand Canon, throu^'h what is called the Grand Canon, but is more correctly a succes- sion of short canons with expanded basins between, filled with eddying currents. In low water the whole of this reach can be easily run in almost any kind of a boat, but in the season of high floods such as it was when we passed through, the water forcing its way through the throat-like contractions is thrown into a commotion too violent for any but the staunchest boats to stand. The canon is reported to have been run in two hours, which would be at the rate of about eighteen miles an hour, an astonishing velocity, but the time was probably underesti- 200 GEOLOGICAL SURVEY OF CANADA. mated. It took us several days to get through, but we were obliged to make a number of short portages and one of over three miles in " length. Descent of We launched our canvas boat and commenced the descent of the Grand Canon, q^^^^^j^ Canon on the 16th July. I had previously explored the river for some distance and knew that we had nothing serious to encounter for se\'eral miles. The river is at first wide and incloses a number of islands. As we proceed the bordering banks close in, become higher and steeper, while the current runs with increasing impetuosity, until, rounding a bend, it breaks into foam against a barrier of rocks which intercepts its course. We passed this by a small portage and then con- tinued our headlong course down the ri\'er, but were soon stopped again by a succession of bad riffles. In the next couple of miles we ran through a number of wild places, the canvas boat riding the waves gallantly, but were forced to make one or two short portages and then enter a deep, gloomy defile, walled in hy black vertical cliffs. The river here, is however, less boisterous and flows with a steadier cur- rent. Part way down the oafion a couple of small islets with steep, rooky banks divide the stream into several channels. These were soon left behind, and then hurrying through a second narrow pass we came out on a wider portion of ,the i-iver. In the next twelve miles the river is generally wide and shallow and filled with gravel bars. The current is still swift, running at the rate of seven miles an hour, but the navigation is easy. Rapids of the A.t the end of this reach it bends to the north, and striking violently Drowned. against some sombre clifls which line the left hand bank, is deflected again to the east with the formation of what are known as the "Rapids of the Drowned." Here, one of the most dangerous spots on the river is formed by water plunging with its whole force, over a ledge of rock which curves outwards and downwards from the left hand bank, into a boiling chaudiere behind. Nameof rapid. The name of the rapid originated from the drowning at this point of a Hudson's Bay clerk named Brown, and a boat load of voyageurs. As the story goes. Brown, disregarding the advice of his steersman, insisted on running close to the northern bank, and the canoe plunging into the hole mentioned above was drawn under. We passed the rapid by letting our boat down cautiously with a rope to the chaudiere, and then making a short portage. With a pro- per boat, however, no diiflculty would be experienced in crossing the river above the rapid, and running down close to the right bank. Below the "Rapid of the Drowned" is a long riffle, down which we ran at an exciting pace, and then the river is closed in by a hard sand- YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 201 stone bank, through a narrow gap in which it forces with dif- ficulty a stormy passage. An examination of this convinced me of the impossibility of running it at the present high stage of the water with a boat such as we possessed, and also disclosed the unwelcome fact that a number of similar obstructions existed ahead. In the next four miles the river is closely caSoned in, five times, and falls over a number of riffles. Only some of these are dangerous, but we were forced to portage around the whole reach, owing to the steepness of the banks and the impossibility of getting down to the bottom of the valley, except at a few points. Three miles of rapid current fol- Hell Gate, lowed, and then we reached Hell Gate, so named because it is the en- trance from below, to the wild portion of the river we have been descending. At this point there is an abandoned channel on the left- hand side, which is navigable in high water, and affords an easy pas- sage through. From the Devil's Portage to Hell Gate, the rocks noticed on the Rooks banks of the river consist mainly of shales, with bands of sandstone at J?^*^?,*^"^ intervals, and occasionally some limestone. Exposures of the latter oc- Portage cur at the east end of the Devil's Portage, where they are brought up '""'^ Hell Gate. by an anticline from below the shales. They are here grayish in colour, are moderately crystalline, and hold fragments of crinoids and other fossils, and evidently belong to the upper part of the Pa- laeozoic series of the mountains. This series was not observed far- ther east. The shales are usually rather hard and weather into high bold cliffs, which often border the river on either side for miles with- out a break. They are dark in colour, and as a rule are rather coarse- ly laminated, but vary greatly in this respect and also in their texture. The shales undulate at all angles, and are continously exposed all the way down the caiion. They are interstratified with a hard grayish and yellowish sandstone, and a dark compact limestone. The sandstone occurs in bands, which form constrictions where they cross the river, and in the adjoining country rise into high hills and ridges. The limestones associated with the shales are usually impure and occur as a rule in thin shaly beds, but heavily bedded varieties were also observed at Hell Gate and at other places. No fossils were obtained in decending the canon until the " Rapids Xriassic of the Drowned " were reached, where the following species, since fossils. described or determined by Mr. Whiteaves and referred by him to the Trias, were obtained.* *Contributions to Canadian Palseontology, yol. I, Part II, pages 127-] 49. 202 GEOLOGICAL SURVEY OF CANADA. Spiriferina horealis, Whiteaves. I'erebratida Liardensis, " Monotis oralis, " Halohia ( Daonella) Lommelli, Wissman. Hcdobia occidentalis, Whiteaves. Nautilus Liardensis, " Popanoceras McGonnelli, " I'rachyceras Canadense, " Some miles farther down, at the lower end of the last canon before coming to Hell Gate, a second fossil locality was discovered which yielded the followiug species. Terebratula Liardensis, Whiteaves. Trigonodus ? productus, " Margarita Triassica, " Fopanoceras McGonnelli, " At Hell Gate, three miles below the last, specimens were obtained of— Trachyceras Canadense, Whiteaves. Terebratula Liardensis, " The localities from which Triassic fossils were obtained, extend along the river for ten miles, but it is highly probable that the rocks of this age have a much wider distribution than this and include the greater part of the barren shales above the " Rapids of the Drowned," as well as those below Hell Gate. It will require, however, more time than could be spared on a rapid reconnaissance, to separate precisely the shales which cap the Palseozoio from those of the Trias, and the latter from the Cretaceous. The three terranes are closely allied lithologically and will have to be defined largely from fossil evidence. Hell Gate to Fort Liard. River bclov/ Escaping from Hell Gate Canon, the river dilates as usual and is bor- dered by large eddies. Below these, it runs swiftlj around a large island, and then enters a cafion-like reach of the river about a mile long. The stream here is narrowed down to about a hundred and fifty yards in width, and flows easily between vertical banks three hundred feet high. This canon proved to be the last on the river, and from this point on the river has an uninterrupted flow, and presents no obstacles to navigation until near its mouth. Five miles below the canon, the undulations in the shales and associated sandstones and Hell Gate. YUKON DISTRICT AND NORTHEEN BRITISH COLUMBIA. 203 limestones cease, and at the same time the ridgy and hilly foot-hill district we have been passing through is replaced by a region of high irregular plateaux. The foot-hills along the Liard have a width of 38 miles; and are Foot-hills. characterized by a much greater irregularity in altitude than is usually the case. South of the Devil's Portage, Mount Prudence, a steep-sided, reddish looking mountain, rises abruptly from a surrounding of round topped hills to an estimated height of over 4000 feet. Going east from Mount Prudence, lower elevations prevail until near the " Rapids of the Drowned " where the ridges again commence to increase in eleva^ tion, and in a few miles culminate in peaks over 4000 feet high. Still going east they gradually diminish in importance and at last die away and are replaced by flat topped plateaux. This region with the excep- tion of the higher peaks, is everywhere densely forested, chiefly with the white spruce, the Banksian pine and the aspen. After leaving the foot-hill country the river runs in a general direc- River east of tion of N. 30° E. for thirty miles. In this reach it has a steady cur- f°ot-l^ills- rent of about four miles and a-half an hour, and varies in width from five hundred yards to over a mile. In the wider portions the river is usually divided into several channels by islands and bars. The valley is narrow and trough-like, with steep sides rising up in places to a height of fully a thousand feet. The bottoms are usually small, and are here chiefly wooded by members of the poplar family. Some im- portant tributaries are received by the Liard in this portion of its course, among which is Crow River, which joins it from the north after entering the plateau country, and Toad River, which comes in from the south throu'gh a deep gloomy valley four miles farther down. Two miles below Toad River, on the opposite side, is situated Toad River post, which was abandoned when the post on the Nelson was established. The buildings are still standing. The geology of the plateau belt is exceedingly simple. The banks Geology of of the valley are usually scarped, and show everywhere extensive sec- ^ ^ '"^^ tions of flat-lying shales. These shales are dark in colour, are soft and finely laminated, and are interstratified with small beds cf sandstone and ironstone, and layers of ironstone nodules. Tliey are of Cretaceous age, but their mode of junction with the Triassic shales of the foot- hills was not clearly ascertained. Some fossils were collected from this formation about four miles below Toad River, among which are specimens of Flaceniiceras Perezianum, a species of Camptonectes, and fragments of an Inoceramus. Towards the eastern part of the plateau bolt the shales along the river are overlain by massive beds of soft sand- 204 GEOLOGICAL SURVEY OP CANADA. stone and conglomerate, which form a steep escarpment running parallel with the river. Enter lower j^t the end of the northerly reach iust described, the river, here over country. ., . , j j a mile wide and filled with islands, bends suddenly at right angles to its former course, and after passing through a narrow gap, enters a much lower country. The steep scarped banks of the plateau district disappear, and are replaced by gently inclined hillsides covered with forest, while the river spreads out and flows for some miles in a multi- tude of channels through a bewildering maze of islands. The eastern edge of the plateau district faces eastward with a steep slope, and has a height of over a thousand feet. It runs nearly due north-and-south and forms an important feature in the general topography of the country. Where it crosses the river it shows exposures of soft conglomerates dipping lightly in an easterly direction. Bast of this escarpment Beaver River joins the Liard from the north. Beaver River. This is reported to be a fair sized stream and to be navigable for canoes for a long distance. It empties into the Liard behind a group of Meet Indians islands and we passed it without seeing it. Near the mouth we saw Indians for the first time since leaving the mouth of the Dease. They belonged to Fort Liard, and were on their way up the Beaver to hunt. We endeavoured to buy some meat from them, but found that they were totally unacquainted with the use of money, and as we were not supplied with trading goods, or, in fact, with anything except what we wore, it was found impossible to strike a bargain. East of the Beaver the Liard runs in a south-easterly direction for a few miles and makes a couple of sharp bends before joining the Nelson. Meet crew I"^ ^he first of these bends we met a crew of Hudson's Bay voyageurs of H.B. in charffe of W. Lepine, who were endeavouring to make their wav un vojageurs. o i ; o j r the river to the mouth of the Dease. Lupine had been employed on the river as a guide, in the old days when goods were taken by this route to the Yukon, and was well acquainted with it. He brought news of a scarcity of provisions in the Mackenzie River District, and Send men back ^^^^ decided me to send my two men back up the river with him, and up river. jq depend on the services of natives for canoemen in the future. Lepine had become disheartened by the continued high water and the difficulties of up-stream navigation, and when we met him talked of returning, but we induced him to persevere. A small spruce-bark canoe which an Indian and his wife built in the afternoon, in addition to the large birch canoe which he already possessed, furnished sufficient accommodation for his increased party, and on the 28th July^ after a day's delay, he proceeded up the river. I afterwards learned that with the exception of one upset, caused by the unskilfulness of YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 205 Trepanier, one of my men, the journey was successfully accomplished and Dease River reached in safety. After separating from Lepine I continued down the river to Fort Liard in the canvas boat, at first in company with an Indian, but for the greater part of the distance entirely alone. After starting we rounded a large bend and then continued in a northerly direction to the mouth of the Nelson, or east branch of the Liard. Nothing has been published concerning the Nelson [1888], but it is Nelson River, reported to be a somewhat sluggish river of about one hundred and fifty or two hundred yards in width. A hundred miles above its mouth is situated Fort Nelson, a Hudson's Bay trading post. Above the fort the river divides into two branches, one of which, named Buffalo River, turns west to the mountains, while the other continues on and interlocks with tributaries of Hay River. *In 1872-73 a party of miners crossed from Peace River to the Liard by way of the Nelson on a prospecting trip. They descended Peace River to Half-way River, so called because it is half way between Rocky Mountain Portage and Fort St. John, and ascended the latter partly in canoes and partly on the ice for a hundred miles. They then made a portage of twenty-five miles, and reached the Nelson, down which they sledded for sixty miles, and then built boats and came the rest of the way by water. They only mention one portage of half a mile, but describe the river as flowing for a Ion'' distance above Fort Nelson, between lofty banks of sandstone and shale. " Colours " of gold were obtained on the Liard at the mouth of the Nelson. At Fort Nelson some farming is annually done, and potatoes and Fort Nekon. other vegetables are grown without difficulty. The .surrounding country is everywhere well forested, and is reported to produce'a bet- ter o-rade of timlier than any other part of the Mackenzie district. Below the Nelson the Liard has a general northerly direction for Li^rd below thirty miles, and then, bending more to the east, follows a N.E. course *•'«■ Nelson. as far as Foi-t Liard, fifteen miles farther down, where I arrived on July 29th. In this reach it has undulating shore-Hnes, but is gene- rally \% ide and filled with sandbars and wooded islands. It is bordered in many places with wide, alluvial flats, covered with tall, straight Cottonwood, large spruce, and canoe birch. Its valley is wide and shallow and lined with gently sloping, spruce-clad banks. On some of the flats the Indians have built houses, and fenced in small plots for farming purposes, for which the greater part of this section of the ' Extracted from letter by Fred. W. Harte. 206 GEOLOGICAL SUEVJiY OF CANADA. Indian f;irms. Tributaries of Liard. Rocks east of plateau district. Quaternary deposits. Eastern drift. Rocky Mountains.' Absence of foot-hills. district seems well adapted. We passed one small Indian farm about thirteen miles below the mouth of the iNelson, and another at the mouth of Fishing Creek, a few miles above Fort Liard, while others were noticed in the lower part of the river. The two principal tributaries of the Liard between the Nelson and Fort Liard are Riviere la Biche and Black River. The former enters it from the N.W., about twenty miles below the Nelson, and the latter from the S.E. at Fort Liard. Black River is the outlet of Lake Bis-tch6, a large lake situated about 120 miles S.S.W. of Fort Provi- dence, and is reported to be navigable with difficulty throughout its entire course in high water. It will afford, with the Nelson and Hay rivers, a ready means of entering and exploring the vast block of unknown country lying between the Liard and Peace River, the Mackenzie and the Rocky Mountains. Since leaving the plateau district the rock exposures observed along the river have been few and small, and consist of dark shales, alternat- ing with sandy shales and sandstones, all of Cretaceous age. Twelve miles below the Nelson an exposure of sandy shales and sandstone yielded an Ostrea like Ostrea subtrigonalis, and at the mouth of Black River, near Fort Liard, in a hard, crumbly shale, specimens of an Inoceramus were obtained, while the surface of the shale in some places is covered with well marked impressions of gigantic palm leaves, probably belonging to the genus Sabal, which occurs in Vancouver Island in rocks of a similar age. . The quaternary deposits in this part of the river are represented by stratified sands and gravels, immediately overlying the shales, and by gneissic erratics, which are distributed everywhere over the surface of the country, and in some places, as at the mouth of Black River, are present in great profusion. The western limit of the eastern drift along the Liard, judging from the river section, appears to be nearly coincident with the eastern edge of the plateau district, although gneissic fragments were found in the mountains opposite Fort Liard at a somewhat higher elevation. Thirty miles below the Nelson, the Liard approaches, and for the next seventy-five miles hugs closely a chain of mountains which may be considered as a northern division of the Rocky Mountain system, and as the complement of the chain around the northern end of which the Liard passes west of the Devil's Portage. The mountains are not fringed with a belt of foot-hills, such as usually accompany them in other places, but rise abruptly from an almost level plain, and attain at once the full height of about 4000 feet. The folds and fracture to which the mountains are due also seem to die away with startling YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 207 rapidity. The Liard for fifteen miles above Black River, cuts at a distance of two to five miles directly across the strike of the ranges, and yet the beds along its valley wherever seen are practically undis- turbed. While waiting at Fort Liard I climbed one of the ranges to Climb ranges. a height of 3000 feet, but obtained little geological information. The I'ocks were only exposed in a few places, and where seen coQsisted of westerly-dipping unfossiliferous chert and cherty limestone. Gneissic fragments apparently belonging to the eastern drift were found up to a height of over 1500 feet. I obtained, however, an extensive view view from top from the summit, over the plains to the eastward. The country in that "^ mountains. direction rises gradually from the river in easy undulations, and appears to culminate at a distance of twenty-five or thirty miles in a low plateau through which Black River has cut a wide gap. A dense forest, relieved in places by gleaming lakes and light green marshes, stretches to the horizon. To the north and north-west the eye is met by a suc- cession of bare topped and nearly parallel limestone ridges striking about N. 20° W. and reaching elevations of from 4000 to 5000 feet. Fort Liard is at present the only fort on the Liard below the mouth of the Dease, and is resorted to for trading purposes by about two hundred Indians, most of whom are known as Nahanni or Mountain Indians. Under this term are included a number of tribal divisions of the Tinneh family, but the names of these I was unable to obtain. They are reported to be fast dying off. The fort is situated on a fertile flat, part of which has been cultivated for years with unfailing success. Agricultural Wheat and barley are grown here year after year, while potatoes, cab- resources. bages, turnips and other vegetables are raised without the least diificulty. At the time of my visit, 1st August, all the crops were well advanced and in good condition ; the barley was just turning colour, and the potatoes were almost large enough to eat. There is no reason, either climatic or otherwise, why the whole country bordering the Liard, from Beaver River to near its mouth, should not, when needed, support an agricultural community. [Near Fort Liard the Liard River crosses the Eastern boundary of British Columbia. AU that part of Mr. McConnelFs report, as origin- ally printed, that relates to the Mackenzie VaUey proper is here omitted.] Rat River. Fort Macpherson was reached on the 25th of June, and as the Mac- Reach Fort kenzie steamer from which I expected to obtain supplies was not ex- Macpherson. pected down until the 10th of July, a short trip was made up Rat 208 GEOLOGICAL SURVEY OF CANADA. Rat River. River while awaiting its arrival. At the fort I was informed that neither boats nor canoas could be obtained west of the mountains, but that the boat I had used on the Mackenzie might be portaged across by taking it up Rat River and then through the MacDougal pass to Bell River on the other side. This plan was adopted, and five Indians were sent to try and take it across. They made the portage in seven days. Rat River is represented in Isbister's map of Peel River, published in 1845, in the journal of the Royal Geographical Society, as rising on the western side of the Rocky Mountains and flowing directly across them. This error was probably caused by the same name being used by the Indians to designate the two streams flowing east and west from the summit of the range. Rat River (east) discharges its waters by two mouths. The north branch empties into one of the delta streams of the Mackenzie, and has not yet been explored, while the south branch Joins Peel River about nine miles below Fort Macpherson. When Peel River is in flood the current in the south branch is reversed. Rat River for some miles above its mouth, winds through a flat alluvial plain, forming part of the Peel-Mackenzie delta. Its width is about fifty yards, and the current is uniform and easy. Above this reach it enters a lake region, and for several miles connects by short channels a number of small irregular-shaped willow-fringed lakes lying in shallow depressions in the delta. West of the lake region the declivity of the stream rapidly increases, and a few hours hard paddling brought us to the foot of a series of strong rapids which marks the Leave canoe, beginning of the ascent towards the mountains. Leaving our canoe here we tramped eight miles across a marshy plain to the foot of the first range, our objective point. Por part of the way we travelled along the top of a low narrow ridge, which is probably of morainic origin and is composed, so far as I could learn from the surface, of quartzite pebbles and boulders carried eastward from the mountains. The eastern range of the Rockies where broken through by Rat River, has a nearly north-and-south trend, and rises to an elevation of about 2800 feet. It is built throughout of evenly bedded hard sand- stones and quartzites, dipping to the east at an angle of about 30°. No fossils were found in rocks in situ, but in the wash of Long-stick Creek, a tributary of Rat River, which descends from the same range and was crossed on the way. Ammonites and other fossils of Cretaceous age were found. These occur in angular sandstone blocks, evidently derived from the neighbouring mountains, and I have little doubt Geology of eastern range of Rockies. YUKON DISTRICT AND NORTHERN BRITISH COLtJMBIA. 209 characterize the formation of which the latter are composed. No limestones, gneisses or granites, were observed. A sub-angular gneissic boulder, representing the eastern drift, was Eastern drift, found on the eastern slope of the mountains at a height of 1500 feet. Peel River Portage. On July 10th I decided not to waste any more time waiting for the steamer, as it was impossible to tell how long it might be delayed, and to trust to the chance of finding supplies at Rampart House. I had sent part of my outfit across by some Indians who were returning to Lapierre House, and Mr. Hodgson, after some trouble, succeeded in Indian engaging five others to accompany me and to carry the remainder. The P^'^l^ers. ordinary load for an Indian on this portage is 40 pounds, exclusive of blankets and supplies for the trip, and the tariff for this load is fifteen skins, or seven dollars and a-half, paid in goods. The portage is about sixty miles long, and from four to five days are occupied in the trip. We left the fort late in the afternoon, and ascending the river for a Leave Fort mile and a-half, landed on the opposite side and strapped on our Macpherson. packs, while some dogs owned by the Indians were loaded so heavily that they were unable to climb the bank of the river and had to be assisted up. After leaving the river we marched for three miles through a thick spruce forest, and then reached the steep edge of a terrace about Terraces at one hundred and fifty feet high, up which we climbed. From the top of mountains. the terrace, a swampy plain destitute of trees, but covered with willows chiefly of the species Salix arctica and >S. glauca, stretches west for nearly four miles, and then a second slope, longer and higher than the first, has to be ascended, at the summit of which the barometer regis- tered a height of twelve hundred feet above the river at the fort. At this elevation the forest has ceased, and the shrubs which occur so abun- dantly on the first terrace have almost disappeared, and are only represented by an occasional stunted specimen of the arctic willow (S. arctica). A small larch (Larix Americana) still survives, but barely attains a height of six or eight feet and a diameter of one to two inches. From the second slope a wide plain reaches westwards to the moun- tains. The walking here is exceedingly difficult, as the surface is DifScult covered with the rounded grassy sods which go in the country by the walking. name of Tetes des femmes. These project a foot or more above the clayey soil, and are the cause of constant stumbling, which becomes somewhat exasperating when one is weighted down with a pack. An attempt to walk on the top of the mounds soon becomes excessively fatiguing, on account of the irregular length of the strides, and a 14 210 GEOLOGICAL SUEVEY OP CANADA. Mountain stream. Mountain sheep hunt. Character of mountains. Wide valley. slight miscalculation as to distance precipitates the. unlucky traveller down into the muddy depths between. When down, the resolve is usually made, and adhered to for awhile, to keep to the lower levels, but the effort required to step over the intervening hillocks presents obvious disadvantages of a different kind. A walk of five hours, at the rate of about a mile and a-half an hour, brought us to a deep valley occupied by a swift mountain stream, which we forded, and then camped on the opposite bank. This stream, half a mile below camp, joins a small river which falls into Peel River three or four miles above the fort. Its channel is filled with sandstone, and quartzite slabs derived from its banks and from the neighbouring mountains. On the 11th we climbed out of the valley, and, following up the main stream, soon reached the " Gap " from which it emerges from the mountains. A halt of a couple of hours was called here for the purpose of allowing the Indians to add to their store of provisions by killing a mountain sheep which one of them observed on a ledge above. The hunt was successful, and the victim proved to be the interesting Ovis Montani, var. Dalli. This animal resembles the bighorn of the Rocky Mountains to the south in form, but is much smaller, weighing scarcely a hundred pounds, and its hair, instead of being tawny, is almost a pure white. The change in colour and size towards the north is evidently a gradual one, as the saddle-backed sheep of the upper Yukon presents characters intermediate between the two extreme varieties. At the "Gap" the pass through the mountains is eight to ten miles in width, but gradually contracts as we advance, and at our second camp, six miles from the mouth, was reduced to about two miles. The inclosing mountains are regular in outline and somewhat tame in appearance, and rise to elevations above the valley from one thousand to two thousand five hundred feet. The bottom of the valley and the lower slopes of the mountains are clothed with coarse grasses and mosses, above which project naked and often precipitous quartzite cliffs. The stream which occupies the valley at the present time is insignificant in size, and is doing little or no erosive work, and this grea't cleft through the hard quartzite strata is evidently the product of an earlier age, marked by a much greater precipitation than the present one. From our second camp a walk of two miles brought us through the first range and out on a wide longitudinal valley which traverses the mountains in a nearly north-and-south direction. The stream we had been following up divides here into two branches, and the prin- YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 211 cipal one, reduced now to a large sized brook, bends away to the south. After leaving it the trail leads up a long slope covered with Tetes des feinmes, and then directly across the valley to an opening in the range ahead. On the farther side of the valley we reached and followed up for a couple of miles a small stream which runs north to Rat River, and directly afterwards crossed the watershed between the Mackenzie and the Yukon and commenced our descent to the latter. The watershed has an elevation measured by the barometer Elevation ot of 2600 feet above the starting point on Peel River, or about 2650 watershed. above the sea, while the neighbouring mountains rise about a thousand feet higher. Two mUes west of the watershed the mountains turn suddenly to the north, and the westward-flowing stream which we were following plunges down a steep decline of fully 1200 feet. At the base of this declivity it is joined by a large tributary from the south, which swells it to a small river. At the confluence of the two streams is a large flat covered with small groves of the white spruce (Picea alba). The elevation here is over a thousand feet higher than the point at which this tree disappeared on the eastern side of the range. From the foot of the mountains a high table-land, swelHog occasion- tt- v ally into considerable eminences, stretches to the westward. Through table-land, this plateau the branch of Bell River down which the trail leads has cut a wild and gloomy chasm fully a thousand feet deep. The walls of the canon are formed of flat-lying sandstones and hard quartzites, and are naked and precipitous above, but fall away in easier slopes below. The bottom of the valley is generally soft and marshy, but hard gravelly terraces, affording good walking, are occasionally crossed. Seven miles from the head of the canon a large stream comes in from the right, and two miles and a-half farther on a second and more impetuous one pours into the main stream from the same direction. After crossing the latter we camped on some firm ground on the farther side. On the 13th we continued our way down the canon. Three miles from camp the river washes up against the walls on the right hand side, and the trail, which has hitherto followed the right bank, crosses over to the left. The ford is a difiicult one, as the stream is here deep and rapid, and its channel is paved with treacherous quartzite boulders. The greatest caution is necessary in crossing, as a stumble or false step Mode of would almost certainly be fatal to one encumbered with a heavy pack, streams In fording these swift mountain torrents, it is customary to adopt a communistic plan. The party line up behind a long pole, and keeping a firm hold of it advance into the stream abreast. In this case the 14i 212 GEOLOGICAL SURVEY OF CANADA. Character of mountains. Geological section. person above sustains the full brunt of the current, but is held up by those below, and a stunibler receives the support of those who have kept their footing. Two miles and a-half below the ford the flanking plateau drops sud- denly several hundred feet in elevation, and the canon disappears. A second crossing is necessary here, as the river after escaping from the confining walls of the carion bends away to the south. At this point the river is wide and shallow, and the crossing was made without diffi- culty. The stony banks were gay with the yellow flowers of Senecio lugens. Four miles farther on the river is again encountered return- ing from its southern bend, and is crossed for the last time. The river is bordered here by a marshy flat, about two miles in width, through which we waded. Beyond the marsh the trail winds up the steep edge of a high terrace, and then a four mile tramp across a succession of muskegs, brought us to the end of this stage of our journey, and we thankfully threw down our packs on the banks of Bell River. The Rocky Mountains, along the Peel River portage, present features which differ greatly from those which characterize them farther to the south. They consist here essentially of two ranges, separated by a wide longitudinal valley, and flanked on either side by high plateaux. The eastern range has a width of seven miles, and its higher peaks were estimated to reach an altitude of 2500 feet above the level of the pass, or about 4000 feet above the sea. The western range is much narrower, and north of the pass does not exceed four miles in width, but spreads out somewhat more towards the south. The valley of Peel River, which skirts the eastern base of the range, is fully 1200 feet lower than the valley of Rat River on the western side, and the drain- age of the mountains is mostly towards the former. The geological section obtained is somewhat imperfect, as our scanty supplies allowed of no delay, but sufficient was learnt to show that the range has on the whole an anticlinal structure, although the gene- ral anticline is obscured in places by subordinate folds, and is probably broken by faults. In the eastern plateau the beds are nearly horizon- tal, but approaching the mountains they incline gently to the east- wards, and in the centre of the eastern range have dips of from 30° to 70° in the same direction. In the western range the same dip prevails, but the inclination is much less, and the beds flatten out when the mountains are replaced by the elevated western plateau. The horizon- tal attitude is maintained for some miles, but before reaching the west- ern edge of the plateau the beds bend down and dip gently to the west. YUKON DISTRICT AND NORTHEKN BRITISH COLUMBIA. 213 No limestones were observed along this section, and the rooks con- sist of sandstones, quartzites and shales, all of which are probably re- ferrable to the Cretaceous. At the starting point on Peel River the banks are formed of shales, interbedded with some hard sandstones hold- ing carbonized fragments of wood and leaves. In the first fourteen miles the beds are concealed, but shales and sandstones are again ex- posed on the banks of the valley in which we made our first camp. Some fossils were collected here, among which is a Biscina, like D. Fossils. 2>ileolus, Whiteaves, a Macira and a Yoldia, both of which are pro- bably new, but the specimens are too imperfectly preserved to admit of specific determination. Six miles farther west, at the entrance to the "Gap,'' the trail passes over beds of a bluish, rather compact calcareous sandstone. The beds are coated in places with calc-spar, are highly ferruginous, and weather to a rusty yellow. A Cardium and some other poorly preserved fossils were obtained here. In the valley of the river the sandstones are underlain by dark shales. After enter- ing the mountains, only alternating sandstones and quartzites were seen. The beds of this series are grayish in colour, are evenly strati- fied and are very uniform in appearance all across the range. They have an estimated minimum thickness of 5000 feet, and may possibly greatly exceed this. The western plateau is built of compact grayish sandstones, passing in places into quartzites, similar to those found in the mountains and evidently belonging to the same formation. Both Isbister and Petitot mention limestone, schists, and granites, as No limestones being found in this section of the mountains, but these rocks were not J^ountains™ seen in the section examined by me, nor were pebbles derived from them found in the wash of any of the streams which we crossed. In the valley of the Mackenzie, however, the Devonian limestones imme- diately underlie the sandstones and shales of the Cretaceous, and it is highly probable that in parts of the range the disturbance has been sutficient to bring these to the surface. On arriving at Bell River, we crossed over to Lapierre House, Lapierre which is situated on a flat on the western side, but found the post Hou.«e. deserted by all but Indians, the officers in charge having left some time before for Fort Macpherson. Lapierre House is simply an outpost of Fort Macpherson, and is kept up principally to faciliate the transit of goods and furs across the mountains, although some trading is also done both with the Loucheux and the Eskimo. It has been in exis- tence about thirty-five years. No farming of any kind is attempted either here or at Fort Macpherson. The boat which I had sent across the mountain by the McDougal Prepare for Pass I was glad to find had reached its destination in good order, and Porcupine. 2U GEOLOGICAL SURVEY OF CANADA. Character of Bell River. Geology of Bell Eiver. no time was lost in preparing for the descent of Bell River and the Porcupine. My crew down these rivers consisted, beside myself, of one man, an Orkneyman and ex-employ^ of the Hudson's Bay Com- pany named Skee, whom I met at Fort Macpherson on his way out of the country, and was fortunately able to induce to come with me and go out by the way of the Yukon. The upper part of Bell River has not been explored. At the fort it is a small sluggish stream of forty or fifty yards in width, and a cur- rent of less than two miles an hour. The banks are low and alluvial, and the bordering region is covered with willows, birch, aspen and spruce belonging to the same species as those noticed east of the moun- tains. Below Lapierre House, Bell River runs a few degrees east of south for seven or eight miles, and then bending to the south-west flows by Sinclair's Rock with a somewhat accelerated current, and con- tinues on in the same direction to Stony River, a stream nearly equal in size to itself. From Stony River its course is north-west, or nearly opposite to its direction in the first reach, for seven or eight miles, and then it bends gradually around to the south-west and keeps this course until it empties into the Porcupine. Its length from Lapierre House to its mouth is about thirty miles. Besides Stony River, which comes in from the left, it receives a short distance farther down, from the same side, the waters of Eagle River, a stream of about a hundred feet in width. Below its junction with these two streams Bell River becomes considerably enlarged, and in its lower part expands to about a hundred yards in width. No rapids were met with below Lapierre House, and the current seldom exceeds three miles an hour, and is usually much slower. The valley of Bell River is shallow, and no rock-sections occur in the banks below Lapierre House until the cliff at Sinclair's Rock is reached. At this point the river cuts through a range of hills, and coarse arenaceous shales are uncovered, which probably represent an extension of the same beds that in the mountains are hardened into quartzites. The shales are unevenly bedded, and under the influence of the weather crumble into a talus of sharp, angular fragments. For some miles below Sinclair's Rock the banks are formed of alluvial clays and sands, but shales are again exposed about half a-mile below Eagle River. The beds here are softer and darker in colour than those found in Sinclair's Rock, and are better stratified and separated into thinner laminse. They alternate with some beds of sandstone and ironstone. The dip is S. 65° W. < 20°. Beds of a similar character are exposed again four miles farther down. Here they are horizontal when first seen, but in a short distance bend down and assume a verti- YUKON DISTRICT AND NORTHEKN BRITISH COLUMBIA. 215 cal attitude. Below this point dark fissile shales, alternating occasion- ally with beds of sandstone and ironstone, undulate at all angles in the bank of the river, and are almost continuously expesed until with- in a few miles of the mouth of the river, when they are overlain and concealed by recent clays and sands. The rocks exposed along Bell Biver are unfossiliferous, but are closely related lithologically to the Cretaceous shales and sandstones of the Lower Mackenzie and the mountains, and I have little hesitation in referring them to the same horizon. Porcupine River. Bell River to Driftwood River. Very little information was available in regard to the Porcupine P9rcupine before the present exploration was undertaken. Short accounts de- rived from hearsay evidence, are given by Richardson and others, but it was not personally visited by any of the numerous travellers who have written of the Mackenzie and the Yukon, although it has been used by the Hudson's Bay Company as a trade route since 1847, and its navigation presents no special difficulty. The Porcupine heads within thirty miles of the Pelly- Yukon, ap- proximately in latitude 65° 30' N., and after describing a great semi- circular curve to the north-east, falls into the same river a hundred and fifty miles farther down. At its most easterly point it approaches within eighty miles of the Mackenzie, but it is separated from it by the main range of the Rocky Mountains. Its total length approxi- mates to five hundred miles. The upper part of the Porcupine was explored by Mr. Ogilvie in, the Exrforation of early part of the present summer. Ogilvie reached the head of the river by a winter traverse from the Pelly- Yukon, and descended it as far as the mouth of Bell River. An account of this part of the river will be found in his report. The present description treats of the part below Bell River. From Bell River to Driftwood River, a distance of about thirty miles Character of , , /! ,1 • ii T. the Porcupine, in a straight line, but over forty by the course of the river, the Porcu- pine has a general north-westerly trend, but makes a couple of minor bends to the north-east. Its width varies from one hundred and fifty to two hundred yards, and its current barely averages two miles an hour. The valley is generally rather wide and shallow, but at one point about ten miles below Bell River, becomes somewhat contracted, and for some miles has the appearance of a wide canon. The banks 216 GEOLOGICAL SURVEY OP CANADA. here are high and steep, and are formed of broken fragments of hard quartzite. Below the contraction it resumes its usual character. The bordering country is very uneven and swells in places into hills and ridges, most of which appear to have a northerly trend. None of these elevations attain any considerable altitude, and like the lowlands they are clothed with the same monotonous forest of spruce and aspen which has been so often described. Cretaceous In this reach the river cuts through a very interesting series of Cre- section. taceous beds, but the haste with which we were travelling prevented mr. from exiiiuining these in as satisfactory a manner as might be de- sired. In the first ten miles below the mouth of Bell River, alluvial Sandstone and g^nds and clays form the banks of the valley, and then hard sandstones quartzite •' . series. and quartzites appear, and are continuously exposed for the next three miles. The sandstones are grayish in colour, very hard, and pass gradually into a compact quartzite. In weathering they break into sharp angular fragments, and the lower slopes of the banks are covered with a steep talus of this material. The dip here is nearly east, and the rooks exposed have a thickness of between three and four thousand feet. Shale series. The sandstone series is succeeded and undei'lain by dark shales inter- bedded with ironstone. The shales have a light easterly dip, and an approximate thickness of eight hundred feet. They yielded some fossils, among which is a very large belemnite, a finely ribbed ventricose scaphite, which has some resemlilance to Scaphites veiUricosus, and a peculiar ammonitoid shell, which shows a ribbed central portion, while the outer whorl is quite smooth. Mr. Whiteaves states that the fos- sils are probably Benton, but the specimens are too imperfect to make the correlation certain. Sandstone, The shales are exposed for about a mile, and are then replaced by quartzite and ■ ^ ^ ^ r j conglomerate shales, hard shaly sandstones and quartzites, interstratified with occa- sional conglomeritic beds. The sandstones are often somewhat greenish on a fresh fracture, but weather to a yellowish or rusty coloui-. These beds are brought up by a light easterly dip from beneath the shales, and occur in frequent exposures for a distance of two miles. They have a minimum thickness of two thousand feet. Some of the shaly sandstones of this series are fossiliferous, affording numerous speci- mens of an Ontrea, along with a Pectan, a Rliynchonella and an appar- ently undescribed species of Pteria. These fossils do not afford any definite proof in I'egard to the age of the formation, but are not at variance with its reference to the Dakota, to which the stratigraphical evidence would assign it. YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 217 Continuing down the river, an interval of about a mile occurs in which the beds are concealed, and then sandstones are again exposed, forming the floor of a wide shelving beach on the right-hand side of the river. The dip is still easterly, and the descending section has evidently been continuous. The sandstone here is a hard grayish variety, and in its compact texture it approaches a quartzite. It is fossiliferous, and some of the beds are covered with casts of Aiicella Aucella beds. mosqueusis var. concentrica, the characteristic fos.=il of the Queen Char- lotte Islands formation. This formation occurs in British Columbia, at the base of the Cretaceous, and has been referred by Mr. Whiteaves, on fossil evidence, to a position immediately below the Dakota, a ref- erence which is apparently justified by the facts detailed in the section just described. Below the point on the river at which the Aucella beds are exposed, the section becomes so confused and is interrupted by so many con- cealed intervals, that it was found impossible to follow closely the sequence of the terranes. A mile and a-half below the Aucella beds, yellowish-weathering shales and sandstones are exposed, which probably belong to the same series. The easterly dip, which has been maintained for so long. Change of dip. changes here, and the inclination of the beds becomes more variable, and is not infrequently to the west. The sandstones are followed after an interval of three miles by black bituminous looking shales, the relations of which are somewhat obscure, and these by a hard conglom- erate, the dip of which is to the south. Five miles farther down fer- ruginous rusty-weathering shales, striking nearly east-and-west, and dipping up stream, are again exposed. The beds first seen are very fissile, and are lighter coloured than is usually the case, but are under- lain by a black carbonaceous variety resembling the Devonian shales seen along the Grand View on the Mackenzie. Fragments of a brachiopod shell were collected here, but the specimens are too inde- finite for determination. The shales are underlain by a great thickness of yellowish-coloured coarse-textured conglomerate, occurring in massive beds and lying in Coarse an almost horizontal position. The pebbles of this conglomerate, be- '=°°glo"ierate. sides the usual quartzose and slate varieties, include boulders of lime- stone often exceeding a foot in diameter. A high range of hills, which follows the left bank of the valley for some miles, is evidently built of this rock. The conglomerate is succeeded by soft bluish-looking shales, but the Bluish shales, junction of the two formations was not observed. The shales after a .■-liort interval, are concealed, but reappear four miles farther down. 218 GEOLOGICAL SURVEY OP CA>fADA. Dolomites. Summary of section, and are replaced shortly afterwards by the conglomerates with which they are associated. The latter evidently constitute the local base of the Cretaceous, as they are underlain by hard bluish unfossiliferous dolomites dipping at high angles, which have a decided Paljeozoic ap- pearance. Below the dolomites the section is again interrupted for a short dis- tance, and the next exposures show soft, bluish, lightly undulating shales, similar to those seen farther up. The shales are interstratified with some harder calcareous strata, the surfaces of which are occasion- ally blackened with carbonized fragments of leaves. The .shales form the banks as far as Driftwood River and beyond. The section obtained in the reach of the Porcupine between the mouth of Bell River and the 'Driftwood may be summarized as con- sisting in descending order of three to four thousand feet of barren quartzites and sandstone, a band of dark shales, probably of Benton age, and a great series of sandstone shales and conglomerates, characterized by holding Aucella mosquensis var. concentrica, the base of which was not determined. The conglomerates and shales which alternate with one another in the lower part of the reach, may possibly represent in a somewhat modified condition, the two lower divisions just enumerated, but their relations were not definitely ascertained. They form the base of the Cretaceous, as Palaeozoic limestone rises from beneath them in the higher undulations. Driftwood Rivee to the Head of the Ramparts. Ch te f Below Driftwood River, the Porcupine makes a sudden bend of seve- the Porcupine pal miles to the north, and then turning to the west it follows a gene- Driftwood ral bearing of about 15° south of west to the head of the Ramparts. River. ijijjg distance between these two points, measured in a straight line, is about fifty miles, but measured along the tortuous course of the river, exceeds seventy-five miles. The river in this reach has a width of from two to three hundred yards. No rapids were met with, and the current does not average over two miles an hour. Alluvial islands occur at long intervals, and sand and gravel bars are more numerous than in the upper part. The valley varies from one to two hundred feet in depth, and occasionally expands to over three miles in width, and in such cases incloses large wooded flats. The volume of the Por- cupine is swelled in this reach by two considerable tributaries, be- sides a number of smaller streams. Old Crow River enters it from the north about half way down, and Blue-fish River from the south a few miles above the head of the Ramparts. The bordering region is undulating and wooded, and is broken below the Old Crow River by a YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 219 short range culminating in high bare peaks, the summits of which attain elevations of over two thousand feet above the river. This range has been called the Old Crow mountains, after a Loucheux chief of that name, in whose hunting grounds they are. The outlines of a number of similar ranges, all apparently independent of one another, were observed cutting the horizon at various points. The geological section becomes much simpler below Driftwood Geological River. Soft bluish shales, with occasional harder calcareous layers, ^®'=''™- are exposed at the mouth of the river, and again three miles lower down. They are nearly horizontal, but can be seen to undulate slightly in the banks. They are succeeded and underlain by a band of hard sandstones and quartzites, which are white on a fresh fracture, but weather to a bright conspicuous yellow. A mile farther down the quartzite band folds over a sudden anticlinel, the central part of which consists of a hard, compact bluish limestone, alternating with a more shaly variety of the same rock. The harder limestone beds have a cracked appearance, and are covered with a network of reticulating calcite veins. They are filled in places with broken crinoid stems and other fossils, and are undoubtedly Palaeozoic. The contact between the limestones and the overlying Cretaceous rooks has the appearance from the river of being slightly discordant. The anticlinal swell just referred to projects above the surface as a Recent distur- hill, and affords evidence in itself of the comparatively recent age of ''^"''^■ the disturbance which produced it, if denudation be accepted as ti measure of time. The curve of the strata is still unbroken by denu- dation, although the fractured condition of the beds shows that they were bent near the surface and not under any great pressure. Three miles below the anticline the same bluish shales which are exposed above it commence again, and are shown in all the scarped banks until within a short distance of Fishing River. They are hori- zontal, or are subject to almost inappreciable dips. As we descend the river they become much softer and lose the fissile shaly cleavage which characterized them where first seen. Interstratified with the shales are occasional harder beds of dark sandstones, which pass in some instances into a conglomeritic condition. The sandstones often show well marked sliokensided surfaces, a feature somewhat unusual Slickensided in soft, apparently undisturbed strata, and the jointage planes are ^|o^i2''^t']' occasionally linei with a thin film of a pinkish mineral, the identity strata. of which has not been made out. Broken coaly particles form a constituent of the shales in some places, but no lignite beds were seen. Limestone concretions, ranging in size up to six feet in diam- eter, and either arranged in lines or scattered irregularly through 220 GEOLOGICAL SURVEY OF CANADA. Recent deposits. Tertiary hed: Re-appear- ance of Cretaceous. Old Crow Mountains. the shales, are abundant in some of the sections, and hold fragments of bones and pieces of the test and occasionally entire specimens of a large Inoceramus. The recent deposits overlying the shales consist of alluvial silts, sands, clays and gravels, and no boulder-clays or other beds referable to the glacial period are seen after leaving the Mackf-nzie. Half a mile above Fishing River the dark Cretaceous shales are overlain unconformably by soft, scarcely stratified, whitish and yel- lowish clays and sands. No fossils were found in these beds, but they evidently represent some horizon in the Tertiary, and their stratigraphical relations would correlate them with the Bear River Tertiary beds of the Mackenzie, which have been referred by Sir William Dawson, on the evidence of their flora, to the Laramie. Two miles below the first exposure. Tertiary beds are again shown in a scarped bank on the left-hand side of the river. They consist here of yellowish and whitish sands and sandstones, interbedded with a grayish, rather hard conglomerate, made up of quartzite pebbles, fragments of shale, and scales of a silvery micaceous schist, unlike any rocks which I have seen in the country. In the next few miles the coloured Tertiary rock? are exposed at all the bends of the river. Bight miles above the mouth of Old Crow River they are raised up by an anticline, and the bluish shales and sandstones of the Cretaceous appear unconformably beneath them. The Cretaceous beds are marked here with numerous impressions of a large Inocera- mus, some of which exceed three feet in length, but the crumbling character of the rock prevented me from obtaining specimens. The Cretaceous shales have a very limited exposure here, and are again replaced in the course of a few hundred yards by the Tertiary. Two miles below the Cretaceous outcrop, a small bed of shaly lignite was observed among the Tertiary beds exposed on the banks. At the mouth of Old Crow River, the dark shales of the Cretaceous again rise to the surface and outcrop along the right bank for a short distance. They are brought up here by the same disturbance to which the Old Crow Mountains, which are directly opposite, owe their existence. The Old Crow Mountains were not examined closely, but they appear to belong to the same system of anticlinal uplifts examples of which are met with so frequently in the valleys of both the Porcu- pine and the Mackenzie. The lower slopes of the range show dark shales of Cretaceous age, while the higher peaks, judging by the wash in the streams which descend from them, are built principally of hard YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 221 quartzites similar to those which characterize the Cretaceous farther up the river. Two miles below the mouth of Old Crow River, the Porcupine bends Re-enter to the left and enters the Tertiary basin again. At the extremity of Tertiary basin, the bend the Tertiary beds are well exposed, and consist of whitish and light-yellowish sands, clays a.nd gravels., The beds have little coherency, and can be crushed between the fingers. Small nodules and beds of ironstone are developed in some places, but none of these proved to be fossiliferous. In the next twenty miles, or as far as the head of the Ramparts, the sands, clays and gravels of the Tertiary appear with little variety in the scarped banks washed out by the river at the elbows of the various bends. The sections have a general whitish appearance when viewed from a distance, but a closer inspec- tion reveals an alternation of whitish, light-yellowish and reddish tints. The rocks of this terrane show little induration, but have suiEcient coherency when freshly uncovered to enable them to stand in almost vertical cliffs of from one hundred to two hundred feet in height; but under a long exposure the face of an escarpment crumbles down into a gentle slope, from which the harder beds occasionally project. The following section was measured a short distance above the head of the Ramparts : — Feet. Light-coloured, slightly indurated sands 10 Grayish sandy clays 20 Friable pebble conglomerate 8 Reddish clays 1 Cream-coloured clays 5 Light-yellowish soft sandy beds 15 59 The rocks change rapidly in composition when traced along the strike, and beds of sand are replaced by clays or conglomerates in the course of a few hundred feet. At the head of the Rampart-, the Tertiary rocks overlap the Creta- ceous and rest directly on hard limestones and quartzites which are probably Palreozoic. Before entering the Ramparts the Porcupine makes a sudden bend to the south. Its course is then comparatively straight along a bearing of S. 60° W. for a distance of thirty miles. Beyond this it bends more Direction of to the south, and its direction as far as Rampart House, a further dis- '''^er. tance of twelve miles, is S. 22° W. Before reaching Rampart House both valley and river run in an easterly direction for about a mile, and then turning again to the south-west the valley becomes somewhat enlarged, and a small flat is interposed between its northern bank and the river, on which the fort is situated. 222 GEOLOGICAL SURVEY OF CANADA. Character of river. Picturesque valley. Band of basalt. Rooks in Ramparts. The Porcupine while passing through the Kamparts contracts con- siderably, and in places does not exceed se-\-enty-five yards in width. Its current is more rapid than in the upper part, and was estimated to run at the rate of about three or four miles and a-half an hour. Short riffles, with a much greater velocity than this occur occasionally, but no rapids or other obstructions were met with, which would prevent the navigation of the stream by small steamers. The Ramparts is a local name employed by the traders to designate a contracted walled valley or canon. The portion of the valley of the Porcupine which passes under this name is exceedingly picturesque. In the upper part the banks rise steeply from the water's edge on both sides to heights of from three to five hundred feet, and their green slopes are everywhere broken by shattered pinnacles and bold crags and cliffs of brilliantly tinted dolomites and quartzites standing almost on edge. As we descend, the inclosing walls become higher and steeper, and the lighter shades are replaced by more sombre colours. Some miles above Rapid River a band of basalt edged with vertical cliffs, appears above, and gradually descends in the banks of the canon until it reaches the bottom, and from this on the gorge is bounded by even, precipitous walls carved out of this rock. The uni- formity of this part of the valley is interrupted at intervals by deep gashes cut by tributary streams through the basaltic covering. Of these the principal one is Rapid River, which enters the Porcupine about seven miles above the post. A mile below Rapid River is the half-way pillar, a projecting column of rock, which was supposed by the traders to be equidistant from Lapierre House and Fort Yukon. The geology of the Ramparts proved to be too complicated to be un- ravelled in the few hours which I was able to devote to it, and can only be indicated in a general way. At the head of the Ramparts the soft unconsolidated Tertiary strata are underlain, as abo\ e stated, by a different and much older series. The beds first seen consist of grayish granular dolomites, interstratified with a close-textured, very compact variety of the same rock, striking nearly north-and-south, and dipping to the west at an angle of 35°. Farther down the dolomites are associated with limestones, quartzites and a band of dark calcareous shales. The rocks have all been subjected to considerable crushing, due to volcanic action, and in some instances have been so shattered that they crumble completely under the influ- ence of the weather or when struck with the hammer. The most important rock in the first twenty miles is a peculiar fresh- looking dolomite, the texture of which is so hard and compact that it might, readily be mistaken in some cases for a quartzite. It contains. YUKON DISTRICT AND NORTHEEN BRITISH COLUMBIA. 223 however, very little siliceous matter, and dissolves almost completely in hydrochloric acid. A remarkable feature of this rock is the varied and changing coloration which it exhibits along the valley. The colours range from pure white through light- and dark-yellow to a bright red, and appear in some places to have an irregular distribu- varied tion, as if due to subsequent infiltration. The beds are usually tilted coloration, at high angles, and the harder strata project above the surface in a more or less ruinous condition, and when brightly coloured, stand out in strong relief against the back ground of dark-green. A band of dark calcareous shales, several hundred feet in thickness, is repeated at several points, and was carefully searched for fossils, but without success. The shales are interstratified at some points with Band of bedded traps, and a number of irregular intrusions of the same rock were also noticed cutting the bedding. They pass occasionally into light-coloured schists, and often show slaty cleavage. Whitish and coloured quartzites, often holding black plumbaginous- looking scales, form an imporiant constituent of the section, and granu- Quartzites. lar dolomites and lead-coloured limestones are also well developed, and occasionally form high cliffs. A noticeable feature of the section is the peculiar alteration of strata softened and decayed for many feet below the surface, with unweather- ed varieties of apparently the same, or only slightly different rocks. I Decayed was at first inclined to believe that some of the soft Tertiary beds which overlie the dolomites and associated rocks at the head of the Ramparts had been folded up with the latter, but was led on farther examination to abandon this view. The succession of the different terranes was not ascertained, as the section is very much confused, and is broken through repeatedly by trap dykes, and would require considerable time for its elucidation. Ten miles above Rapid River the Porcupine has cut through a thick sheet of basalt. The basalt is first seen crowning in long cliffs the banks of the valley, but is brought down by a light westerly incli- nation to the surface of the water, in a distance of six miles. Beyond this if undulates at low angles along the valley, and occasionally rises sufiiciently to expose the underlying shales, quartzites, dolomites and limestones. The basalt shows no evident columnar structure. A well marked Character of horizontal divisional plane was traced for some distance, and may pos- basalt. sibly indicate the junction of two flows, as slight differences in colora- tion were noted above and below it. In texture the basalt ranges from fine-grained to a moderately granular condition, but also becomes vesicular in places, and passes into an amygdaloid. It is composed 224 GEOLOGICAL SURVEY OP CANADA. Rocks underlying basalt. principally of augite and olivine, with some magnetite and titaniferous iron, and is almost destitute, at least in the hand specimens collected, of plagioclase. At Rapid River, and for some distance above and below, the basalt sheet rises in the banks, and the underlying rocks, consisting here of dark shales, old looking cleaved slates, quartzites and dolomites, all dipping at considerable angles, and brought into view. Two miles be- low Rapid River the basalt dips below the surface, and the older rocks are not seen until near Rampart House. Rampaet House to the Mouth of the Porcupine. Reach Rampart House. Travel by night. Rampart House. Indians at Rrimpart House. Arrangements for trip. We reached Rampart House on the 20th July, the descent of the river from Lapierre House having occupied four days. We travelled mostly at night, in order to avoid a strong head wind which blew up the river every day. It is customary, however, in these latitudes to work in the night rather than in the day, during the period of six weeks or so in which the sun remains above the horizon. The light is suificient, while the temperature is somewhat lower, and the mos- quitoes are less troublesome. Rampart House is the most distant of the Hudson's Bay Company's posts, and was established to replace Fort Yukon, after the site of the latter had been determined to be in Alaskan territory. This post was originally situated twelve miles farther down the river, but the posi- tion of the buildings in regard to the boundary being doubtful, these were burnt by the Hudson's Bay Company, and new buildings were erected on the present site. As a fur post it barely pays expenses, owing to the heavy cost incurred in the transportation of furs and goods, and is kept up mainly as a protection against the encroachments of traders from the west. The Indian hunters trading at Rampart House number about eighty. They belong to the Loucheux branch of the Tinneh family, but speak a slightly different dialect from that used by the Mackenzie River Loucheux. They are christianized, and missionaries sent out by the Church Missionary Society have been working among them for some years. A small church has been built in the vicinity of the post. I remained a few hours at Rampart House for the purpose of taking an obseivation for latitude, and making arrangements for the trip to Fort Yukon. An Indian was engaged to accompany us, and I was fortunately able to obtain a supply of dried meat from Mr. Firth, the officer in charge of the post. A report had reached the fort that a steamer belonging to the Alaska Fur Company would pass Fort Yukon YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 225 on its way to Forty-mile Creek, in a few days, aind I decided, as we would now traverse Alaskan Territory, and had no object for delay, to hurry down to the forks as fast as possible and endeavour to secure a passage on the steamer up the Yukon as far as the Canadian boundary. We descended the river, battling against a strong head wind the whole way, in three days, but learned on our arrival, much to our disappoint- ment, that the boat had already passed. The distance from Rampart House to the mouth of the Porcupine, measured in a straight line, is about one hundred miles, but is fully one hundred and fifty by the course of the river. The general direction is about S. 60° W. The current is uniform, with few riffles and no rapids, and has an average rate of about three miles an hour. The Ramparts continue for two miles below the fort, and present the same features which characterize them above. Opposite the fort, the basalt sheet rises again, and alternating bands of shales, slates, limestones, dolomites and quartzites, with, in one place, a white cliff of coarsely crystalline calcspar, are seen beneath. Eight miles further Rooks below down, it sinks beneath the surface, and massive walls of basalt and ^<^^^^- amygdaloid then border the river as far as Howling-dog Rock, at the foot of the Ramparts. Below Howling-dog Rock, the river intersects for some miles a wide gravel flat, the former site of the fort, and then bends away to the „,. , ° „ , . , Wide gravel north. An exposure here of clays, grayish and yellowish sands and flat. soft pebbly conglomerate, shows that the river has entered a second Tertiary basin. The r^icks aro all very soft, and are similar in appear- Tertiary ance to those occurring above the Ramparts. No fossils were obtained I'ooks. from til em. Sections of unconsolidated Tertiary strata are exposed again at the extremity of the bend, and at several points in the next five miles, and are then replaced by shales and slates of an older series. They are overlain by recent silts, sands and gravels. The width of the Tertiary basin along the river does not exceed Width of seven miles, but it is highly improbable that this represents its full jJ^^'J'"^'^ size, and the appearance of the flat country to the north would seem to show that it extends for a considerable distance in that direction. The shales and slates which succeed the Tertiary are overlain in the course of a mile by basalt. The basalt is exposed for a short distance Exposure of on the left-hand bank, and probably represents a projecting spur from ^^^salt the main sheet. It was not seen again. Two miles below the basalt exposure the shales and slates are associated with rusty coloured lime- stones, and at the mouth of Sucker River, three miles farther down, the same rocks are again seen. In the next fifteen miles yellowish- weathering limestones, rising occasionally into abrupt cliffs, are ex- 15 226 GEOLOGICAL SURVEY OP CANADA. Lower Ramparts. A^e of limestones. Wide plain. posed in a number of places. This part of the valley goes by the name of the Lower Ramparts, but the rocky walls are low and disconnected, and are seldom developed on both sides of the valley so ag to form a canon. The limestones yielded Atrypa reticularis, together with some fragmentary specimens of corals, and are probably referable in part at least to the Devonian. At the lower end of the Lower Ramparts, the limestones are interbedded with dark shales similnr to those associated ■with them above Sucker River. The shales are unfossiliferous, but from their position are proba!ilv Cretaceous. They are exposed at in- tervals for two miles, and are then, overlain unconformably by the yel- lowish and light-reddish clays and sands of the Tertiary, and the lat- ter in turn sinks in a short distance, beneath alluvial sands, clays and gravels. From this point on to the mouth of the river, a distance in a direct line of sixty miles, no further exposures of the older rocks are seen. The valley disappears, and the river serpentines through a wide plain, elevated only a few feet above its surface. At the elbows of the nume- rous bends the cut-banks show small sections of clay and false-bedded sands and gravels, either alluvial or lacustral in their origin. The bordering plains extend to the horizon on either side, unbroken by a single elevation, and their extent and uniformity, taken in connection with the character of the beds seen along the valley, afiFord stron" grounds for the assumption that a lake basin or abnormal expansion of the river once existed here and has since been silted up. Tributaries of The principal streams which the Porcupine receives in this part of its course are the Salmon from the right, and Black River and Little Black River from the left. For some miles above its mouth it divides around numerous islands, and branching channels become so frequent that care has to be exercised to select the right one. Before reaching the Yukon we left the main stream, and turning to the left entered a small channel which is reported to flow in opposite directions, depend- ing on the relative heights of the water in the two rivers. Our hopes 'that the current would set in our favour were doomed to disappoint- ment, as we no sooner rounded the bend than we passed from the brown water of the Porcupine into the milky flood of the Yukon, and found ourselves struggling against an impetuous current of five or six miles an hour. Fort Yukon is situated a mile and a-half above the con- fluence of the two streams, and it required nearly two hours of hard work to ascend that distance. Porcupine. yukon district and northern british columbia. 227 Old Fort Yukon to Forty-mile Ceeek. At Fort Yukon I had the pleasure of meeting the Rev. Mr. Canahan, of the Church Missionary Society, who was on his way to Nukluka- hyet to take charge of a mission. Mr. Canahan's news was far from cheering, as he informed us that the steamer which we expected to meet had passed up the day before. Two courses were now open to us in order to reach the coast and Route to the outside communication ; either to descend the river to St. Michaels and *'°^*" sail from that point to Victoria or San Francisco, or to face the nine hundred miles or so of rapid current and difficult navigation above, and ascend it to the head of the Lewes, and then cross the Coast Range by the Chilkoot Pass. The former is by far the easier route, as the lower Yukon possesses a strong steady current, and is free from dangerous rapids, and the temptation to adopt it was almost irresistible, but as it lay altogether outside of Canadian territory I decided, if possible, to try and ascend the stream. To do this, hoN\-ever, promised to be a mat- ter of no ordinary difficulty. The short, square-sterned boat which I had hitherto used, was built to carry » load down stream and was altogether unsuitable to make an up-stream journey in, and an attempt to force it for hundreds of miles against a five oi' six mile an hour cur- rent seemed well nigh hopeless, but no other was available. Our pro- visions were also running short, but Mr. Canahan kindly supplied our deficiencies in this respect to some extent, and we expected to be able to obtain fish from the Indians along the river. Trader John, the In- dian who piloted us down from Rampart House, and had i>roved him- self a capable and willing fellow, was induced, after some persuasion, to accompany us as far as Forty-mile Creek, and on the 25th of July, after a delay of four days, we succeeded in making a start. A sketch of early exploration on the Yukon has been given by Dr. p ^ , G. M. Dawson, [see pages 134 and 177,] and it will be unnecessary to exploration, repeat it here. Fort Yukon, which was originally one of the best built forts in the north, is now a thing, of the past, and with the exception of one of the outbuildings, which has probably also dis- appeared by this time, has been torn down to supply wood for the steamers pi) ing on the river. While in possession of the Hudson's Bay Company some gardening p j • was done in the vicinity of the fort, notwithstanding the fact that it is Fort Yukon, situated almost on the Arctic circle. Potatoes and other vegetables were raised, and barley is reported to have ripened. On the Macken- zie River, gardening,' ceases to the north at almost the same latitude. 15J 228 GEOLOGICAL SURVEY OP CANADA. Numerous channels. Difficult navigation. Length of island-filled stretch. Above Fort Yukon the river cuts through the same low wooded pkiii- which has been described as bordering the Porcupine above its mouth, and like the latter, only on a much greater scale, its waters flow by numberless interbranching channels through a labyrinth of islands. Its width in this reach is reported to range from three to ten miles or more, l.ut as we kept the left bank the whole way up, and the opposite shore was always concealed by islands, I was unable to verify this statement. The river however has here, practically, no confining val- ley, and the soft sands and clays which underlie the adjoining plain offer little obstruction to an indefinite expansion. New channels are opened up with each succeeding ilood, and older ones blocked by the deposition of sediment. Notwithstanding the great width, the velocity of the current is undiminished, and the up-stream navigation of this part, especially in high water, is attended with greater difiiculty than any other portion of the river. Beaches are seldom present, and tracking is impossible except for trifling distances at long intervals. The water along the cut-banks is too deep a few feet from the shore for poling, and to advance we were obliged to combine the use of an oar on the outside, a pole on the inside, while I steered and paddled behind. Even this complicated method of propulsion often became impracticable, and progression in some places was only attainable by clingino- to the overhanging branches and pulling ourselves up foot by foot. For long reaches the banks are undermined by the wash of the stream, and as the boat creeps along beneath the overhanging wall from which masses of sand and other material are momentarily falling, it is constantly threatened with destruction. Vexatious delays are also caused by the numerous trees which have been undermined and have fallen forwards into the stream, but still cling to the bank by their roots. Above these, driftwood accumulates until it forms a projecting point around which the current swirls with redoubled velocity, and" to pass them is often the work of hours. The width of the river is so great that difiicult spots cannot be avoided as in other streams by crossing to the opposite bank, and the ascent must be made entirely on one side.' The ea.stern side is usually preferred. The length of this expanded island-filled stretch above the mouth of the Porcupine, was estimated at seventy miles, althought a miner the only one so far as I could hear, who had ascended it, assured me that it could not be less than three hundred and fifty miles. His esti- mate must, however, be taken rather as the measure of the difficulties he encountered than of the true length. We made the ascent in five days, but to do this were obliged to work at our full capacity for four- YUKON DISTRICT ANU NORTHERN BRITISH COLUMBIA. 229 teen or fifteen hours a day. With a proper boat much better time could liave been made. Tlie older rocks along this stretch are deeply buried beneath a thick older rooks accumulation of recent sands, clays and gravels, and no exposures were "i°a«"- anywhere seen. Before reaching the head of the islands we crossed to the right bank, Chief Senates. and on our sixth day passed the camp of Senatee, the head of the Fort Yukon Indians, and the most powerful chief in the whole Yukon coun- try. Many stories are told of cruel murders and rapes committed by Senatee in days gone by, but time has now tamed his ferocious dispo- sition, and changing circumstances have shorn him of much of his once absolute authority. He received us hospitably, and after an ex- change of lish for tobacco had been made, commenced a long oration, descriptive of his enduring love for the English (Hudson's Bay Com- pany) and his regret that they had left the country, and his general dislike of the Yankees (Alaska Fur Company), but as time was pre- cious the harangue was cut short by the present of a couple of hand- fulls of tea, the probable object for which it was made, and we pro- ceeded on our journey. Above the Indian camp the river commences to contract. It bends River more to the east, and is bordered by narrow beaches, a welcome change, Avhich enabled us, much to our relief, to substitute the tracking-line for the poles and oars, and to make better progress with less exertion. The shore here is strewn with blocks of dark carbonaceous looking cleavable shales, and on the eastern side of the river, bedded rocks, probably of the same description, could be seen in situ, but were not examined, as a crossing meant the loss of several miles wearisomely won, and was only niade when rendered necessary by the exigencies of the navigation. The elevation of the country bordering the river is here higher, increased and a considerable valley is developed. Fifteen miles above the elevation of Indian camp the river is reduced to about halt a mile in width, and reginn. assumes an easterly direction. A crossing to the eastern side was effected here, and a line of cliffs which made their appearance seven miles below, was found to consist of massive eruptive rocks, which are described in my notes as basalts and fine- and coarse-grained dolo- Occurrence of mites but unfortunately the specimens ha\'e mis-carried. The volcanic basalts. rocks here are, as a rule, coarser textured and older looking than those seen at the Ramparts on the Porcupine, and no amygdaloids or vesi- cular varieties such as occur there were noticed. It is uncertain whether they form parts of the same flow or not, although the relative position of the two areas and the northerly trend of the elevated coun- 230 GEOLOGICAL SURVEY OP CANADA. Contact of basalt sheet with underlying rocks. Second basaltic occurrence. Width of igneous area. Conglomer- ates. Depth of valley. Foldings. try leads to the supposition that they do, and in this case the westerly edge of the flow would correspond in a general manner with the east- ern boundary of the great plain which extends up the Pelly- Yukon and Porcupine from their confluence. The basalt and associated igneous rocks inclose numerous detached masses of shales derived from the underlying terranes, and dark shales and limestones tilted at high angles project up into them in many places in a similar manner to that noticed on the Porcupine. They were traced along the river for fifteen or twenty miles, and are then replaced by a confused succession of shales and limestones, probably referable to a large extent to the Carboniferous, alternating with the dark shales, sandstones and quartzites of the Cretaceous. The whole series is greatly disturbed, and folded closely together, and the beds often show vertical dips. The general strike is S. S. E. The shales and associated limestones and sandstones are exposed in continuous sections for seven or eight miles, and are then again almost buried be- neath a second basaltic occurrence, and only the higher points of the old rugose surface rise into view above the bottom of the valley. The line of contact of the Tertiary basalts with the underlying sedimentaries in this region, both on the Yukon and the Porcupine is, speaking generally, nearly coincident with the present level of the rivers. The second basaltic area, where cut by the river has a width of ten or twelve miles. Above it the rocks are concealed on the eastern side for some distance, and then the shales and sandstones of the Cretaceous outcrop on the banks, and are present all the way to the Tatonduc, a distance of about thirty-five miles. The beds are sometimes horizon- tal, but usually undulate at varying angles, but Seldom to such an ex- tent as to expose the underlying Palseozoic limestones. Conglomerates consisting largely of small pebbles of schist, quartzite and slate, im- bedded in a hard sandstone matrix, are occasionally associated with the shales and sandstones, and in one place about eight miles above Charlie's village, attain a great development. They overlie the shales at this point. Some miles below the mouth of the Tatonduc speci- mens of the characteristic Queen Charlotte Islands' formation fossil Aucella Mosquensis var. concenirica, were collected from the beds below the conglomerates. The valley of the Pelly-Yukon becomes gradually deeper as we ascend, and above Charlie's village the banks were estimated to have an elevation in places of from eight hundred to a thousand feet. Above the mouth of the Tatonduc the foldings increase in severity, and the Palseozoic limestones and shales are brought up and alternate YUKON DISTRICT AND NORTHERN BRITISH COLUMBIA. 231 in broad bands with the shales, sandstones and conglomerates of the Cretaceous. A black carbonaceous variety of shale is repeated at a number of points, and often shows polished slickensided surfaces, due to the crushing to which it has been subjected, which simulati^ very closely the appearance of anthracite. The alternation of the Cretaceous beds with the older limestones and shales continues until near the Boundary, [141st meridian] a distance measured along the river of about forty miles. No fossils were found in this part of the river, but lack of time prevented a careful search for them.* A short distance below the Boundary, the rocks referred to above are Schistose replaced by a very different looking set, consisting mainly of altered r"<^^^ replace volcanic rocks, and characterized by a general greenish colour. Among stones, etc. them are important bands of serpentine, hard quartzose, sheared and altered greenish schists, softer greenish chloritic looking schists, silvery mica-schists, diabases, shales and slates. Picrolite was noticed in several places, and beds of chert are not uncommon. The schistose beds must underlie the limestone series (Carboniferous ?) although the dip would place them above, as the latter is overlain directly in many places farther down the river by the shales, sandstones and conglomer- ates of the Cretaceous. East of the river above the Boundary, and running parallel with it Limestone fur some distance, is a high naked range, built of limestones striking ''""&«• * Mr. Ogilvie has furnished me with the following geological notes obtained on his traverse from the mouth of the Tatonduc across to the head water of the Porcupines and down the latter stream to its confluence with Bell River : Dark shales overlain by Bne-grained conglomerates which are doubtless the equivalents of the Cretaceous beds observed on the Pelly-Yukon, occur along the Tatonduc for twenty-seven miles from its mouth. At two points in this distance, limestones rise to the surface from beneath the shales, and by their superior hardness contract the valley into short canons. The conglomerates are described by Ogilvie as weathering in some of the • ridges and mountains into picturesque castellated cliffs. In the upper part of the valley of the Tatonduc the Cretaceous shales and conglomerates are replaced by gray- ish Palaeozoic limestones. Similar limestones were also met with all across the water- shed separating the Tatonduc from the Porcupine, where they form a range of high mountains, and down the latter stream as far as the Cathedral Rooks. At two points on the Porcupine, viz., two miles and ten miles respectively below the forks, the limestones are underlain by reddish mottled sandstones similar in appearance to those occurring at Sault Ste. Marie. Below the Cathedral Rocks dark shales of Cretaceous age appear again, and were the only rocks observed all the way to the mouth of Bell River where connection was made with my traverse down the Porcupine. An interesting fact noted by Ogilvie, is the emission, near Sheep Mountains, of sulphuretted hydrogen gas in large quantities from the surface, and he states that he was informed by his guide that a small lake a short distance from the trail is kept constantly agitated by the escape of similar gas, and that the sulphurous fumes prove fatal to all animals which venture near. East of the Rooky Mountains sulphuretted hydrogen gas issues in numerous places from the petroliferous limestones of the Devonian, but it is jiossible that its jiresence here is due to volcanic action. 232 GEOLOGICAL SURVEY OF CANADA. in a south-easterly direotion and dipping at high angles towards the river. Resting on these, aud apparently overlying them, are the rocks just described. The schistose beds alternate occasionally with limestone bands, altered in places into marble, and are tra\ei^ed in all directions by quartz veins. They are exposed all the way to Forty-mile Creek and Economic beyond. They are of the greatest economic importance, as they con- schist be't. stitute the gold-bearing rocks of the district, and are on the strike of the same metalliferous zone on which Cassiar and the other principal mining camps to the south are situated, and which has now been traced northwards across Canadian territory from the 49th parallel to the eastern boundary of Alaska, a distance of fully tw'elve hundred miles. Four miles below the mouth of Forty-mile Creek, a small stream Coal seam. was passed coming in from the opposite side, on which a coal seam is reported to occur, but nothing definite was learned about it. FoETY-MiLB Creek Anive at We arrived at Forty-mile Creek on the 9th August, fourteen days Forty-nnle having been occupied on the journey up from Fort Yukon. This is the head-quarters of the miners on the Yukon, [1888] and thirty or forty men were camped here waiting for reports from various pro.specting parties which were exploring the surrounding country, and ready to start Unfavourable at a moment's notice. The present season proved very unfavourable for mining- operations on account of the persistent high water, and with the exception of a short period early in the summer, very little work was Yield of gold, done on Forty-mile Creek, and the amount of gold taken out was esti- . mated at scarcely $15,000. A few dnys before my arrival a, stampede had been made for Beaver River, a northern tributary of the Yukon, which is stated to enter the latter about a hundred and twenty miles below the mouth of the Porcupine, but with somewhat disastrous Stampede of results. The amount of information required to stampede a mining miners camp is very small, and in the present case was almost ridiculous. A report was brought up by the men on the steamer that a miner had boarded the boat at the mouth of Beaver River, and after talking in a hurried manner to the captain, had suddenly departed, and in his haste had left his purse behind him. The miners reasoned that noth- ing but a rich find would cause such an excitement, and a hundred and fifty men immediately loaded their boats and started on a wild goose chase down the river only to meet with disappointment at the end of their journey. A few received a passage up again on the steamer, YUKON DISTRICT AND NORTHERN nRITISH COLUMBIA. 233' but the greater number drifted on down towards St. Michaels, and left the mining country altogether. Prospecting in the Yukon country is attended with peculiar disad- Difficulties in vantages, and requires men of more than ordinary perseverance and P™^P'"=''"S- eudui-ance for its successful prosecution. The Yukon itself is very swift and difficult to navigate, but is easy as compared with many of the tributary streams which are simply a long succession of callous, whirlpools and rapids, and skill and courage in a high degr'ee, in addi- tion to a golden bait, are necessary in order to brave the perils of their ascent. Provisions are high-priced and scarce, and in the upper part of the river cannot be obtained at all, and supplies for the summer must be packed across the Coast Range in the early spring and sledded for two hundred miles down the river to the lower end of Lake La- barge before the break up of the ice, in order to be on hand when the season opens. Added to this is the shortness of the period available for work, which under the most favourable circumstances never exceeds three months, and in seasons of exceptional high water, such as the present, is very much less. In view of these drawbacks prospecting must proceed slowly, and up to the present has been confined almost entirely to the larger and more accessible streams. The most important strike made so far [1888] has been on Forty-mile Creek, on which coarse gold was discovered in 1886. This discovery occasioned a rush towards the creek of nearly all the miners in the district. In 18>^7 over 200 men were actively and successfully employed along the numerous bars, and the total yield for the .season was variously estimated from $65,000 up to $150,000. The present season ^'"^'^ °^ ^°'''- has proved much less remunerative, partly owing to the unfavourable state of the water and partly to the fact that the rich claims first dis- covered have been worked out, the auriferous gravels being of little depth and easily exhausted. The most productive part of this stream is west of the Alaska boundary. It has a total length, according to the miners, of about 150 miles, of which only the lower twenty miles are in Canadian territory, and in this part very little work is now being done. The average va^ue of labour is §10 a day, and bars which yield less than this are soon abandoned. No important strike was made anywhere in this district in 1888. A lode of argentiferous galena crosses Forty-mile Creek a short dis- Argentiferous £r£ilfn.£ii tance above its mouth, and a specimen brought back by Mr. Ogilvie was assayed in this office by Mr. Hoffmann, and found to contain 38- "A oz. of silver to the ton. Tellurium is reported from the head of the stream, and seams of coarse serpentine asbestus occur near its mouth. 234 GEOLOGICAL SURVEY OP CANADA. Prepare for trip up river. Some days were spent at Forty-mile Creek in making preparations for the long trip up the river, and in building a long, narrow, sharp- ended boat, modelled after the pattern of those used by miners in ascending swift streams. Trader John, who had helped us gallantly in the ascent from Fort Yukon, returned home from here, and a miner named Buckley, was engaged for the remainder of the trip. Forty-mile Creek to the Mouth of the Stewart. Leave Forty mile Creek. Character of river. Character of valley. We left Forty-mile Creek shortly afternoon on the 14th August, and passed the mouth of the Stewart early on the morning of the 19th. The distance measured along the river is about 120 miles. The direction as far as Fort Reliance is nearly east, but beyond that point the river bends slowly round and runs almost west for a few miles, after which it maintains a general southerly course to the mouth of the Stewart. The width of the river in the lower portion seldom ex- ceeds half a mile, but above Fort Reliance it gradually enlarges, and in the southerly reach occasionally exceeds a mile in width. In the expanded stretch, however, much of the surfa:e is occupied by islands. The current is swift and uniform, and at a medium state of the water runs at the rate of five miles an hour. Beaches line the shore, and tracking, except at very high water, is possible nearly the whole way. The valley of the Yukon between Forty-mile Creek and the Stewart, and on to the mouth of the Pelly, is cut through an elevated undulat- ing plateau, on which rest numerous low ranges of rounded and partially bare hills, but is not crossed by any well defined moun- tain range. It is somewhat uniform in appearance, but affords many picturesque and even grand views. Bluffs of rock of a more or less precipitous nature, are of constant occurrence, and bold rampart- like ranges of interrupted cliffs, separated or continued upwards by steep grassy or wooded slopes, characterize the banks for long reaches. The flats are few and unimportant, and as a rule the river washes the base of the banks on both sides. The width of the valley varies from one to three miles, and its depth from five to fifteen hundred feet. Its great size, taken in connection with the hard character of the crystal- Age of valley, line rocks through which it has been excavated, afford evidence of great age, and point to an origin long antecedent to the glacial period. The same fact is also emphasized by the remarkably uniform grade which the river has worn across terranes of heterogeneous hardness ranging through the whole geological scale in its long course from Rink Rapid to the sea, a distance of nearly 1700 miles. YUKON DISTRICT AND NOKTHERN BRITISH COLUMBIA. 235 Rock-sections are numerous along the valley, but the geology is in- tricate and difficult, and somewhat confusing to a traveller passing hurriedly through along one line. Above Forty-mile Creek, the same series of micaceous and altered greenish schists and slates which obtain Rooks below are continued for six or eight miles, and are then replaced by grayish granite. The schists are traversed by both quartz and calcite veins, usually of small size. They dip at high angles, and have a general south-southeasterly strike. The granite is a medium-grained, some- what altered variety, showing no distinct foliation. It is composed principally of quartz, plagioclase, orthoclase, musoovite, biotite and epidote. From the central granitic mass a set of dykes radiate out into the surrounding schists. The dykes often pass in between the Dykes. beds, but were also seen cutting across them. Above the granite boss the schistose series resumes again and continues as far as the bend above Fort Reliance. These rocks are broken up in several places by granite intrusions, and the granite appeared to be itself cut in one place by a darker and more basic igneous rock of which I did not obtain speci- mens. A specimen of rather coarse textured granite, collected above Fort Reliance, holds numerous grains of copper-pyrites, and may pos- sibly be auriferous, but has not been analysed. The schists occasion- ally show copper stains running parallel with the foliation. Some miles above Fort Reliance, a band of dark argillites, interbed- ded with limestones, crosses the river. Like the schists it is traversed by numerous quartz veins. It is followed for several miles bv schists of various kinds, but usually more or less micaceous, and occasionally passing into a well foliated rather fine-grained mica-gneiss. The lat- ter increases in relative proportion, and assumes a coarser texture as „ ^ ... . Granite we ascend the river. Igneous granitic intrusions occur every few intrusions. miles. Fifteen miles below Sixty-mile Creek, a peculiar looking soft green- ish rock consisting of glossy talc, and green serpentine was found at several horizons interbedded with the schists. The square columnar crvstals from which the serpentine has been derived still preserve their original form, and penetrate in all directions the lighter coloured tal- cose matrix. They show under the microscope traces of the quadran- <'ular cleavau'e of augite, but are otherwise completely altered. Opposite Sixty-mile Creek the exposures show lustrous musoovite- Kxposures mica-schists, dark biotite-micBrSchists grading into mica-gneisses, opposite Si xty- , J! 1 • • 1 • T ■ , ™il^ Creek. hornblende gneissic schists and compact leisitic roclcs indistinctly foliated. With these are associated grayish biotite-granites, and a coloured granitoid or syenitic rock, consisting of orthoclase, plagioclase quartz, chlorite and epidote. The proportion of quartz in this rock is 236 GEOLOGICAL SURVEY OP CANADA. small and the crystals are greatly crushed and broken. A second variety of this rock shows large, coloured crystals of felspar, porphy- ritically distributed through a fine-grained microcrystalline base. Rocks between From the vicinity of Sixty-mile Creek to the mouth of the Stewart, Greek a'nd the *^® rocks are older looking and more distinctly gneissic then those ob- Stewart. served farther down the river. The most abundant variety is a light reddish-coloured, medium-grained, well foliated inica-gneiss, composed of quartz, orthoclase, plagioolase, biotite and muscovite. Epidote, chlorite, calcite and ilmenite are also nearly always present. In some cases the micas are replaced altogether by chlorite. With the gneisses are associated several varieties of mica-schists, and occasionally some hornblendio schists. Numerons veins of coarse pegmatite, quartz, and less frequently calcite, cut the beds in all directions, and are present in nearly every section. Stewart River to the Pelly. The Stewart The Stewart River has not been explored. At the mouth it is two River. hundred yards wide, but the current is not rapid, and it apparently carries less water than either the Pelly or the Porcupine. It is re- ported to be navigable for two hundred miles above its mouth. Gold was discovered on it in 1885, and in that and the following year, gold Yield of gold, to the value of $100,000 was obtained from it, according to an estimate made by Dr. Dawson on information from miners. The principal bars were, however exhausted in these two years, and in 1887 the yield diminished to about $5,000, but this was partly due to the withdrawal of most of the miners to Forty-mile Creek. In the present year an attempt was made by two prospectors to trace the gold to its source but without success. " Colours " were present as far as they ascended, but not in paying quantities, and no tributaries were found down which the gold in the lower part of the river might have descended. They report that the river grows in its upper part by the reception of num- erous small streams heading in swamps, none of which are gold- bearing. An important feature of the river is the constant recurrence Hifjh gravel along it of high gravel terraces, most of which are more or less auri- ferous, and occasionally yield a^ high as a cent a pan, and could pro- bably be worked profitabl)^ on a large scale. It is possible that the gold found on the bars in the river is concentrated from these. From the mouth of the Stewart the Pelly- Yukon trends in a south- westerly direction for ten miles to its junction with White River. In this reach it averages a mile in width, and is filled with islands. The banks of the valley are steep and rooky, and were estimated at from terraces. YUKON DISTRICT AND NORTHERN BRITISH COLOMBIA. 237 800 to 1000 feet in height. Garnetiferous mica-schist was noticed a short distance above the Stewart, and dark hard mica-schists and light- reddish well foliated gneisses are exposed all along. An extensive system of quartz veins is developed along here, and gold-bearing quartz is reported to have been found. White River, like the Stewart, has not been explored, and has only Wliite River, been ascended by miners for a distance of sixty miles from its mouth. A view up it from the opposite bank of the Pelly- Yukon showed a wide valley filled with countless bars and islands between which the divided stream threaded its tortuous course. Its current is swift, and it preci- pitates itself into the Yukon with a force sufficient to drive its muddy water half way across, but scarcely to force it up on the opposite bank in the manner described by Schwatka, and the two streams flow side by side for several miles before their waters become completely fused. The turbid character of the White River is famous, and sufficient sediment is brought down to change the colour of the whole Pelly- Yukon flood from a pale green to a milky white. White River is re- ported to head in glaciers descending from high mountains, but nothing very definite is known concerning it. " Colours " of gold have been obtained from it, but no paying bars have so far been discovered. From White River to the Pelly the distance is ninety miles. The white River direction is wt first south-westerly, but the river soon bends round to to the Pelly. the east and follows an E.S.-E. bearing the greater part of the way. The width of the river is somewhat reduced, and varies from a quarter to half a mile, while the current averages about five miles an hour. Islands occur at intervals, but are less numerous than in the reaches below. The valley preserves its usual depth of from 800 to 1000 feet, but for some distance above White River the banks are more wooded and show rocky bluffs above supported on steep slopes below. Farther up it is again bordered by steep gneissic and basaltic clifis. Gravel terraces occur occasionally and appeared to increase in height as we fiscendcd. The prevalent rock for many miles above White River is a hard, character of eranular and often moderately coarse-grained, Archasan looking ™oks. mica-<'neiss of somewhat varied mineralogical composition. This occasionally passes into mica-, chlorite- and hornblende-schists. The minerals most commonly present are quartz, plagioclase, orthoclase, biotite, muscovite, chlorite, hornblende, epidote, calcite, magnetite and ilmenite. In some cases the rock is greatly crushed and altered, and the micas and hornblendes are replaced almost entirely by chlorite. Plagioclase is always present and often in greater abundance than ■orthoclase. Quartz veins occur less frequently as we ascend, and are 238 GEOLOGICAL SURVEY OF CANADA. replaced by veins of coarse pegmatite. Igneous rocks are almost absent in the lower part of this reach, but coarse granites and diorites were met with about twenty-five miles below the mouth of Tertiary lavas, the Pelly. Three miles above this point the older rooks are covered with vesicular lavas of Tertiary age. The lavas are exposed for a short distance and then recede from the river at a bend which the lat- ter makes around a high terrace built of coarse gravels. Resting on this terrace is a small bed of white volcanic ash, frequent occurrences of which were noticed on the Pelly by Dr. Dawson, [see p. 40.1 Above the gravel terrace the lavas resume, and are present all the way to the Pelly. They are underlain by gneisses similar to those described before, and by granites and diorites. A specimen of diabase-porphyry composed of crystals of augite and plagioclase porphyritically dis- tributed through a micro-crystalline chloritic base, was obtained from the cliffs west of the flat on which Fort Selkirk was situated. After reaching the site of Old Fort Selkirk, at the confluence of the Pelly and Lewes rivers, my journey was continued up the last-named stream and over the Chilkoot Pass to the sea, which was reached on the 15th September, 1888. This part of my route, therefore, coincides with that followed and examined in 1887 by Dr. Dawson, and is described by him [in previous pages.] NOTES ON THE GENERAL GEOLOGY OF PART OF THE YUKON DISTRICT. [The subjoined notes extracted from Mr. McConnell's report are those bearing upon parts of the Yukon District described in the fore- going pages.] ******* In the Rooky Mountains themselves no pre-Cambrian rocks were observed. Crystalline West of the Rocky Mountains crystalline schists are largely the'Yukon.^ developed along the valley of the Pelly- Yukon. They were first met with, in ascending the river, near the International Bound- ary, and were then traced southwards by numerous exposures as far as Fort Selkirk, at the junction of the Lewes and Pelly, and they continue on up the Lewes about thirty miles farther. The belt of crystalline rocks has a width of something over a hundred miles and strikes in a south-easterly direction from the International Boundary, or long. 141° W. diagonally across the Pelly- Yukon and up the Pelly to near its head, and then continues on YUKON DISTRICT A.ND NORTHERN BRITISH COLUMBIA. 239 across the Frances River. The extension of the belt beyond the Fran- ces River is not known. It does not reach the Liard, however, as no crystalline schists were noticed on that stream. South of the latitude of the Liard the crystalline rocks, like the Rocky Mountains themselves, do not continue southwards along their old strikes, but along parallel strikes a couple of degrees farther west. The break in the continuity of the Rooky Mountains, which is referred to on another page, has its counterpart in a break in the continuity of the belt of crystalline rocks which usually accompanies them on the west. The dips of the crystalline rocks along the Pelly- Yukon are usually Dip of rocks, westerly and at high angles. Along the eastern edge of the crystalline belt, the rocks are character- Character of ized by a general greenish colour and consist largely of altered volcanic rocks. rocks. The most important variety is a sheared and altered greenish quartzose schist. With this are associated greenish chlorice-bearing schists, lustrous mica-schists, diabases and serpentines. The schists are interbedded with occasional bands of slates and crystalline lime- stones, and are broken through in many places by igneous intrusions. The green schists are underlain, in ascending the river, by a great thickness of well foliated mica-gneisses alternating with mica and horn- blende-schists, which are distinctly Archajan in appearance and litho- logical characters. They bear a strong general resemblance to the Archaean rocks, recently described by Dr. Dawson from the western part of the Selkirks. Granite intrusions occur in the gneissic area, but less frequently than in the green schist belt. A hundred and twenty miles below the Ramparts, the Mackenzie enters a third Cretaceous area, and the largest one on the river. Cre- taceous beds appear in the banks a short distance below old Fort Good Hope and extend down the ^Mackenzie to the head of its delta, and westwards across the Rocky Mountains and down the Porcupine to about long. 139° W. They consist on the Mackenzie of coarse shales 'nterstratified with some sandstones and fine-grained conglomerates ; in the mountains of several thousand feet of barren sandstones and quartzites underlain by dark shales, and on the Porcupine of the same two series underlain by a great thickness of alternating shales, sand- stones and conglomerates, holding Aucella Mosquensis var. concentrica. The intermediate dark shales are probably of Benton age, while the lower division so far as the fossil evidence goes, represents the Queen Charlotte Island formation and the Dakota. * [See preceding pages.] 240 OEOLOGIOAL SURVEY OJ'' CANADA. Cretaceous Cretaceous shales holding AuceUa and passing upwards into fine- rocks on the grained conglomerates, occur on the Yukon for many miles above and \ ukon. o o > J below the mouth of the Tatonduc and were traced by Ogilvie up the latter stream for some distance. They have been greatly disturbed and are folded up in broad bands with the underlying Palaeozoic lime- stones. It is highly probable that the various Cretaceous areas scattered along the Liard, Mackenzie, Porcupine and Pelly-Yukon, were origi- nally connected and have been separated by denudation. This is shown by the structure of the region, by the fossil evidence so far as it goes, and by the lithological and ttratigraphical resemblances which prevail throughout. Tertiary. Tertiary west West of the mountains soft incoherent beds resembling those at the of mountains, mouth of Great Bear River and probably of about the same age, are ex- posed along the Porcupine River in the vicinity of the Old Crow Moun- tains. They were first seen about half a mile above Fishing River and were traced as far as the head of the Ramparts, a distance in a direct line of forty-two miles. Opposite the Old Crow Mountains and at another point farther up the river they are interrupted by uplifts of the under- lying Cretaceous shales. Like the Great Bear River beds they rest unoon- formably on the rocks beneath. They consist essentially of light-coloured sands, sandy clays, clays and conglomerates, with occasional nodular beds of ironstone, and in one section hold a small lignite sean:. No fossils of any kind were obtained from them. They are horizontal, or nearly so, and have a minimum thickness of 300 feet. Similar beds were noticed in the interval between the Upper and Lower Ramparts, and also occur for a short distance along the river below the Lower Ranjparts. The Tertiary beds occupy basins in the older rock, and evidence an elevation of the land at the end of the Cretaceous period followed by prolonged denudation, and a subsequent depression, accompanied by the formation of the lakes in which they were laid down. ******* Superficial Deposits and Glacial Action. Terraces. In crossing the mountains by the Peel River Portage few glacial features were noticed. A couple of terraces, the higher of which has 1^ ^. m rV 'h^t'' ; ». # k^:' K'^^l'i^,