■m IMAGE EVALUATION TEST TARGET (MT-3) 1.0 I.I 1.25 £^ 1^ 12.0 2.5 2.2 1.4 1.6 v. V] >> :^^v ^^. V '/ CIHM/ICMH Microfiche Series. CIHM/ICMH Collection de microtlches. Canadian Institute for Historical Microreproductions Institut Canadian de microreproductions historiques 1980 Technical Notes / Notes techniques The institute has attempted to obtain the best original copy available for filming. Physical features of this copy which may alter any of the images in the reproduction are checked below. D D D n Coloured covers/ Couvertures de couleur Coloured maps/ Cartes gdographiques en couleur Pages discoloured, stained or foxed/ Pages ddcolordes- tachet^es ou piqu6es Tight binding (may cause shadows or distortion along interior margin)/ Reliure serrd (peut causer de I'ombre ou de la distortion le long de la marge intdrieure) L'Institut a microfilm^ le meilleur exemplaire qu'il lui a 6t6 possible de se procurer. Certains d6fauts susceptibles de nuire d la qualitd de la reproduction sont notds ci-dessous. D D D D Coloured pages/ Pages de couleur Coloured plates/ Planches en couleur Show through/ Transparence Pages damaged/ Pages endommag6es Th po of fill Th CO or ap Th fill in: in up bo fol D Additional comment/ Commentaires &uppl6mentaires Bibliographic Notes / Notes bibliographiques D Only edition available/ Seule Edition disponible Bound with other material/ Relid avec d'autres documents D Pagination incorrect/ Erreurs de pagination Pages missing/ Des pages manquent D Cover title missing/ Le titre de couverture manque D Maps missing/ Des cartes gdographiques manquent D Plates missing/ Des planches manquent D Additional comments/ Commentaires suppidmentaires IS la The images appearing here are the best quality possible considering the condition and legibility of the original copy and in keeping with the filming contract specifications. The last recorded frame on each microfiche shall contain the symbol ^^> (meaning CONTINUED"), or the symbol V (meaning "END"), whichever applies. Les images suivantes ont 6t6 reproduites avec le plus grand soin, compte tenu de la condition et de la nattetd de I'exemplaire filmd, et en conformity avec les conditions du contrat de filmage. Un des symboles suivants apparaftra sur la der- nidre image de cheque microfiche, selon le cas: le symbole — ► signifie "A SUIVRE", le symbole V signifie "FIN". The original copy was borrowed from, and filmed with, the kind consent of the following institution: Library of the Public Archives of Canada Maps or plates too large to be entirely included in one exposure are filmed beginning in the upper left hand corner, left to right and top to bottom, as many frames as required. The following diagrams illustrate the method: L'exemplaire film6 fut reproduit grdce d la g6n6ro8it6 de I'dtablissement prdteur suivant : La bibliothdque des Archives publiques du Canada Les cartes ou les planches trop grandes pour dtre reproduites en un seul clichd sont filmdes d partir de I'angle supdrieure gauche, de gauche d droite et de haut en bas, en prenant le nombre d'images n6cessaire. Le diagramme suivant illustre la mdthode : 1 2 3 1 2 3 4 5 6 F •Repri i I THE PHYSICAL GEOGRAPHY AND GEOLOGY OF CANADA, BY GEORGE M. DAWSON, C.M.G, F.R.S., Director of the Geological Survey of Canada. Reprinted from the Handbook of Canada, issued by the Publicat Committee of the Local Executive of the British Association. ion TORONTO : Rowsell & Hutchison, Printers. 1897. ' i t " } ^t-^ /4fi X ^ n \ « PHYSICAL GEOGRAPHY AND GEOLOGY. THE Dominion of Canada embraces the northern half of the continent of North America, with its adjacent islands, including those of the Arctic Ocean between the 141st meridian and Greenland, but exclu- sive of Alaska, in the extreme north-west, the island of Newfoundland, which still remains a separate Brit- ish colony, and the small islands of St. Pierre and Miquelon, retained by France. In order, therefore, to include the sister colony of Newfoundland, the general name, British North America must still be employed. The total area of Canada is estimated at about 3,574,- 980 square miles, of which the Arctic islands to the north of the continental land make about 309,000 square miles. Thus, with Newfoundland (42,000 square miles), the aggregate area of British North America is 3,616,980 square miles. This area is some- what larger than the United States (including Alaska), and not much less than all Europe. The form of the North American continent may be described as that of an isosceles triangle, of which the narrower part, pointing to the south, constitutes Mexico, a wide central belt, the United States, while the broader base is the Dominion of Canada. The northern base-line of the continental land lies approxi- mately on the seventieth parallel of north latitude, It! 1 - I it 4 PHYSICAL GEOGRAPHY AND GEOLOGY. but is broken into by the geographically important sea named Hudson Bay, 800 miles from north to south, and some GOO miles in width. Historically, this great indentation of the northern land has been notable in connection with the exploration and trade of British North America, and it promises in the future, when modern means of communication have been adapted to the circumstances, again to become important. But the ruling physical features upon which the existence of Canada as a country depends, -and about which its liistor}' has grown up, are, the proximity of the north- eastern part of the continent to Europe, and the exist- ence of a great waterway, the River St. Lawrence, running to the very centre of the continent, and ex- panding there into the group of inland seas generally spoken of as the Great Lakes. The first of these made the landfall of Cabot in 1497 possible, the second led Cartier (in 1535) to the sites now occupied by the cities of Quebec and Montreal, and opened a route of explor- ation and commerce to subsequent explorers and traders of France, by which they overran much of the central and western country of North America before the col- onists of New England, further to the south, achieved a way across the Appalachian highlands which there barred their progress. The course of the St. Lawrence and the position of the Great Lakes may be regarded as being deter- mined b}'' the southern outline of the Laurentian plateau, composed of ancient crystalline rocks, which in U-shaped form surrounds the central depression of Hud- son Bay. Spreading widely in the Labrador peninsula, this tract of relatively high and generally rocky land, runs with narrower dimensions round the southern extremity of the bay, and thence is contirjued north- westward to the Arctic Ocean. With one important exception, that of the Ontario peninsula, which juts far to the south, the great river and its reservoirs lie G. M. DAWSON. 5 along the southern edge of these highlands, while the Winnipeg system of lakes, Athabasca, Great Slave and Great Bear Lakes, occupy a very similar position on the outer rim of the north-western extension of the plateau. Suess has named this plateau, composed of Archcrt of New York, all with very similar physical characters. In the opposite direction it is interrupted by the Gulf of St. Lawrence, but reappears in the great island of New- foundland, still preserving most of its characteristic features, although somewhat modified in appearances by differing climatic conditions. Throughout this reorion, includinor Newfoundland, the ofeoloo-ical struc- ture is alike, the formations represented are nearly the same, and, both in composition and from a palseonto- logical standpoint, they often resemble those of the opposite side of the Atlantic more closely than they do those of other parts of North America. Before speaking of the geological features of the Acadian region in Canada, two exceptional areas within its limits may be referred to. In the semi- circular bay formed by the coasts of New Brunswick and Nova Scotia, separated from the mainland by Northumberland Strait, is Prince Edward Island, a province by itself, although not much more than 2,000 square miles in area. This lies opposite the Carboni- ferous and Permo-Carboniferous lowlands of eastern New Brunswick and northern Nova Scotia, and con- sists entirely of undisturbed and unaltered Permo- Carboniferous and Triassic red sandstones and shales. Characteristic fossils of both formations have been found, but the circumstances render it difficult to draw a line between them. The surface of the island is for the most part fertile and highly cultivated, nowhere exceeding 500 feet above the level of the sea. The second area of an exceptional character, is that of the island of Anticosti, lying in the wide estuary of the St. Lawrence, and 140 miles in length. This again consists of nearly fiat-lying rocks, chiefly of the iti iii 12 PHYSICAL GEOGRAPHY AND GEOLOGY. Silurian with some of Oambro-Silurian or Ordoviciar> age (Hudson River) alone: its northern side. The island evidently represents part of a submerged and undisturbed Cambro-Silurian and Silurian tract of the northern part of the Gulf of St. Lawrence, the rocks of which differ in some respects from their representatives further to the west. The island is generally wo<;ded, and is at present very scantily inhabited. The geological scale is well represented in Nova Scotia and New Brunswick, from the Archtean to the Triassic, but thereafter ensues a long gap, during which no deposits appear to have been formed, pro- bably because the area in question then existed as land, exposed to denuding agencies alone. Closing this unrepresented lapse of time we find only the clayis and sands referable to the glacial period, w^ith still more recent deposits, such as those of the fertile marsh lands of the Bay of Fundy. With the exception of the flat-lying tracts between the several axes of elevation the wide Carboniferous area in New Brunswick and marginal developments of rocks of the same age along the northern coast of Nova Scotia and in Cape Breton, the region must be considered as one of exceptional geological disturbance and complexity, which, notwithstanding the large amount of investigation it has received, is still but imperfectly understood. One cause of difficulty lies in the existence, at several horizons, of thick masses of strata composed of ancient volcanic materials, generally without organic remains. This is a character common to the rock-formations of must parts of the Appala- chian region of North America, which has been at many times the theatre of great volcanic activity. The occurrence of similar rocks in the south-eastern part of the Province of Quebec has been the cause of much of the uncertainty attaching to the understanding of the " Quebec group " there. The region is not a typical one for the Archtean G. M. DAWSON. 13 rocks, but areas of crystalline schists referable to this time occur. Most of these have so far been mapped simply as pre-Cambrian, for no separation into groups has vet been effected. A large tract of the kind occu- pies the northern part of Cape Breton. In the southern highlands of New Brunswick, particularly in the vicin- ity of the city of St. John, the Archaean is better char- acterized and bears a resemblance, almost amounting to identity, with the typical developments of these rocks in Quebec and Ontario. A Lower Laurentian series comparable with the " Fundamental gneiss " is found, with an upper series composed of crystalline limestones, quartzites, etc., like the Grenville series. Newer than these is a mass of strata composed chiefly of volcanic materials, breccias or agglomerates, greenstones and fel sites, which is referred to the Huronian system. Further areas of the same kind occur in central New Brunswick. Forming the backbone of the peninsula of Nova Scotia and bordering its whole Atlantic coast, is a belt characterized by granitic masses and old stratified rocks assigned to the Lower Cambrian. This belt is widest in the south-west part of the province and narrows gradually in the opposite direction. The granites date from about the Devonian period, and it was probably in connection with their intrusion that the bedded rocks were thrown into the great series of parallel, sharp flexures, in which they pow lie and which bear so intimate a relation to the systems of gold- bearing veins. The disturbances incident to this period are, in fact, those which have chiefly given form to the Maritime Provinces. All the older rocks are involved in them, while those of the Carboniferous remain comparatively unaffected. These Cambrian rocks consist of a lower quartzite series and an upper clay-slate or argillite series, the latter generally dark in colour. It must be added that no really characteristic fossils have yet been obtained 14 PHYSICAL GEOGRA.PHY AND GEOLOGY. if 1 from these gold-bearing rocks. Distinctive Cambrian faunas Have been found only in a few places in Cape Breton. In the vicinity of St. John, New Brunswick, a re- markably interesting and complete section of Cambrian rocks, well characterized by fossils, occurs. This may now be regarded as typical for the eastern part of Canada and Mr. G. F. Matthew, to whom its elabora- tion is chiefly due, remarks particularly the resemb- lance of the fauna to that found in rocks of the same age around the Baltic sea, rather than to that of the interior of North America. In time, the series runs from the Etcheminian (older than the Olenellus zone) to the highest Cambrian. Rocks of Cambro-Silurian (Ordovician) age have also been determined in the locality just referred to, and sparingly in other parts of New Brunswick. In Nova Scotia, the only fossils of the kind come from a single place in Cape Breton, but considerable areas supposed to be of this age, chiefly composed of volcanic materials, occur in other parts of the province. Silurian (Upper Silurian) rocks, a'-e widely spread in northern New Brunswick, and in adjacent portions of Quebec, occupying the greater part of the area which drains to the Baie des Chaleurs. They recur in the southern part of New Brunswick and in the northern part of Nova Scotia, and although comprising lime- stones, sandstones and shales, are often greatly inter- mixed with contemporaneous volcanic materials, indi- cating, it will be observed, a third important volcanic interlude in the history of this part of the continent. In Nova Scotia, important bedded iron ores (haamatite) appear in this series. The geological horizons repre- sented, as compared with those of the Ncav York scale, range from the Clinton to the Lower Helderberg. In the rocks of Devonian age, occurring in that por- tion of the Appalachian region covered by Nova Scotia, New Brunswick, the Gasp^ peninsula of Quebec, and CJ. M. DAWSON. 15. a part of the adjacent State of Maine, a remarkably full representation of the flora of that ancient period is found, from which more than 125 species of land plants have been catalogued by Sir J. Wm. Dawson ; while at Scaumenac, on the Bale des Chaleurs, rocks of this age have yielded many interesting tish remains, investigated by Mr. J. F. Whiteaves, which com- pare closel}^ with those of the Old Red Sandstone. Further west, in beds of the same period in Ontario and New York, the fossils are chiefly marine molluscs, with comparatively little evidence of the existence of adjacent land. The lower part of the Devonian of the Maritime Provinces holds a similar fauna, and contains in Nova Scotia bedded fossil iferous iron ores. The geo- graphical extent of the Devonian rocks in Nova Scotia andNew Brunswick is comparatively limited, but in the Gasp^ peninsula of Quebec it becomes somewhat im- portant. The Carboniferous system, both from its extent and because of its economic value, must be considered as one of the most important features of Nova Scotia and New Brunswick, although there is reason to believe that much larger tracts of this formation still lie beneath the waters of the Gulf of St. Lawrence and the Atlantic. Its total thickness is, in some parts of Nova Scotia, estimated at 1G,000 feet, but it is very irregular in this respect and over the greater part of New Brunswick is comparatively thin. At the Joggins, on the north arm of the Bay of Fundy, is a remarkable continuous section show^ing 14,570 feet of strata, including more than seventy seams of coal, which has been made the subject of investigation by Logan, Lyell, and Sir J. Wm. Dawson. From beds in this section numerous specimens of a land-inhabiting reptilian fauna have been described. The flora of the period is well represented in many places, particularly in Nova Scotia, and includes that of several distinct stages, beginning with the Horton group at the base MWftfe>tea>-Vi»ii«»r 12 b GEOLOGICAL SURVEY OF CANADA. "ff Sip-ki-aw Rapid. tinuation of that of the Lakelso. A small Indian village is situated at the mouth of the river. The water between the Lakelso and the Kitsum- galum we found more rapid and difficult than any we had previously met with on the lower part of the Skeena, and it is at about two miles Limit of steam- *^^'"^'*^ ^'i® mouth of the Kitsumgalum that the stern-wheel steamer boat navigation ]yj^^f.(,j.j j-eached her furthest point in 1866. The Mumford was engaged in carrying supplies for the Western Union telegraph com- pany, and the point reached by her probably marks the limit of profit- able steam navigation of the Skeena. It is indeed difficult to under- stand how some of the rapid reaches below this place were ascended by the Mumford. About five miles above the mouth of the Kitsumgalum is the Sip-ki-aw fall or rapid. The course of the river is here interrupted by a mass of granite, forming a low I'ocky projection on the left side, which is submerged when the river is in flood. On the opposite bank the rock rises abruptly as a steep hill 600 to 700 feet high, which is continued further up stream b^' a range similar in elevation. The width of the river at the rapid is 400 to 500 feet only, and at the time of our visit, the water having declined considerably from its highest stage, a half submerged mass of rock was also visible in mid stream. The fall at this stage of water is about two feet only, but a body of water ho great, making this descent at a single bound, forms a pretty wild rapid. Canoes are generally unloiuled here and tracked up light. Ours were portaged across the i-ocky neck on the left bank, the stage of the water rendering that the easiest method. A steamer could only bo got up the Sip-ki-aw Rapid by warping. From the head of the tide to Sip-ki-aw Eapid the river is characterized by a great number of islands of all shapes and sizes. Above this point there are few. The islands are generally found to be composed of coarse gravel, with several feet in thickness of fine sandy soil or silt capping it. They seldom rise nwre than ten or twelve feet above the water level, and are evidently in most instances detached portions of the flat occupying the botfx)m of the valley through which the river flows. The coarse gravel has been deposited in the bottom of the stream, like the bara which now encumber the river, the finer overlying deposits are the result of the overflow on them of the flood waters when the river had cut down to a somewhat lower level. In ascending the river in a canoe the ' sloughs, ' or narrow channels behind the islands, are followed as much as possible to escape the strength of the current, thoUj^ nch ' sloughs ' are usually found to load at the head of the island ^o a rapid and drift-pile. ZymoetE Valley About four miles above Sip-ki-aw the Zymoetz River from the south- east joins the Skeena. This stream, which is of considerable size, was Inlands and channels. G. M. DAWSON. 17 which produced the Appalachian range. Were this all, a careful study o£ the beds on the two sides of the line would readily show r.heir identity ; but, it appears, that previous to the great epoch of disturbance the original physical conditions themselves differed. To the west, a sheltered sea came into existence about the close of the Cambrian period, in which Cambro- Silurian strata, in large part limestones, were laid down. To the east, sedimentation befj^an much earlier, and the circumstances of deposition were dif- ferent and more varied. Even the animal life present in the two districts was largely dissimilar at the same period, probably as a result of different temperatures in the sea water. Thus it was not until much study and thought had been given to the problem that Logan was enabled to afSrm the equivalency of a great part of the strata on the two sides of the St. Lawrence and Champlain fault. To those on the east, differing in composition and fauna from the locks of the typical New York section, he applied the name '* Quebec group." Regarded as a local name attached to the Atlantic type of the lower members of the Ordovician, and distinguishing these from rocks of the same date deposited upon the Continental plateau, this term may still be employed with advantage, expressing as it does a most important fact. Subsequent investigations have shown, however, that in the ridging up of this part of the Appalachian region, not only are some very old Cambrian rocks brought to the suiface, but considerable areas of crj'stalliiie schists which are evidently pre-Cambrian, and very possibly referable to the Huronian. These were originally included in the " Quebec Group," under mistaken ideas of meta- morphism, but their elimination from it, now rendered possible, does not detract from the merit of the original discovery, nor does it impair the value of the term *' Quebec group," as a descriptive one, if properly limited. i; 18 PHYSICAL GEOGRAPHY AND GEOLOGY. In this folded and disturbed region of south-eastern Quebec, copper ores, asbestos and chromic iron are among the more important minerals of value. Silurian rocks and some of Devonian age rest in places uncon- formably upon the older corrugations. Some facts i-especting the glacial deposits of the Appalachian region are given on a later page, with general statements relating to this period in eastern Canada. Lowland R of the St. Lawrence Valley. The tract of country which it is found convenient to include under this name, comprises but a small part of the hydro- graphic basin of the great river, which in all is about 530,000 square miles in extent. Nor is it altogether uninterrupted, although clearly enough defined in a general way by the edge of the Laurentian plateau on the north, the Appalachian highlands to the south- east, and on the south, further west, by the line of the St. Lawrence River and the lower members of the sys- tem of the Great Lakes. It may be described as extend- ing from a short distance below the city of Quebec to Lake Huron, with a length of over 600 miles and an area of more than 35,000 square miles, all of which • may be regarded as fertile arable land — the greatest connected spread of such land in eastern Canada. These lowlands are based upon nearly horizontal strata, ranging in age from the latest Cambrian (Pots- dam sandstone) to the Devonian. On a geological map its limits are readily observable ; but in order to un- derstand its character, it is necessary to consider it somewhat more clearly, and under such scrutiny it is found to break up naturally into three parts. The first of these is divided between the Provinces of Que- bec and Ontario, running west along the St. Lawrence and its great tributary the Ottawa somewhat beyond the 76th meridian, or to a north-and-south line drawn about twenty-five miles west of the city of Ottawa. It is here interrupted by a projecting, but not bold, I G. M. DAWSON. 19 spur of the Laurentian plateau, which crosses the St. Lawrence at the lower end of Lake Ontario, forming there the Thousand Ishmds, and runs southward to join the large Archaean tract of the Adirondacks in the State of New York. This Eastern division, with an area of 11,400 square miles, constitutes what may be called the St. Lawrence plain proper, parts of w^hich were among the first of those occupied by the early French settlers. Much of its surface is almost absol- utely level, and it nowhere exceeds a few hundred feet in elevation above the sea, although a few bold trappean hills stand out in an irregular line, with heights of 500 to 1,800 feet. Mount Royal, at Mon- treal, is one of these, and from it all the others are in sight, while the Laurentian highlands may also be seen thirty miles to the north, and to the southward the Green Mountains and Adirondacks, forming the bound- ary of the plain in that direction, are apparent on a clear day. Beyond the projecting spur of ancient crystalline rocks above referred to, from the lower end of Lake Ontario, near Kingston, to Georgian Bay of Lake Hur- on, the southern edge of the Laurentian plateau runs nearly due west, with a slightly sinuous line, for 200 miles. Between this edge and Lake Ontario on the south, lies a second great tract of plain, the lowest parts of which may be considered as level with Lake Ontario (247 feet) but of which no part exceeds 1,000 feet above the sea. This plain is naturally bounded to the south and west by the rather bold escarpment of the Niagara limestone, which, after giving rise to the Falls of Niagara between Lakes Ontario and Erie, runs across this part of the Province of Ontario to Lake Huron, forming there a long projecting point and con- tinuing still further west, in the chain of the Manitou- lin Islands. The city of Hamilton lies close under a part of this escarpment. The area of this second tract of plain is about 9,700 square miles. It is scarcely * 111 20 PHYSICAL GEOGRAPHY AND GEOLOGY. more varied in its surface than that to the eastward, and most of its extent is a fertile farming country. The third and last subdivision of the lowlands of the St. Lawrence valley, is an area of triangular form included between the Niagara escarpment and Lakes Erie and Huron. This constitutes what is generally known as the Ontario peninsula, and its south-west- ern extremity touches the 42nd parallel, the latitude of Rome. The area of the Ontario peninsula is 14,200 square miles, and both in soil and climate it is singu- larly favored. Grapes, peaches and Indian corn, or maize, are staple crops in many districts. To the north, some parts of this tract are high and bold, but most of its surface varies from 500 to 1,000 feet above the sea. The geological features of the lowlands of the St. Lawrence valley are comparatively simple. The rocks flooring the region lie either horizontally or at very low angles of inclination upon the spreading base of the Archaean mass to the northward, the crystalline rocks of which have frequently been met with in deep bor- ings. The formations represented correspond closely with those of the New York section, which had been rendered typical by the researches of James Hall and his colleagues, before any definite examination of the geology of Canada was begun. Hall's nomen- clature has been adopted for the series, which, begin- ning with the Potsdam sandstones, continues upward without any marked break, to the Chemung or Later Devonian. In the first or eastern subdivision of this region, the Potsdam sandstone, although strictly speaking refer- able to the Upper Cambrian, physically considered is really the basal arenaceous and conglomeritic member of the Cambro-Silurian (Ordovician) series which fol- lows. The several members of the Cambro-Silurian occupy almost the entire surface, diversified merely by a few light structural undulations, which to the soutk G. M. DAWSON. 21 « V and east of Lake St. Peter result in the introduction ot* some higher beds that are referred, although with some doubt, to the Silurian. Fossils referable to this period are also found associated with a small patch of volcanic bieccia on St. Helen's Island, o|)posite Mon- treal. This is probably connected with the adjacent igneous mass of Mount Royal, and by it the age of this mass, as well as that of the other similar promin- ent ])oints of the same character in the vicinity, is pretty definitely fixed as latest Silurian or Lower ]3evonian. It would in fact appear that these igneous masses represent the necks of ancient volcanic erup- tions, which have since been subjected to prolonged denudation. In their vicinity the sedimentary rocks are shattered and disturbed and traversed in all direc- tions by trap dykes. Basic rocks (theralite, etc.,) to- gether with nepheline-syenites are characteristic. Passing to the second, or central subdivision, to the west of Kingston, the Cambro-Silurian formations just referred to are found to be repeated, in ascending or- der, along the north shore of Lake Ontario, with very similar characters and equally undisturbed. The Trenton limestone occupies the greatest area, extend- ing in a wide belt to Georgian Bay of Lake Huron. Above this lie the Utica shales, and over these the Hudson River formation, upon which Toronto is situ- ated. This is the highest member of the Cambro-SiL iirian, but the plain also overlaps the lower members of the succeeding Silurian system irregularly, finding its natural boundary, from a physical point of view, only at the massive outcrop of the Niagara limestone. The course of the escarpment produced by this out- crop has already been traced ; above it and to the south-west, lies the higher plain generally known as the Ontario peninsula, constituting the third sub- division of the St. Lawrence lowlands. More than half of the area of this peninsula is occupied by De- vonian rocks, which succeed the Silurian regularly in 82 PHYSICAL GEOGRAPHY AND GEOLOGY. ascending order, the highest beds being met with in the extreme south-west of Ontario, beyond which, they are soon followed by the Carboniferous basin of tlie Michigan peninsula. The Silurian and Devonian strata are affected only by slight and low undulations, but these are important in connection with the ex- ploitation of the oil and gas of the region. In the two eastern subdivisions of the lowlands of the St. Lawrence valley, with the exception of struct- ural materials, such as stone, lime and clay, minerals of economic value are scarcelv found ; but in the third or westernmost subdivision, in addition to these, gyp- sum, salt, petroleum and natural gas have become im- portant products. The gypsum and salt are derived from the Onondajxa formation of the Silurian. The salt is obtained in the form of brine, from deep wells,, but beds of rock-salt are known to occur at consider- able depths. At Goderich, two beds of very pure salt have been proved by boring to exist at a depth slight- ly exceeding 1,000 feet, with an aggregate thickness of over 60 feet. Petroleum is chiefly derived from the Corniferous limestone of the Devonian, and natural gas is obtained from several horizons both in the Devonian and Silurian. The Laurentian Plateau. The great region thua named, composed of very ancient crystalline rocks, has an area of over 2,000,000 square miles, cr more than one-half that of the entire Dominion of Canada. In a horse-shoe-like form, open to the north, it pnrrounds three sides of the comparatively shallow sea known as Hudson Bay. Its southern part is divided between the Provinces of Quebec and Ontario, its eastern side expanding into the Labrador peninsula, while the western runs, with narrower dimensions, to the Arctic Sea, west of the great bay. In geographical extent it is thus very important, although somewhat monotonous in its physical and geolog-ical features. It contributes little to the fertile 1 • tS-u O. M. DAWSON. 23 . areas of the country in proportion to its size, but in the aggregate, comprises a considerable amount of land which is either cultivated or susceptible of cultivation. Elsewhere, in its southern parts, it carries forests of great value, and its mineral resources are already known in some places to be very important. It constitutes, moreover, a gathering ground for many large and almost innumerable small rivers and streams, which, in the sources of pt/wer they offer in their descent to the lower adjacent levels, are likely to ])rove, in the near future, of greater and more permanent value to the industries of the country than an extensive coal field. Particularly notable from this point of view is the long series of available water powers which runs from the Stiait of Belle Isle nearly to the head of Lake Superior, coincident with the southern border of the plateau. Although it is appropriate.to describe this region as a plateau or table-land, such terms, it must be under- stood, are applicable only in a very general way. Its average elevation of about 1,500 feet is notably greater than that of the adjacent lands, and is maintained with considerable regularity, but its surface is nearly every- where hummocky or undulating. Away from its borders, the streams draining it are, as a rule, extremely irregular and tortuous, flowing from lake to lake in almost every direction ; but assuming more direct and rapid courses in deeply cut valleys as they eventually leave it. Many of the surface features are of very great antiquity, Mr. Low's observations in Labrador going to show that the larger valleys there existed much in their present form before the Cambrian period. The average height of the central parts of the Labrador peninsula is about 1,700 feet, and the most of its drainage is divided between Hudson Bay, Ungava Bay and the Atlantic coast, the main watershed lying not very far to the north of the St. Lawrence estuary and gulf. Along the Atlantic coast, to the north of 24 PHYSICAL GEOGRAPHY AND GEOLOGY. Hamilton Inlet, the region assumes a really mountain- ous character, numerous elevations attaining 8,000 feet, and some as much as 5,000 or 6,000 feet. These are tl;e highest known points connected with any part of the Laurentian region, and are quite exceptional in character. The same hioh range is continued to the north of Hudson Strait in the western part of Baffin Land. 'J'o the south of Hudson Bay, the watershed is, at least in one part, as low as 1,000 feet. East of Lake Winnipeg, the Nelson and Churchill Rivers cross the Laurentian plateau, in a wide depression, to reach Hudson Bay. Still further north, this part of the plateau has a height of ahout 1,100 feet ahove the sea. Generally speaking, the surface of the plateau is harren and rocky, though often with wide swampy tracts towards the height-of-land. 'J'o the south-west of Hudson Bay it is overlapped by an important spread of Silurian and Devonian rocks, over whicli, and the adjacent parts of the crystalline rocks, is rather uniform- ly spread a mantle of alluvial deposits, affording a good soil, which in some places may eventually be of value, although the climate is doubtfully favourable to the growth of ordinary crops. The stiiking features of the Laurentian plateau are immeasurable lakes, large and small, with intervening rounded rocky elevations, wooded, in their natural conditions to the south, rising above the tree line to the northward ; w4iile in the far north, on both sides of Hudson Bay, hills and valleys become eventually characterized by grasses, mosses and lichens alone, con- stituting the great " barren lands " of North America. The rivers and lakes aie everywhere well stocked with fish, while deer and moose in the southern parts, and to the north the caribou, abound wherever the Jndian hunters have not followed them too closely. Thus, where the region can be entered without undue diffi- i i ; G. M. DAWSON. 25 > -culty, it has already become a much favoured resort of the sportsman. The name bv which it is found convenient to refer ft/ to this region, although derived from that of the Laurentian system of geologists, must not be supposed to impl}?" that the rocks attributed to this system occupy the whole area. There are, besides, several wide and many narrower bands of Huronian rocks, as well as outlying areas referred to the Lower Cambrian, and more such exceptional areas of both kinds doubt- less remain to be discovered. It must be understood, too, that in employing the name Lauientian for the most widely represented member of this great Archaean plateau, this is done in a very general and inclusive manner; the term "basement complex" is adopted by some geologists in nearly the same sense. The results of late investigations, particularly those of Dr. F. D. Adams, seem to prove that in the Laurentian, as thus understood, there is a distinctly stratified series of limestones and other sedimentary rocks, all now highly altered and crystalline, as well as another series of moj'e massive gneissic rocks, of which the apparent bedding is really a foliation due to pressure, and which frequently pass into granites by imperceptible grada- tions. The first of these is known as the Grenville series, and the second as the Fundamental Gneiss ; but even wheie most closely studied it has been found impossible completely to separate the two series, except quite locally, and thus, for the purposes of a general geological map, both must of necessity be indicated by a single colour. The complication alluded to is further increased by the frequent occurrence of true eruptive masses of granite or syenite of much later date, as well as by that of numerous great areas of anorthosite, a rock of the gabbro family, composed principally of plagioclase felspar. The Huronian rocks are as a rule darker in colour •and more basic in composition than those of the : 'I i: ! I 26 PHYSICAL GEOGRAPHY AND GEOLOGY. I Laurentian. They are often still distinctly bedded^ although very frequently in the form of schists in which no true bedding is now distinguishable, and some of which have evidently resulted from the crushing of eruptive rocks. Over great areas the original material of the Huronian has clearly been in large part the result of volcanic eruptions of the time, but elsewhere, as on the north shore of Lake Huron, and in the Sudbury and Lake Temiscaming districts, it comprises thick masses of argillite, quartzite and quartz or jasper conglomerate. The rocks composing the Fundamental Gneiss often assume the physical relations of erup- tives, of later date to the Huronian, along the lines of contact ; but in how far this may really be the fact, and in how far this appearance may be attributed to a certain amount of re-fusion of previously existing basal rocks, has not been determined. The Cambrian outliers date from a time subsequent to the main era of folding and crushing of the sub- jacent rocks, in comparison with which they are little disturbed. In the Labrador peninsula, they charac- terize considerable areas, and they form a border, tilted against the edge of the Laurentian plateau on the east side of Hudson Bay. These areas appear to represent the Animikie series of the Lake Superior section ; well seen in similar relations in the vicinity of Port Arthur. These rocks are evidently separated by a vast period of unrepresented time from the Archsean below, and it is assumed that they may be attached with greatest probability to the base of the Cambrian, although no distinct fossils have been obtained from them. To the west of Hudson Bay and far to the north, are other outliers, largely composed of sand- stones, conglomerates and traps, which resemble the Keweenawan or Nipigon of Lake Superior, somewhat later in age than the last. These also are provisionally classed as Lower Cambrian. The Nipigon rocks are well shown along the line of the Canadian Pacific rail- way to the east of Port Arthur. G. M. DAWSON. 27 Minerals of economic value in the Laurentian are almost entirely confined to the Grenville series, which yields, in various parts of its extent, mica, apatite (phosphate), graphite, iron and marble. The Huronian rocks are those in connection with which numerous important gold-bearing veins and deposits are now being found and opened up in western Ontario, as well as the nickel and copper deposits of Sudbury, iron and other metallic ores. In the Cambrian rocks near Port Arthur are silver-bearing veins, some of which have been extensively worked, and iron ores, while in the Labrador peninsula vast quantities of iron ores have been discovei'ed in the same series. Glaclation of Eastern Canada. Something may be added here on the events of the glacial period as affecting the eastern part of Canada as a whole, although many points connected with this particular period, still remain uncertain and the subject of debate. Like the Scandinavian peninsula, the Laurentian plateau at one stage in the glacial period apparently became the seat of a great confluent ice-sheet, which, when at its maximum, flowed down from it in all directions in general conformity with its main slopes. Climatic conditions and relatively local physical features may have conspired to render the discharge of glacier-ice more important in some directions than in others, and it is even possible that at no single time was the whole extent of the plateau equally ice-clad ; but on the h3^pothesis above stated, it has been pro- posed by the writer to designate the ice-sheet of the Laurentian highlands, the Laurentide glacier. If several distinct centres of dispersion existed through- out, these may be named collectively the Laurentide glaciers. It is certain, at least, that such local centres must have giown up at the beginning of the period of cold, and that towards its close similar centres must have again remained as the last seats of ice action. Two of these have been defined by the existing marks : fr: tii 28 PHYSICAL GEOGRAPHY AND GEOLOGY. of rock striation and transport, one to the west of the northern part of Hudson Bay, named by Mr. Tyrreil the Keewatin glacier, the other, resulting from the observations of Mr. Low, to the east of the bay, the Labradorian glacier. During the whole of this period the Laurentian plateau was in the main an area of denudation. From it the surface material was carried awav in all direc- tions, even to the northward, for there is absolutely no evidence that any " polar ice-sheet " ever trenched upon the continent of North America. The generally bare ice-scored rocky surface of these highlands is evidence of this denudation, while the existence of broken, angular masses of unmoved local debris in the central pai't of Labrador and in the central area of the Keewatin glacier, shows that across these neutral gathering-grounds, no ice ever passed. As to the distance to which the solid glacier-ice came southward from the Laurentian plateau, tlie evidence is yet inconclusive, for it is at best a matter of difficulty at the present day to definitely separate true moraines from beaches upon which floating ice impinged. Neither is it certain at how many times or to what extent the glacial period was interrupted by relatively warm epochs, but it may be stated that the flora of at least one of these interglacial epochs, as represented in the vicinity of Toronto, is such as to indicate a climate fully as warm as that at present existing, during which it seems improbable that much, if any, glacier-ice could have persisted on the Laui'en- tian highlands. These problems cannot be discussed here, but it is certain that at or about the decline of the glacial period, the eastern part of Canada as a whole stood at a relatively low level. Without quoting in detail the heights to which the sea is know^n to have reached at this time in various places, it mny be stated that it invaded the St. Lawrence valley as far at least as Lake lit G. M. DAWSON. 20 Ontario. Water-formed terraces are found at eleva- tions of 230 ieet in the Gaspe peninsula, to a maximum of 895 feet in the vicinity of Richmond, midway between Quebec and Montreal. Near Ottawa, the highest recognized shoreline is at 70.5 feet. The plains lying below these are floored by deposits holding marine shells of sub-arctic type, and these have been found at Montreal to a height of 560 feet. It is uncertain whether or not at the time of greatest subsidence the water coverinsf the area of the Great Lakes stood at the same level as, and w^as in direct connection with the sea, but it is not improbable that what has been described as the " Iroquois beach," around these lakes, may actually form the continuation of a liigh marine shore-line. The question is compli- cated by the unecjual amount of the subsidence and re-elevation to which the land has been subjected, a question requiring nmch further investigation. As a result of the circumstances noted, the St. Lawrence plain, as far west as Ottawa and Kingston at least, is deeply covered by deposits due to the glacial period, including boulder-clay, Leda clay and an overlying Saxicava sand, the two last, often full of fossil shells of the period. In the coastal regions of the Maritime Provinces similar deposits occur, to which the same names have been exten feeti 2372 feat above the sea. G. M. DAWSON. 81 • r North Saskatchewan River, the plain becomes essenti- ally a region of forest, with only occasional prairie tracts, such as those of the Peace River valley. By chance rather than by intention, the boundary line on the 49th parallel, to the west of the Red River, nearly coincides with the low watershed which separates the arid drainaije-basin of the Missouri from that of the Saskatchewan and its tributaries, cutting off only 20,000 square miles from the Missouri slope. Another line, nearly coinciding with a second low transverse watershed, may be drawn on the 54th parallel. The watershed crosses this line several times, but in the main it mav be taken as dividincr the Saskatchewan system of rivers from those of the Mackenzie and the Churchill. The belt of country comprised between these latitude lines is 350 miles wide, with a total area of about 295,000 square miles. The whole interior plain slopes eastward or north- eastward, from the Rockv Mountains towards the foot of the Laurentian highlands, so that a line drawn from the base of the mountains near the 49th parallel to Lake Winnipeg, shows an average descent of over five feet to the mile, fully accounting for the generally rapid courses of the rivers of the region. There are, however, in the area to the south of the 54th parallel, two lines of escarpment or more abru])t slope, which serve to divide this part of the plain into three portions, and although such division is by no means definite, it may usefully be alluded to for purposes of description. The first or lowest prairie-level is that of the Red River valley, of which the northern part is occupied by the Winnipeg group of lakes, its average elevation being about 800 feet above the sea, although gradually rising to the southward, along the axis of the valley, till it reaches a height of 9 GO feet about 200 miles to the south of the International boundary. Its area in Canada is about 55,000 square miles, including the lakes, and to the south of Lake Winnipeg it comprises 32 PHYSICAL GEOGRAPHY AND GKOLOGY. some 7,000 square miles of prairie land, which to the eye is absolutely flat, although rising uniformly to the east and west of the river. This is tlie former bed of the glacial " Lake Agassiz," the sediments of which constitute the richest wheat lands of Manitoba. The escai'pmentbounding this plain on the west, begins at the south in what is known as "Pembina Mountain," and is continued northward in the Riding, Duck, Poreu- pine and Pasqua hills, which overlook Manitoba and Winnipegosis Lakes, constituting the main eastern out- crop of the Cretaceous rocks of the plains. From this escarpment, the second prairie-level extends west- ward to a second and nearly parallel marked rise, wdiich, in general, is known as the Missouri Coteau. The area of this plain is about 105,000 square miles, of which more than half is open prairie. Its average eleva- tion is about 1,600 feet, and its surface is more diversified by undulations and low hills and ridges than that of the last, while the river-valleys are often deeply cut as well as wide. The greater part of the surface is well adapted for agriculture, although in places the scarcity of trees constitutes a disadvantage. The character of the soil is also more varied than that of the lower plain. The third and highest plain, lying between the last and the base of the Rocky Mountains, may be stated to have an average height of 8,000 feet, with an area, between tlie parallels of latitude first referred to, of about 184,000 square miles, of which by far the greater part is almost absolutely devoid of foi-est, its wooded area beino- confined to its northern and north-western edges, near the North Saskatchewan River or its tribu- taries. The surface of this plain is still more irregular than that of the last, and it is evident that both before and after the glacial period the di nuding forces of rain and rivers have acted upon it longer and more energetic- ally. Table-lands like those of the Cypress Hills and Wood Mountain, must be regarded as outlying rem- • r .^ G. M. DAWSON. 33 Hants of an older plain of the Tertiary period, and the slopes and flanks of such outliers show that similar processes of waste are still in operation, addino; to the length and depth of the ravines and " coulees," by which the soft Cretaceous and Tertiary rocks are trenched. The deposits of the glacial period, with which even this high plain is thickly covered, have tended to modify the minor asperities resulting from previous denudation. The soil is generally good, and often excellent, but large tracts to the south and west are sub-arid in character, and suited rather for pastur- age than for agriculture. Along the base of the Rocky Mountains is a belt of *• foot-hills," forming a peculiar and picturesque region, of which the parallel ridges are due to the diflering hardness of the Cretaceous rocks, here thrown into wave-like folds, as though crushed against the resistant mass of the older strata of the mountains. Taken as a whole, the central plain of the continent in Canada may be regarded as a great shallow trough, of which, owing doubtless to post-Tertiar}" difi'erential uplift, the western part of the floor is now higher in actual elevation than its eastern Laurentian rim. But although thus remarkably simple and deflnite in its grand plan, there ai'e many irregularities in detail. The second prairie-level has, for instance, some eleva- tions on its surface as high as the edge of the third plain, both to the west and east of the valley of the Assiniboine River, which, again, is abnormally depi-essed. It is not possible here to do more than charac- terize its features in a general way. Ever since an early faleeozoic time, the area now occupied by the interior plain appears to have remained undisturbed, and to have been affected only by wide movements of subsidence or elevation, which, although doubtless unequal as between its diff'erent parts, have not materially affected the regularity of the strata laid down. Upon this portion of the continental platform, in 3 '■ u PHYSICAL GEOGRAPHY AND GEOLOGY. its eastern part, on Lake Winnipeg and its associated lakes, Cambro-Silurian, Silurian and Devonian rocks are found outcropping along the stable base of the Laurentian plateau. Following this line of outcrop northward, the Devonian rocks gradually overlap those of older date and rest directly upon the Archaean. They continue to the Arctic Ocean and there occupy a gi'eat part of the Northern Archipelago. To the south of Athabasca Lake they rest, without any apparent angular unconformity, upon sandstones refer- red to the Lower Cambrian, to which allusion has already been made, giving evidence, in the strati- graphical hiatus, of prolonged periods during Palaeo- zoic time in which land as well as water existed in some parts ol' the area. On the western side of the Great Plains the Palaeozoic strata reappear crumpled and broken in the Rocky Mountains, where the vast crustal movements of the Cordilieran belt found their inland limit. These rocks consist, for the most part, of ))ale-grey or buti", often magnesian limestones, along the eastern outciop, and from them Mr. Whiteaves has described an extensive and somewhat peculiar fauna. Some, at least, of the Palaeozoic formations represented, probably extend beneath the entire area of the Great Plains, but they are wholly concealed there by later strata of Cretaceous age, consisting chiefly of clay-shales and sandstones, generally but little indurated and flat-lying, or nearly so. The uniformity in the surface features of this country is principally due to that of these deposits, which, although since greatly denuded, have worn down very equally, and have apparently never been very long subjected to waste at a great height above the base-level of erosion. The whole area has in fact been one rather of deposition than of denudation up to a time geologically recent, and has very lately been levelled up still further by the super- ficial deposits due to the glacial period. .; • ' G. M. DAWSON. 35 • f The Cretaceous rocks are for the most part dis- tinctly marine, althougli beginning witli the Dakota sandstones, indicative at least of shallow seas, and from which, in the Western States, a considerable flora of bind plants has been recovered. These are followed, in the eastern part of the plains, by the Benton shales, the Niobrara, largely calcareous and foraminiferal in some places, the Pierre shales, and lastly by the Fox Hill sandstones. Further west, in Alberta, a part of the Pierre (with probably also the uppei* portion of the Niobrara), is represented by the Belly River form- ation, with a brackish water fauna, and containing beds of coal or lignite. The Dunvegan series of the Peace River, to the noith, similarly characterized, is perhaps somewhat older. The Cretaceous strata in fact change vei-y materially in composition and charac- ter toward the Rocky Mountains, aid when followed to the north, giving rise to the necessity for local names, and rendering a precise correlation ditticult in the absence of connecting sections over great tracts of level country. All the Cretaceous strata so far referred to belong to the later stages of that system, but in the foot-hills the earlier Cretaceous is represented by the Kootanie for- mation, holding coal, and reappearing as infolds in the eastern ranges of the Rocky Mountains. One of these is followed by the valley of the Bow River between Banff and Canmore, and affords both anthracite and bituminous coal. Overlying the Cretaceous rocks proper, in consider- able parts of their extent, particularly in Alberta, are those of the Laramie, which although perfectly confor- mable with the marine strata beneath, contain brackish water, and in their upper part entirely fresh water forms of molluscs, together with an extensive flora and numerous beds of lignite-coal or coal. As a whole, this formation may be regarded as a transition from the Cretaceous to the Tertiary, with a blending of 'i^ m 86 PHYSICAL GEOGRAPHY AND GEOLOGY. organic forms elsewhere considered as characteristic of one or the other. The lower parts are undoubtedly most nearly related to the Cretaceous and particularly to the Belly River beds, which were laid down under similar physical conditions at an earlier stage. The remains of Dinosau^lan reptiles are still abundant in these. The upper beds, constituting what was origin- ally named the Fort Union group, with its local representatives under different names, is on the con- trary more nearly allied to the Eocene. A still later stage in the Tertiary is represented by beds of Oligocene, or early Miocene age, found parti- cularly as an outlier capping the Cypress Hills. These have afforded numerous mammalian bones, described by Professor Cope and referred by him to the stage of the White River beds of the Western States. The aggregate thickness of the Cretaceous strata of the plains, so far as known, may in the eastern part be stated as about 2,000 feet; in the west, in northern Alberta, it is about the same, but exceeds 2,500 feet in south-western Alberta, without including the Kootanie series of some 7,000 feet or more. The thickness of the Laramie is also great toward the Rocky Mountains, reaching prooably 3,700 feet. The Pliocene (with perhaps the latter part of the Miocene) appears to have been a time of erosion only in the area of the Canadian plains ; wide, Hat-bottomed valleys were cut out in the foot-hills, and to the east of these great tracts of country between the now out- standing plateaux must have been reduced to the extent of 1,000 feet or more in height. Mineral fuels, in the form of coals and lignite-coals, constitute economically the most important products of the Cretaceous and Laramie rocks of the plains. To the south of the 56th parallel in Canada, an area of not less than 60,000 square miles is underlain by beds of such fuel. These consist entirely of lignite to the eastward, but on entering the foot-hills bituminous * » a » O. M. DAWSON. 87 coals are found, and similar coals recur in infolded areas in the Rocky Mountains, toi^ether with anthracite. Natural gas has been found in considerable quanti- ties in l)orings in several places, but is not yet utilized. Great outcrops of Cretaceous sandstones saturated with tar or maltha occur alonj]; the Athabasca River, pro- bably evidencing the existence of important petroleum deposits in the subjacent Devonian rocks. Salt springs appear on the l)orders of Manitoba Lake, and much further north in the Athabasca basin, but have been utilized to a very limited extent only. Gypsum also occurs in the Silurian and Devonian rocks along their eastern outcrop. Gold is washed from the sands of several of the larger livers, but this, it is supposed, is for the most part derived by natural concentration from the drift deposits. The Cordillera. Of this great mountainous region of the Pacific coast, a length of nearly 1,300 miles is included by the western part of Canada. Most of this is embraced in the Province of British Columbia, where it has a width of about 400 miles between the Great Plains and the Pacific Ocean. To the north, it is con- tinued in the Yukon district of the North-West Terri- tory, till it reaches, in a less elevated and more widely spreading form, the shftres of the Arctic Ocean on one side, and on the other passes across the 141st meridian of west longitude into Alaska. Its strongly marked features result from enormous crustal movements parallel to the edge of the Pacific, by which its strata have at several periods and along different lines, been crumpled, crushed and faulted. These movements having continued at intervals to times geologically recent, and the mountains produced by them still stand high and rugged, with streams flowing rapidly and with great erosive power dov/n steep gradients to the sea. Although preserving in the main, a general north- 38 PHYSICAL GEOGRAPHY AND GEOLOGY. north-westerly trend, the orographic features of this region are very complicated in detail. No existing map yet properly represents even the principal physi- cal outlines, and the impression gained by the traveller or explorer may well be one of confusion. Disregard- ing, however, all minor irregularities, two dominant mountain systems are discovered — the Rocky Moun- tains proper, on the east, and the Coast Range of British Columbia, on the west. The first of these, it has been proposed to name, from an orographic point of view, the '' Laramide Range," as it is essentially due to earth movements, occurring about the close of the Laramie period, and rocks of that age are included in its flexures. Although not quite continuous (for there are several echelon -like breaks, in which one mountain-ridge assumes the dominance previously possessed by another), this range, beginning two or three degrees of latitude to the south of the 49th parallel, forms the eastern member of the Cordillera ail the way to the Arctic Ocean, which it reaches not far to the west of the Mackenzie delta. It is chiefly composed of Palaeozoic rocks, largely limestones, and where it has been closely studied is found to be aflected by series of over thrust faults, parallel to its direction, of which the easternmost separates it from the area of the Cretaceous foot-hills. Here the older rocks have been thrust eastward for several miles over the much newer strata. The struc- ture has as yet been worked out in detail only along the line of the Bow River pass, by Mr. R. G. McCon- nell. In width, this range seldom exceeds sixty miles. The heights formerly attributed to some peaks appear to have been exaggerated, but many points in its southern part exceed 11,000 or 12,000 feet. The Coast Range of British Columbia constitutes the main western border of the Cordillera. Beginning near the estuary of the Fraser River, it runs uninter- ruptedly northward, with an average width of about * t» G. M. DAWSON. 8^ 100 miles, for at l(3asfc 900 miles, when it passes inland beyond the head of Lynn Canal. This range is largely composed of granite, with infolded masses of altered Paheozoic strata. It is not, as a rule, so rugged in outline as the last, but its western side, i-ising from the sea, shows the full value of its elevation there, while its main summits often exceed 8,000 or 9,000 feet. Several rivers rising in the plateau country to the east- ward, flow completely across this range to the Pacific, where the lower parts of their valle3's, as well as those of manj^ streams originating in the mountains them- selves, in a submerged state constitute the remarkable S3''stem of fiords of British Columbia. Even in the arrangement of the islands adjacent to the coast, the further extension of these valleys and of others run- ning with the range, may be traced, the evidence being of great subaerial erosion when the land previously stood at a higher stage. The cutting out of these deep valleys probably began in Eocene times, but was renewed and gieatl}' increased in the later Pliocene. Outside the Coast Range and in a i>artlv submeroed condition, lies another range, of which Vancouver Island and the Queen Charlotte Islands are projecting ridges. This stands on the edge of the Continental plateau with the great depths of the Pacific beyond it. The rocks resemble those of the Coast Range but include also masses of Triassic and Cretaceous strata which have participated in its folding, while horizontal Miocene and Pliocene beds skirt some parts of the shores. In the inland portion of British Columbia, between the Coast and Rocky Mountain systems above particu- larly alluded to, are numerous less important mountain ranges, which, while preserving a general parallelism in trend, are much less continuous. Thus, in travel- ling westward by the line of the Canadian Pacific Railway, after descending from the Rocky Mountain summit and crossing the Upper Columbia valley, the i 40 PHYSICAL GEOGRAPHY AND GEOLOGY. ; I I Selkirk Range has to be surmounted. Beyond this, the Columbia on its southward return is a^^ain crossed, and the Gold Range is traversed by the Eagle Pass before entering the Interior Plateau of British Columbia, which occupies the space remaining between this and the Coast Range. The system of ranges lying immedi- ately to the west of the Rocky Mountains proper, not- withstanding its breaks and irregularities, is capable of approximate definition and its components have been designated collectively the Gold Ranges. Further north, it is represented by the Cariboo Mountains, in the mininfj district of the same name. The hin-hest known summit of this system is Mount Sir Donald, 10,645 feet, one of the Selkirk Mountains. This mountain system is believed to be the oldest in British Columbia. It comprises Archaean rocks with granites and a great thickness of older Palaeozoic beds, much disturbed and altered. The Interior Plateau constitutes an important phys- ical feature. Near the International boundary it is terminated southward by a coalescence of rather irregular mountains, and again, to the northward, it ends about latitude 55° 30' in another plexus of moun- tains without wide intervals. Its breadth* between the margins of the Gold Ranges and the Coast Range is about 100 miles, and its length is about 500 miles. It is convenient to speak of the country thus defined as & plateau, because of its difference, in the large, from the more lofty bordering mountains. It comprises the area of an early Tertiary denudation-plain (or pene- plain) which has subsequently been greatly modified by volcanic accunuilations of the Miocene, and by river- erosion while it stood at a considerable altitude, in the Pliocene ; but its true character as a table-land is not obvious until some height has been gained above the lower valleys, where the eye can range along its level horizon-lines. It is highest to the southward, but most of the great valleys traversing it are less in elevation I G. M. DAWSON. 41 than 8,000 feet above the sea. To the north, and par- ticularlj'^ in the vicinity of the group of large bikes occurring there, its main area is less elevated than 3,000 feet, making its average height about 3,500 feet. Beyond this plateau to the north, the whole width jf the Cordillera, very imperfectly explored as yet, ap- pears to be mountainous as far as the 59th parallel of latitude, when the ranges diverge or decline, and in the upper basin of the Yukon, rolling or nearly flat land, at moderate elevations, again begins to occupy wide intervening tracts. As a whole, the area of the Cordillera in Canada may be described as forest-clad, but the growth of trees is more luxuriant on the western slopes of each of the dominant mountain ranges, in correspondence with the greater precipitation occurring on these slopes. This is particularly the case in the coast region and on the seaward side of the Coast Range, where magnificent and dense forests of coniferous trees occupy almost the whole available surface. The Interior Plateau, however, constitutes the southern part of a notably dry belt, and includes wide stretches of open grass- covered hills and valleys, forming excellent cattle ranges. Further north, along the same belt, similar open country appears intermittently, but the forest invades the greater part of the region. It is only toward the Arctic coast, in i-elatively very high lati- tudes, that the barren Arctic tundra country begins, which, sweeping in wider development to the west- ward, occupies most of the interior of Alaska, • With certain exceptions, the farming land of Brit- ish Columbia is confined to the valleys and tracts below 3,000 feet, by reason of the summer frosts occur- ring at greater heights. There is, however, a consider- able area of such land in the aggregate, with a soil generally of great fertility. In the southern valleys of the interior, irrigation is necessary for the growth of crops. 42 PHYSICAL GEOGRAPHY AND GEOLOGY. I ^< The geological structure of the Cordillera is extremely complicated, and it has as yet been studied in detail over limited tracts only. There have been no appro- priate terms of comparison for the formations met with, and these -it has consequently been necessary to investigate independently by the light of first principle. The difficulty is increased by the abund- ance of rocks of volcanic origin referable to several distinct periods, resembling those of the Appalachian mountain region, though on a vastly greater scale, and like them, almost entirely devoi