^> 
 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 V 
 ^ 
 
 
 A 
 
 /!^,^ .. '^s 
 
 ^^ 
 
 f/. 
 
 
 1.0 
 
 I.I 
 
 1.25 
 
 lAilli |2.5 
 
 uo 12.0 
 
 12.2 
 
 1.8 
 
 U nil 1.6 
 
 
 
 
 y 
 
 >^ 
 
 
CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHM/ICMH 
 Collection de 
 microfiches. 
 
 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. 
 
 L'Institut a microfilm^ le meilleur exemplaire 
 qu'il lui a 6x6 possible de se procurer. Certains 
 d6fauts susceptibles de nuire d la qualitd de la 
 reproduction sont not6s ci-dessous. 
 
 / 
 
 Coloured covers/ 
 Couvertures de couleur 
 
 D 
 
 Coloured pages/ 
 Pages de couleur 
 
 n 
 
 Coloured maps/ 
 
 Cartes g^ographiques en couleur 
 
 Pages discoloured, stained or foxed/ 
 Pages d6color6es, tachet6es ou piqu^es 
 
 D 
 
 D 
 
 Coloured plates/ 
 Planches en couleur 
 
 Show through/ 
 Transparence 
 
 D 
 
 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) 
 
 D 
 
 Pages damaged/ 
 Pages endommagdes 
 
 D 
 
 Additional comments/ 
 Commentaires suppldmentaires 
 
 Bibliographic Notes / Notes bibliographiques 
 
 D 
 
 Only edition available/ 
 Seule Edition disponible 
 
 D 
 
 Pagination incorrect/ 
 Erreurs de pagination 
 
 D 
 
 Bound with other material/ 
 Relid avec d'autres documents 
 
 D 
 
 Pages missing/ 
 Des pages manquent 
 
 n 
 
 Cover title missing/ 
 
 Le titre de couverture manque 
 
 n 
 
 Maps missing/ 
 
 Des cartes g^ographiques manquent 
 
 D 
 D 
 
 Plates missing/ 
 
 Des planches manquent 
 
 Additional comm*»-*i/ 
 Commentaires suppl6mentaires 
 
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. 
 
 Les images suivantes ont 4t6 reproduites avec le 
 plus grand soin, compte tenu de la condition et 
 de la nettet6 de I'exemplaire filmd, et en 
 conformity avec les conditions du contrat de 
 filmage. 
 
 The last recorded frame on each microfiche shall 
 contain the symbol -^> (meaning CONTINUED"), 
 or the symbol V (meaning "END"), whichever 
 applies. 
 
 Un des symboles suivants apparaitra sur la der- 
 nidre image de cheque microifiche, selon le cas: 
 le symbole — ► signifie "A SUIVRE", le symbole 
 V sisnifSe "FIN". 
 
 The original copy was borrowed from, and 
 filmed with, the kind consent of the following 
 institution: 
 
 National Library of Canada 
 
 L'exemplaire film6 fut reproduit grAce d la 
 g6n6rosit6 de I'dtablissement prdteur 
 suivant : 
 
 Bibliothdque nationale du Canada 
 
 Maps or plates too large to be entirely included 
 in one exposure are filmed beginning in the 
 upper Iflft hand corner, left to right and top to 
 bottom, as many frames as required. The 
 following diagrams illustrate the method: 
 
 Les cartes ou les planches trop grandes pour 6tre 
 reproduites en un seul cliche sont filmdes i 
 partir de Tangle sup6rieure gauche, de gauche i 
 droite et de haut en bas, en prenant le nombre 
 d'images n^cessaire. Le diagramme suivant 
 illustre la m^thode : 
 
 1 
 
 2 
 
 3 
 
 1 
 
 6 
 
THE 
 
 Geological History 
 
 OF 
 
 LAKE SUPERIOR 
 
 I 
 
 J BY 
 
 DR. ROBERT BELL, F.R.S. 
 
 (Read before the Canadian Institute, Toronto, Af^ril isth, rSgy.) 
 
 TORONTO : 
 
 MURRAY PRINTING COMPANY. GLOBE BUILDING. 
 
 1899. 
 
The Geological History of Lake Superior. 
 
 By Dk. Robert Bell, F.R.S. 
 
 Read before Canadian Institute^ Toronto, April i^th, i8gg. 
 
 In the talk which I am about to give on the Geological History 
 of Lake Superior I propose to endeavour to trace the origin and develop- 
 ment of the lake itself rather than to sketch the geology of the surround- 
 ing country. The genesis of some of the great lakes, or rather of the 
 depressions in the continent which they occupy, has been the subject of 
 some investigation and of much discussion and speculation among 
 geologists, but their researches and controversies have related mostly to 
 the lower lakes, while but little has yet been written as to the geological 
 history of Lake Superior. It was on this account, and because I have 
 worked for many years on and around this inland sea and have paid 
 some attention to this matter, that I have chosen it as the subject of my 
 address to-night. 
 
 Before proceeding to speak of the basin of Lake Superior, I shall say 
 a few words about the basins of the other great lakes, the origin of each 
 of which has been similar to the others, but in some respects different 
 from that of Lake Superior. Sir Andrew Ramsay and Sir William 
 Logan supposed them to be due principally to glacial action; that is to 
 say, that they had been scooped out of the rocks in which they lie, 
 mainly by the denuding force (jf heavy and e.xtensive masses of land 
 ice. Although the eroding or excavating power of thick glaciers is 
 very great, still it does not appear to have been sufficiently powerful to 
 account for all that was formerly attributed to it. It has been pretty 
 satisfactorily shown that during the glacial period their action in modify- 
 ing the surface features consisted largely in the transportation of prc- 
 viousl)' decomposed and loosened rock. 
 
 A former assistant of mine on the Geological Survey, Dr. J. W. Spencer, 
 was, I think, the first to point out that the depressions now occupied by 
 all the lakes east of Lake .Superior represent the wide valleys or hydro- 
 graphic basins of former rivers, more or less modified by subsequent 
 glacial action, together with slight and widespread undulations or warp- 
 ings of the crust of the earth. Although our lakes are .so extensive 
 
and present such a conspicuous figure on the maps, they are quite shal- 
 low relatively to their area, and as the surface of this part of the conti- 
 nent is so generally level, a very small amount of denudation and a very 
 slight undulation in the land would suffice to produce the shallow 
 depressions which they fill. 
 
 Before the advent of the glacial period a great part of North America 
 stood at an elevation of 3,000 feet or more above its present level, and 
 the disintegration and removal of the rocks due to rainfall alone, which 
 was facilitated by the greater elevation, went on for a vast length of 
 time. This is proved by the existence of long and deep river valleys, 
 some of them running down from the central parts of the continent to 
 the present sea coasts, and beyond them into the depths of the ocean. 
 The old valley of the St. Lawrence can be traced for 800 mile.s. The 
 ancient bed of the Hudson has been followed by .soundings, from its 
 pre.sent mouth, down the slope of the bottom of the ocean far (nit to sea. 
 The canon of the Sagucnay, and tho.se of the Grand or Hamilton and 
 other rivers of Labrador, as well as the channels of many of the long 
 straight lakes and river stretches in the Archiuan country to the north 
 and northwest of us, are due partly to atmospheric and aqueous erosion 
 during a long geological period while the land stood at higher levels. 
 It can be shown that the grade of the Mississippi from its source to the 
 sea was much steeper formerly than now, the present modification or 
 lowering of the slope amounting to some 3,000 feet. The valleys thus 
 excavated guided, to a considerable extent, the movements of the great 
 glaciers which ploughed the surface of hill and dale, excavating and 
 carrying forward vast quantities of the .softened surface rock, together 
 with harder portions in the shape of fragments, boulders and pebbles, 
 which together constitute our til! or hard-pan and other deposits, 
 collectively called drift. The general tendency of the glacial action 
 would be to enlarge the main valleys of the preglacial surface and to 
 fill up or modify the smaller ones. 
 
 The geological structure and the relative resisting powers of the rocks 
 were the [)rimary or fundamental causes which predetermined the loca- 
 tion, direction, extent, etc., of the valleys and basins thus formed by the 
 combined action of aqueous denudation and glacial action. 
 
 The theory of preglacial river-erosion as the main factor in originating 
 the lake basins may apply to all of them with certain dififerences in each 
 case. The soundings show the existence in the bottom of each, of deep 
 channels resembling river valleys on the land traversing their beds and 
 leading to former outlets now closed up by drift materials, but which 
 
formerly drained these lakes or connected thein witli one another. We 
 are indebted mainly to Professor J. W. Spencer for the discovery and 
 location of these connecting channels. One of them, starting in Lake 
 Michigan, runs eastward across the State of the same name, through the 
 basin of Lake iiurcjii to the northern extremity of the Indian i'eninsula, 
 thence down the bottom of Georgian Hay and across the neck of land 
 to Lake Ontario. Another of these buried channels runs northward 
 from Lake I'.rie to the western extremity of Lake Ontario. The basins 
 of all the lower lakes are excavated in the softer portions of the almost 
 horizontal pakeozoic strata, so that their shapes and directions conform 
 to some extent with the geological structure or arrangement of the strata. 
 Lake Superior lies in a hollow almost surrounded b)' the primitive 
 crystalline rocks. At one time it vas probablj' filled with newer strata, 
 which have been mostl)' removed by aqueous and glacial denudation ; 
 but small areas of those rocks still remain. 
 
 Lake Superior lies in the region of the general watershed or summit 
 level of the continent east of the Rocky Mountains, although it is only 
 600 feet above the sea. From near its shores the water flows west to 
 the Winnipeg basin, north to Hud.son's Bay, and south to the Mississippi, 
 while its own discharge is eastward to the St. Lawrence. 
 
 The drainage system, or catchment basin of Lake Superior proper, 
 is consequently small, but it has a sort of extension or apijendix in the 
 basin of Lake Nipigon, which is really the uppermost of the Great 
 Lakes of the St. Lawrence. This " f water measures eighty miles 
 
 from north to south by fifty miles Iroiii east to west, with extensive 
 bays on all sides. Its area is about half that of Lake Ontario, or 
 between 3,000 and 4,000 sfpiare miles. Thirteen rivers and many 
 brooks flow into it from vari<jus directions. One of its bays is only 
 twenty-four miles from the nearest point of Lake Superior, and its 
 surface has an elevation of 244 feet above the latter. The Nipigon 
 river, which is the upward continuation of the St. Lawrence, is a clear- • 
 water stream and altogether the largest one flowing inUy Lake Superior. 
 The larger number of the rivers and brooks falling into the great lake 
 are darkly colored. Some of them look almost black as they enter its 
 limpid water, the contrast being very striking. Hut the vegetable 
 matter, to which the colour is due, .soon becomes oxidized and disappears. 
 There is a general absence of mud and also of dis.solved mineral matter 
 in the tributaries of Lake Superior, and hence its waters are not only 
 singularly transparent, but also nearly chemically pure. The sounding 
 lead has shown that the bottom of the Lake in nearly all parts consists 
 of clay. 
 
In tracing the origin of the basin of Lake Superior, we must go back 
 to a very early period in the history of the soh'd earth. It appears to 
 have begun with a depression in the original crust, even before the 
 deposition of the oldest Huronian rocks, and probably before any water 
 at all rested upon the heated surface of the globe. The great basin of 
 Hudson's Bay is an example of a still larger depre.ssion in the first 
 formed crust, which has persisted to the present day. 
 
 The vast Archaean region of North America, embracing perhaps one- 
 third of its area, and which formed the nucleus of the continent around 
 which the land has grown by the addition of one formation outside of 
 another, has a general elliptical outline and extends from the north 
 of Greenland in the far northeast to the State of Mi.ssouri in the 
 southwest. 
 
 The depression of Lake Superior and that of Hudson's Bay and its 
 connecting waters to the northward are within this ellipse. The earliest 
 clastic deposits from the primeval sea might be looked for in these 
 hollows and in the corresponding levels around the primitive nucleus of 
 the continent, and hence, in these situations we find the Newer Laurentian 
 and the Huronian .systems largely developed and followed by the 
 fossiliferous formations while the higher grounds or tho.se more distant 
 frorti these depressions as well as from the general periphere, consist 
 almost entirely of the oldest gneiss which appear to be all modified 
 from a granitic magma. The Huronian series, made up to a great 
 extent of igneous rocks, are very largely developed lo the south and 
 west of Lake Superior in the States of Michigan, Wisconsin, Minnesota 
 and Dakota, as well as throughout the country lying to the northwestern, 
 northern and northeastern sides of the lake. During the succeeding 
 Lower Cambrian period, the region of Lake Superior was the site of 
 great outpourings of igneous matter, which formed the diorites, diabases, 
 melaphyres, gabbros, amygdaloids, etc., of the Animikic and the 
 Nipigon series. Between the Huronian and the older Cambrian 
 periods, a vast interval of time elapsed, of which but little record has 
 been left in the Lake Superior region. The upturned edges of the 
 highly disturbed Huronian rocks were denuded down to a nearly level 
 surface, and upon them were deposited the horizontal and undisturbed 
 beds of the Animikieand Nipigon or the older Cambrian system, which 
 have remained unaltered to the present day. The masses of granite of 
 greater or less extent, which cut the Huronian and sometimes the 
 Laurentian rocks around Lake Superior may have been erupted during 
 this long interval, as we do not find them intersecting the Cambrian or 
 any of the still newer strata, although they are themselves cut by dykes 
 
similar to those which traverse the Huroniaii. Geographically these 
 eruptive granites are scattered around the lake outside of the area of 
 theCambrian igneous overflows, or in other words they lie, in a general 
 way, between this area and the vast region of primitive gneiss beyond. 
 Some of these granitic masses are large and are elongated parallel to 
 the present coast lines, such as the one called the Giant's range, lying 
 back from the northwest shore in Northern Minnesota and the Thunder 
 Ha\' District, some of those near Xipigon Hay. Otter Head, and thence 
 eastward to Michipicoten River, and Pointe Hrule, from which a wide 
 mass runs inland. 
 
 The Nipigon or Keeweenian formation, ccjnsisting largely of the 
 igneous rocks alreath" mentioned, occupies the outermost large points 
 and islands all around the lake, such as those of Xipigon Bay, Keeweenaw 
 Point, Isle Royale, Michipicoten Island, Gargantua and Xamainse,* and 
 it is probable that the bed of the whole lake consists mostly of this 
 formation. The dip of both the Xipigon and the underlying Animikie 
 rocks on both sides is towards the centre or deeper parts of the lake. 
 The lake-basin existed before these rocks were laid down and it is 
 probable that their jjrcsent dip is partly due to a gentle inward slope of 
 the surface on which they were deposited whether as sediments or 
 volcanic sheets, but the higher dips which the\' have now assumed are 
 believed to be owing to the slow sinking or caving-in of the crust 
 following the removal, from beneath the basin, of such vast quantities 
 of igneous matter to form the sheets of greenstones in the Animikie and 
 the amj'gdaloids, etc., of the Xipigon formation, of which only fragments 
 remain at the present day. 
 
 Those portions of the shores of Lake Superior which are occupied b>- 
 the Laurentian, Huronian and older Cambrian rocks are noted for 
 being traversed by great numbers of dykes of various kinds of green- 
 stones. The larger fissures, marked b\- these dykes, may at one time 
 have allowed great quantities of molten matter to escape to form the 
 tra[)pean overflows of the Animikie and the am)gdaloids, etc., of the 
 Xipigon series. On the other hand, the.se rocks may have been derived 
 mainly from volcanic orifices now covered bj- the lake or represented 
 b\- the plutonic necks to be found among the Huronian strata. 
 
 In the rejjort on m)' survey of Lake Xipigon in 1869, I have shown 
 tliat the immense horizontal cappings of diorite forming the upper part 
 of the Xipig(in series around that sheet of water appear to have flowed 
 in from the direction of Lake Superior. 
 
 * Namaiiiae, meaning: the little sturgeon, is the projer spellintf of this naive, but it is ({<-■"':'■•■>">■ 
 incorrectly written, " Mamainse." 
 
The central .subsidence which lias been referred to, ceased before tlie 
 depositicJH of the Cambro-Sihirian strata of the south shore, such as the 
 red and £?'''-'>' sandstones about Grand Island, the Pictured Rocks and 
 Sault Ste. Marie, as well as the small remaining i)atch of fossiliferous 
 beds of Limestone Mountain on Keeweenaw I'oint, lyin<; west of 
 L'Anse, all of which retain the horizontal attitude of their original 
 deposition. 
 
 It is to be noted that around Lake Superior, red sandstones of two 
 formations come together, or nearl_\' so, in flifferent localities ; and at 
 one time they were confounded with one another and both supposed to 
 be of the same age. The one set is associated with the Nipigon or 
 Keeweenian formation while the other is newer. In the region of 
 Keeweenaw Point their contact is not clearly .seen, and there has been 
 some discussion as to their relative ages. Hut here, as in other parts f)f 
 the lake, the tilting of the one and the undisturbed condition of the 
 other serve to distinguish them. 
 
 From what I have said, it would appcir that in the earlier ages of 
 the earth, or until the Cambro-Silurian period, the Lake Superior region 
 was always one of volcanic activity. If we except the unaltered 
 fossiliferous strata in the southeastern part, we shall see that in 
 approaching Lake Superior as a centre from any point on the ancient 
 crystalline rocks which everywhere else surround it, we have furthest out 
 a great area of fundamental gneiss, followed by successive zones of 
 igneous origin. The Huronian rocks, mainly of volcanic character, 
 occupy large areas all around. Then come many masses of intrusive 
 granite of various kinds. These are followed by greenstones of the 
 Animikic formation and finally, inside of all the others, come the more 
 recent amygdaloids, etc., of the Nipigon or Keeweenian formation. 
 These zones are not complete or regular, but the above is the general 
 order of their occurrence. This rude, concentric arrangement of different 
 igneous rocks which become newer and newer as we approach the inner 
 part, would indicate that there was here a deep-seated volcanic centre, 
 dating back to very ancient geological times. The area of volcanic 
 action had thus a great width in the early conditifjn of the earth's crust, 
 but it became more and more contracted as the world grew older and 
 at length the active condition ceased altogether. 
 
 I have thus pre.sented some of the fundamental geological rea.sons 
 \vh}' the present site of Lake Superior was prepared to become a lake- 
 basin, but between the stage in its history at which we have arrived and 
 the next actual evidence as to how the basin was fonned, there was a 
 
long interval, during which we can only surmise what was going on. 
 Cambro-Silurian rocks, forming part of the northern margin of the 
 great area of these strata lying to the southward, occupy the shore of 
 the lake from Marquette to the outlet, a distance of 130 miles, and 
 farther south they are succeeded by the Devonian and (Carboniferous 
 systems in their regular order. Patches of Cretaceous rocks, resting on 
 the Arch;ean, occur in Minnesota at no great distance west of the lake. 
 It is therefore probable that this part of the continent was submerged 
 throughout much, if not all, of the time up to about the commencement 
 of the Tertiary period. During this period there must have been a 
 very long interval of erosion, in which the land surface was deeply 
 sculptured and the present inequalities to a great extent produced. 
 After this, when the glacial period came on, the deeply decompo.sed 
 surface was ploughed up and its materials were transported to greater 
 or Ie.ss distances. Thus its removal from one part of the land and its 
 deposition on another would add to the inequalities of the surface and 
 might deepen and e.xtend the larger existing lake-basins very much, 
 while it would be the means of creating innumerable smaller ones. On 
 the glaciated surface of the crystalline rocks, great numbers of lakes now 
 show at a glance what their history must have been. We seethe glacial 
 furrows and stri.ne descending into the rock-basins on the one side and 
 emerging on the other, while heaps of boulders and drift material are left 
 wherever they could find rest or shelter from the moving ice or where 
 they may have been deposited by the final melting away of the glacier- 
 A few of the lakes have been formed by the damning up of valleys and 
 partial basins by moraines of drift, but the great majorit\- of them are 
 complete rock-basins. There seems to be no limit to the size of the 
 basins which may be formed in this way, and if we extend our 
 conception of the power and volume of these old continental glaciers 
 and imagine them to have acted upon a deeply softened surface, there 
 is no reason why we may not believe that the greatest of our lake-basins 
 might have been excavated in this manner. It is a question of what is 
 most probable, if, in addition to these processes, we take into con- 
 sideration the slow undulation, tilting or warping of the crust of the 
 earth, which is known to have been going on, and which is still in 
 progress all over this part of the continent, we shall have no trouble in 
 accounting for the existence of our great lakes. When we remember 
 how slight is their depth in comparison with their area, we see how 
 easily they could be formed on this e.xtensive plain of the continent. 
 To give you an idea of the insignificance of the actual depth of these 
 great sheets of water as compared with their extent, let us construct a 
 vertical .section through Lake Huron from north to .south on a natural 
 
scale. If the horizontal distance be represented by one yard, we shall 
 find it difficult to draw two lines on paper close enough togcher to 
 show the difference between the surface and the bottom of the lake. In 
 geological as well as engineering profiles we are accustomed to see the 
 vertical measurements so greatly exaggerated that we are apt to over- 
 look the true proportions. 
 
 To unscientific persons who have not given the matter a thought, it 
 may be a surprise to learn that our great lakes with their jjresent 
 outlines and areas, and their existing outlets or connections with one 
 another are all very new, geologically speaking. Indeed all lakes, 
 whether great or small, and in whatever continent ihey may be situated, 
 are necessarily unstable and transient even in their very existence, 
 owing to the movements which are going on perpetually in the crust of 
 the earth and which tend to either raise or lower their outlets — in the 
 latter case tipping the water out — and partly to the fact that the 
 discharging streams are constantly wearing their beds to lower levels. 
 
 The present great lakes of the St. Lawrence arc only remnants lying 
 in the deeper recesses of much larger ones wiiicli existed in compara- 
 tively recent times. There is plenty of evidence to show that in 
 post-tertiary times a fresh water sea e.xtended from the front of our 
 Laurentian highlands southward to the Middle Stales, and that only a 
 ridge west of Lake Sujjerior separated it from another fresh-water sea 
 which covered over all the lakes of the Winnipeg basin and also 
 extended as one sheet, far up the low and level Saskatchewan and Red 
 River valleys. The great valley of the Mississippi has been the site of 
 numerous wide lakes, in the bottoms of which have been preserved the 
 bones of a large nuinbcr of species of curious and interesting mammals, 
 all of which are now extinct. If the northern part of Hutlson's Hay 
 were raised a very little, and its southern part slightly depressed, so as 
 to flood the low lands around it, we should have a fresh-water lake of 
 unexampled extent, rivalling the Mediterranean Sea in area. It is 
 possible that such a lake did rcall)' exist for a short time. Indeed the 
 central part of our continent, ail the way from the Rocky Mountains to 
 the Appalachians and the Labrador Peninsula has been the region of 
 the greatest lakes of the world in tertiary and post-tertiary times, and 
 even their degenerate successors of the present day retain respectable 
 proportions. 
 
 So far as I am aware. Professor Chapman, late of Toronto University, 
 was the first to recognize the former extension of the great lakes in one 
 sheet as the probable explanation of the phenomena of the superficial 
 geology of the whole lake region and sui rounding countr)-. 
 
A great part of the work of excavating the lake basin;; was therefore 
 completed before the glacial period and at its close, the land, which had 
 stood at a considerably higher level than at the present day, had been 
 depressed somewhat below its present position, and the relative eleva- 
 tions of different parts had been slightl\' but sufficiently altered to check 
 the flow of the waters as continuous river-courses, and to flood extensive 
 tracts and form lakes much greater than even the present ones. 
 
 The wide river-valley forming the bottom of Lake Ontario, about 700 
 fett below the present surface of the lake, was not continuous with its 
 present outlet by the St. Lawrence, but turned southward opposite to 
 Oswego, and was continued in the valley of the Mohawk and Hudson. 
 The rivers, whose branching valleys now form Lakes Huron and Erie, 
 probabl)' discharged from the west end of the latter into the old Missis- 
 sippi, as these lakes are known to have done by the Wabash at the close 
 of the glacial period. The ancient river of the valley of Lake Michigan 
 probably flowed south into the old Mississippi, as the lake itself did, at 
 the close of the glacial period, and as it will do again ver\- soon. Lake 
 Superior appears to cover two valleys of ancient erosion, one lying under 
 the western part and the other under the main body of the lake. The 
 bottom or deeper part of the former runs from Dnluth, at the head of 
 tile lake in a very direct course to HIack Bay, keeping quite close to the 
 northwest side. The land along this side of the lake is high and bold, 
 so that with the portion under water added, the total depth of the valley 
 is 2,000 feet. The other valley has several main branches spreading 
 over the remainder of the lake, but all apparently leading also to Black 
 Baj'. The i)re-glacial discharge of the valleys which now hold Lake 
 Superior was most probably by way of Black Bay, the valley of Black 
 Sturgeon River, Lake Nipigon, and thence northeastward directly into 
 the sea, or by way of an ancient river which traversed the site of 
 Hudson's Bay when the continent stood at a higher level. In 
 considering this question, it must be borne in mind that during the 
 period referred to the relative levels of the land to the north and 
 south were reversed, as compared with the present day. 
 
 There is a possibility that the ancient drainage of what is now the 
 basin of Lake Superior was southward, perhaps by way of Train River, 
 near Grand Island. We have thus see.i that the advent of the glacial 
 epoch found all the eastern part of the continent standing jjerhaps 3,000 
 feet higher than at present, with a very old eroded surface total!)- unlike 
 the present one, and with an entirely difTerent topograph)'. There was 
 a general scarcity of lakes and few or no waterfalls or rapids in the 
 larger rivers, as all would be worn down to base levels, 
 
10 
 
 In whole districts the general course of the streams was in the opposite 
 direction from the present. Several <;reat rivers, which existed then, are 
 now entirely wiped out, and the whole topograi^hy is so changed that, 
 looking at the map of the present day, their courses would tiot be sus- 
 pected. Take, for example, that of the old St. Lawrence as mapped by 
 Dr. Spencer, starting from the middle of Lake Michigan, crossing the 
 land to Lake Huron, thence through Georgian Hay, then across country 
 to Lake Ontario, from which it again crossed country to the Mohawk 
 and on down the Hudson to the Atlantic coast, which was then far to 
 the east of its present position. About the same time, as I have stat.id, 
 Lake Superior may have discharged through Hlack Bay and Lake 
 Nipigon into the sea to the north. The present round-about arrange- 
 ment of the discharge of the great lakes, which however is only 
 temporary, has a very unusual appearance from a topographical point 
 of view. The pre-glacial drainage of the valleys which now form the 
 bottoms of three lakes, running in tlie various directions I have 
 indicated, would be more in accordance with what we might expect 
 from the general contour of the country, so that it is not at all extra- 
 ordinary that it followed these lines. In a paper read to the Royal 
 Society of Camda a few years ago, I sketched the hypothetical cour.se 
 and the branching of a great pre-glacial .stream which finally flowed 
 into the nortii Atlantic along the bottom of what is now Hudson's 
 Strait, and which was probably larger than any of the existing rivers of 
 the world. 
 
 The glacial epoch was of long duration and in these latitudes it was 
 broken by intcrglacial periods, each of which probably lasted for a 
 great length of time, and during them the vegetation, which had been 
 driven south, partially returned and must have given the country 
 something of the appearance of the present day. Although the glacial 
 conditions have finall)' retreated as far as Haffinland and Greenland, 
 many of the trees of North America are still in the process of returning 
 as far as po.ssible towards their original home in the north. But we 
 have not the time this evening to pursue this interesting topic. In a 
 general way the climate of these latitudes since the disappearance of 
 the continental glacier has never been better than at the present day, 
 but in certain deposits of more recent date on the north side of Lake 
 Superior, I have found evidence of a milder interval which may never- 
 theless have been some thousands of years ago. 
 
 As the very existence of our great lakes, as well as their former 
 extensions, their successive relations to each other and their drainage 
 
II 
 
 systems are all associated with the depression and subsequent elevation 
 of the land, it will be in order for me here to say something of these 
 moveinent«. Geologists have not yet agreed as to what produced the 
 glacial epoch — a unique event in the history of the earth — but they 
 think the most probable cause of the depression of the northeastern 
 part of the continent during and just following that epoch was the 
 great weight of the mass of ice which had accumulated to a depth of 
 one or two miles over a vast area. The oscillations of '^he land, as 
 compared with the sca-Ievel, which have alwa3's been going on in one 
 part of the earth or another' are mainly due to the shifting of the surface- 
 load by its partial removal from one region to another through the 
 agency of water, volcanic action, etc. The rocks, which appear to be 
 quite rigid on the small scale are really not so, and on the large .scale 
 the)' yield slowly to pressure. The relativeh- small depression 
 produced by this continental ice-sheet was less than might have been 
 expected, and the rising of the land which is still going on, is the 
 rebound, as it were, or the effort to regain its equilibrium after the load 
 has been removed. Around any (jf our great lakes, one may easily 
 observe abundant and distinct evidence of higher stages of the water in 
 the form of terraces, old beaches, ridges, curving spits and other shore 
 phenomena. They are found at many different levels around all the 
 lakes. The principal ones at various heights may be connected .so as to 
 show that the water stood long enough at each of these heights to wear 
 into the land and leave these permanent records. But a curious fact 
 about the beaches and terraces is that in a northeasterly and south- 
 westerly direction they are not horizontal but slope upwards in the 
 former at a rate which is sufficient to be easily measured, amounting to 
 from five inches to three feet or more per mile. This important 
 circumstance was first noted by a Canadian writer in referring to the 
 terraces around Georgian Ha\' about 50 years ago, but I have forgotten 
 at this moment where I read about it. The changes in the elevation of 
 numerous well-marked beaches around the various lakes have been 
 determined throughout long distances by .several well known geologists, 
 among whom I ma)- mention G. K. (lilbert, l*'. H. Taylor, Warren 
 Upham, Frank Leverett and Professor \V. C. Chamberlain of the United 
 States and Drs. J. W. Spencer and A. C. Lawson of Canada, both 
 formerly assistants of my own. The bearing of the line of ma.ximum 
 rise was easily found after the rate of incrca.se in elevation had been 
 ascertained along various lines forming greater or sirialler angles with it 
 and Profes.sor Gilbert gives it as N. 270 E. for the lower lake region. 
 If we look in the opposite direction along this line we would speak of 
 the movement as a depression. We have .seen, iiowever, that it is really 
 
13 
 
 a diflTerential elevation towards the north-northeast, but the rate is not 
 uniform and there may also be local warping of the crust. 
 
 Around the lakes of the Winnipeg basin, a similar phenomenon has 
 been observed. On the western side of Lakes Manitoba and VVinnipego- 
 sis, which were one sheet when the water was slightly higher, the ancient 
 beaches are very v.ell marked, and Mr. J. B. Tyrrell, formerly of our 
 Geolcc[ical Survey, has shown that in going northward they rise at the 
 rate of about one foot per mile or 300 feet in the length of the two lakes. 
 
 Now lines drawn at right angles to the bearing of the maximum rise 
 would represent isobases or axes along which there would be no change 
 of level although the land might be rising to the northward or sinking to 
 the southward. 
 
 The country is tolerably low and level about the outlets of all our 
 lakes and there is no evidence of the former existence of any kind of 
 obstruction to the outflow of the waters — not even of " ice-dams," 
 which used to be convenient suppositions for getting over difficulties of 
 this kind. Before the discovery of the earth movements, which have 
 been referred to, there was thus great difficulty in accounting for the 
 former extensions, higher levels, shifting of outlets and other changes to 
 which we knew the lakes had been subject. 
 
 Around the northern shore of Lake Superior, Dr. A. C. Lawson has 
 ascertained, by the spirit-level, the elevations at a considerable number 
 of different localities of thirty-three of the most conspicuous of the 
 terraces. These elevations range from near the pre.sent level up to 
 more than 600 feet above it. In 1846, Sir William Logan described the 
 remarkable set of distinct terraces at Les Ecrits or Terrace Bay near 
 the mouth of Steel River or Schreiber, on the Canadian Pacific Railway, 
 and Agassiz gives a picture of them in his " Lake Superior" published 
 in 185 1. Dr. A. C. Lawson ascertained that the highest water-mark in 
 this vicinity was 418 feet above the present surface of the lake, but 
 further west near the mouth of Arrow River he found terraces marking 
 old beach lines up to 607 feet above the lake. If the water stood at any 
 of these upper levels at the present day, there is no ground to the 
 southward high enough to prevent it extending to the Gulf of Mexico. 
 But if the sand and gravel in which these terraces are cut had been 
 deposited by the sea, we would be pretty certain to find in them the 
 remains of marine organisms, as we do in the post-glacial deposits of the 
 province of Quebec. Fresh water, especially in a cold climate, produces 
 few or no mollusks, and even if it did, their shells do not last as long as 
 
«3 
 
 those of marine species when embedded in sand or gravel. It is, 
 therefore, probable that at the close of the glacial epoch, an enormous 
 sheet of fresh water covered the whole region of the great lakes of the 
 St. Lavvrence and extended far to the south, and that it was drained 
 away to the southward by a slightly greater elevation of the land in the 
 north. Geologists have named this vanished sea Lake Warren. 
 
 After a time, the land to the south of the general basin of the great 
 lakes had become dry, but all the water above Lake Ontario was united 
 in one body, which included Lake Nipigon and flooded the land north 
 of Georgian Bay as far as Lake Temiscaming. Dr. Spencer has named 
 it Lake Algonquin and it discharged by the Trent valley into his Lake 
 Iroquois which covered the site of Lake Ontario and the surrounding 
 country and flowed out to the sea by the Mohawk valley. 
 
 When the water had fallen to within about lOO feet of the present 
 level of Lake Superior, it remained united with Lakes Michigan and 
 Huron as one sheet in three lobes discharging by way of Lake Nipissing 
 and the Ottawa River. Mr. F. B. Taylor proposes the name Great Lake 
 Nipissing for this former inland sea. The fact that a differential eleva- 
 tion of the land towards the north-northeast has been going on and is 
 still in progress is proved by the undoubted southward inclination of the 
 ancient beaches around the great lakes which succeed one another and 
 together record a movement which was continuous through a great 
 length of time, and also by other phenomena which I have observed in 
 the northern part of the Province of Quebec and in the Labrador 
 peninsula. But we are not dependent on the geological records alone to 
 establish the existence of this movement in the crust of this part of the 
 earth. Professor Gilbert has carefully investigated the readings of 
 various gauges which were placed many years ago at different points on 
 the /Xmcrican side of the lakes and he has found that, after eliminating 
 all disturbing elements, they agree in showing a steady fall in the water 
 towards the north and a corresponding rise towards the south, which 
 amounts to about .42 of a foot per 100 miles per century. I am of the 
 opinion that both the amount and rate of the uplift increase for a certain 
 distance to the northward or until we reach the centre of the nia.ximum 
 height of the ice during the glacial period; and consequently around 
 Hudson's Bay and in the Labrador peninsula the elevation is going on 
 more rapidly than in our lake region. This is only what we might 
 naturally expect if our theory of the cau.se be correct. 
 
 As a consequence of this tipping up, or canting of the lakes, their 
 northern shores are .shoaling, while their southern ones are flooding. 
 
«4 
 
 The quantity of water in the lake will be increasin<j or diminishing 
 according to the position of the outlet. If we draw a straight line 
 through the outlet of each lake at right angles to the line of uplift, this 
 isobase will represent the axes along which its surface is being tilted 
 and on it there will be no change of level, while the further we go from 
 it to the north the greater will be the fall and the further to the south, 
 the greater the rise. This isobase of Lake Superior runs from Sault 
 Ste. Marie to a point on the northwest side near the international 
 boundar)- line. Meron Bay is the most northern and Duluth the inost 
 southern j)art of the lake in reference to this line. At the above rate of 
 tilting, Professor Gilbert calculates that the fall in the lake at the former 
 place is five inches and the rise at the latter six inches per century, or a 
 
 Modern Kiiised Beaches, Pointc linitt', Noilh Shore. 
 
 relative difference of nearly a foot per hundred years. This movement 
 having been going on for a longtime, the contrast in the appearance of 
 the two sides of the lake is quite noticeable. On the north side, we see 
 wide shores and many raised beaches, while on the south shore the lake 
 is washing away the land or the waves are beating against the partly 
 submerged cliffs and the coast has a generally drowned appearance. 
 
 The contrast between these two conditions is well brought out by the 
 accompanying illustrations, one showing modern raised beaches on the 
 north side and the other the rising water eating away the sandstone on 
 the opposite shore. 
 
 Before closing I may say a few words in regard to a possible cause of 
 
«5 
 
 the driftless area to the south of Lake Superior in Wisconsin. The 
 great extent and depth of the basin of Lake Superior, which, as we have 
 seen, was in existence before the advent of the iijlacial period, lyin^f 
 across the course of tlie on-coming land ice, must have had an important 
 
 I'l't'iv near .■ls/i/ii)/(/, Soiilh S/ioir, .s/towiiii; ^laiiiia/ sii/>iiieri;i:inr:, 
 
 effect in checking its progress. This influence of a great depression 
 would be much enhanced by the very considerable height of the land on 
 the south side of the greater part of the lake. 
 
Along the eastern part of the south shore of Lake Superior, glacial 
 strl.t; are not seen, the rocks being mostly buried under sand and other 
 superficial deposits. The general trend of the stri.t is increasingly 
 deflected to the westward of south as we proceed towards the western 
 extremity of the lake. 
 
 The drift or till which is very heavy towards the western part of the 
 south side of Lake Superior is all red, owing to the colour of the 
 sandstones, marls, etc., from which it has been derived, to the east and in 
 the bottom of the lake. Around Thunder Hay and at other places on 
 the northwest side of the lake, the glacial grooving has a westerly 
 course and the debris of the red marl of the lake-basin has been carried 
 out west over the land for a distance of fifty miles or more. In many 
 places on the east shore, the stria; run up from beneath the water and 
 continue inland with an easterly bearing. It is well known that the 
 peculiar form of red jasper conglomerate, which, as far as we are aware, 
 is found in situ only at the north end of Goulais Bay and north of the 
 St. Mark's River, has been carried east with the drift all along the north 
 shore of Lake Huron, from which it has been subsequently moved south 
 and southwest by a more recent and probably lighter system of 
 glaciation. Along the west side of Lake Michigan the stri;^ are 
 reported to have a course west of south. The ice sheet which produced 
 the striie either passed the side o'the driftless area or did not extend to 
 it. We may therefore conclude tnat to whatever extent the basin of 
 Lake Superior was deepened or enlarged by glacial action the excavated 
 material \vas carried mainly to the west and ea.st and failed to reach the 
 driftless area of Wisconsin