Cornell University Library QE 191.B97 ipicoten island. sec 31 499 si UNIVERSITY OF TORONTO STUDIES GEOLOGICAL SERIES No. 3: THE GEOLOGY OF MICHIPICOTEN ISLAND By E. M. BURWASH ee THE UNIVERSITY LIBRARY, 1905: PUBLISHED BY THE LIBRARIAN s COMMITTEE OF MANAGEMENT Chairman: James Loupon, LL.D., President of the University. Proressor W. J. ALEXANDER, Ph.D. PROFESSOR PELHAM EDGAR, Ph.D. PRINCIPAL J. GALBRAITH, M.A, ProressorR R. Ramsay Wricut, M.A., B.Sc. PROFESSOR GEORGE M. Wronae, M.A. General Editor: H. H. Lancton, B.A., Librarian of the University. THE GEOLOGY OF MICHIPICOTEN ISLAND BY E. M. BURWASH, M.A. —_ OE (34 Borr CONTENTS. PAGE I. Inrropucrory. Literature and acknow- ledgments ......... 5 II. GENERAL CONSIDERATIONS AND PLAN OF WORK... 9 Ill. Detar, oF FIELD WorK AND STRATIGRAPH- ICAL RELATIONS OF ROCKS. (1) Section exposed on islands off the south shore, Quebec Harbour and route to Coldwater lake... . : 15 (2) Section exposed on road from Quebec Harbour to Quebec mine.... 22 (3) Section exposed on shores in passing around the eastern end of the island 25 (4) Section on shore from Quebec Harbour to Quebec mine by the west end of the island...... 4 28 IV. PETROGRAPHY....... 30 V. CONCLUSION.... 40 APPENDIX I. Table 1. Glacial striae 47 Table 2. Directions of joints 47 Table 3. Bearings of veins. .. a7 APPENDIX II. Extract from Science, with reference to pyro- clastics....... 47 APPENDIX III. References by R. D. Irving to Michipicoten island in Copper-Bearing Rocks of Lake Supertvor..... ois 6 we 48 THE GEOLOGY OF MICHIPICOTEN ISLAND. I. INTRODUCTORY. Michipicoten Island lies in the north-eastern part of Lake Superior, about ten miles distant from the northern shore, which here for some distance runsa little north of west,curving northward toward the west. The island is roughly elliptical in shape, about sixteen and five-eighths miles long by six miles in greatest width; its longer axis lies nearly east and west or roughly parallel to the coast, and its western end is situated south and a little west of the mouth of the Pukaswa river. It forms one of the largest developments in Canadian territory of the Keeweenawan series of rocks, so extensively exposed and studied by American geologists on the western and southern shores of the lake. Owing to the fact that the copper deposits on the island have not up to the present proved important, its geology has not received at all the same measure of attention that has been bestowed upon the for- mation elsewhere, and a résumé of the literature of the subject may be briefly made. 1. Sir William Logan in the Geology of Canada, 1863, reports upon the island as follows —— ‘The strata of which it is composed have a general dip to the east of south, and the inclination appears seldom to fall short of thirty degrees. The lower strata, toward the north side of the island, particularly as indicated at the upper end, appear to be composed chiefly of amygdaloidal trap, with occasional beds of trap-conglomerates, red sandstones, and shales; while toward the south these are overlaid by a considerable amount of compact earthy or sub-resinous red trap, assuming sometimes an obscure and sometimes a distinct porphyritic character, by the display of ill-defined crystals of red felspar or well-marked crystals of transparent colourless quartz. “ Along nearly the whole of the south side of the island, the trap assumes a more resinous aspect, and its colour be- coming black, it presents the characters of pitchstone and [89] 6 BuRWASH : MICHIPICOTEN ISLAND pitchstone porphyry. Some of the beds associated with these are of an amygdaloidal character and exhibit large agate- veins, which run chiefly in the direction of the strike, but frequently also transverse to it. “ About three-fourths of a mile out in front of the harbour, which is half-way down the south side, a few narrow islands occur, presenting beds of a peculiar character, amounting to between sixty and seventy feet, dipping southward at an angle of twenty degrees. They are of a general red colour, spotted and patched with yellowish white, and wherever a crack exists the rock is blanched to a small distance on each side of it. The surfaces are uneven, and peculiarly marked with festooned and finely-wrinkled forms, composed of very thin close-fitting laminae, with a ligneous aspect, having a thickness sometimes exceeding one or two inches. The rock scarcely resembles a trap, nor does it bear the character of indurated shale; but it may perhaps be an indurated mixture of volcanic mud and ashes, in which the wrinkles result from a partial flow. The total volume of the formation developed in Michipicoten Island, at the most moderate dip observed, would not fall short of 12,000 feet.”’ 2. Thomas Macfarlane, of the Geological Survey of Canada, visited the island in 1865 and reported upon it in his Report on Lake Supertor.' His description of the rocks upon the island occupies several pages, but has been largely rendered obsolete by later methods of petrographic study. His description is chiefly of value for its numerous records of dip and strike, establishing the general trend of the formation and the flattening of the dip toward the southern part of the island. 3. Professor R. D. Irving, in his Copper-bearing Rocks of Lake Superior’ gives the results of a microscopic examin- ation of nineteen rock specimens collected on Michipicoten island by Mr. Macfarlane, comparing thern with rocks of the 1 Geol. Surv. of Canada, Report of Progress from 1863 to 1866, pp. IIS et seq. 2 U.S. Geol. Surv., Monog. V, 1863. [90] BurwasH : MICHIPICOTEN ISLAND 7 Keeweenawan formation on the south and west shores. More detailed reference to his descriptions of these rocks will be made in connection with the study of them in this paper. Not having visited the island, he accepts the stratigraphical account of it given by Logan and Macfarlane. His report establishes the connection of the Michipicoten with the other copper-bearing rocks of the Lake Superior basin, fixing it as a projecting part of the northern edge of the great synclinal trough which underlies nearly the whole of the waters of the lake. The nearest exposures of similar rocks, he states, are at the Battle islands about roo miles in a straight line to the north-west and at a point on the east shore of the lake about two miles north of Cape Choyye, which lies some 18% miles south of the mouth of the Michipicoten river, and about 33 miles east of the eastern end of the island. 4. The Report of the Royal Commission on the Mineral Resources of Ontario (1890) refers to the general structure of the island and its resources in copper and agate (p. 38). Evidence as to the nature of the copper deposits and the extent of the mining operations carried out upon them was given by Messrs. Joseph Cozens, J. S. Williams, and E. P. Borron. From the evidence of the first mentioned gentleman it appears that copper occurs in two distinct classes of deposits on the Charles Jones location, a conglomerate vein some eight feet in width and an amygdaloid, upon which the principal work was done, both deposits being parallel with the general strike and consisting of impregnated beds, as at the principal deposits of Keeweenaw point, and not of veins transverse to the formation. Mr. Williams described the deposit on the Bonner location as consisting of quartz, calcite, epidote, iron pyrites, and native copper, and running about east and west, or parallel with the general strike at that part of the island (the north shore). Mr. Cozens also reports the discovery of native silver on this location (pp. 60 and 103). 5. During the summer of 1898 Professor A. P. Coleman, acting for the Bureau of Mines of Ontario, visited the island, and made a cursory examination of its principal features. [91] 8 BURWASH : MICHIPICOTEN ISLAND In his report (Bureau of Mines of Ontario, 1899) he describes the raised beaches which occur on the island in common with the rest of the Lake Superior shores, and says :— ‘A very interesting group of beaches is to be found in the western end of Michipicoten island, both on the north and on the south sides, easily examined on the road leading over the hills from the Quebec mine to Quebec Harbour. There are several fairly distinct sets of gravel beaches, each having a number of ridges below the highest one. The lowest one is from 15 to 22 feet above the water and is marked by a fine sea-cave whose floor is at the lower level. Next come indistinct beach-lines running up to 65 feet above the lake, where a terrace affords space for several houses. Above this there are faint stages culminating at 78 feet, and then appar- ently a gap until 128 feet, when fairly distinct wave lines begin and run up to 178 feet. At 204 feet there is a well- marked gravel beach, perhaps just the highest stage of the last series; and at this level on the south side of the divide a lake is found, probably dammed by a beach-deposit. Above this there is a wide terrace occupied by a farm, its highest point being 233 feet. There is a lake at this level also. The highest terrace observed reaches 295 feet. The terrace lying somewhat above the 400-foot level, found at so many points to the east and south-east, was not observed, but may yet be found among the higher hills of the island which were not explored. A comparison of the series of beaches just given and those of Dog river, 35 miles to the north-east as measured by Lawson, shows comparatively few points of agreement, the only exact one being the level of the lowest beach in each, 15 and 16™% feet respectively. Whether there has been differ- ential elevation, placing the beaches at Dog river higher than the corresponding beaches at Michipicoten island is hard to determine, particularly when one recalls that almost all these beaches are composite and have a range of a number of feet.” Dr. Coleman also described somewhat in detail the quartz porphyry at the eastern end of the island. His report also [92] BurwasH : MICHIPICOTEN ISLAND 9 mentions the occurrence of Keeweenawan rocks on the north shore between Pilot Harbour and Pukaswa, not noted by Irving, which seems to indicate that the entire channel be- tween the island and mainland is underlain by these rocks, and accounts for the conglomerate, with Keeweenawan peb- bles, which forms one of the lowest members of the series on the island. This point will require attention at a later part of this paper. The writer's sincere thanks for encouragement and assist- ance are due to Professors T. L. Walker and A. P. Coleman, of the University of Toronto, to the Honourable Charles D. Walcott, Director of the United States Geological Survey, to Mr. G. F. Matthew, of St. John, New Brunswick, to Mr. Joseph Cozens, D.L.S., of Sault Ste. Marie, and to Dr. A. G. Wilson, of McGill University, Montreal. II. GENERAL CONSIDERATIONS AND PLAN OF WORK. The island, as already described by earlier explorers, consists of a series of well-defined eruptive beds, with some interstratified sedimentaries. The general strike is a little north of east, or parallel to the longer axis of the oval, and the beds all dip to the southward at angles which range from 55° on the north-western side to about 14° on some islands off the southern shore. By far the greater part of the thickness of the deposits consists of eruptive flows, most of them several hundred feet in thickness, as compared with the sedimen- taries which occupy a thickness in most cases of from one to ten feet, except at the west and north-west parts of the island, where conglomerates of considerable thickness occur. The igneous beds, with one important exception, (that occupying the north shore of Channel lake) are continuous from shore to shore across the island. They seem to attain their greatest thickness in the centre and eastern parts of the island, tapering down to an inconsiderable thickness at the western end, so that nearly all are crossed by the road which runs from Quebec Harbour to the Quebec mine and appear at the west end of the island, while their eastern or north-eastern extremities [93] Io BuRWASH : MICHIPICOTEN ISLAND are found distributed all along the greater part of the eastern and northern coasts, although the strikes at the east end show also a distinct but less marked convergence. To this fan-like arrangement in part is due the apparent confusion found by Macfarlane in endeavouring to trace the series along the eastern shore. The sedimentary rocks, on the other hand, are not continuous across the island, so far as observed, but are largely developed only at the western end, where the igneous flows thin out giving the impression at first sight that the conglomerates found here were deposited upon the flanks of the extruded masses. The igneous beds at this part of the island have also a less perfectly porphyritic and more felsitic texture, a laminated structure due to flow or pressure, and also a steeper dip than at the east end. The igneous flows have in general a tendency to jointing in a direction varying, along the south shore of the island where it is most easily observed, from 315° to 325°, or roughly perpendicular to the general strike. These facts, taken in connection with the former higher levels of the lake and wave action at those levels, explain nearly all the topographical features of the island. Here, as throughout the Keeweenawan formation,’ the gently dipping eruptive beds present ridges, or cuestas, sloping gradually on the dip side and breaking off sharply in cliffs or steep slopes of talus upon the opposite, in this case the northern, side. The intervening valleys of erosion? were no doubt largely produced by the waters of the lake at higher levels, a process which may be seen in operation at the present time along the southern shore of the island. The inland cliffs were then shore-cliffs of straits or bays of the lake, like those on the northern side of Hope, Long, and Stack islands. On the second of these islands a crack several feet in width ex- tends parallel to the cliff and about twenty yards distant from it, showing the undermining action of the strait, not over a mile in width, which separates these islands from the 1 Irving, Copper-beuring Rocks of Lake Superior, p. 141. 2 See Tig. x. Lo4] BuRWASH : MICHIPICOTEN ISLAND II main island! The removal by the waters of the lake of a softer (acidic) bed, through an opening broken in a more resistant basic one, is well illustrated by Quebec and Cozen’'s Harbours, both of which are hollowed out of the same flow, a reddish brown quartzless porphyry, and protected by a barrier of black rock which resembles the ‘‘ ashbed diabases ”’ of Keeweenaw Point, and has in parts the properties of a pitchstone porphyry. The basic rocks (diabases) are the most resistant to erosive forces, especially the compact or vitreous kinds, and these invariably form ridges of considerable height, running with the strike; the highest point on the island, that used by the American Hydrographic Survey for its station, stands 937 feet above the lake and is composed of a dark grey pitchstone porphyrite. The quartz porphyrites attain a considerable height as ridges only at the east end of the island, where their quartz phenocrysts are most fully developed. In the centre they are often eroded down to a level, as at Quebec Harbour, and underlie somewhat extensive swamps, or lakes of rounded outline, the remains of ancient bays similar in character to Quebec Harbour. These flats are intersected with hills some- what irregularly arranged, the remnants of once continuous ridges. The same result is obtained at the lake on the road about two miles south of the Quebec mine, where a number of softer basic beds of steep dip, together with a large porphyry bed, are worn down to a flat surface, at the level of an ancient beach whose sand covers thinly their upturned edges. ? Examples of beds worn down to the level of the water in this way are reported from the Keeweenawan deposits on the eastern shore of Lake Superior. Even the most resistant ridges are traversed by valleys, like the Saw-tooth Moun- tains of the Minnesota coast described by Irving, giving a characteristic serrated appearance, which is well seen from the deck of a steamer passing along the south-eastern shore 1 This crack, together with the depth of the strait (20 fathoms), may, however, be due to a fault along the north side of these islands. 2 Irving, op. cit. pp. 162-164. [95] 12 BurwasH : MicHIPICOTEN ISLAND of the island. These valleys seem to correspond to the lines of jointing already mentioned as being constant in direction, and not being exactly perpendicular to the strike give the successive hills of the same bed an appearance of being set en échelon, or with corners toward each other, rather than in line abreast. The general plan of exploration determined upon before visiting the island involved an examination of its coast line, where the best exposures are to be had, and the making of as many sections from north to south across the island as possible. For this purpose it was thought that some of the survey lines run about 1881 might still be avaiiable, but as this proved to be impossible, on inquiry after reaching the island, owing to the density of the undergrowth in most parts of it, the plan had to be modified. The road from Quebec Harbour to Quebec mine gives one very good section, running as it does across nearly all the principal beds. The lake north of Quebec Harbour, known as Channel lake, is connected by an old road with the harbour and by trail with Birch and Coidwater lakes to its west and north. There is also a trail running about a mile from a point on its north shore, nearly opposite to where the road reaches it, along a trout stream which enters through a valley cutting the cuesta on the north side of Chan- nel lake, and another trail extends from near its east end northward along the east boundary of the Bonner location to a point near the northern side of the island. Apart from these routes, the only possible method of sectioning would be by compass lines through the woods, or possibly with a canoe, and portaging from lake to lake. It was found, how- ever, that the time available for the examination of the island, about three weeks, was fully occupied in the examination of the rocks along the more accessible routes above mentioned. A micrometer survey was made of the road from Quebec Harbour to the mine, plotting the geology, and keeping a record of the principal heights by barometer. As the islands off the south shore are a somewhat important feature in the geology of the whole and are insufficiently indicated on the [96] BurwasH : MICHIPICOTEN ISLAND 13 Crown Lands Department map which was used as the basis of the work, a day was spent in making a survey of them also. The old road to Channel lake, now so grown up as to leave only a path, was surveyed, the shores of the lake examined by means of a boat kept there by the lighthouse-keeper, to whom we were also indebted for information as to trails, and the trails radiating from this lake to the northward were examined, and mapped by pacing with the compass. In this way the shores of Birch and Coldwater lakes were visited, but the day set apart for the exploration of the long trail northward from the east end of Channel lake unfortunately proved so stormy as to prevent our reaching the point where it begins in the small and unsafe boat available for the purpose. The exploration of the coast was finally undertaken, and, with some special attention paid to the neighbourhood of the mines at the north-west shore, occupied the rest of the time. It is very much to be regretted that at least one compass line could not have been made across the eastern part of the island, whose topographical, as well as geological, details thus remain unknown except as we were able to study them from the shore or infer them from facts observed on the western part. The map submitted with this paper, is, as already noted, a copy of that in possession of the Surveys Branch of the Crown Lands Department of Ontario, to which has been added the topographical features given by Joseph Cozens, O.L.5., in the plan of his survey of the lots between the Bonner and Charles Jones locations. Upon this have been plotted the surveys made by the writer, including the islands off the south shore, the road from the harbour to the mine, with the lakes along its course, various trails, the more prominent topographical features noted, the different rock areas and a number of dips and strikes. In plotting the geology of the isand, the lines of contact have been drawn between the points where they were fixed in general as directly as possible, having regard to known strikes of the beds themselves or of parallel beds. These lines have, however, been varied a little in accordance with the topo- [97] I4 BuRWASH : MICHIPICOTEN ISLAND graphical features. In almost every case, throughout the island, the northern side of an important hill marks the north- ern boundary of a flow, and where the lines upon their esti- mated curves ran at all near to such hills they were altered if necessary to include them. Indentations in the boundaries have also been made at the crossings of streams or lakes, as in every valley the northern edges of the flows, all of which dip southward, necessarily recede to the south. The field work, as carried out, gives three complete sec- tions and one partial section, across the island. Commencing at Quebec Harbour and following the coast in either direction, we pass over all the beds exposed, completing both sections at the Quebec mine, where the lowest beds occur. The road from the harbour to the mine gives a third, and the route con- necting the harbour with Coldwater lake part of a fourth, including an important bed not found in the other sections. Of these sections, that along the road and that around the east end of the island are the most complete. The western end of the island, which is the one especially described by Macfarlane, has some very important gaps owing to the fact that a group of softer beds, here very thin, which reach the shore in a small bay toward the north-west angle, are covered by gravel beaches. These beds, however, are well exposed along the northern and north-western shore, where they occupy considerable space. Arepetition occurs in the eastern section on the north-eastern shore, where the strikes bend toward the south. A quartz porphyry flow, which runs into the lake along the southward curving shore on location Y41, reappears again for a short distance to the south-east on Y42, so that this bed and the diabase just above it in the series are encountered twice in succession in following the shore. The very striking flow of diabase-porphyrite seen along the north shore of Channel lake is not met with either on the road or at either end of the island, and therefore runs out at con- jectural distances east and west of the points where it was observed. The rocks described by Macfarlane as forming a succession along the south shore from Quebec Harbour east [98] BuRWaAsH : MICHIPICOTEN ISLAND 15 appear to be for the most part merely variations in texture of the same flow, which includes vitreous, felsitic and por- phyritic parts, and it has been marked as continuous upon the map. For convenience of reference the different beds have been numbered in order from the top downward, no. 1 being the felsite forming the islands off the south shore. In order to give as clear as possible an account of the con- ditions on the ground as traversed and also to consider each important bed as a whole, the next division of the paper will deal with each of the sections already mentioned separately, and in the petrographical division each bed will be considered separately throughout its length. Il. DETAIL OF FIELD WORK AND STRATIGRAPHICAL RELATIONS OF ROCKS. (1) Section exposed on islands off the south shore, shores of Quebec Harbour, and routes north to Birch and Coldwater lakes. 4 The islands off the south shore form the highest member of the series exposed and represent a partly submerged cuesta separ- ated from the nearest shore by a strait about 4400 feet in width. The whole distance from the shore to the outside of these islands, at the lowest dip observed (12°), would represent about 960 feet in thickness of deposits, and it is probable that the thickness does not much, if at all, exceed 1,000 feet. The chain of islands begins a little to the east of the mouth of Quebec Harbour and extends about five miles and a half in a direction 5° south of west. A slight break in the con- tinuity of the line of islands seems to indicate that the two eastern islands represent the upper portion of the flow. The eastern islands are also the highest of the group, which is due no doubt to greater erosion having taken place at the western end of the chain. The easternmost, or Hope island, has a general east and west bearing. The rock here, described by Logan, consists of laminae from a quarter to a third of an inch in thickness, having a general strike parallel with that of the island and 1 See Map of the islands. [99] 16 BurwasH : MICHIPICOTEN ISLAND the bed, though showing many undulations, bearings taken at various points varying from 270° to 290°. The dip of these laminae is still more variable, and appears to be in most cases steeper than the dip of the flow. It is in some cases even in the opposite direction, the readings varying from 40°S. to 70°N. In accordance with its structure this island slopes considerably more rapidly on the north than on the south side, and this applies to the slopes under water as well as above. Long island, the next in the series, is the largest, having a length of about 6,500 feet on a bearing of about 261°, with a maximum width of 600 feet. It affords the best example of the cuesta structure, having a cliff along its north shore 140 feet in height at its highest point, and a gradual ascent on the southern side. Its southern side and its ends are in the form of a terrace, having a general height of about 15 or 20 feet above the water (See Fig. 2). This area was no doubt worn down to the water-level when the lake stood at the level represented by the first of the series of beaches found on the island and by the old sea-cave at the Quebec mine. Its surface is not by any means flat but presents a series of hummocks of approximately equal height formed by the enlargement of the joints in the rock, which are now represented by open trenches reaching down in many cases to the present water-level. This low area extends across both Hope and Long islands at about their centres, so that each was formerly divided into two islands. This gives a general suggestion as to the manner of division of the cuesta ridges into separate hills. In accordance with this corres- pondence may be looked for between the levels of old beaches observed on the island and elsewhere on Lake Superior and the heights of the passes between the summits of the cuestas throughout the island, a correspondence which was in fact observed in all the cases examined. It is also in accordance with this that in most, if not all cases, such passes do not slope up to a summit and fall immediately away again, but continue for some time level at their highest point, being in [100 | BurRWASH : MICHIPICOTEN ISLAND 17 fact a part of the terrace to the level of which they corre- spond. The sketch of Long island from the east affords a good illustration of this arrangement. A channel carved in this way, beginning originally with the enlargement of a joint in rocks projecting a little above the water, might be continued downward through a number of lower levels, or might be left dry at any one of them by the receding waters, dependent partly upon the persistence downward of the joint and partly by the dimensions attained by the channel at its first stage. The question as to the possible depth to which such a channel could be excavated below the water-level by wave action is also of importance in this connection. Sug- gestions are afforded by the depth of water in the channel north of Hope island, which is charted at 20 fathoms, and that toward the south side of Quebec Harbour, 8 fathoms. The entrance to Quebec Harbour is a typical example of a channel crossing a cuesta, and it has at least 8 or 10 fathoms of water. It would appear, therefore, that wave action, combined with the strong currents which move into and out of the harbour under the influence of the wind, are sufficient to excavate channels of a considerable depth, although the first result of wave action alone is merely to reduce the rocks to a flat surface very little below the water level.! This must be thought to be the case in the absence of evidence to prove that the water of the lake ever stood at a lower level than now, which could hardly be premised without demonstrating the existence of level benches at a depth too great to be the result of wave action at the present level. The strikes and dips of the laminae observed on Long island were as follows : Location Dip Strike 100 paces from east end 17° 25° West end 243° The rock on Long island displays good examples of the bleaching of the rock from purplish red to yellowish white along the joints. 1 See footnote on page I1. [ror] 18 BURWASH : MICHIPICOTEN ISLAND The depth of 20 fathoms, already mentioned on the north side of these islands, occurs on the south side, as charted, a considerable distance out, and no greater depth is shown between Long island and Cariboo islands some thirty miles to the south. : Stack island is separated from Long island by a channel whose direction corresponds, not to the direction of the joints, but rather to that of the strike of the laminae, which here bends toward the south across the general strike of the flow. Stack island also no doubt represents a lower part of the flow. The jointing is very prominent on this island, as well as on Long island. The bearing of the joints is 315°, and where excavated by the waves on the level terrace adjoining the south shore they form long trenches, like little canals, as much as roo yards in length. The rock is much the same as on the other islands already visited; the direction of lamination at the west end of the island is 60° dip, S. about 60°. On Ship island, about a mile and a quarter west of Stack island, the cuesta formation is not so apparent, the whole island having been worn down to the level of the terraces already described and presenting a low flat appearance, no- where more than 15 or 20 feet high. The same difference in rate of increase of depth off the north and south shores else- where observed holds here. Lamination is quite irregular in direction; the strikes and dips observed were as follows : Location . Strike Dip S. E. point 45° W. point 66° 35° Average 50° Two principal directions of jointing were observed, viz. 298° and 340° On Green island, which is nowhere of great height, the laminae strike 47° and dip about vertically. This island has a gravel beach on its northern side, the only beach noticed on the entire group. A small island to the west of it has been split into four islets by two large eroded joints crossing at right angles and forming canals through the island. [x02] BuRWASH : MIcHIPICOTEN ISLAND 19 The most western island visited, one of the so-called Gull Rocks, is low and does not exhibit the distinct lamination found in other parts of this flow. The rocks which flank the entrance to Quebec Harbour belong to the flow which extends along a great part of the south coast of the island and furnishes most of the chalcedony found upon it. It varies in colour and texture from a black vitreous to a dark grey or greenish felsitic rock, both varieties exhibiting a marked tendency to columnar structure in parts. The islands within the harbour mouth also belong to this flow. On the south side of Agate island the rock is fairly distinctly marked by columnar divisions, and is black and vitreous in appearance; on the north side it is much more slatey in tex- ture and grey in colour. Large amygdules filled with ame- thyst and agate occur, and also small veins of chalcedony. There are other amygdules filled with crystalline red feldspar and with a green earthy material. The amygdules here are very large and apparently represent a ‘‘ basal amygdaloid,” ! a feature which also appears at other parts of this flow, and throughout the formation. The felsitic parts of the rock on this island are less columnar than the vitreous parts, but on the next islet to the west the felsitic trap shows well de- veloped columnar structure, the columns varying in position from horizontal to a dip to the north of 70° On Fisherman’s island, the largest of the group, there is a vein of red feldspar running east and west and containing brecciated fragments of the wall-rock. The island south of Fisherman’s island has rock of a compact texture and well marked columnar structure, and the same description applies to the rock on the point at the west side of the harbour mouth. The rock on the lighthouse point is black, resinous-looking, traversed by cracks which suggest columnar structure, and affects the compass. This band is no. 2 on the map. Following the harbour shore eastward from the lighthouse, a peculiar detrital rock was observed lying between no. 2 and the next igneous bed, and consisting of a fine-grained 1 Irving, op. cit. p. 135. [103] 20 BurwasuH : MICHIPICOTEN ISLAND green matrix containing angular fragments of the red felsitic rock below; the bed is of an inconsiderable thickness occupying some 20 feet of shore and is followed by the reddish brown porphyritic rock with green spots which forms the bed of the harbour and its northern shore. This rock appears on the south shore of the harbour as far as the fishing station of the Booth Packing Company, beyond which no rock exposures occur round the shores of the harbour until a point on the north side considerably west of the old mining company’s wharf is reached, where the same rock is exposed. It shows here a laminar structure with laminae from one to four inches thick and bearing 333°, dip vertical. The western end of the harbour is also composed of beaches made up of the under- lying rocks. Following the path northward from the old wharf toward Channel lake, we pass over an old beach at a slight elevation for some 375 paces, before encountering an exposure which marks the base of the red quartzless porphyry of Quebec Harbour, and contains inclusions of the rock immediately below. This is followed by the rocks of the next flow, repre- sented at its upper side by a black vitreous rock with pitted surface due to the weathering out of phenocrysts, and inter- sected by what appear to be apophyses of the overlying flow. Immediately beyond these exposures a well marked terrace of sand and gravel occurs, of which the top has an elevation of about 125 feet above the harbour. The level at the foot of this terrace has about 90 feet elevation. Taking a general dip of 17° for the two flows intervening between this point and the lighthouse, a thickness of about 530 feet is given for the pitchstone and of 695 feet for the quartzless porphyry. The pitchstone flow may be thicker, owing to extension under water, but any additional thickness here would have to be taken from the estimated thickness of felsite in the islands to the south; hence the known figures may be allowed to stand. Northward of the terrace mentioned a raised beach occurs 7 feet higher (132 feet), and finally another level occurs at 195, and another at 240 feet, which is also the level of Channel [104] BURWASH : MICHIPICOTEN ISLAND 21 lake, and therefore of its outlet, flowing through a channel worn in this hill as explained on page 17, and the summit of the trail, lying considerably below the hills on either side, is about 330 feet above the lake. The direction of the pass through which it runs is about 30°. The rock at the summit is the same black rock of resinous or pearly lustre, pitted surface, and conchoidal fracture seen at the top of the flow. Along the south shore of Channel lake the rock forms cliffs from 100 to 200 feet high, exhibiting a well marked columnar structure. In places at the foot of the cliffs this rock is more felsitic, and greyish or greenish in colour, with red streaks and spots, and also with a red “ paint,’’ which appears to be mainly iron oxide, marking its joints. The north shore of Channel lake rises by richly wooded slopes to a high ridge, divided at a point just opposite the landing on the south side by a defile through which a trail leads north. The highest point of this defile is about 300 feet above Lake Superior, at which level it expands, on the north side of the ridge, into an extensive level grown up with alder. The rock of which this range is composed is very similar to the top part of the flow on the south side of Channel lake, showing, if anything, a more closely pitted surface and more vitreous lustre.) A remarkable feature of these hills is that, contrary to the gen- eral rule, they exhibit along their southern slope near the top a line of cliff facing south, which might be referred to erosion by the water when at that level, but has no apparent level bench at its foot. It therefore seems likelv that it marks the line of a fault, similar to those observed by Lane on Isle Royale, hading south and with its down-throw on the south side,” the line of fault being parallel to the strike. Lack of time, however, prevented a closer examination. The pitted rock seen on the shore is the only one encountered in the defile between the hills (exposed at 850 paces north of the lake), which militates against the supposition that the irregularity is due to a dyke. 1 The trough of Channel lake 1s possibly formed by a fault. If this is the case, the northern side is a repetition of the southern. 2 Geological Survey of Michigan, vol. 6, pt. 1. [105] 22 BuRWASH : MICHIPICOTEN ISLAND Coldwater lake, half a mile north of Channel lake, lies at a level of forty feet below it (200 feet above Lake Superior), and drains to the north. Passing over the trail from Channel lake, the trail for 250 paces rises only 10 feet above Channel lake (250-foot level); then a steep rise of 75 feet to a level of about 325 feet follows. Just following this the black pitted diabase porphyrite gives way to a red quartz porphyry, and the trail drops to the 300-foot level which it maintains with a variation of a few feet until about 150 paces from Coldwater lake, where the descent to the 200-foot level begins. Cold- water lake lies on the northern edge of the first of three suc- cessive flows of quartz porphyry, which extend through the centre of the island from end to end, and are eroded to a level surface over a large extent of their area, underlying most of the swamps and lakes of the centre part of the island. The thickness of the two basic flows on the south and north sides of Channel lake, measured on the 240-foot level, seems to be about 1660 feet, taking the dip as 17°. From the west end of Channel lake a trail about 1,000 paces long, bearing 315°, runs over the northern ridge to Birch lake. The top of the ridge is 520 feet above Lake Superior, and Birch lake about 420. A schistose rock found on the northern slope suggests alteration along the line of the fault already noted farther east on the same ridge. (2) Section exposed on traverse of road from Quebec Har- bour to Quebec Mine—Leaving the old wharf on the north side of Quebec Harbour, the road winds westward along the shore, from which, however, it gradually recedes, and rising over successive banks of shingle reaches the foot of the terrace already noted at about one and a third miles from the wharf. The rock exposed up to this point is entirely the red quartzless porphyry seen on the shores of the harbour. After ascending the terrace, whose top here is 115 feet above the lake, the road follows a straight west course along a lot boundary for over two miles, the level continuing the same except where traversed by the valley of a creek at about two-thirds the distance. Near the end, however, a gradual descent begins [106] BurwasH : MICHIPICOTEN ISLAND 23 and the red quartzless porphyry is exposed again at a point about half a mile from where the road, which here curves to the south, reaches the lake. Before reaching the lake there are several exposures of the basic flow lying under the quartz- less porphyry, and shortly after reaching it the contact is exposed on the shore. Between the two flows there exists a detrital bed, whose matrix, originally a clay, is now of the nature of a porcelain, and contains fragments of the basic bed below it and of still lower quartz porphyries. This slate also contains numerous small bodies which at first appearance resemble fossils. This is especially true of some whose forms suggest lingula, as the black material of which they are com- posed gives a decided reaction for phosphorus. A number of these were referred to Charles D. Walcott, Director of the United States Geological Survey, and to G. F. Matthew, of St. John, N.B., both of whom decided against the hypothesis of organic structure for these bodies. Dr. Walcott, after referring to somewhat similar forms from the argillites of the Pipestone beds of Minnesota, and the argillites of the Belt Mountain beds of Montana, whose origin is traceable to con- cretionary action, says :— ‘* T very much regret that I cannot identify them as fossils, as it would be a most interesting and valuable contributioa to the pre-Cambrian faunas. The presence of the phosphatic material appears to indicate the presence of organisms, but I have found the same in connection with concretions in the pre-Cambrian Algonkian beds of the Belt Mountains, Mont., and the Grand Canyon series of Arizona. Phosphatic nodules occur in the Cambrian rocks of New Brunswick (described by G. F. Matthew), which appear to be of organic origin. What the life was that produced the phosphatic material in the rocks that you found I think is yet to be determined.” Dr. G. F. Matthew replied as follows -— ““T have examined the objects you sent me from Michipi- coten Island, and cannot recognize them as fossils. They appear to me to have been concretionary growths in what has originally been, perhaps, a bituminous shale. But whether [107] 24 BurwasH : MICHIPICOTEN ISLAND that or not the objects are not now bituminous. They have a fibrous structure which recalls the cone-in-cone concretions of the upper division of the St. John group, which, now in carbonaceous shale, were probably at first in bituminous shale.” (See Fig. 3). North of this belt the black melaphyre of Channel lake succeeds, here at its southern edge displaying occasional large amygdules filled with quartz crystals or chalcedony. The weathering of this rock here also takes on spheroidal forms composed of concentric layers which weather off one after another (pearlite ?). From the contact, the road follows the shore at the level of the lowest raised beach, about 15 feet, the same rock outcropping on the north side of the road, for about a quarter of a mile, then turns away from the lake and ascends to a level about 45 feet above the lake, where the same rock, here exhibiting a definitely por- phyritic character, is exposed for a distance of 177 yards from the lake, when the contact with the underlying red por- phyry is reached. The road follows the contact for some distance, the two rocks being alternately exposed as far as 978 yards from the shore where the last exposure of the mela- phyre is passed. The level here is about 7o feet. At the edge of the first porphyry cuesta the height is 210 feet, and the swampy lowland immediately succeeding is at about 180 feet. The thickness of the first porphyry flow here passed is estimated at 356 feet. The dip at its southern edge is 20° Ascending the slope of the next cuesta, levels occur at 200, 230, followed by a valiey cut down to 180, and finally the ridge of the cuesta rises to 325 feet. A lower level (250 feet) follows, and the northern edge of the flow rises again to 310 feet, then drops off again to the outlet, of the first lake, at 235 feet. The total thickness of this second flow of quartz porphyry at a dip of 20° (estimated) is about 1160 feet. It is immediately succeeded by another, after (apparently) a thin layer of red sandstone. The thickness (both beds) is estimated at 1493 feet. On the flanks of this ridge marked levels occur at 310 feet, and the lake which follows it stands at 300 feet. The [108] BuRWAsH : MICHIPICOTEN ISLAND 25 quartz porphyry here is succeeded by a completely crystalline olivine-diabase, and this in turn by a succession of thin beds which may be described as (1) pitchstone (diabase), (2) quartz porphyry, (3) melaphyre (amygdaloidal at the top), (4) altered olivine diabase, (5) ‘‘ lustre-mottled melaphyre,”’ all of which occur along the shores of the lake. Their dip is not apparent but may be taken as about 30°.! At this dip they have a total thickness of 1575 feet. On the road between this lake and the Quebec mine, for a great part of the distance the rocks are covered with soil. The 300-foot level is main- tained for some distance, to a point about 1,000 yards from the mine opposite a large hill on the west, when the descent begins and continues to the lake. Dr. Coleman has noted beaches here at 233, 204, 128 to 178, 65 to 78, and 15 to 22 feet respectively. Below the lustre-mottled melaphyre bed last mentioned, there is a felsite flow verging upon quartz porphyry, having a thickness of 532 feet, and this is followed by a band of diabase porphyrite of 463 feet, which forms the mass of the lofty hill to the west of the road (thickness esti- mated for 40° dip). Below this flow there follows a slope covered by soil, which is probably underlain by three distinct beds, a conglomerate, a porphyry, and a “‘ lustre-mottled ”’ rock, of which, however, only the conglomerate and the por- phyry are exposed. The total thickness of the three is at least 1140 feet, measuring to the summit of the cliff behind the houses at the mine, which is composed of an amygdaloid agglomerate. This is followed by a lustre-mottled rock, this in turn by a diabase, and this by an amygdaloid which forms the lowest exposed rock, on the point in front of the cave used as a powder house. The thickness of these beds (to the water line) may be fixed with considerable accuracy at 645 feet. (3) Section exposed on the shores mm passing round the eastern end of the island from Quebec Harbour to Quebec mine—Leaving the lighthouse at the harbour mouth and proceeding eastward 1 Dip estimated from west end of the island, where it is somewhat greater than 30° [109] 26 BurwasH : MicHIPICOTEN ISLAND along the shore, the same flow is exposed on the shore and islands lying along it for six miles. On the islands lying somewhat east of the harbour it is traversed by veins of calcite, which have the bearings 183° and 228°. The courses which have the latter direction are filled with brecciated fragments of rock. The rock weathers out on both sides of the veins, which stand upright in the centre of the troughs thus formed. The dip is about 14°. Much of the bed exhibits columnar structure, either distinct or incipient, in some places the cracks produce cubes instead of columns, in most parts the fracture is distinctly conchoidal and the lustre vitreous. The cracks, as also the veins, are stained with a red ‘‘ paint,’’ due to the weathering out of magnetite in the rocks. In the wide shallow bay about a mile west of Cozens Harbour there are several large brecciated veins. These are distinctly banded, the largest is from 4 to 8 feet in width, has a central band of crystalline calcite, with duplicate red and green bands flanking it in the order given. The rock here takes the character of a diabase porphyrite like that at Channel lake. At the eastern end of this bay the upper part of the porphyry of Quebec Harbour is exposed. This is separated from the pitchstone, which is continued on the point to the east of the bay, by a bed of stratified shale containing some calcite. Just above this shale the base of the pitchstone flow is of a dark chocolate colour, and is apparently of the nature of a basal amygdaloid. A body very much resembling an orthoceras was found in this rock (See Fig. 4). It is composed of chalcedony with some calcite and presents a chambered appearance; the septa, however, are perfectly straight instead of being curved. It is surrounded by a somewhat denser layer of the rock itself, which corresponds to Irving’s description of the cylindrical “spike amygdules ’ of basal amygdaloids elsewhere in this formation. He says, ‘‘ The amygdules at times take on a cylindrical form, with the axis at right angles to the bedding, sometimes extending in this way to a length of several inches. I have noticed such ‘ spike-amygdules ’ in a number of cases 1 ‘Op. cit. pp. 135, 136. [110] BurRwasH : MICHIPICOTEN ISLAND 27 in the thin basal amygdaloids.”” Somewhat similar bodies, but resembling fossil forms even more closely, have been de- scribed by M. E. Wadsworth, and ascribed by him to non- organic origin.! An undoubtedly pyroclastic rock was obtained by Dr. Coleman from this locality, and very probably immediately underlies the pitchstone bed, but was not observed by the writer. As the existence of pyroclastics in this formation has been doubted by Irving, who ascribes all the lavas of this region to fissure vents without explosive action, this specimen, now in the collection of the School of Practical Science, Tor- onto, is of importance. Further eastward the pitchstone flow occupies the shore as far as the south-eastern point of the island, with the ex- ception of Cozens Harbour and the small bay just west of it, whose northern shores are composed of the red rock of Quebec Harbour, here a quartzless porphyry with well developed phenocrysts. This rock is again exposed on the eastern shore of the island and is followed to the north by a band of grey diabase, of felsitic texture corresponding to the rocks of Chan- nel lake, which extends as far as the north-east point. Along the northern shore this is followed by the softer porphyry beds and the olivine diabase in the same order observed on the road from Quebec Harbour to the mine, and following them the ‘ ordinary ’ diabase and the narrow quartz-porphyry flow below it are exposed twice as already explained? A diabase which succeeds the quartz porphyry occupies the shore as far as the western boundary of the Bonner location, where the peculiar reddish altered rock underlying it on the road follows. From this point to the mine the series is some- what difficult to follow, owing to the occurrence of gravel beaches backed by earth terraces, but there seems little doubt that all of the igneous beds can be accounted for, nos. 18, 19, 20 and 22 being exposed in that order. 1 Boston Soc. of Nat. Hist., Proc. vol. 23 (1888), p. 208. 2 Supra, p. 14. [rrr] 28 Burwasu : MICHIPICOTEN ISLAND The grey diabases at the north-east point of the island form a high range of hills, but the quartz porphyries to the west of them are uniformly low; the large flat in the centre of the island north of Channel lake seems to extend to the shore at this part of the island. The central one of the three por- phyry flows forms a low cuesta which reaches the shore at a mile and a half east of location Y46. At the point on Y46 some pieces of St. Mary’s sandstone were found, and this rock appears in place under water a little to the east of here, lying in nearly level beds, and probably overlies unconform- ably the Keeweenawan beds which underlie the channel between the island and the north shore. There seems to be nothing in the relationship of these rocks which would suggest a fault running along the north shore of the island, and the trend of the coast seems rather to be determined entirely by the strike and wearing qualities of the rocks than by a fault line, which, one might expect, would follow a more regular course. Were a fault present in line with the northern part of the island, and the receding of the coast to the east due to erosion, the sandstone would not be found so far south as location Y46. These considerations are of interest in view of the debate as to the existence of a fault in the south-east side of the Keeweenawan rocks of Keeweenaw Point, whose formation recalls that of Michipicoten island, in other details as well as in the relationship of these two classes of rocks. TABLE OF STRIKES AND DIPS. Locality. Strike, Dip. Islands east of lighthouse. . 14° 8. Shale 1 mile W. of Cozen’s Harbour 110° 25° 3. N. shore Cozens Harbour..... aa 25° S: Two miles W. of Y46.. ‘ 110° 25° Ds Point on Y39........ 100° 41S, E. boundary Bonner locaticn.... 85° 44° S. Bay on Y38 273° 55° S. (4) Section on shore from Quebec Harbour to Quebec mine by the west end oj the tsland—Westward from the harbour [tra] BuRWASH : MICHIPICOTEN ISLAND 29: mouth the pitchstone flow continues for about two miles to a point facing west, after which it reappears again on two small islands 21%4 and 314 miles farther west. In the bay which fol- lows the point the red quartzless porphyry of Quebec Harbour is exposed, and is succeeded by an argillite as already described, where the road skirts the shore, and this in turn by the black pitchstone (diabase) porphyrite of Channel lake. On ‘“ Junc- tion Point ’’ just west of here, it is a black, vitreous-looking rock, with a tendency to columnar structure, and traversed by veins. Similar rock appears to occupy the shore as far as the next point, where a fine columnar exposure exists. Along the curving shore toward the south-west point of the island the underlying red quartz porphyry comes to the shore, but on the point to the west this is overlaid by a recurrence of the Channel lake rocks in columns with bright red paint in their interstices, black vitreous lustre and conchoidal frac- ture. There are also large red and green banded veins and agate amygdules here similar to those in the higher basic bed (no. 2) toward the east end of the island. Some fragments of argillaceous sandstone occur here; probably a thin sedimentary bed exists between the pitchstone porphyrite and the quartz porphyry which succeeds. The quartz porphyry is here much nearer to a felsitic character, the phenocrysts being often very small and a laminar structure apparent. It is followed to the north by a very coarse conglomerate, having boulders of Laurentian and Huronian origin as well as those derived from the Keeweenawan, the whole cemented with calcite. Thin flows of quartz porphyry seem to occur at intervals through the conglomerate exposure, which occupies nearly a quarter of a mile in width. It is followed on the north by a quartz porphyry, the lowest of the three observed together elsewhere. Following this are diabases which here exhibit a columnar structure. A number of bands are probably concealed, from this point on, under the gravel beach of the most northerly of the two bays on the western end of the island. At its northern point a massive black diabase is exposed, probably corresponding to that observed in the large hill [113] 30 BurwasH : MICHIPICOTEN ISLAND south of the Quebec mine. Below this is another conglomerate of similar materials to the first, which is followed by a fine- grained reddish flow and this by a ‘‘ lustre-mottled melaphyre”’ which occupies a great width on the north-west point. Below it is an amygdaloid agglomerate, immediately succeeded by a conglomerate composed of Keeweenawan materials, whose upper part has been mined for copper at the south-western shaft on the Jones location. Following the shore north- eastward toward the mine, this is succeeded by another band of lustre-mottled rock. Following this is a diabase with an amygdaloidal belt above it, a red felsitic flow, another diabase, and finally an amygdaloid, on the point near the cave at the mine. These lower beds are all quite thin and dip about 55°. Of all the rocks on the west end of the island, it may be said that they dip steeper and show a much finer grain than in the more easterly parts of the same bed. TABLE OF STRIKES AND DIPS. Location. Strike, Dip. S. shore opposite Stack Island. 85° 18° Southwestern point...... go° BP About 4% mile N.of S.W. point... . 30° 36° S. side of N.W. point....... : 84° 34° N. side of N.W point........ 2 wa 107° 35° Phillips shaft..... .. 55° 50° Between shaft and shore.... . 65° 45° Shore half way from Phillips mine to Quebec mine. . on 26° 35° Point in front of cave........ ‘ 55° 55° IV. PETROGRAPHY. 1.1 This flow is represented on the islands to the south of Quebec Harbour. The rock has been described by Irving as identical with that at Mount Houghton on Keeweenaw Point, at the Porcupine mountains, in Michigan, and on the Minne- sota coast of Lake Superior, but here occupying a higher 1 The numbers correspond to the numbers on the map. [114] BURWASH : MICHIPICOTEN ISLAND ai position than elsewhere. His microscopic description is as follows :—‘‘ The groundmass is nearly colourless, cloudy, and thickly dotted with very minute ferrite particles, which are at times aggregated into waving lines. In the polarized light feebly polarizing flecks dot a dark background, some of which are recognizable as quartz-network clusters; no por- phyritic ingredients in the section.”” A number of sections examined enable the writer to add that this even texture is varied at times by augite grains, tabular plagioclases, and aggregates of highly refracting crystalline quartz, the first two generally rare, the latter sometimes quite large and num- erous, but not in clearly defined crystals, and probably of secondary origin filling cavities. 2. The variations in the ‘ pitchstone ’ flow along the south coast have already been partly described. Irving describes five sections of this flow, including probably all of its impor- tant variations. There is in general a glassy groundmass, filled with more or less developed plagioclase microliths, augite grains and magnetite particles. This represents the pitchstone phase. In felsitic varieties these constituents are so enlarged as to fill nearly the whole area, the plagioclases assuming a tabular form, while the porphyritic varieties have a matrix closely resembling the glassy phase, with phenocrysts of plagioclase, augite, and magnetite. The rock may there- fore be described (see Irving) as a diabase porphyrite, although in the felsitic parts, which are nearly wholly crystalline, the ophitic structure is not present. In some varieties, probably belonging to the basal amygdaloid, or what corresponds to it, in this bed, the tabular plagioclases tend to assume a common direction. The felsitic parts of this rock are interbedded with glassy parts, neither occupying a distinctive position in the flow. The veins which are formed throughout this flow are probably, to some extent at least, alteration bands, being made up largely of brecciated fragments of the rock itself with interstices filled with quartz and calcite. The augite in such cases is represented by hornblende, and the residuary magma by iron a a chloritic material. II5, 32 BuRWASH : MICHIPICOTEN ISLAND 3. The pyroclastic specimen obtained by Dr. Coleman from the south-east shore of the island probably belongs to the base of this flow. It is composed of dark green lapilli of rounded form embedded in a lighter green serpentinous ma- terial, and is distinctly different from the breccias described by Macfarlane, which occur lower in the series and are probably, as Irving maintains, of a different origin. The materials in Dr. Coleman’s specimen are certainly more closely related to the basic bed near which it occurs, while Macfarlane’s breccias are connected with the quartz porphyries both in position and composition. 4. Between the pitchstone bed and the quartzless porphyry of Quebec Harbour, a thin band of sedimentary origin inter- venes. At Quebec Harbour this consists of fragments of the quartzless porphyry embedded in a matrix made up of frag- ments of quartz and chloritic material. At a point at the east end of the broad bay west of Cozens Harbour, a finely stratified siliceous argillite with much calcite in crystallized masses is exposed running across the base of the point. 5. The quartzless porphyry found at Quebec Harbour and eastward corresponds to the rock called by Macfarlane por- phyritic melaphyre and described by Irving (p. 346). The section described by him, however, seems to have been taken from a bed lower in the series. In common with many other flows on the island, especially the acidic ones, this bed is fine- grained at its western end and at Quebec Harbour, while toward the east its phenocrysts are large, up to % of an inch, and at the east end of the island it displays large phenocrysts of quartz corroded by the base, altered orthoclases, and oligo- clases with a matrix showing alteration to secondary quartz. (See Fig. 5). This also is described by Irving and by Mac- farlane as occurring confusedly, but there seems no doubt of its belonging to the same flow as the rocks of Quebec and Cozens Harbours! (See Fig. 5, no. 1). 6. The slate or porcelainite bed which follows has a dark grey, very fine-grained appearance and breaks easily into 1 Irving, op. cit. p. 346. [116] BuRwasH : MICHIPICOTEN ISLAND 33. small irregular fragments. It contains rounded pebbles of the diabase porphyrite just beneath it and of the quartz por- phyries which lie below them. The matrix would seem to be chiefly composed of the materials of the first mentioned rock in a fine state of subdivision. Its present appearance gives the impression of change due to the heat of the flow above it, producing the porcelainitic character. 7. Diabase porphyrite extends from the south-west corner of the island to its north-east corner, and is largely exposed along the south shore of Channel lake. This rock is undoubt- edly somewhat simliar to the bed at the mouth of Quebec Harbour with the exception that augite is more prominent in both the felsitic and porphyritic phases. The upper part is porphyritic, the lower more felsitic and columnar. A section taken from the south shore of Channel lake near the eastern end shows a nearly colourless groundmass filled with great numbers of magnetite grains and green chlorite grains, the whole polarizing dark. Quartz grains seem in parts to form a network. A few grains of pyroxene occur. The appearance of the rock microscopically varies from a black resinous porphyrite with pitted surface to a dark grey or grey- green felsitic rock with no phenocrysts apparent. 8. On the north shore of Channel lake a very fine example of diabase porphyrite occurs. It has a black matrix of resinous lustre and cherty fracture and many phenocrysts of well striated plagioclase and magnetite visible to the unaided eye, which weather out, giving the surface a pitted appearance. A thin section shows a colourless or light brownish ground mass sprinkled with magnetite grains and with much more numerous microliths of plagioclase forming an almost con- tinuous network. In this are large phenocrysts of augite (sometimes twinned), olivine, magnetite and plagioclase, the first three usually associated, the plagioclase separate, also calcite (?) in the section examined showing a fine spherulitic arrangement with wandering extinction (Fig. 5, No. 2). 1 As to possible identity of beds 7 and 8, see page 21 supra [117] 34 BurRwasH : MICHIPICOTEN ISLAND 9. Below no. 7 at the west end of the island a small bed of red and green banded argillaceous sandstone occurs. 10. This flow is the first of three successive quartz porphyry flows, between the second and third of which a wedge-like bed of conglomerate intervenes at the west end of the island. The breccias described by Macfarlane seem to be the upper parts of these beds reduced to a fragmental condition by surface cooling of the viscous flow. This, at least, is the origin attributed to them by Irving. The following descerip- tions are from different parts of the island :— (a) From a point half a mile west of the north-east point of the island:—A breccia of quartz fragments, felsite, and quartz porphyry, in a matrix of diabase whose base (‘“‘residual magma’’), or augitic ingredients, is entirely altered to chlorite. There are masses of calcite. This might result from the penetration of the basic flow above into the fragmental upper part of the quartz porphyry. The diabase porphyrite seems in some other cases to have penetrated into fissures in the quartz porphyry to a considerable depth. (b) From the outlet of the more southerly of the two lakes on the road to the Quebec mine:—This belongs to a flow not capped by a diabase, hence we should not expect a basic matrix. A dark reddish earthy matrix with some light coloured soft fragments and some of a dark brown colour, both felsitic (See Macfarlane, quoted by Irving! ). Both of these might be produced in the way suggested for the first. It is impossible to suppose the first to be the result of explosive action alone, while the second might result either in the way suggested, the interstices being in this case filled either with acidic instead of basic material from a succeeding flow, or, possibly, by the molten parts of the flow to which they belong. An explosive origin would also seem to be pos- sible as stated by Selwyn,” but their position at the upper parts of the flows to which they correspond militates against this hypothesis. The basic pyroclastic obtained by Dr. 1 Op. cit. p. 436. 2 Science, vol. I, no. 1 (Feb. 9, 1883) and vol. I, no. 8 (March 30, 1883). [z18] BURWASE : MICHIPICOTEN ISLAND 35 Coleman seems to occur on the other hand at the base of a basic flow. As to the position of the acidic “breccias” there can be no doubt, as this form of rock is found at the south- west corner of the island and on the road to the Quebec mine at the top of the acidic series. In the latter case the matrix is basic as on the north-east shore, and the breccia occurs immediately below a diabase and at the top of the uppermost of the quartz porphyry flows. As to the lower part of these flows they seem to follow the general rule observed already, being felsitic in appearance and laminated at the west end of the island with very few if any visible phenocrysts, and porphyritic at the east end. The uppermost of the three flows consists of quartzless porphyry at the road, but on the shores of Coldwater lake shows well developed phenocrysts of both quartz and orthoclase. The orthoclase in these cases develops first so that the rock, as in the case of the Quebec Harbour rock, passes through quartz- less into quartziferous porphyry. In these cases the quartz- less porphyry commonly shows a large amount of secondary quartz, filling cavities or forming a network in the groundmass. 11. A flow similar to no. 10, but not so well exposed on the road. 12. A thick bed of conglomerate intervenes between two porphyry beds at the west end of the island. The pebbles are rounded and run up to several inches in diameter. The matrix is partly calcite, and partly reddish brown material derived from the Keeweenawan flows adjacent. The pebbles included are (1) biotite gneiss of Laurentian origin, (2) green- stones, possibly Huronian, (3) reddish and purplish porphyries and amygdaloids from Keeweenawan sources. This bed extends only a short distance inland. 13. This porphyry flow, the lowest of the three already mentioned, develops a definitely porphyritic character farther west than the upper ones, showing fairly large quartzes where crossed by the road. It has a brecciated upper part with acidic filling (see under 10), which was observed both on the western shore of the island under the conglomerate and at the outlet of the most southerly of the two lakes on the road. [119] 36 BuRWASH : MICHIPICOTEN ISLAND 14. Underneath the quartz porphyries is a wholly crystal- line rock belonging to the lustre-mottled melaphyres of Irving or the ophites! of Lane. This rock does not present a poikilitic appearance macroscopically, owing to the large size of the feldspars, but in the thin section the augite is found occupying optically continuous areas enclosing plagioclases and excluding the magnetite and olivine. The plagioclase is much altered and cloudy and appears to be of the oligoclase- albite variety. The olivine is partly altered to a reddish mineral and some chlorite also exists. 15. This is a fine-grained dark grey or greenish grey bed, the upper part of which is composed of fairly large angular fragments cemented together by a material which offers more resistance to atmospheric agencies than the fragments, and hence projects on weathered surfaces. The thin section shows a greyish groundmass filled with microliths of plagio- clase and rarer grains of magnetite, and stained red in patches by iron oxide with occasional small phenocrysts of plagioclase and apatite, and infrequent olivine, augite, and orthoclase grains. There are some grains of quartz probably secondary. The rock may therefore be classed with the diabase porphy- rites or with Lane’s Tobin porphyrites or navites.” 16. A thin bed of quartz porphyry exposed on the road at the northern lake and twice on the north-east shore. It has a glassy matrix with grains of magnetite and microliths of feldspar probably oligoclase and prominent phenocrysts of orthoclase and quartz. On the shore at Y42 the quartzes are very large, much exceeding the orthoclases. 17. A fine-grained diabase with amygdaloidal upper part (See Fig. 6). The amygdaloid corresponds with similarly situated rocks elsewhere in the formation, having a nearly opaque reddish matrix with clusters of plagioclase needles not forming a network, and amygdules filled with calcite and quartz. In the lower part the ophitic structure is well de- veloped, the interstices between the plagioclases being occu- 1 Geol. Survey of Michigan, vol. vi, pt. 1, pp. 51 and 169. 2 Geol. Survey of Michigan, vol. vi, pt. 1. [120] Burwasi : MICHIPICOTEN ISLAND 37 pied by augite which here also shows poikilitic massing, and magnetite. There are also large masses of chlorite with associated secondary iron ore, probably alteration products of olivine. There is, however, no olivine present unaltered in the section obtained at the road,' and the rock here might be assigned to the class called by Irving olivine-free diabase. It forms the northern part of the Bonner location and it seems probable that the copper deposit there is a very much altered part of the amygdaloid belonging to this bed. The rock at the eastern boundary of the Bonner location shows the same masses of greenish material observed on the road, which in this case seemed’ to consist partly of prehnite, with some chlorite, and displays some comparatively unaltered olivine. The rock therefore ranks with the ophites (Fig. 6). 18. This rock may on examination prove to be a part of the last described. It is exposed on the road toward the northern part of the northern lake and at the western boundary of the Bonner location on the north shore. At the first point the plagioclase, which is much altered and cloudy, forms a network containing augite and magnetite. There are also large masses of chlorite forming pseudamygdules as in the rock last described. The augites are segregated into areas and there is some ‘“‘undifferentiated magma” largely altered to chlorite, and also defined by the plagioclases. The chlorite masses show vermicular and spherulitic forms. Olivines are not greatly altered. The rock is evidently merely a coarse form of ophite. At the west boundary of the Bonner location the red colour is due to the fact that the plagioclases are stained red by iron oxide. A similar rock obtained from Thunder Bay contains % per cent. of copper. 19. The rock which follows is a finer grained ophite and may possibly be of the same flow as the last two, though a considerable distinction in appearance is found on the north shore. 20. The next rock is a felsite, brecciated at the top; both fragments and matrix show secondary quartz network, and 1 Geol. Survey of Michigan, vol. v1, pt. 1, p. 169. [rer] 38 BurwasH : MICHIPICOTEN ISLAND the rock probably exhibits porphyritic phases in other parts. The specimen observed was obtained at the road. 21. A dark grey or chocolate grey rock, quite fine-grained, follows, forming the hill to the south of the Quebec mine, and exposed at the north point of the small bay about half a mile south of the north-west point of the island and probably on the north shore at Y32. At the west end it dis- plays very little, if any, porphyritic structure. The section shows merely a pale yellowish magma flecked with darker grey, the whole showing a parallelism due to flow or pressure and dotted with magnetite grains. 22. Just north of the flow last mentioned a second con- glomerate is exposed on the west shore. It consists of a green- ish magma, probably derived from the basic flows below it, contains rounded pebbles of granite, purple and flesh-red porphyries, and of a fine-grained green basic rock. This conglomerate may perhaps extend as far as the road which it crosses near the site of the burned farm-house just south of the mine. 23. A purplish felsite follows the conglomerate, having apparently a brecciated upper part. 24. This is succeeded by a lustre-mottled melaphyre or ophite, which occupies a great breadth on the north-west point of the island, where it in parts displays columnar struc- ture. If crossed by the road it lies just south of the cliff at the mine, but is not exposed. The characteristic nodular surface is well shown on this flow. 4. Melaphyre porphyrites of Channellake 1660 5. Quartz porphyries (1) 355 (2) 1160 (3) 1493 [124] BURWASH : MICHIPICOTEN ISLAND 41 6. Beds exposed at lake on road 1575 7. Felsite 513 8. Diabase porphyrite 463 9. Beds underlying farm (3) 1140 10. Several beds at mine 645 Total 11,230 This thickness is somewhat less than Logan’s estimate (12,000 feet), and considerably less than Macfarlane’s which is 18,500 feet. The work of the latter was probably vitiated by his supposing a continuous series from the north-west corner to the south-east corner of the island where he places the highest rocks of the series. The discovery of Keeweena- wan rocks on the north shore of Lake Superior opposite the island renders the thickness important, as a thickness of 18,500 feet on the island would require the lower Keeweenawan here to be regarded as much thicker than at any point deter- mined by Irving. Taking the general strike of the rocks on the island into account, there seems no reason for supposing that the thickness of the rocks underlying the channel should be estimated at the extreme dip (55°) found at the north-west part of the island, for which the sharp curves of the beds suggest a local cause, but rather at the dips observed further east along the north shore, which range from 25° at the east end to 55° at the west. Taking the average 40° and the dis- tance across the channel as 10 miles the thickness represented would be 34,000 feet, which with the thickness already esti- mated on the island would give a thickness of over 45,000 feet with probable additions to be made at the top of the series. Supposing the dip to be 30°, which is a fair average of the dips throughout the island, and above the average if we neglect the north-west part, the thickness beneath the waters of the strait would be 26,400, and the total 37,630, which is not far above the 35,000 estimated thickness of Keeweenaw Point. The thickness of the lower beds on the island, which occur only at the north-west corner and run out on the north-western coast, would have to be ar ta this total, which would 125 42 BurwWaAsH : MICHIPICOTEN ISLAND thereby be reduced by 2,770 feet to 34,860.! This thickness. would therefore appear as a minimum, leaving the steepening of the strata at the north-west part of the island unaccounted for. The fact that the beds at the eastern end of the island narrow considerably, while the dip remains low, makes it probable that the island represents an exceptionally thick part of the formation, and may be situated close to one of the original vents, from which the flows were extruded. The rock observed by Dr. Coleman on the north shore between Pilot Harbour and the Pukaswa river was a basal conglomerate lying unconformably upon the Huronian and composed of Huronian and Laurentian material. The, above considerations make it appear possible to account for the total thickness of rocks here without supposing a great fault along the north shore of the island. The island as a whole might therefore be considered as somewhat similar in formation to the Niagara escarpment. In both cases the beds first laid down on the flanks of the archean have been worn away for some distance to a plane which slopes gradually away from the archean rocks, until beds are reached of sufficiently resistant power to form an escarpment, in this case the wide group of basic beds of the northern part of the island. If we suppose the lower part of the series now submerged to be chiefly acidic, which agrees both with the materials found in conglomerates on the island, and with the observed facts as to the lower part of the forma- tion elsewhere, we may readily suppose them to have had comparatively low resistant powers, as this is so evidently the case with the acidic flows on the island itself. The gradual upheaval of the edges of the synclinal during the period of deposition? has been held to account for the gradual steep- ening of the dip shorewards, found alike on Keeweenaw Point, Isle Royale, and Michipicoten. In this direction a possible explanation of the peculiar curve southward of the beds at the west end of the island may perhaps be found. The western 1 The possible faults already described would reduce this thickness still further. 2 Irving, op, cit. p. 418. [126] BurRWaASsH : MICHIPICOTEN ISLAND 43 end of the island lies, as has been stated, almost opposite to the centre of the curve of the coast from a west to a north-west direction. This bend forms a re-entrant angle in the edge of the great synclinal trough of Lake Superior. If we suppose the trough to have been subjected to lateral pressure, such as would tilt up the flows around its edge, we must also suppose that the resultant of such constriction exerted from all sides of the trough would be exerted in a direction parallel with its margin. The points at which such a pressure could most easily gain relief by folding would be at exactly such a re- entrant angle. The result of pressure exerted from the east acting against pressure from the north-west, would evidently be the raising of the rock at the point of impact of the two forces, with a tendency to move them in a south-westerly direction, and the natural result would therefore be an anti- clinal fold having its axis in a direction west of south, and pitching in the same direction. If we imagine such a fold denuded, the steepest dips would not be at the apex of the fold, where they would exceed, however, the general dips along the coast, but upon its flanks, and the observed facts here correspond exactly to this. As we approach the fold along the strike of any given bed the dip of the bed increases, simu- lating a twisting of the beds, reaches a maximum of 55° at the Quebec mine and falls again to about 35° at the western end of the island, which represents the apex of the anticlinal. The western side has disappeared beneath the surface of the lake. This hypothesis would also account for the progressive opening of fissures indicated by banded veins on the south side of the island. If we examine a map of the Lake Superior synclinal,' a general triangular form is to be observed. About the middles of the three sides of the triangle are three such re-entrant points. The longer south side has a large one represented by Keeweenaw Point with its continuation to the south-east- ward through Stannard’s Rock. The shorter north-east and north-west sides are indented by points outside of which lie 1 Irving, op. cit. pp. 410-412. [127] 44 BuRWASH : MICHIPICOTEN ISLAND Michipicoten island and Isle Royale respectively. Keeweenaw Point, like the western end of Michipicoten island, shows only one side and the crown of the anticlinal arch. The corres- pondence of dips on the flanks and at the apex of the two anticlinals is quite perfect, the greatest dip shown on Irving’s map south of Portage lake (flank of fold) is 55°, and the dip near Copper Harbour at the northern end of the point is 30°. Isle Royale, unlike the other two, would seem to consist of only the summit of the anticlinal, which in this case is much broader and less marked. Its flank may perhaps be repre- sented by Thunder Cape. This theory would also account for the sharp bend in the strike of the Keeweenawan in the Michipicoten bight. At Michipicoten island the copper deposits are entirely in beds parallel to the strike, and occur upon the flank of the fold, not at its apex. On Keeweenaw Point copper is found in beds and transverse veins. The latter occur mainly near the north-eastern end of the Point, i.e., near the summit of the fold, while the beds as at Michipi- coten occur on its flank. On Isle Royale the only copper deposits are in transverse veins. The rock mentioned in another part of this paper from Thunder Bay containing copper is part of a flow and may represent the bedded deposits of this fold. The promontary of Mamainse is probably a minor fold of the same nature. The copper there occurs in transverse veins. A sort of subsidiary fold appears in the Porcupine mountains to the west of Keeweenaw Point, and a similar one causes the repetition of one of the beds (supra p. 14) on the north-east shore of Michipicoten island. These considerations suggest further reflections especially as to the origin of the transverse veins. Irving has ascribed them solely to alteration, but their general position at the summit of such folds would certainly point to another con- clusion, while allowing sufficient time in the opening of the fissures for the brecciated material of their walls to be retained in the vein stone, and even for a considerable amount of such alteration as he describes to go on along the sides of the fissure, accounting for the differences in width in the same vein in [128] BurRwWaASH : MICHIPICOTEN ISLAND 45 different flows. The thinning of the flows upon the flanks of the folds is quite marked at Michipicoten as on Keeweenaw Point, and the deposition of copper in the flows at these points seems to be connected with the alterations produced under the great tension to which these beds must have been sub- jected during the formation of the fold. The question of the faulting which might be expected in the formation of such a fold, and its comparison with the system of faults known to exist on Isle Royale and Keeweenaw Point, does not come within the scope of this paper, except that as the lifting would be greatest at the edge of the synclinal, faults are to be expected. The deposition of the conglomerates, which are evidently in some way associated with the fold at the west end of the island, may perhaps be referred to the raising of the rocks at this point above water and the consequent action of erosive forces upon them, which would cause the curved beds actually observed to be laid down upon the seaward side of the emerging fold. The felsitic nature of the igneous flows at this point, which farther east display a porphyritic character, may be explained by their comparative thinness over the arch of the gradually rising anticlinal. The succeeding flows may, therefore, be said to lie to a certain extent unconformably upon one another, and the upper ones form part of an arch of much broader and more gentle slope than that seen in the lower. This may perhaps account for the fact that as we cross the island from north to south the felsitic or glassy character of the rocks extends further east. Here as at Keeweenaw Point a horizontal sandstone lies unconformably upon the lower members of the series. At Michipicoten, however, it is not possible to observe its relations fully, but it would seem probable that the situation here differs from that to the south and east of Keeweenaw Point in that the Keeweenawan rocks appear on both sides of the sandstone. South of Keeweenaw Point its shoreward edge is covered by newer strata, so that it is impossible to observe [129] 46 BurwasH : MICHIPICOTEN ISLAND the lower limit of the Keeweenawan deposits, even if the east- ern sandstone itself has been, as appears to be the case at Michipicoten, laid down in a trough, whose edges on both sides were Keeweenawan. Rocks. The igneous rocks on the island, apart from those fully described by Irving, seem to be almost entirely of the sort variously known as lustre-mottled melaphyres, olivine diabases, or ophites, with their accompanying amygdaloids. So far as the writer was able to determine without analysis they are probably the same chemically, or quite nearly so, as the melaphyre or diabase porphyrites of Channel lake. The possible connection of the lamination, so noticeable on the islands off the southern shore, with the anticlinal structure above described is suggested by the fact that these laminae become more and more transverse to the flow in following it from east to west, and that their dip to the south corresponds closely with the dips of rocks at the north-west part of the island, and their position would seem in general to be at right angles to probable lines of pressure. The similar rocks de- scribed by Irving are also in part situated in folded parts of the formation, namely, the Porcupine mountains and Mount Houghton on Keeweenaw Point, but do not show clearage parallel to their banding, as these rocks do. [130] BURWASH : MICHIPICOTEN ISLAND APPENDIX I. Table r. Glacial Striae: At east end of the island 171° and 192°. Table 2. Directions of Joints on south shore : Cozens Harbour 145° Long and Stack islands 135° Shore opposite Stack island 140° and 72° Table 3. Bearings of veins on south shore : Cozens Harbour 93° Cozens Harbour (cross courses) 208° W. side of wide bay W. of Cozens Harbour 103° Islands 1 mile east of Quebec Harbour 183° and 228° 220 Conglomerate at west end of the island APPENDIX II. Extract from Science, vol. 1, No. 8, (March 30, 1883). “I think, if Professor Irving could visit Michipicoten Island, he would be able to recognize plenty of volcanic detrital matter, or tuffs, among the copper-bearing rocks. extinct tertiary volcanoes of Australia enables me very readily to recognize such rocks when seen, but their occurrence at Michipicoten, and else where on the north shore, isno proof that they also occur to the south, and therefore I fail to see why Professor Irving should dissent from my statement on this point.’’ (Signed) ALFRED R. C. SELWYN. [131] a 47 The vast areas over which I have examined the ejectamenta of the 48 BURWASH : MICHIPICOTEN ISLAND APPENDIX III. Irving, Copper-Bearing Rocks of Lake Superior. Index references to Michipicoten Island :-— Michipicoten Island and vicinity .p. 341-346, 415. Amyegdaloids of... Dp. 345. Area of Keeweenaw series on. Pp. 27. Diabases of...... Boas: is P- 343. Diabase porphyrite of....... é p. 78, 80, 85, 87, 433. Dip of rocks on...... Pp. 343. Felsiteof.... 0 ....... p- 103, 112, 343, 346. Lower division on........ ‘ si p. 160. Macfarlane’s specimens from, described .. .. ...p. 85, 86, 87, 93, 112, 342-6. Peculiar ‘breccias’ of....... 2: ies Pp. 436, 437. Quartz porphyry of.......... .Pp- 93, 112, 346. Relation of acid and basic rocks of, :p. 433, 434. Stratification on... . = se Spesads Thickness of rocks of...... . Ps 342: [132] (°S payserr yeyy Suraq iat oq1 07 adoys ayy ‘deut aq} lz payxseut MOY oy} Jo sysisuoo punoswsyoVG 911} UE [LET “ |IMos OF Tou WOT Sit jo adeys [vionad Surmoyy ) ‘UNOAAVH SNAZOD AO ANH LSVa—'I “OIA Re ee 4 Bogie ‘dNW’ISI AAOH WOUA ANW ISI DNO'I—'? “OM FIG. 3.—PSEUDO-FOSSIL FIG. 4.—SPIKE-AMYGDULE. FIG. 5. (1) Quartz-porphyry, east end of Michipicoten Island. I, Quartz phenocrysts eaten by matrix and bordered by magnetite- grains, 2, Altered, cloudy, and fibrous-looking orthoclase. 3. Matrix. indicating partial outlining of quartz-network by magnetite- grains. (2) Diabase-porphyrite, north shore of Channel Lake. Plagioclase shown as if between N+. Magnetite black, wavy lines indicate augite. Mag— netite grains and minute plagioclases in base. FIG. 6. (1) Amygdaloid, (17). Amygdule with chlorite calcite and quartz. Matrix of iron-stained unindividualized matter with porphyritic plagioclase, and alteration products of augite (?) and olivine. Vesicular part of a rock similar to (2). (2) Ophite, or lustre-mottled melaphyre. 1. Augite in large areas, optically often continuous. 2. Plagioclase, intersecting augite and base. 3. Chlorite. 4. Magnetite, and dark base with some altered olivines. (‘deur uo se ‘039 SMO Jo Sioqtunu puB ssulyoyeG | Joay UL oNR] aaoqe Ss} Slay ‘ premyyion Sutsearour diq = *auly ay} WO STIIy sali, peyop ‘peor sjaasaidai apyoid prey‘ enba sayeos [eyNOZII0Y pue [ed1}19A ) ‘(puny fo dvjy aa5) (a) MOOM TINO OL (Vv) ANIA O8aan0 WOU AWAONd AGNV NOILOAS JA 3NO tag FIANS THOT oO Gp 2D) [an Basic Erpuprives (Feows) Ke) Alero LE RUPTIVES. Zz a NS si ass Gz) Co) ” THREE MILES. 6 “ & Ww > 9 N So s © 8 = = n wy z 8 K ° Q = = ~~” MMB 2. ro ccssric. 2500 FEET LEGEND. LAMINATION. 3" STRIKE AND OP. 477 ee B52 YOINTING. 2 e EM Barwa ss Det 202 Nee. SB ee ae 2500 FEET UNIVERSITY OF TORONTO STUDIES ‘Review of Historical Publications relating to Canada, edited by Professor GEoRGE M. Wronc and H. H. Lancron. Vols. I.-VIII. Publications of the years 1896-1903. Vols. 2, 3, §, 6, 7, 8, each... .....000. $1.00 ($1.50 in cloth.) Vols. 1 and 4, each..... . «$2.00, only sold with complete sets. ‘History and Economics, Vol. I. Comprising f x, Louisbourg in 1745, the anonymous “ Lettre d’un Habitant de Louisbourg,” edited and translated by Professor GEORGE M. WRONG...... cc cee cee eee eee e nee aes $0.75 2. Preliminary Stages of the Peace of Amiens, by H. M. BOWMANS