%^ \\ ^ v* IMAGE EVALUATION TEST TARGET (MT-3) ^ 1.0 1.1 11.25 ltt|2£ 125 itt IM 122 S? Ufi no ■luu 1I& |U 11.6 .^ ^ >* '/ HiotQgraphJc Scifflices Corporation 33 WIST MAIN STRUT WnSTIR.N.Y. MSM (716)t72-4S03 \ •^ \ :\ V 4^ ^ ^/"^^ ■-«* A* 4^Z<^ CIHM/ICMH Microfiche Series. CIHIVI/ICIVIH Collection de microfiches. Canadian Institute for Historicai IMicroreproductions / Institut Canadian da microreproductions hittoriquat \ \ T«chnical and Bibliographic Notaa/Notas tachniquaa at bibliographiquaa Tha Inatituta haa attamptad to obtain tha baat original copy availabia for filming. Faaturaa of thia copy wvhicti may ba bibliographically uniqua. •vhieh may altar any of tha imagaa in tha raproduction. or which may significantly changa tha uaual mathod of filming, ara chackad balow. □ Coiourad covara/ Couvartura da coulaur pn Covars damagad/ Couvartura andommagia □ Covars rastorad and/or laminatad/ Couvartura raatauria at/ou pallicuiAa □ Covar titia missing/ La titra da couvartura manqua r~n Coiourad maps«' a n D Cartaa gtegraphiquaa ti coulaur Coiourad inic (i.a. othar than blua or black)/ Encra da coulaur (i.a. autra qua blaua ou noira) Coiourad plataa and/or iliuatrations/ I I Planehaa at/ou illuatrationa an coulaur Bound with othar matarial/ RaliA avae d'autraa documants Tight binding may cauaa shadows or diatortion along intarior margin/ La f liura sarrAa paut cauaar da I'ombra ou da la distorsion la kMig da la marga intiriaura Blank laavas addad during rastoration may appaar within tha taxt. Whanavar possibia, thasa hava baan omittad from filming/ II sa paut qua cartainaa pagas blanches ajoutAas lors d'una raatauration apparaiasant dana la taxta. maia. lorsqua cala Atait possibia. caa pagaa n'ont paa *ti filmiaa. L'Inatitut a microfilm^ la maiiiaur axamplaira qu'il lui a it* poasibia da sa procurer. Las details da cat axamplaira qui sont paut-Atra uniquas du point da vua bibliographiqua, qui pauvant modifier una image reproduite. ou qui pauvant exiger une modification dans la m^thoda normaia dc filmaga sont indiqute d-dessous. □ Coloured pagaa/ Pagae da coulaur □ Pagaa damaged/ Pagaa andommagiaa □ Pages restored and/or laminated/ Pages reataurias at/ou peiliculies Pages discoloured, stained or foxed/ Pages dicolories, tacheties ou piqutes Pagaa dttachies Showthroughy Transparence Quality of prir Quality inAgale de I'impression Includes supplementary matarii Comprend du metiriel supplimentaira Only edition available/ Seuie Mition disponibie [~n Pages detached/ r7| Showthrough/ n~] Quality of print variea/ r~n Includea supplementary materiel/ |~n Only edition available/ n Pages wholly or partially obscured by errata slips, tissues, etc., have been refilmed to ensure the best possible image/ Lea pages totalement ou partiellement obscurcies par un feuiilet d'errata, une pelure. etc., ont iti filmAes k nouveau de fapon i obtenir la meilleure imege possible. D Additional comments:/ Commentaires suppiimantairas: Paginstion as follows : [163] - [193] p. This item is filmed at the reduction ratio checked below/ Ce document eat film* au taux de rMuction indiqui ci-deaaous. 10X 14X itx ax 26X 30X y 12X 16X aox 24X 28X 32X Th« eopy filmed h&n hM b««n r«produe«d ttianks to tlM gwMrosity of: Ubrary of PwtianMnt Mid tiM National LNMrary of Canada. Tho imogoo oppooil m horo aro tiM boot quality poaaibla co w l dorlng tho oondMon and lofibMty of tho orl^nal eopy ond in liooping with tho filming contraot apooHleatlono. L'oxomplairo fllmA fut roproduit grioo A la gdndrooltA do: La Bibliothiqaa du Pariamant at la Bibliothkiua national du Canada. aulwantaa ont 4id roprodultaa avoe to aoin, eompto tami do lo oondMon at ' do I'osompMro fUmd. ot it avoe loo eondltione du eontrat do oo 10 notioto Original coplo o in printod bofpnning with tho front eovor and ending on tho ieet pogo with a printod or ilHietrated impree- alon, or tho book eovor when a pproprl o to. All other orlginol eopiee aro fRmod beginning on tho firat pogo with a printed or Hhietrated i m pree alon. and ending en tho loot pogo %irith e printed or illuetretod Impreeeio n . parlo dom plot. !■ (MrnWrO origlnaux dont le eouverture en imprimoo aont fHmee en eommon^ent plat ot en tomdnent aolt per la page qui e o mporto uno em p rointo ou dlNuetration. aolt per le aeoond io eoe. Toue lee eutree OKompleiree aont fNmde en eommon^ent per le pogo qui eomporto uno emprointo ou dlNuetration ot en termlnent per pogo qui eomporto uno toUo Tho loot recorded frame on aooh ahoN eentain tho aymbol ^»( TINUIO"). or the aymbol ▼ d "COW- INO"). Un le aymbolo ▼ auiventa apporaltra aur la da cheque mieroflcho. aolon lo »• aignmo "A SUIVRE", lo aigniflo "RN". MopOt ptotoOt ehenOt etc.> may bo different reduction retloo. Thoeo too entirely included in one oupoeuro beginning in tho uppof loft hond ngnf ana lop id eononi. ■■ many required, ino following dtegromo lorgotobo nod loft to !• ploneheOt tableouji< etc.. pouvent etre fHmde A doo taux da rdduetion d i f fd rentt. Lorequo lo document eet trap grand pour Atre roproduit en un aoul oMchd. II eet fNmd A pertir do Tangle a upd ri our gaudio. do gouoho * draito, ot do hout en boe« en prenawit le nombre dlmogoe nAceeaalra. Lee diogrammee auiventa IHuotront la methodo. 1 2 3 32X 1 2 3 4 5 6 n BULLETIN OF THE QEOLOGICAL SOCIETY OF AMERICA Vol. 1, pp. 163-174; 176-194 NOTE ON THE PRE-PALEOZOIC SURFACE OF THE ARCHEAN TERRANE8 OF CANADA THE INTERNAL RELATIONS AND TAXONOMY OF THE ARCHEAN OF CENTRAL CANADA BT ANDREW C. LAWSON WASHINGTON PUBLISHED BY THE SOCIETY Mabch, 1890 'f ?*■• / NC Intrc The: ( C The] C Rovie Geneii S Discus Sin andn of Noj glacia basis. niulga tary h ice loa as sout decoiu) terrane t.'b-t.aiii^ia^! BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA Vol. 1, pp. 163-174 March 12, i89o y NOTE ON THE PRE-PALEOZOIC SURFACE OF THE ARCHEAN TERRANES OF CANADA. BY ANDREW C. LAWSON, PH. D. {Read before the Society December 27, 1889.) CONTENTS. Pftge Introductory Remarks ^ 168 The Phenomena in Central Canada 164 Contacts between the Animikie and the Archean 164 Contacts between the Nipigon and Older Rocks 166 The Phenomena in Eastern Canada 167 Contacts between the Paleozoic and the Archean 167 Review of the Evidence. 169 General Considerations 169 . Former Extension of the Paleozoic 169 Transgressions and Oscillations in Level .__ 171 The Erosion of the Archean 172 Source of Paleozoic Sediments 172 Discussion 173 Introductory Remarks. Since the establishment of the glacial theory the cause of the hummocky and rtc/ies moutonnees character of the rocky surface of the Archean terranes of North America has generally been ascribed to the action of the ice of the glacial epoch. Two opinions have been prevalent, having this belief as their basis. The first and older view was, in accordance with the theories pro- mulgated by the Scotch geologists, that the hummocks and their complemen- tary hollows were produced by the direct plowing or gouging action of glacier ice loaded with rock debris. The second and more modern view is, that just as south of the terminal moraine we find the crystalline rocks extensively decomposed in situ, so prior to the advent of the glacial epoch the Archean terranes of the north were similarly decomposed, and the present hummocky XXII— Bull. Geol. 80c. Am., Voi,. 1, 1889. (168) 104 A. V. LAWSOX — THE PRE-PALKOZOir STTRPAf'E. surface represents the locus to wliicli rock decay hiul exten• is of the same uneven cliaracter as that of the uncovered, glaciated country to the north. Similar contacts may be seen inland a short distance, near the head of the lake ; and on Gunflint river the Laurentian gneiss, in low roches moutoniues, ai)[)ear8 jjartially encircled by the Animikie rocks. On the north side of North lake there flows in a creek at the bottom of a deep gorge, which cuts down through 200 feet of flat Animikie strata to the basement of Laurentian gneiss upon which they rest; and the basement is distinctly roches rnoutonnres. Similar conditions are observable two miles up the creek which flows into the east end of North lake, and on Sand lake, where escarpments of Animikie strata overlook and appear to overlie a hum- mocky surface of Laurentian gneiss. The same is true of the escarpments in the vicinity of Little Gull lake. To the north and northeast of Little Gull lake is a group of five small steep-sided, flat-topped hills, known as the Outpost hills, which are outliers of the Animikie, capped as usual with a sheet of columnar trap. The dis- tance which separates them from the main area of these rocks varies from one to four miles. This space is occupied by a very hummocky and roches moutonnees stretch of Laurentian gneiss which maintains the general level of a line extending from the base of the Animikie on the face of the escarp- ment to the base of the same series, where it rests on the Laurentian at the foot of the Outpost hills. The writer has been over the ground between the escarpment and the hills ; and Mr. E. D. Ingall, of the Geological Survey of Canada, who has examined the hills carefully, informs the writer that the actual base of the Animikie may be distinctly observed resting upon the uneven, hummocky Laurentian surface, the sections being perfectly ex- posed. Less than half a mile above Kakabeka falls small outlying patches of the basal beds of the Animikie may be seen lying in the hollows of the mam- millated surface of the Laurentian, and the latter, as it rises from beneath the Animikie, above the falls, is exceedingly hummocky. Along the Dawson road, a few miles back of Port Arthur, low, rounded domes of Laurentian gneiss appear in the midst of the Animikie, projecting above the level of the local upper beds. On Current river the Laurentian rises in hummocky hills from beneath the Animikie slates and traps, although the actual contact has not been observed. Between this and McLean's siding, seven miles east of Port Arthur on the Canadian Pacific railway, the Archean rises in the same hummocky hills from beneath the Animikie, the line of contact being concealed by a narrow strip of swamp. At the siding the contact is only concealed by the width of the road-bed, and the surface of the Laurentian gneiss is seen to plunge down under the flat Animikie rocks with the slope of a steep dome, appear- ing again in a less prominent but still hummocky outcrop close to the con- 100 A. C. LAWSON — TIIK I'UK-rALKUZUlC .SURFACE. tact of the Aniinikie, on the wagon trail about midway between Green point and Wihl Goose point. At Silver harbor, farther up the north side of Thunder bay, there is a strip of the Animikie consisting of lo to 20 feet of flat slates and cherty beds, capped 'y 50 feet of trap, from beneath which on the north, across a narrow strip of swamp, rises the Archean surface in well-defined roches moutounees. Contacts between the Nipigon and Older Rocks. — In the viciuity of Loon lake the basal beds of the Nij)igou series overlapJjie northern edge of the Animikie and rest in undisturbed attitudes directly upon the Laurentian. On the north side of Loon lake and eastward to the vicinity of Pearl river the Laurentian rises from beneath the Nipigon sandstones and conglomerates in prominent huramocky hills. These conglomerates are made up very largely of boulders and rounded pebbles of the Laurentian, which are in- distinguishable in general aspect from the more rounded erratics in the glacial drift. In the bed of the creek at the tank of Pearl river station a low, rounded hummock of Laurentian gneiss appiears from beneath the Nipigon sandstones ; and at the first rock cut east of the station, 200 or 300 yards distant, the sandstones may be seen in the vertical section of the cutting, resting upon the slope of a hummock of Archean schists and dipping away from it to the east at an angle of 15°. Here the schists are rotted in places, leaving a few harder nuclei or boulders of disintegration in situ. Half a mile farther east along the track prominent, lumpy knobs of Laurentian rise above the level of the Nipigon sandstones to a height of over 100 feet, and on the east side of these, in a rock cut of the railway, the sandstones may be seen repos- ing directly upon their slopes, as an outlying patch. About ten miles east of Nipigon, on the Canadian Pacific railway, there is a prominent bluff of Nipigon sandstone, capped with a thick sheet of verti- cally columnar trap, the whole presenting escarped faces which rise on three sides precipitously for several hundred feet above the hummocky plain of Laurentian rocks upon which it rests. The bare Laurentian basement is traceable up to the talus at the bases of the cliffs, and presents the appear- ance of passing under the colunni of superincumbent strata in the same hummocky condition as that which it has beyond the cliffs. About ten miles east of Mazokama station on the Canadian Pacific rail- way a prominent point runs out into the lake, the core of which consists of hununocky Laurentian gneiss, and the outer margin or shore of superim- posed Nipigon sandstones and conglomerates. Here again the Laurentian appears to pass under the Nipigon with its characteristically hummocky sur- face, the country being well bared ; and that it does so is proved beyond question by the fact that scattered over the Laurentian area, away from the edge of the Nipigon rocks, there are numerous outlying patches of the basal 4 AUCHEAN .SUUKACI-: (JNCIIANMiKlJ HINCK TJIK NIPIUON. 107 4^ beds of the Nipigon resting in situ in the hollows between the Laurcntian hummocks, both at the bottoms of the hollows and on the steep slo])e8. These patches are usually not more than a few chains in diameter ; and their relation to the Laurentian affords incontestable proof that the surface of the latter has undergone no material change siiiee they were deposited upon it. At Rossport the Aniniikie rocks come in sij^ain between the Archean and the Nipigon, and here also may be seen, near the railway station, in a hollow between the Laurentian hillocks, an outlying patch of the basal beds of these rocks. Along the shore of the lake between liossport atid Black river, north of the Slate islands, there are occasional patches of the Nipigon araygdaloidal traps which have escaped removal by erosive agencies, and these all repose upon a hjmmocky Archean surface. In none of these instances is there any evidence of a perceptible reduction of the mean level of the glaciated surface of the Archean below that upon which the Nipigon or Aniniikie rocks rest. A noteworthy fact also is, that with one exception none of the Archean rocks, where they pass immediately beneath the Animikie or Nipi- gon, show the slightest evidence of decay. On the contrary, they are remarkably fresh and free from even the incipient decomposition of weather- ing. The exception is the case of the schists in the rock cut east of Pearl river mentioned above. All the Laurentian gneisses and granites are per- fectly fresh in their macroscopic aspects. Another interesting point, which will be alluded to again, is the transgression northward of the newer Nipigon rocks beyond the edge of the older Animikie. The Phenomena in Eastern Canada. " v On instituting a comparative inquiry into the conditions which obtain along the escarped line of the abutment of the undisturbed Paleozoic upon the Archean in eastern Canada, it is found that the evidence here confirms the conclusions arrived at on Lake Superior as to the general character of the pre-Paleozoic Archean surface. Contacts betiveen the Paleozoic and the Archean. — Laflamme in his " report of geological observations in the Saguenay region " * seems to have arrived at nmch the same conclusion as the writer. After describing a new area of the Trenton rocks in the vicinity of the Saguenay " which rest directly on the gneiss," and stating that " their thickness is so slight, at least on the border of the formation, that the undulations of the gneiss are. brought to light through their edge," he gives an account of various outliers and says by way of summary : " I have pointed out in the course of these remarks the fact that limestones (Trenton) are often found in nests or outliers amongst the *Geol. Survey of Can»da, Report Progress for 1882-3-4, Part D. 108 A. (!. I-ANVSO.N — TIIK I'KK-I'ALKOZOIC HIKFACK. ^raiiitcH. Tlieroforc, thcHC depresHions aixl iiillH of riaurentian imiat nccc8- Harily have uxiHted at the bottom uf the I'alcoxoic ocean when the linieutone beds were being depoHited."''* Mr. A. r. Low, of the Geoh)jfical Kurvey of (*anada, who hau been more recently eiij^agcd in tracing ont the northern limitH of the Pahur/oic on the north Hide of the St. Jjawrence, west of {^nel)ec city, informs the writer that at Hcveral phiccH he has noted the HnperpoHition of the Trento«i or Lorraine beds directly upon the hummocky Jjaurentian surface, and that there has been no redncticm of the surface where it projects from beneath the escarp- ments, below that where the tlat strata rest upon it. He notes the following localities as aHbrding particularly good sections : — lietween Loretto village and St. Ambrose railway station, (^. L. St. J. railway ; west of Belair station, C. P. railway; Pont Kouge station, C. P. railway (section on Jacques Cartier river); Deschambault, near railway station. Mr. Low also informs the writer that the undisturbed limestones of Lake Mistassini, in southern Lab- ratlor, may be observed to rest upon hununocky Laurentian surfaces; and that on the l^ast-main coast of Hudson's oay similar flat lying strata may be seen in the transverse section aftbrded by Richmond gulf, resting on a very hummocky surface. In eastern Ontario, the best evidence we have bearing on this (juestion is contained on Mr. E. Coste's " Geological and Topographical Map of the Madoc and Marmora Mining District," recently published by the Geological Survey of Canada. No report accompanies the map as yet, but the writer has had the benefit of frequent conversations with Messrs. Coste, Ami, and White, who were employed in the field-work necessary for its construction. From the map and from the information thus supplied, it is clear that in the area mapped we have a remarkably striking illustration of the superposition of flat, undisturbed Paleozoic strata (Birdseye and Blac'i River) upon a very hummocky and mammuillated Archean surface. The northern border of the Paleozoic is here very irregular in outline, and beyond the limit of the main area there are very nunicrous outliers scattered over the country. Both along the edge of the escarpment and at the perii)hery of many of the outliers, the flat strata may be seen resting directly on the rounded hummocks; and these, out beyond the escarpment, often rise high above the lower horizontal strata. Many of the outliers, also, are mere patches resting in situ upon the steep slopes of these hummocks. Many are but a few chains in diameter, and others only a few yards. Further, there may be repeatedly seen projecting through the upper surface of the Birdseye and Black River formations rounded knobs of the Archean, in the shape of in- liers well within the Paleozoic area. These are clearly the crests of partially * Loc. cit., p. 15. 4^ , THK l'«)I'\r>ATlON' FOIt THK WW.Eo'/JiU'. Kin iiiicovenMl huininocks; and the plu'iionienon is ho (ioiiinion us to leave no (loiiht a8 to the character of the undcrlyiiig Hurf'ace. Ukvikw of tiif Evidrnck. Thu8, wherever careful observations have been made as to the nature of the superposition of the undisturbed Paleozoic rocks upon the Archean, whether in the Lake Superior country, eastern Ontario, (Quebec, or Labrador, the evi\ — THK l'l!i:-l'AI,KOZo|C SIIU'ACK. deposition of the Niagara than in earlier epochs. It would follow from these considerations, that as Paleozoic time advanced from Cambrian to late Silurian or Devonian there was a gradual and progressive subsidence of this portion of the continent. As we have no evidence of the deposition of post- Devonian formations anywhere over the Archean " nucleus " till we come down to post-Tertiary, it may be tentatively inferred that after the Devonian it was again elevated, and this elevation proBably only reached its maximum during the glacial epoch, affording the conditions of altitude contended for by nmny writers to explain the great precipitation of snow. In post-glacial tinjes we know from the distribution of such formations as the Leda day and Saxieava sand that the northern part of the continent was again par- tially submerged for several hundred feet, from which depression it has since recovered ; we thus have evidence of a slow vertical pulsation of the surface of this part of the continent, of which there have been at least four great beats since early Cambrian times. But this is a digression, and the argument which has led to these remarks was inaugurated to show simply that the surface of the Archean " nucleus " was once very extensively if not wholly covered by Paleozoic sediments. This covering probably accounts in a large measure for the remarkable ])reservation of the Archean surface in the condition in which pre-Paleozoic denudation left it. There are other considerations which help us to under- stand this preservation, such as the levelness of the plateau and its compari- tively low altitude, combined with the very resistant character of most of its rocks, which appear to be little susceptible to that erosive or corrasive action of streams which is so effective in removing the more yielding strata of post-Archean age. These considerations will not, however, be entered upon here. The Erosion of the Archean. — One is constantly impressed by the perfectly appalling amount of denudation to which the Archean has been subjected in order to truncate its formations down to the surface which it presents to-day. And when we reflect, as a result of the conclusions here arrived at, that this denudation was practically completed before the beginning of earliest Paleozoic times, and has not been, as commonly supposed, the result of later agencies, there looms up a conception of the pre-Paleozoic interval necessary for such denudation which staggers even the most stalwart geological imagination. To say that it must have been comparable with all the time which has succeeded from the earliest Cambrian to the present seems but a feeble way of expressing it. /Source of Paleozoic Sediments. — The conce})tion of a covering of Paleozoic strata over the surface of the Archean '* nucleus," which probably endured into comparatively recent geological times, enables us to a large extent to understand the preservation of the pre-Paleozoic surface, but it also raises the SOIIJCK OF' PA rj<;( )/<>[(• SKDIMKNTS. 1—0 f important question of the source of the sediments composing those strata. If such a wide-spread formation as the rocks of Niagara age was deposited over the surface of the Archean " nucleus," as well as over the regions which encircle it, it is clear that the Archean " nucleus " could not have been the source of supply of those sediments. Some other portion of the continent, or some other region now submerged, must have constituted the dry land of that time. Where that region lies is a question yet to be answered. DISCUSSION. Professor J. \V. Spencer : The facts set forth in this very interesting paper by Dr. Lawson have their counterparts in the geological structure of the South. The hummocky and rounded rock surfaces have always had an interest for me, on account of their common occurrence in regions which have been glaciated, and hence regarded by many as evidence of glacial erosion. But in the paper of Dr. Lawson we learn that such surfaces existed before the formation of the early Paleozoic terranes. Some of you may be familiar with Stone Mountain, about fifteen miles from Atlanta, Georgia. This is a rounded granite hummock of over a mile, in longer diameter, rising 700 feet above the plain. The rock is remarkably free from joints, and is rarely traversed by even an insignificant vein. Thus its structure has been favorable to the preservation of the rounded form, whose outline is as perfect as any of the domes of glaciated Norway or Canada ; or of southeastern Missouri, which lies outside of former glacial action. Stone Mountain rises from beneath very much disturbed strata of gneiss, whose beds dip to the southeast, and there is no gradation of any importance between the granite and the gneiss. The gneiss is decayed to a depth, in some places, of at least sixty feet ; but the granite is compact, without being weakened by even incipient decay. The surface materials, as fast as decomposed, are washed off by the rains. Thus the contrast between the two formations of rocks is preserved. This Stone Mountain is only one of many in Georgia and Ala- bama. Here, then, we have, in the South, pre-Paleozoic surfaces as old as or older than those described by Dr. Lawson in the Lake Superior region, and brought to light by simple atmospheric action. Along the Potomac river we find hummocks being formed by the progress of atmospheric in- vasion along lines of joints, but these are now in process of formation, and do not represent so ancient surfaces as those of the granite hummocks of the South. BULLETIN OP THE GEOLOGICAL SOCIETY OF AMERICA Vol. 1, pp. 175-194 March 20, i890 THE INTERNAL RELATIONS AND TAXONOMY OF THE ARCHEAN OF CENTRAL CANADA. BY ANDREW C. LAWSON, PH. D. (Read before the Society December '2S, \SS9.) CONTENTS. Page. Primary Separation of the Archean into two Divisions 176 The Upper Division 176 Nomenclature ]7<» Petrographical Description 177 Original Characters and Metamorphism 180 Relations between the two Divisions 181 The General Relations 181 Irruptive Contact on L" -i of the Woods 182 Irruptive Contact in Rainy Lake Region 18.S ■ Significance of Relationship 186 Principles of Classification ... 180 Principles applicable to the Upper Division 18f! Principles applicable to the Lower Division 186 Diflxjrent Generations of Laurentian Rocks 187 Other Conditions considered 187 Similar Observations elsewhere 188 Geognostical Equivalents of the Archean 190 The Argument from Analogy 193 Primary Separation of the Archean into two Divisions. Throughout North America, geologists have long recognized in the great fundamental complex of rocks, known generally to-day as the Archean, a natural division into two well-characterized portions, related to each other in space as upper and lower. The lower division is commonly known as the Laurentian, and consists for the most part of an assemblage of rocks of the character of granites, syenites, diorites, and gabbros in mineralogical composition, but more or less foliated or gneissic. Involved with these in a way not hitherto understood there are also, in some regions, portions of 17(5 A. C. [;A\VS(>\ — UKr.ATFoNS OK TIIK AI.MIflCAN ol' CANADA. various j^neisH, schist, limestone, (jiiartzite, and c, p. 62. l.s-j A. C. I,ANVS(»N — m;i,ATInN,s n\ Till; AIM IIKAN U|' ( ANADA. confornmble u|><)ii tlio Liuirontiiiii ; the nsHUinption hein^ al ways that both UHHcinblageH of roeka wore coinixised of inetutnor|)h()He(l Hey.sti;m. ls:5 ^ later foriiiHtioii than the schintM of the Keuwutin Hcriui^, and breaking through them. Irniptlvn Con tact In Jiuiny Luke Region. — The studies here inaugurate*! ahout Lake of the Woods have since been continue.t!) of Professor N. H. Winchell's report. On the maps accompanying the report, however, it is distinguished clearly by a color and named the " Vermilion series," although here including formations that had earlier been designated Keewatin. From this it would appear that the term " Coutchiching" was somewhat prior to " Vermilion," and was more fully and precisely defined as to its geological sig- nificance. Moreover, the term "Vermilion Ijake series" was used earlier by Irving in another sense than that proposed by Professor N. II. Winchell, and in the same Annual Report (Fifteenth) the terms "Vermilion series" and "Vermilion system" are used by Professor A. Winchell, on pp. 192, 1U5, 100, in another and much more comprehensive, but still undefined, sense. ) i 184 A. C. I.AWSON — KKLATJONS Ol" TIIK AUCllKAX UF CANADA. tinuous, forming a great anastomosing mesh-work, the general strike being always concave to the Laurentian areas which they encircle. Sometimes, however, where denudation has exposed their deeper portions along anticlinal or synclinal ares, as in parts of the Lake of the AVoods and Kainy lake regions, and better in Hunter's island, the formations in contact with the Laurentian granite-gneiss are finind to be excessively shattered, and countless numbers of fragments are strewn throughout the nuiss of the ir- ruptive rocks. The country is well bared, and what is stated is clearly visi- ble on well-exposed continuous rock surfaces. These included detached fragments of the formations overlying the granite-gneiss range in size from pieces a few inches across to immense masses. Their longest diameters are, as would be expected, in the plane of schistosity. Where the enclosing rock is gneissic, the inclusions have usually a constant orientation parallel to the foliation of the gneiss, which also coincides, as a rule, with the nearest edge of the belt through which it breaks, where not too remote from the edge. Other inclusions in the Laurentian have been observed whose derivation from the Ontarian rocks cannot be established. Suggestions as to their origin have been thrown out by the writer in his report on the Rainy lake region. Along the edges of the belts of the Ontarian rocks, there may frequently be observed, running out from the niain belt and in continuous strike with it, tongues of schist which taper more or less gradually and eventually end in ])oints. These also are seen to be immersed and congealed in the granite- gneiss ; and numy of the larger detached inclusions are doubtless portions of such tongues which have been septuated from the main belt by the low- ering of the plane of surface truncation by denudation, rather than by actual detachment at the time of disturbance. This would in a large measure account for the fact that the common orientation of the lax'ger fragments, and their parallelism with the edge of the belt, holds for the dip as well as the strike. Numerous long, attenuated, [)arali«!l tongues arc also formed at the edges of the schist belts by the injection along the planes of schistosity of portions of the granite-gneiss magma, forming an evei.ly ribboned alternation which simulates bedding. Its formation by injection is, however, sufficiently ai)[)arent. Similar ribboned alternations are described and figured by Bar- rois* as occurring at the edge of the Cambrian schists of Brittany, where pierced by irruptive granites. The detached inclusions are, also, not in- fre(iuently ribboned, parallel to the schist planes, with apophyses from the main area of the enclosing granite-gneiss. If, at the base of the Ontarian system, wv had bedded rocks which on metamorphisni gave rise to crystalline limestones, (juart/ites, etc., we would * Hull. Soe. (.ieol. de France, ;ttne Serie, t. Xl\', 188ii, p. 833. rossinMo oRiriiN or pskido p.kddino. isr, edges have these involved with the Laureiitian gneiss, just as the hornblende schists and mica schists are, and intercalations would be produced which would, as in the case of the schists, frequently simulate interbedding of quartzite or limestone, as the case might be, with the gneiss. The deception would, of course, be intensified by subsequent further defornuition of the crust by pressure so as to be practically beyond detection, if the clue were not followed up from a starting point where such subsequent dynamic agencies have not obscured the true relationship. This, the writer is per- suaded, is the explanation of many of the intimate associations of gneiss and (|uartzite or limestone, whereby rocks really metamorphic sediments are so involved and welded with rocks of plutonic irruptive origin that they have been taken together as a simple secpience of deposited strata. In some portions of the Laurentian country, which the attitude of the flanking rocks indicates was once arched over by an anticlinal dome of the latter, there .are found patches of schist lying quite flat, or nearly so, upon the granite, showing, in favorable clirt" sections, a brecciated or intrusive contact on the under side. These reinnants seem to show that the anticlinal dome was flat or very lowly rounded, and that only on the flanks of the Laurentian boss did the strata composing the arch i)lunge down at high angles. Significance of Relatlomhlp. — Bearing in mind the essential distinctions which exist between the rock formations of the Ontarlan and Laurentian systems, both as to their lithological character and their mode of occurrence^ and remembering also their relative geographical distribution, the foregoing statement of the relationship which obtains between the two systems leads clearly and unavoidably to this conclusion, viz., that the formations of the Ontarian system at one time rested, as a volume of hard rocks, upon a magma which subsequently crystallized as the Laurentian granite-gneiss ; so that the present line of demarkation between the two systtaus must be regarded as representing the trace of what was once a plane of contact between the then crust and tiie magma upon which it floated. This conclusion aflf'ords us a conception of the Archean which is ideal in its simplicity and which gives us the key to the raveling of the mystery in which the subject has been involved. Tlie fact that the crust, which con- stitutes what we now call the Ontarian system, was crumjjled while it floated on the magma ; the fact that its lower portions were shattered by disturbance so that the magma penetrated the fissures and enclosed detached fragments ; the fact that there were currents in the magma which arranged the inclusions in streams and also i)roduced the foliation of the gneiss ; the fact of contact metamorphism — all these are incidental and concomitant circumstances of the great essential condition ol'a crust resting on a magma. JJut from the nature of the rocks of the Ontarian system it is clear that .i ISO A. ('. T,A\VSON — 1;KF-ATI0\S or TMK Al.TIIKAX OF CAXADA. 1 f they could not have been deposited upon a magnui. There must have been a firm crust presenting a floor uj)on which they were laid down. That floor, together with portions of the system of rocks which lay piled upon it, has disappeared. That it has sunk down to a zone of fusion and become ab- sorbed by liquefaction in a sub-crustal magma, which later crystallized out as the Laurentian, is the only explanation that is open to us. It follows also that the Laurentian rocks are younger than those of the Ontarian sys- tem, as has been before indicated. PRiNrrpLEs OF Classfficatton. I'!- The bearing of the facts and conclusions recorded above upon the tax- onomy of the Archean is apparent. The argument establishes this cardinal princii)le in the classification of that great complex of rocks, viz., that its primary subdivision depends upon a distinction of cosmical importance be- tween an older assemblage of altered normal surface- formed strata and a younger assemblage of rocks resulting from the crystallization of a sub- crustal magma. Principles applicable to the Upper Division. — To the upper or Ontarian system the ordinary stratigraphical methods of classification are applicable. The system separates stratigraphically into two great groups. The lower and older, consisting of strata free from volcanic admixtures, so far as has been observed, is the Coutchiching. It resembles in its lithological charac- ters and in its position the Montalban of Hitchcock. The upper group, consisting of rocks which are dominantly volcanic in composition, is the Keewatin. It rests upon the Coutchiching in probable unconformity, the l)eginning of the period in which these rocks were deposited being signalized by the advent of a widespread and continued volcanic activity. This group falls into line with the Green Mountain series in the position assigned to it by Hitchcock. Other groups may quite possibly be discovered which will swell the volume of the Ontarian system. Principles applicable to the Lower Division. — In the Laurentian the ordi- nary stratigraphical principles of classification do not apply, since there are no strata properly so called ; and we must seek for a principle appropriate to an assenil)lage of rocks essentially different in their development and mode of occurrence from all those of the stratigraphical column. The Lau- rentian is not homogeneous throughout its surface distribution. It is com- l)osed of different members or masses, which, while they present wonderfully constant general characters within themselves, are distinct from one another lithologically. A study of the relationship between the masses thus differen- tiated in space leads us to the chief moment of all geological classification, namely, their differentiation in time ; and we have to consider the possibility SUI'.DI VISIONS OF TIIH LAURENTIAN SYSTF-M. 187 )^ of different generations of Laurentian rocks. This possibility presents itself as soon as we familiarize ourselves with the sub-crustal igneous and later formations of the Laurentian. Different Generations of Laurentian Bocks. — To the writer this conception of different generations has never been more than a possibility till the present year. In his report on the Rainy lake region, two broadly distinct mem- bers of the Laurentian were distinguished, lithologically and on account of their systematic relative distribution, as the " peripheral zone" and " inner nucleus " of the Stanjikoming area, the former being composed chiefly of hornblende-granite and syenite-gneiss, and the latter of very quartzose biotite-gneiss. The relationship in time between these two rock masses re- mained indeterminate. During the past summer, however, he has been able to establish, in the Hunter's island region, chronologically distinct genera- tions of Laurentian gneisses. In that region there are two broadly distinct members of the Laurentian, analogous petrographically and in relative dis- tribution to those of the Stanjikoming area. Belov the Keewatin rocks there is a great mass of hornblende-granite-gneiss, which presents an irrup- tive or intrusive contact against them. Towards the central part of Hun- ter's island this hornblende-granite-gneiss is pierced by an enormous irrup- tion of biotite-grauite, which is sometimes very distinctly gneissic and sometimes quite undifferentiated in structure. In texture it varies from fine-grained, almost micro-granitic, to a moderately coarse granite. This biotite-grauite-gneiss traverses the horublende-grauite-gneiss in innumerable clearly defined dikes cutting it in all directions, and holds innumerable in- cluded blocks of the same rock. It comes up from beneath the hornblende- granite-gneiss, and is unquestionably of later age. Thus we have in this area at least two distinct generations of Laurentian rocks, both the result of the crystallization of a sub-crustal magma. At the time of the second generation the rocks of the first generation constituted the lower portion of the crust. It is upon the recognition of facts of this order that an intelligible and profitable classification of the Laurentian rock masces and the geological events which they represent must be established. Other Conditions considered. — The relationship which has been found to obtain between the upper and lower Archean leads, as has been said, to a conception which is at once grand and simple. So long as we confine our- selves to regions like that northwest of Lake Superior, where no great com- plications have been introduced by post-Archean crust-crumpling agencies, it affords a full explanation of all the phenomena of Archean geology. There is a possible simpler case which would still present the essential conditions of the relationship in question ; i. e., the case in which the sub- XXV— Bum,. Gf.oi.. Soc. Am., Vol. 1, 1889. 188 A. C. LAWSON — KRr-ATIO\S OF TIIK AUrFIKAN OP f'ANADA. crustal magma might be irrupted within the overlying crustal rocks without the intense folding of the latter. Here we should expect to find a less pro- nounced alteration, due only to the proximity of the magma, and an absence of those phases of metamorphism which accompany the rock shearing, crush- ing, and stretching due to dynamic ageinries. In the common case, where the upper crustal rocks are folded, varying phenomena would also be ob- served according as the folding took place before the fusion which produced the magma immediately beneath the crust or while the latter was floating upon the magma. There are also more complicated cases which are doubtless common. These are due to the superimposed action of crust-crumpling, rock-shearing, strata-squeezing forces subsequent to the establishment of the Archean con- ditions in their primal simplicity. These are possibilities which must be borne in mind in attempts to apply the theory here advanced to the Archean in other regions. It is easily conceivable that had the country northwest of Lake Superior been subjected to extensive deformation in post- Archean times, the evidence whereby the irruptive character of the Laureutian has been demonstrated might have been entirely obscured, and the true relation- ship might have remained unsuspected, as appears to have been the case in better known regions. C. Similar Observations elsewhere. In various parts of the world observations have been recorded which show that the phenomena arising from the irruption of a local or general sub- crustal magma through an overlying crust, and the consequent development of a complex of gneissic igneous rocks and nietamorphic strata, are not peculiar to the region studied by the writer. MacFarlane * long ago described and figured good evidence of the irrup- tive character of the Laurentian of the northeast shore of Lake Superior ; but, in accordance with the views of the extreme plutonic school, he regarded the whole complex of intrusive and intruded rocks as the first crust of the earth, and the angular fragments of hornblende schist as earlier separations from the same magma as that which crystallized into the Laurentian granite or syenite-gneiss. Mr. Frank Adams, who has been for some years past engaged in a study of the Laureutian of the Province of Quebec, north of the St. Lawrence, says — " Tlio unexpected tiict was ascertained that the so-called massive and stratified varieties of this rocic [aiiorthosite ; hitherto regarded as upper Laurentian and meta- * Geological Formations of Lake Superioi. Canadian Naturalist, N. S., Vol. Ill, 1867, p. 177. OBSERVATIONS IN EASTERN CANaDA AND EUROPE. 189 7. morphicj are in reality only different portions of one and the same mass. * * * As a re'sult of this summer's work, I thini< it may bo safely concluded that the rocks com- prising the principal area of anorthosito above referred to, us well as most, if not all, of the smaller areas, are of eruptive origin." * He confirms this in his summary for 1888 in the following words : — " All the areas of anorthosite now known to occur in the district have been ex- amined, and mapped, and have proved to bo either eruptive masses cutting through the gneisses, or masses interstratilied with the latter, but probably still of eruptive origin." f Callaway has shown, in his paper on the granitic and schistose rocks of northern Donegal, that the granite-gneisses of that region, which have been regarded as Laurentian and which correspond closely in lithological characters and mode of occurrence with the Laurentian of Canada, are really irruptive through older rocks, which must have arched them over, and present all the evidences of irruption which have been adduced by the writer in support of the irruptive origin of the Laurentian northwest of Lake Superior. He thus states his conclusions : — "1. The granite rock of northern Donegal, originally supposed to be the result of the metamorphism of sedin ents, and recently referred to the Laurentian system, is a true igneous granite, as seen in its intrusion into the adjacent schists, in its inclusions of masses and fragments of other rocks, and in its metaniorphic action on limestone in contact. 2. This granite is distinctly foliated, the gneissic structure being caused by lateral pressure, * * * 3. The granite is intrusive in a thick group of quartz- ites, quartz-schists, hornblendic, micaceous and talcose (?) schists, and crystalline limestones, called the Kilmacrenan series. These rocks are truly crystalline, but usually thin-bedded and fine-grained. 4. The crystalline schists are bounded on the east by a semi-crystalline series, consisting of quartzose grits and itacolumites, quartz- ites, crystalline limestones, compact dolomites, phyllites, interlaminations of grit and schistose matter, and finely foliated micaceous schists." J These conclusions as to the irruptive origin of the gneiss are confirmed by later observations of the same investigator on the Galwuy gneiss. § In the pre-Caiubrian or Arciiean of Brittany, Barrois recognizes the irrup- tive character of the gneisses which correspond to our Laurentian. He says — *' Ces gneiss alternent avec des lits interstratifles de micaschistes et d'amphibolites, et passent h des granites gneissiques qui les penotrent ti la fa9on d'une roche eruptive. L'ensemble des gneiss et micaschistes (/rmutlqaes avec granites gneissiques rappelle par ses caracteres lithologiquos Vetagc ditncticn, propose par M. Hicks, dans le pays de Galles, le gneiss fondamental d'Ecosse, certains gneiss laurentiens du Canada, *Geol. Survey of Canada, Summary Report for 1887 and 1888, 188SJ, p. ZIk. t Ibid., p. 85a. jliuart. Jour. Geol. Soc, Vol. XLI, 1885, p. 239. Itjuart. Jour. Geol. Soc, Vol. XLIII, 1887, p. 517. X 190 A. C. LAW.SON — KKLATIONS OF Tilt; AKCIIEAX OF CANADA. * * *. lis [micaschistos] y alternent iivec des lits subordonnos do gneiss h gmiiis fins, d'amphibulitcs, de chlorito sehistcs, do schistos micuoos, et compronnent des masses interstruti floes de diorites et de granulites, d'origino eruptive. Ces ruclios subordonnees forment uvec les inieiisohistes, diips lesquels elles sont injectees, de longues bandes parallelcs, * * *."* Newtou's description of the geology of the Bhick Hills of Dakota f leaves little room for doubt but that the rocks which he calls Archean correspond to the upper Archean or Ontarian system of central Canada, and that his irruptive granite, though not described as foliated, is the analogue of the commonest phase of the Laurentian. The same relationship holds between the two rock systems in both regions, and many of the Laurentian granites are devoid of foliation. J. Geognostical Equivalents of the Archean. In iissemblages of rocks of indeterminate or post-Archean age complexes of gneissic irruptive rocks and older metamorphic strata of clastic or vol- canic origin are now well known. These cannot be spoken of as the geolog- ical equivalents of the Archean complex on account of their diverse age, but may be referred to as its geognostical equivalents, since their development appears to depend upon universal sub-crustal conditions, which are to a large extent independent of geological age. McMahon,;}: in his studies of the great " central gneiss " formation of the Himalaya mountains, has demonstrated clearly that the formation is not, as was long supposed, the Archean basement upon which the Paleozoic sedi- ments were deposited, but is an irruptive mass breaking up through the Silurian and later rocks, altering them, holding detached fragments of their strata, and being injected within the strata. Speaking of this formation, which he calls gneissose granite, he cites the following evidences in proof of its irruptive origin : 1. The granite has produced a certain amount of con- tact metamorphism on the rocks touching it. 2. Tongues and intrusive veins have been sent from the granite into the adjoining rocks ; in other places the granite appears in sheets between the beds of the sedimentary rocks at some distance from the junction of the latter with the main mass of the granite, and in some cases these sheets or dikes have cut through tlie beds and passed from one horizon to another. 3. The main mass of the granite appears at different geological horizons. § 4. The granite contains '•feii'iicture Oieologique du FinisWre. Bull. Soc. Geol. de France, .3me Serie, t. XIV, 1886, p. 657. t Report on the Geology md Re.Mources of the Black Hills of Dakota. By Henry JJewton and Walter P. .Jenney, Washington, 1880, pp. 4.V8(), 220-2!i2. IGeol. Survey of India, ieconls, Vol. XVIIl, Part 4, 1884, p. 168; ibid., Vol. XVIII, Fart 2,1885, p. 7!t. Geol. Mag., N. S., Lccade III, Vol. IV, 1887, p. 212. ^ Ah it does when it cotncH at one place against the ICeew.itin and at another against the Cout- chiching in the Rainy lake region. OKANITIC IKKll'TlONS UF VAKIOLS AUKS. 191 veins similar to those caused by shrinkage on cooling in granite of admit* tedly eruptive origin, 5. It contains fragments of slates and schists im- bedded in it. He also states that the evidence afforded by the study of thin slices confirms the conclusion arrived at 'jy the stratigraphical evidence, and gives a summary of the microscopic evidence.* The very able and precise descriptions by Barroisf of the various granitic irruptions which have affected Brittany at different ages from the pre-Cam- brian up to the Carboniferous show beyond (luestion that not only in Archean times, but at various subsequent j)eriod8 were the conditions which characterize the Archean of Canada reproduced. He describes par- ticularly the " granite gneissique," demonstrates its irruptive origin, and notes not only the contact metamorphism, but also the injection of these rocks "en filonnets minces et repetes" within the encasing schists. His descriptions and figures of repeated injections of granite within the schists, so as to produce an alternation simulating bedding, closely corresponds with the contact phenomena described by tlie writer as observed between the Laurentian and Keewatin on the Lake of the Woods, the interpretation of which is entirely in accord with that of Barrois, though questioned by Pro- fessor A. Winchell.J It would appear that just as in Hunter's island, north- west of Lake Superior, we have two generations of Laurentian rocks from a sub-crustal magma, so in Brittany there have been several generations of similar rocks breaking through the overlying crust, extending in time as late as the Carboniferous. In Norway Kjerulf 4^ places the " Gebanderte granit, oder gneisgranit " with the eruptive rocks, and states that in numberless places such rocks break through the strata of the gruudgebirges, and also, indeed, through the Bergenschiefer in which Reusch has since found Silurian fossils.] | In the greater part of Norway he says (translated freely) ^j — " What was formerly recognized as gneiss must on the map be now designated as granite. Tlie reasim why the older observers o.scumo it to be gneiss is the granular banded structure, which we must distinguish from the appearance of bedding. On older maps are shown also other great regions in which the dip and s>ike of the beds is given, an attribute which they do not in reality possess; and t'le reason for this lies in the confounding of foliation with beddine;. * * * The rock, according to the old conception, is granite when no bedding occurs in it. The modern view, which had already' been announced by Delesse, says : ' En realite c'est [le gneiss granit] seulement uno variete du granit, qui est voinee et qui parait avoir ete genee dans sa cristallisation.' ■>** *Geol. Mag., loc. cit. t Hull. Soc. Geol. de France, 3me Serie, t. XIV, 1880, pp. fi5,')-8'J8. j(ieol. Sin-vey oj Minnesota, Fifteenth Annual Keport. 1886, p. 201, js 5. gl>ie Geologic des Slid, und Mit. Norwfgen, Honn, 1880, p. '£iT. li Fossil ieii Fulirenden Scliiefer von liergen, Leipi^ig, 188a. 11 Op. cit. p. 282. ** Delesse, Etudes sur le Metamorphism, 1801. V.)2 A. 0. LAWSON — KELATIONS OF THE AKCHEAN OF CANADA. The syenites of the southeast coast of Norway, also, which have been studied particularly by Bnigger, and which are irruptive through fossiliferous Silurian and Devonian strata, are eminently gneissic in places. They are in- distinguishable in this respect from the more distinctly foliated varieties of our Laurentian gneiss. Lehman's masterly work-'- on the rocks of Saxony and other geologically similar regions has clearly established that many of the gneisses of central Europe are irruptive in their origin. The foliated gabbros or gabbro-gneisses of the Lizard are regarded as eruptive by such eminent observers as Teall f and McMahon,| though they differ as to the precise mode of the development of the foliation. Harper § has shown that the "granite and gneissic granite" of Larn, Caernarvonshire, which was formerly held to be Archean, is in reality irrup- tive and of more recent age than the Upper Arenig strata : " The actual contact of the two rocks is easily found, and the granite is seen to send out little tongues between the laminio of the shale. Specimens of the latter rock, indurated and lirnily adhering to the granite, may be obtained. * * ^ The shale is clearly altered and exhibits little spots and nodules supposed to nfpresent the in- cipient development of chiastolite. Another quarry, well within the boundary of the granite, shows entangled masses of baked shales." In a paper submitted to the International Geological Congress at its Lon- don session || in 1888, the writer quoted Dr. G. M. Dawson ^[ at some length to show how entirely the conditions which obtain between the Triassic rocks of the west coast and the younger subjacent irruptive granite are analogous to those which obtain between the rocks of the upper Archean or Ontarian system and the Laurentian granite gneiss. Dr. Dawson's account of the history of geological events in that region in post-Triassic times confirms the correctness of the writer's interpretation of the Archean of central Canada. The interesting geognostical equivalent of the Archean on the Pacific coast is paralleled on the Atlantic coast by the great irruption of "gneissic granites" which in post-Cambrian times, possibly as late as the Devonian, have broken up through the Cambrian slates and quartzites.** These " gneissic granites " are indistinguisable from many of the Laurentian gneisses. ♦Untersuchungen iiberdie Entstehunt; der altkrystallinischen Schiefergeateine, Bonn, 1881. t Origin of Certain Handed Gneisses; Geol. Mag., N. S., Decade III, Vol. TV, 1887, p. 484. fOn the Foliation of the Lizard (Tabbro; ibid., p. 74. jiCiuart. Jour. Geol. Soc, Vol. XXX IV, 1878, p. 442. II Ktude.s siir les schistes eristallins, p. (id. if Geol. Survey of Canada, Annual lleport, 1887, Part H, pp. II-IH. **The Lower Cambrian rocks of (iuysborougli and Halifax Counties, N. S. By E. R. Faribault; (ieol. Survey of Canada, Annual Report, 1880, Part P, p. l-M. t.V f lie lis « The Akoumknt fisom Analogy. These references and (jiiotations by no means exhaust the literature of the subject. They are taken mostly from very recent writings, and much to the same effect might be quoted from the older geologists, such as Von Cotta, Neumann, Darwin, Delesse, and others, who have insisted on the ir- ruptive character of gneissic rocks or have regarded gneiss as but a differ- entiated variety of irruptive granite. But enough has been adduced to show that the writer's interpretation of the Archean geology of central Canada, in so far as it depends upon the irruptive nature of the Lauren tian gneisses, is not without the strong support of many analogies. (198)