THE PRFXAMBRIAN ROCKS OF THE CANTON QUADRANGLE 53 



limestone country rock is a fact of equal prominence, and must also 

 be accounted for in any theory of igneous activity in this region. 

 If one is inclined to accept the hypothesis of contact origin for these 

 associated marginal amphibolites and garnet-schists, in spite of the 

 marked lack of constancy in their development and their failure 

 always to accompany each other, then these rocks would appear to 

 represent different phases of intensity in the contact action, passing 

 from garnetiferous mica schist, the lowest stage, through garnetif- 

 erous amphibolite, to hornblende schist, the highest stage, of such 

 activity. If one is not so inclined, however, it becomes necessary to 

 assume that the peripheral location of the amphibolite with respect 

 to the granite, and of the garnetiferous schist to the amphibolite, 

 or where this is absent, to the granite gneiss, is wholly fortuitous ; 

 and that the intrusive (notably the belt passing through Canton) 

 has merely happened to thrust itself between horizons in the Gren- 

 ville where the formations presented the associations described. 



On the whole, considering these two alternatives, the weight of 

 probability seems to favor the contact action. Accordingly, where 

 garnetiferous mica schist alone is found bordering granite, it may 

 be judged to be due to a lesser degree of metamorphic influence 

 than in the majority of instances where the amphibolite intervenes. 

 It is important in this connection to note that with one exception 

 the amphibolite never forms a rim round the granite without itself 

 being bordered by garnetiferous mica schist or garnet gneiss. 



While it might also seem plausible, on the ground of analog}^ to 

 demand a contact explanation for the already described strip of 

 garnet gneiss which borders one-half the circumference of the 

 granite east of Pyrites, it is to be remembered (i) that this is not 

 in any sense a garnetiferous mica schist, such as that underlying 

 the granite bluff along Harrison creek, but a garnet gneiss identical 

 with the sedimentary gneisses occurring apart from the granite at 

 Little River, (2) that there is in this case nowhere an exposed inter- 

 mediate amphibolite zone representing that extreme phase of meta- 

 morphism to which the garnet rock would be considered transi- 

 tional, but, instead, frequent amphibolite xenoliths and an occasional 

 strip of garnet gneiss are seen at or near the granite edge. The 

 chief points brought out by this discussion of the mutual relations 

 of these types of rock seem to be that the peripheral character of 

 the amphibolites and garnet gneisses favors their being viewed as 

 contact phases, but that their irregular and inconstant occurrence 

 and the development of similar types independently of contact ac- 

 tion necessitate caution in applying this interpretation to particular 

 cases. 



