LATER STAGES OF EVOLUTION OF IGNEOUS ROCKS 67 



is the primary igneous material from which all known igneous rocks 

 are derived, yet it is a logical step.- Any large saHc body, say 

 granitic, whose every exposed portion is simply granite tells us very 

 little about itself. Such are the bodies which give sahc rocks their 

 great preponderance. It appears to be the prevailing opinion, sel- 

 dom exphcitly stated, perhaps, but certainly implied in description 

 and discussion, that such bodies of granite were formed by the 

 crystallization of granitic magma which always existed as such. 

 The opinion is not based on any definite evidence to that effect 

 but merely on a lack of direct evidence to the contrary. Such 

 lack of evidence stands over against a mass of convincing and posi- 

 tive evidence of the origin of granites (and other sahc types) which 

 occur in bodies of such size (or shape) that their relations are clearly 

 shown. 



A broad survey of the Keweenawan eruptives of the Temis- 

 kaming-Lake Superior region lays bare the essential features of the 

 differentiation of basaltic magma as it is controlled by the size of the 

 body. The Keweenawan flows are overwhelmingly normal basalts. 

 The quickly cooled smaller dykes and sills are principally normal 

 diabase, sometimes olivine diabase. The sills of moderate thickness 

 have diabase with micropegmatite interstices as a result of some- 

 what slower cooling. The sills of considerable thickness commonly 

 have diabase with micropegmatite interstices and a small sahc 

 differentiate (with intermediate types) at their upper borders. 

 The writer has seen several examples both in the sills of Gowganda 

 Lake and in the Logan sills of the north shore of Lake Superior."^ In 

 the case of the huge sheetlike mass at Sudbury the sahc differentiate 

 at the top constitutes a considerable area of granite. It is fortunate 

 for the geologic investigator that the Sudbury intrusive sheet hes 

 in a basin-shaped structural depression. If it had assumed the 

 form of an inverted basin, erosion might have exposed the granitic 

 phase only and the derivation of the granite by gravitative differ- 

 entiation would not have been revealed. It would then have 

 passed muster as an ordinary granite bathohth, and, no doubt, 

 many bathoHths (bodies without visible floors) are similar to the 



^N. L. Bowen, Gowganda Sills, Jour. Geol., XXVIII (1910), 658; Logan Sills, 

 Ontario Bur. Mines, Ann. Rept., 1911, p. 127. 



