SOME PRE-CAMBRIAN LITERATURE OF NORTH AMERICA 265 



Ashbed type with scoriaceous sediments and a little conglomerate. 

 The thickness may be between 1,456 and 2,400 feet; (4) the Eagle 

 River group: a group of basic lava flows with frequent beds of 

 sediments; its thickness is between 1,417 and 2,300 feet; (5) the 

 Great Copper Harbor conglomerate: a coarse, heavy conglomerate 

 from 1,800 to 2,200 feet thick; (6) the Lake Shore traps: a series of 

 thin flows having a total thickness of about 800 feet; (7) the Outer 

 Copper Harbor conglomerate : from i ,000 to 5 ,000 feet in thickness ; 

 (8) the Nonesuch shales: black, fine-grained, micaceous sands and 

 grits, about 500 feet in thickness; (9) the Freda sandstones: red, 

 impure sands, shales, and conglomerates; 4,000 or more feet in 

 thickness. 



The Keweenaw rocks dip at a high angle toward the north under 

 Lake Superior and in general strike parallel to the axis of the 

 Keweenawan Peninsula. They have also been disordered by 

 numerous faults. The great Keweenaw fault, a longitudinal fault 

 dipping toward Lake Superior and cutting across the bedding, is 

 regarded by Lane as a thrust fault due either to the contraction of 

 the earth's crust or to uplift produced by the injection of a large sill. 

 Movement along this fault began before Cambrian time and has 

 taken place since the deposition of the Niagara limestone. There 

 are also other faults both parallel to the bedding and longitudinal 

 faults which cut across the bedding. There are numerous vertical 

 or nearly vertical transverse faults which have produced an offset 

 of the beds to the right in going toward the northeast. 



The Keweenaw rocks of northern Michigan as well as those of 

 other parts of the Lake Superior region show strong evidence of 

 consanguinity. Thus in all of these rocks potassa appears to be 

 subordinate to soda except in extremely siliceous varieties, and 

 in general the potass is very low. Free quartz is not abundant. 

 There is never an excess of alumina which would result in the 

 development of corundum nor are there any ultra-alkaline or ultra- 

 basic rocks. The content of iron is decidedly high. According to 

 Iddings' classification, the commonest type would be auverngnose. 

 No rocks are known which contain less of both silica and alkalies. 



Lane finds that the composition of these rocks is on the sodic 

 side of a line which he regards as a hypothetical eutectic between 



