MARINE SHELLS AND DEPOSITS FAR INLAND. 289 



of altered conditions affecting the influence of the sea in one direction or 

 another. For example, a comparatively small depression might establish a 

 wide channel connecting two oceans. Such a thiug might be conceived as 

 taking place between Avaters covering the valley of the Mississippi and 

 Mackenzie rivers. This would at once have an immense effect on the 

 glaciers, which we may suppose to have existed on both the Laurentian and 

 the Rocky Mountain sides of such a great strait. 



Changes in the proximity of the open sea in the valley of the St. Lawrence 

 and elsewhere may help us to account for the different directions followed 

 by the ice-grooves and by the drift materials in these regions, as well as the 

 changes in the elevations or slope of the land, which such alterations in the 

 distribution of land and water would imply. That such changes have taken 

 place appears to be pretty well established. Among other proofs of this is 

 the fact that marine shells are found in the Pleistocene deposits along the 

 St. Lawrence only as far west as Brockville, about 200 feet above the sea, 

 where they have assumed the brackish water forms ; whereas on Montreal 

 mountain they occur up to an elevation of 500 feet, which is sufficient to 

 have carried the sea all over the basin of Lake Ontario had the relative 

 levels of the land remained the same as at the present time. 



The Ante-Pleistocene Surface. — What was the condition of the surface of 

 the northern part of the continent just before the commencement of the 

 glacial period? There is every reason to believe that the Archean rocks, 

 which occupy so large a portion of the glaciated area, had become deeply 

 decayed and softened like those of the southern States, Brazil and Ecuador, 

 at the present day. This softened crust would be easily ground up and 

 swept away by the ice-sheet to form the deep and extensive layers of till 

 which cover such large tracts in the more southern regions of Canada and 

 extend into the United States. These layers have an average depth of perhaps 

 100 feet all over the extensive Paleozoic districts west and south of Hudson's 

 and James's bays and in those of the province of Ontario, and the average 

 depth may amount to 200 feet in Manitoba and a great part of the Northwest 

 Territories. This till is largely mixed with the debris of the local Paleozoic 

 or Mesozoic rocks, but so vast an amount of loose material could not have 

 been produced by the glaciers working on a surface originally as hard and 

 bare as that of the Archean rocks at the present time. 



The rounded bowlders are probably to a great extent the remains of the 

 hard nuclei or kernels, which, for some reason, in the case of crystalline rocks, 

 remain unaffected in the decay of the surrounding mass, although a certain 

 proportion of them, as well as nearly all the angular and sub-angular bowl- 

 ders and the pebbles, have resulted from the breaking and shattering of 

 the rocks along cliffs or about peaks and from the peeling up of beds beneath 

 the glaciers. 



