294 R. BELL — GLACIAL PHENOMENA IN CANADA. 



mosl conspicuously manifested where the more even course of the glacier has 

 been interrupted by a riseorturn, or by sonic hard knob of rock in its bed. 

 The immense pressure and the friction of the rocky debris would generate a 

 certain amount of beat, and the i<v, where very thick and mingled with 

 earthy matter, would tend to retard the radiation of heat from Mother 

 Earth ; for, notwithstanding the fact that transparent ice is a conductor of 

 heat, a mixture of ice and drift material a mile or two in thickness would 

 retain terrestrial heat, although in a less degree than an equal depth of 

 ordinary rock. The water thus produced would often be temporarily im- 

 prisoned and in the course of the movements of the ice would become sub- 

 jected to greal hydrostatic pressure, causing it to force passages for itself 

 among the debris. This might account for some of the singular forms as- 

 sumed by the drift materials. 



The Direction of Glacial Flow. — The courses of the glacial stria' having 

 been noted in all parts of the northern States and the southern parts of 

 Canada before we knew much about them in the more northern region, it 

 was assumed that the general direction was everywhere southward, with 

 local variation- to the east and west of south. This circumstance, along with 

 the stupendous force which it was obvious must have produced the phenom- 

 ena of continental glaciation, gave rise to the theory of a universal ice- 

 sheet covering the northern regions of this hemisphere duriiig the drift 

 period. Our observations throughout "the great north land" have, how- 

 ever, modified this view, and it now appears as if Bomething less would 

 account for the wonderful facts of the great ice age in North America, as 

 well as in the Old World. 



The dispersion of the ice doe- not appear to have been from a Bingle district 

 iii northern Canada, as supposed by some, but from several. One of tb< 

 as already stated, was in eastern Labrador; another lay between Hudson's 

 bay and the Mackenzie river; while the wide, shallow basin of Hudson's 

 bay itself formed the grandest neVe" and collecting ground of all. Besides 



the ice which formed directly from the copious snows falling on this vast ex- 



panse itself, continuous contributions were received from the Labrador penin- 

 sula to the easl and the great region to the northwest, and the mass discharged 



itseli northward into the deep ami wide valley of Hudson's -trait ami south- 

 ward and Bouthwestward over the Paleozoic and Laurentian plateau. The 

 ice-sheet appear- to have flowed outward everywhere from the eastern, 

 southern, and western margins of the greal Laurentian plateau— that i- to 

 Bay, it- general course was eastward on the coast of the North Atlantic, 

 ithward from the Strait of Belle Isle along the St. Lawrence and the 

 Greal Lake- t., the Winnipeg basin, and Bouthwestward ami westward fr 



thence ],, the Mackenzie river. A faint indication that the regions ea-t and 

 Wesl of Hudson's hay, above referred to, were firmer centres of glacial 



