THICKNESS OF PLEISTOCENE ICE-SHEET. 271 



formed by the main axial range, when, being unable to pass this, it was de- 

 flected northwestward in a stream from 1,500 to 2,000 feet deep down the 

 valley of the Mackenzie and thence out to sea."* 



A partial measure of the thickness of the ice-sheet on British Columbia 

 is furnished by the elevation on the sides of mountains to which it has spread 

 its deposits of till. These accumulations of glacial drift are noted in many 

 places by Dr. Dawson, who however refers them to lacustrine or marine 

 action. They reach upward to an approximately horizontal limit or terrace 

 1,000 to 2,500 feet or more above the general level of the country, and to 

 heights from 3,500 feet to more than 5,000 feet above the sea. In some 

 places their upper margin may form a true terminal moraine, marking the 

 boundary of the ice at its maximum thickness or at some stage of pause or 

 readvance during its final melting. Elsewhere it is probably determined by 

 the general upper limit to which englacial drift was carried in the ice in 

 sufficient amount to produce by its deposition a well defined sheet of sub- 

 glacial till or ground moraine. The material of the higher terraces, accord- 

 ing to Dawson, is "identical in character with that of the general covering 

 of bowlder-clay, or so closely alike as to be indistinguishable from it."t 



During the departure of the ice-sheet, its melting was due to the influence 

 of sunshine and rains, the latter being doubtless brought then as now by 

 great storms sweeping across the continent in an eastward and northeast- 

 ward course. In consequence the borders of the ice-sheet appear to have 

 been pushed back generally in the same northeastward direction. Along 

 the valley of the St. Lawrence, the glacial current, which had before passed 

 southeastward transversely across it to the coast of New England, was during 

 this recession of the border of the ice-sheet deflected toward the southwest, 

 conforming to the law that the glacial motion near the edge of the ice turned 

 perpendicularly toward its boundary. 



In the latest stages of the waning ice-sheet it probably became divided 

 into three remnants, one covering northern British Columbia and contiguous 

 portions of the Northwest Territory and Alaska ; another occupying the 

 region west, northwest and north of Hudson's bay, stretching northward to 

 the large islands of the Arctic ocean ; and a third covering Labrador and 

 the country north of the St. Lawrence. From the second of these areas, 

 glacial currents moved south-south westw^ardly across the Churchill river and 

 Reindeer and Athabasca lakes, partly obliterating the earlier w^estward strise, 

 and southeastwardly across Marble island in the northwestern part of Hud- 

 sou's bay. Possibly the recession and final melting of the continental ice- 

 sheet caused it to extend over lands within the Arctic circle which had not 

 been covered by the ice when it reached farthest south. From the melting of 



* Bull. Geol. Soc. Am., vol. 1, p. 543. 



tGeol. and Nat. Hist. Survey of Canada, Annual Report, vol. Ill, 1887-88, pp. 89, 91. 96, 119, 172 B; 

 Trans. Roy. Soc. Canada, vol. VIII, sec. IV, p. 36. 



