138 PROCEEDINGS OF WASHINGTON MEETING. 



belt of country along a sea coast : but the glacial rivers and their large and small 

 branches had much steeper gradients than those of the present river systems on 

 the land surface, and often or generally they Unwed in deep ice- walled channels, 

 more like canons than ordinary river valleys. In the stages of growth and culmi- 

 nation of the ice-sheet it had possessed a nearly level surface of neve or accumu- 

 lating snow, mainly unflecked by any stone fragment or other drift material, or 

 even dust ; but when the final melting had dissolved away its upper portion, the 

 englacial drift begin to be exposed upon the surface, and at last on many areas the 

 ice doubtless became buried and concealed by this deposit, as was supposed by 

 Professor X. H. Winchell many years ago* and as was found by Mr. I. C. Russell 

 last year in his exploration of Malaspina glacier, at the foot of the Mount Si. 

 Elias range.f The completion of the ice-melting allowed much of the englacial 

 drift to fall loosely as an unstratilied deposit, called the upper till, on whatever sur- 

 face was beneath the ice. whether around moraine, oi other subglacial drift, or the 

 bed rocks. Previous to this, while the glacial melting was in progress, other large 

 portions of the englacial drift were washed away by the superglacial drainage and 

 deposited in beds of gravel, sand, clay, and silt, partly in the ice-walled channels 

 of the glacial rivers, but mainly beyond the ice-margin. The various formations 

 thus derived from the englacial drift will be more fully noticed in a later part of 

 this essay ; and our consideration is here directed especially to the process of disso- 

 lution^ the ice, releasing the drift which it had held. 



Conditions analogous respectively with the growth and maximum extension of 

 the Pleistocene ice-sheets, and with their wane and departure, have been lately 

 made known to us in the case of ice-sheets now existing. The stage of growth or 

 ice accumulation is represented by the inland ice of Greenland, explored by Nor- 

 denskiold, Peary, and Nansen; and the stage of glacial recession, attended by 

 deposition of the englacial drift on the wasting ice surface and afterward on the 

 land beneath, is illustrated by the Malaspina glacier, as before noticed. The ex- 

 plorers of the Greenland ice-sheet describe its surface, excepting near the border, 

 as a vast expanse of neve, with no nunatak or peak rising above it, and with no 

 superficial drift. The line, gray powder which occurs somewhat plentifully on the 

 western portion of this ice«gheet, called by Nordenskiold " kryoconite," and believed 

 by him to he cosmic dust, but which Hoist i has regarded as a loess-like part of 

 the englacial drift, brought upward through the ice to its surface, appears instead 

 to be dust blown from a mountainous belt of land forming the western coast. On 

 the eastern side of Greenland, where Nansen's ascent upon the ice was made from a 

 part of the shore having little bare land, no noticeable quantity of this dust was 

 founds Xansen, in his expedition across the ice between latitude <>4° 10 / and lati- 

 tude *>4° 45', where its width is about 275 miles, encountered no streams of water 

 nor any water-courses at distances exceeding 20 miles inside the ice boundary ; and 

 he particularly remarks the absence of moraine d6hris or erratic blocks even on the 

 outer portions of the ice-sheet, excepting for a distance of " no more than a hun- 

 dred yards from the extreme edge."|| But a very remarkable contrast to all this 



*Geol. Survey of Minnesota, First annual report, for 1872, ]» 62 



f'An Expedition u> Mount St. Elias. Alaska," Nat Geogr. Magazine, vol. in. 1891, pp. 53-203. 

 X " Dr. N. O. Hoist's Studies in Glacial Geology," a review by Josua Lindahl, Am. Naturalist, vol. 

 xxii, July, 1888, pp. 594-598. 

 §F. Nansen, Tin- First Crossing of Greenland, vol. i. p. 483; vol. ii. p. 179. 

 || Ibid., vol. ii, p. 479. 



