230 The American Geologist. \|, n i, L896 
superglacial streams, holds that the englacial drift became 
superglacial, by surface ablation of the Lee, in sufficient 
volume to account for practically all the various types of mod- 
ified drift, and that the superglacial streams were adequate 
for its transportation and deposition. That this must be 
the history of a part of the modified drift is obvious from the 
fact that these conditions are realized in modern ice-sheets; 
hut they are only realized, it must be added, to a very limited 
■extent. 
It is a common and, perhaps, a fair assumption that in 
Greenland and Alaska are exhibited now, on a smaller scale, 
nearly all the essential phases of the disappearance of the 
Pleistocene ice-sheet. A general survey of these and other 
regions now undergoing glaciation lias revealed only one 
notable occurrence of superglacial drift, namely, that mantling 
the outer margin of the Malaspina glacier; and that, as we 
have noted, is of somewhat exceptional origin, inasmuch as it 
is not derived from strictly normal englacial drift. Little 
mere can be said for superglacial streams. They are either 
entirely wanting, or they are short-lived, being almost invari- 
ably swallowed up by crevasses and rarely discharging over 
the margins of the ice-sheets. Nowhere, so far as I am aware 
have superglacial streams been observed actively washing and 
distributing superglacial drift. Assuming, as I think we must, 
that the englacial drift is crowded in the basal layers of the 
ice, enormous wastage of the ice must occur before it becomes 
superglacial ; and the extreme brittleness and consequent As- 
suring of the ice protects it from the ravages of superglacial 
streams, until in the course of time it becomes forest-clad and 
assumes the character of an ancient soil. 
It seems to me very probable, however, that when consider- 
able sections or areas of the Pleistocene ice-sheet were so far 
wasted as to be absolutely stagnant, and when superglacial 
drift covered its surface and checked the melting of the ice, 
the still existing crevasses may have become choked with 
drift to such an extent as to keep the streams superglacial or 
in channels open to the sky, and thus to realize the essential 
conditions of the formation of modified drift from superglacial 
drift. Meanwhile, however, or before these conditions are 
realized, the water resulting from the melting of thousands of 
