228 The American Geologist. 
\|>ril, 1896- 
thus becomes simply a broader phase of the principle of the 
absorption of detritus by shearing and flexing. These con- 
siderations lead me to venture the opinion that, while depths 
of englacial drift amounting to 1,500 or 2,000 feet or more, or 
to one-fourth to one-third of the thickness of the ice-sheet at 
its maximum, as held by Upham, cannot be regarded as 
strictly normal or as prevailing over considerable areas of 
plain country, they are possible under the special conditions 
indicated. 
Relations of Englacial Drift to Modified Drift. 
The manifold forms of modified drift, or washed and strat- 
ified gravels, sands and clays, of glacial origin, forming del- 
tas, terraces, overwash or apron plains, eskers, kames, etc., 
although occurring almost wholly in valleys and on lowlands, 
constitute in the aggregate a considerable fraction of the 
total volume of the drift. Concerning the source of the mod- 
ified drift, there is ns } r et no general agreement among gla- 
cialists. The main views, of which all others may be regarded 
as modifications, are: (1) that the modified drift has resulted 
chiefly from the washing and assorting of the till or ground 
moraine by glacial streams, during and following the waning 
and disappearance of the ice-sheet : (2) that it was derived 
from englacial drift through the agency of subglacial streams ; 
and (3) that it had its origin in englacial drift which became 
superglacial by ablation and was washed and assorted b} 7 sn- 
perglacial streams. Undoubtedly all of these theories are re- 
quired to account for the totality of the modified drift; and 
the real question is as to their relative importance. 
It is inconceivable that the ice-sheet could disappear with- 
out some washing or modification of the ground moraine. It 
is, however, a noteworthy fact that the till does not present 
about the heads of southward sloping valleys or elsewhere 
evidences of such extensive erosion as should be required for 
the general or unqualified acceptance of this view. The ab- 
sence, as a rule, of strongly marked erosion lines in the ground 
moraine is particularly striking in the case of the drumlins, 
since their highly definite and characteristic contours due to 
glacial moulding make them extremely delicate records of 
aqueous erosion. Every glacialist know r s how rare and insig- 
nificant are the evidences of drumlin erosion which can be 
