of the North American Ice-Sheet. 87 
due apparently to inequalities in the amount of snow-fall and 
of melting upon adjacent regions, were sufficient to make angles 
and lobes at the termination of the ice-sheet, and also doubtless 
caused downward and upward currents, by which much of the 
drift gathered while crossing a nearly level area, would be dis- 
tributed throughout the lower part of the ice, probably to the 
eight of several hundred feet. The beds of loose material 
which had been produced by long-continued decomposition of 
the ledges or accumulated by previous glacial action, together 
with the thick fluviatile deposits that probably occupied the 
valleys, were ploughed up by this ice-sheet and thoroughly 
kneaded with each other. Very large amounts of detritus were 
added from erosion of the rock-surface. Fragments of all 
sizes and in great profusion were loosened and wrenched away, 
while the ledges were everywhere worn and striated by bowl- 
ders and pebbles, which were rolled and dragged along under 
the vast weight of ice, breaking up and grinding themselves 
- the underlying rock into gravel, sand, and even the finest 
clay. 
The material which was thus gathered, mingled and swept 
along in and beneath the moving ice, upon reaching its termin- 
ation was accumulated in heaps and ridges of unstratified drift, 
full of bowlders, and identical’with the till which generally 
overspreads the ledges and underlies the modified drift of gla- 
cla = The moraines of Long Island and southern 
ng 
and by the abundance and large size of its bowlders, which 
have seldom been worn or rounded except by the weather. _ 
The massive hills of gravel and sand which form so promi- 
nent'a part in this series of drift deposits heaped at the termi-. 
nal front of the ice-sheet, to have been brought by gla- 
cial rivers. The melting of the ice at and near its terminal 
er a change of climate, 
their melting was extended over a very wide area. Their sur- 
face was then hollowed into basins of drainage and channeled 
