154- The American Geologist. Septomijer, i897 
may suffice to lead away a tongue of the ice cap. On the 
southern side of mount Hope such a valley glacier, project- 
ing from the south Cornell glacier, moves at right angles to 
the main axis of tlie ice, then tui-ns again, until at itsend the 
current is actually back t(>ward the direction from which the 
ice came. Here debris brought from the east turns northwarct 
and Hnally back a short distance toward its source in the east. 
Later, if the land is high, the continental glacier becomes 
transformed locally to smaller snow fields, or ice caps, t\nd 
valley glaciers. Such is the present condition of the Upp<'r 
Nugsuak peninsula, and doubtless other parts of the Green- 
land coast.* Then the glaciers may begin to move back tow- 
ard the direction from which the earliest ice cap came, and 
the boulders may be carried back toward their sources. It is 
possible that some of the gneiss boulders mentioned b}- pro- 
fessor Chamberlain as occurring at the terminus of the Blase- 
dale glacier of Disco island may have begun such a journey 
eastward toward their original source. f In the Upper Nug- 
suak there is certainly this movement. When we are consid- 
ering the distribution of drift in the United States we must 
certainly take into account this fact, that ice near the termi- 
nus of a great glacier is decidedly deflected, and that the cur- 
rents in different stages may be not merely at right angles, 
but even parallel to, and back toward some previous position. 
Drift on the Upper Nugsuak Peninsula. As much as 
three-fourths or even four-fifths of the Upper Nugsuak penin- 
sula is either bare rock or talus deposit. The barrenness of 
drift is remarkable. In protected places, on slopes and in 
narrow valleys, there are till beds, and now and then there are 
moraines at the margins or ends of the vallej's : but elsewhere 
there is bed rock. No stratified deposits of drift were found 
on the peninsula. Indeed the place where drift is most abun- 
dant is exactly along the glacier margin. At first I thought 
this indicated a longer stand of the ice at this point; but very 
soon the true explanation was seen to be removal rather than 
failure to deposit. At best the Greenland glacier is not great- 
ly drift laden, but the amount in all parts of the lower ice 
that was visible would even now furnish at least a thin film 
over the land, if the lee should melt. 
*See Tarr, Am. Geol., vol. xix, April, 1897, pp. -262-267. 
tJour. Geol. 
