566 THE JOURNAL VOR GEOLOGY. 
that run alongside the glacier and constitute its main drainage 
system, and in the reflection and radiation of heat from the 
adjoining cliffs. But neither of these afford an explanation of 
the abruptness of the end of the glacier. This terminates upon 
a flat gravel plain produced by the glacial wash, and no cliff lies 
in front between it and the Gulf, a mile or two distant. As this 
glacier has a southern frontage, the phenomenon might perhaps 
at first thought appear to be due to exceptional exposure to the 
southerly sun, but the sun is less partial in its favors here than 
in southern latitudes. During the main melting season it is con- 
stantly circling above the horizon and throwing its rays on all 
sides of the glacier, and, although the southerly sun is more 
effective than any other, the difference is less than in lower 
latitudes. Besides, we shall find in the study of other glaciers 
of the region that eastern and western and even northern expo- 
sures present the same characteristics. The explanation seems 
to lie in the low inclination of the sun’s rays and in their impact 
from all points of the compass in succession. It is obvious that 
rays of low slant strike the back of the glacier at a very acute 
angle and easily glance away with little effect. On the edge of 
the glacier, however, they strike more directly against the sur- 
face and hence have greater effect. In addition to this, the 
slanting rays that impinge on the surrounding surface at low 
angles are again reflected at like low angles, and hence a much 
larger proportion of them strike upon the edge of the ice. 
Thus it appears that a larger proportion of the sun’s rays fall on 
the edges of a glacier in high latitudes than upon the edges of a 
glacier in low latitudes, and it is obvious that it is the propor- 
tional effect of the sun’s rays that determines the contour. For 
like reasons there is a larger proportional reflection from promi- 
nences in the vicinity of the ice, so that, although verticality of 
face is not dependent on the presence of reflecting cliffs near the 
glaciers, it is facilitated by them. It will be found in subsequent 
observations that where promontories of rock rise through the 
of the Greenlanders, the ice, 
”) 
ice-sheet, forming the ‘nunataks 
in many instances, does not crowd against their sides but is. 
——eSC—“‘“C—t;‘C;C;C!U UU 
