138 
GLACIAL PHENOMENA IN MAINE. 
ice passing from the mainland over this arm 
of the sea sunk eastward and westward into 
these two gorges, acquiring, no doubt, addi¬ 
tional thickness thereby, and being, in con¬ 
sequence of this change in its normal course, 
slightly deflected from its usual direction in 
working its way up against the shore of Mount 
Desert. This is shown by the fact that the 
glacial marks on the northwest shore bear, as 
I have already said, slightly to the east, while 
those on the northeast shore bear slightly to 
the west. On approaching the centre of the 
island the marks converge towards each other, 
and regain their primitive direction due north 
and south, on its more elevated positions. I 
have often observed in Switzerland like in¬ 
stances, when from some local cause the direc¬ 
tion of the movement was slightly deflected to 
the right and left, converging again at some 
little distance. In the valley of Hasli, be¬ 
tween the hospice of the Grimsel and Gutta- 
nen, are several knolls which afford examples 
in point. On the upper side of these knolls, 
facing the higher part of the valley, from 
which large glaciers formerly came down, 
marks are carried directly up the slope oil to 
