GEOLOGY 233 
north of the southern watershed, to one some three hundred 
miles north, in the vicinity of the headwaters of the Koaksoak 
river. 
The glaciation of Labrador seems to have been later than 
that of the western side of Hudson bay, as the striae from the 
western glacier are almost obliterated by those from the east 
and northeast along the rivers south of James bay. 
There is a marked difference in the evidence of the intensity 
of glacial action between the southern regions and the eastern 
and northern portions of the great area embraced in this report. 
On the shores and islands of Hudson bay and Hudson strait the 
crystalline rocks have been denuded of every trace of rotted 
surface material; they have been smoothed, polished and in- 
tensely striated, and their present condition is such that little' 
or no change has taken place since the disappearance of the ice, 
which once covered them deeply, the striae being so fresh as to 
appear of the formation of yesterday. When the eastern mouth 
of Hudson strait is left, a change is soon seen in following the 
eastern side of Baffin island northward. The hills are less 
rounded, and talus lies on the slopes of the cliffs; about Cum- 
berland gulf and Cyrus Field bay there is evidence of a univer- 
sal ice-cap having been present, but the rounding, polishing and 
striation of the rocks are markedly less than to the south and 
westward. In these places it is exceedingly difficult to find 
striae upon the rock surfaces, and these when found show that 
the movement was local and from the highlands towards the 
open sea. In the northern part of Baffin island the hills become 
more serrated in outline, and many of the higher points appear 
to have never been subjected to glaciation, the glaciers having 
only filled the valleys; if an ice-cap existed the ice-move- 
ment to the coast was determined by the course of the local 
valleys, and there is no evidence of a movement not depending 
upon local conditions as is the case in the region about Hudson 
