126 The American Geologist. August, i897 
may be partially interlobate in character, but distinct proof 
of this was not seen. The section in the great ballast pit at 
O'Brien shows glacio-fluvial action quite plainly. 
Ilissatuiibi. A canoe trip was made up Dog lake to the 
low col which forms a part of the hight of land. The trough 
in which Dog lake lies is one of those curious deep rock-wal- 
led valleys that cross the wide, low crest of the Laurentide 
ridge at many places. Into this valley near the head of Dog 
lake a small stream cpmes in from the eastern side and the 
present division of the waters is determined by tlie low broad 
delta which it has built since glacial times. The lake was at 
a very low stage at the time of my visit, but the crest of the 
delta was not over three or four feet above it and the creek on 
the north side flows away to Brunswick lake, Moose river and 
Hudson bay. At the high stage of springtime it is said the 
water sometimes flows northward over the delta. Much of tlie 
lake is deep. But there are boulderly narrows at several 
points that suggest either moraines or boulder spits made by 
the jamming of ice-floes. The boulders are mostly quite an- 
gular. The whole region is very boulderly, gneiss and granite 
predominating, and the boulders are almost all sharply angu- 
lar. The shore of the lake is in a few places of solid rock, 
but nearly everywhere it is merely a mass of angular blocks 
with all finer material washed out b}'' the waves. This valley 
may have been the course of glacial drainage for a brief period. 
But I saw no evidence of a great glacial or post-glacial stream, 
nothing resembling the old-channel characters seen in other 
places. No evidence of submergence was seen in this vicinity. 
Striai at the station bear S. 38° W. (Mag.) 
Sinamari) and Conclusions. The higher beaches observed 
may be divided into two groups of localities, those clustered 
around Port Arthur and those near Jackfish. Not counting lake 
Kaministiquia, the upper limit of submergence in the first 
group is about 400 to 450 feet above the lake, while in the sec- 
ond group it is about 400 to 425 feet. Between the two groups 
there is an interval of about 70 miles in which high beaches 
have not yet been reported by anyone. (Highest reported by 
Lawson 360 feet at Mazokamah.) It might be supposed that 
the highest beach in the two groups is one continuous line, and 
it seems probable that this is the case, but further exploration 
