De Geer.] 
472 
[May iS, 
the longer axis of this lake trends in nearly the same direction as 
the greatest warping of Lake Agassiz, it seems probable that the 
limit for this warping and at the same time for the upheaved 
area lies just between Lake Agassiz and Lake Dakota or through 
Lake Traverse. It is by no means certain that the limit for the 
uplifted region or the iso base for zero remained at the same place 
when the marine limit in the St. Lawrence valley was formed ; 
but we may assume it for this part of the continent, since we can- 
not. at present, expect to get more than a general idea of the 
direction of the isobases and their maximum gradient. To judge 
from the probable thickness and direction of the receding ice-border, 
it appears that the formation of the highest or Herman beach of 
Lake Agassiz was probably antecedent to the geoid surface 
which is traced in this paper. Moreover it is quite possible 
that the ice had not receded from the St. Lawrence valley before 
Lake Agassiz received the northward outflow. Yet to be quite 
sure of maximum figures for the gradient, I have used the 
measurements of the highest or Herman beach, though they may 
be too high. 
As to the probable position of the marine limit in the other 
northern portions of the area very little can be added. Some ex- 
plorers, believing that every kind of drift is deposited in the sea, 
have not paid due attention to the determination of the limit for 
the real marine deposits ; others seem to have estimated only the 
height of beaches accidentally discovered and their most reliable 
observations are made with a barometer often at a long distance 
from any known level or base-barometer. 
From Murray’s paper on the glaciation of Newfoundland 1 it 
seems that marine deposits are found on that island at a height 
of about 200 feet. According to R. Bell 2 distinct beaches are 
seen at Nachvak in Labrador at an estimated height of 1,500 feet ; 
but I have not found more precise measurements for these tracts. 
Even if this measurement should be over-estimated, these beaches 
may be among the highest in all the uplifted area. But the low 
levels at which marine deposits are found in the relatively well 
exjflored southern and western parts of Greenland, make it im- 
probable that the extraordinary high levels, reported as occurring 
1 Proc. and trails. Roy. soc. Can., 1883, I, pp. 55-76. 
2 Rept. geol. sprv. Can., 1885, p. 8 DD; and Bull. geol. soc. Amer., 1889, 1, p, 308. 
