14 F.. Leverett—Correlation of New York | 
of ice which were dropped into the gravelly deposits or per- 
sisted during their deposition and which upon melting left the 
basins. No way has suggested itself in which the basins could 
have been produced by the lake waves alone. Were the mate- 
rials sandy we might suppose wind to have been a leading 
agent in their production, but the fact that they are of a gray- 
elly nature forbids this supposition. 
A reconnaissance was made into the district west and north 
from the western terminus of this ridge with a view to deter- 
mining the line of continuation of the ice margin beyond the - 
beach. There is a broad tract of old lake bottom open to 
view between the Crittenden beach and the shore of Lake Erie, 
and the conditions seemed favorable for such a determination. 
There is abundant evidence of a southwesterly ice movement 
across the region, at least to the vicinity of Buffalo, for striee 
in the north part of Buffalo have that bearing and so also have 
striae and drumlins in the district between Buffalo and Lock- 
port. But the ice margin seems not to have been held at any 
one line sufficiently long to build up a strong moraine. Noth- 
ing was found connecting closely with the Alden till ridge. A 
few miles to the north, along a line following the base of the 
Corniferous escarpment from Akron westward to Buffalo, 
there are a series of short ridges and low knolls of drift banked 
somewhat closely against the escarpment and filling recesses or 
embayments in it. Possibly the knolls and ridges represent 
the line of the ice margin at one stage of the retreat. It seems, 
however, quite as probable that they are attributable in some 
way to the resistance of the ledge to the ice movement, in 
which case they may be submarginal instead of terminal 
deposits.* Aside from these knolls and a few drumlins in 
southern Niagara county the surface is very flat over the entire 
district in Niagara and Erie counties southwest from the Lock- 
port moraine. Bowlders are in places abundant, but so far as 
discovered they do not constitute a well defined.belt. Had the 
ice sheet terminated in a deep body of water it would be less 
remarkable that no moraine could be traced, since icebergs 
might transport the material widely or cause the margin to 
have considerable oscillation. The margin might, because of 
* Dr. J. W. Spencer has published the view that a well defined beach line 
called by him the Lundy leads southwestward from Akron near the line of the 
knolls and ridges just mentioned. (This Journal, vol. xlvii, 1894, pp. 207-212.) I 
did not observe such a beach line, perhaps because my studies were not suf- 
ficiently complete. As I have had no opportunity to visit the ground or to con- 
fer with Dr. Spencer since his paper appeared I do not feel warranted either in 
concurring with or dissenting from the interpretation he has made. I can only 
state that the gravelly ridges in the vicinity of Akron have not in any case so 
far as personally observed seemed to present decisive evidence of a lacustrine 
origin. On the contrary they seem in most instances to be decidedly glacial in 
type. 
