172 NEW YORK STATE MUSEUM 
perhaps floating masses of ice. This water level did not reach to the 
divide at Summitville-— it was over 100 feet below it —: there is 
nothing to indicate an outlet along the only possible confining salient 
(the northern end of Shawangunk mountain one mile southwest of 
Rosendale) ; and the ice contact west of Stone Ridge does not mark 
the position of a front of live ice but an edge of motionless ice. 
There is no outlet for this “lake” other than a long course over the 
surface of what was left of the glacier in the Hudson valley with 
escape through the Highland gorge or through one of the low passes 
east of the river, for there is no pass west of the gorge low enough 
to drain it.** 
If we carry the inquiry into the lower courses of Shawangunk 
Kill and the Wallkill river, we find that stratified clays were 
deposited here at about the same level and that there is no confine- 
ment for lake waters in the valley of the latter except a tongue of 
stagnant ice filling the Hudson valley north of the Highlands. This 
must have been more than 4co feet thick over the present river bed 
opposite Newburgh and must also have covered the slopes now. 
drained by the Moodna Kill and its branches. Had the character 
of the recession of the glacier been by the gradual withdrawal of an | 
ice front, the Wallkill could not have failed to turn eastward along 
such a front and to have joined the Hudson by way of the Otter Kill 
and Moodna Kill, between Cornwall Landing and Newburgh. Two 
miles southwest of Montgomery, in the extreme northwestern corner 
of the Schunemunk quadrangle, there is a low point in the eastern 
divide (of the Wallkill basin) at 360 feet, which has not been crossed 
by a stream and which is a few feet Jower, according to hand-level 
measurements, than the plains immediately west and southwest 
which were built among remnant ice masses. And yet, the river 
continues northward for some twenty-five miles before it falls, by 
way of a course protected from clay-filling by persistent ice, into 
Rondout creek and so to the Hudson. 
In the northeastern corner of the Newburgh quadrangle there is a 
still lower point in the divide (about 350 feet) and an eastward 
leading channel crossing it. The channel begins about three miles 
east of Newpaltz and if it ever carried glacial drainage the stream 
must have soon run out on stagnant ice for its course can not be 
traced beyond Lloyd. 
™ There is a possibility that this region was somewhat higher relatively 
to the lower Hudson when this ice- ‘cumbered lake existed but probably 
not enough to make the other outlets to southern drainage available. 
