REPORT OF TIE DIRECTOR, 1922 173 
The same kind of argument applies to the lower body of water 
pended by the Hudson valley ice tongue in the valley west of the 
upturned limestones, extending from the bend in [Esopus creek at 
the ice contact southwest of Kingston, to Catskill creek at Leeds. 
An ice-front recession would have turned the Esopus into the Hud- 
son at Kingston, where today, what is left of this artifically beheaded 
stream hangs above the mouth of Rondout creek in a manner to 
tempt the engineer seeking waterpower. 
If stagnant ice filled the Hudson valley opposite Kingston at this 
stage there ought to be a fair chance of finding direct evidence of 
the fact. The delta sands of Kingston Point (Rhinebeck quad- 
rangle) rise to an elevation of 220 feet; they were built from south 
to north that is to say, up-river; overlie nearly 200 feet of clay; and 
appear to have direct relation with the Rondout valley. At Rosen- 
dale (Rosendale quadrangle) some eight miles to the southwest is 
a glacial sand plain rising to the same elevation and also apparently 
related to the Rondout valley. No sections were found to exhibit 
its structure. If these two deposits so close together were not made in 
the same water body the according levels are difficult to understand. 
Assuming that they are contemporaneous deposits, either they were 
originally separated by remnant ice in the lower Rondout valley or 
they were once continuous. If continuous, they have been separated 
by the removal of the intervening material by subsequent erosion. 
(No evidence of postglacial faulting was noted in the region). The 
ground between the 220 and 240 foot contours was searched for 
remnants of this supposedly continuous plain and below 220 feet for 
evidences of normal dissection or dissection complicated by the 
melting out of ice. Terrace levels of both clay and sand were found, 
some higher and some lower than the required level but nothing to 
indicate that the Rosendale Plains and the Kingston Point delta 
were once continuous. On the contrary, the bare rock surfaces, 
especially north of the creek, give every indication of having been 
protected by a covering of ice from receiving sediment. The dis- 
section of the valley-filling from Rosendale to Rondout and Port 
wen was partly governed by the melting out of ice. It is very 
doubtful if even the sands at the latter place (at 140 feet) were 
built out into open water; while the occupation of the lower valley 
by ice, preventing the deposition of clays there while the Kingston 
Point deposit was building, renders it probable that this latter deposit 
was made while the Hudson valley proper still contained a con- 
siderable amount of ice. 
