REPORT OF THE DIRECTOR I916 ' 37 



At the next later stage of the retreating ice front to be marked 

 by conspicuous moraines, the ice was pushing through the gaps 

 in the northeastern border range between Mount Pisgah and 

 Windham High peak and building massive moraines in the valleys 

 tributary to Batavia kill at and above Windham. The ice from 

 the gap at East Windham bulged eastward into the upper Batavia 

 Kill valley above Hensonville and impounded a lake in which deltas 

 at two levels, approximately those of the cols south of Big Hollow 

 and Hensonville, respectively, were built. At Windham and below, 

 the ice at this stage discharged into the waters of the glacial lake 

 whose outlet was through the cent'ral escarpment at Grand gorge. 



East and northeast of Windham are about a dozen perfectly 

 formed drumlins whose origin is probably connected with the 

 spreading of the ice after its passage through the narrow gaps in 

 the northeastern border range. 



Local glaciation. Special attention was given to the problem 

 of local glaciation, evidences of which had previously been dis- 

 covered at a few points. 



On the north-facing slope of the central escarpment, no less than 

 ten valleys were found in which distinct morainic loops, convex 

 down-valley, testify to the former existence of independent local 

 glaciers moving northward and northeastward in a direction 

 directly opposed to that of the movement of the continental glacier. 

 Many other valleys, similarly situated, contain deposits which, con- 

 sidered individually, do not constitute convincing evidence of local 

 glaciation, but which, taken as a whole in the light of more definite 

 evidence in neighboring valleys, seem to be most reasonably inter- 

 preted as the products of local glaciers. 



A characteristic type of deposit, found in most of the valleys on 

 the north face of the central escarpment, and elsewhere in similar 

 situations, is a mass of smooth, thick drift which makes a dis- 

 tinct step in the valley bottom with a steep slope up to the top of 

 the step and a more gentle slope above. Such deposits commonly 

 present a convex front down the valley and many of them are 

 higher in the middle of the valley than at the sides. In many 

 instances it is clear that they were fashioned by local glaciers 

 moving down the valleys. Field study led to the hypothesis that 

 they were built, in a manner analogous to the formation of drum- 

 lins, by deposition of till beneath the ice where its transporting 

 ability was decreased, either on account of its thinness near the 

 end or on account of its spreading in the form of piedmont bulbs 



