ESKERS IN THE VICINITY OF ROCHESTER, NEW YORK. 197 



drainage was largely subglacial, and that the tunnels, being small, 

 could not carry the amount of water provided by the rapid melting 

 of the ice, the excess being ponded in the crevasses of the ice adja- 

 cent to the subglacial streams for considerable distances back from 

 the border of the ice. This ponded water created great "head"' in 

 the subglacial streams, which therefore eroded rock ridges beneath 

 the ice, and other deposits encountered, sweeping the materials 

 forward and suddenly dropping them at or near the mouth of the 

 tunnel, where the pressure was removed (48). 



Gregory says that an esker is a "fluvioglacial ridge formed of 

 sand and gravel which has been laid down along the course of a 

 glacial river. The deposition has taken place mainly where the river 

 emerged from the glacier, and the course of the esker is usually at 

 a high angle to the edge of the glacier." "They have been built up 

 into long ridges by the overlapping of successive delta fans" (44). 



Still more recently Trowbridge has advocated essentially the 

 same idea that of esker formation at the edge of the ice or within a 

 re-entrant back from the edge (111). In the following statement 

 the important details in this method of esker formation have been 

 taken from a recent paper by the latter writer (111). 



Statciiioif. Trowbridge believes "that most eskers are simply 

 kames drawn out into long lines by the slow retreat of the edge of 

 the ice while kame deposition is in progress. If a kame is being 

 formed at the edge of an ice sheet, and the edge retreats slowly, 

 deposition will continue so long as the re-entrant remains and the 

 stream continues to issue there, and the kame will be drawn out 

 into a long ridge or esker." Discontinuities would result if "during 

 the recession of the edge of the ice the re-entrant ceased to exist, 

 or the stream ceased to issue there ; when a re-entrant and the mouth 

 of a subglacial stream again coincided deposition would begin again. 

 This would make a break in the esker whose length would be deter- 

 mined by the rate of recession of the ice and the length of time dur- 

 ing which deposition was not in progress. Such changes as these 

 would take place suddenly," and would account for the abrupt ter- 

 mination of esker ridges. A slowly retreating ice edge would form 

 a high, thick esker, rapid retreat would form a "thinner, lower one ; 

 rapidly changing rates of recession would cause an esker of varying 



