ESKERS IN THE VICINITY OF ROCHESTER, NEW YORK. 177 



possessed some movement, and the tunnel gradually enlarged by the 

 melting of the ice adjacent to the stream and by mechanical erosion, 

 Tributary tunnels were developed and maintained in the same way 

 as the main course. The subglacial streams issued from beneath 

 the ice into ponded glacial waters, in some cases as along the 

 New England coast ; beneath marine waters, or onto surfaces 

 sloping away from the ice front. In any case there resulted 

 decrease in velocity at the edge of the ice, through loss of 

 "head," and consequent deposition from the waters heavily 

 charged with debris. This clogged the mouth of the tunnel 

 and caused the water to flow at a higher level in the tunnel, eroding 

 and melting the upper surface of the tunnel. As the tunnel got 

 larger and the mouth more and more clogged with debris the main 

 current of the subglacial stream would vary in position and in load 

 leading to aggradation. At first the deposition would be only local, 

 here and there in the channel, but as the tunnel enlarged the deposi- 

 tion would be more frecjuent to a perfect or nearly perfect ridge 

 development. By reason of the velocity the coarser materials would 

 be left, the finer carried out into the deltas, outwash plains, kanies, 

 etc., that were forming at the edge of the ice. Subglacial streams 

 issuing into standing water would have had their velocities checked 

 some distance back from the edge of the ice without regard to ag- 

 gradation at the mouth of the tunnels, thus leading to deposition in 

 the tunnels. 



The supply of debris was obtained from the surface, from 

 englacial material, from the basal portion of the ice heavily laden 

 with material, from tributaries, from the flow of the ice in case it 

 possessed movement, by the erosion of any till in the upper reaches 

 of the stream's course and in the tributary courses that already had 

 been deposited beneath the ice. 



These subglacial streams were for the most part short lived. 

 They maintained their courses for brief periods only. They may 

 have been diverted in part or wholly by the closing of the tunnel 

 through ice movement, by the collapse of the tunnel, by the opening 

 of an easier passage for the escape of the waters, possibly by finding 

 exit to the surface or to an englacial position, and flowing there for 

 some distance before plunging into another crevasse. 



