Davis.] 
492 
[May i8, 
Russell in his recent observations on Alaska make frequent men- 
tion of the prevailingly overloaded character of the streams that 
are discharged from the ice. 
13 . CONCLUSION AS TO THE JOINT ORIGIN OF ESKERS AND 
SAND PLATEAUS. 
In view of all these circumstances, it must be concluded that 
the Auburndale sand plateau and the esker were built by the 
same glacial stream during a comparatively short time ; and that 
both before and after the building of these deposits, the stream 
found some other escape than along the Auburndale esker. Had 
the stream run here at an earlier time, the sand plateau would 
have had its head further south in the then position of the ice 
front ; had the stream continued to run here after the plain had 
reached the present size, the plain must have grown larger and 
the beds at its head, exposed in the railroad cut, must have given 
some indications of backward extension, following the backward 
melting of the ice front. I see no other conclusion than the one 
presented, that the sand plateau was built during a brief time of 
discharge of a sand-laden stream, which before and after that 
time found other avenues of escape. 
14 . ORIGIN OF ESKERS IN SUPERGLACIAL CHANNELS. 
If the stream that fed the sand plateau had lain in a channel 
on the surface of the ice, it must have been open to the sky and 
hence it must have had a descending slope to the level of the 
standing water at the ice margin. Here a difficulty arises in 
supplying such a stream with the fragments of slate and con- 
glomerate that come from the floor of the Charles River lowland 
close by. It may be admitted that after a forward carriage of 
some distance, the drift gathered at the bottom of an ice-sheet 
may sometimes be raised to a considerable height towards its sur- 
face ; but it is difficult to understand how this can be done in so 
short a distance as here lies between the source of these stones 
and their present resting place. But leaving this difficulty aside, 
let us turn to another question. After the brief activity of the 
superglacial stream , during which the esker and the sand plateau 
were both essentially completed , the stream turns elsewhere, and 
leaves both esker and sand plain to their fate. The special ques- 
