SOME GLACIAL WASH-PLAINS. 97 



a rather coarse type. We have here repeated the cross- 

 section of frontal or snbmarginal deposits which appears 

 so distinctly on Nantucket, viz. : going from south to 

 north, (1) an outwash plain; (2) the ice-contact, a ter- 

 race overlooking low ground which may be designated as 

 (3) the fosse, occupied by undifterentiated drift, fre- 

 quently bouldery ; and followed by (4) morainal mounds, 

 with till and underlying wash, to which succeeds on the 

 north the ordinary ground moraine. 



If we suppose that the morainal mounds were built at 

 the fi'ont of the ice when its edge lay on their northern 

 side, then we liave no contemporaneous wash deposits 

 attributable to the discharging streams. It is more 

 rational to suppose that the morainal mounds accumulated 

 under the ice when its front lay along the wash-plain 

 heads, thus correlating extraglacial plain-building by 

 drainage with intraglacial mounding of till by forward 

 ice movement. 



The superposition of till on stratified drift in these 

 morainal mounds in the intraglacial field has elicited two 

 alternate hypotheses, viz. : 1. The deposit is due to the 

 overriding of a small gravel outwash fan built on the site 

 of the mounds in a stage of the ice retreat immediately 

 preceding the Bridgewater stage, when the ice front was 

 along the northern edge of the present morainal area. 

 Outwash fans tend to occur in isolated forms. The over- 

 riding action of the ice would mantle them over with till 

 and destro}^ the form of the original deposit. 2. After 

 a wash-plain has grown up at the ice margin, it forms a 

 mass resisting the forward motion of the bottom ice. 

 The upper ice would tend to shear off from the stagnant 

 prism lying behind the sand-plain head. At the point 

 where the bottom of the live ice began to move up over 

 the inclined i)lane thus formed, the subgUicial till would 



