624 THE JOURNAL OF GEOLOGY. 



englacial material derived from any given elevation would stand 

 a chance of becoming superglacial, because there must be pro- 

 gressively more and more surface-melting, in order to bring the 

 uppermost portion of the englacial material to the surface. 

 Meanwhile basal melting has been going on, and will have 

 brought some of the englacial material to the bottom of the ice, 

 that is, to a subglacial position. Because of its thinness, there- 

 fore, the marginal part of the ice-sheet was likely to secure more 

 englacial material capable of becoming superglacial, than any 

 other part. 



The ratio of surface melting to basal melting is probably 

 greater at the margin of the ice than elsewhere, so that the 

 upper surface of the ice is here lowered more rapidly than else- 

 where. It follows that, as a result of surface melting, a greater 

 proportion of the englacial material acquired by the margin of 

 the ice would be likely to become superglacial, than of that 

 acquired by any other part. Considered from the standpoint of 

 the ice, there are, therefore, two reasons why the marginal part of 

 an ice-sheet should possess more englacial-superglacial drift than 

 any other part. 



There is another set of reasons why superglacial material, 

 derived from an englacial source, must be more abundant near 

 the margin of the ice than elsewhere. They relate to the surface 

 over which the ice passes. The passage of glacial ice over a 

 region of rough topography tends to smooth it. It is when the 

 ice first invades a region that its topography is roughest. Later, 

 after the passage of much ice, the rugosities of surface have 

 been reduced, and from the smoother surface the ice is able to get 

 less detritus. Furthermore, quite apart from considerations of 

 topography, it is when the ice first invades a region that there is 

 most loose surface material in a condition to be removed. Later, 

 after longer passage of the ice the materials antecedently 

 loosened by surface agencies have been taken away, and any 

 further acquisition the ice may make must be made from the more 

 solid rock beneath. Considered from the standpoint of the sur- 

 face over which the ice passed therefore there are two valid rea- 



