STUDIES FOR STUDENTS. 619 



irom a basal to a superficial position, without actually rising. 

 In a mountain glacier, the bed of which is steeply inclined, basal 

 drift would reach the upper surface of the ice if it were carried 

 forward horizontally, or even if its forward path of motion 

 declined at any rate less than that of the bed of the glacier. It 

 has frequently been argued that this is the actual condition of 

 things. While this conception does not involve a rise of mate- 

 rial in terms of absolute altitude, it involves the rise of material 

 through the ice which embeds it, or the rise of ice which embeds 

 drift through that which surrounds it. That basal drift may rise 

 through its embedding ice, or that ice embedding drift may rise 

 through other ice in any such way as would be necessary to 

 bring basal drift to the surface, has not been demonstrated. 

 That stony material may be crowded up some slight distance 

 into the ice from below by the help of other material beneath 

 the ice, is readily understood, but in the present state of knowl- 

 edge there is little warrant for counting upon the rise of material 

 from the bottom of a glacier to its surface. If such rise were a gen- 

 eral fact, there should be much more evidence of wear upon super- 

 glacial boulders than has yet been found. Indeed, the total absence 

 from most Alpine glaciers of surface boulders showing any trace 

 whatsoever of glaciation, seems to go far toward settling in the 

 negative the question of the rise of basal drift through the ice. 



The material which was superglacial at the outset would be 

 likely to remain superglacial to the end, unless it fell into cre- 

 vasses, or unless some other untoward accident befell it. The 

 englacial-superglacial material must likewise have remained at 

 the surface after once reaching it, unless it suffered some acci- 

 dental fate. 



Broadly speaking, the oldest ice of a glacier is at its lower 

 end. Since the lower end of a glacier has had a longer time 

 than any other part in which to gather superglacial material, and 

 since surface melting has here been greatest, making possible the 

 transfer of more drift from an englacial to a superglacial posi- 

 tion at this point than elsewhere, it follows that the lower end of 

 a glacier is likely to have more superglacial drift than any 



