344 THE BEGOLITH 



part, over the surface, forming superglacial streams, or in part 

 upon the surface of the ground beneath as subglacial streams, 

 of which last the river Ehone of to-day is a good example. 

 Presumably also a portion of the water became concentrated 

 and flowed for short distances in the mass of the ice itself, 

 forming thus englacial streams. In all cases the running water 

 would collect, reassort, and variously modify the rock debris 

 found either in immediate connection with the ice itself or at 

 its extremity, in the terminal moraines. There were thus 

 formed hillocks and ridges or low fan-shaped masses of *' modi- 

 fied drift." The sand, gravel, and boulders which collected in 

 the troughs of superglacial streams would, on the final melting 

 of the ice, be deposited as ridges running essentially parallel 

 with that of the movement of the ice on which they formed. 

 Such are known as eshers, or osars. Other deposits closely 

 resembling these and sometimes confounded with them, but 

 formed, it is believed, only by swift and changeable currents 

 near the frontal margin of the ice, present often a rude and 

 disturbed and distorted stratification, and are known as kames. 

 They differ from the eskers in their outlines as well as positions 

 with reference to the glacier from whence their materials were 

 derived, being as a rule in the form of hills, rather than ridges, 

 and with their longer axes at right angles with that of the ice 

 motion. 



Beyond the margin of the ice and its terminal moraines are 

 found still other loosely aggregated deposits of a similar hetero- 

 geneous nature which are likewise due to swiftly rimning water 

 caused by the melting ice. Such, according to their position 

 and form, are known as valley drifts moramic or frontal aprons^ 

 and overwash plains. 



The thickness of these glacial deposits varies greatly, as has 

 been already indicated. Variations of upwards of a hundred 

 feet may occur within the limits of even less than one square 

 mile. Professor Newberry estimated that the area south and 

 west of the Canadian highlands covered with glacial drift was 

 not less than 1,000,000 square miles, and that its average 

 depth would not be less than 30 feet. Other estimates on 

 deposits in Ohio, Indiana, and Illinois give an average thickness 

 in these states of 62 feet. In extreme cases the deposit has 

 been found to extend to a depth of 300 to 500 feet. Bell has 



