THE ESKERS OF NOVA SCOTIA. PREST. 383 



lines of drainage as in Nova Scotia and Maine. To what 

 natural phenomena can we turn to find a power that would 

 drive a single drop of water over a watershed 200 to 440 

 feet liigh, and finally carry millions of tons of debris to the 

 same height. What was the influence that guided the course 

 of the first tiny crack directly across the line of drainage of 

 Nova Scotia. If we admit the possibility of a transverse 

 open crack we kill the siphon theory of subglacial action, 

 as the crack must be filled 400 feet deep before running over 

 a watershed of that height. 



Another noteworthy fact is that on the plains or swamps 

 or Maine many eskers are partly composed of sand. This 

 indicates a moderate current. It needs a rapid current under 

 tremendous pressure to drive large rocks over watersheds 200 

 to 400 feet high. This is also absolutely fatal to the subglacial 

 theory of esker formation. 



In view of these facts we are forced to admit the utter 

 impossibility of accounting for the existence of eskers by 

 any form of marine, fiuvatile, or lacustrine agency. 



New Theory of Esker Formation. — But there have been 

 conditions under which the distorted stratification mentioned 

 by Chamberlain and Salisbury was inevitable. In fact it 

 could not be otherwise. This embodies a theory of formation 

 of eskers different from any yet considered and will be detailed 

 later. 



Other Objections. — Eskers are irregular in size, height and 

 character. Unlike rivers, eskers never increase regularly 

 in size the further they go, except in river valleys. This fact 

 furnishes one of the strongest clues to their origin. If eskers 

 were formed in subglacial channels we would always find the 

 smallest ridges near the surface of the streams and on high 

 lands and the greatest spread of washed material on the plains. 

 That this is not so is strong evidence that eskers were not 



