ESKERS IN THE VICINITY OF ROCHESTER, NEW YORK. 191 



sion will gradually cease, and aggradation will take its place, building 

 an esker ridge in the stream channel. The stream will derive its 

 supply of debris from subglacial or englacial drift that has become 

 superglacial by shearing, by the overriding of the sedentary ice cap 

 that had accumulated previous to the invasion by the ice from the 

 center of ice dispersion, the latter carrying upward its basal load, 

 by the upturning of layers, phenomena observed especially along the 

 front of the Greenland glacier, by surface ablation revealing the 

 englacial and basal material in the ice, and by erosion of nunataks 

 projecting through the ice and of ledges projecting into the. ice. 

 Upham has thought that much drift has become englacial even in a 

 short distance from the point of origin on a relatively flat surface. 

 Stone formerly entertained the idea that much of the drift of the 

 Maine eskers was englacial in source. 



With the cessation of corrasion the esker channels may be 50, 

 100 or more feet above the ground. The question now arises as to 

 how to get the esker down on the ground without the destruction 

 of its ridge-like character. It is necessary that the banks of the chan- 

 nel — the retaining ice walls — be preserved if the esker form is to be 

 maintained. Hence the bottom of the channel must be lowered as 

 rapidly as surface ablation lowers the neighboring ice surface. Cor- 

 rasion has ceased so recourse must be had to the melting of the ice 

 at the bottom of the channel by the sluggish waters of the stream 

 percolating downward through the porous gravels. Russell has 

 described lakelets on the moraine-covered marginal zone of the 

 Malaspina glacier, 50 to 100 feet deep, rarely more than 100 feet in 

 diameter, with bottoms covered with drift constantly augmented by 

 the addition of fresh material from top and sides. These lakelets 

 are deepened as fast or faster than the surface of the ice is melted, 

 by convection, the warmer, denser (near 39° F.) waters of the sur- 

 face sinking and percolating through the porous materials covering 

 the bottom displacing the colder lighter water, near the freezing 

 point, at the bottom. Thus the lake bottom is lowered, and the walls 

 of ice undercut even below the water surface. In a similar manner 

 the bottom of the esker channel is lowered by the melting of the ice 

 as the warmer surface waters percolate through the esker deposit, 

 and the ridge finally comes to rest upon the ground. This is in brief 



