28 W. UPHAM giants' kettles eroded by MOULIN TORRENTS 



mill, is applied to a vertical tunnel, melted at first by the waters of the 

 surface trickling into some very narrow crevasse that has just begun to 

 open, until, after enlargement by this dissolving action, it receives some- 

 times a large stream, such as could not be waded, pouring with a thun- 

 derous roar down a cylindric shaft to the rock floor under the ice. But 

 the streams that bored these potholes or giants' kettles were very small, 

 and were only very scantily drift-laden, in comparison with the glacial 

 rivers which formed the great esker ridges. 



Rock exposures adjoining glacial potholes are often unmarked by 

 other waterwearing ; but in every present stream having falls and 

 eroding potholes, larger spaces are irregularly worn and channeled. 

 The rock kettles of moulin formation, found where no stream now 

 exists nor can be supposed to have ever flowed except when the coun- 

 try was ice-enveloped, are th-e predominant or the only form of water 

 erosion in their vicinity ; but at the falls of ordinary streams, potholes 

 are exceptional or a subordinate feature among more extensive grooving 

 and other fantastically waterworn sculpture. It is evident, too,' that 

 glacial planation, ensuing after the moulin origin of the giants' kettles, 

 although probably in many places effective to intensify this contrast, 

 can not generally be its chief ex))lanation, which is rather to be found 

 in the protection afforded by the ice covering the rock contiguous to the 

 base of the moulin. The rebounding water, indeed, welling up from 

 one side of the rock kettle, may perhaps have usually flowed aw\ay, for 

 its immediate exit, in an englacial tunnel, or at least with some drift 

 between it and the rock. The conditions of erosion of the giants' kettles 

 prevented or minimized contiguous waterwearing, which, on the other 

 hand, is favored and predominant wherever potholes are made by sub- 

 aerial streams. 



Another way in which subglacial streams })robably sometimes or often 

 acted to erode i)otholes has been i)ointed out by Mr T. T. Bouve^-^^ and 

 Professor George H. Stone,t who well remark that the subglacial waters, 

 after falling down the crevasses and moulin shafts, would flow rapidly 

 away upon the surface of the bed-rock, and there might sweep past the 

 mouths of potholes during the process of their erosion, supplying all the 

 current needed for their further deepening by the whirl thus given to 

 the water and stones at the bottom. For the largest and dee{)est of the 

 giants' kettles, however, to be here noted as discovered in many localities 

 of glaciated countries, I can not doubt that the pothole was cut down 



*" Indian potholes, or giants' kettles of foreign writers," Proc. Boston Society of Natural His- 

 tory, vol. 24, pp. 219-220, April, 1889, with ensuing discussion by Warren Upham, pp. 220-228. 



t " Glacial gravels of Maine and their associated deposits," U. S. Geol. Survey, Monograpii ^i, 

 1899, pp. 324-:«0. 



