PHENOMENA AFFECTING RIVER THEORY 233 



When the ice-block was detached from the front of the glacier the tem- 

 perature of the waters and the detritus was near 32 degrees Fahrenheit. 

 The mass of buried ice and its detrital cover were practically at freezing 

 temperature. For a long time, while the ice-front was in the neighbor- 

 hood, the accumulating deposits were very cold. During all the time, 

 certainly thousands of years, that the ice-sheet lay across the Connecticut 

 Valley, the outflow from the glacier was a cold, bottom current or drift 

 southward down the inlet. Then even after the glacial flood ceased the 

 more quiet water at the bottom of the inlet could not readily be warmed 

 above the maximum density, whatever that was, depending on the salinity. 

 Circulation of the water in the cold gravels or sand or clay at the bottom 

 of the standing water was certainly sluggish and ineffective. 



All the conditions would seem to favor the preservation of the buried 

 ice-block, and it seems probable that the larger blocks and those buried 

 more deeply did not melt until the locality was lifted out of the water 

 and exposed to the sun and atmosphere. Doubtless some ice-blocks did 

 melt in time for detrital filling of the depressions, in which case we find 

 as kettles only the few basins that were very late in forming. 



Such kettles as lie in areas of submergence that were never subjected 

 to stream flow or currents, or even to currents which carried no detritus, 

 might have formed quickly and yet have remained unfilled. 



Marine Submergence 

 theoretic plane 



The Connecticut Valley can not properly be studied alone. Any up 

 and down movement of the land must have involved wide territory. The 

 other valleys of New England and the Hudson-Champlain must be used 

 in comparison. 



In the Hudson-Champlain Valley, with its greater dimensions and 

 larger drainage area, the summit-level phenomena are stronger and more 

 definite than in the Connecticut Valley. They afford a clearer history 

 and stronger argument for the reasons that no one could suggest river 

 work and because the great valley extends north at low altitude, so as to 

 connect New York Bay with the Saint Lawrence depression. The phe- 

 nomena due to marine submergence can be compared with present sea- 

 level at both ends of the valley and the marine plane be determined more 

 accurately. The Connecticut Valley rises rapidly in its head-waters sec- 

 tion and the sealevel waters extended north only to the mouth of the Pas- 

 sumpsic River. 



The uplifted and tilted plane of the sealevel waters is indicated in the 



