G. H. Stone — Glacial Sediments of Maine. 143 



We have, then, to consider not only the thick sheets of 

 alluvium covering the bottoms of the valleys above 230 feet, 

 but also the fluviatile off-shore sands deposited at the same 

 time. These are now exposed for our examination. Alto- 

 gether they represent a very large body of sediment. And we 

 rind proof that it was all laid down within a very brief period. 

 If this vast amount of fine sediment was derived from the till 

 after the ice had all melted, we could not fail to find the 

 ravines and other traces of this enormously rapid erosion. The 

 uplands ought to have been dissected into a condition ap- 

 proaching that of the bad-lands of the west. 



The most probable interpretation from these and many other 

 facts is the following : 



While it is true that the morainal matter contained in the lower 

 portion of the ice must often have been irregularly distributed in 

 consequence of local causes, yet on the average it is probable that 

 the proportion increased near the ground. During 1 he final melt- 

 ing a larger quantity of till would each year become exposed on 

 the surface of the ice. During the melting of the last 100 

 feet very little clear water could have escaped from the ice. 

 The surface was covered with muddy till. The sub-glacial 

 tunnels were roofed with muddy ice and the same muddy ice 

 was revealed in the walls of superficial channels and crevasses. 

 The melting of each minute ice grain unlocked more mud. 

 The rains and melting waters washed away this unconsolidated 

 ooze far more easily than they could erode till already com- 

 pacted and settled. Erosion accomplished by waters issuing 

 from beneath is a very different process from ordinary erosion 

 by rains and streams. Moreover this erosion was more de- 

 pendent on the ice-conditions than on the slopes of the under- 

 lying land surface. Here, then, was an ablation diffused over 

 the whole surface. While post-glacial erosion of previously 

 deposited till would be largely localized in the steeper lateral 

 valleys, late glacial erosion would be far more diffused. The 

 coarser or residual matter of the till would be left over the 

 whole surface, not in the ravines and smaller valleys. 



It is a well known law of melting ice that the parts quickest 

 drained of their melting waters melt slowest. So on the melt- 

 ing ice-sheet, wherever the waters gathered, there the melting 

 proceeded most rapidly. This is the principal cause of the 

 great enlargement of the river channels from those of narrow 

 osar to broad osar- plains. 



The facts in Maine reveal very much of the hydrography 

 of the ice-sheet. It is impossible here to go into details or to 

 discuss the effect of hills on the overlying ice- surface. We 

 have also to consider the effect of the till that accumulated on 

 the ice-surface in consequence of the melting of the upper ice, 

 in protecting the ice beneath it from the heat of the sun and 



