2 52 THE JOURNAL OF GEOLOGY, 



what is the effect of the flowing of a glacier down into a body 

 of water upon the enlargement of the subglacial tunnels. In 

 such a case the tunnels and all crevasses opening into them are 

 permanently filled with water up to the level of the surface of 

 this body of water. But it is by the crevasses that the waters 

 of local melting get down into the subglacial tunnel. The per- 

 manent water in the crevasses is at the temperature of 32°. As 

 the waters of surface melting in the region whose basal ice is 

 submerged in the sea or other body of water pour into the 

 crevasses they cannot at once fall to the ground and enter the 

 tunnels, but they fall into the water in the crevasses that already 

 fills them to the level of the permanent body of water. The 

 large streams find their way pretty directly into the tunnels, 

 but all the smaller streams and trickles become so mixed with 

 the cold waters in the crevasses that their heat, instead of being 

 consumed in enlarging the tunnels, is largely expended in melt- 

 ing the ice walls of the crevasses above the level of the tops of 

 the roofs of the tunnels. 



Thus the flowing of a glacier down into the sea interferes 

 with the natural transfer of heat beneath the ice whereby the 

 tunnels are enlarged in large part. But the supply of surface 

 waters is the same over the area whose base is submerged as 

 elsewhere. The conclusion follows, that as we go toward the 

 distal extremity of a glacier that flows into a body of water, the 

 supply of drainage waters would be increased more rapidly than 

 the tunnel capacity. This would result in increased velocity of the 

 rivers, with a corresponding increase in power of transportation. 

 In other words, they would do just as the osar rivers of Maine 

 did as they approached the coast — would deposit sediments at 

 longer and longer intervals, and in smaller quantities, and finally 

 would sweep their tunnels free from all sediments. 



Now it is certain and inevitable that the submergence of the 

 basal ice should restrict the enlargement of the subglacial tun- 

 nels, yet it is an open question whether this was sufficient to 

 account for the peculiar development of the coastal gravels. 



We have seen that these changes take place within a belt not 



