56 WATER. 



rocks, first to fragments, then to powder. During its 

 frozen state, the ice may serve as a cement, and hold 

 together the disintegrated particles ; but with a 

 thaw, the slightest force, currents of water, or the 

 eifect of weight, suffices to carry the fragments to 

 the bottom of the valley; and the rubbing and 

 motion to which these are exposed in torrents, tend 

 to break them still smaller, and reduce them to 

 sand." 



It is well known that water, by an apparent ex- 

 ception to a general law, expands with great force 

 when freezing. Over a large extent of surface the 

 effect may be very considerable, and when boulder- 

 stones lying in shallow ponds become partially 

 enveloped in the ice, they must feel the effects of 

 this expansion, and be driven toward the shore; 

 since the force must always act in that direction. 

 As nothing exists to bring back the rock to its orig- 

 inal position, the ultimate effect must be to crowd 

 it entirely out of the water ; and perhaps to this 

 cause we may impute the fact, that on the margin 

 of some ponds we find a ridge of boulders, while 

 the bottom, to a considerable extent, is free from 

 them.-^ 



" There is, however, one agent of excavation that 

 still operates to some extent, and this is called ice- 

 floods. Their effect is most powerful upon the 

 smaller and more rapid streams. Whoever has not 



1 Geological Survey of Massachusetts, by Dr. nitchcock. 



