122 THE FORMS OF WATER IN 



goes on, and some curious effects are due to it; but 

 not, I think, the one here ascribed to it. The deepening 

 of a shaft implies a quicker melting of its bottom than 

 of the surface of the glacier. It is not easy to see how 

 the fact of the solar heat being first absorbed by water, 

 and then conveyed by it to the bottom of the shaft, 

 should make the melting of the bottom more rapid than 

 that of the ice which receives the direct impact of 

 the solar rays. The surface of the glacier must sink 

 at least as rapidly as the bottom of the pit, so that 

 the circulation, though actually existing, cannot produce 

 the effect ascribed to it. 



§ 46. Consequences flowing from the foregoing Properties 

 of Water. Correction of Errors. 



315. I was not much above your age when the 

 property of water ceasing to contract by cold at a 

 temperature of 39° Fahr. was made known to me, and I 

 still remember the impression it made upon me. For 

 I was asked to consider what would occur in case this 

 solitary exception to an otherwise universal law ceased 

 to exist. 



316. I was asked to reflect upon the condition of a 

 lake stored with fish and offering its surface to very 

 cold air. It was made clear to me that the water on 

 being first chilled would shrink in volume and become 

 heavier, that it would therefore sink and have its plac* 4 



