MEETING OF SNOW RESTING ON THE EARTH. 315 



"the passage ways between the houses were roofed over 

 with flat slabs of extremely hard snow, these are now bind- 

 ing with their own weight and that of the superincumbent 

 snow ; but the latter solidified sufficiently to form its own support j 

 and we can now remove the original roof without danger." : 



This as we conceive exactly describes the process by 

 which the cavities forming at the ground level are 

 supported, the decay of the snow there being of course 

 effected by the drip of the slowly melting mass, which 

 is always wasting over the whole extent of its under 

 surface, while it is renewed as constantly by fresh falls 

 upon its upper. It has been found that snow resting 

 upon the ground invariably has a tendency to melt, 

 the moisture quickly evaporating and becoming dispersed 

 into space, or forming a coating of ice upon the snow 

 roof above. Thus, while working in the snow houses of 

 the expedition already referred to, Dr. Moss, surgeon 

 on board the Alert, states that he observed that " any 

 small quantity of snow left on the floor, evaporated; 

 the moisture rising and collecting on the inside of 

 the dome of the house, as rime;" f and there can be 

 no doubt that the action going on upon the earth's 

 surface, is of a similar nature. 



But why should snow melt, when it touches the 

 ground? Is it because a certain amount of caloric 

 is being constantly exhaled from the earth's surface, 

 proceeding from the stores of internal heat, which \ve 

 know exist within the terrestrial sphere? This is 

 certainly the interpretation we are disposed to place 

 upon this phenomenon. This melting of the snow 

 however proceeds irregularly, certain places from a 



* Voyage to the Polar Sea, by Capt. Sir George Nares, R.N., 1878, 

 Vol. i, p. 221. 



f Ibid., Vol. i, p. 225. 



