260 THE ANTARCTIC. 



and in March the state of the ice does not greatly differ 

 from that of April. As has already been observed, these 

 rough statistics apply only to the regions visited by com- 

 merce ; whilst very little, indeed hardly anything, is known 

 about the state of the ice in higher latitudes — especially 

 in winter. 



The gradual drift northwards leads pretty rapidly to 

 considerable changes in the shape of the bergs. Rains 

 and the increased temperature of the air melt portions of 

 the ice towering above the surface of the sea, and as 

 the water that penetrates the rifts in the ice freezes 

 again, it expands and breaks off fragments near the 

 edge with loud reports ; along the water-line the 

 breakers hollow out cavities, growing larger and deeper 

 till at last the overhanging roof falls in, and finally the 

 sea water forms cavities all round the berg. The 

 material of the iceberg being ice formed from snow, it 

 follows that the surface water of the sea cannot melt 

 this substance unless its temperature is 32° F. or more ; 

 whilst in the water with a temperature lower than the 

 freezing-point, melting cannot take place in the higher 

 latitudes. The breakers, combined with the melting 

 process, form submarine ice terraces, proceeding from 

 and surrounding the berg, which, being hidden, may 

 easily become dangerous to navigation. If the destruc- 

 tive forces have removed a large portion of one side 

 of the berg, then the centre of gravity of the whole 

 mass is changed, and the berg assumes a different 

 position. Thus arise those boldly jagged pinnacles 

 observed in the lower latitudes, whose lofty summits 

 are balanced by a much larger, but far less deep 

 reaching mass beneath the surface of the water. It is 

 seen, then, how unsafe it is to assume, as Dr. Croll 

 for example has done, that the submarine depth of the 

 ice is nine times as great as is the height above the 

 surface of the water. 



