342 NATURE AND MAN. 



that a berg had the regular shape of a box, its entire depth from 

 its upper surface to its base must be ten times its height above the 

 sea-level. Consequently, if the latter be 200 feet, the entire height 

 of the mass would be 2000 feet, which might thus be assumed to 

 be the thickness of the ice-sheet from whose margin it was detached. 

 This estimate must not be accepted, however, as other than ap 

 proximative. The dimensions of these bergs vary greatly. Those 

 seen from the Challenger were generally from one to three miles 

 long ; but single bergs are reported of seven or even ten miles in 

 length ; and an enormous mass of floating ice, probably composed 

 of a chain of bergs locked together, forming a hook 60 miles long 

 by 40 broad, and inclosing a bay 40 miles in breadth, was passed in 

 1854 by twenty-one merchant ships, in a latitude corresponding 

 to that of the northern coast of Portugal. 



The upper part of the ice-cliff that forms the exposed face of 

 the bergs is of a pale blue, which gradually deepens in colour 

 towards the base. When looked at closely, it is seen to be 

 traversed by a delicate horizontal ruling of faint blue lines 

 separated by dead-white interspaces. These lines preserve a 

 very marked parallelism, but become gradually closer and closer 

 from above downwards, their distance being a foot or even more 

 at the top of the berg, but not more than two or three inches near 

 the surface cf the water, where the interspaces lose their dead 

 whiteness, and become hyaline or bluish. There can be no doubt 

 that this stratification is due to successive accumulations of snow 

 upon a nearly level surface, the spaces between the principal blue 

 lines probably representing approximately the snow-accumulations 

 of successive seasons. The direct radiant heat of the sun is very 

 considerable even in these latitudes, so that the immediate surface 

 of the snow is melted in the middle of every clear day; and the 

 water, percolating into the subjacent layers, freezes again at night. 

 The frequent repetition of this process will convert a very con 

 siderable thickness of snow into ice ; the blue transparent lamella: 

 being the most compact, whilst the intervening white veins are 

 rendered semi-opaque by the presence of air-cells. And it is 

 obviously the compression which these undergo that causes the 

 approximation of the blue lines, and the change to a greater com 

 pactness and transparence in the intervening layers, towards 



