AND TRANSPORTATION OF BOULDERS. 181 



In the formation of icebergs sucli as I have to describe, a rocky or 

 precipitous coast, witb deep water close to the shore are requisite, and 

 the precipices should run in a direction at right angles, or nearly so, 

 to the prevailing winds, the face of the rocks being turned away from 

 the direction of the wind. 



During winter, in the Arctic regions, nearly all the gales of wind are 

 from N. and N.W, These gales are very frequent, and either accom- 

 pany or immediately follow every fall of snow, so that where there are 

 precipices having a southern or eastern aspect, large drifts of snow are 

 formed under their lee. If the water is deep, as it generally is, at 

 the foot of these cliffs, the weight of snow forces the ice on which it 

 rests down into the water, submerging with it the superincumbent 

 snow, and all this submerged snow becomes frozen into a solid mass of 

 ice. Every gale adds a fresh stratum of snow, so that in spring there 

 may be a snow-drift more than one hundred feet in depth;* which, 

 under such circumstances as I have described above, is from the same 

 process of submergence and solidification. The process of submer- 

 gence, however, has its limits, and above this the snow-drift goes on 

 accumulating, in like manner with every succeeding gale. When the 

 summer comes, the surface of these drift-banks is thawed, and the 

 water filters through the snow underneath, which, being of a much 

 lower temperature than 32°, causes the whole to freeze, in like manner, 

 into a solid mass of ice. 



A large portion of this ice, which once was snow, if the summer has 

 not been a peculiarly mild one, remains until the following winter, 

 when a fresh deposit takes place, and thus from year to year one accu- 

 mulation succeeds another until the whole height from the base to the 

 top of the cliff is filled up. This extends in a sloping direction (get- 

 ting gradually thinner,) out to the sea, reaching to a greater or less 

 distance according to the height of the cliff and other favouring cir- 

 cumstances. From time to time pieces break off from this accumulated 

 mass of solid ice, and when the water is deep enough they are floated 

 away in the form of icebergs. Frequently the ice separates from the 

 cliff itself, and occasionally tears away large pieces of the rock along 

 with it. Of this fact I saw some striking examples, whilst sailing in a 

 ship through Hudson's Straits, on my way to England, in the autumn 

 of 1854. We were close to the north coast of the Strait, which is 

 very precipitous, and I could distinctly see large icebergs aground near 



• I have seen a snow-drift of more than twenty feet deep formed in one night. 



