PREPAEATIONS FOE WINTERING. 341 



was smooth and firm, but soon we found it damp and 

 slippery, even at 13° to 18° of cold. We now found that 

 this damp coating was strongly saline, and that in places, 

 as, for example, quite in the inside of the creek, where an 

 overflow of water from the shore, or from cracks or holes, 

 was not to be thought of. Later on, after a storm, we 

 found on the surface of the newly-formed ice in front of 

 the creek, a white covering of salt ; more than this, we 

 all found from experience that the snow lying on the 

 ice, even at a great depth, was more or less brackish, 

 so that what was wanted for melting had to be brought 

 from land. As it is now a known fact that in the freezing 

 of sea-water the salt is expelled, and nothing but pure 

 water is found in the ice when formed, the appearance 

 of salt on the surface puzzled us. The same appearance 

 has already been noticed by Wrangell. It is difficult 

 to account for this in any way but the following, viz., 

 that part of the salt water concentrated under the ice 

 being hard to freeze, it is by some power or other forced 

 through the pores of the ice to the surface. The ice 

 in the harbour itself retained its smooth surface and 

 its first appearance, though in front of it the young ice, 

 soon after its formation, was repeatedly, and often to a 

 large extent, thrown one piece upon another, and that 

 fi^equently in sharp detached corner fragments. To the 

 south of the harbour, and the east of Walrus Island, this 

 appearance was most perceptible. The cause of this, as 

 well as of the fractures in the ice, is to be found in the 

 crowding together of heavy masses, which are set in 

 motion either by the current or the wind. In the same 

 way the young crowded ice, when the coast lies before it, 

 is pushed on to it ; or, if the bank is steep, is crushed 



