312 rETTKRSSON, ON WATER AND ICE. 



cryohydrates will escape to the surface of the ice througb 

 these channels and engender tliose unfrozen brines, whicb 

 attracted the attention of Nordenskiöld and his followers. 



The sea-ice is certainly no homogeneous body. I wonld 

 compare it to a crystalline rock, for example a granite, which 

 contains a number of different crystalHzed combinations. 

 Every one of the constituents: the feldspar, the mica, the 

 sihcia is Hable to decomposition in its own peculiar way. The 

 products of this decomposition escape in the form of aqeous 

 Solutions, until the remainder, which can not be decomposed 

 any more in situ, is mechanically carried off as clay and sand 

 with the glaciers or with the water of the rivers to give birth 

 to new geological formations. Thus the crystallized constitu- 

 ents of the sea-ice, one by one, are attacked by decomposition, 

 even before the arctic summer bas broken the ties, which bind 

 the ice-floe to its birth-place. The rest, which can not be 

 attained by metamorphosis in these high latitudes, viz. the- 

 pure ice and the most perdurable cryohydrates, is caught by 

 the mechanical force of the ice-current, which carries the skel- 

 etons of the polar ice back to southern latitudes to unite 

 again with the warmer parts of the ocean. 



CHAPTER 6. 



On the latent heat of fresh and salt water, 

 conclusions. 



Pure water. 



The latent heat of pure water below zero is, as already 

 mentioned, theoretically calculated by Person. From my ex- 

 perimental verification ' of his formula the following table is 

 cited: 



when the temperature approaches the melting-point, and in sea-ice by fall 

 of temperature (see the illustrations on plate 21 & 22). 



' The following table is reproduced here, because the subject is of 

 essential importance to the description of the physical properties of pure 



