It follows from the observations that: 



4. The carbonate content in sea ice is almost constant, increasing only slightly with the 

 depth of the ice layer which is explained by the fact that the carbonates precipitate out of sea water 

 and settle upon the cell walls almost simultaneously with the beginning of ice formation, 



5. Inasmuch as the chloride content in sea ice changes very much (decreasing along with the 

 age and condition of the ice), the ratio A:C1 can fluctuate within considerable limits, attaining its 

 maximum in many-year old ice. 



Thus, direct measurements of the salt composition of sea ice support the selective character 

 of the ice formation process established by Ringer, and the correctness of Petterson's and Ringer's 

 conclusions that during ice formation, a certain excess of chlorides should be observed in the sea 

 water from which the ice had separated, and, conversely, water in which the ice had melted should 

 show an excess of sulphates and above all, carbonates. 



Therefore, waters of arctic and antarctic origin (i. e. , where intensive ice formation occurs) 

 obviously should be outstanding due to a lower alkaline coefficient. 



Determinations of the alkalinity of the surface waters in the northern part of the Kara Sea, 

 conducted by Chigirin during the expedition on the Sadko in 1935 have shown that the alkaline co- 

 efficient decreases with the increase in the salinity, as can be seen from table 38. 



TABLE 38. THE ALKALINE COEFFICIENT IN THE WATERS 

 OF THE NORTHERN PART OF THE KARA SEA 



50/00.... 29-30 30-31 31-32 32-33 



(A:S)104... 755 694 691 688 



LITERATURE: 62, 95, 104, 166. 



Section 60. Specific Heat 



otto Petterson was the first to direct attention to the fact that the heat of fusion and thermal 

 expansion of sea ice discloses anomalies in comparison with the same properties of fresh water ice. 



Thus, in studying the heat of fusion of artificially prepared sea ice within the temperature of 

 limits of -6° to -9°, Petterson obtained the heat of fusion of 60.5 g-cal, for ice of 20 o/oo, and the 

 heat of fusion of 49.5 g-cal for ice of 40 o/oo salinity. In Petterson's eijqjeriments, very salty ice 

 continued to Increase in volume during a decrease in temperature. 



In analyzing Petterson's experiments, Krummel arrived at the conclusion that the abnormal 

 thermal expansion of sea ice should be ascribed to the brine contained in the ice. If the tempera- 

 ture of sea ice is lowered, pure ice would separate from the brine contained in the cells, which is 

 related to the great increase in volume. This increase in volume in the presence of great amounts 

 of brine exceeds the natural increase in volume during a decrease in temperature. 



Malmgren confirmed the correctness of Krummel's hypothesis by his investigations during 

 the expedition on the Maud (1922 to 1924). Additionally, he was the first to point out that the high 

 specific heat of sea ice near the freezing temperature is created by the fact that at temperatures 



150 



