Figure 35. Formation of ice under ice in the area of the Sedou drift. 



A reverse distribution was observed in separate instances: a thin layer of sub-ice, fresh- 

 water ice was separated from the lower surface of the sea ice by a layer of ice porridge. Evidently, 

 the sub-ice is not distributed over the entire ice cover but rather represents a local phenomenon 

 and collects under the ice areas where snow water drains off under the ice, and is held for a com- 

 paratively short time— from five to ten days. 



According to observations on the Sedov , near one of the ice-measuring rods which were set 

 up, the ice thickness increased by 28 cm from 30 July to 10 August 1939. 



It was clarified that in the given instance, a decisive role was played by an accidental condi- 

 tion: under the ice on which the rods stood, a singular ice shed formed consisting of hummocks 

 reaching down deep (figure 35). These hummocks formed a locked circle which hindered the dis- 

 tribution of the fresh water draining down to a wide sub-ice area. A singular ice box developed in 

 which the melt water turned to ice very rapidly. 



In the summer of 1938 no formation of ice under ice occurred, according to the observations 

 on the Sadho and the Sedou. Thus, these observations affirmed the supposition that the sub-ice 

 under ice of noticeable thickness is a local condition. 



LITERATURE: 11, 62, 77, 88, 107. 



104 



