394 Alf Trolle. 



The heat conditions in the deeper parts of the sea outside the 

 coast had the following influence on the formation of the ice. 



When the tidal wave passed in over the barrier at the mouth 

 of the fjord, mixing processes were started which as a rule 

 led to transport of heat and chlorides from the deeper to the upper 

 layers and this heat was led by convection currents and mixing to 

 the surface and counteracted the formation of the ice. 



In this we may seek for the explanation of the striking difference 

 between the steadily ascending course of the curve of ice thickness 

 in 1907 and its course in 1908, when it began to lessen already in the 

 month of February (see PI. XXIII). 



The cause cannot lie in the difference in the air temperature, as 

 the air was on an average colder in the spring months of 1908, when 

 the thickness of the ice was decreasing, than in the corresponding 

 months of 1907, when it was increasing. But we can find the cause 

 in this, that the sea in the spring of 1907 was cold and of uniform 

 salinity almost to the bottom and no heat, therefore, was carried 

 to the surface from the bottom-layers, (nor was there any transport 

 of chlorides), whereas in 1908 the sea possessed a comparatively con- 

 siderable amount of heat in the deeper layers, which was carried up 

 to the upper layers. 



The rapidity with which the ice melted in the summer was de- 

 pendent, however, not only on this transport of heat, but also on the 

 internal structure of the ice, which again was influenced by the pre- 

 vailing, local, meteorological conditions when the ice was formed, 

 especially by the fall of snow. There was but little snow-fall in the 

 autumn of 1906 and the ice formed was typical sea-ice, produced 

 by the direct freezing of the sea-water. This ice was very hard, little 

 porous, with a veined structure, and melted but slowly. 



In the autumn of 1907 there was abundant snowfall, which 

 weighted the ice down and sea-water flowed over it, a mixed layer 

 of snow and water was formed, freezing later to ice. The ice was 

 stratified with veins of very salt water, of a granular structure, and 

 was much softer and much more porous than that of the previous 

 year and melted much more rapidly 



In the spring of 1908 we observed bubbles forming in the pools 

 on the ice, at some places even like fountains. These were air bubbles, 

 which had been inclosed in the ice when it froze and now were set 

 free owing to the heat and rose upwards, carrying ice particles with 

 them. In this way the ice became spongy and brittle. 



In addition, the soft ice on the under edge was steadily eaten 

 into by the flowing water, which carried away ice particles. Lastly, 

 the horizontal veins with high salinity, which had been formed in 

 winter by the freezing of the snow mush, already became fluid at a 

 temperature varying from — 2 to — 3°. 



