CANADIAN FISHERIES EXPEDITION, 1914-15 



251 



In the experiment here described, it was not always possible to maintain a constant 

 degree of salinity in the inflowing water. When the salinity increased, a warm saline 

 layer with a slow inward movement was formed at the bottom of the tank. Above this 

 was the cold outgoing current of lower salinity. ■ And above this again warm water 

 poured in towards the ice, with finally, at the surface, a thin layer of brackish water 

 directed outwards, vide fig. 35. This situation corresponds very closely to the actual 

 conditions in the gulf of St. Lawrence, where a cold intermediate layer is also found. 



-«<3 



S 10 IS 20 1^ 



Fig. 36.^Movement of the water by 

 the melting of the icw in the Gull- 

 marfjord. 



The experiment and observations mentioned above give a good insight into the 

 origin and formation of the intermediate cold water layers in the Newfoundland area 

 {vide tables IV and V). It has plainly been formed by the melting of the ice in the 

 gulf of St. Lawrence and the Labrador current. 



When the ice was melting in the Gullmarfjord I noticed it carefully, and measured 

 thei velocity of the water in a vertical line inside the ice edge, vide fig. 36, and also 

 the temperature of the water in a section at right angles to the ice edge, vide fig. 37. 

 From fig. 36 it will be seen that a strong current of water directed inward towards the 

 ice existed down to 10 metres depth, and beneath this, a slighter outwards current right 

 down to the bottom. Fig. 37 gives a very clear view of the manner in which the cooled 

 water sinks in portions from the ice down into the depth below. The same thing doubt- 

 less takes place on the melting of the ice in the gulf of St. Lawrence, and in the Arctic 

 ocean and other places. 



6553—3^6 



