1906. No.3. AMUNDSEN’S OCEANOGRAPHIC OBSERVATIONS IN 1901. 85 
VI. The Formation of the Bottom-Water of the 
Norwegian Sea. 
It is evident that the conditions required for the formation of the 
bottom-water of the Norwegian Sea, are that there shall be near its 
surface, water of salinity about 34:90 0/0, which by radiation of heat 
during the winter, may be cooled down to temperatures about —1°3 
and —1°4° C.; thus it may obtain a density of between 28'11 and 28:13 
or still higher, and become sufficiently heavy to sink. 
It is only in places where Atlantic water has become somewhat 
mixed with Artic water that there can exist conditions allowing of the 
production of water of salinity about 34'90 001, and of a temperature 
sufficiently low for the quantity of heat contained in the water, at the 
surface as well as in the underlying strata, to be no greater than can be 
gradually given off by radiation from the surface, and the whole bulk of 
water as a consequence cooled to below —1°2° C. during the winter and 
spring. It is necessary that there be no very rapid horizontal circula- 
tion, to bring in new supplies of warmer water. Such conditions are 
found in the northern Norwegian Sea, between Jan Mayen and Spits- 
bergen near the outer boundary of the East Greenland Polar Cur- 
rent, in the region of Amundsen’s Stations 13—23, and towards the 
northeast in the region of Mohn’s Stations 302 and 3032. It was pointed 
out above that at the end of the winter there is in this region a maxi- 
mum of density, and it is probably the centre of a great cyclonic move- 
ment of the northern Norwegian Sea. West and north of this region, 
inside the polar current, the underlying warm water is protected against 
cooling, as was mentioned above, by ice and the overlying layer of cold 
but -much lighter Polar water. East of this region are the waters of 
the warm Atlantic Current (Gulf Stream) of high salinity but holding 
1 If the salinity of the surface-water is below 34'90 °/,, at the beginning of the winter, 
the formation of ice may increase it appreciably, as it does in the Barents Sea (see 
above p. 31); and besides, by vertical circulation there will result an intermixture between 
the waters of the upper strata, and the surface salinity be raised. 
? In the sea between Jan Mayen and Iceland there may possibly be similar conditions 
during the winter, as seems to be indicated by the observations at Ryders Stations II 
and III (see above Chapt. V), in June, 1891. 
