1915- No. 2. SPITSBERGEN WATERS. 33 



formed on the sea-surface, by which process the salinity of the fjord-water 

 had been much increased. 



The nature of the water near the bottom of the fjords varies with 

 the configuration of the fjord. In \'an Mijen Bay, inside Bell Sound, where 

 there is a high sill, and the entrance is much blocked by Axel Island, 

 the coldest water, with temperatures about — 1.2° C, is found near the bottom 

 at 95 and 100 metres (Fig. 29). The salinity at this depth is 34.25 0. go- 

 No such water was found in Bell Sound outside the sill, but at Station 54, 

 outside Horn Sound, a similar water-layer with a temperature of — 1.075° C. 

 and salinity 34.25 00 was found at 50 metres below the surface. The very 

 cold bottom water in \'an Mijen Bay might be a remnant of the cold 

 water formed in the Bay during the cooling of the surface and the forma- 

 tion of ice during the winter, but more probably it has come in over the 

 sill from the water outside, which has certainly been much colder 

 during the winter and the spring, than that found in Bell Sound on August 

 31st, 1912. 



In Ice Fjord (and Cross Ba}-) there is, however, considerable differences 

 in the temperature and the salinity of the water even near the bottom, which 

 proves that these fjords cannot be closed by any sill of importance, but 

 that there must be comparatively free communication with the water-layers 

 of the sea outside, from which the deep waters of the fjords have obviously 

 been carried in, as they cannot possibU' have been formed in the fjords 

 themselves. In the deep hollow at the mouth of Ice Fjord, the temperature 

 and salinity of the water near the bottom varied from one day to another, 

 and there were even variations at stations not for apart and taken with 

 intervals of a few hours only ; but the densities did not vary much. Thus 

 even these deep layers show less uniformity than might be expected, which 

 indicates that there must be a fairly active horizontal circulation, even near 

 the bottom. 



On the shelf to the north of Spitsbergen, vertical series of 

 observations were taken at Stations 18, 22, 23, 31 — 35, 38, and 39. As a 

 rule there is no indication of a cold intermediate layer at these stations. On 

 the whole the temperatures are fairl}- uniform between the surface and the 

 bottom on the shelf north of Spitsbergen, especially in its western part near 

 Stations 22 and 23, while the salinities are gradually increasing towards 

 the bottom (Fig. 14). These conditions are very similar to those observed at 

 the IsACHSEN Stations 38, 39, and 40 north of Spitsbergen in August 1910 

 (see Helland-Hansen and Nansen, 1912, Section IX, PI. VI.) The uni- 

 formity in the vertical distribution of the temperature is very conspicuous 

 Vid. -Selsk. Skrifter. I. M.-N. KI. 1915. No. 2. 3 



