I912. No. 12. THE SEA WEST OF SPITSBERGEN. I5 



surface-layers very much, while the formation of ice during the winter has 

 the opposite effect; but as more ice is probably melted in the region of 

 the Spitsbergen Atlantic Current than is formed there, this process has 

 also a tendency' to lower the salinity of its waters, down to certain depths 

 limited b}- the vertical circulation. 



The differences in the area, in the sections, of the Atlantic water 

 with salinities above 35.0 %o> are not merely due to the differences in 

 the volume of this water caused b}' the intermixture of the waters. 



A comparison between the isopycnals of Sect. IV and those of 

 Section II shows that the velocit}' of the Atlantic current was probably 

 greater in the former than in the latter, the isopycnals having a steeper 

 inclination towards the continental slope. We may therefore conclude that 

 the Atlantic Current was narrower in Sect. IV, but had a greater velocity, 

 than in the sections further south. In Sect. VI the velocity of the current 

 was slower in the direction transverse to the section, but the water with 

 salinity above 34.9 ^/00 had a much wider distribution than in Sect. IV. 



Our maps for 50 to 400 metres (Pis. II and III) probably give the ex- 

 planation. The Atlantic Current sends off its westward branch towards 

 Greenland in this region [cf our description of it, 1909, pp. 280 ct seq., 

 Fig. 93, pp. 316 et scq.\ and the waters have therefore probably moved 

 in directions more or less oblique to that of the section. 



The sections of the Isachsen Expedition may afford a possibility of 

 estimating the rate of the gradual decrease in salinity during the north- 

 ward flow of the water conve3'ed by the Spitsbergen Atlantic Current. 

 The results of such an estimate would naturally be of doubtful value, if 

 the salinity of the current can vary much in the same region from year 

 to year. We think, however, that there is no probability of any great 

 annual variations in the salinity of the Spitsbergen Atlantic Current, con- 

 sidering that we have not been able to find such great annual variations 

 in the salinity of the Norwegian Atlantic Current, during the six years 1900 

 to 1905, from which we have sufficient observations. During this period six 

 vertical sections were taken of the Atlantic Current in the region west of 

 Lofoten and Vesterålen [see our memoir, 1909, PI. II, etc.]. The maximum 

 salinities in all of them were between 35.21 and 35.25*^/00; and these high 

 salinities were only found in some small isolated patches in each vertical 

 section, while otherwise the highest salinities were between 35.16 and 

 35.18 °/oo in all sections ^ The current off" Lofoten had the highest salinities 

 in May, 1904. 



1 The only exception was the section of Februarj', 1903, where the maximum salinity- 

 was 35.15 ''/go; this section was off Vesterålen, and some distance north of the 

 Lofoten sections. 



