19 1 2. Xo. 12. THE SEA WEST OF SPITSBERGEN. 27 



Annual Variations in the Tcnipi-ratiirc of the Spitsbergen Atlantic Current. 



We have found that there are considerable variations in the mean 

 temperature of the Norwegian Atlantic Current [cfr. 1909, p. 171 et seq.]. 

 It is, therefore, probable that there are similar variations in the Spitsbergen 

 Atlantic Current which is a continuation of this current. There is unfor- 

 tunateh' no great material of observations from previous years which may 

 be used for a stud}'' of these annual variations. 



Dr. Hjort's section of September 4th — 6th, 1900 (Fig. 21, Stat. H 62, 

 H 63, and H 64; cfr. PI. \'I, Sect. H), westward from Bear Island, was 

 taken about 40 nautical miles farther south and 72 days later in the season 

 than Isachsen's Sect. I of June 25th, 1910. It might, therefore be expected 

 that the Atlantic water of the former would be warmer than that of the 

 latter. This is not the case, however. The isotherm of 2*^ C. lies higher 

 in the section of 1900 than the isohaline of 35.0 %o; in Sect. I of 1910 

 it lies lower than that isohaline. 



The mean temperature of the water between 50 and 450 m. at Stations 

 2, 3, 4, and 5 of Sect. I was 3.15'^' C. ^. The mean temperature of the 

 water in the same depths at fijort's Stats. 63 and 64 was 2.68^ C. If, 

 in order to make the mean temperature of Section I fully comparable with 

 that of Hjort's section, the former should be corrected for the probable 

 difterence in temperature owing to its more northern latitude and the earlier 

 season (according to the values of these difterences found above, viz. 0.038^ C. 

 for every 10 naut. miles, 0.10° C. for every 10 da3's), we obtain a mean 

 temperature of 4.02° C, which is consequendy 1.34" C. higher than the 

 corresponding mean temperature at Stats. 63 and 64 of 1900. If Isachsen's 

 more northern sections be corrected in the same manner, their mean 

 temperatures of the water between 50 and 450 metres should be: Sect. II 

 (Stats. 7 — II) = 4.1^' C; Sect. IV (Stats. 25—29) -= 3.9'^ C. ; Sect. VI 

 (Stats. 17 — 19) = 4.0*^ C. These values consequently represent the mean 

 temperatures which the volume of Atlantic water of the ditïerent sections 

 might have had when they were in the latitude of Hjort's Stats. 63 and 64 

 (about 74^ 15' X. Lat.), and if they had been there on September 5th. 

 These values agree very well with each other. 



We mav thus conclude that in the sea west and northwest of the 

 Bear Island submarine shelf, the waters of the Spitsbergen Atlantic Current 

 were considerabl}' warmer in the summer of 1910 than in the summer of 



' It makes no great ditference if Stats. 5 and 2 be left out, the corresponding mean 

 temperature of Stats. 3 and 4 being 3.08'^' C, and of Stats. 2, 3, and 4, 3.14" C. 



