28 BJØRN HELLAND-HANSEN AND FRIDTJOF NANSEN. M.-N. Kl. 



1900. According to the mean temperature of this water in Isachsen's 

 Sect. I, the difference should have been 1.34'^ C. 



The vertical series of observations which Dr. Hjort took in the Spits- 

 bergen waters in 1901 (see above, p. 19), give no trustworthy information 

 regarding the temperature of the current in that year, as they are far too 

 few and not deep enough. Hjort's Stat. 92 may, however, give some in- 

 dication, if we compare the temperatures of the Atlantic water, with sali- 

 nities above 35.0 %o ('• ^- in 100 and 200 metres) at this station with 

 the corresponding temperatures of the Atlantic water at Isachsen's Stat. 11. 

 The temperature of the Atlantic water was in 100 metres 0.37^ C, and 

 in 200 metres 1.24° C. colder at Stat. 92, of July 30th, 1901, than at 

 Stat. II, of June 27th, 1910. Owing to the earlier season (33 days earlier) 

 the temperatures of 191 o should be comparatively still warmer; but on the 

 other hand this may be assumed to be approximately counterbalanced by 

 the more southerly situation. It is consequently possible that the Spits- 

 bergen Atlantic Current was somewhat colder in 1901 than in 1910, but 

 how much cannot be determined. If we were to judge from the above 

 temperatures in 100 and 200 metres, the difference between the mean 

 temperatures for 100 —400 metres might have been approximately some- 

 thing like 0.6'^ C. But this is naturally very uncertain, as there are far 

 too few observations. 



The differences in the temperature of the water in different years 

 where there are no complete vertical sections, may be studied by comparing 

 the vertical temperature-curves of stations that have been taken as nearly 

 as possible in the same locality and at the same time of the year. In our 

 Memoir on "The Norwegian Sea" we have proved that owing to various, 

 probabl}' vertical as well as horizontal, movements of the waters, which 

 are still but little known, there may be great differences between the ob- 

 servations at stations that are quite near each other. It is therefore obvious 

 that for the study of the changes in the temperature of a current, it would 

 be desirable to have for comparison as many vertical temperature-curves 

 as possible from the same region. 



Fig. 21 gives the positions of some stations of the Norwegian North- 

 Atlantic Expedition of 1878 (M 283 — M 361), the Andrée Expedition of 

 1896 (A 3— A 10), the Nathorst Expedition of 1898 (L— U), the Makaroff 

 Expedition 1899 (Ma 5 — Ma 34), Hjort's Expedition 1900 (H 62 — H 65), 

 the Belgica Expedition 1905 (B 3, B 11 a). De Geer's Expedition 1908 (G i), 

 and Isachsen's Expedition 1910 (i — 35). 



Fig. 22 gives the vertical temperature-curves of September 7th, 1910, 

 at Isachsen's Stats. 3 and 4, as also those of Hjort's Stats. 63 and 64 (H 63 



