26 



BJØRN HELLAXD-HANSEN AND FRIDTJOF NANSEN. 



M.-N. Kl. 



As the sections were taken at different dates, their mean temperatures 

 have to be corrected by the probable change in the difference of time. This 

 has been done in the last column, where the temperatures have been 

 corrected to what the}- would have been on July 20th, if thev had increased 

 at the rate of 0.10° C. in 10 da3's, as found above. 



The decrease of the mean temperatures thus obtained are the following: 



These values cannot be expected to give the real decrease in the 

 temperature of the water on its way northwards from the one section to 

 the other, as the water which was observed, c. g. in Section M in July. 

 1910, may have had a mean temperature comparatively quite different 

 from that of Sect. I when it passed this same region. 



By comparing the mean temperatures of the Atlantic water of the 

 Sognefjord-section, in May of the years 1901 — 1903, with those of the 

 Atlantic water of the Lofoten-section of the following years, we have found 

 [see 1909, p. 180] an average decrease in the temperature of 1.18'^ C. 

 The distance between the sections is about 360 nautical miles. The de- 

 crease should consequently be about 0.033 ^ ^. in 10 naut. miles. 



The mean temperature of the Atlantic water of the Lofoten Section, 

 in May of the four years 1901 — 1904, was S^s*^ C. According to the pro- 

 bable rate of the increase of the temperature during the summer, this should 

 correspond to a mean of about 5.55° C. in June. The difference between this 

 value and the mean temperature of the water of the Atlantic Current in Isachsen's 

 Sect. I is 2.40*^ C. The distance between the sections is about 470 nautical miles, 

 thus showing a decrease of 0.05'^^' C. in 10 naut. miles. This is of course a 

 rough approximation only, as the annual variations could not be considered. 



