20 



FRIDTJOF NANSEN. 



M.-N. Kl. 



Orrr\ 



3470 



3500 



Î0O - 



200- 



300- 



that the water 

 of the Spits- 

 bergen Atlan- 

 tic Current re- 

 ally was consi- 

 derably colder 

 in the summer 

 of 1912 than 

 in the summer 

 of 1910. 

 Helland-Hansen and I found 

 [191 2, pp. 42 et seq.\ there was 

 probably a certain similarity 

 between the variations in the 

 temperature of the Atlantic Cur- 

 rent west of northern Spits- 

 bergen and the variations in the 

 mean winter-temperature (/. e. 

 for the six months December i 

 to May 31) of the five northern 

 meteorological stations of Nor- 

 wa}', Andenes, Tromsö, Alten, 

 Vardö, and Syd- Varanger, during the winter a 3'ear and a half earlier. 

 The mean temperature anomaly for December ist 1900 to May 31st 191 1 

 was, however, 1.06 ^C. at these five stations, while it was only 0.68 ^^C. 

 in the wnnter 1908 to 1909. This is consequentl}- contrary to what it 

 ought to have been according to our former results. 



But we thought that the variations in the temperature of the Atlantic 

 Current in the region south-west of southern Spitsbergen might have more 

 similarity to the variations in the mean temperature at the five northern 

 Norwegian stations during the preceding winter, or rather the variations 

 in the means of the two preceding winters. 



wo- 



50O- 



60Û 



700 - 



800- 



900 



1000- 



mo - 



noo^ 



Figs. 17 & 18. Curves showing the vertical distri- 

 bution of Salinity (Fig. 17 to the left) and Tempera- 

 ture (to the right) at the Isachsen Stations 17 and 

 37 (of Jul}- 17 and Aug. 19, 1910) and our Stat. 17 

 (of July 30, 1912). 



