22 



FRIDTJOF NANSEN. 



M.-N. Kl. 



Fig. 20. The Stations in Ice Fjord, in July and August 1912. Stat. 7 of July 18, 1912, is in 

 about the same place as Stat. 47 of Aug. 29, 1912. Stat. 8 of July 18, 1912, is in about the 

 same place as Stat. 10 of July ai, 1912, Stat. 44 of Aug. 24, 1912, and Stat. 48 of Aug. 28, 



1912. Scale i : 1,000,000. 



nature of the intermediate layers with a minimum temperature, observed 

 in the summer, especially in Ice Fjord. Our view was that they were 

 due to the vertical circulation of the upper water-strata of the fjord during 

 the winter. The observations of 1912 in Ice Fjord, as well as in other 

 Spitsbergen fjords, seem, however, to prove that these cold intermediate 

 layers cannot be quite so local as we were then inclined to believe; to 

 some extent they seem to have some connection with the water over the 

 shelf outside the mouth of the fjord. 



In very nearly the same localities, at the mouth of Ice Fjord, I took 

 several vertical series of observations on different dates: July 16, 18 

 21, August 24 and 29. They prove that there are considerable variations 

 in the water-strata of the fjords, even within short intervals of time. We 

 see that also the water near the bottom of the fjord may change to some 

 extent from time to time, and may differ even at short distances at the 

 same time, (cf. the water at 380 and 400 metres at Stations 8 and 9, on 

 July i8th, 4.50 and 6.40 p. m., Fig. 23). 



