1906. No.3. AMUNDSEN'S OCEANOGRAPHIC OBSERVATIONS IN 1901. 25 
central part of the great central depression of the eastern Barents Sea 
(see Figs. 2 and 3, pp. 26, 27, and PI. III) frequently with a thin layer 
of higher salinity close to the bottom. 
Horizontal Circulation. It is a striking fact that in most sections, 
at least in the spring and summer (e. g. Pl. IV, Sections I and II, and 
Figs. 2, 3, on pp. 26 and 27) there seems to be no connection between 
the latter bottom-water and that of the banks to the west and south; 
for on the slope of the central depression there is, even in April and 
May (Pl. IV, Sects. I and II), water with temperature above zero; and 
it seems as if no cold bottom-water is formed on this slopel The 
reason is probably that along the slope there is too much horizontal 
movement of the water, which renews the water-masses too rapidly to 
allow the vertical circulation during winter, to cool the whole bulk of 
water down below zero. 
On several previous occasions? it has been pointed out that the 
oceanographical conditions of the Barents Sea beautifully illustrate how 
the currents have a tendency to follow the deepest channels of the sea 
bottom3. Where the moving water meets a projection on the sea- 
bottom, it is deflected towards the sides, and if there be openings the 
water will follow the lines of least resistance. It will, more or less, run 
outside these projections or banks along their side slopes, even though 
! Dr. Breitfuss’s Section IV, from Kola-Fjord to Mototchkin Shar, Aug. 4—9, 1902 
(Petermann’s Mitteilungen, 1904, Pl. 4) seems to form a remarkable exception, as it 
shows no warmer water in the eastern part of the Sea, near the land slope, but there 
have evidently been quite exceptional conditions at that period, and besides the eastern 
part of this section must have been in a deep submarine fjord (cf.. Fig. 1, p. 24) 
approaching the Novaya Zemlya coast so near that there is hardly indication of a coast 
bank or shelf in the section. If the many other soundings known from this region be 
correct, there must be very shallow sea both north and south of this fjord, and if so 
there may be comparatively slow circulation in it. Sections across the same region as 
Breitfuss’s Stations 43 and 44, in other year's — e.g. Breitfuss's Section Fig. 3, p- 27, 
from Aug. 1904; his Section across the same region in Aug. 1903 (Bull. Courses Period., 
PI. V, Arct. II); and Knipowitsch’s Section a little farther north, from July 13—19, 1901 
(Ann. Hydr. u. Marit. Meteor. 1905, PI. 6, Fig. 5) — show the typical separation, by 
warmer water on the slope, between the cold bottom water of the coast bank and that 
of the central hollow. It seems probable that in Aug. 1902, there has been a similar 
temporary displacement of the waters as was observed at Amundsen’s Stats. 9 and 10, 
in May 1901 (see below), and also on the slope on the southern side of the Central 
Hollow in May, 1903 (see below). 
19 
Nansen, op. cit. pp. 260 et seg. See also, Some Oceanographical Results of the 
Expedition with the Michael Sars in the Summer of 1900; Nyt Mag. for Naturvidenskaberne, 
vol. 39 Christiania, 1901, p. 152. 
3 Dr. N. Knipowitsch has in several papers (cf. op. cit.), carried through the same 
principle and has shown in detail, how the Atlantic current of the Barents Sea divides 
into several branches regulated by the depressions of the bottom. 
