18 FRIDTJOF NANSEN. M.-N. Kl. 
salinities below 34'0 and 33'0 000, extended far southwards along the 
meridian of 40° and 42° E. Long., whilst a broad branch with salinities 
above 34'0 %00 extended northwards off the coast of Novaya Zemlya, and 
even towards Franz Josef Land, where Capt. Støkken (of the Capella) found 
salinities of 34°59 and 34'61 0/00, on July 13, 1901. The form of the isohaline 
of 34'0 %/00 evidently indicates roughly the course of the current in this 
region, and here also, there seems to be a tendency towards a greater 
cyclonic movemment, with water flowing eastwards in the region north 
of the Russian coast, northwards along the coast of Novaya Zemlya, west- 
wards, south and southwest of Franz Josef Land, and a southward move- 
ment of colder and less saline water along the meridian of 40° E., on the 
eastern side of the bank in 75° N. Lat.! 
Southeast of Bear Island Amundsen’s observations show two patches 
of water with salinitles above 35'0 0/00; the increase of the salinity in the 
neighbourhood seems to indicate that these values are trustworthy (and 
not errors due to evaporation through the cork-stoppers). The actual 
course along which this water has travelled is difficult to trace; but it 
appears most probable that the water has followed the edge of the bank 
(the Continental Shelf) or very nearly the isobath of 300 metres (see 
Fig. ı, p. 24), and that it is a continuation of the water with salinity 
above 35'0 ‘/00, in about 19° E. Long. along the Capella's route to the south, 
in June. There may be a tendency towards a cyclonic movement as 
indicated by the isohalines on the chart (Pl. I), colder and less saline water 
extending southwards from the vicinity of Bear Island. The observations 
at the stations of the Michael Sars in July 1901, also seem to support 
the correctness of this view. 
It is a noteworthy fact that the surface temperatures and salinities 
found during Dr. Hjort's cruise with the Michael Sars in these waters in 
September 1900, indicate a very similar course of the isohalines and 
isotherms in this region?; which seems to prove that the above peculiar 
distribution of the salinity is not due to accidental conditions prevailing 
I My chart showing the surface temperatures and salinities of the Barents Sea about 
Aug. 1, 1893, (Oceanography of N. P. Basin, Pl. II) was based on much more 
scanty and less accurate material. It gives a somewhat different representations of the 
distribution of salinity; but the general features agree to some extent. The values of 
the salinities are on that chart about o'15 °/y9 higher than the values now obtained by 
Knudsen’s Tables. 
Nansen, Nyt Mag. f. Naturvidensk. vol. 39, Christiania, 1901, Pl. I. See the isotherms 
for 6° C. and 7° C. The values of the salinity are here about 0'12 ®/,, higher than 
the values computed by Knudsen's Tables. South of Bear Island the isohaline for 
to 
/00 
35'0 2/50 (= 349 °/y9) Should probably have had a form more like the isotherm of BAC: 
