1906. No.3. AMUNDSEN’S OCEANOGRAPHIC OBSERVATIONS IN 1901. 103 
with a temperature of 2:46 C., in 400 metres at his Station IV, is 
consequently much too high; it would give the water a density of 
about 28°13, and make it considerably heavier than the underlying 
waters 1. | 
According to information kindly afforded by Dr. Axel Hamberg, 
the Oceanographer of the Nathorst Expedition to Spitsbergen in 1898, 
the highest salinity observed in the warm West Spitsbergen Current 
during that summer, was 35'19 %0, which on being reduced to the 
values of Knudsen’s Tables, would probably be about 35'14 %0. And 
this was only found in one water-sample; the salinities othewise did not 
exceed 35°15 %0, or reduced 35710 °/oo. This agrees fairly well with 
what might be expected on the basis of what has been pointed out above. 
In Chapter IV it was proved that in the Barents Sea there is formed, 
during winter and spring, a very cold bottom-water which has a 
higher salinity than the warmer water in the same region of the sea. 
It was also pointed out that some of the cold bottom-water, for- 
med in this same manner, may probably flow along the bottom 
into the North Polar Basin from the sea north of Novaya Zemlya. And 
this cold water, which has probably a very high salinity, may help to 
form the bottom-water of the North Polar Basin by being intermixed 
with some of the overlying warmer waters. 
There is also a possibility that the water of the warm Atlantic 
Current with a salinity above 35:0 00, may be carried to the surface by 
vertical circulation during the winter, in the region north of Spitsbergen; 
and may be cooled down to about —1° C. or lower, just as in the 
Barents Sea, whilst the salinity may be somewhat increased by the 
formation of ice at the surface. At Arrhenius’s Stations III—VI the 
- surface-water had salinities between 345 °/oo and 34°82 0/00. Mr. Hercules 
Tornöe.of the Norwegian North Atlantic Expedition found a surface- 
salinity of about 346 V/00? farther north on the very ridge, northwest 
1 Tt was already pointed out above (p. 59) that there are several discrapencies in the 
values of salinity and temperature for the different depths of Arrhenius’s Stations, 
published by Pettersson and Ekman. For 850 metres at Stat. IV, there is given 
a salinity of 3510 °/,., (which reduced would be 35703 °/,,) and a temperature of 
2°05° C. Both salinity and temperature are much too high, the former was probably 
really near that of the bottom-water and the temperature near zero at that depth. At 
Stat. III the published values of temperature and salinity would give about the following 
densities, if the salinities be reduced by 0°07 °/,,: In 40 metres 2749; in 60 metres 
2789; in 80 metres 2763; in 200 metres 27°89; in 300 metres 2797; and in 500 
metres 27°88. A vertical distribution of density like this cannot exist in the Ocean, 
the observations from 60 and 500 metres at least, must be erroneous. 
Hercules Tornöe, Chemistry, The Norw. North-Atlantic Exp. 1876—1878, p. 64. 
