40 FRIDTJOF NANSEN. M.-N. Kl. 



At Station 17 still colder deep-water was observed. At 700 metres 

 the temperature was o.i6C, and the salinity 34.895 / 00 , At 1 100 

 metres the temperature was 1.07 C, and the salinity 34.895 / 00 . The 

 values obtained by these very accurate determinations thus agree fairly 

 well with the values obtained by previous expeditions, especially the Bel- 

 gica expedition 1905 [HELLAND-HANSEN and KOEFOED 1909]; the AMUNDSEN 

 expedition 1901 [NANSEN 1906]; the ISACHSEN expedition 1910 [HELLAND- 

 HANSEN and NANSEN 1912]. See also HELLAND-HANSEN'S and my work on 

 the Norwegian Sea [HELLAND-HANSEN and NANSEN 1909]. 



On the whole the value of 34.895 /oo seems slightly lower than the 

 majority of the salinities of the cold deep-water, obtained by the most 

 accurate determinations of earlier years, e. g. those of the Belgica Expedi- 

 tion 1905 (between 34.90 and 34.93 /oo)> and those of the Michal Sars 

 expeditions 1900 and 1904 [HELLAND-HANSEN and NANSEN, 1909, pp. 332 

 333 SS^ 337] which were between 34.905 and 34.945 when obtained 

 by determinations of specific gravity, but on the average 0,008 / 00 lower 

 when opbtained by titrations. As our determinations with the Interfero- 

 meter are based upon standard waters, determined by titration only, this 

 may possibly be the explanation of our comparetively low values. 



Current Measurements. 



In 1905 [1906, pp. 27 et seq.} I recommended a method of measuring 

 the currents of the open sea from a steadily drifting ship held by drift- 

 anchors or by the drifting ice by using simultaneously two current- 

 meters, one lowered to the quiet water near the bottom of the sea, in order 

 to determine the movement of the ship or the drifting ice, and the other 

 instrument lowered to the depth where observations were desired. This 

 method was, however, not tried, by myself or others, until 1910, when Pro- 

 fessor BJORN HELLAND-HANSEN made several measurements of this kind 

 from the "Michael Sars" in the Atlantic, while the ship was drifting slowly, 

 held by the biggest tow-nets. In 1911 I made some experiments with the 

 "Veslemoy", drifting with a big drift-anchor made of canvas. These experi- 

 ments fully proved the possibility of making useful current-measurements 

 in this manner. 



In 1912 I had, however, a better opportunity of trying the method 

 during the days from August 4th to 6th and August i8th and i9th. The 

 ship was then moored to the biggest and heaviest floe that could be found 

 among the ice-floes. A thick, heavy piece of ice such as this, of great extent, 

 has naturally a very steady movement. During August 4th to 6th, at 



