40 FRIDTJOF NANSEN. M.-N. KI. 
bottles with patent india-rubber stoppers, and the possibility of evapora- 
tion was thus excluded. The specific gravity was carefully deter- 
mined by my assistant Mr. Jakob Schetelig with a hydrometer 
of total immersion (made by C. Richter in Berlin, of Jena Glass 
No. 591, and tested at the Reichs Anstalt, Charlottenberg). The 
chlorine of the same samples were determined by Mr. Leivestad, who 
used for control Standard Water No. I, determined by Mr. Schetelig, 
by numerous observations with the Hydrometer of Total Immersion 
(see alove p. 9). Some smaller samples were taken for titration in 
ordinary small bottles with cork-stoppers, and were determined by Dr. 
Heidenreich who used for control a Standard Water from Mr. Martin 
Knudsen in Copenhagen. His values on the whole agreed very well 
with those obtained by Schetelig and Leivestad but were not equally 
accurate. The values of specific gravity, salinity, and density have been 
computed by means of Knudsen’s Tables}. 
D Leivestad by | Schetelig by Hydr. 
epth ; - | 
: Tempe- Titration. of Total Imm. ; | 
in 4 Ot 
Metres pas | 
im OF Cl. ee S. AE Oo S. pe S. alas 
Wollebæk Oo — 122 34°83 > 28'04 
Stat. II 10 — 1'30 || 19'29 '85 27995 | 34'84 +0.01 | ‘05 
May 31, 1900 20 | — 150 ‘295 ‘86 28'03 '88 SN ‘09 
71° 48' N. Lat. | 88 2 | og 
49° 38° E. Long. 50 Ped rs52al “30 '87 ‘03 ‘88 or | ‘09 
70 — 165 | ‘305 ‘88 ‘03 ‘88 ‘oo | Io 
100 —1'65 ‘30 ‘87 ‘03 ‘88 — ‘or | 10 
120 — 780, 7 (455 355 2551 35'105.) Yes 33 
128 | bottom | | | 
It is a noteworthy fact that salinities as high as that of the sample 
from 120 metres, 8 metres above the bottom, have been observed 
nowhere else in the eastern Barents Sea. As before mentioned it is 
evidently due to the formation of ice at the surface in the same region 
during the previous spring or winter. It is very natural that water of 
! The values are here only o'ıı °/,, lower than the values given in the Memoir on the 
Oceanography of N. Polar Basin, p. 273, although they ougth to have been 0'16 °/,, 
lower (cf. op. cit., Preface p. V). The reason is that by means of Knudsen's Tables 
the observations can now be more accurately computed than was then possible. It 
is also seen that the values now obtained for the sample from 120 metres by titration 
and Hydrometer of Total Immersion agree fairly well. 
? Determined by Dr. Heidenreich. 
