1906. No. 3: AMUNDSEN’S OCEANOGRAPHIC OBSERVATIONS IN 1901. IOI 
| 
Depth Temperature | Salinity «co 
800 m. 015° C. | 3499”. | 28°11 
850 „ | —o'or , GE PN 28'136 
These values are, somewhat lower, than those found abovel; but 
the samples were taken with a water-bottle of Blessing's construction 
| which can hardly have closed perfectly tightly?. Some slight quantity 
of water from the upper water-strata may thus have come into the 
samples, and the salinities have become somewhat too low as a result. 
| On the other hand there is also a possibility that they may have come 
| out somewhat too high by reason of some slight evaporation trough the 
glass stoppers of the bottles, while placed, for sterilisation, in boiling 
water for half an hour without however being heated to boiling point3. 
The glass stoppers were afterwards carefully soldered by paraffin-wax; 
an when the bottles were oppened, in 1898, the paraffin was still in 
perfectly good condition; no considerable evaporation is therefore likely 
to have occurred. There is therefore on the one hand a possibility that 
the salinities may have been somewhat higher than indicated by deter- 
mination of the above samples; it may on the other hand have been 
slightly lower. And if the latter has been the case, the salinity of the 
1 By computing the salinity according to Knudsen’s Tables Prof. Pettersson (/. c. pp. 
316—318) comes to the conclusion that there was a considerably higher salinity (of 
35 10 °/,.) at 450 metres than between 800 and 900 metres of the same station. It was, 
however, pointed out (op. cif. p. 213) that the determination of the sample from 450 
metres was made by Dr. Heidenreich by means of an ordinary Specific Gravity Bottle 
with inserted thermometer, and was not sufficiently accurate. It was necessary first to 
5 EG. 
reduce his somewhat inaccurate values of S Fe by a probable error of 0'00006 
(= 0'078 °/,, salinity). But even after this reduction the value for 450 metres is im- 
probable; for it will give the water at 450 metres a density (o,) of 28°17 (temperature 
= 0'73° C.) while the densities at 800 and 850 metres were according to the values 
above, about 28'11 and 28'136. The density at 450 metres cannot, at any rate, have 
been greater than these; the salinity therefore cannot have been above 35'04 °/,, and 
was probably lower, provided of course that the values obtained for 800 and 850 
metres be correct. It is consequently seen that Pettersson is not on very safe ground 
in concluding from this one inaccurate observation that there was a higher salinity (of 
35'10 °/,,) at 450 metres, than at the greater depths. 
It might be mentioned that the values of temperatures given in Pettersson’s table 
loc. cit. p. 316) are not the final values (op. cit. p. 255); but are somewhat too low. 
It may be mentioned that the writer, some years ago made a water-bottle of exactly 
LEZ 
the same construction, which did not, however, close tightly, and therefore had to be 
altered. 
3 Cf. Nansen, op. cit. p. 212. 
VIE ae Go an, 2a u reg re F6 1 KK å Je EO POET 
