ie 
1906. No. 3. AMUNDSEN’S OCEANOGRAPHIC OBSERVATIONS IN 1901. 33 
In the days from April 29 to 30, from May 3 to 9, and from May 
23 to 28, 1901, Amundsen sailed through water which was cooled down 
to its freezing point (—1'9 C.). The water-samples, taken with the glass 
. water-bottle of his own construction, from 5 and 10 metres below the 
surface, give frequently during those days remarkably high salinities, 
sometimes even above 3600. The reason is obviously that ice has, 
frozen out on the walls of the glass collecting bottle before the samples 
had been bottled for storage (see above pp. 7—8). On some occasions 
however, the salinities have been lowered by small ice-needles floating 
in the water, getting into the water-sample. The observations taken in 
these days are therefore not trustworthy. On the whole, however, the 
Pettersson, Vega Exp. Vetensk. lakttagelser, vol. II. pp. 373—374). Observations 
like these are obviously erronous; for it is at any rate certain, that sea-water cannot 
be thus super-cooled, where ice is present on the surface. In spite of hundreds of ob- 
servations made during the polar winter, the present writer has on no occasion observed 
temperatures below the freezing-point of sea-water, where ice occurred on the sea- 
surface. Attention may, however, be drawn to some observations of Amundsen. When 
he sailed across an extensive open lane in the ice on May 4—5, 1901, he remarks in 
the journal (see Table I) that he would not record the temperature-readings as they 
were absurd, the thermometer indicating below —2° C. or even —2'4? C. and he 
therefore thought that there was something wrong with the thermometer (No. 638) 
which, however, always gave quite correct readings, even only a few hours before and 
after these observations (see Table I). It seems difficult to understand what might have 
been the matter with this evidently very trustworthy instrument, and the posibility does 
not seem to be excluded that there has actually been super-cooled water in this open 
lane far from any ice, although it seems hard to understand how such very cold water 
could be stirred and taken on board in a bucket, without being instantly transformed 
into ice, but the salinities of the water-samples taken, do not indicate that this has 
been the case. It is also a remarktable fact that on May 4 and April 29, Amundsen 
repeatedly got readings of —2'0 C.; and his readings have without doubt usually a high 
accuracy. 
Edlund also mentions that Nordenskiöld on some occasions has observed ice on 
the bottom of the sea. The present writer has frequently seen the same thing near shore. 
The explanation was, however, that at these places the ice had been frozen solid to 
the bottom during the winter, afterwards during the summer it had been partly broken 
away by pressure and partly melted near the surface, while it still remained between 
the stones on the bottom. 
It has to be remembered that when the water is exposed to higher pressure by 
sinking, its temperature is slightly raised, while its freezing point is lowered, and it does 
not, therefore, seem very probable that under regular circumstances ice may be formed 
belôw the surface in this manner. 
In another way, however, ice may be formed at some distance below the sea- 
surface. During pressure the ice-floes are broken and piled up in hummocks, which 
may often reach down to depths of 40, 50, or perhaps even 60 metres or more. The ice 
thus pressed down during the coldest part of the vinter, may have temperatures of 
between —20'0° and —30'0° C. (cf. The Norwegian North Polar Exp. 1893—1896, 
Scientific Results, vol. VI, No. 17, pp. 544—557), and before this ice is heated by the 
water to its freezing point, much new ice may naturally be formed, as the writer has 
also verified by direct observations; the salinity of the water is of course increased as 
a result. 
Vid.-Selsk. Skrifter. I. M.-N. Kl. 1906. No. 3. 3 
