80 FRIDTJOF NANSEN. M.-N. KI. 
Direct measurements of the thickness and growth of the ice, con- 
tinued during the whole drift of the Fram, prove that in the North Polar 
Basin it is only during the late part of the summer that there might be 
a slight melting on the under-side of the Polar ice-floes, and this melting 
is due to the heat-wave from above, penetrating down through the ice, 
and also into the water between the ice-floes. When the top-layer of 
nearly fresh water, formed by the melting of snow and ice, on the 
upper surface of the floes, has grown so thick that it reaches down 
below their under-side, new ice may be formed under the latter in 
spite of the heat-wave; the nearly fresh water is cooled down to its 
freezing point and transformed into ice by contact with the underlying 
cold sea-water of temperatures below —1° C. 
During the winter there is no appreciable melting of ice, either in 
the North Polar Basin, or in the East Greenland Polar Current; but 
much ice is formed, and the upper strata of the sea are cooled down 
to freezing point by radiation from the sea-surface. 
Ice floating in the sea outside the boundaries of the Polar Current — 
e. g. between Iceland and Jan Mayen, and between the latter and Spits- 
bergen — may, however, be melted chiefly by heat from the underlying 
water. But this ice is comparatively thin, and is not “Polar” ice from 
the North Polar Basin. It is formed during winter and spring (even 
as late as April) in the same sea, where it melts during summer. 
The writer had a good opportunity of studying the formation and melting 
of this ice during a cruise in the northern seas in March, April, and May 
1882. Ice-masses of this kind may, to a great extent, be carried east- 
wards into regions where they come in more direct contact with the 
underlying warmer water-strata, and the melting of the ice will then be 
chiefly due to the heat of this warmer water. On April 9, 1882, for 
instance, the writer found the boundary of this “western” ice as for east 
as 13° 30° E. Long., in 74° 2’ N. Lat.; which is in the region of the warm 
Atlantic Current west of Bear Island. The appearance of such ice indi- 
cates at once that it is melting chiefly at its under-side. On April 10, 1882, 
(in 73° 14’ N. Lat. and 13° 24° E. Long.) it was 1'9°C. on the sea-sur- 
face, amongst scattered belts of melting ice, and 2°3° C. at about 40 metres 
its melting-point. If such ice, with a melting-point about —o'4° C., floats in water 
which has, for instance, a salinity of about 32°5 9/99 and a temperature of about 
— 1° C., the ice will melt until it has cooled down the water to its freezing-point, which 
is —1°77° C. (cf. M. Knudsen, Publication de Circonstance, Copenhagen 1903, No. 5, 
p. 13), but the melting will be very slow, and slower than if the ice had a melting 
point as low as the freezing point of the water in which it floats. 
