WORK OF ‘‘ MAUD’’ EXPEDITION—-SVERDRUP 229 
The instrument for measuring radiation was loaned to the expedi- 
tion by the Smithsonian Institution, and was used extensively for 
determining not only the loss of heat at night but also the 
amount of heat received from the sky and the sun in the day- 
time. For this purpose it was mounted beside the instrument for 
recording the duration of sunshine and was made self-recording 
thanks to the ingenuity of Mr. Dahl. The recorder was a very 
sensitive galvanometer. The pen of the galvanometer was pressed 
down by an arm operated by an electromagnet at intervals of four 
minutes. 
Our computation of the amount of heat conducted through the 
sea ice was based on measurements of the temperature within the 
ice at various depths. For this purpose we used resistance ther- 
mometers, which were buried in the ice. The leads were taken into 
the ice house, where the readings were made during the first winter. 
In summer the readings were taken on the ice without any shelter. 
In the spring of 1924 the ice floe in which the thermometers were 
buried was carried away from the ship, and we had to start out in a 
boat in search of it in order to obtain the daily reading. The ther- 
mometers were finally lost when the ice floe upon which they were 
mounted was crushed, but not before a sufficient number of observa- 
tions had been obtained. 
Our knowledge of the physical properties of the sea ice was 
materially increased by experimental studies which Mr. Malmgren 
undertook under very trying conditions. His results show that the 
newly frozen sea ice, which contains a great quantity of salt, really 
consists of pure ice with inclosures of brine. With any change in 
temperature, part of the brine is transformed primarily into pure 
ice, or vice versa. ‘The expansion or contraction of the ice and its 
specific heat depend, to a great extent, upon the intensity of this 
process. The problem can be treated mathematically, and there is 
an excellent agreement between the computed and experimental 
results. 
In summer, when the temperature of the ice approaches the melt- 
ing point, the inclosures of brine increase so much that the ice 
becomes porous, the brine trickles down through, and the upper part 
of the ice, which previously was too salty for drinking purposes, 
becomes absolutely fresh. 
Our daily soundings showed that during the whole time of our 
drift we had remained on the continental shelf; the depth varied for 
long periods between 20 and 30 fathoms, although the distance to the 
coast was 300 miles. A hole in the ice was kept open for the sound- 
ings. Once a week we determined the temperature at various depths 
by reversing thermometers and collected water samples for investiga- 
