WORK OF ‘‘ MAUD’’ EXPEDITION—SVERDRUP 228 
instrument for photographic registration of the declination in a 
light-tight case within the smaller tent previously used. 
I shall not enter upon the results of our magnetic observations 
during the drift, but wish to mention the character of the diurnal 
variation of the magnetic declination as recorded during the winter 
of 1924-25. The most remarkable feature is the small range of the 
diurnal variation in the middle of the winter and the rapid increase 
of this range in the spring. It is to be hoped that our records, 
combined with previous results, may furnish sufficient data for the 
application of corrections for diurnal variation to the declinations 
observed on or near the Siberian coast. . 
The records may also be of value in the study of magnetic storms. 
There is a close relation between the occurrence of magnetic storms 
and the occurrence of the aurora borealis. We always had to keep 
night watches. We used to stay up for two hours each, and the 
watchman was instructed to make frequent notes regarding the form, 
amount, and intensity of the aurora. We succeeded in taking sev- 
eral pictures of brilliant displays, using cameras developed by 
Professor Stérmer, of Oslo. 
The atmospheric-electric observations in the winter of 1922-23, 
which were confined to observations of the potential gradient, were 
also taken in the ice house. 
In 1922 the Department of Terrestrial Magnetism had drawn our 
especial attention to the value of observations of the diurnal varia- 
tion of the gradient over the Arctic Sea. One of the most interest- 
ing results of the atmospheric electric work carried out on board the 
Carnegie during 1915 to 1921 was that this variation follows univer- 
sal time over the oceans, the maximum value being reached simul- 
taneously over all the oceans. Our special task was to ascertain 
whether this law for the variation was valid over the Polar Sea as 
well. 
During the first winter the diurnal variation of the potential gradi- 
ent was followed by eye observations through 24 hours, but we 
found that we naturally would save time and materially increase the 
amount of data if we could record the gradient continuously. I, 
therefore, asked our aviator, Mr. Dahl, who is a genius as an in- 
strument designer and maker, to construct a recording electrometer. 
The instrument itself did not present any difficulties, but these arose 
when a perfect electrostatic insulation was to be insured. Amber 
is generally used for insulation, but we had no supply of amber. 
The difficulty was finally overcome by my sacrificing a perfectly 
good amber pipestem. 
Our recording electrometer was placed in an unheated room on 
deck and became, therefore, covered with frost on the outside, but 
