18 AKSEL S. STEEN. TERRESTRIAL MAGNETISM. [NORW. POL. EXP. 
presently at greater length, was also employed as deflected magnet in 
intensity determinations instead of the small declination needle, as the mirror 
of the latter frequently gave an unsatisfactorily indistinct image of the 
telescope’s wire. 
All the settings at the mark done before and after the declination readings, 
are entered under the heading “Mark”, with figures that are the mean of 
the two vernier readings. When azimuth observations are taken with the 
astronomical altazimuth, the value of the angle C —S + A is given, and B 
calculated, the mean of the figures entered under the heading “Mark”, before 
and after the declination readings, being employed as the value of m. If, on 
the other hand, direct azimuth determinations are made with the magnetic 
apparatus itself, only the angle A — 180° —S is specified. Lastly, in the 
column headed D, are placed the values of the absolute declination calculated 
according to formula (1) or (2), which mean may then approximately be re- 
garded as the mean east declination for that day, at the place indicated by 
latitude and longitude. 
As will be seen from the list, there is a considerable number of days 
in which the statement of the angle C— S-+ A, or A — 180° — S, is wanting. 
On this account, it has also been impossible to calulate a value for the ab- 
solute declination. The reason of the omission is naturally that it has not 
always been possible to obtain an azimuth determination in immediate con- 
nection with the readings of the position of the declination needle. Nor is 
this necessary on terra firma, as, with a fixed mark, it is only needful now 
and again to check the azimuth of the mark by astronomical observations. 
This however, it will be easily understood, is not sufficient when the instru- 
ments and mark are set up on drifting ice; for even if the same reading were 
obtained several days running, by setting at the mark, as the list shows was 
not unfrequently the case during the Fram expedition, there is no guarantee 
that the very floe on which the instruments are set up, has not shifted a 
little, thus causing the connecting-line between the declinatorium and the mark 
(the astronomical altazimuth) to change its azimuth. 
I have thought that in addition to their utilisation in the determination 
of the double needle’s error of collimation, these imperfect declination obser- 
vations might be employed in judging of the daily variations of the declination 
needle, and I have therefore represented all the observations graphically on 
