750 BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, igO2 1903. 



Jan. 23, time about I7 h 3o m i8 h 30, and 



jgh ^ o m 20 h g O m 



24, I7 h 3o m i8''3o m . (Note especially the secondary deflection 



at about 17'' 57 m .) 

 Feb. 7, i7 h 25 m 



i8 h . 



I7 h 4o m i8 h 2o m . 



At all the places mentioned, the same condition is found as has been pointed out in the small 

 serrations, namely, an upward deflection in the F-curve answering to a downward deflection in the . 

 and Z?-curves. 



The resemblance is throughout so great that there seems no doubt that to a considerable e: 

 the deflections are due to the direct influence of earth-currents. 



We have endeavoured to determine on the original curves the relation P,fjc for some small 

 oscillations. 



We have found that the numbers oscillate in such a way that the mean values of two consecutive 

 numbers attain a satisfactory constancy. The reason of this is to be found in the fact that during the 

 period of observation the external force changes considerably. Bv taking the mean this external 

 will be more or less eliminated. By the aid of these numbers, we can then approximately eliminate 

 the influence of the earth-current upon the vertical curve. 



We have effected an elimination such as this for March 30 31, 1903. Unfortunately \ve have no 

 earth-current registerings for the time about the commencement of the perturbation. It may perhaps 

 seem that little that is of interest has been gained; but one fact at any rate is very apparent, namely, 

 that the effect of the earth-currents on the vertical intensity curve is very small compared with that of 

 the extraterrestrial currents. 



In order to find out where the main body of the current is to be sought for, we may first con- 

 sider one of the smaller deflections, e.g. the serrations at about 17'' 5 m and 17'* 13'" on the 15* 

 February, Series II. 



At I7 u 5' n we find a current that flows from SW to 1 NE. It seems to occasion an upward deflec- 

 tion in the vertical curve, which answers to a magnetic force directed vertically downwards 



A horizontal current that would produce such an effect and have the the direction observed, must 

 now be looked for in NW, i. e. in the direction of the mountain-ridge. 



At 17'' 13 both the earth-current and the corresponding deflection in the vertical curve have 

 changed their direction. The current is therefore still to be looked for in the same direction, i. e. NW 

 of the place. 



This seems to agree with the assumption that the current follows the well-conducting veins of 

 copper in Grytbotten Mountain. 



At Bossekop, on the other hand, the vertical intensity is apparently more strongly affected by the 

 earth-currents than in Kaafjord. This is easily seen by comparing the part of the vertical curve about 

 o h 2 h for the 23rd March with the corresponding part of the N S curve for the earth-currents; and 

 the resemblance between the curves for March 30 31 is still more distinct. 



It is principally in the N S curve that we find agreement with the magnetic curves at Bossekop. 

 If we here, in the same way, try to determine where the main body of the earth-current is situated, wt 

 meet at the outset with the difficulty that the current may flow under various azimuths, which may 

 possibly indicate that the current-line in the neighbourhood of the place of observation is much curved. 



