PART I. ON MAGNETIC STORMS. CHAPT. I. 49 



If a deflection of length-units on the photogram answers to /, then the scale-value is 



* = f 



n, 



f is to act as the deflecting force for the horizontal variometer along the line of direction of the 

 horizontal component, for the variometer for vertical intensity, in a vertical direction. 



/ is determined in relation to Hg, or the horizontal component of the magnetic force during the 

 determination of sensibility. This is done by letting the deflecting magnet, as before, deflect the decli- 

 nation-needle. During the determination, care must be taken that the deflecting magnet in all three 

 cases is at the same distance from the observation-magnet. 



If the declination-needle undergoes a deflection answering to n d length-units, we obtain 



If this is inserted, we obtain, employing the equation for 



d rr 



ti, = -- . w d rio 







, = . 10 d tif, 



M c 



If we do not here demand greater exactness from s k and c, than i per cent, of the amount, we 

 can in general, as long as the declinometer has the same thread and the same distance, consider io d as 

 constant, x being small in proportion to the unit. We can then generally, instead of H , choose 

 a mean value, H , of the horizontal component. This we can safely do here, as a determination of 

 sensibility made during a great perturbation ought not to be employed. 



We then obtain 



/. = - ' iOd HO 



v ~ ' (-Od **.Q 



For slighter perturbations, we can put, with the same accuracy as before, 



Ht = HO. 



This assumption, which we can probably always make with more southerly stations, is not always 

 permissible for our Norwegian stations in the treatment of perturbations; for at the latter the horizontal 

 component of the magnetic force is very small, while at the same time the variations in it on account 

 of the perturbation may go up to 500 y or even more. We can now put 



In general we have 



,t TJ D [ D 



A fit J\h -f- r\ , 



where R h is the reduction from the mean value to the normal value for the point of time under 

 consideration. PI, is the perturbing force in the direction of the magnetic meridian. 



In the cases in which the equation will come to be employed, P/, is preponderant in relation 

 to /?/,, and if we put 



^ = 6 ' 



Birkelnnd, The Norwegian Aurora Polaris Expedition, 19021903. 7 



