BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, igO2 1903. 



Sodankyla, on the other hand, we find negative precipitation in the vertical intensity, that is to say a 

 direction the very reverse of that which one would have expected. The easiest explanation of the circ- 

 umstance but hardly a permissible one is, that an error has found its way in, either as a conse- 

 quence of a fault in the apparatus, an error in observation, or an error in calculation; for there seems 

 to be no local current-system at work here. Earth-currents might possibly be supposed to exert a con- 

 siderable influence, but scarcely as much as in the present instance. 



Conditions, however, are found at this place which may be capable of accounting for these dis- 

 crepancies; we have just recently ascertained that in the regions round Sodankyla, there are enormous 

 ironfields, the ore of which possesses magnetic properties of extraordinary strength. 



This could affect the perturbing forces in vertical intensity especially, if we imagine the magnetic 

 masses distributed in a horizontal layer. It would be easy to imagine a distribution of magnetic masses 

 which, by means of induction, might be supposed to occasion anomalies such as these which \ve 

 find here. 



If we, for instance, imagine the station to be situated immediately above the one end of a horizontal 

 magnetic shaft, then the horizontal forces in the neighborhood could be expected to induce free magnetism 

 at the ends of this shaft, and that again would be able to produce strong effects in vertical intensity in 

 a station situated directly above. 



At Godthaab we now have no particularly noticeable effect of the positive system. The perturbing 

 forces are of inconsiderable magnitude. 



At the other stations, as will be clearly seen, negative storms are acting, which, during the three 

 epochs here represented, remain more or less unchanged both in form and strength. Fort Conger evi- 

 dently follows closely upon this series of stations, there being a westerly-directed current-arrow there 

 of a strength similar to that at the other stations. 



From the values of P, to be found at the various stations, a few details may be concluded as 

 to the situation of the current-system. At Fort Rae and Uglaamie, we see that the negative preci- 

 pitation takes place north of the former place and south of the latter, and thus, probably more or less 

 in the auroral zone, which just comes between these two stations. 



In connection with this, we should remember the meaning of the two curves drawn, which show 

 the position of the belt of Northern light. The more southerly, shows the places where aurora is most 

 frequently observed. The more northerly, connects points where aurora is seen as frequently in the south 

 as in the north. 



At Cape Thordsen, we also have small negative values of P v . We must not however, conclude directly 

 from this, that the negative precipitation takes place north of that place, as to the south of it there is the 

 positive polar system, which will here just produce negative values of P t . It would therefore be a fairly 

 probable assumption that the negative precipitation occurred a little to the south of, or possibly more or 

 less directly over, the place. If the area of precipitation were to the north of the station, the perturbing 

 forces in the vertical intensity would probably be greater than we here find them to be, as the two sys- 

 tems would then cause vertical forces directed in the same direction. In all probability, this is the case 

 on Jan Mayen; and we also find powerful perturbations in the vertical intensity. 



Chart V, /<$* 2/'", j<f /"', 79* 25'". The sun is now in the vicinity of the meridian of the magnetic 

 pole, which it crosses in this period. 



Here, too, we find the same areas of perturbation as before. The negative storm has now concen- 

 trated itself more upon the night-side of the globe. In the district Cape Thordsen, Jan Mayen and Kingua 

 Fjord, however, there are quite distinct effects of a negative system which is acting there. The area of 

 perturbation here, however, is not so well defined as before. The positive system is distinctly noticed 

 at Bossekop, and at 19'' 5 at Little Karmakul too. This chart also shows with extreme clearness at this 



