22 . AKSEL S. STEEN. M.-N. Kl. 



traced, as the short vertical line with the dot at the lower end, indicating 

 the noon of the southern magnetic pole, marks an exceedingly slight secon- 

 dary maximum in each curve. 



From the graphic illustration in Figs. 2 and 3, we may, I think, 

 draw the following conclusions: 



The force which produces the diurnal variations of terrestrial magne- 

 tism in the horizontal plane, is due to the sun, which apparently operates 

 in a twofold manner, partly directly, by shining on the observation-station, 

 partly indirectly, by shining on the magnetic pole. 



The direct action of the sun is the prevailing one at stations with a 

 lower latitude than about 60°, and at stations with a higher latitude than 

 60° when there is a greater difference in their longitude from that of the 

 magnetic pole than + 90°. 



For stations with a higher latitude than about 60°, and less difference 

 in longitude than ±90° between them and the magnetic pole, the direct 

 action of the sun is effaced, while an indirect action, dependent upon the 

 position of the station in relation to the magnetic pole, seems to be the 

 predominating one. 



The indirect action of the sun can be demonstrated to a greater or 

 less extent, in the case of each of the stations. 



North of the 78th degree of latitude (Cape Thordsen), the low meri- 

 dian altitude of the sun at the equinox, does not seem to be able to pro- 

 duce any markedly regular diurnal variation. 



.V. 



We have here been dealing only with the numerical value of the 

 deflecting force, its intensity. The direction in which it acts in the hori- 

 zontal plane is indicated by the angle a (Table 2). 



If we now wish, according to our assumption, to explain the appea- 

 rance of the deflecting force as the effect of a system of electric currents 

 existing outside the earth, then the horizontal projection of the set of the 

 current must, according to Ampere's law, always be perpendicular to 

 the direction of the deflecting force in the horizontal plane. If, for in- 

 stance, in Bossekop at a given moment, 9'9'^ a. m. (see Table 2, p. 15), 

 a = 136°, i. e. the deflecting force is directed towards SE, the set of 

 the current in space above Bossekop must be N 46° E, or from NE. If 



