Solar-Diurnal Variations of Terrestrial Magnetism* 437 



we know of electrical action. But while I should thus feel 

 disposed to regard these phenomena as effects rather than 

 causes, in their primary sense at least, it is manifest that, when 

 once generated, such currents will influence the needle; and 

 the results of the Greenwich observers induce us to think that 

 such influence is very great and forms perhaps the chief 

 immediate cause of those oscillations of the needle called 

 disturbances, which are quite different from the solar diurnal 

 variation. In one sense therefore such currents may be 

 effects, while in another they are causes, but causes of dis- 

 turbances rather than of the daily variation. 



It may, however, be replied that earth-currents may be 

 complex phenomena, and that while their most important 

 manifestations may be due to induction, thus being primarily 

 effects rather than causes, yet another portion of them may 

 operate as a true cause producing the solar-diurnal variation. 

 Now there are two objections to this hypothesis ; and one of 

 these is, that we know of no physical agency that will produce 

 in the Earth's crust currents of this nature, which, in order to 

 account for the solar-diurnal variation, must be one and a half 

 times more powerful at times of maximum than at times of 

 minimum sun-spot frequency. Another objection is that Sir 

 George Airy has not been able to detect any resemblance in form 

 between the regular diurnal progress of the magnet and that 

 of the earth- currents. We think therefore that we may dismiss 

 this hypothesis. 



Now, if it be unlikely that these magnetic variations 

 are caused either by the direct magnetic action of the Sun, 

 or by earth-currents, or by the heating effect of the Sun 

 on the chief mass of the Earth's atmosphere, we seem to be 

 driven by the method of exhaustion to look for their cause 

 in the upper atmospheric regions. We shall, however, have 

 to show that there is no improbability in locating their cause 

 in these elevated regions, otherwise our method of exhaustion 

 will have done us no service. In the first place, I need 

 hardly say that if the cause we are in search of be in these 

 upper regions, it must either be in the shape of a set of elec- 

 trical currents, or in some other shape with which we are 

 quite unacquainted; but the nature of this discussion precludes 

 us from entertaining the latter supposition; and we are there- 

 fore driven to regard electrical currents as being the only 

 conceivable cause, if this cause is to be located in the upper 

 atmospheric regions. 



I shall now attempt to reply to two imaginary objec- 

 tions that may be raised as to the possibility of such cur- 

 rents. In the first place, it may be said that while un- 



