548 M. A. dc la Rive on the Aurora Borcaks, 



and by Mr. Loomis in America, show that these currents vary 

 every moment during the appearance of an aurora, not only in 

 intensity, but also in direction, flowing sometimes from N. to 

 S., sometimes from S. to N. But bearing in mind that the 

 currents developed in the telegraphic wires are currents derived 

 from large sheets of metal implanted in moist earth, it will be 

 seen that, the plates being speedily polarized under the che- 

 mical action of the current passing through them, they must 

 develope in the wire which unites them an opposite current as 

 soon as that whence they derived their polarization ceases or 

 simply diminishes in intensity. In fact it has been universally 

 remarked that the light of aurorse exhibits a very variable inten- 

 sity and continual oscillations. Moreover, the discharges which 

 take place simultaneously at the two terrestrial poles, and which 

 constitute the aurorse boreales and australes, must, by the influ- 

 ence of variable and local circumstances, be alternately stronger 

 at one pole than at the other, and even momentarily cease at one 

 pole whilst in action at the other. 



The phenomenon occurs thus : the negative electricity with 

 which the earth is charged, arriving at a certain degree of tension, 

 discharges itself in the atmosphere of the polar regions, where it 

 meets the positive accumulated there by the trade-winds. There 

 result on the earth two currents directed from the poles to the 

 equator, the direction being that of the positive electricity ; there 

 is especially a current directed from the N. to the S. in the 

 northern hemisphere. But if the discharge takes place at one 

 pole only, the south pole for instance, there is in the northern 

 hemisphere, instead of a current directed from the N. to the S., 

 a current directed from the S. to the N., but weaker. It results 

 from this that the deflection of the magnetic needle, which under 

 the influence of the first current was west in the northern hemi- 

 sphere, in conformity with laws of electrodynamics, becomes 

 east; hence also the currents exhibited by the electric wires 

 are directed from the S. to the N. instead of from the N. to the 

 S. This cause, added to the secondary polarity which the plates 

 had acquired in transmitting the current directed from the N. to 

 the S., must produce a current almost as strong as the latter. 



But in reality it is only rarely that the discharges cease at one 

 pole to take place exclusively at the other; it is rather in differ- 

 ence of intensity that these variations manifest themselves : the 

 same results, however, ensue both as regards the magnetic needle 

 and the telegraphic wires, only they are less pronounced, and 

 accompanied by numerous oscillations. 



I have succeeded in verifying by experiment all these results ; 

 I have especially convinced myself, by passing the discharge of a 

 lluhmkorff apparatus, which has traversed highly rarefied air in 



