and the Phenomena which attend them, 547 



Not insisting here on the discussion and study of these inves- 

 tigations, from which it results that the aurora is an electro- 

 atmospheric phenomenon, I will simply call attention to the fact 

 that it is satisfactorily accounted for by admitting, in conformity 

 with the data furnished by direct observation, that, the waters of 

 the ocean being continually charged with positive electricity, the 

 vapours which arise from them act as a conductor of this electri- 

 city as far as the upper strata of the atmosphere, where, carried 

 towards the polar regions by the trade-winds, they form as it 

 were a positive envelope to the earth, which itself remains charged 

 with negative electricity. But the earth and the highly rarefied 

 air of the elevated atmospheric regions being perfect conductors, 

 they may be regarded as forming the two conducting-plates of a 

 condenser, of which the insulating stratum is the inferior por- 

 tion of the atmosphere. The two antagonistic electricities must 

 then necessarily be condensed by their mutual influence in those 

 portions of the atmosphere and of the earth to which they are 

 the nearest, consequently in the regions near the poles, and there 

 neutralize themselves in the form of discharges more or less fre- 

 quent as soon as their tension reaches the limit which it cannot 

 exceed. These discharges should take place almost simulta- 

 neously at the two poles, since the earth being a perfect con- 

 ductor, the electric tension should be nearly the same at each ; 

 there can only be differences in the intensity of the discharges 

 in one region and the other, and from one instant to another in 

 the same region, since the resistance of the stratum of air which 

 separates the two electricities must constantly vary from sundry 

 causes. It is evident, too, that the neutralization of the opposite 

 electricities would not be effected instantaneously, but, consider- 

 ing the low conducting-power of the medium through which it 

 takes place, by successive discharges more or less continuous and 

 variable in intensity. 



These principles admitted, I endeavoured to produce artifi- 

 cially in all its details, and under all the attendant circumstances, 

 the phenomenon of the aurora. I have already published the 

 results which I obtained some time since respecting the influence 

 of strong electro-magnets on luminous electric discharges in 

 highly rarefied air — an influence which explains that of terrestrial 

 magnetism on the aurora. The magnetic and electric pheno- 

 mena which attend their appearance may likewise be produced 

 artificially. The first, as is well known, consist of an augmenta- 

 tion of westerly deflection, followed and occasionally preceded by 

 a much weaker and much less durable easterly deflection. The 

 second, the electric phenomena, manifest themselves by the pre- 

 sence of currents, frequently very intense, in the telegraphic 

 wires. Accurate observations made by Mr. Walker in England 



202 



