ON ATMOSPHERIC ELECTEICITY 213 



The principal facts which have been discovered can be stated in a few 

 words. In clear weather, the potential increases as we go higher, at 

 least for certain parts of Europe, and there is a diurnal and annual 

 variation of this quantity which the presence of fogs causes also to vary. 

 The first observers were inclined to attribute the electricity of the 

 atmosphere to the evaporation of water, and an old experiment which 

 consisted in dropping a ball of red-hot platinum into water placed on a 

 gold leaf electrometer, was supposed to confirm this view. Even re- 

 cently a distinguished physicist held this opinion in the case of electric 

 storms. Now when a ball of platinum is thus dropped into water, the 

 excessive commotion thus produced will certainly give rise to electricity; 

 but to assert that this electricity is due to evaporation may very well 

 be an error. It is true that occasionally a red-hot meteorite may fall 

 into the sea, reproducing thus the laboratory experiment; but most of 

 the water is evaporated quietly. Eecently one of my students used 

 under my direction a Thomson quadrant electrometer in order to inves- 

 tigate this question, and although he evaporated large quantities of 

 different liquids, he did not find any trace of electrization. I hope to 

 prove thus conclusively that the electricity of the atmosphere cannot 

 be the result of evaporation. 



Sir William Thomson thinks that the experiments which have been 

 made hitherto indicate that the earth is charged negatively. This con- 

 clusion would certainly explain all the experiments hitherto performed 

 in Europe ; but the only method of reaching certainty on this point is to 

 execute a series of experiments on the whole surface of the globe, and 

 it is this method that I propose to-day. This series of experiments 

 would furnish data for determining not only the fact of terrestrial 

 magnetism, but also by the aid of Gauss's theorem the amount of the 

 charge on the solid portion of the earth; however, this amount cannot 

 be determined for the upper atmosphere. What we want to know is 

 the law according to which the electric potential varies as we ascend 

 on the whole surface of the globe and at the same instant of time, so 

 that it may be possible to obtain the surface integral of the rate of 

 variation of the potential over the whole globe. If the earth were ever 

 to receive an increase of charge coming either from the exterior or from 

 the upper atmosphere, this increase would be known. When, in the 

 London Physical Society, I criticized the theory of Profs. Ayrton and 

 Perry on terrestrial magnetism, I gave at the end of my paper a brief 

 outline of a recent theory on auroras and storms, which was built on 

 the hypothesis of the electrization of the earth. After mature reflec- 



