5i6 THE POPULAR SCIENCE MONTHLY. 



These two conclusions, applicable to condensation, may be applied 

 to the frequently observed fact that a vivid flash of lightning is often 

 quickly followed by a sudden and heavy down-pouring of rain. It is 

 clearly impossible to tell which is antecedent to the other, the discharge 

 or the sudden condensation ; for, while the flash reaches the observer 

 first, it is clear that light travels from the place of action with a vastly 

 greater velocity than that of the falling rain, and the discharge may 

 therefore have been subsequent to the sudden condensation. If the 

 discharge occurs first, then the lowering of the electric potential per- 

 mits approach of aqueous spherules and consequent coalescence after 

 collision, in accordance with Lord Rayleigh's experiments. On the 

 other hand, if the condensation is antecedent, it follows from the result 

 reached above that it must be accompanied by a sudden rise of poten- 

 tial in the enlarged drops, leading to an electric discharge. 



It will be observed that neither of these principles accounts for the 

 original electrification of aqueous vapor in the air. It has been the 

 custom to regard thunder-clouds as primarily charged by some unex- 

 plained process ; and these, acting inductively, as producing a corre- 

 sponding charge of opposite sign in the earth underneath. This view 

 appears to have no conclusive evidence in its favor, but corresponds 

 rather to appearances merely a very unreliable guide. 



In the theory here proposed, the earth is the charged body, acting 

 inductively on the air, aqueous vapor, and clouds about it. Whenever 

 moisture condenses to cloud, a better conductor is thereby formed, and 

 increased inductive action takes place, causing an accumulation of elec- 

 tricity both in the cloud above and the earth beneath. If, then, the 

 lower part of the cloud, under this inductive influence, condenses to 

 rain and falls away from the upper part, a separation of the two 

 electricities is effected, and an increase in potential results from the 

 enlargement of the drops, as explained above. Consequently, a dis- 

 charge may then take place either between the upper and lower cloudy 

 strata, or between the lower portion and the earth, according as one 

 path or the other offers the least resistance. 



Further, when evaporation takes place from an electrified locality, 

 the rising vapor must carry away a charge of electricity by convec- 

 tion, as air and dust carry away electricity from a charged conductor. 

 The condensation of this vapor increases its potential, and, if sufii- 

 ciently rapid, gives rise to electrical displays. It is a fact of recent 

 establishment that " northern lights " occur in various high latitudes 

 only with southerly winds, which come laden with moisture and jDrob- 

 ably with electricity. 



The following considerations in favor of this view of the origin of 

 atmospheric electricity may be briefly enumerated : 



1. Continuous observations of the electrical state of the atmos- 

 phere at Kew Observatory and elsewhere for several years show that 

 the air is always moi-e or less electrified. The average potential of 



