The Discharge of Electricity through Gases. 553 



According to this view we may explain why gases in their sensitive 

 state, like flames, behave differently to positive and negative charges. 

 A positive charge will attract the negative ions, which will arrange 

 themselves on the surface, and the requisite difference of potential 

 will at once establish itself. But if a conductor placed in the flame 

 carries a negative charge, the layer within which the positive ions 

 collect wilhbe deeper, and the potential of the conductor may not be 

 sufficient to complete the layer so as to produce the necessary normal 

 force. 



It also appears that, just as minute chemical changes affect the 

 polarisation in the electrolyte, so will all similar changes affect in 

 the same proportion the fall of potential at the kathode. If I am 

 right, we must consider the conditions of impact between the metal 

 and the ions, or between the gas and the ions, to be different accord- 

 ing as the ions have a positive or negative charge, and this leads 

 us to the next point which it will be necessary to discuss. 



If the law of impact is different between the molecules of the gas 

 and the positive and negative ions respectively, it follows that the 

 rate of diffusion of the two sets of ions will in general be different ; 

 let us see whether we can find any experimental evidence which 

 may throw light on this point. I think there is some reason 

 to believe that the negative ions diffuse more rapidly, and we may 

 at once trace one cf the consequences of such a difference of diffu- 

 sion. Looking at the positive part of the discharge, which shows no 

 signs of a bodily electrification anywhere, at any rate when there are 

 no stratifications, a quicker negative diffusion means, just as in the 

 case of the so-called migrations in electrolytes, an accumulation of 

 ions at the positive pole. That is to say, at the anode a certain 

 number of the ions must recombine again to form a neutral molecule. 

 It has already been mentioned that at the kathode we must imagine 

 decompositions to be going on, continuing during the discharge, 

 because we know that the necessary electrical forces are maintained 

 there. If the discharge is steady, then decomposed atoms must unite 

 somewhere, and, as just suggested, the reunion may take place 

 at the anode ; or it may already take place in or just beyond the 

 negative glow. The two questions are intimately connected. If the 

 molecules are decomposed in one part of the tube and reunite in 

 another, the ions in between cannot travel at the same rate. What 

 leads me to believe in a quicker diffusion of negative ions is the 

 fact described in my former paper, that in the neighbourhood of a 

 discharge, positive bodies apparently become neutralised more quickly 

 than negative ones. I think there runs throughout the whole set of 

 experiments a general tendency for the negative ions to be drawn 

 more quickly than positive ions towards the oppositely charged 

 bodies. Some observations on flames also point the same way. Gold- 



