The Discharge of Electricity through Gases. 557 



potential which determines the edge of glow. Bat this point must 

 be left to be settled by future experiments. 



As regards the inner luminous layer closely adjacent to the negative 

 electrode, it seems due to the positive ions approaching the kathode, 

 and not, like the glow, to the negative ions projected away from it. 

 This is shown by the fact that a wire placed inside this layer casts a 

 shadow towards the kathode, and also by the distortion it experiences 

 in a magnetic field. It is remarkable that this luminosity, due to the 

 impact of positive ions, shows according to Goldstein, at any rate in 

 the case of nitrogen, the spectrum of the positive part of the dis- 

 charge. 



Conclusion. 



We may now in conclusion shortly summarise the results arrived 

 at. A gas in its normal state contains no free ions, but, if through 

 any chemical or physical causes the molecules are broken up in an 

 electric field, ions form, and the gas becomes a conductor. Supposing 

 the difference of potential of two electrodes is gradually increased, 

 a point will be reached at which a spark will pass, that is to say, the 

 molecules will be broken up by electric forces, the positive ions dif- 

 fusing towards the kathode will tend to form a polarising layer of 

 finite thickness, increasing in width as the pressure diminishes. If 

 the discharge becomes steady, the decompositions are continuously 

 kept up at the kathode, the negative ions being projected with great 

 velocity away from it. These ions will move through the so-called 

 dark space without much loss of energy by impacts, but when, prob- 

 ably owing to sufficient diminution in the electric force, the impacts 

 become more frequent, the translational energy becomes transformed 

 into the luminous vibrations of the glow. The positive ions forming 

 an atmosphere round the kathode must have a greater velocity the 

 nearer the kathode, where their energy becomes visible in the first 

 luminous layer. Whether decompositions take place only at the elec- 

 trode or through a finite distance from it is at present an open 

 question, nor can we decide as yet whether the negative molecules 

 projected outwards are the main carriers of the current inside the 

 dark space. In the dark space the negative ions will accumulate and 

 meet the positive ions proceeding from the positive part of the dis- 

 charge. We shall expect at some point towards the outside of the 

 glow the free ions to become more numerous than in other parts of 

 the discharge. Here we find a small fall of potential and no lumi- 

 nosity ; this is the dark interval separating the positive part of the 

 discharge from the negative glow. A number of ions probably 

 reunite in this part to form molecules, and in case it should ultimately 

 be found that positive and negative ions diffuse with the same velo- 

 city, we should have to conclude that as many molecules as are 



