The Discharge of Electricity through Gases. 551 



importance to the clearing np of those questions on which the 

 mathematical analysis of the subject will have to be based. From 

 this point of view, nothing is of greater importance than the 

 investigation of the surface conditions which must hold between the 

 gas and the vessel, and between the gas and the electrodes. In a 

 previous paper,* I have shown that at any part of the surface of the 

 gas through which no electricity passes the normal forces must 

 vanish, which is not a priori evident, if we consider the gas to 

 possess a certain dielectric strength. The fall of potential at the 

 kathode must depend on the surface conditions which hold there, and 

 the following considerations may help to clear up the question. 



Imagine, in the first instance, a gas containing a certain number 

 of charged particles, and enclosed in a vessel kept at zero potential, 

 but having a surface impermeable to electricity. We may not be 

 able to realise these conditions, but we may discuss the problem as an 

 ideal case. 



How will the charged particles arrange themselves under the 

 influence of their mutual forces ? They will, no doubt, travel 

 outwards towards the surface, but will they cling to the surface, 

 condensing, as it were, to form a layer against the solid surface 

 resembling a liquid more than a gas ? Or will they form a gaseous 

 atmosphere, diminishing in density from the surface outwards ? Or, 

 finally, will they resemble the state of a liquid in contact with a gas, 

 that is to say, will they be in a state of movable equilibrium, a 

 certain proportion always clinging to the surface of the solid, others 

 flying away until brought back by impacts and electric forces ? I do 

 not see how the answer to these questions can be given on the 

 theoretical grounds only, and it seems to me that experiment only 

 can decide it. Nor is it necessary, to my mind, that an atmosphere 

 of positive ions should behave exactly in the same way as an 

 atmosphere of negative ions. The only consistent theory of contact 

 electricity we possess is that worked out by Helmholtz, according to 

 which we must, in calculating the work done in the transfer of 

 electricity through a surface, not only take account of electric but 

 also of electro -chemical forces. There are various ways of expressing 

 the same fact. We may say that there must be a definite attraction 

 between matter and electricity, or we may say that the potential 

 energy of a system contains terms involving both electrical and 

 chemical variables, or, finally, that both chemical and electrical 

 forces are due to stresses in the medium, and that in calculating the 

 forces we must add the displacements and not the energies. 



In considering the mutual action between electrified particles at 

 molecular distances, it is quite possible, and even probable, that 

 positive and negative electrification may affect the molecular forces 

 * 'Boy. Soc. Proc.,' vol.42, p. 371 (1887). 



