Ml, 1 . .1 8, TiWXSKND ON THE DIFFUSION OF IONS INTO OASES. 147 



We therefore obtain, by integration from t = to t = 6, 



Substituting for a and ft their values, this equation becomes 



M - N ~ N ' v A "' (/t * ~ ') 

 N ' N < T ' kg*"' 



CHARGE ACQUIRED BY THE GAS. 



The deflections n,, which are obtained by collecting the positive ions after the gas 

 has passed through the long tubes T,, are invariably greater than the corresponding 

 negative deflections n,', which shows that the gas on issuing from the tubes T, has a 

 positive charge proportional to n, 7;,'. In each of the tubes T, there is a small force 

 arising from this charge, which repels the positive ions towards the sides and attracts 

 the negative ions towards the axes of the tubes. It is only when the rays are strong 

 that this force has any appreciable effect on the motion of the ions. 



The deflections 7i 2 are also slightly greater than n./, but the difference between these 

 numbers is not so great as the difference between n, and n,'. The values which are 



obtained for the ratios are greater than the corresponding values of *^ 



which shows that the negative ions diffuse faster than the positive ions.* 



It was found that the rates of diffusion of the positive and negative ions differed 

 more when the gases were dry than when they were moist. Two sets of experiments 

 were therefore made with each gas ; in one set the gas was passed through large 

 tubes of calcium chloride and entered the diffusion apparatus dry ; in the other set the 

 calcium chloride tubes were removed, and long tubes, partly filled with water, were 

 put in their place. In all the experiments the gases passed through plugs of glass 

 wool before entering the tubes A, in order to remove any dust that might be present. 



RESULTS OF EXPERIMENTS WITH AIR. 



The following tables give the numbers ?t,, n/, n?, and n,', which were obtained with 

 different strengths of rays. The intensity of ionization was reduced to any required 

 value by covering the hole in the lead with pieces of aluminium or zinc. Each 

 experiment consists in determining the four electrometer deflections with a constant 

 strength of rays. The positive and negative deflections n, and n/ are given in the 

 same column, the number in the upper line being n, and that in the lower n,'. The 

 corresponding observations n 2 and / are arranged in a similar manner in the next 



* J. ZEI.KNY, 'Phil. Mag.,' July, 1898, describes an experiment to which he gives a similar inter- 

 pretation. 



U 2 



