i 18 



Mi;. .1. s. TOWN'SEND ON THE DIFFUSION OF IONS INTO GASES. 



column. M is the correction for recombination, which has to be added to , and ra/ 



in order to obtain the ratios - 1 - and - , which are the required values of y. 



nt n, 



7 - 31K x 10 

 The values of - ... - are deduced from Curve I., Section I. ; orV, the square of 



' V 



the radius of the tubing T multiplied by the velocity of the gas, is obtained by 

 observing the rate of escape of the gas from the gasometer. The coefficients of 

 diffusion of the ions into the gas are given in the last column. 



Table I. gives the results of experiments with dry air, and Table II. the results 

 obtained with moist air. 



TABLE I. for Dry Air. 



We see that, as the strength of the rays is increased, the values of K for the 

 positive ions appear to increase, and the values for the negative ions to diminish. It 

 will be observed that the charge on the gas (! n,') varies from 5 "6 to 24, and, as 

 already explained, this charge acts in such a manner as to make the calculated value 

 of K for the positive ions too big and that for the negative ions too small. The 

 nearest value for the ratio of the coefficients of diffusion is therefore 1'54, as given by 

 the first experiment. 



Experiments were made to see whether consistent results would be obtained by 

 varying the velocity V, and '036 was obtained for the mean value of K when a 2 V 

 was 1'57. 



An experiment with air, made with the apparatus arranged for the other gases, 

 gave the mean value of K = '034. 



These values are as consistent as could be expected when all possible sources of 

 error are taken into account. 



