150 MR. J. S. TOWNSEND ON THE DIFFUSION OF IONS INTO GASES. 



with the gas. It would require 3 per cent, of air to make a difference of 1 per cent, 

 in the coefficient of diffusion of the ions into carbonic acid. 



The coefficients of diffusion which were obtained for the positive and negative ions 

 in the dry gas are '023 and '026. 



The corresponding numbers for the moist gas are '0245 and '0255. 



The most remarkable difference between the diffusion in carbonic acid and other 

 gases is that the rate of diffusion is nearly equal for the positive and negative ions. 

 In oxygen, air, and hydrogen the rates of diffusion of the positive and negative ions 

 differ by as much as 50 per cent, when the gases are dry, whereas in carbonic acid 

 the difference amounts only to 12 per cent. 



Hydrogen. 



The hydrogen which was used was generated by the action of hydrochloric acid on 

 zinc. The gas was bubbled through three strong solutions of caustic potash and 

 potassium permanganate, in order to remove the acid vapour and the hydrocarbons, 

 and collected in one of the gasometers. The purity of the gas was tested by finding 

 its specific gravity, which is a very sensitive method of detecting the presence of 

 other gases in hydrogen, since the density of the latter is so small. For this purpose 

 a glass flask having a capacity of about 500 cub. centims. was used. Its volume was 

 accurately found, and its loss of weight when dry hydrogen was substituted in it for 

 dry air ; from these two measurements the specific gravity of the gas could be 

 calculated. 



The presence of 1 per cent., by pressure, of air would alter the density by 14 -5 per 

 cent., which can be very easily detected, as 1 per cent, of air in a 500 cub. centim. 

 flask weighs about 6 milligrammes. It was found that the specific gravity of the 

 hydrogen which was prepared did not differ by 2 per cent, from the value '00009. 

 After being in the gasometers and the diffusion apparatus for a few days the gas 

 rose in density, due to air getting in. It would have been a matter of great difficulty 

 to have made an apparatus, which had so many rubber joints, perfectly gas-tight, 

 and it was considered simpler to find the amount of air in the hydrogen after each 

 experiment, and to make a correction in the observed coefficient of diffusion. 



We may here mention an experiment made with the same apparatus as was used 

 for the determination of the rates of diffusion of the ions into oxygen and carbonic 

 acid. The same velocity of gas in the tubing T was used, crV being 2 -08. The 

 positive and negative deflections obtained after the gas had passed through the tubes 

 T 2 were 29 and 27 '5, and the deflections after passing through the tubes T, were 6 -5 

 and 2-2. This shows that in the last nine centimetres of the tubes T, the mean 

 conductivity of the hydrogen fell from 28 '2 to 4 '3. An experiment with oxygen 

 made with the same velocity, a 2 V = 2'08, showed that the mean conductivity of 

 oxygen was reduced from 30'8 to 15'8. The difference in the behaviour of the ions 



