3860 Mr. R. K. McClung on the Relative Amount of 
gases was therefore as follows:—Both of the cylinders were 
filled with air at atmospheric pressure after it had been 
carefully dried, by being passed through a long glass tube 
containing pumice-stone soaked in strong sulphuric acid. 
The cylinders were then carefully adjusted in position until 
the ionization in them was such as to produce equal charges. 
upon a and a’, and no deflexion of the electrometer-needle, 
due to these "charges, took place when the quadrants were 
insulated. When this balance was obtained, one of the. 
cylinders was exhausted and filled with the gas which was to. 
be compared with air. It was exhausted and refilled several 
times so as to get rid of all the air in the cylinder. Now, 
since the ionization in this gas would in general be different 
from that in air, the charges received by a and a’ would not 
now be equal, and the balance would be destroyed. Instead 
of measuring the alteration of the balance by the deflexion 
of the electrometer-needle, the pressure either of the air in 
the one cylinder, or the gas in the other one, was altered 
until the amount of Weeion in each of the 06 cylinders 
was the same and the balance was restored once more. 
It has been shown by Perrin* and also by Rutherford and 
McClung? that the total amount of ionization in a gas is. 
proportional to the pressure of the gas. Suppose that the 
ionization in a gas X at a pressure Ps being compared with 
the ionization in a gas A at the same pressure. The two. 
cylinders are filled with the gas A at the pressure P, and 
the ionization in one balances that in the other. One vessel 
is then filled with the gas X at the pressure P, and suppose’ 
first that the ionization in it is less thanin A. Then the 
g the gas A will have to be partially exhausted 
in order to restore the balance. Suppose that when the 
vessel containing 
balance is restored the pressure of the gas A is P’. Let. 
I, be the ionization in A at the pressure P, and let Ix be the 
ionization in the gas X at the same pressure P. Then if the » 
ionization in He” one cylinder filled with the gas A at the 
reduced pressure P’ balances that in the other cylinder: 
filled with the gas X at a pressure P, we have the ratio 
ines 
gee 
that in A, then it would be necessary to partially exhaust the 
vessel containing X in order to obtain a balance. If P” 
were the pressure of the gas X when a balance was obtained, 
Similarly, if the ionization in X is greater than 
* Perrin, Annales de Physique et de Chime, xi. p. 496 (1897). 
+ Rutherford and McClung, Phil. Trans. cxevi. T1901). 
