[DEMPSTER] MOBILITIES OF IONS IN GASES 9 
The gauze G was made positive and negative by means of a com- 
mutator C and the connections to it are shown in Fig. 1. In making 
measurements the commutator was run at a fixed speed and the charge 
acquired by E was measured for different potential differences between 
Gand$. By drawing a curve to represent these readings it was easy to 
find the voltage just necessary to bring the ions across in the time that 
the ions are travelling. With this voltage known the mobility may be 
readily deduced. 
It may be easily shown that a change in the rate of running 
the commutator has a great influence on the charge received. In fact, 
if Vo is the voltage just required to bring the ions across and if the 
current is being measured with voltage V, then a change in the rate of 
= 
/ 
yee 
V—V, 
times as large. It was therefore necessary to use an accurately made 
commutator and to run it at a uniform speed if an accurate comparison 
of values was to be made. The commutator was the one ordinarily 
used for such experiments, and consisted of two semi-cylindrical sections 
fastened toarevolving shaft and separated by ebonite. The brush used 
was a strip of stiff brass pressing diagonally down on the revolving 
commutator drum. When the brush was on the ebonite insulation, G 
was brought to zero potential by means of the high resistance con- 
nection to earth, HR. 
the commutator produces a change in the charge received 
The rate of running the commutator was kept uniform by keeping 
constant the E.M.F., which was given by a small constant field dynamo 
run by a belt to the shaft of the commutator. The E.M.F.of the dynamo 
opposed by a couple of standard cells was measured on a second electro- 
meter. In this way variations of 0-1 % could be detected and cor- 
rected by means of a variable resistance in the circuit which controlled 
the motor running the commutator. 
In the experiments made up to the present the gas used was air 
compressed into the receiver by the pump of the liquid air apparatus. 
The air was drawn through lime, compressed, passed through water, and 
dried with caustic potash before being admitted to the cylinder. 
The velocities thus far determined are given in Tables I and II, 
where P means the pressure in atmospheres, k the mobility of the ion 
in cm., per sec., per volt per centimetre, and pk the product of the 
mobility and the pressure in atmospheres. The readings were all made 
with a potential difference of more than 300 volts between P and G. 
