32 Mr Wilson, Leakage of Electricity through dust- free air. 
On the leakage of Electricity through dust-free air. By C. T. R, 
Wilson, M.A., Sidney Sussex College. 
\Read 26 November 1900.] 
Elster and Geitel have shown than an electrified body gradually 
loses its charge when freely exposed in the open air or in a room. 
Their results are in agreement with previous experiments of Linss. 
They conclude from their experiments that free ions exist in the 
atmosphere. The experiments described in this paper prove that 
ionisation can be detected in a small closed vessel containing 
dust-free air not exposed to any known ionising agents. To 
eliminate any uncertainty due to leakage through the insulating 
supports, the system from which the leakage was measured was 
fixed by means of a small bead of sulphur to a conducting rod 
passing through the wall of the vessel and kept at a constant 
potential equal to the initial potential of the leaking system. 
To reduce the capacity of the latter to the smallest possible 
amount the whole system from which the leakage was measured 
was reduced to a small brass strip with a narrow gold-leaf attached, 
the deflections of which, read by means of a microscope, served to 
measure the potential. With a capacity of *73 centim. there is a 
nearly constant fall of potential in a vessel containing 163 c.c. 
of air at atmospheric pressure, amounting to 3 volts per hour, 
the initial voltage being 220. 
The rate of leak is the same in filtered air whether the apparatus 
be filled and used in the laboratory (where contamination with 
radio-active substances might be feared) or in the country. 
The leakage takes place in the dark at the same rate as in 
diffuse daylight. 
The rate of leak is the same for positive as for negative 
charges. 
The quantity lost per second is the same when the initial 
potential is 120 volts as when it is 210 volts. Such voltages 
produce the “ saturation ” current and the rate of leak may there- 
fore be used to measure the ionisation. 
The rate of leak is to a first approximation proportional to 
the pressure ; at a pressure of 43 millims. the leakage is about one- 
fourteenth of that at atmospheric pressure. 
If we take the value found by Prof. J. J. Thomson for the 
charge carried by each ion, 6‘5 x 10 -10 E.U., we can take the experi- 
ments as indicating that 20 ions of either sign are produced per 
second in each c.c. of air at atmospheric pressure. 
