117 
1906-7.] Photo-Electric Discharge from Platinum. 
XV. — On the Influence of Temperature on the Photo-Electric Dis- 
charge from Platinum. By W. Mansergh Varley, D.Sc. 
(Manchester and Leeds), Ph.D. (Strassburg), B.A. (Cantab.), Assistant 
Professor of Physics and Electrical Engineering, and Fred. Unwin, 
M.Sc. (Manchester), Assistant Lecturer in Physics, Heriot-Watt 
College, Edinburgh. Communicated by Professor F. G. Baily, 
M.A., M.I.E.E. 
(MS. received March 18, 1907. Read same date.) 
Introductory. 
Zeleny ( Physical Review, xii., p. 321, 1901) has investigated the variation 
with temperature of the photo-electric discharge from a platinum wire in 
air at ordinary atmospheric pressure. 
He found that the photo-electric current decreased some 40 per cent, as 
the temperature was raised to about 200° C. ; after this a continuous increase 
in the current occurred up to 600° C., when its value was twice as great as 
at ordinary atmospheric temperatures. Zeleny further noticed a curious 
hysteresis effect, the photo-electric currents for corresponding temperatures 
being far greater during the cooling of the wire than during the heating. 
These experiments, as already stated, were carried out in air at ordinary 
pressure. The presence of so much gas, especially if the potential gradient 
between the electrodes is not great enough to give the saturation current, 
will complicate the phenomena enormously, and it is really impossible to 
judge, from experiments made under these conditions alone, as to the 
influence of temperature on the actual photo-electric discharge, that is, on 
the rate at which negative corpuscles are emitted per unit area of the 
illuminated surface. 
In order to eliminate any influence of temperature on the passage of the 
current through the gas, and especially on the secondary ionisation in the 
gas, due to “ ionisation by collision,” it is essential to work in high vacua, 
when the mean free path of the ions is large compared with the distance 
between the electrodes, and ionisation by collision is in consequence 
negligible. 
The authors have carried out many series of observations under these 
conditions, viz., at pressures low enough to ensure the practical absence of 
any secondary ionisation in the gas itself, and under potential gradients 
