DISCHARGE FROM iNlETALLIC SURFACES IN DIFFERENT GASES. 
457 
when, iinclGr otlierwisG idGiitica.! ct)iiclitioiis, no siicli dccrGciSG lu tliG ciiri’Giit was 
o1)tained. 
The curves given in fig’. 10, showing the relation between current and potential at 
various low pressures in air, hydrogen, and carbon dioxide, are all readily under¬ 
stood if we remember that the mean free path of the corpuscles iu hydrogen is, from 
the kinetic theory of the gases, about twice tliat in air or carbon dioxide, and that a 
lower potential difference is required to start ionisation by collision in hydrogen 
than in the other gases. 
The curves of figs. II and 12 obtained after substitution for the zinc of a 
platinum electrode are similar to those obtained using the zinc electrode, except 
that at the lower potential differences the currents in hydrogen are slightly greater 
than those in air, whereas the opposite ’was the case with the zinc. 
The currents in air, hydrogen and carbon dioxide were also comjiai’ed using an 
aluminium electrode. The results obtained were identical in their general character 
with those obtained using zinc, though the sensitiveness of the aluminium was much 
smaller than that of the zinc. 
From the curves of fig. 13 we see that the currents in carbon monoxide nsino- a 
zinc electrode, are some 20 per cent, less than in air, being in fact almost the same 
as those in hydrogen until the jiotentials used are great enough to cause ionisation 
by collision, which, sets in in air and carbon monoxide at about tlie same potential 
gradient. 
Experiments have been carried out on the photo-electric effect in certain less 
simple gases, such as marsli gas and benzene vapour. Tlie results obtained are 
comjilicated by the very large alisorption of the ultra-violet liglit in its passage 
through the gas, and are being furtlier investigated. 
My best thanks are due to Professor J. J. Thomson, in whose laljoratory tlie 
above investigation was carried nut, for his advice and encouragement thronghont 
the whole course of the research. 
\_Note added Octoher I8th, lf)03. — During the past Long Vacation Term experiments 
have been made on the photo-electric currents in gases at very low pressures with a 
new form of apparatus, in which the use of the gauze electrode was discontinued as 
not being suitable for work at [iressures below about 0-02 millim. Two parallel 
plates, separated by a distance of 12 millims., were employed as electrodes, tlie 
ultra-violet light falling obliquely on to the sensitive electrode, and a guard ring 
arrangement being used to procure a perfectly uniform field l)etween the electrodes. 
\\ ith this apparatus expeiiments have been made on the magnitude of the photo¬ 
electric currents in the various gases down to pressures of less tiian 0’002 millim. 
Below a pressure of O'Ol to 0’005 millim. the currents seem to lie ipiite Independent 
^'OL. CCII.—A. 3 X 
