by Ultra-Violet Light, etc. 
419 
The results of the measurements were as follows : 
In air, the natural leak was increased 8 times as the maximum, 
when the gas was exposed to ultra-violet light. 
In carbonic acid gas, the natural leak was increased as much 
as 16 times by the ultra-violet light. 
In ammonia, the natural leak was increased as much as 
150 times by the ultra-violet light. 
The ultra-violet light which produces this ionization is as Lenard 
has shown, absorbed by passing through a few millimetres of air 
at atmospheric pressure. If the box B were provided with a 
quartz window and P instead of going in the box were placed 
opposite the window it was found that the ionization was very 
marked when the distance between the windows was only 1 mm., 
but when the distance was increased to 3 mm. it became too small 
to be observed. We confirmed too Lenard’s observation of the 
great absorption of coal gas by noticing that when the windows 
were 1 mm. apart the ionization was very much diminished by 
allowing a stream of coal gas to play between the windows. 
Origin of the ultra-violet light in the tube. To find the 
relative amounts of ultra-violet light given out by the different 
portions of the tube the arrangement shown in Fig. 2 was used. 
The cathode 0 and the anode A were fixed together by a glass 
rod and the system floated on the mercury in a barometer column, 
the wires carrying the current to heat the cathode 0 dipped into 
side tubes filled with mercury. By raising or lowering the level 
of the mercury in the barometer column different portions of the 
discharge could be brought opposite to the quartz window P. 
Measurements of the amount of ionization produced in a stream of 
ammonia gas, which of all gases tried was the most ionized by the 
ultra-violet light, were made when all portions of the discharge 
from A to G were brought in succession opposite to the window. 
In this way it was found that by far the greater portion of the 
effective ultra-violet light came from quite close to the anode, the 
ionization when the anode A was opposite the window being three 
times as large as when the cathode C was opposite to it. 
All the measurements in these investigations were made by 
Mr G. W. C. Kaye of Trinity College, to whom I wish to tender my 
best thanks. 
The study of the electrical effects produced by light raises 
some very interesting questions as to the constitution of light 
waves, questions which hardly occur when we confine our atten- 
tion to purely optical phenomena. I have already considered 
some of these in my Conduction of Electricity through Gases 
and also in Electricity and Matter. I pointed out that the 
fact that when Rontgen rays pass through a gas only an ex- 
ceedingly small fraction of the molecules are ionized by the rays 
