on Discharge in a Vacuum Tube. 473 
hole, h, gave connection between T and the main tube. The joint 
between S and T was made with sealing-wax ( W ). 
To 
the Pumjj 
X 
T 
The experiments were made at a pressure giving a dark space 
of 6 to 8 mm. The source of ultra violet light was the spark 
from a 6-inch induction coil — with a leyden jar across the 
secondary — taken between aluminium terminals. No condensing 
lens was used but the spark gap was placed immediately in front 
of the quartz window. In order to prevent induction effects from 
the spark gap and its leads, it was found necessary to enclose 
them in an earthed metal case, except for a small opening through 
which the rays passed. The effectiveness of the shielding will 
be seen from the fact that although the discharge would start 
in the dark for a potential difference of 555 to 557 volts, and 
while the radiation was falling on the kathode it would start 
at 511 volts — a bit of ordinary window-glass placed in front of 
the quartz window prevented the discharge from starting at 
554 volts. 
When the discharge was started an electrostatic voltmeter, 
whose terminals were connected to the kathode ( K ) and the 
needle (W), showed a potential difference of 375 volts, and no 
change could be detected in this when the rays were put on or 
shut off, though a change of two volts should have been easily 
seen had it occurred. 
