Mr Richardson, On an attempt to detect radiation, etc. 173 
In the form of apparatus first used the air was drawn away 
from the wire and caused to pass between two parallel charged 
plates. One plate was put to earth and the other was connected 
to a sensitive gold-leaf electroscope whose rate of leak was 
measured. The air currents were generally produced by a drop- 
ping water bottle, but for the faster ones a water pump was used. 
The air in entering and leaving passed through sulphuric acid 
wash bottles, and the whole apparatus was further dried by a bulb 
of phosphorus pentoxide inserted in the middle. The supports of 
the plates were insulated with sulphur, and electricity was prevented 
from leaking from the discharge wire to the plates along the glass 
by putting an earthed ring of tin foil round both inside and outside. 
The electroscope was enclosed in a rectangular glass box and 
had a sulphur insulation. The movable leaf was of gold, and the 
readings were taken by a distant telescope. Both leaves were 
generally read in order to correct for any displacement of the 
instrument. It was found necessary to enclose the greater part 
of the apparatus in a wooden box to prevent leaking caused by 
ultra-violet light and gases from the spark gap. The insulation 
of the electroscope itself was very good, the normal leak being not 
more than '4 scale divisions per hour, but that of the whole 
apparatus could not be reduced below two divisions per hour. In 
every case a blank experiment was made immediately after each 
real experiment. For this the outsides of the jars were discon- 
nected from the apparatus and joined by a short wire. In this 
way the effect of the induction coil was made the same as in the 
actual experiments. Thus any appreciable difference in the leak 
in the two cases must be entirely due to radiation coming from the 
wire. 
The following conditions were altered during the experi- 
ments. 
(1) The velocity of the air varied from three small drops per 
second out of the water bottle to a rapid continuous stream drawn 
through by the pump. 
(2) In some of the experiments, by tightening the pinch-cock 
of the inlet tube, the pressure of the air was reduced to about 
5 cms. of mercury. 
(3) The material and diameter of the wires and the capacity 
of the condensers was varied as before. 
