213 
Mr Campbell, The (3 Rays from Potassium.. 
formed the bottom of the box was covered with thin aluminium 
foil 0‘0004 cm. thick supported on a brass frame. Under this 
window was placed the grid of plates for deflecting the rays. 
This grid consisted of 57 zinc plates each 38 cms. long, 40 wide 
and ' 08 thick : the plane of the plates was vertical and the distance 
between the inside faces of consecutive plates was 06 cm. : the 
ends of the plates were supported in two long blocks of paraffin wax 
and were connected to wires so that alternate plates could be joined 
to opposite poles of a small Wimshurst driven by a motor. 
Under the grid was placed a glass tray in which the active 
material was placed : the whole apparatus was placed in a lead- 
lined box to shield it from accidental radiation. 
The Wimshurst was found capable of maintaining a steady 
p.d. of 8000 volts between the plates : the p.d. could be maintained 
at any lower value by means of an adjustable point and ball gap 
placed in parallel with the plates of the grid : it was measured by 
a Kelvin electrostatic voltmeter. A simple calculation will show 
that the P.D. of 8000 volts was sufficient to deflect into the plates 
of the grid all (3 particles emitted in a vertical direction from the 
active material with a velocity of not more than T4 x 10 10 cms. 
per sec. It is therefore of the order of magnitude required for 
these experiments. 
§ 6. The saturation current through the testing vessel was 
measured by the compensation* method which is eminently suit- 
able for the detection of small changes in the current. The pressure 
in the compensator was adjusted until a balance was very nearl} r 
attained and the movement of the leaf of the electroscope in a 
fixed period (10') was noted. The Wimshurst was then set in 
motion and a similar reading taken with a known P.D. between the 
plates of the grid : the first reading was then repeated and so on, 
until five readings, two with and three without the field, had been 
taken. These five readings constitute one observation and occupy 
about an hour. The difference between the means of the deflections 
of the leaf with and without the field was a measure of the magni- 
tude of the change of current produced by the field. By observing 
the difference in the deflection of the leaf on changing the pressure 
in the compensator by a known amount, this change of current 
could be found in absolute measure. 
It may be of interest to give some idea of the sensitiveness of 
the method. It appears from the “ blank ” experiments given 
below, that a change of half a scale-division in the deflection of the 
leaf in ten minutes could be detected with certainty : such a change 
indicates a change in the current of 3 x 10 -10 amperes. The 
saturation current through the testing vessel in the absence of the 
* See Proc. Cavil, Phil. Soc. xm. p. 132. 
