[ALLAN] EXCITED RADIOACTIVITY FROM ATMOSPHERIC AIR 73 
The apparatus used for testing radioactive wire in most of the 
experiments is shown in Fig. 1. It consists of a metal cylinder A 
closed at the bottom, this cylinder serving as the outer electrode. 
The inner electrode consists of an iron frame B of rectangular shape, 
120 cms. long and 10 cms. wide. The radioactive wire was wound 
on this frame from end to end making about 15 turns of wire. 
This frame was hung inside the cylinder by means of an insulated 
cross-bar C. This cross-bar was in three pieces, joined together by 
two pieces of brass D D. These brass strips were connected together 
and earthed. This arrangement acting as a guard ring and prevented 
any leakage at the ends. The outside cylinder was connected to 
one pole of battery E, the other pole of which was earthed. 
The battery consisted of a group of storage cells from which any 
voltage from 2 up to 600 could be obtained. The diameter of 
cylinder was 30 cms. and the length 200 cms. The inner electrode 
was connected to one pair of quadrants of the electrometer F, the 
other pair being earthed. 
The electrometer consisted of a very light needle of silvered 
paper suspended by a very delicate quartz fibre between upper and 
lower pairs of quadrants. The needle was charged up to a potential 
of 200 volts by lightly touching it with a fine wire at a potential of 
200 volts. The charge on the needle kept quite constant, never leak- 
ing away more than 10 per cent in 24 hours, so that for readings over 
a few mins. it would be quite constant. To make sure whether the 
charge on the needle had leaked away much, an arrangement was used 
which consisted of two parallel plates of metal between which a 
standard specimen of uranium was placed. The lower plate was 
charged to 50 volts and the upper plate was connected temporarily 
to the electrometer. The rate of leak induced by the uranium being 
constant, the needle could thus always be standardized. The readings 
of the electrometer were thus always comparable. The movement of 
the needle was observed by means of a telescope and scale. When 
the needle was at a potential of 200 volts, one scale division corres- 
ponded to a P. D. of -0018 volts, or a little over 500 scale divisions 
per volt. We could thus observe very small rates of leak something 
of the order of 10— amperes being the current generally produced. 
When not taking an observation the quadrants were all connected to 
earth. When a reading was to be taken, one pair of quadrants was 
separated from earth by means of a key operated from a distance by 
a string. The time taken for the needle to move over a certain 
number of scale divisions was observed by means of a stop-watch. 
This gave a measure of the ionization current between the electrodes. 
Sec. IIL., 1902. 5. 
