152 THE ROYAL SOCIETY OF CANADA 
path of the recoiling rest-atoms, at the lower pressures. The longer 
path would allow more of the rest-atoms to reach the anode, since 
their motion during recoil is practically unaffected by the field. 
In similar experiments with actinium Russ‘ found that the ratio 
of the active deposit on a central rod when cathode to that on the rod 
when anode at first increased, reached a maximum and then decreased, 
as the pressure was continually lowered. The final decrease with 
low pressures was probably due to the increased length of the recoil 
path as in the case of radium. The initial increase in the ratio may 
be explained by the fact that the density of the ions decreases with 
decrease of pressure, and, therefore that there is less chance for the 
rest-atoms to recombine. Russ also tried experiments with small 
parallel plates for electrodes. These were placed at different dis- 
tances above a preparation of actinium and it was found that the 
percentage of the total active deposit collected on the cathode in- 
creased as the plates were brought nearer the actinium. This was 
doubtless due to the diffusion to the electrodes of different amounts 
of rest-atoms, which had become neutral in the region surrounding 
the electrodes outside the electric field. 
Kennedy® continued these investigations and obtained about 
the same results as Russ. He used as electrodes two parallel plates, 
circular in shape and surrounded by guard rings. These plates could 
be adjusted both as regards their distance apart and their distance 
above the actinium preparation. He found that the active deposit 
on each plate decreased as the distance from the actinium was in- 
creased, due to the’ decay of the actinium emanation. On keeping 
the distance from the actinium fixed and lowering the pressure, 
the active deposit on both plates first increased and then decreased. 
Similar effects were obtained in carbon dioxide and in hydrogen. 
All these results can be explained by the changes in the density of the 
emanation between the plates caused by the combined effects of dif- 
fusion and decay. 
The behaviour of the active deposit of radium has also been 
investigated by Wellisch and Bronson.6 They used a cylindrical 
testing vessel and small central electrode, which was generally made 
the cathode. As the potential gradient was increased the percentage 
of the total active deposit collected’ on the cathode increased rapidly 
at low potentials and then more slowly until at an average potential 
gradient of about 1,500 volts per cm. this percentage was about 91-5 
4 Russ, Phil. Mag., June, 1908. 
5 Kennedy, Phil. Mag., Nov., 1909. 
6 Wellisch and Bronson, Am. Journ. Sci., May, 1912. 
