10 DISTRIBUTION OF THE ACTIVE DEPOSIT OF 



refers only to a time after that infinitesimal fraction of a 

 second during which recoil is operative. This curve differs 

 in two respects from that found by Wellisch and Bronson 

 for radium. (1) It reaches approximate saturation at much 

 lower voltages. (2) There seems to be no evidence that any 

 fraction of the rest atoms is initially unchanged. Experi- 

 ments showed that the dimensions and form of the testing 

 vessel and the concentration of the rest atoms are largely 

 responsible for (1), and that approximate saturation in air is 

 reached with about the same potential gradient for thorium 

 as for radium. 



Thus the evidence in the case of dry air would indicate 

 that initially all the rest atoms of thorium are positively charg- 

 ed and that all the activity on the an ode at low voltages is due to 

 these rest atoms losing their charge by recombination with 

 negative ions in the air. 



It was found by Wellisch and Bronson that 10.4 per cent, 

 of radium rest atoms were initially uncharged in dry air. 

 Experiments were made to test whether this difference be- 

 tween radium and thorium was due to the type of vessel used. 



The percentage cathode activity in the case of radium 

 was measured with parallel plate vessels. The source of 

 radium emanation used was a layer of radium chloride de- 

 posited on a sheet of aluminum. Preliminary measurements 

 with the first type of vessel described, showed that this frac- 

 tion was of the same order as that found by Wellisch and 

 Bronson for cjdindrical vessels. Later the experiment was 

 repeated using the second type of vessel and a potential of 

 about 12,000 volts supplied by the Wimshurst machine. The 

 percentage cathode activity was found to be 5.2. This would 

 make the total fraction unaffected by the electric field 10.4 

 per cent, in good agreement with the latest value (11.8) found 

 by Wellisch^ 



1. Wellisch, Phil. Mag. Oct. 1914. 



