THORIUM IN AN ELECTRIC FIELD. HENDERSON. 15 



To further test this point, the following experiments were 

 tried, the voltage in both cases being 313. 



1st. The testing vessel was placed under a bell jar, the 

 air exhausted and 10.2 cm. of ether vapor allowed to enter. 

 Then air was admitted till atmospheric pressure was reached. 

 The percentage cathode activity was 63.8. 



2nd. The testing vessel was set up under a bell jar as 

 before and the air exhausted to 0.7 cm. Then 10.2 cm. ether 

 vapor were added and the exposure taken under a total pres- 

 sure of 10.9 cm. The percentage cathode activity was 50.8. 



A similar effect in the case of water vapor was sought. 

 The bell jar containing a dish of water in addition to the 

 testing vessel was exhausted to 1.6 cm. and an exposure taken 

 using a potential of 312 volts. Owing to leakage the final 

 pressure was 2.3 cm. The percentage cathode activity was 

 79.9. The exact proportions' of water vapor and air in the 

 bell jar were hard to determine but at least half the pressure 

 was due to water vapor. 



At such low pressures the percentage of the active deposit 

 on the anode is increased due to recoil of the rest atoms. In 

 order to test whether recoil could wholly account for the low 

 value of the percentage cathode activity with water vapor, 

 an exposure was taken using dry air at initial and final pres- 

 sures of 1.35 and 1.55 cm. respectively. The percentage 

 cathode activity was 94.6. 



Thus we see that in an atmosphere of pure ether all the 

 rest atoms are uncharged, which further strengthens the con- 

 clusions drawn from tables 3 and 4. The results with water 

 vapor point towards similar conclusions, notwithstanding the 

 experimental difficulty; the pressure of the water vapor being 

 necessarily small, and there being a considerable fraction of 

 air present. 



Experiments tried in atmospheres of various other sub- 

 stances gave no evidence that in any pure atmosphere the 



