Photo-Electric Effect of some Compounds. 385 



Ferric Chloride. 



Leak. 



First exposure •<* '2 



Exposed 15 mins *9 



Exposed 20 „ 1*0 



Another layer was distilled and kept in the dark for 

 1| hours. On being exposed to the ultra-violet light for the 

 first time, the leak was < *2, as before. 



In view of the results of some experiments on deliquescent 

 compounds, described later, the effect of admitting moist air 

 for two hours was tried. It was then pumped out and a high 

 vacuum was obtained by the liquid air method. The photo- 

 electric leak was practically the same as before, being 1*2; 

 after 40 mins. exposure it rose to 1*5. While the sur- 

 faces of the ZnCl 2 and P 2 5 retained a moist appearance 

 after contact with air, the ferric chloride surface became 

 quite dry again in the liquid air vacuum. This is probably 

 the reason why there is no appreciable increase in the leak. 

 It is known that the hydrates of: ferric chloride have a 

 vapour pressure, and consequently the salt will part with 

 its water of crystallization in a vacuum. 



Zinc Chloride. 



Leak. 



Eirst exposure < '2 



Exposed 1 hour •< -2 



Exposed 1 „ < -2 



This shows that a dry surface of zinc chloride is not photo- 

 electric. Air from the room, not dried in any way, was 

 admitted for 5 minutes at atmospheric pressure and then 

 pumped out again. The surface did not appear to be w r et 

 as far as one could judge from the appearance of the deposit 

 on the disk when viewed through the glass portion of the 

 apparatus. 



After admitting air 160 



Exposed 5 mins 240 



Exposed 5 ,, 255 



Moist air was then passed into the apparatus and pumped 

 out as before. The surface now appeared to be almost 

 liquid. 



After admitting moist air 250 



In dark 5 mins 300 



Exposed 5 „ 340 



