180 SCIENCE PROGRESS. 



Platinum plate dirty with age connected to a measuring 

 electroscope and illuminated by an arc-lamp, showed a 

 definite fall of potential when charged negatively, passing 

 from mark to mark on the scale in 1 20 seconds ; when 

 charged positively it showed the same fall of potential in 

 100 seconds. Platinum plate scratched clean with glass- 

 cloth, and similarly treated, showed the same amount of 

 leak in 6^ seconds when negatively charged, but when 

 positively charged it still took 100 seconds. 



Here is another example. A clean zinc plate, the same 

 size as the above platinum, leaked at exactly the same rate 

 in the strong rays of an arc-lamp, viz., negative in 6^ 

 seconds, positive in 100 seconds ; but, when a dirty piece of 

 zinc was employed, the time required for the same amount 

 of leak was 30 seconds for negative and 65 for positive. 



Most clean metals on this occasion gave about the same 

 rate of leak, but tin plate seemed rather exceptionally 

 quick. Its rate for negative was 4 seconds, and for 

 positive 80 seconds. 



THE QUESTION OF FATIGUE. 



Lapse of time makes any metallic surface less power- 

 fully photo-electric, simply by reason of tarnishing or 

 accumulation of dust ; but the question arises whether 

 light itself has any effect in accelerating this decay of the 

 power. To try this a couple of similar plates have been 

 kept for some hours, one in the blaze of the arc, the 

 other in comparative darkness, and then their rates of leak 

 compared. The only metal which showed a distinct effect 

 of this kind was palladium. Palladium appears to get- 

 rapidly fatigued by the action of light, so much so that its 

 fatigue is noticeable in a simple series of leakage ex- 

 periments taken one after the other. 



Another experiment, something of the same sort, was to 

 reflect light down on to the surface being tested, by a clean 

 silver-on-glass mirror. The mirror could be kept charged 

 either negatively or positively, and, of course, under these 

 conditions it itself discharged rapidly, especially when 

 negative ; but the question was whether light thus reflected 



