Fatigue and Recovery of the Photo-Electric Current 7 



After standing then twelve hours, the following series of ob- 

 servations was taken under the same conditions as before : 



Table III 



21.3 20.9 21.7 



21.5 21.8 22.7 



21.3 ' 21.9 22.5 



17.7 22.7 22.3 



Mean 21.5 



representing 4.1X10"^^" amperes. This shows only a compara- 

 tively slight increase over the values obtained twelve hours be- 

 fore. Following this, hydrogen was again admitted and the 

 metal used in this case as anode during a period of about seven 

 minutes, after which the chamber was evacuated to less than .001 

 mm. and the following series of values obtained for the photo 

 electric current. 



27.5 

 28.0 

 30.1 

 31.5 



or a current of about S-qXto"^*^ amperes which is strikingly close 

 to the first values as given in table I. 



The above results are readily explained in the light of Skin- 

 ner's experiments if w"e assume that hydrogen may also serve as 

 carrier of negative electricity in the photo-electric current. With 

 the unused metal the photo-electric current was 6xio~"- am- 

 peres. By use as cathode with a glow current the quantity of 

 hydrogen present was undoubtedly considerably reduced. The 

 photo-electric current in this condition was also reduced to about 

 two-thirds its original value, 3.8X10"^" amperes. Left at rest 

 for several hours the metal indicated practically no change from 

 this last value, but after use as anode in hydrogen — which 

 charges the metal with this gas — the photo-electric current re- 



131 



