Ultra-Violet Light fnom the Mercury Arc. 



199 



light. For a given plate, when the amount of reflected light 

 is small, it seems safe to assume that the different values of 

 x x are proportional to the corresponding values of .%, and 

 therefore tnat the ratios of various x^s may be used instead 

 of the ratios of # 's. 



Fig. 3. 







ckai 











/ / 

 / / 







.•^~ 



__— — — — ■ 



„_. — bi»<"-"«i.».»i — .....».,..>..„. 





B 



(PctiSllZ'UL 



In Hull's experiments, the leak due to reflected light 

 was ^5 of the leak due to the light incident on the plate ; 

 in these experiments it was about gV? hence the potential 

 X\ required for zero leak is rather less in these experiments 

 than in Hull's. This accounts, in part at any rate, for the 

 difference between the values 2*18 and 2*33 volts. 



Comparing the effects of the light from the two sources 

 when transmitted through quartz, it is seen that the hydrogen 

 discharge produces a greater proportion of faster electrons 

 than the arc. Also the maximum velocity of tne electrons 

 due to the light from the hydrogen discharge and trans- 

 mitted through quartz, corresponds to a potential of 2*18 volts ; 

 while for the light which comes from the arc through quartz, 

 the potential is 1*43 volts. According to Lyman *, the 

 shortest wave-length transmitted by thin quartz from a 

 hydrogen discharge is \ 1450, which corresponds in these 

 experiments to the potential 2*18 volts. Ladenburg's law 

 states that the maximum velocity of electrons due to ultra- 

 violet light is proportional to the frequency of the light. 

 The velocity is proportional to the square root of the potential 

 required to stop the electrons, so that we have \A r ^ X -1 . 

 The wave-length corresponding to 1*43 volts will there lore 

 be X 1780. The experiments show that there is no appreci- 

 able radiation from the mercury arc in the region between 

 * Lyman, Astrophysical Journal, xxv, p. 45 (1907). 



