X-Rays from Boron and Carbon. 151 



obtained (giving 150 volts as before). It is remarkable, 

 however, that whether the break was but slight, or abrupt, 

 it always appeared in the vicinity of 150 volts. When it 

 was found that conditions which gave rise to the curves in 

 fig. 2 could not be reproduced at will, and that generally 



Fiir. 2. 



Photo E/ectr/'c Effect 

 E-/ectron Current. 











BOPC/V 'WPacZ/at/ou. 























/End 





Afea/? Curve-. 









/ t 









/%^ 







Start 



m- 



sd&a=ns»*«»_ .ov 



- 



j?^^ 



s* 







jfr n' Qnr~u« 





A cce 



/erat/ny 



Pott 



=/7t/'a/. 



50 



100 



l r 



200 



SO 



curves with slight breaks, e. g. fig. 3, were obtained, efforts 

 were concentrated on locating these slioht breaks. In view 

 of progressive time- changes (generally smaller than those 

 shown in fig. 2), it was decided to restrict observations to a 

 certain number of selected accelerating potentials, so as to 

 complete a run in a short time (thirty to fifty minutes), and 

 to repeat these runs with increasing and decreasing potentials 

 alternately. (One could have spent the same total time on 

 one run only, but with from four to six times as many 

 points. However, the procedure adopted was considered 

 more satisfactory.) Then, if each run individually showed a 

 definite break, the average of four or six consecutive runs 



