





the Photoelectric 



Effect. 







58 



Metal 



X, 



X. 



Tm. 



TV 



x/x . 



Metal 

 X . 



X. 



-Lwi« 



T r . 



X/Xo. 





436 



0-60 



030 



0-755 





265 



0-92 



0-52 



0-740 



Na 

 57-7 



36-6 



TOO 



0-50 



0635 



Zn 



25-3 



1-25 



0-625 



0-713 



31-3 



1-50 



075 



0-542 



1 35-7 



230 



1-70 



0-862 



0-644 



254 



2-30 



1-15 



0-440 





210 



203 



106 



0-590 





210 



300 



1-50 



0-365 





26-5 



0-80 



0-432 



0-785 





313 



090 



0365 



0-760 



; Sn 



25-4 



110 



0-520 



0-750 



Al 

 411 



27-5 



1-30 



0-60 



0670 



! 33-7 



23-8 



1-33 



0-660 



0-705 



253 



1-50 



0-74 



0615 





22-0 



1-73 



0-810 



0-650 



23-0 



1-90 



0965 



0-560 















200 



2 30 



1-21 



0-485 



Bi 

 33-7 



25-4 



0-60 



0-315 



0-750 













230 



0-90 



0-450 



0-680 





275 



0-85 



0-45 



0-730 



21-0 



1-15 



0-587 



0-620 



Mg 



264 



1-02 



0-585 



0-705 













390 



254 



1-35 



0-638 



0-675 





26-0 



0-35 



0-175 



0-840 





230 



1-80 



0-85 



0-612 



Cu 

 30-9 



25-4 



0-48 



0-230 



0820 













230 



0-73 



0-335 



0-745 



Pt 

 291 



250 



051 



0-266 



0-870 



210 



1-02 



0-475 



0-680 



230 



099 



0-47 



0-790 





20-0 



1-25 



0-550 



0-645 



21-0 



1-45 



0-713 



0-720 













The unit for X is 1 = 10 6 cm. T m and T r are expressed 

 in equivalent volts. 



small, and the maximum potential observed is necessarily 

 that at which the photoelectric current is balanced by the 

 minute back leakage which is unavoidable in an apparatus 

 of this character, and which arises chiefly from the ioniza- 

 tion of the air in the shielding tubes and the like. It is 

 possible that our measurements of the maximum energy 

 might have been improved by using a source of ultra-violet 

 light of greater intensity, as the Heraeus quartz mercury 

 lamp was chosen for its steadiness rather than on account of 

 the intensity which it furnished. However, it is probable 

 that the chief sources of error in experiments on photoelectric 

 action arise from chemical effects at the surfaces investigated 

 rather than from defects in the electrical and optical arrange- 

 ments used. 



When the foregoing values of T TO and T r are plotted against 

 the corresponding frequencies v, the results for each metal 

 exhibit a linear relation between the corresponding variables. 

 Except in the case of copper and bismuth the lines are almost 

 parallel to each other for all the metals. The frequencies 

 v , and the corresponding wave-lengths X, > which correspond 

 to T m and T r =0, are determined by the intersection of these 

 lines with the frequency axis. Together with the constant k, 



