﻿On the Photoelectric Effect. 619 



Standards. This was set up at such a distance from the 

 thermopile as to give rise to an electromotive force con- 

 venient to measure and comparable with those obtained with 

 the monochromatic illuminator. The directions supplied 

 with the standard were carefully followed. As a result of 

 these measurements it was found that the unit of energy per 

 unit time previously employed was equal to 0'404 erg per 

 sec. For example, to reduce the values of E in Table IV, 

 column 6, p. 562, to ergs per second, it is necessary to 

 multiply each number by 0*404. 



At this time the illuminator was readjusted and the energy 

 distribution in the quartz-mercury arc-lamp spectrum re- 

 determined. The values obtained were not sufficiently 

 different from those given in the table just referred to, to 

 call for comment. 



We can now consider the results for each of the elements 

 separately. The value of N, the quantity of* electricity 

 liberated when unit quantity of radiant energy falls on the 

 metal, will be expressed in terms of the unit 1 coulomb per 

 calorie. To reduce the numbers given to electrostatic units 

 of quantity per erg, it is only necessary to multiply them by 

 72. The quantity N may also be appropriately termed the 

 photoelectric yield, although it would be better to restrict 

 the term photoelectric yield to the case in which the light is 

 completely absorbed. The value of the photoelectric yield r 

 in this sense, can be obtained from the data given, together 

 with a knowledge of the reflecting power, in the different 

 parts of the spectrum, of the metals concerned. 



Platinum (loc. cit. p. 561). 



Unit of N= 9-05 x 10" T coul./cah 

 Greatest value of N (at y=l*5 x 10 15 sec. -1 ) 



= 3*0xl0- 5 coul./cal. 

 Aluminium (loc. cit. p, 562). 



Unit of N = l-91 x 10 -6 coul./cal. 

 for curves 1, 3, and 4. 

 Curve 1 (taken immediately after scraping the aluminium 

 and setting up). 



Maximum value of N Tat i/ = l'36 x 10 15 sec. -1 ) 



= 2*25 x 10" l couL/cal. 

 Curve 3 (72 hours later). 



Maximum value of N (at 1*42 x 10 15 sec. -1 ) 



= l-OlxlO - 'couL oak 

 Curve 4. This was for a flat strip at perpendicular 

 incidence and was taken about 6 hours after scraping and 

 setting up. 



Maximum value of N (at 1*38 x 10 15 sec. -1 ") 



= 1-28 X 10-' coul. eah 



