﻿586 Mr. Karl T. Compton on Contact Difference of 



and of the same dimensions. This difficulty was overcome 

 by obtaining the curve for platinum, then electrolytically 

 copper-plating the box and obtaining with it the curve for 

 copper. For convenience in this case the potential between 

 G and B was reduced to 8 volts. Under these conditions 

 the curves of fig. 6 (PI. IX.) were obtained. We see, as 

 before, that the curve for the electropositive metal falls to 

 the left. 



Table II. gives a quantitative estimate of the relation 

 between this shift of the curves and the contact difference of 

 potential between the metals. The second column show T s the 

 measured contact difference of potential between copper and 

 each of the other metals, and the third column shows the 

 amount, w T ith respect to the copper curve, which each curve 

 is displaced along the axis of V. 



These numbers prove convincingly that the contact 

 difference of potential must be taken into account and that 

 the true potential against w^hich the electrons are moving 

 is the reading of the voltmeter plus or minus the contact 

 difference of potential between the plates. 



Table II. 



Metals. 



Contact difference of 

 potential (volts). 



Displacement of 

 curves (volts). 



Cu-Al 



1-02 

 •75 



•55 



•15 



-•11 



1-16 



•70 



•50 



•25 



-•14 



Cu-Zn 



Cu-Pb 



Cn-Fe 



Cu-Pt 



Tables are given * showing the contact difference of 

 potential between different metals, but it is unsafe to rely on 

 these values for accurate work, since the state of polish, the 

 time since polishing, and other factors change these values 

 considerably and frequently even reverse the order of the 

 metals in the series. The only safe procedure when making- 

 measurements in which this phenomenon must be corrected 

 for, is to adopt some method for measuring the contact 

 difference of potential direct at the time of the experiment. 

 Probably many of the inconsistent results which have been 

 published with regard to the photo-electric effect can be 

 explained as being due to the neglect of two factors : (1) the 



* Winkeliriann, Handbvch der Physik, 1st edition, vol. iii. pp. 117-121 



