﻿Potential and the Velocities of Electrons. 585 



for C, it was necessary that these boxes be exactly similar 

 in dimensions and that the field between C and B be kept 

 constant. 



5. The Effect of Contact Difference of Potential. — The first 

 metals investigated were aluminium, zinc, lead, and copper. 

 For this purpose the perforated foil boxes of these metals 

 were successively used as C, and the curve was plotted for 

 each one. It was noted in section 2 that the boxes of these 

 metals were all practically similar, and thus they could be 

 compared with each other without introducing the error 

 entering in when the boxes are geometrically dissimilar. 

 As a preliminary step a very large number of readings were 

 taken and curves plotted which showed conclusively that the 

 actual shape o£ the " distribution of velocity " curve is 

 independent of the metal used as C. When once the shape 

 of this curve had been very carefully determined, only one 

 or two points on each curve were sufficient to completely 

 determine it. This enabled the readings for any curve to be 

 taken rapidly, thus lessening the change in the contact 

 difference of potential during the experiment and permitting 

 the two metals which were to be compared to be investigated 

 the same day. This reduced the danger of introducing any 

 error due to a change in the setting of the apparatus. As 

 an additional precaution, the following procedure was 

 adopted, Take, for instance, the perforated boxes of zinc 

 and copper. A curve was taken using copper, then one 

 using zino, and again one using copper. If there was any 

 difference in the positions of the two curves for copper, the 

 mean of the two was taken as the correct position. In this 

 way the curves of fig. 5 (PL IX.) were obtained. For all 

 of these the difference of potential between C and B was 

 400 volts. 



It is seen that the curves for the case of the electro- 

 positive metals lie to the left and in the order of the electro- 

 positive series. This is just what would be the case if the 

 contact difference of potential between the aluminium plate, 

 D and the perforated box C influences the paths of the 

 electrons. A box C of an electropositive metal would 

 tend to attract the negative electron more than would an 

 electronegative metal. Thus, for the same externally 

 applied difference of potential, many electrons would reach 

 the electropositive box which would fall short of the electro- 

 negative box, and hence the curve for the former would lie 

 to the left of that for the latter. 



In order to test the platinum box special means had to be 

 employed, since there was no other box of a different material 



