﻿590 Mr. Karl T. Compton on Contact Difference of 



aluminium plate, when there is zero contact difference of 

 potential between the plate and the perforated box and no 

 shift due to stray field through C. 



In fig. 7 (PL IX.) the dotted line, ab, cuts the curve for 

 aluminium (1) at *39. Therefore if ab is taken as the 

 current axis and the point V = is shifted to the foot of ab, 

 the curves will be entirely corrected for all differences of 

 potential, and will therefore represent absolute values of the 

 electron velocities. 



It is evident from these curves that the velocities of the 

 electrons given off by different metals under the action of 

 ultra-violet light are at least so nearly equal that the differ- 

 ences are very small compared to the total velocities. Yet 

 a small real difference seems to exist, for the differences 

 between the curves are too large to be attributed entirely to 

 experimental error. The fact that some of the curves are 

 steeper than others suggests the possibility that the differences 

 in the curves may not be due to any inherent difference in 

 the velocities of the electrons, but to selective absorption of 

 different wave-lengths by the metals. Unfortunately the 

 iron arc gives a great range of wave-lengths, but it is hoped 

 to investigate this point later by a different method. It may 

 also be noticed that these results indicate a maximum initial 

 velocity corresponding to about 4 volts. 



7. Electron Reflexion. — That the electrons liberated by 

 ultra-violet light are reflected to a greater or less degree 

 from the surface of the receiving-plate, and that this reflexion 

 can be prevented by the use of a perforated screen and an 

 auxiliary field, was first demonstrated by 0. v. Baeyer * in 

 1908. He showed that, by thus preventing the reflexion of 

 electrons, the " distribution of velocity " curves are shifted 

 to the left. Since that time many investigators have 

 adopted this device for preventing reflexion. Ladenburg 

 and Markau t used an auxiliary field of 50 volts. But in 

 section 4 it was shown that such an auxiliary field produces 

 a shift of the curves to the left which cannot be attributed 

 to the prevention of electron reflexion. There must be some 

 other factor contributing to the shift to the left. 



In section 4, evidence was given to show that this second 

 factor is the stray field through the perforated screen. It 

 was suggested that curve (a), PI. IX. fig. 4, represents the 

 true shape of the " distribution of velocity " curve, and in 

 section 6 it was shown that when the shift due to the stray 



* Verh. d. D. Phys. Ges. x. p. 96 (1908). 

 t Phys. Zeit. ix. p. 821 (1908). 



