﻿542 Dr. J. Robinson on the Photoelectric 



All observations of complete radiation show that H m falls 

 off when h is large, that is when the temperature is small, in 

 some such ratio as e~ a]l where a is inversely as the wave- 

 length. Hence 



j H m dh 



is finite and by integration of (101) it follows that ^r(Ji) 

 bears a finite ratio to -\/r(co ) where 



o 

 This, it can easily be seen, is impossible if f{H m ) is not 

 somewhere infinite. Planck has in fact, while retaining 

 the formula (101), supposed that f(H m ) is zero except at a 

 set of points distributed at equal intervals where it becomes 

 infinite. 



Conclusion. 



It now appears that with no continuous laws of motion is 

 it possible to account for the actual distribution of energy. 

 The proof, however, involves the assumption that the statis- 

 tical method applies to the sether with its infinity of degrees 

 of freedom. Since this paper was completed I have succeeded 

 in extending the direct calculation of emission and absorption 

 to matter obeying any continuous laws of motion whatever. 

 That I reserve for another paper ; in this the direct method 

 has shown that for any material system equipartition follows 

 if the principle of least action is assumed. 

 January 29, 1912. 



XL VIII. The Photoelectric Properties of Thin Metal Films. 

 By J. Robinson, M.Sc, Ph.D., Demonstrator in Physics 



at the University of Sheffield *. 



THE photoelectric effect of thin metal films on quartz 

 has been investigated by Kleeman f and by Stuhlmann {, 

 both of whom found a difference in the effects produced by 

 incident and emergent lights. For thin films the emergent 

 light produces a larger ionization current than incident light. 

 Stuhlmann showed further that the reverse is the case for 

 thick films. Both investigators worked with films in air at 

 atmospheric pressure. Kleeman also made a measurement 

 at a low pressure. 



* Communicated by the Author. 



f Kleeman, Proc. Roy. Soc. lxxxiv. 1910, p. 93. 



% 0. Stuhlmann, Phil. Mag. xx. 1910, p. 331. 



