﻿620 Prof. 0. W. Richardson : Some Applications 



close to the surface of any substace quite apart from the 

 mode of emission. Thus the same type of relation would be 

 satisfied by the electrons emitted thermionically if there 

 were no photoelectric action, and by the electrons emitted 

 by the complete sethereal radiation if there were no thermionic 

 emission. In fact (51) is entirely consistent with the results 

 of a calculation of the thermionic emission based on the 

 simple hypothesis that there are free electrons inside the 

 metal which escape in virtue of their kinetic energy ex- 

 ceeding the work which is necessary for them to escape from 

 the surface. 



The question at once suggests itself as to whether any 

 true thermionic emission really exists, or in other words,, 

 whether what is usually considered to be thermionic emission 

 is not in reality the photoelectric emission of the material 

 arising from the sethereal radiation at the temperature of the 

 experiments. This is a view which occurred to me as a 

 possibility many years ago, and I think the balance of ex- 

 perimental evidence is against it. When one considers the 

 large electronic emission of hot platinum and carbon and 

 reflects upon the intensity of a beam of ultra-violet light 

 which would be required to produce an equal effect photo- 

 elect rically, the probability of such a view being correct 

 becomes exceedingly small. On the other hand the problem, 

 considered from this standpoint, involves the comparative 

 penetrability of light and slow moving electrons, a matter 

 about which very little is known. 



Leaving this question aside, let us see what we can find 

 out about the connexion between photoelectric emission and 

 the light which causes it. For the present we shall suppose 

 that we are dealing with a material which is photoelectrically 

 active but has no thermionic emission. No such material 

 may exist, but the assumption simplifies the discussion and, 

 as we shall see later, the results will probably be applicable 

 to the case of real bodies. We shall suppose the material 

 to be in a perfectly reflecting enclosure and the steady 

 state characteristic of the temperature 6 to have been 

 established. We may assume the photoelectric emission to 

 be dependent either upon the density of the sethereal radia- 

 tion present or upon its rate of absorption. We shall take 

 the latter as being the more general. Our results can then 

 be adapted to the former hypothesis by making the emis- 

 sivity e equal to unity. Let the steady energy density of 

 the vibrations whose frequencies lie between v and v + dv be 

 E(v)dy, then the energy emitted per unit time which is 



