﻿of the Electron Theory of Matter. 605 



function will be unaltered i£ to is any linear function of 0. 

 The experiments have generally been taken to indicate that 

 i is of the form given by (16) rather than (23 a) ; but I am 

 not convinced that any of them are accurate enough to 

 distinguish between the two expressions with certainty. 



By comparing the terms involving 6 in equation (20) and 

 in equation (51) below we see that 



-!{£-$}• • • • • ™ 



and also since ear is small compared with r it follows that 



^ is very nearly equal to R/(y — 1) or to f R, if 7 = 5/3. 



§ 4. The Reflexion of Electrons at the surface of Conductors. 



Let us consider the atmosphere of electrons which surrounds 

 any conductor, and let us define the group of electrons which 

 strike unit area of any surface indefinitely near the conductor 

 in unit time as an aggregate. A little consideration will 

 show that the reflexion of aggregates possesses properties 

 similar to Stewart and Kirch hoff's laws for the complete 

 electromagnetic temperature radiation. 



If the electronic equilibrium pressure is p, then, as we have 

 seen, 



p = A0y~ l e *« E J ° = »R0, 



where n is the number of electrons per unit volume of the 

 atmosphere. The number of electrons which strike unit area 

 of any surface in unit time is 



{*<*> n™ ,»<» / T> \i 



N = n I i I n f( u • v . w) du dv div = n\ - ) 6* 



Jo J—J— ^ 27rm > 



y x _ w _£ p^ °" j 

 = p(27rmR)-%- | = A.(27rmR)" l (9v" ::l "% ™ E Jo * .(25) 



This is also the number of electrons emitted in the same 

 time by the same area of a perfect absorber (i. e. a substance 

 which does not reflect electrons) at the same temperature. 



If the body is not a perfect absorber, then some of the 

 above N electrons will be reflected by it. Let the proportion 

 be r, Then the number actually absorbed by the conductor 

 in time St is (1 — r)l^St. Let the number emitted by the 

 same surface in time 8t be eNBt. Then the equilibrium 

 condition gives 



e = 1 — r (26) 



Thus the emissivity (compared with a perfect absorber) and 

 the reflecting power are complementary. 



