﻿of the Electron llieory of Matter. 601 



Thus all the quantities on the right of (12) are known 

 -except N and n. The best estimates indicate that for the 

 good conductors N is a moderate fraction of the number of 

 atoms jo in unit volume of the material. Let us put N =yp, 

 then 7 probably varies between 0*1 and 0'5 for the different 

 good conductors. Nothing is known of the structure of the 

 centres of force : they may either be atoms or subatomic 

 structures, but it is unlikely that they are polyatomic. If 

 we put n = Sp it is probable that 8 will have unity as a lower 

 limit. 



Let us apply formula (12) to the case of silver. Sub- 

 stituting the values <r = 6xl0~ 4 e.m. unit, vR = 3*72xl0 3 

 ergs/°C, v/w = 5xl0~ 8 gin., e = l'6xl0~ 20 e.m. unit, 



4=273° C., and C&m 2 <f> a da = '396, we find 

 « 



-K = 8-5xl0" 26 (14) 



7 



From the constants in the radiation formula, Thomson 

 estimates K as 2 x 10~ 27 , while Jeans gives 3 x 10 -27 . Thus 

 these very different methods of estimating K lead to quantities 

 of the same order of magnitude. 



Substituting the mean 2*5xl0~ 27 of the two radiation 

 estimates of K in (14), one finds 8/7=34. If 7 = ^ for 

 silver this would give 8=11, indicating that the centres 

 of force are subatomic structures. A similar conclusion 

 results in the case of the other good conductors. 



There is a superficial inconsistency in formula (12), inas- 

 much as cr is known to be proportional to 6~ l for the pure 

 metals. However, N probably varies with 6, and very likely 

 n also. The elementary theory of the Thomson effect leads 

 to the view that N<x(9 +3/2 , which would make c<x. 6 2 if n 

 were constant. It is probable that the N which occurs in 

 thermo-electric phenomena is not the same thing as the IS" in 

 equation (12). (See § 5.) 



At the very lowest temperatures, according to Kammerlingh 

 Onnes, the electrical conductivity of metals becomes enor- 

 mously high. According to the present theory this would 

 mean either that the number of the centres diminishes 

 greatly or that their strength K becomes very small at these 

 temperatures. 



§ 3. The emission of Electrons from Hot Bodies as a function 



of Temperature. 



In 1901 * the writer showed, from considerations similar 



to those made use of in the kinetic theory of gases, that the 



* Camb. Phil. Proc. vol. i. p. 286. 



