﻿670 Sir J. J. Thomson : Further Studies on 



for steady forces. As the period of the longer waves is 

 8*3 x 10 ~ 14 second, we may conclude that the time taken for 

 an electron in a chain to pass over g times the distance which 

 separates it from its next neighbour in the chain cannot be 

 greater than about 10~ 13 second. If we take the distance 

 between 2 electrons as 2 X 10~ 8 , this would make the 

 minimum velocity of the chains about 2g x 10 5 . This refers 

 to the temperature at which Rubens and Hagen made their 

 experiments — presumably about 15° 0. As the velocity of 

 the chains decreases as the temperature falls, the reflexion 

 from a metallic surface should become at very low temper- 

 atures abnormal at longer wave-lengths than those determined 

 by Rubens and Hagen. 



We can get in another way an estimate of the magnitude 

 of the time taken by an electron in a chain to pass over a 

 distance equal to half the distance between two neighbouring 

 electrons in the chain. At the temperature of 15° C. the 

 wave-length of the light of maximum intensity in the black 

 body radiation is 10" 3 cm.; the time of vibration of this 

 light is I x 10 ~ 13 sec. We should expect from the way we 

 have supposed the black body radiation to arise, that this 

 time would be of the same order as that taken on the average 

 by an electron in the chain to pass over g times the distance 

 between two electrons, and so again we arrive at 10" 13 sec, 

 as being a time of this order. 



On the supposition that cjv is proportional to the time of 

 vibration of the light of greatest intensity, we have 



where 7 is a constant which does not depend on the metal. 



nv 2 y 

 Hence ncv=—^-, 



but \mnv 2 — ^R0, 



m being the mass of an electron ; thus 



nv 2 = 5(9 = 1-51 XlO u X0. 

 m 



Thus ncv = 1'51. 7XIO 11 . 



The specific conductivity 



- e 2 nev 

 =9fP f^- 



_ l'5lxl0 u ygfpe 2 



