124 



BRIDGMAN. 



all metals at the same temperature, and the same metal under different 

 pressures at the same temperature is merely a special case of two 

 different metals. As a matter of fact the ratio is not constant, but 

 may either increase or decrease with increasing pressure; in the ma- 

 jority of cases it decreases. The average values of the coefficient 

 between and 12000 kg. are shown in Table III. 



TABLE III. 



My own theory of electrical conduction attempted to explain the | 

 Wiedemann-Franz ratio,® and to do this, I imagined the .same sort of 

 mechanism of conduction as the classical theory. I still can see no 

 reason to suppose that the most important part of heat conduction is 

 not as imagined by the classical theory; the success of the theory in 

 accounting for the numerical value of the ratio, which is approximately 

 constant for the different metals, (it may vary from 6.38 X 10^" for 

 aluminum to 9.14 X 10'" for bismuth), is too striking to be put aside \ 

 with no substitute. At the same time it is evident that the account 

 given by the classical theory cannot be complete; no account has been 

 taken of the conduction by the atoms, and the agreement of the 

 theoretical with the experimental value is not as close as we must ' 

 demand of a finished theory. 



It is natural to look to the still unexplained part of thermal con- 

 ductivity to account for the departures from constancy of the ^Yiede- 

 mann Franz ratio under pressure. The part of the conductivity 



