﻿270 



Prof. 0. W. Richardson' on the Electron 



following thermodynamic argument. Suppose that we have 

 an electrical circuit made up of two metals A and B, such 

 that the parts A and B with the junction J are maintained at 

 the constant temperature 0, and A' and B' are at the constant 



temperature 9'. The connexions between A A / B' and B may 

 be imagined as being made by conductors of small cross- 

 section protected by insulating covering when necessary for 

 the argument. The connexions are removable when required, 

 and the magnitude of the conductors A A 7 B' and B is supposed 

 to be indefinitely large. Each conductor is surrounded by 

 an insulating surface which does not absorb the electrons 

 contained within it. A reversible c\cle may be carried out 

 by evaporating some definite quantity, let us say one gram 

 molecule, of the electrons from A, condensing them at A', 

 making them pass through the junction J' to B', evaporating 

 at the surface of B', condensing at B, and returning through 

 the junction J to A. In order to be sure that the cycle is 

 reversible it is necessary to be careful about the way in 

 which the electrons are transferred from A to A' and from 

 B' to B. Of course as the different parts of the system are at 

 different temperatures the state of each conductor will never 

 be exactly that corresponding to thermal equilibrium at the 

 specified temperature, but it may be made to approach that 

 state with any desired degree of approximation by increasing 

 the size of A A' B' and B in comparison with that of their 

 connexions and the enclosure, the temperatures being main- 

 tained by contact over indefinitely large areas with electrical 

 insulators kept at the specified temperatures. Under these 

 circumstances the conductors will assume values of w, n, and 



