44 Mr. C. V. Burton on 



electricity across the junction, until the difference of their 

 potentials* has an E.M.F. equal and opposite to E. I shall 

 assume that E is always the same so long as we keep to the 

 same substances and the same temperature ; and also that 

 when electric equilibrium has been established each con- 

 ductor is at a uniform potential throughout, except very near 

 the junction, and that there the potential changes in general 

 by a finite amount as we pass through a layer of very small 

 but finite thickness, which includes the surface of separation. 

 Let the two conductors A and B (fig. 1) be in contact over 



Fig. 1. 

 c 



D 



the surface CD, and let the difference of their potentials be 

 finite. The potential is constant throughout the extent of A, 

 and also throughout the extent of B, except within a very 

 thin layer including the surface CD, which may be called the 

 layer of variable potential, and where the potential may be 

 supposed to change continuously from that of A to that of 

 B as we pass from one side of the layer to the other. Now, 

 consider the distribution of electricity on the system. Elec- 

 tric masses will be distributed on the external surfaces of A 

 and B, but not within these surfaces, except within the layer 

 of variable potential. Here the distribution evidently depends 

 upon the thickness of the layer, the law of variation of 

 potential across the layer, and the specific inductive capacity 

 of A and B. This last quantity is evidently the true or 

 finite specific inductive capacity, and not the apparent or 

 infinite specific inductive capacity which conductors appear 

 to have in ordinary electrostatic experiments ; for in the 

 latter case the electric potential energy of the layer CD 

 would be infinite. 



It is usually supposed that the contact E.M.F. between two 

 conductors is due to some molecular action between them. 



* Wherever the potential of a conductor is spoken of, the potential at 

 any point throughout its volume is intended, and not the potential at a 

 very near external point. I must apologise for this unusual use of the 

 term. 



