1892.] 



On the Theory of Electrodynamics. 55 



In conclusion, I would express my obligations to Mr. W. Esson for 

 assistance in the mathematical portion of the paper, and to my 

 colleagues for suggestions made in the course of the investigations. 



VII. " On the Theory of Electrodynamics, as affected by the 

 Nature of the Mechanical Stresses in Excited Dielectrics." 

 By J. LARMOR, M.A., Fellow of St. John's College, Cam- 

 bridge. Communicated by Professor J. J. THOMSON, F.R.S. 

 Received April 25, 1892. 



1. A theory of electrodynamics was first precisely developed by 

 Maxwell, which based the phenomena on Faraday's view of the 

 play of elasticity in a medium, instead of the conception of action at 

 a distance, by means of which the mathematical laws had been 

 primarily evolved. The electromotive equations of Maxwell how- 

 ever involve nothing directly of the elastic structure of this medium, 

 which remains wholly in the background. They involve simply the 

 assumption of a displacement across dielectrics with such properties 

 as to make all electric currents circuital ; all the equations of Ampere 

 and Neumann for closed or circuital currents have then a universal 

 validity, and no further hypothesis is required for the full develop- 

 ment of the subject. 



The theory was next discussed by Helmholtz in his memoirs on 

 electrodynamics, in a way which took direct advantage of the picture of 

 a polarised dielectric supplied by Mossotti's adaptation of the Poisson 

 theory of induced magnetisation. Stated absolutely, this simply 

 builds upon the assumption that at ecah point in the excited di- 

 electric there is something which has the properties of a current 

 element (electric transfer or displacement), which is represented both 

 in direction and magnitude by the electric force at the point multi- 

 plied by a constant factor; no more general starting point seems 

 possible for an isotropic dielectric. The development of this hypo- 

 thesis, exactly on the analogy of a similar discussion with the 

 Poisson-Mossotti phraseology in a previous paper,* leads to the 

 necessity of recognising the existence of absolute electric charges on 

 the faces of an excited condenser ; so that the exciting current causes 

 the accumulation of these charges, and therefore is not circuital or 

 solenoidal. This defect of circuital character however practically 

 disappears in the limiting case when the constant ratio of the polar- 

 isation to the electric force is extremely great ; and then the theory 

 becomes a concrete illustration of' the general statements of Maxwell 

 with respect to electric displacement. 



The Theory of Electrodynamics," ' Koy. Soc. Proc.,' 1891. 



