02 Mr. J. Larmor. Theory of Electrodynamics [June 2, 



or else it must be non-existent. The precise logical statement of 

 Quincke's results is, in fact, that the difference between the electric 

 stress in a ponderable fluid dielectric K and 'the electric stress in a 



vacuum, in a field of force F, consists of a tension -^r^ 2 along the 



lines of force, combined with a pressure of equal amount in all direc- 

 tions at right angles to them ; and this is consistent with a distribu- 

 tion of polarisation energy in the fluid, added to the electric energy 

 for a vacuum with the same intensity of force, but not entering into 

 combination with it. 



Now the propagation of electrical waves across air or vacuum shows 

 that even then, when there is no ponderable dielectric present, there 

 must be a store of statical energy in the dielectric ; and this fact 

 appears to remove the only explanation which seems assignable for 

 the division of the energy into two parts, one located in the dielectric, 

 and the other located on the plates and absolutely independent of the 

 dielectric, viz., that the latter might be the energy of a direct action 

 across space which is not affected by the dielectric. The experimental 

 facts, therefore, so far tend to the conclusion that at any rate the 

 basis of electrical theory is to be laid on Maxwell's lines, with a 

 reservation for possible modification in the form of residual correc- 

 tions, but not for change of principle. 



A theory has been developed by Helmholtz for fluids, and by 

 Kirchhoff, following him, for solid dielectrics, in which slight residual 

 differences between the intensities of the tension and pressure may be 

 accounted for on the supposition that the inductive capacity, instead 

 of being constant, is a function of the electric force. This theory is 

 primarily expounded in terms of a polarisation scheme, and in so far 

 is subject to the remarks of the next section ; but it may in the end 

 be based, as Helmholtz suggested, on the principle of energy applied 

 with the aid of the ascertained form of the characteristic equation of 

 the potential, treated as a condition of internal equilibrium. If we 

 adopt the view that the difference to be explained has not certainly 

 been detected,* this theory need not here be considered. 



Some of the points in the general treatment given above will 



Ether ................. 4' 62 4" 66 



Carbon disulphide ...... 2 '69 2 *75 



Benzol ................ 2 -32 2 '37 



Turpentine ............ 2 '26 2 '35 



Petroleum ............. 2 -14 2 -15 



The chief difficulty seemed to be to avoid conduction, owing to want of purity of 

 the dielectric fluid. 



* Cf. Bos, " Inaugural Dissertation," abstracted in ' Philosophical Magazine ' 

 February, 1891. 



