of Electricity in an Imperfect Insulator. 421 



the induction-tubes. (See fig. 1, where 1-6 may be taken as 

 successive positions of a tube.) During the motion of the 

 tubes, some of their energy was converted into the magnetic 

 form, the coexistence of the two forms, electric and magnetic, 

 being a necessary condition of motion. We may illustrate 

 this from the analogous case of a strained incompressible solid 

 which can be sheared. If there is any mode of escape given 

 to the strain-energy by a slipping of the surface against the 

 constraint, then the state of strain will be propagated outwards 

 from the interior of the solid, but some of the strain-energy 

 will be converted into kinetic energy, and the presence of the 

 two is a necessary condition for the propagation of the strain. 

 Since the energy-length of a tube diminishes as its ends 

 move along the connecting-wires, we may represent this by 

 supposing that parts of the tubes move into the wire. If a 

 similar motion of electric induction took place into a dielectric, 

 it would remain, and the dielectric would become electrically 

 strained ; but in the wire the strain breaks down rapidly, the 

 energy being converted into heat. I think there is good reason 

 to suppose that it is the electric energy which thus breaks 

 down, the magnetic only being dissipated after it has been 

 reconverted to the electric form. 



We may now consider the case in which the total discharge 

 of a condenser takes place through a connecting wire. Con- 

 sidering merely the conducting plates and the wire, we say 

 that the charges move along them towards each other and finally 

 unite, neutralizing each other and producing heat in the wire. 

 Regarding the medium, we must suppose the tubes of induction 

 to move sideways towards the wire, shortening as their ends, 

 which are represented by the charges, approach each other, 

 and finally disappearing into the wire. Faraday describes the 

 process by saying that " when current or discharge occurs be- 

 tween two bodies, previously under inductrical relations to 

 each other, the lines of inductive force will weaken and fade 

 away, and, as their lateral repulsion-tension diminishes, will 

 contract and ultimately disappear in the line of discharge " 

 (Exp. Res. vol. i. p. 529, § 1659). The so-called velocity of 

 electricity is merely the velocity of the ends of the tubes, and 

 this may evidently vary according to the nature of the circuit. 

 It is quite conceivable that if the wire be in a neutral medium, 

 i. e. one in which there is no surface -difference of potential, 

 say gold in air, and if it follow the direction of a tube of in- 

 duction, then a tube may move into the wire throughout its 

 whole length at once. In this case the " velocity of electricity " 

 would be infinite. 



