Page:Elementary Text-book of Physics (Anthony, 1897).djvu/317

§ 265] discharge can be obtained from it. By similar treatment several such discharges can be obtained in succession. The charge which the jar possesses after the first discharge is called the residual charge. It does not attain its maximum immediately, but gradually, after the first discharge. The attainment of the maximum is hastened by tapping on the wall of the jar. This phenomenon was ascribed by Faraday to an absorption of electricity by the dielectric, but this explanation is at variance with Faraday's own theory of electrification. Maxwell explains it by assuming that want of homogeneity in the dielectric admits of the production of induced electrifications at the surfaces of separation between the non-homogeneous portions. When the jar is discharged the induced electrifications within the dielectric tend to reunite, but, owing to the want of conductivity in the dielectric, the reunion is gradual. After a sufficient time has elapsed, the alteration of the electrical state of the dielectric has proceeded so far as to sensibly modify the field outside the dielectric. The residual charge then appears in the jar. This explanation is supported by the fact that no residual charge remains when the dielectric is a fluid.

265. Tubes of Electrical Force.—If we admit that the nature and condition of the dielectric between conductors determine the charge upon them, an admission which the facts of specific inductive capacity and those cited in the last section render necessary, we must conclude that the hypothesis of electrical charges acting on each other directly at a distance, which we have used up to this point, is an artificial one, and that a more accurate representation of the real state of an electrical field will be had by assuming the action between the electrified bodies to be due to an action in the dielectric. We cannot explain the relation between electricity and the condition of the dielectric which will cause the actions observed between the electrified bodies, but we can show that these actions are consistent with certain conditions in the dielectric which are mechanically possible.

Let us consider a positively charged conductor $$A$$ which is evervwhere surrounded with other conductors. We may assume