When a current is maintained in a wire, the molecular vortices in the surrounding space are kept in uniform motion; but if an attempt be made to stop the current, since this would necessitate the stoppage of the vortices, it is clear that it cannot take place suddenly, but the energy of the vortices must be in some way used up. For the same reason it is impossible for a current to be suddenly started by a finite force. Thus the phenomena of self-induction are accounted for by the supposed medium. The magnetic permeability of a medium Maxwell identified with the density of the substance composing the rotating cells, and the specific inductive capacity he showed to be inversely proportional to its elasticity. He then proved that the ratio of the electro-magnetic unit to the electro-static unit must be equal to the velocity of transmission of a transverse vibration in the medium, and consequently proportional to the square root of the elasticity, and inversely proportional to the square root of the density. If the medium is the same as that engaged in the propagation of light, then this ratio ought to be equal to the velocity of light, and, moreover, in non-magnetic media, the refractive index should be proportional to the square root of the specific inductive capacity. The different measurements which had been made of the ratio of the electrical units gave a mean very nearly coinciding with the best determinations of the velocity of light, and thus the truth underlying Maxwell's speculation was strikingly confirmed, for the velocity of light was determined by purely electrical measurements. In the case also of bodies whose chemical structure was not very complicated, the refractive index was found to agree fairly well with the square root of the specific inductive capacity; but the phenomenon of "residual charge" rendered the accurate measurement of the latter quantity a matter of great difficulty. It therefore appeared highly probable that light is an electro-magnetic disturbance due to a motion of the electric particles in an insulating medium producing a strain in the medium, which becomes propagated from particle to particle to an indefinite distance. In the case of a conductor, the electric particles so displaced would pass from molecule to molecule against a frictional resistance, and thus dissipate the energy of the disturbance, so that true (_i.e._ metallic) conductors must be nearly impervious to light; and this also agrees with experience.