of attraction and bombardment would be the void focus of the disk, toward which the resultant of all the forces in the ring would tend, as a corollary from the deduction from synchronism that the force which binds the atomic couples varies directly as the distance, instead of as the inverse squares.
In its passage through the disk-atom the ray takes up and conserves the dropped motion as a transverse vibratory motion of some kind; or, as Maxwell styles it, "some vector property which does not interfere with the motion of translation" ("Encyclopædia Britannica," ninth edition, article "Ether"), and which it can impart again to the revolving systems of atoms through which it travels, in an inverse mode to that of its derivation. The invention of the mechanism of this vector motion has its difficulties. I consider, first, that the radiant energies are not exhibited in matter, except upon a certain degree of disturbance, showing itself in a violent clashing of systems, either from the forming of new combinations or from incandescence resulting from the accession of energy from without or within. These clashings disturb the equilibrium of the atomic orbits, and occasion their rapid deformation by harmonic vibration in elliptic orbits whose rectangular axes rapidly alternate from major to minor while the agitation is kept up. Now, the circular movements of the components do not disturb the uniform transit of the linear ray; but the rapid approach and recession of the components in passing through the violent elliptical transitions cause a rapid alternation of stress in the field of stress through which the rays pass—since the stress varies as the distance of the components—causing a vibratory deflection of the stream of particles, due to the variation of the attractive force, in all rays except the polar ray, and in the plane of the ray normal to the plane of stress. Since the disk-atoms lie in all planes, we shall have transverse vibrations in all directions; but, if the rotating disk is gyratory, as would be the rule and not the exception, the identical ray would receive a vector or corkscrew motion, similar to what is called for by observation. The amount of deflection I take to be the index of refraction of the molecule.
It may seem unaccountable that the whole ray should undulate from passing through a single locality of oscillatory disturbance, being composed of discrete and unconnected particles; but we may compare the parallel phenomenon of a jet of water, forcibly ejected to a great distance through a hose-nozzle, which exhibits to the eye similar undulations when the source of discharge is slightly and rapidly oscillated.
It may also seem incredible that any orbital movements could be permanent enough to sustain oscillatory vibrations of such inconceivable frequency as those which luminous rays are known to execute; but I have computed, from the probable dimensions of the hydrogen molecule as assigned by the molecular physicists, and an orbital veloci-
