778 
MR. J. LARMOR ON A DYNAMICAL THEORY OF 
favours somewhat the exact validity of this principle. The exactness of this circuital 
principle seems to be required also by the argument (§ 79) from the equilibrium of 
exchange in an enclosure. For if when a system of rays pass from a point to its image- 
point their relative differences of phases were not the same to a small fraction of a 
wave-length whether the bodies are at rest or in motion, it would follow that the 
distribution of the energy in the diffraction pattern which forms the physical image 
would depend on the movement of the bodies. Thus concentration of the radiation 
might be produced by movements of the transparent bodies, which are subject to 
control. 
The present discussion supposes the motion of the transparent bodies to be practically 
uniform ; the condition [u dx v dy iv dz) an exact differential would be violated 
inside a transparent body in rapid rotation, but then (§ 98) the formula of Fresnel 
would require correction owing to the space-rate of variation of the velocity of the 
material medium. 
Experiments hy Professor Oliver Lodge. 
81. Since this account of the theory was written, Professor Lodge has kindly 
made some experiments on the effect produced by a magnetic field on the velocity 
of light, which considerably affect its aspect. By surrounding the path of the beam 
of light in his interference apparatus'*" by coils carrying currents, he realized what 
was equivalent to a circuit of 50 feet of air magnetized to i 1400 c.g.s.; and he 
would have been able to detect a shift in the fringes, between beams of light 
traversing this circuit in opposite directions, of 5 ^ of a band, or say with absolute 
certainty - 2 ^- of a band, either way. Four coils v/ere employed, each 18 inches long 
and with 7000 turns of wire; and they were excited by a current of 28 amperes at 
230 volts, involving nearly 9 horse-power. The result was wholly negative; and in 
consequence the velocity of light cannot be altered by as much as 2 millimetres per 
second for each c.g.s. unit of magnetic intensity. The cyclic cethereal flow in a 
magnetic field must therefore be very slow ; but the radiation traversing it is of 
course very fast. 
To bring this result into line with the present theory we are compelled to assume 
that the density of the aether is at least of the same order of magnitude as the 
densities of solid and liquid matter, at any rate if we must adhere to the view that 
the motion of the aether carries the light with it. This hypothesis is of a somewhat 
startling character; the density under consideration belongs however to an intangible 
medium and is not apparently amenable in any way to direct perception ; it is on a 
different plane altogether from the density of ordinary matter, and is in fact most 
properly considered simply as a coefficient of inertia in the analytical expression for 
the energy. 
* 0. J. Lodge, “ Aberration Problems,” ‘ Phil. Trans.,’ A, 1893. [There are also some earlier experi¬ 
ments by CoRND.] 
