290 
MR. J. LARMOR ON A DYNAMICAL THEORY OF 
practically (1 — 7??E/2p,, where m, is the real part of the index of refraction of 
the medium, as measured by the ratio of the velocities in deviation experiments with 
prisms. And in general it appears that it is only absorption, not reflexion, of 
radiation that is accornjaanied by a mechanical forcive, the force on any absorbing 
mass being (1 — ? 7 iE 72 /rC in the direction of the radiation traversing it, where 
E' is the total radiant energy absorbed by the mass per unit time. 
In the case of a transparent medium traversed by two systems of waves, direct and 
reflected, forming stationary undulations, the mechanical force is proportional to 
sin 2qx, and vanishes at both the nodes and autinodes of the electric vibration. 
74. The rationale of the mechanical forcive is that, owing to the absorption of 
energy, which can only occur when phase-diflerences exist, a difference of phase 
becomes established between the two factors of X, the electric current and the 
magnetic field, so that their product contains a non-alternating part. It is known 
that vapours of complex chemical constitution are very powerful absorbers of radia¬ 
tion ; and in this case (if not in all cases) the absorption must he a property of the 
single molecule. By the argument just stated, there must then be difference of 
phase between the electric flux (displacement of electrons) in the molecule and the 
magnetic field, and this wull give a mechanical forcive driving the molecule along the 
path of the radiation. As the tails of comets and the Solar Corona consist of very 
rare distributions of vaporous or other matter, in free space which exerts no retarding 
influence, a comparatively small absolute amount of absorption by them of the Solar 
radiation might account for their observed repulsion from the Sun; in this way a 
definite and actually existing physical agency may be made to take the jflace of 
vague electrical repulsion in Bkedichin’s important analysis of cometary appendages. 
This mechanical action of wmves on absorbing systems placed in their path may be 
roughly illustrated by an arrangement in wdiich a system of sound waves traverses 
a sjoace filled with resonators approximately in unison with them. The open mouth 
of each resonator is repelled,"^ so that in case there is any regularity in their 
orientation, the system as a whole will be subject to mechanical force. The 
resonators might be suspended so that the mechanical forces may themselves produce 
this orientation, and thus form a sort of medium polarizable by waves. A corre- 
ponding electric illustration is the action of long Hertzian waves in orientating and 
repelling mobile conducting circuits wdiich lie in their path. The very considerable 
repulsion of the vanes in the radiometer arises of course from a mutual stress 
betw’een the vanes and walls and the rarefied gas, and so has a null resultant as 
regards the system as a wdiole. 
Stresses and Deformations in Electric Condensers. 
75. The elastic deformation produced in the dielectric of a spherical condenser by 
Cf. Lord Rayleigh, ‘ Theory of Soiiud,’ vol. 2, §§ 2o5a, 319. 
