THE ELECTRIC AND LUMINIFEROUS MEDIUM. 
781 
is required on account of this interaction of the linear motion of the medium with 
the rotational motion of the waves. The chief obstacle in the way of a complete 
account of the magnetic phenomena of reflexion appears to be the uncertainty with 
respect to the proper mathematical representation of ordinary metallic reflexion. 
On Radiation. 
85. In accordance with this theory, radiation would consist of rotational waves 
sent out into the aether from the vibrations somehow set up in the atomic charges. 
It has been observed (§ 65) that the characters and periods of these electric 
vibrations, and of the radiations they emit, depend only on the relative positions and 
motions of the vortex-atoms in the molecule, and are quite unaffected, except 
indirectly, by irrotational motion (magnetic intensity) in the aether which they 
traverse. The mode of propagation of electric vibrations in free aether cannot be 
mterfered with by the bodily motion of the medium, however intense, except in so 
far as the motion of the medium carries the electrical waves along with it; a result 
justifying the Doppler principle which is applied to the spectroscopic determination 
of stellar motions. It also follows that radiation will not be set up by motions of 
the surrounding free aether, except in so far as the molecules are dissociated or their 
component atoms violently displaced with respect to each other. To allow the 
radiation to go on, such displacement must result on the whole in the performance of 
work against electric attractions, at the expense of the heat energy and chemical 
energy of the system, which must thus be transformed into electrical energy before it 
is radiated away. The radiation of an incandescent solid or liquid body is maintained 
by the transfer of its motion of agitation into electrical energy in the molecules, and 
thence into radiation. This action goes on until a balance is attained, so that as 
much incident radiation is absorbed by an element of volume as it gives out in turn; 
when this state is established throughout the field of radiation the bodies must be at 
the same temjDerature. 
Conversely, the absorption of incident radiation by a body results finally in a 
diffusion of its energy into irregular material motions or heat, directed motion always 
implying magnetic force. 
86 . There appear to be exjDerimental grounds for the view that a gas cannot be 
made to radiate [at any rate with the definite jaeriods peculiar to it] by merely 
heating it to a high temperature, so that radiation in a gas must involve chemical action 
or, wffat is the same thing, electric discharge. This would be in agreement with the 
conclusion that motion of a molecule through the aether, ho\vever the latter is 
disturbed, wall not appreciably set up electric vibrations, unless it comes well within 
them could be produced ; also that a €ow along these tubes need not produce any disturbance in the 
other properties of the electric field [; also that the magnetic rotation being a purely material pheno¬ 
menon, whose direction is not subject to any definite law, it most be of a secondary character]. 
