764 
MR. J. LARMOR ON A DYNAMICAL THEORY OF 
Hutchinson’s experiments,'^ or even if there is a single line of division, each portion 
will carry its field of electric displacement along with it, the field preser\fing its 
statical configuration under all realizable speeds of rotation. If the scratches did not 
run up to the centre of the disc, the field of displacement due to the central parts 
would be quiescent, and the displacement-currents would be altered in character.+ 
The dielectric displacement in the experiments above-mentioned, with two parallel 
rotating gilt glass condenser-discs having radial scratches, is across the field from one 
disc to the other, and is steady throughout the motion ; so that the convection- 
currents are completely represented by the simple convection of the electric charges 
on the discs, and are not spread over the dielectric field. 
60. The motion of a dielectric body through a field of electric force ought also to 
carry its system of electric displacement along with it. It appears that RoNTGENij: 
has detected an effect of convection-currents when a circular dielectric disc is spun 
between the two phites of a charged horizontal condenser. In this case, however, the 
displacement-system in the field maintains its configuration in space absolutely 
unchanged; and according to the present view no effect of the kind should exist 
unless it be really caused by convection of an actual charge on the rotating dielectric 
plate (unless we find in it a proof of the convection of actual paired ions, of which the 
material dielectric is constituted. See § 125.) 
On Vortex Atoms and their Magnetism. 
61. Suppose, in the condenser-system described above, that a current is started 
round the circuit by a change of capacity of one of the condensers, and that then the two 
condensers are instantaneously taken out and the wire made continuous ; the current, in 
the alosence of resistance in the wire, will now be permanent. A permanent magnetic 
element will thus be represented by a circuital cavity or channel in the elastic aether, 
along the surface of which there is a distribution of vorticity ; it will in short be a 
vortex-ring with a vacuum (or else a portion of the fluid devoid of rotational elasticity) 
for its core. An arrangement like this must be supposed, in accordance with Ampere’s 
theory,§ to be a part of the constitution of a molecule in iron and other magnetic 
* H. A. Rowland and 0. T. Hutchinson, “ On the electro-magnetic effect of Convection-cnrrents.” 
‘ Phil. Mag.,’ June, 1889, p. 445. 
t [The statement in the text is certainly time if we can regard the disc as a perfect conductor; on 
the other hand if it is an insulator, the charge will be carried along with it. It has been suggested 
that it is ojien to question whether the conductivity of a coating of gold-leaf is great enough to 
practically come under the fii’st of these types. But if we are to adhere to the ordinary idea that the 
free oscillations of an electric charge on such a conductor are absolutely unresisted by any superficial 
viscosity, as they are certainly independent of ohmic resistance, we must, it would seem, regard a 
metallic disc as practically equivalent for the present purpose to a perfect conductor. This view would 
also suggest an explanation of the circumstance that some experimenters have not been able to verify 
the existence of the Rowland effect.] 
J Rowland, loc. cit., p. 446; Rontgen, “ Wied. Ann.,” 35, 1888, 
§ Maxwell, ‘ Treatise,’ vol. 2, chap. 22. 
