THE ELECTRIC AND LUMINIFEROUS MEDIUM. 
765 
metals. As a fundamental structure like the present can hardly be supposed to be 
broken up at the temperature at which iron hecomes non-magnetic, to appear again on 
lowering the temperature, we must postulate that a permanent electric current of tl)is 
kind is involved in the constitution of the atom ; that in iron the atoms group 
themselves into aggregates with their atomic currents directed in such a way as not 
absolutely to oppose each other’s action ; while at the temperature of recalescence 
these groups are broken up and replaced by other atomic groups, for each of which 
the actions at a distance of the different atomic currents are mutually destructive. 
In a material devoid of striking magnetic properties, we may imagine the atoms as 
combined into molecules in this latter way. 
62. If we imagine a vortex-ring theory of atoms, in which the velocity of the 
primeval fluid represents magnetic force, and the atoms are ordinary coreless vortices, 
we shall have made a step towards a consistent representation of physical phenomena. 
In such a fluid the vortices will join themselves together into molecules and molecular 
groups; the vortices of each group will however tend to aggregate in the same way 
as elementary magnets, so that instead of neutralizing each other’s magnetic effects, 
they will reinforce one another ; on this view substances ought to he about ecpially 
magnetic at all temperatures, instead of showing as iron does a sudden loss of the 
quality. We must therefore find some other bond for the atoms of a molecule, 
in addition to the hydrodynamic one and at least of the same order of magnitude. 
This is afforded by the attractions of the electric charges of the atoms, which are 
required by the theory of electrolysis. But even now about half of the molecules 
would be made up so that the atoms in them assist each other’s magnetic effects, 
unless we suppose each molecule to contain more than two atoms, arranged in some 
sort of symmetry. There is however no course open but to take all matter to 
be magnetic in the same way, the only difference being in some very special circum¬ 
stance in the aggregation of the molecules of iron compared with other molecules. 
The small magnetic moment of molecules of most substances may in fact be explained 
more fully on the same lines as their small electric moment (§ 64). The vortices will 
be quite permanent as regards both atomic charge and electric intensity, so that the 
explanation of diamagnetic polarity given by Weber, on the basis of currents induced 
in the atomic conducting circuits, cannot now stand.* 
* [Added June 14.—It has been suggested that the atomic electi’ic charge might circulate round the 
ring under the influence of induction. It would appear however that such a circulation could have no 
physical meaning, for it Avould not at all alter the conflguration of strain in the surrounding medium, 
which is the reallv essential thing. 
It is otherwise with the motion of translation of a small charged body: the intrinsic twist of 
the surrounding medium is carried on with it, and the effect of the movement is thus to impose 
an additional twist or rotation round the line of motion (§ 59). Thus if we imagine an endless chain of 
discrete electrified particles, which circulate round and round, each particle of it will carry on 
independently its state of strain and so be subject separately to forcive; and we shall have the 
dynamical phenomena illustrated by a current of pui’ely convective character, involving no electvic dis¬ 
placement in the dielectvic, and no generator.] 
