21 j 
the forces of electricity and magnetism, appears to be due to ethereal 
streams generated under the condition of variation of interior 
atomic density. Such variation may be supposed to be produced 
in a magnetic bar in the following manner It is evident that in 
any case of uniform atomic density, the molecular attractions 
acting on a given atom will be equal in all directions, and will 
therefore just neutralize each other ; and that similarly the resultant 
ot the atomic repulsions acting on a given atom will be zero. But 
it is quite conceivable that the equilibrium of an atom might result 
irom the counteraction of one set of forces by the other ; only in 
this case there must be a gradation of atomic density , molecular 
attraction always acting in the direction from rarer to denser parts 
and atomic repulsion in the opposite direction. The capability of 
satisfying magnetic physical conditions, which exists in an eminent 
degree in iron, can only be attributed to the particular constitution 
of the substance. The effect of the known processes of magneti- 
zation seems to be to induce a gradation of atomic density • in 
soft iron, temporarily, and in hardened steel, permanently. The 
ethereal steady streams, generated under the conditions in which 
ns gradation of atomic density is maintained by the interior 
molecular attractions and atomic repulsions, produce by their action 
on temporary or permanent magnets, and on galvanic conductors, 
lose movements which experimenters ascribe to magnetic attrac- 
tions and, repulsions. Such phenomena are consequently accounted 
for by the hydrodynamical theory of magnetism. I take occasion 
to state here that I have succeeded in demonstrating, according to 
the principles of that theory, two well-known experimental results 
obtained by Gauss ; namely, that the action of a large bar-magnet 
on a small one, when the axis of one points perpendicularly to that 
of the other through its middle point, varies inversely as the cube 
of the distance between the middle points ; and that, in case the 
axis of. the small one points to the middle point of the lami one 
the action is just half what it is in the contrary case. The theo- 
retical explanations of these facts are specially noteworthy, because 
they depend entirely on the supposition of the spherical form of 
hypothecs ^ regarded aS verifications of that fundamental 
(13.) Like all other magnetism, Terrestrial Magnetism is to be 
referred to the agency of ethereal streams ; but in this instance the 
streams are not due to gradation of atomic density, but are simply 
impressed on the ether by the constituent atoms of the earth in 
. ei \ minal levolution about its axis. The generating cause being 
s ea y, le streams also, are steady, and have always nearly the 
same re ation to the position of the earth’s centre, the system of 
s reams being, as it were, borne along by the earth’s motion in its 
VOL. xr. Q 
