MAGNETIC CONDUCTION— PARAMAGNETIC AND DIAMAGNETIC CONDUCTORS. 33 
2811. If the iron be a spheroid, then its greatest diameter points axially, whether 
it be in the position of unstable equilibrium, nearer to or in contact with the iron 
walls of the field. As the circumstances are now more favourable for the concentra- 
tion of force in the axial line passing through the body than before, so this result can 
be produced by much weaker paramagnetics than iron, and I have no doubt could 
easily be produced by a vessel of oxygen or nitric oxide gas (2/82. 2792.). It now 
becomes indeed a form, though not the best, of that experiment by which the mag- 
netic condition of bodies is considered as most sensitively tested. 
2812. The relative deficiency of power in diamagnetic bodies renders any attempt 
to obtain the converse phenomena to those of iron somewhat difficult ; in order there- 
fore to exalt the conditions, I used a saturated solution of protosulphate of iron in 
the magnetic field ; by this means I strengthened the lines of power passing across it, 
without disturbing its equality in the parts employed, or introducing any error into 
the principle of the experiment, and then used bismuth as the diamagnetic body. A 
cylinder of this substance, suspended vertically, tended well towards the middle 
distance, finding its place of stable equilibrium in the spot where the paramagnetic 
body had unstable equilibrium. When the cylinder was suspended horizontally, then 
the direction it took was equatorial ; and this effect also was very clear and distinct. 
2813. These relative and reverse positions of paramagnetic and diamagnetic bodies, 
in a field of equal magnetic force, accord well with their known relations to each 
other, and with the kind of action already laid down in principle (2807.) as that 
which they exert on the magnetic power to which they are subjected. One may re- 
tain them in the mind by conceiving that if a liquid sphere of a paramagnetic con- 
ductor were in the place of action, and then the magnetic force developed, it would 
change in form and be prolonged axially, becoming an oblong spheroid ; whereas if 
such a sphere of diamagnetic matter were placed there, it would be extended in the 
equatorial direction and become an oblate spheroid. 
2814. The mutual action of two portions of paramagnetic matter, when they are 
both in such a field of equal magnetic force, may be anticipated from the principles 
(2807. 2830.), or from the corresponding facts, which are generally known. Two 
spheres of iron, if retained in the same equatorial plane, repel each other strongly ; 
but as they are allowed to depart out of that plane, they first lose their mutual re- 
pulsive force and then attract each other, and that they do most powerfully when in 
an axial direction. 
2815. With diamagnetic bodies the mutual action is more difficult to determine, 
because of the comparative lowness of their condition. I therefore resorted to the 
expedient, before described, of using a saturated solution of protosulphate of iron as 
the medium occupying the field of equal magnetic force, and employing two cylinders 
of phosphorus, about an inch long and half an inch in diameter, as the diamagnetic 
bodies. One of these was suspended at the end of a lever, which was itself suspended 
by cocoon-silk, so as to have extremely free motion, and the adjustments were such, 
that when the phosphorus cylinder was in the middle of the magnetic field, it was free 
MDCCCLI. p 
