MAGNETIC CONDUCTION — CONDUCTION POLARITY. 
37 
methods in paramagnetic bodies so strongly influential as saturated solution of iron, 
nickel or cobalt, it can hardly be expected to manifest itself by analogous actions 
in the much weaker cases of diamagnetic substances. 
2828. When a spherical paramagnetic conductor is placed midway in a field of 
equal magnetic force, it occupies a place of unstable equilibrium, from which, if it be 
displaced ever so little, it will continue to move until it has gained the iron boundary 
walls of the field (2465. 2810.) ; this is a consequence of its particular polar condition. 
If the sphere were free to change its form, it would elongate in the direction of the 
magnetic axis ; or if it were a solid of an elongated form, it would point axially, both 
consequences of its polar condition (2811.). 
2829. So also in the case of diamagnetic bodies, their peculiar condition of polarity 
is shown by corresponding facts, namely, by a spherical portion having its place of 
stable equilibrium in the middle of the magnetic field (2812.), by a fluid portion tend- 
ing to expand equatorially and become an oblate spheroid (2813.), and by the equa- 
torial pointing of an elongated portion (2812.). If pointed magnetic poles are used, 
then the effects are very much stronger, but are exactly the same in kind, and 
dependent upon the same causes and polar conditions. 
2830. There are another set of effects produced, which are either the results of the 
axial polarity just referred to, or else may be considered as consequences of the con- 
dition of the equatorial parts of the conductors (2824.). Two balls of iron, in a field 
of equal force, if retained in a plane at right angles to the line of force, i. e. with 
their equatorial parts in juxtaposition, separate from each other with considerable 
power (2814.), and the probability is that two infinitely weaker bodies of the para- 
magnetic class would separate in like manner. Two portions of phosphorus, being 
a diamagnetic substance, have been found also to separate under the same circum- 
stances (2815.). 
2831. The motions here are of the same kind, whereas they might have been 
expected to be the reverse (2816.) of each other; still they are perfectly consistent. 
The diamagnetics ought to separate, for the field is stronger in lines of magnetic 
force between them than on the outsides, as may easily be seen by considering the 
two spheres D D in fig. 6 ; and there- 
fore this motion is consistent, and is 
in accordance generally with the open- 
ing or set equatorially, either of sepa- 
rate portions or of a continuous mass 
of such substances (2829.), in their ten- 
dency to go from stronger to weaker places of action. On the other hand, the two 
balls of iron, P P, have weaker lines of force between them than on the outside ; and 
as their tendency is to pass from weaker to stronger places of action, they also sepa- 
Fig. 6. 
