M A GNE CRYSTAL Lie CONDUCTION. 
41 
(2457.) and hung up on one side of the double cone core (2738.), whilst a cylinder of 
flint-glass was opposed to it on the other. The flint-glass was to be a standard of 
reference, and therefore neither its place on the balance nor condition was altered 
during the experiment. The bismuth group was placed with its magnecrystallic axis 
horizontal, and so that it could be turned in a horizontal plane, that the axis might 
be at one time parallel to the magnetic axis (or lines of force), and at other times per- 
pendicular to it, but without any alteration of the distance of its centre of gravity 
from the opposed glass cylinder. Hence, having either one position or the other, it 
could still be compared with the cylinder. 
2841. The magnecrystallic axis was first made parallel to the core or magnetic axis, 
the magnetic power developed, and when the diamagnetic bodies had taken their 
position of rest or stable equilibrium, the place of the balance lever was observed 
and recorded by means of a ray of light reflected from a mirror attached to it. Then 
the bismuth was turned through 90°, or until its magnecrystallic axis was perpendi- 
cular to the axis of the double cone core ; and now, when the magnet was excited, 
the place of the bismuth was found to be further out from the core than before. On 
being turned through 90° more, so as to be in a position diametral to the first (2461.), 
its place was again a little nearer to the magnet ; and when in the fourth position, 
whieh is diametral to the second, then it was further out. Thus the crystallized bis- 
muth proved to be diamagnetic in different degrees, according with certain direc- 
tions of its magnecrystallic axis, being more diamagnetic when this axis was perpen- 
dicular or transverse to the lines of magnetic force, than when it was parallel to 
them ; and thus the expectation founded upon theoretical considerations (2839.) was 
confirmed. 
2842. I tried to obtain similar results with a cube of calcareous spar (2597.) ; foi* 
it is evident that if its optic axis, being in a horizontal plane, is first placed parallel 
to the magnetic axis and then perpendicular to it, the body ought to be more dia- 
magnetic in the first position than in the second, inasmuch as the latter is the posi- 
tion whieh it takes up under the influence of its magnecrystallic or magneoptic con- 
dition. I could not however obtain any distinct results, partly beeause its power is 
in all respects very inferior to the bismuth, partly because of the present imperfec- 
tion of my torsion balance, and partly because of the size and shape of the calcareous 
spar. A sphere or a cylinder, having the optic axis perpendicular to the axis of the 
cylinder, would be more correct as forms of the substances to be tried. 
2843. In concluding this part of the subject relating to the magnetic conducting 
power, I will now refer to some of the cases which I think experimentally establish 
its existence in the two subdivisions of magnetic bodies (2805.). The place and posi- 
tion of iron in a field of equal force (2810. 2811.) is no doubt a result of the extraor- 
dinary power which this body has of transmitting the magnetic force across the space 
whieh it oeeupies, whether the particles of the iron be considered as polar or not 
MDCCCLI. c 
