119 
or equatorially could not be observed. The revulsive 
motion not being exhibited by the disc of metal when it 
had assumed a position either strictly axial or strictly 
equatorial, in order that the phenomena should be fully 
developed, I found it necessary to prevent the disc from 
assuming such positions. With metals tending to the axial 
position this was easily effected, by closing the polar pieces 
so that the disc could not pass to the axial position, being 
stopped by the polar pieces themselves ; but in experiment- 
ing with some metals in which the revulsive action is only 
manifested in a slight degree, — for example, with palladium, — 
some care is needful in ascertaining whether revulsion takes 
place or not, for if the disc of metal be allowed to touch 
the polar pieces, the vibratory action of the magnet appears 
not unlikely to communicate itself to the disc, thus pro- 
ducing the appearance of revulsion ; therefore, in such 
experiments I found it convenient to use a steady support, 
by which the disc under examination is prevented going 
to the axial position. Such a support is also very useful 
in preventing a metal, affected dia-magnetically, from taking 
the exact equatorial position; but when a support is used, 
the motions require to be observed very carefully, for the 
disc appears to be moved from the position in which it 
would hang when not affected by the magnetic or dia- 
magnetic force, and to move, as though revulsively, by 
the mere action of gravitation. But on carefully noticing 
the motion, it may be observed whether it be more rapid 
than that due to gravitation alone. 
The changes from the magnetic to the dia-magnetic 
state are, perhaps, most decidedly exhibited by pure zinc 
or pure lead. When a very considerable magnetic power 
is applied to these metals, either by increasing the power 
of the battery, or by approximating the polar pieces, 
instead of attraction, as magnetic bodies, they show repul- 
