42 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXI.) 
then as the heat continued to diminish, the iron rapidly, but not instantaneously, ac- 
quired its high magnetic power ; at which time it could not be kept from the magnet, 
but flew to it, bending the suspending wire and trembling as it were with magnetic 
energy as it adhered by one end to the core. 
2346. A small bar of nickel was submitted to an experimental examination in the 
same manner. This metal, as I have shown*, loses its magnetism as respects ordi- 
nary tests at a heat below that of boiling oil, and hence it is very well fitted to show 
whether the magnetic metals can have their power entirely rertioved by heat or not ; 
and also, whether the disappearance of the whole or greater portion of their power is 
sudden or gradual. The smallness of the mass to be experimented on assisted much 
in the determination of the latter point. Upon being heated the nickel soon became 
indifferent to ordinary magnets ; but however high the temperature, still it pointed 
to and was attracted by the electro-magnet. The power was very feeble, but certain. 
It was scarcely enough to sustain the weight of the nickel by the magnetic action 
alone ; but was abundantly evident when the metal was supported as described 
(2344.). 
2347- On carefully lowering the temperature of the nickel, it was again found that 
the transition from one degree of magnetic force to the other was progressive, and 
not instantaneous. With iron it is difficult to preserve all the parts, either in heating 
or cooling, so nearly at the same temperature as to be sure that it is not the union of 
hotter and colder portions which gives the appearance of an intermediate degree of 
magnetism ; but with nickel that is not so difficult, for the progression is more gra- 
dual, so that when in cooling the power began to increase, the cooling might be con- 
tinued some time before the full degree of power came on ; at any time in that pe- 
riod the temperature might be slightly raised, and though the power would then 
diminish a little, it could yet be retained at a degree stronger than the weakest. In 
fact it was easy to keep the nickel at many of the intermediate degrees of power, 
and thus to remove all doubt of the progressive assumption of the full degree of 
force. 
2348. I have expressed an opinion, founded on the different temperatures at which 
the magnetic metals appeared to lose their peculiar power-f'-, that all the metals would 
probably have the same character of magnetism if their temperature could be lowered 
sufficiently. The facts just described appear to me entirely against such an opinion. 
The metals which are magnetic retain a portion of their power after the great change 
has been effected, or in what might be called their diamagnetic state ; but the other 
metals, such as bismuth, tin, &c., present no trace of this power, and therefore are not 
in the condition of the heated iron, nickel, or cobalt ; for in fact whilst these point 
axially and are attracted, the others point equatorially and are repelled. I therefore 
hope to be allowed to withdraw the view I then put forth. 
2349. I next proceeded to examine the peroxides of iron, and in accordance with 
* Philosophical Magazine, 1836, vol. viii. p. 179. f Ibid. 177 ; ibid. 1839, vol. xiv. p. 161. 
