46 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXI.) 
pounds ; and, judging from the character of the great class of diamagnetic bodies 
(2275.), that no magnetic compounds would be obtained of a metal not magnetic of 
itself. Accordingly I proceeded to apply this kind of test to the combinations of 
many of the metals, and obtained the following results : — 
2371. Titanium. — Wollaston has described the magnetic effects of crystals of 
titanium, expressing at the same time a belief that they are due to iron*. I took a 
specimen of the oxide of titanium, which I believe to be perfectly free from iron, and 
inclosing it in a tube (2279.), subjected it to the action of the electro-magnet (2246. 
2247.). It proved to be freely magnetic. Another specimen obtained from Mr. 
Johnson, and believed by him to be perfectly free from iron, was also magnetic. 
Hence I conclude that titanium is truly a magnetic metal. 
2372. Manganese. — Berthier, as far as I am aware, first announced that this 
metal was magnetic at very low temperatures'^. On submitting specimens of the 
various oxides, which were considered as pure, to the magnetic force, they were all 
found to be magnetic, especially the protoxide. So were the following compounds 
of manganese in the pure, dry, or crystallized state : — chloride, sulphate, ammonio- 
sulphate, phosphate, carbonate, borate ; and also the chloride, nitrate, sulphate, and 
ammonio-sulphate when in solution. A specimen of the ammonio-sulphate was ren- 
dered alkaline by the addition of a little carbonate of ammonia boiled and then care- 
fully crystallized thrice ; after that the crystals and solution of the purified salt were 
perfectly and well magnetic. I have no doubt, therefore, that manganese is a mag- 
netic metal, as Berthier said. If any opinion may be drawn concerning the magnetic 
force of the metal from the degree of magnetism of the compounds, I should expect 
that manganese possesses considerable power of this kind when at a sufficiently low 
temperature^. 
2373. Cerium. — I am not aware that cerium has as yet been classed with the 
magnetic metals. Having made experiments with the hydrated protoxide, the car- 
bonate, and the chloride of this metal, and also with the double sulphate of the oxide 
and potassa prepared with great care, I found them all magnetic ; and those that 
are soluble are magnetic in the state of solution. Hence, as the compounds are un- 
doubtedly magnetic, there is every reason to believe that cerium also is a magnetic 
metal (2370.). 
2374. Chromium. — The magnetic phenomena of chromium compounds are very 
interesting. Portions of the chromate and the bichromate of potassa were purified 
by three careful crystallizations each ; part of the bichromate was heated in a pla- 
tinum crucible, until the second equivalent of chromic acid was converted into the 
crystallized oxide, and this being washed out and dried was found to be well magnetic. 
So were all the other specimens of oxide of chromium which were examined. 
A specimen of Warrington’s chromic acid was found to be very feebly magnetic. 
* Philosophical Transactions, 1823, p. 400. 
t Traite des fissais par la Voie Seche, tomei. p. 532. Philosophical Magazine, 1845, vol. xxvii. p. 2. 
{ Philosophical Magazine, 1845, vol. xxvii. p. 2. 
