1855.] on Magnetic Philosophy. 9 



ing its power on other magnets, or when left to itself {Experimental 

 Researches, 3119, 3121, 3215, &c.) ; and like means show that 

 the external disposition of the force is so affected : so that the 

 force in right lines through the magnet does not change under the 

 circumstances, whilst the force in external (and necessarily) curved 

 lines does. 



The polarity of bismuth or phosphorus in the magnetic field is 

 one point amongst many others essentially dependent upon, and 

 highly illustrative of the nature of, the magnetic force. The as- 

 sumption that they have a polarity the reverse of that of para- 

 magnetic bodies involves the consequence, that northness does not 

 always repel northness or attract southness ; or else leads to the 

 assumption that there are two northnesses and two southnesses, 

 and that these sometimes associate in pairs one way, and at other 

 times in the contrary way. But leaving the assumptions and 

 reverting to experiment, it was hoped that a forcible imitation of 

 the imagined state of bismuth in the magnetic field, might illustrate 

 its real state, and, for this purpose, recourse was had to the indica- 

 tions given by a moving conductor. Four spheres of copper, iron, 

 bismuth, and hard steel have been prepared, and rotated upon an 

 axis coincident with the magnetic axis of a powerful horse-shoe 

 magnet ; each sphere has a ring of copper fixed on it as an equator, 

 and the ends of a galvanometer wire were brought into contact 

 with the axis and the equator of the revolving globe. Under these 

 circumstances, the electric current produced in the moving globe 

 was conveyed to the galvanometer, and became the indicator of the 

 magnetic polarity of the spheres ; the direction of rotation, and the 

 poles of the magnet, being in all cases the same. When the copper 

 sphere, as a standard, was revolved, deflection at the galvanometer 

 occurred in a certain direction. When the iron sphere replaced 

 the copper and was revolved, the deflection at the galvanometer 

 was the same. When the bismuth sphere was employed, the de- 

 flection was still the same : — and it still remained the same when 

 the steel sphere was rotated in the magnetic field. Hence, by this 

 effect, which I believe to be a truthful and unvarying indication of 

 polarity, the state of all the spheres was the same, and therefore 

 the polarity of the magnetic force in the iron, copper, and bismuth, 

 in every case alike. {Exp. Res. 3164, &c.) The steel sphere 

 was then magnetized in the direction of its axis, and was Tound to 

 be so hard as to retain its own magnetic state when in a reverse 

 direction between the poles of the dominant magnet, for upon its 

 removal its magnetism remained unchanged. Experiments were 

 then made in a selected position, where the dominant magnetic 

 force was not too strong — (a magnet able to lift 430 lbs. was used) 

 — and it was found that when the steel magnet was placed in ac- 

 cordance, i.e. with its north pole opposite the south pole of the 

 dominant magnet, the deflection was in the same direction as with 

 the bismuth sphere ; but when it was changed so as to be in 



