256 Dr. Faraday on the Magnetic Affection of Light. 



upon matter ordinarily magnetic, as iron is. All depends 

 upon the shape and size of the poles, and upon the state of 

 unity or disintegration, more or less complete, of the magnetic 

 matter ; for matter which is much weaker in magnetic power 

 than peroxide of iron, as the solution of nickel, or dilute solu- 

 tion of iron or the metal platina, does not act as the peroxide, 

 but as metallic iron. Still, in every case, the phenomena are 

 phenomena of attraction ; for not only is the centre of gravity 

 of the whole mass attracted, but the parts of the cylinder of 

 peroxide of iron, as well as of the other magnetic substances, 

 are in those positions which the resultants of attractive force 

 would necessarily give them. This is precisely the reverse of 

 what takes place with diamagnetic bodies, for there the centre 

 of gravity of the whole mass is repelled', and whatever form 

 the mass may have, its parts take those positions which are 

 most consistent with the direction and degree of the rejmlsive 

 force. 



All ambiguity and doubt may be removed in such cases as the 

 above by the use of a single pole, either conical, wedge-shaped 

 or round. It is true, that if a wire of iron, two or three inches 

 long, be placed with its middle part near the apex of a conical 

 or wedge-shaped pole, it will stand at right angles across the 

 axis of the cone or wedge, vibrating like a balance-beam ; and 

 also that if a cylinder of bismuth, phosphorus or heavy glass 

 be placed in the same position, it will take a similar position. 

 But no magnetician who looks at the effect can for a moment 

 confound the phenomena ; for he will see at once that the 

 iron as a mass is attracted, and the diamagnetic is, as a mass, 

 repelled : and then, if for this observation of the latter phce- 

 nomena he will take small spheres of iron, peroxide of iron, 

 or solution of muriate of iron on the one hand, and of bismuth 

 or phosphorus, copper or wood on the other, he will have 

 the phenomena in the simplest state, and the fundamental 

 fact will be before him ; for the magnetics will be attracted, 

 the diamagnetics will be repelled. 



I cannot find any case of transverse position which does not 

 enter into one or other of the two kinds referred to above ; that 

 is to say, which is not either a magnetic or a diamagnetic result. 

 Even as regards the effect of ordinary magnetism in peroxide 

 of iron and such like bodies, I see no reason to accept the 

 statement of M. Becquerel, that the distribution of magnetism 

 tends to take place in a transverse direction. It appears to 

 me that the destruction of the continuity of the mass in these 

 cases, combined with the degree of magnetic force in the sub- 

 stance, prevents the transmission of the force by induction or 

 conduction from particle to particle, to the same extent as 



