344 Dr. Tyndall on the Polarity of Bismuth. 



the part played by the sphere itself. We are compelled to regard 

 the sphere as an induced magnet, whose power is directly pro- 

 portional to the inducing one. Were the magnetism of the 

 sphere a constant quantity, a magnet of double power could only 

 produce a double attraction ; but the fact of the magnetism of 

 the sphere varying directly as the source of induction leads us 

 inevitably to the law of squares; and conversely, the law of 

 squares leads us to the conclusion that the sphere has been 

 induced. 



27. These sound like truisms; but if they be granted, there is no 

 escape from the conclusion that diamagnetic bodies are induced ; 

 for it has been proved by M. E. Becquerel and myself*, that the 

 repulsion of diamagnetic bodies follows precisely the same law as 

 the attraction of magnetic bodies ; the law of squares being true 

 for both. Now were the repulsion of bismuth the result of a 

 force applied to the mass alone, without induction, then, with a 

 constant mass, the repulsion must be necessarily proportional to 

 the strength of the magnet. But it is proportional to the square 

 of the strength, and hence must be the product of induction. 



28. In order to present magnetic ph^enomena intelligibly to 

 the mind, a material imagery has been resorted to by philoso- 

 phers. Thus we have the ^magnetic fluids ' of Poisson and the 

 ' lines of force ' of Mr. Faraday. For the former of these Pro- 

 fessor W. Thomson has recently substituted an ' imaginary mag- 

 netic matter.' The distribution of this ' matter ' in a mass of 

 soft iron, when operated on by a magnet, has attraction for its 

 result. We have the same necessity for an image in the case of 

 bismuth. If we imagine the two magnetic matters which are 

 distributed by induction on a piece of iron to change places, we 

 have a distribution which will cause the phsenomena of bismuth. 

 Hence it is unnecessaiy to assume the existence of any new 

 matter in the case of diamagnetic bodies, the deportment being 

 accounted for by reference to a peculiarity of distribution. Fur- 

 ther, the experiments of lleich, which prove that the matter 

 evoked by one'pole will not be repelled by an unlike pole, compel 

 us to assume the existence of two kinds of matter, and this, if I 

 understand the term aright, is polarity. 



29. During this inquiry I changed my mind too often to be 

 over- confident now in the conclusion at which I have arrived. Part 

 of the time I was a hearty subscriber to the opinion of Mr. Fara- 

 day, that there existed no proof of diamagnetic polarity ; and if 

 I now differ from that great man, it is with the honest wish to be 

 set right, if through any unconscious bias of my own I have 

 been led either into errors of reasoning or misstatements of facts. 



Queenwood College, 

 Oct. 1851. 



* Phil. Mag., Sept. 1851. 



