272 Mr. J. Parker on the Theory of Magnetism and 



the fiction of the magnetized soft bodies on one another ; 

 and if a number of small bodies be placed near the earth, 

 we may neglect their gravitational attraction on one another 

 in comparison with that of the whole earth. Lastly, if a 

 soft body be immersed in a gas or liquid, and then placed 

 near the pole of a magnet, it will appear to be attracted or 

 repelled according as it is more or less magnetic than the gas 

 or liquid by which it is surrounded ; and if any body be 

 immersed in a gas or liquid, it will appear to be attracted or 

 repelled by the earth according as it is heavier or lighter than 

 the gas or liquid in which it is placed. 



The theory we have given is beautifully illustrated and 

 confirmed by the following experiments of Faraday's, described 

 in Tyndall's ' Diamagnetism.' Theory and experiment fit 

 together so exquisitely that we cannot but wonder the true 

 theory should not have been seen from the first. 



" If a weak solution of protosulphate of iron, m, be put into a 

 selected thin glass tube about an inch long, and one third or 

 one fourth of an inch in diameter, and sealed up hermetically, 

 and be then suspended horizontally between the magnetic 

 poles in the air, it will point axially, and behave in other 

 respects like iron ; if instead of air between the poles, a solu- 

 tion of the same kind as m, but a little stronger, n, be sub- 

 stituted, the solution in the tube will point equatorially, or as 

 bismuth. A like solution somewhat weaker than m, to be 

 called /, enclosed in a similar tube, will behave like bismuth 

 in air but like iron in water." 



It now remains to describe how it is generally attempted 

 to gloss over the imaginary difficulties of diamagnetism. 



It is generally admitted that the apparent magnetic or dia- 

 magnetic properties of a soft body B immersed in air or any 

 other gas or a liquid, are merely differential — that is, depend 

 on the algebraic excess of the coefficient k b of the body B 

 over the coefficient k a of the substance A in which it is 

 immersed. This result is thought to be " proved " in the 

 following way : — Since every soft body is magnetically equiva- 

 lent to a layer of magnetism on its surface, it follows that 

 there are two layers on the common surface of A and B, one 

 belonging to A, the other to B. It is then supposed that the 

 layer on this surface which properly belongs to A, does not 

 really belong to A at all, but to B. The body A being mag- 

 netically equivalent to a layer on its surface, it is assumed 

 that, as the layer on the common surface of A and B is 

 supposed transferred to B, we may treat A as unmagnetized. 

 With this assumption the pressure of A would be uniform, 



