1898. 



on Magneto-Optic Botation. 



709 



tube filled with bisulphide of carbon, would, if similarly suspended, set 

 itself across the lines of force. The former substances were there- 

 fore called by Faraday paramagnetic, the latter diamaguetic. 



It might be supposed that diamagnetics would show a turning effect 

 opposed to that found in paramagnetics, but this is not the case. 

 As we have seen, bisulphide of carbon and heavy glass, which are 

 diamagnetics, show a turning in the same direction as that jiroduced 

 in iron — as indeed do most solid, fluid and gaseous diamagnetics. 

 Feebly paramagnetic compound substances, on the other hand, produce 

 negative rotation. 



A theory of diamagnetism has been put forward in which the 

 phenomena are explained by supposing that all substances are para- 

 magnetic in reality, but that so-called diamagnetic bodies are less so 

 than the air in which they are immersed when experimented on. Thus 

 the diamagnetic quality is one of the substances relatively to air, in 

 the same kind of way as the apparent levity of a balloon is due to the 

 fact that its total weight has a positive value, but is less than that 

 of the air dispLT,ced by the balloon and appendages. Lord Kelvin's 

 dynamical explanation of magneto-optic ^ 



rotation does not bear out this view of i 



the matter. 



Before passing to the dynamical ex- 

 planation, however, I must very shortly 

 call attention to some remarkable dis- 

 coveries in this subject made by Dr. 

 John Kerr, of Glasgow. I have here 

 an electro-magnet arranged as in the 

 diagram before you (Fig. 6). The light 

 from the lamp is first plane polarised by 

 the Nicol P, then it is thrown on the 

 piece of silvered glass G, and part of it 

 is thereby reflected through this per- 

 forated pole-piece so as to fall normally 

 on the polished point of the other pole- 

 piece. Keflection thus' takes place at 

 perpendicular incidence, and the re- 

 flected light is received by this second 

 Nicol. When the magnet is unexcited 

 the second Nicol is arranged so as to 

 quench the reflected light. The mag- 

 net is then excited, and it is found that 

 the light is faintly restored, showing 

 that an effect on the polarisation of the 

 light has been produced by tlie magnetisation, 

 here that the incident and reflected light is m 

 magnetisation. We shall not 





Fig. 6. 



It is to be noticed 



the direction of 



pause to make this experiment. It 



was arranged this morning and successfully carried out ; but the 



effect is slight, and might not be noticeable without precautions, 



Vol. XV. (No. 92.) 3 a 



