708 Professor Andrew Gray [April 29} 



found correct, the angle of turning will be proportional to the pro- 

 duct of the intensity of magnetisation and to the length of the path ; 

 and the angle observed divided by this product will give another 

 constant, which has been called Kundt's constant. 



The rotation of the plane of j)olarisation in strongly magnetised 

 substances was investigated by Kundt, the very eminent head of the 

 Physical Laboratory of the University of Berlin, who died only a 

 year or two ago. Kundt is remembered for many beautiful methods 

 which he introduced into quantitative physical work ; but no work 

 he did was more remarkable than that which he performed in magneto- 

 optic rotation when he succeeded in passing a beam of plane polar- 

 ised light through plates of iron, nickel and cobalt. Such substances, 

 though apparently opaque to light, are not really so when obtained in 

 plates of sufficient thinness. In sufficiently thin films all metals, so 

 far as I know, are transparent, not merely to Eontgen rays, but to 

 ordinary light. Kundt conceived the idea of forming such films of 

 the strongly magnetic metals, so as to investigate their properties as 

 regards magneto -optic rotation. He succeeded in obtaining them by 

 electroplating platinised glass with such thin strata of these metals 

 that light passed through them in sufficient quantity for observation. 

 The rotation produced by the glass and the exceedingly thin film of 

 platinum was determined once for all and allowed for. Kundt 

 obtained the remarkable result that the magnetic rotatory power in 

 iron is so great, that light transmitted through a thickness of one 

 centimetre of iron magnetised to saturation is turned through an 

 angle of over 200,000°, that is, that light passing through a thickness 

 of an inch of iron magnetised to saturation would have its plane of 

 polarisation turned completely round more than a thousand times ; in 

 other words, one complete turn would be given by a film less than yoVo 

 of an inch in thickness. A scarcely smaller result has been found by 

 Du Bois for cobalt, and a maximum rotation of rather less than half 

 as much by the same experimenter for nickel. 



The direction of turning in all the cases which have so far been 

 specified — that is, Faraday's glass, bisulphide of carbon, iron, nickel 

 and cobalt — is the same as that in which a current of electricity would 

 have to flow round the spires of a coil of wire surrounding the specimen 

 so as to produce the magnetic field. This we call the 'positive direc- 

 tion. There are, however, many substances in which the turning 

 produced by the magnetic field is in the contrary or negative direction ; 

 for example, ferrous and ferric salts of iron, chromate and bichromate 

 of potassium, and in fact most compound substances which are feebly 

 magnetic. 



Faraday established by his experiments the fact that substances 

 fall into two distinct classes as tested by their behaviour under the 

 influence of magnetic force. For example, an elongated specimen of 

 iron, nickel or cobalt, if freely suspended horizontally between the poles 

 of our electro-magnet, would set itself with its length along the lines 

 of force. On the other hand, a similar specimen of heavy glass, or a 



