of the Plane of Polarization of Light, 319 



The molecular condition of the surface must undoubtedly 

 play an important part in reflection; and since this may vary 

 very greatly with the concentration of the solution and the 

 strength of the current, it is hardly to be expected that iden- 

 tical values should be obtained for the rotation of different 

 mirrors. In order to obtain the electromagnetic rotation alone, 

 all the values given above must be diminished by about 10 

 per cent. 



In conclusion, it may further be mentioned that I have also 

 coated platinized glass with iron and then electroplated the 

 iron with silver or copper. These iron mirrors covered with 

 silver or copper showed no perceptible rotation upon reflection. 

 If we adjust matters so that the rays fall accurately vertically 

 upon the metallic surfaces, the glass plate may be dispensed 

 with, replacing it by a perforated mirror of non-magnetic 

 metal, and adjusting it so that the reflected light passes 

 through the hole into the eye of the observer. The rotation 

 observed with this arrangement is produced by reflection at 

 the magnetized mirrors between the poles. 



Y. Refutatio n of Fitzgerald's Theory of Rotation produced 

 by Reflection. 



Mr. Fitzgerald * has endeavoured to give an explanation of 

 the rotation of the plane of polarization produced by reflection. 

 Briefly stated, it is as follows: — Iron, upon magnetization, 

 becomes circularly doubly refractive, just as diamagnetic sub- 

 stances in a magnetic field do. This circular double refraction, 

 as has been shown above, is very considerable in the case of 

 iron. Since the intensity of the light reflected by a body 

 depends upon its index of refraction, right-handed circularly 

 polarized light is reflected by magnetic iron with a different 

 intensity from left-handed. Plane-polarized light may be 

 regarded as consisting of equal quantities of right-handed and 

 left-handed circularly polarized light. Starting from this 

 point, Fitzgerald endeavours to show that plane-polarized light 

 falling at any angle upon a magnetic surface is converted by 

 reflection into elliptically polarized, and that in general the 

 major axis of the ellipse must be rotated towards the direction 

 of oscillation of the incident light. For normal incidence 

 upon the pole of a magnet, however, it is easily seen that the 

 angle between the plane of oscillation of the incident light 

 and the major axis of the ellipse is zero. Mr. Fitzgerald 

 therefore himself draws the following conclusion from his 

 theory: — If plane-polarized light falls normally upon the pole 

 of an electromagnet not excited, and if after reflection it be 

 * Proceedings of Royal Society, xxv. p. 447. 



