316 Prof. A. Kundt on the Electromagnetic Rotation 



With another mirror, whose thickness was 0000069 millim., 

 the rotation of the iron was found to be 35,000 times as great 

 as that of the glass. 



Cobalt has a specific rotation nearly the same as that of 

 iron. Nickel, on the other hand, is decidedly weaker. Two 

 nickel mirrors gave a mean rotation of 14,000 times that of 

 the glass. 



I, of course, regard these numbers as approximate only, 

 since the determination of thickness by weighing is very 

 inexact. They give occasion, however, for some interesting 

 conclusions. 



According to the experiments of Righi *, a circularly- 

 polarized ray in which the motion of the aether- molecules 

 takes place in the same direction as that of the molecular 

 currents, traversing a substance which, in a magnetic field, 

 produces rotation in the same direction as the amperian cur- 

 rents, advances more rapidly than a circularly-polarized ray 

 in which the aether-molecules move in the opposite direction. 

 There is therefore a circular double refraction. If n and n f 

 are the indices of refraction for the two opposite circular rays, 

 and <f) the amount of the observed rotation, cl the thickness of 

 the substance traversed by the ray, and X the wave-length; 

 then 



If we take X — 0*00058, for d the thickness of iron film 

 given above, viz. 0*000055, <£ = 1° 48', we have 



(w-w')=0-l. 



This difference in refractive index amounts to more than 

 half of that between the ordinary and extraordinary rays in 

 Iceland spar, and is about ten times as great as the difference 

 of refractive indices in quartz at right angles to the axis. 



The difference of the refractive indices for the two cir- 

 cularly-polarized rays in the direction of the axis in quartz, 

 according to Lang, amounts only to 0*0000718. 



The film of iron used has therefore, in the magnetic field 

 which I have employed, a rotation 1462 times as great as the 

 natural rotation in quartz of the same thickness as the iron 

 film. It will be necessary to return later on to the circular 

 double refraction of our iron films when we come to speak of 

 Fitzgerald's theory. 



* Nuovo Cimento, iii. p. 212 (1878). 



