1906.] on Recent Progress in Magneto-Optics. 295 



electrons directly, but adds suitable new terms to the equations 

 of motion in an absorbing medium. His method establishes a 

 connection between the rotation of the plane of polaristition and the 

 resolution of the spectral lines, a connection almost simultaneously 

 pointed out by Fitzgerald. This also led to an interesting result, 

 till then missed by the electronic theory, namely : — 



Rotation of the Plane of Polarisation close to an 

 Absorption Band* 



Faraday's rotation of the plane of polarisation is extremely small 

 in all gases, also in sodium vapour. Only within a very narrow range 

 close to the sodium Unes the rotation is positive and very great, a fact 

 discovered by Macaluso and Corbino. In a recent extremely interest- 

 ing paper. Professor Wood has given measurements of observed 

 rotation of four complete revolutions.! This, however, was in rather 

 dense vapour, at least dense in comparison with the vapour used in 

 the experiments now to be described, in which vapour containing 

 about one-milhonth gram of sodium per cm.^ was used. 



The magnitude of the rotation close to the sodium lines is 

 illustrated by measurements made by Dr. Hallo in the Amsterdam 

 laboratory. It is clear that on both sides of an absorption hne the 

 rotation is in the same direction. (Fig. 11.) 



We may attenuate the vapour still further so that the doublet in 

 the direction of the lines of force becomes visible. What is the 

 rotation then between the components of the doublet ? 



It is easily deduced from Professor Voigt's theory that in very 

 diluted vapours the rotation must occur, in a sense, opposite to that 

 outside the components, and therefore negatively, and also that it 

 must be very great. In the case of sodium vapour, I had the pleasure 

 to confirm this theoretical result, and to observe rotations of —400°. 



In these experiments interference fringes in the spectrum were 

 used, estabhshed by means of a system of Fresnel quartz wedges (a 

 method used by Voigt, Corbino, and others in similar cases). I will 

 project these fringes on the screen. 



If a plate of quartz, which rotates the plane of polarisation, is 

 held in the ray, you will notice a displacement of the fringes. A 

 plate of glass has no influence of course. I have here a quartz plate 

 which rotates the plane of polarisation through 90°, and you will notice 

 a displacement of half the distance between two fringes. A displace- 

 ment of the entire distance between two fringes corresponds to a 

 rotation of half a revolution. 



* Zeeman. Proc. Ac. Sciences, Amslerdam, May, 1902. Hallo. Thesis, 

 Amsterdam, 1902, Archiv, Neerl., ser 2, T. 10, p. 148, 1905. 

 f Wood. Phil. Mag., October 1905, 



