254 Mr B. T. Glazebrook, On the theory [April 20, 



propagation is v/(l - ^af/n*), so that in a rotating medium there 

 would be dispersion but the blue rays would move more slowly 

 than the red. 



(2) On the theory of some experiments of Frohlich on the position 

 of the plane of polarization of light diffracted at reflexion from 

 a grating. By R. T. Glazebrook, M.A., F.R.S. 



In Wiedemann's Annalen Vol. I., Frohlich gives an account of 

 some experiments on the polarization of light diffracted by re- 

 flection from two glass gratings, one of which was ruled with lines 

 at a distance of -0506 mm. apart, the other with lines at a distance 

 of -00617 mm. 



The incident light was polarised in a plane at 45° to the plane 

 of diffraction and the plane of polarization observed for a series of 

 angles of diffraction and for angles of incidence of 85°, 55° and 25°. 



The position of the plane of polarization of the light reflected 

 directly at angles varying by 5° between 85° and 25° was also 

 observed. 



It was found that for the diffracted light the direction of 

 vibration depends on the angle of incidence, the angle of dif- 

 fraction and the nature of the reflecting face, but is completely in- 

 dependent of the distance between the lines of the grating and of 

 the overlapping of spectra of different orders at the same point in 

 the field of the observing telescope ; while the position of the 

 plane of polarization of the light regularly reflected agreed fairly 

 with that given by Fresnel's theory, except near the polarising 

 angle where Jamin's effect, which however in glass of refractive 

 index about 1*5 is not very marked, came in. 



The only other experiments which I know of on polarization 

 by diffraction at reflexion are some alluded to in Prof. Stokes' 

 paper in the Transactions of the Camb. Phil. Soc. Yol. IX. on the 

 dynamical theory of diffraction, but not published on account of 

 the difficulties arising from the overlapping of the images formed 

 by light reflected from the second surface of the grating. These 

 were avoided to some extent by Frohlich by using a wedge-shaped 

 piece of glass for his grating. 



Frohlich states in his paper that his results confirm the suppo- 

 sition that the direction of vibration is normal to the plane of 

 polarization, but does not give any theory. 



This has been partly supplied by Rethy (Wiedemann, Vol. XL, 

 On the polarization of diffracted light), who has shewn that 

 Frohlich's experiments may be represented by the formula 



tan (p = tan <£ cos S -f sec (/> sin & cot p (1), 



cp and (f> being the azimuths of the planes of polarization of the 



