of the Metals. 175 



metal could be determined. It was observed that both in the 

 case of transparent gold and of transparent silver plates, for 

 which n>l, the displacement of the interference-bands (in the 

 sense of retardation of the rays passing through metal), for light 

 polarized perpendicular to the plane of incidence of the metal, 

 increased with increasing incidence, and under an incidence of 

 85° amounted to 0*3 of the space between the interference-bands. 

 For light polarized parallel to the plane of incidence of the metal, 

 the displacement was 0, or at most 0*1 of the space between the 

 bands, in the sense of a retardation of the ray passing through 

 metal. 



Thence it would follow, regard being had to the above, that, for 

 rays of light passing through metal, the component polarized 

 parallel to the plane of incidence is accelerated relatively to that 

 polarized perpendicular to the plane of incidence, and that the 

 acceleration increases with increasing incidence. This agrees 

 with the above-given experiments upon the elliptical polariza- 

 tion of the light going through metals, the experiments having 

 been made with Babinet's compensators. These experiments 

 leave it in uncertainty whether the component polarized perpen- 

 dicular to the plane of incidence is retarded continuously with 

 increasing incidence, or whether the component polarized parallel 

 to the plane of incidence is continuously accelerated. Now, since 

 the greater displacement of the interference-bands for light 

 polarized perpendicular to the plane of incidence arises from the 

 superior thickness of the layer of metal passed through at large 

 angles of incidence, it is apparent that the component polarized 

 perpendicular to the plane of incidence suffers a smaller altera- 

 tion of phase on account of simple refraction than the component 

 polarized parallel to the plane of incidence, and that this altera- 

 tion of phase for the last component increases with increasing 

 angle of incidence. 



This circumstance admits of the same considerations in the case 

 of transmitted or refracted light as in that of reflected light, 

 with which the author, in a former communication*, sought to 

 estimate the position of the oscillations of the particles of the 

 luminiferous sether in relation to the plane of polarization. For 

 the oscillations which take place perpendicular to the plane of 

 incidence, the position of the paths of the luminiferous particles 

 remains the same towards the refracting surface when the angle 

 of incidence is altered. But in the case of the oscillations which 

 lie in the plane of incidence (and like the former are perpendi- 

 cular to the ray of light) the inclination of the path of the lumi- 



* See Monatsber. d.Berl. Akad. 1862, p. 714. Pogg. Ann. vol. cxviii. 

 p. 445. Phil. Mag. S. 4. vol. xxvi. p. 190. 



