174 M. G. Quincke on the Optical Properties 



the author has repeated the experiments in vacuo. The plate 

 of glass with the film of metal, or the free suspended gold-leaf, 

 was id these experiments contained in a brass box which was closed 

 in by two plane-parallel and accurately parallel plates of glass. 

 By means of a cock the interior of the box, dried by means of bits 

 of chloride of calcium, was placed in communication with the air- 

 pump and exhausted. After exhaustion, the height of the baro- 

 meter was 4*5 millims. 



The appearances were the same in the atmosphere and in vacuo. 

 This, therefore, was the experimental proof that the velocity of 

 light in the above-named modifications of gold and silver is 

 greater than in vacuo, whilst in platinum it is less. 



Were the transparent metallic plates inclined towards the trans- 

 mitted rays, and these, therefore, not at perpendicular incidence 

 to the same, then the conditions of the experiment become very 

 complex. For, at the same time, the thickness of the plate 

 is augmented, and thereby the difference of velocity of the rays 

 passing through metal and air is increased, and the consequent dis- 

 placement of the interference-bands in the spectrum is increased 

 also ; and further, if Cauchy's theory be true, the index of refrac- 

 tion would increase with an increasing angle of incidence, which 

 in the case of metals which at perpendicular incidence have an 

 index of refraction n > 1 would bring about an increase, and in 

 the case of those which have n < 1 would bring about a decrease 

 in the displacement of the interference-bands. Indeed one sees, 

 from the form of the expression for v (equation 4), that the index 

 of refraction in the latter metals may become > 1, and then, for 

 the angle of incidence at which this takes place, the displace- 

 ment of the interference-bands must pass into one in the opposite 

 direction. 



To this must be added the displacement which is due to the 

 elliptic polarization of the light which passes through metals. For 

 light which is polarized in the plane of incidence of the metal, 

 there is produced, according to what has been said before, an 

 acceleration in relation to the light polarized perpendicular to 

 the plane of incidence of the metal. The displacement of the 

 fringes in the instance of metals having an index of refraction 

 n > 1 will therefore be greater for light polarized perpendicular 

 to the plane of incidence, than for light polarized parallel to the 

 same. 



Experiment supported this conclusion perfectly when the 

 metal plates with the glass which carried them were placed 

 obliquely against the incident rays. When the interference- 

 bands of the spectrum were observed through a NicoFs prism, 

 the displacement for light which was polarized parallel or per- 

 pendicular to the plane of incidence of the transparent plate of 



