REFLEXION AND REFRACTION OF POLARIZED LIGHT. 103 



and the pencil was divided into two in every position of the rhomb. 

 Prom these facts Mains concluded that the difference between 

 metals and transparent bodies consisted in this: that the latter 

 reflect all the light which is polarized in one plane, and refract all 

 the light polarized in the opposite plane ; while metals, on the 

 other hand, reflect light which is polarized in both planes. 



The subject of metallic polarization was next examined by Sir 

 David Brewster ; and his labours on this subject constitute the 

 most important addition which has been recently made to our 

 knowledge of the laws of polarized light.* When light reflected 

 at a metallic surface is analyzed by a double-refracting crystal, 

 it is observed to be partially polarized in the plane of reflexion. 

 The effect is greatest in galena, and least in silver; and the angle 

 at which it is a maximum is about 74, but varies with the metal. 

 Ey successive reflexions in the same plane, Sir David Brewster 

 found that the proportion of polarized light was increased ; and 

 that by a sufficient number of reflexions the light became, as to 

 sense, wholly polarized in the plane of incidence. The number of 

 reflexions required to produce this effect varied widely in different 

 metals. 



In order to determine the nature and laws of this phenomenon, 

 it is necessary to examine the effect produced upon polarized light. 

 Adopting, then, the method of Malus, Sir David Brewster found 

 that when a ray of light, polarized in the azimuth of 45, was re- 

 ceived upon a metallic reflector at an incidence greater than 40, 

 and less than 86, the reflected light was partly depolarized. The 

 effect produced was greatest at an angle of about 74 ; and when 

 the light underwent a second reflexion in the same plane, and at 

 the same angle, it was restored to light polarized in a single plane. 

 This new plane lies always on the other side of the plane of 

 reflexion ; and its azimuth varies within the limits and 45, 

 being greatest for silver, and least for galena. It is evident, then, 

 that the light produced by a single reflexion cannot be common 

 light. Neither is it plane-polarized light, because it does not 

 vanish in any position of the analyzing rhomb. Sir David Brewster 

 concludes that this light has received a species of polarization 

 hitherto unrecognised, intermediate between plane and circular 

 polarization. He calls it elliptic polarization, because the angles of 



* " On the Phenomena and Laws of Elliptic Polarisation, as exhibited in the 

 Action of Metals upon Light." Phil. Tran*. 1830. 



