Sir David Brewster on the Compensations of Polarized Light. 383 



cipal section of the rhomb, we shall observe in polarized light a beautiful system 

 of rectilineal bands, as exhibited in Fig. 5, where mn is a deep black neutral 

 line, with the usual coloured bands on each side of it. With light polarized oppo- 

 sitely, the central band mn is white, as shewn in Fig. 6, in which the tints are 

 complementary to those in Fig. 5. 



Let us now suppose it required to determine the state or degree of polariza- 

 tion of any luminous surface from which light is reflected, or through which it is 

 transmitted, or of any illuminated medium from which both reflected and re- 

 fracted light are transmitted to the eye of the observer. 



If the light is polarized in the plane of the meridian or a vertical plane, it 

 may be more convenient to use the glass plate at cd, and in doing this the ana- 

 lyser with the rock crystal is fixed between the frame cd and the eye of the ob- 

 server upon a pillar, or it may be held in the hand, so that the principal section 

 of the rhomb is in a vertical plane. The rectilineal bands will then be seen dis- 

 tinctly crossing the luminous surface, when cd is perpendicular to the axis of vi- 

 sion. But if we incline cd from 0° of incidence towards 90, by turning round 

 the goniometer, a position will be found when the rectilinear bands are inter- 

 rupted by a neutral line, as in Fig. 7, and the bands at a on one side of the neu- 

 tral line will be complementary to those at b on the other side. After marking 

 the indication of the goniometer, when this takes place, suppose 1 60°, turn back 

 the goniometer till the light from the luminous surface is nearly as much inclined 

 to the plates on the other side of 0° of incidence, and vary the angle till the bands 

 are interrupted as before, when the observer looks at the same point of the lumi- 

 nous surface. Having again observed the indication of the goniometer, suppose 

 10°, then 160° — 10° =. 150° will be the inclination of two rays equally inclined 

 to the plate, and the half of this, or 75°, will be the angle of incidence upon the 

 plates, at which the polarization of the light from the luminous surface is com- 

 pensated. 



If the light from the luminous surface had been polarized horizontally, it 

 would have been most convenient to have used the rock glass, or other reflector 

 not metallic. In doing this, the luminous surface is reflected at the same angle 

 between the polarizing angle and 90°, and also between the same angle and 0°, 

 the analyser and rhomb being in each case interposed between the reflector and 

 the eye, as before, and the angle of incidence varied till the neutral line mn is 

 opposite to or seen upon the same part of the luminous surface. If the compen- 



