[ 335 ] 



XLIV. On the Polarisation of Light reflected by Glass. 

 By M. Ed. Desains*. 



MARAGO has demonstrated by experiment that in two 

 • bundles of hght reflected under angles differing equally 

 more or less from the angle of polarization of the reflecting body, 

 the ratios of the quantities of light polarized to the total quan- 

 tity are sensibly the same. 



For the calculation of the numerical value of these ratios, 



Vi 11 n ^ .1 P 1 cos^(e — ;•)— cos'^(e— r) . , . 



rresnel has lound the lormula — ^r- r~; 97^. — ^> « bema- 



cos''(z — r) + cos^(2 -\-ry ° 



the angle of incidence and r the angle of refraction. He has 



found that this formula was verified by the above law of M. 



Arago. 



The object which I have proposed to myself is to determine 

 by observation these numerical values for various incidences of 

 the ray upon glass ; employing in the determination the photo- 

 metric processes of M. Arago. 



I furnished myself with a pile of thirteen glasses, which I first 

 graduated. For this purpose I used a tube, terminated at its 

 two extremities by two collars, one of which contained a plate of 

 quartz cut parallel to the axis, and the other the pile of glasses. 

 Pex'pendicular upon the plate of quartz 1 caused a ray to fall 

 which was polarized in a plane inclined at an angle a to its 

 axis ; the ratio between the polarized light and the total light of 

 the transmitted ray was cos«^ — sin«^, according to the law of 

 the square of the cosine, demonstrated experimentally by M. 

 Arago ; and the plane of partial polarization of this ray coincided 

 with the axis of the quartz, or was perpendicular thereto, accord- 

 ing as a was greater or less than 45 degrees. 



The transmitted ray fell afterwards upon the pile of glasses, 

 which were so disposed that the plane of incidence and the plane 

 of polarization were coincident. I inclined the pile until it de- 

 polarized the ray, and I measured the angle <^ which this ray 

 then made with a perpendicular to the pile. By causing « to 

 vary I was able to arrange a table of graduation, containing in 

 one columir the ratio between the polarized light and the total 

 light of different rays polarized partially, and in another column 

 the angles which these rays made with the ])er])endicular to the 

 pile when depolarized. 



The pile being thus graduated, I caused rays of un{)olarized 

 light to fall at different incidences,?, upon a dark glass; the rays 

 reflected were received upon the pile, care being always taken to 

 make the plane of reflexion from the glass to coincide with the 



♦ Translated from Compten Rendus, Nov. 1860, No. 20. 



