346 Sir David Brewster on the Polarization of Light 



the room the polarization increases as the glass recedes from the 

 window. In the open air it is a maximum when the plane of re- 

 flexion passes through the sun and the observer. The rotation is 

 then 15°, and it diminishes with the distance of the plane of re- 

 flexion from the plane passing through the sun and the observer*. 



When the second or polished surface is not blackened, it reflects 

 the light scattered by the ground surface, and the maximum 

 polarization of the rough surface is greatly diminished. The 

 rotation is 16° with a gas-burner very near, and 20° when the 

 flame is still nearer. The polarization is diminished by holding 

 a lighted candle on one side of the plane of reflexion. 



With a similar plate having a rougher surface, the degree of 

 the polarization of the sun's rays directly reflected was 16|° close 

 to a south window. Be- F - j 



fore noon the angle of 

 maximum polarization 

 was 44°, and at noon, 

 with a brighter sun, it 

 was 40°. 



When the whole light 

 of the sky fell upon the 

 rough surface M N, the 

 rotation was 21°. When 

 A B was covered up, the 



rotation was diminished ; and when C D was covered up, the rota- 

 tion was increased. 



When both the surfaces of the plate of glass are rough, the 

 polarization is a maximum, at an angle greater than the normal 

 polarizing angle, or 56^°, and the degree of polarization is 

 about 20°. 



In making similar experiments with ivory, bone, porcelain, 

 white or coloured surfaces painted in oil, paper, parchment, silk, 

 linen, and cotton cloths, milk, flour, and other white powders, 

 &c, I found that the polarization was partial in all of them, the 

 normal or complete polarization being reduced by its combina- 

 tion with the oppositely polarized rays produced by refraction. 



In all these experiments the partial polarization was positive 

 or vertical from an incidence of 90° up to an incidence of 0°, as 

 far as it could be observed, owing to the impossibility of examin- 

 ing the bands where they were reflected at or near a perpen- 

 dicular incidence. In order to meet this difficulty, I adopted 

 the following mode of observation : — 



* The roughness of the glass surface used in the preceding experiments 

 is such, that a gas-flame, distant 5i feet, ceases to be visible at an angle of 

 incidence of 7^° 50'. 



