Polarization by Diffusion of Light. 207 



of diffused light, whether by particles disseminated in a trans- 

 parent medium, or by surfaces not polished. This subject, the 

 study of which I am far from having finished, is connected with 

 my preceding researches on the illumination of transparent 

 bodies*. 



I hardly need to recall here the fact, now well known, that a 

 gas or a liquid holding heterogeneous particles in suspension is 

 illuminated by the passage of natural light, and the light ema- 

 nating from this illuminated trace is polarized in the plane of 

 vision f. If the angle of vision is 90°, polarization is complete 

 if the particles are minute and not too abundant. 



This said, I will report the facts I have observed on employ- 

 ing lampblack as the diffuser, both at the moment of its forma- 

 tion and as a deposit on other bodies. 



If in a dark chamber a pencil of solar rays be made to fall on 

 a smoky flame of illuminating-gas, its trace is perceived very di- 

 stinctly : the portion of the flame traversed by the pencil ap- 

 pears bluish white, contrasting with the reddish tint of the 

 adjacent parts. This experiment can be easily made with a cir- 

 cular (Bengel) burner without a chimney ; in these conditions 

 (i. e. when the combustion is incomplete) the trace of the pencil 

 of solar rays, especially if concentrated by a lens, is very visible 

 in every part of the flame except quite at the base. 



If it be observed with a Nicol or any other analyzer, for an 

 angle of vision of 90° the polarization is ascertained to be com- 

 plete, the white trace disappearing when the analyzer is turned so 

 as to intercept the rays which are polarized in the plane of vision. 



If instead of natural light a pencil of polarized solar light is 

 employed, in the path of which a plate of quartz is interposed, 

 the trace presents well-marked differences of coloration accord- 

 ing to the position of the quartz. 



The same phenomena are manifested when the pencil of solar 

 light passes to the smoke above the flame — that is, to the non- 

 incandescent lampblack. 



salts or oxides, which gave to the two preceding flames their brightness, 

 exhibited no sensible polarization as long as their particles were in the 

 flame, while polarization became manifest outside or in the absence of the 

 flame, although all the conditions except the temperature were absolutely 

 the same? Are we not authorized, or even compelled to conclude that an 

 excessively elevated temperature deprives these particles of their reflecting- 

 power ?" 



* See Archives des Sciences, 1870, vol. xxxvii. p. 129, and vol. xxxix. 

 p. 352. 



t I name, as previously, the plane of vision the plane containing the in- 

 cident pencil of light and the direction in which we view the illuminated 

 body. The angle of vision is the angle made by the visual ray with the 

 pencil of light on the side from which the latter arrives. 



