138 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



until, in half or three quarters of an hour, it gives a uniform 

 yellow of the first order. The colors are as brilliant as those 

 given by selcnite films in polarized light. 4 D, 18*74, VII., 

 415. 



POLAKIZATION BY DIFFUSION. 



Soret, commenting upon the theory of Him, adduces a 

 number of experiments made by himself in the same line of 

 research. He states that he has repeated the famous experi- 

 ment of Professor Tyndall on the polarization of light re- 

 flected from a gas, or liquid, holding hydrogeneous particles 

 in suspension. He has employed lamp-black as the diffuser, 

 and detected the polarization of the light immediately that a 

 beam of sunlight was made to fall on a smoky flame of illumi- 

 nating gas, as well as when it fell upon the smoky or non-in- 

 candescent lamp-black above the flame. With a Bengal burn- 

 er and glass chimney, the flame ceased to be smoky, and the 

 luminous pencil was no longer perceived. 7 A, XLVIL, 205. 



THE POLAEIZATION OF FLAMES. 



Hirn, who is known by his investigations in the department 

 of heat as well as in reference to the rings of Saturn, has also 

 lately occupied himself with the optical properties of flames; 

 and, for the explanation of the whole of the phenomena, he 

 suggests the hypothesis that the solid incandescent particles 

 to which the flame owes its brightness become transparent 

 at that high temperature, and have no longer any sensible 

 reflecting power. As is well known, one of Arago's first ob- 

 servations in optics demonstrated that, in general, the light 

 from flames presents no trace of polarization ; and it was this 

 observation that served to establish a most important land- 

 mark in the theory of the sun, by informing us that the solar 

 light emanates from a gas, and not from a solid or liquid. 

 This fact, discovered by Arago, is not inconsistent with other 

 known phenomena when the flame is that of a burning gas, 

 but when it is formed of a mixture of a burning gas with the 

 dust of a solid body in which case not only does each solid 

 particle emit light of its own, but must reflect light from oth- 

 er sources, since it is illumined by other particles then how 

 is the total absence of polarization to be accounted for ? Ac- 

 cording to Hirn, it is not difficult to explain this in the case 



