by Diffusion of Light. 53 



M. Lallemand, however, in a recent communication relative 

 to what he calls the illumination of opaque bodies with a dead 

 surface* (that is to say, on diffusion), seems to me to have taken 

 an important step towards my view. In other interesting ob- 

 servations, he has arrived, for a smoked surface, at a result 

 identical with that which I made known in my previous Note, 

 viz. that the light diffused by lampblack is subject, with respect 

 to its polarization, to precisely the same laws as the light emitted 

 by the trace of a pencil of rays traversing a transparent body f. 



Now the lampblack covering a slip of glass, for example, is 

 only an agglomeration of very minute particles in juxtaposi- 

 tion. It seems evident that these particles must continue to 

 diffuse the light according to the same laws, but with less 

 intensity, when, instead of being sufficiently abundant to be 

 contiguous and heaped up on one another, they are more 

 scattered and form only a light deposit on the glass, which 

 partially retains its transparency. This is, in fact, confirmed 

 by experiment. No more can this property be refused to the 

 same particles in suspension in a gas (that is, in the state of 

 smoke or flame), or in suspension in a liquid (for instance, 

 water containing a little Indian ink). It must hence be con- 

 cluded, therefore, that, given a medium destitute by itself of all 

 power of illumination, it will be sufficient to spread in it very 

 thin particles in order to see produced the phenomenon of the 

 lateral propagation of light polarized according to the laws just 

 mentioned. This is an important point which I sought to de- 

 monstrate in my previous researches J, and of which I shall pre- 

 sently give fresh proofs. 



But if, as I think, we are agreed upon this material fact, we 

 again diverge as regards its interpretation, and even upon the 

 cause of the phenomenon. M. Lallemand considers that, in the 

 smoked surface, it is each molecule of carbon, or rather the atmo- 

 sphere of aether condensed around each molecule, that determines 

 the propagation of the light in all directions. For myself, I do 



* Comptes Rendus de VAcademie des Sciences, May 4, 1874, p. 1272. 



f It is necessary to remark only that with a smoked surface the polari- 

 zation is not complete — a fact, besides, often met with in the illumination 

 of transparent bodies. I insisted on this point in my previous Note, and 

 shall return to it further on. 



I must here mention that MM. de la Prevostaye and Desains in their 

 memoir " On the Diffusion of Heat " (Annates de Chimie, 1852, vol. xxxiv. 

 p. 215 et seq.), published some results accordant with these laws, but without 

 enunciating them in a complete manner. Sir D. Brewster, in his memoir 

 entitled " On the Polarization of Light by Rough and White Surfaces " 

 (Philosophical Magazine, 1863, vol. xxv. p. 344), did not study the case of 

 black bodies. 



X See Archives des Sciences, 1870, vol. xxxvii. pp. 150 et seqq. 



