ivhen reduced to thin PklKcles. 337 



of a lexture of threads, themselves impermeable to light, 

 such as in black gauze or metallic wire cloth ; ami lastly, 

 that on going back to the action of one point, or of a single 

 material molccnle upon the luminous fluid, we then find 

 the most probable origin of the modifications of the direc- 

 tion of rays which twist or fold the molecules of bodies h\ 

 the varions examples cued, and which being influenced dif- 

 ferently, each according to each species, escape by adifi'erent 

 final direction: whenct results a variety of colours upon 

 those bodies, determined only by the number or distance of 

 their molecules, and without any dependence upon then- 

 proper nature. 



Let us now establish the parallel between these kinds of 

 colours, and of those of the molecules of bodies submitted 

 to the laws of absorption. 



In the first place, with respect to the latter, the luminous 

 bundles are not bifurcated as they are with respect to the 

 former: the rays which do not re-appear in a.given direc- 

 tion, are not thrown out in another direction; they remain 

 absorbed in the substance, even when the mass is perfectly 

 transparent. 



In the second place, the colours resulting from absorp- 

 tion are sometimes owing to groups of rays very diflv-^rent 

 from those which thin pellicles can furnish. For instance, 

 the latter never iurnish colours composed like those of bodies 

 tinged violet by the oxide of manganese, nor like the blue of 

 cobalt or of indigo. 



Besides, there is no relation between the colouring de- 

 pending upon the thickness of any body, in the two kinds 

 or phienomcna. 



We may assure ourselves of this by a comparative exami- 

 nation of my table of tlie varieties of a'osorj)lion *, and the 



* This table, which is described and explained in a preceding: part of the 

 work, givc-i eight pariicular merits oi' ci.loiation, which serve a^ a type of all 

 ll.o-e, ihii posbihillty of v/liich is shown by nature in the ph;j;:ioinena of ab- 

 sf.rption. ' It also presents some very singular results upon the action of co- 

 Uiitred bodies upon lijrlit. 



We hope sooi; to communicate this part of the work at full Icn-^lh in a suc- 

 ceeding iiunilier of these AunaU.— A'<-/<f by the Kditors i,flhc^nnalf:j di Chimie, 



Vol. 28. No. 112. 5'c>/. 1807. V ^S^re 



