418 On the Experimental Relations [1857. 



warm rays are reflected by gold, in preference to the others 

 contained in white light. 



The supernatant fluid in specimens that had stood long and 

 deposited, was always ruby ; yet because it showed no dissolved 

 gold, because it showed the illuminated cone by the lens, and 

 because by standing ruby clouds settled in it, there was every 

 reason to believe that the gold was there in separated particles, 

 and that such specimens afforded cases of extreme division, 

 which by long standing would form deposits of the finest kind. 



Those fluids which on standing gave abundance of deposits, 

 transmitting blue light, consisted in the first instance of particles 

 transmitting a ruby light, and in these cases it would seem that 

 the particles at their first separation were always competent to 

 transmit this ruby light ; and if the preparation were not too 

 rich in gold, the ruby condition appeared to be retained, the 

 division being then most extreme. But purple or amethystine 

 fluids could be procured, which, containing no colouring par- 

 ticles other than suspended gold, still retained them in suspen- 

 sion for many months together, so that they must have been 

 as light or as finely divided as those in the ruby fluids. When 

 the phosphorous ether was employed for the reduction of the 

 gold, such fluids occurred ; also when the solution of the phos- 

 phorus in sulphide of carbon was used, provided the solution 

 of gold had a very little chloride of sodium contained in it. 

 They appear to show that the mere degree of division is not 

 the only circumstance which determines the aptitude to trans- 

 mit in preference this or that ray of light. 



Considering the fluids as owing their properties to diffused 

 particles, it may be observed, that many of them which in small 

 quantities in the dark tube transmit an amethystine light, send 

 forward a ruby light when the quantity is increased ; and this 

 appears to be the general progression. 1 have not found any 

 which by increase in quantity tended to transmit the blue rays 

 in preference to the red. 



Elevation of temperature had an effect upon these fluids 

 which is advantageous in their preparation. On boiling an 

 apparently clear ruby fluid for some time, its colour passed a 

 little towards amethystine, and on boiling a like amethystine 

 fluid, its tint passed towards blue. The separation of the gold 

 particles was also facilitated, for now they would settle in three 



