428 On the Experimental Relations [1857. 



until the gold is reduced, it will generally be found that the 

 vapour has carried a portion of gold on to the neighbouring 

 part of the glass, and that this part, when placed over a sheet 

 of white paper, has the ruby tint. With the rock-crystal both 

 ruby and blue parts are produced ; and when the ruby parts 

 are subjected to rock-crystal pressure, they become beautifully 

 green. In the arts also glass is oftentimes coloured ruby by 

 gold ; I think that glass in this state derives its colour from 

 diffused divided gold; and if either the ruby glass or the 

 watch-glass be examined by a lens and the cone of rays, it will 

 be seen that the colours are not due to any gold dissolved, but 

 to solid and diffused particles. There is nothing in any of the 

 appearances or characters, or in the processes resorted to to 

 obtain the several effects, that point at any physical difference 

 in the nature of the results ; and without saying that gold can- 

 not produce a ruby colour whilst in combination or solution, I 

 think that in all these cases the ruby tint is due simply to the 

 presence of diffused finely-divided gold. 



Metallic character of the divided gold. 



Hitherto it may seem that I have assumed the various pre- 

 parations of gold, whether ruby, green, violet, or blue in colour, 

 to consist of that substance in a metallic divided state. I will 

 now put together the reasons which caused me to draw that 

 conclusion. With regard to gold-leaf no question respecting 

 its metallic nature can arise, but it offers evidence reaching to 

 the other preparations. The green colour conferred by press- 

 ure, and the removal of this colour by heat, evidently belong 

 to it as a metal ; these effects are very striking and important 

 as regards the action on light; and where they recur with other 

 forms of gold, may be accepted as proof that the gold is in the 

 metallic state. Although I do not attach equal importance to 

 the fact already described, that gold-leaf frequently presents 

 fine parts that appear to be ruby in colour, I am not as yet 

 satisfied that they are not in themselves ruby ; and if they 

 should be so, it will be another proof by analogy of the metallic 

 nature of other kinds of preparations eminently ruby. 



The deflagrations of gold wire by the Leyden discharge can 

 be nothing but divided gold. They are the same whatever the 

 atmosphere surrounding them at the time, or whatever the 



