﻿of Light by Metallic Particles. 447 



hours, or days, the gold was thrown down as usual in particles 

 of extreme fineness. For the present purpose, however, there 

 is no advantage in using terebene, since any solution which is 

 transparent must be regarded as non-conducting for voltages 

 reversed with the frequency of light-waves. Accordingly 

 the gold deposits which I have examined have been formed 

 in water. When a very dilute solution of the double chloride 

 of gold and sodium — say, one centigram of the crystallized 

 salt per litre— is reduced by the addition of a solution of 

 phosphorus in some organic solvent, the solution is observed, 

 after a time varying from a few hours to several days or even 

 weeks, to become strongly coloured. The colour varies from 

 a clear pink with dilute solutions to an opaque purple when 

 the solutions are strong. A number of observations on these 

 precipitates and on their removal from suspension (for they 

 are too fine to subside of themselves) by moulds growing in 

 the liquids have been published by Prof, Liversidge (Australian 

 Association for the Advancement of Science, Report, 1890). 

 I have to thank Prof. Liversidge for allowing me to examine 

 these solutions. 



There are certain preliminary questions which must be 

 dealt with. In the first place the precipitate formed is really 

 gold, 



(1) Because in rather strong solutions of the chloride the 



precipitate forms a film on the surface which possesses 

 the lustre of gold ; 



(2) Because when the solutions are strong enough to give 



a precipitate coarse enough to sink, or when the fine 

 precipitates are carried down by the growth of mould, 

 the presence of metallic gold can be rendered evident 

 by burnishing. 

 I have no formal proof that the precipitates consist of gold 

 only ; but taking into consideration the number of expe- 

 riments made by Prof. Liversidge and the great variety 

 of reducing solutions used by him, together with the constancy 

 of the phenomena observed, there can be no doubt that the 

 precipitate is practically pure gold. Of course, when the 

 precipitates are large enough to examine there is no doubt 

 about their nature, and we have no chemical reason for sup- 

 posing that the reaction will change in character by the mere 

 influence of further dilution of the solutions. 



The colour of the solutions is certainly to be attributed to 

 the presence of the precipitate ; for when the precipitate 

 is removed by moulds the colour disappears. The colour, 

 therefore, must be due to the fine particles ; but without 

 careful examination it is difficult to believe that the whole 



