﻿of Light by Metallic Particles. 449 



was made at the same time as the one last described. The 

 solution must have been of about the same strength, and the 

 reducing agent was phosphorus dissolved in carbon bisulphide. 

 The smell of that reagent had disappeared and been replaced 

 by that of sulphuretted hydrogen. Observations similar to 

 the last were made with the same general result, except that 

 the absorption was stronger and that it extended rather 

 further towards the red end of the spectrum. The scattered 

 light was again seen to correspond with the part of the 

 spectrum which was cut out in the last observation. The 

 angle of maximum polarization occupied its former position. 



A large number of similar experiments were made from 

 time to time, but they all gave the same general results. 



In order to avoid any risk of contaminating the gold with 

 organic matter derived from the solvent used for the phos- 

 phorus, I shook up a bit of phosphorus with some distilled 

 water, and found that I could obtain a sufficient reduction of 

 the gold to enable me to be sure that the angle of maximum 

 polarization occupied its former normal position. 



1 also set up solutions of different strengths, and observed 

 them from day to day as the process of reduction went on ; 

 but I never observed any change in the position of the 

 maximum angle. 



Most of the observations on the direction of the position of 

 maximum polarization were made on about a litre of the 

 solution contained in a vessel of glass with plane sides ; occa- 

 sionally I used beakers, which did just as well. 



The source of light was an arc lamp, and when the particles 

 were only few and far between lenses were employed to 

 condense the beam. Some experiments were also made with 

 initially polarized light, but without any difference being in 

 any case detected, whether a gold or a mastic sky was used. 



Metals other than Gold. 



1. Copper. — A solution of sulphate of copper was taken and 

 diluted sufficiently with water that had been shaken up with 

 phosphorus. Of course check experiments were made in this 

 case, as in all others, to insure the absence of particles other 

 than those desired. In many cases it was necessary to use 

 platinum-distilled water, as ordinary once- distilled water con- 

 tains too many motes. A very faint sky was obtained with 

 the copper, and this behaved just as did the gold sky. I 

 believe that the sky did actually consist of copper, for the 

 solutions were both free from motes before mixing, or rather 

 they gave a much stronger sky when mixed than when separate. 



2. Silver. — The phosphorus-water was shaken with dilute 



