146 Proceedings of the Royal Society of Edinburgh. [Sess. 
Liberation of Ions . — A small concentration of free silver ions occurs 
in silver sols. The amount, however, appeared to be insufficient to affect 
the conductivity, as determined by the telephone method of the sol 
chiefly used. But in view of the much greater bactericidal power of 
solutions of silver salts as compared with those of colloidal silver, it 
seemed a plausible hypothesis to attribute the effects observed to the 
liberation of free ions from the ultramicroscopic particles. Such an 
explanation occurred to Cernovodeanu and Henri,* for they state that 
they filtered a colloidal silver sol through a collodion filter and found 
the filtrate to be free from bactericidal or antiseptic action. Their 
investigation was unknown to me at the time I made the follow- 
ing experiment. Electragol containing 0‘044 per cent, of silver was 
filtered under a pressure of forty atmospheres through a gelatined 
Chamberland candle, which had been hardened in 10 per cent, alum sol 
and washed in distilled water for several days. A clear, colourless filtrate 
was obtained which was neither bactericidal nor antiseptic. When left 
in a beaker in the laboratory for a few days an abundant growth of 
bacteria developed. 
Effect of Size of Particles . — The size of the particles in the sols used 
was not actually measured ; but from their stability, and the absence of 
precipitate after standing for three years, we may conclude that they 
probably contained few particles much above 60 fj.fi in diameter. In view 
of the fact that the concentration of free silver ions was insufficient to 
explain the pharmacological action o£ silver sols, it was decided to investi- 
gate in some measure the influence of size of particles. It was found that 
a sol prepared with conductivity water containing five-thousandth normal 
sodium hydroxide was suitable from this point of view for comparison with 
the colloid previously used. The amount of silver and the number of 
visible submicroscopic particles above 15 fifi having been estimated, the 
light diminishing (scattering) power of the two sols was compared. This 
was done by means of a spectrophotometer, a beam of light from an 
incandescent mantle being passed through a column of 5 cm. of sol, and 
the absorption estimated in the middle violet, the middle green, and the 
middle red of the spectrum, by comparison with crossed nicols. The angle 
of rotation of the analysing nicol in the violet and green with the number 
of visible submicrons (above 15 /jl/ul) and the percentage of silver, is given 
for the two preparations in the appended table. For convenience the 
electrolyte-free colloid has been termed A ; the colloid prepared with 
sodium hydroxide solution, B. 
* Com'pt. Rend, de la Soc. de Biol ., vol. lxi, p. 123 (1906). 
