1912-13.] The Absorption of Light by Inorganic Salts, No. X. 151 
as a side issue in the earlier papers. It was thought advisable to collect 
here the data already obtained and to obtain additional data. 
Our previous work in the case of cobalt and nickel, and our work 
together with that of Muller * in the case of copper, showed that the molecu- 
lar extinction coefficient did not vary much with the concentration except 
at special points in the spectrum. Such points we might term “ sensitive 
points.” At these points the change was much more marked in the case 
of the chlorides and bromides than in the case of the sulphates and nitrates. 
The data for these points are collected in the table on p. 152, and illustrated in 
the following curves. The values for nickel chloride are taken from No. III. 
of this series, for ferric chloride and bromide from No. VII., and those for 
copper bromide from Muller’s observations. The other values are new. 
For the sake of comparison, the variation with the concentration of jul 0 ,, 
the molecular electrical conductivity at 0° C., as determined by Jones, f 
is shown. 
Our results show that the molecular extinction coefficient remains, 
constant at low concentrations — the single exception in the case of ferric 
chloride is due to the formation of colloid hydroxide — and increases- 
asymptotically as saturation is reached. Its behaviour is of a totally 
different nature from that of the molecular conductivity, as may be seen 
from the curves. There is thus no connection between the two quantities. 
Ionisation has thus nothing whatever to do with the colour changes of 
cobalt, nickel, copper, and iron salts. What is more, supporters of the 
theory of ionic dissociation are forced to assume that the ions have 
exactly the same colour as the neutral molecules. 
The third point investigated was the absorption of anhydrous cobalt 
chloride dissolved in acetone, and the effect on the absorption of adding 
water to this solution. In the sixth article, it will be remembered, the 
effect of adding water to an alcoholic solution was studied very fully, and 
the results were explained by the law of mass action on the assumption 
that the alcoholic solution gave the spectrum of the anhydrous salt, and 
that when water was added a polyhydrate was formed. 
When anhydrous cobalt chloride is dissolved in acetone the solution 
does not keep well. On standing it deposits a black residue ; also there 
is trouble due to evaporation. Consequently the values of A could not 
be determined very exactly. However, they agree roughly with the 
values obtained with ethyl alcohol as solvent. 
* Ann. d. Phys. (4), 12, p. 767 (1903). 
t “ Hydrates in Aqueous Solutions,” Jones, Publication No. 60 of the Carnegie Institution 
of Washington. 
