1912-13.] The Absorption of Light by Inorganic Salts, No. IX. 145 
The curves when produced back met at first at the point c— — *4, A = 215. 
From this we inferred, as was done in the sixth article, that the alcohol 
contained *4 gm.-mol. of water per litre, or *95 per cent, by weight, and 
that the value of A for a solution in pure alcohol would be 215. The 
values of c 2 given in the preceding table are the corrected values, i.e. 
they include this percentage of water present originally in the absolute 
alcohol. 
Assuming that the change in colour is due to the formation of a 
hydrate of the form CoBr 2 a;H 2 0 at the expense of the anhydrous salt, 
and that the alcohol takes no active part in the change, we shall have 
a balanced action between the anhydrous salt, uncombined water, and 
hydrate of the form 
CoBr 2 + ccH 2 0^=±=CoBr 2 . £cH 2 0. 
Assume that A = 0 for the hydrated phase at 
question. Then the active mass of the anhydrous 
therefore that of the hydrated phase ^1 — 9 jg) c r 
the free water is c 2 — as^l — 
The law of mass action gives 
the wave-length in 
phase is A c i> and 
Zlt) 
The active mass of 
where k is a constant and the dashes denote another set of possible values 
of A, c v c 2 . 
Put A = A' and multiply up. Then 
To calculate x, ordinates were drawn in fig. 4 at A = 10, 30, 50, 70, 100, 
and 150, and the length of the ordinate intercepted between curves I. and 
III. or between the curves I. and II. measured. This gave c 2 — c' 2 for 
the value of A in question. c x varies very slowly along each curve, and 
the value of c 1 — c\ could be obtained from the preceding table. The 
table on p. 146 gives the results. 
Of course, the results depend on the way the curves are drawn through 
the points in fig. 4. They could easily be drawn to secure more consistent 
VOL. xxxiii. 10 
