144 Proceedings of the Royal Society of Edinburgh. [Sess. 
There remained, therefore, only cobalt bromide. Here again the 
solution in acetone gave trouble owing to decomposition. We also inferred 
from the fact that cobalt bromide dissolved in acetone with great evolution 
of heat that some chemical change took place on solution. 
We expected the results in the case of the alcoholic solution of cobalt 
bromide to be analogous with the case of the chloride. After making up 
three solutions of anhydrous cobalt bromide in alcohol, of concentrations 
*218, T09, and ‘054, and selecting the point A = 628 for observation of 
the changes in A, water was gradually added to each solution until the 
colour had changed through the various stages from deep blue to the 
pink colour existing in dilute aqueous solution. Thus solutions I.a, 1.6, 
etc., were obtained from I., and solutions II.a, II.6, etc., from II. The different 
values of A, together with c v the concentration of the salt, and c 2 , the 
concentration of the added water, both expressed in gm.-mols. per litre, 
are given below: — 
Solution. 
A. 
c x . 
1 
c 2 . 
Solution. 
\ A * 
c v 
c 2 . 
Solution. 
A. 
c i- 
c 2 . 
I. 
164 
•218 
•4 
II. 
106 
•109 
•4 
III. 
94 
•054 
*4 
I.a 
109 
•217 
•86 
II.a 
34-6 
•108 
1-09 
III. a 
26-2 
•054 
•98 
Lb 
28*4 
•216 
1-80 
II. 6 
7*13 
•107 
2 02 
III.6 
9-83 
•053 
1-68 
I.C 
9-39 
•212 
2*89 
II.c 
2-01 
•104 
3-18 
III.C 
3*83 
•052 
2-40 
I. d 
2-87 
•210 
4-04 
LI.d 
•73 
•095 
9-40 
lll.d 
1*05 
•C52 
4-07 
I.e 
•89 
•182 
11-77 
like 
•61 
•047 
10*27 
A is plotted as a function of c 2 in the curves in fig. 4, the higher 
concentrations of water being omitted to save space. 
