392 Proceedings of the Royal Society of Edinburgh. [Sess. 
be accounted for if we assume that the dissociation of KI for mixtures of 
alcohol and water at 25° C. is larger than the dissociation of KI at 0° C. in 
mixtures of alcohol and water. Now, Jones has shown {Am. Gh. J., xxviii., 
No. 5) from conductivity measurements that this is the case, so that these 
E.M.F. measurements at 0° C. and 25° C. seem to distinctly confirm the 
results obtained from the conductivity measurements, and to show that 
mixtures of alcohol and water at 0° C. have a smaller dissociating power on 
potassium iodide than pure water or pure alcohol. 
The Electromotive Force between Iodine in Alcohol 
and Iodine in Water. 
If solutions of iodine and potassium iodide in water and in alcohol 
respectively, containing equal quantities of these substances, are placed in 
opposition round two platinum electrodes, a considerable E.M.F. is obtained, 
the water solution being positive to the alcohol solution. The action of the 
cell evidently involves, therefore, the transference of iodine from the water 
to the alcohol, and the simultaneous transference of potassium iodide from 
the alcohol to the water. The following values were obtained : — 
E.M.F. at 0° C. 
E.M.F. at 25° C. 
Volts. 
Volts. 
•025 KI + -001 I 2 in water against \ 
*025 KI + -001 I 2 in alcohol J 
•1897 
•1988 
•025 KI + *0001 I 2 in water against \ 
•025 KI + *0001 I 2 in alcohol J 
•1884 
•1979 
The electromotive force is, therefore, independent of the strength of 
the iodine on both sides and of the ratio between the potassium iodide and 
iodine, these readings being the same within the experimental error. The 
combination has a large temperature coefficient in that direction, which 
shows that a considerable amount of heat is being absorbed in the cell, but 
that this is considerably less than that required for a pure osmotic pressure 
cell. As it seemed possible that this E.M.F. was merely a contact E.M.F. and 
not a permanent source of electrical energy, I made up a small cell with a 
parchment-paper diaphragm and two platinum plates, with '01 of I 2 with 
'025 KI in alcohol on the one side, and the '01 of I 2 with '025 KI in water 
on the other side. This, when connected with a galvanometer of 500 
ohms resistance, gave a steady current, falling off a little through polar- 
isation, but continuing for over an hour. 
It is evident that during the flow of this current iodine is being 
transferred from the water to the alcohol solution, and potassium iodide 
