450 Proceedings of Royal Society of Edinburgh. [sess. 
Measurements were repeatedly made on the ammonium salt 
solutions, and on the mixed solutions at the beginning and at the 
end of an experiment, and in every case the above assumptions 
were verified. Immediately after the cell is set up, bridge 
readings are taken at small intervals of time. The conductivity 
first of all increases slightly, which may be a temperature effect, 
then slowly falls over a long period, and finally becomes practically 
constant. The maximum is taken as the correct initial value, and 
the constant end value as the final one. When the final value is 
reached, the mixed solutions may be replaced by the original 
solution of ammonium chloride. 
The whole of the experiments were made at constant temperature, 
generally 25° C. 
Parallel experiments are then carried out, for bromide, with 
solutions of ammonium bromide, aluminium bromide, and. 
ammonia ; and for sulphate, with ammonium sulphate, aluminium 
sulphate, and ammonia. In some of the experiments the solutions 
of the ammonium salts were of equivalent concentration, whilst 
in others they were of the same specific conductivity; the con- 
centrations of the precipitants were always equivalent. 
By this method the only salt produced by the formation of the 
film is the salt which is under investigation ; it is, therefore, 
unnecessary to attempt to remove it, and all that is required is 
that equilibrium between the concentration of the salt in the film 
and in the solution be attained. 
Results. 
Before giving, in the form of tables, the results of the measure- 
ments, we may give in detail some observations which may be of 
interest. 
1. Effect of filter paper and of gelatin on the conductivity of 
solutions. 
A. Potassium chloride solution. 
(1) No diaphragm, conductivity *00422 mho. 
(2) Hardened filter paper, con- 
ductivity . . *00419 „ 99*3 per cent.. 
(3) Gelatinised filter paper, 
conductivity . . *00406 „ 96*2 „ 
