448 Proceedings of Roycd Society of Edinburgh. [sess. 
Apparatus and Method. 
As it is obviously necessary to have a certain amount of 
rigidity of support for the precipitation film, the form of apparatus 
must be adapted to the support selected. In the early experi- 
ments various supports were tried, viz., filter paper, hardened 
filter paper, filter paper impregnated with a solution of hard 
gelatine, and several kinds of porous earthenware. 
For these preliminary experiments a Kohlrausch U cell, with 
large electrodes and a narrower tube connecting the two limbs, 
was employed. It was cut into two equal parts, the edges were 
ground true, and a brass coupling joint was cemented on in such 
a way that the diaphragm could be mounted, liquid-tight, between 
two rubber washers. This type of cell was found to be quite 
satisfactory, and was employed in most of the subsequent experi- 
ments. 
Filter paper, hardened and unhardened, was found to be quite 
useless, nor was the gelatinised filter paper much more satisfactory. 
Incidentally, one or two of the results showing the effect of filter 
paper and of gelatinised paper on the conductivity of solutions 
are included later on. 
A fairly hard and compact porous earthenware was found to 
give the most consistent results. It also had the advantage over 
more open material that the diaphragms could be ground very 
thin, and were less liable to fracture when mounted between the 
washers. The same diaphragm was used as often as possible ; it 
was cleaned each time by prolonged boiling with a mixture of 
potassium chlorate and concentrated nitric acid, with hydro- 
chloric acid, and repeatedly extracted with boiling water. It was 
then allowed to soak for a long time in the solution which was to 
be measured. 
In a few of the experiments a different type of cell was employed. 
It was a modification of the Arrhenius type, but the electrodes 
were much smaller, and the upper one was separated from the 
other by a wide glass tube, the lower end of which was closed by 
porous earthenware cemented on water-tight. This form of cell 
had the disadvantage of not being so easily set up and renewed 
as the other one ; the porous plate was much more troublesome 
