72 
Blackman and Paine.—Studies in the 
millimetres each and were made from platinum wire hammered flat. In 
order to increase their surface the electrodes were platinized and roasted 
before insertion. The capacity of the cell was about i c.c. and the volume 
of liquid usually employed was between 0-5 c.c. and 07 c.c. The ‘ cell 
constant which naturally varied with the height to which the cell was 
filled, was of the order of 3 to 4. Above the cell the apparatus widened 
out to a cup-shaped mouth which could be closed with a cork when not in 
use. During an experiment this mouth was left open to air, but the liquid 
in the cell was protected from dust falling vertically by a wide tin lid B, 
perforated in the centre and soldered to the cork-borer which carried the 
pulvinus. 
The liquid was agitated by a regular intermittent stream of air-bubbles, 
a system which proved very efficient and at the same time prevented the 
accumulation in the liquid of carbon dioxide produced by the respiration 
of the pulvinus. The air was forced from a gas-holder and entered at the 
base of the cell through a capillary of such narrow dimensions that a con¬ 
siderable pressure was required. The magnitude of this pressure was 
observed in the open side tube C shown on the left in the figure. The 
pressure alternately rose to about 13 cm. of water and fell to about 12 cm. 
At the maximum pressure air entered the cell in a rapid stream of some 
five or six bubbles. A short interval of time then elapsed before the 
maximum pressure was again attained ; this interval could be regulated by 
means of the screw-clip D compressing the rubber tubing upon a wire passing 
through it. In most of the experiments this interval was approximately 
ten seconds. The whole apparatus was placed in water in a thermostat 
with glass sides. The air before entering the cell was saturated with water 
vapour at the temperature of the thermostat by passage over moist glass 
beads in the tube E immersed in the bath. In this way evaporation of the 
liquid in the cell was reduced to a minimum. In order to prevent electro¬ 
lytes being carried into the cell the splash-trap F was interposed between 
this vessel E and the capillary entrance to the cell. Most of the experi¬ 
ments were carried out at temperatures between 25 0 and 40° C. 
For estimations of conductivity, instead of the usual method of 
Kohlrausch, an ordinary moving coil galvanometer and a rotating com¬ 
mutator of the Fitzpatrick ( 4 ) type was employed. Owing to the weakness 
of the solutions and the high constant of the cell, a high balancing resistance 
had to be used ; one variable between 10,000 and 100,000 ohms, made by 
Gambrell Bros., was usually employed. 1 In all cases the rate of increase of 
conductivity was taken as the measure of the rate of exosmosis. 
1 In order to make certain that the accuracy of the readings was not vitiated by the capacity of 
the resistance, which was of the ordinary type with non-inductive winding, comparative experiments 
were made with a resistance of the same range made of ‘ anti-capacity gauze * by W. Paul. The 
results were found to be identical. 
