2 14 
R. C. Knight. 
tendency for the pressure difference to cause a curvature of the 
leaf, so that the portion to which the chamber is attached tends to 
become convex on the side towards the chamber. In the case of 
thin leaves especially, this curvature is quite obvious even when a 
small pressure difference is employed. Obviously there is some 
likelihood that this change of shape has some mechanical effect 
upon the size of the stomatal pores. If the pressure difference is 
nearly constant, as in the aspirator method, this artificial alteration, 
if any, of the size of the pores is constant throughout, whereas in 
using the Darwin and Pertz form, if it becomes necessary to alter 
the pressure relations obtaining in the system, there is a consider¬ 
able danger of affecting the stomata differently at different times, 
thereby impairing the accuracy. 
4. The quantity of air in the bubble is, of course, an important 
factor in the aspirator method, since, if the bubbles vary in size 
irregularly, error is unavoidable. 
The pressure difference in the aspirator bottle, being practically 
constant throughout the experiment, does not enter into consideration, 
but there are other conditions which are liable to vary as the experi¬ 
ment proceeds, and which may affect the bubble. Change in the rate 
of the air stream is, of course, of constant occurrence, and experiment 
has shown that an increase in this rate results in an increase in the 
volume of the bubbles, which was perhaps to be expected. An 
increase in the speed of the air stream from 13-7 c.c. per hour to 
259-2 c.c. per hour, which is comparable to the range encountered 
in some plants, resulted in an increase in the size of the bubbles of 
nearly 4%. This is relatively a very small amount, probably within 
the limits of the experimental error, and may therefore be neglected. 
A second factor which varies as the experiment proceeds, and 
which might affect the size of the bubbles, is the depth at which the 
bubbles are discharged. As air is drawn into the aspirator bottle, 
the water level falls, and the bubbles are discharged nearer and 
nearer to the surface. Experiment has shown, however, that the 
variation in size of the bubbles with varying depths is practically 
negligible—the differences observed being less than 1% for a 
difference of depth of 7 cms. (10-5 cms. to 3-5 cms.). 
A change of temperature, by affecting surface tension, is also 
liable to bring about change in the dimensions of the bubble. 
Experiments at different temperatures have been carried out and 
have shown that considerable variations are possible. An appara¬ 
tus which at 15-9 ft C. delivered bubbles of average volume 0-126 c.c., 
