Permeability 235 
could only give a correct indication of the temperature effect if the 
relative amounts of shrinkage at the different temperatures were the 
same at all times during the experiment, which was not the case. Had 
he chosen to compare the shrinkage after three or four hours instead 
of after two, he would have obtained a quite different set of values 
for the relation between temperature and rate of passage of water 
through the cell membranes. The only way to obtain correct in¬ 
formation on the influence of temperature in these experiments is to 
compare the rate of passage of water out of the tissues at different 
temperatures, but under otherwise exactly the same conditions. This 
is to be accomplished, as pointed out by Miss Delf (1916), by con¬ 
structing the time-shrinkage curves for the different temperatures 
and then comparing the rate of shrinkage at the same stage of shrink¬ 
age, as given by the value of tan a, where a is the angle made by the 
tangent to the curve with the time axis. 
The data of van Rysselberghe have been subjected to this treat¬ 
ment by Miss Delf, when they indicate an increase in the rate of 
passage of water through the cell membranes with increasing tem¬ 
perature up to about 20 0 C., but between this and 30° C. the effect 
of temperature is negligible. Miss Delf found that such a result 
is obtainable only when a very strong plasmolysing solution is 
used, the reason being apparently that there is an "upper limit to 
the rate at which the mechanical tissue system of cells can collapse 
and shrink,” so that with the strong solutions used by van Ryssel¬ 
berghe a great part of the contraction curves obtainable from his 
data represent a contraction of the cell wall system over which 
permeability has no control. 
Other experiments of van Rysselberghe, in which only the times 
taken for cells to plasmolyse or deplasmolyse are given, are obviously 
open to still greater objection, both on the grounds already stated 
in the case of the experiments with pith, and also on account of the 
difficulty of carrying out such experiments with exactitude. Further, 
it is doubtful whether sufficient account has been taken of the great 
variability of plant tissue from different individuals of the same 
species, so that it is not certain whether van Rysselberghe’s results 
could be accepted as expressing more than approximately the relation 
between temperature and rate of intake or excretion of water, even 
if they were not open to grave objection on other grounds. No further 
consideration will therefore be given to the results of Krabbe and 
van Rysselberghe in this place. 
16—2 
