592 Williams.—The Influence of Immersion in Certain 
(b) to find if there be any connexion between the concentration of the 
solution in which the cells are soaked and the time of immersion 
necessary to affect the protoplasm. 
Methods and Material. 
It was decided to deal with cells containing tannin, and to use ferric 
chloride as the material to which the membrane was to be rendered 
permeable. 
In many plants cells occur which are rich in tannin. This substance 
occurs, in such cases, in solution in the cell sap, and in the particular plant 
used is usually insufficient by itself to produce coloration of the cell. The 
protoplasm of the healthy cell is impermeable to the tannin and, for very 
long periods at least, it is impermeable to ferric chloride solution in the 
strength used (0-3 gramme per 100 c.c.). 
As a consequence, sections of the plant can be soaked for a matter of 
two hours in the ferric chloride without a blue coloration appearing in the 
cell, because the tannin and iron solution cannot come in contact. When 
the membrane has been altered, however, soaking for three minutes in 
the ferric chloride will produce a clear, easily recognizable blue colour 
in the cell. 
In order, as far as possible, to equalize diffusion effects, sections must 
be examined after a fixed time in the iron solution. The short interval 
above mentioned was chosen in order that the influence of the ferric chloride 
itself upon permeability, if any, should be negligible. 
The cells chosen were situated in surface strips taken from the leaf¬ 
stalks of London Pride (Saxifraga umbrosa). Under the upper epidermis 
the tannin-containing cells occur in chains, separated by tracts of cells 
without tannin. The cells are large, rectangular ones, which are easy to 
study, and the tissue is in good condition for the greater part of the year. 
The particular plants used were all grown upon the same kind of soil, and 
all strips used, except in preliminary experiments, were washed in a large 
quantity of distilled water for twenty-four hours to remove tannin from the 
torn cells, before immersion in the electrolyte chosen. 
Before one could judge whether the abnormal permeability was general 
in any test, it was necessary to know approximately what proportion of the 
cells probably contained tannin in any section. To ascertain this point pre¬ 
liminary experiments were made with strips treated with i per cent, caffeine 
solution. This liquid readily penetrates the cell membrane and unites with 
the tannin to give a heavy precipitate, which causes the cell to darken. 
A number of these experiments were made on various dates, and the 
cells with, and without, the precipitate were counted. No sections showed 
fewer than 45 per cent, of the cells to contain tannin ; the usual result was 
