Permeability 161 
brief, but containing all the essential facts, reference may be made 
to an article by F. F. Blackman (1912). 
Osterhout (1911, 1913 a) concluded that the plasma-membrane 
could not be lipoid in character because, contrary to the finding of 
Overton, Osterhout had shown that a large number of inorganic salts 
penetrate into the cells of Spirogyra. Further, although both sodium 
chloride and calcium chloride when presented in pure solution can 
enter Spirogyra cells, neither of these salts can penetrate into these 
same cells when present together in a solution in certain proportions. 
To this phenomenon of "antagonism” we shall return later. It will 
be sufficient here to remark that Osterhout regards his experiments 
as indicating a protein rather than a lipoid plasma-membrane. 
In summing up the data here presented dealing with the com¬ 
position of the limiting membrane of the protoplasm, it is clear that 
any view which regards the surface layer as composed exclusively of 
one kind of substance, either fatty or protein, must be dismissed at 
once. Both on theoretical grounds, and as the result of experimental 
work, it seems clear that the limiting layer of the cell must contain 
those constituents of the protoplasm and of the external liquid which 
lower the surface tension. These substances will thus constitute a 
colloidal complex, either a sol or gel, but how the molecules of the 
various constituents are arranged in reference to one another our 
present knowledge is insufficient to indicate. 
Finally two points appear to deserve special emphasis. Firstly, 
on the view of the plasma-membrane here adopted, it is clear that 
the limiting membrane of the cell is not an invariable structure, but 
will vary with alterations in the composition both of the external 
medium and the protoplasm. It is clear that the hypothesis of de 
Vries, that the plasma-membranes are permanent structures never 
arising de novo but only from pre-existing membranes, is completely 
untenable. 
Secondly, we can understand how it is impossible to separate the 
limiting membrane from the rest of the protoplasm. It depends for 
its existence on, and is closely related to, the media which it separates, 
and we should expect it to pass over into the two phases on either side 
of it, gradually in some cases, more or less sharply in others, accord¬ 
ing to the composition of the protoplasm and the external medium. 
Thickness of the Plasma-Membrane 
Seifriz (1921) has recently discussed this question. It is obvious 
that the plasma-membrane must be very thin, and that any measure- 
Phyt. XXI. III. 
11 
