Walter Stiles 
154 
rest of the protoplast (Pfeffer, 1890, 1900). Thus, when a cell is 
treated with very dilute hydrochloric acid, the plasmatic membrane 
assumes “a condition of rigor,” but its original diosmotic properties 
are usually at first retained, and a dye, which is unable to penetrate 
the plasma-membrane will rapidly penetrate throughout the dead 
protoplast if it can find entrance through a tear or gap in the mem¬ 
brane. The fact, already referred to, that the vacuolar membrane 
may actually be separated by strong and sudden plasmolysis, is cited 
as a further argument. The reader must form his own judgement of 
the value of these lines of evidence derived from pathological con¬ 
ditions; to the present writer they are very unconvincing, and it is 
to be noted that Pfeffer himself did not press them. 
With regard to non-vacuolate cells, then, it must be admitted 
that the evidence is insufficient to enable one to come to any con¬ 
clusive decision as to the presence of a limiting protoplasmic mem¬ 
brane. But here again, as with the water relations of the cell, it is 
impossible to avoid the conclusion that substances may be able to 
pass into the protoplasm and may be held there simply on account 
of the colloidal properties of the protoplasm. The cases dealt with 
by Bayliss are much simpler than those presented by living cells. 
When the state of equilibrium between the very complex system 
which constitutes the latter, and the external medium, is disturbed 
by changes in the composition of the external medium, it is clear 
that there may be expected far-reaching changes in regard to molecular 
association in the protoplasm, changes which may involve adsorption 
of substances from the outer liquid, chemical combination, or 
breaking down of existing molecular associations in the protoplasm 
to bring about a new condition of equilibrium. While the possibility 
of a limiting semi-permeable plasma-membrane is not ruled out, the 
evidence at present available from a consideration of the relation 
of the cell to substances dissolved in the medium outside it, is 
by no means overwhelmingly in favour of the presence of such a 
membrane. 
An observation made by Osterhout (1913 b) on the marine alga 
Griffithsia may be taken as an example of one line of evidence adduced 
in support of the different permeabilities of the two surfaces of the 
protoplasm. If cells of this alga are placed in a solution having the 
same relative composition as sea-water, but stronger, plasmolysis 
takes place, and on returning the cells to normal sea-water the 
protoplasts regain their original size. But if instead of sea-water a 
solution of ammonium chloride is used as plasmolysing solution, 
