Permeability 223 
soluble in lipoid substances which have been found not to enter living 
cells. 
Boas (1921, 1922) made observations on the action of saponins 
on yeast cells and concluded that his results support the lipoid theory. 
He found that saponins in low concentrations bring about an increase 
in the rate of fermentation by yeast, although in high concentrations 
fermentation is retarded. This action is explained on the ground that 
saponins attack lipoids in the plasma-membrane so that the per¬ 
meability is greatly increased and because of this sugars can be more 
rapidly fermented. The argument is not very convincing. 
A modification of the lipoid theory recently proposed by Niren- 
stein (1920) as a result of work on Paramecium caudatum, need not 
detain us long. According to this theory the living cell behaves as 
if it were a lipoid solvent containing a certain amount of fatty acid 
and organic bases soluble in fats, which between them are responsible 
for the uptake of substances, acid dyes, for example, being absorbed 
by the bases, such as diamylamine, in which they are soluble, and 
basic dyes by the fatty acids, such as oleic acid, in which they are 
soluble. Collander points out that this theory is not directly applic¬ 
able to plant cells because accumulation of dyes takes place in the 
cell sap, but it might be possible to regard the plasma-membrane as 
possessing the solvent properties of Paramecium protoplasm. How¬ 
ever, Collander examined seven of the acid dyes which Nirenstein 
had found soluble in diamylamine and which stain living Paramecium , 
but none of these dyes were taken up to any extent by plant cells. 
Nirenstein’s theory is thus quite unacceptable in regard to plant cells. 
The Colloid Precipitation Theory 
A theory that permeability of protoplasm to any particular salt 
is dependent on the capacity of the salt to precipitate the colloids 
of the protoplasm has been put forward recently by Kahho (1921 d). 
This worker finds (1921 a , d) that the influence of salts on the 
coagulation of cell colloids by heat runs parallel with the penetra¬ 
bility of the salts as determined by the plasmolytic or tissue extension 
methods. The series obtained when kations and anions are arranged 
in order of their penetrability have been stated in the last chapter. 
These series are the reverse of the lyotropic series indicating order 
of capacity for precipitating proteins. If kations and anions are 
arranged in order of toxicity or of capacity for coagulating plant 
protoplasm, they fall into the same reverse lyotropic series (Kahho, 
1921 b) while kations of the heavy metals appear to conform to the 
