505 
Protoplasm to some Reagents. 
on the protoplasmic surface would then in time, even in a 2 per cent, 
solution, be sufficiently high to react with the lipoids. But, after all, there 
is no sound reason for believing lipoids to be the sole constituents of the 
plasma membrane. If we accept Nathansohn’s ( 14 ) mosaic conception of 
membrane constitution—and it seems to me that we must—then we are 
forced to grant the likelihood of alcohol attacking proteins and other 
possible constituents of the cell membrane. 
Saponins of low concentration attack (emulsify) lipoids. That lipoids 
are a part of the chemical make-up of the plasma membrane is generally 
accepted. MacDougal ( 11 ), without emphasizing the presence of any one 
single substance in the plasmatic membrane, regards lecithin as an important 
constituent. Other workers, Boas ( 4 ), Kahho ( 10 ), et al. y have been led by 
their experimental results to the belief that lipoids are one of the chief con¬ 
stituents of the plasma membrane. Bayliss ( 1 , p. 130) regards certain 
permeability and other phenomena as attributable to the presence of 
a lipoid peripheral layer. 
It is unwise, in the present limited state of our knowledge of the 
chemical constitution of protoplasm, to give undue emphasis to any one 
single substance in the membrane, yet that lipoids are one of the possible 
components of the protoplasmic surface layer seems likely. Whether they 
occur as free fat or, more likely, in chemical union with proteins ( 15 ), we do 
not fully know. Whatever the actual chemical make-up of the membrane 
is, the experimental data here given stands in opposition to the assumption 
of Grafe ( 9 , p. 29), that a knowledge of ‘ the purely chemical constitution of 
the cell constituents has not given any explanation of the entrance and exit 
of different substances ’. 
The Overton lipoid theory must still stand as one of the possible 
factors concerned in permeability phenomena. 
It is reasonably clear that both alcohol and saponin increase per¬ 
meability in the Elodea cell. Such an increase in permeability means 
a more dispersed membrane, and therefore a lower surface tension. Con¬ 
versely, a reduction in the surface tension of protoplasm will mean increased 
permeability. There is, consequently, good reason to regard surface 
tension forces as playing a part in changes in permeability. It seems likely, 
therefore, that the surface-tension ideas of Traube and Czapek may not be 
as fallacious as some have implied. That a reduction in surface tension 
actually does take place in the treated cells is confirmed by the following 
observations: 
Untreated Elodea leaf cells mostly plasmolyse with concave surfaces, 
which are often very deeply concave, indicating a high su: rface-tension value 
of the plasma membrane of the normal cell. The treated cell (half an hour 
in 10 per cent, alcohol), on the other hand, plasmolyses, nine times out of 
ten, with a convex surface, even when the plasmolysis is ever so slight, 
