5 « 
D. Thoday. 
membrane to particular solutes, and on the other the increased rate 
of entry and final percentage of water in their presence. Recent 
work has led them to lay stress on the correlation between both 
these effects and the lowering of the surface tension of water 
brought about by those substances in solution which penetrate the 
coat . 1 
In potato parenchyma the first and most important complication 
lies in the complex nature and chemical instability of the chief 
differentially permeable membrane present, the protoplasm. The 
water content of the vacuolated cells is, moreover, considerable, so 
that the entry of water is more an osmotic phenomenon than an 
imbibitional swelling of colloids, although such swelling—of starch 
as well as cellulose and proteids—must occur. It has also to be 
remembered that the turgescent condition of the tissues introduces 
from the very beginning of the experiment what Brown and Tinker 
call a “ backward pressure” retarding the entry of water. In addition, 
the presence of air-spaces introduces a possible source of error, 
especially in dealing with substances that lower considerably the 
surface tension of water in contact with air ; for solutions of these 
substances might, by virtue of this very property, tend to inject 
the air-spaces and so cause an immediate increase in weight, as well 
as a more rapid subsequent entry owing to the larger number of 
cells brought into direct contact with the solution. 
In view of such considerations, although the results obtained 
were interesting, some of them curious, they were put on one side, 
as further work would have led away from the track of the main 
problems under investigation at the time. No opportunity of 
following them up with more critical experiments has since occurred; 
but, in view of the recent appearance of a paper by Stiles and 
Jorgensen 2 describing promising preliminary work 3 along the same 
lines, it seems an opportune moment to make the chief results 
available. 
I. Experimental Results. 
The method used was cruder than that elaborated by Stiles 
and Jorgensen. Nevertheless the errors of manipulation and 
weighing were small, and errors of sampling do not seriously affect 
the form of the individual curves : while even a rough quantitative 
1 Loc. cit. ; also Proc. Roy. Soc., B 89, p. 373. 
J “ Studies in Permeability, V.” Annals of Botany, XXXI, 1917, p. 415. 
" The curves given in their Fig. 4 (loc. cit., p. 422), showing the different 
behaviour of pieces of potato in distilled water at various temperatures, are a 
particularly encouraging example of what can be done by this method, used 
with care and discrimination. 
