Walter Stiles 
54 
permeability to water of suberised walls is indicated by an experi¬ 
ment described by Priestley in which a Soxhlet thimble impregnated 
with a condensation product of potassium phellonate was found to 
be completely impermeable to water. It is significant in this regard 
also that according to van Wisselingh (1886-1892) the middle lamella 
of suberised walls consists of an unbroken layer of suberin, so that 
although cellulose is subsequently deposited on suberised walls, 
there is always present this continuous layer impermeable to water 
separating the cork cell from adjoining cells. 
Cutin forms a continuous layer on the outside of the outer walls 
of the cells of leaves and herbaceous stems. Like suberin it is im¬ 
permeable to water, and it is probably also like suberin in consisting 
of a mixture of several substances, although the evidence is not so 
definite. The cutinogenic acids obtained from cutin are, however, 
not the same as the suberogenic acids obtained from suberin. 
With the chemistry of lignin, suberin and cutin we cannot deal 
further here. Those interested should consult the standard textbooks 
on the subject (especially Czapek, 1913) and the recent paper by 
Priestley already cited. 
The coats of many seeds have a very low permeability to water. 
In some cases this is no doubt due to cutinisation or suberisation of 
cell walls, but in other cases to the presence of tannins, lipoid and 
pectic substances, as treatment of certain seed coats (e.g. pumpkin, 
almond, peanut) with solvents for these substances greatly increases 
the permeability (Denny, 1917 b). 
Before leaving the subject of the cell wall, it should be noted that 
the young cell wall is capable of stretching when acted upon by a 
force, while on removal of the force it tends to return to its original 
condition. It is thus elastic. With change in composition as the wall 
gets older this extensibility and elasticity may be lost, but cellulose 
walls generally retain extensibility and elasticity even when con¬ 
siderably thickened. 
Mention has already been made of the pits in cell walls (cf. 
Chapter II). As the rate of diffusion through a membrane when 
diffusion is proceeding at a uniform rate varies inversely as the thick¬ 
ness of the membrane, it has been supposed that these pits help to 
expedite diffusion (Haberlandt, 1914) as the pit-closing membrane 
is considerably thinner than the thickened part of the cell wall. It is 
to be supposed in this connection that the relation between diameter 
of the pore and rate of diffusion through the pore found by H. Brown 
and Escombe (1900) also holds for the diffusion of substances in 
