136 J. H. Priestley and Edith E. North 
tannin isolated from the cortical tissue by Walter ( 31 ). The brown 
substance present in the membrane of the cortex of frond and rhizome 
of Pteridium aquilinum L. appears to be a phlobaphene arising from a 
phlobo-tannin or catechol-tannin, which can be detected in the young 
fronds by micro-chemical tests, but which is completely absent from 
the mature fronds. Substances of this nature are frequently formed 
from cellulose or other carbohydrate derivatives, as is shown by the 
work of Hoppe-Seyler(i2), Conrad and Gutzheit( 4 ), and Beckley(2). 
As Hoppe-Seyler points out these substances usually only 
arise in dead tissues and he questions the identification with these 
“ humin ” substances, of phlobaphenes appearing in living tissues. 
But in great tracts of the cortical tissue of the Filicinese we have 
a tissue which, if not dead, is dormant, and isolated from the water 
and solute supply in the vascular tract by a most obstructive 
endodermis. The cells as a consequence lack vitality, undergo con¬ 
siderable desiccation and processes are initiated in the ground tissues 
which normally await the death of the plant. Thus phlobaphenes, 
probably anhydride derivatives of catechol-tannins, are formed and 
deposited in the cell membranes and probably contribute towards 
making these membranes relatively impermeable. 
It is a consequence of this view that the starch stored in the 
mature rhizome of a plant like Pteridium aquilinum L. would seem 
to be lost to the plant. We have searched the literature, so far in 
vain, for a description of its subsequent utilisation, and Miss Vera 
Hunter, M.Sc., has made preliminary examination of the rhizome 
during the growth of the leaves in the spring of 1921 without detecting 
any signs of its removal. It may be slowly accessible by cortical 
diffusion to the growing regions but this seems to be unlikely. 
The application of these considerations to phylogeny we leave to 
others, but we would point out that Bower’s ( 3 ) interesting contribu¬ 
tion needs revision in the light of the fact that he assumes an exchange 
under protoplasmic control between cortex and vascular strand in 
plants where the secondary stage of the endodermis is present. The 
general validity of the argument as to the relation between stelar 
bulk and the need for surface exchange of gases and nutrient 
materials may not be affected by such a revision. Such an exchange 
is only essential, probably, so long as the stele is developing, and in 
the short developmental region of the rhizome or frond the endo¬ 
dermis is in the embryonal or primary stage. 
Many problems of great intrinsic interest await the application 
to the Angiosperms of these considerations as to endodermal function 
and structure. Some of these problems are discussed in subsequent 
