520 Boodle . — Comparative Anatomy of the 
generally goes hand in hand with physiological and structural 
identity. On the other hand, it is just possible that the 
presence of cortex-like elements within the inner endodermis 
may be due to a certain correlation in tissue-development 1 , 
the formation of an endodermis leading to the production, on 
its inner side, of elements similar to those on the outer side of 
the outer endodermis. 
3. Comparison with Anemia (which belongs to the same 
Order) is of considerable value. Certain forms of that genus 
(in Prantl’s sub-genus Aneimiorrhiza ), which have a creeping 
rhizome and solenostelic structure, appear to form a series of 
reduction, the more xerophilous forms, some of which are 
adapted to growth on rocks, having thinner rhizomes. It 
is highly probable that the structure of the latter should 
be regarded as reduced from the type found in the larger 
forms of this series. To take two examples, which were 
partially investigated, Anemia mexicana has solenostelic 
structure, and central sclerotic tissue continuous with the 
cortex through the leaf-gaps, while A. coriacea, which is 
a plant with a smaller rhizome than that of A. mexicana , 
has in its stele a central core of sclerotic tissue surrounded by 
an inner endodermis, and also has endodermal pockets (see 
Boodle, ’ 01 , PI. XXI, Figs. 41 and 42, e f ), which are indepen- 
dent of the inner endodermis 2 . The most natural conclusion 
is that the endodermal pockets in A. coriacea are reduced 
remnants of previous funnel-like connexions between outer 
and inner endodermes at each node, i. e. of typical foliar gaps, 
such as are found in species with a thicker rhizome, e. g. 
in A. mexicana. Now as the two structural features just 
referred to in A. coriacea are paralleled in Schizaea dichotoma 
1 The writer hopes to publish at a later date some facts, which probably 
find their explanation in a correlation of this kind. As an example of correlation 
of a different kind, but also apart from function, one may quote the dorsiventrality 
of the floral organs of certain Podostemaceae as explained by Willis (’02, p. 438). 
2 These facts are based on the examination of a small amount of dried material, 
so it cannot be said whether important variations of structure may not occur within 
these two species, but the occurrence of the structure described in them is sufficient 
for our present purpose. 
