154 
A. G. Tansley. 
to a reduction in the need for water conduction brought about by 
the diminished transpiration due to the shade-habit. A similar 
cause probably led to the development of the somewhat similar pith 
in most Hymenophyllaceac. A further complication exists. In at 
least two species, 5. digitata and S. malaccana, there is a local 
development of internal endodermal strands, sometimes consisting 
entirely of endodermal cells, and sometimes enclosing a file of cells 
with the characters of ground tissue and quite distinct from the 
surrounding pith. These structures are usually, though not always, 
decurrent from the leaf-gaps, where they are continuous with the 
outerendodermisand cortex,thus forming pockets in exactly the same 
way as in Gleichenit r, etc. (Fig. 49). Whether the pouches are to he 
Fig. 49. Schizcea digitata. Series of transverse sections of part of the stele 
from below upwards, showing departure of leaf-trace and endodermal pocket. 
<•./>., endodermal pocket ; l.t. leaf-trace ; sclcr., sclercnchymatous pith. After 
Boodle. 
regarded as reduced structures, which they certainly suggest, is 
rather a difficult question (Boodle ’03, Tansley & Lulham ’03) that 
would take us too far to discuss Here. In any case there is no 
evidence of internal phloem having ever existed in the ancestors of 
Schizcea and we must regard this peculiar form of stele as closely 
related to the Lygodium- type. The petiolar bundle of Schizcea is 
collateral and usually band-shaped. That of 5. elegans (Fig. 48) has 
a certain resemblance in form to the typical Lygodium petiolar strand. 
The genus Aneimia is partly solenostelic, but some of the species, 
at least, show signs of reduction, and even suggest the Schizcra-typc, 
though they have not only a regular internal endodermis, but also 
internal phloem. 
