Filicales. 
233 
embedded in central parenchyma and surrounded by two bands of 
tracheides. Mr. Boodle admits that the species with larger solid 
steles, such as T. scandens, might be regarded as more primitive 
than the type of T. reniforme, but he claims that the sub-coriaceous 
fronds of many of them indicate that they have probably become 
adapted to less moist conditions, and that this has brought about 
the increase in the conducting elements of the stele (3). Mr. 
Boodle strongly supports the view that the Hymenopbyllaceae are 
reduced and that the filmy character of the leaf is due to reduction. 
It is therefore hard to see why the conducting elements should be 
more numerous, for example in T. scandens than in T. reniforme, the 
leaf of which may be called filmy, since it has no intercellular 
spaces, although it is several cells thick. But, as Mr. Tansley 
points out, the thickness of the leaf is a point in favour of the primi¬ 
tiveness of T. reniforme (32). It would seem more natural to 
regard the forms with solid steles, such as T. scandens , as relatively 
primitive. Such a type is usually regarded as highly primitive and 
is so considered by Mr. Tansley in Tubicaulis and Grammatopteris. 
Here, as in those genera, the solid stele is associated with exarchy, 
and such an association strongly favours the primitiveness of 
exarchy. There are, however, other species with solid steles that 
are not exarch, as for example Trichomanes radicans. Mr. Boodle 
asserts that T. reniforme appears to be more primitive than T. 
radicans because it has sori on most of its veins. But this 
character is not necessarily a primitive one ; it might result from 
reduction of leaf-surface unaccompanied by decrease in the number 
of sporangia produced. Such reduction may well have occurred, 
since the leaf of this species is exceptional in being simple. Mr. 
Boodle accounts for the origin of the exarch stele by supposing 
that definite protoxylem having been lost, owing possibly to slow 
growth in length, a fresh protoxylem may have reappeared under 
other and more rapid conditions of growth in a different position. 
The difficulty that a change in the position of the protoxylem has 
occurred within the Hymenophyllaceae must be faced; it would 
seem to be best met by Mr. Boodle’s hypothesis. But this change 
of position of the protoxylem may equally well have resulted in the 
evolution of endarch types from exarch forms. This is what the 
present writer believes has occurred, not only in the Hymeno¬ 
phyllaceae, but also in such of the Botryopterideae as were really 
endarch. Mr. Tansley argues that the primitiveness of endarchyin 
the Hymenophyllaceae is supported by its “ apparently general 
