36 Bower. — Studies in the Phytogeny of the Filicales. VI. 
It has been shown how N eocheiropteris has a thick rhizome, covered by 
ciliated scales, and containing a highly perforated solenostele, with divided 
leaf-trace and gritty sclerotic nests. The stalked pedatifid leaf is a modifi- 
cation of the Dipterid type, but related in point of its simplified helicoid 
development to Leptochihis tricuspis and with similar venation. The sori, 
seated often on irregular vascular loops, are elongated parallel to the costa, 
forming at the base an almost continuous elongated soral area (Plate I, 
Fig. 5). The sporangia resemble those of Platy cerium and L. tricuspis , while 
all of these have smooth bilateral spores. Remembering the original refer- 
ence of this Fern by Christ to a Dipterid affinity, it is not difficult to see 
in it another example of that alliance, showing a spread of the sorus over 
the leaf-surface. 
It is in relation to these Ferns that the plant designated in Christensen ’s 
Index as Leptochilus tricuspis (Hk.), C. Chr., finds its natural place. It 
shares with Neocheiropteris the rhizomatous habit, long naked leaf-stalk, 
protecting scales, highly divided leaf-trace, and perforated stele, with 
gritty sclerotic nests. The lamina is referable, as has been seen, to the same 
type of branching. The vascular system of the upper leaf is particularly 
distinctive (Text-fig. 2, p. 6) : the modification of the horseshoe as it passes 
upwards resembles that of Platycerium , taking, as we have seen, a middle 
position between that genus as the extreme type and Cheiropleuria. The 
venation is essentially similar in them all. Further, in the fertile region it has 
been seen that Cheiropleuria has occasional slight extensions of the recep- 
tacular strands beyond the single areola : this is the first step towards a 
‘diplodesmic ’ state. L. tricuspis has a well-developed diplodesmic system, 
but not so complete or extensive as that well known in Platycerium. Hitherto 
these are the only three genera which show this condition. It has not been 
possible to make a complete examination of Neocheiropteris on this point. 
Finally the sporangia and spores of Neocheiropteris , L. tricuspis , and 
Platycerium are very similar. Even the branched soral hairs are present, 
though those of Platycerium are more elaborate, in accordance no doubt 
with the epiphytic habit. It is concluded that Neocheiropteris , L . tricuspis , 
and Platycerium are phyletically related Ferns, while Cheiropleuria is also 
related, but rather more aloof ; and that all of them are Dipterid derivatives . 
On the other hand, Leptochilus tricuspis differs from all investigated 
species of Leptochilus in the fact that in them no diplodesmic structure has 
been found, while they show also a less advanced perforation of the soleno- 
stele and absence of the sclerotic nests. These and other distinctive 
features are of sufficient importance to justify its separation from the genus, 
which may be effected without any introduction of a new name. It was 
named by Hooker AcrosticJnmi ( Gymnopteris ) tricuspe , Hook. Beddome 
later named it Gymnopteris tricuspis , Hook. 1 Meanwhile the genus 
1 Ferns of British India, Text to Plate LIII. 
