22 Bower . — Studies in the Phytogeny of the Filicales. VI. 
If the vascular system thus observed in Leptochilus nicotianaefolius be 
described in the usual terms, it would be a dictyostele, with leaf-gaps greatly 
elongated downwards, so as to overlap. The leaf- trace is much divided 
(‘ perforated ’), and the separate strands are inserted on the margin of the 
gap. The median strands spring somewhat irregularly from the very narrow 
lower end of the leaf-gap, while the larger marginal strands are given off 
near to the upper and wider end of it. The leaf-gaps alternate right and 
left. Thus the vascular skeleton resembles the type shown in any soleno- 
stelic Fern with overlapping leaf-gaps ; for instance, as shown in Pellaea 
rotundifolia by Gwynne-Vaughan , 1 but with this difference: that whereas 
in such a case the leaf-trace is an undivided horseshoe, here it is broken up 
into a number of strands. Clearly it is a type not far removed from 
solenostely, as shown by the gutter-shaped lower vascular tract. All that 
differentiates it from a type like Metaxya or Dipteris is the slight overlapping 
of the elongated leaf-gaps, and the subdivision of the leaf-trace by numerous 
perforations. This subdivision resembles that in Dicksonia ( Cibotium ) 
Barometz , as shown by Gwynne-Vaughan ; 2 but here it is continued down 
to the very base of the leaf. 
Other species of Leptochilus show a similar structure, though varying 
in the proportion and number of the strands and the overlapping of the leaf- 
gaps . 3 But the existence of the smaller (upper) and larger (lower) meristeles 
in the axis is the same. This has been seen in such smaller species as 
L. zeylanicus (Houtt), C. Chr., L. alienus (Sw.), C. Chr., and L. heteroclitus 
(Pr.), C. Chr., and it may be held as probably characteristic for the genus. 
In general their vascular condition differs from that seen in L. tricuspis in 
showing a less advanced state. But this may find its natural explanation 
in their smaller size. It will not be necessary to enter into their details. 
On the other hand, L . cuspidatus (Pr.), C. Chr., which has a much 
more massive rhizome, shows the lower strap large and continuous, and it is 
thus reminiscent of the primitive solenostele. Its leaves are borne on 
the massive creeping rhizome, not in two simple lateral rows but upon 
a more complex and seemingly less regular plan, so that some leaves 
are seated obliquely on the upper surface. This makes the vascular 
arrangements more complex than in some of the smaller species. In 
the case where two lateral leaf-bases are traversed — which is the more 
frequent — the structure is as in Text-fig. io, a , where there is substantial 
agreement with what has been seen in L. nicotianaefolius'. there is a small 
upper strap, with a much larger lower strap, forming almost half of 
a solenostele, from which root-strands arise. Right and left are the leaf- 
1 Solenostelic Ferns, II. Ann. of Bot., vol. xvii, PI. XXIII, Fig. 8. 
2 See Land Flora, Fig. 331. 
3 Leptochilus axillaris (Cav.), Klf., has been examined by Mettenius (Ueber Angiopteris , p. 554) : 
he found only two strands of the leaf-trace, while the stele was found to be much perforated, with 
elongated meshes, but a continuous upper strap clearly marked (Pl. IX, Fig. 3). 
