8 
Tansley and Lulham. 
A careful comparison of the figures illustrating the structure 
of the node of Pteris incisa, var. integrifolia with Figs. 1 and 2, 
which .are taken from Gwynne-Vaughan’s paper on Solenostelie 
Ferns (’03), and represent the vascular structure of the nodes of 
Hypolepis tenuifolia and H. repens respectively, will facilitate the 
interpretation of the structure under consideration in terms of the 
morphologically simpler case. 
The leaf-trace of Hypolepis tenuifolia (Fig. 1) will be seen to 
come off from the lateral surface of the solenostele of the rhizome 
as a curved and folded plate, with its concavity facing the median 
dorsiventral plane of the rhizome. The stele opens dorsally at first, 
but the leaf-gap becomes lateral as soon as the leaf-trace is free. 
As the trace departs it becomes deeply infolded on the basi- 
scopic side, and the infolding so formed soon becomes separated off 
as a gutter-shaped strand (/. sh.) which closes to a cylinder and is 
the stele of the lateral shoot arising on the posterior side of the 
petiole base. Just after the gutter-shaped stele becomes free, a 
small strand (a) runs from its free edge furthest from the leaf-trace 
to join the basiscopic edge of the latter. Now if we imagine the 
stele of the main rhizome to decrease considerably in size, so that 
its continuation and the lateral branch arising from the base of the 
petiole become of about equal importance, we should have a 
•“false” dichotomy of the stem, with the petiole arising in the angle 
of the fork, exactly as in the two nodes of Pteris incisa var. integri¬ 
folia. The parallel is even closer, though not perhaps so legitimate, 
between the node of the latter and the petiole-base with its two 
lateral shoots in Hypolepis repens (Fig. 2) if we imagine the 
continuation of the main rhizome to be suppressed altogether. The 
general resemblance between this and Fig. 3 will be obvious at once 
if we consider the form of the leaf-trace in cross section just below 
the separation of the lateral shoot-steles ( i.e . along the line d — d in 
Fig. 2 corresponding with Fig. 8 in P. incisa var. integrifolia). 
Here the trace is in both cases complicated by the two deep dorso¬ 
lateral gutter-shaped infoldings most of which form the two branch- 
steles. In P. incisa var. integrifolia , however, the branch-stele is 
nipped off as a cylinder from the outer part of the infolding, i.e. from 
the end of the dorsal limb of the H, by the union of the dorsal and 
ventral walls of this limb at z (Fig. 10) so that no gap occurs either 
in the branch-stele or in the adjacent part of the trace, while in 
Hypolepis the stele separates as a gutter-shaped strand whose outer 
edge is united with the edge of the trace by means of the strand a 
