158 
A NOTE ON THE INFLORESCENCE AND 
It may then be considered as probable that the functionally bisexual 
inflorescence from which the unisexual forms now found are derived, bore 
few male flowers at each node, perhaps only a single ring. In this 
connection the course of the vascular trunks which supply the flowers in 
G. Gnemon (Fig. 6) may be suggestive. The main trunk on separating 
from the leaf-trace passes upwards towards the female flower and gives 
off a branch opposite the oldest male flower. From this latter another 
branch passes downwards supplying successively the younger male flowers. 
This structure is at least reconcilable with the view that the two main 
branches, viz. those which supply the female flowers and the ring of male 
flowers innnediately beneath them, represent a relatively primitive vascular 
system from which the smaller branches supplying the lower male flowers are 
of later origin. In G. scandens (Fig. 5 A) the course of the vascular traces 
supplying the flowers is less constant 1 ; the form shewn in the figure is 
frequently seen. It perhaps reflects the decreased importance of the female 
flower compared with that of G. Gnemon. 
Each node is ensheathed in a short cupule consisting of a concrescent 
pair 2 of reduced leaves 3 . In the male inflorescence, the cupule is sometimes 
irregularly split by the pressure of the developing flowers (Fig. 7). While 
the cupules themselves may possibly appear in acropetal succession, this 
order of development does not extend to the products of successive axillary 
meristematic rings in the stage represented by the youngest male inflo- 
rescence available (Fig. 2). Here the median nodes are clearly in advance of 
those above and below them. In all the stages of the female inflorescence 
examined, as in the later stages of the male, the flowers at successive nodes 
appear to be approximately equal in development ; at least, no node is 
distinctly in advance of any other. In Ephedra on the other hand the 
primordia of the youngest pair of male flowers are fairly advanced in 
development before the succeeding pair of bracts ai’e yet visible on the 
surface of the vegetative cone of the axis 4 . The last stage of the inflorescence 
1 Cf. Pearson, 1912, text fig. 2. 
2 Karsten, 1893, p. 340 ; Pearson, 1912, p. 607. 
3 In this respect Gnetum resembles Ephedra (Stapf, 1889, pp. 20, 23) and the male cone of 
Welwitschia (cf. Hooker, 1863, p. 21). With regard to the female cone of Welwitschia (Hooker, 
l.c. p. 24, PI. vm, fig. 29), Church (1914, p. 118) is in error in stating that neither the “sterile 
scales at the base of the (ovulate) cone, nor the fertile bracts are connate below.” In every cone 
examined by the present writer, the condition is exactly as described and figured by Hooker (l.c.). 
In this connection, reference may be made to a feature of the ovulate cone which does not 
appear to have been noticed previously. It is well known that the lower 6 — 8 pairs of bracts of 
the ovulate cone are sterile (Pearson, 1909, p. 333; Church, l.c.). It now appears that between 
the seed-bearing portion of the cone and the succession of empty scales below, there are from 2—4 
pairs bearing in their axils female flowers which do not develop beyond the point at whidh they 
are little more than just visible to the naked eye. (Cf. Goebel, 1905, ii, p. 511.) 
4 Strasburger, 1872, p. 132 ; Taf. xiv, figs. 2, 3. 
