164 
A NOTE ON THE INFLORESCENCE AND 
The antherophore is variously regarded as a cauline or a foliar structure 
—in the latter case it is comparatively unimportant whether it represents 
one filament or more. Characters which appear to be pertinent in this 
connection are: (1) it is entirely intercalary in origin; (2) it is usually 
traversed by bundles equal in number to the anthers 1 , i.e. two ; (3) in 
Ephedra the antherophore is frequently two-lobed and in E. distachya the 
lobes separate near the base 2 ; (4) in E. fragilis the antherophore exhibits a 
circinnate vernation 3 and in several species of Ephedra it possesses a broad 
laminate form 3 . In Ephedra it is therefore probable that the antherophore 
is composed of sporophylls more or less concrescent 4 and, if this is true for 
Ephedra, there is little reason to doubt that it is equally true for Gnetum. 
If so the only structure which can represent the floral axis is the minute 
pedestal which, in the adult flower, supports the antherophore and the envelope. 
The flowers of Gnetum are not clearly axillary if, as is fairly certain, the 
cupule is composed of two concrescent leaf-structures. To regard them as 
axillary in position is to assume that the meristematic ring is derived from two 
opposite axillary primordia. The only evidence by which such a hypothesis 
can be supported is the fact that solitary and apparently axillary flowers 
occur in Ephedra and Welwitschia — evidence whose value as applied to 
Gnetum is very doubtful. 
If the envelope is inserted upon the secondary (floral) axis, it may 
be anticipated that the axis will precede the envelope in ontogeny. If 
further the envelope consists of two concrescent leaves, some indications 
of such a structure may be expected. Such indications would be: (1) the two 
leaves would probably arise simultaneously from opposite primordia — opposite 
leaves being so regular a character of Gnetum, as of the other members 
of the group ; in this case it is probable that at later stages of growth the 
envelope will be more or less lobed, the lobes standing vertically over the 
primordia. Or if (2) the leaves are concrescent from the beginning they may 
arise as an equal collar, whose insertion will be likely to be in a plane per- 
pendicular to the secondary axis and, at least in the early stages of its 
growth, the structure as a whole will be radially symmetrical. (3) Vascular 
bundles will be normally present, probably one along each line corresponding 
to the midribs of the concrescent leaves ; and they will arise by branching 
from the bundle system of the secondar}^ (floral) axis; if bundles are absent, 
their absence must be explained by reduction. 
The envelope arises as a collar. The upper (anterior) part of it precedes 
1 This is not without exception. More than one case has been seen in which an antherophore 
bearing two anthers was traversed by three bundles (Pig. 2B). A case in which four anthers 
were present has been described (Pearson, 1915) but the bundles were not yet differentiated. 
2 Stapf, 1889, p. 21 ; Thodav and Berridge, 1912, p. 971. 
3 Thoday and Berridge, l.c. p. 970. 
4 L.c. p. 974. 
