162 
A NOTE ON THE INFLORESCENCE AND 
A second specimen (female) from the same material shewed three distinct 
whorls below and above them six complete turns of a Ze/£-handed spiral. In 
other cases there are distinct whorls at base and apex separated by a spiral. 
In these, the tissues of the axis shew no indication of torsion. This peculiar 
development is probably due to the conditions under which the young 
inflorescence develops. All the cupules and nodal rings are laid down and 
the uppermost flowers derived from the latter are fairly advanced (Fig. 2) 
before the inflorescence (when axillary) emerges from the subtending bract. 
A slight change in the normal pressure conditions may therefore be presumed 
to exercise a powerful influence upon the adult form of the inflorescence. 
But until more information regarding the early stages of the inflorescence is 
available the phenomenon cannot be fully explained. 
Apart from the occasional development of such a spiral arrangement of 
lateral members, the whorl of two members is established in Gnetum as in 
Ephedra and Welwitschia. The only exceptions in the group are found in 
the disposition of the flowers of Gnetum and of the stamens of Welwitschia 1 . 
According to Church 1 there is no evidence that the whorl of six stamens in 
Welwitschia is derived from a more primitive pair, and the same is true for 
Gnetum. 
On the facts and inferences which have now been stated is based the 
suggestion that the primitive inflorescence of Gnetum consisted of a short 
axis bearing a terminal female flower or a group of female flowers of which 
one was terminal, surrounded at a lower level by a ring of male flowers ; that 
from this structure the existing multinodal inflorescence has arisen by the 
addition of new parts to the axis through the retention of vegetative 
characters by the axis of the terminal flower. 
The lateral female flower is regarded as a secondary axis bearing a 
terminal nucellus (“organised directly from the tip of the bud axis 2 ”) and 
two or three envelopes. Of these the innermost is undoubtedly an integu- 
ment; the outermost, it has been suggested above, is formed of two 
concrescent bracts homologous with those which form the cupule. The fact 
that the middle envelope contains vascular tissue would suggest that it is 
allied in origin to the outermost rather than to the innermost envelope, but 
the evidence is not conclusive. There is however no clear reason why either 
should be regarded as homologous with the angiosperm perianth 3 . 
The male flower has received less attention than the female. It is a 
widely accepted view that it is a “ proanthostrobilus.” Comparatively minor 
differences of opinion prevail as to the precise morphology of the antherophore 
and of the envelope of Ephedra or Gnetum. The antherophore is (1) a 
1 Church, 1914, p. 133. 
2 Coulter and Chamberlain, 1901, p. 122. 
3 Cf. Coulter and Chamberlain, 1910, p. 363. 
