Grietum , with some Notes on the Structure of Inflorescence. 6 1 1 
■{.c 
top of the node ; the descending bundle (of the inner series) arises from it 
close to its point of origin, and the outer flower-trace is given off on its way 
down to the leaf. This is probably to be regarded as a case in which the 
main trunk of the leaf-trace has remained adnate to the ring-bundle through 
the length of the node. And it is not improbable that in the more normal 
case shown in Text-fig. 3, the bundles of the inner series really arise from 
the leaf-traces or from the ring-bundles in the neighbourhood of the point 
of departure of the leaf-traces, and are adnate to the latter until they reach 
the level of the top of the node or some higher point in the lower half of 
the next internode. 
The inner series of bundles 
now described for the male spikes 
of G. africanum and G. Buch- 
holzianum are without doubt to 
be regarded as the homologues 
of the inner series of the female 
spike of G, africanum } which 
they resemble in their origin, their 
course, and their orientation, and 
also of the inner series of the 
female spike, 2 and probably the 
single series in the male spike of 
G. scandens (Text-fig. 2), which 
are orientated in the same manner. 
The ‘ augmenting ’ bundles which 
appear to be a constant feature 
of the more advanced female spike 
in G. scandens and G. africanwn 3 
are not found in the male spike 
of these species nor of G. Buck - 
holzianum. 
The course of these ascending, later descending, bundles of the inner 
series in the male and female spikes of G. africanum and in the male of 
G. Buchholzianum bears a resemblance to that of the sporangiophore trace 
in Palaeostachya vera . 4 This arises from the ring immediately above the 
bract-bundle, 5 and it ‘ does not separate from the main bundle until it has 
ascended through almost exactly half the internode *. It then becomes 
reflexed as in the African Gneta . In the case of Palaeostachya there are 
grounds for the opinion that the course of the sporangiophore trace is to 
some extent determined by the distribution of sclerized tissue. In Gnetum 
Text-fig. 4. Longitudinal section through a node 
and parts of two internodes of G. africanum (an excep- 
tional case). Lettering as in preceding figures, x 15. 
1 1 . c., Text-fig. 3, Biv. 
3 1 . c., Text-fig. 2, Miv. ; Text-fig. 3, om, lom. 
4 Hickling (’ 07 ). 
2 1 . c., Text-fig. 2, biv. 
5 1 . c., Text-fig. 1. 
