EVOLUTION OF THE INFLORESCENCE. 523 
female sporophylls above the male. In this case it may be assumed that in 
the Pteridospermous ancestor a crop of male sporophylls was immediately 
followed by one of female sporophylls, without intervening foliage ; on the 
axis ceasing to lengthen, there would be formed a bisexual strobilus of 
a special character, to which the name anthostrobilus * has been given, as it 
foreshadows the angiospermous hermaphrodite flower, in which the carpels 
are always situated above the stamens on the floral axis. 
This peculiarity in common is one of the facts, among several others, which 
make the origin of the Angiosperms from a stock near the Bennittitales 
distinetly plausible. Hence an examination of the manner in which the 
anthostrobilus of this group is borne becomes of interest. The best-known 
members of this fossil group, however, do not afford evidence for the 
primitiveness of the solitary terminal flower, since the strobili are borne 
in an axillary fashion. It may be suggested that Bennittites is much 
modified in its vegetative parts, and that its sessile strobili, occurring 
laterally on a short stout stem, have been derived from ones borne in a 
lax manner on slender stems. Williamsonia and the recently-described 
genus, Mielandiella, are cases in point. They had comparatively thin 
branching stems, upon which the cones appeared in a terminal position. 
Hence it would seem highly probable that the cones of the Bennittitales 
were primitively borne terminally, and that the lateral position has arisen 
through a great change in vegetative habit. 
The Gnetales deserve a passing remark. Here the flowers or strobili, 
much reduced, are by no means solitary, but clustered together in complex 
inflorescences. Recent research Í suggests, in addition to Angiospermous, 
Bennittitean affinities as well. The hypothetical*group, the Hemiangio- 
sperms i, introduced to contain those supposed plants related to the 
Bennittitales from which the Angiosperms are imagined to have arisen, 
has been considered to include the Gnetales§. On this assumption, the 
Gnetales, in regard to their flower-clusters and to the reduction in the 
individual flower itself, eoupled with diclinism, may be looked upon as 
bearing a somewhat similar connection to the Hemiangiosperms as the 
Amentaceous families do to the Angiosperms as a whole. They are, as 
it were, the Amentiferw of the Hemiangiospermex, and their complicated 
* Arber & Parkin, “On the Origin of Angiosperms," Journ. Linn. Soc., Bot. xxxviii. 
(1907) p. 37. 
+ Sykes, M. G., “The Anatomy and Morpbology of the Leaves and Inflorescences of 
Welwitschia mirabilis,” Phil. Trans. Roy. Soc. London, 1910; Thoday (Sykes) & Berridge, 
“The Anatomy and Morphology of the Inflorescences and Flowers of Ephedra,” Ann. Bot. 
xxvi. 1912, p. 953. I 
r Arber & Parkin, loc. cit. p. 62. 
§ Arber & Parkin, “The Relationship of the Angiosperms to the Gnetales," Annals of 
Botany, xxii. (1908) p. 489. 
