so Transactions of the Royal Society of South Africa. 
Van Tieghem later modified his opinion, and his ovuliferous scale became a 
bicarpellary ovary* without style or stigma. 
In 1903, Lignierf adopted a modification of Van Tieghem's earlier 
hypothesis. The second envelope (second and third in the complete ? of 
Gnetum) is a whorl of concrescent bracts. An ovuliferous scale is axillary 
to one of these bracts ; the unitegumented ovule is inserted on the summit 
of this scale. The scale, however, is so reduced that there is no visible sign 
of its presence, and the unitegumented ovule therefore appears to be a direct 
prolongation of the axis. This hypothesis rests entirely upon a supposed 
analogy with Taxus,X and there is no direct evidence in support of it. It 
has since been abandoned by its author in favour of the view discussed 
above, viz. that the innermost envelope is a pluricarpellary ovary. 
Other authors retain an open mind with regard to the cauline or foliar 
origin of the ovule. Mrs. Thoday states that the female flower of Wel- 
witschia consists of a " short axis terminated by a single ovule " which is 
enclosed in two envelopes. § Later, in the same paper, it is suggested that 
although the ovules in the male cone and the stamens and aborted ovules 
in the male cone are now in an axillary position, they may not originally 
have been cauline." || The meaning of this suggestion as regards the female 
flower is that it " represents a sorus which was originally borne on the 
bract" ^; the male flower presents greater difficulty, but it appears 
possible " that it also is " derived from a form in which the sporophyll bore 
several sori." The author states quite distinctly that this is merely a 
suggestion based on an analogy with the Cycads, just as Van Tieghem's and 
Lignier's earlier views rested upon analogy with the Coniferae. For none 
of the three is there any direct evidence. 
With regard to Ephedra and Welwitschia still another hypothesis, having 
.a similar end in view, has been advanced. It suggested that the single ovule 
^'now differentiated direct from the plastic apex of the axillary bud" is "the 
■equivalent of more than one ovule, each originally borne on a foliar organ, 
but now fused together at the apex of an axis"** — in other words, that the 
apical portion of the ovule-bearing shoot is composed of concrescent leaf- 
structures. This is an interesting speculation suggested by the occasional 
indications of the fusion of two ovules in Ephedra altissima,ff a phenomenon 
which is most simply explained on physiological grounds. 
* Van Tieghem, 1891 {fide Lignier, 1903). 
t Lignier, 1903. 
X Loc. cit., p. 67. 
§ Thoday, 1910, p. 194. 
II Loc. cit., p. 221. 
H Loc. cit., p. 216. 
»* Thoday and Berridge, 1912, p. 979. 
ft Loc. cit., p. 962. 
