Morphology of the Female Flower of Gnetum. 
77 
more strongly reduced. In both cases the ovary contains a single naked 
ovule which is " probably of foliar origin."* 
MM. Lignier and Tison's opinion that the structure of the inner envelope 
in the three cases considered alwve is most naturally explained on the 
assumption that it has undergone strong reduction may be correct. But 
even if we go so far as to assume that in a more primitive state each had a 
strongly developed vascular system, they are not therefore carpellary. 
The utmost that such a fact could be held to prove would be that this 
envelope represents one or more leaf structures — a conclusion which many 
investigators have already adopted. But in view of the known structure 
of the ovular envelope in the Cycads and in a number of the lower 
Angiosperms,t the mere fact that it is vascularised is not sufficient to 
establish the conclusion that it is foliar ; still less does it prove it to be 
the homologue of an ovary ; and if it be foliar it is not therefore impos- 
sible that it may be still an integument. 
With regard to the external form of this envelope in the male flower of 
Welwitschia, Hooker, who regarded it as an integument, comments upon 
its resemblance to a carpel with style and stigma [j: and " speculates " on the 
possibility of its form being inherited from an existing or extinct race of 
plants in which " the office of the stigma of the carpellary leaf was per- 
formed by a stigmatic dilatation of the ovular coat itself." Strasburger, 
for whom it was at first § a carpel, later an integument, || also clearly 
regarded it as a stigma which had lost its stigmatic function. Its form and 
structure so obviously suggest such a function, and incidentally tend to 
strengthen the morphological comparison with an ovary, that it is well to 
remember that there is no direct evidence whatever that it has ever 
functioned as a stigma or that it is a reduced form of a functional stigma. 
There is, however, good reason to believe that it now performs a function 
in connection with the attraction of insect visitors.^ It is possibly an 
•excretory organ** ; it is certainly largely responsible for the conspicuous- 
* Lignier and Tison, 1912, p. 162. 
t Cf. Hemsley, 1906 ; Kershaw, 1909, A. B. C. ; Benson and Welsford, 1909. 
MM. Lignier and Tison's position with regard to these cases is not clear. If the ovule 
is reduced to a nucellus, as appears to be suggested {loc. ext., 1912, p. 161), then the 
first envelope (integument) of Juglans and Myrica is, by inference, carpellary. Prob- 
ably the authors do not intend to convey this impression, and, if so, it is not easy to 
see why the very reduced vascular system in the envelope of Welwitschia or Gnetum 
should mean so much more than the strongly developed system of Juglans or Myrica. 
X Hooker f 1863, p. 24. 
§ Strasburger, 1872, p. 153. 
1) Strasburger, 1879, p. 133. 
1 Strasburger, 1872, p. 272. 
*• Pearson, 1906, p. 274; Lignier and Tison, 1912, p. 122; but see Church, 1914, 
p. 135. 
