Morphology of the Female Flower of Gnetum. 
;75 
I'axe floral, mais est tres probablemexit de nature foliare/' * This type is 
□low most nearly represented by the male flower of Welwitgchia.; AH the$e 
whorls are present only in the male flower of Welwitschia ; in otl^er cases 
«one or more are entirely missing, or their former presence is indicated only 
by peculiarities of the vascular system. The authors' comparison * is here 
reduced to tabular form (Table 3). 
The comparisons contained in column 4 of this table have been recently 
'discussed elsewhere.f Those of columns 6 and 7, except as regards whorl 1, 
have now been modified by the authors. :{: It should be noted, however, that 
the androecium (whorl 2) of the hypothetical type still appears in three 
xlifferent forms — viz. an androecium, a winged envelope and an envelope 
without wings (Ephedra $ ) — the two extremes of these occurring in the 
same species. This in itself would be remarkable ; but evidence that will 
either establish or overthrow the suggested homology is lacking. It has 
been stated more than once that evidence in support of this same compari- 
son was advanced by Bertrand.§ This is not strictly correct, for he did not 
<listinguish the male from the female flowers. In his view the staminal 
tube of the male flower and the winged envelope of the female were two 
stages in the ontogeny of one and the same structure. || He says, for 
•example, that the winged envelope " resulte de la transformation " of the 
staminal tube. This transformation is thus described : ^' Sitot apres la 
pollinisation, les antheres tombent, I'orifice du tube staminal se retrecit, et 
«es bords lateraux s'etendent pour former deux grandes ailes marginales." 
Bertrand's contribution to the discussion is therefore limited to emphasising 
the fact that these two envelopes possess common characters, by regarding 
them as identical.^ 
In this connection it may also be noted that MM. Lignier and Tison 
imply that Bertrand considered this winged envelope to be an ovary.** On 
the contrary, he combats this view, previously propounded by Macnab,tt 
* Idem, 1911, A, p. 1. . ' =: 
t Pearson, 1915, C. , ' . ■{ 
X See Table 1. 
§ Pearson, 1909, p. 336 ; Lignier and Tison, 1912, p. 67. 
il Vide Bertrand, 1878, pp. 63, 82, 86, 89 ; pi. 9, figs. 7, 8, 9. 
IT That Bertrand should have so misinterpreted the relations of the Welwitschia 
:flowers is remarkable, for, although he does not cite Hooker (1863), he refers to 
Strasburger (1872) and to Macnab (1873). With regard to the latter he refuses to 
-admit the accuracy of his figures and says (p. 65) : " Entre les fleurs hermaphrodites 
-et les fleurs femelles, je n'ai trouve de differences que dans Tabsence des antheres au 
tube staminal de la fleur femelle." It is probable that his material was scanty. His 
-earlier account of the anatomy of Welwitschia was based upon the study of a single 
•dried specimen received from Decaisne (Bertrand, 1874, p. 8). 
** Lignier and Tison, Zoc. ci^. ; .^^ 
ft Macnab, 1873. •- ; . 
