Gnetum africctnum, with Notes on Gnetum scandens. 1117 
is withered ; it only projects a little distance beyond the outer integument 
and does not overtop the outermost covering as it does in the younger 
stages . 1 
a In my oldest ovule of G. africanmn the innermost integument is com¬ 
posed almost entirely of parenchymatous cells, though in the micropylar 
tube (Fig. 11, PL LXXXVI) a few short cells with lignified walls are 
scattered ; and it is probable that these would increase in number later, as 
they do in G. scandens. 
A ring of numerous well-developed and branching vascular bundles 
traverse the common base of the integument and nucellus, running, however, 
distinctly in the region of the integument; as a rule the principal part of the 
bundles terminates just at the level of freedom of the integument, but one or 
more spirally thickened tracheides have been in several cases traced quite 
into the base of the free portion of the integument (Fig. 10, PI. LXXXVI), 
and in most cases a few elongated cells with dense contents, such as accom¬ 
pany the xylem elements throughout the course of the bundles, continue 
past their termination for some little distance into the integument. 2 The 
integumentary nature of the vascular strands 3 is still more clearly seen in 
the younger ovules (see photograph shown in Fig. 13, PL LXXXVII), where 
the procambial strands extend right into the base of the integument. 
The micropylar tube is closed at this stage 4 for a short distance 
opposite the tip of the outer integument. Above this level the cavity can 
still be distinguished in the withered tip ; the cells lining the cavity are 
cuticularized. Below this level the cavity again appears and is at first slit¬ 
like, thus imparting to the seed a small amount of bilateral symmetry. 
A projecting flange 5 extends downwards from the closed part of the micro¬ 
pylar tube and is wrapped over the tip of the outer integument, locking the 
two firmly together. 
This method of closing the micropyle, of which details will be given 
later, resembles that recently described by Miss Berridge in G. Gnemon. 
The tip of the inner integument appears to be withered lower down in my 
species, and all the structures at the apex are less conical and pointed 
than in G. Gnemon (cf. Text-fig. 9, p. 1113, with Berridge, Fig. 1, p. 140). 
In G. africanmn the withering of the micropylar tube seems very rapid ; in 
most of my smallest ovules the micropylar tube had already withered 
below the level of the outermost covering (Fig. 14. PL LXXXVII). 
1 Griffith, p. 302, Taf. LV, Fig. 18, shows the long tube of the inner integument before it withers ; 
also Lotsy, 1889, Fig. 18, PL III. 
2 See also Lignier, 1911, p. 3. 
3 Miss Berridge tells me that isolated tracheides occur in the inner integument of Ephedra. 
4 Griffith, p. 302 ; Karsten, p. 200 ; Berridge, 19ti. 
5 Described by Karsten in G. Rumphianum and G. ovalifolium. Cohn’s Beitrage, 1893, p. 377 ; 
Thoday, 1910; Berridge, 1911. 
