ZANICHELLIA. 193 
and the leaf, to which the flower seems axillary, it may be 
considered that the flowers are really terminal, and that were 
the axis prolonged, each prolongation would be a branch axil- 
lary to each leaf. 
Again in some cases where the scale is absent from either 
flower, and intervening between one of the leaves, which are 
almost invariably present (one to each flower) and the deve- 
loped continuation of the axis, such continuation is to be 
considered as lateral, and belonging to the leaf, between 
which and the axis the scale intervenes. And in such a case 
the female flower is really terminal, that is, terminates, the 
axis most to the centre, the male flower somewhat less so, 
and the bud from the remaining leaf or that next the male 
flower is more or less developed, entirely so when the scale 
is absent. 
The flowers are therefore terminal however much to the 
contrary they may seem, for in these dichotomous plants, as 
many forks should occur, as there are leaves, at or about one 
node ; of the setz, I know nothing more than I do of those of 
genuine Aroideze. 
What bears me out in the above, is, that careful ехатійа. 
tion will generally disclose a young bud, obviously belonging 
to the scale which intervenes between it and its leaf, and 
that this scale when of auy development, will be found to en- 
close at the extreme base, the actual elongation of the axis. 
This latter observation may however be wrong, because the 
vagina belongs to this actual elongation, especially as the eon- 
nection between the scales and the leaves is evident. 
Supplementary sketches Pl. CCLVI. Fig. I. 
1, represents a node and the inflorescence; at a the vagina 
laid open; аё 6 the primary leaf, from the axilla of which 
the convolute bud-scale, or vagina is evolved, which again 
belongs to the actual elongation of the axis; at c is this ac- 
tual elongation of axis; at d a supplementary leaf; ate its 
scale; at f its bud; at ^ the stamen; at i the second supple- 
2 
