low, the edge crenulate and revolute all round, nerves about 
ten, radiating, the larger branched, the smaller subsimple. 
Petioles of Very unequal length, round, resembling the 
stem. Stipules intra-foliaceous, large, ovate, acute, keeled, 
marcescent. Peduncle (ten inches long) terminal, becom- 
ing axillary, tapered, similar to the stem but without spots, 
repeatedly dichotomous at the apex (primary branches 
about one inch long, the others gradually shorter). Bracteas 
lanceolato-elliptical at each subdivision. Flowers white, 
rather small. Male flowers in the clefts of the cyme, or on 
the inner side, where the ultimate branches are reduced to 
two flowers (the outer being a female), or, occasionally, 
solitary on the ultimate branches. Petals four, the two 
outer subrotund, slightly crenate, the two inner linear- 
elliptical, very narrow, entire. Stamens numerous ; /fila- 
ments free, excepting at the base, where they are mono- 
delphous ; anthers spathulate. Female flowers with five 
subequal petals, expanding later than the earlier of the 
males. Stigmas pale-rose coloured. Germen with three 
subequal wings. 
I was at some pains to ascertain the relative position of 
the male and female flowers when only these two were 
found at the extremity of the ultimate branch. It seems 
to me that the normal form is the conversion of the last 
dichotomous ramification of the cyme into the pedicels of 
two female flowers, and that the male flower here, as else- 
where, is placed in the cleft; the loss of the inner female 
flower being an illustration of the opinion, that internal 
parts, from pressure, more frequently abort than those 
which are external. As the common support of these two 
flowers generally turns half round on its axis, their true 
position may not be obvious unless examined when they 
are very young. Graham. 
—_ 
