64 CONSPECTUS TABULARUM. 
Drupe 4-5 lines diameter, in my specimens crowned with the persisten 
tube of the perianth: seeds spherical, embryo minute, in the base of the 
albumen. 
I follow A. De Candolle in separating this plant generically ‘from 
Thesium, from which it chiefly differs in its succulent fruit. I do not, 
however, find the tube of the perianth deciduous, as stated by De Can- 
dolle, and in my numerous specimens the flowers are more commonly 
solitary than ternate. The 5 callosities at the base of the perianth tube 
are not clearly shown in the figure, and the two lateral nerves of the — 
leaves have been omitted. 
Fig. 1, a flowering branch; the natural size. ‘Fig. 2, perianth; 8, the same laid 
open, and the inferior ovary vertically cut; 4, cordlike placenta, and its 4 ovules; 5 
’ 
fruit, crowned with the perianth ; 6, section of seed ; 7, embryo; all magnified. 
200. CYCLOSTEMON NATALEN SE, Harv. (Euphorbiacee.) 
C. Natalense; ramulis petiolisque minutissime puberulis, foliis 
breve petiolatis utrinque glaberrimis supra nitentibus oblongis v. ellip- 
ticis subacutis basi obtusis argute serratis. 
Has.—Near D’Urban, Natal, J. 
plant), 862 (female). (Herb. T. C, 
Descr.—A dicecious shrub, or small tree, glabrous, except on the 
young twigs and petioles, which are minutely pubescent. Leaves alter- 
nate, 24 to 5-6 inches long, 14-21 inches wide, glossy above, veiny 
beneath, varying from elliptical to oblong, obtuse at base, more or less 
acute at apex, sharply serrate. Flowers of both sexes fascicled, spring- 
ing either from the larger branches or the trunk. Male flowers on 
slender pedicels, many in the tuft; perianth 4_5 parted, its segments 
ovate, very convex, unequal. Female flowers subsessile, few together; 
perianth of 5 unequal, very convex pieces ; no stamens ; ovary depressed, 
densely velvety, sitting in a fleshy disc, thick-walled, crowned by 8 
divergent styles, with capitate stigmas, 6-celled; ovules solitary. Im- 
mature fruit, apparently Jleshy, velvety like the ovary, subtended by 
the persistent perianth. 
Blume founded his genus Cyclostemon on certain Javanese trees, to 
which Thwaites has since added a Cingalese species, I have now the 
pleasure to figure one from South Africa, agreeing in general habit and 
characters, but in which I find (perhaps erroneously) only a single ovule 
in each cell of the ovary ; in the other species there are two. Mr. San- 
derson informs me that the yellow flowers are very fetid and offensive; 
they spring from knots on the older branches and the stem! 
a gira a see side of a leaf; 3, tuft of male flowers; the natural size. 
A tamen ; 6, q flower ; all magni i immature 4 
fruit; and 8, cross section of the same ; the naeour size, denial sien 
— ! 105; Gerr. & M‘Ken! 724 (the male 
D 
