186 EDIBLE FRUITS OF THE RIO NEGRO. 
horse-bean ; testa thinnish, soft, reticulated ; albumen none; cotyledons 
contortissimo-plicate; radicle next hilum. 
The pulp of this is very pleasant eating, being sweeter than that of 
the Cupá-acá, of which I sent flowering specimens to Mr. Bentham 
from Santarem under n. 954, Genipa macrophylla. The latter is an 
evident Rubiacea, with opposite leaves and interpetiolary stipules, and 
seems closely allied to the Genipépa; but its fruit is so exceedingly 
like that of the Cupua-í externally, that I am not surprised an Indian 
should class them together. 
6. Cocira, Ling. Ger., (distributed as Pourouma retusa, Spruce.)— 
(Nat. Ord. Artocarpee.) Coll. n. 1219. 
Has. Barra; dense forest, in moist situations. Fruit ripe in Ja- 
nuary. 
Tree of 50 feet by 12—18 inches, supported on a cone of half-emersed 
compressed roots, obscurely ringed like a Cecropia, rings 3—2 inches 
apart; central pith very slender; bark smooth, often tinged with light 
purple; ramuli pubescent. Leaves near apex of ramuli, on long floc- 
cose petioles; lamina 9 inches by 6 inches (on young trees much larger), 
ovate subobtuse, very obtuse at base; margins undulate; veins quite 
straight, leaving the midrib at an angle of 45?, and running almost to 
the very margin ; across these extend numerous pellucid secondary veins 
nearly at a right-angle, and between the last are minute reticulations ; 
under side of leaf pubescent, silvery, between the veins plicate. (and 
so folded in vernation). Stipules linear-oblong, apiculate, bicostate, 
thickly pubescent without, silky within, falling with the evolution of 
the leaf.. Cymes terminal, contracted, about five together. Pedicels 
short, very thick. Fruit as large as a small grape, drupaceous (a capsule 
included in a thickish succulent integument), ovate, pubescent, the skin 
and pulp purplish-black, one-celled, one-seeded. Pericarp thin, hard, 
brittle. Seed deep purple; albumen 0; cotyledons amygdaloid, conti- 
guous by a narrow face; radicle opposite the hilum. 1 
This does not agree well with any genus described in Endlicher; it 
is nevertheless perhaps Pourouma, Aubl. A large Cecropiaceous tree, 
with trifid leaves, which often grows along with it, is probably also a 
Pourouma; but of the latter I have not yet seen the flower or fruit.— 
The Cocura is a very handsome tree, especially when profusely laden 
with its clusters of black fruit, which resemble so many small bunches 
of grapes; nor does the taste belie the look, for I have not eaten any 
