Sponpras. TEREBINTHACEJE. 173 
ovule solitary in each cell, suspended from near the apex of the axis. Styles 
5, short and thick, distinct, rather distant at the base, conniving upwards. 
Drupe fleshy : nut 5- (or from abortion 1-3-) celled. Seeds solitary, pendu- 
lous. Embryo straight, inverted.— Trees. Leaves alternate, unequally pin- 
nated, without dots. Panicles axillary and terminal. Flowers whitish or 
ted. Fruit eatable. 
In S. lutea, mangifera, and dulcis, the ovules are certainly solitary and suspended, 
notwithstanding what Gertner thought he saw to the contrary: we have not seen S. 
purpurea. S. dulcis may perhaps form a distinct genus yino pa on account of 
the singular structure of the nut; there, from the axisnot being prolonged while the 
cells are enlarged, not only does the nut become lobed, but the formerly pendulous 
ovules seem at length to form ascending seeds, with an inferior hilum and radicle. 
Blume, trusting to Gertner’s accuracy as to the number and position of the ovules, 
has separated all the Indian species which he found to have solitary ovules, and 
placed them in Poupartia ; df amongst these S. dulcis, which he seems to consider 
as having a centrifugal radicle, while Garin spe it centripetal. 
533. (1) S. mangifera (Pers.:) petiole terete: leaflets 4-5-pairs, ovate or 
elliptic-oblong, abruptly acuminated, oblique at the base, quite entire, veiny 
and glabrous: panicle diffuse: drupe oval pee : nut oblong, woody, very 
hard, outwardly slightly fibrous —DC. prod. 2. p. 75; Roxb. fl. Ind. 2. p. 451 ; 
Wight ! cat. n. 534.—S. amara, Lam.; Spr. syst. 2. p. 439.—8. Amra, Ham. 
in Linn. Soc. Trans. 18. p. 531.—S. paniculata, Roxb. in E. I. C. mus. tab. 
50.—Mangifera pinnata, Koen.; Linn. f. suppl. p. 156 (not Lam.)—Poupartia 
Mangifera, Blume; G. Don in Mill. dict. 2. p. 79.—Rheed. Mal. 1. t. 50; 
Rumph. Amb. 1. t. 612 : 3 : 
The Condondum of Rumph. Amb. 1. t. 60, which Hamilton (in Wern. Soc. 
Tr. 5. p. 359) seems disposed to refer to Sorindeja, we are rather inclined to 
consider the same with S. dulcis, Forst. 
* 534. (2) S. acuminata (Roxb.:) petioles terete: leaflets 5-8 pair, nearly 
Opposite, long-oval, acuminated, remotely crenulated, shining.— Roa. fi. 
Ind. 2. p. 453. Malabar. 
With this we are quite unaequainted, and unfortunately Roxburgh does 
not furnish us with the details of the flower, ovary, or fruit, which, however, 
itis probable that he never saw. Supposing it not to be a Spondias, it may 
Prove to be our Solenocarpus Indica. 
Suborder 3. BunsEREX ( Kunth.) Flowers usually bisexual. Calyx 
Persistent, somewhat regular, 2-5-divided. Petals 3-5, equal : æstiva- 
tion usually valvular. Stamens two or four times as many as petals, 
distinct, perigynous. Torus orbieular. Ovarium 2—5-celled, superior, 
Sessile: ovules in pairs, collateral, suspended: style one or none: 
stigma simple or lobed. Fruit rarely a hard capsule ; usually drupa- 
ceous, 2-5-celled, its outer portion or sarcocarp often splitting into 
valves. Seed solitary. Albumen none. Radicle straight, superior, 
next the hilum: cotyledons fleshy, or wrinkled and plaited.— Trees or 
shrubs abounding in balsam, gum, or resin. Leaves alternate, usually 
not dotted, generally with stipules. 
Rumphia of Linnewus, apparently characterised solely from Rheede’s figure and 
description (Mal. 4. t. 11), geh d referred to the Terebinthacem, cle » syne 
affinity with this suborder than with the others, on account of the 3-ce ex rire 
even although we suppose the calyx to be tubular and trifid, with a 3-lo ne ira ea 
at its base, and the petals to be i and distinct, by which it would xd s er cim 
Ca having 3 stamens, yet the habit is — different. Perhaps, rj 
ever, each apparent stamen may be composed of several filaments, a structure whic 
