SAPINDACEZ. SILVA OF NORTH AMERICA. 67 
SAPINDUS. 
FLoweErs polygamo-dicecious, regular; sepals 4 or 5, imbricated in estivation ; 
petals 4 or 5, naked or appendiculate, imbricated in estivation. Ovary 2 to 4-celled ; 
ovules solitary. Fruit baccate, coriaceous, 1 to 3-seeded. 
Sapindus, Linneus, Gen. 359.— Adanson, Fam. Pl. ii. 
343. — A. L. de Jussieu, Gen. 247. — Cambessedes, Mém. 
Mus. xviii. 26. — Endlicher, Gen. 1070. — Meisner, Gen. 
53. — Gray, Gen. Ill. ii. 213. — Bentham & Hooker, Gen. 
i. 404. — Baillon, Hist. Pl. v. 394.— Radlkofer, Sitz. 
Akad. Miinch. xx. 283. 
Aphania, Blume, Bijdr. Fl. Ned. Ind. 236. — Endlicher, 
Gen. 1070. — Meisner, Gen. 52. 
Didymococcus, Blume, Rumphia, iii. 103. 
Trees or shrubs, sometimes subscandent, with terete branches, thick fleshy roots, and bitter and 
detersive properties. Leaves alternate, destitute of stipules, abruptly pinnate or rarely one-foliolate ; 
leaflets alternate or opposite, entire or occasionally serrate. Flowers minute, in ample axillary or termi- 
nal racemes or panicles. Pedicels short, from the axils of minute deciduous bracts. Sepals unequal, 
slightly united at the base. Petals equal, alternate with the sepals, inserted under the thick edge of the 
disk, unguiculate, naked or often furnished at the summit of the claw, on the inside, with a two-cleft 
scale, deciduous. Disk annular, fleshy, entire or crenately-lobed, hypogynous or perigynous. Stamens 
usually eight or ten, rarely four to seven, inserted on the disk immediately under the ovary, equal ; 
filaments subulate or filiform, often pilose, exserted in the sterile, much shorter in the fertile flower ; 
anthers oblong, attached near the base, introrse, two-celled, the cells opening longitudinally. Ovary 
sessile, entire or two to four-lobed, two to four-celled, contracted into a short columnar style; rudimen- 
tary in the staminate flower; stigma two to four-lobed, the lobes spreading ; ovules solitary in each cell, 
anatropous or amphitropous, ascending from below the middle of the inner angle of the cell; raphe ven- 
tral; micropyle inferior. Fruit usually formed of one globose fleshy or coriaceous carpel, the others 
abortive, their rudiments remaining at its base ; or of two or sometimes of three carpels more or less 
connate by their bases, and then two or three-lobed. Seed solitary im each carpel, obovate or globose, 
destitute of albumen ; testa crustaceous or membranaceous, smooth, black or dark brown; tegmen mem- 
branaceous or fleshy; hilum oblong, surrounded (at least in the North American species) by an ariloid 
tuft of long pale silky hairs. Embryo incurved or straight; cotyledons thick and fleshy, incumbent ; 
radicle very short, inferior, near the hilum. 
The genus Sapindus is widely distributed through the tropics, especially in Asia, occasionally 
extending into subtropical regions. One of these, the 
type of the genus and a common West Indian tree, reaches the shores of southern Florida, and another 
occurs in the southern part of the North American continent from the coast of Georgia to northern 
Sapindus existed in Europe in the Tertiary period, and even earlier, with forms which repre- 
About forty species have been distinguished.’ 
Mexico.” 
sent the ancestors of existing American species.” 
377. — Hemsley, Bot. Biol. Am. Cent. i. 214.—Hillebrand, Fi. 
Haw. Is. 85. 
1 De Candolle, Prodr. i. 607. — Blume, Rumphia, iii. 93. — Wal- 
pers, Rep. i.416 ; v. 362 ; Ann. i. 134; ii. 211 ; iv. 378. — Thwaites, 
Enum. Pl. Zeylan. 55. — Turezaninow, Bull. Mose. i. 401. — Hooker 
f. Fl. Brit. Ind. i. 682. — Miquel, Fl. Ind. Bat.i., ii.551 ; Suppl. 198, 
508; Mus. Lugd. Bat. iii. 92. Harvey & Sonder, Fl. Cap. i. 
240.— Oliver, Fl. Trop. Afr. i. 430.— Bentham, Fi. Austral. i. 
464. — St. Hilaire, Pl. Usuelles Brasil. 368 ; Fl. Bras. Merid. i. 300, 
t. 81. — Gray, Wilkes Explor. Exped. i. 251. —Grisebach, Fl. Brit. 
W. Ind. 126.— Triana & Planchon, Ann. Sci. Nat. ser. 4, xviii. 
2 Radlkofer (Sitz. Akad. Miinch. 1878, 221 ; Durand Index Gene- 
rum, 81) refers many of the species of Sapindus to other genera, 
As his 
paper is an annotated catalogue and not a monograph of the genus, 
reducing the number to ten, with a few doubtful ones. 
it is not easy to judge of the value of his conclusion with regard to 
the limitation of genera and species. 
8 Saporta, Origine Paleontologique des Arbres, 279. 
