Tree, up to 42 m high, dbh up to 1 m. Branchlets terete, up to 1 cm thick, inconspicuously lenti-cellate, brownish to blackish, fulvous tomentellous, glabrescent. Leaves (7-)9-13-jugate, up to 50 cm long, glabrous; Inflorescences up to c. 35 cm long, densely fulvous- to ferrugineous-tomentel-lous. Flowers zygomorphic, white. Sepals flat, hardly petaloid, outside densely appressed fulvous-hairy, outer broad-ovate to suborbicular, 2-3 by 1.5-2 mm, inner obovate, 3-4 by 1.8-2 mm. Petals 4, lanceolate-ovate to elliptic, 3 by 1-2 mm, short-clawed, outside densely appressed long fulvous-hairy, woolly along the margin; Stamens: Fruits: parts subglobular, 2 by 1.8 cm, carinate, red, glabrous. Seeds sub-globular, 1.2-1.5 cm diam.
Asia-Temperate: Taiwan (Taiwan present), Asia-Tropical: Assam (Assam present); Jawa (Jawa present); Malaya (Peninsular Malaysia present); Sumatera (Sumatera present); Thailand (Thailand present), Bangka present, Burma present, Lombok present, Madura present, Sumbawa present
Assam, Burma, Thailand, Indo-China, Taiwan; in Malesia: Sumatra (known from a few localities only), Bangka, Malay Peninsula, Java, Madura, Lombok, Sumbawa; widely cultivated, also in other parts of Malesia. It is difficult to distinguish between wild and naturalized specimens.
Sapindus pinnatus has since De Candolle often been cited as a possible synonym. From Miller's short description — based upon seedlings cultivated in England — it is clear, however, that S. pinnatus cannot be identified with S. rarak. He clearly mentions the leaves as being winged; moreover, he says that the species is more hardy than S. saponaria, which, on comparing the areas of distribution of the two species, is hardly to be expected of S. rarak. Apparently, his name refers either to seedlings of a race of S. saponaria with 8 or 10 pairs of leaflets, more than usual, or to another genus with seeds like those of Sapindus.