Mimosa parvifolia Swartz, Flora Ind. Occid. 2(2): 948. 1800. —. . in arenosis apricis non procul a littoribus mans Antillas affluentis." — Holotypus, S(herb. swartz., 2 sheets)! one = NY Neg. 12581. — Acacia parvifolia (Swartz) Willdenow, Sp. Pl. 4: 1086. 1806. Pithecolobium parvifolium (Swartz) Bentham, London J. Bot. 3: 223. 1844.

Inga marthae K. Sprengel ex de Candolle, Prodr. 2: 441. 1825. — "ad Sanctam-Martham. Bertero [in herb, balbis.]." — Holotypus "misit Balbis, 1822", fl., G-DC! = F Neg. 6971! — Equated with Pithecolobium parvifolium by Bentham, 1844: 223.

Albizzia marthae Britton & Killip, Ann. New York Acad. Sci. 35: 133. 1936. — "Santa Marta, Magdalena, Colombia, about 80 m elevation, H. H. Smith 296!' — Holotypus, US!; isotypi A!, GH!, NY (one sheet labeled: "Near Bonda, below 500 ft," one without locality-data)!, U!, UC 1219189!. - Pithecellobium marthae (Britton & Killip) Niezgoda & Nevling, Phytologia 44: 379. 1979.

Lf-formula iv-ix/14-25(29); lfts linear from obtusangulate base, deltately acute, the larger ones (3.5)5- 8 x 1-1.8 mm, (3.1)3.5-5 times as long as wide. Pods almost straight to falcately recurved, when well fertilized 9-13 x (1-)1.1-1.5 cm, the sutures not or scarcely constricted, the valves coherent between seeds but not produced inwardly as septa.

In thickets and disturbed matorral, sometimes surviving in hedges, below 150 m, locally frequent around Lago Maracaibo in NW Venezuela (Zulia), in the Caribbean lowlands of N Colombia (Magdalena, Atlántico, Bolívar), and in central Panama, and known from single collections in Pacific Honduras, from Chiapas, Mexico (Arriaga), and in coastal SW Ecuador; disjunct in SE Jamaica, where perhaps an introduction of long standing. — Map 41. — Fl. (X-XI) II-V. — Caudero, quebrahacha (Venezuela); carbonero, hoyo de zorra, vainillo (Colombia).

In the protologue, Jacquin ascribed to Mimosa mangensis a leaf-formula of v/16, but in 1791 illustrated it by a leaf showing only 12 pairs of leaflets. Swartz at first identified his Jamaican plant with Jacquin’s, but was dissuaded by an apparent disparity in leaflet- number, redescribing the species as M. parvifolia. Modern material from the western Caribbean and from Jamaica shows almost the same range of variation in leaf-formula and confirms Bentham’s conclusion that these two names apply to one taxon.

The continental dispersal of Ch. mangense follows a long-recognized pattern of dispersal that coincides with a belt of drought-deciduous woodland. The stations in Jamaica are seemingly anomalous, and the species may not be aboriginally native there. It is not recorded where on Jamaica Swartz collected M. parvifolia, but all subsequent collections are from Kingston or the hill country of parishes immediately adjoining the city. It was unknown to (or not recorded by) Patrick Browne, and James Macfadyen, writing in 1837, had it only from Kingston and Linguanea, now part of the city itself.

Inga marthae was recognized as a synonym of Pithecellobium parvifolium by Bentham as early as 1844. Albizia marthae was transferred by Niezgoda and Nevling to Pithecellobium on examination of the pollen, which is discordant with that of typical Albizia; they made no comment on the validity of the species within Pithecellobium.