Chamaecrista repens var. multijuga

Stems 3—11(—14) dm, sometimes as low and diffuse as in var. repens but often taller and more erect, sometimes weakly suffrutescent in age; lvs at middle of stems up to (4—)5—10(—11) cm; lfts (10—) 11—19(—20) pairs mostly linear-oblong to oblong-elliptic, straight or falcate, (2-)2.5-4.5(-5) mm wide, the midrib usually strongly excentric, the secondary venulation variable as described for var. repens, but the first primary vein arising from the pulvinule on broad side of midrib sometimes as strong as midrib itself and produced almost or quite to blade’s apex.—Collections: 56.

Grassy and brushy places in campo, cerrado and caatinga, mostly 450-1200 m, frequent on both slopes of Chapada Diamantina and Sa. do Espinhaço in upland inland Bahia and n. and e. Minas Gerais, extending w. to Espigão Mestre and to the headwaters of Rio Paraná in e.-centr. Goiás, thence s. through Sa. do Rio Preto to headwaters of Rio Paranaíba near 17° on the Goiás-Minas boundary, n. interruptedly to the s.-w. corner of Ceará (Sa. do Araripé) and the middle Parnaiba valley near 7°S in e.-centr. Maranhão, on the Atlantic slope s. from s.-e. Bahia through far e. Minas Gerais to the Doce valley; collected once by St. Hilaire on the Atlantic coast near Ibatapoana in extreme n.-e. Rio de Janeiro and (perhaps as a waif) by Glaziou (no. 12620) within the city of Rio de Janeiro, Guanabara.—Fl. (VII-)IX-III.

The material of var. multijuga seen by Bentham was disposed of in three ways. As noted in the synonymy above, Blanchet 3711 was simultaneously referred to Cassia drepanophylla, C. subtriflora and C. brachypoda. Among Martius's material at München we found collections of var. multijuga identified by Bentham as C. brachypoda (in campis prov. Goyaz) and as C. subtriflora (Sa. das Tres Barras). The latter, associated with an unpublished name of Martius, was cited in the protologue of C. subtriflora and is herein designated as lectotypus of the same While it seems probable that the epithet subtriflora was intended by Martius to apply to his specimens of Ch. nictitans var. ramosa, these cannot be associated with the species actually described as C. subtriflora by Bentham who twice (1870, 1871, l.c.) emphasized the large flowers as the diagnostic character.

Reference has been made in the description of the species to a variation in the petiolar gland in which the stipe becomes squatly cylindric and not or scarcely wider than the head, and sometimes simultaneously glabrous. Such glands characterize most, but not all populations along the Atlantic slope from southeastern Bahia to Guanabara. The old collection from coastal Rio de Janeiro has, however, the mounded pubescent stipe prevalent inland in the São Francisco valley. We find no correlated differences in foliage or flower.