Amply multifoliolate cauliflorous treelets 2-4 m, closely resembling Zygia basijuga in foliage and fruit but different in paleaceous striate stipules and distantly allopatric, the young stems and all lf-axes densely hispidulous with erect straight, yellowish or bronze hairs to 0.15-1 mm, the thin-textured, when dry curling and crumpling, glabrous lfts olivaceous above, pallid beneath, the short-pedunculate few-fld capitula solitary on very short peduncle, the primary inflorescence-axis greatly condensed or obsolete. Stipules erect, narrowly lance-attenuate 7-19 x 0.5 1.5mm, early dry paleaceous, striately (3-)5-9- nerved, dorsally glabrous or (in Guyana) thinly pilosulous, persistent unless broken. Lf-formula iv-v/13-19; lf-stk of longer lvs 8.5-15 cm, the petiole (consisting largely of lf-pulvinus) 3.5-8(-14) x 2-3 mm, the longer interpinnal segments ±25-40 mm; nectary between the short first pinna-pair cupular thick-rimmed 1-1.7 mm diam, similar ones between nearly all further pinna-pairs and yet smaller on pinna-rachis between all but a few proximal pairs of lfts; pinnae abruptly decrescent proximally, not or scarcely so distally, the rachis of longer ones 7—13 cm, the longer interfoliolar segments 6-9 mm; pulvinule of lfts obsolete, the blade sessile against pinna-rachis; lfts strongly decrescent proximally, the first, very unequal pair arising next to pinna-pulvinus, the blades falcately narrow-oblong from obliquely truncate or very broadly inequilaterally flabellate, postically acute-auriculate base, incurved beyond middle and abruptly deltate-acuminulate and mucronate, the larger ones 19—32 x 5—9 mm, 2.9—4 times as long as wide; venation of Z. claviflora. Peduncles 1-6 mm, each subtended by a reflexed bract charged ventrally with a small shallow-cupular nectary; capitula 4-10-fld, the receptacle ±1.5 mm; fls sessile, the 5-merous whitish perianth typically glabrous, but in Guyana sometimes thinly puberulent, everywhere campanulate, including short solid base 2-3.2 x 1-1.3 mm, the teeth 0.15-0.25 mm; corolla narrowly tubular-funnelform 10.5-13 mm, a little ampliate in distal 1/3, the lobes 1.5-2.1 mm; androecium 50-56-merous, ±24 mm, the stemonozone ±1 mm, the slender tube 15-16.5 mm, the free filaments pink distally; ovary sessile linear 2-2.3 mm, obliquely conical at apex, at anthesis glabrous. Pods (not seen fully ripe) pendulous, in profile broad-linear straight 17-20 x 1.8 cm, contracted at base into a stipelike neck 1.5-3 mm, obtuse at apex, 14—16-seeded, the thinly leathery, coarsely venulose, glabrous fuscous-brown valves framed by slightly dilated, almost straight, ciliolate sutures, low-colliculate over each ovule; dehiscence and seed unknown.

In swampy forest and on riverbanks subject to flooding, 130-330 m, well known from the base of Cerro de la Neblina (headwaters of Ríos Baria, Mawarinuma, and Yatua) in the state of Amazonas, Venezuela, and disjunct (but see discussion) on Kaieteur Plateau in centr. Guyana. — Map 20. — Fl. Ill, V, X, XII, perhaps intermittently throughout the year.

Zygia palustris resembles Z. claviflora and Z. basijuga in general aspect and especially in the pallid foliage of thin texture that retains its olivaceous coloring when dried. It further resembles the distantly allopatric, upper Amazonian Z. basijuga in the sharply mucronate incurved tip of the leaflets and in the elongate pod glabrous except for ciliolate sutures. It differs from Z. basijuga, however, in long paleaceous, striately nerved stipules, in the suppressed primary axis of the pseudoraceme, and in the pink, not white tassel of free filaments. The stipules of Z. palustris are nearly those of the almost sympatric Z. claviflora, but the pods, as contrasted in the key to species, are fundamentally different. The leaflets of Z. claviflora are smaller than those of Z. palustris and only minutely mucronate at the deltately obtuse apex.

Typical Z. palustris from the base of Cerro de la Neblina has glabrous stipules and perianth. One collection from Guyana (Cowan & Soderstrom 2013, K, NY, not mapped) has thinly pilosulous stipules and perianth but appears otherwise identical. A second (1879, im Thum, K), from Kaieteur savanna, mentioned by Sandwith in the protologue of Pithecellobium collinum, seems conspecific. Their identity needs confirmation from fruiting material, but they cannot, we think, represent the related and closely sympatric Z. potaroensis, which see for comment.