Rupert C. Barneby
Barneby, Rupert C. & Grimes, James W. 1997. Silk tree, guanacaste, monkey's earring: A generic system for the synandrous Mimosaceae of the Americas. Part II.
Mimosaceae
ZYGIA
Zygia P. Browne, Civ. Nat. Hist. Jamaica 279, t. 22, fig. 3. 1756. — Typus infra sub sect. Zygia indicatur.
Unarmed, macro- and microphyllidious arborescent shrubs and slender, 1—several-stemmed trees, seldom attaining 20 m, with smooth or flaking, whitish or pallid gray annotinous and older branches, the majority (all the common species) cauliflorous. Indumentum of short plain, gray or brownish, sometimes rufescent hairs often confined to lf-axes or inflorescence, the lf-blades often glabrous, rarely pilosulous dorsally, the fruit usually puberulent or pilosulous when young but glabrescent at maturity, sometimes permanently pilosulous. Stipules usually small firm caducous, nerveless or bluntly few-nerved, rarely thick-textured persistent, occasionally papery and striately nerved. Lf-formula variable, most commonly (a) i/1½-5½ (-6½), in a few spp. i/½, with petiole mostly much shorter than the rachis of the 1 pair of pinnae, occasionally as long, the 2 pinnae of any lf sometimes of unequal length and the anterior lfts of proximal pair often wanting, in few species each pinna 1-foliolate, the whole lf consequently 2- to 26-foliolate; or (b) ii- x/6-23, the lfts then diminished with increasing number; or (c) exceptionally (Z. inundata) lvs simply pinnate as in Inga. Lf-nectaries of conjugately pinnate lvs situated at or immediately below tip of petiole and of each pinna-rachis, of pluripinnate lvs at or close below first and often of further pinna-pulvini, and sometimes smaller ones between further lft-pairs, in any case all sessile, shallowly cupular, plane or low-convex (often eaten, exceptionally immersed, in Z. transamazonica lacking). Venation of lfts pinnate, or palmate and pinnate, in some species of sect. Zygia the first 1-2 secondary nerves on posterior side of midrib stronger and longer than the rest, incurved-ascending to or beyond midblade. Inflorescence arising in most species from knots on annotinous and older wood, below current foliage, occasionally (Z. odoratissima, Z. ocumarensis) axillary to coeval lvs, consisting commonly of capitula or spikes, exceptionally of umbelliform capitula or racemes, when cauline arising either directly, solitary or fasciculate and pedunculate or subsessile, from the trunk or branch, or forming a pseudoraceme or weakly branched panicle, each unit subtended either by a simple bract or by a rudimentary lf-stk with nectary; floral axis of each inflorescence-unit 1 mm to 5 dm, the fls contiguous when capitate, but well spaced along extended axes; perianth normally 5-merous (random exceptions); calyx shallowly saucer-shaped to campanulate, 0.5-3(-4) mm, rarely campanulate to 5.5 mm or (Z. codonocalyx) to 11.5 mm, the rim truncate or denticulate, the teeth often of unequal length; corolla subtubular or dilated at the limb, the tube often striately nerved, the lobes at anthesis erect-ascending; androecium 26-130-merous, the stemono- zone short or obscure, the tube either shorter than corolla or far exserted; ovary sessile or almost so, commonly surrounded at base by a crenately lobed nectarial disc (this lacking only in some pluripinnate species), symmetrically attenuate into style, this not or scarcely dilated at the stigma; ovules (7-)8-17. Pods geotropic, either sessile or attenuately pseudostipitate, diverse in size, texture, and compression, either straight, or gently decurved (and sometimes twisted), or evenly retrofalcate to decurved through nearly a full circle, in profile oblong, linear, or undulately linear but only exceptionally torulose or moniliform, overall 3 cm to ±4.5 dm long; valves either leathery and planocompressed except for small umbo or depressed dome over each seed, or lignescent and then often biconvex or subterete at maturity, either smooth, or rugulose, or less often transversely fissured or elaborately sulcate, the sutures forming a frame only around planocompressed valves, otherwise immersed or almost so, exceptionally dilated and wider than valves; cavity of pod usually but not always continuous; dehiscence inert, follicular or more often through both sutures, but often tardy, the valves narrowly gaping or recurving to release the seeds; seeds uniseriate on slender crumpled funicle, either discretely spaced along the cavity or (especially when large or fat) contiguous or imbricate and then distorted by mutual pressure, in orientation conforming to constraint from the valves, in broad view either round or oblong, varying greatly in size, outline, and thickness, when smaller compressed-lentiform, when enlarged biconvex, exceptionally narrow-ellipsoid, the thin testa either smooth or wrinkled, often lustrous, becoming papery and fragile, nerved or narrowly winged around periphery, lacking pleurogram; embryo green when fresh, horny when dried; endosperm lacking.
An American genus of nearly 60 species, the true number contingent on incomplete or unpublished studies of sect. Zygia by M. L. Rico Arce (K) and C. E.Barbosa C. (FMB), several needing redefinition and others awaiting description; widespread from lowland tropical Mexico and Greater Antilles S to SE Brazil, NE Argentina, Paraguay, and Amazonian Bolivia, extending W of the Andean cordilleras only between SW Mexico and N Peru; most diverse in Central America, Colombia, the Guianas, and NW Amazonia, characteristically hygrophilous, of riparian forest and shore habitats, from mangrove swamp at sea level to 500 m, but few attaining ±2600 m in the N Andes, or (Z. ocumarensis) in upland semi deciduous forest of NE Colombia and NW Venezuela.
We here adopt a broad definition of the genus Zygia that transcends that of Britton and Rose by admitting several pluripinnate species, some of them microphyllidious. As here defined, Zygia is essentially equivalent to Pithecellobium sect. Caulanthon Bentham, modified to admit taxa discovered since the Revision of suborder Mimoseae (1875). The crucial characters of Zygia are cauliflory (not quite universal), homomorphic flowers, an intrastaminal disc surrounding the base of the ovary (exceptionally lacking), and a thin seed-coat without pleurogram. In a few exceptional or peripheral members of the genus either cauliflory or the disc is wanting, but never both at once.
When first formulated by Bentham (1844), as a section of Pithecolobium, the genus Zygia accommodated all known cauliflorous, American Ingeae other than such true Ingae as I. glomeriflora Ducke, which following discovery of simply pinnate Z. in undata seem to differ consistently only in indehiscence of the pod. While the protologue of sect. Caulanthon called for "40-50" species, only 11 were described at that time, all but the last of which have exactly 1 pair of pinnae per leaf and not more than 4 pairs of ample leaflets per pinna, in one species no more than a solitary leaflet. The exception was Z. ramiflora, different from the rest in leaf-formula of "ii/5-7" (in hindsight revised to ii-v/6-10). In his summary monograph of Mimoseae (1875), Bentham added four new species of sect. Caulanthon. Three were variations on the conjugately pinnate theme, but one, P. claviflorum, materially expanded the section’s definition by introducing pluripinnate microphyllidious leaves reminiscent of Macrosamanea. At this point, fruits were known only in species with conjugately pinnate leaves, all of one basic pattern, though heterogeneous in size, texture, and curvature. Also, although Bentham did not emphasize the point, all these species, except P. claviflorum, share the feature of an intrastaminal disc.
With the passage of years, surprising additions have been made to the conjugate-pinnate group of zygias, some (e.g., P. odoratissimum Ducke; Z. confusa L. Rico) modified in axillary, not cauline origin of the inflorescence, others (Z. obolingoides L. Rico) in the individual flower or (e.g., Z. rhytidocarpa L. Rico) in the fruit, and one (Z. inundata Ducke, already mentioned) in simplified leaves resembling those of Inga. Yet all these are in other respects compatible with the original nucleus of sect. Caulanthon in the syndrome of leaf-formula and nectarial disc and are unmistakably close allies. In addition, we now possess several other, at once cauliflorous and plurijugate Zygiae that fit less readily into Bentham’s section. Four relatives of P. (sect. Caulanthon) claviflorum unknown to Bentham now form a close-knit satellite group, anomalous in pallid coloration of the foliage and in lack of intrastaminal disc, a group described below as Zygia sect. Zygiopsis. Pithecellobium ramiflorum has been joined by P. collinum Sandwith, Z. tetragona, and Z. transamazonica, which stands somewhat apart in loss of leaf-nectaries; together these form a discrete sect. Parazygia that might be visualized as plurifoliolate precursors of sect. Zygia. Four yet further modified species belong evidently to this same circle of affinity, but each diverges in one or more particulars and could logically be excluded from Zygia as a monotypic genus:
1. Pithecellobium racemosum Ducke, treated in recent years as type of a distinct genus Marmaroxylon, stands apart from the core of Zygia in having multifoliolate leaves combined with relatively large stipules imbricate around the resting buds, a complex pseudoracemose (but still cauline) inflorescence of fasciculate capitula, notably small, often yellowish rather than green-white-pinkish flowers lacking intrastaminal disc, and a relatively small narrow pod, at least incipiently septate and resinous inside. The seeds are smaller than those of typical Zygia, and the heartwood is reportedly (Baretta- Kuipers, 1981: 689) distinct in color, though hardly different in microscopic structure. This set of morphological characters might suffice to support generic status. However, the stipules can be matched in some species of sect. Zygiopsis, as can the lack of intrastaminal disc; its inflorescence is that of conjugately pinnate Z. ampla (except that the fasciculate, not solitary peduncles are subtended by a simple bract rather than by a rudimentary leaf-stalk), while the narrowly septate, internally resinous pod and the small seeds are nearly those of Z. ampla and of Z. ocumarensis, discussed below.
2. Pithecellobium eperuetorum Sandwith, local in Guyana, has leaf-formula of ii/3-5, unusually elongate leaflet-pulvini (scarcely duplicated in sect. Zygia), racemose (not spicate) units of inflorescence, and flowers with a disc. We propose for this enigmatic entity a monotypic sect. Barticaea.
3. Zygia ocumarensis was described from flowering material and referred by Pittier to Klugiodendron, the type of which it resembles in leaf-formula but not in other respects. Its capitate, narrowly cylindric flowers with intrastaminal disc are unexceptional in Zygia, but the leaf-formula of ii/1 and the mostly axillary, solitary or pseudoracemose capitula do not conform to traditional standards for that genus. The fruit of Z. ocumarensis, however, which is narrow, septiferous and resinous inside and follicular in dehiscence, points to a common origin with Z. racemosa alone.
4. Zygia sabatieri has peculiar, slenderly pedunculate capitula of shortly pedicellate, narrowly tubular flowers that recall the genus Cojoba and undulately linear fruits deeply pinched between seeds, unlike any other Zygia. The slenderly ellipsoid, basipetal seeds are likewise unmatched in the genus. It is segregated without apology as a sect. Pseudocojoba.
Inclusion of these four awkwardly idiosyncratic species in an enlarged genus Zygia precludes a neat definition of the genus. It has the merit of bringing into taxonomic and nomenclatural association species that might appear, in light of all macroscopic characters, genuinely related taxa. The alternative is recognition of as many monotypic genera.
Cassens and Miller (1981) found that the wood of Zygia (including Marmaroxylon) is unique among New World members of the Pithecellobium-complex in having confluent parenchyma but lacking septate fibers. They did not examine species from sects. Zygiopsis, Nothellobium, Parazygia, Barticaea, or Ingopsis. Baretta-Kuipers (1981) noted that the wood of Zygia is distinct in having very thick-walled fibers, abundant aliform to aliform-confluent parenchyma, and 1-2 seriate rays and that Marmaroxylon is microscopically similar.
No cladistic analysis of Zygia is presented. Unfortunately more than one third of the species are still unknown in fruit, some only partially known, and a few are known in fruit but not in flower. This lack of data precludes a cogent analysis.