Chamaecrista chamaecristoides

Title

Chamaecrista chamaecristoides

Authors

Howard S. Irwin, Rupert C. Barneby

Scientific Name

Chamaecrista chamaecristoides (Collad.) Greene

Description

44. Chamaecrista chamaecristoides (Colladon) Greene, Pittonia 4: 20. 1899. Cassia chamaecristoides Colladon, Hist. Cass. 134. 1816.—Typus sub. var. chamaecristoide indicatur.

Robust or slender, commonly small-lvd but large-fld, always psammophilous herbs either monocarpic or perennant and suffruticose, the freely, often divaricately branching main stems diffuse or decumbent, forming mats or low thickets (3-)5-18 dm diam, when engulfed by shifting sand sometimes adventitiously rooting, the annotinous branchlets and lf-stalks sparsely to densely and cinereously strigulose or pilosulous with all incumbent, partly incumbent and partly ascending, or all loosely ascending hairs up to 0.2-0.6(-0.9) mm, the foliage concolorous, the lfts as variably pubescent as the stems, often glabrous or glabrate above, rarely quite glabrous beneath, ciliolate.

Stipules erect submembranous, pale green turning brownish-stramineous, from oblique base lanceolate or ovate-acuminate (2-)3-7 x 0.7-1.7(-2.2) mm, rather weakly 5-11-nerved, glabrous to densely puberulent dorsally, persistent.

Lvs usually widely arcuate-spreading, in expanded outline broadly to narrowly ovate, the major cauline ones up to 1.5-5 cm, on some lateral branchlets often all smaller; petiole with wrinkled pulvinus (2-)2.5-5(-6.5) mm, narrowly thick- margined, at middle 0.35-0.6 mm diam, the shallow sulcus open at gland; gland 1 (if 2-3 then very small) situated near middle of petiole proper, urceolate or subdiscoid (0.2-)0.25-0.7(-0.9) mm diam, subsessile or shortly stipitate, in profile obconic to shortly trumpet-shaped, (0.2-)0.25-0.8 mm long, varying from 0.4(-0.5) mm longer to 0.3 mm shorter than diam of head; lfts of major lvs up to 8—17(—19) pairs inserted along rachis 0.6-0.7 mm apart, decrescent both up- and downward from below middle of rachis, in outline obliquely or dimidiately linear or linear-lanceolate or the distal ones, sometimes (var. cruziana) all, semi-oblong or -obovate, obliquely mucronulate, often subfalcately incurved, the larger up to 3-12 x 0.6-2.8 mm, at base cordate on proximal and cuneate or cuneately rounded on distal side, from base 3-5-nerved by a midrib intramarginal by 0.1-0.6(-0.7) mm and dividing the blade l:2+-8, the midrib giving rise on proximal side to 2-7 secondaries, the first of these (or the innermost primary next to midrib) incurved- ascending beyond middle of blade, the venulation immersed or almost so on the (when dry sometimes striately wrinkled) upper face, bluntly, sometimes indistinctly prominulous beneath.

Peduncle and raceme-axis together 2-13(-20) mm, adnate through (0—)1—10(—15) mm, 1-3-fld; pedicels 5-15 mm; sepals brownish or yellowish, ovate- or lance- acuminate up to (6—)7—11; petals usually very unequal, 3 shorter oblong or oblong- obovate beyond the claw, 1 abaxial flabellate up to 9—17(—19) mm, the falcately oblong cucullus ± as long; androecium 10-merous, the anthers mostly red, the longer ones up to (5.5-)6.5-9(-ll) mm; ovary white-pilosulous, pilosulous along the sutures only, or glabrous; style filiform 3.5-6 mm; ovules 10-14.

Pod (25-)30-65 x (3.5-)4.5-6 mm, the purplish-castaneous, ultimately nigrescent valves thinly pilosulous or glabrous; seeds ±3.4-4.3 mm, the dull fuscous testa sometimes paler at hilum or buff-ocher overall, finely shallow-pitted.

The close relationships of Ch. chamaecristoides, as we define it in these pages, lie in two directions: on the one hand toward Ch. rufa, which replaces it at elevations of 850 m upward on the Gulf slope of Cordillera Oriental, and Ch. pedicellaris, lowland but fully disjunct on Hispaniola and eastern Cuba; and on the other to Ch. fasciculata, into which it imperceptibly fades northward by way of the essentially indefinable entity here called var. cruziana. On grounds of overall morphological similarity a strong case could be made for combining Ch. chamaecristoides and Ch. rufa into one major species. Typical var. chamaecristoides and typical var. rufa are hardly different in shape and venation of the leaflets and differ principally in proportion of flower to leaf, var. chamaecristoides retaining the large longistylous flower of Ch. rufa alongside of xeromorphic reduction, at once in length and complexity, of the leaves. But in practice the var. chamaecristoides is so easily recognized and so sharply delimited ecologically that we feel it preferable to maintain the traditional specific separation. The differential characters of Ch. pedicellaris are emphasized under that heading.

More persuasive arguments might be developed for the subordination of Ch. chamaecristoides sens. lat. to Ch. fascicidata, between which we can find no discontinuities, either morphological or ecological. While var. chamaecristoides, sens, str., is separable easily enough by the syndrome of a perennial root and narrow, highly asymmetric or dimidiate leaflets, this is linked to the pubescent monocarpic phases of Ch. fasciculata with subsymmetric leaflets by a series of populations along the Gulf coast northward from Tampico in which the prostrate growth-form and dune habitat coincide with an annual or at best feebly perennant root and broader, moderately (variably) asymmetric leaflets. Texan botanists have long been aware that the distinctive plant of the Gulf dunes southward approximately from Aransas Bay, here called Ch. chamaecristoides var. cruziana, which Turner (1959, p. 79) recognized as Cassia fasciculata var. ferrisiae and which Isely (1975, p. 93) could accept at no higher level than that of unnamed variant of C. fasciculata var. puberula, intergrades freely and insensibly on the south Coastal Plain with erect, obligately annual Ch. fasciculata. Indeed it is difficult to find two populations in this area in which the vesture of the leaflets, the displacement of the midrib, the form and stipitation of the petiolar glands and the number of leaflets are strictly identical. The taxonomic problem posed by this mass of intergradient material is of the sort faced by a surgeon required to separate a pair of Siamese twins with one heart in common; irreparable damage to the integrity of one twin is inevitable. It appears to us equally rational to follow Isely and refer all the intergradient Texan forms, even those with highly asymmetric leaflets, or dune habitat, or both, to Ch. fasciculata, which is left with only the single differential character of an annual root; or to interpret the prostrate dune ecotype as a northern extension of Ch. chamaecristoides in which a shorter life span coincides with slightly broader and, in consequence, less asymmetric leaflets. The decision to adopt the latter course is one of several unpalatable but pragmatic ones made in the course of this revision. A third solution to the problem, a return to Otto Kuntze’s concept of a multiracial megaspecies incorporating not only the taxa under immediate discussion but also Ch. glandulosa and Ch. nictitans sens. lat., may well be the soundest in a biological sense, but has been rejected as taxonomically unacceptable in practice. In shifting the emphasis from the fasciculata-like elements of the intergrades to those nearer Ch. chamaecristoides sens. str., we find it desirable to take up the epithet cruziana rather than the more familiar ferrisiae, herein referred to the synonymy of Ch. fasciculata sens. lat. The isotype of Chamaecrista ferrisiae cited above represents one of the erect pubescent forms of Ch. fasciculata found on coastal prairies inland from the Gulf itself, not the diffuse coastal form with short leaves which we interpret as Chamaecrista cruziana.

Our concept of Ch. chamaecristoides is further expanded to accommodate the controversial Chamaecrista brandegeei, which seems to represent the species very locally on the shore of the Gulf of California in Sinaloa. Known to us from only one collection, lacking mature fruits, obviously, therefore, inadequate to support definitive taxonomic conclusions, Ch. brandegeei was placed by Britton & Rose, because of its supposedly sessile but in reality short-stipitate petiolar glands, in ser. Patellariae, that is among monocarpic relatives of Ch. nictitans to none of which it appears closely akin. On the other hand it seems to differ from var. chamaecristoides sens. str., which can have the same gland, only in the somewhat smaller stipules, slightly less displaced midrib of the leaflets (in which it resembles var. cruziana), and shorter corolla, androecium and style. The habit of growth, so far as can be judged from the specimens available, seems to be that of var. chamaecristoides, and the habitat may well be so. A distributional disjunction between the east and west shores of the continent is not exceptional; compare, for example the range of Dalea scandens (Mill.) Clausen, like Ch. chamaecristoides based on a Houston collection from Veracruz and abundant around the coastal plain between Tamaulipas and Campeche, which is represented locally on the Pacific coast by an endemic var. occidentalis (Rydb.) Barneby.