Mimosa pudica L.

Authors

Rupert C. Barneby

Authority

Barneby, Rupert C. 1991. Sensitivae Censitae. A description of the genus Mimosa Linnaeus (Mimosaceae) in the New World. Mem. New York Bot. Gard. 65: 1-835.

Family

Mimosaceae

Scientific Name

Mimosa pudica L.

Type

379. Mimosa pudica Linnaeus, Sp. pl. 518. 1753. — "Habitat in Brasilia." Lectotypus (Brenan, Kew Bull. 1955(2): 185. 1955), a sterile specimen in hb. Cliffort., BM!, this not identifiable below the level of species and hence constituting of itself the var

Description

Species Description - Weakly frutescent, usually diffuse or scrambling but sometimes independently erect-assurgent subshrubs precociously flowering as prostrate or ascending herbs, potentially attaining 12 dm but commonly less, with rare exceptions (var. pastoris) armed at or shortly below all nodes with a pair of broad-based, straight or declined, stramineous brown-tipped aculei (1—) 1.5—5(—6.5) mm and sometimes further with intemodal or exceptionally with a few petiolar aculei, the simple or widely few-branched stems, lf-stks and peduncles varying from glabrous or almost so to thinly strigose or thinly to densely hirsute with erect-spreading or -declined, fine tapering sordid setae to 1-3.5 mm, the foliage dull olivaceous subconcolorous, the thin-textured lfts commonly glabrous on both faces, sometimes all or only a few proximal ones of each pinna thinly strigose dorsally, rarely all minutely puberulent dorsally, all ciliate with slender, forwardly appressed or narrowly ascending setae not produced backward into a continuous margin, the ellipsoid capitula sometimes solitary in early lf-axils but verticellate by 2-7 in most distal ones, sometimes in early anthesis forming a shortly exserted pseudoraceme but this hysteranthously foliate, the fruits lateral. Stipules firm, lanceolate or linear-attenuate (3—)4—10(—11) x (0.7-)l-2.4 mm, striately (5—)7—1 l(-13)-nerved from base, either glabrous or setose distally, setose- or setulose- ciliate, persistent. Leaf-stalks (1—)2—5(—5.5) cm, the petiole proper not more than 3 mm shorter, at middle (0.4-)0.5-0.9 mm diam., openly grooved ventrally, the one (exceptionally 2) interpinnal segment, if present, 0.5-2.3 mm; pinnae commonly of all lvs 2-jug., or of some upper lvs 1-jug., rarely 1-jug. in all lvs and exceptionally in random lvs 3-jug., the distal pair scarcely or decidedly longer than the proximal, the rachis of longer pinnae of primary lvs 3-6.5(-8) cm, of lvs along some distal branchlets (1-) 1.5-3 cm, the interfoliolar segments 1—2.5(—3) mm; lfts of distal (or only) pair of pinnae 1l-27(-36)-jug., decrescent only at each end of rachis, the first pair (0.3-)0.7-1.8 mm distant from lanceolate, often unequal, 1-nerved paraphyllidia (0.3-)0.5- 2(-2.3) mm, all in outline linear-oblong from obtusangulate base, straight or rarely subfalcate, abruptly acute-apiculate, those near mid-rachis (4—)5—13(—14) x (l-)1.2-2.3(-2.5) mm, ±4-6 times as long as wide, all veinless above, beneath slenderly (4-)5-nerved from pulvinule, the distally subcentric midrib remotely 3-5-branched on each side, the single anterior and inner posterior nerves filiform, shortly intramarginal and produced well beyond mid-blade, the outer posterior one(s) much shorter. Peduncles (0.7-)l-3 (-3.5) cm; capitula without filaments or bracteal setae 3.5-5 mm diam., prior to anthesis either moriform or conelike and cancellately hispid, the linear receptacle ±5-10 mm; bracts linear or linear-oblanceolate, abruptly incurved above or below middle, often accrescent upward along the rceptacle, those near mid-capitulum 1-2.5 x 0.2-0.5 mm, all dorsally glabrous 1-nerved, the margin either glabrous or beyond middle ciliate on each side with 1-3 or rarely several fine ascending setae 0.1-1.4 mm; flowers 4-merous 4-androus, some or many lower ones staminate; calyx a membranous campanulate cup 0.1-0.2 mm, the rim either subtruncate or minutely, often asymmetrically denticulate, one or more teeth sometimes tipped by a short setula; corolla narrowly vase-shaped or subtubular, or some bisexual ones at times narrowly turbinate, the bisexual ones 1.5-2.1 mm, the membranous, whitish or pink-purple, weakly 1-carinate, shallowly concave, ovate lobes 0.45-0.7 x 0.4-0.5 mm, externally either glabrous or puberulent; filaments pink sometimes fading whitish, at base very slender and either free or weakly monadelphous through 0.2 mm or less, exserted 3.5-6 mm; ovary minutely stipitate, at anthesis glabrous, commonly 4-ovulate. Pods several or to ±30 per capitulum, subsessile, in profile undulately narrow-oblong 8-15 x (2.7-)3.5-4.5(-5) mm, (2-)4-seeded, the constricted replum 0.25-0.45 mm wide, contracted at apex into a cusp 0.5-2.5 mm, hispid on dorsal and lateral ribs with stramineous, often brown-tipped tapering setae to 2—4 (-5) mm, the smooth greenish-brown or dark brown papery valves usually glabrous, sometimes densely pilosulous, when ripe breaking up into free-falling biconvex indehiscent articles (2-)2.5-3.5(-4) mm long; seeds plumply lentiform 2.4-3.3 x 2.1-2.8 mm, the dull fawn-brown or fuscous-castaneous testa microscopically granular.

Discussion

Mimosa pudica, the mimosa best known, although not exceptional, for its shrinking response to shock or touch, is closely related to M. polydactyla. From this it differs most obviously in simpler leaf-formula, but also (cf. key to subser. Pudicae) in broader stipules and longer filaments. In some respects it is a variable species, but the task of disentangling the intraspecific taxonomy is lightened by the worldwide revision by Brenan (1955, condensed in 1959: 46), who proposed a lectotype and recognized within the species three varieties distinguished by characters here assembled in key form:

a. Corollas glabrous; floral bracts 0.7-1.1 mm, shorter than submature buds and marginally either glabrous or minutely setulose; stipules 4-7(-8) mm; stems either glabrous or hispid......

.................................var. unijuga.

a. Corollas puberulous distally; other features variable.

b. Floral bracts 1-1.5 mm, either glabrous or ciliate, but marginal setulae not over 0.75 mm; stipules of var. unijuga; stems more often hispid than glabrous.............var. tetrandra.

b. Floral bracts 1.8-2.2 mm, ciliate, the marginal setae 1-1.5 mm; stipules 8-14 mm; stems always hispid. ...............var. hispida.

It will be noted that var. pudica could not be keyed because the identity of the sterile specimen in the Linnaean herbarium cannot be positively identified. The var. pudica is therefore restricted to this one specimen.

When applied to material of M. pudica from Africa, Asia, Malesia and Micronesia, Brenan’s key conditions are neatly met in almost all cases. Although widespread and locally very populous, the races of M. pudica encountered in the paleo-tropics appear relatively simple genetically, being descended probably from a finite number of introductions from the Americas in post-Columbian times, and they are essentially uniform over extensive areas. Thus all Philippine M. pudica that I have seen is pronounced var. hispida, all Hawaiian material pronounced var. unijuga. It is otherwise in tropical America, where I have found many instances of independent variation in length of stipule, length of floral bracts, and length of bracteal cilia that make strict application of Brenan’s criteria difficult. As in many Leguminosae, stipules and bracts tend to vary together, so that the relatively long stipules and bracts attributed to var. hispida may be viewed as two aspects of one basic difference. Similarly, setose pubescence of stems and of bracts is generally, though not invariably, linked. A glabrous corolla, which characterizes var. unijuga, is occasionally associated with the long stipules, and occasionally with the long bracteal setae, of var. hispida. Brenan tentatively suggested that var. unijuga and var. hispida, which are substantially different overall, could be distinct but interfertile species which have given rise by introgression to a polymorphic derivative var. tetrandra. An objection to this hypothesis is that var. hispida, which bulks large among paleotropical populations of M. pudica, is, as Brenan was quick to observe, inexplicably rare and scattered in the Americas, where it lacks a focus of abundance in undisturbed habitats that could be identified as an aboriginal homeland.

Because M. pudica sensu lato is assumed to be of American origin, like all near kindred, it should be possible to learn something of its racial history by close observation of its dispersal in the New World, however modified by man this may be. To this end I divided 180 gatherings of M. pudica on loan or on permanent deposit at NY into Brenan’s three varietal categories and, ignoring a few specimens of controversial identity, plotted each on a separate map. From these maps it appeared that the foci of abundance for vars. unijuga and tetrandra, although overlapping, are substantially different. The dense continuous range of var. unijuga is primarily circum-Caribbean, extending from Jamaica east through the Greater and Lesser Antilles to Trinidad, thence west through northern Venezuela and northern Colombia to Panama, north along the Atlantic coast of Central America to Belize, and west to the southern bend of the Gulf of Mexico between Tabasco and Veracruz. Outside this very natural range var. unijuga occurs, or has occurred, sporadically as follows: on North Bimini in the Bahamas; in urban waste places and around seaports in El Salvador, Pacific Ecuador, and the Guianas; in the Amazon valley at Manaus and Belém only; along the upper rio Madeira in Rondônia, Brazil; and as a roadside weed in coastal and interior southern Bahia, where it was first found by Martius in the year 1818. Native status of var. unijuga within and south of Amazonia seems hardly credible.

The primary range of var. tetrandra, as determined from morphologically unambiguous specimens in which short stipules, bracts, and bracteal cilia coincide with puberulent flowers, is concentrated in central Colombia, from the Pacific coast to the Magdalena valley, and around the southern borders of the Guayana Highland in Venezuela and neighboring Brazil, in which range it is the only form of M. pudica known to occur. It is common also on the Caribbean slope in Venezuela eastward to Trinidad, Tobago, and the southernmost Antilles, where it is sympatric with var. unijuga, but not with var. hispida. Outside these areas of concentration var. tetrandra has been collected in urban Veracruz, Mexico, in urban central Costa Rica (where all three varieties converge), rarely in Cuba, and in Brazil at Manaus, around the bay of San Salvador in Bahia, and southward at scattered points along the coast in Espírito Santo, Rio de Janeiro (with Guanabara), and S. Paulo. It is improbably native anywhere outside of northern South America.

The var. hispida, as already noted, is relatively uncommon in the Americas, and indisputable native status cannot be inferred from its presently known dispersal. It is the most often cultivated form of M. pudica, but has not by this means achieved any secondary dispersal in America comparable to that gained in the Old World tropics. I have the following records: in Antillea, from New Providence in the Bahamas, from suburbs of La Habana and elsewhere in Cuba, from Jamaica and from St. Croix; in Central America one each from capital cities of El Salvador and Costa Rica; in South America one each from Trinidad and Surinam, two from Iquitos in Peru, one from Minas Gerais in southeastern Brazil, and two from extratropical Brazil, clearly escaped from cultivation. In Mexico var. hispida has been found in the garden city of Cordoba, not probably wild, but also in two places on the Pacific slope in Guerrero and Oaxaca, outside the range of the other two varieties and apparently in deep countryside. It is conceivable therefore that var. hispida arose in southwestern Mexico and was thence introduced, at first deliberately as a curiosity and subsequently as an invasive weed in Africa, India and elsewhere, where according to Ridley (Dispersal of plants throughout the world 587, pl. XIX, fig. 7, 8, 1930) it was passed from mission to mission by Jesuit fathers. Brenan’s hypothesis of introgression between var. hispida and var. unijuga is not supported by this analysis of distribution.

While Brenan’s three varieties of M. pudica are strongly marked in the Old World, in the neotropics continuous and independent variation in stipule-length and stem-pubescence makes these characters weak and sometimes contradictory taxonomic markers. The following key is accordingly a simplified version of Brenan’s and includes, moreover, a minor new taxon from Guyana.