Dalea versicolor

Title

Dalea versicolor

Authors

Rupert C. Barneby

Scientific Name

Dalea versicolor Zucc.

Description

132.  Dalea versicolor Zuccarini

(Plates CXVIII-CXXI)

Suffrutescent becoming shrubby, monopodial or polypodial, 2-20 dm tall, with slender but wiry, virgately erect or assurgent, when tall pliantly bending, more rarely diffuse or (in exposed places) depressed and humifuse, when old gray and ± suberous stems, either paniculately or irregularly branching distally, the young branchlets castaneous or purplish, smooth or distantly gland-verruculose toward the heads, varying like the foliage from glabrous to pilosulous, the leaves dimorphic, the primary ones longer and more complex than those of leafy axillary spurs developing late in season, the leaflets green, gray, or silvery, always punctate beneath, sometimes densely minutely gland-sprinkled above; leaf-spurs 0.5-1.7 (2) mm long; stipules commonly livid or castaneous, narrowly lance-caudate to subulate, linear-caudate, or subsetiform, when long becoming dry and fragile in age, 1-6 mm long; intrapetiolular glands 0, 1, or 2; post-petiolular glands always prominent, black or orange, conic or prickleshaped; leaves subsessile or short-petioled, the primary (drought-deciduous) cauline ones 1-4 cm long, with variably margined, ventrally grooved or ventrally carinate, sparsely punctate rachis and (3) 4-17 (18) pairs of oblanceolate, oblong-elliptic, obovate, obovate-cuneate, or linear-oblong, subacute to emarginate but nearly always bluntly gland-mucronate, flat or shallowly cymbiform, rarely folded, submembranous to subcarnosulous leaflets 1-7.5 (9) mm long, the leaves of axillary spurs or those subtending the spikes shorter and simpler; peduncles mostly terminal to branchlets, the first of each main stem-axis often leaf-opposed and up to 7 (10) cm long but mostly less, the rest shorter or obsolete; spikes often racemiform, moderately dense, the larger earlier ones commonly oblong-cylindroid, the later ones ovoid to capitate or subglobose, the densely pilosulous axis becoming 0.5-11 cm long; bracts deciduous, lance-acuminate to narrowly lance- or linear-caudate, (2.5) 3-7.5 mm long, dorsally brown or castaneous, gross-glandular, and ± densely pilose, the tips sometimes glabrate, the margins plumose-ciliate, glabrous within; pedicels 0-1 mm long; calyx densely pilose with silky, at first ascending finally spreading spiral hairs up to (0.6) 0.8-1.4 mm long, the tube (2.1) 2.4-3.5 mm long, ± oblique at orifice, the dorsal sinus sometimes shallowly recessed behind banner, the commonly golden-brown or castaneous ribs becoming thick and prominent, the intervals charged with one row of 3-6 small and pale or larger and orange or black glands, the teeth broadly to narrowly triangular at base and drawn out into laterally spurred, gland-tipped aristiform tips, the dorsal one usually longer than the rest, (2.3) 2.6-5 mm long, with very rare exceptions as long or (most often) decisively longer than tube; petals bicolored, the banner cream-colored or pale yellow, early rubescent, nearly always gland-sprinkled near and above the greenish eye-spot, like the keel and often the wings also tipped with a large elliptic blister-gland, the epistemonous ones pale lavender to bright rosy- purple, dull purplish-violet, exceptionally almost white, perched much below middle of androecium (mostly not over 3 mm distant from hypanthium); banner (4.8) 5-8.2 mm long, the claw (2.2) 2.5-4.1 mm, the deltate-cordate, ± hooded blade (2.4) 2.8-4.6 mm long, 3-5.2 mm wide; wings (4.8) 5-9.2 mm long, the claw (1.3) 1.5-3.2 mm, the oblong or elliptic blade (3.7) 4-6.5 mm long, 1.7-3.3 mm wide; keel (6.4) 6.8-11.5 mm long, the claws (2.1) 2.5-4 (4.8) mm, the broadly oval-obovate blades (4) 4.4-7 (7.5) mm long, 2.7-4.2 mm wide; androecium 10-merous, 7-10.5 (12) mm long, the longest free for 1.8-3.2 mm, the connective gland-tipped, the anthers 0.5-0.85 mm long; pod (poorly known) obtusely deltate in profile, ± 2.4-2.9 mm long, the slightly convex prow prominent but filiform, the valves hyaline in lower 1/3, thence thinly papery, charged with a few castaneous or blackish glands, pilosulous distally; seed (1.4) 1.6-2 mm long.A species of immense range and wide morphological diversity, rich in minor variants of habit and vesture, but relatively uniform in structure of the long-toothed, silky-pilose calyx and prominently gland-tipped petals. Other North American daleas with longtoothed, plumose calyx differ in one or more of these characters: petals yellow; leaflets fewer; interfloral bracts persistent.

The taxonomic problems inherent in D. versicolor sens. lat. are complicated by seasonal dimorphism; the autumnal phase with flowers and primary stem leaves coinciding is often deceptively unlike that assumed by the same plant in late winter or spring, after the primary leaves have disjointed and short-shoots issuing from their axils have taken their place. A case of this sort is illustrated in the accompanying plate (Pl. CXX) of D. versicolor var. argyrostachya. As in the related D. lutea there is a marked instability in pubescence of the foliage, which may be permanently dense and silky, lacking from the upper face of leaflets but persistent on the lower, lacking from primary leaves but present on leaves of the second order, or absent altogether below the always barbate spikes. In each variety the calyx varies from sessile to distinctly pedicelled, in apparently random fashion.

The group of entities united here under the heading of D. versicolor have been studied in their entirety by Rydberg (1920), in part by Macbride (1922, essentially a commentary on Rydberg’s monograph of 1920), and in part by Gentry (1950). The first two authors, working from herbarium material alone, failed to appreciate the importance of seasonal dimorphism and could hardly guess at the fallibility of pubescence, although this is now firmly established. Gentry’s revision is far more sophisticated and penetrating, the work of a man intimately acquainted with the living plants and their environment. Its limitation is geographical, for Gentry considered only those elements of the D. versicolor complex native to the arid grasslands of north-central Mexico and to Sierra Madre Occidental in the wider sense. Continuing study and acquisition of new material has scarcely modified Gentry’s concepts of the entities present in the area that he studied intensively, but has thrown a new light on wider problems of kinship within a group which extends far south and east in Mexico and on into the highlands of southern Guatemala. Gentry’s paper remains, nevertheless, a milestone in the taxonomy of Dalea for its perceptive commentary on the life-history of a type of suffruticose, arid-tropical dalea of which the flowering season begins with the onset of the dry weather in fall and continues, northward interrupted by a period of dormancy near mid-winter, into middle or even late spring.

The races of D. versicolor can be sorted into two groups by means of a character in the leaf-rachis first noted in D. leucantha (our var. argyrostachya) by Gentry. In this variety, as also in what have been called D. wislizeni (= var. glabrescens) and D. lasiostachya (= var. involuta), the green margin is continuously grooved along the ventral side, the groove passing uninterrupted between each pair of petiolules, which perforce are a little separated at base. In typical D. versicolor and in those forms defined by Gentry as D. wislizeni subsp. sessilis and subsp. calcarata (= homonymous vars. of this account) the margins arise so close together on the leaf-stalk’s adaxial side as to become wholly or partly confluent, if wholly so forming a ridge, interrupted by a transverse notch at each pair of petiolules, if partly so then narrowly grooved in the lower part of each segment but confluent directly under the petiolules, which arise subcontiguously from the rachis proper. In the continuously grooved type the margins terminate at each pair of petiolules with a pair of so-called intrapetiolular glands, in the partly or wholly confluent types with a single gland; but in some plants, or in leaves of them, these glands, so handy as markers for the type of rachis-margin, are unfortunately lacking.

The races of D. versicolor as sorted by type of rachis fall into two coherent groups, here evaluated as subspecies, each composed of either three of four geographic varieties. The collective ranges of each subspecies form a surprising, interlocking, almost fully vicariant pattern, that of subsp. versicolor being divided neatly into two lobes by that of subsp. argyrostachya. In the small common range of subsp. versicolor var. calcarata and subsp. argyrostachya var. argyrostachya along the Pacific slope in the south half of Sinaloa, these sympatric (but so far as known nowhere associated) components are the most highly differentiated as to rachis-margins, and within the region assume the character of discrete species. In a perceptive discussion of leaf-ontogeny in D. wislizeni sessilis, Gentry (1950, p. 249) has suggested that a strongly carinate, elliptic-acute leaflet, such as prevails in subsp. argyrostachya, is primitive to a plane obovate membranous leaflet, such as prevails in subsp. versicolor. It would appear that the continuously grooved rachis, strongly correlated with the supposedly primitive leaflet, is likewise antecedent to a rachis with fused margins. If this indicates the phylogenetic history of the complex, the genesis of the species may have occurred in the central Sierra Madrean highlands and given rise to dichotomously divergent derivatives which, migrating northward and southeastward have evolved genetically separate but morphologically parallel forms.

My comprehensive concept of D. versicolor is explained in more detail in discussion of particular varieties below. It may be of interest to note that it is already foreshadowed by Hemsley (1880, p. 242) who listed as D. lasiostachya Benth. an early (Ghiesbreght) collection of typical D. versicolor and (with mark of interrogation) the type collection of argyrostachya. During the course of my study of Dalea I attempted at first to maintain a specific distinction between D. versicolor and its northern kindred, which I grouped at first under D. argyrostachya, but the vanity of this arrangement was finally brought home to me in the field. In the meantime I have, unfortunately, left in herbaria a trail of unpublished combinations, which are, however, readily comprehensible as to intent by comparison with the synonymy accepted herein.