Astragalus bisulcatus

Title

Astragalus bisulcatus

Author(s)

Rupert C. Barneby

Scientific Name

Astragalus bisulcatus (Hook.) A.Gray

Description

108. Astragalus bisulcatus

Coarse or sometimes quite slender, ill-scented, leafy perennials, with a thick, woody, pluricipital taproot, strigulose with fine, straight, appressed hairs up to 0.3-0.6 mm. long, often quite thinly so, the thick-textured herbage either bright green or pallid and subglaucescent, occasionally cinereous when young, the leaflets glabrous or nearly so above; stems several or numerous, either erect and ascending in clumps or decumbent with incurved-ascending tips, 1.5-5 (7) dm. long, simple or branched (spurred) at 1-3 nodes preceding the first peduncle, commonly fistular at base, green or purplish-tinged; stipules submembranous early becoming pallid and scarious, 2.5-12 mm. long, the lowest broadest and shortest, at least in vernation connate into a subtruncate or bidentate (in age fragile and often ruptured) sheath, the median and upper ones progressively less connate upward, the uppermost free or united by a stipular line, with deltoid or triangular-acuminate, spreading or deflexed blades; leaves (3) 4-12 cm. long, the lowest shortly petioled, the rest subsessile, with (7) 15-35 ovate-oblong, oblong-elliptic, lance-elliptic, or oblanceolate, obtuse (but often mucronulate), exceptionally emarginate, or (in some upper leaves) linear-elliptic and acute, flat or loosely folded leaflets 0.5-3.2 cm. long; peduncles 2.5-13 cm. long, erect or (from decumbent stems) vertically ascending or incurved; racemes 25-80-flowered, dense and ovoid to narrowly cylindroid in early anthesis, the flowers early nodding ± retrorsely imbricated, becoming looser (especially at base), the axis elongating, 3-25 cm. long in fruit; bracts membranous, lanceolate, nearly always surpassing the pedicel, deflexed in age; pedicels slender or filiform, straight and ascending or arcuately spreading, at anthesis 1—2.5 (3) mm., in fruit 1.7-3 mm. long; bracteoles 0—2, sometimes conspicuous; calyx 3.5-9.6 mm. long, strigulose with white or some black hairs, the strongly oblique disc 0.6—1.1 mm. deep, the obliquely campanulate, membranous, pallid, purplish, or dark red tube 2.8—5.7 mm. long, either truncate at base and inserted on the pedicel by its ventral corner, or tumid and gibbous-saccate behind the pedicel, the teeth variable, subulate or sometimes setaceous, longer to much shorter than the tube, the ventral pair usually shortest, the whole becoming papery, marcescent unruptured; petals white, ochroleucous, dull straw-yellow, purplish- suffused or -veined, or bright rose-purple (drying violet), when pallid the keel-tip usually masculate; banner oblanceolate to oval-obovate, somewhat sigmoidally arched, the blade recurved through ± 45° (or in one var. more than 90°), 6.7-17.5 mm. long; wings usually shorter, the blades oblong-oblanceolate, obtuse, truncate, or rarely emarginate, straight or slightly incurved; keel 7-13 mm. long, usually shorter than wings or banner (in one var. surpassing the banner), the blades lunately half-elliptic, rarely oblong or half-oblanceolate, gently or abruptly incurved to the bluntly rounded apex; anthers 0.45—0.8 mm. long; pod pendulous, stipitate, the slender, straight stipe ± as long or a little longer than the calyx-tube, the body linear-ellipsoid, ellipsoid, or oblong-ellipsoid, straight or a trifle in- or decurved, 6.5—20 mm. long, 2—4.5 mm. in diameter, usually cuneate but sometimes acuminate or abruptly contracted at either end, cuspidate at apex, obcompressed, with low-convex dorsal face and filiform dorsal suture, the ventral face flattened and excavated lengthwise along both sides of the elevated and ± thickened suture as two deep and narrow, or (when fully mature) shallow and open grooves, the thin, green, strigulose or glabrous valves becoming thinly leathery or papery, stramineous, either nearly smooth, reticulate, or cross-rugulose, not inflexed; ovules 5-15; seeds brown, smooth but dull, 2.5-3.4 mm. long.

The concept of A. bisulcatus as a single polymorphic or pluriracial species equivalent (always excepting the very distinct A. oocalycis) to the whole genus Diholcos as elaborated by Rydberg, or a subgenus of the same name as treated by C. L. Porter (1939), was foreshadowed by Jones as early as 1891 (in Zoë 2: 240). Rydberg’s microspecies are examined in detail below; I am concerned here only with some aspects of the latter revision, with which the present account is most likely to be compared. Porter maintained in this immediate group three species, A. bisulcatus, A. Haydenianus, and A. haydenioides, which he defined principally in terms of calyx- and pod-measurements, and the presence or absence of transverse nervature or wrinkling of the ripe pod-wall. Smooth and relatively large pods (the body 12-20 mm. long) together with large flowers (calyx-tube 4-6 mm., corolla 12-20 mm. long) were attributed to A. bisulcatus, and to the other two a shorter pod (the body 8-10 mm.) and small flowers (calyx-tube 2-3 mm., corolla 12 mm. or less long). In the small-flowered group A. Haydenianus was said to differ from A. haydenioides by its comparatively large flower (8-12 mm., not 6-7 mm. long) and cross-reticulate as opposed to smooth pod. The criteria are often valid; but too many individual plants and populations of like plants, combining characters of more than one of Porter’s species, have come to my attention. Some specimens, for example, from northern New Mexico which could confidently be identified as A. bisulcatus because of their large flowers have small, smooth pods less than 1 cm. long. Similarly the form from southern Utah described by Jones as A. Haydenianus var. major, with flowers about 12 mm. long (on the cusp between A. Haydendianus and A. bisulcatus), often present fruits over 1 cm. long with only faintly reticulate valves. The flowers and fruits of the disjunct var. nevadensis fall most nearly within the range of variation assigned to A. Haydenianus, but the small pod is often nearly smooth. Yet again, plants from the Gunnison Valley in Colorado combine the mediumsized flower attributed to A. Haydenianus with a perfectly smooth pod up to 13 mm. long. Enough has been said to illustrate the absence of correlation in size of flower and fruit which Porter assumed to exist in this group. Falling back finally on a rugulose-reticulate pod as a firm criterion for A. Haydenianus, we encounter new difficulties insofar as a pronounced reticulate condition is not correlated with definite pod- or flower-size. In fact no really sharp discontinuity in variation has been found in any one differential character, so it seems desirable to treat the complex as a single species. I might add that the key offered by Jones (1923, p. 244) to discriminate between A. bisulcatus and A. Haydenianus is irrelevant to perhaps half the material of the group now available.

Among papers found after Sereno Watson’s death was the description of a new astragalus, A. grallator, so named because of the greatly elongated erect pedicels, fancifully reminiscent of stilts. The typus was collected by Alice Eastwood from a colony of A. bisulcatus var. Haydenianus at Steamboat Springs, Colorado, and has been generally and beyond any doubt correctly interpreted as an abnormal state of the latter. A similar form of var. nevadensis was found in northeastern Nevada by Torrey (NY) as early as 1865; and one of var. bisulcatus (Ripley & Barneby 8310, CAS, RSA) in central Colorado, this also associated with normal plants. In the last instance the ovaries and pollen seemed to be normally formed, and there was no obvious sign of insect injury. In the extreme state the altered pedicels may reach a length of 17 mm. and bear a small, amorphously modified, usually sterile pod erect in the same plane. The teratological grallator form is not confined to sect. Bisulcati, having been noted in some Megacarpi, and once in A. alpinus. Its cause remains to be investigated.