Astragalus robbinsii (Oakes) A.Gray
-
Authors
Rupert C. Barneby
-
Authority
Barneby, Rupert C. 1964. Atlas of North American Astragalus. Mem. New York Bot. Gard. 13(1): 1-596.
-
Family
Fabaceae
-
Scientific Name
-
Type
"On rocky ledges, overflowed in the spring, on the banks of the Onion river, Burlington, Vermont. Dr. Robbins, 1829."—Spma. authent., Oakes in May, 1840, GH! and labeled "Burlington, Vt., Robbins, August, 1829" ex herb. Oakes., NY! The spm. from Burlingto
-
Description
Species Description - Variable in stature but usually slender, very thinly to quite densely strigulose, strigulose- or pilose-villosulous with straight and appressed, incumbent, or more rarely spreading hairs up to 0.2-0.8 mm. long, the herbage usually green, the leaflets of membranous texture, when large visibly penninerved, bicolored, pallid and usually pubescent beneath, darker green and glabrous or glabrescent above, the upper surface (when dried) often appearing minutely whitish-tuberculate, the inflorescence commonly but not consistently black-pubescent; stems usually several, erect, incurved-ascending, or (in exposed sites) diffuse or even trailing, arising together from the root-crown or shortly forking caudex, simple, (0.7) 1-4.5 (6) dm. long, at base leafless and commonly purplish, the lowest internodes short or inhibited, the upper 5-9 developed; stipules 2-8 mm. long, dimorphic, the lowest (approximate or loosely imbricated) early becoming papery and brownish, several- nerved, strongly adnate to the suppressed petiole, broadly ovate, obtuse, fully amplexicaul and commonly broader than the stem, either free or united through half their length, the upper ones progressively narrower, ovate or lanceolate, more rarely oblong or triangular-acuminate, thinly herbaceous or submembranous, often purplish, semi- or fully amplexicaul, free, the blades often deflexed, glabrous or glabrate dorsally, thinly ciliate; leaves 3-10 (12) cm. long, the lowest often slender-petioled, the rest (sometimes all) subsessile, with (5) 7-17 lance-oblong, oblong-elliptic, ovate-oblong, ovate, or obovate, obtuse or emarginate, exceptionally subacute, flat leaflets (3) 5-32 mm. long; peduncles incurved-ascending or erect from ± 2-3 distal axils, (3.5) 4-21 cm. long, nearly always surpassing the leaf; racemes (3) 6-21 (25) -flowered, often moderately compact at early anthesis, nearly always moderately or greatly elongating, sometimes loose from the first, the flowers early spreading and ultimately declined, the fruits loosely secund, the axis (1) 1.5-18 cm. long in fruit; bracts submembranous, pallid or purplish, narrowly ovate to linear-oblong or -lanceolate, (1) 1.5-5.5 mm. long; pedicels slender, at anthesis ascending or arcuately spreading, 0.9-2.5 mm. long, in fruit recurved, geniculate at base and horizontally spreading, abruptly dejected, or (from decumbent peduncles) apparently ascending, a trifle thickened, 1.5-5 mm. long; bracteoles usually 0, minute when present, sometimes attached well below the calyx; calyx (3.4) 3.7-8 mm. long, thinly strigulose to densely strigose-villosulous with black or fuscous, commonly mixed with a few pale, more rarely with nearly all or all white hairs, the subsymmetric disc 0.7-1.1 (1.5) mm. deep, the campanulate, pallid or more often purple- or red-tinged tube (2.4) 2.6—4.5 mm. long, 1.8-3 (3.4) mm. in diameter, the subulate, more rarely linear- or triangular- subulate dorsal and lateral teeth 0.8-3.4 mm. long, the ventral pair often broader and shorter, the whole becoming papery, marcescent unruptured; petals pale purple with whitish claws, pale slate-blue, or more rarely whitish, but the keel-tip then usually pink- or purple-maculate; banner gently recurved through ± 40° broadly oblanceolate to obovate-cuneate, shallowly or deeply notched, 7.2-11.5 mm. long, 3.8-7 mm. wide; wings 6.1-9.3 mm. long, the claws 2.8-4.2 mm., the narrowly oblong or oblong-oblanceolate, nearly straight, obtuse, truncate, or shallowly (deeply) emarginate blades (4) 4.3-7 mm. long, 1.4-2.2 mm. wide; keel 5.4-7.9 mm. long, the claws 2.7-4.2 mm., the half-obovate blades 3.1-4.8 mm. long, 1.7-2.5 mm. wide, abruptly incurved through 90-105 (110)° to the bluntly rounded (exceptionally subporrect and then deltoid) apex; anthers 0.3-0.5 mm. long; pod essentially pendulous (sometimes spreading from horizontal peduncles), stipitate, the slender stipe usually 1.5-5.5 mm. (in western United States sometimes only 0.5-1.5 mm.) long, the subsymmetrically ellipsoid, half-ellipsoid, oblong- or rarely lance-ellopsoid, straight or very slightly decurved body 8-25 mm. long, (3) 3.5—5.5 (6) mm. in diameter, cuneately or acuminately tapering into the stipe, contracted distally into a conic-subulate or subulate, cusplike beak 0.5-3 mm. long, obscurely triquetrous, with the three faces almost equally broad or the low-convex to depressed dorsal face commonly a little narrower than the lateral ones, the ventral angle carinate by the filiform, convexly arched suture, the thin, green or purplish valves becoming papery-membranous and delicately cross-reticulate, minutely strigulose to densely pilosulous with black or fuscous, commonly mixed with a few (many, rarely all) white hairs up to 0.2—0.5 mm. long, inflexed as a hyaline septum 0.2—1.6 (2) mm. wide; ovules 7—11, rarely 3—6; seeds brown, sometimes purple-speckled, smooth and somewhat lustrous, 1.4-2.3 mm. long.
-
Discussion
The Robbins milk-vetch is a collective species, with two centers of dispersal, one Cordilleran from Colorado to eastern Alaska, the other extending from New England into eastern Canada. The dispersal of the species is like that of many boreal-montane phanerogams which bore the brunt of the Tertiary glaciations. In each of the main areas it is represented by three virtually monomorphic and probably genetically fixed varieties confined to a narrow ecological niche, but there is one relatively widespread and variable entity common to both areas, and it is this var. minor which forms the nucleus of the species. The Robbins milk-vetch has been known for well over a century, but the racial situation was only investigated comparatively recently, a beginning being made (if we discount the discovery of var. occidentalis) in the nineties by W. W. Eggleston in New England. Between the years 1923 and 1933 the results of studies undertaken by three botanists of widely divergent philosophy appeared in print, and it is interesting to compare them. The account of Jones (1923, p. 133-5), whose inclusive concept of A. Robbinsii is adopted in these pages, is the least critical in detail. Jones mistakenly took up for the species the epithet labradoricus properly belonging (although a misnomer) to A. alpinus var. Brunetianus, and established three varieties: a typical one ( = our vars. Jesupi, Fernaldi and the New England aspect of var. minor); a var. Robbinsii ( = our var. Robbinsii with an admixture of var. minor and the latter’s range in the East); and a var. occidentalis (= our var. minor in its western aspects, var. occidentalis sens, str., and var. alpiniformis). Rydberg (1928, pp. 124—9) described nine species of Atelophragma covering the same range of variation, four of them new, and divided them between two sections, Robbinsiana and Oroboidea, distinguished simply by the length of the stipe, an example of how far one can be led astray by over emphasis of a single character. Finally Rousseau (1933, passim, in relation to A. alpinus and A. Fernaldi in Quebec) published notes on the eastern as well as the Rocky Mountain components of the complex, accepting them in Astragalus at Rydberg’s specific level. I must reluctantly remark that the drawings of floral parts and fruits in Rousseau’s figure (op. cit. fig. 8) are at best no more than idealized average examples; and since no attempt is made to illustrate variation, an altogether false picture of the so-called species is drawn. Rousseau’s table of relationships (op. cit. fig. 9), which outlines a phylogenetic scheme for our sects. Cenantrum, Astragalus, Oroboidei, and Hemiphragmium, is open to criticism not only on theoretical grounds, but also because it presents a narrow provincial view of a problem involving four or five times as many species in Asia and several groups not even represented in America. The Alaskan relatives of A. Robbinsii have been elaborated more recently by Hulten (1944). In the present account the species is considered to consist of seven varieties, one of them extinct.
In all the herbaria that I have consulted I have found A. Robbinsii confused with A. alpinus. The two species are much alike in coloring of the leaves and flowers, in the often dark- hairy inflorescences and fruits, and in their preference for streamsides and moist places, where they are often found together. The following contrasts should serve to forestall confusion in the future:
1. Petals regularly graduated, the keel shorter and little or no wider than the wings, much shorter than the banner; stems arising together from the root-crown at soil- level; leaflets 7-17; pod obscurely and bluntly trigonous, depressed or flattened but not or scarcely sulcate dorsally………A. robbinsii
1. Petals (in the range of A. robbinsii) irregularly graduated, the prominent keel longer or as long as and also wider than the wings, equaling or little shorter than the banner; stems arising singly or few together from buds on slender, subterranean caudex-branches; leaflets mostly 15-25; pod more sharply trigonous, strongly grooved dorsally………..A. alpinus
The species is also sometimes confused with forms of A. aboriginum, which has much the same wide and interrupted range; the latter is easily distinguished by its apically cleft and not obtuse or merely emarginate wing-petals and also by its usually more strongly compressed, decidedly two-sided pod.