Astragalus tennesseensis

Low, usually coarse and robust, with a thick taproot and knotty root-crown, hirsute nearly throughout with fine, straight, lustrous, spreading or feebly retrorse, spirally twisted hairs up to 2-3.5 mm. long, the stems densely so and shaggy, the leaflets more thinly so beneath, glabrous above, ciliate, the herbage green; stems several or numerous, decumbent and ascending, (0.4) 1—5 dm. long, arising together in well-furnished, leafy clumps from a point at or just below soil-level, simple or spurred at 1-3 nodes preceding the first peduncle; stipules mostly very large and conspicuous, pale green, submembranous, becoming papery-scarious in age, several-nerved, 0.5—2 cm. long, the lowest ones often (but not always) amplexicaul and shortly connate, the median ones (or all) asymmetrically ovate- acuminate and broader than the stem but not fully amplexicaul, the uppermost often lance-acuminate, all hirsute-ciliate; leaves (6) 8—15 cm. long, shortly petioled or the uppermost subsessile, with (15) 23—33 oblong-elliptic, obtuse, emarginate, or mucronulate-acute, or (in some lower leaves) obovate-obcordate, thin-textured, flat leaflets (4) 6-23 mm. long; peduncles incurved-ascending, 5-10 (14) cm. long, mostly shorter than the leaf, becoming stouter and decumbent in fruit, racemes rather compactly 9—20-flowered, the flowers at first ascending, becoming horizontal or a trifle declined, the axis little elongating, 2—5 (6) cm. long in fruit, bracts resembling the stipules in texture, narrowly to broadly lanceolate, lance- acuminate, or subulate-caudate, 5—12 mm. long, spreading or deflexed in age, pedicels ascending, straight or a trifle arched outward, at anthesis slender, 1.5—2.6 mm long, in fruit thickened, 2—3.5 mm. long; bracteoles 0; calyx 8.3—15.1 mm. long, thinly hirsute with white or sometimes a few fuscous hairs, the strongly oblique disc 1.2-1.7 mm. deep, the membranous, pallid tube 5.9-11.4 mm. long, 4.2—4.7 mm in diameter, gibbous-saccate at base behind the pedicel, this inserted laterally at or above the end, the lance-subulate, thinly herbaceous teeth 2.4—4 mm long, the whole becoming papery, ruptured, marcescent, petals whitish or pale cream-color, immaculate, sometimes drying yellow; banner oblanceolate, 14-19 mm. long, (5.5) 6.7-8.6 mm. wide, somewhat sigmoidally arched, the claw incurved to conform with the calyx, the blade recurved through 40 or ultimately further; wings 13-16 mm. long, the claws 7-8.6 mm, the oblong or oblong- oblanceolate, nearly straight, shallowly 2-lobed or emarginate blades 7.2—9 mm. long, 2.8-3.5 mm. wide; keel 11-13.6 mm. long, the claws 6.7-8.1 mm., the broadly triangular-obovate blades 5-6.3 mm. long, 3.2-3.7 mm. wide, abruptly incurved through 95-100° to the rounded apex; anthers 0.6-0.85 (0.9) mm. long; pod ascending (nearly always humistrate), sessile on a stout, glabrous gynophore (1) 1.3-2 mm. long, the body oblong or lance-elliptic in profile, nearly straight or gently incurved, 2.5-4 cm. long, 7-13 mm. in diameter, rounded at base, acuminate distally, slightly obcompressed and shallowly sulcate along both sutures, the fleshy, thinly hirsute valves when first formed either green or pink-tinged, somewhat transparent, 2.5-3 mm. thick, in ripening becoming spongy, stramineous, and by collapse of the mesocarp onto a thick, at first immersed reticulate nervation becoming alveolate-rugulose when fully ripe, inflexed as a complete septum 3-4 mm. wide, the funicular flange well developed, 1.7—2.1 mm. wide; seeds indigo- green or black, smooth but dull, 2.8-3.1 mm. long.—Collections: 18 (iii); representative: Gattinger distrib. Curtiss 593 (NY, SMU); Ripley & Barneby 11,037 (CAS, NY, RSA, US, WS), 11,039 (CAS, NY, RSA); E. J. Palmer 35,493 (NY); Rollins & al. 6017 (GH, NY); Umbach (from Utica, Illinois) in 1901 (WIS).

Openings in cedar barrens and at the edge of thickets, on limestone pavement or stony calcareous hillsides, locally plentiful at about 750-850 feet, in central Tennessee (Davidson, Rutherford, and Wilson Counties) and adjoining Alabama (Morgan and Lauderdale Counties), also in unrecorded habitats (but sometimes along railroad rights-of-way) along the upper Illinois and Rock Rivers in north- central Illinois (Ogle, LaSalle, and Grundy Counties); reported, probably erroneously, from Iowa and Missouri.—Map No. 103.—April to June, the fruit ripening slowly and often dehiscent only after winter weathering.

Astragalus tennesseensis (of Tennessee) Gray ap. Chapman, Fl. S. U. S. 98. 1860 ("Tennesseensis’).—"Hills near Nashville, Tennessee, Lesquereux, and Lagrange, Alabama, Prof. Hatch."—Holotypus, Lesquereux in 1855, GH! paratypi (Hatch), GH, NY (fragm.)!— A. plattensis var. tennesseensis (Gray) Gray in Proc. Amer. Acad. 6: 193. 1864. Geoprumnon tennesseense (Gray) Rydb. ap. Small, Fl. S.E. U. S. 1332. 1903.

The Tennessee milk-vetch is a handsome astragalus of strong, leafy growth, easily recognized as the only hirsute species of the genus native to states lying between the Appalachians and the Mississippi River, and well characterized by its pithy, curiously honeycomb-sculptured, thinly long-hairy fruit. The flowers have been described as yellow from dryed specimens, but in central Tennessee at least the petals when fresh are ochrolucous or almost white. The species is quite local, but the precise limits of its dispersal and the delimiting factors are not fully understood. In Tennessee and probably in Alabama, it is restricted to the calcareous cedar glades, but its habitat in Illinois has not been reported. The Lagrange in Alabama, where Prof. Hatch collected the species a century ago, has not been identified, and the map is incomplete in this respect. Fernald (1950, p. 942) reported the Tennessee milk-vetch from southern Illinois, but this was probably a mistake, for all collections associated with credible labels originated from the Illinois and Rock River Valleys in the northcentral part of the state, a situation corroborated by the modern state floras and by Welsh (1960, Map 18). The range-extension to "southeastern Iowa... and Missouri," given by Jones (1923, p. 238), has not been taken up subsequently and is certainly based on error. Granted that the unidentified Alabama station marks at best a small southward extension of the principal cedar glade area, it appears that the dispersal of A. tennesseensis is bicentric, and it is even possible that the populations in the two areas are not identical. The flowers in Illinois seem to be smaller than those in Tennessee, but the point requires confirmation from more ample material. A specimen from Ottawa, Illinois (Bebb in 1897, NY), distinguished by unusually narrow stipules, is apparently no more than a minor variant already mentioned as such by Rydberg (1926, p. 164).