Astragalus Purshii var. glareosus

Habit and vesture of var. Purshii or var. tinctus, either subacaulescent or shortly caulescent, the stems up to 5 cm. long; leaves (2) 3-12 cm. long, with (7) 9—17 (21) nearly always narrowly elliptic and acute or short-acuminate, rarely oval-oblanceolate leaflets (2) 4-14 (17) mm. long; racemes 2-5 (7)-flowered; calyx 12-16 (18.7) mm. long, the tube (8.5) 9.5-12 (13) mm. long, 3.6-4.8 (5) mm. in diameter, the linear-lanceolate or subulate teeth 2.3—4.8 mm. long; petals purple, sometimes pale; banner 19-25 (26.5) mm. long, 7-10.5 (12) mm. wide; wings 19-24.5 mm. long, the claws 10-14.8 mm., the blades 8.8-11.3 (13) mm. long, (1.7) 2-3 mm. wide; keel 16.7-21 (22.5) mm. long, the claws 9.4-13.5 (14.5) mm., the blades 6.4-8.4 (10) mm. long, 2.3-3.5 mm. wide; pod obliquely ovoid, ovoid-ellipsoid, or lance-ellipsoid, lunately incurved or sometimes abruptly hooked, (1.3) 1.5-2.6 (3) cm. long, 7-11 mm. in diameter, strongly depressed-sulcate in the lower ½, the sutures contiguous or approximate within, the valves villous-hirsute with hairs up to (1.2) 1.5-4 mm. long; ovules (17) 22-30 (35); seeds (1.8) 2-3.7 mm. long.—Collections: 131 (xi); representative: J. W. Thompson 11,499 (CAS, GH, NY, POM, RSA, WS, WTU); Cronquist 5648 (ID, RSA, SMU, WS), 7040 (NY, RSA); Ownbey & Ward 3108 (ID, RSA, SMU, TEX, WS); Maguire & Holmgren 26,278, 26,302 (NY, RSA, UTC).

Dry sandy and gravelly plains, foothills, bluffs, and canyon benches, nearly always among sagebrush, but ascending into openings in yellow pine forest, commonly on basaltic bedrock but also on granites and various alluvia, 200-5000 (6650) feet, widespread and locally abundant in the Columbia Basin, from southern British Columbia, where apparently localized in the Kamloops area and along the Okanogan River, through transmontane Washington to northern Oregon, south and southeast through the Blue Mountains and up the Snake River canyon into southeastern Oregon, southern Idaho, and rarely into extreme northwestern Utah and northeastern Nevada.—Map No. 87.—April to June, exceptionally again in fall.

Astragalus Purshii var. glareosus (Dougl.) Barneby in Amer. Midl. Nat. 37: 503. 1947, based on A. glareosus (of gravelly places) Dougl. ex Hook., Fl. Bor.-Amer. 1: 152. 1831.— "Plentiful on dry gravelly banks of rivers, from the confluence of Lewis and Clarke’s River with the Columbia to the mountains. Douglas."—Holotypus, from "the barren, sandy grounds," K! isotypi ("N.-W. America"), GH (fragm.), P!—Tragacantha glareosa (Dougl.) O. Kze., Rev. Gen. 945. 1891. Astragalus inflexus var. glareosus (Dougl.) Jones, Contrib. West. Bot. 10: 62, PI. 5. 1902. Phaca glareosa (Dougl.) Piper in Contrib. U. S. Nat. Herb. 11 (Fl. Wash.): 369. 1906. Xylophacos glareosus (Dougl.) Rydb., Fl. Rocky Mts. 1063. 1917.

Astragalus allanaris (white-woolly) Sheld. in Minn. Bot. Stud. 1: 141. 1894.—" ... Rattlesnake mountain, Yakima County, Washington, June, 1884, also near Walula, Walla County, Washington, April, 1891, by Mr. W. H. Suksdorf."—Lectotypus, Suksdorf 1811 from Rattlesnake Mountain, MINN! isotypus, WS! paratypus, Suksdorf 2004, MINN!

Astragalus lanocarpus (with woolly pod, the fruit likened to the oak-galls of Andricus lanae) Sheld. in Minn. Bot. Stud. 1: 144. 1894.—"Klickitat Prairie, Washington, June, 1880, by Thomas J. Howell; also at Reno, Nevada."—Lectotypus, Howell, MINN! isotypus, NY! the paratypus from Reno = var. tinctus, MINN!—A. leucocystis (with white bladder) Greene in Erythea 3: 76. 1895, a superfluous substitute.

 Astragalus Booneanus (Rev. William Judson Boone, 1860—1936, founder and president of College of Idaho, Caldwell) A. Nels. in Bot. Gaz. 53: 224. 1912.—“The species rests upon several very representative collections as follows: President W. J. Boone... at Caldwell, no. 2, ... C. N. Woods,... Hailey, nos. 5 and 25a; Merrill and Wilcox, Leckie, Wyo., no. 583; and J. Francis Macbride, Falk’s Store, Idaho, no. 57.”—Lectotypus, Boone 2, collected in May, 1910, RM!—The paratypi (RM) are conspecific with the exception of Merrill & Wilcox 583, which = A. argophyllus var. Martini.

(?) Xylophacos ventosus (windy, the type from "windy, rocky places”) Rydb. in Bull. Torr. Club 52: 370. 1925.”—‘ ...several kilos east of Bingen [Washington], Nov. 10, 1920, Suksdorf 10662."—Holotypus, NY! isotypus, WS!

The gravel milk-vetch, A. Purshii var. glareosus, is the common representative of its species at low elevations in the Columbia Basin, whence it extends in a southeasterly direction through the Blue Mountain system and reappears in abundance in the foothills enclosing the Snake River Plains. Its principal features are the narrow, acute leaflets, large and few, ordinarily bright purple flowers, and ventrally grooved legume. The pod’s compression, stressed in the key as the principal differential character separating var. glareosus from the more southern var. tinctus (as well as, in the fruiting condition, from the sympatric but white-flowered var. Purshii), needs observation in the field by local botanists. The flowering season of the Pursh milk-vetch is early and short, whereas the pod matures over a period of several weeks, and it is uncommon to find fresh flowers and ripe fruits simultaneously in a colony. As a result, there is too little correlation between the form of the pod and the color of the petals. The outlying records of var. glareosus from Nevada, Utah, and southeastern Idaho are based on mature fruiting specimens, and proof that these populations are purple-flowered is still to be obtained. Conversely, a number of fruiting plants from the Columbia Basin have been identified as var. Purshii because the pod lacked a ventral groove, but these plants are in other respects closely similar to sympatric var. glareosus known to have purple petals. The dispersal of the two varieties needs accurate local mapping.

The typus of Xylophacos ventosus was collected in November and is, perhaps, not characteristic of its sort. It was interpreted (Barneby, 1947, p. 500) as representing an outlying population of var. tinctus, but this has not been traced farther north than the headwaters of the Deschutes River, at points a hundred miles south of the Columbia Gap. Other material from low elevations along the Columbia is too scanty to provide a clear picture of X. ventosus; it is suspected that it belongs with var. glareosus rather than var. tinctus.

In eastern Oregon, especially along the upper Malheur River, a form of var. glareosus occurs in which the pod is only shortly and relatively thinly villous, the longest hairs only up to ± 1.2 mm. long. In the extreme state the vesture of the fruit is hardly denser than that found in the more loosely hirsute states of A. argophyllus var. Martini, from which, however, the cottony leaf-pubescence readily distinguishes it. Plants intermediate between this extreme and typical var. glareosus are common in the region (cf. Ripley & Barneby 6110, 6117, RSA).

Occasional specimens of the gravel milk-vetch, especially a shortly caulescent phase found along the upper reaches of the Grand Canyon of the Snake River, superficially resemble and have been confused with A. inflexus. The following contrasts may forestall confusion in the future:

1. Stems 0-5 cm. long; leaves all petioled, with (7) 9-17 (21) leaflets; racemes 2-5 (7)-flowered, the axis 0.5-2 cm. long in fruit....................A. Purshii var. glareosus

1. Stems (5) 10-35 (50) cm. long; upper leaves subsessile, with (9) 17-23 (27) leaflets; racemes (5) 8-18-flowered, the axis (1.5) 2.5-6 (8) cm. long in fruit----A. inflexus