Astragalus purshii var. tinctus

Dry gravelly or sandy hillsides, plains and mesas, nearly always among sagebrush, most commonly on basaltic or granitic bedrock, 3750-8000 (9500) feet and descending along the Klamath River into pine forest as low as 1500 feet, widespread and locally common along the east base of the Sierra Nevada from northern Inyo County, California, north through western Nevada to Lake, Harney, Deschutes, and Wheeler Counties, Oregon; west in northern California from near Truckee Pass to the foothills of Mt. Shasta, the Klamath Basin, and south in the inner North Coast Range to Anthony Peak, Mendocino County; also in scattered stations through the higher ranges of the northern and western Mohave Desert, from the Death Valley region to the Granite, Ord, and east slope of the Tehachapi Mountains, thence west and northwest through the southern Sierra Nevada (Greenhorn Range; Baker Ridge) to Mt. Pinos and the inner South Coast Range in San Benito County; apparently isolated on the Wallowa Mountains, northeastern Oregon.—Map No. 87.—April to June (July), rarely again in fall.

Astragalus Purshii var. tinctus (dyed, of the purple flowers) Jones in Zoe 4: 269. 1893. —"Edgewood near Mt. Shasta and also in Ventura County, Cal., Brandegee; Olancha and Keeler, Inyo County, Cal., Brandegee; the former also by Miss Eastwood, Soda Spring, Nevada County, Cal., 1882, Jones..."—Lectotypus (Barneby, 1947, p. 502), collected at Soda Springs, Nevada County, California, Jones in 1881, POM!—The paratypi represent various minor variants mentioned below.—Phaca Purshii var. tincta (Jones) Piper in Contrib. U. S. Nat. Herb. 11 (Fl. Wash.): 369. 1906.

Astragalus Purshii var. longilobus (with long calyx-teeth) Jones in Zoe 4: 269. 1893.— ‘Tehachapi, June, 1884, Brandegee; Aurum, Nevada, May 4, 1893, Jones (not in fruit) ... "— No specimen so labeled found at CAS, POM, UC, the only spm. at UC so annotated by Jones, Brandegee from Reno, Nevada = var. Purshii. Lectotypus, collected at Tehachapi, without date or number, K. Brandegee (UC, 2 sheets); Jones’s collection from Aurum likewise = var. Purshii.

Astragalus consectus (cut up, the application obscure) Sheld. in Minn. Bot. Stud. 1: 143. 1894.—"Collected in California, 1846, Fremont; Carson Valley, Utah, 1859, Engelmann; Utah, 1874, C. C. Parry; and at Tejon Pass, southern California, June, 1887, S. B. Parish ... The last specimen ... may be taken as the type."—Holotypus, Parish 1890, MINN! isotypus, NY!—The paratypi include elements of A. utahensis and A. Newberryi.—Xylophacos consectus (Sheld.) Rydb. in Bull. Torr. Club 40: 49. 1913. A. leucolobus subsp. consectus (Sheld.) Abrams, I11.. Fl. Pac. St. 2: 578, fig. 2809. 1944.

Astragalus candelarius (of Candelaria) Sheld. in Minn. Bot. Stud. 1: 142. 1894.— near Candelaria, Esmeralda [now Mineral] County, Nevada, April and May, 1888, by W. H. Shockley."—Holotypus, Shockley 33, MINN! isotypi, ND, NY, US!—Xylophacos candelarius (Sheld.) Rydb. in Bull. Torr. Club 52: 370. 1925.

Astragalus candelarius var. exiguus (slight and small) Sheld. in Minn. Bot. Stud. 1: 143. 1894.—"... beside the road to Sierra Valley, Nevada county, California, May, 1886, by C. F. Sonne; also near Candelaria, Esmeralda county, Nevada, by W. H. Shockley."—Cotypi, Sonne 66, collected 23 May, 1886, and Shockley 33 in part, collected in May, 1886. MINN! '

Astragalus inflexus var. flocculatus (beset with flocs of wool) Jeps., Man. Calif. 573, fig. 564. 1925.—"Panamint Mts., Hanaupah Canon, Jepson 6999, type."—Holotypus collected May 9, 1917, JEPS!

Astragalus inflexus var. ordensis (of Ord Mountain) Jeps., Man. Calif. 573. 1925.—"Ord Mt., Mohave Desert (Jepson 5877, type)."—Holotypus, collected May 3, 1914, JEPS!—A. Purshii var. ordensis (Jeps.) Jeps., Fl. Calif. 2: 360. 1936.

Xylophacos subvillosus (somewhat villous, less so than A. utahensis) Rydb. in Bull. Torr. Club 52: 368. 1925.—"...in Sierra Valley, California, in 1873; Lemmon (herb. N. Y. Bot. Gard.)."—Holotypus, NY! isotypus, GH!

Astragalus Purshii var. gavisus (delighted, meaning delightful) Jeps., Fl. Calif. 2: 360. 1936.—"Scott River Valley, Siskiyou Co., T. H. Gilbert (type)."—Holotypus, collected June 15-July 15, 1899, UC!

The concept and circumscription of var. tinctus, as defined in these pages, are substantially more inclusive than those adopted in the preliminary revision (Barneby, 1947, p. 500, sequ.), and the range of the variety is greatly expanded in consequence. It now embraces the relatively small-flowered populations of northeastern California formerly referred to var. lectulus, as well as var. longilobus, the Pursh milk-vetch of the Mohave Desert and South Coast Range, and the formerly obscure var. gavisus. Material acquired during the past decade has demonstrated complete intergradation in flower-size between the so-called var. lectulus of the North Coast Range in Mendocino and Siskiyou Counties, California, and the lowland, mostly larger- flowered typical var. tinctus prevalent in the Klamath Basin. Furthermore, small-flowered plants, although usually less compact in growth-habit than those from the crest of the Coast Range, are known to occur with some frequency on the sagebrush plains of Modoc County and of adjoining Oregon and Nevada. It seems more natural to associate these sympatric forms under one name than to abstract the small-flowered extreme and attach this arbitrarily to the otherwise more southern and montane var. lectulus. Deprived of these northern elements, var. lectulus can be redefined as endemic to the Sierra Nevada and San Bernardino Mountains. The disadvantage of this rearrangement is a continuous series of flower-sizes from the smallest in var. lectulus to the largest in var. tinctus; almost nothing remains to distinguish two varieties except that in the northern populations, which most nearly approach genuine var. lectulus in length of the petals, the calyx-teeth are longer than in the Sierran plant. There can be no doubt that var. lectulus has been derived from var. tinctus, but it has achieved a more compact and better-defined range than the incipiently small-flowered, parallel development in the North Coast Range. It is interesting to find that a second small-flowered form simulating var. lectulus but with slightly larger flowers (keel ±11 mm. long) has developed in isolation high in the Wallowa Mountains (Ice Lake, Peck 18,485; Aneroid Lake, Peck 17,976, both WILLU) in northeastern Oregon.

Little can be added to what was said (Barneby, 1947, p. 508) concerning variation in var. longilobus, herein reduced to var. tinctus. The length of the peduncles and number of leaflets, formerly thought to be good differential characters, have not proved to be very constantly associated within a natural area of dispersal. Typical var. tinctus in southwestern Nevada passes insensibly into the desert form (var. flocculatus and var. ordensis), and this in turn, in the Tehachapi Mountains, passes into the matted state (A. consectus) prevalent in the South Coast Range from Mt. Pinos northward.

The typus of var. gavisus, tentatively referred in my revision to var. lectulus, has turned out to be a large-flowered form (calyx ± 14 mm., keel almost 17 mm. long) with obovate-cuneate leaflets. It represents an ecotype of var. tinctus found along the Scott and adjoining Klamath Rivers at elevations (low for the species) of about 1500—1800 feet and distinguished by a loosely matted, caulescent habit of growth.

The diverse climates and environments tolerated by var. tinctus doubtless favor the process of differentiation into geographic races. The more notable trends, readily perceived but not at present usefully definable in taxonomic terms, may be described as minor variants:

M. v. 1. Leaflets mostly 5-13, either acute or obtuse; flower large, the banner over 18 mm. long; peduncles shorter than the leaves (var. tinctus, sens. str.)—Nevada and northeastern California.

M. v. 2. Similar, but the foliage very densely pannose-tomentose; calyx-teeth unusually short, triangular.—Typically developed in sinter gravel about hot springs in southern Washoe County, Nevada, but forms transitional to typical var. tinctus are common in the region. Specimens of this variant were identified by Jones (in Zoë 4: 270) and by Rydberg (in Bull. Torr. Club 52: 366) as A. utahensis, but have the characteristic pod of A. Purshii.

M. v. 3. Similar to var. tinctus, sens. str., but flower smaller (banner 14.5-18 mm. long). —Klamath Basin, California, and adjoining Nevada and Oregon, mostly at middle elevations but reappearing on the crest of the North Coast Range from the Marble Mountains south to Mendocino Pass.

M. v. 4. Apparently the same as the montane phase of the last, but isolated on the Wallowa Mountains, northeastern Oregon.—Little known and to be studied further.

M. v. 5. Acaulescent or nearly so, densely long-villous; leaflets mostly 11-17; peduncles commonly equaling or surpassing the leaves; flowers of var. tinctus; pod often, but not invariably, semibilocular toward the base, the septum 0.5-1.4 mm. wide.—Mohave Desert, Inyo, northern San Bernardino and eastern Kern Counties, California (var. ordensis and var. flocculatus Jeps.).

M. v. 6. Commonly caulescent, loosely matted, the stems up to 14 cm. long; leaflets mostly 11-17; peduncles usually shorter than the leaves; petals commonly, but not invariably, little graduated, the flower closely resembling that of the related and vicariant A. leucolobus but the pod unilocular and long-hirsute.—Inner South Coast Ranges, California, San Benito County south to the head of San Joaquin Valley (A. consectus Sheld.).