Astragalus racemosus var. racemosus

Gullied bluffs, barren knolls, and alluvial bottomlands, on clay, shale or gypsum, 1700-4300 feet, locally abundant, southern Saskatchewan (local near Moose Jaw and on the headwaters of Qu’Appelle River) and North Dakota to western Oklahoma, the Texas Panhandle, and the Pecos Valley in eastern New Mexico, west just into eastern Wyoming and eastern Colorado, east to southeastern Nebraska and central Kansas; greatly isolated on dry lake beds at ± 6700 feet in western San Luis Potosi; reported from Minnesota (Fernald, 1950, p. 911, but no specimens at GH, MINN, or elsewhere).—Map No. 46.—May to July, flowering in late March southward.

Astragalus racemosus (with many flower-clusters) Pursh, Fl. Amer. Sept. 740. 1814.— "In Upper Louisiana. Bradbury."—Holotypus, labeled "Louisiana, Bradbury. Pursh’s specimens," PH! isotypus, dated 1811-12, BM!—A. galegoides (resembling Galega, or goat’s-rue) Nutt., Gen. 2: 100. 1818, technically an illegitimate substitute for the preceding. Tragacantha racemosa (Pursh) O. Kze., Rev. Gen. 947. 1891. Tium racemosum (Pursh) Rydb. in Bull. Torr. Club 32: 659. 1905. A. racemosus var. typicus C. L. Porter in Madrono 8: 99. 1945.

Astragalus racemosus var. brevisetus (with short bristles, in reference to the bracteoles) Jones in Proc. Calif. Acad. Sci. II, 5: 662. 1895.—"Ramos, Zacatecas, Mexico, May 5, 1892, Jones."—Holotypus, POM! isotypi, GH, NY, US!—Ramos is ± 40 miles e. of Ciudad Zacatecas within the state of San Luis Potosí.—Tium brevisetum (Jones) Rydb. in N. Amer. Fl. 24: 386. 1929.

The alkali milk-vetch, A. racemosus, is a handsome floriferous species, impossible to confuse with any astragalus of the Prairie States other than the white-flowered forms of A. bisulcatus and distinguished from these by its sharply three-angled and three-sided pod. It is particularly common along dry water courses and on banks of running streams and rivers, and is likely to be met with wherever an accident of the terrain or of erosion has exposed a seleniferous formation; and it becomes abundant where destruction of the sod has given rise to a badlands topography. It is thence carried down to the alluvial flats and bottomlands, but in such situations is less frequent and less vigorous than A. bisulcatus. Its primary natural habitat on the open prairie was apparently around calcareous or shaley outcrops.

The flowers of var. racemosus are ordinarily pure white or creamy white with pinkish or lilac keel-tip. In Kansas and Oklahoma, possibly elsewhere, it is common to find a color- form in which a blush of anthocyanin extends from the base of the banner-blade upward around its margins and into the wing-tips, or is concentrated into a purple keel-tip and a fan of reddish-purple veins in the fold of the banner. Possibly this color-form is commoner than has been realized, for the pigment fades rapidly in drying and disappears completely after a few months. The type-collection of var. brevisetus, according to Jones’s field note (NY, POM), was such a form. Despite its remote origin at a point in Mexico over seven hundred miles south of the alkali milk-vetch’s southernmost outposts in Texas or New Mexico, var. brevisetus cannot be distinguished from typical material from the United States. In the original publication Jones referred to his new variety, which he did not separate from typical A. racemosus in specific terms, specimens from the Cimarron Valley, Oklahoma (Carleton 221, NY, US) and from "Fremont," properly Frontier County, Nebraska (Rydberg 70, NY); and I agree that these are exactly conspecific. Rydberg maintained a purely Mexican Tium brevisetum, supposedly differing in its short calyx-teeth (1.5-2 as opposed to 2.5-4 mm. long) and thick- textured foliage, the latter being a feature characteristic of A. racemosus sens. lat. Only two collections of A. racemosus have been seen from south of the border, and the species was possibly introduced in San Luis Potosí.