Astragalus mollissimus var. irolanus

Loosely tufted, the stems 5-29 (30) cm. long, exceptionally subacaulescent, the herbage densely hirsute-tomentose with shorter, curly together with longer, spirally twisted hairs up to (0.7) 1-2 mm. long, the vesture becoming tawny in age; stipules 2-16 mm. long, deltoid or lance-acuminate; leaves (6) 10-20 (26) cm. long, with (13) 23-33 obovate to elliptic, obtuse or more rarely acute leaflets 6-25 mm. long; peduncles 6-18 (20) cm. long; racemes 10-25-flowered, the axis (1.5) 3-7 cm. long in fruit; calyx 9-13 mm. long, densely villous or villosulous like the herbage, the tube (5) 5.8-7.5 (8) mm. long, (2.5) 3-5 mm. in diameter, the teeth 3-6 (9) mm. long; petals ochroleucous with pale wing-tips and purple-tipped keel, or ± suffused with lilac-purple; banner (15.3) 16-23.5 mm. long, 7—9 mm. wide, sometimes subacute or mucronate in the apical notch; wings (14.5) 15.2-20.2 mm. long, the claws 6-7.6 mm., the blades 8.5-13.2 mm. long, 2—3.5 mm. wide; keel 11—13.6 mm. long, the claws 6—8 mm., the blades 5.5—7.6 (7.9) mm. long, 2.7—3.2 mm. wide; pod narrowly to broadly and plumply ovoid, (1) 1.1-2.3 (2.5) cm. long, 4-13 mm. in diameter, straight or somewhat incurved, the beak unilocular, the valves hirsute or hirsutulous with ± spreading hairs up to 0.5-1.2 mm. long; ovules 22-28.—Collections: 30 (o); representative; Palmer 14, 49 (NY, US); Johnston 7274 (US), 7469 (GH); Schneider 1006 (NY, US); Rose, Painter & Rose 9173 (US), 9756 (NY, US), 8810 (NY), US); Matuda 21,159 (MICH, NY), 18,538, 21,498 (NY); Pringle 8565 (DAV, NY, POM, US); Galeotti 3438 (G, P).

Plains, dry hillsides, fallow fields, roadsides, reaching 13,000 feet on Mt. Orizaba but mostly 5500-9000 feet, apparently common and locally plentiful, central Durango to southern Coahuila and southwestern San Luis Potosi, south to the Valley of Mexico and the Orizaba massif in Puebla (and possibly adjoining Vera Cruz).—Map No. 98.—May to October.

Astragalus mollissimus var. irolanus (Jones), comb. nov., based on A. orizabae var. irolanus (of Irola, Hidalgo) Jones, Rev. Astrag. 235, Pl. 60. 1923 ("Irolanus").—"Collected near Irola Mexico by J. N. Rose, June 14, 1899, Nat. Herb. No. 346530, the upper plant on the sheet... "—Holotypus, Rose & Hough 4557, US! isotypus, NY!—A. irolanus (Jones) Rydb. in N. Amer. Fl. 24 : 444. 1929.

Phaca mollis (soft, of the vesture) Kunth in H. B. K., Nov. Gen. Sp. 6: 388, PI. 585. 1823. —"in convalli Mexicana prope Gasave [near Pachuca, fide Sprague in Kew Bull., 1924, p. 22]." —Holotypus, P (herb. Humb.)!—Astragalus Humboldtii (Friedrich Heinrich Alexander von Humboldt, 1769-1859) Gray in Proc. Amer. Acad. 6: 195. 1864, a legitimate substitute (non A. mollis M. B., 1819).

Astragalus orizabae (of Mt. Orizaba) Seat. in Proc. Amer. Acad. 28: 117. 1893 ("Orizabae").—"Ledges and cliffs, Mount Orizaba, 9000 feet, August. (Seaton 262)"—Holotypus, collected August 8, 1891, GH! isotypus, NY!

Astragalus niquiriciaefolius (meaning obscure, unless a misprint for "liquiritiaefolius," with leaves of liquorice or Glycyrrhiza) Sessé & Mociño, Fl. Mex., Ed. 2. 168. 1894; Ed. 1. 184.

1895.—“Habitat in agris arenosis Regionum frigidarum, ubi vulgo Garbancillo indigenae appellant ... ”—Lectotypus, “Astragalus equicida,” Sessé & Mociño Herb. No. 3750, MA! isotypus, OXF!

Humboldt’s woolly locoweed, var. irolanus, is very close to var. Earlei. Indeed the two varieties are not reliably distinguished by any one differential character, and in the latitude of southern Coahuila (±25° N.) there exist transitional forms, already mentioned under var. Earlei, which are difficult to assign to either. For the most part, and nearly always southward from San Luis Potosi, the calyx of var. irolanus is broader than that of var. Earlei and has longer teeth; at the same time, the banner tends to be slightly longer and the pod more tumid or distinctly inflated and of plumper outline. Even in southcentral Mexico, however, the pod varies considerably in size and profile and provides no solid criterion. The plant collected at Gasave in southern Hidalgo by Humboldt and Bonpland, the original Phaca mollis, was in flower only. Had the bilocular pod been known, the species would have been described as an Astragalus, not as a Phaca. Perhaps it will never be known how the fruit would have turned out, for narrower and broader types occur together in the Valley of Mexico. When Seaton described A. orizabae, he assumed that Phaca mollis (A. Humboldtii) had a small, narrow pod as in var. Earlei. Jones (1923) also associated the name A. Humboldtii with narrow-fruiting plants, but his concept was an extremely muddled one, based on a mixture of var. Bigelovii, var. Earlei, and perhaps a small element of our var. irolanus. On the other hand, Rydberg (1929) assumed that A. Humboldtii would have had the plump tumescent pod of A. orizabae, and he treated these two as synonyms, taking up the name A. irolanus for the southern plant with narrow pod. However, I feel pretty sure that there is only one woolly locoweed in southcentral Mexico and that the variation in the pod is of a superficial nature. The var. irolanus was originally described as differing from A. orizabae in a proportionately longer (although still tumid) pod and leaflets of acutely rhombic outline. These features are not correlated in other collections, and there is no reason to interpret the type-collection as a mixture of significantly different forms.

I have listed A. niquiriciaefolius as a synonym of var. irolanus, even though no Sessé & Mociño specimens so labeled are known to exist. The description of the species fits var. irolanus well, and the plants were said to be poisonous to sheep, cattle, and horses. The vernacular name garbancillo, used in Spain for A. lusitanicus Lamk. and in Latin America (both North and South) for a variety of toxic astragali, was already current for A. niquiriciaefolius in Mexico at the end of the XVIII century. The choice of Sessé & Mociño’s “Astragalus equicida,” a name obviously applied to a known locoweed, seems appropriate. Other specimens from “Nueva Espana,” all probably dispersed by Pavon out of the collections of Sessé and his collaborators, are labeled: “A. contortuplicatus” (BM, MA), this possibly the basis of A. con-tortuplicatus sensu S. & M., PI. Nov. Hisp. 119. 1889 (non L.); “ex herb. Pavon.” (G, herb. Moricand.); and “Crotalaria sp. nov. de Mexico” (BM).