130a.  Dalea lutea (Cavanilles) Willdenow var. lutea

(Plate CXV)

To 1 m tall, mostly less; stems and foliage variably pubescent to completely glabrous, the foliage deep green (blackish when dry) to gray-silky, the leaflets when pubescent often glabrous above; leaflets of larger leaves mostly 3-8 (10) mm long; spikes (1) 1.5-10 cm long; calyx (3.6) 4-7 (7.7) mm long, the tube pubescent from base upward, or glabrescent at base, or glabrous and lustrous up to the plumose teeth, the dorsal tooth (1.3) 1.7-4.5 (5.2) mm long; petals pale but clear yellow, in age or dried bright yellow, ochroleucous tinged or veined with brown, less often nigrescent; petals variable in size, the keel-blades (4.2) 4.8-7 (7.3) mm long; androecium mostly 7.5-10 mm long. — Collections: 110 (vii).

Open brushy and thinly wooded hillsides and grassy plains, sometimes colonial along hedges or on terraced milpa, n.-ward in canyons of desert mountains, e.-ward in mesquite grassland, s.-ward entering the oak-pine belt, 750- 2400 m (2500- 8000 ft), most prolific on limestone bedrock but not confined to it, widespread and locally abundant through Sierra Madre Oriental from Monterrey s. to Hidalgo, on the Gulf slope extending to the piedmont and e. into Sierra de Tamaulipas, n.-w. over scattered desert mountains to Sierra del Carmen in n. Coahuila, and w. to the foot of Sierra Madre Occidental in n.-centr. Chihuahua (to nearly 30° N), thence s. over the Central Plateau to Aguas Calientes, Sierra de Guanajuato, and valley of Mexico, apparently disjunct in extreme s. Puebla. — Flowering June to November. —Representative: Chihuahua: Pringle 1010 (F, NY, UC); Stewart & Johnston 2143 (GH, TEX); Correll & Johnston 21,709 (RENNER). Coahuila: Stewart 1705 (GH, TEX); Johnston 8946 (GH); Marsh 590 (GH, OKLA, MEXU); Johnston & Mueller 1174 (GH, MICH, TEX). Nuevo Leon: Mueller 2303 (ARIZ, F, NY, OKLA, TEX); M. Taylor 81 (ARIZ, F, NY, OKLA, UC); Ripley & Barneby 13,550 (CAS, MEXU, NY, US). Tamaulipas: von Rozynski 163 (F, NY, UC); Stanford, Retherford & Northcraft 736 (ARIZ, NY, UC); R. M. King 4517 (NY, US); Dressier 2288 (MEXU, UC). Durango: Gentry 6922 (F, GH, NY); Reko 5317 (F). Zacatecas: Ripley & Barneby 14,141 (CAS, DS, NY, US); Pringle 1754 (ARIZ, F, L, M, W); Gentry 6988 (F, GH, NY). San Luis Potosi: Parry & Palmer 153 (BR, F, GH, NY, WIS); Lundell 5120 (ARIZ, F, UC, US); Rzedowski 3104, 5246, 8461, 13,758 (ENCB). Aguas Calientes: McVaugh 23,691 (MICH, NY). Guanajuato: Ripley & Barneby 13,367 (MEXU, NY, US). Queretaro: Pringle 10,027 (L, M, NY, UC, Z); Painter & Rose 9746 (NY, US). Hidalgo: Hartweg 282 (OXF, K, NY, W); Pringle 6632 (M, NY, UC, US, W, Z); Ripley & Barneby 13,604 (CAS, NY, US), 13,624 (CAS, NY). Mexico (Edo. and D. F.): Bourgeau 1059 (BR, P); Matuda 19,532, 21,732 (NY); Rzedowski 1597 (ENCB); Ripley & Barneby 13,381 (NY). Puebla: Purpus 3882 (UC), 3213 (NY, UC).

Dalea lutea (Cav.) Willd., Sp. Pl. 3: 1341. 1801, based on Psoralea lutea (yellow) Cav., Ic. 4: 12, PI. 325. 1797 (Sept-Dee; cf. Taxon 10: 78). — "Habitat in Nova Hispania...Floruit in laudato horto mense Octob. 1796." — Holotypus, MA (herb. Cav., excluding small plant labelled "Queretaro, Nee")! isotypus, BM! — Parosela lutea (Cav.) Cav., Descr. Pl. 186. 1802; Rose, Contrib. U. S. Nat. Herb. 12: 273. 1909.

Dalea ovalifolia (with oval leaflets) Ort., Dec. 1: 30, tab. 3. 1797 (? mensis). — "Habitat in nova Hispania. Floret in Reg. Horto Matrit. mense Augusto e seminibus missis per D. Sesse."— No typus known to survive; lectotypus, the cited plate.

Dalea leucostoma (white-mouthed, of the ciliate calyx) Schlechtd., Linnaea 12: 294. 1838. —"Pr. Regia, Oct. (C. Ehrenberg)." —No typus seen, but the detailed protologue agrees well with material from the Gulf slope in centr. Hidalgo. — Parosela leucostoma (Schlechtd.) Rose, Contrib. U. S. Nat. Herb. 10: 106. 1906.

Dalea plumosa (feathered, of the calyx-teeth) Wats., Proc. Amer. Acad. 21: 448. 1886.— "In the shade of cliffs near Chihuahua; C. G. Pringle, October, 1885 (n. 621)." —Holotypus, GH! isotypi, BR, NY, US, W! — Parosela plumosa (Wats.) Rose, Contrib. U. S. Nat. Herb. 10: 106. 1906.

Parosela painteri (Joseph Hannum Painter, 1879-1908) Rose, Contrib. U. S. Nat. Herb. 10: 105. 1906. —"Type...collected by J. N. Rose and Jos. H. Painter on a stony hillside near San Juan del Rio, Queretaro, August 18, 1905 (no. 9526)." — Holotypus, US! isotypi, MEXU, NY! — Dalea painteri (Rose) Bullock, Kew Bull. 1939: 197. 1939.

Parosela wardii (Lester Frank Ward, 1841-1913) Rydb., N. Amer. FI. 24: 112, exclus. syn. 1920.— "Type collected at San Luis Potosi, September, 1891, Lester F. Ward..."—Holotypus, US!

Parosela caudata (tailed, of the bracts) Rydb., N. Amer. Fl. 24: 112. 1920.— "Type collected at Alvarez, San Luis Potosi, September 5-10, 1902, Edward Palmer 110..." — Holotypus, US! isotypus, NY! — Parosela lutea var. caudata (Rydb.) Macbr., Contrib. Gray Herb., New Ser. 65: 21. 1922. Dalea caudata (Rydb.) Bullock, Kew Bull. 1939: 195. 1939.

Parosela argyrostachya sensu Rydb., N. Amer. Fl. 24: 110. 1920, exclus. basonym.

Parosela macrostachya sensu Rydb., N. Amer. Fl. 24: 110. 1920, exclus. basonym.

This is the common yellow-flowered dalea of the Mexican Plateau, known in Tamaulipas, Nuevo Leon, Coahuila, and probably elsewhere as Ramon or Escobita de Ramon, in Hidalgo sometimes as Anicillo. Unlike those of most yellow-petalled species, the petals often retain much of their color in fading, the banner, however, always early rubescent and the wings and keel often becoming brown-veined. It differs from the rarer, marginally sympatric D. capitata in its more numerous (5-10 not 3-5) pairs of generally expanded and thin-textured leaflets and in the aristiform tips of the calyx-teeth, these almost always longer than the tube; and from the more closely related D. melantha, which has a similar calyx, in the persistent interfloral bracts.

Variation in var. lutea finds expression in stature and growth-habit, in density and dispersal of trichomes on leaves and calyx, in length of bracts and flower-parts, and in development of glands, whether on the panels of the calyx-tube, or as spurs on the calyx- teeth, or as blisters terminal to the midvein of banner and keel. The region in which var. lutea is most abundant lies within and on both slopes of Sierra Madre Oriental, from Monterrey into Hidalgo, and onward into the Valley of Mexico. Here the average plant has moderately pubescent leaves and stems, a calyx-tube at least thinly hairy below the orifice, relatively small (to 0.45 mm) intercostal glands, and spurless calyx-teeth. The latter, however, vary greatly in length, often in company with bracts and sometimes with stipules, the divaricate dorsal teeth and the bract-tails when long giving the spike a distinct teasel-like appearance. From San Luis Potosi northward the prevailing habit is diffuse and scarcely suffruticose, and the flower is relatively small, with glandless keel-blade ±5-6 mm long; in Hidalgo and southward the plants become distinctly shrubby when adult and the flower is often larger, with gland-tipped keel 6-7 mm long. These contrasts would suggest the possibility of separating from typical southern var. lutea a northern segregate corresponding with Parosela wardii and P. caudata (the minor variant with long-tailed bracts), but occasional large-flowered individuals are found as far north as Monterrey (Edwards s. n., NY) and a small, glandless keel as far south as Hidalgo and Mexico (Matuda 4865, Hartweg 282, NY). Sporadically within this range and west to Zacatecas (Gentry 6988, NY) occurs a densely and loosely pilose phase, the gray or silvery pubescence extending upward from stems and foliage to the bracts and calyx-tube; this, corresponding with the original Psoralea lutea Cav. lacks differential characters correlated with pubescence, and seems clearly to represent no more than a minor variant.

Westward toward the foothills of the Western Sierra, in the dry grasslands of the Central Plateau, var. lutea presents a series of glabrescent or truly glabrous forms, differing from their eastern kindred in having a glabrous calyx-tube charged with larger, sometimes confluent glands up to 0.5-1 mm diameter. Except that the calyx-teeth are here more prominently spurred, reflecting a greater development of glands in the tube, the flower is essentially identical with that of large-flowered var. lutea from Hidalgo and Mexico. The variant truly glabrous to the spikes, corresponding with Parosela argyrostachya sensu Rydb. (but not the true D. argyrostachya H. & A.), and its thinly pilosulous twin described as D. plumosa Wats, might be interpreted as a geographic variety, except for the existence in Queretaro and Mexico of plants combining large glands with both large and small keel- petals and a calyx-tube either glabrous below the teeth or partly hairy. These intermediate forms, among them the type-collection of Parosela painteri, provide all the necessary transitional steps between the most nearly glabrous and most densely pubescent aspects of var. lutea. Moreover a fully glabrous form has been collected far out on Sierra de Tamaulipas, at the point furthest in the range of var. lutea from the focus of glabrescent phases on the Central Plateau.

The modes of variation described in the foregoing paragraphs act independently of each other, yielding in a variety of combinations a number of individually distinct forms. Upon close examination these are found to stand away from the rest only in a particular combination of characters each of which is known to be inconstant elsewhere in the species. Most of the names listed in the synonymy of var. lutea were applied in the first instance to such individual variants.

It is possible that the earliest name for D. lutea is really D. ovalifolia. Cavanilles and Gomez-Ortega independently described the same plant growing in the Madrid garden from seed sent home by Sesse. Ortega’s specimen was said to be in flower in August, that of Cavanilles in October of 1796, and both descriptions appeared during the course of the following year. The description of Cavanilles has been dated to the last quarter of 1797; the precise date of Ortega’s is unknown. By 1802 Cavanilles (Descr. Pl. 186) transferred D. ovalifolia to the synonymy of his own Parosela lutea and this synonymy, not necessarily reflecting the facts of priority, has been accepted universally since. Unless and until the contrary can be demonstrated, tradition is best followed.