132g.  Dalea versicolor Zuccarini var. argyrostachya (Hooker & Arnott) Barneby

(Plate CXX)

Basally lignescent tall herbs or finally shrubby, at anthesis 5-15 dm tall, the primary inflorescence a terminal to few-branched panicle but this succeeded by lateral subcapitate or shortly ovoid spikes arising from axillary short-shoots; stipules 0.5-2 mm long; foliage thinly pilosulous with weak hairs or glabrous; leaflets of primary leaves (7) 8-11 pairs, strongly carinate dorsally; spikes racemiform, without petals (10) 11-15 mm diam, the axis of those terminal to principal branchlets (2) 3-11 cm, of those on axillary short-shoots ± 0.5-2 cm long; calyx (5) 5.8-7.2 mm long, the tube 2.5-2.8 mm, the teeth (2.5) 3-4.4 mm long; epistemonous petals pale lilac or whitish. — Collections: 9 (o).

Dry grassy hillsides, sometimes in oak woodland, ± 500-1350 m, local, known only from the piedmont and detached sierras or low hills rising from the Pacific coastal plain along the w. slope of Sierra Madre Occidental in Nayarit and Sinaloa, the known localities lying between Tepic and Sierra Tacuichamona in lat. 21° 30- 24° 30 N. — Flowering October to April. — Material: Sinaloa. Cosala: Sierra Tacuichamona, Gentry 5685 (ARIZ, MEXU, MICH, NY). San Ignacio: San Xavier, Ortega 1220 (MEXU), 5090 (MEXU); Cordon de Balboa, Montes & Salazar 787 (US). Nayarit. Tepic: Tepic, L. Paray 2708 (ENCB); Tepic to junction of San Bias road, Alava & Cook 1565 (MICH, UC). : Territoire Huichol, Sierra de Nayarit, "Jalisco", Diguet s. n. (NY).

Dalea versicolor Zucc. subsp. argyrostachya (H. & A.) var. argyrostachya. D. argyrostachya (with silvery spikes) H. & A., Bot. Beechey Voy. 285. 1836, sens. str. — No particular station indicated, therefore from environs of Tepic. — Holotypus, labelled "Talisco, Beechey", i.e. Jalisco, a few km s. of Tepic, K! — Parosela argyrostachya (H. & A.) Rose, Contrib. U. S. Nat. Herb. 10: 106. 1906.

Parosela leucantha (white-flowered, rarely if ever appropriate) Rydb., N. Amer. FI. 24: 103. 1920.— "Type collected in Tepic, [Jan] 1892, Edward Palmer P, in part (herb. N. Y. Bot. Gard.)." —Holotypus, 2 branches + some fragments in packet, all of one species, the "in part" of protologue unexplained, NY! isotypus, Palmer 1871, GH! — Dalea leucantha (Rydb.) Gentry, Madrono 10: 230. 1950.

Gentry (1950, p. 226, in key) separated D. leucantha from D. wislizeni on the basis of long spikes, more openly grooved leaf-rachis, relatively long dorsal calyx-tooth, and paler petals. The first is true of primary spikes only; those from axillary short-shoots produced in spring (cf. Paray 2708 collected April 1) are no longer. The length of calyx-teeth is variable, and the petal-color is apparently not paler than in some var. glabrescens of the interior plateau. Ultimately the best differential character is found in the openly sulcate leaf-rachis which coincides with a distinctive pattern of dispersal. The one other element of the polymorphic D. versicolor present on the Pacific slope of Sierra Madre, var. calcarata, is much more easily distinguished by its confluent rachis-margins with solitary intrapetiolular gland, flat, thin-textured (not carnosulous and cymbiform) leaflets, and bright rosy- pink or purple epistemonous petals.

The types of D. argyrostachya and D. leucantha were both collected at Tepic and are narrowly conspecific. Rydberg interpreted the former by means of a drawing (NY) prepared at Kew in 1918 which he mistakenly associated with a form of D. lutea, believing the petals to have been yellow, in this being misled by a remark of Hooker’s. No yellow- flowered dalea is known to occur in Nayarit, and all forms of D. lutea differ from D. argyrostachya in persistent interfloral bracts.