Dalea bicolor var. bicolor

Dalea bicolor Humb. & Bonpl. ex Willd., Hort. Berol. pl. 89. 1809. - "Habitat in America meridionali. Dalea bicolor de Humb. & Bonpland." - No typus seen, but a spm is to be sought in herb. Willd., B.-D. bicolor sensu Bonpl., PI. Rar. Malm. 142, tab. 58. 1819, the locality there corrected to Mexico. — Parosela bicolor (H. & B. ex Willd.) Rydb., N. Amer. Fl. 24: 83. 1920, quoad nom., exclus syn. et descr.

Dalea tuberculata (warty, of the stems) Lag., Gen. & Sp. Nov. 23. 1816. — "Habitat in Nov. Hisp. Semina adportavit prael. D. Sesse..." — No typus found at MA, but the traditional interpretation is confirmed by the protologue and Plate 136 of Sesse & Mocino Caiques at G (= Field Neg. 30,607, F), so labelled. — Parosela tuberculata (Lag.) Rose, Contrib. U. S. Nat. Herb. 10: 104. 1906.

Dalea thymoides (resembling in habit common thyme, T. vulgaris) Schlechtd. & Cham. Linnaea 5: 580. 1830.— "Inter Hacienda de Tenestepeque et Hacienda de Quantotalapa in sylvaticis. Sept." —No collector mentioned in the protologue, but this supplied by Schlechtd., Linnaea 12: 292. 1838: holotypus, Schiede & Deppe 617, formerly B, not seen; isotypus W!

Dalea laevigata (smooth) G. Don, Gen. Hist. Diehl. Pl. 2: 224. 1832. — "Native of Mexico...(v. s. in herb. Lambert.)." —Presumed holotypus, labelled "Dalea sp. nov. de Mexico" with epithet "laevigata" added in another hand, perhaps G. Don’s, OXF!

Dalea verrucosa (warty) G. Don, Gen. Hist. Diehl. PI. 2: 225. 1832. — "Native of Mexico...(v. s. in herb. Lambert;)." — Presumed holotypus, labelled "Dalea sp. nov. de Mexico" with annotation in same hand as above "D. verrucosa", OXF!

Dalea comosa (with flowers arranged in a pigtail plait, as Polygala comosa Schkuhr) Schlechtd., Linnaea 12: 289. 1838. — "Mineral del Monte; Cuesta blanca, Nov., Dec. (C. Ehrenberg)." —Holotypus, formerly B, not seen, but the protologue decisive.

Dalea ehrenbergii (Carl August Ehrenberg, 1801-1849) Schlechtd. in op. cit. 290. 1838. — "...ad aquas calidas pr. Grande [ = Atotonilco, n. of Pachuca, Hidalgo] (C. Ehrenberg)."—Holotypus, formerly B, not seen, but the detailed protologue decisive.

Dalea seemanni (Berthold Carl Seemann, 1825-1871) Wats., Proc. Amer. Acad. 22: 470. 1887 ("Seemani"). — "Collected by Seeman in the Sierra Madre, probably in Durango, and by C. R. Orcutt at Socono [= Socorro], in Lower California, April, 1886." —Holotypus, Seeman s. n. [from Durango], GH! the paratypus (Orcutt) = var. orcuttiana Barneby. — Parosela seemanni (Wats.) Rose, Contrib. U. S. Nat. Herb. 10: 106. 1906.

Parosela tuberculina (tuberculate, of the stems) Rydb., N. Amer. FI. 24: 89. 1920.— "Type collected in the vicinity of San Luis Tultitlanapa, Puebla, July 1908, Purpus 3215..."—Holotypus, NY! isotypi, F, GH, UC, US!— Dalea tuberculina (Rydb.) F. J. Hermann, Jour. Wash. Acad. Sci. 38: 237. 1928.

Parosela longeracemosa (with long spikes) T. Brandg., Univ. Calif. Pub. Bot. 10: 184. 1922. — "Collected at Minas de San Rafael, San Luis Potosi, No. 4839." — Holotypus, collected by C. A. Purpus, Nov 1910, UC!

The var. bicolor is the commonest purple-flowered shrubby dalea of the Mexican Plateau, characterized by the combination of warty stems, relatively few (usually only 3-6) pairs of leaflets, and narrow spikes of short-toothed calyces subtended by long-persistent bracts. The variety is composed of a mosaic of local races that differ in amount and orientation of pubescence, in density and length of spikes, in exact number (and to some degree shape) of leaflets, and in color of the epistemonous petals, which vary from pale rose-purple through vivid phlox-purple to violet or (less often) cobalt blue. To the field observer, however, these variable features are quite overshadowed by the extraordinary plasticity of habit and stature. On the Central Plateau north of the Bajio, the average adult plant of var. bicolor forms a rounded, more or less erect, many-stemmed bush up to about a meter tall, but the potential stature is not achieved in exposed places; on the other hand even in desert Coahuila plants twice that height are found in sheltered hollows of canyon washes. In the more benign climate of the Neovolcanic mountains a stature of two meters is normal, the plants there forming thin erect shrubs consisting of a few crooked leafless trunks bearing aloft a bushy head. In eastern San Luis Potosi, where var. bicolor descends into the edge of the relatively moist oak woods of the Gulf slope, it is possible to stand in the shadow of arborescent specimens and reach or even jump for the lowest flowers. By contrast on the thin pumice soils at the foot of Cofre de Perote and Orizaba in eastern Puebla and adjoining Veracruz var. bicolor is represented by a lowly depressed or humifuse ecotype, described long ago as D. thymoid.es Schlechtd., of which the most vigorous individual rises no more than one or two decimeters from the ground.

Field experience suggests that the extent to which habital differences are inherited and imposed on the individual by environment varies from one population to another. Possibly, therefore, within comprehensive var. bicolor there may lie concealed by random variation elsewhere in the complex some genetically stabilized entities. But when a summit branchlet taken from a tall shrub in Michoacan is (in the herbarium) indistinguishable from the whole stem, cut at the root, of some northern xerophytic bushlet, taxonomic discrimination founders. The discovery by Mosquin of tetraploid var. bicolor in Aguas Calientes and Zacatecas indicates genetic ferment within the species, but unfortunately it seems impossible to distinguish by means of exterior morphology between known diploid and known tetraploid specimens.

Notable variations other than those involving habit and stature and largely independent of either are expressed in length of spike, leaflet number, and pubescence of the calyx. The extreme conditions of the spike are on one hand a subcapitate to shortly oblong cluster of 7-20 flowers with calyces subcontiguous, on the other a series of 50 or more flowers strung out, distant each from the next by its own wideth or more, along a flexuous axis up to two decimeters long. In the arid climate of the Central Plateau the leaflets are commonly 3-6 pairs in primary cauline leaves, but these rise (inconsistently) to 5-9 pairs in Michoacan and southeast Mexico. In the barrancas high on the Gulf slope in Hidalgo, and sporadically elsewhere, the calyces become densely villous with longer hairs, the minor variant recognized by Rydberg (1920, p. 88) as Parosela seemanni.

Along the eastern edge of its range var. bicolor passes successively into the next three varieties, only the last of which is at all strongly characterized. The dominant race in western Coahuila and adjoining central Chihuahua is at once densely and loosely pubescent with ordinarily short spikes (as in var. argyrea) but villosulous calyces. A prostrate or diffuse variant of this more commonly bushy and erect variant, similar to the diffuse states already noted from further south, was annotated by I. M. Johnston (see especially Stewart 52,515 etc., GH) as a species nova, but lacks convincing differential characters.

The var. bicolor is sympatric over parts of its range with several more or less closely related species and seems to give rise occasionally to fertile hybrids. Near Zimapan, Hidalgo, colonies of D. dorycnioides, there extremely abundant on arid limestone, interfinger along water-courses through the hills with thin stands of D. bicolor, and seemingly intermediate individuals are found between them (Ripley & Barneby 13,618, NY). In municipio Jacala, Hidalgo, a plant combining the growth-habit and narrow panicle of spikes characteristic of D. tomentosa with the relatively large bicolored flower of D. bicolor (Moore 5030, BM, GH, MEXU, UC) suggests another interspecific cross.

On the headwaters of Rio Conchos in southern Chihuahua a form of var. bicolor is recorded as furnishing "significant animal food" (Pennington 349, TEX) with a vernacular name carnero.