Dalea multiflora

  • Title

    Dalea multiflora

  • Authors

    Rupert C. Barneby

  • Scientific Name

    Dalea multiflora (Nutt.) Shinners

  • Description

    42. Dalea multiflora (Nuttall) Shinners

    (Plate LV)

    Closely related to D. Candida, differing somewhat in habit of branching and leaflets; stems of adult plants several or numerous, mostly erect and ascending in clumps, 3-9 dm tall, either virgate and paniculately branched distally or branched nearly throughout, never monocephalous; foliage either deep or pallid green; stipules 0.4-2 mm long; primary cauline leaves 2-3 cm long, with 3-6, mostly 4 or 5 pairs of narrowly oblong to elliptic-oblanceolate, less often narrowly obovate or linear-oblong leaflets up to 7-13 mm long; spikes permanently dense, headlike, subglobose to shortly oblong, without petals 7-9 mm diam, the glabrous axis 4-10 (12) mm long, concealed by calyces; calyx of D. Candida, always glabrous except for ciliolate teeth, the intercostal panels of tube charged at apex with 1 (2) glands; epistemonous petals 3.5-5 mm long, the oblong blades obtuse, usually truncate at base, 2.3-3.5 mm long, 1.5-2.1 mm wide.— Collections: 86 (i).

    Prairies and stony hillsides, in great variety of xeric microhabitats, in acid sands, rich clays, and calcareous gravels, near sea-level on the Gulf Coast up to ± 660 m (2200 ft) on Edwards Plateau, widespread and locally common from the coastal prairies between Galveston and Corpus Christi, Texas, w. to lower Devils River in Sutton and Valverde counties, n. through central Oklahoma to e.-centr. Kansas and immediately adjacent w.-centr. Missouri (Jackson County); s. interruptedly and apparently rarely to centr.' Coahuila and the n. extremity of Sierra Madre Oriental in Nuevo Leon; cf. Wemple, 1970, map 5.— Flowering (mid-May) June to September.—Representative: UNITED STATES. Missouri: Bush 257, 826 (NY). Kansas: W. T. Barker 4349 (NY); Bartley 1126, 1207 (NY); McGregor 17,332 (NY). Oklahoma: Hopkins & Cross 6447 (NY, OKLA, UC); Waterfall 12,431 (OKLA, UC, WIS); Demaree 13,275 (NY, UC). Texas: Berlandier 2467 (GH, NY); Wemple 242 (NY); A. Heller 1814, 1903 (NY, UC); Whitehouse 16,155 (NY, UC); Cory 53,312 {NY, UC); E. J. Palmer 13,585 (WIS). MEXICO. Coahuila: Marsh 30, 1311 (F, OKLA, TEX); Wynd & Mueller 223 (ARIZ, NY). Nuevo Leon: C. & M. Mueller 505 (F, MEXU, TEX).

    Dalea multiflora (Nuttall) Shinners, Field & Lab. 17: 82. 1949, based on Petalostemon multiflorum (many-flowered) Nutt., Jour. Acad. Philad. 7: 92. 1834.— "In the plains of Red river." —Holotypus, PH, verified by Wemple, 1970, p. 41; isotypi, labelled by Nuttall "Petalostemon *multiflorum. Red River", NY (herb. Torr.) and "Arkansa", probably = Choctaw County, Oklahoma, collected by Nuttall in May 1819, K!—Kuhnistera multiflora (Nutt.) A. Heller, Mem. Torrey Club 5: 197. 1894. K. Candida multiflora (Nutt.) Rydb., Contrib. U. S. Nat. Herb. 3: 154. 1895.

    Closely akin to D. Candida, of which it has exactly the individual flower, but differing in the short, always compect spikes combined with relatively many leaflets. The spikes may be arranged either in a narrow thyrse at top of a tall virgate stem, or corymbosely at tips of all branchlets of a more diffuse and shorter primary axis. The plants vary much in stature, those of lowland prairies between Kansas and the Gulf coast resembling sym- patric var. Candida, whereas those of more arid hill-country of Edwards Plateau are lower and bushier. The range of the tall lowland ecotype of D. multiflora coincides extensively with that of var. Candida, and occasional intermediate forms, with foliage of the first but cylindric spikes of calyces, are difficult to assign to either species. During my independent study of sect. Kuhnistera I interpreted these intermediate types as evidence of intergradation and, following an earlier estimate of Rydberg, annotated many sheets of D. multiflora as D. Candida var. multiflora (nom. nud.). Wemple, however, has synthesized a hybrid D. Candida X multiflora and finds the plants very similar to the naturally occurring intermediates, which may well be first-generation hybrids also. The lower ecotypes of D. multiflora are found in central Oklahoma to overlap the range of D. Candida var. oligophylla, but from this region no intermediates are known as yet and the two entities become almost completely vicariant in range south of Red River. Occasionally the spike of var. oligophylla fails to lengthen much, even in fruit, and it then very closely resembles the spike of D. multiflora, but is generally then combined with substantially simpler stem- leaves that are the most stable diagnostic feature of var. oligophylla. I therefore follow Wemple in treating D. multiflora as specifically distinct.

    According to Shinners (1949, p. 82) the androecium of D. multiflora is sometimes hexamerous, but I have seen no instance of this certainly anomalous condition. Wemple by a slip traces D. multiflora to Chihuahua; doubtless Coahuila is meant.