4. Liatris punctata Hook. var. punctata (prairie snakeroot)

Pl. 270 f, g; Map 1124

Rootstock a more or less thickened, elongate taproot, vertical or occasionally somewhat spreading. Stems 20–90 cm long, glabrous. Basal and adjacent lower stem leaves sessile to short-petiolate, the blades 7–15 cm long, 1–4 mm wide, linear, the narrow, light margins occasionally curled under, the surfaces glabrous but the margins of at least some leaves hairy (note that the hairs break off with age, leaving minute, stubby bases), grayish green, with 3(5) main veins, grading toward the stem tip to shorter leaves, these mostly sessile, 1.5–12.0 cm long, linear. Inflorescences elongate spikes, the heads densely spaced (the axis mostly not visible between heads), sessile or nearly so, with 1 basal bract. Heads with 3–7 disc florets, the terminal head sometimes slightly longer than the others. Involucre 10–14 mm long, narrowly cup-shaped to nearly cylindrical, with 4 or 5 unequal, overlapping series of bracts (the outer series appearing somewhat shorter). Involucral bracts broadly lanceolate to narrowly oblong-obovate, tapered to a sharply pointed, ascending to somewhat spreading tip, mostly with narrow, thin, pale to transparent margins, these sometimes slightly to strongly purplish-tinged, more or less entire but usually cobwebby-hairy, the main body appearing flat below the tip. Corollas 9–11 mm long, glabrous. Pappus bristles plumose. Fruits 5.5–7.5 mm long. August–October.

Uncommon along the western border of the state, mostly in the Glaciated Plains Division (South Dakota to New Mexico east to Michigan, Tennessee, and Louisiana). Loess hill prairies and rarely upland prairies.

Steyermark (1963) followed Gaiser (1946) in accepting Missouri plants as part of var. nebraskana Gaiser based upon minor differences in leaf length and width, as well as involucral bract shape and pubescence. However, he questioned whether these taxa should be formally recognized, as specimens from Missouri and adjacent states are somewhat intermediate between var. nebraskana and var. punctata. Further study of additional collections has reinforced the futility of attempting to separate populations into two varieties based upon these minor and inconsistent variations. The wisdom in recognizing somewhat better-marked populations from Texas, New Mexico, and adjacent Mexico that have a more loosely flowered inflorescence as var. mexicana Gaiser requires further study. See the treatment of L. mucronata for further discussion of problems of species recognition in this complex. The specimen of L. punctata cited by Steyermark (1963) as having been collected in Bates County could not be located during the present study.