4. Ambrosia psilostachya DC. (western ragweed)

A. coronopifolia Torr. & A. Gray

Pl. 272 f, g; Map 1145

Plants perennial, usually colonial from deep, often widely creeping rhizomes. Stems 30–70(–100) cm long, moderately to densely pubescent with relatively short, appressed-ascending hairs usually with minute, pustular bases, sometimes also with longer, spreading hairs. Leaves opposite, the uppermost sometimes alternate, sessile or with very short, narrowly winged petioles. Leaf blades 3–10 cm long, lanceolate to ovate in outline, mostly 1 time pinnately lobed with more than 5 lobes (the uppermost leaves sometimes few-lobed to nearly entire), the lobes narrowly lanceolate to narrowly triangular, entire or few-toothed, the surfaces moderately to densely pubescent with somewhat pustular-based hairs and often appearing somewhat grayish, the undersurface not or only slightly paler than the upper surface. Staminate heads in spikelike racemes, these usually not in paniculate clusters, the staminate involucre 2–3 mm wide, with 3–9 shallow lobes, usually moderately hairy. Pistillate heads in small axillary clusters (or sometimes solitary), the involucre enclosing 1 floret and with 1 beak, 2.5–3.5 mm long at fruiting, more or less ovoid, with 1 series of not or only slightly flattened, short tubercles in a ring toward the tip, these sometimes absent, moderately to densely hairy, especially above the midpoint. 2n=36, 72, 108, 144. August–October.

Scattered, mostly in the Unglaciated Plains Division and the western portion of the Glaciated Plains (western U.S. east to Michigan and Louisiana; Canada; introduced eastward to Maine and Florida). Upland prairies, loess hill prairies, and sand prairies; also railroads, roadsides, and sandy, open, disturbed areas.

Many earlier botanists segregated midwestern plants from those farther south in the range as A. coronopifolia based on slight morphological differences as well as different ploidy (Steyermark, 1963). Following more detailed investigations of morphological variation and chromosome number, Payne (1970) concluded that the complex is best treated as a single variable species.