1. Mirabilis albida (Walter) Heimerl (white four-o’clock, pale umbrellawort, hairy four-o’clock)

Oxybaphus albidus (Walter) Sweet

Pl. 459 f–h; Map 2086

Stems 20–100 cm long, glabrous to densely hairy, the hairs sometimes in 2 vertical lines, when hairy sometimes also glandular, when glabrous appearing light brown, tan or silvery-whitened. Leaves sessile or with petioles to 0.8 cm long. Leaf blades 3–10(–12) cm long, linear to more commonly lanceolate or oblong-elliptic, rarely narrowly ovate, narrowed or tapered at the base, narrowed or abruptly tapered to a rounded or bluntly pointed (rarely sharply pointed) tip, glabrous to densely hairy, sometimes also somewhat glandular, when glabrous the undersurface pale or glaucous. Inflorescences terminal and axillary or sometimes only axillary. Involucres 4–5 mm long at flowering, becoming enlarged to 8–12 mm long at fruiting, glabrous or sparsely to densely hairy on the surfaces, sometimes also glandular, with (1)2–5 flowers. Perianth 6–10 mm long, white to pink. Fruits (including the hardened perianth tube) 4–6 mm long, noticeably warty on the angles and sides, each tubercle of the angles with a tuft of short hairs at the tip, otherwise minutely hairy, olive green to dark brown. 2n=58. May–October.

Scattered nearly throughout Missouri, but apparently absent from most of the northeastern quarter of the state (eastern [mostly southeastern] U.S. and adjacent Canada west to North Dakota, Nebraska, and Texas). Upland prairies, sand prairies, glades, tops and exposed ledges of bluffs, banks of streams, and less commonly savannas and openings of dry to mesic upland forests; also pastures, roadsides, railroads, and open disturbed areas.

Mirabilis albida is quite variable in most vegetative features, including hairiness. Steyermark (1963) and most other earlier authors accepted M. hirsuta (Pursh) MacMill. (hairy four-o'clock) as a species separate from M. albida (white four-o’clock, pale umbrellawort), based on its densely hairy leaves and stems (vs. glabrous or nearly so). B. L. Turner (1993b), who treated the genus in Texas, accepted this species with reservation and remarked that, “...I suspect that, as treated by most American workers, it is a hodge-podge of hirsute specimens belonging to several species, mainly M. albida and M. nyctaginea. For example, Steyermark (1963), in his Flora of Missouri, retained the species, but it seems clear from his key and distribution maps that it might be better treated as a leaf form of M. albida.” In his studies of the genus for the Flora of North America Project, Spellenberg (2003) also noted that fruits of plants attributed to M. hirsuta always seem to resemble those of other species more common in the region, and placed M. hispida into synonymy under M. albida, based on examination of the type specimen.