2. Rorippa curvipes Greene (blunt-leaved yellow cress)

R. truncata (Jeps.) Stuckey

Pl. 326 c, d; Map 1382

Plants annual or biennial, with taproots, not rooting at the nodes or rooting sporadically only at the lowermost few nodes. Stems 6–50 cm long, spreading or less commonly erect or ascending, usually several-branched from the base, glabrous. Leaves basal and alternate, 2–12 cm long, the basal and lowermost stem leaves mostly short-petiolate, the bases usually somewhat clasping the stem with small, rounded auricles, pinnately divided with 3–17 divisions, the lobes linear to oblong or ovate, the margins irregularly toothed, glabrous or the lower leaves sparsely hairy on the petioles and larger veins. Sepals 0.8–1.5 mm long. Petals 1.0–1.5 mm long, yellow. Styles not apparent or up to 1 mm long. Fruits 2.5–6.5 mm long, 1–2 mm wide, ovoid or oblong, straight, usually slightly constricted at about the midpoint, the surface smooth. Seeds mostly 20–80 per fruit, in 2 rows in each locule, 0.5–0.7 mm long, circular in outline, usually with a small notch at the base, the surface slightly uneven or finely bumpy, brown. May–September.

Uncommon, sporadic along the floodplains of the Mississippi and Missouri Rivers (western U.S. and adjacent Canada east to Missouri and Texas). Banks of rivers and margins of ponds.

Steyermark (1963) included R. obtusa (Nutt.) Britton in his treatment of the genus. Stuckey (1972) untangled the several taxa in North America that had been referred to collectively under that name. Stuckey noted that R. teres (Michx.) Stuckey is the correct name for R. obtusa, and that this taxon occurs primarily along the Atlantic Coastal Plain and in Mexico, not in Missouri. According to him, the specimens from Missouri constitute a mixture of two different species, R. tenerrima and R. truncata. The taxonomic disposition of these segregates from R. teres has not been without controversy, however. Gleason and Cronquist (1991) accepted only a single segregate under the name R. tenerrima but also suggested that true R. teres occurred in the St. Louis area. Examination of specimens has not supported this claim. Rollins (1993) also accepted R. tenerrima but treated R. truncata as part of another segregate species, R. curvipes var. truncata (Jeps.) Rollins. Rollins further suggested that the western var. curvipes probably was introduced in Missouri, based apparently upon collections from St. Louis of plants with a growth form typical of this taxon. The two growth forms also were noted by Stuckey (1972). The association between these two taxa as components of R. curvipes appears quite sound; however, there is so much overlap in the fruit and other characters Rollins used to distinguish his varieties that there is no utility in attempting to formally recognize them taxonomically. For further discussion on the separation of R. curvipes from R. tenerrima, see the treatment of that species.