1. Sanicula canadensis L. (Canada snakeroot)

Pl. 211 i, j; Map 879

Plants biennial, with fibrous roots. Stems 15–100 cm long. Leaf blades 1.5–14.0 cm long, deeply palmately 3- or 5(7)-lobed and/or compound. Involucre with the bracts 8–35 mm long. Rays 0.2–3.0 cm long. Umbellets all with a mixture of staminate and pistillate flowers, the staminate flowers usually 1–7 per umbellet. Sepals 0.4–1.1 mm long, fused only at the base, narrowly lanceolate, with the tip tapered to a sharp point. Petals greenish white, usually shorter than the sepals. Stamens with the anthers white. Styles shorter than the bristles of the fruit, shorter than to about as long as the sepals. Fruits 2–5 mm long, the stalks 1.0–1.5 mm long. 2n=16. May–July.

Scattered to common throughout the state (eastern U.S. west to Wyoming and Texas; Canada). Bottomland forests, mesic upland forests, acid seeps, banks of streams and spring branches, and margins of upland prairies and glades; rarely also shaded, disturbed areas.

Steyermark treated a few specimens scattered through the range of S. canadensis in Missouri as var. grandis Fernald. This variety originally was defined to include robust plants with larger leaves and clusters of fruits and was thought to be unworthy of formal taxonomic recognition by Shan and Constance (1951) and most later authors. Phillippe (1978) and Pryer and Phillippe (1989) redefined var. grandis to include plants generally intermediate between S. canadensis and S. marilandica in inflorescence, fruit, and stylar morphology, and Phillippe cited a single specimen from St. Louis County as conforming to this concept of the taxon. This specimen was relocated through the kindness of Dr. Bruno Wallnöfer of the Natural History Museum in Vienna, Austria, and shown not to have been collected in Missouri after all. Also, it is of interest that although Phillippe and Pryer and Phillippe suggested that their variety did not represent plants of hybrid origin, many of the specimens from other states annotated by Phillippe as var. grandis earlier had been ascribed provisionally to S. canadensis × marilandica by Shan and Constance. The identity of these morphologically intermediate plants remains in question and requires further study using modern methods of genetic analysis. Thus, it seems premature to recognize varieties within S. canadensis at the present time, and even if the var. grandis were to be recognized, it would have to be excluded from the Missouri flora.