Mapping here is highly provisional. Miscellaneous records of broad-leaved variants within series Crus-galli are grouped here, pending further revision. Sources for these records include several old colls. at out-of-state herbaria (GH, MICH, MO); see also Slack (1941; check Cornell) and Wharton (1945).The coll. from MONT (GH) has been named C. persimilis Sarg. (verified by J.B. Phipps), which may be a distinct midwestern segregate (including C. hannibalensis Sarg.), perhaps transitional from crus-gallii to succulenta or calpodendron (F, Y, FNA 9). Included here are records of C. fecunda (FNA 9), C. palmeri Sarg. (CLAR, TRIG), C. regalis Beadle (ESTI, LYON, MCRA), and C. denaria Beadle (LYON, JEFF, MCRA; see also C. acutifolia Sarg. in D, perhaps transitional to C. viridis). Some of these taxa may be derived from hybridization (D, F). They are mostly distributed in midwestern regions, especially Ind., Ill., Iowa, Kans., Mo. and Ark., but with eastern records from N.Y., N.C. and Ky. (D, F). C. fecunda is a possible transition to series Macracanthae that is similar to persimilis, but it has not been recorded since the 1930s (FNA 9). J.B. Phipps (FNA 9) has treated palmeri as a large-leaved, large-flowered, northern variety of C. reverchonii Sarg. That species differs from crus-galli in its suborbicular to broadly elliptic leaves (versus spathulate to narrowly elliptic), and it usually has 3-5 seeds per fruit (versus 1-2); it is concentrated on lowlands in south-central states. He mapped var. palmeri at scattered locations in east-central states from Kans. to Del. Three colls. from TRIG at APSC are referable to palmeri, although they have more or less pubescent lower leaf surfaces and young twigs, which also have pale lenticels: E.W. Chester 86-669 ("wet woods and thickets at head of Barnett Bay" with "5 pyrenes per fruit") and 86-755 ("mesic slopes above Rhodes Bay"); R.L. Thompson 3-326 ("Elk and Bison Prairie... N of Golden Pond").