Flora of North America species comparison

Subspecies 4 or 5 (4 in the flora).

Potentilla anserina is polymorphic in most features, especially in hairiness, but also in size and in degree of dissection of leaflets, epicalyx bractlets, and sepals. While most of the variation described by A. G. Blytt (1906) is taxonomically insignificant, A. Rousi (1965) found support for three northern races: subspp. anserina, egedei, and pacifica. He also suggested that P. yukonensis Hultén might qualify as a separate subspecies. This treatment follows J. Soják (1994) in accepting four northern subspecies of P. anserina, all present in North America.

A. Rousi (1965) demonstrated partial interfertility between the races of Potentilla anserina, which form intermediates in all zones of contact. The distinctness of the four races, in spite of intermediates, is upheld by being partly allopatric and occupying different habitats where they are sympatric. As partly interfertile parapatric entities, they are best treated as subspecies.

The vast majority of chromosome counts are tetraploid. Tetraploid plants (2n = 28) are fully fertile; hexaploids (2n = 42) are largely pollen and seed sterile; pentaploids (2n = 35) are probably occasional hybrids (S. Erlandsson 1942, 1942b; A. Rousi 1965). A. Kurtto et al. (in J. Jalas et al. 1972+, vol. 13) considered numbers above tetraploid level as cases of occasional autopolyploidy.

(Discussion copyrighted by Flora of North America; reprinted with permission.)

The present circumscription of Potentilla drummondii differs from that of B. Ertter (1992, 1993), which encompassed P. breweri and P. bruceae as varieties (see 28. P. bruceae discussion for details). The description above focuses on large, eglandular, subpinnate plants that occur in mountains from the central Sierra Nevada and northern Coast Ranges of California to the Olympic Peninsula of Washington, east to southern Alberta. At least some collections previously identified as P. glaucophylla (as P. diversifolia Lehmann) from the Kenai Peninsula in Alaska can be readily accommodated within this concept, as can collections from the Cabinet Mountains in Montana. The distinction between P. drummondii and subpalmate P. glaucophylla can be vague; in general, the proximalmost leaflets of P. glaucophylla are relatively small and often entire; in P. drummondii proximalmost leaflets are similar in size and dissection to other leaflets.

Pinnate-leaved regional variants merit further attention; in particular the smaller form described as Potentilla cascadensis, and glandular plants in the northern Coast Ranges of California. Relatively small plants forming uniform populations in the southern Sierra Nevada may represent stabilized hybrids with P. breweri.

The inclusion of Potentilla bruceae and P. drummondii in sect. Graciles differs from that of J. Clausen et al. (1940) and B. C. Johnston (1980, 1985), who allied both species with P. breweri in sect. Multijugae. Although subpalmate to subpinnate leaves are anomalous in primarily palmate sect. Graciles, doubled proximal leaflets and an indistinct transition between the rachis and axis of the pinnatisect terminal leaflet in these two species suggest that this state is derived from a palmate progenitor. Habit and petioles also are more compatible with sect. Graciles than sect. Multijugae.

(Discussion copyrighted by Flora of North America; reprinted with permission.)