Flora of North America species comparison

P. A. Rydberg (1898) used var. alpicola (as Horkelia gordonii var. alpicola) for compact, high-elevation plants throughout the range of Ivesia gordonii. As circumscribed here, var. alpicola is restricted to plants occurring disjunctly in an arc around the northwestern part of the species range. Populations occur in the northern Coast Ranges and central Sierra Nevada of California, sporadically to Mount Adams and the Wenatchee Mountains in Washington, and from there to the mountains of central Idaho and the Bitterroot Range of Montana. The variety is also disjunct on the Owyhee Mountains of southwestern Idaho.

Plants of var. alpicola average shorter than those of var. gordonii, which occurs farther east and south. In var. alpicola, stems with only glandular indumentum are common; such stems are rare in var. gordonii. Unlike those of var. gordonii, the leaflets of var. alpicola are frequently glandular with only marginal hairs.

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

Species 30 (30 in the flora).

The often aromatic (resin-scented) ivesioid genera (Horkelia, Horkeliella, and Ivesia) have generally been assumed to be derived from Potentilla in the broad sense, an interpretation confirmed by molecular analyses (T. Eriksson et al. 1998, 2003; C. Dobeš and J. Paule 2010; M. H. Töpel et al. 2012). Continued recognition of the three genera (and Duchesnea) accordingly creates a paraphyletic Potentilla; the senior author does not accept the dictum that paraphyletic groups are inherently unnatural and finds it more useful to continue recognizing the ivesioid genera as here circumscribed.

In spite of the morphologic and ecologic extremes encompassed within Ivesia, spanning the continuum from planar-leaved desert chasmophytes to alpine species with mousetail-like leaves, molecular phylogenetic analyses of representatives from all sections of Ivesia, Horkelia, and Horkeliella result in a monophyletic near-polytomy, indicating rapid morphologic radiation not matched by molecular divergence (C. Dobeš and J. Paule 2010; M. H. Töpel et al. 2012). This interpretation is not fully supported by M. Töpel et al., but until the lack of concordance between molecular-based clades and even the most well-defined morphology-based groups is addressed, any conclusions based on molecular analysis alone are inconclusive.

The hypothesis on which the sequence of species used here is based is that the ivesiod genera represent a threshold-crossing radiation in response to the late Tertiary development of xeric conditions in western North America, with adaptations that allow drought-avoidance or minimize water loss. A recurrent trend is the progressive reduction in size and number of floral parts occurring independently in multiple lineages. The end product is a flower in which the nectariferous, patelliform hypanthium itself becomes the main pollinator attractor. Much of this evolutionary radiation has occurred in the various island-like habitats that characterize the arid West, with the result that the ivesioid genera include a high percentage of narrowly endemic species.

In the following descriptions, stems refer to flowering stems, with length including the inflorescence. For chasmophytic (growing on cliffs) species in sect. Setosae, terms such as decumbent are in reference to the vertical rock face. The inflorescence is composed of the branched portion of the flowering stem, including the proximalmost branched node. Cauline leaves are on the portion of the flowering stem proximal to the inflorescence. Planar leaves are those in which the leaflets are largely arranged in a single plane on both sides of the rachis; in cylindric leaves the lobes of deeply incised leaflets are arranged evenly around the rachis and thus appear verticillate. Well-developed blades of cauline leaves are similar to those of basal leaves, though progressively reduced distally; vestigial blades are reduced to a simple lobe, which, with the stipules, has the appearance of a 3-lobed bract. Leaf incisions are referred to as teeth if less than half to base, as lobes if more than half. Counts of leaflet lobes are of ultimate segments, including teeth, not just of primary segments incised to the base or nearly so. Inflorescence structure is described at peak anthesis; at early anthesis flowers are often more congested than indicated here and sometimes become more distant in fruit. Pedicels at proximal nodes are often significantly longer than others; those in most sections remain more or less straight in fruit or bend proximal to the flower, in contrast to those of sect. Setosae that generally become strongly curved or even sigmoid. Petal color is described at peak of flowering; yellow petals often fade afterward. Vestiture descriptions are primarily of non-glandular hairs; subsessile glands or evidently septate-glandular trichomes are universal in Ivesia, though often sparse or hidden by dense non-glandular vestiture.

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