Flora of North America species comparison

Subspecies 2 (2 in the flora).

Lepidium virginicum is perhaps the only species in the genus with accumbent cotyledons; the other species have incumbent cotyledons, except for Australian ones with diplecolobal cotyledons. This contradicts N. H. Holmgren’s (2005b) assertion that the genus characteristically has accumbent cotyledons.

Examination of thousands of specimens provides good evidence (e.g., Florida: Lakela et al. 27038, GH; Missouri: Raven & Raven 27501, GH, MO) that Lepidium virginicum hybridizes with L. densiflorum. The ranges of both species overlap in much of the flora area, and this intergradation is, perhaps, the reason behind the recognition of some infraspecific taxa in both species. Molecular studies, along with a critical evaluation of morphology, are needed. Typical L. virginicum is easily distinguished by having well-developed or, rarely, rudimentary petals, accumbent cotyledons, orbicular fruits, and raceme rachises with curved cylindrical trichomes. By contrast, L. densiflorum has rudimentary or, often, no petals, incumbent cotyledons, obovate fruits, and raceme rachises with straight, often subclavate trichomes.

Of the seven varieties recognized by C. L. Hitchcock (1945) and R. C. Rollins (1993) in Lepidium virginicum, three do not occur in the flora area, and they most likely belong to other species. The four present in our area clearly fall into two groups. The first, which corresponds to the species lectotype, has accumbent cotyledons and terete fruiting pedicels. The second group, which includes the holotypes of vars. medium, pubescens, and robinsonii, has incumbent cotyledons and flattened fruiting pedicels. The type of var. medium has completely glabrous raceme rachises and fruiting pedicels, but most authors (e.g., Hitchcock 1936, 1945; Rollins) assigned this varietal name to glabrous plants regardless of whether they have accumbent or incumbent cotyledons. Indeed, the cotyledonary position and indumentum absence (or presence) do not always covary, and some glabrous plants have accumbent cotyledons (e.g., Demaree 47912 (GH), from Arkansas) or incumbent cotyledons (e.g., Demaree 43698 (GH) from Arizona). Therefore, var. medium does not merit recognition. As for var. robinsonii, it was based solely on being shorter plants with or without divided leaves. We believe that delimitation is artificial; such plants occur sporadically in the ranges of the two groups noted above. With the elimination of vars. medium and robinsonii, L. virginicum consists of two infraspecific taxa recognized herein as subspecies.

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

North American records of Lepidium apetalum Willdenow mostly represent misidentifications of L. densiflorum. The latter has obovate fruits widest beyond the middle, whereas L. apetalum has elliptic fruits widest at the middle.

The number and limits of the varieties recognized in Lepidium densiflorum, as well as the characters used to delimit them, vary among authors (A. Thellung 1906; C. L. Hitchcock 1936; G. A. Mulligan 1961; R. C. Rollins 1993; N. H. Holmgren 2005b). The variation almost always does not correlate with geography, and the recognition of varieties in this species is neither practical nor very useful. All of those authors admitted that these varieties are “very weak at best” (Rollins, p. 554). Of them, perhaps var. pubicarpum (including var. elongatum) might merit recognition. It is distributed in almost all of the Mountain and Pacific states and is distinguished from the other varieties solely by the presence of trichomes or minute papillae on the fruit valves. The density of these trichomes ranges from moderate and covering the entire valve surface to very sparse and represented by individual papillate trichomes restricted to the valve margin. Furthermore, the length of these trichomes may vary from ca. 0.01 to 0.3 mm. In some species (e.g., L. dictyotum) both glabrous- and pubescent-fruited forms occur, yet none of the above authors gave formal recognition to both forms. It is not known if both glabrous and puberulent fruits occur within the same population in L. densiflorum. The species is autogamous, but nothing is known about the rates of gene flow between and within populations.

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