The genus Psorothamnus, here expanded to embrace Psorodendron Rydb. (cf. sect. Xylodalea, infra), is related on the on$ hand to Errazurizia and on the other to Marina, and only through the latter, therefore more remotely, to Dalea as here received. It belongs with Errazurizia, Marina, Eysenhardtia, and Amorpha in the group of Amorpheae with ten pairs of chromosomes, and resembles all of these, except Marina, in having petals inserted directly on the hypanthium rim. It resembles Marina, but not others mentioned, in having an apparently normal papilionaceous keel, the two leaves of which are folded around and conceal the androecium at anthesis. The sect. Xylodalea, with its large, little compressed, exserted pod and single massive seed, makes a close approach to Errazurizia; the sect. Psorothamnus leans toward Marina in its smaller, compressed pod. However the ovule of Marina is solitary, those of Psorothamnus, like other Amorpheae, paired or exceptionally (Ps. spinosus) more numerous, and Marina has the inner petals elevated on a column continuous with the androecium, appearing epistemonous. It is this character which links Marina to Dalea sens. lat. and separates both from Psorothamnus.
The genus Psorothamnus as here redefined embraces not only all of Psorodendron Rydb. but also Parosela thompsonae Vail (misplaced by Rydberg in Parosela subgen. Trichopodium) and Dalea whitingi Kearn. & Peeb., described long after Rydberg’s monograph and now interpreted as only varietally distinct from the last. The dichotomy stressed by Rydberg (1919, and again in 1928, p. 198, in clave) between a genus Psorodendron characterized by a pedicellate flower and exserted (i.e., large, errazurizia-like) pod and a genus Psorothamnus different in its sessile flower and included (more compressed and dalea-like) pod cannot be followed literally. The flower of Psorothamnus polydenius (Torr.) Rydb. is often shortly pedicelled, while the fruit of Psorodendron kingii (Wats.) Rydb., though pedicelled, does not surpass the calyx-teeth. Moreover Psorothamnus thompsonae (Vail) Welsh & Atwood, with a compressed included pod combined with the long pedicel of Psorodendron but the peculiar retrorse stem-pubescence found otherwise only among original members of Psorothamnus sens, str., sits squarely between the two. It should be mentioned that W. F. Mahler, working primarily from the palynological angle, arrived independently at the same inclusive generic concept.
Morphological features not or only rarely duplicated elsewhere in tribe Amorpheae but seen in one or more (but not all together in any one) species of Psorothamnus are both numerous and, in some instances, suggestive of intergeneric relationships: creeping rhizomes (Ps. kingii); leaves all (Ps. spinosus), most (Ps. scoparius), or some (Ps. emoryi) reduced to the terminal leaflet; stem-pubescence retrorse (sect. Psorothamnus except Ps. emoryi); leaflets decurrent on the rachis (Ps. arborescens, Ps. emoryi, but both in part only); axis of the raceme (sects. Capnodendron and Winnemucca) or of sterile branchlets immediately preceding it (Ps. thompsonae) spinose; pedicels bibracteolate (in all but sect. Psorothamnus, where bracteoles are represented, as in all species of Marina, by sessile glands); petals marcescent (Ps. thompsonae); more than one seed maturing from a pair of collateral ovules (Ps. kingii)', more than 2 ovules (Ps. spinosus always, Ps. thompsonae occasionally). The large pod and massive seed of sect. Xylodalea are closely matched in Errazurizia, of which the one Chilean species has, unlike the North American, keel-petals lightly adherent by the overlapping margins, at least in bud. The small compressed pod of sect. Psorothamnus hardly differs from that of some Marina. However these odd or aberrant features of individual Psorothamnus species are, as implied in the list, poorly correlated, and contribute little to clarification of cross-relationships within the framework of the genus. Note, for example, how the characters of the inflorescence and the incidence of a glandular connective cross over, as it were cancelling each other out.
A notable feature of Psorothamnus is the high percentage of taxonomically isolated types. Despite a constitution, at the generic level, adapted to extreme desert conditions, the genus has developed no monocarpic species, as has occurred in Dalea and Marina under comparable climatic stresses. The generally woody character of its members, their mutually remote taxonomic relationships, and the connection via sect. Xylodalea with the apparently relict genus Errazurizia (and thence with Eysenhardtia) mark Psorothamnus as probably ancient and primitive relative to the two epistemonous genera of Amorpheae. It is, however, evidently closer kin to Marina than to Dalea itself.