Phytologia (November 1991) 71(5):420-422. NEW COMBINATIONS IN POTENTILLA AND HORKELIA (ROSACEAE) IN CALIFORNIA Barbara Ertter University and Jepson Herbaria, University of California, Berkeley, California 94720 U.S.A. ABSTRACT Three new combinations are proposed for use in the upcoming re- vised Jepson Manual of California plants. Potentilla rimicola (Munz k. I.M. Johnston) Ertter replaces P. wheeleri subsp. rimicola, while Horkelia californica subsp. dissita (Crum) Ertter and subsp. fron- dosa (E.L. Greene) Ertter replace H. elata and H. frondosa respectively. KEY WORDS: Potentilla, Horkelia, Rosaceae, California, taxon- omy As a result of preparing the treatment of herbaceous Rosaceae for the forthcoming The Jepson Manual: Higher Plants of California, several new combinations were found to be necessary. Key characters and descriptions will be found in the Manual and are therefore not included here. Potentilla rimicola (Munz & I.M. Johnston) Ertter, comb, et stat. nov. BASIONYM: Potentilla wheeleri S. Wats. var. rimicola Munz & I.M. Johnston, Bull. S. Calif. Acad. Sci. 24:18. 1925. The Potentilla wheeleri complex (Rydberg's Subviscosae) includes a series of biogeographically interesting taxa on isolated mountain ranges in Cahfor- nia, Arizona, New Mexico, and northern Mexico. Typical P. wheeleri occurs from the southern Sierra Nevada to northern Baja California. Although the distinctiveness or circumscription of possible segregates var. paupercula Jep- son (Mount San Gorgonio, San Bernardino County, Cabfornia), P. luteosencea Rydb. (northern Baja California), or P. viscidula Rydb. (southern Arizona) are thus far uncertain, my initial studies of the complex indicate that at least var. rimicola is worth recognizing at the species level. A new combination is therefore needed at this time for the Jepson Manual. 420 Ertter: New combinations in Potentilla and Horkeha 421 The most distinctive difference between Potentilla nmicola and other mem- bers of the complex is that it grows in crevices of vertical rock faces; morpho- logical differences include pedicel length and characters of the fruit. Potentilla nmicola occurs only in the San Jacinto Mountains of Riverside County, Cah- fornia, and the Sierra San Pedro Martir of northern Baja California, Mexico. Although P. wheelen occurs in the same mountain ranges, no intergradation between the taxa has been found. The petrophytic habit is very unusual in Potentilla s. str. but characterizes many species of Ivesia Torrey k A. Gray (Ertter 1989). These same species of Ivesia differ from members of the P. wheelen complex in having pinnately rather than palmately divided leaves, but otherwise share several other intrigu- ing biogeographical and morphological similarities, including glandular hairs, ridged seeds, and ± recurved pedicels. Horkelia californica Cham. k Schldl. subsp. dissita (Crum) Ertter, comb, et stat. nov. BASIONYM: Potentilla elata E.L. Greene var. dissita Crum in Jepson, Fl. Calif. 2:197. 1936. Potentilla elata E.L. Greene, Pittonia 1:100. 1887. Potentilla califor- nica (Cham. k Schldl.) E.L. Greene var. elata (E.L. Greene) E.L. Greene, Fl. Franciscana 1:66. 1891. Horkeha elata (E.L. Greene) Rydb., Bull. Torrey Bot. Club 25:54. 1898. Horkelia glandulosa Eastw., Bull. Torrey Bot. Club 32:195. 1905. Horkelia californica Cham. k Schldl. subsp. frondosa (E.L. Greene) Ert- ter, comb, et stat. nov. BASIONYM: Potentilla frondosa E.L. Greene, Pittonia 1:300. 1889. Horkeha frondosa (E.L. Greene) Rydb., Bull. Tor- rey Bot. Club 25:54. 1898. Potentilla californica (Cham, k Schldl.) E.L. Greene var. frondosa (E.L. Greene) Jepson, Man. Fl. PI. Calif. 494. 1925. Although most treatments during the last few decades treat Horkeha (or Potentilla) californica, elata, and frondosa as distinct species, this glosses over the large number of intermediate specimens that do not fall conveniently into one taxon or another. Numerous collections from the North Coast ranges of California are particularly frustrating in their intermediacy between typical H. elata and typical H. californica (e.g., size of leaflets and degree of lobing). Intermediates in leaf morphology also blur the distinctiveness of H. frondosa, supporting the treatment of all three as infraspecific taxa. Unfortunately, the famibar epithet "elata" must be replaced, in that Po- tentilla elata E.L. Greene is a later homonym of P. elata Salisb. (Prodr. Stirp. Chap. Allerton 1796, p. 362). Britten (1916) argues convincingly that Sabs- bury's numerous names, although largely superfluous, should not be ignored. 422 PHYTOLOGIA volume 71(5):420-422 November 1991 Subspecies are used rather than varieties, both to parallel usage elsewhere in Horkelia and to avoid problems caused by the questionable identity of Poten- tilla glandulosa [/?] incisa Lindl. The possible synonymy with subsp. frondosa of this name and the equally problematical Horkelia grandis Hook. & Am. is discussed in detail by Keck (1938). LITERATURE CITED Britten, J. 1916. The plants of Salisbury's "Prodromus" (1796). J. Botany 54:57-65. Ertter, B. 1989. Revisionary studies in Ivesia (Rosaceae: Potentilleae). Syst. Bot. 14:231-244. Keck, D.D. 1938. Revision of Horkelia and Ivesia. Lloydia 1:75-142. 1938.