The wolf spider Pardosa sierra was described and illustrated by Banks in 1898 based on specimens from the Sierra de la Laguna, Baja California Sur. Later, two morphologically similar species, P. atromedia Banks 1904 from Claremont, California, and P. sura Chamberlin & Ivie 1941, also from California, were described. However, the latter two species were subsequently synonymized with P. sierra, due to similarities in male genitalia. In this study we test the species limits within this group. We suggest that the details of the epigynum are different enough among the genitalic morphs studied to consider them different species as originally designated. We conducted a morphological and genetic-distance analysis of a fragment of the cytochrome c oxidase subunit I gene sequences of some species of lapidicina group, as well as some sequences of Pardosa astrigera L. Koch 1878 from the GenBank database. Genetic analysis revealed greater genetic distances (GD) among haplotypes of P. sierra, P. atromedia, and P. sura (GD  =  0.053–0.069) than with other species of the lapidicina group. Moreover, P. sierra was closest to P. sura (GD  =  0.053), P. sura was closest to P. vadosa Barnes 1959 (GD  =  0.040), and P. atromedia was closest to P. steva Barnes 1959 (GD  =  0.052). Overall, morphological and genetic differences, and disjoint distributions, suggest that the synonymy of P. sierra, P. atromedia, and P. sura was in error, and that these “morphs” do indeed represent different species.