Neglected taxonomy of rare desert fishes: congruent evidence for two species of leatherside chub.

Conservation biologists rely heavily on taxonomy to set the scope for biological monitoring and recovery planning of rare or threatened species. Yet, taxonomic boundaries are seldom evaluated as falsifiable hypotheses that can be statistically tested. Here, we examine species boundaries in leatherside chub (Teleostei, Cyprinidae), an imperiled desert fish native to the Bonneville Basin and upper Snake River drainages of western North America. Recent molecular data hint that this fish could be composed of two distinct taxa that are geographically separated into northern and southern species. To formally test this hypothesis, we evaluated leatherside chub using several different categories of species concepts, including criteria dependent on phylogenetic, morphological, and ecological data. We found that leatherside chub is composed of two reciprocally monophyletic clades (candidate species) characterized by numerous fixed genetic differences for both mitochondrial and nuclear DNA markers; mtDNA sequence divergence between the two clades approached 8%. The candidate species also showed significant differences in cranial shape, revealed by morphometric analysis. Finally, controlled growth and foraging experiments using representative populations from each clade show that candidate species appear to be locally adapted to the thermal environments where they now occur. Combined, these three lines of evidence support the hypothesis that leatherside chub is composed of two species. Moreover, all lines of evidence place these two species within the genusLepidomeda, a group consisting of three additional species of endangered spinedace fishes, and one extinct species, all native to the Colorado River system. Hence, we elevate the two clades of leatherside chub to distinct species status (Lepidomeda copeiin the north andL. aliciaein the south), and argue that each warrants independent conservation and recovery action.