The subspecies concept has received considerable debate throughout the past century. Subspecies were originally used to delineate potential incipient species, but were later employed to simply capture geographical variation. There is a recent trend to eliminate the trinomial in light of new evidence. Discrete, diagnosable lineages are elevated to specific status, while those that show clinal variation and/or appear to represent ecological pattern classes are placed in synonymy with the parent species and the subspecific epithets are disregarded. Here, I examine the species boundaries of nightsnakes (Hypsiglena torquata) using standard phylogeographic methods and mtDNA data from 178 individuals. Previously, seventeen subspecies of H. torquata were described. In this study, I recognize six species in what was previously considered H. torquata: one is novel, two were previously recognized subspecies, while the remaining three are wide-spread, polymorphic lineages, composed of multiple subspecies. I make the case to maintain the subspecific lineages in these wide-ranging species because they are geographically cohesive, morphologically discrete, and may represent incipient species within each complex, which have not yet achieved speciation. These subspecies are maintained, not only pending future investigations, but because they provide a useful identity for the taxonomy of this diverse lineage.