Do substitute species help or hinder endangered species management

Abstract Substitute species (common species used to represent endangered species) are used to evaluate a range of conservation strategies globally. However, the effectiveness of this approach has not been empirically evaluated. We leveraged a large-scale habitat restoration experiment to test the validity of the substitute species concept. We selected a common butterfly, Satyrodes appalachia , that is on first inspection as near a substitute as possible - it is closely related to, overlaps in distribution, habitat requirements, host use, and life history with Neonympha mitchellii francisci , an endangered butterfly. We integrated small-scale measures of behavior, habitat preference, and demography of both species in our test, demonstrating that subtle differences between two species cause the substitute relationship to fail. Despite nearly identical habitat requirements, we found the endangered butterfly used different host plants, had higher larval survival in restored sites, and was found in more open habitat than the common butterfly. These differences added up to differences in abundances; the endangered species was more abundant than the common species in restored sites, the opposite was true in un-restored sites. Management decisions based on unvalidated substitute species run the risk of doing more harm than good for endangered species conservation. Instead, using experiments to evaluate a target species' response to management will result in effective recovery strategies.