Complementary Roles of Phenotype and Genotype in Subspecies Delimitation

Recent exchanges in this journal about subspecies taxonomy (Fredrickson et al. 2015; Weckworth et al. 2015; Cronin et al. 2015) signal an increased attention to this Linnaean rank, yet it is clear from some of the discourse that misconceptions about subspecies definitions persist. We emphasize that our goal is not to evaluate the validity of named subspecies of the gray wolf (Canis lupus), even if we refer to this particular example because it was the basis of the published exchange. Rather, our goals are to outline current thinking and methodology of practitioners who use subspecific taxonomy and to highlight the difficulty, short of further theory (e.g., development of a sound null model), inherent in use of neutral genetic markers to establish the validity of subspecies. Wilson and Brown’s (1953) commentary about the futility of subspecies taxonomy ignited a controversy that flared then in the pages of Systematic Zoology (now Systematic Biology) and that continues today (e.g., Winker and Haig 2010, a special issue of the Auk in 1982, and, most recently, Taylor et al. 2017). Wilson and Brown’s point of view was easy to understand because in that era many subspecies were being described for trivial or inconsistent reasons and with no little effort to establish standards. Nevertheless, we think that their commentary went too far in its advocacy. Wilson and Brown argued that the taxonomic rank ought to be discarded because some taxonomists applied it incorrectly, an argument analogous to decrying null hypothesis tests because many practitioners fail to use statistics properly. In any classification scheme that attempts to impose a typology on a continuum, the inevitable problems that arise are not necessarily reasons to abandon the system.