Dense Geographic and Genomic Sampling Reveals Paraphyly and a Cryptic Lineage in a Classic Sibling Species Complex

Incomplete or geographically biased sampling poses significant problems for research in phylogeography, population genetics, phylogenetics, and species delimitation. Despite the power of using genome-wide genetic markers in systematics and related fields, approaches such as the multispecies coalescent remain unable to easily account for unsampled lineages. The Empidonax difficilis/Empidonax occidentalis complex of small tyrannid flycatchers (Aves: Tyrannidae) is a classic example of widely distributed species with limited phenotypic geographic variation that was broken into two largely cryptic (or "sibling") lineages following extensive study. Though the group is well-characterized north of the US Mexico border, the evolutionary distinctiveness and phylogenetic relationships of southern populations remain obscure. In this article, we use dense genomic and geographic sampling across the majority of the range of the E. difficilis/E. occidentalis complex to assess whether current taxonomy and species limits reflect underlying evolutionary patterns, or whether they are an artifact of historically biased or incomplete sampling. We find that additional samples from Mexico render the widely recognized species-level lineage E. occidentalis paraphyletic, though it retains support in the best-fit species delimitation model from clustering analyses. We further identify a highly divergent unrecognized lineage in a previously unsampled portion of the group's range, which a cline analysis suggests is more reproductively isolated than the currently recognized species E. difficilis and E. occidentalis. Our phylogeny supports a southern origin of these taxa. Our results highlight the pervasive impacts of biased geographic sampling, even in well-studied vertebrate groups like birds, and illustrate what is a common problem when attempting to define species in the face of recent divergence and reticulate evolution.