The systematics of Speyeria butterflies has historically been complicated by intraspecific variability that has challenged efforts to delimit species and reconstruct phylogenies. Our study presents a phylogenetic comparison of genomic single nucleotide polymorphisms (SNPs) and mitochondrial COI gene sequences, with comprehensive taxon sampling that includes 15 species and 46 subspecies. Increased sampling of genetic markers and taxa improved the match between genetic clusters, obtained with both phylogenetic and cluster‐based analyses, and species previously detected using morphology, as well as showing two species delimitations that may need revision. We also recovered extensive mitonuclear discordance between genomic SNPs and the COI gene, confirming that mitochondrial DNA does not reliably identify several species at broad geographic scales. Resolution of the relationships of Speyeria species demonstrates the importance of sampling variation across the whole genome, and provides an essential foundation for understanding the evolution of this charismatic clade of North American butterflies.

Recent advances in both genomics and ecological modelling present new, multidisciplinary opportunities for resolving species boundaries and understanding the mechanisms that maintain their integrity in regions of contact. Here, we use a combination of high-throughput DNA sequencing and ecological niche modelling to resolve species boundaries and niche divergence within the Speyeria atlantis-hesperis (Lepidoptera: Nymphalidae) complex, a confusing group of North American butterflies. This complex is notorious for its muddled species delimitations, morphological ambiguity, and extensive mitonuclear discordance. Our admixture and multispecies coalescent-based analyses of single nucleotide polymorphisms identified substantial divergences between S. atlantis and S. hesperis in areas of contact, as well as between distinct northern and southern lineages within S. hesperis. Our results also provide evidence of past introgression relating to another species, S. zerene, which previous work has shown to be more distantly related to the S. atlantis-hesperis complex. We then used ecological models to predict habitat suitability for each of the three recovered genomic lineages in the S. atlantis-hesperis complex and assessed their pairwise niche divergence. These analyses resolved that these three lineages are significantly diverged in their respective niches and are not separated by discontinuities in suitable habitat that might present barriers to gene flow. We therefore infer that ecologically-mediated selection resulting in disparate habitat associations is a principal mechanism reinforcing their genomic integrity. Overall, our results unambiguously support significant evolutionary and ecological divergence between the northern and southern lineages of S. hesperis, sufficient to recognize the southern evolutionary lineage as a distinct species, called S. nausicaa based on taxonomic priority.

Species delimitation is essential to understanding and categorizing our planet's biodiversity, particularly amidst rapid changes to environmental conditions and natural landscapes. However, the process of speciation is heterogenous and often complex, and robust characterization of species boundaries has remained a challenge for many taxa. Recent advances in both genomics and ecological modelling have been a boon for research focused on population dynamics, and present new, multidisciplinary opportunities for clarifying species boundaries in taxa that have been difficult to classify otherwise. Here, we present an approach to combining ecological niche models with next-generation sequence data to aid in integrated species delimitation. We apply this approach to the Speyeria atlantis-hesperis (Lepidoptera: Nymphalidae) species complex, which is notorious for its muddled species delimitations, morphological variation and mito-nuclear discordance. Using genomic SNPs, we recovered substantial divergence, not only between S. hesperis and S. atlantis, but also within S. hesperis, which may be attributed to a combination of past introgression with another species, S. zerene, and post-glacial range expansion. We then applied niche modelling to assess ecological divergence and barriers to gene flow among the recovered genomic lineages. Results of these analyses suggest that adaptation to ecological conditions is hindering contemporary gene flow between northern and southern populations of S. hesperis, contributing to and reinforcing their genetic integrity. We suggest that the current species delimitation of S. hesperis should be revised, and demonstrate the utility of an approach to integrated species delimitation that combines ecological and genomic data and reconciles related species concepts.