Genomic and morphological analysis of an American crow hybrid zone

Abstract

The Northwestern Crow (Corvus caurinus) and American Crow (Corvus brachyrhynchos) are continuously distributed sister taxa that lack reliable traditional characters for identification near their range boundary. We used genomic SNPs (nuDNA) and mtDNA to investigate the degree of genetic differentiation and hybridization between these crows. Our results indicate that American and Northwestern crows have distinct evolutionary histories, supported by two nuDNA ancestry clusters and two 1.1%-divergent mtDNA clades dating to the late Pleistocene, when glacial advances may have isolated crow populations in separate refugia. We document extensive current hybridization, with geographic overlap of mtDNA clades and admixture of nuDNA across >900 km of western Washington and western British Columbia. This broad hybrid zone consists of late-generation hybrids and backcrosses, but not recent (e.g., F1) hybrids. Nuclear DNA and mtDNA clines had concordant widths and were both centered in southwestern British Columbia, farther north than previously postulated. Overall, our results suggest a history of reticulate evolution in American and Northwestern crows. The two major hypotheses that have been put forward to explain the origin of this broad hybrid zone differ in both mechanism and timing. The first hypothesis is that post-glacial expansion allowed crows to expand from formerly isolated Pleistocene refugia into newly available habitats, leading to lineage fusion. The second is that the hybrid zone is a more recent artifact of anthropogenic habitat changes, resulting from the relatively recent forest fragmentation and land use changes associated with European colonization. Our objective was to differentiate between these two hypotheses by sequencing mtDNA from >160 years of museum specimens and assessing the timing of secondary contact within the hybrid zone. Sequencing a 90-bp fragment of mtDNA ND2 diagnostic for American and Northwestern haplogroups, we detected breeding season co-occurrence between American and Northwestern haplogroups at three different localities in 1889-1892, prior to the bulk of European-associated land use changes. We also detected geographic overlap of haplogroups in the southern Puget Sound of Washington in the 1850s, but not all of these individuals were sampled during the breeding season. We did not detect significant changes in haplogroup frequency within localities over time. Overall, our results are most consistent with the hybrid zone resulting from non-anthropogenic habitat changes since the last glacial maximum, or from anthropogenic influences of Native American/First Nations peoples prior to the late 19th century. Although our power to detect mtDNA haplogroup overlap and changes in haplogroup frequency over time was limited by sample size, using historical museum specimens is a direct and promising way of assessing changes in secondary contact over time. The broad, sigmoidal molecular cline across the hybrid zone is consistent with a general pattern of decreasing body size along the Pacific coast reported by an earlier morphological study. However, the data in the morphological study were not presented in a way that facilitates a quantitative comparison with the shape of the molecular cline. In this study, we reassessed morphological variation in American/Northwestern crows along the Pacific coast and across North America, and compared morphological and molecular variation along the Pacific coast using modern cline methods. The cline for body size was centered near Seattle, 355 km and 158 km south of the clines for mtDNA and nuDNA, respectively. The ± 2 LL intervals for cline width overlapped between body size and both the mtDNA and nuclear DNA clines. The smallest crows in North America occur in Alaska, coastal British Columbia, coastal Washington, and coastal Oregon, respectively, mirroring the American/Northwestern crow hybrid cline based on molecular data. The largest crows are resident in Florida. Overall, body size variation in the American/Northwestern crow complex provides a counterexample to Bergmann's rule, which states that animals have larger body size at higher latitudes and/or in regions with cooler climates. One alternative explanation for geographic variation in body size of American crows is ecological character displacement as a consequence of competition with other co-occurring Corvus species. The smaller and gregarious Fish Crow reaches its peak abundance in Florida, where American Crows are raven-like, achieving their largest body size and tending to be more solitary. Conversely, the larger Common Raven is abundant in Alaska, where it co-occurs coastally with the smallest American (Northwestern) crows.