RADseq data reveal widespread historical introgression in four familiar North American songbirds.

Population genetic structure is influenced by a combination of contemporary and historical events; however, this structure can be complicated by ongoing gene flow. While it is well known that contemporary hybridization occurs frequently among many closely related species, it often remains uncertain as to which populations are involved in introgression events, and this can be even more difficult to infer when introgression is historical. Here we use restriction-site associated DNA sequencing to look at the level of introgression among four species of songbirds in North America: the black-capped, mountain, boreal, and chestnut-backed chickadee. Samples from both sympatric and allopatric sites across the species' ranges supported limited ongoing mixing among the four species with Bayesian clustering and principal component analyses. In contrast, f4-statistics and admixture graphs revealed extensive historical introgression among geographically structured populations. Almost all historical admixture events were among populations west of the Rocky Mountains, and almost all populations west of the Rocky Mountains, excluding island and coastal populations, showed evidence of historical admixture. The inclusion of all four chickadee species proved crucial in differentiating which species were involved in hybridization events to avoid erroneous conclusions. Taken together, the results suggest a complex pattern of divergence with gene flow.

Hybridization between closely related songbirds is related to human habitat disturbance.

Human habitat disturbances can promote hybridization between closely related, but typically reproductively isolated, species. We explored whether human habitat disturbances are related to hybridization between two closely related songbirds, black-capped and mountain chickadees, using both genomic and citizen science data sets. First, we genotyped 409 individuals from across both species' ranges using reduced-representation genome sequencing and compared measures of genetic admixture to a composite measure of human landscape disturbance. Then, using eBird observations, we compared human landscape disturbance values for sites where phenotypically diagnosed hybrids were observed to locations where either parental species was observed to determine whether hybrid chickadees are reported in more disturbed areas. We found that hybridization between black-capped and mountain chickadees positively correlates with human habitat disturbances. From genomic data, we found that (1) hybrid index (HI) significantly increased with habitat disturbance, (2) more hybrids were sampled in disturbed habitats, (3) mean HIs were higher in disturbed habitats versus wild habitats, and (4) hybrids were detected in habitats with significantly higher disturbance values than parentals. Using eBird data, we found that both hybrid and black-capped chickadees were significantly more disturbance-associated than mountain chickadees. Surprisingly, we found that nearly every black-capped chickadee we sampled contained some proportion of hybrid ancestry, while we detected very few mountain chickadee backcrosses. Our results highlight that hybridization between black-capped and mountain chickadees is widespread, but initial hybridization is rare (few F1s were detected). We conclude that human habitat disturbances can erode pre-zygotic reproductive barriers between chickadees and that post-zygotic isolation is incomplete. Understanding what becomes of recently hybridizing species following large-scale habitat disturbances is a new, but pressing, consideration for successfully preserving genetic biodiversity in a rapidly changing world.

Comparison of genotyping by sequencing procedures to determine population genetic structure.

Advancements in technology over the past few decades have resulted in the development of genome sequencing at lower costs. Protocols, costs, and the amount of data produced by different sequencing technologies are highly variable. Ion Torrent and Illumina sequencing instruments are two sequencing technologies which use very similar library preparation procedures. Enzymatic combinations can be changed in genotyping by sequencing (GbS) library protocols without significant adjustments. To compare the outputs from two different GbS procedures, we sequenced samples of two sister species of yellow-nosed albatross collected at multiple geographic locations. The data sets involving different sequencing instruments and enzymatic combinations were analysed using the Stacks pipeline and aligned to the same reference genome. Both procedures identified the same genetic clusters separating Atlantic and Indian yellow-nosed albatross and substructure within Indian yellow-nosed albatross.

Sympatry leads to reduced body condition in chickadees that occasionally hybridize.

Both abiotic and biotic drivers influence species distributions. Abiotic drivers such as climate have received considerable attention, even though biotic drivers such as hybridization often interact with abiotic drivers. We sought to explore the (1) costs of co-occurrence for ecologically similar species that hybridize and (2) associations between ecological factors and condition to understand how abiotic and biotic factors influence species distributions. For two closely related and ecologically similar songbirds, black-capped and mountain chickadees, we characterized body condition, as a proxy for fitness, using a 1358-individual range-wide dataset. We compared body condition in sympatry and allopatry with several abiotic and biotic factors using species-specific generalized linear mixed models. We generated genomic data for a subset of 217 individuals to determine the extent of hybridization-driven admixture in our dataset. Within this data subset, we found that ~11% of the chickadees had hybrid ancestry, and all hybrid individuals had typical black-capped chickadee plumage. In the full dataset, we found that birds of both species, independent of demographic and abiotic factors, had significantly lower body condition when occurring in sympatry than birds in allopatry. This could be driven by either the inclusion of cryptic, likely poor condition, hybrids in our full dataset, competitive interactions in sympatry, or range edge effects. We are currently unable to discriminate between these mechanisms. Our findings have implications for mountain chickadees in particular, which will encounter more black-capped chickadees as black-capped chickadee ranges shift upslope and could lead to local declines in mountain chickadee populations.

Habitat-linked genetic structure for white-crowned sparrow (Zonotrichia leucophrys): Local factors shape population genetic structure.

Ecological, environmental, and geographic factors all influence genetic structure. Species with broad distributions are ideal systems because they cover a range of ecological and environmental conditions allowing us to test which components predict genetic structure. This study presents a novel, broad geographic approach using molecular markers, morphology, and habitat modeling to investigate rangewide and local barriers causing contemporary genetic differentiation within the geographical range of three white-crowned sparrow (Zonotrichia leucophrys) subspecies: Z. l. gambelii, Z. l. oriantha, and Z. l. pugetensis. Three types of genetic markers showed geographic distance between sampling sites, elevation, and ecosystem type are key factors contributing to population genetic structure. Microsatellite markers revealed white-crowned sparrows do not group by subspecies, but instead indicated four groupings at a rangewide scale and two groupings based on coniferous and deciduous ecosystems at a local scale. Our analyses of morphological variation also revealed habitat differences; sparrows from deciduous ecosystems are larger than individuals from coniferous ecosystems based on principal component analyses. Habitat modeling showed isolation by distance was prevalent in describing genetic structure, but isolation by resistance also had a small but significant influence. Not only do these findings have implications concerning the accuracy of subspecies delineations, they also highlight the critical role of local factors such as habitat in shaping contemporary population genetic structure of species with high dispersal ability.

Geolocator tagging links distributions in the non-breeding season to population genetic structure in a sentinel North Pacific seabird.

We tested the hypothesis that segregation in wintering areas is associated with population differentiation in a sentinel North Pacific seabird, the rhinoceros auklet (Cerorhinca monocerata). We collected tissue samples for genetic analyses on five breeding colonies in the western Pacific Ocean (Japan) and on 13 colonies in the eastern Pacific Ocean (California to Alaska), and deployed light-level geolocator tags on 12 eastern Pacific colonies to delineate wintering areas. Geolocator tags were deployed previously on one colony in Japan. There was strong genetic differentiation between populations in the eastern vs. western Pacific Ocean, likely due to two factors. First, glaciation over the North Pacific in the late Pleistocene might have forced a southward range shift that historically isolated the eastern and western populations. And second, deep-ocean habitat along the northern continental shelf appears to act as a barrier to movement; abundant on both sides of the North Pacific, the rhinoceros auklet is virtually absent as a breeder in the Aleutian Islands and Bering Sea, and no tagged birds crossed the North Pacific in the non-breeding season. While genetic differentiation was strongest between the eastern vs. western Pacific, there was also extensive differentiation within both regional groups. In pairwise comparisons among the eastern Pacific colonies, the standardized measure of genetic differentiation (FꞌST) was negatively correlated with the extent of spatial overlap in wintering areas. That result supports the hypothesis that segregation in the non-breeding season is linked to genetic structure. Philopatry and a neritic foraging habit probably also contribute to the structuring. Widely distributed, vulnerable to anthropogenic stressors, and exhibiting extensive genetic structure, the rhinoceros auklet is fully indicative of the scope of the conservation challenges posed by seabirds.

Landscape effects on the contemporary genetic structure of Ruffed Grouse (Bonasa umbellus) populations.

The amount of dispersal that occurs among populations can be limited by landscape heterogeneity, which is often due to both natural processes and anthropogenic activity leading to habitat loss or fragmentation. Understanding how populations are structured and mapping existing dispersal corridors among populations is imperative to both determining contemporary forces mediating population connectivity, and informing proper management of species with fragmented populations. Furthermore, the contemporary processes mediating gene flow across heterogeneous landscapes on a large scale are understudied, particularly with respect to widespread species. This study focuses on a widespread game bird, the Ruffed Grouse (Bonasa umbellus), for which we analyzed samples from the western extent of the range. Using three types of genetic markers, we uncovered multiple factors acting in concert that are responsible for mediating contemporary population connectivity in this species. Multiple genetically distinct groups were detected; microsatellite markers revealed six groups, and a mitochondrial marker revealed four. Many populations of Ruffed Grouse are genetically isolated, likely by macrogeographic barriers. Furthermore, the addition of landscape genetic methods not only corroborated genetic structure results, but also uncovered compelling evidence that dispersal resistance created by areas of unsuitable habitat is the most important factor mediating population connectivity among the sampled populations. This research has important implications for both our study species and other inhabitants of the early successional forest habitat preferred by Ruffed Grouse. Moreover, it adds to a growing body of evidence that isolation by resistance is more prevalent in shaping population structure of widespread species than previously thought.

Spatiotemporal Patterns of Infection for Emerging Larval Liver Fluke ( Dicrocoelium dendriticum) in Three Species of Land Snail in Southern Alberta, Canada.

The control of emerging parasites requires a fundamental knowledge of where and when rates of transmission are high. Data on spatiotemporal patterns of infection are challenging to obtain, particularly for complex life cycle parasites that involve transmission into multiple obligate hosts. The lancet liver fluke, Dicrocoelium dendriticum, has a long history of colonization outside its native host and geographical range in continental Europe. Infection patterns involving adult and metacercarial stages have been characterized for this trematode in a region of emergence in western Canada within co-grazing herbivores and ants, but infection patterns in snail intermediate hosts in this region are unknown. We combined spatiotemporal prevalence surveys with sequence analyses of the cytochrome c oxidase subunit 1 ( COI) barcoding gene from samples of sporocyst tissue in infected snails to confirm that D. dendriticum utilizes 3 sympatric species of Oreohelid land snail ( Oreohelix subrudis, Oreohelix sp., and Oreohelix cooperi) as first intermediate host. Mean prevalence within a total sample of 900 adult snails collected over 1 field season from 6 sites was 9.9 ± 2.4%. For each species of snail, prevalence ranged between 5-30% within monthly samples, with peaks in mid-summer followed by declines in fall. Between-site variation in prevalence was low and non-significant, implying that rates of transmission of D. dendriticum miracidia from domestic stock and wildlife into snails are similar within localized sites, despite high variation in local habitat characteristics and in the structure of the definitive host community.

Population Genetics and Speciation of Yellow-Bellied, Red-Naped, and Red-Breasted Sapsuckers (Sphyrapicus varius, S. nuchalis, and S. ruber).

The root of understanding speciation lies in determining the forces which drive it. In many closely-related species, including Sphyrapicus varius, S. nuchalis, and S. ruber, it is assumed that speciation occurred due to isolation in multiple Pleistocene refugia. We used genetic data from 457 samples at the control region (CR), cytochrome oxidase I (COI), and chromo-helicase DNA binding protein (CHD1Z) to examine range-wide population genetic structure and differentiation amongst these 3 species across each species' breeding range. In addition, we modeled these species' ecological niches for the Holocene (~6000 years ago), Last Glacial Maximum (~22000 years ago), and Last Interglacial (~120000-140000 years ago) to determine if Pleistocene glaciations could have contributed to allopatric distributions, therefore allowing these groups to differentiate. Population genetic data show a potential Pleistocene refugium in Haida Gwaii, an east-west split among S. varius, and low genetic differentiation within each species. Our CR data show some polyphyly, while COI and CHD1Z data show differentiation among species using composite genotypes. Ecological Niche Modeling shows a large amount of niche overlap at each time period suggesting that S. varius, S. nuchalis, and S. ruber may not have been completely allopatric, and these species likely had repeated intermittent contact. Our data support the growing body of research that suggests differentiation despite gene flow.

Multilocus genetic analyses and spatial modeling reveal complex population structure and history in a widespread resident North American passerine (Perisoreus canadensis).

An increasing body of studies of widely distributed, high latitude species shows a variety of refugial locations and population genetic patterns. We examined the effects of glaciations and dispersal barriers on the population genetic patterns of a widely distributed, high latitude, resident corvid, the gray jay (Perisoreus canadensis), using the highly variable mitochondrial DNA (mtDNA) control region and microsatellite markers combined with species distribution modeling. We sequenced 914 bp of mtDNA control region for 375 individuals from 37 populations and screened seven loci for 402 individuals from 27 populations across the gray jay range. We used species distribution modeling and a range of phylogeographic analyses (haplotype diversity, ΦST, SAMOVA, FST, Bayesian clustering analyses) to examine evolutionary history and population genetic structure. MtDNA and microsatellite markers revealed significant genetic differentiation among populations with high concordance between markers. Paleodistribution models supported at least five potential areas of suitable gray jay habitat during the last glacial maximum and revealed distributions similar to the gray jay's contemporary during the last interglacial. Colonization from and prolonged isolation in multiple refugia is evident. Historical climatic fluctuations, the presence of multiple dispersal barriers, and highly restricted gene flow appear to be responsible for strong genetic diversification and differentiation in gray jays.

Telomere length measurement by qPCR in birds is affected by storage method of blood samples.

Given the potential role of telomeres as biomarkers of individual health and ageing, there is an increasing interest in studying telomere dynamics in a wider range of taxa in the fields of ecology and evolutionary biology. Measuring telomere length across the lifespan in wild animal systems is essential for testing these hypotheses, and may be aided by archived blood samples collected as part of longitudinal field studies. However, sample collection, storage, and DNA extraction methods may influence telomere length measurement, and it may, therefore, be difficult to balance consistency in sampling protocol with making the most of available samples. We used two complementary approaches to examine the impacts of sample storage method on measurements of relative telomere length (RTL) by qPCR, particularly focusing on FTA (Flinders Technology Associates) cards as a long-term storage solution. We used blood samples from wandering albatrosses collected over 14 years and stored in three different ways (n = 179), and also blood samples from captive zebra finches (n = 30) that were each stored using three different methods. Sample storage method influenced RTL in both studies, and samples on FTA cards had significantly shorter RTL measurements. There was no significant correlation between RTL measured in zebra finch blood on FTA cards and the same samples stored either as frozen whole blood or as extracted DNA. These results highlight the importance of consistency of sampling protocol, particularly in the context of long-term field studies, and suggest that FTA cards should not be used as a long-term storage solution to measure RTL without validation.

Population genetic isolation and limited connectivity in the purple finch (Haemorhous purpureus).

Using a combination of mitochondrial and z-linked sequences, microsatellite data, and spatio-geographic modeling, we examined historical and contemporary factors influencing the population genetic structure of the purple finch (Haemorhous purpureus). Mitochondrial DNA data show the presence of two distinct groups corresponding to the two subspecies, H. p. purpureus and H. p. californicus. The two subspecies likely survived in separate refugia during the last glacial maximum, one on the Pacific Coast and one east of the Rocky Mountains, and now remain distinct lineages with little evidence of gene flow between them. Southwestern British Columbia is a notable exception, as subspecies mixing between central British Columbia and Vancouver Island populations suggests a possible contact zone in this region. Z-linked data support two mitochondrial groups; however, Coastal Oregon and central British Columbia sites show evidence of mixing. Contemporary population structure based on microsatellite data identified at least six genetic clusters: three H. p. purpureus clusters, two H. p. californicus clusters, and one mixed cluster, which likely resulted from high site fidelity and isolation by distance, combined with sexual selection on morphological characters reinforcing subspecies differences.

Gene Flow of a Forest-Dependent Bird across a Fragmented Landscape.

Habitat loss and fragmentation can affect the persistence of populations by reducing connectivity and restricting the ability of individuals to disperse across landscapes. Dispersal corridors promote population connectivity and therefore play important roles in maintaining gene flow in natural populations inhabiting fragmented landscapes. In the prairies, forests are restricted to riparian areas along river systems which act as important dispersal corridors for forest dependent species across large expanses of unsuitable grassland habitat. However, natural and anthropogenic barriers within riparian systems have fragmented these forested habitats. In this study, we used microsatellite markers to assess the fine-scale genetic structure of a forest-dependent species, the black-capped chickadee (Poecile atricapillus), along 10 different river systems in Southern Alberta. Using a landscape genetic approach, landscape features (e.g., land cover) were found to have a significant effect on patterns of genetic differentiation. Populations are genetically structured as a result of natural breaks in continuous habitat at small spatial scales, but the artificial barriers we tested do not appear to restrict gene flow. Dispersal between rivers is impeded by grasslands, evident from isolation of nearby populations (~ 50 km apart), but also within river systems by large treeless canyons (>100 km). Significant population genetic differentiation within some rivers corresponded with zones of different cottonwood (riparian poplar) tree species and their hybrids. This study illustrates the importance of considering the impacts of habitat fragmentation at small spatial scales as well as other ecological processes to gain a better understanding of how organisms respond to their environmental connectivity. Here, even in a common and widespread songbird with high dispersal potential, small breaks in continuous habitats strongly influenced the spatial patterns of genetic variation.

Limited geographic genetic structure detected in a widespread Palearctic corvid, Nucifraga caryocatactes.

The Eurasian or spotted nutcracker (Nucifraga caryocatactes) is a widespread resident corvid found throughout the Palearctic from Central Europe to Japan. Characterized by periodic bouts of irruptive dispersal in search of Pinus seed crops, this species has potential for high levels of gene flow across its range. Previous analysis of 11 individuals did not find significant range-wide population genetic structure. We investigated population structure using 924 base pairs of mitochondrial DNA control region sequence data from 62 individuals from 12 populations distributed throughout the nutcracker's range. We complemented this analysis by incorporating additional genetic data from previously published sequences. High levels of genetic diversity and limited population genetic structure were detected suggesting that potential barriers to dispersal do not restrict gene flow in nutcrackers.

Molecular markers reveal limited population genetic structure in a North American corvid, Clark's nutcracker (Nucifraga columbiana).

The genetic impact of barriers and Pleistocene glaciations on high latitude resident species has not been widely investigated. The Clark's nutcracker is an endemic North American corvid closely associated with Pinus-dominated forests. The nutcracker's encompasses known barriers to dispersal for other species, and glaciated and unglaciated areas. Clark's nutcrackers also irruptively disperse long distances in search of pine seed crops, creating the potential for gene flow among populations. Using the highly variable mitochondrial DNA control region, seven microsatellite loci, and species distribution modeling, we examined the effects of glaciations and dispersal barriers on population genetic patterns and population structure of nutcrackers. We sequenced 900 bp of mitochondrial control region for 169 individuals from 15 populations and analysed seven polymorphic microsatellite loci for 13 populations across the Clark's nutcracker range. We used species distribution modeling and a range of phylogeographic analyses to examine evolutionary history. Clark's nutcracker populations are not highly differentiated throughout their range, suggesting high levels of gene flow among populations, though we did find some evidence of isolation by distance and peripheral isolation. Our analyses suggested expansion from a single refugium after the last glacial maximum, but patterns of genetic diversity and paleodistribution modeling of suitable habitat were inconclusive as to the location of this refugium. Potential barriers to dispersal (e.g. mountain ranges) do not appear to restrict gene flow in Clark's nutcracker, and postglacial expansion likely occurred quickly from a single refugium located south of the ice sheets.

Sampling for microsatellite-based population genetic studies: 25 to 30 individuals per population is enough to accurately estimate allele frequencies.

One of the most common questions asked before starting a new population genetic study using microsatellite allele frequencies is "how many individuals do I need to sample from each population?" This question has previously been answered by addressing how many individuals are needed to detect all of the alleles present in a population (i.e. rarefaction based analyses). However, we argue that obtaining accurate allele frequencies and accurate estimates of diversity are much more important than detecting all of the alleles, given that very rare alleles (i.e. new mutations) are not very informative for assessing genetic diversity within a population or genetic structure among populations. Here we present a comparison of allele frequencies, expected heterozygosities and genetic distances between real and simulated populations by randomly subsampling 5-100 individuals from four empirical microsatellite genotype datasets (Formica lugubris, Sciurus vulgaris, Thalassarche melanophris, and Himantopus novaezelandia) to create 100 replicate datasets at each sample size. Despite differences in taxon (two birds, one mammal, one insect), population size, number of loci and polymorphism across loci, the degree of differences between simulated and empirical dataset allele frequencies, expected heterozygosities and pairwise F(ST) values were almost identical among the four datasets at each sample size. Variability in allele frequency and expected heterozygosity among replicates decreased with increasing sample size, but these decreases were minimal above sample sizes of 25 to 30. Therefore, there appears to be little benefit in sampling more than 25 to 30 individuals per population for population genetic studies based on microsatellite allele frequencies.

Genetic signals of demographic expansion in Downy Woodpecker (Picoides pubescens) after the last North American glacial maximum.

The glacial cycles of the Pleistocene have been recognized as important, large-scale historical processes that strongly influenced the demographic patterns and genetic structure of many species. Here we present evidence of a postglacial expansion for the Downy Woodpecker (Picoides pubescens), a common member of the forest bird communities in North America with a continental distribution. DNA sequences from the mitochondrial tRNA-Lys, and ATPase 6 and 8 genes, and microsatellite data from seven variable loci were combined with a species distribution model (SDM) to infer possible historical scenarios for this species after the last glacial maximum. Analyses of Downy Woodpeckers from 23 geographic areas suggested little differentiation, shallow genealogical relationships, and limited population structure across the species' range. Microsatellites, which have higher resolution and are able to detect recent differences, revealed two geographic groups where populations along the eastern edge of the Rocky Mountains (Montana, Utah, Colorado, and southern Alberta) were genetically isolated from the rest of the sampled populations. Mitochondrial DNA, an important marker to detect historical patterns, recovered only one group. However, populations in Idaho and southeast BC contained high haplotype diversity and, in general were characterized by the absence of the most common mtDNA haplotype. The SDM suggested several areas in the southern US as containing suitable Downy Woodpecker habitat during the LGM. The lack of considerable geographic structure and the starburst haplotype network, combined with several population genetic tests, suggest a scenario of demographic expansion during the last part of Pleistocene and early Holocene.

Effects of Pleistocene glaciations on population structure of North American chestnut-backed chickadees.

The postglacial recolonization of northern North America was heavily influenced by the Pleistocene glaciation. In the Pacific Northwest, there are two disjunct regions of mesic temperate forest, one coastal and the other interior. The chestnut-backed chickadee is one of the species associated with this distinctive ecosystem. Using seven microsatellite markers we found evidence of population structure among nine populations of chestnut-backed chickadees. High levels of allelic variation were found in each of the populations. Northern British Columbia and central Alaska populations contained a large number of private alleles compared to other populations, including those from unglaciated regions. The disjunct population in the interior was genetically distinct from the coastal population. Genetic and historical records indicate that the interior population originated from postglacial inland dispersal. Population structuring was found within the continuous coastal population, among which the peripheral populations, specifically those on the Queen Charlotte Islands and the central Alaska mainland, were genetically distinct. The pattern of population structure among contemporary chickadee populations is consistent with a pioneer model of recolonization. The persistence of genetic structure in western North American chestnut-backed chickadees may be aided by their sedentary behaviour, linear distribution, and dependence on cedar-hemlock forests.

Rapid divergence and postglacial colonization in western North American Steller's jays (Cyanocitta stelleri).

Post-Pleistocene avian colonization of deglaciated North America occurred from multiple refugia, including a coastal refugium in the northwest. The location of a Pacific Coastal refugium is controversial; however, multiple lines of evidence suggest that it was located near the Queen Charlotte Islands (also known as Haida Gwaii). The Queen Charlotte Islands contain a disproportionately large number of endemic plants and animals including the Steller's jay Cyanocitta stelleri carlottae. Using five highly variable microsatellite markers, we studied population structure among eight populations of Steller's jay (N = 150) from geographical areas representing three subspecies in western North America: C. s. carlottae, C. s. stelleri and C. s. annectens. Microsatellite analyses revealed genetic differentiation between each of the three subspecies, although more extensive sampling of additional C. s. annectens populations is needed to clarify the level of subspecies differentiation. High levels of population structure were found among C. s. stelleri populations with significant differences in all but two pairwise comparisons. A significant isolation by distance pattern was observed amongst populations in the Pacific Northwest and Alaska. In the C. s. carlottae population, there was evidence of reduced genetic variation, higher number of private alleles than northern C. s. stelleri populations and higher levels of divergence between Queen Charlotte Island and other populations. We were unable to reject the hypothesis that the Queen Charlotte Islands served as a refugium during the Pleistocene. Steller's jay may have colonized the Queen Charlotte Islands near the end of the last glaciation or persisted throughout the Pleistocene, and this subspecies may thus represent a glacial relic. The larger number of private alleles, despite reduced genetic variation, morphological distinctiveness and high divergence from other populations suggests that the Queen Charlotte Island colonization pre-dates that of the mainland. Furthermore, our results show rapid divergence in Steller's jay populations on the mainland following the retreat of the ice sheets.

Perils of paralogy: using HSP70 genes for inferring organismal phylogenies.

Conserved genes have found their way into the mainstream of molecular systematics. Many of these genes are members of multigene families. A difficulty with using single genes of multigene families for phylogenetic inference is that genes from one species may be paralogous to those from another taxon. We focus attention on this problem using heat shock 70 (HSP70) genes. Using polymerase chain reaction techniques with genomic DNA, we isolated and sequenced 123 distinct sequences from 12 species of sharks. Phylogenetic analysis indicated that the sequences cluster with constituitively expressed cytoplasmic heat shock-like genes. Three highly divergent gene clades were sampled. A number of similar sequences were sampled from each species within each distinct gene clade. Comparison of published species trees with an HSP70 gene tree inferred using Bayesian phylogenetic analysis revealed several cases of gene duplication and differential sorting of gene lineages within this group of sharks. Gene tree parsimony based on the objective criteria of duplication and losses showed that previously published hypotheses of species relationships and two novel hypothesis based on Bayesian phylogenetics were concordant with the history of HSP70 gene duplication and loss. By contrast, two published hypotheses based on morphological data were not significantly different from the null hypothesis of a random association between species relatedness and the HSP70 gene tree. These results suggest that gene tree parsimony using data from multigene families can be used for inferring species relationships or testing published alternative hypotheses. More importantly, the results suggest that systematic studies relying on phylogenetic inferences from HSP70 genes may by plagued by unrecognized paralogy of sampled genes. Our results underscore the distinction between gene and species trees and highlight an underappreciated source of discordance between gene trees and organismal phylogeny, i.e., unrecognized paralogy of sampled genes.