Integrative tracking methods elucidate the evolutionary dynamics of a migratory divide

Migratory divides, the boundary between adjacent bird populations that migrate in different directions, are of considerable interest to evolutionary biologists because of their alleged role in speciation of migratory birds. However, the small size of many passerines has traditionally limited the tools available to track populations and as a result, restricted our ability to study how reproductive isolation might occur across a divide. Here, we integrate multiple approaches by using genetic, geolocator, and morphological data to investigate a migratory divide in hermit thrushes (Catharus guttatus). First, high genetic divergence between migratory groups indicates the divide is a region of secondary contact between historically isolated populations. Second, despite low sample sizes, geolocators reveal dramatic differences in overwintering locations and migratory distance of individuals from either side of the divide. Third, a diagnostic genetic marker that proved useful for tracking a key population suggests a likely intermediate nonbreeding location of birds from the hybrid zone. This finding, combined with lower return rates from this region, is consistent with comparatively lower fitness of hybrids, which is possibly due to this intermediate migration pattern. We discuss our results in the context of reproductive isolating mechanisms associated with migration patterns that have long been hypothesized to promote divergence across migratory divides.     

A Continent-Wide Migratory Divide in North American Breeding Barn Swallows (Hirundo rustica)

Populations of most North American aerial insectivores have undergone steep population declines over the past 40 years but the relative importance of factors operating on breeding, wintering, or stopover sites remains unknown. We used archival light-level geolocators to track the phenology, movements and winter locations of barn swallows (Hirdundo rustica; n = 27) from populations across North America to determine their migratory connectivity. We identified an east-west continental migratory divide for barn swallows with birds from western regions (Washington State, USA (n = 8) and Saskatchewan, Canada (n = 5)) traveling shorter distances to wintering areas ranging from Oregon to northern Colombia than eastern populations (Ontario (n = 3) and New Brunswick (n = 10), Canada) which wintered in South America south of the Amazon basin. A single swallow from a stable population in Alabama shared a similar migration route to eastern barn swallows but wintered farther north in northeast Brazil indicating a potential leap frog pattern migratory among eastern birds. Six of 9 (67%) birds from the two eastern populations and Alabama underwent a loop migration west of fall migration routes including around the Gulf of Mexico travelling a mean of 2,224 km and 722 km longer on spring migration, respectively. Longer migration distances, including the requirement to cross the Caribbean Sea and Gulf of Mexico and subsequent shorter sedentary wintering periods, may exacerbate declines for populations breeding in northeastern North America.     

Contrasting Patterns of Genetic Differentiation among Blackcaps (Sylvia atricapilla) with Divergent Migratory Orientations in Europe

Migratory divides are thought to facilitate behavioral, ecological, and genetic divergence among populations with different migratory routes. However, it is currently contentious how much genetic divergence is needed to maintain distinct migratory behavior across migratory divides. Here we investigate patterns of neutral genetic differentiation among Blackcap (Sylvia atricapilla) populations with different migratory strategies across Europe. We compare the level of genetic divergence of populations migrating to southwestern (SW) or southeastern (SE) wintering areas with birds wintering in the British Isles following a recently established northwesterly (NW) migration route. The migratory divide between SW and SE wintering areas can be interpreted as a result of a re-colonization process after the last glaciation. Thus we predicted greater levels of genetic differentiation among the SW/SE populations. However, a lack of genetic differentiation was found between SW and SE populations, suggesting that interbreeding likely occurs among Blackcaps with different migratory orientations across a large area; therefore the SW/SE migratory divide can be seen as diffuse, broad band and is, at best, a weak isolating barrier. Conversely, weak, albeit significant genetic differentiation was evident between NW and SW migrants breeding sympatrically in southern Germany, suggesting a stronger isolating mechanism may be acting in this population. Populations located within/near the SW/SE contact zone were the least genetically divergent from NW migrants, confirming NW migrants likely originated from within the contact zone. Significant isolation-by-distance was found among eastern Blackcap populations (i.e. SE migrants), but not among western populations (i.e. NW and SW migrants), revealing different patterns of genetic divergence among Blackcap populations in Europe. We discuss possible explanations for the genetic structure of European Blackcaps and how gene flow influences the persistence of divergent migratory behaviors.     

Fuel reserves affect migratory orientation of thrushes and sparrows both before and after crossing an ecological barrier near their breeding grounds

Fat reserves influence the orientation of migrating songbirds at ecological barriers, such as expansive water crossings. Upon encountering a body of water, fat migrants usually cross the barrier exhibiting 'forward' migration in a seasonally appropriate direction. In contrast, lean birds often exhibit temporary 'reverse' orientation away from the water, possibly to lead them to suitable habitats for refueling. Most examples of reverse orientation are restricted to autumn migration and, in North America, are largely limited to transcontinental migrants prior to crossing the Gulf of Mexico. Little is known about the orientation of lean birds after crossing an ecological barrier or on the way to their breeding grounds. We examined the effect of fat stores on migratory orientation of both long- and short-distance migrants before and after a water crossing near their breeding grounds; Catharus thrushes (Swainson's and gray-cheeked thrushes, C. ustulatus and C. minimus ) and white-throated sparrows Zonotrichia albicollis were tested for orientation at the south shore of Lake Ontario during spring and autumn. During both spring and autumn, fat birds oriented in a seasonally appropriate, forward direction. Lean thrushes showed a tendency for reverse orientation upon encountering water in the spring and axial, shoreline orientation after crossing water in the autumn. Lean sparrows were not consistently oriented in any direction during either season. The responses of lean birds may be attributable to their stopover ecology and seasonally-dependent habitat quality.     

A place to land: spatiotemporal drivers of stopover habitat use by migrating birds

Migrating birds require en route habitats to rest and refuel. Yet, habitat use has never been integrated with passage to understand the factors that determine where and when birds stopover during spring and autumn migration. Here, we introduce the stopover‐to‐passage ratio (SPR), the percentage of passage migrants that stop in an area, and use 8 years of data from 12 weather surveillance radars to estimate over 50% SPR during spring and autumn through the Gulf of Mexico and Atlantic coasts of the south‐eastern US, the most prominent corridor for North America’s migratory birds. During stopovers, birds concentrated close to the coast during spring and inland in forested landscapes during autumn, suggesting seasonal differences in habitat function and highlighting the vital role of stopover habitats in sustaining migratory communities. Beyond advancing understanding of migration ecology, SPR will facilitate conservation through identification of sites that are disproportionally selected for stopover by migrating birds.     

Wind patterns as a potential driver in the evolution and maintenance of a North American migratory suture zone

Suture zones are areas where range contact zones and hybrid zones of multiple taxa are clustered. Migratory divides, contact zones between divergent populations that breed adjacent to one another but use different migratory routes, are a particular case of suture zones. Although multiple hypotheses for both the formation and maintenance of migratory divides have been suggested, quantitative tests are scarce. Here, we tested whether a novel factor, prevailing winds, was sufficient to explain both the evolution and maintenance of the Cordilleran migratory divide using individual‐based models. Empirical observations of eastern birds suggest a circuitous migratory route across Canada before heading south. Western breeders, however, travel south along the Pacific coast to their wintering grounds. We modeled the effect of wind on bird migratory flights by allowing them to float at elevation using spatially explicit modeled wind data. Modeled eastern birds had easterly mean trajectories, whereas western breeders showed significantly more southern trajectories. We also determined that a mean airspeed of 18.5 m s^–1 would be necessary to eliminate this difference in trajectory, a speed that is achieved by waterfowl and shorebirds, but is faster than songbird flight speeds. These results lend support for the potential importance of wind in shaping the phylogeographic history of North American songbirds.     

Migratory patterns and connectivity of two North American grassland bird species

Effective management and conservation of migratory bird populations require knowledge and incorporation of their movement patterns and space use throughout the annual cycle. To investigate the little‐known migratory patterns of two grassland bird species, we deployed 180 light‐level geolocators on Grasshopper Sparrows (Ammodramus savannarum) and 29 Argos‐GPS tags on Eastern Meadowlarks (Sturnella magna) at Konza Prairie, Kansas, USA, and six US Department of Defense (DoD) installations distributed across the species' breeding ranges. We analyzed location data from 34 light‐level geolocators and five Argos‐GPS tags attached for 1 year to Grasshopper Sparrows and Eastern Meadowlarks, respectively. Grasshopper Sparrows were present on the breeding grounds from mid‐April through early October, substantially longer than previously estimated, and migrated on average ~2,500 km over ~30 days. Grasshopper Sparrows exhibited strong migratory connectivity only at a continental scale. The North American Great Lakes region likely serves as a migratory divide for Midwest and East Coast Grasshopper Sparrows; Midwest populations (Kansas, Wisconsin, and North Dakota; n = 13) largely wintered in Texas or Mexico, whereas East Coast populations (Maryland and Massachusetts, n = 20) wintered in the northern Caribbean or Florida. Our data from Eastern Meadowlarks provided evidence for a diversity of stationary and short‐ and long‐distance migration strategies. By providing the most extensive examination of the nonbreeding movement ecology for these two North American grassland bird species to date, we refine information gaps and provide key insight for their management and conservation.     

Hybrid songbirds employ intermediate routes in a migratory divide

Migratory divides are contact zones between populations that use different routes to navigate around unsuitable areas on seasonal migration. Hybrids in divides have been predicted to employ intermediate and potentially inferior routes. We provide the first direct test of this hypothesis, using light‐level geolocators to track birds breeding in a hybrid zone between Swainson's thrushes in western Canada. Compared to parental forms, hybrids exhibited increased variability in their migratory routes, with some using intermediate routes that crossed arid and mountainous regions, and some using the same routes as one parental group on fall migration and the other on spring migration. Hybrids also tended to use geographically intermediate wintering sites. Analysis of genetic variation across the hybrid zone suggests moderately strong selection against hybrids. These results indicate that seasonal migratory behaviour might be a source of selection against hybrids, supporting a possible role for migration in speciation.     

Genomic analysis of a migratory divide reveals candidate genes for migration and implicates selective sweeps in generating islands of differentiation

Differential gene flow, reductions in diversity following linked selection and/or features of the genome can structure patterns of genomic differentiation during the process of speciation. Possible sources of reproductive isolation are well studied between coastal and inland subspecies groups of Swainson's thrushes, with differences in seasonal migratory behaviour likely playing a key role in reducing hybrid fitness. We assembled and annotated a draft reference genome for this species and generated whole‐genome shotgun sequence data for populations adjacent to the hybrid zone between these groups. We documented substantial genomewide heterogeneity in relative estimates of genetic differentiation between the groups. Within population diversity was lower in areas of high relative differentiation, supporting a role for selective sweeps in generating this pattern. Absolute genetic differentiation was reduced in these areas, further suggesting that recurrent selective sweeps in the ancestral population and/or between divergent populations following secondary contact likely occurred. Relative genetic differentiation was also higher near centromeres and on the Z chromosome, suggesting that features of the genome also contribute to genomewide heterogeneity. Genes linked to migratory traits were concentrated in islands of differentiation, supporting previous suggestions that seasonal migration is under divergent selection between Swainson's thrushes. Differences in migratory behaviour likely play a central role in the speciation of many taxa; we developed the infrastructure here to permit future investigations into the role several candidate genes play in reducing gene flow between not only Swainson's thrushes but other species as well.     

Ticks collected from migratory birds,including a new record of Haemaphysalis formosensis,on Jeju Island,Korea

Migratory birds may disperse parasites across ecological barriers, and recent climate change may alter the pattern of ectoparasite dispersal via changed patterns of bird migration. In order to document the parasitization of migratory birds by Ixodidae ticks on Jeju Island in Korea, we examined 934 migratory birds comprising 75 species for ticks from 2010 to 2012. In total, 313 ticks were collected from 74 migratory birds across 17 avian species and identified based on morphological keys. These ticks represented six species: Haemaphysalis flava, H. formosensis, H. longicornis, H. concinna, Ixodes turdus and I. nipponensis. Of particular note was the presence of H. formosensis, a species not previously reported to have been found in Korea, and H. concinna, which had not been previously reported on Jeju Island. The dominant tick species found were H. flava (226 ticks, 72.2 %) and I. turdus (54 ticks, 17.3 %), and ground-dwelling thrushes such as Pale thrushes (Turdus pallidus; 39 birds, 52.7 %) were the most important hosts. Although H. longicornis is the most abundant and prevalent terrestrial tick on Jeju Island, the species accounted for only 3.8 % of the total ticks collected in this study, suggesting that ticks on migratory birds may differ from the local tick fauna and that exotic ticks may be introduced via migratory birds. Therefore, long-term programs for tick and tick-borne disease surveillance are recommended to understand the role of migratory animals in the introduction of exotic species and associated pathogens and in life cycles of ticks at different stages in this region.     

Lack of genetic differentiation between monarch butterflies with divergent migration destinations

Monarch butterflies are best known for their spectacular annual migration from eastern North America to Mexico. Monarchs also occur in the North American states west of the Rocky Mountains, from where they fly shorter distances to the California Coast. Whether eastern and western North American monarchs form one genetic population or are genetically differentiated remains hotly debated, and resolution of this debate is essential to understand monarch migration patterns and to protect this iconic insect species. We studied the genetic structure of North American migratory monarch populations, as well as nonmigratory populations in Hawaii and New Zealand. Our results show that eastern and western migratory monarchs form one admixed population and that monarchs from Hawaii and New Zealand have genetically diverged from North American butterflies. These findings suggest that eastern and western monarch butterflies maintain their divergent migrations despite genetic mixing. The finding that eastern and western monarchs form one genetic population also suggests that the conservation of overwintering sites in Mexico is crucial for the protection of monarchs in both eastern and western North America.     

Migratory divides and their consequences for dispersal, population size and parasite-host interactions

Populations of migratory birds differ in their direction of migration with neighbouring populations often migrating in divergent directions separated by migratory divides. A total of 26% of 103 passerine bird species in Europe had migratory divides that were located disproportionately often along a longitudinal gradient in Central Europe, consistent with the assumption of a Quaternary glacial origin of such divides in the Iberian and Balkan peninsulas followed by recolonization. Given that studies have shown significant genetic differentiation and reduced gene flow across migratory divides, we hypothesized that an absence of migratory divides would result in elevated rates of gene flow and hence a reduced level of local adaptation. In a comparative study, species with migratory divides had larger population sizes and population densities and longer dispersal distances than species without migratory divides. Species with migratory divides tended to be habitat generalists. Bird species with migratory divides had higher richness of blood parasites and higher growth rates of Staphylococcus on their eggs during the incubation period. There was weaker cell-mediated immunity in adults and stronger cell lysis in species with migratory divides. These findings may suggest that migratory divides constitute barriers to dispersal with consequences for ecology and evolution of distributions, population sizes, habitats and parasite-host interactions. They also suggest that migratory divides may play a role in local adaptation in host-parasite interactions.     

Ground-feeding migratory songbirds as cellular slime mold distribution vectors

Cosmopolitan species of cellular slime molds occur continents apart in both tropical and temperate zones of the world though the spore masses are too heavy to be wind borne, and water dispersal is limited to the watercourses. A highly mobile distribution vector was found in ground-feeding migratory song birds. Nine ubiquitous species and 2 ecologicially distinct species of dictyostelid cellular slime molds were isolated from the feces of ground-feeding eastern North American migratory thrushes, finches, sparrows and warblers, both on breeding and winter grounds. Three propagules of slime molds, amoebae, spores and macrocysts survive passage through the avian digesive tract and remain in the gut long enough to be transported by major bird migrations. Habitats with the greates species diversity of both cellular slime molds and ground-feeding passerines concur in both eastern North America and Central America. Birds actively seek their prefered habitats; the cellular slime molds have arrived at these habitats as passengers. Rare slime molds can serve as a marker to the habitats that migratory birds have visited, or birds with known habitats can provide clues as to the distribution of rare species of cellular slime molds.     

Not as the crow flies: a historical explanation for circuitous migration in Swainson's thrush (Catharus ustulatus)

Many migratory songbirds follow circuitous migratory routes instead of taking the shortest path between overwintering and breeding areas. Here, we study the migration patterns in Swainson's thrush (Catharus ustulatus), a neartic-neotropical migrant songbird, using molecular genetic approaches. This species is presently separated into genetically distinct coastal and continental populations that diverged during the Late Pleistocene (as indicated by molecular dating), yet appear to have retained ancestral patterns of migration. Low nucleotide diversity, a star-like haplotype phylogeny and unimodal mismatch distributions all support the hypothesis that both the coastal and the continental populations have undergone recent demographic expansions. Nearctic-neotropical banding and genetic data show nearly complete segregation of migratory routes and of overwintering locations: coastal populations migrate along the Pacific Coast to overwintering sites in Central America and Mexico, whereas continental populations migrate along an eastern route to overwintering sites in Panama and South America. Nearctic-neotropical banding data also show that continental birds north, northwest and east of this migratory divide fly thousands of miles east before turning south. We conclude that circuitous migration in the Swainson's thrush is an artefact of a Late Pleistocene range expansion.     

Migratory divides coincide with reproductive barriers across replicated avian hybrid zones above the Tibetan Plateau

Migratory divides are proposed to be catalysts for speciation across a diversity of taxa. However, it is difficult to test the relative contributions of migratory behaviour vs. other divergent traits to reproductive isolation. Comparing hybrid zones with and without migratory divides offers a rare opportunity to directly examine the contribution of divergent migratory behaviour to reproductive barriers. We show that across replicate sampling transects of two pairs of barn swallow (Hirundo rustica) subspecies, strong reproductive isolation coincided with a migratory divide spanning 20 degrees of latitude. A third subspecies pair exhibited no evidence for a migratory divide and hybridised extensively. Within migratory divides, overwintering habitats were associated with assortative mating, implicating a central contribution of divergent migratory behaviour to reproductive barriers. The remarkable geographic coincidence between migratory divides and genetic breaks supports a long‐standing hypothesis that the Tibetan Plateau is a substantial barrier contributing to the diversity of Siberian avifauna.     

The role of migration for spatial turnover of arctic bird species in a circumpolar perspective

Several different factors may determine where species range limits are located within regions of otherwise continuously available habitat and suitable climate. Within the Arctic tundra biome many bird species are migratory and their breeding distributions are affected by migration routes that are in turn limited by factors such as suitable winter habitat, migratory stopover sites, geographical barriers and historical routes of colonization. We identified longitudinal zones in the circumpolar Arctic of pronounced changes in the avian species composition (high species spatial turnover; ‘species divides’). We tested for the association between migratory status and the geographical location and numbers of such species divides for species with non‐breeding habitats mainly within terrestrial, pelagic and coastal ecosystems. Our results demonstrate that migration is of profound importance for both the number and locations of species divides in the Arctic. Long‐distance migration is associated with a large number of divides among terrestrial and coastal arctic birds but with a reduced number of divides among pelagic birds. We suggest that long‐distance migration permits pelagic but not terrestrial and coastal birds to colonize large winter ranges, which in turn causes expansion of breeding ranges, with more homogenous communities and reduction of species divides as consequences, among the long‐distance migrants of pelagic but not of terrestrial and coastal birds. Furthermore, the divides among long‐distance migrants are situated in two main regions, the Beringia and Greenland zones, while divides among short‐distance migrants are more evenly spaced throughout the circumpolar Arctic. The Beringia and Greenland divides result largely from inter‐continental colonization of new breeding ranges but retainment of original winter quarters in a process of evolution through extension of migration routes, leading to aggregated divides in the meeting zones of major global flyways.     

Connecting breeding and wintering grounds of Neotropical migrant songbirds using stable hydrogen isotopes: a call for an isotopic atlas of migratory connectivity

There is an overdue and urgent need to establish patterns of migratory connectivity linking breeding grounds, stopover sites, and wintering grounds of migratory birds. Such information allows more effective application of conservation efforts by applying focused actions along movement trajectories at the population level. Stable isotope methods, especially those using stable hydrogen isotope abundance in feathers (δ²H_f) combined with Bayesian assignment techniques incorporating prior information such as relative abundance of breeding birds, now provide a fast and reliable means of establishing migratory connectivity, especially for Neotropical migrants that breed in North America and molt prior to fall migration. Here we demonstrate how opportunistic sampling of feathers of 30 species of wintering birds in Cuba, Venezuela, Guatemala, Puerto Rico, and Mexico, regions that have typically been poorly sampled for estimating migratory connectivity, can be assigned to breeding areas in North America through both advanced spatial assignment to probability surfaces and through simpler map lookup approaches. Incorporating relative abundance information from the North American Breeding Bird Survey in our Bayesian assignment models generally resulted in a reduction in potential assignment areas on breeding grounds. However, additional tools to constrain longitude such as DNA markers or other isotopes would be desirable for establishing breeding or molt origins of species with broad longitudinal distributions. The isotope approach could act as a rapid means of establishing basic patterns of migratory connectivity across numerous species and populations. We propose a large‐scale coordinated sampling effort on the wintering grounds to establish an isotopic atlas of migratory connectivity for North American Neotropical migrants and suggest that isotopic variance be considered as a valuable metric to quantify migratory connectivity. This initiative could then act as a strategic template to guide further efforts involving stable isotopes, light‐sensitive geolocators, and other technologies.     

Geolocators reveal migratory connectivity between wintering and breeding areas of Golden‐winged Warblers

The conservation of migratory songbirds is often impeded by a lack of understanding of how populations in breeding and wintering areas are geographically linked (migratory connectivity). In recent years, light‐level geolocators have improved our understanding of migratory connectivity. Such information is valuable for evaluating how conservation efforts align between the breeding and non‐breeding areas of at‐risk species, and help to more effectively prioritize the allocation of conservation funding. Golden‐winged Warblers (Vermivora chrysoptera) are imperiled migratory songbirds, but the extent to which conservation efforts in their breeding and non‐breeding areas coincide with patterns of migratory connectivity are not well known. We used light‐level geolocators to evaluate the extent to which conservation actions targeting Golden‐winged Warblers in Nicaragua and in their breeding range in North America align with patterns of migratory connectivity. We recovered six of 22 geolocators that had been deployed on male Golden‐winged Warblers at the El Jaguar Reserve during the winter of 2015–2016. All six males migrated to breeding areas in the western Great Lakes region that includes eastern Minnesota, northern Wisconsin, southwestern Ontario, and Michigan's Upper Peninsula. All six males also had similar migration routes, with spring stopovers in southern Mexico, Guatemala, and Belize, a trans‐Gulf flight, and a stopover in the region of Louisiana, Arkansas, eastern Oklahoma, and Texas. Our results, in combination with those of previous studies, demonstrate strong migratory connectivity between portions of the breeding and winter distributions of Golden‐winged Warblers currently targeted for conservation. However, additional studies are needed to improve our understanding of the stopover ecology of Golden‐winged Warblers, especially in areas where they remain for extended periods of time. Finally, patterns of migratory connectivity revealed in our study should be used in combination with existing demographic parameters for Golden‐winged Warblers in the western Great Lakes and Nicaragua to help inform full life cycle population models for this imperiled songbird.     

Energy reserves stored by migrating Gray‐cheeked Thrushes Catharus minimus at a spring stopover site in northern Colombia are sufficient for a long‐distance flight to North America

Stopover sites used to accumulate the energy that fuels migration, especially those used prior to crossing ecological barriers, are regarded as critically important for the survival of Nearctic?Neotropical migratory birds. To assess whether South American stopover sites are used to store the energy required to cross the Caribbean Sea and the Gulf of Mexico to North America by a Neotropical migratory landbird, we studied Gray‐cheeked Thrushes in northern Colombia through constant effort mist‐netting during spring migration in 2010 and 2011. We combined stopover duration estimates and models of body mass change based on recaptures to estimate departure body mass and potential flight range from our study site. We recaptured 62 birds, the majority of which gained mass. Models indicated significant differences in rates of mass gain between years and age groups and with arrival date. Estimated total stopover durations varied between 15.4 (2010) and 12.5 days (2011). Predicted departure mass ranged between 41.3 and 44.9 g, and potential flight range was estimated at between 2727 and 4270 km. Gray‐cheeked Thrushes therefore departed our study site with sufficient energy reserves to cross the Caribbean Sea and the Gulf of Mexico (2550 km). As the first demonstration that birds departing from South American stopover sites can reach North America without refuelling, this has important implications for stopover site protection. Strategic conservation measures in the Sierra Nevada de Santa Marta could protect habitats in which up to 40% of the energy required to complete spring migration is stored by a Neotropical migratory land bird.     

Constraints on dispersal and the evolution of the avifauna of the Northern Hemisphere

One could predict that the capacity for travelling great distances might predispose long-distance migratory birds to be good colonists and to exhibit wider geographic distributions than their non-migratory or short-distance migratory relatives or non-volant mammals. This prediction is not supported by the data on avian biogeography. The distributions of species, genera and families of North American and Eurasian birds and mammals are indeed related to migratory status, but long-distance migratory birds exhibit a great deal of biogeographic regionalism. In particular, at all taxonomic levels their distributions tend to be confined to either the Eastern or Western Hemisphere, suggesting that there has been little successful east–west dispersal between North America and Eurasia. Compared to non-migratory birds, short-distance migratory birds and non-volant mammals, long-distance migrants appear to be subject to severe constraints on their physiology, behaviour and ecology, w hich have prevented colonization of distant regions.