Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hybridization and speciation in the Yellow-rumped Warbler complex Brelsford, Alan 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2010_fall_brelsford_alan.pdf [ 2.61MB ]
JSON: 24-1.0071052.json
JSON-LD: 24-1.0071052-ld.json
RDF/XML (Pretty): 24-1.0071052-rdf.xml
RDF/JSON: 24-1.0071052-rdf.json
Turtle: 24-1.0071052-turtle.txt
N-Triples: 24-1.0071052-rdf-ntriples.txt
Original Record: 24-1.0071052-source.json
Full Text

Full Text

HYBRIDIZATION AND SPECIATION IN THE YELLOW-RUMPED WARBLER COMPLEX by ALAN BRELSFORD Hon.B.A. Williams College, 2001  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES  (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2010 © Alan Brelsford, 2010  ABSTRACT Hybrid zones provide insight into the speciation process. They show which characteristics have evolved differently between recently separated populations, and which differences promote reproductive isolation upon secondary contact. In the Yellowrumped Warbler complex, the Myrtle and Audubon's forms hybridize in the Rocky Mountains. I investigated the history of the complex and the current evolutionary dynamics of the hybrid zone. In chapters two and three, I used genetic, plumage, and song analysis to show that the hybrid zone is stable and maintained by postmating barriers. Association between two genetic markers on different chromosomes showed that the hybrid zone is maintained by moderately strong selection against hybrids. Comparison between my sampling and data collected in 1965 showed that the hybrid zone remained stable during that time. Myrtle and Audubon's songs were weakly differentiated, with hybrid songs intermediate between them. Both parental types reacted as strongly to hybrid songs as to songs of their own type, and there was no evidence of assortative mating. In chapter four, I investigated the history of divergence and hybridization among the four Yellow-rumped Warbler subspecies. Previously, conflicting relationships had been hypothesized. Based on appearance, the Audubon's form had been grouped with two southern subspecies from Mexico and Guatemala. Another study of mitochondrial DNA had supported a very recent divergence between Audubon's and Myrtle and more distant relationship with the southern forms. Using nuclear markers, I showed that the Audubon's Warbler is closely related to the Mexican subspecies, but has acquired its mitochondrial DNA and a substantial fraction of its nuclear genome from the Myrtle form. This is the first multilocus genetic evidence for the hybrid origin of a widespread bird taxon. This thesis provides evidence for the diversity of mechanisms that may drive species divergence. Unlike many well-studied examples, neither premating barriers nor song maintain the distinctness of Myrtle and Audubon's Warblers. My work provides further evidence that hybridization can be a source of evolutionary novelty in animals, and highlights the importance of surveying multiple genetic markers in phylogeography. Finally, this study is a point of comparison for several other hybridizing species with similar biogeographic histories. ii  TABLE OF CONTENTS Abstract..................................................................................................................... ii Table of contents...................................................................................................... iii List of tables ............................................................................................................. v List of figures........................................................................................................... vi Acknowledgements ................................................................................................. vii Co-authorship statement......................................................................................... viii CHAPTER 1: General introduction....................................................................... 1 CHAPTER 2: Incipient speciation despite little assortative mating: the Yellowrumped Warbler hybrid zone................................................................................. 7 2.1 Introduction...................................................................................................... 7 2.2 Materials and Methods ..................................................................................... 9 2.2.1 Field sampling ........................................................................................ 9 2.2.2 Molecular methods ............................................................................... 10 2.2.3 Cline width analysis.............................................................................. 11 2.2.4 Linkage disequilibrium analysis............................................................ 12 2.2.5 Tests for assortative mating................................................................... 14 2.3 Results............................................................................................................ 15 2.4 Discussion ...................................................................................................... 17 CHAPTER 3: No reproductive barrier by song in the Yellow-rumped Warbler hybrid zone............................................................................................................ 26 3.1 Introduction.................................................................................................... 26 3.2 Methods ......................................................................................................... 27 3.2.1 Song recording....................................................................................... 27 3.2.2 Song analysis ......................................................................................... 28 3.2.3 Playback trials ....................................................................................... 29 3.2.4 Playback analysis................................................................................... 29 3.3 Results............................................................................................................ 30 3.3.1 Song analysis ......................................................................................... 30 3.3.2 Playback analysis................................................................................... 31 3.4 Discussion ...................................................................................................... 31  iii  CHAPTER 4: Hybrid origin of Audubon’s Warbler........................................... 42 4.1 Introduction.................................................................................................... 42 4.2 Methods ......................................................................................................... 44 4.2.1 Field sampling ....................................................................................... 44 4.2.2 Morphology........................................................................................... 44 4.2.3 Molecular techniques ............................................................................. 44 4.2.4 Analysis................................................................................................. 45 4.3 Results............................................................................................................ 46 4.4 Discussion ...................................................................................................... 47 CHAPTER 5: General conclusions....................................................................... 58 References.............................................................................................................. 65 Appendix 1: Details of loci sequenced for chapter 2 ............................................... 80 Appendix 2: Correlation coefficients among two genetic markers (CHD1Z, Numt-Dco1) and five plumage traits (Throat, Auricular, Spot, Line, Wing) among samples within 20 km of the hybrid zone centre ...................................................... 81 Appendix 3: Linkage disequilibrium at the centre of the hybrid zone increases due to the Wahlund effect as samples farther than 30 km from the hybrid zone centre are included. ........................................................................................................... 81 Appendix 4: AFLP primer combinations used in chapter 4, and number of polymorphic loci obtained from each....................................................................... 82 Appendix 5: Primers used for amplification and sequencing of nuclear loci in chapter 4 ................................................................................................................. 83  iv  LIST OF TABLES Table 2.1: Location, sample size, and marker frequencies for sampling localities used in this study ............................................................................................... 21 Table 2.2: Details of loci sequenced ....................................................................... 22 Table 2.3: Proportion of randomization trials resulting in less-than-observed correlation between genotypes or phenotypes of social mates ............................ 23 Table 3.1: Means and standard deviations of 8 acoustic variables measured from songs of Audubon’s, hybrid, and Myrtle warblers.............................................. 35 Table 3.2: Factor loadings on first two principal components of song variation ...... 36 Table 3.3: Significance of linear regressions of each acoustic variable against individual hybrid index, and correlations between each acoustic variable and both individual and site average hybrid index ................................ 37 Table 3.4: Results of statistical tests of whether birds responded to playbacks of controls, Audubon’s, hybrid, and Myrtle songs at five sites, and whether responses to hybrid and local songs differed ...................................................... 38 Table 4.1: Divergence among the three widespread forms of Yellow-rumped Warbler at 12 nuclear loci, measured by FST and net sequence divergence ......... 51 Table 4.2: AFLP-based differentiation among Yellow-rumped Warbler forms coronata, auduboni north clade, auduboni south clade, nigrifrons, and goldmani. .......................................................................................................... 52  v  LIST OF FIGURES Figure 2.1: Location of five hybrid zone transects and allopatric reference sample sites in British Columbia and Alberta ................................................................ 24 Figure 2.2: Linkage disequilibrium estimated from diagnostic genetic markers and from plumage traits for five transect centres and for a pooled sample from all transect centres ............................................................................................. 25 Figure 3.1: Locations of song sampling, playback trials, and hybrid zone centre and edges, and response to each song type by location....................................... 39 Figure 3.2: Slight differentiation between songs of Myrtle and Audubon’s Warblers and hybrid intermediacy, as shown by principal component analysis of acoustic variables. ......................................................................................... 40 Figure 3.3: Representative spectrograms from six of the 16 Yellow-rumped Warbler recordings used as playback stimuli ..................................................... 41 Figure 4.1: Range map, sampling locations, and groupings based on AFLP, mtDNA, and phenotype for four Yellow-rumped Warbler subspecies................ 53 Figure 4.2: Multidimensional scaling plot of AFLP variation ................................. 54 Figure 4.3: Haplotype networks for 5 loci that differ between coronata and nigrifrons .......................................................................................................... 55 Figure 4.4: Geographic variation in traits and genetic markers that differ between coronata and nigrifrons demonstrate incongruent patterns in the putative hybrid auduboni................................................................................... 56  vi  ACKNOWLEDGMENTS Thanks first to my advisor Darren Irwin for ideas and support during my time at UBC, and for introducing me to the study of hybrid zones, speciation, and molecular genetics. I also thank my supervisory committee, Peter Arcese, Dolph Schluter, Eric Taylor, and Jeannette Whitton for their suggestions and feedback. Trevor Price, Hopi Hoekstra, Nick Barton, and four anonymous reviewers provided invaluable advice on chapter two, as did Loren Rieseberg on chapter four. Discussions with members of the Irwin and Schluter labs and the SOWD/DeltaTea group were instrumental in shaping the ideas I present here. Alana Demko, Allison Patterson, Olga Lansdorp, and Michael Moretti provided valuable assistance in the field, and Kristen Haakons helped extract and analyse data from song recordings. Borja Milá generously shared unpublished data, genetic samples, and photographs from his studies of Yellow-rumped Warblers, which were conducted in Tom Smith’s lab at the UCLA Center for Tropical Research. John Pollinger helped with shipment of samples from UCLA to UBC. Funding was provided by an NSERC Discovery Grant to DEI, an NSF Graduate Research Fellowship to AB, and an Alberta Conservation Association Grant in Biodiversity to AB. The molecular analysis was conducted in the UBC Laboratory for Molecular Biogeography, funded by grants to DEI from the Canadian Foundation for Innovation and the British Columbia Knowledge Development Fund. I gratefully acknowledge the Canadian Wildlife Service, Parks Canada, Environment Canada, Alberta Community Development, and the Animal Care Committees of University of British Columbia and University of Calgary for permission to conduct fieldwork. Finally, thanks to Jessica Purcell for love, support, scientific insight, and editorial comments at all stages of my degree.  vii  CO-AUTHORSHIP STATEMENT Chapter 2: I planned the study, collected and analysed data, and wrote the manuscript with the help of Darren Irwin. Chapter 3: I planned the study, collected and analysed data, and wrote the manuscript with the help of Darren Irwin. Michael Moretti helped carry out playback experiments, and Kristen Haakons helped measure acoustic variables from song recordings. Chapter 4: I planned the study, collected and analysed data, and wrote the manuscript with the help of Darren Irwin. Borja Milá collected most of the blood and DNA samples, provided mitochondrial DNA sequences and photographs of birds from the U.S., Mexico, and Guatemala, and contributed ideas during the writing of this chapter.  viii  CHAPTER 1: GENERAL INTRODUCTION Hybrid zones form when divergent populations come into contact and interbreed. They have long been recognized as “natural laboratories” for studying speciation (Barton and Hewitt 1989), since they allow the study of which traits have diverged between populations and which of those divergent traits contribute to reproductive isolation. Some divergent organisms avoid mating with each other, showing that they have diverged in traits (e.g. male sexual characteristics or female preferences) that cause premating isolation. Pied and Collared Flycatchers (Ficedula hypoleuca and F. albicollis), for example, hybridize much less often than would be expected by chance (Veen et al. 2001). Other hybridizing species interbreed extensively but produce offspring with lower fitness. For example, hybrid fire-bellied toads have reduced viability in captivity relative to their parental species (Bombina bombina and B. variegata), showing intrinsic postzygotic isolation (Szymura and Barton 1986). Hybrid irises have ecological requirements intermediate between those of their parental species (Iris fulva and I. hexagona), an example of ecological post-mating isolation (Anderson 1949). The acoustic mating signals of hybrids between the grasshoppers Chorthippus brunneus and C. jacobsi are avoided by females, an example of behavioural post-zygotic isolation (Bridle et al. 2006). By investigating hybrid zones, we can determine the strength of these effects and the relative importance of each in speciation. Several classes of models have been proposed to describe hybrid zones. Perhaps the simplest hybrid zone model is that of a neutral, transient zone. If two differentiated populations come into contact and interbreed without any selective consequences, they will blend together over time at a rate proportional to their dispersal distance (Endler 1977), leading to an expanding hybrid zone and the eventual fusion of the two hybridizing populations. This model is often treated as a null hypothesis in studies of hybrid zones (e.g. Gay et al. 2008, Ruegg 2008, Carling and Brumfield 2008, Cheviron and Brumfield 2009), but I am not aware of any well-supported empirical examples of this phenomenon. Two reasons may account for the rarity of observed neutral-diffusion hybrid zones: first, they are by definition transient, so we can only expect to observe them in cases of very recent secondary contact or in taxa with very restricted dispersal. 1  Second, if the extent of divergent evolution between two populations has been so small that no selective differences exist between them, they are likely to be so similar in all observable traits that they would not be recognized as forming a hybrid zone. More often, some form of selection limits the expansion of hybrid zones. One possibility is that each parental type is best adapted to its respective habitat, restricting the hybrid zone to an ecotone where the two parental ranges abut. For example, highelevation and low-elevation morphs of the big sagebrush (Artemisia tridentata) each have high fitness in their own habitat and low fitness in the other habitat, as shown by a reciprocal transplant experiment (Wang et al. 1997, Miglia et al. 2005). A special case of the ecotone hybrid zone is the bounded hybrid superiority model (Anderson 1949, Moore 1977), in which hybrids have higher fitness than parental types, but only in a restricted area of ecologically intermediate habitat. Another possibility is that the hybrid zone is maintained by intrinsically (i.e. not environment-dependent) lower hybrid fitness. If hybrids have reduced viability or fertility due to genetic incompatibilities, or if they are less able to attract mates, the hybrid zone will be stabilized in width. This type of hybrid zone, (termed “tension zone,” Key 1968) is theoretically free to drift in position, but will become trapped in place by any area of reduced dispersal or population density (Barton 1979). Many examples of tension zones have been documented, including buntings (Carling and Brumfield 2008), mice (Macholán et al. 2007), and salamanders (Alexandrino et al. 2005). Of course, ecological differences and intrinsically low hybrid fitness are not mutually exclusive. The Bombina toad hybrid zone is one of the best-studied examples of a tension zone, and hybrids have a lower hatch rate and high prevalence of developmental abnormalities (Szymura and Barton 1986). Nevertheless, their hybrid zone coincides with differences in the frequency of small temporary and large permanent pools, which are associated with B. variegata and B. bombina respectively (Vines et al. 2003). Similarly, hybrids between the monkeyflowers Mimulus lewisii and M. cardinalis have low viability and fertility (Ramsey et al. 2003), and each parental species also has low fitness in the other’s range (Angert and Schemske 2005). Whether a hybrid zone is maintained by either or both of these factors, its width will represent a balance between homogenizing forces (dispersal and interbreeding) and the disruptive force of divergent selection. This 2  theory has been worked out most extensively in the case of tension zones: Barton (1982) shows that the associations between alleles at different loci (linkage disequilibrium) will be proportional to selection against hybrids. Knowing the strength of selection and the hybrid zone width also allows the calculation of an approximate dispersal distance. Botanists have long argued that hybrid zones are important to evolutionary biologists not only as illustrations of incomplete speciation, but as sources of evolutionary novelty in their own right (Stebbins 1959, Arnold 1997). In the past, this perspective was thought to apply only to plants (Mayr 1963). Polyploidy is common in plants but rare in animals (Otto and Whitton 2000), and can provide an instantaneous route to reproductive isolation (Coyne and Orr 2004). Comparative genomic analyses of plants suggest that speciation by polyploidy has occurred frequently (Wood et al. 2009), often through allopolyploidy (formation of polyploid hybrids from diploid parents of two different species). Hybrid speciation can also occur without polyploidy (“homoploid hybrid speciation”). Some of the earliest and best-studied examples of homoploid hybrid speciation are three sunflower species that arose independently from hybridization between Helianthus annuus and H. petiolaris (reviewed in Rieseberg 2006). The hybrid species occupy different habitats from their parental species, and their genomes are composed of a mosaic of annuus-like regions and petiolaris-like regions. In the last few years, several cases of hybrid speciation have been identified in animals (reviewed in Mavárez and Linares 2008): The cyprinid fish Gila seminuda, the butterfly Heliconius heurippa, and unnamed lineages of Cottus sculpins and Lycaides butterflies have all been shown to have hybrid origins based on molecular markers (DeMarais et al. 1992, Mavárez et al. 2006, Nolte et al. 2005, Gompert et al. 2006). Two birds, the Adelaide Rosella and the Italian Sparrow, have been proposed as possible hybrid species (Price 2008). Genetic evidence in the rosella is consistent with a hybrid origin but does not rule out other explanations (Joseph et al. 2008), while little genetic data is available on the Italian Sparrow. Hybridization can also contribute to evolutionary novelty through adaptive introgression, the spread of an allele from one species to another under the influence of positive selection. Adaptive introgression differs from hybrid speciation in that no new species results from hybridization, but it does form novel combinations of traits from 3  different species (Jiggins et al. 2008). An allele that causes dark coat colour in wolves of the Canadian boreal forest has been shown to originate from dogs (Anderson et al. 2009). This allele is much rarer in wolves in adjacent tundra habitats, suggesting that it may provide a selective advantage in certain environments, although the adaptive significance of this introgression has not been proven (Anderson et al. 2009). Hybridization has provided genetic variability in Darwin’s finches Geospiza spp. on Daphne Major in the Galapagos, allowing these small populations to respond to strong fluctuating selection associated with El Niño and La Niña climatic events (Grant and Grant 2010). The fruit flies Drosophila yakuba, D. saotomei, and D. teissieri share the same mitochondrial haplotypes despite considerable nuclear divergence and near-complete reproductive isolation (Bachtrog et al. 2006), and the authors of that study argue that positive selection must have been involved to overcome the reproductive barriers between the species. Mitochondrial DNA has been a widely used genetic marker for phylogeographic studies in animals for several reasons. In general, it does not recombine, its mutation rate is much higher than nuclear sequences, and it is inherited only through the maternal line giving it a lower effective population size than nuclear loci. Because of these characteristics, it is expected that recently diverged populations will show greater structure in mitochondrial than nuclear markers (Zink and Barrowclough 2008). However, the mitochondrial gene tree may differ from the underlying species tree for many reasons, including the stochastic nature of the coalescent process, sex differences in dispersal, or selection on mitochondrial genetic variation (Irwin 2002, Ballard and Whitlock 2004, Edwards et al. 2005, Bazin et al. 2006). As DNA sequencing technology has advanced over the past decade, inclusion of multiple nuclear markers in phylogeographic studies has become much more common. Hybridization requires divergent populations, and allopatry is one of the most straightforward ways that differentiation can occur. The Pleistocene epoch was characterized by climatic oscillations. Northern landmasses were repeatedly covered by ice sheets, altering sea levels and rainfall patterns and fragmenting habitats worldwide (Hewitt 2004). The role of Pleistocene glaciations in generating biodiversity has been extensively debated (Klicka and Zink 1997, Avise and Walker 1998, Klicka and Zink 1999, Weir and Schluter 2004, Lovette 2005). Before the advent of molecular 4  phylogenetics, some authors (e.g. Rand 1948, Hubbard 1969) argued that many east-west pairs of bird species originated during the most recent glacial period. An influential study by Klicka and Zink (1997) and subsequent refinements (Avise and Walker 1998, Klicka and Zink 1999) showed that most of these species pairs originated before the late Pleistocene. However, many sister species of boreal forest birds did diverge in the early Pleistocene (Weir and Schluter 2004), suggesting that the onset of glacial cycles may have been responsible for increased speciation rates in northern areas. Many species or populations that persisted in refugia during the Pleistocene have come into contact in the current interglacial period. These contact zones are clustered in areas known as “suture zones” (Remington 1968, Swenson and Howard 2005). The eastern edge of the Rocky Mountains in western North America forms one of the most extensive of these suture zones (Swenson and Howard 2005). In the Canadian Rockies, this suture zone provides a natural experiment of sorts: Several pairs of bird taxa with similar ecological requirements began to diverge from each other in the early Pleistocene, have likely experienced several episodes of allopatry and parapatry during glacial cycles, and now have ranges that overlap slightly. Studying the traits that have diverged between members of these taxon pairs and how they affect reproductive isolation may allow us to find general patterns in the factors promoting speciation in birds. One prominent pair of bird taxa that meet in the Canadian Rockies is found in the Yellow-rumped Warbler (Dendroica coronata) complex. This group has been the subject of considerable taxonomic debate over the past century, mainly as a result of hybridization among geographic variants. Four visually distinct forms were originally described as separate species: the Myrtle Warbler (D. coronata, Linnaeus 1766) is found in Alaska, most of Canada, and the northeast U.S.; the Audubon’s Warbler (D. auduboni, Townsend 1837) in southern British Columbia and the western U.S.; the Black-fronted Warbler (D. nigrifrons, Brewster 1889) in western Mexico; and the Goldman’s Warbler (D. goldmani, Nelson 1897) in the highlands of Guatemala and southern Mexico. These forms were considered subspecies after the discovery of a hybrid zone between Myrtle and Audubon’s Warblers in the Canadian Rockies (AOU 1973). Two previous studies of the hybrid zone have reached conflicting conclusions. Hubbard (1969) suggested that the  5  zone is maintained by selection, while Barrowclough (1980) argued that its width is compatible with selectively neutral expansion and eventual fusion of the two forms. More recently, Milá et al. (2007) found considerable differentiation in mitochondrial DNA between the two widespread, northern, seasonally migratory subspecies (Myrtle and Audubon’s) and the two southern non-migratory subspecies with smaller ranges (Black-fronted and Goldman’s), with little differentiation between Myrtle and Audubon’s. This pattern was surprising, since the male breeding-season plumage of the Myrtle form is quite distinct from the other three. The migratory and non-migratory forms were estimated to have diverged approximately 1.5 million years ago, and the Myrtle and Audubon’s divergence was estimated to have occurred in the last 20,000 years. Milá et al. hypothesized that the distinct Myrtle plumage has evolved rapidly as a result of strong sexual selection, but called for further work using nuclear genetic markers to confirm the relationships estimated from mitochondrial DNA. In this thesis, I use the Yellow-rumped Warbler complex as a case study of the factors that contribute to speciation. I aim to resolve conflicting interpretations from previous work on the system, and provide a point of comparison for other boreal forest taxa that meet in the Rocky Mountain suture zone. In chapter two, I compare new and historical plumage clines to test whether the hybrid zone is stable. I use linkage disequilibrium between two genetic markers on different chromosomes to estimate the strength of selection that stabilizes the zone, and use observed breeding pairs to test for evidence of assortative mating. In chapter three, I investigate whether song is involved in maintaining the hybrid zone. In chapter four, I extend the genetic analysis to all four Yellow-rumped Warbler subspecies, using nuclear sequence and amplified fragment length polymorphism (AFLP, Vos et al. 1995) markers to estimate relative divergence times and examine whether hybridization has played a creative role in the evolution of this group.  6  CHAPTER 2: INCIPIENT SPECIATION DESPITE LITTLE ASSORTATIVE MATING: THE YELLOW-RUMPED WARBLER HYBRID ZONE1 2.1 INTRODUCTION Hybrid zones form when genetically divergent populations come into contact and interbreed (Harrison 1993, Barton and Hewitt 1989). They represent an intermediate stage in the process of speciation, when genetic differences have evolved but are insufficient to completely prevent interbreeding (Hewitt 1988, Barton and Hewitt 1989, Harrison 1993, Jiggins and Mallet 2000). Hybrid zones may be ephemeral, resulting from the recent meeting of two divergent forms that are now blending together (Endler 1977), or may be maintained indefinitely by a balance between selection and dispersal, resulting in a stable cline (Key 1968, Bazykin 1969, Slatkin 1973, May et al. 1975, Endler 1977, Barton 1979, Mallet 1986). By studying variation in a sample of traits across a hybrid zone, we can infer (1) whether two forms are likely to have distinct evolutionary trajectories due to partial reproductive isolation, and (2) the factors that contribute to reproductive isolation. This approach can help resolve ongoing debates regarding the relative importance of premating and postmating reproductive isolation in the early stages of speciation (Coyne and Orr 2004; Price 2008), and the role of hybridization in preventing speciation (e.g. Mayr 1942) or enhancing it through the process of reinforcement (Dobzhansky 1940). A measure of dispersal is essential to distinguishing between ephemeral and stable hybrid zones. In the case of neutral expansion, dispersal can be used to estimate the time since secondary contact necessary to explain the width of the cline (Endler 1977). If that estimated time is unreasonably small, then a model of neutral expansion can be rejected. In the alternative model of a tension zone (i.e. a hybrid zone maintained by a balance between dispersal and selection against hybrids, Key 1968), cline width and dispersal can be used to estimate the strength of selection maintaining the hybrid zone 1  A version of this chapter has been published. Brelsford, A., and D. E. Irwin. 2009. Incipient speciation despite little assortative mating: the Yellow-rumped Warbler hybrid zone. Evolution 63:3050–3060. 7  (Barton 1979). A given cline width can be explained by high dispersal and strong selection, or low dispersal and weak selection, but these two scenarios differ in the distributions and associations of traits in the centre of the zone. In the former case, many individuals in the centre of the zone resemble the pure forms, leading to high trait variance and high linkage disequilibrium between distinct traits. In the latter case, the centre of the zone consists of individuals that are the product of many generations of hybridization, leading to lower trait variance and lower linkage disequilibrium. Thus the measurement of linkage disequilibrium allows a simultaneous estimation of both dispersal and selection, given a known hybrid zone width (Barton 1982, Szymura and Barton 1986, Barton and Gale 1993). Dispersal distance is notoriously difficult to measure directly in many organisms (Sotka and Palumbi 2006, Mallet et al. 1990), making this approach of using tension zones to estimate dispersal quite useful. Here, we apply both the neutral expansion and tension zone models to a well-known hybrid zone within the North American avifauna, with the goal of understanding the evolutionary dynamics occurring in the zone. A hybrid zone that has been much debated in the literature (Hubbard 1969, Barrowclough 1980, Zink and McKitrick 1995, Rohwer and Wood 1998, Johnson et al. 1999, Milá et al. 2007) is found within the Yellow-rumped Warbler (Dendroica coronata) complex. Two of the four well-marked subspecies differ noticeably in plumage patterns, the Audubon’s Warbler (D. c. auduboni) in the west and the Myrtle Warbler (D. c. coronata) in the east, such that they were originally considered separate species (Hubbard 1969). Discovery of the hybrid zone where their ranges meet in western Canada (Alexander 1945, Hubbard 1969) led taxonomists to lump the two taxa into a single species (AOU 1973). Recently, Milá et al. (2007) showed that ancestral common mitochondrial DNA haplotypes are shared between the forms, and used a coalescent model to infer a divergence date between the forms of approximately 16,000 years ago (but see Arbogast et al. 2002 for caveats on estimation of divergence times). Milá et al. (2007) suggested that male breeding-season plumage evolved rapidly in the Myrtle Warbler, likely driven by strong sexual selection, and that partial reproductive isolation may exist between the two forms.  8  Two previous studies on the hybrid zone reached conflicting conclusions about the degree of reproductive isolation between the subspecies. Hubbard (1969) mapped plumage and morphometric variation along two transects across the hybrid zone, and found that nearly all individuals at the centre of the zone showed some evidence of admixture in plumage pattern. He speculated that the hybrid zone was narrow enough that some selection against hybrids was necessary to maintain it. In contrast, Barrowclough’s (1980) analysis of Hubbard’s data, pooled with additional samples he collected, concluded that the cline could be explained by neutral expansion since secondary contact 5 to 10 thousand years ago. However, Barrowclough’s analysis was based on an assumed dispersal distance (1 kilometre per generation) that subsequent research (e.g. Moore and Dolbeer 1989, Paradis et al. 1998, Ruegg 2008) suggests may have been too low by as much as two orders of magnitude. Given the contradictory conclusions of Hubbard (1969) and Barrowclough (1980) regarding stability of the hybrid zone, here we test whether partial reproductive isolation has evolved between Myrtle and Audubon’s Warblers. We show that the observed cline width and realistic ranges of dispersal and time since contact are incompatible with a neutral model of cline expansion. We test for premating isolation by comparing observed and expected differences (both genetic and phenotypic) between social mates. Since we are able to reject strong or moderate levels of premating isolation, we use cline width and linkage disequilibrium between physically unlinked diagnostic markers to estimate dispersal and the strength of selection maintaining the hybrid zone. 2.2 MATERIALS AND METHODS 2.2.1 Field sampling We sampled breeding warblers along five transects across the hybrid zone in British Columbia and Alberta (Figure 2.1). Birds were captured using song playback and mist nets between late April and early July of 2005-2007, for a sample size of 661 individuals at 43 locations (Table 2.1). We used non-lethal sampling methods to enable observation of behaviour of marked birds subsequent to capture, and because large series of specimens from this hybrid zone already exist (Hubbard 1969, Barrowclough 1980). To ensure that we sampled locally breeding warblers, individuals that were not observed 9  singing (males only) or defending a territory (both sexes) were excluded from the analysis. For each captured bird, we scored five plumage colour traits that differ between coronata and auduboni, following Hubbard’s (1969) hybrid index. The scored traits were throat colour (yellow in auduboni, white in coronata), auricular colour (grey in auduboni, black in coronata), white supraloral spot (absent in auduboni, present in coronata), white postocular line (absent in auduboni, present in coronata), and wing pattern (single broad white patch in auduboni, two distinct white bars in coronata). A sixth trait used by Hubbard (1969), tail pattern, was excluded from analysis due to concerns over its repeatability. Auricular colour was excluded from analysis in females, because in both subspecies female auriculars were brown rather than black or grey. Digital photographs were taken to verify consistency of plumage scores across observers, and a blood sample was obtained by brachial venipuncture and stored in Queens lysis buffer (Seutin et al. 1991). We attempted to identify the social mate of each female warbler captured by observing mate-guarding behaviour; if a female was consistently followed or guarded by a single male, we designated that male her social mate. 2.2.2 Molecular methods We extracted DNA from blood samples using standard phenol-chloroform methods. We then screened loci from both autosomes and sex chromosome Z (in birds, males are ZZ and females ZW) for diagnostic variation between coronata and auduboni by sequencing each locus in birds sampled far from the hybrid zone. Initial screening used two birds from each subspecies; diagnostic status (i.e. frequency difference >0.9 between allopatric populations) of potentially informative markers was confirmed by sequencing or genotyping additional individuals (see below). We tested primers for 18 introns chosen from two studies on Ficedula flycatchers (Borge et al. 2005, Backström et al. 2006), of which 9 primer pairs reliably produced single PCR products. Additionally, we screened an intron of CHD1Z that has been used for molecular sexing of birds (Fridolfsson and Ellegren 1999), and a nuclear sequence of mitochondrial origin (“numt”) that we name Numt-Dco1, which was initially sequenced inadvertently in an effort to sequence a fragment of the mitochondrial control region. Numt-Dco1 almost certainly originated from a mitochondrial sequence based on its 83-90% sequence identity to other warbler control region sequences, and is clearly located on an autosome, since many 10  individuals of both sexes carry two different copies of this sequence, both of which are substantially (15-17 %) divergent from the Yellow-rumped Warbler mitochondrial version. In total, we screened six autosomal and five Z-linked loci; details of sequenced loci are found in Appendix 1. A typical PCR reaction for sequencing included 1x PCR buffer (Invitrogen), 1.5 mM MgCl2 (Invitrogen), 0.2 mM dNTP mix (New England Biolabs), 0.5 µM forward and reverse primer, 1.2 units Taq DNA polymerase (New England Biolabs), and 75 ng genomic template DNA, in a total volume of 30 µl. The thermal cycling profile was 3 minutes at 94ºC followed by 35 cycles of 30 seconds at 94ºC, 30 seconds at an annealing temperature that varied by locus (Appendix 1), and 45 seconds at 72ºC, ending with 5 minutes at 72ºC. Amplified PCR products were sequenced along both strands by Macrogen, Inc., using an ABI 3730 XL sequencer. Forward and reverse sequences were aligned in BioEdit (Hall 1999) and checked by eye against electropherograms. Haplotypes were inferred using PHASE (Stephens and Donnelly 2003), and within-taxon polymorphism, FST , and net sequence divergence were calculated for each locus using DnaSP (Rozas et al. 2003). We designed RFLP assays for diagnostic SNPs using NEBcutter (Vincze et al. 2003), and used these assays to genotype all sampled warblers. PCR conditions for RFLP genotyping were identical to those for sequencing, but reactions were run at 10 µl volumes. Restriction digests were carried out on 2 µl of PCR product, using 2 units of restriction enzyme (HindIII for CHD1Z, MnlI for Numt-Dco1) and 1x concentration of the appropriate buffer (New England Biolabs) in a total volume of 6 µl. Products were digested for 2 hours at 37ºC, and restriction enzymes inactivated by 15 minutes at 70ºC. Digested DNA was visualized by electrophoresis on 2% agarose gel stained with SYBRSafe (Invitrogen). 2.2.3 Cline width analysis To enable comparisons between transects, we converted our two-dimensional location data to a set of five linear transects by calculating the distance from the midpoint of the hybrid zone (Figure 2.1) for each sampled individual. The location of the cline midpoint was determined as follows: First, we calculated the average plumage-based hybrid index for each sampling site, and for each transect we found the two sites that bracket the midpoint value of 0.5. We then plotted a line between the centres of these two 11  sampling sites, and interpolated the location along this line at which the expected phenotypic hybrid index would equal 0.5. For example, if the two bracketing sample sites had average hybrid index values of 0.4 and 0.6, the centre of the hybrid zone would be at the midpoint of the line connecting the two sample sites. We then drew a curve (representing the hybrid zone centre; Figure 2.1) connecting the five transect centres; over most of the distance, this curve paralleled the local orientation of the Rocky Mountains, since the hybrid zone is broadly coincident with the Rockies over long distances. Finally, we measured minimum straight-line distances from each sampled bird to this curve, giving a single spatial dimension that is comparable among multiple transects. All spatial measurements and plotting were conducted using Google Earth. We estimated cline width using the program C-fit (Gay et al. 2008). C-fit uses maximum likelihood to fit sigmoid curves with exponential tails of introgression on each side of the cline, using methods described by Szymura and Barton (1986). Concordance of clines among markers, among transects, and between plumage data from 1965 (taken from Hubbard 1969) and 2005-2007 (this study) was evaluated using likelihood ratio tests, following the approach of Fel-Clair et al. (1996). Cline width was calculated as the inverse of the maximum slope (Endler 1977). Samples that were taken more than 100 km from the centre of the zone were used in the cline analysis for every transect, whereas those taken within 100 km of the centre of the zone were used only in a single transect. 2.2.4 Linkage disequilibrium analysis We estimated composite digenic linkage disequilibrium (LD) between two diagnostic markers (Weir 1996, ch. 3) among birds sampled within 20 km of the centre of each of the five transects. Lumping samples collected a range of distances from the cline centre might spuriously inflate estimates of linkage disequilibrium by the Wahlund effect (Sinnock 1975). To account for this, we re-calculated LD for samples ranging from within 1 km to 75 km of the hybrid zone centre. Confidence intervals were estimated by bootstrapping with 10000 replicate samples using Matlab (The MathWorks, Inc.). When the origin of dispersing organisms differs in allele frequencies from the destination, dispersers carry gametes that retain their original associations between alleles at different loci, causing linkage disequilibrium (Barton 1982). In a tension zone of a given width, Barton (1982) showed that linkage disequilibrium is related to the square of 12  dispersal: D =  "2 , where D is linkage disequilibrium, σ is the standard deviation of rw 2  parent-offspring distance (hereafter, “dispersal”), w is cline width, and r is recombination rate between loci. Barton and Gale (1993) modified the equation for samples taken after ! " 2 (1+ r) dispersal and before breeding D = . We used this second equation (for samples rw 2 taken after dispersal) along with our measurements of LD and cline width to estimate dispersal σ (Barton and Gale 1993). This model assumes that mating is random, the ! hybrid zone is maintained by selection against intermediate genotypes, and that there are no epistatic interactions among the loci used to measure linkage disequilibrium (Barton 1982, Szymura and Barton 1986, Barton and Gale 1993). The model was also developed using an assumption of weak selection, but subsequent simulations incorporating stronger selection gave results consistent with the model (Barton and Gale 1993, Kruuk et al. 1999). The width of a tension zone at equilibrium is given by w =  4" where s is 2s  selection against a heterozygote (Bazykin 1969); this assumes that the cline is maintained by heterozygote disadvantage at a single locus, but authors have argued that the ! relationship holds approximately for more realistic multilocus clines and a wide variety of types of selection (Moore and Price 1993, Kruuk et al. 1999, Price 2008), and the equation is routinely used in analyses of hybrid zones (e.g. Alexandrino et al. 2005, Raufaste et al. 2005, Sequeira et al. 2005, Macholán et al. 2007). Using this equation, we estimated the strength of selection (s) necessary to maintain the cline. We used a second method of estimating LD from variance in a hybrid index (Barton and Gale 1993) using plumage. Treating five plumage colour traits as independent “loci”, we obtained a phenotypic estimate of LD to complement our estimate based on genetic markers. (In reality, plumage traits are unlikely to be genetically independent; see Results.) Correlation coefficients among five plumage traits and two genetic markers in populations within 20 km of the hybrid zone centre were calculated using Matlab.  13  2.2.5 Tests for assortative mating To test whether Yellow-rumped Warbler subspecies mate assortatively, we used randomization tests to determine whether mated pairs (n = 77) were more genetically or phenotypically similar than would be expected by chance. First, we calculated a genetic hybrid index based on the two diagnostic genetic markers. This index was defined as the number of Audubon’s alleles carried by an individual divided by the total number of alleles (two CHD1Z alleles plus two Numt-Dco1 alleles for males; one CHD1Z allele plus two Numt-Dco1 alleles for females), with possible values from zero to one in increments of 0.25 for males and 0.33 for females. We then calculated the correlation between genetic indices of females and their mates. To determine the expected level of association, we randomly assigned each female a male from the same sampling locality (generally within 10 km) and calculated the overall genetic correlation between paired individuals. This randomization was repeated 100,000 times, producing a distribution of within-pair correlations expected under random mating. Randomization tests were also used to assess the power of our mate-choice sample to detect varying strengths of assortative mating. In these tests, we modelled mate choice as a process in which a female warbler sequentially evaluates a number n of potential mates sampled from the males caught at the same study site, choosing to pair with the first male she finds acceptable. If no acceptable males are encountered, the females pairs with the nth male. Acceptability is defined by the parameter A, the maximum difference between a female’s genetic hybrid index and that of an acceptable mate. For example, a pure Myrtle female (hybrid index 0) would pair with a pure Audubon’s male (hybrid index 1) only if A = 1, or if the male is the nth potential mate she has evaluated. We simulated pairings according to this model for the 77 females with known mates, with 100,000 replicates for each parameter combination. We then compared the simulated correlation to the observed value to determine the parameter combinations compatible with the observed pattern of mate choice. This analysis was repeated using Hubbard’s (1969) phenotypic hybrid index.  14  2.3 RESULTS We identified two diagnostic markers among the eleven screened. One diagnostic marker (CHD1Z) is Z-linked; the other (Numt-Dco1) is autosomal; both exhibited very nearly fixed differences between allopatric Myrtle and Audubon’s Warblers (Table 2.1, localities Hope, Whistler, Cold Lk.) as well as very low levels of polymorphism within subspecies (Table 2.2). Sequences have been deposited in GenBank (accession numbers GQ457569–GQ457722). Patterns of variation across the hybrid zone closely conformed to the expected sigmoidal pattern predicted by tension zone models (Barton and Gale 1993). Cline widths at diagnostic markers and plumage traits did not differ significantly among the five sampled transects (Figure 2.1, Pairwise comparisons using likelihood ratio test described by Fel-Clair et al. 1996: χ2 ranged from 0.00 to 1.25, d.f. = 1, P ranged from 0.26 to 1.), and overall marker cline widths (i.e. when the five transects were combined) were similar to each other and to the cline in plumage pattern (Figure 2.1, χ2 ranged from 0.01 to 0.16, d.f. = 1, P ranged from 0.69 to 0.92). Cline location (i.e. the location of the centre of the clines) did not differ between CHD1Z and plumage (χ2 = 0.18, d.f. = 1, P = 0.67), but did differ between Numt-Dco1 and those traits (CHD1Z: χ2 = 14.56, d.f. = 1, P = 0.0001; plumage: χ2 = 10.27, d.f. = 1, P = 0.0014), with the Numt-Dco1 cline displaced toward the Audubon’s side by 14 km, a small distance compared to the width of the clines. Clines in plumage hybrid index from samples taken in 1965 (Hubbard 1969) and 20052007 did not differ in width (χ2 = 0.53, d.f. = 1, P = 0.47) or position (χ2 = 0.02, d.f. = 1, P = 0.66). Like Hubbard (1969), we found that the vast majority (191 of 200) of individuals near the hybrid zone centre showed some evidence of admixture. The cline width estimated from all transects using plumage and both genetic markers was 132 km. Correlation between the hybrid indices of female warblers and their mates was 0.69 (phenotypic) and 0.54 (genetic). Observed correlations were slightly higher than the median expected under local random mating for phenotypic (0.60) and genetic (0.52) hybrid indices. However, the difference was not significant; phenotypic correlation between randomly assigned pairs was higher than the observed value in 9.5 percent of trials, while genetic correlation was higher than the observed value in 35 percent. Power analysis indicated that the observed pairing pattern was consistent with random or very 15  weakly assortative mating (females may avoid mates that differ from their genotype by at least 0.75, or from their phenotype by at least 0.67), but moderate or strong assortative mating would have resulted in a significantly higher female-male correlation (Table 2.3). Without access to family groups, we were only able to assess social mates; extra-pair parentage is common in passerines, including some warblers (Griffith et al. 2002). Because of this constraint, we assumed that patterns of mate choice for extra-pair copulations do not differ substantially from patterns of choice of social mates. Linkage disequilibrium between CHD1Z and Numt-Dco1 was significant at two of five transects (Figure 2.2), with estimates ranging from -.007 (transect A) to 0.120 (transect C). Pooling the samples from all five transect centres gave an LD estimate of 0.067 (95% C.I. 0.029 to 0.104). This level of LD is substantial; the theoretical maximum for this measure of disequilibrium is 0.25, which would be expected in the absence of hybridization. Assuming for the moment that there is no strong epistasis between the two marker loci, we can use the model of Barton and Gale (1993) to calculate a dispersal distance of 20 (13 to 25) km per generation, and a strength of selection of 0.18 (0.08 to 0.28). We note that this estimate of selection refers to the “effective selection pressure at a single locus that would be required in order to maintain a cline of the observed width” (Szymura and Barton 1986). In reality, selection is likely to act on multiple loci; hence, this estimate of selection represents the sum of selection acting on each locus multiplied by its linkage disequilibrium with the markers we use in this analysis at the hybrid zone centre. We observed evidence of the Wahlund effect only at sample threshold distances of 30 km or greater, whereas LD values calculated from samples within 30 km of the hybrid zone centre were relatively insensitive to sample area (Appendix 3); we used samples within 20 km of the centre to calculate LD. Using the alternate method of estimating linkage disequilibrium from variance in a hybrid index (Barton and Gale 1993), in this case based on plumage, we obtained an LD estimate of 0.14. Linkage disequilibrium estimated from plumage traits was consistently higher than LD between genetic markers at each transect centre (Figure 2.2), but the two measures were strongly correlated (R2 = 0.79, p=0.04). These observations suggest that the five plumage traits used to calculate LD are not completely genetically independent, due to either pleiotropy or physical linkage of genes. Further support for the 16  non-independence of plumage traits is given by the correlation coefficients among traits, which ranged from 0.43 to 0.75, considerably higher than the correlation (0.25) between two physically unlinked genetic markers in the same sample (Appendix 2). Because of this non-independence, we do not use plumage-based LD to estimate dispersal. 2.4 DISCUSSION Our results strongly suggest that partial reproductive isolation exists between Myrtle and Audubon’s Warblers. Given a conservative estimate of migratory songbird dispersal of 10 km/generation (based on band recovery data, Moore and Dolbeer 1989, Paradis et al. 1998, and Ruegg 2008 estimated 10-100 km/generation), a neutrally diffusing cline would reach a width of 132 km in approximately 60 generations, or 108 years using a generation time of 1.8 years (Milá et al. 2007). Higher estimates of dispersal would lead to even more recent estimates of the time since initial secondary contact. A neutral cline of such recent origin would be expected to show noticeable expansion over the past four decades; thus the consistency of width of the hybrid zone between 1965 and 2005-07 is incompatible with selectively neutral mixing of the taxa. Assortative mating in the hybrid zone is absent or very weak, suggesting that premating isolation is unlikely to be a major factor in stabilizing the hybrid zone. The observed pairing data does not allow us to rule out assortative mating between the two pure forms, but does rules out assortative mating between hybrids and either pure form. These results are in line with Coyne and Orr’s (2004, ch. 2) argument that postzygotic barriers often play a major role in the early stages of speciation. In the absence of strong assortative mating, we are able to apply tension zone models (Barton and Gale 1993) to the system to investigate the strength of the partial reproductive barriers between the two forms. Assuming that all selection maintaining the hybrid zone falls on hybrids, we find selection equivalent to a single-locus heterozygote disadvantage of 18% is necessary to account for the cline width and LD we observe. We emphasize, however, that this result is approximate. Uncertainty in LD and cline width due to sampling, assumptions inherent in the tension zone model (see Methods), and the possibility of habitat-dependent (e.g. Moore 1977) or frequency-dependent (Mallet 1986) selection on parental types indicate a low degree of confidence in the specific numerical result. In particular, we cannot rule 17  out weakly assortative mating. To the extent that assortative mating causes sexual selection against hybrids or reduces the frequency of hybridization, this selection would be incorporated in our estimate of selection maintaining the hybrid zone. Nevertheless, our findings of a narrow and stable cline, non-zero linkage disequilibrium at the cline centre, and little to no assortative mating provide strong evidence that some form of postmating isolation maintains this hybrid zone. The mechanism by which this selection acts remains unknown, and will require detailed behavioural, ecological, and genetic study to determine. Differences in migratory pathway (e.g. Helbig 1991, Irwin and Irwin 2005), adaptation to different environments (e.g. Price 2008, ch. 15), and intrinsic genetic incompatibilities (e.g. Bronson et al. 2005) may all play a role. Regardless of mechanism, the presence of substantial LD long after secondary contact suggests that some form of selection impedes the fusion of the Myrtle and Audubon’s forms. When two incipient species interbreed, genetic differentiation between them will erode at neutral loci but persist where selection reduces introgression (Wu 2001, Turner et al. 2005, Via and West 2008). In the context of an old and stable clinal hybrid zone, markers closely linked to selected loci will exhibit clinal variation while neutral, nonhitchhiking loci will not (Barton and Bengtsson 1986, Durrett et al. 2000). In the presence of linkage disequilibrium, selection on any one locus will reduce introgression and narrow the cline for all other selected loci, such that the cline width for all selected loci will be determined by the total effective selection on all of them (Barton 1983). In this study, we did not sample enough loci to precisely determine the fraction of the genome over which selection maintains differences between the two incipient species, but our estimate of the strength of selection incorporates the selection acting on all loci that differ across this hybrid zone. Of the 11 loci we screened, only two were reciprocally monophyletic between Myrtle and Audubon’s Warblers. Our subsequent analyses focused on the two diagnostic markers to make use of the substantial body of hybrid zone theory developed for diagnostic markers. However, consideration of the other nine markers can provide additional insights into the maintenance of the hybrid zone. One interpretation of the lack of reciprocal monophyly in 9 of 11 screened loci is that the proportion of the genome 18  subject to restricted introgression is roughly 2/11, or 0.18 (keeping in mind that 11 loci, six of which are on a single chromosome, is a minuscule sample relative to the total genome size). However, some of the non-diagnostic loci we screened may also be subject to restricted introgression across the hybrid zone, and the shared sequence variation could result from incomplete lineage sorting (Hudson and Coyne 2002). The SNP frequency differences we observe in several introns suggest that gene flow between the taxa may be restricted at some non-diagnostic loci. In the future we plan to analyze changes in SNP frequency across the hybrid zone at the 9 non-diagnostic introns and mitochondrial DNA; we hypothesize that several clines in SNP frequency may prove concordant with the diagnostic clines we have measured in the current study. Regardless of the genomic extent of restricted introgression across this hybrid zone, we note that any incomplete barrier to gene flow between divergent populations can be viewed as partial reproductive isolation (e.g. Mayr 2001, Rundle et al. 2001, Vines et al. 2003, Jones et al. 2006), even if such a barrier applies only to certain loci (e.g. Barton and Bengtsson 1986, Wu 2001, Navarro and Barton 2003, Gay et al. 2007). Our findings indicate that the Myrtle and Audubon’s Warblers are stable and genetically distinct forms, that parts of their genomes remain distinct despite extensive hybridization, and that selection maintains differences between the taxa. We therefore suggest that these two taxa may meet the current criterion for full species status between hybridizing North American birds, that the hybrid zone be “narrow and stable” (AOU 1998). If the recent divergence date estimated by Milá et al. (2007) from mitochondrial DNA (roughly 16,000 years) is even remotely accurate, our results would suggest a relatively rapid evolution of partial reproductive isolation between the forms. However, many authors have criticized reliance on mtDNA for dating recent species divergences (Arbogast et al. 2002, Ballard and Whitlock 2004, Edwards et al. 2005). Thus, we suggest that an alternative hypothesis merits consideration: mitochondrial introgression from one form to the other may have obscured a long history of independent evolution of each form. This phenomenon, termed “cytoplasmic capture,” has been documented in plants (Rieseberg and Soltis 1991), fish (e.g. Wilson and Bernatchez 1997), birds (e.g. Weckstein et al. 2001, Irwin et al. 2009a), insects (e.g. Llopart et al. 2005), and mammals (e.g. Good et al. 2008). Evaluating this alternative hypothesis of ancient divergence and 19  recent cytoplasmic capture will require analysis of multiple nuclear markers throughout the Yellow-rumped Warbler species complex. The Yellow-rumped Warbler provides an intriguing example of bird speciation in progress. Partial postmating isolation has evolved between the forms before the onset of any measurable premating isolation, and has not yet led to reinforcement. In the face of extensive hybridization, selection maintains differences between the Audubon’s and Myrtle forms over at least part of their genomes, including the two diagnostic markers we have found as well as the loci controlling plumage pattern differences. Given the long period of time that complete reproductive isolation usually takes to evolve (usually more than a million years; Coyne and Orr 2004, Price 2008) and the many climatic cycles that occur over such a long period, many diverging pairs of taxa likely come into secondary contact multiple times on their way to becoming full species. The Yellow-rumped Warbler hybrid zone, in which there is moderate postmating isolation but little if any premating isolation, may thus represent a typical early stage in the divergence of many species. If so, the study of such hybrid zones could play a central role in developing a full understanding of speciation.  20  Table 2.1: Location, sample size, and marker frequencies for sampling localities used in this study. Negative distances from the hybrid zone centre are on the Audubon’s (southwest) side of the zone, and positive distances are on the Myrtle (northeast) side. Site name Hope Whistler Lac Le Jeune Hazelton Nass R. Gavin Lk. Yard Ck. Crowsnest Pass Golden Whiskers Point Bob Quinn Lk. Mt. Robson Blaeberry Kennedy Lk. Lougheed Evan Thomas Yellowhead Pass Willow Ck. Sibbald Ck. W. Rafter 6 Yamnuska Pine Pass W. Sibbald Ck. E. Pyramid Lk. Saskatch. R. X-ing Pine Pass E. Todagin Palisades Waiparous Cline R. Pocahontas Tumbler Ridge Abraham Lk. Morchuea Lk. Hinton Moberly Lk. Rocky Mtn. House Prophet R. Fox Ck. Toad R. Watson Lk. Slave Lk. Cold Lk.  Lat.  Long.  49.21 50.16 50.50 55.25 55.70 52.49 50.90 49.70 51.40 54.94 57.06 53.02 51.61 55.12 50.78 50.89 52.89 57.44 51.03 51.07 51.12 55.49 51.04 52.90 52.00 55.52 57.62 53.02 51.38 52.16 53.20 55.10 52.34 57.98 53.39 55.77 52.43 57.97 54.37 58.87 60.08 55.48 54.72  -121.36 -122.94 -120.51 -127.40 -128.79 -121.71 -118.82 -114.56 -116.99 -122.99 -130.27 -119.28 -116.77 -122.81 -115.15 -115.13 -118.39 -130.24 -115.02 -115.02 -115.08 -122.78 -114.91 -118.09 -116.66 -122.67 -130.08 -118.09 -115.01 -116.46 -117.92 -121.11 -116.34 -130.07 -117.74 -121.60 -115.00 -122.78 -116.89 -125.38 -128.86 -114.85 -110.09  Distance from centre (km) -440 -435 -294 -240 -199 -183 -179 -143 -61 -53 -40 -37 -33 -33 -31 -19 -10 -5 -2 1 1 4 4 7 8 10 12 15 23 30 36 49 51 52 59 74 111 146 182 182 289 358 515  n  Plumage  CHD1Z  12 10 30 10 10 20 14 11 17 13 11 15 16 13 11 25 11 13 28 7 23 17 20 13 31 8 29 16 14 26 9 13 13 13 12 12 11 12 13 13 12 12 13  0.88 0.92 0.94 0.88 0.93 0.91 0.93 0.93 0.82 0.85 0.75 0.81 0.80 0.75 0.73 0.67 0.60 0.62 0.57 0.49 0.48 0.48 0.40 0.40 0.45 0.33 0.34 0.51 0.28 0.43 0.26 0.21 0.24 0.21 0.20 0.17 0.17 0.09 0.20 0.12 0.10 0.09 0.16  1.00 1.00 1.00 0.80 0.85 0.89 0.96 1.00 0.91 1.00 0.84 0.58 0.81 0.60 0.84 0.73 0.43 0.77 0.67 0.50 0.54 0.29 0.43 0.43 0.53 0.46 0.33 0.41 0.27 0.30 0.19 0.22 0.43 0.11 0.25 0.22 0.10 0.05 0.20 0.04 0.09 0.09 0.04  NumtDco1 1.00 1.00 0.93 0.80 0.85 0.88 0.93 0.86 0.94 0.65 0.82 0.64 0.68 0.58 0.77 0.54 0.55 0.46 0.38 0.36 0.39 0.28 0.39 0.38 0.40 0.43 0.28 0.34 0.29 0.35 0.17 0.06 0.31 0.19 0.13 0.22 0.00 0.05 0.04 0.04 0.00 0.08 0.04  21  Table 2.2: Details of loci sequenced. Sample size indicates the number of chromosomes sequenced (two chromosomes per individual). Diagnostic loci (CHD1Z and Numt-Dco1) have much greater FST between Myrtle and Audubon’s forms, and lower within-taxon polymorphism (πA for Audubon’s, πM for Myrtle) than other loci. Net sequence divergence (DA) varies among the loci. Average polymorphism is more than two times higher among autosomal than Z-linked loci. Locus  Sample size  Chromosome  Total base pairs  Coding bp  Noncoding bp  FST  πA  πM  DA  GH1  8  1  594  8  586  -0.1783  0.0077  0.0139  -0.0016  RPL30  8  2  945  0  945  0.1827  0.0198  0.0127  0.0036  LHCGR  22  3  648  6  642  0.0045  0.0126  0.0167  0.0001  NumtDco1  28  4  343  0  343  0.6709  0.0004  0.0020  0.0025  RPL5  8  8  573  24  549  0.2821  0.0059  0.0023  0.0016  CEPU1  8  24  567  0  567  0.0513  0.0072  0.0039  0.0003  24555  8  Z  500  56  444  0.1458  0.0060  0.0077  0.0012  ADAMTS6  8  Z  546  49  497  -0.0667  0.0037  0.0012  -0.0002  BRM  8  Z  332  35  297  0.0000  0.0015  0.0015  0.0000  CHD1Z  40  Z  613  127  486  0.6882  0.0006  0.0013  0.0021  PTCH  8  Z  554  22  532  -0.0208  0.0127  0.0021  -0.0002  0.6795  0.0005  0.0017  0.0023  Mean, Diagnostic Mean, Other Z  Z  0.0146  0.0060  0.0031  0.0002  Mean, Other Auto.  various  0.0685  0.0106  0.0099  0.0008  22  Table 2.3: Proportion of randomization trials resulting in less-than-observed correlation between genotypes or phenotypes of social mates. Parameter A is the maximum difference between a female’s hybrid index and that of an acceptable mate; a low A signifies a stringent mate preference. Parameter n is the maximum number of potential mates a female evaluates before mating randomly; an n of 1 or an A of 1 corresponds to random mating. Values in bold indicate parameter combinations incompatible with observed pairing data (α = 0.05). n A  1 0.25 0.5 0.67 0.75 1.0  0.650 0.654 0.652 0.651 0.649  Genetic hybrid index 2 3 5 0.037 0.021 0.101 0.239 0.648  0.001 0.001 0.026 0.170 0.649  0.000 0.000 0.008 0.147 0.651  8  1  0.000 0.000 0.006 0.137 0.653  0.905 0.905 0.907 0.905 0.908  Plumage hybrid index 2 3 5 0.076 0.182 0.565 0.730 0.907  0.002 0.035 0.466 0.690 0.905  0.000 0.008 0.437 0.681 0.905  8 0.000 0.005 0.439 0.682 0.906  23  Figure 2.1: Location of five hybrid zone transects and allopatric reference sample sites in British Columbia and Alberta. Parallel curves on map denote hybrid zone centre and 100km buffer. Plots of plumage pattern and two genetic markers along five transects show the consistent width of the hybrid zone between Myrtle and Audubon’s Warblers. Shaded areas of plots contain samples outside the 100-km hybrid zone buffer; these same samples are plotted on graphs for all five transects and were used in the cline analysis of each transect.  24  Figure 2.2: Linkage disequilibrium estimated from diagnostic genetic markers and from plumage traits for five transect centres (all samples within 20 km of each centre) and for a pooled sample from all transect centres. Error bars represent bootstrap 95% confidence intervals. Sample sizes for each transect are 39 (A), 19 (B), 33 (C), 22 (D), 87 (E), and 200 (Pooled).  25  CHAPTER 3: NO REPRODUCTIVE BARRIER BY SONG IN THE YELLOW-RUMPED WARBLER HYBRID ZONE2 3.1 INTRODUCTION Both pre- and post-zygotic isolating barriers may contribute to the maintenance of species cohesion when closely related species coexist, but they often evolve at different rates (Coyne and Orr 2004). While intrinsic post-zygotic isolation is often incomplete millions of years after divergence, pre-zygotic isolating mechanisms such as assortative mating are often thought to evolve much more quickly (Coyne and Orr 1989, Price and Bouvier 2002). In birds, the finding that complete postzygotic isolation evolves more slowly than new species form (Price and Bouvier 2002) has led to the hypothesis that prezygotic isolation is a critical part of the speciation process in this group (Slabbekoorn and Smith 2002, Edwards et al. 2005). Hybrid zones offer an ideal opportunity to explore the relative influences of pre- and post-zygotic isolating mechanisms, because they represent an intermediate stage in the speciation process, where reproductive barriers have evolved but are not complete (Barton and Hewitt 1989, Harrison 1993). One mechanism of prezygotic isolation that has received particular attention in birds is territorial song. Males of many species use song to defend territories from other males and to attract females, and songs are often distinct between closely related species (e.g. Grant and Grant 1997, Irwin 2000, Payne et al. 2002, Toews and Irwin 2008). Testing female mate choice in birds is difficult, so male-male interactions are often used as an indicator of a species' ability to distinguish conspecific from heterospecific signals (e.g. Ratcliffe and Grant 1983, Baker 1991, Irwin et al. 2001, Grant and Grant 2002a, 2002b, Patten et al. 2004, Balakrishnan and Sorenson 2006, Seddon and Tobias 2007, Uy et al. 2009a, 2009b). In the Yellow-rumped Warbler (Dendroica coronata) complex, a hybrid zone between two visually distinct populations has been well-studied, but the role of song remains unclear. The eastern Myrtle and western Audubon's forms, currently designated 2  A version of this chapter will be submitted for publication. Brelsford, A., K. Haakons, M. Moretti, and D. E. Irwin. No reproductive barrier by song in the Yellow-rumped Warbler hybrid zone. 26  subspecies, meet and interbreed in the Canadian Rocky Mountains. Three studies of the hybrid zone over a 40-year time span have found consistent position and width (Hubbard 1969, Barrowclough 1980, Brelsford and Irwin 2009), and the most recent study concluded that the evidence was incompatible with a neutral, expanding hybrid zone. Brelsford and Irwin (2009) also found no significant assortative mating between the subspecies, and concluded that some form of post-mating isolation was responsible for maintaining the hybrid zone. Anecdotal reports suggest that slight differences in song exist between Myrtle and Audubon's Warblers (Hubbard 1969), but song has not been studied quantitatively. To our knowledge, only one previous study has tested the response of parental bird species and their hybrids to hybrid vocalizations (den Hartog et al. 2007). In this study, we measured differences between territorial songs of Myrtle, Audubon’s and hybrid warblers in the vicinity of the hybrid zone in the Canadian Rockies. We also tested the response to these three types of song in five locations spanning the hybrid zone. If Brelsford and Irwin's (2009) hypothesis (that post-mating isolation rather than pre-mating isolation stabilizes the hybrid zone) is correct, we would expect to find no difference in song response across the hybrid zone. Alternatively, if song is important in maintaining incipient species boundaries in this complex (as it is in many others; Grant and Grant 1997, Irwin et al. 2001, Payne et al. 2002), we would expect to find strong differences in song between the parental types, with reduced response to hybrid and heterospecific songs. 3.2 METHODS 3.2.1 Song recording We recorded songs of 58 territorial male yellow-rumped warblers at 12 locations in Alberta and British Columbia (see figure 1) in April-June of 2005. Songs were recorded in WAV format at a sampling rate of 44.1 kHz, using a shotgun microphone (Audio-Tecnica) and an R1 digital recorder (Edirol). Whenever possible, we recorded song samples of at least two minutes. Of the 58 individuals sampled for this study, 48 were captured for genetic and morphological analysis (see Brelsford and Irwin 2009). Each captured bird was scored for a hybrid index developed by Hubbard (1969) and modified by Brelsford and Irwin (2009). Briefly, five traits were scored as resembling 27  Audubon’s (score 0), Myrtle (score 1), or intermediate (score 0.5), and an individual’s hybrid index was calculated as the average of its five trait scores. We recorded songs either immediately prior to capture or at least two days after capture; birds recorded postcapture were identified by a unique combination of coloured leg bands. 3.2.2 Song analysis We used hybrid index scores to designate individuals as Audubon’s ( index ≥ 0.8), hybrid (0.2 < index < 0.8), or Myrtle (index ≤ 0.2). Study sites were designated as Audubon’s, hybrid, or Myrtle locations according to which of the three types was most common. For each recorded individual, we chose three songs randomly and used the Raven program (Cornell Laboratory of Ornithology) to generate spectrograms and extract acoustic data from the raw sound files. The variables we used were maximum and minimum frequency, peak frequency (the frequency with the highest amplitude across the song), number of syllables (the basic units of a song), number of syllable types (discrete classes of syllable), and song length. From these variables, we calculated the additional variables average syllable length (number of syllables / song length) and frequency range (max. frequency – min. frequency). We tested for differences in each variable among Audubon’s, hybrid, and Myrtle sites using ANOVA. We also performed a principal component analysis implemented in R (R Development Core Team) using the command 'princomp'. We then performed ANOVA on the first two principal components. Classifying warbler songs by the most common type at each site may obscure within-site variation between hybrids and pure parental types. To account for this possibility, we fit a linear regression of each song trait by individual hybrid index for birds at majority-hybrid locations. Song and plumage pattern are inherited differently, since oscine songbirds may learn their songs from neighbouring males other than their fathers. To test whether song variation is decoupled from plumage variation in the hybrid zone, we calculated correlation coefficients between each song variable and individual hybrid index, and between each song variable and site average hybrid index. A higher correlation between song and site average hybrid index would indicate that the plumage pattern of a warbler’s 28  neighbours conveys more information about that individual’s song than does the individual’s own plumage pattern. 3.2.3 Playback trials We selected 16 Yellow-rumped Warbler recordings from six locations (two each of Audubon's, hybrid, and Myrtle) for use in playback trials to test male song response. Recordings were chosen on the basis of sound quality and low background noise, and each location was represented by at least two recordings. Two additional recordings of different species (Warbling Vireo Vireo gilvus and Wilson's Warbler Wilsonia pusilla) were used as controls. Peak amplitude was adjusted to be constant across all songs. We carried out a total of 189 playback trials between 16 May and 22 June 2006 at five locations: Lesser Slave Lake (Myrtle Warblers, 358 km east of hybrid zone centre), Waiparous River (Myrtle Warblers, 23 km east of hybrid zone centre), Kananaskis (hybrid warblers, at hybrid zone centre), Peter Lougheed Provincial Park (Audubon's Warblers, 31 km west of hybrid zone centre), and Lac Le Jeune (Audubon's Warblers, 294 km west of hybrid zone centre). Each trial was run as follows: A focal singing male was chosen, and a portable CD player was placed at the base of a tree 40 metres from the focal male and at least 40 metres from any other Yellow-rumped Warbler. One of the 18 recordings was chosen at random and played on continuous loop for ten minutes. During each one-minute interval, an observer recorded the closest approach of any Yellowrumped Warbler to the CD player. Distances from the CD player were later subtracted from 40 metres to obtain the approach distance (net movement of the bird toward the CD player), which was used in subsequent analyses. 3.2.4 Playback analysis To reduce the number of variables in the playback response dataset, we performed a principal component analysis of the minute-by-minute approach distances. Principal component 1 accounted for most of the variation in the data, and was highly correlated with the average approach distance across the 10 minutes of each trial (see Results). For this reason, we used average approach distance in all subsequent analyses of playback response. We compared average approach distances for each 10-minute trial across song types using linear mixed-effect models. First, we fit a model to the entire data set in 29  which approach distance is a function of song type (Audubon's, Myrtle, hybrid, control), locally common form of warbler (Audubon's, Myrtle, or hybrid), and song type * locally common warbler interaction as fixed factors, with the specific song recording used as a random factor. A significant interaction term indicates that the two pure species and their hybrids respond differently to the different song types. Second, for each of the five sites, we fit a model in which response is a function of song type, with song recording as a random factor. In these models, a significant term for a song type indicates that the response to that song type at that location is greater than zero. Third, for three sites where the comparison is possible, we tested whether the response to hybrid song is significantly different from the response to the song of the locally common type. (Audubon’s for site B, Myrtle for sites D and E). 3.3 RESULTS 3.3.1 Song analysis We found significant differences among the three types of warbler in 5 of 8 acoustic variables (ANOVA; Table 3.1). Differences were significant in maximum frequency, minimum frequency, peak frequency, number of syllable types, and mean syllable length. No significant differences were observed in frequency range, duration, or number of syllables. Principal components 1 and 2 account for 63.7% and 28.3%, respectively, of the variation in the song characteristics we measured. Principal component 1 was positively correlated with number of syllables, frequency range, and maximum frequency, while principal component 2 was negatively correlated with peak frequency and minimum frequency, and positively correlated with song duration and number of note types (Figure 3.2, Table 3.2). Myrtle, Audubon's, and hybrid songs differed from one another most strongly in principal component 2 (ANOVA, F=6.81, df=2,55, p=0.0023) but also in principal component 1 (ANOVA, F=3.71, df=2,55, p=0.031). Myrtle songs differed from Audubon’s songs in both principal components, but there were no significant differences between hybrids and either parental form (Tukey's HSD test). 30  Among songs at majority-hybrid sites, we found a significant relationship between hybrid index and two of the eight acoustic variables (Table 3.3). Number of syllables and song duration were negatively correlated with hybrid index (i.e. higher in Myrtle-like and lower in Audubon’s-like birds). Principal components 1 and 2 showed no significant relationship with hybrid index. Six of the eight acoustic variables, as well as the first two principal components, showed higher correlation with site average hybrid index than with individual hybrid index (Table 3.3). Number of syllables and mean syllable length were more strongly correlated with individual than site hybrid index. 3.3.2 Playback analysis Principal component 1 explained 74% of the variation in playback response data, and was strongly correlated with average approach distance (correlation coefficient = 0.9994). We used average approach distance in subsequent analyses because it is biologically more meaningful than unitless principal component scores. We found a significant effect of interaction between song type and locally common warbler (linear mixed effects model, F6,163=6.61, p<0.0001) in our simultaneous analysis of all playback trials. Responses to each song type by site, with standard errors, are shown in Figure 3.1. The significant interaction was largely due to the low response to Myrtle song by Audubon’s Warblers in sites A and B. We found Audubon's and hybrid songs to elicit a significant response at all sites tested, while Myrtle songs elicited a significant response at sites B, C, D, and E (Table 3.4). Response to hybrid song was not significantly different from Audubon’s song at site B (linear mixed effects model, p=0.86), and did not differ from Myrtle song at sites D (p=0.57) and E (p=0.56). 3.4 DISCUSSION In many of the best-studied examples of ongoing or recent bird speciation, songs have been implicated as important reproductive barriers between species (Irwin et al. 2001, Grant and Grant 1997, Payne et al. 2002, Toews and Irwin 2008, Uy et al. 2009, Patten et al. 2004). In contrast, we find that differences among Myrtle, Audubon’s, and 31  hybrid warblers are subtle (Figures 3.2, 3.3), and that hybrid songs are effective territorial displays both within and outside the hybrid zone. This confirms earlier anecdotal evidence that song differences between the subspecies are minor (Hubbard 1969, Dunn and Garrett 1997). Although the pure Myrtle and Audubon’s songs were differentiated along PC1 and PC2, there was considerable overlap between them on both axes. Thus, it is not surprising that the responses to intermediate hybrid songs did not differ from the responses to the songs of the locally common subspecies. We did find asymmetric song response between the pure subspecies (Myrtles recognized Audubon's song, but not the reverse). However, the evolutionary consequences of this difference may be minor, since birds can learn to recognize an initially unfamiliar song (Rohwer 1973, Price 2008). One of us (AB) observed two cases of nearly-pure male Myrtle Warblers holding territories west of the hybrid zone, and the pure Audubon's Warblers surrounding both of these territories were observed to recognize and respond to the Myrtle songs. In one of these cases, the territorial Myrtle male was observed paired with an Audubon's female, and in the other case the male was carrying food for nestlings, implying that Myrtle Warblers can successfully communicate with Audubon’s Warblers of both sexes despite the lack of song recognition that we document in this study. It could be argued that the strong response we observed for hybrid songs is spurious, since some of the recordings used for playback stimuli were obtained from Myrtle-like or Audubon’s-like hybrids rather than true intermediates. We find this argument unpersuasive for the following reasons. First, song and phenotype are decoupled in the hybrid zone: a warbler’s appearance is less predictive of its song than the appearance of its neighbours (Table 3.3), and all of the hybrid songs used as stimuli were recorded near the centre of the hybrid zone where the site average hybrid index was close to 0.5. Second, if the hybrid songs we used as stimuli were instead mixture of songs resembling pure Myrtle and Audubon’s Warblers, we would expect to see a reduced response in site B, since Myrtle songs elicited only a weak response there (Figure 3.1). Instead, there is no indication of a weaker response to hybrid songs at that site. Finally, the songs we used as playback stimuli are typical of the songs recorded throughout the 32  hybrid zone in PC1 and PC2 (Figure 2), suggesting that on average, hybrid songs are as effective as parental songs both within and outside the hybrid zone. If certain types of hybrid songs are indeed less effective, the difference is not large enough to cause a measurable reduction in the average response to hybrid song. We are aware of only one previous study that has compared responses to hybrid and pure vocalizations within and outside of a hybrid zone. That study of the hybrid zone between the doves Streptopelia vinacea and S. capicola (den Hartog et al. 2007) found that hybrid territorial display vocalizations received as strong a response as pure parental vocalizations only within the hybrid zone, and were less effective outside the zone. In contrast to their results, we find that hybrid songs are effective in all locations. The contrasting results of these two studies highlight the fact that the mechanisms maintaining species boundaries are idiosyncratic, and that while song is important in reproductive barriers in some cases (Grant and Grant 1997, Irwin et al. 2001, Payne et al. 2002, Patten et al. 2004, Toews and Irwin 2008, Uy et al. 2009), it plays a smaller role in others. Previous work in the Yellow-rumped Warbler system hypothesized that reproductive barriers maintaining the stable hybrid zone between Myrtle and Audubon’s Warblers were likely post-mating, since no evidence of assortative mating by plumage or genotype was observed (Brelsford and Irwin 2009). This study’s results complement that hypothesis, since hybrid songs are effective territorial signals within and outside the hybrid zone. Male song response is often used as a proxy for female song response, and the conclusions of this study and Brelsford and Irwin (2009) suggest that hybrids do not suffer from an inability to attract mates or defend territories as a result of their songs. Not surprisingly given their history of independent evolution, Audubon’s and Myrtle Warblers do exhibit some differences in song, although these differences are subtle. Myrtle songs have, on average, a higher pitch, shorter duration, and fewer syllable types than Audubon’s songs, although there is considerable overlap. Our results are consistent with a previous anecdotal report indicating slight song differences (Hubbard 1969), which also noted the higher pitch of Myrtle songs. Calls, which are not learned, seem to be more distinct than songs between Myrtle and Audubon’s forms (personal observation, Dunn and Garret 1997, Hunt and Flaspohler 1998), although no data are available to confirm this. It is possible that song learning in and around the hybrid zone 33  has led to convergence in song between the two subspecies, while gene flow restricted by postmating isolation has allowed distinct calls to persist, similar to the differentiation observed in plumage and genetic markers. In an analogous case, the Greenish Warbler ring species, Irwin et al. (2008) found that patterns of call variation were more consistent with genetic variation than with song variation. Our study shows that the slight song differences that exist between Myrtle and Audubon’s Warblers are insufficient to cause reproductive isolation between them upon secondary contact. Hybrid songs are effective territorial signals within and outside the hybrid zone. To our knowledge, this is only the second study to test the effectiveness hybrid and parental vocalizations across a hybrid zone (after den Hartog et al. 2007), and the first in a species with learned vocalizations. Theory predicts that song learning may speed or hinder the evolution of reproductive isolation depending on context (Irwin and Price 1999, Lachlan and Servedio 2004, Servedio et al. 2009), and the Yellow-rumped Warblers may be a case where learning of heterospecific song has prevented the evolution of premating isolation despite substantial postmating isolation.  34  Table 3.1: Means and standard deviations of 8 acoustic variables measured from songs of Audubon’s, hybrid, and Myrtle warblers. Maximum frequency, minimum frequency, peak frequency, number of syllable types, and mean syllable length are significantly different among the three song types. Variable  Audubon's mean ± SD  hybrid mean ± SD  Myrtle mean ± SD  F2,55  p  Maximum frequency  6190 ± 380  6430 ± 420  6700 ± 620  4.8  0.01  Minimum frequency  2840 ± 210  2990 ± 290  3150 ± 280  6.33  0.003  Peak frequency  4250 ± 420  4500 ± 440  4830 ± 370  8.84  0.0005  Frequency range  3350 ± 410  3440 ± 560  3550 ± 620  0.6  0.55  Song duration  1.59 ± 0.20  1.50 ± 0.23  1.49 ± 0.17  1.49  0.23  Number of syllable types  2.30 ± 0.68  1.99 ± 0.65  1.74 ± 0.44  3.73  0.03  Number of syllables  8.19 ± 1.34  8.65 ± 1.58  8.82 ± 1.10  1.06  0.35  0.199 ± 0.030  0.175 ± 0.018  0.171 ± 0.020  8.19  0.0008  Mean syllable length  35  Table 3.2: Factor loadings on first two principal components of song variation. Max. frequency Min. frequency Peak frequency Frequency range Song duration # syllable types # syllables Syllable duration Eigenvalue % variance explained  PC1 0.466 -0.040 0.188 0.471 0.281 0.295 0.503 -0.324  PC2 -0.246 -0.476 -0.491 0.012 0.444 0.404 0.085 0.323  2.74 63.7  2.17 28.3  36  Table 3.3: Significance of linear regressions of each acoustic variable against individual hybrid index (d.f. = 19), and correlations between each acoustic variable and both individual and site average hybrid index. Variable  Linear regression against individual hybrid index  Correlation coefficient  t19  p  Maximum frequency  0.08  0.94  Individual hybrid index -0.32  Minimum frequency  -0.15  0.89  -0.29  -0.37  Peak frequency  0.28  0.78  -0.36  -0.42  Site hybrid index -0.37  Frequency range  0.12  0.9  -0.17  -0.19  Song duration  -2.53  0.02  0  0.21  Number of syllable types  -0.02  0.99  0.26  0.37  Number of syllables  -2.47  0.02  -0.37  -0.18  Mean syllable length  1.14  0.27  0.5  0.47  0.14 0.12  0.88 0.9  -0.29 0.3  -0.33 0.37  Principal component 1 Principal component 2  37  Table 3.4: Results of statistical tests of whether birds responded to playbacks of controls, Audubon’s, hybrid, and Myrtle songs at five sites, and whether response to hybrid and local songs differed. Response to hybrid songs is not significantly different from response to Audubon’s in site B, or to Myrtle in sites D and E. Response to control songs was not significant in any site, while all other song/site combinations aside from Myrtle song in site A produced significant responses.  site A B C D E  Hybrid < parental  control ≠ 0  df=7, F1,7=0.034, p=0.86  t10=0.069, p=0.95 t11=.46, p=.6553 t11=-.24, p=.8118 t13=.47, p=0.6432  df=6, F1,6=0.36, p=0.57 df=8, F1,8=0.37, P=0.56  Audubon ≠ 0 t2=17.7, p=0.0032 t10=10.4, p<0.0001 t11=4.49, p=0.0009 t11=9.64, p<0.0001 t13=4.66, p=0.0004  hybrid ≠ 0 t10=12.2, p<0.0001 t11=4.77, p=0.0006 t11=8.42, p<0.0001 t13=5.48, p=0.0001  Myrtle ≠ 0 t2=4.04, p=0.056 t10=3.46, p=0.0061 t11=4.23, p=0.001 t11=9.85, p<0.0001 t13=6.82, p<0.0001  38  Figure 3.1: Locations of song sampling, playback trials, and hybrid zone centre and edges. Songs of Audubon’s (A) and hybrid (H) warblers elicited strong responses at all sites tested, while Myrtle (M) songs elicited strong response only at sites C, D, and E. Control recordings (C) did not elicit responses at any site.  39  Figure 3.2: Slight differentiation between songs of Myrtle and Audubon’s Warblers and hybrid intermediacy, as shown by principal component analysis of acoustic variables. Triangles denote songs used in playback trials. Black symbols represent songs from majority-Audubon’s sites, white symbols majority-Myrtle sites, and grey symbols majority-hybrid sites. 40  Figure 3.3: Representative spectrograms from six of the 16 Yellow-rumped Warbler recordings used as playback stimuli.  41  CHAPTER 4: HYBRID ORIGIN OF AUDUBON’S WARBLER3 4.1 INTRODUCTION Botanists and zoologists have traditionally held different views on the evolutionary importance of hybridization (Harrison 1993, Arnold 1997). Zoologists have generally emphasized the role of hybridization in preventing or reversing divergence between incipient species (e.g. Mayr 1963), while botanists have pointed to the possibility of cross-species spread of beneficial mutations and the formation of novel lineages (e.g. Stebbins 1959). Hybrid speciation has long been known to occur in plants, often involving polyploidy (Otto and Whitton 2000) but in some cases without it (Rieseberg 1997). In recent years, several cases of homoploid hybrid speciation have been documented in animals (reviewed in Mavárez and Linares 2008), including fish (e.g. Nolte et al. 2005, Meyer et al. 2006), insects (e.g. Mavárez et al. 2006, Gompert et al. 2006), and Daphnia (Taylor et al. 2005), suggesting that this phenomenon may be more important in animals than was previously thought. Hybrid origins have been proposed for a few bird taxa, but evidence for these cases is sparse or equivocal. The Adelaide Rosella Platycercus elegans adelaidae was proposed as a possible hybrid species (Price 2008) on the basis of intermediate colour pattern between two related forms with nearby distributions, the Crimson Rosella P. e. elegans and Yellow Rosella P. e. flaveolus. Joseph et al. (2008) examined mtDNA and microsatellite variation across this complex with ambiguous results; the intermediate form may have originated through hybridization, but may have been ancestral to the two extreme types, forming an "incomplete ring species." The Italian Sparrow (Passer domesticus italiae) presents another possible case of hybrid species (Price 2008) on the basis of morphological and geographic intermediacy between the House Sparrow (P. d. domesticus) and Spanish Sparrow (P. hispaniolensis). Experimental hybrids between house and Spanish Sparrows resemble wild Italian Sparrows (Johnston 1969), but molecular evidence is lacking.  3  A version of this chapter will be submitted for publication. Brelsford, A., B. Milá, and D. E. Irwin. Hybrid origin of Audubon’s warbler. 42  Finally, Grant and Grant (2009) show that a lineage of Darwin’s finches descended from a pair of hybrid Geospiza scandens x G. fortis individuals have been reproductively isolated from sympatric populations of both parental species for three generations due to their unique song, but the authors note that a small population size makes it unlikely that this incipient species will persist indefinitely. The Yellow-rumped Warbler (Dendroica coronata) complex is composed of four well-marked forms, the relationships and taxonomic status of which have been extensively debated (Figure 4.1; Hubbard 1969, Barrowclough 1980, Zink and McKitrick 1995, Rohwer and Wood 1998, Johnson et al. 1999, Milá et al. 2007). The four forms were initially described as separate species Dendroica coronata (Linnaeus 1766), D. auduboni (Townsend 1837), D. nigrifrons (Brewster 1889), and D. goldmani (Nelson 1897). Later, nigrifrons and goldmani were considered subspecies of auduboni on the basis of similar plumage (Oberholser 1921), while coronata was maintained as a full species, presumably due to its distinct facial plumage pattern (Figure 4.1). All four taxa have been considered conspecific since 1973 (AOU 1973) based on hybridization between coronata and auduboni. Milá et al. (2007) found a surprising similarity in mtDNA between auduboni and coronata, and a deep divergence between these forms and the two southern forms. The suggested interpretation of this pattern was that coronata diverged very recently from auduboni, and rapidly evolved its distinctive male breedingseason plumage under strong sexual selection (Milá et al. 2007). Using nuclear DNA variation in the hybrid zone, Brelsford and Irwin (2009) showed that the coronata/auduboni hybrid zone is stable and maintained largely by post-mating reproductive barriers. Intrinsic post-zygotic barriers evolve slowly in birds (Price and Bouvier 2002), leading Brelsford and Irwin (2009) to hypothesize that the mtDNA similarity between the two forms could be a result of cytoplasmic capture rather than very recent divergence from a common ancestor. Here, we survey mitochondrial and nuclear genetic variation across the range of the species complex to test three alternative hypotheses: auduboni is 1) a close relative of coronata, as mtDNA suggests (Milá et al 2007); 2) a close relative of nigrifrons, with mitochondrial introgression from coronata to auduboni obscuring this relationship (Brelsford and Irwin 2009); or 3) a hybrid taxon, with substantial genetic contributions 43  from both coronata and nigrifrons. If auduboni is a linage of hybrid origin, it raises the possibility that hybrid speciation may be occurring in this and other groups of birds. 4.2 METHODS 4.2.1 Field sampling Warblers were captured on their breeding grounds in 2001-2007 (see Figure 4.1 for locations) using mist nets and song playbacks. We chose sampling sites to broadly cover the range of the species complex and to enable analysis of clinal genetic variation along a transect spanning the coronata, auduboni, and nigrifrons forms. Blood samples were obtained by brachial venipuncture, and stored in Queens lysis buffer (Seutin 1991). 4.2.2 Morphology We extracted data on wing length and throat colour from previously published studies (Hubbard 1969, 1970). One of us (AB) scored the extent of black in the cheek patch from photographs taken in the field. Scoring criteria were as follows: 0) lores and auriculars completely grey, 1) black feathers in lores and around the eye, 2) black extends beyond eye into the auriculars, but doesn't reach the posterior edge of the auricular region, 3) black feathers present throughout the lores and auriculars, but grey covers >5% of the area, and 4) lores and auriculars >95% black. 4.2.3 Molecular techniques i. Mitochondrial DNA We used mitochondrial ATPase 6 and 8 sequences obtained by Milá et al. (2007, GenBank accession numbers DQ855192 - DQ855209), supplemented by sequencing 81 additional auduboni individuals sampled along a transect from Idaho to Arizona (Figure 4.1, localities G, I, J, K, L). PCR and sequencing protocols followed Milá et al. (2007). ii. AFLP Amplified fragment length polymorphism (AFLP) reactions were carried out according to Vos et al. (1995), with minor modifications described by Toews and Irwin (2008). Fluorescently labelled AFLP products were separated by 6.5% polyacrylamide gel electrophoresis, visualized on a 4300 DNA analyzer (LI-COR, Lincoln, Nebraska), and scored manually using SAGA 2.0 (LI-COR, Lincoln, Nebraska). AFLP analysis was carried out on 119 individuals, 38 of which were run twice starting from a new DNA 44  extraction. Only loci that could be scored unambiguously were analysed. We used 10 primer combinations (Appendix 4), which yielded 311 polymorphic markers with average repeatability of 98.6%. iii. AFLP sequences We isolated and resequenced two AFLP markers that showed high differentiation across the species range, following the protocol of Brugmans et al. (2003) with slight modifications. We followed Brugmans et al.'s (2003) minisequencing procedure, using a generalized set of 12 degenerate primers to determine three additional selective bases for each M primer. We then carried out a PCR reaction using the band-specific M primer and original E primer, separated the multiple PCR products on a 4% agarose gel, and excised a plug from the target band. The plug was incubated in 50 µl of water for 1 hour at 55°C, and this water was used as template for an additional PCR reaction using the bandspecific M primer and original E primer. After the amplification of a single PCR product of the appropriate length was confirmed, the fragments were sequenced bidirectionally by Macrogen, Inc., using the amplification primers. We then sequenced outward in both directions from each AFLP fragment using a DNA Walking SpeedUp kit (SeeGene, Inc., Seoul, Korea), and designed primers to amplify a fragment of approximately 800 bp spanning the entire AFLP marker. These sequences aligned unambiguously to the zebra finch genome; one maps to chromosome 3 near the predicted gene KIAA1383, while the other maps to chromosome 6 within an intron of the gene ARID5B. iv. Nuclear sequences We sequenced 12 nuclear loci in at least 5 individuals of each species. Loci sequenced included 9 introns, 2 AFLP-derived loci, and one numt (See Table 4.1 for details). Because of limited quantities of template DNA available from some sites, we used nested PCR to amplify loci for sequencing. Primers and reaction conditions are shown in Appendix 5. Amplicons were sequenced bidirectionally by Macrogen Inc. or by the Genome Quebec McGill Innovation Centre, using ABI 3730XL sequencers. 4.2.4 Analysis Sequences were aligned and proofread using BioEdit (Hall 1999). For nuclear loci, the phase of heterozygous sites was determined using Phase (Stephens and Donnelly 2003). We used TCS (Clement et al. 2000) to reconstruct haplotype networks, and 45  DNAsp (Rozas et al. 2003) to calculate pairwise FST values between subspecies. Multidimensional scaling based on Jaccard distances (Kingston and Rosel 2004) was used to graphically display AFLP-derived genetic distances among individuals in two dimensions; this analysis was implemented in R (R Development Core Team) using the package Vegan (Oksanen et al. 2008). We used Bayesian clustering implemented in Structure 2.3 (Hubisz et al. 2009) to assess population structure, and used the methods of Evanno et al. (2005) to determine the appropriate number of populations in the sample. To enable comparisons of clinal genetic and phenotypic variation, we measured the minimum straight-line distance from each sampled population to the centre of the coronata/auduboni hybrid zone (as described in Brelsford and Irwin 2009). 4.3 RESULTS Mitochondrial and AFLP analysis identified three distinct clusters in the Yellowrumped Warbler complex (Figures 4.1, 4.3E), but the clusters identified by each type of marker were incongruent. Both types of markers separated coronata, nigrifrons, and goldmani from each other, but the placement of auduboni differed. Structure, using AFLP data, grouped auduboni with nigrifrons, while mitochondrial DNA of most auduboni grouped with coronata, consistent with the results of Milá et al. (2007). However, 38 of 41 auduboni sampled in Arizona resembled nigrifrons in their mtDNA. While mtDNA shows most auduboni to be similar to coronata, nuclear markers show auduboni to be intermediate between coronata and nigrifrons. Multidimensional scaling of AFLP data shows that while the auduboni cluster overlaps partially with that of nigrifrons, its centroid is located between the centroids of the coronata and nigrifrons clusters (Figure 4.2). This suggests that most of the genetic differentiation between auduboni and nigrifrons is a result of the presence of alleles characteristic of coronata, which are more common in northern auduboni populations. The goldmani subspecies is the most genetically distinct form on the basis of AFLPs, largely due to its much lower within-population variation (Figure 4.2, Table 4.2). Nuclear sequences are consistent with the close relationship between auduboni and nigrifrons seen in AFLPs. Of 10 non-AFLP-derived nuclear loci sequenced, most showed very little differentiation among the four taxa (Table 4.1). Two of these loci, 46  CHD1Z and Numt-Dco1, were highly differentiated between coronata and auduboni (Brelsford and Irwin 2009), and in both cases showed little differentiation between auduboni and nigrifrons. Haplotype networks of nuclear loci with strong differentiation among subspecies support a close relationship between auduboni and nigrifrons, with considerable introgression from coronata. In three of the four informative loci (Figure 4.3B, C, D), the distribution of haplotypes in auduboni and nigrifrons is nearly identical, with coronata having substantially different haplotypes. In the fourth locus (Figure 4.3A), differentiation between coronata and nigrifrons is high, but auduboni contains haplotypes characteristic of both forms. These results are highly consistent with the AFLP data, which integrate patterns from a much larger number of markers. Clinal variation in different markers and traits also show congruence among some traits and incongruence among others. The frequency of southern-clade mtDNA has a sharp break near the Arizona/Utah border (Figure 4.4D), while the two nuclear markers CHD1Z and Numt-Dco1 show sharp changes in frequency at the hybrid zone between coronata and auduboni (Figure 4.4H, I). The population mean of AFLP dimension 2 (which incorporates the bulk of differentiation among the three northern subspecies) shows a small but abrupt change across the coronata/auduboni hybrid zone, and a gradual change across the range of auduboni (Figure 4.4G). Wing length (measured by Hubbard 1970) shows a similar pattern to AFLPs (Figure 4.4C), while throat colour (scored by Hubbard 1969, 1970) is similar to CHD1Z and Numt-Dco1 (Figure 4.4A). 4.4 DISCUSSION Taken together, our evidence indicates that the history of the taxa in the Yellowrumped Warbler complex cannot be represented as a simple bifurcating tree. The auduboni subspecies closely resembles its southern neighbour nigrifrons in appearance and nuclear genetic variation, but is not identical. The similarities decrease with distance from the U.S./Mexico border region, where the two forms come into contact. It shares with nigrifrons most of its male breeding-season plumage traits (including yellow throat, white wing patch, and lack of eye stripe) which are thought to be under sexual selection (Milá et al 2007). In contrast, over most of its range auduboni shares with coronata its 47  mtDNA clade and migratory behaviour. Additionally, auduboni appears to have incorporated AFLP alleles otherwise found only in coronata. In nuclear allele frequencies (Figures 4.2, 4.4) and wing length (Figure 4.4), northern auduboni are intermediate between coronata and nigrifrons, while southern auduboni closely resemble nigrifrons. The auduboni subspecies clearly represents an admixture between two long-divergent lineages, coronata and nigrifrons. What processes could have generated this pattern? One possibility is that the hybridization that gave rise to auduboni may have occurred during a previous interglacial period. Range shifts and fragmentation associated with glacial cycles could have caused geographic isolation between auduboni and one or both of its parental forms, with subsequent expansion from a refugium producing a widespread taxon of hybrid origin. Alternatively, auduboni may have formed at the beginning of the current interglacial period as a result of secondary contact and hybridization between coronata and nigrifrons. If this contact occurred in the south-western U.S., the coronata alleles (including mtDNA) found in auduboni would represent a neutral wake left behind by a hybrid zone that has moved northward (e.g. Krosby and Rohwer 2008). If instead the initial contact between coronata and nigrifrons occurred in the Canadian Rockies, the coronata alleles found in auduboni would have introgressed southward under positive selection. We find the latter possibility more plausible for two reasons: First, the transition between the two mitochondrial clades in Arizona and Utah occurs within less than 150 km, a distance similar to the width of the coronata/auduboni hybrid zone in Canada (132 km), which is maintained by selection (Brelsford and Irwin 2009). If the northern-clade mtDNA is neutral and was left behind after hybrid zone movement, it seems likely that more mixing would have occurred at the interface of the two clades. Second, the Canadian Rockies form a suture zone for many east-west pairs of boreal forest taxa (Swenson and Howard 2005, Toews and Irwin 2008, Irwin et al. 2009b), making it a likely location for secondary contact. A final possibility that does not involve secondary contact is that coronata and nigrifrons have independently diverged from auduboni, and that the intermediate characteristics of auduboni are the ancestral condition, rather than a result of admixture. However, we find this scenario unlikely, as we would expect to find distinctive allele 48  frequencies in all three widespread forms. Instead, we find distinctive allele frequencies only in coronata and nigrifrons, while at all loci auduboni has allele frequencies similar to coronata, similar to nigrifrons, or intermediate between the two. This suggests that only the auduboni form lacks a long history of independent evolution. The four taxa in this complex were originally described as distinct species (Linnaeus 1766, Townsend 1837, Brewster 1889, Nelson 1897). If auduboni truly represents an evolutionarily distinct form, it may be one of the best-documented cases thus far of the origin of a bird taxon through hybridization. However, the broad clines in phenotype and AFLPs between nigrifrons and auduboni call into question the distinctness of these two subspecies from one another. Further study of the contact zone between these forms is clearly warranted, as information is especially lacking from the area south of the US/Mexico border. We wish to draw a distinction between hybrid speciation and hybrid origins of a taxon (cf. Mavárez and Linares 2008): hybrid speciation requires that hybridization is directly responsible for reproductive isolation from both of the parental species (e.g. Mavárez et al. 2006, Rieseberg 2006). In the case of D. c. auduboni, reproductive isolation from nigrifrons has not yet been demonstrated, and if it exists, we have no evidence that gene flow from coronata is responsible for that isolation. Still, it is plausible that hybridization with coronata would be responsible for any reproductive barriers between auduboni and nigrifrons, since the only documented differences between these taxa are the presence of coronata-like traits (smaller size, less black coloration, more pointed wing, seasonal migration; Hubbard 1970, Milá et al. 2008) and alleles (mtDNA, Milá et al. 2007; AFLPs, this study) in auduboni. The geographically isolated fourth subspecies in the complex, D. c. goldmani, showed a much lower level of nuclear genetic diversity than in the other three forms, confirming this aspect of Milá et al.'s (2007) mtDNA results. The goldmani form has a small effective population size, restricted range, and long history of independent evolution, which suggests the need for increased conservation attention. This result also highlights the sensitivity of allele-frequency-based measures of genetic differentiation (i.e. most analyses of microsatellites and AFLPs) to differences in population size (Ehrich  49  et al. 2009), unlike measures based on DNA sequence divergence (Zink and Barrowclough 2008). We have shown that the coronata and nigrifrons subspecies diverged from each other long ago, and that the auduboni form represents an admixed population in which alleles from the two divergent parental types can assort independently, generating novel combinations. The two parental types remain distinct despite this admixture, providing additional evidence for the prevalence of speciation with gene flow (Nosil 2008). Previous research has demonstrated partial reproductive isolation between auduboni and coronata (Brelsford and Irwin 2009), and it is possible that some reproductive isolation exists between auduboni and nigrifrons, making this a potential case of hybrid speciation. Regardless, such a widespread taxon of hybrid origin has not yet been shown in birds, but others may exist. Such cases can best be recognized through surveys of many independent genetic markers, thus ongoing advances in genomic methods and greater attention to the possibility of hybrid speciation may reveal many more examples in birds and other animals.  50  Table 4.1: Divergence among the three widespread forms of Yellow-rumped Warbler at 12 nuclear loci, measured by FST and net sequence divergence (Da). Loci are arranged in ascending order of FST between coronata and nigrifrons. Locus RPL30 CEPU RPL5 GH1 13446 5087 9300 LHCGR KIAA1383 ARID5B CHD1Z Numt-Dco1  FST cor/aud 0.017 0.044 0.061 -0.037 0.104 -0.014 0.081 0.009 0.491 0.235 0.623 0.631  FST aud/nig 0.007 0.036 0.039 0.049 -0.057 0.047 0.007 0.123 -0.027 0.155 0.053 0.039  FST cor/nig 0.013 0.049 0.053 0.080 0.099 0.100 0.107 0.143 0.472 0.530 0.737 0.783  Da cor/aud 0.00026 0.00025 0.00049 -0.00026 0.00040 -0.00009 0.00028 0.00012 0.00281 0.00109 0.00104 0.00318  Da aud/nig 0.00014 0.00022 0.00027 0.00021 -0.00017 0.00031 0.00002 0.00131 -0.00006 0.00108 0.00001 0.00011  Da cor/nig 0.00024 0.00030 0.00026 0.00046 0.00038 0.00070 0.00039 0.00174 0.00254 0.00361 0.00132 0.00312  51  Table 4.2: AFLP-based differentiation among Yellow-rumped Warbler forms coronata, auduboni north clade, auduboni south clade, nigrifrons, and goldmani. Distances above diagonal are pairwise FST values calculated by AFLP-SURV (Vekemans 2002); values in bold are significant at the p<0.05 level. On the diagonal are average within-population Jaccard distances, and below the diagonal are average between-population Jaccard distances. cor aud N aud S nig gol  cor 0.5358 0.5690 0.5789 0.5868 0.6197  aud N 0.0289 0.5328 0.5485 0.5552 0.6074  aud S 0.0333 0.0000 0.4941 0.5513 0.5850  nig 0.0577 0.0127 0.0007 0.5201 0.6090  gol 0.2264 0.2095 0.1575 0.2058 0.3028  52  Figure 4.1. Range map, sampling locations, and groupings based on AFLPs, mtDNA, and phenotype for four Yellow-rumped Warbler subspecies. Both mtDNA and AFLPs show three clusters in the species complex, corresponding to coronata, nigrifrons, and goldmani. In AFLPs, auduboni groups with the nigrifrons cluster, while in mtDNA, auduboni from Arizona locations group with nigrifrons and other auduboni group with coronata. The coronata form has a facial pattern distinct from the other three forms, which differ from each other mainly in the extent of black plumage. Sampling locations (A - O) were distributed throughout the range of the four currently recognized subspecies.  53  Figure 4.2. Multidimensional scaling plot of AFLP variation. Dimension 1 separates the disjunct goldmani subspecies from the other three more widespread forms, which are arrayed along dimension 2. Northern populations of auduboni are intermediate between nigrifrons and coronata, while southern auduboni (sampled in Arizona) are closer to nigrifrons.  54  Figure 4.3. Haplotype networks for 5 loci that differ between coronata and nigrifrons. Northern populations of auduboni group with nigrifrons in three loci (B, C, D) coronata in one (E), and share common haplotypes with both divergent groups in one (A).  55  56  Figure 4.4. Geographic variation in traits and genetic markers that differ between nigrifrons and coronata demonstrate incongruent patterns in the putative hybrid auduboni. The auduboni form resembles nigrifrons in one trait (A) and three nuclear markers (E, H, I). In contrast, in one trait (C), one nuclear marker (E), and multilocus AFLP score (G) auduboni is intermediate between nigrifrons and coronata, with gradual change across its range. The sharp phylogeographic break between northern and southern mtDNA lineages (D) occurs within the range of auduboni, and is not accompanied by an abrupt cline in phenotype or any genetic marker. Throat colour and wing length (A, C) were taken from Hubbard (1969, 1970). “X” symbols denote sample locations near the Pacific coast (California, Oregon, Washington, western British Columbia, Yukon, and Alaska); other sites are represented by diamonds.  57  CHAPTER 5: GENERAL CONCLUSIONS Hybridization is common in birds (McCarthy 2006) and other taxa (Arnold 1997). This study provides a close examination of a hybrid zone that may be representative of many others. The results point to the importance of postmating isolation in maintaining the stability of the zone, and the surprisingly small role of song and plumage differences. Overall, hybridization in this complex has not caused an erosion of biodiversity, but has instead played a creative role. In chapter two, I showed that the hybrid zone has remained stable in location and width over the past 40 years, based on Hubbard’s (1969) plumage scores collected in 1965 and my own from 2005-2007. Assortative mating was absent or very weak, allowing me to use a tension zone model (Barton 1982, Szymura and Barton 1986, Barton and Gale 1993) to investigate the balance between selection and dispersal that maintains the hybrid zone. Linkage disequilibrium between two physically unlinked SNP markers was significant (0.067 out of a maximum of 0.25). Combined with the 132 km measured width of the hybrid zone, this produced estimates of per-generation dispersal (20 km/gen) and strength of selection (0.18) required to maintain the hybrid zone. Selection against hybrids can take many forms. One possible mechanism is behavioural sterility, in which the abnormal or intermediate behaviour of hybrids causes reduced mating success. In chapter 3, I studied variation in territorial song and male song response across the hybrid zone, with the goal of determining whether hybrid songs are less effective signals than Myrtle or Audubon’s songs. I found significant but slight differences in song between Myrtle, Audubon’s and hybrid warblers. Both parental types responded just as strongly to hybrid songs as to songs of their own type. The results of chapter three rule out one possible mechanism of selection against hybrids, and add an independent line of evidence suggesting that premating isolation is of little importance in maintaining boundaries between these incipient species. This result is unexpected based on the conclusions of Milá et al. (2007). That study found very low divergence in mitochondrial DNA between Myrtle and Audubon’s Warblers, and suggested that the distinct male breeding-season plumages of the two forms had evolved in the last 20,000 years, probably under strong sexual selection. If that 58  were the case, presumably females in the hybrid zone would not mate randomly with respect to plumage. In chapter four, I resolved this discrepancy by showing that the Audubon’s Warbler originated through hybridization between the Myrtle and Blackfronted Warblers, which share a common ancestor roughly 1.5 million years ago. In the nuclear genome, the Audubon’s is most closely related to Black-fronted, although it has acquired a substantial fraction of its nuclear genome from Myrtle, and Myrtle mitochondrial DNA has replaced the original Audubon’s mtDNA over most of its range, possibly through a selective sweep. My research answers several questions that have been debated since the discovery of hybridization between Myrtle and Audubon’s Warblers, but also raises many new questions. First, what are the mechanisms of partial reproductive isolation between the two forms? Second, what is the genetic architecture of species differences? Third, why have distinct male breeding-season plumages evolved given the lack of discrimination among them by females? Finally, how has selection shaped the combination of traits and alleles that the Audubon’s Warbler acquired from its two parental species? Two possible mechanisms of reproductive isolation, premating isolation and inability of hybrids to obtain mates, have been ruled out (Chapter 2, Chapter 3). Remaining alternatives include extrinsic ecological selection against hybrids and intrinsic genetic incompatibilities. One possible mechanism of ecological selection against hybrids is specific to seasonally migratory animals: hybrids between two species with different migration routes may have intermediate or otherwise maladapted migratory behaviours (Helbig 1991, Irwin and Irwin 2005). Myrtle and Audubon’s Warblers differ in migratory behaviour, but there is some overlap in both migratory routes and wintering locations (Hunt and Flaspohler 1998); thus, it is unclear whether intermediate migratory behaviour would result in strong selection against hybrids. Another type of ecological selection against hybrids would result if they are intermediate in one or more ecologically relevant traits, which could restrict them to a narrow zone of intermediate habitat. This mechanism would define the hybrid zone as an ecotonal hybrid zone rather than a tension zone. The width of an ecotone hybrid zone depends partly on the width of the transition between habitat types favouring each parental species: a wider ecotone would result in a wider hybrid zone (Endler 1977). The 59  Myrtle-Audubon’s hybrid zone coincides with the Rocky Mountains over most of its length, and it seems reasonable to expect that any ecological differences involved in maintaining the hybrid zone would be associated with the east and west slopes of the mountain range. If this were the case, the hybrid zone would be expected to change in width in parallel with changes in width of the associated mountain range. Instead, I found that the hybrid zone width was consistent among five transects, despite the considerable difference in the width of the Rockies among the transects (Chapter 2). Finally, intrinsic genetic incompatibilities between the two forms may reduce the fertility or viability of hybrids. These incompatibilities evolve slowly in birds (Price and Bouvier 2002), and would be unlikely to have arisen since the 20,000 year divergence date estimated by Milá et al. (2007). However, the much longer divergence date between Myrtle and Black-fronted Warblers (1.5 million years, Milá et al. 2007) may be sufficient for some incompatibilities to evolve. My finding that the Audubon’s Warbler shares much of its nuclear genome with the Black-fronted Warbler implies that genetic incompatibilities are a plausible mechanism for the stability of the hybrid zone. Incompatibilities between the Myrtle and Black-fronted genomes would not preclude the formation of a hybrid species; these barriers exist between the parents of several wellstudied hybrid species including Helianthus sunflowers (Rieseberg et al. 1999) and Heliconius butterflies (Jiggins et al. 2008). Little data is available to discriminate between these non-exclusive hypotheses (migratory divide, adaptation to different breeding habitats, and genetic incompatibilities), but for reasons discussed above I suggest that genetic incompatibilities are likely to be more important than ecological or migratory differences. To test this hypothesis, it would be useful to look for direct evidence of genetic incompatibilities between Myrtle and Audubon’s Warblers. The earliest such incompatibilities to evolve should cause a reduction in fertility of hybrids, and according to Haldane’s rule should affect females more than males (Price and Bouvier 2002). A study of nests in the hybrid zone could determine whether hybrid females have smaller clutch sizes or lower hatching rates than Myrtle or Audubon’s females. Because this thesis is the first study of the Yellow-rumped Warbler hybrid zone to use DNA-based genetic markers, only limited conclusions can be drawn about the genetic 60  architecture of species differences. I was not able to infer the proportion of the genome that remains isolated between the two warblers. It is possible that the loci that cause the plumage differences and the two diagnostic markers analysed in chapter 2 are broadly representative of the genome, but we cannot rule out the possibility that strong selection on an unknown subset of the genome maintains the hybrid zone, and that relatively free gene flow occurs at other loci. The fraction of the genome that does not cross the hybrid zone must be substantial, since at least two of the 11 sequenced loci fall within it. Some of the other nine loci may also be subject to restricted introgression; the lack of fixed differences we observed could result from either incomplete lineage sorting or ongoing gene flow. Approaches that are often used to separate these effects, such as fitting an isolation-with-migration model (Hey and Nielsen 2007, Becquet and Przeworski 2009), may be problematic if one of the species in the analysis originated through hybridization. Other studies of avian hybrid zones (Sattler and Braun 2000, Sætre et al. 2003, Carling and Brumfield 2008) have found a large effect of the Z chromosome, consistent with the large X-effect observed in other taxa (Coyne and Orr 2004, Price 2008). I found fixed differences at one of five Z-linked and one of six autosomal loci, which does not on its face provide evidence of a large Z-effect in the Yellow-rumped Warblers. However, the number of sampled loci is quite small, and advances in high-throughput sequencing will make a more comprehensive genomic approach feasible in the near future, using the samples I collected. Genotyping a large number of markers in a large number of individuals across the hybrid zone (perhaps using the “RAD-tags” approach of Baird et al. 2008) will give a more complete view of whole-genome patterns of gene flow and differences between sex-linked and autosomal loci, as well as enabling the detection of outlier loci putatively under selection. Another question raised by this thesis is what drove the evolution of distinct male breeding-season plumage in the four Yellow-rumped Warbler taxa. The much less obvious differences in female and non-breeding appearances suggest some role for sexual selection. The lack of assortative mating between Myrtle and Audubon’s Warblers seems to indicate that female preferences are not strongly divergent. However, random mating and evolution driven by female choice are not completely incompatible: a lack of female discrimination against hybrid phenotypes does not necessarily imply a lack of 61  discrimination against the phenotype of the common ancestor of all Yellow-rumped Warblers. Alternatively, the divergent plumages of Myrtle and Audubon’s Warblers may have evolved as signals in male-male competition. Plumage signals have been shown to affect competition for territories in Yellow-browed Leaf Warblers Phylloscopus inornatus (Marchetti 1998) and Golden-winged Warblers Vermivora chrysoptera (Leichty and Grier 2006), but this possibility remains unexplored in Yellow-rumped Warblers. I established in chapter four that the Audubon’s Warbler originated through hybridization between the Myrtle and Black-fronted Warblers, but noted that extensive gene flow between Audubon’s and Black-fronted forms could not be ruled out. Broad clines in morphometrics and nuclear AFLP markers, along with narrow but noncoincident clines in plumage colour and mtDNA, call into question the validity of these two forms as distinct taxa. However, this study should be viewed as only a preliminary examination of the Audubon’s/Black-fronted hybrid zone: only one transect was sampled, and sample sizes were sufficient only to characterize broad-scale differences among the four Yellow-rumped Warbler taxa, and not for fine-scale analysis of the two hybrid zones in the complex. Further work is clearly necessary, with greater sampling of individuals, multiple transects, and more genetic markers. One aspect of the Audubon’s/Black-fronted hybrid zone that is particularly intriguing is the narrow contact zone between the two most divergent mtDNA clades in the complex. The mitochondrial cline has a width of the same order as the Myrtle/Audubon’s hybrid zone, and therefore is probably maintained by selection, either on loci in linkage disequilibrium with mtDNA or on the mitochondrial genome itself. I view the latter hypothesis as more likely, based on the absence of strong phenotypic or nuclear AFLP differences across the mitochondrial break. The boundary between mitochondrial clades coincides with the northern limit of the Audubon’s winter range in the Rockies, which may indicate a boundary between seasonally migratory and yearround resident populations. One can speculate that long-distance migrant and year-round resident life histories would place different selective demands on metabolism and mitochondrial function. Research into the migratory behaviour and mitochondrial function of Audubon’s Warblers in the south-western U.S. may provide insight into some 62  of the metabolic adaptations required for long-distance migration as well as the role of introgression in providing adaptive genetic variation. The Yellow-rumped Warbler complex has been the subject of considerable taxonomic controversy over the past century, with particular focus on whether the Myrtle and Audubon’s Warbler are conspecific. This thesis informs the debate by showing that the hybrid zone between them has remained stable over the past four decades despite the lack of assortative mating. As a result, the International Ornithological Congress has elevated these subspecies to full species (Gill and Donsker 2010), and the American Ornithologists’ Union is considering a similar proposal. Hybridizing taxa present a conundrum for taxonomists under many species concepts. The key criterion for the biological species concept is reproductive isolation, which in a strict sense is only complete when hybridization is absent or hybrids are completely infertile. However, many species occasionally hybridize without losing their distinct identity, and even the originators of the biological species concept allowed for some gene flow between taxa they considered species (Dobzhansky 1951, Mayr 1963). If a hybrid zone between two distinct taxa is stable over time and narrow relative to dispersal, this signifies that the taxa are on separate evolutionary trajectories and some form of selection maintains the differences between them. For this reason, taxa connected by a narrow, stable hybrid zone are considered species by the American Ornithologists’ Union (AOU 1998), although the British Ornithological Society treats such taxa as semispecies (Helbig 2002). My findings corroborate those of Milá et al. (2007) that the Goldman’s Warbler has a long history of independent evolution and a very small effective population size, as shown by its high divergence from its closest relative and very low diversity in both mitochondrial (Milá et al. 2007) and nuclear (chapter four) DNA. The conservation status of the Yellow-rumped Warbler is currently “Least concern” (IUCN 2010), but the Goldman’s Warbler is clearly an evolutionarily significant unit and possibly a distinct full species. Its range consists of a small area of highland coniferous forest in southern Mexico and Guatemala that is highly threatened by deforestation in both countries (Cayuela et al. 2006, Andersen et al. 2006). Thus, it should be treated separately from the abundant and widespread Myrtle and Audubon’s forms for conservation purposes.  63  This thesis provides a detailed examination of hybridization and speciation in a widespread, abundant species complex. While song and premating isolation are implicated in many well-studied examples of bird speciation, neither factor is important in maintaining the species differences between Myrtle and Audubon’s Warblers. I have shown that hybridization is not leading to the erosion of evolutionary distinctiveness in this complex, but instead has led to the formation of a new lineage. My research resolves a long-standing controversy over the extent of reproductive isolation between Myrtle and Audubon’s Warblers by showing the temporal stability of the hybrid zone, and documents for the first time a hybrid zone between Audubon’s and Black-fronted Warblers. Finally, this study provides a point of comparison for numerous other east-west species pairs that diverged during the Pleistocene glaciations (e.g. Weir and Schluter 2004, Ruegg 2008, Irwin et al. 2009b, Rush et al. 2009). Examination of many pairs of species or subspecies that share the same biogeographic history will show whether the factors that have led to speciation in Yellow-rumped Warblers are part of a general pattern in birds of the boreal forest. Complete reproductive isolation rarely arises within the time of a single glacial cycle (Price 2008), so it is likely that episodes of secondary contact and hybridization have occurred during the speciation process in many of these species.  64  References: Alexander, G. 1945. Natural hybrids between Dendroica coronata and D. auduboni. Auk 62:623–626. Alexandrino, J., S. J. E. Baird, L. Lawson, J. R. Macey, C. Moritz, and D. B. Wake. 2005. Strong selection against hybrids at a hybrid zone in the Ensatina ring species complex and its evolutionary implications. Evolution 59:1334–1347. American Ornithologists’ Union. 1973. Thirty-second supplement to the American Ornithologists’ Union (AOU) check-list of North American birds. Auk 90:411– 419. –––––. 1998. Check-list of North American Birds, 7th ed. American Ornithologists' Union, Washington. Andersen, U.S., J. P. Prado-Córdova, M. Sørensen, and J. Kollmann. 2006. Conservation and utilization of Abies guatemalensis Rehder (Pinaceae) – an endangered endemic conifer in Central America. Biodivers. Conserv. 15:3131–3151. Anderson, E. 1949. Introgressive Hybridization. Wiley, New York. Anderson, T. M., B. M. vonHoldt, S. I. Candille, M. Musiani, C. Greco et al. 2009. Molecular and evolutionary history of melanism in North American gray wolves. Science 323:1339–1343. Angert, A. L., and D. W. Schemske. 2005. The evolution of species’ distributions: reciprocal transplants across the elevation ranges of Mimulus cardinalis and M. lewisii. Evolution 59:1671–1684. Arbogast, B. S., S. V. Edwards, J. Wakeley, P. Beerli, and J. B. Slowinski. 2002. Estimating divergence times from molecular data on phylogenetic and population genetic timescales. Annu. Rev. Ecol. Syst. 33:707–740. Arnold, M. L. 1997. Natural Hybridization and Evolution. Oxford Univ. Press, New York. Avise, J. C. , and D. Walker. 1998. Pleistocene phylogeographic effects on avian populations and the speciation process. Proc. R. Soc. Lond. B 265:457–463. Bachtrog, D., K. Thornton, A. Clark, and P. Andolfatto. 2006. Extensive introgression of mitochondrial DNA relative to nuclear genes in the Drosophila yakuba species group. Evolution 60:292–302. 65  Backström, N., M. Brandström, L. Gustafsson, A. Qvarnström, H. Cheng, and H. Ellegren. 2006. Genetic mapping in a natural population of collared flycatchers (Ficedula albicollis): conserved synteny but gene order rearrangements on the avian Z chromosome. Genetics 174:377–386. Backström, N., S. Fagerberg, and H. Ellegren. 2008. Genomics of natural bird populations: a gene-based set of reference markers evenly spread across the avian genome. Mol. Ecol. 17:964–980. Baird, N. A., P. D. Etter, T. S. Atwood, M. C. Currey, A. L. Shiver, Z. A. Lewis, E. U. Selker, W. A. Cresko, and E. A. Johnson. 2008. Rapid SNP discovery and genetic mapping using sequenced RAD markers. PLoS ONE 3:e3376. Baker, M. C. 1991. Response to male indigo and lazuli buntings and their hybrids to song playback in allopatric and sympatric populations. Behaviour 119:225–242. Balakrishnan, C. N., and M. D. Sorenson. 2006. Song discrimination suggests premating isolation among sympatric indigobird species and host races. Behav. Ecol. 17:473–478. Ballard, J. W. O., and M. C. Whitlock. 2004. The incomplete natural history of mitochondria. Mol. Ecol. 13:729–744. Barrowclough, G. F. 1980. Genetic and phenotypic differentiation in a wood warbler (genus Dendroica) hybrid zone. Auk 97:655–668. Barton, N. H. 1979. The dynamics of hybrid zones. Heredity 43:341–359. –––––. 1982. The structure of the hybrid zone in Uroderma bilobatum (Chiroptera: Phyllostomidae). Evolution 36:863–866. –––––. 1983. Multilocus clines. Evolution 37:454–471. Barton N. H., and B. O. Bengtsson. 1986. The barrier to genetic exchange between hybridising populations. Heredity 56:357–376. Barton, N. H., and K. S. Gale. 1993. Genetic analysis of hybrid zones. Pp. 13–45 in R. G. Harrison, ed. Hybrid zones and the evolutionary process. Oxford Univ. Press, New York. Barton, N. H., and G. M. Hewitt. 1989. Adaptation, speciation and hybrid zones. Nature 341:497–503.  66  Bazin, E., S. Glemin, and N. Galtier. 2006. Population size does not influence mitochondrial genetic diversity in animals. Science 312:570–572. Bazykin, A. D. 1969. Hypothetical mechanism of speciation. Evolution 23:685–687. Becquet, C., and M. Przeworski. 2009. Learning about modes of speciation by computational approaches. Evolution 63:2547–2562. Brelsford, A., and D. E. Irwin. 2009. Incipient speciation despite little assortative mating: the Yellow-rumped Warbler hybrid zone. Evolution 63:3050–3060. Brewster, W. 1889. Descriptions of supposed new birds from western North America and Mexico. Auk 6:85–98. Bridle, J. R., C. I. Saldamando, W. Koning, and R. K. Butlin. 2006. Assortative preferences and discrimination by females against hybrid male song in the grasshoppers Chorthippus brunneus and Chorthippus jacobsi (Orthoptera: Acrididae). J. Evol. Biol. 19:1248–1256. Bronson, C. L., T. C. Grubb Jr., G. D. Sattler, and M. J. Braun. 2005. Reproductive success across the Black-capped Chickadee (Poecile atricapillus) and Carolina Chickadee (P. carolinensis) hybrid zone in Ohio. Auk 122:759–772. Brugmans, B., R. G. van der Hulst, R. G. F. Visser, P. Lindhout, and H. J. van Eck. 2003. A new and versatile method for the successful conversion of AFLP markers into simple single locus markers. Nucleic Acids Res. 31:e55. Borge, T., M. T. Webster, G. Andersson, and G.-P. Sætre. 2005. Contrasting patterns of polymorphism and divergence on the Z chromosome and autosomes in two Ficedula flycatcher species. Genetics 171:1861–1873. Carling, M. D., and R. T. Brumfield. 2008. Haldane’s rule in an avian system: using cline theory and divergence population genetics to test for differential introgression of mitochondrial, autosomal and sex-linked loci across the Passerina bunting hybrid zone. Evolution 62:2600–2615. Cayuela, L., Golicher, J. D., and Rey Benayas, J. M. 2006. The extent, distribution, and fragmentation of montane cloud forest in the highlands of Chiapas, Mexico. Biotropica 38:544–554.  67  Cheviron, Z. A., and R. T. Brumfield. 2009. Migration-selection balance and local adaptation of mitochondrial haplotypes in Rufous-collared Sparrows (Zonotrichia capensis) along an elevational gradient. Evolution 63:1593–1605. Clement, M., D. Posada, and K. A. Crandall. 2000. TCS: a computer program to estimate gene genealogies. Mol. Ecol. 9:1657–1660. Coyne, J. A., and H. A. Orr. 1989. Patterns of speciation in Drosophila. Evolution 43:362–381. –––––. 2004. Speciation. Sinauer Associates, Sunderland. DeMarais, B. D., T. E. Dowling, M. E. Douglas, W. L. Minckley, and P. C. Marsh. 1992. Origin of Gila seminuda (Teleostei: Cyprinidae) through introgressive hybridization: implications for evolution and conservation. Proc. Natl. Acad. Sci. USA 89:2747–2751. den Hartog P. M., S. R. de Kort, and C. ten Cate. 2007. Hybrid vocalizations are effective within, but not outside, an avian hybrid zone. Behav. Ecol. 18:608–614. Dobzhansky, T. 1940. Speciation as a stage in evolutionary divergence. Am. Nat. 74: 312–321. –––––. 1951. Genetics and the Origin of Species. 3rd edition. Columbia Univ. Press, New York. Dunn, J. L., and K. L. Garrett. 1997. A Field Guide to Warblers of North America. Houghton Mifflin, New York. Durrett, R., L. Buttel, and R. Harrison. 2000. Spatial models for hybrid zones. Heredity 84:9–19. Edwards, S. V., S. B. Kingan, J. D. Calkins, C. N. Balakrishnan, W. B. Jennings, W. J. Swanson, and M. D. Sorenson. 2005. Speciation in birds: Genes, geography and sexual selection. Proc. Natl. Acad. Sci. USA 102:6550–6557. Ehrich, D., P. B. Eidesen, I. G. Alsos, and C. Brochmann. 2009. An AFLP clock for absolute dating of shallow-time evolutionary history – too good to be true? Mol. Ecol. 18:4526–4532. Endler, J. A. 1977. Geographic variation, speciation, and clines. Princeton Univ. Press, Princeton.  68  Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol. Ecol. 14:2611–2620. Fel-Clair, F., T. Lenormand, J. Catalan, J. Grobert, A. Orth, P. Boursot, M. C. Viroux, and J. Britton-Davidian. 1996. Genomic incompatibilities in the hybrid zone between house mice in Denmark: evidence from steep and non-coincident chromosomal clines for Robertsonian fusions. Genet. Res. 67:123–134. Fridolfsson, A. K., and H. Ellegren. 1999. A simple and universal method for molecular sexing of non-ratite birds. J. Avian Biol. 30:116 –121. Gay, L., G. Neubauer, M. Zagalska-Neubauer, C. Debain, J.-M. Pons, P. David, and P.A. Crochet. 2007. Molecular and morphological patterns of introgression between two large white-headed gull species in a zone of recent secondary contact. Mol. Ecol. 16:3215–3227. Gay, L., P.-A. Crochet, D. A. Bell, and T. Lenormand. 2008. Comparing clines on molecular and phenotypic traits in hybrid zones: A window on tension zone models. Evolution 62:2789-2806. Gill, F., and D. Donsker, eds. 2010. IOC World Bird Names (version 2.4). Gompert, Z., J. A. Fordyce, M. L. Forister, A. M. Shapiro, and C. C. Nice. 2006. Homoploid hybrid speciation in an extreme habitat. Science. 314:1923–1925. Good, J. M., S. Hird, N. Reid, J. R. Demboski, S. J. Steppan, T. R. Martin-Nims, and J. Sullivan. 2008. Ancient hybridization and mitochondrial capture between two species of chipmunks. Mol. Ecol. 17:1313–1327. Grant, B. R., and P. R. Grant. 2002a. Lack of premating isolation at the base of a phylogenetic tree. American Naturalist 160:1–19. –––––. 2002b. Simulating secondary contact in allopatric speciation: an empirical test of premating isolation. Biol. J. Linn. Soc. 76:545–556. Grant, P. R., and B. R. Grant. 1997. Hybridization, sexual imprinting, and mate choice. Am. Nat. 149:1–28 –––––. 2009. The secondary contact phase of allopatric speciation in Darwin’s finches. Proc. Natl. Acad. Sci. USA 106:20141-20148.  69  –––––. 2010. Conspecific versus heterospecific gene exchange between populations of Darwin's finches. Philos. Trans. R. Soc. Lond. B 365:1065–1076. Griffith, S. C., I. P. F. Owens, and K. A. Thuman. 2002. Extra pair paternity in birds: a review of interspecific variation and adaptive function. Mol. Ecol. 11:2195–2212. Hall, T. A. 1999. BioEdit: a user friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41:95–98. Harrison, R. G. 1993. Hybrids and hybrid zones: historical perspective. Pp. 3–12 in R. G. Harrison, ed. Hybrid zones and the evolutionary process. Oxford Univ. Press, New York. Helbig, A. J. 1991. Inheritance of migratory direction in a bird species: a cross-breeding experiment with SE- and SW- migrating blackcaps (Sylvia atricapilla). Behav. Ecol. Sociobiol. 28:9–12. Helbig, A. J., A. G. Knox, D. T. Parkin, G. Sangster, and M. Collinson. 2002. Guidelines for assigning species rank. Ibis 144:518-525. Hewitt, G. 1988. Hybrid zones – natural laboratories for evolution studies. Tr. Ecol. Evol. 3:158–166. ———. 2004. Genetic consequences of climatic oscillations in the Quaternary. Philos. Trans. R. Soc. Lond. B 359:183–195. Hey, J., and R. Nielsen 2007. Integration within the Felsenstein equation for improved Markov chain Monte Carlo methods in population genetics. Proc. Natl. Acad. Sci. USA 104:2785–2790. Hubbard, J. P. 1969. The relationships and evolution of the Dendroica coronata complex. Auk 86:393–432. ———. 1970. Geographic variation in the Dendroica coronata complex. Wilson Bull. 82:355–369. Hubisz, M.J., D. Falush, M. Stephens, and J. K. Pritchard. 2009. Inferring weak population structure with the assistance of sample group information. Mol. Ecol. Resour. 9:1322–1332 . Hudson, R. D., and J. A. Coyne. 2002. Mathematical consequences of the genealogical species concept. Evolution 56:1557–1565.  70  Hunt, P. D., and D. J. Flaspohler. 1998. Yellow-rumped Warbler (Dendroica coronata). No. 376 in A. Poole and F. Gill, eds. The Birds of North America. The Academy of Natural Sciences, Philadelphia. Irwin, D. E. 2000. Song variation in an avian ring species. Evolution 54:998–1010. –––––. 2002. Phylogeographic breaks without barriers to gene flow. Evolution 53:2383– 2394. Irwin, D. E., and J. H. Irwin. 2005. Siberian migratory divides – the role of seasonal migration in speciation. Pp. 27–40 in R. Greenberg and P. Marra, eds. Birds of two worlds – the ecology and evolution of migration. Johns Hopkins Univ. Press, Baltimore. Irwin, D. E., and T. Price. 1999. Sexual imprinting, learning and speciation. Heredity 82:347–354. Irwin, D. E., S. Bensch, and T. D. Price. 2001. Speciation in a ring. Nature 409:333–337. Irwin, D. E., M. P. Thimgan, and J. H. Irwin. 2008. Call divergence is correlated with geographic and genetic distance in greenish warblers (Phylloscopus trochiloides): a strong role for stochasticity in signal evolution? J. Evol. Biol. 21:435–448. Irwin, D. E., A. S. Rubtsov, and E. N. Panov. 2009a. Mitochondrial introgression and replacement between yellowhammers (Emberiza citrinella) and pine buntings (E. leucocephalos; Aves, Passeriformes). Biol. J. Linn. Soc. 98:422-438. Irwin, D. E., A. Brelsford, D. P. L. Toews, C. MacDonald, and M. Phinney. 2009b. Extensive hybridization in a contact zone between MacGillivray's and mourning warblers (Oporornis tolmiei and O. philadelphia) detected using molecular and morphometric analyses. J. Avian Biol. 40:539-552. IUCN 2010. IUCN Red List of Threatened Species. Version 2010.1. Jiggins, C. D., and J. Mallet. 2000. Bimodal hybrid zones and speciation. Tr. Ecol. Evol. 15:250–255. Jiggins, C. D., C. Salazar, M. Linares, and J. Mavárez. 2008. Hybrid speciation in Heliconius butterflies. Philos. Trans. R. Soc. Lond. B 363:3047–3054.  71  Johnson, N. K., J. V. Remsen Jr., and C. Cicero. 1999. Resolution of the debate over species concepts in ornithology: a new comprehensive biologic species concept. Pp 1470–1482 in N. J. Adams and R. H. Slotow, eds. Proc. 22 Int. Ornithol. Congr., Durban. BirdLife South Africa, Johannesburg. Johnston, R. F. 1969. Taxonomy of House Sparrows and their allies in the Mediterranean Basin. Condor 71:129–139. Jones, F. C., C. Brown, J. M. Pemberton, and V. A. Braithwaite. 2006. Reproductive isolation in a threespine stickleback hybrid zone. J. Evol. Biol. 19:1531–1544. Joseph, L., G. Dolman, S. Donnellan, K. M. Saint, M. L. Berg, and A. T. D. Bennett. 2008. Where and when does a ring start and end? Testing the ring-species hypothesis in a species complex of Australian parrots. Proc. R. Soc. Lond. B 275:2431–2440. Key, K. H. L. 1968. The concept of stasipatric speciation. Syst. Zool. 17:14–22. Kingston, S. E., and P. E. Rosel. 2004. Genetic differentiation among recently diverged delphinid taxa determined using AFLP markers. J. Hered. 95:1–10. Klicka, J. and R. M. Zink. 1997. The importance of recent ice ages in speciation: a failed paradigm. Science 277:1666–1669. ———. 1999. Pleistocene effects on North American songbird evolution. Proc. R. Soc. Lond. B 266:695–700. Krosby, M., and S. Rohwer. 2009. A 2000 km genetic wake yields evidence for northern glacial refugia and hybrid zone movement in a pair of songbirds. Proc. R. Soc. Lond. B 276:615–621. Kruuk, L. E. B., S. J. E. Baird, K. S. Gale, and N. H. Barton. 1999. A comparison of multilocus clines maintained by environmental adaptation or by selection against hybrids. Genetics 153:1959–1971. Lachlan, R. F., and M. R. Servedio. 2004. Song learning accelerates allopatric speciation. Evolution 58:2049–2063. Leichty, E. R., and J. W. Grier. 2006. Importance of facial pattern to sexual selection in golden-winged warblers (Vermivora chrysoptera). Auk 123:962–966 Linnaeus, C. 1766. Systema Naturæ. 12th ed. Laurantius Salvius. Stockholm.  72  Llopart, A., D. Lachaise, and J. Coyne. 2005. Multilocus analysis of introgression between two sympatric sister species of Drosophila: Drosophila yakuba and D. santomea. Genetics 171:197–210. Lovette, I. J. 2005. Glacial cycles and the tempo of avian speciation. Trends Ecol. Evol. 20:57–59. Macholán, M., P. Munclinger, M. Sugerkova, P. Dufkova, B. Bimova, J. Zima, and J. Piálek. 2007. Genetic analysis of autosomal and X-linked markers across a mouse hybrid zone. Evolution 61:746–771. Mallet, J. 1986. Hybrid zones in Heliconius butterflies in Panama, and the stability and movement of warning colour clines. Heredity 56:191–202. Mallet, J., N. H. Barton, G. M. Lamas, J. C. Santisteban, M. M. Muedas, and H. Eeley. 1990. Estimates of selection and gene flow from measures of cline width and linkage disequilibrium in Heliconius hybrid zones. Genetics 124:921–936. Marchetti, K. 1998. The evolution of multiple male traits in the yellow-browed leaf warbler. Anim. Behav. 55:361–376. Mavárez, J., and M. Linares. 2008. Homoploid hybrid speciation in animals. Mol. Ecol. 17:4181–4185. Mavárez, J., C. A. Salazar, E. Bermingham, C. Salcedo, and C. D. Jiggins. 2006. Speciation by hybridization in Heliconius butterflies. Nature 441:868–871. May, R. M., J. A. Endler, and R. E. McMurtie. 1975. Gene frequency clines in the presence of selection opposed by gene flow. Am. Nat. 109:659–676. Mayr, E. 1942. Systematics and the Origin of Species. Columbia Univ. Press, New York. –––––. 1963. Animal Species and Evolution. Harvard Univ. Press, Cambridge. –––––. 2001. Wu's genic view of speciation. J. Evol. Biol. 14:866–867. McCarthy, E. M. 2006. Handbook of avian hybrids of the world. Oxford Univ. Press, Oxford, U.K. Meyer, A., W. Salzburger, and M. Schartl. 2006. Hybrid origin of a swordtail species (Teleostei: Xiphophorus clemenciae) driven by sexual selection. Mol. Ecol. 15:721–730.  73  Miglia, K., E. McArthur, W. Moore, H. Wang, J. Graham, and D. Freeman. 2005. Nineyear reciprocal transplant experiment in the gardens of the basin and mountain big sagebrush (Artemisia tridentata: Asteraceae) hybrid zone of Salt Creek Canyon: the importance of multiple-year tracking of fitness. Biol. J. Linn. Soc. 86:213– 225. Milá, B., T. B. Smith, and R. K. Wayne. 2007. Speciation and rapid phenotypic differentiation in the yellow-rumped warbler (Dendroica coronata) complex. Mol. Ecol. 16:159–173. Milá, B., R. T. Wayne, and T. B. Smith. 2008. Ecomorphology of migratory and sedentary populations of the yellow-rumped warbler (Dendroica coronata). Condor 110:335–344. Moore, W. S. 1977. An evaluation of narrow hybrid zones in vertebrates. Q. Rev. Biol. 52:263–278. Moore, W. S., and R. A. Dolbeer. 1989. The use of banding recovery data to estimate dispersal rates and gene flow in avian species: case studies in the Red-winged Blackbird and Common Grackle. Condor 91:242–253. Moore, W. S., and J. T. Price. 1993. Nature of selection in the Northern Flicker hybrid zone and its implications for speciation theory. Pp. 196–225 in R. G. Harrison, ed. Hybrid zones and the evolutionary process. Oxford Univ. Press, New York. Navarro, A., and N. H. Barton. 2003. Accumulating postzygotic isolation genes in parapatry: a new twist on chromosomal speciation. Evolution 57:447–459. Nelson, E. W. 1897. Preliminary descriptions of new birds from Mexico and Guatemala in the collection of the United States Department of Agriculture. Auk 14:42–76. Nolte, A. W., J. Freyhof, K. C. Stemshorn, and D. Tautz. 2005. An invasive lineage of sculpins, Cottus sp. (Pisces, Teleostei) in the Rhine with new habitat adaptations has originated from hybridization between old phylogeographic groups. Proc. R. Soc. Lond. B 272:2379–2387. Nosil, P. 2008. Speciation with gene flow could be common. Mol. Ecol. 17:2103–2106. Oberholser, H. C. 1921. A revision of the races of Dendroica auduboni. Ohio. J. Sci. 21: 240–248.  74  Oksanen, L., R. Kindt, P. Legendre, B. O’Hara, G. L. Simpson, P. Solymos, M. Henry, H. Stevens, and H. Wagner. 2008. VEGAN: Community Ecology Package. R package version 1.15-1. Otto, S. P., and J. Whitton. 2000. Polyploid incidence and evolution. Ann. Rev. Genet. 34:401–437. Paradis, E., S. R. Baille, W. J. Sutherland, and R. D. Gregory. 1998. Patterns of natal and breeding dispersal in birds. J. Anim. Ecol. 67:518–536. Patten, M. A., J. T. Rotenberry, and M. Zuk. 2004. Habitat selection, acoustic adaptation, and the evolution of reproductive isolation. Evolution 58:2144–2155. Payne, R. B., K. Hustler, R. Sternstedt, K. M. Sefc, and M. D. Sorenson. 2002. Behavioural and genetic evidence of a recent population switch to a novel host species in brood parasitic indigobirds Vidua chalybeata. Ibis 144:373–383. Price, T. D. 2008. Speciation in Birds. Roberts and Co., Greenwood Village. Price, T. D., and M. M. Bouvier. 2002. The evolution of F1 postzygotic incompatibilities in birds. Evolution 56:2083–2089. Rand, A. L. 1948. Glaciation, an isolating factor in speciation. Evolution 2:314–321. Ramsey, J., H. D. Bradshaw, Jr., and D. W. Schemske. 2003. Components of reproductive isolation between the monkeyflowers Mimulus lewisii and M. cardinalis (Phrymaceae). Evolution 57:1520–1534. Ratcliffe, L. M., and P. R. Grant. 1983. Species recognition in Darwin’s finches (Geospiza, Gould). 1. Discrimination by morphological cues. Anim. Behav. 31:1139–1153. Raufaste, N., A. Orth, K. Belkhir, D. Senet, C. Smadja, S. J. E. Baird, F. Bonhomme, B. Dod, and P. Boursot. 2005. Inferences of selection and migration in the Danish house mouse hybrid zone. Biol. J. Linn. Soc. 84:593–616. Remington, C. L. 1968. Suture-zones of hybrid interaction between recently joined biotas. Pp. 321–428. in T. Dobzhansky, M. K. Hecht, and W. C. Steere, eds. Evolutionary biology. Plenum Press, New York. Rieseberg, L. H. 1997. Hybrid origins of plant species. Annu. Rev. Ecol. Syst. 28:359– 389. –––––. 2006. Hybrid speciation in wild sunflowers. Ann. Mo. Bot. Gard. 93:34–48. 75  Rieseberg L. H., and D. E. Soltis. 1991. Phylogenetic consequences of cytoplasmic gene flow in plants. Evol. Tr. Plants. 5:65–83. Rieseberg, L. H., J. Whitton, and K. Gardner. 1999. Hybrid zones and the genetic architecture of a barrier to gene flow between two sunflower species. Genetics 152:713–727. Rohwer, S. A. 1973. Significance of sympatry to behavior and evolution of Great Plains meadowlarks. Evolution 27:44–57. Rohwer, S., and C. Wood. 1998. Three hybrid zones between Hermit and Townsend's warblers in Washington and Oregon. Auk 115:284–310. Rozas, J., J. C. Sánchez-DelBarrio, X. Messeguer, and R. Rozas. 2003. DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19:2496–2497. Ruegg, K. 2008. Genetic, morphological and ecological characterization of a hybrid zone that spans a migratory divide. Evolution 62:452–466. Rundle, H. D., F. Breden, C. Griswold, A. Ø. Mooers, R. A. Vos and J. Whitton. 2001. Hybridization without guilt: Gene flow and the biological species concept. J. Evol. Biol. 14:868–869. Rush, A., R. C. Cannings, and D. E. Irwin. 2009. Analysis of multilocus DNA reveals hybridization in a contact zone between Empidonax flycatchers. J. Avian Biol. 40:614–624. Sætre, G. P., T. Borge, K. Lindroos, J. Haavie, B. C. Sheldon, C. Primmer, and A. C. Syvänen. 2003. Sex chromosome evolution and speciation in Ficedula flycatchers. Proc. R. Soc. Lond. B 270:53–59. Sattler, G. D., and M. J. Braun. 2000. Morphometric variation as an indicator of genetic interactions between black-capped and Carolina chickadees at a contact zone in the Appalachian mountains. Auk 117:427–444. Seddon, N., and J. A. Tobias. 2007. Song divergence at the edge of Amazonia: an empirical test of the peripatric speciation model. Biol. J. Linn. Soc. 90:173–188. Sequeira, F., J. Alexandrino, S. Rocha, J. W. Arntzen, and N. Ferrand. 2005. Genetic exchange across a hybrid zone within the Iberian endemic golden-striped salamander, Chioglossa lusitanica. Mol. Ecol. 14:245–254. 76  Servedio, M. R., S. A. Sæther, and G.-P. Sætre. 2009. Reinforcement and learning. Evol. Ecol. 23:109–123. Seutin, G., B. N. White, and P. T. Boag. 1991. Preservation of avian blood and tissue samples for DNA analysis. Can. J. Zool. 69:82–90. Sinnock, P. 1975. The Wahlund effect for the two-locus model. Am. Nat. 109:565–570. Slabbekoorn, H., and T. B. Smith. 2002. Bird song, ecology, and speciation. Philos. Trans. R. Soc. B 357:493–503. Slatkin, M. 1973. Gene flow and selection in a cline. Genetics 75:733–756. Sotka, E. E., and S. R. Palumbi. 2006. The use of genetic clines to estimate dispersal distances of marine larvae. Ecology 87:1094–1103. Stebbins, G. L. 1959. The role of hybridization in evolution. Proc. Am. Philos. Soc. 103:231–251. Stephens, M., and P. Donnelly. 2003. A comparison of Bayesian methods for haplotype reconstruction from population genotype data. Am. J. Hum. Gen. 73:1162–1169. Swenson, N. G., and D. J. Howard. 2005. Clustering of contact zone, hybrid zones, and phylogeographic breaks in North America. Am. Nat. 166:581–591. Szymura, J. M., and N. H. Barton. 1986. Genetic analysis of a hybrid zone between the fire-bellied toads, Bombina bombina and B. variegata, near Cracow in southern Poland. Evolution 40:1141–1159. Tarr, C. L. 1995. Primers for amplification and determination of mitochondrial controlregion sequences in oscine passerines. Mol. Ecol. 4:527–529. Taylor, D. J., H. L. Sprengher, and S. Ishida. 2005. Geographic and phylogenetic evidence for dispersed nuclear introgression in a daphniid with sexual propagules. Mol. Ecol. 14:525–537. Toews, D. P. L., and D. E. Irwin. 2008. Cryptic speciation in a Holarctic passerine revealed by genetic and bioacoustic analyses. Mol. Ecol. 17:2691–2705. Townsend, J. K. 1837. Description of twelve new species of birds, chiefly from the vicinity of the Columbia River. J. Acad. Nat. Sci. Phila. 7:187–193 Turner, T. L., M. W. Hahn, and S. V. Nuzhdin. 2005. Genomic islands of speciation in Anopheles gambiae. PLoS Biol. 3:e285.  77  Uy, J. A. C., R. G. Moyle, and C. E. Filardi. 2009a. Plumage color and song differences mediate species recognition between incipient flycatcher species of the Solomon Islands. Evolution 63:153–164. Uy, J. A. C, R. G. Moyle, C. E. Filardi, and Z. A. Cheviron. 2009b. Difference in Plumage Color Used in Species Recognition between Incipient Species Is Linked to a Single Amino Acid Substitution in the Melanocortin‐1 Receptor. Am. Nat. 174:244-254. Veen, T., T. Borge, S. C. Griffith, G.-P. Sætre, S. Bures, L. Gustafsson, and B. C. Sheldon. 2001. Hybridization and adaptive mate choice in flycatchers. Nature 411:45–50. Vekemans, X., T. Beauwens, M. Lemaire, and I. Roldan-Ruiz. 2002. Data from amplified fragment length polymorphism (AFLP) markers show indication of size homoplasy and of a relationship between degree of homoplasy and fragment size. Mol. Ecol. 11:139–151. Via, S., and J. West. 2008. The genetic mosaic suggests a new role for hitchhiking in ecological speciation. Mol. Ecol. 17:4334–4345. Vincze, T., J. Posfai, and R. J. Roberts. 2003. NEBcutter: a program to cleave DNA with restriction enzymes. Nucleic Acids Res. 31:3688-3691. Vines, T. H., S. C. Köhler, M. Thiel, T. R. Sands, C. J. MacCallum, N. H. Barton, and B. Nürnberger. 2003. The maintenance of reproductive isolation in a mosaic hybrid zone between the fire-bellied toads Bombina bombina and B. variegata. Evolution 57:1876–1888. Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. van de Lee, M. Hornes, A. Frijters, J. Pot, J. Peleman, M. Kuiper, et al. 1995. AFLP—a new technique for DNA fingerprinting. Nucleic Acids Res. 23:4407–4414. Wang, H., E. D. McArthur, S. C. Sanderson, and J. H. Graham. 1997. Narrow hybrid zone between two subspecies of big sagebrush (Artemisia tridentata: Asteraceae). IV. Reciprocal transplant experiments. Evolution 51:95–102.  78  Weckstein, J. D., R. M. Zink, R. C. Blackwell-Rago, and D. A. Nelson. 2001. Anomalous variation in mitochondrial genomes of White-crowned (Zonotrichia leucophrys) and Golden-crowned (Z. atricapilla) sparrows: pseudogenes, hybridization or incomplete lineage sorting? Auk 118:231–236. Weir, B. S. 1996. Genetic Data Analysis II. Sinauer Associates, Sunderland. Weir, J. T., and D. Schluter. 2004. Ice sheets promote speciation in boreal birds. Proc. R. Soc. Lond. B 271:1881–1887. Wilson, C. C., and L. Bernatchez. 1998. The ghost of hybrids past: fixation of Arctic charr (Salvelinus alpinus) mitochondrial DNA in an introgressed population of lake trout (S. namaycush). Mol. Ecol. 7:127–132. Wood, T. E., N. Takebayashi, M. S. Barker, I. Mayrose, P. B. Greenspoon, and L. H. Rieseberg. 2009. The frequency of polyploid speciation in vascular plants. Proc. Natl. Acad. Sci. USA 106:13875–13879. Wu, C.-I. 2001. The genic view of the process of speciation. J. Evol. Biol. 14:851–865. Zink, R. M., and G. F. Barrowclough. 2008. Mitochondrial DNA under siege in avian phylogeography. Mol. Ecol. 17:2107–2121. Zink, R. M., and M. C. McKitrick. 1995. The debate over species concepts and its implications for ornithology. Auk 112:701–719.  79  Appendix 1: Details of loci sequenced for chapter 2. 1: Chromosome and nucleotide number in Zebra Finch genome sequence (, assembly July 2008) 2: Approximate position of sequence homologous to the flanking region of Numt-Dco1 (data not shown). 3: The forward primer we use is the sequence reported by Milá et al. (2007) for LGL2, which because of a typo differs by one substitution from the original primer sequence (Tarr 1995). Milá et al. used Tarr’s original LGL2 primer sequence. Locus  Start pos. in zebra finch1  End pos. in zebra finch1  Forward Primer  Reverse Primer  Source  Annealing Temp. (°C)  GH1  1 112878784  1 112879348  AACCTGTTTGCCAACGCTGT  CTGGTCCTCCGGAATATAGGTG  Borge et al. 2005  55  RPL30  2 134052606  2 134051596  CCAAGTTGGTCATCCTAGCCA  GCCACTATAATGATGGACACCAGTC  Borge et al. 2005  62.5  LHCGR  3 21664487  3 21663828  TGCCTTCAATGGGACCAAG  CCGCCTGAGGTTTTTGTTGT  Borge et al. 2005  55  Numt_Dco1  4 ~96370002  4 ~96370002  GGCCACATCAGACAGTCCAT  AGTAGCTCGGTTCTCGTGAG  Milá et al. 2007, Tarr 19953  61.5  RPL5  8 10441755  8 10441184  GTTGGCCTGACCAATTACGC  CTTCAACTTGGCCTTCATAGATCTT  Borge et al. 2005  55  CEPU1  24 6314566  24 6315079  GTGCAGTGCCTCCAACGAC  TCGCATCCGAGATGTACGG  Borge et al. 2005  55  24555  Z 6656177  Z 6656632  CCTCCAGATATTTCATTCCC  AATGGAAATGGCTGAACTTG  ADAMTS6  Z 50815125  Z 50815666  GGAGAGAATGGATTTCTGCC  TGATTCCAGTCTAGGAAACG  BRM  Z 64937586  Z 64937254  AGCACCTTTGAACAGTGGTT  TACTTTATGGAGACGACGGA  Borge et al. 2005  62.8  CHD1Z  Z 24736734  Z 24736133  GTTACTGATTCGTCTACGAGA  ATTGAAATGATCCAGTGCTTG  Fridolfsson and Ellegren 1999  58  PTCH  Z 9746535  Z 9747057  CCATTTTCTTCCAAGCAATA  TTTCTTGACAGTCCATAGCA  Borge et al. 2005  57.5  Backström et al. 2006 Backström et al. 2006  61.5 62.8  80  Appendix 2: Correlation coefficients among two genetic markers (CHD1Z, Numt-Dco1) and five plumage traits (Throat, Auricular, Spot, Line, Wing) among samples within 20 km of the hybrid zone centre. Trait CHD1Z CHD1Z --Numt-Dco1 .25 Throat .51 Auricular .37 Spot .42 Line .42 Wing .50  Numt-Dco1  Throat  Auricular  Spot  Line  Wing  --.35 .26 .35 .35 .27  --.56 .46 .53 .63  --.43 .56 .44  --.75 .50  --.49  ---  Appendix 3: Linkage disequilibrium at the centre of the hybrid zone increases due to Wahlund effect as samples farther than 30 km from the centre are included. We used samples within 20 km of the centre to calculate LD in chapter 2.  81  Appendix 4: AFLP primer combinations used in chapter 4, and number of polymorphic loci obtained from each. EcoRI Primer E-AAC E-AAC E-AAG E-AAG E-ACC E-ACT E-AGC E-AGC E-AGG E-AGG  MseI Primer M-CAA M-CAG M-CAC M-CAT M-CATT M-CATT M-CAA M-CAG M-CAC M-CAT  Number of loci 34 17 16 22 37 45 38 16 37 49  82  Appendix 5: Primers used for amplification and sequencing of nuclear loci in chapter 4. External primer sources are (1) Borge et al. 2005, (2) Bäckstrom et al. 2006, (3) Fridolfsson and Ellegren 1999, and (4) this study; all internal primer sequences were developed for this study. All PCR reactions consisted of an initial 3 minutes at 94 ºC followed by 35 cycles of 30 seconds at 94 ºC, 30 seconds at annealing temperature (see table), and a variable extension time (see table) at 72 ºC, ending with 5 minutes at 72 ºC. Locus RPL30 CEPU RPL5 GH1 13446 5087 9300 LHCGR KIAA1383 ARID5B CHD1Z Numt-Dco1  External primer source  External annealing temp. (ºC)  Internal annealing temp. (ºC)  Extension time (s)  1  54  56  1:15  1  54  56  1:15  1  52  56  1:00  1  52  56  1:00  2  52  56  1:00  2  52  56  1:00  2  52  56  1:00  1  52  56  1:00  4  52  56  1:00  4  52  56  1:00  n/a  3  58  n/a  0:45  CCTTCCTCTAATTCCCTACTGTCA, CAGAGTGACCCCGAGAAAAG  4  54  56  1:15  External primer pair (f, r)  Internal primer pair (f, r)  CCAAGTTGGTCATCCTAGCCA, GCCACTATAATGATGGACACCAGTC GTGCAGTGCCTCCAACGAC, TCGCATCCGAGATGTACGG GTTGGCCTGACCAATTACGC, CTTCAACTTGGCCTTCATAGATCTT AACCTGTTTGCCAACGCTGT, CTGGTCCTCCGGAATATAGGTG GATGTCATTTCAAGGTTTGC, TCAGGTGTATGACCCTCTC CTGGCAAATCATGATGACAG, CTGGATGGGCACTTGATCAG GATTACCTTCATGCCACTGC, CATACGAGGTGCCTTGAAGT TGCCTTCAATGGGACCAAG, CCGCCTGAGGTTTTTGTTGT ATGCTCGCTTCCATCTGC, TGGACCTGCTCCTCCTCA AGGCAGCTACCCTGCTCA, AGGCCTGGAAAGGCAAGA GTTACTGATTCGTCTACGAGA, ATTGAAATGATCCAGTGCTTG CCCTACATTGTCCAAGCCATT, TGACGTACAGCCCTACATTGTC  TGGGAGAAAACWATTCTRAKATGR, TGARATCCAAAAAYTGGGTGT GCAGCCTTTGGSACTCCT, TCACCGTCAACTGTGAGTSC TACGCTGCTGCCTACTGC, GGCCAAATTTATTCAGAAGCTG TGGCTGCCGAGACATACA, GCCCATGTTCTGGAGARG AAGGTTTGCACAGYTGGAGT, TGMAAGAAACCAAAGYGATTC CTATGCCATMCCAGGTGAGT, CAGTGGAGCTGTAGGGAAGG CACTGCTTACCAGGGGACTT, AGCCAGATCTSCTCGAGTCA GGGACMAAGCTGAATCAA, TGAGGTTYTTGTTGTCYTTCAG GCATTTCAGGGGCAAGTG, TGGTGGGAAAGGGACAGA GCTCACTTGCTCAGAAAATTCA, AAGAAATGTGGAGCAAGCAATTA  83  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items