Patterns of Molecular and Morphological Variation in Some Solomon Island Land Birds

Jun 10, 05:56 AM

Current Headlines: By Smith, Catherine E Filardi, Christopher E

ABSTRACT.- The Solomon Archipelago is the largest and most biologically complex archipelago in northern Melanesia. We collected tissues and voucher specimens from codistributed bird species found on five island groups that provided the first opportunity to apply molecular methods to this avifauna. Using the mitochondrial marker ND2, we constructed a series of intraspecific phylogenies for 23 ecologically and taxonomically diverse species (13 families from 5 orders). Intraspecific comparisons across islands revealed a broad range of genetic differentiation, from 0% in widespread dispersive species such as Eurystomus orientalis and Nectarinia jugularis, to as high as 4-8% in species such as Ceyx lepidus and Dicaeum aeneum. Fifteen of the 23 species had well-supported phylogeographic structure, and for many of these, endemic and otherwise, molecular phylogenetic relationships closely matched those delineated by morphology. However, degrees of genetic differentiation appeared to be inconsistent across taxonomic designations, and the monophyly of some endemic taxa was not well supported. The data reveal the limits of morphology in tracking complex evolutionary histories and suggest that taxonomic revision of some Solomon Islands birds is warranted. As the first molecular analyses of this avifauna, results presented here create a platform for further biogeographic and evolutionary studies of bird diversity in this influential region. Received 17 December 2004, accepted 5 April 2006.

Key words: comparative phylogeography, genetic variation, ND2, northern Melanesia, taxonomy.

Patrons de variation moleculaire et morphologique chez quelques oiseaux terrestres des iles Salomon

RESUME. - L'archipel des iles Salomon est, du point de vue biologique, le plus grand et le plus complexe de tous les archipels du nord de la Melanesie. Nous avons recolte des tissus et des specimens d'especes d'oiseaux co-distribuees sur cinq groupes d'iles. A l'aide du marqueur mitochondrial ND2, nous avons etabli une serie de phylogenies intraspecifiques pour 23 especes differentes du point de vue ecologique et taxonomique (13 familles et 5 ordres). Les comparaisons intraspecifiques entre les iles ont revele un large eventail de differentiation genetique, de 0% chez les especes a grande dispersion tel que Eurystomus orientalis et Nectnrinia jugularis, jusqu'a 4-8% chez les especes tel que Ceyx lepidus et Dicaeum aeneum. Quinze des 23 especes avaient une structure phylo-geographique bien etablie. Pour plusieurs d'entre elles, qu'elles soient endemiques ou non, les relations phylogenetiques moleculaires s'apparentaient fortement a celles depeintes par la morphologie. Cependant, le degre de differentiation genetique etait inconstant parmi les designations taxonomiques et la monophylie de certains taxons endemiques n'etait pas bien etablie. Les donnees revelent les limites de l'utilisation de la morphologie pour retracer les histoires evolutives complexes et suggerent une revision de la taxonomie de certains oiseaux des iles Salomon. En tant que premieres analyses moleculaires de cette avifaune, les resultats presentes ici servent de base pour des etudes plus poussees sur la biogeographie et l'evolution de la diversite des oiseaux dans cette region.

THE BIRDS OF the Solomon Islands (Fig. 1) constitute one of the most distinctive avifaunas in the world. Of 163 breeding land birds, 72 (44%) are found nowhere else in the world, and 62 (38%) occur elsewhere but are represented by distinct races or subspecies in the Solomons (Mayr 1945, Mayr and Diamond 2001). These exceptional levels of endemism, coupled with rich patterns of intraspecific variation, create an unparalleled biogeographic laboratory that has inspired biogeographers and evolutionary thinkers alike. Ernst Mayr's formulation of the biological species concept (Mayr 1942, 1963), MacArthur and Wilson's (1967) theory of island biogeography, and Jared Diamond's provocative ideas on dispersal, competition, and gene flow (e.g., Diamond 1970, 1974) were all influenced by apparent patterns of speciation in this region.

In a recent monograph, Mayr and Diamond (2001) compiled data on the ecology, distribution, and morphology of birds from northern Melanesia and presented a comprehensive taxonomic synthesis based on their analyses. However, without well-established phylogenetic hypotheses, explanations for dramatic biogeographic patterns in the Solomon Islands are exceedingly difficult to test. Until now, the extreme paucity of both modern data-rich scientific specimens and tissue samples from Solomon Islands birds prevented deductive treatment of this avifauna that goes beyond what can be gleaned from the external characters of study skins.

Molecular techniques offer an additional lens into the rich patterns of morphological diversity characterized by Mayr and Diamond (2001). Using molecular data, we describe some intraspecific phylogeographic patterns for 23 codistributed Solomon Islands bird species and compare them with morphological patterns of differentiation (from Mayr and Diamond 2001). Here, we present the first molecular data comparing multiple Solomon Islands bird lineages, which we hope will form the beginnings of a platform from which theories, hypotheses, and concepts of speciation in this region can be evaluated with genetic -in addition to morphological, ecological, and behavioral -data.

Taxonomy of Solomon Islands land birds. - Mayr and Diamond (2001) used terminology that clearly defined graded levels of morphological differentiation in Melanesian birds and follows the tenets of the biological species concept (Mayr 1942). Assuming that allopatric speciation is the predominant mode of avian speciation in this region and that geographic differentiation proceeds gradually, subspecies are expected to differ continuously in their distinctness. To capture this continuum taxonomically, Mayr and Diamond (2001) characterized subspecies as weak-, sub- and mega- subspecies, corresponding to subtle versus more dramatic degrees of morphological distinctness. All subspecies are considered to belong to the same species and are assumed to interbreed. If closely related allopatric populations are sufficiently distinct (but do not yet have any geographic overlap, i.e., sympatry of sister species pairs), they are termed "allospecies." Allospecies are assumed to be reproductively isolated, as inferred from differences in mating behavior, song, morphology, or all three. Sets of closely related allospecies are grouped as superspecies, indicating that speciation, as defined by the biological species concept, has not been unequivocally completed. Thus, full species status is conservatively reserved for taxa that have achieved some degree of stable sympatry and, in turn, ecological differentiation.

If the morphological taxonomies put forth by Mayr and Diamond (2001) are good predictors of genetic isolation, their designations of subspecies and allospecies can be used as hypotheses of relative degrees of genetic differentiation. Although limited sample sizes preclude any definitive assessment of gene flow between island populations, a preliminary species-by-species assessment of phylogeographic structure across the archipelago will give qualitative indications of degrees of isolation to compare with taxonomic designations and can identify taxa appropriate for future population-level analyses.

METHODS

Taxon sampling.-Between 1997 and 2001, we collected tissue samples and voucher specimens that constitute the only systematically sampled modern collection of Solomon Islands birds. Over the course of five expeditions (June-August 1997, June-August 1998, April-June 1999, January-March 2000, and February-May 2001), we visited five of the seven major island groups in the Solomon Archipelago (Isabel, Choiseul, Guadalcanal, Malaita, and New Georgia), and one outlying island (Rennell; see Fig. 1). In addition to collecting samples, these expeditions enabled us to gather new knowledge of the distributions and life histories of Solomon Islands birds (Filardi et al. 1999; Kratter et al. 2001 a, b; Filardi 2003; Smith 2003).

The 23 focal taxa in the present study include all species for which we had individuals from at least four main island groups and for which we could make a reasonable assumption of monophyly (i.e., the white-eyes [Zosterops spp.) were excluded because they are variably classified into 7-10 species of disparate affinities across the five focal island groups in the present study). Unpredictable encounter rates, access to appropriate habitat, seasonal changes, and limits on field time impeded complete taxonomic sampling. However, focal species comprise a taxonomically diverse (13 families from 5 orders) and ecologically varied array of birds, with contrasting distribution patterns including Solomon Islands endemics, Melanesian endemics, and more widespread species (Table 1). Voucher specimens are housed at the University of Washington Burke Museum (UWBM). Tissue samples (heart, liver, muscle) were collected in the field and stored in lysis buffer or ethanol until frozen (at -80[degrees]C) on return to the UWBM.

Through loans from existing tissue collections at otherutions, as well as a UWBM expedition to Papua New Guinea in 2001, we acquired many tissue samples helpful in outgroup analyses. The absence of regional generic-level phylogenies including Solomon Islands taxa makes it difficult to confidently identify sister taxa. Because recent comparative studies have emphasized the importance of comparisons with sister taxa (see Johnson and Cicero 2004), we included, whenever possible, multiple closely related outgroups to identify the most probable sister taxa and to verify ingroup monophyly. For Solomon Islands endemics, ideal outgroups are often sister taxa from adjacent areas such as New Guinea and Australia. For more widespread Australasian species, appropriate outgroups are conspecific individuals from adjacent regions. Because Melanesian birds are poorly represented in existing tissue collections, optimal outgroups were often unavailable. The museum or voucher numbers of all samples used in the study are listed in the Appendix. Molecular markers. - We used the NADH dehydrogenase unit 2 gene (ND2) as the focus for our phylogenetic analyses. Preliminary genetic data from both cytochrome b and ND2 suggested that ND2 evolved slightly faster in some focal Solomon Islands taxa and that variation in ND2 was sufficient to resolve intraspecific phylogenies. Preliminary ND2 data for some species yielded quite dramatic genetic breaks, indicating that population-level analyses with more rapidly evolving noncoding markers may be inappropriate. Although the ability of mitochondrial DNA (mtDNA) gene trees to infer trees for taxa is increasingly questioned (e.g., see Ballard and Whitlock 2004), and we acknowledge the need for additional, independent estimates of phylogenies, such an undertaking was beyond the scope of this preliminary look at Solomon Islands birds. However, to mitigate concerns over lineage-specific differences in the rate of evolution of single mitochondrial markers, we also sequenced a portion of cytochrome b for more than half of the focal species. Although results must be interpreted with caution, congruence of pattern for both markers adds confidence to conclusions.

DNA extraction, amplification, and sequencing.- Whole-genomic DNA was digested and extracted using Qiagen DNeasy tissue kits (Qiagen, Valencia, California), following the manufacturer's protocols. After extraction, DNA was amplified by polymerase chain reaction (PCR) in 30-[mu]L reactions in an Applied Biosystems PCR System 9700 thermocycler (Applied Biosystems, Foster City, California) with the following reaction conditions: 2 min at 94[degrees]C, followed by 30 cycles of 30 s at 94[degrees]C and at 48-50[degrees]C, 30-45 s at 72[degrees]C; followed by a final extension of 7 min at 72[degrees]C. The two primers that amplified ND2 for all species and outgroups have the following sequences: L5215 (Hackett 1996; 5'- TATCGGGCCCATACCCCGAATAT-3') and H1064 (Drovetski et al. 2004; 5'- CTTTGAAGGCCTTCGGTTTA-3'). The PCRs included at least one negative control, and products were examined on agarose gels stained with ethidium bromide.

We prepared PCR products for sequencing using a Qiaquick PCR Purification Kit (Qiagen) according to the manufacturer's protocols and sequenced using 2.0 [mu] of ABI Prism BigDye Terminator (Applied Biosystems) in 10-[mu]L reactions for 30 cycles under the following conditions: 10 s at 96[degrees]C, 5 s at 50[degrees]C, and 4 min at 60[degrees]C. Sequencing was performed on an ABI Model 377 (Applied Biosystems). A series of species-specific internal sequencing primers was designed, resulting in most sequenced samples with >30% overlap. Difficult samples were re-extracted, then reamplified or resequenced or both, until unambiguous results were obtained.

We sequenced >800 base pairs (bp) of ND2 for 23 widespread species (Table 1) from at least four main island groups in the Solomon Islands (Fig. 1). When possible, we sequenced ND2 from at least two individuals from each available population (= island). Phytogenies presented here rely almost exclusively on ND2 sequence data (see exception below). All sequences were aligned in SEQUENCHER, version 3.0 (Gene Codes Corporation, Ann Arbor, Michigan) and verified manually. Sequences are deposited in GenBank (accession numbers: DQ468876-DQ469086).

The problem of nuclear copies.-We assumed that mitochondrial gene trees recovered from sequence data accurately reflect species-tree topologies (see Moore 1995). The sequencing of nuclear copies of these same genes (hereafter "pseudogenes"), however, is difficult to detect and can grossly confound phylogenetic inference (e.g., Arctander 1995, Kidd and Friesen 1998). To minimize the possibility of sequencing pseudogenes, we (1) translated all sequence data into codons to confirm the absence of stop codons; (2) plotted the transition-transversion ratios by codon position to confirm expected contrasts for mitochondrial coding sequence (e.g., Edwards 1997); (3) sequenced >/=500 bp of cytochrome b for 13 focal taxa, assessed signal compatibility with ND2 using partition homogeneity tests (Bull et al. 1993, Harris et al. 1994, Swofford 2002), and plotted relative rates of gene evolution to detect anomalous sequence; and (4) performed mitochondrial enrichments (Beckman et al. 1993) on a subset of individuals to verify that sequence data from the resulting extracts matched sequence data obtained from total genomic extractions for the same individuals.

Phylogenetic analyses. - Intraspecific phylogenies were constructed using both maximum-likelihood (ML) and maximum- parsimony (MP) optimality criteria in PAUP*, version 4 (Swofford 2002). For ML analyses, we constructed intraspecific phylogenies using the HKY85 substitution model. We estimated the gamma- distribution parameter, transition-bias parameter, and proportion of invariable sites separately from each of the data sets during tree construction. For MP analyses, all sites were weighted equally. Because results for ML and MP analyses were consistent, we present only ML results here.

We used bootstrap analyses with 100 bootstrap replicates to provide a preliminary indication of support for tree topology (Felsenstein 1981). To present as much structure as possible, we used 50% bootstrap support as a cut-off when showing species topologies. Although this value may prove, on occasion, to misrepresent phylogenetic rank, it does not affect the strong support for most area groupings of individuals. Ideally, a more stringent test of topology would be employed (e.g., determining whether the data can reject monophyly of designated taxa), but larger sample sizes and greater outgroup representation are needed- especially given the apparent recency of some intraspecific splits (e.g., Filardi and Smith 2005). Tree topologies were then compared with taxonomic limits described by Mayr and Diamond (2001). This coarse comparison provides a framework for a preliminary, qualitative mtDNA assessment of phylogeographic patterns.

RESULTS

Identification of possible nuclear copies.-All analyses suggested that, barring two exceptions, gene-sequence data were from mitochondrial rather than nuclear genes. Consistent with previous studies of avian mtDNA, mean base-pair frequencies for ND2 and cytochrome b were strongly AC-skewed and, in general, both genes provided concordant signal across lineages (data not shown). The absence of stop codons and expected transition-transversion ratios (not shown) suggested that targeted mitochondrial markers were amplified. In all but two taxa (see below), linear regression indicated significant correlation between the two markers, and partition-homogeneity tests revealed that genes contained compatible signal (results not shown).

Nuclear copies were, however, suspected in Myiagra ferrocyanea and Micropsitta finschii. The partition homogeneity tests for Myiagra ferrocyanea revealed incompatible signal between ND2 and cytochrome-b data sets (P = 0.017). Although the same test for Micropsitta finschii was not significant (P = 0.24), the P value is also much lower than the values obtained for all other taxa (P > 0.6), and low sample size reduced the power of the test to detect significant signal incompatibility. Analyses of pairwise sequence divergences also suggested incompatibilities (R^sup 2^ = 0.68 and 0.23 for Micropsitta finschii and Myiagra ferrocyanea, respectively; R^sup 2^ > 0.75 for all others; analyses not shown).

Amplification and sequencing of ND2 from mitochondrial extractions of two M. ferrocyanea individuals resulted in matching sequences, which tentatively suggests that cytochrome-b data contained nuclear copies. Independently obtained cytochrome-b sequence data from an individual of Micropsitta finschii (GenBank accession number: U89176; de los Monteros 2000) from Isabel Island (reported erroneously in de los Monteros [2000] as coming from Rennell Island; M. LeCroy pers. comm.) corresponded very nearly to our cytochrome-b sequence for individuals from Isabel and Choiseul, which suggests contamination of the ND2 data set. Data sets with suspected nuclear copies were omitted from further analyses.

Phylogenetic analyses.-On the basis of ND2 results, taxa can be divided into those with phylogenetic structure across the archipelago and those without (see above). Figure 2 presents intraspecific phylogenies for the 15 (of 23) focal taxa with genetic structure. The remaining eight had little or no genetic differentiation (see Table 2) and no resolvable phylogeographic structure (not shown). Average uncorrected pairwise genetic distances between islands varied substantially for species with phylogeographic structure (Table 2). Dietician aeneum, Myzomela lafargei, and Ceyx lepidus all had pairwise base differences >6% for some island comparisons. By contrast, Coracina papuensis, C. tenuirostris, Rhipidura rufifrons, Myiagra ferrocyanea, and Aplanis grandis did not vary >2% from island to island (Table 2).s for the latter species and "deeper" tree topologies for the former (see Fig. 2). Comparing genetic and morphological patterns. - The morphological subspecies described by Mayr and Diamond (2001; Table 3) generally coincide with patterns detected in the genetic data (Table 2 and Fig. 2). Widespread, generally monotypic species show no phylogeographic structure across the Solomon Islands. By contrast, 10 of 16 species with subspecific differentiation show genetic breaks that coincide with morphological breaks (Table 3). Genetic data for the six remaining species contradict or simply do not identify subspecies limits as predicted by morphology. In three cases (Coracina papuensis, C. tenuirostris, Aplanis grandis), morphology fails to identify genetically distinct populations on the New Georgia group (Fig. 2). In a fourth case (Rhipidura rufifrons), morphological patterns have resulted in one designated subspecies shared by Choiseul and Isabel with a distinct subspecies on Guadalcanal (Table 3), whereas genetic data suggest equally distinct populations on all three islands (Fig. 2). In the fifth case, genetic data for Monarcha castaneiventris reveals a marked break between Malaita and all other islands (Fig. 2), despite the Malaitan population being morphologically indistinguishable from Choiseul, Isabel, and Guadalcanal (Mayr and Diamond 2001; Table 3). In the final case, a separate morphological subspecies on Malaita for Myiagra ferrocyanea is not supported by any detectable genetic break in the mtDNA gene trees.

Outgroup analyses and evidence of paraphyly. Outgroup analyses established monophyly for Solomon Islands representatives of 20 taxa but failed to do so for Monarcha castaneiventris, M. barbatus, and Myzomela lafargei (Fig. 3A-C), all classified morphologically as Solomon Islands endemics (Mayr and Diamond 2001). Monarcha castaneiventris forms a polytomous clade with an allopatric and a sympatric congener (M. melanopsis and M. cinerascens, respectively; Fig. 3A). Similarly, M. barbatus forms a clade with M. guttula and M. trivirgatus (both allopatric congeners), within which the branching hierarchy is unresolved (Fig. 3B). For Myzomela lafargei, including M. cardinalis (a partially sympatric congener) in the phylogeny contradicts monophyly for this putatively endemic lineage (Fig. 3C). Additionally, outgroup analyses called into question previous species delimitations for the exceptionally complex Pachycephala pectoralis group. Although monophyly of the Solomon Islands P. pectoralis was strongly supported, analyses with conspecifics from Australia and West New Britain as well as the variably classified P. melanura (see Beehler et al. 1986, Mayr and Diamond 2001) nested this latter species within the clade of Australian P. pectoralis (Fig. 3D).

Other notable patterns also emerged from outgroup analyses. In particular, many widespread (i.e., nonendemic) Solomon Islands species were substantially genetically distinct from conspecifics outside the archipelago. Outgroup samples from adjacent regions of Papua New Guinea for Ceyx lepidus and P. pectoralis (both species with well-supported genetic structure across the archipelago and known patterns of morphological differentiation across their ranges) were >6% and >4% divergent, respectively (not shown). Although optimal outgroups were unavailable, three species with little or no evidence of genetic structure across the archipelago (Collocalia esculenta, Halcyon chloris, and Nectarinia jugularis) were 5.0%, 2.7%, and 7.0% divergent, respectively, from conspecific samples from the Philippines.

DISCUSSION

Molecular analyses. - Successful resolution of phylogeographic structure in many Solomon Islands taxa speaks to the unique characteristics of both this geographic region and its avifauna that have made it remarkable for morphological (and now genetic) differentiation. We uncovered dramatic genetic breaks across very short distances in species such as Ceyx lepidus, R. cockerelli, Monarcha barbatus, P. pectoralis, D. aeneum, and Myzomela lafargei (Fig. 2), all of which have one or more populations with intraspecific differences comparable with species-level distinctions suggested by previous studies (Prychitko and Moore 1997, Johns and Avise 1998, Slikas et al. 2000). On a broader scale, genetic differences between widespread Solomon Islands taxa and conspecific outgroups outside the Solomon Islands (see above) reveal previously undocumented genetic structure supporting classification of all focal species as endemic to northern Melanesia at least at the subspecies level (Mayr and Diamond 2001). Collectively, these preliminary patterns of genetic differentiation further identify this region as a virtually unparalleled center of avian endemism (e.g., Stattersfield et al. 1999) and support Mayr and Diamond's (2001) long-standing contention that the Solomon Islands are an important natural laboratory for the study of speciation.

Comparing genetic and morphological patterns. Mayr and Diamond's (2001) interpretations of morphological differentiation have proved to be remarkably good predictors of genetic patterns of differentiation across this island archipelago (Table 3). However, two groups of exceptions are notable. First, not surprisingly, Mayr and Diamond's (2001) taxonomic designations do not consistently correlate with degrees of molecular differentiation. Second, the monophyly of three Solomon Islands endemics (Monarcha castaneiventris, M. barbatus, and Myzomela lafargei) and one widespread Australasian species (P. pectoralis) is not well supported by ND2 genealogies (Fig. 3).

Ideally, taxonomy should reflect evolutionary relationships and, by inference, degrees of genetic differentiation. Significantly, we do not expect taxonomic equivalency across lineages, because idiosyncrasies in the mode and tempo of speciation within and among lineages should cloud any general genetic thresholds for taxonomic rank. However, if taxonomy is to reflect evolutionary processes, within a lineage we expect, for example, that the classification of subspecies will generally correlate with divergences that are less extreme than divergences at the species level. The most morphologically distinct populations of Solomon Islands birds are designated as allospecies (Mayr and Diamond 2001; see Table 3), yet genetic differences do not consistently track morphological patterns. For example, although genetically distinct, the New Georgia allospecies for Monarcha castaneiventris and M. barbatus (M. richardsii and M. browni, respectively) are no more distinct than their comparable Malaitan populations.

Three additional cases suggest complex evolutionary histories and notable decoupling of morphological and (presumably neutral) molecular evolution. Three Solomon Islands endemics (M. barbatus, M. castaneiventris, and Myzomela lafargei) are paraphyletic with either allopatric or sympatric species from outside their Northern Melanesian ranges (Fig. 3A-C). In each case, the lack of resolution is evidence of very short internodes among internal branches of these trees. With increased taxon sampling and a more rapidly evolving marker, Filardi and Smith (2005) revealed that populations of Solomon monarchs are in fact monophyletic. More detailed analyses of the Myzomela lafargei-cardinalis group may similarly resolve this polytomy. Perhaps rather than failures of morphological analyses, these three examples expose the limitations of our preliminary molecular results, particularly in light of complex, and often explosive, insular radiations (see Filardi and Smith 2005).

Finally, outgroup analysis for the P. pectoralis species group disclosed that P. melanura grouped neatly with Australian samples of P. pectoralis (Fig. 3D). Notably, taxonomists have long been challenged by the classification of the exceptionally complex and diverse subspecies of P. pectoralis and its close allies (Galbraith 1956, Mayr 1967). Pachycephala melanura has generally been considered a separate species, morphologically distinct and segregating on a finer geographic scale from the mainland P. pectoralis by virtue of inhabiting smaller and more remote islands (Beehler et al. 1986). Ecological segregation and limited hybridization between these two taxa (Diamond 2002) suggest that P. melainira's status as a separate biological species may be merited. Available genetic data suggest, however, that unless P. pectoralis is split into multiple species (e.g., a Solomon species and two Australian species; Fig. 3D), P. melamtra may not comprise a natural species grouping (Fig. 3D). What emerges as critical to unraveling the questions raised by this preliminary work is dense sampling and thorough, geographically driven outgroup selection when constructing phylogenies of closely related, poorly circumscribed species.

The inability of morphology to predict genetic breaks in the above cases can be better understood by considering (1) a long history of idiosyncratic taxonomic methods and (2) the disproportionate weight that plumage characters have traditionally had when delimiting species, allospecies, or subspecies status. Importantly, Mayr and Diamond's (2001) taxonomic treatments rest on the systematic and taxonomic work of many investigators. As a result, these taxonomies are a product of a range of judgment calls made over a number of decades. Although the taxonomy of Solomon Islands birds generally reflects patterns of plumage differentiation among island populations, it is neither internally consistent nor impartially quantified. Consequently, it is difficult to judge whether intraspecific phylogenies that show a discrepancy with taxonomy are a result of (1) poor correspondence between evolution of morphological traits and genetics or (2) inconsistent taxonomies.

Furthermore, plumage characters have traditionally carried disproportionate weight when delimiting species, allospecies, or subspecies status. It follows that if natural selection is exerting strong pressure on island populations to diverge or conserve plumage patterns, genetic differentiation will (respectively) be less than or exceed what morphology might predict. A recent analysis of Pacific radiations in the genus Monarcha exposes the difficulty of relying solely on integumentary characters and geography to decipher complex evolutionary histories (Filardi and Moyle 2005). Thus, from both morphological and genetic perspectives, many taxonomic designations within Solomon Islands bird lineages are not equivalently defined, and the distinction between subspecific and species-level classifications is blurred. The continuous spectrum of differentiation evident in this data set clearly illustrates the challenges of assigning discrete levels of genetic and morphological differentiation to taxonomic levels. However, the island populations of many species have strongly supported clades, and although sample sizes preclude our ability to rule out low-frequency gene flow, these clades may represent phylogenetic units with independent evolutionary trajectories (Fig. 2). But because gene trees reflect the coalescent process, which does not necessarily encapsulate the process of speciation, molecular data alone will often be insufficient to define species limits. Taxonomic revision will require careful and consistent consideration of ecological, morphological, and distributional data as well.

In addition to informing taxonomy, the intraspecific phylogenies presented here provide a framework for testing evolutionary hypotheses in a region that has contributed disproportionately to concepts of species and speciation (e.g., Mayr 1963; MacArthur and Wilson 1967; Diamond 1970, 1974; Diamond and Mayr 1976; Diamond et al. 1976). For example, lower sea levels in the Pleistocene are believed to have connected Choiseul, Isabel, and possibly Guadalcanal (Fig. 1), which explains, at least in part, the consistent relationship between these three islands for many of the shallower, presumably more recent lineages (Fig. 2). Patterns are preliminary but suggest powerful influences of both contemporary geography and historical geological and climatic change (see also Mayr and Diamond 2001). Ultimately, the placement of this region's avifauna in a comparative and historical phylogeographic context can help elucidate both proximate and ultimate mechanisms of genetic and morphological differentiation among island birds.

ACKNOWLEDGMENTS

The Ferguson Foundation, University of Washington Burke Museum Ornithology Endowment, and the Wildlife Conservation Society all generously supported fieldwork necessary to collect tissue samples in the Solomon Islands and New Guinea. Eddy and Leader's five fellowships from the Burke Museum to both authors and a National Science Foundation predoctoral grant to C.E.S. supported both field and laboratory components of this work. Most of the tissue samples used in the study are housed at the Burke Museum. Additional tissue samples were generously granted by Commonwealth Scientific and Industrial Research Organization (Australia), University of Kansas, the Field Museum of Natural History, and the Royal Ontario Museum. Special thanks to J. Diamond for sharing some of his exceptional knowledge of the northern Melanesian avifauna that helped set the tone of our dissertation work, and to S. Rohwer, S. Birks, S. Edwards, and R. Huey for thoughtful and constructive feedback. Comments from T. K. Pratt and an anonymous reviewer improved the manuscript. M. Biliki and J. Horokou at the Solomon Islands Ministry of Forests, Environment and Conservation; and D. Malasa at the Solomon Islands Ministry of Education provided invaluable assistance. We thank D. Kwon and Eagon Resource Development Company for access to field sites on New Georgia and Choiseul. Finally, we thank the many Melanesians who have generously shared their wisdom and their homes with us and who have been instrumental in facilitating all facets of this work.

LITERATURE CITED

ARCTANDER, P. 1995. Comparison of a mitochondrial gene and a corresponding nuclear pseudogene. Proceedings of the Royal Society of London, Series B 262:13-19.

BALLARD, J. W. O., AND M. C. WHITLOCK. 2004. The incomplete natural history of mitochondria. Molecular Ecology 13:729-744.

BECKMAN, K. B., M. F. SMITH, AND C. ORREGO. 1993. Purification of mitochondrial DNA with Wizard Minipreps DNA Purification System. Promega Notes Magazine 43:10-13.

BEEHLER, B. M., T. K. PRATT, AND D. A. ZIMMERMAN. 1986. Birds of New Guinea. Princeton University Press, Princeton, New Jersey.

BULL, J. J., J. P. HUELSENBECK, C. W. CUNNINCHAM, D. L. SWOFFORD, AND P. J. WADDELL. 1993. Partitioning and combining data in phylogenetic analysis. Systematic Biology 42: 384-397.

DE LOS MONTEROS, A. E. 2000. Higher-level phylogeny of Trogoniformes. Molecular Phylogenetics and Evolution 14:20-34.

DIAMOND, J. M. 1970. Ecological consequences of island colonization by southwest Pacific birds, II. The effect of species diversity on total population density. Proceedings of the National Academy of Sciences USA 67: 1715-1721.

DIAMOND, J. M. 1974. Colonization of exploded volcanic islands by birds: The supertramp strategy. Science 184:803-806.

DIAMOND, J. M. 2002. Dispersal, mimicry, and geographic variations in Northern Melanesian birds. Pacific Science 56:1-22.

DIAMOND, J. M., M. E. GILPIN, AND E. MAYR. 1976. Species- distance relation for birds of the Solomon Archipelago, and the paradox of the great speciators. Proceedings of the National Academy of Sciences USA 73: 2160-2164.

DIAMOND, J. M., AND E. MAYR. 1976. Speciesarea relation for birds of the Solomon Archipelago. Proceedings of the National Academy of Sciences USA 73:262-266.

DROVETSKI, S. V., R. M. ZINK, S. ROHWER, I. V. FADEEV, E. V. NESTEROV, I. KARAGODIN, E. A. KOBLIK, AND Y. A. RED'KIN. 2004. Complex biogeographic history of a Holarctic passerine. Proceedings of the Royal Society of London, Series B 271:545-551.

EDWARDS, S. V. 1997. Relevance of microevolutionary processes to higher level molecular systematics. Pages 251-278 in Avian Molecular Evolution and Systematics (D. P. Mindell, Ed.). Academic Press, San Diego, California.

FARRIS, J. S., M. KOLLERSJO, A. G. KLUGE, AND C. BULT. 1994. Testing significance of incongruence. Cladistics 10:315-319.

FELSENSTEIN, J. 1981. Evolutionary trees from DNA sequences: A maximum likelihood approach. Journal of Molecular Evolution 17:368- 376.

FILARDI, C. E. 2003. Speciation and the biogeographic history of Tropical Pacific flycatchers (genus Monarchn). Ph.D. dissertation, University of Washington, Seattle.

FILARDI, C. E., AND R. G. MOYLE. 2005. Single origin of a pan- Pacific bird group and upstream colonization of Australasia. Nature 438:216-219.

FILARDI, C. E., AND C. E. SMITH. 2005. Molecular phylogenetics of monarch flycatchers (genus Monarchn) with emphasis on Solomon Island endemics. Molecular Phylogenetics and Evolution 37:776-788.

FILARDI, C. E., C. E. SMITH, A. W. KRATTER, D. W. STEADMAN, AND H. P. WEBB. 1999. New behavioral, ecological and biogeographic data on the avifauna of Rennell, Solomon Islands. Pacific Science 53:319- 340.

GALBRAITH, I. C. J. 1956. Variation, relationships, and evolution in the Pachycephala pectoralis superspecies (Aves, Muscicapidae). Bulletin of the British Museum (Natural History) Zoology 4: 131- 222.

HACKETT, S. J. 1996. Molecular phylogenetics and biogeography of tanagers in the genus Ramphocelits (Aves). Molecular Phylogenetics and Evolution 5:368-382.

JOHNS, G. C., AND J. C. AVISE. 1998. A comparative summary of genetic distances in the vertebrates from the mitochondria cytochrome b gene. Molecular Biology and Evolution 15: 1481-1490.

JOHNSON, N. K., AND C. CICERO. 2004. New mitochondria! DNA data affirm the importance of Pleistocene speciation in North American birds. Evolution 58:1122-1130.

KIDD, M. G., AND V. L. FRIESEN. 1998. Sequence variation in the Guillemot (Alcidae: Cepphus) mitochondrial control region and its nuclear homolog. Molecular Biology and Evolution 15:61-70.

KRATTER, A. W., D. W. STEADMAN, C. E. SMITH, AND C. E. FILARDI. 2001a. Reproductive condition, moult, and body mass of birds from Isabel, Solomon Islands. Bulletin of the British Ornithology Club 121:128-144.

KRATTER, A. W., D. W. STEADMAN, C. E. SMITH, C. E. FILARDI, AND H. P. WEBB. 2001b. Avifauna of a lowland forest site on Isabel, Solomon Islands. Auk 118:472-483.

MACARTHUR, R. H., AND E. O. WILSON. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, New Jersey.

MAYR, E. 1942. Systematics and the Origin of Species. Columbia University Press, New York.

MAYR, E. 1945. Birds of the Southwest Pacific: A Field Guide to the Birds of the Area between Samoa, New Caledonia, and Micronesia. MacMillan Company, New York.

MAYR, E. 1963. Animal Species and Evolution. Belknap Press of Harvard University Press, Cambridge, Massachusetts.

MAYR, E. 1967. Subfamily Pachycephalinae, whistlers or thickheads. Pages 3-51 in Check-list of Birds of the World, vol. 12 (R. A. Paynter, Jr., Ed.). Museum of Comparative Zoology, Harvard University Press, Cambridge, Massachusetts.

MAYR, E., AND J. M. DIAMOND. 2001. The Birds of Northern Melanesia: Speciation, Ecology, and Biogeography. Oxford University Press, New York.

MOORE, W. S. 1995. Inferring phylogenies from mtDNA variation: Mitochondrial-gene trees versus nuclear-gene trees. Evolution 49: 718-726.

PRYCHITKO, T. M., AND W. S. MOORE. 1997. The utility of DNA sequences of an intron from the a-fibrinogen gene in phylogenetic analysis of woodpeckers (Aves: Picidae). Molecular Phylogenetics and Evolution 8:193-204.

SLIKAS, B., I. B. JONES, S. R. DERRICKSON, AND R. C. FLEISCHER. 2000. Phylogenetic relationships of Micronesian white-eyes based on mitochondria! sequence data. Auk 117:355-365.

SMITH, C. E. 2003. Comparative phylogeography of some Solomon Island land birds. Ph.D. dissertation, University of Washington, Seattle. STATTERSFIELD, A. J., M. J. CROSBY, A. J. LONG, AND D. C. WEGE. 1999. Endemic Bird Areas of the World: Priorities for Biodiversity Conservation. BirdLife International, Cambridge, United Kingdom.

SWOFFORD, D. L. 2002. PAUP*: Phylogenetic Analysis Using Parsimony (*and Other Methods), version 4.0b10. Sinauer Associates, Sunderland, Massachusetts.

Associate Editor. R. C. Fleischer

CATHERINE E. SMITH1 AND CHRISTOPHER E. FILARDI2

University of Washington Burke Museum and Department of Biology, Box 353020, Seattle, Washington 98105, USA

1 Present address: 205 Brooks Street, Missoula, Montana 59801, USA.

2 Address correspondence to this author. Present address: Center for Biodiversity and Conservation, American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024, USA. E-mail: filardi@amnh.org

Copyright American Ornithologists' Union Apr 2007

(c) 2007 Auk, The. Provided by ProQuest Information and Learning. All rights Reserved.