Molecular Phylogenetic Relationships of Xiphidiopicus Percussus, Melanerpes, and Sphyrapicus (Aves: Picidae) Based on Cytochrome B Sequence

Home , Dendropicos, Hispaniolan woodpecker, Melanerpes, Pici (taxon), Picoides, Piculus

Molecular Phylogenetics and Evolution 41 (2006) 288–294 www.elsevier.com/locate/ympev

Molecular phylogenetic relationships of Xiphidiopicus percussus, Melanerpes, and Sphyrapicus (Aves: Picidae) based on cytochrome b sequence

Lowell C. Overton a,¤, Douglas D. Rhoads b

a Department of Biological Sciences, University of Windsor, Windsor, Ont., Canada N9B 3P4 b Department of Biological Sciences, University of Arkansas, Fayetteville, AR 72701, USA

Received 25 November 2005; revised 10 May 2006; accepted 15 May 2006 Available online 22 May 2006

Abstract

The endemic woodpecker, Xiphidiopicus percussus, from Cuba has been postulated as the sister taxon to the Hispaniolan woodpecker (Melanerpes striatus) and its relationships to the genera Sphyrapicus and Melanerpes have been speculated. We used mitochondrial cytochrome b sequences from a collection of New World picids to investigate the phylogenetic relationships among these species using maximum parsimony and maximum likelihood approaches. Our data suggest that X. percussus is the sister taxon to the Melanerpes woodpeckers, which appear to group into a single distinct clade. Xiphidiopicus percussus is not the sister taxon to M. striatus as has been postulated [Olson, S., 1972. The generic distinction of the Hispaniolan Woodpecker, Chryserpes striatus (Aves: Picidae). Proc. Biol. Soc. Wash. 85, 499–508]. The genus Sphyrapicus appears to have diverged earlier than Xiphidiopicus. Divergence estimates from the cytochrome b sequences indicate that Xiphidiopicus probably diverged sometime in the late Miocene-early Pliocene, and the endemic contem- porary species X. percussus on Cuba may be a relict from a group that originated in Central America or North America. © 2006 Elsevier Inc. All rights reserved.

Keywords: Cytochrome b; Xiphidiopicus percussus; Cuba; Endemic

1. Introduction ulets are the sister group to the woodpeckers and the wrynecks are the sister group to the piculet–woodpecker clade The Order Piciformes comprises the Jacamars (Galbuli- (Sweirczewski and Raikow, 1981; Short, 1982; Burton, dae), PuVbirds (Bucconidae), Barbets (Capitonidae), Tou- 1984). cans (Rhamphastidae), Honeyguides (Indicatoridae), and The monophyly of woodpeckers and their sister status to true woodpeckers (Picidae) (Short, 1982). The family Pici- piculets have never been seriously disputed. However, the dae consists of the subfamilies Picinae (woodpeckers), Jyn- placement of certain genera and species within the Picinae ginae (wrynecks), and the Picuminae (piculets). has not yet been established. One explanation given for the Woodpeckers comprise approximately 218 species within poorly resolved classiWcation of many woodpecker genera 23 genera in the subfamily Picinae and are fairly cosmopol- is rapid diversiWcation of the basal picinae lineages from an itan in distribution (except Australia and Antarctica). The early common ancestor may have obscured the develop- accepted relationships among the three subfamilies are: pic- ment of synapomorphies during incipient diVerentiation (Webb and Moore, 2005). Mitochondrial DNA analyses from DeWllipis and Moore (2000); Prychitko and Moore * Corresponding author. Present address: Institute of Environmental (2000), Weibel and Moore (2002a,b), and Webb and Moore Health Sciences, Wayne State University, 2727 Second Avenue, Detroit, (2005) have provided a substantial database of DNA MI 48201, USA. E-mail addresses: [email protected], [email protected] sequences for many woodpecker species and genera, much (L.C. Overton). of which is in disagreement with Short’s (1982) taxonomic

1055-7903/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2006.05.014 L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294 289 scheme for picids, and Olson’s (1983) polyphyletic argu- proven useful in a number of studies involving taxa of vary- ment for woodpeckers. The molecular evidence from these ing evolutionary depths, from population level analyses studies shows that woodpeckers (excluding barbets, jaca- (Smith and Patton, 1991; Patton and Smith, 1994) to stud- mars, toucans, and honeyguides) comprise a monophyletic ies of much older (>50 mya) divergences (Meyer and Wil- group, with Melanerpes and Sphyrapicus as sister genera, son, 1990). As woodpeckers are considered to have and Colaptes, Dryopcopus, Piculus, Geocolaptes, Campephi- diverged relatively rapidly over short periods of time (DeWl- lus, Celeus, and Chrysocolaptes forming a separate distinct lipis and Moore, 2000; Moore and DeFilippis, 1997; Webb group closest to a Melanerpes–Sphyrapicus–Picoides–Venil- and Moore, 2005), and since nucleotide sequence from the iornis–Dendropicos subclade. cytochrome b gene has been used successfully to examine Within the picid woodpeckers, however, the placement the phylogenetic relationships for most piciforms (DeWllipis of the monotypic species Xiphidiopicus percussus from and Moore, 2000; Moore and DeFilippis, 1997; Prychitko Cuba has been of question for some time. The species has and Moore, 2000; Weibel and Moore, 2002a,b; Webb and been thought to be nearest Melanerpes and Sphyrapicus, Moore, 2005) and other vertebrate species (e.g., Irwin et al., although some speculation regarding where X. percussus 1991; Graybeal, 1993), we felt that the cytochrome b gene sits relative to both of those genera exists. Short (1974) would be an appropriate genetic marker for this study. thought it showed some resemblance to Melanerpes formi- civorous and some species of Sphyrapicus, and may have 2. Materials and methods diverged earlier from a group that gave rise to Melanerpes (including M. striatus from Hispaniola). However, it has 2.1. Taxon sampling been suggested that plumage convergence among woodpeckers has caused errors in generic-level classiWcations Species chosen for inclusion in the phylogenetic analyses (e.g., Veniliornis and Piculus) (Webb and Moore, 2005). As were from the Melanerpes, Sphyrapicus, Picoides, Venilior- the Melanerpes woodpeckers are the closest relatives to the nis, and Dendropicos genera. The following species were sapsuckers (Sphyrapicus) (DeWllipis and Moore, 2000; sampled for this study: Sphyrapicus varius, M. striatus, Mel- Moore and DeFilippis, 1997; Weibel and Moore, 2002a,b; anerpes carolinus, Melanerpes supercilliaris, Melanerpes por- Prychitko and Moore, 2000; and Webb and Moore, 2005), toricensis, and X. percussus (GenBank Accession Nos.: it’s possible that X. percussus is a relict that shares common AF441649–AF441659). Sequence data from a minimum of ancestry to Melanerpes; however, no analysis to address two individuals per species was chosen for reproducibility, this hypothesis has ever been done. with the exception of X. percussus for which only one indi- Olson (1972) proposed that X. percussus and M. striatus vidual was used. Samples (blood or tissue) of Melanerpes, (previously Chryserpes striatus) from Hispaniola were Sphryapicus, and Xiphiodiopicus were obtained from Weld probably sister species based on osteological evidence. He collections by the Wrst author, collaborators, or from tissue felt that M. striatus was too distinct to be considered close collections at the Museum of Vertebrate Zoology (UC to any other Melanerpes species and as such suggested plac- Berkeley) and Louisiana State University Museum of Nat- ing M. striatus nearest Xiphidiopicus, indicating the possi- ural Science. To increase the phylogenetic robustness and bility of M. striatus not being a member of Melanerpes. statistical power of our tree topologies we included the Olson (1972) argued that X. percussus and M. striatus be cytochrome b sequences from GenBank of Veniliornis call- removed from a “melanerpine” grouping and placed near onotus (AY942892), Veniliornis nigriceps (AY942893), Pico- the African and Neotropical members of the “campether- ides albolarvatus (AY942887), Dendropicos griseocephalus ine–colaptine” assemblage (Campethera, Geocolaptes, Den- (AY942884), Dendropicus fuscescens (AY942883) from dropicos, Colaptes, Piculus, and Picoides) (Olson, 1972). Webb and Moore (2005), and Picoides pubescens Olson (1972), however, did not perform any phylogenetic (AF389324) from Weibel and Moore (2002a). All of these reconstruction of the characters he used to propose these species are the closest relatives to Melanerpes and Sphyrapi- relationships. cus, so we expected that their inclusion would provide ade- Although the evolutionary placement of X. percussus quate resolution of the phylogenetic position of X. has been the subject of speculation for some time, no percussus. We rooted our trees with Veniliornis callonotus molecular analysis of this taxon and its relationships to (Webb and Moore, 2005) to evaluate the Xiphidiopicus– other woodpecker genera has ever been reported. We used Melanerpes–Sphyrapicus relationships. nucleotide sequence data from the mitochondrial gene cytochrome b to assess the phylogenetic placement of X. 2.2. DNA sequencing percussus relative to other species of the woodpeckers (including M. striatus) within the subfamily Picinae. The Genomic DNA was extracted from tissue or blood sam- cytochrome b gene codes for the central subunit of the ubi- ples by lysis in 0.5% SDS, phenol:chloroform:isoamyl alco- quinol: cytochrome c reductase (bc1 complex) present hol (25:24:1) extractions, and ethanol precipitation. within the inner mitochondrial membrane as part of com- Extracts were then digested with 100 g/ml RNAse A at plex III of the electron transport chain (Hauska et al., 1983; 37 °C for 24 h, followed by ethanol precipitation. DNAs Howell, 1989; Degli Eposti et al., 1993). The gene has were dissolved in 10 mM Tris–Cl, 1 mM EDTA (pH 7.5) 290 L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294 and DNA concentrations were measured by Xurorometry polyacrylamide buVer gradient gel for 60,000 V h. The gel (TKO, Hoefer). All DNAs were diluted to 50–100 ng/l for was transferred to electrophoresis paper and X-OMAT X- use. ray Wlm placed on the gel. Exposures were done at ¡70 °C The majority of the cytochrome b gene (1080 bp of for 24–40 h depending on signal intensity before the autora- 1143 bp) was ampliWed via the polymerase chain reaction diograph was developed. (PCR) using primers described by Kocher et al. (1989) and Double stranded amplicons of cytochrome b were Helm-Bychowski and Cracraft (1993): L14841 (5Ј-AAAA sequenced for both H- and L-strands with overlap of 50– AGCTTCCATCCAACATCTCAGCATGATGAAA-3Ј), 100 bp. Sequence reads from both strands were assembled L14990 (5Ј CCATCCAACATCTCAGCATGATGAAA-3Ј), and then the consensus L-strand for each species used in L15311 (5Ј-CTACCATGAGGACAAATATC-3Ј), “H15656” the phylogenetic analyses. (5Ј-TCTGGGTCTCCTAGTAGGTT-3Ј) (reverse comple- ment of L15656), L15656 (5Ј-AACCTACTAGGAGACCC 2.3. Sequence alignments AGA-3Ј), and H16065 (5Ј-GAGTCTTC AGTCTCTGGT TTACAAGAC-3Ј). We used 1022 bp of this sequence The concatenated sequences were initially aligned by eye for alignment with the GenBank sequences of cytochrome (GeneDoc, ver. 2.6.001, Nicolas et al., 1997) then aligned b from Webb and Moore (2005). using the default settings in Clustal 1.8X to insure accurate Polymerase chain reaction ampliWcations (25 l) were positional homology. One ingroup taxon (M. supercilliaris) carried out in an AirThermo Cycler (Idaho Technologies, was used as a reference sequence for both manual and com- Inc.) with the following parameters: initial denaturation, puter-generated alignments. 90 °C, 15 s; then for 33 cycles, denaturation, 90 °C, 15 s; annealing, 45 °C, 15 s; and extension, 72 °C, 1 min; fol- 2.4. Phylogeny reconstruction lowed by a Wnal 5 min extension at 72 °C. Each reaction consisted of: 1X Mg2+ Free Taq Polymerase BuVer We used likelihood and parsimony approaches to (Promega Corp.), 1.5 mM MgCl2, 1 M of either forward assess the phylogenetic position of X. percussus. A maxi- or reverse primers, 4U Taq Polymerase, 0.5 g of BSA, mum likelihood analysis employing a Bayesian strategy and 50–100 ng genomic DNA. AmpliWcation reactions was computed using Mr. Bayes, ver. 3.0 (Ronquist and were concentrated to 10 l under vacuum and electropho- Huelsenbeck, 2003). A series of Monte Carlo Markov resed in 0.7% agarose (Low EEO, Fisher ScientiWc) for Chains (MCMC) were simulated for 200,000 generations 72 V h. AmpliWed fragments were puriWed using a modi- and sampled every 100 generations; four chains were run Wed GlassMilk procedure (Bio101, LaJolla): the appropri- and 25% of the initial trees were discarded (“burnin” ate-sized fragment was excised from the gel, melted at option). The program MODELTEST (ver. 3.06, Posada 65 °C in 5 M NaI solution (Sigma–Aldrich), and silica and Crandall, 1998) was employed to determine the glassmilk (Sigma–Aldrich) was used to bind DNA. The appropriate substitutional model for the likelihood analy- glassmilk-bound DNA was then puriWed using three ses of the cytochrome b sequences. For comparison to the washes of a 25 mM NaCl, 1:1 95% Ethanol double-dis- substitutional model and parameters of the likelihood tilled water solution. PuriWed cytochrome b amplicons results, a maximum parsimony analysis (PAUP, ver were eluted from the silica using double-distilled water 4.0b10, SwoVord, 2000) was also performed. In the event (5 min at 65 °C) and by spinning the samples at 3000 rpm of multiple-equally parsimonious trees, a strict consensus for 1 min. approach was used. Node strength for the parsimony Sequencing reaction mixtures (10 l) consisted of: 2X analysis was assessed using the bootstrap technique. We Taq DNA Polymerase BuVer (Promega Corp.), 1 mM used the approximately unbiased Shimodaira and Hase- W V MgCl2, 5 M dNTPs, 4U Taq Polymerase (Promega Corp.), gawa test (Shimodaira, 2002) to test for signi cant di er- 32 1 M primer, and 125 mCI [ - P]dATP (Amersham-Phar- ences among any alternative tree topologies (e.g., Olson’s, macia). Termination mixtures were determined empirically 1972 hypothesis) compared to our likelihood tree. (Overton and Rhoads, unpublished data) and consisted of the following components: 30 M ddGTP, 5 M dNTPs; 3. Results 600 M ddATP, 5 M dNTPs; 900 M ddTTP, 5 M dNTPs; 300 M ddCTP, 5 M dNTPs. Reactions were We used 1022-bp of the mitochondrial cytochrome b cycled in an Air Thermo-cycler (Idaho Technologies, Inc.) gene to investigate the placement of X. percussus within the programmed for: initial denaturation 90 °C, 30 s, followed woodpecker subfamily Picinae and to test its relationship to by cycling for 30 cycles of denaturation for 90 °C, 30 s; the Melanerpes woodpeckers, Sphyrapicus sapsuckers, and annealing at 45 °C, 30 s; and extension at 72 °C, 1 min. Four M. striatus, a proposed sister taxon (Olson, 1972). microliters of “stop solution” (0.1% bromophenol blue/ Sequences from a total of 20 individuals from 11 wood- xylene cyanol, 20 mM EDTA, 95% formamide) was added pecker species were included in our analyses; sequences for upon completion of the reactions. Samples were heat dena- six of these species were obtained from the GenBank cyto- tured at 90 °C, 2 min and snap cooled on ice. Three microli- chrome b sequences from Webb and Moore (2005) and ters of each reaction were electrophoresed through a 6% Weibel and Moore (2002a). L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294 291

3.1. Sequence composition supported by woodpecker fossil evidence dating from the early Pliocene (approximately 5 mya; Short, 1982; Feduccia Among the 1022 nucleotide sites (nt) utilized for the and Wilson, 1967; Cracraft and Morony, 1969). phylogenetic analyses, 452 nt were variable, and 299 nt were Parsimony analysis yielded a single minimum-length tree phylogenetically informative. Considerable among-site rate of 839 steps (Fig. 2) with a topology almost identical to that variability existed for all sequences ( D 0.29414). Base com- of the Bayesian tree. We also found well-supported nodes position analyses show that the sequences are comparable (>90% support) within the parsimony tree for X. percussus to cytochrome b sequences from other vertebrates with being the sister group to Melanerpes (94%), and Sphyrapi- 25.7% A, 35% C, 26.4% T, and 12.8% G. Xiphidiopicus per- cus being the next closest group after X. percussus (91%). cussus had similar percentages to the other species for all The bootstrap results from the parsimony analysis show four bases: A: 25.8%, C: 35.1%, G: 11.5%, and T: 27.5%. good support for the tree’s nodes, all of which are >70% First, second, and third codon positions for all taxa show and therefore likely to be accurate (Hillis and Bull, 1993). dissimilar percentages of nucleotides, especially at third We tested Olson’s (1972) hypothesis of X. percussus being positions with a lack of G (3.3%) and an abundance of C the sister species to M. striatus, although our trees strongly (49%). Second positions had an overabundance of T suggest that such a relationship is unlikely. If X. percussus (40.4%) and a low percentage of G (13.3%). First positions and M. striatus are constrained to share a common ances- had similar percentages for all four bases (A: 25.5%, C: tor, the likelihood of the resulting tree is signiWcantly lower 29.5%, T: 23.1%, G: 21.9%). than that of the tree we obtained in our analyses (lnL D¡5042.45860, compared to lnL D¡4992.71; Shimo- 3.2. Tree estimation daira–Hasegawa test: 2 D 126.51 P D 0.000, one-tailed test).

MODELTEST (ver. 3.06, Posada and Crandall, 1998) 4. Discussion revealed that the general time-reversible model with a dis- crete gamma distribution (six rate categories for the dis- The woodpeckers comprise one subfamily (Picinae) crete approximation of the gamma distribution, Yang and among three within the Family Picidae (true woodpeckers). Kumar, 1996) of independent changes (GTR+G+I) was the The New World origin for many members of the Family appropriate substitutional model for the likelihood Picidae are supported by diverse sets of data: morphology searches (lnL: ¡4989.703, AIC: 10037.405). and behavior (Sweirczewski and Raikow, 1981; Short, A single likelihood tree was produced from the Bayesian 1982), and molecular (Moore, 1995; Moore and DeFilippis, analysis (Fig. 1). The majority of nodes within the topology 1997; and DeWllipis and Moore, 2000). Short (1982) divided are supported by posterior probabilities of greater than the Picinae subfamily into six distinct tribes: Melanerpini 95%. The likelihood analysis strongly supports (98%) X. (Melanerpes), Campetherini (Picoides), Colaptini (Colaptes, percussus as the sister group to the Melanerpes clade, and Piculus, and Veniliornis), and Campephilini (Campephilus Sphyrapicus as the next closest group after X. percussus and Dryocopus). The placement of X. percussus among (100%). The Melanerpes species, including M. striatus, were these tribes, however, has remained unclear. Olson (1972) found to cluster into a distinct group, suggesting mono- argued for a sister-taxa relationship between X. percussus phyly for the genus. The patterns found among the other and M. striatus, which would place X. percussus within the woodpecker species within the topology are exactly the Melanerpini tribe. same as those relationships found by Webb and Moore The results of our sequence analyses presented in this (2005). We found that imposing a molecular clock on the study indicate that the Cuban endemic X. percussus is the data did not change the taxon relationships (data not sister taxon to the Melanerpini with Sphyrapicus being next shown), although the cytochrome b sequences were found most closely related to Melanerpes after X. percussus. This not to evolve in clocklike fashion (2 D 48.34, df D 9, result was consistent across parsimony and likelihood anal- P < 0.001). The branch lengths (with and without a clock yses, and suggests that the genus Xiphidiopicus may have a enforced) for X. percussus from the likelihood analysis were North American or Central American origin which is more 0.203877 (non-clock) and 0.163255 (clock). These distances recent than the split that gave rise to Sphyrapicus. Xiphidi- were used only as a heuristic estimate of divergence time opicus percussus, however, is not closely related to M. stria- between Xiphidiopicus and Melanerpes and suggest that the tus, contrary to the suggestion by Olson (1972). Our results split between Xiphidiopicus and a Melanerpes ancestor may also suggest that the Melanerpes genus may comprise a have occurred approximately 7–10 mya (Late Miocene- monophyletic group. early Pliocene) using an average rate of 2% per million The relatively large genetic distance between Xiphidiopi- years for mtDNA in birds (Shields and Wilson, 1987; Tarr cus and Melanerpes suggests that the split between these and Fleischer, 1993). Although the cytochrome b sequences two genera is older than the Pleistocene and could be as old were found not to evolve in clocklike fashion (typical for as late Miocene. It is therefore possible that an ancestor of most mitochondrial sequences), these distances are in line X. percussus colonized Cuba from either North America or with the suggestion that the diversiWcation of woodpeckers northern Central America sometime pre-Pleistocene, as began approximately 7–8 mya (Moore et al., 1999) and is suggested by Bond’s (1963, 1966) hypothesis that the 292 L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294

Melanerpes striatus 100

100 Melanerpes portoricensis

Melanerpes 98 Melanerpes carolinus

98 Melanerpes supercilliaris

100 Xiphidiopicus percussus Xiphidiopicus

Sphyrapicus varius Sphyrapicus

Dendropicos fuscescens

100

Dendropicos griseocephalus

Picoides albolarvatus

Picoides pubescens

Venilornis callonotus 0.1

Fig. 1. Phylogram produced from the likelihood analysis (lnL D¡4992.71) using the GTR+G+I algorithm with a Bayesian approach. Scale bar indicates the number of substitutions/site for the tree. Numbers above branches indicate the posterior probability from a 200,000 generation Monte Carlo Markov Chain simulation.

Caribbean avifauna followed colonization routes from ted diversiWcation of new species (Ricklefs and Berming- those two landmasses. ham, 2001). Other studies have suggested, however, that a Because of the east-to-west movement of ocean currents Central American or North American dispersal route and hurricanes in the eastern Caribbean, it has been sug- existed for other West Indian avian lineages (Seutin et al., gested that most West Indian terrestrial groups have their 1995; Overton, 2001). Hence, it is plausible that Xiphidiopi- closest relatives within South America given the water cur- cus could have been derived from a North American or rents and hurricane tracks (Hedges, 1996b). This view is Central American ancestor, and colonized Cuba from contrary to Bond’s (1963, 1966) assertion that the avian either the northern portion of Central America or south- species within the region were principally derived from eastern North America to reach that island. The lack of fos- Central American or North American groups. Some studies sil evidence for the genus, however, leaves unresolved the have suggested that certain avian groups within the region most likely phylogeographic pathway for the taxon. appear to have been derived from either an eastern South In summary, our sequence analyses show that the American pathway through the Lesser Antilles, or from endemic Cuban woodpecker X. percussus is the sister taxon inter-island radiations within the West Indies which permit- to the Melanerpes woodpeckers. This result is consistent L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294 293

Melanerpes supercilliaris 92

Melanerpes carolinus 94 Melanerpes

Melanerpes striatus 100 91

Melanerpes portoricensis

88 Xiphidiopicus percussus Xiphidiopicus

82 Sphyrapicus varius Sphyrapicus

Dendropicos fuscescens 100

Dendropicos griseocephalus

Picoides albolarvatus

Picoides pubescens

Veniliornis callonotus

Fig. 2. The most-parsimonious tree found from the parsimony analysis using Veniliornis as the root from the cytochrome b sequences (TL D 839 steps, CI D 0.642, HI D 0.358). The topological relationships are in agreement with the likelihood tree. Numbers above the branches reXect the percentage a node occurs for a 2000 replicate bootstrap analysis (Hedges, 1991). across tree reconstruction methods and supports the notion F. Sheldon, and J.V. Remsen of the Louisiana State Univer- that this Cuban endemic is New World in origin. Given the sity Museum of Natural Science for kindly contributing tis- extent of sequence divergence between Xiphidiopicus and sue samples of M. portoricensis (B-11314, B-11475). The Melanerpes, it is possible that divergence may have Ministry of Agriculture and Fisheries of the Bahamas occurred sometime late Miocene to early Pliocene. The bird kindly granted the appropriate collecting permit (permit# also does not appear to be the sister taxon to the wood- 96/035) to the Wrst author to collect blood samples of the pecker species M. striatus from Hispaniola as Olson (1972) Abaco form of M. supercilliaris. U.S.D.A. granted the suggested. The species, therefore, can be considered a relic. appropriate import permit (permit #43909) to the Wrst author for importation of the biological samples from Acknowledgments exemplars for M. supercilliaris and X. percussus. We also thank A. Town Peterson and three anonymous reviewers We thank N.K. Johnson and C. Cicero of the Museum for helpful comments on previous versions of this manu- of Vertebrate Zoology, UC Berkeley for their generosity of script. Funding was provided to the Wrst author from Sigma donating tissue samples from M. striatus (MVZ#81649, Xi, the ScientiWc Research Society, Frank M. Chapman 81650). Pascal Villard also contributed an M. striatus sam- Memorial Fund, American Museum of Natural History, ple for this work. G. Wallace and A. Kirkconnell of the David Causey Grant-In-Aid, and Doctoral Dissertation Museo Nacional de la Habana, Cuba contributed samples Research Award from the Fulbright College of Arts and of X. percussus, M. supercillaris. We also thank D. Dittman, Sciences, University of Arkansas, Fayetteville. 294 L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294

References (Eds.), Proceedings of the 22nd International Ornithology Congress. University of Natal, Durban, South Africa, pp. 745–753. Bond, J., 1963. Derivation of the Antillean avifauna. Proc. Acad. Nat. Sci. Nicolas, K.B., Nicholas Jr., H.B., DeerWeld II, D.W., 1997. GeneDoc: anal- Phil. 115, 79–98. ysis and visualization of genetic variation. EMBNEW. NEWS 4, 14. Bond, J., 1966. AYnities of the Antillean avifauna. Carib. J. Sci. 6 (3-4), Olson, S., 1972. The generic distinction of the Hispaniolan Woodpecker, 173–176. Chryserpes striatus (Aves: Picidae). Proc. Biol. Soc. Wash. 85, 499–508. Burton, P.J.K., 1984. Anatomy and evolution of the feeding apparatus in Olson, S., 1983. Evidence for a polyphyletic origin of the Piciformes. Auk the avian orders Coraciiformes and Piciformes. Bull. Br. Mus. Nat. 100, 126–133. Hist. Zool. 47, 331–443. Overton, L.C., 2001. Molecular evolutionary biogeography of some Cen- Cracraft, J., Morony, J.J., 1969. A new Pliocene woodpecker, with com- tral American-derived Greater Antillean bird lineages: a mitochondrial ments on the fossil Picidae. Am. Mus. Novit. 2400, 1–18. gene/single-copy nuclear gene approach. Ph.D. dissertation, University DeWllipis, V.R., Moore, W.S., 2000. Resolution of phylogenetic relation- of Arkansas, 185pp. ships among recently evolved species as a function of amount of DNA Patton, J.L., Smith, M.F., 1994. Paraphyly, polyphyly, and the nature of sequence: an empirical study based on woodpeckers (Aves:Picidae). species boundaries in pocket gophers (Genus Thomomys). Syst. Biol. Mol. Phylogenet. Evol. 16, 143–160. 43, 11–26. Degli Eposti, M., DeVries, S., Crimi, M., Ghelli, A., Patarnello, T., Meyer, Posada, D., Crandall, K.A., 1998. MODELTEST: testing the model of A., 1993. Mitochondrial cytochrome b: evolution and structure of the DNA substitution. Bioinformatics 14 (9), 817–818. protein. Biochemica et Biophysica Acta 1143, 243–271. Prychitko, T.M., Moore, W.S., 2000. Comparative evolution of the mito- Feduccia, A., Wilson, R.L., 1967. Avian fossils from the lower Pliocene of chondrial cytochrome b gene and nuclear -Wbrinogen intron 7 in Kansas. Occas. Pap. Mus. Zool. Univ. Mich. 655, 1–6. woodpeckers. Mol. Biol. Evol. 17, 1101–1111. Graybeal, A., 1993. The phylogenetic utility of cytochrome b: lessons from Ricklefs, R.E., Bermingham, E., 2001. Nonequilibrium diversity dynamics bufonid frogs. Mol. Phylogenet. Evol. 2, 256–269. of the Lesser Antillean avifauna. Science 294 (5546), 1522–1524. Hauska, G., Hurt, E., Gabellini, N., Lockau, W., 1983. Comparative Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes3: bayesion phylogenetic aspects of quinol-cytochrome c/plastocyanin oxidoreductases. Bioche- inference under mixed models. Bioinformatics 19, 1572–1574. mica Biophysica Acta 726, 97–133. Seutin, G., Klein, N.K., Ricklefs, R.E., Bermingham, E., 1995. Historical Hedges, S.B., 1991. The number of replications needed for accurate estima- biogeography of the Bananaquit (Coerebra Xaveola) in the Caribbean tion of the bootstrap P value in phylogenetic studies. Mol. Biol. Evol. 8, region: a mitochondrial DNA assessment. Evolution 43, 1041–1061. 366–369. Shields, G.F., Wilson, A.C., 1987. Calibration of mitochondrial DNA evo- Hedges, S.B., 1996b. Historical biogeography of West Indian vertebrates. lution in geese. J. Mol. Evol. 24, 212–217. Annu. Rev. Ecol. Syst. 27, 163–196. Shimodaira, H., 2002. An approximately unbiased test of phylogenetic tree Helm-Bychowski, K., Cracraft, J., 1993. Recovering phylogenetic signal selection. Syst. Biol. 51, 492–508. from DNA sequences: relationships within the Corvine assemblage Short, L., 1974. Habits of three endemic West Indian woodpeckers (Aves, (Class Aves) as inferred from complete sequences of the mitochondrial Picidae). Am. Mus. Novitates. 2549, 1–44. DNA cytochrome b gene. Mol. Biol. Evol. 10, 1196–1214. Short, L., 1982. Woodpeckers of the World. Delaware Museum of Natural Hillis, D.M., Bull, J.J., 1993. An empirical test of bootstrapping as a History. Greenville, Delaware. 676pp. method for assessing conWdence in phylogenetic analysis. Syst. Biol. 42, Smith, M.F., Patton, J.L., 1991. Variation in mitochondrial cytochrome b 182–192. sequence in natural populations of South American akodontine Howell, N., 1989. Evolutionary conservation of protein regions in the pro- rodents (Muridae: Signodontinae). Mol. Biol. Evol. 8, 85–103. tonmotive cytochrome b and their possible roles in redox catalysis. J. Sweirczewski, E.V., Raikow, R.J., 1981. Hindlimb morphology, phylogeny, Mol. Evol. 29, 157–169. and classiWcation of the piciformes. Auk 98, 466–480. Irwin, D.M., Kocher, T.D., Wilson, A.C., 1991. The evolution of cyto- SwoVord, D.L., 2000. PAUP*: Phylogenetic Analysis using Parsimony, chrome b gene of mammals. J. Mol. Evol. 32, 128–144. ver. 4.0. Sinauer Assoc., Sunderland, MA. Kocher, T.D., Thomas, W.K., Meyer, A., Edwards, S.V., Paabo, S., Villa- Tarr, C.L., Fleischer, R.C., 1993. Mitochondrial-DNA variation and evo- blanca, F.X., Wilson, A.C., 1989. Dynamics of mitochondrial DNA lutionary relationships in the Amakihi complex. Auk 110, 825–831. evolution in animals: ampliWcation and sequencing with conserved Webb, D.M., Moore, W.S., 2005. A phylogenetic analysis of woodpeckers primers. Proc. Natl. Acad. Sci. USA 86, 6196–6200. and their allies using 12S, cyt b, and COI nucleotide sequences (class Meyer, A., Wilson, A.C., 1990. Origin of tetrapods inferred from their Aves; order Piciformes). Mol. Phylogenet. Evol. 36, 233–248. mitochondrial DNA aYliation to lungWsh. J. Mol. Evol. 31, 359–365. Weibel, A.C., Moore, W.S., 2002a. Molecular phylogeny of a cosmopolitan Moore, W.S., 1995. Inferring phylogenies from mtDNA variation: mito- group of woodpeckers (Genus Picoides) based on COI and cyt b mitochondrial gene trees versus nuclear gene trees. Evolution 49, 718–726. chondrial sequences. Mol. Phylogenet. Evol. 22, 65–75. Moore, W.S., DeFilippis, V.R., 1997. The window of taxonomic resolution Weibel, A.C., Moore, W.S., 2002b. A test of a mitochondrial gene based for phylogenies based on mitochondrial cytochrome b. In: Mindell, phylogeny of woodpeckers (Genus Picoides) using an independent D.P. (Ed.), Avian Molecular Evolution and Systematics. Academic nuclear gene, -Wbrinogen intron 7. Mol. Phylogenet. Evol. 22, 247–257. Press, San Diego, California, pp. 83–119. Yang, Z., Kumar, S., 1996. Approximate methods for estimating the pat- Moore, W.S., Smith, S.M., Prychitko, T., 1999. Nuclear gene introns versus tern of nucleotide substitution and the variation in substitution rates mitochondrial genes as molecular clocks. In: Adams, N., Slotow, R. among sites. Mol. Biol. Evol. 13, 650–659.