Evolutionary and ecological forces influencing population ...lithornis.nmsu.edu/~phoude/Chua et al 2017 Evolutionary...d Sabah Parks, P.O. Box 10626, 88806 Kota Kinabalu, Sabah, Malaysia

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/317152138

Evolutionary and ecological forces influencing population diversification in

Bornean montane passerines

Article  in  Molecular Phylogenetics and Evolution · May 2017

DOI: 10.1016/j.ympev.2017.05.016

CITATIONS

2READS

318

8 authors, including:

Some of the authors of this publication are also working on these related projects:

Evaluating the evolutionary history and systematics of herons (Ardeidae) using genomic data View project

Modeling nest site selection and fledging success of Amur Falcons (Falco amurensis) in central Mongolia View project

Vivien L. Chua

Louisiana State University

8 PUBLICATIONS   115 CITATIONS   

SEE PROFILE

Brian Tilston Smith

American Museum of Natural History

52 PUBLICATIONS   1,158 CITATIONS   

SEE PROFILE

Ryan C. Burner

Louisiana State University

11 PUBLICATIONS   10 CITATIONS   

SEE PROFILE

Mustafa Abdul Rahman

University College Sabah Foundation

53 PUBLICATIONS   370 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Vivien L. Chua on 08 June 2017.

The user has requested enhancement of the downloaded file.

Molecular Phylogenetics and Evolution 113 (2017) 139–149

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier .com/ locate /ympev

Evolutionary and ecological forces influencing population diversificationin Bornean montane passerines

http://dx.doi.org/10.1016/j.ympev.2017.05.0161055-7903/� 2017 Elsevier Inc. All rights reserved.

⇑ Corresponding author.E-mail address: [email protected] (V.L. Chua).

Vivien L. Chua a,⇑, Brian Tilston Smith b, Ryan C. Burner a, Mustafa Abdul Rahman c, Maklarin Lakim d,Dewi M. Prawiradilaga e, Robert G. Moyle f, Frederick H. Sheldon a

aMuseum of Natural Science and Department of Biological Sciences, Louisiana State University, Baton Rouge, LA 70803, USAbDepartment of Ornithology, American Museum of Natural History, New York 10024, USAc Faculty of Resource Science and Technology, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysiad Sabah Parks, P.O. Box 10626, 88806 Kota Kinabalu, Sabah, MalaysiaeDivision of Zoology, Research Centre for Biology-LIPI, Jl. Raya Bogor km. 46, Cibinong-Bogor 16911, IndonesiafBiodiversity Institute and Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, KS 66045, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 27 September 2016Revised 28 April 2017Accepted 16 May 2017Available online 22 May 2017

Keywords:BorneoComparative phylogeographyDispersalHand-wing indexND2Population structure

The mountains of Borneo are well known for their high endemicity and historical role in preservingSoutheast Asian rainforest biodiversity, but the diversification of populations inhabiting these mountainsis poorly studied. Here we examine the genetic structure of 12 Bornean montane passerines by compar-ing complete mtDNA ND2 gene sequences of populations spanning the island. Maximum likelihood andBayesian phylogenetic trees and haplotype networks are examined for common patterns that might sig-nal important historical events or boundaries to dispersal. Morphological and ecological characteristics ofeach species are also examined using phylogenetic generalized least-squares (PGLS) for correlation withpopulation structure. Populations in only four of the 12 species are subdivided into distinct clades or hap-lotype groups. Although this subdivision occurred at about the same time in each species (ca. 0.6–0.7 Ma), the spatial positioning of the genetic break differs among the species. In two species, northeast-ern populations are genetically divergent from populations elsewhere on the island. In the other two spe-cies, populations in the main Bornean mountain chain, including the northeast, are distinct from those ontwo isolated peaks in northwestern Borneo. We suggest different historical forces played a role in shapingthese two distributions, despite commonality in timing. PGLS analysis showed that only a singlecharacteristic—hand-wing index—is correlated with population structure. Birds with longer wings, andhence potentially more dispersal power, have less population structure. To understand historical forcesinfluencing montane population structure on Borneo, future studies must compare populations acrossthe entirety of Sundaland.

� 2017 Elsevier Inc. All rights reserved.

1. Introduction

In contrast to the Neotropics and Africa, where comparativephylogeographic investigations of Pleistocene refuges, centers-of-endemism, and dispersal patterns of animals (Cracraft and Prum,1988; Fjeldså and Bowie, 2008; Haffer, 1987) are common (e.g.,Burney and Brumfield, 2009; Ditchfield, 2000; Huhndorf et al.,2007; Smith et al., 2014), such studies in Sundaland—the biogeo-graphic region of Southeast Asia encompassing the Sunda conti-nental shelf and its current land (Malay Peninsula, Sumatra, Java,Borneo, Palawan and small islands)—are rare. Only a handful exist

(Demos et al., 2016; Gorog et al., 2004; Lim et al., 2010; Lim andSheldon, 2011). This is surprising, not only because studies in Sun-daland helped inspire the science of Biogeography (Wallace, 1860),but because the region’s potential for comparative phylogeo-graphic study of Pleistocene vicariance, habitat refuges, coloniza-tion, community relaxation, and other modes of diversificationand extinction is substantial and has been recognized for years(Brandon-Jones, 1996; Cracraft, 1988; Diamond et al., 1987;Heaney, 1991).

Among biogeographic features in Sundaland that merit greaterstudy, the mountains of Borneo are paramount because of theimportant role they played in preserving and fomenting SoutheastAsia’s rainforest diversity (de Bruyn et al., 2014; Merckx et al.,2015; Sheldon et al., 2015). Borneo’s main mountain chain runsdiagonally from Sabah in the northeast, along the Sarawak-

140 V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149

Kalimantan border, toward the southwestern side of the island(Fig. 1). By virtue of their magnitude, longevity and position (espe-cially in the northeast), these mountains helped maintain rain-forest continuously throughout much of the Cenozoic, evenduring the cooler, drier times of the Oligocene, Pliocene, and Pleis-tocene, when rainforest was reduced in large areas of Sundaland(Morley, 2012). In respect to birds, Borneo’s mountains not onlyhelped preserve ancient (perhaps Eocene) rainforest lineages, butalso fomented more recent diversification as a result of periodicrainforest isolation in the Pliocene and Pleistocene (Sheldonet al., 2015). In the warm, moist Miocene, when most of southernAsia was tropical (Morley, 2012), rainforest taxa surviving the Oli-gocene appear to have dispersed from Borneo, spreading to theother Sunda islands and mainland Southeast Asia (de Bruynet al., 2014; Price et al., 2014). In the same epoch, taxa from theHimalayas and other mountains in Southeast Asia, including war-blers and babblers, invaded Borneo (Moyle et al., 2012; Päckertet al., 2012). As a result of these dynamics, Borneo’s montane avi-fauna now comprises a rich mixture of geologically ancient ende-

125 250 500 Kilometers

Sarawak

Kaliman

Brune

Fig. 1. Elevation map of Borneo with the Malaysian states of Sabah and Sarawak oupopulations: Padawan, Penrissen, and Pueh. Mid-Bornean populations: Lumaku, MaligaRange, Trus Madi, Meliau, and Tawau. Inset shows the region of Southeast Asia.

mic, middle-aged endemic and non-endemic, and youngerimmigrant taxa (Sheldon et al., 2015).

Although this scenario outlines the general forces responsiblefor assembling Borneo’s montane avifauna, it fails to describe themore subtle dynamics that have shaped the composition and tex-ture of Borneo’s montane populations as they now appear. In truth,we know very little about the structure of Borneo’s montane birdpopulations because almost all of our understanding comes fromsubjective assignment of subspecies based on plumage and sizedifferences among populations separated from one another byhundreds of kilometers. We have no idea what these morphologi-cal differences signify in terms of evolutionary history.

Based on recent geographic events, we would expect little dif-ferentiation among bird populations in Borneo’s mountains, firstbecause montane habitat is largely continuous along the island’smain mountain chain, and second because Borneo’s montane habi-tat was certainly more expansive for most of the last 2.6 millionyears than it is currently (Manthey et al., 2017). During the Pleis-tocene, cooler climate resulting from long-lasting global glaciationevents should have caused Sundaic montane forest to descend in

KinabaluCrocker RangeTrus MadiMeliauTawau HillsLumakuMaliganBruneiBarioMuluPadawanPenrissenPueh

Sabah

tan

i

tlined and sampling sites marked with colored circles. Southwest (SW) Borneann, Brunei, Mulu, and Bario. Northeast (NE) Borneo populations: Kinabalu, Crocker

V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149 141

elevation (Cannon et al., 2009). Some habitat reconstructions of theLast Glacial Maximum (LGM, ca. 21,000 years ago) indicate thatmontane habitat covered much of central Borneo (Manthey et al.,2017), if not virtually the entire island (Cannon et al., 2009). Withmultiple, long lasting glacial events occurring in the last 2 millionyears (Woodruff, 2010), repeated and extended opportunitiesexisted for montane bird populations to expand and intermix.Thus, long-term isolation and consequent diversification isexpected to have occurred only on mountains outside the centralmountain chain that were continuously disconnected by low ele-vation habitat, or among populations restricted to habitat on thechain’s highest peaks (e.g., ericaceous heath scrub).

To date, the molecular population structure of only one mon-tane vertebrate has been examined across the breadth of Borneo:Chlorocharis emiliae, the mountain black-eye (Gawin et al., 2014;Manthey et al., 2017). This bird is a Bornean endemic with a uniquesky-island distribution; it is the only Bornean bird restricted to thetops of tall or isolated peaks across the island. Phylogeographicanalysis and LGM niche modelling indicate that genetic variationamong C. emiliae populations derives mainly from isolation-by-distance and unsuitable habitat surrounding outlying inhabitedmountains (Manthey et al., 2017). Such a pattern of variationmakes sense in a species with a sky-island distribution. Even ifits populations did expand in range during the glacial events, C.emiliae would retract to high mountain tops during interglacialperiods, as it has currently. But what about Borneo’s numerousmontane species that are distributed more broadly at lower eleva-tions than C. emiliae? Several of these are divided into subspeciesbased on plumage differences, even within the largely continuouscentral mountains of Borneo. Do these plumage differences reflectsubstantial genetic variation? If so, do they derive from historicalevents (barriers or invasion) or from ecological limitations to dis-persal? To understand the forces that have influenced populationstructure in Borneo’s montane birds requires a comparative exam-ination of genetics, ecology, and morphology in co-distributed spe-cies: genetics to determine patterns and degree of populationvariation (Arbogast and Kenagy, 2001; Carstens and Richards,2007), and ecology and morphology to understand habitat require-ments and dispersal potential of populations (Burney andBrumfield, 2009; Claramunt et al., 2012; Smith et al., 2014).

In this paper, we compare phylogeographic structure in 12passerine species across the mountains of Borneo using mitochon-drial DNA sequences. The species are as follows (as classified byGill and Donsker, 2016): in Pycnonotidae, ochraceous bulbul (Alo-phoixus ochraceus) and cinereous bulbul (Hemixos cinereus); inDicruridae, ashy drongo (Dicrurus leucophaeus); in Muscicapidae,snowy-browed flycatcher (Ficedula hyperythra) and Bornean whis-tling thrush (Myophonus borneensis); in Pachycephalidae, Borneanwhistler (Pachycephala hypoxantha); in Phylloscopidae, yellow-breasted warbler (Seicerus montis); in Rhipiduridae, white-throated fantail (Rhipidura albicollis); in Timaliidae, Temminck’sbabbler (Pellorneum pyrrogenys) and grey-throated Babbler (Sta-chyris nigriceps); in Leiothrichidae, chestnut-hooded laughingth-rush (Garrulax treacheri); and in Zosteropidae, chestnut-crestedyuhina (Yuhina everetti). These species were selected based mainlyon variability in life history, occurrence in different taxonomicfamilies, and availability of specimens. They represent a range inglobal distribution, from Bornean endemism to widespread occur-rence in Indochina, the Philippines, and Wallacea (Table S1). Someof the species tend to be social (bulbuls, babblers, and white-eyes);most others are solitary. Their preferred foraging sites vary fromlow in the forest (P. pyrrogenys and S. nigriceps) to the canopy(e.g., D. leucophaeus and Seicercus montis); their main foods fromfruit (bulbuls and yuhinas) to small vertebrates (whistlingthrushes) and arthropods (all others); their feeding methods fromflycatching (flycatchers, R. albicollis, and D. leucophaeus) to hover-

gleaning (S. montis); and their habitats from open forest (e.g., D.leucophaeus), to rivers (M. bornensis), to closed forest (all others).

Given the general lack of knowledge about population structurein Bornean montane birds, our study is one of discovery. The speci-fic goals are to: (1) identify common patterns in phylogeographicstructure among the species, which may help us postulate histori-cal events for future testing; (2) identify morphological and ecolog-ical characteristics of the species that may have influenced theirpopulation diversification; and (3) assess the subspecific classifica-tion of the species being compared.

2. Materials and methods

2.1. Phylogenetics and population genetics

We sampled 188 vouchered individuals representing 12 spe-cies, nine families, and 13 localities across the mountains of Malay-sian Borneo (Fig. 1, Table S2). Sites were grouped a priori into threegeneral regions based on their spatial concentration and on theirpotential isolation from one another by lower elevation habitat(Fig. 1): SW Borneo (western Sarawak, including Mts. Pueh, Penris-sen and Padawan), mid-Borneo (Brunei, eastern Sarawak and west-ern Sabah, including the Kelabit Highlands, Meliau Range, and Mts.Mulu and Lumaku), and NE Borneo (including the Mts. Kinabalu,Trus Madi, Meliau, and Tawau Hills). Tissues and sequences(n = 30) of extralimital populations were obtained from museumcollections and GenBank, when available, to provide outgroup per-spective on some of the Bornean populations (see Table S2 for adetailed listing of samples and localities).

We sequenced mitochondrial NADH dehydrogenase subunit 2(ND2). Although comparing a single locus can result in an incorrectgene tree (Edwards and Bensch, 2009; Zwickl, 2006), the rapid rateof mitochondrial DNA evolution renders their sequences useful asfirst-approximation markers of population genetic structure (Zinkand Barrowclough, 2008). We extracted total genomic DNA fromtissue samples using a DNeasy tissue extraction kit (Qiagen, Valen-cia, CA) following the manufacturer’s protocol. We amplified all1041 base pairs (bp) of ND2 using primers L5215 (Hackett, 1996)and HTrpC (Spellman and Klicka, 2006). Amplifications were per-formed in 12.5 lL reactions under the following conditions: denat-uration at 94 �C for 10 min, followed by 40 cycles of 94 �C for 30 s,54 �C for 45 s, and 72 �C for 2 min. This series was followed by a10 min extension at 72 �C and a 4 �C soak. Amplicons were visual-ized using electrophoresis in 1% agarose gel. PCR products weresent to Beckman Coulter Genomics (Danvers MA) for all subse-quent sequencing steps. We manually aligned complementaryND2 strands in Geneious 8.0.5 http://www.geneious.com (Kearseet al., 2012). Sequences were submitted to GenBank: KY547843–KY548031.

To infer population trees, we applied maximum likelihood (ML)methods using Garli 2.01 (Zwickl, 2006) and RAxML 8.1.11(Stamatakis, 2014), and a Bayesian approach using MrBayes 3.2.5(Ronquist and Huelsenbeck, 2003). First, aligned sequences wererun in jModelTest 2 (Darriba et al., 2012; Guindon and Gascuel,2003) to estimate appropriate models of sequence evolution. Thedata were then partitioned by codon positions (1st, 2nd, and 3rdcodon positions separate; 1st and 2nd codon positions together,and 3rd codon position separate; 1st codon position separate,and 2nd and 3rd codon positions together). We ran the partitioneddata in PartitionFinder v1.1.1 (Lanfear et al., 2012, 2014) using theAkaike Information Criterion (AIC) to select the best-fit models forphylogenetic analyses. In Garli 2.01, bootstrap analysis of nodalsupport was performed using 1000 pseudoreplicates. We usedSUMTREES 3.1 (Sukumaran and Holder, 2010) to generate consen-sus trees. In RAxML analyses, the rapid bootstrap algorithm was

142 V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149

used to generate 1000 nonparametric bootstrap replicates. InMrBayes, two simultaneous independent runs were conductedfor 2 million MCMC generations, with trees retained as samplesevery 1000 cycles. The first 10% of trees were discarded as burn-in, and a 50% majority rule consensus tree was generated. We usedTracer v 1.6.0 (Rambaut and Drummond, 2007) to inspect the like-lihood plots and assess convergence in estimated parameters. Mostof the analyses were conducted using the CIPRES Science Gatewaycluster (Miller et al., 2010).

To visualize genetic diversity and geographic relationships, weconstructed median joining networks (Bandelt et al., 1999) ofND2 haplotypes for each species using PopArt (http://popart.otago.ac.nz). We also calculated population genetic summarystatistics, including nucleotide diversity (p), number of segregatingsites (S), and number of parsimony-informative sites. Uncorrectedproportional genetic distances among populations within eachspecies for both complete and pairwise deletion were calculatedusing MEGA6 (Tamura et al., 2011). We examined populationstructure within and among populations of each species (usuallywith at least five individuals each) using analysis of molecular vari-ance (AMOVA) in ARLEQUIN 3.5 (Excoffier and Lischer, 2010). Forthe AMOVA, individuals were divided into the three a priori regionsnoted above: SW Borneo, mid-Borneo, and NE Borneo (Fig. 1).

To calculate divergence times, we used a Bayesian framework inBEAST ver. 1.8.2 (Drummond et al., 2012). Appropriate fossil cali-brations were not available, so we used the ND2 mutation rate of1.25 times the Cytochrome b determined by Smith and Klicka(Smith and Klicka, 2010), with a lognormal distribution(mean = 0.0125 substitutions/site/million years, SD = 0.1), and arelaxed uncorrelated lognormal molecular clock model. The analy-sis was run for 60 million generations, sampling parameters every1000 trees for all species. To check for convergence in parameterestimates, we examined effective sample size (ESS) values in Tracerv.1.6.0 (Rambaut and Drummond, 2007). Burn-in was set at 10%,and a maximum clade credibility tree (MCCT) for each specieswas generated using Tree Annotator v 1.8.2 (Drummond et al.,2012).

2.2. Causes of population structure

Under the assumption that phylogeographic structure amongpopulations has been influenced by dispersal and, hence, gene flow(Gavrilets and Vose, 2005; Kisel and Barraclough, 2010; Mayr,1963), we selected ecological characteristics that might affect dis-persal in each of the 12 species. We then examined the influence ofthese characteristics using phylogenetic generalized least-squares(PGLS) analyses (Martins and Hansen, 1996) in ‘Caper’ (Ormeet al., 2012) and ‘Geiger’ (Harmon et al., 2008) packages in the pro-gram R (Team, 2012). PGLS is a weighted regression model thatincorporates the phylogenetic non-independence of the ecologicaldata by estimating the expected co-variance from the phylogeneticrelationships among species (Symonds and Blomberg, 2014).

The characteristics we compared by PGLS were: abundance, ele-vational range, foraging stratum, foraging guild, and hand-wingindex (Table 1). To produce objective assessments of abundanceand elevational range, we used occupancy modelling (Royle andDorazio, 2008) and extensive survey data from Mt. Mulu, Sarawak(Burner et al., 2016), one of the few Bornean mountains still cov-ered by primary forest from 50 to >2000 m. As an index of abun-dance, we averaged the occupancy probabilities (Psi) for eachspecies from within their core elevational range. Elevational rangelimits for each species included all elevations for which Psi was noless than 10% the max Psi at their most abundance elevation. Forforaging stratum and feeding guild, we assigned categorical valuesinferred from publications and our own experience (Myers, 2009;Phillipps and Phillipps, 2014; Robson, 2000; Sheldon et al., 2001).

Hand-wing index was obtained by measuring museum specimensusing the formula 100 � ((WL � SL)/WL), in which WL is the stan-dard length of the closed wing and SL is the distance from the car-pal joint to the tip of the first secondary feather.

The logic behind the choice of these characters was as follows.Abundance: Greater abundance implies increased dispersal poten-tial because of area effects and increased competition. Elevationalrange: Species restricted to higher elevations should be less likelyto cross lowland barriers and thus have lower dispersal potential.Foraging Stratum: Canopy feeders are expected to range morewidely in search of food than understory foragers (especially ter-restrial species), and to have greater tolerance for exposure tospace, sun, reduced humidity, etc., and thus have greater dispersalpotential. Food: We divided species into frugivores/nectarivoresversus obligate insectivores, under the assumption that frugi-vores/nectarivores range more widely in search of food than insec-tivores, and thus have more dispersal potential. Hand-Wing Index:This, like Kipp’s Index (Swaddle and Lockwood, 2003), indicatesflight capability in birds and hence dispersal potential (Burneyand Brumfield, 2009; Claramunt et al., 2012).

Previous work has shown a positive relationship between phy-logeographic structuring and species age (Smith et al., 2014), butspecies ages were not available for our study. Moreover, we wereinterested in more fine-scale genetic partitioning within speciesthan Smith et al. (2014), who examined variation in lowland rain-forest birds across the entire Neotropics. Thus, we calculateduncorrected mean proportional nucleotide dissimilarity (p-distance) among populations within a species to provide a phylo-genetic metric that was (1) appropriate given our sample sizes,(2) comparable across species, and (3) captured the range ingenetic variation across species. Because of our sample size of spe-cies (n = 12), we performed univariate PGLS analysis in which wetreated mean p-distance as the response variable and each of theecological variables as a predictor variable. To account for thenon-independence of species data due to shared ancestry, we esti-mated the maximum clade credibility (MCC) tree from a matrixcomprising a single exemplar of each species, and the tree wasused in the PGLS function to calculate the expected co-variancestructure in the data. The MCC tree was determined using the samesettings as the within species’ trees estimated in BEAST. We reportonly model results from the MCC tree because phylogenetic uncer-tainty was low and our preliminary PGLS analyses using multipletrees from the posterior distribution showed negligible differencesin the model output.

3. Results

The complete ND2 gene (1041 base pairs) was sequenced ineach of 207 individuals representing 12 montane species, andthese data were supplemented with sequences from GenBank(Table S2). Nucleotide diversity ranges from 0.003 to 0.008 amongthe species (Table S3), with D. leucophaeus, P. hypoxantha, R. albicol-lis, Y. everetti, and S. montis having the lowest levels and P. pyrro-genys and S. nigriceps the highest.

Maximum Likelihood, Bayesian analyses, and the haplotypenetwork yielded similar topologies within each species, but differ-ent phylogeographic patterns among species (Fig. 2, SupplementalFig. 1). Three clear topological patterns are apparent among thespecies, if we define distinct groups (SW, mid-, and NE Borneo)by reciprocal monophyly, strong branch support, and more thanfour nucleotide changes separating haplotypes: (1) a phylogeo-graphic break between NE populations and combined mid-Bornean and SW populations (A. ochraceus); (2) a break betweenSW populations and combined mid-Bornean and NE populations(P. pyrrogenys and S. nigriceps); and (3) no phylogeographic struc-

Table 1Characteristics tested by phylogenetic generalized least squares.

Species Mean p-distance Hand-Wing Index Foraging Stratum Average Occupancy Elevation

Range Min Max

Alophoixus ochraceus 0.0054 16.7 Upper understory 0.3825 2000 650 2650Hemixos cinereus 0.0071 17.2 Lower canopy 0.5388 2250 500 2750Dicrurus leucophaeus 0.0027 27.88 Canopy 0.2196 1700 700 2400Pachycephala hypoxantha 0.0033 16.99 Upper understory 0.5987 1700 1250 2950Ficedula hyperythra 0.0036 17.42 Understory 0.3655 2650 1050 3700Pellorneum pyrrogenys 0.0073 10.37 Understory 0.5266 950 600 1550Seicercus montis 0.0031 15.58 Understory 0.5602 1700 750 2450Stachyris nigriceps 0.0059 9.88 Understory 0.3591 1900 300 2200Rhipidura albicollis 0.0036 15.78 Upper understory 0.5396 6000 1150 2750Yuhina everetti 0.0029 18.35 Canopy 0.4327 2000 650 2650Garrulax treacheri 0.0063 3.54 Upper/under story 0.656 2400 950 3350Myophonus borneensis 0.0044 11.99 Understory N/A 1200 1000 2200

V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149 143

ture among populations (D. leucophaeus, F. hyperythra, M. borneen-sis, P. hypoxantha, R. albicollis, Y. everetti, and S. montis). In G. trea-cheri, mid- and NE populations are distinct from one another, butno populations occur on the SW mountains for comparison. In F.hyperythra, we were unable to obtain specimens from the SW pop-ulations, thus cannot tell if the SW groups differs from the mid/NEpopulations. In the H. cinereus, the SW group forms a well-supported clade, but only four nucleotide substitutions distinguishit, the same number of changes found between other individualsand populations. We considered testing agreement among thephylogeographic estimates using concordance factors (e.g., Satlerand Carstens, 2016), but inconsistent sampling precluded suchanalysis.

In the three species exhibiting marked phylogeographic struc-ture, the branching position of the isolated Tawau Hills populationvaries. In A. ochraceus, the Tawau Hills population is closest toother NE Borneo populations; in P. pyrrogenys, it is closest to theextreme SW population; and in S. nigriceps it lies among mid-and NE Borneo populations.

AMOVA supports the structure evident in trees and networks ofA. ochraceus, S. nigriceps, and P. pyrrogenys (Table 2). Substantialvariation is evident within each population, and little variationoccurs between populations within the three groups (SW, mid-Borneo, and NE), suggesting incomplete lineage sorting or geneflow among grouped populations. Garrulax treacheri exhibited asimilar pattern as the three structured species, but it lacks a SWpopulation for comparison. In H. cinereus, despite support of aSW clade, AMOVA indicated little structure.

For the three species with defined population structure, diver-gence dates were estimated to be in the Middle to Late Pleistocene.In A. ochraceus, the mean date of divergence between NE Borneoand the SW/mid-Borneo clades was 0.717 Ma (95% highest poste-rior density (HPD): 0.314–1.12). In P. pyrrogenys, the split betweenSW Borneo and the rest of Borneo occurred about 0.748 Ma (95%HPD: 0.366–1.13). This was followed by divergence of the TawauHills population about 0.613 Ma (95% HPD: 0.269–0.957), and thenby the N/mid-Borneo clades about 0.395 Ma (95% HPD: 0.171–0.618). In S. nigriceps, SW Borneo and the rest of Borneo divergedaround 0.575 Ma (95% HPD: 0.3–0.850).

When Bornean populations were compared to extralimital pop-ulations, some interesting patterns emerged. In D. leucophaeus, theJavan individual is closer to Bornean populations than to Indochi-nese and Chinese populations. In F. hyperythra, Javan and Sumatranindividuals form a clade separate from Bornean birds, but with lowsupport. In Myophonus, Himalayan M. caeruleus and Sumatran M.melanurus form a strongly supported clade distinct from the Bor-nean M. borneensis.

PGLS analyses using p-distances (Table S4) indicated that hand-wing index (adjusted R2 = 0.3341) explains the highest variation in

population genetic structure across our 12 species on Borneo(Table 3, Fig. 3). No ecological characteristic explained populationstructure (Table 3).

4. Discussion

Our phylogeographic comparison of 12 Bornean montanepasserines revealed substantial population subdivision in only fourspecies: Alophoixus ochraceus, Garrulax treacheri, Stachyris nigriceps,and Pellorneum pyrrogenys. In A. ochraceus, mid-Borneo populationsare markedly distinct from those of the NE and are closer to SWpopulations. Garrulax treacheri also exhibits a mid-NE Borneobreak, but we did not have sampling of its SW populations todetermine if it matched the A. ochraceus pattern. (G. treacheri doesnot occur on Pueh and Penrissen, but does occur in SW portions ofthe main mountain chain.) In contrast to those two species, S. nigri-ceps and P. pyrrogenys populations in mid-Borneo are more similarto those in the NE than SW. The other species—Hemixos cinereus,Ficedula hyperythra, Dicrurus leucophaeus, Myophonus borneensis,Pachycephala hypoxantha, Rhipidura albicollis, Seicercus montis, andYuhina everetti—do not have substantial structure. For F. hyperythra(like G. treacheri), we lacked sampling from the SW and do notknow if those populations differ substantially from mid- and NEBorneo populations.

4.1. Geography and population structure

Spatial proximity and habitat connectivity explain much of thestructure among Borneo’s montane bird populations. In NE Borneo,the populations in the main mountain chain--Kinabalu, CrockerRange, Trus Madi, and Meliau--are linked by largely continuousmontane habitat. Thus, little structure is expected among themand little occurs. The one ‘‘NE” population that stands out fromthe rest is Tawau Hills Park. The Tawau Hills are 1200–1300 m vol-canoes (Hall, 2013), isolated from the main Bornean chain by ca.200 km of low elevation forest (Fig. 1). We sampled five speciesfrom Tawau Hills: A. ochraceus, S. nigriceps, P. pyrrogenys, H. ciner-eus, and Y. everetti. The relationships of these populations to thosein the main mountain chain are idiosyncratic: Y. everetti exhibitsno discernable relationship (Fig. S1.2); A. ochraceus and H. cinereusare closest to mid-Bornean populations (Fig. 2b); S. nigriceps isclosest to a cluster of mid- and NE Borneo populations; and P. pyr-rogenys is closest to the SW population of Penrissen. One explana-tion for a close relationship between Tawau Hills and mid-Borneo(and even SW Borneo) populations is that during the LGM, whenmontane forest was lower in elevation, mid-Bornean mountainsreached Tawau Hills (Cannon et al., 2009; Manthey et al., 2017).However, we cannot discount the possibility that Tawau Hills’ var-

B88030_MULUB88197_PENRISSENB88193_PENRISSEN

B88037_MULUB88038_MULU

B61587_MALIGANB61589_MALIGAN

B88165_PENRISSEN

B51006_LUMAKUB51187_TAWAU

B88028_MULUB52634_T.MADI

B61645_KINABALUB36311_CROCKER

B36336_CROCKER

B38614_KINABALUB51176_TAWAUB52607_T.MADIB61635_KINABALU

YPM143000_BRUNEI

B88177_PENRISSENB88198_PENRISSENB88172_PENRISSEN

B78667_BARIOB78715_BARIO

YPM143002_BRUNEIB88045_MULU

5.0E-4

A. o. fowleri

A. o. ruficrissus

A. ochraceus

P. pyrrogenys

B78716_BARIOB85180_MULUB88033_MULU

B88042_MULUB78721_BARIO

YPM142991_BRUNEI

8.0E-4

B85106_MULU

B51018_LUMAKUB51010_LUMAKU

B78757_BARIOB85180_MULU

B36316_CROCKERB36335_CROCKER

B52568_T.MADIB61643_KINABALU

B52655_T.MADIB57456_MELIAU

B57474_MELIAUB61637_KINABALU

B88132_PENRISSENB79662_PENRISSEN

B88124_PENRISSEN

B88147_PENRISSENB88164_PENRISSENB88166_PENRISSEN

B51195_TAWAU

P. p. canicapillus

P. p. longstaffi

P. p. erythrote

S. n. borneensis

S. n. hartleyi

4.0E-4

B79595_PENRISSENB79596_PENRISSEN

B52146_PADAWANB79620_PENRISSENB79668_PENRISSEN

B88206_PENRISSENB79799_PENRISSENB79737_PUEHB88130_PUEHB79481_PUEHB79483_PUEH

B88022_MULUB51036_LUMAKU

B85184_MULUB88031_MULU

B51030_LUMAKUB78763_BARIO

B88041_MULUB61556_MALIGANB36328_CROCKER

B36331_CROCKERB51181_TAWAUB51184_TAWAU

B52556_T.MADIB52559_T.MADI

B36288_KINABALUB61565_MALIGANB61633_KINABALUB61636_KINABALUB78756_BARIO

B79667_PENRISSENB88131_PENRISSENB88168_PENRISSEN

B79582_PENRISSEN

S. nigriceps

Fig. 2. Phylogeographic trees, haplotype networks, and maps with sampling sites (small circles) and subspecies distributions (large circles). Each tree was constructed byBayesian majority-rule consensus, with thickened branches representing Bayesian posterior probabilities >0.90 or maximum likelihood bootstrap nodal support of >70%. Sizeof circles in the haplotype networks indicates the proportion of individuals with a given haplotype, and cross hatches indicate nucleotide changes. Subspecies names are fromGill and Donsker (2016). Color of small circles corresponds with colors in the haplotype network. Large circles outline approximate distributions of subspecies. Illustrations byVLC.

144 V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149

ied relationships result from founder effects or drift due to smallpopulation sizes and long term isolation.

In mid-Borneo, the populations of Lumaku, the Maligan Range,Mulu, and Bario are connected by low elevation hills (<900 m),which might reduce gene flow. However, in all 12 species mid-Borneo populations are genetically similar to one another. A closerelationship between mid-Borneo and NE populations, as in S.

nigriceps and P. pyrrogenys, makes especially good sense in termsof current and Pleistocene geography. Both areas lie in Borneo’smain mountain chain, are relatively close to one another (ca.200 km), and would have experienced greater connectivity duringcolder Pleistocene times.

The SW populations of Penrissen and Pueh (Fig. 1) are geo-graphically more distant from mid-Borneo populations (ca.

H. cinereus

H. c. connectens

4.0E-4

0.9

G

B88040_MULU

B61548_MALIGAN

B78696_BARIO

B61647_KINABALU

B52552_T.MADI

B52549_T.MADI

B61651_KINABALU

JN980724_G.CHINENSIS

B38649_KINABALU

B52610_T.MADI

. treacheri

B88043_MULU

B85203_MULU

G. t. treacheri

G. t. damnatus

G. t. griswoldi

B52612_T.MADI

B78688_BARIO

B38659_T.MADI

B51128_TAWAU

B78719_BARIO

B88171_PENRISSEN

B88173_PENRISSEN

B88186_PENRISSEN

B88187_PENRISSEN

Fig. 2 (continued)

Table 2Results of AMOVA among NE, mid-, and SW Borneo groups.

Alophoixus ochraceus Pellorneum pyrrogenys

Source of variation % ofvariation

Source of variation % ofvariation

Among groups 44.64 Among groups 58.04Among populations

within groups11.53 Among populations

within groups13.66

Within populations 43.83 Within populations 28.31

Hemixos cinereus Stachyris nigriceps

Source of variation % ofvariation

Source of variation % ofvariation

Among groups 4.61 Among groups 24.32Among populations

within groups34.45 Among populations

within groups9.61

Within populations 60.94 Within populations 66.07

Garrulax treacheri

Source of variation % ofvariation

Among groups 30.94Among populations

within groups3.21

Within populations 72.27

V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149 145

600 km) and isolated from the main mountain chain by ca. 200 kmof low elevation habitat (Banks, 1952; Harrisson, 1956). Their iso-lation was also likely during the LGM (Manthey et al., 2017). Like

the Tawau Hills, both SW mountains are low in elevation (ca.1200 and 1500 m, respectively) and small in area, increasing thelikelihood of founder events and drift.

4.2. Comparative phylogeography

Several factors potentially played a role in structuring A. ochra-ceus, G. treacheri, S. nigriceps, and P. pyrrogenys populations, espe-cially: (1) geographic distance and lowland barriers, which wouldreduce gene flow; (2) morphological and ecological characteristicsof the species that tend to increase or decrease gene flow; (3) his-torical barriers or links increasing or decreasing gene flow in thepast; and (4) independent invasion of Borneo (non-monophyly).Factors (1) and (2) likely play a role currently, especially in distin-guishing the isolated SW and Tawau Hills populations. Factor (3)would seem of little importance to populations in the centralmountain chain, where montane forest connections would havebeen greater during Pleistocene glacial events than at present,but it could be important to the SW and Tawau Hills populations.Factor (4) is untestable with our data, because we have no out-groups for the four structured species. It might, however, explainthe difference in population structure between A. ochraceus/G. trea-cheri and S. nigriceps/P. pyrrogenys.

Although the main division in the structured Bornean speciesoccurred at about the same time (ca. 0.6–0.7 Ma), it seems unlikelythat a single geographic event caused it. The structure of S. nigri-ceps and P. pyrrogenys probably results from long-term isolation

Table 3Results of phylogenetic generalized least squares analyses of the effects of ecological variables on mean p-distance.

Effect Estimate Standard Error t value P AICc Output

Hand Wing Index �1.79E�04 7.02E�05 �2.5534 0.0287 �120.9563 Adjusted R2 = 0.3341; Fs(df) = 6.52 (1,10)Occupancy 3.75E�03 4.04E�03 0.9274 0.3779 �105.987 Adjusted R2 = �0.0142; Fs(df) = 0.8601 (1,9)Elevation Range �1.58E�07 8.23E�07 �0.1917 0.8518 �115.4297 Adjusted R2 = �0.096; Fs(df) = 0.03675 (1,10)Elevation Max �6.96E�07 8.55E�07 �0.8143 0.4344 �116.1669 Adjusted R2 = �0.0316; Fs(df) = 0.6631 (1,10)Elevation Min �2.86E�06 1.64E�06 �1.7355 0.1133 �118.0918 Adjusted R2 = 0.1546; Fs(df) = 3.012 (1,10)

Hand Wing Index

Mea

n P

Dis

tanc

e

Fig. 3. Relationship of hand-wing index to genetic distance. Each circle represents adifferent species. (Adjusted R2 = 0.3341; Fs(df) = 6.52 (1,10)).

146 V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149

of their SW populations on Pueh and Penrissen. As understoryinsectivores without much dispersal potential (see below), theywould have found it difficult to cross the wide expanse of lowlandforest between these mountains and the main Bornean range. Theirpopulations in the main range, however, would easily move backand forth, uniting mid- and NE Borneo populations. Alophoixusochraceus presents more of a conundrum. In what circumstanceswould the isolated SW populations mix with mid-Borneo but notthe NE populations? Separate invasion from western Sundalandby a genetically divergent (possibly reproductively isolated) popu-lation is a possibility (this scenario might also apply to G. treacheri)Alophoixus ochraceus and G. treacheri both have potentially closelyrelated populations on Sumatra and the Malay Peninsula (includ-ing, for G. treacheri, the subjectively synonymous G. mitratus).These populations may have invaded western Borneo recentlyand expanded eastward, much as several lowland species havedone (Lim et al., in press; Lim et al., 2010). The actual reason forthe disparity in populations’ structures, however, will have to waitfor more extensive comparisons using genomic DNA and out-groups (Lim et al., in press).

4.3. Ecology and population structure

We anticipated the greatest phylogeographic structure in spe-cies whose dispersal potential is limited by habitat and morphol-ogy. Thus, we expected understory species to display the mostpopulation structure, especially those that are insectivorous,uncommon, and have narrow elevational ranges (Moore et al.,2008; Sekercioglu et al., 2002). We expected the least populationstructure in canopy species, especially frugivorous or nectarivorousspecies (Burney and Brumfield, 2009; Sekercioglu et al., 2004).

Canopy species, especially frugivores and nectarivores, tend torange widely in search of patchy food resources, and thus encoun-ter and tolerate relatively wide variation in temperature, sunlight,and habitat types (Moegenburg and Levey, 2003; Terborgh, 1986).In contrast, understory insectivores tend to sally, flutter, hop, orwalk to forage in a more limited areas, often defined by distinctforaging microhabitats (Mansor and Ramli, 2017; Powell et al.,2015; Styring et al., 2016). Also, compared to canopy species,understory birds possess a wing shape designed better for maneu-verability than long distance flight, and vice versa (Burney andBrumfield, 2009). Thus, understory species are physically as wellas behaviorally limited in their dispersal potential.

Most of these expectations did not hold up when tested withPGLS analysis. The only characteristic we tested that showed astrong correlation with population structure was hand-wing index(Table 3, Fig. 3). Mean p-distance decreased as hand-wing indexincreased. Our interpretation is that species with longer wingshave greater dispersal potential, causing increased gene flowamong their populations. It is worth noting, however, that hand-wing index may have provided a clearer relationship with popula-tion structure because it was the only characteristics we measuredprecisely, and our measurements indicated relatively little vari-ance within populations. The other characteristics we examinedare more subjective and vary substantially, depending on weather,food seasonality, and other unpredictable or unknown factors.

4.4. Historical geography

Divergence date estimates for A. ochraceus, P. pyrrogenys, S.nigriceps, and G. treacheri suggest their populations diverged inthe mid-Pleistocene (ca. 0.6–0.7 Ma). This timing—though approx-imate and subject to wide variance--is more recent than dates ofdivergence between most lowland rainforest bird populations onBorneo, which diverged >1–2 Ma (Chua et al., 2015; Lim et al., inpress, 2010; Lim and Sheldon, 2011). In addition, lowland rain-forest populations of birds and several other groups tend to transi-tion near the Sabah-Sarawak border (Lim et al., in press 2010; Limand Sheldon, 2011). Although this is true of A. ochraceus and G.treacheri, it is not the case in S. nigriceps and P. pyrrogenys.

Temporal and spatial differences between lowland and mon-tane populations make sense given the effect of Pleistocene glacialevents on habitat distribution (Cannon et al., 2009; Heaney, 1991;Manthey et al., 2017). Global glaciation caused sea levels to dropnumerous times, exposing the Sunda continental shelf as land thatunited the Sunda Islands and Malay Peninsula into a single subcon-tinent for extended periods of time (Woodruff, 2010). The center ofsubaerial Sundaland is believed to have been quite dry, developinginto seasonal woodland or savannah (Bird et al., 2005; Heaney,1991). Although biogeographers do not agree whether a dry inte-rior existed in Sundaland during the LGM, it certainly existed inthe early Pleistocene when Java was invaded by the Asian mam-malian megafauna and man (Sheldon et al., 2015; Wurster andBird, 2014). At that time, Sundaland’s dry interior would havevicariated rainforest organisms into rainforest refugia in easternBorneo and western Sumatra/Java, which were maintained by oro-graphic rainfall in coastal mountains and on offshore islands

V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149 147

(Gathorne-Hardy et al., 2002). The rainforest refugia, in turn, wouldhave fomented population diversification. As differentiated easternand western lowland species expanded across Sundaland in theLate Pleistocene with increased forest coverage (Cannon et al.,2009; Raes et al., 2014), they met in central Borneo forming thecontact zones we see today (e.g., near the Sabah-Sarawak border).While Pleistocene glacial events caused the vicariance of lowlandrainforest populations, the colder climate associated with theseevents would have caused montane rainforest to descend in eleva-tion and broaden in range (Cannon et al., 2009), allowing montanepopulations to expand and colonize areas not previously accessi-ble. Montane forest expansion would have unified previously iso-lated montane populations, while further subdividing lowlandpopulations.

The outcome of these dynamics is that lowland species formcontact zones between differentiated populations on Borneo, oftennear the Sabah-Sarawak border (Lim et al., in press). Montane pop-ulations, in contrast, are not expected to exhibit such structure.They should be structured mainly by geographic distance and low-land barriers, as seems likely for the Pueh/Penrissen populations ofS. nigriceps and P. pyrrogenys, and which also applies to the Puehpopulation of Chlorocharis emiliae (Manthey et al., 2017). However,like lowland species, disjunct populations of A. ochraceus (andprobably G. treacheri) may have formed between western Sunda-land and eastern Borneo. Subsequently, the western Sundaic pop-ulations invaded and expanded in Borneo to yield a lowland-likecontact zone in the mountains near the Sabah-Sarawak border.

4.5. Taxonomy

Subspecific classification of the four structured species is largelyconsistent major phylogeographic breaks (Fig. 2). Alophoixus ochra-ceus consists of A. o. ruficrissus described from Mt. Kinabalu(Sharpe, 1879) and A. o. fowleri from the Kelabit Highlands inmid-Borneo (Amadon and Harrisson, 1956). Populations of thisspecies on Mts. Pueh and Penrissen were not examined when thesesubspecies were designated, but fit genetically within A. o. fowleri.Garrulax treacheri is divided into three subspecies, but we com-pared only two, G. t. treacheri described from Mt. Kinabalu(Sharpe, 1879) and G. t. damnatus described from Mt. Dulit inmid-Borneo (Harrisson and Hartley, 1934). Stachyris nigriceps isdivided into S. n. hartleyi from Mts. Pueh and Penrissen (Chasen,1935) and S. n. borneensis fromMt. Dulit (mid-Borneo) and Mt. Kin-abalu (Harrisson and Hartley, 1934; Sharpe, 1887). Pellorneum pyr-rogenys is the only one of the four species with a slightly inconstantsubspecific classification. It is divided into P. p. erythrote from Puehand Penrissen (Sharpe, 1883), P. p longstaffi from Mt. Dulit in mid-Borneo (Harrisson and Hartley, 1934), and P. p. canicapillus fromMt. Kinabalu (Sharpe, 1887). The status of P. p. longstaffi shouldbe re-examined.

Five of the remaining species are placed in a single Bornean sub-species and exhibit no substantial or consistent variation acrossthe island (Yuhina everetti, Seicercus montis, Myophonus borneensis,and Dicrurus leucophaeus, and Hemixos cinereus).

The last three species, Pachycephala hypoxantha, Ficedula hyper-ythra, and Rhipidura albicollis, are divided into subspecies on Bor-neo. In P. hypoxantha, P. h. sarawacensis is described from Mt.Pueh and P. h. hypoxantha from Mt. Kinabalu (Chasen, 1935). Ourgenetic assessment (based on only one specimen, from Pueh) indi-cates no obvious differences among populations. Ficedula hypery-thra populations from NE and mid-Borneo are placed in the samesubspecies as the Malay Peninsula and Sumatran populations, F.h. sumatrana (Hachisuka, 1926), whereas SW Bornean populationsare in a different subspecies, F. h. mjöbergi (Mjöberg, 1925). Unfor-tunately, we did not have samples from SW Borneo for comparison.Our genetic assessment of Sumatran and Javan F. hyperythra sug-

gests these populations are distinct from mid- and NE Borneanpopulations and more closely related to one another. Properassessment of their status must wait until SW Borneo and MalayPeninsula specimens are available. Rhipidura albicollis is split intotwo subspecies on Borneo, R. a. kinabalu from Mt. Kinabalu andR. a. sarawacensis from Mt. Pueh (Chasen and Hoogerwerf, 1941).Phylogeographic comparison do not indicate substantial geneticdifferences between these populations.

5. Conclusions

Comparative phylogeography of 12 Bornean montane passerinespecies indicates that most do not have deep genetic structureamong populations. Four species, however, are divided into twomain clades: Alophoixus ochraceus and Garrulax treacheri into NEBornean populations (essentially in Sabah) versus those on the restof the island; Stachyris nigriceps and Pellorneum pyrrogenys into themain Bornean mountain chain versus two isolated peaks in west-ern Sarawak (Pueh and Penrissen). Division in all four speciesappears to have occurred about the same time (ca. 0.6–0.7 Ma),but the cause is not clear and the timing of subdivision does notcoincide with lowland species subdivision (ca. 1–2 Ma). The onlyecological or morphological characteristic of the species that isstrongly correlated with population structure is hand-wing index:long winged birds (which presumably disperse more effectively)have less population structure. These findings provide specificgoals for future studies of Bornean montane bird phylogeography:obtain and compare western Sundaic outgroups of the structuredspecies to assess possible non-monophyly of Bornean populations,and apply genomic comparisons that can measure gene flow andchanges in population size.

Acknowledgments

We would like to thank those who helped collect specimens forthe Louisiana State Museum of Natural Science used in this study.We are also grateful to Kristof Zyskowski and Richard Prum fromYale Peabody Museum for providing tissue samples from Brunei.We thank the Malaysian Chief Minister’s Department and the gov-ernments of Sabah and Sarawak for the help and permission toundertake research in Malaysia Borneo. J.V. Remsen, R.T. Brumfield,J. Brown, S. Taylor, J. A. Oswald, Z. Rodriguez, and C.E. Brown pro-vided helpful discussions and comments. Financial and logisticalsupport were provided by NSF DEB-0228688, NSF DEB-1241059,NSF DEB-1241181, Coypu Foundation of Louisiana, National Geo-graphic Society (8753-10), Louisiana State University, Universityof Kansas, Sabah Parks, Sabah Museum, Universiti Malaysia Sara-wak, and a collection study grant from the Frank M. ChapmanMemorial Fund of the American Museum of Natural History.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.ympev.2017.05.016.

References

Amadon, D., Harrisson, T., 1956. A new bulbul from the Kelabit Uplands. SarawakMus. J. (new series) 7, 516–517.

Arbogast, B.S., Kenagy, G.J., 2001. Comparative phylogeography as an integrativeapproach to historical biogeography. J. Biogeogr. 28, 819–825.

Bandelt, H.-J., Forster, P., Röhl, A., 1999. Median-joining networks for inferringintraspecific phylogenies. Mol. Biol. Evol. 16, 37–48.

Banks, E., 1952. Mammals and birds from the Maga mountains in Borneo. Bulletin ofthe Raffles Museum 24, 160–163.

148 V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149

Bird, M.I., Taylor, D., Hunt, C., 2005. Palaeoenvironments of insular Southeast Asiaduring the last glacial period: a savanna corridor in Sundaland? Quatern. Sci.Rev. 24, 2228–2242.

Brandon-Jones, D., 1996. The Asian Colobinae (Mammalia: Cercopithecidae) asindicators of quaternary climatic change. Biol. J. Lin. Soc. 59, 327–350.

Burner, R.C., Chua, V.L., Brady, M.L., Van Els, P., Steinhoff, P.O.M., Rahman, M.A.,Sheldon, F.H., 2016. An ornithological survey of Gunung Mulu National Park,Sarawak, Malaysian Borneo. Wilson J. Ornithol. 128, 242–254.

Burney, C.W., Brumfield, R.T., 2009. Ecology predicts levels of genetic differentiationin Neotropical birds. Am. Nat. 174, 358–368.

Cannon, C.H., Morley, R.J., Bush, A.B.G., 2009. The current refugial rainforests ofSundaland are unrepresentative of their biogeographic past and highlyvulnerable to disturbance. Proc. Natl. Acad. Sci. U.S.A. 106, 11188–11193.

Carstens, B.C., Richards, C.L., 2007. Integrating coalescent and ecological nichemodeling in comparative phylogeography. Evolution 61, 1439–1454.

Chasen, F.N., 1935. Four new races of Malaysian birds. Bull. Raffles Museum 10, 43–44.

Chasen, F.N., Hoogerwerf, A., 1941. The birds of the Netherlands Indies Mt. Leuserexpedition to north Sumatra. Treubia 18 (supplement), 1–125.

Chua, V.L., Phillipps, Q., Lim, H.C., Taylor, S.S., Gawin, D.F., Rahman, M.A., Moyle, R.G.,Sheldon, F.H., 2015. Phylogeography of three endemic birds of Maratua Island, apotential archive of Bornean biogeography. Raffles Bull. Zool. 63, 259–269.

Claramunt, S., Derryberry, E.P., Remsen, J., Brumfield, R.T., 2012. High dispersalability inhibits speciation in a continental radiation of passerine birds. Proc.Roy. Soc. Lond. B: Biol. Sci. 279, 1567–1574.

Cracraft, J., 1988. From Malaysia to New Guinea: evolutionary biogeography withina complex continent-island arc contact zone. Acta Congr. Intern. Ornith. 19,2581–2593.

Cracraft, J., Prum, R.O., 1988. Patterns and processes of diversification: speciationand historical congruence in some neotropical birds. Evolution 42, 603–620.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. JModelTest 2: more models,new heuristics and parallel computing. Nat. Methods 9. 772-772.

de Bruyn, M., Stelbrink, B., Morley, R.J., Hall, R., Carvalho, G.R., Cannon, C.H., van denBergh, G., Meijaard, E., Metcalfe, I., Boitani, L., Maiorano, L., Shoup, R., vonRintelen, K., 2014. Borneo and Indochina are major evolutionary hotspots forSoutheast Asian biodiversity. Syst. Biol. 63, 879–901.

Demos, T.C., Achmadi, A.S., Giarla, T.C., Handika, H., Rowe, K.C., Esselstyn, J.A., 2016.Local endemism and within-island diversification of shrews illustrate theimportance of speciation in building Sundaland mammal diversity. Mol. Ecol.25, 5158–5173.

Diamond, A.W., Bishop, K.D., Van Balen, S., 1987. Bird survival in an isolated Javanwoodland: Island or mirror. Conserv. Biol. 1, 132–142.

Ditchfield, A.D., 2000. The comparative phylogeography of Neotropical mammals:patterns of intraspecific mitochondrial DNA variation among bats contrasted tononvolant small mammals. Mol. Ecol. 9, 1307–1318.

Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogeneticswith BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29, 1969–1973.

Edwards, S., Bensch, S., 2009. Looking forwards or looking backwards in avianphylogeography? A comment on Zink and Barrowclough 2008. Mol. Ecol. 18,2930–2933.

Excoffier, L., Lischer, H.E., 2010. Arlequin suite ver 3.5: a new series of programs toperform population genetics analyses under Linux and Windows. Mol. Ecol.Resour. 10, 564–567.

Fjeldså, J., Bowie, R.C.K., 2008. New perspectives on the origin and diversification ofAfrica’s forest avifauna. Afr. J. Ecol. 46, 235–247.

Gathorne-Hardy, F.J., Syaukani Davies, RG., Eggleton, P., Jones, DT., 2002. Quaternaryrainforest refugia in south-east Asia: using termites (Isoptera) as indicators.Biol. J. Lin. Soc. 75, 453–466.

Gavrilets, S., Vose, A., 2005. Dynamic patterns of adaptive radiation. Proc. Natl. Acad.Sci. U.S.A. 102, 18040–18045.

Gawin, D.F., Rahman, M.A., Ramji, M.F.S., Smith, B.T., Lim, H.C., Moyle, R.G., Sheldon,F.H., 2014. Patterns of avian diversification in Borneo: the case of the endemicMountain Black-eye (Chlorocharis emiliae). Auk 131, 86–99.

Gill, F., Donsker, D., 2016. IOC World Bird List (v 6.1).Gorog, A.J., Sinaga, M.H., Engstrom, M.D., 2004. Vicariance or dispersal? Historical

biogeography of three Sunda shelf murine rodents (Maxomys surifer,Leopoldamys sabanus and Maxomys whiteheadi). Biol. J. Lin. Soc. 81, 91–109.

Guindon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimatelarge phylogenies by maximum likelihood. Syst. Biol. 52, 696–704.

Hachisuka, M.U., 1926. Dendrobiastes hyperythrus sumatranus, subsp. nov. Bulletinof the British Ornithologists’ Club 47, 52.

Hackett, S.J., 1996. Molecular Phylogenetics and Biogeography of Tanagers in theGenus i Ramphocelus/i (Aves). Mol. Phylogenet. Evol. 5, 368–382.

Haffer, J., 1987. Biogeography of neotropical birds. In: Whitmore, T.C., Prance, G.T.(Eds.), Biogeography and Quarternary History in Tropical America. ClarendonPress, Oxford.

Hall, R., 2013. The palaeogeography of Sundaland and Wallacea since the LateJurassic. J. Limnol. 72 (s2), 1–17.

Harmon, L.J., Weir, J.T., Brock, C.D., Glor, R.E., Challenger, W., 2008. GEIGER:investigating evolutionary radiations. Bioinformatics 24, 129–131.

Harrisson, T., 1956. A new mountain black-eye (Chlorocharis) from North Borneo.Sarawak Museum J. 7, 518–521.

Harrisson, T.H., Hartley, C.H., 1934. Descriptions of new races from mountain areasin Borneo. Bull. Brit. Ornithologists’ Club. 378, 148–161.

Heaney, L.R., 1991. A synopsis of climatic and vegetational change in Southeast Asia.Clim. Change 19, 53–61.

Huhndorf, M.H., Peterhans, K., Julian, C., Loew, S.S., 2007. Comparativephylogeography of three endemic rodents from the Albertine Rift, east centralAfrica. Mol. Ecol. 16, 663–674.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S.,Cooper, A., Markowitz, S., Duran, C., 2012. Geneious Basic: an integrated andextendable desktop software platform for the organization and analysis ofsequence data. Bioinformatics 28, 1647–1649.

Kisel, Y., Barraclough, T.G., 2010. Speciation has a spatial scale that depends onlevels of gene flow. Am. Nat. 175, 316–334.

Lanfear, R., Calcott, B., Ho, S.Y., Guindon, S., 2012. PartitionFinder: combinedselection of partitioning schemes and substitution models for phylogeneticanalyses. Mol. Biol. Evol. 29, 1695–1701.

Lanfear, R., Calcott, B., Kainer, D., Mayer, C., Stamatakis, A., 2014. Selecting optimalpartitioning schemes for phylogenomic datasets. BMC Evol. Biol. 14, 82.

Lim, H.C., Gawin, D.F., Shakya, S.B., Harvey, M.G., Rahman, M.A., Sheldon, F.H., 2017.Sundaland’s east-west rainforest population structure: variable manifestationsin four polytypic bird species examined using RAD-Seq and plumage analyses. J.Biogeogr. (in press).

Lim, H.C., Rahman, M.A., Lim, S.L.H., Moyle, R.G., Sheldon, F.H., 2010. RevisitingWallace’s haunt: coalescent simulations and comparative niche modelingreveal historical mechanisms that promoted avian population divergence inthe Malay Archipelago. Evolution 65, 321–334.

Lim, H.C., Sheldon, F.H., 2011. Multilocus analysis of the evolutionary dynamics ofrainforest bird populations in Southeast Asia. Mol. Ecol. 20, 3414–3438.

Mansor, M.S., Ramli, R., 2017. Foraging niche segregation in Malaysian babblers(Family: Timaliidae). PLoS ONE 12, e0172836.

Manthey, J.D., Moyle, R.G., Gawin, D.F., Rahman, M.A., Ramji, M.F.S., Sheldon, F.H.,2017. Genomic phylogeography of the endemic Mountain Blackeye of Borneo(Chlorocharis emiliae): montane and lowland populations differ in patterns ofPleistocene diversification. J. Biogeogr. http://dx.doi.org/10.1111/jbi.13028.http://wileyonlinelibrary.com/journal/jbi1.

Martins, E.P., Hansen, T.F., 1996. The statistical analysis of interspecific data: areview and evaluation of phylogenetic comparative methods. In: Martins, E.P.(Ed.), Phylogenies and the Comparative Method in Animal Behavior. OxfordUniversity Press, Oxford, pp. 22–75.

Mayr, E., 1963. Animal Species and Evolution. Belknap Press, Cambridge,Massachusetts.

Merckx, V.S., Hendriks, K.P., Beentjes, K.K., Mennes, C.B., Becking, L.E., Peijnenburg,K.T., Afendy, A., Arumugam, N., de Boer, H., Biun, A., 2015. Evolution ofendemism on a young tropical mountain. Nature 524, 347–350.

Miller, M., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway forinference of large phylogenetic trees. Gateway Computing EnvironmentsWorkshop (GCE), 2010. IEEE, pp. 1–8.

Mjöberg, E., 1925. An expedition to the Kalabit country and Mt. Murud, Sarawak.Geographical Review 15.

Moegenburg, S.M., Levey, D.J., 2003. Do frugivores respond to fruit harvest? Anexperimental study of short-term responses. Ecology 84, 2600–2612.

Moore, R., Robinson, W., Lovette, I., Robinson, T., 2008. Experimental evidence forextreme dispersal limitation in tropical forest birds. Ecol. Lett. 11, 960–968.

Morley, R., 2012. A review of the Cenozoic palaeoclimate history of Southeast Asia.Biotic evolution and environmental change in Southeast Asia. Syst. Assoc.Special 82, 79–114.

Moyle, R.G., Andersen, M.J., Oliveros, C.H., Steinheimer, F.D., Reddy, S., 2012.Phylogeny and biogeography of the core babblers (Aves: Timaliidae). Syst. Biol.61, 631–651.

Myers, S., 2009. A Field Guide to the Birds of Borneo. New Holland Publishers,London.

Orme, D., Freckleton, R., Thomas, G., Petzoldt, T., Fritz, S., Isaac, N., Pearse, W., 2012.CAPER: Comparative Analyses of Phylogenetics and Evolution in R. R packageversion 0.5 <http://CRAN.R-project.org/package=caper>.

Päckert, M., Martens, J., Sun, Y.H., Severinghaus, L.L., Nazarenko, A.A., Ting, J., Töpfer,T., Tietze, D.T., 2012. Horizontal and elevational phylogeographic patterns ofHimalayan and Southeast Asian forest passerines (Aves: Passeriformes). J.Biogeogr. 39, 556–573.

Phillipps, Q., Phillipps, K., 2014. Phillipps’ Field Guide to the Birds of Borneo. JohnBeaufoy, Oxford.

Powell, L.L., Cordeiro, N.J., Stratford, J.A., 2015. Ecology and conservation of avianinsectivores of the rainforest understory: a pantropical perspective. Biol. Cons.188, 1–10.

Price, T.D., Hooper, D.M., Buchanan, C.D., Johansson, U.S., Tietze, D.T., Alström, P.,Olsson, U., Ghosh-Harihar, M., Ishtiaq, F., Gupta, S.K., 2014. Niche filling slowsthe diversification of Himalayan songbirds. Nature 509, 222–225.

Raes, N., Cannon, C.H., Hijmans, R.J., Piessens, T., Saw, L.G., van Welzen, P.C., Slik, J.F.,2014. Historical distribution of Sundaland’s Dipterocarp rainforests atQuaternary glacial maxima. Proc. Natl. Acad. Sci. 111, 16790–16795.

Rambaut, A., Drummond, A.J., 2007. Tracer v1. 4.Robson, C., 2000. A Guide to the Birds of Southeast Asia. Princeton University Press,

Princeton, New Jersey.Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference

under mixed models. Bioinformatics 19, 1572–1574.Royle, J.A., Dorazio, R.M., 2008. Hierarchical Modeling and Inference in Ecology: The

Analysis of Data from Populations, Metapopulations and Communities.Academic Press, New York.

Satler, J.D., Carstens, B.C., 2016. Phylogeographic concordance factors quantifyphylogeographic congruence among co-distributed species in the Sarraceniaalata pitcher plant system. Evolution 70, 1105–1119.

V.L. Chua et al. /Molecular Phylogenetics and Evolution 113 (2017) 139–149 149

Sekercioglu, C.H., Daily, G.C., Ehrlich, P.R., 2004. Ecosystem consequences of birddeclines. Proc. Natl. Acad. Sci. U.S.A. 101, 18042–18047.

Sekercioglu, C.H., Ehrlich, P.R., Daily, G.C., Aygen, D., Goehring, D., Sandi, R.F., 2002.Disappearance of insectivorous birds from tropical forest fragments. Proc. Natl.Acad. Sci. U.S.A. 99, 263–267.

Sharpe, R.B., 1879. 2. On Collections of Birds from Kina Balu Mountain, in North-western Borneo. In: Proceedings of the Zoological Society of London, pp. 245–249.

Sharpe, R.B., 1883. Malacopteron erythrote Sharpe. Catalogue of Birds in the BritishMuseum Volume 7. British Museum, London, p. 567.

Sharpe, R.B., 1887. Notes on a collection of birds made by Mr. John Whitehead onthe mountain of Kina Balu, in northern Borneo, with descriptions of newspecies. Ibis 1887, 435–454.

Sheldon, F.H., Lim, H.C., Moyle, R.G., 2015. Return to the Malay Archipelago: thebiogeography of Sundaic rainforest birds. J. Ornithol. 156 (Supplement 1), S91–S113.

Sheldon, F.H., Moyle, R.G., Kennard, J., 2001. Ornithology of Sabah: history,gazetteer, annotated checklist, and bibliography. Ornithological Monogr. 52,1–285.

Smith, B.T., Klicka, J., 2010. The profound influence of the Late Pliocene Panamanianuplift on the exchange, diversification, and distribution of New World birds.Ecography 33, 333–342.

Smith, B.T., McCormack, J.E., Cuervo, A.M., Hickerson, M.J., Aleixo, A., Cadena, C.D.,Perez-Eman, J., Burney, C.W., Xie, X., Harvey, M.G., Faircloth, B.C., Glenn, T.C.,Derryberry, E.P., Prejean, J., Fields, S., Brumfield, R.T., 2014. The drivers oftropical speciation. Nature.

Spellman, G.M., Klicka, J., 2006. Testing hypotheses of Pleistocene populationhistory using coalescent simulations: phylogeography of the pygmy nuthatch(Sitta pygmaea). Proc. Roy. Soc. Lond. B: Biol. Sci. 273, 3057–3063.

Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30, 1312–1313.

Styring, A.R., Ragai, R., Zakaria, M., Sheldon, F.H., 2016. Foraging ecology andoccurrence of 7 sympatric babbler species (Timaliidae) in the lowland rainforestof Borneo and peninsular Malaysia. Curr. Zool. 62, 345–355.

View publication statsView publication stats

Sukumaran, J., Holder, M.T., 2010. DendroPy: a Python library for phylogeneticcomputing. Bioinformatics 26, 1569–1571.

Swaddle, J.P., Lockwood, R., 2003. Wingtip shape and flight performance in theEuropean Starling Sturnus vulgaris. Ibis 145, 457–464.

Symonds, M.R., Blomberg, S.P., 2014. A primer on phylogenetic generalised leastsquares. Modern phylogenetic comparative methods and their application inevolutionary biology. Springer, Berlinn, pp. 105–130.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5:molecular evolutionary genetics analysis using maximum likelihood,evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28,2731–2739.

Team, R.C., 2012. R 2.15. 2: a language and environment for statistical computing. RFoundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0,URL: http://www.R-project.org.

Terborgh, J., 1986. Community aspects of frugivory in tropical forests. In: Estrada, A.,Fleming, T.H. (Eds.), Frugivores and Seed Dispersal. Dr. W Junk Publishers,Dordrecht, pp. 371–384.

Wallace, A.R., 1860. On the zoological geography of the Malay Archipelago. J. Proc.Linnean Soc. 4, 172–184.

Woodruff, D.S., 2010. Biogeography and conservation in Southeast Asia: how 2.7million years of repeated environmental fluctuations affect today’s patterns andthe future of the remaining refugial-phase biodiversity. Biodivers. Conserv. 19,919–941.

Wurster, C.M., Bird, M.I., 2014. Barriers and bridges: early human dispersals inequatorial SE Asia. In: Harff, J., Bailey, G., Lüth, F. (Eds.), Geology andArchaeology: Submerged Landscapes of the Continental Shelf. GeologicalSociety, London, pp. 235–250.

Zink, R.M., Barrowclough, G.F., 2008. Mitochondrial DNA under siege in avianphylogeography. Mol. Ecol. 17, 2107–2121.

Zwickl, D., 2006. GARLI—genetic algorithm for rapid likelihood inference. See<http://www.bio.utexas.edu/faculty/antisense/garli/Garli.html>.