Mathematisch-Naturwissenschaftliche Fakultät

Renata F. Martins | Anke Schmidt | Dorina LenzAndreas Wilting | Joerns Fickel

Human-mediated introduction ofintrogressed deer across Wallace’s line

historical biogeography of Rusa unicolor and R. timorensis

Postprint archived at the Institutional Repository of the Potsdam University in:Postprints der Universität PotsdamMathematisch-Naturwissenschaftliche Reihe ; 617ISSN 1866-8372https://nbn-resolving.org/urn:nbn:de:kobv:517-opus4-423843DOI https://doi.org/10.25932/publishup-42384

Suggested citation referring to the original publication:Ecology and Evolution 8 (2018), pp. 1465–1479 DOI https://doi.org/10.1002/ece3.3754ISSN 2045-7758

Ecology and Evolution. 2018;8:1465–1479.  | 1465www.ecolevol.org

Received:21August2017  |  Revised:23November2017  |  Accepted:27November2017DOI:10.1002/ece3.3754

O R I G I N A L R E S E A R C H

Human- mediated introduction of introgressed deer across Wallace’s line: Historical biogeography of Rusa unicolor and R. timorensis

Renata F. Martins1  | Anke Schmidt1 | Dorina Lenz1 | Andreas Wilting1 |  Joerns Fickel1,2

ThisisanopenaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium,providedtheoriginalworkisproperlycited.© 2017 The Authors. Ecology and EvolutionpublishedbyJohnWiley&SonsLtd.

1DepartmentofEvolutionaryGenetics,LeibnizInstituteforZooandWildlifeResearch,Berlin,Germany2InstituteforBiochemistryandBiology,UniversityofPotsdam,Potsdam,Germany

CorrespondenceRenataF.Martins,DepartmentofEvolutionaryGenetics,LeibnizInstituteforZooandWildlifeResearch,Berlin,Germany.Email:martins@izw-berlin.de

Funding informationLeibniz-Gemeinschaft,Grant/AwardNumber:SAW-2013-IZW-2

AbstractInthisstudywecomparedthephylogeographicpatternsoftwoRusaspecies,Rusa unicolor and Rusa timorensis,inordertounderstandwhatdroveandmaintaineddifferentiationbetweenthesetwogeographicallyandgeneticallyclosespeciesand investigatedtherouteof introductionofindividualstotheislandsoutsideoftheSundaShelf.Weanalyzedfullmitogenomesfrom56ar-chivalsamplesfromthedistributionareasofthetwospeciesand18microsatellitelociinasubsetof16individualstogeneratethephylogeographicpatternsofbothspecies.Bayesianinferencewithfossilcalibrationwasusedtoestimatetheageofeachspeciesandmajordivergenceevents.OurresultsindicatedthatthesplitbetweenthetwospeciestookplaceduringthePleistocene,~1.8Mya,possiblydrivenbyadaptationsofR. timorensistothedrierclimatefoundonJavacom-paredtotheotherislandsofSundaland.Althoughbothmarkersidentifiedtwowell-differentiatedclades,therewasalargelydiscrepantpatternbetweenmitochondrialandnuclearmarkers.WhilenDNAseparatedtheindividuals intothetwospecies, largelyinagreementwiththeirmuseumlabel,mtDNArevealedthatallR. timorensissampledtotheeastoftheSundashelfcarriedhaplo-typesfromR. unicolor and one Rusa unicolorfromSouthSumatracarriedaR. timorensishaplotype.OurresultsshowthathybridizationoccurredbetweenthesetwosisterspeciesinSundalanddur-ingtheLatePleistoceneandresultedinhuman-mediatedintroductionofhybriddescendantsin

allislandsoutsideSundaland.

K E Y W O R D S

Cervidae,humanintroduction,hybridization,Phylogeography,Sundaland,Wallace’sline

1  | INTRODUCTION

Biogeographic barriers interruptmigration and reproduction amongpopulations and thus are an important force responsible fordrivingandmaintaininggeneticdifferentiation,potentiallyleadingtospecia-tion.Sundaland,aSoutheastAsianbiodiversityhotspot,isborderedintheEastbyoneofthebestknownfaunalboundaries—theWallaceline(Bacon,Henderson,Mckenna,Milroy,&Simmons,2013).ThisbarrierisresponsibleforasharpbreakbetweenfaunalcompositionsofSundaandWallacea.Atthesouthernborder,theWallacelinerunsbetweenthe islandsofBaliandLombok,andat thenorthernedge, itdividesfaunaandfloraofBorneoandthePhilippinesfromthatofSulawesi

(Figure1).Althoughsomespeciesandpopulationshavenaturallydis-persedacross thisbarrier (squirrel sps.;MercerandRoth2003), thepresenceof themajorityofSundaicmammal speciesoccurringpasttheWallacelineintotheeasternislandsofWallaceaisassociatedwithhuman transportation (Groves, 1983; Heinsohn, 2003;Veron etal.,2014).

Sundaland’s dynamic geologic and climatic history, especiallyduringthePlio-Pleistoceneepochs,resultedinsealevelchangesthatrepeatedlyexposedthecontinentalshelfconnectingthemajorislandsofthisarchipelago(Voris,2000).Itisbelievedthattheseavailablelandbridgeswouldallowpopulationspreviouslyisolatedonsingleislandsto disperse, creating a largepanmictic populationwithin thiswhole

1466  |     MARTINS eT Al.

system(Latinneetal.,2015;Demosetal.,2016).However,deepgeo-graphical barriers like theWallace linewould remain through evenlowsealevelperiods,thuscreatingpatternsofgeneticdivergencebe-tweentaxaonbothsidesofthesebarriers.

Here,weinvestigatedthephylogeographicpatternsoftwoRusaspecies:thesambar,Rusa unicolor,andtheJavandeer,Rusa timorensis. WhileR. unicoloriswidespreadthroughoutSouthandSoutheastAsia(fromIndiaandSriLanka,SouthernChinaandmostof IndochinatoBorneoandSumatra,thetwolargestSundaIslands),R. timorensis has itsnativerangeonJavaandBalionly(Figure1).ThepresenceofJavandeer on islands east of theWallace line (e.g., Lesser Sunda Islands,Sulawesi,andtheMoluccas)isdescribedtobetheresultofprehistorictohistorichuman-mediatedintroductionsduringtheHolocene.Thesehuman-mediated introductionswere attributed, for example, to theAustronesian-speaking peoples’ migrations, ca. 4,000years ago, re-sponsibleaswellfortheintroductionofotherdeerspecies(e.g.,munt-jacs),pigs,macaques,andcivets; (Heinsohn,2003;Groves&Grubb,2011).

Theaimofthisstudywastocomparephylogeographicpatterns,geneticdiversity,andevolutionaryhistoryofthesetworelatedspeciesinordertoanswerthefollowingthreequestions:(1)InthepresenceoflandbridgesconnectingislandsoftheSundaShelf,whatgeograph-icalorclimaticbarrierswereresponsibleforspeciationbetweenthetwospeciesandisthereevidenceofadmixturebetweenthem?(2)DopopulationsofthewidelydistributedR. unicolorshowsignsofgenetic

structuringcorrespondingtoknowngeographicalbarriers?(3)DoesR. timorensisshowageneticsignatureofnon-naturaldispersalandwhatisthemostlikelysourcepopulationoftheintroducedpopulationsEastoftheWallaceline?

2  | MATERIALS AND METHODS

2.1 | Sampling and DNA extraction

Wesampled110individualslabeledasRusa unicolor(RUN)andRusa timorensis (RTI) from European museums, aged between 180 and61yearsold.Wecollectedeitherturbinalbonesfromthenasalcav-ity,skin,drytissuefromskeletons,andantlerdrillsexclusivelyfromindividuals with known locality. All molecular work, including DNAextraction and sequencing library preparation, was conducted in alaboratorydedicatedtoworkwitharchivalsamplestoreducetheriskof contamination.DNAextraction followed theDNeasyTissue andBloodkitprotocol (Qiagen,Hilden,Germany),withovernightdiges-tionofsamples inLysisbufferandProteinaseKat56°Candapre-elutionincubationfor20minat37°C.

2.2 | Mitochondrial genome

All extractions, includingnegative controls,werebuilt into individualsequencinglibrarieswithsingle8-ntindexes(Fortes&Paijmans,2015),

F IGURE  1 Distributionmapofbothspeciesandsamplinglocation.LightgrayindicatesthedistributionrangeofRusa unicolor(a).DarkgrayindicatesthenativedistributionofRusa timorensis,(b)anddasheddarkgrayareasindicateintroductionrangeofR. timorensis(C).FilledcirclesanddiamondsindicateR. unicolor and R. timorensis,respectively,andsizeisproportionaltothenumberofsamples.FordetailedinformationaboutallsamplesseeTableA1

North India

India

Sri Lanka

Myanmar

Indochina

Thailand

South Thailand

West Sumatra

MentawaiSumatra

North Borno

Borneo

West Ja va

Java

Bal i

Sulawesi

South Sulawesi

Timorand LSI

Moluccas

B C

(a) (b)

(c)

     |  1467MARTINS eT Al.

whichwerethensequencedonanIlluminaMiSeqtoassesssampleDNAquantityandquality(150cyclesv3kit,Illumina,CA,USA),asdescribedbelow. Samples with low-quality DNA were consequently enrichedfortheirmitochondrialDNA,usinganin-solutiontargethybridizationcapturetechnique(Maricic,Whitten,&Pääbo,2010).Baitsforhybridi-zationwere obtained by amplifying three overlappingmitochondrialfragments fromonefresh tissuesampleofR. unicolor (fromthe IZWarchive)whichwereconsequentlypreparedintocapturebaits(Maricicetal.,2010;primersandPCRconditionsasdescribedinMartinsetal.,2017).Afterhybridizationcapture,librarieswereamplifiedfornomorethan18cyclesandsequencedagainontheIlluminaMiSeqplatform.

2.3 | Bioinformatics

Sequencing readswere first demultiplexed into respective sampleswithBCL2FASTQv2.17(Illumina,CA,USA).CUTADAPTv1.3(Martin,2011)wasusedtofindandremoveadaptersequencesfromthese-quenced reads. Adapter-clipped reads were then quality trimmedthroughaslidingwindowapproachof10bpforaphredscoreofatleastQ20.Finally,readsshorterthan20bpwereremovedfromfur-ther analyses.Mapping of quality-filtered reads was carried out intwophases:AfirstmappingrunwasperformedwithBWAv.0.7.10(Li&Durbin, 2009), using a genome reference fromR. unicolor de-jeani (NCBI accessionno.NC_031835).Clonal readswere removedfrom the mapped reads using MARKDUPLICATES v1.106 (http://picard.sourceforge.net/picard-tools).SAMTOOLSMPILEUPv1.1andBCFTOOLS v1.2 (http://github.com/samtools/bcftools) were usedforvariantcalling(SNPsandInDels).Aconsensussequencewasthengeneratedforeachsampleusingathresholdofminimum3×cover-age andmajority rule (>50%) for base calling. The secondmappingstepusedthenewlygeneratedconsensussequenceasreferenceforeach sample, in order to increasemapping quality and base cover-age.BWAandMARKDUPLICATESwereusedasbefore,butGATKv1.6(McKennaetal.,2010)wasappliedforvariantcallingofthefinalconsensussequence.Positionswithcoveragelowerthan5×wereN-masked, aswere ambiguous heterozygous positions. A final qualityfilteringstepwasperformedtoremoveallsampleswithlessthan80%oftheirmitogenomecoveredatleastwith5×depth.Duetothelimita-tionsofthesequencingmethod(readsnolongerthan75bp),wewerenotabletoclearlyresolvetherepeatregionofthed-loop.Therefore,wetrimmedallsequences,byremoving460bpofthed-loopregion.Mitogenomes obtained were deposited in Genbank (for accessionnumbersseeTableA1).

2.4 | Microsatellite DNA

Microsatellitegenotypingwasachievedbyamplificationof18locionall56samplesforwhichmitochondrialDNAwasobtainedasdescribedabove (microsatellite loci and references in TableA2). All sampleswereamplifiedthroughPCRwiththeType-itMicrosatellitePCRkit(Qiagen,Hilden,Germany),with1μmol/Lofeachprimer.Annealingtemperaturesfollowedagradientfrom63°Cto55°Cin2°Cstepsandfinal amplificationoccurred for40cyclesat55°C.Allele sizeswere

determinedonanABI3130xlGeneticAnalyserusingGeneScan™ 500 ROX(bothThermoFischerScientific,Darmstadt,Germany)asinter-nalsizestandard.AlleleswerescoredwiththesoftwareGeneMapperv.4.0(AppliedBiosystems,Germany).

2.5 | Genetic diversity, phylogeography, and differentiation times

All mitochondrial sequences obtained were aligned using the autosettingasimplementedinMAFFTv7.245(Katoh&Standley,2013).The relationship among all haplotypes was reconstructed by amedian-joining(MJ)networkusingthesoftwareNETWORKv.4.6.1.4(Bandelt,Forster,&Röhl,1999).Haplotypesweregeneratedby re-movingnoninformativesitesandpositionswithgapsormissingdata.Haplotypicandnucleotidediversitiesforthefulldatasetandforeachspecieswere assessedwith the softwareDNASPv.5.10 (Librado&Rozas, 2009).We estimated genetic differentiation through FST as implemented in ARLEQUIN v.3.5.12 (Excoffier, Laval, & Schneider,2005).Forthisanalysis,wecreatedtwodatasets:(1)twopopulationscorrespondingtospeciesasdeterminedbythemuseumidentificationand(2)populationscorrespondingtomajorhaplotypeclades.

Thebestfittingsubstitutionmodelforthefullmitogenomedata-set(GTR+G+I)wasobtainedbythehierarchicallikelihoodratiotestasimplementedinJMODELTESTv2.1.7(Darriba,Taboada,Doallo,&Posada,2015).Wereconstructedphylogeneticrelationshipsthroughmaximumlikelihood(ML)withRAXMLGUIv1.5(Silvestro&Michalak,2012)andBayesianinference(BI)asimplementedinMRBAYESv3.2.6(Ronquist&Huelsenbeck, 2003), applying the determined substitu-tion model. Both approaches were congruent with the haplotypicnetworkandwitheachother.WedatedtheBItreebasedonfossilin-formationfortheCervidaestemoftheArctiodactylfamily(18.4Mya;Bibi,2013).Forthat,wefirstdetermineddivergencedatesforadata-set containing Bos javanicusasanoutgroup(JN632606.1),Muntiacus reevesi (AF527537.1), Axis axis (NC_020680.1) and A. porcinus (JN632600.1),D. dama (NC_020700.1),Cervus nippon (JN389444.1)and C. elaphus(AB245427.2),andthetwoRusaspeciesinvestigatedhere,usingthesoftwareBEAST(Drummondetal.2012).WerantwoMCMC chainswith 50million,with a lognormal uncorrelated clockandaYulespeciationtreeasfurtherestimationpriors.WeusedthecalibratedtimeofthesplitbetweenCervusandRusa(2.1My;HighestPosteriorDensity[HPD]=1–2.8My)tobesetasapriorfortherootheightofthegenetreeoftheRusasamplesobtainedinthisstudy,asbefore.TraceresultswereanalyzedwithTRACERv.1.6(implementedinBEASTv.1.8),forparameterconvergenceandESSvaluesabove200.TREEANNOTATORv.1.8.1(BEASTsoftwarepackage)wasusedtoan-notatealltrees,afteraburn-inof10%ofthetrees.AlltopologieswerevisualizedandeditedwiththesoftwareFIGTREEv.1.4.2(http://tree.bio.ed.ac.uk/software/figtree/).

2.6 | Population structure analyses

Analyses of microsatellite data proceeded by removing all indi-vidualswithmissingdata atmore than two loci.Weestimated the

1468  |     MARTINS eT Al.

probability for the presence of null alleles on our datasetwith thesoftwareFREENA(Chapuis&Estoup,2007).TestsforthepresenceoflinkagedisequilibriaandanexacttestfordeviationsfromHardy–WeinbergEquilibrium (HWE)wereperformedwithARLEQUIN, ap-plyingtheBonferronicorrectionformultipletests.Levelsofobserved(HO)andexpected(HE)heterozygosityandinbreedingcoefficient(FIS)werecalculatedinGENETIXv.4.05.2(Belkhir,Borsa,Chikki,Raufaste,&Bonhomme,1996–2004).

A Bayesian approach was used to test for population struc-ture with the software STRUCTURE 2.3.4 (Pritchard, Stephens, &Donnelly,2000).TheλvaluewasestimatedbyrunningaprospectiverunofK=1with10 iterationsandaburn-inof10%after15×104 generations.A secondMCMC simulationwas run for 20×104 gen-erations,witha10%burn-in.The likelihoodswereestimated forKvalues from1to6,because6washigher thanthemaximumnum-ber of mitochondrial clades obtained in our analyses. The admix-turemodelwasappliedwithcorrelatedallelefrequenciesandλ=3.STRUCTURE HARVESTER v.0.6.94 (http://taylor0.biology.ucla.edu/structureHarvester/;Earl&vonHoldt,2012)wasused toestimatethemostlikelynumberofKusingthe∆Kmethod(Evanno,Regnaut,&Goudet,2005).PopulationdifferentiationwascalculatedwiththesoftwareARLEQUINbyestimatingFSTbothamongtheclustersiden-tifiedby∆Kandbyspecies.

3  | RESULTS

3.1 | Mitochondrial genome analyses

Thefinaldatasetconsistedof56individualsforwhichamitogenomeof16,064bpwasobtained.Ofthese,23sampleswerelabeledinthemuseumcollectionsasR. timorensisand33individualswerelabeledasR. unicolor(TableA1).TheoriginofthreespecimenslabeledasR. uni-color(RUN37Moluccas,RUN39Timor,andRUN61Java)suggeststhatthesespecimensareactually Javandeer,R. timorensis.Allothermu-seumlabelsmatchedthegeographicaldistributionrangesofthetwospecies.The56deershared46haplotypeswithanoverallhaplotypediversityofH=0.983(SD0.010)andaverylownucleotidediversityofπ=0.00941(SD0.00118).WithinindividualslabeledasRTI,wefound18haplotypeswithH=0.972(SD0.026)andπ=0.0085(SD0.0024).WithinRUN-labeledindividuals,therewere28uniquehaplotypes,andonehaplotypewassharedduetoadmixturewithRTI (H_3).Overall,the29haplotypeshadH=0.991(SD0.011)andπ=0.011(SD0.0012).

ThefullmtDNAhaplotypenetworkseparatedtwomajorcladesbyaminimumof168mutationalsteps(Figure2).Thesmallercladecom-prised six individuals from Java, Bali, Moluccas, and South Sumatra(Figure2)ofwhichfourhadbeenlabeledinthemuseumcollectionsasR. timorensis(Javandeer,RTI)andtwoasR. unicolor(sambar,RUN;RUN41:SouthSumatra;RUN37:Moluccas,seeabove).Thesecondmajorcladecomprisedallothersamples,includingsampleslabeledasRTIfromJavaand fromthe introduction rangeofJavandeer (LesserSunda Islands,Sulawesi,andtheMoluccas).BothMLandBItreetopologieswerecon-cordantwiththeoverallpatternrecoveredbythehaplotypicnetwork,alsorevealingtheexistenceoftwowell-differentiatedclades(Figure3).

Generally,geneticdiversityofhaplotypesshowedgeographicalstruc-tureonlyinsomepartsofthetree(subcladesA,B,andC).SamplesfromtheMoluccas,Sulawesi,andTimor(outsideSundaland)werepresentinmorethanonebranchofthephylogenetictree(subcladeD).

TheresultingnodeagesfortheestimatedageofCervidaespe-ciesweresimilartothosereportedinotherstudies(Table1).Usingthose,wethenestimatedthetimingofdivergencebetweenthetwomaincladestohavestartedintheearlyPleistocene,about1.8mil-lion years ago (Mya) (HPD=0.95–3.1). The position of all cladeswassimilartothetopologyrecoveredbyML,withtheexceptionoftheSriLankancladeposition (cladeB,Figures3and4),whichdi-vergedmorerecentlyintheBItree.Thissubcladewasalsoaccom-panied by lowBayesian Posterior Probability values (BPP=0.34).Nevertheless, ourdivergenceestimates indicated that subcladeAdiverged first within the RUNmitogenome clade about 1.4Mya(HPD=0.7–2.3); subclade C diverged 1.185 Mya (HPD=0.6–2)andsubcladeDataround1.13Mya (HPD=0.6–2) (Figure4).FST showed significant population differentiation only when popula-tionswerebasedonmtDNAcladeassignment,butnotwhentheywere based on species assignment from the museum collections(Table2).

3.2 | Microsatellite analyses

Ofallarchivalsamplesforwhichmitogenomeswereobtained,16in-dividuals (~29%) could be successfully genotyped at 18 loci. Theseindividualsweredistributedrelativelywellacrossthecladesofthemi-togenometree(Figure3).Linkagedisequilibriawerefoundat10%ofallpairwiselocicombinations,yetwithoutanyconsistency,andper-centageofnullalleleswas0.18.Therefore,weretainedalllociforfur-theranalyses.WedetectedsignificantdeviationsfromHHWE,whichindicatedprobablepopulationstructurewithinourdataset.Expectedandobservedheterozygositiesateachlocusrangedfrom0.23(locusMu_4D)to0.92(locusMu_1_51)andfrom0(locusMu_4D)to0.61(locusRoe09),respectively(TableA2).Numberofallelesvariedamongloci,withthehighestnumberfoundatlociMu_1_51andNVHRT48(16alleles)andthelowestfoundatlocusMu_4Dwithonlytwoalleles(TableA2).

According to the∆Kapproach, themost likelynumberofgeno-typicnDNAclusterswasK=2(Figure5).Thesetwomainclusterscor-respondedwelltothetwospecies,sambar(R. unicolor,greencluster)andJavandeer (R. timorensis, redcluster).Populationdifferentiation(FST)wasalwayssignificantbetweenthetwospecies,independentofthegroupingmethod(museumassignmentsorSTRUCTUREanalyses,Table2).

4  | DISCUSSION

ThemitochondrialgenomesandthenDNAlociofsambarandJavandeer investigatedhere revealedan intriguingandsurprisingpatternofgeneticdiversityandpopulationdifferentiationbetweenthetwospecies.AlthoughmonophylyofR. timorensis and R. unicolor remain

     |  1469MARTINS eT Al.

undisputed,our resultspoint toamorecomplexhistoryofhybridi-zationbetweenspeciesandmultiplehuman-mediated introductionsoutsidetheSundaShelf.

Thepresenceoftwodivergentmatrilineagesclearlyindicatesmo-leculardifferentiationbetweentwogroupsofRusadeer,whichwein-terpretasthehistoricalcladogenesisofbothR. timorensis and R. unicolor. Ourfossil-calibratedestimatesarecorroboratedbyrecentstudies(Bibi,2013;Escobedo-Morales,Mandujano,Eguiarte,Rodriguez-Rodriguez,&Maldonado,2016;Table1)andareinaccordancewiththedatessug-gestedbyotherauthorsfortheageofthegenusRusa(e.g.,2–2.5Mya;DiStefano&Petronio,2002).Theseparationofthetwodeerspeciesinvestigatedherehadbeenchallengedinthepast (mentionedinVanBemmel,1949),yetsubsequentstudiesfoundrobustsupportfortheirdistinctiveness(morphological:vanBemmel,1994;Meijaard&Groves,2004; andmolecular: Emerson&Tate, 1993; Pitra, Fickel,Meijaard,& Groves, 2004). Our comprehensive molecular study corroboratedthese findings,butalsoprovidesevidence foramuchmorecomplexevolutionaryhistoryoftheRusadeer.IthasbeenproposedthatRusa-like deer have appeared in Northern India, around 2.5Mya, wheretheyadaptedtodenseforesthabitatswithsomeopengrassvegeta-tion(Geist,1998).However,duringthePleistocene,subtropicalforest

shiftedsouthwards, completelydisappearing fromChina (Meijaard&Groves, 2004).Thiswould have also shifted the distribution area ofsubtropicalforest-adaptedRusa(orRusa-like)speciessouthwardstoo.When low sea level allowed,Rusadeer couldhave reachedSundaicislands,includingJava.Sealevelremainedlowuntil1.4Mya,maintain-ingconnectionsbetweenlandmassesthroughtheemergedcontinentalshelf(VanDenBergh,DeVos,&Sondaar,2001).By1Mya,sealevelhadrisenagainandhadreachedahighstandat+5mcomparedtopresentday(Zazo,1999),therebyinterruptinglandbridgesbetweenislands.Atthistime,RusapopulationsofJavaandSumatra(cladeD)wouldhavebecomeisolatedandhabitatavailabilityforforest-dependentspecieswouldhavebeenreduced.

Our data indicate that therewas also a secondwaveof colo-nization to Sundaic islands by Rusa unicolor, likely from Thailand(Mainland). This second wave would have likely occurred duringtheLatePleistocene,withdropsinsealevelsandonceagaincoolerand drier climates. This southward expansion brought previouslyisolated Rusa unicolor in contact with Rusa timorensisfromJava,fa-cilitatedbythepresenceoftheemergedSundaStrait,astraitthatsubmerged just ~10kya (Sathiamurthy & Voris, 2006). This factraisestheobviousquestionofwhatthenmaintaineddifferentiation

F IGURE  2 Haplotypicnetworkforall46haplotypessharedamongthetwospecies.Circlesizeisinaccordancewithfrequencyandcolorrepresentssamplinglocation.Smallblackcirclesrepresentmedianvectors.Allbranchesrepresentonemutationstep,exceptwhenindicatedotherwisebynumbersonbranches.Twomajorcladeswererecoveredandareindicatedbythespeciesnames.HaplotypesaredescribedinTableA1

H_39

H_16H_14

H_46 H_21

H_44

H_40H_45

H_31

H_20

H_7H_2

H_43

H_28 H_29

H_25

H_17

H_19H_13

H_23 H_30H_41

H_35

H_34 H_5

H_6

H_33

H_9

H_42

H_18

H_32

H_8H_4 H_15

H_38

H_26

H_22

H_24

H_27

H_1

H_37

H_10

H_36

H_12

H_11

H_3

4

4 5

168

35

4

104 26 4

12

3 8

10

5511

6

2

33512

3

2

2

10

3

26

7 3

4

2

4

51

71

5147

28

2

3

36

18

9

420

IndochinaMentawai

SumatraBorneo

The Moluccas

JavaBaliSulawesiTimor and LSI

IndiaSri Lanka

R. timorensis R. unicolor

1470  |     MARTINS eT Al.

betweenthetwospeciesand/orrestrictedhybridizationtoasmallsecondarycontact region.Wehypothesize that thedifferenteco-logicalnichesonJavaandSumatramighthavehadacentral role.Speciation may have resulted from the ecological adaptation ofJavandeerRusa timorensistotheprevailingvegetationtypeonJava,separatingitfromitssisterspecies,thesubtropicalforest-adaptedsambar Rusa unicolor. Java and Bali, although part of Sundaland,had(andstillhave)differentclimaticconditionsthanSumatraand

Borneoandthusalloweddifferentiationbetweenspeciesbasedonevolvedecological adaptations (Leonardetal., 2015).The climateonJavaischaracterizedbyaWest–Eastgradient,atransitionfromaslightlyseasonalclimate in theWest toastronglyseasonaloneintheEast.CentralandEastJavaarecharacterizedbydrier,coolerclimate(climate-data.org),andthevegetationhasmoregrassareasthan on the surrounding islands (Heany, 1991; Mishra, Gaillard,Hertler,Moigne,& Simanjuntak, 2010).Therefore, it is likely that

F IGURE  3 Mitogenomemaximumlikelihoodtreeofbothspecies.Colorsontipsrepresentsamplinglocation(asinFigure2)andstarsrepresentspliteventswithbootstrapvalues/Bayesianposteriorprobabilitieslowerthan90/0.95(butbiggerthan50/0.5).RedandgreendotsrepresentsamplesforwhichweobtainednDNA;reddot:assignedtotheRusa unicolorgenotypicclusterandgreendot:assignedtotheRusa timorensisgenotypiccluster.MajormtDNAcladesandsubcladesarelabeledwithcurvedbrackets.Scalebarindicatesnumberofsubstitutionsperposition

0.004

RTI27_Sulawesi

RTI8_Java

RUN61_Java

RUN21_Borneo

RUN3_SriLankaRUN45_ESumatra

RUN41_Sumatra

RUN37_Seram

RTI31_Sulawesi

RTI20_Sulawesi

RUN35_SriLanka

RTI30_Java

RUN55_India

RUN12_Borneo

RTI19_SSulawesi

RUN10_SriLanka

RTI17_Sulawesi

RTI3_Moluccas

RUN42_Sumatra

RTI44_Bali

RUN28_Thailand

RUN19_NBorneo

RTI18_SSulawesi

RUN65_Sumatra

RUN38_Sumatra

RTI38_WJava

RTI22_Lombok

RUN6_NBorneo

RTI33_Moluccas

RUN5_NBorneo

RUN58_Myanmar

RUN56_SriLanka

RUN20_Mentawai

RUN34_Java

RTI9_Java

RUN66_Sumatra

Run33_India

RTI39_Timor

RUN62_India

RTI43_Timor

RUN26_Borneo

RUN2_Sumatra

RTI32_JavaRUN32_Sumatra

RUN43_Indochina

RUN47_SThailandRUN44_Thailand

RTI2_Dompoe

RTI37_WJava

RUN54_NIndia

RUN39_Timor

RTI6_Timor

RUN51_SThailand

RTI21_Timor

RTI13_Java

RTI45_Bali

Rusa unicolor

mtD

NA clade

Subclade A

Subclade B

Subclade C

Rusa timorensismtDNA clade

Subclade D

Split

Age

Other studiesMedian Minimum Maximum

Bovidae/Cervidae 16.8 10.7 23.3 18.4(Bibi,2013)

Muntiacus/Cervus 7.2 4.1 10.2 7.5(Martinsetal.2017)

Axis/Cervus/Dama 4.6 2.6 5.4 6 (DiStefanio&Petronio,2002)

Cervus/Rusa 2.1 1 2.8 3.4(Pitraetal.2004)

R. unicolor/timorensis 1.4 0.7 2.2 –

Ages(inmillionyears[My])representthemedianobtainedforeachofthedescribedsplit.Valuesinboldrepresentfossil-basedcalibrations.

TABLE  1 CalibrateddivergencedatesestimatedfortheCervidaetree

     |  1471MARTINS eT Al.

R. timorensis, being better adapted to drier climate, would havecrossed the dry central Sundaland during Pleistocene glacials tocolonizeeastJava (Sheldon,Lim,&Moyle,2015),where itstayedisolatedfromitssisterspecies.Afterthisinitialseparation,wefindevidenceofrangeexpansion,likelyduringconsequentdropsinsealevels,demonstratedbytheintrogressioninJavaandpossiblySouthSumatra.

One sambar individual from South Sumatrawas found to carryRTImtDNA.This indicatesthepossibilitythat individualsofR. timo-rensisalsomigratedtoatleastSouthSumatra,wheretheyhybridizedwith R. unicolor.SucharangeexpansionwouldhavebeenenabledbythedrierandcoolerclimatesandtheemergedlandcorridorbetweenSumatraandJavaduringtheLatePleistocene,asatthetimeofLGM,WestJavapresenteddrierandcoolerclimates in the lowlands (Sun,Li,Luo,&Chen,2000).Becausewedidnotobtain thegenotypeof

F IGURE  4 MitochondrialDNAdatedtreeaccordingtoBEASTanalyses,from3Myatopresent.Bluebarsrepresentassociateddeviationsforthemostimportantsplits.Atimescaleinmillionsofyearsandaroughestimateofsealevelchangesthroughtime(adaptedfromPatouetal.2010)arepresentedbelow

0.511.522.53

RTI20_Sulawesi

RTI13_Java

RUN5_NBorneo

RTI3_Mollucca

RTI38_WJava

RTI6_Timor

RTI19_SSulawesi

RTI17_Sulawesi

RTI32_Java

RTI27_Sulawesi

RTI2_Dompoe

RUN28_Thailand

RUN66_Sumatra

RUN2_Sumatra

RTI9_Java

RUN10_SriLanka

RUN55_India

C_elaphus

RUN26_Borneo

RUN56_SriLanka

RTI22_Lombok

RUN32_Sumatra

RTI33_Mollucas

RTI8_Java

RUN43_Indochina

RUN44_Thailand

RUN35_SriLanka

RUN51_SThailand

RTI39_Timor

RTI45_Bali

RTI37_WJava

RUN3_SriLanka

RUN62_India

RUN20_Mentawai

RUN45_ESumatra

RUN6_NBorneo

RUN54_NIndia

RUN42_Sumatra

RTI21_Timor

RUN38_Sumatra

RUN19_NBorneo

RTI31_Sulawesi

RUN41_Sumatra

RUN39_Timor

RUN12_Borneo

Run33_India

RTI43_Timor

RTI18_SSulawesi

RUN37_Seram

RUN58_Myanmar

RTI30_Java

RUN34_Java

RUN21_Borneo

RUN61_Java

RUN65_Sumatra

RUN47_SThailand

RTI44_Bali

C

A

B

+50

– 50

0 m

D

TABLE  2 Populationdifferentiationestimates(FST)accordingtomarkerandgrouping

FST p- value

mtDNA

MuseumID 0.085 >.001

Results(2clades) 0.75 <.001

nDNA

MuseumID 0.12 <.001

Results(K=2) 0.14 <.001

EstimationswereperformedbothforthemtDNAandnDNA.PopulationsweregeneratedbyeithermuseumID(bothformtDNAandnDNA)andbyassignmentof individualstooneofthetwomajorclades (mtDNA)ortooneofthegenotypicclusters(nDNA).Statisticallysignificantcomparisonsareindicatedbyboldp-values.

1472  |     MARTINS eT Al.

thissample,moreintensivesamplingofSouthSumatranpopulationswouldberequiredtoconcludethattheseresultsreflectevidenceofreciprocal hybridizationbetween two sister speciesofdeer.On theotherhand,wefoundastrongevidencethatthesambar,likelyduringwetterinterglacialperiods,expandeditsrangefromSumatraatleasttowesternJavawhereithybridizedwiththeJavandeer.Thoseintro-gressedJavandeermostlikelyconstitutedthesourcepopulationfortheintroductionseastoftheWallaceline.

4.1 | Phylogeography and taxonomy of Rusa timorensis

WithinR. timorensis,individualsfromBaliandWestJavashowedgeneticdivergenceatmtDNA.ThissubstructurecouldindicatelimitedgeneflowduringpartsoftheLatePleistocene,whichmightcorroboratetheclas-sificationofBalipopulationsasR. t. renschi,withgeneticisolationmostlikelybeingtheresultofa“smallpopulationeffect”andalimited/inter-ruptedgeneflowtoJavanpopulations.However,becauseweonlyhadtwosamplesfromBaliandnosamplesfromEastJava,ourassessmenthastobeviewedcautiously.Itdoeshoweverindicatetheespeciallyur-gentneedtoassessifhybridizationoccurredaswellinthesepopulations,throughmoreextensivesamplingandinclusionofnuclearmarkers.

ThemtDNAofRTI-labeledsamplesfromislandsbeyondtheWallaceline clustered with R. unicolormtDNAbutdidnotshowanycleargeo-graphical distribution pattern. These RTI hybrids shared haplotypeswithRUN-labeledsamplesfromSumatraandBorneo(Figures2and3).GeneticdistancesamongRTIhybridhaplotypesandtootherRUNhap-lotypeswereverylow,indicatingarecent,thushuman-mediatedintro-ductiontotheseWallaceaislands.Quiterecently,ithadbeensuggestedto splitR. timorensis into seven subspecies according to their occur-renceonislandswithinandoutsideoftheSundashelf(Mattioli,2011;Hedges,Duckworth,Timmins,Semiadi,&Dryden,2015).However,ourdatadonotsupportsuchasuggestion,asallsamplesfromtheintro-duction range shared haplotypes, indicating a lack of differentiationamongindividualsfromtheseWallaceanislands.Furthermore,thefewsamplesfromJavaandTimorthatcouldbegenotypedshowedgeneticsimilarity,againindicatingthelackofdifferentiation.

4.2 | Phylogeography and taxonomy of Rusa unicolor

Sambar is currently subdivided into five subspecies: R. u. unicolor (India,Nepal,Bangladesh,andSriLanka),R. u. brookei (Borneo),R. u. cambojensis(mainlandSoutheastAsia,fromSouthChina/HainanandMyanmartoPeninsulaMalaysia),R. u. equine(SumatraandMentawai),and R. u. swinhoei(Taiwan;Mattioli,2011).Themitogenomestructurerecoveredhere,however,didnotsupportanyofthedescribedsub-species,asitindicatedgeneflowbetweenallpopulations.Especiallyamong populations of Sundaic islands, we found a lack of geneticstructurethatwouldcorrespondtoisolatedislands,evidencedbythepresenceofindividualsdistributedthroughoutthetreetopologiesandhaplotypicinferences.

Amongpopulationsofsambar,wefoundevidenceofatleastthreedeepsplit (>1My)subcladeswhichwerenot inaccordancewith thecurrentsubspeciesassignment.SubcladeAcomprisedhaplotypesfromMyanmarandIndia,withanageofabout1.36My;thesecondcladeincluded Sundaic populations from Sumatra, Mentawai, and Borneo(cladeC)andwasdatedtobe~1.18Myold;andthethirdoneincludedall haplotypes fromSri Lanka (cladeB) and split from the remainingpopulations ~1.13Mya. Subclade D encompassed all remaining in-dividuals of sambar, both fromMainland South and Southeast Asiaand theSundaic islands.Despite the ancient split of cladeA furthersampling ofmainland southeastAsia, particular India,Myanmar, andBangladesh isneededtorevealwhethercladeA is indeedgeograph-icallyseparatedfromtheothermainlandAsianpopulations,particularthoserepresentedinsubcladeD.Thus,albeithistoricallycladeAandDwereisolated,ourdataindicaterecentgeneflowbetweenthetwoclades,andthus,itremainsuncertainwhetherourdatawouldsupportasubspecificstatusofindividualsfromcladeA.Verysimilarly,SumatranindividualswerepresentinthedistinctmitochondrialcladesCandD,andthus,ourdatadidnotsupportseparatingthesecladesindistincttaxonomicunits.

These subclades could then rather represent centers of sambardistribution, which would have remained in place during times ofwarmerandwetterconditions,whichcontractedsambarpopulationsto subtropical refugia. From these centers, we observed waves of

F IGURE  5 Genotypingresultsfrom16individualsgenotypedfor18loci,showingastructureplotforK=2,withR. timorensis samplesingreenandR. unicolor individuals inred.Eachcolumnrepresentsasingleindividual,asidentifiedbelow

0.80

0.60

1.00

0.40

0.20

0.00

I45_

Bali

RTI4

3_Ti

mor

RTI3

9_Ti

mor

RUN

61_J

ava

RTI4

4_Ba

liRT

I6_T

imor

RUN

58_M

yanm

arRU

N56

_Sri

Lank

aRU

N43

_Ind

ochi

naRU

N3_

Sri L

anka

RUN

12_B

orne

oRU

N19

_Bor

neo

RUN

20_M

enta

wai

RUN

66_S

umat

raRU

N5_

Born

eoRU

N32

_Sum

atra

     |  1473MARTINS eT Al.

expansion.ThebranchingorderofthesethreeoldsubcladesindicatedcolonizationfromnorthernIndochinasouthwardstoSriLankaandtotheSundaShelf, respectively.The “younger” individualswithin sub-cladeDfromIndia,aswellasfromSumatraandBorneo,appearthentobedescendantsfromasecond/thirdnaturaldispersionwave(pos-sibly fromThailand) during glacial periods of thePleistocene,whenlow sea level again exposed the shallow Sunda shelf connecting allmajor islands (Voris, 2000; Bird, Taylor, &Hunt, 2005). During gla-cialperiods, climatewasdrierandcooler in tropical regions (Gorog,Sinaga,&Engstrom, 2004).However, species that retained a broadecological niche such as Rusa unicolorwouldhavebeenabletoutilizethenewly emergedhabitats Such a scenariowould likely cause thehaplotypicdistributionpatternweobservedhere.Duringglacialperi-ods,SundalandwasalsoconnectedtoSoutheastAsianmainland,al-lowingsecondaryadmixturebetweenformerlyseparatedpopulations,thus generating the patternswe observe between haplotypes fromThailand,India,andSundaland.

Incontrast, thedistinctSri LankancladeBprovidesadditionalev-idence for the recognitionofSriLankansambarasbeingdistinct.Thissupportcomesbothfrommorphologicalassessments(Groves&Grubb,2011)andkaryotypedifferences(2n=56inSriLankansambarvs.2n=58inIndianand2n=62inChineseandMalaysiansambar;Leslie,2011).SriLankanpopulationsareoftenmoregenetically related to theWesternGhatsthantootherIndianregions.Veryrecently,a40bpinsertionwasdetectedinthecontrolregionofthemitochondrialDNAinsamplesfromtheWesternGhats(Gupta,Kumar,Gaur,&Hussain,2015),whosepres-encewe,however,wereunabletoverifyduetomethodlimitations.

Thus,furtherstudiesonthemainlandpopulationsthatcouldcon-firmthepresenceofoldsplitswithinR. unicolorareofurgency.Increasedsamplingand,especially,theinclusionofnuclearmarkerscouldconfirmtheextentofgeneflowbetweenthesepopulationsor,conversely,truegeneticdivergencebetweenthesubcladesrecoveredhere.Ifconfirmed,thesepopulationsmight represent subspeciesof sambaroccurring inhighlydisturbedregionsofMainlandSoutheastAsia.

4.3 | Introductions past the Wallace line

It is generally accepted that the presence of Rusa deer on islandsbeyond Sundaland (excluding Philippines) was the result of humaninterference (Long, 2003; Groves & Grubb, 2011; Hedges etal.,2015).However,untilnow,theseindividualswereassumedtohavebeenpureR. timorensis,collected,andtransportedforvenisonandasgamespeciesbyhumansfromtheislandsofJavaandBaliduringtheHolocene(Heinsohn,2003).WhilethenDNAdataclearlyseparatedthetwospecies—beinghighlyconcordantwiththeirdescriptionfromthemuseumcollections—themitochondrialgenomespointtoamoresurprisingpatternofpastPleistocenehybridizations.Human-mediatedintroductionsofthesedeeroccurredca.3,000yearsago(Heinsohn,2003),and,toourknowledge,theseresultscouldthereforeconstituteevidencefortheearliesttransportofintrogresseddeer.

AllmtDNAhaplotypesofsampleslabeledinthemuseumcollec-tions as R. timorensisandsampledfromWallacean islands (Sulawesi,Lesser Sunda Islands and TheMoluccas) were of R. unicolor origin,

renderingtheseindividualshybriddescendants.Themostparsimoni-ousexplanation for themolecularpatternsobtained in this study isthathybridizationoccurredonJava(centerofstar-likepatterninthehaplotypicnetwork),withnaturaldispersionoffemalesambar.Theseimmigrated individuals were likely from Sumatra and/or the Thai-MalayPeninsula,asindicatedbythebasalpositionofthetwoSouthernThailandindividuals(RUN51andRUN57),andtheywouldhaveusedtheconnectinglandbridges.AftertheintrogressionofSambarhaplo-types intoJavanpopulations, humanswould thenhave transportedthe introgressed descendants of Pleistocene hybrids from Java toTimor,theMoluccasandSulawesi.Despiteevidenceformultiple in-dependentintroductions(Section4),almostallintroducedindividualscarriedsambarhaplotypes(exceptRUN37fromSeram,theMoluccas).Thisindicatesthateitherhumansselectedforindividualstobeintro-duced(e.g.,carryingaparticulartraitonlyfoundinintrogressedJavandeer);thattheintrogressedindividualshadahighersurvivingprobabil-ityaftertheirintroduction;orthatmostintroductionsoccurredfromasingleregion(e.g.,WestJava)whereRUNhaplotypesgotfixed,possi-blethroughmitochondrialcaptureofsambarhaplotypes.

Introduction of Javan deer to Timor seems to have occurredonlyonceand,presumably,withveryfewfoundersbecauseofthelackofmtDNAdiversity foundamongall individuals. In fact,pop-ulations recently introduced toAustralia andNewCaledonia froma low,knownnumberof individualsfromTimor,havebeenshowntohavevery lowgenomicdiversity,whichwouldbe theexpectedresultafteranintroductionofindividualsthathadcomefromanal-readygeneticallyimpoverishedpopulation(Webley,Zenger,English,&Cooper,2004;deGarine-Wichatitskyetal.,2009).AlthoughweobtainedonlyveryfewsamplesfromtheMoluccasandotherLesserSundaIslands(DompoeandLombok),theydidnotsharehaplotypes,indicatingeithermultipleintroductionsorahighernumberoffound-ers.SulawesihadbyfarthemostgeneticallydiverseJavandeerpop-ulationofalltheWallaceanislands.ItshaplotypeswerepresentinalmostallyoungercladesofthemtDNAtree.ThispatternindicatedthatRusadeerreachedSulawesimultipletimes.Onesample(RTI18)was closer related to Bornean populations than the other haplo-types from Sulawesi. Although natural dispersal from Borneo toSulawesiovertheMakassarStraitisconceivable,itishighlyunlikelyasthelastpossibleconnectionbetweenthesetwolandmasseswasduringthelatePliocene/earlyPleistocene~2.5Mya,adatethatbyfarpredatestheemergenceofthismtDNAlineage.Themostlikelyscenario is a human-mediated introduction of Bornean sambar toSulawesiwhereithybridizedwiththeintroducedJavandeer.Iftrue,this represents a second hybridization event on Sulawesi (com-paredtotheLatePleistocenehybridization inJavaandpotentiallySouthSumatra)and further studiesonSulawesiJavandeerwouldberequiredtotestthishypothesis.TheremaininghaplotypesfromSulawesiindividualswerecloselyrelatedtothehaplotypesfromin-dividuals introducedtotheMoluccasandTimor Islands, indicatingeitherthatallofthemhavebeenintroducedinonewaveoratleastfromasimilarsourcepopulationfromJava.

This is the first reportofhistoricalhybridizationsbetweensam-bar (R. unicolor) and Javan deer (R. timorensis). Occurrence of such

1474  |     MARTINS eT Al.

hybridizationhadbeendocumentedbefore,namelybetweenR. timo-rensisindividualsintroducedtoBorneowiththelocalBorneanR. uni-color(WestKalimantan,nowpossiblyextinct,Hedgesetal.,2015)andattainedthroughhusbandrybefore(Leslie,2011).Hybridizationswithfertileoffspringhavealsobeenreportedtooccurbetweenotherdeerspecies,amongothersbetweensambarandreddeerCervus elaphus (Muiretal.,1997)andreddeerandSikaCervus nippon(Smith,Carden,Coad,Birkitt,&Pemberton,2014).

This fact has potentially important conservation implications forthetwoRusaspeciesanalyzedinthisstudy.DespitebeingoneofthemostwidespreaddeerspeciesinsouthernAsia,R. unicoloristodaynolonger abundant throughoutmost of its native range (Timmins etal.,2015).Likewise,R. timorensis iscurrentlyconsideredapestspecies inareaswhere it has recently been introduced (e.g., Australia) but has,however,decreasedlargelyinpopulationnumbersinnativeandhistor-icalintroductionregions(Java,LesserSundaIslands,Sulawesi,andtheMoluccas).BothRusaspeciesstudiedherearenowconsideredvulner-ablebytheIUCN/RedListofThreatenedSpecies(Hedgesetal.,2015;Timminsetal.,2015).Therefore,geneticmonitoringofindividuals,bothatmtDNAandparticularlyalsonucleargenomes,isnecessarytoassesswhetherpureRTIandRUNindividualsarebeingintroduced(orrepro-ductivelyassistedintheirnativeranges)andnotintrogressedindividu-als.Moreover,moreintensiveandextensivesamplingofR. timorensis on theirnativerangeisnecessarytodiscernwhetherpureRTIpopulationsstillremaininJavaandBaliorwhethertheyarecomposedintheirma-joritybyhybridindividuals.

5  | CONCLUSION

In addition to representing the first comprehensive phylogeo-graphicalstudyonR. unicolor and R. timorensis,thisstudyrevealedsurprising evolutionary histories of these two sister species.Answering the questions poised before, we hypothesized thatwhile climateadaptationswere likely responsible formaintainingspecies monophyly, Pleistocene climate changes were responsi-ble forsecondarycontactandconsequenthybridizationbetweensambar and Javan deer.We recovered a pattern of (possibly re-ciprocal) introgressions between the two species, facilitated bythepresenceof landcorridorsduringperiodsof lowsealevels inSundaland. The introgressed populations of Javan deer on Javawere then the source of all human-mediated introductionwavestotheislandseastoftheWallaceline,aswefoundthatallR. timo-rensis individuals carried R. unicolorhaplotypes.Additionally,thesedramaticclimatechangeswerealsolikelyresponsibleforthediver-genceofpopulationswithinR. unicolor.

ACKNOWLEDGMENTS

The authors would like to thank all curators listed in TableA1 forprovidingaccesstotheircollections.ThisprojectwasfundedbytheLeibnizcompetitivefundSAW-2013-IZW-2.

CONFLICT OF INTEREST

Nonedeclared.

AUTHOR CONTRIBUTIONS

RFM,AW,andJFdesignedthestudy;RFMandASperformedthelab-oratoryprocedures;RFMandDLperformedthebioinformaticanaly-ses;RFM,AW,andJFwrotethemanuscript.Allauthorscontributedequallyforthefinalversionofthismanuscript.

ORCID

Renata F. Martins http://orcid.org/0000-0001-8145-6998

REFERENCES

Bacon,C.D.,Henderson,A.J.,Mckenna,M.J.,Milroy,A.M.,&Simmons,M.P.(2013).GeographicandTaxonomicdisparitiesinspeciesdiversity:Dispersal and diversification rates acrossWallace’s line. Evolution, 67,2058–2071.https://doi.org/10.1111/evo.12084

Bandelt,H.J.,Forster,P.,&Röhl,A.(1999).Median-joiningnetworksforinfer-ringintraspecificphylogenies.Molecular Biology and Evolution,16,37–48.https://doi.org/10.1093/oxfordjournals.molbev.a026036

Belkhir,K.,Borsa,P.,Chikki,L.,Raufaste,N.,&Bonhomme,F.(1996–2004)Genetix4.05,logicielsousWindowsTMpourlagenetiquedespopula-tions.CNRSUmr5000.

Bibi, F. (2013). A multi-calibrated mitochondrial phylogeny of extantBovidae (Artiodactyla, Ruminantia) and the importance of the fossilrecord to systematics. BMC Evolutionary Biology, 13, 166. https://doi.org/10.1186/1471-2148-13-166

Bird, M. I., Taylor, D., & Hunt, C. (2005). Palaeoenvironments of insularSoutheast Asia during the Last Glacial Period: A savanna corridor inSundaland? Quaternary Scientific Reviews, 24, 2228–2242. https://doi.org/10.1016/j.quascirev.2005.04.004

Chapuis,M.P.,&Estoup,A. (2007).Microsatellitenull alleles andestima-tion of population differentiation.Molecular Biology and Evolution, 24,621–631.

Darriba,D.,Taboada,G. L.,Doallo, R., & Posada,D. (2015). jModelTest 2:Moremodels,newheuristicsandhigh-performancecomputing.Nature Methods,9,6–9.

deGarine-Wichatitsky,M.,deMeeus,T.,Chevillon,C.,Berthier,D.,Barre,N.,Thevenon,S.,&Maillard,J.-C.(2009).Populationgeneticstructureofwildandfarmedrusadeer(Cervus timorensis russa)inNew-Caledoniainferredfrompolymorphicmicrosatelliteloci.Genetica,137,313–323.https://doi.org/10.1007/s10709-009-9395-6

Demos,T.C.,Achmadi,A.S.,Giarla,T.C.,Handika,H.,Rowe,K.C.,&Esselstyn,J.A.(2016).Localendemismandwithin-islanddiversificationofshrewsillustrate the importance of speciation in building Sundalandmammaldiversity.Molecular Ecology, 25, 5158–5173. https://doi.org/10.1111/mec.13820

Di Stefano, G., & Petronio, C. (2002). Systematics and evolution of theEurasian Plio-Pleistocene tribe Cervini (Artiodactyla, Mammalia).Geologica Romana,36,311–334.

Drummond,A. J., Suchard,M.A., Xie,D.,&Rambaut,A. (2012).Bayesianphylogenetics with BEAUti and the BEAST 1.7.Molecular Biology and Evolution,29,1969–1973.https://doi.org/10.1093/molbev/mss075

Earl,D.A.,&vonHoldt,B.M.(2012).STRUCTUREHARVESTER:AwebsiteandprogramforvisualizingSTRUCTUREoutputand implementingtheEvannomethod.Conservation Genetic Resources,4,359–361.https://doi.org/10.1007/s12686-011-9548-7

     |  1475MARTINS eT Al.

Emerson,B.C.,&Tate,M.L.(1993).Geneticanalysisofevolutionaryrelation-shipsamongdeer(SubfamilyCervinae).Journal of Heredity,84,266–273.https://doi.org/10.1093/oxfordjournals.jhered.a111338

Escobedo-Morales, L. A., Mandujano, S., Eguiarte, L. E., Rodriguez-Rodriguez, M. A., &Maldonado, J. E. (2016). First phylogenetic anal-ysis of Mesoamerican brocket deer Mazama pandora and Mazama temama (Cetartiodactyla:Cervidae)basedonmitochondrial sequences:ImplicationsonNeotropicaldeerevolution.Mammalian Biology,81,303–313.https://doi.org/10.1016/j.mambio.2016.02.003

Evanno, G., Regnaut, S., & Goudet, J. (2005). Detecting the num-ber of clusters of individuals using the software STRUCTURE: Asimulation study. Molecular Ecology, 14, 2611–2620. https://doi.org/10.1111/j.1365-294X.2005.02553.x

Excoffier, L., Laval, G., & Schneider, S. (2005). Arlequin (version 3.0): Anintegrated software package for population genetics data analysis.Evolutionary Bioinformatics Online,1,47–50.

Fickel,J.,&Reinsch,A.(2000).MicrosatellitemarkersfortheEuropeanRoedeer (Capreolus capreolus).Molecular Ecology,9,993–1011.

Fortes,G.G.,&Paijmans,J.L.A.(2015).Analysisofwholemitogenomesfromancient samples.Whole Genome Amplification Methods Protocols,1347,179–195.https://doi.org/10.1007/978-1-4939-2990-0

Geist, V. (1998). Deer of the world. Their evolution, behaviour, and ecology. Mechanicsburg,PA:StackpoleBooks.

Gorog, A. J., Sinaga, M. H., & Engstrom,M. D. (2004). Vicariance or dis-persal? Historical biogeography of three Sunda shelf marine rodents(Maxomys surifer, Leopoldamys sabanus and Maxomys whiteheadi).Biological Journal of the Linnean Society, 81, 91–109. https://doi.org/10.1111/j.1095-8312.2004.00281.x

Groves,C.P. (1983).PigseastoftheWallaceLine.Journal de la Societe des Oceanistes,39,105–119.https://doi.org/10.3406/jso.1983.2792

Groves,C. P.,&Grubb, P. (2011).Ungulate Taxonomy. Baltimore,MD:TheJonhsHopkinsUniversityPress.

Gupta,S.K.,Kumar,A.,Gaur,A.,&Hussain,S.A.(2015).Detectionof40bpinsertion-deletion(INDEL)inmitochondrialcontrolregionamongsam-bar(Rusa unicolor)populationsinIndia.BMC Research Notes,8,1–7.

Heany, L. R. (1991). A synopsis of climate and vegetational change inSoutheast Asia. Climatic Change, 19, 53–61. https://doi.org/10.1007/BF00142213

Hedges, S., Duckworth, J. W., Timmins, R. J., Semiadi, G., & Dryden, G.(2015). Rusa timorensis. IUCN Red List Threatened Species, https://doi.org/10.2305/IUCN.UK.2015-2.RLTS.T41789A22156866.en

Heinsohn,T. (2003).Animal translocation: Long-termhuman influencesonthevertebratezoogeographyofAustralasia(naturaldispersalversuseth-nophoresy).Australian Zoologist,32,351–376.https://doi.org/10.7882/AZ.2002.014

Katoh, K., & Standley,D.M. (2013).MAFFTmultiple sequence alignmentsoftwareversion7:Improvementsinperformanceandusability.Molecular Biology and Evolution, 30, 772–780. https://doi.org/10.1093/molbev/mst010

Kühn,R.,Anastassiadis,C.,&Pirchner,F. (1996).Transferofbovinemicro-satellites to the cervine (Cervus elaphus).Animal Genetics,27,199–201.

Kühn, R., Schröder, W., Pirchner, F., & Rottmann, O. (2003). Geneticdiversity, gene flow and drift in Bavarian red deer populations(Cervus elaphus). Conservation Genetics, 4, 157–166. https://doi.org/10.1023/A:1023394707884

Latinne,A.,Meynard,C.N.,Herbreteau,V.,Waengsothorn, S.,Morand, S.,&Michaux, J. R. (2015). Influence of past and future climate changesonthedistributionofthreeSoutheastAsianmurinerodents.Journal of Biogeography,42,1714–1726.https://doi.org/10.1111/jbi.12528

Leonard,J.A.,denTex,R.-J.,Hawkins,M.T.R.,Munoz-Fuentes,V.,Thorington,R., & Maldonado, J. E. (2015). Phylogeography of vertebrates on theSunda Shelf: A multi-species comparison. Journal of Biogeography, 42,871–879.https://doi.org/10.1111/jbi.12465

Leslie Jr, D. M. (2011). Rusa unicolor (Artiodactyla: Cervidae).Mammalian Species,43,1–30.https://doi.org/10.1644/871.1

Li, H., & Durbin, R. (2009). Fast and accurate short read alignment withBurrows–Wheelertransform.Bioinformatics,25,1754–1760.https://doi.org/10.1093/bioinformatics/btp324

Librado,P.,&Rozas,J.(2009).DnaSPv5:Asoftwareforcomprehensiveanal-ysisofDNApolymorphismdata.Bioinformatics,25,1451–1452.https://doi.org/10.1093/bioinformatics/btp187

Long,J.L. (2003). Introduced mammals of the world. Their history, distribution and influence.Clayton,VIC:CSIROPublishing.

Maricic, T.,Whitten, M., & Pääbo, S. (2010). Multiplexed DNA sequencecapture of mitochondrial genomes using PCR products. PLoS ONE, 5,e14004.https://doi.org/10.1371/journal.pone.0014004

Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet Journal, 17, 10. https://doi.org/10.14806/ej.17.1.200

Martins,R.F.,Fickel,J.,Le,M.,vanNguyen,T.,Nguyen,H.M.,Timmins,R.,…Wilting,A.(2017).Phylogeographyofredmuntjacsrevealsthreedis-tinctmitochondriallineages.BMC Evolutionary Biology,17,34.https://doi.org/10.1186/s12862-017-0888-0

Mattioli,S.(2011).Cervidae.InD.Wilson,&R.Mittermeier(Eds.),Handbook of the mammals of the World,Vol.2 (pp. 350–443)., HoofedmammalsBarcelona,Spain:LynxEdicions.

McKenna,A.,Hanna,M.,Banks,E.,Sivachenko,A.,Cibulskis,K.,Kernytsky,A.,…DePristo,M.A.(2010).ThegenomeanalysisToolkit:AMapReduceframeworkforanalyzingnext-generationDNAsequencingdata.Genome Research,20,1297–1303.https://doi.org/10.1101/gr.107524.110

Meijaard,E.,&Groves,C.P. (2004).MorphometricalrelationshipsbetweenSouth-east Asian deer (Cervidae, tribe Cervini): Evolutionary and bio-geographic implications.Journal of the Zoological Society of London,263,179–196.https://doi.org/10.1017/S0952836904005011

Mercer,J.M.,&Roth,V.L.(2003).TheeffectsofCenozoicglobalchangeonsquirrel phylogeny. Science,299, 1568–1572. https://doi.org/10.1126/science.1079705

Meredith,E.P.,Rodzen,J.A.,Levine,K.F.,&Banks,J.D.(2005).Characterizationof an additional 14microsatellite loci inCalifornia Elk (Cervus elaphus)foruseinforensicandpopulationapplications.Conservation Genetics,6,151–153.https://doi.org/10.1007/s10592-004-7735-8

Mishra, S., Gaillard, C., Hertler, C., Moigne, A. M., & Simanjuntak, T.(2010). India and Java: Contrasting records, intimate connections.Quaternary International,223–224,265–270.https://doi.org/10.1016/j.quaint.2009.11.040

Muir,P.,Semiadi,G.,Asher,G.W.,Broad,T.E.,Tate,M.L.,&Barry,T.N.(1997).SambarDeer (Cervus unicolor) ×RedDeer (C. elaphus) interspecieshy-brids.Journal of Heredity,88,366–372.https://doi.org/10.1093/oxford-journals.jhered.a023120

Patou,M.-L.,Wilting,A.,Gaubert,P.,Esselstyn,J.A.,Cruaud,C.,Jennings,A.P.,…Veron,G.(2010).EvolutionaryhistoryoftheParadoxuruspalmciv-ets—Anewmodel forAsianbiogeography. Journal of Biogeography,37,2077–2097.https://doi.org/10.1111/j.1365-2699.2010.02364.x

Pitra,C.,Fickel,J.,Meijaard,E.,&Groves,C.(2004).Evolutionandphylogenyof oldworld deer.Molecular Phylogenetics and Evolution,33, 880–895.https://doi.org/10.1016/j.ympev.2004.07.013

Pritchard,J.K.,Stephens,M.,&Donnelly,P.(2000).Inferenceofpopulationstructureusingmultilocusgenotypedata.Genetics,155,945–959.

Røed,K.H.,&Midthjell,L.(1998).Microsatellitesinreindeer,Rangifer taran-dus,andtheiruseinothercervids.Molecular Ecology,7,1771–1788.

Ronquist,F.,&Huelsenbeck,J.P.(2003).MrBayes3:Bayesianphylogeneticinferenceundermixedmodels.Bioinformatics,19, 1572–1574.https://doi.org/10.1093/bioinformatics/btg180

Sathiamurthy, E., & Voris, K. H. (2006). Maps of Holocene Sea LevelTransgression and Submerged Lakes on the Sunda Shelf. The Natural History Journal of Chulalongkorn University, Suppl.,2,1–43.

Schröder,O., Lieckfeldt, D., Lutz,W., Rudloff, C., Frölich, K., & Ludwig,A.(2016).LimitedhybridisationbetweendomesticandtheEuropeanmou-floninWesternGermany.European Journal of Wildlife Research,62,307–314.https://doi.org/10.1007/s10344-016-1003-3

1476  |     MARTINS eT Al.

Senn,H.V.,&Pemberton,J.M.(2009).Variableextentofhybridizationbe-tween invasive sika (Cervus nippon) and native red deer (C. elaphus) ina small geographical area.Molecular Ecology,18, 862–876. https://doi.org/10.1111/j.1365-294X.2008.04051.x

Sheldon, F. H., Lim, H. C., & Moyle, R. G. (2015). Return to the MalayArchipelago: The biogeography of Sundaic rainforest birds. Journal of Ornithology,156,90–113.

Silvestro, D., & Michalak, I. (2012). RaxmlGUI: A graphical front-end forRAxML. Organisms Diversity & Evolution, 12, 335–337. https://doi.org/10.1007/s13127-011-0056-0

Slate, J., Coltman, D.W., Goodman, S. J., MacLean, I., Pemberton, J. M.,& Williams, J. L. (1998). Bovine microsatellite loci are highly con-served in red deer (Cervus elaphus), sika deer (Cervus nippon) andSoay sheep (Ovis aries). Animal Genetics, 29, 307–315. https://doi.org/10.1046/j.1365-2052.1998.00347.x

Smith,S.L.,Carden,R.F.,Coad,B.,Birkitt,T.,&Pemberton,J.M.(2014).AsurveyofthehybridisationstatusofCervusdeerspeciesonthe islandofIreland.Conservation Genetics,15,823–835.https://doi.org/10.1007/s10592-014-0582-3

Sun,X.,Li,X.,Luo,Y.,&Chen,X.(2000).ThevegetationandclimateatthelastglaciationontheemergedcontinentalshelfoftheSouthChinaSea.Palaeogeography, Palaeoclimatology and Palaeoecology, 160, 301–316.https://doi.org/10.1016/S0031-0182(00)00078-X

Talbot,J.,Haigh,J.,&Plante,Y.(1996).AparentageevaluationtestinNorthAmericanElk(Wapiti)usingmicrosatellitesfromovineandbovineor-igin. Animal Genetics, 27, 117–119. https://doi.org/10.1111/j.1365- 2052.1996.tb00480.x

Timmins,R.J.,Kawanishi,K.,Giman,B.,Lynam,A.,Chan,B.,Steinmetz,R.,…SambaKumar,N.(2015)Rusa unicolor.IUCNRedListThreatenedSpecies,e.T41790A85628124.

vanBemmel,A.C.V.(1949).RevisionoftheRusineDeer.Treubia,20,191–262.VanDenBergh,G.D.,DeVos,J.,&Sondaar,P.Y.(2001).TheLateQuaternary

palaeogeography ofmammal evolution in the IndonesianArchipelago.Palaeogeography, Palaeoclimatology and Palaeoecology, 171, 385–408.https://doi.org/10.1016/S0031-0182(01)00255-3

Veron,G.,Willsch,M.,Dacosta,V.,Patou,M.-L.,Seymour,A.,Bonillo,C.,…Wilting,A.(2014).ThedistributionoftheMalaycivetViverra tangalunga (Carnivora:Viverridae)acrossSoutheastAsia:Naturalorhuman-mediateddispersal?Zoological Journal of the Linnean Society,170,917–932.https://doi.org/10.1111/zoj.12110

Voris, H. K. (2000). Maps of Pleistocene sea levels in Southeast Asia:Shorelines,riversystemsandtimedurations.Journal of Biogeography,27,1153–1167.https://doi.org/10.1046/j.1365-2699.2000.00489.x

Webley, L. S., Zenger, K. R., English,A.W., & Cooper, D.W. (2004). Lowlevelsofgeneticvariationwithin introducedJavanrusadeer (Cervus ti-morensis russa) in Australia. European Journal of Wildlife Research, 50, 137–140.

Zazo,C.(1999).Interglacialsealevels.Quaternary International,55,101–113.https://doi.org/10.1016/S1040-6182(98)00031-7

How to cite this article:MartinsRF,SchmidtA,LenzD,WiltingA,FickelJ.Human-mediatedintroductionofintrogresseddeeracrossWallace’sline:HistoricalbiogeographyofRusa unicolor and R. timorensis. Ecol Evol. 2018;8:1465–1479. https://doi.org/10.1002/ece3.3754

     |  1477MARTINS eT Al.

TABLE A1 Completedatasetdetails

Spec

ies

Sam

ple

IDO

rigin

MtD

NA

hap

loty

peA

cces

sion

Num

ber

nDN

A a

ssig

nmen

tM

useu

m C

olle

ctio

nM

useu

m ID

Cura

tor

Rusa

uni

colo

rRUN2

Sumatra

H_13

MF176993

BerlinNHM

7515

0F.Mayer

Rusa

uni

colo

rRUN3

SriLanka

H_14

MF1

7701

8RUN

BerlinNHM

75133

F.Mayer

Rusa

uni

colo

rRUN5

NorthBorneo

H_37

MF1

7701

9RUN

BerlinNHM

1125

9F.Mayer

Rusa

uni

colo

rRUN6

NorthBorneo

H_38

MF1

7702

0BerlinNHM

1126

1F.Mayer

Rusa

uni

colo

rRUN10

SriLanka

H_14

MF1

7699

4BerlinNHM

12368

F.Mayer

Rusa

uni

colo

rRUN12

Sarawak,Borneo

H_15

MF1

7699

5RUN

BerlinNHM

1470

2F.Mayer

Rusa

uni

colo

rRUN19

NorthBorneo

H_37

MF1

7702

1RUN

BerlinNHM

103292

F.Mayer

Rusa

uni

colo

rRUN20

Men

taw

aiH_16

MF1

7699

6RUN

Naturalis

139–50

P.Kamminga

Rusa

uni

colo

rRUN21

Borneo

H_37

MF1

7702

2StuttgartNHM

1690

0S.Merker

Rusa

uni

colo

rRUN26

Borneo

H_39

MF177023

StuttgartNHM

1590

0S.Merker

Rusa

uni

colo

rRUN28

Thai

land

H_17

MF1

7699

7BonnNHM

ZFMK_

MAM_2013.618

R.Hutterer

Rusa

uni

colo

rRUN32

Sumatra

H_18

MF1

7699

8RUN

ViennaNHM

7333

F.Zachos

Rusa

uni

colo

rRUN33

Indi

aH_19

MF1

7699

9ViennaNHM

4086

7F.Zachos

Rusa

uni

colo

rRUN34

Java

H_20

MF1

7700

0SenckenbergFrankfurt

15937

I.Ruf

Rusa

uni

colo

rRUN35

SriLanka

H_21

MF1

7700

1ParisNHM

1877-10

G.Veron

Rusa

uni

colo

rRUN37

Burum,Moluccas

H_22

MF2

7925

0MunichNHM

1906/3021

M.Hiermeyer

Rusa

uni

colo

rRUN38

Sumatra

H_23

MF1

7700

2MunichNHM

1905

/46

M.Hiermeyer

Rusa

uni

colo

rRUN39

Timor

H_3

MF177003

MunichNHM

1911

/214

7M.Hiermeyer

Rusa

uni

colo

rRUN41

SouthSumatra

H_24

MF2

7925

1MunichNHM

1908

/465

M.Hiermeyer

Rusa

uni

colo

rRUN42

Sumatra

H_25

MF1

7700

4MunichNHM

1908

/466

M.Hiermeyer

Rusa

uni

colo

rRUN43

Indo

chin

aH_26

MF1

7700

5RUN

MunichNHM

1962/235

M.Hiermeyer

Rusa

uni

colo

rRUN44

SouthThailand

H_27

MF1

7700

6CopenhagenNHM

M15

94D.K.Johansson

Rusa

uni

colo

rRUN45

EastSumatra

H_40

MF1

7702

4CopenhagenNHM

M16

50D.K.Johansson

Rusa

uni

colo

rRUN47

SouthThailand

H_41

MF1

7702

5CopenhagenNHM

M1593

D.K.Johansson

Rusa

uni

colo

rRUN51

SouthThailand

H_42

MF1

7702

6CopenhagenNHM

M1583

D.K.Johansson

Rusa

uni

colo

rRUN54

BhutanDooars,India

H_43

MF1

7702

7CopenhagenNHM

M534

D.K.Johansson

Rusa

uni

colo

rRUN55

BhutanDooars,India

H_28

MF1

7700

7CopenhagenNHM

M533

D.K.Johansson

Rusa

uni

colo

rRUN56

SriLanka

H_44

MF1

7702

8RUN

CopenhagenNHM

M1236

D.K.Johansson

Rusa

uni

colo

rRUN58

SouthMyanmar

H_29

MF1

7700

8RUN

Naturalis

RMNH.MAM.693.a

P.Kamminga

Rusa

uni

colo

rRUN61

CentralJava

H_3

MF1

7700

9RT

INaturalis

RMNH.MAM.33832

P.Kamminga

(Continued)

APP

END

IX

1478  |     MARTINS eT Al.

Spec

ies

Sam

ple

IDO

rigin

MtD

NA

hap

loty

peA

cces

sion

Num

ber

nDN

A a

ssig

nmen

tM

useu

m C

olle

ctio

nM

useu

m ID

Cura

tor

Rusa

uni

colo

rRUN62

Bengala,India

H_30

MF1

7701

0Naturalis

RMNH.MAM.51460

P.Kamminga

Rusa

uni

colo

rRUN65

SouthSumatra

H_45

MF1

7702

9Naturalis

RMNH.MAM.51453

P.Kamminga

Rusa

uni

colo

rRUN66

WestSumatra

H_46

MF177030

RUN

Naturalis

RMNH.

MAM.1033.b

P.Kamminga

Rusa

tim

oren

sisRT

I2LesserSundaIslands

H_31

MF1

7701

1BerlinNHM

92305

F.Mayer

Rusa

tim

oren

sisRTI3

Mol

ucca

sH_1

MF1

7698

1BerlinNHM

30840

F.Mayer

Rusa

tim

oren

sisRT

I6Timor

H_2

MF1

7698

2RT

IStuttgartNHM

1587

8S.Merker

Rusa

tim

oren

sisRT

I8Java

H_3

MF176983

StuttgartNHM

1589

2S.Merker

Rusa

tim

oren

sisRT

I9Java

H_4

MF1

7698

4StuttgartNHM

1589

7S.Merker

Rusa

tim

oren

sisRTI13

Java

H_1

MF1

7698

5StuttgartNHM

1589

1S.Merker

Rusa

tim

oren

sisRT

I17

SouthSulawesi

H_5

MF1

7698

6BonnNHM

58.1

09R.Hutterer

Rusa

tim

oren

sisRT

I18

SouthSulawesi

H_32

MF1

7701

2BonnNHM

58.7

5R.Hutterer

Rusa

tim

oren

sisRT

I19

SouthSulawesi

H_33

MF177013

BonnNHM

58.1

05R.Hutterer

Rusa

tim

oren

sisRT

I20

SouthSulawesi

H_6

MF1

7698

7BonnNHM

58.1

12R.Hutterer

Rusa

tim

oren

sisRT

I21

Timor

H_3

MF1

7701

4ViennaNHM

231

F.Zachos

Rusa

tim

oren

sisRT

I22

LesserSundaIslands

H_34

MF1

7701

5ViennaNHM

2049

F.Zachos

Rusa

tim

oren

sisRT

I27

Sulawesi

H_35

MF1

7701

6ViennaNHM

7334

F.Zachos

Rusa

tim

oren

sisRTI30

Java

H_7

MF1

7698

8DresdenNHM

B730

C.Stefen

Rusa

tim

oren

sisRTI31

Sulawesi

H_3

MF1

7698

9DresdenNHM

B2686

C.Stefen

Rusa

tim

oren

sisRTI32

Java

H_8

MF1

7699

0SenckenbergFrankfurt

1546

2I.Ruf

Rusa

tim

oren

sisRTI33

Mol

ucca

sH_9

MF1

7699

1SenckenbergFrankfurt

5562

I.Ruf

Rusa

tim

oren

sisRTI37

WestJava

H_10

MF2

7924

7Naturalis

RMNH.MAM.5679

P.Kamminga

Rusa

tim

oren

sisRTI38

WestJava

H_11

MF2

7924

8Naturalis

RMNH.MAM.5680

P.Kamminga

Rusa

tim

oren

sisRTI39

Timor

H_3

MF1

7699

2RT

INaturalis

RMNH.MAM.51419

P.Kamminga

Rusa

tim

oren

sisRTI43

Timor

H_3

MF1

7701

7RT

INaturalis

RMNH.MAM.51417

P.Kamminga

Rusa

tim

oren

sisRT

I44

Bali

H_12

MF2

7924

9RT

INaturalis

RMNH.MAM.33828

P.Kamminga

Rusa

tim

oren

sisRT

I45

Bali

H_36

MF2

7925

2RT

INaturalis

RMNH.MAM.33827

P.Kamminga

Samplesareindicatedaccordingtothespeciesassignmentfromthemuseumcollections.mtDNAhaplotypeandgenotypicclusterassignmentareprovidedaccordingtotheresultsobtained.

TABLE A1 (Continued)

     |  1479MARTINS eT Al.

Locus Allelic range Na HE HO FIS (W&C) Reference

Haut14 116–130 6 0.66 0.25 0.628 Kühn,Anastassiadis,andPirchner(1996)

VH110 101–153 12 0.89 0.44 0.516 Talbotetal.(1996)

NVHRT48 71–120 12 0.8 0.31 0.615 RøedandMidthjell(1998)

BM757 167–199 12 0.92 0.5 0.424 Slateetal.(1998)

CSSM39 158–200 12 0.87 0.37 0.575 Slateetal.(1998)

CSSM41 120–146 11 0.85 0.37 0.566 Slateetal.(1998)

FSHB 171–206 13 0.92 0.5 0.426 Slateetal.(1998)

Roe09 167–199 12 0.9 0.63 0.31 FickelandReinsch(2000)

CSSM14 131–161 10 0.82 0.13 0.851 Kühn,Schröder,Pirchner,andRottmann(2003)

T115 139–184 9 0.81 0.44 0.467 Meredith,Rodzen,Levine,andBanks(2005)

C143 154–174 4 0.71 0.19 0.741 Meredithetal.(2005)

C180 138–160 6 0.67 0.37 0.448 Meredithetal.(2005)

INRA6 103–141 10 0.78 0.37 0.529 SennandPemberton(2009)

Mu_4D 111–113 2 0.23 0 1.000 Schröderetal.(2016)

Mu_1_51 114–152 16 0.95 0.75 0.216 Schröderetal.(2016)

C183 119–133 5 0.65 0.25 0.62 Schröderetal.(2016)

Mu_1_25 98–110 6 0.78 0.37 0.529 Schröderetal.(2016)

Mu_1_550 139–173 14 0.94 0.37 0.608 Schröderetal.(2016)

Allelicrange,numberofalleles(Na),expectedandobservedheterozygosity(HE and HO),andinbreedingcoefficient(FIS)aregivenforeachloci,ascalculatedforthe16individualsgenotyped.

TABLE  A2 Microsatellite loci details and results