Repeated and Time-Correlated Morphological Convergence in Cave-Dwelling Harvestmen (Opiliones, Laniatores) from Montane Western North America


previous item | next item

Citation
Repeated and Time-Correlated Morphological Convergence in Cave-Dwelling Harvestmen (Opiliones, Laniatores) from Montane Western North America

Material Information

Title:
Repeated and Time-Correlated Morphological Convergence in Cave-Dwelling Harvestmen (Opiliones, Laniatores) from Montane Western North America
Series Title:
PLoS ONE
Creator:
Derkarabetian, Shahan
Steinmann,, David B.
Hedin, Marshal
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Cave Dwelling ( local )
Montane Western North Africa ( local )
Morphological Convergence ( local )
Genre:
serial ( sobekcm )

Notes

Abstract:
Many cave-dwelling animal species display similar morphologies (troglomorphism) that have evolved convergent within and among lineages under the similar selective pressures imposed by cave habitats. Here we study such ecomorphological evolution in cave-dwelling Sclerobuninae harvestmen (Opiliones) from the western United States, providing general insights into morphological homoplasy, rates of morphological change, and the temporal context of cave evolution.
Original Version:
PLoS ONE, Vol. 5, no. 5 (2010-05-07).

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
This item is licensed with the Creative Commons Attribution License. This license lets others distribute, remix, tweak, and build upon this work, even commercially, as long as they credit the author for the original creation.
Resource Identifier:
K26-05121 ( USFLDC: LOCAL DOI )
k26.5121 ( USFLDC: LOCAL Handle )

USFLDC Membership

Aggregations:
University of South Florida
Karst Information Portal

Postcard Information

Format:
serial

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

RepeatedandTime-CorrelatedMorphological ConvergenceinCave-DwellingHarvestmen(Opiliones, Laniatores)fromMontaneWesternNorthAmericaShahanDerkarabetian1,DavidB.Steinmann2,MarshalHedin1*1 DepartmentofBiology,SanDiegoStateUniversity,SanDiego,California,UnitedStatesofAmerica, 2 DepartmentofZoology,DenverMuseumofNatureandScience, Denver,Colorado,UnitedStatesofAmericaAbstractBackground:Manycave-dwellinganimalspeciesdisplaysimilarmorphologies(troglomorphism)thathaveevolved convergentwithinandamonglineagesunderthesimilarselectivepressuresimposedbycavehabitats.Herewestudysuch ecomorphologicalevolutionincave-dwellingSclerobuninaeharvestmen(Opiliones)fromthewesternUnitedStates, providinggeneralinsightsintomorphologicalhomoplasy,ratesofmorphologicalchange,andthetemporalcontextofcave evolution.Methodology/PrincipalFindings:WegatheredDNAsequencedatafromthreeindependentgeneregions,andcombined thesedatawithBayesianhypothesistesting,morphometricsanalysis,studyofpenismorphology,andrelaxedmolecular clockanalyses.Usingmultivariatemorphometricanalysis,wefindthatphylogeneticallyunrelatedtaxahaveconvergently evolvedtroglomorphism;alternativephylogenetichypothesesinvolvinglessmorphologicalconvergencearenotsupported byBayesianhypothesistesting.Inoneinstance,thismorphologyisfoundinspecimensfromahigh-elevationstonydebris habitat,suggestingthattroglomorphismcanevolveinnon-cavehabitats.Wediscoveredastrongpositiverelationship betweentroglomorphyindexandrelativedivergencetime,makingitpossibletopredicttaxonagefrommorphology.Most ofourtimeestimatesfortheoriginofhighly-troglomorphiccaveformspredatethePleistocene.Conclusions/Significance:WhileseveralregionsintheeasternandcentralUnitedStatesarewell-knownhotspotsforcave evolution,fewmodernphylogeneticstudieshaveaddressedtheevolutionofcave-obligatespeciesinthewesternUnited States.Ourintegrativestudiesrevealtherecurrentevolutionoftroglomorphisminaperhapsunexpectedgeographic region,atsurprisinglydeeptimedepths,andinsometimessurprisinghabitats.Becausesomenewlydiscovered troglomorphicpopulationsrepresentundescribedspecies,ourfindingsstresstheneedforfurtherbiologicalexploration, integrativesystematicresearch,andconservationeffortsinwesternUScavehabitats.Citation: DerkarabetianS,SteinmannDB,HedinM(2010)RepeatedandTime-CorrelatedMorphologicalConvergenceinCave-DwellingHarvestmen(Opiliones, Laniatores)fromMontaneWesternNorthAmerica.PLoSONE5(5):e10388.doi:10.1371/journal.pone.0010388 Editor: RobertC.Fleischer,SmithsonianInstitutionNationalZoologicalPark,UnitedStatesofAmerica Received December30,2009; Accepted March24,2010; Published May7,2010 Copyright: 2010Derkarabetianetal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermits unrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited. Funding: ThisresearchwassupportedbytheCaveConservancyFoundation(FellowshipinKarstStudies),theSocietyforIntegrativeandComparativeBiology (GrantsinAidofResearch),theAmericanArachnologicalSociety(VincentRothFundforSystematicResearch),acollaborativeUSNationalScienceF oundation grantawardedtoM.Hedin(AwardNo.0640173)andDr.JeffShultz(AwardNo.0640179),agrantfromtheLloydDavidandCarlyeCannonWattisFoundation, andtheZocchiFoundation.Thefundershadnoroleinstudydesign,datacollectionandanalysis,decisiontopublish,orpreparationofthemanuscrip t. CompetingInterests: Theauthorshavedeclaredthatnocompetinginterestsexist. *E-mail:mhedin@sciences.sdsu.eduIntroductionCavehabitatshavelonginterestedevolutionarybiologistsand ecologists–suchhabitatscombine geographicisolation ,promoting speciationandendemicity,with selectivesimilarities ,promoting convergenceinlifehistoryandmorphologicalform[1].This combinationofdivergenceandselectiveconstraintsleadsto evolutionarytrendsthatarepredictableandoftenreplicated independentlywithinandamonganimallineages,makingcave organismsexcellentcandidatesforcomparativeevolutionary analyses[2–6].Anexampleisthestudyofecomorphological evolution,whereecomorphsaredefinedasecologicalforms characterizedbysuitesofphenotypictraitsthathavesome conspicuouscorrelationwithecologicalattributes[7].Cave ecomorphsarecalled troglomorphs [8],atermcoinedtodescribea commonmorphologicalsyndromedisplayedbycaveanimals(e.g., pale,reducedeyes,elongateappendages,etc).Assummarizedby Culveretal.,[1],‘‘ troglomorphsareformswherethebodyisclearlymodified forcaveexistence….whicharequitedifferentfromallnormalcavernicolous animals.Suchformsare,onthewhole,easilyidentifiableascaveanimalsand havemanycommontraitsevenwhentheybelongtodifferenttaxonomicgroups ’’. IncontinentalNorthAmerica,areasofhighdiversityfor terrestrialtroglomorphsincludetheAppalachianandInterior Lowlandsregions,thecentralTexasuplands,andtheCalifornia ‘‘MotherLode’’region[9,fig.2;10].Theseregionsarehometo terrestrialcave-obligatefaunasthatarespecies-richandhighlyendemic,reflectingacombinationofhabitatavailability(i.e., manycaves),habitatpersistencecombinedwithperiodicisolation, andavailabilityofsourcefaunasforcavecolonizationand subsequentevolution.Incomparison,muchlesshasbeenwritten PLoSONE|www.plosone.org1May2010|Volume5|Issue5|e10388

PAGE 2

abouttheGreatBasinandRockiesasahotspotforcaveevolution. ExamplesofexceptionsincludepapersbyPeck[11,12],andShear andcolleagues[13].Ingeneral,westernstates(exceptfortheCA MotherLoderegion)mayhavefewerdocumentedcave-obligate speciesbecausetherearefewerregionalbiospeleologists,fewer sourcelineages,orcavehabitatslessfavorablefortroglomorphic evolution(e.g.,cavesthataresmaller,younger,drier,shallower, nutrient-poor,etc).Surfaceconditionsoverevolutionarytimemay alsoplayarole,withperiodicgeneflowfromsurfaceforms swampingincipientevolutionoftroglomorphs. Herewedescribethediscoveryofafascinatingpatternof recurrenttroglomorphisminharvestmenfromthesouthern RockiesandadjacentwesternstatesofNorthAmerica.Laniatoreanharvestmen,membersofthearachnidorderOpiliones,are conspicuouspredatorsincavesthroughouttheworld[14–17] includingNorthAmerica.Over50speciesrepresentingmorethan adozenLaniatoreangeneraarefoundinNorthAmericancaves, andmanyofthesetaxaareobviouslytroglomorphic,withpale integuments,longlegsandpalps,andreducedorabsenteyes [6,18–25].Thesetroglomorphictaxaalsooftenhaveverysmall geographicdistributions,andthreespeciesarelistedasfederallyendangered[26,27].ThisstudyconcernsSclerobunineharvestmen,cryophilicarachnidsfromwesternNorthAmericathatare phylogeneticallyplacedinalargerHolarcticclade,theTravunioidea[28,29].Threegeneraarecurrentlyclassifiedas Sclerobunines,including Cyptobunus , Zuma ,and Sclerobunus .Membersofallthreegeneraprefermoist,dark,surfacemicrohabitats (e.g.,underrocksandlogs),butaresometimesfoundincaves.For example,alldescribed Cyptobunus taxaareknownonlyfromcaves, andaredistinguishedfromothersclerobuninesinelaborationof troglomorphicfeatures(Table1,Fig.1).Cavernicolous Sclerobunus arealsoknownfrommuseumcollections,buthaveneverbeen formallystudied–whetherornotthesetroglomorphic Sclerobunus areindependentfromorphylogenetically-relatedto Cyptobunus is unknown. Inthispaperweexamineinternalrelationshipsofthe SclerobuninaeusingDNAsequencedatafromthreeindependent generegions.ThesegenetreedataarecombinedwithBayesian hypothesistesting,morphometricsanalysis,studyofpenis morphology,andrelaxedmolecularclockanalysestounderstand thebiogeographyandtimingoftroglomorphicevolutionwithin thisgroup.Centralquestionsincludetheprevalenceofhomoplasy introglomorphism,theageoftroglomorphictaxa,andthe geographiccontextofthisevolution.Wefindthattroglomorphic populations/specieshaveevolvedmultiple(3–5)timesindependentlywithinsclerobunines,arerelativelyold,andgenerallyhave smallgeographicdistributions.Someofthesetroglomorphic populationsalmostcertainlyrepresentundescribedspeciesdeservingconservationattention.Onetroglomorphicpopulationisfound inahigh-elevationnon-cavesituation,suggestingthattroglomorphssometimesevolveinhabitatsotherthancaves.Overall, thisstudyrevealsasurprisinglydynamicpictureofcaveevolution inmontanewesternNorthAmerica.MaterialsandMethods TaxonSamplingMostofthespecimensusedinthisstudywerecollectedinrecent fieldworkconductedbytheauthors-whencollectingaseriesof individuals,somewerepreservedin80%EtOHformorphological analysis,whereasspecimensdestinedformolecularanalysiswere preservedin100%EtOH(andlaterstoredat 2 80C).Other specimensusedinbothmolecularandmorphologicalanalysis wereborrowedfrommuseums(seeAcknowledgements). ThecladeTravunioideahasrecentlybeenrecoveredin phylogeneticanalysesofLaniatores.Themonophylyofthisgroup issupportedbybothmorphologicalfeatures[28]andmolecular data[29].TravunioidscomprisethreeHolarcticfamilies(Travuniidae,Cladonychiidae,andBriggsidae),andseveralnorthern hemispheresubfamiliesthatwererecentlytransferredfromthe Triaenonychoidea(Sclerobuninae,Paranonychinae,Kaolinonychinae,andNipponychinae).Withinthislargergroup,the northernhemispheretriaenonychidsmayformaclade,basedon themusculatureofthepenis,andthetridentshapedtarsalclaw withvariablenumbersofsidebranches[30].Asoutgroupsforthis studyweincludedNorthAmericantaxarepresentingthe Travuniidae( Speleonychia ),Cladonychiidae( Erebomaster,Cryptomaster, Theromaster , Speleomaster ),Briggsidae( Briggsus ),andtheParanonychinae( Metanonychus and Paranonychus ). Fumontanadeprehendor and Equitiusdoriae wereusedasdistantoutgroupstoTravunioidea, followingresultsofGiribetetal.,[29]. Alldescribedsclerobuninetaxawererepresentedinoursample (TableS1).Thisincludestwonarrowly-distributed Zuma species fromCalifornia,andalldescribedtaxa(speciesandsubspecies)of both Cyptobunus and Sclerobunus .Thespecies Sclerobunusrobustus includestworelativelywidespreadsubspecies( S.r.robustus and S.r. glorietus ),whichmaycontaincrypticlineages–wefirstconducteda comprehensivemitochondrialphylogeographicsurveyofthe entiregeographicdistributionofeachsubspecies[31],andinclude heregeographicrepresentativesfromallprimarymitochondrial lineages. Weemphasizedthesamplingofcavepopulations.Twospecies of Cyptobunus weresampledfromgeographically-isolatedcavesin UT,NVandMT.Asdiscussedabove, Cyptobunus taxaareknown onlyfromcaves(Table1),areobviouslycave-modified(Fig.1), andlack known neighboringsurfacepopulations(Fig.2).Wealso includedfournewly-discoveredcave-dwellingpopulationsof S. robustus fromColoradocaves(CaveoftheWinds,MalloryCave, FaultCave,SkeletonCave). Apriori ,wequalitativelyclassified thesepopulationsastroglomorphicbasedonoverallmorphologicalappearance.Thesecavepopulationsarelessgeographically isolatedfromnearbysurfacepopulations,andinsomeinstances areadjacenttoforestswithsurface-dwelling Sclerobunus .Finally,we sampledfromahigh-elevationmontanesiteinnorth-centralNew Mexico(nearTaos)wherewediscoveredqualitativelytroglomorphicanimalsfromdeepstonydebris(0.25–1meterbelowthe surface),syntopicwith‘‘typical’’specimenscollectedclosertothe surface(seeFig.1).Wearenotawareofcavehabitatsnearthis montanelocation.MorphometricsAnalysisLinearmeasurementsofthebody,eyemound,chelicerae,palps, andthefirsttwolegsweretakenusingadissectingmicroscope fittedwithanocularmicrometer(TableS2).Characterschosenfor analysishavebeenusedinotherstudiesoflaniatoreanmorphology [32].Allmeasurementsweretakenfromadultmales,exceptfor theinclusionoffemalespecimensfromTaos(bothsurfaceand troglomorphicforms),SkeletonandFaultCaves,andtheir respectivenearestsurfacelocations(HangingLakeandApex ValleyRoad).Wealsomeasuredspecimensofslightlytroglomorphic Sclerobunus fromTerreroCave(NM)andCementCreek Cave(CO),butwereunabletoacquiremoleculardataforthese cavepopulations. WeconductedaPrincipalComponentsAnalysis(usingSystat 11,SystatSoftware,Inc.)toassesswhethercave-derivedsamples (plusdeepstonydebrisTaosspecimens)consistentlyoccupya distinctregionofmultivariatemorphologicalspace.Basedon resultsofPCAanalyses,we quantitatively classifiedpopulations/taxaConvergenceinCaveOpiliones PLoSONE|www.plosone.org2May2010|Volume5|Issue5|e10388

PAGE 3

astroglomorphic.WethenusedBayesfactoranalysestotest severalalternativehypothesesregardingtheevolutionoftroglomorphisminthecontextofmolecularphylogeneticdata(see Table2).Weusedthe‘‘prsettopologypr’’settinginMrBayes,and setthepriorprobabilityoftheconstrainedgroupto100[33].The Bayesfactor(BF)wascalculatedbytakingtwicethedifferenceof the–lnLharmonicmeansbetweenunconstrainedandconstrained analyses[34],andinterpretedfollowingKassandRaftery[35].MolecularDataCollectionGenomicDNAwasextractedfromlegtissueusingtheQiagen DNeasyKit,permanufacturer’sprotocol.Threeseparategene regionswerePCR-amplifiedandsequenced–theseincluded portionsofthemitochondrialcytochromeoxidaseI(COI)gene, thenuclearribosomal28Sgene,andthenuclearelongation factorIalpha(EF-1 a )gene(seeTableS3forprimers).EF-1 a sequencesweregeneratedforareducedtaxonsamplethatdid notincludealltravunioidoutgroups.Amplificationsfollowed previously-publishedmethods[6,36,37],exceptthataMasterAmpkit(EpicentreBiotechnologies)wasusedforsome28S reactions,andsomeEF-1 a reactionswereconductedina nestedfashion,usingamplificationproductsfromaninitial reactioninasecondreactionwithhigherannealingtemperatures.PCRproductswerepurifiedusingPolyethyleneGlycol (PEG)precipitationorgelpurific ation(QIAquickGelExtraction Kit,QIAGEN),andsequencedusingBigDyeVersion3dye chemistry(ABI).AllDNAsequencingwasperformedattheSan DiegoStateUniversityMicrochemicalCoreFacility.PhylogeneticAnalysisContiguoussequenceswereassembledfrombi-directionalreads andmanuallyalignedusingSequencher4.5(GeneCodes Corporation,MI)andMacCladev4.06[38].TheCO1alignment included3-bpindelsin Fumontana and Equitius ;thesewere manuallyaligned,guidedbyaminoacidtranslations.Thelength variable28Sdatawasfirstalignedusingdefaultparametersin ClustalXv1.83[39],thenmanuallyadjusted.TheEF-1 a sequencesincludedbothexondata(alignedmanually),anda rapidlyevolvingintron;thisintronwasdifficulttoalignacrossthe divergenttaxonsample,andwasthereforeremovedfrom phylogeneticanalyses. BothunpartitionedandpartitionedBayesiananalysesof combineddatawereusedtoinferphylogeneticrelationships, conductedusingthesoftwareMrBayes3.0b4[40].Modelsusedin analyseswerechosenwithMrModeltest[41].Severaldifferent partitioningstrategiesweretestedusingBFanalysestodetermine anoptimalpartitioningstrategy(TableS4).Asaconvergence diagnostic,allanalyseswerepermittedtorununtilthestandard deviationofsplitfrequenciesfellbelow0.01[33].Afterdiscarding thefirst40%oftreetopologiesasburn-in,topologieswere summarizedasamajority-ruleconsensustree.Taxonbipartitions withposteriorprobabilitiesover95%areconsideredstrongly supported.Bayesfactoranalysis(asdescribedabove)wasusedto testthemonophylyoftheSclerobuninae(i.e., Zuma , Sclerobunus , and Cyptobunus wereconstrainedtoformamonophyleticgroup).DivergenceTimeEstimatesTheconcatenatedmatrixwasanalyzedusingBEASTsoftware [42,43],whichusesaMetropolis-HastingsMCMCalgorithmto simultaneouslyestimateevolutionaryparameters,phylogeny,and divergencedatesunderarelaxedclockmodel.Relaxedclock methodsusingBEASThavebeenusedtoestimatethetimingof troglomorphicevolutioninothercave-dwellingtaxa[5,44–46]. Datawerepartitionedbygene(CO1,28S,EF-1 a )andbycodon position(CO1andEF-1 a )usingthesamemodelsselectedforeach partitionasinMrBayesanalyses,andruntwice(toassess convergence)for30milliongenerationswith50%burnin.Use ofanuncorrelatedrelaxedclockbranchratemodelallowsfor variableratesofmolecularevolutionamongbranches.Bothfossil andbiogeographicevidencewasusedtoplaceclockcalibrations (seeResults).PenisMorphologyMalegenitaliaareoftenthemostconspicuouslydifferent charactersobservedwhencomparingclosely-relatedterrestrial arthropodtaxa[47].Thisisalsooftenthecaseforlaniatoreans, Table1. TaxonomicSummaryoftheSclerobuninae.GenusSpeciesGeographicdistributionDistinguishingFeatures Zuma GoodnightandGoodnight,1942 Z.acuta GoodnightandGoodnight,1942 coastalcentralCAIntegumentyellow-brown,withsome blackpigment Z.tioga Briggs,1971a YosemiteNP,CAIntegumentyellow,withoutblack pigment Cyptobunus Banks,1905 C.cavicolens Banks,1905 LewisandClarkCaverns,MTTroglomorphic,someblackpigment,with reducedlateralbranches C.u.ungulatus Briggs,1971a cavesinGreatBasinNP,NVHighlytroglomorphic,absentbranches C.u.madhousensis Briggs,1971a MadhouseandProfessorBussCaves,UTHighlytroglomorphic,absentbranches, largecornea,retinadisplaced Sclerobunus Banks,1898 S.nondimorphicus Briggs,1971a WA,OR,andBCPalpalfemurofmalesnotswollen S.r.idahoensis Briggs,1971a ID,southernBCScutewithmuchblackpigment S.r.glorietus Briggs,1971a northernNMScutewithblackpigment, , 2mm S.r.robustus Packard,1877 AZ,CO,NM,UTScutewithsomeblackpigment, . 2mm Notes: BC,BritishColumbia;CA,California;ID,Idaho;MT,Montana;NV,Nevada;OR,Oregon;UT,Utah;WA,Washington. doi:10.1371/journal.pone.0010388.t001 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org3May2010|Volume5|Issue5|e10388

PAGE 4

wherestudiesofmalepenismorphologyarestressedinthe taxonomyofclosely-relatedtaxa[48].Forarepresentativesample ofsclerobuninesandcloseoutgroups,malepeniseswereextracted bypushingthepenisthroughthegenitaloperculumwithaninsect pininsertedthroughtheopeningoftheanaloperculum.Thepenis wasthensoakedinacoldsolutionof10%KOH,andslidemountedusingasmallamountofKY-Jellyimmersedinfiltered 70%EtOH.PhotographsweretakenwithaVisionaryDigital imagingsystem(http://www.sci.sdsu.edu/eb/imaging.html).Individualimageswerecompiledintoacompositeimageusing HeliconFocus(http://www.heliconsoft.com/heliconfocus.html) andeditedusingAdobePhotoshopCS3. A B C D E F Figure1.Representativesclerobuninemorphologicaldiversity. A) S.r.robustus (ApexValleyRoad);B) S.r.robustus (FaultCave);C) S.r. glorietus (Taos);D) S.r.glorietus ‘‘trog’’(Taos);E) Sclerobunussp (CaveoftheWinds);F) C.u.ungulatus (ModelCave).Allscalebars=1mm. doi:10.1371/journal.pone.0010388.g001 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org4May2010|Volume5|Issue5|e10388

PAGE 5

Figure2.DistributionoftheSclerobuninae,emphasizingsamplinglocationsofSclerobunusrobustus. Colorscorrespondtoclades recoveredincombinedanalysis.MapnumbersasinTableS1. doi:10.1371/journal.pone.0010388.g002 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org5May2010|Volume5|Issue5|e10388

PAGE 6

Results DataAvailabilityVoucherspecimensformostsamplesusedinDNAanalysesare currentlyhousedintheSDSUTerrestrialArthropodscollection– thesespecimenswillultimatelybedepositedattheCalifornia AcademyofSciences.OtherDNAspecimensresideattheDenver MuseumofNatureandScience.AllDNAsequenceshavebeen depositedtoGenBank(accessionnumbersinTableS1).AGoogle Earthkmzfileisavailableuponrequestfromthecorresponding author.MorphometricsAnalysisTocontrolforbodysize,aPCAwasruninwhichtherawdata wasscaledbyscutelength,whichisanappropriatemeasureof overallsize.Thisanalysisresultedintwoprincipalcomponents accountingfor76%ofthetotalvariation(PC1–53.8,PC2–22.2). PC1accountsformuchoftheoveralllengthofthespecimen, specificallyleglength,andPC2accountsforpalpalandcheliceral size(basedoncomponentloadings,notshown).Thesize-scaled PCArevealssexualdimorphism,wheremalespecimensloadmore positivelyonPC2(e.g.,Taossurfacemaleversusfemale,Hanging LakeTrailmaleversusfemale,seeFig.3A).Cave-derivedsamples of Cyptobunus ,and Sclerobunus fromMalloryCaveandCaveofthe Winds,clustertogetherinmorphologicalspace(Fig.3A).Deep stonydebrisspecimensfromTaosclusterwiththesesamples, ratherthanwith‘‘surface’’samplesfromTaos.Less-modified specimensfromTerrero,CementCreek,Fault,andSkeletoncaves are‘‘intermediate’’betweencaveversussurfacemorphologies.For FaultandSkeletoncaves,thesedifferslightlyfromgeographicallyproximatesurfacesamplesthatarephylogenetically-related.CombinedAnalysisResultsThecombineddatamatrixincludedapproximately2900total characters,comingfromCO1(1140bp,someshorter),28S (1200bp,includinggaps),andtheEF-1 a exon(570bp).Allthree genescontributedparsimonyinformative(PI)characters( Sclerobunus + Cyptobunus only)asfollows:CO1(335PI),28S(49PI),and EF-1 a exon(54PI).Eachdifferentconcatenatedsequencewas derivedfromasinglespecimen,butnotallspecimensincluded dataforallgenepartitions(TableS1).EF-1 a sequencescouldnot becollectedforthree Sclerobunus samples(SkiApache,FaultCave, SkeletonCave),andEF-1 a outgroupsamplingincludedonly Paranonychus and Zuma . AnalysisofcombineddatarecoveredamonophyleticTravunioideawithrespecttodistantoutgroups(Fig.4).TheCladonychiidaearesplitintoeasternandwesternNorthAmerican clades,withtheeasterncladonychiids( Erebomaster and Theromaster ) recoveredassistertoacladecontainingthewestern Briggsus and Speleonychia .Northern‘triaenonychids’arerecoveredasmonophyletic(althoughwithweaksupport),withanearly-diverging Paranonychus ,then Zuma ,then Metanonychus sisterto Sclerobunus plus Cyptobunus .ThealternativehypothesisofamonophyleticSclerobuninae(( Zuma ( Sclerobunus , Cyptobunus ))isnotsupportedbyBF analysis(Table2). The Sclerobunus plus Cyptobunus cladeisstrongly-supported,with anearly-divergingPacificNorthwestcladecontaining S.nondimorphicus and S.robustusidahoensis .Describedspeciesof Cyptobunus form acladesisterto Sclerobunusrobustus (excluding S.r.idahoensis ).The S. robustus cladeincludesthreeprimarylineages,includingthe describedsubspecies S.r.glorietus and S.r.robustus ,andaCaveof theWindsplusMalloryCaveclade.Thiscombinedanalysistree impliesatleastthreeseparateoriginsoftroglomorphywithinthe sclerobunines,includingonetransitionintheancestorof Cyptobunus ,andtwoadditionalindependenttransitionsintroglomorphic Sclerobunus fromTaos(within S.r.glorietus ),andthe ancestorofCaveoftheWindsplusMalloryCave(Fig.4).Ifthe non-sisterFaultCaveandSkeletonCavesamples(within S.r. robustus )areconsideredtobetroglomorphic,thenfiveseparate originsareimplied.UsingBFanalyses,werejectthealternative singleorigin,twoorigins,andfouroriginshypotheses(Table2).DivergenceTimeEstimatesMultiplecalibrationpointswereusedinBEASTanalyses.First, therecoveredsisterrelationshipbetween S.nondimorphicus and S.r. idahoensis correspondstoawell-knownbiogeographicbreakinthe northernCascadesandnorthernRockyMountains[49,50].The estimateddivergencetimefordisjuncttaxaintheserespective regionscorrespondstotheupliftoftheCascadeMountains,2– 5Ma.Weusedanormally-distributedpriorwithameanof3.5 (95%confidenceintervalof2.03–4.97Ma).Second,aBaltic amberfossilclassifiedasacladonychiid[ Protoholoscotolemon ;51], whichdatesfrom38–54Ma,wasusedasaminimumestimatefor allTravunioidea(cladonychiidsspantherootnodeofTravunioideainBEASTtopologies,Fig.5).Hereweusedagammadistributedprior,withanalphaof2,betaof7,andoffsetfromzero by35(95%confidenceintervalof38–80Ma).Agamma distributionwasusedsuchthatthenodeatthebaseof Travunioideacouldbenoyoungerthan38Ma,butcouldbe older,reflectinguncertaintyinphylogeneticplacementanddateof thefossilcalibration.Finally,theCaveoftheWindsplusMallory Cavesplitwascalibratedtobenoolderthan5Ma,usinga lognormaldistribution(meanof1,standarddeviationof0.4). Althoughdatingcaveformationtimesisoftendifficult,itis generallyagreeduponthatmostextantlimestonecavesworldwide arelessthan10millionyearsold,withmostbeingnomorethan Table2. ResultsofBayesFactorhypothesistesting.Constraint-lnLHarmonicmean(postburnin)BayesFactorSupportforRejection Noconstraint 2 25191.16---Sclerobuninaemonophyletic 2 25202.2422.16VeryStrong Alltroglomorphsmonophyletic(singleorigin) 2 25446.55510.78VeryStrong Cyptobunus separatefromtroglomorphic Sclerobunus (twoorigins) 2 25286.94191.56VeryStrong Cyptobunus ,COW + Mallory,Taos,Fault + SkeletonCave(fourorigins) 2 25214.4646.6VeryStrong Notes: Singleoriginhypothesis-allcave-modifiedsamples(including Cyptobunus andtroglomorphic Sclerobunus )constrainedtoformaclade.Twooriginshypothesis -troglomorphic Sclerobunus constrainedtoformacladeindependentof Cyptobunus .Fouroriginshypothesis-troglomorphic Sclerobunus fromFaultandSkeleton Cavesconstrainedtoformaclade,independentofothertroglomorphic Sclerobunus and Cyptobunus . doi:10.1371/journal.pone.0010388.t002 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org6May2010|Volume5|Issue5|e10388

PAGE 7

severalmillionyearsold[52,53].Usingseveraldatingtechniques, Luiszer[54]concludedthatthemajorityoftheformationofCave oftheWindsoccurredmorerecentlythan5Ma.Wealso conductedBEASTanalyseswithoutthisthirdcalibration(Fig.S1). EffectiveSampleSize(ESS)valuesforallparameterswere consideredtobesufficient(over200)inallrunsforbothtwoandthree-calibrationBEASTanalyses. Thethree-calibrationsBEASTtreetopologywasalmost identicaltothecombinedBayesiantopology,exceptforminor differenceswithinsubclades(Fig.5).Afullsetofdateestimatesis foundintheTableS5–herewestressestimateddatesfor troglomorphicspeciesandpopulations.Thesetimeestimatesare generallyPlioceneorMioceneinage,includingthecommon ancestorof Cyptobunus (95%HPD3.7–10.7Ma),andthecommon ancestorofsyntopicformsatTaos(95%HPD2.9–8.2Ma).Time estimatesfordivergenceofFaultandSkeletoncavepopulations fromgeographically-proximatesurfacepopulationsarePleistocene inage(95%HPD0.4–1.6,0.7–1.2,respectively).Two-calibration BEASTanalysesresultinsimilar,butslightlyoldertimeestimates (Fig.S1).Intheseanalyses,thecommonancestoroftheCaveof theWindsplusMallorycladeisestimatedtohavedivergedinthe lateMiocene(95%HPD7.2–13.4Ma). Toassesstherelationshipbetweentroglomorphyandestimated divergencetimes,weusedPC1loadingsasaproxyfor troglomorphy.Becausethisproxyisalsorelatedtoincreasedleg length,thereexistsapossiblefunctionalrelationship,asthesecond ‘‘antenniform’’legofharvestmenisasensoryappendage[22,55]. ThecorrelationbetweenPC1loadingsandestimatedtimesince divergencefromthenearestnon-troglomorphicrelative(mean valuesfromtwo-calibrationBEASTanalyses)isnearlylinear (correlationcoefficientof0.952,Table3andFig.3B).PenisMorphologyTwenty-twomalespecimenswereexamined,including samplesfor Metanonychus , Paranonychus , Zuma ,andall Cyptobunus and Sclerobunus taxa.Consistentwithmolecularphylogenetic data,penismorphologysuggeststhat Metanonychus ismore similartothesclerobuninesthanis Paranonychus or Zuma (data notshown).All Sclerobunus and Cyptobunus shareatridentshaped stylus. Sclerobunusnondimorphicus and S.r.idahoensis arevery similarinthefinestructure,spination,andsizeofthepenis, whichisnoticeablydifferentfrom S.robustus . Cyptobunus taxa shareamorphologythatisquitedifferentfrom Sclerobunus (Fig.6E,F). Sclerobunusr.glorietus and S.r.robustus penisesare Figure3.Resultsofprincipalcomponentsanalysis,andmorphologyversustimerelationship. PanelA:Principalcomponentsscatterplot ofsize-correcteddata.Unlessotherwiseindicated(F),allspecimensaremale.PanelB:Relationshipbetween‘‘troglomorphyindex’’anddivergen ce timeestimatesfortroglomorphicpopulations/taxa. doi:10.1371/journal.pone.0010388.g003 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org7May2010|Volume5|Issue5|e10388

PAGE 8

nearlyidenticalinstructure,butdifferinoverallsize( S.r. robustus beingnearly50%largerthan S.r.glorietus ).TroglomorphicmalesfromCaveoftheWindsdifferfromother S.r. robustus intherecurvedspinesofthestylus(Fig.6I,J),whereas malesfromMalloryCavepossess areducedstylus,somewhat resemblingthestylusof Cyptobunus (Fig.6K,L).Thereareno obviousdifferencesinpenisstructurebetweenmoderately troglomorphic S.r.robustus (FaultCave)andgeographicallyproximatesurfacesamples(Ap exValleyRoad;da tanotshown). MalesfromthetroglomorphicTaospopulationremain unknown.DiscussionTheevolutionofsimilartroglomorphiccharacteristicsin distantly-relatedanimallineages(e.g.,fish,salamanders,spiders, beetles,harvestmen,etc)clearlyillustratesthepowerofevolutionaryconvergenceatdeepphylogeneticlevels.Overthepast10 years,studieswhichhaveviewedmorphologicalevolutionthrough thelensprovidedbymolecularphylogeneticshavelikewise revealedtroglomorphicconvergence within groups.Examples includeMexicancavefish[56],cavecatfishes[57],salamanders [3],shrimp[45],amphipods[5],divingbeetles[2],spiders[58] Figure4.Combined-data50%majorityruleconsensusBayesiantree. Majorcladesarenamed,thickbranchesaresupportedbyaposterior probabilityof95orhigher.Troglomorphicpopulations/taxaindicatedbyspecialicons. doi:10.1371/journal.pone.0010388.g004 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org8May2010|Volume5|Issue5|e10388

PAGE 9

Equitius doriae Fumontana OP281 Erebomaster OP522 Theromaster OP359 Briggsus OP1052 Speleonychia OP1683 Cryptomaster OP1641 Speleomaster OP1691 OP1671 OP1784 OP1053 OP750 OP748 OP2445 OP1966 OP974 OP985 S.r.idahoensis OP1649 S.r.idahoensis OP1635 S. nondimorphicus OP222 S. nondimorphicus OP1056 Cyptobunus cavicolens OP2143 Cyptobunus cavicolens OP2144 Taos OP882 “trog” Taos OP972 OP907 OP890 OP885 OP2115 OP893 OP956 OP1039 OP905 OP913 OP951 OP964 OP1130 OP1203 OP1149 OP1223 OP1140 OP1122 OP1164 Hanging Lake OP2679 Skeleton Cave OP2567 Apex Valley Rd OP2686 Fault Cave OP2569 OP1215 OP2109 OP1210 10 20 30 40 50 60 43.6 78.7 32.6 70.5 S. sp Cave of the Winds OP1127 S. sp Mallory Cave OP2568 C. u. madhousensis Prof Buss Cave OP239 C. u. ungulatus Ice Cave OP1229 C. u. ungulatus Model Cave OP1232 PNW Sclerobunus S. r. robustus Cyptobunus Paranonychus Zuma Metanonychus MiocenePlio. Oligocene Eocene S. r. glorietus Figure5.BEASTphylogram. Basedonthreecalibrationpoints(arrows),with95%confidenceintervalsforallestimatednodes.Confidence intervalsfortherootofthetreeandtheoutgroupcladearegivenbelowthenodes.Posteriorprobabilitieslessthan95areshownwithrednodebars. Troglomorphicpopulations/taxaindicatedbyspecialicons.TimeforgeologicepochsfromtheU.S.GeologicalSurveyGeologicNamesCommittee, 2007,Divisionsofgeologictime—Majorchronostratigraphicandgeochronologicunits:U.S.GeologicalSurveyFactSheet2007-3015,2p. doi:10.1371/journal.pone.0010388.g005 Table3. Valuesfor‘‘troglomorphyindex’’andcorrespondingdivergencetimes.TroglomorphicSampleAvgPC1Surfacesistertaxon Estimateddivergencetime (mean) SkeletonCave 2 3.402HangingLake0.74 FaultCave 2 1.254ApexValleyRoad1.0 Taos2.165Taos5.9 CaveoftheWinds/Mallory5.030 S.r.robustus 10.1 Cyptobunus 4.702 S.robustus 17 Notes: Timeestimatesfromcomparisonsbetweentroglomorphicsamplesandnon-troglomorphicsistertaxa.Divergencetimesfromthetwo-calibrationsBEA ST analysis. doi:10.1371/journal.pone.0010388.t003 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org9May2010|Volume5|Issue5|e10388

PAGE 10

andotherharvestmen[6].Itisbecomingclearthatconvergent evolutionoftroglomorphismwithinlineagesistheexpectation, ratherthantheexception.Consistentwiththisexpectation,we haveshownthattroglomorphyhasevolvedindependentlyatleast threetimeswithinthe Sclerobunus plus Cyptobunus complex,and perhapsasmanyasfivetimes.Thisinferenceisfoundedon multigenicphylogenetics,explicithypothesistesting,newly-developedrelaxedclockmethods,andquantitativeconsiderationof morphologicalvariation.Ourperspectiveisuniqueinthatwe uncoveredanapparentlytime-correlateddynamic,inanunexpectedgeographicregion,insometimessurprisingmicrohabitats. Thesampleavailableforthisstudyalmostcertainlyunderestimatestheactualnumberofindependenttransitionstotroglomorphy,forseveralreasons.First,therearemanyknown populationsofcave-inhabitingtroglomorphic Sclerobunus forwhich molecularphylogeneticdatahasyettobegathered(e.g.,Terrero Cave,CementCreekCave).Also,Briggs[20]describeda troglomorphicjuvenile‘triaenonychid’fromacaveinsouthern Idahothatwasplacednear Cyptobunus ,andWelbourn[59]reports populationsof Sclerobunus fromcavesinsouthernArizona,some wellbelowtheelevationatwhichsurface Sclerobunus canbefound. Finally,alikelymajorityofwesterncaveshaveyettobediscovered and/orexplored.Clearly,afullaccountoftroglomorphic evolutioninthisgrouprequiresmorefieldeffort. Wemayhavealsounderestimatedtransitionstotroglomorphybecause clades ofexclusivelytrogl omorphictaxa(e.g., Cyptobunus ,CaveoftheWindsplusMallory)couldrepresent instancesof multiple independentderivations .Thisisparticularly likelyfor Cyptobunus ,wherecurrentlydescribedspeciesand subspeciesarefoundingeograp hicallydisjunctcavesystems, separatedbyhundredsofkilom etersoflowelevationhabitat thatisobviouslyuninhabitableandlackssubterraneanconnectivity(seeFig.1). Cyptobunus taxaaremostlikelyindependently derivedfromsurfacepopulationsthathavesincegoneextinct, leavinganarrayofextant,troglomorphicpopulationsthatshare commonancestry,butnotasi ngletroglomorphiccommon ancestor.TimeandMorphologicalDivergenceAlogicalexpectationforcave-dwellingorganismsisthatthe degreeoftroglomorphismiscorrelatedwithtime,i.e.,more highly-modifiedtaxaareexpectedtoberelativelyold.While quantifyinga‘‘troglomorphyindex’’isreasonablystraightforward, quantifyingtimeavailableforevolutionismoreproblematic,for severalreasons.Intheabsenceofmolecularclockinformation,one mightrelyoncaveagestoplaceupperlimitsoncavespeciesages, butstudieshaveshownthattroglomorphicspeciesaresometimes olderthanthecavestheyinhabit[60,61;seebelow].With molecularclockdata,thereisinherenterrorintimeestimates themselves,buttherearealsopotentialproblemswithincomplete taxon/populationsampling,andteasingapartcladogenicfrom post-speciationtroglomorphicevolution[44,62].Forexample,a sparsephylogeneticsampleofcave-dwellers,withmissingsurface taxa(eitherbecauseofincompletesamplingorextinction),would beexpectedtooverestimatethetimeavailablefortroglomorphic evolution.Oneexampleofastudythatassessedtherelationship betweentime(basedongeologicalevidence)andtroglomorphyis thatofWesseletal.,[60];theseauthorsfoundnocorrelationin caveplanthoppers. Toinvestigatethetroglomorphy/timerelationshipinsclerobunines,weusedPC1loadingsasaproxyfortroglomorphism.We correlatedthisvariablewithestimatedmeantimesincedivergence Figure6.Ventralandlateralviewsofpenisforrepresentativesamples. A,B) S.nondimorphicus ,C,D) S.r.idahoensis ,E,F) C.u.ungulatus – ModelCave,G,H) S.r.glorietus –GlorietaCanyon,I,J) Sclerobunussp. –CaveoftheWinds,K,L) Sclerobunussp. –MalloryCave,M,N) S.r.robustus – HangingLake,Allscalebars=0.5mm. doi:10.1371/journal.pone.0010388.g006 ConvergenceinCaveOpiliones PLoSONE|www.plosone.org10May2010|Volume5|Issue5|e10388

PAGE 11

from non-troglomorphic sistertaxa/populations(valuesfromtwocalibrationBEASTanalyses),andfoundastrongpositive relationship(Table3andFig.3B).Thiscorrelationraisesthe interestingpossibilityofbeingabletopredicttaxonagefroma simpletroglomorphyindexinthisgroup.Wehavereasonable confidenceinthisrelationshipforthreereasons.First,our samplingofknown extant taxaandpopulationsisrelatively complete.Wenotethatoursamplingofsurfacetaxanear Cyptobunus appearssparse(Fig.2),butsurfacesclerobunineshave infactneverbeenfoundinthesegeographicregions.Second,the observedcorrelationbetweentroglomorphyandtimedoesnot dependupon absolute accuracyoftimeestimates,butratheronlyon therelativetemporalorderingofdivergenceestimates.Wehave nobasistoquestiontherelativetemporalorderingofthese divergenceestimates.Finally,acloserconsiderationofmorphologyshowsaperhapsorderedevolutionoftroglomorphic characters,notcapturedbythesimplemorphologicalproxy. ‘‘Intermediate’’forms(FaultandSkeletonCaves)showslight depigmentationandasmallincreaseinleglength,whereas Cyptobunus showfurtherappendageandspineattenuation.Eye reductionmaybethelastcharacteraffected,asthisisnotseenin Cyptobunus ,butisseeninapparentlymoreancienteyeless travunioids( Speleonychia and Speleomaster ;Fig.5).PleistoceneEffectsModelAfrequently-cited temporal modelfortheevolutionoftemperate cave-dwellersisthePleistoceneEffectsmodel[63,64].Underthis model,moistconditionsofPleistoceneglacialperiodsallow connectionstolower-elevationcaves,whiledrierconditionsof interglacialperiods‘‘strand’’cryophilicpopulationsunderground, withsubsequentevolutioninisolatedcavehabitats. In Sclerobunus ,itappearsthattheevolutionof‘‘weakly’’ troglomorphicforms(i.e.,withslightlyreducedpigmentation, slightincreaseinleglength,noreductionineyemound)is consistentwiththePleistoceneEffectsmodel.Three-calibration BEASTestimatesforthedivergenceofFaultandSkeletoncave populationsfromgeographically-proximatesurfacepopulations arePleistoceneinage(95%HPD0.4-1.6,0.7-1.2,respectively).In eachcase,thegeographicsettingisalsoconsistentwitharecent surface-to-caveclimaticshiftmodel.FaultCaveislocatedona south-facinghillsideat1850meters,surroundedbychaparral/ grasslandhabitats(closestknownsurfacepopulationis23km distant,butlikelycloser),whileSkeletonCaveislocatedonly200 metersupslopefromaforestedhabitat(closestknownsurface population3.5kmdistant).Theseinstancesofcaveandsurface populationsincloseproximityinvitepopulationgeneticsstudies, withlargergeneticsamplesizesanddensersurfacesamplingcloser tothecaves.Suchstudiesmightbeabletodisentanglearecurrent geneflowmodelfromsharedancestralgeneticpolymorphism,and allowmoreaccurateestimatesofdivergencetime[4,5]. Absolutetimeestimatesforthemorehighlytroglomorphic sclerobuninelineagespredatethePleistocene,ashasbeenfound forothertemperatecave-dwellingtaxa[2,45,46].Three-calibrationestimatessuggestPlioceneorMiocenedivergencesfor Cyptobunus andTaostroglomorphs(Fig.5),whereastwo-calibration BEASTanalysessuggestMiocenedivergencesforthecommon ancestoroftheCaveoftheWindsplusMalloryclade(Fig.S1). Othertroglomorphictravunioidsinoursample( Speleonychia and Speleomaster )alsoappearrelativelyancient.Weadmitthatour temporalanalysesarehinderedbyalackofadditionalcalibration points,butthisisadifficultyoftenfacedinanalysesof understudiedarthropodgroupswithasparsefossilrecord.We contendthatourtimeestimatesarereasonable,ifnot underestimates . Giribetetal.,[29,fig.10],usingarelaxedclockanalysisfor multiplegenesandmultiplefossilcalibrationpoints,estimateda divergencevalueforthecommonancestorof Erebomaster plus Theromaster atover100Ma.Ourestimatesforthissamenodeare aboutfourtimesshallower.Wealsonotethatestimatedabsolute ratesofsclerobunineCOIevolution,asimpliedbyrelaxedclock analyses,areconsistentwitharthropodratesreportedinthe literature(TableS6).Evolutionof‘‘Surface’’TroglomorphsAtTaos,troglomorphicspecimenswerefoundindeepstonydebris inanorth-facingforestatanelevationof2850meters,syntopicwith typicalformsfoundunderrocksandlogsrestingonthesurface(Fig.1). ThetroglomorphicformsinhabitamicrohabitatclassifiedasMSS [milieusouterrainsuperficial–superficialundergroundcompartment;65],whichinmontanesouthernEuropehasbeenshownto housemanypopulationsandspeciesoftroglomorphicinvertebrates [66,67].Suchhabitatsaregenerallypoorly-studiedinNorthAmerica. Interestingly,Peck[68]long-agonotedtroglomorphiccavebeetles fromdeeplitterinhigh-elevationforestsofcentralNM.Indeed,the taxonomiccompositionoftheselitterfaunasremindedhimsogreatly ofcavefaunasthathereferredtotheseasthe‘‘southwesternmontane forestcavecommunity’’.Non-cavetroglomorphshaveactuallybeen foundinmanyarthropodtaxa[23,25,69,70],buttheirrelationshipto eithersurfaceorcave-dwellingformshasrarelybeenstudied. ThesyntopicsurfaceandtroglomorphicTaosformsare recoveredasphylogeneticsistertaxa,butaregeneticallydivergent forallgenes,differby8.7%forCO1(uncorrecteddistance),and areestimatedtohavedivergedintheMio-Pliocene(95%HPD 2.9–8.2Ma).Becausethetroglomorphsareextremelyrare–we havevisitedthissiteonfourseparateoccasionsandhavecollected atotalof3specimens(noadultmales)–weareuncertainofboth thefullgeographicdistributionofthisform,andthegeographic contextofdivergencefromsurfaceancestors(e.g.,inallopatry versusparapatry).MoreworkinwesternMSShabitatsisneeded, astheevolutionoftroglomorphisminthesehabitatsgreatly impactsinterpretationoftheevolutionofcaveforms.Forexample, itcouldbearguedthattroglomorphsoriginateinMSShabitats, withsuchformssecondarilyinvadingcavehabitats.Anexample mightbethetroglomorphsfoundinCaveoftheWindsand Mallorycaves,asbothofthesecavesarefoundalongtheeastern edgeoftheColoradoFrontRange,inpotentialcontactwith montaneMSS.Underthismodel,cave-dwellingformssharea troglomorphiccommonancestor,whichevolvedtroglomorphism ina non-cave habitat.Thehighly-troglomorphic(andapparently ancient,Fig.5) Speleonychia ,foundinCascadianlavatubesdatedat lessthan1Ma[71],isanotherpossibleexampleofthisdynamic.SpeciesDiscoveryandConservationWehypothesizethatatleastthreenewspecieshavebeen discoveredviaourcombinationoffieldworkandsystematic research.Theseinclude S.r.idahoensis fromtheNorthernRockies ofIdaho,geographically-isolatedfromCascadian S.nondimorphicus withoutknownpopulationsininterveningxerichabitats.Interestingly,thesetaxaarequitegeneticallydivergent(7.63%average uncorrectedpairwisedivergenceforCO1),butsurprisinglysimilar inbothsomaticandpenismorphology(Figs.3and6).We hypothesizethatTaostroglomorphicspecimensrepresenta uniquespecies,followingtheabove-mentionedgeneticand morphologicaldivergencefromsyntopicforms.Thisspeciesis currentlyonlyknownfromasinglesite,butlikelyhasalarger geographicdistribution.Finally,troglomorphicpopulationsfrom CaveoftheWindsandMalloryCavelikelyrepresentone,or perhapstwo,undescribedspecies.Thishypothesisisbasedon observeddivergenceinsomaticmorphology,penismorphology,ConvergenceinCaveOpiliones PLoSONE|www.plosone.org11May2010|Volume5|Issue5|e10388

PAGE 12

andconsistentdivergenceformultiplegenes.Shearetal.,[13] recentlyreportedonsimilarspeciesdiscoveryincave-limited millipedesfromthewesternUS,andstressedtheneedforfurther caveexploration,integrativesystematicresearch,andconservation attention.AsnotedbyShearetal.,[13],climatechangemay ultimatelydramaticallyimpacttheseeasily-perturbedecosystems, andtheirmanyendemic,habitat-specializedinhabitants.SupportingInformationFigureS1BEASTphylogram.Basedontwocalibrationpoints (arrows),with95%confidenceintervalsforallestimatednodes. Confidenceintervalsfortherootofthetreeandtheoutgroupclade aregivenbelowthenodes.Posteriorprobabilitiesoflessthan95are shownwithrednodebars.Troglomorphicpopulations/taxa indicatedbyspecialicons.TimeforgeologicepochsfromtheU.S. GeologicalSurveyGeologicNamesCommittee,2007,Divisionsof geologictime-Majorchronostratigraphicandgeochronologicunits: U.S.GeologicalSurveyFactSheet2007-3015,2p. Foundat:doi:10.1371/journal.pone.0010388.s001(0.62MBEPS)TableS1SampleInformationandGenBanknumbers. Foundat:doi:10.1371/journal.pone.0010388.s002(0.16MB DOC)TableS2MorphometricMeasurements.Bodyandscutelength measurementsweretakendorsallyatthemidline,whilelengthwas takenatthewidestpoint.Eyemoundheightwastakenfromthe tallestpointofthemoundfromthebottomoftheeyeandwidth wastakenfromtheoutsideedgeofbotheyesatthewidestpoint. Cheliceraelengthandwidthweretakenatthelongestpointsina dorsalviewforbothsegments.Palpallengthanddepth measurementsweretakenforeachsegmentinlateralviewexcept forthetarsalsegment,whichwastakenindorsalview.Leglength measurementsweretakenretrolaterallyforthetrochanter,femur, patella,tibia,andmetatarsusandweremeasuredbetweenthe mostdistantpointsoneithersideofthesegment. Foundat:doi:10.1371/journal.pone.0010388.s003(0.30MB DOC)TableS3PCRPrimers. Foundat:doi:10.1371/journal.pone.0010388.s004(0.04MB DOC)TableS4Modelselectionandtestingresults. Foundat:doi:10.1371/journal.pone.0010388.s005(0.03MB DOC)TableS5EstimateddivergencetimesfromBEASTanalyses (meanandconfidenceintervals). Foundat:doi:10.1371/journal.pone.0010388.s006(0.04MB DOC)TableS6EstimatedratesofCOIevolutionasimpliedby2calibrationand3-calibrationBEASTanalyses. Foundat:doi:10.1371/journal.pone.0010388.s007(0.04MB DOC)AcknowledgmentsForloansofmaterialfromColoradocaveswethanktheDenverMuseum ofNatureandScience.AdditionalloanswereprovidedbytheCalifornia AcademyofSciencesandAmericanMuseumofNaturalHistory.Samples fromGreatBasinNationalPark(GBNP)werecollectedunderapermit grantedtothefirstauthor.GretchenBakerandBenRobertshelpedwith permitsandcollectingintheGBNP.JeremyStilesgrantedaccesstoCave oftheWindsandJamesReiterhelpedcollectspecimens.LynetteKemp grantedaccessandRheaArmstronghelpedwithcollectingatLewisand ClarkCaverns.Allotherwildcavesweresampledwithpermissionor permitsfromtherespectiveorganizationsgrantedtoDavidSteinmann. SkeletonCavewassampledwithaSpecialUsePermitfromtheWhite RiverNationalForest;MalloryCavespecimenswerecollectedundera researchpermitfromtheCityofBoulderOpenSpaceandMountain Parks;FaultCavespecimenswerecollectedwithpermissionfromJefferson CountyOpenSpace.MeganPortercollected Cyptobunus specimensfrom nearProvo,UT.Foradditionalassistanceinfieldworkwewouldliketo thankJoeDeas,DamianElias,RyanFawcett,LarsHedin,Maureen McCormack,AdrienneRichart,NickRichart,CaseyRichart,Steven Thomas,andJeffreyUnderwood.CommentsbyDavidCulver,Francis Howarth,DeanLeavitt,JimStarrett,andStevenThomashelpedto improvethemanuscript.AuthorContributionsConceivedanddesignedtheexperiments:SDDBSMH.Performedthe experiments:SD.Analyzedthedata:SDMH.Contributedreagents/ materials/analysistools:SDDBSMH.Wrotethepaper:SDMH.References1.CulverDC,KaneTC,FongDW(1995)AdaptationandNaturalSelectionin Caves.TheEvolutionofGammarusminus.Cambridge:HarvardUniversity Press. 2.LeysR,WattsCHS,CooperSJB,HumphreysWF(2003)Evolutionof subterraneandivingbeetles(Coleoptera:Dytiscidae:Hydroporini,Bidessini)in thearidzoneofAustralia.Evolution57:2819–2834. 3.WiensJJ,ChippindalePT,HillisDM(2003)Whenarephylogeneticanalyses misledbyconvergence?AcasestudyinTexascavesalamanders.Systematic Biology52:501–514. 4.NiemillerML,FitzpatrickBM,MillerBT(2008)Recentdivergencewithgene flowinTennesseecavesalamanders(Plethodontidae: Gyrinophilus )inferredfrom genegenealogies.MolecularEcology17:2258–2275. 5.VillacortaC,JaumeD,OromiP,JuanC(2008)Underthevolcano: phylogeographyandevolutionofthecave-dwelling Palmorchestiahypogaea (Amphipoda,Crustacea)atLaPalma(CanaryIslands).BMCBiology2008,6: 7.doi:10.1186. 6.HedinM,ThomasSM(2010)MolecularsystematicsofeasternNorthAmerican Phalangodidae(Arachnida:Opiliones:Laniatores),demonstratingconvergent morphologicalevolutionincaves.MolecularPhylogeneticsandEvolution54: 107–121. 7.LososJB,MilesDB(1994)Adaptation,constraint,andthecomparativemethod: phylogeneticissuesandmethods.In:WainwrightPC,ReillySM,eds.Ecological Morphology:IntegrativeOrganismalBiology.Chicago:UniversityofChicago Press.pp60–98. 8.ChristiansenKA(1962)Propositionpourlaclassificationdesanimaux cavernicloes.Spelunca2:76–78. 9.CulverDC,MasterLL,ChristmanMC,HobbsHH,III(2000)Obligatecave faunaofthe48contiguousUnitedStates.ConservationBiology14:386–401. 10.CulverDC,ChristmanMC,ElliottWR,HobbsHH,III,ReddellJR(2003)The NorthAmericanobligatecavefauna:regionalpatterns.Biodiversityand Conservation12:441–468. 11.PeckSB(1980)Climaticchangeandtheevolutionofcaveinvertebratesinthe GrandCanyon,Arizona.NationalSpeleologicalSocietyBulletin42:53–60. 12.PeckSB(1981)TheinvertebratefaunaofthecavesoftheUintaMountains, northeasternUtah.GreatBasinNaturalist41:207–212. 13.ShearWA,TaylorSJ,JudsonWynneJ,KrejcaJK(2009)Cavemillipedsofthe UnitedStates.VIII.Newgeneraandspeciesofpolydesmidanmillipedsfrom cavesinthesouthwesternUnitedStates(Diplopoda,Polydesmida,Macrosternodesmidae).Zootaxa2151:47–65. 14.ForsterRR(1965)HarvestmenofthesuborderLaniatoresfromNewZealand caves.RecordsOtagoMuseum2:1–18. 15.MauryEA(1988)TriaenonychidaesudamericanosV.UnNuevogenerode opilionescavernicoladellaPatagonia(Opiliones,Laniatores).Memde Biospelogie15:117–131. 16.Pe ´rezGonza ´lezA,KuryAB(2002)Anewremarkabletroglomorphic GonyleptidfromBrazilArachnida,Opiliones,Laniatores).RevistaIbe ´ricade Arachnologia5:43–50. 17.KaramanIM(2005) Trojanellaserbicagen.n.,sp.n. ,aremarkablenewtroglobitic travunioid(Opiliones,Laniatores,Travunioidea).RevueSuissedeZoologie112: 439–455. 18.BriggsTS(1971)TheharvestmenofthefamilyTriaenonychidaeinNorth America.OccasionalPapersoftheCaliforniaAcademyofSciences90:1–43.ConvergenceinCaveOpiliones PLoSONE|www.plosone.org12May2010|Volume5|Issue5|e10388

PAGE 13

19.BriggsTS(1971)RelictharvestmenfromthePacificNorthwest.Pan-Pacific Entomologist47:165–178. 20.BriggsTS(1974)TroglobiticharvestmenrecentlydiscoveredinNorthAmerican lavatubes(Travuniidae,Erebomastridae,Triaenonychidae:Opiliones).Journal ofArachnology1:205–214. 21.CokendolpherJC(2004)Revalidationoftheharvestmangenus Chinquipellobunus (Opiliones:Stygnopsidae).TexasMemorialMuseum,SpeleologicalMonographs 6:143–152. 22.UbickD,BriggsTS(1992)TheharvestmenfamilyPhalangodidae.3.Revisionof Texella (Opiliones:Laniatores).TexasMemorialMuseum,Speleological Monographs3:155–240. 23.UbickD,BriggsTS(2002)TheharvestmenfamilyPhalangodidae4.Areviewof thegenus Banksula (Opiliones,Laniatores).JournalofArachnology30:435–451. 24.UbickD,BriggsTS(2004)TheharvestmanfamilyPhalangodidae.5.New recordsandspeciesof Texella GoodnightandGoodnight(Opiliones:Laniatores). TexasMemorialMuseum,SpeleologicalMonographs6:101–141. 25.UbickD,BriggsTS(2008)TheharvestmanfamilyPhalangodidae6.Revisionof the Sitalcina complex(Opiliones,Laniatores).ProceedingsCaliforniaAcademy Sciences59:1–108. 26.ChambersSM(1993)DepartmentoftheInterior,FishandWildlifeService,50 CFRPart17,RIN1018-AC06.Endangeredandthreatenedwildlifeandplants: CoffinCavemoldbeetle( Batrisodestexanus )andtheBoneCaveharvestmen ( Texellareyesi )determinedtobeendangered.FederalRegister56:43818–43819. 27.LongacreC(2000)DepartmentoftheInterior,FishandWildlifeService,50 CFRPart17,RIN1018-AF33.FinalruletolistnineBexarCounty,Texas invertebratespeciesasendangered.FederalRegister65:81419–81433. 28.GiribetG,KuryAB(2007)Chapter3.PhylogenyandBiogeography.In:Pintoda-RochaR,MachadoG,GiribetG,eds.Harvestmen:TheBiologyof Opiliones.Cambridge,MA:HarvardUniversityPress.pp62–87. 29.GiribetG,VogtL,Pe ´rezGonza ´lezA,SharmaP,KuryAB(2009)Amultilocus approachtoharvestman(Arachnida:Opiliones)phylogenywithemphasison biogeographyandsystematicsofLaniatores.Cladistics25:1–30. 30.MartensJ(1986)DiegrossgliederungderOpilionesunddieevolutionder ordung(Arachnida).XInternationalCongressofArachnology1:289–310. 31.DerkarabetianS(2010)PhylogeneticsoftheSclerobuninae(Arachnida: Opiliones:Laniatores):Systematics,TroglomorphicEvolution,andPhylogeography.MSThesis,SanDiegoStateUniversity. 32.ThomasSM,HedinM(2006)Naturalhistoryanddistributionoftheenigmatic southernAppalachianopilionid, Fumontanadeprehendor Shear(Laniatores: Triaenonychidae),withanassessmentofmorphologicalvariation.Zootaxa 1242:21–36. 33.RonquistF,HuelsenbeckJP,vanderMarkPMrBayes3.1Manual,Draft5/26/ 2005,onlineathttp://mrbayes.csit.fsu.edu/manual.php. 34.NylanderJAA,RonquistF,HuelsenbeckJP,Nieves-AldreJL(2004)Bayesian phylogeneticanalysisofcombineddata.SystematicBiology53:47–67. 35.KassRE,RafteryAE(1995)BayesFactors.JournaloftheAmericanStatistical Association90:773–795. 36.ThomasSM,HedinM(2008)Multigenicphylogeographicdivergenceinthe paleoendemicsouthernAppalachianopilionids Fumontanadeprehendor Shear (Opiliones,Laniatores,Triaenonychidae).MolecularPhylogeneticsandEvolution46:645–658. 37.HedinM,DerkarabetianS,McCormackM,RichartC,ShultzJW(2010)The phylogeneticutilityofthenuclearprotein-codinggeneEF-1aforresolvingrecent divergencesinOpiliones,emphasizingintronevolution.JournalofArachnology 38:9–20. 38.MaddisonDR,MaddisonWP(2003)MacClade4,ReleaseVersion4.07. Sunderland,MA:SinauerAssociates,Inc. 39.ThompsonJD,GibsonTJ,PlewniakF,JeanmouginF,HigginsDG(1997)The ClustalXwindowsinterface:flexiblestrategiesformultiplesequencealignment aidedbyqualityanalysistools.NucleicAcidsResearch24:4876–4882. 40.HuelsenbeckJP,RonquistF(2001)MRBAYES:Bayesianinferenceof phylogeny,Version3.0b4.Bioinformatics17:754–755. 41.NylanderJA(2003)MrModeltest1.1b.Dept.SystematicZoology,EBC, UppsalaUniversity,Sweden. 42.DrummondAJ,HoS,PhillipsM,RambautA(2006)Relaxedphylogeneticsand datingwithconfidence.PLoSBiology4:e88.doi:10.1371/journal. pbio.0040088. 43.DrummondAJ,RambautA(2007)BEAST:Bayesianevolutionaryanalysisby samplingtrees.BMCEvolutionaryBiology7:214–222. 44.PorterML,DittmarK,Pe ´rez-LosadaM(2007)Howlongdoesevolutionofthe troglomorphicformtake?Estimatingdivergencetimesin Astyanaxmexicanus .Acta Carsologica36:173–182. 45.PageTJ,HumphreysWF,Hughe sJM(2008)Shrimpsdownunder: EvolutionaryrelationshipsofsubterraneancrustaceanfromWesternAustralia (Decapoda:Atyidae:Stygiocaris).PLoSONE3:e1618. 46.FailleA,RiberaI,DeharvengL,BourdeauC,GarneryL,etal.(2010)A molecularphylogenyshowsthesingleoriginofthePyreneansubterranean Trechinigroundbeetles(Coleoptera:Carabidae).MolecularPhylogeneticsand Evolution54:97–106. 47.EberhardWG(1985)Sexualselectionandanimalgenitalia.Cambridge: HarvardUniversityPress. 48.Pinto-da-RochaR,GiribetG(2007)Chapter4,Taxonomy.In:Pinto-daRochaR,MachadoG,GiribetG,eds.Harvestmen:TheBiologyofOpiliones. Cambridge,Massachusetts:HarvardUniversityPress.pp88–246.49.BrunsfeldSJ,SullivanJ,SoltisDE,SoltisPS(2001)ComparativephylogeographyofnorthwesternNorthAmeric a:aSynthesis.In:SilvertownJ, AntonovicsJ,eds.IntegratingEcologyandEvolutioninaSpatialContext. Williston,Vermont:BlackwellPublishing.pp319–339. 50.CarstensBC,BrunsfeldSJ,DemboskiJR,GoodJM,SullivanJ(2005) InvestigatingtheevolutionaryhistoryofthePacificNorthwestmesicforest ecosystem:Hypothesistestingwithinacomparativephylogeographicframework. Evolution59:1639–1652. 51.UbickD,DunlopJ(2005)OntheplacementoftheBalticamberharvestman Gonyleptesnemastomoides KochandBerendt,1854,withnotesonthephylogenyof Cladonychiidae(Opiliones,Laniatores,Travunioidea).Mitteilungenausdem Museumfu ¨rNaturkundeinBerlin–GeowissenschaftlicheReihe8:75–82. 52.WhiteWB(1988)Geomorphologyandhydrologyofkarstterrains.Oxford UniversityPress. 53.MooreGW,SullivanN(1997)Speleology–Cavesandthecaveenvironment: Thirdedition.Ohio:CaveBooks. 54.LuiszerFG(2007)TimingofpassagedevelopmentandsedimentationatCaveof theWinds,ManitouSprings,Colorado,USA.ActaCarsologica36:157–172. 55.ShultzJW,Pinto-da-RochaP(2007)Chapter2.MorphologyandFunctional Anatomy.In:Pinto-da-RochaR,MachadoG,GiribetG,eds.Harvestmen:The BiologyofOpiliones.Cambridge,Massachusetts:HarvardUniversityPress.pp 14–61. 56.DowlingTE,MartasianDP,JeffreyWR(2002)Evidenceformultiplegenetic formswithsimilareyelessphenotypesintheblindcavefish, Astyanaxmexicanus . MolecularBiologyandEvolution19:446–455. 57.WilcoxTP,Garc ´ adeLeo ´nFJ,HendricksonDA,HillisDM(2004) Convergenceamongcavecatfishes:long-branchattractionandaBayesian relativeratestest.MolecularPhylogeneticsandEvolution31:1101–1113. 58.ArnedoMA,Mu ´rriaC,Mac ´ as-Herna ´ndezN,Orom ´ P,RiberaC(2007)The darksideofanislandradiation:Systematicsandevolutionarypatternsof troglobiticspidersofthegenusDysdera(Araneae,Dysderidae)intheCanary Islands.InvertebrateSystematics21:623–660. 59.WelbournWC(1999)InvertebratecavefaunaofKartchnerCaverns,Kartchner Caverns,Arizona.JournalofCaveandKarstStudies61:93–101. 60.WesselA,ErbeP,HochH(2007)Patternandprocess:Evolutionof troglomorphyinthecave-planthopperofAustraliaandHawaii–preliminary observations(Insecta:Hemiptera:Fulgoromorpha:Cixiidae).ActaCarsologica 36:199–206. 61.HochH,HowarthFG(1989)Sixnewcavernicolouscixiidplanthoppersinthe genus Solonaima fromAustralia(Homoptera:Fulgoroidea).SystematicEntomology14:377–402. 62.PorterML(2007)Subterraneanbiogeography:whathavewelearnedfrom moleculartechniques?JournalofCaveandKarstStudies69:179–186. 63.BarrTC,Jr.(1968)Caveecologyandtheevolutionoftroglobites.In: DobzhanskyT,HechtMK,SteereWC,eds.EvolutionaryBiology.NewYork: Appleton-Century-Crofts.pp35–102. 64.BarrTC,Jr.,HolsingerJR(1985)Speciationincavefaunas.AnnualReviewof EcologyandSystematics16:313–337. 65.JuberthieC,DelayB(1981)Ecologicalandbiologicalimplicationsofa ‘‘superficialundergroundcompartment’’.ProceedingsoftheInternational CongressofSpeleology8:203–206. 66.RuzickaJ(1998)Caveandrockdebrisdwellingspeciesofthe Cholevaagilis speciesgroupfromcentralEurope(Coleoptera,Leiodidae:Cholevinae). RegionalMuseumofNaturalScience,Torino.pp261–286. 67.RuzickaJ(1999)Thefirststepsinsubterraneanevolutionofspiders(Araneae)in centralEurope.JournalofNaturalHistory33:255–265. 68.PeckSB(1978)Newmontane Ptomaphagus beetlesfromNewMexicoand zoogeographyofsouthwesterncaves(Coleoptera,Leiodidae,Catopinae). SouthwesternNaturalist23:227–238. 69.HowarthFG(1983)Ecologyofcavearthropods.AnnualReviewofEntomology 28:365–389. 70.OromiP,MedinaAL,TejedorML(1986)Ontheexistenceofasuperficial undergroundcompartmentintheCanaryIslands.CongresoInternacionalde Espeleologia2:147–151. 71.GreeleyR,HydeJH(1972)Lavatubesofthecavebasalt,MountSt.Helens, Washington.GeologicalSocietyofAmericaBulletin83:2397–2418.ConvergenceinCaveOpiliones PLoSONE|www.plosone.org13May2010|Volume5|Issue5|e10388


printinsert_linkshareget_appmore_horiz

Download Options

close


  • info Info

    There are only PDFs associated with this resource.

  • link PDF(s)



Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.