Homo naledi, a new species of the genus Homo from the Dinaledi Chamber, South Africa


previous item | next item

Citation
Homo naledi, a new species of the genus Homo from the Dinaledi Chamber, South Africa

Material Information

Title:
Homo naledi, a new species of the genus Homo from the Dinaledi Chamber, South Africa
Creator:
Berger, Lee R.
Hawk, John
de Ruiter, Darryl J.
Publisher:
eLIFE
Publication Date:
Language:
English
Physical Description:
1 online resource

Subjects

Subjects / Keywords:
Fossil huminids ( lcsh )
Genre:
serial ( sobekcm )
Location:
South Africa

Notes

Abstract:
Homo naledi is a previously-unknown species of extinct hominin discovered within the Dinaledi Chamber of the Rising Star cave system, Cradle of Humankind, South Africa. This species is characterized by body mass and stature similar to small-bodied human populations but a small endocranial volume similar to australopiths. Cranial morphology of H. naledi is unique, but most similar to early Homo species including Homo erectus, Homo habilis or Homo rudolfensis. While primitive, the dentition is generally small and simple in occlusal morphology. H. naledi has humanlike manipulatory adaptations of the hand and wrist. It also exhibits a humanlike foot and lower limb. These humanlike aspects are contrasted in the postcrania with a more primitive or australopith-like trunk, shoulder, pelvis and proximal femur. Representing at least 15 individuals with most skeletal elements repeated multiple times, this is the largest assemblage of a single species of hominins yet discovered in Africa.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-05064 ( USFLDC DOI )
K26.5064 ( USFLDC 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

elifesciences.org RESEARCHARTICLE Homonaledi ,anewspeciesofthegenus Homo fromtheDinalediChamber, SouthAfricaLeeRBerger1,2*,JohnHawks1,3,DarrylJdeRuiter1,4,StevenEChurchill1,5, PeterSchmid1,6,LucasKDelezene1,7,TracyLKivell1,8,9,HeatherMGarvin1,10, ScottAWilliams1,11,12,JeremyMDeSilva1,13,MatthewMSkinner1,8,9, CharlesMMusiba1,14,NoelCameron1,15,TrentonWHolliday1,16, WilliamHarcourt-Smith1,17,18,RebeccaRAckermann19,MarkusBastir1,20, BarryBogin1,15,DebraBolter1,21,JulietBrophy1,22,ZacharyDCofran1,23, KimberlyACongdon1,24,AndrewSDeane1,25,ManaDembo1,26, MichelleDrapeau27,MarinaCElliott1,26,ElenMFeuerriegel1,28, DanielGarcia-Martinez1,20,29,DavidJGreen1,30,AliaGurtov1,3,JoelDIrish1,31, AshleyKruger1,MyraFLaird1,11,12,DamianoMarchi1,32,MarcRMeyer1,33, ShahedNalla1,34,EnquyeWNegash1,35,CaleyMOrr1,36,DavorkaRadovcic1,37, LaurenSchroeder1,19,JillEScott1,38,ZacharyThrockmorton1,39, MatthewWTocheri40,41,CarolineVanSickle1,3,42,ChristopherSWalker1,5, PianpianWei1,43,BernhardZipfel1 1EvolutionaryStudiesInstituteandCentreofExcellenceinPalaeoSciences,University oftheWitwatersrand,Johannesburg,SouthAfrica;2SchoolofGeosciences,University oftheWitwatersrand,Johannesburg,SouthAfrica;3DepartmentofAnthropology, UniversityofWisconsin-Madison,Madison,UnitedStates;4Departmentof Anthropology,TexasA&MUniversity,CollegeStation,UnitedStates;5DepartmentofEvolutionaryAnthropology,DukeUniversity,Durham,UnitedStates;6AnthropologicalInstituteandMuseum,UniversityofZurich,Zurich,Switzerland;7DepartmentofAnthropology,UniversityofArkansas,Fayetteville,UnitedStates;8SchoolofAnthropologyandConservation,UniversityofKent,Canterbury,United Kingdom;9DepartmentofHumanEvolution,MaxPlanckInstituteforEvolutionary Anthropology,Leipzig,Germany;10DepartmentofAnthropology/Archaeologyand DepartmentofAppliedForensicSciences,MercyhurstUniversity,Erie,UnitedStates;11CenterfortheStudyofHumanOrigins,DepartmentofAnthropology,NewYork University,NewYork,UnitedStates;12NewYorkConsortiuminEvolutionary Primatology,NewYork,UnitedStates;13DepartmentofAnthropology,Dartmouth College,Hanover,UnitedStates;14DepartmentofAnthropology,Universityof ColoradoDenver,Denver,UnitedStates;15SchoolofSport,ExerciseandHealth Sciences,LoughboroughUniversity,Loughborough,UnitedKingdom;16Department ofAnthropology,TulaneUniversity,NewOrleans,UnitedStates;17Departmentof Anthropology,LehmanCollege,Bronx,UnitedStates;18DivisionofPaleontology, AmericanMuseumofNaturalHistory,NewYork,UnitedStates;19Departmentof Archaeology,UniversityofCapeTown,Rondebosch,SouthAfrica;20PaleoanthropologyGroup,MuseoNacionaldeCienciasNaturales,Madrid,Spain;21Department ofAnthropology,ModestoJuniorCollege,Modesto,UnitedStates;22Departmentof GeographyandAnthropology,LouisianaStateUniversity,BatonRouge,United States;23SchoolofHumanitiesandSocialSciences,NazarbayevUniversity,Astana, Kazakhstan;24DepartmentofPathologyandAnatomicalSciences,Universityof* Forcorrespondence: Lee.Berger@wits.ac.za Competinginterests: The authorsdeclarethatno competinginterestsexist. Funding: Seepage32 Received: 19June2015 Accepted: 04August2015 Published: 10September2015 Reviewingeditors :Johannes Krause,UniversityofT ¨ ubingen, Germany;NicholasJConard, UniversityofT ¨ ubingen,Germany CopyrightBergeretal.This articleisdistributedunderthe termsofthe CreativeCommons AttributionLicense ,which permitsunrestricteduseand redistributionprovidedthatthe originalauthorandsourceare credited. Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 1of35

PAGE 2

Missouri,Columbia,UnitedStates;25DepartmentofAnatomyandNeurobiology, UniversityofKentuckyCollegeofMedicine,Lexington,UnitedStates;26Human EvolutionaryStudiesProgramandDepartmentofArchaeology,SimonFraser University,Burnaby,Canada;27Departmentd’Anthropologie,Universit ´ edeMontr ´ eal, Montr ´ eal,Canada;28SchoolofArchaeologyandAnthropology,AustralianNational University,Canberra,Australia;29FacultyofSciences,BiologyDepartment,UniversidadAut ` onomadeMadrid,Madrid,Spain;30DepartmentofAnatomy,Midwestern University,DownersGrove,UnitedStates;31ResearchCentreinEvolutionary AnthropologyandPalaeoecology,LiverpoolJohnMooresUniversity,Liverpool, UnitedKingdom;32DepartmentofBiology,UniversityofPisa,Pisa,Italy;33DepartmentofAnthropology,ChaffeyCollege,RanchoCucamonga,UnitedStates;34DepartmentofHumanAnatomyandPhysiology,UniversityofJohannesburg, Johannesburg,SouthAfrica;35CenterfortheAdvancedStudyofHumanPaleobiology,GeorgeWashingtonUniversity,Washington,UnitedStates;36DepartmentofCell andDevelopmentalBiology,UniversityofColoradoSchoolofMedicine,Aurora, UnitedStates;37DepartmentofGeologyandPaleontology,CroatianNaturalHistory Museum,Zagreb,Croatia;38DepartmentofAnthropology,UniversityofIowa,Iowa City,UnitedStates;39DepartmentofAnatomy,DeBuskCollegeofOsteopathic Medicine,LincolnMemorialUniversity,Harrogate,UnitedStates;40HumanOrigins Program,DepartmentofAnthropology,NationalMuseumofNaturalHistory, SmithsonianInstitution,Washington,UnitedStates;41DepartmentofAnthropology, LakeheadUniversity,ThunderBay,Canada;42DepartmentofGenderandWomen’s Studies,UniversityofWisconsin-Madison,Madison,UnitedStates;43Departmentof Paleoanthropology,InstituteofVertebratePaleontologyandPaleoanthropology, Beijing,ChinaAbstractHomonaledi isapreviously-unknownspeciesofextincthominindiscoveredwithinthe DinalediChamberoftheRisingStarcavesystem,CradleofHumankind,SouthAfrica.Thisspeciesis characterizedbybodymassandstaturesimilartosmall-bodiedhumanpopulationsbutasmall endocranialvolumesimilartoaustralopiths.Cranialmorphologyof H.naledi isunique,butmost similartoearly Homo speciesincluding Homoerectus , Homohabilis or Homorudolfensis .While primitive,thedentitionisgenerallysmallandsimpleinocclusalmorphology. H.naledi hashumanlike manipulatoryadaptationsofthehandandwrist.Italsoexhibitsahumanlikefootandlowerlimb. Thesehumanlikeaspectsarecontrastedinthepostcraniawithamoreprimitiveoraustralopith-like trunk,shoulder,pelvisandproximalfemur.Representingatleast15individualswithmostskeletal elementsrepeatedmultipletimes,thisisthelargestassemblageofasinglespeciesofhomininsyet discoveredinAfrica.DOI:10.7554/eLife.09560.001IntroductionFossilhomininswerefirstrecognizedintheDinalediChamberintheRisingStarcavesystemin October2013.Duringarelativelyshortexcavation,ourteamrecoveredanextensivecollectionof 1550homininspecimens,representingnearlyeveryelementoftheskeletonmultipletimes( Figure1 ), includingmanycompleteelementsandmorphologicallyinformativefragments,someinarticulation, aswellassmallerfragmentsmanyofwhichcouldberefitintomorecompleteelements.Thecollection isamorphologicallyhomogeneoussamplethatcanbeattributedtonopreviously-knownhominin species.Herewedescribethisnewspecies, Homonaledi .Wehavenotdefined H.naledi narrowly basedonasinglejaworskullbecausetheentirebodyofmaterialhasinformedourunderstandingof itsbiology. Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 2of35 Researcharticle Genomicsandevolutionarybiology

PAGE 3

OrderPrimatesLINNAEUS1758 SuborderAnthropoideaMIVART1864 SuperfamilyHominoideaGRAY1825 FamilyHominidaeGRAY1825 TribeHomininiGRAY1825 Genus Homo LINNAEUS1758 Homonaledi sp.nov.urn:lsid:zoobank.org:pub:00D1E81A-6E08-4A01-BD98-79A2CEAE2411EtymologyTheword naledi means‘star’intheSotholanguageandreferstotheDinalediChamber’slocation withintheRisingStarcavesystem.LocalityTheDinaledichamberislocatedapproximately30metersunderground,withintheRisingStarcave systematabout261 13 S;2742 43 E.ThesystemlieswithintheMalmanidolomites,approximately 800meterssouthwestofthewell-knownsiteofSwartkransintheCradleofHumankindWorld HeritageSite,GautengProvince,SouthAfrica.HorizonandassociationsThepresentsampleofskeletalmaterialfromtheDinalediChamberwasrecoveredduringtwofield expeditions,inNovember2013andMarch2014. Sixspecimensfromanexsitucontextcanbeidentifiedasbirdbones,andfewfragmentaryrodent remainshavebeenrecoveredwithintheexcavationarea.Neitherofthesefaunalconstituentscan presentlybeassociatedwiththehomininfossilcollection( Dirksetal.,2015 ). Asidefromtheselimitedfaunalmaterials,theDinaledicollectionisentirelycomposedofhominin skeletalanddentalremains.Thecollectionsofarcomprises1550fossilhomininspecimens,this numberincludes1413bonespecimensand137isolateddentalspecimens;anadditional53teethare presentinmandibularormaxillarybonespecimens.Asidefromthefragmentaryrodentteeth,all dentalcrowns(n = 179)arehominin,recoveredbothfromsurfacecollectionandexcavation.Likewise, asidefromthefewbirdelements,allmorphologicallyinformativebonespecimensareclearlyhominin. Inallcaseswhereelementsarerepeatedinthesample,theyaremorphologicallyhomogeneous,with eLifedigestModernhumans,or Homosapiens ,arenowtheonlylivingspeciesintheirgenus. Butasrecentlyas100,000yearsago,therewereseveralotherspeciesthatbelongedtothegenus Homo .Togetherwithmodernhumans,theseextincthumanspecies,ourimmediateancestorsand theircloserelatives,arecollectivelyreferredtoas‘hominins’. NowBergeretal.reporttherecentdiscoveryofanextinctspeciesfromthegenus Homo thatwas unearthedfromdeepundergroundinwhathasbeennamedtheDinalediChamber,intheRisingStar cavesysteminSouthAfrica.Thespecieswasnamed Homonaledi; ‘naledi’means‘star’inSotho(also calledSesotho),whichisoneofthelanguagesspokeninSouthAfrica. Theunearthedfossilswerefromatleast15individualsandincludemultipleexamplesofmostof thebonesintheskeleton.Basedonthiswiderangeofspecimensfromasinglesite,Bergeretal. describe Homonaledi asbeingsimilarinsizeandweighttoasmallmodernhuman,withhuman-like handsandfeet.Furthermore,whiletheskullhadseveraluniquefeatures,ithadasmallbraincasethat wasmostsimilarinsizetootherearlyhomininspeciesthatlivedbetweenfourmillionandtwomillion yearsago. Homonaledi ’sribcage,shouldersandpelvisalsomorecloselyresembledthoseofearlier homininspeciesthanthoseofmodernhumans. The Homonaledi fossilsarethelargestcollectionofasinglespeciesofhomininthathasbeen discoveredinAfricasofarand,inarelatedstudy,Dirksetal.describethesettingandcontextfor thesefossils.However,sincetheageofthefossilsremainsunclear,oneofthenextchallengeswillbe todatetheremainstoprovidemoreinformationabouttheearlyevolutionofhumansandtheirclose relatives.DOI:10.7554/eLife.09560.002 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 3of35 Researcharticle Genomicsandevolutionarybiology

PAGE 4

Figure1 .Dinalediskeletalspecimens.Thefigureincludesapproximatelyallofthematerialincorporatedinthisdiagnosis,includingtheholotype specimen,paratypesandreferredmaterial.Thesemakeup737partialorcompleteanatomicalelements,manyofwhichconsistofseveralrefitted specimens.Specimensnotidentifiedtoelement,suchasnon-diagnosticlongboneorcranialfragments,andasubsetoffragilespecimensarenotshow n here.The‘skeleton’layoutinthecenterofthephotoisacompositeofelementsthatrepresentmultipleindividuals.Thisviewisforeshortened;the table uponwhichthebonesarearrangedis120-cmwideforscale. DOI:10.7554/eLife.09560.003 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 4of35 Researcharticle Genomicsandevolutionarybiology

PAGE 5

variationconsistentwithbodysizeandsexdifferenceswithinasinglepopulation.Theseremains representaminimumof15homininindividuals,asindicatedbytherepetitionandpresenceof deciduousandadultdentalelements. Thegeologicalageofthefossilsisnotyetknown.Excavationshavethusfarrecoveredhominin materialfromUnit2andUnit3inthechamber( Dirksetal.,2015 ).Surface-collectedhomininmaterial fromthepresenttopofUnit3,whichincludesmaterialderivedfrombothUnit2andUnit3, representsaminorityoftheassemblage,andismorphologicallyindistinguishablefrommaterial excavatedfrominsituwithinUnit3.Inadditiontogeneralmorphologicalhomogeneityincluding cranialshape,distinctivemorphologicalconfigurationsofalltherecoveredfirstmetacarpals,femora, molars,lowerpremolarsandlowercanines,areidenticalinbothsurface-collectedandexcavated specimens(seeFigure14laterinthetext).Theseincludetraitsnotfoundinanyotherhomininspecies yetdescribed.Theseconsiderationsstronglyindicatethatthismaterialrepresentsasinglespecies, andnotacommingledassemblage.Holotype,paratypes,andreferredmaterials HolotypeDinalediHominin1(DH1)comprisesthepartialcalvar ia,partialmaxilla,andnearlycompletemandibleof apresumedmaleindividual,basedonsizeandmorphologywithinthesample( Figure2 ; Supplementary file1 ).TheholotypewasrecoveredinsituduringexcavationswithintheDinalediChamberinMarchof 2014,embeddedinunconsolidatedfineclaymatrix( Dirksetal.,2015 ).Theholotypeishousedinthe EvolutionaryStudiesInstituteattheUniversityo ftheWitwatersrand,Johannesburg,SouthAfrica.ParatypesDinalediHominin2(DH2)isapartialcalvariathatpreservespartsofthefrontal,leftandrightparietals, righttemporal,andoccipital( Figure3 ; Supplementaryfile1 ).DinalediHominin3(DH3)isapartial Figure2 .Holotypespecimenof Homonaledi ,DinalediHominin1(DH1).U.W.101-1473craniumin( A )posteriorand ( B )frontalviews(frontalviewminusthefrontalfragmenttoshowcalvariainterior).U.W.101-1277maxillain( C ) medial,( D )frontal,( E )superior,and( F )occlusalviews.( G )U.W.101-1473craniuminanatomicalalignmentwith occludedU.W.101-1277maxillaandU.W.101-1261mandibleinleftlateralview.U.W.101-1277mandiblein( H ) occlusal,( I )basal,( J )rightlateral,and( K )anteriorviews.Scalebar = 10cm. DOI:10.7554/eLife.09560.019 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 5of35 Researcharticle Genomicsandevolutionarybiology

PAGE 6

calvariaofapresumedfemaleindividualthatpreservespartsofthefrontal,leftparietal,lefttemporal, andsphenoid( Figure4 , Supplementaryfile1 ). DinalediHominin4(DH4)isapartialcalvariathat preservespartsoftherighttemporal,rightparietal, andoccipital( Figure3 ; Supplementaryfile1 ). DinalediHominin5(DH5)isapartialcalvaria thatpreservespartofthelefttemporaland occipital( Figure3 ; Supplementaryfile1 ).U.W. 101-377isamandibularfragmentthatpreserves dentalanatomyinanunwornstate;atpresentit cannotbedefinitivelyassociatedwithanyof theseDinalediHominin(DH)individuals,and indeedmightrepresentanotherindividual ( Figure5 ; Supplementaryfile1 ).Thesecranial specimensagreecloselyinnearlyallmorphologicaldetailswheretheyoverlapinareaspreservedexceptthoseweinterpretasrelated tosex. Dinaledihand1(H1)isanearlycomplete (missingonlythepisiform)righthand,found articulatedinassociation,comprisingspecimens U.W.101-1308to 1311, 1318to 1321, 1325to 1329, 1351, 1464,and 1721to 1732 ( Figure6 ; Supplementaryfile1 ).U.W.101-1391isaproximalrightfemurpreservingpartofthe head,theneck,someofthelesserandgreatertrochanter,andtheproximalshaft( Figure7 ; Supplementaryfile1 ).U.W.101-484isarighttibialdiaphysismissingonlytheproximalend ( Figure8 ; Supplementaryfile1 ).Dinaledifoot1(F1)isapartialfootskeletonmissingonlythemedial Figure3 .Cranialparatypes.( A )DH2,rightlateralview. ( B )DH5,leftlateralview.( C )DH4,rightlateralview. ( D )DH4,posteriorview.Scalebar = 10cm. DOI:10.7554/eLife.09560.005 Figure4 .ParatypeDH3.( A )Frontalview.( B )Leftlateralview,withcalvariainarticulationwiththemandible (U.W.101-361).( C )Basalview.Mandiblein( D )medialview;( E )occlusalview;( F )basalview.DH3wasarelativelyold individualattimeofdeath,withextremetoothwear.Scalebar = 10cm. DOI:10.7554/eLife.09560.006 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 6of35 Researcharticle Genomicsandevolutionarybiology

PAGE 7

cuneiformandthephalangesofraysII–V.Foot1 iscomposedofspecimensU.W.101-1322, 1417 to 1419, 1439, 1443, 1456to 1458, 1551, 1553, 1562,and 1698( Figure9 ; Supplementaryfile1 ).ReferredmaterialReferredmaterialisalsolistedin Supplementary file1 .Wereferto H.naledi allhomininmaterial fromtheDinaledicollectionthatcanbeidentified toelement;intotal,theholotypes,paratypesand referredmaterialcomprise737partialorcompleteanatomicalelements. Specimennumbersinthecollectionare assignedatthepointofexcavation.Laterlaboratoryanalysesallowedustorefitspecimensinto morecompleteelements,whichwehaveusedas unitsofanatomicalstudy.Herewereferto refittedelementsbyonlyasinglespecimen number;eitherthenumberofthemostconstitutivespecimen,orthefirstdiagnosticparttobe discovered.DHdesignationsarereservedfor clearlyassociatedindividuals;atthistimethese arelimitedtothefivepartialcraniadesignated above.Futureexcavationandanalyseswillcertainlyuncovermorerefitsamongspecimens.Asrefitsare found,allnumbersassignedtorefittedelementswillremainstable,andallnumbersin Supplementary file1 willberetained. Thecollectionismorphologicallyhomogeneousinallduplicatedelements,exceptforthose anatomicalfeaturesthatnormallyreflectbodysizeorsexdifferencesinotherprimatetaxa.Therefore, althoughwerefertotheholotypeandtheparatypesfordifferentialdiagnoses;thesectiondescribing theoverallanatomyencompassesallmorphologicallyinformativespecimens.DifferentialdiagnosisThiscomprehensivedifferentialdiagnosisisbaseduponcranial,dentalandpostcranialcharacters.The hypodigmsusedforotherspeciesaredetailedinthe‘Materialsandmethods’.Weexaminedoriginal specimensformostspecies,exceptwhereindicatedinthe‘Materialsandmethods’;whenwereliedon othersourcesforanatomicalobservationsweindicatethis.Asummaryoftraitsof H.naledi in comparisontootherspeciesispresentedin Supplementaryfile2 .Comparativecranialandmandibular measuresarepresentedin Table1 ,andcomparativedentalmeasuresareprovidedin Table2 .Cranium,mandible,anddentition(DH1,DH2,DH3,DH4,DH5,U.W. 101-377)Thecraniumof H.naledi doesnothavethewell-developedcrestpatternsthatcharacterize Australopithecusgarhi ( Asfawetal.,1999 )andspeciesofthegenus Paranthropus ,northederived facialmorphologyseeninthelattergenus.Themandibleof H.naledi isnotablymoregracilethan thoseof Paranthropus .Althoughmaxillaryandmandibularincisorsandcaninesof H.naledi overlapin sizewiththoseof Paranthropus ,thepost-canineteetharenotablysmallerthanthoseof Paranthropus and Au.garhi ,withmandibularmolarsthatarebuccolinguallynarrow. H.naledi differsfrom Australopithecusafarensis and Australopithecusafricanus inhaving apentagonal-shapedcranialvaultinposteriorview,sagittalkeeling,widelyspacedtemporallines, anangulartorus,adeepandnarrowdigastricfossa,anexternaloccipitalprotuberance,ananteriorly positionedrootofthezygomaticprocessofthemaxilla,abroadpalate,andasmallcaninejugum lackinganteriorpillars.Theanteriorandlateralvaultof H.naledi differsfrom Au.afarensis and Au.africanus inexhibitingonlyslightpost-orbitalconstriction,frontalbossing,awell-developed supraorbitaltoruswithawell-definedsupratoralsulcus,temporallinesthatarepositionedonthe Figure5 .U.W.101-377mandible.( A )Lateralview; ( B )medialview;( C )basalview;( D )occlusalview.( D )The distinctivemandibularpremolarmorphologywith elongatedtalonidsinunwornstate.Scalebar = 2cm. DOI:10.7554/eLife.09560.007 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 7of35 Researcharticle Genomicsandevolutionarybiology

PAGE 8

posteriorratherthanthesuperioraspectofthesupraorbitaltorus,arootofthezygomaticprocessof thetemporalthatisangleddownwardsapproximately30relativetotheFrankfortHorizontal(FH)and whichbeginsitslateralexpansionabovethemandibularfossaratherthantheEAM,amandibular Figure6 .Hand1.Palmarviewonleft;dorsalviewonright.Thishandwasdiscoveredinarticulationandallbonesare representedexceptforthepisiform.Theproportionsofdigitsarehumanlikeandvisuallyapparent,asarethe expandeddistalapicaltuftsonalldigits,therobustpollicalray,andtheuniquefirstmetacarpalmorphology. DOI:10.7554/eLife.09560.008 Figure7 .U.W.101-1391paratypefemur.( A )Medialview;( B )posteriorview;( C )lateralview;( D )anteriorview.The femurneckisrelativelylongandanteroposteriorlycompressed.Theanteversionoftheneckisevidentinmedialview. Scalebar = 2cm. DOI:10.7554/eLife.09560.009 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 8of35 Researcharticle Genomicsandevolutionarybiology

PAGE 9

fossathatispositionedmedialtothewallofthetemporalsquame,asmallpostglenoidprocessthat contactsthetympanic,acoronallyorientedpetrous,andasmallandobliquelyorientedEAM.The H.naledi mandibleexhibitsamoregracilesymphysisandcorpus,amoreverticallyinclinedsymphysis, aslightmandibularincurvationdelineatingafaintmentaltrigon,andasteeplyinclinedposteriorface ofthemandibularsymphysiswithoutapostincisiveplanum.Theincisorsof H.naledi overlapinsize withsomespecimensof Au.africanus ,thoughthecaninesandpost-caninedentitionarenotably smaller,withrelativelynarrowbuccolingualdimensionsofthemandibularmolars.ThemaxillaryI1lacksamedianlingualridgeandexhibitsabroadanduninflatedlingualcervicalprominence,the lingualmesialanddistalmarginalridgesdonotmergeontothecervicalprominenceinthemaxillaryI2, themandibularcanineexhibitsonlyaweaklingualmedianridgeandabroadanduninflatedlingual cervicalprominence,andthebuccalgroovesonthemaxillarypremolarsareonlyweaklydeveloped. H.naledi exhibitsasmallandisolatedCarabelliÂ’sfeatureinthemaxillarymolars,unlikethemore prominentandextensiveCarabelliÂ’sfeatureof Australopithecus .Moreover,the H.naledi mandibular molarspossesssmall,mesiobuccallyrestrictedprotostylidsthatdonotintersectthebuccalgroove, differingfromthetypicallyenlarged,centrallypositionedprotostylidsthatintersectthebuccalgroove in Australopithecus . Thecraniumof H.naledi differsfrom Australopithecussediba ( Bergeretal.,2010 )inexhibiting sagittalkeeling,amorepronouncedsupraorbitaltorusandsupratoralsulcus,aweaklyarched supraorbitalcontourwithroundedlateralcorners,anangulartorus,awell-definedsupramastoidcrest, acurvedsuperiormarginofthetemporalsquama,arootofthezygomaticprocessofthetemporal thatisangleddownwardsapproximately30relativetoFH,aflattenednasoalveolarclivus,weak caninejuga,ananteriorlypositionedrootofthezygomaticprocessofthemaxilla,andarelatively broadpalatethatisanteriorlyshallow.The H.naledi mandible(DH1)hasamentalforamenpositioned superiorlyonthecorpusthatopensposteriorly,unlikethemid-corpusheight,morelaterallyopening mentalforamenof Au.sediba .Themaxillaryandmandibularteethof H.naledi aresmallerthanthose of Au.sediba ,withmandibularmolarsthatarebuccolinguallynarrow.Thelingualmesialanddistal marginalridgesdonotmergeontothecervicalprominenceinthemaxillaryI2,theparaconeofthe maxillaryP3isequalinsizetotheprotocone,the protoconidandmetaconidofthemandibular molarsareequallymesiallypositioned,andthe lingualcuspsofthemolarsarepositionedatthe occlusobuccalmarginwhilethebuccalcuspsare positionedslightlylingualtotheocclusobuccal margin.Also, Au.sediba shareswithother australopithsaprotostylidthatiscentrallylocatedandwhichintersectsthebuccalgrooveof thelowermolars,unlikethesmallandmesiobuccallyrestrictedprotostylidthatdoesnotintersect thebuccalgroovein H.naledi . Thecraniumof H.naledi differsfrom Homo habilis inexhibitingsagittalkeeling,aweakly archedsupraorbitalcontour,temporallinesthat arepositionedontheposteriorratherthanthe superioraspectofthesupraorbitaltorus,an angulartorus,anoccipitaltorus,onlyslightpostorbitalconstriction,acurvedsuperiormarginof thetemporalsquama,asuprameatalspine, aweakcristapetrosa,aprominentEustachian process,asmallEAM,weakcaninejuga,andan anteriorlypositionedrootofthezygomaticprocessofthemaxilla.Mandiblesattributedto H. habilis showaweaklyinclined,shelf-likepost incisiveplanumwithavariablydevelopedsuperiortransversetorus,unlikethesteeplyinclined posteriorfaceofthemandibularsymphysisof H. naledi ,whichlacksbothapostincisiveplanumor Figure8 .U.W.101-484paratypetibia.( A )Anteriorview; ( B )medialview;( C )posteriorview;( D )lateralview.Thetibiae arenotablyslenderfortheirlength.Scalebar = 10cm. DOI:10.7554/eLife.09560.010 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 9of35 Researcharticle Genomicsandevolutionarybiology

PAGE 10

superiortransversetorus.The H.naledi mandible (DH1)hasamentalforamenpositionedsuperiorly onthecorpusthatopensposteriorly,whilethe mentalforamenof H.habilis isatmid-corpus heightandopensmorelaterally.Themaxillary andmandibulardentitionsofDH1aresmaller thantypicalfor H.habilis .ThemandibularP3of H.naledi ismoremolarizedandlackstheocclusal simplificationseenin H.habilis ;ithasasymmetricalocclusaloutline,andmultipleroots(two roots:mesiobuccalanddistal)notseenin H. habilis .Themolarsof H.naledi lackcrenulation, secondaryfissures,andsupernumerarycuspsthat arecommonto H.habilis .Theprotoconidand metaconidofthemandibularmolarsareequally mesiallypositioned. Thecraniumof H.naledi differsfrom Homo rudolfensis byitssmallercranialcapacity,andby exhibitingfrontalbossing,apost-bregmaticdepression,sagittalkeeling,awell-developedsupraorbitaltorusdelineatedbyadistinct supratoralsulcus,temporallinesthatarepositionedontheposteriorratherthanthesuperioraspect ofthesupraorbitaltorus,anoccipitaltorus,anexternaloccipitalprotuberance,onlyslightpost-orbital constriction,asmallpostglenoidprocess,aweakcristapetrosa,alaterallyinflatedmastoidprocess, acaninefossa,incisorsthatprojectanteriorlybeyondthebi-canineline,andashallowanteriorpalate. Asin H.habilis ,mandiblesattributedto H.rudolfensis showaweaklyinclined,shelf-likepostincisive planumwithavariablydevelopedsuperiortransversetorus,unlikethesteeplyinclinedposteriorface ofthemandibularsymphysisofDH1,thelatterofwhichlackseitherapostincisiveplanumorsuperior transversetorus.Themandibularsymphysisandcorpusof H.naledi aremoregracilethanthose attributedto H.rudolfensis ,andthe H.naledi mandible(DH1)hasamentalforamenpositioned superiorlyonthecorpusthatopensposteriorly,unlikethemid-corpusheight,morelaterallyopening mentalforamenof H.rudolfensis .Themaxillaryandmandibulardentitionof H.naledi issmallerthan thatofmostspecimensof H.rudolfensis ,withonlyKNM-ER60000andKNM-ER62000appearing similarinsizeforsometeeth( Leakeyetal.,2012 ).Themolarsof H.naledi lackcrenulation,secondary fissures,orsupernumerarycuspscommonin H.rudolfensis .Thebuccalgroovesofthemaxillary premolarsareweakin H.naledi ,andtheprotoconidandmetaconidofthemandibularmolarsare equallymesiallypositioned. H.naledi lacksthetypicallydistinctivelongandlowcranialvaultof Homoerectus ,aswellasthe metopickeelingthatistypicallypresentinthelatterspecies. H.naledi alsodiffersfrom H.erectus in havingadistinctexternaloccipitalprotuberanceinadditiontothetuberculumlinearum,alaterally inflatedmastoidprocess,aflatandsquarednasoalveolarclivus,andananteriorlyshallowpalate.The parasagittalkeelingthatispresentbetweenbregmaandlambdain H.naledi (DH1andDH3)isless markedthanoftenoccursin H.erectus ,includinginsmallspecimenssuchasKNM-ER42700andthe Dmanisicranialsample.Alsounlikemostspecimensof H.erectus , H.naledi hasasmallvaginal process,aweakcristapetrosa,amarkedEustachianprocess,andasmallEAM.Themandibleof H. erectus showsamoderatelyinclined,shelf-likepostincisiveplanumterminatinginavariably developedsuperiortransversetorus,differingfromthesteeplyinclinedposteriorfaceofthe H.naledi mandibularsymphysis,whichlacksbothapostincisiveplanumorasuperiortransversetorus.The mentalforamenispositionedsuperiorlyandopensposteriorlyinDH1,unlikethemid-corpusheight, morelaterallyopeningmentalforamenof H.erectus .Themaxillaryandmandibularincisorsand caninesof H.naledi aresmallerthantypicalof H.erectus.ThemandibularP3of H.naledi ismore molarizedandlackstheocclusalsimplificationseenin H.erectus ,theyrevealasymmetricalocclusal outline,andmultipleroots(2R:MB + D)nottypicallyseenin H.erectus .Furthermore,themolarsof H.naledi lackcrenulation,secondaryfissures,orsupernumerarycuspscommonin H.erectus . H.naledi lacksthereducedcranialheightof Homofloresiensis ,anddisplaysamarkedangulartorus andparasagittalkeelingbetweenbregmaandlambdathatisabsentinthelatterspecies. H.naledi Figure9 .Foot1in( A )dorsalview;and( B )medialview. ( C )Proximalarticularsurfacesofthemetatarsalsof Foot1,showninarticulationtoillustratetransversearch structure.Scalebar = 10cm. DOI:10.7554/eLife.09560.011 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 10of35 Researcharticle Genomicsandevolutionarybiology

PAGE 11

Table1 .Cranialandmandibularmeasurementsfor H.naledi ,earlyhominins,andmodernhumans Measurement definitions asin Wood (1991) P. aethiopicus P. boisei P. robustus Au. afarensis Au. africanus Au. sediba H. naledi H. habilis H. rudolfensis H. erectus MP Homo H. sapiensCranium Cranialcapacity–41048549345746742051361077686512661330 Porionheight67274–86706781779094101112 Posteriorcranial length 35847546044–656070799981 Bi-parietalbreadth99498–9099100103107118129142132 Bi-temporalbreadth10110109108115104101107112126131146127 Closestapproachof temporallines –crest*crest*crest*crest*21565235517210196 Supraorbitalheight index –465350516056566459566271 Minimumpost-orbital breadth –626670776770687578899697 Superiorfacial breadth 49100107109–95868697113110124107 Post-orbital constrictionindex† –626164–6981797274818091 EAMarea(asan ellipse)‡ –77801037096–3876–958561 Rootofzygomatic processorigin –P4P4P3toM1P4toM1P4toM1P4P3 toP4 P4 toM1 P4toM1P4toM1M1M1 Petromedianangle1375045503133–5548–525546 Maxilloalveolar process Maxilloalveolar length 87947869677163576568666955 Maxilloalveolar breadth 88837669686663716872707262 Palatebreadth91324035303629443840385640 Table1.Continuedonnextpage Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 11of35 Researcharticle Genomicsandevolutionarybiology

PAGE 12

Table1.ContinuedMeasurement definitions asin Wood (1991) P. aethiopicus P. boisei P. robustus Au. afarensis Au. africanus Au. sediba H. naledi H. habilis H. rudolfensis H. erectus MP Homo H. sapiensPalatedepthat incisivefossa –31110109105101311109 PalatedepthatM110371811111310101216151813 Mandible Symphysisheight141374942393732333137353434 Symphysiswidth142262825202118182024181714 Symphysisarea atM1(asan ellipse)‡ 1467571114835623606452467393723519474365 Corpusheight atM1 150384236343230262936313128 Corpusbreadth atM1 151252926202118162022191913 Corpusareaat M1(asan ellipse)‡ 152742955736540539405326425631458469296 Mentalforamen heightindex § –515054585350404649484850 *Atleastinpresumedmales. †Post-orbitalbreadth/superiorfacialbreadth × 100. ‡Followingtheformula( × (corpusheight/2) × (corpusbreadth/2)). § Heightofmentalforamenfromalveolarborderrelativetocorpusheightatthementalforamen. MP,MiddlePleistocene. Unlessotherwiseindicatedmeasurementsaredefinedasin Wood(1991) .Chorddistancesareinmm.Datafor H.naledi collectedfromoriginalfossilsorlaserscansbyDJdeRandHMG; comparativedatacollectedbyDJdeRonoriginalfossilsandcastsandsupplementedbydatafrom Wood(1991) . DOI:10.7554/eLife.09560.012 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 12of35 Researcharticle Genomicsandevolutionarybiology

PAGE 13

Table2 .Dentalmeasuresfor H.naledi andcomparativehomininspecies Maxillary I1I2CP3P4M1M2M3MDLLMDLLMDLLMDBLMDBLMDBLMDBLMDBLAu.anamensis n35–2677653121010898 mean10.88.7–7.311.010.69.912.68.913.611.512.913.014.412.514.2 range9.1–12.48.2–9.3–7.0–7.59.9–12.39.1–11.88.2–11.810.1–14.37.2–12.112.6–14.27.8–14.39.0–16.710.9–16.312.9–16.111.1–15.713.0–1 5.7 Au.afarensis n7899151512101812161310111111 mean10.78.47.57.29.910.88.812.49.112.412.013.412.914.612.714.5 range9.9–11.87.1–9.76.6–8.26.2–8.18.8–11.69.3–12.57.7–9.711.3–13.47.6–10.811.1–14.510.5–13.812.0–15.012.1–13.613.4–15.210.9–14. 813.1–16.3 Au.africanus n15151110161326252020212023242728 mean10.78.36.96.89.910.39.212.79.513.412.913.914.115.714.216.0 range9.4–12.57.4–9.15.8–8.05.6–7.98.8–11.08.7–12.08.5–10.210.7–14.57.2–11.012.4–15.311.7–14.412.9–15.312.1–16.312.8–17.911.2–16 .913.1–18.6 Au.sediba n1111111111111122 mean10.16.97.26.69.08.89.011.29.312.112.912.012.913.713.013.5 range––––––––––––––12.6–13.312.9–14.1 H.naledi n–548109101077121311977 mean9.46.56.66.28.18.68.010.58.111.011.611.712.212.811.612.4 range8.8–9.86.3–7.06.3–7.05.8–6.67.3–8.98.0–9.67.7–8.49.8–11.0 7.7–8.710.5–11.210.5–12.411.2–12.411.0–13.011.9–13.611.0–12.711.4 –13.4 H.habilis n224423778813137777 mean10.68.07.46.69.09.89.011.99.212.112.713.012.714.312.314.7 range10.1–11.17.3–8.76.7–8.16.0–7.98.5–9.48.5–11.68.1–9.611.0–12 .78.5–9.911.0–13.111.6–13.912.1–14.111.8–13.513.5–16.211.3–13.9 13.2–16.6 H.rudolfensis n11––111122222211 mean12.37.7––11.512.510.513.610.212.514.014.014.315.813.313.5 range––––––––9.7–10.711.1–13.813.9–14.213.3–14.814.1–14.614.1–17.6–– H.erectus n111266121227273029343222221616 mean10.38.17.78.09.510.08.511.88.111.612.213.212.013.310.512.8 range8.1–12.67.0–11.76.0–8.36.9–8.58.5–11.19.0–11.87.1–10.19.5–13.87.0–9.49.9–13.410.1–14.611.0–15.910.3–13.610.9–15.58.7–14.71 0.4–15.8 H. neanderthalensis n28373541282916172119232427282221 mean9.78.58.08.48.810.18.010.67.810.611.612.310.912.59.912.3 range8.2–11.87.3–9.95.8–9.35.8–9.97.2–10.07.6–11.46.6–9.38.4–11.85.9–11.58.3–11.79.5–13.511.0–14.28.9–15.910.8–14.68.2–11.49.8– 14.6 Table2.Continuedonnextpage Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 13of35 Researcharticle Genomicsandevolutionarybiology

PAGE 14

Table2.ContinuedMaxillary I1I2CP3P4M1M2M3MDLLMDLLMDLLMDBLMDBLMDBLMDBLMDBLH. heidelbergensis n21231921272925252223252424232627 mean9.67.87.77.88.89.87.910.67.610.311.211.910.212.38.911.6 range8.7–10.77.1–9.97.2–8.47.3–8.68.1–11.08.8–11.87.1–9.09.2–12.27.0–8.89.1–11.59.9–12.310.3–13.28.1–12.111.1–13.87.6–11.010.0– 13.2 MP/LPAfrican Homo n6678446610101414202099 mean9.07.87.47.28.99.78.410.88.110.812.313.211.012.99.211.7 range6.3–10.96.6–8.76.0–9.36.1–8.58.2–9.58.8–10.08.1–8.79.9–11.87.5–9.39.4–12.810.4–14.012.0–15.07.8–13.011.0–15.07.6–10.210.0– 13.2Mandibular I1I2CP3P4M1M2M3MDLLMDLLMDLLMDBLMDBLMDBLMDBLMDBLAu.anamensis n21437788889107788 mean6.97.47.88.310.010.412.49.29.111.312.912.314.013.415.313.4 range6.8–6.9–6.6–8.77.9–8.66.6–13.99.2–11.411.3–13.48.6–10.07.4–9. 89.6–13.211.6–13.810.2–14.813.0–15.912.3–14.913.7–17.012.1–15 .2 Au.afarensis n7876131627262421322631272623 mean6.77.16.78.08.810.49.610.69.811.013.112.614.313.415.313.5 range5.6–7.75.6–8.05.0–8.06.7–8.87.5–11.78.0–12.47.9–12.68.9–13.87.7–11.49.8–12.810.1–14.811.0–14.012.1–16.511.1–15.213.4–18.11 1.3–15.3 Au.africanus n11121213232520212523293238383435 mean6.26.77.27.99.410.19.711.510.411.614.013.015.714.516.314.6 range4.8–6.95.7–7.95.6–8.16.6–9.28.5–10.78.2–12.18.8–11.09.9–13.98.7–12.39.3–13.212.4–15.811.2–15.114.2–17.712.8–16.813.5–18.51 2.2–16.8 Au.sediba n–1–1221111222222 mean–5.9–6.67.78.08.19.28.89.713.111.414.512.814.913.2 range––––7.3–8.07.4–8.6––––13.1–13.111.3–11.514.4–14.512.3–13.214.9–14.912.5–13.6 H.naledi n7756779106611119965 mean6.15.46.95.97.17.19.08.88.79.112.210.713.311.213.412.1 range5.7–7.05.3–5.96.6–7.45.9–6.06.4–7.56.9–7.48.2–9.48.2–9.7 8.3–9.08.5–10.211.3–12.710.3–11.412.3–14.010.7–12.212.9–13.711.7–1 2.8 H.habilis n2222324433554444 mean6.46.87.47.68.79.09.69.69.910.513.711.915.013.515.413.3 range6.4–6.56.7–7.07.2–7.77.6–7.67.6–9.67.9–10.09.0–10.68.6–11.19.0–10.59.9–11.013.0–14.810.9–12.814.2–15.712.0–15.114.8–15.912 .4–14.4 Table2.Continuedonnextpage Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 14of35 Researcharticle Genomicsandevolutionarybiology

PAGE 15

Table2.ContinuedMandibular I1I2CP3P4M1M2M3MDLLMDLLMDLLMDBLMDBLMDBLMDBLMDBLH.rudolfensis n–1–1–13366556533 mean–5.4–6.7–8.39.911.110.111.414.012.716.013.716.414.1 range––––––9.0–10.79.5–12.38.8–11.89.8–12.212.8–15.211.4–13.214.0–18.312.7–14.915.6–17.013.1–14.6 H.erectus n11121416141630302526434341402627 mean6.26.477.28.79910.18.710.112.711.913.312.512.711.7 range4.8–7.45.8–7.15.3–8.16.4–8.57.0–10.38.0–10.47.0–12.08.2–12.07.2–10.38.0–12.59.9–14.810.1–13.311.3–15.310.8–14.310.0–15.210 .0–14.2 H.neanderthalensis n9162331364120212325384026271820 mean5.67.26.87.87.88.87.99.17.89.411.811.112.111.312.011.0 range4.2–6.45.2–8.85.9–7.56.8–9.06.7–8.86.8–10.36.6–9.18.0–10.36.5–9.48.5–10.510.1–13.610.2–12.99.3–14.08.8–12.411.2–13.99.9–12 .2 H.heidelbergensis n21221920232422222626292929293232 mean5.66.76.57.37.68.77.98.97.28.711.310.611.210.511.510.0 range4.8–6.56.0–7.56.0–7.26.6–8.06.9–9.07.3–10.07.2–9.07.6–11.66.6–8.87.2–11.710.4–13.89.6–13.09.7–14.68.5–13.99.7–13.28.6–12.5 MP/LPAfrican Homo n55888888129161620201313 mean6.06.86.87.28.89.68.69.88.610.313.111.812.511.712.411.5 range5.7–6.46.1–7.25.6–8.36.4–8.07.8–10.08.8–10.37.7–9.08.6–11.26.9–9.69.3–11.410.7–14.210.0–13.010.8–15.09.2–13.610.6–13.59.9– 12.7 MP,MiddlePleistoceneandLP,LatePleistocene. DOI:10.7554/eLife.09560.013 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 15of35 Researcharticle Genomicsandevolutionarybiology

PAGE 16

furtherhasaflatandsquarednasoalveolarclivus,unlikethepronouncedmaxillarycaninejugaand prominentpillarsof H.floresiensis .Themandibleof H.floresiensis showsaposteriorlyinclinedpost incisiveplanumwithsuperiorandinferiortransversetori,differingfromthesteeplyinclinedposterior faceofthe H.naledi mandibularsymphysis,whichlacksbothapostincisiveplanumorasuperior transversetorus.Dentally, H.naledi isdistinguishablefrom H.floresiensis bythemesiodistal elongationandextensivetalonidofthemandibularP4,andthelackofTomes’rootonthemandibular premolars.Themolarsizegradientof H.naledi followstheM1 < M2 < M3pattern,unliketheM3 < M2 < M1patternin H.floresiensis ,andthemandibularmolarsarerelativelymesiodistallylongand buccolinguallynarrowcomparedtothoseof H.floresiensis . H.naledi differsfromMiddlePleistocene(MP)andLatePleistocene(LP) Homo (hereweinclude specimenssometimesattributedtotheputativeEarlyPleistocenetaxon Homoantecessor ,andMP Homoheidelbergensis , Homorhodesiensis ,aswellasarchaic Homosapiens andNeandertals)in exhibitingasmallercranialcapacity. H.naledi hasitsmaximumcranialwidthinthesupramastoid region,ratherthanintheparietalregion.Ithasaclearlydefinedcaninefossa(similarto H.antecessor ), ashallowanteriorpalate,andaflatandasquarednasoalveolarclivus. H.naledi lacksthebilaterally archedandverticallythickenedsupraorbitaltorifoundinMPandLP Homo . H.naledi alsodiffersin exhibitingarootofthezygomaticprocessofthetemporalthatisangleddownwardsapproximately 30relativetoFH,aprojectingentoglenoidprocess,aweakvaginalprocess,aweakcristapetrosa, aprominentEustachianprocess,alaterallyinflatedmastoidprocess,andasmallEAM.The H.naledi mandibletendstobemoregracilethanspecimensofMP Homo .Themandibularcanineretains adistinctaccessorydistalcuspulidnotseeninMPandLP Homo .Molarcuspalproportionsfor H. naledi donotshowthederivedreductionoftheentoconidandhypoconidthatcharacterizesMPand LP Homo .ThemandibularM3isnotreducedinDH1,thusrevealinganincreasingmolarsizegradient thatcontrastswithreductionoftheM3inMPandLP Homo . H.naledi differsfrom H.sapiens inexhibitingsmallcranialcapacity,awell-definedsupraorbital torusandsupratoralsulcus,arootofthezygomaticprocessofthetemporalthatisangleddownwards approximately30relativetoFH,alargeandlaterallyinflatedmastoidwithwell-developed supramastoidcrest,anangulartorus,asmallvaginalprocess,aweakcristapetrosa,aprominent Eustachianprocess,asmallEAM,aflatandsquarednasoalveolarclivus,andamoreposteriorly positionedincisiveforamen.The H.naledi mandibleshowsaweaker,lesswell-definedmentum osseumthan H.sapiens ,aswellasaslightinferiortransversetorusthatisabsentinhumans.The mentalforamenispositionedsuperiorlyin H.naledi ,unlikethemid-corpusheightmentalforamenof H.sapiens .Themandibularcaninepossessesadistinctaccessorydistalcuspulidnotseenin H. sapiens .Molarcuspalproportionsfor H.naledi donotshowthederivedreductionoftheentoconid andhypoconidthatcharacterizes H.sapiens .ThemandibularM3isnotreducedin H.naledi ,thus revealinganincreasingmolarsizegradientthatcontrastswithreductionoftheM3in H.sapiens .Hand(H1)H.naledi possessesacombinationofprimitiveandderivedfeaturesnotseeninthehandofanyother hominin.H1isdifferentiatedfromtheestimatedintrinsichandproportionsof Au.afarensis inhaving arelativelylongthumb((Mc1 + PP1)/(Mc3 + PP3 + IP3))( RolianandGordon,2013 ; Alm ´ ecijaand Alba,2014 ).Itisfurtherdistinguishedfrom Au.afarensis , Au.africanus, and Au.sediba inhaving awell-developedcrestforboththe opponenspollicis andfirstdorsal interosseous muscles, atrapezium-scaphoidjointthatextendsontothescaphoidtubercle,arelativelylargeandmore palmarly-positionedcapitate-trapezoidjoint,and/orasaddle-shapedMc5-hamatejoint. H.naledi also differsfrom Au.sediba inthatitlacksmediolaterallynarrowMc2-5shafts( Kivelletal.,2011 ).Manual morphologyof Au.garhi iscurrentlyunknown. H1isdistinguishedfrom H.habilis inhavingadeepproximalpalmarfossawithawell-developed ridgedistallyfortheinsertionofthe flexorpollicislongus muscleonthefirstdistalphalanx,and amoreproximodistallyorientedtrapezium-secondmetacarpaljoint.Italsodiffersfromboth H.habilis and H.floresiensis byhavingarelativelylargetrapezium-scaphoidjointthatextendsontothe scaphoidtubercle,andfrom H.floresiensis inhavingaboot-shapedtrapezoidwithanexpanded palmarsurface,andarelativelylargeandmorepalmarly-positionedcapitate-trapezoidjoint( Tocheri etal.,2005 , 2007 ; Orretal.,2013 ). H1isdissimilartohandremainsattributedto Paranthropusrobustus /early Homo fromSwartkrans ( Susman,1988 ; Susmanetal.,2001 )inhavingarelativelysmallMc1baseandproximalarticular Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 16of35 Researcharticle Genomicsandevolutionarybiology

PAGE 17

facet,asaddle-shapedMc5-hamatejoint,andmorecurvedproximalandintermediatephalangesof ray2–5. Manualmorphologyof H.rudolfensis iscurrentlyunknown,andthatof H.erectus islargely unknown.Still,H1differsfromathirdmetacarpalattributedto H.erectuss.l. ,aswellasfrom Homo neanderthalensis and H.sapiens bylackingastyloidprocess( Wardetal.,2013 ). H1isfurtherdistinguishedfrom H.neanderthalensis and H.sapiens byitsrelativelysmallfacetsfor theMc1andscaphoidonthetrapezium,itslowanglebetweentheMc2andMc3facetsonthe capitate,andbyitslongandcurvedproximalandintermediatephalangesonrays2–5. H1isdifferentiatedfromallknownhomininsinhavingaMc1thatcombinesamediolaterallynarrow proximalendandarticularfacetwithamediolaterallywidedistalshaftandhead,adorsopalmarlyflat andstronglyasymmetric(withanenlargedpalmar-lateralprotuberance)Mc1head,andthe combinationofanoveralllater Homo -likecarpalmorphologycombinedwithproximaland intermediatephalangesthataremorecurvedthanmostaustralopiths.H1alsodiffersfromallother knownhomininsexcept H.neanderthalensis inhavingnon-pollicaldistalphalangeswithmediolaterally broadapicaltufts(relativetolength).Femur(U.W.101-1391)Thefemurof H.naledi differsfromthoseofallotherknownhomininsinitspossessionoftwowelldefined,mediolaterally-runningpillarsinthefemoralneck.Thepillarsrunalongthesuperoanterior andinferoposteriormarginsoftheneckanddefineadistinctsulcusalongitssuperioraspect.Tibia(U.W.101-484)Thetibiaof H.naledi differsfromthoseofallotherknownhomininsinitspossessionofadistinct tubercleforthepesanserinustendon.Thetibiadiffersfromotherhomininsexcept H.habilis, H.floresiensis ,and(variably) H.sapiens initspossessionofaroundedanteriorborder.Foot(F1)Thefootof H.naledi differsfromthepedalremainsof Au.afarensis,Au.africanus, and Au.sediba in havingacalcaneuswithaweaklydevelopedperonealtrochlea.F1alsodiffersfrom Au.afarensis in havingahigherorientationofthecalcanealsustentaculumtali.F1canbefurtherdistinguishedfrom pedalremainsattributedto Au.africanus inhavingahighertalarheadandnecktorsion,anarrower Mt1base,adorsallyexpandedMt1head,andgreaterproximolateraltodistomedialorientationofthe lateralmetatarsals.The H.naledi footcanbefurtherdifferentiatedfromthefootof Au.sediba in havingaproximodistallyflattertalartrochlea,aflatsubtalarjoint,adiagonallyorientedretrotrochlear eminenceandaplantarpositionofthelateralplantarprocessofthecalcaneous,anddorsoplantarly flatarticularsurfaceforthecuboidontheMt4( Zipfeletal.,2011 ).Pedalremainsof Au.garhi are currentlyunknown,andthoseof P.robustus aretoopoorlyknowntoallowforcomparison. The H.naledi footcanbedistinguishedfromthefootof H.habilis bythepresenceofaflatter, non-slopingtrochleawithequallyelevatedmedialandlateralmargins,anarrowerMt1base,greater proximolateraltodistomedialorientationofthelateralmetatarsals,andametatarsalrobusticityratio of1 > 5 > 4 > 3 > 2.Pedalmorphologyin H.rudolfensis iscurrentlyunknown,andthatof H.erectus is toopoorlyknowntoallowforcomparison.The H.naledi footcanbedistinguishedfromthefootof H.floresiensis byalongerhalluxandshortersecondthroughfifthmetacarpalsrelativetohindfoot length,andhighertorsionofthetalarheadandneck. Thefootof H.naledi canbedistinguishedfromthefootof H.sapiens onlybyitsflatterlateraland medialmalleolarfacetsonthetalus,itslowangleofplantardeclinationofthetalarhead,itslower orientationofthecalcanealsustentaculumtali,anditsgracilecalcanealtuber.DescriptionH.naledi exhibitsanatomicalfeaturessharedwith Australopithecus ,otherfeaturessharedwith Homo , withseveralfeaturesnototherwiseknowninanyhomininspecies.Thisanatomicalmosaicisreflected indifferentregionsoftheskeleton.Themorphologyofthecranium,mandible,anddentitionismostly consistentwiththegenus Homo ,butthebrainsizeof H.naledi iswithintherangeof Australopithecus . Thelowerlimbislargely Homo -like,andthefootandankleareparticularlyhumanintheir configuration,butthepelvisappearstobeflaredmarkedlylikethatof Au.afarensis .Thewrists, Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 17of35 Researcharticle Genomicsandevolutionarybiology

PAGE 18

fingertips,andproportionsofthefingersaresharedmainlywith Homo ,buttheproximaland intermediatemanualphalangesaremarkedlycurved,eventoagreaterdegreethaninany Australopithecus .Theshouldersareconfiguredlargelylikethoseofaustralopiths.Thevertebraeare mostsimilartoPleistocenemembersofthegenus Homo ,whereastheribcageiswidedistallylike Au.afarensis . H.naledi hasarangeofbodymasssimilartosmall-bodiedmodernhumanpopulations,andis similarinestimatedstaturetobothsmall-bodiedhumansandthelargestknownaustralopiths.We estimatedbodymassfromeightfemoralspecimensforwhichsubtrochantericdiameterscanbe measured(‘Materialsandmethods’),withresultsrangingbetween39.7kgand55.8kg( Table3 ).No femurspecimenissufficientlycompletetomeasurefemurlengthaccurately,buttheU.W.101-484 tibiapreservesnearlyitscompletelength,allowingatibialengthestimateof325mm( Figure10 ). EstimatesforthestatureofthisindividualbasedonAfricanhumanpopulationsamplesrangebetween 144.5and147.8mm.Again,thisstatureestimateissimilartosmall-bodiedmodernhuman populations.ItiswithintherangeestimatedforDmanisipostcranialelements( Lordkipanidzeetal., 2007 ),andslightlysmallerthanestimatedforearly Homo femoralspecimensKNM-ER1472and KNM-ER1481.Somelargeaustralopithsalsohadlongtibiaeandpresumablycomparablytallstatures, asevidencedbytheKSD-VP1/1skeletonfromWoranso-Mille( Haile-Selassieetal.,2010 ). Theendocranialvolumeofall H.naledi specimensisclearlysmallcomparedtomostknown examplesof Homo .Wecombinedinformationfromthemostcompletecranialvaultspecimensto arriveatanestimateofendocranialvolumeforbothlarger(presumablymale)andsmaller(presumably female)individuals(largercompositedepictedin Figure11 ).Theresultingestimatesofapproximately 560ccand465cc,respectively,overlapentirelywiththerangeofendocranialvolumesknownfor australopiths.Withinthegenus Homo ,onlythesmallestspecimensof H.habilis, onesingle H.erectus specimen,and H.floresiensis overlapwiththesevalues. Despiteitssmallvaultsize,thecraniumof H.naledi isstructurallysimilartothoseofearly Homo . Frontalbossingisevident,asisamarkeddegreeofparietalbossing.Thereisnoindicationofmetopic keeling,thoughthereisslightparasagittalkeelingbetweenbregmaandlambda,andsome prelambdoidalflattening.Thecranialvaultbonesaregenerallythin,becomingsomewhatthickerin theoccipitalregion.Thesupraorbitaltorusiswelldeveloped,thoughweaklyarched,andisbounded posteriorlybyawell-developedsupratoralsulcus.Thelateralcornersofthesupraorbitaltorusare roundedandrelativelythin.Thetemporallinesarewidelyspaced,andthereisnoindicationof asagittalcrestortemporal/nuchalcresting.Thetemporalcrestispositionedontheposterioraspect ofthelateralsupraorbitaltorus,ratherthanonthesuperioraspectasinaustralopiths.Atthe posteroinferiorextentofthetemporallines,theycurveanteroinferiorlypresentingawell-developed angulartorus.Thecraniahaveapentagonaloutlineinposteriorview,withthegreatestvaultbreadth locatedinthesupramastoidregion.Thenuchalregionexhibitssexuallydimorphicdevelopmentof nuchalmusclemarkingsandtheexternaloccipitalprotuberance,andthereisaclearindicationof Table3 .Dinaledibodymassestimatesfromfemurspecimenspreservingsubtrochantericdiameters SpecimenIDSide APsubtrochanteric breadth MLsubtrochanteric breadthMass(a)Mass(b)U.W.101-002R18.523.640.044.7 U.W.101-003R21.631.454.555.8 U.W.101-018R18.123.839.744.4 U.W.101-226L19.124.041.345.7 U.W.101-1136R16.925.539.744.4 U.W.101-1391R18.823.940.845.3 U.W.101-1475L18.829.046.549.7 U.W.101-1482L20.728.949.752.1 Regressionequationsdescribedin‘Materialsandmethods’.Mass(a)basedonforensicstaturesfromEuropean individuals.Mass(b)basedonmultiplepopulationsample.Thetwoestimatesdivergesomewhatforsmallerfemora. DOI:10.7554/eLife.09560.014 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 18of35 Researcharticle Genomicsandevolutionarybiology

PAGE 19

atuberculumlinearuminadditiontotheexternaloccipitalprotuberance.Insuperiorviewthevault tapersfromposteriortoanterior,thoughpost-orbitalconstrictionisslight.Thesquamosalsutureis lowandgentlycurved,andparietalstriaearewelldefined.Thelateralmarginsoftheorbitsface laterally.Asmallzygomaticofacialforamenistypicallypresentnearthecenterofthezygomaticbone. Therootofthezygomaticprocessofthemaxillaisanteriorlypositioned,attheleveloftheP3orthe P4.Thereisnoindicationofazygomaticprominence,andthezygomaticarchesdonotflarelaterally toanyextent.Therootofthezygomaticprocessofthetemporalisangleddownwardsapproximately 30relativetoFH.Therootofthezygomaticprocessofthetemporalbeginstolaterallyexpandabove thelevelofthemandibularfossa,ratherthanabovetheleveloftheEAMasinaustralopiths.The mandibularfossaissomewhatlarge,andmoderatelydeep.Thearticulareminenceofthemandibular fossaissaddle-shaped,andorientedposteroinferiorly.Almosttheentiremandibularfossais positionedmedialtothetemporalsquama.Theentoglenoidprocessiselongatedandfacesprimarily laterally.Thepostglenoidprocessissmallandcloselyappressedtothetympanic,formingpartofthe posteriorwallofthefossa.Thepetrotympanicisdistinctlycoronallyoriented.Thevaginalprocessis smallbutdistinct.Thecristapetrosaisweaklydevelopedandnotnotablysharpened.Thereisastrong Eustachianprocess.Theexternalauditorymeatusissmall,oval-shaped,andobliquelyoriented,and adistinctsuprameatalspineispresent.Themastoidregionisslightlylaterallyinflated.Themastoid processistriangularincross-section,witharoundedapexandamastoidcrest.Thedigastricgrooveis deepandnarrow,alongsideamarkedjuxtamastoideminence.Thecaninejugaareweaklydeveloped andthereisnoindicationthatanteriorpillarswouldhavebeenpresent.Ashallow,ill-definedcanine fossaisindicated.Thenasoalveolarclivusisflatandsquare-shaped.Theparabolic-shapedpalateis broadandanteriorlyshallow,becomingdeeperposteriorly. Themandibulardentitionof H.naledi isarrangedinaparabolicarch.Thealveolarandbasalmargins ofthecorpusdivergeslightly.Asingle,posteriorlyopeningmentalforamenispositionedslightlyabove Figure10 .Maximumtibialengthin H.naledi andotherhominins.MaximumtibialengthforU.W.101-484, comparedtoothernearlycompletehominintibiaspecimens. Australopithecusafarensis representedbyA.L.288-1 andKSD-VP-1/1( Haile-Selassieetal.,2010 ); Homoerectus representedbyD3901fromDmanisiandKNM-WT 15000; Homohabilis byOH35; Homofloresiensis byLB1andLB8( Brownetal.,2004 ; Morwoodetal.,2005 ). ChimpanzeeandcontemporaryEuropeanancestryhumansfromClevelandMuseumofNaturalHistory( Lee,2001 ); AndamanIslandersfrom Stock(2013) .Verticallinesrepresentsampleranges;barsrepresent1standarddeviation. DOI:10.7554/eLife.09560.015 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 19of35 Researcharticle Genomicsandevolutionarybiology

PAGE 20

themid-corpuslevel,betweenthepositionoftheP3andtheP4.Themandibularcorpusisrelatively gracile,withawell-developedlateralprominencewhosemaximumextentistypicallyattheM2.Aslight supremelateraltorus(ofDart)weaklydelineatestheextramolarsulcusfromthelateralcorpus.The superiorlateraltorusismoderatelydeveloped,runninganteriorlytothementalforamenwhereitturns uptoreachtheP3jugum.Themarginaltorusismoderatelydeveloped,anddefinesamoderate intertoralsulcus.Theposteriorandanteriormarginaltuberclesareindicatedonlyasslightroughenings ofbone.Thegracilemandibularsymphysisisverticallyoriented.Awell-developedmental protuberanceandweaklateraltuberclesaredelineatedbyaslightmandibularincisure,thereby presentingaweakmentumosseum.Thepost-incisiveplanumissteeplyinclinedandnot-shelf-like. Thereisnosuperiortransversetorus,whileaweak,basallyorientedinferiortransversetorusispresent. Theanteriorandposteriorsubalveolarfossaearecontinuousanddeep,overhungbyawell-developed alveolarprominence.Theextramolarsulcusismoderatelywide.Therootoftheramusofthemandible originateshighonthecorpusattheleveloftheM2.Strongectoangulartuberositiesareindicated. Alargemandibularforamenispresent,withadiffuselydefinedmylohyoidgroove. Liketheskull,thedentitionof H.naledi comparesmostfavorablytoearly Homo samples.Yet comparedtosamplesof H.habilis , H.rudolfensis ,and H.erectus ,theteethof H.naledi arecomparatively quitesmall,similarindimensionstomuchlatersamplesof Homo .Withbothsmallpost-canineteethand asmallendocranialvolume, H.naledi joins Au.sediba and H.floresiensis inanareadistinctfromthe generalhomininrelationofsmallerpost-ca nineteethinspecieswithlargerbrains( Figure12 ). Incomparisonto H.habilis , H.rudolfensis ,and H.erectus ,theteethof H.naledi arenotonlysmall, butalsomarkedlysimpleincrownmorphology.Maxillaryandmandibularmolarslackextensive crenulation,secondaryfissuresandsupernumerarycusps.TheM1hasanequal-sizedmetaconeand paracone,andhasaslightexpressionofCarabelliÂ’straitrepresentedbyasmallcusporshallowpit. I1exhibitsslightocclusalcurvaturewithtracemarginalridgesandvariablysmall tuberculumdentale . I2exhibitsgreaterocclusalcurvatureand tuberculumdentale expressionbutneitherupperincisorhas doubleshovellingorinterruptiongroove.Themandibularcaninesof H.naledi haveasmallocclusal area,andhaveadistalmarginalcuspuleasatopographicallydistinctexpressionofthecingular margin.TheP3isdouble-rooted,fullybicuspidwithmetaconidandprotoconidofapproximately equalheightandocclusalareaseparatedbyadistinctlongitudinalgroove,hasadistallyextensive talonid,andanocclusaloutlineapproximatelysymmetricalwithrespecttothemesiodistalaxis. P4likewisehasadistallyextensivetalonidandapproximatelysymmetricalocclusaloutline( Figure5 ). M1andM2lackcusp6andcusp7,exceptforveryslightexpressioninasmallfractionofspecimens, andhaveaveryfaintsubverticaldepressionratherthanadistinctorextensiveprotostylid.Like australopithsandsomeearly Homo specimens, H.naledi hasanincreasingmolarsizegradientinthe mandibulardentition(M1 < M2 < M3). Thelowerlimbof H.naledi isdefinednotonlybyauniquecombinationofprimitiveandderived traits,butalsobythepresenceofuniquefeaturesinthefemurandtibia.Likeallotherbipedal Figure11 .VirtualreconstructionoftheendocraniumofthelargercompositecraniumfromDH1andDH2overlaid withtheectocranialsurfaces.( A )Lateralview.( B )Superiorview.Theresultingestimateofendocranialvolumeis560cc. Scalebar = 10cm. DOI:10.7554/eLife.09560.016 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 20of35 Researcharticle Genomicsandevolutionarybiology

PAGE 21

hominins, H.naledi possessesavalguskneeandvarusankle.Thefemoralneckislong,anteverted,and anteroposteriorlycompressed.Muscleinsertionsforthe M.gluteusmaximus arestrongandthefemur hasawell-markedlineaasperawithpilastervariablypresent.Thepatellaisrelativelyanteroposteriorly thick.Thetibiaismediolaterallycompressedwitharoundedanteriorborder,alargeproximal attachmentforthe M.tibialisposterior ,andathinmedialmalleolus.Thefibulaisgracilewithlaterally orientedlateralmalleolus,arelativelycircularneckandaconvexsurfacefortheproximalattachmentof the M.peroneuslongus .Uniquefeaturesinthelowerlimbof H.naledi includeadepressioninthe superioraspectofthefemoralneckthatresultsintwomediolaterallyorientedpillarsinferoposteriorly andsuperoanteriorly,andastrongdistalattachmentofthepesanserinusonthetibia. Thefootandankleof H.naledi arelargelyhumanlike( Figure9 ).Thetibiastandsorthogonally uponthetalus,whichismoderatelywedged,withamediolaterallyflattrochleahavingmedialand lateralmarginsatevenheight,aformdis tinctfromthestrongkeelingseeninOH8( H.habilis )and severaltalifromKoobiFora.Thetalarheadandneckexhibitstrong,humanliketorsion;the horizontalangleishigherthaninmosthumans,si milartothatfoundinaustralopiths.Thecalcaneus isonlymoderatelyrobust,butpossessestheplan tardeclinationofther etrotrochleareminence andplantarlypositionedlateralplant arprocessfoundinbothmodernhumansand Au.afarensis . Theperonealtrochleaissmall,unlikethatfo undinaustralopithsandsimilaronlytothatin H.sapiens andNeanderthals.Thetalonavicular,subta larjointsandcalcaneocuboidjointsare humanlikeinpossessingminimalrangesofmotion andevidenceforalocking,rigidmidfoot.The intermediateandlateralcuneifo rmsareproximodistallyelongated.Thehallucaltarsometatarsal jointisflatandproximodistallyalignedindicatingthat H.naledi possessedanadducted,nongraspinghallux.Theheadofthefirstmetatarsalism ediolaterallyexpandeddorsally,indicativeof Figure12 .Brainsizeandtoothsizeinhominins.Thebuccolingualbreadthofthefirstmaxillarymolarisshownhere incomparisontoendocranialvolumeformanyhomininspecies. H.naledi occupiesapositionwithrelativelysmall molarsize(comparabletolater Homo )andrelativelysmallendocranialvolume(comparabletoaustralopiths).The rangeofvariationwithintheDinaledisampleisalsofairlysmall,inparticularincomparisontotheextensiverangeof variationwithinthe H.erectussensulato .Verticallinesrepresenttherangeofendocranialvolumeestimatesknown foreachtaxon;eachverticallinemeetsthehorizontallinerepresentingM1BLdiameteratthemeanforeachtaxon. Rangesareillustratedhereinsteadofdatapointsbecausetherangesofendocranialvolumeinseveralspeciesare establishedbyspecimensthatdonotpreservefirstmaxillarymolars. DOI:10.7554/eLife.09560.017 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 21of35 Researcharticle Genomicsandevolutionarybiology

PAGE 22

ahumanlikewindlassmechanism.Thefootpossesseshumanlikemetatarsallengths,head proportions,andtorsion. Thephalangesaremoderatelycurved,slightlymoresothanin H.sapiens .Theonlyprimitive anatomiesfoundinthefootof H.naledi arethetalarheadandneckdeclinationandsustentaculumtali angles,suggestiveofalowerarchedfootwithamoreplantarlypositionedandhorizontallyinclined medialcolumnthantypicallyfoundinmodernhumans( Harcourt-Smithetal.,2015 ). Theaxialskeletonpresentsacombinationofderived(mainlyaspectsofthevertebrae)and seeminglyprimitive(mainlytheribs)traits.ThepreservedadultT10andT11vertebraeare proportionedsimilarlytoPleistocene Homo ,withtransverseprocessmorphologymostsimilarto Neandertals.Theneuralcanalsofthesevertebraearelargeincomparisontothediminutiveoverall sizeofthevertebrae,proportionallyrecallingDmanisi H.erectus ,Neandertals,andmodernhumans. The11thribisstraight(uncurved),similarto Au.afarensis ,andtheshapeoftheupperribcage appearsnarrow,asassessedfromfirstandsecondribfragments,suggestingthatthethoraxwas pyramidalinshape.The12thribpresentsarobustshaftcross-sectionmostsimilartoNeandertals. Thiscombinationisnotfoundinotherhomininsandmightreflectallometricscalingatasmall trunksize. TheDinalediiliacbladeisflaredandshortenedanteroposteriorly,resembling Au.afarensis or Au.africanus. Theischiumisshortwithanarrowtuberoacetabularsulcus,andtheischiopubicand iliopubicramiarethick,resembling Au.sediba and H.erectus .Thiscombinationofiliacand ischiopubicfeatureshasnotbeenfoundinotherfossilhominins( Figure13 ). Theshoulderof H.naledi isconfiguredwiththescapulasituatedhighandlateralonthethorax, shortclavicles,andlittleornotorsionofthehumerus.Thehumerusisnotablyslenderforitslength, withprominentgreaterandlessertuberclesboundingadeepbicipitalgroove,withasmall,nonprojectinghumeraldeltoidtuberosityandbrachioradialiscrest.Distally,thehumerushasawide lateraldistodorsalpillarandnarrowmedialdistodorsalpillar,andamedially-displacedolecranon fossawithseptalaperture.TheDinalediradiusandulnadiaphysesexhibitlittlecurvature.Theradius hasaglobularradialtuberosity,prominentpronatorquadratuscrest,andreducedstyloidprocess. Thehandsharesmanyderivedfeaturesof modernhumansandNeandertalsinthethumb, wrist,andpalm,buthasrelativelylongand markedlycurvedfingers( Kivelletal.,2015 ).The thumbislongrelativetothelengthofthe otherdigits,andincludesarobustmetacarpalwith well-developedintrinsic( M.opponenspollicis and M.firstdorsalinterosseous )muscleattachments ( Figure6 ).Thepollicaldistalphalanxislargeand robustwithawell-developedridgealongthedistal borderofadeepproximalpalmarfossaforthe attachmentof flexorpollicislongus tendon.Ungual spinesalsoprojectproximopalmarlyfromaradioulnarlyexpandedapicaltuftwithadistinctarea fortheungualfossa.Thewristincludesabootshapedtrapezoidwithanexpandednon-articular palmarsurface,anenlargedandpalmarly-expanded trapezoid-capitatejoint,andatrapezium-scaphoid jointthatextendsfurtherontothescaphoid tubercle.Overall,carpalshapesandarticular configurationsareverysimilartothoseofmodern humansandNeandertals,andunlikethoseof greatapesandotherextincthominins.However, the H.naledi wristlacksathirdmetacarpalstyloid process,hasamoreradioulnarlyorientedcapitateMc2joint,andhasarelativelysmalltrapeziumMc1jointcomparedtohumansandNeandertals. Moreover,thephalangesarelong(relativetothe palm)andmorecurvedthanmostaustralopiths. Figure13 .Selectedpelvicspecimensof H.naledi .U.W. 101-1100iliumin( A )lateralviewshowingaweakiliac pillarrelativelyneartheanterioredgeoftheilium,with nocristaltubercledevelopment;( B )anteriorview, angledtodemonstratethedegreeofflare,whichisclear incomparisontothesubarcuatesurface.U.W.101-723 immaturesacrumin( C )anteriorview;and( D )superior view.U.W.101-1112ischiumin( E )lateralview;and( F ) anteriorview,demonstratingrelativelyshorttuberacetabulardiameter.Scalebar = 2cm. DOI:10.7554/eLife.09560.018 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 22of35 Researcharticle Genomicsandevolutionarybiology

PAGE 23

DiscussionTheoverallmorphologyof H.naledi placesitwithinthegenus Homo ratherthan Australopithecus or otherearlyhominingenera.Thesharedderivedfeaturesthatconnect H.naledi withothermembersof Homo occupymostregionsofthe H.naledi skeletonandrepresentdistinctfunctionalsystems, includinglocomotion,manipulation,andmastication.Locomotortraitssharedwith Homo includethe absolutelylonglowerlimb,withwell-markedlineaaspera,strong M.gluteusmaximus insertions, gracilefibulaandgenerallyhumanlikeankleandfoot.Theseaspectsofthelowerlimbsuggest enhancedlocomotorperformanceforastridinggait.The H.naledi handsharesaspectsof Homo morphologyinthewrist,thumbandpalm,pointingtoenhancedobjectmanipulationabilityrelativeto australopiths,including Au.sediba ( Kivelletal.,2011 ; Kivelletal.,2015 ). H.naledi lacksthe powerfulmasticationthattypifies Australopithecus and Paranthropus ,withgenerallysmallteeth acrossthedentition,gracilemandibularcorpusandsymphysis,laterally-positionedtemporallines, slightpostorbitalconstrictionandnon-flaringzygomaticarches.Theupperlimb,shoulderandribcage haveamoreprimitivemorphologicalpattern,butdonotprecludeaffiliating H.naledi with Homo , particularlyconsideringthatpostcranialremainsof H.habilis appeartoreflectanaustralopith-like bodyplan( Johansonetal.,1986 ).Locomotor,manipulatory,andmasticatorysystemshaveboth historicalandcurrentimportanceindefining Homo ( WoodandCollard,1999 ; Holliday,2012 ; Ant ´ on etal.,2014 ),and H.naledi fitswithinourgenusintheserespects. Thestructuralconfigurationofthe H.naledi cranium,beyondthefunctionalaspectsof mastication,islikewisesharedwith Homo .Asinmanyspecimensof H.erectus and H.habilis ,the H.naledi vaultincludesawell-developedandmoderatelyarchedsupraorbitaltorus,markedfrom thefrontalsquamabyacontinuoussupratoralsulcus,frontalbossing.Further,asinmany H.erectus crania, H.naledi exhibitsamarkedangulartorusandoccipitaltorus.The H.naledi faceincludesaflat andsquarednasoalveolarclivus,comparableto H.rudolfensis ( Leakeyetal.,2012 ),andweak caninefossae.Whileitsanatomyplacesitunambiguouslywithin Homo ,the H.naledi craniumand dentitionlackmanyderivedfeaturessharedbyMPandLP Homo and H.sapiens. Theaustralopith-likefeaturesofthepostcranium,includingtheribcage,shoulder,proximalfemur,andrelativelylong, curvedfingers,alsodepartsharplyfromthemorphologypresentinMPhumansand H.sapiens . Thesimilaritiesof H.naledi toearliermembersof Homo ,including H.habilis , H.rudolfensis ,and H.erectus ,suggestthatthisspeciesmayberootedwithintheinitialoriginanddiversificationofour genus. Thefossilrecordofearly Homo and Homolikeaustralopithshasrapidlyincreasedduringthelast15 years,andthisaccumulatingevidencehaschangedourperspectiveontheriseofourgenus.Many skeletalandbehavioralfeaturesobservedtoseparatelater Homo fromearlierhomininswereformerly arguedtohavearisenasasingleadaptivepackage,includingincreasedbrainsize,toolmanipulation, increasedbodysize,smallerdentition,andgreatercommitmenttoterrestriallong-distancewalkingor running( WoodandCollard,1999 ; Hawksetal.,2000 ).Butwenowrecognizethatsuchfeatures appearedindifferentcombinationsindifferentfossilsamples( Ant ´ onetal.,2014 ).TheDmanisi postcranialsample( Lordkipanidzeetal.,2007 )andadditionalcranialremainsof H.erectus from Dmanisi( Gabuniaetal.,2000 ; Vekuaetal.,2002 ; Lordkipanidzeetal.,2013 )andEastAfrica( Spoor etal.,2007 ; Leakeyetal.,2012 ),demonstratethatlargerbrainsizeandbodysizedidnotarise synchronouslywithimprovedlocomotorefficiencyandadaptationstolong-distancewalkingorrunning in H.erectus ( Holliday,2012 ; Ant ´ onetal.,2014 ).Further,thediscoveryof Au.sediba showedthat amosaicof Homo -likehand,pelvisandaspectsofcraniodentalmorphologycanoccurwithinaspecies withprimitivebodysize,limbproportions,lowerlimbandfootmorphology,thoraxshape,vertebral morphology,andbrainsize( Bergeretal.,2010 ; Carlsonetal.,2011 ; Kivelletal.,2011 ; Churchill etal.,2013 ; DeSilvaetal.,2013 ; Schmidetal.,2013 ). H.naledi presentsyetadifferentcombination oftraits.Thisspeciescombinesahumanlikebodysizeandstaturewithanaustralopith-sizedbrain; featuresoftheshoulderandhandapparentlywell-suitedforclimbingwithhumanlikehandandwrist adaptationsformanipulation;australopith-likehipmechanicswithhumanliketerrestrialadaptationsof thefootandlowerlimb;smalldentitionwithprimitivedentalproportions.Inlightofthisevidencefrom completeskeletalsamples,wemustabandontheexpectationthatanysmallfragmentoftheanatomy canprovidesingularinsightabouttheevolutionaryrelationshipsoffossilhominins. Arecentphylogeneticanalysisoffossilhomininsbasedoncraniodentalmorphologyplaced Au.sediba atthebaseofthegenus Homo ( Demboetal.,2015 ),inagreementwithearlieranalysesof Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 23of35 Researcharticle Genomicsandevolutionarybiology

PAGE 24

thisspecies( Bergeretal.,2010 ).Thecranialanddentalaffinitiesidentifiedbetween Au.sediba and Homo includemanyfeaturessharedby H.naledi. But H.naledi and Au.sediba sharedifferent postcranialfeatureswithotherspeciesof Homo .Resolvingthephylogeneticplacementof H.naledi willrequirebothpostcranialandcraniodentalevidencetobeintegratedtogether.Suchintegration posesachallengebecauseofthepoorrepresentationofseveralkeyspeciesbothwithinandoutside of Homo ,mostnotably H.habilis ,forwhichpostcranialevidenceisslight,and H.rudolfensis forwhich noassociatedpostcranialremainsareknown.Weproposethetestablehypothesisthatthecommon ancestorof H.naledi , H.erectus ,and H.sapiens sharedhumanlikemanipulatorycapabilitiesand terrestrialbipedality,withhandsandfeetlike H.naledi ,anaustralopith-likepelvisandthe H.erectuslikeaspectsofcranialmorphologythatarefoundin H.naledi .Enlargedbrainsizewasevidentlynot anecessaryprerequisiteforthegenerallyhuman-likeaspectsofmanipulatory,locomotor,and masticatorymorphologyof H.naledi . Althoughitcontainsanunprecedentedwealthofanatomicalinformation,theDinaledideposit remainsundated( Dirksetal.,2015 ).Consideringthat H.naledi isamorphologicallyprimitivespecies withinourgenus,anagemayhelpelucidatetheecologicalcircumstanceswithinwhich Homo arose anddiversified.Ifthefossilsprovetobesubstantiallyolderthan2millionyears, H.naledi wouldbethe earliestexampleofourgenusthatismorethanasingleisolatedfragment.Thesamplewouldillustrate amodelfortherelationofadaptivefeaturesofthecranium,dentitionandpostcraniumduring acriticaltimeintervalthatisunderrepresentedbyfossilevidenceofcomparablecompleteness.A dateyoungerthan1millionyearsagowoulddemonstratethecoexistenceofmultiple Homo morphs inAfrica,includingthissmall-brainedform,intothelaterperiodsofhumanevolution.Thepersistence ofsuchaspecieswithclearadaptationsformanipulationandgrip,alongsideMPhumansorperhaps evenalongsidemodernhumans,wouldchallengemanyassumptionsaboutthedevelopmentofthe archaeologicalrecordinAfrica. Thedepthofevidenceof H.naledi mayprovideaperspectiveonthevariationtobeexpected withinfossilhominintaxa( Lordkipanidzeetal.,2013 ; Berm ´ udezdeCastroetal.,2014 ).Theentire Dinaledicollectionisremarkablyhomogeneous.Thereisverylittlesizevariationamongadult elementswithinthecollection.Eightbodymassestimatesfromthefemur( Table2 )haveastandard deviationofonly4.3kilograms,forabodymasscoefficientofvariation(CV)ofonly9%.TheCVof bodymasswithinmosthumanpopulationsissubstantiallyhigherthanthis,withanaveragenear15% ( McKellarandHendry,2009 ).Likewise,thesizevariationofcranialanddentalelementsisminimal. With11mandibularfirstmolars,theCVofbuccolingualbreadthisonly3.2%andfor13maxillaryfirst molarstheCVofbuccolingualbreadthisonly2.0%(buccolingualbreadthisusedbecauseitisnot subjecttovariancefrominterproximalwear).Notonlysize,butalsoanatomicalshapeandformare homogeneouswithinthesample.Almosteveryaspectofthemorphologyofthedentition,including thedistinctiveformofthelowerpremolars,thedistalaccessorycuspuleofthemandibularcanines, andtheexpressionofnonmetricfeaturesthatnormallyvaryinhumanpopulations,isuniforminevery specimenfromthecollection.Thedistinctiveaspectsofcranialmorphologyarerepeatedinevery specimen,andeventheaspectsthatnormallyvaryamongindividualsofdifferentbodysizeor betweensexesexhibitonlyslightvariationamongtheDinalediremains.Oneofthemostunique aspectsof H.naledi isthemorphologyofthefirstmetacarpal;thederivedaspectsofthisanatomyare presentineveryoneofsevenfirstmetacarpalspecimensinthecollection( Figure14 ).Unlikeanyother fossilhomininsiteinAfrica,theDinalediChamberseemstopreservealargenumberofindividuals fromasinglepopulation,onewithvariationequaltoorlessthanthatfoundwithinlocalpopulationsof modernhumans. TheDinaledicollectionistherichestassemblageofassociatedfossilhomininseverdiscoveredin Africa,andasidefromtheSimadelosHuesoscollectionandlaterNeanderthalandmodernhuman samples,ithasthemostcomprehensiverepresentationofskeletalelementsacrossthelifespan,and frommultipleindividuals,inthehomininfossilrecord.Theabundanceofevidencefromthis assemblagesupportsouremergingunderstandingthatthegenusHomo encompassedavarietyof evolutionaryexperiments( Ant ´ onetal.,2014 ),withdiversitynowevidentforfossil Homo ineachofthe fewintensivelyexploredpartsofAfrica( Leakeyetal.,2012 ).Butasmuchasitadvancesour knowledge, H.naledi alsohighlightsourignoranceaboutancient Homo acrossthevastgeographic spanoftheAfricancontinent.Thetreeof Homo -likehomininsisfarfromcomplete:wehavemissedkey transitionalformsandlineagesthatpersistedforhundredsofthousandsofyears.Withanincreasing paceofdiscoveryfromthefieldandthelaboratory,morelightwillbethrownontheoriginofhumans. Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 24of35 Researcharticle Genomicsandevolutionarybiology

PAGE 25

MaterialsandmethodsComparativehomininspecimensexaminedinthisstudyInthedifferentialdiagnosisof H.naledi ,wehavecomparedtheholotypeDH1,paratypes,andother referredmaterialtofossilevidencefrompreviously-identifiedhominintaxa.Ourgoalistoprovide adiagnosisfor H.naledi thatisclearinreferencetowidelyrecognizedhomininhypodigms.Different specialistscontinuetodisagreeaboutthecompositionandanatomicalbreadthrepresentedbythese hominintaxaandattributionofparticularspecimenstothem(seee.g., WoodandCollard,1999 ; Lordkipanidzeetal.,2013 ; Ant ´ onetal.,2014 onearly Homo taxa).Wedonotintendtotakeany positiononsuchdisagreementsbyourselectionofcomparativesamplesfor H.naledi . Wehavebeencautiousinourattributionofpostcrani alspecimenstohominintaxa,particularlyinthe AfricanPlio-Pleistocene,whereithasbeendemonstrate dmultiplehominintaxacoexistedintime,ifnotin geographicalspace.Becausethepurposeofthisstudyisdifferentialdiagnosisinreferencetoknowntaxa, unattributedspecimensarenotgermane,althoughince rtaincasestherearewell-acceptedattributionsto genusforspecimens(e.g., Homo sp . or Australopithecus sp . )ascitedbelow.Wehaveincludedsome specimensincomparisonsbecausetheyarerelativ elycomplete,eveniftheycannotbeattributedto aspecies,becausefewhominintaxaarerepresent edbyevidenceacrosstheentireskeleton.Forsome anatomicalcharacters,partsarepreservedonlyforM Porlaterhomininsamples,sowehaveincludedsuch comparisonstomakeclearhow H.naledi comparesintheseelementstothe(few)knownfossilexamples. Thisstudyreliesuponobservationsandmeas urementstakenfromoriginalfossilsbythe authors,observationstakenfromcasts,andobservationstakenfromtheliterature.These observationsareinlargepartstandardanat omicalpractice;wherefeaturesarespecially describedinpreviousstudieswehavereferenced thosehere.Forthisstudy,acastcollectionwas assembledincludingthePhillipV.Tobiasre searchcollectionattheUniversityofthe Witwatersrandandloansofcastmaterialsfromth eUniversityofWisconsin–Madison,University ofMichigan,AmericanMuseumofNaturalHistory, NewYorkUniversity,UniversityofColorado– Denver,UniversityofDelaware,TexasA&MUniv ersity,andthepersonalcollectionsofPeter Schmid,MilfordWolpoffandRobBlumenschine.We extendourgratitudetothecuratorsoffossil Figure14 .Firstmetacarpalsof H.naledi .SevenfirstmetacarpalshavebeenrecoveredfromtheDinalediChamber.U.W.101-1321istherightfirst metacarpaloftheassociatedHand1foundinarticulation.U.W.101-1282andU.W.101-1641areanatomicallysimilarleftandrightfirstmetacarpals ,which wehypothesizeasantimeres,bothwererecoveredfromexcavation.U.W.101-007wascollectedfromthesurfaceofthechamber,andexhibitsthesame distinctivemorphologicalcharacteristicsasallthefirstmetacarpalsintheassemblage.Alloftheseshowamarkedrobusticityofthedistalhalfo fthebone, averynarrow,‘waisted’appearancetotheproximalshaftandproximalarticularsurface,prominentcrestsforattachmentof M.opponenspollicis and M.firstdorsalinterosseous ,andaprominentridgerunningdownthepalmaraspectofthebone.Theheadsofthesemetacarpalsaredorsopalmarlyflat andstronglyasymmetric,withanenlargedpalmar-radialprotuberance.Thesedistinctivefeaturesarepresentamongallthefirstmetacarpalsinth e Dinaledicollection,andareabsentfromanyotherhomininsample.Theirderivednatureisevidentincomparisontoapesandotherearlyhominins,her e illustratedwithachimpanzeefirstmetacarpalandtheMH2firstmetacarpalof Australopithecussediba . DOI:10.7554/eLife.09560.004 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 25of35 Researcharticle Genomicsandevolutionarybiology

PAGE 26

collectionsandthegenerosityoftheseinstitutionsi nfacilitatingthisresearch,bothinSouthAfrica andthroughouttheworld. Thislistofskeletalmaterialsextendsthelistofcraniodentalcomparativematerialusedin diagnosing Au.sediba ,withmanyofthehypodigmsidenticaltothatstudy( Bergeretal.,2010 ). Wherewehavehadfirst-handaccesstooriginalspecimens,werelyuponourownobservations;we thereforedonotreferreaderstoothersourcesforthesedata.AustralopithecusafarensisThesamplesattributedto Au.afarensis fromHadar,Laetoli,theMiddleAwash,Woranso-Milleand Dikikawereutilized.Forthistaxonwerelieduponpublishedreports( Johansonetal.,1982 ; Kimbel etal.,2004 ; Drapeauetal.,2005 ; Alemsegedetal.,2006 ; Haile-Selassieetal.,2010 ; Wardetal., 2012 ),inadditiontoourownobservationsonoriginalfossilsandcasts.AustralopithecusafricanusThesamplesattributedto Au.africanus fromTaung,SterkfonteinandMakapansgatwereemployed. Originalspecimenswereexaminedfirst-handbytheauthors.AustralopithecusgarhiThecraniumBOU-VP-12/130fromBouriwasincluded,withdatatakenfromapublishedreport ( Asfawetal.,1999 ).AustralopithecussedibaThepartialskeletonsMH1andMH2fromMalapa,SouthAfricawereincludedinthisstudy,basedon examinationoftheoriginalspecimensbytheauthors.ParanthropusaethiopicusThecraniumKNM-WT17000wasexaminedfirst-handforthisstudy.ParanthropusboiseiSamplesfromtheOmoShungurasequence,EastLakeTurkana,OlduvaiGorgeandKonsowere includedinthisstudy.OriginalspecimensfromOlduvaiGorgeandEastLakeTurkanawereexamined first-hand,whilecastsandpublishedreports( Tobias,1967 ; Suwaetal.,1996 , 1997 ; Dom ´ nguezRodrigoetal.,2013 )wereusedtostudytheOmoandKonsomaterials.Ourpostcranial considerationsof P.boisei areverylimitedandwedidnotrelyupontheassociationofKNM-ER 1500( Grauszetal.,1988 )toderiveinformationaboutthepostcranialskeletonof P.boisei .ParanthropusrobustusThesamplesfromKromdraai,Swartkrans,Sterkfont ein,Drimolen,Gondolin,andCooperswereincluded inthisstudy.First-handobservationsoforiginalspec imensfromalllocalitieswereusedwiththeexception ofDrimolenfossils,whichwerecomparedusingpublishedreports( Keyser,2000 ; Keyseretal.,2000 ).HomohabilisSamplesfromOlduvaiGorge,EastLakeTurkana,theOmoShungurasequence,Hadar,and Sterkfonteinwereincludedinthisstudy.OriginalOlduvaiGorgeandEastLakeTurkanafossilswere examinedfirst-hand,whilefortheOmoandHadarmaterialswereliedonouroriginalobservationson castsandoriginalsandpublishedreports( BoazandHowell,1977 ; Tobias,1991 ; Kimbeletal., 1997 ).Weincludethefollowingfossilsinthehypodigmof H . habilis :A.L.666-1,KNM-ER1478,KNMER1501,KNM-ER1502,KNM-ER1805,KNM-ER1813,KNM-ER3735,OH4,OH6,OH7,OH8,OH 13,OH15,OH16,OH21,OH24,OH27,OH31,OH35,OH37,OH39,OH42,OH44,OH45,OH 62,OMO-L894-1,andStw53.Werecognizethatsomeauthors(includingsomeoftheauthorsofthis paper)prefertoclassifyOH62,Stw53andA.L.666-1outsideof H.habilis ,(e.g.,as Homo gautengensis whichwedonotrecognizeasvalid),orevenoutsidethegenus Homo ;thesespecimens expandthemorphologicalandtemporalvariabilityencompassedwithin H.habilis.HomorudolfensisSamplesfromOlduvaiGorge,EastLakeTurkana,andLakeMalawiwereincludedinthisstudy.The EastLakeTurkanafossilsavailablepriorto2010wereexaminedfirst-hand,whilefortheOlduvaiand LakeMalawifossilsandKNM-ER60000,62000,and62003wereliedonoriginalobservationson fossilsandcastsaswellaspublishedreports( Schrenketal.,1993 ; Blumenschineetal.,2003 ; Leakeyetal.,2012 ).Weincludethefollowingfossilsinthehypodigmof H . rudolfensis :KNM-ER819, Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 26of35 Researcharticle Genomicsandevolutionarybiology

PAGE 27

KNM-ER1470,KNM-ER1482,KNM-ER1483,KNM-ER1590,KNM-ER1801,KNM-ER1802,KNM-ER 3732,KNM-ER3891,KNM-ER60000,KNM-ER62000,KNM-ER62003,OH65,andUR501.Wedo recognizethatKNM-ER60000andKNM-ER1802presentsomeconflictinganatomythatsome authorshavearguedprecludesthemasconspecificspecimens( Leakeyetal.,2012 );byconsidering both,weaimtobeconservativeastheyencompassmorevariationwithin H.rudolfensis .HomoerectusSamplesfromBuia,Chemeron,Daka,Dmanisi,EastandWestLakeTurkana,Gona,Hexian,Konso, Mojokerto,OlduvaiGorge,Sangiran,Swartkrans,Tri nil,andZhoukoudianwereincludedinthisstudy. SouthAfricanmaterialisofspecialinterestinthisc omparisonbecauseoftheg eographicproximity,and becauseofthedifficultyofclearlyidentifying Homo specimenswithinthelargefossilsamplefrom Swartkrans.Inparticular,thef ollowingspecimensfromSwartkransareconsideredtorepresent H.erectus : SK15,SK18a,SK27,SK43,SK45,SK68,SK847,SK878,SK2635,SKW3114,SKX257/258,SKX267/ 2671,SKX268,SKX269,SKX334,SKX339,SKX610,SKX1756,SKX2354,SKX2355,SKX2356,andSKX 21204.Ithasbeensuggested( Grineetal.,1993 , 1996 )thatSK847andStw53mightrepresentthe sametaxon,andthatthistaxonisacurrentlyundiagnosedspeciesof Homo inSouthAfrica.However,we agreewith Clarke(1977 ; 2008) thatSK847canbeattributedto H . erectus ,andthatStw53cannot. Becausethereisnoclearindicationthatmorethanonespeciesof Homo isrepresentedintheSwartkrans sample,weconsiderallthismaterialtobelongto H.erectus .Weconsidered‘ Homoergaster ’(andalso ‘ Homoaff.erectus ’from Wood,1991 )tobesynonymsof H.erectus forthisstudy;TurkanaBasin specimensthatareattributedto H.erectus thusincludeKNM-ER730,KNM-ER820,KNM-ER992,KNMER1808,KNM-ER3733,KNM-ER3883,KNM-ER42700,KN M-WT15000.OlduvaispecimensincludeOH 9,OH12andOH28.OriginalfossilmaterialsfromC hemeron,LakeTurkana,Swartkrans,Trinil,and Dmanisiwereexaminedfirst-handbytheauthors,whiletheremainderwerebasedoncastsandpublished reports( Weidenreich,1943 ; Wood,1991 ; Ant ´ on,2003 ; Rightmireetal.,2006 ; Suwaetal.,2007 ). AlargenumberofpostcranialspecimenshavebeencollectedfromtheTurkanaBasinandappear consistentwiththeanatomicalrangeotherwisefoundin Homo ,andinconsistentwithknownsamplesof Australopithecus and Paranthropus fromelsewhere.TheseincludeKNM-ER1472,KNM-ER1481,KNMER3228,KNM-ER737,andothers.Wemayaddotherfossilsfromothersiteslackingassociationwith craniodentalmaterial,suchasthepartialBOU-VP12/1skeletonandeventheGonapelvis.These specimensattributableto Homo butnotnecessarilytoaparticularspeciesdidinformourunderstanding ofvariabilitywithinthegenus,butforthemostpartthesespecimensdonotinformourdifferential diagnosisof H.naledi relativetoparticularspecies.Forexample,thekeyelementoffemoral morphologyof H.naledi incontrasttootherspeciesisthepresenceoftwowell-definedmediolaterally runningpillarsinthefemoralneck;theisolatedspecimensofearly Homo donotcontradictthisapparent autapomorphy.Likewise,noisolatedspecimensinformusaboutthehumanlikeaspectsoffoot morphologyin H.naledi .Inthesecases,thelackofassociationsforthisevidenceactuallyisless importantthanthelackofspecimensthatreplicatethedistinctivefeaturesofthe H.naledi morphology.MiddlePleistocene HomoSpecimensfromthelatestLowerPleistoceneandMPofEuropeandAfricathatcannotbeattributed to H.erectus wereincludedinourcomparisons.Theseincludefossilsthathavebeenattributedto H.heidelbergensis , H.rhodesiensis ,‘archaic H.sapiens ’,or‘evolved H.erectus ’byavarietyofother authors.SpecimensattributedtoMP Homo includematerialsfromEliyeSprings,Arago,Atapuerca SimadelosHuesos,Bodo,BrokenHill,CaveofHearths,Ceprano,Dali,Elandsfontein,Jinniushan, Kapthurin,Mauer,Narmada,Ndutu,Petralona,Reilingen-Schwetzingen,Solo,Steinheim,Swanscombe.Thisgroupingincludesthefollowingspecimens:KNM-ES11693,Arago2,Arago13,Arago 21,Atapuerca1,Atapuerca2,Atapuerca4,Atapuerca5,Atapuerca6,CaveofHearths,SAM-PQEH1,Kabwe,Mauer,Ndutu,Sal ´ e,Petralona,Reilingen-Schwetzingen,Steinheim.HomofloresiensisSpecimensfromLiangBua,Floresasdescribedby Brownetal.,2004 ; Morwoodetal.,2005 , Jungersetal.,2009a , Jungersetal.,2009b ,and Falketal.,2005 wereincludedinthisstudy.ScanningandvirtualreconstructionmethodsThecalvariae(DH1-4)werescannedusingaNextEnginelasersurfacescanner(NextEngine,Malibu,CA) atthefollowingsettings:Macro,12divisionswithauto-rotation,HD17kppi.Dependingonthe Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 27of35 Researcharticle Genomicsandevolutionarybiology

PAGE 28

complexityofthesurfacerelief,eithertwoorthree completescanningcycleswerecompletedper specimen,resultinginmultiple360scans.Each individualscanwastrimmed,aligned,andfused (volumemerged)intheaccompanyingScanStudio HDProsoftware.Foreachspecimen,theindividual360scanswerethenalignedandmergedin GeoMagicStudio14.0(RaindropGeomagic,ResearchTrianglePark,NC),creatingafinalthreedimensionalmodelofthespecimen.Giventhe fragmentednatureofthecalvariaespecimens, boththeectocranialandendocranialsurfaces werecapturedinthescans. DH3consistedprimarilyofportionsofthe rightcalvaria.However,asmallsectionofthe frontalandtheparietalcrossedthemid–sagittal plane.Forthisreason,itwaspossibletomirror imagethesurfacescantoapproximatetheleft calvariaandobtainamorecompletevisualization ofthecompletecalvaria( Figure15 ).Thevirtual specimenofDH3wasmirroredinGeoMagic Studio,andmanuallyregistered(aligned)using commonpointsalongthefrontalcrestand sagittalsuture.TheregistrationprocedureinGeoMagicStudioisaniterativeprocessthatrefines thealignmentofspecimenstominimizespatialdifferencesbetweencorrespondingsurfaces.Inthis manner,theprogramisabletomatchthepositionoverlappingsurfaces,inadditiontotheir angulationandcurvature. ThesameprocedureswereusedtomirrorimageandcreateavirtualreconstructionofDH2andthe occipitalportionofDH1( Figure16 ).TheoccipitalandvaultportionsofDH1werereconstructed basedontheanatomicalalignmentofthesagittalsuture,sagittalsulcus,parietalstriae,andthe continuationofthetemporallinesacrossboththespecimens.Virtualreconstructionofcompositecrani aandestimationofcranialcapacityInordertovirtuallyestimatethecranialcapacity,compositecraniawereconstructedfromthesurface scansandmirrorimagedscansofthecalvariae.Twoseparatecompositecraniawerecreated;the relativelysmaller-sizedcalvariae(DH3andDH4)werecombinedintoonecomposite,andthelargersizedcalvariae(DH1andDH2)composedthelargercompositecranium. Figure15 .Posteriorviewofthevirtualreconstructionof DH3.Theresultantmirrorimageisdisplayedinblue. Theantimereswerealignedbythefrontalcrestand sagittalsutureusingtheManualRegistrationfunctionin GeoMagicStudio14.0. DOI:10.7554/eLife.09560.020 Figure16 .Virtualreconstructionof( A )DH2and( B )occipitalportionofDH1.Theactualspecimendisplaysits originalcolorationandthemirrorimagedportionisillustratedinblue. DOI:10.7554/eLife.09560.021 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 28of35 Researcharticle Genomicsandevolutionarybiology

PAGE 29

Thesmallercompositecranium,DH3wasmirrore dinGeoMagicStudio14.0,andmergedwiththe originalscanasoutlinedabove.Th esurfacescanofDH4wasuploadedandregistered(aligned)totheDH3 modelusingoverlappingtemporalfeatures(e.g.,theexternalauditorymeatus).Noscalingwasperformed. DH4wasthenmirrorimagedtocompletetheoccipital contour.Theresultantmodelsuggestsageneral concordancebetweenthespecimensinbothsizeandshapewithaclosealignmentofvaultsurfacesand anatomicalfeaturesbetweenspecimens( Figure17 ). Forthelargercompositecranium,thesurfacemodelofDH2anditsmirrorimagewasthen uploaded,registered(aligned),andmergedwiththemirror-imagedmodelofDH1.Noscalingwas performed.ThecongruencybetweenthespecimensintheresultantmodelsuggeststhatDH1and DH2aresimilarinbothsizeandvaultshape( Figure18 ).VirtualreconstructionofcranialcapacityThecompositemodelofDH3andDH4wasusedtoestimatethecranialcapacityforthesmaller morphotype.InGeoMagicStudio14.0,theendocranialsurfaceofthecompositewascarefully selectedfromtheectocranialsurfaceandcopiedasanewobject.Inordertoobtainavolume calculationthemodelhastobeaclosedsurface,meaningthatalloftheholesinthesurfacemodel hadtobefilled.Smallholesinthemodelwerefilledusingthe‘FillbyCurvature’function.Largerholes werefilledinbysections.Forexample,thecranialbasewasfilledinusinganumberoftransverse sections,sothatintheabsenceofthecranialbasethecontourofthevariouscranialfossaeandthe petrousportionsofthetemporalcouldbepreservedasbestaspossible.Whenappropriate (e.g.,aroundangularportionsofthepetrousbone),smallsectionswerefilledusingaflatholefilling function.Thenewsurfacescreatedbythehole-fillingmechanismwerecarefullymonitoredand repeateduntilanacceptablemodelthatappearedtobestapproximatethemissingportionswas obtained.Theresultisaclosedmodelapproximationoftheendocranium,ofwhichavolumecanbe Figure17 .Postero-lateralviewofthevirtualreconstructionofacompositecraniumfromDH3andDH4.( A )The surfacescanofDH3wasmirrorimagedandmergedasdescribedinSupplementaryNote8.( B )ThescanofDH4was alignedtotheDH3model.( C )DH4wasthenmirrorimagedtocompletetheoccipitalcontour( D ). DOI:10.7554/eLife.09560.022 Figure18 .VirtualreconstructionofacompositecraniumfromDH1andDH2.ThesurfacemodelofDH2(blue), consistingoftheoriginalscanmergedwiththemirrorimage,wasthenuploadedandalignedwiththemirrorimagedDH1model(pink).NotethesimilarityinsizeandshapebetweenDH1andDH2observedintheposterior ( A )anterior( B )lateral( C )andsuperior( D )views. DOI:10.7554/eLife.09560.023 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 29of35 Researcharticle Genomicsandevolutionarybiology

PAGE 30

calculatedbyGeoMagicStudio( Figure19 , Figure20 ).Thevolumeofthesmallercompositecranium (DH3andDH4)indicatesacranialcapacityofapproximately465cm3. Inordertodeterminewhethersignificanterrorswerebeingintroducedinthemannerthatthe cranialbasewasfilledintheaboveprocedures,theendocranialvolumeofDH3/DH4wasalsovirtually calculatedusingthecranialbaseofSts19asamodel.A3DmodelofSts19wasmirroredandaligned totheDH3/DH4modelusingtheexternalauditorymeatusandcommonpointsontheinternalsurface ofthepetrousportionasaguide( Figure21 ).TheSts19modelwasthenscaledby0.97toobtainan optimalfitbetweenthetwomodels. AftertheSts19modelwasmergedwiththeDH3/DH4model,theendocranialsurfacewas extractedandreconstructedasdescribedabove( Figure22 ).Theresultantendocranialvolumeusing theSts19cranialbasewas465.9cm3.Thisvalueisinagreementwiththefirstestimateandsuggests thatusingamodelcranialbasedidnotsignificantlyaltertheresults. Thelargercompositecranium,consistingofDH1andDH2,lacksmostofthefrontalregion.In ordertocreateaclosedendocranialsurfaceforavolumeestimate,thefrontalregionfromthesmaller compositecraniumwasscaledby5%,andthenregistered(aligned)andmergedtothemodelofthe largercompositecranium.Aswiththesmallercompositecranium,theendocranialsurfacewasthen selectedandconvertedtoanewobject,andtheremainingholesfilledbasedonthecurvatureofthe surface.ThevolumeoftheclosedendocranialmodelwascalculatedusingGeoMagicStudio.The cranialcapacity(endocranialvolume)ofthelargercompositemodelisapproximately560cc.BodymassestimationmethodsEightfemoralfragmentsfromtheDinaledicollectionallowadirectmeasurementofthe subtrochantericanteroposteriorandmediolateraldiameters( Table3 ).Wedevelopedtworegression Figure19 .VirtualreconstructionoftheendocraniumofthecompositecraniumfromDH3andDH4.( A )Lateralview. ( B )Superiorview.( C )Inferiorview.Inallviews,anterioristotowardstheleft. DOI:10.7554/eLife.09560.024 Figure20 .VirtualreconstructionoftheendocraniumofthecompositecraniumfromDH3andDH4overlaidwiththe ectocranialsurfaces.( A )Lateralview.( B )Superiorview. DOI:10.7554/eLife.09560.025 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 30of35 Researcharticle Genomicsandevolutionarybiology

PAGE 31

equationstoestimatebodymassfromthesediametersbasedonthemassesofmodernhuman samples.MCEmeasuredbodymassesofasampleof253modernEuropeanindividuals,128males and125females,collectedfromtheInstituteforForensicMedicineinZurich,Switzerland.Body massesweretakenattimeofforensicevaluation.Thissampleyieldsthefollowingregressionequation relatingbodymasstosubtrochantericdiameter,whereFSTprreferstotheproductofthefemoral subtrochantericmediolateralandanteroposteriorbreadths:BodyMass = 0 : 060 × FSTpr + 13 : 856 ; SEE = 6 : 78 ; r = 0 : 50 ; p =< 0 : 001 :Wefurtherexaminedabroadersampleof276modernhumanstakenfromanumberofpopulations aroundtheworld,withdatameasuredbyTWH.Thebodymassesofindividualswereestimatedfrom femurheaddiameter,usingtheaverageofresultsobtainedfrom Grineetal.(1995) and Ruffetal. (1997) .Thesampleincludes115females,155males,and6individualsofindeterminatesex.BodyMass = 0 : 046 × FSTpr + 24 : 614 ; SEE = 5 : 82 ; r = 0 : 82 ; p < 0 : 001 :StatureestimationmethodsWecollecteddatafromskeletalmaterialrepresentingtwoAfricanpopulationsamples.Weuseonly Africanpopulationsinthiscomparisonbecausetheratiooftibialengthtofemurlength,andthereby theproportionofstatureconstitutedbytibialength,variesbetweenhumanpopulationsbothtoday andprehistorically.Althoughwedonotknowthisproportionfor H.naledi ,weadoptthenull hypothesisthattheylikelyhadtibia/femurproportionssimilartootherAfricanpopulationsamples. 95maleandfemaleKulubnartiindividualsfrommedievalNubiaarecuratedattheUniversityof Colorado,Boulder.DatawerecollectedbyHMG,includingestimatesoflivingstaturebasedonthe Fullymethod( Fully,1956 ; Raxteretal.,2006 ),andthesewereusedtodeveloparegression equationrelatingtibialengthtostature.The resultingequationis:Stature = 0 : 295 × TML + 48 : 589 ; SEE = 3 : 13 ; r = 0 : 90 ; p < 0 : 001 :We(HMGandTWH)collectedmeasurements from38Africanmalesand38femalescurated withintheDartCollectionoftheUniversityofthe Witwatersrand.SpecimenswererandomlychosenwithnopreferenceforspecificAfricanethnic groups.Cadavericstaturesaredocumentedfor thiscollection,theregressionequationrelating tibialengthtostatureinthissampleis:Stature = 0 : 223 × TML + 75 : 350 ; SEE = 6 : 50 ; r = 0 : 63 ; p < 0 : 001 : Figure21 .VirtualreconstructiontheDH3/DH4cranialbaseusingamodelofSts19.( A )Rightlateralview.( B )Left lateralview.( C )Posteriorview.( D )Inferiorview. DOI:10.7554/eLife.09560.026 Figure22 .VirtualreconstructiontheDH3/DH4 endocranialvolumeusingacranialbasemodelofSts 19.Rightlateralview. DOI:10.7554/eLife.09560.027 Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 31of35 Researcharticle Genomicsandevolutionarybiology

PAGE 32

NomenclaturalactsTheelectroniceditionofthisarticleconformstotherequirementsoftheamendedInternationalCode ofZoologicalNomenclature,andhence,thenewnamecontainedhereinisavailableunderthatCode fromtheelectroniceditionofthisarticle.Thispublishedworkandthenomenclaturalactsitcontains havebeenregisteredinZooBank,theonlineregistrationsystemfortheICZN.TheZooBankLSIDs(Life ScienceIdentifiers)canberesolvedandtheassociatedinformationviewedthroughanystandardweb browserbyappendingtheLSIDtotheprefix‘http://zoobank.org/’.TheLSIDforthispublicationis: urn:lsid:zoobank.org:pub:00D1E81A-6E08-4A01-BD98-79A2CEAE2411.Theelectroniceditionofthis workwaspublishedinajournalwithanISSN(2050-084X)andhasbeenarchivedandisavailablefrom thefollowingdigitalrepositories:PubMedCentralandLOCKSS.AccesstomaterialAllDinaledifossilmaterialisavailableforstudybyresearchersuponapplicationtotheEvolutionary StudiesInstituteattheUniversityoftheWitwatersrandwherethematerialiscurated(contact BernhardZipfel[Bernhard.Zipfel@wits.ac.za]).Three-dimensionalsurfacerenderingsandotherdigital dataareavailablefromtheMorphoSourcedigitalrepository(http://morphosource.org).AcknowledgementsTheauthorswouldliketothankthemanyfundingagenciesthatsupportedvariousaspectsofthis work.InparticulartheauthorswouldliketothanktheNationalGeographicSociety,theSouthAfrican NationalResearchFoundationandtheGautengProvincialGovernmentforparticularlysignificant fundingofthediscovery,recoveryandanalysisofthismaterial.Otherfundingagenciesincludeand thePalaeontologicalScientificTrust,theTexasA&MCollegeofLiberalArtsSeedGrantProgram,the LydaHillFoundationandtheWisconsinAlumniResearchFoundation.WewishtothanktheJacobs FamilyforaccesstothesiteandtheSouthAfricanHeritageResourceAgencyandCradleof HumankindUNESCOWorldHeritageSiteManagementAuthorityforissuingthevariouspermits requiredforthiswork,includingtheexcavationpermit(PermitID:952).Wewouldalsoliketothank theUniversityoftheWitwatersrandandtheEvolutionaryStudiesInstituteaswellastheSouthAfrican NationalCentreofExcellenceinPalaeoSciencesforcuratingthematerialandhostingtheauthors whilestudyingthematerial.AdditionalinformationFundingFunder Author NationalGeographicSocietyLeeRBerger TheNationalResearch FoundationofSouthAfrica LeeRBerger ThePalaeontologicalScientific Trust LeeRBerger LydaHillFoundationLeeRBerger WisconsinAlumniResearch Foundation(WARF) JohnHawks TexasAandMUniversityDarrylJdeRuiter Thefundershadnoroleinstudydesign,datacollectionandinterpretation,orthe decisiontosubmittheworkforpublication.Authorcontributions LRB,JH,DJR,SEC,Conceptionanddesign,Acquisitionofdata,Analysisandinterpretationofdata, Draftingorrevisingthearticle,Contributedunpublishedessentialdataorreagents;PS,Conception anddesign,Acquisitionofdata,Analysisandinterpretationofdata,Contributedunpublished essentialdataorreagents;LKD,TLK,HMG,SAW,JMDS,MMS,CMM,NC,TWH,WH-S,MB,BB,DB, JB,ZDC,KAC,ASD,MD,MCE,EMF,DG-M,DJG,AG,JDI,MFL,DM,MRM,CMO,DR,LS,JES,ZT, MWT,CVS,CSW,Acquisitionofdata,Analysisandinterpretationofdata,Contributedunpublished Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 32of35 Researcharticle Genomicsandevolutionarybiology

PAGE 33

essentialdataorreagents;RRA,MD,Analysisandinterpretationofdata,Contributedunpublished essentialdataorreagents;AK,SN,EWN,PW,BZ,Acquisitionofdata,Analysisandinterpretationof dataAdditionalfilesSupplementaryfiles·Supplementaryfile1.Holotypeandparatypespecimensandreferredmaterials.DOI:10.7554/eLife.09560.028·Supplementaryfile2.Traitsof H . naledi andcomparativespecies.DOI:10.7554/eLife.09560.029ReferencesAlemsegedZ ,SpoorF,KimbelWH,BobeR,GeraadsD,ReedD,WynnJG.2006.Ajuvenileearlyhomininskeleton fromDikika,Ethiopia. Nature 443 :296–301. doi:10.1038/nature05047 . Alm ´ ecijaS ,AlbaDM.2014.Onmanualproportionsandpad-to-padprecisiongraspingin Austalopithecus afarensis . JournalofHumanEvolution 73 :88–92. doi:10.1016/j.jhevol.2014.02.006 . Ant ´ onSC .2003.Naturalhistoryof Homoerectus . AmericanJournalofPhysicalAnthropology Suppl37126–170. doi:10.1002/ajpa.10399 . Ant ´ onSC ,PottsR,AielloLC.2014.Humanevolution.Evolutionofearly Homo :anintegratedbiological perspective. Science 345 :1236828. doi:10.1126/science.1236828 . AsfawB ,WhiteT,LovejoyO,LatimerB,SimpsonS,SuwaG.1999. Australopithecusgarhi :anewspeciesofearly hominidfromEthiopia. Science 284 :629–635. doi:10.1126/science.284.5414.629 . BergerLR ,deRuiterDJ,ChurchillSE,SchmidP,CarlsonKJ,DirksPH,KibiiJM.2010. Australopithecussediba : anewspeciesof Homo -likeaustralopithfromSouthAfrica. Science 328 :195–204. doi:10.1126/science. 1184944 . Berm ´ udezdeCastroJM ,Martin ´ on-TorresM,SierMJ,Mart ´ n-Franc ´ esL.2014.OnthevariabilityoftheDmanisi mandibles. PLOSONE 9 :e88212. doi:10.1371/journal.pone.0088212 . BlumenschineRJ ,PetersCR,MasaoFT,ClarkeRJ,DeinoAL,HayRL,SwisherCC,StanistreetIG,AshleyGM, McHenryLJ,SikesNE,VanDerMerweNJ,TactikosJC,CushingAE,DeocampoDM,NjauJK,EbertJI.2003. LatePliocene Homo andhominidlandusefromwesternOlduvaiGorge,Tanzania. Science 299 :1217–1221. doi:10.1126/science.1075374 . BoazNT ,HowellFC.1977.AgracilehominidcraniumfromuppermemberGoftheShunguraformation,Ethiopia. AmericanJournalofPhysicalAnthropology 46 :93–108. doi:10.1002/ajpa.1330460113 . BrownP ,SutiknaT,MorwoodMJ,SoejonoRP,SaptomoEW,DueRA.2004.Anewsmall-bodiedhomininfromthe LatePleistoceneofFlores,Indonesia.Nature 431 :1055–1061. doi:10.1038/nature02999 . CarlsonKJ ,StoutD,JashashviliT,deRuiterDJ,TafforeauP,CarlsonK,BergerLR.2011.TheendocastofMH1, Australopithecussediba . Science 333 :1402–1407. doi:10.1126/science.1203922 . ChurchillSE ,HollidayTW,CarlsonKJ,JashashviliT,MaciasME,MathewsS,SparlingTL,SchmidP,deRuiterDJ, BergerLR.2013.Theupperlimbof Australopithecussediba . Science 340 :1233477. doi:10.1126/science. 1233477 . ClarkeRJ .1977. UnpublishedPhDdissertation. Johannesburg:UniversityoftheWitwatersrand. ClarkeRJ .2008.LatestinformationonSterkfontein’s Australopithecus skeletonandanewlookat Australopithecus . SouthAfricanJournalofScience 104 :443–449. doi:10.1590/S003823532008000600015 . DemboM ,MatzkeNJ,MooersAØ,CollardM.2015.Bayesiananalysisofamorphologicalsupermatrixshedslight oncontroversialfossilhomininrelationships. ProceedingsoftheRoyalSocietyB 282 :20150943. doi:10.1098/ rspb.2015.0943 . DeSilvaJM ,HoltKG,ChurchillSE,CarlsonKJ,WalkerCS,ZipfelB,BergerLR.2013.Thelowerlimbandmechanics ofwalkingin Australopithecussediba . Science 340 :1232999. doi:10.1126/science.1232999 . DirksPHGM ,BergerLR,RobertsEM,KramersJD,HawksJ,Randolph-QuinneyPS,ElliottM,MusibaCM,Churchill SE,deRuiterDJ,SchmidP,BackwellLR,BelyaninGA,BoshoffP,HunterKL,FeuerriegelEM,GurtovA,duG HarrisonJ,HunterR,KrugerA,MorrisH,MakhubelaTV,PeixottoB,TuckerS.2015.Geologicalandtaphonomic evidencefordeliberatebodydisposalbytheprimitivehomininspecies Homonaledi fromtheDinalediChamber, SouthAfrica. eLife 4 :e09651. doi:10.7554/eLife.09561 . Dom ´ nguez-RodrigoM ,PickeringTR,BaquedanoE,MabullaA,MarkDF,MusibaC,BunnHT,UribelarreaD, SmithV,Diez-MartinF,P ´ erez-Gonz ´ alezA,S ´ anchezP,SantonjaM,BarboniD,GidnaA,AshleyG,YravedraJ, HeatonJL,ArriazaMC.2013.Firstpartialskeletonofa1.34-million-year-old Paranthropusboisei fromBedII, OlduvaiGorge,Tanzania. PLOSONE 8 :e80347. doi:10.1371/journal.pone.0080347 . DrapeauMS ,WardCV,KimbelWH,JohansonDC,RakY.2005.Associatedcranialandforelimbremainsattributed to Australopithecusafarensis fromHadar,Ethiopia. JournalofHumanEvolution 48 :593–642. doi:10.1016/j.jhevol.2005.02.005 . Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 33of35 Researcharticle Genomicsandevolutionarybiology

PAGE 34

FalkD ,HildeboltC,SmithK,MorwoodMJ,SutiknaT,BrownP,Jatmiko,SaptomoEW,BrunsdenB,PriorF.2005. ThebrainofLB1, Homofloresiensis . Science 308 :242–245. doi:10.1126/science.1109727 . FullyG .1956.Unenouvellem ´ ethodeded ´ eterminationdelataille. AnnalesdeM ´ edecineL ´ egale,Criminologie, PoliceScientifiqueetToxicologie 36 :266–273. GabuniaL ,VekuaA,LordkipanidzeD,SwisherCCIII,FerringR,JustusA,NioradzeM,TvalchrelidzeM,Ant ´ onSC, BosinskiG,J ¨ orisO,LumleyMA,MajsuradzeG,MouskhelishviliA.2000.EarliestPleistocenehominidcranial remainsfromDmanisi,RepublicofGeorgia:taxonomy,geologicalsettingandage. Science 288 :1019–1025. doi:10.1126/science.288.5468.1019 . GrauszHM ,LeakeyR,WalkerA,WardC.1988.Associatedcranialandpostcranialbonesof Australopithecus boisei .In:GrineFE,editor. Evolutionaryhistoryofthe‘Robust’Australopithecines. NewYork:AldinedeGruyter. p.127–132. GrineFE ,DemesB,JungersWL,ColeTMIII.1993.Taxonomicaffinityoftheearly Homo craniumfromSwartkrans, SouthAfrica. AmericanJournalofPhysicalAnthropology 92 :411–426. doi:10.1002/ajpa.1330920402 . GrineFE ,JungersWL,TobiasPV,PearsonOM.1995.Fossil Homo femurfromBergAukas,northernNamibia. AmericanJournalofPhysicalAnthropology 97 :151–185. doi:10.1002/ajpa.1330970207 . GrineFE ,JungersWL,SchultzJ.1996.Pheneticaffinitiesamongearly Homocrania fromEastandSouthAfrica. JournalofHumanEvolution 30 :189. doi:10.1006/jhev.1996.0019 . Haile-SelassieY ,LatimerBM,AleneM,DeinoAL,GibertL,MelilloSM,SaylorBZ,ScottGR,LovejoyCO.2010.An early Australopithecusafarensis postcraniumfromWoranso-Mille,Ethiopia. ProceedingsoftheNational AcademyofSciencesofUSA 107 :12121–12126. doi:10.1073/pnas.1004527107 . Harcourt-SmithWEH ,ThrockmortonZ,CongdonKA,ZipfelB,DeaneAS,DrapeauMSM,ChurchillSE,BergerLR, DeSilvaJM.2015.Thefootof Homonaledi . NatureCommunications 6 :8432. doi:10.1038/ncomms9432 . HawksJ ,HunleyK,LeeSH,WolpoffM.2000.PopulationbottlenecksandPleistocenehumanevolution. Molecular BiologyandEvolution 17 :2–22. doi:10.1093/oxfordjournals.molbev.a026233. HollidayTW .2012.Bodysize,bodyshape,andthecircumscriptionofthegenus Homo . CurrentAnthropology 53 : S330–S345. doi:10.1086/667360 . JohansonDC ,MasaoFT,EckGG,WhiteTD,WalterRC,KimbelWH,AsfawB,ManegaP,NdessokiaP,SuwaG. 1986.Newpartialskeletonof Homohabilis fromOlduvaiGorge,Tanzania. Nature 327 :205–209. doi:10.1038/ 327205a0 . JohansonDC ,TaiebM,CoppensY.1982.PliocenehominidsfromtheHadarformation,Ethiopia(1973–1977): Stratigraphic,chronologic,andpaleoenvironmentalcontexts,withnotesonhominidmorphologyand systematics. AmericanJournalofPhysicalAnthropology 57 :373–402. doi:10.1002/ajpa.1330570402 . JungersWL ,Harcourt-SmithWE,WunderlichRE,TocheriMW,LarsonSG,SutiknaT,DueRA,MorwoodMJ. 2009a.Thefootof Homofloresiensis . Nature 459 :81–84. doi:10.1038/nature07989 . JungersWL ,LarsonSG,Harcourt-SmithW,MorwoodMJ,SutiknaT,AweRD,DjubiantonoT.2009b.Descriptionsofthe lowerlimbskeletonof Homofloresiensis . JournalofHumanEvolution 57 :538–554. doi:10.1016/j.jhevol.2008.08.014 . KeyserAW .2000.TheDrimolenskull:themostcompleteaustralopithecinecraniumandmandibletodate. South AfricanJournalofScience 96 :189–192. KeyserAW ,MenterCG,Moggi-CecchiJJ,PickeringTR,BergerLR.2000.Drimolen:anewhominin-bearingsitein Gauteng,SouthAfrica. SouthAfricanJournalofScience 96 :193–197. KimbelWH ,RakY,JohansonDC,Minugh-PurvisN.2004. TheskullofAustralopithecusafarensis. NewYork:Oxford. KimbelWH ,JohansonDC,RakY.1997.SystematicassessmentofamaxillaofHomofromHadar,Ethiopia. AmericanJournalofPhysicalAnthropology 103 :235–262. doi:10.1002/(SICI)1096-8644(199706)103:2 < 235::AIDAJPA8 > 3.0.CO;2-S . KivellTL ,KibiiJM,ChurchillSE,SchmidP,BergerLR.2011. Australopithecussediba handdemonstratesmosaic evolutionoflocomotorandmanipulativeabilities. Science 333 :1411–1417. doi:10.1126/science.1202625 . KivellTL ,DeaneAS,TocheriMW,OrrCM,SchmidP,HawksJ,BergerLR,ChurchillSE.2015.Thehandof Homo naledi . NatureCommunications 6 :8431. doi:10.1038/ncomms9431 . LeakeyMG ,SpoorF,DeanMC,FeibelCS,Ant ´onSC,KiarieC,LeakeyLN.2012.NewfossilsfromKoobiForain northernKenyaconfirmtaxonomicdiversityinearlyHomo. Nature 488 :201–204. doi:10.1038/nature11322 . LeeSH .2001.Assignedresamplingmethod:anewmethodtoestimatesizesexualdimorphisminsamplesof unknownsex. Przegla dAntropologicznyAnthropologicalReview 64 :21–39. LordkipanidzeD ,JashashviliT,VekuaA,PoncedeLe ´ onMS,ZollikoferCP,RightmireGP,PontzerH,FerringR, OmsO,TappenM,BukhsianidzeM,AgustiJ,KahlkeR,KiladzeG,Martinez-NavarroB,MouskhelishviliA, NioradzeM,RookL.2007.Postcranialevidencefromearly Homo fromDmanisi,Georgia. Nature 449 :305–310. doi:10.1038/nature06134 . LordkipanidzeD ,PoncedeLe ´ onMS,MargvelashviliA,RakY,RightmireGP,VekuaA,ZollikoferCP.2013.A completeskullfromDmanisi,Georgia,andtheevolutionarybiologyofearly Homo . Science 342 :326–331. doi:10.1126/science.1238484 . McKellarAE ,HendryAP.2009.Howhumansdifferfromotheranimalsintheirlevelsofmorphologicalvariation. PLOSONE 4 :e6876. doi:10.1371/journal.pone.0006876 . MorwoodMJ ,BrownP,SutiknaT,SaptomoEW,WestawayKE,DueRA,RobertsRG,MaedaT,WasistoS, DjubiantonoT.2005.Furtherevidenceforsmall-bodiedhomininsfromtheLatePleistoceneofFlores,Indonesia. Nature 437 :1012–1017. doi:10.1038/nature04022 . OrrCM ,TocheriMW,BurnettSE,AweRD,SaptomoEW,SutiknaT,Jatmiko,WasistoS,MorwoodMJ,Jungers WL.2013.Newwristbonesfor Homofloresiensis fromliangbua(Flores,Indonesia). JournalofHumanEvolution 64 :109–129. doi:10.1016/j.jhevol.2012.10.003 . Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 34of35 Researcharticle Genomicsandevolutionarybiology

PAGE 35

RaxterMH ,AuerbachBM,RuffCB.2006.Revisionofthefullytechniqueforestimatingstatures. AmericanJournal ofPhysicalAnthropology 130 :374–384. doi:10.1002/ajpa.20361 . RightmireGP ,LordkipadnizeD,VekuaA.2006.Anatomicaldescriptions,comparativestudiesandevolutionary significanceofthehomininskullsfromDmanisi,RepublicofGeorgia. JournalofHumanEvolution 50 :115–141. doi:10.1016/j.jhevol.2005.07.009 . RolianC ,GordonAD.2013.Reassessingmanualproportionsin Australopithecusafarensis . AmericanJournalof PhysicalAnthropology 152 :393–406. doi:10.1002/ajpa.22365 . RuffCB ,TrinkausE,HollidayTW.1997.BodymassandencephalizationinPleistocene Homo . Nature 387 :173–176. doi:10.1038/387173a0 . SchmidP ,ChurchillSE,NallaS,WeissenE,CarlsonKJ,deRuiterDJ,BergerLR.2013.Mosaicmorphologyinthe thoraxof Australopithecussediba . Science 340 :1234598. doi:10.1126/science.1234598 . SchrenkF ,BromageTG,BetzlerCG,RingU,JuwayeyiYM.1993.Oldest Homo andPliocenebiogeographyofthe Malawirift. Nature 365 :833–836. doi:10.1038/365833a0 . SpoorF ,LeakeyMG,GathogoPN,BrownFH,Ant ´ onSC,McDougallI,KiarieC,ManthiFK,LeakeyLN.2007.Implications ofnewearly Homo fossilsfromIleret,eastofLakeTurkana,Kenya. Nature 448 :688–691. doi:10.1038/nature05986 . StockJT .2013.Theskeletalphenotypeof‘Negritos’fromtheAndamanIslandsandPhilippinesrelativetoglobal variationamonghunter-gatherers. HumanBiology 85 :67–94. doi:10.3378/027.085.0304 . SusmanRL .1988.Handof Paranthropusrobustus fromMember1,Swartkrans:fossilevidencefortoolbehavior. Science 240 :781–784. doi:10.1126/science.3129783 . SusmanRL ,deRuiterD,BrainCK.2001.Recentlyidentifiedpostcranialremainsof Paranthropus andearly Homo fromSwartkransCave,SouthAfrica. JournalofHumanEvolution 41 :607–629. doi:10.1006/jhev.2001.0510 . SuwaG ,AsfawB,BeyeneY,WhiteTD,KatohS,NagaokaS,NakayaH,UzawaK,RenneP,WoldeGabrielG.1997. Thefirstskullof Australopithecusboisei . Nature389 :489–492. doi:10.1038/39037 . SuwaG ,WhiteTD,HowellFC.1996.MandibularpostcaninedentitionfromtheShunguraformation,Ethiopia: crownmorphology,taxonomicallocations,andPlio-Pleistocenehominidevolution. AmericanJournalofPhysical Anthropology 101 :247–282. doi:10.1002/(SICI)1096-8644(199610)101:2 < 247::AID-AJPA9 > 3.0.CO;2-Z . SuwaG ,AsfawB,Haile-SelassieY,WhiteT,KatohS,WoldegabrielG,HartWK,NakayaH,BeyeneY.2007.Early pleistocene Homoerectus fossilsfromkonso,southernEthiopia. AnthropologicalScience 115 :133–151. doi:10. 1537/ase.061203 . TobiasPV .1967. ThecraniumandmaxillarydentitionofAustralopithecus(Zinjanthropus)boisei. Cambridge: CambridgeUniversityPress. TobiasPV .1991. OlduvaiGorgevolume4:theskulls,endocastsandteethofHomohabilis. Cambridge:Cambridge UniversityPress. TocheriMW ,OrrCM,LarsonSG,SutiknaT,Jatmiko,SaptomoEW,DueRA,DjubiantonoT,MorwoodMJ,Jungers WL.2007.Theprimitivewristof Homofloresiensis anditsimplicatiosnforhomininevolution. Science 317 : 1743–1745. doi:10.1126/science.1147143 . TocheriMW ,RazdanA,WilliamsRC,MarzkeMW.2005.A3Dquantitativecomparisonoftrapeziumandtrapezoid relativearticularandnonarticularsurfaceareasinmodernhumansandgreatapes. JournalofHumanEvolution 49 :570–586. doi:10.1016/j.jhevol.2005.06.005 . VekuaA ,LordkipanidzeD,RightmireGP,AgustiJ,FerringR,MaisuradzeG,MouskhelishviliA,NioradzeM,De LeonMP,TappenM,TvalchrelidzeM,ZollikoferC.2002.Anewskullofearly Homo fromDmanisi,Georgia. Science 297 :85–89. doi:10.1126/science.1072953 . WardCV ,KimbelWH,HarmonEH,JohansonDC.2012.Newpostcranialfossilsof Australopithecusafarensis from hadar,Ethiopia(1990–2007). JournalofHumanEvolution 63 :1–51. doi:10.1016/j.jhevol.2011.11.012 . WardCV ,TocheriMW,PlavcanJM,BrownFH,KyaloManthiF.2013.EarlyPleistocenethirdmetacarpalfrom Kenyaandtheevolutionofmodernhuman-likehandmorphology. ProceedingsoftheNationalAcademyof SciencesofUSA 111 :121–124. doi:10.1073/pnas.1316014110 . WeidenreichF .1943.Theskullof Sinanthropus . PalaeontologiaSinica 10 :1. WoodBA .1991. KoobiForaresearchprojectIV:hominidcranialremainsfromKoobiFora. Oxford:Clarendon. WoodBA ,CollardM.1999.Thehumangenus. Science 284:65–71. doi:10.1126/science.284.5411.65 . ZipfelB ,DeSilvaJM,KiddRS,CarlsonKJ,ChurchillSE,BergerLR.2011.Thefootandankleof Australopithecus sediba . Science 333 :1417–1420. doi:10.1126/science.1202703 . Berger etal .eLife2015;4:e09560. DOI:10.7554/eLife.09560 35of35 Researcharticle Genomicsandevolutionarybiology


printinsert_linkshareget_appmore_horiz

Download Options

close
No images are available for this item.
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.