No known hominin species matches the expected … Robles et...No known hominin species matches the expected dental morphology of the last common ancestor of Neanderthals and modern

No known hominin species matches the expecteddental morphology of the last common ancestorof Neanderthals and modern humansAida Gómez-Roblesa,b,1, José María Bermúdez de Castroc, Juan-Luis Arsuagad, Eudald Carbonelle, and P. David Pollyf

aDepartment of Anthropology, The George Washington University, Washington, DC 20052; bKonrad Lorenz Institute for Evolution and Cognition Research,A-3422 Altenberg, Austria; cCentro Nacional de Investigación sobre la Evolución Humana, 09002 Burgos, Spain; dCentro de Investigación Sobre la Evolucióny Comportamiento Humanos, Universidad Complutense de Madrid–Instituto de Salud Carlos III, 28029 Madrid, Spain; eInstitut Català de PaleoecologiaHumana i Evolució Social, Universitat Rovira i Virgili, 43007 Tarragona, Spain; and fDepartments of Geological Sciences, Biology, and Anthropology, IndianaUniversity, Bloomington,IN 47405

Edited by Jukka Jernvall, University of Helsinki, Helsinki, Finland, and accepted by the Editorial Board September 19, 2013 (received for reviewFebruary 9, 2013)

A central problem in paleoanthropology is the identity of the lastcommon ancestor of Neanderthals and modern humans ([N-MH]LCA). Recently developed analytical techniques now allow thisproblem to be addressed using a probabilistic morphological frame-work. This study provides a quantitative reconstruction of theexpected dental morphology of the [N-MH]LCA and an assessmentof whether known fossil species are compatible with this ancestralposition. We show that no known fossil species is a suitable candi-date for being the [N-MH]LCA and that all late Early and MiddlePleistocene taxa from Europe have Neanderthal dental affinities,pointing to the existence of a European clade originated around1 Ma. These results are incongruent with younger molecular diver-gence estimates and suggest at least one of the following must betrue: (i) European fossils and the [N-MH]LCA selectively retainedprimitive dental traits; (ii) molecular estimates of the divergencebetween Neanderthals and modern humans are underestimated;or (iii) phenotypic divergence and speciation between both specieswere decoupled such that phenotypic differentiation, at least indental morphology, predated speciation.

phylogeny | node reconstruction | geometric morphometrics |morphospace | European Pleistocene

Uncertainty about ancestry is typical for many nodes of thehominin phylogeny. Species are usually suggested as com-

mon ancestors based on their geography, chronology, and mor-phological affinities (1–3), but these criteria can be unsatisfactoryfor two fundamental reasons: first, the inherent incompletenessof the hominin fossil record precludes a clear understanding ofthe spatial and temporal range of variation of hominin species;and second, comparative morphological analyses are frequentlydescriptive in nature, making them difficult to evaluate objectively.Fossil dental remains are central to our understanding of

hominin evolution due to their commonness in archaeologicaland paleontological sites, their normally good state of preserva-tion, and the minor influence that environmental factors—apartfrom wear—have on their morphology during an individual’s life.Teeth provide the most consistently complete representation ofthe hominin fossil record and are thus fundamentally importantfor assessing the plausibility of competing evolutionary scenariosand avoiding inadvertently biased interpretations (Tables S1 andS2). The combination of quantitative evolutionary theory, geo-metric morphometrics, and dental anthropology provides a pow-erful framework for reconstructing statistical expectations of themorphology of unknown ancestors (ref. 4 but see also ref. 5) andfor making posterior comparisons with actual fossil remains.The main aim of our study is to evaluate whether several known

candidate species have morphologies that are statistically consistentwith them being the last common ancestor of Neanderthals andmodern humans ([N-MH]LCA). To be a credible ancestor, a fossilmust have lived at the appropriate time and be statistically close tothe ancestral morphology, taking into account the uncertainty of the

ancestral reconstruction that arises from lability of the evolutionaryprocess and alternative hypotheses of the phylogeny on which it isbased. To implement this, we used geometric morphometric data toreconstruct ancestral dental shapes and their confidence ellipsoidsusing the phylogenetic generalized linear model (GLM) approach(ref. 6; see Materials and Methods and SI Text for detailed explan-ations). Ancestral reconstructions are based on an independentphylogeny, ideally one built from nondental characters to avoidcircularity (Fig. 1 and Fig. S1). Note, however, that the phylogeneticrelationships of hominin species are vigorously debated and thatthey are based mainly on craniodental traits, although not on thequantitative dental shape traits we use (Figs. S2 and S3). The an-cestral dental reconstructions were based only on species whosephylogenetic position is relatively uncontroversial; species whosephylogenetic position is contested were evaluated as potentialcandidate ancestors for nodes of relevant age. We considered 12phylogenetic topologies that are collectively representative of therange of credible alternative hypotheses about hominin rela-tionships (7). The ages of fossils and nodes were calibrated usingcomprehensive reviews of the chronology, geographical location,morphology and ecology of hominin species (8). In addition tothe [N-MH]LCA, we evaluated ancestor candidates at two otherimportant nodes of hominin phylogeny, the last common ances-tor (LCA) of the genus Homo (excluding transitional homininssuch as Homo habilis and Homo rudolfensis) and the LCA ofParanthropus and Homo, as controls for whether our statisticalframework is too strict. Our method is not intended to be

Significance

The identity of the last common ancestor of Neanderthals andmodern humans is a controversial issue. This debate has beenoften addressed by means of descriptive analyses that aredifficult to test. Our primary aim is to put questions abouthuman evolution into a testable quantitative framework andto offer an objective means to sort out apparently unsolvabledebates about hominin phylogeny. Our paper shows that noknown hominin species matches the expected morphology ofthis common ancestor. Furthermore, we found that Europeanrepresentatives of potential ancestral species have had af-finities with Neanderthals for almost 1 My, thus supportinga model of early divergence between Neanderthals andmodern humans.

Author contributions: A.G.-R. and P.D.P. designed research; A.G.-R., J.M.B.d.C., J.-L.A.,E.C., and P.D.P. performed research; P.D.P. contributed new reagents/analytic tools;A.G.-R. analyzed data; and A.G.-R. and P.D.P. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission. J.J. is a guest editor invited by the Editorial Board.1To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1302653110 PNAS Early Edition | 1 of 6

EVOLU

TION

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental/pnas.201302653SI.pdf?targetid=nameddest=ST1


http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental/pnas.201302653SI.pdf?targetid=nameddest=STXT

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental/pnas.201302653SI.pdf?targetid=nameddest=SF1



mailto:[email protected]

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental

www.pnas.org/cgi/doi/10.1073/pnas.1302653110

a phylogenetic reconstruction technique, but a means to evaluatecompatibility of potential ancestral species within the frameworkof established phylogenies.Phylogenetic scenarios differed from one another in the

number and age of tip species and in the age of the nodes(branch lengths; Fig. 1 and Fig. S1). European Middle Pleisto-cene fossils were grouped into one of two operational taxonomicunits based on previously recognized morphological differences(9): the Homo heidelbergensis group (represented in our studyby fossils from Arago Cave) and the Sima de los Huesos (SH)group. SH hominins, unlike other European Middle Pleistocenefossils (such as those from Arago, Mauer, and Petralona), arewidely considered to be an early chronospecies of classic Nean-derthals (3, 9–11). However, the markedly derived dentitions ofSH hominins (12) appear to contradict their placement as anearly member of the Neanderthal lineage and their morphologysubstantially influences the reconstruction of the [N-MH]LCA;therefore, we evaluated SH both as a taxon on the phylogenetictree and as a potential candidate for the [N-MH]LCA. For theformer, SH was considered to be a node in the Neanderthallineage at 350 ka following ref. 13, an age that is less contro-versial than a subsequent, older assessment of circa 600 ka (14).We considered two alternative estimates for the divergence timebetween Neanderthals and modern humans (1, 11). One scenariodates this divergence to 450 ka based on molecular clock esti-mates (11, 15); the other dates this divergence to 1 Ma based onmorphological affinities observed in the fossil record (see Results

and Discussion). We also considered two dates for Homo erectus,1 Ma or 500 ka, to account for the different age estimates of keyJavanese (16) and Chinese fossils (17). We included H. erectusas a potential candidate species for the [N-MH]LCA in somephylogenetic scenarios. For each candidate, we evaluated itscompatibility with the statistical reconstruction of the morphol-ogy of the LCA under each of the 12 phylogenetic scenarios.Species were considered implausible as ancestors if one or moreof their teeth was incompatible with the expected ancestralmorphology under all 12 of the phylogenetic scenarios.

ResultsThe statistical reconstruction of the dental morphology of the[N-MH]LCA is characterized by symmetric and lingually re-duced upper premolars (P3 and P4) and by slightly asymmetriclower premolars (P3 and P4). The upper first molar (M1) hasa slightly skewed morphology and the lower first molar (M1),a Dryopithecus pattern (also called “Y5”) with five well-developed main cusps. The Dryopithecus pattern is the sharedancestral condition for all hominin lower molars, and consists ofa five-cusped molar in which the mesiolingual cusp (metaconid)and the distobuccal cusp (hypoconid) are in contact and createa Y-shape fissure pattern. The upper second molar (M2) of thereconstructed common ancestor has a mildly reduced hypocone,whereas the upper third molar (M3) has strong reduction suchthat a high proportion of individuals are expected to lack thedistolingual cusp. Lower second molars (M2) of the recon-structed [N-MH]LCA are expected to have marked reduction ofthe fifth cusp and frequent occurrence of non-Y5 morphologies(mainly cruciform patterns). The lower third molar (M3) ofthe expected ancestral phenotype has five cusps that are notarranged in a Y pattern, and a reduction of the lingual half of themolar relative to its buccal half (Fig. 2). This reconstruction ofthe dental morphology of the [N-MH]LCA is generally consis-tent with all 12 phylogenies, one of which is shown in Fig. 2 (theone that includes the SH group as a pre-Neanderthal population,that uses the earlier date of 1 Ma for the [N-MH]LCA, andwhich dates H. erectus to 1 Ma; Fig. 3 and Figs. S4 and S5 arebased on the same phylogeny).Our results reveal that no known species of the relevant age

is statistically consistent with the expected morphology of the[N-MH]LCA, even given the considerable uncertainty of thereconstructed ancestral morphologies. Graphical reconstructionsof the nodes in dental morphospace are shown in Fig. 3 and Fig.S4, along with confidence ellipsoids that show the associateduncertainties that arise from the stochastic nature of the evolu-tionary process. The size of each confidence ellipsoid is directlyrelated to the lengths of the branches connected to the node. Atleast two postcanine teeth fall outside the 95% confidenceintervals of the [N-MH]LCA reconstruction for all candidatespecies under all 12 phylogenetic scenarios (Table 1). Further-more, taxa such as H. heidelbergensis have several postcanineteeth that are morphologically consistent with the [N-MH]LCAonly in one or two phylogenetic scenarios (Table S3). The ad-dition of other European fossils to the H. heidelbergensis sample,such as those from Mauer and Petralona, could alter this result,but this is unlikely because they are similar (not only in dental,but also in cranial and mandibular morphology) to the Aragospecimens relative to other taxa in our analysis (2, 3, 9, 18). Ourresults demonstrate that morphological evidence for any of thesetaxa being the [N-MH]LCA is, at best, weak. For these species tobe credible ancestors, the age of the Neanderthal–modern hu-man split would have to be radically different from the range weconsidered or selection on the dentition would have to havecaused the ancestor to radically deviate from the expectations ofBrownian motion evolution. In addition to ruling out thesecandidate species, our results indicate that all three Europeancandidate groups are statistically closer to the Neanderthal lin-eage than to the modern human lineage (Table 2 and Table S4),which suggests that European hominins have had significantdental affinities with Neanderthals for almost 1 My (the oldest

Fig. 1. Phylogeny used for ancestral state reconstruction and candidatespecies. Numbers in parentheses represent the averaged chronology of thecorresponding species (although see ref. 56); italic numbers represent branchlengths in millions of years; bold gray numbers represent ages of nodes.Branch lengths corresponding to a 1-Ma divergence between Neanderthalsand modern humans are represented to the left of the slashes and valuescorresponding to a 0.45-Ma divergence to the right. H. erectus has beendated to 1 Ma (values before the slash in the H. erectus branch) or to 0.5 Ma(values after the slash). Candidate species are represented below the phy-logenetic tree with bars showing their approximate temporal spans (black,candidates for the [P-H]LCA; dark gray, candidates for the [H]LCA; light gray,candidates for the [N-MH]LCA). H. erectus and the SH group have been in-cluded in the framework phylogeny in four (H. erectus) and six (SH) phylo-genetic scenarios (Fig. S1) and evaluated as candidate species in theremaining scenarios. H. ergaster has been evaluated as a candidate speciesfor the [H]LCA and the [N-MH]LCA. H. habilis has been evaluated as a can-didate for the [P-H]LCA and the [H]LCA. Vertical guides indicate the age ofthe evaluated nodes. Fig. S1 shows individualized representations of the 12phylogenetic topologies.

2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1302653110 Gómez-Robles et al.











species, Homo antecessor, has been dated to ca. 950 ka; ref. 19).African Homo ergaster and Asian H. erectus do not share thisEuropean affinity, but they have nearly equal dental affinitieswith Neanderthals and with modern humans (Table 2 and TableS4). H. ergaster was evaluated as a [N-MH]LCA candidate be-cause it is the youngest African species in our sample, despite itbeing considerably older than this ancestor is expected to be.Interestingly, H. ergaster fits the expected ancestral phenotypebetter than some of the younger candidate species (Table 1).Asian H. erectus was also evaluated as an ancestral candidatebecause the fossil specimens included in our study lived duringthe right time interval (0.5–1 Ma), but it has generally a worse fitwith the [N-MH]LCA than H. ergaster.Our method identified compatible candidates for other nodes

in the phylogeny (Fig. S5). Our data show that H. ergaster and,especially, H. habilis fit the expected morphology of the LCA ofthe genus Homo ([H]LCA) at all tooth positions in one or moreof the 12 phylogenies (Table 1 and Table S3). Similarly, bothAustralopithecus afarensis and Australopithecus africanus fit theexpected morphology of the LCA of Paranthropus and Homo ([P-H]LCA) at all tooth positions in most or all of the phylogeneticscenarios (Table 1 and Table S3). Despite being age-in-appropriate, we also evaluated H. habilis as a candidate for the[P-H]LCA because of its suggested affinities with Australopithecusspecies (20). Interestingly, it has even stronger compatibility withthe expected morphology of this ancestor than do A. afarensis andA. africanus (Table 1 and Table S3). The fact that several can-didates are compatible with deeper nodes is not surprising be-cause the long branch lengths leading to older nodes result inbroader confidence ellipses than at shallower nodes with shorterbranches (21). Regardless of uncertainty, Procrustes distances

show that the fit between candidates and their correspondingnodes is substantially better at the older nodes than at the [N-MH]LCA (Table 1). The main exception is the SH group, whichhas some teeth that are comparatively close to the [N-MH]LCAin terms of Procrustes distances, but which is incompatible as anancestor because some of its teeth are strongly derived and havea very low probability of conforming to the expected ancestralmorphology.

DiscussionOur results demonstrate that, even though there are plausiblecandidate species for deep nodes of the hominin phylogeny (A.afarensis, A. africanus, and H. habilis for the [P-H]LCA and H.habilis and H. ergaster for the [H]LCA), we still do not havea good fossil candidate for the phenotypic LCA of Neanderthalsand modern humans. This result is striking because confidenceintervals of ancestral reconstructions are very broad, whichmeans that existing candidates are very incompatible. Species

Fig. 2. Dental morphology of the [N-MH]LCA. Estimated ancestral mor-phologies are represented as TPS-grids showing the deformation from thebase of the phylogeny into the estimated ancestral shape. Photographs(with landmarks represented as blue points) show examples of typicalmorphologies found in the Neanderthal lineage (NEA) and in the modernhuman (H. sapiens) lineage (SAP). Mesial margins are represented to the left,distal margins to the right, buccal margins to the top, and lingual margins tothe bottom of the figure. Not to scale.

B

A

Fig. 3. Projection of the simplified hominin phylogenetic tree (representedby lines connecting terminal species and nodes) with terminal species (largedark-gray filled circles), ancestor shapes (small light-gray filled circles) andcandidate species (open circles) into the upper and lower first molar mor-phospaces. Confidence ellipses (95%) around the three nodes discussed inthe text are represented with the three ancestors located in their centers(solid line, confidence ellipse of the [N-MH]LCA; dashed line, confidence el-lipse of the [H]LCA; dotted line, confidence ellipse of the [P-H]LCA). (A)Upper first molar morphospace. (B) Lower first molar morphospace. Plotscorresponding to premolars and second and third molars are provided in Fig.S4. Note that confidence ellipses in these graphs correspond only to the firstand second principal components (PCs), whereas P values in Table 1 are basedon all of the PCs. Phylogeny: BOI, P. boisei; ERE, H. erectus; NEA, H. nean-derthalensis; ROB, Paranthropus robustus; SAP, H. sapiens; SH, SH group.Candidates: AFA, A. afarensis; AFR, A. africanus; ANT, H. antecessor; ERG, H.ergaster; GEO, Homo georgicus; HAB, H. habilis; HEI, H. heidelbergensis.

Gómez-Robles et al. PNAS Early Edition | 3 of 6

EVOLU

TION









falling outside the confidence intervals for a given node can beexcluded from ancestry with high confidence presuming themodel of Brownian motion evolution is applicable. The use ofseveral different phylogenetic topologies in our study makes therejection of these candidates even stronger.Even though none of the candidate species we evaluated were

compatible with the [N-MH]LCA, note that our study did notinclude African populations of H. heidelbergensis or Africanspecimens dated to ca. 1 Ma. These groups deserve scrutiny aspotential ancestors, especially because they lived at the righttime and their cranial morphological affinities are ambiguous (2,18, 22–24). Middle Pleistocene fossils from Africa (sometimesreferred to as Homo rhodesiensis) have been considered by someas part of the lineage leading to H. sapiens (22) and by othersas belonging to the ancestral species of Neanderthals andmodern humans (3, 18, 23, 24). African fossils dated to ca. 0.5–1Ma thus merit continued study and are currently the most prom-ising source of candidates for the [N-MH]LCA.The lack of a plausible [N-MH]LCA candidate among Euro-

pean groups and their dental affinity with Neanderthals back to

1 Ma suggests three possible scenarios (covered in the next threesections), which are not necessarily mutually exclusive.

Neanderthals and the [N-MH]LCA Could Be Primitive in Their DentalMorphology. One possible explanation of our results is that theconstellation of characters shared by European Pleistocene fos-sils represents the primitive condition for the clade Homoneanderthalensis–H. sapiens and that the true ancestral pheno-type was closer to the Neanderthal condition than expectedunder a Brownian motion model. This scenario would requirerelative stasis in the evolution of dental morphology in the Ne-anderthal lineage and either neutral or directional evolution inthe H. sapiens lineage. If this were the case, our ancestral re-construction method would fail to recover the correct mor-phology because it assumes similar, nondirectional evolutionarychanges in the lineages stemming from a node and subtending it.We have tested and rejected this possibility elsewhere (25), andso consider this scenario unlikely. Whereas some authors havesuggested that Neanderthals are morphologically more primitivethan modern humans (26), quantitative analysis of evolutionary

Table 1. Procrustes distances between expected ancestral morphologies and candidate species

Procrustes distance* (P value)†

P3 P4 M1 M2 M3 P3 P4 M1 M2 M3

[N-MH]LCASH group 0.045

(<0.001)0.027(0.072)

0.024(0.373)

0.034(0.710)

0.049(0.392)

0.051(0.042)

0.042(0.103)

0.023(0.432)

0.031(0.169)

0.044(0.074)

H. heidelbergensis 0.046(0.785)

0.061(0.147)

0.052(0.006)

0.072(0.001)

0.059(0.255)

0.057(0.310)

0.036(0.487)

0.025(0.991)

0.034(0.194)

0.042(0.980)

H. antecessor 0.038(0.234)

0.037(0.280)

0.051(0.001)

0.058(0.001)

N/A 0.081(0.036)

0.046(0.259)

0.030(0.934)

0.048(0.016)

0.041(0.360)

H. erectus 0.031(0.999)

0.023(0.805)

0.028(0.867)

0.053(0.015)

0.067(0.030)

0.084(0.039)

0.043(0.622)

0.029(0.733)

0.038(0.628)

0.038(0.678)

H. ergaster 0.051(0.206)

0.044(0.920)

0.053(0.013)

0.077(0.001)

N/A 0.061(0.250)

0.032(0.608)

0.038(0.913)

0.034(0.670)

0.042(0.811)

[H]LCAH. georgicus 0.061

(0.456)0.028(0.580)

0.050(0.918)

0.049(0.307)

0.091(<0.001)

0.059(0.009)

0.030(0.988)

0.033(0.983)

0.036(0.955)

0.050(0.583)

H. ergaster 0.044(0.394)

0.040(0.995)

0.050(0.060)

0.060(0.170)

N/A 0.057(0.256)

0.035(0.694)

0.035(0.894)

0.034(0.765)

0.037(0.805)

H. habilis 0.041(0.627)

0.037(0.596)

0.026(0.868)

0.032(0.300)

0.039(0.602)

0.036(0.813)

0.046(0.497)

0.021(0.965)

0.030(0.919)

0.034(0.971)

[P-H]LCAH. habilis 0.028

(0.847)0.027(0.944)

0.024(0.948)

0.029(0.751)

0.023(0.953)

0.046(0.569)

0.040(0.724)

0.022(0.990)

0.031(0.965)

0.037(0.947)

A. africanus 0.031(0.970)

0.029(0.647)

0.029(0.732)

0.031(0.934)

0.056(0.410)

0.043(0.932)

0.026(0.992)

0.025(0.925)

0.035(0.745)

0.043(0.573)

A. afarensis 0.041(0.962)

0.033(0.868)

0.030(0.688)

0.055(0.282)

0.049(0.509)

0.055(0.951)

0.031(0.960)

0.037(0.664)

0.035(0.729)

0.037(0.642)

Bold represents P < 0.05 in all phylogenetic scenarios (candidate species is incompatible with ancestry in the specified tooth). N/A, not applicable (no M3 ofH. antecessor and H. ergaster are included in the sample).*Mean Procrustes distance obtained in the 12 evaluated phylogenetic scenarios.†Highest P value observed in the 12 different phylogenetic scenarios. P values represent the probability that a fossil species falls within the expecteddistribution of an ancestral node under the assumption of a Brownian motion model of evolution.

Table 2. Morphological affinities of candidate species with Neanderthals and with modern humans

Candidate P3 P4 M1 M2 M3 P3 P4 M1 M2 M3

SH group NEA NEA NEA — SAP NEA SAP NEA NEA NEAH. heidelbergensis SAP NEA — NEA NEA SAP NEA NEA NEA NEAH. antecessor NEA NEA NEA NEA N/A SAP NEA SAP NEA NEAH. erectus SAP NEA SAP NEA NEA SAP NEA SAP SAP SAPH. ergaster SAP NEA SAP NEA N/A SAP SAP SAP NEA NEA

Affinities based on Procrustes distances (Table S4). N/A, not applicable; NEA, higher affinity with Neanderthals; SAP, higher affinitywith modern humans; —, higher but not significant affinity with Neanderthals.




patterns in these lineages are consistent with a Brownian processof nonadaptive, neutral evolution in both dental (25) and cranialfeatures (27). Moreover, detailed, comprehensive evaluations ofpostcanine dental morphology have revealed that several dentaltraits shared by European late Early and Middle Pleistocenepopulations and classic Neanderthals are, in fact, derived (28–31).The suite of dental traits shared by European hominins sug-

gests that a separate European clade with Neanderthal affinitiesexisted since at least 1 Ma. This does not necessarily implya continuity between all European Pleistocene populationsduring the last 1 My (32), but it points to all these taxa being partof a Neanderthal phenotypic clade separate from the lineageleading to modern humans. It has been suggested that this Eu-ropean group could have been composed of successive migrantpopulations who shared their common phenotypes via gene flow(ref. 33, see also refs. 15 and 34), resulting in a lack of linearity inthe pattern of acquisition of Neanderthal traits (33). It is worthnoting that extensive hybridization would hamper ancestral re-construction because the expected morphological differenceswould be attenuated by gene flow between diverging populations.

Molecular Divergence Times May Be Underestimated. The paleon-tological estimate of the divergence time between Neanderthalsand modern humans at ca. 1 Ma is substantially older than mostmolecular estimates, which set the divergence at less than 500 ka(11, 15). The divergence between Neanderthals and modernhumans has also been dated to 300–400 ka using neutrallyevolving cranial measurements (35), an estimate that is highlydependent on assumptions and parameters that are not knownwithout error, but which is broadly consistent with the youngestmolecular clock estimates. In light of the young molecular andcranial dates, several analyses have dismissed the early di-vergence model of Neanderthal origins (1, 3, 11). However,a late divergence model has been recently challenged on thebasis of slower mutation rates that result in earlier split times inape and human evolution (refs. 36–38 but see also refs. 39 and40). Attempts to reconcile paleontological and molecular dateshave to face the fact that alternative molecular estimates for thetiming of the [N-MH]LCA vary considerably, with confidenceintervals ranging from 300 to 850 ka (11, 37). Our results arehowever incongruous with even the lowest boundary of molecularconfidence intervals, and thus lend support to the early divergencemodel and suggest that molecular estimates of divergence timecan be underestimated.

Phenotypic Divergence and Speciation Could Be Decoupled. A thirdpossible alternative that can conciliate molecular and paleonto-logical observations is that phenotypic divergence in dental traitsbetween Neanderthals and modern humans significantly pre-dated the completion of speciation. This scenario could explainwhy someEuropean Pleistocene populations appear to be derivedin dental traits but still show primitive cranial and postcranialfeatures. Such mosaicism is particularly evident in H. antecessor’scombination of some derived Neanderthal dental features (29–31) and generally plesiomorphic nondental traits (at least for theclade H. neanderthalensis-H. sapiens; refs. 41–43), and in SH’scranial morphology, which is transitional to the Neanderthal form(10), combined with its strongly derived dental phenotype (12).It is most likely that the large discrepancy between molecular

dates and phenotypic differentiation results from the wide con-fidence intervals of the molecular estimates and from the un-certainty associated with dating fossil materials. Regardless, ourresults indicate that dental morphology developed derived fea-tures as early as almost 1 Ma in the Neanderthalization process(12), raising interesting questions about which evolutionaryforces may have caused this early dental differentiation withrespect to other traits and over what time interval occurred. Theevolutionary mechanisms behind this process are likely to consistof a mosaic of random factors (such as genetic drift) and non-random factors (such as lineage-specific patterns of environ-mental selection and rates of evolution of different traits; ref. 44).

New paleontological, molecular, and quantitative phenotypic stud-ies are needed to elucidate the exact nature of these factors.

ConclusionsOur study presents a statistical framework for addressing ques-tions about ancestry and the relationships of particular fossilsthat can serve as a model for making paleoanthropological hy-potheses more testable. The quantitative approach we proposeoffers an objective means to test hypotheses that capitalize onthe comparatively rich dental fossil record and address appar-ently unresolvable debates about hominin phylogeny. The sameapproach can be applied to other skeletal parts via 2D and 3Dquantifications of morphology, and it can be extended to in-clude models of evolution other than Brownian motion pro-cesses should there be evidence that they are appropriate. Ourapproach thus provides a promising and extensible statisticalcontext for evaluating fossil remains of uncertain phylogeneticposition.

Materials and MethodsMaterials. Our study analyzed the morphology of 1,200 hominin teeth rep-resenting their 10 postcanine tooth positions (upper and lower premolars andmolars) from 13 different species/morphotypes (Tables S1 and S2). We choseto split controversial named taxa for which all teeth are represented in oursamples to evaluate their ancestral status independently on their ownmerits, but we lumped taxa when their dentitions were incompletely rep-resented so as to provide more statistical power (e.g., Paranthropus aethio-picuswas included in Paranthropus boisei and H. rudolfensis in H. habilis). Theonly exceptions were H. ergaster and H. antecessor, which were retained asseparate candidate species despite their lack of M3 due to their strong can-didacy as [N-MH]LCA. Although other taxonomic arrangements are plausible(e.g., merging H. erectus, H. ergaster, and Dmanisi fossils under H. erectussensu lato), the purpose of our study is not to address taxonomic questions,but to evaluate the plausibility that previously recognized taxa occupyancestral positions. In this regard, the most informative approach is a de-tailed evaluation of morphologically, geographically, and chronologicallycohesive groups.

Geometric Morphometrics. Dental morphology was quantified using 2Dlandmarks and sliding semilandmarks (45) digitized from photographs of theocclusal surface of premolars and molars. Depending on the tooth, 4–8landmarks and 30–40 semilandmarks were used (Fig. S2 and S3). Coordinateswere digitized using tpsDig2 software (46). Orientation of dental specimenswas standardized such that the occlusal surface was parallel to the lens ofthe camera. Unworn and moderately worn teeth were included in ourevaluation, whereas severely worn teeth were excluded. A generalizedProcrustes analysis (47) and a principal components analysis (48) were carriedout using the mean configurations of the species (Dataset S1) to provideshape variables. Ancestral state calculations were carried out on these var-iables in the principal components space (Fig. 3 and Fig. S4) using all prin-cipal components (5).

Ancestral State Reconstruction. The GLM method for estimating ancestraltraits (6, 49, 50) was used, implemented in Mathematica by one of us (51).As is standard practice (52), our implementation of ancestral state recon-struction assumes that the traits are evolving under a Brownian motionmodel of the sort that results either from genetic drift or from selectionwhose direction and magnitude varies randomly through time and acrossspace. Our previous study indicated that hominin dental evolution is con-sistent with a Brownian motion model, although considerable stabilizingselection has been identified in first molars and slight directional trends inlower third premolars and upper third molars (25). Estimates of ancestralstates are not badly affected by stabilizing processes, only by long-termunidirectional selection that occurs in parallel in different lineages (21, 53).

Ancestral node reconstruction identifies the single most likely ancestralcondition and a multivariate confidence ellipsoid (95%) containing otherplausible but less probable ancestral phenotypes (6, 50, 54). The likelihooddistribution of the ellipsoids was used to calculate the probability that fossils ofappropriate age could feasibly occupy those nodes (SI Text).

The reconstructed ancestral shape variables were converted back intolandmark configurations (55), after which the Procrustes distances of can-didate species to the expected ancestor were calculated. Procrustes distanceis a measure of shape difference between superimposed conformations of

Gómez-Robles et al. PNAS Early Edition | 5 of 6

EVOLU

TION





http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental/sd01.txt



landmarks calculated as the square root of the sum of the squared differencesbetween the positions of homologous landmarks. The species that fallswithin the confidence ellipsoid and has the smallest Procrustes distance tothe hypothetical ancestor can be considered the best candidate to occupythe ancestral position. Procrustes distance was also used to calculate theaffinity of candidate species with Neanderthals and modern humans takinginto account the phylogenetic topology (SI Text).

ACKNOWLEDGMENTS.We thank Emília Martins, James Rohlf, Michelle Lawing,Bernard Wood, Asier Gómez-Olivencia, and David Sánchez-Martín for dis-cussion on several aspects of the manuscript. We also thank the editor and

three anonymous reviewers for their help in improving our manuscript. Ac-cess to fossil specimens was made possible by O. Kullmer, B. Denkel, andF. Schrenk (Senckenberg Institute); D. Lordkipanidze, A. Vekua, and G. Kiladze(Georgian National Museum); H. de Lumley, M. A. de Lumley, and A. Vialet(Institut de Paléontologie Humaine); P. Mennecier (Musée de l’Homme);I. Tattersall, G. Sawyer, and G. García (American Museum of Natural History);L. Jellema and J. Haile-Selassie (Cleveland Museum of Natural History); andM. Botella (Laboratory of Physical Anthropology, University of Granada). Thisworkwas partially supported by funding by the Spanish Ministry of Science andInnovation (Project CGL2009-12703-C03-01-02-03). Fieldwork at Atapuerca issupported by the Consejería de Cultura y Turismo (Junta de Castilla y León)and the Fundación Atapuerca.

1. Hublin JJ (2009) Out of Africa: Modern human origins special feature: The origin ofNeandertals. Proc Natl Acad Sci USA 106(38):16022–16027.

2. Schwartz JH, Tattersall I (2010) Fossil evidence for the origin of Homo sapiens. Am JPhys Anthropol 143(Suppl 51):94–121.

3. Stringer C (2012) The status of Homo heidelbergensis (Schoetensack 1908). Evol An-thropol 21(3):101–107.

4. González-José R, Escapa I, Neves WA, Cúneo R, Pucciarelli HM (2008) Cladistic analysisof continuous modularized traits provides phylogenetic signals in Homo evolution.Nature 453(7196):775–778.

5. Adams DC, Cardini A, Monteiro LR, O’Higgins P, Rohlf FJ (2011) Morphometrics andphylogenetics: Principal components of shape from cranial modules are neither ap-propriate nor effective cladistic characters. J Hum Evol 60(2):240–243.

6. Martins EP, Hansen TF (1997) Phylogenies and the comparative method: A generalapproach to incorporating phylogenetic information into the analysis of interspecificdata. Am Nat 149(4):646–667.

7. Strait DS (2013) Human systematics. A Companion to Paleoanthropology, ed Begun DR(Blackwell Publishing, Oxford), pp 35–54.

8. Wood B, Lonergan N (2008) The hominin fossil record: Taxa, grades and clades. J Anat212(4):354–376.

9. Tattersall I (2007) Neanderthals, Homo sapiens, and the question of species in pa-leoanthropology. J Anthropol Sci 85:139–146.

10. Arsuaga JL, Martínez I, Gracia A, Lorenzo C (1997) The Sima de los Huesos crania(Sierra de Atapuerca, Spain). A comparative study. J Hum Evol 33(2-3):219–281.

11. Endicott P, Ho SY, Stringer C (2010) Using genetic evidence to evaluate four palae-oanthropological hypotheses for the timing of Neanderthal and modern humanorigins. J Hum Evol 59(1):87–95.

12. Gómez-Robles A, Bermúdez de Castro JM, Martinón-Torres M, Prado-Simón L,Arsuaga JL (2012) A geometric morphometric analysis of hominin upper second andthird molars, with particular emphasis on European Pleistocene populations. J HumEvol 63(3):512–526.

13. Bischoff JL, et al. (2003) The Sima de los Huesos hominids date to beyond U/Thequilibrium (>350 kyr) and perhaps to 400-500 kyr: New radiometric dates. J ArchaeolSci 30(3):275–280.

14. Bischoff JL, et al. (2007) High-resolution U-series dates from the Sima de los Huesoshominids yields 600 kyrs: Implications for the evolution of the early Neanderthallineage. J Archaeol Sci 34(5):763–770.

15. Green RE, et al. (2010) A draft sequence of the Neandertal genome. Science328(5979):710–722.

16. Antón SC (2003) Natural history of Homo erectus. Am J Phys Anthropol 46(Suppl 37):126–170.

17. Shen G, Gao X, Gao B, Granger DE (2009) Age of Zhoukoudian Homo erectus de-termined with (26)Al/(10)Be burial dating. Nature 458(7235):198–200.

18. Rightmire GP (2008) Homo in the Middle Pleistocene: Hypodigms, variation, andspecies recognition. Evol Anthropol 17(1):8–21.

19. Parés JM, et al. (2013) Reassessing the age of Atapuerca-TD6 (Spain): New paleo-magnetic results. J Archaeol Sci 40(12):4586–4595.

20. Wood B, Collard M (1999) The human genus. Science 284(5411):65–71.21. Polly PD (2001) Paleontology and the comparative method: Ancestral node re-

constructions versus observed node values. Am Nat 157(6):596–609.22. Bräuer G (2008) The origin of modern anatomy: By speciation or intraspecific evo-

lution? Evol Anthropol 17(1):22–37.23. Rightmire GP (1998) Human evolution in the Middle Pleistocene: The role of Homo

heidelbergensis. Evol Anthropol 6(6):218–227.24. Stringer C (2002) Modern human origins: Progress and prospects. Philos Trans R Soc

Lond B Biol Sci 357(1420):563–579.25. Gómez-Robles A, Polly PD (2012) Morphological integration in the hominin dentition:

evolutionary, developmental, and functional factors. Evolution 66(4):1024–1043.26. Trinkaus E (2006) Modern human versus Neandertal evolutionary distinctiveness. Curr

Anthropol 47(4):597–620.27. Weaver TD, Roseman CC, Stringer CB (2007) Were neandertal and modern human

cranial differences produced by natural selection or genetic drift? J Hum Evol 53(2):135–145.

28. Bailey SE (2004) A morphometric analysis of maxillary molar crowns of Middle-LatePleistocene hominins. J Hum Evol 47(3):183–198.

29. Gómez-Robles A, et al. (2007) A geometric morphometric analysis of hominin upperfirst molar shape. J Hum Evol 53(3):272–285.

30. Martinón-Torres M, et al. (2007) Gran Dolina-TD6 and Sima de los Huesos dentalsamples: Preliminary approach to some dental characters of interest for phylogeneticstudies. Dental Perspectives on Human Evolution, eds Bailey SE, Hublin J-J (Springer,Berlin), pp 65–79.

31. Gómez-Robles A, de Castro JM, Martinón-Torres M, Prado-Simón L (2011) Crown sizeand cusp proportions in Homo antecessor upper first molars. A comment on Quamet al. 2009. J Anat 218(2):258–262.

32. MacDonald K, Martinón-Torres M, Dennell RW, Bermúdez de Castro JM (2012) Dis-continuity in the record for hominin occupation in South-Western Europe: Im-plications for occupation of the middle latitudes of Europe. Quat Int 271:84–97.

33. Dennell RW, Martinón-Torres M, Bermúdez de Castro JM (2011) Hominin variability,climatic instability and population demography in Middle Pleistocene Europe. QuatSci Rev 30(11-12):1511–1524.

34. Jolly CJ (2001) A proper study for mankind: Analogies from the Papionin monkeys andtheir implications for human evolution. Am J Phys Anthropol 116(Suppl 33):177–204.

35. Weaver TD, Roseman CC, Stringer CB (2008) Close correspondence between quanti-tative- and molecular-genetic divergence times for Neandertals and modern humans.Proc Natl Acad Sci USA 105(12):4645–4649.

36. Hawks J (2012) Longer time scale for human evolution. Proc Natl Acad Sci USA109(39):15531–15532.

37. Langergraber KE, et al. (2012) Generation times in wild chimpanzees and gorillassuggest earlier divergence times in great ape and human evolution. Proc Natl Acad SciUSA 109(39):15716–15721.

38. Scally A, Durbin R (2012) Revising the human mutation rate: Implications for un-derstanding human evolution. Nat Rev Genet 13(10):745–753.

39. Fu Q, et al. (2013) A revised timescale for human evolution based on ancient mito-chondrial genomes. Curr Biol 23(7):553–559.

40. Green RE, Shapiro B (2013) Human evolution: Turning back the clock. Curr Biol 23(7):R286–R288.

41. Bermúdez de Castro JM, et al. (1997) A hominid from the lower Pleistocene of Ata-puerca, Spain: Possible ancestor to Neandertals and modern humans. Science276(5317):1392–1395.

42. Arsuaga J-L, et al. (1999) The human cranial remains from Gran Dolina Lower Pleis-tocene site (Sierra de Atapuerca, Spain). J Hum Evol 37(3-4):431–457.

43. García-González R, et al. (2009) Étude analytique d’une clavicule complète de sub-adulte d’Homo antecessor (site de Gran Dolina, Sierra d’Atapuerca, Burgos, Espagne).Anthropologie 113(1):222–232, French.

44. Pérez SI, Bernal V, González PN, Sardi M, Politis GG (2009) Discrepancy betweencranial and DNA data of early Americans: Implications for American peopling. PLoSONE 4(5):e5746.

45. Bookstein FL (1996) Applying landmark methods to biological outline data. ImageFusion and Shape Variability Techniques, eds Mardia KV, Gill CA, Dryden IL (LeedsUniv Press, Leeds, UK), pp 79–87.

46. Rohlf FJ (2005) tpsDig2 software. (State Univ of New York, Stony Brook, NY).47. Rohlf FJ, Slice D (1990) Extensions of the Procrustes method for the optimal super-

imposition of landmarks. Syst Zool 39(1):40–59.48. Dryden IL, Mardia KV (1998) Statistical Shape Analysis (Wiley, New York).49. Polly PD (2008) Adaptive zones and the pinniped ankle: A three-dimensional quan-

titative analysis of carnivoran tarsal evolution.Mammalian Evolutionary Morphology,a Tribute to Frederick S. Szalay, eds Sargis E, Dagosto M (Springer, Dordrecht, TheNetherlands), pp 167–196.

50. Rohlf FJ (2001) Comparative methods for the analysis of continuous variables: Geo-metric interpretations. Evolution 55(11):2143–2160.

51. Polly PD (2012) Geometric morphometrics for Mathematica. Version 9.0. (IndianaUniversity ScholarWorks). Available at http://hdl.handle.net/2022/14613. June 1, 2013.

52. Pagel M (2002) Modelling the evolution of continuously varying characters on phy-logenetic trees: The case of hominid cranial capacity. Morphology, Shape and Phy-logeny, eds MacLeod N, Forey PL (Taylor & Francis, London), pp 269–286.

53. Oakley TH, Cunningham CW (2000) Independent contrasts succeed where ancestorreconstruction fails in a known bacteriophage phylogeny. Evolution 54(2):397–405.

54. Felsenstein J (2004) Inferring Phylogenies (Sinauer, Sunderland, MA).55. Polly PD (2004) On the simulation of the evolution of morphological shape: Multi-

variate shape under selection and drift. Palaeontol Electronica 7(2):1–28.56. Hunt G (2004) Phenotypic variation in fossil samples: Modeling the consequences

of time-averaging. Paleobiology 30(3):426–443.



http://hdl.handle.net/2022/14613


Supporting InformationGómez-Robles et al. 10.1073/pnas.1302653110SI TextDefinition of Landmarks. The configurations of landmarks andsemilandmarks used in our study are as noted in the following 10sections (Figs. S2 and S3).Upper third premolars (P3).

1 = mesial fovea

2 = tip of the buccal cusp or paracone

3 = distal fovea

4 = tip of the lingual cusp or protocone

5–43 = equidistant sliding semilandmarks

Upper fourth premolars (P4).

1 = mesial fovea

2 = tip of the buccal cusp or paracone

3 = distal fovea

4 = tip of the lingual cusp or protocone


Upper first molars (M1).

1 = tip of the mesiolingual cusp or protocone

2 = tip of the mesiobuccal cusp or paracone

3 = tip of the distobuccal cusp or metacone

4 = tip of the distolingual cusp or hypocone

5–34 = equiangular sliding semilandmarks

Upper second molars (M2).




4 = central fovea (intersection between the buccal groove andthe central groove)

5 = distal fovea (intersection between the central groove, thedistobuccal groove and the distolingual groove)


Upper third molars (M3).




4 = central fovea (intersection between the buccal groove andthe central groove)


Lower third premolars (P3).

1 = tip of the buccal cusp or protoconid

2 = distal fovea

3 = tip of the mesial cusp or metaconid

4 = mesial fovea


Lower fourth premolars (P4).

1 = tip of the buccal cusp or protoconid

2 = distal fovea

3 = tip of the mesial cusp or metaconid

4 = mesial fovea


Lower first molars (M1).

1 = mesial fovea or most mesial extreme of the central groove

2 = intersection between the central groove and the mesio-buccal groove

3 = intersection between the central groove and the lingualgroove

4 = intersection between the central groove and the distobuc-cal groove

5 = tip of the mesiobuccal cusp or protoconid

6 = tip of the mesiolingual cusp or metaconid

7 = tip of the distobuccal cusp or hypoconid

8 = tip of the distolingual cusp or entoconid


Lower second molars (M2).

1 = mesial fovea or most mesial extreme of the central groove








Lower third molars (M3).








Definition of landmarks and semilandmarks followed previouslypublished studies (1–6). Equiangular landmarks were used for thoseteeth in which the centroid defined by homologous landmarksapproximates the anatomical center of the molar/premolar, thusgiving rise to approximately equispaced semilandmarks (2). For

Gómez-Robles et al. www.pnas.org/cgi/content/short/1302653110 1 of 10

www.pnas.org/cgi/content/short/1302653110

those tooth positions in which the centroid defined by landmarksis off-centered with respect to the anatomical center, we usedequispaced semilandmarks. Both strategies intend to obtain ap-proximately equally spaced semilandmarks that were later slid us-ing the criterion of minimization of the Procrustes distance (7).

Reconstruction of Ancestral Dental Morphology. Ancestral dentalmorphology and associated confidence ellipsoids were estimatedusing the generalized linear model (GLM) method (8). Thismethod regresses traits, which here are shape variables in theform of principal components (PC) scores, onto a phylogeny tosimultaneously estimate the rate of trait evolution (in varianceunits), the most likely ancestral value for the traits at each node,and confidence ellipsoids associated with each estimate. Thismethod can be combined with geometric morphometrics to es-timate ancestral shapes, to estimate confidence ellipsoids aroundthe nodes in shape space, and to project a phylogenetic tree intogeometric morphometric shape space (9, 10). A Brownian mo-tion evolutionary process, which can arise from drift or fromselection that changes direction and magnitude randomly acrossthe tree, is standardly used as a null model for reconstructingancestral trait values (11, 12). Parallel directional selectionacross lineages can strongly undermine the reconstruction ofancestral traits, however the best ancestral estimates are thesame under stabilizing selection and Brownian motion (13, 14).A previous evaluation of the same taxa as we use in this studyindicated that hominin dental evolution is generally consistentwith a Brownian motion model (15). The reconstructions fromGLM using a Brownian motion model are identical to thosefound by squared-change parsimony (16) and maximum-likeli-hood (17).Each node reconstruction has an associated confidence ellip-

soid that describes the uncertainty due to the evolutionary process(e.g., different evolutionary pathways consistent with the Brow-nian motion model) and parameter estimation (e.g., error inestimating the tip phenotypes and the phylogenetic tree; ref. 8).Our ellipsoids take into account only the uncertainty due to theevolutionary process because we have dealt with the uncertaintydue to the tree by performing the analysis on 12 different phy-logenetic topologies. Each ellipsoid is multivariate normal anddepends on the rate variances of the traits, the temporal lengthof the branches, and the tree topology (8, 17–19). The shortbranches required for a late split (Fig. S1) constrain the range ofplausible ancestral phenotypes and reduce the confidence ellip-soids around the nodes; thus, the ancestral candidates becomeless plausible as the divergence becomes younger.The likelihood distribution of the ellipsoids was used to cal-

culate the probability that real fossils of appropriate age couldfeasibly occupy those nodes. The probability that a fossil could bedrawn from the multivariate ellipsoid at the node is based on thesquared Mahalanobis distance between it and the node. P valuesare calculated using the PC scores of the fossil, the PC scores ofthe node, and the covariance matrix of the node (which is a di-agonal matrix of the squared SDs associated with the evolu-tionary process). P is taken from a χ2 distribution whose degreesof freedom equal the number of morphospace dimensions beingconsidered. The P value is thus the probability that the fossilcould represent the ancestral population at that node givena Brownian motion process of evolution, the rate of evolutionestimated from the tree, the topology of the tree, and the shapesof the tip taxa.All calculations were implemented inMathematica by one of us

(20) using the functions ProbabilitiesOfShapesAsAncestors[],ReconstructAncestorShapes[], TreeAndFossilsToMorphospace[], and ReconstructNodes[].

Morphological Description of the [H]LCA and [P-H]LCA. The followingdescriptions of ancestral dentitions are based on the ancestralreconstructions of tooth shapes described above.The last common ancestral species to the genus Homo ([H]

LCA) is expected to have nearly symmetric upper premolars buta strong asymmetry in its lower premolars. The upper first molarswould have a more squared morphology than that of the LCA ofNeanderthals and modern humans ([N-MH]LCA) and lowerfirst molars would be very similar to those observed in the [N-MH]LCA. Both upper second and third molars are expected tohave four fully developed cusps, although upper third molarsshow certain reduction of the distobuccal cusp (metacone). Bothlower second and third molars are expected to have in the ma-jority of individuals five well developed cusps arranged in a Dry-opithecus (Y5) pattern in second molars and with a cruciformpattern in third molars (Fig. S5).The common ancestor of Paranthropus and Homo [P-H]LCA

is expected to have asymmetric upper premolars and stronglyasymmetric lower premolars with a large distal talonid involvingboth the lingual and buccal aspects of the distal surface. Upperfirst molars would have a squared morphology and lower firstmolars would present five well-defined cusps, a Y5 pattern,a broad buccolingual diameter, and an indentation separatingthe talonid from the trigonid. Both upper second and thirdmolars would present four well-defined cusps, including a well-defined hypocone. Lower second molar morphology would besimilar to lower first molar’s, whereas lower third molars wouldhave five cusps but would show in some cases non-Dryopithecuspatterns (Fig. S5). The main difference between the dentalmorphology of the [H]LCA and the [P-H]LCA would be foundin the relative reduction on the talon (upper premolars andmolars) and talonid (lower premolars and molars) basins ob-served in the [H]LCA.

Calculation of Morphological Affinities of Candidate Species toRepresent the [N-MH]LCA. Our analysis was based on phyloge-netic topology and thus took into account that candidate speciesfor the [N-MH]LCA are expected to be more similar to Nean-derthals than to modern humans because modern humans havea longer divergence time from this ancestor than Neanderthals. Acandidate species was considered to have higher affinity with H.sapiens if the Procrustes distance between that species and H.sapiens was shorter than the distance between it and H. nean-derthalensis (Table S2). A candidate species was considered tohave significantly higher affinity with Neanderthals if the Pro-crustes distance between that species and Neanderthals waslower than the distance between that species and H. sapiens bymore than

0:1Chrono½CLCA�PDist½½N−MH�LCA−MH�;

where 0.1 is the difference in age of modern humans (dated to0 Ma) versus Neanderthals (dated to 0.1 Ma); Chrono[CLCA]is the age of the candidate species to represent the [N-MH]LCA (0.5 Ma for H. heidelbergensis; 0.95 Ma for H. antecessor;1.65 Ma for H. ergaster; 1 Ma for H. erectus); and PDist½½N−MH�LCA−MH� is the Procrustes distance between the expectedmorphology of the [N-MH]LCA and H. sapiens. As the morphol-ogy of the [N-MH]LCA is involved in the calculation of morpho-logical affinities, these affinities are susceptible to change whenevaluating the 12 different phylogenetic scenarios. However, ourresults demonstrate that those changes are minimal such that in3 cases of the 50 comparisons (5 candidate species and 10 toothpositions) a species with a general significant higher affinity withNeanderthals shows no significant affinity in three phylogeneticscenarios.



1. Gómez-Robles A, Bermúdez de Castro JM, Martinón-Torres M, Prado-Simón L,Arsuaga JL (2012) A geometric morphometric analysis of hominin upper second andthird molars, with particular emphasis on European Pleistocene populations. J HumEvol 63(3):512–526.

2. Gómez-Robles A, et al. (2007) A geometric morphometric analysis of hominin upperfirst molar shape. J Hum Evol 53(3):272–285.

3. Gómez-Robles A, et al. (2008) Geometric morphometric analysis of the crownmorphology of the lower first premolar of hominins, with special attention toPleistocene Homo. J Hum Evol 55(4):627–638.

4. Gómez-Robles A, Martinón-Torres M, Bermúdez de Castro JM, Prado-Simón L,Arsuaga JL (2011) A geometric morphometric analysis of hominin upper premolars.Shape variation and morphological integration. J Hum Evol 61(6):688–702.

5. Gómez-Robles A, Olejniczak AJ, Martinón-Torres M, Prado-Simón L, Bermúdez deCastro JM (2011) Evolutionary novelties and losses in geometric morphometrics: Apractical approach through hominin molar morphology. Evolution 65(6):1772–1790.

6. Martinón-Torres M, et al. (2006) Hominin lower second premolar morphology:Evolutionary inferences through geometric morphometric analysis. J Hum Evol 50(5):523–533.

7. Pérez SI, Bernal V, González PN (2006) Differences between sliding semi-landmarkmethods in geometric morphometrics, with an application to human craniofacial anddental variation. J Anat 208(6):769–784.

8. Martins EP, Hansen TF (1997) Phylogenies and the comparative method: A generalapproach to incorporating phylogenetic information into the analysis of interspecificdata. Am Nat 149(4):646–667.

9. Rohlf FJ (2001) Comparative methods for the analysis of continuous variables:Geometric interpretations. Evolution 55(11):2143–2160.

10. Polly PD (2008) Adaptive zones and the pinniped ankle: A three-dimensionalquantitative analysis of carnivoran tarsal evolution. Mammalian EvolutionaryMorphology, a Tribute to Frederick S. Szalay, eds Sargis E, Dagosto M (Springer,Dordrecht, The Netherlands), pp 167–196.

11. Pagel M (2002) Modelling the evolution of continuously varying characters onphylogenetic trees: the case of hominid cranial capacity. Morphology, Shape andPhylogeny, eds MacLeod N, Forey PL (Taylor & Francis, London), pp 269–286.

12. Felsenstein J (2004) Inferring Phylogenies (Sinauer, Sunderland, MA).13. Oakley TH, Cunningham CW (2000) Independent contrasts succeed where ancestor

reconstruction fails in a known bacteriophage phylogeny. Evolution 54(2):397–405.14. Polly PD (2004) On the simulation of the evolution of morphological shape:

Multivariate shape under selection and drift. Palaeontol Electronica 7(2):1–28.15. Gómez-Robles A, Polly PD (2012) Morphological integration in the hominin dentition:

evolutionary, developmental, and functional factors. Evolution 66(4):1024–1043.16. Maddison WP (1991) Squared-change parsimony reconstructions of ancestral states

for continuous valued characters on a phylogenetic tree. Syst Biol 40(3):304–314.17. Schluter D, Price T, Mooers AO, Ludwig D (1997) Likelihood of ancestor states in

adaptive radiation. Evolution 51(6):1699–1711.18. Garland T, Jr., Ives AR (2000) Using the past to predict the present: confidence

intervals for regression equations in phylogenetic comparative methods. Am Nat155(3):346–364.

19. Polly PD (2001) Paleontology and the comparative method: ancestral nodereconstructions versus observed node values. Am Nat 157(6):596–609.

20. Polly PD (2012) Geometric morphometrics for Mathematica. Version 9.0. (IndianaUniversity ScholarWorks). Available at http://hdl.handle.net/2022/14613. AccessedJune 1, 2013.

Fig. S1. Phylogenetic trees used in our study. Top row (from left to right): Tree 1: Sima de los Huesos (SH) group included in the phylogeny as a pre-Neanderthal population, [N-MH]LCA dated to 1 Ma, H. erectus dated to 1 Ma (this is the phylogeny used for results in Figs. 2 and 3 and Figs. S4 and S5). Tree 2: SHgroup excluded from the phylogeny, [N-MH]LCA dated to 1 Ma, H. erectus dated to 1 Ma. Tree 3: SH group included in the phylogeny as a pre-Neanderthalpopulation, [N-MH]LCA dated to 1 Ma, H. erectus dated to 0.5 Ma. Tree 4: SH group excluded from the phylogeny, [N-MH]LCA dated to 1 Ma, H. erectus dated to0.5 Ma.Middle row: Tree 5: SH group included in the phylogeny as a pre-Neanderthal population, [N-MH]LCA dated to 0.45 Ma, H. erectus dated to 1 Ma. Tree 6:SH group excluded from the phylogeny, [N-MH]LCA dated to 0.45 Ma, H. erectus dated to 1 Ma. Tree 7: SH group included in the phylogeny as a pre-Neanderthalpopulation, [N-MH]LCA dated to 0.45 Ma, H. erectus dated to 0.5 Ma. Tree 8: SH group excluded from the phylogeny, [N-MH]LCA dated to 0.45 Ma, H. erectusdated to 0.5 Ma. Bottom row: Tree 9: SH group included in the phylogeny as a pre-Neanderthal population, [N-MH]LCA dated to 1 Ma, H. erectus excluded fromthe phylogeny. Tree 10: SH group excluded from the phylogeny, [N-MH]LCA dated to 1 Ma, H. erectus excluded from the phylogeny. Tree 11: SH group includedin the phylogeny as a pre-Neanderthal population, [N-MH]LCA dated to 0.45 Ma, H. erectus excluded from the phylogeny. Tree 12: SH group excluded from thephylogeny, [N-MH]LCA dated to 0.45 Ma, H. erectus excluded from the phylogeny. The same abbreviations as in Fig. 3 are used.


http://hdl.handle.net/2022/14613


Fig. S2. Landmarks and semilandmarks used in the analysis of upper third premolars (P3) and upper fourth premolars (P4) (upper left and upper right in thepanel) and lower third premolars (P3) and lower fourth premolars (P4) (lower left and lower right in the panel). Premolar orientation: buccal: top; lingual:bottom; mesial: left; distal: right.

Fig. S3. Landmarks and semilandmarks used in the analysis of upper first molars (M1), upper second molars (M2), and upper third molars (M3) (upper left,center, and right in the panel) and lower first molars (M1), lower second molars (M2), and lower third molars (M3) (lower left, center, and right in the panel).Molar orientation: buccal: top; lingual: bottom; mesial: left; distal: right.



Fig. S4. Principal components analyses of upper third premolar (P3), lower third premolar (P3), upper fourth premolar (P4), lower fourth premolar (P4), uppersecond molar (M2), lower second molar (M2), upper third molar (M3), and lower third molar (M3) morphologies. Analyses are based on species mean shapes. Thesame conventions as in Fig. 3 have been used.



Fig. S5. Dental morphology of the [H]LCA and the [P-H]LCA. Expected ancestral morphologies are represented as TPS-grids showing the deformation of themorphology corresponding to node 0 (base of the tree) into the estimated ancestral shape. TPS-grids corresponding to the [P-H]LCA are not deformed becausethis is the node corresponding to the base of our phylogeny. Main morphological differences between the [P-H]LCA and the [H]LCA are located at distal areasof premolars and molars, which are more developed in the [P-H]LCA. Photographs do not show the expected ancestral morphologies, but they represent modelteeth for each dental position represented below as a TPS-grid.

Table S1. Number of specimens per species and dental classincluded in the analysis

Species P3 P4 M1 M2 M3 P3 P4 M1 M2 M3

A. afarensis 4 3 6 2 2 7 8 6 9 7A. africanus 7 7 9 9 8 5 2 4 5 8P. robustus 4 5 5 4 4 1 4 4 4 3P. boisei* 3 4 2 3 3 2 2 2 3 4H. habilis† 4 5 12 7 6 5 3 5 4 3H. ergaster 2 2 2 2 — 3 3 3 3 2H. georgicus‡ 1 2 2 2 1 2 2 2 2 1H. erectus 8 6 5 7 8 10 5 12 11 5H. antecessor{ 2 2 3 1 — 2 3 3 3 2H. heidelbergensis§ 2 1 2 2 2 4 3 3 4 2SH groupjj 15 16 15 17 22 15 19 19 20 21H. neanderthalensis 16 20 18 18 11 20 18 22 18 14H. sapiens** 46 43 54 45 37 53 44 47 41 36

Total 114 116 135 119 104 129 116 132 127 108

*Including P. boisei and P. aethiopicus.†Including H. habilis and H. rudolfensis.‡Plio-Pleistocene fossils from Dmanisi site (Georgia).{Lower Pleistocene fossils from Atapuerca-TD6 site (Spain).§Including only Arago fossils.jjIncluding SH, Steinheim, Pontnewydd, and Montmaurin fossils.**Including fossil and recent H. sapiens.



Table S2. List of specimens and sites used in our study

Species Specimens

A. afarensis Hadar: AL 128–23, AL 145–35, AL 188–1, AL 199–1, AL 200–1a, AL 207–13, AL 266–1, AL 288, AL 333–1, AL 333–2, AL 333W-57,AL 333W-60, AL 400–1a, AL 486–1

Laetoli: LH 2, LH 3, LH 4, LH 6, LH 17, LH 21A. africanus Makapansgat: MLD 2, MLD 4, MLD 6, MLD 18, MLD 19, MLD 23, MLD 28

Sterkfontein: Sts 1, Sts 8, Sts 17, Sts 21, Sts 22, Sts 24, Sts 52, Sts 56, Sts 57, Sts 60, Stw 2, Stw 3, Stw 6, Stw 14, Stw 73, Stw 183,Stw 252, Stw 384, Stw 498, TM 1511

TaungP. robustus Sterkfontein: TM 1517

Swartkrans: SK 11, SK 13/14, SK 23, SK 46, SK 48, SK 55, SKW 5, SKW 10, SKX 4446P. boisei Chesowanja: KNM-CH1

East Turkana: KNM-ER 729, KNM-ER 3230, KNM-ER 15930Olduvai: OH 5, OH 30Peninj: 1West Turkana: KMN-WT 17400Omo: 51–79, L338X-40, L427-7

H. habilis East Turkana: KNM-ER 1590, KNM-ER 1802, KNM-ER 1805, KNM-ER 1813Olduvai: OH 4, OH 6, OH 7, OH 13, OH 16, OH 21, OH 24, OH 39, OH 41, OH 44Sterkfontein: SE 255, Stw 53, Stw 80, Stw 151Swartkrans: SK 27

H. ergaster Baringo: BK 67East Turkana: KNM-ER 820, KNM-ER 992, KNM-ER 3733Olduvai: OH 22Swartkrans: SKX 268West Turkana: KNM-WT 15000

H. georgicus Dmanisi*: D211, D2282, D2700, D2735, D3672H. erectus Sangiran*: S1a, S1b, S4, S6, S7†

Trinil: 5Zhoukoudian: 35.77, 22687, A2, B3.9, F1.5, G1.6, K1.96, PA327 and isolated teeth

H. antecessor Atapuerca-TD6*: ATD6-3, ATD6-4, ATD6-5, ATD6-7, ATD6-9, ATD6-12, ATD6-18, ATD6-69, ATD6-93, ATD6- 96, ATD6-103,ATD6-113, ATD6-125

H. heidelbergensis Arago*: 7, 10, 13, 14, 16, 21, 26, 28, 31, 40, 54, 69, 71, 75, 89, 103SH group‡ Atapuerca-SH*: Ind I-XXVIII and isolated teeth

Montmaurin*Pontnewydd: 2, 4, 8, 11, 12, 15, 17, 18, 19, 21Steinheim

H. neanderthalensis{ Amud, Arcy-sur-Cure (Grotte du Renne)*, Arcy-sur-Cure (Grotte du Bison)*, Arcy-sur-Cure (Grotte de l’Hyene)*, Breuil,Ehringsdorf, Guattari, Hortus*, Krapina, Kulna, Le Moustier, Malarnaud*, Monsempron*, Pech-de-l’Azé*, Petit-Puymoyen*, Pinilla del Valle*, St. Cesaire, Tabun, Vindija

Fossil H. sapiens{ Abri Pataud*, Almonda, Dolni Vestonice, Grimaldi, Isturitz*, Les Rois*, Mladec, Qafzeh*, St. Germain la Rivière*, TrouMagritte

Recent H. sapiens{ Heidenheim (Germany)*La Torrecilla (Spain)*CENIEH (Spain)*

*Original collections studied.†Including all isolated teeth in the S7 sample.‡The SH group includes European Middle Pleistocene fossils with clear Neanderthal affinities. The Steinheim skull has been classified within SH group followingthe morphological affinity described in ref. 1. Montmaurin and Pontnewydd affinity with Neanderthals is demonstrated by their frequent adscription to thisspecies (e.g., ref. 2). These fossils have been classified as part of SH group due to the younger age of classic Neanderthals.{H. neanderthalensis and H. sapiens are listed by site due to the large number of isolated teeth. All of the other species are listed by specimen.

1. Tattersall I (2007) Neanderthals, Homo sapiens, and the question of species in paleoanthropology. J Anthropol Sci 85:139–146.2. Bailey SE, Hublin J-J (2006) Dental remains from the Grotte du Renne at Arcy-sur-Cure (Yonne). J Hum Evol 50(5):485–508.



Table S3. Compatibility of candidate species with ancestral positions

P3 P4 M1 M2 M3 P3 P4 M1 M2 M3

[N-MH]LCASH group 0/6 2/6 6/6 2/6 2/6 0/6 2/6 6/6 2/6 2/6H. heidelbergensis 3/12 2/12 0/12 0/12 1/12 6/12 12/12 10/12 2/12 9/12H. antecessor 4/12 6/12 0/12 0/12 N/A 0/12 12/12 12/12 0/12 7/12H. erectus 4/4 4/4 4/4 0/4 0/4 0/4 4/4 4/4 4/4 4/4H. ergaster 3/12 7/12 0/12 0/12 N/A 1/12 12/12 12/12 12/12 9/12

[H]LCAH. georgicus 4/8 8/8 8/8 8/8 0/8 0/8 8/8 8/8 8/8 6/8H. ergaster 6/8 8/8 1/8 4/8 N/A 5/8 8/8 8/8 8/8 8/8H. habilis 6/8 8/8 8/8 4/8 8/8 8/8 8/8 8/8 8/8 8/8

[P-H]LCAH. habilis 12/12 12/12 12/12 12/12 12/12 11/12 12/12 12/12 12/12 12/12A. africanus 12/12 10/12 12/12 12/12 8/12 12/12 12/12 12/12 12/12 12/12A. afarensis 10/12 12/12 12/12 11/12 12/12 8/12 12/12 12/12 12/12 12/12

Number of phylogenetic scenarios (out of the total number of phylogenies in which a given species has beenevaluated as a candidate) for which each tooth and candidate is compatible with the corresponding expectedancestral morphology. N/A, not applicable.



Table

S4.

Procrustes

distancesfrom

candidatesp

eciesto

representthe[N

-MH]LCA

toNea

nderthalsan

dmodernhuman

s

Can

didate

P3P4

M1

M2

M3

P 3P 4

M1

M2

M3

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

NEA

SAP

SHgroup

0.03

7*0.05

20.02

1*0.03

90.01

7*0.04

30.03

50.03

70.05

30.05

0†0.03

7*0.05

00.04

70.04

4†0.02

1*0.02

90.02

8*0.04

00.03

7*0.05

1H.heidelbergen

sis

0.04

30.04

0†0.05

9*0.06

40.05

30.05

70.07

7*0.08

50.06

1*0.07

70.06

10.05

6†0.02

3*0.05

20.02

8*0.03

30.02

5*0.04

70.03

8*0.04

6H.an

tecessor

0.03

7*0.04

00.03

7*0.04

80.04

7*0.06

80.05

1*0.06

9N/A

0.08

70.08

0†0.04

6*0.06

00.03

50.03

4†0.03

7*0.05

90.03

5*0.04

9H.erectus

0.03

10.03

1†0.01

9*0.03

50.03

60.03

1†0.05

3*0.06

00.06

5*0.08

30.08

80.07

8†0.03

4*0.05

50.03

40.03

0†0.04

30.04

0†0.03

80.03

2†

H.ergaster

0.04

90.04

4†0.04

1*0.04

80.06

70.04

2†0.07

0*0.08

0N/A

0.06

70.05

6†0.04

20.03

3†0.03

90.03

7†0.02

9*0.04

30.03

9*0.04

4

N/A,notap

plicab

le.

*Significantlyhigher

affinitywithNea

nderthals.

†Significantlyhigher

affinitywithmodernhuman

s.



Other Supporting Information Files

Dataset S1 (TXT)


http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1302653110/-/DCSupplemental/sd01.txt


Documents

No known hominin species matches the expected … Robles et...No known hominin species matches the expected dental morphology of the last common ancestor of Neanderthals and modern